Journal of Volcanology and Geothermal Research, 8 (1980) 315-336.

315
Elsevier Scientific Publishing Company, Amsterdam - Printed in Belgium

A WORKING TERMINOLOGY OF PYROCLASTIC DEPOSITS

JOHN V. WRIGHT’, ALAN L. SMITH’ and STEPHEN SELF2

‘Department of Geology, University of Puerto Rico, Mayaguez, PR 00708 (U.S.A.)
‘Department of Geology, Arizona State University, Tempe, AZ 85281 (U.S.A.)
(Received December 14, 1979; revised and accepted March 13, 1980)

ABSTRACT

Wright, J.V., Smith, A.L. and Self, S., 1980. A working terminology of pyroclastic de-
posits. J. Volcanol. Geotherm. Res., 8: 315-336,

A nomenclature for pyroclastic deposits is proposed which it is hoped will provide a

working terminology. Three basic types of deposit may be distinguished: fall, flow and
surge, and descriptions are given of their dominant characteristics. No unique classifica-
tion for pyroclastic rocks can be made and at least two systems are required : (1) a genetic
classification to interpret the genesis of a deposit, and (2) a lithological classification which
may be solely descriptive, but also which may be used to discriminate on a lithological
basis the mechanisms which produced a particular pyroclastic deposit. Genetic classifica-
tion schemes are presented for various types of fall, flow and surge deposits. A lithological
classification is given based on grain size limits and distribution, constituent fragments and
degree and type of welding. A glossary of some other terms in use to describe pyroclastic
deposits is also given.

INTRODUCTION

The purpose of this paper is to propose a nomenclature for pyroclastic de-

posits and rocks; this system, however, is not meant to be a review of all the
terms now in use (for a comprehensive review the reader is referred to
Fisher, 1966). Work on young pyroclastic deposits by us has shown that the
existing terminology is of limited use for our studies. Accordingly, we pro-
pose a genetic classification relating deposits to eruptive mechanisms, to-
gether with a descriptive lithological classification.
Pyroclastic deposits are those formed by the fragmentation of magma and
rock by explosive volcanic activity. Three kinds of component are found in
any pyroclastic deposit, namely juvenile vesiculated fragments, crystals and
lithics :

Juvenile vesiculated fragments. This term includes highly vesiculated pumice

and denser, less well vesiculated juvenile magmatic fragments. Indeed the

0377-0273/80/0000-0000/$02.25 01980 Elsevier Scientific Publishing Company

316

o+ ‘

l-o l-5 xl 25
DENSITY fpzrn4

Fig. 1. Variation in density of juvenile fragments in a single pyroclastic deposit. The ex-
ample is a block and ash deposit from Mt. P&e, Martinique.

same pyroclastic deposit may contain a complete range in density from pum-
ice to dense non-vesiculated juvenile fragments (termed cognate lithics, see
below). This is illustrated in Fig. 1 with an example from Mt. PelBe,
Martinique.
Pumice is the common product of explosive eruptions involving viscous
magmas. In the literature the term has become inextricably linked to the
larger grain sizes. Consequently we believe it necessary to emphasize grain
size in order to avoid confusion. The terms pumice block or bomb (> 64 mm),
pumice lupilli (< 64 mm to > 2 mm) and ash (< 2 mm) should be used when
appropriate (see Table 5 in section “Lithological classification”). In addition
the terms glass shard and ash shard for ash-sized particles which result from
the fragmentation of pumice vesicle walls is in wide use and cannot be aban-
doned. Morphologically glass shards have various Y and cuspate shapes. The
reader is referred to Heiken (1974) for scanning electron photomicrographs
of various types of glass shards and ash particles.
Scoriu is the term used to describe vesiculated fragments of basalt and
basaltic andesite compositions. The larger fragments tend to have a ropey or
breadcrust surface texture.

Crystals. During the explosive disruption of porphyritic magmas the crystals

are released and can be regarded as forming a discrete juvenile component,
often having their own behavior patterns during transport.

Lithics. This term is generally used to describe the dense components in a

pyroctlastic deposit. Lithics may be subdivided into non-vesiculated juvenile
fragments (cognate Eithics) and country rock which has been explosively
ejected during eruption (access lithics) or, in the of pyroclastic flows
and surges, clasts picked up lo during transport identul Ethics).
 317

Following Sparks and Walker (1973) three basic types of pyroclastic de-
posits are distinguished :
(1) Pyrdastic full deposits. These are produced when material is explosive-
ly ejected from the vent into the atmosphere producing an eruption column
in the form of a convective plume. The plume expands and pyroclasts fall
back under the influence of gravity, down wind, at varying distances from
the source. The geometry and size of the deposits reflects the eruption column
height, and wind velocity and direction (Eaton, 1964; Knox and Short, 1964;
Shaw et al/, 1974; Wilson, 1976; Blackburn et”al., 1976; Wilson et al., 1978).
Fall deposits show mantle bedding, maintaining a uniform thickness over re-
stricted areas while draping all but the steepest topography (Fig. 2a). They
are generally well sorted and sometimes show internal stratification due to
variations in eruptive column conditions. Carbonized wood is generally lack-
ing in these deposits, when it does occur it is invariably restricted to near
vent deposits.

FALL

FLOW

SURGE

Fig. 2. Geometric relations of the three main types of pyroclastic deposit overlying the
same topography.

(2) Pyroclastic flow deposits. PyrQ~l~tic flows involve the lateral move!-
ment of pyroclasts as a gravity controlled, hot, high ~o~~@~t~a~io~gas/sol~~
dispersion, which may in some instances be partly fluidized (Sparks, 1976).
318

Deposits are topographically controlled, filling valleys and depressions (Fig.

2b). They are poorly sorted and sometimes show coarse-tail grading (Smith,
1960a; Sparks, 1976). Poor sorting in flow deposits is attributed to high par-
ticle concentrations and not turbulence, with the dominant flow mechanisms
probably being l&ninar and/or plug flow (Sparks,-1976; Wright, 1979; Wright
and_ Walker, 1930; Wilson, 1980). Individual deposits generally lack
internal stratificatron, although the superposition of a number of flow
units (each flow unit being regarded as the deposit of a single pyroclastic
flow) can give the appearance of stratification. They sometimes contain fossil
fumarole pipes in which the fine ash fraction has been lost making the pipes
enriched in crystals, lithics or vesicular fragments (Walker, 1971,1972;
Roobol and Smith, 1976). Carbonized wood may also be present.
(3) P~roclustic surge deposits. Surges involve the lateral movement of pyro-
clasts as expanded, turbiulent, low-concentration gas/solid dispersions. Depos-
its do mantle topography but tend to accumulate thickest in depressions
(Fig. 2~). Characteristically they show unidirectional sedimentary bedforms
(cross-stratification, dunes, planar lamination, antidunes, pinch and swell
structures, and chute and pool structures) and individual laminae are generally
well sorted (Fisher and Waters, 1970; Waters and Fisher, 1971; Crowe and
Fisher, 1973; Schmincke et al., 1973; Sparks, 1976; Wohletz and Sheridan,
1979). They also can contain carbonized wood.
In detail it seems probable that there is a complete gradation in concentra-
tion from high-concentration, high-density pyroclastic flows to low-concen-
tration, low-density surges. Indeed ash-cloud surge deposits represent the
lateral equivalents of pyroclastic flows (Fisher, 1979; Fisher et al., 1980). Of
course surges are a type of flow, but the term pyroclastic flow has tradition-
ally been associated with the high-concentration flows and it is appropriate
to classify the fundamentally different types of deposits produced by flows
and surges separately.

