ALLUVIAL SAPPHIRES AND DIAMONDS

OF THE NEW ENGLAND GEM FIELDS,

NEW SOUTH WALES,
AUSTRALIA

by

Robert Raymond Coenraads

Honours Degree of Bachelor of Arts, Macquarie University, Sydney, 1978.
Master of Science, University of British Columbia, Vancouver, 1982.
Diploma in Gemmology, Gemmological Association of Australia, 1990.

A Thesis Submitted in Fulfillment

of the Requirements for the Degree of
Doctor of Philosophy

School of Earth Sciences

Macquarie University

June, 1991
 TABLE OF CONTENTS
Table of Contents ......................................................................................................... ii
Summary ...................................................................................................................... vi
Certificate ..................................................................................................................... vii
Acknowledgements ...................................................................................................... viii
List of Colour Plates..................................................................................................... ix
List of Plates ................................................................................................................. x
List of Figures .............................................................................................................. xi
List of Tables................................................................................................................ xvii
Chapter 1
Sapphire and Diamond Exploration in the New England Gem Fields, New
South Wales, Australia. ................................................................................................ 1
1.1 Aims .......................................................................................................... 1
1.2 Introduction ............................................................................................... 3
1.2.1 Corundum deposits of metamorphic association. ....................... 3
1.2.2 Corundum deposits of igneous association ................................ 3
1.2.3 Australian sapphire deposits ....................................................... 5
1.2.4 Volcanism in the Central Province ............................................. 5
1.3 The alluvial sapphire and diamond mining industry in eastern
Australia. .......................................................................................................... 7
1.3.1 Development of the industry in Australia .................................. 7
1.3.2 Sapphire mining and processing. ................................................ 7
1.3.3 Sapphire exploration methodology............................................. 9
1.3.4 Sapphire sorting and grading ...................................................... 11
1.3.5 Gemstone processing .................................................................. 11
1.3.6 Gemstone enhancement .............................................................. 12
Chapter 2
The Nature and Origin of Corundum associated with Volcanic Provinces ................. 16
2.1 Introduction ............................................................................................... 16
2.2 Properties of corundum ............................................................................. 16
2.3 Form of sapphires from the Central Volcanic Province ............................ 17
2.4 Surface features of rubies and sapphires from volcanic provinces ........... 20
2.4.1 Negative crystal impressions ...................................................... 20
2.4.2 Surface etch features ................................................................... 24
2.4.3 Healed fracture surfaces and damage resulting from
transport in an alluvial environment ..................................................... 30
2.4.4 Discussion ................................................................................... 30
2.5 Inclusions in sapphires and zircons from the Central Volcanic
Province ............................................................................................................ 31
 2.6 U-Pb dating of the zircon inclusions in Central Volcanic
Province sapphires ............................................................................................ 39
2.7 Sapphire-bearing rocks in the Central Province volcanic
sequence ........................................................................................................... 45
2.8 Conclusions ............................................................................................... 46
Chapter 3
Key Areas for Alluvial Diamond and Sapphire Exploration in the New
England Gemfields, New South Wales ........................................................................ 47
3.1 Introduction ............................................................................................... 47
3.2 Central Province volcanism....................................................................... 47
3.2.1 Structural control of Central Province volcanism ...................... 49
3.2.2 Timing of Central Province volcanism....................................... 49
3.3 Sapphire and diamond occurrences in the Central Volcanic
Province. ........................................................................................................... 50
3.3.1 Quaternary alluvium. .................................................................. 50
3.3.2 Tertiary deep leads. .................................................................... 50
3.4 Identification of exploration targets for diamond and sapphire ................ 52
3.4.1 Analysis of sub-volcanic topography ......................................... 52
3.5 Geologic and geomorphic evolution of the Central Volcanic
Province. ........................................................................................................... 58
Chapter 4
Delineation of Alluvial and Deep Lead Exploration Targets: Case Studies. ............... 60
4.1 Introduction ............................................................................................... 60
4.2 Diamond prospectivity in the palaeo-Gwydir River system ..................... 60
4.3 Kings Plains and Reddestone type alluvial sapphire deposits................... 63
4.4 Braemar type deep-lead sapphire deposits ................................................ 68
4.4.1 Sapphire-bearing palaeodrainage systems near Braemar ........... 80
4.4.2 Extent of the Eocene/Oligocene sapphire-bearing
sediments below basalt near Braemar .................................................. 82
4.4.3 Implications of the Braemar deposit in the search for
similar deposits in the zone-of-overlap between the East and
West Central Provinces ........................................................................ 84
4.4.4 Conclusions and exploration-mining problems .......................... 85
Chapter 5
Heavy Mineral Suites within the Central Province, New South Wales ....................... 88
5.1 Introduction ............................................................................................... 88
5.2 Preparation of heavy mineral concentrates ............................................... 88
5.3 Heavy minerals present in the Central Volcanic Province. ....................... 90
5.4 Comparison between heavy mineral suites from collection sites
within the Central Volcanic Province. ............................................................. 92
 5.5 Ilmenite mantled rutile crystals from the Uralla district ........................... 97
5.5.1 Occurrence. ................................................................................. 97
5.5.2 Description of the ilmenite-rutile crystals. ................................. 97
5.5.3 Comparisons with ilmenite-rutile crystals and ilmenite
found elsewhere .................................................................................... 101
5.6 Garnets from Horse Gully in the New England gemfields. ...................... 107
5.6.1 Occurrence .................................................................................. 107
5.6.2 Features and properties of the gem garmets ............................... 107
5.7 Implications for genesis of heavy mineral placer deposits........................ 111
5.8 Conclusions and interpretation. ................................................................. 112
Chapter 6
Evaluation of Sapphire Source Rocks and Potential Source Structures in the
Central Volcanic Province. .......................................................................................... 115
6.1 Introduction ............................................................................................... 115
6.2 Evaluation of the natural lagoons of the Central Volcanic
Province -Are they sapphire-producing maars ? .............................................. 115
6.3 Detailed work at Kings Plains and Dunvegan lagoons ............................. 119
6.4 Sapphire-source rock evaluation in the Central Volcanic
Province ............................................................................................................ 132
6.4.1. Watershed analysis of the Central Volcanic Province. .............. 133
6.4.2 Definition of specific catchment areas for sapphire
exploration. ........................................................................................... 133
6.4.3 Source rock evaluation program ................................................. 133
6.4.4 Chemical features of the Mount Buckley lavas .......................... 143
6.5 Comparison between volcanics of the sapphire-bearing East and
sapphire-barren West Central Province ............................................................ 148
6.6 Comparison between the Central Province and other volcanic
provinces in northeastern Australia and southeast Asia ................................... 149
6.7 Conclusions ............................................................................................... 149
Chapter 7
Summary and Conclusions ........................................................................................... 151
7.1 Origin of corundum associated with volcanic provinces .......................... 151
7.1.1 Aims and achievment of aims .................................................... 151
7.1.2 Inclusion studies ......................................................................... 151
7.1.3 Surface features .......................................................................... 152
7.2 Division of the New England gem fields into three exploration
regions .............................................................................................................. 153
7.2.1 Diamond-bearing deep leads of the West Central
Volcanic Province................................................................................. 153
 7.2.2 Holocene alluvial sapphire deposits of the East Central
Volcanic Province................................................................................. 154
7.2.3 Braemar-type deep lead deposits in the zone-of-overlap ........... 154
7.3 Heavy mineral suites in the New England gem fields............................... 155
7.3.1 Sapphire associates of the East Central Volcanic
Province ................................................................................................ 155
7.3.2 Ilmenite-mantled rutile crystals of the Uralla area ..................... 155
7.3.3 Pyrope-almandine garnets from Horse Gully ............................. 155
7.4 Investigation of sapphire source rocks and potential source
structures in the Central Volcanic Province ..................................................... 156
References ................................................................................................................... 158
Appendix 1
Pre-Volcanic Tertiary Landscape and Drainage of the New England Region,
New South Wales ......................................................................................................... 173
Appendix 2
Description of Water Bores drilled in the New England Region, New South
Wales ............................................................................................................................ 225
Appendix 3
Geophysical Assessment .............................................................................................. 230
Appendix 4
Drilling Programme...................................................................................................... 260
Appendix 5
Mt Buckley Basalt Flows; Thin Section Descriptions ................................................. 271
Appendix 6
Material published during Doctor of Philosophy Candidature .................................... 274
Summary

The Central Volcanic Province in northern New South Wales, also known as the
New England Gem Field, is one of Australias’ most important sapphire producing areas.
Large-scale mechanized mining and restoration techniques are the most effective means of
recovering this resource from beneath valuable farm land.
The alluvial sapphires are shown to be derived from the 32-38 million year old
alkaline basaltic volcanic rocks comprising the eastern portion of the Central Province via
uranium-lead dating of zircon inclusions within the sapphires. Corundum, zircon and other
exotic minerals crystallized in coarse aggregates when high proportions of incompatible
elements and volatiles were present in early melt fractions. These unusual crystallization
products were subsequently transported to the surface by voluminous basaltic magmas.
Surface dissolution or etch features on the corundum occurred in the melt environment en
route to the surface.
In the East Central Province the post-eruptive fluvial history was vital in
controlling the concentration of economic sapphire deposits from lower grade source
rocks. The sapphire-bearing placer deposits occur as "shoestring" type accumulations
occupying channels within basalt-filled valleys. The highest grades generally correspond
to areas where channels are deepest. Minimal evidence of abrasion due to fluvial transport,
coupled with the spatial variability in the character of the corundum, and the differences in
proportions of heavy minerals in placer deposits, collectively indicate minimal amount of
fluvial transport, reworking or mixing. Within these low energy systems the heavy
minerals have moved vertically downwards and become concentrated with time. Such
observations indicate that the corundum is derived from local multiple sources.
Highly prospective diamond-bearing "deep leads" associated with the palaeo-
Gwydir River have been traced under the 19-23 million year old basalts of the West
Central Province via reconstruction of the pre-volcanic topography. The
palaeotopographic reconstruction indicates at least 3 separate systems of diamond bearing
alluvials and therefore, local, multiple sources for the diamonds.
The Central Volcanic Province has been divided into three prospective alluvial
target regions. These comprise a western region with Tertiary deep-lead diamond deposits,
a central belt with Eocene-Oligocene deep-lead sapphire deposits, and an eastern division
with Holocene sapphire deposits.
 CERTIFICATE

I certify that the work presented in this thesis is original unless

otherwise referenced and has not been submitted for a higher degree at any
other university or institution.

. . . . . . . . . . . . .

Robert R. Coenraads
Acknowledgements

I would like to thank Mr. Tom Nunan of T.J & P.V. Nunan Pty Ltd, Mr. Bruce
Hyman of Jingellic Minerals N.L., Mr. Peter Kennewell of Cluff Resources Pacific
Limited, and Mr. Colin and Mrs. Joan Rynne of Braemar for allowing publication of
information concerning their exploration and mining areas. Dr. Steve Webster of the
Department of Minerals and Energy and Mr. Zoltan Beldi of Geo-Instruments Pty Ltd are
thanked for generously supplying the geophysical instrumentation, Prof. Don Emerson for
the geophysical measurements of the drill core and Dr. Stephen Bannister, Dr. Graeme
Chapman and Miss Sue Froud for their assistance with the geophysical field programme.
Funding for the geophysical and drilling progromme was provided by T.J & P.V. Nunan
Pty. Ltd. and Jingellic Minerals N.L. Sapphires, rubies and other heavy minerals for
inclusion and surface feature studies were generously supplied by T.J. & P.V. Nunan Pty
Ltd, Colin and Joan Rynne, John MacCormack, S.A.P. Mining Co, Bo Ploi and Nakorn
Gems, Thailand; garnets from Horse Gully were provided by John Joris of Joris Gemstone
Traders Pty. Ltd, Sydney and faceted by Henk van der Graaf; and ilmenite mantled rutile
crystals were provided by Simon Paige. I would like to thank the Australian Museum for
providing funds for the ion probe analyses, and Dr W. Compston, Dr F. L. Sutherland and
Dr P. D. Kinny for their interest and support of this project. I would also like to thank Mr.
Jim Stroud and his team at the Department of Minerals and Energy, Armidale Office, for
providing the 1:25,000 scale geologic maps that were essential for this study. Many
individuals have contributed generously to this project by offering suggestions, advice and
critical reviews of manuscripts. It is therefore a pleasure to thank Dr Jane Barron, Dr.
Larry Barron, Dr. Bill Birch, Mr. Bob Brown, Dr. Graeme Chapman, Mr Terry Coldham,
Dr. Peter Flood, Prof. Brenda Franklin, Dr. A.L. Jacques, Mr. David Jones, Dr. John Lusk,
Prof. Cliff Ollier, Mr. Simon Pecover, Mr. Ross Pogson, Dr. A.J. Stoltz, Dr. F. Lin
Sutherland, Mr. Pongsak Vichit, Prof. J.F.G. Wilkinson, and Prof. Ken Williams. John
Coenraads' patient assistance with manuscript typing and fieldwork, and the assistance of
Mr David Barnes, Dr. Günther Bischoff, Dr. Ron Oldfield, Mr Jack Thompson with
photography are gratefully acknowledged.
I would like especially to thank Simon Pecover for his initial encouragment to
undertake a research project in sapphires, John Lusk for supervising my project, and Lin
Sutherland and Günther Bischoff for their untiring support during the project. Their
continuing enthusiasm, inspiration and friendship during the project has been greatly
apprecciated.
My sincere appreciation is extended to many individuals and families who offered
hospitality and friendship of during my fieldwork in the New England Area. Special
mention is due to Tony and Lucia Boomgaard, Bruce Mair, Tom and Pat Nunan, Cliff and
Jeannetta Ollier, Alan and Elizabeth Uebergang, Joyce Willows, Ralph and Pam Paige,
Col and Joan Rynne, and Len and Lou Stewart.
LIST OF COLOUR PLATES

Colour Plate 1. Partially resorbed sapphire "dog tooth" crystal from the New England gem
fields, New South Wales................................................................... xviii

Colour Plate 2. Heavily included sapphire "barrel" crystal from the New England gem
fields, New South Wales................................................................... xviii

Colour Plate 3. "Dog tooth" and "barrel" sapphire crystals from the New England gem
fields, New South Wales................................................................... xix

Colour Plate 4. Inclusions in sapphire from the New England gem fields.............. xix

Colour Plate 5. Landsat Image No. 095-081 (Australian Centre for Remote Sensing)
showing the Maybole volcanic centre in the New England gemfields, New South
Wales, together with the lineament interpretation................................................. xx

Colour Plate 6. Parcel of "fancy" sapphires from the New England gem fields, New South
Wales collected by T.J. & P.V. Nunan Pty. Ltd. over several years.................. xx

Colour Plate 7. Quarry at Braemar (circa 1987). Sapphire-bearing fluvio-lacustrine

sediments are exposed in the southwest face below a protective capping of 23 million
year old basalt................................................................................. xxi

Colour Plate 8. Highly ferruginous volcanic rock intruding hydraulically fractured

Devonian-Carboniferous metasiltstone at Braemar in the New England gemfields, New
South Wales................................................................................... xxi

Colour Plate 9. Small sapphire recovery plant of Mr. Col Rynne at Braemar, New South
Wales........................................................................................... xxii

Colour Plate 10. Sapphire recovery plant of T.J. & P.V. Nunan Pty. Ltd. at Reddestone
Creek in the New England gemfields, New South Wales................................ xxii

Colour Plate 11. Alluvial plains of Back Plains Creek in the New England gemfields, New
South Wales, showing the layout of a well designed exploration programme of T.J. &
P.V. Nunan Pty. Ltd......................................................................... xxiii

Colour Plate 12. Sapphire mining operation of T.J. & P.V. Nunan Pty. Ltd. on the Kings
Plains Creek Eastern Feeder, low angle aerial photograph, looking west..............xxiii
LIST OF PLATES

Plate 1. 1. Negative crystal socket in blue sapphire. 2. Negative crystal in blue-purple

sapphire. 3. Negative crystal socket in purple ruby, all from the Ruby Well Mine,
Chanthaburi-Trat. 4. Sapphire from Reddestone Creek, No. 2 plant, New England. Flat
topped triangular hillocks and detail showing relief of hillocks and their irregular,
eroded apprearance, note damage to top edge of the hillock due to fluvial
transport.......... 23

Plate 2. 1. Blue sapphire from Reddestone Creek, No. 1 plant, New England, surface
perpendicular to the c-axis, showing oriented flat, shallow hexagonal etch depressions
with central hillocks. 2. Blue sapphire from SAP mine, Bo Ploi, hexagonal etch
depressions with a central hillock. 3. Pink-purple ruby, Ruby Well mine, Chanthaburi-
Trat, showing early stages of formation of triangular and hexagonal hillocks. 4. Deep
prominent etch sculpturing on a pink-purple ruby, Ruby Well mine, Chanthaburi-Trat,
showing detail of etch channels............................................................. 25

Plate 3. Blue sapphires from 1. New England gem fields. "Shark-skin like" etch channels.
2. SAP mine, Bo Ploi. "Intaglio-like" surface due to extensive etching showing
structural control of "intaglio" pattern and microlamination due to succesive layer
dissolution. 3. Braemar, New England. Etch pattern resembling trigonal prismatic
"stacked bricks" on a surface oriented at a low angle to the c-axis. 4. Braemar, New
England. Triangular pyramidal depressions resulting from etching of a surface at an
intermediate angle to the c-
axis........................................................................................... 26

Plate 4. Blue sapphires from Braemar, New England. 1. Etch surface at a low angle to the
c-axis resembling trigonal prismatic "bricks". 2. Chatter marks or sub-parallel curved
grooves attributed to mechanical stress. 3. Sapphire traversed by numerous fractures. 4.
Fracture radiating from an inclusion in sapphire.......................................... 27

Plate 5. Blue sapphires 1. Ruby Well mine, Chanthaburi-Trat showing surface covered
with triangular hillocks. Detail of triangular hillocks, some chipped due to fluvial
damage. 2. Reddestone Creek No. 1 plant; specimen showing a "bladed" surface
texture. Crystallographic control as the raised "blades" form two sides of triangular
pyramidal shaped pit. 3. SAP mine, Bo Ploi. Rebroken healed fracture surfaces reveal a
"fingerprint-like" pattern of exposed, formerly fluid filled, tubes. 4. Reddestone Creek,
No. 1 plant, New England. Conchoidal fracturing on the edge of this sapphire is
evidence of damage in the fluvial environment. 5. Ruby Well mine, Chanthaburi-Trat.
Protruding hillock broken off at its base leaving a clean fracture surface is evidence of
damage in the fluvial environment.......................................................................... 28
LIST OF FIGURES

Figure 1-1. Location of the Central Province in northeastern New South Wales. Inset
shows Mesozoic-Cenozoic volcanic provinces which form a discontinuous band within
or adjacent to the eastern Australian highlands............................................ 2

Figure 1-2. Operations flowchart for a typical large scale sapphire mine................. 10

Figure 2-1. Ruby and sapphire crystal forms belonging to the rhombohedral division of
the trigonal system................................................................................ 18

Figure 2-2. Sapphire-anorthoclase xenolith from basalt, Mt Leura, Hoy Province,

Queensland..................................................................................... 19

Figure 2-3. Sapphire crystal from the Central Province, N.S.W., showing a negative
crystal shape in its faces...................................................................... 21

Figure 2-4. Sapphire crystal from the Central Province, N.S.W., showing uneven chemical
resorption...................................................................................... 21

Figure 2-5. Scenario explaining uneven chemical resorption such as seen on the sapphire
crystal shown in Figure 4.................................................................... 22

Figure 2-6. Compositional ranges of feldspar inclusions in sapphires and zircons from the
Central Volcanic Province.................................................................... 33

Figure 2-7. Backscatter photographs of a euhedral feldspar inclusion in a sapphire from

Braemar........................................................................................ 34

Figure 2-8. Spinel inclusions in New England sapphires plotted within the spinel family
prisms.......................................................................................... 35

Figure 2-9. Gahnospinel (sample nos. 1-17) and hercynite (nos. 18-24) inclusions in
sapphire from the Central Volcanic Province.............................................. 36

Figure 2-10. Niobium-rutile (sample nos. 1-3) and columbite (nos. 4-13) inclusions in
sapphire from the Central Volcanic Province.............................................. 37

Figure 2-11. Melt inclusions in sapphire (nos. 1-15) and zircon (18-24) from the Central
Volcanic Province............................................................................ 38
Figure 2-12. Euhedral zircon crystals included in sapphire, transmitted light
photomicrographs............................................................................ 40

Figure 2-13. Zircon inclusion 28/11, showing the sputter-hole excavated by the ion-
beam............................................................................................ 42

Figure 3-1. Structural elements of the Central Province, New South Wales, determined by
landsat lineament analysis.................................................................... 48

Figure 3-2. Drainage pattern of the Central Province and location of heavy mineral
collection sites listed in Table 5-1........................................................... 51

Figure 3-3. Illustration of the Technique by which sub-basaltic topography is

determined..................................................................................... 53

Figure 3-4. Present-day and palaeodrainage patterns in the Central Province, New South
Wales........................................................................................... 54

Figure 3-5. Progressive stages in the development of relief inversion by lateral stream
formation....................................................................................... 56

Figure 3-6. Stages in the development of stream piracy illustrated by two examples in the
Central Province............................................................................... 57

Figure 4-1. Central Province, New South Wales divided into broad zones indicating
suitability for sapphire or diamond exploration............................................ 61

Figure 4-2. Diamond-bearing palaeodrainage and distribution of historic mines of the

Copeton Field.................................................................................. 62

Figure 4-3. Economic Holocene sapphire deposits of the Kings Plains

palaeochannel.................................................................................. 64

Figure 4-4. Economic Holocene sapphire deposits of the Reddestone Channel.......... 65

Figure 4-5. The development of economic sapphire deposits through stream

concentration................................................................................... 67

Figure 4-6. Reddestone Creek sapphire deposit and sapphire concentrations in river
gravels.......................................................................................... 69
Figure 4-7. Kings Plains Creek sapphire deposits and sapphire concentrations in river
gravels.......................................................................................... 69

Figure 4-8. Location of the Braemar sapphire deposit in the zone-of-overlap between the
East and West Central Provinces............................................................ 71

Figure 4-9. Location of the Braemar sapphire deposit in relation to the geology of the
Elsmore-Swan Brook area................................................................... 72

Figure 4-10. Reconstruction of the surface geology at Braemar........................... 73

Figure 4-11. Stratigraphy of the old quarry (south-west face) at Braemar (circa 1986).74

Figure 4-12. Stratigraphic and intrusive relationships evident in costeans A to D at

Braemar........................................................................................ 75

Figure 4-13. Stratigraphic and intrusive relationships evident in costeans E to J at

Braemar........................................................................................ 76

Figure 4-14. Breccia dyke intruding Devonian-Carboniferous metasiltstone, exposed in

costean B at Braemar......................................................................... 77

Figure 4-15. Distribution of Eocene/Oligocene deep lead sapphire deposits and Holocene
sapphire-bearing alluvium at Braemar...................................................... 78

Figure 4-16. Proposed extent of Eocene/Oligocene sapphire-bearing sediments at Braemar

beneath a capping of 23.2 Ma. basalt ...................................................... 79

Figure 4-17. Positions of the Braemar and Swan Brook palaeochannels as determined by
basement palaeotopographic analysis on the Elsmore 1:25,000 sheet.................. 81

Figure 4-18. Cross sections, AB and CD, of the Braemar palaeochannel looking
north............................................................................................ 83

Figure 4-19. Inversion of the palaeo-Swan Brook by lateral streams, Swan Brook and
Kings Creek and potential deep lead sapphire deposits................................... 86

Figure 5-1. Central Province spinel plotted within the spinel family prisms.............. 91

Figure 5-2. Central Province chromium-spinel compositions.............................. 93

Figure 5-3. Central Province pleonaste compositions....................................... 94

Figure 5-4. Central Province titanium-magnetite compositions............................ 95

Figure 5-5. Central Province ilmenite compositions......................................... 96

Figure 5-6. Geology of the Uralla district, New SouthWales and location of the ilmenite
coated rutile crystals.......................................................................... 98

Figure 5-7. Ilmenite mantled rutile crystals from the Rocky River Goldfield near
Uralla........................................................................................... 99

Figure 5-8. Transmitted light photomicrographs of sections cut through ilmenite mantled
rutile crystals.................................................................................. 100

Figure 5-9. Summary of microprobe results across 4 ilmenite mantled rutile crystals cut
perpendicular to the C-axis.................................................................. 102

Figure 5-10. Summary of the main elemental oxide ranges in the Uralla ilmenite
compared to those in literature sources................................................................. 105

Figure 5-11. Location of Horse Gully in the New England Gemfields, New South
Wales........................................................................................... 108

Figure 5-12. Relationship between refractive index, specific gravity and end member
composition for the Horse Gully Garnets.................................................. 110

Figure 6-1. Location of the natural lagoons of the Central Province, N.S.W........... 116

Figure 6-2. Kings Plains Lagoon, aerial photograph....................................... 117

Figure 6-3. Dunvegan Lagoon, aerial photograph.......................................... 118

Figure 6-4. Clarevaulx and Dunvegan Lagoons, aerial photograph...................... 118

Figure 6-5. John Ryall Lagoon, aerial photograph.......................................... 118

Figure 6-6. Barley Field Lagoon, aerial photograph......................................... 118

Figure 6-7. Basement topography in the vicinity of Kings Plains Lagoon................ 118
Figure 6-8. Basement topography in the vicinity of Clarevaulx and Dunvegan
Lagoons........................................................................................ 121

Figure 6-9. Basement topography in the vicinity of John Ryall Lagoon.................. 122

Figure 6-10. Basement topography in the vicinity of Barley Field Lagoon............... 123

Figure 6-11. Basement topography in the vicinity of Inverell Racecourse................ 124

Figure 6-12. Cross sections through; A; Kings Plains Lagoon. B; Dunvegan Lagoon. C;
Clarevaulx Lagoon. D; John Ryall Lagoon. E; Barley Field Lagoon. F; Sheep Station
Creek in the vicinity of Inverell Racecourse............................................... 125

Figure 6-13. Kings Plains Lagoon geophysical survey area, elevation................... 127

Figure 6-14. Dunvegan Lagoon geophysical survey area, elevation....................... 128

Figure 6-15. Model explaining the development of the lagoons in the Central
Province........................................................................................ 130

Figure 6-16. Surface configuration of Kings Plains, Dunvegan, Clarevaulx, John Ryall
and Barley Field Lagoons.................................................................... 131

Figure 6-17. Watershed analysis for the Central Volcanic Province in northeastern New
South Wales................................................................................. 134

Figure 6-18. Central Province watershed analysis and location of Mt Buckley study
area............................................................................................. 135

Figure 6-19. Location of basalt sample localities at Mount Buckley....................... 136

Figure 6-20. Total alkalis versus silica plot for the Central Province and Mt Buckley
analyses........................................................................................ 141

Figure 6-21. C.I.P.W. normative plagioclase versus differentiation inde for the Central
Province and Mt Buckley analyses......................................................... 141

Figure 6-22. C.I.P.W. nepheline or hypersthene versus normative plagioclase for the
Central Province and Mt Buckley basalt analyses........................................ 142
Figure 6-23. C.I.P.W. nepheline or hypersthene versus normative plagioclase for the Mt
Buckley basalt analyses...................................................................... 143

Figure 6-24. Na20 versus K2O plot for the Central Volcanic Province and Mt Buckley
basalt analyses................................................................................. 144

Figure 6-25. C.I.P.W. nepheline or hypersthene versus flow number for Mt Buckley
flows............................................................................................ 145

Figure 6-26. Major elements versus flow numberm at Mt Buckley........................ 146

Figure 6-27. Trace elements versus flow number at Mt Buckley.......................... 147

Figure 6-28. Total alkalis versus silica plot, compositional fields of the Central Volcanic
Province, northeastern Australian corundum-bearing provinces and the corundum-
bearing and corundum-less southeast Asian volcanic provinces........................ 144
LIST OF TABLES

Table 1-1. Age data available for the Central Province...................................... 6

Table 1-2. History of the east Australian sapphire and diamond industry................. 8

Table 2-1. Properties of Corundum............................................................ 13

Table 2-2. Inclusion suite for Central Volcanic Province Sapphires....................... 32

Table 2-3. Inclusion suite for Central Volcanic Province Zircons.......................... 32

Table 2-4. U-Pb ages of zircon inclusions in sapphire...................................... 43

Table 5-1. Heavy mineral suites from soil and placer collection sites within the Central
Province............................................................................................ 89

Table 5-2. Weight percent oxide content for the Central Province heavy mineral
suite................................................................................................. 89

Table 5-3. Main elemental oxide ranges in the Uralla ilmenite compared to those in
literature sources......................................................................................... 103

Table 5-4. Composition and gemmological properties of the Horse Gully Garnets..... 109

Table 6-1. Location, chemical analyses and CIPW norms for the Mt Buckley basalts.. 139
 Chapter 1

SAPPHIRE AND DIAMOND EXPLORATION IN THE NEW ENGLAND

GEM FIELDS, NEW SOUTH WALES, AUSTRALIA.

1.1 Aims

The principal aims of this study are to examine the occurrence and origin of
Australian sapphires which are found associated with Tertiary Central Volcanic Province
of eastern Australia (the New England Gemfields), (Figure 1-1). This province was chosen
for study because it is Australia's principal producer of sapphire.
Sapphires occurring in the Central Province are principally extracted from
unconsolidated Quaternary alluvial placer deposits. Sapphires and diamonds also occur in
basalt covered Tertiary deep lead placer deposits. The successful exploration for new
deposits, together with the effective mining of existing deposits requires an integrated
understanding of geological, geomorphic and hydrologic factors which contributed to the
concentration of the sapphires from their host rocks into economically workable alluvial
deposits. The present study therefore attempts to explain the distribution of Quaternary
alluvial sapphire, Tertiary deep lead concentrations of diamond and sapphire, and
associated heavy minerals in the Central Province in terms of these factors. The
Reddestone Creek and Kings Plains sapphire-bearing placers; the Braemar deep-lead
sapphire deposit, and the Copeton deep-lead diamond deposits of the New England
Gemfields are used as case studies with this aim in mind. The model developed for the
Central Volcanic Province and its exploration implications may be used in other volcanic
provinces to assist in the development of a systematic exploration programme for
commercial sapphire concentrations.
A study of the characteristics of the sapphires themselves, such as their surface
morphology and inclusion suite, aims to positively link their formation with that of the
associated Central Volcanic Province.
The following chapters will deal with the characteristics of sapphires of volcanic
origin, as well as aspects of the Central Volcanic Province, including controls on it's
emplacement, timing of emplacement, the volcanic rocks which make up the province,
eruptive mechanisms, the pre-emplacement surface, pre and post volcanic drainage
systems and heavy mineral assemblages associated with the province.
1.2 Introduction

Gem quality corundum occurs in both igneous and metamorphic rocks, however, it
is most commonly found and mined from concentrations in alluvial systems draining these
source terrains. Corundum is concentrated by the processes of weathering and erosion into
alluvial deposits by virtue of its high specific gravity. It is not broken down owing to its
high hardness and lack of cleavage and occurs together with other resistant, heavy
minerals such as zircon, garnet, spinel and ilmenite.

1.2.1 Corundum deposits of metamorphic association.

Corundum may occur as an accessory mineral in metamorphic rocks, such as

metamorphosed limestones and dolomites, mica schists, gneisses and amphibolites.
Gemfields associated with metamorphic terrains include Mogok (worlds finest ruby ) in
Burma (Iyer, 1953; Webster, 1975), Hunza valley (ruby) in Pakistan, Ratnapura and
Elahera (many colours and star stones) in Sri Lanka (Munasinghe & Dissanayake, 1981;
Zwaan, 1982; Gunawardene & Rupasinghe, 1986; Heilmann & Henn, 1986), Umba Valley
and in the western Uluguru mountains near Morogoro (all colours) in Tanzania (Hanni,
1987), the Chimwadzulu hill in southern Malawi (Webster, 1975), Jagdalak (ruby) in
Afganistan (Webster, 1975), and India (star ruby).

1.2.2 Corundum deposits of igneous association

A large number of important sapphire and ruby gemfields throughout the world are
associated with igneous rocks. These are predominantly volcanic terrains of alkaline type,
and also pegmatites.
Gem fields associated with alkaline basaltic volcanism include New England
(sapphire) and Anakie (worlds best green and golden yellow sapphire) in Australia
(MacNevin, 1972; Broughton, 1979; Coenraads, 1990); Pailin (sapphire) in Cambodia
(Jobbins and Berange, 1981); Chanthaburi-Trat (ruby and sapphire), Denchai, Bo Ploi
(sapphire) and Khorat Plateau in Thailand (Vichit et al, 1979; Barr and MacDonald, 1981;
Keller, 1982; Yaemniyom, 1982; Gunawardene and Chawla, 1984; Vitchit, 1987); Bokeo
Plateau, Xuan Loc Plateau, Cardamones Massif, Solovens Plateau and Kassens Plateau in
Kampuchea (Lacombe, 1969-70); Haut Chalong Plateau, Pleiku Plateau, Darlac Plateau
and Djiring Plateau in South Vietnam; Kouang Tcheoci Wan, Fujian Province and Hainan
Island in Southern China (Keller and Keller, 1986; Keller and Wang, 1986); Mercaderes
Rio Mayo area in Colombia (Keller et al, 1985); Gimi Valley near Jemaa and Kaduna
Province in Nigeria (Wright, 1969; Kiefert and Schmetzer, 1987). Sapphires may be found
in matrix in basic dykes of alkaline affinity at Yogo Gulch near Utica, Montana, U.S.A.
(Brownlow and Komorowski, 1988), and at Loch Roag on the Isle of Lewis, Scotland
(Jackson, 1984).
 Gemfields associated with pegmatites include Soomjam Village in the Pader
District, Kashmir (Webster, 1975; Atkinson and Kothavala, 1983) and Barauta, Zimbabwe
(Sweeney, 1971).

Striking similarities between the gemfields of volcanic association can be seen.

(a) They are associated with extrusive basaltic lava fields or eroded remnants, or also
laterites.

(b) In most cases, crater lakes, cones, cone remnants or plugs can be found.

(c) Alkaline basalt types appear to be always present.

(d) Ultramafic xenoliths (predominantly mantle lherzolites) are commonly reported.

(e) In a number of the fields, older basalt flows are tholeiitic with younger flows being
increasingly alkaline.

(f) The corundum and associated megacrysts are corroded implying disequilibrium
with the carrier magma.

(g) In-situ corundum megacrysts are rarely seen within the basalts, apart from rare
mafic dykes (such as Loch Roag, Jackson, 1984).

(h) Different areas within a gem-producing terrain may yield different coloured stones,
with characteristics changing in distances under 10 kilometres; this implies
multiple sources.

(i) Crystal inclusions found in sapphire include; apatite, columbite, rutile, pyrrhotite,
boehmite, zircon, hercynite (Fe spinel), gahnospinel (Fe, Zn spinel), almandine,
pyrope, hotchetelite (U, Ti pyrochlore), alkali-feldspar and plagioclase.

(j) Sapphires are characteristically associated with zircon, spinel and ilmenite in placer
deposits. Occasionally they may also be associated with olivine, clinopyroxene,
garnet, magnetite or feldspar.
1.2.3 Australian sapphire deposits

Sapphires have long been recognized in areas of basaltic volcanism (Dunstan,

1902), being found, in varying quantity and quality, in streams and rivers draining many
Cenozoic-Mesozoic "lava field" provinces in eastern Australia, (Spencer,1983). The two
lava field provinces which contain known economic concentrations of sapphire are the
Central Province in northern New South Wales (the New England gem field) and the Hoy
Province in central Queensland (the Anakie gem field). Recent exploration suggests that
the other provinces in central Queensland (the Lava Plains gem field) also contain
potential deposits (S.R. Pecover, pers comm.).

1.2.4 Volcanism in the Central Province

The Central Volcanic Province, in northeastern New South Wales is one of a

number of Mesozoic-Cenozoic intraplate volcanic provinces which form a discontinuous
belt stretching over a distance of 4400 kilometers and up to 300 kilometers wide, within or
adjacent to the Eastern Australian Highlands (Johnson, 1989). The locations of east
Australian volcanic provinces are shown in Figure 1-1.
The Central Volcanic Province, comprising Tertiary basaltic volcanics, intrusives
and minor sediments, locally overlies and intrudes Devonian to Triassic volcanics,
metasediments and plutonics of the Woolomin-Texas Block of the New England Fold Belt
(Leitch, 1974). It extends from Armidale in the south, through Glen Innes and Inverell, to
the Queensland border in the north (Figure 1-1).
The basaltic rocks of the Central Province yield potassium-argon ages of 19 to 38
Ma (Cooper et al, 1963; McDougall & Wilkinson, 1967; Wellman & McDougall, 1974;
Smith, 1988; C.D. Ollier, pers. comm. 1989; Coenraads et al, 1990; Sutherland & Raynor,
unpubl. data) and from 2 to 49 Ma based on reset zircon fission track ages (Hollis &
Sutherland, 1985; Sutherland & Hollis, unpubl. data). The ages available for the Central
Province are shown on Figure 1-1 and are listed in Table 1-1.
The lavas are predominantly alkaline to strongly alkaline. They include alkali
olivine basalts, basanites, hawaiites and nepheline hawaiites (Wilkinson, 1962; Wilkinson,
1966; Binns, 1969; Binns et al, 1970; Duggan, 1972; Wilkinson, 1973; Wilkinson &
Duggan, 1973; Street, 1974; McKay, 1975; McQueen, 1975; Barron, 1987; Coenraads, in
prep). Tholeiites occur in the vicinity of Inverell (Duggan, 1972; Wilkinson & Duggan,
1973). Breccias and volcaniclastic rocks have been reported by various workers (
Lishmund & Oakes, 1983; Sutherland, 1985; Temby, 1986 and Barron, 1987) and occur
throughout the Inverell-Glen Innes region at or near the base of the volcanic pile (Brown
& Pecover, 1986a,b; Pecover & Brown, 1986; Brown, 1987; Pecover, 1987; Pecover and
Coenraads, 1989). Some of the volcaniclastic rocks are known to contain sapphire
(Pecover and Coenraads, 1989).
 The structural control and timing of Central Province volcanism has been described
by Coenraads (1988, 1990) and the province may be divided into western and eastern parts
on this basis. Sapphires are found in variable quantities in almost every gully, creek and
river draining the East Central Province and some of the alluvial deposits are currently
being mined by large scale, mechanized operations (Nunan, 1989).

1.3 The alluvial sapphire and diamond mining industry in eastern Australia.

1.3.1 Development of the industry in Australia

The existing gem industry in Australia has developed mainly around sapphire,
opal, some imported diamond and other semi-precious stones. The first reported
occurrence of sapphires was in 1853 when the Reverend W.B. Clarke found sapphire
associated with alluvial tin in the vicinity of Inverell (MacNevin, 1972). Early sapphire
mining took place in the Anakie and New England fields from the 1890's to the 1920's
with most of the production going to Czsarist Russia and Germany. It was not until the
1960's, however, that commercial mechanized mining began to take place on a large scale
due to an increase in demand for rough sapphires throughout Asia.
Diamonds were first recovered as a result of gold or tin mining operations in
Tertiary deep leads, or in recent alluvial deposits derived from them. The Bingara and
Copeton diamond fields were exploited from 1872 and 1883 respectively, however there
has been little production since 1922. The total production recorded from these fields is
202,000 carats, although actual production probably exceeded this amount (MacNevin,
1977). The history and development of the sapphire and diamond industry in eastern
Australia is summarized in Table 1-2.
In recent years Australia has been a major producer of sapphire, supplying between
65 and 70 percent of the total world sapphire production (Spencer 1983). The New
England field contributes in excess of 50 percent of Australia's sapphire production which,
although not well documented, is probably in excess of 5 million carats annually (T.J.
Nunan Pty Ltd, unpublished data, 1989).

