aot GA gE CLE ELIE OO LEAN AANA AF @ os . 4 me 3 nti
peering gO Pre” f a 2 cat atc tee " a - . Ss pen Royster SIT DON
me ign GA alee pirsoSn co Pass eT wie bao siege entainn ne
wie . pen e one
Sn i OOS ET Te ae = ere : °
a te eer RSE een ng rag : a tartans patina
r 7 nag, og TT el Ne
Sl ie Be ee ae dd
a a ee
be 5 ee ee ee nae ne Se
ee dal . vs . . vas
died : “ — yee - oe en a or somes . soil “ i —Gz ee i ee te
3 ‘. — 2 Satie eI ee Ne ee a ost ae

AE eo eptie oet pate eee e Be

ee ee

he agree PALER NG ca GL AT OIA A BCE Secret es
—- perpen cee i PIR ie as SOA OI Sa See
Dg hen ng rege cohen e re
— Fe eo ae PAE me
Sabre eT
CTD ENF
ted s

Te tre Piety Tain hee AON antite —canrmee alt emo —_=
ee eed Ace ao gta tie eitgnin te

cad wee ape area OLE RE

tag gg rreiaeng ae EE i

eee me etn

alee
eaten eral a te ee ne RO ER ae ee ee

pp ae VOCS an on saiegen paises

Or eel on OE a

ee? eter

ee ee ee ee ie -
So ieee

a ec toaSeraae es Ae ak eri aera e tna, 2 faa
me ge gee ee melee at Neer ane yt - 7

S06,7¢4Y¥

JOURNAL AND PROCEEDINGS

OF THE

ROYAL SOCIETY

OF NEW SOUTH WALES

FOR

1952
(INCORPORATED 1881)

VOLUME LXXXVI

Parts I-IV

EDITED BY

IDA A. BROWNE, D.Sc.

Honorary Editorial Secretary

THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE
STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN

SYDNEY
PUBLISHED BY THE SOCIETY, SCIENCE HOUSE

GLOUCESTER AND ESSEX STREETS

Issued as a complete volume, April, 14 1953

yor ss i
f , Ie 7 i
\ 4
I 1
=X,
; 1)
|
re
i { i
5 4 ne ,
ri {
ant 7 I y
; i)
x *
. %
y e
.
\ °
k
&
i i
I ¥
. hoe
ii
J ee)
A ' i 7 if
; i i am
i ; os
.
‘
~ j
i
.
l
’ n im
+ f

CONTENTS
VOLUME LXXXxXVI

Part I*
TITLE PAGE
OFFICERS FOR 1952-1953
NOTICES
List OF MEMBERS
AWARDS
ANNUAL REPORT OF COUNCIL
BALANCE SHEET
REPORT OF SECTION OF GEOLOGY
OBITUARY

Art. I.—Presidential Address. By R. C. L. Bosworth—

General

Transport Problems in Applied Chemistry ..
Art. II.—A Geological Account of Heard Island. By A. J. Lambeth ..

Art. III.—Occultations Observed at Sydney Observatory during 1951. By W. H.
Robertson and K. P. Sims

Part Il+

Art. IV.—Climate and Maize Yields on the Atherton Tableland. By D. 8S. Simonett
and N. T. Drane

ArT. V.—Palladium Complexes. Part V. Reactions of Palladium Compounds with
2:2’ Dipyridyl. By S. E. Livingstone .

Art. VI.—Liversidge Lecture. Electron Diffraction in the Chemistry of the Solid
State. By A. L. G. Rees ..

Art. VII.—Permian Spirifers from Tasmania. By Ida A. Brown

* Published October 10, 1952.
+ Published January 21, 1953.

iil

1V

vi

vil

XV

22

38

55

i SNe Nai Dw |) Dac

CONTENTS

Part II

Art. VIII.—Induced Optical Activity of the Tris-1 : 10-Phenanthroline and Tris-2 : 2’-
Dipyridyl Copper II Ion. By N. R. Davies and F. P. Dwyer

Art. I[X.—Soil Horizons and Marine Bands in the Coastal Limestones of Western
Australia. By R. W. Fairbridge and C. Teichert

Part IVy

Art. X.—A Contribution to the Geology and Glaciology of the Snowy Mountains. By
A. 8. Ritchie .

Art. XI.—Graptolite Zones and Associated Stratigraphy at Four-Mile Creek, South-west
of Orange, N.S.W. By N. C. Stevens and G. H. Packham

Art. XII.—Martiniopsis Waagen from the Salt Range, India. By Ida A. Brown

Art. XIII.—Contributions to a Study of the Marulan Batholith. Part II. The Grano-
diorite-Quartz Porphyrite Hybrids. By G. D. Osborne and J. 8. Lovering

Art. XIV.—The Replacement of Crinoid Stems and Gastropods by Cassiterite at Emma-
ville, New South Wales. By L. J. Lawrence

* Published March 9, 1953.
+ Published April 14, 1953.

64

68

88

94

100

108

119

v

PART I

Ki

~ JOURNAL AND PROCEEDINGS

wy
Misc

ae | ane OF THE bon

ROYAL SOCIETY |
ate NEW SOUTH WALES | |

FOR
1952

eae ; _ (INCORPORATED 1881)

PART [I

| OF
| VOL. LXXXVI | oe |
Ek sining Report of Council, Balance Sheet, Presidential” Address ae
and Papers read in April and May, 1952. ee

~

EDITED BY

IDA A. BROWNE, D.Sc.

| ; : Honorary Editorial Secretary

a THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE
|| —__—s STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN

* 3 SYDNEY oy ;
PUBLISHED BY THE SOCIETY, SCIENCE HOUSE
GLOUCESTER AND ESSEX STREETS 3 5 0 4 2 a

———

1952

TrrLE Pace

; oct ror 1952-1953 RAS: a fopcatneee at 4 re ae.
“Nonna : ee en ta o's +s .s as Bh pie yee Fee Ae 3 |
“List oF Mamas,’ 1, eee ae ot eee “Pagans -

Aguva ais ite Beira hivenchis 1h ae oe Pg Meet A ee ee

_AnnuAL REpPorT oF CouNcIL .. e: ba Be Oe f, he ae tae

BALANCE SHEET at eal ee oe ee ee ee ee ave mae _ oe
_ Report or SecTION or GEOLOGY 9 Hs ae we , raat

.

_ Oprruary ee ee ae ee ee mais aes ete Snes ee ;

ak

an oS 4) Presidential Address. By R. ome Pe Bosworth—

thes Mott ie: General ee ‘ Bas Te pe ee Seer ee oe ee Tae Rs

De # . _ Transport Problems in Applied Chemistry Se, ees eee oa se

Ant. IL—A Geological Account of Heard Island. By A. J. Lambeth

Prt Be . * z NR

Arr. 11. PiGeewlenmon Observed at Sydney Observatory during 1951.
ee and K. P. Sims bles Geers aoe

JOURNAL AND PROCEEDINGS

OF THE

BROYAL SOCIEIY

OF NEW SOUTH WALES

FOR

1952

‘INCORPORATED 1881)

——<——— ee

VOLUME LXXXVI
Part I

—_—_—————

EDITED BY

IDA A. BROWNE, D.Sc.

Honorary Editorial Secretary

—

THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE
STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN

SYDNEY
PUBLISHED BY THE SOCIETY, SCIENCE HOUSE
GLOUCESTER AND ESSEX STREETS

Vi pili v9 ade
Nah ,

i rit | ving Na Ci
fey Ai
" \ Tekan ‘th (N as
Ae f ay He AOR AN
\ f HA eA i} i)
i i, ¢

{fants

Royal Sorivty of Nem South Wales

OFFICERS FOR 1952-1953

Patrons:
His EXCELLENCY THE GOVERNOR-GENERAL OF THE COMMONWEALTH OF AUSTRALIA
THe Rr. Hon. Srr Witwtiam J. McKELL, G.c.M.G., P.C., Q.C., LL.D.

His EXCELLENCY THE GOVERNOR OF NEw SoutTH WALES,
LIEUTENANT-GENERAL Str JOHN NORTHCOTT, k.c.M.G., ©.B., M.V.0., D.Litt.

President :
C. J. MAGEE, D.sc.agr. (Syd.), M.sc. (W2s.).

Vice-Presidents :
R. C. L. BOSWORTH, M.sc., D.Sc. (Adel.), | PHYLLIS M. ROUNTREE, pD.sc. (Méib.),

Ph.D. (Camb.), F.A.C.1., F.Inst.P. Dip.Bact. (Lond.). |
D. J. K. OPCONNELL, s.J., D.se., Ph.D., | W. B. SMITH-WHITE, m.a. (Cantab.),
F.R.A.S. | B.Sc. (Syd.).

Honorary Secretaries :
IDA A. BROWNE, D.sc. | K. E. BULLEN, M.a., se.D., F.R.S.

Honorary Treasurer :
H. A. J. DONEGAN, 4.8.7.c., A.A.C.I.

Members of Council:

J. P. BAXTER, B.Sc., Ph.D., A.M.I.Chem.E. | F. D. McCARTHY, pip.anthr.

G. BOSSON, m.sc. (Lond.). | P. R. McMAHON, m.agr.sc. (N.Z.), Ph.D.
H. B. CARTER, B.v.sc. | (Leeds), A.R.1.C., A.N.Z.I.C.

N. A. GIBSON, M.sc., A.R.1.C. | C. E. MARSHALL, Ph.p., D.sc.

T. IREDALE, D.sc., F.R.1.c. | G. D. OSBORNE, p.sc. (Syd.), Ph.p. (Camb.).,
A. V. JOPLING, B.sc., B.E. | F.G.S.

iV NOTICES.

NOTICE.
Tue Royau Socrety of New South Wales originated in 1821 as the “ Philosophical Society
of Australasia ’’; after an interval of inactivity, it was resuscitated in 1850, under the name

of the ‘‘ Australian Philosophical Society *’, by which title it was known until 1856, when the
name was changed to the “‘ Philosophical Society of New South Wales ”’ ; in 1866, by the sanction
of Her Most Gracious Majesty Queen Victoria, it assumed its present title, and was incorporated
by Act of the Parliament of New South Wales in 1881.

TO AUTHORS.

Particulars regarding the preparation of manuscripts of papers for publication in the
Society’s Journal are to be found in the “ Guide to Authors’, which is obtainable on appli-
cation to the Honorary Secretaries of the Society.

FORM OF BEQUEST.

f) bequeath the sum of £ to the Royaut Society or New SoutH WaALEs,
Incorporated by Act of the Parliament of New South Wales in 1881, and I declare that the receipt
of the Treasurer for the time being of the said Corporation shall be an effectual discharge for the
said Bequest, which I direct to be paid within calendar months after my decease,
without any reduction whatsoever, whether on account of Legacy Duty thereon or otherwise,
out of such part of my estate as may be lawfully applied for that purpose.

[Those persons who feel disposed to benefit the Royal Society of New South Wales by Legactes
are recommended to instruct their Solicitors to adopt the above Form of Bequest.]

The volumes of the Journal and Proceedings may be obtained at the Society’s Rooms, Science
House, Gloucester Street, Sydney.

Volumes XI to LIII (that is to 1919)
> LIV ,, LXVI (1920 to 1932)
- LXVIII (1936)
5 LXX ,, LXXXITI (1938 to 1948)

» UXXXIIT and LXXXIV
Volumes I to X (to 1876) and LX VII and LXIX (1935 and 1937) are out of print.
Reprints of papers are available.

LIST OF MEMBERS.

A list of members of the Royal Society of New South Wales up to Ist April, 1951, is included

in Volume LXXXV.
During the year ended 3lst March, 1952, the following have been elected to membership

of the Society : ;

Banks, Maxwell Robert, B.sc.(Hons.), Lecturer in Geology, University of Tasmania, Hobart, Tas.

Basden, Kenneth Spencer, 4.s.T.c., Technical Officer, Department of Mining and Applied Geology,
N.S.W. University of Technology, Broadway, Sydney.

Booker, Frederick William, M.sc., c/o Geological Survey of New South Wales, Mines Department
Sydney.

Bosson, Geoffrey, m.se. (Lond.), Professor of Mathematics, N.S.W. University of Technology,
Broadway, Sydney.

Charlwood, Joan Marie, B.Sc., Biochemist, 184 Queen-street, Concord West.

Crane, Roslyn Ann, B.Sc., Librarian, Australian Leather Research; p.r. 6 Chesterfield-road,
Epping.

Darvall, Anthony Roger, M.B., B.s., D.o., Medical Practitioner, Royal Prince Alfred Hospital,
Missenden-road, Camperdown.

Dunn, Thomas Melanby, B.sSc.(Hons.), Chemistry Department, University of Sydney, Sydney.

French, Oswald Raymond, Research Assistant, University of Sydney ; p.r. 66 Nottinghill-road,
Lidcombe.

Goldstone, Charles Lillington, B.agr.sc. (N.Z.), Lecturer in Sheep Husbandry, N.S.W. University
of Technology, c/o East Sydney Technical College, Darlinghurst.

Heard, George Douglas, B.se., Maitland Boys’ High School, East Maitland, N.S.W.

Holmes, Robert Francis, 15 Baden-street, Coogee.

Jamieson, Helen Campbell, a.s.t.c.; p.r. 3 Hamilton-street, Coogee.

Johnson, William, Geologist, c/o The Supervising Engineer, Warragamba Dam, N.S.W.

Jones, Roger M., Laboratory Assistant, Sydney Technical College; p.r. 69 Moore Park-road,
Centennial Park.

Lawrence, Laurence James, B.Sc.(Hons.), Lecturer in Geology, N.S.W. University of Technology ;
p-r. 28 Church-street, Ashfield.

Livingstone, Stanley Edward, 4.s.T.c.(Hons.), A.A.C.1., Lecturer in Inorganic Chemistry, N.S.W.
University of Technology ; p.r. 5 Parker-street, Rockdale.

Lovering, John Francis, B.sc., Assistant Curator, Department of Mineralogy and Petrology,
Australian Museum, College-street, Sydney.

Mallaby, Hedley Arnold, B.Sc.(For.), Dip.For. (Canberra), 114 Kurrajong-avenue, Leeton.

Males, Pamela Ann, 81 Frederick-street, St. Peters.

Minty, Edward James, B.Sc., 2 Drayton Court, Mosman-street, Mosman Bay.

Murray, James Kenneth, B.sc., 237 South-road, Broken Hill, N.S.W.

O’Dea, Daryl Robert, a.s.t.c.; p.r. 123 Perouse-road, Randwick.

Packham, Gordon Howard, 61 Earlwood-avenue, Earlwood.

Rector, John, B.sc., 46 Sir Thomas Mitchell-road, Bondi Beach.

Robinson, David Hugh, 4.s.T.c., Chemist, 21 Dudley-avenue, Roseville.

Sellgren, Elise Evelyn, B.sc., Teaching Fellow, Department of Geology, The University of Sydney ;
p-r. 76 Bream-street, Coogee.

Stevens, Robert Denzil, 32 Menangle-road, Camden.

Stuntz, John, B.sc., 511 Burwood-road, Belmore.

Weatherhead, Albert Victor, F.R.M.S., F.R.P.S., Technical Officer, Geology Department, N.S.W.
University of Technology; p.r. 3 Kennedy-avenue, Belmore.

Whitley, Alice, B.sc., Teacher, 39 Belmore-street, Burwood.

Whitworth, Horace Francis, m.sc., Mining Museum, Sydney.

Honorary Member.
Fairley, Sir Neil Hamilton, C.B.E., M.D., D.Sc., F.R.S., 73 Harley-street, London, W.1.

Obituary.

1916 Septimus Birrell.

1913 Edwin Cheel.

1916 Henry James Hoggan.

1909 Thomas Harvey Johnston (Corresponding member since 1912).
1930 William Perey Judd.

1915 Andrew Gibb Maitland (an Honorary member).

1951

NOTICES.
AWARDS.

The Clarke Medal.
Stillwell, Frank L.. p.sc., C.S.I.R.O., Melbourne.

The James Cook Medal.
Gregg, Norman McAlister, m.B., B.s., Macquarie Street, Sydney.

The Edgeworth David Medal.
Bolton, John Gatenby, s.a., Division of Radiophysics, C.S.I.R.O., Sydney.

The Society's Medal.

Penfold, Arthur Ramon, F.R.A.C.1I., F.c.s., Director, Museum of Applied Arts and
Sciences, Sydney.

Royal Society of Nem South Wales

REPORT OF THE COUNCIL FOR THE YEAR ENDING 3lst MARCH, 1952.
PRESENTED AT THE ANNUAL AND GENERAL MONTHLY MEETING OF THE SOCIETY,
2npD APRIL, 1952, IN ACCORDANCE wItTH RULE XXVI.

The membership of the Society at the end of the period under review stood at 385, an increase
of 16. Thirty-five new members were elected during the year and 13 members were lost by
resignation. Six members have been lost to the Society by death during the year 1951 :

Septimus Birrell (elected 1916),
Edwin Cheel (elected 1913),
Henry James Hoggan (elected 1916),
Thomas Harvey Johnston (elected 1909) (Corresponding member since 1912),
William Percy Judd (elected 1930), and
Andrew Gibb Maitland (elected 1915), an Honorary member, died in January, 1951.
During the year nine General Monthly Meetings were held, at which the average attendance
was thirty-six. Nineteen papers were accepted for reading and publication by the Society,
twelve less than the previous year.
An Exhibit, ‘‘ Bananas containing Seeds ”’, was discussed by Dr. C. J. Magee at the meeting
on 6th June, 1951.
Lecturettes given during the year were as follow:
2nd May, 1951: ‘‘ Colloids”’, by Prof. A. E. Alexander.
4th July, 1951: ‘* Polyploiedy and Evolution ’”’, by Mr. 8S. Smith-White.
Addresses of general interest were given at several meetings.

6th June, 1951: The evening was devoted to a discussion on “* Fluorine ’’, and the following
speakers gave addresses :

Prof. J. P. Baxter: ‘‘ New Derivatives of Fluorine.”’
Mr. H. R. Sulliven: ‘‘ Some Biological Aspects of Fluorine.”
Dr. D. L. Ingles: ‘‘ Defluorination of Artesian Waters.”

5th December, 1951: Mr. G. G. Blake spoke on “‘ Xerographic Processes ’’ and Dr. G. D.
Osborne and Mr. J. S. Proud showed “ Coloured Slides of Central Australia, mostly of a Geological
Interest ”’.

Question.—At the meeting held on the 4th July, 1951, Mr. H. W. Wood answered the question
‘““ How many Star Systems are known ?”’

The meeting held on the Ist August, 1951, was devoted to a Symposium on ‘“‘ Temperature ”’,
at which the following addresses were given:

*“* The International Temperature Scale in Science and Industry ’’, by Mr. W. R. G. Kemp.
‘“* Some Physiological Aspects of Temperature ”’, by Prof. F. 8. Cotton.

* Vulcanological Aspects’, by Dr. G. D. Osborne.

‘The Temperature of the Earth’s Interior’, by Prof. K. E. Bullen.

A Film Evening was held on 3rd October, 1951, and, through the courtesy of the Australian
Museum, Prof. A. P. Elkin and the Motion Picture Division of the U.S. Information Service,
three films were shown :

(a) “‘ Two Hundred and Fifty Million Years Ago.”
(6) ‘‘ Arnhem Land Dances.’’
(c) ‘““ The Story of Palomar.”

The meeting devoted to the commemoration of great scientists was held on 7th November,
1951. The following addresses were given :

“William Harvey and the Circulation of the Blood ”’, by Prof. E. Ford.
“Linneus ”’, by Prof. L. G. M. Baas Becking.
“What Science Owes to the 1851 Exhibition’, by Dr. T. Iredale.

vill REPORT OF COUNCIL.

Five Popular Science Lectures were delivered during the year, and were appreciated by
members of the Society and the public :

17th May: ‘‘ Wool’’, by Dr. P. R. McMahon.

19th July: ‘‘ Science, Medicine and Health in the Twentieth Century ’’, by Prof. Harvey
Sutton.

16th August: ‘‘ Shore Fluctuations: Their Causes and Effects ’’, by Dr. W. R. Browne.

20th September: ‘‘ Chemical Discovery and Invention through the Last Half Century ”’,
by Prof. R. J. W. Le Fevre.

18th October: ‘Seismology’, by Rev. D. J. K. O’Connell.

Jubilee Conversazione.—To commemorate the Jubilee of the Commonwealth of Australia,
our Society, collaborating with other scientific bodies, organized a conversazione, which was
held in the Great Hall of the University of Sydney on 18th April, 1951. This successful and
well attended function was arranged to illustrate the scientific achievements of Australia during
the last fifty years.

The Annual Dinner of the Society was held in the Withdrawing Room of the Union, University
of Sydney, on the 27th March, 1952. There were present fifty-five members and friends.

The Section of Geology, whose Chairman was Mr. R. O. Chalmers and Hon. Secretary Mr.
N. C. Stevens, held eight meetings during the year, at which the average attendance was eighteen
members and six visitors. The activities included the showing of films, exhibits, notes and
lecturettes.

The Council of the Society held eleven ordinary meetings during the year, at which the
average attendance was twelve.

. On Science House Management Committee the Society was represented by Mr. F. R. Morrison
and Mr. H. O. Fletcher ; substitute representatives, Dr. R. L. Aston and Mr. H. H. Thorne.

On Science House Extension Committee the Society was represented by Dr. R. C. L. Bosworth
and Dr. C. J. Magee.

Sir Neil Hamilton Fairley was elected an Honorary Member of the Society at the Annual
and General Monthly Meeting held on the 4th April, 1951.

The Clarke Memorial Lecture for 1951 was delivered by Dr. A. B. Edwards on the 21st June,
1951, the title being ‘‘ The Ore Minerals and their Textures ”’.

The Clarke Memorial Medal for 1952 was awarded to Professor J. G. Wood, of the University
of Adelaide, for his distinguished work both on the vegetation of arid Australia and on mineral
nutrition and metabolism in plants.

The Medal of the Royal Society of New South Wales for 1951 was awarded to Mr. A. R.
Penfold in recognition of his outstanding services to the Society and his valuable researches in
the chemistry of essential oils.

The Edgeworth David Medal for 1951 was awarded to Mr. J. G. Bolton for his outstanding
researches in the field of radio-astronomy.

The James Cook Medal for 1951 was awarded to Dr. N. McAlister Gregg for distinguished
contributions to medical science, particularly the discovery of, and work on, the connection
between congenital defects in children and the occurrence of Rubella in the mother during
pregnancy.

During the year overseas visiting scientists entertained by the Council were :

Sir Edward and Lady Mellanby, on 21st September.

The scientific members of the Danish ‘“‘ Galathea ’”’ Deep Sea Expedition, on 22nd November.

Sir Neil Hamilton Fairley, F.R.S., on 22nd November.

The financial position of the Society, as disclosed by the annual audit, is not a satisfactory
one. The deficit for the year is £736, as compared with the previous year of £687. The largest
increase in a single item of expenditure was £203 for salaries.

Although the Council has been applying rigid standards and fewer papers are being accepted
for publication, the cost per page of printing the Journal and Proceedings has increased still
further, and meeting this charge must remain one of our most serious problems.

At the General Monthly Meeting of 3rd October, 1951, Rule IX was altered to read: “ The
Annual Subscription shall be three guineas payable in advance, but members who are under
twenty-eight years of age shall be required to pay only one and a half guineas.

‘“‘ The amount of thirty-five guineas may be paid at any one time by a financial member as a
life composition for the ordinary annual payment.”’

The Society’s share of the profits from Science House for the year was £430, a decrease of
£20 on the previous year.

The grant from the Government of New South Wales of £400 has been received. The
continued interest of the Government in the work of the Society is much appreciated.

REPORT OF COUNCIL. 1x

The Library.—The amount of £60 16s. 6d. has been spent on the purchase of periodicals,
and £17 8s. 6d. on binding.

Exchange of publications is maintained with 417 societies and institutions, an increase
of 11.

The number of accessions entered in the pple us during the year ended 29th February, 1952,
was 3,392 parts of periodicals.

The number of books, periodicals, etc., borrowed by members, institutions and accredited
readers was 316.

Among the institutions which made use of the library through the inter-lbrary borrowing
scheme were : National Standards and Radiophysics Laboratories ; Waite Agricultural Research
Institute ; Royal Society of Tasmania; Fisher Library ; Colonial Sugar Refinery ; McMaster
Laboratory ; Sydney Technical College ; Public Library of Victoria ; Forestry Commission,
N.S.W.; Sydney Hospital Library ; Department of Mines, Victoria; Department of Health,
N.S.W. ; Department of Public Works, N.S.W.; University of Adelaide ; Department of Wood
Technology ; Food Preservation Laboratory, Homebush; New England University College ;
C.S.L.R.O. Irrigation Research Station; Standard Telephones and Cables; Department of
Agriculture, N.S.W. ; Bureau of Mineral Resources ; M.W.S. and D. Board, N.S.W. ; Institution
of Engineers, Australia; C.S.I.R.O. Division of Industrial Chemistry ; C.S.I.R.O. Division of
Fisheries ; Geological Museum and Library, Melbourne; Snowy Mountains Hydro-Electric
Authority ; Taubman’s Ltd. ; C.S.I.R.O. Division of Entomology ; State Fisheries Laboratory ;
University of Melbourne ; Standards Association of Australia; Granville Technical College ;
Australian Museum; Botanic Gardens, Sydney; Queensland Institute of Medical Research ;
Glenfield Veterinary Research Station; Commonwealth Observatory ; Newcastle Technical
College.

R. C. L. BOSWORTH,
President.

7,007
24,886

£32,802

10,160

25
6

£32,802

BALANCE SHEET.

THE ROYAL SOCIETY OF NEW SOUTH WALES.
BALANCE SHEET AS AT 29th FEBRUARY, 1952.
LIABILITIES.

1952.
y S, 3a. NAY So
Accrued Expenses " 269 ll 9
Subscriptions Paid in Advance Reh 22 11 6
Life Members’ Subscriptions — Amount carried
forward 113 11 O
Trust and Monograph Capital Funds (detailed below)—
Clarke Memorial 1,941 16 9
Walter Burfitt Prize 1,081 16 8
Liversidge Bequest . 744 5 1
Monograph Capital Fund 3,659 18 4
a 7,427 16 10
ACCUMULATED FUNDS 24,088 1 2
£31,921 12 3
ASSETS.
1952.
x s. d. £. ad.
Cash at Bank and in Hand . 572 9 11
Investments—Commonwealth Bonds and Inscribed Stock,
etc.—at Face Value—
Held for—
Clarke Memorial Fund 1,800 0 0O
Walter Burfitt Prize Fund 1,000 0 0O
Liversidge Bequest .. 700 0 O
Monograph Capital Fund 3,000 0 0
General Purposes 2,860 0 0
9,360 0 0
Debtors for Subscriptions... 102 7 6
Deduct Reserve for Bad Debts LOZ Wii
Science House—One-third Capital Cost 14,835 4 4
Library—At Valuation 6,800 0 0
Furniture—At Cost—less Depreciation 324 18 0
Pictures—At Cost—less Depreciation 24 0 0
Lantern—At Cost—less Depreciation 5 0 0
£31,921 12 3

BALANCE SHEET.

TRUST AND MONOGRAPH CAPITAL FUNDS.

Walter
Burfitt
Prize.

Clarke

Memorial.

5 gs. d.

Capital at 28th February, 1951 1,800 0 O
Revenue——

Balance at 28th February, 1951 126 2,10

Interest for Twelve Months 57 ney fas KY)

183 10 8

Deduct Expenditure 41 13 11

Balance at 29th February, 1952 £141 16 9

£

s.

d.

1,000 0 0

126 10 11
oe WW A: 5

158 8 3
76 11

7

Sol Ojos

ACCUMULATED FUNDS.

Balance at 28th February, 1951
Less—

Increase in Reserve for Bad
Debts .. Ps eed te

Loss on Sale of Inscribed Stock

Deficit for twelve months (as
shown by Income and Ex-
penditure Account)

Bad Debts written off

Liversidge
Bequest.

xl

Monograph
Capital
Fund.

a on £ s. d.
700 0 O 3,000 0 0O

21 18 9 Voor: Oke 6
ZI a (Gr A! 95 12 10
7G a sn | 828 13 4
ates 168 15 0O
$44.5 t £659 18 4
£ si di
24,886 2 9
798) bo 7

£24,088 1 2

The above Balance Sheet has been prepared from the Books of Account, Accounts and
Vouchers of the Royal Society of New South Wales, and is a correct statement of the position

of the Society’s affairs on the 29th February, 1952, as disclosed thereby.

We have satisfied

ourselves that the Society’s Commonwealth Bonds and Inscribed Stock are properly held and

registered,

Prudential Building,
39 Martin Place,
Sydney, 18th March, 1952.

HORLEY & HORLEY,

Chartered Accountants.

xi

1950-1.

141

579
21
£2,365

1950-1.
£
624

BALANCE SHEET.

INCOME AND EXPENDITURE ACCOUNT,

Ist March, 1951, to 25th February, 1952.

To Annual Dinner—

Expenditure
Less Received

Audit. .

Cleaning

Depreciation

Electricity ; we

Entertainment Expenses :

Insurance...

Library Purchases and Binding .

Miscellaneous : : Ree

Postage and Telegrams : ae

Printing and Pe J ournal—Vol. 84
Less Received . Ls te 2

Printing—General ..

Rent—Science House Management Committee

Reprints—
Expenditure
Less Received

Salaries
Telephone

By Membership Subscriptions

Proportion of Life Members’ Subscriptions
Government Subsidy ite
Science House—Share of Surplus

Interest on General Investments

Other Receipts

Deficit for Twelve Months os

829 14 9
75° 0 ©

380 12 O
25219 4

128
782
24

WH AADWODWOS

ow
Doo

8
a
£2 0
6

16

£2,325

£2,325 16 6

1951-52.

£
632

10
400
430
116

6.)
12 6
10 O
0 0
0 0
12 11

1,589
736

15
ia

Ol

ABSTRACT OF PROCEEDINGS

OF THE SECTION OF

GEOLOGY

Chairman: R. O. Chalmers, A.S.T.C.
Honorary Secretary: N. C. Stevens, B.Sc.

Meetings.—Eight meetings were held during the year 1951, the average attendance being
eighteen members and six visitors.

April 20th.—Annual Meeting. Election of Office-bearers for 1951: Chairman, Mr. R. O.
Chalmers ; Honorary Secretary, Mr. N. C. Stevens.

Films: Colour slides were shown by Mr. A. V. Jopling (Broken Hill), Mr. G. EK. McInnes
(Warrumbungle Mtns. and the A.C.T.), Mr. N. C. Stevens (Belubula River district), and
Mr. T. G. Vallance (New England and North Coast).

Exhibit: By Mrs. K. M. Sherrard, of Upper Ordovician graptolites from Carboona
Gap, near Jingellic.

May 11th.—Lecturettes : Wallabadah District, by Mr. J. Hallinan ; Bungonia District, by Mr.
B. H. Flinter and Mr. N. O. Jones ; Wollongong District, by Mr. C. T. McElroy.

June 15th.—Address : ‘‘ Some Aspects of a Trip to Great Britain and America ’’, by Dr. G. D.
Osborne.

July 20th.—Notes and Exhibits: By Mr. A. 8. Ritchie: Glacial features near Mt. Kosciusko.
By Mr. G. E. McInnes: Magnetic Is. and Townsville. By Mr. N. C. Stevens: Geophysical
survey of bathyliths in N.S.W. By Mr. T. G. Vallance: Pleistocene glacial banded clays
from Trapyard Ck., Kosciusko. By Mr. J. M. Hallinan: Tertiary volcanic rocks from
Lamington National Park, Q. By Mr. G. Packham: Silurian plant remains from west of
Sofala, N.S.W. By Mr. W. H. Williamson: Very porous sandstone from Eumungerie,
N.S.W. By Mr. Chalmers: Photographs of parabolic current bedding in sandstone at
Glenorie and Eastwood ; and curved crystals of prehnite from Prospect, N.S.W.

August 17th.—Discussion: ‘“ Contributions to the Geology of the Marulan-Wombeyan- Yerran-
derie District.” By Mr. L. Lawrence (Yerranderie), Mr. J. Lovering (Tallong-Marulan)
and Dr. G. D. Osborne (who presented the work of Mr. C. V. Phipps in the Wombeyan-
Taralga area).

September 2ist.—Lecturettes : ‘“‘ On the Work of the Geological Survey during the Past Year ’’,
by Officers of the Survey, including a general account by Mr. C. St. J. Mulholland ; investiga-
tions in the Western Coalfield, by Mr. E. O. Rayner, and Contour Trench formations in
upland plains of N.S.W., by Mr. C. T. McElroy.

Exhibit : By Mrs. Sherrard : Upper Ordovician graptolites (zone of Nemagraptus gracilis)
from two localities north and north-east of Walli, N.S.W.

October 19th.—Addresses, illustrated by colour slides: ‘‘On N.W. Queensland and Central
Australia’”’, by Miss F. M. Quodling and Dr. H. Narain.

November 16th.—Notes and Exhibits: By Dr. C. J. Magee: Mt. Lamington, N.G. By Dr.
W. R. Browne: Deuteric alteration of South Coast lavas. By Dr. G. D. Osborne: Dunite
with phlogopite xenoliths from Mt. Strangways, Central Australia; reaction rims with
humite, chondrodite, cummingtonite and anthophyllite. By Miss F. M. Quodling: Kyanite
from Harz Ra., Central Australia. By Mr. N.C. Stevens: Crinoidal limestone from Borenore,
N.S.W. By Mr. Chalmers: Copper and coalfield material from Long Reef, Narrabeen,
N.S.W.

Obituary

Septimus BrrRRELL, who died on 17th July, 1951, had been a member of this Society since
1916. In his early days he vacated a position in the Physiology School in the University of
Sydney to join the newly-established Bureau of Microbiology of the Department of Public Health,
where he worked on matters related to pathology and dairy bacteriology.

He became Dairy Bacteriologist under Dr. Darnell Smith in the Biological Branch of the
Department of Agriculture, N.S.W., in 1913, and was directly associated with the establishment
of the course in Dairy Bacteriology for Diploma Students at Hawkesbury Agricultural College.
In 1915 he became a commercial biochemist, and was recognized as an authority on the
bacteriology and chemistry of fats and oils. He took an active part in the Australian Society
of Dairy Technology, of which he was a foundation member.

EDWIN CHEEL, who was born in England on 14th January, 1872, and died in Sydney on
19th September, 1951, had been a member of this Society since 1913.

On his arrival in Australia as a young man he worked as a gardener in Queensland and in
Sydney. In 1897 he was appointed to the staff of Centennial Park, Sydney, and later transferred
to the gardening staff of the Sydney Botanic Gardens, where he was entrusted with the care and
maintenance of the Cryptogam collections of the Herbartum. His keen interest in the knowledge
of his work led to his appointment to the botanical staff in the National Herbarium in 1908, and
to his appointment in 1933 to the position of Chief Botanist and Curator, which he held until
his retirement in 1936.

He had a wide knowledge of many groups of Australian plants, but his greatest interest
was in the Myrtacez, many species of which he cultivated and observed at his home at Ashfield
and on private ground at Hill Top, south of Picton. His plant collections added much to the
resources of the National Herbarium.

He contributed many botanical papers and articles to scientific journals in N.S.W., including
21 to This Society. He took an active part in the work of the scientific societies in Sydney and
was President of the Naturalists’ Society of N.S.W. in 1924, of the Linnean Society of N.S.W.
in 1930, of This Society in 1931, and of the Botany Section of A.N.Z.A.A.S. in 1937. He was a
delegate of the A.N.R.C. to the Fifth Pacific Science Congress in Canada in 1933.

He was a warm supporter of the Friendly Society movement, and served in the highest
position in the Manchester Unity Order of Oddfellows.

In 1943 This Society awarded its Bronze Medal to Edwin Cheel “ in recognition of his contri-
butions in the field of botanical research and to the advancement of science in general ”’.

Henry JAMES Hocean, a member of This Society since 1916, died on 25th December, 1951,
at the age of 74. Born in Scotland, he came to Australia at an early age, and received his schooling
in Australia.

For nearly thirty years he was a teacher of Engineering Trades Drawing at St. George
Technical College (1910-1939) and from 1934 to 1939 also taught at the Sydney Technical College.

As Consulting Engineer to St. George County Council Electricity Supply Undertaking, he
was responsible for the planning and construction of the distribution system during the period
from 1920 to 1923.

He was a foundation member of The Institution of Engineers, Australia, and was formerly
a member of the Engineering Association of Australia.

WILLIAM PERCY JUDD, who died on 27th November, 1951, at the age of 77 years, had been a
member of This Society since 1930.

THOMAS HARVEY JOHNSTON was born and educated in Sydney and died at Adelaide on 30th
August, 1951, aged 70. He became a member of the Society in 1909, and a Corresponding
member in 1912. He graduated in Arts at the Sydney University in 1904 and in Science in
1906. He obtained his M.A. in 1907 and D.Sc. in 1911. He was Lecturer in Zoology and
Physiology at the Sydney Technical College (1907-1909) and Assistant Microbiologist in the
Bureau of Microbiology, N.S.W. (1909-1911), in which State he was responsible for the revival
of plant pathology after the departure, in 1905, of Dr. N. A. Cobb to take up a post in Hawaii.
Among his early studies of note in this field were those associated with first outbreaks of Irish
blight of potatoes in Australia. He was Lecturer, and later Professor, of Biology in the University
of Queensland (1911-1922), and Professor of Zoology in the University of Adelaide (1922-1951).

OBITUARY. XV

Professor Johnston has been one of the leading Australian zoologists for many years and was
Honorary Zoologist to the Australian, Queensland and South Australian Museums.

In the course of his work he travelled widely : as a member of the Prickly Pear Travelling
Commission, which visited many countries of the world in search of a means of controlling prickly
pear (and it was on his recommendation that the Commission introduced the cochineal parasite
to control the pest) ; on several anthropological expeditions from Adelaide into Central Australia ;
‘ and twice to the Antarctic with the British, Australian and New Zealand Antarctic Research
Expeditions.

He published more than 200 scientific papers, chiefly on Australian parasitology. He was
President of the Royal Society of Queensland, Queensland Field Naturalists’ Club, Royal Society
of South Australia, Entomological Club of South Australia, Anthropological Society of South
Australia, and Zoological Section of the Australian Association for the Advancement of Science.
In recognition of the excellence of his work he was awarded the David Syme Memorial Medal
(1913), the Polar Medal (1934), the Sir Joseph Verco Medal (1935), and the Mueller Memorial
Medal (1939).

ANDREW GisB MAITLAND, who was elected an Honorary Member of This Society in 1915 and
died on 27th January, 1951, was the last of the pioneer geologists of Australia. Born in Hudders-
field, Yorkshire, on 30th November, 1864, he studied geology at Yorkshire College, Leeds, under
Professors A. H. Green, W. W. Watts and J. E. Marr.

At the age of twenty-four, he was appointed Assistant Geologist to the Geological Survey
of Queensland, where, under the direction of the late Dr. R. L. Jack, he spent eight years, some of
his most important work being connected with the survey of the north-eastern intake beds of
the Great Artesian Basin. During his period of service in Queensland, he was seconded in 1891
to accompany Sir William Macgregor in his explorations in New Guinea, and produced the first
systematic account of the geology of Papua as known at that time.

In 1896 he went to Western Australia as Government Geologist and Director of the Geological
Survey, a position in which he laboured for thirty years. Apart from administrative
achievements, which included the organization and enlargement of the Geological Survey and
the determination and implementation of its policy, he made important and far-reaching contri-
butions to the general and economic geology of the State. Outstanding among these were the
discovery of artesian water and the elucidation and mapping of the geology of the Pilbara area in ©
the ‘‘ north-west ”’.

It was due to his careful field work that he was able to predict the occurrence of valuable
supplies of artesian water in the country between Shark Bay and North-West Cape (an area of
pastoral possibilities but poorly provided with surface water), and later in the West Kimberley
region, around Perth and in the Nullarbor Plain.

He initiated a geological survey of the Pilbara region (which became noted as a producer of
gold, tin, tantalum and the rare-earth metals) and was responsible for the publication of several
important Bulletins of the Geological Survey dealing with the geology of various mineral fields
im the State. The extensive knowledge of the geology of the State gained by himself and his
staff was epitomized in his ‘‘ Summary of the Geology of Western Australia ’’, published in 1919.
accompanied by a coloured geological map.

Maitland took an active interest in the advancement of Science. He played an important
part in the foundation of the Royal Society of Western Australia, of which he was twice President.
and was for twenty-five years Local Secretary of A.N.Z.A.A.S. and President of Section C (Geology)
in 1907.

He was awarded the M ueller Medal of A.N.Z.A.A.S. in 1924, the Clarke Memorial Medal of
This Society in 1927, and the Kelvin Memorial Medal of the Royal Society of Western Australia
in 1937.

Ag

i . r j ah
‘ ue “
4
a: oa
i) Pete tive [eiite sain d poe ey Drct otuad phat eng
VORA SAAS bine, ahah ee Pek Yee fae fats, eect Ri
pea Mae ts ae Chee ey be Qe ee Te
ay hy ae ati i Fi , E Pitas Hye eS iy M i etae (Abert a ea bobeds
Ae? hy te 1 arory PTS FRY) yeh of Luly ff SPE 4 15 3 bi phat CESERTIZ IY et ianel te) Hagety
PR Ly baa aN al? eeees He rey pa pnts he awh’? rth ike
aa Beer ded ' buy mel Weer

t Lehsy hh Pies
ih - 3
qin et hs ;
= Dei | baa Pel s
: an
i i bei 4 * Siu iy Pig's
aU ax ;
. % t i i We
+ ave |
7 ‘
yo i ‘
‘ ts Y
7 P De,
i
2 see
sf U > pn
|
ra f
‘ . gre
. a
¥ P, ' a at) 4
j ; y P
| {
r a ‘ f
©
4 j Me
i We 19, 7 hc
FER HS apeitey yal ended Da ‘ oan ei ie
rn i i
:
»
i wa aa {
j ! | 4 é
i
* '
‘
) j '
a ve ts
, ’
' J ' , >
= : , t
Phy t R'
> t4
fi % i soos PERE Ki, 5
&
>| 4 i ¥
i yt a PACE ve
i \ +
1 ohh ates ‘
kK ‘
. f ii io j
~~, \
.
%
; 2 shel tes
if ome PRRAAT ET
r : fs /
+ Dae {i
S 4 Fe &
x {
; te ) bree i ‘
Ugh ) { p ~ 4
. } ‘ Rae Oe
ed Uy £75 z z SLA Cae te ee ie
+ fr
‘
.
if v
“
{
' ,
)
{
¥ j

PRESIDENTIAL ADDRESS

By R. C. L. BOSWORTH, Ph.D., D.Sc.

Delivered before the Royal Society of New South Wales, April 2, 1952,

Part I.

THE SOCIETY’S ACTIVITIES DURING THE PAST YEAR.

The year under review will long be remembered for two important events :
the Jubilee of the Commonwealth of Australia, followed shortly afterwards by
the tragic news of the death of His Well-Beloved Majesty King George VI.
I was among one of the many millions who were shocked by the solemn news,
which came through so unexpectedly just eight weeks ago. On behalf of the
Members and Council I sent a letter to Her Majesty Queen Elizabeth expressing
our respectful sympathy.

In the Jubilee celebrations, your Society played its part as the leading
organizer in the Jubilee Conversazione held in the Great Hall of the University
of Sydney. Other scientific and professional bodies co-operating with the Royal
Society in this event were: The Institution of Engineers, Australia; The
Royal Australian Chemical Institute ; The Institute of Physics, N.S.W. Division ;
The Institute of Mining and Metallurgy ; and the Linnean Society of N.S.W.
The Conversazione provided a brilliant display on the evening of Wednesday,
April 18th, and was much appreciated when the exhibition was thrown open to
the general public on the following day. I think it is only fit at this stage that
we should express our appreciation of the pioneer organizational work done for
this function by your last year’s Council, and in particular by my immediate
predecessor in office, Mr. F. R. Morrison.

As another fitting mark of the Jubilee year, the Society arranged to cover,
in its normal course of popular science lectures, representative scientific achieve-
ments of the last half century, with special reference to Australia. Lectures
were given on the following subjects :

On May 7th Dr. P. R. McMahon spoke on ‘ Wool ”’.

On July 18th Prof. Harvey Sutton spoke on “Science, Medicine and
Health in the 20th Century ”’.

On August 16th Dr. W. R. Browne spoke on ‘‘ Shore Fluctuations, Their
Causes and Effects ”’.

On September 20th Prof. R. J. W. Le Fevre spoke on “‘ Chemical Discovery
and Invention Through the Last Half Century ”’.

On Octcber 18th Rev. J. L. K. O’Connell spoke on ‘“ Seismology ”’.

While in some instances a combination of inclement weather and electricity
blackouts resulted in audiences of a disappointing size, their enthusiasm
invariably paid tribute to the lecturer, and I would like again to express to our
popular science lecturers the appreciation of the Society.

The general meetings of the Society have, this year, been somewhat more
varied than usual. Fewer papers have been read before the Society, and in

B

2 R. C. L. BOSWORTH.

their place lecturettes and exhibitions have been given. Two symposia were
arranged and one evening was devoted to the showing of scientific films. The
subject matter at the meetings generally was well balanced and was appreciated
by the audiences.

On Monday, June 18th, your President and Secretary waited on His
Excellency the Governor, Sir John Northcott. His Excellency expressed keen
interest in the affairs of the Society and was looking forward to an occasion
when he might be present at one of our functions.

The Council, to whose report you have listened earlier this evening, has
been untiring in applying itself to the business of the Society and in giving me,
as your President, unstinted co-operation. In a year of financial stringency,
there have been many controversial matters brought before the governing body
of the Society, and while initial opinions held by Councillors were often very
varied, yet when the matter in question had been thoroughly discussed and the
time had come to take the vote it was frequently an unanimous one. I repeat,
your Council has been one which it has been a pleasure to lead, and I am grateful
to each and every one of them. Particularly, however, am I indebted to the
members of your Executive Committee. I was fortunate in that this body
contained two past Presidents of the Society, who were able to give me valued
advice aS well as co-operation, namely Mr. Morrison as the immediate past
President, and Mr. Wood who was President in 1949 and who, in spite of the
fact that he has already served the Society for many years, yet accepted again
the responsibilities of Business Secretary, responsibilities which he shouldered
So willingly and so well that the task of your President was made so much the
lighter. Dr. Ida Browne has again acted as Editorial Secretary, and in that
position she has done, I assure you, more than any other single person in restoring
our Journal to a position where it appears reasonably on time. In addition,
Dr. Browne has been tireless in exploring all possibilities of saving costs by
efficient editorship.

And now to our Honorary Treasurer, Dr. C. J. Magee, there is due a special
word of appreciation. Under present conditions the position of treasurer in
any society such as ours is a most unenviable one, and while, as you have heard,
we have recorded a large deficit, I can assure you that, had it not been for the
efficient management of Dr. Magee, the deficit would have been a far larger one.

Among the other functions performed for the Society by Dr. Magee has
been that of placing our financial position before persons and bodies likely to
be interested. Not long ago he visited officers of the Rural Development
Branch of the Commonwealth Bank, and shortly afterwards we heard that the
Society had been made a grant of £400 from the fund administered by this
branch of the Bank. The grant arrived too late for inclusion in this year’s
balance sheet but will provide a very favourable starting point for the new
financial year under our new President, Dr. C. J. Magee. Let me here express
our deep appreciation of the action of the Bank in making this grant.

I must also at this stage refer to the valuable work done by our Honorary
Librarian, Mr. F. N. Hanlon. The duties which he has performed have largely
relieved the Secretary of responsibility for library matters, permitting that officer
to devote himself more fully to the other business of the Society. I, as President,
the Council and every member are indebted to Mr. Hanlon in this matter.

