•k'. K ^ -■ [Registered at the General Post Orlice, Perth, for transmission by post as a Peri odical.] JOURNAL or THE ROYAL SOCIETY or WhSTLRN AUSTRALIA, INC. Founded 1913 :: :: Incorporated 1937 Tlie Authors of Papers are alone rosponsilile for tlie statements a nd the opinions expressed therein. Printed for the Society by Kobkht 11. Millkk, (lovennnenl Printer. Perth. l!)4;j. j 8572/4:i JoxjR. Roy. Soc., Western Australia, Vol. XXVII. THE ROYAL SOCIETY OF WESTERN AUSTRALIA, Ixc. OFFICERS AND COUNCIL, 1940-41. Patron : * His Majesty the King. Vice-Patron : His Excellency Sir James Mitchell, K.C.^LG., Lieut. -Governor of the State of AVestern Australia. President : Professor A. H. Ross, M.A., D.Sc., F.R.S.E., E.R.A.S., F.I.P. Past President : H. Rowley, F.A.C.L Vice-Presidents : 0. A. Gardner, R. C. AA’ilson, B.Sc., B.E. Joint Hon. Secretaries : R. T, Prider, B.Sc., Ph.I). C. F. H. Jenkins, ALA. Hon. Treasurer : E. AL Watson, B.Sc., Ph.D., D.LC., A.I.C., F.A.C.I. Hon. Editor : B. L. Southern, A.A.C.I. Hon. Librarian : Mrs. C. F. H. Jenkins, M.Sc. Council : Aliss A. M. Baird, B.Sc. H. W. Bennetts, D.V.Sc. F. G. Forman, B.Sc. L. Glauert, B.A., F.G.S. R. J. Little. G. R. W. Meadlev, B.Sc. H. P. Rowledge, A.A.C.I., A.W.A.S.AI. L. J. H. Teakle, B.Sc. (Agric.), Ph.D., A.A.C.I. Hon. Auditors ; J. Shearer, B.A., M.Sc. W. F. Cole, B.Sc. NfcTiONAL MUSEUM OF VICTORIA JouK. Roy. Soc., Western Australia, Vol, XXVII. iii. CONTENTS, VOLUME XXVII. Page Office Bearers ... ... ... ... ... ... ... ... ii. Annual Report ... ... ... ... ... ... ... ... iv. Treasurer’s Report ... ... ... ... ... ... ... ... ... ... vi. Proceedings — Abstract of ... ... ... ... ... ... ... ... ... vii. General Index ... ... ... ... ... ... ... ix. 1. Results of a Microscopic Studyof Some Soil Colloids. By W. F. Cole ... ... 1 2. Some Kyanite-Bearing Rocks from the Eastern Goldfields, Western Australia. By R. T. Piider 9 3. The Contact between the Granitic Rocks and the Cardup Series at Armadale. By R. T. Prider 27 4. Further Permian Corals from Western Australia. By Dorothy Hill ... ... 57 5. New Forms of Australian Birds. By Gregory Mathews ... ... ... ... 77 6. Some Basalts from the North Kimberley, Western Australia. By A. B. Edwards 79 7. Note on Fossil Corals from Langley Park Bore, Perth. By John W. Wells ... 95 8. Foraminifera and a Tubicolous M'orm from the Permian of the North-West Division of Western Australia. By Walter J. Parr ... ... ... ... 97 9. The (Seeds of Slnjrhnos lucida^ R.Br., and their Alkaloid Content. By E. M. Watson 117 10. Notes on the Vegetation of the North-Eastern Goldfields. By Nancy T. Burbidge 119 11. Beraunite — from Dandaragan, Western Australia. By C. R. Le Mesurier ... 133 12. Marine Copepoda from Western Australia. By A. G. Nicholls ... ... ... 135 13. New Leaf-Hoppers {(Homoptera, Jassoidea) from Western Australia. By J. AV. Evans ... ... ... ... ... ... ... ... ... ... 143 14. Contributiones Florae Australiae Oceidentalis XI. By C. A. Gardner 165 15. Mineralogy the Donnybrook Sandstones, Western Australia. By Dorothy Carroll 211 16. The Ocean Birds of the Perth Beaches. By L. Glaucrt ... ... ... ... 219 17. X-ray Analysis (by the powder method) and Microscopic Examination of the Products of AVeathering of the Gingin Upper Greensand. By AV. F. Cole ... 229 18. The Geology and Physiography of the Gosnells Area. By C. E. S. Davis ... 245 19. The Geology and Physiograi)hy of the AVongong-Cardup Area. By B. P. Thomson ... ... ... ... ... ... ... ... ... ... 265 IV. Jour. Roy. Soc., Western Australia, Vol. XXVII. The Royal Society of Western Australia (Inc.). ANNUAL REPORT OF THE COUNCIL FOR THE YEAR ENDING 30th JUNE, 1941. Ladies and Gentlemen, Your Council begs to submit the following report for the year ended 30th June, 1941. Council.— The Council of the Royal Society met on five occasions and the Executive Committee on six occasions. As in previous years an Executive Committee was appointed to deal with the routine business. FinanGe . — ^The General Fimd shows a balance of £135 1 (is. 3d. The Endow- ment P\ind now amounts to £245. It may be noted that the cost of tlie steel library shelving purchased by the Society for installatioTi in the Glerlden Building was met by the Entlowment Fund. The surplus of assets over liabilities now amoimts to £507. Volume XXVI. has not yet been com- pletely paid for, so that there are cej'tain commitments to the Government Printer against the General Fund. The very satisfactory condition of tlie Society's finances are due to the very able nianagement of the Honorary- Treasurer, Dr. E. M. Watson. Dr. Watson retires from the office of Hon. Treasuu'r to become a Vice-President of the Society and the Council wishes to express their thanks for fus services as Treasurer during the past four years. Membership.-— There has been a slight increase in membership of the Society during the year. Four ordinary members and one associate have resigned and two associates ha\^e transferred their membership — one to ordinary and the other to corresponding membership. The names of four associates liave been removed from the register in accordance with Rule 6. Members elected during the year include ten ordinary members and tliree associates. There are at ])re‘sent 154 members of tJie Society, made up as follows : - - Honorary Mi'inbers .... .... 7 C Corresponding Members .... .... 8 Life Member .... . .. .... , ] Ordinary Members .... .... .. . 98 Associate Members .. . .. . .... 35 Student Members .... .... .... 5 Journal .- — Volume XXVI. has been completed and is at present being distributed to members and scientific institutions with which the Society is in exchange. During the j^ear twenty papers were presented to the Society for publica- tion in the Journal. Fourteen of these are already' in the hands of the Gov- ernment Printer. JouK. Roy. Soc., Western Australia, Vox.. XXVII. V. Mr. Sniithern remains in the office of Hon. Editor and tJie Society is indebted to him for continuing to place his experience in this work at its disposal. The Council also desires to thank the Government Printer and his staff for their co-operation with the Hon. Pklitor in connection with the pub- lication of the Journal. Lihntn^.- -Tho Lifu'aiy is now housed in the Professional Societies’ Rooms at the Gledden Ihiildijig, wliere it forms a distinct section of the combined Library of the \arious Proh'ssionat 8ocieti(‘s. During the year new steel' shelving to conform with that in tin* Gledden Building Library was purchased by the Society from the Enduwmt*nt Fund. The Society has now entered int(> exchange of publications with a total of 184 scientific institutions, of wliich ;34 are in Australia, 16 in the LTnited Kingdom, 23 in otlier parts of the British Einpirt*, 46 in North and South America, 42 in Europe, and three in Asia. The Council wishes to express its thanks to Mr. A. Gibb Maitland for a number of donations to the Socitdy’s Library made during the past year. Housing of the Socieiif. — ^En-rly in the present year ari'angements \\^(U'e finalised with the Institution of Engineers, anrl the Royal Society has since been housed, along with the other Professional Societies, in the Gledden Buikl- ing. In addition to the improved meeting rooju, members will bejietit by this change, as they -will now be able to refer to \'olumes in the libi'aries of the \-arious other Professional Societies. Award of the Medal of the Rojjal Society of IF. H.