1.3.2 Sapphire mining and processing.

The types of operation found today vary from single prospectors, working in creeks
and rivers with picks, shovels and sieves, through to large mechanized operations
employing about 30 people and capable of processing 60 cubic yards of unconsolidated
alluvium per hour. Although the prospector's pick and shovel are being replaced by
bulldozers, backhoes and small processing plants (see Colour Plate 9), the principle of
gemstone separation is essentially the same; the gemstone-bearing alluvial gravels or
"wash" is excavated, cleaned, sized and the heavy minerals removed by gravity separation.
The larger mechanized operations, such as those run by T.J. & P.V. Nunan Pty Ltd,
(Nunan, 1989), are by far the most significant contributors of sapphire to the market place.
A typical large-scale operational procedure is described below and illustrated in the
flowchart, Figure 1-2.
Mining large areas of alluvium is carried out via overburden stripping and an
advancing trench (Colour Plate 12). A backhoe digs the trench and loads the loose alluvial
gravels into trucks for transport to the processing plant (Colour Plate 10). The trench is
backfilled and restored as the mining operation advances. The gravels are dumped into a
feed box at the upper end of the plant and, with the assistance of high pressure water jets,
are gravity fed into huge rotating mesh cylinders called trommels. The trommels size the
material by allowing it to pass over progressively coarser meshes. The sized factions pass
onto a series of pulsating jigs which effectively trap heavy minerals of a certain size in
trays or "hutches" which are emptied by hand once or twice a day. The pulsating action
causes liquefaction of the gravel allowing the heavy grains to sink rapidly. The undersize
"heavies" pass through a screen at the base of the hutches and the "lights" are washed over
the top. A log washer may be included in the circuit if the alluvial material is clay-rich and
therefore difficult to break down. Recovery rates of sapphire are of the order of 90% or
better (T.J. Nunan Pty Ltd, unpublished data, 1989).

1.3.3 Sapphire exploration methodology

Bulk sampling of a drainage system is commonly used to determine sapphire

grades in the alluvium. Sapphire is usually associated with other heavy minerals such as
spinel, zircon and ilmenite which yeild a visible "blacksand". This is usually a good sign
of potential economic grades. Exploration drilling is performed on a grid pattern (see
Colour Plate 11) using a large diameter drill rig (diameter approximately 1 metre) which
allows a reasonably large sample of the alluvial gravels to be processed. In a thorough
exploration programme at least 0.5 to 1 loose cubic yards of sample are processed to yield
a data point at each sample locality (J. McPhee pers. comm., 1987). The advantages of this
method are that the depth and location of the ancient stream are determined together with
the thickness and type of alluvial gravel and its expected potential in grams of sapphire per
cubic metre. The colour, quality, size, etc. of the sapphire can also be assessed. In essence,
a well planned exploration drilling programme will indicate the value of the ground and
the expected returns prior to mining. Economic grades are dependant on a number of
variables, including; current sapphire prices; type and efficiency of plant; quality, quantity
and size of sapphire in the wash and; depth, thickness and lateral extent of the wash. Under
economic conditions prevalent in 1988, material with grades as low as 5 grams per loose
cubic yard could be worked profitably by a highly efficient mechanized operation (T.
Nunan pers. comm. 1988).
When the data from the alluvial sampling have been plotted on a map, the highest
sapphire grades commonly conform reasonably to an ancient drainage axis. The
exploration drilling data for Reddestone Creek and Kings Plains Creek, discussed in
Chapter 6, illustrate this point. High grades are followed upstream until an economic
cutoff is reached, that is, beyond which fluvial concentration has been insufficient to
create economic grades. High grades are followed downstream, away from the source
region, until dilution of the gem gravels also renders the grades unworkable.
Systematic exploration for sapphires has only recently been applied in Australia
due to the small size of the companies and partnerships involved in the industry.
Furthermore, there had been little input of geological or geophysical expertise in sapphire
exploration, prior to the directive issued by Department of Minerals and Energy at their
seminar on Tertiary volcanics and sapphires in the New England district in May, 1987.
It had been recognized that exploration should be directed at defining
sapphire/ruby bearing drainage systems associated with Cenozoic lava field provinces
(Spencer 1983). What was poorly understood, however, is why only certain areas are rich
enough to be worked economically, and why, within those areas, certain streams carry
large concentrations whereas others are uneconomic.

1.3.4 Sapphire sorting and grading

The heavy mineral concentrate from the processing plant consists of spinel,
ilmenite, zircon and corundum together with ironstone pisolites. This is dried and passed
through magnetic separators which remove the magnetic ironstone, spinel and ilmenite.
The non-magnetic residue is passed onto a team of skilled sorters and graders. Sorting and
grading is normally carried out on mirrors using natural lighting. Mirrors reflect incoming
light back through the stone allowing the rapid assesment of colour, clarity and the
presence of flaws, without the need to pick up each stone. Zircons are removed, as are
parti-coloured and other coloured sapphires (called "fancies", see Colour Plate 6). The
remaining blue sapphires are graded into 3 or 4 grades (depending on the requirements of
the buyer), with the sapphires in the highest grade being the best blue colour, transparent
and free of flaws and inclusions.

1.3.5 Gemstone processing

Thailand is the major processor of all coloured gemstones and recent Thai statistics
indicated that between 70 and 80 percent of all sapphires imported into Thailand
originated from Australia (PA Technology, 1987). Australian sapphire processing
technology uses machinery to set faceting angles, whereas Thailand sapphire faceting is
done "by eye". The major advantage of southeast Asian processing is the large quantity of
low-cost, efficient labour. Comparing average 1987 Thai and Australian working rates and
salaries, the cost of cutting a stone in Australia is 16 to 18 times greater (PA technology
1987). This labour force is particularly important for the sorting and heat treatment of
large numbers of gemstones.
1.3.6 Gemstone enhancement

It is an accepted practice to improve the quality of a sapphire or ruby permanently

through heat treatment. Thailand has accumulated extensive practical experience in the
heat treatment of sapphire and over 95% of rubies and sapphires exported from Asia are
heat treated. Treatments include the removal or precipitation of fine rutile needles (silk)
and the enhancement or reduction of the intensity of the blue colour by influencing the
oxidation state of the trace amount of iron present. The furnace conditions necessary to
achieve improvement in stones with various "problems" are outlined in Table 2-1 (Nassau,
1984).
Processes to remove fine epigenetic rutile inclusions or "silk" via rapid cooling, or
oxidative heating to lighten blue colour are used by the Thais, mainly on Australian and
Thai stones.
 Chapter 2

THE NATURE AND ORIGIN OF CORUNDUM ASSOCIATED WITH

VOLCANIC PROVINCES

2.1 Introduction

One aim of this study was to determine the origin of corundum associated with
volcanic provinces ("volcanic corundum") based on the surface morphology of the
corundum crystals and the nature of their inclusions. Some 255 inclusion bearing
corundums were collected from streams and rivers draining the Central Volcanic Province,
New South Wales (The New England Gemfields) to provide data for the project. A key
result was the age relationship between the corundums and associated volcanic rocks.
The reported failure of experimental attempts to grow corundum from a corundum-
bearing basaltic composition, (Green et al. 1978), and more significantly, the abundance
of incompatible elements such as U, Th, Zr, Nb and Ta in inclusion minerals, indicate that
the crystallization process is not simple. Corundum and the other minerals found as its
inclusions (zircon, columbite, thorite, uranium pyrochlore, alkali feldspar etc.) could not
have crystallized from most basaltic compositions. A more complex process must occur in
which crystallization takes place when there are high proportions of incompatible
elements and volatiles in an early melt fraction.

2.2 Properties of corundum

In gemstone form corundum is called sapphire if it is blue. The term ruby is

reserved for reds and pinks; padparadscha for a pink-orange colour, and the term
"wattle" is sometimes given to green and yellow parti-coloured stones. All other colours
are called sapphire, with the particular colour as a prefix, and are also known as "fancies".
Pure alumina is transparent and colourless and is known as leuco-sapphire, however, most
natural stones contain trace amounts of elements which cause the variety of allochromatic
colours ranging through blue, green, yellow, pink, and mauve. Nassau (1984) stated that
the blue colour of sapphire is due to the interaction between iron (Fe2+) and titanium
(Ti2+), with the appearance of the colour depending not only on the relative amounts of
each present but also on their valence states. This is controlled by the oxidizing-reducing
conditions during formation and possibly subsequent heating events. The red colour in
rubies is due to chromium (Cr). A clear, transparent and faultless ruby of uniform deep
red colour is the most valuable precious stone known. In sapphire, a fine "cornflower" blue
is the most sought after, with paler blues, blackish or green blues being less valuable. In
more recent times, fine golden stones have become popular. Stones displaying strong
colour zoning are often called parti-coloured sapphires and are common, particularly in
Australian material.
 Corundum crystallizes in the rhombohedral division of the trigonal system,
however because of equal development of the trigonal prisms and bipyramids, the crystals
have a hexagonal appearance. Ruby and sapphire often display quite distinct crystal forms.
Crystals of ruby typically display short six-sided prisms and large basal pinacoids, with
only minor rhombohedral and bipyramidal development between the pinacoid and prism
faces. The result is a flat to tabular prismatic habit shown in Figure 2-1a. Crystals of
sapphire show a more pronounced development of the bipyramidal faces with only minor
prismatic and pinacoid development, such as shown in Figure 2-1c,d. Frequent repetition
or oscillation between different pyramid forms result in horizontal striations at right angles
to the c-axis giving the crystals a stepped tapering appearance (Fig 2-1d) or barrel shaped
appearance (Fig 2-1c, Colour Plates 2 & 3). Terms used to describe sapphire crystals
include "dogs tooth" or "monkeys tooth" (Fig 1d, Colour Plates 1 & 3).
The properties of corundum are summarised in Table 2-1, (Bauer, 1968;
Crowningshield, 1983 Nassau,1984; Gemmological Association of Australia, Gem
Material Constants, 1986; Coenraads, 1990):

2.3 Form of sapphires from the Central Volcanic Province

The large scale mining operations of T.J.& P.V.Nunan Pty.Ltd. allow the viewing
of a very large number of sapphires as well as low value corundum making up the daily
mine run. The stones are predominantly well formed crystals or fragments showing crystal
faces. Pointed crystal-terminations, or "dogs teeth" and flat hexagonal prisms known
locally as "flats" are common, indicating that crystals are often broken along the basal
{0001} parting plane. Large whole crystals are rare but sections bounded by crystal faces
and parting planes with diameters of 3 to 4 cm indicate that some crystals grew to at least
12 cm in length. Sharp crystal edges are not seen, crystal edges are slightly rounded and
crystal faces are often smooth and glossy due to chemical corrosion (see Colour Plate 1).
Zoning is common, both parallel to the prism and pyramid faces forming concentric colour
bands, and parallel to the basal plane. Zoned crystals with greenish or yellowish cores (
"pipes") and blue rims may be seen. Red and pink pipes have been reported (T. Coldham,
pers. comm., 1989)
Sapphires are rarely seen intergrown with other minerals such as anorthoclase,
spinel and zircon, however limited evidence suggests that they probably grew in coarse
crystallized aggregates. Some examples follow:
1. A sample of sapphire intergrown with anorthoclase is known from the Hoy Province,
Queensland, and is held by the Australian Museum (Figure 2-2). An anorthoclasite
xenolith containing corundum has been reported from Ruddons Point, U.K. by Upton et
al (1983). A sapphire intergrown with pleonaste spinel was found in Reddestone Creek
in the Central Volcanic Province. It was registered in the New South Wales Geological
Survey collection but is now missing. A greenish blue sapphire intergrown with an
 octahedral iron-rich spinel crystal and a red zircon from Khao Wua, Thailand was
reported by Vichit (1987).

2. Sapphire crystals from the Central Volcanic Province occasionally show negative
crystal shapes in their faces as shown in Figure 2-3. These suggest growth against other
crystals even if only other sapphires.

3. Some sapphires from the Central Volcanic Province show evidence of markedly uneven
chemical resorption indicating that they have been partially protected at some stage,
possibly by being partially included in another mineral or aggregate of minerals. Such
an example is shown in Figure 2-4 and Colour Plate 1, and Figure 2-5 shows a possible
scenario explaining its morphology.

2.4 Surface features of rubies and sapphires from volcanic provinces

The surface features of a large number of stones from placer deposits associated
with volcanic provinces in both eastern Australia and Thailand were examined.
Australian sapphires were collected from Reddestone Creek No 1 plant
(29°41'30"S, 151°38'50"E), No 2 plant (29°43'00"S, 151°38'20"E), and Kings Plains
Creek (29°41'00"S, 151°27'50"E) near Glen Innes, New South Wales; and Braemar
(29°47'20"S, 151°17'30"E), near Inverell, New South Wales. Thai sapphires were
collected from the SAP Mining Co, near Bo Phloi, Kanchanaburi region (14°20'30"N,
99°29'30"E); and the Elem Mine near Ban Khlang, Chanthaburi-Trat region (12°36'10"N,
102°18'10"E). Thai Rubies were collected from the Elem Mine and the Ruby Well Mine,
near, Noen Chali, Chathaburi-Trat region, (12°32'0"N, 102°30'10"E).
Samples were examined with a binocular microscope, then selected specimens
were cleaned for 5 to 10 minutes in an ultrasonic bath of water or hydrogen peroxide,
mounted on scanning electron microscope stubs and gold coated. Photographs were taken
with a JEOL JSM 840 scanning electron microscope.
The observed features may be broadly grouped as; negative crystal impressions,
surface resorption or etch features, including chatter marks, and surface damage consistent
with transport in the alluvial environment.

2.4.1 Negative crystal impressions

Plates 1-1, 1-2 and 1-3 show sharp edged, deep, geometric shaped holes, or crystal
impressions of the order of 0.5 - 1.0 mm across on sapphire and ruby crystal faces. The
angles between the impression faces in 1-1 and 1-2 are 120°. The negative crystal shape in
Plate 1-3 is particularly interesting in that it appears to show tetragonal symmetry, thus
precluding the possibility of it having been a fluid filled negative crystal cavity. These
impressions developed as the corundum grew as part of a coarsely crystallized aggregate
with minerals such as anorthoclase and zircon (Coenraads et al, 1990) and spinel (Vichit,
1987).

2.4.2 Surface etch features

The above Plates, as well as Plates 1-4, 2-4a, 3-1a, 4-3a, 5-1a and 5-2a show the
corundum to be sub-rounded to rounded in appearance. This rounding, although once
attributed to alluvial wear, is clearly the result of resorption by magmas responsible for
carrying the corundum to the surface. The grains in Plates 1-4 and 5-1a show the surface to
be covered with flat topped triangular hillocks of the order of 15-20 μm which reflect the
internal trigonal crystal symmetry. The only exception is where the bottom of the grain in
Plate 1-4 has broken away along a conchoidal fracture surface. A further enlarged view of
the sapphires' surface in Plate 1-4b shows the relief of the triangular hillocks and their
irregularity in detail. They stand like sculptured "monoliths" on the surface of the
sapphire, roughly equidimensional and have a relief of about 20 μm. Plates 1-4c and 1-4d
show the gradual rise of the surface from the floor in between the hillocks into their walls,
and the surface in detail, which is covered with sub-geometrically regular to rounded pock
marks or shallow dishes about 1 μm in diameter. These features give the hillocks a
distinctly eroded appearance.
The hillocks are most likely of similar origin to the pyramidal or drop shaped hillocks on
rounded surfaces of diamonds (Orlov, 1977, p.93) and also reproduced in etching
experiments (Patel and Agarwal, 1966).
Plates 2-1a and 2-1b show relief on a face perpendicular to the c-axis {0001} of a
sapphire crystal. The etching is variable in distinct bands. Plate 2-1b shows detail of the
etching as oriented flat shallow hexagonal depressions of the order of 4-10 μm in diameter
with a hexagonal shaped remnant hillock centrally positioned in each.
The probable sequence of formation of the etch features is illustrated in Plate 2-1c
where all stages are present on the one crystal, and where each form merges into another
across the surface. In the bottom centre of the photograph, below the large chip displaying
conchoidal fracture, the surface is covered by sub-triangular to rounded hillocks described
earlier. In the right of the photograph can be seen what may be the early stages of
formation of such a surface. Here an originally flat surface has been etched to produce a
number of hexagonal depressions, each with a centrally placed hillock. In the left of the
photograph the surface has a spongy or porous appearance, probably as a result of
extensive etching. Plates 2-3a and 2-3b show detail of the early stages of formation of
hexagonal and triangular hillocks on a Thai ruby. Crystal terminations ranging up to 50 μm
are surrounded by a moat into which the outer wall gradually slopes. In some cases
preferential alignment of etching has led to the formation of ridges as seen in Plate 2-3b.
The hillocks are similar in appearance to the hexagonal etch marks (of the order of 10-20
μm) seen on a Thai sapphire (Plate 2-2 and also Plate 2-1b, described above), only in this
case, the outer walls of the moats are roughly sculptured as opposed to being smoothly
sloping. The reason for the development of a crystal termination centrally within an etch
pit is not clear, although it could be a response by the crystal to minimize its surface free
energy when it was in contact with the hostile carrier magma.
As the dissolution process proceeds, etch features appear to grow and interfere
with one another producing deep and prominent sculpturing ("shark skin" appearance)
such as seen on the surface of the Thai ruby (Plates 2-4a and 2-4b) and Australian sapphire
(Plates 3-1a and 3-1b). Orlov (1977, p.98) found similar features on the octahedral faces of
diamond resulting from continued dissolution which caused the etch trigons to grow larger
and crowd in on one another. Such extensive dissolution also seems to have been
responsible for the "intaglio-like" surface produced on the sapphire in Plate 3-2a. The
outer surface has been removed in places to a depth of about 15 μm, to a lower level which
appears to be more resistant and only lightly etched. Detail in Plate 3-2b reveals a fine
microlamination which has been emphasised by the dissolution as a series of steps of the
order of 1 μm in size. Such microlamination is also observed in diamond (Orlov, 1977,
p.87) and has been attributed to successive layer dissolution.
On surfaces which appear to be at a lower angle to the c-axis of the corundum
crystal (see Plates 3-3a, 3-3b and 4-1), the etch patterns are quite different, and resemble
inclined stacks of elongated, triangular shaped bricks. These structures clearly reflect the
internal crystal structure of the corundum. Plate 3-3a shows two surfaces; in the lower part
of the photograph, a surface roughly parallel to the c-axis displays triangular prisms whilst
in the upper part of the photograph a surface perpendicular to the c-axis, partially covered
by lichen, shows typical triangular hillocks described earlier. Plate 3-4a and the detail in
Plate 3-4b, show triangular depressions of the order of 1-2 μm across, with the depression
walls forming a negative triangular pyramid. Such features appear to be the result of
etching on a surface at an intermediate angle to the c-axis.
The surface features of most grains studied, even those with an apparently random
"blade-like" texture such as seen in Plate 5-2a, could be ultimately explained in terms of
surface etching. Under magnification (Plate 5-2b), a crystallographic control of the
features is revealed. In Plate 5-2b the raised "blades" are seen to form the sides of
triangular shaped pits.
Chatter marks are occasionally present and are shown in Plates 4-2a, 4-2b and 4-
2c. These features are attributed to some form of mechanical stress leading to the
formation of a series of parallel or curved marks on the surface of the grain. Detail in Plate
4-2c shows that the marks may be of 1-2 μm in width and as deep or deeper. The finely
striated faces within the curved chatter mark grooves show a pattern, not unlike that
described in Plates 3-3a, 3-3b and 4-1. This suggests that the grooves have been enhanced
by dissolution and implies that the corundum displaying chatter marks has been subject to
stresses prior to, or possibly during transport to the surface.
2.4.3 Healed fracture surfaces and damage resulting from transport in an alluvial
environment

Plates 4-3a, 4-3b and 4-3c show narrow (1-2 μm) but deep fractures which traverse
the stone irregularly or in a semi-oriented manner. These start and end abruptly and
probably represent healed or partially healed stress fractures. They are described by Orlov
(1977, p.53) as planar dislocation growth defects originating from a defect centre such as
an impurity or inclusion in the interior of the crystal and radiating to the faces. Such may
be the case in Plate 4-4. It is logical that such planes of weakness would be likely sites for
breakage in the alluvial environment. Plates 5-3a and 5-3b show such rebroken healed
fracture surfaces revealing a "fingerprint-like" pattern of tubes of about 1-2 μm in size.
Such fingerprints are commonly photographed within gem rubies and sapphires using light
microscopes (Gübelin and Koivula, 1986, p.340).
Damage in the alluvial environment results from impact during transport. This is
seen as conchoidal fracturing particularly around the more exposed edges of crystals as
shown in Plate 5-4, or as chipping on protruding hillocks (Plates 1-4d and 5-1b). In Plate
5-5, an impact has completely broken off a protruding hillock leaving a smooth fracture
surface.

2.4.4 Discussion

Surface features on rubies and sapphires from volcanic provinces may give an
insight into the history of such crystals from the time of their growth to their eventual
recovery by mining.

1. Negative crystal impressions suggest that the corundum grew as part of coarsely
crystallized aggregates together with minerals such as anorthoclase, zircon and spinel.

2. Surface resorption or etching and layer dissolution features clearly result from
reaction with the magmas responsible for carrying the corundum to the surface. These
features include, triangular and hexagonal hillocks and depressions on faces perpendicular
to the c-axis, and stacks of triangular prisms on surfaces parallel to the c-axis of the
corundum crystal. Chatter marks imply that the corundum has been subject to stresses
prior to, or possibly during transport to the surface.

3. Surface damage including impact marks displaying conchoidal fracture, broken off
protrusions and exposed healed fracture surfaces indicate the degree of reworking in the
alluvial environment. The grain surface features generally indicate a minimal amount
damage due to fluvial transport. This observation, when considered in conjunction with the
spatial variability in the physical characteristics of corundum in a given placer deposit
(Coenraads, 1990), indicates minimal degree of downstream reworking and mixing, and
suggests that the corundum sources are local to the placer deposits.

2.5 Inclusions in sapphires and zircons from the Central Volcanic Province

Inclusions may shed light on the origin of sapphires and zircons. However, in order
to do so the inclusions must be syngenetic, that is, they must have grown
contemporaneously under the same pressure-temperature conditions and compositional
regime as their host. Gübelin and Koivula (1986) defined criteria for inclusions to be
identified as syngenetic with their host. They noted that inclusions should be euhedral and
unbroken, and show no evidence of surface etching, corrosion or oxidation which would
indicate an earlier period of crystallization or derivation from the wall rocks. Inclusions
should also be independent from any cracks, fractures or epigenetic (later) mineralisation
within the host. Trapped melt droplets are another type of inclusion perhaps capable of
more directly revealing the environment in which sapphires and zircons grew. These will
be have smooth surface and spherical to highly irregular shapes. Contraction bubbles may
be visible where inclusions are sufficiently transparent and light in colour.
Samples of sapphire and zircon from the Central Volcanic Province were examined
in a high-refractive index medium (glycerine) using a binocular microscope. Those found
to contain inclusions were mounted in resin disks, ground down to the inclusion level,
polished and photographed. Inclusions which had been exposed by this process were
analysed using an electron microprobe and their identities deduced. Compositional ranges
determined for the inclusions are listed in Tables 2-2 and 2-3.
Inclusions identified in New England sapphires during this study include zircon,
hercynite, gahnospinel, columbite, niobium-rutile, ilmenite, thorite, alkali feldspar,
plagioclase, pyrrhotite, pyrrhotite-pentlandite intergrowth, and iron-rich melt inclusions
(see Table 2-2). Uranium pyrochlore (Gübelin and Koivula, 1986), hornblende and mica
(Coldham, 1985) have also been found in Australian sapphires.
Inclusions identified in New England zircons during this study include zircon,
zirconia?, ilmenite, alkali feldspar and iron-rich melt inclusions (see Table 2-3).

2.5.1 Discussion of inclusion compositional ranges

Alkali feldspar inclusion compositions are plotted on the ternary diagram shown in
Fig. 2-6. Compositional ranges are essentially the same for both sapphires (circles) and
zircons (triangles). Of the 10 inclusions analysed, two were large enough to show
compositional variation, one inclusion (19/19-1; No. 5 on Fig. 2-6) in zircon and the other
in sapphire (5/10; No. 1 on Fig. 2-6). Backscatter photographs of the inclusion in sapphire
(Figure 2-7) show an irregular shaped core of alkali feldspar mantled by plagioclase. The
boundary between the zones is sharp and non conformable with the outer euhedral crystal
faces. The plagioclase appears to be a later overgrowth.
 Spinel inclusions in sapphire plot in two distinct fields within the magnetite and
ülvospinel prisms shown in Figure 2-8:
a) Hercynite group (circles):- Inclusions in this group extend from the hercynite field
towards the pleonaste field. There is no Cr2O3 or ZnO present and MgO is low (0.1-6%).
b) Gahnospinel group (squares):- Inclusions form a tight group and are richer in ZnO (10-
16%) and Al2O3 and poorer in FeO (24-28%) and MgO (0.1-2%) than the hercynites.
The compositional fields for the inclusion spinels do not overlap those for the spinels and
magnetites associated with sapphires in alluvial heavy mineral concentrates (compare
Table 2-2 with Table 5-2; and Fig. 2-8 with Fig. 5-1).
Ilmenites included in both sapphire and zircon also have low MgO content (<1%,
Tables 2-2 & 2-3) compared to those found in the heavy mineral concentrates (Table 5-2
and Fig 5-5).
Common in both sapphire and zircon are opaque to brownish-red, translucent,
subequant to amoeboid shaped inclusions. These may neck down into thin tubes or
dumbbell shapes with no apparent relationship to the crystallographic structure of the host.
These are interpreted to be primary melt inclusions. They are soft, do not polish well, and
give low totals (70-85 weight %). They are composed predominantly of iron (60-75% FeO
in sapphire; 45-70% FeO in zircon, melt inclusions 1, Tables 2-2 & 2-3), or a mixture of
FeO, Al2O3 and SiO2 (melt inclusions 2, Tables 2-2 & 2-3)
Spinel, niobite and melt inclusions were found in sufficient numbers to permit
arbitrary arrangement into listings of oxides, in ascending or descending order for a given
oxide. The following relationship between major and minor oxides for respective minerals
are apparent:
a) Hercynite (Fig. 2-9):- Increasing FeO (total) corresponds to decreasing Al2O3 and
weakly increasing TiO2.
b) Gahnospinel (Fig.2-9):- Increasing FeO (total) corresponds to decreasing ZnO and
increasing MgO.
c) Columbite and niobium-rutile (Fig. 2-10):- Inclusions appear to form two distinct
groups on this figure although this observation is based on only 13 inclusions.
d) Melt inclusions (Fig 2-11):- Increasing FeO (total) corresponds to decreasing Al2O3,
decreasing SiO2 and, for those included in zircon, decreasing CaO and increasing Na2O.
This diagram suggests that with more analyses a continuum may exist between the two
melt inclusion groups. It is proposed that these melt inclusions should be re-analysed using
a high resolution analysis technique, but this task was beyond the scope of the present
study.
2.5.2 Conclusions

1. Syngenetic inclusions such as zircon, uranium pyrochlore, thorite, columbite, alkali

feldspar and gahnospinel (Gübelin and Koivula, 1986; Coenraads et al, 1990.) as well as
abundant carbon dioxide filled fluid inclusions (R. Wilkins, pers. comm. 1989) indicate
that the sapphires must have grown in an environment rich in incompatible elements (eg.
U, Th, Nb, Ta, Zr), alkalis (Na, K) and volatiles. Abundant iron rich melt inclusions and
crystal inclusions of iron-spinel and iron -sulphides indicate a parent melt rich in iron and
poor in silica and magnesium.
2. Compositional ranges for melt inclusions in sapphire and zircon show extensive
overlap. This, coupled with the similarity between other common inclusions (alkali
feldspar, ilmenite and zircon), suggests that sapphire and zircon grew under similar, if not
identical pressure-temperature conditions and compositional regimes.
3. Inclusion minerals such as alkali feldspar, iron-spinel and zircon are also
represented in rare xenoliths; thus all of these minerals appear to have formed
cogenetically from an exotic melt enriched in incompatible elements.
4. Compositional ranges for the inclusion spinels and ilmenites are distinct from those
similar mineral species encountered in the sapphire and zircon bearing alluvial
concentrates. This implies that the sapphires and zircons, together with their exotic
inclusion suite, formed prior to the crystallisation of the abundant spinels and ilmenites
(comparatively richer in MgO) that occur as xenocrysts in the Central Province volcanic
rocks.

2.6 U-Pb dating of the zircon inclusions in Central Volcanic Province sapphires

2.6.1 Introduction

Four small, transparent, euhedral zircon inclusions were identified during 1988
using the electron microprobe at Macquarie University. They are shown in Figure 2-12,
and Colour Plate 4.
Zircons contain trace amounts of U isotopes taken into the lattice at the time of
crystallization. By measuring the amount of breakdown to daughter Pb isotopes and
knowing the rates at which this occurs, the zircons can be dated. It is clear that such
inclusions are extremely valuable as a means of providing the age of formation of the host
sapphires. However, it is only recently that U-Pb dating of such samples has become
possible with the development of the Sensitive High Mass Resolution Ion MicroProbe
(SHRIMP) at the Australian National University, which permits in situ isotopic analysis
on a microscopic scale.
2.6.2 Analytical methods (by P. Kinny in Coenraads et al, 1990)

The SHRIMP technique for U-Th-Pb isotopic analysis of zircons was first
described by Compston et al. (1984) with subsequent modifications outlined by Kinny et
al. (1988). The electron probe mounts in which the zircon inclusions in sapphire were
identified, were modified for ion probe analysis by the addition of a hole to accommodate
the laboratory standard zircon, then repolished and recoated with carbon. The two largest
inclusions were chosen for analysis, both being of sufficient size to accommodate the 25
μm analytical spot without overlap onto the sapphire matrix (Fig. 2-13) which in any case
was found to contain no detectable lead or uranium. Each analysis consisted of three
counting cycles of the Pb, Th, and U species of interest, after which raw Pb/Pb and Pb/U
isotopic ratios, and total U and Th contents were calculated. Both analyses were
"duplicated" by resumed excavation of the same sputter-holes.
The young (Cenozoic) ages of the zircon inclusions necessitated special treatment
of the data. For young zircons, there is little value in using the 207Pb/206Pb ratio as a
measure of age, because the majority of the measured 207Pb is of non-radiogenic origin.
For this reason, however, the 207Pb provides the most effective means of monitoring the
contribution of "common" Pb. Assuming that the U-Pb isotopic systems are concordant
and undisturbed, the measured 206Pb/238U ratio, normalised to that of the laboratory
standard zircon SL3 (206Pb/238U=0.0928), provides an initial estimate of the true age, from
which the expected radiogenic 207Pb/206Pb can be calculated. Using this value together
with a modelled common Pb composition (that of Broken Hill ore Pb in this instance,
because the detectable common Pb was derived principally from surface contaminants
rather than intrinsic non-radiogenic Pb), the proportion of non-radiogenic 206Pb in the total
206Pb can be accurately determined. This yields a revised estimate for the radiogenic

206Pb/238U, and a corresponding age. The quoted uncertainties represent combined

estimates of the errors associated with ion-counting and the uncertainty associated with
normalization of the measured 206Pb/238U ratio to that of the standard zircon SL3, the
latter being based upon the reproducibility of measurements of the standard zircon during
the analytical session.

2.6.3 Results

Results are shown in Table 2-4. The duplicate analyses of both inclusions are each
in agreement to within their assigned experimental uncertainties. Grain 30/26 gave a mean
206Pb/238U age of 35.9±1.9 Ma, whereas grain 28/11 gave 33.7±2.1 Ma. Since there is no

significant difference between these two ages with respect to experimental error, they may
be combined to yield a mean age estimate for both inclusions of 34.9±1.4 Ma (2σ). Strictly
speaking, this is only a minimum estimate for the age of the inclusions, because of the
possibility that at some stage of their history an unknown amount of the radiogenic Pb
accumulated in the zircons may have been lost. However, we would argue against such a
possibility for two reasons. Firstly, the young apparent age of the inclusions implies that
up to the present time only minor damage would have been caused to the zircon structures
by the in situ disintegration of U atoms (the principal cause of Pb loss from zircons), and
secondly, being wholly enclosed within inert sapphire megacrysts the zircon inclusions
would have been protected throughout their history from interaction with permeating
fluids.
Another consideration is that Pb might not have accumulated in the zircons at all
prior to the time of eruption of the host basalt, owing to the high temperature of the source
regions, in which case they could be much older than their indicated age. The effective
"blocking temperature" for Pb in zircon has never been established. However, Rudnick
and Williams (1987) reported the preservation of Proterozoic zircon ages in lower crustal
xenoliths incorporated into a Cenozoic basalt cinder cone in north Queensland, and Kinny
et al. (1989) reported analyses of kimberlitic zircons from Botswana which were erupted
in the Permian but which nevertheless preserved Archaean ages. Both examples suggest
that if these zircons and their host sapphires were in fact ancient xenocrysts incorporated
into the basaltic magmas, it is likely that the old U-Pb ages would be preserved.
On the basis of this evidence it was concluded that the true age of both inclusions,
and hence of the sapphires themselves, is 35 Ma.

2.6.4 Discussion

The 35 Ma age of the zircon inclusions in the sapphires lies within the range of K-
Ar ages for the basalts of the East Central Volcanic Province and within the range of
zircon fission track ages for the area (Figure 1-1, Table 1-1) thus suggesting a genetic link.
It is believed that the sapphires formed in some process associated with the basaltic
magma generation or evolution. The surface morphology of the stones and their inclusion
suite indicate, however, that this process is not simple. This is reinforced by the presence
of reaction rims of spinel on some corundum crystals (Stephenson, 1976), by the reported
failure of experimental attempts to grow corundum from a corundum-bearing basaltic
composition under different pressure, temperature and hydrous conditions (Green et al.
1978), and by the anomalously high abundances of incompatible elements such as U, Th,
Zr, Nb and Ta in the inclusion minerals (Gübelin and Koivula, 1986; Coenraads, 1990).
Crystallization may be taking place as partial melting of the mantle occurs and
such a process might involve generation of more evolved magmas at depth, crystallization
of mineral assemblages including the sapphires and their inclusion minerals from these
magmas rich in volatiles and incompatible elements, and finally the disintegration and
partial resorption of the rock types during transport to the surface in later magmas. Such
minerals may crystallize from liquids whose composition represents very tiny amounts of
upper mantle partial melting. As such, these crystallisation products could then be
expected to be out of equilibrium with basalts produced by larger proportions of melting.
 Irving and Price (1981) describe a fractionation model which allows the generation
of evolved magmas (phonolites) at upper mantle depth, via the fractional crystallization of
kaersutitic amphibole, olivine, iron-titanium spinel, aluminous clinopyroxene, mica and
apatite. Their model is supported by the observation of lherzolite-bearing phonolitic lavas
from Bokkos (Nigeria), Phonolite Hill (Australia) and Heldburg (East Germany). The
crystallization products of such magmas may accumulate in pockets or fissures, or as
plating on conduit walls (Irving, 1980), until such time as partial melting has generated
sufficient magma to carry them to the surface.
Upton et al (1983) suggested that the megacrysts (eg. anorthoclase, sanidine,
clinopyroxene, kaersutite, Ti-biotite, Ti-magnetite, Mg-ilmenite, apatite, zircon and
corundum) found in upper Palaeozoic alkali basalts of the British Isles are the result of the
disintegration of such rock types. Evidence in the form of megacryst surface morphology
and the occasional observation of corundum-bearing xenoliths discussed earlier support
their suggestion.
The growth zoning, CO2-rich fluid inclusions and high inclusion homogenization
temperatures (greater than 685°C) observed in Queensland sapphires led Stephenson
(1976) and Irving (1986) to assign a high temperature and pressure magmatic origin to
their formation. They were presumed to result from fractionation of basanitic magmas, at
least within the deep crust.

2.7 Sapphire-bearing rocks in the Central Province volcanic sequence

The conclusion drawn by Thompson et al (1986) for the British Tertiary Volcanic
Province and the observations of Pecover (1987) suggest that the following sequence of
events may have occurred in the Central Volcanic Province :- Magma movements "ream
out" conduits containing crystallization products of the highly evolved magmas or
products of metasomatizing fluids, carry the evolved megacryst assemblage upwards along
fracture sets and erupt explosively onto the surface. Thompson et al (1986) suggested that
once the initial magmas ream out the magmatic plumbing system then the remaining
batches of liquid would be able to rise without significant interruption. The implications of
this model are that the volcaniclastics, or products of initial explosive volcanism,
occurring towards the base of the volcanic pile are likely to contain the most abundant
sapphires.
Lacombe (1969-70) noted that in the Bokeo Plateau region of Kampuchea, zircons,
corundum, garnet, titanomagnetite, spinel and anorthoclase megacrysts, although present
in the majority of basalts, are most abundant in the products of explosive eruptions and
their weathering products. Similarly, Barr and MacDonald (1981) noted for the
Chanthaburi-Trat area of Thailand that megacrysts of aluminous clinopyroxene, ilmenite,
garnet, zircon and spinel are abundant in the pyroclastic debris in the vicinity of vents.
Variation in the fine details of such a process, such as the time elapsed between
initial partial melting and megacryst formation, and their final transportation to the
surface, may determine the amount of sapphires in a particular source rock and possibly
their size and quality. When combined with the necessary geomorphological conditions for
the development of a sapphire deposit, the above-mentioned process of megacryst
formation may determine whether or not a particular volcanic province has economic
potential.

2.8 Conclusions

An ion microprobe age of 34.9±1.4 Ma for syngenetic zircon inclusions in New

England sapphires shows that the the minerals originated during Cainozoic volcanism in
this area. The corundum probably crystallized from strongly evolved magmas held in the
deep crust or upper mantle and was carried up in profuse abundance during the volatile-
rich, explosive eruptions.
Surface features on rubies and sapphires from volcanic provinces also indicate the
history of such crystals.
1. Negative crystal impressions suggest that the corundum grew in coarsely crystalline
aggregates along with minerals such as anorthoclase, zircon and spinel.
2. Surface resorption or etching and layer dissolution features clearly result from reaction
with the magmas responsible for carrying the corundum to the surface.
3. Etched chatter marks imply that the corundum has been subject to stresses prior to, or
possibly whilst in transit to the surface.
4. Surface damage, including impact marks displaying conchoidal fracture, broken off
protrusions and exposed healed fracture surfaces, result from reworking in the alluvial
environment. Such damage is usually minimal indicating the crystals to be near source.
 Chapter 3

KEY AREAS FOR ALLUVIAL DIAMOND AND SAPPHIRE

EXPLORATION IN THE NEW ENGLAND GEMFIELDS, NEW SOUTH
WALES

3.1 Introduction

The aim of this chapter is to define key areas for diamond and sapphire exploration
within the New England Gemfields of eastern Australia through an understanding of the
geologic and geomorphic processes that operated within the Tertiary Central Volcanic
Province. This chapter describes the structural control and timing of volcanism in the
Central Province. These, combined with geomorphic factors, have determined the
distribution of Quaternary alluvial sapphire and Tertiary deep lead concentrations of
diamond and sapphire.
The palaeo-Hobbs and the palaeo-Gwydir diamond-bearing deep lead Tertiary
systems, the Reddestone Creek and Kings Plains Creek sapphire-bearing placers, and the
Braemar deep-lead sapphire deposit are presented as case studies in Chapter 4.

