SAND-BURIED

RUINS
OF
KHOTAN
SAND-BURIED
RUINS OF KHOTAN
PERSONAL NARRATIVE OF A JOURNEY OF
ARCH/EOLOGICAL AND GEOGRAPHICAL
EXPLORATION IN CHINESE TURKESTAN
BY
M. AUREL STEIN
CHEAPER EDITION
WITH A MAP FROM ORIGINAL SURVEYS AND
NUMEROUS ILLUSTRATIONS
LONDON
HURST AND BLACKETT, LIMITED
13, Great Marlborough Street
1904
Ail rights reserved
AV
TO
THE MEMORY
OF
MY BROTHER
WHOSE LOVING CARE EVER FOLLOWED ME THROUGH LIFE,
THIS ACCOUNT OF MY JOURNEY,
FIRST RECORDED FOR HIM
IS INSCRIBED
IN UNCEASING AFFECTION AND SORROW
INTRODUCTION
The journey described in these pages was carried out in the year
igoo-oi, under the auspices of the Government of India. Its
main object was the systematic exploration of ancient remains
about Khotan and in the adjoining parts of the great desert of
Chinese Turkestan. The fresh materials thus brought to light
for the study of the early history and culture of those regions
were so extensive that my full scientific report must, by reason
of its bulk and cost, necessarily remain beyond the reach of the
general public. I have therefore gladly availed myself of the
permission accorded to me to publish independently the present
narrative, which is intended to record for a wider class of readers
my personal experiences and observations, as well as the main
facts concerning my antiquarian discoveries.
I have spared no trouble to render my account of the latter
accurate in its details and yet thoroughly inteUigible to the non-
Orientahst. It has been my hope to attract his interest to a
fascinating chapter of ancient history which witnessed inter-
change between the civilisations of India, China, and the
Classical West in that distant part of Central Asia, and which
seemed almost completely lost to us. If this hope is fulfilled,
and if at the same time these pages convey adequate impres-
sions of the strange scenes and conditions amidst which I passed
that year of trying but happy toil, I shall feel repaid for the
additional labour involved in the preparation of this narrative.
The circumstances which induced me to form the project of
these explorations, and the arrangements by which I was enabled
to carry it into execution, have already been explained in my
"
Preliminary Report on a Journey of Archaeological and Topo-
viii INTRODUCTION.
graphical Exploration in Chinese Turkestan," published in
1901
under the authority of the Secretary of State for India. Hence
a succinct notice of them may suffice here. The idea of archaeo-
v logical work about Khotan first suggested itself to me in the
spring of
1897,
in consequence of some remarkable antiquarian
acquisitions from that region. Among the papers left by the
distinguished but ill-fated French traveller, M. Dutreuil de
Rhins, were fragments of ancient birch-bark leaves, which had
been acquired in the vicinity of Khotan. On expert examina-
j
tion they proved to contain a Buddhist text in an early Indian
< script and language, and were soon recognised as the oldest
1 Indian manuscript then known, going back to the first centuries
of our era.
About the same time the
"
British collection of Central-Asian
antiquities
"
formed at Calcutta through the efforts of Dr. A. F.
Rudolf Hoernle, c.i.e., received from the same region notable
additions, consisting of fragments of paper manuscripts, pieces
of ancient pottery, and similar relics. Thej^ had been sold to
representatives of the Indian Government in Kashgar, Kashmir
and Ladak as finds made by native
"
treasure-seekers
"
at ancient
sites about Khotan. Similar purchases had reached public
collections at St. Petersburg through the Russian Consul-General
at Kashgar and others. A curious feature of all these acquisi-
tions made from a distance was that, besides unmistakably
genuine documents in Indian and Chinese writing, they included
a large proportion of texts displaying a strange variety of
entirely
"
unknown scripts," which could not fail to arouse
suspicion. While the materials thus accumulated, no reliable
information was ever forthcoming as to the exact origin of the
finds or the true character of the ruined sites which were supposed
to have furnished them. No part of Chinese Turkestan had then
^been explored from an archaeological point of view, and it
struck me that, however much attention these and other future
discoveries might receive from competent Orientalists in Euro})e,
their full historical and antiquarian value could never be realised
without systematic researches on the spot.
The practicable nature of the project was proved in the mean-
time by the memorable march which Dr. Hedin made in the
PLAN OF EXPLORATIONS.
ix
winter of 1896
past two areas of sand-buried ruins in the desert
north-east of Khotan. Though the distinguished explorer,
during his necessarily short halt at each place, was unable to
seciue any exact evidence as to the character and date of the
ruins, this discovery (of which the first account reached me in
1898)
sufficed to demonstrate both the existence and the com-
parative accessibility of ancient sites likely to reward exca-
vation.
It was only in the summer of 1898^that I found leisiu-e to
work out the detailed plan of my journey and to submit it with
Dr. Hoernle's weighty recommendation to the Indian Govern-
ment, whose sanction and assistance were indispensable for its
execution. Generously supported first by Sir Mackvvorth
Young, K.C.S.I., late Lieutenant-Governor of the Punjab, and
subsequently on my temporary transfer to Bengal by the late
Sir
John
Woodburn, K.c.s.i., the lamented head of that adminis-
tration and a zealous friend of Oriental learning, my proposals
met with favourable consideration on the part of Lord Curzon's
Government. In
July, 1899,
the scheme, in which Sir Charles
Rivaz, K.C.S.I., then Member of the Viceregal Council and now
Lieutenant-Governor of the Punjab, had from the first shown
kind personal interest, received the final sanction of the Govern-
ment of India. A resolution in the Department of Revenue and
Agriculture provided for my deputation on special duty to Chinese
Turkestan, during a period of one year. At the same time a
grant of Rs.
9,000
(;^6oo), partly from Imperial resources and
partly from contributions by the Local Governments of the
Punjab and Bengal, was placed at my disposal to meet the esti-
mated expenditure on the journey and explorations.
That, notNsithstanding the great distances and physical
obstacles to be overcome and in spite of all the uncertainties
attending an enterprise in a new field, I succeeded in accom-
pHshing the whole of my programme strictly within the sanc-
tioned estimates of time and expense, is a fact which from a prac-
tical and
quasi-administrative point of \'iew I feel proud to
record. How much anxious thought, calculation and effort its
attainment cost me, need scarcely be detailed here. Considering
the nature and extent of the ground covered by my travels, and
X INTRODUCTION.
the difficulties of work in the desert, the relatively low expendi-
ture involved in my explorations has since been noted with sur-
prise by brother archaeologists and others.
Long experience of marching and camping gained on Indian
ground certainly helped in restricting the cost. But even thus
the expenses of my expedition would certainly have been higher,
had not the Survey of India Department liberally offered its
assistance. Previous antiquarian tours in Kashmir, the Punjab,
and on the Afghan Frontier had taught me the importance of
exact topographical observation as an adjunct of my researches.
The necessity of fixing accurately the position of ancient sites
and generally elucidating the historical geography of the country
was bound to bring surveying operations in Chinese Turkestan
into the closest connection with my immediate task. But in
addition I was anxious from the first to utilise whatever oppor-
tunities the journey might offer for geographical work of a more
general character in regions which had hitherto remained without
a proper survey or altogether unexplored.
Colonel St. George Gore, r.e., c.s.i., Surveyor-General of
India, proved most willing to further this object. He kindly
agreed to depute with me one of the native Sub-Surveyors of his
Department, and to provide the necessary equipment of surveying
instruments, together with a special grant of Rs.
2,000 (;^I33), in
order to cover the additional expenses. Of the excellent services
rendered by Babu Ram Singh, the Sub-Surveyor selected, my
narrative gives ample evidence. With his help
a continuous
system of surveys, by plane-table, astronomical
observations and
triangulation, was carried on during the whole of my travels in
Chinese Turkestan. The results of these surveys, which in the
mountains I was able to supplement by photogrammetric survey
work of my own, and the direction and supervision of which
throughout claimed nmch of my time and attention, are now
embodied in maps published by the Trigonometrical Branch of
the Survey of India. From these the small scale map was pre-
pared which, with the kind permission of the Royal
Geographical
Society, has been reproduced for the present volume.
For the generous consideration and aid of the Indian Govern-
ment that alone enabled me to undertake the scientific enter-
AID OF INDIAN GOVERNMENT. xi
prise I had planned, I shall ever retain the feeling of deep and
sincere gratitude. Through it, I had secured at last the long
and eagerly sought chance to serve, in a new field and with a
measure of freedom such as had never fallen to my share, those
interests of Oriental research which had claimed me from the
commencement of my student days, and which had brought
me to India.
The twelve years since passed, mainly in the service of the
Punjab University, had taught me fully to appreciate the im-
portance of both time and money in regard to archaeological
labours. Though placed tantalisingly near to the ground which
by its ancient remains and historical associations has always
had a special fascination for me, I had rarely been able to devote
to antiquarian work more than brief intervals of hard-earned
leisure.
The fact that my administrative duties had no direct connec-
tion with my scientific interests, might well have made me feel
despondent about the chance of ever obtaining the means needed
for systematic archaeological explorations, even on well-known
ground and in easily accessible regions. For with, I fear, the
majority of fellow-workers I had failed to profit by the example
of the late Dr. Schhemann, who, before attempting to realise
his grand projects at Troy and Mykene, had resolutely set him-
self to assure that safest base of success, personal independence
and an ample reserve of funds.
The
exceptional help which the Indian Government, inspired
by Lord Curzon's generous interest in the history and anti-
quities of the East, had accorded to me, for a time removed the
difficulties
against which I had struggled, and brought the
longed-for
opportunity within my reach. But remembering
the
circumstances under which it had been secured, I could not
prevent anxious thoughts often crossing my mind in the course
of my
preparations and after. Would Fate permit the full exe-
cution of my plan within the available time, and would the
results prove an adequate return for the liberal consideration
and aid that the Government had extended to me ?
I knew well that neither previous training and experience, nor
careful
preparation and personal zeal, could guarantee success.
xii
INTRODUCTION.
The wide extent of the region to be searched and the utter
insufficiency of rehable information would alone have justified
doubts as to how much those sand-buried sites would yield up
during a limited season. But in addition there was the grave
fact that prolonged work in the desert such as I contemplated
would have to be carried through in the face of exceptional
physical difficulties and even dangers. Nor was it possible to
close my eyes to the very serious obstacles which suspicions of
the local Chinese administration and quasi-political apprehen-
sions, however unfounded, might raise to the realisation of my
programme.
When I now look back upon these anxieties and doubts, and
recognise in the light of the knowledge since gathered how much
there was to support them, I feel doubly grateful to the kindly
Destiny which saved my plans from being thwarted by any of
those difficulties, and which allowed my labours to be rewarded
by results richer than I had ventured to hope for. In respect
of the efforts and means by which these results were secured, no
remarks seem here needed ; the reader of my present narrative,
whatever his knowledge of Central Asia and its historical past
may be, can safely be left to judge of them for himself. But in
regard to the scientific value of the results similar reticence
would scarcely be justified, however much personal feelings might
make me incline towards it.
It is impossible to overlook the fact that archaeological researcli
in great fields like India and Central Asia, which lie beyond the
stimulating influence of Biblical associations, has not as yet
succeeded in gaining its due share of symj^athyand interest from
the wider public. In consequence the latter has so far had
little opportunity of learning to appreciate the great historical
problems which are involved in those researches. In the absence
of such prei)aratory information the non-Orientalist could not
be expected to form for himself a correct estimate of the im-
portance of the discoveries resulting from my explorations without
the guidance of expert opinion. I must therefore feel grateful that
the generous attention paid to my labours by the most
rei)re-
sentativc body of qualified fellow-scholars j)ermits me to supply
expert opinion in a clear and conclusive form.
SCIENTIFIC RESULTS OF JOURNEY. xiii
The International Congress of Orientalists, assembled at
Hamburg in September,
1902,
before which I was privileged to
give an account of my journey and excavations, adopted the
following resolution, proposed by Professor Henri Cordier, the
representative of the French Government, and Dr. A. A. Mac-
donell, Boden Professor of Sanskrit at Oxford, and recommended
by the combined Indian, Central-Asian, and Far-Eastern Sec-
tions
:

"
The Xlllth International Congress of Orientalists held at
Hamburg beg to express their thanks to His Excellency the Viceroy
and the Government of India for the great encouragement they
have extended to Oriental learning and research by granting to Dr.
M. A. Stein the necessary leisure and means for the prosecution of
his recent explorations in Ecistem Turkestan. They desire at the
same time to express their appreciation of the highly important
results which have rewarded the labours of the scholar selected by
the Government of India, and which represent an ample return
for the outlay incurred, owing to the practical nature of the opera-
tions conducted by him. They would also venture to express the
hope that faciUties will be given to him for completing the pub-
lication and elaboration of the results obtained, and that the Govern-
ment wiU be pleased to sanction any necessar\- extension for this
purpose of Dr. Stein's present deputation. Finally, they venture
to express the hope that, when circumstances permit, the interests
of archaeological research will be allowed to benefit by Dr. Stein's
special experience and previous knowledge, which are Ukely to
facilitate considerably the further explorations which it is desirable
should be entrusted to him in the interests of India."
As far as the space and the limited means of illustration avail-
able in this personal narrative would permit, I have endeavoured
to explain to my readers the significance of the mass of anti-
quarian materials brought to light by my excavationswhether
in the form of objects of ancient art and industry
;
or in those
hundreds of old manuscripts and documents which the desert
sand has preserv^ed in such surprising freshness
;
or finally in the
many curious observations I was able to make on the spot about
the conditions of every-day life, etc., once prevailing in those
sand-buried settlements. But of the great historical questions
xiv
INTKODUCTION.
which all these finds help to
illuminate, it was impossible to show
more than the bare outlines, and those only in glimpses. This
cannot be the place for their systematic discussion. But I
may at least indicate here the main directions in which those
discoveries are likely to open new vistas into obscure periods of
Central-Asian civilisation.
The early spread of Buddhist teaching and worship from India
into Central Asia, China and the Far East is probably the most
remarkable contribution made by India to the general develop-
ment of mankind. Chinese records had told us that Buddhism
reached the
"
Middle Kingdom
"
not directly from the land of
its birth, but through Central-Asian territories lying northward.
We also knew from the accounts left by the devoted Chinese
pilgrims who, from the fourth century a.d. onwards, had made
their way to the sacred Buddhist sites in India, that Sakyamuni's
creed still counted numerous followers in many of the barbarian
"
Western Kingdoms
"
they passed through. But these Chinese
travellers, best represented by the saintly
"
Master of the Law,"
Hiuen-Tsiang, our Indian Pausanias, had their eyes fixed on
subjects of spiritual interest, on holy places and wonder-working
shrines, on points of doctrine and monastic observance. Of the
many things of this world about which their observations would
have been of far greater interest for the historical student, they
have rarely chosen to inform us even within the sacred bounds of
India. Hence their brief notices of Central-Asian countries,
visited merely en route, fail to supply definite indications of the
extent to which Indian culture, language and art had spread with
Buddhist propaganda across the Himalaya and the Hindukush.
That such influences had been at work there for long centuries,
and sometimes penetrated even much further to the East, occa-
sional references in the Chinese Annals and elsewhere had led us to
suspect. But of those indigenous records and remains which
might enable us to reconstruct that bygone phase of civilisation
in its main asi)ects, all trace seemed to have vanished with the
Muhammadan conquest (tenth-eleventh century).
Chance finds of ancient manuscripts, in Sanskrit and mostly
Buddhistic, which commenced in
1890
with Captain (now
Colonel) Bower's famous birch-bark leaves from Kucha, were
BUDDHISM IN CENTRAL ASIA. xv
the first tangible proof that precious materials of this kind
might still be preserved under the arid soil of Chinese Turkestan,
The importance of these literary relics was great, apart from
their philological value
;
for they plainly showed that, together
with Buddhism, the study of the classical language of India also
found a home in that distant land beyond the Himalaya. But
on the cultural entourage in w^hich this far transplanted Indian
learning had flourished, such chance acquisitions, of uncertain
origin and unaccompanied by archaeological evidence, could
throw little light.
For systematic excavations, which alone could supply this
evidence, the region of Khotan appeared from the first a field of
particular promise. In scattered notices of Chinese records
there was much to suggest that this little kingdom, situated
on the important route that led from China to the Oxus Valley
and hence to India as well as to the West, had played a prominent
^art in developing the impulses received from India and trans-
mitting them eastwards. The close connection with ancient
Indian art seemed particularly marked in whatever of small
antiques, such as pottery fragments, coins and seals, native
agency had supplied from Khotan. And fortunately for our
researches, archaeology could here rely on the help of a very
effective allythe moving sand of the desert which preserves
what it buries. Ever since human activity first created the oases
of Khotan territory, their outskirts must have witnessed a
continuous struggle with that most formidable of deserts, the
Taklamakan ; while local traditions, attested from an early
date, told of settlements that had been abandoned before its
advance.
The ruined sites explored by me have more than justified the
hopes which led me to Khotan and into its desert. Scattered
over an area which in a straight Une extends for more than three
hundred miles from west and east, and dating back to very
different periods, these ruins throughout reveal to us a uniform
and well-defined ci\'ilisation. /It is easy to recognise now that
this bygone culture rested mainly on Indian foundations. But
there has also come to fight unmistakable evidence of other
powerful influences, both from the West and from China, which
XVI
INTEODUCTION.
helped to shape its growth and to invest it with an individual
character and fascination of its own^
The origin and history of the culture that once flourished in
Buddhist Khotan, are faithfully reflected in the remarkable
series of sculptures and paintings which the ancient shrines and
dwelling places, after long centuries of burial beneath the dunes,
have yielded up. Exact archaeological evidence enables us to
determine the various periods at which these settlements were
invaded by the desert sand. Though these periods range from
the third to the close of the eighth century of our era, yet the pre-
ponderance of Indian art influences is attested by the latest as
well as by the earliest of these finds. The rich statuary of the
Rawak Stupa Court, and the decorative wood carvings of the
ancient site beyond Niya, reproduce with astonishing fidelity
the style and motives of that fascinating
*
Graeco-Buddhist
'
art which, fostered by Hellenistic-Roman influences grew up
and flourished in Gandhara (the present Peshawar Valley) and
other neighbouring tracts in the extreme North-West of India,
during the centuries immediately preceding and following the
commencement of our era. Yet when we turn from those remains
to the frescoes on the walls of the small Buddhist shrines at
Dandan-Uiliq, dating some five hundred years later, we recog-
nise with equal distinctness the leading features of ancient
Indian pictorial art as preserved for us in the Ajanta Cave
paintings.
The records of the Chinese Annals ])lainly showed us that for
considerable periods under both the Later Han and the Tang
dynasties China had maintained effective political control over
the kingdom of Khotan. My excavations have confirmed
these records, and from the finds of Chinese documents on wood
or paper, Chinese coins, articles of manufacture, etc., it has
become abundantly clear that Chinese civilisation no less than
political ascendency asserted there a powerful influence. Seeing
how close for centuries were the relations between Khotan and
the great empire eastwards in matters of administration, trade
and industrial intercourse, we cannot feel surprised to find a
connection in art also attested by manifest traces. It is China
whicli in this direction appears the main borrower
;
for besides
ART OF ANCIENT KHOTAX. xvii
such distinct historical evidence as the notice about a scion of the
royal house of Khotan, whom the Annals name as the founder
of a new pictorial school in China in the seventh century a.d.,
there is much to suggest that the Indian element which so con-
spicuously pervades the whole Buddhist art of the Far East had
to a very large extent found its way thither through Khotan.
Yet a careful analysis of the composition and drawing in more
than one of the frescoes and painted panels of Dandan-Uiliq
will show that Chinese taste also had its influence on the later art
of Khotan.
For us still greater interest must attach to the convincing
evidence disclosed as to the question how far into Central Asia
the classical art of the West had penetrated during the first
centuries of our era. We see its triumphant advance to Khotan
half-way between Western Europe and Peking, strikingly demon-
strated by the remarkable series of classical seals, impressed on
clay and yet preserved in wonderful freshness, which still adhere
to a number of the many ancient documents on wood discovered
at the sand-buried site beyond Niya. As explained in Chapter
XXV., where I have discussed and illustrated some of these
important finds, we cannot make sure in each case where the
well-modelled figures of Greek deities, such as Pallas Athene and
Eros, or the classically treated portrait heads that appear in
these seals, were actually engraved.
But it is certain that the
seals themselves were currently used by officials and others
resident within the kingdom of Khotan, and that classical models
greatly influenced the work of local lapidaries and die-sinkers.
The remarkable diversity of the cultural influences which met
and mingled at Khotan during the third century a.d. is forcibly
brought home to us by these records from a remote Central-Asian
settlement, inscribed on wooden tablets in an Indian language and
writing and issued by officials
with strangely un-Indian titles,
whose seals carry us to the
classical world far away in the West.
The imitation of early Persian
art of which, five centuries
later, we find unmistakable
traces in some of the paintings of
sacred Buddhist subjects recovered from the ruins of Dandan-
UiUq, is a curious parallel, and from an historical point of view
almost equally instructive.
B
xviii
INTEODUCTION.
The dwelling places, shrines, etc.,\of those ancient settlements
had, no doubt, before the desert sand finally buried them, been
cleared by the last inhabitants and others of everything that
possessed intrinsic value. But much of what they left behind,
though it could never tempt the treasure-seekers of succeeding
ages, has acquired for us exceptional value^ The remains of
ancient furniture such as the wooden chair reproduced on
p. 356 ;
the shreds of silks and other woven fabrics
;
the tatters
of antique rugs
;
the fragments of glass, metal and pottery
ware
;
the broken pieces of domestic and agricultural imple-
ments, and the manifold other relics, however humble, which
had safely rested in the sand-buried dwellings and their deposits
of rubbishthese all help to bring vividly before our eyes details
of ancient civilisation that without the preserving force of the
desert would have been lost for ever.
But however interesting and instructive such details may be,
they would, by themselves, not permit us with any degree of
critical assurance to reconstruct the life and social organisation
which once flourished at these settlements, or to trace the
historical changes which they have witnessed. The hope of ever
elucidating such questions was dependent on the discovery of
written records, and it is fortunate indeed that, at the very sites
which proved richest in those relics of material culture, the finds
of ancient manuscripts and documents were also unexpectedly
ample and varied. The Sanskrit manuscripts excavated at
Dandan-Uiliq acquaint us with that class of canonical Buddhist
literature which we may assume to have been most cherished in
the monastic establishments of ancient Khotan. The series of
Chinese documents discovered in ruins of the same site is of
particular historical interest. The exact dates recorded in
them
(781-790
A.D.), in combination with other evidence, clearly
indicate the close of the eighth century as the time when the
settlement was deserted, while their contents throw curious
side-lights on the economical and political conditions
of the
territory immediately before Chinese suzerain
power finally
abandoned these regions to Tibetan invasion. Sanskrit manu-
scrij)ts and records in Chinese mark foreign imports in the culture
of Khotan. All the more interest attaches to the numerous
DISCOVERIES OF
ANCIENT RECORDS. xix
documents and fragmentary texts from the same site which show
an otherwise
unknown language, manifestly non-Sanskritic yet
written in Indian Brahmi characters
;
for it appears very pro-
bable that in them we have records of the tongue actually spoken
at that period by the indigenous population of Khotan.
We see Sanskrit, Chinese and the same non-Sanskritic language
similarly represented among the Uterary finds from the ruined
temple of Endere, in the extreme east of the territory explored.
But here in addition there appears Tibetan, as if to remind us of
the prominent part which Tibet too has played in the history of
Central Asia. A curious Chinese graffito found on the waU
of the Endere temple clearly refers to the Tibetans, and gives
a date which, since its recent examination by Sinologists, can
be safely read as
719
a.d. It is probable that these finds of
Tibetan manuscripts are directly connected with that extension
of Tibetan power into Eastern Turkestan which the Chinese
Annals record for that very period.
But much older cind of far greater importance than any of
these finds are the hundreds of Kharoshthi documents on wood
and leather brought to light from the ruined houses and the
rubbish heaps of the ancient settlement discovered beyond
the point where the Niya River now loses itself in the desert.
Their peculiar writing material (so much older than the paper
of my other Uterary finds), their early Indian script and language,
and the surprisingly perfect state of preservation of many among
them would alone have sufficed to invest these documents
with special interest. But their exceptional historical value
is derived from the fact that they prove to contain records
written as early as the third century of our era, and dealing
with a wide range of matters of administration and private
hfe.
In Chapter XXVI. I have endeavoured to indicate the varied
nature and abounding interest of the information which this
mass of official reports and orders, letters, accounts, and miscel-
laneous
"
papers
"
(to use an anachronism) is bound to reveal
to us. The results already obtained have opened new and
far-reaching vistas. It is no small discovery to find the old
local tradition of a colonisation of Khotan from the extreme
B 2
XX INTRODUCTION.
North-West of India confirmed by the use, in ordinary practical
intercourse, of an Indian language and a script peculiar to
the very region from which those Indian immigrants were
believed to have come.
The thought of the grave risks with which nature and, still
more, human activity threaten all these relics of antiquity, was
ever present to my mind, and formed an urgent incentive to
unwearied exertion, however trying the conditions of work
might be. On the one hand I had ample occasion in the
desert to observe the destructive
effect -i)f
erosion by wind]^
and sand on whatever of ancient remains is left exposed to its
slow but unrelenting power. On the other I could not fail
to be impressed by the warnings of impending destruction
through the hand of man : there were the evident traces of
the mischief done by Khotan
"
treasure-seekersjj at the more
accessible sites, and also, alas ! a vivid remembrance of the
irretrievable loss which-the study of Indian art and antiquities
has suffered through 'l^irresponsible digging
"
carried on until
recent years by, and for, amateur collectors among the ruined
Buddhist shrines of the North-West Frontier of India.
Though the climate of the Turkestan desert is not inferior in
conserving capacity to that of Egypt, yet neither Khotan nor
any other territory bordering on that desert could ever compare
with the land of the Pharaohs in wealth of antiquarian remains
awaiting exploration,
"
Ancient cities," complete with palaces,
streets, markets, etc., such as are pictured by Turkestan folklore,
and also by indiscriminating European imagination, as lying
submerged under the sand-dunes through a kind of Sodom
and Gomorrah catastrophe, are certainly not to be looked for.
The sites where settlements abandoned in early times could
be located, with ruins still capable of excavation, were few
in number, and even those among them which, being further
removed from the present inhabited area, had so far escaped
the ravages of the
"
treasure-seekers," could not be expected
to remain safe much longer. The time seems still distant
when Khotan will see its annual stream of tourists.
Yet the
extensive industry of forged
"
old books
"
which had grown
up in Khotan during recent years, and which I was able to
RISKS TO KHOTAN ANTIQUITIES. rri
trace and expose in detail (see Chapter XXXI.
),
sufficiently
shows how dangerous a factor "collecting" has already become
even in Chinese Turkestan.
In the face of such difficulties as work in the
Taklamakan
presents I could never have made my
explorations sufficiently
extensive and thorough without the active co-operation of the
Chinese administrators of the districts from which I had to
draw guides, labour, supplies

^in fact, whatever was needed

during my winter campaign in the desert. I had the good
fortune to find in the Ambans Pan-Darin and Huang-Daloi,
then in charge of Khotan and Keriya, reliable friends, thoroughly
interested in my work and ever ready to help me with all that
was in their power. I look back to the invariable kindness
and attention I received from these amiable Mandarins \viih
all the more gratitude as it was sho\vn at a time when, as they
well knew, the conffict with the European powers was con-
xiilsing their empire. They were fully aware, too, that the
services rendered to my scholarly enterprise could earn them
neither material advantages nor honom^.
The tnie historical sense innate in educated Chinese and the
legendary knowledge I found to prevail among them of[Hiuen-
Tsiang, the great Buddhist pilgrim, whom I claimed as my guide
and patron saint, certainly helped me in explaining the objects
of my explorations to my Chinese friends and enlisting their
personal interest. J But I cannot doubt that the sympathetic
attitude adopted from the first by the pro\Tncial administration
towards my work was directly due to the efforts made on my
behalf by Mr. G. Macartney, c.i.E., the representative of the
Indian Government at Kashgar, whose personal influence
among all Chinese dignitaries of the province is as great as it
is well deserved. My narrative shows the manifold benefits
I derived from the unfailing care of this kind and accompUshed
friend, who from afar never ceased to follow my
explorations
with watchful interest. For the important help he thus rendered
towards their success, and for all his personal kindness, I am
anxious to record here the expression of my sincere gratitude.
The resume given above of the aims and results of my archaeo-
logical work will, I hope, help to account for the character of my
xxii INTKODUCTION.
present narrative and the labour involved in its preparation.
The interests of science obliged me to concentrate my efforts
on a series of well-defined tasks and to avoid whatever might
interfere with their carefully prepared execution. Mine was
not a journey leaving much range for those chance incidents
which may at times lead to exciting personal experiences, but
are far more likely to cause loss in substantial results through
waste of time, energy and means. I can only hope that my
book may reach readers ready to find compensation in the
thought that long-continued study of the ancient East and
familiarity with modern India and its northern borderlands
permit me to offer them guidance in regard to much that is
of general human interest both in the present conditions and the
historical past of the regions traversed.
The critical standards to which I am pledged by my work as a
scholar would not allow me to compile a narrative by the mere
reproduction of those diary leaves which were intended to
convey the first records of my personal experiences and im-
pressions to dear eyes since closed for ever by Death. Though
my account was intended for a wider public than that of
Orientalist or antiquarian scholars, yet I felt it incumbent
to take every care^that it should neither contain statements
which further scrutiny might require to be modified in my
scientific Report, nor pass over unnoticed any essential facts
connected with my archaeological discoveries.
The preparation of my narrative on these lines has implied far
more labour than may, perhaps, appear on the surface. It
would, in fact, have been impossible to accomplish it with the
scanty leisure left from official duties as Inspector of Schools
in the Punjab, to which I had to return on the conclusion of
my explorations, in the autumn of
1901.
Fortunately, how-
ever, the Government of India, on the proposal of the Punjab
Government and with the concurrence of the Secretary of State
for India, granted to me in the following year a period of
deputation to England in order that I might be enabled to
elaborate the results of my journey with the help of the original
finds temporarily deposited in the British Museum.
For the generous consideration thus shown to me I feel it my
PREPARATION OF PERSONAL NARRATIVE, xxiii
duty to record here my deep sense of gratitude to His Excellency
the Viceroy and the Indian Government.
Just
as my explora-
tions were rendered possible only through their powerful aid,
so, too, I owe to their liberality the temporary freedom for
scholarly labour which has permitted me to complete the present
narrative. I feel confident that its contents will be found
in more than one respect a necessary complement to my Detailed
scientific Report which is still under preparation. On the other
hand, I must refer my readers to the latter publication for
many illustrations of antiquities, ruins, scenery, etc., which
to my regret it was found impossible, on account of technical
difficulties and other reasons, to provide here.
It remains for me to record my grateful acknowledgments
for the manifold assistance which I have received while preparing
this volume. To none do I feel more indebted than to my
artist friend, Mr. Fred. H. Andrews, who ever since my return
from Chinese Turkestan has furthered my labours with
enthusiastic devotion. His wide knowledge of ancient Indian
art, acquired in his late post as Principal of the School of Art
and Curator of the Museum at Lahore, and his own high artistic
abilities, have rendered his co-operation in the arrangement
and description of my collection of antiquities of the utmost
value. He has never wearied in giving me the full benefit
of his expert advice in questions affecting the technical aspects
of my finds, and he has spared no trouble to make the illustrations
of this book as effective as their number and the available means
of reproduction would permit.
Beside drawings and diagrams embodied in these pages I owe
to his skill the design reproduced on the cover of this volume
and the Black and White drawing for the Vignette which adorns
the title-page. This represents a restored yet faithfully con-
ceived enlargement of the figure of Pallas Athene as seen in
several of the ancient seal impressions on clay excavated by
me from the desert sand. I could scarcely have wished for my
narrative to issue under a more felicitous emblem.
Dr. A. F. Rudolf Hoemle, the eminent Indologist, who from
the first had shown the warmest interest in my explorations,
was kind enough to place at my disposal valuable information
xxiv
INTKODUCTION.
in respect of the ancient manuscripts in Brahmi characters,
the pubhcation of which has been undertaken by him
;
he
has further rendered me the great service of reading a revision
of this book. I owe a similar debt of gratitude to my friend
Mr. E.
J.
Rapson, of the British Museum, who not only charged
himself with the care of my collection while I was absent in
India, but has also allowed me to benefit at all times by the
results of the most painstaking researches he has devoted to the
decipherment of the ancient Kharoshthi documents. To Dr.
Percy Gardner, Professor of Archaeology in the University
of Oxford, I am indebted for most competent guidance in respect
of the objects of classical art contained in my collection, and for
much kind encouragement besides.
For the interpretation of my important Chinese records I must
consider myself particularly fortunate in having enjoyed the
assistance of such distinguished Sinologist experts as Dr. S. W.
Bushell, C.M.G., and Professors
E.
Chavannes and Douglas.
The complete translation and analysis of those documents
with which Professor Chavannes, of the College de France,
has favoured me for publication in my Detailed Report, has
already proved of very great value for the study of Chinese
influence in Turkestan. Dr.- Bushell and Professor Douglas,
of the British Museum, have never failed to help me with learned
advice on questions concerning Chinese lore.
If I have left it to the last to mention my obligations to my
friends Mr.
J.
S. Cotton, late editor of the Academy, and Mr.-
P. S. Allen, of Corpus Christi College, Oxford, it is only because
theirs was the help benefiting more directly the Western or
modern aspect of the work now presented. The former did me
the great favour of revising my manuscript with special regard
to the requirements of the general reader, a task for which he
was exceptionally qualified by his literary experience
;
while
the other kind friend cheerfully charged himself with a revision
of my jiroofs and greatly helped me by its thoroughness. To
his kind offices and the generous mediation of Mr. Cuthbert
Shields, I owed the jxiaccful retreat for scholarly work which
the hospitality of the President and Fellows of Corpus Christi
College assured to me during the summer of 1902.
With those
ACKNOWLEDGMENT OF FRIF:NT)LY HELP, xxv
inspiriting precincts, full of great memories from Erasmus to
Ruskin, I shall always associate the recollection of the pleasantest
part of my work in England.
********
The narrative here presented still leaves me far from the
conclusion of the labours which the antiquities and obsen'ations
brought back from Chinese Turkestan have entailed upon me.
Yet even thus I cannot prevent my eyes from looking beyond
towards other fields of archjeological exploration, no less closely
linked with the sphere of Indian historical interests and equally
likely to j^ield a rich harvest. On some my thoughts had been
fixed long before I was able to \-isit India
;
but the years which
have since passed by, though as full of scholarly labours as other
duties would permit, have seemingly not brought me nearer
to the longed-for chance of exploring them.
Life seems short where the range for research is so vast as in
the case of ancient India and the regions through which it com-
municated with the classical West, But life must appear shorter
still when the chosen tasks cannot be done in the study, when
they call for the exertions of the scholar and explorer combined,
such as are readily faced only while the optimism of com-
parative youth and physical vigour endures. To Fateand
to those who dispense itI offer due thanks for having allowed
me to work on Indian ground and at Icist, after years of toil,
to attain for a time freedom and the means to serv'e science.
Yet when I look back upon all the efforts that had to precede
this opportunity, I am tempted to regret that I cannot share
the Indian belief in those
'
future births ' which hold out
promise of appropriate reward for
'
merits,' spiritual and
other. For on the strength of such a belief I might feel more
hopeful of meeting yet with that reward for my work at Khotan
which I should prize highestthe chance of repeating it else-
where.
M. AuREL Stein.
British Museum,
April i6,
1903.
CONTENTS
INTRODUCTION.
PAGE

First plan of explorationsSanction and assistance of Indian Govern-

mentEstimates of time and expenseHelp of Survey of India
DepartmentPrevious obstacles to circhaeological laboursUn-
certainties of enterpriseValue of scientific resultsResolution
of International Oriental CongressBuddhism in Central Asia

Antiquarian acquisitions from Chinese TurkestanIndian in-

fluences at KhotanAncient art of KhotanCultural connection
with ChinaRelics of classical artInterchange of ancient
civilisationsDiscoveries of Sanskrit, Chinese, Tibetan manu-
scriptsKharoshthi documents on woodRuins threatened by
"
treaisure-seekers
"
Help from Chinese ofl&cialsPurpose of
Personal NarrativeConsideration shown by Indian Government
Acknowledgment of scholarly assistanceProspects of future
archaeological work ....... viixxv
CHAPTER I.
CALCUTTA TO KASHMIR.
On Mohand MargPrevious antiqucirian toursStart from Calcutta

Up
the Jhelam ValleyArrival in KashmirPreparations at
SrinagarThe Kashmir ChronicleCamp on Sind RiverThe
Bagh of Buchvor

Joined by Sub-SurveyorFormation of camp

Start from Srinagar . . ...... 1-9
CHAPTER II.
TO ASTOR AND GILGIT.
Start on Gilgit routeA
"
Master of the Gates
"
Over theTragbal
PassDescent over snow-bridgesValley of the
"
Black Ganga
"
xxviii CONTENTS.
PAGE
Among the DardsHalt at GurezIn the Burzil Valley

MinimargCrossing of Burzil PassBelated Commissariat

Views of Nanga-Parbat PeaksThe Astor capitalFirst sight

of IndusHalt at DuyanThe British BabyDown to the
IndusA night ride from BunjiAt the Gilgit AgencyLast
Indian
"
Station
"
. . . . . . . . 10-26
CHAPTER III.
THROUGH HUNZA.
Start from GilgitIn the old Dogra ServiceGlories of Mount
RakiposhiThe storming of NilthA historic gorgeOld
adversariesStupa of TholLanguage of HunzaHalt at
AliabadOld raiding daysRelations with ChinaThe Hunza
LeviesVisit to Mir's CastleFirst march in Hunza Gorge

Ghammesar landslipClimbs over

'
Rafiks
'
GhulmitWakhi
settlementsGlaciers near PasuCrossing of Batur Glacier

'
Darband
'
of KhaibarMore alpine climbsMarching of
KanjutisA polyglot campClimbs to MisgarHunza hillmen
dischargedA Celestial soldierYaks from the Pamir

Among Sarikoli herdsmen ....... 27-53

CHAPTER IV.
ON THE TAGHDUMBASH PAMIR.
Crossing of Kilik PassCamp at Kok-torokSurvey work com-
mencedWatershed towards OxusCrossing of Wakhjir Pass

Oxus Source GlaciersOn Afghan soilA vista into Wakhan

First news of Chinese troublesMarch down the Taghdumbash
PamirChance meeting with German ofiicerKirghiz and
Wakhi settlersOver pleasant grazing-groundsRide to Tash-
kurghanA difficult fording
54-66
CHAPTER V.
IN SARIKOL.
Tash-kurghan, Ptolemy's
"
Stone Tower
"
On the track of Hiuen-
TsiangRuins of old townDiplomatic surveyingChinese
garrisonMeeting with Sarikoli headmenApplicants for
"
Tabloids
"
Requisitions for kitchenPlain of Tagharma

Muztagh'Ata sightedMarch along Russian PamirA Kirghiz

shrine

At Karasu postCrossing of Ulugh-Rabat PassOld

friends, the marmotsChinese at Subashi post

Arrival at
Little Kara-kul
67-78
CONTENTS.
xx
CHAPTER
VI.
ON MUZTAGH-ATA-
PAGE
Lakes of L. Kara-kul and Basik-kulDay of alpine rainApprehen-
sions of Kirghiz BegKirghiz hospitalityA
hitch about
transportPhoto-theodolite surveyStart for Muztagh-Ata

Reconnoitring the great peaksHeavy mantle of snow and ice

Along Yarabulak GlacierTrouble with YaksPrehminarj- climb

Reconnaissance of Hunza LeviesWaiting for fair weather

Start for higher slopesClimb in deep snowMountain-sickness

of followersHighest point reachedView over
"
Roof of the
World
"
Plucky Hunza guidesFreebooting VisionsDescent
to Yambulak Glacier CampTriangulation from Shamalda spur
Return to Kara-kul 79-9*
CHAPTER Vn.
THROUGH THE GEZ DEFILE TO KASHGAR.
Departure from Kara-kulTarns of Basik-kulObstruction at
BulunkulA pUable interpreterEntrance of Gez Defile

Chinese engineeringCrossing Koksel GlacierAt Gez Karaul

Absconding of KirghizOpportune rehefOver the

"
Nine
Passes
"
Through tortuous gorgesSohtary springDistant
view of plainsScrambles over decomposed rangesArrival
in Tashmalik plainPassage of Yamanyar R.Through the Opal
oasisFirst impressions of rural TurkestanRide into Kashgar
99-112
CHAPTER Vni.
ST-\Y AT KASHGAR.
Mr. Macartney's hospitable roofStay at Chini-BaghOrganisation
of caravanPurchase of camels and poniesEngagement of
Turki followersThe Kashgar craftsmenWater-tanks for the
desertCongenial studiesVisits to Chinese officicdsAppeal to
Hiuen-Tsiang's memoryNews of fighting at PekingDangers
of local troublesLife at Chini-BaghA polyglot clientele

Picnics of KashgarisThe Russian ConsulateRemains of

[
StupasVisit to Liu-Kin-tang's templeChinese historical
paintingsIn the Chinese cantonmentVisit to an artist-official
113-129
XXX CONTENTS.
CHAPTER IX.
KHANUI AND ORDAM-PADSHAH.
PAGE
First start from KashgarVia Appia of KashgarCollation at Besh-
karimShrine of holy MairyamCamp at KhanuiSite of
Hasa-TamRuins of Topa TimThe Mauri Tim Stupa
'
The
pigeon house
"
ruinDeparture from KashgarRide to Khana-
rikHindu moneylenders March to AchchikOn edge of desert
First crossing of sand-dunesArrival at Ordam-Padshah

Shrines in the desert

' Yolchi Beg

'
on a camelTomb of
Hazrat-BegimOasis of KizilA dreary caravan-route . 130-149
CHAPTER X.
YARKAND AND KARGHALIK.
Reclamation of desert groundEntry into YarkandPalatial
quartersHalt at YarkandCosmopolitan visitorsStrange
mixture of racesOld Turkestan art-wareInterviews with Liu-
DarinChinese dinner-partyCurrency complicationsDepar-
ture from YarkandCrossing of Yarkand RiverHalt at Kar-
ghalikVisit of Karghalik AmbanBazars of Karghalik

Buddhist monk from China . . . . . . 150-166

CHAPTER XL
ON THE ROAD TO KHOTAN.
Ancient desert routeOn the track of Marco PoloAn exile in the
desertIn Guma oasisEnquiries after alleged
"
old books
"

Karakul MazarGlimpse of Karakorum RangeStupa of Topa

TimD6bris-strewn 'Tatis'Eroded ancient sitesAn evening
in weird desolationOld remains at MojiOases of Zanguya
and PialmaComfortable homesteads .... 167-178
CHAPTER XII.
ARRIVAL IN KHOTAN.
"
The Pigeons' sanctuary
"
Legend of sacred ratsSurvival of
Buddhist local cultRural environsEntry into Khotantown

Camp in residential gardensFirst meeting with Pan-Darin

Despatch of
"
prospecting
"
partiesPreparing for the moun-
tainsInteresting geographical taskForged birch-bark manu-
scriptSuspected forgeries
179-189
CONTENTS.
xixi
CHAPTER XIII.
TO THE HEADWATERS OF THE YURUNG-KASH.
PAGE
Start for the mountainsDebouchure of Yuning-kash R.Through
the outer rangesCrossing of Ulugh-DawanTrying march to
BuyaIn the Pisha ValleyA centenarian hillman

" Kuen-
luen Peak No.
5
"
A grand panoramaPrecipitous descent

Arrival in Karanghu-taghA penal settlementA gloomy vale

Start from Karanghu-taghMarch to Yurung-kash Gorge

Hot springWild river-bedAttempt to penetrate gorge

Forced to turn backClimbing on Yaks .... 190-207

CHAPTER XIV.
OVER THE KARA-KASH RANGES.
A terra incognitaA hidden mountain trackCrossing of Pom-tagh
PassIn the Nissa ValleyMythic Xysa of DionysusDifficult
tent-pitchingSurvey above Brinjak PassPanorama of glaciers
A trying descent^^ombination of ice and dustDistrustful
hillmenView from Yagan DawanMaze of eroded ridges

Through fantastic gorgesWant of waterMarch in Mitaz

ValleyClimb to Ulughat-DawanExtensive panorama

Identification of ice-peaksConnection with Indian Surveys

Waiting for waterNight on Ulughat-DawanMoonlight over

desert plainsA magic cityTriangulation of KhotanDescent
to Kara-kash ValleyReturn to arid gorgesClimb of Kauruk-
kuzSuccessful triangulation ...... 208-225
CHAPTER XV.
ANTIQUARIAN PREPARATIONS AT KHOTAN.
Visit to Mount GosringaLegend of hidden ArhatIn a sacred cave
Easy-going MuhammadansReturn to Khotan townAn-
tiques brought by TurdiVisits of Pan-DarinThe blessing of
'
Tang-Seng
'
Camp in Akhun Beg's gardenMedical func-
tionsRana Singh's departureRemains of Chalmakazan

Among the jade-pitsSpeculative jade-miningFame of KhotEin

jade
226-236
CHAPTER XVI.
YOTKAN, THE SITE OF THE ANCIENT CAPITAL.
Camels stuck in
'
Yars
'
First discovery of Yotkan siteWashing
for goldAntiques as secondary productsAncient coins and
potteryCulture-strata of YotkanSilt over culture-strata

xxxii CONTENTS.
Alluvial deposits on irrigated groundRising of ground level

Site of ancient capitalPosition of Sa-mo-joh ConventLinger-

ing local worshipIdentification of Buddhist shrinesThe
marsh of HalalbaghA learned MullahAbu Bakr's excavations
Preparations for winter ....... 237-250
CHAPTER XVII.
TO THE RUINS OF DANDAN-UILIQ.
Start for desert campaignAt Tawakkel oasisRecruiting of diggers
Hunters as guidesPreparations at TawakkelFailure of
local dentistryStart into desertCamping in wintry desert

Arctic clothingTurdi guides through sand-dunesArrival at

Dandan-UiliqFirst survey of ruinsFuel from ancient or-
chards .......... 251-261
CHAPTER XVIII.
EXCAVATION OF BUDDHIST SHRINES.
Stucco-relievos of ruined shrineExcavation of temple-cellasCon-
struction of wallsDecoration of cellasBuddhist frescoes

Sculptures in chapelPainted panelsAncient broomsIn-

teresting relievosScenes represented in frescoesLegend of
Naga ladyPicture of Buddhist scholars .... 262-275
CHAPTER XIX.
FIRST FINDS OF ANCIENT MANUSCRIPTS,
Excavation of monastic dwellingFirst leaf unearthedSanskrit
manuscript findsDiscovery of Sanskrit
'
Pothi
'
Buddhist
canonical textCook-room of monastic dwellingReturn of Ram
SinghRendezvous in desertAccuracy of surveys

" The
Aksakal of the Taklamakan
"
Turdi's old ponyA failed specu-
lationAn antique fodder storeEnd of ill-fated animal . 276-287
CHAPTER XX.
DISCOVERY OF DATED DOCUMENTS.
Finds in ruined monasteryPicture of rat-kingDocuments in
unknown languageProbable contentsFinds of Chinese re-
cordsAn ancient petitionAncient name of Dandan-Uiliq

Christmas Day in desertLost among dunes

' Yolchi Beg,'

CONTENTS.
xxxiii
PAGE
fox-terrierThe Hu-kuo monasteryIts Chinese documents

Date of its abandonmentMore painted panelsAncient

"
horse-millinerv
"
Persian art influenceBuddhLst pictorial art
Traces of old cultivationAbandonment of settlementPro-
bable cause of abandonment 288-305
CHAPTER XXI.
THROUGH THE DESERT TO KERIYA.
Departure from Dandan-UiliqRemains of RawakDismissal of
Tawakkel labourersStart for Keriya DaryaFormidable sand-
dunesCrossing of
'
Dawans
'
Arrival at frozen riverDesert
shrine of Burhanuddin

Jungle of Keriya R.Welcome at

KeriyaThe Amban of KeriyaVisits of stateHalt in Keriya
town
306-318
CHAPTER XXn.
TO NIYA AND IMAM JAFAR SADIK.
View of Kuen-luenYesyulghun and OvrazOasis of NiyaRam-
zan festivalA promising findFirst Kharoshthi tabletsMarch
along Niya RiverReclamation of jungle groundWinter
atmosphere of desertThrough riverine jungleShrine of
Imam Jafar SadikExhibition of textile ex-votosTomb on
sacred hillStart for ancient siteTransport of iceW'here the
Niya R. loses itselfThrough dead forestAncient houses
sightedArrival at ruins ....... 319-333
CHAPTER XXni.
FIRST EXCAV.\TIOX OF KHAROSHTHI TABLETS.
Search of first find-placeIbrahim's chance discoveryAn abundant
haulVarieties of inscribed wooden tabletsT\-pcof Kharoshthi
writingFirst deciphermentLanguage of recordsClearing of
roomsMore documents discoveredWedge-shaped and oblong
tabletsDated recordsErosive power of windStructures
attacked^y erosion ........
334-348
CHAPTER XXIV.
EXCAVATION OF ANCIENT RESIDENCES.
Decayed records on wood.\ncient ice-pitLarge dwelling-houses
An old reception-hallExcavations in deep sandRemains of
c
xxxiv
CONTENTS.
PAGE
coloured rugBroken arms from storage roomAncient carved
chairAntique furniturePlan of an ancient gardenDead
poplars and fruit treesTroublesome followersNiaz Akhun's
truculent propensities

Quarrel with camel-manAffray in

desert camp

" The evil spirits of the Desert

"
. . . 349-364
CHAPTER XXV.
DISCOVERIES IN AN ANTIQUE RUBBISH HEAP.
Discovery of ancient rubbish layersSurvey of Stupa ruinAncient
"
waste-paper
"
depositsClearing of consolidated refuse

Antique microbesKharoshthi documents on leatherTechni-

calities of wooden stationeryFastening of ancient envelopes

Classical seals in clayPallets Athene and ErosSymbols of

clcissical influence ........ 365-377
CHAPTER XXVI.
DECIPHERMENT OF ANCIENT DOCUMENTS ON WOOD
AND LEATHER.
Prakrit language of Kharoshthi recordsAncient official corre-
spondenceTitles of officialsSanskrit introductionsOld
names of KhotanTradition of Indian immigrationKhotan
colonized from TaxilaUnique tablet in BrahmiChronological
evidenceChinese dated recordCommercial relations with
ChinaKelics of ancient industryArchitectural wood-carving
378-387
CHAPTER XXVH.
THE RUINS OF ENDERE.
Return to Imam
Jafar SadikTo the Yartungaz RiverA forlorn
colonyVagaries of Yartungaz RiverThrough the desert to
Endere RiverArrival at Endere StupaA successful concen-
trationAncient circumvallationExcavation of temple

Manuscript findsLeaves in unknown languageTibetan manu-

scriptsOldest known Tibetan writingRecords of Tibetan in-
vasionDate in Chinese graffitoAncient ex-votes of rags

Remains of ancient rampartsSurvey of Stupa . .

"*
. 388-401
CHAPTER XXVni.
EXPEDITION TO KARADONG RUINS.
Return towards NiyaOn the desert track to Cherchen

" Home
mails
"
Forced marches to KeriyaHelp of AmbanStart for
CONTENTS.
XXXV
PAGE
KaradongAlong the Kenya RiverThe shepherds of the
riverine jungleGuides from Tonguz-basteThe first sand-
stormArrival at KaradongA legend of Hiuen-Tsiang

Ancient fortified postExcavation of ruinsFinds of ancient

cerealsReturn of
'
Buran
'
Sad news 402-412
CHAPTER XXIX.
THE SEARCH FOR HIUEN-TSIANG'S PI-MO.
March back along Keriya RiverThrough the Shivul swampsIn
new cultivationDifficulty about guidesDeserted village sites
Shifts of irrigated areaLegend of Ho-lo-lo-kiaAn Odyssey
amidst the dunesRemains at Uzun-tatiLachin-Ata Mazar

The oasis of GulakhmaFarewell visit to KeriyaOases of

Chira and Sampula

' Tati
'
of HanguyaReturn to Khotan
environs
.......... 413-423
CHAPTER XXX.
AK-SIPIL AND THE SCULPTURES OF THE RAWAk STLPA.
Halt at Yurung-kash

" Culture-strata
"
of Tam-6ghilMarch to
Ak-sipilRemains of ancient fortSculpture from Kighillik
A
huge refuse-heapDiscovery of Rawak StupaSuccession of
sand-stormsTrying heat and glarePlan of StupaExcava-
tion in Stupa courtClearing of colossal statuesThreatened
collapse of imagesRisks of excavationWealth of statuary

Interesting relievos

" Guardians of the Gates "A quaint

ex-votoAffinitv to Graeco-Buddhist artNumismatic finds-
Date of StupaRemoval of relievosBurying of sculptures . 424-446
CHAPTER XXXI.
ISLAM AKHUN AND HIS FORGERIES.
Return to Khotan

Quarters at Nar-BaghInterviews with Pan-

DarinPurchases of
"
old books
"
Suspicions about Islam
AkhunArrest of Islam AkhunHis previous impostures

Queer papers seized on himAn improvised

"
Cutcherj-
"

Cross-examination of forgerConvicted by his own statements

Islam Akhun's admissionsAssociates in the factoryManu-
facture of
"
old books
"
Methods of productionThe forger's
confessionsWit and humour of Islam Akhun . . . 447-459
xxxvi CONTENTS.
CHAPTER XXXII.
LAST DAYS IN KHOTAN OASIS.
PAGE
Farewell to Pan-DarinDeparture from Khotan town

" Tips
"
in
TurkestanLast visit to YotkanPetty trade of oasisForeign
coloniesVisit to Kara-kash townSite of Kara-dobeLeave-
taking of TurdiNiaz Akhun's matrimonial entanglement

Farewell to Khotan friendsAii offering at the

"
Pigeons'
Shrine " 460-467
CHAPTER XXXIII.
FROM KHOTAN TO LONDON.
Rapid marches to YarkandReturn to KashgarDemobilisation of
caravanAssistance from Russian Consul-GeneralAmong old
friendsPacking of antiquitiesFarewell to faithful companions
Start for OshOver the Alai PassesDown the Gulcha
ValleyWelcome at OshIn Russian TurkestanThe Andijan
BazarsMoghul monumentsFrom Samarkand to the Caspian
Arrival in LondonTemporary work on collectionConclu-
sion .......... 468-480
INDEX . .
481-503
LIST OF ILLUSTRATIONS.
SEATED BUDDHA, BETWEEN TORSOS OF COLOSSAL STATUES,
STUPA. .....
MOHAND MARG, KASHMIR
ANCIENT TEMPLE AT PANDRENTHAN, KASHMIR
VIEW IN BLRZIL VALLEY
MIR'S CASTLE AT BALTIT
CLIFFS OF NILTH GORGE, NAGIR .
STUPA OF THOL, NAGIR ....
MOU.NT RAKIPOSHI, SEEN FROM ALIABAD .
HUNZA COOLIES, BEFORE START FROM ALIABAD
FORT-VILLAGE OF ALTIT
RAFIK ABOVE .ATAABAD
WAKHI VILLAGERS, GHULMIT .
VIEW TO NORTH-EAST OF P.ASU VILLAGE .
BATUR GLACIER, SEEN FROM SOUTH-EAST .
HUNZA VALLEY BELOW KHAIBAR .
RAFIK NEAR MURKHUN
KANJUTIS CARRYING MERCHANDISE
KANJUTI HILLMEN, DISCHARGED AT MISGAR
YAKS STARTING FOR KILIK PASS .
KILIK PASS, SEEN FROM KHUSHBEL
SNOWY RANGE SOUTH OF HEAD OF AB-I-PANJA VALLEY
PHOTO-THEODOLITE VIEW OF OXUS SOURCE GLACIERS
VIEW DOWN AB-I-PANJA VALLEY FROM NEAR WAKHJIR PASS
KIRGHIZ
'
AK-UIS
'
AT TIGH.\RMAN-SU
WAKHIS AND KIRGHIZ AT DAFDAR
CHINESE FORT WITHIN RUINED TOWN OF TASHKURGHAN
IN THE CEMETERY OF TIZNAF ....
CHINESE GARRISON OF SUBASHI ....
MUZTAGH-.ATA PEAKS, SEEN FROM ABOVE YAMBUL.\K VALLEY
ICY RANGE, WITH SARGULUK PEAK, TO NORTH-EAST OF K
LAKE .........
MUZTAGH-ATA PEAKS, SEEN FROM C.^^MP SOUTH OF KARAKUL
RAWAK
Frontispiece
PAGE
I
9
lO
27
31
n
35
i7
39
41
43
44
45
46
47
49
51
54
55
57
58
59
6x
63
67
71
77
79
83
8s
XXXVlll LIST OF ILLUSTEATIONS.
NEAR KARGHALIK
VIEW FROM ABOVE YAMBULAK GLACIER LOOKING WESTWARDS.
ICY RANGE WITH PEAKS ABOVE KONGUR-DEBE AND KOKSEL GLACIERS
START FOR GEZ DEFILE . . . . . .
ASCENT OF SHAGILDIK DAWAN ......
ROAD TO MR. MACARTNEY'S HOUSE, WITH CITY WALL, KASHGAR
PRIEST IN LIU-KIN-TANG'S SHRINE .....
IN THE BAZAR OF THE "NEW CITY, KASHGAR
"
MY SERVANTS FROM KASHGAR AND YARKAND
CARAVAN STARTING FROM KASHGAR
BEGS AND AKSAKAL OF BESHKARIM
RUINED STUPA OF MAURI TIM
HINDU MONEYLENDERS
pilgrims' sarai at ORDAM-PADSHAH
entrance to the yamen, yarkand
BADAKHSHANI TRADER, YARKAND .
LIU-DARIN, AMBAN OF YARKAND .
YETIMLUKUM MAZAR, WITH CEMETERY,
BUDDHIST MONK FROM CHINA
MENDICANT, OR
'
DIWANA
'
HOUSE OF TOKHTA AKHUN, KHOTAN
PAN-DARIN, AMBAN OF KHOTAN, WITH PERSONAL ATTENDANTS
MUZTAGH PEAK, IN KUEN-LUEN RANGE .....
TAGHLIKS AND EXILED CRIMINALS AT KARANGHU-TAGH .
VIEW UP THE YURUNG-KASH GORGE, WITH SPURS OF PEAK K;
5
ON LEFT 202
YAKS CARRYING BAGGAGE IN YURUNG-KASH GORGE, NEAR KARANGHU-
TAGH ........... 206
KUEN-LUEN RANGE, WITH GLACIERS OF NISSA VALLEY, SEEN FROM BRIN-
JAK ........... 2X2
ERODED RANGES TO NORTH-WEST, SEEN FROM ABOVE YAGAN-DAWAN 2X6
TURDI,
"
TREASURE-SEEKER "....... 23O
KHOTANESE WAITING FOR MEDICINES . . . . . . 232
JADE-PIT WITH DIGGERS, NEAR DEBOUCHURE OF YURUNG-KASH . 235
NORTH-WEST CORNER OF EXCAVATED AREA AT YOTKAN, WITH ENTRANCE
TO
'
YAR '..........
239
ANTIQUES FROM VOTKAN . . . . . -24'
TERRA-COTTA FIGURINES FROM YOTKAN ..... 242
OLD VILLAGERS OF SOMIYA ........ 247
CAMELS STARTING FOR DANDAN-UILIQ . . . . -251
AHMAD MERGHEN AND KASIM AKHUN, OF TAWAKKEL .
253
TAWAKKEL LABOURERS TAKEN TO DANDAN-UILIQ ....
255
CAMP IN THE DESERT, DANDAN-UILIQ ......
262
RUINS OF BUDDHIST SHRINE, D. II., AT DANDAN-UILIQ, BEFORE EXCA-
VATION
.......... 264
CELLA OF BUDDHIST SHRINE, D. II., AT DANDAN-UILIQ, AFTER EXCAVA-
TION
........... 266
PAGE
90
97
99
107
I II
126
127
128
130
132
137
141
146
150
154
IS8
163
166
172
183
185
190
199
LIST OF
ILLUSTRATIONS.
XXXIX
FRESCO FROM OUTER WALL OF SHRINE, D. II., DANDAN-UILIQ
SMALLER CELLA OF BUDDHIST SHRINE, D. II., AT DANDAN-UILIQ. AFTER
EXCAVATION
.......-
ROOM OF MONASTIC DWELLING, D. III., DANDAN-UILIQ, FIND-PL.\CE
OF ANCIENT MANUSCRIPTS
......
LEAF OF BR.\HMI MANUSCRIPT, IN NON-INDIAN LANGUAGE. FROM
MONASTIC DWELLING, D. III.
......
OBVERSE PORTION OF LEAF, OF BUDDHIST TEXT IN SANSKRIT
(VAJRACCHEDIKA), from D. III. .....
ANCIENT
'
TAKHTA
'
FOR WRITING
......
CHINESE DOCUMENT (D. VII.
2),
CONTAINING BOND, FROM DANDAN-UILIQ
D.ATED A.D. 782
........
CHINESE WOODEN TABLET, N. XV.
315
STREET IN SUBURB OF KERIYA ......
HUANG-DALOI, AMBAN OF KERIYA ......
VILLAGE BOYS AT NIYA .......
TREES WITH EX-VOTOS, ON PATH TO IMAM JAFAR
SADIK'S TOMB
RUINS OF ANCIENT DWELLING-HOUSE (N. III.), WITH GARDEN-
RUINED BUILDING (N. I.), FIRST FIND-PL.\CE OF INSCRIBED T.ABLETS
AFTER EXCAVATION .......
KHAROSHTHI DOCUMENTS ON WOODEN TABLETS
KHAROSHTHI DOCUMENT ON WOOD, UNDER-TABLET (N. IV.
1
39)
ANCIENT WOODEN PEN, WITH BONE KNOB (FROM N. XV.)
SPECIMENS OF ARCHITECTURAL WOOD-CARVING, FROM RUINED DWEL
LING HOUSE (N. VIII.) .......
REMAINS OF ANCIENT HOUSES, AT GROUP OF RUINS N. II. .
PORTION OF ANCIENT DWELLING-HOUSE (N. III.), BEFORE EXCAVATION
NORTH WALL OF CENTRAL HALL OF ANCIENT DWELLING-HOUSE (N
III.), DURING EXCAVATION ......
ANCIENT WOODEN CH.AIR, FROM RUINED DWELLING-HOUSE N. III. .
ANCIENT HOUSEHOLD IMPLEMENTS, ETC., MAINLY FROM RUINED
DWELLING N. IV. .
PL.AN OF .'ANCIENT DWELLING-HOUSE N. IV. ....
CLAY IMPRESSIONS OF CLASSICAL SEALS, FROM KHAROSHTHI T.^BLETS
RUINED STUP.A, AT .ANCIENT SITE BEYOND IMAM
JAF.AR SADIK
RUINED DWELLING-PLACE, CONTAINING .ANCIENT RUBBISH HE.\P (N
XV.), SEEN FROM SOUTH-EAST .....
ANCIENT RUBBISH HE.AP (N. XV.), IN COURSE OF EXCAVATION .
ANCIENT KHAROSHTHI DOCUMENT ON LEATHER (N. XV. 3IO) .
KH.\ROSHTHI DOCUMENT ON DOUBLE WOODEN TABLET (N. XV. 1
37)
DIAGRAMS OF WEDGE-SHAPED T.ABLET .....
COVERING TABLETS OF KHAROSHTHI DOCUMENTS ON WOOD (N. XV. I
33
167, 330),
WITH SE.\LS .......
KHAROSHTHI
DOCUMENT ON DOUBLE OBLONG T.ABLET (N. XV. 166)
SEAL-IMPRESSIONS IN CLAY, FROM KHAROSHTHI TABLETS
PAGE
268
X
LIST OF ILLUSTKATIONS.
PAGE
REMAINS OF ANCIENT TREES, NEAR SAND-BURIED DWELLING-PLACE
(n. VIII.)
378
ANCIENT KHAROSHTHI DOCUMENT ON LEATHER (N. XV.
305) . . 381
RUINED BUILDINGS WITHIN ENDERE FORT ..... 388
INTERIOR OF RUINED TEMPLE CELLA, ENDERE, AFTER EXCAVATION .
394
TWO LEAVES IN CENTRAL-ASIAN BRAHMI, FROM PAPER ROLL (E. I.
7),
FOUND IN ENDERE TEMPLE .......
395
HALF-LEAF OF TIBETAN MS. ON PAPER, FROM ENDERE TEMPLE (E. I.
32)
. 396
LOWER FLOOR ROOM OF RUINED DWELLING-PLACE, ENDERE FORT .
399
INTERIOR OF RUINED QUADRANGLE, KARADONG .... 4O9
WOODEN GATEWAY OF RUINED QUADRANGLE, KARADONG, AFTER EX-
CAVATION . . . . . . . . . .411
BOYS AND GIRLS AT KERIYA, IN HOLIDAY DRESS . . . . 42
1
VILLAGE CHILDREN, KERIYA .......
423
EXCAVATIONS PROCEEDING ALONG SOUTH-EAST WALL OF RAWAK STUPA
COURT .......... 424
RAWAK STUPA, SEEN FROM SOUTH CORNER OF COURT . . . 428
COLOSSAL STATUES WITH SEATED BUDDHA, IN SOUTH CORNER OF RAWAK
STUPA COURT, AFTER EXCAVATION ...... 432
RELIEVO STATUES OF RAWAK STUPA COURT, SOUTH-WEST WALL, AFTER
EXCAVATION .........
434
RELIEVO SCULPTURES ON OUTER SOUTH-EAST WALL OF RAWAK STUPA
COURT, IN COURSE OF EXCAVATION .....
436
RELIEVO STATUE OF BODHISATTVA (R. IV.), ON SOUTH-WEST WALL,
RAWAK STUPA COURT ........
437
COLOSSAL STATUES ON OUTER WALLS OF SOUTH CORNER OF RAWAK
STUPA COURT .........
439
TORSOS OF COLOSSAL STATUES ALONG INNER SOUTH-EAST WALL . . 44O
TORSOS OF STATUES (DVARAPALAS) AT GATE OF RAWAK STUPA COURT 44I
STUCCO HEAD OF SMALL BUDDHA OR BODHISATTVA, ORIGINALLY
PAINTED, FROM RAWAK STUPA COURT ..... 442
STUCCO HEAD OF SMALL BUDDHA OR BODHISATTVA, RETAINING
ORIGINAL COLOURING, FROM RAWAK STUPA COURT . . .
445
ISLAM AKHUN
.......... 45O
IN A KHOTAN BAZAR ......... 460
BADRUDDIN KHAN AND AFGHAN TRADER, KHOTAN ....
463
HALT ON THE MARCH DOWN THE GULCHA VALLEY, lAKuHANA . 468
RAM SINGH AND JASVANT SINGH, WITH
'
YOLCHI BEG.' IN MR.
macartney's GARDEN, KASHGAR ......
472
IN THE BAZAR OF OSH, FARGHANA ......
475
AT SAMARKAND : MARKET WITH RUINED MOSQUES IN BACKGROUND
478
KIRGHIZ FAMILY ON THE MARCH ....... 480
Sand-buried Ruins of Khotan.
CHAPTER I.
CALCUTTA TO KASHMIR.
MOHAND MARG, KASHMIR.
glow of satisfaction I could look up to
It was from the
Alpine plateau of
Mohand Marg, my
beloved camping-
ground for three
Kashmir summers,
that I had in June,
1898,
submitted to
the Indian Govern-
ment the first scheme
of the explorations
which were to take
me across the great
mountain barriers
northward and into
the distant deserts
of Khotan. Almost
two years had passed
when I found myself,
early in May,
1900,
again in Kashmir and
within sight of Mo-
hand Marg. With a
the crest of the high
1
2 CALCUTTA TO KASHMIE.
spur, some 10,000 feet above the sea and still covered with snow,
on which my tent had stood, and where my plans had been
formed. It had taken two years, and bulky files of correspond-
ence
;
but at last I had secured what was neededfreedom to
move, and the means requisite for my journey.
In the meantime official duty, and minor archaeological tours
to which I devoted my vacations, had taken me over widely
different parts of India. From Lahore, where during eleven
long years, amidst the worries and cares of University office
work, I had ever felt the refreshing touch of the true East and
the fascination of a great historical past, I had been transferred
to Calcutta. With its strangely un-Indian conditions of life,
its want of breathing space, and its damp heat, the
"
city of
palaces
"
appeared to me like a tropical suburb of London.
From there I had visited Sikkim, that strange half-Tibetan
mountain-land where true Alpine scenery is invaded by the
luxuriant vegetation of the tropics. I had wandered in South
Bihar, the ancient Magadha, tracing the footsteps of Hiuen-
Tsiang, the great Chinese pilgrim, among the ruins of the sacred
Buddhist sites which he had seen and described more than
twelve hundred years ago. Also the fascinating tracts along
the Indus and the North-West Frontier, where the influence of
classical art has left its witnesses in the ancient
"
Graeco-
Buddhist " sculptures of so many a ruined monastery and
shrine, had seen me once more on a flying visit.
The thought of the task which was drawing me beyond the
Himalaya had followed me everywhere. But it was only when
the final sanction for my proposals reached me on a sultry
monsoon night down in Calcutta that I had been able to start
.some of the multifarious preparations which the journey de-
manded. Busy as I was with official duties and literary work
that had to be concluded before leaving India, I managed to
arrange for the supply of many articles of equipment, both
personal and scientific. The tents wliich I had ordered from
the Cawnpore Elgin Mills
;
the galvanised iron water-tanks,
made at Calcutta workshops, that were to serve in the desert
;
the stores of condensed food, the photographic outfit, and the
semi-arctic
winter clothing which I had indented for from London
UP THE JHELAM VALLEY. 3
all were slowly moving up to Srinagar, whence my little ex-
pedition was to start.
But only in Kashmir itself, and not in over-civilised Calcutta,
was it possible to complete my practical preparations. So I
could not entirely suppress a feeling of unholy joy when an
increase of plague, or rather the fear of it, caused Calcutta
colleges to be closed some weeks in anticipation of the usual
summer vacation. On the loth of April I was free to escape
northward. It was a source of satisfaction to me that on the
day of my departure I was able personally to take leave of the
late Sir
John
Woodburn, Lieutenant-Governor of Bengal, and
to express my deep gratitude for all the kind help and interest
with which he had furthered my undertaking.
The week I spent in Lahore in order to pick up various por-
tions of my outfit and to supervise their despatch passed rapidly
amid old friends and surroundings dear to me. After Calcutta
the Punjab spring appeared still comparatively cool. All the
same I enjoyed as keenly as ever the invigorating change to the
fresh air of the hills when the Tonga carried me from Rawal-
pindi first to the fir-covered heights of Murree, and then along the
Jhelam Valley up towards Kashmir. Often had I done this
journey along the ancient Hydaspes, where it rushes down
towards the plains in an almost uninterrupted succession of
rapids and cataracts, but never so early in the year. Whether
it was the sight and fragrance of the shrubs still in blossom along
the road, or the glittering caps of snow still lying on many of
the higher spurs, or simply the prospect of a year's explorations,
never had this drive of nearly two days appeared to me so
enjoyable.
On the 25th of April I passed once more into the Kashmir
Valley by the gorge of Baramula, now as in ancient days the
"
Western Gate of the Kingdom." The snow still lay low down
the mighty Pir Pantsal range which forms the southern rampart
between Kashmir and the outer world. But the great riverine
plain which opens out just beyond Baramula was decked in all
the gay colours of a Kashmir spring, blue and white irises growing
in profusion over village cemeteries and other waste spaces.
At Baramula, where my servants, sent ahead with the heavy
1*
4 CALCUTTA TO KASHMIR.
baggage, awaited me, I took to boats for the remaining journey
to Srinagar
;
for old experience had shown me the convenience
and attractions of river communication in Kashmir. The day
I spent ghding in my comfortable
"
Dunga
"
through the limpid
water of the great lagoons which fringe the Volur Lake, and
along the winding course of the
Jhelam, gave delightful repose
such as did not again fall to my share for many months.
Familiar to me as are quce loca fabulosus lanihit Hydaspes, there
was plenty to feast my eyes upon. The floating meadows of
water-lilies and other aquatic plants which cover the marshes
;
the vivid foliage of the great Chinar trees which shade all
hamlets and Ghats along the river banks
;
the brilliant snow-
fields on the Pir Pantsal, and the higher ranges to the north
over which my road was soon to leadthese and all the other
splendours of Kashmir spring scenery will never lose their
charm for me.
During the second night the boat passed the winding reaches
in which the river traverses Srinagar, and the next morning
found me once more in the Chinar Bagh, my old camping-
ground in the Kashmir capital. With the increasing crowd of
European visitors from the Indian plains, the shady grove by
the side of the
"
Apple Tree Canal
"
has long ago ceased to be a
place suited for work or even quiet enjoyment. But haunted
as it is at all hours of the day by the versatile Kashmir traders
and craftsmen who jirovide for the Sahibs' camping require-
ments, it was just the place adapted for the purpose of my first
stay at Srinagar. There were plenty of orders to give for mule
trunks and leather-covered baskets or
"
Kiltas," in which stores,
instruments, etc., were to be packed. Fur coats and warm
winter clothing of all sorts had to be provided to protect myself
and my followers against the cold of the Pamirs and the Turk-
estan winter
;
bags to carry provisions, and all the other para-
l)hernalia which my previous experience showed to be necessary
for a protracted campaign in the mountains. Clever and intelli-
gent as the Kashmir craftsman ordinarily is, it requires pro-
tracted interviews to ensure that the work he is going to execute
is really that intended. So what with endless particulars to be
explained, and all the bargaining which local custom renders
V
PREPARATIONS IX KASHMIR. 5
indispensable, there remained little time dming these busy daj^
to collect information on the important questions affecting the
first part of my jovu-ney.
The Government of India in the Foreign Department had
granted me permission to use the Gilgit-Hunza route for my
journey to Kashgar. The special conditions prevaiUng along
the
"
Gilgit Transport Road
"
made it necessary to give timely
and exact intimation as to the amount of transport required,
the number of followers, etc., all the more as the time I had fixed
for my start, the end of May, was in advance of the regular trans-
port season. Luckily, Captain G. H. Bretherton, d.s.c,
Assistant Commissary-General for Kashmir, to whom I had to
apply in the matter of these arrangements, proved exception-
ally able and willing to afford information. Guided by his
experience, I was soon in a position to prepare with fair accuracy
my estimates as to the time, means of transport, and suppUes
needed not only up to Hunza, but also beyond towards the
Chinese frontier. It was no small advantage to obtain quickly
a clear working plan of these practical details. For upon the
exact information which I could send ahead to Gilgit and Kash-
gar depended my hope of securing, without loss of time, all that
was needful for the onward jovurney.
I was heartily glad when I succeeded within five busy days in
disposing of these preliminaries. The few weeks which remained
to me in Kashmir were none too long for the literary tasks that
had to be completed before my departure. For over ten years
past I had devoted whatever leisure I could spare from official
duties to work connected in one form or another with Kalhana's
>/
"
Chronicle of the Kings of Kashmir." The Sanskrit text of
the great poem, the only record of a truly historical nature that
exists in the classical literature of India, and one full of interest
for the student of Indian antiquities, rehgion, geography, etc.,
had long ago been edited by me. But my translation and com-
mentary required protracted researches into all that has survived
of ancient Kashmir in records, traditions, and
antiquarian lore,
and the two stout quarto volumes which they filled in print were
only now approaching completion. The introduction which was
to give an accoimt of these labours still remained to be written,
6 CALCUTTA TO KASHMIR.
and in order to complete it in time, together with some minor
tasks of a similar kind, seclusion was indispensable.
"To go into Purdah," as our Lahore phrase ran, within
Srinagar or its immediate environs, was well-nigh impossible,
and Mohand Marg, my mountain retreat of former seasons, was
still covered with snow. My knowledge of Kashmir topography,
however, stood me in good stead, and after a short search at the
debouchure of the great Sind Valley over which Mohand Marg
rises, I found near the hamlet of Dudarhom a delightfully quiet
grove by the river-bank where I could pitch my tents. There
under the shade of majestic Chinars and within view of the snow-
covered spurs of Mount Haramukh, I was soon hard at work
from morning till evening. It was not an easy task to sum up
and review the results of labours that had extended over so long
a period and over so wide a field. Yet I felt grateful that I was
able to bid farewell to them, while having that Alpine scenery
before my eyes with which I shall ever associate the happiest
recollections of my Kashmir researches. But still more cheer-
ing, perhaps, was the thought of the new field of exploration
that awaited me northward, far beyond the ranges I had viewed
from my
'
Marg.' Undisturbed by intrusion of any kind, these
three short weeks afforded leisure for concentrated work which,
after the preceding
"
rush," seemed almost as enjoyable as if it
had been a period of rest.
On the 23rd of May I completed the last of the tasks for the
sake of which I had retired to my peaceful camping-ground.
The date fixed for my start was drawing near, and with it came
the necessity for returning to bustling Srinagar for the last pre-
parations. Thanks to the convenient water-way provided by
the Anchar Lake and the ancient Mar Canal, a single night passed
in boats sufficed to bring me into the Kashmir capital. I found
the grounds usually occupied by European visitors more crowded
than I had ever seen them. Lines of house-boats along the river-
banks and endless rows of tents in all the
'
Baghs
'
seemed to
leave no room for a new arrival. Fortunately, in years gone
by I had had ample occasion to study the topography of Srinagar,
in its modern as well as its ancient aspects, and thus I discovered
at last a spot for my camp, on the narrow strip of ground which
CAMP ON SIND
RIVER. 7
lines the west foot of the Takht-i-Sulaiman
hill towards the Dal
Lake. Hidden behind willow
plantations and
"
floating
gardens
"
pecnhar to the lake, the httle Bagh of Buchvor offered
the needed quiet to complete my arrangements.
Busy indeed were the days I passed there. All details of
the camp outfit had to be revised
;
the freshly arrived stores to
be sorted and packed into loads for pony transport ;
surveying
and other instruments to be tested and protected against
damage
;
and amid these preparations there were accounts to
be settled and farewell visits to be received. Numerous were
the questions of my Pandit friends which I had to answer as to
the place and object of my journey. More conversant though
they are with mythical than with real geography, yet I found
that my reference to the
"
Uttarakurus
"
(the Ultima Thule of
Indian mythology) as the land for which I was about to set out,
did not altogether satisfy their curiosity.
Ram Singh, the Gurkha Sub-Surveyor, whose services Colonel
St. George Gore, r.e., the Surveyor-General of India, had very
kindly placed at my disposal, together with a complete outfit
of surveying instruments, joined me punctually on the day of
my arrival at Srinagar. He had accompanied Captain Deasy
in his recent travels near the sources of the Yarkand River and
in the Kuen-luen mountains, and the practical acquaintance he
had thus gained of the regions I was about to visit proved useful
at once in the course of my preparations. With Ram Singh came
Jasvant Singh, a wiry little Kangra Rajput, who was to attend
to the Sub-Surveyor as cook and personal servant. He too had
travelled in Chinese Turkestan as one of Captain Deasy's fol-
lowers.
On the 28th of May there arrived Sadak Akhun, the Turkestan
servant whom Mr. George Macartney, CLE., the British repre-
sentative at Kashgar, had been kind enough to engage for me.
He had left his home in the first half of April and came just in
time to start back with me. He was to act as cook and
'
Karawan-bashi
' combined, and was welcomed with no small
satisfaction by honest Mirza Alim, my Kokandi servant, whom
I had engaged four months earlier in Peshawar for the pmrpose
of myjourney.
'
Mirza ' had been useful to me by giving me the
8
CALCUTTA TO KASHMIK.
needed opportunity of practising Turki conversation, but willing
as he was to pick up the novel art of attending to the wants of a
'
Sahib,' his acquirements did not reach far in regard to the
kitchen department. His earlier career as a petty trader in
Kabul and Peshawar had not been a special preparation for these
functions
;
and yet his straightforward ways made me anxious
to retain him. Sadak Akhun's timely arrival relieved both him
and his master of all uneasiness as to the future arrangements
of the travelling household. For Sadak Akhun had brought
with him not only the appearance of a smart
'
Karawan-bashi,'
but a training in the mysteries of European cuisine amply suffi-
cient for my wants. When he turned up in his fur-lined cap and
coat of unstained azure, and red leather top-boots of imposing
size, my camp seemed to receive at once a touch of Central-
Asian colour.
But it was not only from the Far North that I was anxiously
expecting during these days a much-needed complement of my
camp. Knowing that no European traveller in the parts I was
bound for could wholly refuse the role of the
*
Hakim
'
forced
upon him by popular belief, I had early ordered my medicine
case from Messrs. Burroughs Wellcome & Co., the great London
firm of
"
Tabloid
"
fame. The South African War and other
incidents delayed its arrival for months, and even when it had
at last been reported by telegram as landed at Calcutta, it seemed
doubtful whether it would reach me in time. The Indian Post
Office does indeed provide with its usual efficiency for the wants
of the distant frontier post of Gilgit. But its power cannot level
mountains, and as the transport of heavy articles across the
snow-covered passes was not to begin till later in the season,
there seemed little chance of that eagerly looked-for case ever
catching me up if not received before my start from Srinagar.
Fortune seemed to offer a small mark of favour at least in
this direction. For when, on the evening of the 29th of May,
the time of departure fixed weeks before, my little flotilla of
boats was lying
opposite to the Srinagar Post Office, worthy
Lala Mangu Mai, the attentive postmaster, triumphantly re-
l)orted the arrival of the box. When it was at last safely
deposited in my hands it was time to set out from the Venice
START FROM SRINAGAR. 9
of India. Gliding down the dark river under the seven bridges
which have spanned it since eariy times, and between the mas-
sive embankments built with the slabs of ruined temples, I could
not fail to be impressed with

quod mihi supremum tempus in Urbe fuit.