PROPOSED CLASSIFICATION

There is no unique classification system for pyroclastic rocks and at best

two quite different systems are needed. One to interpret the genesis of de-
posits which can then be related to a volcano’s history, behavior pattern and
eruptive mechanisms. The other system is lithological. Such a system is primari-
ly descriptive, describing major characteristics of a deposit such as grain size.
However, these features are themselves often diagnostic of a particular process
and allow conclusions to be made on the deposit’s genesis.

Genetic classification

The genesis of-a pyroclastic deposit is partly deduced from its lithology
but also from its overall geometry and field relations. In ancient volcanic ter--
rains where rocks may have suffered much erosion .and even tectonic deforma-
 319

tion the record may not be good enough to determine parameters such as
geometry and gram size, and hence, a genetic classification cannot be rigor-
ously applied. A genetic classification system can indeed only be rigorously
applied to very young, Quaternary deposits.
For pyroclastic fall deposits the classification scheme of Walker (1973) is
adopted (Fig. 3). This quantitative scheme relies on accurate mapping of the
distribution of a fall deposit and detailed granulometric analysis (for methods
see Walker, 1971; and Walker and Croasdale, 1971). The term phreatoplinian
has been introduced by Self and Sparks (1973) for the salic analog of a
surtseyan deposit (which could more strictly be termed phreatostrombolian).
Ultraplmian has been recently introduced by Walker (1980) to describe the
most widely dispersed plinian fall deposits.

100 I
1
S”RTSEY*N P”RE.4TOP‘,N,A,.I
/-C-
’ I
ULTRA-

FX ,p+ /’ 1
PLINIAN

/ ’

A’ ’ I
//
I
AN’ I
/
--f
---
-“IIw*,,AN1 - - - - - ; SUB-PLINIIN 1 ,/’
STROMBOLllN
I I /
0 I I , I I I
O-05 5 500 5000

D km’

Fig. 3. Classification scheme of pyroclastic fall deposits after Walker (1973). Plot of F
(weight percentage of deposit finer than 1 mm on the axis of dispersal where it is crossed
by the ‘0.1 T,, isopach) against D (the area enclosed by the 0.01 Tmax isopach). Ultraplinian
has been introduced by Walker (1980) for the most widely dispersed plinian fall deposits.
The field of vulcanian deposits is described in Fig. 4 and discussed in text.

Vulcanian pyroclastic fall eruptions (short-lived explosions) appear to be

common on andesite and basaltic-andesite strato-volcanoes (Self et al., 1979)
and they generally produce small-volume (< 1 km3), fine-grained ash deposits
with a wide dispersal (Fig. 4). However, during recently observed eruptions,
for example Fuego 1974 (W.I. Rose, personal communication) and
Ngaurohoe 1975, coarser-grained scoria fall deposits of more limited dispersal
were produced during periods of more intense, maintained explosions. These
occurred intermittently with short-lived explosions and hence two types of
deposit were formed during eruptions which have overall been termed vulcani-
an. These coarser scoria fall deposits seem to have similar fragmentation and
dispersal indices to those deposits termed violent strombolian by Walker
(1973). The approximate field of vulcanian deposits is shown in Fig. 4 by a
few examples.
320

TABLE 1

Genetic classification of Pyroclastic flows

Pyroclastic flow Deposit

:ulated

1‘

_w--
pumice flow ~ ignimbrite;
pumice and ash _q
deposit
/

Eruption COlumn CotiPse scoria flow scoria and ash -

I\ deposit

Decreasing average
density of juvenile
clasts

\
semi-vesicular - semi-vesicular -
andesite flow andesite and ash
deposit

explosive __ block and ash flow;- block and ash -

nub ardente deposit
/

Lava/dome collapse

Non-vesiculated gravitation&--- block and ash flow;_ block and ash y

m&e ardente deposit

1
I
 321

Historic examples Comments

~ Large-volume deposits formed by

/_A g~;yJa~;~;gy!$;2 continuous collapse of a plinian
(Fenner, 1926; Curtis, 1968) eruption column as envisaged by
Sparks et al. (1978). Salie in
composition

_- Komagadake, Japan, 19 29 Small-volume deposits probably

Aramaki and Yamasaki, 1963) formed by interrupted column
collapse as in the case of scoria
flows described below.
Intermediate to salic in
composition

SoufrGre, St. Vincent, 1902 ------Small-volume deposits probably

_---- (Anderson and Flett, 1903; Roobol formed by interrupted eruption
and Smith, 1975); Hibok-Hibok, column collapse produced by short
Philippines, 1950-51 (Macdonald explosions (see Nairn and Self.
and Aicarez, 1956); Mount 1978). Basalt to andesite
Lamington, Papua, 1951 composition. Some deposits do
(Macdonald, 1972); Mayon, contain large unvesiculated blocks,
Philippines, 1968 (Moore and e.g. those of Ngauruhoe 1975
I Melson, 1969); Ngauruhoe, North
Zealand, 1975 (Nairn and Self,
1978)

Asama, 1783 _ Small-volume deposits composed of

(Aramaki, 1956,1957) semi-vesiculated angular andesite
clasts

Mt. P&e, Martinique, 1902 - Small-volume deposits, usnall~

(LaCroix, 1904: Roobol and Smith, andesitic or dacitic in composition.
1975; Fisher et al., 1980); Produced both by explosive
Bezymianny, Kamchatka. 1956 collapse of an actively growing
(Gorshkov, 1959); Santiaguito, dome or lava flow and by the
Guatemala, 1973 (Rose et al., 1977) collapse of a vertical eruptive
column as recognized in the early
eruptions (e.g. May 8 and 20) of Mt.
P&e 1902 (Fisher et al., 1980)

Merapi, Indonesia, 1930 (Van ~ Small-volume deposits usually

Bemmelen, 1949); Santiaguito, andesitic or dacitic in composition.
Guatemala, 1967 (Stoiber and Rose, These are the hot-avalanche
1969) deposits of Francis et al. (3.974)
 322

F % 46% /
“4_, *
5Q-
/
A”75
A, ,q PLINIAN ,‘I
’ I
,--’ I /
/’ /
__--- /
I
/
SUB-PLINIAN I /
I ,
01 I -I

D km’

Fig. 4. Fragmentatian F plotted against dispersal D (terms defined in Fig. 3) to show the
field of vulcanian deposits which is discussed in text; ce8 , 1: eruptions of Cerro Negro,
Guatemala, in 1968 and 1971; e: an old undated fall desosit of Mt. Egmont, New Zea-
land; e,,,,: fall deposit of the 1665 eruption of Mt. Egmont; fT1: eruption of Fuego,
Guatemala, in 1971; i,,: eruption of Irazu, Costa Rica in 1963; n,4 75: eruption of Ngauru-
hoe, New Zealand, in 1974 and 1975.