3.2 Central Province volcanism

3.2.1 Structural control of Central Province volcanism

The elongated shape of the Central Volcanic Province implies that volcanism was
localized by preferred planes of weakness in the earth's crust. These were probably of the
order of 100 to 200 kilometres in length. A Landsat lineament analysis and analysis of
physiographic features in the Central Province were carried out to evaluate these
relationships.
The lineament analysis was confined to the Woolomin-Texas structural block
which is bounded by the Peel Fault to the west (Fig. 3-1), and by the Demon Fault to the
east (just outside the eastern border of Fig. 3-1). The analysis covered portions of the
Inverell, Manilla, Grafton and Dorrigo-Coffs Harbour 1:250,000 Geological Sheets (New
South Wales Government Department of Minerals and Energy) and utilized Landsat
images 090-080F, 096-081 and 095-081 (Australian Centre for Remote Sensing) which
cover the area of interest with considerable overlap. All visible lineaments were
highlighted on plastic overlays (Colour Plate 5) at 1:250,000 scale and the data presented
in the form of a rose diagram (inset Fig. 3-1) constructed by summing the lineament
lengths (in kilometres) corresponding to their respective azimuths (in degrees from north).
Analysis of physiographic features was carried out at 1:100,000 scale, using the
Inverell, Glen Innes, Guyra, Bundarra, Armidale, Yallaroi and Bingara topographic sheets
(New South Wales Government Department of Lands). Map sheets were joined and
colored in order to strongly enhance prominent ridge and valley trends as well as areas of
high ground.
The major trends determined from the observed physiographic features were
complimented by the Landsat interpretation. The rose diagram indicates four major groups
of lineament development. The most significant are the NNW-SSE trending structural
elements, (numbered 1, 2 and 3 on Fig. 3-1), which correspond to prominent long, straight
ridges and areas of maximum basalt thickness. These coincide with the long axis of the
Central Volcanic Province and are sub-parallel to the Peel and Demon fault traces. They
will be referred to as "axial lineament trends".
The NNW-SSE trends appear to have acted as the loci for emplacement of volcanic
centres and plugs located along the length of the lava field. Six named centres lie along
lineament trends numbered 1 and 2 on Figure 3-1, and four named volcanic plugs lie
parallel to the landsat lineament trend numbered 3. Landsat lineaments were not visible on
this portion of the volcanic province itself owing to the higher landuse on the fertile
volcanic soils.
A series of radial and ring fractures centered near the Gough Sugarloaf plug are
clearly visible on the Landsat images (Fig. 3-1; Colour Plate 5). This structure, described
as the Maybole Volcano by Pecover (1987), appears to be the most significant volcanic
centre in the Province. It is located at the intersection of the four prevalent lineament sets
and coincides with a maximum thickness of basalt. The Maybole Volcano is also the
centre of a large radial drainage system.
A number of other important structural trend directions are evident within and
around the Central Volcanic Province (Fig. 3-1). These cluster around 0°-20°, 60°-80 and
95°-110° and are reflected mainly as lineament-controlled drainage (Coenraads, 1988).

3.2.2 Timing of Central Province volcanism

The locations of age determination sites in the Central Volcanic Province ages are
shown in Figure 1-1 and are listed in Table 1-1. The ages of volcanism associated with
axial lineament trends 1 and 2 differ from those associated with, and west of, axial trend 3.
Accordingly, an older East Central Province (32 - 38 Ma) and a younger West Central
Province (19 - 23 Ma) can be delineated. Lineament trend 3 indicates that there is a
structural link between the West Central Province and the volcanics to the south and west
of Armidale, some of which have a similar age (20.7 Ma).
The results of this study show a correlation between the lineament maximum at
150° - 180° and the axis of elongation of the Central Volcanic Province. Volcanism in the
Central Province has thus been controlled by NNW/SSE trending crustal fractures.
3.3 Sapphire and diamond occurrences in the Central Volcanic Province Quaternary
alluvium and Tertiary deep leads.

3.3.1 Quaternary alluvium.

Figure 3-2 highlights the major rivers and creeks associated with the Central
Volcanic Province. The most obvious feature is the radial pattern of drainage centered on
the Ben Lomond-Maybole area. It has an approximate radius of 50 kilometres. Sapphires,
together with zircon, spinel and/or ilmenite, may be found in Quaternary gravels in most
gullys, creeks, and rivers draining the 32 - 38 Ma East Central Province.
Creeks and rivers flowing westward and north-westward, such as the Gwydir,
Copes, Swan Brook, Middle and the Macintyre, appear to have been diverted from their
original flow direction by the 19 - 23 Ma volcanic activity to the west and northeast of
Inverell. Swan Brook, Middle Creek and the Macintyre have been deflected to the north in
the vicinity of Inverell, and Copes Creek and the Gwydir deflected to the west (Fig. 3-2).
The 19 - 23 Ma West Central Province created its own radiating drainage network.
Croppa, Gourmana, Mosquito, Reedy, Kelly's, Sheep Station and Myall Creeks flow
westerly from the volcanic high ground. Oatleys and Bannockburn creeks flow northward,
and Auburn Vale Creek flows to the south. Numerous small creeks drain to the east and
southeast from the volcanic high ground, but the radial pattern is not fully developed due
to the pre-existing high ground of the East Central Province.
Sapphires have not been reported in the streams draining the basalts of the West
Central Province.

3.3.2 Tertiary deep leads.

The volcanics of the West Central Province overlie and protect alluvial sediments
forming "deep leads" which have been recognized for their diamond bearing potential
since 1872, (MacNevin, 1977). Cassiterite, gold, zircon, garnet, sapphire, tourmaline, and
topaz are also found in the deep leads. The deep leads outcrop extensively around
Armidale, in the Copeton-Gilgai-Stannifer area (south east of Inverell) and also to the
north west of Inverell extending to the Queensland border (Fig. 3-1).
The possibility of two periods of volcanism, separated by a period of drainage
development and alluvial deposition, was noted last century. Anderson (1888 p.156)
observed boulders of basalt in deep lead alluvium which "could not have been derived
from the Tertiary basalt overlying in the wash but must have come from an older bed of
basalt". Using the relative heights of the bases of Tertiary deep lead channels in the
Copeton, Inverell and Gragin areas, Cotton (1914) showed that the Tertiary land surface
fell away to the north and west. It is likely, therefore, that the deep lead zircon and
sapphire were derived from the older volcanics of the East Central Province.
3.4 Identification of exploration targets for diamond and sapphire

3.4.1 Analysis of sub-volcanic topography

3.4.1.1 Objective: Deep lead alluvials, worked for diamonds, gold and tin, have long been
recognized as channel deposits with the richest grades of heavy minerals being found
along the palaeochannel axes. Many attempts have been made to map the location of the
palaeochannels (Anderson, 1888; Stonier, 1895; Curran, 1897; Pittman, 1897; Cotton,
1914; Gibbons & Pogson, 1963). The palaeotopographic analysis presented here makes
use of large scale topographic maps, borehole data and recently available, detailed
geologic maps.
The present study shows the precise location of the diamond-bearing
palaeodrainage beneath the basalts of the West Central Province and also highlights likely
source areas for the alluvial diamonds.
The development of post-volcanic drainage controls the location and grade of
alluvial sapphire deposits derived from non-economic source rocks. Comparison of pre-
and post-volcanic drainage patterns makes it possible to gain an insight into the
mechanisms involved in this process and to predict areas where these mechanisms have
been most effective.

3.4.1.2 Method: Analysis of the sub-basaltic topography was carried out at a scale of
1:25,000 over most of the Central Volcanic Province (see Appendix 1). The geological
boundaries between Central Province volcanics and older basement rocks were overlain on
the topography. Elevations for each mapped boundary were plotted and then contoured as
shown in Figure 3-3. The interpretation is conservative, that is to say, in areas of basalt
cover the basement surface is interpreted to dip gently and smoothly underneath. The least
amount of basalt cover required was interpreted in areas of poor control.
In order to improve the analysis, data from some 430 water bores (New South Wales
Government Department of Water Resources; see Appendix 2) were also plotted on the
1:25,000 sheets. The logs are poorly described but serve to distinguish between basalt and
basement rocks, or at least indicate a minimum depth of basalt in a given borehole. The
bore data generally supported the basement topographic analysis and, in many cases,
indicated that the basalt filled valleys are steeper and deeper than are given by the
conservative analysis. After contouring the pre-volcanic topography for each 1:25,000
sheet, the information was linked to the surrounding sheets and the whole reduced to
1:100,000 scale. At this scale the pre-basaltic drainage was compared with the present
drainage pattern.

3.4.1.3 Results and discussion: Figure 3-4 shows the pre-volcanic and present day
drainage patterns in the Central Volcanic Province. The lavas flowed into the
palaeovalleys and progressively filled them. The areas of basement rocks exposed today
are generally of equal or higher elevation than the immediately adjacent volcanic rocks.
These acted as highs around which the lava initially flowed. In some cases, it is clear that
the basement was covered and has now been exhumed. Very little basement rock is visible
through the thickest and highest part of the volcanic pile along the East Axial Lineament
between Glen Innes in the north and Guyra in the south. In this area the pre-basaltic
drainage cannot be determined. A radial and annular pattern, controlled soley by the
volcanic pile, has developed. The headwaters of the Macintyre River, Paradise Creek,
Wellingrove Creek, Furracabad Creek, Stonehenge Creek, Ben Lomond Creek, Gara River
and Moredun Creek are radially arranged. Beardy Waters and Grahams Valley Creek form
an annular system around the Maybole Structure (Pecover, 1987).
The palaeodrainages in the East Central Province (Fig. 3-4) indicate that a radial
drainage pattern was present at the onset of Central Province volcanism. This may imply
pre-volcanic doming, such as was reported by Wellman (1988) for the large shield
volcanos, or uplift along the Great Divide, or perhaps even an earlier eroded centre. This
doming or uplift may have been responsible for opening the NNW-SSE trending axial
fractures sub-parallel to the pre-existing fabric of the New England Fold Belt, which acted
as the loci for Central Province volcanism.
The lavas of the West Central Province have acted similarly to those in the East,
filling pre-existing valleys. Again, however, it is difficult to determine the exact position
of the palaeochannels in areas where the volcanic pile is thickest, that is, to the north-west
of Inverell along the West Axial Trend (Fig. 3-4). As the pre-volcanic surface was buried
by basalts, the drainage was severely or entirely disrupted. Streams were reset in different
positions and with different local base levels. Unlike in the East, some of the major
palaeorivers in the West Central Province such as the Gwydir, Hobbs and Macintyre, have
been markedly altered and even reversed in direction from their original north-easterly and
north-westerly trending channels. Today, some of these palaeochannels are crosscut by
numerous westerly flowing streams, including Warialda, Sheep Station, Hobbs, Myall,
Staggy and Sandy Creeks (Fig. 3-4).
Based on the sub-volcanic topographic mapping, the following drainage
features were developed on the post-volcanic landscape of the Central Province:

1. Streams following their original pre-basaltic course: These streams have

remained in their original valleys even after infilling by basalt. The valleys are thus
generally flat floored resulting in streams which are underfit, low energy and meander
over their floodplains. Kings Plains and Reddestone Creeks are examples (Fig. 3-5A & 3-
6A). This configuration is extremely important for the concentration of economic sapphire
deposits.

2. Lateral and twin lateral streams: These streams flow alongside and parallel to
the original basalt filled valleys. They flow along the contact between the basalt flows and
the basement rocks where there is a convenient plane of weakness (Fig. 3-5B). Once fixed
in this position, continued downcutting may isolate the original basalt filled valley as a flat
topped or slightly concave upwards mesa. This relief inversion by lateral stream activity is
illustrated by portions of Swan Brook, Kings Creek and Paradise Creek (Fig. 3-5C).

3. Cross cutting and pirate streams: The process of stream piracy is illustrated in
Figure 3-6 using Kings Plains and Reddestone creeks as examples. Kings Plains Creek is
sketched in part A at the point where its broad basalt filled valley swings round to the
west. Arrawatta Creek has not yet captured its flow. Reddestone Creek, however, (Fig. 3-6
part B), has been diverted from its original broad valley into a steep narrow gully flowing
to the northeast.

4. Reverse direction streams: The palaeo-Gwydir and palaeo-Hobbs Rivers once

flowed to the north and northeast under what is now the thickest and highest part of the
West Central Province volcanic pile. These courses have since been disrupted and even
reversed in flow directions in places due to volcanic activity and/or uplift. The palaeo-
Gwydir (Fig. 3-4) once flowed northward and then eastward, to finally join the Macintyre
River in the vicinity of Inverell. That course was blocked and today the Gwydir River
flows westward. Auburn Vale Creek flows southward (a reversal) into the present Gwydir
River.

3.5 Geologic and geomorphic evolution of the Central Volcanic Province.

The geological and geomorphic evolution of the Central Volcanic Province is

summarized as follows:-

a/ Pre-volcanic doming, and/or uplift along the Great Divide altered the pre-
volcanic topographic surface producing high ground and radial drainage centred on the
East Central Province. This was accompanied by block faulting and the opening of deeply
penetrating NNW trending fractures sub-parallel to the pre-existing fabric of the New
England Fold Belt.

b/ Injection of breccias and explosive eruption of pyroclastic material took place

prior to 36 Ma, forming a blanket of volcanic debris, (some sapphire-bearing), on the pre-
basaltic topography.

c/ Explosive volcanic eruption continued, as well as basalt extrusion, with a

general decrease in the amount of explosive activity with time. Early basaltic lavas flowed
down valley systems and may be found directly on basement rocks in places.
 d/ Major effusive basaltic eruptions took place along the Eastern Axial and
Wellingrove lineaments and parallel fracture sets at about 32 - 36 Ma. This produced a
"lava field" with an overall SSE-NNW elongation.

e/ Erosion and reworking, particularly of the less resistant volcaniclastic rocks,

took place over some 10 million years and this material was deposited around the edges of
the East Central Province. Alluvial concentration of tin, tourmaline, topaz and gold (from
the granites), diamonds (from local intrusives and breccias) and sapphire and zircon (from
the volcaniclastics and basalts) took place.

f/ The next phase of volcanism, also controlled by SSE/NNW trending planes of

weakness, occurred further to the west. Basaltic eruptions grading from tholeiitic through
to alkaline occurred at about 19-23 Ma and formed the West Central Province, which also
includes volcanics south and west of Armidale. The basalts flowed out and covered the
Tertiary alluvium, and thus formed the deep leads that have been periodically exploited for
diamonds. To the east of the East Central Province these alluvials have long since been
removed due to the westward migration of the Great Escarpment and the lack of a
protective basalt cap. Pockets remain however to the northwest, west and south.

g/ The development of the West Central Province deflected the west and
northwesterly flowing drainage system radiating from the Maybole high and developed a
radial drainage pattern centered on the Delungra-Mount Russell areas.
 Chapter 4

DELINEATION OF ALLUVIAL AND DEEP LEAD EXPLORATION

TARGETS: CASE STUDIES.

4.1 Introduction

The Central Province has been divided into three prospective target regions for
sapphires and diamonds (Fig. 4-1) based on the detailed palaeotopographic reconstruction
technique described in the previous chapter These regions comprise a western region with
a potential for "palaeo-Gwydir type" alluvial diamond deposits, a central belt with
"Braemar type" sapphire deposits, and an eastern division with "Kings Plains- Reddestone
type" alluvial sapphire deposits.
Case studies of these "type" deposits of diamond or sapphire in the Central
Province will be described in this chapter.

4.2 Diamond prospectivity in the palaeo-Gwydir River system

The Palaeo-Gwydir River system was buried by 19-23 Ma volcanism of the West
Central Province. Prospectivity for diamond-bearing deep lead alluvium is considered to
be favorable in two major palaeodrainage systems that extend northward from the present
Gwydir River in the vicinity of Copeton Dam. These are described below as the palaeo-
Hobbs and palaeo-Gwydir systems and are shown in Figure 4-2.

Palaeo-Hobbs: A tributary of the palaeo-Hobbs heads northward from a

palaeosaddle near Copeton Dam (GR:962922 Copeton Dam 1:25,000, Fig. 4-2). Basalt at
the base of the channel at this position indicates an original elevation for the palaeoriver of
650 m asl. It can be traced for 20 kilometres heading in a north to northwesterly direction,
over which distance the elevation of the channel falls to 480 m asl (GR:900080 Delungra
1:25,000, Fig. 4-2). At this location the channel joins the palaeo-Hobbs and the palaeoaxis
heads in a north northeasterly direction where it eventually becomes obscured by extensive
basalt cover.
The palaeochannel is presently cut by westerly flowing streams, including Sandy,
Hughies, Staggy, Myall and Spring creeks, which dissect the palaeovalley walls. This is
particularly obvious in the western palaeovalley wall where downcutting is of the order of
100 m.
The Staggy Creek (GR:955975 Gum Flat 1:25,000, Fig. 4-2) and Wonderland
(GR:968954, Fig. 4-2) diamond bearing leads form part of the NNW trending palaeo-
Hobbs system. They are not part of the same system as the leads occurring at a lower
elevation in the vicinity of Copeton Dam, but are from a separate source as was first
suggested by Stonier (1895).
 Palaeo-Gwydir: The elevations of the deep leads fall from the south, and from the
west, to a point where two tributaries of the palaeo-Gwydir join, near Round Mountain
(GR:087890 Tingha 1:25,000, Fig. 4-2). Progressing from the west (Fig. 4-2), Oaky Creek
& Kirks Hill (GR:010920 Copeton Dam 1:25,000), The Banca (GR:022919) and
Stockyard Hill (GR:043927) are at 620 m, Malacca (GR:052918) is at 600 m and Soldiers
Hill (GR:067906) is at 590 m. Moving from the south, Ryders Lead (GR:075804 Tingha
1:25,000) is at 700 m, Kenzies Claim (GR:085860) is at 680 m and Collas Hill
(GR:085860) is at 620 m. Stonier (1895) first recognized this channel and his ideas were
verified by Cotton (1914) on the basis of numerous aneroid elevation readings. The
palaeo-Gwydir leaves the Copeton Dam area in the vicinity of Auburn Vale Creek
(GR:081918). At this point alluvium is exposed below the basalt fill and the channel base
is at 580-590 m. asl. The channel can be traced heading northwards for approximately 8
kilometres, but becomes obscured beyond that point by extensive basalt cover. It is likely
that it flowed to the NE to join the palaeo-Macintyre as was suggested by Cotton (1915),
or perhaps flowed to the NW around a basement high (GR:060040 Gum Flat 1:25,000,
Fig. 4-2). Today Auburn Vale and Long Gully creeks flow southward forming "twin
laterals" on either side of the basalt filled valley.
The present palaeotopographic reconstruction indicates that the Gwydir River has
been diverted, at least once, by the volcanic activity of the West Central Province. It also
indicates at least 3 separate systems of diamond bearing alluvials and, therefore, local,
multiple sources for the diamonds. MacNevin (1977 p.27) also states that "the great
variety, both in size and character of the diamonds from different part of the leads" is
consistent with multiple sources. Local sources for the diamonds were noted by Cotton
(1914) at Oakey Creek (GR:010920, Fig. 4-2) and have been described as lamprophyric
dykes intruding shattered granitic basement (Cluff Resources Pacific Limited, Annual
Report, 1989). Cluff Resources Pacific Limited are currently evaluating the economic
prospectivity of a lamprophyric pipe complex with associated blast breccias in the Round
Mountain - Star Hill area (GR:100900, Fig. 4-2).

4.3 Kings Plains and Reddestone type alluvial sapphire deposits.

Two major sapphire deposits are situated in the Kings Plains and Reddestone
Creeks of the East Central Province (Figs 4-3 and 4-4). These Holocene alluvial deposits
are currently being worked by T.J. & P.V. Nunan Pty Ltd using mechanized mining
methods. Each mine processes approximately 1000 tonnes of alluvium per day, yielding 1
tonne of heavy mineral concentrate containing about 10 kilograms of corundum (Nunan,
1989). The heavy minerals associated with the sapphire in the alluvial gravels are
pleonaste, ilmenite, chrome-spinel, titanium-magnetite, corundum, zircon and minor
chrysoberyl. There is a clear association between these deposits which have developed
since the filling of the palaeovalleys with basalt, and the present drainage. However the
locations of the sapphire deposits do not relate in detail to the position of the present low
energy streams which meander across the alluvial floodplain.
Stream energies in the fluviatile system must have been high enough to remove
most weathered material, such as clays, downstream, but at the same time, not sufficiently
high to disperse the heavy minerals of interest. In an ideally balanced system the heavy
minerals tend to move vertically downwards, rather than downstream, and become
concentrated with time. This is evident at a Public Fossicking Area (formerly a mining
area) located at grid reference 630188 on the Glen Innes 1:25,000 sheet. Here, in a small
valley, sapphires and zircons may be found in alluvium trapped in crevasses, and between
boulders on Permian granodiorite. No basaltic and/or volcaniclastic source rocks remain in
the present catchment area.
Concentration of sapphires is clearly favored by transporting weathering products
of source rocks into smaller volumes or areas. The longer the process continues, the more
effective will be the concentration of heavy resistant minerals. In contrast, a diverging,
radially distributed drainage system will lead to a dispersion of the sought after gems.
Both scenarios are illustrated in Figure 4-5.
The palaeotopography, palaeodrainage and present-day drainage at Kings Plains
and Reddestone Creeks are shown in Figures 4-3 and 4-4. The present Kings Plains Creek
flows within the palaeochannel until it is captured and diverted northward (GR:456220,
Fig. 4-3). Borderline economic sapphire deposits, associated with Kings Plains Creek,
exist to this point. When Arrawatta Creek eventually captures Kings Plains Creek (at
GR:505215), the downstream deposits will no longer show any obvious association with
the Kings Plains Creek. It is these types of abandoned channel deposits, which may
otherwise be difficult to find, that are highlighted by palaeotopographic reconstruction.
Within the Reddestone palaeovalley, the present day creek has been captured and
diverted northeastward (at GR:730204, Fig. 4-4). Economic sapphire deposits are
associated with the present creek above and below this point.
The present study indicates that the Kings Plains palaeovalley should be explored
further downstream from where the present Kings Plains Creek has been captured (at
GR:456220). A comparison between Figures 4-3 and 4-4 indicates that there is the
potential for sapphire deposits to be discovered along the now-abandoned palaeovalley,
which continues northwestward from the point of capture of the present Reddestone Creek.
Such channels may contain economic deposits if the original drainage existed for
sufficient time to allow concentration to take place. The area is currently held under
exploration licence by T.J. & P.V. Nunan Pty Ltd with a view to testing this prediction.
Detailed maps of the sapphire deposits are shown in Figures 4-6 and 4-7. The data
were collected by T.J. & P.V. Nunan Pty Ltd, using one metre diameter bucket drills to
penetrate the alluvium through to weathered bedrock (as shown in Colour Plate 11). Any
gravels encountered, (generally just above the bedrock) were tested for their heavy mineral
content. Data points are shown as dots, and contour lines indicate the depth to bedrock in
metres below the floodplain. The hatched areas indicate the position of sapphire bearing
gravels, with concentrations of greater than 15 gm/m3 at Reddestone Creek (Fig. 4-6) and
concentrations of greater than 50 gm/m3 at Kings Plains Creek (Fig. 4-7). Additional
stipple indicates grades higher than 100 gm/m3 rising to an excess of 500 gm/m3 in some
holes.
The maps indicate that the sapphire-bearing 'shoestrings' roughly coincide with the
axis of the bedrock valley and that highest sapphire concentrations occur in deeper pockets
or traps in the bedrock.
The bedrock is generally weathered basalt, or in some areas in Kings Plains,
weathered silicic volcanics. Gravels generally lie on the weathered rock and include
weathered basalt fragments, hard white nodules (cavity fillings of secondary
mineralization from the basalt) and ironstone in a clay matrix. Quartz is noticeable where
silicic volcanics form part of the channel. The sapphire-bearing "wash" occurs typically at
a depth of about 4 to 9 metres, and is around 0.5 - 1 metre thick. The "wash" is overlain by
fine sediments of the present floodplain.
The Reddestone and Kings Plains deposits postdate the majority of the basaltic
activity (32-38 Ma), but weathered basalt overlies portions of the wash in the Western
Feeder at Kings Plains (T. Nunan, pers.comm.) and at Braemar, where a K-Ar age of 23
Ma. was obtained for the overlying basalt (C.D. Ollier, pers comm.). Clays overlying the
deposits at Reddestone Creek contain vertebrate fossils (Horton and Connah, 1981) which
indicate a minimum age of 5000 years.

4.4 Braemar type deep-lead sapphire deposits

Potential sapphire deposits have been delineated in the Braemar palaeochannel and
its tributaries up-palaeostream from the Braemar sapphire mine, in which sapphires are
recovered along with other heavy minerals, with the highest concentration occurring at the
base of the channel. "Braemar-type" deep lead deposits considered in this study had
approximately 10 million years in which to form, and required the presence of a younger
basalt capping (19-23 Ma) for their preservation.
Potential sapphire-bearing alluvials are also delineated along the topographically
inverted palaeo-Swan Brook. The extent of these "deep lead" deposits to the east of
Braemar, in palaeochannels that drained the sapphire-bearing East Central Province, is
related to the easternmost incursions of the lavas of the West Central Province.
The Braemar sapphire deposit is located along the Gwydir Highway 18.5
kilometres east of Inverell, New South Wales (Figs. 4-8, 4-9 and 4-10). The sapphire-
bearing material, exposed in Mr. Col Rynnes quarry on the side of a hill, has a
tuffaceous/brecciated appearance and is overlain by a flow-basalt (Fig. 4-11; Colour Plate
7). Braemar was recognized by the New South Wales Government Department of
Minerals and Energy geologists as being particularly important because of its non-
conventional setting (being on a hillside), and thereby presenting new opportunities for
sapphire exploration in the New England region, and also in many of the other eastern
Australian volcanic provinces. A seminar entitled "Tertiary Volcanics and Sapphires in the
New England District" was held by the New South Wales Department of Minerals and
Energy on the 1st May 1987 to promote research and exploration.
Since an initial model for Braemar was proposed by Lishmund and Oakes, (1983),
in which the deposit was considered to be tuffs and breccias formed around a diatreme, a
significant amount of further work has been carried out in the area. The sapphire-bearing
sediments at Braemar were first considered to be palaeochannel deposits by Temby (1986)
owing to their appearance and the discovery of cassiterite, derived from the Permian
basement, in heavy mineral concentrates. Drilling by Brown and Pecover (1986)
confirmed that the sediments lie in deep basement channels.
The physical processes involved in the concentration of sapphire and other heavy
minerals as placer deposits in the Braemar palaeochannel appear to be similar to those
which formed the major sapphire deposits at Kings Plains and Reddestone creeks
(Coenraads, 1990). Braemar is unique, however, in that it is situated in an area in which
the younger (19-23 Ma.) volcanics of the West Central Province overlap onto the older
(32-38 Ma.) volcanics of the East Central Province (Fig. 4-8). The sapphire-bearing,
intrabasaltic sediments contain minerals which are derived from the older volcanic rocks,
as demonstrated by Hollis & Sutherland, (1985) from a fission track age of 37 Ma. for a
zircon from Braemar. The sediments are overlain by a basalt flow dated at 23 Ma. (C.D.
Ollier, pers. comm.) and must therefore have an age of between 23 and 37 Ma.. McMinn
(1989) reached a similar conclusion, dating the sediments as Eocene/Oligocene on the
basis of palynology.
At least two generations of sapphire-bearing alluvial deposits exist at Braemar
(Temby, 1986; Pecover & Coenraads, 1989) (Figs. 4-10, 4-12 and 4-13). The oldest
economic deposits lie in a palaeochannel exposed in a quarry (Fig. 4-11; Colour Plate 7),
and are worked intermittently by Mr. Col Rynne. The deposits are Eocene/Oligocene in
age, and are clay-rich, white and grey, fluvio-lacustrine and tuffaceous sediments with the
highest grades of sapphires occurring at the base of the channel amongst weathered basalt
boulders which range in size up to half a metre. These earlier sapphire-bearing alluvials
have been preserved by a capping of basalt. The youngest deposits at Braemar are
Holocene alluvial gravels which are situated in Schumachers Gully and its tributaries
(Figs. 4-15 and 4-16). They have undergone at least two cycles of reworking and
deposition and, as a result, are the richest deposits with the highest proportion of gem-
quality sapphire (C. Rynne, pers. comm.).
Zircon, ilmenite and chrome-spinel are associated with the sapphire at Braemar
(Temby, 1986; Slansky, 1987; Coenraads, 1990). Cassiterite is also found in small
quantities where Permian basement rocks have been reworked (Temby, 1986; Slansky,
1987) and two diamonds have been reported (J. Rynne, pers. comm.). Chrome-spinel and
ilmenite have also been recovered from the intrusive red breccias at Braemar (Fig. 4-14;
Colour Plate 8).
Braemar is the only mine known to the author where sapphires are recovered from
below basalt, although, further west in the Copeton and Bingara areas, diamond-bearing
deep leads have been worked sporadically since 1872.

The aims of this section are as follows:

1. To use a palaeotopographic reconstruction technique, based on the elevation of the

basalt-basement contact and available drill hole data, to determine the course of the
palaeochannel in which the Braemar sapphire deposit is situated.

2. To define the extent of the Eocene/Oligocene sapphire-bearing sediments lying below

23 Ma. basalt in the Braemar palaeochannel and its tributaries.

3. To define other palaeochannels in the area of influence of the 19-23 Ma. basaltic lavas
likely to have undergone a similar series of events and hence also likely to contain
deep lead sapphire.

4.4.1 Sapphire-bearing palaeodrainage systems near Braemar

The palaeotopography of the Braemar area was mapped using the elevation of the
basalt-basement contact based on detailed geologic mapping by Temby (1986), Brown &
Pecover (1986), Brown (1987), Stroud (1989) and Pecover & Coenraads (1990). The
positions of the palaeochannels can thus be better constrained, particularly in the Braemar
area, than the position inferred by Temby (1986) based on the regional distribution and
grainsize of outcropping sub-basaltic sediments
The sapphire-bearing Eocene/Oligocene fluvio-lacustrine sediments are situated in
a basalt-filled, northeasterly trending tributary of the palaeo-Swan Brook, referred to as the
"Braemar palaeochannel" (Figs. 4-15 and 4-17). A tongue of basalt extending towards
Dodds Hill, south of Braemar (Figs. 4-16 and 4-17), is apparently an upstream
continuation of the Braemar palaeochannel. Two smaller tributaries join the Braemar
palaeochannel in the vicinity of Braemar Homestead (Fig. 4-16); the larger of the two
trends southeasterly from Carinda Homestead (the Carinda Branch) and the smaller,
trending westerly (the Costeans Branch), has been exposed in the exploration costeans
mapped by Pecover and Coenraads (1989) (Figs 4-12 & 4-13). Diamond drill holes DDH 1
and DDH 7 at Braemar (Brown & Pecover, 1986) indicate that this channel is filled with at
least 60 m of flow-basalt. Fig. 4-18 shows cross sections, AB and CD, of the Braemar
palaeochannel. Section AB incorporates the data from drill hole DDH 1 (Brown &
Pecover, 1986). The locations of the sections are shown on Fig. 4-16. The present day
topography (Fig. 4-16) and the sections, indicate that lateral stream activity, post dating
the 19-23 Ma. basalt flows, has begun to invert the topography.
Diamond drill holes DDH 2 at McCarthys Knob and DDH 3 at Dodds Hill (Brown
& Pecover, 1986) support the presence of another palaeochannel that heads in a
northwesterly direction from Dodds Hill (Figs. 4-15 & 4-17) and eventually joins the
palaeo-Swan Brook. DDH 2 & 3 did not reach basement rocks, but penetrated some 45
metres of white and grey, tuffaceous and fluvio-lacustrine channel fill sediments similar to
those exposed in the Braemar quarry. The accumulation of such a major sedimentary
sequence is thought to be due to a damming of the westerly flowing palaeochannels by
faulting associated with the Severn Thrust (Temby, 1986). Alternatively, the damming
may have occurred as a result of uplift and volcanism of the West Central Province which
also appears to have been responsible for the deflection of the major westerly flowing
rivers such as the Gwydir and Macintyre (Coenraads, 1990). Local evidence for the
damming of streams by lava flows is reported by Smith (1989).
The present-day Schumachers Gully and its tributaries flow at right angles to the
former flow direction of the Braemar palaeochannel (Fig. 4-16). They flow to the
northwest across rocks and sediment filling the Braemar palaeochannel and then onto late
Palaeozoic metasedimentary basement. The Holocene alluvials, which have been mined
for sapphires in the tributaries of Schumachers Gully, are apparently reworked mainly
from the Eocene/Oligocene sediments associated with the Braemar palaeochannel
(Pecover and Coenraads, 1989).

4.4.2 Extent of the Eocene/Oligocene sapphire-bearing sediments below basalt near

Braemar

The Eocene/Oligocene sapphire-bearing sediments at Braemar are associated with

the Braemar palaeochannel and were preserved by a protective capping of 23 Ma. basalt.
Locally the base of the young basalt appears to be at about 720 m above sea level. White
and grey, tuffaceous and fluvio-lacustrine sapphire-bearing sediments are exposed below
the basalt at this level in the quarry at Braemar (Pecover & Coenraads, 1989) (Fig. 4-11;
Colour Plate 7). Similar material outcrops close to this level, south of the Gwydir Highway
and near Carinda (Fig. 4-16) and has been mapped by Brown & Pecover, (1986). Based on
this level, the 23 Ma. basalt cap is interpreted to extend up-palaeostream in both the
Braemar palaeo-channel and the Carinda tributary. It is predicted that sapphire-bearing
sediments exist beneath a thin basalt cover within the stippled area in Fig. 4-16. White and
grey volcaniclastic sediments have also been mapped by Brown & Pecover (1986) at about
690 to 710 m along the sides of the palaeochannel heading northwest from Dodds Hill, in
the vicinity of Dodds Hill and also south of the Gwydir Highway in the vicinity of
McCarthys Nob.
 Diamond drill holes DDH 2 and DDH 3 indicate the sediment in the Dodds Hill
Palaeochannel (Fig. 4-15) to be more than 5 times thicker than that in the Braemar
palaeochannel. Hence the axial deposits, with presumably the highest sapphire
concentrations, are predicted to be deep and therefore difficult to test and extract. The
exploration licence (E.L.2987) for this area was held by Hooker Resources Pty. Ltd.
(Pithers, 1988) and at the time of preparation of this thesis, their findings were
confidential.
In the downstream direction along the Braemar palaeochannel, in the vicinity of
Schumachers Gully and its tributaries, the basalt cap is absent. It is unknown however,
whether the young basalt is again present as the terrain rises at, and beyond, the
intersection of the Braemar palaeochannel with the palaeo-Swan Brook.

4.4.3 Implications of the Braemar deposit in the search for similar deposits in the zone-of-
overlap between the East and West Central Provinces

The sapphire-bearing Eocene/Oligocene sediments at Braemar are associated with

the basalt-filled Braemar palaeochannel which is a tributary of the palaeo-Swan Brook.
Similar sediments also exist immediately to the south of Braemar in the Dodds Hill
palaeochannel which also joins the Swan Brook palaeochannel. The deposits are protected
by a capping of younger basalt, and demonstrate that similar deposits may exist elsewhere
in the zone-of-overlap between the East Central Province and the West Central Province.
Owing to the limited number of ages available for the Central Volcanic Province, the
extent of the zone-of-overlap can only be approximately defined by the lines shown on
Fig. 4-8. The existence of similar "Braemar-type" deep leads further to the east is
dependent on the presence of younger basalts further east of Braemar in the Swan Brook
and other palaeochannels.
The ages of the volcanic rocks in the vicinity of Braemar are shown in Figs. 4-8
and 4-15. An age of 20.6 Ma. (Smith, 1988) at grid reference GR:389001 on the Elsmore
1:25,000 topographic sheet (Fig. 4-15) demonstrates that young basalt, and hence the
potential for deep lead "Braemar-type" deposits, extends eastward at least as far as this
location. Such deposits, if present, would exist on or close to the axis of the relief-inverted
palaeochannels, such as the flat-topped basalt ridge running along the axis of the
topographically inverted palaeo-Swan Brook (indicated by question marks on Fig. 4-15).
Therefore, the Windy Ridge near Bellview (GR:440050 Elsmore 1:25,000), Bald Hills
near Golden Grove (GR:470050 Elsmore 1:25,000) and possibly Table Top Mountain
(GR:503022 Elsmore 1:25,000) are worthy of more detailed investigation.
If "Braemar-type" sapphire-bearing deep leads exist in these areas, then the
presence of sapphire in the present-day lateral streams, Swan Brook and Kings (Newstead)
Creek (Fig. 4-15), may well be explained in part by the exhumation of such deep lead
deposits. This process is shown schematically in Fig. 4-19. This hypothesis may also
explain the occurrence of sapphires in soils (Mr. Doug Erry, of Golden Grove property,
pers. comm.) below a certain level on hillsides on the southern side of Swan Brook which
form the northern flank of the topographically inverted palaeo-Swan Brook.

4.4.4 Conclusions and exploration-mining problems

Potential sapphire-bearing deposits within the zone-of-overlap between the 32-38

Ma. East- and the 19-23 Ma. West Central Volcanic Provinces have been delineated in the
Braemar palaeochannel and its tributaries, up-palaeostream from the Braemar sapphire
mine. Potential sapphire-bearing alluvials are also delineated along the topographically
inverted palaeo-Swan Brook.
Deep lead deposits within the zone-of-overlap would be poorly exposed and may
be situated at a level well above the present day alluvials, as is the case at Braemar. The
only clue to their presence may be trails of sapphires and associated heavy minerals in
soils and recent alluvials downslope from the deposits.
The "Braemar-type" deep lead deposits require a capping of younger basalt for
their preservation. Hence, in any exploration program for deep leads, the first step is to
determine of the presence or absence of young basalts. This could be assessed by K-Ar
dating basalts along the relief inverted palaeochannels, such as along the palaeo-Swan
Brook at the sites mentioned earlier. If K-Ar dating were to yield young ages (19-23 Ma),
then deep leads may exist below these basalts. An attempt to distinguish between the older
and younger basalts, based on their chemistry, is not considered as useful an exploration
tool as K-Ar dating. Coenraads (in prep), using all available analyses for the Central
Province, found the compositional ranges of major and minor elements for basalts of the
West Central Province to largely overlap those of the East Central Province. The only
exception being the tholeiites found by Duggan (1972) in the vicinity of Inverell.
Once the extent of potential cover rocks has been established, the presence and
position of any palaeochannel axes may be identified using a palaeotopographic
reconstruction, such as that carried out at Braemar. In the vicinity of Braemar it was
possible to determine the course of the palaeochannel axes to an accuracy of ± 200m, or to
within ± 50m in areas of high drill hole control. The position of such palaeoaxes could
subsequently be appraised by geophysical techniques relying on detectable property
differences between the basement rocks of the palaeovalleys and the material filling them.
Gravity surveying has been tried in this regard, yielding an accuracy of the order of ± 50m
in other basalt filled palaeovalleys of the Central Province (Coenraads, 1989). It must be
noted however, that the course of potential sapphire-bearing intrabasaltic channels may not
conform exactly to the oldest palaeochannel axis, especially in wide valleys.
Ultimately, the presence of sapphire-bearing deep lead sediments and their
economic potential must be verified by costeaning or drilling which may prove to be the
economic restraint on exploration for this type of deposit. The more expensive exploration
and mining program must be supported by the commodity price. At Braemar, two lines of
holes are proposed (Fig. 4-16), comprising a maximum of 12 holes spaced at 50 m to a
depth of between 5 and 20 m, to test Eocene/Oligocene sediments below a capping of hard
basalt, and to determine the position of potentially sapphire rich placer deposits.
Exploration holes would have to be drilled in excess of 45 m in the Dodds Hill
palaeochannel.
An efficient mechanized mining operation can profitably recover grades as low as
5 grams per cubic metre (Mr. T.J. Nunan pers. comm.), from Holocene placer deposits, if
the quality of stone is high. However these deposits are mined at depths of 4 to 9 metres
from below soft alluvium using backhoe excavators and from areas whose grades have
been proved by large diameter-bucket drilling (Coenraads, 1990). In the case of deep lead
sapphire deposits, both hard basalt and excessive depth would preclude the use of cheaper
large-diameter bucket drills and necessitate diamond drilling yielding a much smaller
sample that cannot be used to estimate grade or quality.
 Chapter 5

HEAVY MINERAL SUITES WITHIN THE CENTRAL PROVINCE, NEW

SOUTH WALES

5.1 Introduction

Twelve suites of heavy mineral concentrates were analyzed, from sapphire placer
deposits and from in-situ soils in close proximity. The aim of this part of the study was to
shed light on the provinance of the heavy minerals and determine their behaviour during
weathering and subsequent concentration in alluvial deposits.
The sampling sites for heavy minerals were chosen to test for differences in
mineral constitution and composition within and between drainage catchments. Sampling
(Fig. 3-2 & Table 5-1) was undertaken within the catchments of Reddestone Creek (5
sites), Kings Plains Creek (1 site), Yarrow River (2 sites), Schumachers Gully/Swan
Brook (2 sites) and Dumaresq Creek/Gara River (2 sites). The locations of the sample sites
are shown in Figure 3-2. Samples were collected at or near the tops of hills from thin soils
directly overlying weathered basalt (in situ sites) and from river gravels (placer sites).