It was midnight before I had seen the last of my old Pandit
friends, who were waiting each at the Ghat nearest to his home
to bid me farewell.
ANCIENT TEMPLE AT PANDRE;. -...:., KASHMIR.
CHAPTER II.
TO ASTOR AND GILGIT.
When I awoke in the morning
my boat had just entered the
lagoons which fringe on the east
the great Volur Lake. A look
towards the mountain range which rises above it on the north
showed that the heavy rain of the last few days meant fresii
snow on the passes I had to cross. Bandij)ur village, which
forms as it were the port for the route connecting Kashmir
with Gilgit and the regions beyond, was soon reached. It
VIEW IN BUKZIL VALLEY.
THE NEW OILGIT ROAD. U
appeared as I had seen it in 1889
on my march to Skardo,
pre-eminently a place
differtum
natUis cauponibus atqtie malignis.
But as regards transport arrangements it was easy to realise
a marked change. Since an Imperial garrison was placed in
Gilgit and the new
"
Gilgit Transport Road
"
was constructed,
the Indian Commissariat
Department has taken charge of the
means of transport on this route. Timely arrangements had
been made on my behalf by Captain Bretherton, and a
reference to the warrant officer on the spot brought the quick
assurance that ponies and cooUes would be available whenever
wanted. The time when the intending traveller on this route
had to press his transport as best he could has passed, let us
hope, for ever. If restrictions have to be placed on the number
of private xasitors in the interest of the commissariat work
on which the supplies of the Gilgit garrison depend, the dis-
advantage is amply compensated by the benefit to the Valley
at large. There was a time, still vividly remembered, when
the demand for coolies to carry military baggage or supplies
moving to Gilgit would spread terror through Kashmir villages.
Of the thousands of cultivators used annually for this corvee,
a large proportion never saw their country again
;
for, ill-fed
and still worse clad, the
'
Begaris ' succumbed only too readily
to the inclemency of climatic conditions or the epidemics
favoured by them. All this has changed since Imperial advice
and control has made itself felt in Kashmir, and the construction
of the new Gilgit road, fit throughout for laden animals, including
camels, during three summer months, has rendered the use of
human labour altogether superfluous.
On the morning of the 31st of May sixteen ponies were ready
to receive the loads which were made up by our tents, stores,
instrmnents, etc. Formidable as this number appeared to
me, accustomed as I was to move hghtly on my wanderings
in and about Kashmir, I had the satisfaction to know that my
personal baggage formed the smallest part of these impedi-
menta. When the string of animals had filed off together
with the Sub-Surveyor and
servants, there were yet
imposing
12
TO ASTOR AND aiLGlT.
indents to sign and bills to pay which the obliging Commissariat
Conductor kept ready for me under a group of fine Chinars
by
the roadside where on sunny days he transacts the business
of his office. No transport can move up the road without
his permission, and though the procedure he superintends
is modern in its ways, yet it seemed to me as if this modest
British official had simply taken the place of those
"
Masters
of the Gates
"
who used in ancient Kashmir to guard all routes
leading into the valley.
The road, after leaving the straggling line of wooden huts
which form the Bazar of Bandipur, leads for about four miles
up the open valley of the Madhumati stream. In the irrigated
fields the fresh green of the young rice-shoots was just ap-
pearing, while the hamlets on either side were half hidden
under the rich foliage of their Chinars and walnut-trees. It
was the typical spring scenery of Kashmir to which I here
bade farewell. Near the village of Matargom the road turns
to the north to ascend in long zigzags the range which
forms the watershed between Kashmir and the valley of
the Kishanganga. From the spur up which the road winds
I had a splendid view of the Volur Lake and the snow-
covered mountains to the east which encircle the hoary
Haramukh Peaks. At a height of about
9,000
feet a fine
forest of pines covers the spur and encloses a narrow glade
known as Tragbal. Here the snow had just disappeared, and
1 found the damp ground strewn with the first carpet of
Alpine flowers.
A rude wooden rest-house begrimed with smoke and mould
gave shelter for the night, doubly welcome, as a storm broke
soon after it got dark. The storm brought fresh snow, and
as this was sure to make the crossing of the pass above more
difficult I started before daybreak on the ist of
June.
A steep
ascent of some two thousand feet leads to the open ridge which
the road follows for several miles. Exposed as this ridge is
to all the winds, I was not surprised to find it still covered
with deep snowdrifts, below which all trace of the road dis-
appeared. Heavy clouds hung around, and keeping off the
rayi of the sun let the snow remain fairly hard. Soon, how-
OVER THE TRAGBAL PASS. 13
ever, it began to snow, and the icy wind which swept the ridge
made me and my men push eagerly forward to the shelter
offered by a Dak runners' hut. The storm cleared before
long, but it sufficed to show how well deserved is the bad repute
which the Tragbal
(11,900
feet above the sea) enjoys among
Kashmirian passes.
For the descent from the pass I was induced by the
'
Markobans
' owning the ponies to utilise the winter route
which leads steeply down into a narrow snow-filled nullah.
Though the ponies slid a good deal in the soft snow of the slope,
we did not encounter much difficulty until we got to the
bottom of the gorge. Here the snow bridges over the stream
which flows from this valley towards the Kishanganga had
begun to give way, and the high banks of snow on either side
were in many places uncomfortably narrow. At last our
progress was stopped at a point where the stream had washed
away the whole of the snow vault. To take the laden animals
along the slatey and precipitous side of the gorge, which was
free from snow, proved impracticable. To return to the top
of the gorge, and thence follow the proper road which descends
in long zigzags along a side spur, would have cost hours. So
the council of my
'
Markobans,' hardy hill-men, half Kashmiri,
half Dard, decided to try the narrow ledge of snow which
remained standing on the right bank of the stream. The
first animal, though held and supported by three men, slipped
and rolled into the stream, and with it Sadak Akhun, who
vainly attempted to stem its fall. Fortunately neither man
nor pony got hurt, and as the load was also picked out of the
water the attempt was resumed with additional care. Making
a kind of path with stones placed at the worst points, we
managed to get the animals across one by one. But it was
not without considerable anxiety for my boxes, with survey
instruments, and similar contents, that I watched the operation.
Heavy rain was falling at the time, and when at last we had
all the ponies once more on a safe snow-bridge, men and animals
were alike soaked. By one o'clock I reached the Gorai rest-
house, down to which the valley was covered with snow, having
taken nearly seven hours to cover the eleven miles of the march.
14 TO ASTOE AND GILGIT.
The little rest-house, looking doubly bleak in the drizzling
rain, held already three
'
Sahibs,' officers who were returning
from their shooting nullahs to Kashmir and the plains.
Refreshed by their hospitality, I decided to push on to the
next stage, Gurez, where better shelter and supplies were
available. The offer of some Bakhshish, and the hope of a
dry and comparatively warm corner for the night, overcame
the remonstrances of the
'
Markobans,' and the little caravan
moved on. Some four miles lower down 1 reached the main
valley of the Kishanganga, and in it the first Dard village.
Another ten miles' march up the valley brought me to Gurez,
a collection of villages at a point where the valley widens to a
little plain, about a mile broad.
Sombre and forbidding the valley looked between its high
pine-covered mountains and under a dark, rainy sky. The
effect was heightened by the miserable appearance of the rude
log-built dwellings scattered here and there along the slopes,
and by the dark-coloured sand in the bed of the river. The
latter bears, not without good reason, the name of 'the
"
Black
Ganga " (in Sanskrit, Krishnaganga). The backward state
of the vegetation showed that spring had only just commenced
in the valley, which here has an elevation of about 8,000 feet
above the sea. With its short summer and scanty sunshine it
can raise but poor crops of barley and 'Trumba,' and the popu-
lation is accordingly thin.
The mountain range towards Kashmir marks also a well-
defined ethnographic boundary. The Dard race, which inhabits
the valleys north of it as far as the Hindukush, is separated
from the Kashmiri population by language as well as by physical
characteristics. The relation between the language of the
Dards and the other Indo-Aryan vernaculars of North-Western
India is by no means clearly established. But whatever the
linguistic and ethnic affinities of the Dard race may be, it is
certain that it has held these valleys since the earliest time
to which our historical knowledge can reach back. Herodotus
had heard of them in the same region they now inhabit ; for
he mentions the gold-washing operations still carried on by
them within modest limits on the Indus and the Kishanganga.
AMONG
THE DARDS.
15
There is little in the Dard to enlist the sympathies of the casual
observer. He lacks the intelligence, humour, and fine physique
of the
Kashmiri, and though undoubtedly far braver than
the latter, has none of the independent spirit and martial bearing
which draws us towards the Pathan, despite all his failings.
But I can never see a Dard without thinking of the thousands
of years of struggle these tribes have carried on with the harsh
chmate and the barren soil of their mountains. They, like
the Afridis,who also are mentioned by the Father of History,
have seen all the great conquests which swept over the North-
West of India, and have survived them, unmoved as their
mountains.
Gurez was once the chief place of a little Dard kingdom
which often harassed the rulers of old Kashmir. But I confess,
when I approached it at the close of my fatiguing double march,
this antiquarian fact interested me less than the
comfortable
shelter which I found for my men and myself in Mr. Mitchell's
new bungalow.
The following day was a halt, for my people needed rest
and my baggage drying. There were besides fresh arrange-
ments to be made for the transport ahead. In Srinagar I had
been told officially that the Burzil Pass, which had to be
crossed between Gurez and Astor, would, owing to the deep
snow, be open only for coolie transport. However, from the
parties of Dards whom I met on the road, and who had brought
their unladen ponies safely across from Astor, I gathered better
news. As the use of coolies meant a complete rearrangement
of the loads, and still more trouble for the scanty population
of the valley, which had already been obliged to furnish a
hundred carriers for a survey party ahead of me, I decided
to take ponies. These were easily forthcoming, and on the
morning of the 3rd of
June
I set out from Gurez much as I had
reached it, except that the more dehcate instruments,
like
theodolite and photographic cameras, were entrusted to the
safer backs of coolies.
The weather had cleared at last, and the march from Gurez
up the side valley of the stream which comes from the Burzil
was most enjoyable. To the south there was the view of the
16 TO ASTOR AND GILGHT.
fine snow-covered mountains which divide the Kishanganga
Valley from Kashmir, while along the route leading north-
wards the slopes of the valley refreshed the eyes with their
rich green of Alpine meadows and pine forests. Of avalanches
which had swept down on the road there were many to cross.
But the task of taking the ponies over them was trifling after
the Tragbal experiences. I halted for the night at Pushwari,
and next morning continued the march in the same direction
and amidst similar scenery up to Minimarg
There, at an altitude of nearly 10,000 feet above the sea,
the valley widens to a little plain with plenty of grazing and
a little collection of huts used by Gujar cattleherds for their
summer quarters. The snow had melted here about ten days
before, and the meadow land was already covered with young
shoots of grass and a variety of hardy Alpine flowers, mostly
old acquaintances from my beloved Kashmir
'
Marg.' But
a glance at the telegraph ofiice placed here to keep watch over
the line across the pass was sufficient to show the rigour of
the winter season. Raised high above the ground, and enclosed
with heavy palisaded verandahs and sheds, the building looked
more like a small fort than an office. These precautions are,
indeed, necessary in order to make the place inhabitable during
the long winters with their heavy snowfall.
At Minimarg the route to the Burzil strikes off to the north-
west, and ascending the valley some five miles higher I reached
the rest-house at the foot of the pass. The snow began to cover
the ground soon after Minimarg was left behind, and at the
foot of the pass it was a true winter scenery which met the
eye. The sky was of a dazzling blue, and so clear that I felt
quite reassured as to the result of taking my laden ponies across
the pass.
The only condition to be observed was an early ascent before
the snow should become soft. I therefore got up at one o'clock,
and an hour later my caravan was plodding up the snow-filled
ravine which forms the winter route to the pass. Of the road
no trace could be seen. After two hours' steady ascent we
arrived at the point where the Burzil defile is met from the
north-east by another pass leading down from the high plateau
CROSsma of burzil pass. 17
of the Deosai. A telegraph shelter-hut raised on a wooden
scaffolding some thirty feet high serves as a guiding-post to
the parties of Dak runners who are obUged to carry the Gilgit
mail during the winter. The structure was even now some
ten feet deep in the snow. Fortunately the temperature was
so low that the hard snow offered comparatively good going
to the animals. By the time that the first rays of the sun
swept across the higher ranges to the east, we had gained safely
the summit of the pass,
13,500
feet above the sea. The six
miles from the rest-house had taken over three hours. There
was no distant view from the pass, which lies between winding
spurs, but the glittering snowfields all around, covered with
a spotless crust of fresh ice, were a sight not to be forgotten.
The temperature was only 35^ F. when I took a hurried breakfast
under the shelter of the Dak hut.
The descent on the north side was long and tiring. The
snow lay for some eight miles from the top of the pass, and
as the morning advanced the going necessarily became heavier.
The only living beings that inhabit this Arctic waste are big
marmots. Sitting on the top of their burrows as if to warm
themselves in the sun, they did their best to attract attention
by shrill whistling, only to disappear with lightning speed at
the approach of danger. It took some time before my little
fox-terrier realised this, and refrained from spending his breath
in vain attempts to rush the provoking animals.
'
Dash,'
or
'
Yolchi Beg
'
("
Sir Traveller "), as he had been renamed
since I took to Turki with Mirza, proved true to his name.
He marched as cheerily over the miles of snow as on earlier
wanderings through the dusty dry Punjab plains or in the
dripping jungle of Sikkim. My Turki servants soon grew
fond of
'
Yolchi Beg,' and, being untrammelled by the caste
conventions of India, never hesitated to show their affection
for my faithful companion.
It was one o'clock when I arrived at Chillum Chauki, the
first rest-house on the Astor side of the pass, having left the
snow behind about two miles before. All the ponies came in
safely except one, the absence of which was soon noticed when
I was looking^out^for breakfast. The pony carrying the
18 TO ASTOR AND GrlLGlT.
kitchen
*
Kiltas
'
had lagged behind, and I became painfully
aware that something had gone wrong when hour after hour
passed in vain expectation. My Surveyor, who had marched
in the rear, brought news of the animal having broken down
in the softening snow, and though I at once despatched coolies
to its assistance it was not till after six in the evening that
Mirza turned up with his charge. As if to console me for the
delay in bodily comforts, I got in the evening the cheerful news
of the occupation of Pretoria from Mr. M., the road engineer,
who arrived at the rest-house from a shooting excursion. News
travels fast along the telegraph line, and although there is,
apart from the Political Agent at Gilgit, no subscriber to
Renter's messages this side of the Burzil, telegraph masters
in Astor and their friends were evidently well informed of what
was happening far away in South Africa.
The set of ponies which I had brought from Gurez, and
which were the first laden animals that crossed the Burzil
this year, were relieved at Chillum Chauki by fresh ones sent
up for me from Astor. My march on the 6th of
June, down
the valley leading to Astor, was recreation after the previous
one. Notwithstanding the brilliant colour imparted to the
scenery by a blue sky, glittering bands of snow in the ravines,
and the green tossing stream at the bottom of the valley, it
was easy to realise that the crossing of the water-shed between
the drainage areas of the Jhelam and Indus meant the entry
into a sterner region. The hillsides were no longer clothed
with verdure as in Kashmir and the Kishanganga Valley. On
the slopes of bare decomposed rock cedars and a kind of juniper
showed themselves only in scanty patches. Cultivation lower
down also bore evidence of the unfavourable conditions
of soil and climate. All the more cheerful it was to behold,
by the side of the little terraced fields of more than one hamlet,
an oblong sward carefully marked off with stonesthe polo
ground of the villagers. Polo is the national game of all Dard
tribes
;
and that even the inhabitants of these poor mountain
hamlets make a sacrifice of valuable soil for its sake attests their
devotion to this manly pastime.
At Gadlioi, where a march of about seventeen miles brought
THE ASTOR CAPITAL. 19
me, it was already distinctly warmer than I had felt it since
leaving Kashmir, though the aneroid still indicated an deviation
of about
9,000
feet. On the 7th of June I continued my journey
to Astor, the chief place of the hill district, to which from early
times it has given its name. Some miles below Gadhoi there
showed themselves above the bare rocky mountains along
the valley the icy crests of the great mass of peaks culminating
in Nanga Parbat. That giant of mountains (26,600 feet above
the sea), the ice-clad pjTamid of which I had so often admired
from Kashmir Margs, and even from above Murree itself,
remained hidden behind lower ranges, though only about ten
miles away, as the crow flies. Yet even its bodyguard of minor
peaks, ranging between 18,000 and
23,000 feet, was a sufficiently
inspiriting sight.
I felt the need of looking up to their glacier walls
;
for down
on the road it got warmer and warmer. From Gurikot onwards
where the two branches of the Astor Riv^er unite, the road,
dusty and hot, winds up the steep scarp on the left side of the
valley until at last the group of villages known as Astor came
in sight spread out over a mighty alluvial fan. The view that
opened here was striking in its ruggedness. For a wall of
rocky ridges seems to close the valley to the north, while the
deep ravines cut by the mountain torrents into the allu\4al
plateaus on either side give them a look of fantastic diversity.
I reached at
3
p.m. the bungalow of Astor, situated on a
dominating point of the plateau, and felt heartily glad of its
shade and coolness. Below me lay the Fort of the Sikhs, now
used for the accommodation of a battery of Kashmir Imperial
Service troops, while on the south there stretched the orchards
and fields of the Astor
"
capital." The Rajas of Astor have
become
"
mediatised" since the advent of the Sikhs, and their
power, such as it was, is now wielded by a modest Tahsildar
of the Kashmir administration. Generosity was not a fault
of Sikh rule in these mountain regions, and the deposed family
of hill chiefs have little left to support the pride of their ancient
hneage.
Though Astor lies about
7,700
feet above the sea, the air
would have been decidedly oppressive but for a storm which in
2*
20 TO ASTOR AND GILGIT.
the evening swept over the valley. It left plenty of clouds
behind to screen the sun on the next morning
(June
8th) when
I resumed the march down towards the Indus. The valley
became bleaker and bleaker as the route descended, and the
streaks of red, yellow, and grey displayed by the rocky hillsides
offered poor compensation for the absence of vegetation. Of
flowering shrubs only a kind of wild rose seems to thrive on
the barren soil, and being just in full bloom caught the eye
by its purple patches. A few green fields perched on the top
of small alluvial fans were all the cultivation visible on the
fifteen miles' march to Dashkin village, the first stage from
Astor.
As I had heard of the arrival at the next stage, Duyan, of
Captain
J.
Manners Smith, whom I was anxious to meet, I
decided to push on. The slow rate of progress made by the
baggage animals confirmed the objections which my pony-men
had raised. But otherwise, this extra march of twelve miles
proved a pleasant surprise. The road, rising gradually to
about
5,000
feet above the tossing river, took me through a
charming forest of pines, which in the shadows of the setting
sun looked its best. This forest evidently owes its growth to
its sheltered position on the north-east flank of a great ridge,
which on its top was still covered with snow. It was a pleasure
to behold once more green moss and ferns along the little streams
which rush down through the forest. But when this was left
behind at the turning of a cross-spur there spread a grander
view before me.
Through the gap between the mountains enclosing the Aster
Valley there appeared the broad stream of the Indus and beyond
it range after range towards the north. Thin clouds hung over
the more distant ranges, yet I thought I could recognise rising
above the fleecy mist the icy mass of Mount Rakiposhi. Father
Indus was greeted by me like an old friend. I had seen the
mighty river at more than one notable point of its course, where
it breaks through the rocky gorges of Baltistan, where it bursts
forth into the Yusufzai plain, and in its swift rush through the
defiles below Attock. But nowhere had it impressed me more
than when I now suddenly caught sight of it amidst these
HALT AT DUYAN. 21
towering mountain walls. The shadows of evening fell quickly
in its deep-cut valley and the glittering xdsion of the river had
vanished when, somewhat tired, I reached the end of my double
march.
My stay at Duyan was prolonged in the pleasantest manner.
Early on the morning of
June
9
Captain
J.
Manners Smith,
v.c, C.I.E., the Political Agent of Gilgit and the adjacent hill
tracts, on his return from a shooting excursion, came to see
me and kindly invited me to spend the day in his camp. I
was most glad to accept the hospitality of the distinguished
officer, then acting as
"
Warden of the Marches
"
for the moun-
tain region I was about to traverse
;
and after despatching
my party ahead, soon found myself riding on one of his hill
ponies up to the mountain-side occupied by his tents. It was
a charming spot on a httle shoulder of the fir-covered slope,
some
1,500
feet above the road, where the ground was carpeted
\vith wild Wolets, forget-me-nots, and other mountain flowers,
and where a bright little stream added to the attractions of
the scene. Picturesque, indeed, it was with the well-fitted
hill tents of the PoHtical Agent and the motley crowd of his
followers haiUng from all parts of Gilgit, Chilas, and Hunza.
In the amiable society of my host and Mrs. Manners Smith
I passed a day which I shall long remember for its varied
enjo\Tnents. Anglo-Indian ladies know how to carry true
refinement into camp life even at the most distant points of
the Empire, and here Nature had surrounded the tasteful
comforts of a well-arranged camp with special glamour. The
hours I spent at this dehghtful spot fled only too fast. Captain
Manners Smith, who has been connected with the political
administration of this region for the last twelve years, and
whose Victoria Cross was earned at one of the most striking
incidents of its modern history, the storming of the Hunza
fastness beyond Xilth, knows these mountains and their races
better probably than any European.
What added to the interest of his varied commvmications
about the old customs and traditions of the people was the
illustration which his remarks received from the hillmen attending
his camp. The petty headmen from the vaUeys towards Chilas
22 TO ASTOR AND GILGIT.
and from Punyal furnished me with more than one curious
fact bearing on the earlier social and religious condition of
the tract. Muhammadanism is a comparatively recent growth
here, and the traditions as to former worship and rites have
survived in many a valley. One grey-bearded village headman
from Gor in particular seemed full of old-world lore. He had
investigated the relics of an old burial-place near his home,
where the burnt bodies of his ancestors in pre-Muhammadan
times used to be deposited, and was not shy about relating
the drastic punishment which as a boy he had received from his
mother when disturbing the spot. In these mountains, as
elsewhere throughout the world, it is the women-folk who act
as the best guardians of all old lore and tradition.
The close contact with the Far West into which modern
political conditions have brought these once secluded valleys
was illustrated by the fact that I could read at Captain Manners
Smith's table the latest Renter telegrams just as if it had been
in the Club at Lahore. But the presence in camp of my host's
pretty little children offered an even more convincing indication
how far European influence has penetrated across the mountains.
Bright and rosy-cheeked, they were worthy representatives of
the British Baby which in the borderlands of India has always
appeared to me as the true pioneer of civilization. I have
come across it in many a strange place, and its manifest
happiness amongst surroundings which often seemed in-
congruous with the idea of a nursery has ever forced me to
admiration. The British Baby has never been slow to follow the
advance of British arms in India. Occasionally it has come
early enough to see some fighting : witness Fort Lockhart and
the Malakand. But on the whole its appearance on the scene
marks the establishment of the pax britannica, and for this
mission of peace and security it well deserves that thriving
condition which it usually enjoys in the mountains around
Kashmir.-
For afternoon tea my hosts took me to a pretty
'
Marg
'
on the top of the ridge above their camp. From this height
the Indus Valley, in its barrenness of rock and sand, could be
seen descending far away towards Chilas and Darel. The
DOW^f^ TO THE IXDUS.
23
day will come when this natural route to the Indian plains
will be open again as it was in old times. Then the last bit
of terra incognita along the Indus, which now extends from
Chilas down to Amb, will be accessible, while the difficulties
inseparable from a line of transport crossing the great barriers
of the Kashmir ranges will no longer have to be faced.
On the morning of the.ioth of June
I took leave of my kind
hosts and hurried down towards Bunji to catch up my camp.
As I descended the defile of the Astor River, where the road
leads along precipitous cliffs and past shingly ravines, the
heat rose in a marked degree. I could well reahse what the
terrors of this part of the route, known as Hatu Pir, abso-
lutely waterless and exposed to the full force of the sun, must
have been for the Kashmiri coolies of old days. On the
eleven miles which brought me down to the level of the
Indus close to the point where the Astor River joins it, I did
not meet with a single traveller. Equally desolate was the
ride from Ramghat, where the road crosses the Astor River,
to Bunji, some eight miles higher up on the Indus. The broad
rocky plain which stretches from the bank of the great river
to the foot of the mountains showed scarcely a trace of vege-
tation. The radiation of the sun's rays was intense, and I
was glad to reach by one p.m. the shelter of the Bunji Bungalow.
The neighbouring fort is still held by some detachments of
Kashmir troops, though the ferry over the Indus which it once
guarded has become disused since the construction of the new
road. During the hot hours I spent at Bunji there was little
to tempt me outside. A hazy atmosphere hung over the valley
and deprived me of the hoped-for view of Nangaparbat, which,
rising fully 22,000 feet above the level of the Indus, dominates
the whole scenery in clear weather. A strong wind blowing
down the valley carried the fine sand of the river-bed even
into the closed rooms. Bunji altogether seemed by no means
a desirable place to spend much time in, and strongly reminded
me of the hot days I had once passed in the low hills of Jammu
territory.
Fodder is practically not to be got at Bunji, and this accounts
tor the difficulty I found in procuring a pony that was to take
24 TO ASTOR AND GILGIT.
me in the evening to the next stage where my baggage had
marched ahead. At last the local Tahsildar had to lend me
his mount, but it was already evening before I could set out.
A lonely ride across a sandy plain brought me to the imposing
suspension bridge which spans the Indus, just as it was getting
dark. In the dim light of the moon which was then emerging
for a time from the clouds the deep, rockbound gorge of the
river looked quite fantastic. And so did the rugged mountains
further east through which the Gilgit River comes down to
meet the Indus. To ride along the face of the rocky spur
which rises in the angle of the two rivers was slow work in the
scanty light of a fitful moon, and by the time I had turned
fully into the Gilgit Valley, and reached safer ground, rain came
on and brought complete darkness. Mile after mile passed
without my coming upon the longed-for rest-house where I
could rejoin my camp. At last it became clear that I must
have passed it by, and I had only the choice of continuing my
ride straight into Gilgit or returning to search for the missed
bungalow. Dark as it was I preferred the latter course, and
ultimately discovered a side path which brought me to the
expected shelter fully half a mile away from the main road.
It was close on midnight when I sat down to the dinner which
my servants had duly kept ready for me, though it had never
struck them that I might require a light to show me the way
to it.
Pari, where I spent what remained of the night, proved in
the morning a desolate spot by the sandy bank of the river,
enclosed by an amphitheatre of bare reddish-brown mountains.
The scenery remained the same for the next nine miles or so
until after rounding one of the countless spurs along which the
road winds the open part of the great Gilgit Valley came into
view. Minaur is the first village where cultivated ground
is again reached, and thereafter every alluvial fan on the left
bank was green with carefully terraced and irrigated fields. A
few miles further on the valley of the Hunza River opens from
the north, and beyond it stretches the collection of hamlets
to which the name Gilgit properly applies. It was a cheerful
sight to view this expanse of fertile fields and orchards from
AT THE GILGIT AGENCY. 25
the height of an old moraine issuing from a side valley. \\'hile
riding through it I was met by a note from Captain H. Burden,
I. M.S., the Agency Surgeon, offering me that hospitable reception
for which Captain Manners Smith's kindness had prepared
me.
I soon was installed in a comfortable set of rooms, and
realised that for my stay at Gilgit I was to be the guest of the
officers remaining at the headquarters of the Agency. Small
as their number was I found among them most attractive and
congenial company. Each of them, whether in charge of the
Kashmir Imperial Sers-ice troops supplying the local garrisons,
or of the Commissariat, the Public Works, or the hospitals of
Gilgit, showed plainly that he knew and liked these hills. For
each the semi-independence secured by the arrangements of
an out-l>ang frontier tract under
"
pohtical
"
management had
been a source of increased activity and consequent experience
in his own sphere. That the pohtical interests which necessitated
the garrisoning of Gilgit with Imperial officers and troops have
benefited this region in more ways than one was apparent from
a stroll through the little
"
station." I found there a
well-built hospital, neat offices for the various departments
of the administration, a clean and airy bazar, and even sub-
stantial buildings for a school and a zenana hospital. Small
but comfortable bungalows have been built for the European
officers on the terraced slopes overlooking the valley, and in
their midst there has quite recently risen even a substantial
club with an excellent though necessarily select library. It
is only some eleven years since the new era set in for Gilgit, and
yet it is already difficult to trace the conditions which preceded
it. The fort, built of rubble with a wooden framework, after
the usual Sikh fashion, alone reminds one of the days when
Gilgit was the prey of an ill-paid and badly disciplined soldiery,
when years of unabated exactions had laid great parts of the
cultivable land waste and driven the now peaceful Dards into
violent rebellions.
I had originally intended to stop only one day at Gilgit in
order to give my men a much-needed rest and to effect some
repairs in the equipment. But difficulty arose about getting
26 TO ASTOR AND GILGIT.
fresh transport for the march to Hunza, and my stay was
of
necessity extended to three days. Ample work and the amiable
attention of my hosts scarcely allowed me to notice the delay.
Though all Government transport was occupied in out-lying
camps, and the local ponies were grazing far away in distant
nullahs, Captain E. A. R. Howell, the energetic Commissariat
Officer, provided by the third day a train of excellent animals
to which I could safely trust my baggage up to Hunza. Little
defects in my outfit which the experience of the previous marches
had brought to light were easily made good in the interval,
since every member of the
"
station
"
offered kind help. While
the Commissariat Stores supplied what was needed in the way
of followers' warm clothing, foodstuffs, etc., Mrs. W., the only
lady left in the
"
station," kindly offered threads of her own
fair hair for use in the photo-theodoHte. How often had I
occasion to feel grateful thereafter for this much-needed
reserve store when handling that delicate instrument with half-
benumbed fingers on wind-swept mountain-tops !
MIK 6 CASTLE AI BALTIT.
THROUGH HUNZA.
On the afternoon of the
15th of
June I left Gilgit
full of the pleasant impres-
sions
from my cordial re-
ception at this last Anglo-
Indian
outpost. The first
march of eighteen miles was
to Nomal, a green oasis in the
otherwise
barren valley of the
river which comes from Hunza.
The
preceding days in Gilgit
had been abnormally cloudy and cool, and this weather made
marching pleasant enough: Since the little war of
1891,
which had asserted British authority in Hunza, the road up
the valley has been greatly
improved. Nevertheless, it is but
28 THROUOH HUNZA.
a narrow bridle path, and as it winds along precipitous spurs
many hundred feet above the stream, it required such a steady
hill pony as that kindly lent to me by Major E.
J.
Medley,
of the 17th Bengal Lancers, then Commanding the Force in
Gilgit, to ride with any feeling of comfort.
From Nomal and upwards the river has cut its way through
a succession of deep gorges, lined often with almost perpen-
dicular chffs. The path is carried in long zigzags over the
projecting cross-ridges, and more than once traverses their face
by means of galleries built out from the rock. At Chalt, the
end point of my second day's march, I reached the limit of
Gilgit territory. Here the valley widens considerably and
takes a sharp turn eastwards. As a reminiscence of an earlier
state of things the place is garrisoned with a company of
Kashmir Imperial Service troops. Their commandant, an
aged Subahdar from the Garhwal district, came to call on me
soon after I had arrived at the comfortable bungalow of the
Military Works Department. In the course of our long con-
versation he gave me graphic accounts of what Gilgit meant
to the Kashmir troops twenty and thirty years ago ; of the
hardships which the want of commissariat arrangements
caused both to the soldiers and the inhabitants. From the
description of these sufferings it was pleasant to turn to other
aspects of soldiering in the old Dogra service, e.g., the quaint
Sanskrit words of command concocted under Maharaja Ranbir
Singh, and still in use not many years ago.
On the 17th I intended to make a double march, pushing on
straight to the centre of the Hunza valley, where baggage
animals were to be left behind and coolies taken for the rest
of the journey to the Taghdumbash Pamir. After leaving
Chalt the road crosses to the left bank of the river by a fine
suspension bridge, hung like the rest of the more important
bridges on the route from Kashmir, from ropes made of
telegraph-wires. This mode of construction, first tried in these
parts by Colonel Aylmer, of the Royal Engineers, has proved
everywhere a signal success ; its advantages are easily appreci-
ated in a country where other suitable materials could scarcely
be carried to the spot.
MOUNT
RAKIPOSHI. 29
It was after rounding a long massive spur which causes a
great bend in the river-bed that I first beheld the ice-clad
peaks of Mount Rakiposhi in their glory. The weather had
been too cloudy during the
preceding days to see much of this
giant of mountains while I was marching in the valleys which
flank it to the south and west. Now that I had got to its
north side a day of spotless clearness set in, and the dazzling
mass of snow and ice stood up sharp against the blue sky.
Rakiposhi, with its towering height of over
25,500
feet,
commands completely the scenery in the Upper Hunza Valley.
Though several peaks run it close in point of elevation, none
can equal it in boldness of shape and noble isolation. All day
long I revelled in this grand sight, hidden only for short dis-
tances by the spurs which Rakiposhi sends down into the
valley. Between them lie deep-cut side valleys through which
the streams fed from the glaciers of Rakiposhi make their
way to the main stream. The ample moisture supplied by the
eternal snows of the higher slopes has not only brought
verdure to the cultivated terraces in the valley. High above
the walls of bare rock which bound the latter, patches of pine
forest and green slopes of grazing land can be seen stretching
up to the edge of the snow line. Glaciers, of spotless white
on their higher parts, but grey with detritus below, furrow
the flanks of the mountain mass and push their tongues almost
down to the level of the main valley which here rises from six
to seven thousand feet above the sea.
At Nilth, some eight miles above Chalt, the first Nagir
village is reached.
It was the scene of the notable fight
which decided in
1891 the fate of Hunza and Nagir. The
two little hill states which di\'ide between them the right and
left sides of the valley jointly known as Kanjut, had stoutly
maintained
their independence against all Dogra attempts at
conquest. No wonder that people to whom their own
mountains
offer so scanty room and sustenance proved
troublesome
neighbours.
Slave raiding into the lower valleys
had for a long time been a regular source of revenue for the
chiefs or Mirs of Hunza. The plundering expeditions of the
sturdy
Kanjutis were feared by caravans far away on the
30 THEOUGH HUNZA.
Pamirs and on the trade routes towards the Karakorum.
Across the great glaciers which stretch along the flanks of the
Muztagh range parties of Kanjuti freebooters used to break
into the valleys of Baltistan. I well remember the rude
towers near the mouth of the great Biafo glacier which I saw
on my visit to the Braldo Valley eleven years before. They
plainly showed that even in that forbidding region raids from
Hunza had to be guarded against.
All this has changed with the brilliant little campaign which
began and ended at Nilth. The graphic accounts of Mr.
Knight, who accompanied the small force from Gilgit as
correspondent of the Times, has made all the incidents well
known. From the shady little Bagh in front of Nilth where I
halted for breakfast, I could conveniently survey the fortified
village which Colonel Durand's force stormed, and the pre-
cipitous gorge behind, which stopped his further progress for
nearly three weeks. The sangars which had crowned the
cliffs on the opposite side and from which the men of Nagir
had offered so stout a resistance, were already in ruins. But
of their defenders, several joined me in a friendly chat, and
pointed out all the important positions.
Nothing speaks more for the policy and tact of the victors
than the good feeling with which the people of the valley
remember the contest. The men of the local
"
Levies
"
who
showed me the precipitous cliffs of conglomerate over i,ooo
feet high, seen on the left of the accompanying photograph,
which Captain (then Lieutenant) Manners Smith scaled with
his handful of Dogras and Gurkhas, seemed almost as proud
of the daring exploit that had won that gallant officer his
Victoria Cross as if it had been done by one of themselves.
The explanation lies probably in the fact that all interference
with the habits of the people and their traditional rulers has
been scrupulously avoided. The small garrison of Kashmir
Imperial Service Troops which was quartered in the centre
of the valley for a few years has been removed. The British
Political Officer who was left to advise the chiefs of Hunza
and Nagir, has now also been withdrawn, and of the visible
effects of the conquest there now remains nothing in the valley
NILTH GORGE. 31
but a well-made road and absolute security for the traveller.
The zeal and bravery which the Kanjuti le\ies displayed when
called to aid in the Chitral campaign are the best proof of the
loyal spirit with which the changed situation has been accepted.
Yet this population of brave mountaineers, small as it is, has to
CLIFFS OF NILTH GORGE, XAGIR.
struggle harder than ever to maintain itself amidst these gorges
bound by rock and ice, now that the days of raiding are gone.
From Nilth onwards the road leads over a succession of
highly cultivated plateaus, separated by deep-cut glacier
ra\-ines.
Everywhere there were little clumps of fruit trees,
of which the mulberries were just ripening. The Ullages
which I passed were distinctly picturesque, being all enclosed
with walls of rough stone and square loop-holed towers. Their
32 THROUGH HUNZA.
position, which is usually on the very edge of the plateau,
falling off in precipitous banks towards the river, also shows
that safety was a consideration. Old are these sites in all
probability, but the only remains of antiquity that I could see
or hear of above ground are those of a small Buddhist Stupa
or relic tower passed on the road close to the hamlet of Thol.
Built of solid masonry, it rises on a base of ten feet square to
a height of nearly twenty feet, and is remarkably well pre-
served. The only damage done is at the corner, where the
masonry of the base has been knocked off to save the detour
of a few feet to the road which passes by the side of the monu-
ment. It is evident that even at so remote a spot the
"
Public
Works
"
of modern India involve the same danger to ancient
monuments which they have unfortunately proved throughout
the peninsula.
While the Nagir side of the valley shows a cheerful succes-
sion of villages, the opposite side, which belongs to Hunza,
is here for the most part a rocky waste. The difference is easily
accounted for by the increased supply of water which Raki-
poshi provides. Among the people of Nagir no marked
difference from the Dard type is noticeable. Shina, the lan-
guage of Gilgit, seems to be spoken in most of the lower
villages, though Burisheski, the language of Hunza, is also
understood. The latter has no apparent connection with
either the Indian or the Iranian family of languages, and
seems an erratic block left here by some bygone wave of con-
quest. In its stock of words it shows no resemblance to the
Turki dialects, but is closely allied to the Wurshki tongue
spoken in the northern valleys of Yasin. How the small race
which speaks the language of Hunza has come to occupy these
valleys will perhaps never be cleared up by historical evidence.
But its preservation between the Dards on the south and the
Iranian and Turki tribes on the north is clearly due to the
isolated position of the country. It was curious to me to watch
the rapid inroads which Hindustani has made in this linguistic
area during the last few years. The few hundred men placed
in garrison along the valley and the passage of the convoys
bringing their supplies have sufficed to spread a knowledge of
LANGUAGE OF HUNZA. 33
Hindustani, or rather Punjabi, among the villagers, which
considering the brief time is quite surprising. In view of this
experience the rapid spread of Arabic and Persian words on
STUPA OF THOL, NAGIR.
the line of early
Muhammadan conquest throughout Asia
becomes
more easily intelligible.
The
constant ups and downs of the road seemed to spread
out
considerably
the distance of twenty-six miles between
Chalt
and Aliabad,
the end of my march. Below the fort
3
34 THEOUGH HUNZA.
village of Tashsot the route crosses the rock-bound bed of the
river by a bold bridge, and then continues along absolutely
barren slopes of rock and shingle for several miles. In the
light of the evening the steep walls of rock rising on either
side fully five or six thousand feet above the river, with the
icy crests of Rakiposhi in the background, formed a picture
worthy of the imagination of Gustave Dore. By the time I
had cleared the worst parts of the road along sliding beds of
detritus it had got quite dark. For two hours more the road
wound round deep side-valleys from the north until I emerged
on the open plateau which bears the village and lands of
Aliabad. Here a little fort had been erected during the tem-
porary occupation of Hunza, and close to it stands the bun-
galow of the Political Officer. Though Captain P.
J.
Miles,
the rightful occupant, was absent on leave, I was able to find
shelter under its hospitable roof. Cheerful enough the httle
luxuries of this frontier-officer's home appeared to me. His
servants too, sturdy Hunza men, knew how to help a belated
Sahib to an early meal and rest.
When I awoke in the morning a view of unexpected grandeur
greeted me. Rakiposhi, seen now from the north-east, reared
its crown of ice and snow more imposingly than ever, and
without a speck of cloud or mist. To the north mighty peaks,
also above 25,000
feet in height, frown down upon the valley,
while eastwards I could see the range along which my onward
route was to lead. The two days which I had saved by the
double marches between Gilgit and Hunza, were used for a
short halt at Aliabad. I required it in order to distribute my
baggage into loads suitable for coolie transport, and also to
dispose of arrears of correspondence, etc. Hunza, it is true,
does not boast as yet of a post office. But a
"
Political Dak
"
connects it every second day with Gilgit, and in view of the
long journey before me it seemed right to utilise to the full
this last link of regular postal communication.
The first morning brought the Mir's Wazir, who came to
assure me of the arrangements that had been made for the
onward journey. Wazir Humayun is no small jiersonage in
the Hunza State, being the chief adviser and executive officer
k
36 THROUGH HUNZA.
of the Mir, which rank he holds by hereditary right. He is
a tall, well-built man of about fifty years, with an imposing
beard, and makes a striking appearance, even in the semi-
European costume he has chosen to adopt, evidently as a
mark of his progressive ideas. It must have been different
in former years, when the Wazir led Kanjuti raids into Sarikol,
Gilgit, and Baltistan. A pleasant fire lit up his eyes as he
talked to me of his expeditions to Tashkurghan and into the
Braldo Valley. Now that the days of fighting are gone he
evidently does his best to develop the internal resources.
It is no easy task, for the cultivable land is far too limited to
provide for the increase of population. Only by elaborate
irrigation can produce be wrung from the rock-strewn slopes
of the valley, and the long courses of
'
kuls ' (water-channels)
winding along the foot of the mountains often in double and
treble tiers, show how carefully the available
supi)ly
of water
from the glacier-fed streams of the side valleys has been
utilised.
Curious, too, was the information about the relations of
Hunza with the Celestial Empire. Hunza people have for
a long time back occupied valleys like that of the Oprang
stream draining into the Yarkand River ; and their continued
occupation of these tracts, which plainly fall within the natural
boundaries of Chinese Turkestan, is probably the reason why
the further periodical transmission of presents to the Kashgar
authorities has been acquiesced in even after the enforcement
of British sovereign rights. On the other hand Hunza enjoys
the benefit of Chinese return
"
presents " considerably in
excess of those sent, an arrangement manifestly representing
the blackmail which the Chinese had to pay to safeguard their
territory between Sarikol and the Karakorum from Kanjuti
raiding. On my enquiring after records of the relations with
the Chinese authorities, the Wazir informed me that a quantity
of documents, mostly Chinese, with Persian or Turki transla-
tions, had been removed from the Mir's residence at Baltit
to Simla, after the occupation in
1891.
It would be interesting
to ascertain from these or from the Chinese archives, what official
status was accorded by Chinese diplomacy to the Kanjuti chiefs.
HALT AT ALIABAD.
37
Though British supremacy in Hunza, very
different from
Chinese fictions, is a thing of manifest reahty, it is maintained
without material force. The httle fort built in the open fields
of Aliabad is now mainly used as a
commissariat
"
Godown,"
and guarded only by a few local levies raised among the neigh-
bouring villagers. Yet these levies, of whom there are about
one hundred and eighty in the state, proved useful during
HUNZA COOLIES, BEFORE START FROM ALIABAD.
the Chitral campaign. As elsewhere along the North-West
border, these local militia supply an excellent instrument for
the political management of their own territory.
Regular
pay and easy service are effective in attaching them to the
ruling order of things. The additional advantage which
levies
on the Afghan border offer for the safe employment of noto-
riously bad characters that would otherwise be likely to give
trouble, need fortunately not be considered in Hunza. The
38 THROUOH HUNZA.
people have been described by those best quahfied to judge
as thoroughly tractable and obedient to constituted authority,
and notwithstanding their old raiding reputation, this descrip-
tion seems fully justified.
On the second day of my stay at Aliabad I received the visit
of the Mir of Hunza, Muhammad Nazim, who had been in-
stalled after the occupation in
1891.
He is a man of about
thirty-five, of open and manly bearing, and evidently deserves
the reputation for intelligence and firmness which he enjoys.
Our conversation, carried on in Persian, turned naturally
more to the old conditions of the country than to the reforms
about which the Mir is said to be energetic. Road-making,
vaccination, and similar Western improvements seem strange
as objects of genuine interest in the representative of a family
for which intrigue and murder were down to the present
generation the main incidents of life. This transformation
in its rapidity and evident thoroughness is a striking proof
of the results of the pax britannica.
Through the Wazir I had engaged two Hunza levies who had
been on the Pamir before, to accompany my camp to Sarikol
as guides. Muhammad Rafi, the commandant of the Mir's
bodyguard, was sent to organise and supervise the transport,
represented by sixty coolies. Swelled by these numbers my
caravan looked alarmingly large as it moved off on the morning
of
June
20th. The first march was only a short one, to Baltit,
the chief place of Hunza, and the Mir's residence. Rising on
a cliff from an expanse of terraced fields and orchards, the
Castle of Baltit looks imposing enough with its high walls and
towers. Below it, closely packed on the hillside, are the rubble-
built houses, some two hundred in number, of the Hunza
capital. The newly built bungalow which received me lies
immediately below the fine polo ground, offering a cheerful
sight with its green turf and shady Chinar trees. On the
opposite southern side of the valley a striking view opened on
the Sumair glacier with a hoary ice peak l)ehind it.
The visit which I paid to the Mir in the late afternoon, gave
me an opportunity to inspect more closely the time-honoured
castle of the Hunza rulers (see
p. 27).
The high, massive
VISIT TO MIR'S CASTLE.
39
walls of the foundation upon which the inhabited quarters are
raised, are said to have been the work of Balti workmen who
came in the train of a Balti princess, and from whom the
place has derived its name, Baltit, From the roof of the
castle where I found the Mir with his numerous retainers,
a superb view extends over the main portion of the Hunza
valley. A newly-built pavihon-like structure
where I was
FORT-VILLAGE OF ALTIT.
subsequently entertained to tea and cake, occupies the same
elevated jx)sition and offers the same delightful prospect.
Notwithstanding some European articles of furniture of doubt-
ful taste which had already found their way to this
apartment,
the whole showed clearly the prevalence of Central-Asian
manufacture. Carpets from Yarkand, Chinese silks and gaily-
coloured prints from Kashgar could indeed make their way to
Hunza far more easily over the Sarikol passes than
Indian
articles before the opening of the Gilgit route.
Even now
40 THEOUGH HUNZA.
the latter is open to trade for a far shorter period than the
passes from the North.
Returning from my visit to this interesting place I noticed
several small mosques constructed of wood, and showing on
their beams and posts a good deal of effective carving.
Rougher in execution than old Kashmir woodwork, it yet
displayed, just like the latter, decorative elements of a dis-
tinctly early Indian type, e.g., the double
'
Chaitya ' orna-
ment, the Sacred Wheel, the Svastika. The work I saw was
said to be of comparatively recent date, which makes the
survival of these patterns borrowed from the South so much
the more curious.
My march on
June 21st looked short on the map, but the
accounts I had collected of it prepared me for its difficulties.
Soon after passing, about two miles above Baltit, the pic-
turesque fort-village of Altit, the valley contracts to a gorge
of rugged rocks, almost without a trace of vegetation. A
narrow path winds along the cliffs, sometimes close by the
swollen river, sometimes several hundred feet above it. A
small alluvial plateau, reached some four miles beyond, bears
the little village of Muhammadabad, But the track leads
far below over the sandy bed of the river. This bed indeed
forms the easiest route up the valley, and only when^its water
is low in the winter can ponies be brought up or down. The
frequent crossing of the river which this winter route neces-
sitates is altogether impossible when the snow on the mountains
has once begun to melt.
Accordingly a high rugged spur had to be climbed and the
d6bris of an enormous old landslip to be traversed before I
could descend again to the riverside and reach the camping-
ground of Ataabad. The hamlet which gives this name was
scarcely to be seen from below, and shut in by an amphi-
theatre of absolutely bare rocky heights, our halting-place
looked a dismal spot. About half a century ago the Gham-
mesar landslip, already referred to, is said to have blocked
the whole valley, when from Ataabad upwards an enormous
lake was formed. The black glacier-ground sand, which the
Hunza River brings down and deposits in large quantities,
CLIMBS OVER RAFIKS. 41
rose in thick dust with the wind which blew down the valley
in the evening. Drink and food tasted equally gritty ; it
seemed a foretaste of the Khotan desert. In so desolate a
neighbourhood I felt doubly grateful for the Dak-runner who
at nightfall brought a long-expected home mail.
The march of the ne.xt day proved a trying experience. A
short distance above Ataabad the river passes along a series of
cross-spurs which at their foot are almost perpendicular.
So the path climbs up their sides, and clings to them where
they are too steep by means of narrow galleries. These are
carried in parts over branches of trees forced into fissures of
the rock and covered with small
stones. Elsewhere narrow natural
ledges are widened by flat slabs
packed over them. In some
places these galleries, or
'
Ra-
tiks,' as they are locally called,
turn in sharp zigzags on the
side of chffs where a false step
would prove fatal, while at
others again they are steep
enough to resemble ladders.
To carry loads along these
galleries is difficult enough,
and for cattle as well as
ponies, surefooted as the
latter must be in Hun-
za, they are wholly
impassable. At
more than one
place even
'
Yolchi Beg,'
my little ter-
rier, had re-
luctantly to
submit to the
indignity of
RAFiK ABOVE ATAABAD.
bcing Carried,
42
THROUGH HUNZA.
though on our cHmbs in Kashmir I had found few rocks
that would refuse him a foothold. Scrambles of this kind
alternated along the whole march with passages over
shingly slopes and climbs over rock-strewn wastes. Only at
a few spots the barren grey and yellow of the rocks was
relieved by some green shrubs growing where scanty water-
courses forced their way down the fissured slopes. After six
hours' steady climbing and scrambling it was a relief to see
at last the valley widen again, and two hours more brought
me to Ghulmit village. It occupies a wide alluvial fan on the
flank of a considerable glacier, the white crest of which could
be seen from a distance rising above the orchards and fields.
At Ghulmit that part of the Hunza Valley is entered which
is known as Little Guhyal. It takes this name from its
inhabitants, Wakhi immigrants from Wakhan or Guhyal on
the Oxus. It was easy to notice the change of race in the
assembly of well-built handsome village headmen which
received me some distance from the village. Headed by the
Mir's relation, Muhammad Nafiz, who acts as his representa-
tive among the villages of this part of the valley, they escorted
me in stately procession to the little orchard of apricot trees
where my camp was to be pitched. I was delighted to hear
at last the language of Wakhan, which had attracted my
attention years before I first came to India, as a remarkably
conservative descendant of the ancient tongue of Eastern Iran,
It seemed strange that I should have first touched the lin-
guistic borders of old Iran, high up in these mountains. The
fact was bound to remind me that the Pamirs which I was
about to approach, mark the point of contact not only of great
geographical divisions, but also of equally great language
families and of the races speaking them. Close to the Kilik
Pass is the point where the watersheds bounding the drainage
areas of the Oxus, Indus, and Yarkand Rivers meet ; and it is
plain that as far as history can take us back, these areas
belonged to the sphere of the dominant races of Iran, India,
and Turkestan.
The Wakhis of Little Guhyal, numbering altogether about
a thousand souls, are a line stalwart type, taller than the men
WAKHI SETTLEMENTS.
43
of Hunza and usually showing clear-cut and intelligent features.
The characteristic eagle-nose of the true Iranian was well
represented, and their complexion, too, seemed to me distinctly
fair. Many of them talk Persian with more or less fluency,
and I was thus able to indulge in short chats. The
connection
with the people of Wakhan and Sarikol is still
maintained by
WAKHI VILLAGERS, GHULMIT.
occasional marriages, and the original immigration from the
Oxus Valley is distinctly remembered. How the Hunza people
proper, undoubtedly more warlike and so pressed for land,
acquiesced in this invasion, seems difficult to explain. The
peaceful character of the Wakhis is curiously s}Tiibolised by
the implement which every respectable householder carries
about with him on state occasions. It is a long staff with
a small heart-shaped shovel of wood at the end, used for
k
44 THEOUGH HUNZA.
opening and damming up the irrigation courses that bring
fertiHty to the laboriously cleared terrace lands.
Ghulmit cannot have seen many Sahibs, for a large assembly
of villagers remained for a long time round the neat little fruit
garden where I was encamped. Next morning we made a late
start owing to the change of coolies, when time is always lost
until every one settles down to the load he fancies. But the
VIEW TO NORTH-EAST OF PASU VILLAGE.
march to Pasu proved short, and after the previous days'
experience unusually easy. This does not mean, of course,
that the track is as yet fit for perambulators. For a short
distance above Ghulmit the Ghulkin glacier comes down close
to the river, and the numerous channels in which its ash-grey
waters rush forth, are troublesome to cross at this season.
But the valley is open, and the stony plateaus along the right
riverbank afforded easy going. Just
before the end of the
march the road passes in front of the Pasu glacier, which comes
BATUR GLACIER. 45
down with its debris-covered masses of ice from a great peak
of over
25,000 feet, also visible from Aliabad and Baltit. An
enormous side moraine, which is crossed by the route, shows
that the glacier must have advanced further at a former period.
The little village of Pasu, situated immediately to the north
of the glacier-head, formed with its green fields and orchards
a pleasant contrast to the bleak scenery around. It owes its
BATUR GLACIER, SEEN FROM SOUTH-EAST.
existence to the irrigation cuts which catch some of the water
issuing from the glacier. A little orchard in the midst of the
few scattered homesteads which form the village, was my
cheerful camping-ground for the day. The cooler air and
the backward state of the crops of oats and millet were indi-
cations of the elevation of the place (circ. 8,000 feet above
the sea). The flowers by the side of the fields, scanty as they
were, gave the whole a springlike look which was most pleasing.
46
THEOUGH HUNZA.
HUNZA VALLEY BELOW KHAIBAR.
The inarch of
June
24th brought me first to
the huge Batur glacier,
some three miles above
Pasu. Probably over
twenty-four miles long,
it fills completely a large
side valley which des-
cends from the north-
west, and unlike the glaciers previously passed, it advances
its frozen walls down to the river-bed. They are covered
for miles up the valley with an extraordinary mass of
detritus, and thanks to this thick crust of rock and
shingle the crossing of the glacier was comparatively easy.
All the same it took me nearly an hour to scramble
across the mile and half of glacier, and the slippery ground
delayed the coolies still longer. There are years' when masses
of ice pushed down from the unexplored upper reach of the
glacier make the crossing far more difficult even for men,
and altogether close the route for animals. It is in view of
such obstacles, which no skill of the engineer can ever com-
pletely overcome, that one realises the great natural defences
of the Hunza Valley route against invasion from the North.
Above the Batur glacier the valley contracts and continues
between bare walls of rock and shingle to Khaibar, the next
inhabited place above Pasu. The river, no longer fed by the
MARCHING OF KANJUTIS. 47
glacier streams from the high ranges, is now far smaller in
volume, yet still quite unfordable in summer. The mountains
on either side culminate in serrated rock pinnacles of fantastic
forms, but \'ie\vs of mighty masses of ice and snow no longer
meet the eye.
The hamlet of Khaibar, which I reached after a tiring
march of six hours, lies on an alluvial fan at the mouth of a
narrow side valley. Scanty indeed are the fields of the place,
and one wonders how they can support even the half-dozen
homesteads. Yet even here where Nature is so harsh, defence
against human foes was not so very long ago a necessary
condition of existence. The path which leads to the plateau
is guarded at a point of great natural strength by
a rude gateway or
'
Darband,' a necessary jmt
"
precaution seeing that the opposite bank of the ?^^
river was easily accessible to the people of
Nagir, the hereditary enemies of Hunza.
'
From Khaibar to Misgar
there are two
routes available, one leading
through
the
hamlet of Gircha by the left bank of the
river, and the other through
Khudabad
on the right. The former, which was
said to be easier if the water of the river
was not too high, was reported im-
practicable soon after I had
started on the morning
of
June 25th. Hence the
track on the right bank had
to be taken. Without offer-
ing exceptional difficulties
that day, it was trjdng
enough, leading almost the
whole length over boulder-
strewn slopes and along
banks of slatey shingle.
Just opposite to the
hamlet of Murkhun,
RAFIK NEAR MURKHUX.
48 THROUGH HUNZA.
where a route to the Shimshal Valley opens eastwards, the path
descends over a long Rafik built out in the usual fashion from
an almost perpendicular rock face. Curiously enough at one
point of the narrow ledge which bears the gallery, there issues
a little spring of deliciously clear water, offering welcome
refreshment to the wayfarer.
Not far beyond I met, to my surprise, the messenger whom
the Wazir of Hunza had despatched to Tashkurghan to notify
to the Political Munshi there my approaching arrival. The
man had left Hunza on the morning of the i8th, and now he
was returning with the Munshi's reply and a considerable
load of merchandise which he was bringing back as a private
venture. As an illustration of the marching powers of the
men of Hunza this feat deserves record. The distance from
Hunza to the Kilik is about eighty-one miles, and of the
character of the track my experiences so far described will
suffice to give an idea. In addition to this and half the return
journey, the man had covered twice the route along the Tagh-
dumbash Pamir to and from Tashkurghan, a distance of at least
eighty miles each way. Performances of this kind make it
easy to understand how the raids of Kanjuti parties could be
carried to so great distances, and thanks to the rapidity of their
movements, usually with impunity.
At Khudabad, a hamlet of eight houses, my day's march
ended. Here I passed once more out of the Wakhi area into
that of small Hunza settlements. The fact reminds me of
the strange variety of tongues which at that time could be
heard in my camp. Apart from Turki conversation with my
personal servants, Persian served me as a convenient medium
with my Wakhi guides and the more intelHgent villagers. My
coolies spoke partly Wakhi, partly Burisheski, while the Dard
dialect of the Shinas was represented by
"
Raja
"
Ajab Khan,
a relative of the hill chiefs of Punyal, whose services as an
orderly Captain Manners Smith had kindly secured for me,
and by his retainer. In addition to these languages there was
Hindustani talked between my Sub-Surveyor and Jasvant
Singh, his Rajput cook. Had I brought the Kashmiri servant
whom I had first engaged before Sadak Akhun joined me from
CLIMBS TO MISGAR. 49
Kashgar, I should have had an opportunity to keep up my
Kashmiri also. Notwithstanding this diversity of tongues
things arranged themselves easily, for everybody seemed to
know something at least of another's language.
The march from Khudabad to Misgar which I did on the
26th of
June had been described to me as the worst bit of the
route, and as an Alpine climb it certainly did not fall- short
of the estimate I had been led to form of it. The
Chaparsun
River,
which comes down from the glaciers near the Irshad
#-ii
and Chillinji
Passes in the north-west, was fortunately
low at
the early morning hour, and could be forded immediately
above Khudabad. A long detour and the use of a rope bridge
were thus avoided. But the succession of cUmbs which followed
in the main valley beat all preNnous experience. Scrambles
up precipitous faces of slatey rocks, alternated with still more
trying descents to the river-bed
;
'
Rafiks
'
and ladders of the
t5rpe already described were in nmnerous places the only
possible means of getting over the cliffs, often hundreds of feet
above the river. The previous five days, however, had accus-
50 THROUGH HUNZA.
tomed me somewhat to such modes of progress, and it was in
comparative freshness that 1 emerged at last in the less con-
fined portion of the valley above its junction with the gorge of
the Khunjerab River. Some miles before Misgar I was met
on a desolate little plateau by the levies of that place, a re-
markably striking set of men, and conducted to their village.
After the barren wilderness of rocks and glacier streams
through which I had passed, the smiling green fields of Misgar
were a delight to the eye. They are situated on a broad
plateau some
300
feet above the left river-bank, and amply
irrigated by channels fed from a stream of crystal-clear water
which issues from a gorge to the east. The millet and
'
Rishka
'
were still in young shoots, since the summer comes late at this
the northernmost village of the valley. In the midst of the
fields and the scattered homesteads I found an uncultivated
spot just large enough for my tent, and enjoyed again the
pleasure of camping on a green sward. Close by was the
Ziarat of a local saint, Pir Aktash Sahib, a simple enclosure
adorned with many little flags which fluttered gaily in the
wind, just as if they marked the approach to a Buddhist estab-
lishment in Sikkim or Ladak. The open view across the broad
valley was most cheerful after the gloomy confinement of the
previous camping grounds. Far away to the north-west I
even beheld a snowy ridge which clearly belonged to the water-
shed towards the Oxus. I felt at last that the Pamir was near.
At Misgar I was able to discharge the hardy hillmen who
had carried our impedimenta over such trying ground without
the slightest damage, and on the morning of June
27th I moved
on with fresh transport. This consisted chiefly of ponies, as
the route further on is open to baggage animals at all seasons.
Though the road no longer offered special difficulties, it was
tiring owing to the boulder-strewn wastes it crosses for a great
part. At Topkhana, where there stands a half-ruined watch-
tower amidst traces of former habitations and fields, I was mot
by a jolly-looking
young Sarikoli, whose aj)pcarance and outfit
at once showed that he came from Chinese territory. It was
one of the soldiers of the
'
Karaul
'
or guard kej^t by the
Chinese on the Mintaka Pass who had been sent down to
4*
52 THKOUGH HUNZA.
inquire as to my arrival. He carried a long matchlock with the
gable-ended rest sticking out beyond it, an indispensable
implement of the Celestial soldiery of the old type all through
the empire. Ruddy-cheeked and clothed in fur cap, mighty
boots, and a series of thick
'
Chogas ' or coats, the young
fellow looked serviceable enough. Less so his matchlock,
which had lost its breach-piece, and in the barrel of which
a broken ramrod had stuck fast evidently for many a long
day. He assured me that the expected yaks and ponies were
already waiting for me, and tried to make himself as useful
on the rest of the march as if he belonged to my following of
Hunza levies.
In reality the frontier line seems of little consequence to
the Wakhi herdsmen who live on either side of it. When after
a march of over twelve miles I arrived at Murkushi, where
the routes to the Kilik and Mintaka diverge, there was a set
of picturesque Wakhis from across the border waiting for me.
They had left their yaks on this side of the pass, where they
found better grazing. It was a pleasure to behold these sturdy
fellows in their dresses of Yarkand fabrics showing all colours
of the rainbow. Their clear-cut Iranian features, almost
European in complexion, seemed to contrast pleasantly with
their Kirghiz get-up. Down in the little wood of stunted
birch-trees by the river where I camped for the night, it was
scarcely as cold as might be expected at an elevation of nearly
12,000 feet. At 6 a.m. on the following morning the ther-
mometer showed 47 F.
A march of four hours brought me on the 28th of
June
to the
high grazing ground known as Shirin Maidan
("
the Milky
Plain "), close to the foot of the Kilik Pass. Here the change
in the temperature due to the great elevation made itself
most perceptible. When the sun i)assed
behind light clouds
at noon and a fresh breeze blew down the pass it was bitterly
cold, and I was glad to get into my fur coat as soon as the
baggage arrived.
The range immediately to the north which is crossed by the
pass, appeared low by the side of the rugged peaks which
show their snowy heads further down in the valley. More
II
CAMP AT SHIRIX MAIDAN.
53
imposing than the watershed towards the Taghdumbash, looked
a distant glacier-covered ridge \'isible through a side valley
westwards. Behind it lay the sources of the Oxus, or more
exactly of the Ab-i-Panja branch.
My Guhyal coolies and Hunza levies had now all been dis-
charged, and I was left to enjoy the change in my camp sur-
roundings.
Muhammad Yusuf, the Sarikoli headman, and
his seven relatives who brought the yaks that were to take my
baggage onwards, were cheerful to look at and talk to. They
understood Turki quite well and were most communitative.
In their midst I felt that I had passed out of India.
yLi i
V
YAKS STARTING FORIkILIK PASS.
CHAPTER IV.
ON THE TAGHDUMBASH PAMIR.
When early on the morning of the 2gth of June
I struck camp
to move over the Kihk or
'
Kahk
'
Pass, as it is called by
Kanjutis, the ground was covered with hoar frost and the little
streams which came down from the pass were partly frozen.
I tried to start early in order to find the snow still hard
;
but the
packing of the baggage on yaks proved a lengthy affair, and it
was not till 8 a.m. that the caravan moved off. I had the satis-
faction of seeing the servants whom the previous marches had
tried a great deal, now comfortably mounted on yaks. Tin-
ascent lay northwards through a comparatively open though
steep nullah for about an hour. Then the ground widened, and
the flat watershed still covered with snow came into view. On
the cast the
i)ass
is flanked by spurs of a rugged peak, which
rises to a height of nearly 20,000 feet. On the west two small
glaciers stretch down to it from a somewhat lower range, the cul-
CROSSING OF KILIK PASS. 55
minating peak of which seems to mark the point where the
drainage areas of the Oxus, Indus, and Yarkand Rivers meet.
On the flat plain, about half a mile broad, which forms the top
of the Kilik, it was not easy to fix the actually lowest point, the
true watershed. When I had ascertained the spot that looked
like it, a halt was made to boil the water for the hypsometer.
It proved a troublesome business in the bitterly cold wind which
was blowing across, and by the time that I got the readings
which gave the height as circ. 15,800
feet, it began to snow.
Bleak and shrouded in clouds looked the range to the north,
which marks the boundary of the Russian Pamirs, but there was
nothing striking in its outlines, nor was the amount of snow as
great as on the serrated high peaks towards Hunza. The ride
down in the soft snow and in the face of the cutting wind was
not a pleasant experience, but the yaks proved most useful as
snow-ploughs, and by i p.m., after a descent of over two hours.
56 ON THE TAGHDUMBASH PAMIR.
I found myself at Kok-torok
("
the Blue Boulder," in Turki)
on the fiat of the Taghdumbash Pamir.
An imposing cavalcade met me as I approached the place
where my camp was to be pitched. Munshi Sher Muhammad,
the Political Munshi stationed at Tashkurghan under the orders
of Mr. Macartney, had come up from his post to greet me
;
and
attracted, no doubt, by his example, the Sarikoli Begs in charge
of the several portions of the country above Tashkurghan, also
awaited my arrival. Munshi Sher Muhammad, a fine-looking,
active man, introduced himself as an old pupil of the Oriental
College at Lahore of which I had held charge so long. The
arrangements he had made for my journey down to Tash-
kurghan were all that could be desired and showed his influence
with the local authorities as much as his eagerness to help me.
It was bitterly cold during this first day on the Taghdumbash
as, soon after my arrival, a strong wind sprung up blowing across
the valley from the north-east and bringing light snow at
intervals. The observation of Captain Deasy, who had encamped
at the same spot in
1897,
shows that its elevation is close on
14,000 feet.
On the 30th of
June
the sun shone brightly when I rose, and
though the temperature at 6 a.m. was only 37 F. in the shade,
it felt pleasant enough as the air was still. The surrounding
ridges, all snow-capped, stood out with perfect clearness against
the blue sky. The conditions were exceptionally favourable for
the survey work which was to be commenced here, and by 8
a.m. the surveyor and myself were on our way to the top of the
Khushbel spur which was to serve as a station. This spur descends
from the high range on the east of the Kilik Pass towards the
valley, and by its detached position offers an extensive view
over the upper portion of tlie Taghdumbash. We were able
to ascend close to its top, 16,820 feet above the sea, by means
of yaksan advantage which, in view of the subsequent work,
was not to be despised. The way in which the sure-footed
animals carried us and our instruments steadily up, first over
steep grassy slopes, then over fields of snow, and finally over the
shingly beds of rock, was to me a novel and gratifying exjieri-
ence. It^was clear that by a judicious use of the yak the difii-
r
WATERSHED TOWARDS OXUS. 57
culties which the high elevations offer to mountaineering in
these regions could be reduced for the initial stages. From the
top of Khushbel we were able to identify some peaks both
towards the Murghab V'alley and Hunza which had been trian-
gulated by Captain Deasy. While Ram Singh was busy with
his plane-table, I did my first work with the Bridges-Lee photo-
theodolite, an excellent instrument, which was now on its first
SNOWY RANGE SOUTH OF HEAD OF AB-I-PANJA VALLEY.
trial in Central Asia. By noon the wind began to blow again,
which seems a regular feature of the atmospheric conditions at
this season, and I was glad when by 6 p.m. the shelter of the
tent was reached.
Koktorok is so near to the Wakhjir Pass, which marks the
watershed between the Oxus and the Yarkand River drainage
systems, that I could not resist the temptation of visiting it
during the two days which were
required for the Sub-Surveyor's
58 ON THE TAaHDUMBASH PAMIR.
work round this camp. It would have weighed on my topo-
graphical conscience to have passed by without seeing at least
the head of the Wakhan Valley and the glacier which Lord Curzon
first demonstrated to be the true source of the Oxus. Accord-
ingly, leaving all heavy baggage with the Sub-Surveyor's party
at Koktorok, I set out on the morning of the ist of
July
towards
the Wakhjir Pass. The road led first up the open valley towards
PHOTO-THEODOLITE VIEW OF OXUS SOURCE GLACIERS.
the west, and then after some five miles turned into a narrower
side valley in a south-westerly direction. Large patches of
snow and the gradual disappearance of the thick, coarse grass,
which was to be seen round Koktorok Camj), marked the higher
elevation. I pitched my tent at the jioint which offered the last
bit of comparatively dry ground, circ.
15,300
feet above the sea.
Higher uj) there was snow at the bottom of the valley, or boggy
soil where the snow had just melted. In front I had the view of
OXUS SOURCE GLACIERS. 59
numerous small glaciers, which clothes the slopes of the range
south of the pass. My intention of going up to the
latter the same day was frustrated by a storm which
brought sleet and snow. In the cutting cold -my people
felt the scarcity of fuel ; for even the coarse grass known to
the Sarikolis as
'
Dildung ' and to the Kirghiz as
'
Burse,'
the dry roots of which supply the only fuel of this region

VIEW DOWN AB-I-PANJA VALLEY FROM NEAR WAKHJIR PASS.

apart from dry yak dungwas no longer to be found at this
altitude.
By the morning of the next day the weather had cleared,
and the ascent to the pass could be effected without difficulty.
One and a half hour's ride on a yak over easily sloping snow
beds and past a small lake brought me to the watershed. It
was clearly marked by the divergent direction of the small
streams which drained the melting snow
;
and the
hypsometer,
60 ON THE TAGHDUMBASH PAMIR.
which I boiled on a boulder-strewn patch of dry ground close
by, gave the height as close on 16,200 feet.
A glacier of pure white ice pushes its tongue to within a few
hundred yards from the north. The descent to the west of the
pass took me into Afghan territory, but in this mountain solitude
there was no need to consider whether this short inroad into His
Highness the Amir's dominions was authorized or not. The
soft snow impeded my progress for about a mile and a half, but
then the ground got clear, and I was able to follow without
trouble the stream from the pass down to where it joins the far
greater one which drains the glaciers at the true head of the
Wakhan Valley. A climb of some eight hundred feet up the
mountain-side to the north gave me a splendid view of the
valley through which the collected waters of the Ab-i-Panja
flow down towards Bozai Gumbaz and Sarhad. The glaciers,
too, from which they chiefly issue, were clearly in view. An
hour's work with the photo-theodolite enabled me to retain the
whole of this impressive panorama. It left no doubt as to Lord
Curzon being right in placing here the true source of the great
river.
The high ranges which line the valley precluded a view
further west towards Wakhan proper and Badakhshan. Yet
it was a strange and joyful sensation to know that I stood at last
at the eastern threshold of that distant region, including Bactria
and the Upper Oxus Valley, which has had a special fascination
for me ever since I was a boy. How I wished to have been able
to follow the waters of the Oxus on their onward course ! All
the interests of ancient Iran cluster in one form or the other
round the banks of the great stream. Since the earliest times it
has brought fertility and culture to the regions which it waters.
Here at its source there was only a silent, lifeless waste of rock
and ice. Yet I found it hard to leave this desolate scene.
The evening glow was spreading over the valley when I
retraced my route to the pass, and it was dark before I returned
to camp. I found there to my delight an eagerly-expected
home mail, which the attentive Wazir of Hunza had sent by
special messenger over the Kilik. With it came a batch of the
latest telegrams of Reuter, which were to be forwarded to Mr.
A CHANCE MEETING. 61
Macartney at Kashgar after perusal. They brought news of
the attack on the Peking Legations and of the fighting about
Tien-tsin. It was strange to read here at the westernmost
extremity of the Chinese Empire of the events which had con-
vulsed its capital in the far East scarcely more than a week
before. I thought it fortunate indeed at the time that this dis-
turbing news would probably take months to reach the popu-
KIRGHIZ 'AK-UIS' AT TIGHARMAX-SU,
lation of the outl\-ing pro\-ince of Kashgar. And I felt still
more grateful for the time-honoured decentralisation of the
Celestial Empire which made any immediate influence of those
troubles on Chinese Turkestan and on my programme of explora-
tions appear distinctly improbable.
On the 3rd of
July
I marched back to Koktorok, and haNnng
picked up the Sub-Sur\'eyor and heavy baggage, continued in
the broad, grassy valley down to Tigharman-su.
There I camped near two Kirghiz felt huts or
'
Ak-uis
'
62 ON THE TAaHDUMBASH PAMIR.
pitched by Muhammad Yusuf's people, who graze their flocks
of sheep and yaks here during the summer. On the following
morning appeared Karakash Beg, the Sarikoli headman in
charge of the Mintaka route, to escort me further down the
valley. We were nearing, after a march of some six miles,
the post at the northern foot of the Mintaka Pass, when a report
was brought to my guide of a
"
Russian officer," who had just
reached the post via the Payik Pass from the north. Having
heard nothing before of such a visitor being expected, I rode up
with some curiosity, and soon found myself face to face with the
reported arrival. It was a young German officer. Lieutenant
F., of the Bavarian Foot Guards, who had just travelled down
from the head of the Russian railways in Farghana, and was
now intending to make his way to Gilgit and India. He knew
nothing of the special permission of the Indian Government,
without which the Hunza route is closed to European travellers,
and was also surprised to hear of the time required for the journey
down to Kashmir. Finding that his leave would not suffice for
this extension of his trip, Lieutenant F. there and then, while
refreshing himself at the breakfast my men had soon got ready
for us, made up his mind to visit Kashgar instead. Accepting
my invitation to share my camp, he accompanied me to Sarik-
Jilga,
the end of my march.
On the way, and then at table, my young guest told me
much of interest concerning his ten days' ride over the Russian
Pamirs. Though far too rapid for close observation, it was a
performance highly creditable to his endurance. Of outfit and
provisions he had brought scarcely more than is wanted for a
few days' outing in the Bavarian Alps, but he had soon been
obliged to provide himself against the rigours of a Pamir summer,
for which he was little prepared, by purchasing a large fur coat
off the back of a Kirghiz. I wondered inwardly how he managed
to get rid of the livestock likely to be involved in this transaction.
As the Kirghiz had so far been his only hosts except at the
Russian fort of Pamirski Post, and as he could not make himself
readily understood by them, his bodily wants had found but
scant satisfaction. His two ponies were also nearly done up
liy
the hardships of these precipitous marches. On the other hand,
KIRGHIZ AT DAFDAR.
63
there was no need for the two revolvers which he was carrying in
his belt, and after our conversation he soon found for them a less
prominent receptacle. For, indeed, if the Pamir region does not
yet offer inns and rest-houses after the fashion of the Alps, it
may boast of an equal degree of security.
Comparing notes from north and south we spent a cheerful
WAKHIS AND KIRGHIZ AT DAFDAR.
evening
together.
Karakash Beg and his followers shared my
satisfaction
at this chance meeting. For the assurance that the
unexpected
arrival
was after all not a
'
Rus
'
relieved them of
11 responsibihty.
On the 5th of
July
we rode down together
-ome fifteen
miles to Ghujakbai (the
'
Ujadbhai
'
of former
maps),
where the valley turns to the north and considerably
widens.
The
snow-capped
ranges on both sides now receded,
and the
widening
expanse of the valley vividly demonstrated
64 ON THE TAGHDUMBASH PAMIE.
the importance which belongs to the Taghdumbash from ancient
times as a great natural thoroughfare over the
"
Roof of the
World." Here my newly-found companion left me in order to
hurry onwards to Tashkurghan. He had been fitted out with
what was needed in the way of tinned provisions, etc., in order
to take him comfortably through to Kashgar
;
and M. Sher
Muhammad, for whom I gave him a letter, subsequently secured
for him the change of animals and the passport of the local
Chinese commandant which were required for his further pro-
gress.
My way on
July
6th lay from Ghujakbai first over a broad
alluvial plateau which stretches for miles up the valley of the
stream coming from the Khunjerab Pass. As it approaches
the Taghdumbash Darya it spreads out fan-wise, and resembles
most closely the
'
Karewa
'
plateaus which form so characteristic
a feature of the Kashmir Valley. A ride of some five miles
across this barren waste brought me to Dafdar, where, near a
couple of
'
Ak-uis,' I found a picturesque assembly of Wakhis
and Kirghiz awaiting me. The latter had come from Pisling, a
small settlement across the river. By the side of their stalwart
and handsome Wakhi neighbours they looked somewhat insig-
nificant
;
but their cheerful expression and joviality amply made
up for the defects of stature and countenance. A short distance
below Dafdar I came upon the first traces of cultivation. On
the scattered fields which little channels from a side stream
irrigate, the crops of oats and barley have evidently a hard
struggle. All the same it was a pleasure to meet again with this
evidence of permanent habitation. It is only during the last
ten years that the latter has become possible, since Kanjuti raids
have ceased and order has been secured for the valley.
I It was a novel sensation, after the weeks passed in narrow
gorges and amidst snow-covered
heights, to ride along these
broad, smiling slopes gently descending from the foot of the
mountains. Wherever water reaches them from the side valleys,
the ground was covered with a carpet of flowers and herbs which
scented the air quite perceptibly. When about midway of the
march I made a short halt on the green meadows of Ghan, a
summer grazing-ground, 1 could easily imagine myself enjoying
I
APPROACH TO TASHKURGHAN. 65
a bright summer day on a Hungarian
"
puszta." A troop of
ponies tinned loose to graze around were lustily enjoying the
delights of freedom and rich pastme. To watch their lazy,
happy ways was a pleasant distraction.
Light, fleecy clouds himg over the mountains, and it was
only in the afternoon when approaching the end of my march
of some eighteen miles that I could perceive, rising above them
in the north, the glistening mass of a great sno\vy dome. This
was Muz-tagh-Ata,
"
the Father of Ice Mountains," which I
had so long wished to behold. At Yurgal Gumbaz, where I
pitched my camp by the side of the river now grey and swollen
with the water of glacier streams, it was distinctly warm until
the wind began to blow up the valley.
Next morning we were ready for an early start, for the
neighbourhood of Tashkurghan and such comforts as it could
offer was an attraction for my people no less thcin myself. Muz-
tagh-Ata, stiU so distant, showed itself in fascinating clearness
during the early horns of the morning. Its grand dome of ice
filled the vista behind the north end of the valley. After a few
miles' ride over a stony level
'
Dasht,' my guide, Rashid Beg,
the Ming-bashi
("
head of a thousand men
")
of Tashkurghan,
broke his usual silence, and indicated
a
white spot in the far
distance
as the goal of our march. It was the Fort of Tash-
km-ghan,
rising over the western bank of the river. Then I
reached a strip of deUghtfully green sward stretching along the
irrigation
channel which carries the water of the river to the
fields of Tughlanshahr,
the collection of hamlets opposite Tash-
kurghan.
For miles the path winds along it, and ultimately
reaches
the fertile tract where the water spreads itself over care-
fully-terraced
fields.
Whether it was the bright surroundings or the historical
interests
associated with the place, the sight of the walls of
Tashkurghan
rising higher and higher above the flat filled me
with
emotion.
I knew that they did not hide imposing struc-
tures or special comforts. Yet they marked the completion of a
considerable
part of my journey and my entry upon the ground
which
was to occupy my researches. The swoUen state of the
river
prevented the use of the nearest route, and I had to descend
66 ON THE TAaHDUMBASH PAMIR.
almost to the foot of the spur which projects into the valley below
Tashkurghan from the eastern range, before a practicable ford
was reached. Even here, where the river spreads in about half-
a-dozen branches over the flat meadow land, the crossing was
no easy matter. For the water reached almost up to the saddles,
and flowed with great rapidity. At last, however, though wet
to the waist, we got safely across, and leaving the care of the
baggage to the village headmen who had assisted me in the
passage, I galloped over the rich meadows towards the foot of
the cliffs on which the fort stands.
M. Sher Muhammad awaited me near the comfortable Kirghiz
'
Yiirt ' (felt hut), once belonging to Major F. E. Younghusband,
which he had pitched for my accommodation, and which in the
meantime had proved useful for my fellow-traveller of the
previous days. The news that the Chinese Amban of the Sarikol
district raised no objection to my proceeding westwards of Muz-
tagh-Ata was a welcome piece of intelligence. Less so that Mr.
Macartney's Dak for Gilgit, with which I had hoped to post my
Europe mail, had already started by the route on the left river-
bank, and had consequently missed me. Fortunately it is easier
to rectify such postal mishaps in Central Asia than in civilised
Europe, and after an evening busily spent in writing, a special
messenger rode off with my own mail bag, which was to catch
up the Dak courier before he had started from his first night's
quarters.
CHINESE FORT WITHIN RUINED TOWN OF TASHKDRGHAN.
CHAPTER V.
IN SARIKOL.
The
8th of
July and the day following were given up to a halt
at
Tashkurghan.
There were not only fresh supplies and trans-
port to be arranged for, but also much information to be collected
on
points
of historical and archaeological interest. For Tash-
kurghan,
the chief
place of the mountain tract kno\sTi as Sarikol,
is
undoubtedly
a site of considerable antiquity. Its importance
reaches
back
to the days when the traders from the classical
West
exchanged
here their goods for the produce of ancient
China.
As far
as loccd observations go, everything tends to
support
the
view
first expressed by Sir Henry Rawlinson, that
Tash-kurghan,
"
the Stone Tower," retains the position as well
as
the
name
of the >J6iv<y; inpyos, which Ptolemy, and before
ik
68 IN SARIKOL.
him the great geographer, Marinus of Tyre, knew as the empo-
rium on the extreme western frontier of Serike, i.e., the Central
Chinese Dominions. Nature itself has marked the site not only
as the administrative centre for the valleys forming the Sarikol
region, but also as the most convenient place for trade exchange
on an ancient and once important route connecting great por-
tions of Central Asia with the far West and East. From Tash-
kurghan the road lies open equally to Kashgar and Khotan, and
thus to both the great routes which lead from Turkestan into the
interior of China. Here also the two best lines of communica-
tion across the Pamir converge. The Taghdumbash Valley,
giving direct access to the Upper Oxus, is met by the route
which crosses by the Naiza-Tash Pass into the Aksu Valley and
thence by the Great Pamir leads down to Shighnan and
Badakhshan.
At Tashkurghan I had the satisfaction of finding myself once
more on the track of Hiuen-Tsiang, the great Chinese pilgrim,
whose footsteps I had traced to so many a sacred Buddhist site
of ancient India. Travelling about a.d.
649
from Badakhshan
to Khotan, he passed through the district of Kie-pan-to, long ago
identified by Sir Henry Yule as the modern Sarikol. Examin-
ing on the spot the description he and the earlier Chinese pilgrim,
Sung-yun (circ.
500
a.d.), give of the old capital of that territory,
I found it to agree most closely with the position and remains
of Tashkurghan. The ruined town, within which the modern
Chinese fort is built,
"
rests on a great rocky crag and is backed
by the river Sita
"
{i.e., the Taghdumbash branch of the Yarkand
River), on the east, exactly as the pilgrims describe it.
A line of massive but crumbling stone walls crowns the edges
of a quadrangular plateau of conglomerate cliffs, roughly one-
third of a mile in length on each of its faces. A small jiortion
of the area thus enclosed, on the east side facing the river, is
occupied by the Chinese fort. Its high and carefully plastered
walls of sun-dried bricks stand undoubtedly on far more ancient
foundations. Outside them now all is silence and desolation.
The rubble-built dwellings, whose ruins fill part of the area,
were tenanted as long as the insecure condition of the valley
made it impossible for the scanty cultivators to live near their
1
RUINS OF TASHKURGHAN. 69
fields. Since peace has come to Sarikol new villages have sprung
up near all the cultivated patches of land, and the stronghold
has become deserted. WTien the earthquake of
1895
shook
down most of the dwellings, there was no need to rebuild them.
The walls of the town had already suffered by earlier earthquakes,
and show in many places wide gaps as if they had been breached.
Rebuilt undoubtedly again and again after successive periods of
neglect, and always of unhewn stone, they cannot afford any
distinct criterion of age. But the high mounds of debris over
which the extant walls rise, in some places to a height of over
25
feet, show plainly that these fortifications mark the lines
of far more ancient ones.
In order to prove my identification of these and other old
remains, such as that of a ruined Stupa, just beyond the north
wall, an exact survey of the site was essential. To make it
required some diplomatic caution, as the Chinese commandant
or his subordinates might easily have mistaken its object. M.
Sher Muhammad's local experience obviated any trouble on
this score. After I had gone over the site with the Sub-Surveyor
in an apparently casual fashion, we waited with the surveying
until the hours after midday when the whole garrison is wont
to take its siesta. When the work continued beyond this safe
period, the clever diplomatist went to see the Amban and so
skilfully occupied his attention with various representations con-
cerning my journey that he and his underhngs had no time to
grow suspicious about the work around their stronghold.
What I saw of the Celestial soldiery quartered at this frontier
station, showed them as peacefid gardeners or harmless idlers.
One or two of the soldiers, clad in blue cotton fabrics, were
loitering
about my camp to satisfy their curiosity. Neither
Persian,
Turki, nor Wakhi could draw any conversation from
them.
According to the Munshi's statement scarcely any of
the men, who have now passed close upon eight years in the
district,
have even an elementary knowledge of the language
spoken
around them. Considering that the same observation
holds
good of the few officials, and that the military force at their
disposal
is really
insignificant, the order maintained by the
Chinese
administration appeared truly admirable. The success
70 IN SAEIKOL.
may largely be due to the wise arrangement by which all local
affairs are left in the hands of local chiefs and headmen. Taxa-
tion in these frontier districts is very light, and as the Chinese
are anything but exacting masters the people seemed perfectly
contented. Facts like these make one appreciate the power
which an ancient culture and the political wisdom resulting from
many centuries' experience give to the Chinese administration
even in these days of apparent political dissolution.
The Amban had just returned from leave in Kashgar in a
somewhat ailing condition, and as it seemed doubtful whether
he would be able to make the return call demanded by etiquette
during the short time available, I had by the Munshi's advice
to forego my intended visit, however much I should have liked
to profit by the first opportunity to see something of the repre-
sentatives of the Imperial power. All the local dignitaries, with
Karim Beg, the chief of Sarikol, at their head, came, however,
to pay their respects, and with M. Sher Muhammad's assistance
the little Durbar in my
'
Kirgha ' or Yiirt proved quite a
success. The Begs told much that helped me to understand the
former condition of Sarikol and the curiously mixed aspect of
its population. Among the better-class people it seems difficult
to trace any whose families are indigenous to the soil. Some
are descended from Wakhi immigrants
;
a few from Chitrali
and Kanjuti refugees
;
more numerous are those who have come
from Shighnan. It seems that Sarikol, exposed to inroads from
all sides, has been a kind of happy hunting-ground for adven-
turous spirits of the neighbouring tracts who for one reason or
the other found their own valleys too hot for them. This
curious mixture is reflected in the polyglot faculties of the people,
who seem all more or less familiar with Sarikoli, closely akin to
Wakhi, as well as with Persian and Turki.
M. Sher Muhammad had done his best to explain that I was
no
'
Hakim.' All the same, the applications for medicines
from among my visitors were numerous. I could in conscience
do nothing for the aged relative of one of the Begs, whose eye-
sight had grown dim with his burden of years. Still less was
there a remedy in my little medicine-case for the initial stage
of leprosy from which the youthful son of another Beg mani-
PEOPLE OF SARIKOL.
71
nevertheless to be prescribed, and as these would not be con-
sidered sufficiently efficacious without stringent orders as to diet
^
etc., I found myself compelled to add verbal prescriptions also
on matters of my patient's daily hfe, which lay quite beyond
my ken. Spells, if I could have offered them, would undoubtedly
have been still more appreciated.
On the loth of
July
I was able to continue my journey, all
arrangements for transport and such supplies as the place could
offer having been completed. The valley, fully
10,000 feet above
the sea, grows only oats and pulse. Vegetables there were none
to be had. M. Sher Muhammad, Nvith due forethought of the
inhospitable region before me, had all the hamlets ransacked
for eggs, and succeeded in furnishing my
"
chef," Sadak Akhun,
with three score of them. This requisition had eNadently ex-
hausted local resources
;
for before I started I was very politely,
and with many excuses, asked to favour the Amban ^vith half-a-
dozen of these precious eggs, as they were urgently wanted for
making up a medicine ! Of course, I felt happy to obhge that
dignitary.
My route took me first for about three miles down the left
bank of the river to the fortified village of Tiznaf. There my
attention was attracted by a large cemetery with a number of
mud-built domes (Gumbaz), of which the photograph reproduced
here gives a view. A short distance fmiiher down the Tagh-
dumbash River turns to the east and enters the narrow gorge
of Shindah, by which it has forced its way through the meri-
dional range. All the mountains around looked bleak and bare
of
vegetation, forming a striking contrast to the green fields and
meadows of the riverine flat.
The winter route along the Tagharma-su, which joins the
Taghdumbash
River from the west slopes of Muztagh-Ata, was
closed
by the depth of the water. So we had to turn off from
Tiznaf to the North-West, and make for a low pass over the spur
which
descends in the angle formed by the two rivers. From
the
top of the Shush or Kum-Dawan
("
the Sandy Pass
"),
though
it scarcely exceeds 12,000 feet in height, there opened
an
extensive
view over the Tashkurghan Valley southwards.
72 IN SAKIKOL.
The distant snowy peaks, half-enveloped in clouds, which rose
behind it in the South, were the last glimpse I had of the border
of India. The view to the North was still more extensive.
The great mass of Muztagh-Ata, with its mantle of ice, rose up
clearly from the broad valleys which encircle its base on the west
.<s?^
'^m^
IN THE CEMETERY OF TIZNAF.
and south; Imposing as the great mountain looks from its mass
and its crown of glaciers, it did not seem to me from this dis-
tance to equal in grandeur and picturesque form those moun-
tain giants of the Himalaya I had seen, Nangaparbat, Mount
Godwin Austen
("
K. 2"), Rakiposhi, still less Kinchanjanga.
The fact that the relative elevation of the highest dome of
Muztagh-Ata above the broad, undulating plain of Tagharma
at its southern foot is only about 14,000 feet, largely accounts
PLAIN OF TAGHARZHA. 73
I
^^or this ; equally, perhaps, also the absence of boldness in its
^^Drm, and the great height of the permanent snow-hne which
^^^owards the south does not seem to reach down much below
17,000
feet.
After the world of soaring peaks, glaciers, and deep gorges
through which the way from India had taken me, I felt it diffi-
cult to believe myself still in an Alpine world in view of the
broad, rolling plains before me and of the low-looking ranges
which fringe them towards the Pamir. It was a novel type of
mountain scenery that greeted me, and I confess it looked some-
what tame by the side of the views which have indehbly im-
pressed themselves on my memory between Kashmir and the
Taghdumbash. A descent of about one thousand feet brought
me to the irrigated fields of Tagharma, which were clothed in
the fresh green of young shoots of oats and barley. Without
raising one's eyes to Muztagh-Ata I might have thought myself
on the steppe of some northern region. The felt-covered Kirghas
scattered over the plain did not dispel this impression ; the
yaks contentedly grazing on the young grass of the meadows
were the sole feature suggestive of the high elevation at which
we still moved. Safsgos, where I encamped for the night, is
one of the small Sarikoh summer settlements spread over the
Tagharma plain. The inhabitants of the three Karghas, as far
as I could see them, the men and children, were aU singularly
good-looking. Milk and delicious cream were obtainable in
plenty.
On the morning of
July
nth the air was comparatively still
and warm, and only the highest parts of Muztagh-Ata were
enveloped in clouds. Riding along the open grassy plain I
enjoyed distant \4ews, both to the East and West. In the
latter direction the passes of Ghulan, Sarik-tash, and Berjash (or
Berdasht), all leading across the range into Russian territory,
came consecutively into view. Though snow-covered on the
highest shoulders over which they pass, these routes are all
exidently
easy enough at this season. Near the small hamlet
of
Sarala, where SarikoUs carry on some cultivation, we passed
a
Uttle Chinese post, enclosed by loopholed mud walls. It is
intended
to maintain some control over the small detachments
74 IN SARIKOL.
of Sarikoli levies
('
Karaulchi,' as they are called) which guard
the approaches from the Russian side.
After Sarala cultivation ceased, and the irrigated grassy ground
became more and more cut up by patches of sandy soil scantily
covered with hardy herbs. The few Yiirts we now passed were
tenanted by squalid-looking but jovial Kirghiz herdsmen. After
Kukyar the route enters a broad stony nullah, enclosed East and
West by low walls of conglomerate, which looked like remains
of ancient moraines. Above them to the east towered
the snow-capped heights of a great spur known as Kara-
korum, which projects from Muztagh-Ata southwards. By
noon a strong wind began to blow down from the north, and
I was glad to reach the little Kirghiz settlement of Ghujak,
which offered a suitable place for camping. The wind
brought light rain soon after the tents were pitched, and as
the temperature descended rapidly I was glad to get again
into my fur coat, discarded since our entry into the open
Sarikol Valleys. The hypsometer showed an elevation of
about 1
1,
600 feet.
The next day's march was to be a short one, and accordingly
I utilised the morning to ascend with Ram Singh to the top
of the steep spur of conglomerate which rose immediately to the
East in front of the Karakorum peaks. Light clouds, foreboding
a change in the weather, had settled everywhere around the
higher ranges. But the view over the great Tagharma Valley,
and far beyond it to the peaks South-East of Tashkurghan, was
unobstructed, and the plane table work benefited no little by
this excursion. To the North, unfortunately, Muztagh-Ata,
with its glaciers, hid itself in a thick veil of mist and cloud.
After descending again to our last camping-place we resumed
the route to the North. A little beyond I passed the mouth of a
narrow side valley running to the west, known as Khayindi.
It contains a little Mazar or shrine much frequented by the
Kirghiz who grjize around Muztagh-Ata. A little heap of stones
on the road, adorned with horns of Ovis Poli and the wild goat,
and a few sticks bedecked with rags of various hues, direct the
attention of the wayfarer to the neighbouring shrine. The
bits of rag, as throughout the hills of northern India, mark the
KIRaHIZ HERDSMEN.
75
ex-voto offerings of those who have turned to the saint for help
in sickness or some other trouble.
A ride of a little over two hours along the gradually diminish-
ing stream, and between gently sloping ridges of disintegrated
rock and gravel, brought me to Kara-su. There I found a
small post, or
'
Karaul,' enclosed by loopholed mud walls, and
my servants comfortably established in the few huts built inside.
The garrison, the last on this side subject to the authority of
the Tashkurghan Amban, consisted at the time of just three men.
Considering that the ramparts of the post are commanded by
the rising ground to the West within a hundred yards, the
defensive purpose seems to have been less in the mind of those
who built it, than the wish to secure a wind-sheltered comer
for the garrison. Immediately to the South-West a series of
broad, undulating downs leads up to the Kulma Pass, apparently
the easiest of all routes which cross the watershed into the valley
of the Aksu. A Kirghiz whom I met riding on a heavily-laden
pony, some miles below Karasu, had left the Russian outpost
on the other side of the pass that very morning.
The meadows round Karasu w^ere carpeted with the few
varieties of red and white flowers which had greeted me on
the Taghdumbash
;
else, the scenery looked gloomy enough,
for
the clouds were hanging still lower than in the morning.
The
hypsometer gave the elevation as 12,100 feet. Next morn-
ing, the 13th of
July,
the temperature was not as low as I ex-
pected, being 46
F. at
6.30 a.m., but the air was full of mist and
rain threatened. I left the Sub-Surveyor behind to wait for
better weather to continue his work, and marched off by
9
a.m.
The ponies seemed to have a presentiment of the bad time
before them and gave trouble when their loads were being packed.
One of them managed to knock off my travelling bookcase with
such
impetus that its internal fittings were rudely dislocated.
Soon after marching off a violent blast from the pciss before us
brought
icy rain and sleet, and, driving it right into our faces,
made
progress both slow and disagreeable. As far as I could
see
the road led between low, bare ridges by the side of a little
brook,
the head-waters of the Tagharma-su. As, after two
hours'
marching, we were nearing the summit of the pass, the
76 IN SARIKOL.
Ulugh- Rabat
("
High Station "), the rain stopped a httle, and
soon it was noticeable that this bleak upland was not altogether
untenanted. The shrill, whistling voices of the Himalayan
marmots were heard all round, and more than half-a-dozen of
these brown guardians of the passes, so well known to me from
beyond Kashmir, could be seen sitting, with seeming unconcern,
on the little mounds over their holes.
At 11.30 a.m. I reached the pass, which seems to be only
a slight depression in a broad transverse ridge connecting the
Muztagh-Ata massive with the so-called Sarikoli range, the
eastern brim of the Russian Pamirs. The pass, a little over
14,000 feet above the sea, is marked by a stone heap, the tra-
ditional resting-place of some saint. Popular lore about moun-
tain passes does not seem to differ much northwards of the great
Himalayan watershed from what I know it to be on the other
side. Heavy mist on right and left prevented a view of the
higher ranges, but just in front to the North I could look down
into the open, flat valley which descends to Subashi and the
Little Karakul Lake. I had not far advanced on the small
spur over which the path leads steeply downwards, when icy-cold
rain, mixed with snow, began to come down again. It was far
heavier than before, and by the time I passed the first Aul
(herdsmen's camp), called Igrikyok at the bottom of the hill
amphitheatre, I felt nearly drenched. However, there was
little hope of the weather getting better, and I therefore deemed
it best to push on to Su-bashi
("
Head of the Waters "), the
Chinese post in the valley, where better shelter and supplies
could be expected. In the drizzling rain I passed some half-
decayed Kirghiz graveyards and a stone-built Gumbaz, evi-
dently the remains of some older structure.
At last, by 2 p.m., the Chinese post came in view, and with
heartfelt gratitude I greeted its shelter. Inside a neglected
stone enclosure I found, besides a number of tumbledown build-
ings, a row of mud-built huts, representing the quarters of the
garrison. The latter soon emerged in its full strength of eight
men, and their commandant, a sort of corporal, hospitably
invited me to his state-room. It was, in truth, a poor enough
hovel, lighted by a hole in the roof which, closed on account of
CHINESE AT SUBASHI. 77
^Band warm and it felt cheerful amid the felts and quaint articles
^^^of equipment which covered the raised sleeping platform and
the walls. A fire was lit under the hole already mentioned, but
its smoke drove me into the interior apartment adjoining, long
before the tea was ready which it was to warm. Perhaps my
little terrier felt happiest, who, shivering with cold and wet,
could scarcely wait for the host's good-natured invitation to bury
CHINESE GARRISON OF SUBASHI.
himself in the bundle of quilts marking the bed in one corner of
the platform. That he met there a little pet cat without picking
a quarrel with it was the surest proof of his usual temper having
softened under the influence of exposure.
Whether it was the hospitable reception they gave me or
their
neat look and get-up, the little Chinese garrison made
by no means a bad impression on me. The men were mostly
big,
well-set fellows, talking Turki more or less fluently, and
seemed
intelligent enough. When the rain stopped they turned
out
to be photographed in their parade dressblue
velvet
78 IN SARIKOL.
trousers, red cloth tunics, with Chinese letters in black velvet
sewn on them, and neat black felt boots. All these articles
were in good order, less so their Enfield carbines bearing the
"
Tower
"
mark. In the meantime the news of my arrival had
been sent on to Karm Shah Beg, the chief of the Kirghiz herds-
men in the valley North of the Ulugh-Rabat, who duly came to
welcome me. As the rain had stopped I moved my baggage
down two miles from the post of Subashi to where his Kirghas
stood. One of them was readily vacated for the accommoda-
tion of my servants, while a short break in the rain sufiiced for
pitching my tent on a dry, sandy spot by the side of one of the
numerous branches by which the stream of the Subashi Valley
finds its way down to the Karakul Lake. The glittering surface
of the latter, one and a half miles further North, could just be
seen from my camp.
MUZTAGH-ATA PEAKS SEEN FROM ABOVE YAMBULAK VALLEY.
CHAPTER VI.
ON MUZTAGH-ATA
The 14th of
July
brought no change in the weather, and was
by necessity a day of repose. I used it to collect information
as to my intended excursion up the western slopes of Muztagh-
Ata and to pay off the Sarikolis who had so far supplied my
transport. Previously, however, I took the opportunity of
effecting anthropometric measurements on them. After Shams
Beg, the Yiiz-bashi
("
Head of a Hundred "), who had escorted
me from Tashkurghan, had set the example, they readily sub-
mitted to the various operations, each victim in turn affording
amusement to his companions.
In the afternoon the rain grew less, and I rode out with
Karm Shah Beg to pay a short visit to the Little Kara-kul and
the neighbouring tarns of Basik-kul. The detailed descriptions
of Dr. Sven Hedin, who studied these little lakes for weeks
and with loving interest, have made the readers of his work
fully
familiar with all aspects of the neighbovu^hood. Riding
round
the West shore of Karakul I had a full \'iew of the grand
moraine which borders the lower edge of the lake and originally
caused its formation. Above it only a mass of cloud indicated
the
high range which closes the valley to the
North-East.
80 ONJMUZTAGH-ATA.
The bleakness of the hills which rise on the West to a height
of about four to five thousand feet above the lake and the
low mounds of old moraines stretching along the shore, under
the grey sky, gave a desolate, sombre look to the little lake.
As the glaciers of Muztagh-Ata kept wholly invisible, this
impression was not relieved by the grandeur of the more distant
surroundings. Mournful, too, looked the still smaller Basikkul
basins and wild the confusion of ice-ground mounds of rock
and detritus which ancient moraines have left in the narrow
space between them. It was evident that the icy splendour
of the great range eastwards is required to give to this group
of little lakes its true Alpine beauty.
I returned by the east shore of Karakul, past the little bay
where Dr. Sven Hedin's camp had been pitched.
Considering
how long he stopped in the neighbourhood, and how closely
acquainted he became with the Kirghiz then encamped there,
I was surprised how liitle my guides could tell me of this dis-
tinguished visitor. But the nomadic ways of the Kirghiz fully
explain this scant recollection. The families then grazing
around Muztagh-Ata have wandered elsewhere. Togdasin Beg,
Dr. Hedin's friend, has since died far away on the Russian
Pamir, and the other companions of his excursions in these
mountains seem also to have scattered to other grazing-grounds.
It was instructive evidence how little local tradition can be
expected among the wandering tribes that frequent these
valleys. The path back to camp took me along the cliffs which
run down into the lake from Kara-kir
("
Black Ridge "), a
bleak height of dark rock rising immediately to the East of
it. As soon as camp was reached at six o'clock the rain began
to pour heavily again. It plainly meant snow in the higher
regions and consequent delay in my excursions. My diary
entries for
July
15th, which I here transcribe, show that I had
not been mistaken in my apprehension.
"
It rained and snowed through the whole night, and
mist
and grey, drizzling rain covered what little I could see of the
valley when I got up. There was nothing for it but to sit
in the tent and write up notes and letters that were to go down
to Tashkurghan to catch the next Dak for India and
Europe.
KIRGHIZ HOSPITALITY.
81
Larm Shah Beg came to pay his respects and to sit in happy
jkregard of time and weather under the httle awning in front
my tent, but what I could ehcit from him as to the arrange-
lents for my further jovuney was far from cheerful. It was
>y
to notice that the want of instructions from the Chinese
Bulunkul was sorely disturbing his peace of mind. In a
)ne intended to convey a sense of mystery and secret devotion,
assured me that he was ready to render any ser\'ice
if
were not for the distrustful Chinese. To give me yaks and
len for my intended visit to the Yambulak Glacier and the
^opes of Muztagh was a thing he could do in safety. But to
ipply animals for a move to Kashgar might bring down upon
the wrath of the Amban. Even to send a few yaks to
iy last camp at Karasu for the baggage of the Sub-Surs-eyor,
^ho was to join me, seemed an act of grave risk. The Beg's
Itering excuses gave me a good idea of how well the Chinese
lanage to keep their roving Kirghiz in hand, but equally little
hope of the help I needed for my immediate movements.
I
accordingly sent the Sarikoli Beg, who had come with me from
Tashkurghan, back to Karasu with orders to pro\4de there
locally for the transport needed by Ram Singh, At the same
time I got Karm Shah Beg to despatch a messenger to Bulunkul
who was to show to the Amban the local passport issued to me
by his Tashkurghan colleague and to bring back orders for
my Kirghiz host.
"
In the afternoon the clouds hfted a little and showed the
mountain slopes down to a few hundred feet above the level
of the valley clad in fresh snow. No encouraging
prospect
for my Muztagh-Ata excursion, which if to be made at all
must be made within the next few days ! I used the short
interval when the rain stopped in the evening for a visit to
the Beg's Yiirt. He seemed to appreciate the compliment, and
whatever doubts he may feel as to the results of any assistance
he may render me, they did not interfere with a display of
cordial
hospitality. In the middle of the Yiirt a big cauldron
('Kazan')
of milk was boiling over the fire. One of the Beg's
wives,
no longer young, but of a pleasing expression and cleanly
dressed,
was attending to the fire of dwarf juniper (' Teresken
').
6
82 ON MUZTAGH-ATA.
"
While the dish was getting ready, I had time to look
about and to examine the homestead. Comfortable it looked
in contrast to the misty, grey plain outside. The wicker-
work sides and the spherical top of the Yiirt are covered with
coloured felts, which are held in position by broad bands of
neatly-embroidered wool. All round the foot of the circular
wall lie bundles of felt rugs and bags of spare clothes, evidently
stored for a more rigorous season. A screen of reeds, covered
with woollen thread worked in delicate colours and bold but
pleasing pattern, separated a little segment of the Yiirt ap-
parently reserved for the lady of the house, who again and
again dived into it, to return with cups and other more precious
implements. The floor all round, except in the centre where
the fire blazed, was covered with felts and thick rugs made of
yak's hair
;
for my special accommodation a grey-coloured
Andijan carpet was spread on one side. The warm milk,
which was offered from the cauldron by the presiding matron,
tasted sweet and rich. I had it presented in a large Chinese
cup, while the rest of the company, which comprised over a
dozen of the Beg's male relatives and neighbours, helped them-
selves from a number of bowls in wood and iron. Milk is a
staple article of food with the Kirghiz, and the healthy look
of the men around me, young and old, showed how well it
agrees with them.
"
Towards the end of my visit Karm Shah Beg produced
a big sheep that I was to accept as a token of hosj^itality and
good-will. I should gladly have taken a smaller one, since
for weeks past I had occasion to notice tliat the sheep which
my men selected for purchase were as distinguished for tough-
ness as for size. Karm Shah Beg, however, had different
notions on this point, and was not to be denied. So I consoled
myself with the thought that at least there would be satisfaction
among my men. The Kirghiz are a matter-of-fact people, with
a
keen eye for money. Hence I did not fail to assure my host
that his present would be returned by more than its equivalent
in value before I left the valley.
"
Late in the evening, as 1 was comfortably settled in my
tent
and busy writing, Karm Shah Beg turned up with
A HITCH ABOUT TRANSPORT. 83
triumphant mien to announce the arrival of a Chinese officer
from Bulunkul who had brought orders to supply me with
transport. It was clear that a great load had been taken
from the Beg's mind. Glad as I was for this early settlement
of the question, I thought it right to treat the news as a mere
matter of course. I could not have expected it otherwise !
Karm Shah Beg was accordingly told to keep his Chinaman
ICY
RANGE, WITH SARGULUK PEAK, TO NORTH-EAST OF KARAKUL LAKE.
and the message he was to deliver until I should find it con-
venient
to receive them on the morrow."
The night brought at last a change in the weather, and when
on the morning of the i6th the Beg turned up with the Amban's
messenger and the commandant of the Subashi post, I could
receive
them in the open. I had tea passed round in cups
and
bowls which my servants procured

I do not know from

where
and^then received the assm-ance that whichever way
6*
^
84 ON MUZTAGH-ATA.
I should choose for my journey to Kashgar, transport would
be forthcoming. There seemed little hope of the shortest
route down the valley, by the Gez defile, being available
;
for the river, swollen by the melting snows, was said to have
carried away one if not two bridges, which could scarcely be
repaired before the autumn. But there is another, if more
difficult, route round by the northern spurs of Muztagh-Ata
and over the Karatash Pass, and though this was represented
as nearly blocked by snow, I made it clear that if the Amban
and his people wished to get rid of me, it would have to be
by either of the above two routes. On both of them there
was surveying work to be done, which explains my insistence.
When I had finally dismissed my visitors with a clear notion
of what I expected from them, I set out for the Karakir Hill,
east of the lake, which by its central position promised to be
a good station for work with my photo-theodolite. The rain
had stopped during the night, but the clouds were still hanging
round the peaks, and icy gusts of wind were shifting them
continually. The yaks carried me and my instruments easily
enough to the long ridge which crowns the hill and is seen
in the foreground of the view reproduced on
p. 83 ;
but it wi
only after a long wait, made trying by the cold wind whic
passed through all my thick clothing, that the clouds lifted
sufficiently to permit of satisfactory work. Then glacier after
glacier emerged from the great white wall to the north and east
formed by a succession of ice-crowned peaks, the worthy rivals
of Muztagh-Ata
;
the deep valley of the Ekkibel-su, which
drains Muztagh-Ata from the north, also lifted its veil, and by
3
p.m. the tiring work on the wind-swept height was rewarded
by a complete round of accurately fixed views which, I could
hope, would prove a useful supplement and check to the plane-
table work. That in the midst of the operation the tangent
scale of the photo-theodolite broke, and had to be replaced
with what primitive tools I managed to procure about my
person, was an incident taxing what little I possess of mechanical
skill. It was no surprise that my benumbed fingers, while
replacing the scale, broke one of the cross-hairs of the camera.
But
this mishap was repaired too, thanks to
the ample suj^ply
86 ON MUZTAaH-ATA.
of delicate threads of hair which Mrs. W.'s kindness had provided
in Gilgit.
It was bitterly cold by the time I descended, and all the
more grateful I felt for the shelter of my little tent. Its
warmth was increased by the use of a small
"
Stormont-
Murphy Arctic Stove," burning cakes of compressed fuel, with
which I had provided myself from the Military Equipment
Co., London. Thus comfortably ensconced within my tent-
walls of eight feet square, it was a pleasure to work away till
midnight at a mail that was to carry my news to distant friends.
On the 17th of
July
I awoke to a glorious clear morning.
Without a speck of cloud or mist the gigantic mass of Muztagh-
Ata towered above my camp. I had counted on this chance
for my projected visit to its higher slopes. The rain of the
previous days had interfered with the Sub-Surveyor's work,
and while he was making up for the delay I could effect my
excursion without having to accuse myself of any waste o^
time. I had soon separated the outfit most needed for th|j
tour from the rest of my baggage. The ten yaks that were
to move it and to serve as mounts were also soon procured
from among Karm Shah Beg's herd that was grazing near
the lake. All surplus stores and baggage not needed were
to be left behind in charge of Mirza, my Turki servant from
Peshawar, who, not equal to the fatigues of the long journey,
seemed manifestly in need of rest. With him I also decided
to leave
'
Yolchi Beg,' who was to be spared unnecessary
climbs in ice and snow. He had so far borne the long marches
wonderfully well, and had lost none of his vivacity and high
spirits.
It was midday by the time I moved off, accompanied by
Sadak Akhun, my Kashgar servant, and the three followers
from Hunza and Punyal, who were now to have a chance of
showing what they were worth on the mountain-side. The
air was delightfully still and warm, and as we rode along the
rich grazing land at the bottom of the valley, the fragrance of
the flowers and herbs was most perceptible. Passing the
Subashi post, which now in full sunshine looked far more
dilapidated than when it first offered me its shelter, we turned
RECONNOITRING THE GREAT PEAKS. 87
round the foot of the great spur of Shamalda into the valley
leading towards the Yambulak Glacier. The latter descends
in a westerly direction from the col connecting the two main
peaks of Muztagh-Ata, and is flanked by mighty ridges both
to the north and south. From Dr. Sven Hedin's experience,
fully detailed in his fascinating volumes, it was clear that the
(jnly part of the great mountain from which access might be
gained to its higher slopes, if not to one of its summits, was
the spur rising above the north edge of the Yambulak Glacier.
From a height near the head of the valley, where the fine view
reproduced at the head of this chapter lay before me, I surveyed
through my small telescope this great ridge as it stretches
up in apparently unbroken line to the northern summit. The
t
orresponding ridge on the south side of the glacier could be
>een to be coated with a huge crust of old ice, which, furrowed
by crevasses up to the very highest summit, manifestly left
no chance of ascent. The rocky spur which this mantle of
ice covers, rises above the glacier in an almost perpendicular
face of cliff several thousand feet high. The rock wall on the
opposite, northern side of the Yambulak Glacier, is not only
lower, but its slope is less steep and seemed less encrusted with
ice. Further to the north the sides of the mountain are far more
precipitous and packed with glaciers.
My preliminary examination of the northern ridge which Dr.
Hedin had followed in his three attempts to ascend Muztagh-
Ata, fully bore out his description, except in one important
particular. His ascents in
1894
had taken him along ground
that up to a height he estimated at over 20,000 feet, was almost
clear of snow. But now I coiild not fail to note even from a
distance
that snow of considerable depth covered the identical
ridge
down to a level of less than 17,000
feet. Satip Aldi,
my
Kirghiz guide, who had accompanied Dr. Hedin on one
of his ascents, was aware of a change which the heavy snow-
fall
of the last two years had brought about in the condition
of
this part of the mountain. It was evident that I could
not
possibly hope to reach the height to which that
distinguished
explorer's
party had ridden up on yaks, in the same convenient
manner.
88 ON MUZTAOH-ATA.
The night from the 17th to the i8th of
July
was spent in
camp near the few Yiirts in the upper part of the valley which
bears the name of Yambulak. Reckoning with the increased
difficulties which deep snow was likely to offer, I decided to
obtain spare yaks for myself and the men who were to accom-
pany me. Animals broken for riding were secured after some
delay, and it was only by
7
a.m. on the following morning
that I was able to move off. An hour later I had reached the
foot of the great moraine which flanks the lower portion of the
Yambulak Glacier on the north, and with it the last bit of
fairly level ground. Leaving my baggage behind with orders
to pitch the tents, I then rode up the steep slope of moraine
debris and gravel to reconnoitre the ridge above. The point
where the baggage was left lay already at an elevation of over
15,000 feet, and the panting of the yaks as they struggled up
over the trying slopes of loose stones and shingle showed plainly
that these hardy animals felt the effects of the elevation. With
their wonderful surefootedness the yaks combine a sluggishness
of temper which at all times makes a ride on them a trying
mode of locomotion. But I never felt this more than when
we had to make our way over these steep and slippery slopes
and at the same time to drag along the spare yaks that were to
relieve our mounts. Comparatively safe from sticks, the use
of which alone could keep the yaks to an upward track, these
extra animals were ever and again twisting themselves into
the wrong place.
More than an hour passed before we reached the lower end
of the rocky ridge above described over which we were to
make our way. The ground now became firmer, but with it
too we had reached the line of snow. It lay thin at first and
did not hamper our progress. But after half a mile of ascent
along the crest it became deeper, and at an elevation of about
16,500
feet j)ractically forced us to dismount. It was
half-
past ten by this time, and the clouds which seemed thin and
fleecy in the early morning were now gathering in heavier
masses above us. The point to which we had been able
to
force our yaks seemed the last where we could jiitch a camp.
The snow which covered the top of the ridge had melted
on
CLIMB TOWAEDS YAMBULAK GLACIER. 89
the slope which descended to the glacier several hundred feet
below. The slope was not too steep for tents, but seemed
otherwise to offer little advantage. Far above the region
where even yaks could secure food, the spot was yet decidedly
too low to serve as a convenient starting-point for a long climb
on the follownng day.
Considering the height above us, the selection of a suitable
spot for a camp seemed all-important. From where the yaks
had brought us nothing was to be seen but a broad slope of
snow fringed on its southern edge by precipitous cliffs falling
towards the glacier. In order to make sure of the chances
for camping higher up I despatched the two Hunza levies on
a reconnaissance. They were to examine the conditions of
the snow, and to look out for some shelter in the rocks which
might enable us to pass a night at a greater elevation
eventually without tents. I myself remained behind to use
the comparatively clear weather for work with the photo-
theodolite. The clouds that were gathering and the high wind
that sprang up were a warning not to lose time.
The view which the place of my halt offered, and part of
which is shown by the photograph reproduced on
p. 90,
was
grand indeed. It comprised to the west range after range of
the Pamirs, from the distant peaks of Wakhan far away to the
Alai mountains. The mountains lining the valley below me
on the west seemed nowhere higher than my place of obser\'a-
tion, for which the hypsometer reading indicated 16,820 feet.
From the same point splendid views were obtained up and
down the Yambulak Glacier. Compressed between mighty
waUs of rock the stream of ice seemed in a state of petrified
convulsion. From its highest point where its firn filled the
space between the twin peaks of Muztagh-Ata down to the
opening of its rock-bound gorge, the glacier displayed a be-
wildering
maze of huge crevasses. Their greenish depths
contrasted vividly with the spotless white of the snow-crust
that covered the surface of the ice.
Opposite to us rose the
almost
perpendicular wall of rock which faces the great
ridge ascending straight to the southern and
highest peak
of
Muztagh. Above this rock-wall there showed the thick
I RECOXNAISSANCE OF HUNZA GUIDES. 91
ice of the glacier-mantle covering that side of the
I
mountain.
Though the sun was hidden only for short periods b}- light
clouds, it felt cold enough in the strong breeze. So it took
time to work the delicate apparatus of the photo-theodohte.
The two Kirghiz who had come up with us in charge of the
'
yaks had already complained of headache, and by the time
I completed my work succumbed to mountain sickness in that
most drastic form affecting the digestive organs. To get from
j
them the needed topographical information was for the time
quite impossible. I was watching the snow-covered ridge by
which my Hunza followers had ascended, with some anxiety
as the day wore on, when at last by
5
p.m. they returned.
Hardy and bom climbers as they are, Wali Muhammad
and Ghun looked thoroughly exhausted. They reported that
they had climbed in deep snow shoulder after shoulder of the
I
great ridge above until they were stopped by a precipice
I
of sheer rocks descending to a side glacier which separates
the ridge they followed from the main mass of the northern
peak.
Their account confirmed the doubt which the observation
made on the preceding day had left in me as to the continuity
iof the spur apparently leading to the summit. High up, at
an
altitude approximately estimated at 22,000
feet, 1
had
I
noticed what looked like crevassed masses of ice protruding
I
to the north from below one of the buttresses of the ridge.
From
the description of my men I was forced to conclude that
this
ice was in reality the end of a
transverse glacier hidden
from
view
by the ridge on which we stood. Both men de-
scribed
a descent over the cliffs down into the glacier-filled
i gorge as wholly impracticable, and the ascent on the opposite
face
giving
access to the summit as of equal difi&culty.
They
; complained
bitterly of the cold they had experienced in the
higher
part of their ascent and of the dif&culty of breathing.
j
Though
they had followed closely the Une of the cliffs
overlooking
I
the
Yambulak glacier, they had found no possible place
of
j
shelter
among the rocks nor even a spot where a
small
tent
could
be pitched. Ever>^vhere the snow was too deep
and
92 ON MUZTAGH-ATA.
tolerably level space wholly wanting. This report showed
clearly the limits which existing conditions imposed on any
attempt at ascending Muztagh-Ata from this side. I realised
that without the possibility of camping higher up for one night
all that could be aimed at was to penetrate the gorge which had
stopped my men. The discovery of this formidable obstacle
was in itself an interesting fact ; for Dr. Sven Hedin, who
in
1894
had ridden on yaks to a height estimated at over 20,000
feet, seemed to have remained wholly unaware of its existence.
At the same time the report showed that the ascent up to that
point was only a question of endurance and fair weather. I
accordingly determined to use the chance of the next day
keeping clear for an ascent by the track which my Hunza men
had followed.
This chance seemed doubtful indeed, for clouds settled
around the summits, and violent gusts of wind made us gla
to seek the shelter of the tents below. I found them pitche
by the side of the great moraine wall and Ram Singh dul
arrived from Karasu. The sky was cloudy when I turned ii
and when I rose on the 19th at
3.30
a.m. there was ever
indication of a storm. It did not take long before the snov
driven by a hard gale, came down. Wrapped in furs I wa
sorrowfully watching this atmospheric interference with ra;
plans when by 6 a.m. the sky began rapidly to clear. I saW*
that it had been only one of those short-lived gales which,
according to Dr. Hedin's testimony, almost daily visit one
or other face of the great mountain. The yaks were kept
ready, and when the sun broke through, a little before
7
a.m.,
I gave the order to start. Instead of the Kirghiz who had
proved so useless on the mountain-side, only my Hunza men
and Ajab Khan, the Punyali orderly, were to accompany me
and the Sub-Surveyor. The latter's instruments were to be
carried as far as practicable by honest Hai Bai, a
Turki
'
Kirakash
'
(pony-man) from Kokyar, who had accompanied
Captain Deasy's caravan to Ladak and Kashmir, and who,
anxious to return homewards after a winter's rest, had attached
himself to my camp at Srinagar.
This time we were not encumbered by spare yaks for
which
I
HIGHEST POINT KEACHED. 93
no use could be found, and a little over an hour from the time
of the start saw us at the point I had reached the day before.
The weather kept clear though there was much wind. The
yaks toiled on through the snow, which gradually grew deeper,
but their progress was slow, and the task of keeping them
ahead trying for the riders. More and more frequently we had
to dismount and drag the stubborn animals out of the deep
snowdrifts into which they had plunged. At last, when an
hour's toil had advanced us only some
500
feet above the previous
day's station, it became necessary to leave the yaks behind.
The snow by this time had increased to an average of five feet
in depth, and in many places where some projection of the ridge
had favoured the formation of drifts, our alpenstocks altogether
, failed to strike the rock. The surface snow was crisp and
'
granular, hence so easily shifted by the wind that in most
places the footprints left by the men on their ascent the day
before could barely be distinguished. The snow along the
edge of the rock-wall, which we were careful to follow, seemed
at this altitude to have accumulated only during recent winters.
Hence it had scarcely yet had time to be compressed by its
own weight into ice
;
and the glittering snow sheet over which
we were ascending
rested firmly on the rock. Against slips
of snow and av^alanches we were thus safe enough, while from
the
opposite side of the glacier where the southern wall of
rock
was topped by a thick layer of ice, httle avalanches
would
glide down more and more frequently as the day
wore
on.
It
was not easy work to ascend in the soft snow, where we
continually
sank in for a couple of feet, and where a slight
deviation
from the track of the front man would land one up
to the
waist. Though the high elevation we gradually at-
tained
caused me neither headache nor any other symptoms
of
mountain
sickness,
yet the rarity of the air necessarily
j
imposed
slow progress and frequent though short halts. The
wind
grew
stronger as the day advanced, and brought passing
I showers
of snow. Yet more troublesome was the snow which
,
the
force
of the wind swept up at times from the slope
before
us.
Shortly
before midday I reached a point where a few
94 ON MUZTAGH-ATA.
dry rocks at the edge of the spur protruded from the snow.
They offered a convenient spot for a halt and refreshment.
Immediately below the nearly perpendicular cliffs there stretched
the contorted icy surface of the Yambulak Glacier. Contrary
to my expectation, the altitude of close on
19,000
feet which
we must have reached by this time interfered in no way with
my appetite.
After another short snow squall had passed we resumed our
climb, but the Sub-Surveyor and Ajab Khan began to complain
of headache and general lassitude, Bichlorate of potash
tabloids, with which after Dr. Bellew's advice I had provided
myself, proved of little avail, and by 1.30 p.m. Ram Singh
had to remain behind. The wind had by this time driven
away all clouds that hung over the ranges west and north,
and he was thus able to check his plane-table work by rays to
a number of distant peaks previously sighted or triangulated.
Twenty minutes later Ajab Khan, skilled mountaineer as he
is, fell out and received permission to descend. There remainec
now Wali Muhammad and Ghun, my two Hunza levies, anc
they without a word of complaint steadily plodded on witl
me. The snow became still deeper, and the mist that settle^
on the peaks above us showed clearly that a further ascen|
would offer no chance of a close survey of the summits. A'
change in the weather seemed also to threaten, and after due
deliberation I fixed upon the buttress of the ridge just before
me as the final pbject of the climb. By 2.30 p.m. I had reached
its top and settled down by the side of the precipitous rock
wall descending to the glacier.
In the piercing wind it was not easy to boil the water for
the hypsometer. By scooi)ing a hole in the snow, however,
sufficient shelter was at last secured, and repeated careful
readings of the thermometer gave a mean of 177.8 Fahr. as
the boiling point. Taking into calculation the temperature
of the air, which was 33 Fahr., this corresponds to an elevation
of almost exactly 20,000 feet. Our bodily condition would
have allowed a further climb, though I as well as my Hunza
followers felt the effect of our six hours' ascent through the
snow. But neither the state of the weather nor the remaining
I
FREEBOOTING VISIONS.
95
lylight justified the hope that we could reach this day the
id of the spur reconnoitred on the previous day. I ac-
jrdingly decided to descend and to let a fresh attempt at
iching that point depend on the chance of the weather
iproving on the next day. The soft snow rendered even the
scent by the route we had come a slow and arduous afifair
;
it the grand view which lay before me amply compensated
^r the delay. To the west the clouds had lifted completely,
id the multiplicity of the ranges over which my gaze
ivelled was the best demonstration of the height we
ire at.
^Straight in front, where the view must have extended prac-
ly across the whole breadth of the Pamir region, there
;re no notable landmarks to attract attention. But this
seemingly endless succession of valleys and ranges was perhaps
i best calculated to impress me with a sense of the vastness of
'
the
"
Roof of the World." To the south-west there glittered
white pinnacles of bolder shape far away on the horizon, and
j
in them I thought I could recognise the mountain-giants that
i guard the approach to the Indus Valley. They had worthy
rivals
to the north in some towering masses of ice and snow,
which from a reference to the map I could not fail to identify
with
Mount Kaufmann and other great peaks of the Trans-
Alai
range. Their highest points were shrouded in fleecy
clouds.
The
grand spectacle which made me stop again and again,
heedless
of the cold and wind, also impressed my companions,
though
from another point of view. Here was a vast region
full of rich grazing grounds, greater than the dwellers of the
narrow
valleys of Hunza could ever imagine. It was not
difficult
to guess what were the prominent thoughts that
. passed
through
my Kanjutis' minds, and a few s3Tnpathetic
' hints
soon brought them to their tongues. What a vast field
for
raids
and
conquest lay there before the hardy, brave
;
hillmen
of Hunza ! The old freebooting spirit broke forth
I
again
in their talk, together with their contempt for the meek
I
Kirghiz,
those willing servants of whoever lets them graze in
peace.
Wall
Muhammad revelled in recollections handed
96 ON MUZTAGH-ATA.
down by his father how the men of Hunza had raided the rich
flocks of Tagharma, to the very foot of the great mountain
on which we stood. But now the
'
Sirkar ' has made its
will felt, and no Kanjuti dare disturb the peaceful dwellers in
these valleys. I could not cheer my plucky guides with promises
of a return to those happy days, but I must own to sympathy
with their views in my innermost heart. Were it not for the
great powers that keep watch from south to north, there is no
doubt that little Hunza would with ease sweep across all the
valleys from the Oxus to the Kashgar border.
When by 6 p.m. we had descended to that portion of the
ridge where the snow had left some patches of bare rock, I
was cheered to find the tents pitched on the steep declivities
towards the glacier. The place was the best that could be
had for the purpose, but the angle of the slope was anything
but adapted for a tent. When I retired to its shelter, I felt
as if it were a cabin on board a ship rolling badly. With
some trouble the nearest approach to a horizontal position
was secured for the camp-bed, and as it was the only thing on
which it was possible to sit or lie in comfort, I soon succumbed
to its attraction. Before, however, I finally got to rest the
patients had to be attended to. Ram Singh and Ajab Khan,
'
huddled up in their little tent, still complained of splitting
headache and nausea. Outside, undismayed by cold and
wind, my hardy Hunza men were feasting in great glee on the
big mutton-legs which had been assigned for their refresh-
ment. Their cheery talk, alas ! unintelligible to me, was
still in my ears when I fell asleep after a frugal dinner. Was
it of the pluck and prowess of their little race that they
chatted, or of the happy hunting-ground for slaves and sheep
which the Pamirs had so long offered to it ?
The night brought violent gusts of wind and several light
falls of snow. The noise of the avalanches falling over the
cliffs on the south side of the Yambulak Glacier woke me at
frequent intervals. It was a comfort to think that there was
no
danger of that kind to fear on the ridge we occupied. When
1 woke up at 6 a.m. there was fresh snow to a depth of two
inches covering the ground. The sky was still cloudy.
SURVEY WORK ON SHAMALDA.
97
There was nothing for it but to wait in patience for a change
in the weather. But the change would not come, and as the
temperature still kept at freezing-point we felt badly the
exposed position of the camp. At last all hope of an ascent
that day had to be abandoned, and as to wait for better weather
would have meant loss of more time than I could afford, I had
by midday reluctantly to give the order for the move to a less
inclement region.
On the evening of the 20th of
July
I camped once more by
the side of the Kirghas in the Yambulak
Jilga.
On the
ICY RANGE WITH PEAKS ABOVE KONGLR-DEBE AND KOKSEL GLACIERS.
following day Ram Singh and myself ascended the high side
spur, called Shamalda, which descends from Muztagh-Ata
northwards of the Kampar-Kishlak Glacier, for survey work,
ilcy
blasts of wind blew in turn from different directions, and
kept now one and now the other side of the mountain
shrouded in clouds and mist. Work with the photo-theodolite
iwas trying under such conditions, yet by using the favourable
Smoments views were secured of the valleys and ranges opposite,
[which
from this height
14,570
feet above the seashowed
iClear and sharp, as if on a relievo map spread before me. The
jtriangulation,
too, was extended to the great peaks north-
east
of Lake Karakul. After long and trying hours on the
7
98 ON MUZTAGH-ATA.
wind-swept bleak ridge I was glad to hurry down to my camp,
which in the meantime had been moved again to its old place
south of the lake. To my pleasant surprise there awaited
me a troop of
'
Kirakash ' men with ten ponies, whom Mr.
Macartney, most thoughtfully anticipating my need of fresh
transports, had sent from Kashgar to meet me. Through
them there came, to my intense joy, a packet of home letters
which had reached Kashgar from Europe by the Russian post
via Samarkand. The latest bore the date of the 24th of
June,
a proof how near the railway has brought even the slopes of
Muztagh-Ata.
On the 22nd of
July
the weather cleared very suddenly,
and the day of rest and quiet work in camp was made doubly
enjoyable by a perfect view of the grand mountain. Even
the great glacier-clad range to the north-east, dubbed
"
Kongur
"
on our maps, but locally bearing neither that nor any other
general name, lifted for an hour the veil of clouds from its
highest peaks. Those above the glaciers of Kongur-debe and
Koksel rise, according to our triangulation, to elevations of
23,600
and
23,470
feet respectively, and thus fairly rival
Muztagh-Ata, for which
24,321
feet is the latest ascertained
height. All the same I was glad to learn from the Kashgar
pony-men that I should not have to force my way through
the gap known as the Kara-tash
("
Black stone
")
Pass, which
on the south separates that range from the
"
Father of Ice
Mountains." Instead of this difficult route the Gez gorge
was reported to be available, though in the flooded condition
of the Yamanyar River I had scarcely ventured to hope for
this.
START FOR GEZ DEFILE.
CHAPTER VII.
THROUGH THE GEZ DEFILE TO KASHGAR.
The morning of the 23rd of
July
found me ready for the start
northwards on the way to Kashgar, while Ram Singh, with a
small camp, was to move to the north-east to complete the
survey up to the Karatash Pass and towards Bulunkul. He
was then to follow me to Kashgar a week in the rear. To
provide against possible loss of the plane-table work so far done,
through accident or Chinese interference, I photographed the
section of the map that Ram Singh was to keep. My two Hunza
men and Ajab Khan were discharged to their homes, cheered by
the prospect of returning to their own mountains and by the
substantial rolls of rupees representing their pay. Before I
said good-bye to Karm Shah Beg, who in a handsome
Bakhshish
100 THEOUGH THE GEZ DEFILE.
in cash had found compensation for the revolver he had pre-
viously coveted as a
"
keepsake," I had the satisfaction of seeing
my Indian mail arrive from Sarikol. It was a big one, and brought
besides welcome letters and papers from home and India eagerly
expected little parcels, the result of orders I had sent to Lahore
six weeks before from Gilgit, after the first experience had shown
me the lactmce of my equipment.
While the baggage marched on to Bulunkul I took the oppor-
tunity of completing my photo-theodolite survey from an isolated
hill rising due west of Karakul, and about
1,500
feet above it.
The panoramic view I enjoyed from the height of Kok-tumshuk
Hill was perfect. The lake at my feet glittered in the changing
tints of emerald and chrysoprase.
Just
opposite on the eastern
shore the telescope showed Ram Singh working from my former
station on Karakir with the theodolite. To the north there lay
peacefully the little tarns of Basikkul, deep green in colour, and
in the soft clear light even the bleak old moraines around them,,
with their
"
cirques
"
and walls of rocky debris, looked less
desolate and dreary. By
3
p.m. I had exposed the last plate
and hurried down to where the limpid stream leaves Karakul.
Then I marched along the eastern shore of the Lower Basikkul,
and realised for the first time the charms of this pretty
"
eye of
the sea
"
as it would be called in the Carpathians. Sleepy it
looked on the quiet summer afternoon, and inviting for rustic
repose the grassy strip on its eastern bank. But I had little
time to spare for such pleasures, and soon had to tear myself
away from this newly discovered attraction.
Beyond, vegetation soon diminished as I marched along
the stony
'
Dasht ' that forms the bottom of the valley further
down. The river, swelled by the great glacier stream known as
Ekkibel-su, that joins it from the south-east, occupies a gradually
broadening bed of rubble and boulders. I crossed it with some
little difficulty about four miles below Karakul, its depth there
reaching
4
to
5
feet. The rest of the march lay alternately over
stony
"
alluvial fans
"
spreading in front of the glaciers which
descend from the great icy range eastwards, and over narrow
strips of rich meadow land fringing here and there the steeply-
cut conglomerate banks. The flora seemed more or less the same
i
OBSTRUCTION AT BULUNKUL. 101
as about Karakul, but the growth and the scent of the herbs
growing in these sheltered nooks was stronger.
It was close on 8 p.m. when I reached my camp pitched near
the Chinese post of Bulunkul on the swampy piece of meadow-
land that fills a bend of the river. The military Amban of the
place is supposed to watch the neighbouring passes which lead
across to Rang-kul and adjoining parts of the Russian Pamirs.
From the report brought to me he seemed httle inclined to help
me on my
j
ourncy. Though he sent fuel and a sheep as a present,
Osman Beg, the influential headman of the Kirghiz grazing in
the neighbourhood, to whom I had been recommended from Tash-
kurghan, found it advisable to pay his visit by stealth and
under the cover of darkness. Ne.xt morning the attitude of the
Amban made itself palpably felt. Karm Shah Beg's men, with
their ponies, which had brought part of my baggage from
Karakul, had disappeared during the night. To move on with
the five Kashgar animals (the rest I had left for Ram Singh's
camp) was manifestly impossible. From the Amban, to whom
I sent, came nothing but a rude reply, leaving it to my own choice
how I should make my way beyond. I thought of Dr. Sven
Hedin, who had met with an even worse reception at Bulunkul,
and consoled myself with the conjecture that possibly the climate
of the place disagrees with the liver of successive Chinese com-
mandants.
Fortunately the Amban's obstnictiVeness was purely passive.
I found little difficulty in persuading his interpreter
('
Tol-
mach
'
) that it was to his material advantage to supply the
needful animals. He took the hint, and by ii a.m., by taking
Dak ponies and pressing Kirghiz animals, the needed complement
of my transport was made up. The pliable factotum of the
Chinese
"
Warden of the Passes
"
was for a
consideration found
ready even to issue in his chief's name an order for the supply
of other animals at the Karauls further down the valley. WTien
my httle caravan was fairly started I could not deny myself the
satisfaction of returning to the Amban the sheep he had sent, as a
present that was not acceptable. The Kirghiz, in whose pre-
sence I sent my message,
were evidently
much tickled by the
announcement.
102 THROUGH THE GEZ DEFILE.
The weather had again become cloudy and cold, quite wintry
in aspect. With the clouds covering the mountains almost to
their foot the wide valley, through which the river beyond
Bulunkul spreads in numerous broad branches, looked like a
dreary steppe in the autumn. About five miles north of Bulunkul
the expanse of muddy glacier water contracts and enters by a
sudden bend to the east the long defile known as Gez-Darra.
At its entrance we passed a loneh^ Karaul, square-walled and
garrisoned by a dozen Chinese soldiersscarcely tenable as a
defensive work, even in the days of matchlocks. The rest of
this day's march lay along the right bank. The gorge kept
narrow, and the road almost throughout led over old moraines
and stony
"
fans." By 6.30 p.m. we passed the first serious
impasse of the gorge called Janguruk. Great serrated coulisses
of rugged rock, several thousand feet in height, descend from the
main mountain spurs on both sides. Along the face of one of
them the road is carried by a gallery, a true
'
Rafik
'
of Hunza
recollection, only that the one here was well built and gave a
roadway of at least four feet breadth. It was getting dark
between the high mountain walls when this awkward part of
the route had been passed. So we had to stop for the night's
camp as soon as the necessary bit of level space could be found
by the river side. Here amidst sombre Alpine scenery, with snowy
heights gleaming up side gorges, I was to be reminded of the
very different region that awaited me eastwards. Heavy
yellowish clouds overspread the narrow bit of sky, visible between
the mountains, and soon the tail end of a duststorm, wafted from
the eastern plains swept up the valley. The night, too, was
warm for this elevation.
On the morning of the 25th we had to cross to the left of
the tossing river a little below the spot of our camp, known as
Ilegorum. The river is compressed there by mighty rocks to a
width of some
45
feet, and the chasm is spanned by a wooden
bridge 6 feet broad, quite a creditable specimen, I thought, of
Chinese engineering. The sides were protected by a substantial
railing, and the whole painted bright yellow. The opposite
bank for more than a mile further down was formed by a high
and precipitous wall of rock wholly impassable to man or beast.
CROSSING
KOKSEL GLACIER.
103
After some three miles we recrossed by a similar bridge to the
right bank, and could have continued our march there with
ease had it not been that the bridge across the swollen glacier-
stream from a side-valley to the south had been washed away.
The stream was wholly unfordable, and it was necessary to climb
up for some three miles to the mouth of the huge Koksel glacier
from which it issues. It was a trying detour, for the whole
valley is blocked by enormous old terminal moraines. WTien at
last the present end of the glacier was reached, it was with difl&-
culty that we dragged up the ponies to the top of that mass of
ice rising in a bank of at least
150
feet above the river. It was
fortunately thickly coated with glacier mud and detritus, and
in half-an-hour we had safely got the first pony across. From
the eastern side moraine the glacier could be seen stretching
away for miles up the valley to the slopes of high peaks which
were enveloped in clouds. Subsequent surveying showed that
the highest summit of this mountain mass is identical with the
ice-clad Koksel or Sarguluk Peak
(23,470
feet), which rises
prominently at the salient angle of the great range north of
Muztagh-Ata, and is visible even from Kashgar, in the form of a
great truncated cone.
This detour had delayed us much, and it was getting late
in the afternoon when the baggage arrived at the Uttle Karaul
of Gez, from which the valley takes its name. Fresh animals
were to be taken here, but they were said not to have arrived
as yet from their grazing-grounds.
So we pitched camp among
the fantastic
conglomerate formations which line the river-bank.
The evening
brought a fresh duststorm, and notwithstanding
the elevation
about
8,000 feet above the seait felt oppres-
sively warm. On the morning of the 26th of
July
I awoke to
hear the news that the promised ponies had not arrived, and
that the men from Bulunkul together with their animals had
decamped during the night. To make matters worse, the
'
Karaulchis
'
Kirghiz posted at the Karaul for carrying the
Chinese Dakhad also for some unknown reason disappeared.
I did not
appreciate the prospect of being detained inde-
finitely at so dreary and desolate a spot, and the hours of weary
waiting sorely tried my patience. Amidst the sand and bare
104 THKOUGfH THE GEZ DEFILE.
rocks the air grew almost hot as the day advanced. Satip Aldi,
the Kirghiz I had engaged at Karakul, was despatched to the
herdsmen high up the Koksel Valley, but I knew that assist-
ance from that side could not come till nightfall. All the greater
was the feeling of relief when by 2 p.m. a party of traders with
four ponies came in sight, the first travellers we had met since
leaving Karakul. With some persuasion and gentle pressure my
men prevailed on the party to unload their animals and to help
to take our baggage down to the next Karaul, Kauriik-Kurghan.
Heavy loads had to be made up, and we all had to walkno
great sacrifice, for the ten miles down the valley proved excep-
tionally easy-going. The scenery became less sombre as the
valley broadened, and after returning once more to the right bank
at Kok-moinak, a picturesque little cross-spur, I was gratified
by the first view of cultivated fields. The reception that met me
at Kauriik-Kurghan was a cheerful contrast to our Gez expe-
rience. The Kirghiz Karaulchis, among them some picturesque
old men in padded coats of wonderful hues, turned out for my
reception in great style. All the ponies we required for the
difficult marches of the next two days were supplied most readily,
and the services of extra men offered to help the animals over
the difficulties of the track. Kauriik-Kurghan proved a far
cooler place than Gez. The evening, after a light shower from
the east had sprinkled the ground, was delightful, and I felt
refreshed by the scent of the thyme growing profusely about my
camping-place.
The ordinary route to the plains below Kauriik-Kurghan runs
along the bottom of the river gorge, but this is impassable during
the summer months owing to the flood. Then communication
can be maintained only by the circuitous track through the
mountains, known characteristically as Tokuz-Dawan,
"
the
Nine Passes." The first of these passes was up a steep spur a
couple of miles below the Karaul. The hillsides were clothed
with plenty of shrubs, and reminded me of the scenery I had seen
in Buner and in Hazara, east of the Indus. On the top of the
pass a pleasant surprise awaited me. Unexpectedly the patli
opened on a charming glen with trees and fertile patches of oat-
fields. The height was over
9,000
feet. Kauriik-Bel looked a
OVER THE "NINE PASSES." 105
veritable oasis after the stony barrenness of the Gez defile, and
might with some imagination be turned into a suitable site for
a cosy
"
hill-station." Unfortunately the glen holds practically
no water. For a couple of miles we descended it, and then
turned up a narrow side gorge to the north-east.
The scenery had by this time changed considerably. All
vegetation disappeared from the gravel-strewn bottom of the
gorge, and the sides were formed by bare rocks of reddish-
grey tint, worn into fantastic shapes by the influences of chmatic
extremes. At the bottom of the narrowing gorge, cut into the
rock by the action of rain and melting snows, we wound our way
onwards, often in the welcome coolness of overhanging cUffs.
In a little cul-de-sac of rocks there was a tiny stream of cool
water dripping over the stones and losing itself below in the
line sand. Not far from it the Ularlik Jilga
contracts to a narrow
fissure, some 8 to lo feet across, closed in by unscalable crags.
The large rocks which form the bottom are too steep to be climbed
by laden animals. So all the baggage had to be taken off and
carried by men for a short distance. Emerging from this gloomy
fissure we had fresh trouble in pulling the ponies along a series of
rocky ledges and up an exceptionally steep spur. At last the
top of the second pass was reached, with a distant view to the
snowy peaks south-west and a succession of bare serrated ridges
in the foreground looking like lines of petrified waves. Far-
advanced decomposition was plainly written on all features of
the landscape. Winding round the highest ridges of some neigh-
bouring
peak, at an elevation of about 10,500 feet, we reached
at last the head of the Khush-kishlak Valley, the only one in
this maze of mountains which contains a permanent spring, and
where a longer halt is practicable. Dreary and miserable the
place looked, which we reached after a descent of some three
miles between bare hillsides, apparently sandstone. But there
was the spring, fresh and clear, and after the long, hot climb
men and beasts were equally grateful for its blessing.
The march of the 28th of
July
was to bring me right down
into the plains to the large oasis of Tashmalik at the entrance
of the Gez defile. It was a double march, and we started early.
A slight storm had cleared the air remarkably, and when the
106 THROUGH THE GEZ DEFILE.
top of the next pass, the Aktiken-Bel, was reached in the fresh
morning air I was surprised by a dehghtfully distant view. On
the west it extended to the great icy peaks which He between
Muztagh-Ata and the Gez-Darra, most prominent among them
the ghttering cone of Sarguluk. To the east the plains could
be seen far away to the green tracts under irrigation round Yangi-
Hissar, Opal, and Tashmalik. A light haze of dust that hung over
the plain on the pale-blue horizon was all that suggested to the
mind the great desert beyond. Immediately before me was a
maze of bare rocky ridges. The eye revelled in the bright and
wonderfully varied tints which they exhibited. From bluish
grey to terra-cotta every shade of colour glistened in the full
sunlight. It was a view which will long live in my memory
for its vastness and fascinating variety. For a couple of hours
it remained before my eyes as I crossed in succession the Talantik,
Sarvai-Bel, and Topalu-Bel Passes. The valleys between them
were not deep, and progress was easy. Then at last there was
a decided descent into the Kizil
Jilga
("
the Red Valley "), not
inappropriately so-called from the reddish-brown hills that
enclose it. A six miles' march in the dry bed of the stream that
in the spring drains the nullah, brought us at last back again to
the bank of the Gez River. It was flowing here over a bed of
rubble nearly a mile broad, divided into numerous channels, but
as rapid as above.
Only for one and a half miles was there a way along the river-
bank. Then a precipitous spur of conglomerate, which is washed
at its foot by the current of the river, intervenes, and we had to
wend our way again into a tortuous gorge. It felt hot in its
still air, though the thermometer in the shade showed only
83
F. At the end of the gorge there was an extremely steep
ascent at an angle of over 30, where the ponies even without
their loads had difficulty to scramble up. After crossing this
the Shagildik Dawanthere was once more a descent to the
main valley. But our way there was soon blocked by a fresh
spur, and instead of emerging on the level plain for which, I
confess, I was by this time longing, a last great detour into the
wild barren waste of conglomerate hills had to be made. One
after the other of the ponies that l^ad come all the way from
ARRIVAL IX TASHMALIK PLAIN. 107
Karakul broke down, and I was heartily glad that the forethought
of our guides from Kauriik-Kurghan had provided spare animals
to shift their loads on to. After a tiring ascent of close on 2,000
feet the Yamala or Kepek Pass, the last obstacle, was sur-
mounted, and I hurried ahead of my caravan to reach Tash-
malik if possible before dusk. The plain at the point where I
ASCENT OF SHAGILDU
struck it at last by
6.30 p.m. was a stony waste, but pursuing
my way by the side of the river I soon came to a canal and then
to fertile irrigated lands.
Men were still working in the fields
and in all directions rills of water, betokening
by its colour its
glacier
sources, were spreading fertility over the rich soil. It
was a picture of life doubly
impressive after the stony wilderness
from
which I had
emerged.
108 THROUGH THE GEZ DEFILE.
It was getting dark when I arrived at the first houses of
Tashmahk. Thus the oasis is called, and not
"
Tashbalik
"
or
"
Tashbulak," as, distorted by a sort of
"
popular etymology,"
its name appears in almost all maps and books except the
records of the old
Jesuit Surveyors of the eighteenth century.
The mud-walls of the buildings and those enclosing many
gardens looked quite imposing in the dim light. The roads
were lined with willows and poplars. I enquired at first after
the Beg's house, to whom I had sent word regarding fresh ponies.
An elderly
'
Dihkan
'
(cultivator) riding along on a lively donkey
offered to take me there. We passed miles that seemed to me
endless between fields and gardens and little groups of houses.
Yet the Beg's place was ever ahead. I had forgotten that in
Eastern Turkestan extensive groups of villages or hamlets,
spreading over a wide area of cultivated ground, bear a common
name, representing in reality that of a little district. When,
tired out by a ride and walk of nearly fourteen hours, I arrived
at the house, I found to my annoyance that the Beg was away
in Kashgar and that whatever arrangements were possible would
have to be made at the Karaul. It meant riding back in the
darkness for over two miles. But at last I reached the place,
though the pony stumbled and nearly broke down with weari-
ness. The baggage, too, turned up at last, and my tent was
pitched on a field close to racks of fresh-cut scented
*
Beda,'
a kind of lucerne. But it was long after ten o'clock before I
managed to get a
"
wash
"
and close on midnight before I could

sit down to a well-earned dinner.