- -The award for 1946-41 was made to Professor E. de (V Clarke, M.A., for his work in connection witli the geology of Western Australia. The medal w ill be presented to him later this evening by His Fxcelkaicy the Lieutenant-(h)V(‘rnor, vSir Jamt's Mitchell. Prograwnies for General Meetings.- The ('ouncil has endeav oured, during the past year, to present at tlie monihly meetings programmes whicli contain some variety of interest for the different sciences. In addition t(3 the record number of original research papers. {‘ommuni('at('d talks of general interest to members have been arranged for nearly' (U’cry meeting. A. D. ROSS, President. REX T. PRIDER. C. H. E. JENKINS, Joint Hon. Secretaries. VI. Jour. Roy. Soc., Western Australia, Vol. XXVII. ",' 9 i "d r- ^ ^ 0 —I CO 3 ; ts- C'l U'^ pH 055 IC 0 fH CO •i* 0 0 uO 01 0 55 CO pH (M uo «rt -4 X 32 50 W >o o 50 5^ ; 'O o 55 O rH iM a . c- :CC .25 r o ^ =! c; ^ ^v. 7i X {« ti ; O ti •/) 42 C ■ Ci s o : £ tr^ ■ I; S t- t; . ^ "zJ £2 . t- 42 ^ O -S^ X 02:? •-c’'-)5 ■7.2 c ffl 2 .S" 2 2-1 .«*■ — H .SH .2 j=: X . 5 ! 3 - 2 »J ^ X '.O o ?ti 3. un J; o o S 3"^ 45 02 O 42 Ci< s :?o I <1 X CO ^• — a .M 32 s ^ 3 H "2 d a o j2 5?; ^ O *< H 3 ; Jour. Roy. Soc., Western Australia, Vol. XXVII. vii. ABSTRACT OF PROCEEDINGS, 1940-1941. 9th July, 1940 — Annual General Meeting in Gledden Hall. Presidential Address : “ The Ceramic Resources of SoutliAVestern Australia” by Mr. H. Rowley. ISth August, 1940 — Papers — “ Further Permian Corals from Western Australia.” Dr. Dorothy Hill, com- municated by Dr. C. Teichert. “ The Results of a Microscopical Examination of Some Soil Colloids,” Mr. W. F. Cole. “ The Granite-('arduj) Series Contact at Armadale.” Dr. R. T. Prider. 10th September, 1940 — Paper — “ Some Kyanite-bearing Rocks from the Eastern Goldfields, Western Australia.” Mr. K. R. iVIiles. Talks — " His life work in Ornithology.” Mr. Gregory Mathews. “ Criticism of Motor Head-lights Regulations illustrated with Models and Demonstra- tions.” Professor A. D. Ross. 8th October, 1940 — Papers — “ Permian Productinae and Strophalosiinae of Western Australia.” Dr. K. Prendergast, communicated by Professor E. de C. Clarke. “ New Forms of Western Australian Birds.” Mr. Gregory Mathews, communicated by Mr. Glauert. Talks — “ Chemistry of Strategic Materials.” Professor N. S. Bayliss. “The Habits of the Mallee Fowl.” Mr. F. Lewis. 12th November, 1940 — Papers — “ A Note on Fossil Corals from Langley Park Bore, Perth.” Dr. J. W. Wells, communicated by Dr, Dorothy Carroll. “ Some Basalts from the North Kimberley.” Dr. A. B. Edwards, communicated by Professor E. dc C. Clarke. Talk — “ Some Phases of the Work of the Commonwealth X-ray and Radium Labora- tory.” Dr. C. K. Eddy. 10th December, 1940 — Paper — “ Foraminifera and a Tubicolous Worm from the Permian of the North-Western Division of Western Australia.” Mr. W. J. Parr, communicated by Dr. R. T. Prider. Talks — “Night Photography.” Mr. B. L. Southern. “The Work of the Perth Observatory.” Mr. H. S. Spigl. 11th March, 1941 — Papers — “ Notes on the Vegetation of the North-Eastern Goldftelds.” Miss N. Bur- bidge. “ Contributiones Florae Australiae Occidentalis, No. XI.” Mr. C. A. Gardner. “ Seeds of Strychnos lucida and their Alkaloid Content.” Dr. E. M. Watson. Talk — “The History and Development of King’s Park.” Mr. J. E. Watson. 8th April, 1941 — Papers — “ Beraunite from Dandaragan.” Mr. C. R. Le Mesurier, “Marine Copepods from Western Australia, Part I.” Dr. A. G. Nicholls. “ New Leaf-Hoppers (Homoptera Jassoidea) from Western Australia.” Dr. J. W. Evans, communicated by IMr. C. F. H. Jenkins. Talk — “ Development of Western Australian Fisheries.” Dr. A. G. NiehoUs. Exhibits — Mr. L. Glauert exhibited three rare specimens of Cirripedes or Barnacles, Lepas fascicularis Ellis and Solander, from Albany together with Poecilasma fissa Darwin and Alepas tuhulosa. (Quoy and Gaimard) found on crayfish Panulirus longipes A.M. Ed. caught at Houtinan’s Abrolhos. All are additions to the Western Australian list. VI 11. JoTTR. Hoy. Soc., Western Australia, V^ol. XXVIT. 13tii May, 1041 — Paper — Mineralogy of the Donnyitrook Sandstone.” Dr. Dorothy Carroll. palk — Artificial Hadio-activity and some Applications.” Mr. J. L. Farrant. 10th dUNE, 1041 — “ The Geology and F'hysiography of the Gosnells Area.” Mr. C'. E. S. Davis. The Geology and I’hysiography of the Wongtingd'ardnp Area.” Mr. B. P. Thompson, commnnicated by Bf- E. 'I'- Pridor. “ X-niy Analysis (by the •|)o\vder methtjd) anri Microscopic Examination of the Weather- ing Proiiucta of the Gingin Djjpcr Greensand.” Mr. "W. F. Cole. Ocean Birds of Perth Beaches.” Mr. L. Glaiiert. K.ihihits— Mr. Glanert exhibited fragments of hioceramm slu‘11 from a depth of 550 feet l)elo\v the surface Woomerangie No. 1 Bore Tomala Station, Shark Buy, sug- gesting the ]>resence of Gingin chalk. Dr. Prider exhibited specimens of Sericite Schicit closely resembling Cardup Slate, collected at Black Boy Hill by Mr. J. E. NATIONAL MUSEUM OF VICTORIA JOURNAL OF THE ROYAL SOCIETY OF WESTERN AUSTRALIA. Volume XXVII. 1.— RESULTS OF A MICROSCOPIC STUDY OF SOME SOIL COLLOIDS. JTy W. F. Cole. Read 13th August, 1940 ; Published 14th November, 1941. INTRODUCTION. In tv o previous pu})lieatious (1) and (2) -I. Shearer and the author have doserib(Ml the X-ray analysis of some soil colloids from Western Australia and Tasmania. In this paper an account is given of a microscopic examina- tion of these soil colloids. In the lirst jnihlication cit(xl, the crystalline eonti'nt of colloid separations (effective diameter 2^ and less) from two soil samples from the Salmon Gums district in Western Australia was dotormiiu^d. These two samples, designattnl as A 1340 and A 1351, represent the subsoil and dtH'p subsoil respectively. In the S(>cond publication cited, the crystalline content of colloid separa- tions (etfective diameter l-4p, and less) from four soil samples from Tasmania was determined. The details rega.rntmorillonite group and the mica grou]) respectively. This is lu^cessary because at present it is not possible to idtsntify minerals within a grou]> in the dilfVaction pattern of a soil colloid. w. y. coi.K. 