5.2 Preparation of heavy mineral concentrates

Soil samples of 5-10 kg were concentrated by sieving and panning. Gravels from
creeks and rivers were likewise concentrated or taken from the mining operations of T.J.&
P.V. Nunan Pty Ltd. Each concentrate was examined under a binocular microscope and
blue sapphires, together with red, orange or colorless zircons, bright-green clinopyroxene
and yellow-green olivine, were identified visually. Minerals chosen for detailed
investigation were those with the potential for isomorphous substitution, being those most
likely to show discernable differences between collection sites. These "black" minerals
were difficult to identify visually, but were abundant in both the soil and placer sites.
Representative samples from each concentrate were selected, mounted in plastic discs,
polished and analyzed using an electron microprobe. At least two analyses were carried
out on each mineral grain to check for homogeneity and the presence of zoning. Only if
the analyses were sufficiently similar, was the average taken and used for data comparison.
The majority of the mineral grains analyzed proved to be homogeneous, with the
exception of a large number of ilmenites from the Reddestone and King Plains catchments
which showed irregular variation across individual grains.

5.3 Heavy minerals present in the Central Volcanic Province.

The minerals in the concentrates and their compositional ranges are listed in Tables
5-1 & 5-2. The problem of interpreting the large number of mineral analyses was
approached by, firstly, characterizing the minerals for the entire volcanic province, and
then, secondly, checking for subtle differences between individual sites.
The minerals studied were subhedral to rounded. Sapphires and zircons are often
glossy in appearance and spinels and ilmenites are usually pitted and dull. Detailed
scanning electron microscopic examination of the grain surfaces revealed little evidence of
abrasion due to fluvial transport, and indicated that pitting and etching are the result of
surface biodegradation, which was particularly pronounced on spinels (Bischoff and
Coenraads, in prep.). Rounded and glossy mineral surfaces are attributed to disequilibrium
and resorption in the carrier magmas (Coenraads et al, 1990).

The spinel-group plots in four parts of the magnetite and ulvospinel prisms shown
in Figure 5-1. Their compositional features are:

1) Chromium-spinel group: These trend from the spinel field towards the chromite-
picrochromite field. Cr2O3 ranges from 8 to 23 percent with FeO enrichment and MgO
depletion corresponding to Cr2O3 enrichment. One sample from the breccia dyke at
Braemar contains 39.3 percent Cr2O3.

2) Pleonaste Group: These lie along the spinel-hercynite join from 52 to 78 percent end
member spinel, and are essentially chrome free (Cr2O3 < 0.52%).

3) Titanium Magnetite-Ulvospinel Group: These trend towards magnetite in the magnetite

prism and plot near the ulvospinel end member in the lower prism. They have low
Al2O3 (1-8%), MgO (0.5-6%) and are chrome free (<0.06%). Plots nearer the
ulvospinel corner have slightly higher Al2O3 and MgO contents.

4) Magnetite Group: The few magnetites found plot close to the magnetite end member
and have lower TiO2 than the titanium-magnetites (maximum 6.5%).

The ilmenites for the Central Volcanic Province contain up to 8 percent MgO, and
MgO is highest in those ilmenites with high TiO2 and low FeO.
Rutile crystals with ilmenite rims have been found in heavy mineral concentrations
from the Uralla area (Coenraads et al, in prep). Gem quality pyrope-almandine series
garnets have been obtained from Horse Gully in the New England Gemfields (Coenraads
and van der Graaf, 1991). These apparently localized and uncommon heavy minerals are
described in sections 5.5 and 5.6
5.4 Comparison between heavy mineral suites from collection sites within the
Central Volcanic Province.

To enable detection of subtle differences between individual collection sites, each

mineral species from a particular site was plotted alongside similar species from the other
sites (Figs 5-2 to 5-5). The data for each site is arbitrarily arranged in ascending or
descending order for a particular major oxide so that the relationship between the major
oxides and minor oxides can be seen at a glance.
The figures indicate appreciable variations in weight percent oxide abundance as
well as some systematic relationships between oxides as follows:-

1. Chromium-spinel ; (Fig. 5-2) decreasing Al203 corresponds to increasing Cr2O3 and

MnO, decreasing MgO, and weakly decreasing NiO.

2. Pleonaste ; (Fig. 5-3) increasing FeO (total) corresponds to increasing TiO2, and
decreasing Al2O3 and MgO.

3. Titanium-magnetite; (Fig. 5-4) increasing FeO (total) corresponds to decreasing Al203,

MgO and TiO2.

4. Ilmenite ; (Fig. 5-5) increasing FeO corresponds to decreasing MgO and TiO2.

The breccia dyke at Braemar (Fig. 4-14) described by Pecover and Coenraads,
(1989) yielded 22 chromium spinels (including one grain with 39.3% Cr2O3), five
ilmenites and one pleonaste. The minerals from the dyke showed a slight core to rim
zoning with enrichment in FeO (2.77% in pleonaste, up to 0.37% in Cr-spinel, and 1.46%
in the ilmenite). The chromium-spinels also showed an outward enrichment in Cr2O3 (up
to 0.77%) and both the chrome spinels and pleonaste, a depletion in Al2O3 (1.56% in the
pleonaste and up to 1.47% in the chrome spinels). The breccia dyke and various basalt
flows are believed to be responsible for contributing to the alluvial heavy mineral deposits
worked for sapphire at Braemar.
Each mineral species shows a remarkable similarity, both in ranges for given
oxides and in relationships between oxides, for the sample sites. There are no significant
differences between mineral compositions from the various drainage systems. However,
some sites differ in the presence or absence of certain mineral species, and/or in the
relative mineral abundances. Corundum and zircon appear to be common associates.
5.5 Ilmenite mantled rutile crystals from the Uralla district

5.5.1 Occurrence.

Ilmenite-mantled rutile crystals of enigmatic origin occur in Late Tertiary

conglomeratic arkoses and Quaternary alluvium west of Uralla, New South Wales (Figure
5-6).
The dense, dull-black mantled crystals ranging from 0.5 to 4 cm in length are found
within all types of arkose (Tca on map, Figure 5-6) and are always concentrated, by virtue
of their specific gravity, with other heavy minerals within conglomerate bands. They also
occur in the Rocky Creek, Kentucky Creek and the upper Rocky River drainages, being
profusely abundant in areas where concentration of heavy components in gravel is high.
(Qa, on map, Figure 5-6). Other heavy minerals found with them include zircon, spinel,
ironstone pisoliths, tourmaline and rare corundum and diamond.
The mantled crystals were briefly described by David (1886) as "Titaniferous Iron"
comprising part of the heavy mineral assemblage found in Recent alluvium along with
zircon, spinel and quartz. He also reported that "nodules of titaniferous iron are abundant
at Wallaby Gully, where a pebble of that mineral has been found 2 inches in diameter"
(GR 509108 9136-I-N Balala 1:25000).
The ilmenite-rutile crystals have not been found to occur in any igneous rocks, or
metamorphic rocks which outcrop in the Uralla area. Nor have they been found to occur in
covering basalt soils or Tertiary deep-lead deposits (Coenraads et al, in prep). Their source
is therefore problematic.
These crystals are of interest because of their restricted occurrence and their physical
resemblence to crystals known from kimberlite pipes (Mitchell, 1979).

5.5.2 Description of the ilmenite-rutile crystals.

The crystals are shown in Figure 5-7. They are stumpy to cylindrical in shape and are
always rounded. This is probably due to magmatic corrosion with only minor modification
due to transport. A core of rutile is exposed below the mantle of ilmenite if the specimen
has been broken or is cut. In thin slices the rutile is a deep translucent red.
Four crystals were selected for detailed study. Polished blocks were prepared for
electron microprobe analysis and thin sections were cut parallel to, and perpendicular to,
the long axis of the cylinder, (photomicrographs shown in Figure 5-8). Uniform extinction
parallel to the long axis of the cylinder (Figure 5-8d), indicates that the rutile cores are
single tetragonal crystals with the C-axis parallel to the direction of elongation of the
cylinder. In reflected light the ilmenite mantle comprises a number of subgrains. Figure 5-
8c shows the interface between the rutile core and ilmenite mantle. Blades of ilmenite
interlock with those of rutile thereby creating a large surface area of contact. The ilmenite
appears to be replacing the rutile crystals and its planes of penetration appear to be
controlled by the rutile crystal structure. The planes are exactly 90° apart and are
penetrating at right angles to the prism faces, that is, in the (100) direction. Also apparent
in Figure 5-8c are the good (110) cleavage planes which are free of ilmenite replacement,
so possibly opened during the thin sectioning procedure. In Figure 5-8d the dark ilmenite
blades are parallel to the C-axis of the rutile crystal, confirming replacement along the
{100} planes.

Electron microprobe results: Analyses approximately 1 mm apart, were performed on the

polished sections (cut normal to the c-axis) in transect lines running from rim to core and
then out to the opposite rim. Initially the full spectrum of each representative mineral was
examined using Macquarie University's Link 1000 Energy Dispersive Analytical System
and analyses of the elements detected were carried out using the Siemens wavelength
dispersive system.
Four analyses near each desired location were averaged for plotting, and the bars in
Figure 5-9 indicate the calculated standard deviations.
The rutile cores contain 0.34 to 0.44% FeO(total), but there is no discernable
variation in FeO across the cores indicating them to be homogeneous crystals.
The variation in ilmenite compositions are 50.5-55.7% TiO2, 40.3-45.0% FeO, 1.9-
3.0% MnO and 0.3-1.4% MgO. In individual samples TiO2 increases and FeO decreases
by about 1% from the outside to the inside of the rim. Mn shows no variation. In some
ilmenite rims MgO increases towards the core, with the highest values recorded within the
ilmenite replacement blades within the rutile core (1.4%, not shown on Figure 5-9).
A detailed transect of 11 data points, 0.3mm apart, was made across one section of
the ilmenite rim in grain 3 (Figure 5-9). The overall fall in TiO2 and a rise in FeO from the
rutile-ilmenite contact to the outer edge is neither smooth nor gradational. However the
changes between successive points generally lie within the error bars.

5.5.3 Comparisons with ilmenite-rutile crystals and ilmenite found elsewhere

The main elemental oxide contents of the Uralla ilmenite relative to published
ilmenite compositions from many different rock associations is presented in Table 5-3 and
summarized in Fig. 5-10. This highlights their somewhat unusual composition and helps to
suggest potential sources.
Discrete crystals of rutile commonly mantled by magnesian ilmenite are found in
heavy mineral concentrates from some kimberlites, such as at Somerset Island, Canada
(Mitchell, 1979). The rutile is homogeneous but the ilmenite may be compositionally
zoned with increased Mg, Cr and Mn towards the edges. Dawson (1980) proposed that
fragmented eclogite xenoliths are the probable source of the rutiles prior to their mantling
with kimberlitic ilmenite. Samples from the Wesselton Pipe, South Africa (Mitchell, 1973)
show that the ilmenite is a replacement of the rutile. Kimberlitic ilmenites are well known
Footnotes to accompany Table 5-3.
1 This paper. 2 Carmichael (1967, p.140-2, Tab.2-4, analy.1-29). 3 Haggerty (1976,
p.Hg258, analy.9-18). 4 Wass (1973, p.430, table III, analy.1-17; p.434, Tab.IV). 5
Cundari (1973, p.478, Tab.VI, analy. BQ-PEGM, BEH-C, BEH-D). 6 Birch (1979, p.379,
Tab.6, analy.CLOG-44, CLOG-45, CLOF-53). 7 Birch et al. (1982). 8 F.L.Sutherland &
D.F.Hendry, unpublished data. 9 Haggerty (1976, p.Hg258 & 260, analy.1, 3-4, 6-8, 45).
10 Haggerty (1976, p.Hg259, analy.21-27). 11 Deer et al (1975, p.29, table 5, analy.3-4).
12 Haggerty (1976, p.Hg260, analy.41-44, 46-47). 13 Cawthorne et al. (1988, p.149,
Tab.1A, 21/1-15, 16/1-10, NGL/9,11, 54, 56, 62, 110). 14 Haggerty (1976, p.Hg260,
analy.48-54). 15 Simpson (1951, p.651, Tab.33, analy.1-6). 16 Deer et al. (1975, p.29,
Tab.5, analy.5). 17 Rumble (1976, p.R4, table R2, analy.1). 18 This paper. 19 Sutherland
(1980, v.2, Tab.6e, analy.1D). 20 Jaques et al. (1989, p.133, Tab.7.6, analy.7), 21 Cassidy
et al. (1988, p.1055, Tab.3, analy.1) 22 Coenraads (1990, p.1205, Fig.19, 55 analy.). 23
Coenraads (1990, p.1205, Fig.19, 45 analy.). 24 Hollis et al. (1983, p.191, analy. Bal.149,
Nan.Il, Mt.Mit). 25 Sutherland (1980, v.2, Tab.5g, analy.1-7). 26 Jaques & Perkin (1984,
p.35, analy.5). 27 Haggerty (1976, analy.1-15). 28 Dawson (1980, Tab.11, p.78, 302
analy.). 29 Haggerty (1976, p.Hg258, analy.13,15-18). 30 Atkinson et al. (1984, p.208,
Tabs.1&2, 6 analy.). 31 Neal & Davidson (1989, p.1980, Tab.5, 12 analy.). 32 Scott Smith
et al. (1989, p.196-7, Tab.11.3, analy.P6/2). 33 Clarke & MacKay (1990, p.231, Tab.1). 34
F.L.Sutherland, A.D.Robertson & J.D.Hollis unpublished data. 35 Wass et al. (1980,
p.338-9, Tab.1, analy.K41, K3, K13A1). 36 Harte et al. (1987, pp.187-188, Tabs.3&4, 4
analy.). 37 Moore (1987, p.248, Tab.1, 3 analy.). Oxides are in weight percent. For
comparison purposes all iron is presented as FeO and oxides are listed in decreasing order
of abundance for the Uralla samples.
for high MgO, (see Table 5-3E), with enrichment towards the edges because of magmatic
reaction with the more magnesian kimberlitic melt.
The Uralla rutiles, unlike cores of rutile in kimberlite, show reaction relationships for
a magma rich Fe but poor in Mg. Rutiles with replacement rims of ilmenite occur in
granulitic xenoliths from Australian basalts but these ilmenite rims are also more
magnesium-rich. The Uralla ilmenite compositions are compared with those from many
different rock associations, as well as those derived from the volcanic rocks of the adjacent
Central Province, in Table 5-3 and Fig 5-10. The Uralla ilmenites differ from the Central
Province and kimberlitic ilmenites, being significantly higher in MnO, lower in MgO and
having no Cr2O3.
In general there is too much MnO and/or too little MgO in the Uralla ilmenites to
match typical ilmenites in alkali basalts, high Mg-tholeiite and mafic-ultramafic suites,
including highly undersaturated melts from deep sources such as kimberlites, carbonatites,
picritic monchiquites, alnoites and lamproites, or those in mantle xenoliths and xenocrysts.
A rare exception is ilmenite inclusions in diamonds from Brazil, though low Mg ilmenites
are not found in most other diamond suites. The contrast in size is a marked point of
difference between the Brazilian and Uralla ilmenites which makes a related source
unlikely. Ilmenite attributed to immiscibility processes in kimberlite melts at crustal levels
(Clarke and MacKay, 1990) also differs from the Uralla compositions.
Uralla ilmenites are more akin to those found in salic igneous rocks. Comparison of
their compositions with those of silicic volcanics show only a partial overlap as the latter
are mostly lower in TiO2 and higher in FeO. Ilmenites from the intrusive equivalents,
granites and pegmatites are distinctly lower in TiO2, and those from less silicic intrusions
(syenite-monzonite-diorite-foyaite) barely overlap the Uralla range.
Metamorphic rocks rarely show ilmenite compositions within the Uralla range. One
is a crustal xenolith of ilmenite-garnet-amphibole-plagioclase-quartz granulite from
Western Australia and another crustal xenolith comes from low grade metamorphism of
mafic-ultramafic rocks (see Table 5-3). However the latter are not typical of the great
majority of ilmenites found in various grades of metamorphism of these rocks (Cassidy et
al, 1988).
Low MgO in the Uralla ilmenite might suggest a relatively low pressure origin, as
most high pressure ilmenites show high MgO contents. However, the amount of Mg
entering the mineral is regarded more as a function of melt composition rather than a
reflection of a depth or pressure effect (Haggerty, 1976).
5.6 Garnets from Horse Gully in the New England gemfields.

5.6.1 Occurrence

Deep red to purplish, gem quality garnets occur in sapphire bearing river gravels at
Horse Gully in the New England Gemfields. The garnets were recovered by Arrawatta
Sapphires Pty Ltd (J. Joris; Joris Gemstone Traders Pty Ltd, pers. comm.) within their
Horse Gully mining areas near the abandoned township of Sapphire, 27 kilometres
northeast of Inverell (Figure 5-11).
The garnets appear to be derived from a local source situated within the catchment
area of Horse Gully.

5.6.2 Features and properties of the gem garmets

The garnets range in size up to 15 mm, are irregular and rounded in shape, and
show a chipped and crazed water-worn surface. Some of the stones display a slight colour
shift from a red or purplish hue in natural light to an orange red or almost pink in
incandescent light. Seven garnets were polished to facilitate refractive index
measurements, to provide a window into the stone for viewing inclusions, and to provide a
surface for microbe analyses.
Three microprobe analyses of each garnet proved each to be homogeneous in
composition with any differences being less than one percent. The average composition of
each garnet and its gemmological properties are listed in Table 5-4. They are members of
the pyrope-almandine intermediate series ranging in composition from pyrope 55% -
almandine 34% to pyrope 39% - almandine 49%. The stones with the highest end-member
pyrope content (highest magnesium and lowest iron) display a most attractive mauve-
purple to pinkish hue.The garnets have a specific gravity of 3.837 to 3.958 and a refractive
index of 1.755 to 1.772. The relationship between these properties and the chemical
composition is clearly shown in Figure 5-12. A higher pyrope content and a lower
almandine content correspond to a lower specific gravity and refractive index, fewer
inclusions and generally a lighter, more purplish colour.
The stones all show anomalous double refraction under the polariscope which is
very strong in some cases (indicating a high degree of internal strain). The garnets are inert
under both short wave and long wave ultraviolet light. Inclusions include, short to long,
fine, oriented needles, which are probably rutile; short, colourless, oriented, cylindrical
inclusions; and colourless crystals ranging from subrounded in shape to well formed
hexagonal prisms. The latter are suspected to be apatite (Gübelin and Koivula, 1986), and
a handsome trail of such crystals were found in garnet HG6 (Coenraads and van der Graaf,
1991).
The spectroscope revealed characteristic Fe2+ bands at 573 nm, 520 nm and 504
nm with another weak band at 460 nm (stones HG2, HG5 and HG7). In stones HG1
&HG2, the 504 nm and 520 nm bands merge to form a broad absorption band and stones
HG1, HG2 & HG4 show a cutoff at the violet end of the spectrum. Based on their
gemmological properties, the garnets would be classified as belonging to the pyrope-
almandine intermediate series with HG5 being of sufficiently purple hue and light in
colour to be called rhodolite.

5.7 Implications for genesis of heavy mineral placer deposits.

The mineral suite comprising pleonaste, chrome-spinel, titanium-

magnetite/ulvospinel and ilmenite is well represented in both placer and basaltic soil
concentrates. These all occur as accessory minerals or as megacrysts in the Central
Province volcanic rocks (Binns, 1969 and Binns et al., 1970). Sapphire and zircon are also
usually present, and a genetic link between these two minerals and the volcanic rocks has
been established by U-Pb dating of zircon inclusions contained in sapphire (Coenraads et
al,1990). Ages of 34 Ma (at site DUN) and 36 Ma (at YAR) lie within the K-Ar age range
for basalts of the Central Volcanic Province.
Small systematic differences in major and minor oxide content of the heavy opaque
oxides in placer and soil samples from several drainage systems are not found to be
statistically significant. In spite of this, subtle visual differences in the nature of the heavy
mineral concentrates and in the colour, shape and size of the sapphires have been reported
by miners and sorters. For example, as mining proceeds along the length of a deposit
stones may vary from being predominantly parti-coloured in a particular area, to
predominantly blue with a blue cross-table, or to predominantly darker blue with a green
cross-table. Shape may vary from predominantly equant crystal fragments to slender
pointed hexagonal pyramids (dogs teeth). Sapphires recovered from one area of Kings
Plains were reported to be extensively internally fractured and jagged in appearance (K.J.
Walker, pers. comm.). Corundum recovered from Furracabad Creek, in the catchment
adjacent to Reddestone Creek, is largely black (T.J. Nunan, pers comm.). The ratio of
zircon to sapphire has also been observed to vary from place to place (K.J. Walker, pers.
comm.). These differences occur within the order of kilometres
Weathering of basalts and their volcaniclastic associates liberates all of the heavy
minerals found in in situ basalt soils and in placer deposits. The less resistant silicates
(olivine, orthopyroxene, clinopyroxene and amphibole) break down leaving the resistant
heavy minerals which are concentrated into placer deposits by fluvial action. Differences
in proportions of heavy minerals in placer deposits are obviously influenced by hydraulic
and comminution factors. However, it is likely that they reflect net differences in the
proportions of heavy minerals shed from various local basalt flows and/or nearby
volcaniclastics. For instance, the breccia dyke (BRE) at Braemar is believed to have
contributed partly to nearby placer deposits (Pecover and Coenraads, 1989). This, coupled
with the spatial variability in the character the sapphires and grain surface features which
indicate minimal amounts of fluvial transport, suggests local multiple sources and minimal
downstream reworking and mixing.
The concentration of sapphire to yield important deposits, such as those at
Reddestone and Kings Plains creeks, results from favorable hydrologic conditions,
including appropriate palaeostream energy and gradient, combined with the availability of
suitable source rocks within the catchment.

5.8 Conclusions and interpretation.

The analysis of heavy mineral concentrates from different catchment areas in the
Central Volcanic Province has led to a number of interesting findings. These are:

1) The only ubiquitous associate of sapphire is zircon. It is likely that both formed
under similar conditions based on ages of zircon inclusions in sapphires (34-36 Ma) and
related ages of older basalts from the Central Province (32-38 Ma). Abundant quantities of
spinel or ilmenite are not necessarily indicators for sapphires or zircons, but do indicate
that fluvial concentration has been successful.

2) In situ soils from basalts and their volcaniclastic associates yield suites of heavy
minerals that contain various combinations of sapphire, zircon, pleonaste, chrome-spinel,
titanium magnetite-ulvospinel and ilmenite. Less resistant heavy minerals (olivine,
pyroxene, and amphibole) were found only in basaltic soils and immature alluvials.

3) From all the in situ soil sites visited, only one small zircon was recovered by the
author (UNE). One large sapphire was found by a property owner (HIL) and small
quantities of sapphire and zircon were reported to have been recovered from the red
breccias at Braemar. This scarcity of sapphire and zircon demonstrates the enormous
amount of concentration required to produce the economic placer deposits.

4) Placer concentrations of heavy minerals typically contain most members of the

suite, although proportions of individual minerals can vary appreciably. This variability is
believed to largly reflect mineralogical variation in local alkali basaltic and volcaniclastic
sources for the placer concentrations. Visual characteristics of the sapphires also change
locally.

5) The abundances of major and minor oxides in pleonaste, chrome-spinel,

titanium-magnetite and ilmenite in both soils and placer deposits, within and between
drainage systems, are remarkably uniform. Subtle differences are not significantly clear to
be of use in exploring for sapphire deposits in the Central Province.
 6) The gem quality pyrope-almandine garnets from the New England gemfields are
very local, being found with sapphires only in Horse Gully. For this reason they are
probably not derived from the widely distributed sapphire bearing volcanic rocks of the
Central Province. They may be derived from a small isolated diatreme-like intrusion
similar to those which produce gem quality garnets at Ruby Hill near Bingara , N.S.W.
(Lovering, 1964), at Brigooda and Ballogie, near Proston, Queensland (Hollis et al, 1983;
Sutherland et al, 1990) and Bogie River, north Queensland (Jones, 1983) . However, the
Horse Gully garnets show significantly lower CaO and MgO and higher FeO than these
garnets.
The Horse Gully garnet source is located within the 18 km2 catchment area of
Horse Gully and Mary Ann Creek (Figure 5-11). Possibly it remains hidden by basalt that
also fills the, now extinct, White Hill palaeochannel presently being exhumed by Horse
Gully and Mary Ann Creek. The position of the palaeochannel is indicated as a series of
dots on Figure 5-11.
The Horse Gully garnets are quite different to the spessartine-almandine garnets
found in the diamond-bearing Tertiary deep leads at Copeton (D. Jones pers. comm.), and
also different to the garnets derived from acid to intermediate igneous sources as these
contain little CaO (0.5-1.5%) and MgO (1-6%) (W. Birch pers. comm.). It is thus most
likely that the Horse Gully garnets are derived from a mafic source.

7) The source of the ilmenite mantled rutile crystals has not yet been identified. The
crystals are found, concentrated by virtue of their size and weight, within conglomerate
bands in arkoses of possible late Tertiary to Quarternary age. They also occur in high
concentration in the modern streams draining the arkoses; however, they are seen to be
fragmented within a short distance downstream. All this implies a proximal source.
Other heavy minerals associated with the arkoses include zircon, spinel, gold and
sapphire, but areas of high concentration of a particular mineral, such as zircon, do not
always correspond directly to areas of high ilmenite-rutile crystal concentration.
No metamorphic rocks (Sandon Beds and Dummy Creek Group) exist in the areas of
highest concentration of the crystals so are unlikely to be the original source. Panned
concentrates of numerous samples of sub-basaltic Tertiary gravels, chiefly from mullock
heaps around shafts and from rare outcrops, contained no ilmenite-rutile crystals. Panned
soil samples from around basalt outcrops produced no crystals and none were seen in
hand-specimens of a wide variety of basalts from numerous locations around Uralla.
Owing to the restricted distribution of the crystals, the host rock possibly had a
restricted outcrop area, thus excluding the Uralla granodiorite, and it must be local because
of the immaturity of the arkoses and the inability of the crystals to withstand fluvial
transportation. The chemical comparisons, discussed above, also indicate that the source
rock is unlikely to be mafic or ultramafic in composition. Possibly the host rock was easily
weathered and hence difficult to recognize, or has been covered by younger deposits.
 The summary of the main elemental oxide contents of the Uralla ilmenite relative to
ilmenite compositions in the literature (Figure 5-10) highlights their somewhat unusual
composition and helps to suggest potential sources. The Uralla TiO2, FeO(total), MgO and
MnO contents only intersect four of the thirty two ranges listed from the literature (salic
volcanics, syenites-foyaites, coarse phase leucitites and the quartz granulite xenolith).
One possible source is a magmatic precipitate from a magnesium poor, iron and
titanium enriched melt. If there is a link with the abundant near-source zircon crystals in
the district, then these minerals would represent late-stage or cavity crystallizations from
fractionated felsic magmas in subvolcanic chambers or conduits before eruption. This
would be compatible with ilmenite compositions overlapping those of salic volcanics.
Zircons from the Rocky River give mid to late-Tertiary ages of formation and eruption
(red to orange crystals, 27 to 3 Ma; F.L. Sutherland and P.D. Kinny, unpubl. data). This
relates these zircons to the Tertiary volcanic activity in the area. Such an association for
these minerals could account for their apparent absence from the sub-basaltic Tertiary
sediments.
Another possible source may be a course grained metamorphic or vein rock related
to the obscured Uralla Granodiorite contact in the area. The least likely association is as an
indicator for diamond source rocks.

The present study of the evolution of the Central Volcanic Province and its
drainage patterns has provided valuable insight into the occurrence and mechanisms of
concentration and distribution of diamond and sapphire deposits. Diamond bearing deep
leads occur below the lavas of the West Central Province and sapphire bearing alluvial and
deep lead deposits are associated with the East Central Province. In particular, the present
study allows the prediction of possible extensions to known deposits and identifies
potentially favorable areas for future exploration.
The absence of less resistant minerals and relative abundance of sapphires and
zircons in the placer deposits indicate a high degree of concentration. Variability in the
mineral species, and character the sapphires within short distances, however, suggests
local sources. Grain surface features also indicate minimal amounts of fluvial transport.
This, combined with the low gradient and low energy drainage systems suggests that
reworking occurs largely in a vertical sense. The fine or light material is "winnowed out"
leaving a heavy residue. A rich upstream source, in contrast, would yield a constant suite
of minerals of uniform appearance downstream with more evidence of fluvial abrasion.
These observations are of vital importance in exploring for new sapphire deposits.
They demonstrate that a search for elusive rich sources is inappropriate. Rather, the
exploration effort should be directed towards gaining a thorough understanding of the
geomorphic processes involved in the development of economic placer deposits from low-
grade source rocks.
 Chapter 6

EVALUATION OF SAPPHIRE SOURCE ROCKS AND POTENTIAL

SOURCE STRUCTURES IN THE CENTRAL VOLCANIC PROVINCE.

6.1 Introduction

This chapter examines the natural lagoons found on the lavas of the Central
Volcanic Province. These enigmatic features have been described as maar volcanoes by
Pecover (1987) who also proposed that they are possible sources of sapphire-bearing
volcaniclastic rocks. They thus represent potentially important economic targets.
Sapphire source-rock compositions were studied and compared with those from
sapphire-bearing and sapphire-barren provinces. If difinitive differences exist, these could
be used as indicators in sapphire exploration programs both for new gem fields and for
high grade areas within existing gem fields.

6.2 Evaluation of the natural lagoons of the Central Volcanic Province -Are they
sapphire-producing maars ?

Lagoons of the Central Volcanic Province are roughly circular to amoeboidal in

shape and range from less than 100 metres to several hundreds of metres in diameter. They
are usually flat swampy areas or lakes surrounded by a bank up to several metres high. The
sediment in the lagoons is a grey clay and the rims or banks consist of ironstone pisoles,
either cemented or loose, within a red clay matrix. The rims are altered to the extent that
no relict primary textures are observable. In all the lagoons examined, basalt is present in
portions of the rims where it occurs as flows or discrete pebbles, cobbles and boulders
within the clay and pisolitic material. These are usually the steepest and highest parts of
the rims.
Maars are described by Ollier (1967) and Ollier & Joyce (1976). Typical maars in
Victoria are circular depressions usually containing lakes or swamps. They are "explosion
craters extending below the general ground level and surrounded by a low rim of
pyroclastics, including some country rock. The pyroclastic rim is steep on the inside and
very gentle on the outside, merging into the surrounding plain. The craters are 500 m to 2
km across with walls up to about 20 metres high, or higher in exceptional cases such as
Bullenmerri. Ash is often asymmetrically distributed by wind, with high walls on the east
side and low or no walls at all on the west" (Ollier & Joyce, 1976). They are occasionally
filled with sediment and have flat crater floors. However, the maars of Victoria show
internal structure in the form of characteristic bedded pyroclastic material in their rims and
do not have associated ferricrete.
 Six natural lagoons in the Central Volcanic Province were chosen for the
investigation. Their locations are shown in Fig. 6-1 and tabulated in Appendix 3. Figures
6-2, 6-3, 6-4, 6-5 and 6-6 are aerial photographs of the lagoons:
Dunvegan Lagoon and Kings Plains Lagoon, have been examined geophysically
(Appendix 3) and bulk tested in order to ascertain their economic potential. Kings Plains
Lagoon has been drilled extensively (Lishmund, 1987; Coenraads, 1988b). The
relationship of these features to the underlying valley fill lava flows and Permian basement
topography has also been examined at a scale of 1:25,000.
The geology in the vicinity of the lagoons appears in Figures 6-7, 6-8, 6-9, 6-10 &
6-11. The present edge of the basalt flows is shown as a solid thick line and areas of
exposed pre-volcanic basement rocks can be seen. The contours indicate the interpreted
pre-volcanic topography; solid contours are used where the the basement is exposed and
dashed contours indicate the interpreted basement below the basalt. The topography has
been determined by plotting a series of elevations for the mapped boundaries between the
Central Province volcanics and the older basement rocks. The data points have been
contoured conservatively, that is, in areas of basalt cover the basement is interpreted to dip
gently and smoothly underneath and the least amount of basalt cover was inferred in areas
of poor control.
The cross sections (see Figure 6-12) show the relationship of the lagoons to the
present surface topography, to the valley flow basalts, and to the Permian basement
channels. The interpreted Permian basement is shown dotted, and indicates that the
channel axis is roughly below the surface lagoonal feature in every case and that there is at
least 10 metres of valley flow basalt below each feature. Drilling at Kings Plains found the
valley axis to be under the feature at a depth of 46 metres. At least two lava flows,
separated by a woody horizon, were seen in holes KP3, KP4 and KP1.
Gravity data at Dunvegan support a similar model of a basalt filled channel with its
axis beneath the centre of the feature. The gravity data indicate that the axis of the channel
is trending towards Clarevaulx lagoon.
At Dunvegan sub-economic to marginally economic concentrations of sapphire
were found by bulk testing in the rim, on the flanks of the rim, and below the lacustrine
clay within the lagoon. No sapphires, however, were recovered during the investigation of
Kings Plains Lagoon. It is noteworthy that abundant spinel and the occasional sapphire
may be easily "specked" by eye from around the edges of Dunvegan Lagoon, whereas
spinel is not seen around the edges of the Kings Plains feature, nor on the rims of John
Ryall Lagoon and Barley Field Lagoon.

6.3 Detailed work at Kings Plains and Dunvegan lagoons

The following points describe chronologically the exploration work carried out at
Kings Plains and Dunvegan lagoons.
1) A test trench 180 metres long was cut to a depth of 7 metres by Jingellic Minerals N.L.
to test the material on the flanks of the Kings Plains structure. The location of the
trench is shown on Figure 6-13. No sapphires were found in the concentrate.

2) Diamond drill holes KP1 and KP2 were designed by the Department of Minerals and
Energy to test if the Kings Plain feature is a maar (Lishmund, 1987). KP1 was drilled in
the geometric centre of the feature and KP2 on the rim as shown on Figure 13. The
holes encountered basalt flows to a depth of 44 metres overlying basement. The core is
located at the Department's core library at Londonderry.

3) Geophysical surveys (magnetics, gravity and seismic reflection; see Appendix 3) were
carried out for T.J. & P.V. Nunan Pty Ltd in order to assess the subsurface structure of
the Kings Plains and Dunvegan features and provide a basis for the location of any
future drill holes. The surveys covered an area of one square kilometre (the extent of
Figs. 6-13 and 6-14) and models were constructed using the information from the Kings
Plains drilling data. The data are held by R. Coenraads at Macquarie University.

4) Two tests pits (numbered 1 and 2 on Fig. 6-13) were opened by Jingellic Minerals N.L.
to test the economic potential of the rim material at Kings Plains. Eight cubic metres of
material from the surface to a depth of 1.5 metres were removed and processed in a test
plant. Only trace amounts of sapphire and zircon were found in pit 1 (0.3 grams/cubic
metre) and none in pit 2. No spinel or ilmenite were recovered.

5) A line of 15 shallow bulk test holes (numbered on Fig. 6-14) were drilled across the
Dunvegan structure by T.J. & P.V. Nunan Pty Ltd in order to test the economic
potential of the near surface material. A Cauldwell rig with a bucket diameter of 1 m
was used. Sapphires were found in almost all holes with some good grades. Grades of
13.0 g/m3 at 2.1-2.4 m. in hole 14, 10.7 g/m3 at 0.5-1.2 m. in hole 4 and 9.5 g/m3 at
2.4-2.7 m. in hole 13 were recorded.

6) Percussion holes KP3 and KP4 were drilled by Jingellic Minerals N.L. in the most
likely position to encounter a vent below the Kings Plains feature based on the
geophysical data (see Fig. 6-13). The drill cuttings are held by Jingellic Minerals N.L.
in Inverell and the logs are described in Appendix 4, (Coenraads, 1988b). The drilling
programme was designed to test the following:
(a) To test for the presence of any volcaniclastic material within the crater that may
have undergone some secondary reworking and contain economic sapphire
concentrations;
(b) To test for the possibility of a vent structure or breccia pipe in the eastern and
northern portion of the crater based on the geophysical data (KP3 was positioned on
the gravity high according to a geophysical model in which the pipe material is
 fresher than the surrounding rocks and KP4 on the gravity low according to a model
in which the pipe material is highly altered);
(c) To determine the depth to the basement rocks (Permian silicic volcanics);
(d) To examine the nature of the contact between the Permian basement and overlying
basalt flows and to determine whether any alluvial deposits encountered in this
position are economically prospective for sapphires;
(e) To take the opportunity to examine the concentrate from large quantities of crushed
basalt and see whether any indication can be found of sapphires originating from
this material.