On the 29th of
July
we were up as soon as the day broke.
The vicinity of Kashgar was an irresistible attraction to hurry
on, and though the available information allowed me to estimate
the distance correctly at close on fifty miles, I was anxious to
cover it that day. There was the usual difficulty of getting
fresh ponies to replace those hired at Kauriik-Kurghan, whose
owners naturally enough would not consent to their going on
to Kashgar. But Alia Beg, the
'
Dakchi ' in charge of the
Chinese post, proved more helpful than the Beg's people, and
by six a.m. I could start with the most needful loads of baggage
packed on a couple of post-horses. The rest of my camji was
THROUGH OPAL 0A616. 109
to follow as soon as the needed animals had been secured. I
was glad to leave its encumbrances behind, for I knew that at
the other end of my long march a hospitable roof was awaiting
me. A greyish haze covered the sky and effaced all view of
the higher hills to the west, but to the north I could dimly
discern the low broad ridge which is fringed by the cultivated
lands of Opal, our immediate goal.
To reach it we had to cross the river from Gez, the Yamanyar
as it is here called, which, notwithstanding all the water drawn
off for irrigation, still spreads in half-a-dozen broad branches
over the plain. The water was
4-5
feet deep in most of them,
and the flow so rapid that it required careful guiding of the
animals by special men stationed to assist at the fords, to effect
a safe passage. After an hour I reached the other side of the
broad river-bed, wet above the knees but without damage to the
baggage. Then followed a delightful ride through the green
grazing land that stretches by the side of the river for several
miles. A little Chinese garrison occupies a dilapidated
ix)st
at
the foot of the low plateau which bears the lands of Opal. Out-
side a circular Karaul we managed to obtain a change of ponies,
but the gain in time it was intended to assure was more than
compensated by the delay which ensued by a quarrel among the
post-men. It was eWdently the question who were to accom-
pany me to Kashgar which excited the commotion. Ulti-
mately I found the baggage subdi\-ided into four small loads and
a villager perched on the top of each laden animal. I acquiesced
in an arrangement which seemed to solve the difficulty, and had
no reason to regret it, for the little caravan moved gaily along
and never stopped till I reached Kashgar.
Opal is a conglomeration of numerous hamlets spread between
fields and irrigated
meadows. To ride along its lanes shaded by
willows and poplars was a delightful change after the dreary
wilderness of stone and sand we had lately passed through. In
the fields the melons were ripening, and richly cultivated gardens
displayed a profusion of vegetables. Everyw^here was the wel-
come presence of water, irrigation cuts of all sizes following and
intersecting the roads. The quantity of reddish mud deposited
by these httle streams was a notable feature. By the side of one
110 THEOUGrH THE GEZ DEFILE.
I made a brief halt to refresh myself with a modest
"
Tiffin
"
carried in a saddle-bag and some apples and plums I had bought
from a wayside stall. It was the first fruit I had tasted for
months.
After an hour and a half we had passed through the whole
breadth of the oasis of Opal. On the stretch of sterile sandy
plain to be crossed eastwards I gathered my first impressions
of the Turkestan desert. Here its look of barren desolation
was mitigated by tamarisks and other low scrub growing pro-
fusely along the now dry ravines that intersect it. To my
surprise the temperature kept comparatively cool even after
midday
;
93 F. in the sun, with a refreshing breeze from the
east, was nothing to complain of. But the glare was trying and
soon forced me to protect my eyes with goggles of neutral-
coloured glass. As we passed one low sandy dune after the other
the pony-men struck up singing, and their tunes, surprisingly
melodious, brought life and a feeling of cheeriness into the soli-
tude. By
3
p.m. we left the desert track behind and emerged
on the edge of the cultivated lands of Tokuzak. Here by the
side of the little Sarai of Saibagh several parties of travellers,
with ponies and rude carts, were resting in the shade of a small
poplar grove. Water kept ready in big jars and a stock of
melons were the refreshments provided at this Turkestan edition
of a roadside inn.
The hours of a pleasant ride which followed will long keep
fresh in my memory. Hamlet after hamlet was passed, en-
sconced among green avenues of poplars, mulberry, apricot,
and other fruit trees. The mud walls of the houses with their
bright yellow and brown looked singularly neat in this setting
of gardens and orchards. Here and there a cemetery with
tombs of sun-dried bricks and a crumbling mosque or Ziarat
built of the same material added a jMcturesque touch to the
rural landscape. The high-pitched songs of my pony-men as
we moved at a rapid amble through the lanes always brought
the children and women before the doors of their homesteads.
The men were busy in the fields, and until we got to the Bazar
of Tokuzak I saw scarcely any grown-up person that was not
occupied in some way. We had so far suffered little from the
RIDE INTO KASHGAR. Ill
dust, for the route we had followed was not the high-road from
Tashmalik, but a more circuitous track connecting hamlet \N4th
hamlet. But at last we emerged on the road and found it as
dusty as it was broad. High lumbering carts or
'
Arabas
'
dragged along by little ponies and droves of donkeys kept up a
continuous cloud. I was getting eager to reach the end of the
journey and too readily gave credence to Sadak Akhun, who
about
5
p.m. assured me that Kashgar was now \\-ithin one
'
Tash's
' distance. I had not yet learned that, away from the
main caravan roads where the Chinese administration has marked
RO.\D TO MR. MACARTNEY'S HOUSE, WITH CITY WALL.
the distances of ten
*
Li ' (approximately two miles)
by small
mud-built towers popularly known as
'
Tash '
(stone), this
measure in Turkestan conveys only the vaguest
estimate of
distance.
It was disappointing when after an hour's ride there was
still no sign of the river-bed which I knew we should have to
cross before approaching the town. At last at a turn of the
road a broad nullah came in sight, with a shallow
stream.
But
beyond it no trace of walls. Minars or other tokens of an Eastern
city. The river was not the Kizil-su which flows past Kashgar,
but only a branch of it known as Ak-su
("
White Stream "), or
112 THEOUGH THE GEZ DEFILE.
Telwichiik. Another three miles or so of endless rice fields
seemed a long distance in the failing light and on the tired
ponies. But at last they too came to an end, and we forded the
truly red water of the Kizil-su
("
Red Stream
").
Beyond it we
wound along dusty suburban lanes where the women with quaint
caps of imposing height sat in groups enjoying a chat in the
twilight.
It was almost dark when the walls of the city suddenly rose
before me. Mud-built as they are, they looked massive and
imposing, while the quaint regularity of their battlements and
square bastions vividly reminded me of many a picture of
mediaeval towns seen in old books of travel. Outside the city
walls all was quiet and dark. The gates were already closed.
At last Sadak Akhun struck off to the left, along a short, poplar-
lined avenue, and the light of a lantern showed me the outer
gate of Mr. Macartney's residence that was to offer me a home
for the next few weeks. Belated as I was, my arrival was not
unexpected, and as I descended from the spacious court to the
terraced garden I found myself welcomed in the heartiest fashion
by Mr. and Mrs. Macartney. Comfortable quarters adjoining
the garden were awaiting me, and when after a needful change
I joined my hosts in their dining-room, there was every little
luxury to favour the illusion that I was in an English home far
away from the Heart of Asia.
f
113
CHAPTER VIII.
STAY AT KASHGAR.
The cheerful impressions of that first evening under Mr.
Macartney's
hospitable roof were a true indication of the happy
circumstances under which the busy weeks of my stay at
Kashgar were to pass. Busy, indeed, they were bound to be
;
for numerous and urgent tasks had to be completed before
I could set out from the capital of Chinese Turkestan for the
proper goal of my explorations. For almost every one of these
tasks I stood in need of Mr. Macartney's experience and active
help. But great as the facilities were which his ofl&cial position
and local knowledge assured to me, I could scarcely have
availed myself of them with full advantage, had not his friendly
care surrounded me from the first with all personal comfort
and encouragement. After two months of almost incessant
mountain travel I felt the need of some bodily rest. Nowhere
could I have combined it more pleasantly with active pre-
parations for the long joiuney before me than in the charming
residence to which my kind Kashgar friends welcomed me.
Chini-Bagh had been a simple walled-in orchard wath a
little garden house, such as every respectable Kashgari loves
to own outside the city walls, when Mr. Macartney, more than
ten years before my visit, took up the appointment of the
Indian Government's Political Representative at Kashgar.
Continuous improvements effected with much ingenuity and
trouble had gradully changed this tumbledown mud-built
garden house into a residence which in its cosy, weU-fumished
rooms now offered all the comforts of an English home, and in
8
114 STAY AT KASHGAR.
its spacious out-houses and
"
compound
"
all the advantages
of an Indian bungalow. Built on the very brink of a high
loess bank overlooking the broad bed of the Tiimen-Darya, the
house and the adjoining terraces command a delightfully open
view over the fertile belt of village land and gardens which
skirt the city to the north. Even through the light dust haze
which is so common a feature of a Turkestan summer day, the
picturesque outlines of the low hill-range beyond gave a setting
to the scene. On repeated occasions when rain had cleared
the atmosphere, I enjoyed distinct views of the great ice-
crowned range north and north-east of Muztagh-Ata, and
also of distant snowy peaks belonging to the Thian-shan
mountains.
After the fatigues and daily
"
rush
"
of the preceding two
months of rapid marching, it was a delightful change to the
well-ordered surroundings of my friends' home. Free for a
time from the petty cares of camp life, I could enjoy in their
genial company all that makes the contact with European
ways of life and thought attractive. Yet I felt as safe as
before from the bustle and outside interference which seem so
difficult to evade during an ordinary European existence.
There was ample reason to feel grateful for the peace and
leisure thus assured to me
;
for I needed them badly for the
manifold labours that now claimed my attention.
Foremost among them was the organisation of the fresh
caravan which was required for my onward journey, and the
explorations in the desert. I realised that my chances of
success in covering within the limited period allowed the
whole of the wide area I desired to visit, depended largely
on the careful selection of the men and animals that were to
make up my party. It was essential to limit the baggage
with a view to rapidity of movement, and at the same time
to ensure that all stores and equipment required for travels
likely to spread over eight months, and under widely varying
conditions of ground and climate, should be kept within easy
reach. I found that, including riding animals, eight camels
and twelve ponies would lie needed for my caravan. The
season vvas not favourable for tlie j)urchase of camels, for
ORGANISATION OF CARAVAN. 115
most of the caravan animals were away engaged on the brisk
summer traffic towards the Russian trade-centres of Andijan
and Almati. But after lengthy trials and negotiations, in
the course of which the local experience and help of Munshi
Bahadur Shah and other members of Mr. Macartney's estab-
lishment proved of great advantage, the necessary complement
of transport and riding animals was gradually secured. The
trouble taken about their selection was, as subsequent ex-
p)erience showed, fully rewarded by the result. For notwith-
standing the fatigues and hardships implied by travels which
covered an aggregate of more than
3,000 miles, none of the
animals I brought from Kashgar ever broke dov^Ti. The
average price paid for the camels amounted to
624 Tangas per
animal, representing approximately Rs.
91,
at the then current
rate of exchange. The cost of the ponies varied considerably,
an average of 260 Tangas, or
38
rupees, being paid for each
of a serviceable lot of baggage animals.
In regard to the personnel, too, of my caravan, it was
necessary to exercise careful selection, in order to keep the
number of followers down to the minimum indicated by con-
siderations for economy and for facihty of supply arrange-
ments. For Mirza Alim, my personal servant from Peshawar,
who had proved not quite equal to the fatigues of rough and
rapid marching in the mountains, I found a very useful sub-
stitute in Muhammad-J
u,
a hardy
'
Kirakash
'
of half Yarkandi,
half Kashmiri extraction, who from long trading experience on
the Karakorum route had acquired much useful knowledge
about ponies. Having served Captain Deasy on his return
journey to India, he had also learned the indispensable rudi-
ments of the art of looking after a
"
Sahib's
"
kit and of serving
at table,
Niaz Akhun, a Chinese-speaking native of Kashgar, whom,
after several unsuccessful experiments ^\'ith other individuals,
I managed with Mr. Macartney's help to secure for the com-
bined duties of
'
Tungchi
' (Chinese interpreter) and pony
attendant, was outwardly a person of more imposing
appear-
ance and of manners to match. He had accompanied
Mr.
and Mrs. Littledale on their great journey through Tibet and
8*
116 STAY AT KASHGAR.
China, and not unnaturally assumed an air of superiority
towards the rest of my Turki followers who had not seen
'
Bajin ' (Peking) and the other wonders of Cathay. It is
only fair to record that they were ever ready to retaliate by
artfully conveyed doubts whether he was truly to be reckoned
among faithful followers of Islam or not rather among his
much-extolled paragons, the heathen Chinese. His relative
intelligence made him useful for his particular function, and
as an interpreter he served me honestly. Perhaps it was just
as well that during the weeks of our stay in the well-ordered
surroundings of Chini-Bagh he had no opportunity to display
before me those little personal failings, such as his inordinate
addiction to opium and gambling, and his strong inclination
to qualified looting, which subsequently caused occasional
trouble. Two young
'
Tugachis,' or camel-men, were engaged
with less difficulty through the traders who had sold me the
camels. Neither Roza Akhun nor Hassan Akhun had seen much
of the world beyond the caravan routes northward. But young
as they were they well knew the difficult art of camel manage-
ment, and prompted perhaps by youthful curiosity and love of
adventure, proved readier to face the hardships and supposed
risks of desert journeys than their elders. It mattered little
that they made up for their cheerfulness and steady conduct
on the march through the sand-wastes by an irrepressibly pug-
nacious disposition whenever the varied temptations of a
Bazar were near.
The numerous repairs and additions the camp outfit needed
were also among the practical preparations demanding early
attention. Saddlery, mule trunks,
'
Kiltas
'
and most other
articles of equipment bore marks of the rough wear to which
they had been subjected on the long journey from Kashmir.
Ever since we emerged from the gorges of Hunza, yaks and
Kirghiz ponies seemed to have vied with each other in doing
damage by knocking and rubbing their loads against every
rock passed on the mountain tracks. The stay at Kashgar
seemed none too long for effecting the needful repairs, for the
Turkestan artisan has none of the imitative skill of the average
Indian craftsman
;
and finding it apparently easy to make
THE KASHGAR
CRAFTSMEN. 117
a
living, he takes little trouble to accommodate himself to the
requirements of the passing European traveller. In view of
such leisurely habits of work and the need of constant super-
vision, I soon ceased to be surprised at seeing Mr. Macartney's
outer courtyard more or less permanently occupied by the
few
'
Ustads
'
(masters) who cared to attend to my orders.
The most
troublesome operation of all proved to be the
preparation of the additional water-tanks which I decided
to get made for use in the desert. The pair of galvanised
iron tanks which had been specially constructed for me at
Calcutta had been safely transported across the mountains.
But the total quantity of water which with due regard to
the carrying capacity of camels they had been designed to
hold, amounted only to seventeen gallons a-piece. The supply
of water thus assured would have been, of course, wholly
inadequate for the needs of a party such as I proposed to take
to sand-buried sites in the desert, and the construction of at
least four more tanks proved indispensable. It was found
that the only material locally available for this purpose con-
sisted of iron tanks in which kerosene oil is transported into
Turkestan from the Transcaspian railway. The adaptation
and strengthening of these much-battered
"
tins," together
with the making of the wooden covers needed for their safe
transport, was a task that taxed the combined resources of
my blacksmith
'
Ustads
'
for weeks.
But luckily the necessity of attending to aU these practical
arrangements did not prevent me from finding time also for
more congenial and equally pressing tasks. Sitting in the
cool shade of the poplar groves of Mr. Macartney's garden, I
spent long and pleasant hours, refreshing by systematic study
my knowledge of the ancient accounts of Eastern Turkestan,
such as the Chinese historical Annals, the narratives of old
Chinese pilgrims and of the earliest European travellers, have
preserved for us. To me it is always a source of pleasure to
be able to read such old records on the very soil to which they
refer. At Chini-Bagh I enjoyed exceptional advantages for
this favourite occupation
;
for Mr. Macartney, whom long
residence and the power of keen observation have made
118
STAY AT KASHGAR.
thoroughly conversant with the economic and social conditions
of modern Turkestan, was ever ready to allow me to ransack
the storehouse of his knowledge for that information without
which the ancient accounts of the country cannot be properly
understood. Often when matters of Chinese lore were con-
cerned Mr. Macartney would summon to our discussion Sun-
Ssu-yieh, the
"
Chinese Munshi " of the Agency, a
literatus
thoroughly versed in his classics, and yet keenly alive to the
things of this world. As I listened to his vivacious expla-
nations, which Mr. Macartney kindly interpreted, I could not
help thinking of my dear old Kashmirian Pandit Govind
Kaul, and the converse I used to hold with him in Sanskrit
during the long years of common scholarly labour. Bitterly
I regretted the great gap in my philological equipment, my
ignorance of Chinese. But how should I ever find the leisure
to fill it, except perhaps in that
"
fresh birth
"
to which, in
accordance with the Indian notion, I used to refer my Chinese
friends ?
It was an important object of my stay at Kashgar to
familiarise the officials of the Chinese provincial Government
with the purpose of my intended explorations and to secure
their goodwill, which I realised would be an indispensable
condition for the practical execution of my plans. In this
direction too I could not have wished for more efifective help
than that which Mr. Macartney accorded to me. Already the
initial visits which I was able to pay in his company to the
Tao-tai, or Provincial Governor, and the other chief digni-
taries were under such expert guidance most instructive to
me and full of interest. In the course of these visits, followed
as they were by
"
return calls
"
and other less formal inter-
views, I was introduced to at least a rudimentary knowledge
of the
"
form
"
and manners which Chinese etiquette considers
essential for polite intercourse. It was no small advantage
to receive this instruction through a mentor so familiar with
all Chinese notions and ways as Mr. Macartney. Every little
act and formality, quaint and strangely contrary to our habits
as it often seemed, thus acquired its due significance, until in
the end when visiting strange
'
Yamens
'
far away from my
VISITS TO CHINESE OFFICIALS.
119
Kashgar friends, I found a comforting assurance in the rigid
uniformity of these observances.
It was essential to secure from the Tao-tai the issue of clear
instructions to the Amban of Khotan which were likely to
assure me all needful assistance in regard to transport, supplies,
and labour, as well as full freedom for my movements. In
view of the serious difficulties which through a sort of demi-
official obstruction Captain Deasy had experienced in the same
region only eighteen months earlier, it seemed doubtful at
first whether the way would be effectively cleared for my
operations, particularly so far as they related to survey work
south of Khotan. That the efforts which Mr. Macartney under-
took on my behalf proved in the end ^tirely successful was
due largely, I believe, to the personal influence and respect
he enjoys among all Chinese officials of rank in the Province.
After a series of interviews and a lengthy correspondence the
Tao-tai agreed to issue the desired instructions. The result
showed that he faithfully carried out his promise, and that
Mr. Macartney's representations, coupled with what explana-
tions I could give through him of the historical connection
of ancient Indian culture and Buddhist religion with Central
Asia, had effectually dispelled any suspicions which might
otherwise have been roused by the intended excavations and
surveys.
In the course of these interviews my references to the
'
Si-5ai-ki,' the records of Hiuen-Tsiang's travels, proved
singularly helpful. All educated Chinese officials seem to
have read or heard legendary accounts of the famous Chinese
pilgrim's visit to the Buddhist kingdoms of the
"
Western
countries." In my intercourse with them I never appealed in
vain to the memory of the
"
great monk of the Tang dynasty
"
(Tcing-Seng). Endeavouring as I now was to trace his foot-
steps through Turkestan as I had done before in more than
one part of India, I might well claim that saintly traveller as
my special patron in the heaven of Arhats.
Strange enough it seemed to me at the time, this pleasant
intercourse with the friendly old Tao-tai, his colleague, the
"
Hsieh-tai
"
or General, and the rest of the local Mandarins,
120 STAY AT KASHGAE.
when I thought of the great pohtical upheaval far away in
the east of the empire. Through the Renter telegrams trans-
mitted from Gilgit and the news indirectly conveyed to us
from Russian sources we knew of the fierce fighting around the
legations and the danger surrounding European settlements
elsewhere in China. Through the telegraph line from Urumchi
to Kashgar the Chinese officials too were receiving accounts
of the great conflagration, and apparently fairly correct ones.
For while Europe was held in horrified suspense by the false
news of a general massacre at Peking, the reports communi-
cated to us from the Yamen, though admitting much fighting,
stoutly maintained that the legations kept up their defence and
were safe.
Disturbing rumours about the conflict with the European
Powers had already in
July
spread through the Bazars of
Kashgar, and just a week before my arrival a feeling of mutual
apprehension and distrust threatened for a moment to bring
about a collision between the Muhammadan population and the
Chinese garrison quartered in the Yangi-Shahr, or
"
New
City." The commotion luckily died away when it was found
that the Chinese Commander-in-Chief, whose visit to the
"
Old
City
"
with an unusual escort had given rise to alarm, had
only come to play a harmless game of cards at the Hsieh-tai's
Yamen ! The recollections of the last great rebellion against
Chinese rule
(1863-77)
have, indeed, not disappeared from
the
"
New Dominions
"
;
but the peaceful cultivators of the
oases and the easily cowed petty traders and artisans of the
towns have little reason to wish back the times when the
turbulent
'
Andijanis ' carried on their exactions in the name
of Islam.
Kashgar, since the days of Sir Douglas Forsyth's mission,
has so frequently received the visits of European travellers
that I may be excused from attempting within the limited
space of this narrative to describe the general features and
life of the city. Most of my time was spent in busy work at
Chini-Bagh
;
and to the little oasis of Anglo-Indian civiliza-
tion which my kind hosts had created around themselves cling
my main recollections of Kashgar. There was little contact
DANGER FROM CHINESE TROUBLES. 121
wnth the outer world to vary the pleasant round of our daily
life. Though the worst of the summer heat of the Turkestan
plains had passed, it was still warm enough during the
middle part of the day to make the freshness of the
morning particularly attractive for work.
So I was regularly
astir with the break of day, and 6 a.m. found me established
beside my books and papers under the tall poplars of the
terraced garden.
The fruit season had fully begun. The closely planted
apricot-, peach-, and plum-trees of the orchard occupying the
upper terrace were already bending low under an abundance
of luscious fruit ; while a little later a fine bower of vines
supplied an equally welcome complement to my
"
Chota
Hazri." As I sat at work I could see and hear the little caval-
cades of cultivators and their women-folk as they gaily rode
along the road between the river and the garden, bringing
their produce to the city markets. Only beggars seemed
to walk on foot, and even they were often provided with
donkeys. Breakfast assembled us as the morning wore on,
in a stately little arbom-, where rows of tail poplars planted
in a square, after the fashion of all Turkestan gardens, gave
grateful shade at almost all hours. There were luckily no
morning papers and daily mails to delay attention to the
work of the
'
Ustads,' who had in the meantime leisurely
settled down to their several tasks. A short stroll taken
round the courtyards after breakfast, usually in Mr. Macart-
ney's company, enabled me to control the progressor other-
wisethat their labours of repair or construction were making.
Then my friend retired to his
"
Daftar ' to wTite his reports
or to go into the cases of his polyglot clientele from across the
Indian borders. Punjabi traders, Hindu money-lenders from
Shikarpur, Ladaki carriers, Kanjuti settlers in Raskam, and
hoc genus otnneall had occasion at one time or other to seek
the presence of the
'
Mulki Sahib
'
(Pohtical Officer) whom
the
'
Sirkar's
' paternal care had planted far away in the
Turkestan capital to protect their persons and interests. I
myself, though
plentifully provided with writing work, ordi-
narily
managed to give an hour or two about midday to the
122 STAY AT KASHGAR.
study ot Turki texts with grave Mullah Abdul Kasim, a
shining academical light of the chief Madrasah of Kashgar.
Muhammadan learning, such as the country knows in these
days of infidel rule, is purely theological. I have no doubt
that the good Mullah would have preferred a discussion on a
knotty passage of some Arabic manual of rehgious law to our
reading of vulgar Turki, even though they concerned the
exploits of that royal champion and martyr of Islam, holy
Satok Boghra Khan. The hottest hours of the day, in the
early afternoon, usually found me in the dark-room I had
improvised in the Hospital Assistant's empty quarters, busy
with the developing of the many photographs I had taken
on the preceding part of the journey. Later on, after tea-
time, a walk with my hosts along the shady village lanes,
or frequently a ride on my newly-bought Andijani mare would
bring welcome recreation. But perhaps the pleasantest hours
were those when after dinner we would sit in the mild evening
air on the flat roof of my quarters adjoining the main house.
From there we could watch the frequent picnic parties of
Kashgar families which had gone out early in the day to feast
on the profusion of fruit in the orchards owned by almost
all respectable citizens in the environs, and which were now
gaily returning in long cavalcades of men, women and children.
Their songs sounded to me very melodious, often strangely
reminding me of airs I had heard long ago on road and river
in Hungary.
As if to remind us of the West, which seemed so distant,
there reached us at times as we sat in the evenings snatches
of Russian airs wafted across by a breeze from the grounds of
the Russian Consulate, half a mile away, while the men of
the Cossack guard were singing in chorus. Frequently we
saw the men on our rides in and about the city, but no oppor-
tunity offered for making the acquaintance of their
'
Sahibs,'
as we should say in India. M. Petrovsky, the Imperial Consul-
General of Russia at Kashgar, to whom, in view of his scholarly
interest in the ancient history and ethnography of these
regions and his activity as a collector of Central-Asian anti-
quities, I was particularly anxious to pay my respects, was
LIFE AT CHINI-BAGH.
123
indisposed and could not receive me. It was not until my
return from Khotan. nine months later, that I had the satis-
faction of making the acquaintance of this accomplished
official.
Apart from the small Russian colony gathered at the Con-
sulate, Father Hendricks and a Swedish missionary, Mr. G.
Raquette, with his wife, were then the only Europeans at
Kashgar, and them we saw often. Father Hendricks, whom
Catholic mission labours had brought many years ago from
Holland, his native land, to Mongolia and hence to Kashgar,
seemed to exemplify in his person the principles of inter-
national amity by being an equally frequent and entertaining
\isitor at Chini-Bagh, the Russian Consulate, the Swedish
Mission, and the Chinese Yamens. The visits of the kindly
Abbe, always bringing a plentiful budget of news and rumours,
impartially gathered from these often conflicting sources of
information, might have gone a long way to console any one
likely to regret the absence of a local newspaper.
I can only briefly mention the remains of ancient structures
which were the object of my first short excursions in the
vicinity. Considering that the site of Kashgar in all pro-
bability corresponds to that of the capital of the ancient terri-
tory of
'
Kie-sha,' which Hiuen-Tsiang describes as possessing
hundreds of Buddhist monasteries, the remains of the pre-
Muhammadan period still traceable above ground are scanty
indeed. The most conspicuous is a much-decayed mound
of sun-dried brick masonry rising over the deep-cut northern
bank of the Tiimen-Darya, about a mile and a half to the
north-west of Chini-Bagh, which undoubtedly represents the
remains of a large Stupa. The present height of the mound
is
85
feet, and the diameter of its base from east to west about
160 feet. But notwithstanding the exact surv'ey made I found
it impossible to ascertain the original form of the whole Stupa,
or even to fix its centre, to such an extent have the masses
of soft brick-work fallen or crumbled away. It was for me
an instructive observation to find that fully
15
feet of the
masonry base now lie below the level of the irrigated fields
close by. I had here the first indication of that remarkable
124 STAY AT KASHGAR.
rise in the general ground level, mainly through silt deposit,
which my subsequent observations on the site of the ancient
capital of Khotan clearly demonstrated. Of a similar though
smaller Stupa mound, a mile and a half to the south of the
city, no details need be given here. Nor is it possible to find
space for descriptions of more modern places of interest which
I visited in the environs. But I may, perhaps, make an
exception in the case of my visit to the
"
New City," the
Chinese cantonment of Kashgar, of which I here give my im-
pressions such as my diary records them.
Mr. Macartney wished to return the call of the Chu-kuan's
or City Prefect's chief assistant, and I myself wanted to profit
by the occasion to do some
"
shopping
"
in things Chinese.
The day was gloriously clear and yet comparatively cool. So
our ride of some eight miles along the broad, well-shaded road
which connects the two cities, was enjoyable even though
the sun still stood high. The branch of the Kizil-su, which
is crossed about midway, was full of reddish-brown water, a
sign that the heavy rain of a few days before had left its mark
in the mountains. Close to the north-west corner of the
"
New City," and not far to the left of the road, rises a stately
complex of buildings, the Chinese temple consecrated to the
memory of Liu-Kin-tang, the great general who, after Yaqub
Beg's death in
1877,
reconquered Turkestan. To this we
rode first.
The temple stands in the midst of a large and well-kept
arbour of poplars, and already its outer court showed, by its
clean appearance and the evident care bestowed on repairs,
that the means are not wanting to maintain worthily the
memory of this modern hero of Chinese power in the
"
New
Dominions." This is fully accounted for by the fact that ever
since the reconquest the general's relatives have exercised a
preferential claim to all the best appointments in the Province.
Through a high gateway decorated with wonderful stucco
volutes we entered the second court. Adjoining the gateway
an4 facing towards the innermost court, is a fine wooden stage
raised some eight feet above the ground. As we ascended
it we noticed with amusement that the walls of what may be
VISIT TO
LIU-Kiy-TAXG'S
TOIPLE.
125
called a sort of greenroom, bore cleverly executed drawings
in charcoal of Europeans.
The temple itself,
approached through a third colonnaded
court, is an imposing hall, lavishly decorated
outside \vith
boards blazing forth auspicious inscriptions in scarlet and
gold. But it is in the inside of the hall that we had
occasion
to admire the generosity and good sense of those who raised
this monument. The whole of the side walls, right and left,
are covered with series of large paintings representing in
sequence the victorious career as well as the administrative
activity and private hfe of Liu-Kin-tang. There are
quaint
but graphic pictures of the battles and sieges by which he
reduced the rebellious province. The characteristic features of
the Andijanis who played the leading part in the revolt, and
of the dark-skinned Tartar people of Urumchi and other
centres to the east, are reproduced with striking fideUty. We
see the general sitting in court, punishing malefactors, and in
all the chief functions of a provincial governor. Other pictures
show him returning to his ancestral home, his meeting \vith
his aged mother, &c. It is a remarkable cycle of illustrations
of a great career, and when properly reproduced, it would be
no mean acquisition for an Ethnographic Museum in Europe.
The richly gilt altar or central fane contains curiously enough
a very Western souvenir of the Chinese general, his portrait
in the form of a photograph enlarged to hfe size.
I could find no representation of any Chinese deity sharing
the honours of the place with the departed hero. But on the
parts of the main wall not occupied by the altar, two large
representations of mythological animals attracted attention.
To the right is a remarkably spirited picture of a storm-pro-
ducing dragon, drawn with a verve and power of imagination
which betoken no mean artist. As I looked at the wonderful
masses of cloud propelled by the dragon's blast, I thought of
old Hiuen-Tsiang's account of the fearful
'
Xagas
'
which guard
the Pamir heights with their winds and icy showers. Chinese
eyes seem still to see the powers of nature as they saw t^jem
then. The royal tiger drawn on the opposite side of the wall
is no match for this noble dragon, and bears in his half-human
126 STAY AT KASHGAR.
face an expression of such utter dejection that we were unable
to withhold our mirth. I wondered whether it was the inten-
tion of the artist to show the noble beast in a mood of hopeless
resignation at its own wickedness.
The attending priest who greeted
"
Ma-shao-yieh," as Mr.
Macartney is called in Chinese, with unfeigned reverence,
seemed a shy and modest specimen of his class. His ways and
dress reminded me of a Lama, freshly emerged from some
forlorn little monastery. He talked glibly enough of the
benefits which the temple enjoys owing to
the generosity of the high officials of the
province, who look to Liu-Kin-tang as a
patron saint watching over their interests.
But when it came to being photographed
an operation to which the average
Chinaman in these parts submits with
relishhe grew hopelessly nervous and
looked as if he were preparing himself for
the worst. The operation passed off all
the same, and a good douceur revived the
spirits of this humble devotee.
The little suburb, through which we
had to ride on our way to the north gate
of the Yangi-Shahr, presented a markedly
Chinese appearance. There were groups
of Chinese soldiers everywhere with the
strangely mixed womenfolk they take
unto themselves in these foreign parts.
The petty traders are also largely Chinese,
and the contents of their booths showed
that they cater for Chinese tastes.
I saw, in fact, the Chinese counterpart
of the Bazar of an Indian cantonment,
only with that easy disregard for order
and appearances which the difference of
European and Chinese notions justifies. Mutatis mutandis, the
soldiers' marketing place outside a Roman Castrum planted some-
where in the East might have offered a spectacle not unlike this.
PKIEST IN LIU-KIN-TANG's
SHRINE,
CHINESE CANTONMENT.
12';
^
jThis suggestion of a Castrum was curiously maintained by
the broad street we entered on passing the gate. It contains
the main Bazar of the cantonment besides various public
buildings, and bisects the whole
"
New City
"
from north to
south. The Chinese shops are more numerous here than in
the
"
Old City," and owing, perhaps, to the ampler light, looked
neater and more inviting. I do not like
"
shopping
"
as an
IN THE BAZAR OF THE
"
NEW CITY,
KASHGAR.
occupation. But there was so much of quaint wares in the
way of dress, household utensils, &c., to be examined that
the time spent in looking for suitable presents to be sent
homewards, such as silks, &c., seemed all too short.
It struck me as an intelligent application to modern con-
ditions that among the neatly-docketed files of correspondence
which one of the merchants showed us in the pigeon-holes
of his snug office, there were letters sent to him by business
128 STAY AT KASHOAR.
friends in Ho-nan through Shanghai and the Indian Post
Office. The addresses on the envelopes, neatly printed in
English, gave the directions as to the route via Colombo-
Rawalpindi-Gilgit and the numerous
"
c b's " needed for
safe transit with far greater clearness and accuracy than one
is accustomed to look for in the case of native Indian corre-
spondents.
Passing on from the shops we paid the intended visit to
Mr. Macartney's Chinese friend, Liu-Lai-chin, who owns a
modest but clean and well-arranged house close to the City
MV SERVANTS FROM KASHGAR AND VARKAND.
Prefect's Yamen. Our host, apart from his official
work,
has no mean reputation as an artist, and I had already had
occasion to admire the charmingly decorated fan which he had
painted as a present for Mrs. Macartney. Sitting in the well-
lit and large room which served as office, studio, and drawing-
room, it was a pleasure to examine the delicately
worked
scrolls and paintings with which the artist-official had adorned
his walls. Even more, perhaps, I enjoyed watching his eager
and animated conversation with Mr. Macartney. It turned
in part on antiquarian questions which my references to Hiuen-
Tsiang's account of the
"
New Dominions " had raised.
Though only small portions of the discourse could be made
AN ARTIST-OFFICIAL. 129
intelligible to me through Mr. Macartney's kindness, it was
impossible to mistake in his friend's remarks the spirit of true
historical interest, that connecting link of Chinese and Western
thought, I shall ever retain the pleasantest recollection of
the air of culture and refinement which seemed to fill this quiet
household.
CARAVAN STARTING pROM KASHGAR.
CHAPTER IX.
KHANUI AND ORDAM-PADSHAH.
On the morning of the 4th of September, I was ready to take
the field again. The five weeks of refreshing rest at Kashgar
had passed so fast that there seemed at the end scarcely time
enough for the completion of the multifarious preparations
that my journey eastwards demanded. My start for Khanui,
where ancient remains had been reported, was a move of dis-
tinct practical utility. It gave me an opportunity for a kind
of experimental mobilisation of my caravan. As the site to
be examined lay only one long march to the north-east of
Kashgar, and as I should have to return to the latter place
before proceeding on the journey to Khotan, it was easy to
discover and remedy in time any deficiencies which in tlu
ENVIRONS OF
KASHOAR.
131
hurry of the preparations had been overlooked. On the other
hand, it meant a definite start, and duly impressed all
'
Ustads
'
with the necessity of completing
their tasks.
My amiable hosts had made light of my announcement that
the morning of the 4th of September would see my caravan on
the march and their compound clear of the motley accession
of servants and followers I had brought there. They had seen
several visitors start with a delay of half a day or more, such as
illustrates the delightful dilatoriness of Central-Asian travel, and
were no doubt prompted in their sceptical predictions also by
the wish to extend their kind hospitality yet a little longer. I
was, therefore, not a little pleased to find that men, camels, and
ponies all fell into their places without much trouble in the early
morning. A preliminary weighing of all baggage allowed its
quick arrangement and loading. The
'
Ustads
'
had managed
to finish their labours late on the preceding evening. No time
was taken up with the men's leavetakingit could be left for
the final moveand thus the caravan, to my friends' sur-
prise, was ready to start when I joined them at breakfast.
The camels seemed anxious to emphasise their exemplary
punctuality. For loaded as they were, they started off, nobody
quite knew by whose order, before breakfast was finished and
my hosts prepared to take their intended photograph of my
caravan. They had not got far, however, and were promptly
brought back to figure in front of my friend's newly arrived
camera. The Begs, whom the City Prefect or Hsien-kuan of
Kashgar had sent to escort me, were also in attendance, and
gave in their Chinese get-up an additional element of pictures-
queness to the scene.
When the procession of trimly packed camels with their
little escort of mounted Begs and servants had passed on for a
couple of miles, I left the hospitable roof of Chini-Bagh, glad
at heart that it was not yet a real good-bye to the friends who
had treated me with such kindness. My way lay past the
Yarbagh Gate of the city and along the foot of its northern
walls. Then the river had to be crossed by the bridge of Tar-
bugaz, with its picturesque little Minars at either end. It was
no market day, yet the stream of mounted peasants, droves of
9*
132 KHANUI AND ORDAM-PADSHAH.
loaded donkeys and ponies that passed through the adjoining
Bazars, might elsewhere have suggested a great fair. Where
the road, flanked by cemeteries, turns off to Hazrat Apak's
shrine, the true Via Appia of Kashgar, I overtook the camels,
and then rode on between suburban gardens and through shaded
BEGS AND AKSAKAL OF BESHKARIM.
village
lanes northwards to Beshkarim. This large collection of
hamlets
which lies on the caravan route towards the distant
Narym
and which has since terribly suffered in the great earth-
quake of
1902,
was reached about midday.
In the central village, known as Beshkarim Bazar, a grand
SHRINE OF HOLY MAIRYAM.
133
reception awaited me. There was the Beg of the Httle district,
a cheerful Kashgari dressed in orthodox Chinese fashion, ready
to welcome me, and on the terrace of his
'
Yamen,' under shady
elms, a plentiful
'
Dastarkhan
'
of fruit, tea, and sweetmeats.
It was a pleasure to sit down in the airy v^erandah and to partake
of the delicious fruit then abounding all over the country. For
my companions and servants there was no lack of more sub-
stantial refreshments in the form of soups, large plates of
mutton, and mountains of flat cakes. The broad market-place
in front was filled with men and ponies, each \'illage headman
summoned for the occasion having evidently brought a little
troop of mounted followers. To walk on foot is an exertion
left in these parts only to the poorest.
At Beshkarim Bazar nearly an hour passed, there being no
need to hurry on in the heat of the sun before the baggage turned
up. Then we mounted and with a follo\nng swelled by the local
Beg's people and the headmen under him, pursued our eastward
way. For some eight miles the road wound through highly
cultivated strips, along irrigation canals of varying size, all fringed
with rows of poplars and willows. Towards three, we emerged
on less fertile ground, and I was expecting to get near the edge
of the cultivated area when there rose before me a grove of
magnificent old poplars. It was the arbour enclosing the shrine
of Bu (Bibi) Mairyam Khanum, the saintly wife of Satok Boghra
Khan, the popular hero of Muhammadan local tradition.
Entering the courtyard through a small gate in the high mud wall
enclosure, I found a delightfully shady quadrangle of old poplars,
and in its centre under a grand elm tree a fresh collation. I was
not prepared like my better-trained followers to attack large
heaps of melons and other fruit after so short an interval. So
I strolled away into the inner part of the garden, where there
were cool, shady walks between roomy tanks of water, and a
fine
wood-built prayer-hall. In smaller buildings half-hidden
behind the trees and intended for the accommodation of pil-
grims, I could hear voices reading chapters of the Koran. But
it was no time for popular pilgrimages, and the inmates were
only a few itinerant mendicants and
'
Talib-ilms ' or theo-
logical students.
134 KHANUI AND ORDAM-PADSHAH.
Beyond the garden stretched barren, sun-baked ground,
filled with graves and tombs in all stages of decay. In its centre
rises the simple but massive cupola which covers the resting-
place of the holy Mairyam (Miriam). Yaqub Beg, or Bedaulat
as he is popularly known, had raised it with hard-burnt bricks,
and the good condition of the building, which has seen no repairs
since the death of its founder, speaks well for the solidity of the
construction. All around are only crumbling ruins, mud walls
slowly mingling with the loess dust from which they were made.
The shrine has kept some of the land left to it by former pious
benefactors, but it would be against all Eastern notions if any
of the proceeds were spent on repairs. The feeding of poor
pilgrims and of the ever-present Darwishes is a more urgent task.
At four my hosts and guides had finished their feast of melons
and meat with a pious prayer, and we resumed the march.
The single canal along which we rode could not supply water
for the whole plain. From where we entered the limits of the
land of Khan-ui
("
the Khan's residence "), stretches of desert
ground could be seen both to the north and south. By
5.30
p.m. the last hamlet eastwards was reached. To proceed for
camp to the old site which I intended to visit was impossible
on account of the want of water. So I gladly assented to the
proposal to pitch my camp in a large
'
Bostan
'
or arbour of
the hamlet. I found it half-swamped by an excess of irrigation
such as this thirsty soil is, no doubt, in periodical need of. But
the raised avenues between the cross rows of poplars were above
the water and at one point left room enough for my tent. So
the pleasure of being among trees and green hedges overcame all
sanitary scruples, and when the camels arrived liy nightfall I
felt quite pleased with my new quarters. The moonlight
glittered brilliantly on the water-logged ditches and fields, and
around ruled a delightful silence, a foretaste of the desert that
lay so near.
On the following morning I started with quite an imposing
cavalcade for the ancient site. A mile to the east of my camp
all cultivation ceased, and a little further all trace of vegetation
disappeared. At a distance of about two miles the low ridges
of hard-baked loess were covered with fragments of old pottery,
STUPA RUINS NEAR KHANUI. 135
glass, and slag, unmistakable evidence of ancient habitations.
But no other indications remain of the buildings
;
the walls of
mud or sunburnt bricks of which they must have been con-
structed, have long ago disappeared, mainly, as subsequent
experience showed me, through the erosive action of wind and
sand. The people know this ruined waste by the name of Hasa-
Tam, and suppose it to have once been the capital of a
'
Chinese
Khakan ' until
'
Hazrat Sultan,' i.e., Satok Boghra Khan,
destroyed it.
In the middle of the pottery-strewn area, where the ground
is slightly raised, I found a tent pitched by the Hsien-kuan's
order and a fresh Dastarkhan spread. But it was too early
to indulge in such comforts, and the hopes of my guides for a
continuation of yesterday's picnic series were doomed to dis-
appointment. The view from the rising ground displayed to
the south nothing but a desert plain hidden on the horizon by
an ominous dust haze. But to the east and north I could make
out a few mounds rising high above the low banks of loess and
sand. Sop Niaz Baba, the fine-looking old Aksakal of Besh-
karim, who knew the neighbourhood well, spoke of these emi-
nences as
'
Tims,' the designation current in these parts for
mounds formed by ancient structures. So I set off straight to
the mound on the eastern horizon, and when I arrived there after
a three miles' canter found to my satisfaction that I had got to
the remains of a Stupa. The ravages of time had reduced it to a
shapeless little hill. But the masonry of sun-dried bricks of
which it was formed, displayed itself plainly at several points
below the covering crust of earth. Immediately to the S.W. of
the mound I could trace in low banks rising above the level of
the surrounding country the remains of a great quadrangle,
measuring
260 by
170
feet, undoubtedly the monastery once
attached to the Stupa.
I had scarcely began a rough survey of the site when a \nnd
of increasing violence rose from the north. The mountains that
were at first visible in the distance soon disappeared in the thick
haze of dust, and with them too the mounds that had been pre-
viously
pointed out to me. But my guides knew their position
well and set off without hesitation when our work at Topa Tim
136 KHANUI AND ORDAM-PADSHAH.
("
the Sand Mound
")
was finished and I wished to proceed to
the
'
Tims
'
northward. The four miles' ride in the face of a
sandstorm was not pleasant in itself, but revealed an interesting
topographical fact. Unexpectedly I came upon a deep-cut
river-bed, now almost completely dry, but showing by its great
breadth, of some
500
feet, that it must once have carried a con-
siderable quantity of water. My guides assured me that only
after exceptionally heavy rain in the mountains does this ravine
now contain any water. It must have been different in earlier
times, for it is evident that the water supply of the ancient settle-
ment was derived from it.
There was another interesting discovery awaiting me at the
end of the ride. As the mounds I was bound for emerged from
the thick yellow haze, I saw to my surprise that one of them
was a comparatively well preserved Stupa, closely resembling in
its dimensions and proportions the Buddhist monuments of this
kind on the Afghan border and in the northern Panjab. It
rises on a three-storied square base, with the hemispherical
dome above almost intact, to a height of nearly forty feet. The
position it occupies, on an isolated tongue of high ground sloping
down from the foot of the mountains, makes the structure look
still more conspicuous. Behind the Stupa, which still retains
on the less exposed southern side portions of the original coating
of plaster, I found a great oblong mound of far greater dimen-
sions but much more dilapidated. Traces of niches or cells in a
three-fold row preserved on the side least exposed to the winds
and the rain suggest that it may be the remains of a monastery
attached to the Stupa. On the platform connecting the two
structures I could trace low crumbling walls of several small
buildings.
The old Aksakal of Beshkarim told me tluit the people know
the site by the name of Mauri Tim, and look upon the
'
Gum-
baz,' i.e., the Stupa, as the watch-tower of the fabulous
"
King
of Chin and Machin
"
who resided in the ancient city before
Harun Boghra Khan destroyed it. That the Stupa goes back
to pre-Muhammadan times is quite certain, and judging from
its shape and proportions I should be inclined to date it several
centuries previous to the arrival of Islam. It has not escaped
2
-
138 KHANUI AND ORDAM-PADSHAH.
the ravages of man, for on the western side I found a deep cut-
ting, in all probability made long ago by treasure-seekers. It
has been carried to the centre of the hemispherical dome and
reveals the interesting fact that this Stupa contained, like others
examined by me in Swat and Buner, a small square chamber
probably intended for the deposit of relics. This chamber was
near the top of the dome, and below it a narrow square shaft
can still be made out descending into the base.
It must be due to the dryness of the air and the absence of
destructive climatic influences that the rows of sticks supporting
the plaster mouldings of the circular base immediately below
the dome were found still intact. The wood, though certainly
over a thousand years old, seemed scarcely to differ in touch and
toughness from tamarisk branches dried for a few weeks.
The wind with the blinding dust of sand it carried along
was too violent that day to permit the taking of photographs
and measurements. But I returned from Khanui on September
6th, and a day's work with the plane-table and photo-theodolite
gave me a complete survey of the ruins. I chose for my second
day's quarters Eski, a pretty village some eight miles south-west
of Mauri Tim, where my tent was pitched among groves of vines
and luxuriant fields of Indian corn. On my way back about
three miles from Mauri Tim, I examined a curious structure
about 22 feet square, open at the top and showing thick walls
of clay cast in moulds. The name Kaptar-Khana
{"
the pigeon
house "), by which the people know it, is derived from the rows
of little niches, about ten inches square, which line the whole of
the inner sides of the walls still rising to a
height of sixteen feet.
The ground inside was thickly strewn with fragments of human
bones, and local tradition asserts that it has always been in
this condition. Nothing at or near this desolate structure
afforded evidence as to its date, but its shape and apparent
purpose curiously recalled a
*
Columbarium.' Neither Bud-
dhist nor Muhammadan custom would allow of such a disposal
of human remains. Is it possible, then, that this strange ruin
is a relic of the times when Kashgar held a considerable popula-
tion of Nestorian Christians ?
On September 7th Ram Singh was sent on survey work to
DEPARTURE FROM KASHGAR.
139
the south-east while I rode back to Kashgar, greatly pleased
with the instructive little tour I had made and the attention
shown to me by the local officials. The short excursion to
Khanui had been useful in bringing to notice various deficien-
cies in the outfit of my caravan, chiefly concerning camel gear.
Some of the animals showed bruises due to ill-fitting saddles,
while the knocks suffered by some of my bo.xes plainly indicated
the necessity of proper packing crates. So the
'
Ustads
'
were
set to work again, and by dint of continual pressure managed
to complete the desired alterations and additions within two
days. On September loth the camels with the main body of
my camp establishment marched off to Khan-arik, where Ram
Singh had proceeded direct from Khanui. Thus, when on
the morning of September nth, I set out from Kashgar on the
journey that was to take me via Yarkand to Khotan and the
field of my explorations, there was no imposing caravan to
give eclat to my departure, but also no final preparations to
cause worry or delay. On the preceding evening, a dinner
given by the Macartneys allowed me to say a quiet good-bye
to those members of the European community with whom
I had become acquainted.
On the morning of the nth I bade farewell to my hosts,
whose inexhaustible attention and help had rendered the long
halt at Kashgar far more pleasant than I could reasonably have
hoped. In the outer courtyard of Chini-Bagh there was quite
a little crowd, composed of Mr. Macartney's native staff and
others connected with the Agency. Mr. Macartney himself
accompanied me round the city walls and through the suburbs
to the point where the high road towards the south enters open
country.
For my march to Yarkand I had chosen a route east of
the ordinary caravan road, so as to traverse the desert tract
containing the famous pilgrimage site of Ordam-Padshah.
Though visited before by members of Sir Douglas Forsyth's
mission and by Dr. Sven Hedin, the exact position of this
shrine had never been fixed. The opportunity thus offered
for new topographical
work and the useful experience of a short
desert trip preliminary to longer excursions were an ample set-off
140 KHANUI AND ORDAM-PADSHAH.
for the slight detour. The first few miles of my ride took me
along the road leading to the
"
New City," but when I had once
passed the busy Bazar under its bastioned wall with the lounging
crowd of Chinese soldiers and those who live on them, I was
able to pursue my way free from the bustle and dust of the high
road. The village lanes along which I rode, guided by the Beg
whom the attention of the Chinese district officials had provided
for me, gave welcome shade from poplars and willows. At
Yonduma, some twelve miles from Kashgar, I passed over one
of the streams into which the Yamanyar, from Tashmalik.
divides. Beyond it lay a wide tract of fields of Indian corn and
grazing lands, irrigated by a network of shallow canals. At
Dangalchi I made a short halt in a shady little garden, and then
at a distance of some twenty-eight miles from Kashgar reached
Khanarik,
The Bazar into which we rode turned out to be only one of
the five market villages which belong to Khanarik, and on
inquiry I found that my camp had moved from this, the
"
Monday Bazar," to the
"
Sunday (Yak-shamba) Bazar,"
ten miles further east. The ride had been long and fairly hot,
so this announcement was not particularly welcome. But there
was nothing for it except to ride on. Towards six in the evening,
after passing a strip of barren land which intervenes between
these parts of the Khanarik tract, I was met to my surprise by
a solemn assembly of well-dressed men.
They turned out to be the Hindus of Khanarik, Khattri
moneylenders from Shikarpur, who had ridden ahead to welcome
the
"
Sahib." It was strange to meet in these rural surround-
ings, so closely resembling those of a European village, the
representatives of a class which thrives all through the Punjab.
However little sympathy the calling and general character of
these men can claim, it is impossible not to feel some satisfaction
at the pluck and enterprise which enables them to carry on their
operations so far away from their home. The connection of the
Shikarpur Banias with Central Asia is undoubtedly an old one.
Already in the eighteenth century Forster found them estab-
lished as far as Samarkand and the Caspian. But the opening
up of Eastern Turkestan to Indian trade within the last thirty
HINDUS AT KHANAEIK. 141
years seems to have attracted them to these parts in increased
numbers.
From the men who greeted me on my approach to Yak-
shamba Bazar, I learned that Khanarik supports no less than
eighteen Shikarpuris. Such an allotment of Hindu usurers to
a single village tract, however large, can only imply the pro-
gressive indebtedness of the cultivators, and my informants
readily admitted that business was brisk. They had all settled
down here during the last eight years, and their well-to-do
1
142 KHANUI AND ORDAM-PADSHAH.
The garden of the Beg in which my camp had been pitched
was a large and well-secluded place, and consequently I enjoyed
a quiet evening after my long ride. The Hindus, true to their
native custom, brought a
'
Dali ' of fruit and sweets, and
would not rest satisfied until I accepted some pomegranates and
almonds for myself and melons and sugar-balls for my people.
The spokesman of the guild was Parmanand, the wealthy banker
from Aksu, who had left his distant place of business to look
after some debtors in this neighbourhood. He assured me in
advance that I should find no difficulty in getting my cheques
cashed in Aksu !
On the following day my march lay to Achchik, the last village
of Khanarik southwards in the direction of Ordam-Padshah.
The distance was only about twelve miles, but a deep-cut river-
bed about half-way proved a serious obstacle for the camels.
The rickety bridge that spans this branch of the Yamanyar,
was scarcely safe even for the ponies
;
it was certain that camels
could iiot be taken across it. So we had to wait patiently in
the scanty shade of some willows until the slowly-moving
animals arrived
;
then to arrange for the unloading of the
baggage, which had to be carried over piece by piece. Finally
a suitable spot was found nearly a mile higher up, where the
banks sloped down less steeply. It was a troublesome affair
to drag the shaggy quadrupeds into the water one by one. But
once in it they swam better than I had expected, and guided by
two villagers swimming in front managed to reach the opposite
bank safely. The sum of money invested in them amounted to
over seven hundred rupees. So it was with a feeling of relief
that I saw, after a delay of nearly three hours, the whole caravan
on the march again.
Defective cultivation and patches of barren land on the
way to Achchik showed that we were approaching the edge of
the desert. But Achchik itself proved a cheerful place. The
Yiizbashi, or village headman, had prepared his house for my
reception. The rooms looked inviting with their freshly-
plastered walls and the plentifully-spread carpets of Khotan
felt. But the light and air were rather scanty, and so I pre-
ferred to pitch my little tent in a neighbouring field, where the
ON THE DESERT EDGE.
143
lucerne crop had just been cut. It was a delightful evening,
with a distant \'ista over fields of wheat and Indian com, hedged
with poplars and mulberry trees. Like many a rural view of
these parts, it carried me back to the fertile Alfold of Hungary.
At Achchik I heard of a
"
Kone-shahr," i.e., a ruined site
of some sort, that was said to be on the edge of the desert to
the south-east. My informants were unable to give any idea
of its exact distance, but believed that it could be reached by
a detour the same day while my camp moved to Ordam-Pad-
shah. A short visit seemed all that was needed, as no buildings
were said to be extant, only scattered heaps of bricks and patches
of ground covered with potsherds.' So I directed my caravan
to start with one guide straight to the desert shrine southwards,
while I myself with another guide and the Sub-Surveyor rode off
at eight o'clock to take Baikhan, the old site mentioned, on the
way. Soon beyond Achchik cultivation ceased, and we entered
a wide, scrub-covered plain of sand and loess. Neither beast
nor man was seen until we reached Khuruz, a miserable hamlet,
about four miles to the south-east. A little watercourse allows
the lonely dwellers of the few scattered huts to irrigate some
fields. Another four miles' ride over a similar waste brought
us to Nurunam, where some shepherds live in a couple of
wretched hovels. Every tree forms a distinct landmark on this
dreary plain. So we had no difficulty in fixing our position on
the plane-table as we moved along. By midday, when the heat
grew intense, we reached Bekhtauruk, another collection of huts
by the side of a swampy depression fed by a little canal. But
of Baikhan nothing was to be seen, and e\"idence as to its dis-
tance was conflicting. The prospect of being benighted over
our search, or of havdng to traverse the sand-dunes towards
Ordam-Padshah with no daylight to guide us, was distinctly
uninviting. I therefore felt bound to abandon the projected
visit and to make for the shrine where I supposed my caravan
to have preceded me.
So after securing a shepherd guide, we set off to the south,
and gradually approaching the line of white sandhills, after
an
hour and a half's ride entered the true desert area. All scrub
disappeared, and only hardy tufts of grass known as
'
Kumush
'
144 KHANUI AND OEDAM-PADSHAH.
covered in patches the glittering sand. With the hope of a more
extended view we made for a higher sandhill. Far away to the
south stretched a sea of sand, curiously resembling the ocean
with its long wavelike dunes. Through the dust-haze that lay
over the long succession of ridges there appeared to the south-
west a darker range of low hills for which the extant maps had
in no
way prepared me, and nearer to us a series of high posts
marking the sacred sites to be visited by the pilgrims. With
the help of these far-visible marks it was easy to ascertain the
position of Ordam-Padshah, as well as of the subsidiary
'
Mazars
'
of Dost-bulak, Sultanim, and Kizil-jaim.
Following our guide, we struck to the south until we reached
the main route of the pilgrims near a lonely rest-house known
as Uftu Langar. There we looked in vain for the track of the
camels which we expected to have passed long before. After a
long and somewhat anxious waitfor it was getting latewe saw
at last far away to the north the caravan emerging from behind
the sandhill. Assured that I should not have to wait in vain for
the baggage, however belated it might be, I rode on in the twi-
light towards our destination.
The sand-dunes to be crossed steadily increased in height, and
the going became more difficult. Even in the faihng light
it was easy to make out the series of semi-lunes into which the
drift-sand forms under the action of wind. In the intervals
between these ridges the ground was fairly hard and white
with alkaline salts. The ponies' feet sank deep into the loose
sand, and each ascent of thirty to forty feet was thus a tiring
performance. The lines of sandy ridges ran mostly from south-
west to north-east, the steep inner sides of the semi-lunes facing
all to the south-east. After a tiresome march of some five miles
from where we first entered the moving sand region, we drew near
to our goal. A long, open valley appeared between the dunes,
and at its entrance from the north we could make out a group of
stunted poplars. They grow near a well of brackish water,
which is carefully protected by a wooden shed from the advance
of the neighbouring sandhills. The water-surface was at the
time some six feet below the level of the artificially cleared ground
in front of the shed.
FIRST CROSSING OF SAND-DUNES. 145
It was nearly dark when we reached there. But the water
tasted so bad and the neighbouring rest-house looked so dilapi-
dated, that I readily moved on to the main settlement of the
desert shrine some half a mile off. There I found a collection
of huts and Sarais built for the accommodation of the local
custodians or
*
Mujawirs
'
and their pilgrim visit(ws.
One of
the rest-houses had been cleared for my party, and
there our
ponies found grass and water. I myself was glad to discover
at some distance a spot where the ground was firm enough
for
pitching my tent, and where I was safe from the odours
that
rose from the accumulated refuse-heaps of this strange
settle-
ment. It was a long wait till the baggage turned
up, towards
eight o'clock
;
but in the pure desert air the evening
breeze from
the east felt delightfully fresh, and when at last the late
dinner
appeared and I could retire to rest I had almost
forgotten
the
fatigue and heat of my first day in the long-looked-for
desert.
The morning showed me my surroundings in their
true
colouring
; the little plain on which my camp
was
pitched
;
the waves of drift-sand in front and behind
; the
dilapidated
mud-built huts and Saraisall displayed the same
monotonous
khaki. Even the sun while low down seemed to shed a grey
light. I felt pleased to note how well my tent and clothes
harmonized with this monochrome
picture.
A sand-dune
some thirty-five feet high, which rises immediately
behind the
mosque and threatens to bury this modest
structure
before long,
gave a panoramic view convenient
for the plane table.
From
its top we could make out the various
'
Langars
'
(travellers'
shelters) and shrines to the north, and thus
exactly
fLx our
position.
A reference to the available
maps
showed that
Ordam-Padshah
had been
placed fully half a degree of longitude
out of its true position.
The miserable looking
Mujawirs of the place had followed
me to the dune, and now related
the story
how the holy Sultan
Arslan
Boghra had
succumbed
on this plain to the attack of
unbelievers,
i.e., the
Buddhist
antagonists
of Islam,
and how
by a miracle the slain
bodies of the faithful
were found turned
towards
Mecca,
whereas
the sand
swallowed
up the remains of
ihe infidels. Half
a mile to the west there
rises a stack of high
10
146 KHANUI AND OEDAM-PADSHAH.
poplar staffs, marking the supposed resting-place of the sainted
king. Like the staffs over all Ziarats in the country, they were
covered with little flags and rags of all kinds, ex-votos of pious
^
pilgrims. In a depression about half-way to the proper
'
Mazar
'
is the well used by the attendants of the shrine. They all claim
to be descendants of the Sultan. Low mud walls on a fiat piece
of ground, a little to the west of the line of sandhills now ap-
proaching the extant houses, were shown to me as the remains of
PILGRIMS SARAI AT ORDAM-PADSHAH.
a former settlement. These ruins probably mark the position
of houses which have been overwhelmed at a previous date by
the advancing dunes and left bare again when the latter had
passed by in their gradual movement to the south-east. The
same process may repeat itself in due time with the present
houses of Ordam-Padshah.
Notwithstanding the holiness of this curious place of
i>il-
grimage my men were anxious to leave it as soon as possible.
So my caravan was already far ahead when I started from Ordam-
Padshah. The route to Yarkand lay to the south vid Hazrat-
SHRINES IN THE DESERT. 147
Begim, another sacred site on the edge of the desert. The going
was heavier even than on the preceding day, for the Hnes of
sandhills were closer together and the direction to be followed
made it difficult to utilize the narrow strips of comparatively firm
ground that separate the successive waves of sand. My little
dog felt so miserable in the basket in which he was to ride on a
camel that I had allowed him to follow me on foot. But the
sand and the heat told on him before long, and I was glad when,
after about four miles, I picked up the camels again and could
safely instal
'
Yolchi Beg
'
in his lofty seat. A hole pro\4ded
in the top of the basket allowed him to look about \\-ithout
gi\nng a chance of escape.
The sand-dunes seemed to grow in height as we slowly ap-
proached the previously mentioned ridge to the south-west.
But at last the patches of hard loess became larger and larger
as the level rose and the ascent became perceptible; The
ridge which had looked so high from a distance through the haze
proved only about
300
feet above the sandy plain. Its pebble-
strewn slopes bore a curiously scarred and withered look, testify-
ing to the force of long-continued erosion by wind and sand.
No stone or distinct formation of conglomerate appeared on
the slope, swept clean as if with a brush.
On the top of the ridge a number of high staffs serve as a
directing mark for the pilgrims. So the place bears appro-
priately the name of Ulugh-Nishan
("
the High Standard
").
Arslan Padshah is believed to have addressed from there his
last prayer to the holy Beg, his adviser, who lies buried at
Hazrat-Begim
("
My Beg of holiness
").
The latter shrine
was visible to the south-west, and as the slope is far steeper
on that side and quite clear of drift-sand, we soon reached it.
Hazrat-Begim has little to detain the traveller, for around
the modest mud-built quadrangle enclosing the saint's tomb
there are only a few wretched huts of Mujawirs and a sandy
plain strewTi with bones and refuse. The camels were, how-
ever, tired by the ten miles' march through the deep sand, and
Kizil, the next inhabited place, was too far to be reached that
day. So my tent was pitched on what I suppose to have been
an old burial-ground near the shrine. The water from the
10*
148 KHANUI AND ORDAM-PADSHAH.
well close by tasted extremely brackish, and neither filtering nor
the lavish use of
"
Sparklets
"
could make it palatable.
On the morning of the
15
th of September I resumed my
march across the plain, which gradually turned into a scrub-
covered Dasht of hard loess. At Saduk-Langar, some four miles
off, I hailed with pleasure a little green oasis, created by a small
watercourse. It is a Waqf or endowment for the benefit of pil-
grims
;
so we could with a good conscience allow our ponies
to graze awhile in the few lucerne fields. By 2 p.m. Kizil was
reached, a large village on the main road that connects Kashgar
with Yarkand. The sight of its green fields and gardens was
truly delightful after the mournful desert behind us. My
servants made straight for the Chinese rest-house and seemed
surprised when I objected to putting up at that dusty caravan-
serai, with its courtyards full of carts, donkeys, ponies, and
their attendants. Sadak Akhun gravely asserted that the
'
Sahibs ' coming from Kashgar
"
always " stopped there.
But then I came from Hindustan, and had learned by long ex-
perience that the places where
"
everybody
"
camps are usually
the least attractive. So I set out to search for a camping-
ground, and after a while found what I had looked for. A
charming little orchard surrounded by open fields gave
room and shade for my tent, while the owner hospitably
welcomed my followers in his house a short distance off.
Grapes and excellent peaches were soon forthcoming, and I
feasted on them in honour of my return from a first visit to
the desert.
The following day, the i6th of September, was spent on a
long march through an arid waste to Kok-robat, the western
limit of the great oasis of Yarkand. For a distance of close
on twenty-four miles there was neither a tree nor even a shrub
to be seen, only the gravel-covered grey
'
Dasht ' far away
to the dusty horizon. As we were now on a post route I found
a square, mud-built tower marking each
'
Potai,' the
Chinese road measure equivalent to ten 'Li.' As the Potai
seems to correspond closely to a distance of two English miles
it is evident that the value of about one-fifth of a mile still
holds good for the Li
in Turkestan, as it does by computation,
OASIS OF KIZIL. 149
for the road distances recorded in India by Hiuen-Tsiang and
other Chinese pilgrims.
I stopped awhile at midday at Ak-robat
("
the White
Station
"),
a solitary Sarai in the desolate waste. I found
the little rest-house within the enclosure, evidently intended
for Chinese officials and better-class travellers, surprisingly
clean, and gratefully availed myself of its deep, shady veranda
for a short rest while the camels came up. It was nearly five
o'clock before my eyes again rested on green fields and trees.
Kokrobat
("
the Green Station
")
receives its water, and with
it fertility, from a stream coming from the hill range that was
dimly visible in the west. I had to ride through the main
village, spreading its houses in a single street over a mile long,
before I found an arbour suitable for my camp. I could not
have desired a shadier or more secluded grove. Curiously enough
there was no proper entrance through the wall enclosing it.
But sun-dried bricks are a material easily handled and
replaced. So when my choice was made the owner without
much trouble knocked a hole in the wall and thus established
easy communication between the
'
Bostan
'
and his courtyard,
where my servants were quartered. The yellow leaves lay thick
under the walnut and other fruit-trees, a sad memento of rapidly
advancing autumn.
ENTRANCE TO THE YAMEN, YARKAND.
CHAPTER X.
YARKAND AND KARGHALIK.
A MARCH of about eighteen miles brought me on the 17th of
September from Kokrobat into Yarkand. The scenery had
undergone a welcome change compared with that of the previous
day, for along the whole route there was no piece of barren
ground to be seen. Sandy as the soil almost everywhere is,
ample water is brought to it by canals large and small. Long
avenues of poplars and willows give shade along the greater
part of the road. Work with the plane-table was not easy in
this terrain coupe. But when more open ground was reached
about half-way, near Toguchak village, and the direction of
Yarkand could be clearly made out, we had the satisfaction
to note that the distance and bearing of this previously fixed
PALATIAL QUARTERS. 151
position coincided exactly with that derived from our
survey.
For several miles before and after Toguchak the road
traverses country that has only within the last few years been
brought under cultivation by means of newly opened canals
from the Yarkand River. It was a pleasure to see a sandy
waste thus reclaimed by dint of skilful labour. The crests of
the low sand-dunes still retain their original scrub-covered
surface, but everywhere around them spread carefully terraced
fields, which were said to have already yielded this year a
bountiful crop of wheat. The arrangement of the canals
along the road, often crossing each other at different levels,
indicated a systematic scheme of irrigation. The result is
creditable to the enterprise of Liu- Darin, then Amban of
Yarkand, who seems to have carried through a piece of truly
productive work with remarkable energy. The labour em-
ployed, it is true, is said to have been wholly unpaid, i.e.,
'
Begar.' Yet are not all great engineering feats of the East
due to this agency ? The cultivators to whom I talked
acknowledged that they had been forced to the work. But
now they were glad to occupy the ground they had reclaimed
for cultivation, and thus to reap the direct benefit of their
labours. The saying
of the Kashmir cultivators,
"
We do
not want money, we want the slipper," i.e., compulsion, for
any work of general utility, evidently holds true also in
Turkestan.
When I had crossed a broad canal from the Yarkand River
known as Opa, about three miles from the city. I found the
whole colony of Indian traders, with Munshi Bunyad Ali, the
"
Newswriter
"
employed
by Mr. Macartney, at their head,
waiting
to give me a formal reception. Most of the traders
from
the Punjab had akeady left for Ladak, and the fresh
contingent
of the
year had not yet arrived from across the
mountains.
All the same it was quite an imposing cavalcade,
at the head of which I rode into Yarkand. There were hardy
Khattris from
various parts of the Punjab, whom I felt tempted
to greet as
quasi-countrymen
;
men from
Jammu territory,
equally
familiar to me
;
and a sprinkling of Muhammadan
152 YARKAND AND KAROHALIK.
Kashmiris, of whom there is quite a settled colony here. They
were all in their best dresses, decently mounted, and unmis-
takably pleased to greet a
'
Sahib.' So it was only natural
that they wished to make some show of him. Accordingly
I was escorted in great style through the whole of the Yangi-
Shahr, or
"
New City," and the Bazars that connect it with
the old one. Our clattering cavalcade was undoubtedly a
little event for the people that thronged the Bazars. These
all seemed broad and fairly clean
;
in point of picturesqueness
far more attractive than those of Kashgar.
Then we turned off to the right and rode round the crenellated
walls of the
"
Old City
"
into an area of suburban gardens.
Here lies the Chini-Bagh which Mr. Macartney had in advance
engaged for my residence. It proved quite a summer-palace
within a large walled-in garden. Passing through a series
of courts, I was surprised to find a great hall of imposing
dimensions, with rows of high wooden pillars supporting its
roof. Beyond it I entered a series of raised apartments, once
the reception-rooms of Niaz Hakim Beg, the original owner
of these palatial quarters. There was no mistaking the marks
of departed glory. The gilding of the latticework screens
separating the rooms had faded, and other signs of neglect
were numerous. Yet good carpets covered the floors and the
raised platforms ; tasteful dados ran along the walls, and
over the whole lay an air of solemn dignity and ease. When
alone in my temporary mansion I felt the reality of the charms
which such an abode offers even more than I had in the old
Moghul and Sikh garden-residences, once my favourite haunts
in the Campagna of Lahore.
The days which followed my arrival at Yarkand passed with
surprising rapidity. I had intended from the first a stay
of five or six days, in order to make use of the opportunities
which Yarkand offers for the collection of anthropological
materials and old art ware
;
but several circumstances helped
in extending it. I had been assured in Kashgar that Yarkand
was the j)lace where I could most conveniently arrange for
the money needed on my further journey. There I was to
find the Indian traders eager to take Government Supply Bills
I
COSMOPOLITAN VISITORS. 153
and my cheques and convert them into cash. Unfortunately,
those who needed drafts on Indian Treasuries had already
started on their way to Ladak, and the remaining Khattris
had taken the opportunity to remit with them whatever ready
money they had cleared by the sale of their goods. So it was
no easy matter to find a market for my drafts, and seeing
that only a portion of the money I needed could be raised
at a reasonable rate, I was ultimately obliged to despatch a
messenger to Kashgar. Until my messenger had returned
with the desired cash in silver and gold a start appeared scarcely
desirable.
Another discovery made soon after my arrival, and equally
annoying, was that two of the camels and two of the ponies
had developed sore backs, which needed cure. The fact had
before been carefully screened from my knowledge, with the
natural result that the evil had got worse than it need have.
So nothing less than a week's rest would do to make the trans-
port fit again, and accordingly the camels, after careful ex-
amination and dressing of sores, were sent to have an easy
time grazing in a wooded tract, a day's march southwards.
The experience was not thrown away on me. Thereafter
inspections of the animals were held almost daily, and those
responsible for their loading learned to understand that the
hire of transport in place of animals rendered temporarily
unfit would be recovered from their own pay.
It was lucky that my Yarkand quarters were of such delightful
spaciousness, for from the first day of my stay there was no
want of visitors. Yarkand is the great commercial centre
of Chinese Turkestan, and owing to its geographical position
at the point where the routes to India, Afghanistan, and to
the North meet, has something of a cosmopolitan air about
it. The colonies of Kashmiris, Gilgitis, Badakhshanis, and
people from other parts of the Indian frontier regions are large,
and each of them has members of some position anxious to
show attention to a
'
Sahib.' So I had to hold regular
receptions with the assistance of old Munshi Bunyad Ali, and
"
the carpet of my presence
"
was rarely clear of more or less
picturesque
visitors. If their statements as to the strength
154 YARKAND AND KARGHALIK.
of their respective com-
munities can be trusted,
Yarkand must have a
strangely mixed popula-
tion. Immigrants from
Wakhan, Shighnan, Ba-
dakhshan, and the other
Iranian tracts westwards
abound
;
Kashmir and
Ladak are strongly repre-
sented
;
even little Bal-
tistan has sent its colony
from beyond the ice moun-
tains. So there was no
want of materials for
anthropometric work, and
I did all I could to benefit
by it.
The types of my visitors
made me feel far nearer
again to India and its
borderlands. Hearing
Kashmiri, Panjabi, and
Pushtu spoken at all hours
of the day in a place so
closely resembling the na-
tive country residences of
Northern India I might
have felt myself on In-
dian soil. That European
fashions have not yet in-
vaded this corner of Asia
helped to throw one back
in regard to time too.
Turki is, of course, the
language of general intercourse among the different colonies of
immigrants, and as the latter scarcely ever bring wives from
their own homes but marry in the country, it is natural that by
1. \l) AKllSHANI TKAUKU, YAKKAND.
OLD TURKESTAN ART-WARE.
155
the second or third generation the knowledge of their father's
tongue is akeady lost. But physical features are not so easily
effaced, and a stroll in the Bazars is enough to convince the
observ^er how large an infusion of foreign, particularly Iranian,
blood there is in the Yarkand population.
Apart from my visitors, there was for a great part of the
day another crowd to fill the outer hall of my temporary
palace. Munshi Bunyad Ali had taken care to let it be known
that I wished to acquire things of old local art, and in consequence
improvised agents of such articles were pouring in every morning.
I was specially looking out for specimens of that ornamented
brasswork which had once its home at Khotan and which has
become well known to lovers of Central-Asian art, though
its local connection does not appear to have been realised.
Judging from the quantities of beautifully worked
'
Aptabas,'
'
Chaugans
'
(tea-pots),
'
Chilapchis ' (water-basins), jugs, and
other metal articles brought to me from houses of once well-
to-do families, the supply must still be considerable.
Among the pieces offered for sale I was able to pick out
some excellent specimens of open metal work, all showing
most clearly the influence of Persian floral design, yet with a
distinct individuality of treatment. By the side of these
indigenous art products, queer odds and ends of Chinese work
turned up

pottery of evident merit and age, and beautiful

pieces of embroidery from the Far East. The rich Begs of
old days must have looked truly gorgeous in their dresses of
state, heavy red or blue silk Chogas emblazoned with artistic
designs of pure Chinese style. Fine carpets of Turkoman
and Khorasan make indicated that importations from the West
had been equally frequent. Thus Yarkand shows also in this
respect that it has been a point where, since early Muhammadan
times, Chinese influences have mingled with the culture of
Iran and Turkestan. It is Khotan, however, which seems to
have been the place of origin and the true home for most of
the indigenous industries. Curiously enough, almost all the
people who brought me these deUghtfully varied specimens of
old art-ware were Kashmiris. The pedlars' instinct, which
is so conspicuous in the urban population of the valley, has
156 YARKAND AND KARGHALIK.
evidently not been extinguished by emigration across the
mountains. So I often felt as if I were besieged again in the
camping-grounds of Srinagar by the voluble and irrepressible
agents of Kashmirian craftsmen.
Liu-Darin
('
Darin ' represents the local pronunciation of
the Chinese title
'
Ta-jen
'),
the Amban of Yarkand, was
absent on tour when I arrived. But he soon returned, and
after the due preliminaries had been arranged, I made my
call at his Yamen. I found Liu-Darin a very amiable and
intelligent old man. Conversation through a not over-intelligent
interpreter is not the way to arrive at a true estimate of character.
But somehow Liu-Darin's manners and looks impressed me
very favourably. On the next day I received the return visit
of the old administrator, and found occasion to show him the
Si-yu-ki of Hiuen-Tsiang and to explain what my objects
were in searching for the sacred sites which the great pilgrim
had visited about Khotan, and for the remains of the old settle-
ments overwhelmed by the desert. It was again reassuring
for me to find how popular the figure of the pious old traveller
still is with educated Chinamen. Though Tang-Seng,
"
the
monk of the Tang Dynasty," is evidently credited with many
wonderful relations for which we should look in vain in his
"
Description of the Western Countries," this scarcely need
disturb our conscience.
On the 22nd of September Liu-Darin insisted on entertaining
me at a Chinese dinner. Well-meant as the invitation no doubt
was, I confess that I faced the entertainment with mixed feelings.
My Kashgar experiences had shown me the ordeal which such
a feast represents to the average European. However, things
passed better than I had ventured to hope. The dinner consisted
of only sixteen courses, and was duly absorbed within three
hours. It would be unfair to discuss the strange mixture of
the menu, especially as I felt quite incompetent to analyse
most of the dishes, or the arrangements of the table. Having
regard to my deficient training in the use of eating-sticks I
was provided with a fork (never changed or cleaned) and a
little bowl to eat from. As my host insisted on treating me
personally to choice bits, a queer collection accumulated
on
INTERVIEWS WITH LIU-DARIN. 157
this substitute for a plate. I felt more comfortable when
I managed to get it cleared from time to time. For the hot
spirit, a kind of arrack it seemed, served in tiny square cups
as the only beverage, there was no such convenient depository,
and in reply to the challenges of my conWves I had to touch
it more frequently than I could have wished. Besides my
host, two of his chief ofl&cials, jovial-looking men, were keeping
me company.
It was a little pathetic when, in the course of dinner,
Liu-Darin pressingly inquired as to what news I had about
the capture of Peking by the allied forces and the flight of the
Emperor. I had no direct news from Europe later than the
end of
July,
and thus could not satisfy his curiosity. So I
contented myself with describing the relief with which the
safety of the legations had been greeted in Europe. The old
"
Political " would not credit my ignorance, and attributed my
reticence to the wish to keep back unpleasant information.
Whatever the reliability of the news may have been that reached
Chinese Yamens in Turkestan through the wire to Kashgar,
it seemed clear that they reaUsed the great danger to the
Government they served. There may have been anxiety about
the future
;
but if my Kashgar friends' views were right, it
was the doubt about their o\nti individual fortimes, not those
of their nation, which secretl}' troubled the minds of the officials
in this land of exile.
It was arranged that after the dinner I should photograph
my host and some of his people. So Liu-Darin at the end of
the feast duly installed himself on a raised chair of office,
with his little daughter and son by his knees, and some
implements of western culture, in the shape of sundry clocks,
etc., on a small table close by. A crowd of more or less
ragged attendants formed the background. The photos were
easy to take, as my sitters kept as quiet as if they were
sculptured. Then we parted in all friendship. Liu-Darin
talked of retiring soon to his native province of Hu-nan. May
he return to it in peace and, as my Chinese patron saint,
'
Tang-
Seng,' did of old, enjoy the rest he was looking for.
The last days in Yarkand were busily spent in completing
158 YARKAND AND KARaHALIK.
the winter outfit of my men and in sorting and packing my
purchases. Accounts too had to be settled, and in this
respect I was glad to avail myself of the skilled assistance of
LU' DARIN, AMBAN UF VAKKAND.
Lala Gauri Mall, the Ak-sakal (headman, literally
"
white-
beard
")
of the Hindu traders. Apart from the question of
priceno small matter in a country where it would ajiparently
be against all business principles to ask less than double the
right amount even from local customersthere is enough
CURRENCY COMPLICATIONS. 159
trouble in the mere payment. The Chinese currenc}^ with
its
'
Sers ' or
*
Tels,'
'
Miskals
'
and
'
Fens,' arranged on a
plain decimal system, would be as convenient as can be desired.
But its simplicity is of little avail in this outlying province
of the empire, which stubbornly clings to its time-honoured
reckoning in
'
Tangas
'
and
'
Puis.' Each of the little square
Chinese coppers known in Turkestan as
'
Dachins
'
is reckoned
in Kashgar and Yarkand as equal to two Puis, and twenty-five
of them make up a Tanga. The Khotan Tanga is worth twice
as much as the Kashgar Tanga.
Coins representing this local unit of value there are none
;
so all sums have to be converted into Miskals, the smallest
available silver coins, at the ratio of eight Tangas to five
Miskals, unless one is prepared to handle the dirty rolls of
Chinese coppers which the local trader keeps strung up like
sausages. But the exchange rate between silver and copper
is not stable, and the silver Miskal was just then considerably
above the value of forty copper pieces which the ratio just
mentioned would indicate. So after successfully converting
Tangas into the legal coin, a varying discount has to be
calculated before payment can be effected. It only adds to
these monetary complications that prices of articles imported
from Russia are reckoned in
'
Soms
'
(Roubles), which in the
form of gold pieces of five or ten Roubles widely circulate
through the markets of Turkestan, while the heavier Chinese
silver
'
Yambus,' of horse-shoe shape and varying weight,
have a discount of their own. During my stay in the cormtry
the value of the gold Rouble as against the local currency of
Tangas represented by Chinese silver and copper pieces,
steadily declined, and with it unluckily fell the Rupee too,
the exchange value of which seems in Turkestan to depend
mainly on the Rouble rate. For the well-trained arithmetica
faculties of the Hindu trader these tangled relations of^er no
difficulty. But I confess I sadly reflected on the loss of time
which they implied for me.
On the 24th of September the weather became cloudy and
the temperature distinctly cold. A yellow haze hung all day
low over the ground and intensified the effect of the atmospheric
160 YARKAND AND KARGHALIK.
change. It felt autumnally chilly in the wide halls of my
palace, and I realised how different life in them would be when
the winter set in. The haze still continued when early on
the morning of the 27th of September my caravan was again
set in march. It was market-day, and the endless stream
of villagers that passed along the roads with their manifold
produce and belongings was a welcome distraction. Women
of the cultivating class play a prominent part in all the marketing.
I met them in large groups or accompanied by their men and
children, but in either case almost invariably mounted. The
large caps with peaks of velvet which the women wear looked
more comfortable on this chilly morning than when I had first
seen them in the heat of early August.
There was little else to occupy my attention after this stream
of market visitors ceased with the advance of the morning.
The broad and fairly straight road, lined with poplars and
mulberries, runs through flat, fertile country. In the fields
of maize the harvest was proceeding
;
for the rice, apparently,
the return of warmer days was expected. About five miles
from the city I passed the large Bazar of Manglik, a long row
of clean mud huts with booths opening on the road, but almost
completely deserted, as it was not the local market day. Such
Bazars are met at varying distances all along the route to
Karghahk, a sure indication of the thick population of the
fertile belt of land through which it passes. After a ride of
about eight miles we reached the bank of the Zarafshan or
Yarkand River. Fed by the streams which drain the whole
mountain region between Muztagh-Ata and the Karakorum
range, it must carry a mighty mass of water in the height of
the summer. Even now it flowed in three arms which had
to be crossed by boat. The clumsily built ferry boats could
not take the laden animals, so the loads had to be unpacked
and refitted again and again. Each of the branches was about
forty yards broad, and the depth well up to a camel's girth.
It took my caravan three hours to effect the passage, and all
through that time the traffic of laden ponies and donkeys was
suificient to fill the two or three ferry boats at each crossing
as quick as they could be worked. On the opposite bank of
CROSSING OF YARKAND RIVER. 161
this river-bed, which has the total breadth of a mile, we passed
the large Bazar of Painap, and by five o'clock I reached Posgam
Bazar, the end of the day's march. Inside the large inner
quadrangle of the Sarai I found my tent pitched. In the rooms
of the spacious rest-house, which, dating from the time of
Yaqub Beg, is kept clean and in a very fair state of repair, I
should have been -warmer
;
but I preferred to stick to my little
tent and its fresh air as long as possible.
My march on September 28th to Karghalik was a fairly
long one, about twenty-four miles, but very enjoyable. A
light storm overnight, though accompanied only by dust
without a drop of rain, had thoroughly cleared the atmosphere.
It was pleasant to walk in the fresh morning air between the
carefully cultivated fields and orchards that cover the ground
south of Posgam. Irrigation from the Yarkand River provides
plenty of water, and the comparative pro.ximity of the villages
and Bazars along the route testifies to the prosperity of the
tract. About nine miles from Posgam there followed a grassy
plain known as Tiigiilaz, which is intersected by numerous
clear streamlets said to be fed by springs further west. The
sight of their limpid water, so different from the red, grey,
or brown colouring of the larger streams seen since Kashgar,
was in welcome harmony with the view of the distant snowy
ranges that now showed themselves in the south-west. The
mountains which I could see for a great portion of the march
belong to the ranges through which the Zarafshan forces its
way down from Sarikol. All through my stay at Yarkand
the haze had hidden them from view.
Beyond the Tiigiilaz plain we came to the Tiznaf River,
now reduced to a number of narrow channels, but evidently
fed with plenty of water when the snow melts in the advanced
ranges of the Kuen-luen. The well-constructed bridge which
leads over the main river-bed was built, according to the
Chinese and Turki inscription at its head, some twenty-five
years ago, and measures fully
250
steps. Beyond followed
a rich tract with smiling fields of lucerne, Indian corn, and
cotton, dotted with comfortable-looking villages. At Charvak,
the
"
Tuesday Bazar " of this neighbourhood, I found an
11
162 YARKAND AND KAROHALIK.
animated scene. The Amban of Karghalik was expected to
pass through on his way to Yarkand, where he was proceeding
to welcome Liu-Darin's successor. He had postponed his
journeyas I was told, on account of my approaching visit