'L'aulk I. . {'OMPOSITIOX OK THK SOIL COLLOIDS. (lU'sult.s of X-rny analysis.) Tasuuiniaii. West Australian. Mineral. Tml. 449. A 1351 and A 1340. (huu'tz .Mucii < ^ -Much <- Much Very J^ittle i'ossibJo trace Ivaoliiiite. Little Much Miieh Much little Mica ... — — — Much .Montmorillonite Little, Possible trace Possible trace — — liai'matite — Very Little \'ery Little Little — Xole : An arrow (— >) indicates increasing amounts of quartz. A similar assessment witli regard to kaolinite in ool, odO and 499 could not b(‘ made. Th(‘ soil colloid fi'H('ti()ns from VVi^sttu’u Australia wtnv separated accord- ing to lli(‘ Interiiatioiuil M(‘thod which giv('s particles' of an effective- dianieO'r of 2jLt and less, d'he soil colloid fractions from 4''asniania wei‘(* se])arated accord- ing to the fornun* British systtnn whicdi gives particles of an (*ffecti\‘e diameO'r of 1 ■ 4 |Lx and less. From the point of view of this study the main differences iTi thes(^ two methods of preparation are ( I ) the 'Tasmanian soil colloid fractions ar(^ of a liruu* grad(' than the soil colloid fractions from Western Australia, aiul (2) tlte fornuM' wen^ ov<'n-drieii wlannais the latter were air-dried. As su]>pUetl for examination the local matcn’iai was still attaclied to tin.' filter paper and luid to be removed by scraping witli a glass sli when viewed under tlu* microsciqu' were sei'ii to ])e in the form of [)articles up to 50^ aiuL gri'ater in diameter. Since (it was claimi'd that) tlu' clay si'parations were (‘omposed of pai-liclcs of 2p, and less in effectiv'c diameter, it was assumed that tlu' par- ticles viewed beneath the microscope wim-c aggregates of tlie sod colloiil par- ticles. In that they exhibited definite iqitical propi'rlies, all these aggregate's, in both samj)l('s, elosefv' resembl(>d fragments of n singli' rniuei'al. ddu' optieal propt'i’tii'S of these, aggregates, it is eoasideri'd, ure due to the individual com- ponents exhibiting uniformity of crystallogruphie orientation. 'Kliis orienta- tion of soil colloid and clay partieli's into aggri'gates after iLrying was originally obstu'ved by Hendricks and Fi'v (3) and lias betni fnrtlu'r iuv'estigated liy Bray, (h-im, and Kerr (4). The aggregates appeai’iug in the soil colloitl fractions of A 1351 and A 1340 are larger than those noted l)y Jh*ay, ({rim, and Kerr. Wk, SI LTS (jv A Mi('j{()s(.’()[’ir Sti'dv of Soml Soil Colloids. 3 In sain})loi? A 1351 mitl A 134-6 the ]jerfcc‘tion of the unifomiity of orienta- tion of the individual eoinponents irdo aggregates is amazing. It was not }K)ssible to identify infjre than one t>'pe of aggrt^gate. Tins indicated that such aggregates were characteristic of the Manptle as a wliole. 'The refractive'- indict's W(U‘e doternuiu'd on aggr(>gates by hnnu'rsion in mixtures of clove oil and r(- -monoc]ilort)tuiphl lial^au'. Tlie main difftmaices between tlie two colloid se]iai'ations A I34(i ami xV 1351 is on(' of colour. Und('r the tnicros(*oj)e A 1346 is light brown ; A 1351 is ('hocolate brown. All the mati'rial of both samples is doubly refracting ; A 1351 bemg the more l:>iriTringent. 'Fhe extinction of the aggregates is in all cases sharp. The indices in difteront din'ctions \\'('re easily d(‘tt‘rminal)le. 44ie optical interference figure of aggrtgates of both samples is biaxial negativi^ and the axial angle is small. Tliese tigiires are similar in character to thos(' mentioned in the original woi-k of Hendricks and Fiy (3). The o])tical properti(‘s of aggregates of the' soil colloid fractions A 1351 and A 1346 ar(> suinmarisod in TabU' 2. Tablk 2. OPTICAL DATA KELATIXG 'I’O THE SOU. CObbOll) ERACTIOXS FROM a\t:stehx acstralia. ()j)tical Properties. A 1346 (Water Dispersed). A 1351 (Water Dispersed). (( 1 ■ 562 1 • 555 y • • « • . . I -572 L570 6- 010 (0015 ’IV Small Small Optical character ( - ) ( - ) It may bt' seen from Table' 3 that these optieal properties are not those? of any cominon clay mineral. X-ray ewidemcc indic,ates that both the soil e?olloid samplers are composed of a mixture of two clay Toinorals, one be'longing to tlm micH group and the' otlier te) the' kaolinitc group. If the aggregate's under e?xaminat ion eu'e’ to bi' e-euiside'red roprese'atati\'o of the sam})li‘ as a whole, then these optical pi'ope'vties e'oulel re'sult from an aggregatieni ejf tlu' above two clay mineu'als. That twej clay mineu'als bHonging to different groups may closely intergrown iide) aggregate's which possess uniformity of crystallographie* orientation has alnmdy bc>en noted by Bray, Grim, and Kerr (4). Hence tlie optical ])ropcrtie's of aggregates of the soil colloid frac- tions A 1351 and A 1346 arc ce)nsistent with the? X-ray conclusions. Table 3. DPTTCAL PROPERTIES OF CLAY MINERALS. Kaolixite Grout. 51ineraJ. 7 (( 7 - « 2V. Sign. Kaoliuite Nacrite Diekite Halloysite Hydrous Halloysite 1-570— 1-560 1 • 566— 1-563 1-556 Mean Index Mean Index 1 -5(33— 1-553 1 -560— 1 -557 1 ■ 560 i ■561—1-549 1 -542—1 -526 0-006 0-006 0-006 Very slight \Yry slight 40^—90° 52°— 80° (--) ( + ) or (— ) ( + ) 4 W. V. COLK. Montmorjij.onjtk Okoijp. M iiK'ral. y (( 1 y - 2V. Higu. Meait morilloiiite' . . . 1 ■5;!(i - 1 - Mm 1 1 -505 1 -47(1 (>■035 - t)*()35 r- 25° ( — ) P>.'i(l('llit(‘ 1 ■5()(l 1 - .-.30- 1 -5(10 (1-935 — 9-935 Small {— ) Xont remit 1 -(>10 l-5(i(» 1-575 1 -530 9-935- 9-()39 Small {-) Sa|)e)nilc ... 1 lulice's e>f re'fractioii similiir ti) .Monttiioriliemite* j Mjca (jRorr. Miru'ral. r i( y - a 2V. Sign. niite' 1 .(ior,— ! .51)5 1 •579—1 -535 9-035— 0-939 Small {-) Xole : This Tahh* is based on that ])ul>!is]ied ))V (dim (5). SOIL ('OLI.OID KKACTIONS KKOM TASMANIA. Ill ^(‘iK'ral Mu' nMimiTs jiuulo in tlu^ pi'nvious section as r('pir(ls ag^rc'ga- tion of soil colloid mutt'i'ial aftta* drying apply also to tlio Tasnuuiian soil colloid fractions. W’luai \ic\\cd nndci' the inii'roscopc* all four sample's woi'o sc.<>n to !)(' in IIk' form of [lartTde^s up to and greater in diam<;t('r. In view of what ha-s hei'ii previously stated, tJies(> })art'ich‘s are eonsidt'iHMl to bo aggi*('gat('s of the soil colloid particles which Imvt^ a.n td’tV'ctive diamedcr (in tliis ('as(') td 1-4/^ and h'ss. Sample 222 is, howevea*, the only fraction whi(*h shows any deHlucihle uniformity of crystallographic orientation of individual compoiu'uts. d'he three samples, 551, Still, 4d9, and tlu^ grt'ater [lart of 222, ai'(^ dark helwe'eai (U'osst'd nicols. In siudi sample's a nuam ind(>x of I'efraction onl> A\as di'termined. N'ariations from I hi* nu'an \'alue are small. The soil colloid sample's 551, Slit), and 4tU) are' \ i'ry similar. Aggj-('gat('s (oi'ined from (>ach of llu'st' thre'i' sainph's art' cliaracti'rise'd by theii- t'olour, high indict's of refraction, a-nd a,ppar('ntly isotrtipic character. Tlu' colour of 4i)t) uiuh'r tile microscojH' is chocolate' brown, that of 551 and Still dark brown. 