Lastly, I must refer to the work of our Assistant Secretaries, Miss M. Ogle
and Mrs. M. Golding, who have managed the office and library of the Society
in a most efficient and courteous manner. Few members are aware that the
smooth running of these general meetings of ours, and likewise our Council
meetings, are due to the very complete and orderly agenda prepared in advance
by our Assistant Secretaries.

PRESIDENTIAL ADDRESS. 3

And now it is my regretful duty to refer to the balance sheet. We are
passing through a period when the cost of every service we offer to members is
erowing, and in many cases growing at an alarming rate. While our Society
can still claim to be a comparatively wealthy one, we cannot continue to sustain
such deficits and must either contract our services or seek additional sources of
revenue. For some little time your Executive has been pressing the State
Government for an increase in our grant—so far unsuccessfully. However,
you have already heard of one successful venture in this field by our President-
elect, and it will be one of the most urgent tasks of the new Council you have
just elected to continue to explore all other possible lines of action in this matter
of securing additional revenue from those bodies who, directly or indirectly,
derive benefit from the activities of the Society.

During the year we have had somewhat fewer papers accepted for publication
than in the immediately previous years. Not only have there been fewer papers
offered, but, in addition, your Council has adopted a more stringent editorial
policy. The world standing of the Society is judged, primarily, on the contents
of its Journal, and if the rising costs of publication are to enforce restriction of
our output it is doubly imperative to ensure that those papers which are accepted
are of the highest possible quality.

I might, at this stage, be permitted a word to the authors of our papers.
I have gained the impression, both from the chair, and, in earlier days, in the
audience, that authors, and particularly young authors, when reading a paper
are very much on the defensive and are attempting to justify their methods and
conclusions before the audience in this hall. I would remind these authors
that before their papers are even accepted by the Council for reading and/or
publication that these papers have been submitted to the highest authorities
in the appropriate subject which your Editor and Council could find, and found
acceptable. When it comes to reading to the persons present in this hall, the
author is usually by far the highest authority on the subject matter of his paper
and the audience are interested primarily in comprehending sufficient of the
subject matter to be able to judge what relation new facts brought forward by
the author bear to the structure of science as a whole and each to their own
little field in particular. As I said at the dinner six days ago, one of the most
pressing problems facing scientists as a class is that of breaking down the growing
barrier of incomprehensibility between experts in different fields, and I believe
that the author who is reading an abstruse paper before the body of this Society
and attempting to make the contents comprehensive to the audience is making
a genuine effort in that regard.

The translation of scientific achievement to improved standards of living
usually demands the co-ordination of the experts in many different fields, not
only scientific—a fact recently observed by Meier (1951)—and the Royal Society
would serve an end valuable not only to scientists themselves but to the industrial
world as well, if it could promote better understanding between scientists in
different fields.

PART If,

TRANSPORT PROCESSES IN APPLIED CHEMISTRY.

(I) INTRODUCTION.

While a presidential address has to be acceptable to an audience of mixed
scientific interests, it is also traditionally expected to deal with some of the
more recent developments in one particular branch of science. In an attempt
to make some kind of a bridge between the requirements of generalization on

BB

4. R. C. L. BOSWORTH.

one hand and close specialization on the other, I shall follow the course of first
defining the relation between the subject matter of this address and the better
known physical sciences.

The translation of a chemical reaction or synthesis from a primary gain in
chemical knowledge made in the laboratory to an industrial achievement involves
the co-ordination of chemical developments with those gained in other fields.
This has now been known for a considerable time and has frequently formed the
bases of discussions, especially in the last decade. An important member of
these allied studies is that of Physics, concerned as that subject is with the
measurement, control and co-ordination of the properties of reactants, inter-
mediates, resultants and equipment and with the balance of the forces acting
between these bodies (Bosworth, 1950).

While physicists in general would agree that their subject finds important
application in the field of chemical industry, there appears to be a fertile field
for discussion in considering just what branches of physics are of major
importance in such an application. There is fairly general agreement, for
example, that methods of measurement of the physical properties of materials
of construction and process materials form an important section of such a
study. It is more difficult to get agreement as to the value of the other less
direct applications, and it is on one of these applications that I have elected to
speak this evening.

Many of the engineering operations of the chemical and other industries
involve the transport of matter and/or physical properties from one point to
another in a material assembly. The chemical engineering unit operation of
drying, for example, involves the transfer of water molecules from one medium
(the material to be dried) to another (the drying agent), and simultaneously
and necessarily the transfer of heat in the opposite direction. The study of
this and other allied unit operations is of subject matter proper to chemical
engineering, but the study of the transport processes in themselves is, I claim,
an aspect of physics as applied principally, but by no means exclusively, to the
chemical industry.

(II) THE Two WAYS OF STUDYING TRANSPORT PROCESSES.

In the discussion of the transport of physical properties such as heat,
momentum and the electric current, it is convenient to consider the property
concerned as transported by the movement of certain mobile parts of the assembly
which, in moving, carry the property concerned with them. Heat and electric
current in metals, for example, are conducted by the movement of the so-called
free electrons. In gases, both heat and momentum may be transported by the
random movement of the molecules, whereas electric current may only be
transported by ionized particles. In other systems, non-material entities—
electromagnetic or acoustical waves—can act as carriers. It is convenient
to extend the concept of carrier from physical properties to material particles.
Oxygen diffusing through air, for example, can be transported by the movement
of oxygen molecules, which thus constitute the carriers. Carbon dioxide
diffusing through, for example, a limed sugar solution, can be carried as carbon
dioxide molecules or in the form of carbonate or bicarbonate ions.

One particular class of carrier, aS we have seen in the example above, may
often transport two or more properties, and there necessarily arises a certain
degree of parallelism between the different transport processes so that it is
frequently possible and often desirable to argue from one process to another
by analogy.

There are two Ways in which transport processes. may be studied. We
may take a mechanical picture and consider the different carriers operative

PRESIDENTIAL ADDRESS. 5

and the mechanism of their movement. On the other hand the flow involved
in a transport process may be considered as a physical property in its own right
and related to other properties of the system. The best known example of this,
the phenomenological approach, is of course the study of current electricity.
The study of heat transfer is, of course, also well known, as in the possibility of
arguing from electric current to heat flow by means of setting up the equivalent
electric circuit. The use of such devices for studying complicated heat (and
incidentally also mass) transfer by analogy with an electrical circuit has recently
been considered in detail by Paschkis and his co-workers (1942, 1946).

(III) RECENT DEVELOPMENTS.

The mechanical study of transport phenomena on the molecular basis was
first encountered in the Kinetic Theory of gas as developed by Joule, Clausius,
Maxwell, Crookes, Jeans, Loeb, Knudsen (1934), Chapman and Cowling (1939)
and others. Here the movement of individual molecules in a molecular assembly
results in a scattering of local differences in composition or in the concentration
of such physical properties as momentum and energy; and also, if the gas is
ionized, of the electric charge.

Fundamental study on the molecular basis in condensed phases has not
proved as easy as in gaseous systems, but we have recently seen the development
of an embryo Kinetic Theory of Liquid by Born and Green (1949) and also by
Eisenschitz (1950). |

The study of transport phenomena on the molar as distinct from the
molecular bases has been developed in the theory of turbulence by Prandtl,
von Karman, G. I. Taylor, E. G. Richardson and others. Similarly, transport
produced by convective mechanism has been studied in connection with the
flow of heat, evaporation processes and, more recently, in the corrosion of
metals (Bosworth, 1950a, 1951).

On the basis of the phenemological approach to transport processes a
major development has occurred in the growth of the thermodynamics of
irreversible phenomena developed by Onsager (1931), Prigogine (1949), Cox
(1950), de Groot (1951) and others, and in the recent publication of a book on
this subject by de Groot.

Just what significance the ultimate development of theory in this field
may have for the more humble practising chemist may be appreciated by
reference to a very old physico-chemical principle. I refer to the principle of
Le Chatelier, which says in effect that any chemical reaction gives rise to physical
conditions which tend to stop and ultimately to reverse the reaction. Thus
conditions for a reaction which is exothermic are rendered less favourable by a
rise in temperature. Continuation of any reaction is thus ultimately contingent
on the transport processes which dissipate the unfavourable conditions produced
by the reaction. Further, the dissipation of any changed physical property is
usually rendered more difficult by an increase in the size of the containing vessels.
The large scale manufacturer of a chemical product is usually much more
dependent on the transport processes than the laboratory worker. Where one
or more transport processes become rate determining factors for any particular
reaction, some knowledge of these processes is imperative if the chemical engineer
is to face his design problem with intelligence.

(IV) TYPE OF TRANSPORT PROCESS.

A flow process is characterized by the occurrence of a flux or current of
some particular substance or property. This current may always be expressed
as the temporal derivative of a particular parameter of the system. The current

6 R. C. L. BOSWORTH.

further proceeds under the influence of a driving force or potential which takes
different values in different parts of the assembly.

As examples of such flow processes we may cite:

Hydraulic or mechanical flow.
The electric current.

The flow of heat.

Chemical diffusion.

Momentum flow or fluid friction.
Magnetic displacement.

Elastic and plastic deformation.
Relaxational phenomena.
Chemical reaction.

This list of processes may be divided into three classes, depending on the
tensor rank of the flux. In the first four processes the flux is a directed vector
quantity. In the next three the flux is a tensor of the second rank. In the
last two processes the flux is that of a change between two co-existing and
spatially superimposed distinct states of the assembly and as such is a scalar
quantity.

Again, magnetic displacement differs from the other processes listed above
in that continual magnetic flux can only occur in a magnetic system in which
the driving force is continually changing. The flow in such a system is referred
to as a displacement current. Pure displacement currents also occur in ideally
elastic bodies—+z.e. those not subject to plastic deformation, and in electric
phenomena in dielectric bodies.

—
SESE

SESS

i)
—

(V) GENERAL PROPERTIES OF TRANSPORT SYSTEMS.

While the properties of transport systems differ according to whether the
flux is a scalar, vector or tensor quantity, and to whether it constitutes a real
or a displacement current, there are certain general properties shown by all
such systems.

Every transport process, a8 we have seen, consists of a flux produced by
the influence of a potential difference and carried by the movement of mobile
carriers. The types of carrier operative differ both with the properties carried
and with the systems concerned. A table of some of the important carriers
with the properties for which they may be effective in maintaining flow is given
below (Table I).

TABER ,§.
Carrier Operative for the Different Transport Systems.

Carrier. | Effective in the Flow of.

Steady electromagnetic field .. | Electric current, magnetic flux.

Alternating electromagnetic field Electric current, magnetic flux, heat flux, momentum flow.

Steady elastic field 3 .. | Elastic deformation.

Alternating elastic field .. .. | Elastic deformation, heat flux, momentum flow.

Electrons Ms Me he .. | Electric current, heat flux.

Tons ae ii, hs .. | Electric current, heat flux, chemical diffusion, momentum
flow.

Molecules... Hs ro .. | Heat flux, chemical diffusion, momentum flow.

Molar aggregates... we .. | Heat flux, hydraulic flow, chemical diffusion, momentum
flow. i

Convection currents als .. | Heat flux, hydraulic flow, chemical diffusion, momentum
flow.

Metastable molecules ae .. | Heat flux, chemical diffusion.

Sard
6

PRESIDENTIAL ADDRESS.

In any transport system the flux may be regarded as proceeding from an
initial point or source to a final point or sink. The possibility of endless flux
lines (or toroidal fluxes), however, is not excluded. Given a distribution of
carriers continuous along some path or paths from the source to the sink, flux
will only occur if there exists a potential difference between the source and the
sink and, usually, if there is a continuous potential fall along the “ conducting
path ’’ joining these two points. Current or flux is the time rate of change of
some physical parameter of the system which we may call the change (Q). In
classical transport systems we may define the potential difference between any
two points (say A and B) as the work done in taking unit charge from B to A.
In considering the thermodynamics of irreversible processes, it has been found
more convenient to take a definition of the potential difference as the entropy
generated when unit charge is taken in the reverse direction, namely from
Ato B. If AS is the entropy of the system measured with respect to a state
of equilibrium, and if Q, is the charge, X, the potential and J; the flow, all
measured with respect to the it property, then we have as a definition of X,
and J,

Ley OOS)

X,= i eal cai ee (1)
dQ;

0 fee at. Siistinitishione: lal eb ieurejre ies) -el\/eue7pe) er ele \s: (6. “) 76, Fe (2)

It is, as a matter of observation, frequently noted that, at least for a limited.
range of X,; over a conducting path, the flux is directly proportional to the
potential difference

5 MSY Gi a (3)

where the property(ies) L; may be known as the conductance or phenomenological
constants. Applied to current electricity, equation (3) becomes a statement
of Ohm’s law; to chemical diffusion, Fick’s law ; to heat flow, Fourier’s law ;
to the transport of momentum, a definition of Newtonian behaviour ; and to
hydraulic flow, a statement of the Poiseuille equation.

In any transport process proceeding from a source A to a sink B under a
potential difference which is maintained at some fixed value, the flux at all
points along the conducting path eventually settles down to a condition in which
just as much charge flows into any and every elementary volume of that path
in unit time as flows out of that same volume. Such a transport process is
said to have attained steady flow conditions and is associated with a steady
potential fall along the conducting path. If now, however, the potential
difference between source and sink were suddenly altered, the current would
not immediately settle down to conditions of steady flow. Before such is
attained, a certain time must elapse, during which the current flows into local
sources and sinks all along the conducting path until such points are brought
up to the new operating local potential. While charge is thus going into storage
along the path, the flow is said to be in the unsteady state. The magnitude of
the charge required to raise any volume of the conducting path by unity is
referred to as the capacity of that volume. For a path having a capacity C
and conductance L, the condition of unsteady flow persists for a time proportional
to C/E.

For some transport processes, particularly current electricity and hydraulic
flow in pipes, it is often permissible to regard the conducting path as a linear one.
With other processes, particularly heat flow and chemical diffusion, the con-
ducting path is often very diffuse and we may prefer to have local measures of
the intensity of the current and potential properties. The most convenient

§ R. C. L. BOSWORTH.

of such local measures are the field and the flux or current density.. By the
field (#), we mean the local measure of the gradient of the potential, and by the
flux density (7) we mean the quantity of charge crossing unit area normal to the
direction of flow

and feb land). tenes aa (5)

If we choose the « direction to be locally the line of steepest descent of the
potential, then

ax
Se
uy
sae

Boj ....0.- +p 10a (6)

where o, the specific conductance or conductivity, is related to the conductance
LE by

_ Lye

=
When dealing with local unsteady state flow the capacity of the path as a whole

may be replaced by the concept of the specific capacity (y) or capacity per unit
volume

Both o and y are properties of the medium constituting the conducting
path. The flow, while in the unsteady state, proceeds by a diffusive mechanism

with a diffusivity De (Bosworth, 1949). This diffusivity has the physical

p
dimension of - for every type of transport process. On the other hand, the

physical dimensions of conductivities (and specific capacities) in general differ
with each different transport process.

(VI) THE ENTROPY SOURCE.

Whenever a flow process takes place work is dissipated or entropy generated.
For any single transport process, the rate of generation of entropy is equal to
the product of the flux by the potential, namely

The local entropy generated per unit volume per unit time in a large con-
ducting path is likewise given by the product 7x #H. Entropy is generated
only by the flowing in conductances. Entropy is not generated by the flow of
charge into capacities. Such a process is reversible while the flow in con-
ductances is not.

(VII) Cross POTENTIAL DIFFERENCES. THE ONSAGER RELATIONSHIPS.

The single transport process discussed above is comparatively rare.
Frequently a single potential difference gives rise to two or more fluxes. Thus

PRESIDENTIAL ADDRESS. 9

a temperature difference in a circuit involving two different metals gives rise
not only to a heat current but also to an electric current—the so-called thermo-
electric effect. In a mixture of two different chemical substances, a temperature
can give rise both to a heat flow and to the differential migration of one of the
chemical constituents in the phenomena of thermal diffusion. When coupled
transport processes of this nature occur we must modify equation (3) to the
form

SOS Cel Mek Oe (9)
3 |

where the current now depends on a whole number of different potentials and
the potential conversely depends on a whole number of different currents.
The phenemological coefficients L,, with 14k now represent the effect of cross
potential differences such as the thermoelectric and Peltier coefficients, electro-
osmosis and the streaming potential and so on. The fundamental equation to
these phenemological constants for cross fluxes is given by the Onsager reciprocal
relations (Onsager, 1931; de Groot, 1951) which state that, for all 7’s and k’s

Tig Totnes cients ate ty, (10)

provided the fluxes and potentials have been defined in a proper manner. A
proper choice of these quantities demands that the fluxes should be the time

derivative of the state parameters A,,....A,, and that the entropy change
produced in the system by a departure of these state parameters by amounts
Q;,-..-Q,, ete., from their equilibrium value should be a quadratic expression
of these quantities, namely
AS=34 = Ce ONO Ochs sei gs Pray Shaus 2 (11)
Jj
Since the @,’s commute, the quantities J;, also may be taken as symmetric
Dh pad [ee COR OOo Le rcm ute ONC) Oath Datla co Con (12)
Since the potentials X, are defined by
o( AS)
Sa setts ae Gian, SAIN MEE TT Shieh Geuze apa ¥en eli i
saa)
It follows that
A — Xd .Q, ot Sis ne ke\es Is hetelerte eis) 6) 6 6 leeks. (14)

and that the J;,’s are the capacity factors.

The Onsager reciprocal relationship may be proved on the basis of the
principle of detailed balancing, and provides a basis for the theoretical treatment
of all examples of coupled transport processes.

Particular examples occur in electric and thermal conduction in anisotropic
media. In such media the resultant flux density is not necessarily co-directional
with the applied potential field. Ina field with components LZ, H,, FE. the resultant
flows are j,,j,,j- given by the equation

B.+k,,B, +h,

J "2, -h Hk, 2,
jh, +k, EL, +k,E

ZU 2 Zommiee

The k,,’s are subject to the Onsager reciprocal relationship
Kiyy=Kyxy ete.

ay

jah

10 R. C. L. BOSWORTH.

This particular symmetry with regard to the thermal and electric conductivity
tensors, even when the crystal order did not demand such symmetry, has been
known for a long time.

A limitation on the possible degrees of coupling between different transport
processes is provided by the principle referred to as Curie’s law (Curie, 1908),
which states that coupling is only possible between transport processes of the
same rank. In other words, coupling is not possible between scalar chemical
reaction and vector mass diffusion or between vector heat conduction and tensor
momentum flow.

(Vill) INDUCTIVE PHENOMENA.

In certain types of coupled transport processes it is possible to trace a
cyclic relationship between the flow of the properties concerned. This condition
arises when the second p.d. in the circuit is produced as a result of the first
current and conversely the first p.d. is produced by the second current. Thus
in an appropriate circuit a magneto-motive force may be set up by an electric
current and an electromotive force as the result of a (displacement) magnetic
current. Again, when a fluid flows in a pipe, the hydraulic current gives rise
to velocity differences across any one sectional area. These velocity differences
constitute potential differences for the transport of momentum and thus tend
to drive a momentum current across these sectional areas. The momentum
flow in turn constitutes a stress and as such provides the potential for an
hydraulic flow. Other examples of this type of cyclic coupling could be given.

Pairs of transport processes coupled in such a way that the X, depends on
the J, and the X, on the J, have some interesting properties concerned with the
transient state. Any sudden increase in X, gives rise to transient processes
resulting in a storage of charge in the capacities associated with circuit 1. The
effect of these capacities alone results in the change in flux J, lagging behind
the change in potential X,. On the other hand, change in J, results in an
instantaneous change in X, which leads to storage of charge 2 in the capacities
associated with circuit 2. The action of these capacities alone results in the
change of flux J, and thus the potential X, lagging behind the change in J.
The capacities for circuit 2 thus act as the inductances for circuit 1 (and vice
versa). Again, any increase in the resistance of circuit 2 acts as an increase in
the conductance of circuit 1 (and again vice versa). If, however, one of the
coupled transport processes is of a pure displacement type, this process will
make no contribution to the steady state conductance of the other process, but,
by adding inductances to the capacities may profoundly effect the transient
phenomena.

Inductive phenomena are, of course, best known in electrical and mechanical
transport systems but other coupled transport processes also give rise to these
phenomena. Examples connected with heat flow by natural convection have
been given by the author (Bosworth, 1946, 1948).

(IX) TREATMENT OF COUPLED TRANSPORT PROCESSES.

Some of the most complicated examples of coupled transport phenomena
occur in the field of chemical industry. Even comparatively simple chemical
engineering operations involve consideration of several transport processes.
In the process of fractional distillation, for example, there occurs, at each stage,
an exchange of two or more chemical components as well as the associated latent
heats between the two interacting streams. At the same time the two streams,
one liquid and one vapour, are flowing in opposite directions under driving
forces ultimately provided by the overall temperature difference across the

PRESIDENTIAL ADDRESS. 11

still. An interesting discussion of the performance of a fractionating column,
from the point of view of thermostatic equilibrium rather than in terms of the
irreversible transport processes, has been given by Rossini (1950).

A continuous chemical reaction provides a still more complicated example of
coupled and interlocked transport processes. In any typical reactor in which
reagents flow into the reaction zone under prescribed physical conditions,
react there with resultant change in heat content and volume, and then flow out
- again, there occurs simultaneously the following transport processes :

(a) Flow of heat.

(6) Flow of momentum.

(c) Flow of each of the chemical reagents and resultants.

(d) Chemical reaction or reactions. And possibly there may occur
(e) Relaxational phenomena.

Progress of each flow process depends not only on its own phenemological
constants and storage capacities, but also on the degree of coupling with other
transport processes. In each process too there may be several carriers operative.

Attempts at a theoretical treatment of these coupled processes have been
made in several different manners :

(a) By using the thermodynamics of irreversible processes.
(b) By dimensional analysis.
(c) By using model methods.

A comprehensive discussion of the application of the first method has
been given by de Groot (1951). The second (the method of dimensions) has
been discussed by a number of authors. An early paper by Greenewalt (1926)
discussed application of dimensional methods to the problem of the absorption
of water by sulphuric acid. The application of the same method to a continuous
chemical reactor has been discussed by Damkohler (1936), by Edgeworth-
Johnstone (1939) and by Bosworth (1946), all of whom were concerned with
laying down conditions of similarity between reactors of different sizes. The
conditions, however, are so stringent that the method is only applicable in
certain special instances.

The application of model methods to heat and mass flow problems has
become a tool of power and importance in the hands of Paschkis and his co-
workers. <A similar comprehensive attempt to apply such methods to all the
transport processes occurring in a chemical reaction has not been reported,
although Wilhelm, Johnson, Wynhoop and Collier (1948) have applied this
method to the transport processes occurring in fixed-bed catalyst converters.

(X) GRAND TRANSPORT PROCESSES.

The concept of an irreversible transport process or flow occurring under the
influence of a potential or driving force at a rate dependent upon conductance
terms when in the steady state, and on combined conductance and capacity
terms when in the transient state, is by no means restricted to phenomena on the
molecular scale. On the contrary, many industrial and other operations
involving the irreversible generation of entropy may be regarded as examples
of transport processes. The distribution of steam from boiler house to the
various thermal sinks of a factory is a typical example. Because, however,
such a process, while involving an overall transport under a driving force, may
also be resolved into a series of microscopic and molecular transport processes—

12 R. C. L. BOSWORTH.

heat flow, chemical diffusion, etc., it is perhaps appropriate to refer to the
aggregate process as a Grand Transport Process. The operation of the fractiona-
tion column considered in the section above constitutes another example of
such a grand transport process.

(XI) FEEDBACK.

A type of flow circuit which is involved in many different forms of engineering
equipment causes the main stream of the grand transport process to divide at
some point or points in the conducting path. Part of the stream then goes on
and another part is fed back to an earlier stage in the system. Circuits with
feedback of this nature are particularly well known in radio physics, but the
general effect of similar feedback in hydraulic and chemical systems is not at
all well understood and would appear to offer a particularly apt subject for
study by the method of models mentioned above. A heat exchanger is a
particularly simple example of an engineering equipment based on a transport
process with feedback. The performance of a factory as a whole, considered -
as a flow from sources (raw materials) to sinks (scrap heaps, sewers, chimney
stacks and sales departments), could also usually be represented as that of a
grand transport process with multiple feedback. The influence of this feedback
on the time of the transient period—or the time taken to settle down after a
break—is frequently sensed by factory personnel but has never received the
detailed study it undoubtedly merits.

(XIT) EFFICIENCY.

In the study of thermodynamics we are familiar with the concept of efficiency
of operation as an inverse measure of the useful energy wasted in carrying out
the operation; and also with the fact that certain classes of operation are
inherently more efficient than others. Such a concept has yet to develop with
respect to industrial chemical operations. The efficiency, just like the efficiency
of a prime mover, could be judged on the balance of the entropy of the raw
material coming in through the various sources—including fuel, power and
services—and on the finished materials going out through all sinks. The
difference is a measure of the entropy generated when the irreversible process
constituted by the factory operation takes place and is minimized when that
factory reaction is carried out in the most efficient manner. As far as I know,
no attempt to carry out such a balance has yet been made. A very rough
preliminary survey of a particularly simple industrial chemical operation has,
however, indicated a very low overall efficiency.

The present industrial index of efficiency is cost. A process is judged good
if of low cost, and there is probably some truth in such a judgment, but the
correlation between cost and thermodynamical efficiency is by no means perfect.
Examples could be given of operations judged desirable on a cost basis which to
future generations appeared disastrous. Witness, for example, the profligate
waste of potential organic chemicals in coal, and, as a simpler example, the
direct-acting steam-reciprocating pump, which while praised by old hands for
its reliability, has a thermodynamical efficiency of 2° or under and is a source
of profligate waste of power.

Again, analysis on the cost basis gives little real indication as to what benefits,
if any, may be expected from more concentrated industrial research, whereas
analysis of the factory operations into steps with the efficiency shown for each
would indicate at once where intensive research would not likely result in
improved performance.

I do not think I am being fanciful here. Analysis of prime movers thermo-
dynamically has provided just such information and has resulted in a steady

PRESIDENTIAL ADDRESS. 13

improvement in efficiency. Further, while the science of thermodynamics was
developed over 100 years ago from a fundamental study of the performance of
heat engines it has, particularly in the last half century, increasingly been
applied to chemical phenomena and now provides the theoretical basis of chemical
energetics. The concerted use of thermodynamical principles to the
improvement of the efficiency of thermally operated equipment is much more
recent. A similar application to improve the efficiency of industrial chemical
operations has yet to begin. Most of the fundamental information necessary
for finding, for example, the relative rates of entropy generation in the different
steps of the grand transport process constituting a factory operation, or for
choosing the most efficient (7.e. the least wasteful) way of preparing a given
product, are now available. Perhaps some day in the not too distant future
we shall hear that sufficient data has accumulated to enable these applications
to be made.

REFERENCES.

Avrami, M., and Paschkis, V., 1941. Bull. Amer. Phys. Soc., 16, 15.

Born, M., and Green, H. 8., 1949. ‘‘ A General Kinetic Theory of Liquids.”’ C.U.P.

Bosworth, R. C. L., 1946. Nature, 158, 309.

—_—_—_———. 1947. Tris JourNAL, 81, 15-23.

—_—_—_________—— 1948. Nature, 161, 166-7.

——_—_—_—__—_—_——— 1949. Jour. and Proc. Aust. Chem. Inst., 460-82.

—_—— 1950. ‘* Physics in Chemical Industry.’? Macmillan.

——__-+—__—_—__——. 1950a. Tuts JourNaAt, 83, 8-30, 124-33.

—__—_— 1951. Tris JOURNAL, 84, 53-8. ;

Chapman, 8., and Cowling, T. G., 1939. ‘‘ The Mathematical Theory of Non-Uniform Gases.”’
@.ULP.

Cox, R. T., 1950. Reviews of Mod. Phys., 32, 238-48.

Curie, P., 1908. ‘‘ Ghuvres.’’ Gauthier-Villars, Paris, 127.

Damkohler, G., 1936. Z. Electrochem. 42, 846-62.

Edgeworth-Johnstone, R., 1939. Trans. Inst. Chem. Engrs., 17, 129-36.

Hisenschitz, R., 1950. Nature, 167, 216-20.

Groot, 8S. R. de, 1951. ‘* Thermodynamics of Irreversible Processes.” North Holland Pub-
hshing Co., Amsterdam.

Greenwalt, C. H., 1926. Ind. Eng. Chem., 18, 1291-95.

Knudsen, M., 1934. ‘‘ The Kinetic Theory of Gases.” Methuen, London.

Meier, R. L., 1951. Research, 4, 463-70.

Onsager, L., 1931. Phys. Rev., 37, 405-26; 38, 2265-79.

Paschkis, V., and Baker, H. D., 1942. Trans. Amer. Soc. Mech. Engrs., 64, 105-11.

Paschkis, V., and Heisler, H. P., 1946. J. of Appl. Phys., 17, 246-54.

Prigogine, I., 1949. Physica, 15, 272-84.

Rossini, F. D., 1950. ‘‘ Chemical Thermodynamics ”’, 458. Wiley, New York.

Wilhelm, R. H., Johnson, W. C., Wynkoop, R., and Collier, D. W., 1948. Chem. Eng. Progress,
44, 105-16.

A GEOLOGICAL ACCOUNT OF HEARD ISLAND.

By A. JAMES LAMBETH, B.Sc.
With Plate I and one Text-figure.

Manuscript received, December 11, 1951. Read, April 2, 1952.

CONTENTS.

Page

I. Introduction ae we bl oe ule ies oe

Il. Stratigraphy—

(a) Laurens Peninsula Limestones ay we 2S ify ae

(b) Drygalski Agglomerates av ae rue dist he

(c) Lavas— 16

(i) The Mt. Olsen Lavas Ap bie oe a

(ii) The Coast Lavas .. “0 Ae bho) cia

(ii) The High Lavas .. ae 4h heehee

KII. Tectonics and Structure iN a
IV. Correlation with the Kerguelen Archipelago whe oh 24
V. Summary.. ah Romie
Acknowledgments e 5 aM ot me oink
References as ae iy ey a ig

I. INTRODUCTION.

A general account of the geographical features of Heard Island together
with notes on the glaciology has already been given (Lambeth, 1951).

Prior to 1947 only four scientific expeditions had called at Heard Island,
and little is recorded of the stratigraphical and tectonic features. The observa-
tions contained herein were made by the first party of the Australian National
Antarctic Research Expedition during 1947-49.

Heard Island is glaciated throughout the year and because of the sey
and discontinuity of outcrops much reliance has been placed on collections
made from moraines.

As far as is known, three formations exist, as follows :
Uppermost : Lavas.
Intermediate : Drygalski Agglomerates.

Basal: Laurens Peninsula Limestones.

II. STRATIGRAPHY.
(a) Laurens Peninsula Limestones.

The lowest formation, here called Laurens Peninsula Limestones, outcrops
on both the south and north-east coasts of Laurens Peninsula, in latitude
53° O01’ S., longitude 73° 20’ E. The existence of this underlying pelagic lime-
stone was suspected in 1908, but it is believed that the outcrops reported above
are the first observed in situ.

The limestones are thinly bedded and intercalated with thin soft tuffaceous
shales, and are folded about an E.-W. magnetic axis with north and south dips
varying between 25° and 35°. The colour varies from white or grey to blue or

A GEOLOGICAL ACCOUNT OF HEARD ISLAND. 15

brown, the texture being even and fine-grained with conchoidal fracture. There
are abundant foraminifera which indicate a Paleogene age.*

The upper surface of the formation is plane and sub-horizontal, the greatest
elevation being approximately 250 ft. above sea level on the south coast of
Laurens Peninsula. From here it dips gently southwards to below sea level.
The attitude indicates that the formation is not very far below sea level through-
out the north-eastern part of the island.

On the south coast of Laurens Peninsula the formation has been intruded,
prior to folding, by concordant sills of fine-grained, non-porphyritic, holo-
crystalline trachybasalt, varying in thickness from a few inches to five feet.
These are the only igneous rocks observed.

GENERALISED GEOLOGICAL SKETCH MAD

WHICH THE ICE COVER IS ASSUMED ABSENT
Ps LEGEND
O Red Island |
fad Alluvium (7) laurens Peninsula Linestanes

= / Olsen lavas (A) Zur
pa pa ZZ) Cost Lavas Scoriae Cone

Rernehapeethcgs orc _ |
X cate Sips Corinth Head PS aba 53°0
"Corinthian Bay BAA yt , ape Bidlingmaiar High lavas P3- Hoe-Hoe Structure

(TIM) 22447 Agglomersies Aa-Aa Structure

Rovad Hil

Cape Lockyer

Cape Lambeth

HEARD ISLAND

MAGNETIC DECLINATION 48°40’ W 1947

(b) Drygalski Agglomerates.

These were first reported by Phillipi from the flanks of Mt. Drygalski near
Atlas Cove, and as the widespread nature was not realized the occurrence was
described as a crater ruin. They occur over most of Laurens Peninsula and the
east and south coasts of the island. Their threefold nature is apparent at Mt.
Drygalski and the southern and north-eastern parts of Laurens Peninsula, which
may be taken as the type area. (Mt. Drygalski, lat. 53° 02’S., long. 73° 23’ E.)

The agglomerates are sub-horizontal in attitude, overlying the Laurens
Peninsula Limestones with an angular unconformity of approximately 35°, the
maximum thickness on the Laurens Peninsula being 1100-1200 ft.

Their lower division consists largely of agglomerates, but near the top of the
division local thinly-bedded tuffaceous shales are developed. Volcanic bombs
are recognizable, while the variable-sized, angular to sub-angular, and rounded
agglomerate pebbles are mainly porphyritic olivine-basalt, limburgite and
limburgitic scoriez. The bond is tuffaceous, shaly, palagonite and occasionally

* Glaessner, M. F., personal communication, 8th September, 1950, Globigerina sp. and
Gumbelina sp. were determined.

16 A. JAMES LAMBETH.

calcareous. Much of the material was deposited under water. Locally there ~
is little sorting, but broadly there are significant differences in grainsize, the
grainsize diminishing towards the top. The maximum thickness of this division
on the Laurens Peninsula is about 350 ft.

The igneous rocks are of three types: minor gabbroic stocks and bosses,
contemporaneous trachytic necks, and the feeder dykes of the middle division.

The middle division overlies the lowest on an approximately even erosional
surface, some of the trachytic necks being truncated. It is entirely igneous and,
unlike the other divisions, is not continuous but is a series of volcanic outpourings
within the Drygalski Agglomerate. Thin flows emanating from small dykes
have produced local thicknesses of up to 300 feet. Columnar structure is
common, the rock types being olivine-basalts and feldspar-basalts.

The upper division, thickness approximately 400 ft., shows a return to
the agglomeratic facies of the lowest, with a finer grainsize in the agglomeratic
particles, whilst tuffaceous shales and grits are more common than elsewhere.
Their presence assists in the differentiation of the lower and upper divisions,
where the middle is not developed. Contemporaneous brecciated plugs are
the only igneous rocks.

The upper surface of the Drygalski Agglomerates is roughly plane, con-
forming to the general attitude of the formation.

(c) Lavas.

These are known mainly from moraines, although those near sea level can
be examined in detail. Geographically, the lavas fall into three groups ; these,
however, are not necessarily in sequence and may be contemporaneous.

(i) The Mt. Olsen Lavas: Situated on the heights of Laurens Peninsula.
(ii) The Coast Lavas: Parasitic cones adjacent to sea level.
(iii) The High Lavas: Situated on the main mass of the island.

(i) The Mt. Olsen Lavas. These are situated on the heights of Laurens
Peninsula about Mt. Olsen and Mt. Anzac, and are mostly glaciated (Mt. Olsen,
lat. 53° 01'S., long. 73° 20’ E.). They overlie the Drygalski Agglomerates with
disconformity, and stratigraphically are in a situation similar to the High Lavas,
although they may not be contemporaneous.

The basal beds are trachyandesites, approximately 100 ft. maximum
thickness, with traces of columnar structure, overlain by an accumulation of
trachyte, up to 1300 ft. thick, which forms the various peaks. Moraine material
does not suggest that any other rock type is present.

(ii) The Coast Lavas. These are diverse. The limburgite of Rogers Head-
land is noteworthy. This is a crater floor-relic, remnants of the sides occurring
in the limburgitic tuff relics of Rogers Head, portion of Corinth Head and
Church Rock. This cavernous lava issued from many centres in the floor and
is similar to occurrences at Saddle Point, Cape Bidlingmaier, Scarlet Hill, Red
Island and Mt. Macey. At these places, Mt. Macey and Cape Bidlingmaier
excepted, the containing walls are absent. They all show pa-hoe-hoe structure,
and columnar structure on a small scale, while tumuli are common. The
conspicuous and almost linear scoria-cones developed at these places with their
abundant lapilli ejecta, are of an earlier date than the surrounding limburgites.

The lavas on the northern and western coasts of Laurens Peninsula centre
about Mt. Dixon. This mound-like structure is completely glaciated and
outcrops can be seen only at the base. They are trachyte, overlain by massive
basalts, followed by vesicular aa-aa basalts. The accumulation of cinders
probably represents the final outburst.

A GEOLOGICAL ACCOUNT OF HEARD ISLAND. iM

The Cave Bay trachytes with the overlying scoriaceous lavas appear to be
allied to this Mt. Dixon suite, both forming part of a former widespread occurrence
extending to the islands (e.g. Pulpit Rock) lying off Cape Gazert, where the
aa-aa type overlies a massive basalt.

(iii) The High Lavas. The high lavas appear to extend upwards from the
top of the Drygalski Agglomerate to the culminating peak Mt. Mawson. Conse-
quently they represent a piling up of nearly 8000 ft. of volcanic material. They
are known almost entirely from moraines, as those few outcrops which do exist
are either difficult of access or unapproachable.

The lavas appear to have built up Big Ben Range by emanating from vents
situated about the centre of the island. Seen from the air, the plateau-like
upper surface of this mass resembles an infilled crater, in which case Mt. Mawson
is a@ cone-in-cone structure, a feature commonly developed in the vents of the
Laurens Peninsula coast lavas. The great height of this mountain mass
developed in the narrow island area suggests lavas of high viscosity. In the
samples collected limburgites, olivine-augite-basalts and trachybasalts pre-
dominate, with some subordinate plagioclase-basalts and trachytes, whilst
more coarsely grained olivine-augite-types probably represent local intrusions
into the lavas.

III. TECTONICS AND STRUCTURE.

Three distinct formations are superimposed in a simple structure. The
elevation and folding of the Laurens Peninsula Limestones indicates movements
of great magnitude and, since these sediments are Paleogene in age, the
movement may have been contemporaneous with the Alpine of the Northern
Hemisphere. Only minor trachybasalt sills occur.

After a period of erosion, widespread explosive volcanic activity occurred
from many centres, the ejecta being mostly limburgite and basalt, but some
trachyte necks were formed. ‘These formed the lowest division of the Drygalski
Agglomerate. A hiatus followed during which thin fissure basalts formed
discontinuous local accumulations. Explosive vulcanism followed, the average
grainsize of the ejecta being finer than previously.

So far there was no sign of glaciation, it being inferred that the glacial
epochs of the Pleistocene had not yet intervened. Consequently the Drygalski
Agglomerates are probably late Tertiary, when this formation was undoubtedly
of much greater extent than at present.

_ During the break in deposition which followed, the upper surface of the
Drygalski Agglomerates was eroded to a roughly plane surface..

Igneous activity on a grand scale then commenced, localized about Big
Ben. Other centres probably existed and may be represented by the various
neighbouring islands. Lavas of high viscosity rapidly built up the mass of
Big Ben. The main island fault may have had its beginnings here, providing
a fissure through which the lavas were extruded. Most of the vulcanism appears
to have finished before the Pleistocene glaciation, but its recurring and
diminishing nature is clearly indicated in the numerous cone-in-cone structures.
Some of the more recent flows overwhelmed the eroded edges of the Drygalski
Agglomerates, indicating a general erosion of this formation. Fumaroles and
hot springs do not occur on Heard Island.

The main island fault appears to have been most active after glaciation.
It has truncated and destroyed the old trunk glacier flowing down Atlas Cove,
as well as the headlands of the Jacka Glacier. The downthrow side was on the
south-west and the throw could not be determined. Evidence of its existence
igs the non-occurrence of the Drygalski Agglomerates on the downthrow side

18 A. JAMES LAMBETH.

either in situ or in moraines, as well as the truncation of the glaciers. It is
represented physiographically by the escarpments of South Barrier and North
West Cornice, which appear to represent the eroded scarps, at the latter place
slickensided pebbles being found. The movement of the fault was associated
in the Laurens area with trachytic vulcanism followed by basalts of increasing
viscosity, culminating in aa-aa lavas, ashes and cinders.

The east coast may have been influenced by faulting. The almost linear
arrangement of centres of eruption of limburgitic tuff, limburgite with pa-hoe-hoe
structure, and scoriz, extending from Mt. Macey to Scarlet Hill is significant.
The tuff of Rogers Head contains fragments which, judging from their in situ
position in the Drygalski Agglomerate and the attitude of this formation, could
not have been brought from below if this area were not downfaulted. No other
evidence was apparent, but the existence of such a fault would account for the
absence of the eastern wall of the old Atlas Cove Glacier, which would have
been downfaulted. The island appears to have had its origin in a tectogene
in the early Tertiary, and is now a horst with downthrows to the north-east and
south-west.

IV. CORRELATION WITH THE KERGUELEN ARCHIPELAGO.

A threefold division of the strata on the Kerguelen Archipelago has been
indicated by Mawson (1933). A basal series of early Tertiary age is overlain
by fluviatile conglomerates, which are in turn overlain by a great thickness of
voleanics, the ash beds of which contain molluscan fauna of Pliocene age.

The centrally situated conglomerates are of basalts and trachytes, in which
tuffceous beds are common. Contemporaneous trachyte plugs occur as well as
scorie and agglomerate. Thin intercallated lignites contain Araucarian flora
indicating a late Oligocene age (Mawson, 1933; Crié, 1869; de la Rue, 1929).
The section displayed in the ravine of the Port Jeanne D’Arc rivulet through
the heights dominating the old whaling station there is the best known, and
must for the time being be considered the type. (Port Jeanne D’Are, lat.
49° 54S, long. 69° 517 ,)

The presence of globigerina limestone in the basement rocks has been
indicated by Roth (1875) and Mawson, and detrital material was found in the
central conglomerates. Consequently both islands stand on pelagic sediments.
While the Port Jeanne D’Arec conglomerates are not directly comparable with
the Drygalski Agglomerates, the two formations appear to be coeval. It is
clear that explosive volcanic activity was happening in both places, but whilst
it was dominant at Heard Island, it was of a lesser degree at Kerguelen Land
and the products are mingled with erosion débris. As these places are approxi-
mately two hundred miles apart topographical differences at the time are to be
expected.

Other points of correlation may be revealed by petrological methods.

V. SUMMARY.

Three formations were recognized on Heard Island. The basal pelagic
limestones of early Tertiary age are unconformably overlain by agglomerates,
which in turn are overlain by an immense thickness of lavas. The area has
characteristics similar to those of a tectogene. Faulting has occurred and the
island now appears to stand as a horst.

ACKNOWLEDGEMENTS.

The writer wishes to thank Mr. George 8S. Compton, of Kalgoorlie, for
unfailing help throughout the duration of the expedition, and to whom the
observations in the vicinity of Round Hill are due; and E. O’Driscoll, B.E., of
Sydney, for many helpful discussions.

Journal Royal Society of N.S.W., Vol. LDXXXVI, 1952, Plate I

A GEOLOGICAL ACCOUNT OF HEARD ISLAND. 19

REFERENCES.

Crié, L., 1869. ‘“* Die fossile Flora des Kerguelen Archipels.”’ Beitrage der fossilen Flora einiger
Inseln des sudpacifischen und indischen Oceans, Jena.

Lambeth, A. J., 1951. ‘‘ Heard Island, Geography and Glaciology.”’ THis JouRNAL, 84, 92
(with bibliography).

Mawson, D., 1933. ‘“‘ The Geology and Glaciation of Some Islands of the Southern Ocean and
the Newly Discovered Antarctic Mainland.” Quart. Jour. Geol. Soc. London, 89, cxili-cxv.

Phillipi, E., 1908. ‘‘ Geologie der Herd Insel.” Deutsche Sudpolar Exped. 1901-3, Bd. 11,
Heft. 3, Geogr. u Geol., 241-50.

Roth, 1875. ‘‘ Uber die gesteine von Kerguelen’s Land.’ Monatsberichte der Konig. Preuss.
Akademie der Wissenschaften zu Berlin.

EXPLANATION OF PLATE I.

Fig. 1.—Laurens Peninsula seen from Mt. Drygalski. West Bay on the left, Atlas Cove on
the right with the moraine-strewn plain of Atlas Cove. Top centre is Mt. Olsen (2080 ft.). The
glaciated heights about Mt. Olsen are composed of Mt. Olsen Lavas, the cliffs below are of Drygalski
Agglomerate. The Laurens Peninsula Limestones outcrop at sea level on West Bay below the
cliffs. Left distance, the slopes of Mt. Dixon composed of Coast Lavas.

Fig. 2.—The culminating point, Mt. Mawson (9005 ft.), surmounting the plateau-like top of
Big Ben Range (approx. 8000 ft.) suggesting a cone-in-cone structure. The rib-like outcrops are
of High Lavas. Air photo from the south.

These photographs are reproduced by kind permission of the Department of External
Affairs, Antarctic Division.

OCCULTATIONS OBSERVED AT SYDNEY OBSERVATORY
DURING 1951.

By W. H. ROBERTSON, B.Sc.
and K. P. SIMS, B.Sc.

Manuseript received, February 8, 1952. Read, April 2, 1952.

The following observations of occultations were made at Sydney Observatory
with the 114-inch telescope. A tapping key was used to record the times on a
chronograph. The reduction elements were computed by the method given
inthe Occultation Supplement to the Nautical Almanac for 1938 and the reduction
completed by the method given there. The necessary data were taken from
the Nautical Almanac for 1951, the Moon’s right ascension and declination
(hourly table) and parallax (semi-diurnal table) being interpolated therefrom.
No correction was applied to the observed times for personal effect but a correc-
tion of —0-00076 hour was applied before entering the ephemeris of the Moon.
This corresponds to a correction of —1”-5 to the Moon’s mean longitude.

Table I gives the observational material. The serial numbers follow on
from those of the previous report (Robertson, 1951). The observers were
H. W. Wood (W), W. H. Robertson (R) and K. P. Sims (S). In all cases the
phase observed was disappearance at the dark limb. ‘Table II gives the results

TABLE [.

Serial N.Z.C.