but the preparations for his reception were complete. Broad

strips of scarlet cloth were stretched across from house to
house under the matted awnings that cover the whole long
Bazar street ; the latter itself was thronged with a crowd
apparently
making holiday. The local Beg received me in
his official Chinese garb, and politely invited me to the large
shop that had been fitted up with carpets and felts as a kind
of
reception-room. So I had to partake of tea that was
welcome
enough after the dusty ride, and of a fine collation
of fruit.
By
half-past four I had approached Karghalik through a
belt of villages rich in orchards and shrines of all kinds. The
pebble-strewn bed of a half-dry stream, which I passed shortly
before
entering the town, betokened the vicinity of the hills.
I soon passed into the tangled net of Bazars that form the
centre of
Karghalik town, and was struck with their com-
parative
cleanliness and the thriving look of the whole place.
It is clear at the first glance that Karghalik derives no small
amount of profit from its position at the point where a much-
frequented route to the Karakorum Passes joins the great
road
connecting Khotan with Yarkand. After a long search
among the suburban gardens to the south I found a large plot
of meadow land with some beautiful old walnut-trees that
carried me back in recollection to many a pretty village in
Kashmir.
It was a delightful camping-ground for myself,
and, as my people found quarters in a cottage close by and the
ponies
excellent grazing, everybody was satisfied.
On the morning of the 29th some Begs sent by tlie Amban
brought a present consisting of a sheep and fodder for my
animals.
I returned the attention with a collection of Russian
sweets,
sardine tins, and highly scented soap of German make,
bought
for such ])urposes at Kashgar. About noon I went
to pay my visit to Chang-Darin, the Amban, at his official
residence or Yamen. I was received with all the ceremony
HALT AT KARGHALIK. 163
due to the occasion, a salute of three popguns included, and
soon found myself face to face with my host in his neat little
reception-room. Chang-Darin impressed me very favourably
with his liveliness and unmistakable intelligence. He had
heard from the Tao-tai at Kashgar of my visit and its object.
^o no lengthy explanations were required as to what I was
YETIMLUKUM MaZAR WITH CEMETERY, NEAR KARGHALIK.
looking for along the route to Khotan, and all needful help
about transport and supplies was readily offered.
Chang-Darin insisted on treating me to a light kind of lunch
consisting of only a few plates. In addition to these I had to
taste a good deal of wnne. Fortunately it was not the strong
Chinese spirit that appeared on Liu-Darin's table, but a kind
of
Madeira, possibly from the Caucasus or the Crimean vine-
yards.
Two little wine-glasses and proper forks by the side
11*
164 YAKKAND AND KARGHALIK.
of the orthodox eating-sticks also betokened the progressive
attitude of my host in matters cuHnary.
We parted in mutual good feeling, and I utilized the occasion
to pay a cursory visit to the Bazars. There were plenty of
shops open, though it was not market day, and I was soon
able to make the needed purchases in felt materials required
for my men's winter outfit. The hills about Kokyar are re-
nowned for the
'
Paipaks
'
or felt socks there produced, and
Karghalik is the great market for them. All the Bazars are
covered with substantial matting which gives shade for the
summer and keeps off the glare. Canals neatly bridged over
cross the Bazars at numerous points, and the luxuriant trees
that grow by their side give a welcome change of colour. Open
kitchens or eating-houses were to be met at numerous spots,
and, as a rush of customers was not great on this day, I could
conveniently inspect their arrangements. They resemble far
more those to be found in similar establishments of the West
than anything that could be seen in an Indian town. There
were pots and cauldrons kept boiling in regular stoves, plates
with breads and cakes, dishes of vegetables, etc. One of these
street restaurants struck me particularly by the elaborate floral
designs on its whitewashed front walls.
When I returned to my camp much pleased with the stroll
I found the Amban already waiting to return my call. He
had whiled away the time by a careful inspection of my camp
furniture, which evidently met with his approval, as he sent
next day a carpenter to take measurements of the neat folding
table Messrs. Luscombe & Co., of Allahabad, had made for
me. We talked a good deal about old Hiuen-Tsiang and his
account of the country. I showed Chang-Darin the Chinese
glossaries attached to Julien's translation of the Si-yu-ki, and
the plates of Dr. Hoernle's publications on the antiquities
of Khotan and Kucha. The ancient Chinese coins and the
few fragments of Chinese manuscripts there depicted excited
a
good deal of curiosity on the part of my visitor. I felt more
than ever the disadvantage of my ignorance of Chinese, for
it was no easy task to give intelligible answers to the many
queries of my visitor through an interpreter so little versed
VJSIT OF KARGHALIK AMBAX. 165
in literary matters as Niaz Akhun. He had, however, been
to
'
Bajin
'
(Peking), and this supreme achievement gave
him an air of assurance which made him, if not others also,
forget the limitations of his intelligence.
On the morning of the next day there arrived the consign-
ment of money, sent by Mr. Macartney from Kashgar in pay-
ment for my drafts on Lahore. My halt at Karghahk had
been made partly in expectation of it. With the bags of
Chinese silver coin and the smaller packet of newly-coined
gold Rouble pieces, Mr. Macartney's
'
Chaprassi ' brought
home letters also. He was to return the next day and carry
my own mail to Kashgar. So I was kept busy all day with
letters and with accounts that were to be despatched to
Government. I sometimes wondered how the Babus of the
Calcutta office would take to the currency complications
reflected in my
"
Monthly Cash Accounts." The shady grove
of walnut-trees in front of my tent made a delightful Daftar.
In the evening I strolled up the bank of the stream that flows
to the west of the town and got a distant glimpse of the hills
towards Kokyar. They produce a great quantity of wool,
and by the side of the stream I came upon a place where an
enterprising Kashgar trader, who exports to Andijan, has an
establishment for washing and cleaning the wool. The people
I met there accounted by these exports for the reduced output
of felts in the neighbourhood.
The ist of October was the Karghalik market, and I had
decided to wait for it, in the hope that it might bring to the
town specimens of the curious hill-people known as Phakhpo,
that live in the valley south of Kokyar. Anthropological
data regarding them would be most welcome, in view of the
doubts that exist as to their ethnic affinity. From the de-
scription recorded by members of the Yarkand Mission it appears
that the Phakhpo in build and featiu-es are eminently
"
Aryan
"
looking. Yet they are distinct from the Tajiks of Sarikol,
and are believed to speak a Turki dialect. Their small numbers,
distant home, and shy habits have so far prevented any close
observation. I was no luckier in this respect, as, notwith-
standing the watch kept by the Amban's order, no Phakhpos
166 YARKANP
AND KARGHALIK.
could be discovered among the market crowds. To wait for
the arrival of men who might have been fetched from their
mountains would have meant the delay of a week, which I could
not afford.
Instead of these hill-folk the day brought another interesting
acquaintance, a travelUng
Buddhist monk from the East of
China who had begged his way
through to Aksu and Khotan and
was now again on his way north-
ward. He had somehow heard of
the respect I paid to
'
Tang-Seng's
'
memory, and not unnaturally hoped
for some help on his onward
j
ourney.
His was evidently not a pilgrimage
in search of sacred sites connected
with Buddhism. Yet his simple,
jovial way appealed to me, and I
was glad to return his gift of a
religious tract, nicely painted on
red paper, with an offering of silver
that sent the humble devotee away
quite happy and contented.
I shall always look back with
pleasure to the short stay at Kar-
ghalik, or rather Yetimlukum, as
the village is called where my camp
stood. There was nothing to remind
me of the neighbourhood of the desert or the equal barrenness
of the outer hills. As far as the eye could reach over the large
plots of fields and gardens fertility and jjlenty reigned. Much
reminded me of Kashmirthe variety and luxuriant growth
of the trees, the numerous picturesque Ziarats with shad\"
groves near my camp, and in the Bazars the quaintly carved
wooden houses.
BUDDHIST MONK FROM CHINA.
167
CHAPTER XI;
ON THE ROAD TO KHOTAX.
On the morning of the 2nd of October my caravan wended
its way through the busy little town towards the East gate,
from whence the road to Khotan starts. On my way I visited
a large Madrasah called after the Ghujak Masjid, opening on
one of the principal Bazars. It comprises a large quadrangle
with rows of vaulted rooms for about
150 students, and at its
west end an open hall of imposing dimensions. The wooden
pillars supporting the roof as well as the roof itself are painted
in lively colours, chiefly shades of red, suggesting the poly-
chrome splendour of some classical building.
Chang-Darin had sent his principal
'
Tungchi
'
or interpreter
to accompany me on my start and bring me his farewell good
wishes. Within a mile of the Khotan gate where I parted
from the good-looking old man the caravan road emerges on
barren desert. A few miles further on this gave way to narrow
strips of cultivation forming the little oasis of Besharik, but
this was soon traversed, and beyond there received us an un-
mitigated wilderness of gravelly Dasht. The road is marked
all along by wooden posts erected at short intervalsno useless
precaution considering how easy it would be for the traveller
to lose his way at night or in a sand-storm. At Kosh Langar,
where the day's march ended, I was surprised to find in the
midst of the barren waste a commodious Sarai built of hard-
burned bricks, with vaulted rooms and ample out-houses.
This building attests the enterprise of Niaz Hakim Beg, governor
of Khotan in the days of Yaqub Begthe same whose country
168 ON THE EOAD TO KHOTAN.
mansion gave me shelter at Yarkand. A tank is provided
into which water flows for one day in the week by a small canal
brought down from the outer hills. A ruined mound of sun-
dried bricks, some
30
feet high, which rises from the desert
plain about a mile and a half northwards, may possibly mark
the remains of a Stupa. But its decay was too far advanced
to permit any certain conclusion.
On the 3rd of October my march lay entirely through
desert ground. To the south the line of outer hills was
faintly visible through the haze, but no canal or watercourse
of any kind descends from them to the plain. After the chilly
nights the heat and glare of the midday hours were very per-
ceptible. The dreary route we were now following along the
southern edge of the great sandy desert, the Taklamakan,
had for me a special historical interest. It was undoubtedly
the ancient line that led from the Oxus region to Khotan and
China. Walking and riding along the track marked here and
there by the parched carcases and bleached bones of animals
that had died on it, I thought of travellers in times gone by
who must have marched through this same waterless, unin-
habited waste. Hiuen-Tsiang, who travelled here on his way
back to China, has well described the route. After him it
had seen Marco Polo and many a less-known mediaeval traveller
to distant Cathay. Practically nothing has changed here
in respect of the methods and means of travel, and thus my
thoughts could wander back into the past the more readily.
It is certain that, with the caravans that once trod this sand,
the Buddhist religion and the elements of Indian as well as
of classical culture and art travelled to the land of the Sins.
Shall we ever learn how much they brought back that has
influenced the civilization of the ancient world ?
Cholak Langar, which I reached in the early afternoon, has
a Sarai closely resembling that of the preceding station. It
stands on the edge of a broad, sandy ravine, that descends
from a gap in the low hill-range south. A large andwell-filled
tank close below the Sarai was the only redeeming feature in
the bleak landscape. The course of the small watercourse
that feeds it at times is marked far away northward by a line
IN GUMA OASIS. 169
of low shnibs. In the evening the Chinese clerk in charge of
the post station came to call on me. He seemed a quiet, well-
educated man, not over-pleased with his desert surroimdings.
He has nine Dak horses and four postmen under him to carry
letters to Guma and Karghahk. The
"
mails " along the
postal routes of the country are ordinarily restricted to the
ofi&cial correspondence of the Chinese authorities. My visitor
told me that he had come two years ago from Urumchi with
the Karghalik Amban, whose district in the Ho-nan was also
his own. That his \nllage was near to the birth-place of
'
Tang-
Seng,' i.e., Hiuen-Tsiang, was a piece of information, not
indifferent to me. So I treated this modest exile to tea and
cake, and tried to cheer him with the hope of an early transfer
to a more congenial post.
The march from Cholak Langar to Guma was reckoned
a long one, nine
'
Potais.' So the camels marched off with
the heavy baggage by daybreak. My tent and the kitchen
things going by ponies could start later. While they were
being packed I strolled into the courtyard of the Sarai, the
vails of which I found plastered over vvith official edicts in
hinese and Turki. Conspicuous among them was a long
trilingual proclamation, in Chinese, Mongolian, and Turki,
printed on large sheets of yellow paper. It was an edict in
the name of the Emperor, referring to the trade intercourse
with the Russian markets, and had been issued some two
months before.
Niaz Akhun, my interpreter, was eager to
know
whether it was likely to be the last edict of the Emperor,
whose flight from Peking was being reported all through the
Bazars.
After a ten miles' march over bare gravel and sand Siligh
Langar was reached, a collection of wTetched mud-hovels,
with a Uttle tank fed by a small watercourse. The tank was
full and the water flowed away into the sand. Beyond Siligh
Langar
scanty scrub and
'
Kumush ' appeared again and
covered the sandy soil up to Hajib Langar, another uninviting
wayside station two and a half miles beyond. Then the ground
began
to show pebble-strewn beds of shallow ravines, and in
a long dark line the trees of the oasis of Guma appeared on
170 ON THE EOAD TO KHOTAN.
the horizon. It was a dreary ride of some six miles before
we struck the river-bed, then dry, that marks the western edge
of the lands of Guma. Beyond it I passed scattered fields
and groves half-buried under drifting sand that seems to
advance from the west, and at last, after riding up a steep
bank, some
30
to
40
feet high, I was once more amidst fertile
gardens and fields. Close to a large canal that skirts the Bazar
of Guma I discovered a camping-ground just as I wanted it,
in a quiet garden enclosed by a hedge of high willows and
poplars. While my tent was being pitched I rode off again
towards the Bazar, where the weekly market was still in full
swing. The large crowds buying and selling cattle, fruit,
cotton stuffs, and other local produce were an indication of
the extent of the oasis. Over rows of stalls high boots of red
leather were hanging, an article evidently in great demand
owing to the approach of winter.
The 5th of October was given up to a halt needed for
antiquarian inquiries. Among the purchases of Central-Asian
antiquities made for the Indian Government by Mr. Macartney
and other political officers, paper manuscripts and
"
block-
prints
"
in
"
unknown characters " had since
1895
become
more and more frequent. These and similar acquisitions
which had reached Russian and other public collections in
Europe, were all supposed to have been unearthed from sand-
buried sites in the Khotan region. Islam Akhun, the Khotan
"
treasure-seeker " from whom most of these strange texts
were acquired, had in statements recorded at Kashgar by
Mr. Macartney and subsequently reproduced in Dr. Hoernle's
learned report on the Calcutta collection, specified a series of
localities from which his finds were alleged to have been
obtained. Most of these were described as old sites in the
desert north of the caravan route between Guma and Khotan.
Information that had reached me at Kashgar helped to
emphasise the doubts which had previously arisen as to the
genuineness of his
"
finds." But it was at Guma that I first
touched the ground where it was i)ossible to test the
"
treasure-
seeker's
"
statements by direct local inquiries.
When the local Begs together with the several Yuz-bashis
KARAKUL MAZAR.
171
of the main villages joined me in the morning, I ascertained
that there was an extensive debris-covered area known to all
as a
'
Kone-shahr,' close to the road between Guma and Moji,
the next oasis eastwards. But nobody
had ever heard of the
discovery of
"
old books
"
either at this or any other site.' Of
the string of localities named as find-places in the detailed
itinerary which Islam Akhun had given of one of his desert
journeys, only two were known to them. As both lay close
to the oasis it was easy to arrange for their inspection. Riding
to the north-east with a lively follov^ing of Begs and their
attendants,
I soon reached the area of moving sand-dunes
20 to
30
feet high which encircles Guma from the north. Near
to the little hamlet of Hasa, passed en route on the edge of
this area, the dunes had within the memory of the villagers
encroached considerably on the original holdings. A portion
of the scattered homesteads was believed also to lie buried
under the advancing sands. But as they had been abandoned
only within a comparatively recent period, they could never,
as my guides sensibly pointed out, have furnished antiquities.
A ride of a little over three miles sufficed to bring me to Kara-
kul Mazar
("
the Mazar of the Black Lake
")
which figured
prominently in Islam Akhun's itinerary. By the side of a
little lake of saline water, hcdf-covered by reeds, there rises
a
semicircle of sandhills. On the top of one, the customary
erection of poles hung with votive rags, yaks' tails, and skins,
indicates the supposed resting place of a saint. Of his life
and deeds I could gather nothing except that the holy man
came
to live here when his beard was black, and died here
as an
'
Ak-sakal '
("
a white-beard ").
Of the vast cemetery
round this shrine where Islam Akhun alleged that he had made
finds of ancient block-prints, I could discover no sign.
The lake is fed by a small riviilet, which flows in a broad,
tortuous bed about a quarter of a mile eastwards. It rises
from a series of springs and pools about Hasa, and accordingly
is known by the name of Kara-su
("
black-water ")
;
but
during the spring and early summer it is swelled by flood water
(*
ak-su ' or
"
white water
")
when the snow melts on the
mountains southward. In its bed, which we followed for about
172 ON THE EOAD TO KHOTAN.
three miles to the oasis of Karatagh-aghzi, or Karataghiz,
I came for the first time upon the jungle that thrives along
the watercourses that penetrate into the desert. Reeds of
various kinds, the hardy
'
Yulghun ' plant with its heather-
like small red flowers, and other shrubs filled the dry bed of
the Kara-su in picturesque confusion. The autumn had already
turned the leaves of many to various tints of yellow. So
there was a feast for the eyes, doubly welcome after the dreary
monochrome view of the Dasht. At Karatagh-aghzi I found
luxuriant groves of poplar-, mulberry-, and other trees scattered
among ripe fields of Indian corn. The other produce had
already been harvested. The part of the cultivated land
which I saw was said to have been reclaimed only some fifteen
years ago. The size and luxuriance of the trees that had grown
up
in this short time was a striking
illustration of the capability of the
desert soil if once reached by water.
From Karatagh-aghzi Islam Akhun
alleged that he had visited various
ruined sites which yielded him
"
old
books
"
and other strange finds. But
the inhabitants, when closely questioned,
knew nothing of such sites and still less
of such discoveries. So assured of the
negative result of my inquiry I turned
I^mmnYttJ^
back to Guma. We took the track
* ^^flWllS' H
across the sand to Towen-Bazar, one of
w^
0^^'-'
^^^
more northerly villages which
^'
fliF
a
merges imperceptibly into Guma Bazar.
^X;2..
"
*
It was pleasant to ride in the shady
village lanes, with a peep again and
MKNP.CANT.OK-UIWANA.-
^^^.^
-^^^
j^^^^^j^
j-^^j^
j^^j^ gardCHS.
The profuse growth of melons and
cucumbers was a characteristic feature of all ; I passed several
open-air paper factories, the pulp, prepared from the bark of
the mulberry-tree, drying on little sieve-like screens.
I also met a troop of fantastically clad
'
Diwanas,' or
beggars, bent apparently on collecting in alms their share
STUPA OF TOPA TIM. 173
of the villagers' harvest. The lanes of the main Bazar through
which I returned to camp looked singularly empty after the
busy life vsitnessed on the preceding market-day.
When I left my cheerful Guma camp on the morning of the
6th of October the sky was of radiant clearness, with scarcely
a trace of haze. So when I emerged from the shady lanes
of the southern part of the Guma oasis on the open Dasht I
was not surprised to find parts of the great snowy range distinctly
visible. The snows I saw glittering far away over the dark
lines of the outer mountains evidently belonged to the main
range about the Karakorum Passes. Distances seemed to
shrink strangely when I thought that behind those stupendous
mountain ramparts lay valleys draining to the Indus. Mist
and clouds hung over other parts of the range, and, as the
sun rose higher, drew a veil also over the ice-covered ridges
first sighted. A couple of miles further, after crossing a broad
but now entirely dry river-bed which, lower down, receives
the water of the Kara-su, I came upon the first of the old sites
which earlier reports had led me to expect on the march from
Guma to Moji. All along the right bank of the ravine the
ground was thickly strewn with fragments of coarse red pottery.
No ornamented pieces could be found, but the exceptional
hardness and glaze of these potsherds showed that they
belonged to a period far removed from the present. The
extent of the area covered by these scattered fragments
plainly indicated the site of a large and thickly inhabited settle-
ment. But no other trace now remained of its existence.
The innumerable potsherds invariably rested on the bare surface
of loess, with never a trace of walls or more substantial remains
below.
When to the east of this old site I had crossed the narrow
belt of irrigated ground occupied by the hamlets of Mokuila
and was passing once more over a barren scrub-covered Dasht,
I sighted to the north-east the mound of which my Guma
informants had spoken as Topa Tim.
'
Tim '
is the desig-
nation given to all ruined mounds about Kashgar, and as the
one now within reach looked through my glasses much like an
ancient Stupa, I made haste to reach it. It proved a longer
174 ON THE KOAD TO KHOTAN.
business than I anticipated. For our guide insisted on our
first following the road towards Moji and crossing the deep-cut
bed of a watercourse, now dry. I accepted his guidance much
against my instinct, with the tantalizing result that when we
had got abreast of the mound after a two miles' ride, a caiion-
like ravine absolutely cut us off from it. In vain we searched
for a place where the perpendicular banks of loess would admit
of a descent to the bottom of this fissure, "40
to
50
feet deep.
There was nothing for it but to ride back to the road and start
afresh on the other side.
The Stupa, when at last reached, was a sight that cheered
my archaeological heart. In size and proportion it closely
resembled the Mauri Tim Stupa near Khanui. Though its
exterior had suffered more decay, and an excavation on the
top showed that it had not escaped the ravages of the
"
treasure-seeker," it still rose to a height of nearly
29
feet.
Immediately around the Stupa I found the ground strewn
with broken bits of ancient pottery, exactly as seen at Mokuila
and on the great site subsequently traversed. So the conclusion
seems justified that the habitations which these scanty remains
indicate belonged, like the Stupa, to the Buddhist period.
On the first attempt to reach Topa Tim on the other side
of the ravine I had passed a debris-strewn area far more
extensive than those seen before. Stretching to the north
of the caravan route it seemed to cover fully three square
miles. My guides called it the
'
Tati ' of Kakshal. The
relics of ancient habitations that lay scattered here in patches
of varying extent and thickness comprised, besides pottery
fragments of all sorts : pieces of burned brick, slag, broken
bits of bone and metal, and similar hard refuse. The conditions
in which these remains presented themselves appeared at first
very puzzling. But the examination of similar sites which I
subsequently traced at many points beyond the limits of the
present
cultivated area in the Khotan region, and which are
all known
by the general name of
'
Tati,' gradually furnished
a
convincing exi)lanation.
The most striking feature noticed
at
Kakshal, as well as at all other
'
Tatis,' was that the above-
named fragments rest on nothing but natural loess, either hard
DEBRIS-STREWN
'
TATIS.' 175
or more or less disintegrated into a sandy condition. It was
easy to ascertain that the soil underneath contained neither
walls nor other structural remains
;
for the small banks of
loess which rise here and there from the general level of a Tati,
sometimes to a height of lo to
15
feet, and on the top of which
the fragments usually lie thickest, invariably displayed on their
bare sides the natural soil without any trace of ancient deposits.
In the formation of these banks, as in all other features of
such sites, it was impossible to mistake most striking evidence
of the erosive action of the winds and sandstorms which sweep
the great desert and its outskirts for long periods of the spring
and summer. Only the fragments above described could, by
the hardness and weight of their material, survive, sinking
lower and lower as the ground beneath gets more and more
eroded, while everything in the shape of mud walls, timber,
etc., as ordinarily used in the construction of Turkestan
houses, has long ago decayed and been swept away. Even
the potsherds which have withstood destruction
bear plain
evidence of the slow but continuous onset to which they have
been exposed, in their small size and in their j:)eculiarly rough
surface, that looks as if it had been subjected to "grounding."
It is evident that such a process of erosion at sites of ancient
habitations could not have gone on during the long centuries
since their abandonment without also considerably lowering
the ground level. But the erosion has not proceeded uniformly
over an entire area, as shown by the banks of loess already
referred to, which are now seen rising like small plateaus or
islands above the more disintegrated parts of a
'
Tati.'
Whether they derived comparative protection from the greater
abundance of hard debris with which they are ordinarily
covered, or from some other special feature, it is certain that
they are most useful to the archaeologist as evidence of the
original ground level. Coins, much corroded metal orna-
ments, stone seals and similar small objects which can \\ith-
stand the force of the winds, are occasionally picked up from
Tatis. A few of the latter situated beyond Guma were found
to be named in the list of places where Islam Akhun alleged
he had made his discoveries of paper manuscripts or
"
block-
176 ON THE EOAD TO KHOTAN.
prints." But the examination of the very first sites passed
sufficed to show that the physical conditions absolutely precluded
the possibility of such relics surviving there.
It was not in the hope of striking finds of this kind that I
wandered for a long time over the debris-strewn waste of
Kakshal, though it was getting late and Moji, the end of the
march, was still far off. There was a weird fascination in
the almost complete decay and utter desolation of the scanty
remains that marked once thickly inhabited settlements.
Occupied in the examination of small pottery pieces with
ornamental designs, etc., which my men picked up again and
again, I found it difficult to tear myself away even when the
last red rays of the sinking sun had strangely illumined the
yellow soil and its streaks of reddish-brown pottery. The
route to which I rode off at last was difficult to see, for
invading waves of low sand dunes had to be crossed for
several miles before scrubby ground was reached again near
the little village of Chudda. The moon had come up by that
time, and as I was riding comfortably along guided by its light
I could indulge in reflections regarding the strange places I
had seen, without risking loss of the track. It was close on
eight o'clock when I arrived at last at Moji, where my tent
was ready to receive me.
On the 7th I was induced to make a halt at Moji by the
quantity of old coins that were brought to me, almost all of
an early Muhammadan ruler who calls himself in the legend
Sulaiman Khagan. The site from which they had been
obtained, and which I proceeded to examine early in the
morning, lies only a mile to the north of the village, and is
known as Togujai. There I found a number of loess banks
covered with broken })ottery, similar to that seen on the previous
day, but less affected by erosion
;
and the now dry ravines
which the flood water of the early summer had cut through
them were the place where the old coins had been extracted.
A number of men had accompanied me from the village by
the local Beg's order, and their search soon furnished me with
numerous i)ieces of pottery showing ornamental designs and
often glazed in bright colours. In the bed of the ravine others
OLD KEMATNS AT MOJI. 177
set to work to seek for coins, and from the burrows they made
half a dozen copper pieces were dug out in my presence. There
can be no doubt that these coins have been washed out
originally from the same debris layers to which the pottery
belongs. Thus a clue is gained for the date of the latter, which
may help in regard to the chronology of other sites. With
the pottery there is found a great deal of broken glass and
small bits of jade. Among the former I noticed a number
of pieces with that iridescence which is so frequent in the ancient
glassware of the West. The production of glass is a long-
forgotten art in Turkestan.
From Togujai I rode to another old site, one and a half
miles north-east of Moji, known as Hasa, which had already
attracted Dr. Hedin's notice. It is undoubtedly a Muham-
madan cemetery, but there is no clear indication as to its date.
On a small hillock, from which skulls and skeletons were
protruding, I found a number of graves covered with wooden
boards. One of them which I opened showed the remains of
a child, wrapped in the cotton stuff of the country, and turned
towards the Qibla in accordance with orthodox practice.
Though the graves are supposed to be those of Shahids, i.e.,
Musulmans who fell fighting the infidel, the men with me had
no scruples whatever in exposing their contentsa proof that
fanatical superstition can have no deep hold on them. The
-and of the desert has invaded this resting-place, and em-
phasises its look of desolation. The sky was in full accord
with the scene, dust-laden and hazy. Of the mountains no
trace could be seen, though Sanju, whence the Karakorum
route starts, lies only some twelve miles south of Moji. It
seems to me probable that Moji occupies the position indi-
cated by Hiuen-Tsiang for the town of Po-kia-i, where a famous
Buddha statue brought from Kashmir was worshipped in the
pilgrim's time.
On the 8th of October an easy march of fourteen miles over
a gravel-covered Dasht with scanty patches of scrub brought
me to Zanguya. The bed of the stream, which is crossed
immediately before entering the fields of Zanguya, was entirely
dry,
the water being at this season used up for irrigation. Zan-
12
178 ON THE ROAD TO KHOTAN.
guya is a fairly large oasis, counting over five hundred houses
in its several hamlets. I crossed through a long covered Bazar
and found beyond it, near the eastern end of the village, a
pleasant camping-ground in a field of lucerne. In the evening
I visited an old village site, called Kul-Langar, some two miles
to the north-west on the edge of the desert. Besides old
pottery and the like I here found the remains of two large
tanks still clearly traceable.
'
On the 9th I marched to Pialma, some nineteen miles from
Zanguya. The first couple of miles of the road lies through
irrigated land
;
but as the water supply is scant, cultivation
shifts every year in turn to one of the four great plots into
which the land on this side of the oasis is divided. Were it
possible to secure more water by storage or otherwise, no
doubt most of the barren Dasht which lies towards Pialma
could be brought under cultivation. Light dunes of sand
appeared again about the middle of the march, and continued
up to the strip of raised ground appropriately known as Bel-
kum ("the top sand"). Some miles beyond I sighted the
ruined mound of Karakir, which proved to be an ancient
Stupa, much decayed but still holding its own among the high
dunes of the surrounding drift. The base of the structure
when intact must have been about
65
feet square. The size
of the bricks agreed closely with that observed in the Stupa of
Mauri Tim.
At Pialma, which is quite a small place, counting only
about a hundred houses, I reached the last oasis of the
Karghalik district eastwards. My camp was pitched in a
little fruit garden, the trees of which were still laden with
excellent peaches. For my servants the house of the owner
offered ample room. Felt carpets and mats are quickly spread
over the raised platforms that surround the principal room of
these pleasant dwellings, and every time I ins])ccted my servants'
quarters along the route I was surprised by their improvised
comfort. Not only the average standard of living but also
the housing of the agricultural population of Eastern Turkestan,
seemed far above the level observed among the corresponding
class in any part of India.
179
CHAPTER XII.
ARRIVAL IN KHOTAX.
A LONG march on the loth of October was to bring me at last
to the very confines of Khotan.
Up to Ak-Langar, the regular
stage some fourteen miles from Pialma, the route lay over an
absolutely barren plain of hard loess and grav'el. Two half-
decayed pillars on the road a few miles from Pialma mark the
boundary between Karghalik and Khotan. At Takhtuwen,
about half-way, there is a well sunk to a depth of nearly 200
feet, and at Ak-Langar another, almost equally deep. After
the long lonely marches on the flat of the desert, I hailed with
delight the appearance of the mountains which from Pialma
onwards showed themselves more and more to the south, though
the light haze hanging over the landscape never lifted com-
pletely. After Ak-Langar sand appeared in low dunes forming
the semi-lunes so familiar to me from Ordam-Padshah. By
the time I reached the Mazar of Kum-rabat-Padshahim
("
My
Lord of the Sands Station
")
we were again in a sea of sand.
Amid these surroundings the lively scene that presented
itself at the shrine popularly known as
"
Pigeons' Sanctuary
"
(Kaptar-Mazar) was doubly cheerful. Several wooden houses
and sheds serve as the residence for thousands of pigeons,
which are maintained by the offerings of travellers and the
proceeds of pious endowments. They are believed to be the
offspring of a pair of doves which miraculously appeared
from the heart of Imam Shakir Padshah, who died here in
battle with the infidel, i.e., the Buddhists of Khotan. The
youthful son of one of the Shaikhs attached to the shrine was
12*
180 AREIVAL IN KHOTAN.
alone present to tell me the story. Many thousands had fallen
on both sides, and it was impossible to separate the bodies of
the faithful
'
Shahids
'
from those of the
'
Kafirs.' Then at
the prayer of one of the surviving Musulmans the bodies of those
who had found martyrdom were miraculously collected on one
side, and the doves came forth to mark the remains of the fallen
leader. From gratitude, all travellers on the road offer food to
the holy birds. I too bought some bags of Indian corn from
the store of the shrine, and scattered their contents to the flut-
tering swarms.
While watching the pretty spectacle I could not help being
reminded of what Hiuen-Tsiang tells us of a local cult curiously
similar at the western border of Khotan territory. Some thirty
miles before reaching the capital,
"
in the midst of the straight
road passing through a great sandy desert," the pilgrim describes
"
a succession of small hills," which were supposed to be formed
by the burrowing of rats. These rats were worshipped with
offerings by all the wayfarers, owing to the behef that in ancient
times they had saved the land from a great force of Hiung-nu,
or Huns, who were ravaging the border. The Khotan king had
despaired of defending his country, when in answer to his prayer
myriads of rats led by a rat-king destroyed over-night all the
leather of the harness and armour of the invading host, which
then fell an easy prey to the defenders.
"
The rats as big as hedgehogs, their hair of a gold and silver
colour," of which Hiuen-Tsiang was told as inhabiting this
desert, are no longer to be seen even by the eyes of the pious.
But the locality he describes corresponds exactly to the position
of the
'
Kaptar-Mazar ' relative to ancient Khotan, amidst
dunes and low conical sandhills covered with tamarisk bushes,
while the manner in which the pigeons kept at the shrine are
propitiated with food offerings by all modern wayfarers mani-
festly marks a survival of the Buddhist legend.
Just
like
Hiuen-Tsiang's rats, so now the holy pigeons which have taken
their place are supposed to recall the memory of a great victory. .
It was in fact the first striking instance of that tenacity of
localBJ
worship which my subsequent researches showed for almost all
sacred sites of Buddhist Khotan. In Kashmir and on the Indus
RURAL EWIRONS. 181
it had been no small advantage for me to find the position of
old Buddhist or Hindu shrines I was in search of marked in-
variably by Muhammadan Ziarats. So I might well take it for
an auspicious omen that my entry into Khotan territory brought
me across a pious local custom which the Muhammadans of
this region had derived from their Buddhist ancestors.
Three miles beyond this curious shrine the road emerged
from the sandy billows on to a low-lying marshy plain. Here
we halted near the solitary hut of Tarbugaz Langar. The news
of my coming had preceded me from Pialma. So late in the
evening I received the visit of the Beg of Zawa. the next \nllage
tract. He was a fine-looking, genial old man, and I appreciated
his kindly welcome on this my first night on the soil of Khotan.
Camels and baggage ponies had felt the length of the pre\'ious
day's march. So I decided to divide the remainder of the journey
into two short stages. Soon after leaving Tarbugaz cultivated
ground was entered, and three miles from it I passed the mud
fort which was erected in Yaqub Beg's time, closing the road to
Zawa, the first large viUage of Khotan.
From there onwards there lay an unbroken succession of
gardens, hamlets and carefully cultivated fields on both sides.
The road itself is flanked by shady avenues of poplars and
willows for almost its whole length. Autumn had just turned
the leaves yellow and red on most of the trees, and after the
monotonous khaki of the desert marches this display of colour
was doubly cheerful. On the road the dust lay ankle deep.
It was easy to realise the vicinity of a great trade centre from
the hvely traffic which passed us. I saw strings of donkeys
carrying
'
Zhubas,' the lambskin coats for the manufacture
of which Khotan is famous. Few, indeed, were the passers-by
that did not ride on some kind of animal

pony, donkey, or
bullock.
To proceed to any distance on foot must seem a real
hardship even to the poorer classes. No wonder that the people
see no reason to object to the ridiculously high heels of their top
boots.
When riding the inconvenience cannot be felt. But
to see the proud possessors of such boots waddle along the road
when
obliged to use their legs is truly comical.
Some seven miles from Zawa I passed the stony bed of the
182 ARRIVAL IN KHOTAN.
Kaia-kash
("
Black-jade
")
Darya, the second main river of
Khotan. Its bed, fully three-quarters of a mile broad, betokened
the great volume of water it carries down in the summer from
the glaciers towards the Karakorum. But at this season the
river, diminished no doubt by the demands of irrigation, finds
room in a single channel, about thirty yards broad and one to
two feet deep. I was delighted to come at a distance of about
a mile and a half beyond upon a second river-bed, that of a
branch of the Kara-kash known as the Yangi-Darya,
"
the New
River." Whatever the age of the designation may be, the posi-
tion of this bed agrees most accurately with the accounts which
Chinese historical records give as to the rivers west of the old
capital of Khotan. The site of
"
Borazan," which I knew to
contain in all probability the remains of this ancient capital,
lay too far off the road to be visited immediately.
In one of the hamlets of Sipa, east of the
"
New River," I
found a garden that offered a quiet camping-ground. While
watching the unloading of my baggage I was not a little surprised
by the appearance from a neighbouring house of a man chained
by the neck to a heavy iron rod of almost his own length. It
was a cultivator who had been sentenced to this punishment
some months ago for grievously assaulting a neighbour. Cruel
as the weight of the chain looked, I could not help thinking that
the mode of punishment had its practical advantages. Instead
of being imprisoned the man could remain with his family and
follow any occupation not requiring quick movements. At the
same time the sight of the inconvenient appendage he has to
carry must act as a sufficient deterrent to others, and the guilt
of the culprit is constantly brought to notice.
On the morning of the 13th of October I was just about to
start from my camp at Yokakun for Khotan when the Beg
arrived whom the Amban, on hearing of my a})proach, had
deputed to escort me. The Beg was in his Chinese gala garb
and had his own little retinue. So we made quite a cavalcade,
even before Badruddin Khan, the head of the Afghan mer-
chants in Khotan and a large trader to Ladak, joined me a few
miles from Khotan town with some of his fellow-countrymen.
I rode round the bastioned walls of the great square fort that
CAMP IX EESIDENTIAL GARDENS. 183
forms the
"
New City
"
of the Chinese, and then through the
outskirts of the
"
Old City
"
to the garden belonging to Tokhta
Akhun, a rich merchant, which Badruddin Khan had previously
taken up for my residence. The narrow Bazars passed on the
way were more than usually squalid. The number of people
afflicted with diseases whom I saw in them was also depressing.
In the garden which lay close to the southern edge of the suburb
HOUSE OF TOKHTA AKHUN, KHOTAN.
of Gujan I found a large though somewhat gloomy house, but
none of the attractions of my Yarkand residence. The maze
of little rooms all lit from the roof and badly deficient in ventila-
tion could not be used for my own quarters. Outside in the
garden there was a picturesque wilderness of trees and bushes,
but little room for a tent and still less of privacy. So after
settling down for the day and despatching my messages and
presents for the Amban, I used the few remaining hours of
184
AKRIVAL IN KHOTAN.
daylight for a reconnaissance that was to show me the imme-
diate environs, and also a more congenial camping-ground.
There is a charm about the ease with which, in these parts,
one may invade the house of any one, high or low, sure to find
a courteous reception, whether the visit is expected or otherwise.
So when after a long ride through suburban lanes and along the
far-stretching lines of mud-built fortifications erected after the
last revolt against the Chinese, but already crumbling into ruin,
I came about half a mile from Tokhta Akhun's upon another
residential garden, enclosed by high walls and surrounded by
ifields, I did not hesitate to have my visit announced to the
owner. Through a series of courts I entered a large and airy
reception hall, and through it passed into a large open garden
that at once took my fancy. Akhun Beg, a fine-looking, portly
old gentleman, received me like a guest, and when informed
of the object of my search readily offered me the use of his resi-
dence. I had disturbed him in the reading of a Turki version
of Firdusi's Shahnama. My acquaintance with the original
of the great Persian epic seemed to win for me at once the good-
will of my impromptu host, and I hesitated the less about
accepting his offer. So when next morning my tent was pitched
on the lawn in front of a shady clump of trees, I again enjoyed the
peace and seclusion of a country residence.
At noon I paid my first visit to Pan-Darin, the Amban,
after the usual preliminaries required by Chinese etiquette. I
found him a quiet, elderly man, with features that seemed to
betoken thoughtfulness and honesty of purpose. His kindly
though somewhat abstracted look and his gentle manners of
gesture and speech impressed me from the first as entirely in
agreement with the reputation for learning and piety which
has followed this Mandarin wherever he was employed in the
province. Dressed in his state clothes and surrounded by
numerous attendants, Pan-Darin received me with every mark
of attention. He had long before been informed from Kashgar
of the objects of my visit, and I was curious to see what his
attitude would be, both as to ex]ilorations in the desert and
my proposed survey of the mountains about the sources of the
Khotan river.
FIRST MEETING WITH PAN-DARIN. 185
To my delight there was no trace of obstruction to be dis-
covered in what Pan-Darin had to tell me as to either project.
He had no doubt that ancient places amidst the dreaded sands
of the
'
Gobi,' if they existed at all, were difficult to reach, and
that the statements made about them by natives were not to
be trusted too readily. In the mountains again the routes were
PAN-DARIX, AMBAN OF KHOTAN, WITH PERSONAL ATTENDANTS.
bad. imphing hardships and risks, and beyond the valleys of
Karanghu-tagh there lay the unknown uplands of Tibet where
Chinese authority ceased, and where, under the strict orders of
the Tsung-li-Yamen, no assistance was to be rendered to tra-
vellers. But apart from these natural difficulties and political limi-
tations Pan-Darin offered to give me all help that lay in his power.
The Amban's simple, earnest ways, his eWdent comprehension of
the scientific objects^j^in view, and the scholarly interest with
186 AREIVAL IN KHOTAN.
which he followed my explanations about Hiuen-Tsiang's travels
and the old Buddhist culture of Khotan, induced me to put
reliance in this promise of help. And subsequent experience
showed me how well it was justified. Without his ever ready
assistance neither the explorations in the desert nor the survey
work in the mountains which preceded it could have been
accomplished.
As soon as I had arrived in Khotan I had commenced the
local inquiries which were to guide me as to ancient sites par-
ticularly deserving exploration and as to the best means for
organising a systematic search for antiquities. Apprehensions
about possible forgeries had prevented me from sending in
advance to Khotan information as to the main object of my
journey. I now found that some time would have to be allowed
for the collection of specimens of antiquities from the various
old sites which
"
treasure-seekers
"
were in the habit of visiting.
"
Treasure-seeking," i.e., the search for chance finds of precious
metal within the areas of abandoned settlements, has indeed
been a time-honoured occupation in the whole of the Khotan
oasis, offering like gold-washing and jade-digging the fascina-
tions of a kind of lottery to those low down in luck and averse
to any constant exertion. In recent years, owing to the con-
tinued demand of European collectors from Kashgar and else-
where, the small fraternity of quasi-professional treasure-seekers
had learned on their periodical visits to ancient sites to pay
attention also to antiquities as secondary proceeds. Neverthe-
less, all the information that could be elicited about such
localities, even from persons who seemed reliable, was exceed-
ingly vague, and I soon realised that if I were to set out without
having before me specimens distinctly traceable to specific sites,
much valuable time might be lost and labour wasted. In order
to secure such specimens, Badruddin Khan, who had previously
rendered useful services to Mr. Macartney, offered to organise
and send out small
"
prospecting " parties. Their return,
however, could not be expected before a month, and I decided
to utilise this interval for the interesting geograj^hical task which
I had already marked out for myself in the mountains south of
Khotan.
PREPARING FOR THE :M0UNTAINS. 187
That portion of the Kuen-luen range which contains the
headquarters of the Yurung-kash or Khotan River had hitherto
remained practically unsurveyed, the scanty information avail-
able being restricted to the sketch map of the route by which Mr.
Johnson, in
1865,
had made his way from Ladak down to
Khotan. Colonel Trotter had, in
1875,
expressed the belief
that the head waters of the Yurung-kash were much further to
the east than shown on that map, and probably identical with
a stream rising on the plateau south of Polu. Captain Deasy,
working from the side of Polu in
1898, succeeded in reaching the
sources of this stream at an elevation of over
16,000 feet, but
was prevented from following it downwards. Thus the true
course of the main feeder of the Yurung-kash, together with
most of the orography of the surrounding region, still remained
to be explored.
The close approach of winter made me anxious to set out
for this task with as little delay as possible, while it was neces-
sary to equip properly the men as well as the ponies that were to
accompany me, for the cold mountain region to be visited. My
camels could be of no use in that direction, and extra ponies
were needed for the baggage with which I was to move up, greatly
reduced as it was. The animals of the
*
Kirakash
'
or profes-
sional caravan men were all away on the Karakorum route,
where the autumn months are the busy time for the trade with
Ladak. To buy ponies for this comparatively short tour would
have been jm expensive arrangement. So I felt glad when the
Amban, on returning my visit the next day, issued orders to
supply me \vith the transport needed on hire from neighbouring
villages.
While Badruddin Khan busied himself with procuring the
fur-clothing for my men and the felt covers for the ponies, I
managed to pay a visit to the village of Yotkan, the site of
the old capital of Khotan and a well-known find-place of anti-
quities of all sorts. It was an interesting day I spent at that
locality, where the accumulated debris layers of the old city,
embedded deep below the present level, are being regularly
washed for gold, and in the course of these operations \ield up
also
ancient pottery, coins, seals, and similar remains. But I
188 ARRIVAL IN KHOTAK
need not here detail the impressions of that first hurried visit
;
for subsequent investigations were to render me far more familiar
with this important site.
During the few days of my stay at Khotan much of my time
was taken up with the inspection of the coins, terra-cotta figures,
and other antiquities that were brought for sale by villagers and
"
treasure-seekers." Most of the bagfuls contained only the
broken pottery and copper coins found so plentifully at Yotkan,
and already fairly well known from previously formed collec-
tions. But their inspection was a useful training to me, and I
thought it advisable to make at first ample purchases so as to
stimulate the zest of professional searchers.
I was naturally on the look-out too for those
"
old books
"
written or
"
block-printed
"
in a variety of unknown characters
which, as already mentioned, had during the last five or six years
been sold from Khotan in increasing numbers to European col-
lectors at Kashgar. In regard to these acquisitions the suspicion
of forgery had before presented itself to competent scholars, but
evidence was wanting to substantiate it, and in the meantime
these strange texts continued to be edited and analysed in learned
publications. Offers in this article were surprisingly scanty at
Khotan itself, and curiously enough the very first
"
old book
"
that was shown to me supplied unmistakable proof of forgery.
Hearing of my presence at the place, a Russian Armenian from
Kokand brought me for inspection a manuscript on birch-bark,
consisting of some ten ragged leaves covered with an
"
unknown
"
script. He had bought it for forty roubles, undoubtedly as a
commercial speculation, and now wished to have his treasure
properly appraised.
I saw at once that the birch-bark leaves had never received
the treatment which ancient Bhurja manuscripts, well known
to me from Kashmir, invariably show. Nor had the forger
attempted to reproduce the special ink which is needed for
writing on birch-bark. So when I applied the
"
water-test
"
the touch of a wet finger sufficed to take away the queer
"
un-
known characters" both written and block-printed. It was
significant that the
"
j^rintcd matter
"
of this manifest forgery
showed a close resemblance to the formulas of certain
"
block-
SUSPECTED FORGERIES.
189
prints
" contained in the Calcutta collection. In fact, my
inquiries indicated a close connection between the person from
whom the Armenian had purchased the leaves and Islam Akhun,
the treasure-seeker whose alleged places of discovery I had vainly
endeavoured to locate about Guma. Local rumour credited
Islam Akhun with having worked a small factory for the pro-
duction of
"
old books." But at this time he was keeping away
from Khotan, and there were reasons to postpone personal in-
vestigations about him.
On the day preceding my start for the mountains I was
cheered by the opportune arrival of my Dak from Yarkand.
The contents of my home mails, despatched via India, did not
come down later than the 17th of August. But the evening
before I had received a letter sent to Kashgar through the
Russian post and thence forwarded with the official Chinese
Dak, which had been written as recently as the 19th of Sep-
tember. No more convincing proof is needed of the compara-
tive proximity to which the advance of the Russian railway
system has brought even this distant corner of Turkestan,
described by Sir Henry Yule in
1865 as
"
the most inaccessible
and least known of Asiatic States." The quotation is from the
great scholar's
"
Cathay and the Way Thither," a work which
followed me every\vhere on my travels, and the reading of
which never failed to provide both learned guidance and amuse-
ment.
MUZTAGH PEAK, IN KUEN-LUEN RANGE.
CHAPTER XIII.
TO THE HEADWATERS OF THE YURUNG-KASH.
At midday of the 17th of October I set out for my journey
into the mountains, after taking a friendly leave of Akhun Beg,
my white-haired host. A five-rouble gold piece, presented in
a little steel purse, as a return for the use of his garden, was
accepted without much difficulty. I was glad to leave behind
in Badruddin's care all stores and other articles not imme-
diately needed. Nevertheless our baggage, including the
survey instruments and food supplies for a full month,
required ten ponies. The first march was luckily a short and
easy one. For about six miles we proceeded south through
cultivated land, dotted with hamlets, to the village of Jamada,
not far from the left bank of the Yurung-kash. Beyond it
the bare Dasht rises gently towards the foot of the mountains,
which now stood clear of the haze that had veiled them at
Khotan.
On the sandy plain south of Jamada I found a
'
Tati
'
with relics of ancient settlement. Fragments of pottery are
strewn over the site, and some villagers brought me old coins,
beads, and a few small seals, one showing the figure of a
Cui)id.
We then rode for four miles over the high banks of
DEBOUCHURE OF YURUNG-KASH. 191
stone and gravel which the river has brought down from its
course in the mountains, and at last crossed to the right bank.
The bed of the Yurung-kash is ov-er a mile broad at this point,
but the water flowed only in a few narrow channels. The
rest is diverted into the canals that feed the villages of the
eastern part of the Khotan oasis. Our night's quarters were
at Bizil, a small village close to the river-bed, where man}'
burrows and pebble heaps showed the working of jade-seekers:
The stone, which has from ancient times been so highly prized
in China, and to which the river owes its name,
"
White-jade,"
is still an important product. As I crossed the river-bed I
thought of the distant lands to which it has carried the name
of Khotan.
Beyond Bizil, to the south, low, undulating slopes of much-
decayed conglomerate ascend towards the mountains. Over
these we travelled on the morning of the i8th of October:
Several ridges, fairly steep on the north side, but joined by
almost level terraces on the south, form natural steps in the
ascent. Gravel and coarse sand, with scarcely a trace of vege-
tation, covers the ground
;
and the landscape, save for the
distant view of the Khotan oasis below, was one of complete
desolation. When the last of the steps was crossed by the
Tashlik-Boyan Pass, I found myself in full view of the outer
ranges through which the Yiu-ung-kash flows in a tortuous
gorge, and greeted with relief some snowy peaks that raised'
their heads above them, far away to the south, A long descent
over a sandy slope brought us to the Kissel Stream, along
which our onward route lay. Half smothered by the dust
that the ponies raised as they scrambled down, we reached
the bottom of the valley at the little hamlet of Kumat. A
narrow strip of level ground by the side of the Kissel and
irrigated from it, supports some fifteen
famihes. It was soon
dark in the deep and narrow glen, and the four miles we had
to march to Yangi-Langar,
our night
quarters, seemed very
long. The night air was still and warmer than in the plain
of Khotan, the thermometer showing
48''
F. at 8 p.m.
On the 19th of October a march of some eighteen miles up
the winding gorge of the Kissel brought us to Tarim-Kishlakj
192 HEADWATERS OF YURUNG-KASH.
On the whole way there was no habitation, nor indeed room
for one. The rough path crossed innumerable times the stream
that flows between high and precipitous spurs of conglomerate
and what looked to me like sandstone. In more than one
place there was a difficulty in getting the laden ponies over
the rocks that fill the narrow bottom of the gorge. As this
jumbled mountain mass has never been surveyed, it was
tantalising to wind along between the rocky walls without
a chance of an open view. But there was no time to be lost
with climbs to points that might give one. Tarim-Kishlak
("
cultivated holding
")
consists of a single miserable mud
dwelling amid a few fields of oats. Apart from the small
patch of sloping ground that is irrigated from the stream,
there is nothing around but decayed rock and ravines filled
with gravel. Compared to the absolute barrenness of these
hill-sides, the vegetation of the Hunza or Sarikol glens would
look quite luxuriant.
On the morning of the 20th of October I found the little
stream, by the side of which my tent was pitched, half covered
with ice. The boiling-point thermometer indicated an eleva-
tion of close on
9,000
feet, and the air at
7
a.m. was just at
freezing point. The gorge we ascended continued for another
eight miles in a south-easterly direction. Then the path
leaves the stream which comes from a high mountain capped
with snow, and strikes up a dry side gorge to the south. Here
all trace of rock disappeared from the surface of the hill-sides.
Loose earth and detritus were alone to be seen, with scanty
patches of hardy scrub. Before we reached the pass, a strong
wind sprung up that overcast the sky with clouds and shrouded
us in dust. So when at last by 2 p.m. we stood on the Ulugh-
Dawan
("
High Pass
"),
the distant view to the south was
seen through a haze. All the same, when I had climbed with
the Sub-Surveyor a ridge rising about
500
feet above the pass,
we were rewarded by the sight of a grand glacier-girt mountain
rising in solitary splendour to the south-east. It was im-
possible to mistake the
"
Kuen-luen Peak, No. 5," which the
tables of the Indian Trigonometrical Survey showed with a
height of
23,890
feet. Right and left of it stretched a chain
I
CROSSING OF ULUGH-DAWAN. 193
of ice mountains, but their crests were hidden in clouds, and
our endeavour to recognize among them other peaks fixed
from the southern side was in vain. The wind on the pass
was cutting and the temperature close to freezing-point. By
boiling-point thermometer we found the height to be over
12,000 feet.
I was glad to leave by
4
p.m. the cheerless ridge. The
descent into the Buya Valley, which runs from east to west
draining by an inaccessible gorge into the Yurung-kash, was
very steep and trying. The bleak mountain-side is fissured by
narrow ravines, and the path follows the ridges between them;
The landscape looked wild and lifeless in the extreme. It was
quite dark before we had extricated ourselves from the rocky
ledges that project from the decomposed slopes and lead
ladder-like down to the valley. With some difficulty our
guide found the way to the main group of huts of Buya, but
the straggling baggage animals were much belated, and I had
to sit till midnight in a smoke mud hovel before my tent was
pitched and my dinner ready.
Next morning when I rose I found to my delight that the
sky had completely cleared. In order not to lose the good
chance for survey work, I decided to push on to Pisha, though
men as well as animals seemed in need of a day's rest. The
valley of Buya, about a mile broad at the principal hamlet,
supports from its scanty fields of oats a population of thirty
odd holdings. The level of our camp was close on 8,000 feet.-
To the south of the valley rises a series of plateaus showing
on the surface only detritus and gravel, with conical hills
crowning them at intervals. When we had climbed the crest
of the nearest plateau the whole of the great snowy range
towards Ladak and the westernmost border of Tibet lay spread
out before us. Over the whole chain towered the great Kuen-
luen Peak already referred to, with its glaciers now clearly
visible. The Un-bashi
{"
head of Ten
")
of Buya, an uncouth
looking hillman or
'
Taghlik,' knew the peak only by the
name
of
'
Muz-tagh '
("
the Ice-mountain
").
Apart from
the
glittering
walls of snow and ice in the far south, there was
nothing
to be seen before us but the yellowish slopes of the
13
194 HEADWATEES OF YUEUNG-KASH.
plateaus that mark where transverse ridges must once have
risen. The extremes of temperature, and possibly the ex-
cessive dryness of the climate, with the consequent absence of
vegetation, may partly account for the extraordinary disin-
tegration of the soil. In colour and outlines the near view
reminded me of the hill ranges that are seen when passing along
the Egyptian coast of the Gulf of Suez. The plateaus are
separated by broad depressions in which tiny streams of saltish
water try to make their way towards the Yurung-kash.
Except when the snow melts on the distant mountains east-
wards, there is no moisture to fill these ravines.
Thus we marched for about ten miles to the south-west,
glad that the ground offered no difficulty to the tired ponies.
From a high ridge that crowns the last plateau southwards, I
sighted the broad and partly cultivated valley of Pisha, and on
its other side the ridge that still separated us from Karanghu-
tagh, the last inhabited valley at the northern foot of the
Kuen-luen, our immediate goal. At
5
p.m. I arrived at Kul-
dobe, the main hamlet of the Pisha Valley, where two dozen
or so of Taghliks were assembled to welcome me. There
seemed little in their speech or manners to distinguish them
from the people of Khotan. But their sheepskin coats and
hard weather-beaten faces indicated the difference of the
climatic conditions. Many among them had never seen the
plains. Harsh and bare of all graces are their surroundings.
I wondered whether they ever see flowers such as carpet the
Pamir grazing-grounds.
The 22nd of October was needed as a day of rest for men
and beasts, and I was glad to grant it in a locality where
there was at least plenty of shelter. The sky was heavy with
clouds, and cold blasts swept up the valley from time to time,
enveloping it in a haze of dust. After a morning spent over
notes and letters I went for a walk along the stream through
cheerless fields and with nothing in view but the bare grey
spurs that line the valley. On my return I found the whole
grown-u]) male j)opulation of Pisha assembled in the courtyard
of the mud dwelling where my men had established them-
selves. It seemed that for many years past Pisha had known
I
VALLEY OF PLSHA.
195
no such time of excitement and novel interest. In Hakim
Shah, the oldest man of the valley and father of the local
Yiizbashi, I found an intelligent interlocutor. He claimed an
age of fully a hundred years, and his \VTinkled face and snowy
hair seemed to support his assertion. Though bent by the
burden of his years, the old man was still active enough in
mind, and he talked glibly of the days of early Chinese rule
before the Muhammadan revolt. He had once in his life been
to Khotan, and was evidently in the eyes of his people a man
well-up in the affairs of the world.
My men had been told that a difficult and long march lay
before us. So on the morning of the 23rd they were quicker
than usual about the start. When I got outside my tent a
little after six o'clock I saw to my delight a gloriously clear
sky. The cold was also a surprise. Even at
7
a.m. the ther-
mometer showed
23
F.
;
the little watercourse near my tent
was hard frozen. As soon as we had climbed the edge of the
plateau some
500
feet above Pisha, a grand view opened out
upon the whole ice-crowned range. Kuen-luen Peak No.
5
now lay in full view to the south-cast, and its glacier-crowned
head appeared quite close in the absolutely clear atmosphere.
For about eight miles we rode over a broad, barren plateau that
rose with an easy gradient towards the south. Then I turned
off the track and climbed a high ridge eastwards that from a
distance promised a good surveying station.
Its height,
13,950
feet above the sea, commanded a panorama
more impressive than any I had enjoyed since I stood on the
slope of Muztagh-Ata. To the east there rose the great Kuen-
luen Peak with its fantastic ridges separated by ghttering
glaciers and its foot rising from a belt of strangely eroded bare
ridges, as shown by the photograph at the head of this chapter.
By its side the gorge of the main branch of the Yurung-kash
could clearly be made out as it cuts through the series of
stupendous spurs that trend northwards from the main snowy
range of the Kuen-luen. From the latter the great peak was
thus entirely separatedan interesting observation fully in
accord with the orography of the Karakorum and Hindukush.
There it has long ago been remarked that the points of greatest
13*
196 HEADWATEES OF YUEUNG-KASH.
elevation are not to be found on the actual watershed, but
on secondary spurs detached from it.
The deep-cut valleys and serrated ridges descending from
the main range presented a most striking contrast to the flat,
worn-down features of the plateaus behind us. To the west
the course of the Yurung-kash was lost in a jumble of rocky
walls that gradually sank away towards the plain. In the
north there showed itself as one unbroken mass the gaunt
conglomerate range which we had crossed on the way to Buya,
culminating in a broad, snow-covered peak, the Tikehk-tagh,
some distance to the east of the Ulugh-Dawan. Nature
could not have created a better survey-station than the ridge
on which I stood. With the enjoyment of the grand pano-
ramic view there mingled the satisfaction of seeing so large
and interesting a tract hitherto unsurveyed suddenly spread
out before me as if it were a map. While Ram Singh worked
away at his plane-table I was busily engaged in taking a
complete circle of views with the photo-theodolite. Not-
withstanding the perfectly blue sky it was bitterly cold on
that height, as my fingers soon felt in handling the delicate
instrument.
It was nearly three o'clock before our work was done, and I
was able to hurry down hill. I had noticed how distant the
valley of Karanghu-tagh was where we were to finish the day's
march, and the guides from Pisha had, with unwonted anima-
tion, dwelt on the badness of the track leading to it. After a
comparatively easy descent of two miles we reached the line
where the high plateau so far followed falls off towards the
Yurung-kash Valley in a scries of precipitous ravines. The one
which the track follows at first looked exactly like the gorges
I had seen in Astor leading down to the Indus. High rt)ck-
faces lined its sides, and the withering effects of atmos])heric
influences seemed here less marked than on the ranges passed
northwards. At an elevation of about ii,ooo feet the patli
crossed a rocky neck eastwards, and then led down prccijii-
tously to the river flowing more than
3,000
feet below.
It was just getting dark as we began this trying part of tlie
descent, but even if it had been broad daylight it would have
I
PRECIPITOUS DESCENT. 197
been impossible to ride. The angle at which the path zigzags
down the precipitous cliff was so steep that the ponies could
be dragged forward only with difficulty. The loose stones that
cover the path increased the trouble, while the deep dust in
which they are imbedded at times almost smothered us.
Never had I marched in such a dust-cloud as that which
enveloped us until, after an hour and a half's scramble, the
bottom of the valley was reached at the point where the
Yurung-kash is joined by the Kash stream flowing out from
the side valley of Karanghu-tagh.
It was perfectly dark when we crossed to the left bank of
the Yurung-kash by a rickety bridge consisting of three badly
joined beams laid over a chasm some
70
feet wide. The foam
of the river tossing deep down in the narrow bed of rocks
could be made out even in the darkness. In daylight, and in
a less tired condition, the crossing might have affected one's
nerves more. As it was, I felt heartily glad when I saw the
ponies safely on the other side. Karanghu-tagh means
"
Mountain of blinding darkness," and at the time of our
approach the appropriateness of the name could not have been
doubted. For about an hour we and our tired beasts groped
our way between the boulder-strewn bank of the Kash stream
and the foot of steep hill-slopes before we reached at last the
village that bears that cheerful name. The baggage had
arrived safely, but also with great delay, and thus it was late
in the night before I could retire to rest.
The 24th of October was spent at Karanghu-tagh, where-
arrangements had to be made for men and yaks to take us
further into the mountains. The survey of the previous day
had shown me that the only way by which the source of the
main branch of the Khotan River might possibly be approached
would lie in the gorge of the river itself. The Yiizbashi and
the old men of the little village, whom I summoned in the
morning,
at first denied stoutly that the valley of the Yurung-
kash was accessible beyond the point where we had crossed it.
By-and-bye,
however, I eUcited the fact that there were
summer
grazing-grounds in some of the nullahs descending
from
Muztagh, and then the fact of their being reached by a
198 HEADWATERS OF YURUNa-KASH.
track up the Yurung-kash had to be acknowledged. Of a
route across the main range south, by which Mr. Johnson
appears to have come on his rapid descent from Leh to Khotan
in
1865,
I could get absolutely no information. It was evident
that the hill-men feared the trouble and exposure of a tour
in those high regions. At the same time the serious and very
puzzling discrepancies I discovered between the sketch-map of
Mr. Johnson's route and the actual orography of the mountains
south of Pisha convinced me that I could not dispense with
local guidance. My interest, however, lay eastwards where
the course of the Yurung-kash was to be traced. After a
time Islam Beg, a young and energetic attendant of the Khotan
Yamen, whom Pan-Darin had despatched with me, succeeded
in making it clear to the surly Taghliks that the Amban's
order for assistance to me must be obeyed. So those who
rule Karanghu-tagh set about to collect the yaks which were
to take on my baggage and the men who were to accompany
me.
It was no difficult task, for Karanghu-tagh, though hidden
away amid a wilderness of barren mountains, is a place of
some resources. When I inspected it in the morning I was
surprised to find a regular village of some forty closely packed
houses. The scanty fields of oats below and above could
scarcely support this population. But Karanghu-tagh is also
the winter station for the herdsmen who graze flocks of yaks
and sheep in the valleys of the Uj)per Yurung-kash. These
herds belong mostly to Khotan
'
Bais,' or merchants, and
the visits of the latter seem the only tie that connects this
strangely forlorn community with the outer world. From
time to time, however, Karanghu-tagh receives a })ermanent
addition to its poj)ulation in the j)ersons of select malefactors
from Khotan, who are sent here for banishment.
It would indeed be difficult to find a bleaker place of exile.
A narrow valley shut in between absolutely bare and pre-
cipitous ranges, without even a view of the snowy peaks, must
appear like a prison to those who come from outside. It was
strange to hear the hill-men, who during the summer lead a
solitary life in the distant glens, speak of Karanghu-tagh as
A PENAL SETTLEMENT.
199
their
'
Shahr 'or
"
town." For these hardy sons of the
mountains this cluster of mud-hovels, with its few willows
and poplars, represents, no doubt, an enviable residence.
To me the strange penal settlement somehow appeared far
more lonely and depressing than the absolute solitude of the
mountains.
I was glad to start soon on a climb to one of the steep
TAGHLIKS AND EXILED CRIMINALS AT KARANGHU TAGH.
ridges north-east of the village, which offered a convenient
station for further survey work. But the day was far less
clear than the preceding
one, and the views too were less
inspiriting.
On my return I passed the cemetery of Karanghu-
tagh. The number of tombs it contains may, in view of the
very
scanty population
(barely amounting
to 200 souls), be
taken as a sign of long-continued
occupation. There were
plenty of decayed little domes of mud and wooden enclosures
marking
graves. Over them
rose high staffs, invariably hung
200 HEADWATERS OF YUEUNO-KASH.
with a yak's tail. I counted also two mosques in the place,
and half-a-dozen simple Mazars, where a bundle of sticks
bedecked with rags and yaks' tails marks the reputed resting-
place of some holy man. I could well believe that the dreari-
ness of their earthly surroundings might turn the minds of the
dwellers in this gloomy vale to a happier world beyond.
The information extracted with no little trouble from the
Yiizbashi of Karanghu-tagh and his people about a route
up the main valley of the Yurung-kash was by no means
encouraging. They acknowledged that a little settlement
existed in the Omsha
Jilga,
one march up the main valley, and
that a path accessible to yaks led beyond to a point where a
hot spring flows into the river. But after this no possible
track could be found through the mountains. Whether this
was true or not could be made certain only by personal
inspection. Yaks were to carry the indispensable baggage
and to serve as riding animals for myself and my men. The
ponies which had been severely tried by the preceding marches
were to remain at Karanghu-tagh in charge of Niaz Akhim,
the Chinese interpreter. He had complained of the hardships
previously experienced. It was easier for me to part with
him than with
'
Yolchi Beg,' my little terrier. He had
bravely kept up so far, but the long marches had evidently
told on him, and a rest would give him fresh strength for the
fatigues still before us.
By 10 a.m. on the 25th of October the yaks were packed and
the caravan was ready to start. With each animal I took a
hill-man to guide it. Yaks are as sluggish as they are sure-
footed, and without a man to drag each animal by the rope
which is passed through its muzzle the rate of progress would
be amazingly slow. I arranged that each man should be pro-
vided with food for ten days, and secured extra yaks to carry
these rations. Karanghu-tagh has perhaps never seen so
grand a procession as when my caravan set out on the march.
The whole village turned out to witness the spectacle.
After passing down the Kash valley for about two miles we
struck to the east, and, crossing the spur I had before ascended,
moved into the side-valley of Busat. Not far from the point
GORGE OF YURUNG-KASH. 201
where it bifurcates into two narrow gorges leading up to the
mountain wall southwards, the path ascended a high cross-spur.
From its top, at an elevation of close on 12,000 feet, the glaciers
of the great Muztagh, and all the gorges leading down to the
main stream, were visible in great clearness. So the photo-
theodolite was brought to work again, though the weather was
not as favourable as on the day when I marched to Karanghu-
tagh. Early in the afternoon for several days past I had noticed
the same atmospheric change, a strong north wind rising and
bringing clouds and a dust haze that soon covered the sky.
From the Boinak spur an easy path led down for some
three miles to where the mouth of the Omsha Valley descending
from the west face of Muztagh opens into the Yurung-kash
gorge. The river, which we here crossed to the right bank,
was about
50
yards broad, and nowhere deeper than
3
feet-
Its water had a delightful bluish-green tint, and reminded me
by its limpidity of the mountain streams of Kashmir and the
Alps. I wondered how to account for this clearness of the
water, seeing that the Yurung-kash must be fed very largely
by the glacier water of the Muztagh and other peaks. Of the
large volume of water which it carries down during the summer
months, the broad strips of boulder-strewn ground were a
plain indication.
On the 26th of October I woke again to a gloriously clear
morning, and soon forgot in the rays of the rising sun that it
had been 24 F. at
7
a.m. From Terek-aghzi, where I had
camped by the river-bank, a steep path led up to a long grassy
spur known as Zilan, jutting out from the mountain side
northward. On reaching its top, after a climb of two and a
half hours, I was rewarded by a splendid view of the glacier-
girt Muztagh and the rugged snowy range southwards. Some
four miles to the south-east the Yurung-kash gorge com-
pletely disappeared between the series of stupendous spurs
of rock which descend from the great peak on its left and the
main range opposite. Looking up towards the mighty
southern buttresses o:
"
K.5," and the frowning ice-peaks
showing their heads above them, it required almost an effort
of
imagination to believe that behind lay those Pamir-like
202 HKADWATERS OF YUEUNG-KASH.
uplands in which, as I knew from Captain Deasy's explorations,
the Yurung-kash takes its rise. That there was no practicable
route over the rock-walls through which the river has cut its
way past Muztagh, was absolutely clear from the view before
us. But there remained the chance of the river-bed itself
offering the desired passage. This hope occupied my mind as I
i
*l!*f^lt),(5,^l^ljl|i^
VIEW UP THE YURUNG-K.^SH GORGE, WITH SPURS OK PKAK K.
5
O.N LlCri.
descended by a difficult track just practicable for yaks to the
left bank of the river, at the point known as Issik-bulak,
"
the
hot spring." On the sheer cliffs opposite my camping-ground,
and at a height of about 300
feet above the level of the river-
bed, I could see a hot spring issuing in considerable volume.
The hill people are said to bathe in its water when the winter
makes the river easily fordable. The half-a-dozen herdsmen
ATTEMPT TO PENETRATE GORGE.
203
of Omsha, who had joined me on the way, unanimously declared
that they had never passed beyond this spot, and that the
gorge further up was inaccessible for human feet. Whether
their assertion was true, or whether the formidable ravine
ahead would not yield us an opening, was a question that only
the morrow's exploration could answer.
On the 27th of October a day's hard climbing among the
rocks, shingle, and boulders of the Yurung-kash gorge verified
the Taghliks' prediction. As soon as the sun had fairly risen
over the great mountain walls to the east I started with Ram
Singh, Tila Bai, the most active of my people, and three hill-
men from Omsha. Foreseeing that we should have to cross the
river in the course of our reconnaissance, I had three of the
biggest yaks taken along. At first we followed the steep hill-
side above the right bank where our camp was pitched, as its
height promised a better view of the ground ahead. We had
made our way for about a mile and a half onwards when all
further progress was barred by a ravine descending from a
great height and flanked b}- wholly unscaleable rocks. The
view I had before me was wild in the extreme. I could now
clearly make out the walls of frowning cliffs which, broken
only by almost equally precipitous shoots of rock and shingle,
lined the foot of the great spurs falling off to the river. The
passage left for the river seemed nowhere more than 200 feet
wide, and at places considerably less. The volume of water
reduced by the autumn now filled only one-half to three-
fourths of this space. But the beds of huge boulders seen
along the actual channel were not continuous, but alternately
on the left and right bank. Where the river flowed with light
green colour over boulders and ledges, we might hope to effect
a crossing.
But where it whirled round the foot of sheer cliffs
the water showed a colour of intense blue, and was manifestly
far
deeper. Yet it was clear that our only hope lay in being
able to follow up the river-bed.
To descend to it was no easy matter from where w^e stood.
But
after marching back for half a mile we found a practicable
slope
and managed to scramble down to the edge of the water.
When
the yaks had been dragged down too, with much trouble.
204 HEADWATERS OF YUEUNO-KASH.
we began to make our way up the ravine. A wall of im-
passable rock, with a stretch of deep water at its foot, forced
us soon to search for a ford to the opposite side. This we
found, and thanks to the yaks, which waded splendidly in the
ice-cold water undismayed by the rapid current, we managed
to get safely across. The yak is a difficult animal to guide,
even on the best ground
;
when in the water any attempt to
control its movements would be useless. So it was with a
feeling of relief that I noticed the instinctive care with which
our yaks made their way from one convenient boulder to the
other. The limpid water made it possible for them to see
their way as much as to feel it.
On the left bank we had scarcely advanced a few hundred
yards over jumbled masses of rock that had been swept down
from the slopes above, when we were stopped again by a
precipitous rock-face washed at its foot by the ominous blue
water. To cross over to the opposite bank, where a stretch of
boulder-strewn ground might have allowed an advance, was
quite impossible. The yak we drove into the water to test its
depth was soon obliged to swim, and had we attempted the
passage we should have had to follow its example. In order
to effect a crossing here with the needful baggage a raft or
boat was manifestly indispensable. But how could we secure
it in this forlorn region, where wood was practically unobtain-
able, and where the people had never even heard of that most
useful implement, the
'
Massak,' or inflated skin ?
The only chance of progress left was to take to the crags
above us, and to trust that further on a descent might be
found again to a practicable portion of the river-bed. After a
difficult climb of some 500 feet I managed to bring myself
and my men safely to a narrow flat ledge, but the yaks had
to be left below. We followed the ledge for some hundreds
of yards until it ended at the flank of a ravine that woukl
have defied any cragsman. A careful search for a point
where we might descend again to the river was in vain. The
steep shingly slope terminated everywhere in cliffs that offered
no foothold. Battled in these endeavours, I climbed uj) the
precipitous hillside above the ledge that liad brought us so
FORCED TO TURN BACK. 205
far, in the hope of turning the ravine. But after an ascent of
about 1,000 feet I convinced myself that the ground beyond
was one over which I could never hope to move either yaks or
men with loads.
While I was resting on a little projecting ridge the noise of
faUing stones drew my attention to a herd of wild goats (Kiyik)
that were evidently about to descend from the cliffs opposite.
The tracks of these animals I had already noticed on the hill-
side. They alone are likely ever to have penetrated into the
wild gorge that lay before me. The point where a large
stream from the glaciers of Muztagh falls into the Yurung-
kash seemed temptingly near. Once beyond this junction
there would be less difficulty in crossing the river, and conse-
quently in ascending its bed. Yet there was no hope of
reaching this point until perhaps the river was completely
frozen, an eventuality for which it was impossible to wait.
Even then I doubt whether a practicable passage could be
found, considering the climatic conditions and the masses of
fallen rock likely to be encountered.
All day an icy wind had been blowing down the valley,
giving a foretaste of the cold that might be encountered at
this season on the elevated plateau where, in view of our
survey results, the source of the river can now be definitely
j
located. I did not envy the yaks the bath they got in crossing
j
back to the right bank, and was heartily glad to reach the
j
shelter of my tent at the hour of dusk. The night was cloudy
1
and still, and on the following morning snow was falling on
I
the mountains down to about
3,000 feet above our camp,
!
the elevation of which by aneroid was close to
9,000
feet.
j
Down in the river gorge the temperature at
7
a.m. was a
little higher than on previous days (34^ F.), but as soon as
we ascended by the path we had come before it became bitterly
cold, and the wind was piercing. Winter had already set in
fnr
these regions.
For the return to Karanghu-tagh I chose the route through
tne Omsha Valley, into which we crossed without much diffi-
culty over the ridge of Soghak-Oghil, at an elevation of about
11,500 feet. At the central hamlet of Omsha I found two
206 HEADWATEKS OF YUKUNG-KASH.
low mud-built houses among a few fields of oats and some
troglodyte shepherds' dwellings. The weather cleared in the
afternoon, and I felt grateful for the warming rays of the sun
before he set behind the mountains. The valley of Omsha,
though scarcely a quarter of a mile broad, looked quite spacious
and inviting after the awful gorge of the main river. Notwith-
YAKS CARRYING BAGGAGE IN YURUNG-KASH GORGE, NEAR KARANGHU-TAGH.
standing the elevation of about 10,000 feet, oats are said to
grow well in years when a sufficient snowfall on the mountains
around assures irrigation.
The elevation of Omsha, together with the change in the
weather, made itsoJf felt by a truly cold night. On the morn-
ing of the 29th of October the thermometer at
7
a.m. showed
only
17
F. But the sky was of dazzling clearness, and in
the crisp mountain air the cold had an almost exhilarating
effect. After a pleasant march of two hours we reached the
OMSHA VALLEY. 207
right bank of the Yurung-kash, close to Terek-aghzi. Instead
of the previous route, I now followed the path by the river-
side. It crosses the Yurung-kash about two miles below the
above junction, and then winds along the precipitous cliffs of
the left bank for another three miles. The ups and downs
over slopes of loose conglomerate were very fatiguing, but the
picturesque views of the wild river-gorge amply made up for
this. At one point the river has cut its way through walls of
solid rock, scarcely
50
feet apart, for a distance of several
hundred yards. Elsewhere the vehemence of floods has
excavated yawning caverns from the huge alluvial fans. Not
far from the point where the Kash Valley from Karanghu-tagh
joins this gorge, the path led over a succession of rocky ledges
of remarkable steepness. The ascent indeed looked hke a
huge flight of stairs built by nature along the brink of a pre-
cipice more than
500 feet high. The yaks climbed it with
astonishing
surefootedness, but it was uncomfortable to look
down on the track over which they had carried us.
208
CHAPTER XIV;
OVER THE KARA-KASH RANGES.
Our previous survey, including the expedition up the Yurung-
kash gorge, had cleared up the important question as to the
true origin and course of the main feeder of the Khotan River.
The next and equally interesting task was to map the head
waters of the streams which drain the portion of the Kuen-luen
range south and south-west of Karanghu-tagh, and are mani-
festly the principal tributaries. In the course of my inquiries
from the Omsha herdsmen about dominant points that would
enable me to sight again the series of magnificent glaciers which
feed the Kash River, I had ascertained that there was a difficult
path just practicable for laden yaks crossing the transverse range
north-west to Karanghu-tagh. It was said to lead to the Nissa
Valley, whence a track could be found to the mountains on the
upper Kara-kash River. I was delighted at this intelligence. For it
showed not only, what the Karanghu-tagh i)eople had carefully
hidden from me, that there was a connection with the outer
world besides the route via Pisha, but also that this connection
would take me into a region which had so far remained an abso-
lute terra incognita.
The start for Nissa, on which I accordingly decided for the
morning of October 30th, was attended with some difficulty.
The Yiizbashi of Karanghu-tagh, who had before proved ob-
structive, evidently did not cherish the idea of helping us to
follow a route the knowledge of which he seemed anxious to
keep for his own people. So, notwithstanding the previous
orders, no yaks turned up in the morning. When the man
IN NISSA
VALLEY. 209
saw that I was in earnest and[that further delay was likely to
involve him in more serious consequences than the voluminous
objurgations
to which Islam Beg and Niaz, the Chinese inter-
preter, had treated
him already, the yaks were dragged out
from the
neighbouring
glens. But we had lost two hours
a
long time at that season when night falls so early in the narrow
valleys.
At lo a.m. we started up the Kash stream, and after about
two miles turned into a narrow glen known as Gez
Jilga. When
after a
toilsome climb of close on three hours we had reached
the Pom-tagh Pass, about
12,400
feet above the sea, a grand
view opened to the east and south. It comprised the whole
glacier-crested
range from
'
Muztagh
'
on the extreme left to
the hoary peaks which showed their heads above the glaciers
closing the Karanghu-tagh Valley. No visible point in the
glittering crest-line which filled about one-third of the horizon
could be much under 20,000 feet, while quite a number of the
peaks, as subsequent triangulation showed, reached
22,000 to
23,000
feet. Nearer to the south-west and west there rose a
perfect maze of steep serrated ridges and steeple-like peaks.
Embedded among them, but quite invisible lay the narrow valleys
forming the grazing grounds of Nissa. I climbed a knoll on the
water-shed ridge some
400
feet above the pass, where work with
the plane-table and photo-theodolite kept us busy for a couple
of hours. It was an ideal day for survey work
;
scarcely a cloud
lay on the horizon, and the air, with 50
F. in the shade, felt
deliciously warm.
An extremely steep track, by which our ponies
were led
with difficulty, took us first along a bare rocky ridge and then
down, at least
3,000
feet, by a narrow ravine to the Karagaz
gorge. WTien we had reached its bottom by half-past
four
it was getting quite dusk between the high and precipitous
rock-walls. As we descended for about two miles in this narrow
defile to where it joins the gorge of the Nissa stream, the red-
dish glow of the evening sun that had set for us long before
lit up some towering pinnacles in front. It was like a magic
illumination, this display of red light on the yellowish crags
devoid of all trace of vegetation.
Only in the Tyrol Dolo-
14
210 OVER. THE KARA-KASH RANGES.
mites, and on a smaller scale in the defiles where the Indus
breaks through the Salt Range, had I seen the like.
The Nissa gorge which we had next to ascend was equally
confined, and the darkness which now completely overtook us
made the long ride, with our ponies slowly groping their way
between the boulders of the river-bed or along the narrow
ledges, most wearisome. Here and there in bends of the defile
we passed scanty patches of cultivated ground, with low mud
huts inhabited only during the summer months. The wicked
Yiizbashi who by his delay had caused this trying night march,
and who was now accompanying the baggage, came in for some
blows from my men as we passed the belated yaks, a long way
yet from the end of our march:
When at last we arrived at Nissa, I was glad of the tem-
porary shelter which the hut of the
'
Bai ' of the little settle-
ment offered. My host owed this proud title to the possession
of some yaks and a flock of sheep, and his habitation was but
a mud-built hovel. All the same, it was a cheerful change from
the raw night air to the warmth and light of his fire-place.
The 31st of October we halted at Nissa, The men needed
rest and Ram Singh time for astronomical observations, I used
the day to collect information regarding the mountain routes
that lead to the Kara-kash Valley westwards and towards
Khotan, but found it no easy task
;
for the apprehension of the
trouble that my tours might cause made the hillmen more than
usually reticent. Nissa counts some twenty houses, but most
of the men who inhabit it during the winter were still away
with the sheep and yaks on the higher grazing-grounds. Apart
from a few willows and a bold snowy peak visible at the head of
the valley, there was nothing to break the monotony of the dusty
grey of the rocks and the little plain between them. But the
sky showed the purest blue and the sun shone warmly. So
the day passed pleasantly even in these surroundings.
It is lucky for historical geography that the name Nissa is
not that of a locality further West. Else it could scarcely have
escaped identification, at the hands of amateur antiquaries,
with Nysa, the mythic residence of Dionysus in the Indian Cau-
casus, which Alexander too is supposed to have visited. It
SURVEY ABOVE BRENJAK PASS. 211
amused nie to think of the flights of imagination that would be
required in order to clothe these most barren of rocks with the
vines sacred to the god whom the great conqueror flattered him-
self by imitating in his Indian conquest.
On the morning of November ist I set out for the Brinjak
Pass, which connects the Nissa Valley with the mountain defiles
northward. As I was anxious to utilise the extensive view likely
to be obtained from its height for a final survey of the head-
waters of the Yurung-kash, I decided to camp as near as possible
to the pass in order to secure plenty of time for the morrow's
work. It was not easy to carry out this plan, as the steep rocky
ravine in which the ascent lay was exceptionally narrow. But
at last a point was reached about 12,800 feet high by aneroid,
where the narrow bottom just left room for a couple of tents.
So giving order to pitch the camp here, I climbed the steep
ridge south of the ravine. My reconnaissance showed that a
splendid survey station could be secured by ascending a high
arete north-east of the pass. The piercing cold wind soon drove
me down to my tent, which seen from above in the narrow gorge
looked curiously like a stretched-out bat, the outer flaps touch-
ing the rocky slopes on either side. The interior did not give
ease, for the steep slope allowed the use of neither table nor
chair, and the camp-bed, too, could not be placed at an angle
of less than 25.
Whether it was through the unaccustomed
position or the continual slipping away of the rugs that were to
keep off the bitter cold, I got little sleep that night.
At
7
a.m. the temperature was only
21^
F., and the little,
stream close by was frozen soUd. An hour's stiff cUmb brought
me up to the Brinjak pass, for which the aneroid showed a
height of about
14,000 feet. To ascend the steep ridge previously
singled
out for survey work was no easy task, as the whole of it
proved to be covered with confused masses of boulders and flaked
rock, showing the force with which decomposing agencies are
at work at this altitude. After a few hundred feet the yaks
carrying
the instruments could be got no further. The theodolite
could
not be exposed to the risk of this scramble from rock to
rock,
but the Taghlik to whom I entrusted the photo-theodolite
managed
to follow though with great difficulty. The ridge
212 OVER THE KARA-KASH RANGES.
gradually narrowed to a precipitous grat. After an hour and
a half's climbing I had reached its highest knoll, where hard
frozen snow filled the interstices of the rocks.
To the north-east, but separated from us by a great dip in the
ridge, rose a steeple-like peak, the Mudache-tagh, we had already
sighted from the Pom-tagh Pass. To climb it would have been
a stiff piece of mountaineering, even if time had sufficed. This
KUEN-LUEN RANGE, WITH GLACIERS OK-NISSA VALLEY, SEEN FROM BRINJAK.
peak,
17,220
feet high, shut off the view of the second trian-
gulated peak above Buya, upon which we should have had to
rely for theodolite work. But otherwise the view was as grand
and clear as could be desired.
'
Muztagh
'
showed itself in full
majesty, and beyond it to the south-east there now appeared
several distant snowy ridges previously invisible that guard the
approach to the main Yurung-kash source. How should we have
fared between them if the passage above Issik-bulak could have
been negotiated ? Further to the south the line of the horizon
A TRYING DESCENT. 213
for a distance of close on one hundred miles was crowned by
an unbroken succession of snowy peaks and glaciers.
The nearest to us were those at the head of the Nissa Valley
below a prominent cone, for which subsequent triangulation
showed a height of
23,070
feet. But bigger still looked the
ice-streams that descend in a huge amphitheatre above the
valley of Karanghu-tagh. Further to the south-west and west
the steep crags of the Chankul and other neighbouring peaks
shut off a distant view. They were aU gUttering with fresh
snow, probably from that fall which we had witnessed at Issik-
bulak
;
but the beds of snow filling the ravines of the Iskuram
valley enclosed by these peaks looked old, more like incipient
glaciers.
The sky was the brightest azure, and its colour only
heightened the effect of the dazzling glacier panorama south-
wards. Though it was midday and the actinic power of the
sun's rays considerable, the temperature in the shade kept about
25
F. Fortunately there was little wind, so I managed to do
the photo-theodolite work without much trouble. But I was
glad when, after an hour and a half's exposure, I could again
warm my benumbed fingers. The aneroids showed a height of
15,300
feet.
By half-past one our work was finished
;
Ram Singh had been
able to verify by good intersections the plane-table work of
the last ten days. Once back on the pass our yaks could be
used again for the descent northwards into the valley which
drains the Iskuram peaks. But an unexpected difficulty
retarded the descent. About half a mile from the pass where
the track enters a narrow ravine we suddenly came on hard
ice below a crust of detritus dust. It was the recent snow that
had melted in the few hours of sunshine, and had subsequently
got frozen. Even the yaks slid uncomfortably on this treach-
erous ground, and the slopes below the path were suf&ciently
steep to make a shp dangerous. The leather mocassins
('
Charuk
')
of my companions here gave safer foothold than
my boots with Alpine nails worn flat by previous marches. So
I gladly availed myself of their assistance at the worst bits.
Ice and dustthe combination appealed to me as charac-
214 OVER THE KARA-KASH RANGES.
teristic of this strange and forbidding mountain-land of Khotan.
But I felt grateful when, after about an hour's cautious progress,
we had got clear of this trying ground. Lower down the ravine
somewhat widened, and just as it was getting dark we arrived
at the little grazing-ground of Chash, which gives its name to
the valley. My tent was pitched on a small plot of withered
grass
;
behind it under the shelter of a projecting rock-wall
my men established themselves. Close by, huddled under the
side of some rock cavities, I found a couple of small felt huts
inhabited by Taghlik families who live here summer and winter.
They owned only a few sheep, and were said to subsist mainly
upon charitable gifts from the shepherds of the Borazan canton
who drive their flocks up here during the summer months.
The ample scrub growing in the valley enabled these poor people
to withstand the rigours of the winter which, at an elevation of
about io,ioo feet, must be considerable.
In the course of the evening four Taghliks arrived from
Mitaz, the nearest hamlet northwards, in response to the
summons sent by my Beg. They assured us that fodder had
been sent ahead to an intermediate halting-place. This was
welcome news, as our supply from Nissa was running out
;
but the hoped-for information as to a route across the moun-
tains to the Kara-kash Valley was not to be got out of the
distrustful hillmen. Every question about localities was met
with a stereotyped
'
bilmaidim
'
("
I do not know "), until
even the stolid herdsmen from Nissa laughed at this pretended
ignorance. It was evident that the arrival of strangers, such
as they had never before beheld or perhaps even heard of, filled
these good people with all kinds of apprehensions.
After the hard work of the previous day I was glad that on
the 3rd of November my men could start late when the air had
warmed up a little in the bright sunshine. For about three
miles we descended the Chash Valley, until it turns eastwards
to flow through an impassable rock defile towards the Yurung-
kash. Our way continued to the north up a narrow side valley
flanked by sheer cliffs of conglomerate. At its entrance we
watered the ponies
;
for the glen higher up is absolutely waterless,
except for a salt spring unfit for drinking. After another eight
VIEW FROM
YAGAN-DAWAN. 215
miles we arrived at the foot of the Yagan-Dawan, and pitched
camp at the highest point where there was still room for a tent
in the steep ravine leading up to the pass. Three bags of ice
had been brought from Chash to provide us with water.
The night, thanks to the sheltered position, was passed in
comparative comfort, and next morning the bright sunshine
induced me and Ram Singh to clamber up the pass long before
the baggage was ready to start. Some of the Nissa men had
bolted overnight, and this caused trouble, for the yak is an
obstinate animal and each wants one man quite to himself when
carrying baggage. That day one man had to sufi&ce for three
or four of them, and the poor fellows left behind were mani-
festly in for a bad time. The Yagan-Dawan proved a very
narrow saddle flanked by steep ridges on the east and west.
In order to get a full view we climbed the western ridge, and
reached its top at an elevation of about 12,000 feet. It
was a splendid survey station, completely commanding the con-
fused network of rocky ridges and deep-cut raxines which ex-
tends between the middle courses of the Yurung-kash and Kara-
kash. We now stood on the watershed between the two rivers.
But the high serrated range we had crossed from Nissa shut off
the view of the great snowy mountains south, and even of the
dominating Muztagh we could only sight the glacier-covered
northern buttresses. So the hope of triangulation was once
more doomed to disapp>ointment.
I shall never forget the view that opened westwards and in
the direction of the distant plains. There were lines upon
lines of absolutely bare rocky spurs, closely packed together
and running mostly from south to north
;
between them, shut
in by unscalable rock slopes, was a maze of arid gorges, of
which the bottom could not be seen. It was like a choppy
sea, with its waves petrified in vN-ild confusion. Far away on
the horizon this rocky waste was disappearing in a yellow haze,
the familiar indication of another region which knows no life
the distant sea of sand.
The impressions gathered in front of this panorama were
heightened
when, after three hours' busy work, we descended
into
the ravine leading down from the pass to the north-west.
216 OVER THE KARA-KASH RANaES.
About 1,50b feet below the saddle the bottom was reached,
and then began a passage of fantastic rock defiles, the like of
which I had never seen. For nearly three hours I marched
between walls of conglomerate and apparently chalky rock
rising thousands of feet above the narrow fissure at the bottom.
As it appeared to me in my total want of geological training,
only the erosive action of water, aided by extreme disintegra-
EKODED RANGES lO NORTH-WEST, SEEN FROM ABOVE YAGAN-DAWAN.
tion of the rocks under peculiar climatic conditions, could have
produced these extraordinary formations. But of water there
was no trace, only ankle-deep dust overlying the detritus. For
the first four or five miles there was scarcely even scrub growing
in these terribly barren gorges
;
animal- life seemed completely
absent. The want of water did not physically distress me, as
it did our ponies and yaks, which had tasted no drop for more
than twenty-four hours. Yet my attention was ever turned to
it in contrast, by the sight of the huge, overhanging cliffs, tlic
THROUGH FANTASTIC GORGES.
217
cavities, and isolated pinnacles, which all looked as if water had
worked them.
Above my head the sky was still blue, and the higher cliffs
reflected bright sunHght
;
yet the gloom of these ravines and
their desolation were depressing. I also knew that my baggage
was painfully straggling, the yaks proving unmanageable with
so few men, and knocking off their loads whenever they found
a conveniently projecting rock. So I was doubly pleased when
after a march of about eight miles from the pass I emerged into
the fairly open valley of Mitaz. There I found still warm sun-
shine and a lively stream from which my pony drank in long,
long draughts. I enjoyed the splash and sound of the water
after those silent dead ravines, and sat cheerfully by its side
until my baggage appeared at dusk. It was pleasant to read in
the tiny seventeenth-century edition of Horace, which always
travels in my saddlebag, of the springs that gave charm for the
poet to another mountain region far away in the West. And
then the question touched my mind : WTiat is this vast mountain
world in human interest compared to the Sabine Hills ? It has
no past history as far as man is concerned, and what can be its
future
?
unless destiny has reserved the prospects of another
Klondyke for the auriferous rivers of Ivhotan.
On the 5th of November our start was late
;
for the men from
Nissa had to be paid off, and it took time before those of Mitaz
had got their animals ready and loaded. Mitaz is a very small
hamlet, and its eight or nine holdings lie scattered higher up the
valley. The latter after our previous route, looked comparatively
open, but in reality the only available track lay close along, or
in, the river-bed. The water, beautifully clear, was nowhere
more than two feet deep. So our continual crossings, neces-
sitated by projecting rock spurs, caused no great trouble except
to
'
Yolchi Beg,' who had to be caught each time and carried
across on horsebacka procedure to which the httle fellow never
submitted in good grace.
We marched this day some sixteen miles down the stream
to the north, but saw no human being, except the children of
a
shepherd
family hving in a httle cave close to where the
Sukosai
Valley runs down from the west. The eldest of four
218 OVER THE KARA-KASH RANGES.
children was a blind boy of seven. Smallpox had deprived
him of his eyesight, but he knew his way about the valley, and
I had less trouble than usual in getting from him the local
names of the immediate neighbourhood. The only reward
I had at hand was a silver piece, which he promised to give
to his mother. We camped at the point where an alternative
route to the plains, by the Kunat Pass, leaves the Mitaz valley
eastwards. It was said to be impassable for horses, and its
entrance, a narrow rockbound gorge, looked sufficiently for-
bidding.
At Kunat-aghzi, where the hypsometer showed a height of
only
6,890
feet, and where the temperature at
7
a.m. was just
at freezing-point, I had the feeling of nearing the plains. But
the Ulughat Pass that was still to be crossed had a surprise
in store. On the 7th of November we marched for about eight
miles down the Mitaz stream, when the view to the right showed
us a broad, sandy slope leading up to a high ridge. In striking
contrast to the serrated cliffs of the ranges around, no rock pro-
truded from this uniform slope. Hence it looked far lower
than in reality it was. I knew the optical deception which
made Ram Singh estimate the height before us at only about
1,000 feet
;
yet I was not prepared for the climb that awaited
us. For two and a half hours our ponies toiled upwards in zig-
zags along a slope of which the angle seemed nowhere less than
25
degrees. The soil was gravel and loose earth, the last
remains of rock formations that had withered away during un-
known ages. The longer tlie climb lasted the higher rose my
almost abandoned hope of getting a panorama of the whole
range that would give us at last a simultaneous view of several
peaks already triangulated from the Ladak side. On this
depended the chance of fixing our position with absolute cer-
tainty and ultimately connecting Khotan itself with the Indian
5
Trigonometrical Survey. The great Muztagh, which had again
|
and again during our ])revious climl^s appeared before us in un-
mistakable majesty, could alone not suffice for this purpose
;
and other triangulated peaks on the main watershed we had
been unable to recognise with any certainty as long as we were
comparatively near to the unbroken screen of icy ridges.
A GRAND PANORAMA. 219
It was thus with a feehng of eager expectation that I pushed
upwards. The long-stretched back of the mountain forming
the Ulughat-Dawan had become visible when the slope changed
into a series of less steep shoulders. But a projecting spur shut
off the view to the south and kept me in suspense. An hour
ahead of my people and followed only by Ram Singh, I gained
at last a small saddle in the main ridge. By ascending a broad
knoll to the south I should soon learn whether my hope was to
be fulfilled. So we left the ponies and hurried up. It was a
moment of intense joy when, arrived at the top, I beheld the
grand panorama that suddenly revealed itself. The whole of
the mountain-world traversed during the last three weeks lay
before me, and beyond it a semicircle of great snowy peaks which
had been hidden hitherto by nearer ranges. Far beyond Muz-
tagh we could see glittering ranges in the direction of the main
Yurung-kash source. The glaciers we had passed at the head
of the valleys between Issik-bulak and Nissa were now seen to
be surmounted by ice-peaks of the most varied shapes, domes,
pyramids, and bold steeple-like cones. To the west there rose
a grand chain of snowy mountains encircling the head waters
of the Kara-kash River. No European eye had ever seen them
from the south. Towards the north only a narrow belt of eroded
rocky ridges separated us from the great desert plain and its
fringe, the Khotan oasis.
The sky was brilliantly clear all around, but over the plains
there hung the ever-present haze of dust. It covered and
effaced with its tinge of brownish-yellow alike the sand of the
desert, the river courses, and the belt of cultivated land. Where
it touched the horizon, far away in the Taklamakan, the skyhne
showed a brilliant light green. Yet in height this cover of dusty
atmosphere could scarcely exceed i,ooo ft. For we could clearly
see the foot of the outer range rising above the bed of the Kara-
kash where the latter winds through the low glacis-like plateaus
stretching
away northwards.
It was three in the afternoon when I arrived on this com-
manding
height.
It was manifest that no time remained for
theodolite
work, for which nature herself seemed to have des-
tined
the position, and that we should have to remain there
220 OVER THE KARA-KASH RANGES.
for the night. The saddle on the main spur offered a con-
venient spot on which to place the camp, but the want of water
was a difficulty. Fortunately I had foreseen this chance and
sent Islam Beg ahead to Pujia, a village on the Kara-kash.
He had orders to meet us on the pass with fresh ponies and a
supply of water. So when my baggage arrived a little before
sunset, the tents were pitched close below our survey station.
Before this the plane-table had come up, and we eagerly searched
the horizon southwards for points previously triangulated and
shown on our section sheet.
This time my hopes were not to be disappointed. Having
once determined our position on the plane-table, it was easy
to recognise in a great ice-pyramid towering above the Kash
valley glaciers the Kuen-luen Peak No. i of the Indian Survey,
21,750
feet high. Its position coincided most accurately with
the direction indicated by our map. In the east the identity
of another high landmark, the
"
Tartary Peak No.
2,"
was
equally assured, and in order to dispel any lingering doubt,
there appeared in a gap of the Iskuram range the glittering snowy
top of a far more distant peak, exactly where the Survey tables
place the
"
Kara-kash Peak No.
2,"
also reaching close to 22,000
feet. This rapid survey made it certain that it was possible to
triangulate the surrounding region down to Khotan itself with
assurance. The direct connection of Khotan with the system
of the Indian Surveys, on which the determination of its exact
longitude depends, had long been sought for in vain. Yet here
a position within a few days' march from Khotan, to which luck
and, perhaps, a little topographical instinct had guided us, gave
the desired opportunity. It only remained to pray for a clear
sky on the morrow.
The sunset on the grand chain of the south was a sight of
incomparable beauty. Long after the serrated crests of the
intervening ranges had sunk into blueish shadows, the icy peaks
beyond the glaciers which feed the western tributaries of the
Yurung-kasli continued in brilliant sunlight. Then one after
the other shone in rosy tints until the glow became a deeper and
deeper red, to pass away into jnuplc and darkness. At last,
only the grand dome of
'
Muztagh,' with its highest pinnacle
NIGHT ON ULUGHAT-DAWAN. 221
shaped like a Phrygian cap, and our newly discovered Kuen-luen
Peak No. i reflected the Hght of the sun that had long before
set for us.
The changes of colour in the tints of yellowish haze over
the plains were dehghtful to watch. But the increasing cold and
the wind that sprung up from the east soon drove me down to
the tent. There a cup of tea boiled from the water I had
brought up in my water-bottle was for hours the only refresh-
ment my establishment could offer. There was no trace of
Islam Beg and his water supply. But I cheerfully put up with
the prospect of not eating my dinner until next morning, in
view of the result which to-day's work promised. The glorious
sight of the full moon rising below us soon drew me outside the
tent. Her light was as clear at our altitude as I had ever seen it
in India, and showed up every crag and recess in the withered
conglomerate ridges eastwards. She looked as if rising from the
sea when first emerging from the haze of dust that hid the plains,
and her light shimmered on its surface. But when she climbed
high up in the sky it was no longer a meek reflection that lit up
the plain below. It seemed as if I were looking at the hghts of
a vast city lying below me in the endless plains. Could it really
be that terrible desert where there was no life and no hope of
human existence ? I knew that I should never see it again in
this alluring splendour. Its appearance haunted me as I sat
shivering in my tent, busy with a long-delayed mail that was to
carry to distant friends myChristmas greetings. At last, about
ten o'clock, a cheerful commotion in the camp announced the
arrival of Islam Beg and the water-filled gourds he had managed
to get brought up. The supply was small, and scarcely sufficed
for a cup of tea for each man. Nevertheless, Sadak Akhun
succeeded in cooking my modest dinner, and after a last look
at the magic city below I could retire to rest close upon
midnight.
Next day when I rose a little before
7
a.m. the sun was just
rising above a lower ridge to the east. He shone brightly into
the tent, but light fleecy clouds were floating in the sky. For-
tunately the horizon to the south above the mountains was
clear,
and I lost no time in beginning the work of triangulation
222 OVEK THE KARA-KASH RANGES.
on our
"
hill-station
"
close by. It was no easy task to select
in this vast panorama the peaks that were the best landmarks
of the numerous ranges within view and also likely to be recog-
nised again from other positions. But after five hours' steady
work
twenty-six prominent points were safely triangulated. The
light clouds that gathered as the day advanced luckily kept
clear of the mountains
;
but coupled with a breeze from the
north-east they made it cold on the exposed height, for which
the triangulation results have indicated an elevation of
9,890
feet.
I took a round of photo-theodolite views, and then we set
about building a mark to enable us to identify our position
with accuracy from the next triangulation station. No stone
could be found anywhere. So the men from Pujia had to collect
the low withered scrub and heap it up mixed with loose earth.
When I descended to the tent I was glad of a cup of tea. But
even more delightful it was to get enough water for washing
hands and face. A fresh supply had been sent up from Popuna,
the next village north on the bank of the Kara-kash. So even
my men ceased grumbling at the halt on this inhospitable Dawan,
and were cheered by the prospect of our early descent to the
plains.
On the morning of the 8th of November we left Ulughat-
Dawan under a sky of speckless blue. Notwithstanding the
elevation the temperature was a little above freezing-point at
7.30
a.m., an indication of the atmospheric influence of the neigh-
bouring plains. For an hour and a half the path led down steeply
over disintegrated slopes of earth and sand which completely
covered the rock structure of the mountain. Only when close
to the head of a narrow gorge did I see rocks showing strata of
mica exposed. Down the bottom of this gorge, scarcely two
or three yards broad, a little stream of water wound its way.
It was so saline that the ponies would not drink from it. After
a mile or two its water was lost in the ground. For fully three
hours the route led between high cliffs of conglomerate and
slate, until a turn round a projecting screen of rock suddenly
brought us out into the open valley of the Kara-kash, just below
Popuna. It was pleasant to see a stretch of level ground again
DESCENT TO KAKA-KASH VALLEY. 223
and rows of trees in their vivid autumn tints. The valley of the
Kara-kash, about half a mile broad, was bounded to the north
by a bank of gravel some 200 feet high, sloping like a natural
glacis gently away towards the plains. Twice we crossed the
Kara-kash, now a stream of beautifully clear greenish water,
some forty yards broad and two to three feet deep, before
Langhru was reached three miles below Popuna. The village,
though counting only about sixty houses, looked quite a large
place to me after my wanderings amidst the solitary mountains.
I could let my men enjoy its comforts only for a single night.
For I knew that a wind raising the haze would effectually stop
further survey work. So I felt anxious soon to reach another
high ridge called Kauruk-kuz, which had appeared from the
Ulughat-Dawan, the only point in the neighbourhood sufficiently
elevated for a second triangulation station, and at the same
time accessible with instruments.
In order to reach it I started on the morning of the 9th of
November back into the arid range southwards by the valley
which leads towards the Kunat Pass. Against all expectation
this valley proved fairly open for a distance of about nine miles.
Then it contracted to a narrow gorge at a point known as
Kuchkach-bulaki, where a little stream of brackish water trickled
down between the rocks, covering the bottom with a saline de-
posit that looked like ice. The cliffs on either side grew higher
and wilder as we advanced up the ravine, and I began to doubt
whether after all a practicable way would offer out of this maze
of contorted rocks to the high ridge I had sighted from Ulughat.
It was getting dark by
4
p.m. when the highest point was reached
to which ponies could advance.
But to my relief there rose on the left a steep slope of detritus,
much like that leading to Ulughat, and evidently the hoped-for
route to the Kauruk-kuz ridge. Camp was pitched in the narrow
ravine, at an elevation of about
8,000 feet by aneroid. I took
it as a lucky omen that just there I came upon a little party from
Nissa,
who had crossed the Kunat Pass with four yaks and were
now waiting for the flour that was to be brought up to them
from
Khotan, The yaks had tasted no water for the last two
days,
but were all the same fit to help us.
224 OVEE THE KARA-KASH RANGES.
The next day's climb proved a stiff one. The ridge which
I had singled out for our station was close on
3,000
feet above
our camp, and the slope was exceptionally steep. But the yaks
carried us safely over the most trying part of the ascent, and
when after three hours the top was reached, Ram Singh as well
as myself was ready to set to work at once. The view was in
some directions more extensive even than that from Ulughat.
But the sky was less clear, and from the first I noticed an ominous
haze that made me hurry on the observations. It was not long
before my apprehensions were verified. A strong wind, passing
from the plains southward, carried the haze further and further
into the mountains
;
there was no mistaking the dust of the
desert that was threatening to overtake our work. Luckily the
identification of the peaks to which previously angles had been
measured by us, caused no delay, and though it seemed like a
race with the veil of dust that was steadily rising, the round of
theodolite observations could be carried through with all need-
ful accuracy. The peaks in the outer range of hills nearest to
Khotan, by which the longitude of the town itself might be
determined thereafter, were first in danger of being wiped from
our horizon. But we were still in time
;
and when the haze,
two hours later, had also obscured the view of the distant high
ranges above the Kara-kash Valley, all but three out of the
twenty-six peaks requiring triangulation had been safely ob-
served. It was with a feeling of relief that I saw this task com-
pleted
;
for I knew how persistent an obstacle the foglike haze
of this region can prove to survey operations. Had I delayed
but for a single dayand, I confess, there had been strong temp-
tationthe chance of this triangulation might have been lost
to us completely. The triangulated height of the ridge was
10,820 feet.
An hour's scramble down the steep slopes brought me again
into the ravine, where the ponies were waiting. As there was
no water at the camp beyond that which had been brought up
from Langhru on donkeys, I had sent word to my people earlier
in the day to move back to the village. The ponies which had
been left behind for us seemed eager too to get at water, and
j
hurried down the valley at a good pace. But it soon got dark
SUCCESSFUL TRIANGULATIOX. 225
and our progress slackened. In the end our guide missed the
track, and in order to make sure of nothing worse happening,
took to the boulder-strewn bed of the dry stream. It was terribly
bad ground for the ponies, and we all felt thoroughly tired by
the time when a big camp-fire guided us late at night to camp in
a field near Langhru.
15
226
CHAPTER XV.
-
'
ANTIQUARIAN PREPARATIONS AT KHOTAN.
On the nth of November the short march to the village of
Ujat, some eight miles lower down on the left bank of the
Kara-kash, was made in an atmosphere so thick and grey that
I had the sensation of a foggy autumn day somewhere near
London. All view of the mountains, near as they were, was
effaced as if with a brush, and from where my tent was pitched
even the bluff spur just across the river at scarcely a mile's
distance loomed only in faint lines through the dust-laden air.
It was this spur, known as Kohmari, the last offshoot of the
Ulughat range towards the plains, which made me place my
camp at Ujat.
Topographical indications that need not be detailed here had
convinced me that M. Grenard, the companion of M. Dutreuil
de Rhins, was right in identifying Kohmari with the lioly
Mount Gosringa which Hiuen-Tsiang describes as a famous
pilgrimage place of Buddhist Khotan. A Vihara, or monastery,
raised on it marked the spot where Sakyamuni was believed
to have preached a
"
digest of the Law
"
to the Devas. A
cave in its side was venerated as the approach to
"
a great
rock dwelling
"
where popular legend supposed an Arhat to
reside
"
plunged in ecstasy and awaiting the coming of Maitreya
Buddha." The Muhammadan Mazar, worshipped as the
resting-place of the saintly
"
Maheb Khwoja," which now
occupies the crest of the conglomerate cliff rising almost per-
pendicularly above the right river-bank, has inherited the
religious merit of the old Buddhist shrine. It forms a
VISIT TO MOUNT GOSRINGA. 227
favourite place of pilgrimage for the faithful of Khotan, who
believe that the intercession of the saint is most efficacious
when the low state of the rivers makes the cultivators fear a
failure of their crops. On this account official recognition,
in the form of a liberal offering from Amban Pan-Darin, was
said to have recently been accorded to the shrine.
The cave which the Chinese pilgrims saw still exists in the
side of the cliff some fifty feet below the crest. It is approached
along a ledge of rock which contains the semi-troglodyte
dwelling of the Sheikhs attending the Mazar. The cave itself,
which is about
40
feet deep and from 8 to 10 feet high, is
believed to have been the refuge of the saint whom the infidels
killed here with smoke. Thus the legend accounts for the
black soot that covers the rock walls. Pious pilgrims are wont
to sit and pray in the cave, and the fires they light to keep
themselves warm in winter time have naturally left their traces
on the rock, A small upper chamber, approached from below
by a ladder, shows above a narrow fissure running into the
rock. The legend heard by the Chinese pilgrim represented
this fissure as a passage which had been miraculously blocked
by fallen rocks to hide the Arhat.
Apart from its association with Hiuen-Tsiang's visit, the
Kohmari cave possessed for me a special interest. From it
the fragmentary birch-bark leaves of the ancient Indian manu-
script in Kharoshthi characters, now known as the Dutreuil de
Rhins MS., were alleged to have been obtained. M. Grenard's
account shows that the leaves were delivered to him and his
companion on two successive visits to Kohmari by natives
who professed to have found them with other remains inside
the grotto. But it is equally clear that neither of them was
present on the occasion or was shown the exact spot of dis-
covery. The men who sold those precious leaves to the
French travellers seem to have prevented them from a
personal inspection of the cave by alleging religious objec-
tions.
No difficulty whatever was raised in my case. I found
the Mullahs, jovial, well-fed fellows, curiously resembling in
their ways my old Purohita friends at Indian
'
Tirthas,' ready
15*
228 ANTIQUARIAN PEEPARATIONS.
enough for a consideration to show me the cave, including its
mysterious recesses. The close examination I was thus able
to effect gave me strong reason to doubt the possibility of the
manuscript having been really found there. Though the visit
of the French explorers was well remembered by the Sheikhs,
nothing was known to them or the villagers of the alleged
discovery in the cave. Taking into account that other frag-
ments of the same manuscript had been sold separately into
Russian hands at Kashgar, it appears probable that the native
"
treasure-seekers
"
concerned made the statement connecting
their find with the cave simply in order to disguise the true
place of discovery.
In the course of my inspection of this sacred cave I had
occasion to appreciate the easy-going ways of Khotan local
worship. Nobody, however good a Musulman he may be,
thinks of taking off his boots on approaching a sacred spot.
Those who wear a kind of over-shoes with their top-boots
leave them outside, it is true. But the common people not
possessed of such refined footgear freely retain their high
leather
'
Charuks ' (mocassins) or the sandals fastened with
long cloth bandages. The winter is cold in this region, and I
wonder how frequent the occasions are when the Khotanese
really do remove their footgear during the winter months. I
have always managed to make friends with the priestly atten-
dants of Indian shrines, be they Hindu or Muhammadan, and
have almost invariably escaped the necessity of taking off my
bootsa kind of deshabille which for a European is incon-
gruous and inconvenient, without in reality marking in any
way
religious conciliation. But in Khotan there seemed nc
need
even for the little diplomacy which elsewhere is usual!]
required to save oneself this chance of catching cold.
Ujat is a large village, its straggling dwellings surroundec
by
grape-gardens, for which it is famous. The dried grape
and
currants of the place are said to find their way as far
the
markets of Aksu, Kashgar, and Turfan. The vines are
trained, as Ihrougliout Chinese Turkestan, along low fencesj
ranged in parallel lines. The work of covering up the stei
with earth for the winter was just proceeding. The peo])le oi
IN A SACRED CA\^ 229
Ujat seem to have retained for a long time after the accep-
tance of Islam the reputation of being weak in the faith and
addicted to heretical ways. I wonder whether the extensive
cultivation of the \ine has something to do with this.
My local enquiries and the arrival of a long-expected mail
from Kashgar, which brought me home and Indian letters of a
whole month and required early disposal, helped to detain me
at Ujat. But on the 15th of November I marched back to
Khotan by the shortest route, crossing the bleak pebble
'
Sai
'
that stretches from the Kohmari ridge to the southern edge
of the cultivated area near the village of Kosa. I was sur-
prised to find how rapidly the fertile tract towards the city
liad assumed its winter aspect. Tlte long alleys of poplars
and willows stood leafless
;
the same storm that put a stop to
our survey work on the mountains had brushed away the
bright autumn colours which greeted me on my first descent
to the Kara-kash.
At Khotan it became necessary to make a short halt in
order to give to my men and ponies the rest they required
after the month of fatiguing marches. I also wanted time for
the examination of the antiques which had found their way
from various localities into the hands of the agents sent out
on my behalf after my first visit. The small parties despatched
to ancient sites in the desert also turned up with their spoO
during my week's stay. The party which had gone out under
the guidance of Turdi, an old, and as experience showed,
reliable
"
treasure-seeker
"
from a village of the Yurung-kash
canton, had visited the most distant of the locally known sites,
called by them Dandan-Uiliq ("the houses with ivory").
Among the specimens brought back by them I foimd to my
great satisfaction several pieces of fresco inscribed with Indian
Brahmi characters, fragments of stucco relievos representing
objects of Buddhist worship, and also a small but imdoubtedly
genuine piece of a paper document in cursive Central-Asian
Brahmi.
It turned out, on further examination of the
"
treasme-
seekers,"
that the ruins from which they had unearthed these
remains,
and ^which they described as reached after nine
230 ANTIQUAEIAN PEEPARATIONS.
to ten marches north-eastwards
through the desert, were apparently
identical with the site which Dr.
Hedin had seen on his memorable
march to the Keriya Darya, and
which is spoken of in the narrative
of his travels as the "ancient city
Taklamakan." He had reached it
by another route from Tawakkel
on the northern edge of the oasis.
So Pan-Darin, whom I informed of
the results of this reconnaissance,
sent word to the Beg of Tawakkel
to produce the two hunters who
had guided Dr. Hedin on his jour-
ney. On November 20 Ahmad
Merghen and Kasim Akhun, the
men I wished to examine, were
duly produced by the Beg himself,
who had brought them to Khotan
in person. Their examination in
the presence of Turdi, the leader
of my pioneer party, left no doubt
as to the identity of Dandan-Uiliq.
I was thus able to arrange definitely
the programme of my tour for the
exploration of this site, which in view of the specimens secured
by Turdi seemed the best place for commencing systematic
excavations.
Immediately after my return I visited my kind friend the|
Amban, and thanked him for the thorough-going help by]
which he had made my survey in the mountains possible. On;
that occasion I invoked again the evidence of the great
'
Tang-
*
Seng,' in order to explain to
Pan-Darin the object of my
desert journey. When after two days he returned the visit!
I was able to show him the finds brought in by Turdi. Soi
Pan-Darin by ocular inspection became convinced that I hadi
a good guide in the famous old pilgrim, and promised to doj
TURDI,
"
TREASURE SEEKER."
ANTIQUES BROUGHT BY TURDI. 231
all he could to further my explorations. I thought that I could
not more fittingly express my gratitude than by wishing that
the blessed spirit of Hiuen-Tsiang himself might reward the
Amban for the assistance he was rendering me. Niaz, the
interpreter, managed to reproduce this pious compliment
better than I had expected ; for the Amban answered it by
asking quite seriously whether I believed in the continued
existence of
'
Tang-Seng's
'
soul ! It seemed indeed that in
the memory of Chinese Buddhists Hiuen-Tsiang lives like a
glorified Arhat or Bodhisattva. If so, Indian archaeologists
would be still better justified in proclaiming him as their own
patron saint.
I had pitched my tent again in the garden of dear old
Akhun Beg, my former host. But though the place gave the
desired privacy it offered no protection whatever against the
increasing cold. Tokhta Akhun's house seemed too gloomy
and close after the long journey in the free mountain air. So
I preferred to put up for the time with the cold and to stick to
my little tent outside. Many repairs of outfit, saddlery, etc.,
required my attention too
;
for the terribly rough tracks of the
"
Mountains of Darkness
"
and the wily ways of the 'yaks had
caused damage of all kinds. So the saddler, blacksmith, and
tailor were kept busy under my eyes. Vendors of antiques,
bringing seals, coins, old pottery, and similar small objects,
mostly from Yotkan, frequently presented themselves. But
of the
"
old books
"
none were offered. It seemed as if the
particular "treasure-seeker" to whom I had reason to trace
them, credited me with a more inquisitorial turn of mind than
'
was convenient for himand his factory.
But my days at Khotan were taken up not only with these
avocations. There had been since I returned an increasing
rush of people seeking benefit from my medicine case. Patients
from among the local Begs and the Chinese officials could not
be denied, and though my
"
Tabloids
"
could scarcely effect
the wonderful cures expected by these visitors, they evidently
spread my fame as a
"
Hakim
"
throughout the district. From
what I saw and heard Khotan seems to be a hotbed^ofjdiseases
of all kinds. Numerous
''
cases " of a sickening type were
232 ANTIQUARIAN PKEPAEATIONS.
daily brought to me, though rarely was I able to administer
remedies from which I could expect any real good. A medical
man would find here a splendid field of work, but I doubt whether
his fees would suffice even to balance the charities expected
by a large portion of those seeking relief. Chinese mendicants
and'loafers were frequent among mypatients, and their condition
KHOTANESE
WAITING FOR
MEDICINES.
fully justified the requests for a present which were invariably
made after I had attended to their ailments. I wondered
whether the Chinese officials realized how detrimental to tiieir
regime must be the ])resence of large numbers of these destitute
compatriots, living on charity and, no doubt, occasional
loot.
It was manifest that my desert campaign would necessitate
a prolonged absence from the oasis. Accordingly I decided to
MEDICAL FUNCTIONS. 233
make, previous to my start, a thorough examination of old
localities within the oasis itself, with a view to settling its
ancient topography. At the same time I decided to send out
Ram Singh independently for a survey of the high range east
of the Kuen-luen Peak No.
5,
by which the gap could be filled
that was left between our recent survey and the tract explored
by Capt. Deasy about Polu. On completing this task within
about a month Ram Singh was to march to Keriya and then
join me eventually at Dandan-Uiliq.
On the 23rd of November we both left Khotan. Our way
was in common as far as Jamada, the village on the Yurung-
kash which I had passed before when marching to Karanghu-
tagh. I halted here for the night and received a cheerful welcome
from Wang-Daloi, a Chinese acquaintance of my pre\nous visit.
For the last ten years the little Chinaman had lived there,
trading in jade, which is washed from the Yurung-kash bed in
the neighbourhood. He seemed to have ventured occasionally
on speculative jade mining too, but fortune had never shown
him favour
;
for my interpreter told me that he was still a
long way from the sum that might take him back to Peking,
apparently the life ambition of this exile. I found in Wang-
Daloi an intelligent guide to the old sites which extend from
Jamada to the south along the left river-bank, and also genial
company, as he talked a little Turki. Next morning I passed
over the eroded old site known simply by the general designation
of
'
Tati,' forming an area of about a square mile covered
with fragments of pottery. Chinese coins up to the time of
the Tang dynasty are also found, but of structural remains
j
there was no trace.
^
Some six miles beyond we entered the region of the jade-
diggings. On the fiat plain, from half a mile to one mile broad,
which extends between the left bank of the river and a gently
sloping ridge of gravel westwards, the precious stone is found
among the beds of rubble deposited by the river at earher
periods.
Jade is the produce that has made Khotan famous
aU over the east since ancient times. In China it has ever
-
been
valued more than anywhere else, and most of the
information which the Annals of the Celestial Empire give
234 ANTIQUARIAN PREPAEATIONS.
about old Khotan, we owe mainly to the interest attaching to
its jade.
It was therefore with a good deal of interest that I examined
the burrows crossing the barren plain in all directions. For the
first mile or two they seemed to have been deserted long ago, as
sand had partly filled the great hollows. But higher up we came
upon diggings of more recent date not far from the old site
known as Chalmakazan. A vast quantity of pottery fragments,
mixed here and there with bits of broken glass and slag, strews
the plain for about a mile and a half, from the river to the foot
of the ridge. In the middle of this area a low mound, covered
with large stones from the river bed, attracted my notice. Its
round shape suggested a Stupa, and a closer examination proved
this to be true. Unfortunately, others before me had guessed
the nature of the structure, and a large trench run down into
the very centre of the mound showed that
"
treasure-seekers
"
had been at work. The mound in its present condition has a
diameter of about ninety-eight feet, and rises about fifteen
feet above the ground. From the excavation made, it could be
seen that it was constructed of closely packed layers of rough
stones as a base, with a circular wall of similar material above
it. A kind of well in the centre filled with loose earth probably
contained the relic deposit.
There can be little doubt that the old settlement indicated by
these remains was connected with the jade mining of the
immediate neighbourhood. On the southern edge of the site
the jade pits are still worked. For a mile and a half we had to
thread our way between them before reaching the little miners'
camp of Sirik-Toghrak, where I pitched my tent. The pits
vary greatly in size and shape. Usually, a square or oblong
cutting is made through the layer of gravel and river sand.
At a depth from ten feet downwards strata of rubble are
reached, and in these search is made for the pieces of jade that
the river once washed down. Finds of great value occur very
rarely
;
but there is always the chance of sudden wealth, and
this suffices to attract at all times
'
Bais,' i.e., small capitalists
from Khotan and other Turkestan towns. They engage ])arties
of labourers, ten to thirty strong, from among the jioorcst
JADE PITS OF CHALMAKAZAN. 235
of the agricultural class, and set them to work on a digging of
proportionate size. The men receive food, clothing, and six
Khotan Tangas (say two Rupees) as monthly pay. They have
no share in the jade finds, but get extra rewards in case of
special profits. According to Wang-Daloi's testimony, many never
see any return for the money they have sunk in these mining
ventures. Yet from time to time great hits are made. A Kashgar
JADE PIX WIIH DIGGRi, ^'EAR DEBOUCHCiiii OF VURL'Nu-KASH.
Bai, whom I found at one of the diggings superintending his
twenty men. acknowledged that dming the last three years
he had cleared a hundred Yambus of silver (say Rs.
13,000)
worth of jade at an expense of some thirty Yambus.
Though the Chinese administration exercises no control
whatsoever over the jade mining,
"
claims
"
once opened are
scrupulously respected by other prospectors. I saw diggings
which had been left partially unexploited for many years
;
yet I
was assured that the right of the original workers would never
236 ANTIQUARIAN PREPARATIONS.
be disputed. None of the diggings went to a greater depth
than twenty feet from the surface
;
lower down, I assume,
the water from the river would probably percolate and stop
the work. The flat deposits along the river banks for a day's
journey up the valley, up to the point where the latter becomes
a narrow gorge, are visited by jade-diggers. But the work is
carried on only intermittently and by small parties at the
various points which bear the general designation of
*
Kumat.'
Now in the winter months only about two hundred men were
engaged in mining, and even in the summer, when the privations
of life in this barren region are less, the number probably is
not more than doubled.
Quite distinct from this jade-mining, the ancient industry
of
"
fishing" for jade in the river bed after the summer floods
still continues all along the valley above Jamada, just as described
in the old Chinese chronicles. No capital is wanted for this
kind of search
;
so annually for a short period it attracts a
large number of the poorer agriculturists of the oasis, who look
to it as a kind of lottery. Very few find their labours rewarded,
but the hope of turning up a valuable piece of jade among the
rubble is as strong now among the poor of Khotan as it has
been for many centuries.
The Annals of the old Chinese dynasties, from the Han period
downwards, contain many curious data and anecdotes about
the jade
('
yii
')
which made the little kingdom of Yii-t'ien
or Khotan famous in the Celestial Empire. Abel Remusat,
the Sinologist, collected and translated many of these notices
in his Histoire de la ville de Khotan (Paris :
1820),
and it was
a satisfaction to me to read this earliest contribution to the
European literature on Khotan near the very pits which furnish
the precious stone so learnedly discussed in it.
237
CHAPTER XVI.
YOTKAX, THE SITE OF THE ANCIENT CAPITAL.
My
march on the 25th of November to Yotkan, the site of the
old Khotan capital, took me over ground that I had partly seen
before, but the day did not close without a novel, though some-
what annoying, experience. Coming from the south, I had,
within a couple of miles from my destination, passed two deep
ravines, or
'
Yars ' as they are called, cut into the loess beds
by the action of flood water. Though the banks were steep,
the ponies found no difficulty in crossing, and I did not give
a thought to the question how the camels with the baggage
would fare at these obstacles. I reached Yotkan, to which I
had already paid a preliminary visit in October, about sunset,
and selected a suitable ground for my tent close to the Yiiz-
bashi's house, overlooking the area where the excavations of
treasure-seekers have laid bare the soil of the ancient capital.
The best room of the well-to-do villager was quite a cosy place,
with its carpets and coloured Khotan felts, and with a cheerful
log fire burning in the little fire-place. So the time of waiting
for the arrival of the baggage passed quickly at first. The
Yiizbashi's little red-cheeked son kept me company and amused
'
Yolchi Beg,' my faithful follower.
At last, long after it had got pitch-dark outside, one of the
camel men arrivednot with the eagerly expected animals, but
with the news that they had stiick fast at the bottom of the first
ravine and could not be got to move further. So a rescue party
was despatched under the orders of the village headman. I have
reason to suspect that easy-going young man did not move far
238 ANCIENT SITE OF YOTKAN.
beyond the neighbours' houses, but left the task to some myr-
midons of his who managed to mistake the place and never came
to help the belated party. When another hour had passed,
Islam Beg and Tila Bai were sent out into the night. But it
was not until about lo p.m. that the unfortunate camels turned
up at last. An attempt had been made to send on one of the
animals that carried my tent and bedding, with the result that
it slipped in crossing a canal and gave a thorough ducking to
its load. When dragged out of its bath, this camel with the rest
had to be taken by a great detour round the heads of the two
ravines. The late arrival of the party was thus accounted for
;
but the explanation did not exactly console me for a half-wet
tent, and bedding that had first to be dried. It was nearly
midnight when dinner appeared, and some of the rugs had been
made fit for use.
The ravines which proved such an obstacle to my clumsy
camels had little claim to my regard. And yet my archaeo-
logical conscience felt grateful to them
;
for without the forma-
tion of one of them, known as the
'
Yotkan Yar,' that has cut
through the fields of the village of Yotkan, the remains of the
old Khotan capital might have been left buried for ages to come.
From the statements of the old villagers which I tested with care
in the course of my stay, I ascertained that no finds of any kind
indicating that an ancient site was buried here below the ground
had been made, until the time of Niaz Hakim Beg, the first
governor under Yaqub Beg. Two or three years after his ap-
pointment, which took place about 1866, the small canal con-
veying water from the Kara-kash River for the irrigation of the
Yotkan fields began -to cut for itself a deeper bed in the soft
loess, that is, to turn into a
'
Yar.' This is the origin of the
ravine which begins about one and a half miles to the west of
Yotkan at the village of Chalbash, and joins the Yars of Kashe
about a mile to the east of the site presently to be noticed. A
small marshy depression
('
kul
')
formed a little to the east of
Khalche, as that portion of Yotkan is called which lies to the
north of the excavated area, and there the villagers accidentally
came across little bits of gold amidst old pottery and other
petty debris. The latter objects possessed, of course, no interest
WASHING FOR GOLD REMAINS. 239
for them
;
but the gold naturally excited the cupidity of the
villagers, many of whom had, like the rest of the poorer agri-
cultural population, tried their luck
"
prospecting
"
for jade and
gold in the river beds. So they set to wash the soil near the
incipient Yar, and the proceeds were so rich that they came to
the governor's knowledge.
Niaz Hakim Beg was an administrator of considerable enter-
NORTH-WEST CORNER OF EXCAVATED AREA AT YOTKAN, WITH ENTRANCE
TO
'
YAR.'
prise. He sent to Yotkan large parties of diggers from Kara-
kash tovm whom he employed like the men working in the jade
pits. The owners of the fields which were gradually cut away
by these
"
washings," received compensation. Subsequently
the excavations were continued by private enterprise, the usual
arrangement being that the owners of the soil and the diggers
-hare the proceeds equally. The earth excavated from the
banks has to be washed, just like the river deposit. The larger
240 ANCIENT SITE OF YOTKAN.
supply of water needed for this purpose caused the Yotkan canal
to cut its bed deeper and deeper and to form the extant Yar,
the bottom of which is from twenty to thirty feet below the
ground level. Finally the canal had to be diverted to a higher
level, but springs came to the surface at the bottom of the ravine,
and these account for the swampy condition of the excavated
area. In the recollection of old villagers the land of Yotkan
was everywhere a level flat ; there were no springs or swampy
groundnor any knowledge or tradition of the
"
old city
"
below.
Former travellers, who paid to Yotkan only a flying visit,
have spoken of "the frightful ravages in the soil" and attri-
buted them to some extraordinary flood catastrophes of which,
it is true, they were unable to trace any recollection. But in
reality the great extent of the excavated area which forms, as
my careful survey showed, an irregular oblong of more than half
a square mile, is almost exclusively due to systematic digging
and washing for gold, as it still continues to this day on the north
and west sides. The banks there are yielding a small but
"
pay-
ing
"
quantity of gold, and in recent years antiques, such as
ornamented fragments of pottery, engraved stones, and coins
have come to be counted as a kind of secondary product. The
gold is usually found in tiny flakes of leaf-gold, of which I was
able to secure samples. It is easily distinguished by the villagers
from the gold-dust
(*
kepek-altiin
')
washed from the river-beds.
No gold coins or solid ornaments of gold and silver are admitted
to have been found. But I have my doubts on this point, as
the villagers or miners would have reason to be reticent about
such finds. In any case it is acknowledged that during the first
j
years and near the original spot the workings yielded rich quan-
tities of gold. I myself subsequently purchased at Yotkan
tiny figure of solid gold of excellent workmanship, representing
monkey, that had been found during the year's washings. Larger
articles of this kind are doubtless melted down speedily after]
discovery.
It seems at first difficult to account for the prevalence of golc
in the form described and over sc large an area. But the use
of leaf-gold on an extensive scale in the decoration of statues]
ANCIENT COINS AND POTTERY. 241
and buildings offers a probable explanation. From the detailed
description which the earher Chinese pilgrim Fa-hien gives of
the splendid Buddhist temples and monasteries he saw on his
c
ANTIQUES FROM YOTKAN.
A Terra-cotta head. B
Relievo in ivory. C Fragment
of
relievo in
stone showing seated Buddha ; also atuietu Khotan and Chinese
coins. D Terra-cotta vase with monkey-shaped handle. E Piece
of
decorated Txtse. (Scale one-half of originals.)
visit to Khotan (circ.
400
a.d.), it is certain that not only images
but many parts of sacred buildings
were richly overiaid \Wth
leaf-gold.
Much of this must have fallen off and mingled with
16
242 ANCIENT SITE OF YOTKAN.
the dust when these structures crumbled away, not to be re-
covered until the soil could be washed by the method now
followed;
The stratum from which this gold is obtained consists of
decomposed rubbish
and humus, in which
are embedded frag-
ments of ancient
pottery, plain or
ornamented, bones of
animals, pieces of
much decayed wood,
and ashes, all indica-
tions that we have
here the debris that
accumulates on a site
occupied by buildings
for centuries. The
copper coins, which
are found plentifully,
range from the bilin-
gual pieces of the
indigenous rulers,
showing Chinese
characters as well as
early Indian legends
in Kharoshthi, struck
about the commence-
ment of our era,
to the square-holed
issues of the Tang dy-
nasty
(618-907
A.D.)-
The stratum which
represents the deposits of these and possibly also of earlier cen-
turies, shows a uniform brownish colour, but varies in thickness.
On the south and west it is on the average from five to eight feet
deep. But on the north of the excavated area, the banks worked
immediately below the houses of Khalche, where the proceeds
I I
!
1: \
'
! ; \ !h,l IxIN I'S FROM YOTKAN.
(Monkeys playing nnisical instruments, eating,
&c. Scale two-thirds of original.)
SILT OVER CULTURE-STRATA. 243
in antiques, such as terra-cotta figurines, seals, etc., are richest,
show a
"
culture-stratum
"
thirteen to fourteen feet thick. It
is evident that this varying depth is due to the different length
of the periods during which particular localities were occupied,
and to the different character of the uses to which they had been
put. The frequency of pottery fragments and of bones also
varies at different points.
But in one respect all portions of the
"
culture-strata
"
ex-
posed show a regrettable uniformity
;
nowhere did I come upon
traces of remains of buildings, nor could I hear of such having
been found during previous excavations.
\
This is easily
accounted for by the fact that, owing to the total absence of
suitable stone, sun-dried bricks and clay supplemented by timber
must have been in old days, just as now, the only obtainable
materials for the construction of houses in the Khotan region.
Whatever of the mud walls of buildings had not crumbled into
dust, was bound to decay completely in the course of the cen-
turies during which the site was taken up for cultivation and
the soil kept constantly moist by the percolation of irrigation
water. The same fate overtook whatever of the wood once
contained in the buildings had not been extracted and utilised
by successive occupiers of the soil. It might have been different
if the old town had been overwhelmed by some sudden catas-
trophe and its site left deserted. Then we should expect to
find under the ruins the original materials preserved in a recog-
nisable form. But there is nothing to support the assumption
of such a catastrophe.
The strata containing the old remains are everywhere covered
by a considerable layer of alluvium from nine to twenty feet
thick at various points. This layer, which by its light colour is
easily distinguished from the "culture-strata" below and is
absolutely free from remains indicating subsequent occupation
of the site, interested me greatly. Some of the earUer European
visitors to Yotkan have hazarded the assumption that the thick
cover of earth under which the relics of the old town are hidden
was due to a great flood, and they accordingly attributed its
1
destruction
to this supposed catastrophe. But a few hours'
careful
examination of the excavated banks sufiiced to dispel
16*
244 ANCIENT SITE OF YOTKAN.
such a notion once for all. Nowhere did I find the slightest
trace of that stratification in the soil which such a flood or
series of floods would necessarily have left behind. At every
point the earth immediately above the
"
culture-strata
"
proved
exactly the same in substance and colour as that which is to-day
turned up by the plough of the Yotkan cultivator.
What, then, is the explanation of this deep cover under
which the remains of this old town have rested ? I think it is
not far to seek. Cultivation in Khotan, as everywhere else in
Turkestan, demands constant and ample irrigation
;
and as the
river from which the water for the Yotkan fields is drawn in
the spring and summer carries down enormous quantities of
disintegrated soil from the mountains, the accumulation of silt
over the fields on which the earth thus suspended is ultimately
deposited, must be comparatively rapid. Thus the level of the
cultivated portions of the oasis is bound to rise steadily
;
and
considering how near these lands are to the region where the
river collects most of this silt on its passage through the outer
ranges, the thickness of the deposit left during a thousand years
can by no means surprise us.
Observations I had occasion to make again and again after
my first visit to Yotkan fully supported this explanation. Every-
where in the oasis I noticed that the main roads were sunk
considerably below the surrounding level where they pass through
cultivated land, while elsewhere, on waste or within the village
areas, they kept flush with the adjoining ground. This low
position of the roads is so uniformly observable and so marked
that it is impossible not to seek for a natural cause. And none
I could think of seemed more probable than that the level of
the fields is constantly rising by irrigation, while that of the
roads cannot undergo any marked variation. This
observation
led me to notice an equally characteristic factthe low jiosition
of all the old cemeteries that are surrounded by fields.
Cenit
teries of^^any age are easily distinguished by their extendiiiji
around some Mazar or shrine, and in their case I invariably found
a ground-level considerably below that of the neighbouring
fields. This curious fact becomes easily intelligible if we
re-
member that the fields are continually recefving a deposit
oi
POSITION OF SA-MO-JOH CONVENT. 245
silt from irrigation, while the cemeteries are naturally kept clear
of water and consequently of this accretion.
The Yar which passes through Yotkan from west to east,
and the excavations of the gold washers to the south of it, enable
us to form some idea as to the position and extent of the old
town. The banks of the Yar cease to jield any remains about
200 yards below the houses of Khalche. Accordingly, digging
has stopped there. In the south the diggings near the portion
of Yotkan known as Allama have been discontinued, as the
ground did not yield the coveted gold in paying quantities. It
is on the banks to the west and north-west that the work of
washing the soil still continues vigorously, and it is under the
fields lying in that direction that the remaining parts of the old
town are likely to have been situated. The Yars which intersect
the ground to the south and east of Yotkan nowhere cut through
layers containing old remains. The negative evidence thus fur-
nished excludes the idea of the town having ever extended in
those directions.
There can be no doubt that the site discovered under the fields
of Yotkan is that of the old capital of Khotan, as already sug-
gested by M. Grenard. The proof, however, does not lie in an
alleged tradition of the villagers (this could only be a very modern
growth if it really existed), but in the exact agreement of the
site with the topographical indications furnished by the early
Chinese Annals, and in the ease with which I was able to identify
from this starting-point the positions assigned by Hiuen-Tsiang's
narrative to the most prominent Buddhist shrines he visited
in the vicinity of the capital.
On the morning of the 28th of November I started on a survey
of the villages to the west of Yotkan, in order to trace, if possible,
the positions of these sacred places. Nearest among them was
the Stupa and convent of
'
Sa-mo-joh
'
which the pilgrim visited
it a distance of five or six li (a little over a mile) to the west of
he city. It was founded in honour of an Arhat who had by
/arious miracles won the special worship of one of the first
luddhist
kings of the country. Under its Stupa, which was a
undred feet high, a great collection of sacred relics from
uddha's body had been deposited. Fa-hien also, two and a
246 ANCIENT SITE OF YOTKAN.
half centuries earlier, had seen this monastery, and describes
"
the magnificent and very beautiful hall of Buddha
"
that rose
behind its Stupa.
Judging from what previous experience has
taught me of the fate which has overtaken all ancient structures
within the cultivated area of the oasis, I did not expect to find
remains of what was undoubtedly only a pile of sun-dried bricks
doomed to rapid decay. All the more delighted was I when
among the villages westwards I heard the name of Somiya
mentioned. Other phonetic analogies prove that this repre-
sents the direct derivative of the ancient local name which is
intended by the Chinese transcription of
'
Sa-mo-joh,' and to
the evidence of the name there was soon added topographical
confirmation.
Leaving the excavated area of the ancient city at its north-
west corner, I reached first the hamlet of Eskente half a mile
to the west. There I was told of a
'
Dobe ' or mound that
exists near the cemetery of Somiya. The latter place I found to
be situated only three-fourths of a mile further west, and to
consist of some thirty scattered dwellings. I went at once to
the local Mazar, which is surrounded by an extensive cemetery,
and on asking for the
'
Dobe
'
was taken to a field adjoining its
north-eastern corner. A little low mound, rising scarcely five
feet above the surrounding ground, is respected by the
villagers
with a kind of superstitious fear, though it shares in no orthodox
way the sacred character of the neighbouring Mazar and ceme-
tery. I soon had the oldest men of the village summoned to
the spot, and in what they told me of the mound we may, I
think, yet trace the last lingering recollection of the ancient
shrine that has left its name to Somiya. Shami Sope, a withered
old man of about ninety, had heard from his father and grand-
father, who had both died at a great age, that the little mound
had ever been respected by the folk of Somiya as a hallowed s])ot
not to be touched by the plough-share. Some unknown saint
is supposed to have sat at the spot, and evil would befall
thost
who should touch the ground. The name of the saint is forgotten
and the villagers would not assert whether he rests under tin
mound or not. But the people of Somiya never pass withoui
saying a prayer, and according to the testimony of Shami
Sojh
OLD LOCAL WORSHIP.
247
and his forbears, they have clung to this custom for the last two
centuries.
I take it as a sign of the antiquity of the tradition that no
name is assigned to the saint whose memory lingers about the
'
Dobe,' whereas the names of the three Mullahs who are sup-
posed to sanctify the Mazar of the \allage are currently known
to young and old. Nobody seemed to know of any other spot
similarly surrounded with superstitious awe in the neighbour-
hood. Considering the concordant evidence of the name and
position of Somiya, I think it highly probable that the worship
of this nameless mound is the last trace left of the
'
Sa-mo-joh
'
Stupa of Buddhist days. And if this assumption be correct, we
have here another proof of the tenacity of local worship which in
Khotan, as elsewhere in the East, has survived all changes of
creed.
The day's search enabled me to identify in all probability yet
another sacred site mentioned by the Chinese pilgrims. Hiuen-
Tsiang saw at a distance of ten li (two miles) to the south-west
L
V
248 ANCIENT SITE OF YOTKAN.
of the capital the monastery of
'
Ti-kia-po-fo-na,' which was
distinguished by the possession of a miraculous statue of
Buddha. The name in this case can no longer be traced, but
exactly in the direction and at the distance indicated there
lies the popular Ziarat of
'
Bowa-Kambar
'
visited by people
from all parts of the Khotan district. I found it to consist of a
large square cemetery enclosing the high mud tomb of the saint,
who is supposed to have acquired holiness as the groom of
'
Ali
Padshah,' The level of the cemetery lies some twelve feet
below the surrounding fieldsa certain indication of its anti-
quity according to my previously detailed observations. A grove
of fine old trees faces the eastern entrance, and a row of booths
testifies to the popularity of the fairs which take place here at
the time of pilgrimages.
It was dark when I returned from Bowa-Kambar, else I
should have paid another visit to the still more popular shrine
of Imam Musa Kasim at Kosa, which I had already passed on
my way from Ujat. Its position due south of Yotkan makes
me suspect that it has taken the place of the Virochana-Sangha-
rama which was famous in the days of Hiuen-Tsiang as one
of the earliest sanctuaries of Buddhism in Khotan. Its distance,
a little over three miles from Yotkan, is somewhat in excess of
the ten li south of the capital which the pilgrim indicates as its
position. But then we do not exactly know the extent of the
old city, and in any case there is no shrine of any note due south
of Yotkan that comes nearer to the distance indicated.
On the 29th of November I left Yotkan to return to Khotan
town, where the preparations for my desert journey were now
urgently calling me. It was a misty cold morning as I bade
good-bye to my host the Yiizbashi and rode along the Yotkan
Yar eastwards. About two miles from the village I crossed by
a bridge the fairly deep stream formed by the united waters of
the Yars of Yotkan and Kashe, and on the other bank of the
ravine reached the lands of Halalbagh, a collection of large
hamlets which I was anxious to see once more, as a local tradition
connects the site with the ine-Muhammadan rulers of the
country. Close to the central hamlet there stretches a marsh,
known as Aiding-kul, coverii)g about a square mile. It is over-
I
THE AIDING-KUL MARSH. 249
grown with reeds and fed by copious springs which form quite
a little stream at the northern end where the marsh drains
towards the Yurung-kash.
Islam Beg secured me here a very intelligent guide in the
person of Ibrahim Mullah, a man well known for his learning
and piety. Though eighty-six years old at the time of my visit,
he was still quite active. His comfortable embonpoint and his
showy silk dress well-lined with fur showed plainly that, despite
Koran and pilgrimages, he had not neglected the good things of
this world. Ibrahim Mullah owns Turki
*
Taskiras
'
of the
varous Imams who are worshipped at the most ix)pular of
Khotan Mazars, and soon showed me in them chapter and verse
for his assertion that it was at Halalbagh that there once stood
the city of the
'
Khalkhal-i Chin-u-Machin,' the legendary
heathen ruler of Khotan. According to the popular tradition
recorded in these texts, the four Imams whose blessed bodies
now rest in a famous Mazar at Hasha, killed this opponent of
Islam, and his city became a waste. The shrine of Kum-i-
Shahidan, about half a mile to the west of the marsh, is supposed
to mark the spot where three hundred and sixty faithful fol-
lowers of the Imams found martyrdom in the final struggle.
According to Ibrahim Mullah, Mirza Abu-Bakr, the ruler of
Kashgar and Khotan in the early part of the sixteenth century,
had the old site excavated for the sake of its hidden treasures.
He brought river-water to the place to enable his workmen
to wash the soil