For pyroclastic flows the proposed classification is presented in Table 1

and our system is compared with other classifications in Table 2. For pyro-
elastic surges the proposed classification is presented in Table 3. A brief de-
scription of both pyroclastic flow and surge deposits is given in Table 4.
With regard to the classification of pyroclastic flows and their deposits
several points are worth further discussion:
(1) The scheme is based on a limited number of observations of actual
eruptions, which have invariably produced small volume pyroclastic flows
composed of relatively dense material. The mechanism of continuous column
collapse as envisaged by Sparks and Wilson (1976) and Sparks et al. (1978)
for the formation of ignimbrites has as of yet not been observed. Field evi-
dence in support of continuous collapse has however been described from an
ignimbrite in Mexico, with the recognition that an extremely coarse, lithic-
rich airfall deposit termed a co-ignimbrite Zag-falldeposit (Wright and,Walker,
1977) shows the same compositional zoning as the ignimbrite. This deposit
consists of pyroclasts that are too large and heavy to be carried away in the
pumice flows and is therefore thought to indicate the site of the eruptive
column collapse.
(2) Another type of deposit formed by the eruption of a pyroclastic flow
is the vitric air-fall ash deposit (co-ignimbrite ash-fall of Sparks and.Walker,
1977). This consists of the fine vitric ash lost during the eruption and trans-
port of a pyroclastic flow; this forms the upper part of the turbulent cloud
seen in the observed historic pyroclastic flows (for photographs see Moore
and Melson, 1969; Nairn and Self, 1978). Studies have shown that ignimbrites
and other types of pyroclastic flow deposits are crystal enriched due to loss
of this vitrjc component (Hay, 1 9; f&man, 1967; ker, 1972; Sparks
TABLE 2

Comparison of various classifications of PYroclastic flows

-
This paper: Murai (1961): Williams and McBir- Smith (1979);
based on eruptive mechanism and based on type eruptions ney (1979): based on volume of the deposits
&aracteristics of the deposits and characteristiciof based on site of and
the deposits type eruptions
VaBes-type huge-volume ash-flow tuffs (100-1000 km?
Pyrodastic flow Deposit
VTTS-type* VTTS-type intermediate-volume ash-flow
(large volume)
Krakatoa-type KrakatOa-tYPQ r tuffs (l-100 km3)
I I
/ I
Pumice flow ignimbrite; / I
I **
pumice and ash\
(small volume)
St. Vincent-type
Scotia flow scoria and ash St. Vincent-type
Semi-vesicular semi-vesiclllar intermediate-type Asama-type
andesite flow andesite and ash

Block and ash flow;’ block and ash Sakurajima-type

nu& ardente Fe&m-type mm&volume ash-flow tuffs (O.OOl-
1.0 kmsj
Block and ash;= block and ash Pelee-type
uue’e ardente

Block and ash? block and ash Merapi-type Merapi-type

m&e ardente
1
“Produced by collapse of vertical eruptive column accompanying lava/dome coBaPse.
2 Produced by explosive collapse of an actively growing lava flow or dome.
%roduced by gravitational collapse of an actively growing lava flow or dome.
*VTTS - Valley of Ten Thousand Smokes.
**overlap between boundary of in&‘mediate- and small-volume ash-flow tuffs required to compare Smith’s Classification with other Cla&fiCations is
shown by dashed line.
TABLE 3
E
Genetic classification of pyroclastic surges

Essential fragment Eruptive mechanism Type of surge Historic examples Comments

--
Vesiculated- -collapse of a phreatomag- -base surge - Taal, Philippines, 1965, Base surges result from the explosive
(non-vesiculated) matic eruption column 1966 (Moore et al., 1966; interaction of magmatic material and
Waters and Fisher, 1971); water and are consequently cool. They
Capelinhos, Faial, Azores, are often associated with maar volcan-
1957-1958 (Machado et oes and tuff rings. The historic ex-
al., 1962; Waters and amples described are both basaltic but
Fisher, 1971) older phonalitic base surges are known
from the Laacher See area, Germany
(Sehmincke et al., 1973) and rhyolitic
ones from the Minoan eruption of San-
torini (Bond and Sparks, 1976)
Vesiculated- -accompanying pyroclastie associated with examples Ground surge, although originally in-
(non-vesiculated) flows erupted by mechanisms of pyroclastic flows given troduced by Sparks and Walker (1973)
given in Table 1 in Table 1 to encompass all pyroclastic surges, is
\ ground’ surge here used to describe those surges found
Vesiculated also associated with air- Askja, Iceland 1875 at the base of pyroclastic flow deposits,
(non-vesiculated) fall deposits by collapse / (Sparks and Wright, 1979); as well as those produced without any
of an eruption column but Vesuvius A.D. 79 (B.S.& accompanying pyroclastic flow
without generation of Sparks, personal eommuni-
pyroclastic flow cation)

Vesiculated- ash-cloud surge Mt. Pelke, Martinique, Ash-cloud surges (Fisher, 1979) are
non-vesiculated 1902* (Fisher et al., the turbulent, low-density flows deriv-
1980) ed from the overriding gas-ash cloud
of pyroclastic flows. These may in
some cases become detached from the
parent pyroclastic flow and move in-
depently

*Ground surges were also produced by the 1902-03 eruptions of Mt. Pelge, e.g. August 30, 1902 (Fisher et al., 1980).
 325

and Walker, 1977). Vitric air-fall ash deposits may be very extensive and have
volumes comparable with those of the parent pyroclastic flow deposits
(Sparks and Walker, 1977).
(3) The most obvious surficial manifestation of a pyroclastic flow is the
overriding ash cloud. It was this feature that influenced LaCroix (1903) to
call the pyroclastic flows generated by Mt. Pelee in 1902 nuees ardentes. It is,
however, obvious from his work (LaCroix, 1904, p. 350) that he meant the
term to refer to the complete phenomenon, that is, to both the overriding ash
cloud and the basal avalanche or underflow. In a later publication LaCroix
(1930) expanded the use of the term to include all types of pyroclastic flows.
This use of the term n&e ardente for pyroclastic flows produced by different
mechanisms has persisted to the present day with the result that the exact
meaning of the term is now rather ambiguous. This has been especially so dur-
ing the last few years with the recognition that the ash cloud produces distinct
deposits (surge deposits) from the underflow, pyroclastic flow sensu stricto. In
view of these problems the authors feel that either the term should be avoided
altogether, the pyroclastic flows produced by “Pele’an-type” eruptions being
called block and ash flows (Perret, 1937); or nuee ardente should be restricted
to the original definition and only be used for those pyroclastic flows produced
by the collapse of an actively growing lava flow or dome (Rose et al., 1977;
Smith and Roobol, 1980). Both definitions are used in Tables 1 and 2.
(4) The deposits from pumice flows are termed ignimbrites or pumice and
ash deposits. Such deposits are found in oceanic islands (Iceland, Azores),
island arcs (Lesser Antilles), continental margins (Andes) and continental in-
teriors (Western United States). They may be subdivided on the basis of lateral
extent and volume into large-volume and small-volume ignimbrites (or pumice
and ash deposits). The former are restricted in their occurrence to continental
margins and interiors, and large islands (e.g. New Zealand); generally they re-
sult from eruptions of salic talc-alkaline magmas and tend to form large
ignimbrite sheets (Smith, 1979). The latter are characteristically developed on
island arc volcanoes and oceanic islands but can also be found in a continental
setting as well (Francis et al., 1974; Sparks, 1975; Wright, 1979). Thus modern
island arcs are characterized by small-volume ignimbrites while those
produced at the active margins and interiors of continents are often of
large volume and cover wide areas. This distinction should be borne in
mind in interpreting the volcano-tectonic setting of deposits from pumice
flows in ancient environments.
(5) Concerning the classification scheme of Williams and McBirney (1979,
and Table 2), the Valley of Ten Thousand Smokes type and the Valles type
are included as a category of pyroclastic flows discharged from fissures. Many
workers have suggested that ignimbrites are erupted from fissures or ring frac-
tures (Smith, 196Oa,b; Van Bemmelen, 1961; Rittman, 1962; Smith and
Bailey, 1968; Macdonald, 197~2);this type of eruption has been r~g~ded as
one way of accounting for the more voluminous deposits.
The fissure by~~t~~s~s originated from the o negation of lines of fum
326