No. No. Mag Date. Ut: Observer.

h m) “4s

216 855 6:8 Feb. 16 11 23 26-8 W
217 -—— 7:4 Feb. 16 11 24 38-6 W
218 864 6-7 Feb. 16 12 42 28-6 W
219 1018 5:5 Feb. 17 12 47 08:0 WwW
220 = 7:9 May 16 14 17 59-9 W
221 1676 6:7 May 16 14 22 40-0 W
222 1888 6:2 May 18 13 03 36:7 W
223 1907 6°7 July 12 9 45 01-6 W
224 1913 7:1 July 12 11 32 52-2 W
225 2039 5:6 July 13 13 49 51-4 W
226 2157 6-1 July 14 9 49 03-1 W
227 2470 6-1 July 16 7 55 09-7 R
228 2505 5:4 July 16 13 30 37-5 W
229 2108 6:4 Aug. 10 9 08 37-1 R
230 2268 4°8 Aug. Ill 13 04 04-9 W
231 2273 5:9 Aug. ll 13) 30 lee W
232 2536 7:4 Sept. 9 10 30 20:4 WwW
233 2545 6:4 Sept. 9 12 55 31-1 W
234 2554 5:4 Sept. 9 14 23 35:8 W
235 2723 6°7 Sept. 10 11 19 24-6 S
236 2735 7-2 Sept. 10 14 03 51-4 W
237 2740 6:3 Sept. 10 14 46 27-8 W
238 2907 6°3 Sept. 11 14 06 04-9 W
239 2852 7:4 Oct. 8 10 24 08-2 R
240 2864 4-7 Oct. 8 13 05 26-4 W
241 3416 5:6 Oct. 12 13 36 04-1 W
242 2804 5:9 Nov. 4 9 23 08-7 W
243 68 5°7 Nov. 10 9 53 12-8 W

OCCULTATIONS OBSERVED AT SYDNEY OBSERVATORY DURING 1951. 21

TABLE II.
Coefficient of

Serial | Luna-

No. tion. p q Pp? Pq q? Ao | pAo | qAo

Aa AS

216 348 + 88 | +48 77 +42 23 |—1-1 |—1-0 |—0-°5 | +11-3 | +0-50
217 348 + 86 | +51 74 +44 26 |—1-8 |—1-5 |—0:9 | +11-1 | +0-53
218 348 + 89 | —46 79 —4] 21 |j|—1-0 |—0-9 |+0°5 | +11-7 | —0-44
219 348 + 83 | +56 69 +46 31 |—1-4 |—1-2 |—0-8 | +11-5 | +0-48
220 351 +100 | — 3 | 100 — 3 0 |—0-2 |—0-2 0-0 | +13-0 | —0:50
221 351 + 98 | +20 96 +20 4 |—0-3 |—0:3 |—0-1]! +14:4 | —0-29
222 351 + 79 | —62 62 —49 38 |—0O-2 |—0-2 |+0-1 | + 6-3 | —0-90
223 353 + 96 | +29 92 +28 8 |—0-5 |—0-5 |—0-1 | +14-5 | —0-16
224. 353 + 98 | —22 95 —22 5 |—0-7 |—0-7 |+0°-2 | +11-4 | —0-63
225 353 + 49 | —87 24 —43 76 |+2-2 |/4+1-1 |—1:9 | + 1-8 | —0-99
226 353 + 53; +85 28 +45 2, 0-0 0-0 0-0 | +10-7 | +0:64
224 353 +100 | + 3 100 + 3 0 |—1-2 |—1-2 0-0 |; +13-3 | —0-05
228 353 + 99 | —13 98 —13 2 |—0-8 |—0-8 |+0:-1 | +13-0 | —0-18
229 354 ee 2a —-96 7 —26 93 |+2:-5 |+0-7 |—2-4 | — 0-8 | —1-00
230 354 + 98 | +19 96 +19 4 |—0:-5 |—0-5 |—0-1 | +13-6 | —0-03
231 354 + 83 | +56 69 +46 31 |—1-6 |—1:-3 |—0-9 | +12:7 | +0-36
232 355 + 94 | —34 88 —32 12 |—2-0 |—1-9 |+0-7 | +12-3 | —0:35
2a0 355 + 31 | +95 10 +29 90 |—0-6 |—0-2 |—0-6 | + 4-1 | +0-95
234 355 + 79 | +61 63 +48 37 |—1-0 |—0-8 |—0-6 | +10°5 | +0-61
235 355 + 96 | —28 92 —27 8 |—l-1 |—1-1 |+0-3 | +13-1 | —0-16
236 355 + 24) +97 6 +23 94 |—0-8 |—0-2 j—0-°8 | + 1:5 | +0-99
237 355 + 36 | +93 13 +34 87 |—1-0 |—0-4 |—0-9 | + 2-9 | +0-97
238 355 + 64 | +77 4] +49 59 |—1-6 |—1-0 |—1-2 | + 5-7 | +0-91
239 356 + 61 | —79 od —48 63 |+1-2 |+0-7 |—0-9 | +10°3 | —0-65
240 356 + 95 | +31 90 +29 10 |—0:7 |—0-7 |—0-2 | +11-7 | +0:51
241 356 + 73 | —69 53 —50 47 |+0-°5 |+0-4 |—0°3 | +14-5 | —0:-27
242 357 + 87 | +50 75 +43 25 |—1-1 |—1-0 |—0-6 | +10-3 | +0:65
243 357 + 96 | +29 92 +28 8 |—2-0 |—1:-9 |—0-6/} +10°6 | +0-70

of the reductions which were carried out in duplicate. The N.Z.C. numbers
given are those of the Catalog of 3539 Zodiacal Stars for the Equinox 1950-0
(Robertson, 1940), as recorded in the Nautical Almanac.

The stars involved in occultations 217 and 220 were not included in the
Nautical Almanac list. Their apparent places were as follows :

No. Star. R.A. Dec. Catalogue.

217 G.C. 7088 5h 38m 158-34 +29° 28’ 25’°8 G.C.

220 B.D. +3°2518 11 30 53-57 + 2 43 52-1 Yale Vol. 20
REFERENCES.

Robertson, A. J., 1940. Astronomical Papers of the American Ephemeris, 10, Part II.
Robertson, W. H., 1951. TuHis JourRNAL, 85, 13; Sydney Observatory Papers, No. 17

AUSTRALASIAN MEDICAL PUBLISHING Company Liarep
Arundel and Seamer Streets, Glebe, N. Ss. Ww.
1952

OF NEW SOUTH WALES

1952

(INCORPORATED 1881)

~

Po

PART II
ae OF ?
VOL. LXXXVI |
"Containing Liversidge Lecture and Papers read in
July and August, 1952

EDITED BY

IDA A. BROWNE, D.Sc.

Honorary Editorial Secretary

Vile 2

H so WE
iia ae meaty

a —— Cmaves 1953

| LIBRARY.
: iS SYDNEY _ ;
PmLISHIED BY THE SOCIETY, 8

ne CONTENTS

cc ee VOLUME LXXxXxXVI_
. Aa Sale: | Part II | 3 F é.

“Arr. IV.—Climate and Maize Yields on the Atherton Tableland. By D. 38. )

ka and N. T. Drane .... ee oe oa ae t. oe >

“Arr. V.—Palladium Complexes. Part V. Reactions of Palladium Compounds with an
_ 2:2’ Dipyridyl. By S. E. Livingstone .. oe hice Sie oid 10 a ee

~ : pies PP as

_ Arr. VI.—Liversidge Lecture. Electron Diffraction. in the Chemistry of the Solid
; State. By A. L. G. Rees .. i Pi coe a: Sa are PS ae

Arr. VII.—Permian Spirifers from Tasmania. By Ida A. Brown

JOURNAL AND PROCEEDINGS

OF THE

me yYAL SOCIETY

OF NEW SOUTH WALES

FOR

1952

(INCORPORATED 1881)

VOLUME LXXXVI

Part II

EDITED BY

IDA A. BROWNE, D.Sc.

Honorary Editorial Secretary

THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE
STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN

SYDNEY
PUBLISHED BY THE SOCIETY, SCIENCE HOUSE
GLOUCESTER AND ESSEX STREETS

CLIMATE AND MAIZE YIELDS ON THE ATHERTON TABLELAND.

By D. 8S. SIMONETT, M-Sc.,
and N. T. DRANE, B.Ec.
Department of Geography and Faculty of Economics, University of Sydney.
With one Text-figure.

Manuscript recewed, May 13, 1952. Read, July 2, 1952.

The Atherton Tableland, situated in the coastal ranges of north Queensland
near Cairns, virtually lies in the cloud zone of the north-west monsoon
and the south-east trade winds for the greater part of the year. Due to this,
as much as to its height, it experiences an equable temperature and humidity
regime, very like that of the northern rivers of N.S.W., and its rainfall pattern
shows less of the rigid wet-dry division so characteristic of tropical Australia.
The wet season opens earlier with storms in November, rainfall intensities are
lower, the proportion of cloudy days significantly higher, and the end of the wet
Season is blurred by the ‘‘ drizzle ”’ season of April, May, and occasionally June,
the dry season being limited to July to October in the maize area centred on
Atherton. Add to the above a major wet season cyclone every four or five years,
and the combination is peculiar to the Atherton Tableland.

In land use the Tableland falls into three distinct divisions (see Skerman,
1947). In the hilly, high rainfall area south of Malanda (65”) there is no cultiva-
tion and dairying is based entirely on introduced pastures. Between Malanda
and the Barron River (c. 55”) is a rolling fragmented maize belt, the small
cultivated plots forming an integral part of a dairy-maize economy. North
and west of the Barron River is the major maize producing area, almost 85 per
cent. being grown on undulating land within a three-mile radius of Atherton
and the two small silo townships of Tolga, three miles north, and Kairi, six
miles north-east of Atherton. MRainfalls here are the lowest on the Tableland,
Atherton obtaining 54 inches per annum, Tolga and Kairi 48 inches.

RELATIONSHIP BETWEEN CLIMATE AND MAIZE YIELDS.

With relatively few exceptions, and these mainly in the intermediate
maize-dairy belt between Malanda and the Barron River, maize production on
the Tableland is a monoculture, neither animals nor subsidiary crops entering
the economy. No moisture conserving practices are employed, indeed they are
unnecessary, fertilizers are used only to an insignificant degree, the soils are
extremely uniform kraznosems on basalt, cultivation practices are sensibly
uniform throughout, and the varietal composition of the crop has not undergone
Serious changes during the last three decades.?

Night temperatures are never a limiting factor in growth, and both moisture
and temperature conditions are usually very favourable, and it is doubtful if
adverse effects from high temperatures and low moisture during tasseling have
been experienced in more than two seasons since 1900.

1W. R. Straughan, Manager, Atherton Tableland Maize Marketing Board, personal communi-
cation on the varietal composition of the maize crop.

CLIMATE AND MAIZE YIELDS ON THE ATHERTON TABLELAND. 23

With so many factors constant or inapplicable, since they are not limiting
in this area, two appear to outweigh all others in influencing ultimate yields,
namely the growing season rainfalls from December to February, including the
tasseling period, and the rainfalls of the cob-development stage during March
and April, when the maize is severely affected by ear-rots. These relationships,
together with the influence of time (hypothetical declining soil-fertility) form
the basis of the following study.

DATA.

Rainfalls. Monthly rainfalls at Atherton were used for they provided
the only reliable, unbroken figures covering the last 35 years. Rainfalls within
the maize area vary from 65 inches at Malanda on the southern boundary to
c. 45 inches at Rocky Creek on the northern limits of cultivation. Atherton
is centrally placed, and moreover, since this pattern persists in most years,
rainfall at Atherton is a good guide to totals elsewhere in the maize area.

Maize Yields. The Petty Sessions District of Atherton covers all the
Tableland, and up to 1940 formed the unit of collection for production data.
Since that date collection has been by shires, the two shires Atherton and Kacham
covering the area of the former P.S.D. Post 1940 figures are for the two shires ;
this introduces no change in the area involved.

The Period. Maize production on the Tableland commenced over 70 years
ago. The shorter period 1913-14 to 1948-49 was used to keep as many factors
constant as possible. Prior to 1914 acreages were below 15,000 acres. After
1914 acreages fell below 15,000 in only four years. From the little that is
known of the composition of the crop prior to 1913-14 it appears that numerous
varieties were introduced and the varietal picture probably fluctuated consider-
ably from year to year. Since then, however, it has not appreciably changed.
Until November, 1907, when a large area was thrown open for selection, and to
a lesser extent after the First World War, when the area under cultivation was
expanded for soldier settlement, maize growing was widely spread over the
Tableland and hand cultivation was generally practised following ‘‘ scrub ”’
burns. The present methods of cultivation did not become really widely used
until about 1915-1920.

In Table 1 for the period 1913-14 to 1948-49 are given the growing season
rainfalls (December-February) for Atherton, the totals of the ear-development
stage (March-April), and the yield of maize in bushels per acre for Atherton
Petty Sessions District. Those years in which gales accompanying cyclones
in the Coral Sea caused destruction of a considerable portion of the crop are
marked by an asterisk.

For the analysis of the influence of time and the two rainfall periods on
maize yields certain seasons have been rejected :

(1) Those in which excessive damage to maize resulted from cyclones.
In addition to bringing heavy rains, cyclones frequently flatten portion
of the maize crop.

(2) Seasons in which the December-February rainfall was below 22 inches.
The choice of a 22-inch rainfall in December-February as a rejection
level was based on the fact that preliminary inspection of the data
(see Table 3) indicated that below this level of rainfall maize yields
ranged widely and showed little relation to the actual rainfalls received.
In other words factors such as the incidence of rainfalls within the
growing season, a8 well as the amount, are apparently important.
Since the present analysis 1s concerned with the influence of heavy
rainfalls on maize yields, it was considered advisable to exclude these

DD

24

Rainfalls at Atherton During the Different Stages of Crop Growth for Maize, and Seasonal Yrelds of
_ Maize for the Atherton Tableland during 1914-1919.

SIMONETT AND DRANE.

seasons a8 they would only weaken the relation being studied. While
this rejection level was thus arbitrarily chosen, inspection of the data
of Table 3 in the light of the following analysis appears to confirm

such rejection.

TABLE l.

1944-45
1948-49
1933-34
1938-39
1920-21
1928-29
1926-27
1939-40
1945-46
1932-33
1924-25
1929-30
1921-22
1942-43
1916-17
1927-28
1917-18
1935-36
1913-14
1931-32
1936-37
1922-23
1937-38
1947-48
1934-35
1946-47
1943-44
1925-26
1915-16
1941-42
1930-31
1923-24
1919-20
1918-19
1914-15

Rainfall.
Dec.-Apr.

ase
. ° e e =
WIOWAANOONANAOREUAONTINRAOWFRNKROWNOCORWR WY

Yield Maize.
Bushels/Acre.

Ze
32°
13-
25:
29:
29>
29:
22°
23°
43°
37°
38°
35°
34:
37:
42-
30°
34:
47:
42-
42:
43°
40-
29:
23°
37°
36:
49-
50:
24:
35°
27°
43-
42:
PHU

98
75
22
79
64
dl
24
44
34
33
79
34
46
40
28
11
23
16
46
45
87
35
80
68
26
34

Eighteen seasons were then available for study, covering the 36-year period
The results of the simple correlations are given below.

x,=Yield of maize in bushels per acre.
x,=March-April rainfall.
x,—=December-February rainfalls.

1913-14 to 1948-49.

Tyg —0-5044
Yy3= —0°5581
Tus —0-4314
Tog = —0°0837 x,== Time.
T34= +0-5060
Tog = —0-0328

We then applied a confluence analysis to these simple correlations, following
Frisch (1934), in order to test the usefulness of including all three independent
variables in the explanation of yields. In the circumstances, we think this is a

ol:
47:
48-
4l-
22°
44:
44:
29°
40>
37°
28:
44-
38°
4l-
31:
33°
26-
24-
23°
31:
29:
23°
33°
19-
13-
18-
22-
23°
23°
20:
19-
18-
21+
10-

9:

Rainfall.
Dec.-Feb.

=

PON ROT OH KH ORORDORWRKHOOWnwPrPACWOKFUGSANWNSOS

Rainfall.
Mar.-Apr.

_
ht et Or 00 G2 Sd =]1

* Years in which considerable damage to maize resulted from cyclonic storms.

CW DWE WONODNWOONDWEH WHE ONORHOOKOG

*

CLIMATE AND MAIZE YIELDS ON THE ATHERTON TABLELAND. 25

more useful technique to use than the standard method of applying significance
tests to the partial correlation coefficients. These tests of significance are, we
feel, suspect in this case, for it is difficult to justify the assumption that our data
constitute a ‘‘ sample’’—and a random sample at that. Our methods of
Selection are not designed to preserve randomness, and in any case our data
cover all (suitable) available observations, and hence seem nearer to the concept
of ‘“‘ population ’’. Even if this is no objection, however, it would seem that the
inter-correlation between x, and x, must make the significance tests for these
variables of doubtful validity.

THE CONFLUENCE ANALYSIS, AND INTERPRETATION OF THE
BUNCH MAPS.

The confluence analysis is designed to help overcome some of the difficulties
which arise in interpretation of the results of multiple regression analysis when
more than one relation might be present among the explanatory variables.
This problem is particularly present in economic data since most economic
quantities are the results of simultaneous activities in different sectors of the
market, e.g., prices are determined by simultaneous behaviour of both buyers
and sellers. Thus if we were to construct a relation between price and quantity
sold (or quantity bought), we would have difficulty, after confronting this with
the data, in saying whether this was a demand or a supply equation, since the
same factors enter largely into both. (SeeJ. R. N. Stone, 1945, for a more detailed
explanation of these problems, and for practical applications of the confluence
technique.) But the problem is a more general one, and arises whenever we
have more than one relation between the determining variables, so that the
influence of the second (or other) relation(s) tends to distort the ‘‘ true ’’ influence
of the first relation, which we assume to be the significant one for the purpose
in hand.

The method developed by Frisch for determining where such multicol-
linearity exists consists in plotting the complete sets of regression coefficients
in the form

gi _ _ Byig- ++ +n,
ij(a** +n) Riigy- - + +n)

where Rij, Ri; are co-factors of the matrix of zero-order correlation coefficients,

e
e °
e e
e e
e e
e e

and the 6’s are in terms of “ normalized ”’ variates which can be transformed
to the regression coefficients in terms of the original variables by the formule

p) se) pee

ij(a** °°) Seda ** oN) Lx?
Thus, he starts from the theorem that if perfect correlation is present
without multicollinearity, all regression equations lead to the same result. This
is the familiar result from simple regression analysis that if r,,=I=r,, then the

two regression coefficients 1/r,,, r.,;/1 will be identical. This holds too in the
general case.

The first step, then, is to plot the bunch map for the two-sets. . . 12, 13,
14, 23, 24, 34 from which a general idea of the relations can be obtained, Then

26 SIMONETT AND DRANE.

by systematic inclusion of additional variables it is possible to deduce from the
bunch map whether or not the relation is ‘‘ improved ” by this new explanatory
factor. In the present example, since there is no difficulty in determining the
dependent variable, we are only concerned with the various combinations of the
explanatory variables with this dependent variable.

In interpreting the bunch maps, we must consider the possibility of the
new variable being useful, superfluous, or detrimental to the analysis. Frisch
provides several criteria, given below, for interpreting these cases, but they
must be used with care, since they are highly subjective. Outside, or a priori
theoretical considerations may have to be relied upon to make the final decision,
particularly where the criteria appear to be contradictory.

Indeed, if theoretical considerations indicate that the variable is very
important, whilst our statistical analysis shows it to be detrimental, then it is
the statistical analysis, at least in this linear form, which must be discarded.
All we can deduce, then, is that we cannot measure the influence of that variable
by this method. There will remain, of course, the possibility that the data
might fit some non-linear function.

Frisch states that a variate is useful if the bunch is tightened by its inclusion ;
if the beam representing this variate falls inside the sector of the other beams ;
and if the general direction of the bunch is changed (p. 100). <A variate is
superfluous if “ (1) the bunch does not tighten by the inclusion of the new
variate ; (2) the general slope of the bunch does not change (or more specifically,
each of the beams in the bunch remains unchanged) ; (3) the beam of the new
variate falls outside the sector of the other beams; (4) the beam of the new
variate is much shorter than the other beams in the new bunch; (5) the beams
of the other variates are not appreciably shortened by the inclusion of the new
variate ”’.

‘* If all these criteria are simultaneously fulfilled, the variate in question is
decidedly superfluous. The variate must, however, be considered superfluous
even if only some of the above criteria are fulfilled, particular importance must
then be attached to the criteria (1)-(3)”’ (p. 102).

Finally, ‘‘ if the bunch explodes by the inclusion of the new variate, that
is to say, if it becomes much less tight than before, the new variate must be
considered as detrimental’ (p. 103).

In the present application (Figure 1), we see from the two-sets that time (4),
December-February rainfalls (3), and March-April rainfalls (2), all show a
moderate correlation with yields (1), whilst (23), (24) are non-correlated. (34)
however, does reveal some positive correlation. This appears to be an accidental
bias introduced because of the several criteria of selection. It does serve,
however, to impair the reliability of our regression coefficients for both (3)
and (4) when we include time as a separate explanatory variable in the four-set,
even though some decline in yields due to fertility deterioration over time is to
be expected from general considerations. This is already evident in the three-sets,
which reveal considerable improvement in the relations, following the inclusion
of the third variable, in the bunches (12:3), (13-2), (12-4), (14-2), whilst (13-4)
and (14-3) are ‘‘ exploded’. This is because (3) and (4) are intercorrelated,
and are therefore to some extent incompatible variables. In the fowr-sets,
the (12-34) bunch shows a slight further improvement, whilst the (13-24) and
(14-23) plots are exploded. This seems to indicate that in our four variable
multiple regression analysis, if we hold (3) and (4) constant together we can get
a somewhat better relation than by using simply (12-3) or (12-4) alone. Whilst
(3) and (4) are intercorrelated, when taken together the result is as if we were to
use a third variable z compounded of x; and x, by some such relation as

z=Ax,+BU

CLIMATE AND MAIZE YIELDS ON THE ATHERTON TABLELAND. 27

where (U) represents the residual time element which is required to give full
effect to the influence of time, part of which is contained in x,. Since, however,
we cannot separate this residual in practice, we use another variable, time (x,),
which is already partly contained in x, and this then explodes the relation for
any but the (12-34) case. We feel, however, that since the (12-34) bunch
map is a slight improvement over the (12-3) bunch as a result of including (4)
(the bunch is tightened slightly ; all beams are slightly shortened ; and beam (4)
lies very nearly along the (1) and (3) beams), we are justified in regarding time

4

(13-4) (14:3) (12-34) (13:24) (14:23)

Fig. 1.—Confluence analysis bunch maps.
1=Yield of maize in bushels per acre.
2= March-April rainfalls.
3=December-February rainfalls.

4— Time.

aS a useful variable—a conclusion we could not arrive at from the ordinary
partial correlation coefficients, without confluence analysis.

Yi9-34— —0°6791 Significant at 1% level.
43-04 —0-°5902 Significant at 1% level.
Ty4-03 = —0-2693 Not significant at 5% level.

This is reinforced to some extent by consideration of the multiple correlation
coefficients, and standard errors of estimate.

Weve — 0° 8031 S 1-034 = 4 °0269
Naas =(0-7856 Sie. =—4-1816

28 SIMONETT AND DRANE.

The fact that R,.., is considerably better than R,.,, indicates that (4) has
only a fairly small influence on the regression, but nevertheless R,..5, is still
better than R,..3, reducing the 8.H.E. by the amount of some 3%. This is small
but does indicate that the residual amount of time does have some effect, lending
some support to our argument that if the time-element in x, could be removed in
Some way, then the additional variable time (x,) (not now intercorrelated with
x3) would be significant. We conclude, then, that the influence of time is a
‘‘ useful ’’ one.

By including (4) then we are able to get a better estimate of the influence of

March-April rainfalls on yields. This regression coefficient is
D45-34== —0 4380.

When we come to the other regression coefficients, however, between
yields and December-February rainfalls, and between yields and time, the results
obtained from the four-set are biassed due to intercorrelation between December-
February rainfalls and time, hence we cannot measure the ‘‘ true ”’ influence of
each of these variables. The nature of this interrelation is difficult to unravel,
but for our purposes—prediction—we may tentatively use the estimates from
the four-set, provided that the interrelation persists in its present form, and with
the reservation that they are probably somewhat lower than the “true ”’
regression coefficients, because of this intercorrelation. The estimates are

Di3-04= —0°4021
Dya-03 = —0°1335.
The full regression equation is
X,= —0-4380x, —0-4021x, —0-1335x,-+59-7493.

RESULTS AND DISCUSSION.

From a study of Table 1, the regression analysis, and a close inspection of
the monthly rainfall distribution for the 35 years studied, the following points
emerge :

1. The maximum mean yield of maize on the Atherton Tableland was
50-28 bushels per acre in the 1915-16 season, when rainfalls at Atherton were
29-6 inches from December to April, distributed as follows :

Dec. Jan. Feb. Mar. Apr.
Sj ba SME Oa Satta pc) Uleateee 3) Seiler san 372 Total, 29-6 inches.

Inspection of the distribution of rainfall for years with yields close to the
maximum suggests that the highest yields would be obtained with the following
rainfall distribution at Atherton :

Dec. Jan. Feb. Mar. Apr.
ff 8 7 3 2 Total, 27 inches.

This is not strictly the ‘‘ optimum ”’ rainfall for the crop since portions of the
maize lands would receive some inches more or less depending on their position.
Effectively, however, it represents the optimum rainfall for the area as a whole.

2. In slightly less than 80 per cent. of the years under review, rainfalls
during the December-February period were in excess of the optimum of 22
inches.

3. In slightly less than 90 per cent. of seasons, rainfalls during the March-
April period were in excess of the optimum of five inches. The analysis of the
rainfall-yields relationship for these two critical periods thus devolves largely
into an assessment of the influence of excessive rainfalls on maize yields.

CLIMATE AND MAIZE YIELDS ON THE ATHERTON TABLELAND. 29

4, Cyclones caused major damage to the maize crop in 8 of the 35 years
studied, depressing yields by an average of the order of 10 bushels per acre
below the expected yield for the various rainfall levels.

5. In years either not affected, or only slightly affected, by cyclonic storms,
and with a December-February rainfall exceeding 22 inches :

(i) Rainfalls greater than 22 inches during the growing period December-
February significantly depress yields, each additional inch reducing the crop by
approximately 0-402 bushel per acre. This may be due to weed competition
for nitrogen in excessively wet seasons when inter-row cultivation is abandoned
earlier than in more favourable seasons, or it may be due to excessive leaching
of nitrates, or it may be partly a function of the decreased light intensity during
cloudy, rainy weather. Whatever the reason, or reasons, the net effect of the
factors associated with very heavy rainfalls during this period is to reduce the
growth rate of the maize plant and hence both the number and size of the cobs.

(ii) Rainfalls above five inches during cob-development in March-April
Significantly depress yields, each additional inch by 0-438 bushel per acre below
the level produced by the December-February rainfall. The high humidities
and the moist state of the cobs which are a natural corollary of high rainfalls
during this season, together with almost perfect incubation temperatures—
average day temperatures of 80° F., night temperatures of 60° F.—favour cob
diseases, particularly Diplodia ear rot. At present no economic control of cob
rots is available, and in almost every season some loss is experienced from these
diseases.

In addition to causing large-scale rejection of Diplodia—infested, dead
grain at the harvest—inclusion would reduce the sale-value of the crop, as well
as increasing the silo-storage difficulties—high rainfalls in March and April
result in a high moisture content in the grain, particularly in that pulled in late
May or June. Consequently, before storage in the nests of silos at Atherton,
Tolga and Kairi, most of the early harvest must be artificially dried to a moisture
content of 14 per cent. (Straughan, 1949). |

6. During the period under review, yields declined to a slight but significant
degree—probably of the order of 0-1335 bushel per acre per annum—and
present-day yields would appear to be at least five bushels per acre below those
obtaining in the beginning of the period.

No precise data are available as to the exact causes of this decline, but on
general grounds we may reasonably expect that there would be a time-nitrogen
regression (in the absence of fertilizers and nitrogen-building rotations), a soil-
structure deterioration, and erosion loss similar to those noted by Cornish (1949)
in his study of yield trends in the wheat belt of South Australia.

A suggestion of such a nitrogen decline in the maize lands is found in the
following figures (Table 2), taken from Teakle (1950).

If the nitrogen-loss of the maize area north of Tolga, which was originally
under Hucalyptus woodland, parallels that of the rain forest soils to the south
(Atherton, Table 2), there is every reason to suspect a nitrogen-deficiency in
them at least, for even under the virgin woodlands (Mapee et seqg., Table 2)
they were only of moderate to low nitrogen content.

Teakle further notes that ‘“‘ under paspalum (pasture) or cultivation the
crumb structure is greatly reduced and there is some sub-surface compaction ”’.

Sheet erosion, though masked by the uniform red colour of the soils, is
evident on long gentle slopes where drifting against fences is characteristic.

It appears likely, then, that all three factors are contributory to the
diminishing maize yields.

30 SIMONETT AND DRANE.

7. The suggested March-April optimum rainfall of five inches at the 22-inch
level almost certainly does not apply through the whole range of December-
February rainfalls. Logically, we would expect that stored soil moisture from
heavy rains in the first period, and the residual effect of the higher humidities
attending these heavy rains would reduce the requirements during ear-develop-
ment. Unfortunately, no data are available on this point. The converse is
probably not true, except when the distribution in the preceding months is
irregular.

TABLE 2.

Nitrogen Content of Selected Red Loams on Basalt, Atherton Tableland.
(After Teakle.)

Approx. Depth,

District. Rainfall. Ins. Condition. N%.
Atherton ae a 50 0-6 Rain forest. 0-613
Yungaburra .. a 55 0-6 Cleared and under Wild 0-559

Tobacco and regrowth.
Atherton ae Ae 52 0-7 Maize. 0-235
Sth. Mapee .. ge <50 0-6 Eucalyptus woodland. 0-257
Mapee : am <50 0-6 0-202
Sth. Carbeen ne 40 0-6 0-155
Carbeen Be iy 40 0-2 0-194
Nth. Carbeen af 40 0-6 0-113
Nth. Carbeen as 40 0-6 0-114

8. Seasons little affected by cyclones, and in which the December-February
rainfalls were below 22 inches, ranged widely in yields from 21-50 to 43-12
bushels per acre (Table 3).

TABLE 3.
Maize Yields and Rainfalls in Low-rainfall Years, Atherton Tableland.

|
| Growing Yield,
Season. Dec. Jan. Feb. Mar. Apr. Season Bushels/
Total. Acre.

1914-15 .. 386 256 294 69 59 10-64 21-50
1918-19 .. 412 410 181 599 591 21-93 42-56
1919-20 .. 113 1,299 711 102 408 26°33 43-12
1923-24 .. 651 576 610 671 156 26°64 27-52
1930-31 .. 450 1,042 487 287 403 26-69 35°31
1941-42 .. 333 164 1,498 187 734 29-16 24°73
1946-47 .. 202 442 1,239 1,089 200 31-72 37°34
1947-48 .. 380 200-4 365 1,107 319 33°71 29-68

From the information in Table 3 it is clear that growing season rainfalls
alone are not a good indicator of crop success when rainfalls are below the
optimum. Nor are the pre-seeding rainfalls of September to November which
in the above cases bear no relation to yields. The important factors are probably
rainfall incidence and (possibly) heat wave spells in relation to critical stages in
growth.

CLIMATE AND MAIZE YIELDS ON THE ATHERTON TABLELAND. oe

SUMMARY.

Maize yields on the Atherton Tableland in north Queensland are significantly
depressed by rainfalls exceeding 22 inches in the growing season of December-
February (including the tasseling period), each additional inch reducing yields
by approximately 0-402 bushel per acre. Rainfalls above five inches during
cob-development in March-April also reduce yields, each inch by 0-438 bushel
per acre. During the period 1913-14 to 1948-49 maize yields declined to a
slight but significant degree and present day totals are at least five bushels per
acre below those of the early years. The factors responsible for the time-yields
decline, and the factors associated with rainfall which appear to be responsible
for the rainfall-yield suppression were discussed. The statistical technique
used was that of confluence analysis.

ACKNOWLEDGEMENTS.

Grateful acknowledgement is made to Mr. R. 8S. G. Rutherford for advice
and criticism of the statistical treatment, to Mr. E. J. Breakwell and Dr. L. G.
Miles for helpful criticism, to Miss Margaret Oates for help in computations, to
the Central Weather Bureau, Melbourne, for supplying rainfall data, and the
Queensland Government Statistician for supplying maize production data.
One of us (D.S.8.) is indebted to Professor J. Macdonald Holmes for his continued
interest, and to the Commonwealth Research Grant, University of Sydney, for
financial assistance.

REFERENCES.

Cornish, H. A., 1949. Aust. J. Scr: Res., Ser. B, 2, 83.

Frisch, Ragnar, 1934. Statistical Confluence Analysis by Means of Complete Regression
Systems, Oslo.

Skerman, P. J., 1947. Bureau of Investigation, Queensland, 4th Annual Report.

Stone, J. R. N., 1945. J. Roy. Stat. Soc., Parts III-IV, 108, 286.

Straughan, W. R., 1949. Queensland Ag. J., 68, No. 6, 311.

Teakle, L. J. H., 1950. Univ. Queensland Papers, Faculty of Agriculture, 1, No. 1, xx.

PALLADIUM COMPLEXES.

PART V. REACTIONS OF PALLADIUM COMPOUNDS WITH 2: 2’ DIPYRIDYL.

By 8S. E. LIVINGSTONE, A.S.T.C., B.Sc.

Manuscript received, June 18, 1952. Read, July 2, 1952.

An investigation of the types of compounds formed by palladium with
2:2’ dipyridyl has been carried out similar to that describing the reactions of
palladium compounds with 1:10 phenanthroline (Livingstone, 1951).

Some complexes of palladium containing 2: 2’ dipyridyl (dipy) have been
previously reported: viz. Pd dipy Cl, (Morgan and Burstall, 1933); Pd dipy
(NO,). and Pd dipy C,O, (Mann and Purdie, 1936). Compounds of the type
Pd dipy X,, similar to these, are described here (X=Br, I, CNS, NO,). These
compounds are all insoluble in water and organic solvents—except Pd dipy
(NO3),, which is moderately soluble in water—and are precipitated when a
solution of PdX, or K,PdX, is treated with an aqueous solution of 2 : 2’ dipyridyl.
These compounds probably have the structure

Baa
Pd
[\7 _e

the dipyridyl acting as a bidentate ligand and thus giving the palladium atom
a coordination number of four, which is usual for divalent palladium. Pre-
cipitates are not formed when dipyridyl is added to solutions of (i) PdSQ,,
(ii) PASO, + NaF, (ii) K,Pd(CN),. All the above compounds of geneval formula
Pd dipy X, dissolve in cold aqueous solutions of KCN to give colourless solutions
liberating 2:2’ dipyridyl, in analogy wita Pd[phen|X, (where phen=1: 10
phenanthroline) (Livingstone, 1951).

A tetravalent palladium complex tetrachloro-?: 2’ dipyridyl palladium

(IV) chloride is formed as reddish-orange crystals by the oxidation of Pd dipy Cl,
with chlorine.

CHCl,
— [Pd dipy Cl,]
VII.

[Pd dipy Cl,]+Cl,
II.

This compound VII, unlike its phenanthroline analogue (Livingstone, 1951),
does not liberate chlorine in moist air, nor on boiling with water, and is stable

PALLADIUM COMPLEXES. ao

on heating to 160°C. It appears to have much greater stability than other
tetrachlorodiammine palladium (IV) type compounds previously reported
(Rosenheim and Maass, 1898; Drew et alii, 1932).

The reactions of [Pd dipy Cl,] (II) with different amines were investigated.
II dissolves in excess aqueous 2: 2’ dipyridyl to give a yellow solution. Con-
centration and precipitation with acetone yields the original [Pd dipy Cl,].
However, salts of the ion [Pd dipy,|*+*+ can be obtained with the anions per-
chlorate and 2-naphthol-6-sulphonate.

[Pd dipy Cl,] 38 dipy ea 4 yellow Solution
I]

evaperation
--
acetone

(Pd dipy,) (C10,), (Pa dipy,| (CioHOH SO), 3H,O (Pa dipy C\,)
VITI [X Il

Existence of the equilibrium
[Pd dipy Cl,] + dipy = [Pd dipy.|+t*+ + 2Cl-

is demonstrated by the fact that addition of a large excess of NaCl to an aqueous
solution of [Pd dipy,] (ClO,), immediately precipitates [Pd dipy Cl,].

Solutions of mixed tetrammines of the type [Pd dipy Am,|*+ (where
Am=ammonia, pyridine, 2Am-=ethylenediamine, propylenediamine) are
obtained by treating IT with an excess of amine in dilute aqueous solution.
Addition of NH,C1O, to the pale green solution precipitates [Pd dipy Am, ]|(C1O,4)..
These perchlorates, surprisingly, are fairly soluble in cold water, but all excepting
[Pd dipy en](ClO,), could be recrystallized from water. A considerable excess
of 1 : 10 phenenthroline must be added to [Pd dipy Cl,] and the mixture boiled to
effect solution; addition of NH,ClO, precipitates [Pd phen,]|(ClO,),. This
is probably due to the fact that 2:2’ dipyridyl is appreciably volatile under
these conditions.

Comparison of the results obtained with palladium and 2: 2’ dipyridyl with
those for palladium and 1:10 phenanthroline (Livingstone, 1951) and for
platinum and 2: 2’ dipyridyl (Morgan and Burstall, 1934) shows that there is a
considerable similarity between them. Failure was experienced in all cases to
isolate [M(chel),|Cl, (where M=Pt, Pd and chel=dipy or phen), [M chel Cl, |
being obtained in each case. However, the perchlorates and 2-naphthol-6
sulphonates of [Pd phen,]++ and [Pd dipy,]**+ could be prepared. Morgan
and Burstall (1934) claimed to have isolated (i) [Pd dipy,|PtCl, and
(ii) [Pt dipy py,]PtCl,, but this seems unlikely since their products were yellow
and one would expect these compounds to be pink or possibly green (Drew
et alii, 1932) and hence the substances obtained were probably (i) [Pt dipy Cl,]
and (ii) a mixture of [Pt dipy Cl,| and [Pt py,Cl,]. Using similar procedures
with the palladium phenanthroline complex, the author obtained yellow
[Pd phen Cl,]. However, by altering the conditions, viz. addition of a solution
of [Pd phen,](ClO,), to excess of concentrated ice-cold solution of K,PdCl,,
salmon pink products resulted, which were undoubtedly impure [Pd phen, |PdCl,.
However, this latter compound could not be prepared in the pure condition.

E

34 S. E. LIVINGSTONE.

It is possible that it may have been contaminated with a bridged compound
having the structure

sae

N
Ni e
Pd Pd
Aan A ie ‘

| Cl
N

Support for this is given by the fact that if the solutions are mixed at the boiling
point, which usually favours formation of bridged compounds, the palladium
content of the resulting product is increased. A similar result was obtained
using K,PdBr,, but no pure compounds could be isolated. Chatt and Mann
(1939) attempted to prepare bridged palladium complexes of this particular
type (t.e. having four halogen atoms attached to one palladium atom) but
reported negative results. The matter is being further investigated as part of a
study of the formation of bridged compounds of palladium.

EXPERIMENTAL.
(III) Dibromo-2 : 2’ dipyridyl palladium (I).

A hot aqueous solution (50 ml.) of K,PdBr, (0-5 g.) was mixed with a warm solution of
0:16 g. of 2: 2’ dipyridyl—I— in water (30 ml.) containing 3 ml. of alcohol. Orange prisms of
III separated immediately ; the product was insoluble in water and organic solvents, but readily
soluble in aqueous KCN to a colourless solution. Yield 0-385 g.

Found: Pd, 25-69%; Br, 38°4%.
PdC,,H,N,Br, requires: Pd, 25-24%; Br, 37-82%.

(IV) Diido-2 :2’ dipyridyl palladium (II).

(i) A solution of K,PdI, was treated with a hot aqueous solution of I. A pinkish brown
precipitate resulted.

Found: Pd, 20°7%.

(ii) Dichloro-2 : 2’ dipyridyl palladium (II)—compound II—(0-45 g.) and 0:35 g. of I were
dissolved in hot water containing a few drops of alcohol. The yellow-brown solution was poured
into 20 ml. of water containing KI (5 g.). 0-63 g. of pinkish brown crystals of IV were pre-
cipitated.

Found: Pd, 20:6%; I, 48:1%.

PdC,,H.N.I, requires :/ Pd, 20-65%; Ic) 40- 1200,

(V) Dithiocyanato-2 : 2’ dipyridyl palladium (II).

K,PdCl, (0:3 g.) was dissolved in water (30 ml.) and KCNS (2-5 g.) added. On addition of
an aqueous solution containing I (0:14 g.), pale yellow crystals were thrown down. Yield,
0-33 g. .

Wound':) Pd, 28-19): (Ci-37*89) >| Eo 2507.

PdC,,H,N.(CNS). requires: Pd, 28:15%; C, 37-92%; H, 2-13%.

PALLADIUM COMPLEXES. 35

(VI) Denitrato-2 : 2’ dipyridyl palladium (LJ).

I (0:15 g.) dissolved in hot water (30 ml.) was added to a solution of Pd(NQs3), (excess) in
dilute HNO, and the mixture heated to boiling. On cooling, pale yellow crystals, appreciably
soluble in water, were deposited. Yield, 0-20 g.

Found: Pd, 27-3%.

PdCO,,H,N.(NO3). requires: Pd, 27-58%.

(VII) Tetrachloro-2 : 2’ dipyridyl palladium (IV).

II (0:4 g.) was suspended in chloroform and chlorine passed into the suspension for 40
minutes. On filtering, the product was washed with chloroform, then dry ether, and dried
an vacuo over P,O;. Yield, 0:43 g. The reddish orange crystals had no smell of chlorine and
were not decomposed by heating to 160° C., nor by boiling with water.

Found: Pd, 26:5%; Cl, 35-1%.

PdC,,H,N.Cl, requires: Pd, 26-36% ; Cl, 35-05%.

Reactions of II with Amines.
(A) 2: 2’ Dipyridyl.
II (0:2 g.) and I (0:2 g.) were heated to boiling in water (15 ml.) to give a deep yellow solution.
On concentration to 5 ml., acetone was added. Pale cream needles were precipitated.
Found: Pd, 32:0%.
Calculated for PdC,,H,N,Cl,: Pd, 31-96%.

(VIII) Bis (2: 2’ dipyridyl) palladium (IL) perchlorate.

IT (0-2 g.) and I (0-2 g.) were warmed with water (25 ml.) and alcohol (3 ml.).. On addition
of an aqueous solution of NH,ClO, 0°35 g. of deep yellow dentritic crystals of VIII separated.
On recrystallization from water (150 ml.) 0-3 g. of deep yellow prisms were obtained.

Found: Pd, 17:3%.

Pd(C),H.N.).(ClO,), requires.: Pd, 17-29%.

0-15 g. of VIII were dissolved in boiling water (100 ml.) and NaCl (3 g.) added ;_ pale yellow
needles were deposited while the solution smelt strongly of 2 : 2’ dipyridy]l.

Hound): Pd, 32°0%.

Calculated for PdC,,H,N,Cl,: Pd, 31-96%.

(IX) Bis (2: 2’ dipyridyl) palladium (IL) 2-naphthol-6-sulphonate trihydrate.

II (0-1 g.) and I (0:1 g.) were warmed with water to give a yellow solution and excess of a
solution of sodium 2-naphthol-6-sulphonate was added. Deep yellow crystals (0:25 g.) separated ;
these were washed with hot water, then acetone and dried in vacuo over P,O;. On heating in a
closed tube, water was evolved at a temperature just below decomposition.

Raumnd:) Pd5 11:99 3) Cy 5:99 sc HH, 3° 8%.
Pd(C,>H,N2).(C;>H ,OHSO;),.3H,O requires: Pd, 11-60%; C, 52-24%; H, 3°95%.

(B) Ammonia.
(X) Diammino-2 : 2’ dipyridyl palladium (LI) perchlorate.

II (0-35 g.) was treated with water (30 ml.) and IN NH,OH (4 ml.) and the mixture warmed
to 40°C. After 15 minutes a pale greenish yellow solution resulted. On addition of NH,C10,

thin needles of X were slowly precipitated. Recrystallization from water (8 ml.) yielded 0-14 g.
of colourless prisms.

Found: Pd, 21:4%.
PdC,,HsN,(NH;),.(ClO,). requires: Pd, 21-52%.

36 S. E. LIVINGSTONE.

(C) Pyridine.
(XI) Dipyridine-2 : 2’ dipyridyl palladium (IL) perchlorate.
II (0:35 g.) was placed in water (40 ml.) and treated with drops of pyridine in the cold to

give a pale greenish yellow solution. Addition of NH,Cl1O, caused acicular prisms of XI (0:51 g.)
to slowly separate. Recrystallization from water gave very pale green elongated prisms (0°43 g.).

Found: Pd, 17-1%.
PdC,,H,N.(C;H;N).(ClO,), requires: Pd, 17-21%.

(D) Hthylenediamine.
(XII) Hthylenediamine-2 : 2’ dipyridyl palladium (II) perchlorate.

II (0:3 g.) was treated with water (10 ml.) and 3% aqueous ethylenediamine (3 ml.). After
10 minutes at room temperature a clear pale green solution was obtained. Addition of NH,C1O,
caused slow crystallization of thin needles (0-10 g.) of XII, very soluble in water.

Found: Pd, 20-1%.
Pd(C,,H,No(C»H,N.)(C1O,). requires: Pd, 20-44%.

(E) Propylenediamine.
(XIII) Propylenediamine-2 : 2’ dipyridyl palladium (£1) perchlorate.

II (0-3 g.) was treated with water (7 ml.) and 3% aqueous propylenediamine (3 ml.). To
the resulting pale yellow green solution, NH,ClO, was added, precipitating colourless needles
(0:23 g.). These were recrystallized from water (8 ml.).

Hound?) (ed; 19-19%.

Pd(C,,>HsNe2)(C3HipN2)(C1O4), requires: Pd, 19-91%.

(F) 1:10 Phenanthroline.

II (0:3 g.) and 1: 10 phenanthroline (0-2 g.) were heated in water (30 ml.) ; an extra 0-2 g.
of 1: 10 phenanthroline was required to achieve a clear solution. NH,ClO, precipitated yellow
prisms of (Pd phen,)(C1O,)., which were recrystallized from water (300 ml.). Yield, 0-37 g.

Pound = "Pd, 15289, 25-8%,.

Calculated for Pd(C,,H,N.).(ClO,),: Pd, 16-02%.

Attempted preparation of bis (1 : 10 phenanthroline) palladium (L1) chloropalladate (II).

(i) Bis (1:10 phenanthroline) palladium (II) perchlorate—compound XIV—(0:14 g.)
dissolved in boiling water (80 ml.) was added slowly to a solution of K,PdCl, (0-45 g.) in ice cold
water (100 ml.). The salmon pink product was filtered, washed with cold water, then acetone.

Found: Pd, 33-8%.

(ii) XIV (0-25 g.) in boiling water (100 ml.) was slowly added to a solution of K,PdCl, (1 g.)
in ice cold water (80 ml.). Yield of salmon pink crystals, 0-30 g.

Found: Pd,'33°6%; <Cl, 22°2%, 2.e. Pd 1Ci= F007. 1-99:

Pd(C,,H,N,),PdCl, requires: Pd, 29-82%; Cl, 19-83%.

Pd,C,,.H,N,Cl, requires: Pd, 39:86%; Cl, 26°49%.

Attempted Preparation of bridged chloro compound.

A solution of XIV (0-25 g.) in boiling water (100 ml.) was slowly poured into a boiling
aqueous solution (100 ml.) of K,PdCl, (0-7 g.). The solution was kept boiling for a further 30
minutes. The orange brown product (0:33 g.) was filtered, washed well with hot water, then

acetone, and dried over P,Os.
Found: Pd, 35-6%.

PALLADIUM COMPLEXES. at

Attempted preparation of bridged bromo compound.

XIV (0:15 g.) dissolved in boiling water (80 ml.) was added to a solution of K,PdBr, (0-7 g.)
in boiling water (50 ml.) over a period of 20 minutes and the mixture kept at the boil for a further
20 minutes. The precipitate of reddish-brown crystals was filtered, washed with hot water,
then acetone, and dried over P,O;.

Found: Pd, 28:2%.