just as is now done at Yotkanand in the

hollow left by his diggings there formed the marsh of Halalbagh.
.
No old remains of any kind are now found, and it is thus diffi-
cult to judge whether there is any historical foundation for the
story. Mirza Abu-Bakr, about whose doings we are well in-
formed by the Tarikh-i-Rashidi, the interesting chronicle of
Mirza Haidar, his nephew, certainly carried on treasure-seeking
operations on a great scale at various old sites in his dominions.
But whether Halalbagh was really among the places he exploited,
or whether his reputation alone induced the local literati to con-
nect with his name the supposed origin of the Aiding-kul marsh,
can scarcely be decided without tracing old remains at the site or
earlier evidence of the tradition. The mound called
'
Nagara-
250 ANCIENT SITE OF YOTKAN.
khana
'
("
the hall of kettle-drums
')
which rises to a height
of about twenty-seven feet close to the southern edge of the
marsh, is popularly supposed to represent a remnant of the
ancient city wall. But on close examination it proved to be a
natural bank of loess, without a trace of brickwork or other
mark of artificial construction.
The rigour of winter was now setting in rapidly. So I was
glad to avail myself during the short halt at Khotan, which
the final preparations for the desert journey demanded, of the
shelter offered by Tokhta Akhun's suburban residence. In its
dimly lighted but tolerably warm rooms I was hard at work
writing up accounts of my geographical and antiquarian observa
-
tions for despatch to Europe, and carefully sorting and re-packing
the baggage. In order to keep the camels, on which we should
have to depend entirely for the difficult marches through the
sands, as lightly laden as possible, I decided to leave behind in
charge of Badruddin, the Afghan Aksakal, a depot of all stores
and materials not immediately needed. The elimination was
no easy matter. On the one hand it was impossible to estimate
the length of time during which my explorations would keep
me away from Khotan
;
on the other it was clear that in the
inhospitable regions in which we were to pass the winter, any
deficiency in the necessary supplies and equipment might have
a very serious effect on our health and thus hamper my move-
ments. It was largely through the care bestowed on transport
and supplies, that I was subsequently able to carry my opera-
tions so much farther from the Khotan base than originally
anticipated.
CAMELS STARTING FOR .DANDAN-UILIQ.
CHAPTER XVII.
TO THE RUINS OF DANDAN-UILIQ.
The morning of the 7th December, a misty and bitterly cold
day, saw our start for the winter campaign in the desert. My
goal was Dandan-Uiliq, the ancient site I had decided upon
for my first explorations. To reach it I chose the route via
Tawakkel
;
for though longer than the track leading straight
into the desert north-east of Khotan which Turdi, my
"
treasure-
seeking
"
guide, was in the habit of follo\N'ing, it somewhat
reduced the extent of actual desert-marching with its inseparable
privations for men and animals. The first day brought me,
at Yangi-arik, to the edge of the cultivated area north of Khotan
town. The next two days were passed in dreary marches
along the barren left bank of the Yurung-kash, where there
was nothing to be seen but sand-dunes to the west and reed-
covered strips near the winding course of the river.
It was getting dark when we crossed on the evening of the
third day to the right bank and approached the southern end of
the Tawakkel oasis. It was formed some sixty years ago by
the construction of an irrigation channel, which takes off the
river waters a few miles further south. It has since developed
into a prosperous settlement estimated at some thousand
households. Its Beg, instructed in advance from Pan-Darin's
252 TO THE KUINS OF DANDAN-UILIQ.
Yamen, had awaited me at the crossing, and now escorted
me with an imposing array of followers to the southernmost
hamlet. The big bonfires which lit up our way, and the pre-
valence of wood in the construction of the houses indicated
the proximity of the forest belt which accompanies the Khotan
River on its course through the desert and furnishes a plentiful
supply of wood to this outlying colony.
On the following day I moved my camp to the Beg's house
at Atbashi, some six miles further north, where the arrange-
ments were to be completed for the party of labourers I wished
to take along as well as for our supplies. In view of the
observations already detailed as to the rise of the ground level
in the old cultivated area of Khotan, I was interested to note
that in this comparatively recent oasis the roads and waste
spaces lay nowhere more than about one foot below the level
of the neighbouring fields. It was evident that the period of
irrigation and consequent silt deposit had been too short here to
permit of any appreciable rise in the level of the fields. Still
less was I surprised to hear that the area of the colony might
be greatly extended towards the desert by the construction
of additional irrigation channels. The abundant sujjply of
water which the river carries down during the spring and summer
months might bring fertility to large tracts now covered by
low dunes. But here, as elsewhere along the southern edge of
the great Turkestan desert, there is no surplus of population
available for such extended cultivation, nor an administration
capable of undertaking fresh irrigation works on a large scale.
Thanks to the stringent instructions issued by Pan-Darin, I
was able to collect at Atbashi a party of thirty labourers for my
intended excavations, together with four weeks' food sujiply.
Owing to superstitious fears and in view of the exj^ected rigours
of the winter, the cultivators were naturally reluctant to venture
so far into the desert, though they appreciated the pay offered,
I
J Miskals per diem, which was more than twice the average
wages for unskilled labour. Fortunately, the Amban's authority
was not to be denied; and there were also the two Tawakkel
hunters, Ahmad Merghen and Kasim Akhun, to ins[)ire
confidence.
PREPAEATIONS AT TAWAKKEI.. 253
I had already from Khotan secured their services as guides,
and soon found them most useful in looking after the labourers.
They were indeed
splendid men, in-
ured to all hardships
by their roving life
in the desert and
river jungle, and by
their experience in-
telligent far beyond
the horizon of the
villagers. They did
their best to con
vince the more
fainthearted that
this season, when
neither sand-storms
nor thirst need be
apprehended, w a s
in reality the safest
for the work I had
in view; that
travelling in so great
a party they had
nothing to fear from
the*
J
inns' or demons
of the desert ; and
that plenty of dead
wood would be
forthcoming to keep them from getting frozen to death.
I on my part took care to select the physically fittest from
among the men brought before me, and made the respective
village headmen responsible for their being supplied with all
needful warm clothing and food sufficient to last for four weeks.
The liberal cash advance I paid myself into the hands of each
of my
"
Levies
"
facilitated these domestic preparations. In
order to provide for professional help far away in the desert,
I arranged to include among the labourers a young cultivate-
AHMAD MEKGHKN AND KASIM AKHLN, OF TAWAKKEL.
254 TO THE RUINS OF DANDAN-UILIQ.
who had been to a Mosque school and had acquired the art
of writing Turki, not according to any high standard of ortho-
graphy, it is true, but still legibly. Another was used to practise
tailoring in his spare hours, while a third was proficient in
leather work and could look after the men's boots. Each
man had to bring his
'
Ketman,' the hoe in common use
throughout Turkestan, which proved an excellent implement
for excavation work in the sand. Steel shovels of German
make I had brought along from Kashgar
;
but I soon found
that, except where there was a risk of causing damage to buried
remains, the
'
Ketman
'
to which the men were accustomed
yielded much better results.
For the carriage of the men's food, supplies, and other impedi-
menta the few camels I could spare were not sufficient. So a
dozen donkeys were engaged which offered the advantage of
needing a minimum of fodder. For the camels only a quantity
of oil made of rape seed could be taken along. Half a pound
daily of this evil-smelling liquid for each animal proved
wonderfully effective in keeping up their stamina during the
trying desert marches, when they had to go without grazing
or fodder of any kind and sometimes for a number of days
without water. Our ponies, for which the desert to be crossed
offered neither sufficient water nor fodder, were sent back
to Khotan in charge of Niaz, the interpreter. The dejected
faces of my servants, when it was made clear to them that
they would have to trudge through the sands on foot like my-
self, were truly amusing.
A severe cold brought on by exposure made me glad for the
day's halt at Tawakkel which these various
preparations
demanded, and which was the last I could pass in comparative
comfort. My attempt to utilise it also for getting rid of a
troublesome tooth through the local barber's help proved a
painful failure. This worthy first vainly tortured me with a
forceps of the most primitive description, then grew nervous,
and finally prayed hard to be spared further efforts. Perhaps
he had lost confidence in his hands and instrument, since I had
insisted on seeing them thoroughly cleaned witli soap and hot
water previous to the intended operation.
I
START INTO DESERT. 255
When at last on the forenoon of the 12th December the
camels were ready with the freshly packed loads and my troop
of labourers duly collected, half the population of Tawakkel
seemed to be assembled to witness our departure. Those
who had come to wish luck and a safe return to relatives among
my party followed us to the northernmost hamlet. Then
beyond, where cultivated land gave way to scrub-covered
low dunes by the river bank, the Beg of Tawakkel, who escorted
me with two picturesque attendants carrying falcons as a sign
KEL LABOURERS TAKn..
of his dignity, took leave. A present of some Russian ten-
rouble gold pieces was to reward him for the services he had
rendered, not too willingly perhaps, but still effectively, and
to assure his good will in keeping open our communications
while we were in the desert. The first march was a very short
one, to a deserted shepherd station
('
satma
')
by the side
of the river
;
for follo^^ing the advice of the guides, I wished
to let all our animals have a plentiful drink in the evening before
entering the sands eastward. On the following morning we
struck to the east, and soon found the track marked by the
footprints of the small advance party which I had sent ahead
256 TO THE EUINS OF DANDAN-UILIQ.
two days previously under the guidance of Kasim. He had
orders to dig wells at all places suitable for camps, and after
reaching the ruins of Dandan-Uiliq to push on to the Keriya
Darya, whence Ram Singh was to join me.
Marching in the drift sand was slow work, though the dunes
were low, rising only to 6 to lo feet in the area crossed during
the first two days. The feet of men and animals sank deep
at every step into the fine sand, and the progress of the
heavily laden camels was reduced to about i^ miles per hour.
In view of the want of sufficient fodder and water, it was
essential to save them all over-exertion
;
hence I soon found
that the direct distance covered by a day's march could rarely
exceed
9
to 10 miles. The tamarisk and
'
Kumush
'
scrub
which was plentiful at first, grew rare in the course of our second
march, while the wild poplars or
'
Toghraks ' disappeared
altogether as living trees. Luckily amidst the bare dunes
there rose at intervals small conical hillocks thickly covered
with tamarisk scrub, the decayed roots of which supplied
excellent fuel. Close to these hillocks there were usually to
be found hollows scooped out of the loess soil, evidently by
the erosive action of wind. These hollows, which reach down
to at least 10 to
15
feet below the level of the little valleys
separating the neighbouring sand dunes, offer of course the
nearest approach to the sub-soil water. It was accordingly
invariably in these depressions that Kasim's advance party had
dug their wells, which we also chose for our camping places.
The water, which was reached after digging to an average
dejUh of
5
to
7
feet, was very bitter at the first two cam])s
and scarcely tit for human consumption. But as we moved
further away from the Khotan River it became comparativel}'
sweet. I have no doubt that geology would furnish a satis-
factory explanation for this observation, which was well known
to my guides as generally applicable to these parts of the
Taklamakan and has been noticed already by Dr. Hedin. The
supply of water furnished by these wells was decidedly scanty
for so large a party as mine
;
and as it was stopped altogether
by the damj) soil getting frozen overnight, men had in the
evening to be detailed gradually to collect spare water in two
CAMPING IN WINTRY DESERT. 257
of my iron tanks where it could be stored as ice for use on the
next day.
The winter of the desert had now set in with full vigour.
In daytime while on the march there was little to complain
joi
;
for though the temperature in the shade never rose above
freezing point, yet there was no wind, and I could enjoy without
discomfort the delightfully pure air of the desert and its repose
which nothing living disturbs. But at night, when the ther-
mometer would go down to minimum temperatures from o to
io Fahr. below zero, my little Kabul tent, notwithstanding its
extra serge lining, was a terribly cold abode. The
"
Stormont-
Murphy Arctic Stove
"
which was fed with small compressed
fuel cakes (from London
!)
steeped in paraffin proved very
useful
;
yet its warmth was not sufficient to permit my dis-
carding the heavy winter garb, including fur-lined overcoat
and boots, which protected me in the open. The costume
I wore would, together with the beard I was obliged to allow
to grow, have made me unrecognisable even to my best friends
in Europe. When the temperature had gone down in the
tent to about
6
Fahr. below freezing-point, reading or \VTiting
became impossible, and I had to retire among the heavy
blankets and rugs of my bed. There
'
Yolchi Beg
'
had usually
long before sought a refuge, though he too was in possession
of a comfortable fur coat of Kashmirian make, from which he
scarcely ever emerged between December and March.
To protect one's head at night from the intense cold while
retaining free respiration, was one of the small domestic problems
which had to be faced from the start of this winter campaign
in the desert. To the knitted Shetland cap which covered the
head but left the face bare, I had soon to add the fur-lined
cap of Balaclava shape made in Kashmir, which with its flaps
and peak pulled down gave additional protection for every-
thing except nose and cheeks. Still it was uncomfortable
to wake up with one's moustache hard frozen with the respir-
ation that had passed over it. Ultimately I had to adopt
the device of pulling the end of my fur-coat over my head and
breathing through its sleeve ! Also in another way these first
campings in the wintry desert brought some trying experiences.
17
258 TO THE EUINS OF DANDAN-UILIQ.
The tooth I had vainly endeavoured to get rid of at Tawakkel
continued to cause trouble, and the neuralgic pains it gave
me were never more exquisite than at night. The only remedy
I had within reach to secure some rest was chlorodyne, and to
take its drops I had need of water. But for this it was first
necessary to melt the solid lump of ice contained in my
aluminium tumbler, and the minutes which passed until I
had secured over my candle the little quantity of liquid, were
enough to benumb hands and fingers.
On the evening of the fourth day after entering the desert, as
we were pitching camp amidst desolate sand dunes covering dead
tamarisk scrub, two of the men sent ahead returned to report
that Kasim's party had failed to trace the ruined site we were
in search of. It was now the turn of old Turdi, my
"
treasure-
seeking
"
guide and factotum, to prove his knowledge of this
dreary region. He had only once in his life approached Dandan-
Uiliq from this side, and had apparently, from a feeling of
professional etiquette or pride, refrained from pressing his advice
against the guidance of the two Tawakkel hunters. But he
had more than once on the march told me that he thought
our route was leading too far north, and now, on the plain
avowal of their inability to discover our goal, I could see a
gleam of satisfaction pass over his wrinkled face. A short
conversation with the returned men sufficed for him to locate
the point which Kasim's party had reached, and early next
morning they were sent ahead again with full instructions that
were to guide Kasim back into the right direction. We our-
selves set out later, now under the guidance of old Turdi, wh0|
with an instinct bred by the roamings of some thirty years
and perhaps also inherited

^his father had followed the fortunes

of a
treasure-seeker's life before himfound his bearings even
where the dead uniformity of the sand dunes seemed to offer
no possible landmark;
Skirting the foot of several higher ridges of sand or
'
Dawans
'
running as usual from N.W. to S.E., we crossed in the evening
a belt of ground where dead trees were seen emerging from
heavy sand. Shrivelled and bleached as they appeared, Turdi
and the men could recognise among them trunks of the^' Terek^'
i
FIRST SURVEY OF RUINS. 259
r poplar, the willow and other planted trees, unmistakable
proofs that we had reached the area of ancient cultivation.
About one and a half miles further to the S.E. we came upon
stretches of bare loess with an extensive line of hollows, curiously
resembling a dry river course, yet undoubtedly only a result
of wind erosion. In one of these steep-banked hollows we
succeeded in digging a well, and thus saved ourselves a search
in the dark for the spot which alone, according to Turdi's know-
ledge, offered water in the immediate vicinity of the ruins.
On the following morning, the i8th of December, after turning
a
great Dawan, Turdi guided us to this spot, and a couple of miles
further south I found myself amidst the ruined houses which
mark the site of Dandan-Uihq.
Scattered in small isolated groups over an area which my
subsequent survey showed to extend for about one and a half
miles from north to south with a width of three-quarters of
a mile, there rose from among the low dunes the remains of
buildings modest in size, but of manifest antiquity. WTiere
the sand had blown away, the walls constructed throughout
of a timber framework with thick layers of plaster were exposed
to view, broken down to within a few feet from the ground.
Elsewhere in places covered by low dunes the walls could stiU
be made out by the rows of wooden posts rising from the sand;
All structural remains left exposed showed signs of having
been
"
explored " by treasm-e-seekers, and the marks of the
damage done by their operations were only too e\'ident. Yet
even thus the ruins, on a first hurried inspection, furnished
unmistakable proofs of their character and approximate date.
In the remains of frescoes which had once adorned the much-
injured walls in some of the larger rooms, I could easily recognise
representations of Buddhas and Bodhisattvas. These plainly
indicated that I was standing in the ruins of Buddhist places
of worship. Peculiarities in the style of the frescoes seemed
to mark the last centuries preceding the introduction of Islam
as the probable date when these shrines and the settlement
to which they belonged had been deserted. And this conclusion
received curious support on the first day by Chinese copper
coins bearing the date of the Kai-yuen period (a.d.
713-741),
IT
260 TO THE RUINS OF DANDAN-UILIQ.
which were picked up under my eyes from the debris-strewn
ground near the buildings.
Old Turdi felt quite at home among these desolate sur-
roundings, which he had visited so frequently since his boy-
hood. It was the fascinating vision of hidden treasure which
had drawn him and his kinsfolk there again and again, however
scanty the tangible reward had been of their trying wanderings.
Yet the familiarity thus acquired with the silent relics of the
past had developed in him an instinctive interest in all traces
of the life that once reigned there. As he guided me among
these ghostly wrecks of houses and answered the many questions
I put to him about his own observations, I could see the shy
man grow more and more animated. It was evident from
his communications that the conditions of the dunes were
changing very slowly at this site. Consequently Turdi had no
difficulty in recognising the places where he and his companions
had been at work during previous visits. Luckily their scanty
resources had never allowed them to overcome the difficulty
experienced in carrying to this distant site supplies sufficient
for a prolonged stay or to bring working parties of more than
a few men.
Hence the structures more deeply buried in the sand had
escaped unopened. It was important to select these in the
first place for my excavations, and I felt grateful for Turdi's
excellent memory and topographical instinct which enabled him
readily to indicate their positions. Guided by this first rapid
survey, I chose for my camp a spot from which the main ruins
to be explored were all within easy reach. There were practical
considerations which compelled me to make my choice care-
fully. For in order to keep my men at work as long as possible
every day, it was necessary to spare them tiring tramps through
the drift sand. It was still more essential that we should keej)
to a spot at which fuel could be readily obtained in the ample
quantities I foresaw would be needed for our long stay. The
dead
trees of ancient orchards could alone supply it, and tiieir
occurrence in various i)arts
of the ruined site was very unequal.
As soon as the baggage had been unloaded at the point which
appeared best to answer these conditions, I hastened to des-
CAMP AT DANDAN-UILTQ. 261
patch the camels under Ahmad Merghen's guidance on their
three days' journey eastwards to the Keriya Darya. There in
the jungle lining the river's course they were to find the fodder
they so badly needed, and to gather fresh strength for sub-
sequent desert marches. The donkeys, too, which had carried
the men's food supplies, with the meagrest rations for them-
selves, were sent back to Tawakkel under the care of two of
the \allagers.
CAMP IN THE DESERT, DANDAN-UILIQ.
CHAPTER XVIII.

EXCAVATION OF BUDDHIST SHRINES.
On the
morning of the 19th of December I commenced my
excavations by clearing the remains of a small square building
immediately to the south of my camp. Turdi knew it as a
'
But-khana ' or
"
temple of idols," and well remembered once
having searched it in his own fashion. But the sand, though lying
only two to three feet high, had not been removed, and by laying
bare the foundations and floor I could expect to gain useful })re-
liminary knowledge as to the general construction and arrange-
ment of such shrines. In this I was not disappointed. A careful
examination of the remains of walls which were brought to light
on the north and west sides showed that there had been an inner
square cella enclosed by equi- distant outer walls twenty feet
long, forming a kind of corridor or passage on each side. Both
inner and outer walls consisted of hard plaster laid on a frame-
work of wood and reed matting, which itself was held in position
by massive square posts fixed at regular intervals.
/ Of the manner in which the upper portions of the inner cella
/
walls, long ago decayed, had once been decorated, I could not
remain in doubt when fragments of flat stucco relievos, which
must have originally belonged to plaques of regular
patterns,
\
RELIEVOS OF BUDDHIST SHRINES. 263
turned up in dozens from the sand covering the interior. Mixed
with frequently repeated architectural ornaments there were
numerous reproductions in low reUevo of the figure of Buddha,
in the orthodox attitudes of teaching \Nith hand raised or seated
in meditation. Other small rehevos showed attendant figures
in adoration, such as the graceful garland-holding woman rising
from a lotus and probably meant for a Gandharvi, which has
been reproduced on the cover of this book. Conventional as
all these representations are and evidently casts from a series
of moulds, they at once arrested my interest by their unmis-
takable affinity to that style of Buddhist sculpture in India
which developed under classical influences. Nor was I less
gratified to observe how well many of these small relievos re-
tained the bright colours \vith which they had been painted.
Equally reassuring proof of the preserving power of the desert
sand was furnished by the remarkable freshness in which ela-
borately painted figures of Buddhist saints appeared on pieces
of wooden posts and beams that evidently once belonged to the
ceiling.
The clearing of this single small shrine not only yielded some
one hundred and fifty pieces of stucco relievo fit for transport to
Europe, but supplied me with the indications I needed in order
to direct the systematic excavation of structures more deeply
buried in the sand. So when on the next day I proceeded to a
group of small buildings buried below six to eight feet of sand
by the slope of a fairly high dune, just half a mile south of my
camp,
I was able correctly to gauge their construction and
character, though only the broken and bleached ends of posts
were visible above the sand. They are seen in the accompany-
ing photograph, which shows the place before excavation. The
posts soon proved to belong to the walls of two temple cellas
(marked as D. II.), once richly decorated with frescoes and
stucco images.
As their constructive features and adornment are typical of
those
observed in other shrines subsequently excavated at this
site I shall describe them here briefly. The larger cella forms a
square
of ten feet inside, with a door opening from the north.
The walls, which here again were constructed of a wooden frame-
CELLA OF
BUDDHIST
SHRINE.
265
work with layers of hard plaster on either side, showed a uniform
thickness of six and a half inches. The cella was enclosed by
a quadrangular passage about four and a half feet wide, with
outer walls of the same materials. This passage, which almost
certainly ser\'ed for the purposes of the circumambulation
("
pradakshina
')
common to all traditional forms of Indian
worship, also had its entrance in the centre of the north wall.
The interior of the cella was once occupied by a colossal statue
made of stucco and painted, which most probably
represented
a Buddha. But of this only the feet remained, about thirteen
inches long, raised on an elaborately moulded oblong base about
three feet high. The other parts of the statue had crumbled
away long ago, and the fragments comprising parts of the legs
and of the lower drapery which were found in the sand above
the base broke at the slightest touch. Of the wooden framework,
too, which once supported the heavy image, only the lowest
part was still intact, fixed \%ithin the left foot. Each of the four
comers of the cella was occupied by a draped stucco figure
standing on a lotus-shaped pedestal. But of these statues only
the one in the north-west corner was found intact up to the
waist. A photograph of this cella, taken after the excavation,
is reproduced on
p.
266.
The walls of the cella, which, judging from the size of the
statue, must have been of considerable height, were decorated
inside with frescoes showing figures of Buddhas or Bodhisattvas
enveloped in large halos. As these too were over life-size, only
the feet with the broad ;painted frieze below them showing
lotuses and small figures of worshippers, could be seen on the
walls still standing. The colours looked faded and worn, as if
the frescoes had been exposed for a considerable time before
the protecting sand invaded the building. But the outlines,
drawn mostly in a kind of terra-cotta colour on the fine-grained,
well-prepared plaster surface, were still sharp and clear. The
decoration of the outside of the cella walls consisted mainly of
fresco bands containing small representations of seated Buddh-
ist saints in the attitude of meditation, only the colours of the
robes and halos varpng. But amidst these conventional
designs there was found a picture which, though much effaced,
DUDDHIST FEESCOES. 267
at once attracted my interest as representing some sacred legend,
\
perhaps of a local character. It shows rows of youths riding
on horses or camels each holding a cup in his outstretched right
hand, while above one of the riders a bird, perhaps meant for a
falcon, is swooping down on this offering. The popularity of
the subject was subsequently attested by my discovery of a
well-preserved painted tablet in another temple ruin on which
->
a similar scene is figured.

>,^
Frescoes of Buddhist saints over-lifesize, similar to those \
found in the cella, once adorned the inside of the walls enclosing
the passage. Below them there ran a decorative frieze in which
lotuses floating in the water and small human figures, perhaps
meant for Nagas or deities of springs, supporting the feet of the
sacred personages above, could still be made out. From the
south wall of the passage I succeeded in removing the piece of
painted plaster which is seen on
p.
268,
and which is now safely
deposited in the British Museum. It shows the figure of a seated
Buddha or Bodhisattva, occupying the triangular space left
between the lower portions of two larger frescoes. The inscrip-
tion painted beneath in black colour is in a cursive variety of
the Brahmi script ; its language, however, like that of some
other short inscriptions found on the frescoed walls of the
Dandan-Uiliq ruins, is not Indian, but probably the same as
appears in the Non-Sanskritic Brahmi documents I discovered
at this site.
The excavations, when extended on the 21st of December
to the remains immediately adjoining the west wall of the shrine
just described, brought to light another Buddhist temple cella
which, notwithstanding
its smaller dimensions, proved parti-
cularly rich in interesting art relics. This little chapel, as it
might be called, measured only 12 ft. 8 in. from north to south
with a width of 8 ft. 8 in., and had no enclosing square passage.
Its walls, built of the usual wooden framework and plaster, were
only four inches in thickness and had in consequence crumbled
away to within a foot or two from the ground, except on the east
side, where the closely adjoining outer wall of the larger cella
gave support, and on the south where a long platform, sur-
mounted by a massive base for the principal image, had been
268 EXCAVATION OF BUDDHIST SHRINES.
built against the wall. Of the stucco statue which once occupied
this base only the scantiest fragments could be found
;
for with
a pedestal raised close on four feet above the ground it must have
long remained without the protecting cover of drift sand, as
testified by the extremely friable condition of the few recovered
fragments of coloured stucco that belonged to this image. The
FRESCO FROM OUTER WALL OF SHRINE, D, IL, UANDAN-ULIQ.
(Scale one-eighth of original.)
front of the base, which was nearly three feet broad, proved to
be flanked on either side by the half-detached figure of a lion,
and was thus manifestly meant to represent the
'
Simhasana,'
or
'
lion's throne,' on which ancient Indian tradition seated
both heavenly and earthly rulers. The heads of the lions had
decayed long ago, but the curls indicating the manes falling
over the fore part of the bodies were still clearly recognisable.
In the top layer of the sand which covered the south-eastern
DECORATION OF CELLAS. 269
corner of the cella (showTi by the photograph reproduced on
p.
270),
were found numerous small relievos in stucco, from
5
to 8
in. high, representing Buddhas or Bodhisattvas with the right
hand raised in the attitude of teaching. The robes of these
figures are painted reddish brown, the orthodox colour for the
garb of Indian monks, while their heads are surrounded by halos
in a light green shade. Some sj^ecimens were still attached to
portions of a hard stucco ground, decorated in relievo with
elaborate wreaths and rosettes and gaily coloured. It was
e\4dent that all these pieces once belonged to plaques which,
perhaps in the form of large halos, had adorned the uppermost
part of the walls in this corner. The fact of these fragments
being found in loose sand, several feet above the platform already
referred to, is a clear indication that the interior of the little
shrine had been invaded by the drift-sand while its walls were
still intact to a considerable height.
-.^
Comparatively well protected as this south-eastern portion of
the cella was, the stucco image which once occupied the corner
had decayed, just as in the case of the other three corners, down
^
to the feet. These, however, as well as the elegantly moulded
[
lotus-pedestal of circular shaj:)e, can still be made out in the
photograph. Luckily this corner had afforded better protec-
tion for some other adornments of the shrine. On clearing the
platform between the corner pedestal and the base of the prin-
cipal statue, I found a small detached statue in stucco about
i^ ft. high and well preserv^ed but for the head and arms. The
photograph shows it placed subsequently on the main base.
The seated figure represented in it must, by evidence of the
carefully indicated dress, have been intended for a Buddha or
,
Bodhisattva. The colour of the robe, a reddish brown, has
survived very well. The small wooden board on which the
image had been set up, evidently for the purpose of convenient
transport, was still intact, and as the stucco too had kept com-
paratively hard I was able to risk its removal. Carefully packed
away in one of my mule-trunks, amidst cotton wool and plenty
of soft country paper I had provided myself with at Khotan, the
little statue accomplished its long journey to London far better
than I had expected.
PAINTED PANELS ANT) RELIEVOS. 271
At the foot of the principal base and leaning against it were
found five painted panels of wood, all oblong, but of varying
sizes. The largest measures ii in. in length, with a height of
5^
in., and has a thickness of about a quarter of an inch. Owing
to their position near the ground the wood of these panels and
also the thin layer of water-colour with which they are painted
has suffered much, evidently through damp. For the same
reason the removal of the crust of sand and siliceous matter
which adheres to the surface proved a very delicate task. But
even the imperfect cleaning I could attempt at the time sufficed
to show that these little paintings represent personages of
Buddhist mythology or scenes bearing on Buddhist worship
and legends. On one of them two figures evidently meant for
Bodhisattvas, can be seen seated on lotus-flowers, vvith coloured
vesica and halo behind them. In another I could, notwith-
standing the much-faded outlines, recognise the quaint features
of that popular figure of the Indian pantheon, the elephant-
headed god of learning, Ganesha. A third panel exhibits the
figure of a dancing-woman, drawn in full movement and with
remarkable freedom. From the head, which is thrown back,
there flows downwards a quantity of black tresses, while the
left hand holds the loop of a sash or veil poised in graceful
cur\'e over the head.
These painted tablets, like all the others subsequently dis-
covered at the bases of sacred images in ruined temples of
Dandan-Uiliq, were undoubtedly still in the same position in
which they had originally been deposited as the votive offerings,
of pious worshippers. And curiously enough, as if to show the
care which must have been taken by the last attendants of the
little shrine to keep the sacred objects clear of the invading dust
and sand. I discovered several ancient brooms both near the
principal base and in other parts of the ceUa (reproduced
p.
358).
They were about 16 in. long and constructed in a very
ingenious way from stalks of some hardy grass. At their bottom
these stalks were plaited into a continuous strip subsequently
rolled up tightly and bound round with twisted grass, while
their feathery ends, being thus brought into a bunch, form a
convenient birch-like broom. The sand against which these
272 EXCAVATION OF BUDDHIST SHRINES.
humble implements were once used to wage war had been the
means of preserving them in almost perfect condition
;
and the
same was the case with another curious relic, a little cloth bag
filled with fragments of bones and human teeth, which turned
up in the south-eastern corner close to the small seated image
already described. Had they been brought here by some visitor
as reputed relics of the body of a saintly personage, or were they
ex-votos deposited with some superstitious object ?
As the work of clearing proceeded along the east wall of the
cella it revealed a series of very interesting frescoes, together with
a relievo statue in stucco of a peculiar character. As seen in the
photograph it is a male figure, complete but for the head and
left arm, standing close to the platform already mentioned and
over the body of a prostrate foe. The figure, which measured
a little over three feet from the heel to below the arm-pit, is
clad in a coat of mail reaching below the knees and elaborately
decorated. The gay colours of the successive rows of small
plates which form the mail, alternately red-blue and red-green,
were remarkably well preserved, and not less so all the details
of the ornaments which are shown along the front and lower edge
of the coat and on the girdle around the waist. Even the
arrangement of the rivets which join the plates of mail, and the
folds of the garment })rotruding below the armour, are indicated
with great accuracy. There can be no doubt that the artist has
carefully reproduced here details of armour and dress with which
he was familiar from his own times. The feet, which seem to
be clad in wide top-boots of leather, just like the
'
Charuks
'
still worn throughout Eastern Turkestan, are placed over the
contorted body manifestly of a vanquished demon. The fea-
tures of the lattcr's head, which alone is raised somewhat from
the ground, with the eyes wide open and the teetli displayed,
express terror. The representation of the thick hair by elabo-
rately worked spiral tufts strongly reminded me of the treatment
of the hair in many a sculpture of Gneco-Buddhist type familiar
to me from the Lahore Museum. The body a}^j)cars to have
been painted dark blue, but owing to the low jwsition of this
relievo the stucco retained little of the original coating of colour.
The standing figure probably represents one of the Yakshas,
SCENES OF FRESCOES.
273
or divine
"
guardians of the gate "
popular in Buddhist
mythology.
The cella wall immediately adjoining this relievo group
revealed a series of small fresco paintings which, by their uncon-
ventional subjects and their spirited drawing, at once attracted
my attention. The one to the left of the mail-clad statue shows
a woman bathing in a square tank of water, enclosed by a
tesselated
pavement and filled with floating lotuses. The figure
is nude, except for a large red headdress resembUng an Indian
Pagri and profuse ornaments round the neck, arms and wrists,
and is drawn with remarkable verve in simple yet graceful out-
lines. The right hand with its shapely fingers rests against the
breasts, while the left arm is curved down towards the middle
of the waist.
Fourfold strings of small bells are shown hanging
around the hips, just as seen in representations of dancing-girls
in early Hindu sculpture
;
while, curiously enough, an elaborate
vine-leaf appears where post-classical convention would place its
fig-leaf. The face of the bather is turned to her proper right,
down towards a small male figure, apparently a boy, who is
shown as if trying to rise from the water by holding to her side.
The delineation of the lotus-flowers which rise from the tank
in a variety of forms, closed or half-open, as well as their colours,
ranging from dark blue to deep purple, seemed remarkably true
to nature, and distinctly suggested that these sacred flowers
were familiar to the painter from personal observation.
I
remembered the splendid tank of lotuses I had seen at the
Tao-tai's Yamen in Kashgar which had been grown from seed
imported from China. In view of this pictorial representation I
feel convinced that already ancient Khotan had known the
graceful plant dear to the gods of India. Considering the close
historical connection between Kashmir and Khotan which the
local traditions recorded by Hiuen-Tsiang indicate, it needs no
effort of imagination to beUeve that the lotuses that once adorned
the gardens of settlements now buried by the desert sand were
originally derived from the great Himalayan Valley, on the lakes
of which I had so often admired them;
The appearance of a riderless horse in front of the tank an<i
some
other features of the fresco suggest that its subject may
18
274 EXCAVATION
OF BUDDHIST
SHEINES.
perhaps be identical with the curious legend which Hiuen-Tsiang
relates of a Naga lady residing in a stream east of Khotan and
j
of her strange wooing by a pious mortal. But the point is too
! uncertain to permit more than an allusion here.
Of the adjoining frescoes, however, it is impossible to mistake
the significance. A well-drawn though now much effaced male
figure of youthful appearance, seen seated in cross-legged fashion
and dressed in a dark-blue cloak that leaves the right shoulder
bare, is manifestly that of a Buddhist scholar. His right hand
holds the oblong leaves of a
'
Pothi,' or manuscript book
arranged in the traditional Indian fashion, on which the eyes are
fixed in intent study. By the side of this figure and likewise
turned to the proper right, an old man is depicted in the act of
teaching. His robe, which seems to be made up of patches of
varying shades of brown, curiously suggests the orthodox garb of
mendicant monks of all Indian sects, termed
'
chiravastra
'
in
Sanskrit. While the right hand, with the second and third
fingers stretched out, is raised in the gesture of teaching, the palm
of the left supports a closed
'
Pothi.' The two boards of thin
wood between which the leaves are placed after a fashion still
commonly observed in the case of Sanskrit manuscripts, are
quite distinctly marked. The cleverly drawn features of the
old man's face seem to express complacent assurance in his
teaching and full abstraction in its subject. In front of him,
too, a tank is shown with open lotuses floating on the surface.
Two birds, looking like wild geese, disport themselves in the
water, and with necks marked dark-blue and green, raise their
heads towards the old teacher.
To remove any portion of these interesting frescoes proved
quite impracticable, owing to the friable condition of the plaster on
which they were painted. Nor can it, in view of the faded state
of the colours, be surprising that the photographs which I secured
of them do not permit of satisfactory reproduction by any
mechanical process. But drawings have been made from these
as well as other photographs of Dandan-Uiliq frescoes by the
hand of my artist friend, Mr. F. H. Andrews. These, when
^
published in my Detailed Report, will, I hope, render it easier
\
to judge of the remarkable resemblance which, in style of com-
SCENES OF FRESCOES. 275
position and the drawing of figures, exists between these frescoes
and the later of the Indian paintings in the cave temples of
Ajanta.
Little, indeed, of early Indian painting has survived in India
itself. Hence all the more interest must attach to the speci-
mens which the frescoes and painted tablets of Dandan-Uiliq
shrines have preserved for us of that selfsame Indian art as trans-
planted to the Buddhist region of Khotan.
18
ROOM OF MONASTIC DWELLING, D. III., DANDAN-UILIQ, FIND-PLACE OF
ANCIENT MANUSCRIPTS.
CHAPTER XIX.
FIRST FINDS OF ANCIENT MANUSCRIPTS.
It had not needed the discovery of the pictorial representation
of
'
Pothis
'
to make me eagerly look out for finds of ancient
manuscripts. None had turned up during the excavations oi
the first three days. But as if to revive my drooping hopes,
a painted tablet, badly defaced by decay and accretion of
siliceous matter, which was found in the last cleared cella, dis-
played a narrow strij) of paper with three lines of Indian Bralnjii
characters sticking to the top edge and running transversely
across the panel. The paper, which, as it covers part of the
painting, is plainly j)roved to be a subsequent addition, liad
decayed even more than the tablet itself. While on the latter
two female figures, each holding a swathed infant, could just
be made out, it was impossible to read more than a few detaclied
characters in each line of the rotten paper. But these letters
FIRST LEAF UNEARTHED. 277
were in a bold literary hand, very different from the cursive
writing seen below a few of the frescoes, and clearly suggested
a Sanskrit text.
The little temples so far excavated had shown me some-
thing of the cult and art which this sand-buried settlement
possessed before its abandonment. But for indications of the
conditions of everyday life and for other documentary evidence
it was manifest that I should have to turn to remains of a
different character. So, on the 22nd of December, I directed
my men to the excavation of a structure close by, which by its
position and ground-plan as deducible from the arrangement
of the wooden posts that were seen sticking out above the
sand, appeared to suggest an ancient dwelling-place. It lay
about twenty yards to the north-west of the temple-cella last
described, just at the northern end of a sand-dune which, vnth
its crest, rose to a height of fully 16 ft. above the original
ground level. The bleached trunks of dead fruit-trees, which
were visible around where the sand was less high, indicated
that this building, together with the cellas already excavated,
had stood in the midst of an orchard or garden. The digging
started on the west side soon brought to light the top part
of massive and fairly well-preserved walls in wood and plaster,
belonging to what was evidently the lowest storey of a dwelling-
house. The apartment formed by them had been an oblong
of
23
by 20 ft., and about 10 ft. high. The photograph at the
head of this chapter shows a part of it after excavation.
By noon, at a depth of 2 ft. from the surface, a small scrap
of paper showing a few Brahmi characters was foimd in the
loose sand which filled the building. I greeted it with no small
satisfaction as a promise of richer finds. In order to stimulate
the efforts of my labourers, who, with the sand continually
falling in from the side of the adjoining dune, had no easy task
in effecting a clearance, I offered a small reward in silver to the
man who should be lucky enough to hit upon the first real
manuscript.
Barely an hour later a cheerful shout from one
of the men working at the bottom of the small area so far
excavated on the north-west side of the apartment announced
the
discover)' of a
'
Khat,' or wTiting.
278 FINDS OF ANCIENT MANUSCKIPTS.
\ Carefully extracted with my own hand and cleared of the
adhering sand, it proved a perfectly preserved oblong leaf of
paper,
13
inches long and
4
inches high, that had undoubtedly
I
formed part of a larger manuscript arranged in the shape of
'
an Indian
'
Pothi.' The circular hole intended for the string
that was passed through the separate leaves in order to keep
them together and preserve their order is placed on the left
side of the leaf, as in most of the ancient manuscripts that have
previously been acquired from Chinese Turkestan. The six
lines of beautifully clear writing which cover each side of the
leaf show Brahmi characters of the so-called Gupta type, but
a non-Indian language. The photograph reproduced on
p.
279
shaws one side of this leaf.
While the men gathered around to watch me cleaning the
precious find, I heard more than one humorous remark about the
chance which had placed this first
'
Khat,' as well as the cash
reward for it, in the hands of the youth who alone of the party
could read and write, and whom, as already related, I had
brought along from Tawakkel just on account of these ac-
quirements. Niaz, our
'
Mullah,' to give him his proud title
(he is seen kneeling on the extreme left in the group repro-
duced in my photograph,
p.
255),
was himself beaming with
boyish delight at his good luck, and subsequently did his best
to prove worthy of it by additional care in digging and in
penning my Turki
"
despatches."
The interesting find just described was made at a depth of
I
about
5
feet from the surface and close to the rough wooden
post seen upright on the left side of the apartment as photo-
graphed after excavation
p. 276.
It was quickly followed by a
series of other manuscript finds, either in loose leaves, more or
less complete, or in little sets of fragments. They all showed
Brahmi writing, of an early type and had, as their conformity
in paper, size and handwriting showed, originally belonged to
at least three distinct
'
Pothis,' or books. Their contents
were soon recognised by me as Sanskrit texts treating of
\
Buddhist canonical matter. The position in which all these
manuscript pieces were found, embedded in loose sand several
feet above the original flooring, proved beyond all doubt tliat
H
O
<
T
in
^
o
2
O g
Q -^s
Q >
09
^
u
280 FINDS OF ANCIENT MANUSCRIPTS.
they could have got there only by accident. Their distribu-
tion in varying depths and places suggests that they had
fallen in from an upper storey, while the basement was gradually
filling up with drift-sand. This assumption was fully borne out
by the small pieces of felt, leather, oilcakes
('
kunjara
'),
and
similar refuse which turned up in the same layers. The
pagination numerals which I could make out on the margin of
some leaves, and which in one instance go up to
132,
plainly
showed that the pieces thus rescued were mere fragments of
larger texts which had probably perished with the destruction
of the upper floor.
The earlier these fragments had reached the safe resting-
place offered by the sand-covered basement, the more ex-
tensive they might reasonably be expected to be. So I watched
with growing eagerness the progress my men made on the
23rd of December in clearing the sand nearer down to the
original floor. It was no easy task, for the drift-sand from the
slope of the dune to the south was ever slipping to fill the
space laboriously cleared, and as the wall on that side had
apparently decayed long ago, additional exertions were needed.
As the work proceeded towards the centre of the room a massive
beam of poplar wood, nearly a foot in thickness, was laid bare.
Its length, close on
19
feet, and its position showed that it
had once stretched right across the room, undoubtedly sup-
porting its roof. Two well-carved octagonal posts with bell-
shaped capitals surmounted by a circular band, in which I
easily recognised the Amalaka ornament of Indian architecture,
had turned up before
;
they had undoubtedly served to support
this central beam.
A little beyond the latter, towards the east, the men clearing
the sand just above the floor came upon a closely packed
bundle of manuscript leaves, evidently still retaining the order
they had occupied in the original
'
Pothi.' A little later two
more packets of leaves belonging to the identical manuscript
were brought to light, practically intact, though the action ot
moisture to which these leaves must have once been subjected,
owing to their position not far above the ground, had stuck
them closely together and made them so very brittle that
their
DISCOA'ERY OF SANSKRIT
*
POTHL' 281
successful separation could only be accomplished in London
through the expert help of the Manuscript Department of the
British Museum. The ends of the leaves had been bent over
near the usual string-hole already referred to, and had often
got detached through this folding of centuries
;
but they could
be fitted again without difficulty to their proper places.
The leaves in their complete state measure
14
inches in
length and show on each side six lines of bold Brahmi writing
of the so-called Gupta type. The text, which is Sanskrit,
deals with some subject of Buddhist
'
Dharma,' or canonical
law. In view of the extent of the well-preserved portions it
will in all probability permit of an exact identification by Dr.
A. F. Rudolf Hoernle, the distinguished Indologist, who has
undertaken the decipherment and publication of all manu-
script materials in Brahmi characters discovered by me.
[WTiile these pages are passing through the press, Dr. Hoernle
informs me that he has recognised in this manuscript almost
the whole of the Vajracchedika, a famous Sutra text of the
Mahayana School of Buddhism.] Certain palaeographic features
of the \\Titing, which need not be set forth here in detail, make
it difficult to assign to this and the other Sanskrit manuscripts
recovered from this ruin a date later than the seventh century
A.D. But as far as other observed criteria go, some of them
might well have been written a century or two earlier.
The religious character of their contents makes it appear
highly probable that these manuscripts formed part of the
library
of a Buddhist monastic estabhshment, or
'
Vihara,'
that had once occupied the structiue and no doubt supplied
the
attendant priests for the adjoining small temples. That
the
basement room I was actually excavating had offered only
accidental
shelter to these fragmentary relics of Buddhist
literature,
and had originally served the more prosaic purposes
of a cook-room for the little monastery, became abundantly
clear as the work of clearing proceeded towards the east wall.
Built
against the latter we found a big fireplace, constructed
of hard
plaster \nth an elaborately moulded chimney that
reached
to a height of over 6 feet from the floor.
By the side of it a broad wooden bench filled a kind of recess.
282 FINDS OF ANCIENT MANUSCRIPTS.
Judging from a similar arrangement still observed in Turkestan
houses, and from the broken pottery discovered below it, this
bench probably served for the handy storing of cooking utensils.
In front of it, and not far from the fireplace, there stood a rough
wooden tripod, such as is still used throughout the country to
support large water-jars required for kitchen purposes, while a
short post with branching head, which I found fixed in the
ground close to where the first manuscript leaf was discovered,
certainly served to hang a kettle from. Remains of animal
bones, oilcakes, and small layers of charcoal found scattered
over the floor in various places fully bore out the conclusion
indicated.
My attention was still fixed on the manuscript remains that
were successively emerging from the depth of this sand-buried
dwelling, when at noon on the 23rd of December the sound
of a distant gun-shot was heard over the silent dunes eastwards.
Old Turdi, who with me was keenly watching the excavation
work, at once interpreted the faint sound as a signal that Ram
Singh was approaching from the direction of the Keriya Darya.
An hour later the Sub-Surveyor was by my side, together with
faithful Jasvant
Singh, his Rajput cook and companion, both
manifestly as pleased as I was at our successfully effected
junction. Considering the distances covered and the various
incidents for which it was impossible to make proper allowance
in our respective programmes, the rendezvous I had arranged
for had been kept most punctually.
I was greatly relieved to find from Ram Singh's report that
he had fully carried out the topographical task I had assigned
to him, and had experienced no difficulties from either the
Chinese or the native local authorities. Marching back by our
former route to above the Pisha Valley, he had effected a
supplementary triangulation of tiie great peaks above the
headwaters of the Yurung-kash. He had then made his way
north of the massive of Peak K.
5,
or
'
Muztagh,' to the
slopes of the great glacier-crowned range which sends its
numerous streams down to the small oases fringing the desert
west of Keriya. By plane-table work and triangulation
carried across the elevated spurs which descend from this
KENDEZVOUS IN DESERT. 283
range, he had succeeded in connecting our surveys eastwards
with the work done by Captain Deasy around Polu. In
accordance with my instructions he then proceeded to Keriya,
and thence a fresh
'
Darogha,' supplied as a guide and escort
by the local Amban, had taken him through the jungle belt
flanking the Keriya River down to the point where he fell in
with the party I had sent ahead under Kasim, the hunter.
Ram Singh, reticent at all times, had little to relate either
that day or thereafter of the mountain tracts he had passed
through, apart from topographical details which appealed to
his professional training, and which had already been duly
recorded in his plane-table
"
sections." That his little party
had undergone considerable hardships both on account of the
cold and the want of all local supplies, I could readily believe
from my own experience. But what had ev-idently impressed
him more than anything else, and what prompted him to a
short outburst of quite unusual communicativeness, was the
weird desolation of the desert, the total absence of life of any
kind among the high waves of sand he had crossed since leaving
the banks of the frozen river. I could see that the curiosity
excited by the manifestly Indian character of the sculptures
and paintings I had unearthed was by no means suflScient
to counter-balance the uncanny feelings which these strange
surroundings, pregnant with death and solitude, had roused in
my otherwise hardy Hindu followers. I did my best to cheer
them
by
sending as a welcome gift when their camp was pitched
such little luxuries as I had