TABLE 4

Summary descriptions of types of pyroclastic flow and surge deposits

Deposit Description
_
Ignimbrite; Unsorted ash deposits containing variable amounts of rounded salic
pumice and ash pumice lapilli and blocks up to 1 m in diameter. In flow units pum-
ice fragments can be reversely graded while the lithic clasts can show
normal grading; ungraded flow units are as common. A fine-grained
basal layer is found at the bottom of flow units. They sometimes
contain fossil fumarole pipes and carbonized wood. The coarser
smaller-volume deposits usually form valley infills while the larger-
volume deposits may form large ignimbrite sheets. Sometimes they
may show one or more zones of welding
Scoria and ash Topographically controlled, unsorted~ash deposits containing basalt
to andesite vesicular lapilli and scoriaceous ropey surfaced clasta up to
1 m in diameter. They may inssome circumstances contain Iarge non-
vesicular cognate lithic clasts. Fine-grained basal layers are found at
the bottom of flow units. Fossil fumarole pipes and carbonized wood
may also be present. The presence of levC?es,channels and steep
flow fronts indicate a high yield strength during transport of the mov-
ingpyroclastic flow
Semi-vesicular Topographically controlled unsorted ash deposits containing inter-
andesite and ash mediate vesicular (between pumice and non-vesicular juvenile clasts)
andesite lapilli, blocksBnd bombs. F~ine-gra:ned basal layers, fossil
fumarole pipes and carbonized wood all may be present
Block and ash Topographically controlled, unsorted ash deposits containing large,
generally non-vesicular, jointed, cognate lithic blocks which can ex-
ceed 5 m in diameter. The deposits are generally reversely graded.
Fine-grained basal layers are again present. Again they may contain
fossil fumarole pipes and carbonized wood. Surface manifestations
include the presence of levees, steep flow fronts and the presence of
large surface blocks, all of which again indicate a high yield strength
during transport of the flow
Base surge Stratified and laminated deposits containing juvenile vesiculated
fragments ranging from pumice to non-vesiculated cognate lithic
clasts, ash and crystals with occassional accessory lithics (larger bal-
listic ones may show bomb sags near-vent) and deposits produced in
some phreatic eruptions w_hich are composed totally of accessory
lithics. Juvenile fragments are usually less than 10 cm in diameter due
to the high fragmentation caused by the water/magma interaction.
Deposits show unidirectional bedforms. Generally they are associated
with maar volcanoes and tuff rings. When basaltic in composition
they are usually altered to palagonite
Ground surge Generally less than I m thick, composed of ash, juvenile vesiculated _
fragments, crystais and litbics in varying proportions depending on
constituents in the eruption column. Typically enriched in denser
corn~.onents (less well vesiculated juvenile fragments, crystals and
lithies) compared to accompanying pyroclast%floW. Again they
showyunidhec nal bedforms; eiirboni wood and small fumarole
pipes may be present
 327

TABLE 4 (continued)

Deposit Description

Ash-cloud surge Stratified deposits found at the top of and as lateral equivalents of
flow units of pyroclastic flows. They show unidirectional bedforms,
pinch and swell structures and may occur as descrete separated
lenses. Grain size and proportions of components depend on the
parent pyroclastic flow. Can contain small fumarole pipes
,/
in the Valley of Ten Thousand Smokes ignimbrite, 1912 eruption of Katmai.
This led Penner (1920) to conclude that the ignimbrite erupted from fissures
in the floor of the valley. More recently, Curtis (1968), from detailed mapping
of the closure of isopach maps of pyroclastic fall deposits produced in the
same eruption, located the central vent of Novarupta as the source. Now
Wright et al. (1979) have suggested that, at least initially, the Bandelier Tuffs
from Valles caldera (VaIles “type example” of Williams and McBirney, 1979)
were erupted from a central vent.
Korringa (1973) and Ekren and Byers (1976) have documented large welded
ignimbrites erupted from linear vent systems in Nevada. However, the field evi-
dence for such fissure eruptions is meagre in many cases. The authors believe
that ignimbrites are generally erupted from central vents.
(6) The group of rocks known variously as foam lavas, froth flows, tuff
lavas and tuffolavas (Vlodavetz, 1963; Shirinian, 1962,1963; McCall, 1964;
Cook, 1966; Macdonald, 1972) have been interpreted as intermediate between
lavas and pyroclastic flows (Panto, 1963) or as part of a gradational sequence
from lava to ignimbrite (McCall, 1964). According to M.R. Sheridan (personal
communication) the classic Russian examples from Armenia are ignimbrites,
while the so-called froth flows of Vulsini volcano (Locardi and Mittempergher,
1965,1967) can be interpreted as other types of pyroclastic rocks (Sparks,
1975). A study of rocks corresponding to published accounts of tuffolavas by
one of the authors on Pantelleria (Wright, 1979,198O) agrees with the view of
Sparks. In the opinion of the authors tuffolavas, etc., do not exist as a separate
category, but represent misinterpretations of other kinds of volcanic rock.

Lithological classification

The main bases of lithological classification are:

(1) The grain size limits of the pyroclasts and the overall size distribution
of the deposit.
(2) The constituent fragments of the deposit.
(3) The degree and type of welding.
Both (1) and (2) can be used to help discriminate the genesis of a partic
pyroclastic deposit in the older geological record.

Grain size. With regard to the grain size ~irnit~of ~~~~c~

system of Fisher (1961) is adopted (Table 5).
328

TABLE 5

Grain size limits for pyroclastic fragments after Fisher (1961)

-7
Grain size Pyroclastic fragments
(mm)

coarse blocks
256 and
fine bombs
64
lapilli
2
coarse
l/l6 ash

fine
T
/ .__ -~~_L_-

Concerning the overall grain size distribution, granulometric analysis of

non-welded and unlithified pyroclastic deposits can be an important dis-
criminant in deducing their mechanism of formation (Murai, 1961; Walker,
1971; Sheridan, 1971; Sparks, 1976). Pyroclastic flow deposits generally
show extremely poor sorting, while pyroclastic fall deposits are better sorted.
A convenient way of representing grain size data is by Md, /o@ plots (Fig. 5)“.
Using such a plot Walker (1971) was able to define separate field for the de-

O#

4 :

0
-8 -6 -4 -2 0 2 4 6 Md$
256 64 16 4 1
%a l/l6 l/64 Mdmm

Fig. 5. Mdo/cm plot used to show the fieldsof pyroclastic fall and flow fields after Walker
(1971). Dotted line iathe l%co&ur of thepyroclastic flowfield from Walker (1973).
Dashed line isthe 1% contour for the field of pyroclastic fall deposits modified by Sparks
and Wright (1979) after Walker (1973).