6.3.1 Discussion

The lack of internal structures and unaltered rock types in the lagoon rims makes it
impossible to identify these features positively by field inspection alone. Based on the
observational data available at the time, (circular shape, raised rim, arcuate drainage
around the features etc), the description of these features as maars, (Pecover, 1987),
seemed justified.
In light of the failure to locate feeder pipes or brecciated material within the
structures, using geophysical methods and a subsequent drilling programme, a new model
is required. The present model proposes filling of pre-Tertiary basement channels with
lava flows as illustrated in Figure 6-15 (A-D). Basalt lavas flowed down these channels
(diagram A) and filled them to a uniform level (diagram B). On cooling, shrinkage
occurred, tending to leave the deeper centre sections at lower levels. The creation of a
broad flat river channel (diagram C) led to the development of an extremely low energy
system of interconnected lagoons and swamps. Lagoons and swamps occur in lower lying
areas of the broad flat-floored valley associated with the old channel axis. The geophysical
modelling (Appendix 3, Figs. A3-21 and A3-22), suggests that the lagoons are located
over the old drainage axis, possibly in areas where the palaeochannel was slightly wider
and the basalt surface slightly lower.
During dry periods the lagoons and swamps dry out and if there is a strong
prevailing wind, the floor of the lagoon may be scoured to the level of the watertable.
Transported clays and fine particles form a smooth crescentic dune, or lunette on the
downwind side (Fig. 6-15, diagram D), leaving a cliffed margin on the lee side. The
process is similar to that described by Mabbutt (1977) for seasonally dry saline playas.
This process will occur selectively within the lagoons where there is no vegetation binding
the soil. Eventually, when the lunette reaches a sufficient height, the lagoon will become
isolated and the throughgoing streams will be diverted around them. The similarity of the
position and shape of the pisolitic lagoon rim or lunette with respect to the cliffed basaltic
portion of the rim for the cases studied, as shown in Figure 6-16, suggests that a regional
southwesterly palaeowind direction has been responsible for these structures in the New
England region. Furthermore, the positions of the lunettes on the northeast to easterly
margins of the lagoons are consistent with the positions, observed by Mabbutt (1977), for
the late Pleistocene and Recent clay lunettes occurring south of 29°S in Australia. The
alternate filling and drying of the lagoons, with associated movement of water through the
lunette material, has led to the development of pisolites and the destruction of all internal
structures in the New England area.
Lagoons with downwind lunettes occur along the entire length of the Great Divide
from the Cooma region to northern Queensland (Ollier, 1982; Ollier, 1976), and some are
found in valleys where the floor consists of material other than basalt (Ollier, pers. comm.
1988). Ollier (1979) suggested that these lagoons are formed by tectonic activity, with
uplift of the Eastern Highlands levelling formerly westerly flowing streams and thereby
causing the drainage to become essentially stagnant.
The aim of the geophysical surveys and drilling was to provide data which would
test the hypothesis that the observed topographic features are maars and therefore possible
economic sources of sapphire. This work has provided no evidence that these lagoons are
maars and a preferred explanation, consistent with the present observations, is that they are
wind blown features. Sufficient work has been carried out to indicate that most, if not all,
of these features do not themselves constitute an economic target.

6.4 Sapphire-source rock evaluation in the Central Volcanic Province

To find areas likely to contain sapphire-source rocks and evaluvate their potential
as indicators in sapphire exploration, it was necessary to:

1. Differentiate the Central Volcanic Province into water catchment areas along drainage
divides or interfluves.

2. Identify individual catchments, based on those drainages from which sapphires have
been recovered, as potentially economic (i.e. worthy of further exploration) or non-
economic. This allows definition of exploration targets using the boundaries identified
in point 1, for both alluvial deposits and their potential source rocks.

3. Evaluate potential sapphire source rocks within one of the catchments defined as being
potentially economic. Major element and trace element data, together with field data,
were used to test for differences between sapphire-bearing and sapphire-barren parts of
the Central Province.

4. Compare lava compositions of the Central Province with those of other sapphire-
bearing and sapphire barren volcanic provinces in northeastern Australia and
southeastern Asia.
6.4.1 Watershed analysis of the Central Volcanic Province.

The watershed analysis was carried out over the Central Volcanic Province as
outlined in Figure 6-17. Lines following the highest topographic points or water divides
were drawn on the 1:100,000 scale Inverell, Glen Innes and Guyra topographic sheets to
delineate the catchment areas shown in Figure 6-18. Each catchment is named after the
principal creek or river flowing within it. Material cannot be moved across water divides
by the normal processes of fluviatile erosion; it can only move within its watershed
downslope from its original location. Conversely, the source rocks for a sapphire deposit
located within a particular catchment must be, or have been, located upslope within that
catchment area.
In some areas source rocks may no longer exist in a catchment area. At a Public
Fossicking Area, formerly a mining area, (grid reference GR:630188, Glen Innes 1:25,000
sheet), sapphires and zircons occur in alluvium trapped in cracks and crevasses in Permian
granodiorite whilst former basaltic and/or volcaniclastic rocks have been completely
eroded.
The watershed analysis was used to define distinct catchment areas which could
then be superimposed on maps showing economic sapphire deposits (Fig. 6-18).

6.4.2 Definition of specific catchment areas for sapphire exploration.

Although sapphire is ubiquitous in almost all East Central Province drainages, only
some contain mineable to very rich deposits. Using this knowledge (MacNevin, 1977;
Tom Nunan and John McPhee pers. comm., 1988), drainages with significant sapphires
were identified by Coenraads and Lawrence (1989). The geology and setting of the
alluvial sapphire deposits, and in particular the important Kings Plains Creek and
Reddestone Creek deposits currently being worked by T.J. & P.V.Nunan Pty.Ltd, were
discussed by Coenraads (1990). Figure 6-18 defines the extent of the catchments of these
economic deposits.
Four catchment areas within the Central Volcanic Province were identified as
economic, and worthy of greater exploration for alluvial deposits and their potential source
rocks, i.e. the Frazers, Kings Plains, Reddestone and Marowan catchments.

6.4.3 Source rock evaluation program

6.4.3.1 Sampling
(i) Basalts: Mt Buckley summits the water-divide between Kings Plains Creek, to
the north, and Swan Brook, to the south (Figs. 6-18 & 6-19); both catchments are mined
for alluvial sapphires. Mt Buckley is therefore an obvious place to sample a representative
cross section of Central Province basalt types, which may contain sapphire source rocks. It
also exposes the thickest vertical section.
 Thirty one samples (Fig. 6-19) covering some 300 metres of vertical section were
collected from locations separated by obvious breaks in slope. These breaks in slope were
interpreted to separate individual flow units. From sample B1 at the top of Mt. Buckley
(Matheson Trig Station elevation 1168 m GR:547077 Inverell 1:100,000), the traverse
followed the ridge down to the north-west, around the head of Chinamans Gully
(GR:525100) then south along a spur to Vol Ashton Hill (GR:523069), and finally down a
spur to the lowermost flow, B27, resting on basement behind "Mindora" shearing sheds
(840 - 860 m elevation, GR:519059). A further 4 samples B28 to B31 were collected from
another spur off Vol Ashtons Hill, GR:524064, with the lowermost flow, B31, also resting
on basement.
While none of the basalts have been dated, the Mt Buckley flows are most likely to
be of similar age to those at Spring Mountain, 8 km to the southeast of "Mindora". The
latter were dated by Cooper et al (1963) and are 35-36 Ma (Fig. 6-17). Younger ages of 21
Ma (Smith, 1988) and 23 Ma (Coenraads et al, 1990) exist 15 to 20 km to the west of
"Mindora", however these are at a lower elevation and therefore it is unlikely that there are
any younger flows in the Mt Buckley sequence.
(ii) Volcaniclastic Rocks: Whole-rock chemistry of the volcaniclastic rocks, as
potential sapphire sources in the economic catchments, was not considered appropriate
because all observed volcaniclastic rocks are entirely weathered to clays. Additional
complications which would make any conclusions highly equivocal include; unknown
percentage contamination by country rocks during explosive implacement; unknown
original lava type; and unknown amount of reworking.
Barron (1987) presented chemical analyses of these rocks and has attempted to unravel
their origin. Observations on volcaniclastic rocks occuring at Braemar are discussed by
Pecover and Coenraads (1989), and their residual heavy minerals are described by
Coenraads (1990).
No volcaniclastic rocks were encountered in the Mt Buckley area.

6.4.3.2 Analytical methods:

The basalt samples were sawn to remove all weathering surfaces, cracks and
fractures. Each sample was sawn into slices, from which a representative thin section was
made, then broken and crushed in a tungsten carbide N.V. Tema mill. Sources of
contamination, such as xenoliths, megacrysts and vugs were removed prior to crushing.
Major and trace element abundances were obtained by X-ray fluorescence
spectrometry. Fused borate buttons were used for the major elements and pressed pellets
were used to determine 15 trace elements. All samples were prepared in duplicate.
Measurements were made on a Siemens SRS1 sequential X-ray spectrometer with a
Siemens Kristallaflex 800 X-ray generator.
FeO analyses were made using the technique of hydrofluoric acid digestion and
titration with cerric sulphate. The volatiles H2O-, H2O+ and CO2 were determined by the
fusion and collection method using a Leco induction furnace.
 Chemical analyses of the Mt Buckley basalts are presented in Table 6-1

6.4.3.3 Treatment of the analytical data.

In order to make meaningful comparisons between rock chemistries, the raw
analytical data were treated as follows.
(a) The maximum Fe2O3/FeO ratio was set at 0.2. If the original ratio exceeded this value
it was recalculated to be O.2. Although post eruptive alteration of the volcanic rocks
may have caused some oxidation of FeO to Fe2O3, this value is somewhat arbitrary, as
discussed by Wass (1980), due to the uncertainty of the pre-eruptive Fe2O3/FeO ratio.
(b) Samples showing post-eruptive alteration were interpreted with caution. Alteration was
identified in thin section (see Appendix 5), or determined to be present from the
chemical analyses as high H2O+ (>4%) or high CO2 (>0.5%) indicating presence of
secondary minerals such as zeolite or calcite, and high Fe2O3/FeO ratio (>0.75)
indicating oxidation of ferrous iron. The Mt Buckley analyses interpreted with caution
are B15 (approximately 20% modal zeolite, 5% H2O); B21 (1.8% CO2), B22
(alteration of olivine phenocrysts and along fractures, carbonate veining, 2.8% CO2,
Fe2O3/FeO=0.8); B24 (alteration of olivine phenocrysts, alteration along fractures,
Fe2O3/FeO=1.0); B27 (alteration of olivine phenocrysts, approximately 20% modal
zeolite); B28 (alteration of olivine phenocrysts, presence of zeolite-filled vesicules).

6.4.3.4 Nomenclature.
The basalts are named on the basis of their chemical composition and normative
components following Green & Ringwood (1967) and Coombs & Wilkinson (1969), see
Table 6-1. The criteria used by Johnson (1989, p.13) to separate sub-alkaline (tholeiitic)
from alkaline types requires that sub-alkaline rocks have normative quartz, or more than
10% normative hypersthene. However, this scheme classifies as sub-alkaline a number of
earlier Central Volcanic Province analyses (in which the percentage SiO2 was calculated
by difference) which are clearly alkaline based on their mineralogy (McKay, 1975;
McQueen, 1975). Analyses are presented on an alkalis (Na2O + K2O) versus silica plot
(Fig. 6-20) with the dividing lines of MacDonald & Katsura (1964) and Saggerson &
Williams (1964) provided as a reference. This diagram has limitations for those rocks
which plot close to the alkalic-tholeiitic dividing line, (Wilkinson 1974).
The alkaline rocks (Fig. 6-21) have been classified using the scheme proposed by
Coombs & Wilkinson (1969) based on their normative plagioclase versus differentiation
index (D.I. = normative quartz + orthoclase + albite + nepheline + leucite). Rocks with
D.I. under 75 and greater than 5% normative nepheline include basanite, or are prefixed
with "nepheline" as shown in Figure 6-22.
6.4.4 Chemical features of the Mount Buckley lavas

All the Mt Buckley Series rocks are basic, with SiO2 contents between 45 - 50%
(Carmicheal et al., 1974, p. 29). Plots of normative hypersthene or nepheline against
normative plagioclase composition (Fig. 6-23) shows most are alkali olivine basalts and
basanites. The uppermost flows are predominantly basanite and nepheline hawaiites
whereas the lowermost flows are predominantly alkali olivine basalts and hawaiites.
A positive correlation can be seen on the normative plagioclase versus
differentiation index plot (Fig. 6-21), but there is no systematic variation between
stratigraphic occurrence and position on this plot.
In terms of the alkalis ratio, K2O : Na2O, the Mt Buckley flows (Fig. 6-24) are
neither distinctly sodic nor potassic but tend to cluster about a line, K2O : Na2O = 1 : 2,
between the sodic Hawaiian alkalic suite (McDonald & Katsura, 1964) and the potassic
Tristan de Cunha alkalic series (Baker et al, 1964).

6.4.4.1 Chemistry versus stratigraphic position at Mount Buckley

Normative hypersthene or nepheline (Fig. 6-25), major oxides (Fig. 6-26), and
trace elements (Fig. 6-27) were plotted versus flow number to discern any trends.
A trend towards increasing undersaturation with time is revealed in Figure 6-25.
The uppermost flows, B1 to B10, are all moderately silica undersaturated (>5% normative
nepheline) and are all basanites or nepheline hawaiites, with the exception of B9 which is
an alkali olivine basalt. B3 is chemically quite distinct having the highest Zn (119 ppm),
Zr (377 ppm), K2O (2.71%), Na2O (4.63%), P2O5 (1.01%) and Ga (25 ppm), high Sr
(1237), low Ca (7.02%), Ni 92 (ppm), MgO (6.27%) and V (103 ppm) and the lowest Cr
(94 ppm) and Cu (27 ppm). B9 is also different from its neighbours being coarse grained
with distinct CaO, Zr, Nb and Cu values. Their field relations, however, are conformable.
The lowermost flows, B11 to B27, are mildly undersaturated alkali olivine basalts or
hawaiites with the exception of B23 which is a basanite. The additional four lower flows,
B28 to B31, are also alkali olivine basalts. The only flows plotting within or close to the
tholeiitic compositional field on Figure 6-25 are B15 and B27 respectively; however, their
modal mineralogy is distinctly alkaline.
Figures 6-26 and 6-27 show systematic correlations between various elements,
such as MgO, Ni, Cr, Al2O3, Na2O, P2O5, CaO and V.
The followng adjacent flows within the Mt Buckley sequence appear to behave as
groups, with members of each group having similar chemical compositions and apparently
common genesis. These include; flows B16, B17, B18 and B19 (all coarse grained alkali
olivine basalts), flows B10, B11, B12 and B13 (alkali olivine basalts except B10 a fine
grained basanite), and flows B5, B6, B7 and B8, (basanites except B8 a nepheline
hawaiite). Some elements in these groups of flows change in a regular manner from flow
to flow suggesting a probable fractionation pattern.
 For the overall Mt Buckley sequence (Figs. 6-26 & 6-27), when chemically distinct
flows B3 and B15 are ignored, the SiO2 content decreases from about 48% in B19 to 45%
in B1. V decreases from flows B18, B19 to a minumum in flow B7 then increases again
from flow B6 to B1. CaO follows the same trend with the minimum occurring at B8. MgO
also shows a minimum in B7.

6.4.4.2 Xenolith-bearing and primary lavas at Mount Buckley.

Ultramafic xenoliths occur in flows B1, B2, B4, B6, B7, B10, B12, B24 and B25.
The location of the pyroxenite and lherzolite bearing flows, described by Wilkinson
(1973), is in a small tributary of Youngs' Creek (GR:520075 Inverell 1:100,000, Fig. 6-19,
Len Stewart, pers. comm.). The tributary incises the flows comprising Vol Aston Hill,
numbered B24 and B25. Wilkinson (1973) determined the pyroxenites and peridotites to
be samples of essentially unmodified, layered upper mantle. These xenoliths are
characteristic of more alkaline and undersaturated flows which must have moved directly
and rapidly from upper mantle levels to the surface. Such flows are therefore either
primary melts or produced by crystal fractionation processes at high pressures (Irving and
Green, 1976; Wass, 1980). Criteria for recognition of primary alkaline magmas include;
a) An Mg-value (100Mg/[Mg+Fe2+]) of 66 to 75 which represents liquids that could be in
equilibrium with upper mantle residual olivine compositions (Fo89 - Fo99)
b) high Ni values (>300 ppm)
Flow B25 satifies these criteria with a Mg value of 69, Ni = 429 ppm and has the
highest observed MgO, Cr and Ni in the entire sequence. The rest of the xenolith-bearing
flows have Mg values of less than 65 and Ni less than 300 ppm, suggesting fractionation in
the mantle through high pressure crystallization (Green et al, 1974) or direct derivation
from more iron rich peridotite (Wilkinson and Binns, 1977). Such conditions of rapid
transport to the surface are also consistent with the preservation of sapphires in their
carrier magmas (Coenraads et al, 1990)
Xenoliths were notably absent in the coarse grained flows (B9, B11, B14, B15,
B16, B17, B18, and B19) as was also observed by Wass (1980) for flows in the Southern
Highlands. This may indicate a longer transit time to the surface, and/or residence in
reservoirs, for such flows. If this is the case, they are also less likely to be sapphire
carriers.

6.5 Comparison between volcanics of the sapphire-bearing East and sapphire-barren

West Central Province

Chemical analyses for the Central Volcanic Province have been separated into
those from the sapphire-bearing East and sapphire-barren West Central Province. Analyses
for samples close to the boundary between the two (i.e. the MacIntyre River region) were
included with the eastern group as these constitute the majority and will include both
sapphire-bearing and sapphire-barren flows. The purpose of this comparison was to test
whether the sapphire-barren flows form a separate subset or a distinct, smaller
compositional field, as was observed by Vichit et al. (1978) for the southeast Asian
volcanic provinces. However, apart from more tholeiitic compositions, the rest of the
sapphire-barren West Central Province analyses range over the entire compositional field,
(see samples flagged with "w" in Fig. 6-20). The tholeiites found by Duggan (1972) are
distinctly different from any basalts in the East Central Province. The identification of
such tholeiites may be a discriminant against finding sapphire as these lavas show
evidence of some low pressure fractional crystallisation (Wilkinson and Duggan, 1973),
which would involve residence at higher levels.

6.6 Comparison between the Central Province and other volcanic provinces in
northeastern Australia and southeast Asia

Stephenson et al (1980) compare the compositional fields for the volcanic

provinces of north Queensland on a number of variation diagrams (i.e. alkalis versus silica,
K2O versus Na2O, AFM, and normative plagioclase versus differentiation index plots).
The compositional fields largely overlap one another, and those known to have associated
sapphire (Atherton, Chudleigh and McBride) are not significantly different from the rest of
the north Queensland provinces. Not surprisingly, the Central Volcanic Province
compositional field also overlaps with those for Queensland (Fig. 6-28). The implications
are that, either all the north Queensland provinces and the West Central Province are
potentially sapphire-bearing, or more likely, that this type of data and its scatter are
insufficient to predict likely sapphire-bearing and sapphire-barren provinces.

6.7 Conclusions

1. The basalt flows examined at Mt Buckley show a trend of increasing undersaturation

with time, and range from predominantly alkali olivine basalts and hawaiites to
predominantly basanites and nepheline hawaiites. They probably represent intercalated
flows from several different mantle and crustal sources, particularly as many of the
major and trace element values for flows, such as B3, B9 and B15, vary markedly from
their neighbouring flows. The group of hawaiites B21, B22, B24 and B25 show widely
varying values but consistently differ from their neighbours by having higher K2O and
Na2O and lower CaO and V. These flows include a possible primary representative
(B25). A large number of the flows carry high pressure xenoliths, implying their rapid
movement from upper mantle depths.

2. There appears to be little distinction between the compositions from the sapphire-
bearing East Central Province and the sapphire-barren West Central province on the
variation diagrams. Tholeiites occurring to the west of Inverell appear to be an
exception.
3. The compositional fields for the northeastern Australian volcanic provinces and the
Central Volcanic Province overlap extensively and there is no apparent compositional
distinction between fields known to contain corundum and those which do not.

4. The compositional field for the Central Volcanic Province overlaps both the corundum-
bearing and corundum-barren fields for southeast Asian volcanic provinces.

5. Common compositional variations in basaltic fields do not appear to be useful indicators

for predicting sapphire prospectivity in volcanic provinces.
 Chapter 7

SUMMARY AND CONCLUSIONS

7.1 Origin of corundum associated with volcanic provinces

7.1.1 Aims and achievment of aims

A principal aim of this study was to shed light on the origin of corundum
associated with volcanic provinces, using the Central Volcanic Province in northeastern
New South Wales as a case study. The surface morphology of the corundum crystals and
their inclusions were studied with a view to determine likely source rocks, the age
relationship between the sapphires and the basalts, and the effects of transport in magma
and fluviatile environments on the corundum.
A second principal aim of this study was to investigate hydrologic and geomorphic
factors which contributed to sapphire concentration to form known economic deposits and
to assess prospectivity for sapphire and diamond exploration within the New England Gem
Fields.
The aims of this study have been achieved. A summary of results and conclusions
are presented in the sections following. Publications generated during this study are listed
in Appendix 6.

7.2.1 Inclusion studies

Uranium-lead isotope dating of two zircon inclusions in sapphires from the Central
Province, New South Wales gave ages of 35.9 ± 1.9 and 33.7 ± 2.1 million years (Ma).
These ages fall within the range of basalt potassium-argon ages (19 to 38 Ma) and zircon
fission track ages (2 to 49 Ma) for the timing of volcanism in the Central Volcanic
Province (Coenraads et al, 1990). These data, combined with the observation that
corundum is found associated with many alkali basaltic provinces, indicate a genetic
link between the growth of large corundum crystals and contemporaneous processes
involved in alkali basaltic magma generation.
The reported failure of experimental attempts to grow corundum from a corundum-
bearing basaltic composition, (Green et al. 1978), and more significantly, the abundance of
incompatible elements such as U, Th, Zr, Nb and Ta in inclusion minerals, indicate that the
melting/crystallization process is not simple. Corundum, and other minerals (such as
zircon, columbite, thorite, uranium pyrochlore, alkali feldspar etc.) found as inclusions in
corundum, could not have crystallized from most basaltic compositions. A more complex
process must occur in which crystallization of coarse aggregates takes place when high
proportions of incompatible elements and volatiles are present in early melt fractions.
These unusual crystallization products are subsequently transported to the surface by
voluminous basaltic magmas. The extent to which this process occurs, together with
the rate of transport to the surface, presumably determine whether a particular
basaltic province carries sufficient corundum to be worked into economic
concentrations of sapphire.

7.1.2 Surface features

The majority of surface features observed on natural rubies and sapphires of

volcanic origin reflect their trigonal crystal structure and are the result of layer dissolution
or etching that occurred while the crystals were exposed to the hostile melt environment
whilst en route to the surface. Such features generated by magmatic resorption include
trigonal prismatic hillocks, trigonal pyramids, inverted trigonal pyramids, flat floored
trigonal and hexagonal depressions (often with a central rhombohedral or pyramidal
termination), and "brick-like" stacks of trigonal prisms. The style of features observed on a
given surface is dependent both on the angle of the surface to the c-axis of the corundum
crystal and the degree of subsequent etching. More irregular sculpture-like features appear
to arise from continued dissolution, and the growth and merging of the above-mentioned
features. Many of the crystal surfaces also show a superimposed surface texture consisting
of fine corrosion-pits.
Fluvial transport and residence in the the alluvial environment are responsible for
additional surface features. These include conchoidal fracture surfaces radiating from
points of impact (particularly along exposed edges on some stones), clean fracture surfaces
where pyramidal and prismatic protrusions have been broken, and exposed healed-fracture
surfaces. There is often little evidence of abrasion due to fluvial transport.
It has been observed that pitting and etching on associated heavy minerals is
commonly the result of surface biodegradation and this is particularly pronounced for the
spinels.
The grain surface features, coupled with the spatial variability in the
character the corundum, and the differences in proportions of heavy minerals in
placer deposits collectively indicate minimal amount of fluvial transport, and
minimal degree of downstream reworking and mixing. Such observations indicate
that the corundum is derived from local multiple sources.
7.2 Division of the New England gem fields into three exploration regions

Key areas for diamond and sapphire exploration may be defined within the New
England gem fields of eastern Australia through an understanding of the geologic and
geomorphic processes that operated within the Tertiary Central Volcanic Province.
At least two distinct periods of volcanic activity - one at 32-38 Ma and the other at
19-23 Ma - were controlled by SSE/NNW trending fracture sets revealed by landsat data.
The locus of volcanism stepped westwards with time from the Glen Innes-Ben Lomond-
Guyra area (the East Central Province) to the area west of Inverell and Armidale (the West
Central Province). Drainage and pre-volcanic topography indicate a radial pattern of
drainage associated with East Central Province uplift and volcanism (32 - 38 Ma). This
was modified by the later West Central Province volcanism (19 - 23 Ma) which
established its own radial pattern. This division of the Central Volcanic Province defines
broad regions suitable for sapphire and diamond exploration.
Highly prospective diamond-bearing "deep leads" associated with the palaeo-
Gwydir River can be traced under the younger basalts of the West Central Province via
reconstruction of the pre-volcanic topography.
In the East Central Province the post-eruptive fluvial history was vital in
controlling the concentration of economic sapphire deposits from lower grade source
rocks. The sapphire-bearing placer deposits occur as "shoestring" type accumulations
occupying channels within broad, flat, basalt-filled valleys. The highest grades generally
correspond to areas where channels are deepest.
The Central Province has been divided into three prospective target regions. These
regions comprise a western region with a potential for "palaeo-Gwydir type" alluvial
diamond deposits, a central belt with "Braemar type" sapphire deposits, and an eastern
division with "Kings Plains- Reddestone type" alluvial sapphire deposits.
A detailed palaeotopographic reconstruction technique had not been applied
previously to the search for economic deposits of diamond or sapphire in the Central
Volcanic Province. This method has proven to be a valuable exploration tool for
delineating palaeochannels containing the above-mentioned deposits.

7.2.1 Diamond-bearing deep leads of the West Central Volcanic Province

The Palaeo-Gwydir River system was buried by 19-23 Ma volcanism of the West
Central Province. Prospectivity for diamond-bearing deep lead alluvium is considered to
be favorable in two major palaeodrainage systems that extend northward from the present
Gwydir River in the vicinity of Copeton Dam. These are the palaeo-Hobbs and palaeo-
Gwydir systems. The palaeotopographic reconstruction indicates at least 3 separate
systems of diamond bearing alluvials and therefore, local, multiple sources for the
diamonds.
7.2.2 Holocene alluvial sapphire deposits of the East Central Volcanic Province

Two major sapphire deposits are situated in the Kings Plains and Reddestone
Creeks of the East Central Province (Figs 4-3 and 4-4). These Holocene alluvial deposits
are currently being worked by T.J. & P.V. Nunan Pty Ltd using mechanized mining
methods. Each mine processes approximately 1000 tonnes of alluvium per day, yielding 1
tonne of heavy mineral concentrate containing about 10 kilograms of corundum (Nunan,
1989). There is a clear association between these deposits which have developed since the
filling of the palaeovalleys with basalt, and the present drainage. Higher grades generally
correspond to areas where channels are deepest, and sapphire grades in excess of 500
grams per cubic metre have been recorded. In this ideally balanced system the heavy
minerals have moved vertically downwards, rather than downstream, and become
concentrated with time.
Palaeotopographic reconstruction highlights potential abandonned channel
deposits which may no longer show any obvious association with the present day
drainage. Such sapphire deposits may occur along abandoned palaeovalleys which
continue below the points of capture by the present drainage; they may be economic
if the original drainage existed for sufficient time to allow concentration to take
place.

7.2.3 Braemar-type deep lead deposits in the zone-of-overlap:

Exploration areas for deep lead sapphire deposits, such as those found at Braemar
were defined via mapping of palaeochannels in which the deposits are situated.
Palaeotopographic reconstruction, based on the elevation of the contact between the
Tertiary volcanic rocks and the Palaeozoic basement, together with available drill hole
data, were used. The potential sapphire-bearing palaeochannels have been delineated
within the zone in which the 19-23 million year old volcanics forming the western portion
of the Central Volcanic Province (the West Central Province) overlap onto 32-38 million
year old sapphire bearing volcanics forming the eastern portion of the province (the East
Central Province). In this zone, the 19-23 Ma basaltic lavas flooded a number of major
palaeodrainage systems already containing 32-38 Ma basalt flows and alluvial deposits
reworked from them.
Potential sapphire deposits have been delineated in the Braemar
palaeochannel and its tributaries up-palaeostream from the Braemar sapphire mine
where sapphires are recovered along with other heavy minerals. The highest concentration
occurring at the base of the channel. "Braemar-type" deep lead deposits considered in this
thesis had approximately 10 million years in which to form, and required the presence of a
younger basalt capping (19-23 Ma) for their preservation.
Potential sapphire-bearing alluvials are also delineated along the
topographically inverted palaeo-Swan Brook. The extent of these "deep lead"
deposits to the east of Braemar, in palaeochannels that drained the sapphire-bearing
East Central Province, is related to the easternmost incursions of the lavas of the
West Central Province.

7.3 Heavy mineral suites in the New England gem fields

7.3.1 Sapphire associates of the East Central Volcanic Province

Heavy minerals associated with sapphire in the alluvial gravels are pleonaste,
ilmenite, chromium-spinel, titanium-magnetite, magnetite, corundum, zircon and minor
chrysoberyl. Additional minerals found in nearby in situ basaltic soils comprise olivine,
clinopyroxene, enstatite and amphibole, with zircon and sapphire being extremely rare.
Investigation of heavy mineral suites from 12 collection sites indicated significant
variability in mineral species within and between particular drainage catchments. Visual
characteristics of the sapphires also vary within and between catchments. This implies that
sources are 'local' to the placer deposits. The similarity in composition of minerals from
soil and placer sites indicates that the sapphire-bearing placers have formed through
extensive reworking of the immediately surrounding alkali basaltic and volcaniclastic
rocks. Concentration was largely through vertical movement, with fine or light material
being winnowed downstream leaving behind a heavy gem-bearing residue, within low
gradient, low energy drainage systems.

7.3.2 Ilmenite-mantled rutile crystals of the Uralla area

Ilmenite-mantled rutile crystals of enigmatic origin are found in Late Tertiary

conglomeratic arkoses and Quaternary alluvium in the vicinity of Uralla, New South
Wales. They comprise homogeneous, single, rutile crystal cores with ilmenite replacing
the crystals and penetrating along (100) planes. The ilmenite ranges from higher TiO2
adjacent to the rutile core to higher FeO at the outer edge. MgO decreases away from the
core. This is consistent with reaction in a magma containing iron but poor in magnesium.
The crystals are not associated with any obvious parent rock type. Comparisons of the
chemistry of the ilmenites with a wide range of known ilmenite associations indicate that
the source rock is unlikely to be mafic or ultramafic and that they are an unlikely to be an
indicator for diamond source rocks. The minerals may have formed as late-stage or cavity
crystallizations from fractionated felsic magmas before eruption.

7.3.3 Pyrope-almandine garnets from Horse Gully

Deep red to purplish, gem quality garnets have been recovered from sapphire-
bearing river gravels at Horse Gully in the New England gem fields. Seven garnets were
faceted, gemmologically tested and analyzed with an electron microprobe. They are
members of the pyrope-almandine intermediate series ranging in composition from pyrope
55%-almandine 34% to pyrope 39%-almandine 49%. The stones with the highest end-
member pyrope content (highest magnesium and lowest iron) display a most attractive
mauve-purple to pinkish hue. They also have the lowest refractive index and specific
gravity. The garnets are derived from a local, probably mafic, source situated within the
catchment area of Horse Gully.

7.4 Investigation of sapphire source rocks and potential source structures in the
Central Volcanic Province

Basaltic and volcaniclastic rocks in the East Central Volcanic Province are
potential sources of the alluvial sapphires being mined in the New England gem fields.
The associated drainage catchment areas generally contain sapphire, but only four major
catchments contain rich deposits. These are the Frazers, Kings Plains, Reddestone and
Marowan catchments and are the most likely targets for sapphire source rocks.
The basalt flows and intrusives comprising Mount Buckley divide Kings Plains
Creek and Swan Brook, two important catchment areas for alluvial sapphire. Thirty one
whole-rock analyses show mostly alkaline to strongly alkaline types and a general trend of
increasing undersaturation with time. The top flows are predominantly basanite and
nepheline hawaiites whilst the lowest are predominantly alkali olivine basalts and
hawaiites. Ultramafic xenoliths in several flows suggest rapid movement from upper
mantle levels. Marked variation in major and trace elements between groups of flows
suggests that the sequence contains intercalated flows from different levels. No
volcaniclastic rocks were encountered in the Mt Buckley area.
Within the Central Province, variation diagrams show no significant difference
between the 32-38 Ma sapphire-associated eastern basalts and the 19-23 Ma sapphire-
barren western alkali-basalts. The compositional fields also overlap the Atherton, McBride
and Chudleigh provinces in north Queensland, known to contain sapphire, as well as those
of the sapphire-barren northeastern Australian volcanic provinces. These Australian
compositional fields also overlap the corundum-bearing and corundum-barren fields
defined by Vichit et al (1978) for the southeast Asian volcanic provinces.
Major and minor element chemistry of basaltic rocks is not considered to be a
useful exploration tool for discriminating sapphire-bearing from sapphire-barren
volcanic provinces, nor for recognizing potentially high-grade areas within a
sapphire-bearing province. This suggests that basaltic rocks are not the initial
sapphire source rocks, but only one of the potential carriers.

Lagoons located in the basalts of the Central Province were investigated to

determine whether they could be maars and therefore potential sources of sapphire-bearing
volcaniclastic rocks. This was found not to be the case and the lagoons are interpreted to
have formed within wind blown lunettes occupying the wide, flat floors of basalt filled
valleys. This interpretation was based on drilling, geophysical modelling and geological
mapping which revealed the lagoons to be axial to pre-basaltic drainage channels
overlying at least 10 metres of valley fill flow basalts. Drilling at Kings Plains Lagoon
intersected basalt flows overlying silicic volcanic basement and failed to encounter any
breccia within, or feeder vent below, the structure. Bulk testing at Kings Plains and
Dunvegan lagoons indicates that these features do not constitute economic targets.
REFERENCES

Atkinson, W.J., Hughes, F.E. and Smith, C.B. (1984). A review of the Kimberlitic rocks of
Western Australia, in J.Kornprobst (Ed.). Kimberlites l. Kimberlites and Related Rocks,
195-224, Developments in Petrology, 11A, Elsevier, Amsterdam.

Anderson, W. (1888). Appendix No. 8 in Report in progress for 1887 by the Geological
Surveyor in Charge. Report, Department of Mines N.S.W. for 1887 155-156.

Atkinson, D. and Kothavala, R.Z. (1983). Kashmir sapphire. Gems and Gemmology, 19,
64-76.

Australian Centre for Remote Sensing, Landsat images, 090-080F, 096-081, and 095-081:
New South Wales Government Department of Lands, Remote Sensing Unit, 23-33
Bridge Street, Sydney, Australia.

Baker, P.E., Gass, I.G., Harris, P.G., and Le Maitre, R.W. (1964). The volcanological
report of the Royal Society expedition to Tristan de Cunha, 1962. Philosophical
Transactions of the Royal Society of London, 256, 439-578.

Barr, S.M. and MacDonald, A.S. (1981). Geochemistry and geochronology of late
Cenozoic basalts of Southeast Asia: Summary. Geological Society of America Bulletin,
92 (1), 508-512.

Barron, L.M. (1987). Summary of petrology and chemistry of rocks from the sapphire
project, New South Wales Geological Survey Report GS1987/050, 1-106. (Unpubl.).

Binns, R.A. (1969). High-pressure megacrysts in basanitic lavas near Armidale, N.S.W.
American Journal of Science 267-A, 33-49,

Binns, R.A., Duggan, M.B. and Wilkinson, J.F.G. (1970). High pressure megacrysts in
alkaline lavas from northeastern New South Wales, American Journal of Science 269,
132-168.

Birch, W.D. (1979). Mineralogy and Geochemistry of the leucitite at Cosgrove, Victoria.
Journal of the Geological Society Of Australia, 25, 369-385.

Birch, W.D., Smith, N. and Hately, J. (1982). Andesine, augite and ilmenite in vesicles in
an alkali olivine basalt from Portland, Victoria. Australian Mineralogist, 37, 195-198.
Bischoff, G.C.O. and Coenraads, R.R. (in prep). Fossil and Recent Traces of
Biodegradation on Heavy Minerals.

Broughton, P.L. (1979). Economic geology of the Anakie sapphire mining district
Queensland. Journal of Gemmology., 16, (5), 318-337.

Brown, R.E. (1987). Detailed geological mapping in the Elsmore and Kings Plains areas,
New South Wales Geological Survey - Report GS1987/058, 22-24. (Unpubl.).

Brown, R.E. and Pecover, S.R. (1986a). The geology of the Braemar sapphire deposit.
New South Wales Geological Survey - Report GS1986/270. (Unpubl.).

Brown, R.E. and Pecover, S.R. (1986b). The geology of the Kings Plains sapphire field.
New South Wales Geological Survey - Report GS1986/271. (Unpubl.).

Brownlow, A.H. and Komorowski, J.C. (1988). Geology and origin of the Yogo sapphire
deposit, Montana. Economic Geology 83, 875-880.

Brownlow, J.W. (1989). Guyra 1:100,000 Geologic Sheet, Department of Minerals and
Energy, Armidale Branch, State Government Office Building, cnr. Faulkner and
Dumeresq streets, Armidale, N.S.W., 2350.

Carmichael, I.S.E. (1967). The iron titanium oxides of salic volcanic rocks and their
associated ferromagnesian silicates. Contributions to Mineralogy and Petrology, 14,
36-64.

Carmichael, I.S.E., Turner, F.J. and Verhoogen, J. (1974). Igneous Petrology, McGraw
Hill Inc. New York.

Cassidy, K.F., Groves, D.I. and Binns, R.A. (1988). Manganoan ilmenite formed during
regional metamorphism of Archaean mafic and ultramafic rocks. Canadian
Mineralogist, 26, 999-1012.

Cawthorne, R.G., Maske, S., de Wet, M., Groves, D.I. and Cassidy, K.F. (1988).
Contrasting magma types in the Mount Ayliff (Insizwa complex), Transkei: evidence
from ilmenite compositions. Canadian Mineralogist, 26, 145-160.

Clarke, D.B. and MacKay, R.M. (1990). An ilmenite-garnet-clinopyroxene nodule from

Matsoku: Evidence for oxide-rich liquid immiscibility in kimberlites? Canadian
Mineralogist, 28, 229-239.
Coenraads, R.R. (1988a). Kings Plains drilling project. Report for Jingellic Minerals N.L.,
11th March. (Unpubl.).

Coenraads, R.R. (1988b). Six monthly report to the Department of Mines, E.L.'s 2988 and
2989 for the period to 12 July 1988. New South Wales Geological Survey - Reports
GS1988/250, GS1990/003 and GS1990/314. (Unpubl.).

Coenraads, R.R. (1988c). Structural control and timing of volcanism in the Central
Province. Implications for regional targeting of prospective areas for sapphire and
diamond exploration, in New England Orogen Tectonics and Metallogenesis.
J.D.Kleeman (Ed.). Department of Geology and Geophysics, University of New
England, Armidale, Australia, pp.302-307.

Coenraads, R.R. (1989). Evaluation of the natural lagoons of the Central Province, N.S.W.
- are they sapphire producing maars? Bulletin of the Australian Society of Exploration
Geophysicists, 20, 347-363.

Coenraads, R.R. (1990). Key areas for alluvial diamond and sapphire exploration in the
New England gem fields, New South Wales, Australia. Economic Geology, 85, 1186-
1207.

Coenraads, R.R. (in press). Palaeogeography of the Braemar sapphire deposit:

Implications for deep-lead sapphire exploration in the Central Volcanic Province, New
South Wales. Journal and Proceedings Royal Society of New South Wales,

Coenraads, R.R. (in prep). Evaluation of potential sapphire source rocks within the
catchments of Kings Plains Creek and Swan Brook, near Inverell, New South Wales.
Records of the Australian Museum

Coenraads, R.R. and Lawrence, D. (1989). Six monthly report to the Department of
Mines, E.L.'s 2988 and 2989 for the period to 12 January 1989. New South Wales
Geological Survey - Reports GS1988/250, GS1990/003 and GS1990/314. (Unpubl.).

Coenraads, R.R. and van der Graaf, R. (1991). An occurence of gem garnets from Horse
Gully in the New England gem fields, New South Wales. The Australian Gemmologist ,
17(10)
Coenraads, R.R., Paige, S.C.B. and Sutherland, F.L. (in press). Ilmenite mantled Rutile
Crystals from the Uralla District, New South Wales. Journal and Proceedings Royal
Society of New South Wales

Coenraads, R.R., Pope, G.J. and Whittle, M.A. (1989). Six monthly report to the
Department of Mines, E.L.'s 2988 and 2989 for the period 15 January, 1989 to 14 July,
1989. New South Wales Geological Survey - Reports GS1988/250, GS1990/003 and
GS1990/314. (Unpubl.).