Pd,C,,H,N.Br, requires: Pd, 29-91%.

Pd(C,,H,N.),.PdBr, requires: 23-88%.

SUMMARY.

An investigation of compounds formed by palladium with 2: 2’ dipyridyl
(compound I) acting as a chelating group, has been described. I reacts with
PdX, and K,PdX, to form [Pd dipy X,] (where dipy=2: 2’ dipyridyl and
X=Br, I, CNS, NO;). Similar compounds (where X =Cl, NO, and 2X =C,0,)
have been previously reported. [Pd dipy Cl,] (compound II) is oxidized to the
orange-red compound [Pd dipy Cl,] by passing chlorine through a suspension
of II in chloroform. II dissolves in excess of aqueous solution of I to form a
solution of the tetramine chloride [Pd dipy,|Cl,, which could not be isolated from
solution, but addition of ammonium perchlorate and sodium 2-naphthol-6-sul-
phonate yields VIII, [Pd dipy,](ClO,)., and IX, [Pd dipy,|(C,,H ,OHSO;),..3H,O
respectively. II also dissolves in aqueous ammonia, pyridine (py), ethylene-
diamine (en) and propylenediamine (pn) to give pale yellow green solutions
which upon addition of NH,ClO, yields X, [Pd dipy (NH3).|(ClO,),, XI,
[Pd dipy py.|(ClO,),, XII, [Pd dipy en](ClO,),. XIII, [Pd dipy pn](ClQ,),.
Attempts to prepare [Pd phen,|PdCl, (phen=1 : 10 phenanthroline) gave salmon
pink products with high Pd and Cl content, possibly due to contamination
with a bridged compound, which could be isolated in the pure state.

ACKNOWLEDGEMENT.

The author is indebted to Dr. F. P. J. Dwyer for his generous gift of 2 : 2’
dipyridyl used in this investigation and also to Dr. E. Challen for the carbon
and hydrogen analyses.

REFERENCES.
Chatt, J.. and Mann, J. G., 1939. J.C.S., 1622.
Drew, H. D. K., Pinkard, F. W., Preston, G. H., and Wardlaw, W., 1932. J.C.S., 1895.
Livingstone, 8. E., 1951. Tuis Journat, 85, 151.
Mann, F. G., and Purdie, D., 1936. J.C.S., 873.
Morgan, G. T., and Burstall, F. H., 1933. J. Indian Chem. Soc., P. C. Ray Commemoration,
Vol., 1-16.
Morgan, G. T., and Burstall, F. H., 1934. J.C.S., 965.
Rosenheim, A., and Maass, T. A., 1898. Z. anorg. Chem., 18, 331.

Inorganic Chemistry Department,
School of Applied Chemistry,
N.S.W. University of Technology.

LIVERSIDGE RESEARCH LECTURE*

ELECTRON DIFFRACTION IN THE CHEMISTRY OF THE
SOLID STATE.

By A. L. G. REES, D.Sc., Ph.D., D.I.C.
Chemical Physics Section, Division of Industrial Chemistry, C.S.I.R.O.,
Melbourne.

With 17 Text-figures and Plates IT-IV.

Technical rather than scientific progress over the last 20 years has been
largely responsible for the useful application of physics to the problems of
chemistry. Naturally enough physical methods find immediate application
in the identification and analysis of chemical species ; later, as a proper under-
Standing of the physical phenomenon is acquired, application to the more
fundamental problems of structure, energetics and dynamics becomes possible.
Electron diffraction has been understood fairly completely, in a formal way at
least, from a time shortly after the establishment of wave mechanics, but recent
experimental advances have only now permitted some realization of its full
potential in chemistry.

The terms of the Liversidge Bequest require that the subject of this lecture
should be such as to encourage research and stimulate interest in some aspect of
chemistry. The importance of the chemistry of the solid state, both in academic
and technological spheres, the relevance and uniqueness of the information on
the solid state provided by electron diffraction and the prospect of exciting
experimental and theoretical development constitute a topic which, I venture
to submit, satisfies these terms.

THE ELECTRON DIFFRACTION METHOD.

Electrons, of mass m and charge e, accelerated by a potential V of the order
of 50 kV., have, from the relation A=h/(2meV)?, an equivalent wave-length ’
of ~0-06 A, somewhat smaller than the normal X-ray wave-lengths used in
diffraction studies, but still of the right order of magnitude for diffraction by
regular arrangements of atoms or molecules. One would expect an electron
diffraction pattern to have a formal resemblance to an X-ray diffraction pattern ;
this is so, but there are important differences. Indeed, it is these differences
which represent the value of electron diffraction in chemistry. By reason of its
charge a 50 kV. electron can penetrate only some hundred or so atomic planes
without suffering considerable inelastic scattering (7.e., scattering in which
kinetic energy of translation is not conserved). Moreover, electrons are scattered
elastically by atoms some 10’ times more efficiently than are X-rays. Electron
diffraction is restricted then to the study of solid surfaces and of extremely small
crystals, for both of which X-ray diffraction is of indifferent value.

Those refinements to the technique of electron diffraction responsible for
widening its scope have been largely electron-optical in character. An electron
diffraction camera is simply an instrument for producing a narrow beam of

* Delivered to the Royal Society of New South Wales, July 17, 1952.

ELECTRON DIFFRACTION IN THE CHEMISTRY OF THE SOLID STATE. 39

high-velocity electrons by means of a system of electromagnetic lenses and small
apertures, and recording on a photographic plate held normal to the beam direc-
tion the beams diffracted by the specimen. Figure 1 is a diagrammatic repre-
sentation of the principle of a high-resolution electron diffraction camera.

For high resolution in single-crystal reflections, with which we shall be
particularly concerned here, the electrons in the narrow pencil impinging on the
specimen must be of uniform velocity, as this determines their wave length, and
must have small angular spread. To fulfil these conditions stability in

APERTURE 2

[2] LENS 2.

—- SPECIMEN

PLATE

Fig. 1.—Principle of high-resolution electron diffraction camera.

accelerating voltage (determining the electron velocity) and in lens current
(determining the focal length of the lens) and freedom from external perturba-
tions, either electrical, magnetic or mechanical, are required. For a given
resolution, the tolerable fluctuations in any of these may be calculated. This
problem was analysed before construction of the diffraction camera in Melbourne
(Cowley and Rees, 1952), which is shown in Plate II, Fig. 2.

Many of the applications of electron diffraction to solid state studies require
high resolution, since it is the fine structural detail of these patterns which
provides the relevant information. Recent work indicates that experimental
conditions must be controlled even more rigorously if all useful information is
to be extracted from the method.

Diffraction patterns obtained from specimens in which the crystallites are
in random orientation, either by transmission through a powder specimen or by
reflection from the surface of a polycrystalline solid, consist of circles centred
on the undeflected beam. The radii of the rings give the diffraction angle 26,,.,,
related to the lattice spacings d,,, by the Bragg law

(2/A) sin 0,,,=1/dpg)-

Patterns from a single crystal or from a number of crystals in identical orientation
consist of single reflection spots or, in special cases, an array of spots, which has
become known as a cross-grating pattern owing to its resemblance to the diffrac-
tion pattern from a two-dimensional grating.

40 A. L. G. REES.

It is convenient to discuss diffraction from single crystals in terms of a
concept introduced by Ewald, that of the reciprocal lattice. This concept is one

Reciprocal
Lattice.

Fig. 3.—Ewald’s reciprocal-lattice construction for diffraction conditions.

of the most useful in diffraction studies; as it is necessary for subsequent
discussion it will be outlined briefly here. The vector equivalent of Bragg’s law
is given by a construction wherein the incident beam is represented by a vector

Fig. 4.—Reciprocal-lattice construction for diffraction by a thin
crystal plate. Electron beam parallel to short dimension of
erystal.

S, of magnitude 1/4 and the diffracted beam by a vector S,, of identical magnitude
but making an angle 20 with the incident beam. The difference between the
vectors S, and S, is a vector of magnitude 1/d,,, if the diffraction condition is
satisfied. All diffracted beam vectors must terminate on the surface of a sphere

ELECTRON DIFFRACTION IN THE CHEMISTRY OF THE SOLID STATE. AL

centred on the origin of the beam vectors and of radius 1/A. If d,,, is a proper
lattice spacing then it is obvious from Figure 3 that Bragg’s law is implicit in

this construction. The vector OB, having the direction of a normal to the
diffracting planes and of magnitude 1/d,,, is termed a reciprocal lattice vector.
For any given crystal lattice, a reciprocal lattice of terminal points of vectors
1/d,,,; maay be constructed. The directions of the permitted diffracted beams

for a specific orientation of the crystal to the incident beam vector AO will then

Fig. 5.—Cross-grating pattern from crystal of boric acid.

be given by the vectors AB terminated at the intersections of the Ewald sphere
and the integer points of the reciprocal lattice. A change in orientation of the
real crystal leads to a rotation of the reciprocal lattice about its origin O through
the same angle. In this way different integer points of the reciprocal lattice
intersect the surface of the Ewald sphere and diffracted beams are permitted.
Reflections for which 1/d,,, is greater than 2/A will not be permitted as the corres-
ponding reciprocal lattice points lie outside the volume swept out by the Ewald
Sphere pivoted on O.

In the same way that the resolving power of an optical diffraction grating
is determined by the total number of lines, the angular breadth of a reflection
from a three-dimensional crystal grating is greater the smaller the number of
Scattering points in the crystal. For an infinite crystal the reciprocal lattice
is a lattice of points and the angular range of any reflection is infinitesimal ;
for a finite crystal the deterioration in resolution appears in the reciprocal lattice

42 A. L. G. REES.

as a Spreading of each reciprocal lattice point. This spread will be greatest in
the direction of the smallest dimension of the crystal. The Ewald sphere will
now pass through the region around a reciprocal lattice point for a range of
directions of the incident beam vector. This is the origin of the dependence of
breadth of reflections on crystal size.

The construction for the diffraction of fast electrons is simplified as 1//
is much greater than the unit reciprocal lattice spacings and the Ewald sphere
may be approximated by a plane or near-plane sectioning the reciprocal lattice.
Moreover, since the small penetration of electrons limits the effective dimensions
of the three-dimensional grating, it is to be expected that reciprocal lattice
points will be invariably extended into the space between the integer points.
For a thin crystal plate the diffraction conditions are as shown in Figure 4.
The reciprocal lattice points are drawn out into spikes parallel to the short
dimension of the crystal and the Ewald sphere of large radius can section many
of these spikes simultaneously. The pattern resulting from such a diffraction
experiment is a cross-grating pattern, a symmetry-true projection of a section
of the reciprocal lattice. An example of such a pattern is given in Figure 5.

THE STRUCTURAL PROBLEM.

Application of electron diffraction to chemical problems followed its
discovery very promptly ; particularly successful application came from the
laboratories of G. I. Finch, G. P. Thomson, 8. Kikuchi, J. J. Trillat and L. H.
Germer (Thompson and Cochrane, 1939; Finch and Wilman, 1937). Studies
of the growth and structure of thin films, of the structure of metal oxides,
particularly surface oxides, of surface films and lubrication were among the
important fields of application to chemistry. The useful fundamental informa-
tion extracted from these investigations was confined to cell dimensions, crystal
symmetry, orientation and habit, and to some extent structure. No structural
problems were attempted from single-crystal data and Fourier methods were
not applied. The reason for this lay in the anomalies in observed electron
diffraction intensities and in the presence of reflections forbidden on space-group
grounds. These anomalies were ascribed to various causes, among them
dynamic interaction, limitation of crystal dimensions and secondary scattering.

Over the last five years Pinsker and his colleagues (Vanchstein and Pinsker,
1949, 1950) have used intensity data from powder patterns to analyse crystal
structures by Fourier methods. The disadvantage of using patterns of this
type, from which only one parameter, the Bragg angle, is available for determining
the geometry of the lattice and in which different reflections often appear in
coincidence, is familiar to all crystallographers. The possible use of single-
crystal data for structure analysis was examined in Melbourne several years
ago as the result of the study of secondary scattering of electrons (Cowley ef al.,
19d51a).

The high efficiency of elastic scattering of electrons results in the appearance
of a considerable proportion of the incident energy in the diffracted beams ;
strong primary reflections may therefore be rediffracted by underlying crystallites
or mosaic elements of the same crystal as shown in Figure 6. This secondary
scattering may lead to many unusual and anomalous features in patterns ;
in polycrystalline specimens extra spots, extra rings, groups of spots, diffuse
bands with sharp edges and rings centred on primary spots may occur; in
single-crystal patterns a redistribution of the intensity among the reflections in
the pattern and the appearance of the forbidden reflections may result. These
features are illustrated in the patterns as shown in Plate ITI, Figs. 7, 8, 9 and 10.

In the cross-grating pattern of Plate III, Fig.10, there are superimposed
secondary patterns identical with the primary pattern and of the same orienta-
tion, but centred on each of the primary reflections. The secondary reflections

ELECTRON DIFFRACTION IN THE CHEMISTRY OF THE SOLID STATE. 43

consequently coincide with primary reflections of different indices. This is
illustrated in Figure 11. Each reflection in the pattern has lost intensity by
contributing to all other reflections and has gained by secondary contributions
from every other spot. This modification of the intensities is the obstacle to
their use for structure analysis.

Incident
Beam

TLL. } oa
TELLER. > - ili begs

1

Undeflected Primary Secondory

Beam Reflectian Reflection
(4,k,1l) (hake, about hy kyl

fram 1.)

Fig. 6.—Diagrammatic representation of secondary elastic
scattering of primary diffracted beams by underlying crystallites.
See Plate III, Figs. 7-10.

The theoretical problem is relatively simple. The contribution by secondary
scattering from a reflection h,k,0 to another reflection h.k,0 is proportional to
the product of the theoretical intensities

Thizo ° La.thpike tho

and the total contribution from all primary reflections to this reflection is
proportional to the sum of such products over all indices h, and k,

XBT wo * La. rhe +o
A mathematical analysis on this basis leads to an expression for the correction of
observed intensities for secondary scattering. The magnitude of the corrections
and the way in which they account for the intensity anomaly is shown in Figure

44 A. L. G. REES.

a be
‘ ° ® Or SS
Seconcdcory

Origin.

Primory
Oriqin.

Strong Primar
e Rencchon

f7, Corresponding
‘4 Secondary

. f F : Contribution.

Fig. 11.—Diagram showing coincidence of secondary reflections with primary reflections
of different indices in a cross-grating pattern.

5000 |-

@— lobs. against corrected Icale

germ Ce against uncorrected Vente:

4000

] obs

1500

1000 2000 3000 4000 5000
I calc

Fig. 12.—Comparison of observed and calculated intensities of reflections from dicetyl
single-crystal. Satisfactory correspondence of observed and calculated intensities is
obtained only when a correction for secondary scattering is introduced.

ELECTRON DIFFRACTION IN THE CHEMISTRY OF THE SOLID STATE.

Fig. 13.—Basal-plane Fourier projection of potential distribution in dicetyl using intensity
data uncorrected for secondary scattering.

Fig. 14.—Basal-plane Fourier projection of potential distribution in dicetyl using intensity

data corrected for secondary scattering. Full circles denote carbon atom positions and

open circles hydrogen atom positions deduced from data on bond lengths and angles. Note
that the hydrogen atom positions are clearly indicated in the contour map.

45

46 A. L. G. REES.

12. The importance of this correction in structure analysis was tested by
obtaining basal-plane Fourier projections of the hydrocarbon dicetyl (C3.H,¢e.).
The Fourier projections obtained from diffraction of electrons are projections
of the distribution of potential in the lattice and not of the electron density, to
which, however, they are closely related. In Figure 13 the contour map obtained
from uncorrected intensities shows spurious and misleading detail; in Figure 14
the contour map obtained from corrected intensities not only shows the carbon
atom positions but also gives an indication of the hydrogen atom positions.

This particular test structure analysis made it clear that the application
of electron diffraction to structural problems in chemistry had specific merit.
Crystals of very small dimensions only are necessary and light atoms, notably
hydrogen, can be located more readily than by X-ray methods owing to their

Fig. 15.—(111)-Fourier projection of y-Al,OQ,; obtained from intensities in an electron
diffraction cross-grating pattern. This projection must be interpreted in terms of a
stacking disorder in the lattice.

higher relative scattering power for electrons than for X-rays. New problems.
are, however, introduced, a major one being the measurement of intensities.
Intensities must be measured with greater accuracy than in the X-ray method,
but this is not difficult to achieve.

The results of subsequent structural studies by Dr. J. M. Cowley and Mr.
A. F. Moodie have justified development of this method and illustrate its value
in chemistry.

It has been known for some time that the electron diffraction pattern of
polycrystalline gold foil developed several extra reflections if the specimen was
heated at ~500° C. in air or oxygen. Nothing of this sort occurred if larger
Specimens treated in the same way were studied by X-ray methods. Moodie
(unpublished work) succeeded in obtaining cross-grating patterns after protracted
heating of gold films in oxygen and was able to identify several Au-0 phases of
different symmetry from these patterns. The intensity data were complete
enough to obtain Fourier projections of two of these phases. The significant
feature of this result is that gold dissolves oxygen to form well-characterized
phases of low oxygen content, even though on a bulk specimen this may be
confined to a surface layer less than 100 A thick. This is, moreover, a problem

BOS etl oe

fond

ELECTRON DIFFRACTION IN THE CHEMISTRY OF THE SOLID STATE. 44

which could not be investigated by X-ray methods as the low heat of formation
of these phases and the high heat of activation of the diffusion of oxygen into

Fig. 16.—Basal-plane Fourier projection of boric acid crystal showing distribution of potential
in single two-dimensional layer of the lattice. The trigonal groups are BOs each forming

three pairs of hydrogen bonds with the three neighbouring BO” groups. Note the distribu-
tion of potential in the region of the hydrogen-bond pairs.

gold preclude the preparation of any but a small amount of the material in a
reasonable time.

48 AL. Go BEES.

Cowley (unpublished work) has studied the structure of y-alumina and
related compounds. A (111) Fourier projection of a disordered y-Al,O, (cubic,
spinel-type) is shown in Figure 15. The X-ray work on y-Al,O; has been
confined to powder patterns largely because of the small crystal size and the
extensive disorder.

The structure of boric acid has been determined by Zachariasen (1934) by
X-ray methods in so far as the boron and oxygen locations are concerned, but
no information on the location of the hydrogen atoms was obtained. It was
inferred from the intermolecular O-—O distances, however, that hydrogen-bonds
occur extensively in the structure. Cowley (1952) has developed new methods
of dealing with stacking disorder in layer-lattice structures and has been able
to obtain a clear projection of a single two-dimensional atom layer showing

©) BORON @ OXYGEN e 4 HYDROGEN

Fig. 17.—Diagram showing the eight possible canonical structures for a pair of hydrogen bonds
in the boric acid structure and the resulting “‘ resonance ”’ structure.

trigonal BO3- groups and also the hydrogen bonds interlinking these groups.

The projection is shown in Figure 16. It is to be noted that (i) the hydrogen
atom in the bond tends to be near either one oxygen or the other and not
symmetrically disposed between them, and (ii) hydrogen positions do not lie
on the internuclear line between the hydrogen-bonded oxygens. Evidently
the explanation is to be found in thé interaction of the two parallel orthodox
hydrogen bonds. There are, in fact, eight different canonical structures that
may be written down showing different possible positions of the protons. These
structures are given in Figure 17 together with the resultant distribution. This
type of hydrogen-bond ‘‘ resonance ”’ must be common among oxyacids of the
non-metals and probably occurs in protein structures also. This type of
structural study would certainly be impossible by X-ray methods at present.

Of further chemical interest is the fact that structural disorder is common
in small crystals. For layer-lattice structures it is invariably present. It is
possible that the 3-dimensional structure of larger crystals is not the thermo-
dynamically stable configuration when crystal dimensions are small. Moreover,
there is evidence (Rees and Spink, 1950a) that the lattice parameters change
significantly when crystal dimensions are reduced. Further study along these
lines is desirable.

ELECTRON DIFFRACTION IN THE CHEMISTRY OF THE SOLID STATE. 49

FINE STRUCTURE IN ELECTRON DIFFRACTION PATTERNS AND THE
SIZE AND HABIT OF CRYSTALS.

High-resolution diffraction patterns show fine structure of the individual
single-crystal reflections which is closely related to the size and habit of the
crystal. The origin of this fine structure is to be found in two effects, namely
_ (i) the deviation of an electron beam on crossing a crystal boundary (refraction)
(Sturkey and Frevel, 1945; Hillier and Baker, 1945, 1946 ; Cowley and Kees,
1946, 1947; Honjo, 1947), and (ii) the extension of reciprocal lattice points
arising from small crystal dimensions (Rees and Spink, 1950b). These effects
may be discussed independently by means of a kinematic approximation,
adequate for most purposes, or more rigorously by a dynamic treatment.

Since the potential inside a crystal is different from that of free space, the
electron velocity is also different and a refraction phenomenon is to be expected.
Actually the square root of the inner potential is the electron analogue of the
refractive index for light. Angular deviation of an electron beam will occur at

Fig. 18.—Diagram of deviation of an electron beam at entrant and exit faces of a
crystal at diffraction plane. 8,, S,, S, and 8, are beam vectors ; M, and M, are crystal

—

normals; N is the diffracting plane normal; 6 is the Bragg angle.

the entrant and exit faces of a crystal and also, if the diffraction conditions are
satisfied, at the diffracting planes inside the crystal, as shown in Figure 18.
Since the deviations due to refraction are normally very small, electrons entering
non-parallel faces of the crystal may still satisfy the somewhat relaxed diffraction
conditions associated with small crystals. Each single reflection will conse-
quently be broken up into a number of components corresponding to the number
of possible entry and exit faces presented to the beam by the crystal in its given
orientation. Typical spot groups attributable to refraction are shown in Plate IV,
Fig. 19. The spot configuration is characteristic of the orientation and habit
of the crystal.

The effect of crystal shape can be discussed only in terms of the reciprocal
lattice. As we have seen earlier, the shape of the relevant reciprocal lattice
region about an integer point depends on the external form of the crystal and in
fact the distribution of scattering amplitude around each integer point of the
reciprocal lattice is accurately described by the square of a function of the
reciprocal lattice co-ordinates known as the shape transform. This amplitude
is greatest along directions perpendicular to crystal faces and the transform is in
fact a 3-dimensional periodic function. For a plain parallel slab the scattering
amplitude is extended along the c*-direction and is of the form sin? Nxu/(Nzxu)?,

¥F

50 A.D. G. BEES.

where wu is the reciprocal lattice co-ordinate measured from the integer point
and N is the number of scattering points in that direction. Referring now to
Figure 20, in which the situation for refraction and diffraction at (110) planes
of a small crystal of cubic habit is represented in real and reciprocal space, we
see the way in which the more complicated spot groups originate. A contour
map of the distribution of scattering amplitude around a reciprocal lattice point
in a small crystal of cubic habit is shown in Figure 21.. There is now a different
Ewald sphere for each face of entry and each sphere makes a different section
of the amplitude function. Each refraction spot is broken up into a configuration
determined by the shape transform. In some instances the sphere sections the

Fig. 20.—Diagram illustrating splitting of the 110 diffraction spot into group of reflections
by a very small crystal of regular habit (cubic) in real and reciprocal space. The vectors S,

denote beams prior to diffraction, the vectors Si denote beams after diffraction. The
distribution of scattering amplitude (square of the shape transform) around the (110)*
reciprocal-lattice point is shown on a considerably enlarged scale.

transform in such a way as to show the subsidiary maxima, from the positions
of which the dimensions of the crystallite may be determined uniquely (Rees
and Spink, unpublished). Examples of this are shown in Plate II, Fig. 22.
Further development of this could provide one with a powerful method of
Studying crystal growth.

An interesting example is provided by zinc oxide prepared by burning zine
in air or oxygen. The particles grow as ‘“‘ fourlings ’’, four elongated hexagonal
prisms roughly in tetrahedral configuration. Often one finds thin sheets forming
webs between one spine and each of the other three. Further spines may grow
from these sheets. It has been possible to use electron diffraction methods to
provide a complete morphological description of these crystals (Cowley et al.,
1951b). A drawing of an idealized crystal is shown in Figure 23 and an electron
micrograph in Plate IV, Fig. 24. The reason for this fantastic habit is not known,
but it does represent an interesting problem in the mechanism of crystal growth.

The possibility of deducing the shape of the molecules of crystalline proteins
is within reach. These molecules are well ordered units of dimensions of the

ELECTRON DIFFRACTION IN THE CHEMISTRY OF THE SOLID STATE. 51

order of 50-100 A and should give reflections from individual molecules in a
dried molecular dispersion. The water necessary to preserve long-distance
order in protein crystals is intermolecular and drying should not therefore
interfere with the order in each individual molecule. Shape transform. detail
should allow a complete and unique description of the polyhedral shape of the
molecule to be given, an important step in the structure analysis of large complex
molecules.

The rigorous (dynamic) treatment of diffraction (Bethe, 1928 ; Heidenreich,
1950; and Kato, 1951) predicts a different effective inner potential for different

Fig. 21.—Contour map of scattermg amplitude around a reciprocal lattice point
for a crystal of cubic habit.

reflections, a prediction which is confirmed by experiment. Experimental data
on the dimensions and configurations of spot groups such as those of Figure 19
may be used to deduce the Fourier coefficients V,,, of the three-dimensional
potential distribution in the unit cell. Goodman (unpublished work) has been
able to compute the V,,, values for several reflections in MgO from his experi-
mental data and has shown that they are in good agreement with values obtained
from the known structure of MgO. This provides us with a new approach to
structure analysis. The Fourier coefficients may be determined for all observable
reflections by measurements of spot group dimensions and the structure analysis
performed in the usual way. This could be of considerable value for crystals for
which intensity measurements are difficult to make. It is my opinion that this
represents the most promising approach to the problem of structure analysis by
electron diffraction.

52 A. L. G. BEES.

REACTIONS IN THE SOLID STATE.

At some stage in a reaction involving the solid state a new solid phase must
be formed from a solid reactant phase. The mechanism by which this is accom-
plished is a problem of some importance in chemistry. For simplicity we may
examine two types of solid state reactions, namely

(i) tarnish reactions, in which an oxide, sulphide, halide, etc., film is formed
on the surface of a metal by reaction with the corresponding non-metallic
element ; and

(ii) simple dissociation processes, in which an ionic compound is decomposed
by thermal or photo-chemical means into a gas and a solid product.

Fig. 23.--Diagram of ideal crystal form of “‘fourling”’ in ZnO smoke. Three

needles of hexagonal section are twinned on (1122) with the fourth needle (vertical

in the diagram) making an angle of 116° with it. The webs grow as continuations

of the lattice of the vertical needle and have a twin relationship with the three
lower needles.

The nucleation and growth steps are always the result of aggregation of defects
in the crystal, e.g. in simple dissociation reactions the step may be either the
ageregation of interstitial cations or of vacant anion lattice sites trapping
electrons. A schematic representation of the aggregation of vacancy defects
is given in Figure 25. The aggregate of defects is in fact simply a lattice of
cations and electrons, that is, a small crystal of the metal with somewhat enlarged
lattice dimensions and perhaps different symmetry. It is clear that the defect
aggregate will at some point become thermodynamically unstable and break
away from the parent lattice to form a small crystal of the metal of correct
lattice dimensions. The metal lattice will have an orientational relationship
to the parent lattice which will reflect the mechanism of nucleation and this
can be established by electron diffraction studies. Moreover, the size and shape
of the precipitated particles can be deduced from the fine structure.

ELECTRON DIFFRACTION IN THE CHEMISTRY OF THE SOLID STATE. 53.

An illustration of this type of study is Pashley’s (1950, 1951) recent work on
the photolytic and electron-induced decomposition of silver halides. Pashley
was able to show that the lattice of silver produced from silver chloride is in
parallel orientation to the silver chloride lattice, even though the cell dimensions
of the two are such that no good atomic fit is to be expected. AgCl is a cubic
crystal of the NaCl- type; the Agt ions form a face-centred cubic lattice of
side 5:55 A. Removal of Cl- ions by aggregation of F-centre defects will leave
a face-centred lattice of Ag of side 5-55 A, which at some stage collapses to
give the face-centred cubic lattice of silver metal of side 4-08 A without change
of orientation. This particular study is of importance in the mechanism of the

ax *
ae sy ot) Che i 8 a eM AU ew gos Noe iy eA ly 2 A
Ch ot Ate i+ Sala ci res toe ila cs
Sa en See ye Na NOES ON
+—- +- the — + - + -!+teltle iltie]+ —-
> apa X of -~ a-~ NN uL— \ 14> Pd
/ \ _— 4 N ba acta Sach “ye
eed ik i) A Slt Serine tect cee
fans - x / BEE ON eZ ESS ved
4 YY. / woe _—_——— oO NN ee eS,
t ) aa ty th ; 2 / N 7 i
+i e I+ Bs ley ict + =p ye J te Mere Po —
naa / \ — A ale fT Neat Oe ne s
Wes a 1 Md yale Ne Sad Nn
aay Ree a ge - at outa EW eunte ce +
ils ore aN u> 2 Sire A eng a
+{ east Sayed! yi Brains, Teer) Pa osb se eet anny ees ee es es
~— ~ cf
iat a a Ne te peg
+ + +[ © un ar -— + —- + -—- + =— + = +
~~’
Dispersed Defects Agqqreqate

Fig. 25.—Illustration of the aggregation of F-centre defects in an ionic lattice to give a
metallic nucleus.

photographic process. Reactions in the solid state are intimately associated
with the existence of crystal defects, the structural consequences of which are
almost unknown. In my opinion electron diffraction is capable of providing
this information with little improvement on present experimental methods.

CONCLUSION.

Our understanding of the chemistry of the solid state, particularly those
aspects which concern small crystals and defect solids, is still elementary and
any prospect of increasing this understanding is of no small consequence.
Recent developments in electron diffraction show promise of improving this
knowledge and point the way to undoubtedly fertile fields of scientific work.

The work we have done in Melbourne in an effort to pioneer some of these
fields has been carried out by a small group, namely Dr. J. M. Cowley, Messrs.
A. F. Moodie, P. Goodman and J. A. Spink. They are responsible for the
diffraction patterns used for illustration here, and I gratefully acknowledge my
indebtedness to them.

REFERENCES.

Bethe, H. A., 1928. Ann. Physik, 87, 55.
Cowley, J. M., 1952. Nature. (In press.)
Cowley, J. M., and Rees, A. L. G., 1946. Nature, 158, 550.
—_-— -—_________——._ 1947. Proc. Phys. Soc., 59, 287.
—__- 1952. J. Ser. Instr. (In press.)
Cowley, J. M., Rees, A. L. G., and Spink, J. A., 195la. Proc. Phys. Soc., A 64, 609.
1951b. Proc. Phys. Soc., B 64, 638.
Finch, G. I., and Wilman, H., 1937. Hrgebn. exakt. Naturw., 16, 353.
Heidenreich, R. D., 1949. J. Appl. Phys., 20, 993.
——_—_—____—_—_—————. 1950. Phys. Rev., 77, 271.
Hillier, J., and Baker, R. F., 1945. Phys. Rev., 68, 98.
—_——_—— 1946. J. Appl. Phys., 17, 12.

54

A. L. G. REES.

Honjo, G., 1947. J. Phys. Soc. Japan, 2, 133.
Kato, N., and Uyeda, R., 1951. Acta Crystallographica, 4, 227, 229.

Pashley, D. W., 1950. Acta Crystallographica, 3, 163.

1951. ‘‘ Fundamental Mechanisms of Photographic Sensitivity ’’, Butter-
worth, London, p. 39.

Rees, A. L. G., and Spink, J. A., 1950a. Nature, 165, 645.

--—— 19506. Acta Crystallographica, 3, 316.

Sturkey, L., and Frevel, L. K., 1945. Phys. Rev., 68, 56.
Thomson, G. P., and Cochrane, W., 1939. ‘‘ Theory and Practice of Electron Diffraction ”’,

Macmillan, London.

Vanshstein, B. K., and Pinsker, Z. G., 1949. J. Phys. Chem., U.S.S.R., 23, 1058.

Fig.

Fig.

Fig.

Fig.

Fig.

1950. Dokl. Akad. Nauk., S.S.S.R., 72, 53.

Zachariasen, W., 1934. Z. Krist., 88, 150.

e EXPLANATION OF PLATES.

PLaTE II.

2,.—Electron difiraction camera in Chemical Physics Section, Division of Industrial Chemistry,
C.S.1.R.O., Melbourne.

22.—Fine structure of single reflections in patterns from small crystals of ZnO of elongated
hexagonal prismatic habit.

Pruate III.

‘, 7.—Hlectron diffraction pattern from crystals of a long-chain paraffin hydrocarbon. Secondary

scattering is responsible for diamond-shaped groups of reflections and extra spots inside
first strong ring.

. 8.—Electron diffraction pattern from dicetyl showing secondary rings centred on strong

primary reflections. Secondary rings originate from diffraction by numerous small crystals
in random orientation underlying a large crystal giving the strong primary spots.

g. 9.—Electron diffraction pattern from dicety] illustrating the origin of “‘ extra’”’ rmgs. The

envelope (distinguished by an arrow) of the numerous secondary rings, which originate from
either (i) 200 secondaries around 110 primaries, or (ii) 110 secondaries around 200 primaries,
forms a non-legitimate ‘“ extra’’ ring centred on the undeflected beam.

10.—Cross-grating pattern from single-crystal of dicetyl showing effects attributable to
secondary scattering. The odd orders on both axes are forbidden on space-group grounds,
but appear in the pattern as a result of secondary scattering.

PuaTE IV.
19.—Typical spot groups in patterns from MgO crystals of cubic habit. These spot groups,
which comprise the fine structure of ring patterns from specimens consisting of randomly
oriented crystals of regular habit, are attributable in part to refraction effects.
24.—An electron micrograph of an imperfect ZnO ‘“‘ fourling ’’ showing webs. The dotted
outline is a dark-field image of this crystal.

Journal Royal Society of N.S.W., Vol. LXXXVI, 1952, Plate II

gz
he

ck, aa

Fig. 22.

iii
AU Tice ‘i

ay ie

Journal Royal Society of N.S.W., Vol. DXXXVI, 1952, Plate III

Any
TAF as

Journal Royal Society of N.S.W., Vol. LXXXVI, 1952, Plate IV

ae SSBC LASS Scat ELIE AAMOMESI DS pie sere seeeltie egaennecipransnascennsies/suiapronenmaoanscpangronansaneonartitrmoonmernonrances

Atle Se Se Ny ee ee oe me ee
, AN cee eee ore sae ore aa

Fig. 24,

mar,
tis
wk
te,

Mee

he
ate

5
te beled:
ped rier,

PERMIAN SPIRIFERS FROM TASMANTA.

By IbA A. BROWN, D.Sc.
[Mrs. W. R. BROWNE. |

With Plates V and VI.

Manuscript received, July 14, 1952. Read, August 6, 1952.

Abstract.—Descriptions are given of two species of Permian Spirifers
(Brachiopoda) from the Berriedale (or Granton) Limestone near Hobart,
Tasmania; one, Trigonotreta stokesii Koenig, 1825, based on the holotype and
on topotype material collected by the writer, the other, Trigonotreta stokesi
auctt. (non Koenig, 1825), for which the name Grantonia hobartensis gen. et
spec. nov. is proposed. It is shown that Spirifer tasmaniensis Morris, 1845,
is a Synonym for Trigonotreta stokesti Koenig, 1825.

INTRODUCTION.

The early collections of Australian Permian Spirifers made by Robert
Brown, Darwin, Strzelecki, Dana, W. B. Clarke and others were sent to Europe
or America for identification and description and most of them were retained
in museums abroad. Thus the holotypes of many of the most common and
important species of south-eastern Australian fossils have not been re-examined
Since their original descriptions, in some cases more than one hundred years
ago.

In 1938 the writer examined all the known holotypes of eastern Australian
Permian spirifers in museums in England, and during 1946 and 1947 studied the
extensive brachiopod collections from all parts of the world in museums in
U.S.A., particularly at the U.S. National Museum, Washington, D.C.; the
American Museum of Natural History, New York; the Peabody Museum,
Yale ; and the Museum of Comparative Zoology, Harvard. Museum collections
in Australia have supplemented the writer’s own collections made in Tasmania
in January, 1940, and in New South Wales over many sears.

This work has revealed some rather unexpected misconceptions, two of
which are discussed below. These concern specimens collected by R. Brown
and P. E. de Strzelecki.

Robert Brown, naturalist to the Joseph Banks Expedition to New Holland
(Australia) under Capt. Matthew Flinders (1798 to 1802), made a collection of
fossils, which was described briefly by Buckland (1821, p. 481). Apparently
the collection was acquired by Charles Stokes, a member of Council of the
Geological Society of London, who donated it to the Society’s Museum in 1821
(Stokes, 1821 : 1854, xxvi). It was transferred to the British Museum in 1911.
The specimen of Spirifer in this collection was described by C. Koenig (1825)
and named by him Trigonotreta Stokesii, ‘‘ a supra laudato Carola Stokes nobiscum
communicata’”’. A wax cast of the specimen was presented by Chas. Stokes
to the British Museum (Nat. Hist.), specimen B4798, and a plaster cast of this
is in the University of Sydney, by courtesy of the British Museum.

P. E. de Strzelecki collected a number of fossils during his visit to this
country in 1839 to 1843 and the Permian spirifers in his collection were described

56 IDA A. BROWN.

by J. Morris (in Strzelecki, 1845). The specimens are in the British Museum,
and plaster casts of them were presented to the University of Sydney and the
Australian Museum in 1939.

Although frequent reference has been made to the genus Trigonotreta,
no description of the genotype based on the type specimen has been published
since Koenig’s original identification. The misinterpretation of this species by
Morris (1845) has been the cause of much confusion in the literature dealing with
the affinities of the genus and also the recognition of the species.

The following description is based on an examination of the holotype, on a
topotype collection made by the writer, and a study of comparative material
in the U.S. National Museum, Washington, D.C. The two species concerned
both occur in the Berriedale or Granton Limestone of the Granton Stage (Voisey.
1938) of the Permian, in the vicinity of Hobart, Tasmania.

No classification (other than generic) of the spirifers is offered at this stage,
since the writer is of the opinion that such is futile in the present state of know-
ledge of the group. The ‘‘ Classification of the Spiriferide ’’ by Fredericks
(1926) is admitted by him to be purely artificial and unnatural.

The terminology used in this paper is that defined by Cloud, 1942, except
where otherwise stated,

ACKNOWLEDGEMENTS.

The writer wishes to thank all those who have assisted in this work, particu-
larly Dr. W. D. Lang and Dr. H. M. Muir-Wood of the British Museum; Dr.
G. A. Cooper of U.S. Nat. Museum; Prof. C. O. Dunbar, Yale; and Mr. H. O.
Fletcher of the Australian Museum. The late Dr. A. N. Lewis gave valuable
assistance in the field-work in Tasmania and Prof. L. A. Cotton arranged for the
writer’s leave of absence from teaching duties on two occasions for study abroad,
for which she is deeply appreciative.

A grant from the Commonwealth Research Fund administered by the

University of Sydney to cover portion of the field expenses in Tasmania is
gratefully acknowledged.

Genus Trigonotreta Koenig, 1825.

Plate V, figs. 1-5.

KOENIG, C., 1825. IJcones Fossilium Sectiles, p. 3, pl. 6, fig. 70.

Bronn, H. G., 1837. Lethwa Geognostica, Vol. I, p. 77, Tab. II.

KinG, W., 1850. The Permian Fossils of England, Pal. Soc. of London, Vol. 3,
Pt. 1, pp. 81, 122, 126-134.

DAVIDSON, T., 1854. British Fossil Brachiopoda, Pal. Soc. of London, Vol. VII
(for 1853), p. 79.

MEEK, F. B., and HAYDEN, F. V., 1865. Pal. of Upper Missouri, Smiths. Cont.
No. 172, Wash., p. 19.

SCHUCHERT, C., 1893. American Geologist, Vol. XI, p. 141.

HALL, J., and CLARKE, J. M., 1894. Natural History of New York, Palwontology,
Vol. VIII, Brachiopoda, Part II, p. 8.

HALL, J., and CLARKE, J. M., 1894. 13th Annual Report of the State Geologist
of New York (for 1893), Vol. Il, p. 751.

ScHUCHERT, C., 1897. Synopsis of American Fossil Brachiopoda, Bull. U.S.
Geol. Surv., No. 87, p. 127.

BUCKMAN, S., 1908. Brachiopod Homeomorphy: Spirifer glaber, Quart.
Jour. Geol. Soc. London, Vol. 64, p. 30.

SCHUCHERT, C., 1913. In Zittel: Text-book of Paleontology, p. 410.

PERMIAN SPIRIFERS FROM TASMANIA. 57

SCHUCHERT, C., and LE VENE, C. M., 1929. Fossilium Catalogus, Animalia,
Pars. 42, Brachiopoda, p. 125. W. Junk, Berlin.

PAECKELMANN, W., 1932. Neues Jahrb. fur Min., Geol. u. Pal., Ablandl. 67.
BB. Abt. B, Erstes Heft, z. 9,36.

Genotype (by subsequent designation). (?) King, 1850, or Buckman, 1908.

Trigonotreta stokesit Koenig, 1825. Locality, Isle of Van Diemen, New Holland
(=Tasmania).

Diagnosis. Impunctate, spiriferoid brachiopods ; biconvex, width greater
than the length ; high ventral interarea, open delthyrium, strong medial fold in
dorsal valve and corresponding sulcus in ventral valve. Anterior commissure
parasuleate.

Surface ornamentation characteristic: coarsely costate, not fasciculate :
four angular cost on the dorsal fold interlock anteriorly with three strong cost
in the sulcus of the ventral valve ; six or seven primary cost on both sides of
fold and sulcus, the outer of which may bifurcate once only. Concentric growth
lines.

In ventral valve receding dental plates do not reach the floor of the valve,
but support pyramidal articulating processes or teeth, which fit into corres-
ponding sockets in dorsal valve. Cardinal process of dorsal valve sessile, with
myophore vertically striated for reception of diductor muscles.

Heavy deposit of callus in ventral posterior region.

Discussion. Koenig (1825) gave a generic description intended to dis-
tinguish Trigonotreta from Spirifer Sowerby, Pentamerus and other Brachiopoda,
and illustrated his remarks by two figures, ‘“ Fig. 70. TRIGONOTRETA
Stokesw. n.’? from the Permian of Tasmania, and ‘ Fig. 71. TRIGONOTRETA
speciosa. (Terebratulites speciosus Schloth.) ”’ from the Eiffel, without designating
a type species. The latter species has since been assigned to the genus Hystero-
lites Schlotheim.

Bronn (1837) accepted Trigonotreta as a genus distinct from Spirifer and
described additional species.

King (1850) cited Trigonotreta stokestt Koenig as a Permian form in his
tabular ‘‘ Classification of the Various Groups constituting the Class Palio-
branchiata ”’, and this was regarded by Hall and Clarke (1894, p. 8) and by
Schuchert and Le Vene (1929, p. 125) as selection of the genotype, although it is
not always accepted. King (1850, pp. 125-126) considered Anomites cuspidatus
Martin to be the type of Spirifer Sowerby, and thus retained ‘‘ for another group
of Spiriferide, a name which many conchologists have considered a cancelled
synonym. Genus Trigonotreta Koenig.’ He discussed the genus based on
Koenig’s diagnosis and described five other species as belonging to Trigonotreta.

Davidson (1854, p. 79) in England and Meek and Hayden (1865, pp. 17-19)
and others in America have regarded Trigonotreta as a synonym for Spirifer ;
however, Meek and Hayden indicated ‘‘ Spirifer cuspidatus Sowerby ’’ as the
type of Spirifer Sowerby and recognized as a sub-genus ‘“ Trigonotreta Koenig —
Spirifer of most authors . . . Example.—Spirifer striatus, Sowerby.’ Subse-
quent suspension of the Rules (Opinion 100) by the International Commission on
Zoological Nomenclature has fixed the genotype of Spirifer Sowerby as Anomia
striata Martin. (Smithsonian Mise. Coll., Vol. 73, no. 5.)

Buckman (1908, p. 30) recognized Trigonotreta as a valid genus, indicating
T. stokesit Koenig as the type.

Schuchert’s choice (1913, p. 410) of Spirifer aperturatus (Schloth.) as the
genotype of Trigonotreta is invalid under International Rules of Zoological

58 IDA A. BROWN.

Nomenclature, Article 30, e, a, a8 this species was not included under the generic
name at the time of its original publication.

Schuchert and Le Vene (1929, p. 125) cite as ‘‘ Genolectotype (King)
T. stokesti Koenig, 1825 ”’, but place Trigonotreta as a synonym for Spirifer (s.8.)
Sowerby, 1818.

In his discussion of the classification of the Spiriferide King, Paeckelmann
(1932, pp. 36-37) recognizes Trigonotreta as a sub-genus of Spirifer (s.1.) in the
sub-family Elythine Fredericks.

Comparison. Trigonotreta differs from Spirifer Sowerby [genotype S. striatus
(Martin), Lower Carboniferous| in gross form, surface ornamentation of few
coarse coste, and probably in internal characters. It also differs in geological
age. No recent study of S. striatus has been published to the writer’s knowledge.
Martin’s specimen is supposedly lost, although the writer, in December, 1938,
saw in the Department of Geology, Manchester University, England, a collection
labelled ‘‘ Types of (?) Martin ...” containing a specimen which exactly
matched in appearance and dimensions that illustrated by Martin (1809, T.23,
figs. 1-2) from Castleton, Derbyshire. A small specimen collected by the writer
from this locality is shown in Plate VI, fig. 6. As Martin states, the valves are
‘* convex, semicircular, and longitudinally striated on every side. The strize
close, nearly equal, and prominent. In the smaller valve, a convex wave, which
is answered by a concave one (scarcely distinguishable) in the larger valve ;
both terminating in a small wave at the margin.”

The specimen illustrated by Davidson (1857, Pl. II, figs. 19, 20) as Spirifer
striatus var. attenuatus Sowerby is a much more alate form, often mistakenly
called S. striatus in text-books ; it is in the Sedgwick Museum, Cambridge.

Trigonotreta Koenig differs from Neospirifer Fredericks, 1924 (genotype
Spirifer fasciger Keyserling, 1846) by the coarseness of the costz and the lack of
fasciculation. Photographs of specimens of Neospirifer fasciger from the
Upper Productus Limestone, Salt Range, India, and of Neospirifer condor from
the Wolfcamp (Permian) of Texas, U.S.A., are reproduced on Plate V, figs. 7
and 8, for comparison with the Tasmanian spirifers.

Trigonotetra stokesii Koenig, 1825.
Plate V, figs. 1-5.

Trigonotreta Stokesw Koenig, 1825. Icones Fosstlivum NSectiles, p. 3, Pl. 54, fig. 70.
B.M. (N.H.) specimen B4798.

Spirtfer Tasmaniensis Morris, 1845. In Strzelecki, 1845, p. 280, Pl. XV, fig. 3.
B.M. (N.H.) BB6246.

non Spirifer stokesit Morris, 1845. In Strzelecki, ‘‘ Physical Description of
New South Wales and Van Diemen’s Land ”’, p. 283, Pl. XV, figs. 1 and la.
B.M. (N.H.) 96859.

non Spirifer stokesii d’Orbigny, 1846. In Dumont d’Urville, ‘‘ Voy. au Pole
Sud ’’, Géologie, Atlas, t. 9, ff. 12-14.

non Spirifer stokesii Eth. fil., 1892. Geol. and Pal. Queensland and New Guinea,
Dp. 232, Pl. LO nes... 2). 3.

Holotype. Brit. Mus. (Nat. Hist.) specimen B4798. Plastotype: Univ. of
Sydney, Spec. 8469.