*Cumulative curves of the grain size distribution are drawn on arithmetic probability paper
and the Inman (1952) parameters of ian diameter Mdd, (= $J,,) and ff@ (= & - (P,,/2),
which represents the graphical stand eviation and a measure of sorting, are derived. For
discussion of the validity-of these stati cs when applied to p astic deposits see Sparks
(1976) and’Bond an&Sparks (1976).
 posits produced by pyroclastic flows and falls. Pyroclastic flow deposits usu-
ally have o@ values greater than 2.0, while in general pyroclastic fall deposits
have values less than 2.0. Based on the supposed dominance of ash sized par-
ticles in pyroclastic flows Smith (1960a,b) and Ross and Smith (1961) intro-
duced the term ash-flow tuff. This term has now become almost synonymous
with the older term ignimbrite. The authors feel that, despite some vagueness
of the original description of Marshall (1935), the definition of ignimbrite by
Marshall (p. 38, and reproduced in Chapin and Elston, 1979, p. 2) as applied
here, for deposits solely from pumice flows, is valid and its use should be con-
tinued, especially in view of the fact that in most small volume pumice flows
the modal grain size is in the lapilli or bomb size range (Smith and Roobol,
1980). It is worth pointing out here that ash-flow tuff has been used to de-
scribe block and ash-flow deposits (included in small volume ash-flow tuffs
of Smith, 1979, and Table 2) although these deposits only contain between
15 and 40% ash (A.L. Smith and M.J. Roobol, unpublished data). The authors
feel that the use of ash-flow tuff for such deposits is unteneable.
Md, /u$ plots of pyroclastic surge deposits tend to overlap both pyroclastic
flow and fall fields (Roobol and Smith, 1974; Sparks, 1976). Individual
laminae of deposits can be well sorted while channel samples through several
laminae are often more poorly sorted (see Bond and Sparks, 1976). Ground
surge deposits seem in general to be better sorted than base surge deposits
(Handler et al., 1980); this may be due to cohesion of water-saturated fine
ash to larger particles in the base surge cloud. Ash-cloud surge deposits formed
from block and ash flows of the 1902 eruption of Mt. Pelee can be extremely
coarse; some contain blocks up to 70 cm and are more poorly sorted than
many pyroclastic flow deposits (Fisher et al., 1980).
In certain welded and lithified rocks grain size analyses can be attempted
using thin sections and photographs. Such analysis can be of value in deciding
the mechanism of formation of deposit, while for ancient, pyroclastic rocks
even a qualitative assessment of the size distribution is of value.

Constituent fragments. A summary of the dominant components in a pyro-

elastic deposit provides a qualitative lithological description as well as provid-
ing some information as to the genesis (Table 6).

Welding. This post-depositional process involves the sintering together of hot

vesicular fragments and glass shards under a compactional load (Smith, 1960a,b;
Ross and Smith, 1961; Ragan and Sheridan, 1972; Sparks and Wright, 1979).
Welding has only generally been described from ignimbrites; no descriptions
in the literature are known to the authors of welding occurring in other
denser-clast pyroclastic flows, however certain semi-vesicular andesite and ash
deposits from Mt. Pelee, Martinique are welded (A.L. Smith and M.J. Roobol,
unpublished data). Welded air-fall Luffs are known in the literature and their
characteristics and the criteria for distinguishing them from welded ignim-
b~ites have been d~c~rn~~t~ by Spank and W ht (~9~~~ and Wriest (19
TABLE 6

Summary of the components in pyroclastic deposits

A. Pyroclastic flows and surges

Type of flow or surge Essential components Other components

vesicular non-vesicular
_l_py

pumice crystals Accessory and accidental lithics

scoria crystals ‘cognate, accessory and accidental
poor-moderate cognate lithics and accidental lithics
vesicular clasts crystals

3, Pyroclastic falls -

Type of fall Essential components* Other components

Grain size
(mm) vesicular non-vesicular

agglomerate pumicejscoria cognate and accessory lithics

64
breccia cognate and/or accessory lithics
lapilli deposit pumice/scoria cognate and/or accessory lithics crystals
.>2
<2 ash deposit pumice/scoria crystals and/or, cognate and/or
accessory lithics

*Depending on type of deposit.

 331

Features of these tuffs indicate post-emplacement compaction and welding

over a relatively wide area so that they cannot be simply ascribed to the flatten-
ing of spatter lumps on impact (see agglutinate in the glossary below). Welded
pyroclastic surge deposits are known to occur in the products of the 1875
eruption of Askja, Iceland (Sparks and Wright, 1979) from Mayor Island, New
Zealand (S. Self, unpublished data), and in the Bandelier Tuffs, New Mexico
(R.V. Fisher, S. Self and J.V. Wright, unpublished data).
Those ignimbrites in which welding is fully developed show three character-
istic zones of dense, partial and incipient, ana no welding (Smith, 1960b). In
the zones of dense welding glass shards and larger, commonly glassy, flattened
pumioe fragments (fiamme) define a planar foliation or eutaxitic texture. In-
cipiently welded tuffs are here distinguished from partially welded tuffs on
the basis that although coherent, they show no textural evidence of deforma-
tion of the constituent fragments. The term siblar (Fenner, 1948) is used here
for incipiently welded tuffs. Coherence in sillars is generally due either to the
sintermg of clasts at their points of contact, or to vapor phase crystallization
of minerals in pore spaces (see Sheridan, 1970) rather than to compactional
welding.
Ignimbrites which are totally non-welded or only a small proportion of
which is welded can be termed low-grade ignimbrites, in contrast to high-grade
ignimbrites which show well-developed zones of dense welding and in which
the proportion of welding is high.

GLOSSARY

In order to broaden the scope of this classification some terms introduced

earlier and a number of other terms commonly used to describe pyroclastic
deposits are described below:

Accretionary lapilli. Lapilli-sized pellets of ash, commonly exhibiting a concentric internal

structure which have been recorded from pyroclastic fall, base- and ground-surge deposits;
also from fossil fumarole pipes in pyroclastic flow deposits. They are believed to form by
the accretion of fine ash around a nucleus of water or some solid particle. This may occur
in the downwind eruption cloud, where a common mechanism of formation is by rain
flushing (Moore and Peck, 1962; Walker, 1971) or in the vertical eruption column of
phreatomagmatic eruptions (Self and Sparks, 19’78), or by gases (in fossil fumarole pipes)
streaming up through pyroclastic flow deposits (Walker, 1971).

Agglutinate. Coherent deposit of lava spatter and poorly vesiculated juvenile pyroclasts
associated with strombolian and hawaiian eruptions (Macdonald, 1972).

Fiamme. Flattened glassy juvenile clasts in welded pyroclastic deposits (Zavaritsky, 1947).
In most cases fiamme result from the deformation of original pumice clasts. However,
Gibson and Tazieff (1967), Schmincke and Swanson (1967) and Self (1973) have sug-
gested that some Eiamme represent the flattening of non-vesiculated juvenile clasts.

Fused tuff. At the contact with lava flows and dikes older pumiceous pyroelastic deposits
may become fused and resemble welded tuffs (Cbrist~a~se~ and ~~~rn~, 1966; Schmincl~e~
1967).
332

PaZagonite tuf7”.The original glass in deposits of basaltic phreatomagmatic eruptions (surtsey-

anand base surge) commonly alters by hydration to palagonite (Hay and Iijima, 1963).
Such deposits may be lithified and could be confused with welded tuffs.