'The' high indice's of re-fraction are' probably due' to tlie high iron conte'iit ol the' seiil e'olloid sample's. This iron e'onte'iit elot>8 not ('liter th(> struedme eif the' clay inine'rais pre'se'ut in the* soil colloids (a cotu'lusion arrived at freiin the* X-ray analvsis) but jirobably coats tlu* colloid jiarticles in the. form t)f ii’on o\id(\s (.ir liyelroxidos. Such a pejssiliility has already bet'ii noted by Nagelselnnult (ti). These indice's eil refraction, tlua'C'fore*, eh'ai’ly do not r(>pivHent the* refracti\'(' inde^x of the clay mineral shown io be* pn^sent by X-ray examination. Ibaulrieks and Kry (M) have' note'd that a high iron contr*nt in a soil colloid frae*tion te'uds to mask the doubly re'fracting mat(''i'ial pre'sent. In N'ie'w ol tii(> X-ray i‘e.»nclusioii that the' elay miiu'ral ])rese’nt in those' tliree' eolloid fVaetieaxs is preHleaninantly kaeilinit.e', it is belit'\-eMl that, in the ahsenee of a liigli iron eonle*nl, aggre'gati's fornu'd from eaeli of these thre'C colloid fractions would still posse>ss only a \ery weak hin'fringonce. Hemce, generally speaking, the optica! data relating to aggregate's of the soil colloid fractions 551, 560, and 499 (which are summarised in Table 3) are consistent with the X-ray conclusions. Kkstlts or A Mk'koscopio Studv or So.mk Soil ('oi.loids. vSaiopk' 222 is (lifiVront froTii the oIIk'v 'I'asitianian sain]iU>s in tliat th(' colour Ixaioatii tlie iniorosropo is ^royish-x cllow, tho itidicc^s of i'(>fi*ac'tinn art' lowof, and tlio ajjcu;rt'^atos a>'i> nu)l't^ hirofringtait. Many a^^iu'^att's art* dai'k Ix'twci'ii arossod tiicols but soino. slunv fairly unifonn orystallograpliir oricntalion oi indix idual aoinpononts and for tliosf^ y n ^ ()*005. Such aggr('jiatos iailod to show int(M'f(M-('n('o liguros suiliaiiaitly satisfaaior\ Ibr tho (>(►, and 491), containing kaolinit(‘ as t hi' only clay initu'ral art' nou-bireli'ingent , the slight ly biri'fringent character of 222 ('fable 4) is consist I'nt with the al)o\(' X -I'ay i-onchision. 'fliat is, if the aggregates under ('xainination are to be I'onsidered re|)resentati\ e of till! sanipk' as a wholes tlu'u thesis two minerals (kaolinite and montmorillonite) are occasionally closely iiitergi-own into aggri^gatos which possi'ss unit'ormit\' of crystallographic oruaitatiou. Tarlk 4. OPTK'AL DATA liK.LATTXtJ TO TIIK SOIL C(.)Lb01l) FKACTTOXS FROM TASMANIA. Sample. (( Mean Refractive iiide.v. y - c 9a»> 1 • 507 1 • 572 0-0( 15 AM 1 • 00.7 -MiO ! -OLS 499 1 ■ 05,7 RK-DISPF.RSION AND RF-ACKiRFOATlON OK THK SOIL COLLOIDS. In view of criticism of the o]‘)S(‘rvations made in this study that fii'stly the particles examined wi'n* aggregati's (of a size ii]) to and greatei- in diameder) of soil colloid partii'les rathi'r than original single crystals wiiiedi had not been broken down in the |)re])arat ion, and that sec'ondly two clay minerals bi'longing to diftei'iait groups nia\' l)e closidy intt'rgre)wn into aggre- gates whicli possi’ss uniformity of crystallographic oi-icnt ation, it ^^’as of inte'rest to restori' tlu' particles to their original sus]>ended stati' where they liad an effective diana-ter of 2^ aiul to re|-)roduce tla^ aggri'gafes by remox'ing the ])artiid<;s from suspension. 'This applii'd ])articularly to tlui soil colloid fractions from Western Australia \\-hicli exhibit well marked crystallinity. Hv-disijcr-sion.^-'Vhv re-dispiM’sion of one of these samj)les (A Dial) was undertaken ])y Mr. I>ur\ ill, of llie State Department of Agriculture, who was n^sponsiblo for the original si'paration of the West Australian soil i-olloids. By repeating ]a'ecisely the procedure adopted in the prepai-at ion, Mi‘. Uur\ iil was able to re-disperse the greater t>art of the soil colloid sajupk' A Biol. 'Vlu' individual partick's of this susiiended mati'rial wi're not \'isihle under liigh power magnitieation. 'riial sonu' matei'ial was not i-e-dispersi'd is ol little eon- set[uence when it is realised that, as a result of the grinding a sample- i'eeei\es in a mi'elianieal analysis, the amount of clay recorded is rather arbitrary. Re-iiggrcgatioa . — By removing the clay particles from suspension in the same maiau'r as oiiginally carried out by Mi*. Ihir\ill, the author was able to reproduce the state of large-sized crystal aggregates in wtiich tlie sample originally existed. These aggregates again exhibited a well marked uni- formity of crystallographic orientation. 6 \V. F. Cole. I’luloubtodiy tlu'ii tiio iargo-sized particles iisod in the o])tical oxaniina- tioii wort* not original unbroken partick'^, but on the contrary, were aggregates of soil colloid particles (th(^ ]iarticleK being of 2jLt an [)arti('k's stdl led in siudi a maniuu* tiuit all the material colloetiMl remained dark i)c‘tv\'e(Mi crossinl nicols no mattcu' wh(dh(M‘ <>vaporation was slow or fast, whetlua* tlocculatcxl or unflocculati'd material was u.sed. Secondly, the detinition of tlu‘ interference figure improved as th(‘ material lost water. 'Llu^ intiatcrc-nci^ figure biM'.auK* only iletlnitely recognizable when the sample first showinl tlu' production of contraction cracks. This is in (3onfoimity with the ol)S('r\ at ions of other worki'rs (iirindously noted) that indix idual partick's (Existing in a soil colloid sus])(‘nsion t(>nd to orient thon.\s(d\'es aftei' drying into aggri?gat(‘s whiidi possess uniform optical pi'operties. That thi' whole surface was s(*t in an optically uniform manni'r was seen from tlu' fact that the inter- feriavce figun' produciKl was th(^ sami' fo?’ all the material collected. This is com})let(dy in accord with (Irim's obs('r\-at ions (7) on the pro cla_\' miiu'i'als bf'kmging to diPiV'rt'nt groups may bt' clostdy intergi'ow'u into aggregates which possess unifoi'inity of crystallographic orientation. SUMMARY. An account is given of the results of a niicrosco]'ic examination of soil colloid materials from M estern Australia and Tasmania. These results are consistent with the luineralogical composition as deduced from X-ray analysis. The methods employed and the obsoiw'ations made are in conformity with those of other workers in this field. At'KXOWLFDrLMENTS. These investigations were carried out in the Physics Department of the University of Western Australia during the tenure of a Hackett Researcli Studentship and a Uommonwca-lth I'csearch grant. The autlior wishes to express his thanks to the 1 nive-rsity of Western Australia for the former and to th(' Council for Siuontific and rndustriid Reseavi’ch for the latter. The author s thanks are also due to Dr. R. T. Pricier, of the Department of Geology, and to Mr. J. Shearer, of the Department of Physics, for helpful criticism, and to Mr. Burvill, of flie State Department of Agriculture, for the redispersion of sample A 1351. RE.sriiTS 01’ A Microscopic Study of Some Soil C'olt.oids. BlBLiOCJRAPHV. (1) Shearer, J., and (hie, W. F.— “ Analysis of Soil Colloids by X-ray Diffrac- tion Methods.” Joiu\ Roy, Roc. W. Aust., 1940, XXVI., p. 121. (2) Shearer, J., and Cole, \V. F.- — X-ray Analysis of some Fasinanian Soil C'Olloids.” Jour. Roy. Roc. IF. Aust., 1940, XXVL., p. 133. (3) Hendricks, S. B., and Fry, W. H.— “ The Results of X-ray and Micro- scopic examination of Soil Colloids.” Roil Scl., 1930, XXIX., pp. 457-479. (4) Bray, R. H., Chhn, R. F., and Kerr, P. Ap])li(*ation of C3ay Mineral TechnifpK' to Illinois Clay and Shale.” Bull. Gcol. Roc. A))(cr., 1935, XLVl., PI). 1909-1920. (5) (jrim, R. E.- — “ Properties of Clay.” Recent Marhu' Sediments, A Sym- posium, London. (0) Xagelschmidt, G.