Locality. ‘Isle of Van Diemen, New Holland ”’ (Tasmania).

According to Buckland (Trans. Geol. Soc., 1st Series, Vol. V, 1821, p. 481)
this or a similar specimen came from “ the south side of the Table Mountain,
near Hobart’s Town, Van Diemen’s Land.”

PERMIAN SPIRIFERS FROM TASMANIA. 59

Horizon. Berriedale (Granton) Limestone, Granton Stage (Voisey, 1938).
Permian.

Description of Holotype. The specimen is a somewhat crushed external
cast of the dorsal valve and the cardinal area and beak of the ventral valve of a
typical Spiriferoid shell (Plate V, fig. 1) embedded in rock.

The dorsal valve is 3-4 cm. in length and 5-0 cm. in greatest width, slightly
below the hinge-line, which is long (4:5 cm.) and straight. The cardinal
extremities are slightly rounded. There is a marked median fold, which is
almost carinate posteriorly, but wide at the anterior margin; radiating costz
are superimposed on the median fold and six or seven folds occur on each side
of it. At a distance of about 12 mm. from the beak smaller secondary ribs
appear on one side of most of these lateral costz. The alar portions of the shell
show less pronounced folding. Concentric growth lines occur close together
over the anterior and lateral parts of the shell and they indicate the existence
of a pronounced broad sinus in the anterior commissure, otherwise obscured by
rock matrix.

The ventral valve is shown by its sharply pointed and overhanging beak and
the cardinal area, which reaches a height of 8 mm. in the centre. The features
of the delthyrium are not clearly shown in this specimen: the sides and base
of the triangular delthyrium are each about 9 mm. and its remarkable size
evidently inspired the name Trigonotreta.

The writer has numerous specimens from various localities along the outcrop
of the Berriedale Limestone including an almost perfect specimen from the slopes
of Mt. Dromedary, on the north side of the Derwent River, three miles N.W. of
Bridgewater, near Hobart (Plate V, figs. 2a-2d). This specimen is not crushed
(as indeed the majority of specimens are), and the cost of the dorsal valve can
be matched exactly against those of the holotype (Plate V, fig. 1). The ventral
valve of this specimen shows the presence of a deep sulcus containing three
prominent cost, each lateral part of the shell being ornamented by about seven
coste, the outer of which tend to bifurcate halfway to the margin ; anterior
commissure parasulcate. This valve can be matched exactly with the specimen
(Brit. Mus. (N.H.) BB6246) of a ventral valve figured by J. Morris (1845, Pl. XV,
fig. 3) as the type of Spirifer Tasmaniensis (see Plate V, fig. 5).

Thus S. Tasmaniensis Morris, 1845, becomes a synonym for Trigonotreta
stokesvt Koenig, 1825.

The identification of S. tasmaniensis auctt. is the subject of further investiga-
tion. A common species in the Lower Marine of New South Wales often referred
to this species is not the same as the Tasmanian form.

Morris was in error therefore in the identification of a Spirifer showing
fasciculate ornamentation, which occurred with Trigonotreta stokes, and which
he mistook for it. This form is described below as a new species.

The internal characters of 7. stokesit have been studied by means of various
preparations. Plate V, fig. 3, shows the interior of a typical ventral valve :
the wide, open delthyrium is bounded on its inner surface by the dental plates
which converge towards the inside of the valve, but they do not reach the floor
of the valve. At the cardinal margin they produce small, pyramidal projections,
which fit into the sockets of the dorsal valve. The old positions of these teeth
form narrow triangular structures or false areas on each side of the delthyrium,
but these are not true deltidial plates. No supporting or apical plates are
present. The adductor muscle-scars are long and narrow and are placed slightly
posteriorly ; they are surrounded by the diductor sears, which are relatively
large. The whole of the region below the palintrope is filled with a dense deposit

60 IDA A. BROWN.

of secondary shell, and the delthyrial cavity is almost half-filled with callus.
This covers over the trace of the muscle-track, which appears as a dark line in
sections of the shell (Plate V, fig. 4).

The interior of the dorsal valve shows widely divergent crural plates bounding
the inner sides of the sockets. The crura give rise to descending lamelle, thin
ribbon-like structures supporting the spiralia, which are directed postero-
laterally towards the ends of the hinge-line. There are about twenty turns in
each spire. There is no jugum.

The cardinal process is sessile and the myophore is vertically striated for
the reception of the diductors. The adductor scars are long and narrow,
situated high inside the fold.

Genus Grantonia Brown, gen. nov.
Plate VI, figs. 1-8.

Genotype. Grantonia hobartensis Brown, sp.n.

Diagnosis. Large spiriferoid shell; strongly biconvex, length about equal
to the width, somewhat greater than the thickness ; hinge-line sub-megathyroid,.
cardinal extremities obtuse; beak of the ventral valve prominent, incurved ;
wide ventral inter-area, striated parallel to the hinge-line, but curved in section.
at right angles to it; large open delthyrium, unmodified in adult. High,
almost carinate, median fold in dorsal valve and deep broad sulcus in ventral
valve. Multicostate over the entire shell, three prominent folds or plice each
bearing three or five fasciculate coste on each side of the sulcus, with corres-
ponding ornamentation on the dorsal valve, giving the shell a rough appearance.
Three primary plications in the sulcus. Concentric lamella most pronounced
near the margin of the shell. Fine surface ornamentation, seldom preserved, of
radiating costelle and concentric file, producing a cancellated effect. Anterior
commissure modified uniplicate, or parasulcate. Shell impunctate.

Interior of the Ventral Valve. Cardinal teeth supported by receding dental
plates or delthyrial ridges, which do not reach the floor of the valve ; no support-
ing accessory plates or apical lamelle. Adductor muscle-scars centrally placed,
long, narrow; diductor muscle-scars large, heart-shaped, situated slightly
above the centre of the valve. Pallial markings, pittings in the posterior region.

Interior of Dorsal Valve. Cardinal process sessile, myophore vertically
striated for the reception of diductor muscles; adductor muscle-scars long,
narrow. Socket plates divergent, convex to the median plane of the shell,
supported only by adventitious tissue below the palintrope. The crura give
rise to parallel descending lamella, with no jugum, which support spiralia
directed postero-laterally towards the ends of the hinge-line. Twenty to twenty-
three turns in each spire.

Discussion. Grantonia is distinguished from Trigonotreta Koenig and
Spirifer Sow. by major differences in gross form and surface ornamentation of
fasciculate costae. From MNeospirifer Fredericks, 1924 (genotype Spirifer
fasciger Keyserling, 1846) it is distinguished by the kind of fasciculation. The
type of Neospirifer comes from Pechora Land, and the genus has a wide distribu-
tion over Eurasia, North America, Timor, Western Australia and Tasmania.
The generic characters are discussed in some detail by Dunbar and Condra
(1932), who give the diagnosis: ‘‘ Shells differing from true Spirifer (Spirifer
striatus stock) in having the ribs fasciculate.”” However, examination of
specimens from Russia, and from the Salt Range, India, and the Permian of
Texas, U.S.A., as illustrated on Plate VI, figs. 7, 8, shows that the fasciculation.
in the shells of the two genera is quite distinct. Whereas in Neospirifer the
numerous coste are finely and evenly distributed over the major folds of a thin

PERMIAN SPIRIFERS FROM TASMANIA. 61

shell, and may even be reflected on the inner surface of the shell, in Grantonia
there are only one or two pairs of costze about a central one, superimposed on
the surface layers of the principal folds of a thick shell.

Grantonia hobartensis Brown, spec. nov.
Plate VI, figs. 1-9.

{?) Spirtfer trapezoidalis G. Sowerby, 1844. In Darwin, ‘‘ Geological Observa-
tions in Volcanic Islands ...”’, p. 159. (Specimen lost).

Spirifer Stokes Morris, 1845. In Strzelecki, ‘‘ Physical Description of New
South Wales and Van Diemen’s Land ’’, p. 283, Pl. XV, figs. 1, la, B.M.
(N.H.) 96859.

Spirifer stokesti d’Orbigny, 1846. In Dumont d’Urville, ‘‘ Voy. au Péle Sud ”’,
Géologie, Atlas, t. 9, ff. 12-14.

Spirifer crassicostatus Jukes, 1847. (Nomen nudum) Quart. Journ. Geol. Soc.
Lond., Vol. III, pp. 248, 249, B.M. (N.H.) 81606.

Holotype. Aust. Mus. specimen F44666, Coll. I.A.B. (Plate VI, figs. la, 16).

Locality. Rathbone’s Quarry, New Norfolk Road, near Granton, about
12 miles N.N.W. of Hobart, Tasmania. On aerial photo No. 9744, Run 1,
Hobart (Land and Surveys Dept., Tasmania) appears about 5 cm. north of
centre point.

Horizon. Berriedale or Granton Limestone, Granton Stage (Voisey, 1938),
Permian.

Description of Holotype. The specimen is well-preserved shell consisting
of a complete dorsal valve and the posterior half of the ventral ee (Plate VI,
fig. 1).

The dorsal valve is 54 mm. wide and 34 mm. long. The hinge-line is the
full width of the shell. There is a high median fold and there are three prominent
folds on each side of it. There are four coste on each flank of the median fold,
and the lateral folds bear one pair of coste on their flanks. Several minor cost
occur on the alar part of the shell, and the whole surface is ornamented by
concentric growth lines.

The ventral valve, 45 mm. long, has a long, wide cardinal area, divided by
a large open delthyrium. The beak is high and curved over that of the dorsal
valve. A deep sulcus is flanked by three marked folds corresponding with those
of the dorsal valve, and with similar ornamentation. The sulcus bears three
primary cost in addition to the pair belonging to the lateral folds.

Morris (1845) gave a brief description and illustrations (Pl. XV, figs. 1, 1a)
of specimens of this species from Mt. Dromedary, Tas., under the title Spirifer
Stokesit (p. 283). These specimens are both in one block of limestone (B.M.
(N.H.) 96859), a plaster cast of which is in the Australian Museum, Sydney.

Etheridge (1892) followed Morris in identifying specimens of Grantonia
hobartensis as Spirifer stokesu, and gave a good description of specimens from
the Gympie Beds, Queensland ; specimens in the Australian Museum from
Mt. Britton, Q., appear to the present writer to show some variation from the
Tasmanian material.

Preparations have been made from topotypes to show all the internal
characters of the shell. The internal structures of the ventral beak (Plate VI,
figs. 3b, 3c, 6, 7 and 8) and the cardinalia (Plate VI, figs. 4a and 5) are essentially
similar to those already described for Trigonotreta stokesii Koenig, and are
undoubtedly of family value in the classification of these forms (see Ulrich and
Cooper, 1938, p. 6; Cloud, 1942). The receding dental plates (Plate VI,

62 IDA A. BROWN.

figs. 3b, 6) do not reach the floor of the valve, but heavy callus fills the umbonal
cavities and part of the delthyrial cavity (Figs. 3b and 7). This is best shown in
the thin section of the beak of a ventral valve (Fig. 8); the covered track of the
muscle-scars appears as a horizontal dark band within the secondary shelly
material. The cardinalia are shown in Fig. 4a, and Fig. 5 shows the natural
mould of the striated myophore of the cardinal process.

Serial sections of an average uncrushed specimen revealed that the
brachidium consists of two descending lamella about 5 mm. apart and 20 mm.
in length from which arise two postero-laterally directed spiralia, each with
twenty-three turns in the spire.

Internal moulds of Grantonia hobartensis show distinctly the three lateral
folds on each side of the medial fold and the sulcus, and thus may be distinguished
from internal moulds of Trigonotreta stokes, which are relatively smooth.

REFERENCES.

Buckland, W., 1821. Trans. Geol. Soc. Lond., 1st Ser., 5, 481.

Cloud, P. E., 1942. Geol. Soc. Amer., Spec. Pap. No. 38.

Davidson, T., 1857. Pal. Soc. Lond., 10 (for 1856), Pl. II, 19-20.

Dunbar, C. O., and Condra, G. E., 1932. Bull. Geol. Surv. Nebraska, 5 (2nd Ser.), 326.

Fredericks, G., 1924. Russie Comité Géologique, Bull. 38 (3) (for 1919), 311.

Keyserling, A. G., 1846. Wussenschaftliche Beobachtungen auf eine Reise in das Petschora-Land,
p. 231, Tab. VIII, figs. 3, 3a, 3b.

Koenig, C., 1825. Icones Fossiliwm Sectiles, p. 3, Pl. 6, fig. 70.

Martin, W., 1809. Petrificata Derbiensia, Pl. 23, figs. 1-2.

Morris, J., 1845. In Strzelecki, 1845.

Stokes, C., 1821. Trans. Geol. Soc. Lond., 1st Ser., 5, 560.

—-—-—_— 1854. Obituary Notice, Q.J.G.S., Lond, 10, xxvi.

Strzelecki, P. E. de, 1845. Physical Description of New South Wales and Van Diemen’s Land.

Ulrich, E. O., and Cooper, G. A., 1938. Geol. Soc. Amer., Spec. Pap. No. 13, p. 6.

Voisey, A. H., 1938. Proc. Linn. Soc. N.S.W., 63, 309.

EXPLANATION OF PLATES.
All photographs were taken by the author, and are natural size, except Plate VI, figs. 5,
6 and 8, which are mag. x3.
Tasmanian figured specimens are in the Australian Museum, Sydney.

PLATE V

Fig. 1.—Trigonotreta stokes Koenig, 1825. Plaster cast of holotype, B.M. (N.H.) 4798.
““TIsle of Van Diemen, New Holland.”’

Fig. 2a-d.—T rigonotreta stokesiz. Aust. Mus. (F44663). Coll. 1.A.B. Near Old Lime Kiln,
Upper Dromedary Rd., 3 mls. N.W. of Bridgewater, Derwent River, Tasmania. (a) Dorsal view.
(b) ventral, (c) side, and (d) anterior views, showing anterior commissure.

Fig. 3.—Trigonotreta stokesii. Interior of the ventral valve, showing open delthyrium,

dental plates and teeth, muscle scars and pallial markings. A.M. (F44664), Rathbone’s Quarry,
Granton.

Fig. 4.—Trigonotetra stokesii. Vertical median section of ventral valve to show callus in
delthyrial cavity, also dental plate. A.M. (F44665), Rathbone’s Quarry, Granton.

Fig. 5.—Trigonotetra stokesii. Plaster cast of ventral valve, B.M. (N.H.) 6246, figured by
Morris as Spirifer Tasmaniensis on his Pl. XV, fig. 3.

Fig. 6.—Spirifer striatus (Martin). A.M. (F44669). Coll. I.A.B. Loc. Treak Cliff, near
Castleton, Derbyshire, England. (Lower Carb.)

Fig. 7.—Neospirifer fasciger (Keyserling). Ventral valve. Syd. Univ. Coll. 9487. (Geol.
Surv. Coll. K7/353). Loc. Warcha, Salt Range, India. (Upper Productus Limestone.)

Fig. 8.—Neospirifer condor (Orbigny). Loc. N.W. of Marion School, Texas, U.S.A. (Early
Wolfcamp, Permian.)

Pirate VI.

‘Grantonia hobartensis gen. et spec. nov.

Figs. la, 1b.—Holotype. Aust. Mus. (F44666). Coll. I.A.B. Loc. Rathbone’s Quarry
New Norfolk Road, near Granton, about 12 miles N.N.W. of Hobart, Tasmania. x 1.

Journal Royal Society of N.S.W., Vol. LXXXVI, 1952, Plate V

Journal Royal Society of N.S.W., Vol. LXXXVI, 1952, Plate VI

PERMIAN SPIRIFERS FROM TASMANIA. 63

Fig. 2.—Rubber mould of portion of ventral valve showing details of external ornamentation.
A.M. (F44670). Loc. 1,000 feet above s.l. on Huon Road from Hobart. Coll. I.A.B. Mag. x1.

Figs. 3a-d.—Ventral valve showing (3a) exterior, (3b) interior, and (3c, 3d) posterior views.
A.M. (F44671). Rathbone’s Q. Mag. x1.

Figs. 4a, 4b.—Dorsal valve showing (4a) interior, and (4b) exterior views. A.M. (F4466 ).
Mag. x1.

Fig. 5.—Natural mould of cardinal process, showing vertical striations. A.M. (44668).
Rathbone’s Quarry. Mag. x3.

Fig. 6.—Portion of beak of ventral valve (left) and dorsal valve (right) showing curved dental
plate over callus fitting into socket behind crus of dorsal valve. Mag. x3.

Fig. 7.—Vertical median section of ventral valve, showing callus in delthyrial cavity, A.M.
(F44672). Mag. xl.

Fig. 8.—Thin section through beak of ventral valve, showing folding and fasciculation of
outer layers of shell, and microscopic structure of callus. Divergent receding dental plates and
trace of muscle-track, sub-parallel to the sulcus. A.M. (F44673). Mag. x3.

AUSTRALASIAN Mepicat PuBLisHine Company LIMITED

Seamer and Arundel Streets, Glebe, N.S.W.
1953. |

PART I

=

JOURNAL AND PROCEEDINGS

OF THE

ROYAL SOCIE
OF NEW SOUTH WALES

1952

(INCORPORATED 1881)

he
ee 3 "1

*

PART III
er |
VOL. LXXXVI

= Containing Papers read in October, 1952 C

EDITED BY

IDA A. BROWNE, D.Sc.

Honorary Editorial Secretary

| THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE
_ STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN

ae SYDNEY :
_ PUBLISHED BY THE SOCIETY, SCIENCE HOUSE
GLOUCESTER AND ESSEX STREETS

1953

CONTENTS — SS aaa ea
ooo VOLUME LXXXVI ee
i A ietid; its dys se alee . Y
PAS oe oa ae
Ree | ee Ea aa Part III io fe
cs ee vere
Aa Vit A Beiibehd ptical Activity of the Tris-1 : 10-Phenanthroline and Tris-2:2’- §
yl Copper II Ion. By N. R. Davies and F. P. Dwyer .. a Rea oe, y
i Art. IX.—Soil Horizons and Marine Bands in the Coastal Limestones of Western : ae
i: Australia. By R. W. Fairbridge and C. Teichert a ae 4 we EGS:
d ie
t

JOURNAL AND PROCEEDINGS

OF THE

mOYAL SOCIEIS

OF NEW SOUTH WALES

FOR

1952

(INCORPORATED 1881)

ae

VOLUME LXXXVI
Part III

ee ened

EDITED BY

IDA A. BROWNE, D.Sc.

Honorary Editorial Secretary

ere ES

THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE
STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN

SYDNEY
PUBLISHED BY THE SOCIETY, SCIENCE HOUSE
GLOUCESTER AND ESSEX STREETS

INDUCED OPTICAL ACTIVITY OF THE TRIS-1 : 10-PHENANTHROLINE
AND TRIS-2: 2'-DIPYRIDYL COPPER II ION.

By N. R&R. DAVIES, B.Sc., F.B.L.C.
and Ff. P. DWYER, D.Sc.

With one Text-figure.

Manuscript received, August 21, 1952. Read, October 1, 1952.

Hexacovalency is relatively rare among copper II compounds and with
the exception of anhydrous copper II chloride and bromide, which in the solid
state show a weak octahedral arrangement of the halogen atoms about the metal
(Wells, 1947 ; Helmholtz, 1947), appears to be confined to complexes containing
nitrogen, such as hexammine copper II iodide (Peyronel, 1941), and hexakis
pyridine copper II nitrate (Pfeiffer and Pimmer, 1905), both of which are
relatively unstable. Chelate derivatives with ethylene diamine have been
prepared, but in solution appear to eliminate one molecule of the ligand and
pass to the familar 4-covalent planar bis-ethylene diamine ion (Dubsky and
Tritilek, 1934). However, Rosenblatt (1932) concluded from absorption spectral
curves that the ionic species [Cu(en),(H,O),|+*+ exists in aqueous solutions and
Wahl (1927) claimed to have resolved this cation through the tartrate. This
claim was not substantiated by later workers (Johnston and Bryant, 1934).

When 2: 2’-dipyridyl or 1:10-phenanthroline are added to solutions of
copper IT salts, the deep blue colour of the bis chelate complex ion results, but
with excess of the bases the colour then lightens and stable pale blue salts,
Cu(dipy);X>, Cu(phenan),X,, can be isolated (Jaeger and van Digk, 1934).

The octahedral configuration of the metallic atom in these complexes could
not be demonstrated by resolution through the tartrate, bromcamphor sulphonate
or antimony tartrate. With the latter anion, which is particularly suitable
for the resolution of phenanthroline- metallic complexes (Dwyer and Gyarfas,
1950), antimony oxide rapidly precipitated, showing that extensive dissociation
of the complex cation had occurred in solution.

Pfeiffer and co-workers (1932) noted considerable rotational changes when
various optically active anions were added to solutions containing the ion
M.(A)3++, (M=Zn, Cd, Ni, Fel!, Coll, Mn!!; A=dipy. phenan.) and ascribed
the rotation to excess of either the dextro or levo complex cation in the mixture.
Recently, following the work of Turner and Harris (1948) on the phenomenon
of asymmetric induction in organic compounds, the above changes in rotation
have been, related to the movement of the equilibrium d*++tzl** consequent on
differential activity changes on the enantiomeric ions with the added optically
active anion (Dwyer, Gyarfas and O’Dwyer, 1951).

Since the iron and nickel compounds can be resolved, but are optically
labile, as demanded by this theory, rotational changes of the kind observed by
Pfeiffer can be regarded as diagnostic of potentially active complex salts that
are too labile to be separated into discrete enantiomeric forms. The time
required to reach the maximum rotation change after addition of the optically
active anion can be used to determine whether resolution is practicable. Thus
the zine and cadmium compound, which showed the effect instantly, has

INDUCED OPTICAL ACTIVITY. 65

resisted all efforts at resolution, whilst the nickel phenanthroline compound,
which reaches the maximum rotation only after several days, has a half-life for
the active forms of 16 hours at 25° C.

In the present work the equilibrium :
CuA,*++A 2 d CuA,tt = 1 Cu(A,)tt

(A=1:10-phenanthroline; 2: 2’-dipyridyl; ethylene diamine) has _ been
studied in the presence of ammonium d@ bromcamphor sulphonate, ammonium d
camphor sulphonate and d trisethylene diamine cobalt III nitrate.

Addition of the phenanthroline and dipyridyl copper complexes to
ammonium bromcamphor sulphonate solution led to marked increases in the
rotation. The change was increased with increasing concentration of the free
chelate bases or of the ammonium bromecamphor sulphonate. Since the effect
was immediate, it can be concluded that the enantiometric forms of the copper
compounds are too labile to be separated. No rotational change was observed
in the presence of ammonium camphor sulphonate or tris-ethylene diamine
cobalt III nitrate. Negative results were obtained in all experiments using the
ethylenediamine copper complex. This could be due to the fact there was no
appreciable concentration of the tris ethylenediamine copper II ion in solution
or to the necessity of using very dilute solutions owing to the intense colour
of the complex.

An appreciable lightening in colour was also observed when ammonium
bromcamphor sulphonate was added to the phenanthroline and dipyridyl
complexes. This effect did not occur when ammonium camphor sulphonate
or the optically inactive ammonium (8 naphthalene sulphonate were added.

Absorption spectral curves of solutions containing various ratios of 1:10
phenanthroline to cupric ion were found to show the development of a pro-
nounced maximum at 520u and a weaker maximum of 475-480u due to the
tris-1 : 10 phenanthroline copper II ion. Ammonium bromcamphor sulphonate
added to solutions containing both the bis and the tris phenanthroline complexes
led to an appreciable enhancement of the maxima due to the tris phenanthroline
complex ion.

EXPERIMENTAL.

The observations were carried out at 17° C. in a 1 d.m. tube using either a
sodium amp ora mercury lamp, with filter to isolate the green line 5461, as light
‘ sources. The copper complex solution was placed in a polarimeter tube, and in

TABLE I.
Conc. Brom. Change in Rotation.
Phenan. Added. Camphor
Mols. Sulphate.
% Nap. | Hi 5161:
3 0-5 0-03° =
6 0-5 0-05 —
6 CD) 0-09 0-11°
Dipyridyl added
Mols.
3 0-5 0-02 0-03
3 2-0 0-04 0-09

Initial rotations of ammonium d bromcamphor sulphonate solutions :
0-5%; op —0-43° 3 1%, ap=0-85° 5 W5ag3— 1-04"; 2%, ap =1-70°;
Ore OO.

66 DAVIES AND DWYER.

a second tube the optically active salt solution, diluted with an equal volume of
water. The rotation of the optically active salt was determined with both tubes
in the polarimeter. In this way corrections were made for errors due to lack of
uniformity in the illumination of the field of the polarimeter. Equal volumes
of the copper complex and the optically active salt were then mixed and the
rotation again measured. The difference between the readings was ascribed
to the optical activity of the complex.

Copper sulphate solution, M/100 was used in all experiments and treated
with solid 1:10 phenanthroline or 2: 2’ dipyridyl in order to obtain various
molecular proportions of salt and chelate groups. The results are shown in
Table 1.

Transmission curves in the range 400-700u are shown in Fig. I.

Absorption Curves of Copper Sulphate and 1:10 Phenanthroline.

TRANSMITTANCE

WAVE-LENGTH IN MILLIMICRONS

Fig. I.
(a) System CuSO,+2 phenan.
(b) System CuSO,+3 phenan.
(c) System CuSO,+6 phenan.
(d) System CuSO,+3 phenan-+Ammonium-bromcamphorsulphonate. —
All solutions were 0:01 molar with respect to copper.

SUMMARY.

Addition of solutions containing the tris-1:10-phenanthroline and tris-
2: 2'-dipyridyl copper II ions to ammonium d bromcamphor sulphonate solutions
lead to appreciable increases in the rotation. The increases are roughly pro-
portional to the concentration of the bromcamphor sulphonate and to the excess
of the chelate bases. The rotational changes have been ascribed to the movement

INDUCED OPTICAL ACTIVITY. 67

of the equilibrium d complex=l complex and to the existence of optically labile
forms of an octahedral complex copper ion. This conclusion is supported by
absorption spectral curves.

REFERENCES.

Dwyer, F. P., and Gyarfas, E. C., 1950. THis JourNAL, 83, 263.

Dwyer, F. P., Gyarfas, E. C., and O’Dwyer, M. F., 1951. Nature, 167, 1036.
Dubsky, J. V., and Tritilek, J., 1934. J. Prkt. Chem., 2, 140, 47.
Helmholz, L.,.1947. J. Am. Chem. Soc., 69, 886.

Jaeger, J. M., and van Dijk, J. A., 1934. Proc. K. Akad. Amst., 37, 395.
Johnston, C. H., and Bryant, S. A., 1934. J. Chem. Soc., 1783.

Peyronel, G., 1941. Gazze Ha, 71, 363.

Pfeiffer, P., and Pimmer, V., 1905. Z. anorg. Chem., 48, 98.

Pfeiffer, P., and Quehl, K., 1932. Ber., 65, 560; 66, 415.

Rosenblatt, F., 1932. Z. anorg. Chem., 204, 351.

Turner, E. E., and Harris, M. H., 1948. Chem. Soc. Quart, Reviews, 1, 299.
Wahl, W., 1927. Soc. Sct. Fennica, 4 (14), 1.

Wells, A. F., 1947. J. Chem. Soc., 1670.

Departments of Chemistry,
N.S.W. University of Technology and
University of Sydney.

SOIL HORIZONS AND MARINE BANDS IN THE COASTAL
LIMESTONES OF WESTERN AUSTRALIA.

By RHODES W. FAIRBRIDGE
and CURT TEICHERT.

With Plates VII and VIII and five Text-figures.

Manuscript received, July 29, 1952. Read, October 1, 1952.

CONTENTS.
Page
I. Introduction ae oye mis ae _ sls ls Zt 68
II. Soil Horizons... wa ws at ah we 71
(a) Hamelin Bay and Island .. ap ss ai ae 71
(b) Other Fossil Soil Localities .. - ee oe ae 74
III. Marine Bands .. sis me es ays mba teen
(a) Cowaramup Bay At bts oie!) EE
(b) Other Occurrences of Marine “Members ie ais ee 77
IV. Interpretation of the Aolianites and Their Soils a so ae
V. Correlation with et and North Africa ae P le 80
Conclusions a i per Ph 82
Bibliography Ake ap ae Me as ee ws a OS
ABSTRACT.

Formed under eustatic and climatic controls, a cyclic sequence of marine
bands, zolianites, and travertine crusts in Western Australia correspond to the
Riss/Wirm interglacial, the three substages of the Wurm, and the post-Glacial
(Flandrian) transgression.

Dune building took place with periods of high eustatic sea level, each of
which was relatively short and coincided with the beginning of each arid inter-
stadial. During each: succeeding eustatic low (corresponding to a glacial
maximum) the base-level was lowered and rapid consolidation took place with
travertine crust and karst formation. Thus conglomerates of the latter
mark the following marine invasion. In the south, as at Cowaramup Bay,
only the youngest marine band and eolianite are represented, resting with
a conglomerate of granite-gneiss boulders on the pre-Cambrian basement.

Numerous horizons of immature soils intercalated in each eolianite, as at
Hamelin Bay and elsewhere, do not represent interglacial pauses, as postulated
for similar soils in Bermuda by Sayles, but merely minor oscillations in short
periods of dune building.

I. INTRODUCTION.

The ‘‘ Coastal Limestone’ of Western Australia is a formation of more
than local interest. It is of Pleistocene age and includes marine and continental
sediments which reflect the extraordinary changes of climate and the eustatic
oscillations of sea level characteristic of that epoch. Since Western Australia
is largely a great pre-Cambrian shield and suffered no glaciation whatever during
the Pleistocene, its coasts are not unnaturally regarded as amongst the most

SOIL HORIZONS AND MARINE BANDS. 69

stable in the world. It is here then that we may expect to find traces of
Pleistocene eustatic events most faithfully recorded.

The Coastal Limestone is a rock essentially similar to that for which Sayles
(1931) in Bermuda introduced the term ‘ calcareous eolianite ’’, that is to say, a
lithified dune-rock. For general descriptions of the Western Australian
eolianites see Teichert (1947a, 1947b) and Fairbridge (1950). It is, as a rule, a
medium-grained clastic sediment, in which the grains consist mainly of frag-
mental calcareous alge, mollusca, foraminifera, and bryozoa, with varying
amounts of inorganic constituents and a cement of secondary calcium carbonate.
The CaCO; may vary as much as 10 to 90 per cent. Nevertheless, the rock
reacts to weathering as a ragged limestone, almost regardless of the amounts of
insolubles, which normally increase considerably in the proximity of the pre-
Cambrian basement. Intercalated in the dune rock are found prominent bands
of red and brown fossil soils (terra rossa and rendzina), and in the lower parts
of the formation there are generally wedges of shallow-water marine beds in
reef, beach or estuarine facies.

Identical formations are found on Bermuda (Sayles, 1931) on the Bahamas,
Barbados, Madeira, Cape Verde Islands, at Alexandria, and on the coasts of
Palestine, of Morocco, of Ecuador, of South Africa, on the south coast of Arabia
and on the western shores of India, in Hawaii, and elsewhere.

In Australia the Coastal Limestones are found intermittently along 4,000
miles of coastline and are of great value to those interested particularly in the
last 50,000 years of our history. In South Australia they are found on Eyre
Peninsula, Yorke Peninsula, Kangaroo Island, and in south-eastern South
Australia (Tindale, 1947 ; Hossfeld, 1950; Sprigg, 1952). Their characteristics
are similar to those of Western Australia. Fossil soils developing on them are
mostly of the terra rossa type, but rendzinas occur intermixed with them on
Eyre Peninsula (Crocker, 1946). In Victoria both exolianites and fossil soils are
known from Portland Bay to Barwon Heads (Hills, 1939; Coulson, 1940 ;
Gill, 1943; Keble, 1946). Johnston’s (1888) ‘‘ Helicidz Sandstones ”’ of the
Bass Strait Islands are the same.

This paper will be devoted largely to a description of certain features of
the Western Australian Coastal Limestones where they overlie a ridge of pre-
Cambrian granitic gneiss, extending in a north-south direction for 60 miles
between Cape Naturaliste (33° 32’8., 115° 00’ EH.) and Cape Leeuwin (34° 23’8.,
114° 38’ EK.) and rising to 750 feet in places (Fig. 1).

The presence of Coastal Limestone here was reported as early as 1849 by
J. W. Gregory, Senr., who described the lower part as a ‘‘ grey marble ’”’ with
marine shells and with 2-3 foot blocks of gneiss cemented to the basement in a
coarse conglomerate. He also mentioned ‘‘ the remains of an ancient sea-beach
near Cape Naturaliste’’ and its water-worn boulders, a statement which was
later misquoted as ‘‘ 15 miles south (sic) of Cape Leeuwin ”’ by F. T. Gregory
(1861) and others. The general relationships were seen again by H. Y. L. Brown
(1872), who observed the garnet-gneisses of the pre-Cambrian basement. E. T.
Hardman (1884) expressed the opinion that the overlying limestones were
Cretaceous or Eocene in age and noted the presence of caves therein, which
would be likely spots for digging for extinct marsupial remains. As predicted,
they were later found in great quantities (see Glauert, 1910). An Eocene age
was also attributed to the ‘‘ coast limestone ’’ by H. P. Woodward (1894), and
in another report he mentions sending the marine fossils from the lower part to
England for description.

The first general account of the geology of this area came from E. 8. Simpson
(1902), who recognized the mainly eolian character of the Coastal Limestone
and its interesting karst features. Further notes on the area appeared in survey

70 FAIRBRIDGE AND TEICHERT.

ICAPE NATURALISTE,
| GEOGRAPHE

arr DUNSBOROUGH

Nose Me hee
~ BUSSELTON _

GEOLOGICAL SKETCH MAP

NATURALISTE -

COWARAMUP oe LEEUWIN
BAY ps! we REGION
WESTERN AUSTRALIA
Be 2

25, 120 8 19

MARGARET____
+ RIVER ; MILES
of Aisa:

Lose ay on surveys by Sarnt-
Smith, (W/é and onrecopnaissante by
RW Farbridge x C.Terchert, 1948.

Legend:
&..MID-RECENT !IOFOOT BENCH
C...KARST LIMESTONE CAVES
—| Alf COASTAL LIMESTONE
—. ~MESOZOIC-TERTIARY
bs

CAPE FREYCINET y, at SEDIMENTS

ae + PRE-CAMBRIAN

+ GRANITE-GNEISS

HAMELIN BAY

Hamelin 15/377 ogf

FLIINDERS

“LEEUWIN) sities une)

Text-fig. 1.—Geological Sketch-map of Naturaliste-Leeuwin Region,
W.A.

SOIL HORIZONS AND MARINE BANDS. G1

bulletins by Saint-Smith (1912) and Woodward (1915), but since then it appears
to have been neglected, apart from a short paper on granitization of the Leeuwin
eneiss by Carroll (1940).

The Coastal Limestone tends to dominate the landscape with karst features,
caves and underground streams. Instead of it being barren, however, there
are rich soils derived from the old rocks, which, coupled with the heavy rainfall,
support a very flourishing vegetation, except immediately on the coast. On the
shoreline the Coastal Limestone is etched by spray into bizarre shapes and in
the intertidal zone the evidence of marine solution of limestone is demonstrated
by the deep, etched undercuts and broad horizontal reef platforms. As indicated
elsewhere (Fairbridge, 1948a, 1950) the effect on limestones of this corrosive
activity in sea-water and spray near the shore is to produce perfectly truncated
horizontal reefs at low-tide level.

On a visit to this area in January, 1948, we paid attention mainly to two
aspects of the Coastal Limestone, both connected with climate and eustatic
processes. These less usual features were soil horizons with a fauna of land
snails and a thin marine member with shell beds and massive conglomerates.
Both of these are better developed here than anywhere else we have seen on the
Western Australian coast, but some other occurrences will be briefly described
for comparison.

II. Som HORIZONS.
(a) Hamelin Bay and Island.

Fossil soils in the Coastal Limestone may be seen at many points between
Cape Naturaliste (Plate VII, fig. 3) and Cape Leeuwin, as well as elsewhere
along the Western Australian coast, but the finest development is at Hamelin
Bay (Plate VII, figs. 1, 2) and on Hamelin Island, about one mile offshore
(Text-fig. 2). The calcareous xolianite between the soil bands here is normal,
except that vertical pipes are rather rare.

The average soil horizon is about 1-3 feet in thickness, but in hollows it
was found up to 8 feet. Features of the soil are its lack of stratification, its soft,
‘‘ earthy ”’ feel (where not travertinized) and fineness of grain, all in contrast
to the exolianite. A typical section of one of these soils shows from top to
bottom :

(iv) Travertine ‘‘ cap-rock ’’, smooth and hard on top, soft below; inter-

fingered by fossil root structures, also preserved in travertine
4-1 foot
2

(ii) Soft, brown humic layer ; often full of land snails (Bothriembryon spp.)

and ‘slightly calcified insect cocoons (probably of the weevil Leptops

duponti, see Lea, 1925) at : 1-3 feet
(ii) White, leached layer; also overated be tone are anode in places
this contains knobby concretions of travertine. ve 1-4 feet

(i) Basal travertine; in places merely a thin ees. on the irregular
underlying surface, but sometimes cementing a layer of breccia or
conglomerate (pebbles from an older travertine cap) .. +-1 foot

These divisions are not constant. In places the travertinized cap-rock is
missing ; in others the whole band is travertinized from top to bottom. Thus in
places the soil horizon stands out ruggedly, while in others it is so weathered
away as to look like a narrow cleft cut in the cliff wall.

On, the mainland cliff, four soil zones (Text-fig. 3) are marked on the attached
diagram (A, B, C and D), while on Hamelin Island no less than six were noted
(A to F), rising to about 65 feet above sea level. There are local differences in
each soil (varying travertinization, varying humic content), but they do not

72 FAIRBRIDGE AND TEICHERT.

seem to be constant. For that reason it would not seem desirable to apply
stratigraphic names to each horizon (as, for example, in Bermuda).

The soils are rich in carbonates (Table I). The calcium carbonate and
magnesium carbonate content is generally only slightly lower than that of fresh
seolianite (in one case even slightly higher) and the insolubles are more concen-
trated in the two lower soils. There is a varying percentage of organic matter

GEOMORPHOLOGICAL
SKETCH MAP

HAMELIN BAY

AND VICINITY
WESTERN AUSTRALIA

"MNE

“a y aS
HAMELIN 443° ‘

Fresh- Water

LEGEND:

'.: Loose Sand& Beach
Dune fatterns
Coastal Limestone

escarpment
he Mid-Fecent 10 Benth
+ Lre-Carmbrian
granite-Greiss
woe fPeets

wm,

Based on air photos 29629 & 29696. RWE

Text-fig. 2.—Geomorphological Sketch-map of Hamelin Bay and vicinity.

which is absent from the parent rock. The test for phosphate was negative in
every case. On the whole these soils seem to be of the rendzina type. Their
special interest lies in the fact that they represent fossil rendzinas in which the
development of a mature soil profile has been arrested. An important point is
the absence of thick travertine crusts, pipes and other karst features, indicating
that the periods of soil formation were not of long duration and that the dune
developments were in rapid sequence.

pte eee | Yon wese hao nal

SOIL HORIZONS AND MARINE BANDS. 73

All soil horizons contain abundant specimens of the land snail Bothriem-
bryon. This genus is at present restricted to the south-west of Australia, where
it occurs from Eucla to Shark Bay. It has developed into a large number of
local races or subspecies, which were given specific rank by Iredale in his revision
of the land molluscs of Western Australia (Iredale, 1939).

65 Fr
N ; F A Prevailing Wind

VY
6)
Zug
LL ll

°o
Soll with travertine Beach -rock coarse-grained 8/4 Soil Locket Aolanite Aipping IZ,
concretions calcareous sandstone with with Bothriembryon
Shells

Text-fig. 3.—Pleistocene xolianites and fossil soil horizons at Hamelin
Bay, W.A. Length of section about 300 yards.

Recent shells collected on Hamelin Island and Hamelin Bay seem to belong
to Bothriembryon trilineatus Kobelt, which Iredale regards as a synonym of
Bothriembryon kingii Gray (Iredale, 1939, Plate 2, fig. 27). He records this
species from the King George’s Sound area. The same form also occurs in
Soil Horizon D, but not in the earlier horizons. The material at our disposal is
insufficient for a thorough study of the affinities of the species from the various
horizons. This is, moreover, made difficult by the fact that Iredale is silent on
the subject of variation of the individual ‘‘ species ”’, for each of which he figures

TABLE I.

Analyses of Soils and Parent Rock in Coastal Limestone of Hamelin Bay.

] 2, 3 4 5
% | % % % %
Calcium carbonate a: eect EI) 78-98 91-57 88-51 89-94
Magnesium carbonate ifs sag 5°75 4-33 4-87 4-77 7:68
Insoluble in acid .. — he 2°95 9-99 0-49 0-51 0-46
Organic matter as humus pele led b 2-08 1-73 2-29 —
1. Soil Horizon A, Hamelin Bay. 3. Soil Horizon C, Hamelin Bay.
2. Soil Horizon B, Hamelin Bay. 4. Soil Horizon D, Hamelin Bay.

5. Fresh eolianite, Hamelin Bay.

one representative only. However, specimens from Soil Horizons A and B
resemble in general the specimen figured as Bothriembryon perditus by Iredale
(1939, Plate 2, fig. 32) and specimens from Soil Horizon C are most similar to
Bothriembryon leeuwinensis Smith (Iredale, Plate 2, fig.17). Both these ‘‘ species”’
are now restricted to the extreme south-west of Western Australia. The
occurrence of different forms in different soil horizons does not, therefore, seem
to indicate climatic fluctuations. The meaning of these changes in the com-
position of the snail fauna cannot be evaluated until systematic collecting of
large series of both recent and fossil shells from many localities has been made.

74 FAIRBRIDGE AND TEICHERT.

(b) Other Fossil Soil Localities.

Fossil soil horizons of the rendzina type are well exposed along the coast,
especially near the Margaret River mouth. Farther north at Cowaramup Bay a
soil horizon has formed on beach conglomerate and marine shell beds. It
contains Bothriembryon trilineatus as in Soil D of Hamelin Bay. At Cape
Naturaliste, owing to recent disturbance to the vegetation, the old dune covering
is partly stripped off by wind erosion, exposing a large area of the fossil soil, in
which the travertinized roots are standing up. The local people refer to it as
‘* The Petrified Forest ’’.

Some 100-150 miles to the north, at Penguin Island and Point Peron
(Fairbridge, 1950) and at Rottnest Island (Teichert, 1950), there are further
developments of fossil soil, but there are generally not more than two bands,
and these are rather thin, though a pocket at Point Peron is 8 feet thick. These
more northerly soils generally have the appearance of terra rossa, though locally
they are dark brown and phosphatized. Deep solution pipes, descending beneath
sea level, are more important in this area, and the walls of these pipes are
characteristically banded with travertine and terra rossa. The egg-shaped
fossil cocoons of Leptops are also common here.

Still farther north, in the Wallaby Group of Houtman’s Abrolhos Island
(Teichert, 1947b) there are only a few thin bands of travertinized ‘‘ root horizons ”’
in the Coastal Limestone.

It appears therefore that the number and thickness of the soils decrease as
we go north. At the same time there is a change from rendzinas to travertine
and terra rossa, indicating, as might indeed be expected, climatic zoning and
decreasing humidity from south to north. It should be remembered that
rendzinas form at present in central and western Europe, whereas terra rossa
is the product of weathering in Mediterranean climates. It is the rendzina
which is forming to-day on the limestones of the south-west, while the terra
rossa soils are restricted to the coast about 500 miles to the north. Evidence
from the soils confirms our impression that at times during the Pleistocene almost
the whole of south-west Australia suffered from hot and arid conditions, but that
it was frequently relieved by wetter conditions in the south-west. This line of
reasoning may also perhaps help explain the extraordinary concentration of
marsupial life in this area at that time, for nowhere else in Western Australia
has such a rich and varied fossil marsupial fauna been found (e.g. Margaret
River Caves, see Glauert, 1910).

ITT. MARINE BANDS.

As indicated above, the Coastal Limestone is essentially an xolian, con-
tinental deposit. It is accordingly difficult to date except on geomorphological
evidence. The development in it of sporadic marine bands and littoral con-
glomerates is thus of particular interest, and for this reason Cowaramup Bay
is selected as a type locality.

(a) Cowaramup Bay (33°51'S.,114°59’H.). This is a rocky horseshoe-shaped
anchorage about 1,000 yards across (Text-fig. 4). The basement rocks outcrop
in rugged cliffs on the north side and on the floor of the bay. The hills inland
and at Cowaramup Point (on the south side) consist of Coastal Limestone, but
close to sea level around the south side of the bay there is a basal conglomerate
of granite-gneiss boulders cemented into a thin deposit of hard cream-coloured

SOIL HORIZONS AND MARINE BANDS. 75

marine limestone with pockets of sheils, and overlain by thin layers of beach
rock and less consolidated shell-beds, followed by a fossil soil and eventually
seolianites. The section is indicated in Text-fig. 5. The boulders in the con-
elomerate are mostly 2-4 feet across; they are well rounded and thus differ
markedly from the angular blocks of the same gneissic rocks being thrown up on
the shore to-day. In places on the present beach the two are mixed, the former

CEOMORPHOLOGICAL SKETCH MAP (oe
or COWARAMUP. BAY ee
WESTERN AUSTRALIA. 7
° V4 Ya ! Mile if + 4

LEGEND: | foes
we Toose Sand & Beach ay ;
; y
Ip Dune Patterns a3
mrem Loastal Limestone
Escarpment

Aw Mid-Recent 1/0 bench
+ fre-Camorian

Grante-Gnelss
a reets

COWARAMUP POINT BAY
oo aS

N \: a
{ yf

%

Based on air photo 26262: RWF

Text-fig. 4.—Geomorphological Sketch-map of Cowaramup Bay.

being reworked. In the overlying beach-rock there are also some boulders but

they are more rounded and smaller (1-2 feet). Springs of fresh water occur at
the base of this beach-rock.

The bed of soft, unconsolidated calcareous sand which follows is apparently
an old storm beach and is full of Patella, Nerita, Turbo, Marginopora, sharks’
teeth, ete. Itis proposed to call the whole of this marine sequence, conglomerate,
limestone, beach-rock and shell beds, after its dominant feature, the ‘‘ Cowaramup
Conglomerate ”’ as part of the ‘‘ Coastal Limestone ”’ Group (Plate VIII, figs. 1-4).

During its formation the sea level must have risen at least 10 to 12 feet
above its present level.

76 FAIRBRIDGE AND TEICHERT.

The age of these various deposits is not apparent from the contained fossils,
all of which seem to be of living species, although not necessarily from the local
station.*

NI Das
Deen Soil
Shell Bed

6 @ uss Conglomeratic Beach - Rock

| Massive marine limestone
with conglornerate & shell pockets

Approx. scale :

Uneven basement of

0 /0 20 f of
Pre-Cambrian granite-gneiss a

RWE

Text-fig. 5.—Section of Coastal Limestone on south side of Cowaramup Bay,
showing Cowaramup Conglomerate resting on basement of pre-Cambrian
granite-gneiss, and containing rounded blocks of the latter up to 4 feet long.
The lower part of the conglomerate has a matrix of massive marine limestone,
with fossil shells in pockets, and the upper part passes into a conglomeratic
beach-rock (calcareous sandstone). There is next a shell-bed (a Pleistocene
‘“‘ raised beach ’’), followed by a fossil soil band and steeply bedded eolianite.

* The fauna from the shell bed, overlying the Cowaramup Conglomerate, has been kindly
identified by Miss E. Dixon, as follows:

Phylum .. Protozoa.
Marginipora vertebralis.
Phylum .. Arthropoda.