Rheomorphic welded tuff. Rheomorphism (Ylow change”) of a welded tuff is the second-
ary mass flowage of the tuff as a coherent viscous fluid. Rheomorphic welded tuffs are
characterized by stretched fiamme which define a lineation parallel to the flowage direc-
tion. Flowage is capable of producing folding and other internal flow structures similar to
those found in salic lava extrusions. The term rheoignimbrite was first introduced by
Rittmann (1958) to describe welded ignimbrites showing secondary flowage features. Such
features in the ignimbrites of Gran Canaria were thought by Schmincke and Swanson
(1967) to indicate primary laminar flow of the pyroclastic flows. However, Wolff and
Wright (1981) reinterpret these as secondary flow features and document rheomorphism in
both welded ignimbrites and welded air-fall tuffa.

Tephra. Thorarinsson (1954) proposed the term tephra as a collective term for all pyro-
elastic material transported from the vent through the air. In this original definition the
term strickly applied only to: pyroclastic fall deposits, however it was later re-defined
(Thorarinsson, 1974) to include both pyroclastic fall and flow material.

Tuff. This is used here loosely as a collective term for all consolidated pyroclastic rocks.

Vitrophyre. This term is used to describe a densely welded tuff which in handspecimen
has a glassy appearance.

Volcanic mudflows (lahars). These consist of volcanic debris mobilized by water. They are
often generated during eruptions, as in the May 1902 eruptions of Mt. Pelee, Martinique
and Soufrihe, St. Vincent (Roobol and Smith, 1975), but can also”be formed without any
associated volcanic activity. They are a common feature of all recent large volcanoes and
are sometimes hard to distinguish in the field from pyroclastic flow deposits. Excellent de-
scriptions are given of lahars by Neal1 (1974) and from Mount Rainier by Crandell(1971).

ACKNOWLEDGEMENTS

For financial support in our studies of pyroclastic rocks we would like to

thank the National Science Foundation (A.L.S., grants EAR 73-00194 and
77-17064), NASA (SS., grant NSG 5145), the Natural Environment Research
Council of Great Britain (S.S. and J.V.W.), a University of Wellington post-
doctoral Fellowship (S.S.), the Lindemann Trust Fellowship (J.V.W.), the
American Philosophic@ Society (A.L.S.) and the University of Puerto Rico
(A.L.S. and J.V.W.).
We would like tomthank the following people for thought provoking discus-
sions, criticism and encouragement during the preparation of this manuscript
over the past two and a half years B. Booth, B.J. Clough, R.V. Fisher, E.H.
Francis, 0. Smarason, M.J. Roobol, R.S.J. Sparks, G.P.L. Walker, C.J.N.
Wilson and J.A. Wolff. R.S.J. Sparks gave a careful and throughtful review of an
earlier d&t of the -~~uscript whi&tvas also review& by G. Green, C.J.N.
Wilson and J.A. Wolff.
 333

REFERENCES

Anderson, T. and Flett, J.S., 1903. Report on the eruptions of the Soufriere in St. Vincent
in 1902, and on a visit to Montagne Pelee in Martinique, 1. Philos. Trans. R. SOC.
London, Ser. A, 200: 353-553.
Aramaki, S., 1956. The 1783 activity of Asama volcano, Part 1. Jpn. J. Geol. Geogr., 27:
189-229
Aramaki, S., 1957. The 1783 activity of Asama volcano, Part 2. Jpn. J. Geol. Geogr., 28:
11-33.
Aramaki, Sand Yamasaki, M., 1963. Pyroclastic flows in Japan. Bull. Volcanol., 26:
89-99.
Blackburn, E., Wilson, L. and Sparks, R.S.J., 1976. Mechanisms and dynamics of stromboli-
an &tivity. J. Geol. Sot. London, 132: 429-440.
Bond, A_ and Sparks, R.&J., 1976. The Minoan eruption of Santorini, Greece. J. Geol.
SDC. London, 132: l-16.
Chapin, C.E. and Elston, W.E., 1979. Introduction. In: C.E. Chapin and WE. El&on
(Editors), Ash-flow Tuffs. Geol. Sot. Am., Spec. Paper, 180: 1-4.
Christiansen, R.L. and Lipman, P. W., 1966. EmpIacement and thermal history of a rhyolite
lava flow near Fortymile Canyon, Southern Nevada. Geol. Sot. Am. Bull., 77: 671-
684.
Cook, E.F:(Editor), 1966. Tufflavas and Ignimbrites. Elsevier, New York, N.Y., 212 pp.
Crandell, D.R., 1971. Postglacial lahars from Mount Rainier volcano, Washington. U.S.
Geol. Surv., Prof. Paper, 667: 75 pp.
Crowe, B.M. and Fisher, R.V., 1973. Sedimentary structures in base-surge deposits with
‘special reference to crossbedding, Ubehebe Craters, Death Valley, California. Geol.
Sot. Am. Bull., 84: 663-682.
Curtis, G.H., 1968. The stratigraphy of the ejecta from the 1912 eruption of Mount
Katmai and Novarupta, Alaska. In: R.R. Coats, R.L. Hay and C.A. Anderson (Editors),
Studies in Volcanology (Howell Williams Volume). Geol. Sot. Am. Mem., 116: 153-
210.
Eaton, G.P., 1964. Windborne volcanic ash: a possible index to polar wandering. J. Geol.,
72: l-35.
Ekren, E.B. and Byers, F.J., 1976. Ash-flow fissure vent in west-central Nevada. Geology,
4: 247-251.
Fenner, C.N., 1920. The Katmai region, Alaska, and the great eruption of 1912. J. Geol.,
28: 569-606.
Fenner, C.N., 1948. Incandescent tuff flows in southern Peru. Geol. Sot. Am. Bull., 59:
879--893.
Fisher, R.V., 1961. Proposed classification of volcanoclastic sediments and rocks. Geol.
Sot. Am. Bull., 72: 1409-1414.
Fisher, R.V., 1966. Rocks composed of volcanic fragments and their classification. Earth-
Sci. Rev., 1: 287-298.
Fisher, R.V., 1979. Models for pyroclastic surges and pyroclastic flows. J. Volcanol.
Geotherm. Res., 6: 305-318.
Fisher, R.V. and Waters, A.C., 1970. Base surge bed forms in maar volcanoes. Am. J. Sci.,
268: 157-180.
Fisher, R.V., Smith, A.L. and Roobol, M.J., 1980. Mt. Pelbe, Martinique: 1902 pyroclastic
flows. Abstr., IX Caribbean Geol. Conf., Santo Domingo (in press).
Francis, P.W., Roobol, M.J., Walker, G.P.L., Cobbold, P.R. and Coward, M.P., 1974. The
San Pedro and San Pablo volcanoes of Northern Chile and their hot avalanehe deposits.
Geol. Rundsch., 63: 357-388.
Gibson, I.L. and Tazieff, H., 1967. Additional theory on the origin of fiamme in ignim-
brites. Nature, 216: 1473-1474.
Gorshkov, G.S.,1959. Gigantic ~~pt~a~ of the volcano e~yrnia~n~~ Bull. Volcanol,, 20:
77-112.
334