packets of my own compressed

tea, some frozen eggs, raisins and almonds, etc.
In the evening when the dusk had put a stop to excavation
and I could tramp back through the sand to the shelter of my
little tent, I lost no time in sending for Ram Singh to examine
his plane-table work. A comparison of the position indicated
in it for Dandan-Uihq with my own fixing of the site would be
a decisive test for the accuracy of our respective surveys.
Considering the very deceptive nature of the desert ground over
which
we had carried so much of our route traverses, I could
not help feeling uneasy about the result. All the greater was
my
delight when I ascertained that the difference between
284 FINDS OF ANCIENT MANUSCRIPTS.
the positions which our wholly independent surveys showed
for our actual camp, amounted only to about half a mile in
longitude and less than a mile in latitude. If it is taken into
account that, since leaving our common camping place in
Khotan, Ram Singh had brought his survey over approximately
500
miles of route (on which for the last
130
miles or so no
intersections could be obtained owing to the absence of all
prominent landmarks), while my own marches extended over
about 120 miles, and lay almost wholly through desert, this
slight difference represents in reality a very striking agreement.
It could not fail to assure me as to the accuracy of our survey
work even far away in the desert sands where the frequent
dust-haze, if not the great distances from any elevated point,
practically exclude all hope of exact checks
by
means of trian-
gulation.
Neither Ram Singh nor Jasvant Singh took at first kindly to
life in the wintry desert. They both complained bitterly of
the badness of the water which our single brackish well yielded.
With the true Indian belief in the omnipotence of
'
pine-ka
pani,' they were eager to ascribe to this sole factor the un-
pleasant symptoms for which the combination of trying climatic
conditions and previous fatigues and exposure were mainly
responsible. Jasvant Singh, probably in consequence of the
total want of fresh vegetables, showed signs of an incipient
scorbutic affection, which- however, I was soon able to stop
by the administration of lime juice I had brought with me
from Gilgit. That it had first to be melted did not reduce
the effectiveness of this remedy.
For Ram Singh, who anticipated a return of the rheumatic
complaint which he had originally contracted while employed
on Captain Deasy's explorations, the work requiring plentiful
movement which I could assign to him in the jirejnaration of a
general survey of the ruins and of detailed plans of the structures
excavated,
proved perhai)s
the best antidote. He subsequently
stood the undoubted hardships which our winter campaign in the
desert entailed far better than I had ventured to hope. He also
bravely held out against the pains inflicted by his old enemy,
when at last in the spring it seriously attacked him. Nor did
AN ANTIQUE FODDER STORE. 285
my servants from Kashgar and Yarkand, though better pre-
pared for the rigours of the desert winter, escape without
suffering from the inevitable exposure. One after the other in
the course of this and the next two months was attacked by
painful swellings on the legs or arms, resulting in large boils,
which for the time being incapacitated the victim from any
useful service.
Old Turdi, whom many former
'
treasure-seeking ' expedi-
tions had inured to heat and cold alike and to all sorts of priva-
tions, was not likely to come on the sick list. Yet the quaint
pleasvue he took in showing me over what I used jokingly to
call his own village and temples, and the honest pride that lit
up his %\Tinkled face whenever I had occasion to appeal to his
quasi-antiquarian instinct and his experience of desert con-
ditions, were soon overcast by a cloud. Its real significance
I failed to comprehend during my first few days at these ruins.
While I felt overjoyed by the interesting discoveries which the
first excavations had pelded, poor Turdi Khwojam
('
my
pious Sir Turdi
'),
"
the Aksakal of the Taklamakan," as we
soon got to call him, was contemplating with sad apprehension
the imminent failure of a commercial venture quite serious for
his modest resources. The spirit of speculation, perhaps
natural in a
'
treasure-seeker,' had induced him to invest the
greatest part, if not the whole, of the advance of pay I had
given him before leaving Khotan in the purchase of an old pony.
The intention was that it should carry Turdi's provisions and
slender outfit to Dandan-Uiliq and should then be killed there
to provide meat for sale to my labourers.
Turdi would, no doubt, have reaped due benefit from this
ingenious combination of
"
transport and supplies
"
if the men
I brought from Tawakkel had not been mean enough to seize
upon the idea for their own advantage. They took along an
old cow as a joint-stock affair and duly slaughtered her near
their camping-ground soon after our arrival at the ruined site.
Both these time-honoured animals had been carefully kept
out of my sight by their respective owners while on the march,
probably from a correct siuTiiise that I should have insisted
upon the carriage of adequate fodder for them to prevent
286 FINDS OF ANCIENT
MANUSCRIPTS.
downright starving. Nor did I learn the facts until several
days after our arrival, when Turdi had at last to resort to a
strange and desperate expedient in the vain hope of saving
his pony. He had failed to come to terms with the Tawakkel
labourers for the sale of the animal, and had also let the oppor-
tunity go by of sending it either back to their oasis or on to
the Keriya River. So he tried to keep it alive by sending it
with a young fellow of his own fraternity, who accompanied
him as a kind of acolyte, for a considerable distance to the
south, where it might get some grazing on dry tamarisk leaves
and Kumush.
Just
when the poor creature had no more strength left for the
daily journey in search of this miserable diet, Turdi made a
curious discovery, which to his confiding soul appeared at first a
quasi-miraculous saving. Scraping the sand-covered bank of a
small depression that had formed through wind erosion by the
side of a ruined dwelling house, the remains of which were
visible about a hundred yards to the south of my tent and which
he had searched years before, he laid bare a closely-compressed
mass of straw. It had evidently been once deposited in a
corner of the fenced court-yard of that house and had, like the
fence itself, remained in a remarkably well-preserved condition,
though darkened and, of course, completely dried by the
long centuries that had passed since the sand covered it.
Turdi was exultant over his discovery, and at once brought
his starving pony, the existence of which could no longer be
hidden from me, to feed on this providentially jireserved antique
fodder store ! I had, of course, from the first strong doubts as
to the nourishing capacity of this perhaps the most
"
desic-
cated
"
fodder stuff that was ever offered to a horse. But the
poor famished creature swallowed it ravenously at the beginning
and seemed to justify old Turdi's hopeful expectations. How-
ever, a day's experience sufficed to prove that Turdi had badly
over-estimated the feeding value of his ancient straw, or rather,
as I ought perhaps to put it from consideration for my honest
"
treasure-seeking" guide, that he was less of an authority on
the keeping of live animals than on most matters dead and
buried beneath the desert sands. So when, a day later, Ram
TURDI'S PONY. 287
Singh arrived with some of my camels that carried his baggage
from the Kenya Darya, I had as much as possible of the httle
store of dry Kumush they had brought along given to the
poor famished pony.
Turdi, still finding no purchasers, was now anxious to have it
returned to Tawakkel, and, as the return of Kasim's small
party made it easier now to spare men from the work, I arranged
at once to have it sent back there in charge of two labourers,
who were also to take my mail-bag for transmission to Khotan.
I had, of course, great doubts as to whether the victim of Turdi's
ill-fated experiment would be able to cover the sixty miles or so
of desert marching to the river bank. Hence it was no surprise
to me when the first arrivals from that sidetwo men who
brought me news from the kindly Amban of Khotanreported
that the pony had succumbed to its sufferings two marches
from Dandan-Uiliq.
The last I heard of the ill-fated animal was a request which
Turdi addressed to me, before I finally left the ruined site, and
which showed how curiously the rules of the local administration
would affect the quondam owner even far away in the desert.
Turdi wanted to have at least the pony's skin carried back to
Tawakkel by his companion and sold there. In order to save
the tax which would otherwise have to be paid to the local Beg
on account of this sale, I was obliged to endorse in due form the
quaintly-worded and still more queerly penned Turki applica-
tion setting forth all the sad circumstances of the case which
Turdi got written out by the hand of Niaz, our
'
Mullah.' I
thought at the time how puzzling a document it would be for
an archaeologist who might have the good fortune to light upon
It in the desert sand some two thousand years hence !
288
CHAPTER XX.
DISCOVERY OF DATED DOCUMENTS.
Christmas Day was spent in clearing a group of ruined struc-
tures situated about half a mile to the north-east of my camp,
in which I could without difficulty recognise the remains of a
square temple cella and of an adjoining dwelling-place, probably
a monastic habitation. These ruins had suffered badly from
erosion, which, in the unprotected soil immediately to the north
and east of them, had produced broad depressions to a depth
of about twenty feet below the original ground level. The ruins,
owing to this lowering of the surrounding ground, seemed now
to occupy a raised tongue of land quite clear of dunes, and nowhere
retained more than two or three feet of covering sand. Above
this there rose the splintered short stumps of posts that once
formed the framework of wattle and plaster walls, their rows
clearly marking the original division of the rooms. The exposed
condition of these ruins had, of course, attracted the visits of
"
treasure-seeking
"
parties, including some Turdi had person-
ally conducted in former years, and the debris of plaster, timber,
ancient pottery, etc., scattered about on the surface plainly told
of their burrowings.
Notwithstanding the damage thus caused, there remained
some very curious relics to reward my careful clearing. In tho
western part of the quadrangular passage that enclosed the cella
we
found two painted panels of wood, showing on both sides
representations of sacred personages and undoubtedly the votive
offerings of some pious worshipper. On the larger and better
I preserved of these panels, which measures i8 by
4
inches, thei>
AN UNKNOWN l.ANGUAGE. 289
appears seated between two attendants a half-length human
figure with the head of a rat, and wearing a diadem. It was
only long afterwards, when the little painting had been cleaned
of its adhering layer of sand in the British Museum, and ex-
amined by the trained eye of my friend, Mr. F. H. Andrews, that
I reahsed the pecuhar shape of the figure and its true signifi-
cance. It is manifestly meant to represent the king of those holy
rats which, according to the local legend related by Hiuen-
Tsiang and already referred to, in connection with the Kaptar-
Mazar
(p.
i8o), were worshipped at the western border of the
Khotan oasis for ha\'ing saved the kingdom from a barbarian
invasion. The sacred character of the rat-headed figure is
suf&ciently marked by the semi-eUiptical vesica or halo which
encloses it, and by the worshipping attitude of the attendant
figure on the left, which carries in one hand a long-stemmed,
leaf-shaped fan or punkah.
In a comer of the temple-cella, close to the floor, there turned
up two scraps of thin water-lined paper, shouing writing on one
side only, and that in characters which I could at once recognise
as belonging to that peculiar cursive form of Brahmi already
known to us from certain ancient documents in a non-Sanskritic
language that had reached Dr. Hoernle's collection through pur-
chase from Khotan. On clearing the largest of the rooms in the
ruined
dwelhng-house adjoining the shrine, I found several small
sheets of the same coarse paper and with similar writing, either
crumpled
up or folded into narrow rolls, just like the Chinese
documents
I subsequently unearthed at this site. It was no
easy
task to open out these flimsy papers with fingers half-
benumbed
with cold, and the more dehcate part of such work
was
accomplished only in the British Museum. But the
cursory
examination that was possible on the spot showed that
these
more or less fragmentar}* sheets could not have belonged
to
manuscript books or
'
Pothis,' but evidently contained
detached
records of some kind.
The
impression I had gained from the outward appearance of
these
and similar documents in cursive Brahmi found in other
ruins
of Dandan-Uiliq, has been fully borne out by the result
of
Dr.
Hoernle's painstaking researches, as since pubhshed in the
19
290 DISCOVERY
I
OF DATED DOCUMENTS.
second part of his
"
Report on the British Collection of Anti-
quities from Central Asia
"
(1902).
The materials upon which
that eminent scholar worked comprised a considerable number
of well-preserved documents of this type which had been pur-
chased in the years
1895-97
by Mr. Macartney and Captain
Godfrey from Badruddin, the Afghan Aksakal of Khotan.
Internal evidence, as well as the information secured by me,
makes it highly probable that these documents represent chance
finds made by Turdi during his earlier visits to Dandan-Uiliq.
Their minute analysis has enabled Dr. Hoernle to establish a
series of philological facts which are of great interest, and possess
considerable importance also from a historical point of view.
He has succeeded in determining a number of words, either
names, or terms, or numerals, which
"
seem to prove clearly
that the language of the documents is an Indo-Iranian dialect,
having affinities both with Persian and the Indian vernaculars,
in addition to peculiarities of its own," pointing towards a con-
nection with the so-called Ghalchah dialects of the Pamir region.
He has also ascertained the interesting fact that the majority
of the complete documents are fully dated, though the key to
^
the chronology has yet to be discovered.
f
A number of ingenious observations, such as the discovery of
lists of names at the end of certain documents, accompanied by
what manifestly are the marks of witnesses, support Dr. Hoernle's
conclusion that we have in them records of official or private
transactions similar in character to the deeds of loan, requisi-
tion orders, etc., which are contained in the Chinese documents
from Dandan-Uiliq to be described below. The detailed exami-
nation of my finds of this kind which Dr. Hoernle has very kindly
undertaken has not proceeded sufficiently far to throw further
light on the interesting questions thus raised. But the certainty
which exists as to all circumstances attending the discovery of
the documents contained in my collection has already helped us
to settle definitely the period to which these records in cursive
Brahmi
belong.
I
Dr.
Hoernle, judging from palfeographic evidence, had sul
gested the eighth century of our era as their probable date.
The
correctness of this approximate dating is now fully proved
FINDS OF CHINESE RECORDS. 291
by the fact that I found some of these documents mixed up in
the same place and conditions with the Chinese records to be
mentioned below which bear definite dates ranging from
781
to
787
A.D. If we are right in supposing for these documents in
cursive Brahmi such practical contents as above indicated, it
follows with great probability that their language was that
actually spoken by the inhabitants of the ruined settlement
immediately before it was abandoned. It still remains to be
ascertained whether this language is identical with the unknown
tongue already mentioned above which appears in some frag-
mentary manuscript books or Pothis from Dandan-Uiliq written
with Brahmi characters. Judging from certain Sanskrit terms
found interspersed in the latter texts, it seems probable that they
treat, like the Sanskrit Pothis with which they were found, of
Buddhist religious matters. They may possibly prove to be
translations of Sanskrit treatises from the Buddhist Canon into
a Central-Asian language that had obtained hterary use in the
Buddhist Church of these regions.
The room which had yielded those paper documents in cursive
Brahmi held nowhere more than three feet of sand. Yet, even
this scanty layer had sufficed to protect a variety of interesting
remains, found mainly at the foot of the walls adjoining the
south-west corner. Two small oblong tablets of thin wood,
rounded off on the right end and provided with a string-hole,
proved to contain several lines of the same cursive Brahmi
script already discussed. Another and somewhat larger tablet,
about fourteen inches long and three inches broad, at once
attracted my interest by showing in
its shape and handle the closest re-
semblance to the
*
Takhta,' that
traditional wooden board which in
ANCIENT
'TAKHTA' FOR
all uativc schools of Northern India
WRITING.
mis the place of our slate. This tablet
was found blank, but the marks of
plentiful
scraping plainly show that it had once been used for
writing.
In the light of subsequent discoveries we must look
upon these few tablets, just like the Indian
'
Takhta
'
itself,
as
quasi-archaic
survivals. But at the time of their discovery
19*
292 DISCOVERY OF DATED DOCUMENTS.
I little suspected what much more extensive finds of the same
ancient writing material were awaiting me elsewhere.
The elaborate floral decoration on a portion of a lacquered and
painted wood bowl which I found in the same corner clearly
betokened Chinese work. And very soon after, as if to confirm
a conjecture as to the dwellers in this room, the first finds of
Chinese documents rewarded my search. One consisted of a
stick of tamarisk wood, about fourteen inches long and one inch
wide, partly flattened on two sides, on each of which there
appear in vertical lines about a dozen Chinese symbols. The
ink of most has badly faded, and no certain interpretation of
the few clear characters has as yet been obtained, though it
seems probable that the stick had once been used as a kind of
tally, making mention of a certain load. More important from
a historical point of view is the second document, a sheet of
thin water-lined paper, originally folded up into a narrow roll
and recovered almost complete.
According to the provisional translation which Mr. Macartney
kindly supplied to me at Kashgar, and which has been con-
firmed in all essentials by Professor Chavannes, of the College
de France, the main purport of this paper is a petition for the
recovery of a donkey which had been let on hire to two indivi-
duals, who after a lapse of ten months had failed to come back
or to return the animal. It is precisely dated on the sixth day
of the second month of the sixteenth year of the Ta-li period,
which corresponds to
781
a.d. The locality from which the
petition originates is referred to by a name which, owing to
certain doubts as to the phonetic value of the two characters
composing it, may be variously read as Li-sieh, Lieh-sieh, or
Li-tsa. Now this find is of special value, not merely because
it
supplies an exact date but also because, in conjunction with thn
Chinese documents which Mr. Macartney had obtained in
189^
through Badruddin, and which have been jniblished in Dr.
Hoernle's Second Report already quoted, it makes it possiblt
to fix with great probability the name of the settlement reprr
sented by the ruins of Dandan-Uiliq, as well as the Chinese
administrative division to which it belonged.
The three documents 1 refer to in general appearance and style
ANCIENT NAME OF DANDAN-UILIQ. 293
of writing closely resemble those excavated by me at Dandan-
Uiliq, and are, like the latter, official records of a public or private
character. The translations with which Mr. Macartney kindly
supplied me at Kashgar show that the first of them, dated in the
third year of the Ta-li period, corresponding to a.d.
768,
con-
tains the draft of a report from the officer in charge of Li-sieh
(Li-tsa) on a petition from the people of that locahty. The
report recommends a postponement of the collection of miscel-
laneous taxes in view of the distress caused by the depredations
of robbers. Another document, dated only by month and day,
is a military requisition sent from the Li-sieh camp to a civil
authority for a skin to re-cover a drum and for feathers to re-fit
arrows. The third, dated in the seventh year of the Chien-Chung
period, corresponding to a.d.
786,
records the issue of a loan of
15,000
cash on the security of a house in a village (name not
deciphered) belonging to Li-sieh
In \'iew of the close agreement shown by the dates and con-
tents of these documents with those of the Chinese papers which
I discovered in various ruined buildings at Dandan-Uihq, it
appears to me practically certain that they represent some of
the finds of Chinese manuscripts which Turdi well remembered
to have made on a visit to the site some years previously, and
which with other
"
old things
"
he had sold to Badruddin, his
usual employer at Khotan. It is possible that these particular
documents came from one of those rooms in the ruined house
D.V. which I found already thoroughly searched, or from some
I
other ruins that had similarly been
"
explored " before by .
I
Turdi's parties. In any case their comparison with the first
i
Chinese document I unearthed at the site, the petition (D.V.
6)
I already referred to, leaves no reasonable doubt as to Li-sieh
I (Li-tsa) being the name of the settlement or small tract to which
the ruined shrines and dweUings of Dandan-Uiliq belonged.
It makes it further clear that the administrative di\'ision in
^
which this settlement was included bore a Chinese designation
meaning
"
Six Cities
"
(Liu-Cheng)
;
for this identical termjs
used in the title of the
"
Inspecting Superintendent of the Six
Cities
"
to which the report preserved in Mr. Macartney's first
document
was addressed, and it also occurs in my first find imme-
294 DISCOVERY OF DATED DOCUMENTS.
diately after the name of Li-sieh. According to the informa-
tion supplied by Sun, Mr. Macartney's Chinese literatus, and
verified also from other sources, the term
"
Six Cities
"
is still
well known by Chinese administrators in the
"
New Dominions
"
as an old designation of the Khotan District. It is supposed
to be derived from the six towns (Ilchi or Khotan, Yurung-kash,
Kara-kash, Chira, Keriya, and a sixth of doubtful identity)
which were reckoned to belong to it previous to the modern
division of the Ambanships of Khotan and Keriya.
Those few interesting finds, together with hours of bright
sunshine that gave relief after the bitter cold of the night, had
helped me to pass Christmas Day in good heart. In the evening
on leaving the buildings excavated, I had occasion to learn by
experience how easy it was to lose one's way amidst the mono-
tonous sand-waves. The discovery of a well-preserved Chinese
coin bearing the mark of the Kai-yuen period (a.d.
713-741),
which I picked up at the foot of a dune close to the south of the
ruin, and the search for more relics made me tarry behind the
men who were hurrying back. Remembering how near the
camp was, I did not think it necessary to retain any one with
myself. When after a while I set about to return in the twilight
I mistook the track, and then after tramping through the low
dunes for about a mile vainly attempted to locate my cam]\
There was no sound nor any other indication to guide me.
Realising the risk of completely losing my bearings as it was
getting dark (my magnetic pocket compass had unfortunately
been left in the tent that day), I was just about to retrace my
footprints while I could still distinguish them, when I suddenly
recognised sticking out from the sand some remains of walls
which days before I had noticed at a considerable distance to the
south-east of my camping-ground.
Trusting to my recollection of their relative position, I turned
off to my right and, keeping along the crest-line of the dunes
which I knew to be running mainly from north-west to south-east,
made my way slowly onwards until I heard my shouts answered
by some of my men. Old Turdi and Islam Beg, my faithful
Darogha, had grown uneasy at my absence, and had sent the
men out in couples to search for me. The shelter of my tent and
THE HU-KUO MONASTERY. 295
I
the hot tea that awaited me were doubly welcome after this little
incident. There was nobody to share my Christmas dinner but
*
Yolchi Beg,' my sociable little terrier, who was ever ready to
let his own dinner outside the tent get frozen hard while sitting
up amidst the rugs of the bed for choice bits from my table. I
sometimes doubt whether even the friends whose kind thoughts
turned towards me that evening from the distant South and West
could realise how cheerful is the recollection of the Christmas
spent in the solitude and cold of the desert.
The ruined structures which had helped to direct me that
evening were the next to be excavated. Curiously enough, the
finds made in them formed the best complement of the results
of the previous day's work. A small Buddhist temple, con-
structed in the usual style of these ruins, with square cella and
enclosing passage, was first brought to light, and furnished a
number of interesting frescoes, as well as some painted panels
and manuscript fragments in cursive Brahmi characters. When
subsequently the ground floor rooms of a small dwelling-house
(D. \TI.) close to the north of the shrine were cleared from the
deep sand that filled them, we came on the floor of the central
room, measuring about i8 by
13
feet, upon quite a small collec-
tion of Chinese documents on paper. They were all folded into
narrow rolls, just as the one found in the ruin D.V., and lay
scattered about on the ground near the well-preserved fireplace,
ther separate or sticking together in little packets. The leaves,
ui which the rolls had been made up, proved on the average
eleven inches high.
Owing to the damp that must have once reached them through
the mud floor some of these rolls had decayed in parts. But
others were recovered more or less complete, and though the
translations of four of these, which I obtained through the
kindness of Professors Chavannes and Douglas, are only pro-
visional,
they amply suffice to settle all doubts as to the date and
character of the records. Two of them, dated in the third year
of the Chien-Chung period, corresponding to a.d.
782,
are bonds
for small loans specified in copper cash or grain issued on interest
to
different borrowers by one Chien-3dng, who is designated as a
priest of the Hu-kuo monastery. As security for these loans the
DATE OF ABANDONMENT. 297
borrowers, whose names and ages are appended together with
those of certain relatives as sureties (mother and sister, wife and
daughter), pledge the whole of their household goods and cattle
irrespective of any valuation. A third document, dated in the
year equivalent to a.d.
787,
records an agreement of similar
purport, without clearly showing the name of the lender. That
this unknown capitalist was also in some way connected with the
Hu-kuo monastery becomes very probable from the contents of
the fourth paper (reproduced already in my
"
Preliminary Re-
port
").
This, undated, conveys instructions from the three
priests superintending the Hu-kuo convent to the monk or care-
taker in charge of some outlying landed property of theirs. He
is therein directed how to carry out certain urgent field-labours,
employing all available men in cutting grass for three days after
receipt of the order, while one labourer is to be retained to irrigate
the fields, etc.
From the nature of these petty records it can safely be con-
cluded that the ruined dwelling-house in which they were found,
together with the adjoining shrine, represents either the Hu-kuo
convent, or a monastic residence directly attached to it. The
Chinese designation of the convent (Hu-kuo, literally
"
country-
protecting
")
and the Chinese names of the superintending priests,
|
which are recorded in the last-named document, leave little I
doubt as to the nationality of the monkish establishment. But
/
that the population which supported it was not Chinese is plainly I
indicated by the transcribed names of the borrowers and their
sureties, as well as by the short inscriptions in cursive Brahmi
'_
found beneath some of the frescoes of the temple. The more
learned of the monks may be supposed to have been versed also
in Sanskrit, the sacred language of the Buddhist Church through-
out the North
;
a small fragment of a Sanskrit
'
Pothi ' was, in
J
fact, among the manuscript finds of this ruin.
^
The very pettiness of the affairs recorded in the Chinese papers
of this small convent increases to no small extent their value from
a chronological point of view. Unimportant in character and
insignificant in size and material, it is highly improbable that
these
documents should date back to a period preceding by any
great length of time the final abandonment of the building. Now
L
298 DISCOVERY OF DATED DOCUMENTS.
it deserves to be noted that all the papers from this ruin which
can be dated with accuracy, belong to the years
782-787
a.d.
Taking into account that the first-described Chinese document,
found in the ruined building D.V. under exactly similar con-
ditions, also bears the date of a.d.
781,
we are almost by necessity
led to the conclusion that the settlement to which these shrines
and dwelling-houses belonged was deserted about the close of
the eighth century of our era. In each case the papers were
discovered closely adhering to the original floor, which proves
that the sand must have entered the rooms very soon after
these petty records had been scattered about there. For light
and flimsy as they are, the little paper-rolls could not have
resisted very long the force of the storms that pass over the
country each successive spring and summer.
T It is a particularly fortunate circumstance that such unmis-
\ takable chronological evidence has been obtained in the very same
structure which has furnished us with the best preserved, and
! perhaps the most interesting specimens of contemporary paint-
i ing from this site. The three painted panels of wood I refer
I
to were found lying in the loose sand a few inches above the
/
flooring, and not far from the east wall of the central room. From
[
this position, and from the dowels still clearly marked on the
'
back of two of them, it is evident that these pictures had once
been fixed high up on the wall, from which they dropped only
when the little monastic building was gradually being filled with
sand. This accounts for the remarkable preservation of the
colours and the wood underlying them. No verbal description
can take the place of adequate reproductions in colour which
it has been impossible to provide for the present publication.
But until such will be made accessible in my Detailed Report,
now under preparation, I may at least draw attention to one of
these pictures which by its subject presents unusual interest, and
which can already be seen on a reduced scale in one of the
coloured collotype plates illustrating my
"
Preliminary Report."
This panel, which has a rectangular shape, with pointed
arched top, is fifteen inches high and seven inches broad. It
shows two figures, both mounted and manifestly of high rank,
one above the other. The upper figure is seen riding on a high-
PAINTED PANELS. 299
stepping horse, the colour of which, white with large spots of
black, curiously recalls the appearance of the piebald
'
Yar-
kandi ' horse, which, until very recent times, was so much fancied
by natives of Northern India. The rider, whose handsome,
youthful face shows an interesting combination of Indian and
Chinese features, wears his long black hair tied in a loose knot
at the crown, while a yellow band passes round the head holding
in front a large elliptical jewel. The long pink tunic, and the
narrow light scarf that descends from the back of the head, with
its two ends floating behind the arms to indicate rapid move-
ment, are drawn with the same care and freedom of outline as
the rest of the details. The feet are cased in high black boots
with felt soles, very much like those still worn by men of means
in Chinese Turkestan, and are placed in stirrups. While the
left hand holds the rein, the right raises a patera, towards which
a bird is shown swooping down in full flight. From the girdle
hangs a long sword, nearly straight, and of a pattern that
appears early in Persia and other Muhammadan countries of the
East.
The horse, which is remarkably well drawn even to its legs and
hoofs, carries a deep and narrow saddle over a large
'
Numdah
'
or felt-cloth, and shows elaborate trappings. These include a
single bridle, a surcingle, breastplate and crupper, as well as a
large ornamental plate that covers the forehead and nose. Two
curious horns project from this plate, the one at the forehead
carrying what might be the Trisula, or Indian trident, while the
other above the noseband is surmounted by what looks like a
mango-shaped tassel in red silk, represented also on other parts
of the trappings. We could not have wished for a more accurate
picture of that
"
horse miUinery
"
which in the eighth century
evidently flourished throughout Turkestan as much as it does
nowadays. No less interesting is the representation of the
second figure below, riding on a two-humped camel, shown in
full movement and with striking fidelity to nature.
~^
The rider, whose face is partly deleted, wears over his short curly
hair a curious sugar-loaf hat, with its broad brim turned up into
*
Vandyke
'
points. Marks on the hat indicate some spotted
fur
as its material. The long and loose-fitting green garment
300 DISCOVERY OF DATED DOCUMENTS.
worn by the rider is gathered below the knee into the wide tops
of red boots, or mocassins without soles, closely resembling the
'
Charuks
'
still used through the whole of Eastern Turkestan,
particularly during the winter months. While the left hand is
guiding the camel by a nose-cord, the right, in the same pose as
that of the rider above, raises a shell-shaped cup. The elaborate
fittings of the saddle and the stirrups show that the animal
bestridden by this personage is meant for a riding camel, a mount
rarely used nowadays in these regions. Some freely drawn
contour-lines visible behind the camel indicate a hilly country,
or else high ridges of sand. The nimbus painted round the head
of each of the mounted figures shows plainly their holy character,
and the identity of their attitude leaves no doubt as to their
connection in what evidently was a sacred legend. But as to
the nature of this legend, which, as already stated, appears also
elsewhere in the decoration of Dandan-Uiliq shrines, no clue has
as yet been traced.
It is impossible to give here either reproductions or detailed
accounts of the other panels, one of them painted on both sides,
recovered from what we may briefly call the refectory of the
'
Hu-kuo
'
monastery. The paintings of Buddhas or Bodhi-
sattvas which they contain must prove of exceptional interest
for the study of the history of Buddhist art in Central Asia.
For, apart from the high opinion which their good drawing,
graceful composition, and free execution must give us of the
merits of the Indian art transplanted to Khotan, these paintings
also strikingly illustrate the early development of the elaborate
conventional details in emblems, poses, etc., which are so charac-
teristic for all representations of the Buddhist Pantheon in
Tibetan art. As the latter is so far known to us only from com-
paratively late specimens, and as scarcely any remains exist of
early Indian paintings illustrating Buddhist worship of the
Northern or so-called Mahayana type, the value can scarcely
be over-rated of such well-preserved and approximately dateable
pictures treating subjects thoroughly familiar to that system of
worship.
But to me perhaps even more curious, because certainly
less
expected, are the traces of the influence of Persian art which
BUDDHIST PICTORIAL ART. 301
appear unmistakably in the treatment of face and dress on some
of these and subsequently discovered paintings. The evidence
of this remarkable fact will be duly furnished by means of proper
illustrations in my scientific pubUcation. In the meantime,
however, I may point out that this connection with later Iranian
art finds its exact parallel in the still more remarkable influence
of classical art imported from the Far West, of which my sub-
sequent discoveries have revealed such tangible proofs for a
much earUer period.
The description so far given of the structures I excavated
during the first part of my stay at Dandan-Uihq will suffice to
indicate the general character and contents of the ruins of this
site. The survey and excavation of other detached shrines and
dwelling-places scattered amidst the dunes kept me for another
week busy from morning till evening. It brought the number
of buildings that were thoroughly cleared and examined up to a
total of fourteen. But though the results of these excavations
helped to confirm and supplement my former obsersations, they
did not bring to light novel features requiring detailed mention
here. In regard to materials and mode of construction all the
buildings resembled each other closely, showing plainly that they
had been built and probably also deserted in approximately the
same period. But their state of preservation greatly differed,
according to the measure in which they had been exposed to the
erosive action of the winds and the equally destructive diggings
of
"
treasure-seekers." In some instances I came upon unmis-
takable evidence of their operations even in structures that had.
since been covered up again by deep sand.
Thus in a small temple-cella the well-moulded octagonal base
of the principal image had been dug into from behind in search
of
supposed treasure. Yet the votive tablets that had been
leaning
against its foot in front were left untouched, and equally
so a fairly thick packet of manuscript leaves comprising probably
more than one
'
Pothi.' But, alas ! the damp rising from the
floor
when water still reached this neighbourhood had here done
its
work of destruction far more effectively than the
"
treasure-
seekers
"
ever could. The leaves of paper had decayed com-
pletely
into compact layers, which could not be detached from
302 DISCOVERY OF DATED DOCUMENTS.
the hardened crust of what once was dust enveloping them.
Ultimately these lumps of earth had to be cut off as a whole
from the ground to which they adhered
;
but even with the ut-
most care in slicing, the flakes of paper since laid bare reveal
only fragmentary groups of Brahmi characters. The upper
board of wood which once covered one of these books, and which
I was able to remove in a half-rotten condition, figures now as
an instructive relic in my collection of ancient manuscripts and
writing implements.
But opportunities for interesting antiquarian observations
were by no means restricted to the finds which the excavation
of extant ruins yielded. A careful survey of their surroundings
also taught much that helps us to reconstruct the general aspects
and conditions of the life which once flourished here. I have
already referred to those strange witnesses of ancient orchards
and avenues, the gaunt, splintered trunks of poplars and various
fruit trees, which could still be seen, half-buried in the sand, near
most of the buildings. On some patches of the original ground
left uncovered between the moving dunes the traces of old irri-
gation channels, running between small banks of earth, and
evidently constructed after the fashion that still prevails in the
country, were easily recognisable. But owing to the dunes or
else the effects of erosion it was impossible to follow them for
any distance.
In many places between the scattered ruins, the ground was
thickly strewn with fragments of coarse pottery, small corroded
pieces of metal, and similar debris. These remains, found in
places where at present no traces of old structures survive,
probably mark the positions occupied by less pretentious dwell-
ings which, like the houses of common Khotan cultivators of the
present day, were built wholly of sun-dried bricks or stamped
clay. These were likely to crumble away far more quickly
than buildings with a timber frame-work covered by hard plaster.
The latter
mode of construction also is still used in the towns
and villages of Khotan, but being far more expensive, owing to
the
distance from which wood has to be brought, it is restricted
to the
houses of the well-to-do and to Mosques, Sarais, and
similar
buildings. This observation helps to explain, at least
I
ABANDONMENT OF SETTLEMENT. 303
partly, why, at sites like Dandan-Uiliq which must for various
reasons be supposed to have been occupied by comparatively
large settlements, the extant structural remains are hmited in
number and so widely scattered.
But the striking preponderance of rehgious buildings among
the Dandan-Uiliq remains also suggests the possibility that these
local shrines and their small monastic establishments continued
to be kept up and visited, perhaps as pilgrimage places, for some
time after the rest of the settlement had been abandoned. The
conditions in which Muhammadan Ziarats are now often found
beyond the present cultivated area of oases, would furnish an
exact parallel. In this case the complete decay of the deserted
village structures was likely to have been accelerated by the
demands that the attendants of the shrines, as well as the pil-
grims, would necessarily make upon them for whatever in the
way of wood and other useful materials had remained in them.
However this may be, it must be considered as certain that the
abandonment of the settlement was a gradual one, and in no
way connected with any sudden physical catastrophe such as
some European travellers have been only too ready to assume,
'^n account of popular legends they had heard about the so-called
ancient cities" of the Taklamakan. The Sodom and
Gomorrah legends related all over Eastern Turkestan about
"
old toN\Tis
"
suddenly submerged under the sand-dunes, are
more ancient than the ruins of Dandan-Uiliq themselves. Hiuen-
Tsiang had already heard them more or less in the same form
in which they are now current, as is seen from the story of the
town of
'
Ho-lo-lo-kia,' which we shall have occasion to refer
to in connection with the site of Pi-mo. These legends un-
doubtedly are interesting as folk-lore. But where we have such
plain archaeological evidence to the contrary as the examina-
tion of the Dandan-Uiliq ruins, and in fact of every other ancient
site in this region, has suppUed to me, scientific inquiry need have
no
concern with them.
My detailed survey of Dandan-Uiliq, together with other ob-
servations of a semi-topographical, semi-antiquarian nature
which
gradually accumulated during my explorations at this
and other sites, make it very probable that the lands of Dandan-
304 DISCOVERY OF DATED
DOCUMENTS.
Uiliq were irrigated from an extension of the canals which, down
to a much later date, brought the water of the streams of Chira,
Domoko and Gulakhma to the desert area due south of the ruins.
The debris-covered site of Uzun-Tati, which I discovered there
amidst the sand-dunes, is identical with the
'
Pi-mo
'
of Hiuen-
Tsiang, Marco Polo's
'
Pein,' and can be proved by unques-
tionable evidence to have been occupied for at least live cen-
turies longer than Dandan-Uiliq. A number of historical as well
^
as topographical considerations, for a detailed discussion of
^
which I must refer to my scientific publication, point to the con-
clusion that the successive abandonment of both Dandan-Uiliq
and
'
Pi-mo
'
was due to the same cause, the difficulty of main-
""
taining effective irrigation for these out-lying settlements.
I cannot attempt here to investigate the question to what
extent this receding of the cultivated area may be attributed to
neglect of irrigation works, caused probably by political troubles
and consequent depopulation, or to a change in the physical
conditions attending the supply of water from those streams. I
may, however, with advantage call attention here to my subse-
quent observations at certain villages of the Gulakhma and
Domoko oases, the cultivated area of which has, owing to the
difficulty of carrying the irrigation water sufficiently far, been
shifted, within the memory of living men, as much as six to
eight miles further to the south. The crumbling ruins of the
deserted village homesteads which I saw there, stripped of
materials that could be of use, and the miles of once cultivate(
ground which the desert sand is now slowly over-running, butj
on which the lines of empty canals, irrigation embankments
etc., can still be made out, were the best illustration of the process!
by which the lands of Dandan-Uiliq became finally merged in!
the desert.
In this connection I may note that our survey furnished nc
evidence in support of the assumption put forth by Dr. Svei
Hedin, that the Keriya Darya in historical times flowed close t(
Dandan-Uiliq,
and that the abandonment of the site was con-
nected with the subsequent shifting of the river to its present
bed, some
twenty-eight miles in a direct line further to the
east.
Probably,
the distinguished explorer would have hesiis
STOPPAGE OF IRRIGATION. 305
tated to make this suggestion had he known the indisputable
antiquarian evidence which shows that the ruins to which,
mainly on the basis of conjectural calculations as to the move-
ment of the sand-dunes, he was prepared to assign an age of
about two thousand years, were in reality abandoned only
about the close of the eighth century of our era.
20
306
CHAPTER XXL
THROUGH THE DESERT TO KERIYA.
On the 3rd of January, 1901,
the explorations
at Dandan-
Uihq were completed. The previous evening my long-expected
mail had arrived from Kashgar, a heavy bag this time with
the postal accumulations of some six weeks. The latest of
the letters and papers sent from Europe via India dated from
the
beginning of October. One of the most welcome letters
was a
communication from the Indian Foreign Office, whicli
informed me that the request I had made some nine months
before from Calcutta, to be allowed eventually to return through
Russian
Turkestan, had received the sanction of the Russian
authorities. The safe packing of my fragile antiquarian finds,
and the making up of my own mail kept me busy all day and
as long as work was possible in the tent. The camels had,
in accordance with previous instructions, duly arrived from
the river, where they had managed to gather fresh strength
even on the scanty fare offered by the wintry jungle. From
the unusual animation with which the preparations for tlic
start were proceeding, it was easy to see how much all my men,
from Ram Singh downwards, enjoyed the prospect of saying
good-bye to this trying camping-ground. So there was some
disappointment when they learned that before altogether
leaving the desolate neighbourhood I intended to visit some
ruins of which Turdi had spoken as situated to the north and
known to treasure-seekers by the name of Rawak
("
High
Mansion
").
On the morning of tlie 4th of
January I paid off and dismissed
REMAINS OF RAWAK. 307
to Tawakkel a portion of my little force of labourers who had
worked so valiantly. With the rest I set out to the north,
and after a march of about seven miles across gradually rising
sand-ridges reached again ground where broken pottery between
the dunes indicated the former existence of habitations. We
halted at a spot between deeply eroded banks of loess, where
Kasim's party eighteen days before had camped and found
water. But the well would yield no water now, as the ground
was frozen quite hard, and when at last water was reached
by fresh digging it proved even more brackish than that we
had to drink at Dandan-Uiliq.
Though Turdi had not visited the place for nine years he
guided me on the next day without hesitation to where behind
a long-stretching ridge of sand, some 60 feet high, the ruins
were situated. They proved to consist mainly of two much-
decayed mounds, lying quite close together, composed of fairly
hard sun-dried bricks, probably the remains of small Stupas.
They had evidently been dug into repeatedly and had suffered
badly ; but in the case of the larger one it was still possible to
make out what looked hke a circular base about
32
feet in
diameter. From among the small debris of ancient pottery,
broken glass, etc., strewn over the ground near the mounds I
picked up a fragment of remarkably hard stucco on which the
practised eye of Turdi at once discov^ered traces of a thin
gold-layer.
Judging from its shape this stucco piece is
likely to have belonged to a statue that had once been fully
gilded.
Old Chinese coins without legend, as issued under the Han
dynasty, also turned up at various places among the pottery
debris which covered the low ground between the dunes. As
the latter rise here to heights over
25
feet and are proportionately
large, it was scarcely surprising that we could trace the ruins
of only one house built with timber. Its walls had decayed
by erosion to within a few feet of the ground, and the high
dune rising immediately above it made it impossible to clear
more than a single room. Within it and close to^the floor
we found two small wooden tablets of oblong shape, inscribed
on one side with cursive Brahmi characters. The socket which
20*
308 THROUGH THE DESERT TO KERIYA.
appeared on the back of one of them was proved by subsequent
discoveries elsewhere to have once held a clay seal.
The fact that only Han coins were found here, as well as other
indications, make it appear probable that Rawak was deserted
a considerable time before Dandan-Uiliq. But until the peculiar
physical conditions of the various parts of the Taklamakan, and
particularly those concerning the movement of the sand-dunes,
have been systematically studied for a prolonged period, it would
be hazardous to draw conclusions as to the rate of progress
in the general advance of the desert southwards. And even
when such observations are availableand I for one shall
respect the devotion of those who may hereafter charge them-
selves with their collection on this forbidding groundit is
very doubtful whether their results could be relied upon to give
a true view of the conditions prevailing at earlier periods.
The examination of the scanty remains at Rawak completed
the task for which I had set out just a month previously from
Khotan. So on the morning of the 6th of January I began
the march to the Keriya River after paying off Merghen Ahmad
with the last batch of the Tawakkel labourers. They parted
from me in good spirits, well satisfied with the reward their work
had earned them, and evidently none the worse for their long
camping in the desert. Islam Beg, too, who had managed
these people so well, now left me for Khotan, glad to regain
once more the comforts of a warm homestead. He was to carry
to the Khotan Amban my news and thanks for his help which
had rendered the exploration of Dandan-Uiliq possible
;
also
my mail-bag was entrusted to him to commence its long
journey westwards.
It was /with mixed feelings that I said farewell to the sileiii
sand-dunes amidst which I had worked for the last three weeks.
They had yielded up
enough to answer most of the questions
which arise about the strange ruins they have helped to preserve,
and on my many walks across these swelling waves of sand I
had grown almost fond of their simple scenery. Dandan-
Uiliq was to lapse once more into that solitude which for a
thousand years had probably
never been disturbed so
long
as during my visit. For me the recollection of this fascinating
FORMIDABLE SAND-DUNp:S. 309
site will ever suggest the bracing air and the unsullied peace
and purity of the wintry desert.
The day was cloudy and as my caravan slowly moved off about
II a.m. a steady breeze met us from the north-east. About two
miles from Rawak camp we passed a broad strip of ground where
broken pottery, glass fragments, etc., cropped up again on the
hard loess banks between the dunes. Beyond, all traces of
ancient habitation ceased, and soon I saw also the last of the
shrivelled dead trees standing in little clumps, the sight of which
had become so familiar to me during these weeks. As the day
wore on the breeze increased sufficiently to treat us to the
spectacle of a gentle dust-storm. The air became filled with a
gray fog, and the dust carried by the wind threatened to efface
the track marked by the footprints of Kasim and Turdi, who
marched ahead as guides. So I had to keep our party, which
now looked much reduced, close together. The sand-hills
rose gradually as we advanced to the east, and I realised that
the apprehensions of old Turdi about the difl&culty of finding
water north of the line which the camels had previously followed
to and from the river, were not without foundation. At the
foot of a great ridge of piled-up dunes we had to halt for the
night, though there was no chance of getting water there. Some
withered roots of tamarisk supphed scanty fuel, but there
was no living scrub to show where we might dig a well. Fortu-
nately a small supply of water, or rather ice, was kept available
for this emergency in the tanks which had already rendered
good service at Dandan-Uiliq for storage.
I had at first intended to steer due east for Gharib-Chakma,
'
which the Sub-Surv'eyor's map showed as the nearest point on
the river. But the thought that in doing so we were likely
to get beyond the line up to which water could be got by digging,
induced me on the next day to change our course to the south-
east, with a view of reaching the track by which Kasim had
previously brought the camels to and from the river. The wind
had subsided during the night, and the haze slowly dissolved
in the course of the day. The individual sand-dunes we passed
were all between
30
and
50
feet high, but the line of march
ed also across three great
'
Dawans ' running from south
310 THROUGH THE DESERT TO KERIYA.
to north. Their height above the Httle valleys between the
ordinary dunes on either side seemed over
150 feet. It was with
a feeling of relief that after having covered about eleven miles
in a straight line and reached the ridge of the third Dawan
we noticed on the easier ground beyond a few sand cones covered
with live tamarisks. Kasim at once declared that water might
be found at their foot. His prediction proved correct. After
digging to a depth of 6 feet through ground which for about
2 feet from the surface was frozen, the men got at water. It
was very salt but none the less most welcome. I in particular
was glad of the wash which I had to deny myself at the previous
camp for the sake of economising the water supply in the tanks.
The camels, too, were glad to get a drink again
;
for heavily
laden as they necessarily were now, they had felt the long march
over these formidable sand-ranges.
After a cold night, when the thermometer fell to
5
Fahr.
below zero, we started early. Everybody was eager to pass
out of the region of sand and reach the river. After about
two miles' marching Kasim's sharp eyes discovered faint traces
of the track which the camels had left in the sand when returning
to fetch us from Dandan-Uiliq, and by following this track
we soon came upon the well previously dug by Kasim's party.
Four Dawans had then to be crossed in succession, each piled
up of terrace-like dunes and apparently between 120 and
150
feet in height. The dunes in the broad valleys between them
sank now to about 20-30 feet
;
yet there was no other indication
that we were approaching the river, until at last from the top
of the last huge ridge of sand the dark line of trees fringing the
Keriya Darya came into view.
Four miles more we tramped on over dunes that showed
broad backs and gradually diminished in height, until a
belt
was reached where tamarisks and
'
Kumush
'
grass was growing
freely. When passing a last low bank of sand, I suddenly
saw the glittering ice of the river before me. While I was glad
to sit down on its bank after the tiring walk of some fourteen
miles, Kasim went to search for the ponies which had brought
the Sub-Surveyor's ])arty from Keriya and were to await
us
here. Half an hour later they turned up under the escort
of
BURHANUDDIN MAZAR. 311
Ibrahim, the
'
Darogha ' whom the Amban of Kenya had
sent to look after my camp. A cheerful fire was then lit under
the poplars that line the river-bank, and by its side I sat
contentedly until the camels turned up in the darkness. It
was pleasant to view in the dusk the high trees still bearing
partly their red autumn foliage, the thick shrubs and the wide
ice-belt of the river, after those weeks when one's eyes had
rested only on yellow sand and the wavy lines of its expanse.
On the following morning Kasim with a single companion left
us to start on the march back to the Khotan Darya, while
I was grateful to get into my saddle once more for the rest
of our march to Keriya. The river along which the route
led was now almost ever>^vhere completely frozen ov^er. It
flows in a deep and extremely tortuous bed about 50-60 yards
across at the narrowest points, but widens at occasional great
bends to fully three times as much. The ground on the left
bank, along which the day's march took us, is covered for a
breadth of about a mile with patches of forest and a belt of
reed-jungle. Beyond stretch the sand-dunes westwards. On
the right bank a high and well-defined ridge of sand, known as
KLzil-kum ("
the Red Sands "), which seems to rise 300
feet
or so, could be seen following the river-course. The growth
of willows and poplars seemed equally luxuriant on either
side.
I passed a number of shepherds' huts
('
Satmas
')
built of a
rude framework of wood with walls of rushes closely packed,
but met no human being until after a ride of about sixteen
miles I reached the shrine which was to offer me shelter for the
night.
The tomb of Saiyid Burhanuddin Padshahim
("
my
Lord S. B.") seems a very popular place of pilgrimage for
the people of the Keriya and Khotan districts, and the com
fortable
quarters and appearance of the five Sheikhs in at-
tendance on the saint's resting-place attest the veneration
enjoyed
by the latter. The Sheikhs were unable to tell me
any
particulars of the holy man's story, except that he was
connected in some way with the still hoUer Imam
J
afar Sadik,
worshipped
at a famous desert shrine where the Niya River
ends;
312 THROUGH THE DESERT TO KERIYA.
The Sheikhs, who receive so many hundreds of pilgrims every
year, know how to prepare for the comfort of
"
paying guests."
So I found a neat little room with felt carpets and a blazing
fire ready to receive me by the side of the saint's tomb. While
waiting for my baggage, which did not arrive till late in the
evening, I had plenty of time to think of the curious inroad
made by civilization, as represented by this sacred establish-
ment, into the solitude of the desert. The shepherds who
frequent the lonely grazing-grounds of the Keriya River, cannot
fail to benefit largely as regards their knowledge of the outer
world by the stream of pilgrims that passes in the autumn
and spring to the local saint's tomb. Is it possible that the
Buddhist shrines I unearthed at Dandan-Uiliq had also been
once the object of similar pilgrimages ?
Three fairly long marches brought me from the
'
Mazar
'
into Keriya. They led along the side of the Keriya River and
through scenery very much like that passed on the first day
after we had struck its bank. Every day we saw some reed-
huts of shepherds, but their occupants seemed to have moved
away from the river. The belt of vegetation grew broader as
we progressed further south, but the thickets of trees gradually
became rarer and most of the ground was covered only with
tamarisk scrub and Kumush. These alone will grow in the
loose sand which the fertilising water is unable to reach. The
spots where we camped for the night, Bulak and Chogalma,
jhowed as little sign of human habitation as the rest of the
jungle we had been traversing. But our guide the
'
Darogha
'
knew the camping-places of the shepherds and always managed
to produce some of this folk to help in collecting fuel. The
days were hazy, and the murky atmosphere made me regret the
clear, bracing air of the desert.
On
January
12th, about noon, I arrived at Bostan Langar, a
tiny hamlet in the midst of a wide, marshy plain where the
river gathers the outflow of numerous springs. Now all the
water-logged ground was hard-frozen, and there was no need
to
follow the turns and bends by which the road avoids quagmires.
At Bostan Langar I was met by Abdullah Khan, an Afghan
merchant from Pishin, who had been settled in Keriya for some
WELCOME AT KERH^A.
313
fifteen years. He was a fine-looking old man and e\ndently
anxious to make himself useful to the
'
Sahibs,' towards whom
he, like his fellow-countrymen all over Turkestan, pretends to
cherish a feeling of allegiance. Unfortunately disease seems
to have played havoc with his constitution, and his utility was
further impaired by a strange confusion of tongues. Persian had
long ago passed from Abdullah Khan's knowledge
;
Turki he
did not appear to have fully learned
;
and Hindustani he heard
so rarely nowadays that conversation in it also presented
difficulties. My knowledge of Pushtu was too scant to permit
my judging how much he remxcmbered of his mother-tongue.
However, the message sent to him in advance had been duly
grasped, and he had accordingly arranged for quarters during
my stay at Keriya.
Soon after meeting this claimant of the
"
Sirkar's protection
"
I was welcomed in the Amban's name by a cavalcade of local
Begs and their followers. The Begs were fat and jovial, and
when they had convinced themselves that I could really talk
their own tongue we kept up quite a lively chat while riding
on towards the town. Etiquette evidently required that they
should meet me in their quasi-Chinese official garb. The fur-
lined little cape of
'
Khitai' fashion was easily worn over
their wann, homely
'
Chappans ' or long coats. But the black
silk cap with the red button of office is a poor head-covering
for a good Turki Muhammadan, accustomed to shelter his
shaven head under a substantial fur cap when the temperature
is so low as it was just then. So my Begs soon compromised
comfort and appearances by making one of their attendants
wear the cap imposed by their Cathay masters, while they them-
selves kept their heads warm with mighty furs.
About four miles from Bostan Langar we reached the edge of
the cultivated area of the oasis. I was once more among the
I
hamlets with their canals and poplar avenues, so tmiform in
appearances all over Turkestan. After a month's life in the
solitude of the desert the bustle of these homesteads was a
ivelcome sight.
'
Yolchi Beg,' my little terrier, also felt this
Jtirring effect of seeing fellow-creatiures once more, and we had
ao little trouble in protecting him from the large village dogs
314 THROUGH THE DESERT TO KERIYA.
which he persisted in provoking by his self-assertive behaviour.
Keriya town presents no very striking appearance even for
the wanderer from the wilderness, and I was far advanced amidst
the low mud-houses of its outskirts before I realised that I
had entered the headquarters of a territory that extends over
some five degrees of longitude.
I was glad to find that the quarters Abdullah Khan had
THE a:mban of keriya. 315
and factotum presented himself to deliver his master's greetings
and presents of welcome. They were all of a thoroughly practical
nature, including firewood, fodder for my ponies, sheep and
fowls for myself, and on so lavish a scale as to render a worthy
return a httle difficult. However, the Amban's kind intentions
were amply testified, and. I felt sure, would not suffer from any
HUANG-DALOI, AMBAX OF KERIYA.
inadequacy
of a traveller's counter-gifts. Etiquette permits
the
feeing
of the bringer of such presents, and I did not fail
to
impress the interpreter with a due sense of my satisfaction.
My
state visit to the Yamen was duly notified for the morrow.
My
interview
with Huang-Daloi, the Amban, passed off most
-tisfactorily.
At i p.m. I rode to his Yamen through the
modest
Bazar of the
'
Yangi Shahr.' The tortuous lanes of
316 THROUGH THE DESERT TO KERIYA.
the equally humble
"
Old Town
"
I avoided by riding outside
along the foot of the high loess-bank on which it is built. The
Yamen of Keriya closely resembles the similar structures I
saw at Yarkand, Karghalik and Khotan. Architectural
etiquette evidently prescribes all details as to the direction
of the gates, the arrangement of the courtyards, nay, even
the position of the table and seats in the reception room. The
salute of three pop-guns solemnly fired at my entry through
the outer gate showed that the Amban wanted to be polite,
and the reception he himself accorded to me at the inner gate
was distinctly hearty.
Huang-Daloi seemed a man of about forty-five, well built
and with a ruddy face which bore a look of good-nature and
humour. He was dressed in elegant Chinese silks
;
the
brocaded yellow petticoat and the fine embroidered centre-
piece of his state jacket particularly attracted my attention.
I found the little table on the raised dais, between the two
seats which host and guest must always occupy, decked with
sweets which looked clean and wholesome, and the customary
cups of tea replaced by little wineglasses of European make.
A kind of Madeira had to serve in place of the orthodox Cathay
beverage
;
whether from a wish to please supposed European
taste or from the Amban's own predilection I did not know.
Though my interpreter, Niaz Akhun, the humorous
Tungani,
had not arrived from Khotan with the ponies I had left behind
on entering the desert, our conversation went on with some ease.
The Amban's intelligence made up for the imperfect compre-
hension his
'
Tungchi ' showed for my Turki. I had to relate to
him at length how I fared at Dandan-Uiliq, and, of course,
took occasion to tell him how all I saw and found there agreed
with the account old
'
Tang-Seng,' i.e.,
Hiuen-Tsiang,
has
given us of Buddhism in these regions, I only wished that I
knew how to talk about Buddhist things through the ordinary
Chinese interpreter, invariably a
Muhammadan
with very
hazy notions on the religious systems en vogtie among his
infidel
masters. When I told the Amban of my wish to visit an ancient
site which had been reported to me north of Niya, the
Ni-jang
of Hiuen-Tsiang, he readily promised the issue of all needful
VISITS OF STATE. 317
orders for help. My thanks and little compliments were always
requited by a smile so cordial and amiable, that the best
diplomatic actor might have envied its expressiveness. When
I left, escorted according to etiquette by the Amban to the side of
my pony, I found the whole of his retinue, down to the scarlet-
dressed executioners, drawn up on the way through the inner
gate. The dresses of the men looked clean and new, and
altogether there was an air of neatness and order about the
place which seemed a reflex of the Amban's personal habits.
I had scarcely left the Yamen when information reached me
that the Amban was starting immediately to return my visit.
So I rode back in haste and just arranged to get tea ready
in time and the little inner room of my airy villa tidied up.
A cover for my camp table was difficult to improvise.
WTiite
is the colour of mourning in China, and hence no ordinary
table-cloth would do. If I visit Chinese territory again I shall
bring a table-cloth of auspicious red. This time a Ught rug
from my bed had to do instead. I thought I was playing at
European court etiquette when I had to receive
as my guest
the host of a few minutes ago. But it was easy enough to
continue a conversation so recently broken off, and volumes
from my little travelling book-case, among them
Stanislas
Julien's edition of Hiuen-Tsiang, helped to entertain. I showed
Huang-Daloi some of the manuscripts found at Dandan-Uiliq
and let him read the legends of the Chinese coins of by-gone
dynasties in my collection. He was too polite to show whether
they interested him as much as my camp table, chair, and bed,
which he also closely examined.
I had originally planned a halt of three days at Kenya, but
the non-arrival of the ponies expected from Khotan caused an
extension of two more days. To tell the truth, all of us were
not particularly sorry for the delay. My men. Ram Singh
included, were glad to have a little respite after the roughing
and exposure in the desert. I myself, the interviews with the
Amban once over, found so much to do in the w^ay of notes to
be WTitten up, accounts to be prepared, etc., that I could scarcely
spare an hour for a daily walk in the dusk. Dandan-Uiliq was
no place for clerical work, and it was only now that I was able
318 THROUGH THE DESEET TO KERIYA.
to write a short account of my work there for the Royal Asiatic
Society's Journal. The last four days at Keriya were cold and
gloomy. On the 14th and 15th of January
there were light falls
of snow, scarcely more than an inch deep each day, yet sufficient
to give a thoroughly wintry aspect to the bare fields along the
river. It was cheerful to have a fire by which to spend the days
of busy work. But I missed in my little den the hght and
'
plein air ' of my camp life. As usual in all houses of Khotan
and the regions around, light is admitted to the inner rooms
inhabited during the winter only by a small square opening in
the roof. It is naturally scanty and ill-distributed.
My inquiries for antiquities made through Abdullah Khan and
others brought me little that was of interest. Keriya is not
itself an old place and the
"
treasure-seeking
"
profession does
not flourish as in Khotan. But on the first day after my arrival
I received information about ancient remains in the desert
north of Niya which decided me to extend my tour in that
direction. Abdullah, a respectable cultivator of Keriya, told
me of having seen about ten years earlier houses evidently of the
same type as at Dandan-Uiliq, half buried in the sand, some
marches beyond the famous Mazar of Imam
J
afar Sadik. Others
too had heard stories of this
"
old town." The eagerly ex-
pected ponies for some unaccountable reason had failed as
yet to arrive. But in a country where there is neither telegraph
nor a real post-office, one learns to take such little delays
calmly, and the Amban's help amply provided for all needful
transport.
319
CHAPTER XXII.
TO NIYA AND IMAM
J
AFAR SADIK.
On the morning of the i8th of January
I started in glorious
sunshine, doubly enjoyable after the confinement of the last
few days. As usual after a halt of some days, my caravan
took a good deal of time to set out again. So there was sufl&-
cient delay to allow half the boys and idlers of
'
Old Keriya
'
to gather in the road ajid on the roofs of surrounding houses
to watch the exciting spectacle. The jovial fat Begs of Keriya
and Niya (the latter away from his charge for the time) duly
saw me off. After crossing the river-bed, about a quarter of a
mile broad, but now all dry but for a modest streak of water,
\ve passed the little villages of Besh-toghrak and Ghadghang.
Scarcely two miles beyond the town we were again in barren
sands, the outskirts of the great desert northwards. On the
right an absolutely bare plain of coarse sand and pebbles leads
up gradually like an enormous glacis to the foot of the mountain-
wall rising to the south. It was the outer range of the Kuen-
luen, east of Polu, usually hidden by the haze from the eyes of
the traveller who follows this ancient route to the Lop-nor
region and the confines of true Cathay. One or two peaks,
which Ram Singh had triangulated on his trip east of Pisha,
were clearly recognisable again and offered safe points for
further survey work. The outer range was completely covered
with recent snow and thus looked more imposing than it
probably does at other times. The high peaks about Polu
and behind, which reach up to 21,000 ft. and more, glittered
dimly in the distance.
320 TO NIYA AND IMAM JAFAR SADIK.
Ui-toghrak, reached after an easy march of about fourteen
miles, is a small oasis of some two hundred houses, scattered
about in hamlets. Under the trees and elsewhere in the shade
a good deal of snow still lay on the ground. The wind all day
was biting cold, and I gladly availed myself of the shelter pre-
pared for my party in the roomy house of a local
'
Bai.' Clean
mud walls and gaily-coloured Khotan felts ('Kirghiz') make
even a bare little room look cheerful and homely on a winter
evening.
Clouds had come up in the evening and stopped the astro-
nomical observations for latitude. On the morning of January
19th it was snowing hard when I got up, and the white trees
of the orchard behind the house looked delightfully European.
The temperature at 8 a.m. was
9
Fahr. The snow stopped
by the time I got my caravan to move off, but all day the
clouds hung low and the mountains were hidden. The ground
traversed was a pebbly
'
Sai ' very much like the soil on most
of the marches to Khotan. High ridges of sand were visible
on the left, stretching away to the north. After a ride of about
sixteen miles we passed the broad and shallow bed of a stream
now completely dry, and a little beyond arrived at the tiny
oasis of Yesyulghun. It consists of about a dozen mud-hovels,
which serve as wayside quarters for travellers to Niya and the
goldfields of Surghak. The few fields irrigated in the summer
would not suffice for the keep of the inhabitants, who derive
their
maintenance from providing quarters and supplies.
The water of the place is obtained from a well, said to be
40
*
Gulach
'
(fathoms) deep, and it was curious to observe how
this form of water supply has affected the topography of the
hamlet. Whereas in Turkestan villages the houses are usually
scattered about among fields and gardens, the dwellings of
Yesyulghun range themselves neatly round the open space
with the well in the centre just as if it were a market-place.
Some
line old poplars growing in a group near by on the edge
of a
storage tank give a j)icturesque look to the spot.
The clouds cleared overnight, and the minimum thermometer
showed
1
Fahr. The mountains immediately south again
became
visible and I could recognise the valley at the entrance
YESYULGHUN AND OVRAZ. 321
of which the Surghak gold mines He. The march was over a
stony steppe and almost too short, only about eleven miles.
But our
'
Darogha ' thought, probably rightly, that the camels
could not cover the whole distance to Niya in one day, and as
there is neither water nor shelter to be found on the remaining
twenty-four miles, I had to acquiesce. Ovraz Langar consists
of a solitary mud-house, tenanted by a
*
Langarchi.' Supplies
and ice had been sent on from Ui-toghrak, so we were fairly
comfortable. The room I occupied was low, and the fireplace
smoked badly ; but even thus it was preferable to a tent in the
sharp wind that blew from the east.
The start for Niya next morning was made cheerful by the
arrival of Niaz Akhun, who, with the ponies and my mail from
Khotan, overtook me before I had left Ovraz Langar. He
brought me letters from home which had travelled by the
Farghana route to Kashgar and thence by Chinese post. The
latest of them had left my brother's hand on the yth of December,
and seemed quite recent, considering that the mail from Europe
that reached me via Gilgit bore dates in the second week of
October. It was impossible to ignore the postal advantages
which the Trans-Caspian railway has secured for the European
in Central Asia, though for safety I preferred to rely on the
Indian post-office and its Dak via Hunza.
The whole of the twenty-four miles' march from 0\Taz
Langar lay over a pebble-strewn
'
Sai,' the detritus washed
down from the great southern range. Here and there sand-
dunes advancing from the desert stretched their last offshoots
across the hard
'
Sai.' There was no trace of vegetation until
we got within about six miles of the Niya oasis, when tamarisks
and some hardy brushwood appeared in small patches. The
oasis of Niya is formed by a series of hamlets and villages ex-
tending along the river that leaves the mountains near Surghak.
In its upper com-se it is known as the
'
Darya '
of the Ulugh-
Sai Valley. After
3
p.m. I had reached the western edge of
the cultivated area at the hamlet of Kang-sarigh, and a further
two miles brought me into the central village containing the
Bazar of Niya. I was received in due form by the local Beg^s
deputy, and found decent quarters prepared in a Bai's house
21
322 TO NIYA A:ND IMAM JAFAR SADIK.
close to the entrance of the Bazar. It was the weekly market
day of Niya, and though it was getting dark in the narrow
street overhung by awnings, there was still busy life in front
of the booths that line its sides for about a furlong. There
were plenty of dried fruits, plums, raisins from Khotan, with
tea and various condiments of Chinese origin. Excellent
walnuts and fair red grapes were local produce. People seemed
busy buying such little luxuries on account of the Id festival,
marking the close of the Ramzan fast.
My people had as travellers taken dispensation from the
observance of Ramzan, yet they were anxious to celebrate the
day in due fashion as good Muslims, and hence asked for a halt
on
January 22nd. I could not well refuse the request, par-
ticularly as arrangements had to be made for the labourers
and the supplies which were to be taken to the desert site I
was bound for. All Niya was in holiday attire, and the prayers
from the mosque sounded sonorously into my room. I was
busy with making up my mails for India and home, but used
the bright midday hours to take photographs of local people.
There were plenty of fine-looking greybeards to choose from,
and no want of nicely-dressed children. Shy at first, the little
ones were readily enticed before my camera by the present of
a few coppers for sweets.
'
Diwanas,' too, or wandering
mendicants, in fantastic rags showing patches in all colours of
the rainbow did not object to giving a sitting in return for my
alms. The rural population here, as at Khotan, shows on the
whole remarkably good featuresof course, Caucasian as the
popular term has it. Noticing the thoroughly European ap-
pearance of physiognomies in the great mass of this Turki
population, I feel inclined to wonder at all the efforts that have
been made to account for the same fact in the Western Turks
and their kindred in Europe.
P
Niya is an ancient place. Hiuen-Tsiang, travelling towards
Lop-nor and China, duly notices the town of Ni-jang, i.e., Niya,
1 which
"
the king of Khotan makes the guard of his eastern
frontier." Niya remained, indeed, the easternmost of the
smaller oases included in the Khotan district until the constru(
tion in recent times of Keriya as a separate administrative unit.
A PROMISING
FIND.
323
The pilgrim's description shows that the desert pressed then, as
now, close round the small oasis. A huge jar of ancient pottery,
nearly
3
feet in diameter, which had been found years ago at
the old site to be visited, was at first the only antiquity that
Niya could show me. But in the afternoon I
received un-
expected proof of the great age of the ruined site I was to visit.
Hassan Akhun, my inquisitive young camel-man. had acci-
VILLAGE BOYS AT NIYA.
dentally come across a villager possessing two inscribed wooden
\
tablets brought away from that site. When these objects
were produced before me, I discovered to my joyful surprise
that they contained writing in that ancient script of the ex-
treme North-West of India known as Kharoshthi, and of a
type which closely agreed with that prevailing during the
period of Kushana rule in the first centuries of our era.
The man who brought me the tablets had picked them up
21*
324 TO NIYA AND IMAM JAFAR SADIK.
on the road to Imam
Jafar's Mazar. But I soon ascertained
the original finder in the person of Ibrahim, an enterprising
young villager who had dug them out from a
"
house of the
old town
"
in the desert beyond. He had gone there a year
before in search of treasure, but had found only a number of
these, to him, useless tablets. He brought away six, only to
throw some away on the road and to give the rest to his children
to play with. The latter specimens were soon destroyed, and
Ibrahim now greatly regretted their loss, when he saw how
well I rewarded the more sensible man who had picked up
what he had cast aside. I tried to hide my delight as well as I
could, but did not fail to secure Ibrahim as a guide for my
party, and to assure him of a good reward if he could show me
the ruined structure where he made his find. Kharoshthi
writing had before been found in Central Asia only on the
earliest Khotan coins approximately assigned to the first and
second centuries of our era, and in those remarkable fragments
of a birch-bark codex which M. Dutreuil de Rhins, the ill-
starred French traveller, acquired in Khotan in 1892. It was
a happy evening when I examined these most promising finds.
The very cursive form of the writing and the faded appearance
of the ink prevented any attempt at immediate decipherment.
Certain linguistic features seemed to prove that the tablets I
held in my hands contained documents with an early Indian
text, and the writing alone sufficed to assure me of the antiquity
of the ruins that had furnished them. But full of expectation
as I was, I little anticipated at the time what a rich harvest
was awaiting me there.
I A three days' march from Niya brought me to Imam
J
afar
Sadik's shrine, the starting-point of my fresh expedition into
the desert. Scenery as well as the weather helped to make
these days pleasant. I left Niya on a delightfully sunny
morning, and the sky kept clear all the way, but the cold was
still severe, the temperature at night falling to somewhere
about
8
Fahr. below zero, and in daytime never rising much
above
22
Fahr.
The route lay, of course, all along the Niya River, as the
"
old town
"
of my present quest had to be reached from
MARCH ALONG NIYA RIVER. 325
where the river dies away in the sand. Its course proved
almost as winding as that of the Keriya Darya, but its volume
far smaller. Just
like the river of Keriya, the Niya stream
gathers water from springs and marshes a short distance below
the town. This is, of course, the water that has been absorbed
higher up by irrigation channels, and comes again to the surface
lower down. Very soon after losing sight of the cultivated
area we were in a broad belt of jungle land covered with
luxuriant Kumush and forest vegetation. The sands receded
to more than two miles from the left river-bank, and nearly
as much from the right. The route, owing to the large number
of pilgrims who annually frequent Imam Jafar's Mazar in the
autumn, had the well-trodden look of a high-road. On the
first day it touched again and again the bank of the stream^
now a glittering sheet of ice. Its breadth was there usually
30-35
yards, its depth as far as I could ascertain from holes
that had been cut into the thick ice, nowhere more than about
3
feet. As the banks were only about 2-3 feet above the
surface of the ice, it is probable that during the time of the
melting snows a good deal of overflow must occur. This may
account for the luxuriance of the jungle growth that distin-
guishes the riverine belt. The grazing-grounds of the Niya
shepherds begin, therefore, close below the little oasis, and
evidently maintain a considerable number of flocks. They
are said to be divided among ten shepherd stations, and all
belong to
'
Bais
'
of Niya.
The thought that all this fertile stretch of ground might
well be brought under cultivation had occupied me as I rode
along. It was, therefore, a pleasant sight to me when a little
below Nagara-khana, the shepherd's hut where my first night's
camp had been pitched, I came upon the head of a canal begun
only two years previously under the Amban's orders. From
this point, which is about nineteen miles distant from Niya
Bazar, the fertile belt of soil widened considerably, and the
ridges of the desert sand disappeared from view. The river
winds away on the eastern side, while the route led through
the central part of what looked Hke a small tract of jungle.
Close to the route runs the new canal, a modest work so far.
326 TO N]YA AND IMAM JAFAE SADIK.
only
6-8 feet broad, yet likely to bring life and wealth to this
lonely woodland. The soil is a fertile loess, and the level of
the ground so uniform that its irrigation will be easy when the
jungle is once cleared away.
For over eight miles we followed the canal, and I pictured
to my mind the changes it is likely to bring soon to this silent
scene. No doubt in ancient times irrigation was carried all
along the streams which cut into the desert area, and by a
careful storage of their waters probably much ground beyond,
that now seems irretrievably lost to the moving sands, was
secured for cultivation. A strong and capable administration,
whether on European or Eastern lines, might any day take
up again the old struggle with the desert and successfully
push forward the borders of human habitation, just as it has
in the Turkoman steppes and the Doabs of the Punjab,
by nature scarcely less arid. But whence is that impulse to
come ?
Wherever the forest left sufficient open ground I could see
the distant snowy range rising far away to the south of Niya.
The atmosphere kept so clear that even up to Otra Langar,
where a few reed huts form a half-way rest-house for the pil-
grims, our position on the plane-table could always be fixed
by intersections from the prominent points in the great moun-
tain range. Truly a remarkable testimony in favour of the
winter atmosphere of the desert, considering that at Otra
Langar we were close on seventy miles away from the nearest
of hose peaks.
But these distant vistas ceased when the thickets of poplars
and tamarisks were entered a little beyond that station. Here
the woodland seemed to expand considerably over ground
that bears ample evidence of having once been occupied by the
shifting bed of the river. According to our shepherd guides
the width of the jungle tract is here 8-10 miles, and the bear-
ings obtained from certain elevated points on the day's march
seemed to confirm this estimate. Under the trees and in all
depressions of the ground there remained a thin layer of snow,
evidently of the previous week's fall. With the bare trees
and their thick undergrowth it made up a landscape that re-
THROUGH RIVERINE JUNGLE. 327
minded me more of a winter scene in Northern Europe than
anything I had seen for long years.
For over twelve miles we rode through the forest without
getting a glimpse of the river or of a shepherd's hut that might
have indicated its vicinity. At last a little before sunset we
reached the deserted reed-huts of Dobe-Bostan, the second
camping-ground. Here the sand-hills of the desert-edge re-
appeared westwards, while the river came again within reach
about a mile off to the east. The camels with the baggage
did not arrive until
7
p.m. My men kept up splendid bon-
fires with the fallen trees close at hand, and thus I was able to
enjoy the delightfully clear night sky without feeling the cold
too acutely. But when my tent was ready at last the tem-
perature in it was
10
Fahr.
My march on the 26th of January was considerably shorter,
only about thirteen miles, and brought us to the famous Mazar
that was to be our point of departure for the desert. The
river, when we met it again in the morning, appeared as a narrow
band of ice scarcely over 20 feet in width, and yet constantly
sending off little branches. It looked as if the much-reduced
stream were trying where it could bury itself quickest. Yet
just here, so near its end, the fertilising power of its water
visibly reasserted itself. The trees increased in size and the
shrubs in height as we came nearer to the shrine, and the
scenery around the latter looked quite pretty even in the bare-
ness of its winter dress. Scattered at various points amidst
groves of large poplars are huts intended for the shelter of
pilgrims. On the ridge of what looked like a huge sand-dune
to the west there appeared a tangled mass of staffs and flags
of all kinds, marking the resting-place of the mart}T prince
whose memory renders the place sacred. At last we were in
front of the collection of mosques, Madrasahs and houses of
the hereditary attendants which make up the
'
Mazar.' It
looked imposing enough to eyes that had seen besides the
desert only the mud huts of Keriya and Niya for the last two
months. A group of little lakes, formed apparently by the
river between the group of buildings and the hill which the
tomb crowns, greatly added to the picturesqueness of the scene.
328 TO NIYA AND IMAM JAFAR SADIK.
After a rapid inspection of the buildings, among which only
a [large quadrangular Madrasah built by Niaz Hakim Beg
with burnt bricks can claim some merit, I proceeded across
the ice of the northernmost lake to the hill opposite. Its foot
is occupied bygroves of fine old trees, amidst which pious donors
have erected praying platforms and various little Sarais for
pilgrims and the scholars who attend the school of the shrine.
TREES WITH EX-VOTOS, ON PATH TO IMAM JAFAR SADIK's TOMB.
All the trees were decked with little flags, yak tails or simple
rags, the votive offerings of visitors.
The path to the hilltop
ascends through a large number of rough wooden arches, all
bearing the same marks of pilgrims'
devotion. At the first of
these arches there is to be seen the motliest collection imagin-
able of rags. All colours and
materials are represented, from
fine Indian muslin to Birmingham
cotton prints, Chinese silks,
Russian chintzes, and the coarse
'
Kham
'
of the country
SHEINE OF IMAM JAFAR. 329
The custom
which prompts pilgrims to leave behind these
tokens of their devotion prevails equally at Muhammadan and
Hindu pilgrimage places throughout India. I thought at the
time how curious an archaeological find this exhibition of textile
samples would make if it were safely buried beneath the sands
and laid bare again after centuries.
The pilgrims' path to the tomb winds round the hill, and on
its inner side are everywhere little heaps of earth arranged like
graves. They are intended to symbolise the resting-places of
the
'
Shahids,' the faithful Muslims who fell here with Imam
Jafar
Sadik, their holy leader, fighting the infidels of
'
Chin
and Machin,' i.e., Khotan, as related in the Taskirah or legen-
dary of the shrine. More curious to me, however, was the
observation that the hill, which rises about
170
feet above the
lake, does not consist of sand but of stony detritus overlying
reefs of salt. The latter crops out at several points and is of
greyish-white colour. The presence of rock-salt, and of the
gravel which covers it, is remarkable enough in this locality.
Both to the west and east of the riverine belt there is nothing but
sand. If the sanctity of the hill goes back to pre-Muhammadan
times, as is likely enough in view of what has been observed
of other pilgrimage places in this region, this natural peculiarity
would suffice to explain it.
From the hilltop we enjoyed an extensive view over the
desert northward. The forest, which marks the extent of the
river's fertilising influence, seemed to die away some six or
seven miles beyond the Mazar. The stream, before losing
itself in the sand, takes a turn to the north-west, and that
direction, too, my guides indicated for the ancient site.
It took a long time to get my caravan into marching order
on the morning of January 26th. The men went in turns to
pay their respects to the saint's tomb on the hilltop, and later
on the filling and securing of the water-tanks that were to supply
us at
"
the old town
"
caused further delay. No water is to
be got by digging at the site I was about to visit, and accordingly
I knew that we should have to depend for a lengthy stay on the
tanks. Those two, which had come all the way from Calcutta,
being constructed of strong galvanised iron, had already during
330 TO NIYA AND IMAM JAFAE SADIK.
their use at Dandan-Uiliq proved equal to the strain caused by
the freezing of their contents. In hot Calcutta, I confess,
neither the makers (Messrs. Thomson) nor I myself had thought
of nights like the last one, when the minimum thermometer
showed
12
F. below zero, our lowest temperature yet recorded.
The other tanks, notwithstanding the precautions taken, proved
to have sprung leaks owing to expansion, when the ice into
which their contents had turned was again melted. Luckily
the great cold prevailing permitted the transport of additional
ice in improvised sacks and netsa most useful expedient which
materially facilitated the regular supply of the indispensable
minimum of water while my camp, counting from forty to fifty
people, was pitched far out in the desert.
The day was brilliantly clear and the sunshine and the gay
colours of the landscape made the march quite enjoyable
through the forest land where the river finally loses itself in the
sand. About three miles below the Mazar the tiny river-
course spreads out in some shallow marshes and then finally
disappears, at least in winter-time. During summer the flood
water, as shown by the deep-cut
'
Yars
'
or ravines we passed
several miles lower down, is carried for some distance further.
Near the marsh known as Tulkuch-kol lie the huts and the
sheep-pens of NuruUah, the guardian of the flocks which belong
to the shrine. These were said to number over four thousand
sheep
;
and Nurullah, who acted as our guide up to the end
of the grazing-grounds, plainly showed by his get-up and
manners that he was more than a common
'
Koichi,' or
shepherd. He cultivates close to his huts a plot of land which
produces wheat and maize sufficient to maintain his family.
At this little farm the ponies were to remain during our stay
in the desert.
The trees grew so thickly in this amply-watered tract that
the camels had often to halt until a path could be cleared for
them. Everywhere the traces of deer, hares and other game
were to be seen. Gradually the jungle area became more and more
invaded by drift-sand
;
clumps of trees which had withered and
died showed themselves more frequently
;
and at last, some
eight miles below the Mazar, the forest changed to a wide
START FOR ANCIENT SITE. 331
expanse of low sand-cones thickly overgrown with tamarisks
and a hardy shrub known as Ak-tiken. Groups of dead poplars
and other trees rose between, their large stems now gaunt and
twisted by age, bearing evidence of a time when the river
carried life further into the desert. From a high sand-hill
close to my camp I could see how the scrubby jungle spread
out between the great ridges of sand that mark on the east
and west the commencement of the true desert. The breadth
of this area was here fully four miles, and at various points it
formed bays that indent still further into the true desert. The
old course of the river must have extended towards the north-
west
;
for in that direction the jungle-scrub could be seen for
a considerable distance, spreading over ground nowhere broken
by high ridges of sand.
The surmise I formed, that the ancient Site would be reached
by following these traces of the former river-course, was con-
firmed by the next day's march. This also showed, for the
first time in my experience of the desert, that the distance given
by the local guides was exaggerated. I had been told that the
ruins to be visited would be reached in three marches from
Imam
Jafar's shrine. In reality we reached the southern
edge of the area containing them by a second march of about
fourteen miles on the 27th of January. It lay all along in the
directionmore exactly N.N.W.in which on the previous
evening I had sighted the continuation of the old river-bed.
For the first fiv^e miles or so the patches of dead forest were so
thick that we had often to pick with;care a way for the camels.
Tamarisk brushwood still grew vigorously amidst the dead
trees, chiefly Toghrak. The time when the latter fioiuished
equally cannot have been very remote. For many of the
lifeless trees still retained their branches, unlike the shrivelled
skeletons of trunks seen elsewhere. A dry channel, about
4
feet deep, could be traced for some distance, winding along
the eastern edge of the old jungle. The men promptly called
it the
'
Ustang
'
(canal) of the
"
old towTi." But I was unable
to find any proof of its artificial origin.
Further down we had to pass through a belt of steep, conical
sand-knolls from
15
to
30
feet high, rising close together and all
332 TO NIYA AND IMAM JAFAE SADIK.
covered on their tops with tangled masses of hving and dead
tamarisks. On the northern slopes the snow that had fallen
a week before still lay plentifully to the depth of an inch. In
the midst of this belt, extending for a distance of about three
miles from south to north, I came upon broken pottery remains
and an enclosure made of thickly-packed rushes. Inside it
the men recognised trunks of fruit-trees and planted poplars,
or
'
Terek.' We had evidently passed the site of some ancient
farm. Beyond, the sand-hills were lower, but also bare.
Living tamarisk bushes could be seen only on isolated sand-
cones rising here and there over the low dunes. Pottery frag-
ments strewn over the sand, with bits of slag and similar hard
refuse, assured my guides that we were near the goal.
Soon I sighted the first two
"
houses," standing on what
looked at first sight like small elevated plateaus, but which
subsequent observation proved to be merely portions of the
original loess soil that had escaped the erosion proceeding all
round. The wooden posts of these buildings rose far higher
above the sand than in the case of the dwellings at Dandan-
Uiliq. A rapid inspection sufficed to show that their mode
of construction was materially the same
;
but the dimensions
here were larger and the timber framework was far more
elaborate and solid. That these remains were of far greater
antiquity became evident almost immediately when, in a room
of one of the houses, I came upon some finely carved pieces of
wood lying practically on the surface, which displayed orna-
ments of a type common to early Gandhara sculptures. March-
ing about two miles further north across fairly high dunes, I
arrived at a ruined structure of sun-dried bricks, which Abdullah
had already mentioned to me at Keriya as a
'
Potai.' It
proved, as I had expected, to be the remains of a small Stupa,
buried for the most part by the slope of a high conical sand-
hill, and hence comparatively well preserved.
Close to it I pitched my camp, in a position conveniently
central for the exploration of the scattered ruins. The ground
in the immediate vicinity seemed greatly eroded, and where
not actually covered by dunes, displayed in profusion large
pieces of broken pottery, withered trunks of poplars and garden
ARRIVAL AT RUINS. 333
trees, as well as much decayed remains of ancient timber that
splintered and broke almost as soon as lifted. Even more than
this debris, the fragments of stone that covered the bare loess,
evidently the remains of larger pieces that must have been
brought here from the river-bed near the foot of the mountains
for use in the houses, attested the destructive force of the desert
winds and of the extremes of climate. As I retired to my
first night's rest among these silent witnesses of ancient habita-
tions, I wondered with some apprehension whether Ibrahim's
story would prove true, and how much of the other precious
documents on wood which he declared to have left behind at
the time of his
"
prospecting
"
visit were still waiting to be
recovered by me.
RUINS OF ANCIENT DWELLING-HOUSE (n. III.), WITH GARDEN.
CHAPTER XXIII.
FIRST EXCAVATION OF KHAROSHTHI TABLETS.
My first business on the morning of the 28th of
January was to
despatch Ram Singh with Ibrahim Akhun, our plucky httle
Darogha, on a reconnoitring expedition westwards. They were
to observe how far fuel and ground easy enough for laden camels
could be found in that direction. My object was to make sure
whether, on the conclusion of mywork at this site, an attempt
could safely be made to move my caravan straight across the
desert to the ruins of Aktiken or Karadong, which were to be
visited on the lower Keriya Darya. In the event of no objective
offering for exploration further to the east, this march of some
fifty miles through the desert might save us the great detour via
Niya and Keriya.
As soon as I had seen the little party safely off with camels
to ride and a sufficient store of ice and food, I hastened to set
out to the ruined building, where Ibrahim had a year previously
picked up his ancient tablets inscribed with Kharoshthi cha-
racters. At Niya he had declared that plenty more of them
were left in situ. It had been impossible to hide from him the
SEARCH OF FIRST
FIND-PLACE. 335
value which I attached to these tablets, and as he subsequently
seemed to regret not having himself made a haul of them, I had
him watched en route, and after our arrival, to prevent his escape
or any possible interference with the spot. The mingled feelings
of expectation and distrust with which I now approached it soon
changed to joyful assurance. About a mile to the east of the
camp I sighted the ruin towards which Ibrahim was guiding us,
on what looked Uke a little terrace rising high above the depres-
sions of the ground caused by the erosive action of the wind.
On ascending the slope I picked up at once three inscribed tablets
Ipng amidst the debris of massive timber that marked wholly
eroded parts of the ruined structure
;
and on reaching the top,
I found to my delight many more scattered about within one
of the rooms, still clearly traceable by remains of their walls.
Only a year had passed since Ibrahim had thrown them down
there, and the layer of drift-sand was so thin as scarcely to
protect the topmost ones from the snow that still lay on the
ground. Ibrahim at once showed me the spot where he had
unearthed the rehcs he had treated in such utter ignorance of
their value. It proved to be the south-west comer of a small
room, which was situated between other apartments in the
northern wing of the building, and is seen on the left in the photo-
graph showing this ruin (N. I.) after excavation. There in a
little recess, about four feet wide, between the large brick-built
fireplace, well recognisable above the sand and the west wall of the
room, he had come upon a heap of tablets by scooping out the
sand with his hands. The
"
treasure
"
he looked for was not
there, and the ancient documents which he found, apparently
still lying in horizontal rows with some sort of arrangement,
were thrown away into the ne.xt room. I blessed the good luck
which had brought me to the site so soon after this discovery.
For, fully exposed to wind and sun, these thin wooden boards
could not long have retained their uniting in such wonderful
freshness as they had during their safe interment of many cen-
turies beneath the drift of sand. As it was, the sun of one year
and perhaps the recent snow (patches of it are seen on the ground
in the photograph on the next page) had bleached and partly
effaced the fvdly exposed writing of the topmost tablets.
AN ABUNDANT HAUL. 337
My first task was to put a guard over the place where Ibrahiin
had scattered these precious finds, so as to prevent further
injury or abstraction. Then the men were set to work to clear
the room where he had first come upon them. It was an easy
matter, as the room measured only
14
by 16 feet, and the sand
which covered its floor was not deep. On the north side, near
the eroded slope, it only lay to a depth of about two feet, which
increased to about four feet towards the south wall. While this
clearing proceeded, I had time to examine more carefully the
character of the whole structure. It was essentially one based
on the use of timber, which the forest land along the river and
the plantations of
'
Terek
'
or white poplar subsequently traced
at many points of the ancient site must have supplied in abund-
ance. Massive wooden beams, which surprised my workmen by
their thickness and perfect finish, formed a kind of foundation.
On this were set wooden posts about four inches square, which
supported the roof and at the same time served as a frame for
the walls. These and smaller intermediary posts, fixed at
regular intervals of about a foot, were joined by light cross-
beams, of which some were still found in position. To this
framework was fixed a strong kind of matting of tamarisk
branches woven diagonally, which again was covered on each
side with layers of hard, white plaster of varying thickness.
The walls had completely decayed where not actually covered
by sand, but most of the posts originally holding them, now
bleached and splintered, still rose high above the surface.
As the room was gradually cleared, about two dozen inscribed
tablets were found at various points of the original floor and on
the raised platform that flanked the fireplace on the west.
There was nothing to indicate whether they had been separated
from the main deposit of documents which Ibrahim had lighted
upon in the recess at the south-west corner of the room. The
careful search which I then made myself for the scattered remains
of his find, resulted in the recovery of no less than eighty-five
tablets, and as the clearing of the remaining rooms of the north
wing still further added to their number, I found myself before
the day's work was done in possession of materials far more
abundant than
I could reasonably have hoped for.
22
1
338 EXCAVATION OF KHAROSHTHI TABLETS.
The remarkable state of preservation in which a considerable
portion of the wooden tablets was found made it easy for me,
even during a first cursory examination on the spot, to acquaint
myself with the main features of their use and outward arrange-
ment. With the exception of a few oblong pieces, all tablets
I
found that day were wedge-shaped, from
7
to
15
inches long,
1 and showed evidence of having been originally arranged in pairs.
\ These pairs had been held together by a string which passed round
\
the square ends of the wedges, usually from
2
J to i^ inches
I
broad, and also through a hole drilled into both tablets where
their left end tapered to a point. In the case of many such double
I
tablets, even where the two pieces had got detached, portions
;of the ancient string were still intact, and on some also the clay
sealing which had been inserted over the string in a specially
prepared square socket. But it was only by the subsequent
discovery of practically perfect specimens that I was able to
I
ascertain all details of the ingenious method of fastening adopted
[
for these ancient documents on wood. I therefore leave their
full description, as well as that of other technicahties
connected
with their use, for a later chapter.
The text, invariably written in Kharoshthi
characters, and
running from right to left parallel to the longer side, occupied the
inner surfaces of the tablets, i.e., those originally turned towards
each other when the tablets were still fastened in pairs. On
the outside surface which bore the clay seal, and soon proved to
have served the purposes of a kind of envelope,
there usually
appeared brief entries in the same script, forming
a single line.
Their very position and form at once suggested
that they were
intended either as records of the contents or else to convey
the
names of the sender or addressee. It was naturally
with intense
curiosity that I examined the writing of each tablet as it was
extracted from the sand. Where double
tablets had remained
together and thus protected each other, the black ink of the
Kharoshthi lines written on the inner surface looked
as fresh
as if penned yesterday. On others it was necessary to apply the
brush to clear away an adherent crust of sand, but only on com-
paratively few had the writing
faded so far as to become
illegible.
I
22*
340 EXCAVATION OF KHAROSHTHI TABLETS.
It was easy to recognise that the tablets, though written by
many different hands, showed throughout the characteristic
peculiarities of that type of Kharoshthi writing which in India is
invariably exhibited by the inscriptions of the so-called Kushana
or Indo-Scythian kings. The period during which these kings
ruled over the Punjab and the regions to the west of the Indus
falls within the first three centuries of our era. The earliest
coins of Khotan and the fragmentary birch-bark leaves of the
i
Dutreuil de Rhins Manuscript, which were the only relics of
j
Kharoshthi writing so far known in Central Asia, have with good
jjreason been assigned to the same period. Thus even while
still engaged in gathering the remarkable documents that were
coming to light here in such surprising numbers, and long before
any careful examination became possible, I felt absolutely assured
as to their high antiquity and exceptional value.
And yet during that day's animating labours and as I marched
back to camp in the failing light of the evening, there remained a
thought that prevented my archaeological conscience from
becoming over-triumphant. It was true that the collected text
of the hundred odd tablets, which I was carrying away carefully
packed and labelled as the result of my first day's work, could not
fall much short of, if it did not exceed, the aggregate of all the
materials previously available for the study of Kharoshthi,
whether in or outside India. But was it not possible that these
strange records, with the striking similarity of their outward
form and almost all, as I had noticed, showing when complete
an identical short formula at their commencement, might prove
to be mere replicas of the same text, perhaps a prayer or an
extract from sacred Buddhist writings ? The care taken about
the sealing of most of the tablets seemed indeed to point to
contents of a more practical natureletters, perhaps, and con-
tracts or documents of some official character. I knew well
that the historical and antiquarian interest of the finds, if they
were really to furnish such records, would be increased beyond
all proportion. Yet the supposition seemed almost too good to
be readily indulged in.
Once in the comparative shelter of my tent, as soon as the
detailed account of the first excavations had been written up. 1
FIRST DECIPHERMENT. 341
began with impatience to compare and study the best preserved
of those remarkable tablets. I knew from the experience fur-
nished by the Dutreuil de Rhins fragments and more than one
Kharoshthi stone inscription from the Punjab, how serious a
task the proper decipherment of these documents would neces-
sarily prove even to the epigraphist working in his study. I was
thus prepared for the exceptional difiSculties likely to be presented
by the cursive character of the \\Titing and cdl the uncertainties
as to the language and contents. Yet sitting up that evening
wrapped in my furs until the increasing cold drove me to seek
refuge in bedthe thermometer showed next morning a mini-
mum of
9
F. below zeroI gained assurance on two important
points.
"^
A series of philological observations bearing on the phonetic
value of the characters, single or compound, that could be read
with certainty, and on the recurrence of particular inflectional
endings, etc., convinced me that the language was an early Indian
Prakrit, probably of a type closely akin to the dialect found in
the legends of the oldest Khotan coins and in the Dutreuil de
Rhins fragments. It became equally certain from a cursory
comparison of the tablets that their text varied greatly both in
extent and in matter, notwithstanding the brief initial formula
with which most of them opened. It was only some days later
that I succeeded in definitely deciphering the latter, when its
wording

mahanuava maharaya lihati,

"
His Highness the
Maharaja writes [thus] :
"

plainly estabhshed that these par-

ticular documents conveyed official orders. In the meantime,
however, the previous observations together with others, such
as the occurrence of numerical figures in the body of various
tablets, sufficed to reheve me of the doubt that this unexpected
wealth of epigraphical finds might after all resolve itself into
numerous repetitions of identical religious texts so much in
favour with the pious among all Buddhist communities.
Though I could not feel sure as yet in regard to the real nature
of the contents, there was enough in the day's discoveries to
justify the conclusion that, with the Kharoshthi script trans-
planted from the extreme North-West of India, an early form of
Indian speech had also been brought into popular use within the
342 EXCAVATION OF KHAEOSHTHI TABLETS.
/ territories of ancient Khotan, probably from the same region.
Such a fact could be accounted for only by historical events of
fair-reaching importance, which hitherto seemed wholly lost to
our field of vision. The fascinating prospect of bringing them
again to light made me look out with intense interest for such
additional finds as the site might have in store.
That my hopes in this direction had indeed been well founded
was proved on the following morning, when I began the clearing
of the southern wing of the ruined building. This adjoined at
right angles the eastern end of the row of apartments excavated
on the preceding day, and communicated with them by a door
leading at first into a small room, only lo ft. broad, which might
have served as a kind of ante-chamber. A broad platform,
built of plaster some
3
ft. above the floor, and extending along
most of the length of the room, looked as if intended to accom-
modate attendants, an exactly similar arrangement being ob-
('
served in modern Turkestan houses. A well-preserved oblong
tablet, which was the sole find made here, shows a handle at one
end and thus closely resembles the Indian
'
Takhta ' or wooden
writing board to which I have already referred in connection
with a discovery at Dandan-Uiliq. It also attracted my atten-
tion by exhibiting on both sides narrow vertical columns of
writing which suggested either a metrical text or else lists.
There was little time to bestow on individual finds when the
clearing of the large apartment (N. iv.) immediately adjoining
on the south had begun
;
for from the shallow sand which
covered it inscribed tablets of all forms and sizes soon began
to crop up in unexpected numbers. It was a room 26 ft. square,
with a raised platform of plaster running round three of its sides,
while the remains of eight posts arranged in a square indicated
a central area which probably had a raised roof to admit light
and air, after the fashion still observed in the halls of large Tur-
kestan houses. As the protecting layer of sand was here only
2 ft. deep, httle more was left of the walls than rows of broken
posts. The first inscribed tablets, too, which turned up in the
sand close to the surface, had suffered greatly, their warped and
split wood showing plainly the effect of the terrible summer
heat to which they must have been exposed since the winds
MORE DOCUMENTS DISCOVERED. 343
had carried away most of the sand that originally protected
them.
All the more delighted was I when I found that even the light
remaining layer of sand had sufficed to preserve in a more or
less legible condition the threescore of tablets that were found
covering the platform along the southern side of the room. In
some places, particularly near the centre of the wall, they were
rising in small closely packed heaps above the plaster flooring,
evidently just as left by the last occupants. But a considerable
number of other inscribed tablets showed plainly by their posi-
tion that they had been disturbed at some early period, appa-
rently not long after the building had been deserted. For, in
addition to some twenty tablets which were found scattered
about in the loose sand covering the floor just in front of the
south and east platforms, I unearthed over two dozen more from
the southern part of the central area of the room marked by the
posts already referred to.
As the layer of these tablets was being removed, it was seen
that they had rested on a square piece of strong matting which,
supported by some light rafters also recovered, must have
once formed a roof over the central area. The matting was
found lying about a foot above the floor, thus showing the depth
to which the invading sand had accumulated before the roof
fell. The tablets found above the matting could only have got
there subsequently. They may well have been thrown there
when the abandoned dwelhng was visited by some one anxious
to search its remaining contents after the fashion of the modem
"
treasure-seekers." The ancient records left behind by the
last occupiers as so much
"
waste-paper
"
(to use an anachro-
nism) were not hkely to have been treated by him with more
respect than Ibrahim had shown for the collection of tablets
he so luckily unearthed in another part of the building.
Below the matting I discovered some more tablets which owed
their excellent state of preservation evidently to this safe
covering, and then I came upon a small oval-shaped platform
of plaster which, judging from the raised rim enclosing it, must
have served as an open fireplace. But more interesting and
^
more puzzling too than these structural details were the epi-
344 EXCAVATION OF KHAEOSHTHI TABLETS.
graphical finds. Their variety in respect of shape and size was
truly remarkable. The wedge-shaped tablets familiar from the
first day's work reappeared again
;
but in numbers they were
far surpassed by inscribed boards of wood, to which, notwith-
standing great variations in detail and proportions, the desig-
nation of oblong appears generally applicable. The use of
those which were provided with a handle, usually rounded or
pentagonal, and exhibited Kharoshthi writing on both sides,
could readily be accounted for by their resemblance to the Indian
Takhta (for a specimen, see
p.
339).
yi^
Other tablets attaining considerable dimensions in length,
up to
30
in., but comparatively narrow, curiously reminded me,
by their appearance and the hole regularly found near one end,
of the palm leaves which we know to have been used from the
earliest times for manuscripts in India. Useful as this hole
must have been for handling and storing these tablets of wood,
it was nevertheless evident that it could not have been intended,
as in the case of palm-leaf manuscripts, for a string to unite a
series of them into a sort of
'
Pothi.' For not only were such
tablets of entirely different sizes, but their great majority (close
on thirty pieces) showed plainly by the irregular arrangement
of their writing, in small columns and often running in different
directions and concluding with numerical figures, by the appear-
ance of various handwritings, erasures, bracketings, and similar
indications, that they did not contain texts, or even connected
communications, but in all probability memoranda, tabular
statements, accounts and other casual records.
Two series of oblong tablets largely represented among the
finds of N. iv. (as this particular room was designated by me)
showed far greater regularity and care in writing as well as work-
manship, without being the less puzzling at the time of their dis-
covery. These were tablets of rectangular shape, varying in
length from
4
to 16 inches, which soon attracted my attention
by the raised rim resembling a margin they invariably ex-
hibited at the narrower sides of their single inscribed surface.
The writing extending between these rims in five to thirteen lines
always parallel to the longer side, ordinarily bore at the com-
mencement a Kharoshthi numerical figure preceded by a word
DATED RECORDS.
345
which I read before long as samvatsare, meaning in Sanskrit or
Prakrit
"
in the year." In the text immediately following there
appeared with equal regularity figures preceded by the words mase
and divase,
"
in the . . . month
"
and
"
on the . . . day."
There could be no doubt that I held in my hands dated docu-
ments or records of some kind. Yet there was nothing in this
to enlighten me as to the peculiar form of these tablets or the
manner in which they might have been used.
Busily occupied as I was in directing the excavation and
clearing and numbering each find, I failed to reahse at the time
the close relation that existed between these tablets and another
class of which the same ruined apartment furnished numerous
specimens. They consisted of rectangular pieces of wood not
exceeding eight inches in length and five in width, and often
much smaller, which on their fiat reverse rarely had any writing,
while the obverse in its raised centre invariably showed a square
or oblong socket, manifestly intended for the insertion of a seal,
together with a transversely written line or two of Kharoshthi
characters. It was only later, when the remarkable rubbish-
heap to be described below had yielded up its antiquarian trea-
sures, that an explanation, as definite as it was simple, revealed
itself of these curious seal-bearing tablets, and of the rims ap-
pearing on the wooden documents to which they had once been
fitted as envelopes (for illustrations, see
pp. 346, 374, 375).
The structure which had so richly rewarded my first two days'
labours was too far decayed to permit of any certain conclusion
as to its real character. The records which had escaped destruc-
tion, buried low on its floors, would, no doubt, furnish the clue,
though I foresaw that their complete elucidation might be a
labour of years. Whether, however, the building I had ex-
cavated had once served as the residence of some local ofl&cial or
perhaps as a monastic dweUing, it was clear that the writings
found in it could only be the remains of a collection that had
gradually accumulated and been left behind by chance when
the place was abandoned.
<^
It was hence a fact of archaeological significance that among
all this wealth of written documents not a single scrap of actual
paper was found. Nor could I discover a trace of paper among
Q
^
'I
D
S
o
o
O 1)
u
2;
^
o
^
EROSIVE POWER
OF WIND. 347
the miscellaneous rubbish which was brought to light in other
rooms of the building. It was evident, therefore, that the use
of this writing material, so much more convenient than wood,
however old it may have been in China, had not spread in Eastern
Turkestan at that early date which palaeographic considerations
indicated for my discoveries. Wooden tablets for writing pur-
poses are indeed mentioned in very early Indian te.xts, particu-
larly Buddhist ones
;
and it is easy to realise that their use
recommended itself in a country like Turkestan which produces
neither palm-leaves nor birch-bark, the other ancient writing
materials known in India. None the less was it a deUghtful
surprise when, far away at this sand-buried site on the eastern
limits of ancient Khotan territory*, I found myself in possession
of the first specimens ever discovered of Indian records on wood.
All that day a breeze was blowing from the north-east, light
yet sufficiently piercing in the prevailing cold and strong enough
to drive before it a light spray of sand. As this passed over the
ancient tablets laid out in rows as they were being recovered, it
ever threatened to efface the pencil figures which I wrote with
half- benumbed fingers on the often soft wood of their surface
to mark the succession of the finds. I scarcely needed to be
thus
gently reminded of the erosive power of the desert winds.
With the view of the ruined building before me as reproduced
in the photograph taken from a sand-hill close by
(p. 336),
it
was
impossible to ignore the extent to which this and other
structures of the site have suffered by erosion. The small
plateau
which the ruin is seen now to occupy, raised some twelve
to fifteen feet above the immediately surrounding ground, is
unmistakably
due to the erosion which proceeded around the
building.
While the strip of ground actually covered by the
debris of the structure retains the original level, the open sur-
face near by, consisting of mere loess, has been lowered more
and more by the action of the wind. The drift-sand carried
along over this portion of the area, which was once watered
from the Niya River, is not sufficient at present to fill the
depressions thus created or to protect the ruins. Broad ravines,
from
fifteen to thirty feet deep, were to be seen in many places
where the excavating force of the winds could freely assert itself
348 EXCAVATION OF KHAROSHTHI TABLETS.
in the bare loess soil. That part of the ground, too, still occupied
by ancient buildings, is being slowly cut into and undermined,
just as if it were exposed to the erosive action of running water.
The result finally produced by this slow process of destruction
"is aptly illustrated by the photograph just referred to
;
for the
heavy timber debris there seen on the slope of the foreground
marks a part of the original building which has completely fallen,
owing to the soil beneath having been eroded. Thus at more
than one spot near my camp I found scattered fragments of
beams and posts as the sole remains of ancient structures.
Ultimately the wood, rendered brittle by long exposure, breaks
up into splinters which the winds are able to carry away with
ease, and only potsherds and small fragments of stone or metal
remain to indicate the place of ancient habitations.
SPECIMENS OF ARCHITECTURAL WOOD CARVING, FROM RUINED DWELLING-
HOUSE (N. VIII.).
CHAPTER XXIV.
EXCAVATION OF ANCIENT RESIDENCES.
The danger to ruins involved in erosion was strikingly demon-
strated by the condition of a large group of ancient dwellings
which I found about half a mile to the north-west of the building
first excavated, and which I next proceeded to examine. Here
an area roughly measuring
500
square feet proved to be closely
occupied by the timber debris of ancient houses. But as the
dunes were only a few feet high, and the ground everywhere
was greatly eroded, very little remained of the walls, and still
less of the contents of the rooms. Even thus careful search
led to interesting finds of inscribed tablets in a detached room
immediately to the south of the best preserved of these houses,
which is seen in the foreground of the photograph reproduced
on
p. 351.
The sand covering the floor of the detached room (N. v.) lay
350 EXCAVATION OF ANCIENT
KESIDENCES.
only half a foot to one foot deep. Being thus poorly protected
against atmospheric influences, the majority of the fifty odd
tablets of wood we here picked up had withered and bleached
until all trace of writing was lost. Others, though much warped
and often perished on the surface, still show Kharoshthi writing.
Oblong tablets of considerable length and irregular shape are
particularly frequent among these finds, and, so far as legible,
usually exhibit matter in narrow, closely-written columns.
The list of names and items of accounts which seem to be
contained in them, again point to records kept in some office.
The extent of the clerical labour once carried on here could
be measured by the size of these tablets, one piece, unfortunately
entirely bleached, attaining the inconvenient dimensions of
7
feet 6 inches in length, with a width of
4^
inches.
The slight depth of the sand covering this area permitted me to
clear in the course of two days a considerable number of small
ruined houses. They served to acquaint me with the typical
arrangement of the rooms, cattle-sheds, etc., composing these
homesteads, though, owing to the far-advanced decay, finds of
interest were scarce. Among these I may mention an ancient
ice-pit in the outhouse of a modest dwelling-place on the western
edge of the area. Here in a small room my labourers came upon
two unhewn trunks of Toghrak, lying close together and parallel.
Abdullah, my guide from Keriya, at once suggested that we had
found a' Muz-khana,' or ice store-room, trunks being used now in
exactly the same way to keep the ice from touching the ground.
The thick bed of ancient poplar leaves, which were soon brought
to light in the space of about two feet between the trunks, proved
the correctness of Abdullah's conjecture, heaps of such leaves
being still the usual covering for the ice which well-to-do villagers
store for use in the summer season.
On the 30th of January
my Darogha returned safely from his
reconnaissance into the desert. For about three marches west-
wards he had found the sands practicable, and the supply of
fuel yielded by occasional patches of tamarisk and Kumush
growth sufficient. This showed that the short route to the
Keriya River could eventually be taken by us, even if the snow
which Ibrahim Akhun reported to be about a span deep there,
352 EXCAVATION OF ANCIENT RESIDENCES.
should have completely melted. There were already signs
that the rigour of the desert winter could not be depended upon
to help us for many weeks longer. The minimum temperature
on the 30th of January was still
-4
Fahr., but at midday the
thermometer rose to 42 Fahr. in the shade. I had now got so
accustomed to the cold as to find
26
Fahr. quite a comfortable
temperature for writing, and while walking that day to some
ruins yet to be explored, I felt it almost too warm. The thought
of what life in the desert would be like after March had once
passed was in itself a sufficient inducement to hurry on work.
The object of my next excavations was the remains of two
large dwelling-houses which I had passed on my first arrival,
about two miles to the south of the Stupa. Their position
on isolated terrace-like banks of loess, due to erosion of the
neighbouring soil, and the rows of fallen poplars of great size
marking ancient gardens and avenues close by, made these
ruins particularly prominent. The one to the east, N. Ill,
(see photograph,
p.
334),
which, judging from the size and
number of the apartments, must have been the residence of a
man of position, was far deeper in the sand than any of the
buildings previously examined, and consequently proved much
better preserved. Its excavation occupied my band of labourers
for fully four days, though their numbers had by this time
been strengthened by every able-bodied man from Imam Jafat
Sadik, a reinforcement which I summoned in haste as soon ag
the first inspection had shown me the great extent and importance
of the site.
The most striking feature of this building was the large
central apartment, which measured
40
by 26 feet, and probably
served as a kind of reception-hall. Four massive beams of
poplar wood, full
40
feet long, once supported the roof. The
corbel, which was fixed beneath the two central beams, ws
also a fine piece of wood, nearly 8 feet long and 10 inches* in|
thickness, showing bold mouldings. The post, which once
carried this corbel, had fallen long ago
;
yet when the excavatior
began the beams were still in place, resting on the deep sandj
that had meanwhile filled the hall. As the work of clearing
proceeded, the stuccoed walls revealed remains of a carefullj
AN OLD RECEPTIOX-HALL.
353
executed decoration in fresco. This proved to consist of
horizontal bands, painted in dark red and black, on a plaster
ground of creamy white. The broad upper band contained
a scroll ornament of large lotus flowers
;
below it extended
a narrow band in black with leaves like those of a fern painted
in white, and from this again descended elegant festoons of
budding lotuses.
To prop up the walls and prevent their falling in during
PORTION OF ANCIENT DWELLING-HOUSE (X. Ill), BEFORE EXCAVATION.
'
xcavation was no easy task. The photograph on
p.
355
shows
part of the north wall with a side door leading from the hall into
a sort of ante-room. On the extreme left of the photograph is
seen the top of the single wooden leaf which once closed another
small door connecting the ante-room with an apartment west-
wards. This leaf was found in perfect preservation and still
on its hinges, leaning against the wall, just as when it had
been last opened.
Another curious memento of Ufe long departed were the
23
354 EXCAVATION OF ANCIENT RESIDENCES.
remains of embers I found on the small raised platform of
plaster occupying the floor in the centre of the hall, and evidently
once serving for the reception of an open fire.
The hall had been completely cleared by the last dwellers or
visitors of any articles it may have contained. But the smaller
rooms adjoining it to the north furnished a very interesting series
of relics, illustrating the manufactures and arts of the period.
'
The specimens of textile industry were particularly numerous.
Besides pieces of felt and coloured cotton cloth, not unlike the
modern
'
Kham ' of the country, there turned up portions of a
delicately worked rug, resembling in make an Indian
'
Durrie,'
and showing elaborate geometrical patterns and harmoniously
blended colours which only wanted a little brushing to reappear
in their original brilliancy. The coloured reproduction of part
of this rug, given in a plate of my
"
Preliminary Report," will
show the interest which these finds possess for the history
of Central-Asian handicrafts. Small pieces of carved ivory,
including an ivory-tipped baton, attested the skill of the ancient
turner.
A number of small wooden tablets with Kharoshthi writing,
wedge-shaped, or else having the appearance of labels, turned
up in the refuse that covered the floor of an outer apartment,
and what seems to have been once the kitchen. In the kitchen
there was, just as in modern Turkestan houses, a mud platform
to hold the water-jars and other stores, also a large wooden
trough. More curious were the finds in a small closet-like
apartment, which evidently served as a storage room. There
I found a bow of tamarisk wood, still crisp and capable of use
;
carefully turned shafts of light poplar wood, broken, yet still
over six feet long, which must have once served for spears
;
a section of a shield of wood (willow) about
3
feet 6 inches high
;
spindles and other small household implements, all of wood,
including a stout walking-stick of apple-wood that I found
to come in very handy for use.
Among the articles of ornamental wood-carving found in the
building none can compare in point of artistic interest with the
ancient chair reproduced on
p. 356.
Its pieces, though disjointed,
lay close together on the floor of one of the outer rooms and
23*
356 EXCAVATION OF ANCIENT EESIDENCES.
could easily be fitted together. The excellent preservation of
Vhe elaborate carvings was all the more surprising as progressive
erosions had left scarcely any of the sand which no doubt
previously covered and protected them. The decorative motives
shown in these carvings are familiar to us from relievo sculpture
ANCIENT WOODEN CHAIR FROM RUINED DWELLING-HOUSE (N. III.).
(Scale one-eighth.)
that once adorned the Buddhist monasteries of Yusufzai and
Swat, the ancient Gandhara. I was glad to note at the time
how closely the date thus indicated agreed with the chronological
evidence deducible from the Kharoshthi writing of the tablets.
My men were duly impressed by the fine appearance of this
ancient piece of art furniture when it was set up before their
wondering eyes.
ANTIQUE FURNITURE. 357
The large size of the house, which covered an area nearly
100 feet long by 80 in width, also helped to suggest that this
was the residence of a man of means, and possibly in authority.
So my labourers promptly christened the place as the
'
Yamen.'
It is true they did not find in it the hoped-for horseshoes of
silver, but several Chinese copper coins were picked up from
the sand in its immediate vicinity. The fact that these coins,^
as well as all subsequent numismatic finds on the site, belonged
to issues of the second Han dynasty, greatly strengthened 1
my reliance on the palaeographic evidence of the tablets.
^
Equally curious results rewarded the clearing of another large
dwelling-house, N. IV., some three hundred yards to the south-
west, which occupied us during the 4th and 5th of February.
The plan of this house and of the garden adjoining it, repro-
duced on
p. 360,
will help to illustrate the internal arrangement
of these ancient residences. It strangely recalled the disposition
of rooms, etc., I had observed in modern Khotan dwelling
places of some pretensions. In a room, which seems to have
served as an office, there were found, besides a number of in-
scribed tablets of varying shape, apparently orders and
memoranda, several tablets that had never been used, blank
stationery left behind by the last occupants ; also writing
pens of tamarisk wood
;
eating-sticks of wood like those still
used by the Chinese ; and a large sleeping-mat or hammock
made of tamarisk rushes. In the long, narrow passage that
traverses this house I came upon the well-preserved upper
part of a guitar, resembling the
'
Rabab
'
still in popular use
throughout Turkestan, and retaining bits of the ancient string,
as well cLS upon more samples of carpet materials.
-^
But even more interesting were the remains of an elaborately
decorated wooden arm-chair which were also successively
recovered there. The legs represent standing lions, in evident
imitation of the Indian
'
Simhasana
'
("
lion-seat "), and retain
in part their original colouring in red and black. The arm-
rests are formed by a pair of well-carved monsters, male and
female. The heads and busts are shown as human, the parts
from the waist downwards apparently birdlike, while the legs
are those of a horse with strongly-marked hoofs. The terra-
AN ANCIEXT GARDEX. 359
cotta ground colour is well preserved, and over it appear traces
of black and dark blue paint marking the plumage and hoofs.
Have we to recognise in these strange creatures a reminiscence
of the Kimnaras of Indian mythology, or of a still more distant
hybrid, the Centaurs of Greece ?
A very curious feature of this ruin was the clearness vAth which
the arrangement of the adjoining garden could be traced. The
trunks of the poplars, which still rise 8 to lo feet from the
original surface, and are thus clearly \dsible above the drift-
sand, are grouped in the same little squares and enclosing
rectangular avenues which can be seen in every well-kept
'
Bostan ' from Kashgar to Keriya. The trees were planted at
regular distances, which are marked on ray plan. The hedgerow
or rush fence enclosing the garden was also mostly intact,
though covered in parts by the sand. It was with a strange
feeling, obliterating almost the sense of time, that I walked
between two parallel fences of rushes that still form a little
country lane just as over i,6oo years ago. The wind had
swept the ground between them clear at various places, thus
displaying the pottery fragments, bits of charcoal, and decayed
foliage on which the last inhabitants must have trodden.
Searching at the bottom of the fences my antique walking-
stick disclosed the rustling dead leaves of poplar and fruit
trees, perhaps the same that still raised their shrivelled trunks
in scattered groups near by. Among these withered relics
of ancient orchards, which I observed here as at some other
points of the site, my diggers had no difficulty in distinguishing
various fruit-trees, such as the peach, plum, apricot, mulberry,
the wood of which they knew from their own homes.
There might have been during those days a temptation for
me to forget altogether my li\-ing surroundings through the
antiquarian fascination of the dead past. It was, perhaps,
in order to proxade against any such eventuality that my faith-
ful myrmidons from Kashgar took the opportunity to attract
my
attention to their several human faiUngs and consequent
mutual
bickerings.
The incidents which arose thereby have
their
hmnorous
interest when looked back upon ; but at the
time, I must own, I should have gladly done \\-ithout them.
TROUBLESOME FOLLOW^ERS.
361
It was with Niaz Akhun, my Chinese interpreter, that these
troubles began. Knowing how Httle disposed he was to make
himself useful during work in the desert, or to bear patiently
with its inherent hardships, I had left him behind in charge
of the ponies which were to await our return at Nunillah's,
the shepherd's, huts beyond Imam
J
afar Sadik. There were
no Chinese there to gamble with, and, in fact, scarcely any
people to fight or to bully. So I thought my troublesome
follower fairly out of the reach of mischief. Reports, however,
which began to arrive with the convoys bringing ice, and with
the men ordered up from the shrine as reinforcements, soon
convinced me that I had underrated Niaz Akhun's truculent
propensities.
That he was supplementing the ample rations left for him by
requisitions of whatever the resources of the Sheikhs of the
shrine could supply in the way of eatables would scarcely have
brought matters to a crisis. But his amorous demands on the
attentions of the women-folk living at the Mazar grew in excess
of what even the hospitality and easy morals of these parts
would tolerate. The result was the despatch to me in the
desert of a pitiful petition praying for the holy settlement
to be speedily relieved of the infliction represented by my semi-
Chinese attendant. There was only too good reason to believe
in the substantial truth of the complaints, and as the only
safe course was to keep Niaz Akhun under my own eye, I sent
peremptory orders for him to give over charge of the animals
to another of my men, and to join camp at once. It was a
truly comic sight when the sinner arrived, weary with the
two days' tramp through the sand to which he was little ac-
customed, yet in his genuine dejection acting with consummate
skill the part of injured innocence. According to his own story,
he was the victim of a wicked conspiracy between some of his
enemies among my own men and the shameless mendicants at
the shrine, etc. In order to make his appeal for justice still
more impressive he had donned over his comfortable coat white
rags to indicate mourning. He pretended to have just received
news of the death of his mother, far away at Aksu, but he
failed to explain how the sad intelligence could have arrived
362 EXCAVATION
OF ANCIENT KESIDENCES.
1
SO opportunely. At first he offered to commit suicide in order
to clear his reputation, and to rid me of further trouble on
his account. Gradually, however, he settled down to a more
resigned behaviour, and I began to hope that he would submit
with good grace to the privations of desert life, the worst
punishment I could inflict, when on the next day the storm
broke.
It pleased Niaz Akhun to attribute his ignominious recall from
the fleshpots of the Mazar to the enmity of Hassan Akhun, one
of my young camel-men. So he promptly denounced the latter
to me as having secreted an antique gold ring which he had
picked up near one of the ruined houses, against my standing
order that all chance finds were to be reported, thus giving
me an opportunity to acquire them for an adequate com-
pensation. When Hassan Akhun arrived with the next ice-
transport, he readily delivered the ring, which proved to be of
brass, and took the reward I offered for it. But the incident
stirred up still more the spite felt by my Muhammadan followers
against Niaz Akhun, so that on returning to camp in the evening
he was attacked in open fight by the pugnacious little camel-
man. The encounter might have been as amusing to watch
as that other heroic fight described by Horace in his journey
to Brundisium, could I only have trusted the temper and balance
of mind of my Chinese interpreter. Half-maddened by his
disgrace and the taunts of the other men, he seemed quite
prepared to give a tragic turn to the affair by the use of his
knife, when I managed to separate the combatants not without
vigorous application of the antique walking-stick I happened
to have in hand. Just
as I had succeeded in this, with the
help of Ram Singh, Sadak Akhun, the cook, came rushing up
in wild excitement, brandishing the sword which he prided
himself upon carrying about as an old
'
Dakchi ' of the British
Agency at Kashgar. I thought at the time that he might
have run
"
amuck
"
;
for Sadak Akhun's conduct had grown
very queer for weeks past under the baneful influence of the
'
Charas
'
drug to which he was addicted, and of which he took
increasing quantities as the hardships of our winter campaign
continued. Luckily his brain had not given way, as his frantic
AFFRAY IN DESERT CAMP. 363
behaviour might have made one believe. So while xaolently
protesting his wish to avenge his misery as well as Islam
(!)
on
the renegade Niaz, he at last allowed himself to be disarmed
and led away. The interval during which my attention had
been diverted by Sadak Akhun's appearance on the scene,
was promptly utilised by Niaz for a dramatic attempt at suicide.
With astonishing rapidity he had loosened his waistband, and
drawing it tightly round and round his neck, was doing his best
to strangle himself in a fit of frenzied energy. His face was
getting to bear a strange look by the time we succeeded in
loosening the convulsive grip of his hands, and the utter
exhaustion which then overcame him con\'inced me that it
was not a mere exhibition of mimic power to which he had
treated us.
It was evident that, to prevent fresh affrays with possibly
more serious results, Niaz had to be kept away from the rest
of my Muhammadans. So I was glad that my two Hindu
followers, Ram Singh and Jasvant Singh, were ready to share
their camp-fire with him and in a general way to keep an eye
on him. Hassan Akhun, for his unwarranted attack, was
sentenced to a number of stripes, which Ibrahim, my excellent
Darogha from Keriya, administered next morning with an
arm practised in such functions and which had a very salutary
effect upon the young offender. There remained only Sadak
Akhun to worry me by mad fits of despondency and sneaking
attempts at insubordination. To his repeated demands for
discharge I could not possibly consent ; for like the rest of my
men he had been expressly engaged for the whole of the journey,
and I was not prepared, if I could help it, to go without European
food which he alone in camp knew how to cook. To the rest of
my men he would wildly talk of running away at night ; but the
fear of losing his way and still more the knowledge that I could
easily get him stopped at Keriya or Khotan and detained in the
mban's Yamen until my return, sufficed to prevent the execution
'I such foolish plans. AH the same I thought I had reason to
compliment myself on successful management when I brought
Sadak Akhun in the end safely back to Kashgar. Restored to
the
pleasures of Bazar life, ^vith substantial accmnulations of
364 EXCAVATION OF ANCIENT RESIDENCES.
pay to spend upon his elegant person, he was then wont to
attribute the troubles he had caused, not to his own innate
fickleness and the effects of the
'
Charas ' habit, but to the
'
Jinns
'
or evil spirits of the sand-buried sites who had gained
possession of his mind during the long nights in the desert. May
he keep clear of them thereafter !
\
CLAY IMPRESSIONS OF CLA^-i^Ai- ai^Ai.:,. r i^oM KHAROSHTHI TABLETS.
CHAPTER XXV.
DISCOVERIES IN AN ANTIQUE RUBBISH HEAP.
The excavations previously described plainly showed me that
the ancient houses of this site had been cleared by their last in-
habitants, or soon after their departure, of everything possessing,
intrinsic value. It was evident that I must base my hopes for
further archaeological finds mainly on any rubbish remains which
might have been left behind. These hopes received gratifying
confirmation in the course of a reconnaissance to ruins reported
north of my camp. I sighted on that occasion over half-a-dozen
groups of old structures scattered over an area of about three
and a half miles from south to north and more than two miles
across. In one ruin, greatly decayed and in no way attracting
special attention, I had come upon a number of bleached tablets
lying exposed, and a little digging had within half-an-hour
brought to light nearly thirty inscribed pieces. Among them
there were two novelties which, though small in size, could not
fail to arouse my utmost interest. One was a fragment of a
narrow piece of wood showing Chinese characters
;
the other a
small strip of well-prepared leather, also fragmentary, with a
line of Kharoshthi characters recording a date. These finds
clearly betokened a rich deposit, and as the ruin in question lay
366 AN ANTIQUE RUBBISH HEAP.
nearly two miles to the north, I decided to move my camp
there.
While my men were busy on the morning of February 6th
effecting the shift, I found time to make a close examination
of the little Stupa below which my camp had stood. At first
sight it appeared that the small hemispherical dome about seven
feet high rose on a single base
13
feet 6 inches square and 6 feet
RUINED STUPA, AT ANCIENT SITE BEYOND IMAM JAFAR SADIK.
6
inches high. But observations as to the original ground level
made me suspect that the structure now visible above the sand
on the southern slope of a large tamarisk-covered sand-cone did
not represent the whole of the Stupa. And in fact, on clearing
the sand heaped up on the east face by the prevailing winds, I
came upon a second base below. It was also square, projecting
on each side
3
feet beyond the upper base and 6 feet high. The
whole of the Stupa thus rose to a height of about 20 feet. I
ANCIENT "WASTE-PAPER" DEPOSITS. 367
was much interested to note how closely the arrangement and
proportions of the dome and its bases agreed with that observed
in the ruined Stupas of Khanui, Moji, and Pialma. The size
of the bricks too (circ. 22 by
17
by
4
inches) proved nearly the
same as in those structures. Nor was the shaft wanting in the
centre of the Stupa dome. It was only one foot square and had
been laid open from the west by a cutting into the brickwork.
That treasure-seekers had been at work was shown also by two
large holes excavated in the upper base. WTiat relic deposit
there once was in this modest
"
memorial tower," as Hiuen-
Tsiang would have called it, must have been abstracted long ago.
Promising as the finds were which my pre\'ious
"
prospecting
"
had yielded, I little anticipated how extraordinarily rich a mine
of ancient records I had struck in the ruin I proceeded to ex-
cavate. On the surface there was nothing to suggest the wealth
of relics contained within the half-broken walls of the room,
23
by 18 feet large, which once formed the western end of a
modest dwelling-place. But when systematic excavation, beguiT
at the north-western corner of the room, revealed layer upon
layer of wooden tablets mixed up with refuse of all sorts, the
truth soon dawned upon me. I had struck an ancient rubbish
heap formed by the accumulations of many years, and containing
also what, with an anachronism, we may fitly call the
"
waste-
paper
"
deposits of that early time.
>^
It was not sand from which I extracted tablet after tablet, but
a consolidated mass of refuse lying fully four feet above the
original floor, as seen in the photograph reproduced
p. 369.
All the documents on wood, of which I recovered in the end more
than two hundred, were found scattered among layers of broken
pottery, straw, rags of felt and various woven fabrics, pieces of
leather, and other rubbish. It was evident that the consistency
which these varied remains had acquired in the course of cen-
turies had more than anything else helped to protect them
against the erosive action of the winds, from which the other
parts of this ruin had suffered considerably. Thus it is mainly
to the unsavoury associations of the dustbin that we must ascribe
the remarkable state of preservation shown by the great mass of
these precious records.
368 AN ANTIQUE RUBBISH HEAP.
I had ample occasion to console myself with this thought
while engaged in the laborious task of clearing this room (N. xv.).
As soon as I had realised the peculiar character of its deposits
it became a matter of importance to keep accurate record of the
relative position in which each object turned up. This would
thereafter help to ascertain the chronological order, and possibly
the internal connection of the various documents. Accordingly,
RUINED DWELLING-PLACE, CONTAINING ANCIENT RUBBISH HEAP (N. XV.),
SEEN FROM SOUTH-EAST.
every inscribed piece had to be carefully tabulated before it was
removed and cleaned, no easy task with fingers half benumbed
by cold and in the dust which a fresh north-east breeze raised
from the dug-up refuse heap. For three long working days I
had to inhale its odours, still pungent after so many centuries,
and to swallow in liberal doses antique microbes luckily now dead.
But in the full enjoyment of my great antiquarian haul I did
not think of these little discomforts.
The diversity in form and material of the documents which
24
{
370 AN ANTIQUE EUBBISH HEAP.
came to light from amidst all this ancient litter and dirt, was not
less remarkable than their good preservation. The first few
hours' work was rewarded by the discovery of complete
Kharoshthi documents on leather. The oblong sheets of care-
fully prepared smooth sheepskin, of which altogether two dozen
came to light here, showed different sizes, up to fifteen inches in
;^^^
b-JSj^
ANCIENT KHAROSHTHI DOCUMENT
ON LEATHER (N. XV. 3IO).
(Scale one-half.)
A
Unfolded.
B In origitialfolded
stale.
length. They were invariably found folded up into neat little
rolls, but could be opened out in most cases without serious
difficulty. The Kharoshthi text, which covers the inner surface,
is usually written in a neat clerical hand, and its black ink has
retained a remarkable freshness. At the head of each document
I was now able to read with certainty the introductory formula
previously seen less clearly on many of the wedge-shaped
\^
tablets : Mahanuava maharaya lihati,
"
His Highness the
DOCUMENTS ON LEATHER. 371
Maharaja orders in writing," There could be no further doubt
that these were official documents. Most of them, I could see,
were dated, but only by month and day, while the single lines
on the otherwise blank reverses were manifestly addresses. In
them I thought I could recognise two personal names or titles
appearing again and again. But who were the recipients of
these and so many other documents, the administrative officers
or simply the clerks of this ancient settlement ?
Quite apart from their contents, these documents have a
special interest as the first specimens as yet discovered of leather
used for writing purposes among a population of Indian lan-
guage and culture. Whatever the religious objections may
have been, it is evident that in practice they had no more weight
with the pious Buddhists of this region than with the orthodox
Brahmans of Kashmir, who for centuries back have used
leather bindings for their cherished Sanskrit codices. The finish
given to the leather of those ancient documents indicates ex-
tensive practice in the preparation of the material. Small
pieces of blank leather of this kind, unmistakably shreds left
after the cutting of full-sized sheets and subsequently swept out
of the office room, turned up plentifully among the rubbish.
The discovery of an ancient pen made of tamarisk wood (see
p.
346),
in the same refuse heap, helps us still better to realise the
conditions of clerical work in that period. The bone knob of
the pen had probably served as a burnisher.
But interesting as these details were, they could not compare
in importance with the information yielded by the far more
numerous kinds of Kharoshthi tablets. Many of those unearthed
ifrom N. XV. were in excellent preservation and retained
intact
Ithe original clay seals and strings with which they were fastened.
iThere could be no doubt as to wood having been the general
iting material, and it was hence particularly fortunate that
[I was thus enabled definitely to ascertain all technicahties
con-
[nected with its use. The wedge-shaped tablets, which seem to
[have been in favour for short communications, invariably
con-
sisted of pairs of pieces fitted exactly to match each other in
size, as seen in the specimen reproduced on
p. 372.
One end of the
[double tablet thus formed was cut off square
;
the other runs
out
24*
372 AN ANTIQUE RUBBISH HEAP.
into a point, near which a string-hole is drilled through both
pieces. The text occupies the smooth obverse of the under-
tablet and is protected by the upper or covering tablet, which
res'ts on it and serves as a kind of envelope. If the length of
the communication required it, the writing was continued on
KHAROSHTHI DOCUMENT ON DOUBLE WOODEN TABLET (N. XV. 1
37).
,
(Scale one-third.)
A Obverse
of
coveritii; tablet with seal. B Keversc
of
ttndcr-tablct. C Obverse oj
tinder-tablet.
the reverse of the covering tablet. The wood of the latter shows
greater thickness towards the square end, and in this raised
portion of the outside surface a square socket was neatly cut,
intended for the reception of a clay seal. A string of hemp was
passed in a cleverly devised fashion through the string-hole and
then drawn tightly over both tablets near the square or right
FASTENING OF ANCIENT ENVELOPES. 373
end. Grooves communicating with the seal socket held the string
in regular cross-folds. The socket was then filled with clay,
covering these folds of the string. When once the seal of the
sender had been impressed into the clay, it became impossible to
separate the pair of tablets without either breaking the seal im-
pression or cutting the string.
The ingenious arrangement here briefly described, which the
accompanying diagrams of Mr. F. H. Andrews' drawing will help
Diagram
of
wedge-shaped double tablet, showing oiverse
of
covering tablet, with
string-hole (A) and stringfastetud in seal socket (B).
Diagram
of
wedge-shaped double tablet, showing reverse
of
under-tablet, with
string-hole (A) andfolds
of
string held by grooves (B).
to illustrate, rendered the communication written on the inner
sides of the' two'tablets absolutely safe against unauthorised in-
spection. If the recipient desired to preserve the sealing and
also retain a convenient fastening for the two tablets after having
acquainted himself with the contentsan obvious advantage
when such letters had to be kept for recordhe had only to
cut the string near the string-hole. The under-tablet could
then easily be sUd out from the folds of string running beneath
the seal, and after being read be passed back again into its
original position, just as we can do this now, after so many cen-
374 AN ANTIQUE RUBBISH HEAP.
turies, in the case of the double tablet reproduced on
p.
372.
Usually to the right of the seal on the obverse of the covering
tablet there appears the name of the addressee, while entries in
a different hand, traceable in some instances on the reverse of
the under-tablet, may possibly represent
"
docket
"
notes by the
official who received the communication.
Scarcely less ingenious is the method of fastening which the
Detached seal-socket
(N.xv.
133.)
(Scale one-half.)
Covermg tablet
of
oblong Kha-
roshilii document {J\l. xv.
167),
showing double seal.
(Scale one-half.)
Covering tablet
of
KharoshtX
document {N. xv. 33c
showing seal with Eros.
(Scale one-half.)
finds of N. xv. prove to have been used for the oblong tablets.
From a number of double tablets of that shape which I recovered
here practically intact, it became clear that the under-tablet
was in this case provided with a raised rim on either of the shorter
sides. Between these rims fitted exactly a covering tablet,
the obverse of which, in its raised centre, had a square or oblong
socket for the reception of a clay seal. Here, too, a string passed
transversely over both tablets and secured below the seal,
effectually prevented any unauthorised opening and reading of
MiiiiMifa
KHAROSHTHI DOCUMENT ON DOUBLE OBLONG TABLET (N. XV. l66).
(Scale two-thirds.)
A
Double tablet unopened.
B Reverse
of
cornering tablet. C Obverse oj under-tahlet.
376 AN ANTIQUE EUBBISH HEAP.
the document written on the inner sides of the two tablets. The
accompanying plate
(p.
375)
shows such a double tablet which
was found with the string broken but otherwise intact, both
before and after its opening,
"
Envelopes
"
of oblong double
tablets, the corresponding under-tablets of which have not been
found or identified, are seen on
p.
374.
There also a wooden
seal-socket found detached, with its clay seal-impression still
intact, has been reproduced.
I cannot attempt, from want of space and of adequate illustra-
tions, even briefly to indicate here all the curious discoveries
made in connection with this ancient stationery in wood. But
some notice must be accorded to that remarkable series of clay
SEAL-IMPRESSIONS IN CLAY FROM
seal-impressions which were found still intact on a
number of
tablets and some of which can be presented here from well-
executed drawings of my friend Mr. F. H. Andrews (see above
and
p. 365).
From an historical point of view they claim ex-
ceptional interest, for they have furnished convincing evidence
of the way in which the influence of classical Western art asserted
itself even in distant Khotan. It was a delightful surprise when,
on cleaning the first intact seal impression that turned, up I
recognised in it the figure of Pallas Athene, with aegis and
thunder-bolt, treated in an archaic fashion. This particular
seal (seen, though not so distinctly as elsewhere, on the covering
tablet reproduced on
p. 365)
was found thereafter to recur fre-
quently, and probably belonged to an official who was directly
connected with the administration of the ancient settlement.
CLASSICAL SEAL-IMPRESSIONS. 377
Another and larger seal (seen in the reproduction of the covering
tablet C,
p.
374,
also in the headpiece of this chapter) shows a
well-modelled naked figure of pure classical outline, perhaps a
seated Eros. Another Athene, a standing Eros, and probably
Heracles, are also to be found among the Greek deities repre-
sented. On other seals, again, there appear portrait heads of
men and women showing classical modelling, though barbarian
features, etc.
Just
as in the case of the engraved stones of similar make
found in the debris layers of Yotkan, it is impossible to make
sure which of these seals were actually engraved in Khotan
territory and which were imported from the West or other parts
of Asia reached by classical art. But though we have yet to
learn the exact functions or place of residence of those who once
used the seals, there can be no reasonable doubt that the docu-
ments bearing their impressions originated in the vicinity of the
ancient site or at least \^'ithin the borders of the Khotan kingdom.
As the date, too, of the documents can, as we shall see, be fixed
with fair accuracy, these seal-impressions are to us far more
valuable than if chance had preserved the original seals. The
vista thus opened out to us is one of far-reaching historical
interest. We already knew that classical art had established
itself in Bactria and on the north-west frontier of India. But
there was little to prepare us for such tangible proofs of the fact
that it had penetrated so much further to the east, half-way
between Western Europe and Peking. As if to symboUse this
strange mixture of influences from the Far West and the Far
East, the covering tablet reproduced in Fig. A,
p.
374,
presents
to us a seal with Chinese lapidary characters in juxtaposition
with one showing a portrait head unmistakably cut after Western
models.
REMAINS OF ANCIENT TREES NEAR SAND-BURIED DWELLING-PLACE (N. VIII.).
CHAPTER XXVI.
DECIPHERMENT OF ANCIENT DOCUMENTS ON WOOD AND
LEATHER.
The historical importance attaching to the records themselves
was ever vividly before my mind, and in each fresh Kharoshthi
document which this precious refuse-heap continued to yield up
I had reason to welcome additional help towards the study
of both script and contents. But from the first I recognised
that the decipherment of this wealth of materials would require
much time and patient labour. The many exacting tasks
which have claimed my attention since my return have left
me no leisure to supplement the cursory examination I was able
to make at the time of discovery and in the rare moments of
rest on my subsequent travels. Fortunately, however, my
friend Mr. E.
J.
Rapson, of the British Museum, to whose care
I had for a time to entrust the whole of my antiquarian col-
lections, readily agreed to charge himself with the study of all
CONTENTS OF KHAROSHTHI EFXORDS. 379
epigraphical finds in Kharoshthi. The painstaking researches
concerning them which he has carried on for the past year
with scholarly zeal and acumen are not likely to be concluded
for a long time. But they have advanced sufficiently to make
it possible for me, with his permission, to indicate in broad
outlines certain main results and some of the more curious
details.
It is a source of gratification to me that the conclusions I first
arrived at regarding the language and general character of these
documents have been fully confirmed by Mr. Rapson's labours^
From his exact analysis of a considerable number of Kharoshthi
documents on wood and leather, it can now be asserted with
certainty that the language is an early Indian Prakrit, possessing
a large admixture of Sanskrit terms. The latter are particularly
prevalent in the introductory and other formal parts of the
letters and reports, that is, exactly where the epistolary custom
of modern Indian vernaculars has large recourse to phrases of
the classical language. As regards the great mass of the docu-
ments there can be no doubt that they contain, as surmised
by me from the first, official correspondence of various kinds.
Reports and orders to local officials on matters of administration
and police, complaints, summonses, safe-conducts, and similar
communications seem to constitute the bulk of the documents.
Others, again, may prove to be records of payments or requisi-
tions, agreements, bonds, and the like. Accounts, lists of
labourers, etc., are probably the contents of the mass of miscel-
laneous
"
papers
"
wTitten on single tablets of irregular shapes
and usually in columns ending with numerical signs.
Exceptionally great as are the difficulties with which the work'
of detailed decipherment has to contend, on account of the
very cursive character of the Kharoshthi script and the puzzling
phonetic peculiarities of the Prakrit dialect employed, we
already obtain many interesting glimpses from the passages
which can be definitely interpreted. The titles given to the
ruler in whose name orders are issued, and with reference to
whose reign the more elaborate documents are dated
('
Maharaja,'
'
devaputra,' i.e.,
"
son of the gods," etc.),
are
purely Indian. They agree strikingly with the official
380 DECIPHERMENT OF DOCUMENTS.
nomenclature observed under the Kushana or Indo-Scythian
princes who ruled the extreme North-West of India and Afghan-
istan during the first centuries of our era. The majority of
the persons to or by whom documents are despatched bear
purely Indian names, among them appearing a Kushana-sena,
as if to emphasise some connection with Indo-Scythian dominion
far away to the South-West.
In strange contrast to the names, some of the titles borne by
these officials are distinctly non-Indian (Chodbo, Shodhoga,
Kala, etc.). But we meet also with official designations familiar
from ancient Indian usage
{'
rajadvara-purasthita,'
"
president
of the royal court of justice ;
"
'
dibira,'
"
clerk," etc.).
Letter-carriers,
'
lekhaharaka,' are frequently referred to by
their Sanskrit designation. The often recurring introductory
formulas, with their stereotyped greetings, honorific addresse
('
priyadarshana
;
'
'
devamanushya
-
sampujita,'
'
priyadeva
manushya,'
"
dear to gods and men," etc.), and polite enquiries
after the health and spiritual welfare of the addressees, possesi
a distinct flavour of that quaint phraseology to which th
Sanskrit correspondence of my Kashmirian Pandit friends hai
accustomed me. But in other documents we find a style far lesi
/
ornate, in fact quite peremptory, as, e.g., in some office
"
memos
f
ordering the submission of affidavits
('
savatha
')
according to
\ a specified list ; the production of certain witnesses
;
arrests of
^
individuals, etc.
The particular interest attaching to some petty records is
well illustrated by an oblong tablet, dated in the ninth year
of King Jitroghavarshman, which relates a transaction by a
certain Buddhagosha, slave of the Sramana or Buddhist monk
Anandasena, concerning some household goods, pawned perhaps
or taken over on mortgage. The articles are enumerated
in detail and their value indicated in a currency that we may
yet succeed in determining. It is curious to find that this list,
besides sheep, vessels, wool-weaving appliances and some other
implements, enumerates also
'
Namadis.' We may well
recognise here the earliest mention of the felt-rugs or
'
Numdahs
'
so familiar to Anglo-Indian use, which to this day form a
special
product of Khotan home industry and of which large consign-
1
H
X
H
<
2
C
C/3
O
Q ^
382 DECIPHERMENT OF DOCUMENTS.
ments are annually exported to Ladak and Kashmir. In
another document we read that all the
*
Shodhogas ' and
'
Drangadharas,' evidently local officials, of the district are
complaining of the want of water
;
and many of the tablets
seem to have reference to disputes about water used for
irrigation.
The frequent references in the tablets to
'
Khotana ' and its
officials shows us not only how ancient the name of Khotan is
in its present phonetic form, but also that the district containing
this settlement was part of the kingdom of Khotan. It is
of interest that, alternating with that old popular name, we
also find the duplicate form Kustana[ka] known to us from
Hiuen-Tsiang's records. It represents, in all probability, a
learned adaptation of the local name made for the sake of a
Sanskrit etymology, which the pilgrim duly relates to us together^
with its attendant pious legend
('
ku-stana
'
meaning in Sanskrit
"
the breast of the Earth
").
As if to remind us of the positioi
which the ruined settlement must have occupied on the out-|
skirts of the cultivated territory, we meet with frequent mentior
of
"
frontier-watch stations," designated by the Sanskrit terr
'
Dranga,' the true significance of which I first demonstratec
years ago in Kashmir. That the faith of Buddha must have
been widely spread among the people can be proved by a number
of passages. Thus the Buddhas, Arhats and other sacred
categories of the Buddhist Pantheon are distinctly enumerated
in one tablet, while in another the addressee is with polite unction
designated as a
''
Bodhisattva incarnate."
Not the least curious among the facts revealed by the work of
decipherment is the discovery that there existed a recognised
official terminology for the various classes of stationery repre-
sented. With unchanging regularity the wedge-shaped tablets
are designated in their context as
'
kila-mudra,' literally,
"
sealed wedges
"
;
the
'
Takhtas
'
with handles, apparently
used for files, as
'
stovana
'
;
the oblong tablets with envelopes
as
'
lihitaka,' or
"
letter "
;
the documents on leather as
'
anadilekha,' or
"
rescript," etc. It is evident that the
clerks of those ancient offices had quite as keen a sense for
bureaucratic distinctions of this kind as the Babu of modern
TRADITION OF INDIAX BDIIGRATIOX. 383
India, who would never make a mistake about supplying him-
self with
"
octavo note
"
for his D.O.'s, foolscap for his
"
fair
dockets," or slips for his
"
ofi&ce memos."
The necessarily brief notes here presented will suffice to show
that these Kharoshthi documents are bound to bring back to
light many aspects of life and cultiue in an early period of
Central-Asian history which seemed almost entirely lost from
our field of vision. The very nature of the contents and the
complete absence of similar records of ancient date in India
itself will render their full elucidation a slow and laborious
task. But whatever revelations of interesting detail may
be in store for us, one important historical fact already stands
out clearly. The use of an Indian language in the vast majority
of the documents, when considered together with their secular
character, strikingly confirms the old local tradition recorded
by Hiuen-Tsiang and also in old Tibetan texts, but hitherto
scarcely credited, that the territory of Khotan was conquered
and colonised about two centuries before our era by Indian
immigrants from Takshasila, the Taxila of the Greeks, in the
extreme North-West of the Punjab. It is certainly a significant
fact that within India the Kharoshthi script used in our tablets
was peculiar to the region of which Taxila was the historical
centre. Neither the language nor the script presented by our
documents can satisfactorily be accounted for by the spread
of Buddhism alone, seeing that the latter, so far as oiu" available
e\idence goes, brought to Central Asia only the use of Sanskrit
as the ecclesiastical language, and the writing in Brahmi
characters.
It seemed strange that these ruins far away in the barbarian
North, overrun by what Hindu legends vaguely knew as the
"
great sand ocean," should have preserved for us in an Indian
language records of every day life older than any written docu-
ments (as distinguished from inscriptions) that have as yet come
to fight in India proper. But from the first there was ample
evidence pointing to this chronological conclusion. The
Kharoshthi writing of the tablets and leather documents, as
already stated, showed close agreement in its palaeographic
features with the Kharoshthi inscriptions of the Kushana kings,
384 DECIPHEKMENT
OF
DOCUMENTS.
whose rule in North-Western India
undoubtedly falls mainly
within the first two centuries of our era. This testimony
was
fully supported by the fortunate discovery in another ruin,
N. VIII., of a unique tablet showing by the side of Kharoshthi
some lines of Brahmi characters which clearly display the
peculiarities of Brahmi writing of the Kushana period. The
evidence of the coins was equally eloquent, since the numerous
finds made during my stay included only Chinese copper pieces
of the Later Han Dynasty, whose reign came to a close in
A.D. 220. Finally there was in the use of wood as the only
writing material, apart from leather, another proof of considerable
antiquity. Though the use of paper is attested in Chinese
Turkestan from at least the fourth century a.d. onwards,
yet I failed to discover even the smallest scrap of paper among
1 all the ruined houses and ancient rubbish heaps.
r"^^^^
But with all these indications at hand, I felt particularly
gratified when a recent discovery revealed the incontrovertible
chronological evidence for which I had always longed. It came
from one of the small pieces of wood inscribed with single lines
of Chinese characters, of which the excavation of N. xv.
ultimately
yielded up over twoscore (for an illustration, see
*
p.
296).
Their preliminary examination at Kashgar and in
the British Museum seemed to show that they contained brief
orders, chiefly concerning the movements of specified individuals
who were to be arrested or allowed to pass certain posts, etc.
References to ancient localities such as Kucha, Shen-shen,
Su-le or Kashgar, and the description of two persons as
'
Ta-
Yue-chi,' i.e.,
Indo-Scythians, were points possessed of con-
/siderable
historical interest. But it was only when Dr. S. W.
Bushell, the distinguished Sinologist, had occasion to examine
/
these
Chinese tablets that one of them was discovered to be
I fully and precisely dated. The initial characters, as since
I verified
also by my friend Professor Chavannes, plainly and un-
\
mistakably
indicated the fifth year of the Tai-shih period of the
v Emperor
Wu-ti, corresponding to
269
a.d.
Thus an exact date has at last been found which fixes the
period when this remarkable collection of documents accu-
I mulated.
A careful comparison of the years recorded in the
CHINESE DATED RECORD. 385
Kharoshthi tablets with the relative depths of the layers of
rubbish in which they were found, may yet enable us to determine
a number of important chronological details. A discussion of
these would be out of place here, but I may call attention tc
a point of more general historical interest. We know from
the Chinese Annals that the sovereignty of the Imperial Govern-
ment over Eastern Turkestan, which during the Later Han
Dynasty (24-220 a.d.) had vigorously asserted itself, was
rudely shaken and for long periods practically effaced under
the far less powerful dynasties which followed until the advent
of the Tangs (a.d. 618). Of Wu-ti, however, the first Emperor
of the Western Tsin Dynasty (a.d.
265-290),
it is distinctly
recorded that he succeeded in re-establishing Chinese authority
in the westernmost provinces during his reign. The discoveries
just described fully confirm this, as they show that Chinese
posts then existed at this ancient settlement and probably also
in other parts of Khotan territory. It is difficult to believe
that the buildings of the ruined site continued to be inhabited
for many years after Wu-ti's time. We are thus tempted to
connect its abandonment with the great political and economic
changes which undoubtedly accompanied the withdrawal of
Chinese authority from these parts.
But, whatever the historical events may have been, there
was ample evidence in this refuse-heap for the closeness of
commercial relations with China. The pieces of remarkably
well-finished lacquered ware and the bits of delicately woven
silk fabrics, which lay embedded here with other litter, could
only have come from the far eastern parts of the Empire,
Whether the fragments of cut green and yellow glass, showing
great transparency, and very different from the coarse material
found at other sites, were also of foreign origin, has not as yet
been definitely decided. But the discovery of a seal made of a
piece of ancient Chinese porcelain plainly points to such imports.
Far more numerous, of course, were the objects for which local
manufacture may be assumed. Mixed up with pottery frag-
ments of all kinds there were rags of cotton and woollen materials
some sho\ving delicate patterns and colours
;
eating-sticks and
spindles of wood
;
remains of leather shoes and women's slippers
25
386 DECIPHERMENT OF DOCUMENTS.
coloured red just like the
'
Charuks
'
still in favourite use
;
thick wooden horse-combs
;
spoons made of bone
;
and other
articles of domestic use. The large number of sheep's knuckle-
bones, often painted red or yellow, shows that gambling with this
simple form of dice must have had its votaries in the house-
hold. Besides these there were found also an ivory die, of
the peculiar elongated shape still popular in India, and marked
with round punches on its four sides.
When the rubbish had all been cleared out, I found that one
corner of the room was occupied by a circular mud-platform,
about
5
feet in diameter and
3
feet high, with its centre hollowed
out to a depth of 10 inches. The men from Niya at once ex-
pressed the belief that it was a trough, such as is used to this
day in the houses of better-class people for keeping flowers
fresh under water or wetted leaves. If the contrivance really
served this object, it must have continued in use during all
the time the rubbish around it was accumulating. For the
hollow on the top was found fiiled only with drift-sand.
The other rooms of the house had evidently been cleared long
ago. Yet even here the search was not entirely fruitless. In the
sheltered corner of the apartment next adjoining N. xv. I came
.upon a heap of wheat straw which, as the piece of matting below
it showed, must have fallen from the roof. Among the straw
there were stalks still retaining their grains in perfect preser-
vation. There was no pony at hand like Turdi Khwoja's
venerable animal at Dandan-Uiliq on which to try the value
of the antique straw as a foodstuff. But my quaint old guide
himself, the
"
Aksakal of the Taklamakan
"
as he was called
in camp, just then turned up from Khotan. His arrival was
greeted by me with joy ; for instead of doubtful antiquarian spoil
Turdi brought this time my long-expected mails, the postal accu-
mulations of more than a month, and various much-needed stores.
It is impossible to refer here in detail to the ruins subsequently
explored at this site. They were found scattered in small
detached groups over a wide semicircular area, up to a distance
of one mile and a half to the north of my second cam]). Inter-
esting as these excavations were, they yielded but a comj^ara-
tively small harvest in written documents. The two dozen
I
RELICS OF ANCIENT INDUSTRIES. 387
tablets brought to light comprised, however, an important find._
In one of the houses belonging to the northernmost group I found
the small tablet which furnished the unique specimen of Brahmi
writing already referred to. In the same dwelling some fine
specimens of architectural wood-car\'ing (see
p.
349)
came to
light in the shape of massive corbels, showing flower ornaments
which are closely allied in style to those found in the Graeco-
Buddhist sculpture of ancient Gandhara. Less artistic but
decidedly curious were the wooden boot-lasts we discovered
in the same house
;
also a large cupboard had been left behind
by its last inhabitants. A few hundred feet eastwards, close
to some high sand dunes, the embankments of a small tank,
48
feet square, could be clearl}- made out. One of the poplars that
once gave shade to its water stiU raised its gaunt, bleached
trunk to a height of 12 feet, as seen in the photograph at the
head of this chapter.
By the 13th of February I had completed the examination
of every ruined structure that could be traced under the sand.
From a high ridge rising about three miles beyond the northern-
most ruins I searched the ground with field-glasses further
towards the desert. But no indication of structural remains
could be discovered over the great expanse where absolutely
bare dunes alternated with equally denuded banks of loess. I
was thus able to leave this fascinating site, which had yielded
such precious antiquarian spoil, with a good conscience.
Almost the whole of my stay had been a succession of
deliciously clear days \vith bitterly cold nights and mornings,
the minimum thermometer usually showing temperatm-es
from
6"^
to
9- F. below zero. It was striking evidence of the
remarkable clearness of the air that early on the morning of the
nth of February the Sub-Surveyor's sharp eyes distinctly
sighted the snowy mountains south of Niya, some 120 miles
away. Yet I knew that such favom-able conditions for desert
work could not be expected to serve us much longer. I thought
of the number of sites that still remained to be explored before
the season of sand-storms would put an end to my explorations,
and consequently realised the necessity of setting out for those
fresh fields of work as early as possible.
25*
RUINED BUILDINGS WITHIN
ENDERE FORT.
CHAPTER XXVII.
THE RUINS OF ENDERE.
On the 13th of February I once more started my caravan, back
to Imam
J
afar Sadik. As I passed one ruin after the other
famihar to me from the incessant work of the last weeks, I took
occasion to collect specimens of the various kinds of wood from
the withered trees of ancient orchards. Where will it be next
that I can walk amidst poplars and fruit trees planted when the
Caesars still ruled in Rome and the knowledge of Greek writing
had barely vanished on the Indus ?
I had already heard at Niya, of ancient remains in the desert
near the Endere stream about half-way towards Cherchen, and
subsequent information decided me to select them for my next
explorations. It would have been difficult to take all my former
labourers along to this new site, as the distance was great and
TO THE YARTUNGAZ
RIVER. 389
the men were exhausted by the hard work of the last three
weeks. The fresh set of men needed could only be secured from
Niya. It was hence a welcome surprise when on arrival at that
evening's camping-place I was met by the Deputy of the Beg of
Niya, who brought not only a fresh Kashgar mail sent on by the
thoughtful Amban of Keriya, but also assurance that all arrange-
ments had been made for the timely dispatch of the fresh con-
tingent. The next day's march to Imam
Jafar
Sadik was easy,
and it rejoiced me to hear once more the rusthng of the leaves
in the luxuriant jungle that marks the end of the Niya stream.
There was no sign yet of the approach of spring, but even in its
winter sleep this living forest was a great change after the silent
sands and ruins among which I had dwelt. At the Mazar
hospice I enjoyed for one brief afternoon the cheerful warmth
of a fireplace and indulged in that long-desired luxury, a thorough
'
tub.' But there was plenty to do besides, as I despatched
from there my mails to Europe and India with the first notice of
my recent discoveries, and also settled all accounts with the
labourers and the Sheikhs of the Mazar.
I had all along thought that the Endere ruins might be reached
by striking straight across the desert to the east of Imam Jafar
Sadik, instead of returning first to Niya and thence marching
along the Cherchen road. At first all knowledge of such a direct
route was stoutly denied, but in the end one of the shepherds
from the Mazar acknowledged that he had more than once Nisited
flocks grazing on the Yartungaz stream, the one flowing into the
desert next east of the Niya River. So he and Abdurrahman,
a half-crazy devotee of the shrine, who claimed to have paid
a visit to those ruins, were engaged as guides, and the 15th. of
February saw us once more steering amidst the sand dunes.
Two miles beyond the Mazar all vegetation was left behind.
Then we crossed two steep Dawans rising to about 150
feet and
toiled on through high sand-hills for about six miles until large
patches of gravel soil were struck where camels and ponies
marched with ease. A supply of ice brought along from the
Mazar enabled us to camp that evening at a spot half-way to the
Yartungaz stream, while low tamarisk scrub and some Kumush
supplied fuel.
390 THE RUINS OF ENDERE.
Marching south-eastwards on the following day over gravel-
covered slopes and low sand-dunes, we crossed a wide belt of
tamarisks and reeds which was said to be reached at times by
a small stream known as the Suziije Darya. Three miles west
of the present course of the Yartungaz stream an older bed of
it was passed, completely dry. It was at once succeeded by high
sand-ridges, such as usually accompany these desert rivers, and
I felt heartily glad for the sake of our tired animals when at last
in the evening the glittering ice of the stream came in sight,
which meant for them rest and water.
On February 17th we followed the Yartungaz River, which
at our camping-place had a breadth of about thirty yards, down
to the point where it is absorbed by the sands. It was a march
of some eighteen miles, all through a belt of jungle which
gradually widens out to a breadth of three or four miles. Every-
where we saw the tracks of sheep, but met no living creature
until we emerged on the clearings which have been made by a
small colony in the fertile area marking the furthest reach of the
summer flood water. There is ample land and probably also
irrigation water available for several villages. But only four
families of agriculturists have established themselves here
during the time of Niaz Hakim Beg
;
and though they are
sufficiently well off to employ labourers, no more ground has been
cleared than twenty to thirty hands are able to cultivate. Even
thus, it was a strange feeling for me to ride once more past fields
and irrigation channels. Our excellent Darogha had marched
ahead. So Abdul Karim, the foremost
*
Dehkan
'
of the little
settlement, a fine-looking old man, was ready to welcome us.
His father had come from Faizabad in Badakhshan on a pil-
grimage to Imam Jafar
Sadik, and on settling down received a
grant of land here. Abdul Karim was evidently proud to air
what little knowledge of Persian he retained and to do the
honours of this forlorn outpost in the desert.
For the last five or six years the Yartungaz stream has shown
a tendency to shift its final course westwards. Hence the pre-
sent main channel loses itself in the sand some four miles to the
west of the little colony. The water needed for irrigation is
diverted into the old bed by a
'
band
'
which I had passed some
ARRIVAL AT ENDERE 8TUPA. 391
seven miles higher up. But even thus the supply is getting pre-
carious, and as the labour available is quite insufficient to cope
with the vagaries of the stream, the people of Yartungaz Tarim
have started fresh cultivation at the new debouchure as an
alternative in years when their old irrigation channel is likely
to fail. Given an adequate supply of laboiu: for the maintenance
of dykes and canals, it is certain that the area of cultivation could
be greatly extended. If fields were to replace the expanse of
jungle, covering at present an area of at least twenty-live square
miles, the terminal oasis of the Yartungaz River might well
present the conditions we must assume to have once existed
round the ancient site below Imam
Jafar
Sadik.
Here, too, it was impossible to obtain any clear information
as to the ancient remains for which I was bound. But our
energetic Darogha succeeded at last in hunting up reliable guides
to the grazing-grounds of the Endere shepherds, and our supplies
of foodstuffs and fodder could be replenished from the surplus
stock of the little colony. Two long marches brought my cara-
van across the region of sand-dunes to the forest belt of the Endere
stream. Immediately after leaving Yartungaz Tarim a formid-
able
'
Dawan
'
of sand, apparently about i8o feet high, gave
trouble to the camels and ponies. But as usual the ridges further
away from the river grew lower, and broad depressions between
them covered with soda efflorescence offered easier ground.
Ice brought from the Yartungaz and the contents of my water-
tanks enabled us to camp half-way without the necessity of
digging a well, and the quantity of dead tamarisk stems close at
hand though scanty sufficed for the camp fires. After we left
Yartungaz a fairly strong wind began blowing from the north
or north-east, and the dust-haze it raised displayed an ominous
persistence.
Late on the 19th of February we crossed the chain of high
dunes which skirt the left bank of the Endere stream, and then
continued
to the south-east along what our guides called the
'
old Darya
'
of Endere. Here, too, the river has in recent
years been shifting
westwards, so that we found quite a respect-
able sheet of ice, from ten to twenty yards in width, covering
what previously had been a deserted, dry bed. On the other
392 THE RUmS OF ENDEEE.
hand this return to the eadier channel has led to the abandon-
ment of the little
'
Tarim
'
or colony that had been formed
some miles further east on the
"
new river," now in process of
drying up. It was manifest that these constant changes in the
river courses, just before the desert absorbs them, account for
the many dry depressions we had crossed since leaving Imam
J
afar Sadik.
On the following day we moved up the Endere stream for
about ten miles to where a deserted shed of rushes marked the
shepherd station of Kara-ochke-oltiirgan
("
Where the black
goat sat
"),
From there our guides struck into the desert south-
eastwards, and after another ten miles' march next morning
across low dunes I arrived at what they called the
'
Potai ' of
the
'
Kone-shahr.' It was, of course, a brick Stupa, as I had
assumed when this feature of the site was first vaguely men-
tioned to me at the Mazar. Nor was I surprised to find that it
had been dug into, no doubt in the hope of treasure. All the
same I pushed on with increased eagerness southwards, where
the remains of old houses were said to exist. The eroded ground
around the Stupa was thickly strewn with pottery fragments,
many of them coloured
;
but no trace of structural remains
appeared until I had arrived quite close to low dunes enclosing
the ruins. The rows of wooden posts that rose above the sand
were indeed a familiar sight. But the high brick walls of some
large building, and the remnants of a massive rampart encir-
cling the ruins, presented a novel feature.
The contingent of labourers from Niya had arrived just when
I was nearing the Stupa. Considering the great distance,
some 120 miles, from which the men had been brought, and the
difficulty of communicating with them over wholly uninhabited
ground, I felt not a little pleased at this well-managed concen-
tration, which enabled me to start work at once. Going over
the ground once enclosed by the circumvallation which had a
diameter of about
425
feet, I noticed near the centre rows of
wooden posts just reaching above the sand-dune which covered
this part of the area. Their arrangement in two concentric
squares at once recalled the temple cellas with enclosing passages
I had excavated at Dandan-Uiliq. A little experimental digging
MANUSCRIPT FINDS IN TEMPLE. 393
at one of the corners of the inner square soon brought to Hght
stucco fragments which had belonged to a large-sized image.
So the whole of my httle force, counting over twenty
'
Madigars,'
and supplemented by shepherd guides and other followers, was
at once set to work here.
Within an hour I had conclusive proof that my surmise was
correct. From the loose sand that filled the building more and
more pieces of sculpture in stucco turned up, resembling in
make and colouring the material used in the Dandan-Uihq
statues. My conclusion as to the approximate date of this
shrine was soon verified by successive finds of portions of paper
manuscripts. They comprised several leaves, broken in the
middle, of a Sanskrit text in ancient Brahmi characters, as well
as scraps of sheets written in that very cursive form of Brahmi
which appeared so often in the non-Indian docimients of Dandan-
Uiliq. Here, too, the language for which this script was used
was clearly not of Indian origin. The expectations roused by
these first finds were not disappointed. It took nearly two days
to clear the temple of the sand that had covered and preserved
it. The shrine seen on
p.
394
consisted of a cella, 20 feet square,
having on each side a passage
5
feet wide, and was occupied in
its centre by a large pedestal bearing originally four seated
stucco images, presumably Buddhas. But of these only the legs
and the lowest portions of the robes had survived. Life-sized
statues in the same material, all broken above the waist, but
retaining in part the vivid colouring of their robes, occupied the
foiu: corners.
At the feet of these statues and around the hexagonal base of
the central pedestal our excavation revealed in rapid succession
manuscript leaves on paper, evidently once deposited there as
votive offerings. Among the finds made close in front of the
central base the fragments of a Sanskrit text deaUng with matters
of Buddhist worship, apparently after the fashion of the Maha-
yana school, were particularly numerous. Judging
from the
palaeographical featin-es displayed by the very clear Brahmi hand
this manuscript may well have been wxitten as early as the fifth
century. The manuscript, which had the shape of the usual
Indian Pothi, must have been broken in the middle, either owing
394
THE EUINS OF ENDERE.
to folding or in the course of some previous digging by
"
trea-1
sure-seekers
"
;
for among the separately recovered packets of
leaves, right and left halves were represented in about equal]
numbers. The pagination numbers read by me ran up to forty-
six, and probably about one-half of the original foha may yet bej
restored from these pieces.
Another very remarkable find was a closely-packed roll of
INTERIOR OF RUINED TEMPLE CELLA, ENDERE, AFTER EXCAVATION.
paper, about
4
inches high and half-an-inch thick, from which
under the expert treatment secured in the British Museum the
four large folia reproduced in my
"
Preliminary Report
"
have
since been unfolded. The writing is Brahmi, of the well-known
Central-Asian type, while the text is in a non-Indian language,
perhaps identical with that represented by various manuscript
finds from Dandan-Uiliq.
As the excavation proceeded on the other sides of the cella,
curious evidence came to light of the various nations from which
TIBETAN MANUSCRIPTS.
39.3
the worshippers of this shrine were drawn. Pieces of thin paper
with a few Chinese characters and small coloured drawings were
found at the pedestal of several images. Far more interesting,
historically, was the discovery of numerous paper leaves with
Tibetan writing. They are all written on a peculiar paper easily
distinguished by its toughness and yellowish colour, and in-
variably on one side only. Already at the time of their discovery
it was easy to recognise that, with the exception of three detached
sheets showing very cursive characters, which have since been
proved to contain Buddhist prayers and religious poems, all
TWO
LEAVES IX CENTRAL-ASIAN BRAHMI, FROM PAPER ROLL (E. I.
7)
FOUND IN ENDERE TEMPLE.
(Scale one-fourth.)
Tibetan leaves and fragments had formed part of a single Pothi.
The manner in which the pieces of manuscript were found de-
posited before the various images and on the mouldings of the
central pedestal leaves no doubt that they had been distributed
purposely. In order to propitiate as many divinities as possible,
the pious owner on his visit to this shrine seems to have first cut
up the manuscript in the middle, and then proceeded to deposit
the halved leaves all round the cella.
The careful examination since made of these disjecta membra
by Mr. D. L. Barnett, of the British Museum, has proved that
they
belonged to an early Tibetan version of the Salistamba-
sutra,
a Buddhist treatise on philosophy, the Sanskrit original
396 THE EUINS OF ENDERE.
of which is known only from quotations. The great import-
ance of the Tibetan text, of which about one-half has thus been
,
recovered, has been lucidly discussed by Mr. Barnett in his
j
"
Preliminary Notice " of these discoveries published in the
Royal Asiatic Society's Journal for
1903.
Here it can be indi-
cated only in the briefest outlines. The exceptional interest of
the fragments rests not merely on the fact that they supply a
valuable criterion for the comparatively modern version of the
same text embodied in the Kanjur, and thus for the critical
analysis of this great canon of Tibetan Buddhist literature.
Being the oldest known specimens of Tibetan writing, they fur-
nish a wealth of fresh material for the study of Tibetan paLneo-
HALF-LEAF OF TIBETAN MS. ON PAPER, FROM ENDERE (E. I.
32).
(Scale three-eighths).
graphy and orthoepy. But still greater value must be claimed
for them on account of the historical significance with which
the place and circumstances of their discovery invest them.
There can be no doubt as to the political conditions with
which we must connect the finds of Tibetan texts in this ruined
temple, as well as the appearance of Tibetan graffiti that I found
covering its stuccoed walls in several places. We know from the
Chinese Annals of the Tang dynasty that, during the second half
of the eighth century, Tibetan invasion seriously threatened,
and towards its close actually destroyed, the authority of the
Imperial Government in Eastern Turkestan. These records, of
which Professor Chavannes has kindly communicated to me a
series of very interesting extracts, indicate plainly that, though
DATE IN CHINESE GEAFFITO. 397
the Tibetan advance had already (about
766
a.d.) led to the
isolation of those outlying provinces, the local Chinese adminis-
trators succeeded for a time in maintaining their authority, at
least over part of their territories. In
781
a.d. they managed
to transmit pathetic appeals for help to the Imperial court, from
which, however, there came no succour, only grants of laudatory
titles and liberal acknowledgments of official merits. In
784
A.D. their position appeared so desperate that the Central
Government considered the advisabihty of their recall. Finally,
from
791
A.D. onwards, the Tibetans possessed themselves of
Turfan and the adjoining region, and nothing more was heard
of Eastern Turkestan or the
"
Four Garrisons," as the Chinese
then styled the territories controlled from Kucha, Khotan,
Karashahr, and Kashgar.
It is a fortunate circumstance that a Chinese inscription
scratched into the cella wall close to the image in the north-west
corner (visible to the right in the photograph,
p.
394)
renders it
certain that we must read the evidence of the Tibetan finds in
the light of the Chinese records just summarised. From the
photographs taken by me of this curious graffito, Professor
Chavannes has been able to make out a considerable part of its
contents, probably as much as the loosely scratched
characters
will ever permit to be read. It mentions the return of Tsin-kia-
hing, a dignitary of the Chinese administration
charged with
official sacrifices, to his own district, apparently after the receipt
of a report concerning the death of certain military
officers with
whom he was associated. Twice the
'
Ta-fan
'
or Tibetans are
mentioned, and a reference is also made to the
"
Four
Garri-
sons." The date when this record was incised
is given as the
seventh year of a period which, owing to the defective
preserva-
tion of the first character, may be read as Kai-yuen
or Cheng-
yuen, the year meant corresponding either to a,d,
719
or
791.
In the present state of our knowledge it seems
hazardous
to
decide definitely
between these two dates. The
later one,
791
a".d., would singularly agree with that recorded
by the Annals
for the final
subversion of Chinese rule, and seems to find some
support
also in the fact that the latest Chinese
document
of
Dandan-Uiliq
dates from
790.
In favour of the
earlier
date.
>
398 THE RUINS OF ENDEKE.
719
A.D., it may be mentioned that only Han coins were founc
among these ruins, and also that the sculptural remains of the
Endere temple seemed to me to bear a somewhat older charactei
than those of the Dandan-Uiliq shrines. However this may be,
it can be considered quite certain that the date when this Chinese
graffito was scratched into the wall could not have preceded by
many years the deposition of the various votive manuscripts
and the subsequent abandonment of the shrine. The rough, and
not very hard, plaster of these cella walls was not a material that
could remain intact for a long period without repair, and with
r its renewal all these casual scratchings of pious visitors would
no doubt have vanished. This consideration fixes the second
half of the eighth century as the latest possible time for the
production of the Tibetan as well as the other manuscripts found
here. The same applies to the Tibetan graffiti which, owing to,
their very cursive writing, have not yet been fully deciphered.
r"'
The very numerous little rags which were found scattered ii
/ front of the various image bases are undoubtedly votive of^ering^
of a humbler kind. They comprise shreds of many fabrics, fror
elaborately woven Chinese silks and printed cottons to the
simple
'
Kham,' a kind of buckram, mentioned in Hiuen-Tsiang's
account of Khotan, and still worn by the common folk through-j
out this region. The variety of this collection vividly remindec
me of the wonderful display of rags that graces the approaches t(
the resting-place of the holy Imam
J
afar and other saints
throughout Turkestan. Islam has indeed little changed the
popular type of ex-votos which were in vogue during Buddhist^
times, and which in this case have provided for us a sample 'H
collection of ancient fabrics of no small archaeological interest.
|
During the days following I had almost all the buildings
within the enclosing ramparts cleared of sand. These excava-
tions furnished interesting data as to the methods of construction
employed, but failed to throw much light on the original destina-
tion of the whole of this ruined settlement. The large brick
building to the east of the temple, of which a portion left exposed
by the sand is seen on
]). 388,
occupies with its massive walls of
sun-dried bricks three sides of a quadrangle over 100 feet square.
The dimensions of its rooms suggest public use
;
but as, with the
KEMAINS OF ANCIENT RAMPARTS. 399
exception of a walled fireplace or two, they were found com-
pletely empty, there was nothing to prove the true character
of the structure. Were these the quarters of a well-to-do
monastic establishment which found it advisable to protect
itself by walls and ramparts ? Or do the latter mark a fortified
frontier-post which sheltered also a Buddhist temple ?
LOWER FLOOR ROOM OF RUINED DWELLING-PLACE, ENDERE FORT.
In a row of small rooms built of timber and plaster,
which
stood to the north of the shrine, there was one that appeared to
have served as a httle chapel. Its wall on one side was occupied
by an elaborate fresco, which seems to have represented
a Buddha
surrounded by his former epiphanies. The wall had been broken
at five feet from the ground, but the colours and outlines of the
remaining part were in very fair preservation.
In the same
little room we found a well-executed small painting on wood,
L
400 THE EUINS OF ENDERE.
\ showing the famihar figure of the elephant-headed Indian god of
I ^wisdom, Ganesha. To the south of the temple my excavations
revealed a small double-storied building, of which, however, only
the lower floor rooms remained. They had no doors, and were
evidently underground apartments intended for use in the
winter. The large fireplace found in one of them, with its
elaborate mouldings, is seen in the photograph reproduced on
p.
399.
The circumvallation, which originally consisted of a solid
rampart of clay about
30
feet broad at the base and
17I
feet high,
had survived only in parts of the south face, flanking a gate, and
in much decayed segments elsewhere. On the top of the ram-
parts ran a parapet of brickwork
5J
feet high, and behind it a
platform that seems to have been paved with bundles of brush-
wood, manifestly for the sake of greater consistency. There
was nothing to tell of the attacks from Tibetans or other foes
which this little stronghold may once have resisted. But its
walls have certainly helped to ward off that worst danger ofj
ancient remains in the desert regionthe erosive action of winds
;
and moving sands. Looking around from the ruined ramparts,
it was easy to realise that the original level of the surrounding
ground had been lowered at least ten feet by erosion. Inside
the circumvallation the drift-sand, once accumulated, was less;
liable to be shifted by the winds, and thus provided a protecting
cover for the ruins.
The extent of the erosion which the whole area has undergone]
was brought home to me by the closer examination I made of the^
Stupa already noticed. On surveying the structure I found that
'
the lowest line of bricks in its foundation was about ten feet
higher than the level of the ground in the immediate vicinity.
Kept down by the weight of the masonry, the loess soil below
the Stupa has resisted erosion and now appears like an additional
base, making the structure look considerably higher than it is in
reality. Though the outer casing of bricks has suffered a great
deal, I succeeded with a little trouble in ascertaining the original
proportions of this
"
memorial tower." On a square base,
measuring
23
feet on each side and about
7
feet in height, there
rose a solid dome 16 feet in diameter and approximately of the
DIMENSIONS OF STUPA. 401
same height. Thus the total dimensions agree very closely with
\
those of the little Stupa at the Niya River Site, though the
arrangement there of a double basement introduces a marked
difference in the detailed proportions. In Indian Stupas the in-
crease of height relative to diameter maygenerally be considered
a sign of later date, and a comparison of these two Stupas, the
time of which can be approximately determined, seems to con-
firm this observation also for this northern region.
On the 26th of February my explorations at this site were
completed, thanks to the energy with which the work of exca-
vation had been carried on, from early morning until after night-
fall by the light of bonfires. The interesting finds showed plainly
that I had reached the border-line beyond which Indian influences
yielded to Chinese. A move further east would have carried
me beyond the limits of the territory with the archaeological
exploration of which I was concerned. Besides, the time that
remained at my disposal seemed none too ample for the expedi-
tions I had yet to make to ancient localities north of Keriya and
Khotan. So I reluctantly decided that the time had come to set
my face again westwards. There was consolation in the thought
that it meant a start on the journey which was to bring me
back to Europe and to dear ones not seen for long years.
26
402
CHAPTER XXVIII.
EXPEDITION TO KARADONG RUIN.
On the 26th of February I left the Endere ruins on my rapid
march back to Niya. The river, which was crossed at a point
some ten miles to the south-west of the old site, had through
the warm sunshine of the last few days lost its coating of ice in
the shallower parts
;
but over the main current of the stream,
about
15
yards across, the ice was still strong enough to bear
heavy loads. One by one our camels were got safely across.
Then the water-tanks were filled once more for the passage
of the broad desert belt towards the Yartungaz River. March-
ing along an old bed of the Endere River to the south, we
passed through luxuriant Toghrak jungle for another ten miles,
and late in the evening reached the deserted shepherd's hut of
Tokuz-kol. The name means
"
nine lakes," but of water there
was none.
On the morning of the 27th we still steered due south through
a level plain covered with Kumush. All traces of the true sand
desert for a time disappeared. The change of scenery was all
the more striking as the distant mountain rampart of the Kuen-
luen was clearly visible during the morning. Though sixty
to eighty miles away from the range, we could distinguish a
series of prominent peaks with the glaciers descending around
them. They were duly recorded on the plane-table, but only
just in time
;
for after midday, when we had struck the track
which leads from Keriya to Cherchen, a strong north wind
raised such a dust-haze that soon the mountain view vanished
like a vision. Here the Endere shepherd guide was discharged,
i
KETURN TO KERIYA 403
not until I had given him a little present for his children. I
had met them, two sturdy little boys, on the way, but had then
nothing to offer but bits of chocolate from my saddle-bag.
This they could not be induced to try, though my little terrier,
just to encourage them, readily swallowed some of it. I hope
that the piece of a Russian sugarloaf I got out from the baggage
and sent them was received with more confidence.
After reaching the Cherchen road, once the great line of
communication to China, but now a lonely desert track with
practically no traffic, we still had a long march to do that day.
For over twenty miles we rode to the south-west, over hard-
grained sand, with scarcely any dunes and bare of vegetation,
until Yoke-toghrak was reached. There a small patch of
tamarisks and Toghraks offered scanty fuel, and some brackish
water was found in wells about 6 feet deep. The camels did
not turn up until close on midnight, and dinner was accordingly
an affair of the small hours of the morning. The next day
brought, however, an easy march to the Yartungaz River,
where men and beasts could be made comfortable. On the
way I had the satisfaction to fall in with Tila Bai, my honest
pony-man, who was now bringing mails and sadly wanted
articles from the stores I had left at Khotan. It is always
a pleasure to receive bags full of
"
home mails." But I enjoy
the sensation most when it comes unexpectedly, and there is time
to sit down by the roadside and pore peacefully over the con-
tents, as I could this time. A look over the Weekly Times,
nearly three months old, put me again in contact with the
affairs of the far-off West and East.
The two long marches which brought me back again to Niya
yielded pleasing variety in the little lakes and marshes we passed,
They are fed by springs, the water of which, just released from
the grip of frost, was flowing plentifully into the reed-covered
lagoons. From the Shitala Darya, a watercourse similarly
formed, at which we halted, there stretched an uninterrupted
jungle to within some miles of the Niya River. Its eastern
bank proved to be flanked, just like that of all the other rivers
that flow into the desert east of Khotan, by a high range of sand
dunes. This was crossed near the Mazar of Shitala Padshahim,
26*
404 EXPEDITION TO KARADONG RUIN.
a simple collection of posts with yak tails and fluttering rags,
and at last, after a ride over much boggy ground, I again
entered the little oasis. In the twilight it seemed like a return
to civilisation. Since starting from Niya on the 23rd of January
I had covered over 300
miles in a great oval loop. Yet when
the positions indicated on the plane-table for the starting and
closing points of our route came to be compared, the difference
proved to be only three-fourths of a mile in longitude and a
little over a mile in latitude.
Leaving my
"
goods train " of camels to follow behind, I
covered the distance from Niya to Keriya, some eighty miles,
in two stages. There I was busy at work with reports and
letters, and with rearranging my baggage. The weather was
rapidly getting warmer ; hence all heavy winter clothing was
to be left behind before the start for the ancient site I next
intended to visit far down the Keriya River. As arrange-
ments also had to be made for labourers to accompany us for
excavation work, and for the supplies that men and animals
needed, the two days, which were all I could allow myself,
were indeed no time of rest for me or Ibrahim, my energetic
Darogha.
Huang-Daloi, the kindly Amban, opportunely returned the
day after my arrival from a little tour of criminal investigation.
So I had the wished-for opportunity of thanking him personally
for all the help which I had experienced at a distance. From
the stores that had arrived from Khotan I had sent him the
best selection of tinned goods I could offer. Everything
seemed to be duly appreciated, and the return presents in the
shape of fodder supplies, sheep, etc., were quite overwhelming.
When the Amban, a few minutes after my departure, paid his
return visit, I was able to satisfy his curiosity about my
finds
with specimens of ancient tablets, etc. With the historical
sense which all educated Chinese seem to possess, he at once
rightly surmised that the use of wood as a writing
material
indicated a period corresponding to that when split bamboos
were employed in China previous to the invention of paper.
Extravagant
rumours as to my discovery of coffers full of
gold, etc., seemed to have run through all the Bazars far away
ALONG THE KERIYA RIVER. 405
to Khotan and beyond. But I could see that Huang-Daloi had
quite sensible notions of what I was looking out for, and of what
I had found. So we parted once more in the friendliest fashion,
and with mutual confidence. --(
Thanks to the Amban's energetic assistance, I was able to
set out on the 7th of March for my next objective. It was
the ruined site of Karadong, situated in the desert some
150 !
miles north of Keriya, to which Dr. Hedin had paid a short
^
visit in
1896,
on his memorable march down the Keriya Darya.
I knew from the accounts given by Turdi, whose
"
treasure-
seeking
"
expeditions had twice extended to this place, that the
remains of this so-called
"
ancient city " (which he called
Aktiken) were very scanty. Yet I felt that my duty demanded
a personal examination of the ruins. For the loss of time
which their great distance implied I decided to make up by
hard marching.
With the baggage lightened and my camels partially relieved
by the hire of fresh animals, I was thus able to push on in three
days to the point where I had first struck the Keriya Darya
from Dandan-Uiliq. The aspect of the river-banks was still as
bleak and bare as two months before, but in place of the glassy
sheet of ice there now rolled a muddy current, fed by the melting
of the ice that had covered the marshes and pools about Keriya.
It was a regular spring flood from the
'
Kara-su,'
{"
black
water
")
feeders of the river, while months would yet pass
before the flood of
'
Ak-su '
("
white water
")
would bring
down the melting snows of the mountains. At the Burhan-
uddin
Mazar I was cheerfully welcomed by the
'
Sheikhs,'
who evidently remembered the handsome offering made at
my
first visit
;
and I spent there pleasant hours, busily writing in
the sheltered little loggia of the mosque. When I left to
catch up my caravan, Ghazi Sheikh, the senior of the priestly
fraternity, insisted on accompanying
me. He was a jovial old
man and quite looked the
'
Bai,' or capitahst, which he was
according
to local notions, having at least a thousand sheep
grazing along the river. He knew, of course, every Uving soul
of the little community of nomadic herdsmen and the name of
every
grazing-ground.
So it was easy for our Darogha to
406 EXPEDITION TO KAKADONO EUIN.
strengthen the band of labourers I had brought from Keriya
by fresh recruits from among the shepherds. The men joined
us readily enough
;
for uncouth and
"
jungly
"
as their appear-
ance was, in rough furs and sandals made of goatskins, yet
these supposed
"
semi-savages
"
were quite alive to the chance
of earning a little hard cash that might come in useful on their
periodical visits to Keriya and its Bazars. Thus my band
kept swelling on the way like a small avalanche.
The route which we followed for three days from Kochkar
Oghil downwards was new to me, but space does not permit
more than a passing reference to one distinct change in scenery.
The river, which down to this point had occupied a deep and
narrow bed winding in rapid turns, now spread itself out in
broad reaches. Though the channel actually filled with water
was at the time only 80 to 100 yards wide, yet the clearly
marked bed of the summer floods attained in places the im-
posing breadth of quite half a mile. The belt of vegetation,
which accompanies the river on its course through the desert
sand, did not spread out in the same proportion. But the
increasing height of the Toghraks and the thickness of the
Kumush beds showed that the moisture received from the
river was plentiful wherever it reached.
On the 12th of March we crossed a high Dawan appropriately
named
'
Yoghan-kum '
("
High Sands
"),
which juts out
transversely into the river-bed, and is faced on the opposite
eastern bank by similar high ridges of sand. But this obstacle
once passed, wide room offered itself to the vagaries of the
stream. From the height of the Yoghan-Kum I could make
out no less than three dry beds spreading in different directions
like the fingers of a hand. We followed the middle one
a
wide, flat Nullah in which the yellow Kumush beds swayed by
the breeze looked curiously like fields of ripe corn, down to
where it met the actual river-course again near the shepherds'
station known as Tonguz-baste. Here Ghazi Sheikh's flock
was established for the time being. So hospitable offerings of
sheep and milk turned up that evening. As usual, the end of the
former bed was marked by a lagoon of fresh water comnnini-
cating with the actual river-course, yet keeping its water clear
THE FIRST SAND-STORM. 407
of the mud which discoloured the latter. It seemed a favourite
haunt of wild duck, of which hundreds were disf)orting them-
selves on this pretty sheet of water. That my camp was
placed within 200 yards of its bank seemed in no way to disturb
the birds. Their loud calls sounded strange to me after the
stillness of my desert camps.
I knew from Dr. Sven Hedin's account that the ruins I had
come in search of lay in the desert, within a march to the north-
west of Tonguz-baste. Mullah Shah, an experienced shepherd
who was to guide us, turned up late at night, and after prolonged
protestations of ignorance, acknowledged that he had twice
visited Karadong. With him came another shepherd, Mu-
hammad Shah,
"
the hunter " (Merghen), an active young
fellow who had also once seen the place. He was to help
Mullah Shah, his
'
Ustad,' or master, in finding the track.
This turned out no easy task. The morning was very hazy,
and by the time the water-tanks had been filled and a depot
made of supplies not immediately needed, a stiff north wind
sprung up which by degrees developed into a regular
'
Buran,'
the first of the season. We followed for about seven miles a
course almost due north, until we passed the westernmost
of the former river-beds above mentioned, near a little pool,
known as Toldama. retaining some flood water. Then our
guides struck to the north-west.
So far we had marched in a whirl of dust. But now, with the
increasing force of the storm, the air became so thick that it
was difficult to see even for a hundred yards. The assurance
with which Mullah Shah and his pupil continued to guide us
was doubly welcome under such circumstances. With the
sand driving into my face and accumulating under the eye-
lashes in spite of goggles, it was difficult to see much of the
route.
But I noted that after a couple of miles the scattered
groups of Toghraks were left behind and the sand dunes rose in
height.
After plodding on among them for another hour, our
guides
declared that we were near the tamarisk-covered copses
that
have given the site its name, Karadong
("
the Black
Hillocks ").
But as in the blinding dust they could not make
sure of the exact direction, I let them go on ahead while we
408 EXPEDITION TO KAEADONGf EUIN.
sought shelter behind a tamarisk-covered cone. It was curious
to watch how the sand was driven in a thick spray over the
crest-Hne of the dunes, just as if they were storm-tossed waves.
After half an hour Muhammad Shah returned with the re-
assuring news that the ruins were due west of us and not far
off. In proof he brought a piece of old pottery he had picked
up. So the march was resumed just as the force of the storm
showed signs of abating, and after another two and a half
miles, over fairly high dunes, we arrived at the ruins.
The remains of Karadong proved to consist mainly of a
ruined quadrangle, which was formed by a mud rampart about
235
feet square, with rows of timber-built rooms over it. Within
this quadrangle, which was crossed by two large dunes rising to
about 20 feet above the original ground-level, I could just recog-
nise the timber debris of a much-decayed structure sticking
out from the deep sand. The photograph on
p.
409
shows this
ruin after excavation, together with the interior of the quad-
rangle as seen from the north-east corner. Excessive erosion
had played havoc with the structures which once occupied the
exposed top of the rampart, and the lines of walls could there
be traced with difficulty. Potsherds, small broken pieces of
glass and metal, shreds of felt and scanty refuse of a similar
kind, were all that marked their former occupation.
The trying conditions under which I had reached this site
curiously reminded me of what Hiuen-Tsiang tells of the sand-
buried town of Ho-lo-lo-kia, which a local legend heard by
him at Pi-mo (see
p.
417),
placed somewhere in the great
desert to the north of Khotan. It was believed to have been
covered up by a rain of sand in consequence of the curse of a
holy visitor whom its king had treated with ignominy, and
who foretold the destruction of the town after seven days.
Only one pious man took warning and escaped to Pi-mo by
means of an underground passage.
"
On the seventh day, in
the evening, it rained sand and earth, and filled the city."
Ho-lo-lo-kia, the pilgrim tells us,
"
is now a great sand mound.
The kings of the neighbouring countries and persons in power
from distant spots have many times wished to excavate the
mound and take away the precious things buried there
;
but
ANCIENT
FORTIFIED POST. 409
as soon as they have arrived at the borders of the place a furious
wind has sprung up, dark clouds have gathered together from
the four quarters of heaven, and they have become lost and
unable to find their way."
I had not gone to the Karadong site to look for treasures such
as old Hiuen-Tsiang's
"
persons in power from distant sp)ots
"
sought
;
nor had the Buran that greeted me on the day of my
INTERIOR OF RUINED QUADRANGLE, KARADONG.
arrival made me lose the way. But the result was not very
different if judged by the
"
finds
"
which rewarded the work
here. I soon convinced myself that no other ruins of any kind
could be traced in the neighbourhood besides those already
briefly described. Even pottery fragments were scarce and
limited to a small area. The natives may indeed call these
remains a
'
kone-shahr,' using the term which is applied
throughout the country to old ruins of any kind. But to talk
410 EXPEDITION TO KARADONG RUIN.
of an
"
ancient city
"
here would imply more imagination than
an archaeologist need care to take credit for.
The excavation of the ruined structures in the interior of the
quadrangle kept my little force of diggers hard at work for two
days, but there was little to reward their labour. Long prior
to the accumulation of the present covering of sand, the build-
ings must have been exposed to the full force of erosion. The
plaster of the walls had completely disappeared, and much of
their wooden framework had also crumbled into loose debris.
From the general arrangement of the rooms, however, which
could still be made out, it was evident that these structures
must have once served as dwellings.
Only Toghrak wood appears to have been used in the con-
struction of these houses and of the enclosing quadrangle.
This species of poplar grows plentifully in the jungles of all
the rivers which lose themselves in the desert. But its twisted
knotty trunks and branches by no means furnish as good a
building material as the Terek or white poplar, the
Jigda,
and
other trees planted in the cultivated areas. At Dandan-Uiliq
and at the ancient site beyond Imam Jafar
Sadik only timber
of these latter trees seems to have been used for the framework
of houses, which accordingly there bore a far more finished look
than at Karadong. At those other old sites the dead trunks of
Terek and other trees dependent on cultivation formed a con-
spicuous feature. But around Karadong I looked for them in
vain. Dead trees rising from between the sand dunes were
plentiful, but they were all old Toghraks, just as are still found
growing luxuriantly along the recent river-beds a few miles
to the east. I concluded from this observation that cultivation
could not have existed to any extent in the vicinity of the
Karadong site at the period from which its buildings date.
What then can have been the purpose of the latter, situated
as they evidently were in the narrow belt of forest land between
the desert and the river ? Keeping in view the position and the
peculiar plan of the ruined structures, I think the suggestion
may be hazarded that we have here the remains of a fortified
frontier post or roadside Sarai. A remark of Mirza Haidar. the
Moghul leader and historian, makes it very probable that the
FIND OF ANCIENT CEREALS. 411
Keriya River reached the Tarim as late as the sixteenth century.
Its old course across the -desert can be followed even now with-
out serious difficulty, and certainly forms the most direct route
between Khotan and the ancient settlements of Kucha and
further north-east. Karadong lies about half-way between the
Tarim and the line of oases stretching to the east of Khotan,
and a small post established here would have answered the
purpose of guarding the route and protecting the approaches
WOODE.N
GATEWAY OF
RUI.NED
QUADRANGLE,
KARADONG,
AFTER
EXCAVATION.
from
the northern
region.
The age of the ruined
structure
is
approximately
indicated
by the coins I picked up in its imme-
diate
vicinity,
which
were all copper
pieces of the Later Han
dynasty,
showing
long wear.
The best
preserved
portion of this ancient Sarai or post was
the large
wooden
gateway
which my subsequent
excavations
brought
to light on the
eastern face of the
quadrangular en-
closure.
It formed
a square
of 22 feet, and its roof, perfectly
intact,
reached
the top level of the rampart.
Besides a broad
central
gateway,
closed
by a massive wooden door of two leaves.
412 EXPEDITION TO KARADONG- RUIN.
there were two narrower passages flanking it on either side. I
noticed exactly the same arrangement in the gates of all Yamens
I visited. As the whole of this gate was filled up with sand
to its ceiling,
14
feet above the floor, it took two days' hard
digging before we had cleared it. Above the gate there once
rose another storey, but of this there remained only a few posts
and a thick earth flooring. Embedded in the layer of rubbish
that covered this we came upon a little store of remarkably
well-preserved cereals. There were a couple of pounds of
'
Tarigh,' a kind of pulse still cultivated about Keriya, to-
gether with small quantities of wheat, rice, oats, another sort
of pulse, some roots apparently used as condiments, and a
capful of large black currants dried perfectly hard. I had a
small quantity of the
'
Tarigh
'
boiled, and found the antique
porridge made of it useful for glueing envelopes.
While this excavation was still proceeding, we had a return
of the Buran that greeted us on our arrival. Though the force
of the wind, this time from the south-west, was somewhat less,
the driving sand made it decidedly uncomfortable both in and
outside the tent. As the supply of sufficient water for my
comparatively large number of men was also a serious difficulty,
I felt heartily glad when by the evening of the 17th of March
our work at this desolate spot was concluded. Next morning
I left Karadong, just as I had reached it, in an atmosphere thick
with dust and quite oppressive by its haziness. The look of the
desert harmonised with the mournful news conveyed in the
small mail sent on from Keriya which met me half-way that
day. A short letter from home transmitted via Samarkand
and Osh, and a communication from Mr. Macartney based on
Russian intelligence, informed me of the death of our Queen-
Empress. I could see that my two Indian followers, to whom
I communicated the news, understood, and in their own way
shared the deep emotion which filled me. There were no
details to distract attention from the momentous main fact, the
disappearance from this world-wide scene of the greatest ruler
England has known since her expansion over the seas began.
413
CHAPTER XXIX:
THE SEARCH FOR HIUEX-TSIANG'S PI-MO.
My eyes were now turned to the south again, where a number
of archaeological tasks still awaited me in the vicinity of the
inhabited area. First among them was a search for the site
of the ancient town of Pi-mo. Hiuen-Tsiang had visited it
on his way from Khotan to Niya, and its probable mention
also by Marco Polo under the name of
'
Pein
'
made me all
the more anxious to identify its position. The distance and
direction which the Chinese pilgrim indicated from Pi-mo, viz.,
three hundred Li (or about sixty miles) to the east of the
Khotan capital, had long before made me look out for the place
somewhere to the north-west of Keriya. I was hence much
pleased when, on my last visit to the latter place, I heard from
the Amban himself of a
'
Kone-shahr ' said to exist in the
desert beyond Gulakhma, an oasis on the Khotan road some
thirty miles west of Keriya. Ram Singh, too, had heard about
the ruins, and in order to save time I decided to reach them now
by striking across the desert south-westwards.
Of the series of rapid marches by which I endeavoured to
effect my object, the briefest account must sufl&ce. During the
four days which saw us returning along the Keriya Darya as
fast as camels and ponies could move, I still looked in vain
for any sign of approaching spring in the vegetation of the
riverine jungle. Small wonder, considering that after the
\nndy days of Karadong the temperature showed a marked
fall, down to a minimum of 14 Fahr. of frost on the 19th of
March.
414 SEARCH FOR HIUEN-TSIANG'S PI-MO.
At the familiar shrine of Burhanuddin I picked up en route
the two guides who, under the Amban's order, had been sent
by the Beg of Gulakhma. They looked unusually reticent
and stupid, but it was too late when we found out that they
knew nothing of such a route as I wished to take. Too timid
to aver their ignorance, they thought it safest to guide us further
and further south, where at least there was no risk from the
dreaded Taklamakan. Thus, after leaving the left river-bapk
and crossing a belt of high sand dunes, we found ourselves,
on the 23rd of March, in a wide area of swampy jungle watered
from the marshes of Shivul, west of Keriya. As the local
knowledge of the guides quite gave out here, we had great
difficulty in extricating our animals from the boggy marsh,
treacherously covered with light sand, in which the Shivul
stream ends. Though there were everywhere the tracks of
flocks that had grazed here during the winter, we did not succeed
in finding a single shepherd to help in guiding us. Fortunately
we came at last upon firmer ground, where the Shivul Darya
flowed as a limpid stream in a winding but well-defined bed,
about fifteen feet broad. This helped once more to guide
our
"
guides," and ultimately, after a long and tiring march
through the dusk, we arrived at the solitary little shrine of
Arish-Mazar. Though the rustic Sheikh living near the saint's
tomb was at first greatly alarmed by the arrival of so
large a party, fodder was soon forthcoming for the tired
ponies, and big fires were lit to guide the belated part of the
caravan.
After the experience I had gained of the value of our guides
there was no alternative but to resign myself to letting them
reach again familiar ground in the oasis itself before striking
off into the desert. Accordingly we made our way south-west-
wards, through the sandy jungle in which the water of another
marsh-fed stream, the Karakir Darya, finally loses itself. The
track we followed led through a maze of tamarisk-covered
sand-cones, standing closer together than I had seen them any-
where on the borders of the true desert. Unexpectedly we
came in their midst upon the unmistakable remains of some
ancient settlement, which the few shepherds whom our Darogha
THROUGH THE SHIVUL SWAMPS. 415
had succeeded in discovering called Aktaz or simply
'
Tatilik
'
(ruins). On several small pieces of open ground, showing evidence
of considerable erosion, pottery fragments abounded and founda-
tions of mud-walled houses could be distinguished. But the
latter had been levelled to within a few inches from the ground,
and the few small objects, such as a ring of brass picked up by
my men, gave no clear indication of date.
Ultimately we emerged on cultivated land at the village of
Malakalagan, which had been formed about fifteen years ago
by
people from Domoko, the main oasis due south on the Khotan-
Keriya route. The reclamation of desert soil going on here was
a sight as cheering as it was instructive. Small irrigation cuts
were seen winding along the old tamarisk-covered hillocks of
sand that had not yet been levelled down, while between them
extended carefully-fenced fields. Here and there the Toghraks
of the desert jungle had been spared, particularly near the
huts of the settlers. But it was clear they would soon dis-
appear in a hopeless minority by the side of the avenues of
young poplars,
Jigda, and other fruit trees that were rapidly
growing up along all irrigation channels.
My previous information had clearly indicated the ruined site
I was in search of as being situated in the vicinity of the Mazar
of Lachin-Ata. The people of Malakalagan, whom I closely
questioned, did not deny their knowledge of this popular desert
shrine, but none of them would acknowledge ever having visited
the
'
Kone-shahr
'
near it. Our wanderings of the previous
two days had left me no illusions as to the capacity of our two
worthy guides. Yet as better were not to be got and time
was getting more than ever precious, I decided to set out with
them on the morning of March 25th. Old Turdi, after a long
absence on mail duty, had just joined me from Khotan, and I
could rely on the old
"
treasure-seeker's
"
instinct and experience
coming to the help of whatever local knowledge the Domoko
guides might prove to possess. Turdi had brought along two
more water-tanks previously left at Khotan, and as I took the
precaution of having all six tanks filled before the start, we
were safe from immediate risk. The three days' Odyssey which
followed, and which is illustrated by the tortuous line of route,
416 SEARCH FOE HIUEN-TSIANCx'S PI-MO.
fully bore out my misgivings. Yet there was interesting ex-
perience to compensate for the trouble and fatigue.
The first couple of miles in a north-westerly direction had
brought us to the limit of the newly irrigated land, when to my
surprise I came upon unmistakable marks of earlier cultivation
beyond. Old fields overgrown with tamarisk and thorny scrub
could be clearly distinguished by the little embankments
dividing them, as well as by the lines of dry
'
Ariks ' that once
carried water to them. My guides explained that these were the
fields of
'
old Ponak
'
village, which had been abandoned
"
in
their grandfathers' time," i.e., forty or fifty years ago. Passing
along a road still frequented by the people visiting the cemeteries
of the deserted villages, I arrived some three miles further
north-west at the southern end of the area known as
'
old
Domoko.' Here the ruins of mud-built dwellings, constructed
exactly like the modern villages of this tract, seemed to extend,
together with the interspersed orchards and cemeteries, for
fully three miles from east to west. The mud walls, strengthened
by the insertion of vertical bundles of Kumush, still often rose
4
to
5
feet above the ground, and the massive fireplaces were
intact even to a greater height. The deserted homesteads had
been stripped of all materials that could be of use, such as
beams, wooden doorposts, etc.
;
and as scarcely any sand had
accumulated about the crumbling ruins, their complete dis-
appearance was only a question of time.
The villagers accompanying me, as well as the people I
subsequently examined on my return to the oasis, all agreed
in asserting that the gradually increasing difficulty of conducting
the irrigation water sufficiently far had caused the cultivated
area of these and some other villages of the Begships of Domoko
and Gulakhma to be shifted as much as six to eight miles further
south within the memory of living men. Local tradition, in fact,
maintained that such shifts of the cultivated land, backwards
and forwards, had occurred repeatedly in the case of these
small oases along the road from Keriya to Chira. Evidence that
cannot be detailed here seemed to support the belief that this
tradition had some substantial basis, and I felt inclined to regard
the gradual change of levels consequent on irrigation deposits
DESERTED VILLAG-E SITES. 417
as a possible explanation of these repeated shifts. It would
need a prolonged investigation of local conditions, particularly
those connected with the supply of irrigation water, which is
here largely dependent on springs, in order to arrive at any
safe conclusion. But anyhow, there could be no doubt that
the ruins I saw here were the best illustration of the course
of decay through which the
'
Kone-shahrs,' or Tatis, found
along the western route to Khotan and on the outskirts of
the oasis, must be supposed to have passed once. There, too,
villages were deserted owing to irrigation ceasing from one
cause or other
;
and as they were so much further away from
the desert centre than the terminal oases of Dandan-Uihq or
the Niya River site, the heavy drift-sand could not arrive in
time and in sufficient quantity to give protection to the ruins.
For nearly three miles we traversed the desolate remains of
these village homesteads, but it was not until some miles further
north that the region of low sand dunes was entered near a
little wooden tomb, worshipped as the supposed resting-place of
a saintly associate of Lachin-Ata. The Mazar of the latter was
not in view, nor could our guides give any clear idea where we
should find the ancient site previously described as in its vicinity.
As we plodded on amidst the gradually rising sand dunes, the
villagers I had taken along for eventual excavations became
more
communicative. They professed never to have seen that
site, but they were well aware of its legend.
These same villagers had before shown a very matter-of-fact
perception of the true cause which had led to the abandonment
of their old lands. All the greater was my surprise to find that
the legend they now proceeded to tell me of the
'
Kone-shahr
in the desert beyond was in all substantial points the same
as the one which more than twelve centuries ago Hiuen-Tsiang
had heard at Pi-mo of the sand-buried city of Ho-lo-lo-kia,
and which has been briefly quoted already in my account of
Karadong. A holy man who had reproved the wicked in-
habitants for certain offences, was treated by them with con-
tempt. He thereupon cursed the town and foretold its ap-
proaching destruction. While they still mocked at his prophecy,
sand began to rain from the skies and continued for seven
27
418 SEAECH FOE HIUEN-TSIANG'S PI-MO.
days and nights until the whole of the buildings were buried.
Only seven pious people who had shown respect for the holy
one managed to save their lives, through a curious device which
varies from Hiuen-Tsiang's story. The seven wise men are
supposed to have clung to ropes fixed to a high pole after the
fashion of a merry-go-round. Being whirled round and round
by the raging storm they rose steadily higher above the ground
while the sand accumulated, and thus escaped.
Similar stories, no doubt, are current throughout Turkestan
of ruins buried in the Taklamakan
;
but it was of particular
interest to note how the continuity of local traditions had here
transferred the legend which Hiuen-Tsiang heard at Pi-mo
of a still earlier site, to the remains of Pi-mo itself. For these
I could safely recognise in the extensive debris-covered area,
a portion of which we managed to trace in the course of the
following day. The previous evening our luckless guides had
dragged us aimlessly far out into the desert, until at last the
weariness of animals and men and the difficulty of getting
the caravan in the darkness over the rising dunes had forced
us to pitch camp. During the night one of the guides
deserted, having probably got bewildered by his own display
of deficient local sense. The other, however, a timid young
fellow whom Turdi, my desert factotum, kept under his eye
and encouraged by advice drawn from his own lifelong
"
treasure-seeking " experience, recovered his bearings, and
setting out before daybreak succeeded in finding the ruined
area far away to the south-west.
Uzun-tati