Family Xanthide, Actewa sp. (1 only claw).
Family Grapside (1 only claw): Cirripedia: Tetracleta.
Phylum .. Mollusca.
Amphineura.
Ischnochiton. .
Pelecypoda.
Brachyodontes erosus, Divalucina occidua.
Gastropoda.
Marinauris roei, Ianthina violacea, Huplica bidentata, Josepha
tasmanica, Stomatella imbricata, Cocozeliana granosa, Campanila
levee, Propesium sp., Melarhaphe unifasciata, Nerita lineata,
Melanerita melanotragus, Cantharidus lehmanni, Mimeclanclulus
ventricosa, Vicimitra rhodia, Tallopia callifera, Patelloida alticostata,
Siphonaria baconi, Cellana limbata, Sabia conica, Antisabia erma.
It is interesting to note that this assemblage is a typical reef group (dominated by the “ peri-
winkle ’’ Melarhaphe unifasciata and numerous limpets) commonly found associated with con-
temporary intertidal reef facies near Perth, some 150 miles to the north. The foraminifer
Marginopora is also a standard warm-water reef form.

SOIL HORIZONS AND MARINE BANDS. fel

Physiographic features around Cowaramup Bay shed some light on the age
problem. The gneiss hills around the bay appear to show a rough terracing, at
about 25, 40 and 100-120 feet (illustrated by Clarke et al., 1944, p. 174), but this
does not extend to the Coastal Limestone hills, so if they are true coastal terraces,
they must be older than the last period of dune-building. If they are taken to
be Pleistocene eustatic terraces (see discussion in Section V), the age of the
Cowaramup Conglomerate must be very late Pleistocene.

The head of the bay is cut off by a bay-head bar, a “‘ raised’? beach now
rising to 13 feet above datum, preserving a coastal flat within at about 9 feet
above datum, the inner edge of which is marked by subdued ‘‘ fossil ”’ cliffs and
truncated spurs. On the south side of the bay there are several indications of a
10-foot bench eroded in the Coastal Limestone. This is clearly the mid-Recent
(Flandrian) shore-platform. In places it is overlain by a plaster of Recent
shelly beach-rock.

It may be seen that the coincidence in heights between this mid-Recent
10-foot sea level and the Pleistocene sea level of about the same elevation, when
the shell beds and older beach-rock were formed, is a pregnant source of confusion,
unless great care is taken. Below the 10-foot level the lithology is so varied with
conglomerates, massive limestone and granite-gneiss in situ that it is impossible
to trace out the 5- and 2-foot benches often found elsewhere, though small patches
of lower beach-rock plasters were found.

On the south side of Cowaramup Point there is a particularly fine con-
temporary limestone bench about 200 feet wide, studded with gneissic boulders.
The limestone is reduced by marine solution to low-tide level, while the resistant
conglomerate boulders stand up from it like giant cannon-balls. There is a
somewhat similar ‘‘ plum pudding ”’ bench like this in Hawaii (Wentworth and
Hoffmeister, 1940, p. 57).

(b) Other Occurrences of Marine Members.

We also found traces of the Cowaramup Conglomerate at several other
points in the Naturaliste-Leeuwin region. At Cape Naturaliste there is a
conglomerate with lenticular shell beds 10—23 feet above datum, unconformable
on the gneiss, and capped by fossil soil and wolianite. A mid-Recent 10-foot
bench cuts into it and there is also a 5-6 foot beach-rock plaster.

The granitic gneiss conglomeratic phase is seen at Yallingup and the shelly
limestone is undercut in cliffs and forms the floor of the present shore platform.
Ordinary «olianite overlies it at 10-12 feet above datum. At Canal Rocks, a
25-foot erosion terrace (shore-platform) in the pre-Cambrian is overlapped by the
conglomerate, again with shell beds and followed by eolianite.

About 4 mile north of Cape Leeuwin the shelly conglomerate may be traced
over an uneven gneiss from 3 to 25 feet above datum, but in places passes into an
evenly bedded grey sandy limestone (marine “ calcarenite ’’), a rather unusual
rock type. There is the usual 10-foot bench and a 5-foot beach-rock occurs. At
Hamelin Bay a beach-rock facies, underlying soil A in the eolianites, extends
from datum to 15 feet above, thus indicating a sea rising to 10 feet (assuming
a tidal range of 5 feet) prior to the dune formation here.

Outside our area, along the south coast the basal conglomerate has been
noted at several points: at Malamup near Cape d’Entrecasteaux (by
Montgomery, 1904), at Knapp Head (R.W.F.), with the shell beds underlying the
eolianite at Point Hillier (R.W.F.), and so on. Along the west coast on the
Swan River near Perth, the Peppermint Grove and Minim Cove shell beds
(3-24 feet above datum), characterized by Anadara trapezia, and underlying
the eolianite, have long been known (Somerville, 1920), and they have recently
been found at intermediate points: Mandurah and Myalup (R.W.F.).

78 FAIRBRIDGE AND TEICHERT.

Somewhat farther inland the Anadara beds are reported in the Guildford
Clay, which is clearly an estuarine facies. At Cottesloe and Fremantle there is
a lower, beach-rock and sandstone reef facies with coral boulders, rising from
5 to 15 feet above datum. On Rottnest Island at Salmon Bay an analogous
position (up to 8-10 feet) is occupied by a coral reef (Teichert, 1950), and north
of Perth, at Sorrento, a Lithothamnion reef was recently found at the same
horizon. Similar coral reefs and lagoon limestones (up to 18 feet above datum)
occur in the Abrolhos Islands (Teichert, 1947) ; Fairbridge, 1948a). From the
evidence of bores in the western parts of the Coastal Plain and on Rottnest, these
marine intercalations may be quite thin, and eolianites appear to extend down
to at least 230 feet below sea level.

While it is impossible to go into more detail here, it appears from a general
correlation of the Western Australian Quaternary (Fairbridge, in press) that these
Pleistocene marine shoreline deposits may be divided into two categories : those
that rise up to about 25 feet, and a slightly younger set that do not rise more
than 10-15 feet above datum. These separate different generations of eolianite.
The Peppermint Grove beds and Cottesloe beach-rock carry conglomerates of
earlier eolianites and travertines.

IV. INTERPRETATION OF THE AXOLIANITES AND THEIR SOILS.

Interpretation of the dune rocks of Australia can probably benefit from
comparison with well known dune systems of Western Europe. The analogies
are closest with those of two outstanding belts:

(1) The great dune area of the Gascony district in France, about 150 miles
long and 2—5 miles wide, and

(2) the great dune systems along the south coasts of the North Sea and the
Baltic Sea, which, with interruptions, are over 2,000 miles long and are
similar in width.

These dune belts are very similar to the dune limestone belt of Western Australia
in dimensions, but differ in material, the Kuropean dunes being mainly of quartz
sand. ,

In the Gascony district two generations of ancient dunes are present. Both
are of Pleistocene age and are now fixed by vegetation.

The origin of the North German dune belt on the other hand probably
only dates back to about 3500 to 4000 B.c. The development of the latter dune
system is best known and has been well described by Keilhack (1917 ; see also
review in English by Johnson, 1919). The coastline then (part of the Littorina
time) differed but little from that of to-day and the sandy beaches which are now
generally less than 100 yards wide, were probably not much wider when the
dune formation began. :

The main body of the dunes was probably formed within a few hundred
years. They then became vegetated and soil formation began. The old podsols
show a thickness of up to 3 or 4 feet. In favourable areas dunes continued to be
built up in successive belts right up to the present time. One such area is the
gap between the islands of Usedom and Wollin, at the mouth of the Oder, where
150 dune ridges have been formed. Keilhack (1912) concluded that 35 years
are sufficient for the building up of one dune ridge about 20 feet high. As
evidence of the speed of dune formation, there is the record of the rapid migration
of sand on the Kurische Nehrung after the deforestation of the 18th century,
when, within 50 years, enormous wandering dunes had been formed, piling up to
heights of 200 feet, burying many villages.

SOIL HORIZONS AND MARINE BANDS. 79

Pertinent to the Australian development are the following significant points
in the formation of these North German dunes:

(1) The restricted width of the beaches in Littorina times, which would
have required a ready supply of sand by waves and longshore currents.

(2) A climate which was temporarily unfavourable to fixation by plant
growth.

(3) The rapid accumulation of sand, the main dune belt being formed within
a few centuries.

The main development of the solianite belt in Western and southern
Australia is in close proximity to the present shoreline and on islands that rise
from the continental shelf. The dunes were probably built up in successive
belts from a depth of about 200 feet below sea level to several hundred feet above,
though a number of marine intercalations occur up to 25 feet above present
sea level.

Following Sayles and others, one of us (Teichert, 1947b) has previously
held that the calcareous dunes of Western Australia were formed when ‘“‘ sea
level stood lower than now and large parts of the Western Australian shelf were
exposed to wind erosion’’. Although similar suggestions have been made for
the South Australian eolianites by Tate (1879) and by Crocker (1946) we now
believe that such statements need modification.

First of all, it would seem that dunes may be built either during emergence or
submergence ; in the former case they would tend to be left behind with every
temporary pause in the sea’s retreat ; in the latter case they would tend to be
driven forward to the maximum limit of the sea’s advance.

It follows from this that the Western Australian eolianites may have been
formed at any time when sea level stood anywhere between +25 feet and —230
feet relative to its present height, and that the distribution of these rocks is
satisfactorily explained by assuming that they were formed along a shoreline
that was not very different in character from that of the present. It would
not be necessary to imagine a wide desert of sand during the low Glacial sea
levels, but merely a normal width of beach under conditions of advance or retreat.
The climatic conditions need not have been particularly arid, since even to-day,
if the normal vegetation on the very youthful soft dunes of Western Australia
is disturbed or destroyed, a new dune is rapidly initiated.

From observations in Hamelin Bay and elsewhere it is clear that the main
eolianite ridges were built up in stages, quiescent periods being indicated by
formation of soil horizons. The next question to be answered is how long are
the time intervals which these soil horizons represent.

Sayles (1931) was the first to consider the problem of soil horizons in dune
limestones. In Bermuda he observed six generations of eolianites separated
by five fossil soils, and he named them as stratigraphical units. His conclusions,
which have since been widely quoted, were that each eolianite corresponded to
a glacial stage of the Pleistocene, when the sea levels were low and a broad
insular shelf was exposed to wind erosion. The fossil soils formed during inter-
glacial stages when sea level rose and dune formation temporarily ceased.

Sayles’ ideas were strongly influenced by Verrill (1902-05), who published
analyses of the limestones which showed that ‘‘ impurities ’’ in them were often
‘not more than 0-5 per cent.’’ From this he concluded that 100-200 feet of
dune limestone were required to form 1-2 feet of soil. This figure may be
seriously challenged because the soil horizons are obviously not merely residuals
from solution.

Furthermore, it is generally accepted that sea level rose well above its
present stand several times during the Pleistocene, but of such multiple high

80 FAIRBRIDGE AND TEICHERT.

sea levels there is no sign in Bermuda. It is, therefore, almost certain that the
eolianites of this island and their soils represent a very much shorter span of
time than envisaged by Sayles—probably only a short period of the late
Pleistocene.

Few facts seem to be known which would allow one to form an accurate
estimate of the rate of soil formation under various conditions. As is well
known, depth of weathering has been used to calculate the length of Pleistocene
interglacial periods, e.g. by Kay (1931), whose figures were, somewhat uncritically,
used by Sayles for Bermuda. He proceeded on the assumption that post-glacial
soils in the U.S.A. were as a rule 6 inches thick and from this concluded that
1 foot of Bermuda soil might have accumulated in 50,000 years. There is little
doubt that such figures are much too high. Thick mature brown-earth and
podsol profiles of north-western Europe are not more than 10,000-15,000 years
old, and maturity may be reached fairly quickly.

Under entirely different conditions in Alaska, Judson (1946) found three
generations of dunes, each with a soil (typical podsol) 2-5 feet thick, overlying
weathered till; it would seem that the formation of the oldest dunes could
hardly have begun before the latest Pleistocene, or even post-Pleistocene.

From the analyses of the Western Australian fossil soils (see Table I), it is
apparent that the profile has been arrested at an early stage by renewed dune-
building.

Any explanation of the origin of the eolianites of W.A. must take into
account their repeated generations within narrow belts. Obviously the formation
of any one dune ridge required a rather specialized combination of conditions
which could not be expected to recur within a narrow zone several times during a
period of several hundred thousand years. The superposition of half a dozen
dune generations, as seen for example at Hamelin Island, suggests, on the
contrary, that all of them were formed within a comparatively short time. Thus
we are forced to the conclusion that, given the right combination of physio-
graphical, geological and climatic conditions, the formation of the calcareous
dunes, their fixation, accompanied by soil development, followed by the destruc-
tion of their vegetation and renewed growth of sand, a cycle repeated again and
again, and finally the gradual consolidation into dune limestone, were processes
that followed each other in fairly rapid succession.

V. CORRELATION WITH ENGLAND AND NORTH AFRICA.

Since it has been generally conceded that there has been no appreciable
warping of southern England since the close of the Pleistocene, and south-
Western Australia also seems to be a pretty stable sector of an ancient pre-
Cambrian shield, a tentative correlation may be attempted between, the two for
the late Pleistocene-early Recent history, based on absolute eustatic levels and
paleoclimatic oscillations. It is also fairly generally agreed that the ‘‘ 25 foot ”’
or ‘‘ pre-Glacial Raised Beach ”’ and fluvial Muscliff Terrace of southern England
may be correlated, with Zeuner (1945), on absolute elevation and other evidence,
with the Mediterranean Late Monastir eustatic sea level (7-5 m.). Precise
correlation of the latter, however, is still conflicting. Zeuner puts it as most
probably pre-Wiirm I (his Last Interglacial), but Cooke (1930) puts the 25 foot
terrace as mid-Wisconsin (thus post-Witrm I—probably pre-Wurm II). The
same course is now followed by Gigout (1951), as will appear in the comparison
with North Africa.

Precise data are scarce on the younger terraces and raised benches in
southern Britain, the bulk of the evidence probably being destroyed owing to the
large tidal ranges. Green (1946, p. 92) defined a fluvial Christchurch Terrace,

SOIL HORIZONS AND MARINE BANDS. 81

which may have been formed when the sea level stood about 15-17 feet O.D.
(personal communication). This, he says (1946, p. 94), ‘‘ must be very recent
geologically ’’, seeming to correspond with Daly’s ‘‘ six-metre bench ”’, but
this is not very helpful, since Daly confused both late Pleistocene and early
Recent levels (Fairbridge, 1948b). In a later paper Green (1949) compared the
Christchurch terrace with the Middle Sebilian of Egypt, which, according to Ball
(1939), corresponds to a 10-foot sea level that antedated the last big eustatic
drop at the end of the Pleistocene (and formed the ‘‘ Lower Buried Channel ”’
of Britain).

A still younger and very recent terrace was reported at Southampton,
Christchurch and Totnes (Green, 1949, p. 117), at 10-12 feet O.D., and this may
correspond to the Tilbury Stage of King and Oakley (1936). In a personal
communication (to R.W.F.) he correlated this ‘‘ beach’? with the Recent

TABLE ITI.

Late Quaternary Correlation in the South of England.

Absolute

Chronology Culture. Terrace. Beach. Kustatic Substage

Sea Level. Correlation.

115,000 Mousterian. —First Buried Channel — Low Wirm I.
85,000 Early Muscliff 41 ft. “‘ Pre-Glacial 25 ft.”’ 25-30 ft. Ouljian =

Aurignacian. ? Late Monastir
70,000 Solutrian. ——Second Buried Channel—— Low Wirm II.
45,000 Magdalenian. | Christchurch 22 ft. “ 10 foot”? in 10-15 ft. (2)
part.
25,000 Mesolithic. — —-Third Buried Channel— Low. Wurm III.
4,000 Neolithic. Tilbury (3) 10 ft. ‘* Post-Glacial ’’, ? 5-10 ft. Recent
- years B.P. “10 foot”? in part. (‘* Atlantic ’’).
|

(1) Zeuner (1945). These figures are given here as a general indication. No unconditional
acceptance of the ‘‘ Absolute Chronology ”’ is implied.

(2) Unnamed; Zeuner’s “ Interstadial slightly above O.D.”’ (p. 252).

(3) Tilbury Stage of King and Oakley (1936); also described from Southampton by Green
(1949).

‘¢ Atlantic ”’ stage; he indicated, however, that the ‘‘ 10-foot ’’ beach is often
only the seaward end of a buried 25-foot beach, which is also our experience
sometimes in Australia. At Gower, T. N. George (1932) demonstrated that
there is a distinct cemented beach-rock (Heatherslide Beach Conglomerate) at
about 10 feet O.D., which is definitely post-Glacial and plastered on the late-
Pleistocene terrace.

On this basis there are in the south of England three sets of river terraces
and raised beaches of late- and post-Glacial age, as indicated in Table II. In
this connection Zeuner (1952, p. 47) writes: ‘‘ Altogether, exceptionally favour-
able circumstances are required for the distinction of the three latest high sea
level phases, namely the Late Monastirian, the Wtirm Interstadial and the
post-Glacial (which may in itself be double). These shorelines are less than 8 m.
apart from each other, so that detailed evidence for high water mark is required
to sort them out on morphological grounds. Their independent existence,
however, can no longer be doubted. . .”

In North Africa, along the western shores of Morocco, the events of late
Quaternary history correspond more closely to those of the Australian coastline
than anywhere else in the world. Here the general picture may be obtained from
two fine memoirs by Neuville and Ruhlmann (1941) and by Gigout (1951).

I

82 FAIRBRIDGE AND TEICHERT.

Valuable additional material is contained in papers by Bourcart (1949, etc.)
but his interpretations are made under the assumption of tectonic warping, his
‘* continental flexure ’’, which though a most valuable concept, does not seem
to be applicable for the very short duration of the late Quaternary of either
Morocco or south-western Australia. Successive generations of eolianites during
the Pleistocene are found to correspond to the interglacial arid periods. These
dune rocks are separated by marine intercalations, corresponding to the high
eustatic sea levels, which initiate the pluvial stages. The maxima of the pluvials
correspond to the glacial maxima, when all the climatic belts migrated equator-
wards, and the temperate-wet climates interrupted the desert conditions when
the eolianites were formed. However, with the glacial culminations the sea
levels dropped sharply, separating each major cycle of dunes and high shorelines
from the next by the erosional phenomena of a short period when the base level
was considerably lowered.

This interpretation of Pleistocene events is quite different from that generally
adopted in Australia, where we have been inclined to regard the glacial low
sea levels as inducing the aridity which led to extensive dune formation ; and
these would be interrupted by the high sea levels corresponding to the pluvial
stages.

The evidence derived from this study of the Naturaliste-Leeuwin area seems
to correspond precisely with the conclusions reached in Morocco, where, in
addition to geological and geomorphological evidence, there are extensive
paleontological data and material of human industries, to support a general
correlation with the classical Pleistocene divisions of Europe. Gigout (1951)
appears to have solved the problem of the 25-foot sea level, placing it in the first
Wurm Interstadial (Mousteriam), and proposes the name ‘ Ouljian ”’ for it ;
this seems much better than Zeuner’s problematic ‘‘ late Monastir’’. He also
emphasizes the great importance of the Flandrian (Neolithic) 10-foot shoreline
on the Moroccan coast and in western Europe.

Turning now to the Australian evidence, it may be assumed that if limestone
terraces and ‘‘ raised beaches ”’ do not display any karst features which descend
beneath the present sea level, then these terraces are younger than the last low
sea level (Wurm III). Those Quaternary formations in Western Australia which
do show far-reaching karst channels are taken to be Wurm III or older. In
these pre-Wurm III formations we believe we can detect two high sea levels
reaching to 5-15 and 25-30 feet, which would logically belong to warm inter-
stadials prior to Wurm III and Wurm II.

Our correlation may now seem almost complete, but further surveys and
paleontological studies in Australia are badly needed. In the south-western
province a general correlation has recently been prepared by one of us in connec-
tion with a regional stratigraphy (see Fairbridge, in press), but it is still with
diffidence that this picture of local stratigraphic history is presented, as appears
in Table III.

CONCLUSIONS.

Observations along the Western Australian coasts have shown that the
Coastal Limestone is older than the early Recent, or ‘‘ Flandrian ’’ period, when
the sea level was 10 feet higher than it is to-day, because these rocks have been
widely benched at that level.

In many places along the coast the eolianites are intercalated with marine
deposits of reef or shoreline facies which, on the Abrolhos Islands, reach 18 feet,
and near Perth (Peppermint Grove) reach up to 25 feet above sea level. These
deposits are correlated with the First Wurm Interstadial (Ouljian of Morocco or
? Late Monastirian level in Europe and the Mediterranean).

Absolute
Chronology.

130,000
115,000

85,000

80,000

70,000

PLEISTOCENE.

45,000

4,000-
2,300 B.c.

600 B.c.—
1200 a.p.

Post-GLACIAL OR RECENT.

SOIL HORIZONS

AND MARINE BANDS.

TABLE III.

83

Late Quaternary Correlation in Western Australia.

South-west and
South Coast.

Cowaramup Con-
glomerate (also
Hamelin Bay beach-
rock, etc.).

Third AXolianite (with
Hamelin Bay souls).

Consolidation, karst,
etc. (as in Margaret
R. Caves).

West Coast.

First Holianite (mostly
below sea level).

Consolidation,
travertine and karst.

Peppermint Grove For-
mation (beach-rock
shell beds and cal-

carenite).
Guildford Clay

(estuarine facies) with

Anadara trapezia.

Pelsart Reef Lime-
stone.
Second Avolianite

(slightly lowered sea
level).
Consolidation,
travertine, karst.
Cottesloe beach-rock.
Salmon Bay reef-rock.

Third A#olianite (with
soils and snails).

Consolidation,
travertine, karst.

Bre-arc hy- rock “(at
Cowaramup Bay).
Ten-foot Shore Plat-

form.

Beach-rock
Cowaramup Bay).

Five-foot Shore Plat-
form.

(at

——(2)

od

(1) The Cowaramup Conglomerate rests on Tertiary or pre-Cambrian on the south coast.

‘* Raised Beach ’”’ and
Beach-rock (as at Pt.
Peron).

Ten-foot Shore plat-
form (3).

Emerged reef - rock
beach-rock (as in
Abrolhos Islands).
Five-foot shore plat-
form.

Beach-rock
Rottnest I.).
Two-foot shore plat-
form.

(as at

European Correlation
(R.W.F.).

Riss-Wutrm Interglacial.

Wurm I gilacial-pluvial

maximum.

25 foot shore (beginning
of First Wurm Inter-
stadial).

First
stadial.

Wurm Inter-

Wurm IT glacial-pluvial

maximum.

5-15 foot shore (be-
ginning of Second
Wurm Interstadial).

Second Wtrm _Inter-
stadial.

Wirm iI! | giacial-
pluvial maximum.

Flandrian transgression
‘* Atlantic ’? warm stage
10-foot shore.

Calais Beds.
‘“‘Subboreal’”’ climate (2).

Dunkerque Beds.
‘““Subatlantic” climate

(2).

nl

eolianites may lie buried inland or on continental shelf.

(2) On the south coast greater exposure and greater tidal range may explain general lack of
Same may be true in western Europe.

lower terraces.

(3) Recent platforms and shell beds have not yet received specific names,

examples are cited.

Older

Well described

84 FAIRBRIDGE AND TEICHERT.

At Cottesloe and on Rottnest Island there are slightly lower (5-15 feet)
marine intercalations, and these may belong to the Second Wirm Interstadial.
In the Naturaliste-Leeuwin area, described in this paper, these marine beds form
a basal member, the Cowaramup Conglomerate, which lies on the pre-Cambrian
basement. Elsewhere, the marine beds rest on older eolianites which occur
mostly below sea level.

We conclude that most of the Coastal Limestones of Western Australia,
both eolianites and marine members, have been formed during the latest
glacial stage, the Wurm, because if they were older there would be signs on them
of earlier, higher Pleistocene sea levels of which, in fact, there is no trace, except
those cut into the pre-Cambrian rocks of the south coast. However, it is quite
possible that truncated cores of still older eolianites exist beneath the present
ridges.

The general evidence in W.A. is of a cyclic succession of marine bands and
dunes separated by erosive intervals, but the bulk of the eolianites now exposed
above sea level along the coast were formed during the last (second) interstadial
of the Wiirm and the sea level then cannot have been very much lower than its
present stand.

These dunes most probably started to build up as soon as the sea level
reached its highest point (and consequently did not destroy them by further
advance). The dune-building then continued as the sea level dropped, probably
with a sequence of short oscillations, ceasing in the present belt as the shoreline
finally receded (probably most rapidly) to its minimum level, which would
coincide with the maximum glaciation.

It is an important conclusion of this study that the significant periods of
dune formation did not coincide with the glacial maxima, but were in fact
initiated with the highest eustatic sea levels during the interstadial warm-arid
periods.

The glacial maxima (eustatic lows) are marked by a lowering of the water
table, and solution by underground water caused deep sink holes and underground
rivers to develop, reaching down to the lowest base-level. These karst features
probably required a much longer period to form than the dunes themselves.

The short duration of the last dune building prior to the Recent epoch
hardly needs more emphasis. In places on the coast the younger (third) ceolianite
is over 300 feet thick, and this itself most likely accumulated in short phases,
interrupted by non-xolianite periods when soils developed. The fossil soil
horizons which thus separate a number of dune generations in the eolianites may
indicate minor and short-lived climatic cycles. They change from rendzinas
in the south to terra rossa soils at the latitude of Fremantle and grade into mere
travertine crusts farther north. We have therefore concluded that it is utterly
wrong to correlate each soil with an interglacial stage, as was done by Sayles in
Bermuda, and that any empirical formula deducing an absolute duration of
time from a certain thickness of soil accumulation is equally fallacious.

An interesting feature of the Cowaramup Conglomerate in the Naturaliste-
Leeuwin area is the light it sheds on the rate of coast erosion here. This basal
conglomerate was being formed under conditions of littoral erosion perhaps
45,000 years ago. It was then cut off by the fall in sea level, covered over by a
protective blanket of dune sands at least 60 feet thick, which was then cut
through again by marine erosion in Flandrian times. On the south side of
Cowaramup Point the width of the contemporary bench cut in these late Wurm
eolianites is not less than 200 feet; this has taken 4,000 years to form (aided
by a progressively lowered sea level), which gives an average rate of about 0-05
foot per year. But the eustatic drops have been short and sharp, so that the

SOIL HORIZONS AND MARINE BANDS. 85

actual rate must have been faster. The boulders of the Cowaramup Con-
glomerate, however, are now being reworked by contemporary erosion, so that
we are now back on the shoreline of the Wiirm II-III Interstadial, and so the net
amount of coastal retreat over the last 45,000 years has been nil.

We may close with the hope that these notes may stimulate further interest
in the possibilities of coastal geology as a means of unravelling the history of
eustatic and climatic events over the last 50,000 years or so.

BIBLIOGRAPHY.

Ball, J., 1939. Contributions to the Geography of Egypt. Surv. and Mines Dept., Cairo,
308 pp.
Potente a ., 1949. Nouvelles observations sur le Quaternaire du littoral marocain entre
Casablanca et Safi. Bull. Soc. Géol. France, Ser. 5, 19, 453.
Brown, H. Y. L., 1872. Geological Reports on Country in the Neighbourhood of Stanton Springs,
Williams and Canning Rivers, Pinjarrah, Bunbury and Cape Leeuwin. Parl. Pap. (W.A.),
Perth, No. 6 (for 1872), 6.
Carroll, D., 1940. Granitization at Cape Leeuwin. Aust. Journ. Sct., 2, 167-170.
Clarke, E. de C., Prider, R. T., and Teichert, C., 1944. Elements of Geology, for Western
Australian Students. Crawley (University Bookshop). 301 pp.
Cooke, C. W., 1930. Correlation of Coastal Terraces. Journ. Geol., 38, 577-589.
Coulson, A., 1940. The Sand Dunes of the Portland District and Their Relation to the post-
Pliocene Uplift. Proc. Roy. Soc. Vic., 52, pt. 2, 315-335.
Crocker, R. L., 1946. Post-Miocene Climatic and Geologic History and Its Significance in
Relation to the Genesis of the Major Soil Types of South Australia. Council Sci. Ind.
Res. (Melbourne), Bull. No. 193, 56 pp.
Fairbridge, R. W., 1948a. Notes on the Geomorphology of the Pelsart Group of the Houtman’s
Abrolhos Islands. Journ. Roy. Soc. West. Aust., 33 (for
1946-47), 1-43.

——_____+___—— 1948). Problems of Eustatism in Australia. In: Problémes des Terrasses
(6 Rapp. Comm. Terr. Plio. Pleist., Union Géogr. Int., Louvain),
47-51.

—_—__—_—_——— 1950. The Geology and Geomorphology of Point Peron, Western Australia.
Journ. Roy. Soc. West. Aust., 34 (for 1947-48), No. 3, 35-72.

— —_—_—— Stratigraphy of South-western Australia. In press.

George, T. N., 1932. The Quaternary Beaches of Gower. Proc. Geol. Assoc., 43, 291-324.

Gigout, M., 1951. Etudes géologiques sur la Méséta Marocaine Occidentale (arriére-pays de

Casablanca, Mazagan et Safi). Notes et Mémoires (Serv. Géol., Maroc), No. 86.
Gill, E. D., 1948. The Geology of Warrnambool. Proc. Roy. Soc. Vic., 55, Pt. 2, 133-152.
Glauert, L., 1910. The Mammoth Cave. Rec. W.A. Mus. and Art Gallery, 1, 11-36 (cont.
1912, 39-46; 1914, 244).

Green, J. F. N., 1946. The Terraces of Bournemouth, Hants. Proc. Geol. Assoc., 57, 82-101.

—_—_—_—___+__—— 1949. The History of the River Dart. Proc. Geol. Assoc., 60, 105-124.

Gregory, J. W., 1849. Notes on the Geology of Western Australia. The West. Aust. Almanac

(for 1849), 107-112.
Gregory, F. T., 1861. On the Geology of a Part of Western Australia. Quart. Journ. Geol. Soc.,
17, 475-483.

,Hardman, E. T., 1884. Report on the Geology and Supposed Gold-bearing Rocks in the
Neighbourhood of Bunbury, Blackwood, etc. Parl. Paper (W.A.), Perth, No. 18.

Hills, E. S., 1939. The Age and Physiographic Relationships of the Cainozoic Volcanic Rocks of
Victoria. Proc. Roy. Soc. Victoria, 51, Pt. 1 (n.s.), 112-139.

Hossfeld, P. 8., 1950. The Late Cainozoic History of the South-east of South Australia. Trans.
Roy. Soc. S.Aust., 73, 232-279.

Iredale, T., 1939. A Review of the Land Mollusca of Western Australia. Journ. Roy. Soc.

West. Aust., 25 (for 1938-39), 1-88; also Rec. W.A. Mus., 2, Pt. 1.

Johnson, D. W., 1919. Shore Processes and Shoreline Development. New York (Wiley),
revised ed., 1938.

J ohnston, R. M., 1888. Systematic Account of the Geology of Tasmania, Hobart. (Govt.

Printer.)
Judson, 8., 1946. Late Glacial and Postglacial Chronology on Adak. Journ. Geol., 54,
376-385.

Jutson, J. T., and Simpson, E. 8., 1917. Notes on the Geology and Physiography of Albany.
Journ. Roy. Soc. West. Aust., 2 (1915-16), 45-48.

Kay, G. F., 1931. The Relative Ages of the Iowan and Wisconsin Drift Sheets. Amer. Journ.
Sci., 221, 158-172.

Keble, R. A., 1946. The Sunklands of Port Phillip Bay and Bass Strait, Mem. Nat. Mus. Vic,

(Melbourne), No. 14, Pt. 2, 69-122.

86. FAIRBRIDGE AND TEICHERT.

Keilhack, K., 1912. Die Verlandung der Schwinepforte. Jahrb. Preuss. Geol. landesanst.
(1911), Berlin, 32, Pt. 2, 209-244.
—_——__—_—— 1917. Die grossen Ditinengebiete Norddeutschlands. Zeitschr. Dtsch. Geol.
Gesellsch., 69, B, Monatsber., 2-19.
King, W. B. R., and Oakley, K. P., 1936. The Pleistocene Succession in the Lower Part of the
Thames Valley. Proc. Prehist. Soc. (London), 2 (n.s.), 52-76.
Lea, A. M., 1925. Notes on Some Calcareous Insect Puparia. Rec. South Aust. Mus., 3, 35-36.

Montgomery, A., 1904. On the Prospects of Discovering Petroleum by Boring in the Warren
River District. Dept. of Mines, Perth, Ann. Rept. (for 1903), 95—98.
Neuville, R., and Ruhlmann, A., 1941. La place du Paleolithique ancien dans le Quaternaire
marocain. Inst. Hautes-Etudes Marocaines (Casablanca), No. 8, 156.
Saint-Smith, E. C., 1912. A Geological Reconnaissance of a Portion of the South-west Division
of Western Australia. Geol. Surv. West. Aust., Bull. No. 44.
Sayles, R. W., 1931. Bermuda during the Ice Age. Proc. Amer. Acad. Arts Sci., 66, 381-468.
Simpson, E. 8., 1902. Geological Features of the South-western Caves District. West. Aust.
Year Book (for 1900-01), 124-129.
Sprigg, R. C., 1952. Stranded Pleistocene Sea-beaches of South Australia and Aspects of the
Theories of Milankovitch and Zeuner. Int. Geol. Congr. (XVIII G.B., 1948), Pt. 13, 226-237.
Somerville, J. L., 1920. Evidences of Uplift in the Neighbourhood of Perth. Journ. Roy. Soc.
West. Aust. 6 (1919-20), Pt. 1, 5-20.
Tate, R., 1879. Natural History of the Country Around the Head of the Great Australian Bight.
Phil. Soc. Adelaide, 2, 114.
Teichert, C., 1947a. Stratigraphy of Western Australia. Tuts JouRNAL, 80, 81-142; Bull.
Amer. Assoc. Petr. Geol., 31, 1—70.
1947b. Contributions to the Geology of Houtman’s Abrolhos, Western Australia.
Proc. Linn. Soc. N.S.W., 71, 145-196.
—_—_—_——— 1950. Late Quaternary Sea-level Changes at Rottnest Island, Western Australia.
Proc. Roy. Soc. Vic., 59, 63-79.
Tindale, N., 1947. Subdivisions of Pleistocene Time in South Australia. Rec. S. Aust. Mus.,
8, No. 4, 619-652.
Verrill, A. E., 1902-5. The Bermuda Islands: Their Scenery, Climate, Productions, Physio-
graphy, Natural History, and Geology ... Trans. Conn. Acad. Arts Sci., 11, 413-956 ;
12, 45-348.
Wentworth, C. K., and Hoffmeister, J. E., 1940. Hybrid Shore Bench at Ulupau Head, Oahu.
Journ. Geomorphology, 3, No. 1, 57-58.
Woodward, H. P., 1894. Notes on the Geology of Western Australia, Geol. Mag., dec. IV,
1, 545-551.
—_-—_______—__—_ 1915. The Geology and Mineral and Allied Resources of the Coastal Plains
of the South-west Division. Geol. Surv. West. Aust., Ann.
Progr. Rept. for 1914, 15-18.
Zeuner, F. E., 1945. The Pleistocene Period, Its Climate, Chronology and Faunal Succession.
London (Ray Society, No. 130), 322 pp.
—_______-——— 1952. Pleistocene Shorelines. Geologische Rundschau, 40, 39-50.

EXPLANATION OF PLATES.
PLATE VII.

Fig. 1.—Coastal Limestone zolianites at Hamelin Bay, showing Fossil Soil Horizon B (dark —

band), which just here is 6-18 inches thick. Immediately beneath it (marked by black notebook)
is an irregular band of travertine pebbles. The jagged part of the underlying zolianite is due to
travertinized root structures. The upper eolianite shows false-bedding at 25-30° N.E. (away
from the prevailing wind).

(Photo.—R.W.F. 4080.)

Fig. 2.—Two Fossil Soil Horizons (A and B) intercalated in the Coastal Limestone eolianites
ut Flamelin Bay. The lower band (A) is more heavily indurated with travertine than the upper
one and varies from 2—5 feet in thickness here. Only a thin layer of eolianite separates it here
from the upper soil (B), which is 3—6 feet thick ; note the transition from the dark upper part,
rich in organic matter, passing down into a white, leached zone with travertinized roots.

(Photo.—R.W.F. 4081.)

Fig. 3.—Fossil Soil Horizon at Cape Naturaliste, 3 feet thick, showing fine travertinized root

structures.
(Photo.—R.W.F, 4045.)

fi
)

Journal Royal Society of N.S.W., Vol. DXXXVI, 1952, Plate VII

Journal Royal Society of N.S.W., Vol. LXXXVI, 1952, Plate VIII

als

.

SOIL HORIZONS AND MARINE BANDS. 87

Pruate VIII.

Fig. 1.—Cowaramup Conglomerate, in Cowaramup Bay ; granite-gneiss boulders (A), 4 feet
long embedded in creamy white marine limestone (with few shelly fossils), passing upwards into
(B) finely current-bedded beach-rock (calcareous sandstone) the top of which is here marked by
the mid-Recent 10-foot eustatic bench (C). The beach-rock is overlain by hard and soft shell
beds (D), and a 12-inch fossil soil (E), to be capped by the steeply bedded zolianites in the back-
ground (F).

(Photo.—R.W.F. 4060.)

Fig. 2.—Cowaramup Conglomerate, seen near Cowaramup Point, partly broken up by
contemporary erosion. Boulder-bands alternate with white marine limestone. Granite-gneiss
basement in situ in background.

(Photo.—R.W.F. 4064.)

Fig. 3.—Contemporary coastal erosion platform cut in the Cowaramup Conglomerate and its
limestone matrix, which dissolves rapidly under inter-tidal marine erosion, leaving the resistant
boulders on the reef platform. Looking south from Cowaramup Point.

(Photo.—R.W.F. 4065.)

Fig. 4.—Cape Naturaliste. Detail of contact between deeply weathered pre-Cambrian
gneiss and schist and the overlying Cowaramup Conglomerate and its matrix of calcareous shell
beds.

(Photo.—R.W.F. 4046.)

a

‘ - AUSTRALASIAN MepicaL PusLisHing Company Lim TED
3 Arundel and Seamer Streets, Glebe
bp. 1953)" es
ne
so “ i

ISSUED APRIL 14, 1953

|| VoL. Lxxxvi ee | 2 = PART iV

SS : _ JOURNAL AND PROCEEDINGS

OF THE

| ROYAL SOCIETY

(OF NEW SOUTH WALES Ze

FOR

1952

(INCORPORATED 1881)

JAuSON,

Uy
CF F EB 1 1954
Nel IBRARY XK”

PART IV
OF
VOL. LXXXVI

"Containing Papers read in mieondteL and December, 1952

EDITED BY

IDA A. BROWNE, D.Sc. -

Honorary Editorial Secretary

te THE AUTHORS OF PAPERS ARE ALONE RESPONSIBLE FOR THE
STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN

% SYDNEY
PUBLISHED BY THE SOCIETY, SCIENCE HOUSE
GLOUCESTER AND ESSEX STREETS

1953

CONTENTS

VOLUME LXXxXVI

Part IV

fe st

Ar S. Ritchie’ .
¥ 1 J

RS
Art! XI. | Fg astdttte ones and Associated Stratigraphy at Four-Mile Creek, South-west
of Orange, N.S.W. By N. C. Stevens and G, H. Packham

Art’.X.—A Speen te the Geology and Glaciology of the Snowy Mountains. By

Art. XII.—Martiniopsis Waagen from the Salt Range, India. By Ida A. Brown

Art. XIII.—Contributions to a Study of the Marulan Batholith. Part II. The Grano-
diorite-Quartz Porphyrite Hybrids. By G. D. Osborne and J. 8. Lovering

Art. XIV.—The Replacement of Crinoid Stems and gate ote: Py Cassiterite at Emma-
ville, New South Wales. By L. J. Lawrence ..

100
108

119

n=

JOURNAL AND PROCEEDINGS

OF THE

ROYAL SOCIETY

OF NEW SOUTH WALES

FOR

1952

(INCORPORATED 1881)

VOLUME LXXXVI

Part IV

EDITED BY

IDA A. BROWNE, D.Sc.

Honorary Editorial Secretary

THE AUTHORS OF PAPERS ARE ALONE RESPONS18LE FOR THE
STATEMENTS MADE AND THE OPINIONS EXPRESSED THEREIN

SYDNEY

PUBLISHED BY THE SOCIETY, SCIENCE HOUSE
GLOUCESTER AND ESSEX STREETS

A CONTRIBUTION TO THE GEOLOGY AND GLACIOLOGY
OF THE SNOWY MOUNTAINS.

By A. 8S. RITCHIE.
With one Text-figure.

Manuscript received, October 14, 1952. Read, November 5, 1952.

INTRODUCTION.

Considerable evidence of Pleistocene glaciation in the Kosciusko summit
area was recorded by David (1908) and David and others (1901). Later, Browne
and others (1944, 1946) recorded further evidence both in the summit area and
in the country north of it.

Some years ago the writer turned his attention to an area of about twenty
square miles, between the two already investigated, in what might be called the
Granite Peaks-Rolling Ground region. The eastern boundaries of the region
are the Snowy and Munyang Rivers, while the northern and western boundaries
are Dicky Cooper’s Creek and the precipitous gorge of the Geehi River
respectively. In the south it is bounded by the Snowy River, the Guthega
River and Three Rocks Creek.

The present contribution is a progress report on investigations made during
visits to the region in January, 1945; January, 1948; January, 1949;
November, 1950; and March, 1952. The area is not easily accessible and all
work was done on foot. The weight of food carried limited each visit to eight
or ten days’ duration. Stormy weather with snow, even in summer, often
restricted field work.

Text-figure 1 is mainly a copy of the map of the Snow Leases and Permissive
Occupancies of the Department of Lands. Minor variations have been made in
the light of field observations. The field notes were related to the streams since
anomalous variations in the magnetic declination have been suspected by the
writer. The simple grid system of the map has been introduced to facilitate the
location of points in a region where few physiographic features are named. The
main streams and their tributaries are lettered for the same reason.

PHYSIOGRAPHY.

The Granite Peaks-Rolling Ground area is the northern continuation of the
Main Range of the summit area, the Great Divide extending from Consett
Stephen Pass (location F4) near Mt. Tate (F3) in the south to the saddle (B5
and C5) between the sources of Munyang River and Dicky Cooper’s Creek in the
north. The greatest elevation is at Dicky Cooper’s Bogong (B4), which stands
at 6,570 feet above sea level. The whole area is mostly above 6,000 feet, although
the lower valleys of the streams descend to about 4,200 feet on the east and to
about 3,000 feet on the west of the divide.

The drainage is effected chiefly by four streams which show remarkable
parallelism in a sub-meridional direction. These streams are the Munyang and
Guthega Rivers on the east of the divide and Dicky Cooper’s Creek and Windy

THE GEOLOGY AND GLACIOLOGY OF THE SNOWY MOUNTAINS. 89

Creek on the west. The collinearity of (a) Windy Creek and Guthega River,
(b) Dicky Cooper’s Creek and Munyang River and the parallelism of all these
with each other, with some of their tributaries and with Finn’s River is striking.
The explanation probably lies in the step-faulting as suggested by Browne and
others (1944). Some of the tributaries of these streams are large enough to
deserve names. Similarly, numerous peaks in the area should be named.

ATW)

Wa ww)

al 17d
N

»
7 =

° A i ‘Live qual anys

PHULES SS es

1, < % =
ETA aon 3
2 $04 PS
‘ o = 14° %G
ys 4 = Wy Is
veal 2 a i == definite |
Ma S mU --- indefinite
Ss A i WY definite |
~ = e a
= ated Oufcrop)

Text-fig. 1.—Granite Peaks—Rolling Ground Region, Snowy Mountains, N.S.W.

D.C.—Dicky Cooper’s Creek. T.R.—Three Rocks Creek.

M. —Munyang River. W. —Windy Creek.

F.R.—Finn’s River. G.U.—Guthega River.
EROSION.

Most of the area is above the tree-line and is well grassed except for the
rocky pinnacles. In sheltered places a thick stunted heather-like growth
abounds. Soil erosion seems to be at a minimum and, in normal weather, the
streams are very clear and flow through grassy banks. After heavy rain,
however, the streams become very muddy and leave no doubt that very active
erosion is in progress. |

Snow-patch erosion and nivation (Browne and others 1944) were noted
throughout the area. The role of snow daisies as a retardant should be recorded.
Where the ice thins around the margin of the snow-drifts the snow daisies, as if
anticipating their release, make considerable growth while still covered by up
to half an inch of ice. During this growth the leaves are sheathed in a gelatinous
film. Thus the soil surface is covered to a considerable extent before the snow
melts.

West of the divide the stream gradients are very steep and hence very
active stream erosion is in progress.

In the upland valleys and swamps the present streams are incised into the
floors of the valleys to a depth of from two to twenty feet.

K

90 A. 8. RITCHIE.

EVIDENCE OF GLACIATION.

There is widespread evidence of glaciation throughout the area. This
may best be described by considering each main stream system in turn. The
observations are merely recorded. Conclusions mostly have been left to the
future, when sufficient data are available for the whole Snowy Mountains area.

1) Munyang River (also known as White’s River).
(1) yang

This stream has its source on the south-eastern slopes of Dicky Cooper’s
Bogong (B4) and flows in a south-easterly direction into the Snowy River. The
upper valley is glaciated. The most prominent evidence is a terminal moraine
just downstream from White’s River Hut (C5) at an elevation of 5,500 feet (by
aneroid). It is proposed to call this White’s Moraine. Large faceted boulders
here form a pronounced barrier which is slightly breached to permit the stream
to cascade down some thirty feet to a rocky valley floor. This was observed
independently by Browne and others (1946). Upstream from the moraine the
valley is wide and the main stream and its tributaries meander in front of the
hut. A spur near the hut appears to have been truncated. Further upstream
there is less distinct evidence of two smaller terminal moraines at elevations of
5,670 and 5,760 feet. There is definite evidence of ground moraine right up to
the saddle between this stream and Dicky Cooper’s Creek, the elevation of the
saddle being 5,930 feet. Some of the drift appears to have come down from
Disappointment Spur. Only minor tributaries feed the river above the 5,670
feet level. Two large creeks (M4 and M5) join the river in the upper part, one
on either side of White’s River hut. The stream M5 has faceted boulders near
the hut but these may have been left as marginal moraines or kame-terraces by
the glacier in the main valley through which this stream (M5) flows. Further
upstream in M5, the sources of this stream are found in two indefinite large
cirques on the south-eastern slopes of Dicky Cooper’s Bogong. The elevation
of these cirques is about 6,400 feet and the westerly one has a terminal moraine
across it. Thestream M4 has faceted boulders and is fed in summer by a number
of snow-drifts which are cradled in as many cirques on the eastern slopes of the
Granite Peaks (C4 and D4). Some of these cirques, at elevations of from 6,200
to 6,420 feet, are very definite, possessing steep walls in a semi-circle with flat
floors sometimes terminated by small moraines now breached by the small
streams. The stream M4 rises at the head of a long, wide upland swamp.
Remarkably constant changes of height on both the eastern and western sides
of the swamp suggest step-faulting with strikes parallel to those already men-
tioned. An indefinite terminal moraine is located at the lower end of this upland
swamp (D5). During the ice age, much ice must have accumulated on the site
of the present swamp. The main egress was probably down the valley of M4
but some spillways over the eastern side of the valley into M6 seem possible.
Downstream from the top of White’s Moraine, a great accumulation of large
boulders extends for almost a mile and is probably the lower part of White’s
Moraine. The elevation of the lowest of these boulders is 5,250 feet. Thus a
considerable gradient exists and the stream appears somewhat youthful in
character. The valley is much wider than the stream, however, and is clothed
in an almost impenetrable scrub. This forms the most difficult country on the
eastern side of the divide. Below the moraine the valley opens out and, just
below the tributary M3, displays valley-in-valley structure and some degree of
maturity due to a more gentle gradient. The altitude here is about 5,000 feet.
About a quarter of a mile upstream from M1, the Munyang River becomes quite
youthful and pot-holes and cascades are prominent. The sudden change of
gradient here suggests more faulting. The lower mile of the Munyang River is
entrenched about thirty feet into a broader valley. The stream M1 has not yet
been examined.