Coenraads, R.R., Sutherland F.L. and Kinny P.D. (1990). The origin of sapphires: U-Pb
dating of zircon inclusions sheds new light. Mineralogical Magazine, 54, 113-122.

Coldham, T. (1985). Sapphires from Australia, Gems and Gemology, 21, 130-146.

Compston, W., Williams, I.S. and Meyer, C. (1984). U-Pb geochronology of zircons from
Lunnar breccia 73217 using a sensitve high mass-resolution ion microprobe. Journal of
Geophysical Research, 89, 525-534.

Coombs, D.S. and Wilkinson, J.F.G. (1969). Lineages and fractionation trends in
undersaturated volcanic rocks from the East Otago volcanic province (New Zealand)
and related rocks. Journal of Petrology, 10(3), 440-501.

Cooper, J.A., Richards, J.R. and Webb, A.W. (1963). Some potassium-argon ages in New
England, New South Wales, Journal of the Geological Society of Australia, 10, 313-
316.

Cotton, L.A. (1914). The diamond deposits of Copeton, New South Wales, Proceedings of
the Linnean Society of New South Wales, 39, 803-838.

Cundari, A. (1973). Petrology of the leucite bearing lavas in New South Wales. Journal of
the Geological Society of Australia, 20, 465-492.

Curran, J.M. (1897). On the occurrence of precious stones in New South Wales and the
deposits in which they are found. Journal of the Proceedings of the Royal Society of
N.S.W., 30, 228-237.

Dalrymple, G.B. (1979). Critical tables for conversion of K-Ar ages from old to new
constants. Geology, 7, 558-560.
David, T.W.E. (1886). Report on the geology of the Rocky River goldfields. New South
Wales Department of Mines Annual Report for 1885, pp. 153-158.

Dawson, J.B. (1980). Kimberlites and their Xenoliths (Minerals and rocks; 15). Springer-
Verlag, Berlin. J.B. Dawson (Ed.).

Deer, W.A., Howie, R.A. and Zussman, J. (1975). Rock-Forming Minerals 5. Non-
Silicates, Longmans Green and Co., London. W.A. Deer, R.A. Howie and J. Zussman
(Eds.).

Duggan, N.T. (1972). Tertiary volcanics of the Inverell area, Honours Thesis (unpubl.),
University of New England, Armidale, Australia.

Dunstan, B. (1902). The sapphire fields of Anakie. Queensland Government Mining

Journal 3, 239-246 and 293-298.

Gibbons, G.S. and Pogson, D.J. (1963). Diamond Deposits at Mount Ross. Rep. Geol.
Surv. N.S.W., GS1963/002 (unpubl.).

Green, D.H., Edgar, A.D., Beasley, P., Kiss, E., and Ware, N.G. (1974). Upper mantle
source fof some hawaiites, mugearites and benmoreites. Contributions to Mineralogy
and Petrology, 48, 33-43.

Green, D.H. and Ringwood, A.E. (1967). The genesis of basaltic magmas. Contributions
to Mineralogy and Petrology, 15, 103-190.

Green, T.H., Wass, S.Y. and Fergusson, J. (1978). Experimental study of corundum
stability in basalts. Abstracts of the Third Australian Geological Convention,
Townsville, August 28-31, Geological Society of Australia, p.34.

Gubelin, E.J. and Koivula, J.I. (1986). Photoatlas of Inclusions in Gemstones. ABC
Edition, Zurich.

Gunawardene, M. and Chawla, S.S. (1984). Sapphires from Kanchanaburi Province,

Thailand. Journal of Gemmology, 19 (3), 228-239.

Gunawardene, M. and Rupasinghe, M.S. (1986). The Elahera gem field in central Sri
Lanka. Gems and Gemology, 22(2). 80-95
Haggerty, S.E. (1976). Opaque mineral oxides in terrestrial igneous rocks. in Oxide
Minerals. Reviews in Mineralogy 3, Hg-101 to Hg-300, Mineralogical Society of
America, Washington. D.C. P.H. Ribbe and D. Rumble III (Eds.).

Hänni, H.A. (1987). On corundums from Umba valley, Tanzania. Journal of Gemmology,
20 (5), 278-284.

Harte, B., Winterburn, P.A. and Gurney, J.J. (1987). Metasomatic and enrichment
phenomena in garnet peridotite facies mantle xenoliths from the Matsoku kimberlite
pipe, Lesotho. in Mantle Metasomatism, pp. 145-220. Academic Press Inc., London.
M.A. Menzies and C.J. Hawkesworth (Eds.).

Hollis, J.D. and Sutherland, F.L. (1985). Occurences and origins of gem zicons in Eastern
Australia, Records of the Australian Museum, 36, 299-311.

Hollis, J.D.., Sutherland, F.L. and Pogson, R.E. (1983). High pressure minerals and the
origin of the Tertiary breccia pipe, Ballogie Gem Mine, near Proston, Queensland.
Records of the Australian Museum, 35, 181-194.

Horton, D.R., and Connah, G.E. (1981). Man and megafauna at Reddestone Creek, near
Glen Innes, northern New South Wales. Australian Archaeology 13, 35-52.

Irvine, T.N. and Barager, W.R.A. (1971). A guide to chemical classification of the
common volcanic rocks. Canadian Journal of Earth Science, 8, 523-548.

Irving, A.J. (1980). Petrology and geochemistry of composite ultramafic xenoliths in

alkalic basalts and implications for magmatic processes within the mantle. American
Journal of Science, 280A, 389-426.

Irving, A.J. (1986). Polybaric magma mixing in alkalic basalt and kimberlites: evidence
from corundum, zircon and ilmenite megacrysts. Fourth International Kimberlite
Conference. Extended Abstracts. Perth. Geological Society of Australia Abstracts
Series, 16, 262-263.

Irving, A.J. and Green, D.H. (1976). Geochemistry and petrogenesis of the Newer Basalts
of Victoria and South Australia. Journal of the Geological Society of Australia, 23 45-
66

Irving, A.J. and Price, R.C. (1981). Geochemistry and evolution of Iherzolite-bearing
phonolitic lavas from Nigeria, Australia, East Germany and New Zealand. Geochimica
et Cosmochimica Acta, 45, 1309-1320.
Irving, A.J. and Price, R.C. (1981). Geochemistry and evolution of Iherzolite-bearing
phonolitic lavas from Nigeria, Australia, East Germany and New Zealand. Geochimica
et Cosmochimica Acta, 45, p. 1309-1320.

Iyer, L.A.N. (1953). The geology and gem-stones of the Mogok Stone Tract, Burma.
Memoirs of the Geological Society of India, 82, 1-100

Jackson, B. (1984). Sapphire from Loch Roag, Isle of Lewis, Scotland. Journal of.
Gemology, 19 (4), 336-342.

Jaques, A.L., Kerr, I.D., Lucas, H., Sun, S.S. and Chappell, B.W. (1989). Mineralogy and
petrology of picritic monchiquites from Wandagee, Carnarvon Basin, western
Australia, in Kimberlites and Related Rocks. I. J. Ross (Ed.). Geological Society of
Australia, Special Publication, 14, 120-139.

Jaques, A.L. and Perkin, D.J. (1984). A mica, pyroxene, ilmenite megacryst-bearing
lamprophyre from Mt Woolooma, north eastern New South Wales. Bureau of Mineral
Resources Journal of Australian Geology and Geophysics, 9, 33-40.

Jobbins, E.A. and Berrange, J.P. (1981). The Pailin ruby and sapphire gemfield,
Cambodia. Journal of Gemmology, 17 (8), 555-567.

Johnson, R.W. (1989). Volcano distribution in and classification; distribution in eastern

Australia, in Johnson, R.W. and Taylor, S.R., eds, Intraplate Volcanism in Eastern
Australia and New Zealand: Cambridge University Press. p. 9-10.

Jones, D.R. (1983). Difficulties associated with using indicator minerals for diamond
exploration in north Queensland. M.Sc. Thesis, James Cook University of North
Queensland. (Unpubl.).

Keller, A.S. and Keller, P.C. (1986). The sapphires of Mingxi, Fujian Province, China.
Gems and Gemology, 22 (1), 41-45.

Keller, P.C. (1982). The Chanthaburi-Trat gem field, Thailand. Gems and Gemology, 18
(2), 186-196.

Keller, P.C. and Wang, F. (1986). A survey of gemstone resources of the People's
Republic of China. Gems and Gemology, 22 (1), 3-13.
Keller, P.C., Koivula, J.I. and Jara, G. (1985). Sapphire from the Mercaderes-Rio Mayo
area, Cauca, Colombia. Gems and Gemology, 21 (1), 20-25.

Kiefert, L. and Schmetzer, K. (1987). Blue and yellow sapphire from Kaduna Province,
Nigeria. Journal of Gemmology, 20 (7-8), 427-442.

Kinny, P.D., Williams, I.S., Froude, D.O., Ireland, T.R. and Compston, W. (1988).
Early Archaea zircon ages from orthogneisses and anorthosites at Mount Narryer,
Western Australia. Precambrian Research, 38, 325-341.

Kinny, P.D., Compston, W., Bristow, J.W. and Williams, I.S. (1989). Archaean mantle
xenocrysts in a Permian kimberlite: Two generations of kimberlitic zircon in
Jwaneng DK2, southern Botswana. in: Kimberlites and Related Rocks, Geological
Society of Australia Special Publication, 14.

Lacombe, P. (1969-1970) Le massif basaltique quaternarie a zircon-gemmes de

Ratanakiri (Cambodge-oriental). Bull. Bureau Recherches Geol. Min. (2nd series)
Sect. iv: No. 3-1969, p.31-91; No. 2-1970, p.29-57; No. 4-1970, p.33-79. Paris.

Leitch, E.C. (1974). The geological development of the southern part of the New
England Fold Belt. Journal of the Geological Society of Australia, 21 133-156.

Lishmund S.R. (1987), Regional distribution of sapphire, diamond and volcaniclastic

rocks, Department of Mineral Resources, extended abstracts from a seminar on Tertiary
volcanics and sapphires in the New England District, 1st May, 9-12- Report GS
1987/058.

Lishmund, S.R. and Oakes, G.M. (1983). Diamonds, sapphires and Cretaceous/Tertiary
diatremes in New South Wales. New South Wales Geological Survey - Quarterly Notes,
53, 23-27.

Lovering, J.F. (1964). The eclogite-bearing basic igneous pipe at Ruby Hill near Bingara,
New South Wales. Journal and Proceedings, Royal Society of New South Wales, 97,
73-79

Mabbutt, J.A. (1977), Desert Landforms. ANU Press, Canberra.

MacDonald, G.A. and Katsura, T. (1964). Chemical composition of Hawaiian lavas.

Journal of Petrology, 5, 82-133.
MacNevin, A.A. (1972). Sapphires in the New England District, New South Wales.
Records of the Geological Survey of New South Wales, 14(1), 19-35.

MacNevin, A.A. (1977). Diamonds in New South Wales. Department of Mines,

Geological Survey of New South Wales, Sydney. Mineral Resources No. 52. 1-125.

McMinn, A. (1987), Miocene palynology of DM Elsmore DDH2 and 3, near Inverell,

Geological survey of N.S.W., Palynological Report 1987/005. (Unpubl.)

McMinn, A. (1989). Tertiary palynology in the Inverell area. New South Wales Geological
Survey - Quarterly Notes, 76, 1-10.

McDougall, I. and Wilkinson, J.F.G. (1967). Potassium-argon dates on some Cainozoic

volcanic rocks from northeastern New South Wales, Journal of the Geological Society
of Australia, 14, (2), 225-233.

McKay, G.J. (1975). Volcanic and other petrology of the Wellingrove area, New England.
Honours Thesis, University of New England, Armidale, Australia. (Unpubl.).

McQueen, L.B. (1975). Permian acid volcanic rocks and Tertiary alkaline basaltic rocks of
an area northeast of Inverell, N.S.W, Honours Thesis, University of New England,
Armidale, Australia. (Unpubl.).

Mitchell, R.H. (1973). Magnesian ilmenite and its role in kimberlite petrogenesis. Journal
of Geology, 81, 301-311

Mitchell, R.H. (1979). Mineralogy of the Tunraq kimberlite, Somerset Island, N.W.T.
Canada. Proceedings of the Second International Kimberlite Conference. Volume 1:
The Mantle Sample: Inclusions in Kimberlites and other Volcanics. F.R. Boyd Jr and
H.O.A. Meyer (Eds.). American Geophysical Union, Washington D.C.

Moore, A.E. (1987). A model for the origin of ilmenite in kimberlite and diamond:
implications for the genesis of the discrete nodule (megacryst) suite. Contributions to
Mineralogy and Petrology, 95, 245-253.

Munasinghe, T. and Dissanayake, C.B. (1986). The origin of gemstones of Sri Lanka.
Economic Geology, 76, 1216-1225.
Neal, C.R. and Davidson, P. (1989). An unmetasomatized source for the Malaitan alnoite
(Solomon Islands): Petrogenesis involving zone refining, megacryst fractionation and
assimilation of oceanic lithosphere. Geochemica et Cosmochimica Acta, 53, 1975-1990.

New South Wales Government Department of Lands, 1:25,000 and 1;100,000 topographic
sheets, Aerial Photography and Map Sales, 23-33 Bridge Street, Sydney 2000,
Australia.

New South Wales Government Department of Minerals and Energy: Dorrigo-Coffs

Harbour, Grafton, Inverell and Manilla, 1:250,000 geological sheets, Publications
Section, 29-57 Christie Street, St Leonards, N.S.W. 2065.

New South Wales Government Department of Minerals and Energy. (1987). Tertiary
Volcanics and Sapphires in the New England District, seminar held in Sydney, 1st May
1987. 29-57 Christie St, St. Leonards, New South Wales, 2065.

New South Wales Government Department of Water Resources, 10 Valentine Ave,

Parramatta, New South Wales, 2150.

Nunan, T.J. (1989). The mining of sapphires. The Australian Gemmologist, 17(1) 7-12.

Ollier C.D. (1967). Maars: their characteristics, varieties and definition, Bulletin of
Volcanology, 31, 46-73.

Ollier C.D. (1979). Evolutionary geomorphology of Australia and Papua- New Guinea,
Transcripts of the Institute of British Geographers, 4, 516-539.

Ollier C.D. (1982). Geomorphology and tectonics of the Armidale region, Voisey
Symposium 1982, New England Geology. P.G. Flood and B. Runnegar (Eds.).

Ollier C.D. and Joyce E.B. (1976). Newer volcanic landforms, in Geology of Victoria. J.G.
Douglas and I.A. Ferguson (Eds.). Geological Society of Australia, Special Publication
5, 341-345.

Orlov, Y.L. (1977). The Mineralogy of the Diamond. John Wiley and sons, New York,
Brisbane.

Parasnis, D.S. (1975), Methods in Geochemistry and Geophysics 3, Mining Geophysics.

2nd edition Elsevier Scientific Publishing Company.
Patel, A.R. and Agarwal, M.K. (1966). Unusual microstructures on some South African
natural diamonds. Industrial Diamond Review, 26 (307), 240-244.

Pecover S.R. (1987). Tertiary maar volcanism and the origin of sapphires in northeastern
New South Wales, Department of Mineral Resources, extended abstracts from a
seminar on Tertiary volcanics and sapphires in the New England District, 1st May, 13-
21. New South Wales Geological Survey - Report GS 1987/058.

Pecover, S.R. and Brown, R.E. (1986) Tertiary volcanism in northeastern New South
Wales: Implications for sapphire and diamond exploration. New South Wales
Geological Survey - Report GS1986/272. (Unpubl.).

Pecover, S.R. and Coenraads, R.R. (1989). Tertiary volcanism, alluvial processes, and the
origin of sapphire deposits at "Braemar" near Elsmore, northeastern New South Wales.
New South Wales Geological Survey - Quarterly Notes, 77, 1-23.

Pithers, G. (1988). Final Report for E.L. 2987, Inverell-Tingha Area, Hooker Resources
Pty. Ltd. Report New South Wales Geological Survey - Report GS1988/248.
(Unpubl.).

Pittman, E.F. (1897). Report on the Boggy Camp diamond fields: A Report to the
Department of Mines for 1896, Geological Survey of New South Wales, Sydney, 98-99.

Porter, H.M. (1899). Diamonds at Inverell, N.S.W. Australian Min. Stand., 16, 169-170.

Rudnick, R.L. and Williams, I.S. (1987). Dating the lower crust by ion microprobe. Earth
and Planetary Science Letters, 85, 145-161.

Rumble, D. (1976). Oxide minerals in metamorphic rocks, in Oxide Minerals. Reviews in

Mineralogy 3, R1-R24, Mineralogical Society of America, Washington D.C. P.H.
Ribbe and D.Rumble (Eds.).

Saggerson, E.P. and Williams, L.A.S. (1964). Ngurumanite from southern Kenya and its
bearing on the origin of rocks in the northern Tanganyika alkaline district. Journal of
Petrology, 5, 40-81.

Scott Smith, B.H., Skinner, E.M.W. and Loney, P.E. (1989). The Kapamba lamproites of
the Luangwa Valley, Eastern Zambia, in Kimberlites and Related Rocks I. J. Ross
(Ed.). Geological Society of Australia Special Publication, 14, 189-205.
Simpson, E.S. (1951). Minerals of Western Australia 2. Government Printer, Perth,
Western Australia.

Slansky, E. (1987). X-ray examination of concentrates from sapphire-bearing rocks. New

South Wales Geological Survey - Report GS1987/12, pp. 1-8. (Unpubl.).

Smith, N.M. (1988). Reconstruction of the Tertiary drainage systems in the Inverell
Region, Honours Thesis, University of Sydney, Australia. (Unpubl.).

Spencer, L.K. (1983). Sapphires in Australia. AMAX Office, Inverell, N.S.W. 2360,
report, August 1983, 83 p. (Unpubl.).

Spencer, R. (1987). Ground geophysical surveys conducted for the sapphire project,
Department of Mineral Resources, extended abstracts from a seminar on Tertiary
volcanics and sapphires in the New England District, 1st May, 36-39.

Stephenson, P.J. (1976). Sapphire and zircon in some basaltic rocks from Queensland,
Australia. Abstracts, 25th International Geological Congress, Sydney, 2, 602-603.

Stephenson, P.J., Griffin, T.J. and Sutherland, F.L. (1980). Cainozoic volcanism in eastern
Australia, in Geology and geophysics of northeastern Australia. R.A. Henderson and
P.J. Stephenson (Eds.). Geological Society of Australia, Queensland Division,
Brisbane.

Stonier, G.A. (1895). Report on the Bingarra diamond fields. A Report to the Department
of Mines for 1894, Geological Survey of New South Wales, Sydney, 131-136.

Street, G.J. (1974). An examination of the Permian granitic plutons and Tertiary alkaline
basaltic rocks, east of Glencoe, northern N.S.W., Honours Thesis, University of New
England, Armidale, Australia. (Unpubl.).

Stroud, W.J. (1989). Bingara, Bukkulla, Cherry Tree Hill, Delungra, Dinoga, Elsmore,
Gum Flat, Hurricane Hill, Indianna, Inverell, Mt Rodd, Nullamanna, Sapphire,
Tingha, and Warialda 1:25,000 Geologic Sheets. Department of Minerals and
Energy, Armidale Branch, State Government Office Building, cnr Faulkner and
Dumeresq streets, Armidale, N.S.W., 2350.

Sutherland, F.L. (1980). The Geology and Petrology of some Tertiary Volcanic Rocks in
the Bowen-St.Lawerence Hinterland. PhD. Thesis, James Cook University of North
Queensland, Townsville, Australia. (Unpubl.).
Sutherland, F.L. (1985). Regional controls in eastern Australian volcanism, in Volcanism
in Eastern Australia with case histories from New South Wales. F.L. Sutherland, B.J.
Franklin and A.E. Waltho (Eds.). Geological Society of Australia, New South Wales
Division Publication 1, 13-32.

Sutherland, F.L., Kinny, P.D. and Hollis, J.D. (1991). Use of zircons to resolve origins of
sapphires in east Australian gem fields - the mysterious case of the shuffling sapphire
volcanoes, in "Exploration in a Changing Environment", The Australian Society of
Exploration Geophysicists 8th Conference and Exhibition and The Geological Society
of Australia Exploration Symposium, Sydney 1991, Abstracts No. 30. p. 227-228.

Sutherland, F.L., Robertson, A.D., Hendry, D.F. and Hollis, J.D. (1990). Xenolith suites
from southern Queensland: The lithosphere below volcanic regions, in The Eromanga-
Brisbane Geoscience Transect. D.M. Finlayson (Ed.). Department of Primary Industries
and Energy, Bureau of Mineral Resources, Geology and Geophysics.

Sweeney, J.W. (1971). Rhodesian sapphire deposits. Lapidary Journal, November, 1076-
1077.

Temby, P.A. (1986). Exploration Licence 2285, Elsmore, Final report covering the period
24/10/85 - 24/10/86. Blue Circle Southern Cement Limited. New South Wales
Geological Survey - Report GS1985/246. (Unpubl.).

Thompson, R.N., Morrison, M.A., Dicken A.P., Gibson, I.L. and Harmon, R.S. (1986).
Two contrasting styles of interaction between basic magmas and continental crust in
the British Tertiary volcanic province. Journal of Geophysical Research, 91 B6,
5985-5997.

Upton, B.J.G., Aspen, P. and Chapman, N.A. (1983). The upper mantle and deep crust
beneath the British Isles: evidence from inclusions in volcanic rocks. Journal of the
Geologicl Society of London, 140, 105-121.

Vichit, P. (1987). Gemstones in Thailand. Journal of the Geological Society of Thailand, 9

(1-2), 108-133.

Vichit, P., Vudhichativanich, S. and Hansawek, R. (1978). The distribution and some
characteristics of corundum-bearing basalts in Thailand. Journal of the Geological
Society of Thailand, 3 (1), M4-1 to M4-38.
Wass, S.Y. (1973). Oxides of low pressure origin from alkali basaltic rocks, Southern
Highlands, N.S.W., and their bearing on the petrogenesis of alkali basalt magmas.
Journal of the Geological Society of Australia., 20, 427-448.

Wass, S.Y. (1980). Geochemistry and origin of xenolith-bearing and related alkali basaltic
rocks from the Southern Highlands, New South Wales, Australia. American Journal of
Science, 280A, 639-666.

Wass, S.Y., Henderson, P. and Elliot, C.J. (1980). Chemical Heterogeneity and
metasomatism in the upper mantle: Evidence from Rare Earth and other Elements in
Apatite-Rich Xenoliths in Basaltic Rocks from Eastern Australia. Philosophical
Transactions of the Royal Society of London, A., 297, 333-346.

Webring M. (1985). SAKI: A fortran program for generalized linear inversion of gravity
and magnetic profiles, Open file report 85-122, United States Department of the
Interior.

Webster, R. (1975). Gems, Their Sources, Descriptions and Identification. Butterworths &
Co. Ltd, London, Sydney.

Wellman, P. (1971). The age and palaeomagnetism of the Australian Cenozioc Volcanic
Rocks. Ph.D. Thesis, Australian National University, Canberra, Australia. (Unpubl.)

Wellman, P. (1988). Intrusions beneath large intraplate volcanoes. Geological Society of

Australia, Abstracts No. 21. 9th Australian Geological Convention, February 1-5.

Wellman, P. and McDougall, I. (1974). Cainozoic igneous activity in eastern Australia,

Tectonophysics, 23, 49-65.

Wellman, P. and McDougall, I. (1974). Potassium- Argon ages on the Cainozoic volcanic
rocks of New South Wales. Journal of the Geological Society of Australia., 21, 247-
272.

Wilkinson, J.F.G. (1962). Mineralogical, geochemical and petrogenetic aspects of an

analcite-basalt from the New England district N.S.W., Journal of Petrology, 3 (2), 192-
214.

Wilkinson, J.F.G. (1966). Residual glasses from some alkali basaltic lavas from New
South Wales, Mineralogical Magazine, 35, 847.
Wilkinson, J.F.G. (1973). Pyroxenite xenoliths from an alkali trachybasalt in the Glen
Innes area, northeastern N.S.W., Contributions to Mineralogy and Petrology, 42, 15-32.

Wilkinson, J.F.G. (1974). Garnet clinopyroxenite inclusions from diatremes in the

Gloucester area, New South Wales, Australia, Contributions to Mineralogy and
Petrology, 46, 275-299.

Wilkinson, J.F.G. and Binns, R.A. (1977). Relatively iron rich lherzolite xenoliths of the
Cr-diopside suite: a guide to the nature of anorogenic tholeiitc andesite magmas,
Contributions to Mineralogy and Petrology, 65, 199-212.

Wilkinson, J.F.G. and Duggan, N.T. (1973). Some tholeiites from the Inverell area,
N.S.W., and their bearing on low pressure tholeiite fractionation. Journal of Petrology,
14, 339-348.

Willis, I.L. (1989). Dundee, Glen Innes, Maybole, Mount Slow, Rangers Valley, Red
Range, Shannon Vale, and Stonehenge 1:25,000 Geologic Sheets, Department of
Minerals and Energy, Armidale Branch, State Government Office Building, cnr
Faulkner and Dumeresq streets, Armidale, N.S.W., 2350.

Wright, J.B. (1969). High pressure phases in Nigerian Cenozoic lavas, distribution and
geotectonic setting. Paper presented at the IAVCEI symposium, "Volcanoes and Their
Roots" Oxford, Sept. 1969.

Yaemniyom, N. (1982). The petrochemical study of corundum bearing basalts at Bo

Phloi district, Kanchanaburi. M.Sc. Thesis, Chulalongkorn University, Thailand.
(Unpubl.).

Zwaan P. (1982). Sri Lanka: The gem island. Gems and Gemology, 18(2), 62-71.
 Appendix 1

PRE-VOLCANIC TERTIARY LANDSCAPE AND DRAINAGE OF THE

NEW ENGLAND REGION, NEW SOUTH WALES

The accompanying catalogue of maps provides a coverage of the topography and

drainage that existed prior to the Tertiary volcanic activity of the Central Province. Figure
A1-1 shows the 25 map sheets used in the analysis. The interpretation of sheets 1 to 24
was carried out at the 1:25,000 scale, and sheet 25 (Guyra) at 1:100,000 scale. Geological
maps, prepared by Stroud (1989), Willis (1989) and Brownlow (1989) were used as a base.
Geological boundaries between Central Province volcanics and older basement rocks were
overlain on the topographic sheets (New South Wales Government, Department of Lands).
Elevations for each mapped boundary were plotted and then contoured. The interpretation
used is conservative, that is to say, in areas of basalt cover the basement surface is
interpreted to dip gently and smoothly underneath and the least amount of basalt cover was
inferred in areas of poor control.
In order to improve the analysis, data from some 430 water bores (New South
Wales Government, Department of Water Resources) were also plotted on the 1:25,000
and 1:100,000 sheets. The logs, tabled numerically for each map sheet, are included in
Appendix 2. The logs are poorly described but serve to distinguish between basalt and
basement rocks, or at least indicate a minimum depth of basalt in a given borehole. The
bore data generally supported the basement topographic analysis and, in many cases,
indicated that the basalt filled valleys are steeper and deeper than are given by the
conservative analysis. After contouring the pre-volcanic topography for each 1:25,000
sheet, the information was reduced to 1:100,000 scale and linked to the surrounding sheets
in order to interpret the pre-volcanic drainage. Palaeodrainages are shown as heavy black
lines, which are dashed where the interpretation is less confident. It is important to note
that the interpretation of the pre-volcanic drainage becomes more uncertain in areas of
extensive basalt cover, especially where borehole information is sparse, or where different
relative ground movements have occurred during and after volcanic activity. The full size
plastic transparencies lodged with the New South Wales Government Department of
Minerals and Energy, Armidale Branch (cnr Faulkner and Dumeresq streets, Armidale,
N.S.W., 2350) should be used in conjunction with their equivalent topographic sheets
(New South Wales Government, Department of Lands) and should be refined as more
detailed geological boundary information and borehole data become available. The use of
these maps in exploration for diamond and sapphire deposits in the New England region is
discussed by Coenraads (1990).
The map sheets, as numbered in Figure A1-1, are described below:
 Appendix 1

PRE-VOLCANIC TERTIARY LANDSCAPE AND DRAINAGE OF THE

NEW ENGLAND REGION, NEW SOUTH WALES

The accompanying catalogue of maps provides a coverage of the topography and drainage
that existed prior to the Tertiary volcanic activity of the Central Province. Figure A1-1
shows the 25 map sheets used in the analysis. The interpretation of sheets 1 to 24 was
carried out at the 1:25,000 scale, and sheet 25 (Guyra) at 1:100,000 scale. Geological
maps, prepared by Stroud (1989), Willis (1989) and Brownlow (1989) were used as a base.
Geological boundaries between Central Province volcanics and older basement rocks were
overlain on the topographic sheets (New South Wales Government, Department of Lands).
Elevations for each mapped boundary were plotted and then contoured. The interpretation
used is conservative, that is to say, in areas of basalt cover the basement surface is
interpreted to dip gently and smoothly underneath and the least amount of basalt cover was
inferred in areas of poor control.
In order to improve the analysis, data from some 430 water bores (New South Wales
Government, Department of Water Resources) were also plotted on the 1:25,000 and
1:100,000 sheets. The logs, tabled numerically for each map sheet, are included in
Appendix 2. The logs are poorly described but serve to distinguish between basalt and
basement rocks, or at least indicate a minimum depth of basalt in a given borehole. The
bore data generally supported the basement topographic analysis and, in many cases,
indicated that the basalt filled valleys are steeper and deeper than are given by the
conservative analysis. After contouring the pre-volcanic topography for each 1:25,000
sheet, the information was reduced to 1:100,000 scale and linked to the surrounding sheets
in order to interpret the pre-volcanic drainage. Palaeodrainages are shown as heavy black
lines, which are dashed where the interpretation is less confident. It is important to note
that the interpretation of the pre-volcanic drainage becomes more uncertain in areas of
extensive basalt cover, especially where borehole information is sparse, or where different
relative ground movements have occurred during and after volcanic activity. The full size
plastic transparencies lodged with the New South Wales Government Department of
Minerals and Energy, Armidale Branch (cnr Faulkner and Dumeresq streets, Armidale,
N.S.W., 2350) should be used in conjunction with their equivalent topographic sheets
(New South Wales Government, Department of Lands) and should be refined as more
detailed geological boundary information and borehole data become available. The use of
these maps in exploration for diamond and sapphire deposits in the New England region is
discussed by Coenraads (1990).

The map sheets, as numbered in Figure A1-1, are described below:

 Map 1. Warialda 1:25,000 Sheet

The steep scarp of the Peel Fault, dominant on the Dinoga, Bingara and Mt Rodd Sheets, is no longer as
evident. It trends north from the Mt Rodd Sheet at about GR625210 where the scarp is of the order of 100 m
but this quickly reduces to zero. North of the township of Warialda the Peel Fault is no longer visible. Basalt
flows on the eastern portion of the Warialda Sheet indicate the existence of two northwesterly trending
palaeosystems. Only a small area of basalt is present on the western portion of the map lying in a
palaeotopographic low. Water bore 16326 (GR599313) supports this interpretation.
The westerly trending palaeochannel described on the Delungra Sheet continues via the Mount Rodd sheet
onto the southeastern edge of the Warialda Sheet, (GR755202), and northwest to GR700234, beyond which
point it cannot be traced due to the absence of basalt. Waterbore 49253 (GR775215) supports the presence
of this palaeochannel. Kellys Gully and its tributaries flow within these palaeovalleys.
The Warialda Creek palaeochannel trends westerly from the Cherry Tree Hill Sheet onto the Warialda Sheet
at GR820267. It possibly continues from here to GR760270 although no basalt remains to support this
suggestion. The presence of a palaeochannel is again supported by basalt from GR760270 to GR700290.
Warialda Creek flows within its palaeovalley and has cut through the basalt to the underlying basement
rocks along its entire length. The following tributary channels are defined from their up-channel limits to
where they join the main palaeochannel; GR745250 northwestwards to GR737277; GR758280
southwestwards to GR748270; and GR740300 southwestwards to GR710285.
Extensive basalt cover obscures the palaeotopography to the northeast of the Warialda Sheet. Water bore
14305 (GR773317) indicates however that the basalt has filled a palaeotopographic depression.
 Map 2. Mt Rodd 1:25,000 Sheet

The scarp of the Peel Fault continues to be a dominant feature on the Mt Rodd Sheet extending from
GR630070 to GR630200 although, as on the Dinoga and Bingara sheets, the throw on the fault scarp
appears to have become progressively less with distance north. Rugged land to the east of the fault ranges in
elevation from about 600 m asl in the south to about 500 m asl in the north, while on the west or
downthrown side the elevation is 400 m asl. Oakey Creek flowing west cuts across the Peel Fault at
GR633162. On the eastern half of the Mt Rodd Sheet basalt flows nested between basement high ground,
indicate a south-southwesterly trending palaeochannel, the palaeo-Sheep Station, joined by a number of
tributary channels. Sheep Station Creek flows within its palaeochannel from about GR790105 to GR755070
and beyond onto the Bingara Sheet. It has cut through the basalt to the underlying basement rocks. The
following tributary channels are defined from their up-channel limits to where they join the main
palaeochannel; GR797082 northwards to GR775112; GR795095 northwestwards to GR765115; GR820140
trending southwesterly from the Delungra Sheet to GR765115; GR755160 southwestwards and GR710150
southeastwards to GR750145; and GR700100 eastwards to GR760095.
More extensive basalt obscures the palaeotopography to the north of the Mt Rodd Sheet. Water bore 51276
(GR759187) indicates however the presence of a northerly trending palaeochannel or channels. Relief
inversion is taking place here with the basaltic high ground being drained to the south by tributaries of
Reserve Creek.
 Map 3. Bingara 1:25,000 Sheet
The steep scarp of the Peel Fault, as on the Dinoga Sheet, is a dominant feature on the Bingara Sheet
extending from GR660930 near Bingara to GR630060 in the north. Rugged land to the east of the fault is at
a general elevation of about 600 m asl, while on the west or downthrown side, the general elevation is 400m.
The Gwydir River (GR660940), Whitlow Creek (GR650970), Myall Creek (GR640000) and Harts Creek
(GR640023) cut across the Peel Fault and flow into the Stoddarts Valley. The Gwydir River is at a elevation
of 290 m asl at this point whereas the adjacent land, Molroy Trig (GR668954), is up to 590 m asl implying
some 300 m of downcutting.
On the western half of the Bingara Sheet basalt flows indicate a palaeodrainage system comprised of a
westerly trending palaeochannel joined by a number of northerly trending palaeochannels and a
southwesterly palaeosystem. The westerly trending palaeochannel continues from the Gum Flat Sheet
(GR820025) and can be traced to about (GR730036). Myall Creek flows within this palaeochannel and has
cut through the basalt into the underlying basement. The longest northerly trending basalt filled
palaeochannel can be traced for about 10 kilometres from GR810935 to GR795030 where it joins the
westerly trending Myall palaeochannel. The interpretation is supported by water bore 31250 (GR798011).
The palaeochannel is cut by the westerly flowing Whitlow Creek (GR787962) and northwesterly flowing
Horseshoe Creek (GR792990). Relief inversion has begun in places along the palaeochannel particularly
noticeable at GR795995 where the basalt is at 440 m asl with the land surface falling off into the adjacent
creeks. Two northerly flowing creeks, (GR792010 and GR805000), form twin laterals at this point
Another palaeochannel trends north from GR775975 to the Myall palaeochannel at GR780035, Horseshoe
Creek flows within this channel and has cut through the basalt into the underlying Carboniferous meta-
sedimentary basement. The third northerly trending palaeochannel is supported by water bore 31746
(GR748008) and joins the Myall palaeochannel at GR750035.
The Sheep Station palaeochannel trends southwest from the Mt Rodd sheet and probably joined the Myall
palaeochannel approximately in the position of their present confluence (GR724039). A westerly trending
tributary from GR791065 joins the Sheep Station palaeochannel at GR755060. Sheep Station Creek and the
tributary creek flow within the palaeochannels at this point, however at GR791065 relief inversion has
occurred with two southerly flowing tributaries (GR788065 and GR794065) of Rainbow Creek, forming
twin laterals.
Basalt remnants in a valley at about 450 m asl (GR767945) surrounded by Late Devonian-Early
Carboniferous metasediments up to 150 m higher in elevation indicates the existence of a westerly trending
palaeochannel here.
Whitlow Creek cuts across the metasediment ridge at 450 m asl flowing west (GR772955) indicating that
since the time westerly drainage was established downcutting of the order of 150 m has taken place.
 Map 4. Dinoga 1:25,000 Sheet

The Dinoga sheet is dominated by the steep scarp of the Peel Fault. Rugged land to the east of the fault
ranges in elevation from 700 m asl in the south to 600 m asl in the north along Serpentine Ridge while on
the west or downthrown side, the elevation is about 400 m asl in Stoddarts Valley. The Gwydir River flows
northwest, crosscutting the resistant fault parallel ridges of late Devonian-early Carbonifeous meta-
sediments coming across the Peel Fault into Stoddarts Valley just north of Bingara.
Isolated pockets of basalt, now relief inverted, indicate former valleys at GR820920, GR760920 and
GR755880.
To the west of the Peel Fault patches of basalt indicate the position of two northwesterly trending
palaeochannels at GR600810 and GR600850. Pallal Creek and Five Mile Creek flow within these valleys.
 Map 5. Cherry Tree Hill 1:25,000 Sheet

The palaeotopography falls from the east, where the highest basalt basement contact is 670 m asl
(GR060230), to the west. Although basalt covers a large portion of the Cherry Tree Hill Sheet, key
waterbores support the interpretation that the basement is lower below the basalt and that the basalt
represents flows within palaeotopographic lows. Water bores 619 (GR037229), 51481 (GR058252), 5919
(GR048276) and 6675 (GR964313) indicate the presence of a northerly trending palaeochannel which is
possibly a continuation of the Hobbs palaeochannel described on the Delungra Sheet.
Another poorly defined palaeochannel exists in the vicinity of Gunnee Hill (GR923255). Some relief
inversion has occurred here with the tributaries of Boundary Creek and Warialda Creek draining the basaltic
topographic high. Joined by a tributary from GR930240 the palaeochannel trends in a northerly direction
and possibly joins the palaeo-Hobbs.
The Gragin "deep lead" (GR820283) mentioned by MacNevin (1977) indicates the potential for these
palaeochannels to have diamond and tin bearing alluvium associated with their axis. The present drainage,
Black Creek, Boundary Creek, Warialda Creek and their tributaries drain the land surface to the west,
similarly to that described for the Copeton Dam, Gum Flat and Delungra Sheets. It is proposed that the
disruption of the northerly trending palaeodrainage and the establishment of the westerly system over these
sheets occurred 18-22 Ma due to West Central Province volcanism and associated uplift. Downcutting by
the post basaltic drainage becomes progressively less severe moving northwards from the Copeton area.
 Map 6. Delungra 1:25,000 Sheet