— “ The Identiiication of Minerals in Soil Colloids.” Jour. Agr. Rci., 1939, XXIX. [4J, pp. 477 501. (7) Grim, R. E. — The Petrographic Study of Clay Minerals. A Laboratory Note.” Jour. Red. Retr., 1934, IV. [1], pp. 45-4(). note added IX IdPlOF-ltEADIXO. Since this ])a])cr was first ]n'e])are(] the antlior has Ixhui successful in obtaining chai'actcrisi ic relleetioiis fioin oriented aggregates of soil C()1!{)1<1 >ianiples. A lilni of an oriented aggi’cgate of 'U^inple A Dol shows strong basal reflections at 7 A and 10 A indicating the in'cscnu'c of lioth kaolinitc and mica in the same aggregate and thus fully coulirming tlic conclusions r(uiclnal above on llie aggregation of soil (U)!loids. Some Kyaxite-bearikg Eock^ prom the Eastern Goldfieeds, Western Australia. 9 2.— SOME KYANITE-BEARING ROCKS FROM THE EASTERN GOLDFIELDS, WESTERN AUSTRALIAN By Keith K. Mtles^ B.Se. (Hons.). Uend KUh August, 1940: Pulilished I’dth Xovuud)er, 1941. COXTEXTS. Tnti'oduetion The Camel Humps, Mt. Margaret Goldfield Field oceurrenee Petrography . Mt. Leonora, Alt. Alargaret Goldfield Field 0(‘eiivrence Petrography Alt. Kenneth, Valgoo Goldfield Fiehl oeeurrenee Petrography LJuinu’s, ^riireliison Goldfield Field oeeiirreiiee Petrography Origin and ATetamorphism The Camel Humps and Alt. Leonora Alt. Kenneth and Quinn’s Roeks Summary Ackuowletlgments Referenees Roc-ks Page 9 10 10 12 14 14 Id 18 18 18 19 19 19 20 23 24 24 IXTRODCCTIOX. AA^'hile engaged in a ))rograinnie. of Held Mork iiiuh'rtaken by the Geo- logical Survey in the country north of Laverton, Alt. Margaret Go]dfi(4d, during the winter of 1939, the \\'riter noted and mapped at a spot known ns the Camel Humps, an interesting occurrence of a variety of rock Intherlo unrecorded from this district. This is a kyaniU'-bearing (juailz schist, a fairly high grade metamorphic type, and part of the Older Greenstone Series (Pre-Cambrian) which forms a narrow belt of anrifiTous c(umtrY in this portion of the State. Comparison of this i-ock witli rallu'r similar looking imdamorpincs from a numb(‘r of other localilies in the Eastern Goldfields has led to the discovery or contirmation by the writer, of the existence of kyanite in several of thes(‘ latter rocks also, and the object of this iiaper is to give* a brief desci i])tion of the occurrence and petrology of these interesting rocks. The presence of the mineral kyanite in ci'ystalliiK' scliists is of ju'cnliar interest, because of the light that it may shed upon flu* origin, constitution and metanuH'phic history of the rock in Avhich it is developed. It is a silicate of aluminium, AUO.SiO.,, and is generally recognised as one of the few tyidcal ‘^stress minerals’’ (1), and appears to form only at definite stages in progressive regional metamorphism. It is a mineral which is peculiarly sensitive to changes in temperature and i)ressure conditions, and requires * Piibliyhetl by permission of the Government Geologist of W'estern Australia. 10 Keith K. Miles for its (levelopiiiout, ap])ropriate conditions of moderately hi^h t(miperature in conjunction witli essential shearing- stinsss. It appears to liave only a temporary status or field of stability in regional inetanioi-phisni, and together with its ])olymorphs, andalnsite and sillinianit(‘, has been for a long time i-egarded as a *1), (-). Kyanib* has been recorded IVom comparatively few localities in Western Australia, these being mostly confined to tlu' Chiltering Valley^ (5), (()), (^)t and (11), and the Clnttering-Jiini>erding-Clackline-Y()ilv belt of metamorphic rocks generally (8). Specimens ol kyanite haV(* been obtainese strueTure. The* knots consist of kyanite in either rounded vathei* stumpy crystals ave'i'agiug about 2 mms. in eliameter and 2.5 mms. long, v)r longe'r Hat blaeleel crystals u]> te) (1.5 mms. in le*ngtb. Me)st e)f these crystals are* reuighly oriente'd so thal the*ir shorte'v axes lie* appre)ximate!y tunaual to the plant* e)i' schistosity. They are e'olourless and have* a lustre- ranging from dull vitreems to distine*tly pe'arly ui)e)n cleavagt* facers. The*se knots e)f kyanite are* st't in a "flowing’’ e)r folialt'd ai)hanitie' mat- rix of whitish, very lincly gi*annlate'd (piartz, usually tlt'cke'd with liny (lakes e)f se*ricite gi\ing a silky sh('e*n to surl"ae'e*s of the* ve)ck wiucli have bee*n bre)ke‘n parallel te) the* schistosity. This tpiartzose matrix is ln'i’e and the'i'e* staineel yelle>wish or pink by a little* intre)dne'eel limonito. If has a porpliyi'oblastic gneissic structure* — the pe)rphyroblasts of kyanite being arrange'el in laulely oidenteel bands of very irregular shaped evystals whose be)unelaries are fre(|uently eremulate'el and embaye'd, whilst the* teuiniua- tie>ns are gemwally resolved into a granular diablastic aggregate e)f' kyanite* and (piartz. Individual crystals of kyanite are occasionally curved and bent due to rotation during growth. In thin slice they are colourless and uou-pleoehroic and have characteristic high relief. Many elongated sections show a wedl defined cleavage (100) while sections cut paralled to this disedose an iuqierfect (010) cleavage. The cross parting (OOl) at a])proxiinately 85° to Hu* length of flu* crystals, i.e., the c-axis, is seldom well d(*velo])ed (See Fig. vlA.). End sej.ctious (cut approximately normal to the c-axis) disclose* the i>erfect (100) cleavage whilst the (010) is re[)re*sented by a strong e-ross parting at from 80°-85® to (100) (See* Fig. MB.) In such ■sections extinction is [)arall(d to (100) cleavage. * All numbers in parentheses represent either field or registered number.^ of specimens in the Geological Survey Kock Collection — except those preceded by the letter U, which helong to the General Collection of ihe Dept, of Geology. University of W.A, So:^IE KvAXJTK-IJKAKIXG KOCKW FRO-M tbe Kasterk Goldfiflds Westerx Australia. 13 ^ B 1 MM I Fig. 3. A. j\ y;iuit<^-(|u;irtz .schist (suvciiiieu L3117) from tlio (’anu‘l Hiiiniis. Showing- ('harnct(M-istio hlnded form of kyaiiite cut parallel to (100) (clear (u-ystai). Rutile inclusions in clustered aggregate (black). B. Tlio same, with kyanite crystal showing perfect (100) deuvage and imiDerfect (010) fi‘()SK-]>arting in section cut ap])roxiinatcly parallel to (001). Note diablastic iiitcrgrowtli of granular kyanite witli quartz matrix. The following optical }n'operti(‘S have been determined:— The optica] elongation is i)ositive. Z A c is 29°-3()° on sections cut l)arallel to (100). The mineral is biaxial with negative optical character. Optical axial plane is approximately normal lo (100) and inclijied at 29°-30° to e-axis. 2V is large ( 80'^). The refracti^’(‘ indices are: — a = 1-711 h - 1-718 ^ F 002 r ^ 1-728 J 7 — » = -015 ± -002 Twinning, oft(m a characteristic feature of this mineral, was only occasionally noticed on (100) as twin |>laue. Micrometilc observations in- dicate the presence of 15% to 25% of kyanite in this rock. Inclusions in these kyanite pon)iiyrol)la.sts (A)nsist of ((uartz grains, clustei's of y'ellow rutile prisms and occasonal detrital zircons. The quartz matrix enclosing the kyanite consists of a tiiu‘ grained, even- grannlar mosaic of crystals averaging about .04-. 