THE GEOLOGY AND GLACIOLOGY OF THE SNOWY MOUNTAINS. 91
(2) The streams between Guthega and Munyang Rivers have not been examined.

(3) Guthega River.

This stream rises with several arms, some of which begin in small cirques.
on the eastern side of Mt. Tate (F3). Nivation appears to have played some part
in cirque formation here. The eastern arms of the Guthega rise in slightly
cirquated land a little north-west of the 6,470 feet peak (F4). At first the upper
valley is wide and glaciated but it soon assumes a youthful character and flows
through a rugged valley until it meets the Snowy River. Dr. W. R. Browne, in
a private communication, states that there appear to be truncated spurs on the
right bank of the Guthega valley as viewed from Guthega Camp.

(4) Dicky Cooper’s Creek.

This stream rises in a low col which separates it from the Upper Munyang
River. It receives tributaries from Dicky Cooper’s Bogong on the west and
from the Kerries and Gungartan (B6) on the north and east. A number of small
cirques on the north-eastern slopes of Dicky Cooper’s Bogong are perched high
above the valley of the creek and any relationship between the glaciology of the
two is obscure. The valley of Dicky Cooper’s Creek is wide and displays consider-
able scattered ground moraine in its upper part. The present stream is.
entrenched from about three to seven feet into the wide floor of the valley. The
wide valley narrows at a large terminal moraine (B5), which, it is proposed to
call Dicky Cooper’s Moraine. Its elevation is 5,850 feet and it is comparable in
size to the upper part of White’s Moraine. Below the moraine the valley narrows
considerably and turns sharply to the west. The writer did not follow the
stream further than the track which leads northwards from Dicky Cooper’s Hut.
Here another moraine has been reported by Browne and others (1946). This
valley appears to have received ice from Gungartan and the Kerries but the bulk
of the ice from Dicky Cooper’s Bogong appears to have gone down the Munyang
Valley.

(5) Windy Creek.

Members of this stream system drain to the west all that country from
Consett Stephen Pass (H4 and F4) to Dicky Cooper’s Bogong and its upper valley
is wide and very deep. This valley shows the most pronounced glaciation of
any the writer has seen in the Kosciusko area. The glaciation of this valley
was recorded independently by Browne and others (1946). In the valley there
are four terminal moraines, all breached by the stream which is entrenched from
five feet (near the source) to fifteen feet (two miles downstream) into the wide
valley floor. Most tributaries spill over into this valley from hanging valleys
perched two or three hundred feet above the valley floor. A notable exception
is stream W6, in which the glacial erosion seems to have kept pace with that of
the main valley. Below the confluence with stream W2 the main valley becomes
very youthful and the gradient very steep. Further downstream the Leaning
Rock Falls (C2) make passage on foot impossible.

Stream W6 rises on the west of the Granite Peaks (H4 and D4). The gradient
at first is relatively slight but soon steepens and a wide glaciated valley enters the
main valley. Morainic material in stream W6 is slight and indefinite.

Stream W5 rises in a similar manner to W6 but in its course there are two
terminal moraines. The upper one, the less distinct, is breached, but never-
theless dams the stream to form a small tarn which persists throughout the
summer. Downstream from this moraine the stream cuts into its own valley
to a depth of about five feet and exposes a fine till, which extends downstream for
about two hundred yards. About half a mile below the upper moraine is a

92 A. 8S. RITCHIE.

very large moraine, which appears to have been built up near the end of a
hanging valley. A great mass of morainic matter has tumbled into the main
valley. A small swamp lies upstream from the moraine.

Streams W3 and W4 drain the south-western slopes of Dicky Cooper’s
Bogong. Their upper valleys are wide and swampy. These streams have not
been closely examined. Downstream they assume a youthful condition before
entering Windy Creek. Browne and others (1946) observed the glaciated
character of these streams.

(6) Three Rocks Creek.

This stream drains the western slopes of Mt. Tate, the southern slopes of
Tate West Ridge and the northern slopes of Mann Bluff (G3). The lower parts
of the stream plunge steeply into the Geehi River and display active erosion.
The upper parts show glaciated valleys, two of which, from Tate West Ridge,
are very striking. The west slopes of Mt. Tate and the northern slopes of Mann
Bluff are strongly cirquated but no close examination was made.

(7) Finn’s River.

No thorough examination of this valley was made but strong evidence of
glaciation was noted en route to and from Island Bend on the Snowy River.
The amount of evidence of glaciation here is surprising and the area warrants
closer attention. The following were noted :

(a) The sources of Finn’s River are in large cirques near Tin Hut (A7).
This was noted also by Browne and others (1946).

(b) The upper four miles of the valley is U-shaped, with slight gradient and,
in places, forms upland swamps.

(c) One very large terminal moraine occurs at and above the confluence
with Farm Creek (FR1). This moraine is larger than any in Coota-
patamba Valley (David and others, 1901). Another probable terminal
moraine occurs about one mile upstream from the Snowy River. Above
this point the stream becomes swampy and is very wide and flat. No
close examination of this was made.
Tributaries draining Gungartan (C6) and Disappointment Spur (C6)
and (D7) rise in cirques and some appear to have terminal moraines.
(e) Farm Creek (stream FR1) has a terminal moraine just above its con-
fluence with Finn’s River. This moraine is large but is made up of
much finer material than any other the writer has seen in the Kosciusko
area. Some special circumstances are indicated here. This evidence
indicates a valley glacier from the south-west slopes of Toll Bar Ridge.
It would not be unreasonable to suspect glacial evidence in upper Toll
Bar Creek on the north-west slopes. This locality is well worthy of
further investigation.

—

(8) Petrological Evidence.

The only outcrops of metasediments known to the writer in the area lie to
the west of the divide. Nevertheless, fragments of quartz-schist are common on
the N.E. slopes of Dicky Cooper’s Bogong on the eastern side of the divide.
Similar specimens were found in upper Munyang River and in White’s Moraine.
This may be an important clue to the movement of the ice.

GEOLOGY.

A moderately gneissic granite makes up almost the whole area. Brief
mention should be made, however, of what appears to be a remarkable roof
pendant of metasediments in the gneissic granite. This was first noted about

THE GEOLOGY AND GLACIOLOGY OF THE SNOWY MOUNTAINS. 93

half a mile west of Dicky Cooper’s Bogong. The dip was vertical and the
strike 181°. The width here was about three hundred yards. The rocks were
determined in the field as quartz-schist, phyllite and milky quartz (in wide veins),
quartzites and highly cleavable slates. In 1946 Browne and others independently
observed this occurrence north of Dicky Cooper’s Bogong. Subsequently, the
pendant was traced by the writer northwards along its continuous outcrop to the
saddle near The Ghost on the track from Dicky Cooper’s hut to Mawson’s hut.
The outcrop was lost on this saddle, the last width noted being twenty-five feet.
No search was made further north. From the head of Barber’s Creek (W2),
the outcrop of the metasediments can be seen stretching to the south from near
Dicky Cooper’s Bogong. The outcrop was located again in Windy Creek valley
just downstream from the confluence with W5. Here the width was about one
hundred feet and the nature of the rocks about the same as those further north.
No outcrop of the pendant was found on the south-western side of Windy Creek
nor in the high country beyond. The abrupt cessation of this rock mass in the
creek bed confirms the faulting which has been suggested along these streams.
The writer believes this southernmost outcrop is continuous with those already
noted. It would appear then that this pendant is at least four miles long.

ACKNOWLEDGEMENTS.

The writer wishes to thank Drs. W. R. Browne and G. D. Osborne for advice
and constructive criticism of this contribution ; the Kosciusko State Park Trust
for access to the 1946 Report; Mr. T. Vallance for assistance ; Miss P. Grieves
for the tracing of the map, and the officers of the Snowy Mountains Hydro-
Electric Authority for much assistance and hospitality. Finally, the writer’s
thanks go to Mr. 8S. G. Alley and other field companions without whom the
journeys could not have been undertaken.

SUMMARY.

Observations and evidence of Pleistocene glaciation (in an area between the
two previously described) are recorded. Numerous terminal moraines and
cirques are recorded. A sketch-map is provided to indicate the locations of the
evidence. Brief mention is made of a remarkable roof pendant of metasediments
in gneissic granite.

REFERENCES.

Browne, W. R., Dulhunty, J. A., and Maze, W. H., 1944. Proc. Linn. Soc. N.S.W., 69, 238.

Browne, W. R., Maze, W. H., and Osborne, G. D., 1946. Report to Kosciusko State Park Trust
(unpublished).

David, T. W. E., 1908. Proc. Linn. Soc. N.S.W., 33, 657.

David, T. W. E., Helms, R., and Pittman, E. F., 1901. Jbid., 26, 26.

GRAPTOLITE ZONES AND ASSOCIATED STRATIGRAPHY AT
FOUR MILE CREEK, SOUTH-WEST OF ORANGE, N.S.W.

By N. C. STEVENS
and G. H. PACKHAM.

With two Text-figures.

Manuscript received, October 30, 1952. Read, December 3, 1952.

INTRODUCTION.

In the small area described in this paper, Ordovician, Silurian, Upper
Devonian and Tertiary rocks outcrop, The age of the Lower Paleozoic rocks
has been determined by the graptolites they contain, which show that strata of
Upper Ordovician, and Lower, Middle and probably Upper Silurian age are
present. Lower Silurian fossils have been recorded from only two other places
in New South Wales (Naylor, 1935; Fletcher, 1950), and at no place have
definite Lower, Middle and Upper Silurian strata been found in sequence.

Four Mile Creek Post Office is 14 miles from Orange in a south-westerly
direction and four miles west of the old iron and copper mines of Cadia, which
were believed to be in Ordovician andesites and tuffs (Jaquet, 1901).

Ordovician graptolites from a locality ‘“‘ four miles from Cadia’’ (Smith,
1899) have probably been collected near Four Mile Creek, but apart from this
reference there is no geological literature concerning the area.

STRATIGRAPHY
I. Ordovician.

1. Malongulli Formation (2). Fossiliferous Ordovician rocks occur east of
Four Mile Creek ; they are covered by Tertiary lavas to the north and their
southern and eastern extensions have not been ascertained.

For the most part they are calcareous siltstones ; dark grey and compact
when fresh but light grey and slaty when weathered. They are usually laminated
and show some variation in grain size. Some are impure limestones and resemble
types occurring in the Malongulli Formation of the Cliefden Caves district to the
south (Stevens, 1952); others have dominant intermediate plagioclase with
calcite and chlorite, and appear to have been derived from andesitic rocks.
Fragments of fine-grained andesites are recognizable in some of the coarser
varieties.

Good exposures of strongly folded Ordovician rocks occur between Digger’s
Creek and the Orange-Angullong Road. Along this road north-east of the
turnoff to the Post Office the strata dip gently to the north, but in the creek at
their western margin they dip west at steep angles.

Graptolites (mostly Glyptograptus teretiusculus) have been found at a number
of places along the Orange-Angullong Road (g,, Text-fig. 1). Specimens from
Digger’s Creek in the collection of the Geological and Mining Museum, Sydney,
are probably those recorded by Smith (1899). According to Mrs. Sherrard
(personal communication) these, and others in the collection of the Australian
Museum, Sydney, from the same district, indicate the zone of Nemagraptus
gracilis, the lowest zone of the Upper Ordovician.

GRAPTOLITE ZONES AND ASSOCIATED STRATIGRAPHY. 95

EON ea
=CIN Tie ea
rie (ale la

Bess

XN

—_—_—|
4

NESE

tis
Pe

mS
x ||

MAP OF THE FOUR MILE CREEK AREA

™! eet

Re Bae ees
il | ihe
+ z
ply U
52a ©)
Cee O
aoe G
Es G)
raise
eae
fale

~~

ee

Dee Graptolite Zones

SB ELLCLE as
Seeoeoees 1 TELL LY RL Teese:

siecle ceeee lle

e
et eeidecess es! PERU EU MOUL ED Aer Ee
SS & sccesoor lh ELLE Lee eeee
Bape 2 @£ we Sees EP im ome
a ee F286 2 Neeese ECE EEE eS ae
Zz > ze =F Za Ook = ra Ceoosesoowe i LL EL ELLLLLL
cS z i. S& a 3 =a S35 & raat Seer ee awe
Enis O 8 ae mo 6 Aus UHsé Bielsie cles oer ees Ie EBs
4 vp~vo 2a ee aon YM OWN CEE Le te apse octyl kam ne
FE OO 7) O a eoooeeveee 5
tes

a ecoceoneee Soot! |
ic 2!1L3NDVW SCOKCCCOH ECVE LREe

| Z <—————“} le ecoccsces: poooe

cae || he eeenVe0e0n0ec Yenveee

@ @ ecceeee& ‘eoeee

inne
[223]
TAH

acs delt.

Text-fig. 1.—Geological Map of the Four-Mile Creek Area,
near Orange, N.S.W.

96 STEVENS AND PACKHAM.

2. Angullong Tuff (?). On the west of the Upper Ordovician graptolite-
bearing strata is a belt of andesitic rocks about one mile in width. Andesitic
lavas and tuffs make up the bulk of the formation, with no discernible order.
Breccias and pebbly tuffs grade into conglomerates, and locally, fine and coarse
tuffs are intermingled as a result of slumping, giving the impression that one has
been invaded by the other. Conglomerates with pebbles of sediments similar
to the underlying graptolite-bearing rocks are present near the base of the
formation, suggesting a possible disconformity between the Upper Ordovician
siltstones and these andesitic rocks.

_ This formation, overlying the probable equivalent of the Malongulli Forma-
tion, may well be correlated with the Angullong Tuff to the south.

II. Silurian and Devonian.

The Silurian and Devonian rocks of this area have been divided as follows
in descending stratigraphical order :

Approx.

Thickness
4. Black Rock Sandstone se ye ie 800 feet
3. Bulls’ Camp Rhyolite ie Ree 2 uh 2 OOM
2. Wallace Shale .. a a neh ae 300 ,,
1. Panuara Formation .. é : .. 2,000

de)

The three lower formations have their fyi development on Bulls’ Camp
Creek and the lower part of the Panuara Formation is best seen on Four Mile
Creek, just upstream from the junction with Bulls’ Camp Creek.

1. Panuara Formation. The lowest member of the Panuara Formation
is a conglomerate with a red matrix and pebbles of weathered andesite. It
rests on weathered andesite of the underlying formation and some time break
between the two is indicated.

Above it are fine grey to green felspar sandstones which consist essentially of
material derived from the andesites, and which become more calcareous upwards,
grading into limestone.

This lower limestone bed (here called the Bridge Creek Limestone Member)
has @ maximum thickness of about 50 feet and consists of rhythmically bedded
limestone and marl, richly fossiliferous. The following forms have been identified
(G. HEP):

Halysites orthopteroides Eth. fil. Mycophyllum liluformis (Eth. fil.).
HT. cratus Eth. fil. Cystiphyllum sp.

Al. sp. aff. cratus Eth. fil. ? Lambeophyllum profundum (Conrad).
Heliolites sp. Kloedenia aff. concinna (Jones and
Desmidopora sp. Holl.).

Leperditia sp.
Shales and siltstones rest on the Bridge Creek Limestone, and less than
50 feet above it the lowest Silurian graptolites Monograptus gregarius and
Climacograptus sp. occur in black shales in Bridge Creek (g,). The zone is that
of M. gregarius, equivalent to the top of the Lower Llandovery (Lower Silurian).

About 70 feet above the Bridge Creek Limestone, a graptolite assemblage,
which includes Monograptus exiguus, M. marri, M. (2) galaensis and Retiolites
geinitzianus, is found in sandstones and interbedded shales 30 chains south of the
junction of Bridge and Four Mile Creeks (g,). The zone is that of M. crispus,
about the middle of the Upper Llandovery (Lower Silurian).

Several hundred feet of shales follow, then 50 feet of green felspathic sand-
stone, which may be tuffaceous. Directly on top of this bed is another graptolite

GRAPTOLITE ZONES AND ASSOCIATED STRATIGRAPHY. 97

horizon (g3), which contains Monograptus priodon, Cyrtograptus aff. imsectus
and a Monograptus of the vomerinus group. These forms suggest a zone fairly
low in the Wenlock (Middle Silurian). In the southern part of the area a
lenticular outcrop of massive limestone containing a few corals (Halysites,
Favosites) occurs about this horizon, and it is followed by shales, then fine-
grained quartzose sandstone and interbedded shales.

TABLE I.
Summary of Stratigraphic Relations of Graptolite-bearing Strata.

Lower Ludlow.
Un-named
—_— M. dubsus, M. cf. vulgaris.
member
(interbedded
fine
Wenlock. sandstone,
e siltstone
e) and
S| shale)
7, SI M. priodon, M. sp., Cyrtograptus aff.
< —| & insectus.
ea
5 =
EB <
2) | M. exiguus, M. marri, M. (?) galensis,
2 Retiolites geinitzianus.
<
A,
Llandovery.
M. gregarius, Climacograptus sp.
Bridge Creek Halysites etc. (See list in text.)
Limestone Member
|__| - - ---------- Break in sedimentation
Zz (boundary either near base of Silurian or
< | Ashgillian. high in the Ordovician.)
Ss)
4 Andesitic rocks
2 | Caradocian. (?=Angullong Tuff.)
©
2 Calcareous silt-
& | Llandeilian. stone (?=Malongulli Glyptograptus teretiusculus.
5 Formation.)

To the north, the rocks overlying the g, horizon are shales and siltstones
which show small-scale current bedding. Overlying strata are shales and
sandstones, and a fourth graptolite horizon (g,) occurs in grey shales in Wallace
Creek, a tributary of Four Mile Creek. The graptolites present are Monograptus
dubius and WM. cf. vulgaris, which are either Upper Wenlock or Lower Ludlow
types. The same zone is probably represented by very poorly preserved
graptolites in an impure black limestone in Bulls’ Camp Creek.

The Panuara Formation (which takes its name from Panuara Rivulet, an
alternative name for Four Mile Creek) is thinly bedded, and all the strata dip
to the west at angles which increase from 30° to 70° in this direction.

The Bridge Creek Limestone is the first in eastern Australia for which a
basal Silurian age has been proved. It may in part extend down into the top
of the Ordovician. The fauna contains elements known elsewhere in the Silurian

98 STEVENS AND PACKHAM.

of New South Wales (except for Desmidopora sp. and ? Lambeophyllum pro-
fondum), but none of the forms present in the Cliefden Caves Limestone (Stevens,
1952).

Age determinations of coral faunas in N.S.W. by graptolites are confined
to high Wenlock and Lower Ludlow assemblages. At Yass (Brown and Sherrard,
1952) and in the Nanima-Bedulluck area (Sherrard, 1952) graptolites of the
abovementioned age have been found, but in both cases the Halysites-bearing
fauna is stratigraphically beneath the graptolites. At Four Mile Creek Halysites
occurs in rocks from Lower Llandovery to Lower Wenlock. Unfortunately, no
limestone occurs higher in the sequence.

. . . Amn ot ire
SSS SSE SSS SSE tere ee ee ee eee 048,
afl lenis . sete . . . ° ont ne a ta A ee =—O-

Text-fig. 2.—Diagrammatic Columnar Section illustrating Silurian
Stratigraphy at Four Mile Creek.

A, conglomerate ; B, felspar sandstone ; L,, Bridge Creek Limestone ;

C, shales and sandstones ; D, felspathic sandstone bed ; H, quartzose

sandstone ; 91, 92, 93. Ja, graptolite zones ; L,, massive limestone lens.
-H, Panuara Formation; F', Wallace Shale.

2. Wallace Shale. This formation, which overlies brown shales of the
Panuara Formation with apparent conformity, consists of green and red highly-
jointed shales. The base of the formation is marked by the first thick bed of
green Shale. The bedding is characteristically wider-spaced and less well-defined
than in the Panuara Formation. A few minor sandstone beds a foot or so thick
are present; no fossils have been found.

In the southern part of the area dips are relatively gentle but to the north
they steepen and some minor folds occur.

3. Bulls’ Camp Rhyolite. The Bulls’ Camp Rhyolite, which overlies the
Wallace Shale, consists mainly of rhyolites, which are interbedded with pebbly
tuffs, especially near the top of the formation. The tuffs contain pebbles of
rhyolite, intermediate to basic igneous rocks and rarely, limestones with poorly-
preserved corals.

Structural relations with the underlying rocks are not clear on Bulls’ Camp
Creek, but both dip gently to the west in the southern part of the area. Angles
of dip increase towards the north.

4. Black Rock Sandstone. The rocks of this formation, which rest with
probable unconformity on the rhyolites and tuffs noted above, are interbedded
sandstones and quartzites, somewhat pebbly in places. The western boundary
of this formation has been mapped previously (Stevens, 1950), and it has been
shown that it underlies conglomerates and shales and forms the eastern limb of a
syncline. Further north, Sussmilch (1906) found Camarotechia pleurodon and

GRAPTOLITE ZONES AND ASSOCIATED STRATIGRAPHY. 99

Cyrtospirifer disjunctus at the base of the sandstone, for which he deduced an
Upper Devonian age.

III. Tertiary.

The Tertiary lavas of Mt. Canobolas occur chiefly in the higher country to
the north, but remnants still exist south and south-west of Four Mile Creek Post
Office at altitudes as low as 2,400 feet above sea level. The pre-lava land surface
was irregular but sloped in a general south-westerly direction.

The lavas are mainly alkaline trachytes except along the Orange Road
N.E. of Four Mile Creek, where basalts cap the ridge.

ACKNOWLEDGEMENTS.

We wish to thank Mrs. K. Sherrard for her determination of the Ordovician
graptolites and advice on the Silurian graptolites ; also officers of the Mines
and Forestry Departments, N.S.W., for access to aerial photographs.

REFERENCES.

Brown, I. A., and Sherrard, K. M., 1952. Tuts Journat, 85 (1951), 127.
Fletcher, H. O., 1950. Rec. Aust. Mus., 22, 220.

Jaquet, J. B., 1901. Mem. Geol. Surv. N.S.W., No. 2.

Naylor, G. F. K., 1935. Tuis Journat, 69, 75.

Sherrard, K. M., 1952. TJbid., 85 (1951), 63.

Smith, H. G., 1899. Exhibit. Jbid., 33, xli.

Stevens, N. C., 1950. Jbid., 82 (1948), 319.

—-——___—_———. 1952. Proc. Linn. Soc. N.S.W., 77, 114.

Sussmileh, C. A., 1906. Tuis JourNaAL, 40, 130.

MARTINIOPSIS WAAGEN FROM THE SALT RANGE, INDIA.

By IpA A. BROWN, D.Sc.
(Mrs. W. R. BROWNE.)

With Plate IX and three Text-figures.

Manuscript received, November 12, 1952. Read, December 3, 1952.

Abstract: Description is given of the external and internal characters of the genotype of
the Permian brachiopod Martiniopsis Waagen, 1883, based on complete and serially sectioned
specimens from the Permian Productus Limestone, Salt Range, India.

A new term, adminicula (sing. -wm), is introduced for supporting plates in the umbonal
regions in the Spiriferide.

INTRODUCTION.

In 1883 Waagen described a collection of Permian fossils from the Salt
Range, India, including a new genus of brachiopod, which he named Martiniopsis.
He did not describe fully the internal characters, but suggested that some of the
smooth spirifers occurring in the Permian of eastern Australia might be placed
in his new genus. Etheridge, Junr. (1892) adopted Waagen’s suggestion,
apparently without having examined any specimens from the Salt Range, and
since then the name has been generally accepted for a number of species of
Australian spirifers.

In the course of my revision of the Permian spirifers of south-eastern
Australia it therefore became necessary to investigate the characters of
Martiniopsis Waagen. Waagen had only about 12 specimens from the Salt
Range, which he placed in two species: the shells are still rare in more recent
collections. By the kind permission of the Director of the Geological Survey
of India I was able to examine six specimens of Martiniopsis in a collection being
investigated in 1938 by the late Dr. F. R. Cowper Reed, Paleontologist to the
Geological Survey of India, at Sedgwick Museum, Cambridge, England, and I
am indebted to the Director and to Dr. Reed for permission to make serial
sections of one of these specimens, which form the basis of the present paper.

I have seen one specimen from Amb, Salt Range, in the British Museum
(N.H.), B18596; one complete specimen and six fragments in the Schuchert
Collection, Peabody Museum, Yale; and four recently acquired specimens in
the Australian Museum, Sydney, all from the Productus Limestone of the Salt
Range. :

Since Waagen’s work was published some 15 additional species of
Martiniopsis (other than Australian) have been named by Tschernyschew (1902),
Grabau (1936), Diener (1911), Huang (1933), Fredericks (1929), Reed (1930)
and others, chiefly from the Ural Mountains, Shan States, Tibet and China.
They are all small, smooth spirifers, externally similar to those from the Salt
Range.

I am indebted to Dr. G. Marshall Kay of Columbia University, New York,
for the gift of two specimens of M. uralica Tscher. collected by him in 1938 from
the Pseudoschwagerina princeps zone of the Sakmarian at Tratauchikan,
Bashkiria, Urals, U.S.S.R. (Plate IX, figs. 4a-4c).

MARTINIOPSIS WAAGEN FROM THE SALT RANGE, INDIA. 101

The specimen (K31.326) described below was identified in 1938 by Dr.
F. R. C. Reed as ‘‘ Martiniopsis subpentagonalis Waagen, Horizon: Lower
Productus Limestone (Wynne’s Bed No. II), Locality: Right bank of Baral
Nala, S.W. of Amb, Salt Range. (Sheet 38P/15).” In his monumental work
on the fossils of the Salt Range (1944) Reed gives no description of Martiniopsis,
but quotes identifications of certain specimens as Martiniopsis spp. in the fossil-
lists in the second part of the memoir, and a specimen bearing the same number
as that he gave to me (K31.326) he identifies (Reed, 1944, p. 403) as Squamularia
(Neophricodothyris) baralensis Reed. This is certainly not the correct identitfica-
tion of the specimen given to me, which is undoubtedly a Martiniopsis. Waagen.
named two species of his genus, Martiniopsis inflata and M. subpentagonalis,
based on a dozen specimens, without clearly defining the distinction between

0 5 MD 15 bo 25 m/m

.
-° 8:3

; (ly ‘haa

Sana ey
. ly ! \
= Muay r Nyy! Mul gy” .

Roe te Se cfeursic ee ere

Text-fig. 1.—Camera lucida drawings of serial sections of Martiniopsis inflata

Waagen. The numbers in the upper left corner of each section refer to

numbered cellulose peels in the Australian Museum, Sydney. The figures on

the lower right of each section are distances in mm. from the top of the ventral
valve.

them. A study of the measurements and proportions of the specimens available
suggests that only one, variable, species is involved. It is my opinion that the
specimen, K31.326, agrees with that identified by Waagen as M. inflata, which
was chosen by Hall and Clarke (1894) as the genotype.

The specimen (K31.326) was photographed (Plate IX, figs. la-1d) and
plaster casts were made before mounting it for serial sectioning. It was ground
by hand on glass from the umbonal region to the anterior margin. The sections
are parallel to one another, separated by distances that were carefully measured
by spherometer, so that the position of any point can be fixed by three-
dimensional co-ordinates. Consequently the appearance of any desired section
may be plotted graphically. Cellulose peels were made at intervals of 0-3mm.
(average), but closer at critical points ; in all 76 sections in a depth of 24:20 mm.
These have been mounted and photographed.

Although the surface of the dorsal valve and the anterior commissure of the
specimen were somewhat weathered, the internal structures were perfectly
preserved ; camera lucida drawings of selected sections are reproduced in
Text-fig. 1. The structures revealed by the sections have been plotted in scale
drawings of two aspects : Text-fig. 24, the median vertical view, and Text-fig. 28,
the dorsal view (with the dorsal shell removed).

102 IDA A. BROWN.

The terminology used is in general that of Schuchert and Cooper (1932)
and Cloud (1942) with the following exceptions :

Denial plates, subvertical plates in the ventral valve which bear the hinge
teeth on their outer dorsal ends. They occur along the inner margin of the
delthyrium, increasing in depth from the tip of the umbo, and converging towards
the floor of the valve, although as in the case of all Permian spirifers I have
examined, they do not reach the floor of the valve. Equivalent to the ‘‘ receding
dental plates ’’ of Schuchert and Cooper (1932, p. 7) and Miloradovitch (1937) ;
to the ‘* carine delthyriales (delthyrial ridges) ’’ of Fredericks (1927) ; and to the
‘‘ dental lamelle ”’ (pars) of Muir Wood (1934) and others.

Adminicula, new term (sing. adminiculum), a pair of subvertical plates
that may occur in the apical region of the ventral and sometimes also the dorsal
valves, especially in very rostrate shells, extending from the inner tip of the umbo
to the floor of the valve, and appearing to buttress the arch of the valve. The
ventral adminicula do not reach the hinge-teeth or the margin of the delythrium,

m/m

0

Text-fig. 2.—Scale drawings based on serial sections of Martiniopsis inflata
to show internal structures. (A) Median vertical view. (B) Dorsal view,
with the dorsal valve removed.

v.a.—ventral adminiculum. d.a.—dorsal adminiculum.
d.p.—dental plate. c.p.—crural plate.
d.l. —descending lamella of spiralium.

but converge slightly and unite with the ventral margins of the dental plates
along a suture-line from the tip of the umbo to the antero-ventral ends of the
dental plates. They are not co-planar with the dental plates, but are equivalent
to the ‘‘ septa ”’ of the spondylium duplex of the pentamerids.

In the spirifers the muscle-scars occur on the floor of the ventral valve
between the adminicula, when these are present. The dental plates and the
adminicula form a structure resembling the spondylium discretum of Schuchert
and Cooper (1932, p. 10, Pl. 14, fig. 20). With the deposition of callus in the
posterior portion of the delthyrial cavity, the site of muscular attachment moves
anteriorly and the muscle track is overlain by callus or secondary shelly material
producing a ‘* pseudospondylium ”’ (Schuchert and Cooper, 1932, p. 10).

The ventral adminicula are the ‘‘ dental plates ’’ of many authors, including
Miloradovitsch (1937), the ‘‘lamell# apicales (dental lamella)” (pars) of
Fredericks (1927) and the ‘‘ apical plates ”’ of other authors.

The existing confusion in the nomenclature appears to have been due to the
former lack of recognition of two distinct pairs of plates in the apical region of
the ventral valves of certain spiriferide.

MARTINIOPSIS WAAGEN FROM THE SALT RANGE, INDIA. 103

SYSTEMATIC DESCRIPTION.
Genus Martiniopsis Waagen, 1883.
(Plate IX, figs. 1-4.)

Waagen, W., 1883.—Salt-Range Fossils. Mem. Geol. Surv. India,
Paleontologia Indica, Ser. XIII, Vol. 1, Pt. 4, fase. 2, p. 524, Plate 41,
Hos. ‘7, 8.

Etheridge, R., Junr., 1892.—In Jack and Etheridge: Geology and
Paleontology of Queensland and New Guinea, p. 236. Govt. Printer,
Brisbane, Qld.

Hall, J., and Clarke, J. M., 1894.—Genera of Paleozoic Brachiopoda.
Pal. of New York, Vol. 8, Pt. 2, pp. 34, 40.

Genotype (by subsequent designation, Hall and Clarke, 1894): Martiniopsis
inflata Waagen, 1883.

Diagnosis.—Small (usually less than 40 mm. in width) biconvex spiriferids,
sub-globular. Width approximately equal to length, but greater than the
thickness. Hinge-line short, ventral beak rostrate, tip of umbo incurved, almost
touching that of the dorsal valve ; cardinal extremities rounded. Shell smooth,
no fold or sulcus : anterior commissure rectimarginate or faintly sinuate. Small
interareas on both valves, not visible from outside. Shell impunctate.

Relatively large open delthyrium in the ventral valve, bounded internally
by divergent dental plates bearing small insignificant teeth on their antero-dorsal
edges. The dental plates unite ventrally with well-developed adminicula (see
Introduction) to form a ‘‘ spondylium discretum’’. Muscle-scars not known.
Shell generally is thin; some secondary thickening may occur in umbonal
cavities, but not in the delthyrial cavity.

Notothyrium in dorsal valve is open. There is no cardinal ‘‘ process ”’ ;
a small striated cavity inside the beak received the diductors. Divergent
hinge-plates bear small inner socket-ridges. Small convergent dorsal adminicula
support the hinge-plates, which give rise to the crura, from which the descending
lamelle proceed anteriorly. Near the anterior commissure the lamella swing
laterally and coil towards the ventral valve to commence the spiralia, which are
directed postero-laterally. There are about twenty turns in each spire. No
jugum or other jugal structure is developed. Pallial markings unknown.

Discussion.—Waagen, states (1883, p. 524): ‘‘ It was not possible for me to
expose also the internal characters of the genus, partly on account of the hardness
of the sandy limestone in which these fossils are preserved, and partly on account
of the scarceness of the specimens.’’ However, he was of the opinion that the
shell was punctate; I have been unable to find true puncte in the available
specimens, and suggest that ‘‘ the perforations arranged in quincunx ”’, noticed
by Waagen in the Indian specimens and by Etheridge in some Queensland
Spirifers, are in fact fine surface ornamentation, not uncommon in certain
Spirifers.

Martiniopsis differs from all other spirifers except Squamularia and Phrico-
dothyris by its complete lack of sulcus in the ventral valve and lack of fold in the
dorsal valve, with corresponding rectimarginate anterior commissure. H. and G.
Termier (1948, 1949) have shown that these characters among the brachiopods
are closely related to and dependent on the system of respiration and circulation,
and should therefore be of prime importance in the classification of the group.

Squamularia Gemmellaro, 1899, is based on an early Permian species from
Italy (Spirifer rotundata Gemm.). The genus has been discussed in detail by
Dunbar and Condra (1932) ; it is distignuished from Martiniopsis by its thicker
Shell, fine surface ornamentation and its complete lack of internal plates, ‘‘ dental
or septal lamelle ’’.

104 IDA A. BROWN.

Phricodothyris George, 1932 (genotype: P. lucerna George, 1932) is distin-
guished by its surface ornamentation of biramous, barbed spines and the usual
absence of internal plates.

Waagen and Etheridge apparently placed greater emphasis on the smooth
surface ornamentation of the shell, and made comparisons with Martinia McCoy,
1844. The genotype of Martinia has been regarded as ‘* Conchyliolithus Anomites
(glaber) Martin, 1809 ”’ (see Plate IX, figs. 5a and 5b), but following the ruling
of the International Commission on Zoological Nomenclature that Martin’s
species are invalid, Dr. H. Muir Wood (1951) has applied to the Commission for
the designation of Spirifer glaber J. Sowerby, 1820, as the type-species of the
genus Martinia McCoy, 1844. This form has a smooth surface ornamentation,
but towards the anterior margin develops a marked fold and sulcus, producing a
strongly sinuate anterior commissure. Internally it has no dental lamelle
or septa (adminicula).

Reticularia McCoy, 1844, may be regarded as a homceomorph of Martinia
in which the internal plates (dental plates and adminicula) are well developed.
Thus both of these genera are quite distinct from Martinopsis.

With regard to the designation of the genotype of Martiniopsis, Schuchert
and Le Vene (1929) quote Hall and Clarke (1894) as having selected the type.
It may be pointed out, however, that previously Etheridge (1892, p. 238) had
Stated, after discussing the genus Martiniopsis, ‘‘ Types. Martiniopsis inflata
Waagen (Indian) ;

Martiniopsis inflata Waagen, 1883.
(Plate IX, Figs. la-3c, Text-figs. 1-3.)
Martiniopsis inflata Waagen, 1883. Mem. Geol. Surv. India, Pal. Indica,
Ser. XIII, Vol. 1, Pt. 4, fase. 2, p. 524, Pl. 41, figs. 7, 8.
Martiniopsis subpentagonalis Waagen, 1883. Ibid., p. 527, Pl. 42, figs. 9-10.
Holotype.—Specimen figured by Waagen, 1883, Pl. XLI, figs. 8a-8d, from
the Upper Productus Limestone, Chidru, Salt Range, India.

TABLE l.

Measurements of Specimens of Martiniopsis inflata from Salt Range, India.
All lengths are in millimetres.

Specimen Number. 1 2 3 4 5 6 7 8 9 10
Length of ventral

valve Me 22 25 29 29 30 32 33 38 43 47
Width of valves... 23 25 30 33 35 32 36 36 47 47
Length of dorsal valve 18 21 25 24 25 27 28 30 39 41
Thickness of shell 14 13 19 19 17 18 20 27 29 30
Length of hinge-line 8 10 11 12 15 10 16 20 22 28
Apical angle of ventral

valve . | 108° | 105° | 105°.) 110° | 100° || ¢.82°, |, 103° |) 120? ae
Apical angle of dorsal

valve se fe (i LBO? | 125°) F268 1808 1 187 e T2820 S N20 i aes ie ae
1. Aust. Mus. specimen F39605 ; (G.S. India, K29/242), Loc. Amb.

2. Peabody Museum, Yale, specimen $2948; Loc. Amb, 21.

3. Aust. Mus. specimen F39604; (G.S. India, K31/126); Loc. Buri Khel.

4, Aust. Mus. specimen F44674; (G.S. India, K31/326), Loc. Amb.

5. Waagen’s measured specimen (p. 528, col. IT), Loc. Amb.

6. Waagen’s measured specimen (p. 528, col. I), Loc. Amb.

7. Aust. Mus. specimen F39602; (G.S. India, K30/805), Loc. Trimu Wahan).

8. British Mus. (N.H.) specimen 18596 ; Loc. Amb.

9. Waagen’s measured specimen (p. 526, col. II); Loc. Chidru.
10. Waagen’s measured specimen (p. 526, col. I); Loc. Chidru.

Numbers 1, 9 and 10 are those previously identified as M. inflata and the others as M. sub-
pentagonalis.

MARTINIOPSIS WAAGEN FROM THE SALT RANGE, INDIA. 105

Characters of the shell as for the genus. Waagen’s separation of the few
specimens available to him into two species appears to be unnecessary now that
additional specimens show regular gradation amounting only to variation of a
single species, and I therefore place his second species, M. subpentagonalis, in
Synonymy.

The table above shows the principal dimensions of ten specimens (including
the four measured by Waagen) from the Salt Range, India. They are arranged

00

m/m

40

30

20

10

Specimen No.-
2 345 67

20 30 40 50 m/m
Length of ventral valve in mm.

Text-fig. 3.—Graphical representation of the variation in the measurements
of specimens | to 10 of Martiniopsis inflata listed in accompanying table.
The horizontal axis represents the length (in mm.) of the ventral valves,
against which are plotted vertically
aa. Width of valves (x).
bb. Length of dorsal valve (@).
cc. Thickness of shell (()).
dd. Length of hinge-line (0).

in order of increasing size (length of the ventral valve) and the measurements are
shown graphically in Text-fig. 3. In view of the smallness of the specimens
and the difficulty of making accurate measurements of weathered specimens
there is remarkable approximation to a linear arrangement of corresponding
points for specimens of varying size, which surely indicates identity of species.

The umbonal angles are fairly constant, averaging 108 degrees (with one
abnormality) for the ventral valve, and 128 degrees for the dorsal valve. This is
probably a specific character.

L

106 IDA A. BROWN.

In this species the width of the shell is typically greater than the length.
The convexity of the valves is illustrated in Plate IX, figs. 1-3, and is seen to be
less marked than in M. uralica Tschern. shown in Plate IX, fig. 4. The hinge-line
is about half the width of the shell, and the interareas are small. There is no
trace of fold or sulcus and the anterior commissure is rectimarginate in the young
and mature shells, but there is a tendency for it to become slightly sinuous in the
larger, older shells. No covering plates have been observed either on the
delthyrium or the notothyrium.

The internal characters have been studied by means of serial sections of
specimen 7 in the accompanying table, as described earlier in this paper. Some
of the sections are illustrated in Text-fig. 1, and the internal structures are
indicated in Text-fig. 2. In the ventral valve well-developed dental plates
extend from the margins of the open delthyrium ventrally to meet the large
adminicula, which extend half-way down the valve. The junctions of the latter
with the ventral valve are visible through the translucent shell and have been
called previously the ‘‘ dental plates ”’.

In the dorsal valve there is a short low euseptum, probably not of taxonomic
significance, and small adminicula support the hinge-plates and crura. The
descending lamelle curve smoothly towards the anterior, without any jugal
processes and support the spiralia. These are directed postero-laterally and
have 19 turns in the spire. The ribbon is 1-2 mm. in width and 0-04 mm. in
thickness ; the length of the spire is about 13 mm. and its diameter about 9 mm.

REFERENCES.

Cloud, P. E., 1942. Geol. Soc. Amer., Spec. Pap. No. 38.

Diener, C., 1911. Anthacolitic Fossils of the Shan States. Mem. Geol. Surv. India, Pal. Indica,
N.S. 3, No. 4, p. 7.

Dunbar, C. O., and Condra, G. E., 1932. Brachiopoda of the Pennsylvanian System in Nebraska.
Nebraska Geol. Surv., Ser. 2, Bull. 5, p. 311.

Etheridge, R., Jnr., 1892. In Jack and Etheridge, Geology and Paleontology of Queensland and
New Guinea. Govt. Printer, Qld.

Fredericks, G., 1927. Der Apikalapparat der Brachiopoda Testicardines. Neues Jahrb. f. Min.
Geol. und Pal., Abhandl. LVII, B.-B. Abt. B, 1.

Gemmellaro, 1899. Fauna Calcari con Fusulina. Palermo, Fasc. 4, Pt. 1, pp. 325-326, Pl. 33,
figs. 38-45.

George, T. N., 1932. British Carboniferous Reticulate Spirifers. Q.J.G.S., 88, 546.

Grabau, A., 1936. Early Permian Fossils of China. Mem. Geol. Surv. China, Pal. Simca. B,
VIII (4), 242.

Hall, J., and Clarke, J. M., 1894. Genera of Paleozoic Brachiopoda. Pal. of New York, 8,

Pt. 2, 34, 40.
Huang, T. K., 1933. Late Permian Brachiopoda of South-western China. Pal. Sinica. B,
IX (2), 53.

Martin, W., 1809. Petrificata Derbiensia. Wigan. Pl. 48, figs. 9, 10.
McCoy, F., 1844. Carboniferous Limestone Fossils of Ireland. Dublin, p. 128.

Miloradovitsch, B. V., 1937. Morphogenesis of the Ventral Valve in Spiriferide. Problems of
Paleontology, 2, 3, 526. Moscow University Pub., U.S.S.R.
Muir Wood, H. M., 1934. Structure of Some Mesozoic Brachiopoda. Phil. Trans. Roy. Soc.
Lond., Ser. B, 223.
—_—_—_____———_—— 1951. Brachiopoda of Martin’s “ Petrificata Derbiensia’’. Ann. Mag.
Nat. Hist., Ser. 12, 4, 109.
Reed, F. R. C., 1930. Anthacolithic Faunas of the Southern Shan States. Rec. Geol. Surv.
India, 67 (1), 83-134.
—___—____—_—_—_ 1944. Brachiopoda and Mollusca from the Productus Limestones of the Salt
Range. Mem. Geol. Surv. India, Pal. Indica, N.S., XXIII, No. 2.
Schuchert, C., and Cooper, G. A., 1932. Brachiopod Genera. Mem. Peabody Museum of Nat.
Hist, 4, Pt. 1. New Haven, Conn.

Journal Royal Society of N.S.W., Vol. LXXXVI, 1955, Plate LX

MARTINIOPSIS WAAGEN FROM THE SALT RANGE, INDIA. 107

Schuchert, C., and Le Vene, C. M., 1929. Fossilium Catalogus, I. Animalia, Pars 42,
Brachiopoda. W. Junk, Berlin.
Termier, H., and G., 1948. La respiration et la circulation chez les Brachiopodes: leurs
répercussions sur la coquille. Bull. Soc. d’Hist. Nat. de
VAfrique du Nord, 39, 57.
—__—______—_—_—_—————. 1949. Sur la classification des Brachiopodes. Ibid., 40, 51.
Tschernyschew, 1902. Die Obercarbonischen Brachiopoden des Ural und des Timan. Mém.
du Comité Geolog., 16 (2), 557.
Waagen, W., 1883. Salt Range Fossils. Mem. Geol. Surv. India, Pal. Indica, Ser. XIII, 1, Pt. 4,
Fase. 2, p. 524.

EXPLANATION OF PLATE IX.
(All figures are natural size.)

Figs. la-ld.— Martiniopsis inflata Waagen. Specimen from which serial sections were made,
presented to the author by the Geol. Surv. India, 1938. Horizon: Lower Productus Limestone
(Permian). Locality: Right bank of Baral Nala, S8.W. of Amb, Salt Range. India. (Sheet
38P/15.) Geol. Surv. India (K31.326). Plaster cast of this is in Australian Museum (F44674).

la, dorsal view ; 1b, side view; lc, posterior view with ventral valve above; Id, anterior
view with ventral valve below.

Figs. 2a—2c.—Martiniopsis inflata Waagen. ? Young specimen. Horizon: Lower Pro-
ductus Limestone. Locality: Just N.W. of Amb, Salt Range, Punjab. Geol. Surv. India,
Reg. No. (K29/242). Presented to Australian Museum (F39605).

2a, dorsal view ; 2b, side view ; 2c, anterior view showing rectimarginate anterior commissure.

Figs. 3a—3c.—Martiniopsis inflata Waagen. Horizon: Lower Productus beds. Locality :
About 34 miles W.N.W. of Buri Khel, Salt Range, Punjab. Geol. Surv. Ind. Reg. No. (K31/126).
Presented to Australian Museum (F39604).

3a, dorsal view; 306, side view; 3c, anterior view.

Figs. 4a—4c.—Martiniopsis uralica Tschernyschew. Coll. Dr. G. Marshall Kay. Horizon:
Pseudoschwagerina princeps Zone, Sakmarian. Locality: Trataushikan, Bashkiria, Urals,
U.S.S.R. Australian Museum (F45581).

4a, ventral view ; 4b, side view; 4c, anterior view of ventral valve showing rectimarginate
anterior commissure.

Figs. 5a—5b.—Reproduction of illustration of ‘‘ Conchyliolithus Anomites (glaber) Martin,
1809 ”’, in “‘ Petrificata Derbiensis ’’, Plate 48, figs. 9 and 10, usually regarded as the type of
Martinia McCoy, 1844. (See also H. M. Muir Wood, Annals and Mag. Nat. Hist., Ser. 12, Vol. iv,
p. 109, Feb., 1951.) :

CONTRIBUTIONS TO A STUDY OF THE MARULAN BATHOLITH.
Part II. THE GRANODIORITE—QUARTZ PORPHYRITE HYBRIDS.
By G. D. OSBORNE, D.Sc., Ph.D., F.G.S.
Unwersity of Sydney,

and JOHN F. LOVERING, B.Sc.
Australian Museum, Sydney.

With Plates X, XI and one Text-figure.

Manuscript received, November 18, 1952. Read, December 8, 1952.

INTRODUCTION.