The highest basalt basement contact is 680 m asl (GR985089) along a ridge of Permian granite. For the most
part however, extensive basalt cover obscures the pre volcanic topography of the Delungra Sheet. Numerous
water bores indicate that the basement deepens below the basalt and therefore that the basalt represents lava
flows in palaeotopographic lows. Basalt thickness is in excess of 125 m in places (water bore 56416,
GR916183).
The main feature trending northeasterly across the Delungra Sheet is the Hobbs palaeochannel although the
position and depth of its axis are poorly defined. Commencing on the Gum Flat sheet, it runs onto the
Delungra sheet at GR850070. Tributaries join the Hobbs palaeochannel at GR900100 (which is described on
the Gum Flat sheet) and also at GR950180 and GR930140. Creeks such as Sheep Station, Dumboy and
Hobbs and their tributaries drain this area to the west and southwest.
On the eastern end of the sheet water bores 56938 (GR053093), 32817 (GR041124), 57764 (GR057166),
1106 (GR030169) and 1073 (GR025189) indicate that the palaeosurface falls eastward towards the palaeo-
Macintyre. Another palaeochannel at GR860200 trends westerly onto the Mt Rodd and Warialda Sheets. It is
supported by waterbore 10082 (GR880203)
 Map 7. Gum Flat 1:25,000 Sheet
The highest observed basalt basement contact is 750 m asl at GR035980, on the eastern end of the sheet. The
palaeosurface falls to the west where at GR840020 it is 370 m asl. Three major palaeorivers once traversed
the Gum Flat Sheet from south to north as evidenced by ribbons of basalt lying between basement highs.
The palaeo-Gwydir (GR068960) has been described on the Inverell sheet. Extensive volcanic cover obscures
the fall of the palaeotopography into this palaeovalley.
Easterly trending tributaries (GR020060) of the palaeo-Macintyre are also obscured by volcanics. Water
bores 51329 (GR003048), 51016 (GR017050) and 50877 (GR030035) indicate a fall in the elevation of the
palaeosurface below the volcanic blanket, although the positions of the palaeoaxes are not clear.
The next major palaeochannel to the west begins on the Copeton Dam Sheet trending north from GR960943
to GR902069 and then onto the Delungra Sheet where it joins the northeasterly trending Hobbs
palaeochannel. Water bores 27144 (GR939958), 7073 (GR934030) and 49293 (GR932039) have been
drilled into the basalt filling this palaeovalley. The Wonderland (GR969954) and Staggy Creek (GR955975)
mining areas are located on a northwesterly trending tributary of the above-mentioned palaeochannel. The
Wonderland lead consists of two isolated patches of basalt capped alluvium and the Staggy Creek lead forms
part of a system marked by patches of gravel. The streams flowed from the vicinity of Emu Hills
(GR970965) to the NW, then along the present course of Staggy Creek, joining the main palaeochannel
approximately where Staggy Creek crosses it (GR930015). It is possible that the main palaeochannel also
contains diamond and tin-bearing deep lead alluvium, although there has been no attempt to exploit these as
yet and this is probably due to the fact that this palaeochannel has not been previously recognized as such.
"Deep lead" alluvium has not been seen north of the Staggy Creek crossing due to the thickness of the basalt
cover. This system is now entirely disrupted by a number of westerly flowing creeks. Sandy creek, Staggy
Creek and Myall Creek flow west and cut the palaeochannel at GR918973, GR930015 and GR922055
respectively and then pass into its steep western valley wall composed of resistant early carboniferous
volcanic lithic greywacke. Assuming that the original height of the land surface on which the westerly
flowing streams were set up was at least as high as the greywacke ridge through which they are now cut,
then there has been at least 90 m of downcutting (0.005 mm/year since 18 Ma). The streams have not yet cut
through the basalt flows lying within the palaeovalley. The westernmost palaeosystem is comprised of two
tributaries (GR830970 and GR860962) meeting at GR845985 and trending northwest to GR826023. Pockets
of basalt mark its position and Kangaroo and Ironbark Creeks flow within the palaeovalley for the majority
of their length. Relief inversion has occurred of the uppermost portion of the tributaries on this sheet and the
Copeton Dam sheet. In the vicinity of Leaderville (GR867949), tributaries of Ironbark Creek and Daffeys
Creek flank the palaeochannel, now a basalt ridge. Water bores 34227 (GR866958) and 34936 (GR870958)
support this concept and indicate the presence of a palaeochannel beneath the basalt ridge at "Caroline".
Tributaries of Kangaroo Creek form twin laterals to the palaeochannel (now a basalt ridge) at GR834934.
On the northwestern corner of the Gum Flat sheet (GR840060) the Hobbs palaeochannel trends in a
northeasterly direction onto the Delungra sheet. Hobbs Creek flows in a southwesterly (reversed) direction
within this palaeovalley.
The presence of a formerly northerly trending palaeodrainage direction cross cut by a now predominantly
westerly flowing network is particularly well displayed on this sheet. It is proposed that the disruption of the
old system and establishment of the new, occurred 18-22 Ma due to West Central Province volcanism and/or
associated uplift.
 Map 8. Copeton Dam 1:25,000 Sheet

Only a few isolated areas of basalt remain on the Copeton Sheet. A palaeochannel of historic significance,
the palaeo-Gwydir, has been extremely well defined by the presence of diamond bearing alluvium along the
length of its palaeoaxis. Where pods of this material remain, as "deep leads" protected by basalt caps, they
have been extensively mined for diamond and tin (MacNevin,1977). The palaeo-Gwydir trends east from a
divide at GR962922, 650 m asl and joins the northern flowing tributary at Crown Jewel Hill on the Tingha
Sheet. Old workings indicate the exact position and elevation of the palaeoaxis; from the west Oaky Creek
(GR007920), Kirks Hill (GR014920), The Banca (GR021920) and Stockyard Hill (GR041929) are at 620 m
asl, Malacca (GR052918) is at 620 m asl and Soldier Hill (GR067906) is at 590 m asl.
The fall of this important palaeochannel to the east was originally verified by Cotton (1914) who made
numerous aneroid elevation readings of deep leads. The Staggy Creek and Wonderland diamond bearing
leads on the Gum Flat 1:25,000 sheet form part of a NNW trending tributary of the Hobbs palaeochannel
and are not part of the same system as the above-mentioned leads which occur at a lower elevation in the
vicinity of Copeton Dam. They must come from a separate source, as was first suggested by Stonier (1895).
Another palaeochannel trends northwest onto the Gum Flat sheet from the divide at GR962922.
On the northwest corner of the Copeton Dam sheet remnant tongues of basalt, now relief inverted, indicate
the presence of northerly trending palaeochannels at GR837926 and GR865927.
The present day Gwydir River flows west, having been deflected from its originally northerly trending
channel (described on the Inverell Sheet) by basalt from a source west of Inverell. At the base of Copeton
Dam the Gwydir River is at 471 m asl, indicating some 180 m of downcutting since the original, easterly
flowing tributary through the mining areas was displaced, probably 18-22 Ma. That represents a rate of
downcutting of some 0.01 mm/year. The westerly flowing Sandy, Hughies, Staggy, Myall and Spring creeks
on the Gum Flat 1:25,000 sheet also show evidence of downcutting of the order of 100 m.
 Map 9. Bukkulla 1:25,000 Sheet

The highest observed basalt basement contacts are 660 m asl at GR070270 and 700 m asl at GR293240. The
palaeosurface dips from the east and west into the centre of the map sheet forming a broad valley which has
been filled with basalt. Water bores 55279 (GR204249), 49294 (GR181246) and 33513 (GR203315)
support the existence of a palaeovalley here, although further work to the north of Bukkulla sheet is required
to determine the position of the palaeoaxis and the original flow direction.
The present Macintyre River flows northwards within the palaeovalley to GR170319 at which point it
swings to the northwest.
The northwesterly trending palaeo-Frazer appears to have joined the main Macintyre palaeochannel at
(GR245225), the present stream however appears to have been deflected to the north at (GR253227) by the
18-22 Ma volcanic activity which filled the palaeovalleys. Frazers Creek is now constrained to flow between
volcanic high ground on the west and basement highs on the east.
Bannockburn Creek from GR088215 to GR129301 and Frazers Creek from GR253228 to its junction with
Weean Creek and Weean Creek to GR222352 have incised themselves into the basement on either side of
the basalt filled palaeo-Macintyre and flow parallel to it, forming twin laterals.
 Map 10. Nullamanna 1:25,000 Sheet

The highest observed basalt basement contact is 720 m asl at GR303145, the palaeosurface falling of sharply
(by at least 200 m) to the west into the palaeo-Macintyre. The western two thirds of the Nullamanna sheet is
blanketed by basalt flows.
The palaeo-Macintyre continues north from the Inverell sheet, but north of about GR170100 the lack of deep
bores and extensive volcanic cover make it impossible to determine the position and depth of the palaeoaxis.
The channel may continue north on either the east or west side of the basement high at GR175140) or it may
head northwest. The volcanic pile in the palaeo-Macintyre is at least 200 metres thick. Frazers Creek and
Byron Creek are twin laterals and flank the palaeo-Macintyre valley to the north of Inverell but in this case
the present day Macintyre still flows in the original basalt filled channel, illustrated schematically in
Coenraads (1990). The present Macintyre River meanders north and Spencers Gully, Rob Roy Creek, Jessies
Gully, Hunts Gully and Waterfords Gully flow into it, draining the volcanic high ground to the west. This
drainage probably developed as a result of the volcanic activity 18-22 Ma.
The easterly trending palaeo-Swan Brook enters the Macintyre palaeovalley at GR205085.
Frazers Creek at GR280200 flows northwest along the northeastern edge of the basalt pile. It seems that it
has been deflected to this position from a formerly more westerly trend, as has clearly occurred to the north
on the Bukkulla sheet.
 Map 11. Inverell 1:25,000 Sheet

The palaeosurface falls away from the south where the highest exposed basalt-basement is at 800 m asl
(GR148938) and also from the west and east towards the centre and north of the sheet where, in the vicinity
of Inverell water bore 2524 (GR191047) indicates basement to be deeper than 447 m asl. A network of
palaeodrainage joins in the vicinity of Inverell and heads north. The palaeo-Macintyre heads west from the
Elsmore Sheet marked by a low ridge of basalt hills (GR288030) & (GR260065). The presence of a valley
here is supported by water bore 61426 (GR284061) and numerous bores in the vicinity of Inverell. The
palaeo-Gwydir heads north from the Tingha Sheet at (GR075945) then most likely swings northeast towards
Inverell. The exact position of the palaeoaxis is unclear due to the extensive volcanic blanket here which the
water bores indicate to be of the order of 100 m thick. The "severing" of the northwesterly trending palaeo-
Gwydir by basalt flows approximately 18-22 million years ago was first described by Cotton (1914).
Observations of flow indicators in the basalts by Smith (1988) indicate that the lavas flowed from north to
south in this palaeovalley. Today Cooks Creek (GR080010), Fern Creek (GR100020) and Auburn Vale
Creek (GR110990) flow in a southerly (reversed) direction within the basalt filled channel towards Copes
Creek and the Gwydir River.
To the north east of the present basalt water divide, Spring Creek (GR1600300) and its tributaries flow in a
northeasterly direction into the Macintyre River.
The northerly trending Sheep Station palaeochannel also joins the palaeo-Macintyre and palaeo-Gwydir in
the vicinity of Inverell (GR170010). It is marked by tongues of basalt and is possibly the northerly
continuation of the palaeochannel at Inverell Airport described on the Tingha Sheet. Basalt flows mark two
tributaries to the palaeo-Sheep Station at (GR163980) and at (GR160949). Water bores 53277 (GR160947)
and 29028 (GR157944) help constrain the position and depth of the palaeoaxis of the second tributary.
Sheep Station Creek flows within the palaeovalley until (GR171000) at which point it turns eastwards to
Middle Creek. At Gilgai (GR180960) Gilgai Creek, flowing northeasterly, cuts through the palaeochannel
and is now at a level 20 m deeper than the palaeochannel axis (which is 750 m asl at this position).
 Map 12. Tingha 1:25,000 Sheet

The palaeosurface falls from the east where Tertiary alluvium is found overlying basement at 900 m asl
(GR305895) to 580 m asl in the west (GR083920). The well defined northerly trending palaeochannel, the
palaeo-Gwydir on the western edge of the Tingha sheet was first recognized by Stonier (1895). Remnant
patches of valley flow basalt protect diamond bearing Tertiary alluvium in the palaeochannel and the old
diamond mines mark the exact position of the palaeoaxis. Some of these are Ryders Lead (GR076804),
Kenzies Claim (GR074813), Collas Tunnel (GR088865) and Streak of Luck (GR090884), they are
described in detail by MacNevin (1977). Maids Creek flows in the palaeovalley to (GR079856) at which
point it leaves and forms a lateral on the west side of Collas Hill.
Another palaeochannel from the west joins in the vicinity of Crown Jewel Hill (GR088888).
On the high ground to the northeast of the Tingha sheet, patches of basalt such as that at Inverell Airport
(GR210910) and Tertiary Alluvium such as that at GR250900 and GR290890 indicate the presence of
former, generally northwesterly trending channels. It is possible that the major westerly flowing
palaeochannel on the Indiana Sheet which comes onto the Tingha Sheet at GR310863 and the palaeochannel
at Inverell Airport (GR200920) are part of the same system. However, all evidence that may have supported
this, has been removed by Copes Creek. Copes Creek and Middle Creek flank the high ground on the
southwest and northeast respectively and their small tributaries drain this area. Their positions on either side
of, and parallel with the palaeochannel suggest that they may be lateral streams.
 Map 13. Hurricane Hill 1:25,000 Sheet

The highest observed basalt basement contact on this sheet is seen at the very northern edge of Kings Plains
(GR490279) 990 m asl. The Hurricane Hill sheet covers the northern edge of Kings Plains. Two southern
trending basalt filled palaeochannels can be seen at GR480250 and GR525235 which join the westerly
trending Kings Plains System on the Sapphire sheet. The westernmost palaeovalley contains a present day
stream, the easternmost however is starting to show some relief inversion with Stoney Creek located on the
basalt-basement contact (GR528240) and downcutting at this position. To the north of Kings Plains the
palaeosurface drops rapidly to a level of about 800-850 m asl as indicated by the height of the basalt-
basement contact in the vicinity of Hurricane Hill (GR474334). A northwesterly flowing palaeochannel is
marked by the basalt here and the elevation of Hurricane Hill, 942 m asl, indicates that there would have
been at least 130 m of basalt in this channel.
Northerly trending present day drainage such as Kings Plains Creek, Five Mile Creek, Arrawatta Creek and
Stoney Creek are actively cutting into the plateau of Kings Plains by headward retreat along north-south
trending lineaments or planes of weakness (Coenraads, 1988). The best example is Arrawatta Creek
(GR510290). Kings Plains Creek flows within its basalt filled palaeochannel until it is captured and diverted
to the north at (GR456220). At (GR310260) a tongue of basalt marks the position of a northerly trending
palaeochannel. Tributaries of Weean Creek flank the basalt to the east and west at GR320270 and
GR307270 respectively, forming twin laterals and creating relief inversion.
The palaeochannel probably continued to GR306325 at which point it joins with a south-westerly trending
channel and heads northwest. Mordington Creek (GR330320) flanks the southwesterly trending channel for
a few kilometres as a lateral; it joins Weean Creek which cuts across the main palaeochannel at GR308310.
 Map 14. Sapphire 1:25,000 Sheet
The highest observed basalt basement contact is on Kings Plains, 980 m asl at GR540192. Kings Plains is a
elevated, large, relatively flat area, largely covered by volcanics occupying the eastern portion of the
sapphire 1;25,000 sheet. The basement beneath Kings Plains is also elevated being all over 850 m asl with a
significant proportion over 900 m asl. The highest point is Mt Buckley, 1168 m asl on the southern edge of
Kings Plains (GR548078; Elsmore 1:25,000). Mt Buckley is comprised of basalt flows implying that the
volcanic pile may have been in excess of 200 metres here. The palaeosurface falls sharply to the west and to
the south into the palaeo-Swan Brook, that is of the order of 200 metres over a few kilometres. This
observation and also the observation that that the southern and western boundary of Kings Plains are
relatively straight may lead one to speculate that Kings Plains is a fault bounded, raised, basement block. In
any case, the presence of basalt at all elevations on the escarpment indicates that the Kings Plains was a
basement topographic high prior to the finish of Central Province volcanism.
The Sapphire Sheet is dominated by the westerly flowing Kings Plains-Frazers Creek palaeosystem.
The presence and location of the Kings Plains palaeovalley is quite accurately pinpointed with the aid of
water bores 32244 (GR544135), 32241 (GR525134), 32239 (GR535138), 31273 (GR497177), 31272
(GR507209) and 51436 (GR486201); exploration test holes by T.J.& P.V.Nunan Pty.Ltd., NA1
(GR49951600), NA2 (GR49891600), NA3 (GR49831602) and NA4 (GR49771605); test holes by Dept of
Mineral Resources (Lishmund 1987) KP1 (GR48392064) and KP2 (GR48202086) and test holes by
Jingellic Minerals N.L. (Coenraads, 1989) KP3 (GR48482058) and KP4 (GR48392074). The Kings Plains
Eastern and Western Feeder palaeochannels join at GR507177, the channel continues northwards, turning
sharply west at GR510210 then south-westwards to join the palaeo-Frazer at GR420170. Kings Plains Creek
flows within its palaeovalley, although, in places it has become localized by the basalt basement contact on
the northern side of the basalt filled valley. It follows the contact as a lateral stream to the palaeovalley for
approximately 4 kilometres westerly until it is captured (at GR456220) and diverted northwards into a steep
narrow valley formed by headward erosion controlled by a prominent north-south lineament. Arrawatta
Creek, actively eroding headwards along another north-south lineament, has not yet captured Kings Plains
Creek. Economic quantities of sapphire have been recovered from the Kings Plains System.
Another palaeotributary rising at GR455125 heads west through Whites Hill (GR432121) and joins the
palaeo-Frazer at GR385143. Relief inversion can be seen in the headwaters of this tributary with Horse
Gully and Mary Anne Creek forming twin laterals flanking White Hill to the north and south respectively.
Economic quantity of sapphires have also been recovered from these creeks downstream from the point at
which their valleys widen and their gradients drop.
The palaeo-Weean heads northward from GR369207. Weean Creek flows within its palaeochannel at this
point.
 Map 15. Elsmore 1:25,000 Sheet

The maximum height of the basalt basement contact is 920 m asl at GR540967 and GR547063. The
palaeotopography falls into two major channels trending west. The palaeo-Swan Brook and the palaeo-
Macintyre (Kings Creek).
The palaeo-Swan rises on the Stonehenge Sheet, then heads westward onto the Elsmore Sheet at GR549040.
It joins the main channel in the vicinity of Kerosene Hill (GR430040). The main channel is completely relief
inverted standing out as a west northwesterly trending ridge of flat topped or slightly concave upwards hills
including Table Top Hill (GR492022), Kerosene Hill and Langari Hill (GR361059). Wet Gully and Swan
Brook to GR380068 and Kings Creek to about GR410010 are twin laterals, flanking the palaeo-Swan. Swan
Brook downstream from GR380068 and Kings Creek from GR400986 return to their respective
palaeovalleys which are no longer relief inverted.
In the vicinity of Braemar (GR348033) the basalt basement contact outcrops at a maximum of 750 m asl.
The basement high ground forms a northwesterly trending ridge of height 760 m asl at GR360023, 722 m asl
at GR338040 and 719 m asl at GR344062. The ridge separates the palaeo-Swan from a basalt filled tributary
of the palaeo-Swan at GR320050. The fall of the palaeosurface into this tributary is supported by water bore
19682 at GR329047.
The Braemar sapphire deposit is situated in a saddle in the basement ridge marking the presence of a small
basalt filled palaeochannel trending northeast into the palaeo-Swan. The palaeochannel is well defined here
by Department of Minerals and Energy diamond drill holes 1, 4, 5, 6 and 7 (Brown and Pecover, 1986).
The other major palaeochannel rising on the Maybole and Indiana Sheets is also marked by a prominent
ridge of high ground including Carters Mountain (GR539953) and Eaglehawk Mountain (GR508950). The
fall of the palaeosurface beneath this ridge is supported by waterbore 31080 (GR539971). Kings Creek to
GR410010 and Paradise Creek to GR450940 are twin laterals flanking the ridge to the north and south
respectively.
 Map 16. Indiana 1:25,000 Sheet

The maximum height of the basalt-basement contact is 1130 m asl at GR518820 in the southeast corner of
the Indiana Sheet. The height of the palaeosurface falls progressively to the northwest. Extensive volcanics
cover the eastern side of the sheet (adjoining the Maybole Sheet) obscuring the palaeotopography in this
area. However waterbore 31076 (GR515920) indicates that the height of the basement is lower by at least
30-90 m than the surrounding basement outcrops. The exact trend of this palaeovalley or valleys over which
the the volcanics are situated is not clear, however channels probably head north on either side of the
basement high (GR492932) joining the palaeo-Kings Creek and the palaeo-Macintyre on the Elsmore Sheet.
The palaeo-Macintyre (or Kings Creek) follows a northwesterly path an is traceable from about GR490840.
For the majority of its path, the Macintyre River has cut through the basalt cover exposing basement rocks in
its channel.
A basalt filled palaeovalley at GR362925 containing Oaky Creek within it, flows northward also joining the
palaeo-Macintyre. A northwesterly trending palaeochannel containing tertiary sediment is found at
GR335900. Lowes Creek flows within this channel. Two major basalt-filled palaeochannels run west,
parallel to, and north of Copes Creek, joining at about GR370870. Sheep Station Creek flows, in part, in the
southernmost palaeochannel leaving it at GR392859 to flow southwest into Copes Creek. Snake Creek,
Dead Horse Creek and Boney Creek cut across the palaeochannel at GR354870, GR365870 and GR380865
respectively flowing southward to join Copes Creek. The headwaters of Copes Creek flow for a short
distance in a westerly trending, basalt filled, palaeovalley at GR465835.
 Map 17. Rangers Valley 1:25,000 Sheet

The dominant feature of this sheet is the Reddestone palaeochannel. Patches of basalt mark its former path
from the vicinity of Clarevaulx Lagoon (GR720230) towards the northwest (GR550350). Insufficient data
exist to say wether the major palaeochannel passed through the north or the south of the basement high
ground in the vicinity of Balmoral (GR665250). To the north McCarthys Creek (GR660265) flows within
the palaeochannel and to the south, a tributary of Black Plain Creek (GR670235).
The tongue of basalt at (GR698270) marks a former tributary to the palaeo-Reddestone. Here tin has been
mined from deep-lead gravels below the basalt (T.J. Nunan, pers. comm.). Two more small basalt filled
channels also form tributaries: they are the "Isolation Prospect" channel (GR570300) and the "Lighthouse
Prospect" cannel (GR580270). The northerly flowing Wellingrove Creek cuts across the palaeo-Reddestone
at GR620280
The highest exposed basalt basement contact is 1040 m asl (GR695270). The height of this contact continues
to fall to the northwest along the palaeo-Reddestone to 780 m asl (GR552338).
 Map 18. Glen Innes 1:25,000 Sheet

The palaeodrainage on this sheet flows essentially north. The dominant feature is the Reddestone
palaeochannel and, although the exact position of its axis is obscured by an extensive volcanic blanket, it is
constrained to lie between basement high ground exposed at GR680180 and GR720170. The presence of a
channel below the volcanics is supported by waterbores 29945 (GR702214), 54915 (GR682169), 49192
(GR709144) and 51818 (GR724143).
The present Reddestone Creek flows in its flat-floored, basalt filled palaeovalley until GR730205 at which
point it is captured and diverted to the northeast. Wellingrove Creek (GR:600200) appears to be a lateral to
both the palaeo-Reddestone and the palaeo-Kings Plains.
Furracabad palaeovalley bends to the northwest in the vicinity of Glen Innes township and is supported by
bore 51163 (GR745089). The height of the valley floor under the township of Glen Innes, as indicated by
numerous water bores, is around 1030 m asl (Eg. bore 54918, GR765097) or deeper. It is not clear where the
Furracabad palaeochannel continues from here. To the northeast, the present Furracabad Creek enters an
anomalously narrow valley cut in the basement rocks at 1040 m (GR780158). It seems unlikely that this
marked the trace of the original Furracabad palaeovalley unless post-volcanic block faulting has raised the
area to the northeast of the potential fault line at (GR775155) and downcutting by the present Furracabad
Creek has since taken place. It is more likely that this represents a capture and diversion of Furracabad
Creek and that the palaeo-Furracabad valley swung round to the northwest to join the palaeo-Reddestone.
The highest exposed basalt basement contact is at 1080 m asl (GR723096) on the southern edge of the sheet.
The height of the contact continues to fall to the north.
Other basalt filled palaeovalleys include Back Plains Creek (GR655190), the headwaters of Punchs Creek
(GR630160), Pine Creek (GR625130), Wellingrove Creek (GR620090), Maids Valley Creek (GR590090)
and at (GR583180).
A small valley containing sapphires and zircon trapped amongst basement rock, but with no volcanic rocks
in its catchment, is situated at (GR630188). It was formerly a mining area and now is a public fossiking area.
Wellingrove Creek has cut through the volcanic pile into basement and is a lateral to the palaeo-Reddestone
system to the east and the palaeo-Kings Plains system to the west.
 Map 19. Stonehenge 1:25,000 Sheet

Interpretation of the Stonehenge sheet poses the same problem as that of the Maybole 1:25,000 sheet to the
south, with very few exposures of basement rocks showing through the volcanic cover. The highest basalt
basement contact exposed is at 1240 m asl (GR735040). The elevation of the basement falls to the north,
east and west. Towards the edges of the sheet where basement rocks are again exposed, basalt is seen to
have flowed away from the central portion of the map into the pre existing drainage. On the northern corner
of the sheet, basalt flows at GR750070 and GR764047 surrounded by basement highs indicate pre basaltic
drainage to the northeast. This observation is supported by waterbore 54427 (GR764047).
Basalts fill the palaeo-Reddestone, a major pre basaltic channel trending in northerly direction on the
northern end of the sheet, and at this position the present Reddestone Creek still flows within the confines of
its original palaeovalley. Lying under extensive volcanic cover, the exact position of the headwaters of the
palaeo-Reddestone is unclear.
To the northwest at GR558063, GR572058 and GR574048 tongues of basalt have flowed into the westerly
trending palaeo-Swan Brook. At GR581072, waterbore 28736 indicates the presence of Maids Creek
palaeovalley trending northward.
To the southwest at GR566983 and GR575970 tongues of basalt have flowed into the palaeo-Kings Creek.
The White Rock Volcano centred on Spring Mountain volcanic plug (GR579003) is very similar in
appearance to the Maybole Volcano to the south on the Maybole sheet. White Rock Mountain (GR595010),
elevation 1330 m asl, and Arthurs Seat (GR616025), elevation 1244 m asl, are comprised of basalt flows
putting the height of the volcanic pile at approx. 350 m above the highest observed basement in the area.
Present day streams in this area, controlled by the volcanic pile flow out radially. These are the headwaters
of Spring Creek, Swan Brook, Kennedys Gully, Bessies Creek and tributaries of Falls Creek and Little Oaky
Creek. Falls Creek from (GR608995) to its junction with Wellingrove Creek downstream to (GR630053)
appear to describe and arc around the eastern side of the structure.
 Map 20. Maybole 1:25,000 Sheet

This sheet covers the heart of the East Central Province hence there are very few exposures of basement
rocks showing through the volcanic cover. The highest exposed basalt basement contact is 1360 m asl in the
vicinity of Mt Rumbee (GR: 675855). Mt Rumbee itself (GR653856) is the highest point in the Central
Province at 1503 m asl which puts the volcanic pile at least 140 m above the highest basement in this area.
The Maybole Volcano, described by Pecover (1987), of which the Gough Sugarloaf (GR683916) is a plug,
is situated on this sheet. The present day drainage appears dominated by this structure. Very little basement
rock is visible through the thickest and highest part of the volcanic pile, stretching from south of the Ben
Lomond area on the Guyra Sheet to the north of Maybole on the Glen Innes. In this area the pre basaltic
drainage cannot be determined. A post-volcanic radial and annular pattern, controlled by the volcanic pile,
has been set up. As described by Pecover (1987), the headwaters of the Macintyre River, Paradise Creek,
Falls Creek, Oakey Creek, Wellingrove Creek, Furracabad Creek Stonehenge, Gara River and Moredun
Creek are radially arranged. The headwaters of Beardy Waters and Grahams Valley Creek form an annular
system around the Maybole Structure.
To the east of the sheet basement is exposed at about 1100-1200 m asl and the basalt is seen to have flowed
into palaeochannels of northeasterly orientation. The northeasterly trending palaeochannel, in the vicinity of
Fern Hill (GR736911) was filled by basalt and now has two tributaries of Beardy Waters flowing on it. The
palaeo-Marowan also flowed to the northeast below the course of the present Marowan Creek. Bore Number
54465 (GR743847) supports this.
To the southwest of the sheet basement is again exposed at 1100-1200 m asl. Another palaeochannel marked
by the present position of Back Creek (GR575810) trends in a southerly direction.
 Map 21. Dundee 1:25,000 Sheet

Only two small patches of basalt exist on the Dundee 1:25,000 sheet. The maximum height of the basalt
basement contact is 1100 m asl at GR930340.
 Map 22. Shannon Vale 1;25,000

Volcanics only occupy the south west corner of the Shannon Vale Sheet, the highest point where the basalt
basement contact is exposed is at 1130 m asl at GR850110. The relief inverted palaeochannel at GR840090
continues north from the Red Range Sheet to GR843110 at which point it is not clear whether it continues
north or swings west to join the palaeo-Furracabad. North of GR831132 Beardy Waters enters a narrow
bedrock valley and no more basalt exists in the vicinity of the channel beyond this point. It seems unlikely
that the palaeochannel went this way unless faulting has uplifted the area to the north of the dashed line on
the map and downcutting by modern streams has affected the appearance of the palaeochannel.
 Map 23. Red Range 1:25,000 Sheet

The basalt basement contact falls to the north across the Red Range Sheet, the highest point 1230 m asl is at
GR908952. From this point a prominent palaeochannel marked by basalt flow remnants proceeds north to
GR970070. John Ryall Lagoon (GR955045) and the headwaters of Oaky Creek are situated within the
palaeovalley at its northern end. The palaeochannel is severed by the more northeasterly flowing present
drainage. That is by Sheep Station Creek at GR940994 and Concertina Creek at GR950023.
Another northerly trending palaeochannel is indicated by basalt in between basement highs at GR909045.
Middle Creek flows in this palaeochannel.
At GR828050, waterbore 52317 supports the interpretation that the northerly trending basalt ridge, or
extension to the Great Divide is a relief inverted palaeochannel. Beardy Waters and the Mann River are twin
laterals to this palaeochannel. Bore 53448 (GR808069) indicates that the palaeosurface is falling into the
palaeo-Furracabad below Glen Innes.
A palaeovalley, or valleys exist below the volcanic pile upon which Blair Hill Lagoon is situated
(GR818960) although it is ambiguous as to their direction. They may have flowed to the west or to the north.
 Map 24. Mt Slow 1:25,000 Sheet

Adjoining the eastern edge of the Maybole Sheet, the maximum height of the basalt basement contact is
1330 m asl at Ben Nevis (GR843837). Ben Nevis itself is 1409 m asl, indicating a fall in elevation to the east
from Maybole and that the volcanic pile is at least 80 m thick at this position.
On the western edge of the sheet the continuation of palaeochannels heading to to northeast from the
Maybole high ground can be seen. They are palaeo-Mann (GR800830) and the palaeo-Marowan
(GR796923). Most prominent in the middle of this sheet is the north-south trending string of basalt outcrops.
In the vicinity of GR930900 the basalt flow remnants exist in the palaeochannel at a height of 1180-1240 m
asl surrounded by basement valley walls to about 1270 m asl. Mt Slow Creek flows to the north within this
valley. At GR920840 some relief inversion of the basalts marking the palaeovalley has started to occur. The
Henry River is a lateral for 3 km on the west side of the basalt from GR912836 to GR928862 at which point
it swings east. There is insufficient data to say whether the palaeochannel also went east or north or
northwest. Possible saddle points exist at Mt Slow (GR931876) and (GR945876).
The present day drainage on this sheet trends north-northeasterly flowing almost exclusively on basement
rocks.
 Map 25. Guyra 1:100,000 Sheet.

The present day drainage and palaeodrainage are both radially arranged around the areas of high ground in
the vicinity of The Brothers (GR716778) and Guyra (GR720560). These areas of highest elevation also
represent areas of maximum basalt thickness and cover, making it difficult to determine the nature of the
palaeodrainage.
The natural lagoons, Llangothlin, Little Balbair and Mother of Ducks are situated on the relatively flat basalt
covered areas and are most likely formed in a manner similar to the other Central Province lagoons
described by Coenraads (1989). The bore data, where they are deep enough, support the observation that the
basalt has flowed into, and filled, earlier drainage. A clear example is the large tongue of basalt high ground
projecting to the south east in the vicinity of Chandlers Peak (GR853530). It fills the Boundary
palaeochannel which lies roughly below the present position of Boundary Creek. The Aberfoyle River and
Wollomombi River form twin laterals to this basalt tongue.
Another tongue of basalt at Mt Killalee (GR795351) projecting in a southerly direction towards Armidale
marks the position of the palaeo-Gara River.
The highest basalt/basement contact of 1360 m asl. is recorded in the vicinity of Llangothlin Lagoon
(GR805770). The elevation of the contact falls to the south, east and west.
 Appendix 2

DESCRIPTION OF WATER BORES DRILLED IN THE NEW ENGLAND

REGION, NEW SOUTH WALES

The accompanying table provides a summary of the water bores drilled into the
Tertiary volcanic rocks of the Central Province in the area covered by the 25 map sheets
shown in Figure A1-1. All bores drilled for water are catalogued by the New South Wales
Government, Department of Water Resources. Their offices are located at 10 Valentine
Ave, Parramatta, New South Wales, 2150.
Some 430 water bore logs are tabled numerically for each of the 1:25,000 and
1:100,000 sheets described in Appendix 1. The logs are poorly described but serve to
distinguish between basalt and basement rocks, or at least indicate a minimum depth of
basalt in a given borehole.
Other exploration drilling and testing carried out in the New England region by
T.J. & P.V. Nunan Pty. Ltd and Jingellic Minerals N.L. are described in Appendix 4.
 Appendix 3

GEOPHYSICAL ASSESSMENT; KINGS PLAINS AND DUNVEGAN

LAGOONS

A3.1 Fieldwork

The natural lagoons chosen for the investigation are listed with their grid
references below and aerial photographs are shown in Chapter 6:

Lagoon Grid Reference & Map Sheet

-Kings Plains Lagoon GR:484206 Inverell 1 : 100,000 (Fig. 6-2)
-Dunvegan Lagoon GR:728210 Glen Innes 1 : 100,000 (Figs 6-3 & 6-4)
-Clarevaulx Lagoon GR:720225 Glen Innes 1 : 100,000 (Fig. 6-4)
-John Ryall Lagoon GR:955045 Glen Innes 1 : 100,000 (Fig. 6-5)
-Barley Field Lagoon GR:774905 Glen Innes 1 : 100,000 (Fig. 6-6)
-Inverell Racecourse GR:180030 Inverell 1 : 100,000

The geophysical surveys were designed to assess the hypothesis proposed by

Pecover (1987) that the lagoonal features of the Central Province are maars and to
determine optimum locations for the testing of their economic viability.
Initial computer modelling was carried out at Macquarie University based on the
assumption that any feeder vents could be represented as vertical pipes or dykes. The
purpose of the modelling was to determine the minimum size and physical property
contrasts necessary for such a feature to be detectable and to determine optimum station
spacing. It was determined that a pipe of diameter 100 m would be just within the limits of
detection giving a 2 μms-2 (0.2 mGal) anomaly with a density contrast of 100 kgm-3.
Adequate anomaly definition required a 50 m sample interval. Near surface noise
problems were anticipated for the magnetics as indicated by Spencer (1987) and therefore
a finer sample spacing deemed necessary.
During the period 15-26 July 1987 a field party consisting of 4 persons carried out
geophysical surveys over the Kings Plains and Dunvegan structures in the Glen Innes-
Inverell area.

A3.1.1 Gridding
A grid covering an area of approximately one square kilometre (about 150 stations)
was erected at each locality. A 100 m. grid interval was chosen with a 50 m grid within the
lagoon area to ensure adequate definition of any geophysical anomaly within this area.
A3.1.2 Levelling
The grids were levelled, using a levelling telescope and 4 metre staff, to an
accuracy of 2 cm for the purpose of applying elevation corrections to the gravity data.
Levelling was completed for the Kings Plains grid. Owing to shortage of time, however,
levelling for the Dunvegan grid was carried out by David Morris of J.I. Noad & Co.,
Inverell.

A3.1.3 Gravity
Gravity was measured at each grid locality using a Canadian CG-2 gravimeter
(provided by the New South Wales Department of Minerals and Energy). The aim of the
survey is to define any gravity anomaly that may exist due to a density contrast between
the materials below the lagoon area.

A3.1.4 Magnetics
The earth's total magnetic field was measured at 10 m spacings within the inner
portion of the grid (covering the lagoon floor and flanks) and at a 20 m spacing for the
outer portion of the grid. A Geometrics proton precession magnetometer was used
(provided by the New South Wales Department of Minerals and Energy). The aim of the
survey was to define any magnetic anomaly that may exist due to differing magnetic
susceptibilities between the materials below the lagoon area.

A3.1.5 Seismic
Seven reversed seismic refraction lines were carried out using a Geometrics
Nimbus single channel, signal enhancement seismograph with a hammer source (provided
by Geo-Instruments Ltd.). They were carried out both on the lagoon floor and rim in order
to ascertain the seismic nature of the near surface materials and to define any near surface
sub-horizontal layering. Knowledge of these factors assisted in the gravity and magnetic
modelling studies.

A3.2 Data Reduction

Standard reduction procedures were used on the gravity data. Data were adjusted
for their elevation (free air and Bouguer corrections) with respect to the base station and
their distance north or south of the base station (latitude correction) in accordance with the
formula (Parasnis, 1975).
 Δg = Δgobs + ΔgΦ + ( 3.086 − 4.191x10-4 ρ)h μms-2

where Δgobs is the observed gravity

ΔgΦ = 0.081 sin 2Φ μms-2 per 10 m distance north
Φ is the latitude of the base station
h is the elevation difference in metres
ρ is the density of the slab of material in kgm-3

Note, gravity values are given in micrometres per second squared (10 μms-2 = 1 mGal)

A terrain correction was not applied because topographic variation across the
survey area was only in the order of a few metres.
Bouguer correction densities between 2,100 kgm-3 and 3,000 kgm-3 were applied
in 100 kgm-3 steps to examine the effect on the gravity data. If the density chosen is too
high, the gravity surface moves in a direction opposite to topography and if too low the
gravity moves in accordance with the topography. Within the lagoon area, where there is
no topographic variation, the Bouguer correction density has no effect. The rim consists of
clays and ironstone pisoles as well as pebbles, cobbles, boulders and flows of basalt. A
suitable average density will be somewhat higher than that of 1950-2050 kgm-3 for normal
sediments (Parasnis, 1975) and lower than that of 2700-3300 kgm-3 for basalt (Parasnis,
1975). A Bouguer correction density of 2300 kgm-3 was chosen as a compromise and
supported by the above analysis.
The data were plotted on 1:1,000 scale and both hand and computer contoured.
Three-dimensional perspectives were generated to assist interpretation.