08 mms. in diameter, in- cluded in and surrounding the broken grains of kyanite. (Quartz and kyanite constitute the essential minerals in this rock. 14 Keith R. Miles. Accessory minerals in*esent are : — Rutile in very abundant scattered clusters and separate individuals of tiny o'olden yellow j risinatic crystals frequently occurring in geniculated twins, often as poikilohlastic inclusions in kyanite, but also scattered irn'gularly Ihrougbout the iiuartzite matrix. Sericite or muscovite occurs in some tiny parallel oriented rods “flowing-’ tliroiighout th(' (piart/ite matrix and also as a few larger flakes and string(*rs occasionally moulderobably be found to compare \'ery closely with those of the Mt. Leonora type quoted below. MT. LEONORA, MT. MARCxARET GOLDFIELD. Field occurrence. Mt. Leonora lies about 3 miles to the south-southeast of the townsite oi Leonora and forms tlu' highest and most conspicuous summit of a long line of inl(H’mittent ridg(‘s and low hills which run northwards for many miles. 1 hose also extend southwaixls from the Trig. Station for about three miles in a line of I’ock outcrops which form ridges of g'ratlually decreasing stature ami which tinally disapix'ar beiuMth the alluvial flats of Lake Raeside. The rocks of which Mt. Leonora is composed are creamy yellow to light grey in (-olour and are usually dense, hard (piart/ose scliists of variable grain size, frequently having a knotted chai-acter. These rocks are distinctly Ixalded and dip fairly constantly at 00 ° to the west (S(v Fig. 4.). Tlnw were tlrst described’ in 1904 by C. F. Y. Jackson (15) who considered Mt. Leonora to be an area of “crushed granite,” despite the evidence of a chemical analysis of this rock (5084) published along with his descrij^tion. In 1909 A. (i. Maitland, as a result of a brief re-investigation of the field occurrence of the Mt. Leonora rock, pointed out its obviously bedded nature and from the Held evidence concluded it to be of sedimentary origin and not a metamorphosed granite (16). He also remarked that the quartz schists of the Mt. Leonora type extended noifhwards for over six miles Avith- out interruption to Mt. George. Some Kyaxite-iseartng Rocks from the Eastern Goldfields, Western Australia. 15 In the g-eological map aeeoinpaiiying his report on the geology of Leonora, Jackson (o|>. cit.) shows a broken line of ridges running southward from Mt. Leonora, a})}>arently forming the southern continuation of the strike of the Mt. Leonora rock, for over two miles. Similar ridges are also shown running northwards from the eastern outskirts of L(*onora townshii) to beyond Mt. George. The centre of all this line of ridges is inapp(nl as com- posed of a more or less continuous bed of ‘d)anded hematite-bearing ([uartz.^^ G.N.U'.A. A'r//. So. 19G. Fhoto. C. F. V'. Jackson. Fig. 4. — Outto-ops of audahisite (juavtz seliist, or (pun-tzite, Mt. Leonora, showing ))eddiug idjmes di|ij)ing at about 55® to the west. Outcrops of this rock on the eastern edge of the townsite are certainly those of (|uite typical bedded ferruginous t|uartzite or jaspilite, and conse- quently it would appear that in its liroad geological association with these jaspilite beds the Mt. Leonora rock shows a striking similarity to the Camel Humps occurrence, described above. in his bulletin dealing with the field geology of the Leouora-Duketon District, E. de C. C'larke (17), in an account of the Mt. Leonora rock quoted Farquharson’s determination of it as a ^‘finely foliated, ranch granulated andalusite-c|uaifz schist which he considers to b(‘ a metanior})hosed si'diment.'’ Chemical analy.ses of specimens collected by Jackson (5084) and by Clarke (1/2002) were compared witli two analyses of metainorphic rocks from Quinn^s, (Murchison Goldtield), and their general similarity noted. These analyses are quoted below. Petrography. A recent re-examination by the "writer, of a numlier of the old thin slices together with several new ones cut from dilferent specimens of the Mt. Leonora rock has disclosed the interesting fact that, in addition to andalnsite, some hand specimens also contained the higher grade nietamor])hic minerals, 16 Keith H. Miles. kyaniU* and .silliiniinite. hi fad different specimens of this rock provide 'excellent examplas of arrc'sted mineral development with the incetdion of a higlun' grade faci('s due to change in stability conditions during progressive metamorphi.sni. Hand specimens of this rock collected from various }>oints in the vicinity of the Trig. Statioji and from the western sloju's, are creamy white to gi*ey in colour and of varying dmisity and grain size. Coarser grained specimens are much moi'e foliated or schistose and have a strong knotted structui'e and a characteristic silky lustre on cleavage planes. The knots, presumably repiH'senting andahrsile crystals, are usually rounded and llattened and show ii}) ns slightly lighter coloni- than the foliated matrix. They may be up to 7.5 nims. long and nearly 5 mnis. wide. The finer ginined si>eciniens are much more eompact and though not markedly schistose, have a tendency to cleave along jtarallel ]>Ianes. The andalusile knots are only revealed on ]U)lished surfaces and then appt'ai' as innumerable small ronndisli, light coloured l>atches, averaging less than 1 mm. in diameter. The material surrounding these knots appears dense, tine grained and siliceous. In one hand sj>eeimen (5084), a medium-eoarse grained schistose type, one can distinguish, in addition to the andalusite knots, a few small llat^ pearly lustred laths of kyanite np to 2 mms. long. ^ B IMM Pig. 5. A. Kyaiute-beariug andalusitc-qiiartz schist (Spec. 5082) from near Trig, Station, Mt. Leonora. Showing part of a large granular ])orphyrohlast and some granulated kyanite (darker). The quartzite matrix contains flakes of muscovite and oriented stringers of dark carbonaceous material. B, Sillimaiiite-bearing audalusite-quartz schist (Spec. IT3165) fiaiin near west sio[)e of Mt. Leonora, Showing sillimanite in flowijig needles developing upon the margin of granular poikiloblastic andalusite. Black opaque grains are hematite altering to iiinonite, and some grapliite. Thin slices show a uniform porphyroblastic gneissie lexture. The ‘‘knots^’ or porphyrohlasts consist chiefly of colourless andalusite. This mineral has reached only a very early stage of crystalloblastic development, SoMP: IVYAKlTK-UEARlXCi R()('KS ERO.U TALE FjASTEliX nOEDFTELDS, Westerx Australia. IT tho '‘knots” l)oini>- inon^ or h'ss (‘(iiudinunisional hul olhorwisc witlumt ionn. Indeod so weak has boon tho for(M* ot‘ orystallisatie('iniens (e.g. 5084, 5082, 017205) ami usually 1‘orms rather irregular shap(‘d (‘longat(*d crystJils or occurs in g-ranular ag‘g'r(‘g'ates indifaiting an early stage of gi'OAvth. The kyanite is soiuetinies partially encIos(*d in granular aggn’t^gates of andalusiti' from which it is (Uw eloping, or it may occur as isolatcal porphyrohlasts in the (juarlzite matrix. (See Fig. 5A.) It is (*asil\ di'^l iuguisluMi from