Handler, A., Smith, A.L., Wright, J.V., Roobol, M.J., Self, S. and Fisher, R.V., 1980.
Textural characteristics of pyroclastic surge deposits. Trans. IX Caribbean Geol. Conf.,
Santo~Domingo (in press). ”
Hay, RL,, 1959. Formation of the crystal-rich glowing avalanche deposits of St. Vincent,
B.W.I. J. Geol., 67: 540-562.
Hay, &E-and- Iijima3 A., 1968. Nature and origin of palagonite tuffs of the Honolulu
group on Oahu, Hawaii. In: R.R. Coasts, R.L. Hay and CA. Anderson (Editors}, Studies
in Volcanology (Howell Williams Volume), Geol. Sot. Am. Mem., 116: 331-376.
He&en, G.H., 1974. An Atlas of Volcanic Ash. Smithsonian Contrib. Earth Sci., 12: 100 pp.
Inman; D.L., 1952. Measures for describing the size distribution of sediments. J. Sediment.
~P&Ol., 22: 125-145.
Knox, J.B. and Short, N.M., 1964. A diagnostic model using ash-fall data to determine
eruptian characteristics and atmospheric considerations during a major volcanic event.
BuBVolcanol., 27: 5-24.
Korringa, M.K., 1973. Linear vent area of the Soldier Meadow Tuff, an ash-flow sheet in
Northwestern Nevada. Geol. Soc_~Am. Bull,, 84: 3849-3866.
LaCroix, A., 1903. Sur les principaux r&Bats de la mission de la Martinique. C.R. Acad.
Sci. Paris, 135: 87.1-876.
LaCroix, A., 1904. La Montagne Pelee et Ses Eruptions. Masson et Cie, Paris, 662 pp.
LaCroix, A.,, 1930. Remarques sur les materiaux de projection des volcans et sur la genese
des roches pyroclastiques qu’ils constituent. Sot. Geol. Fr., Livre Jubliaire Centenaire,
2: 431-472,
Lipman, P.W., 1967. Mineral and chemical variations within as ash-flow sheet from Aso
Caldera, south western Japan. Contrib. Mineral. Petrol., 16: 300-327.
Locardi, E. and Mittempergher, M., 1965. Study of an uncommon lava sheet in the Bolsena
District (Central Italy). Bull. Volcanol., 28: 7 5-84.
Locardi, E. and Mittempergher, M., 1967. On the genesis of ignimbrites. How ignimbrites
aud other pyroclastic products originate from a flowing melt. Bull. Volcanol., 31:
131-152.
Macdonald, G.A,, 1972. Volcanoes. Prentice-Hall, Englewood Cliffs, N.J., 510 pp.
Macdonald, G.A. and Alcaraz, A., 1956. N&es ardentes of the 1948-1953 eruption of
Hibok-Hibok. Bull. Volcanol., 18: 169-1’78. J
Machado, F., Parsons, W.H., Richards, A.E. and Mulford, J.W., 1962. Capelinhos eruption
of Fayal volcano, Azores, 1957-58. J. Geophys. Res., 67: 3519-3329.
Marshall, I’., 1935. Acid racks of the Taupe-Rotorua Volcanic District. Trans. R. Sot.
N.Z., 64: l-44.
McCall, G.J.H., 1964; Froth flows in Kenya. Geol. Rundsch., 54: 1148-1195.
Moore, J.G. and Melson, W.G., 1969. Nuees ardentes of the 1968 eruption of Mayon
volcano, Philippines. Bull. Volcanol., 33: 600-620.
Moore, J;G. and Peck, D.L., 1962. Accreticn~ary lapilli in volcanic rocks of the western con-
tine&al thzited States. 6, Geol., 70: 182-193.
Moore, J.G., Nakamura, K. and Aicaraz, A., 1966. The 1965 eruption of Taal volcano.
Science, 155: 955-960.
Murai, I., 1961, A study of the textural characteristics of pyroclastic flow deposits in
Japan. Bull. Earthq. Res. Inst., 39: 133-254.
Nairn, LA. and Self, S., 1978. Explosive eruptions and pyroclastic avalanches from
Ngauruhoe in February 1975. J. Yolcanol. Geotherm. Res., 3: 39-60.
Neall, V.E,, 1974.~Laliars - Global occurrence, origins and annotated bibliography. Publ.
Victoria j&&T Wellington, 5: I8 pp.
Panto, G,, 1963, Ignimbrites of Hungary with regard to their genetics and classification.
Bull. V&cam& 2-5: 175-181.
Perr& F, A., 1EY7i The eruption of M Pelee 1929-1932. Car ie Inst. Washington
Publ:, 45% 126pp.
Ragan, R.M. and Sheridan, M.F., 1972. Compaction of the Bishop Tuff, California. Geol.
Sot. Am. Bull., 83: 95-106.
Rittman, A., 1958. Cenni sulle colate di ignimbriti. Catania Acad. Gioenia Sci. Nat. Boll.,
Ser. 4. 4: 524-533.
 335

Rittman, A., 1962. Erklarungsversuch zum Mechanismus der Ignimbritausbruche. Geol.