In Part I of this series of papers (Osborne 1949) it was explained that a
number of specialized petrological investigations upon various components of
the batholith (particularly regarding hybridism, igneous metasomatism and
related phenomena) would be necessary before the magmatic and tectonic history
of the complex could be elucidated and recorded. The present paper is the
result of intermittent research by the first author over several years, and of
recent study by the second, upon the hybrid-zone.

General Geological Setting.

The zone of hybrid-rocks (see Text-fig.) stretches from the township of
Marulan in a general south-south-east direction to the neighbourhood of South
Marulan, and outcrops over an area about four and a half miles long by one
mile wide.

The surface of the country in the Marulan-Glenrock district is part of the
Southern Tableland of N.S.W., here of gently undulating character, approxi-
mately 2,000-2,100 feet above sea-level, and topped by some residual ridges
rising to a height of 2,300 feet. The latter are composed of Lower Paleozoic
metasediments, mostly quartzites and cherty claystones.

In the region under discussion there is a noticeable ridge surmounted by
two low hills, on one of which stands Barber Trig. Station, 2,232 feet above
sea-level. The hybrid-zone outcrops almost entirely around the periphery of
this ridge, and the topographical details indicate the probability of the under-
surface of the Barber Hills chert, slate and quartzite mass being approximately
horizontal—an idea suggested long ago by W. G. Woolnough (1909).

Contiguous to the hybrid-zone to the west is the Marulan quartz por-
phyrite, outcropping in an elongated area running north and south. Further
westward this is succeeded by the Marulan ‘ granite’? and the Longreach
porphyry, which also trend in more or less meridional belts. The relationships
of these members of the batholith are complex and not yet unravelled, but
it is clear that the Marulan “‘ granite’ is not the product of direct magmatic
crystallization. Contiguous to the hybrid-zone on the east is the Glenrock
granodiorite, exposed on Glenrock Station and neighbouring pastoral properties.

CONTRIBUTIONS TO A STUDY OF THE MARULAN BATHOLITH. 109

FIELD OCCURRENCE OF THE HYBRIDS.

With the data available from the abundant outcrops in the paddocks and the
excellent exposures in portions of Marulan and Tangerang Creeks, it is possible
to delineate three sub-zones of the hybrid-belt. These are respectively :

(a) Sub-zone I, westernmost, passing into the quartz porphyrite,
(b) Sub-zone II, occupying a median position, and
(c) Sub-zone III, easternmost, passing very gradually into the grano-

diorite.
MAP SHOWING
HYBRID ZONE
BETWEEN MARULAN
QS AND
nye SOUTH MARULAN.

is a Upper Ordovician.
o

Des iF
ti Quartz porphyrite.
co Be Granodiorite

Field, petrographic and chemical data establish the hybrid character of all the
rocks in the region intermediate between, the granodiorite and quartz porphyrite.
As will be shown below, some rather unusual features are displayed by some of
the hybrids, such as partial or selective recrystallization ; such conditions have
tended to make difficult the determination of the sequence of intrusions:
responsible for the mixed rocks, but it is our view that the quartz porphyrite
was the earlier, and the granodiorite the later intrusion.

The following criteria have been used to establish this relationship :

(a) There is quite a number of xenoliths of fine-grained partially recrystal-
lized porphyrite in the granodiorite in the north-western part of the
zone.

110 OSBORNE AND LOVERING.

(b) Veins of tourmaline-bearing aplite, practically identical with syngenetic
veins in the Glenrock granodiorite, have occasionally been seen cutting
both the porphyrite and the westernmost Sub-zone I.

(c) All the tourmalinization in the Lower Paleozoic metasediments of the
region, especially near where the Southern Portland Cement Company’s
railway crosses Marulan Creek, is due to the granodiorite, and none is
associated genetically with the porphyrite, where the latter is seen to
invade sediments.

(d) Epidotization has been noted in some of the eastern marginal portions
of the porphyrite and in some of the hybrids. The occurrence of epidote
is a distinctive feature of the Glenrock granodiorite, being related to
deuteric phenomena, but is not characteristic of the quartz porphyrite.
It would be feasible to say that these facts support the view that the
epidote was introduced from the granodiorite, which was thus later
than the porphyrite.

Hybrids of Sub-zone I. The rocks constituting this zone are dark and very
fine-grained with few phenocrysts of quartz and pale-green plagioclase. Iron
pyrites, in small amounts, is widely disseminated. The texture of the rocks
varies and occasionally there are abrupt and patchy relationships between types
of varying grainsize and phenocrystic abundance. Towards associated phases
(either on the west or the east) there is a gradation.

Hybrids of Sub-zoneII. This is the main sector of the hybrids, and embraces
an area with abundant outcrops, where variety of texture and composition can
be seen macroscopically. The rocks are all dark in colour, good examples being
available for study near the Marulan Rectory, in the cuttings of the ©.P.0.
Railway, and to the north and south of Tangerang Creek.

It is clear that at South Marulan both normal hybridism and later silicifica-
‘tion and veining with felspathic material have operated in the production of
these rocks.

The chief varieties to be observed throughout the Sub-zone comprise
(a) almost lithoidal phases, which are seen under the microscope to be practically
biotite-free, and (b) finely crystalline rocks, with ‘‘ schlieren-like’’ patches,
containing much biotite in nests and clots. There are local patches of highly
felspathic material (see petrography) and some ‘ drusy ”’ structure in the light-
coloured veins.

Hybrid Sub-zone III. These rocks are best studied near Marulan Railway
Station and to the north-east thereof. They are also found fairly constantly
as a belt about 400 yards wide adjacent to the granodiorite. They are phanero-
crystalline, medium grained and fairly uniform in texture.

These hybrids resemble the granodiorite when partially weathered. The
freshest rocks are of a pinkish-grey colour and display macroscopically the
following: plagioclase (slightly epidotized), hornblende, biotite both in flakes
and small clots, orthoclase, quartz and occasional scattered pyrites. By increase
in grainsize and decrease in the amount of biotite and quartz the rocks pass into
the normal granodiorite.

Gradations and Widths of the Hybrid Sub-zones. Fortunately there are
sufficient exposures to establish the gradational relationships between all three
hybrid groups, and respectively with the porphyrite and granodiorite, which
were the compositional poles involved in the hybridization processes. From
west to east the relations are such that one can confidently assume that Barber
Trig. Ridge conceals hybrids mainly of the median zone and to some extent of
sub-group III.

CONTRIBUTIONS TO A STUDY OF THE MARULAN BATHOLITH. 111

The maximum width of hybrid-rock development is about 2,000 yards,
to be observed at the latitude of a point half a mile north of Marulan Creek.
In general it may be stated that the average widths of the three zones (from
which averages there are no great departures by extreme measurements) are:
Sub-zone I, 150 yards ; Sub-zone II, 600 yards ; Sub-zone III, 550 yards.

PETROGRAPHY.

The Granodiorite (Plate X, fig. 2). This rock is the western phase of the
Glenrock mass which has received some attention in earlier papers (Woolnough,
1907 ; Osborne, 1931, 1949). The Glenrock type of granodiorite is of more than
passing petrological interest and the authors are now preparing a paper (Part IV
of the present series) in which the petrology of this unit in the batholith will be
discussed.

Along the easternmost margin of the hybrids it is difficult to collect grano-
diorite which is free from some sign of infiltration of quartz, and of other processes
related to the hybridism. However, by collecting from about 50 yards away
from the last definite hybrid of sub-zone III, one obtains material typical of the
granodioritic parent.

The rock is speckled, slightly pinkish in colour and of medium, even grain.
Minerals visible are plagioclase (generally palest green, occasionally white),
amphibole, biotite, quartz and pyrites. Under the microscope it is seen that
some orthoclase (occasionally shightly perthitic), augite and very rarely hypers-
thene, sphene, apatite, zircon and iron oxide complete the list of primary mineral
constituents. The secondary minerals are saussuritic masses, clinozoisite,
epidote, prehnite, lawsonite (?), rarely scapolite, chlorite, leucoxene, hematite
and limonite, and in some cases much kaolin.

The fabric is frequently sub-monzonitic, achieved by the late crystallization
of quartz, and sometimes orthoclase, about the idiomorphic plagioclase.

The soda-lime felspar is zoned (An,,; to An,,), the most acid zone of this being
Sheathed by very clear additions of oligoclase, with irregular periphery but
Sharp interior boundaries against the main crystal. It would appear that the
addition of oligoclase was distinctly much later than the progressive growth of the
other zones.

Alteration to an irresolvable saussuritic material, and also to chlorite,
calcite, kaolin and to much epidote of deuteric origin is characteristic. The
rare plates of scapolite occur in cracks through the plagioclase. The potash
felspar is poorly cleaved and clouded by weathering. In places it is finely
perthitic. Biotite (8=—1-639) is brown and possesses characteristic intra-
cleavage masses of brightly birefringent material, which has caused slight
bulging of the host-mica, due to volume increase.

The nature of these intracleavage masses suggests that they are additions
along the planes of easy access, and are related to deuteric reactions. The
minerals occupying this role, and not to be confused with chloritic and muscovitic
derivatives from biotite, are clinozoisite, prehnite and lawsonite (?). Another
form of alteration is to confused masses of granular sphene plus leucoxene,
indicating a not inconsiderable proportion of TiO, in the mica.

The amphibole comprises light greyish-green hornblende with twinning,
and extinction Z*c=15°. There is an approach to idiomorphism, and chloritic
decomposition is typical. This mineral is seen sometimes to be developing from
pyroxene in the manner of the Reaction Principle. The other amphibole is a
very pale grey-green fibrous uralite, which may be the result of hydrothermal
influence. ‘The pale variety is close to tremolite in some optical properties, such
aS R.I.=1-616, and Z*c=18°, positive elongation.

112 OSBORNE AND LOVERING.

The pyroxene is the least constant of the constituents. There are frequent
cores of the augitic variety with twinning and poor cleavage, surrounded by
hornblende. Occasionally an idiomorphic crystal indicates the primary character
of the augite, and its antecedence to the hornblende in the order of crystal-
lization. The possible petrogenetic significance of the pyroxene in the Glenrock
granodiorite will be dealt with in a later paper. Hypersthene, partly altered
to bastitic material, is not common in the western margin of the granodiorite,
and therefore will not be considered in the present study. An analysis of the
typical western phase of the Glenrock Granodiorite (just slightly more basic
than the rock from much of the pluton) is given below and will be considered
in the chemical discussion.

Quartz Porphyrite (Plate X, fig. 1)—From Marulan, both north and south,
the distinctive Marulan quartz porphyrite may be seen. It is best exposed in
Marulan Creek and in the neighbourhood of the old Limekilns west of South
Marulan village.

The age and magmatic affinities of this mass in the complicated evolution
of the Marulan batholith are not yet known, but we can treat the present
discussion objectively and consider the rock as one of the parent types in the
hybridism.

There is a fair variety of appearance in hand-specimen, but the average
body-colour of the fresh rock is dark-greenish black to greenish grey, the ground-
mass betraying some recrystallization, and showing phenocrysts of pale green
plagioclase, pyroxene and quartz. Some of the rocks are relatively rich in
pyrites and there are fine-grained, almost aphanitic varieties, especially at the
eastern margin of the mass and against the Lower Paleozoic hornfelses, ¢.g.,
near Tangerang Creek.

Microscopically we note phenocrysts of plagioclase (average diameter
2mm.), An;, to Ans, measured from centre to penultimate zone. The outermost
layer is pure albite and represents later sheathing. Alteration is to turbid
masses in which chlorite and epidote can be determined. Quartz phenocrysts
with maximum diameters of about 3 mm. show embayment of periphery.
Orthoclase is sparingly present, but some K,O must be in the groundmass as
indicated by the analysis. The femic minerals are unsatisfactory to deal with.
Pale green hornblende with X=—greenish grey, Y=leek green, and Z=—green,
Y>Z>X, and Z*c=20° may occur in aggregates of ill-defined units with a good
deal of blotching and development of chlorite. The patchy and inhomogeneous
character of the hornblende suggest that it may have suffered some thermal
change, but it is not clear whether or not the aggregates have become stable in
the magma of the porphyrite groundmass. In places, brownish biotite in flakes
and prismatic units appear to be intimately associated with and perhaps to
have been derived partly from the hornblende.

The groundmass has corroded the phenocrysts indicating a former liquid
condition. It consists of cryptocrystalline material comprising silica and other
irresolvable matter. It has in some rocks been partially recrystallized, and this
will be discussed below.

An analysis of the quartz-porphyrite is given in Table I.

Hybrids of Sub-zone I (Plate X, fig. 3; Plate XI, fig. 3). Many of the
rocks are very similar to the porphyrite except that there are fewer phenocrysts
and the groundmass shows a greater amount of recrystallization, so that hitherto
ceryptocrystalline material is now converted to a finely textured quartzose mass
with hornfels structure. The rest of the rock is not so altered, but the pheno-
erystic minerals show some signs of changes having been wrought by fluids
derived from the granodiorite after its first main stage of crystallization had
finished, and hydrothermal or possibly pneumatolytic conditions obtained.

CONTRIBUTIONS TO A STUDY OF THE MARULAN BATHOLITH. 113

The soda-lime felspar has been strongly altered with considerable epidotiza-
tion. Hornblende and pyroxene have been chloritized and thin borders of
chlorite have been deposited around the peripheries of quartz and rarely of
orthoclase grains. Biotite is scarce and always comes from interaction between
hornblende and liquid from the second intrusion.

TABLE I.
(1) (2) (3) (4)
Si0, 73:54 63°95 63°86 57:66
Al,O; 10-95 15:24 15-84 17-36
Fe,O; 1-08 1-03 0:49 0:94
FeO 4°39 4°87 5-31 6:54
MgO 0-08 R53 2-42 4:07
CaO 1-68 D°o2 5-53 7-28
Na,O 2219s 2-40 1:96 1:56
K,O 4-80 PAN ah 2°13 1-38
H,O+ 0:21 Ied3 1-37 2224
H,O — 0:08 0-12 0:33 0:20
TiO, 0:60 0:52 0:66 0:55
MnO tr. 0:09 0:04 0-04
P,O; — | 0-09 0:08 0:14
CO, — tr. abs. abs.
100: 20 99-86 100-02 99-96
Spal. J. aes — PRATT | Dikulcs 2-82
Norms
(1) (2) (3) (4)
Quartz .. oF 33°60 23:04 25-74 17-28
Orthoclase a 28-36 12-79 12:23 8-34
Albite eis a 23°58 20°44 16:24 13-10
Anorthite. . a 3:34 24°19 26°97 35°31
Corundum i, —— —— 0:51 0:41
Diopside ts 4-43 1-17 —- | ~-
Hypersthene ade 3°93 14-68 14-22 20:36
Magnetite S 1-62 1-39 0-70 1-39
Ilmenite .. mes 22, 0:91 pe22 1-06
Apatite .. a — 0:34 0:34 0:40

(1) Quartz porphyrite, South Marulan. Anal. L. A. Cotton.
(2) Hybrid, Sub-zone I, South Marulan. Anal. W. H. Herdsman.

(3) Hybrid, Sub-zone II, near Tangerang Creek. Anal. W. H.
Herdsman.
(4) Granodiorite, South Marulan. Anal. W. H. Herdsman.

In one specimen near the head of Marulan Creek a thin vein of amphibole
(more or less uralitic) one half-millimetre in width was found cutting the por-
phyrite. This had resulted from the segregation of hornblendic molecules during
the hybridization, and later deposition of the same in cracks, which developed
on shrinkage of the porphyrite, following the increase of volume on the first
impact of the granodiorite magma.

114 OSBORNE AND LOVERING.

Hybrids of Sub-zone II (Plate X, fig. 4; Plate XI, fig.1). This is the most
important of the series. Microscopically one sees that there is considerable
variation in grainsize and in the proportions of the various minerals, but no great
differences in general mineralogical constitution. Average grainsize varies from
0-1 mm. to 0:35 mm. In the commonest types the minerals are present in
approximately the following proportions: quartz 28%, plagioclase 30%,
orthoclase 20%, hornblende 10%, biotite 10°, augite and accessories 2%. In
certain types, poor in biotite, there is an increase in hornblende, but the other
constituents remain in about the same proportions.

Quartz forms a kind of groundmass through which most of the other con-
stituents are set, giving a sub-monzonitic fabric. In other cases there are
sectors of crude-graphic, or coarse-aplitic crystallization of orthoclase and quartz.
The plagioclase crystals have many features in common with the phenocrysts
of the porphyrite, but there are distinct additional layers of oligoclase attached
to the altered, zoned units.

Orthoclase is non-perthitic, and the quartz is invariably recrystallized, so
that some large grains show a division into various sectors with sutured or
crenulated boundaries. The biotite is often ragged and wisp-like. Its
pleochroism is X=brass-coloured, Y—=vandyke-brown, Z=dark brown, with
Dy Ne AOS).

A study of the several phases of biotite-development in these rocks indicates
that the evolution has been

hornblende—reddish-brown biotite—brass-coloured biotite,
with evidence of TiO, being more prominent in the lighter-coloured and less
ferriferous varieties.

The hornblende occurs in clots which have aggregate-structure (Plate XI,
fig. 2). This is dependent upon sectors or groups of crystals being orientated in
preferred directions. Some of the hornblende is uralitic and some is ferriferous.
Varying double refraction and lack of uniformity in maximum extinction-angles
indicate the lack of equilibrium at the time of final congealing. As hybridism
has proceeded the passage of hornblende to biotite has increased, and in some
places complete change to mica has been achieved by interaction between
hornblende and the liquids available. Harker (1939) has recorded and figured
clots of biotite-crystals and flakes possessing the same structure as those shown,
in Plate XI, fig. 1. He ascribes them to action of porphyry liquid on granite
xenolith at Dundalk, Ireland. There has been removal of iron in the production
of light-coloured biotite. Intracleavage deposition of epidote and prehnite in
the mica indicates that deuteric conditions had developed.

Some of the biotite clots have been penetrated by salic liquid which has
crystallized as quartz and albite.

Thus the general heterogeneity, the occurrence of clots in various stages of
evolution, and the failure of the pyroxene to pass to lower members of the
reaction-series all suggest that the onset of cooling prevented this group of
hybrids from reaching finality of development.

Recrystallization. Partial recrystallization is common in rocks of Sub-
zones I and II. As a result, the siliceous mesostasis of the less altered por-
phyrites and the dominant quartzose ground of many of the hybrids show a
hornfelsic texture. A less common and somewhat puzzling type of recrystal-
lization is seen in the case of certain dark inclusions scattered through some of the
Sub-zone ITI hybrids, both near Marulan East and a little to the north of South
Marulan. These patches appear to be remnants of primary granodioritic
crystallization. Their petrographic features indicate that when recrystallization
of the quartzose portions of the porphyrites and hybrids was effected, local
greater elevation of temperature led to the formation of hypersthene. Thus we

CONTRIBUTIONS TO A STUDY OF THE MARULAN BATHOLITH. 115

find the following constituents—plagioclase (a little more acid than in the
granodiorite), orthoclase, augite and hypersthene, with a little biotite. The
fabric is sub-ophitic to hornfelsic, in which irregular sutured borders mark the
hypersthene, and the biotite has sieve-structure.

The biotite has resulted from reaction between pyroxene and the liquid in
which these xenoliths were immersed, with or without the intermediate formation
of amphibole. Deuteric action has produced clinozoisite, chlorite and calcite
from the felspars.

Hybrids of Sub-zone III. The rocks in this subzone present greater homo-
geneity over large areas than anywhere else in the main zone. Texturally they
resemble the granodiorite and the main minerals of that type are present, but
in addition there has been a distinct accession of quartz, which has ramified
through earlier-formed minerals. This late quartz has been recrystallized only
in small degree, and not so strongly as in the hybrids of zone IT, but the effects of
its renewal and penetration have led to cracking of the pyrogenetic quartz, due
to volume change under the thermal increase involved.

There is a notable increase in the amount of biotite at the expense of
amphibole, and in some places near Marulan Station deposition of considerable
sodic plagioclase has occurred, mantling the pyrogenetic individuals of the
soda-lime felspar.

No pyroxene is seen in this zone and it is possible that any which existed
has been made over into hornblende and/or biotite.

CHEMICAL DISCUSSION.

In Table I, page 113, are listed analyses of the porphyrite, of two hybrids
(Sub-zones II and III), and of the normal granodiorite from the south-west
portion of Glenrock Estate. Analysis No. 1 is an old one (recorded by Wool-
nough, 1909) and represents one of the phases of the quartz-porphyrites met with
both to the south-west of South Marulan and on the Limekilns Road about
four miles from Marulan. This type always shows some recrystallization and
effects of infiltration of silica and alkalic liquid. This rock becomes a source
of abundant biotite when hybridized.

As to chemical comparison of the rocks in the table, it is very difficult to
adopt any satisfactory basis for quantitative discussion. Number 3 is an
advanced hybrid, but it is not clear from the microscopic evidence whether it
has received much or little from the invaded rock by chemical reaction. Then
again, the quartzo-felspathic mesostasis may reflect some hydrothermal activity
which succeeded the main hybridism.

Concerning the relations of No. 2 and No. 3, it must be remembered that
the rocks as we see them now are the products of early crystallization of pyro-
genetic minerals, which later were affected mechanically by salic infiltration, and
to only a small extent by chemical action. Altogether it is felt that a direct
comparison of the analyses, even conducted with due attention to the relative
densities of the rocks, would not be valid or trustworthy in the present study,
because the hybridism was not carried to completion of equilibrium. Never-
theless we can note certain chemical relationships shown by the analytical data
and gain some help in this discussion by study of the norms.

The true hybrid is distinctly more acid than the unaltered granodiorite,
due to the presence of the acid mesostasis. It is somewhat higher in alkalis,
particularly potash, this being reflected in the abundance of biotite. The
proportions of CaO and MgO are related to the amount of mechanical incor-
poration of material crystallized from the granodiorite magma. The larger
plagioclase crystals and hornblende-biotite aggregates have come from this
source.

116 OSBORNE AND LOVERING.

Examining the hybrids we note further the intermediate position of rocks
with biotite-clots, because No. 1 has practically no MgO and low CaO, while
No. 4 has considerable quantities of both, and the hybrid shows amounts for
each lying between these extremes. The comparative values for the alkalis
again indicate the intermediate position of the hybrids. It is to be stressed that
the variations stated in this discussion would not be eliminated if one made
allowances for possible inequality of the volumes of rocks, based on relative
densities. In other words, the microscopic, field and chemical evidence all
indicate that the hybridism was attended by concentration of MgO and CaO in
the mixed rocks.

An interesting feature displayed by the chemical data is the close approach
to chemical equivalence of rocks No. 2 and No. 3. This bears out the idea
that some of the quartz porphyrite, which has been taken in the field to be un-
hybridized, is on petrological grounds suspect, on account of the origin of the
phenocrysts. It appears possible that some of the eastern marginal porphyritic
rock may be a hybrid more advanced than much of the reaction-zone, having
reached a state of equilibrium. The quartz, orthoclase, biotite and large derived
phenocrysts of plagioclase would then be regarded as having reached stability
in the magma, the last-mentioned having become armoured by the deposition of
thin external zones of albite.

THE COURSE OF THE HYBRIDIZATION.

After the consolidation of the quartz porphyrite (probably as a sill-like
pluton amongst the Lower Paleozoic metasediments) the rocks occupying the
present Marulan-Glenrock-Ballanya region were invaded by a mass of grano-
diorite magma. There is no suggestion of granitization around this mass and
the prevalence of xenoliths at the margins and about the roof of the intrusion,
as well as the petrographic character and textural uniformity, support the view
taken here that the Glenrock granodiorite was emplaced in a liquid condition,
thus being the result of magmatism and not metasomatism.

This intrusion was chilled quickly against the sedimentary roof and wall-
rocks on the north, east and south-east, but against the western wall of quartz
porphyrite hybrid reactions took place. Possibly this was due to the fact that
the porphyrite was not wholly cool at the time of the invasion of granodiorite,
and also because more strongly endothermic reactions would be involved between
granodiorite magma and the metasediments than between porphyrite and
granodiorite. Further, it seems probable that the present western sector of the
district was underlain by an igneous complex, partly solid, partly liquid, and
possessed of sufficient heat to bring about a regional rise of the isogeotherms.
Thus the western invasion-margin of the granodiorite was kept at an elevated
temperature, whereas on the eastern side of the pluton the fall of temperature
was relatively rapid.

Hybrids were developed by the interaction of magma and solid porphyrite.
Three distinct zones developed, as described above. There is evidence of little
more than a limited chemical change, but of considerable mechanical change.

We may trace the hybrid-evolution as follows :

(1) After limited thermal alteration of the quartz porphyrite by the grano-
diorite magma, its primary crystallization-period set in, during which
hybrid relations began to develop. The magma crystallized in pyro-
genetic plagioclase, amphibole, biotite, orthoclase, quartz and limited
pyroxene. Contemporaneously, material of porphyrite groundmass
and smaller phenocrysts was mobilized and resulting salic liquid set
free.

CONTRIBUTIONS TO A STUDY OF THE MARULAN BATHOLITH. ALF

(2) Before the orthomagmatic period had ended, some of the salic (potassic)
liquid which was available near the porphyrite reacted as follows—
(a) chemically with the phenocrysts of the quartz porphyrite,
(b) chemically and mechanically with the partially crystallized
network of the granodiorite.
As a result of (a) reddish-brown ferriferous biotite developed from
hornblende and chlorite from pyroxene. As a result of (b) various
phases of hybrid-margins to the granodiorite developed. These were
produced by the salic liquid disintegrating the pyrogenetic components,
and setting into operation the reaction, series,
pyroxene—amphibole—biotite.

(3) The migration of soda (or albite-oligoclase molecules) from the grano-
diorite led to deposition of fringes or sheaths of sodic felspar in the
hybrids. Sphene dissociated and the CaO and SiO, remained, while
the TiO, moved towards the hybrid to be fixed in biotite.

(4) Potash from the groundmass and from some phenocrystic material

in the porphyrite was fixed in (i) abundant biotite (light coloured) and

in (ii) considerable orthoclase of the hybrids. The light biotite was

stable in the magmatic environment. The following reaction sequence

seems to be a characteristic feature of the hybridism, viz.—
hornblende-+reddish biotite—titaniferous biotite.

Continual partial fluidity in the hybrids (with concentration of apatite

molecules) was effected by strong deuteric or hydrothermal activity in

the granodioritic magma. When the orthomagmatic stages were
completed, vigorous solutions wrought notable changes producing
saussuritic masses, epidote, sericite, etc., through the granodiorite,
the hybrids and the easternmost porphyrite. The effect of these
solutions was to intensify the final aqueous activity which permitted
the evolution of the micas from the hornblende and the latter (in small
measure) from the pyroxene.

(6) Finally, thermal changes caused recrystallization of various components
in the batholith (see above).

The cause of the later thermal change is a matter for further investigation.

The authors think that the intrusion of the granite and porphyry to the west

of the quartz-porphyrite was responsible for a general engulfing of the pre-

existing components of the batholith and its hybrids. This problem will be
taken up in a later contribution.

It remains to be stated that no comparative studies of other hybrid
occurrences are considered here, as we feel that such would be appropriate when
more is known of the history of the batholith.

s

ACKNOWLEDGEMENTS.

The senior author wishes to record his indebtedness to the Commonwealth
Research Grants Committee for financial help in this work, and the junior
author his thanks to the Trustees of the Australian Museum for permission to
collaborate in the present paper.

REFERENCES

Harker, A., 1939. Metamorphism, Plate 45a, p. 114. Pub.
Osborne, G. D., 1931. Contact Metamorphism and Related Phenomena in the Neighbourhood
of Marulan. Geol. Mag., 68, 289.
———_-—————_ 1949. Contributions to the Study of the Marulan Batholith. Part I. Tuis
JOURNAL, 82, 116.
Woolnough, W. G., 1909. The General Geology of Marulan and Tallong. Proc. Linn. Soc.
N.S.W., 34, 782.

118 OSBORNE AND LOVERING.

EXPLANATION OF PLATES.

PLATE X.

Fig. 1.—Quartz porphyrite from one mile §.E. of Marulan. This is adjacent to Hybrid
Sub-zone I and is typical of the least modified parent rock. Corroded quartz, orthoclase (near
centre), plagioclase (in N.E. corner), and chloritized hornblende (dark patches). The groundmass
is cryptocrystalline. Ordinary light. x17.

Fig. 2.—Granodiorite from a little to the north of Marulan Creek at private railway crossing.
Large altered plagioclase (in 8.W.), orthoclase, cleaved hornblende (towards bottom of picture),
dark biotite with intra-cleavage clinozoisite, late crystallizing quartz. Ordinary light. x17}.

Fig. 3.—Hybrid from Sub-zone I. Rock is partially changed by development of large
hornblende phenocrysts. Plagioclase is to be seen to the S.W. and the fine groundmass has been
partly recrystallized. Ordinary light. x 233.

Fig. 4.—Advanced Hybrid from Sub-zone II. Acid mesostasis has been recrystallized.
Biotite on right and saussuritized plagioclase on the left side of picture. Groundmass completely
recrystallized to give hornfels texture. Ordinary light. x 28.

PLATE XI,

Fig. 1.—Hybrid from Sub-zone II at South Marulan. Shows large clot of biotite in a
decussate aggregate surrounded by recrystallized groundmass. Rock has been thermally
metamorphosed after hybrid-reaction. Ordinary light. x40.

Fig. 2.—Partially hybridized rock. Shows development of large patch of hornblende, which
is changing over to aggregate of small biotite units. Above hornblende to N.W. is sericitized
felspar and to right and along base of photograph recrystallized quartz is to be seen. Ordinary
light. x28.

Fig. 3.—Hybrid from Sub-zone I. Large saussuritized plagioclase to the right, quartz in
S.W. quadrant and biotite in N.E. corner, with strongly recrystallized groundmass. The dark
spots in groundmass are chlorite and the light material is quartz. Ordinary light. X57.

eal a

Journal Royal Society of N.S.W., Vol. LXXXVI, 1953, Plate X

Journal Royal Society of N.S.W., Vol. LXXXVI, 1953, Plate XI

THE REPLACEMENT OF CRINOID STEMS AND GASTROPODS BY
CASSITERITE AT EMMAVILLE, NEW SOUTH WALES.

By L. J. LAWRENCE,
Department of Geology, New South Wales University of Technology.

With Plate XII.

Manuscript received, November 18, 1952. Read, December 3, 1952,

INTRODUCTION.

The country rocks of the Emmaville District or Vegetable Creek tin field
consist of Permian claystones, conglomerates and breccias. These are very
sparingly fossiliferous. Late Permian intrusives, notably granites and quartz-
felspar porphyries, have given rise to veins, irregular stockworks and impregna-
tions of cassiterite-bearing quartz. In a few instances stanniferous veins and
stockworks have been developed within the intruded sediments.

Post-Permian erosion has given rise to vast alluvial tin deposits. These
have been covered by Tertiary basalts present now as scattered residuals rising
up to fifty feet above the otherwise peneplaned surface. Tin recovery thus
comprises reef-mining, deep-lead mining (driving into stanniferous alluvium
through or beneath the Tertiary lavas) and sluicing the existing streams.

In the various streams and creeks at Emmaville there occur numerous types
of alluvial cassiterite: ruby, resin, wood and toad’s eye, etc. The most
interesting variety, however, is the one termed by the local miners ‘‘ screw-
tin’’.* These have been identified as crinoid stems replaced by cassiterite.
They have never been found in situ. Associated with the ‘‘ screw-tin”’ are
occasional gastropods, also replaced by cassiterite. Hodge-Smith (1943)
originally drew attention to these remarkable fossils.

The mode of replacement is somewhat obscure. The fossils were
undoubtedly contained in the Permian sediments originally as calcareous
remains. The sediments were intruded and silicified. Judging by the complete-
ness of this silicification one might assume that even the fossil content would have
been silicified. However, the only fossils so far collected from this area were
found in Recent alluvium and have been entirely replaced by tin dioxide.

It is significant that the so-called wood-tin is also found in the same alluvium.
The name is an apt one. The wood-tin is of a pale brown colour ; the fragments
are inclined to be somewhat tabular in shape and up to an inch and a half in
length and are notably dissimilar to any other type of cassiterite found in the
New England. It is not improbable that wood-tin represents fragments of the
sedimentary country rock which have been replaced by stanniferous solutions.
It may be that these solutions, passing through the bedding and jointing in the
sediments, reached a position where metasomatic replacement of both sediment
and fossil content ensued. Subsequent mechanical disintegration might have

* So called because of their striking resemblance to worm-screws.

120 L. J. LAWRENCE.

produced fragments of wood-tin and the stanniferous fossils. It should be
pointed out, however, that metasomatic replacement of the sediments is not a
general feature of the Emmaville mineralization, as far as can be ascertained.

It is interesting to note that throughout this extensive alluvial tin-field
fossils so replaced have been found only in two small non-perennial creeks :
Steele’s Gully and Charcoal Gully, tributaries of Doctor’s Gully. They have
never been found in the main waterway, the lucrative Vegetable Creek, nor in
any of its more prominent tributaries.

The stanniferous crinoid stems were, at one time, fairly plentiful. Many of
them undoubtedly found their way into the smelters together with the associated
alluvial tinstone. The gastropods were very rare; no more than a dozen or so
of these are extant. Further finds are likely, however, since the leases at Doctor’s
Gully are to be reopened in the near future.

DESCRIPTION OF THE FOSSILS.

(1) The Crinoid Stems. These are generally quite circular in cross-section ;
some are oval. The diameter ranges from 1cm.to2mm. In length they rarely
exceed 2:5 cm. The average length, irrespective of diameter, is about 1 cm.

The colour of the specimens ranges from dark steel-grey to reddish-brown,
the latter being due to a thin encrustation of iron oxide. All specimens show
textural homogeneity under megascopic examination.

In each specimen the constituent ossicles are quite apparent. The ratio of
length of ossicle to diameter varies slightly from specimen to specimen. Partial
telescoping of adjacent ossicles was noted in a few cases. There were on the
average, ten ossicles per centimetre of length.

In those specimens that have not suffered excessive attrition, the central
ligament canal is readily discernible. It consists of a small indentation centrally
placed within a larger circular depression. This latter depressed area is rimmed
concentrically by a circular platform which occupies about half the diameter of
each specimen. Radially disposed around this platform are the alternate ridges
and furrows which serve as an interlocking device (Plate XII, figs. la-1d).

Longitudinal thin sections show (Plate XII, fig. 2a) that the crinoid stems
have been completely replaced by cassiterite. The entire stem appears as a
homogeneous aggregate of fine-grained cassiterite. In ordinary light the
aggregate is of a pale brownish colour, with occasional opaque patches of reddish-
brown iron oxide. Running through the centre is a zone of coarser-grained
cassiterite. This zone was originally the central ligament canal. In this region
of greater pore space the cassiterite has assumed a coarser texture. This also
applies to a number of zones set at right angles to the ligament canal. These are
undoubtedly due to replacement of zones where adjacent ossicles have been
parted slightly. Under crossed nicols the replacing grains, especially the larger
ones, show pearly interference colours indicating extremely high birefringence.

In transverse sections (Plate XII, fig. 2b) one sees, again, an aggregate of
cassiterite grains all closely interlocking and showing patches of iron oxide.
In Plate XII, fig. 2b the central ligament canal is not apparent ; this is due to
the fact that the section has been cut through a zone of parting between ossicles.
The effect of this can be seen by studying Plate XII, fig. 2a.

The identity of the crinoids which yielded these stem fragments has never
been established positively. J. E. Carne (1911) reports Mr. W. 8. Dun, former
Paleontologist to the New South Wales Geological Survey, as suggesting that
ealeareous crinoid stems from an adjacent region may belong to the genus
Phialocrinus. No reasons were given for the above suggestion but it is possible
that calices of this genus occur in the area referred to by Dun.

REPLACEMENT OF CRINOID STEMS AND GASTROPODS BY CASSITERITE. WAL

(2) Gastropods. These are also of a reddish-brown colour due to a coating
of iron oxide derived, no doubt, from the decomposition of the Tertiary basalt
which formerly covered the alluvium.

It is probable that two different species (or sub-species) are represented.
Of the specimens examined some had apical angles of from 65° to 75° (Plate XII,
figs. 3a and 3c). These had a ratio of height to diameter of last whorl of 1.: 1.
Others had an apical angle of from 38° to 42° and gave a height-diameter ratio of
3:2 (Plate XII, fig. 3b). Ptycomphalina morrisiana McCoy (Permian) is known
from a number of localities adjacent to the Emmaville District. The specimens
here described would appear to belong to this genus (Hodge-Smith, 1943).
Although most of the cassiterite replacements have suffered abrasion, one or two
Specimens show what is probably a selenizone (Plate XII, fig. 3b). In all cases
the whorls, of which there are four, increase rapidly in size. The mouth is large
and is situated slightly away from the axis of coiling. A small umbilicus was
observed in a couple of specimens. Because of the rarity of the gastropods no
thin sections were made.

The following table of specific gravities is of interest :

Crinoid Stems. Gastropods.
(a) 6°688 (a) 5:°785
(b) 6-378 (b) 5-281
(c) 6-504 (c) 5-511
(d) 6-560

Dana gives the specific gravity of cassiterite as ranging from 6:1 to 6-99
(Palache et al., 1944).

It is interesting to note that in all cases the values for the gastropods are
lower than those for the crinoid stems. This is probably due to the presence of
small quantities of clay or of iron oxide in the body cavities and umbilicus.

Chemical analysis of a crinoid stem gave the result in Column I.

Column I. Column II.
SnO, Sis Hs oe Oac0 Calcium phosphate 80-04
Al,O, es apt me Vela) Calcium carbonate 2-24
Si0, ee aes .. 12-4 Calcium fluoride .. 0-50
Fe,0O, ae bi oye 6-0 Ferrous disulphide 1-66
MgO a ae ae 0-2 Ferric oxide x Oe 0-62
CaO ae the .. trace Stannic oxide 2-60
Silica ve ae a 0-22
Total .. ~=, 99-8 Organic matter and loss. . 12-12
Total ne .. 100-00

It is clear that the analysed specimen had a cassiterite tenor distinctly lower
than would be expected from the specimens whose specific gravity measurements
are given above. These specimens of greater density and higher cassiterite
content were not available for chemical analysis as they constitute type specimens
in the collection of the Australian Museum, Sydney.

J. H. Collins (1888) reported on the replacement of deer horns by cassiterite
in the stanniferous alluvium of Cornwall and gave an analysis quoted in Column IT
above. He stated that ‘‘ The oxides of tin and the iron pyrites have found their
way to the interior of the bone and are visible throughout the structure, although
somewhat more abundant near the periphery ”’.

Since the deer horns could not be replaced by direct emanations from the
Late Paleozoic tin-bearing intrusives of Cornwall, their replacement must be
either the result of penetration by secondary stanniferous solutions or be due to
mechanical infilling by finely-divided mineral

122 L. J. LAWRENCE.

The degree of replacement of the deer horns from Cornwall is very small
compared with that of the Emmaville specimens.

The author wishes to thank Messrs. R. Millett and K. Nash of the C.S.1.R.0O.,
Sydney, for the photographs accompanying this paper.

.

REFERENCES.

Carne, J. E., 1911. Tin Mining Industry of N.S.W. Min. Res. Geol. Surv. N.S.W., 14.
Collins, J. H., 1888. Cornish Tin Stones. Min. Mag., 30, 4. /
Hodge-Smith, T., 1943. Rec. Aust. Mus., 21, 4.

Palache, et al., 1944. ‘‘ Dana’s System of Mineralogy ”’, 1, 575.

EXPLANATION OF PLATE XII.

Figs. la—ld.—Crinoid stem fragments replaced by cassiterite. Loc., Emmaville, N.S.W.
( x8.)

Figs. 2a—2b.—Thin sections of stanniferous crinoid stems. 2a, longitudinal section
2b, transverse section. Ordinary light. (x 10.)

Figs. 3a—3b.—Gastropods replaced by cassiterite. Loc., Emmaville, N.S.W. (x4.)

Journal Royal Society of N.S.W., Vol. LXXXVI, 1953, Plate XII

3a 3b.

Specimens la, 6, c and d are in the Australian Museum; specimens 3a, 6 and c are in the
possession of Mr. L. Collins, Emmaville.

INDEX

A
Page
A Contribution to the Geology and
Glaciology of the Snowy LEN
New South Wales va
A Geological Account of Heard Tsland. 14

Alteration of Rule IX .. ; Ma vorauil
Annual Dinner of the Society Sea Sari
Annual Report of the Council vu
Atherton Tableland, Climate and Maize
Yields on the .. ; 22
Authors, Guide to nee ats Biielar hig
Awards of the Society .. ag A GN
B
Balance Sheet .. tae ae Sa Vina
Bequest, Form of Sh EEN
Birrell, Septimus. — Obituary Notice .. xiv
Bolton, J. G.—Award of Edgeworth
David Medal for 1951 i iat aval
Bosworth, R. CC. L.—Presidential
Address—
I. The Society’s Activities 1
II. Transport Processes in Applied
Chemistry 2 3
Brown, I. A.—
Martiniopsis Waagen from the Salt
Range, India : 22 OO
Permian Spirifers from Tasmania .. 55
Cc
Cheel, Edwin.—Obituary Notice Sp hi
Clarke Memorial Medal—
For 1951 ae ee sos enV
For 1952 : EAL
Climate and Maize Yields on the
Atherton Tableland .. 22

Coastal Limestones of Western patel)
Soil Horizons and Marine Bands in
the oe .. 68
Commemoration a Great Scientists Bian
Contributions to the Study of the
Marulan Batholith. Part II, The
Granodiorite- le Porphyrite
Hybrids . : : 108
Genvermavione—: ubilee Ae Bento bt!
Crinoid Stems and Gastropods, The
Replacement of, by Cassiterite at
Emmaville, New South Wales mai td bli)

D

David Medal, 1951, Awards of the
Edgeworth i ae a av

Davies, N. R., and Dwyer, F. P.—
Induced Optical Activity of the Tris-
1:10-Phenanthroline and Tris-2 : 2’-
Dipyridyl Copper II Ion 3 . 64

Page
Distinguished Visitors .. ool WE
Drane, N. T.—See Simonett, D. Ss and
Drane, N. T.
Dwyer, F. P.—See Davies, N. R., and
Dwyer, F. P.

E

Edgeworth David Medal for 1951 Sv anwall

Emmaville, New South Wales, The
Replacement of Crinoid Stems and
Gastropods by Cassiterite at LG

Exhibits .. : ie aE, vli, Xlll

F
Fairbridge, R. W., and Teichert, C.—

Soil Horizons and Marine Bands in
the Coastal Limestones of Western

Australia As oe Zils it IOS
G

Geology and Glaciology of the Snowy

Mountains, A Contribution to the .. 88
Geology, Report of the Section of ord A
Government Grant to the Society sat VALUE
Grantonia hobartensis Brown, 1953, gen.

et spec. nov. aN 61

Graptolite Zones and Associated Strati-
graphy at Four-Mile Creek, South-

west of Orange, N.S.W. 94
Gregg, N. McAlister—Award of the
James Cook Medal, 1951 oN : vi
Guide to Authors ; ss fat AN
H
Heard Island, A Geological Account of.. 14
Hoggan, H. a —Obituary Notice 2 a SGLV
Honorary Member a V, Vill
I

India, Martiniopsis Waagen from the
Salt Range We 100
Induced Optical Activity of the Tris-
1: 10-Phenanthroline and Tris-2 : 2’-

Dipyridyl Copper II Ion ie .. 64
J
James Cook Medal for 1951 .. Vi

Johnson, T. Harvey.—Obituary Nonicet XIV
Judd, W. P.—Obituary Notice RY p- SAY

XVill
L
Page
Lambeth, A. J.—A Geological Account
of Heard Island ; 14

Lawrence, L. J.—The Replacement of
Crinoid Stems and Gastropods by

Cassiterite at Emmaville, N.S.W. .. 119
Library, Report on Society’s Ma Ula b-¢
List of Members .. Vv

Liversidge Lecture—Electron Diffraction
in the Chemistry of the Solid State.. 38

Livingstone, S. E.—Palladium Com-
plexes. Part V. Reactions of Pal-
ladium Compounds’ with 2: 2’
Dipyridyl 32

Lovering, J. F.—See Osborne, Gg. D.,
and Lovering, J. F.

M

Maitland, A. Gibb—Obituary Notice .. xv
Maize Yields on the Atherton Tableland,

Climate and _.. 22
Martiniopsis Waagen from the Salt
Range, India .. 100

Marulan Batholith. Part II. The
Hybrids, Contributions to the eee.

ofthe s,. ; : 108
Members, List oe af: Be ay Vv
N
Notices... an a x 5a UAL
O
Obituary .. XV

Occultations @heerved oh Sydney Ob-
servatory during 1951 ; 20
‘Officers for 1952-1953... il

Orange, N.S.W., Graptolite Zones ea
Associated Stratigraphy at Four-Mile
Creek, South-West of 94

Osborne, G. D., and eo vorne. a. ipl
Contributions to the Study of the
Marulan Batholith. Part II. The
Granodiorite- eee oe ieee)
Hybrids . a 108

P

Packham, G. H.—See Stevens, N. C.,
and Packham, G. H.

Palladium Complexes. Part V. Re-
actions of Palladium Compounds with
2:2’ Dipyridyl 32

Penfold, A. R.—Award of the Society’s S

Medal, 1951 ; ulna

Permian Spirifers from Tasmania AE TOD

Popular Science Lectures Re Shh gibt

Presidential Address ee at att ]

INDEX.

R

Page

Rees, A. L. G.—Liversidge Lecture—
Electron Diffraction in the ee
of the Solid State

Report of the Council

Ritchie, A. S.—A Contribution to the
Geology and eS of the Snowy
Mountains

Robertson, W. EL .»» and ‘Sime, ice pit
Occultations Observed at pel aes
Observatory during 1951

Rule IX, Alteration of

S

Salt Range, India, Martiniopsis Waagen
from the :

Science House Management Committee,
Society’s Representatives.

Simonett, D. S., and Drane, N. ps
Climate and Maize Yields on the
Atherton Tableland

Sims, K. P.—See Robertson; W. H.,
and Sims, K. P.

Snowy Mountains, A Contribution to the
Geology and Glaciology of the

Society’s Medal for 1951 aa

Soil Horizons and Marine Bands in the
Coastal Limestones of Western
Australia : :

Spirifers from Tasmania

Stevens, N. C., and Packham, Jes
Graptolite Zones and Associated
Stratigraphy at Four-Mile Creek,
South-West of Orange, New South
Wales °c. ae ne aN Ke

Stillwell, F. L.—Award of the Clarke
Medal for 1951 sits

Stratigraphy at Four-Mile Creek, South-
West of Orange, N.S.W., aoe a

Zones and Associated 4

Subscription, Alteration of Rule IX

Sydney Observatory during 1951, Oc-
cultations Observed at

T

Tasmania, Permian Spirifers from :

Teichert, C.—See Fairbridge, R. W., sen
Teichert, C.

The Replacement of Crinoid Stems and
Gastropods by Cassiterite at EKmma-
ville, New South Wales : wif

Transport Processes in Applied
Chemistry—Presidential Address

Trigonotreta stokesii Koenig, 1825

Ww

Western Australia, Soil Horizons and
Marine Bands in the Coastal Lime-
stones of

Wood, J. G. arvana of the Clarke
Memorial Medal for 1952

38

vii

88

20

. Vill

. 100

owl

22

Vili

55

68

. vill

ee

-
Si

—s

ts. Gh 4b

Fer i, ha
a ge

; 92

Beta

Kes.
re

ates

4

and Seamer Str

\SIAN. MEDIC
Arundel

AI

‘WINN

4744