" the distant Tati," as local tradition appro-

priately designates this site

proved to consist of several

extensive patches of ground, one nearly half a mile square,
thickly covered with pottery fragments, and other small debris.
Owing to far-advanced erosion and the destruction dealt by
"
treasure-seekers," the remains of mud-built houses were
too much decayed to permit of excavation or to offer clear
indications as to their date. Yet such chronological evidence
was needed before the identification of this site with Hiuen-
Tsiang's Pi-mo, suggested as it was by every topographical
consideration, could be definitely accepted. Pi-mo was un-
LEGEND OF HO-LO-LO-KIA. 419
doubtedly, as already recognised by Sir Henry Yule, the same
place as
'
Pein,' which Marco Polo visited on his journey east
of Khotan, and must thus have remained inhabited up to the
close of the thirteenth century. The appearance of the bits
of pottery, glass, china, small objects of brass and stone, etc.,
which turned up among the plentiful debris of Uzun-tati,
entirely favoured this assumption. But it was only when
myself picking up a Chinese copper piece of the Southern Sung
dynasty (a.d. 1127-1278),
that I secured conclusive proof that
the site had been occupied up to the Middle Ages.
Our guides had previously spoken of a second
'
Kone-shahr
'
which, from some supposed tombs of saints, they called Ulugh-
Ziarat
("
the shrine of the holy ones
").
Though these remains
were known to them as adjoining Uzun-tati, and in the end
proved to be only about three miles distant in a direct line
to the south-east, it took us nearly two days and very tiring
marches and counter-marches in the sand, over an aggregate
distance of some twenty-five miles, before this second site was
discovered. Less extensive than Uzun-tati, it displayed debris
manifestly of the same period. In addition I found not far
from it the comparatively well-preserved remains of a small
fort, built in the form of an oval of about
480 by
348
feet. The
wall of stamped clay was some 11 feet thick at the base, and,
including the parapet, rose originally to a height of about
14J
feet. No remains of any kind were found in the interior
of this circumvallation, or around it, and consequently I was
unable to form any definite opinion as to its date.
These days in the desert had convincingly demonstrated the
serious difficulties which must always attend a search for scanty
ruins hidden away among deceptive sand dunes if made without
adequate guidance. The rapidly increasing heat and glare

on the 27th and 28th of March the air about midday was
88
Fahr. in the shade, though the minimum thermometer had for
the corresponding nights stUl registered
28
and
30 Fahr.

rendered tramps through the sand very trying, and made us

realise the limitations of the water supply carried in the tanks.
Hence I felt as glad as my men when the satisfactory con-
clusion of my task allowed me to turn back southward to the
27*
420 SEAECH FOR HIUEN-TSIANG'S
PI-MO.
inhabited area. Passing en route the desolate little shrine of
Lachin-Ata and then the hamlet of
'
New Ponak '
on the
fringe of the desert, I reached on the 29th of March the oasis
of Gulakhma, There for the first time I caught sight of the
young green of cultivated fields and orchards.
Gulakhma, which counts about
900 houses in its several
villages and with the adjoining Begship of Domoko is un-
doubtedly the modern representative of the Pi-mo oasis, might
have tempted me to give my caravan the short rest it had
amply earned. But time would not permit. So on the 30th
of March I sent on the main part of my camp under Ram Singh
to reach Khotan by easy stages, while I myself with the minimum
of impedimenta hurried back to Keriya to bid farewell personally
to its kindly Amban. My rapid marches were made pleasant
by noticing on all sides exuberant signs of spring, which seemed
to have come over the land with surprising rapidity. Wherever
the road passed through cultivated tracts the poplars and
willows lining it showed plentiful young leaves in delicate bluish-
green tints. At Yaqa-Langar, where I passed a night in the
garden of a half-ruined Sarai built by Niaz Hakim Beg, my
tent was
pitched under blossoming plum-trees. The mild
evening air and the picturesque neglect of the garden strangely
recalled the surroundings of many a pleasant camp in the Punjab.
I found Keriya looking bright and cheerful in its setting of
sprouting trees, and its whole population en
fete,
celebrating the
festival known to the Muhammadans of India as the
'
Bakri-
Id.' Singing and feasting went all round the
'
Topbashi's
'
garden where I camped. On the following morning, April ist.
I sent my last presents to Huang-Daloi, including a number
of personal
souvenirs, and then jiaid my farewell visit to his
Yamen.
In the course of our long confabulation I did not fail
to emphasise the excellent services of Ibrahim, our energetic
Darogha. So the Amban publicly lauded him and promised to
reward him with a comfortable berth and good emoluments. It
was already well known at Keriya that Pan-Darin on my recom-
mendation had provided Islam Beg, for similar good services
in the Khotan district, with a fat Begship at Kara-kash, and
Huang-Daloi might well feel encouraged to follow the lead
REMAINS AT UZUN-TATI. 421
of his pious colleague. I myself felt pained by the thought
of how Uttle it was in my power to return the Amban's never-
failing help with some substantial service, and how scanty the
hope was of ever seeing his kindly face again. Yet when we
exchanged our final goodbye outside my little camping-ground
he seemed to realise the lasting gratitude I should retain for
him and my sincere regret at the parting.
On the 2nd of April I started back to Khotan by forced
BOYS AND GIRLS AT KERIYA, IX HOLIDAY DRESS.
marches. The first brought me to Karakir Langar, a deserted
roadside Bazar east of the Domoko oasis, where a curious illus-
tration offered itself of the changes affecting cultivation in
this tract. About ten years pre\aously, I was told, abundant
springs had unexpectedly appeared in the sandy jungle some
miles to the south, fed, no doubt, by the Nura and other hill
streams which higher up lose themselves on the pebble
'
Sai,'
that glacis of the mountains. The water supphed by these
springs was so ample that land sufficient for
700
to 800 house-
422 SEAKCH FOE HIUEN-TSIANG'S PI-MO.
holds has since been brought under cultivation in the desert
tract to the north of Karakir Langar, with the result that the
wayfarers' custom has been completely transferred to the new
village of Achma. My second day's ride was to Chira, a large
oasis counting some
3,500
households, and receiving its water
from the river that comes from Hasa, and is fed by the glaciers
north and north-east of the great Muztagh. My night's halt
here was rendered enjoyable by the charming camping-place I
discovered in a terraced orchard, where the white blossoms
of the plum-tree
('
iiriik
')
covered the ground like fresh snow,
while the air was scented with their perfume.
But already on the following morning we had to face a strong
dust-storm blowing from the west, while we covered the forty
odd miles across the dreary plain of sand and pebbles to the
oasis of Sampula. The thick haze which enveloped us all day
made me thankful for the guidance afforded by the rows of
poles marking the road. Sampula, or Lop as it is also called
from its chief village, is a thickly populated tract still included
in the Ambanship of Keriya, though watered chiefly by canals
from the Yurung-kash or Khotan River. I was struck by the
thriving look of its villages, due largely to the flourishing carpet
industry which is centred here. Its products, though un-
fortunately debased by the use of aniline dyes, are still much
prized throughout Turkestan. There is little doubt that the
manufacture of these famous silk carpets, and some other local
industries connected with Khotan, are an inheritance from
ancient days.
The fourth and last day of my journey to Khotan was utilised
for a visit to the extensive debris area spreading on the outskirts
of the desert beyond Hanguya, the northernmost large village
of Sampula. It was a typical
*
Tati,' just as Turdi's report
had led me to expect, covering several square miles. Thanks
to his expert guidance, I had no difficulty in tracing in the
midst of it the much-decayed remains of a Stupa, known to the
people of Hanguya as the Arka-kuduk Tim. The ruin itself
showed no feature of special interest, but it was curious to
note that, owing to deep erosion of the surrounding ground,
the remains of the Stupa now stand on an isolated loess bank
FAREWELL VISIT TO KERIYA. 423
fully 20 feet high. In reality the lowest course of the brick-
work marked the original level, and the mound on which it
now appeared to be raised was but a witness or
"
Zeuge
"
to
use the geologist's termindicating the remarkable depth
to which the slow excavation of the loess soil had been carried
by the force of the \rinds. Ancient coins, seals, and other
small objects are frequently picked up on its site, and the
specimens I acquired on the spot from one of Turdi's associates
were as clear a proof of its antiquity as the extent of erosion.
The dreary
expanse of the Tati looked doubly
doleful in the
yellow dust
haze, and I felt quite relieved when in the evening.
VILLAGE CHILDREN, KERIYA.
after a long ride over much sandy waste interspersed with patches
of young cultivation, I reached the edge of the Yurung-kash
canton. There faithful Islam Beg, with the emblems of his new
dignity, Badruddin Khan, the Afghan Aksakal, and a posse of
local Begs and Yiizbashis were waiting to give me a cheerful
welcome on my retm"n to Khotan territory.
Joined thus by
old friends and an imposing escort, I rode on through shady
lanes where the scent of the fruit trees and weeping wiUows,
now in full bloom, was almost overpowering. When I reached
my re-united camp in a pleasant old garden near the Madrasah
of
Yurung-kash town,
'
Yolchi Beg ' gave vent to his joyful
feelings by the most sonorous of barks.
EXCAVATIONS PROCEEDING ALONG SOUTH-EAST WALL OF RAWAK STUPA COURT.
CHAPTER XXX.
AK-SIPIL AND THE SCULPTURES OF THE RAWAK STUPA.
On the 6th of April I halted in Yurung-kash, where fresh sup-
plies and labourers had to be secured, and many repairs to be
effected in our equipment. Increasing heat by day and recurring
dust storms warned me that the season was close at hand when
work in the desert would become impossible. Instead of taking
the rest we all by this time felt much in need of, I hastened to
set out for the ancient sites which still remained to be explored
in the desert north-east of Khotan. So after discharging
Ibrahim Akhun, our worthy Darogha, with a liberal reward in
glittering gold roubles for himself and an ample supply of speci-
ally desired medicines for his Amban, the caravan was set in
march again on the morning of April 7th.
The ruined site known to treasure-seekers as Ak-sipil
("
the
MARCH TO AK-SIPIL. 425
White Walls "), and situated among high sand dunes, at a
distance of nearly fifteen miles from the right bank of the Yurung-
kash opposite Khotan, was my next objective. On the march,
and close to the edge of the cultivated area, I examined with
interest the site known as Tam-oghil, from an adjoining small
hamlet, where ancient
"
culture-strata," yielding some leaf-gold,
besides old coins, terra-cottas, etc., are worked under exactly
the same conditions as those described at Yotkan. The extent
of the excavations is, however, far more limited, as the avail-
able water supply is scanty and the proceeds are less remunera-
tive. Here, too, the excavations, which now employ about a
dozen people for one and a half to two months in the year, are
said to have been started by the accidental discovery of gold in
a small
'
Yar
'
that had formed about twenty years ago through
the overflow of an irrigation channel. I noticed that the banks
of fertile earth overljang the ancient
"
culture-stratum
"
to a
height of 10 to i8 feet, silt deposits as I take it, showed here and
there distinct traces of stratification. Considering the short
distance, less than three miles, which separates this site from
the present right bank of the Yurung-kash, it appeared to me
possible that these sUght layers, i to ij inches thick, may,
perhaps, be due to exceptional floods from the river. The fertile
soil excavated is used by the \411agers to improve gravelly fields
in the vicinity.
As soon as we had passed the edge of the cultivated area,
fragments of ancient pottery appeared on the bare loess, cropping
up also in large patches between the low dunes over which our
march led for the next four miles. There was ample evidence
that the belt of villages and fields had extended much further
to the north in ancient times. Then the dunes grew remarkably
steep and high, up to sixty feet and more, the coarse, heavy sand
unmistakably showing its origin from the gravel deposits of the
river. Here the uniform direction of the dunes was also clearly
marked, being N.N.W. to S.S.E. After five miles of these
difficult dunes we reached open and in places much eroded
ground near Ak-sipil, where I thought I could distinguish traces
of little embankments dividing ancient fields, and of distributing
'
Ariks
'
along them.
426 SCULPTUKES OF EAWAK STUPA.
At Ak-sipil the most conspicuous remains are ruined portions
of the rampart and parapet of an ancient fort. They have been
visited before by several European travellers, and as some reliable
data concerning them have been published by M. Grenard from
M. Dutreuil de Rhins' notes, the briefest notice will suffice;
The exact survey made by me showed that the extant ruins form
a segment, about 360 feet long, of a circular wall which must
have originally enclosed an area about 1,000 feet in diameter.
Here, as at Endere, the lower portion of the circumvallation con-
sisted of a rampart of hard stamped clay, rising about 11 feet
above the original surface outside, which is still clearly distin-
guishable in places free of sand and uneroded. The rampart is
surmounted by a parapet, 8 feet thick, which, by the large size
of its sun-dried but fairly hard bricks (about 20 by
15
by
4
inches
on the average), as well as by its solid construction, suggests con-
siderable antiquity. The parapet, where in fair preservation,
showed loopholes arranged in two uniform levels, one 16 inches,
the other
5
feet above the base, but at irregular intervals. At
two points of the extant segment the parapet is strengthened
by solid brick platforms projecting about
3
feet on either side,
which were provided with stairs, and probably served as watch-
towers.
With the exception of the small segment facing due north, the
circumvallation of this ancient fort, together with any buildings
the interior might have once contained, has disappeared com-
pletely owing to erosion. The debris which covers the open
ground between the low dunes for some distance around, has
furnished to native
"
treasure-seekers
"
Chinese coins of the Han
period, and plentiful small remains, such as seals, etc. All those
acquired by me point to an early abandonment of this site.
A low mound, some i^ miles south-west, to which Turdi con-
ducted me, proved on excavation to mark the position of an
ancient temple. Though the structure itself had been completely
destroyed, no doubt through the operations of
"
treasure-
seekers," there turned up among the loose debris of plaster and
decomposed timber a considerable quantity of small relievo
fragments in stucco of remarkable hardness. These fragments,
among which pieces of draped relievo figures, as well as of deco
DISCOVEKY OF RAWAK STUPA. 427
rative plaques are largely represented, display a style of modelling
and a technical execution far superior to the stucco work of the
Dandan-UOiq and Endere temples, and recalling the best pro-
ducts of Graeco-Buddhist sculpture in Gandhara. A feature as
remarkable as the hardness of these small stucco relievos is their
cracked and fissured surface, which in places looks as if scorched.
The assumption that these stucco pieces received their present
appearance in a fire that consumed the temple naturally suggests
itself. But whether this accidental burning would also account
for their exceptional hardness is a question still to be settled
by a ceramic expert.
"
Treasure-seekers " call this place
'
Kighillik,' from a large moimd mainly composed of dry
manure
('
kighik
')
that rises quite close to the remains just
mentioned. This huge refuse-heap, which measures, as far as
exposed, some
70
by
50
feet, with a depth of at least 16 feet,
has not escaped the attention of treasure-seekers. The regular
galleries they have tunnelled through it enabled me to ascertain
with comparative ease that its contents, besides manure (appa-
rently horse-dung), were only small bits of bone, charcoal and
fuel.
On the loth of April I left Ak-sipil, and marching due north
for about fourteen miles, partly over dunes of coarse grey sand,
partly along a pebble-covered
*
Sai
'
clearly recognisable as an
ancient river-bed, arrived in the evening at the ruins called
'
Rawak '
("
High Mansion
")
by Turdi and the men of his
craft. Here an unexpected and most gratifying discovery
awaited me. Our honest old guide had spoken only of
"
an old
house
"
to be seen there half-bvu:ied in the sand, but in reahty
the first glimpse showed a large Stupa with its enclosing quad-
rangle, by far the most imposing structure I had seen among the
extant ruins of the Khotan region. Large dunes of coarse sand,
rising over
25
feet in height, covered the quadrangle and part
of the massive square base of the Stupa on the north-west and
north-east faces. But towards the south the drift-sand was
lower, and there great portions of the Stupa base, as well as
the lines of masonry marking the quadrangular enclosure of the
Stupa court, could be readily made out. Near the south corner
of the enclosing wall fragments of the heads of colossal stucco
428 SCULPTURES OF RAWAK STUPA.
statues, the spoil of casual diggings by
"
treasure-seekers,"
were lying on the surface. I realised at once that there was scope
here for extensive excavations, and accordingly lost no time in
sending back urgent orders for a reinforcement of labourers.
Fortunately the position of the ruin, within a day's march of
the oasis, enabled me to secure a large number of willing workers
from the nearest villages of the
Jiya
tract. A favourable factor
II
RAWAK STUPA, SEEN FROM SOUTH CORNER OF COURT.
of still greater importance was the relative ease with which the
question of water supply for such a number of men was solved.
For though the sand dunes surrounding us looked more formid-
able than at any ancient site previously explored, it was possible
to dig a well in a depression within two miles of the Stupa, and
there the labourers' camp was conveniently established. A look
at the map shows that the distance from the Rawak site to tiie
bank of the Yurung-kash is only about seven miles.
In fact
SUCCESSION OF SAND-STORMS. 429
to this comparative proximity of the present river-bed were due
both the forbidding height of the dunes and the sHght depth of
subsoil water.
The season of Burans had now fully set in, and the gales that
were blowing daily, though from different quarters and of vary-
ing degrees of violence, carried along with them a spray of light
sand that permeated everything. I noticed the frequency with
which the wind would shift round to almost opposite directions
on successive days, sometimes even between morning and
eveninga feature of Burans well known to all natives living
near the Taklamakan and observed also by former travellers.
To the discomfort which the constant drifting of sand caused,
and \^^hich we naturally felt in a still more irritating fashion
while engaged in excavation, was added the trying sensation of
glare and heat all through the daytime. The sun beat down
with remarkable intensity through the yellowish dust-haze, and
the reflection of its rays by every glittering particle of sand made
the heat appear far greater than it really was. The quick radia-
tion that set in as soon as the sun had gone down caused rapid
and striking variations in the temperature at different portions
of the day, and I have little doubt that the agues and fevers,
from which all my own followers began to suffer after our start
from Yurung-kash, were mainly brought on by these sudden
changes. It was impossible for me to escape exposure to these
adverse atmospheric influences
;
but luckily the chills I caught
freely could be kept in check by liberal doses of quinine until
my work at these fascinating ruins was done.
-
The excavations, which I commenced on the morning of the
nth of April in the inner south corner of the quadrangle, soon
revealed evidence that the enclosing wall had been adorned with
whole rows of colossal statues in stucco. Those on the inside
face of the wall could still be expected to be in a fair state of
preservation owing to the depth of the sand, which was in no place
less than seven feet, greatly increasing towards the west and
east corners. But I realised that great masses of sand would
have to be shifted belore these sculptures could be systematically
unearthed and examined in safety. For the heavy earth-work
implied by this task it was necessary to await the arrival of the
430 SCULPTUKES OF RAWAK STUPA.
reinforcements already summoned. But in the meantime I was
able to utilise the dozen labourers already at hand for such
clearings as the preliminary survey of the structural remains
demanded.
The result of this survey showed that the Stupa court formed
a great quadrangle
164 feet long from north-west to south-east,
and
143
feet broad. It was enclosed by a solidly built wall of
sun-dried bricks, a little over
3
feet thick, and rising to a height
of over II feet at the exposed south corner of the court, but once
probably higher. The centre of the quadrangle is occupied by
the imposing Stupa base, which rises in two stories to a height
of 20 feet above the floor. The photograph on
p. 428
shows it as
seen from the inner south corner of the court, before any clear-
ing. Owing to bold projections on each face, originally support-
ing well-proportioned flights of steps, the ground plan of the base
showed the shape of a symmetrically developed cross, each
of the four arms of which extended to fifty feet on the lowest
level.
The diameter of the Stupa dome, which was raised on a pro-
jecting circular drum and constructed like the rest of the struc-
ture of sun-dried bricks, measured a little over
32
feet. It seems
to have had an inner chamber about
7I
feet in diameter, but
this could not be exactly determined, as a large cutting had been
made into the dome from the west. The top of the Stupa had
also been broken long ago, the extant masonry reaching to a
height of
33
feet above the level of the court. The dome had pro-
bably always been exposed to the attacks of
"
treasure-seekers
"
as
well as to erosion, and the destruction thus caused made it quite
impossible to determine its original height. The broad flight of
steps which occupied the centre of each of the four faces of the
base, and led up without a break from the court to the foot of
the dome, must have been an imposing feature. The one on the
south-east side, which faces the entrance gate of the quadrangle,
could
alone be cleared. The portions of the base flanking this
flight of steps proved to be coated with a thick layer of white
stucco
which probably once covered the ^whole of the Stupa.
It was
here, sticking to the plaster under a bold moulding at the
foot of the base, that I discovered four well-preserved copper-
EXCAVATIONS IN STUPA COURT. 431
pieces of the Han period showing very Httle wear. Like sub-
sequent finds of such coins, they had manifestly been deposited
as votive offerings, and furnished me with the first indication of
the probable age of the structure.
The great archaeological interest of the ruins, however, does
not centre so much in the Stupa as in the rich series of relievo
sculptures decorating the walls of the Stupa court. These were
brought to light by the systematic excavations which I com-
menced as soon as the bands of labourers, quickly collected and
despatched by the Beg of Yurung-kash, had joined my camp
in the early morning of the 12th of April. In order to avoid
the risks of immediate damage to the friable stucco of the sculp-
tures, and to get sufficient room for photographing them, it was
necessary to open broad trenches at some Httle distance from
the walls and then to proceed towards the latter, carefully
clearing out the sand. Commencing at the inner south corner,
the work of excavation was gradually extended along the south-
west and south-east walls up to the furthermost points which
the high dunes rising over the east and west comers permitted
to be cleared. The photograph,
p. 424,
shows a portion of the
south-east wall with the trenches in course of excavation, along
both its sides, and also helps to convey an impression of the
mighty ridges of sand immediately smrounding the ruins.
As the work of clearing proceeded, I soon recognised that the
main adornment of the walls, both towards the court and outside,
consisted throughout of rows of colossal statues in stucco. All
the large relievos represented Buddhas or Bodhisattvas, but
from the varying attitudes a number of groups could be distin-
guished, arranged apparently on a more or less sjTumetrical plan.
Between the colossal images at frequent intervals were smaller
relievo representations of attendant deities and saints. In
numerous instances the walls were further decorated with
elaborate plaques of stucco forming halos above and around the
more important figures, as well as with small paintings in fresco.
The whole of the relievo work had originally been coloiu^ed, but
the layers of paint had peeled off except where well protected
in drapery folds, etc. Thus the greatest portions of the stucco
images presented themselves in their terra-cotta ground colour.
bl
CLEARING OF COLOSSAL STATUES. 433
I found from the first that the excavation of this wealth of
statuary was attended with serious difficulty. Owing probably
to the moisture rising from the neighbourhood of subsoil water,
the strong wooden framework which once supported internally
the masses of stucco and fastened them to beams let into
the
wall behind, had completely rotted away. The cavities left
by
the beams, which were e\-idently about
5
inches square, and fixed
at a uniform height of 8 feet above the ground, can clearly
be
seen in the photographs,
pp. 432, 439,
440,
while the round
holes visible in the arms of the colossal statues (see
pp.
432,
434)
indicate the position once occupied by portions of the
internal
framework.
Deprived of this support, the heavy stucco images threatened
to collapse when the protecting sand was being removed.
The
Burans greatly added to this risk. They carried away the fine
sand which had filled the interstices between the statues and the
wall behind, and thus placed the friable masses of stucco
in
danger of sliding down through their own weight to immediate
destruction. Experience soon showed me that these risks
could
be obviated only by extreme care in clearing the relievos
and
by covering up again their lowest portions as soon as they
had
been photographed. Even so damage could not altogether
be
prevented. In some instances it was necessary to secure
the
upper portions of statues still intact by means of ropes held
by
my men, even during the few minutes required to obtain
photo-
graphs. Our procedure in these critical cases is illustrated
by
the view on
p.
434.
It shows some of the minor statues ex-
cavated on the inner side of the south-west wall being thus
held,
and also helps to mark the true size of the colossal image seen to
the extreme right by comparison with the laboiu-ers.
The conditions here briefly indicated, which rendered
the
excavation work so difficult and risk}', are also a sufficient
explanation why most of the colossal statues were found without
their heads. Their upper portions, just like the top segments
of the great halo seen on the left of the photograph reproduced
on
p. 432,
had necessarily been left much longer without the pro-
tecting cover of sand, and had accordingly fallen away from the
wall that once supported them. The heads of the smaller images
2S
434 SCULPTUEES OF EAWAK STUPA.
were almost invariably found intact. I may here note that
among all the sculptural decoration of the Stupa court I failed
to trace any evidence of wilful destruction by human agency
as distinguished from such casual damage as the spasmodic
RELIEVO STATUES OF RAWAK STUPA COURT, SOUTH-WEST WALL,
AFTER EXCAVATION.
burrowing of
"
treasure-seekers
"
may have caused at some points
of the more exposed outer face of the enclosing wall. This
observation lends support to the belief, justified by other evi-
dence, that this great shrine was already long deserted and the
ruins of its court covered up by the time when Islam finally
annexed Khotan.
KISKS OF EXCAVATION. 435
It is possible that originally a wooden gallery or some similar
structure projecting from the top of the enclosing wall offered
shelter to the sculptures. But this, if it really existed, must
have been systematically removed even before the sand had com-
pletely invaded the Stupa court, for only in one place near the
inner south-east face did my excavations bring to light some
pieces of timber, about four inches thick, that might have served
for such a structure. Considering how comparatively expensive
an article building timber is to this day in the immediate vicinity
of a large Turkestan town, we could scarcely be surprised at the
early removal of this, the most useful material the deserted shrine
could offer.
The total number of individual relievos of large size, which
were unearthed along the cleared portions of the south-west and
south-east walls, amounted to ninety-one. In addition to these
the finds included many small relievos forming part of halos,
etc., or deposited as ex-votos before the main images. The
position of all statues was carefully shown in the ground-plan
and a detailed description of every piece of sculpture, with exact
measurements, duly recorded. In addition, I obtained a com-
plete series of photographs of whatever sculptural work appeared
on the excavated wall faces, the aggregate length of which
amounted in the end to more than
300 feet. It was no easy task
to collect all these records with the needful accuracy while
directing the successive stages of the excavation in atmospheric
conditions trying alike to eyes, throat, and lungs. Though
Ram Singh and Turdi rendered, each in his own way, very intelli-
gent assistance, I had myself to remain in the trenches practically
from sunrise until nightfall. I could judge from the dust-laden
.
look of the men what an appearance I presented during those
days. Needless to say that the notebook used at this site feels
gritty with sand to this day !
It is impossible to attempt here a description of the mass
of interesting materials which these excavations have yielded
for the study of the ancient sculptural art of Khotan. While
such
a task must properly be reserved for the scientific Report
on my explorations, it will be useful to offer brief notes on those
relievos
which are represented in the accompanying illustrations,-
28*
WEALTH OF STATUARY, 437
Among the sculptural remains occupying the inner south-west
wall nearest to the south corner, shown on
p. 432,
and partly
reproduced also in the gravnre of the frontispiece, the well-
modelled figure of the seated Buddha and the elaborate halo of
the larger standing image behind, filled with representations of
teaching Bodhisattvas or Arhats, deserve special attention. The
three-feet measure placed in front of the seated image, and visible
also in the other photographs, indicates the scale of the sculp-
tures. The statue of a richly-robed Bodhisattva, life-size, seen
on
p.
437,
is of interest both on account of the elaborate drapery
shown in the lower
garment and of the
carefully indicated
strings of jewels which
cover the breast and
arms. Both in style
and arrangement
these jewels bear the
closest resemblance to
those displayed by
many Grasco
-
Bud
-
dhist sculptures from
the ruined Stupas and
monasteries of the
north-west frontier of
India.
On the relievos of
the outer south-east
wall, which the photo-
graph,
p. 436,
repre-
sents just as they were
found in the course of
excavation, the care-
ful execution of rich
drapery and the ele-
gant proportions of
RELIEVO STATUE OF BODHISATTVA (r. IV.), OK
^"^
hauds ajid heads
SOUTH-WEST WALL, RAWAK sTUPA COURT.
Hiay be particularly
438 SCULPTURES OF RAWAK STUPA.
noticed. The photograph opposite shows the grouping of colossal
statues on the outer walls of the south corner. The images on
the extreme right, which still reached with their shoulders to a
height of eight feet, could not be completely cleared of sand, as
the weight of the intact upper portions made a collapse imminent
if the support of the mass of sand that enveloped the lower limbs
'
were removed. The remains in the foreground belong to a kind
of outer passage wall, decorated on both sides, which appears
to have been added at this corner only. The relievos of this apex-
shaped screen, which perhaps represents a later embellishment
of the Stupa court, showed remarkably delicate work in the
plaques of the halos, but had unfortunately suffered much
damage owing to the thinness of the wall and its exposed posi-
tion. The way in which small detached stucco representations
of Buddha, in the attitude of teaching or meditation, mostly
replicas, were found deposited at the feet of the larger images
is illustrated by the view,
p. 440,
showing torsos of colossal
statues along the inner south-east wall.
The entrance gate to the Stupa court leading through this
wall was flanked on either side by two life-sized statues, which
interested me greatly as the only figures found of a quasi-secular
character. Though the upper portions of their bodies had
broken away and were recovered only in fragments, there can
be no doubt as to their representing the
'
Dvarapalas,' or
"
Guardians of the gates," which Indian convention places at
the entrances of all assemblies whether real or mythical. It is
probable that the guardians of the Rawak Stupa court, like
those sculptured at the approaches to many a sacred Buddhist
shrine in India, were meant for Yakshas, a category of attendant
divinities. But Buddhist iconography invariably gave to these
a human appearance, and it was manifest that the figures here
unearthed exhibit the customary dress of the period and of the
country. The boots seen on the feet of the two guardians which
stood inside the gate to the proper right (see
p.
441)
were wide
at the top and showed remains of dark red colouring, with an
ornamental border on their brim. Into them were tucked
bulging trousers, hidden for the greater part by two large coats
hanging down from the waist, one above the other. The bands
I
'440
SCULPTURES OF EAWAK STUPA.
of embroidery running along the hems of the coats showed
elaborate patterns with small circulets and crochets. These,
as well as other details of ancient millinery, such as plaits,
frills, etc., can still readily be made out in the original photo-
graphs.
Want of space does not permit further details about other
TORSOS OF COLOSSAL STATUES ALONG INNEK SOUTH-EAST WALL.
remarkable pieces of statuary. But I may briefly mention the
discovery of remains of gold-leaf stuck originally in small square
patches to the left knee of the colossal image which is seen on the
extreme left of the photograph on
p. 441.
I could not have wished
for a better illustration of the quaint custom which Hiuen-
Tsiang records of a miracle-working Buddha figure of colossal
A QUAINT EX-VOTO. 441
size he saw at Pi-mo.
"
Those who have any disease, accord-
ing to the part affected, cover the corresponding place on the
statue with gold-leaf, and forthwith they are healed. People
who address prayers to it with a sincere heart mostly obtain
their wishes." From the number of gold-leaf plasters of which
the marks remain on this Rawak image, it would seem as if it
'VARAPALAbj AT CATi:: ul KAWAK STLPA
had enjoyed particular fame for healing power in affections of
the knee.
But more important and fascinating than any such details
was the very close affinity in style and most details of execu-
tion which every single find revealed with the so-called Graeco-
Buddhist sculptures of the Peshawar valley and the neighbour-
ing region. Whether that sculptural art, mainly of classical
origin, had been brought direct from the Indus or from Bactria,
there can be no further doubt, in viey,' of these discoveries, that
442 SCULPTURES OF RAWAK STUPA.
II
STUCCO HEAD OF SMALL BUDDHA OR
BODHISATTVA,
ORIGINALLY PAINTED,
FROM RAWAK STUPA COURT.
at an early date it found a true home and flourished in Khotan;
The close study of this wealth of sculpture is a task of great
historical and artistic interest. I hope that it will be possible
AFFINITY TO GR.ECO-BUDDHIST ART. 443
to facilitate it by the publication of adequate reproductions of
all my photographs.
Our data for the chronology of Grjeco-Buddhist art in India
are as yet too scanty to permit any safe conclusion as to the
date of the Rawak relievos. No epigraphical finds of any kind
were made in that part of the ruins which could be cleared, but
I was fortunate enough to secure in situ numismatic evidence
of distinct value. While cleaning the pedestals of various statues
along different portions of the enclosure as well as while exa-
mining the wall where the wooden gate had once been fixed,
we came again and again upon Chinese copper coins bearing
the
'
Wu-tchu
'
sjTnbols and belonging to issues of the Han
dynasty, just like the coins I had discovered below mouldings
at the foot of the great Stupa. These coins were invariably
found within small ca\ities or interstices of the plaster or brick-
work, into which they must have been slipped as votive offerings.
Subsequently, when a detached base only eight feet square,
probably once surmounted by a small votive Stupa, was ex-
cavated near the inner south comer of the quadrangle, many
more coins of the same tjpe came to hght between the masonry
of the base and a much-decayed wooden boarding which en-
cased it.
With this discovery the total number of such coins rose to
close on a hundred. Most of them are in good preser\-ation and
do not show any marks of long circulation. Only current coins
are likely to have been used for such humble votive gifts, and
as no finds of a later date were made, there is good reason to
believe that the latest known date of these issues marks the
lowest chronological limit for the Rawak sculptm^es. The rule
of the Later Han d\Tiasty extended over the period 25-220 a.d.,
but the issue of some of its coin-types appears to have continued
to the close of the fourth centiuy. So fax as minor antiquarian
indications, derived from the construction, the materials, etc.,
of the ruined Stupa and its adornments, permit us to judge at
present, the date of its erection may well fall near the period to
which the ruins of the ancient settlement beyond Imam
J
afar
Sadik have proved to belong.
I soon realised with regret that, owing to the extremely friable
444 SCULPTUEES OF KAWAK STUPA.
condition of the stucco and the difficulties of transport, the
removal of the larger relievos was impracticable. Those pieces
of the colossal images which were found already detached, such
as portions of arms, projecting drapery, etc., usually broke when
lifted, whatever care was used. An attempt to move the com-
plete statues or torsos from their places would have meant only
vandal destruction, unless elaborate appliances, including per-
haps specially constructed coffin-like cases made to measure, as
it were, could be provided. To improvise these I had neither
time nor the technical means, and in any case it would have
been a practical impossibility to arrange for the safe transport
of such loads over the mountains, whether to India or Europe.
All that could be done in the case of these large sculptures was
to bury them again safely in the sand after they had been photo-
graphed and described, and to trust that they would rest un-
disturbed under their protecting coveruntil that time, still
distant it seems, when Khotan shall have its own local museum.
But of the smaller relievos and sculptural pieces already detached,
I succeeded in bringing away a considerable number. I felt
greatly relieved when I found on my arrival at Kashgar, and
later also in London, that the great trouble and labour which
the safe packing of these extremely fragile objects had cost me
was rewarded by their having accomplished the long journey

some six thousand miles by camels, ponies, railway and steamer

without any serious damage. The two heads of saints in alto-
relievo still retaining part of their colouring, which are shown on
pp. 442, 445,
illustrate types frequently recurring in this collec-
tion. Full-size reproductions of other sculptures have been
given among the plates of my
"
Preliminary Report."
By April i8th those portions of the Stupa court which were
not actually buried under sand dunes had been explored. The
proper excavation of the other parts could not have been accom-
plished without months of labour and proportionately heavy
expenditure. A careful examination of the surrounding area
revealed no other structural remains
;
broken pottery found
here and there on some narrow patches of ground between the
swelling sand dunes was the only trace left of what probably
were modest dwelling-places around the great shrine. The
KEMOVAL OF EELIEVOS. 445
STUCCO HEAD OF SMALL BUDDHA OR BODHISATTVA, RETAINING ORIGINAL
COLOURING, FROM RAWAK STUPA COURT.
sand-storms, which visited us daily and the increasing heat and
glare, had made the work very trying to the men as well as my-
self. It was manifestly time to withdraw from the desert.
446 SCULPTURES OF RAWAK STUPA.
Before, however, leaving the ruins I took care to protect the
sculptures which could not be moved, by having the trenches
that had exposed them filled up again. It was a melancholy
duty to perform, strangely reminding me of a true burial, and
it almost cost me an effort to watch the images I had brought
to light vanishing again, one after the other, under the pall of 1
sand which had hidden them for so many centuries.
Jumbe-kum, some four miles beyond Rawak to the north-east,
was the only remaining desert site around Khotan from which
occasional finds had been reported to me. I took occasion to
visit it from Rawak and convinced myself that this debris-
strewn
'
Tati ' contained no remains capable of excavation.
Thus, when on the 19th of April I started back to Khotan. I had
the satisfaction of knowing that the programme of my explora-
tions in the desert was completed.
447
CHAPTER XXXI.
ISLAM AKHUN AND HIS FORGERIES.
The eight days' halt that followed my return to Khotan passed
with surprising rapidity. A severe cold, brought on by the
exposure of the last weeks in the desert, developed into an
attack of what looked like bronchitis. This obliged me to
remain within doors for most of the time, and partly in bed.
But the arrangement of my collections, their partial repacking,
and the endless little agenda which accumulate after a long
season of camp work, kept me so busy that this involuntary
confinement was scarcely realised by myself. I could not
have wished for a more pleasant shelter than that afforded
by Nar-Bagh, the old country residence of Niaz Hakim Beg.
The many-windowed lofty pavilion in the centre of the garden
where I had taken up my quarters, as five years before me
Dr. Hedin had done, secured quiet as well as fresh air. The
trees along the four little avenues which radiate from this
pavilion were still partly in bloom when I arrived, and even
when the last blossoms had withered there was the fresh green
of the leaves to please the eyes which had so long beheld only
the yellow and grey of the sand dunes. Judged by the old
Moghul gardens about Lahore, my cherished haimts in years
gone by, Nar-Bagh would be thought a very plain villeggiatura
of the Eastern type. But here in Chinese Turkestan, where
even the cultivation of a field involves a serious struggle
against sterile nature, real gardens are so few and far between
that Niaz Hakim Beg's creation deserves grateful acknow-
ledgment.
I was glad that, on the morning after my arrival, I still felt
448 ISLAM AKHUN AND HIS FORGERIES.
well enough to call on Pan-Darin,
who received me at his
Yamen like an old friend, and, as I imagined, somewhat like
a fellow-scholar. Much I had to tell him of my excavations
and the finds which rewarded them. When next day the old
Amban came to return the visit, I had ready a little repre-
sentative exhibition of my antiquities to satisfy his curiosity.
Pan-Darin is undoubtedly a man of learning and versed in
Chinese history. All the same I was surprised by the historical
sense displayed in the questions which he put to me regarding
the relative age, the import and character of the multifarious
ancient documents I had discovered. When I attempted to
explain by a reference to the plates in Professor Biihler's
"
Indian Palaeography " how a study of the writing in the
various manuscripts would enable us to fix their dates with
approximate accuracy, Pan-Darin at once showed his apprecia-
tion of this evidence by writing down the modifications through
which Chinese characters have passed in succeeding periods.
I felt almost in company of a colleague, and forgot for a moment
the irksome circumlocution and confusion which conversation
through a not over-intelligent interpreter implies.
Only in one respect did the interest of Pan-Darin in my
finds at first embarrass me. He dwelt on the fact of all these
old records being carried away to the Far West. What could
he show to the Fu-tai or Governor-General at Urumchi, who
had been so inquisitive about the object of my excavations,
and who undoubtedly would wish to hear of the results ? I
knew how sympathetically Pan-Darin had represented my
case, and thanked him heartily for the support he had given
to the cause of science. I assured him against the future
curiosity of the Fu-tai by promising to send from Kashgar
photographs of the various types of ancient documents.
"
But they should be in duplicate," was the cautious demand
of my learned friend. For he seemed eager to retain for him-
self some samples of the strange records which the desert had
yielded up after so many centuries. I feel confident that, of
the copies of my
"
Preliminary Report
"
subsequently trans-
mitted by the Indian Government for presentation to Chinese
officials, his at least was duly appreciated.
PURCHASES OF "OLD BOOKS." 449
My last stay at Khotan had to be utilised also for a curious
semi-antiquarian, semi-judicial inquiry. Its success has been
greeted with no small satisfaction by a number of fellow-
scholars, besides greatly amusing me at the time. It enabled
me to clear up the last doubts as to the strange manuscripts
and
"
block-prints " in
"
unknown characters " which, as
already mentioned, had during recent years been purchased
from Khotan in remarkable numbers, and which had found
their way not only to Calcutta, but also to great public collec-
tions in London, Paris, and St. Petersburg. The grave sus-
picions which my previous inquiries had led me to entertain
about the genuineness of these
"
finds " was strengthened
almost to certainty by the explorations of the winter. Ample
as were the manuscript materials which the latter had >nelded,
and in spite of the great variety of languages and scripts repre-
sented among them (Kharoshthi, Indian Brahmi, Central-Asian
Brahmi, Tibetan, Chinese), I had failed to trace the smallest
scrap of writing in
"
unknown characters." The actual con-
ditions of the sites e.xplored also differed entirely from the
conditions under which those queer
"
old books
"
were alleged
to have been discovered.
There was good reason to believe that Islam Akhun, a native
of Khotan and reputed
"
treasure-seeker," to whom it was
possible to trace most of these manuscripts that had been
purchased on behalf of the Indian Government during the
years
1895-98,
was directly concerned in the forgeries. He
kept away from Khotan during my first visits. He had been
punished some time before on account of other impositions
which Captain Deasy and Mr. Macartney had brought to the
notice of the Khotan authorities, and he evidently did not
think it safe to attempt further deception in my case. I had no
reason to regret the wide berth which Islam Akhun had given
me while I was engaged in my archaeological work about Khotan
and at the ancient sites of the desert. But now when the anti-
quarian evidence as to the true character of those remarkable
literary relics in
"
unknown characters
"
was practically com-
plete, and my departure near at hand, I was anxious for a
personal examination of that enterprising indi\-idual whose
29
450 ISLAM AKHUN AND HIS FOEGEEIES.
>v
productions had engaged so much learned attention in
Europe.
Pan-Darin, to whom I confidentially communicated my wish
to get hold of Islam Akhun, readily granted his assistance.
As an attempt on the part of Islam
Akhun to abscond was by no means
improbable, and as time was getting
short, I took care to impress the
learned Mandarin with the necessity of
prompt and discreet action. Nor did
he disappoint me in these respects ; for
on the morning of April 25th Islam
Akhun was duly produced from Chira,
where he had been practising as a.
'
Hakim
'
during the last winter. He
scarcely anticipated being
"
wanted " now, as when
passing through Chira some
three weeks before I had pur-
posely refrained from making
any inquiries about him. The
Beg who escorted him brought
also a motley collection of
papers which had been seized
partly in Islam Akhun's posses-
sion and partly in his Khotan
house, and which on examination proved rather curious. They
were sheets of artificially discoloured paper, covered with
impressions of the same elaborate formulas in
"
unknown
characters " that appeared in the last batch of
"
ancient
block-prints
"
which had been sold in Kashgar. A manuscript
leaf, also in
"
unknown characters," had evidently remained
over from the earlier manufacture when the forger was still
content to work by mere writing.
Two large sheets of a Swedish newspaper, the Svenska Morgon-
hladet of
July, 1897
I know how these got to Khotan, but
"
that is another story
"
were relics of the imposture when
Islam Akhun tried to make a living by representing himself
ISLAM AKHUN.
PREVIOUS IMPOSTURES. 451
among the ignorant
'
Taghliks
'
south of Khotan and Keriya
as an agent sent by Mr. Macartney to search out owners of slaves
originally carried away by Hunza raiders from Indian territory.
One of these newspaper sheets, nicely mounted on cloth, showed
the portrait of a Swedish missionary in China, for which Islam
Akhun pretended to have sat himself. The few Chinese
characters printed below, giving the Chinese name of the
reverend gentleman, were cunningly represented by him as
containing his own name. Armed with these truly imposing
documents, which he passed off cis his official credentials, the
clever rogue had managed to levy blackmail from innocent
hillmen who feared to be accused of the retention of non-
existent slaves. But after a short period of success he had been
found out, and subsequently, on Mr. Macartney's representa-
tion, had received due punishment at the Khotan Yamen.
WTien about
1898
the sale of
"
old books
"
fell off, owing to the
growing suspicion attaching to them among the European
residents of Kashgar, Islam Akhun had taken to the calling of
a
'
Hakim.' or medicine man. The leaves of a French novel
(left behind perhaps by MM. Dutreuil de Rhins and Grenard)
and the fragments of some Persian texts, which had been found
on his person and were also duly produced, were said to have
figured as part of his latest equipment. But whether he used
the leaves of the French novel merely to read out imaginary
charms from, or administered pieces of them for internal con-
sumption, was, I regret, not ascertained at the time.
The examination of this versatile individual proved a pro-
tracted affair, and through two long days I felt as if breathing
the atmosphere of an Indian judicial court. \\Tien first
arraigned in my improvised
"
Cutchery," Islam Akhun readily
and with contrite mien confessed his guilt in the above
"
per-
sonation case," and also to ha\-ing in
1898
obtained money from
Badruddin, the Afghan Aksakal, by a forged note purporting
to be in Captain Deasy's handwriting. But in the matter of
the
"
old books
"
he for a long time protested complete inno-
cence. He pretended to have acted merely as the Kashgar
sale agent for certain persons at Khotan, since dead or absconded,
who, rightly or wrongly, told him that they had picked them up
29*
452 ISLAM AKHUN AND HIS FOEGERIES.
in the desert. When he found how much such
"
old books
"
were appreciated by Europeans, he asked those persons to find
more. This they did, whereupon he took their finds to Kashgar,
etc. Now, he lamented, he was left alone to bear the onus of
the fraudif such it was. Muhammad Tari, one of those who
gave the
"
books," had previously run away to Yarkand
;
Muhammad Siddiq, the Mullah, had absconded towards Aksu
;
and a third of the band had escaped from all trouble by dying.
It was a cleverly devised line of defence, and Islam Akhun
clung to it with great consistency and with the wariness of a
man who has had unpleasant experience of the ways of the
law. I had thought it right to tell him from the first that I
was not going to proceed against him at the Amban's Yamen
in the matter of these happily ended forgeries
;
for I was aware
that such a step, in accordance with Chinese procedure, was
likely to lead to the application of some effective means of
persuasion, i.e., torture. This, of course, I would not coun-
tenance
;
nor could a confession as its eventual result be to me
of any value. Whether it was from Islam Akhun's reliance
on these scruples of mine, or from his knowledge that direct
evidence could not easily be produced within the time avail-
able, two long cross-examinations, in the interval of which I
had Islam Akhun's wants hospitably looked to by my own
men, failed to bring a solution. However, in the course of his
long protestations of complete innocence, Islam Akhun intro-
duced a denial which seemed to offer some chance of catching
my wary defendant. He emphatically denied having seen
any of the alleged find-places himself, in fact having ever
personally visited any ancient site in the desert.
I had purposely refrained at the time from showing any
special interest in this far-reaching disclaimer. Consequently
I had no difficulty in inducing him to repeat it with still more
emphasis and in the presence of numerous witnesses when he
was brought up "on remand " for a third time. Whether
encouraged by his aj^parent success so far or by the forbearing
treatment I had accorded to him, it was evident that the sly,
restless-looking fellow was for the time being off his guard.
So I promptly confronted him, from the detailed account
CROSS-EXAMIXATIOX OF FORCxER. 453
printed in Dr. Hoemle's Report, with an exact reproduction
of the elaborate stories which he had told, in the course of
depositions made on different occasions before Mr. Macartney,
about his alleged journeys and discoveries in the Taklamakan
during the years
1895-98.
The effect was most striking. Islam Akhun was wholly un-
prepared for the fact that his lies told years before, with so
much seeming accuracy of topographical and other details,
had received the honour of permanent record in a scientific
report to Government. Hearing them now read out by me in
re- translation, he was thoroughly startled and confused. He
appeared also greatly impressed by the fact that, with the help
of the exact information recorded by Mr. Macartney and repro-
duced by Dr. Hoernle, I could enlighten him as to what
"
old
books
"
he had sold at Kashgar on particular occasions, what
remarkable statements he had made about the mcinner of their
discovery
by himself, etc. He was intelligent enough to realise
that he stood self-convicted, and that there was nothing to be
gained by further protestations of innocence. He now admitted
that he had seen manuscripts being written by his above-named
employers {recte accomplices) at a deserted Mazar near Sampula.
Little by little his admissions became more detailed, and ulti-
mately, when assured that no further punishment awaited him,
he made a clean breast of it,
Islam Akhun'
s
subsequent
confessions proved perfectly correct
on many
important particulars when checked from the records
kept at Kashgar, as well as from the evidence of a number of
independent
witnesses.
He showed himself to be possessed of
an excellent
memory, and readily recognised among the nume-
rous
photogravure
plates accompanying Dr. Hoernle's Report
those
representing
specimen pages from the
"
block-printed
"
books in
"
unknown characters
"
which formed his own manu-
facture.
He had, previous to
1894,
been engaged at times in
collecting
coins, seals, and similar antiques from Khotan
villages.
About that time he learned from Afghan traders of
the value which the
'
Sahibs
'
from India attached to ancient
manuscripts.
Genuine scraps of such had indeed been un-
earthed
by Turdi and some other
"
treasure-seekers
" at
454 ISLAM AKKUN AND HIS FORGERIES.
Dandan-Uiliq. But the idea of visiting such dreary desert
sites, with the certainty of great hardships and only a hmited
chance of finds, had no attraction for a person of such wits as
Islam Akhun. So in preference he conceived the plan of
manufacturing the article he was urged to supply the Sahibs
with.
In this enterprise he had several accomplices, among whom a
certain Ibrahim Mullah was the leading man. This person
appears to have made it his special business to cultivate the
Russian demand for
"
old books," while Islam Akhun attended
chiefly to the requirements of British officers and other collectors.
Ibrahim Mullah, from whom the Russian Armenian I met on
my first arrival at Khotan had purchased his forged birch-bark
manuscript, was credited with some knowledge of Russian, a
circumstance which explains the curious resemblance previously
noticed between the characters used in some of the
"
block
-
prints " and the Greek {rede Russian) alphabet. Ibrahim
Mullah gave proof of his
"
slimness," as well as his complicity,
by promptly disappearing from Khotan on the first news of
Islam Akhun's arrest, and could not be confronted with him.
The first
"
old book
"
produced in this fashion was successfully
sold by Islam Akhun in
1895
to Munshi Ahmad Din, who was in
charge of the Assistant Resident's Office at Kashgar during the
temporary absence of Mr. Macartney. This
"
book "
was
written by hand, and an attempt had been made, as also in
some others of the earliest products of the factory, to imitate
the cursive Brahmi characters found in fragments of genuine
manuscripts which Ibrahim was said to have secured from
Dandan-Uiliq. Though the forgers never succeeded in producing
a text showing consecutively the characters of any known script,
yet their earliest fabrications were executed with an amount
of care and ingenuity which might well deceive for a time even
expert scholars in Europe. This may be seen by referring to the
fac-similes which are given in Dr. Hoernle's Second
"
Report on
Central-Asian Antiquities," from
"
codices
"
belonging to the
early output, now deposited with so many other products of
Islam Akhun's factory in the
"
forgery
"
section of the Manu-
script Department of the British Museum. The fac-simile of an
MANUFACTURE OF "OLD BOOKS." 455
"
ancient Khotan manuscript
"
which appears in the German
edition of Dr. Sven Hedin's work,
"
Through Asia," is a con-
veniently accessible illustration of the factory's produce in a
somewhat later and less careful phase of its working.
Seeing that remunerative prices could be obtained for such
articles at Kashgar and, through Badruddin's somewhat careless
mediation, also from Ladak and Kashmir, the efforts of the
forgers were stimulated. As Islam Akhun quickly perceived
that his
"
books
"
were readily paid for, though none of the
Europeans who bought them could read their characters or
distinguish them from ancient scripts, it became unnecessary
to trouble about imitating the characters of genuine fragments.
Thus, apparently, each individual factory
"
hand
"
was given
free scope for inventing his own
"
unknown characters." This
explains the striking diversity of these queer scripts, of which the
analysis of the texts contained in the
"
British collection
"
at
one time revealed at least a dozennot exactly to the assurance
of the Oriental scholars who were to help in their decipher-
ment.
The rate of production by the laborious process of hand-
writing was, however, too slow, and accordingly the factory
took to the more convenient method of producing books by
means of repeated impressions from a series of wooden blocks.
The preparation of such blocks presented no difficulty, as
printing from wooden blocks is extensively practised in Chinese
Turkestan. This printing of
"
old books
"
commenced in
1896,
and its results are partly represented by the forty-five
"
block-
prints
"
which are fully described and illustrated in Dr. Hoemle's
First
Report. These, too, showed an extraordinary variety of
scripts in their ever-recurring formulas, and were often of quite
imposing dimensions in size and bulk.
Islam Akhun, when once his defence had collapsed, was not
chary about giving technical details about the forgers' methods
of work. In fact, he seemed rather to relish the interest I showed
in them. Thus he fully described the procedure followed in
preparing the paper that was used for the production of manu-
scripts or
"
block-prints," as well as the treatment to which
they were subjected in order to give them an ancient look.
456 ISLAM AKHUN AND HIS FOEGERIES.
The fact of Khotan being the main centre of the Turkestan
paper industry was a great convenience for the forgers, as they
could readily supply themselves with any variety and size of
paper needed. The sheets of modern Khotan paper were first
dyed yellow or light brown by means of
'
Toghrugha,' a product
of the Toghrak tree, which, when dissolved in water, gives a
staining fluid.
When the dyed sheets had been written or printed upon they
were hung over fireplaces so as to receive by smoke the proper
hue of antiquity. It was, no doubt, in the course of this mani-
pulation that the sheets occasionally sustained the burns and
scorchings of which some of the
"
old books
"
transmitted to
Calcutta display evident marks. Afterwards they were bound
up into volumes. This, however, seems to have been the least
efficiently managed department of the concern, for the coarse
imitation of European volumes which is unmistakable in the
case of most of the later products, as well as the utter unsuit-
ability of the fastenings employed (usually pegs of copper or
twists of paper), would a -priori have justified grave suspicions
as to their genuineness. Finally the finished manuscripts or
books were treated to a liberal admixture between their pages
of the fine sand of the desert, in order to make them tally with
the story of their long burial. I well remember how, in the
spring of
1898,
I had to apply a clothes brush before I could
examine one of these forged
"
block-prints
"
that had reached
a collector in Kashmir.
All the previously suspected details of this elaborate and, for
a time, remarkably successful fraud were thus confirmed by its
main operator in the course of a long and cautiously conducted
examination. It was a pleasure to me to know, and to be able
to tell fellow-scholars in Europe : hahemns confiteniem retim

and that without any resort to Eastern methods of judicial

inquiry. Yet possibly I had reason to feel even keener satis-
faction at the fact that the positive results of my explorations
were sufficient to dispose once for all of these fabrications so
far as scholarly interests were concerned, even if Islam Akhun
had never made his confession. In the light of the discoveries
which had rewarded my excavations at Dandan-Uiliq and
THE FORGER'S CONFESSIONS. 457
Endere, and of the general experience gained during my work
in the desert, it had become as easy to distinguish between
Islam Akhun's forgeries and genuine old manuscripts as it
was to explode his egregious stories about the ancient sites
which were supposed to have furnished his
"
finds." Not
only in the colour and substance of the paper, but also in
arrangement, state of preservation, and a variety of other
points, all genuine manuscripts show features never to be found
in Islam Akhun's productions. But apart from this, there is
the plain fact that the forgers never managed to produce a text
exhibiting consecutively the characters of any known script,
while all ancient documents brought to light by my explora-
tions invariably show a writing that is other^vise well known
to us. There is, therefore, little fear that Islam Akhun's
forgeries will cause deception hereafter.
This consideration, as well as the fact of the forgers' work
having ceased some three years earlier, had decided me not to
press for Islam Akhun's punishment on the score of this fraud.
I knew besides that my kind-hearted friend Pan-Darin was not
without reason popularly credited with a pious proneness for
pardoning sinners. In fact. I had noticed during our interview
how relieved the old Amban looked when I told him that I did
not consider it a part of my business to demand Islam Akhun's
punishment for antiquarian forgeries, of which Chinese criminal
justice might perhaps take a view very different from ours.
There was also the manifest difficulty of bringing the other
members of the firm to book, not to mention the
"
extenuating
circumstances
"
connected with the way in which encourage-
ment had been afforded to the fraud by undiscriminating
purchasers. Nevertheless, when I remembered the great loss
of valuable time and labour which the fabrications of Islam
Akhun and his associates had caused to scholars of distinction,
it was a satisfaction to know that this clever scoundrel had
already, on one count or another, received from Chinese justice
his well-deserved punishment. For fraudiilently obtaining
from Badruddin a sum equivalent to about Rs. 12 on the strength
of a scrawl which he pretended to be Captain Deasy's order,
he had been made to wear the wooden collar for a good time
;
458 ISLAM AKHUN AND HIS FORaEKIES.
for the imposture practised as Mr. Macartney's Agent he had
suffered corporal punishment as well as a term of imprison-
ment.
I had ample opportunity in the course of these prolonged
"
interviews
"
to convince myself that Islam Akhun was a man
of exceptional intelligence for those parts, and also possessed of
a quick wit and humour, equally unusual among the ordinary
'
Khotanliks.' He was of slender build, with a face and eyes
expressing sharpness as well as sly restlessness. Something in
his looks I thought suggested Kashmiri descent, but this I was
not able to ascertain. He greatly amused me by his witty
repartees to honest old Turdi, whom with humorous impudence
he adduced as a living demonstration of the fact that
"
there
was nothing to be got out of the desert." He was greatly
impressed by seeing his own handiwork so perfectly reproduced
in the photogravure plates accompanying Dr. Hoernle's Report,
and was very anxious to learn how this feat could be accom-
plished.
I had no doubt he was fully alive to the splendid
opportunities for fresh frauds which this
'
Wilayeti ' art might
provide.
How much more proud would he have felt if he could
but have seen, as I did a few months later, the fine morocco
bindings with which a number of his block-printed Codices
had been honoured in a great European library !
I represented to Islam Akhun that, willing as I was to credit
him with a reliable memory concerning the methods and
materials employed in his factory, it would still be desirable
for me to obtain some tangible memento of them. So he at
once volunteered to furnish one or more of the blocks employed
in printing those precious
"
books." As all information had by
that time been duly recorded, I allowed him to be set free con-
ditionally from the lock-up of the Yamen, and on the following
morning he turned up in due course with one of the promised
blocks from his own house. The news of his arrest had of course
long before spread through the town, and hence it was difficult
for him to gain access to the homes of his former associates,
where more of these materials may have been retained.
I
Whether it was from a right perception that his role was now
completely played out, or because he felt that his ignominious
WIT AND HUMOUR OF FORGER. 459
collapse in the course of the inquiry had rendered him ridiculous
before his old friends, Islam Akhun looked far more cowed in the
end, though free, than when first brought up as a prisoner. I
had told him before in jest that I thought him far too clever
a man to be allowed to remain in Khotan among such ignorant
people. A curious incident showed that the remark had not
passed unappreciated. Shortly before my departure Islam
Akhun presented himself with a petition, evidently meant to
be serious, praying that I might take him along to Europe. It
was not quite clear in what capacity he expected me to utilise
his serv^ices en route. But I think there could be no doubt that
the strange request was prompted by the hope of finding in
distant
'
Wilayet ' a wider sphere for his forging abilities I
So I need not regret, perhaps, having shown myself obdurate.
f
&^K