Rundsch., 52: 853-861.
Roobol, M.J. and Smith, A.L., 1974. Mt. Pelee, Martinique. In: Proceedings, I.A.V.
Symposium on Andean and Antarctic Volcanology (Santiago, Chili), pp. 190-202.
Roobol, M.J. and Smith, A.L., 1975. A comparison of the recent eruptions of Mt. Pelee,~
Martinique and Soufriere, St. Vincent. Bull. Volcanol., 39: l-28.
Roobol, M.J. and Smith, A.L., 1976. Mount Pelbe, Martinique: a pattern of alternating
eruptive styles. Geology, 4: 521-524.
Rose, W.I., Jr., Pearson, T. and Bonis, S., 1977. Nuee ardente eruption from the foot of a
dacite @a flow, Santiaguito Volcano, Guatemala: Bull Volcanol., 40: l-16.
Ross, C.S. -and Smith, R-L., 1961. Ash-flow tuffs, their origin, geologic relations and
idenpfication. U.S. Geol. Surv. Prof. Paper, 366: 81 pp.
Schmincke, H.U., 1967. Fused tuff and peperites in south-central Washington. Geol. Sot.
Am. Bull., 78: 319-330.
Schmincke, H.U. and Swanson, D.A., 1967. Laminar viscous flowage structures in ash-flow
tuffs from Gran Canaria, Canary Islands. J. Geol., 7 5: 641-664.
Schmincke, H.U., Fisher, R.V. and Waters, A.C., 1973. Antidune and chute and pool
structures in base surge deposits of the Laacher See area, Germany. Sedimentology, 20 :
553-574.
Self, S., 1973. Recent volcanism on Terceira, Azores. Ph.D. Thesis, Imperial College,
University of London.
Self, S. and Sparks, R.S.J., 1978. Characteristics of pyroclastic deposits formed by the
interaction of silicic magma and water. Bull. Volcanol., 41: 196-212.
Self, S., Wilson, L. and Nairn, I.A., 1979. Vulcanian eruption mechanisms. Nature, 277:
440-443.
Shaw, D., Watkins, N.D. and Huang, T.C., 1974. Atmospherically transported volcanic
glass in deep-sea sediments: theoretical considerations. J. Geophys. Res., 79: 3081-
3094.
Sheridan, M.R., 1970. Fumarolic mounds and ridges of the Bishop Tuff, California. Geol.
Sot. Am. Bull., 31: 851-868.
Sheridan, M.R., 1971. Particle size characteristics of pyroclastic tuffs. J. Geophys. Res.,
76: 5627-5634.
Shirinian, K.G., 1962. Volcanic tuffs and tuff lavas in Armenia. Bull. Volcanol., 24: 171-
175.
Shirinian, K.G., 1963. Ignimbrites and tuffolavas. Bull. Volcanol., 25: 19-25.
Smith, A.L. and Roobol, M.J., 1980. Andesite volcanoes and their products, (b) andesite
pyroclastic flows. In: R.S. Thorpe (Editor), Orogenic Andesite and Related Rocks.
John Wiley and Sons, New York, N.Y. (in press).
Smith, R.L., 1960a. Ash-flows. Geol. Sot. Am. Bulf., 71: 795-842.
Smith, R.L., 1960b. Zones and zonal variations in welded ash-flows. U.S. Geol. Surv.
Prof. Paper, 354-F: 149-159.
Smith, R.L., 1979. Ash-flow magmatism. In: C.E. Chapin and W.E. Elston (Editors),
Ash-Flow Tuffs. Geol. Sot. Am., Spec. Paper, 180: 5-27.
Smith, R.L. and Bailey, R.A., 1968. Resurgent cauldrons. In: R.R. Coats, R.L. Hay and
CA. Anderson (Editors), Studies in Volcanology (Howell Williams Volume). Geol.
Sot. Am. Mem., 116: 153-210.
Sparks, R.S.J., 1975. Stratigraphy and geology of the ignimbrites of Vulsini volcano,
central Italy. Geol. Rundsch., 64: 497-523.
Sparks, R.S.J., 1976. Grain size variations in ignimbrites and implications for the transport
of pyroclastic flows. Sedimentology, 3.3: 147-188.
Sparks, R.S.J. and Walker, G.P.L., 1973. The ground surge deposit: a third type of pyro-
elastic rock. Nature Phys. Sci., 241: 62-64.
Sparks, R.S.J. and WaIka~~G.P.L., 1977. The significance of v~tr~c-a~~cbed air-fall ashes
associated with crystal-enriched ignimbritea. 3. Voleanol. Geotherm. Res., 2: 329-341.
Sparks, R.S.J. and Wilson, L., 1976. A model for the formation of ignimbrite by gravita-
tional column coBapse. J. Geol. Sot. London, 132: 441-462.
336

Sparks, R.S.J. and Wright, J.V., 1979. Welded air-fall tuffs. In: C.E. Chapin and W.E.
El&on (Editors), Ash-Flow Tuffs. Geol. Sot. Am. Spec. Paper, 180: 155-166.
Sparks, R.S.J., Self, S. and Walker, G.P.L., 1973. Products of ignimbrite eruptions.
Geology, 1: 115-118.
Sparks, RS.J_Wilson, L. and Hulme, G., 1978. Theoretical modelling of the generation,
movement and emplacement of pyroclastic flows by column collapse. J; Geophys. Res.,
85B4: 17_27:1739.
Stoiber, R.E. and Rose, WI., Jr., 1969. Recent volcanic and fumarolic activity at Santiaguito
volcano, Guatemala. Bull. Volcanol., 33 : 47 5-502.
Thorarinsson, S., 1954, The tephra fall from Hekla on March 29th, 1947. Nat. Islandica II, 3,2.
Thorarinsson, S., 1974. The terms tephra and tephrochronology. In: J.A. Westgate and
C.M. Gold (Editors). World Bibliography and Index of Quaternary Tephro-chronology.
University of Alberta Press, Edmonton, Alta., pp. xvii-xviii.
Van Bemmelen, R.W., 1949. The geology of Indonesia, IV. General Geology of Indonesia
andAdjacent Archipelago. Government Printing Office, The Hague, 732 pp.
Van Bemmelen, R.W., &96l,Volcanology and geology of ignimbrites in Indonesia, North
Italy and U.S.A. Geol. Mijnb,, 40: 399-411.
Vlodavetz, V.I., 1963. Sur la genese des tufolaves 1 Kamchatka (a l’est du volcan Maly
Sbmiatchik). Bull:Volcanol., 25: 27-30.
Walker, .G.P.L_, 1971. Grain_size.characteristics~of pyroclastic deposits. J. Geol., 7 9: 696-
7 f4.
Walker, G.P.L., 1972, Crystal concentration in ignimbrites. Contrib. Mineral. Petrol., 36:
135-146.
Walker, G.P.L., 1973. Explosive volcanic eruptions - a new classification scheme. Geol.
Rundsch., 62: 431-446.
Walker, G.P.L., 1980. The Taupo Pumice: product of the most powerful known (ultra-
plinian) eruption? J. Volcanol. Geotherm, Res. (in press).
Walker, G.P.L. and Croasdale, R., 1971. Two Plinian-type eruptions in the Azores. J.
Geol. Sot. London, 127: 17-56.
Waters, A.C, and Fisher, R.V., 1971. Base surges and their deposits: Capelinhos and Taal
volcanoes. J. Geophys. Res., 76: 5596-5614.
Williams, H. and McBirney, A.R., 1979. Volcanology. Freeman, Cooper and Co., San
Francisco, Califl, 397 pp,,
Wilson, C.J.N., 1980. The role of fluidization in the emplacement of pyroclastic flows: an
experimental approach. J. Volcanol. Geotherm. Res., 8: 231-249.
Wilson, L,, 1976, Explosive volcanic eruptions, III. Plinian eruption columns. Geophys.
J.R. Astron. Sot., 45: 543-556.
Wilson, L., Sparks, R.S.J., Huang, T.C. and Watkins, N.D., 1978. The control of volcanic
column heights by eruption energetics and dynamics. J. Geophys. Res. 83, B4: ‘1829-
1836.
Wohletz, K.H. and Sheridan, M.F., 1979. A model of pyroclastic surge. In: C.E. Chapin
and W.E.-Elston (Editors), Ash-Flow Tuffs. Geol, Sot. Am., Spec. Paper, 180: 177-
194.
Wolff, 3.8 and Wright, J.V., 1981. Rheomorphism of welded tuffs. J. Volcanol. Geotherm.
Res. (in press).
Wright, J.V., 1979. Formation, transport and deposition of ignimbrites and welded tuffs.
Ph.D. Thesis, Imperial College, University of London, 46X pp.
Wright, J.V., 1980. Stratigraphy and geology of the welded air-fall tuffs of Pantelleria,
Italy. Geol. Rundseh., .69: 263-291.
Wright, J-V. and WnIk_er, G&L., 1977. The ignimbrite source problem: Significance of a
co-ignimbrite lag-falldeposit. _Geology* 5; 729-732.
Wrigh&_J.V,and - ~~Walker G: P_IL *>_~~
_-___1__ 1980
_ uption, transport and
acase Study’~from Mexico. J, Volt _Ge~otherm, Res. (in p
Wright, J.V., Self, S, and Fisher, R.V., 197’9. Eruption sequence of the Bandeller Tuffs,
Jemez Volcano, New Mexico. EOS, 61: 66 (abstract).
Zavaritsky, A.N., 1947. The ignimbrites of Armenia. Akad. Nauk S.S.S.R. Izv. Ser. Geol.,
3: 3-18.