llu‘ andalusite by its high relief and conse(]nent darken' a{>pearance, and by its typii-al l)laded form, cleavage, extinction angle, stronger birefringence and posit i\'e elongation. Twinning on (100) Axas noticed in se'veral crystals. Though characteristically shoAving a. much stronger force of crystallisalion than li’^‘ andalusite, crystals are neven’ entirely fre(? of inclusions of (luartz granuh^s, curl)onaeeous material and some rutile m^edles, AvhiU' tlu‘ terminations of tin' crystal blades are usually i)roken into granula)' aggregates and 0 (-casionally 1 ‘esolved into needle-crystals of sillimanite. Silliiuanite is present in se\'(U'al specinums (I’dlOo, 10833) usually ass). It occurs ill slender elongated and cross-fraclunMl prismatic crystals or in felted masses of acicular fibres, gimeraily lying in parallel opti('al orientation with th(‘ andalnsittL Th(*se are distinguisht'd from ihi' andalusite by their form, higher relief, stronger birefringence and jmsitive elongation, and wher(3 associated wutli granular kyanite, from this mineral by their s1i*aighl (‘xtinc- tion. The matrix enclosing the [lorphyroblasts (-(insists mainly of a line granular lepidoblastic aggregate of quartz, charged Avith some finely dix ided carbonaceous material, t(^gother with s(*attered oriented shiu'ds and Oakes of s(*i‘icite, and fr(>quently abundant tiny vidlow grannl<'s and i>risms of rutile. Accessory minerals im-liule small (|uaii1ities of iron-ore — liotli lieniatite and Jimonite, and also occasional limonite t)seudonu)i'phs aft(*r pyrite. frag- ments of brown mica, and a fexv zicrons, all distributed sporadically through- out both the matrix and the jiorphyroblasts. Thus the Mt. Leonora rock may he .said to vary in mineralogi('al c(mi- Itosition from andalusite-()uartz s(-hist or (puirtzite, to kyanite, and sillimatiile- bearing aridahisite-quartz schist. Their similarity of chemical com])osition is clearly denumstrated, hoAA'ovcr, in the analyses quott'd beloxv. (A) and (!>). Their essentia] similarity to the Camel Humps ro('k is also xery n()ti(-(*al)l(', tho chief mineralogical dilferences arising from diffei-ences in tenqx'raturt' and pressure conditions during nietainorphism, as will b<‘ discussful in a later s^^ction. 18 Keith R. Miles. MT. KENNETH, YALGOO GOLDFIELD. Field ocviir-rence. Another interesting metamorphie which bears a general resemblance to the Camel Humps and Mt. Leonora rocks, comes from several low rocky rises at about s(n'en mih's easLsoutheast of Mt. Kenneth, near the eastern boundary of the Yalgoo Goldfield. This occurrence was described by H. W. B. Ta.Il)ot in IlHi) (14) and a chemical analysis, and l)i’ief petrographical des('i'iplion by H. A. Eanjuharson, were also published, 'flie field relations of tiiis rock are rather obscure, but it is appartmtly eiiciosed and presumably intruded by granite. Fetrography. The rock is pale greenish in colour, is medium grained and has a typical eveiL gTanulated schistose appearance, but has not the mai'ked knotted, or porphyroblastic structure of the two rocks described above, and hand speci- mens ai)i>ear to consist chiefly of Cjuartz and chlorite. The mici’osco})e, however, reveals the presence of numerous ragged prisms and ii‘i-(*gular, j-ather fragmentary crystals of typical kyanite showing usual high I’elief, good (100) cleavage, modeimte birefringence and positive ('lon- gation, with charactenstic maximum extimdion and negative acute bisectrix figure obtained from broad sections cut parallel to (100). These crystals are occasionally enclosed by and appanmtly de^■eloping from chlorite and sericite flakes, ot‘ are scatlered in groups in random orientation through an iri'Cgular cataclastic granidar matrix of quartz, and irregular’ I’agged plates of chlorite and muscovite. The most common accessor\ mineral in this rock occurs in small irregular, rather pointed prisms and rounded granules, scattered liberally throughout. Ihese are pleochroic in dark yellow-brown to red-})rown coloni’s, maximum absorption being pai’allel to the longest axis. Itelief is very high, surfaces of these crystals having a stippled appearance. Birefringence appears weak, and interferema* colours are mostly masked by absorption. Extinction is straight. A few gi'ains show ajij^arent development of penetration twins. Gi'ains were too small to give an interference ligure. They liav(‘ the colour, general form and extremely high refringence of ])ero\'skite, ((’aTiO;;) whilst the ])leochroism am! extinction suggest either brookite, (TiO,), or the meta- mori>hic mineral staurolite (hydrated silicate of Fe and AL). The entire absence of CaO in the chemical analysis of this rock (C) rathei’ dis])oscs of th(‘ possi])ility ot perovskite. The gc'iu'ral distribution is that of a detrital mineral and brookite is suggested. Other accessories inelmie clear yellow-brown rutile in euhedral })risms, pale yelloAV zix’cons often producing ]>leochroie haloes in chlorite, broken apatite needles, and a few grains of cloudy felspar. In 1920 E. de C. Clarke visited the Mt. Kenneth resence of any kyanite hoAvevei’, the indications being that these rocks have apjiarently failed to pass the chloi’ite stage of metamorphie doA^elopment (4, p 209). They consist essentially of (piartz, nuiseovite stained Avitli limonite, and some pale green chlorite. Of the accessories, rutile in bright yelloAv recrystallised prisms and dark detrital grains is most abundant. Others include detrital zircon and apatite. Some Kyaxite-eeakixo Kooks from the Easteex Goldfields, Western Austratra. 19 Chemical analyses of Talbot’s rock and of one of these (1/2695), given below, clearly indicate* their common composition, however, and strongly sug‘g‘(‘st their common origin. (^fCINN’S, ArUKCHTSON GOLDFIELD. F ieJd occurrence. From a little over half a mile to the south-west of (jluinn’s townsite in the Murchison Goldfield, comes another set of metamorphic schists, com- parable with those described above. These rocks were iirst described in 1904 by C. G. Gibson (18) who believ(*d them to be ga-anite schists. In 1921 they were further described and figured by F. K. Feldtmann (19) and a brief account was gi\-(m of tluii’ field occuri'eiice. Petrogra [)hic descriptions of a number of these rocks were provided by K. A. Far(|uharson (19, pl6) wfu» considei’ed sevei'al ol them (1/791-2) (1/794) (53.42-.'l) to lie metanioT’|)hic chloritic ipunlz schists containing a granular mineral which he doulitfully I'cferred to andalusite. Eddtmann concluded that they were contact meta- inori>hosed sediments. Analyses of two of the rocks (1 791 and l '7f)2) Avhidi are (juoted below, were also published. Petrograjdiff. Hand specimens of one of these metamorphics (1/791) are pale greamish grey coloured, rat!i<‘r uneven, medium to hue grained, granulated with an unusual streaky ai)pearanc(‘, consisting mainly of ]nnk and I'olouiless quartz, .and abundatit pale green (ddorite scales. This rock has a granulated ])or])hyroelastic t(‘xture, witli scattered “eyes” of slightly