A3.3 Comparison of the Data Sets

The contoured data sets in this study are shown in the following Figures;

Kings Plains Lagoon

Fig. A3-1 elevation, contour interval 0.5 metres
Fig. A3-2 gravity, contour interval 0.5 μms-2
Fig. A3-3 magnetics, contour interval 50 nanoteslas (nT)
Fig. A3-7 elevation, a 3-D perspective
Fig. A3-8 gravity, a 3-D perspective
Fig. A3-9 magnetics a 3-D perspective

Dunvegan Lagoon
Fig. A3-4 elevation, contour interval 0.5 metres
Fig. A3-5 gravity, contour interval 0.5 μms-2
Fig. A3-6 magnetics, contour interval 200 nT
Fig. A3-7 elevation, a 3-D perspective
Fig. A3-8 gravity, a 3-D perspective
Fig. A3-9 magnetics a 3-D perspective

The advantages of a comparison of data sets over two similar structures are that
subtle, almost unnoticeable, features on one may be better visible on the other. Figures
A3-1, A3-4 & A3-7 show that Kings Plains and Dunvegan Lagoons are remarkably similar
in appearance. Dunvegan Lagoon is approximately twice the area of Kings Plains and
elongated in a NNW/SSE direction. Both gravity data sets (Fig. A3-8), show a gravity high
over the lagoon area nested in a broader regional gravity low, particularly well
demonstrated by the Dunvegan data . The anomaly associated with Dunvegan Lagoon is
approximately twice as large in amplitude, that is 8 μms-2 as compared with the 4 μms-2 at
Kings Plains (Figs. A3-2 & A3-5). The gravity high appears to be part of a throughgoing
gravity ridge, particularly prominent on the Dunvegan data. At Dunvegan, the gravity
ridge trends about N20°W, which corresponds to the long axis of the topographic feature
and is in direct line with Clarevaulx Lagoon, 2 kilometres to the NNW.
A prominent gravity low or "dimple" in the central and northern portion of the
Kings Plains Lagoon area is superimposed on the broad gravity high (Fig. A3-2 & A3-8)
and appears to reflect the presence of the lower density lacustrine clays in the lagoon, as
will be discussed later. A "dimple" is also present in the Dunvegan gravity data towards
the northern portion of the lagoon area, but is less obvious.
The magnetics data (Figs. A3-3 & A3-6), although more variable and erratic owing
to near surface noise, appear to correspond to the gravity features. At Dunvegan a
magnetic high (Fig. A3-6) supports the linear trends observed in the gravity data. The
magnetic anomalies at Dunvegan are some 3 to 4 times as large as those at Kings Plains.
The Kings Plains magnetics have a large amount of near surface interference, probably
due to the ironstone. The gravity and magnetic highs immediately to the south of Kings
Plains Lagoon (Figs. A3-2 & A3-3) reflect the presence of a surface basalt flow here.
The seismic refraction data indicate that the areas have between 1 to 2 m of
unconsolidated sediments, (velocity 250 to 500 ms-1). Below these soils, more
consolidated material, (velocity 750 to 1150 ms-1), extends to a depth of about 5 to 9 m,
which in turn sits on weathered basalt (velocity 2000 to 4200 ms-1). Although it is not a
good idea to generalize from only 7 refraction spreads, the layer 3 velocities seem to be
higher at Dunvegan Lagoon, 2800-4200 ms-1, as compared with 1900-3000 ms-1 at Kings
Plains. This suggests that the shallow basalts are less weathered at Dunvegan Lagoon than
at Kings Plains.
A3.4 Geophysical Modelling

Profiles at locations shown on Figures A3-2 and A3-5 were selected for modelling
using a Fortran program for generalized linear inversions of gravity and magnetic profiles
(Webring, 1985).
Three profiles were chosen at Kings Plains and four at Dunvegan. They were
selected to be at right angles to the observed geophysical strike.
A smooth regional gravity field, also cross-smoothed between profile lines, was
subtracted from each profile in order to remove wavelengths broader than the topographic
feature. An attempt was also made to model this regional field.

A3.4.1 Setting up the Model

A model consisting of 3 layers; weathered basalt, fresh basalt and Permian
basement was designed, based on the data from diamond drill holes KP1 and KP2 at Kings
Plains. Densities for each layer were determined from dry bulk density and apparent
porosity measurements of representative samples from the drill core (courtesy D. Emerson
of Sydney University). The data are included in Figure A3-10.
One of the aims of the geophysical surveying was to assess the existance of a
feeder pipe or pipes below the topographic feature, thus the gravity anomalies were
modelled as simple dykes penetrating the plane layered model. A dyke of uniform density
penetrating a stack of layers of different density may be modelled as a stack of blocks of
different density contrast as shown in Figure A3-11, part A. For example, a basalt dyke of
density 2810 kgm-3 penetrating soil of density 2200 kgm-3 is modelled as a block of 610
kgm-3 positive density contrast. Basalt penetrating basalt has a contrast of zero. Figure A3-
12 is an example of a final model that satisfies the observed field data for profile EF.
The drilling of holes KP3 and KP4 showed this simple model to be invalid. KP3
intersected basement 10 m shallower than expected, KP4 intersected basement 4 m deeper
and both failed to encounter any pipes or brecciated material.
A new, more complex model had to be developed for Kings Plains based on
variations in layer thicknesses and densities and using the data from all four drill holes.
The drill hole data is summarized in Figure A3-10. The depth of weathering of the basalt
and the depth to the Permian basement were allowed to vary. An extra surface layer
representing the lower density lacustrine claypan and surrounding sediments was
introduced to partially account for the dimple in the top of the gravity high. The model is
shown in Figure A3-11 part B. For simplicity of modelling, the density of the weathered
basalt was set equal to that of the Permian silicic volcanics, thus enabling the fresh basalt
to be modelled as a single body of positive density contrast to its surroundings. The
lacustrine clay in the lagoon was modelled as a body of negative density contrast with
respect to the sediments around it.. The densities of the different materials were held
constant from profile to profile within each survey area during the modelling exercise.
A3.4.2 The Modelling Results
Figures A3-13, A3-14 and A3-15 show final models that satisfy the observed
gravity data for the Kings Plains profiles AB, CD and EF and support the hypothesis that a
depression in the basement, filled with fresh basalt, is responsible for the broad gravity
high. The gravity high is flat-topped, as can be seen in profile EF, owing to a deepening in
the level of weathering of the top of the basalt coincident with the axis of the basement
channel. The dimple in the top of the gravity high is due to the slightly lower density
lacustrine clays filling the lagoon.
The gravity high to the south of Kings Plains Lagoon is coincident with fresh
basalt observed on the surface. Profiles AB and CD are affected by this basalt flow and
may be modelled with a similar configuration as used in EF, except with a thin surface
sheet of fresh basalt creating the extra gravity high.
A similar model was used for the Dunvegan gravity data. Figures A3-16, A3-17,
A3-18 & A3-19 show models which fit the observed data for Dunvegan profiles AB, CD,
EF and MN. The larger positive gravity anomaly at Dunvegan Lagoon requires, firstly,
that the fresh basalts have a slightly higher density contrast (440 kgm-3 as opposed to 340
kgm-3), secondly, a deeper Permian basement beneath the topographic structure (60 m as
opposed to 46 m) and thirdly, a shallower depth of weathering. The less pronounced
dimple on the top of the gravity high requires that the lacustrine sediments are not as thick
here.
An attempt was made to explain the longer wavelength regional gravity field which
was removed from the data. Kings Plains profile CD was used for this purpose and the
model shown in Figure A3-20. The regional field is satisfied by decreasing the basement
density by 10 kgm-3 in a series of steps, to a minimum under the deepest point. A
geological explanation for this is based on the observation that Permian basement channels
are localized or controlled by planes of structural weakness. Fracturing and deeper
weathering would be expected along these zones which would lower the bulk rock density
of the basement.

A3.5 Conclusion

a/ The presence of feeder pipes below the topographic structures is not required by
the geophysical data and furthermore, such pipes have not been found by the drilling
programme at Kings Plains designed to encounter them. In fact, drilling at Kings Plains
did not encounter any material that could be described as brecciated (Coenraads, 1988a).

b/ The geophysical anomalies do not require any unusual aberrations in the

channel floor morphology. This is illustrated in the block diagrams ( Figs. A3-21 & A3-
22) showing sections of the basalt filled channels modelled across the geophysical survey
areas. Differences between the gravity profiles were accommodated by minor changes in
the depth of weathering of the basalts, width of pre basaltic channel and presence or
absence of claypan sediments or surface basalt sheets as determined by surface mapping.

c/ The association of these features with older drainage systems that are basalt
filled has been clearly demonstrated and the existence of fresh basalts on a portion of the
rim is probably significant. Figure A3-23 highlights this relationship, the steeper portions
of the rim are edged against basalt shown dashed and the clayey pisolitic portion is shown
stippled.

d/ The fact that sapphires and spinels are found at Dunvegan Lagoon may be
coincidental because they were not seen at any of the other lagoons visited.
 Appendix 4

DRILLING PROGRAMME; KINGS PLAINS & DUNVEGAN LAGOONS,

AND KINGS PLAINS VALLEY

A4.1 Drillhole Locations

This appendix contains detailed descriptions of the exploration drilling and testing
carried out at Kings Plains and Dunvegan Lagoons, and in the Kings Plains Valley by T.J.
& P.V. Nunan Pty. Ltd and Jingellic Minerals N.L.

Exploration 1:25,000 Map Sheet Grid Reference Elevation Total

Hole Depth
(metres) (metres)

KP3 Sapphire 9138-1-S 48482058 921 35.0

KP4 Sapphire 9138-1-S 48392074 921 35.0
TEST PIT#1 Sapphire 9138-1-S 48482090 923 1.5
Dunvegan 1 Glen Innes 9138-4-S 72772134 1007 7.6
Dunvegan 2 Glen Innes 9138-4-S 72762128 1007 7.9
Dunvegan 3 Glen Innes 9138-4-S 72752122 1007 3.7
Dunvegan 4 Glen Innes 9138-4-S 72782140 1007.5 1.8
Dunvegan 5 Glen Innes 9138-4-S 72752116 1007 3.0
Dunvegan 6 Glen Innes 9138-4-S 72742111 1007 2.1
Dunvegan 7 Glen Innes 9138-4-S 72732104 1007 2.4
Dunvegan 8 Glen Innes 9138-4-S 72722096 1010 2.4
Dunvegan 9 Glen Innes 9138-4-S 72712090 1011 3.7
Dunvegan 10 Glen Innes 9138-4-S 72702084 1010
4.1
Dunvegan 11 Glen Innes 9138-4-S 72692077 1009.5
2.1
Dunvegan 12 Glen Innes 9138-4-S 72812163 1006
2.1
Dunvegan 13 Glen Innes 9138-4-S 72822169 1005
3.0
Dunvegan 14 Glen Innes 9138-4-S 72832175 1005
2.7
Dunvegan 15 Glen Innes 9138-4-S 72842180 1005
3.0
NA1 Sapphire 9138-1-S 49791603 935 18.0
NA2 Sapphire 9138-1-S 49851601 935 27.0
NA3 Sapphire 9138-1-S 49921600 935 21.0
NA4 Sapphire 9138-1-S 49981599 935 10.0
NA5 Sapphire 9138-1-S 51291581 935 27.0
NB1 Sapphire 9138-1-S 51191581 935 36.0
NB2 Sapphire 9138-1-S 51141598 935 36.6
NB3 Sapphire 9138-1-S 49921645 931 14.3
NB4 Sapphire 9138-1-S 48501493 938 18.9
NB5 Sapphire 9138-1-S 50121639 931 7.3
A4.2 Kings Plains Lagoon Drilling Programme

Five inch holes were drilled using an Ingersoll Rand T4 Drillmaster high
percussion drill. The material , which varied in size from fine powder to rock chips up to 3
cm, was collected each metre. Sieves were used to separate the coarse material for
geological description and the fine material was panned to obtain the heavy mineral
concentrate.

A4.2.1 Results

KP3 encountered dark lacustrine clay and lighter clay overlying, and probably
derived from, weathered basalt at about 8 m. The basalt becomes increasingly fresh in the
interval from 15 to 21 m. and in the interval from 19 to 20 m. the muddy slurry from the
drill hole changed from a yellow-tan to a blue-grey colour, indicating that the material
becomes predominantly fresh at this depth. Above this depth, an extremely fine grained,
black, magnetic concentrate can be panned from the material and probably represents
magnetite/ilmenite remaining after the decomposition of the basalt. In the interval 30-31m
some coaly fragments were encountered indicating a clear break between two lava flows.
This break can be correlated to similar breaks found in KP1 and KP4. The upper flow is
more altered in places, the basalt is vesicular with open, partially filled and filled vesicles,
whereas in the flow below the woody horizon the basalt is compact, non vesicular and
fresh. Weathered Permian basement, consisting of crystalline quartz in a white clay matrix,
was encountered under fresh basalt at 34 metres.
KP4 encountered similar clay to KP3 overlying weathered basalt at 8 metres which
becomes increasingly fresh in the interval from 19-23m. The woody horizon between the
two distinctive basalt flows was found at 28 metres. Weathered Permian basement was
encountered under fresh basalt at 46.5 metres.
No traces of sapphire were found in either hole.
Test Pit # 1 & 2 were dug into the rim of Kings Plains Lagoon to a depth of 1.5 m.
The material was ironstone pisoles 1 - 2 mm in red clay becoming increasingly cemented
with depth. Eleven cubic yard of material were removed and tested. Only trace amounts of
sapphire and zircon were found in pit #1 (0.2 g/cu.yd) and none in pit #2.
The drilling and outcrop data suggest the presence of a pre basaltic valley some 2.5
kilometres wide and 60 metres deep along which successive basaltic lavas have flowed
forming a broad flat plain. Kings Plains Creek has been displaced from its original position
to a more northerly position on the edge of the basalt- basement contact.
KP3 and KP4 were completed as water bores flowing 1000 an 2000 gallons per
hour respectively.
Detailed logs for KP3 and KP4 are included and the information summarized on
cross section, Fig. A3-10.
 Drill Hole KP3

Depth Interval Description

(metres)
0 -1 Fine olive- grey lacustrine clay
1-2 Clay material becoming a little coarser and lighter
2-3 Yellow- brown clay
3-4 Yellow- brown clay
4-5 Yellow- brown clay
5-6 Yellow- brown clay with some decomposed rock frags
6-7 Yellow- brown wet clay with some larger tan coloured rock fragments
(decomposed basalt)
7-8 Yellow- brown wet clay with some larger tan coloured rock fragments (basalt ?)
8-9 Yellow- brown wet clay with tan coloured altered basalt. Fine black magnetic
concentrate panned from this material.
9 - 10 As above
10 - 11 Clay as above with larger weathered fragments, easily broken, black oxide along
fractures in rock.
11 - 12 Mostly clay with completely altered buff to red- brown basalt fragments, easily
broken.
12 - 13 Very light grey to buff amygdaloidal basalt (texture obvious with altered
plagioclase laths visible), altered, easily broken. Rare blue- grey basalt fragments
more difficult to break.
13 - 14 Altered basalt as above
14 - 15 Altered basalt as above
15 - 16 Altered basalt as above with green, soapy, spherical grains derived from vesicles in
the basalt. Approx 10% blue- grey basalt fragments.
16 - 17 Altered basalt as above with green, soapy, spherical grains derived from vesicles in
the basalt. Approx 10% blue- grey basalt fragments.
17 - 18 As per 12 - 13 but with approx 10% blue- grey basalt.
18 - 19 Altered basalt as above but with blue- grey basalt with zeolite,calcite? filled
vesicles becoming more abundant (80%), moderately altered, more difficult to
break
19 - 20 Mud changes colour at this depth from an oxidized red- brown colour to a blue
colour. Blue- grey vesicular basalt with white (zeolite, calcite) cavities, mod.
altered (80%)
20 - 21 Blue- grey vesicular basalt with white (zeolite, calcite) cavities, mod. altered
(100%).
21 - 22 Basalt as above
22 - 23 Basalt as above
23 - 24 Light grey basalt, hard but altered, veins & vesicles (up to 4mm) filled with green
black soapy material which in some cases shows signs of alteration to a white
material.
24 - 25 Basalt as above but with vesicles up to 12mm.
25 - 26 Basalt as above but vesicles to 5mm, some empty, some lined with botryoidal blue
zeolite.
26 - 27 Basalt as per 25 - 26.
27 - 28 Basalt as above but with blue zeolite lined vesicles to15mm.
28 - 29 Basalt as per 25 - 26.
29 - 30 Basalt as per 25 - 26 but with abundant small (< 1mm) vesicles filled with white
needlelike zeolite.
30 - 31 Appearance of coaly fragments and first appearance of a fresh, hard, dark grey,
non- vesicular basalt, apparently another flow unit.
31 - 32 Fresh basalt as above.
32 - 33 Basalt as above, only moderately altered, softer, cut by fine veinlets and containing
some vesicles filled with white material, probably due to flowing over basement.
33 - 34 Weathered basement (Permian acid volcanics), crystalline quartz (approx 1-3mm
often bipyramidal) in a white clay matrix.
34 - 35 Weathered basement.
 Drill Hole KP4
Depth Interval Description
(metres)
0-1 Olive grey lacustrine clay with vegetation.
1-2 Olive grey clay.
2-3 Oive grey clay.
3-4 Medium grey clay.
4-5 Med. grey clay with some altered tan rock (basalt?) frags.
5 -6 Med. grey clay with some altered tan rock (basalt?) frags.
6-7 Yellow- brown clay becoming wetter (water table).
7-8 Completely altered greenish- grey basalt with red- brown to tan vesicles in tan clay
(similar to KP3 12-13)
8-9 Altered basalt and clay as above.
9 - 10 Greenish- light grey completely altered basalt, soft.
10 - 11 Altered basalt as above.
11 - 12 Altered basalt as above.
12 - 13 Altered basalt as above.
13 - 14 Altered basalt as above.
14 - 15 Altered basalt as above with vesicles up to 4mm partially filled with green and tan
secondary minerals.
15 - 16 Altered basalt as per 14 - 15
16 - 17 Altered basalt as per 14 - 15
17 - 18 Altered basalt as per 14 - 15
18 - 19 Altered basalt as per 14 - 15 with the appearance of some harder, non- vesicular,
blue- grey, altered basalt chips (<5%). Corresponds to KP3 15-16m
19 - 20 Altered basalt as per 18 - 19
20 - 21 As per 18 - 19 but with approx 10% blue- grey basalt.
21 - 22 As per 18 - 19 but with approx 80% blue- grey basalt.
22 - 23 Harder blue- grey, altered basalt (100%) with vesicles, some containing soft, fine,
greenish- black secondary mineral.
23 - 24 Basalt as per 22-23, vesicles increase in size to 6mm.
24 - 25 Basalt as above, some large vesicles coated with transparent secondary mineral.
25 - 26 Basalt becoming fresher, dark grey, cut by veins and fractures filled with white
mineral, vesicles as above.
27 - 28 Basalt as above, evidence of vesicles or veins to 20mm partially filled with yellow-
green translucent botryoidal chalcedony.
28 - 29 Coaly wood fragments at 28m. Fresh, fine grained, dark grey basalt, fresh
pyroxene and plagioclase phenocrysts visible. Some white veinlets and vesicles
filled/ partially filled with transparent secondary mineral. Apparently another flow
unit below woody layer.
29 - 30 Basalt as above.
30 - 31 Basalt as above, minor vesicles.
31 - 32 Fresh, fine grained, dark grey basalt, fresh pyroxene and plagioclase phenocrysts
visible, no vesicles.
32 - 33 Basalt as above.
33 - 34 Basalt as above.
34 - 35 Basalt as above.
35 - 36 Basalt as above.
36 - 37 Basalt as above but with some olivine phenocrysts and plagioclase rich patches
with a stockwork appearance.
37 - 38 Basalt as above.
39 - 40 Basalt as for 36-37.
40 - 41 Basalt as above.
41 - 42 Basalt as above but with some green soapy secondary mineral liberated fronm
fractures.
42 - 43 Basalt as above.
43 - 44 Basalt as above.
44 - 45 Basalt as above, with some small soft red crystals (probably altered olivine).
45 - 46 Basalt as above, with some small soft red crystals and green soapy veins.
46 - 47 Basalt as per 45-46, some vesicles partially filled with botryoidal blue zeolite. Drill
break at 46.5m.
47 - 48 Weathered basement (Permian acid volcanics), crystalline quartz, approx 1-3mm
often bipyramidal, in a white clay matrix.
48 - 49 Weathered basement.
A4.3 Dunvegan Lagoon Drilling and Testing Programme

A line of fifteen holes were drilled accross Dunvegan Lagoon using a Cauldweld
large diameter bucket drill. Bulk samples were processsed and grades are given in grams
per loose cubic yard and are for minerals in the size range 1/16" - 7/8".

A4.2.1 Results

Depth Interval Description Grade Grade

(metres) Sapphire Spinel
g/cu.yd g/cu.yd
Dunvegan 1
0.0 - 2.0 m stiff grey clay (lagoon sediments)
2.0 - 2.4 m transition from clay
to extremely weathered basalt
2.4 - 4.5 m weathered basalt
4.5 - 7.6 m weathered basalt and basalt boulders
becoming quite hard at 7.6 m

Dunvegan 2
0.0 - 1.5 m stiff grey clay
1.5 - 7.9 m soft weathered basalt & basalt boulders
becoming harder with depth

Dunvegan 3
0.0 - 1.5 m stiff grey clay 0.8 42.8
1.5 - 1.8 m wash with basalt pebbles 1.0 42.3
1.8 - 3.7 m weathered basalt

Dunvegan 4
0.0 - 0.5 m ironstone pisoles in red clay 1.5 36.2
0.5 - 1.2 m orange wash 8.2 55.6
1.3 - 1.8 m grey wash with large basalt boulders 0.6 20.9

Dunvegan 5
0.0 - 1.5 m stiff grey clay - 17.0
1.5 - 1.8 m grey wash 0.5 4.8
1.8 - 3.7 m weathered basalt

Dunvegan 6
0.0 - 1.1 m stiff grey clay
1.1 - 1.6 m clay with basalt pebbles, possibly wash 4.5 61.3
1.6 - 2.1 m weathered basalt with basalt cobbles

Dunvegan 7
0.0 - 1.2 m stiff grey clay
1.2 - 1.6 m brown grey weathered basalt 2.5 13.5
1.6 - 2.4 m brown grey weathered basalt

Dunvegan 8
0 - 1.2 m basalt topsoil
1.2 - 2.4 m brown grey weathered basalt
Depth Interval Description Grade Grade
(metres) Sapphire Spinel
g/cu.yd g/cu.yd
Dunvegan 9
0.0 - 0.6 m iron stone pisoles in red clay 0.6 24.6
0.6 - 3.5 m weathered basalt
3.5 - 3.7 m weathered basalt with basalt boulders - 24.0

Dunvegan 10
0.0 - 0.3 m topsoil 3.3 31.8
0.3 - 0.9 m ironstone 0.4 19.8
0.9 - 1.7 m grey clay
1.7 - 2.0 m transition grey clay to weathered basalt 0.5 10.3
2.0 - 4.1 m weathered basalt becoming hard at 4m

Dunvegan 11
0.0 - 0.3 m topsoil
0.3 - 0.6 m ironstone 2 11.3
0.6 - 1.5 m orange grey 0.3 5.0
1.5 - 1.8 m orange clay & weathered basalt - 6.8
1.8 - 2.0 m weathered basalt - 6.8
2.0 - 2.1 m weathered basalt becoming harder 2.5 47.0

Dunvegan 12
0.0 - 1.8 m topsoil
1.8 - 2.1 m clayey orange brown alluvium with - 32.3
2.1 - 2.4 m weathered basalt

Dunvegan 13
0.0 - 2.4 m topsoil
2.4 - 2.7 m clayey orange brown alluvium 7.25 -

Dunvegan 14
0.0 - 2.1 m topsoil
2.1 - 2.4 m clayey orange brown alluvium with
some larger basalt pebbles 10.0 132.3
2.4 - 2.7 m weathered basalt

Dunvegan 15
0.0 - 2.4 m topsoil grading into clay
2.4 - 2.7 m clayey orange brown alluvium 2.0 57.8
2.7 - 3.0 m weathered basalt
A4.4 Kings Plains Valley Deep Drilling Program

The Kings Plains Valley Deep Drilling Program was designed to test for the
possibility of intra-basaltic and sub-basaltic sapphire deposits. The concept of the program
was developed as a result of the Braemar sapphire deposit being recognized as intra-
basaltic, sandwiched between older basalt flows and capped by younger flows; coupled
with the observation of weathered basalts covering portions of some of the alluvial
deposits mined on Kings Plains (T. Nunan pers. comm. 1988)
An initial series of five, one metre diameter holes (the Nunan A series) were
drilled; NA1 on the eastern feeder, and NA2, NA3, NA4 and NA5 on the western feeder,
of Kings Plains Creek (see Figure A4-1). The locations of the exploration holes were
based on the palaeo-topographic analysis and they were designed to pinpoint the exact
position of the palaeo channel axes and to test the nature of any material existing in them.
At the same time, the holes would test for the presence of intra-basaltic deep leads.
Sampling was done at 2 metre intervals, commencing from the top of the basalt. A
half cubic yard sample was washed from each 2 m section and the concentrate examined.
Each hole is described in detail in the results section and a cross section is shown in Figure
A4-2.
Hole NA1 encountered a layer of red clay, between 9 and 12 metres below the
surface, containing bipyramidal quartz sandwiched between two weathered basalt flows. A
horizon of similar description was encountered between 33 and 35 m, directly overlying
silicic volcanic basement.
It is unclear if these horizons represent weathered red volcaniclastic rocks with
basement contamination, such as observed at Braemar, or whether they have been
reworked by fluvial processes.
Trace amounts of sapphire were recovered from these horizons. However, as this
first series of holes were uncased, there remains some doubt as to whether the sapphires
were derived from the horizons of interest or are contamination from the overlying
sapphire bearing alluvium.
A second series of 5 cased holes (the Nunan B series) were drilled to repeat the
experiment under more rigorous conditions (see Figure A4-1). Hole NB1 drilled adjacent
to NA1 experienced recovery problems in the zone of interest and the other holes (NB2 to
NB5) essentially encountered basalt overlying weathered basement. The intra-basaltic
material was not seen in any of these holes although it may have been too thin to be
noticed.
No further sapphires were recovered during the second phase of exploration
drilling.
A4.4.1 Results

The deep drilling program consisting of 10 holes to date supports the presence of at
least one time break between the lava flows that filled the Kings Plains palaeo-valley.
Intra-basaltic and possibly sub-basaltic material was encountered only in hole NA1 and
this material at best contained trace amounts of sapphire.
Although rich, sub-basaltic and intra-basaltic deep leads have not been found
during this program, the possibility of their presence cannot be discounted. If present
however, they will be thin and extremey difficult to find

Grades are given in grams per loose cubic yard and are for minerals in the size range 1/16"
- 7/8"

Depth Interval Description Grade Grade

(metres) Sapphire Spinel
g/cu.yd g/cu.yd

NA1
0 - 4m brown clayey soil (alluvial sediments) - -
4 - 18 m grey weathered basalt overlying - -
weathered Permian basement

NA2
0 - 4m brown clayey soil (alluvial sediments) - -
4 - 27 m soft grey weathered basalt overlying - -
weathered Permian silicic volcanics

NA3
0 - 4m brown clayey soil (alluvial sediments) - -
4 - 21 m soft grey weathered basalt overlying - -
weathered Permian silicic volcanics

NA4
0 - 4m brown clayey soil (alluvial sediments) - -
4 - 10 m soft grey weathered basalt overlying - -
weathered Permian silicic volcanics

NA5
0 - 4m brown clayey soil (alluvial sediments) - -
4 - 5m transition from soil to extremely - -
weathered basalt
5 - 9m weathered basalt - -
9 - 12 m red clay with abundant bipyramidal
quartz up to several cm in size trace trace
12 - 25 m weathered basalt - -
25 - 27 m red clay with bipyramidal quartz, this
may represent weathered Permian silicic
volcanics - -
27 - 30 m Permian silicic volcanics
Depth Interval Description Grade Grade
(metres) Sapphire Spinel
g/cu.yd g/cu.yd
NB1
0.0 - 4.0 m dark brown soil (alluvium) - -
4.0 - 7.0 m transition fromsoil to extremely
weathered basalt
7.0 - 21.0 m weathered basalt no recovery
21.0 - 23.0 m greenish grey to grey weathered basalt,
some vesicular with vesicules filled with
olive green soapy material.- some black
heavy mineral, either spinel
or ilmenite and some water clear grains
showing distinct cleavages and striations
on flat faces, probably feldspar, generally
2-3mm although frag up to 10mm found - trace
23.0 - 25.0 m as for 21.0 - 23.0
25.0 - 27.0 m weathered basalt, fine grained, some
vesicular as above. Some cream
to tan concretions and fragments of
concretions derived from the basalt
spinel and clear quartz
fragments approx 1mm. - trace
27.0 - 29.0 m weathered basalt, ironstone concretionary
fragments becoming more abundant, some
spinel & occasional translucent green
striated pyroxene, occasional clear quartz. - trace
29.0 - 31.0 m ironstone fragments & concretions
abundant, some heavy minerals - trace
31.0 - 33.0 m weathered Permian silicic volcanics,
abundant bipyramidal clear to smokey
quartz in red clay, some basalt fragments
& ironstone concretions. - -
33.0 - 35.0 m weathered Permian silicic volcanics,
bipyramidal quartz, approx 2mm,
grains often showing fresh crystal faces
and edges. - -

NB2
0.0 - 3.7 m dark brown topsoil (alluvium)
3.7 - 34.1 m grey weathered basalt - -
34.1 - 36.6 m clay with abundant bipyramidal
quartz (weathered Permian acid volcanics)

NB3
0.0 - 5.2 m dark brown topsoil (alluvium) - -
5.2 - 14.3 m grey weathered basalt becoming harder
with depth. Hole bottomed at 14.3m as
rock became too hard to proceed - -

NB4
0.0 - 4.3 m dark brown topsoil (alluvium) - -
4.3 - 7.9 m grey weathered basalt
7.9 - 18.9 m clay with abundant bipyramidal
quartz (weathered Permian acid volcanics)

NB5
0.0 - 4.6 m dark brown topsoil (alluvium) - -
4.6 - 7.3 m grey weathered basalt, bore caved in below
steel casing at 7.3 m.
 Appendix 5

MOUNT BUCKLEY BASALT FLOWS: THIN SECTION DESCRIPTIONS

Thin sections made from the thirty one basalt samples collected from Mt Buckley
are described in this appendix. The locations of the sample sites are shown in Fig. 6-19
and grid references provided in Table 6-1.

B1
Very fine grained holocrystalline rock, porphyritic, showing evidence of flow banding.
Subhedral to anhedral equant phenocrysts or megacrysts of olivine give the rock a
porphyritic texture in a groundmass of euhedral plagioclase laths, stubby euhedral augite
prisms and interstitial nepheline.

B2
Same as B1.

B3
Very fine grained rock with euhedral altered olivine and ragged anhedral nepheline
crystals in a groundmass of euhedral plagioclase laths, opaque minerals and glass.

B4
Medium to fine euhedral to subhedral olivine plenocrysts in groundmass of euhedral to
subhedral elongate plagioclase laths, equant euhedral to subhedral opaques and augite.
Some alteration of the olivine is evident.

B5
Same as B4 with some interstitial glass.

B6
Very fine grained hypohaline rock with a porphyritic texture similar to B3. Subhedral to
euhedral olivine, ragged low relief inclusions consisting of equigranular anhedral grains in
a groundmass of smaller plagioclase laths and brown glass.

B7
Euhedral olivine phenocrysts or megacrysts in a grounsmass of short euhedral plagioclase
laths and brown glass.

B8
Similar to B6.

B9
Medium-grained holocrystalline rock. Glomoporphyritic texture with medium-sized,
zoned, single euhedral crystals or clots of subhedral crystals of pink augite and medium
euhedral plagioclase laths. Groundmass comprises of equant subhedral to euhedral finer
plagioclase, olivine, pyroxene and opaques with minor brown glass.

B10
Phenocrysts or megacrysts of equigranular euhedral olivine. Description as per B7 except
groundmass is mostly granular with only minimal glass.

B11
Euhhedral equant and elongate olivine (with some minor alteration along cracks and
around the rims), euhedral elongate plagioclase and some pink augite agregates in a fine
holocrystaline groundmass of plagioclase laths and euhedral augite and opaques.

B12
Suchedral equant olivine in a fine holocrystaline groundmass of euhedral subparallel
plagioclase laths which give the rock a trachytic texture, euhedral equant augite and
opaques.

B13
Medium to fine subhedral equant to elongate olivine phenocrysts in a holocrystaline
groundmass of finer subhedral to anhedral olivine, plagioclastic laths, very fine equant
euhedral laugite prisms and opaques.

B14
Medium sized subhedral to euhedral equant to elongate plenocrysts of olivine and zoned
pink augites with plagioclase displaying subophitic texture in a fine to medium grained
grounsmass of elongate euhedral plagioclase equant subhedral olivine, augite and opaques.
Rock is essentially equigranular holocrystalline.

B15
Phenocrysts of subhedral to euhedral olivine, pink augites displaying a subophitic
relationship with the included plagioclase, equant euhedral opaque minerals. The rock is
essentially holocrystalline with interstitial areas filled with a equant anhedral low relief,
low birefringince zeolite. The presence of approximately 20% modal zeolite renders B15
unsuitable for normative mineralogy comparisons.

B16
Medium to coarse grained glomoporpheritic aggregates of subhedral pink augite an
medium to coarse grained elongate subhedral to euhedral plagioglase which are crudely
alligned imparting a trachytic texture to the rock. Some altered clots of euhedral crystals
probably were olivine phenocrysts or megacrysts. The groundmass consists of fine
euhedral augite and isotopic brown glass with some euhedral, equant and accicular opaque
minerals. Zoning of augite from light centres to pink rims indicates titanium enrichment.

B17
Same as B16 but containes some ragged edged euhedral zeolite with strange curved, less
distinct zoning.

B18
Same as B16, but with more aggregates of ophitic pink augite.

B19
Same as B18.

B20
Fine-medium grainsize, euhedral single augite (zoned with a darker pink rim) and olivine
crystals in a finer granular groundmass comprising euhedral augite, subhedral opaques and
feldspar laths with some minor anhedral granular aggregrates of ?alkali feldspars.
B21
Equant to elongate subhedral olivine megacrysts in a holocrystaline groundmass of
plagioclase, pink augite and opaques.

B22
This sample appears to be quite altered, phenocrysts of olivine have all altered to orange
iddingsite?, and carbonate veins are present. The groundmass consists of subhedral
plagioclase laths, opaque and glass. The extensive alteration renders B22 unsuitable for
normative mineralogy comparisons.

B23
Subhedral equant to elongate olivine phenocrysts, otherwise like B22.

B24
Euhedral, equant olivine phenocrysts in a granular groundmass of finer olivine
(considerably altered red-brown around edges of grains and within fractures, euhedral
plagioclase laths, small euhedral brownish augite and opaques. The extensive olivine
alteration renders B24 unsuitable for normative mineralogy comparisons.

B25
Same as B24 but with much less alteration of the groundmass olivine.

B26
Abundant medium to fine euhedral olivine phenocrysts in a trachytic textured granular
groundmass of fine elongate euhedral plagioclase laths, fine equat augite and opaques.

B27
Similar to B26 but with fewer olivine phenocrysts and fewer opaques in the groundmass.
The phenocrystal olivine shows some alteration ant there is approximately 20% zeolite
thus rendering B27 unsuitable for normative mineralogy comparisons.

B28
Euhedral elongate to equant olivine phenocrysts in a groundmass of plagioclase, augite
and dark brown isotopic glass. Many zeolite-filled vesicules are also present thus
rendering B27 unsuitable for normative mineralogy comparisons.

B30
Similar to B28.

B31
Similar to B28 but groundmass is almost entirely glass with some plagioclase laths.
 Appendix 6

MATERIAL PUBLISHED DURING THE PERIOD OF DOCTOR OF

PHILOSOPHY CANDIDATURE

Coenraads, R.R. (1988), Structural control and timing of volcanism in the central province.
Implications for regional targeting of prospective areas for sapphire and diamond
exploration; New England Tectonics and Metallogenesis, Kleeman J.D. (ed). Proceedings
of a symposium held at the University Of New England, Armidale New South Wales 14-
18 November 1988, 302-307.

Coenraads, R.R. (1989), Evaluation of the natural lagoons of the Central Province, N.S.W. -
Are they sapphire producing maars? Bulletin of the Australian Society of Exploration
Geophysicists, 20, 347-363.

Pecover, S.R. & Coenraads, R.R. (1989). Tertiary volcanism, alluvial processes, and the
origin of sapphire deposits at "Braemar" near Elsmore, northeastern New South Wales.
New South Wales Geological Survey - Quarterly Notes 77, 1-23.

Coenraads, R.R., Sutherland, F.L. and Kinny, P.D. (1990), The Origin of Sapphires: U-Pb
dating of zircon inclusions sheds new light. Mineralogical Magazine , 54, 113-122.

Coenraads, R.R. (1990). Key areas for alluvial diamond and sapphire exploration in the New
England gem fields, NSW, Australia. Economic Geology, 85, 1186-1207.

Coenraads, R.R. & van der Graaf, R. (1991). An occurence of gem garnets from Horse Gully
in the New England gem fields, NSW. Australian Gemmologist, 17(10) 412-415.

Coenraads, R.R. (1990). Palaeogeography of the Braemar sapphire deposit: Implications for
deep-lead sapphire exploration in the Central Volcanic Province, New South Wales.
Royal Society of New South Wales, 123, 75-84

Bischoff, G.C.O., Coenraads, R.R. and Lusk, J. (in press). Microbial gold; an example from
Venezuela. N.Jb.Geol.Paläont.Abh.,

Bischoff, G.C.O., Coenraads, R.R. and Lusk, J. (in press). Microbial gold; an example from
Venezuela. N.Jb.Geol.Paläont.Abh.,

Coenraads, R.R., Paige, S.C.B. and Sutherland, F.L. (1991). Ilmenite mantled rutile crystals
from the Uralla District, New South Wales. Royal Society of New South Wales, 124, 23-
34.

Coenraads, R.R. (under review). Evaluation of potential sapphire source rocks within the
catchments of Kings Plains Creek and Swan Brook, near Inverell, New South Wales.
Records of the Australian Museum,

Coenraads, R.R. (under review). Depositos Diamantíferos del Río Guaniamo del Estado
Bolívar, Venezuela. Boletín de la Sociedad de Geologos Venezolanos,
Coenraads, R.R. (under review). Surface Features of Natural Rubies and Sapphires
associated with Volcanic Provinces. Journal of Gemmology,

Bischoff, G.C.O. and Coenraads, R.R. (in prep). Fossil and recent traces of biodegradation on
heavy minerals.

A6-1 Reports:

Coenraads, R.R. (1988). T.J.& P.V. Nunan Pty. Ltd. Six monthly report to the Department
of Mines, E.L.'s 2988 and 2989 for the period to 12 July 1988, (unpubl.).

Coenraads, R.R. (1988). Jingellic Minerals N.L. Final Report for the Kings Plains Drilling
Project, New England, New South Wales. (unpubl).

Coenraads, R.R. and Lawrence, D. (1989). T.J.& P.V. Nunan Pty. Ltd. Six monthly report to
the Department of Mines, E.L.'s 2988 and 2989 for the period to 12 January 1989,
(unpubl.).

Coenraads, R.R., Pope, G.J. and Whittle, M.A. (1989). T.J.& P.V. Nunan Pty. Ltd. Six
monthly report to the Department of Mines, E.L.'s 2988 and 2989 for the period 15
January, 1989 to 14 July, 1989. (unpubl.).

Coenraads, R.R., Pope, G.J. and Whittle, M.A. (1989). T.J.& P.V. Nunan Pty. Ltd. Six
monthly report to the Department of Mines, E.L.'s 2988 and 2989 for the period 15
July, 1989 to 14 January, 1990. (unpubl.).

A6-2 Symposium abstracts:

Coenraads, R.R. (1991). Sapphires and rubies associated with volcanic provinces:
Inclusions and surface features shed light on their origin. Gemmological
Association of Australia 45th Federal Conference and Seminar, Canberra, May 3-
5, 1991, p. 2-14.