':^
7,
m-^
-jr_-
IN A KHOTAN BAZAR.
CHAPTER XXXIIj
LAST DAYS IN KHOTAN OASIS:
r\
.*WL j:^
On the 27th of April I paid my farewell visit to the Khotan
Yamen with sincere regret. It meant good-bye to Pan-Darin,
who had proved in every way a true friend to me. He was
unmistakably a man of the old school, not over fond of Western
notions and influences. Yet from my first visit I felt that he
understood my scientific aims and was ready to further them.
I soon grew fond of his quiet, unaffected ways, which seemed
to express so plainly his personal character. As an adminis-
trator this learned old gentleman may have his shortcomings.
But all my native informants were unanimous in praising his
integrity and genuine kindness. So I hoped that the literary
DEPARTURE P'ROM KHOTAN TOWN. 461
attainments of my Mandarin friend would carry weight at
Urumchi, whither he was shortly to retire, and would secure
him some comfortable appointment, maybe the Taotai-ship of
Kashgar.
On my way back I treated myself to a last long ride through
the Khotan Bazars. It was the Saturday market of the
"
Old
Town," and its long central street was overflowing with buyers
and sellers. A glorious sunshine, pouring through the shaky
tattered awnings that connect the houses and shops flanking the
street, gave brilliancy to all the gaudy wares exhibited in the
booths from which I selected mementoes. The old skill of the
Khotan workmen still shows itself in the quaint articles of dress
which form a prominent feature of the Bazar stores. But the
universal use of aniline dyes seems, here as elsewhere in the East,
to have destroyed the old sense of colour harmony. The capital
of Khotan is indeed a small place, and in the course of my ride
I revisited almost every picturesque lane and quaint mosque
I knew from my stay in the autiunn. After the long months
in the desert I found a strange pleasure in seeing humanity again
surging around me. But more than anything else the beautiful
green of the young foliage which intruded everywhere into the
lanes and the deep blue sky helped to throw lustre on my last
impressions of Khotan.
On the following morning I said good-bye to Nar-Bagh. I
had started off my heavy baggage under Ram Singh's charge
four days earlier for Yarkand. So the final departiire was not
so troublesome an affair as starts on new journe3's usually are in
Turkestan. All the same I was kept hard at work \rith leave-
taking from local acquaintances who came to see me off, with
the distribution of medicines for cases actual and prospective
among my friends' families, andlast but not leastthe dis-
pensing of
"
tips." Chinese Turkestan is a country where
services whether large or small must be compensated by
"
tips
"
just as much as in the best conducted hotels of European centres
of civilisation. Attendants of the Yamen who had been deputed
to look after my camp
;
visitors who had helped in collecting
information or antiques : Yiizbashis who had arranged for supplies,
et hoc genus om;zt', had to receive appropriate tokens of mysatis-
462 LAST DAYS IN KHOTAN OASIS.
faction. Expensive in a way as this system is, it saves needless
circumlocution and gSne. There is no need to disguise one's
"
tips
"
in the form of presents, or to press them into hands that
for the sake of appearances pretend to refuse them. Silver or
gold, as the case may be, is accepted with the same unblushing
readiness which seems to have been the proper style at Indian
courts before European notions effected a changeon the
surface. Of course, little souvenirs are not rejected by one's
Turkestan friends. But what marks the value of services ren-
dered, and is mainly looked for, is hard cash.
My march of the first day was only a short one. I did not
wish to leave Khotan without a farewell visit to the site of the
ancient capital, Yotkan. The road I followed was the same by
which I had returned from that spot on a gloomy and cold
November day. But what a glorious change in the landscape !
Riding through the hamlets clustering in the fertile cantons of
Tosalla and Borazan, there was nothing but deliciously green
fields and orchards to rest one's eyes on. The first crop of lucerne
!
was already standing high
;
the avenues of poplars, mulberry-
trees, and willows had decked themselves with the richest foliage,
and since the unusual rain that had fallen during my stay in
Nar-Bagh scarcely any dust had had time to settle on the young J
leaves. It was a delightful ride which showed me the oasis'
under its prettiest aspects. When more open ground was
reached beyond Halalbagh, the whole range of the great moun-
tains burst into view. Quite clearly I saw the heights of Ulughat-
Dawan and Kauruk-kuz where we fixed our triangulation sta-
tions. Beyond them, to my surprise, the icy ridges which form
the watershed towards the sources of the Karakash showed
themselves in rugged splendour. The inhospitable mountains
through which I had toiled in November seemed thus to send me
a farewell greeting. Their grand panorama was the finest
setting for the last views I carried away with me of this strange
little world between the desert and the mighty Kuen-luen.
At Yotkan, where I pitched my tent once more in the pretty
orchard below the Yiizbashi's house, I was busy collecting
samples of soil from the different strata which contain the
ancient deposits and the silt that has buried them. I was, also,
LAST VISIT TO YOTKAN. 463
able to acquire an additional number of ancient coins, seals,
teira-cottas, etc., the owners of which had not come forward
on the former occasion. The most notable of these antiques
was a tiny statuette in solid gold, representing a sitting monkey
of exactly the same style and attitude as frequently found
among the terra-cotta figurines from the same site.
On the morning of the 29th of April I left Yotkan for the
canton of Kara-kash, which forms the north-western edge of
the oasis. I had not found a previous opportunity to visit it,
and had now an ad-
ditional reason to look
it up before my final
departure. Islam Beg,
my faithful
'
Darogha
'
of the days of Karanghu-
tagh and Dandan-Uiliq,
had since been ap-
pointed one of the Begs
of Kara-kash.
Rightly
or wrongly, he attri-.
buted his good fortune
to my recommendation
with the Amban.
So
he was anxious
to show
me Kara-kash,
both
as
his native place and the
present sphere of his
official functions.
Both
he and Badruddin
Khan,
the Afghan
Aksakal,
had
followed me from
Khotan and claimed the privilege
of keeping
me company up to the very border
of the oasis.
The weather kept bright and clear, and made the
day's ride
most enjoyable. In the early
morning
we passed through
Bizin,
the
market-place of the
Borazan tract, on the high-road
that
leads from Zawa to Khotan.
It was Monday, the local market-
day, and the long
rows of booths and shops
were
already
thronged with villagers.
But a sight more curious to me was
BADRUDDIN
KHAN AND AFGHAN
TRADER,
KHOTAN.
464 LAST DAYS IN KHOTAN OASIS.
the long stream of petty traders whom we passed along the
country tracks leading from Kara-kash to Bizin. The weekly
market of Kara-kash had been held on the preceding day, and
the same traders who had then exhibited their wares there were
now hurrying on to Bizin. Badruddin Khan, who usually
himself shares these migrations, explained to me the system by
which the week-days are divided between the seven main Bazars
of the oasis. The
"
Old " and
"
New " towns of Khotan,
Yurung-kash, Sampula, Imam Musa Kasim, Bizin, and Kara-
kash have each a weekly market day, and as the distances are
not great and the succession of the several local markets is con-
veniently arranged, the traders make it a point to attend all
these markets in turn. Ponies carry the bales containing the
migratory
"
shops," and, balanced on the top of the loads, their
owners and assistants. Thus that morning the greater part
of the petty trading community of Khotan passed me as it
were on review. Badruddin Khan knew them all well, goods,
ponies and men, and had much to tell of their financial fortunes
and personal characters.
I was surprised at the number of foreigners whom we met
among these hurrying traders. There were Kabulis and
Bajauris, men from Pishin in Baluchistan, and plenty of Andi-
janis. A few Kashmiris, too, I saw in the straggling procession,
but the greeting I addressed to them in familiar
'
Kashiir
kath
'
(Kashmiri) met with no response. They were the sons
of emigrants settled in Yarkand, and had forgotten their fathers'
tongue. Among the Afghans, too, it is rare that the children
know anything of Persian or Pushtu. Once more I had occasion
to reflect on the great power of assimilation exercised by the
Turki-speaking population throughout Turkestan. It quickly
absorbs races which on Indian soil would retain their well-marked
individuality and difference of speech for generations. Whatever
the causes may be, this rapid amalgamation at centres like
Yarkand and Khotan always presents itself to me as an apt
illustration of the historical process by which Turki tribes
far away to the west have j^eacefully absorbed foreign elements
more numerous and cultured than themselves.
I reached Kara-kash town in the afternoon, after crossing tin
1
VISIT TO
KARA-KASH TOWN. 465
wide bed of the river from which it is named, and found it a com-
paratively Uvely and well-built place. The garden of one of
Islam Beg's relations had been hospitably prepared for my recep-
tion, and there I was busy until a late hour with the measure-
ment of many heads for anthropological piuposes and the record
of interesting details about local administration, taxes, etc.,
for which I had in Islam Beg a first-hand authority.
April 30th was to be my last day within the territory of Khotan.
I used it for a long excursion to a
'
Tati ' site called Kara-dobe
(''
the Black Mound "), of which Islam Beg had obtained infor-
mation, away to the west on the edge of the desert. In order
to reach it we had to traverse in succession the remarkably
fertile tracts of Bahram-su. Kayesh, Makuya, and Kuya, all
stretching in long strips of highly cultivated ground with shady
orchards and lanes along their own separate canals fed by the
Kara-kash. No more pleasing picture could I retain as a
souvenir of rural Khotan. The day was hot and close, and the
vision of the mountains had already vanished in the usual haze.
So I was quite glad when, after passing for some seven miles
over a
scrub-covered sandy plain and then through low dunes,
Kara-dobe was reached. I found the ground for about a square
mile covered with ancient pottery, and in the midst of this
debris a small mound of broken masonry. The brick-work was
undoubtedly old, and might well have belonged once to the base
of a Stupa. Elsewhere broken pieces of hard white stucco with
relievo ornament possibly represent the last remains of some
long-decayed shrine. Heavy dunes of coarse sand, very trying
to our ponies, had to be crossed for some four miles before we
struck the western bank of a broad marshy Nullah in which
the
stream of Yawa expands among reed-covered lagoons. And
when by nightfall I arrived at my camp pitched near the village
of Zawa, I might feel as if, by these changes of rich village land,
sandy jungle, high dunes and marsh, Vaisravana, the divine
genius loci of Khotan in Buddhist legend, had wished to let me
once more see, as a parting favour, every type of scenery I had
beheld in the land over which he presided.
By daybreak of the ist of May I set out for my long journey
westwards. Cheered as I was by the thought of the road that
30
466 LAST DAYS IN KHOTAN OASIS.
now lay clear before me to Europe, I felt the sadness of saying
farewell, probably for ever, to a fascinating field of work and to
the last of my faithful local helpmates. At Zawa itself I had to
take leave of Turdi, my honest old guide, whose experience and
local sense never failed me in the desert. I liberally rewarded
his services with more
"
treasure," i.e,, cash, than he had ever
brought back from his wanderings in the Taklamakan. He had
also the expectation of seeing himself, through Pan-Darin's
favour, installed as
'
Mirab
'
or steward of irrigation for his
native village near Yurung-kash. It was a snug though modest
post to which our
'
Aksakal of the Taklamakan ' fondly
aspired, since he thought he was getting too old for the desert,
and in view of his proved honesty I had been able to recommend
him with a good conscience. Yet with this comforting prospect
before him, I could see how genuine the tears were that at our
parting trickled over the weather-beaten face of the old trea-
sure-seeker.
It was easier to leave behind Niaz Akhun, my Chinese inter-
preter. He had fallen into a matrimonial entanglement with
a captivating Khotan damsel of easy virtue, and had decided
to remain, against the emphatic warnings of the old Amban,
who plainly told him that, as a confirmed gambler and without
a chance of employment, he would soon be starving. He had
taken the earliest opportunity to divest himself of all further
responsibihties for his wife and children at Kashgar by divorcing
her
"
through letter post
"
as it were, the necessary document
from a Khotan Mullah costing only a few Tangas. With such
remarkable ease of divorce throughout the country, as illustrated
by this typical case, the organisation of Turkestan family life
has always appeared to me rather puzzling.
Islam Beg and Badruddin Khan, who had reason to be satis-
fied with the rewards their efficient services had earned them,
would not leave me until we reached Tarbugaz, the lonely
Langar on the desert edge where I had passed my first night on
Khotan soil. When they too had bidden me farewell and I was
riding on alone by the desert track to the
"
Pigeons' Shrine,"
my thoughts freely turned to a more cheerful themethe
results I was bringing back from Khotan. When I had passed
FAREWELL TO KHOTAX FRIENDS.
467
here nearly seven months before, there was little to give me
assurance that I should ever see the hopes fulfilled that had
drawn me to this distant land. But now my task was done and
I could rejoice in the thought that my labours had been re-
warded far beyond those long-cherished hopes. Again there
came into my mind a remembrance of the pious custom which
Hiuen-Tsiang had recorded at this very site, of the sacred rats
that once enjoyed the honour now paid to the sacred pigeons.
"
On passing the mounds they descend from their chariots and
pay their respects as they pass on, praying for success as they
worship.
, . . Most of those who practise these religious
rites obtain their wishes." It was true, the sacred birds had not
seen me worship
;
for success too I had not prayed, but onl}^
worked. Yet as success had come, I felt justified in offering to
the birds a liberal treat of maize and com as my grateful ex-voto
on leaving Khotan.
30'
^ '-w-^il
RETUKX TO KASHGAR. 469
into quagmires and the mud-built walls of many houses in town
and villages collapsed. In Yarkand city much distress prevailed
;
and even in the palatial halls of Chini-Bagh, which I again
occupied, the mud roofs were soon leaking so badly that I felt
serious concern about the safety of my antiquities. However,
the heavy downpour had deUghtfully cooled the air, and thus
the 140
odd miles to Kashgar, which I covered in less than
three days, was a thoroughly enjoyable ride.
The morning of May 12th, a brilliantly clear day and full of
the sensation of spring, saw me once more at Chini-Bagh under
Mr. Macartney's hospitable roof, which I had left almost exactly
eight months before. The warmest welcome greeted me there,
and in the company of such kind friends I found it difficult
to realise how long I had been cut off from personal touch with
Europe. I might have feared to tire my hosts by a pent-up
torrent of talk, had I not been assured by so many proofs of
the constant interest with which Mr. Macartney had from afar
followed my explorations. It was a source of keen pleasure to
me to be able to show him what ample results had attended
my work, and how much I owed to that local help which his
influence and care had mainly assured me.
The kind hospitality I enjoyed made my stay at Kashgar a
period of welcome physical rest, notwithstanding the multifarious
preparations that kept me constantly at work. The Govern-
ment of India in the Foreign Department, in accordance with
the request I made before my start from Calcutta, had obtained
tor me permission from the authorities in St. Petersburg to
travel through Russian Turkestan and to use the Trans-Caspian
Railway for my return to Europe. I had also been authorised
to take my archaeological collections for temporary deposit
to England, where alone convenient arrangements could be
made for their scholarly examination. It hence became necessary
at Kashgar to repack all my antiquarian finds with special
regard for safe transport on this long journey, while all surveying
instruments and other equipment, together with the records
of our survey work, were to be sent back to India via Hunza
in charge of the Sub-Surveyor.
While the fresh transport arrangements thus necessitated by
470 FKOM KHOTAN TO LONDON.
our different routes demanded much careful attention, I was also
kept busy with the
"
demobilisation " of my old caravan.
The camels and ponies, which had served us so well during
the journeys of the preceding eight months, could not be taken
any further, and as a not insignificant portion of the grant
allotted for the expenses of my journey was invested in the
animals, their satisfactory disposal was a matter of some
concern. After a good deal of bargaining, which, in view
of the trade customs of the Turkestan
'
Kirakash,' or carriers,
could scarcely be wondered at, I succeeded in this quasi-
commercial task far better than I had ventured to hope at one
time. The ponies sold practically without any loss, while in
the case of our eight camels I realised not less than three-fourths
of the purchase price. If I could have afforded the time to
await the proper season of caravan traffic northward into
Russian territory, I should probably have recovered for Govern-
ment the whole of the original outlay on my Turkestan trans-
port. That, after all the hard marching and exposure of our
winter campaign in the desert the whole of the transport had
been safely brought back, in a condition which allowed of its
sale with such small loss, may justly be claimed as a proof of the
care we had taken of our animals.
The arrangements for my onward journey were greatly
facilitated by the kind help of M. Petrovsky, Imperial Consul-
General of Russia at Kashgar, whose acquaintance I was fortunate
enough to make on this occasion. During a long official career in
Turkestan M. Petrovsky has devoted a great deal of scholarly zeal
to the study of the history and antiquities of the country, as I
had ample occasion to note in the course of the instructive
interviews with which I was favoured within the Russian Con-
sulate. He now did all in his power to ensure the safe transit
of my archaeological finds to England, and to secure for me
the friendly assistance of the authorities in Russian Turkestan.
For the help thus accorded I may be allowed to express here
my grateful acknowledgments.
During my stay at Kashgar I had repeated occasions to meet
again Huang-Kuang-ta, the kindly old Tao-tai, and to assure
him of my gratitude for the most effective co-operation which
AMONG OLD FRIENDS. 471
I had received from the Chinese oflftcials wherever my explor-
ations took me. The amiable old administrator did not deny
the genuine interest and goodwill with which he had followed
my work. But he politely insisted on attributing all the
sympathy and support I had enjoyed from him and his Ambans
to the benediction of my patron saint,
'
Tang-Seng.' He
even suggested as an explanation that we might both, in some
pre\-ious birth, have been together under the direct spiritual
influence of the great Buddhist monk ! The Tao-tai talked
of an early retirement to Hu-nan, and of his wish to end his
days peacefully in a famous Buddhist convent near his home.
This pious hope was not fulfilled
;
for illness and age caused
him to pass away at his post within a year of my departure.
After a fortnight of busy work the demobihsation of my
camp wcis completed, and all my antiquities safely packed
in twelve large boxes. They were duly presented at the Russian
Consulate for purposes of customs examination (a most gently
conducted one), and then received their seals with the Imperial
eagle, which I succeeded in keeping intact until I could unpack
my treasures in the British Museum. I may mention the fact
of my personally taking these boxes unopened over the various
land frontiers from China to England as an indication of how
much ci\'ilisation has done to obliterate in some respects the
great barriers between Kashgar and London.
At last the day came when I had to say farewell to my hosts,
whose unceasing kindness had made this first and practically
only rest after my desert wanderings an experience of which the
pleasure will not easily fade from my recollection. On the
morning fixed for my own departure I saw Sub-Surveyor Ra,m
Singh, the faithful companion of my travels, set out for the
return journey to India. He had rendered excellent services
in accurately surveying the w^hole of the ground covered by
my journeys, and had in addition to his proper duties been
always eager to make himself useful in connection with my
archaeological work. He had at all times cheerfully borne
the fatigues
inseparable from rapid travelling over difficult
ground and often under trying climatic conditions, and had
given me valuable help in the management of my camp. I had
472
FEOM KHOTAN TO LONDON.
indeed every reason to feel grateful to the Survey of India
Department, and in particular to its present head, Colonel St.
G. Gore, c.s.i., for having provided me with so willing and well
trained an assistant. With Ram Singh there left also Jasvant
KAM SINGH AND JASVANT SINGH, WITH
'
YOLCHI BEG,'
IN MR. MACARTNEY'S GARDEN, KASHGAR.
Singh, the wiry little Rajput who had looked after the Surveyor's
bodily comforts with exemplary care and devotion. Cheerful
and contented, however long the march or bleak our camping-
ground, Jasvant Singh could indeed serve as a model to every
one of my followers.
Knowing how great a favourite
'
Yolchi Beg
'
was with both
my Hindus, I could safely entrust the genial little fox-terrier
to their care for the journey back to India. To take him along
FAREWELL TO COMPANIONS. 473
with myself to Europe was out of the question. Exjual as my
Httle companion had proved to all hardships of mountains and
desert, it would have been cruel to subject him to weeks of a
wearisome journey by rail merely to leave him in the end to
the confinement of quarantine on reaching England. Yet I
confess I felt the separation from the devoted comrade of all
my travels, until we joyfully met again one November night on a
Punjab railway platform. He had ailed a little before my
return, but soon picked up his spirits againonly to pine away
in the end when my scientific task had forced me once more to
proceed to England. Fate favoured him in the place of his
death, for he breathed his last in Alpine Kashmir, which he
loved like his master.
On May
29, 1901,
exactly a year after leaN-ing Srinagar, I
started from Kashgar for Osh, the nearest Russian town in
Farghana. My caravan was small, six sturdy ponies carrpng
my antiquities, while two more sufficed for a tente d'abri and
my much reduced camp outfit and personal baggage. Besides
the men attending to the hired animals only Sadak Akhun ac-
companied me. Safely removed from the evil spirits of the
desert (recte the temptation to take too large doses of
'
Charas
')
he had become again a fairly sober character. The caravan
route from Kashgar to Osh, across the Alai mountains, is
reckoned at eighteen marches. Anxious to save time, I managed
to cover it in ten days, keeping in the saddle or on foot from
early morning until nightfall.
Owing to the exceptional rain of the pre\-ious weeks and the
rapid melting of the snows, the feeders of the Kizil-su, which
the route crosses repeatedly before reaching the Russian frontier
towards the Alai, were all in flood. The passage of my precious
loads of antiques across the swollen streams was hence a daily
anxiety. However, with care and some good fortune we managed
to negotiate each of these obstacles without a single box getting
drenched, and on the evening of the fifth day I arrived at
Irkeshtam, the Russian frontier post. Never have I felt so
much the significance of a political barrier. For it seemed
Europe indeed into which I stepped when, a few hundred yards
from the Chinese frontier, I entered the well-built, comfortable
474 FROM KHOTAN TO LONDON.
house, nestling below the Cossack garrison's fort, where M.
Dochenko, the hospitable officer in charge of the Russian
Customs, gave me a warm welcome.
The scenery next morning showed an equally marked and
pleasant change. The barren rock and detritus of the valleys
at the head-waters of the Kizil-su gave way to grassy alpine
slopes soon after I left Irkeshtam. The usual route over the
Terek Pass was closed by the depth and softness of the snow.
So I had to take the more circuitous route over the Alai. On
the Taun-murun Pass (close on
12,000 feet above the sea),
which crosses the water-shed between Tarim and Oxus, and on
which we had to spend a comfortless night, the deep snowdrifts
and inclement weather caused much trouble. The sky did not
clear next day when I rode down the broad rolling
'
Margs,'
as we should call them in Kashmir, of the head of the Alai
Valley, and consequently I lost the chance of sighting Mount
Kaufmann and other high peaks of the Trans-Alai range towards
the Pamir.
The Kirghiz had not yet ventured up to these splendid summer
grazings which would force even the most stolid of Kashmir
Gujars to admiration. The consequent want of shelter and
supplies forced us to attempt the same day the crossing of the
Taldik Pass in order to reach less exposed ground northwards.
We were now indeed on the good bridle road that leads from
Gulcha to the
'
Pamirski post,' the well-known Russian fort
on the Pamirs ; but it was completely obliterated higher
up by deep snow, and a blinding snowstorm came on while
we toiled up to the Pass. But for the excellent guidance of
our plucky
'
Jigit,'
a Nogai or Russified Muhammadan from
Kazan whom the obliging Customs officer of Irkeshtam had
provided as an escort, we might have fared badly. It was
late in the night before we struggled through to the deserted
Kirghiz blockhouses of Och-tobe at the northern foot of the Pass.
It was a wretched shelter, but all my boxes were safe.
After this experience, the rapid marches of the next three
days, which carried me down the valley of the Gulcha River,
were doubly delightful. The Alpine scenery, the luxuriant
growth of herbs and flowers, as well as the abundance of pine
OVER THE ALAI PASSES. 475
forest in the higher side valleys, reminded me at every turn
of familiar views in Kashmir. We met plenty of Kirghiz with
their entire household on camels and ponies, slowly moving
up for their summer
'
Yailaks
'
on the Alai. The fine carpets
IN THE BAZAR OF OSH, FARGHANA.
displayed on the camels which the women-folk rode gave to
these caravans
quite an air of splendour.
I cannot
pause to describe the many signs of prosperity and
rapid
material
development
which met the eye every\vhere as
soon as I had
entered, on the 7th of
June,
the open fertile parts
of the great
Farghana Valley. Through carefully cultivated
fields and
substantially
built villages, where there was much
476 FROM KHOTAN TO LONDON.
to indicate the beneficial results of a well-ordered European'
administration combined with great natural resources, I rode
that evening into Osh, the prettily situated headquarters of
the district. Its cantonment, founded by General Skobeleff
on the conquest of Farghana only some twenty-five years before,
looked, with its clean streets of Russian houses and its fine
park along the broad, tossing river, like a favoured spot of
Eastern Europe. Yet at the same time I was curiously reminded
by many a pleasant feature of Indian
"
stations
"
I knew well
along the foot of the Himalayas.
Colonel Zaytseff, the Chief of the District, and an officer of
distinguished attainments, received me with the greatest kind-
ness. His office, with picturesque Ming-bashis and Kirghiz
headmen in attendance outside, still suggested the
"
Cutchery
"
of an Indian Frontier District. But at the charming villa
where I enjoyed his hospitality, together with a glorious view;
of the snow-covered Alai range in the distance, everything
breathed the air of Europe. The telegraph, which enabled me i
here to get into touch with home, still further strengthened the
illusion that I had reached the confines of the West.
A short halt at Osh gave much-needed rest. I here discharged
|
Sadak Akhun, whose open-air kitchen arrangements had aroused
as much interest in the Russian household of my host, the local
postmaster, as if they had been carried on in the back garden
of a London suburb. I also disposed there of my remaining
j
Indian camp furniture. I had reason to compliment myself on
the lucky inspiration which prompted this last step. For when,]
after a four-hours' drive by the well-shaded road that traverses'
the open fertile plain towards Andijan, I reached this great town
and with it the terminus of the Trans-Caspian Railway, I found
myself in full Europe for all practical purposes. In the com-
fortable hostelry of the
"
Moskwiya Numer " my camp-bed
and camp-chair would have been as much out of place as if
set up in the inn of an English country town.
The Russian part of Andijan, stretching with broad and well
watered roads to the cast of the railway head, presented in all
respects the appearance of a thriving commercial town of Eastern
Europe. There were numbers of well-stocked shops, offices full of
IN RUSSIAN TURKESTAN. 477
Russian clerks, and, in the evening, a large gathering of European
employes listening to the military band that played in the pubUc
gardens surrounding the fine church. The large native city some
miles off bore the same air of bustUng life and prosperity.
Andijan was an important centre long before the Russian
occupation, and the great impetus given by the latter to the
material progress of Farghana had only added to the wealth of
its traders, particularly since the extension of the Trans-Caspian
Railway. While walking through the broad, well-kept Bazars,
stocked with all kinds of European manufacture, as well as the
produce of home industries in Russian and Chinese Turkestan,
how little could I think of the terrible doom awaiting Andijan
in the earthquakes of the last year ! Every Central-Asian race
seemed to be represented in the busy multitude that thronged
the Bazars. Curiously enough I was greeted here by a Kashgar
'
Haji,' who a little over a year before, while on his way to
Bombay, had met me at the Turki Sarai of the Kashmir capital.
Since performing the pilgrimage to Mecca, he had seen Egypt
and Constantinople, and had chosen for his homeward journey
the convenient railway route from the Black Sea and the Caspian.
Our meeting here seemed a striking illustration how small the
'
world
"
is growing, even in Central Asia.
On the nth of
June
I left Andijan by the Trans-Ceispian
Railway, which was now to carry me and my antiquities in
comfort and safety towards real Europe. This journey, however
hurried it had to be under the circumstances, enabled me to
obtain many interesting ghmpses of a part of Central Asia,
which by its historical associations and its ancient culture,
has had a special fascination for me ever since my Oriental
studies began. Though luckily now under a civilised power and
hence fully accessible, how much it still offers to the historian
and archaeologist to explore ! I made short halts at the
provincial capitals of MargUan and Samarkand, where I was
favoured \vith much kind attention by Generals Tchaikowsky
and Medinsky, the respective governors, and offered special
opportunities for examining the antiquities collected in the local
museums. I may add here that, though my knowledge of
P'lssian is as scanty as it could be, I met nowhere with anything
478 FROM KHOTAN TO LONDON.
but courtesy and goodwill among Russian fellow- passengers
and local officials. Tfie impressions of the delightful days I
spent at Samarkand, mainly in visits to the incomparable
monuments of architecture of Timur's period which mark
the height of Muhammadan power and art in Central Asia,
could not be surpassed even by the combined reminiscences
of Lahore, Delhi, and Agra. It was, in truth, another ex-

AT SAMARKAND : MAR
KET WITH RUINKI
MOSQUES IN B\(
K
GROUND.
hibition of Moghul grandeur, but under a sky and in a climate
that even in
June
recalled Kashmir.
A brief stay at Merv allowed me to touch ground full of great
memories of ancient Iran. It was a tantalising pleasure,
perhaps, seeing how little chance there seems for me to follow
up my early historical studies in this field, yet I feel grateful
for it. Then past the ruins of Gok-tepe, an historical site
of more recent memories, the railway carried me to Krasnowodsk:
SAMARKAxND TO THE CASPIAN. 479
From there I crossed the Caspian to Baku, and finally, after
long and tiring days in the train (via Petrovsk. Rostoff,
Podwoloczyska, Cracow, Berlin) I arrived in London on
July
2, 1901.
There I had the satisfaction of depositing the antiquities un-
earthed from the desert sands in the British Museum as a safe
temporary resting-place. Neither they nor my eight hundred
odd photographic negatives on glass had suffered by the long
journey. It was for me finis longce chartceque vicBqiie, but also the
commencement of a period of toil, the more trying because the
physical conditions under which it had to be done were so
different from those I had gone through.
Owing to the great extent of the collections I had succeeded
in bringing back, the task of arranging and cataloguing proved
a very exacting one. As the period of six weeks' deputation
in England originally sanctioned by the Government of India
for this purpose proved wholly insufficient, the Secretary of
State for India was pleased to extend it by another period of
six weeks. I had every reason to feel grateful for this con-
cession
;
but it was only at the cost of great exertions and
through the devoted help of my friend, Mr. F. H. Andrews,
that I succeeded in accomplishing the temporary arrangement
of my collection and the preparation of a
"
Preliminary Report
"
during the allotted period.
When this urgent task was concluded by the close of September,
I felt glad that my impending return to India for ordinary duty
as Inspector of Schools in the Punjab promised at least a change
and temporary respite. The busy weeks spent mainly in the
basement rooms which the authorities of the British Museum had
480 FEOM KHOTAN TO LONDON.
very kindly offered for the first accommodation of my collections,
seemed to me a time of immurement for the sake of science.
How often have I not, then and since, wished myself back in the
freedom and peace of the desert !
KIRGHIZ FAMILY ON THE MARCH.
INDEX.
NOTES.

Transcription. The spelling of Oriental names and terms

in the text and Index conforms to the system of phonetic transliteration
approved by the International Congress of Orientalists and adopted, in a
simplified form, for Indian Government publications. No use, however,
has been made of diacritical marks which would appear out of place in a
book intended for the general reader.
In the Index an endeavour has been made by a uniform use of hyphens
to distinguish the component parts of Turlii words whenever clearly
traceable. In the text the use of such hyphens has been restricted.
.\bbreviations. The following abbreviations have been employed to
mark, where desirable, the derivation of Oriental words : A. Arabic
;
Ch.
Chinese ; I. Indian
; P. Persian
;
Pr. Prakrit
; S. Sanskrit ; T. Turki.
.\bdul Karim,
390
Abdul Kasim, Mullah, 122
Abdullah, of Keriya,
318, 350
.\bdullah Khan, 312 sqq.
Abdurrahman, mendicant,
389
.\b-i-Panja R.,
53,
60
Abu-Bakr, Mirza, excavations of,
249
sq.
accounts, of travelling expenses.
165 ; records of ancient,
379
Achchik, 142
.\chma, 422
administration, ancient records
of,
379
sqq.
affidavits, ancient,
380
Afghan, territory on Pamir, 60
Afghans, at Khotan, 182,
464 ;
allegiance to "Sahibs,"
313
Afridis, known to Herodotus,
15
.\gra,
478
Ahmad Din, Munshi,
454
Ahmad Merghen, of Tawakkel,
230, 252, 253, 261, 308
Aiding-kul, marsh, 248 sq.
Ajab Khan, Raja, orderly,
48, 92,
94. 96.
99
Ajanta, caves, frescoes of. xvi.
275
Akhun Beg, of Khotan,
184, 190.
231
ak,
"
white," T.
Ak-Langar,
179
Akrobat,
149
ak-sakal,
"
white-beard
"
(head-
man), T.
Ak-sipil, site of,
424
;
remains of,
426 sq.
ak-su,
"
white water," T., flood
from melting snows,
171,
405
Ak-su R. (near Kashgar). 1 1
1
Ak-su Valley, on Pamir, 68
Ak-tash Sahib, Ziarat of,
50
Aktaz, ruined site of,
415
ak-tiken, a shrub, T.,
331
Ak-tiken, name of Karadong site,
334. 405
Ak-tiken-Bel, Pass, 106
ak'ui,
"
white hut
"
(felt hut), T.,
61,
64
Alai, mountains,
89, 473
sq.,
476
Alfold, reminiscences of,
143
Aliabad,
^^
sq.
31
482 INDEX.
Alia Beg, io8
Ali Padshah,
248
Allama,
245
Allen, Mr. P. S., help of, xxiv
alluvial fans,
19,
100 sq.,
207
alluvium, layer of,
243
Altit,
40
Amalaka, S., ornament, 280
Amb,
23
Amban, title of district officer,
in
Chinese Turkestan
Amban, of Karghalik, 162 sqq.
;
of Khotan, see Pan-Darin
; of
Keriya, see Hang-Doloi
; of
Sarikol,
66, 70 sq.,
75
Ambanships, division of,
294
anadilekha,
"
rescript," Pr.,
382
Anandasena, monk,
380
Anchar Lake, 6
ancient cities, supposed, xx
ancient site, see Niya
Andijan,
476
sq.
Andijanis,
125 ;
at Khotan,
464
Andrews, Mr. Fred. H., assistance
rendered by, xxiii sq.,
274, 289,
i7i,
376, 479
Annals, Chinese, notices about
Khotan, xvi
;
data about jade,
236
anthropological measurements,
465
antiques, found at Yotkan, 240
sqq.
;
sold at Khotan, 229
sq.,
231
antiquities, transported,
469,
473 ;
deposited in British Museum,
479
aptahas (water jugs), P.,
155
araba, "cart," T., in
Arabic language, spread of,
33
Arctic stove, Stormont-Murphy,
use of, 86,
257
Arhat (Buddhist saint), hidden in
sacred cave, 226 sq. ; Hiuen-
Tsiang regarded as an, 231
;
miracles of,
245
Arhats, mention of, 382 ; shown
within halo,
437
ariks, irrigation canals, ancient,
42s
Arish-Mazar,
414
Arka-kuduk Tim, Stupa, 422
arm-chair, ancient,
357
armour, details of ancient,
272
Arslan Boghra Sultan, legend of,
145
sq.
j!
art, Indian, transplanted to
Khotan, xv,
275 ;
see Buddhist
artist, a Chinese, 128
art-ware, old Turkestan,
152, 155
sq.
Astor, chief place of,
19 ;
Rajas
of,
19;
territory,
15;
Valley,
18,
23
Ataabad,
40
At-bashi, 252
atmosphere, clearness of,
326, 387,
402 ;
see haze, dust, buran
aul,
"
camp," T.,
76
avenues, ancient, 302 sq.
Aylmer, Colonel, r.e., 28
Baby, the British, 22
Bactria, 60
;
classical art in,
377,
441
Badakhshan, 60, 68, 390
Badakhshanis, at Yarkand,
154
Badruddin Khan, trader of
Khotan, 182, 186 sq., 190, 250,
423, 451, 463 ;
MSS. purchased
through, 290, 292,
455
;
his
trade information,
464
;
fare-
well to, 466
Bahadur Shah, Munshi, 1
1
5
Bahram-su,
465
bai, "capitalist," T., 198, 210.
320, 405 ;
Bais work jade-pits,
234. 235
Baikhan, ruined site
(?), 143
sq.
Bajauris,
464
Bajtn.T. name for Peking, 116,
i6s
Baku,
479
Baltis, people,
39
Baltistan, 20, 30
Baltit,
38 sqq.
bamboos, used for writing,
404
Bandipur, port of, 10,
12
Baramula,
3
barber, of Tawakkel,
254
Barnett, Mr. D. L.,
395
base, of statue, 267
Basik-kul, lake,
79
sqq.,
100
Batur glacier, 46 sq.
Bazars, for weekly markets, 160
beda (lucerne), T., 108
Bekhtauruk,
143
hel,
"
ridge,"
"
pass," T.
INDEX. 483
Bel-kum,
178
Bengal, Local Government of, ix
Berjash (Berdasht), Pass,
yi
besh,
"
five," T.
Besh-arik,
167
Besh-karim,
132, 136 sq.
Behs-toghrak,
319
Bhurja (birch-bark) MSS., of
Kashmir, 188
Biafo glacier,
30
bichlorate of potash,
94
birch-bark, forged MS. on, 188
;
used for writing,
347
Bizil,
191
Bizin, market of,
463
sq.
block-prints, supposed ancient,
170 ;
forgeries of, 188 sq.,
449
sqq.,
454
sq.
;
in European
libraries,
458
boards, for fastening MSS., 302
Bodhisattva, Hiuen-Tsiang re-
garded as a, 231
;
person ad-
dressed as
"
Bodhisattva in-
carnate," 382
Bodhisattvas, colossal statues of,
431. 437
;
frescoes of, 265 sqq.,
300
;
paintings of,
27
1
bonds, ancient,
295
bones, remains of, 242 sq.
boot-lasts, ancient,
387
boots, in ancient paintings,
299
sq. ;
in ancient sculpture,
438 ;
removal of, 228
Borazan, canton, 462 sq.
bostan,
"
arbour," P.,
134, 149
Bostan Langax, 312 sq.
bow, ancient,
354
Bowa-Kambar, Ziarat of,
248
Bower, Col., MS. find of, xiv
Bozai Gumbaz, 60
Brahmi, ancient Indian script,
229, 276 sqq., 281, 289 sqq.,
295. 302,
307, 393
sqq.
;
unique
tablet in,
384, 387
Braldo Valley,
30, 36
Bretherton, Capt. C. H.,
5,
11
bricks, sun-dried, construction
with,
243, 307 ;
size of, 426
bridges, built with \s'ire, 28
Bridges-Lee photo-theodolite,
57
British Museum, antiquities de-
posited in,
479
sq.
Brinjak Pass, 211 sqq.
brooms, ancient,
271
Buchvor,
7
Buddha, teaching of, xiv
; body
relics of,
245 ; colossal statue
of,
265, 431
sqq. ; figure of
seated Buddha,
437
;
figure
of teaching B.,
263 ;
miracu-
lous images of,
248, 440 ;
statue
of,
177, 269 ;
coming of Mai-
treya B., 226
Buddhas, mentioned in
tablets,
382 ; frescoes of, 264
sqq.
;
painting of, 300 ;
statues
of.
393
Buddhagosha,
380
Buddhism,
119; spread of, xiv;
brought to China, 168 ; eccle-
siastical language of,
383 ;
referred to in Kharoshthi tab-
lets,
382
Buddhist art, influenced from
Persia, xvii
; Buddhist canon,
xviii,
278 ;
translations,
291 :
Buddhist Church, of Central
Asia, 291 ; Buddhist convent,
in China,
471 ; Buddhist fres-
coes,
259
sq. ; Buddhist icono-
graphy,
438 ; Buddhist legends,
survival of, 180
;
Buddhist
monk, from China, 166 ; Budd-
hist mythology, in paintings,
271 sqq.
;
Buddhist prayers,
in Tibetan,
395
; Buddhist
scholar, representation of,
274 ;
Buddhrist shrine, replaced by
Mazar, 226
;
Buddhist temple,
295
Buddhists, of Khotan,
179
sqq.
bulak,
"
spring," T.
Bulak,
312
Bulun-kul,
81, 83,
loi sq.
Bunji,
23 sq.
Bunyad Ali, Munshi,
151,
153
sqq.
buran,
"
sand-storm," T., first
experience of,
407
sq. ;
return
of,
412, 422, 424, 429
sqq.
;
shifting direction of,
429,
433,
44^,
468 ; buries Ho-lo-lo-kia,
408
Burden, Capt. H.,
25
Burhanuddin Padshahim, shrine
(Mazar),
311, 405, 414
Buriskeski, language,
32
burning of dead, pre-Muham-
madan, 22
31*
484
INDEX.
burse (a grass),
59
Burzil Pass,
15
sqq.
;
Valley, 10
Busat, Valley, 200
Bushell, Dr. S. W., xxiv,
384
Buya, Valley,
193
Calcutta, 2 sq.
camels, fed on oil,
254 ;
grazing
of, 306 ;
in desert,
309
;
riding
of,
299;
purchase of,
114;
sale of,
470 ;
camels, ancient
painting of,
299
camel-men, 116
camp furniture,
164 ;
repairs of,
116; disposal of,
476;
see
equipment
capital, of old Khotan,
237
sqq.,
see Yotkan
caravan, organisation of, 114
sqq. ;
demobilisation of,
470
carpets, ancient,
357
;
of Khotan,
422
castle, of Hunza chief,
38
sqq.
Cathay,
319
cave, sacred to Buddhist legend,
226 sqq.
cella, of shrine, 262, 263, 265
cemeteries,low position of,
244
sq.
Centaurs, reminiscence of,
351
Central-Asia, ancient civilisation
of, xiv sqq. ;
Buddhism in,
383
;
Buddhist art in, 300
;
classical art in, xvii ;
historical
associations of,
477
;
monu-
ments of,
478 ;
Central-Asian
imports, in Hunza,
39
cereals, antique, 412
chair, ancient carved,
354
Chaitya ornament,
40
Chalbash, 238
Chalma-kazan, site of,
234
Chalt, 28
Chang-Darin, Amban, 162 sqq.
Chankul Pk., 213
Chaparsun R.,
49
chappans (long coats), T.,
313
charas, drug, effects of, 362 sq.
charcoal, ancient, 282
charuks
(mocassins), T., 213, 228,
300,
386 ;
in old relievo, 272
Charvak, 161
Chash, 214 sq.
chaugans (tea-jugs), T.,
155
293.
155
its
Chavannes, Prof. E., xxiv,
384,
396,
397
Cheng-yuen, Chinese period,
397
Cherchen,
388, sq. ;
route to,
402
sq.
Chien-chung, Chinese period,
29s
Chien-ying, priest,
295
chilapchis (water-basins), T.
Chilas, hillmen of, 21
Chillinji Pass,
49
Chillum Chauki,
17
China, ancient route to, 168
political control over Khotan,
xvi
;
its relations with Hunza,
36;
troubles in, 61,
157
Chinar Bagh, of Srinagar,
4
Chinese administration, 118
; ob-
struction feared from, xii
;
withdrawn from E. Turkestan,
396, 397
;
Chinese cantonment,
126
;
Chinese civilisation, in-
fluence of, xvi
;
Chinese coins,
see coins
;
Chinese dinner,
156
;
Chinese documents, ancient,
295
sqq. ;
on wood,
384 ;
Chinese documents in Hunza,
36 ;
Chinese frontier adminis-
tration,
69 ;
Chinese frontier
guards,
50 ;
Chinese frontier
posts, see Karaul
;
Chinese,
historical sense in, xxi sq.,
448 ;
Chinese interpreter, see
Niaz Akhun
;
Chinese legends,
on Khotan coins, 242 ; Chinese
mendicants, 232 ;
Chinese MS.
finds, 292 sqq.,
384
sqq.,
395 ;
Chinese Munshi, 118 ; Chinese
official garb,
313
sq. ;
Chinese
pilgrims, travels of, in India and
Central-Asia, xiv
;
Chinese post
office,
189 ;
Chinese priest,
126
;
Chinese records, dis-
covery of ancient, 291 sqq.
;
Chinese seal,
^77
;
Chinese
graffito,
397
sq. ;
Chinese sol-
diers, 69 sq., 76 sq., 102
;
Chinese sovereignty, in E.
Turkestan,
385
;
Chinese
temple, at Kashgar, 124
Chini-Bagh, of Kashgar, 1
13
sqq.,
131. 409
Chini-Bagh, of Yarkand,i52 sqq.,
469
INDEX. 485
"
Chin and Machin," king of, 136
Chira, oasis of,
294,
422
Chira, stream,
304
chiravastra, term for mendicants'
garb, S.,
274
Chitral, campaign in.
37
;
re-
fugees from,
70
Chodbo, ancient title, 380
Chogalma, 312
Cholak Langar, 168 sq.
Christianity, at Kashgar,
138
Christmas Day, in desert, 288,
294
Chronicles, Chinese,
233, 236 ; of
Kashmir Kings,
5
chronology, unknown system of,
290
Chudda, 176
Chu-kuan (City prefect), Ch., of
Kashgar,
124
circumvallation, ancient, 400 sq.
civilisations, mingling of, vii
classical art, in Central Asia, xvii
sq. ;
influence of, 301 ; in
Khotan,
376 ;
classical seal
impressions,
376
sq.
clothing for w-inter campaign,
257
coins, ancient Chinese,
259 ;
an-
cient Khotan coins, 240 sq.
;
Chinese copper coins,
159 ;
finds at Aksipil, 426 ; finds at
Endere,
398 ; finds at Han-
giiya,
423 ; finds at Kara-dong,
41 1 ;
finds at Rawak Stupa,
431, 443
; coins of Han dy-
nasty,
357, 384
;
coins of Tang
dynasty,
233
"
collecting," dangers of, x>d
colossal statues, excavated at
Rawak,
429
sqq.,
437
sqq.
;
fragments of,
427
colouring, of relievo sculpture.
431
colours, of stucco relievos,
263
Columbarium, resemblance t>,
138
cook-room, ancient, 281 sq.
Cordier, Prof. Henri, xiii
Corpus Christi College, Oxford,
xxiv
corvee,
151
Cotton, Mr.
J.
S., help of, xxiv
cotton-prints, ancient.
398
cover of book, design for,
263
criminals, banished to Karanghu-
tagh, 198
;
punishment of. 182
cultivation, causes rise of ground,
244;
shiftingof its area, 416 sq.
culture-strata, of Yotkan,
243
sq. ; of Tam-oghil,
425
currants, excavated,
412
currency, in E. Turkestan,
159
Curzon of Kedleston, Lord, Vi-
ceroy of India, sanctions Stein's
explorations, ix, xxiii
;
iden-
tifies Oxus source,
58,
60
Customs, antiques passed
through,
471
customs, of Gilgit tribes, 21
Dachin. copper coins called,
159
Dafdar,
64
Dashkin, 20
dak,
"
mail," I., arrival or des-
patch of,
41, 60, 66. 80, 98,
100. 165, 189, 229, 287, 306,
321. 386. 389,. 403, 412; to
Gilgit,
17
Dal Lake.
7
Dandan-Uiliq, ruined site of, 229
sq. ; ancient name of, 292
sqq. start for, 251 sq.
;
march to, 256 sqq. ;
ruined
house of,
259
sqq.
;
painted
panels of, 271 sqq. ; frescoes
of, xvi
; fixing of its position,
283 sq. ; scantiness of its struc-
tures,
303
; date of its aban-
donment, 298 ; cause of its
abandonment,
303 ;
ancient
irrigation at,
304
; departure
from,
308 sq. ; timber at, 410
;
MSS. from,
453
Dangalchi,
140
Darband
{"
gateway "),
47
Dards. people.
14 ; race of,
15 ;
games of, 18
;
rebelhons of,
25 ;
t^-pe of,
32
Darel, 22
Darin (for Ta-jen), title, Ch.,
156
darogha,
"
messenger,"
"
agent,"
P.
Darogha. see Ibrahim .\khun
Darogha, see Islam Beg
dasht,
"
waste plain,"
"
stony
desert," P.,
65, 148 sq.,
167,
172, 177 sqq., passim
486
INDEX.
dastar-khan, "collation," P.,
133
dates, in Kharoshthi documents,
371 ;
of Chinese records 298
dawan,
"
pass,"
"
large sand
dune," T.. 258 sq.,
309
sqq.,
passim
dead trees, in desert,
331
sq.,
332,
387, 410
Deasy, Capt. H. H. P., explora-
tions of,
7,
56 sq.,
92, 115,1
19,
187, 202,
233, 283,
449, 451,
457
debris, ancient,
174
sqq. ; see
Tatis
Delhi,
478
Deosai, plateai:,
17
desert, see Taklamakan
;
advance
of dunes in,
303
sq. ;
dead
trees in,
258,
260
;
transport
for,
254
Detailed Report, on Stein's ex-
plorations, xxiii
detritus,
213, 223
devaputra, title, S.,
379
Dharma, Buddhist canon, S., 281
dibira,
"
clerk," S.,
380
die, ancient,
386
dihkan,
"
cultivator," P., 108,
390
di dung, a grass,
59
disintegration of rocks,
194,
222
divorce, in E. Turkestan, 466
diwanas (beggars), P.,
172, 322
"
dobe," debe,
"
mound," T. ; of
Somiya, 246 sq.
Dobe-Bostan,
327
Dochenko, M.,
474
documents, ancient, in unknown
language, 290 sqq. ;
in Kha-
roshthi,
338
sqq.
;
in Chinese,
292 sqq.,
384
sqq.
Dogras, conquest by,
29 ;
mili-
tary service of, 28
Domoko, Begship of, 416 ; irri-
gation in,
304
;
old .site of, 416
;
stream of,
304
Dost-bulak,
144
Douglas, Prof. R. K., xxiv,
291,
295
doves, sacred, 180
dragons, representing Nagas,
125
dranga,
"
frontier watch station,"
S., 382
drangadharas, otticials, S., 382
dress, details of ancient,
438
sq.
Duharhom, 6
dust, in atmosphere, 219, 226,
402
Dutreuil de Rhins, M., MS. find
of, viii, 226 sq.,
324, 340
sqq.
;
visits Ak-sipil,
426
Duyan, 20 sq.
Dvarapalas, statues of,
438
sq.
Earthquake, in Sarikol,
69 ; in
Andijan,
477
Eastern Turkestan, historical ac-
counts of, 117; modern ad-
ministration of, see Chinese
eating-houses, at Karghalik, 164
eating-sticks, ancient,
357
Egypt, climate of Turkestan re-
sembles, XX
Ekki-bel-su, R., 84,
100
embroidery, Chinese,
155 ;
an-
cient,
440
Endere, ruined site of, 388, 391
sqq.
;
Stupa of,
392 ;
circum-
vallation of
392 ;
temple ex-
cavated at,
393
sq. ;
depar-
ture from,
402
Endere colony, abandoned,
392
Endere R., shifting course of,
391
sq. ;
forest-belt of,
402
envelopes, of inscribed tablets,
338. 345. 372
sq.
equipment,
2,
7
; see camp
Eros, represented on ancient clay-
seals, xvii, T,y7
erosion, by water, 216
erosion, by wind
175
sqq.,
301,
400 sq. ;
creates depressions,
256, 259, 288,
307, 332, 335.
347
sq., 400 sq., 422 ;
destruc-
tive to ruins, 408 sq.,
415,
418
439
sq.
Eskente,
246
Eski,
138
etiquette, Chinese, 118; in Chi-
nese architecture, 316
excavation, difficulties of,
433
sq. ;
implements for,
254
ex-votos, of coins,
443
;
of
painted panels, 271
.sqq.
; of
rags,
74,
328 ;
of shreds,
398 ;
of Tibetan MSS.,
395
sq.
;
offered to sacred pigeons,
467
INDEX. 487
Fabrics, excavation of ancient
woven,
367, 385, 398
Fa-hien, Chinese pilgrim, 241
;
visits Stupa of Sa-mo-joh,
245
faience, ancient Chinese,
385
Faizabad
Faizabad,
390
falcons, emblems of dignity,
255
Far East, art of Khotan trans-
mitted to, xvii
Farghana, start for,
473
;
fer-
tility of,
475
felt, ancient, 280
felt-hut, see ak-ui, kirgha, yiirt
felt-rugs, ancient, 380 ;
of Kho-
tan, 142
felt-socks (Paipaks), 164
/en, monetary value,
159
fence, ancient, 286
fields, abandoned to desert, 416
sq. ; rise in level of,
244
figurines, in terra-cotta, 242,
243
fig-leaf, substitute for,
273
fire-place, ancient, 281,
295,
335,
343. 399
.
modem, 416
fodder, antique, 286
folk-lore, about Taklamakan,
417;
quaint custom of ancient,
440
sq.
forest, along Khotan R., 252 ; see
jungle
forging of
"
old books," at Kho-
tan, 188,
449
sqq. ;
details of
process,
456
sq.
Forster, G., travels of, 140
Forsyth, Sir D., 120
fortification, ancient, at Endere,
400 ; at Uzun-tati,
419 ; at
.\ksipil,
426
"
Four Garrisons," Ch. name for
E. Turkestan,
397
framework, of stucco statue, 265
^
frescoes, Buddhist, at Dandan-
Uiliq, 229, 259, 272 sqq. ; in
ancient residence, Niya site,
353
; at Endere,
399
; of
Buddhas, 265 sqq.
frontier-post, ancient,
399,
411
fruit, at Kashgar, 121
; at Kizil,
148
fruit-trees, ancient,
277, 359
Fu-tai, at Urumchi,
448
G.\DHOI, 18
Gandhara, Peshawar Valley, the
ancient, xvi ;
art of,
427 ;
sculptures of,
332, 387
Gandharvi, figure of, 263 >
Ganesha, painting of.
271, 400
garden, plan of ancient,
359
Gardner, Prof. Percy, xxiv
"
Gates, Masters of the," 12
Gauri Mall, Lala, 158
German officer, on Pamir, 62
Gez defile,
84, 98,
102 sqq.
Gez Jilga, 209
Gez Karaul, 103
Ghadghang,
319
Ghalchah dialects, 290
Ghan,
64
Gharib-chakma,
309
Ghazi Sheikh,
405
sq.
Ghujak,
74
Ghujak-bai,
63 sq.
Ghujak Masjid,
167
Ghulan Pass,
73
Ghulkin glacier,
44
Ghulmit, 42 sqq.
Ghun, Hunza guide,
91, 94
sqq.
gilding, of ancient statues, 240,
307
Gilgit, Agency at, 25 sq. ; lan-
guage of, 32 ; mail service to,
17 ;
River of,
24 ;
Transport
Road to,
5,
1 1 ;
Valley of,
24
sqq.
Gircha,
47
glass, ancient,
177, 385
Gobi, i.e., Taklamakan desert, 185
Godwin-Austen, Mt.,
72
Gok-tepe,
478
gold, figurine in, 240
;
gold found
at Yotkan, 238 sqq. ;
in
Khotan rivers,
217 ; washed
from Yotkan soil,
239
sqq.,
245 ;
gold-dust, washed from
rivers,
240
;
gold-leaf, dis-
covery of,
440
sq.
Gor, 22
Gorai, rest-house,
13
Gore, Col. St. George, x,
7, 472
Gosringa Mt., Buddhist site of,
226 sqq.
Govind Kaul, Pandit, 118
Graeco-Buddhist art, reproduced .
in Khotan, xvi
;
Graeco-Bud-
'
hist sculpture,
272, 387, 427,
437. 441
488 INDEX.
graffiti,
396
sqq.
grapes, of Ujat, 228
Greek deities, on seals,
376
sq.
;
Greek writing, on Indus,
388
Grenard, M., 226,
245, 426
ground level, changed by irriga-
tion, 416 ;
rise of,
123 ; raised
through cultivation,
244, 252
Guhyal (Wakhan),
42
Guhyal, Little,
42
Gujan,
183
Gujars (cattleherds), 16
; of
Kashmir,
474
gulach (fathom), T.,
320
Gulakhma, guides from,
414
sqq. ;
irrigation in,
304 ;
oasis
of, 420 ;
shifting of cultivated
area of, 416 sq. ;
stream of,
304
Gulcha,
474
sq.
Guma, oasis, 169 sqq.
gumbaz (dome), P. 72 ;
name for
Stupa, 136
Gupta, characters called, 278,
281
Gurez, villages of,
14, 15
Gurikot, 19
Hai Bai,
92
haji (Mecca pilgrim), A.,
477
Hajib Langar, 169
hakim (medicine-man), A.,
fame as,
231
Hakim Shah,
195
Halalbagh,
249
sq., 462
halos, of Buddhist saints,
289 ;
of relievos,
431
sq.
;
presentation of saints in,
437
Han dynasty, coins of,
307,
357,
398,
426, 431, 443
sq.
Hanguya, ruins beyond, 422
Haramukh, Mount, 6, 12
Hasa, 422
Hasa, near Guma, 171 sq.
Hasa, near Moji,
177
Hasa-Tam, site of,
135
Hasha,
249
Hassan Akhun, camel-man, 116,
323,
362 sq.
hat, in ancient painting,
299
Halu ]*ir,
23
haze, in Khotan mountains,
224
Hazrat-Apak's shrine,
132
Hazrat-Begim, 146 sq.
Hazrat-Sultan,
135
70
271.
re-
Hedin, Dr. Sven, explorations of,
viii,
79, 87, 92, 101,
139, 177,
230, 256, 304, 405, 407, 447,
455
hemp, in ancient string,
372
Hendricks, Father,
123
Herodotus, knows Dards and
Afridis,
14
Hindu, followers,
283 ; money-
lenders in E. Turkestan,
121,
140 sq.
Hindukush,
14
Hindustani, spread of,
32
Hiuen-Tsiang, Chinese pilgrim,
his interest in .spiritual objects,
xiv
;
claimed as patron, xxi
;
routes followed by, 2, 68
;
visits to Buddhist countries,
119 ;
see Tang-Seng, ib.
; 125,
128,
149, 156, 164, 168 ; his
birth-place,
169,
177
; records
local cult, 180, 186; mentions
Mt. Gosringa, 226 sqq.
; de-
scribes shrines of Khotan,
245
sqq.,
247
;
relates legend of
Nagini,
273
; legend of holy
rats,
289 ;
knew legends about
Taklamakan,
303
; his Si-yu-
ki,
317 ;
mentions Niya,
322
;
refers to memorial towers,
367 ;
his etymology of name Khotan,
382
; records tradition about
colonization of Khotan,
383
;
refers to
"
Kham,"
398 ;
relates legend of Ho-lo-lo-kia,
408, 417
;
visits Pi-mo,
413
sqq.
;
his records confirmed,
440
; records pious custom.
467
historical geography, survey work
needed for, x
Hiung-nu (Huns), legend of, 180
Ho-lo-lo-kia, legend of,
303, 408,
417
Hoernle, Dr. A. F. Rudolf, forms
collection of Central-Asian anti-
quities, viii
;
assistance ren-
dered by, ix, xxiii
; 164, 281
;
deciphers non-Indian language,
289 sqi].
; 453, 458
homesteads, plans of ancient,
350
Ho-nan, province,
169
horse, painting of,
299
sq.
;
horse
millinery, ancient,
299
INDEX. 489
horse-combs, wooden,
386
Howell, Capt. E. A. R., 26
Hsien-kuan (City prefect), Ch.,
131.
135
Hsie-tai (general), Ch., 119 sq.
Huang-Daloi. Amban of Kenya,
315
sqq. ; first visit to,
316;
his interest in Tang-seng,
316
;
his help,
318, 389
; his his-
torical sense,
404, 405, 413 ;
farewell \isit to,
420 sqq.
;
assistance rendered by, xxi
Hu-kuo, ancient monastery,
297
sqq..
300
Humayun, Wazir,
34
Hu-nan, province,
157, 471
Hungary, reminiscences of,
65,
122,
143
Huns, attack Khotan, 180
hunters, of Tawakkel,
252
Hunza, chiefs of, 29, 34
sqq.
;
their castle,
27, 38 sq. ;
Hunza
language,
32 sq. ;
Hunza levies,
53
; as mountain guides, 89
sqq..
99
;
Hunza people, their
marching,
48 ; their prowess,
94
; their race,
32 ;
relations
with China,
36 ; road in. ij
sq. ; Hunza River, track along,
40 sqq. ;
Hunza \'allev, 24
sqq.
Hvdaspes, the present Jhelam
R..
3
sq.
I BR.xHiM, villager of Niya,
324,333,
334
sq.,
337, 343
Ibrahim Akhun, Darogha, from
Keriya,
311, 312, 321,
334, 350,
363, 391,404,405,414,420, 424
Ibrahim Mullah, guide,
249
Ibrahim Mullah,
"
treasure-
seeker,"
454
ice, for water-supplv,
330, 389,
391
ice-pit, ancient,
350
iconography. Buddhist,
438
Igrikyok,
76
Ilchi, city of Khotan.
294
Ilegorum, 102
Imam
Jafar Sadik, Mazar, pil-
grimage to,
311, 324
sqq. ; its
shrines,
327
sq. ; tomb of,
329 ; labourers from,
352 ;
return to,
388 sqq.
Imam Musa Kasim, shrine of,
248 ; market of.
464
Imam Shakir Padshaii,
179
Imams, worshipped at Khotan,
249
implements, ancient household,
xviii
India, influence of classical art in,
377
;
classical languages of,
XV
;
Graeco-Buddhist art in,
443
; Khotan colonised from,
383
;
ruins on N.W. frontier
of, XX
India, Government of, acquires
Central-Asian antiquities, viii
;
sanctions Stein's explorations,
ix
;
thanked by Orientalist
Congress, xiii ; enables Stein
to elaborate results, xxii
Indian Caucasus, 210
Indian influences, border line of,
401 ;
Indian shrines, visits to,
227
Indo-Iranian dialect,
290
Indo-Scythians, named in Kha-
roshthi tablets,
384 ;
Indo-
Scythian kings,
340, 380
Indus, course of, 20
; route in
Valley of, 22
industries, of ancient Khotan,
xviii,
354
sqq.
inscribed wooden tablets, first dis-
covery of,
2,}!^
sqq. ; shapes
and construction of,
2>}>^
sqq.
;
preservation of,
11% ; contents
of,
341 ; dates on,
345
;
sealed
envelopes of,
345
; withered
specimens,
350 ;
more finds,
343' 354
''
blank stationery,
357
;
from ancient rubbish
heap,
365
sqq. : with Chinese
characters,
365 ; technicalities
of wedge-shaped tablets.
371
sqq. ;
fastening of oblong
tablets,
Tijl
sqq. ;
classical
seals on,
376
sq. ;
decipher-
ment of contents,
378
sqq.
;
chronological evidence of
Chinese tablets,
384
inscriptions, in Brahmi cha-
racters,
267 ;
in Chinese and
Tibetan,
395
sqq.
Iran, ancient,
478 ;
linguistic
borders of,
42
490 INDEX.
Iranian race,
155 ;
physical type
of,
43
Irkeslitam,
473
sq.
irrigation in Hunza,
32, 34, 44 ;
on Tagharma,
73
; 178, 465 ;
from Niya R.,
325
sq. ; from
springs,
417 ;
from Yar-tungaz
R.,
390
sq.
;
abandonment of,
304
sq.
;
causes silt deposits,
244 sq. ;
difficulties about, 416
sq. ;
implement for,
43
;
ex-
tension of,
151, 415 ;
possible
extension of,
252, 326 ;
men-
tioned in Kharoshthi tablets,
382 ;
irrigation canals, ancient,
302
Irshad Pass,
49
Islam, weakness in,
229
Islam Akhun,
"
treasure-seeker,"
170 sq.,
175, 189 ;
sells old
books,
449
;
arrested,
450 ;
papers seized with him,
450
sq.
;
his previous impostures,
451 ;
his associates,
452 ; his cross-
examination, ib. ;
makes con-
fession,
453
sq. ; describes for-
geries,
455
sqq. ;
his previous
punishments,
457
sq. ;
his sly-
ness and humour,
458
sq.
;
wishes to visit Europe,
459
Islam Beg, from Khotan, Da-
rogha,
198, 209,
220 sqq., 238,
249, 294,
308
;
promoted Beg
of Kara-kash, 420
; 423, 463,
465 ;
farewell to, 466
Iskuram Peaks, 213 sq., 220
Issik-bulak ("hot spring") T.,
202, 212 .sq.
ivory, ancient carving in,
241,
354
; site called
"
the houses
with ivory," 229
Jade, of Khotan, called
"
yii," 236
;
digging of,
234
sqq. ; found in
Yurung-kash R. bed, 191,
233
sqq.
;
profits from,
235
;
valued
in China,
233 ;
jade pits,
234
Jamada, 190, 233
sq.
Janguruk, gorge, 102
jar, ancient,
323
Jasvant Singh, Mian Rajput,
7,
48 ;
rejoins in desert, 282 sqq.
;
his ailment,
284 ;
his excellent
services,
472
Jesuit surveyors, in China, 108
Jhelam R.,
3
sq.
jigda, a fruit-tree, T.,
410,
415
jigit,
"
messenger," T.,
474
jilga,
"
valley," T.
J
ins, evil spirits of desert,
253,
364
Jitroghavarshman, king,
380
Jiya,'428
Johnson, Mr., map of, 187, 198
Jumbe-kum,
446
jungle along Keriya R.,
311 sqq.
;
along Niya R.,
330
sq.
Kabulis,
464
Kafirs, i.e., Buddhists, 180
Kakshal,
"
Tati
"
of,
174
sqq.
Kai-yuen, Chinese period,
259,
294, 397
sq.
Kala, ancient title,
380
Kalik Pass, see Kilik
Kampar-Kishlak Gl.,
97
Kanjur, Tibetan canon,
396
Kanjut, name for Hunza and
Nagir,
29
Kanjutis, raids of,
29, 36
Kaptar-khana, ruin of,
138
Kaptar-Mazar,
179
; legend of.
289
kara,
"
black," T.
Kara-dobe,
465
Kara-dong, ancient site of,
334 ;
start for,
405 ;
arrival at,
407
sq. ;
remains of fortified post,
408 sqq.
;
its purpose, 410
;
coin finds at, 411
;
gateway,
411 ;
departure from,
412
Kara-gaz gorge,
209
Kara-kash R., 182
;
irrigation
from, 238 ;
upper course of,
208,
219 ;
mountain ranges
above, 220, 462 ;
Valley of,
210, 314, 215, 220, 222 sqq.
Kara-kash, town,
294, 464
sq.
;
canton of,
463
sqq. ;
market
of,
464
Kara-kash Beg, 62 sqq.
Kara-kir, ridge, 80 sq., 84
sq.,
100
Kara-kir, stream,
414
Kara-kir, Stupa mound, 178
Kara-kir Langar,
421
Kara-korum, spur called,
74
sq.
Kara-korum Passes, 162,
173,
177.
187
INDEX. 491
Kara-kul, Little, lake,
76, 78 sqq.,
97,
100
Kara-kul Mazar, 171 sq.
Karanghu-tagh, Valley,
194, 197
sqq., 208 sq.
;
glaciers at head
of, 209, 213
Kara-oche-oltiirgan.
392
Kara-shahr, one of
"
Four Garri-
sons,"
397
kara-su, spring flood called,
405
Kara-su, in Sarikol,
75
sq., 81
Kara-su, stream,
171,
173
Kara-tagh-aghzi, or Kara-taghiz,
172
Kara-tash Pass,
84, 98
sq.
karaul, T., watch-station called,
50,
75,
76,
102, 103, 108 sq.
karaulchis,
74,
103 sq.
Karghalik, town, 161 sqq. ; Am-
ban of, see Chang-Darin
;
Bazars of, 162, 164 ; E. border
district,
178
karewas, plateaus called,
64
Karim Beg,
70
Karm Shah Beg,
78,
81 sqq.,
99,
lOI
Kang-sarigh.
321
Kashgar, approach to. 108, iii
;
stay at, 113 sqq. ;
threatening
situation at, 120 ; craftsmen of,
116; Russian Consulate at,
122 ;
"
New City
"
of, 124
sqq.
;
first start from,
139 ;
return to,
469
; departure
from,
473
;
ancient Kashgar,
called Su-le,
384
; old site of,
123 ;
one of
"
Four Garri-
sons,"
397
Kashe,
238 ;
Yar of,
248
Kcishmir, stay in,
3
sqq., 12
;
antiquarian work in,
5
; anti-
quities of,
5 ;
Brahmans of,
371 ; coolies from, 23
;
Chronicle of Kings of,
5
; con-
nection with Khotan,
273 ;
lotuses of,
273 ; Margs of,
474 ;
reminiscences of,
473
sqq.,
475,
478 ;
topography of,
5
;
troops
of,
19, 25, 28,
30
Kashmiri, language.
49, 154, 464
Kashmiris, characteristics of,
4,
15: at Khotan,
464 ; at Yar-
kand,
153, 155
Kash R.,
197
sqq., 200, 207 sqq.
Kash Valley, glaciers of. 220
Kasim Akhun, hunter of Tawak-
kel,
230, 252, 256, 283, 309
sq.,
310
Kaufmann, Mt.,
95, 474
Kauriik-Bel,
104
Kauruk-Kurghan, 104 sq.,
107,
108
Kauruk-kuz,
223 sqq., 462
Kayesh,
465
kazan,
"
cauldron," T., 81
Kazan,
474
Kepek Pass. 107
Keriya, town,
294, 3 14
sqq.
; start
from,
319
; second stay at,
404
;
farewell visit to,
420
sqq.
;
.\mban of, see Huang-
Daloi
; Begs of,
313
sq.
;
Yamen of,
315
Keriya R., 256, 261, 282
;
alleged
old course of,
304 ;
marches
along, 311 sqq.,
405
sqq., ;
once
reached Tarim R., 41 1 ; return
along,
413 ;
marshes of,
405
K.
5,
see Kuen-luen Pk. No.
5
ketman (hoe), T.,
254
Khattri moneylenders, 140 sqq.
Khaibar, hamlet,
46 sq.
Khalche, 238, 242,
245
Khalkhal-i Chin u Machin, legend
of,
249
kham, cotton fabric called, 328
;
ancient fabric like,
354
Khan-arik,
139
sqq.
Kharoshthi, ancient script, docu-
ments in, xix sq., 227 sq. ; coin
legends in, 242 ; first docu-
ments found in,
323
sq.,
334 ;
use on tablets,
338 ;
on Ku-
shana inscriptions,
340 ;
brought to Khotan from India,
341 ;
leather documents in,
365, 370
sqq. ;
decipherment
of Kh. documents,
378
sqq.
:
palaeographic evidence of,
383
sq.
Khan-ui, ancient site,
130, 134
sqq.
khat,
"
writing," T.,
277 sqq.
Khayindi,
74
Khitai,
"
Chinese,"
T-, 313
Khotan, oasis and town, first
arrival at.
179
sqq.
;
second
stay at, 229 sqq. ; third stay
492
INDEX.
at, 250;
return to, 421 ;
de-
parture from, 462 ;
Amban
of, see Pan-Darin ;
Bazars of,
461 ;
carpets of, 422 ;
crafts
of, 461 ;
cultivators' dwellings
in, 302 ;
diseases at, 231 ;
felts
of, 320;
industries of,
155,
380
;
jade of,
233
sqq. ;
local
worship of, 228,
247 ;
longitude
of,
224 ;
"
Old and New
Towns" of,
461, 464;
petty
trade in,
463
sq. ;
rivers of,
182
;
rural,
465
;
Tanga of,
159;
W. boundary of,
179
History of Khotan : Abu-
Bakr's rule at,
249
;
ancient art
of, xvi
;
ancient capital of,
187, 237
sqq., 248 ;
ancient
designation of,
294
;
ancient
industries of, xviii
;
ancient
language at, xix
;
ancient
names of, 382 ; . archaeological
interest of, xv ;
Buddhist cult
in,
179
sq. ;
Buddhist genius
loci,
465
;
Buddhist pilgrim-
age place of, 226 ;
Buddhist
shrines of,
245
;
Buddhist
temples of, 241 ;
Chinese call
Khotan Yii-tien, 236 ;
Chinese
influence at, xvi
;
classical art
j
in,
376 ;
classical, seals used
I at, xvii
;
coins of Khotan with
I Kharoshthi legends,
324 ;
"
Four Garrisons," Khotan
one of,
397
;
Gra;co-Buddhist
art in,
441 ;
historical connec-
tion with Kashmir,
273 ;
his-
torical importance of, xv
;
p
Indian art in, 300
;
prince
of, in China, xvii ;
ruined cities
of, XV ;
"
Six Cities
"
of,
293 ;
tradition of its colonisation
from Taxila,
383
Khotan people, love lot-
teries, 236
;
pilgrimages of,
239 ;
prospecting for gold, 248
;
racial amalgam of,
464
Khotana, ancient name of Kho-
tan, 382
Khudabad,
47
sqq.
Khunjerab Pass,
64
Khunjerab R.,
50
Khuruz,
143
Khush-bel,
56
sqq.
Khush-kishlak Valley, 105
Kie-pan-to, old name of Sarikol,
68
Kie-sha, old name of Kashgar,
123
Kighillik, remains at,
427
kila mudra, Pr.,
382
Kilik (Kalik) Pass,
52 sqq.
kiltas (baskets for baggage),
4
Kimnaras,
359
kirakash,
"
carriers," P.,
92, 98;
of Khotan,
187 ; of Turkestan,
470
kirgha (felt-hut),
70, 72,, 78
Kirghiz herdsmen,
64, 7:^, 78,
80
sqq., loi
; as guides,
91
;
women of, 81
;
on Alai Mts.,
474
sq.
kirghiz, felts called,
320
Kishanganga R., 14 sqq.
Kis-sel, stream, 191
kiyik (wild goat), T.,
205
kizil,
"
red
"
T.
Kizil, oasis,
147
sq.
Kizil-jaim,
144
Kizil-Jilga, 106
Kizil-kum, 311
Kizil-su, R., Ill sq., 124
waters of,
473
sq.
Knight, Mr.,
30
Kochkar-Oghil, 406
Kohmari spur, ancient Mt. Gos-
ringa, 226
koichi, "shepherd," T.,
330
kok (kok)
"
blue,"
"
green,"
Kok-moinak, 104
Kok-robat, 148 sq.
Kok-sel Gl.,
98
Kok-sel Gl. (in Gez Defile), 103
Kok-sel Peak,
103
Kok-torok,
56
sqq.
Kok-tumshuk Hill, 100
Kok-yar, 92, 164 sq.
kone,
"
old," T.
kone-shahr,
"
old town," i.e.,
ruined site, T., 143. 392, 409
Kongur, range called,
98
Kongur-debe Gl.,
98
Kosa, 229
Ko.sh Langar, 167
Krasnowodsk,
478
Kucha, xiv,
384
;
one of
"
Four
Garrisons,"
397
;
route to, 411
Kuchkach-bulaki, 223
INDEX. 493
Kuen-luen, range, 187, 195,
208
sqq. ; Ram Singh's surveys in,
^ii
'> sighted from Keriya,
319
;
sighted from Endere R.,
402 ;
last view of, 462
Kuen-luen Peak No.
5,
192 sq..
195
sq., 201 sqq. ;
see Muz-
tagh
Kukvar, in Sarikol,
74
kul (kol),
"
lake," T.
Kul-langar, old site, 178
Kulma Pass,
75
kumat, name for jade pits,
236
Kumat, hamlet, 191
Kum-i-Shahidan, shrine,
249
Kum-Dawan,
72
Kum-rabat-Padshahim, shrine,
179
sqq.
kumush (a coarse grass), T.,
143,
169, 256,
286 sq., 310 sq.,
325,
350, 389, 402, 406
; used in
walls.
416
Kunat, Pass, 218,
223
Kunat-aghzi, 218
kunjara,
"
oilcake," T., 280
Kushana kings, rulers of Punjab,
323, 340 ;
nomenclature of,
380 ; inscriptions of,
384
Kushanasena,
380
Kustana, ancient name of Kho-
tan,
382
Kuya,
465
Labourers, recruiting of,
253
sq.
;
wages of,
235
Lachin-Ata, pilgrimage place of,
415, 417,
420
lacquer, ancient,
292,
385
Ladak, trade to,
187
Ladakis, 121
Lahore, stay at,
3
;
Campagna
of, 152 ;
Moghul gardens of,
447
;
monuments of,
478
langar,
"
travellers' shelter," T.
langarchi.
"
attendant at rest-
house," T.,
321
Langhru,
223 sqq.
language, of Wakhan,
42 ;
un-
known 1., 278 sqq. : languages
spoken in camp,
48
leaf-gold, used in ancient sta-
tuary,
240 sq.,
307
; on exca-
cavated image,
440
sq.

found
at Tamoghil,
425
leather, ancient documents on,
365, 370
sq. ;
in bindings of
Sanskrit MSS.,
371
legend, of Ho-lo-lo-kia,
408 ; of
Khotan stream,
273
; of So-
miya site, 246 sq. ;
represented
in ancient painting, 300 ;
le-
gends about Taklamakan,
303
Levies, of Hunza,
ij
sq. ; of
Kanjut,
31 ; of Nagir,
30
li, Chinese road measure, iii
Lieh-sieh, see Li-sieh
lihitaka,
"
letter," Pr.,
382
Li-sieh (Lieh-sieh, Li-tsa), an-
cient local name, 292 sqq.
Li-tsa, see Li-sieh
Littledale, Mr. and Mrs.,
115
Liu-Cheng
("
Six Cities "),
293
Liu-Darin, Amban of Yarkand,
151, 156 sqq., 162 sq.
Liu-Kin-tang, temple in memory
of, 124 sqq.
Liu-Lai-chin, 128
local cult, in Khotan, 180 sq. ; at
Somiya, 246
local legend, of holy rats,
289
local tradition, continuity of, 418
local worship, tenacity of,
247
loess, banks of,
316, sSj- ;
erosion
in,
174 sqq.,
347, 400, 422 ;
fertility of. 326;
ravines in,
2^j
Lop (or Sampula), 422
Lop-nor, route to,
319, 322
lotus, represented in frescoes,
273
sq.
;
grown in Kashgar,
273 ;
pedestal in form of, 269
Macartney, Mr. G., 8 ; Kashgar
residence of. 112 sqq. ;
his
influence in E. Turkestan,
119;
called Ma-shao-yieh by Chinese,
126
; 139
;
acquires MSS.,
290,
292 sqq.,
294;
412,
449, 451.
458, 469
;
his services to re-
search, xxi
Macdonell, Prof. A. A., xiii
Madhumati, stream. 12
Madrasahs (schools), of Kar-
ghalik,
167 ;
of Imam Jafar,
327
sq.
Maharajah, title, in Kluiroshthi
tablets,
341
Mahayana, Buddhist school, 281
text,
393
;
worship, 300
494 INDEX.
Maheb Khwoja, saint, 225
mail coat, of ancient statue, 272
mails, see dak
Mairyam, shrine of Bu M.,
133
sq.
Maitreya Buddha, 226
Makuya,
465
Malakalagan,
415
Mandarins, at Kashgar,
119; see
Ambans
Manners Smith, Capt., see Smith
manuscripts, excavated at Dan-
dan-Uiliq,
276 sqq. ; at En-
dere,
393
sqq. ;
fragment of,
229 ;
see inscribed tablets,
Pothis
map, of Stein's surveys, x
marching powers, of Kanjutis,
48
Mar Canal, 6
Marco Polo, 168,
413
sqq.
;
"
Pein
"
of,
304
Margilan,
477
margs, Kashmir plateaus called,
6, 16 ;
of Alai,
474
;
of Duyan,
22
marketting, at Niya, 322 ;
see
Bazars
markobans (pony-men), 13, 14
marmots, Himalayan,
17, 76
Marinus, of Tyre, 68
Matargom, 12
matting, of ancient roof,
343
Mauri Tim Stupa,
135
sqq.
mazar,
"
sacred tomb,"
"
shrine,"
A.
Mazars, cemeteries around,
244 ;
of Kohmari, 226 ; of Somiya,
246 ;
of Ordam-Padshah,
144
sqq., ;
see Ziarats
Mecca, pilgrim to,
477
medicines, dispensing of,
70 ;
sought at Khotan,
231
Medinsky, General,
477
mendicant monks, garb of,
274
merry-go-round, miraculous es-
cape by, 418
Merv,
478
Miles, Capt. IP.
J., 34
millinery, ancient,
440
Minaur,
24
ming-bashi,
"
head of a thou-
sand," T., official called,
65,
476
Minimarg, 16
Mintaka Pass,
50, 52,
62 sq.
mirab,
"
irrigation steward," P.,
466
miracle-working statue,
440
Mirza Alim, from Kokand,
7, 86,
"5
Mirza Haidar, chronicle of,
249 ;
on Keriya R., 410
Misgar,
47, 49, 50
miskal, Chinese coin,
159
Mitaz,
214, 217 sq.
mocassins, see Charuks
Mohand Marg, 1,6
Moji,
173, 176 sqq.
Mokuila,
173
monastery, excavation of, 281
;
the Hu-kuo m.,
295 sqq.
monkey, figurines of, 240
;
gold
statuette of,
465
monks, Chinese,
296 sqq.
mosque school,
254 ; see Ma-
drasahs
mountain sickness,
91,
93
Mudache-tagh Pk., 212
Muhammadabad,
40
Muhammadans, E. Turkestan
conquered by, xiv
Muhammad-Ju, 1
1
5
Muhammad Nafiz,
42
Muhammad Nazim, chief of
Hunza,
38
Muhammad Rah, of Hunza,
38
Muhammad Shah, hunter,
407
Muhammad Sidig,
452
Muhammad Tari,
452
Muhammad Yusuf,
53,
62
mujawirs (custodians of shrines),
A.,
145
sq.,
147
Mullahs, of Kohmari Mazar,
227
Mullah Shah,
407
Murghab Valley,
57
Murkhun,
47
sq.
Murkushi,
52
Muz-tagh, range,
30
Muz-tagh-.\ta,
65, 72 sqq., 80
sq(i.,
84 sqq. ; ascent on. 88
sqq.
;
glaciers of, 87 sqq.. 91 ;
height of,
98 ;
highest {X)int
reached on,
94
sq.
;
peaks
of,
89
Muz-tagh, Kuen-luen Pk.
5
called,
193,
201 sqq., 212,
215,
218,
220
;
glaciers of, 205, 209
;
northern slopes of, 282
INDEX. 495
Xaga, storm deities, represented as,
125 ;
legend of Naga lady,
274
Nagara-khana, mound of,
249
Nagara-khana, shepherd's hut,
325
Nagir, hill tract,
29, 32 ;
Nagir
people,
47
;
their bravery,
30
Naiza-Tash Pass, 68
namadi. term for Numdah,
380
Nanga-Parbat,
19, 23, 72
Nar-Bagh,
447
sq., 461
Nestorians, at Kashgar,
138
"
New City," see Kashgar, Kho-
tan
New Dominions, E. Turkestan
called, 120, 124, 128,
294
Niaz, labourer, 278, 287
Niaz Akhun, Chinese interpreter,
115, 165, 169, 200, 209, 231
;
in charge of ponies,
254, 316,
321 ;
troublesome conduct of,
361 sqq. ;
farewell to, 466
Niaz Hakim Beg, Yarkand palace
of, 152, 167
;
governor of Kho-
tan, 238 ; exploits Yotkan site,
239
;
his Madrasah at Imam
Jafar, 32S
; 390,
420 ;
Khotan
residence of,
447
Ni-jang, ancient name of Niya,
322
Nilth,
29 sqq.
;
storming of, 21,
30
nimbus, in ancient paintings,
300
Nissa Valley,
208T
210 sqq.,
223 ;
glaciers of, 213, 219 ;
men of,
210, 215, 217
Niya, oasis,
319
sqq. : ancient
frontier of Khotan,
322 ; iden-
tified with Ni-jang, 316 ;
anti-
quities acquired at,
^2^
; la-
bourers brought from,
392 ;
return to,
403
;
marshes near,
404
Ancient site beyond Niya,
first reported,
318 ;
arrival at,
332
sq. ;
first excavations,
334
sqq. ;
ruined residences,
352
sqq. ;
ancient rubbish heaps.
365
sqq. ; Kharoshthi docu-
ments,
370
sqq.
Niya R., fed by marshes,
325 ;
course of,
325
sqq. ;
ends in
jungle,
330
sq.
;
sand-dunes
along,
403
Nogai,
474
Nomal,
27
non-Sanskrit language, in ancient
MSS., 289 ; its affinity, 290.
392.
394
Numdah, ancient saddle-cloth,
299
; see namadi
Nura, hill stream of, 421
NuruUah,
330, 361
Nurunam,
143
Nysa, mythic locality, 210
0.\SES, terminal,
391, 417
oats, antique, 412
Och-tobe,
474
official documents in Kharoshthi,
341
sqq.,
371
oilcakes
("
Kunjara "), 280, 282
"
old books," 170 ;
forging of,
452
sqq.
"
Old City." see Khotan, Kashgar
Omsha Valley, 200 sq., 203, 205
sq.
Opa, canal,
151
Opal, 106, 109 sq.
Oprang,
36
orchards, remains of ancient,
260, 302 sq., 388
Ordam-Padshah, shrine of, 142
sqq.
Orientalists, International Con-
gress of, xiii
ornaments, ancient Indian.
40
Osh, start for,
473
;
arrival at,
476
Osman Beg, loi
Otra Langar,
326
Ovis Poli,
74
Ovraz Langar,
321
Oxus, drainage area of,
42,
55 ;
sources of,
53, 57
sqq. ; Upper
Valley of, 60
;
watershed of,
474
P.\IN.\P, 161
painted panels, ancient, from
Dandan-Uiliq, 271 sqq., 276
sq., 288 sqq.,
295, 298 sqq.
;
from Endere,
399
paipaks (felt socks), T., 164
palaeography, Chinese,
448
Pallas Athene, on clay-seal,
376,
x\ii
;
reproduced in Vignette,
xxiii
496 INDEX.
palm-leaves, for writing,
347
Pamirs, ethnic watershed on, 42
;
view over,
95
; Taghdumbash
P.,
54
sqq.
;
Russian Pamirs,
474
sq.
Pamirski Post,
62,
474
Pan-Darin, Amban of Khotan,
184 sqq., 198
;
propitiates
saint,
227
;
gives assistance,
230 sq., 252 ;
seen on return,
448 ;
his learning,
448 ;
re-
ceives Preliminary Report, ib.
;
helps to arrest Islam Akhun,
450 ;
prone to pardon,
457 ;
farewell to, 460 ;
favours
Turdi,
466
;
promotes Islam
Beg,
420 ; services acknow-
ledged, xxi
Pandits, of Srinagar,
7, 9,
380
Pandrenthan, temple of,
9
paper, manufacture of,
172, 456 ;
ancient MSS. on,.
277
sqq., 289
sqq.; in ancient Tibetan MSS.,
395
;
not found at Niya Site,
345 ;
period of its use,
384
parapet, of ancient fort,
426
Pari,
24
Parmanand,
142
Pasu,
44
sq.
Payik Pass, 62
Pein, named by Marco Polo,
304,
now Uzun-tati,
413
Peking, troubles at, 61, 119;
called Bajin, 116, 165;
Em-
peror's flight from,
169,
233
penal settlement,
198
Persia, early art influence from,
xvii,
155,
300;
form of sword
. in,
299
Persian language, spoken in
Hunza,
43,
48 ;
spoken in
Sarikol,
70 ;
rapid spread of,
33 ; 390, 464
petition, ancient,
293
sq.
Petrovsky, M. N., 122
; help
from,
470
Peshawar Valley, ancient sculp-
ture in,
441 ; see Gandhara
Phakhpo, hill people, 165 sq.
photo-theodolite, work with
Bridges-I.ee, 26,
57,
60,
84,
91, 100, 196, 209,
211 sqq.,
222
Pialma, 178 sq.
picnics, of Kashgaris, 122
Pigeons' Sanctuarj' (Kaptar-Ma-
zar),
179,
466 ;
offering to,
467
pilgrimages, Buddhist, xiv
;
Mu-
hammadan, 311 ; to Imam
Jafar Mazar,
325
Pi-mo, site of,
304,
408 ;
men-
tioned by Hiuen-Tsiang,
413 ;
legend of,
417
sq.
;
position
identified, 418 sqq. ; identical
with Marco Polo's Pein,
419 ;
occupied until middle ages, ib.
;
now represented by Gulakhma,
420 ;
miracle-working statue
at,
441
Pir Pantsal range,
3
Pisha Valley,
193
sqq., 208, 282,
319
Pishin,
464
Pisling,
64
pits, of jade-diggers,
234
sqq.
plane-table survey, started,
57 ;
accuracy of, 283 sq.,
404
plaques, in stucco relievo,
269.
431. 438
plaster, in ancient walls, 262,
265,
337. 398
Po-kia-i, identified with Moji,
177
police, ancient orders of,
379
Political Officer, at Ka.shgar, 121
;
in Hunza,
30,
34
Polo, Dard game, 18
Polu,
233, 283, 319
Pom-tagh Pass, 209
Ponak,
420
ponies, purchase of,
115; sale of,
470 ;
story of Turdi's pony,
285 sqq.
poplars, ancient trunks of,
7,^2,
359
; see terek
;
ancient poplar
leaves,
350. 359
population, deficiency of, 252
Popuna, 222
Posgam Bazar, 161
Post Office, Indian, 8
;
see dak
Potai, road measure, 148,
332,
392
Pothis, Indian MS. books, finds
of,
274, 276, 289, 291,
297, 301,
393
potsherds, ancient,
173
pottery, d6bris of ancient,
174,
176, 242 sq.,
302, 307, 300. 425
pradakshina, S., 265
INDEX. 497
Prakrit, in Kharoshthi tablets,
341, 379
sqq.
Preliminary Report, on Stein's
journey, vii,
448, 479
presents, Chinese etiquette about,
315
;
as rewards, 461 sq.
priests, superintending convent,
297
Ptolemy,
6-j
Pujia, 220,
222
pid, monetary value,
159
Punjab, Government of, ix, xxii
;
traders from,
1
5 1 sqq.
Punjabi language, spread of,
a
Punyal, hillmenof,22 ;
chiefs of,
48
Purohitas, of Tirthas, 227
Pushtu language,
464
Pushwari, 16
Puszta, Hungarian,
65
Quadrangle, of Rawak Stupa,
427
sqq. ; at Topa Tim,
135
Rabab, ancient guitar,
357
races, amalgam of,
464
rafiks (galleries),
41, 48, 49,
102
rags, as ex-votos,
398
raids, of Kanjutis,
48, 64, 96
Rakiposhi, Mt., 20, 29
sqq.,
^2
sqq.
Ramghat,
23
ramparts, ancient,
398
sq., 408,
419,
426
Ram Singh, Sub-Surveyor, selec-
tion of, X,
7
;
starts plane-
table work,
57 ; 92, 94, 96 ;
surveys Karatash P.,
99 ;
surveys to Khanarik, 138 ;
196, 210, 213 sq., 218, 224 ;
surveys in Kuen-luen,
233 ;
rejoins in desert, 282 sqq.
;
helps at ruins,
284 ; 306, 319 ;
reconnoitres desert,
334
; 362,
413
sq.,
420,
435
;
starts for
Yarkand,
461 ;
appreciation of
his services,
471 ;
returns to
India,
472
Ramzan fast,
322
Rang-kul, 10
1
Rapson, Mr. E.
J.,
assistance
rendered by, xxiv,
378
sqq.
Raquette, Mr. G.,
123
Rashid Beg,
65
Raskam, 121
rats, legend of sacred, 180 sq.,
289, 467
; king of rats, 180

figure with head of rat,
289
ravines, created by erosion,
347
Rawak, ruins near Dandan-Uiliq,
306 sqq.
Rawak Stupa, discovery of,
427 ;
excavations in Stupa court,
429 ;
survey of Stupa,
430 ;
relievo sculptures of,
43
1 sqq.
;
clearing of statuary,
433
sqq. :
work in trenches,
435
;
details
of sculptures,
437
sqq. ;
outer
screen of court.
437
;
secular
figures,
438 ;
relic of quaint
custom,
440 : affinity of its art,
441 ; its date, ib.
;
packing of
relievos,
444
; burial of sculp-
tures,
446
Rawlinson, Sir Henry C,
67
red, auspicious colour,
317
refuse, excavation of ancient,
367
sqq. ;
antique refuse-heap.
427
relics, of saint, 272
relievos, in stucco, of Dandan- s
UiUq,
263, 269 ;
of Endere,
393
'
"^
sq. ; of Kighillik, 426 sq. ;
of
Rawak Stupa,
431
sqq.
Remusat, M. .\bel,
236
Reuter telegrams, 22, 60, 120
rice, antique, 412
rishka, a cereal,
50
Rivaz, Sir Charles, ix
roads, sunk below fields,
244
"
Roof of the World,"
95
roubles, circulating in Turkestan,
159
routes, ancient,
167
Roza .\khun, camel-man, 116
rubbish-heap, ancient,
365
;
docu-
ments from,
367
rug, ancient,
354
rupee, exchange value of, 1
59
rush fence, ancient,
359
Russian Consulate, at Kashgar,
122; Consul-General purchases
antiquities, viii
; help rendered
by,
470
Russian frontier,
473
Russian Pamirs,
55,
62, 76
Russian Turkestan, journey
through,
474
sqq.
;
postal ser-
vice through,
98
32
498 INDEX.
Sadak Akhun, cook,
7,
8, 13, 86,
III,
148,
221
;
under influence
of Charas,
362 sq.
; 473
; dis-
charge of,
476
saddle, in ancient painting,
300
Saduk-Langar,
148
Safsgos,
73
Sahib, designation for Europeans
in India, A.
sai (stony desert), T.,
229, 320,
etc.
Sai-bagh, no
saint's intercession, for water,
227
Sakyamuni, see Buddha
;
preach-
ing of, 226
saline water, 222 sq.
Salistamba-sutra, Tibetan MS. of,
395
salt, dug at Imam Jafar,
329
Salt Range, 210
Samarkand,
477
Sa-mo-joh, Stupa of,
245 ;
site
identified, 246 sq.
Sampula (or Lop),
422,
453 ;
market of,
464
sand-cones, with tamarisks,
256,
331
sq., 366, 407
sq.,
414
sand-dunes, advance of, 170 sq.
;
direction of, 258, 294, 425
; form
semi-lunes,
144 ;
height of,
256,
277, 309
sqq., 311,
389,
391, 407, 417, 425, 427
; move
ment of, 146, 308 ;
conjectures
about movement of,
304 ;
towards Keriya R.,
309
sqq.
sand-ocean, Indian legend of,
383
sand-storm, see Buran
Sanskrit, known in ancient Kho-
tan, xviii,
297 ;
Sanskrit corre
spondence,
380 ;
Sanskrit ety-
mology, for name Khotan,
382 ;
Sanskrit MSS., finds of,
274, 276 sqq.,
297, 393
;
S.
official designations, 380 ; S.
terms in Kharoshthi docu-
ments,
379
; S. terms in non-
Sanskritic MSS., 291 ;
Sanskrit
words of command. 28
sarai,
"
travellers' rest-house,"
P. ;
ancient,
410
Sarala,
y^
Sarguluk Pk., 103, 106
Sarhad, 60
sarik,
"
yellow," T.
Sarik-Jilga, 62
Sarikol,
67 sqq.
;
Begs of,
56,
62, 70 ;
raids into,
36 sq.
;
range of,
76
Sarikolis,
50, 53
sqq. ;
language
of,
70
Sarik- tash Pass,
y^
Sarvai-Bel Pass, 106
Satip Aldi, 88, 104
satma, "reed hut," T.,
255, 311
sq.
Satok Boghra Khan, king, 122,
13s
Schliemann, Dr., xi
I
sculpture, Buddhist, in India,
263 ;
see relievos
seal impressions, on inscribed
tablets,
338, 345
;
classical,
xvii,
376
sq.
sealing, of Kharoshthi tablets,
372
sqq.
.
ser, monetary value,
159
Serike, 68
servants, engagement of,
7
sq.,
115 sqq.
sgraffiti, at Endere, Tibetan,
396 ;
Chinese,
397
sq.
Shagildik Dawan, 106
Shahids (martyrs), 180
;
supposed
tombs of,
329
shahr,
"
town," P.
Shamalda, spur,
87,
97
Shami Sope, 246 sq.
Shams Beg,
79
Sheikhs (attendant priests), 180
;
of Kohmari cave,
227 ;
of Bur-
hanuddin Mazar,
311, 405
;
of
Imam Jafar Mazar, 361 ;
of
Arish-Mazar,
414
Shen-shen,
384
shepherds, along Keriya R., 312
;
supposed
"
semi-.savages," 406
Sher Muhammad, Munshi,
56,
64,
66,
69, 70
.shield, ancient,
354
Shields, Mr. Cutlibert, xxiv
Shighnan, 68
;
immigrants from,
70, 154
Shikarpur, Hindus from, 140 sqq.
Shina, people, 48 ;
language,
32
Shindah, gorge,
72
Shirin-Maidan,
52
Shitala-Darya,
403
Shitala Padshah Mazar,
403
INDEX.
499
Shivul, marshes of,
414 ; stream
of,
414
Shodhoga, ancient title,
380, 382
shoes, ancient,
385
shrines construction of, at Dan-
dan-Uiliq, 262
Shush Pass,
72
Sikhs, their rule in Gilgit,
25
Sikkim, 2
Siligh Langar,
169
silks, ancient, xviii,
385, 398
silt deposits, due to irrigation,
244
sq., 252 ;
at Tam-oghil,
425 ; specimens from Yotkan,
462
Simhasana (lion's throne), 268,
357
Sinae, land of, 168
Sind Valley, 6
Sipa, 182
Sirik-toghrak,
234
Sita (Tash-kurghan) R., 68
Si-yu-ki, records of Hiuen-Tsiang,
119, 156, 164
"
Six Cities," ancient name of
Khotan district,
293
Skobeleff, General.
476
Smith, Capt.
J.
Manners,
20sqq.,
22
snow, in desert,
332
soda efflorescence, in desert,
391
Soghak-Oghil,
205
som, rouble called,
159
Somiya, site of,
246 sq.
Sop Xiaz Baba,
135
Sramana (monk),
380
speculation, in jade-mining,
235
spoons, ancient,
386
springs, appearance of,
421
Srinagar,
4
sq., 6
;
seven bridges
of,
9
;
ruined temples of,
9
standard of living, in E. Turke-
stan,
178
stationery,
ancient terms for,
382
statue, of colossal Buddha,
265
;
of Yaksha,
2^2 sq.
; see
Buddhas, relievos
Stein, Dr. M. A., forms first plan
of journey, vii sq.
; obtains
sanction of plan, ix
;
previous
archaeological labours of, x sq.,
2
;
surveys of, in E. Turkestan,
X sq.
;
Orientalist
Congress's
resolution
concerning, xiii
;
offi-
cial duties of, xxii ; his pre-
vious Indian tours, 2
;
starts
from Calcutta,
3 ;
prepares
in Kashmir,
4
sqq. ; edits
Kashmir Chronicle,
5
; leaves
Srinagar,
9
; arrives in Gilgit,
24 ;
reaches Hunza, 28
;
crosses Kilik P.,
54
; visits
Oxus source, 60 ; arrives in
Tash-kurghan,
65 ; camps at
Kara-kul,
78 ; climbs on Muz-
tagh-Ata, 88
; travels to Kash-
gar,
99 ;
prepares caravan at
Kashgar,
114; visits Chinese
officials, 118; examines old
sites, 130 ; leaves for Yarkand,
139;
halts at Yarkand,
150;
sets out for Khotan,
167 ; first
arrival in Khotan, 181
;
starts
for Kuen-luen Mts., 190 ;
ex-
plores Yurung-kash gorge,
203 ;
crosses Kara-kash ranges,
214 ;
triangulates Khotan, 221 ; re-
turns to Khotan,
229 ;
visits
Yotkan site,
237
; starts for
desert,
251 ; experiences great
cold,
257
; excavates at Dan-
dan-Uiliq, 262
;
marches to
Keriya,
306 ; sets out for an-
cient site beyond Niya,
319 ;
discovers Kharoshthi records
on wood,
334
; excavates an-
cient rubbish-heap,
365
;
ex-
plores Endere ruins,
392 ;
visits Kara-dong,
408 ; iden-
tifies Pi-mo site, 418 ; last
visit to Keriya, 420 ; explores
Aksipil, 426 ;
excavates Rawak
Stupa,
427 ;
exposes forgeries,
449
;
leaves Khotan,
460
;
returns to Kashgar,
468 ; starts
for Russian Turkestan,
473 ;
arrives in London,
479
; re-
turns to Punjab,
479
"
Stone Tower
"
(Tash-kurghan),
67
storms, of desert region,
298
stovana,
382
stratification, in silt-deposits,
425
straw, ancient, discovery of, 286,
386
string-hole, in ancient MSS.,
278,
281,
291, 372 ; in ancient
tablets,
344
500 INDEX.
stucco, gilt,
307
; relievos in, 262
sq.
;
colossal statues in,
429
sqq.
;
see relievos
Stupas, called "memorial towers,"
367
;
chamber in,
138
;
pro-
portions of, 400 ; Stupa at
Thol, 32 ; at Tcish-kurghan,
69 ;
near Kashgar,
123 ;
near
Khan-ui,
135
sqq. ;
on Afghan
border,
136;
near Mokuila,
173
sq. ;
near Pialma, 178
;
at Chalma-kazan,
234
;
of
Sa-mo-joh,
245
sqq.
;
of Rawak
307
;
of Niya Site,
332, 352,
366 sqq. ;
of Endere,
392,
400
sq.
;
of Hanguya, 422 ;
of
Rawak,
427
sqq. ; at Kara-
dobe,
465
Su-bashi,
76, 83
Suko-sai Valley,
217
Su-le, old name of Kashgar,
384
Sulaiman Khagan, coins of, 176
Sultanim,
144
Sumair glacier,
38
Sung dynasty, coin of,
419
Sung-yun, 68
Sun-Ssu-yieh, Munshi, 118,
294
superintendents of monastery,
297,
sqq.
sureties, in ancient bonds,
297
Surghak, goldfields of, 320 sq.
Survey of India Department, help
of, X,
7
Suziije Darya,
390
Svastika, emblem,
40
Swat, ruins of,
356
Swedish missionary, portrait of,
451
sword, in ancient painting,
299
Tablets, writing on wooden, sec
inscribed tablets
Tabloids, 8, 231
Ta-fan, Chinese name of Tibetans,
397
Tagharma, plain,
73
sqq.
Tagharma-su, R.,
72,
75
Taghdumbash Pamir,
53
sqq.
Taghdumbash R.,
64, 72
taghliks
"
hillmen," T.,
193
sqq.,
etc.
;
deceived by Lslam Akhun,
451
Ta-jen, Chinese title, 156
takhta (writing tablet), P.,
291
sq. ;
ancient documents on,
342, 344,
382
Takhtuwen,
179
Taklamakan, desert, xv, 168, 251
sqq. ;
dust-haze of,
219 ; le-
gends about,
303,
418
;
phy-
sical conditions of, 308 ;
sand-
storm season in,
429 ;
so-called
"
City of T.," 230 ;
"
the
Aksakal of the T.,"
386, 466
Takshasila, the Taxila of the
Greeks,
383
Talantik Pass, 106
Taldik Pass,
474
Ta-li, Chinese period, 292 sqq.
tamarisks, in desert,
256,
286,
309,
310 sq.,
326, 331
sq.,
350, 391, 416;
writing on, 292
Tam-oghil, culture-strata of,
425
tanga, monetary value,
159
sq. ai
Tang dynasty, monk of, 119; Si
coins of,
233, 242
"
Tang-Seng, Hiuen-Tsiang called,
119, 156, 157,
166, 169;
his
evidence invoked, 230 ;
re-
garded as Arhat, 231
;
316 ;
benediction of,
471
tank, ancient,
387
Tao-tai, of Kashgar, 118 sqq.,
273, 470;
death of,
471
Tar-bugaz, bridge of, 131
Tar-bugaz Langar, 181, 466
tarigh (a pulse), T., 412
Tarikh-i-Rashidi, chronicle,
249
tarim,
"
colony," T.
Tarim R., 41 1,
474
Tarim-Kishlak, 191
task
("
stone," T.), road measure,
III
Tashbalik, see Tash-malik
Tash-bulak, see Tash-malik
Ta-shih, Chinese period,
384
Tash-kurghan, town and fort,
65
sqq. ;
raids to,
36
; marches
to,
'48
; ruins of, 68 sqq.
;
Munshi at,
56
Tash-kurghan R.,
65
Tashlik-Boyan Pass, 191
Tash-malik oasis, 105 sqq.
Tashsot,
34
(askirti, legendary called,
249
Tati, ancient sites called,
174
sq.
;
course of decay at,
417 ;
of
INDEX. 501
Jamada, 190,
233
;
of Han-
guya, 422 ;
of Jumbe-kum,
446 ; of Kara-dobe,
465
Tati-lik, ruins called,
415
Taun-murun Pass,
474
Tawakkel, oasis,
230, 251, 254
sq.,
261,287,307;
Beg of, 230, 252,
255
; labourers from,
285, 308
tax on skins,
287 ;,
record of an-
cient taxes,
293
Taxila, immigration from,
383
Ta-Yue-chi, Indo-Scythians,
384
Tchaikowsky, General,
477
teeth, as ex-votos,
272
tel, monetary value,
159
telegraph, to Gilgit, 16, 18
;
to
Kashgar, 120
Telwichuk R., 112
temperatures, in desert,
257, 310,
320. 324. 327, 330, 341, 352,
387,413,419
temple ruins, of Dandan-Uiliq,
288 sqq.
;
of Endere,
392
sqq.
tent,
257
terek, cultivated poplar (populus
alba), T.,
258, 332.
^^j
;
as
building material,
410
Terek-aghzi, 201,
207
Terek Pass,
474
teresken (juniper), T., 81
terra-cotta, figurines,
242 ;
vases,
241
theodolite, work with, x,
224
Thian-shan,
114
Thol, Stupa near,
32
Tibet, art of,
300 ;
invasion from,
396
sq.
;
political power of, xix
Tibetan MSS., found at Endere,
395
sqq.
; oldest specimens of,
396 ;
Tibetan graffiti, at En-
dere,
396 sqq.
Tien-tsin, lighting at, 61
Tigharman-su, 61
Tikehk-tagh, Pk.,
196
Ti-kia-po-fo-na, monastery of, 248
Tila Bai, pony-man,
238, 403
tim, ancient mounds called, T.,
135
sqq.
timber, in ancient buildings,
243,
262,
302,
iij,
348, 349 ;
varieties of,
410
Timur, Emperor,
478
"
tips," in E. Turkestan,
461
Tirthas, of India,
227
Tiznaf,
72
Tiznaf R., 161
Togdasin Beg, 80
toghrak (wild poplar), T., 256,
326, 402, 406 sq.
;
dead trunks
of,
331,
410, 415, 456 ;
ancient
timber of, 410 ;
in ancient ice
pit.
350
toghruga,
456
Toguchak, I50sq.
Togujai, ancient site, 176 sq.
Tokhta Akhun, house of,
183,
231, 250
tokuz,
"
nine," T.
Tokuz-ak, 1 10 sq.
Tokuz-Dawan, Passes, 104 sqq.
Tokuz-kol, 402
Toldama,
407
Tonguz-baste, 406 sq.
Topalu-Bel Pass, 106
Topa Tim, near Guma,
173
Topa Tim, near Khan-ui,
135
top-bashi, small official called,
T., 420
Tok-khana,
50
Tosalla, 462
Towen-Bazar, Guma,
172
trade-routes, across Pamirs,
64
.
68 ;
through Khotan, xv
Tragbal, Pass, 12 sq.
Trans-Alai Mts.,
95, 474
Trans-Caspian railway, 321
;
per-
mission to travel by,
470 ;
ter-
nninus reached,
476
sq.
;
jour-
ney by,
478
transport arrangements, in Kash-
mir,
3,
II
;
in Gilgit, 26; in
Hunza,
34
;
for stucco relievos,
444
treasure-seekers, of Khotan, viii.
186, 188; open Stupa,
.234 ;
diggings of,
259
sq.,
394, 427
sq.,
434
;
destruction dealt by,
XX sq.,
301, 426
trees, dead, in desert, 258, 260,
309
;
see terek, toghrak
triangulation. in Kuen-luen,
215,
223 sqq., 282 ;
connection with
Indian surveys, 218 sqq.
Trisula, S., ornament,
299
Tsin, Chinese dynasty,
385
Tsin-kia-hing,
397
tugachis,
"
camel-men," T., 116
Tughlan-shahr,
65
INDEX.
Tiigiilaz, i6i
Tulkuch-kol,
330
Tiimen-Darya, R., 114
tungchi,
"
interpreter," T.,
167,
316
Turdi Khwoja, treasure-seeker,
229sqq.,25i; guides in desert,
258 ;
familiar with ruins, 260
;
watches excavations, 282
;
his
antiquarian instinct,
285 ;
story
of his pony, 285 sqq. ;
earlier
MS. finds of,
290, 293, 294 ;
306, 309
;
helps to clear refuse-
heap,
386
; 405 ;
rejoins,
415 ;
directs guide, 418
; 422, 426
;
guides to Rawak,
427 ; 435,
453
;
confronts Islam Akhun,
458 ;
farewell to,
466
Turfan,
397
Turkestan, Chinese, called
"
New
Dominions," 120, 124, 128,
294 ;
Chinese sovereignty over,
385 ;
climate of, xxi
;
currency
of,
159
sq. ;
folklore of,
303 ;
gardens in,
447
;
houses in,
342, 354
;
surveys in, x
;
Tibetan invasion of, xix,
396
sqq. ;
see Chinese
Turki language, study of, 8, 122
;
conversation in,
70, 316
;
writing in,
254, 287
Turki people, features of, 322
;
power of assimilation in,
464
Uftu langar,
144
ui,
"
house,"
"
hut," T.
Ui-toghrak,
320
Ujadbhai, see Ghujak-bai
Ujat, 226 sqq. ;
228
Ularlik-Jilga, 105
Ulughat, Pass, 218 sqq. ;
descent
from, 222 sqq. ;
range, 226
;
last view of, 462
Ulugh-Dawan, 192 sq., 196
Ulugh-Nishan Mazar,
147
Ulugh-Rabat Pass,
76
Ulugh-Sai Valley, 321
Ulugh-Ziarat, remains at,
419
un-bashi,
"
head of ten men," T.,
193
under-tablets, of Kharoshthi
documents,
372
sq.
unknown characters, in forged
MSS.,
449
sqq.
Urumchi, 120, 125, 169, 461 ;
governor-general at,
448
liriik, plum-tree, T., 422
ustads
{"
masters," P.), craftsmen
called,
117,
121, 131,
139
ustang,
"
canal," T.
Uttarakurus,
7
Uzun-tati, ancient site of,
304,
418 sqq.
;^
debris of,
419
Vaisravana, as genius loci,
465
Vajracchedika, MS. of, 281
vesica,
289
Victoria, Queen-Empress, death
of, 412
vignette of title-page, design for,
xxiii
Vihara (Buddhist monastery), S.,
226
;
excavation of, 281 sqq.
vines, at Ujat, 228
Virochana monasterv, position of,
248
Volur Lake,
4,
10, 12
votive offerings, at Dandan-Uiliq,
271 sqq., 301 ;
at Rawak
Stupa,
443
;
see ex-votos
votive Stupa,
443
Wages of labourers, 252
Wakhan,
42, 60,
89
Wakhi, language,
70
Wakhis, settlements of, 42
sqq.,
48, 52, 64, 70 ;
herdsmen,
52
sqq. ; at Yarkand,
154
Wakhjir, Pass,
57
sqq.
Wall Muhammad, guide,
91,
94
sqq.
walking-stick, ancient,
354. 359.
362
Wang-Daloi,
233
sqq.
ivaqf,
"
endowment," A., 148
washing, for gold, at Yotkan,
239
sqq.
water, in desert, 256, 284, 307 ;
see wells
;
pilgrimage for, 227
water-lined paper, ancient, 289,
292
water-tanks, used in desert, 2,
117. 257. 309. 329
sq.,
407. 415
wattle and plaster walls, 288
Wazir, of Hunza chief,
34
wedge-shaped tablets,
338
sqq.,
344, 354
;
term for. 382
wells, dug in desert, 256, 259, 284,
307,
310, 428 sq.
INDEX.
503
wheat, antique.
412
wheel, sacred,
40
IVilayet, Europe called. A.,
458
sq.
wood, as writing material,
307,
347
;
camp, inscribed tablets
wood-carving, ancient,
354,
i^j
;
in Hunza,
40
Woodburn, Sir John, ix,
3
wool, exports of,
165
writing pens, in wood,
346, 357,
371
Wurshki, language,
32
Wu-tchu, Chinese symbols,
443
Wu-ti, Emperor,
384
Yagan-Dawan, 215 sqq.
yailaks,
"
summer grazings," T.,
475
Yaks,
52 sqq., 56 sqq., 88, 92 sq.,
200, 203, 207, 208, 215, 217,
223, 231
Yak's tails, on graves. 200
yak-shamba,
"
Sunday," P.
Yak-shamba, Bazar, of Khan-
arik, 140
Yakshas, figures of,
272,
438
Yamala Pass, 107
Yaman-yar R.,
98, 109, 140, 142
yambiis. silver pieces called,
159
Yambulak Glacier, 81,
87,
88 sq..
91 sqq.,
96
Yambulak Valley, 88
Yamen, Chinese official residence
called, 118 sqq., passim
;
gates
of, 412
yangi,
"
young," T.
Yangi-arik,
251
Yangi-Darya, 182
Yangi-Hissar, 106
Yangi-Langar.
191
Yangi-Shahr ("New City"), of
Kashgar, 120, 126 sqq. ; of
Keriya,
315
; of Yarkand,
152
Yaqa-Langar,
420
Yaqub Beg- ruler of Turkestan,
124, 134, 161, 181
yar,
"
ravine," T., cut in loess,
237, 425 ; of Niya R.,
330
Yar-bagh,
131
Yarkand,
150 sqq. ; departure
from,
160, 161
;
return to,
468
sq.
;
foreign colonies at,
153
sqq.,
464 ; racial amalgam at
464 ;
trade relations of,
155
Yarkand R.,
55.
160 sq.
Yarkandi horse,
299
Yar-tungaz R.,
389
sqq.
;
jungle
of,
390 ;
shifting of,
390, 403
Yar-tungaz Tarim,
391
sq.
Yasin,
32
Yawa, stream,
465
Yesyulghun,
320
Yetimlukum, 166
Yoghan-kum,
406
Yokakun, 182
Yoke-toghrak,
403
"
Yolchi Beg," Stein's fox-terrier
,
17, 41, yy,
86,
147,
200,
217,
237- 257. 295, 313, 403, 423 ;
returns to India,
472
Yotkan, site of ancient Khotan,
187, 237
sqq.,
245
sq. ; dis-
covery of, 238 sqq.
;
excava-
tions at,
239
sq. ; antiques
found at, 238 sqq. ; alluvium
at,
243
sqq. ; culture-strata at,
242 sqq. ; last visit to, 462
;
Yotkan Yar, 238 ; its forma-
tion,
240 ;
extent of,
245, 248
sq.
Young, Sir Mackworth, ix
Younghusband, Major F. E., 66
yu,
"
jade," Ch.,
236
Yule, Sir Henry, 68, 189 ; iden-
tifies Marco Polo's Pein,
419
Yurgal Gumbaz,
65
viirt. Kirghiz felt-hut called, T.,
66, 74,
81, 88
Yurung-kash, R.,
187 sq. ; head-
waters of, 190 sqq., 196, 201,
203 sqq., 211, 214, 219, 220,
233 ;
course through desert,
251
sq. ;
canals from,
422 ; floods
from,
425 ;
jade washed from,
233
sqq.
;
percolation from, 428;
Yurung-kash R. gorge,
197,
201
;
advance in,
203 sqq.
;
distant view of, 212
Yurung-kash, canton,
423 sq., 466
Yurung-kash, town, market of,
464
yiiz-bashi,
"
head of a Hundred
"
(headman), T.
Zanguya,
177
Zawa, 181,
465
ziarat,
"
shrine," A.
; see Mazar
Zilan, 201
PRINTED BY
KBLLT'S DIRECTORIES LTD.
LONDON AND KINGSTON.
404
PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET
UNIVERSITY OF TORONTO LIBRARY