broken, cleai* (piartz in a tine grajiulated matrix of rounded (piartz ■gTains and chlorite* shreds. The gratiular mineral referred to andalusite })y Faniuharson occurs in several roughly rectangular patches, enclosing abund- ant grains of <|uarlz and chlorite and some rutile iLcedles. The determinable optical chai’acters of this mineral two those of andalusite present iti a \'(‘ry early stage of sk(‘leton growtli. In addition to the andalusite, tin* specinum also contains kyanite in a tew small ragged elongatetl prisms, usually enclosed by and apparently de- veloping at the expense of chlorite, but also occasionally associated with the andalusite. These crystals have modej'ately high relief, moderate birefring- ence, positive elongation, maximum extinction angle on (100) of about 30° A^Oj negative optical character and large axial angle., and undonbtedly re])resent an incipitait growth of kyanite. Other minerals include scales of muscovite associated with abundant altered chlorite, a few cloiuly grains of knolinisod felsjnxr, rutile in jiale yellow prisms, a few clusters of dark yellow irregulai* shaped granules with extremely high relief (probably brookite), zircon, apatite and a little granular carbonate. A second rock (1/792) from the same area has a similar tf'xture in hand specimens but is of a mottled pink colour. Small clear white crystals of andalusite ean be seen scattei’ed through a fine granular matrix of pink ('luaidz and pale green to colourless mica flakes. The thin slice reveals a ty])ica.l porphyro])lastic granular texture of andalusite in (piite well crystallised produefs crowded with fluid inclusions, grains of (piartz and a fow i-utih* lu'cdles, and usually surrounded by and 20 Keith R. Miles. appavoiitly alteriJi”- to a.^'ii’rog-ates of sealy ('olouvicss ('lilorito, Avith some musf'ovile, s(‘1 in a ground mass of clear broken and irre,i>-ular to rt)iinded (jiiartz grains, and interstitial fibrous chlorit(% and si'rieite. Kyanile occurs in om^ or two scattered <*rystals distiiii^’uislied from the aiidaliisite Ijy their characteristic Ijladed form and iii,u'lier relief. Other a(‘ces- sories include ridile, in tiny rods and o-eniculate twins ran^in”' from almost colourless to dark yellow and occurring- often in clusltu's or in stri?igs of crystals, and fragments of bleached, obviously dtdrital biotite, and a few .grains of zircon. ORIGIN AND MKTAMORPHISIM. FolloAvino- are the chemical analyses of tlie ]\It. Leonora, Mt. Kenmdh and th(> Quinn’s rocks : — ■ A. B. C. 1). E. F. SiO.> 7fi-71 82-74 72-57 75-35 78-28 84-11 AIXX. 20-08 14-84 1015 11 -35 9-48 8-05 Ee.,0.} 1-70 ■43 •49 •79 •40 Nil FcO ... • . • 2-70 ] -70 2-05 •62 MnO Nil Nil ■05 ■12 ■03 Nil MgO ■00 Nil 9-38 4-09 5-00 4-21 CaO •50 •09 Nil Trace •11 •18 Na.,0 •07 ■40 •34 •30 -23 -25 K..0 ■11 •20 -28 2-00 ■19 •18 H.,0 — •21 ■03 •02 ■10 ■00 ■00 H.,0 + ■24 •17 3- 19 3-10 3-19 2 • 15 Tib., ■00 1-33 •57 ■51 •04 •05 P.0, . . . Trace -20 ■07 -05 ■01 ZrO. . • • ... Present 4Vaee Trace CO. • « • . . • • • * ■03 -02 Nil YeS, ... ... ... •17 Nil •02 100-40 100 -29 99 ■ 94 100-40 100-33 99-89 Sp. Gr. 2-81 •? 9 2-72 2-70 2-07 Analyst : C. C. H. Bowlev. H. Bowley. D.G. Murray. H . Bowley. H . Rowley. AVilliams. A. S]K’(*. 508-1 Kyanlte-boaring andahiRite-(iua rtz schist. ATcinit\ ' of Trig.. Station, Mt. Lconoia, , Mt. Margaret G oldfield (15, pl9). 1». Spec. 1/2002 A mlaliisUe-<]Liavtz scliist. WA'st sloi)C of Mt. Leonora, Mt. Margaret (foldfiehl (17, p2.D. ('. Spec. 1/1070 Kvanite-hearing (|uarti^-cliIoi‘itc sdiist. 7 miles east-sfnitln east of Mt. Kenneth, Valgoo GoldtieUl (14). I). 8iiec. 1/200O Quartz-nmscovite schist. Near (Imiel Paddock (102M. on No. 1 Kabbit Proof Pence) vicinity of Mt. Kenneth, Valgoo Goldfield. E. E. S])ec. 1/791 Andalusite hearing quartz-chlorite schist with incipient kyanite. Ouni]) of No. 2 Stock Well, near Water Res. i:U35, Quinn’s, Mureliison Goldfield (19, pl8). S])ec. 1/792 Andalusitesiuartz-clilorite schist with incipient kyanite. Prin- cess Dagmar Water Shaft, GML 843N, Quinn’s, Mur- chison Goldfield (19, plS). An examination of the abo\e analyses clearly indicates the essential similarity of the rocks, and the variations closidy 'reflect the mineral ngical differenees. in general they all show a very high iiercentage of SiO.., and a notable excess of Al.O, over the 1 : 1 ratio necessary to satisfy tlie CaO and the alkalies present. From the alumina eontent the iiereentagc of andalu- site, kyanite, etc., jiresont in the Leonora rock is SO.SPc for A and 20.80% SoitE KvAX ITE-liEAHl X(i KOOKS EK0-\[ TELE EasTERX ( lUEDFiEEDS, W ESTEliX A CSTKALl A. -i for 1>, and in tlu' Quinn's rocks for K and ai)proxiinaIcly 5.(H)Oc ior F. In th<* Alt. Ivcnncth rocks this tiiiui'c lies ladwt'cn lOy; juid tor (\ while for rock 1) the' aniounl of alumina availai)l(‘ after satisfyino- the soda and potash is too small to pennit of the production of any aiidalusite. Tiie percentage of alkalies in most of the rot'ks is very low^ — in all but one specimen (D) beino- h‘ss tlian and of these the Nad) content is generally sliyhily in excess of the Kd)- The All. Kcmiieth ami Quinn's rocks all sliow an o\’e;'- wheliuinj^ (hnninance of AI^’() o\'er C'aO — and the total lime conhmt in all the rocks ('.\cept A is extreiiH^ly loAW Th(» hit’ll silica pmaamtaye, the predominance of alumina o^'('r linu' and the a]kali(*s, tlu> dominance of maj^'uesia over lime and the iow alkali content are criteria I'avoui'inu the sedimentai'v oriiiin ot a nietamorphit' rock. (20). The pi'csence of tai‘,u'(" i|uantities of andalusite or the allied kyanite or silli maiiiti* in i|Uart/ose crystallint* schists retlects tile existtaice ot an abundant alumina content and is usually a reliabh' indication ot‘ conla(d or thermal metamorphism ol' original sedimentary material in the case ol' andalusite, and of a hi^'her yradi* moi'c I'eyional type of metaniorphism wheri' kyanite or sillimaniti* predmninates. The (Utwei orphyrv occurring near the southei'n Camel Hum]). * k>ntact thermal (d’fects of this intrusion can ho assumed negligible. No large body of undoubted intrusiAT gi-anite or other igneous rock is knoAvn to (vvist in the immediate vicinity ol‘ IMt. la'ouora. The general north-south sehistosity of the country in the neighbourhood of both the Camel HiimjAs and Leonora Avas pi-obahly the r('sull of regional shearing pressures mostly contemi)oraneous with the formation of the nudamoridiic minerals. It i^- e\-ident that the I\lt. Leonora ro('k eommenc(’d reeratiir(‘, it luH-aine sidijectful to in- ci'easing sheaiing stu'sses — i>robably as'^ociated with folding movements re- si>onsibl(‘ for the sehistosity' oi' adjaetnit grccmstoiic. 'fhis ])r('ssar(* iinally reached siudi a stage* tlnd further growlh of the so (-allerl “anti-stress’^ mineraL andaiusite was completely' inhibited and it commenced recrystallisation into the* more sta})h* kyanite* and sillimaiute. Ace'ording to Harke*!’ (4, p. 'id'J-b) this association of andaiusite Avitli kyantte is dee*id(*dly' rare. He (piotes an occurrenn-e* in the* Fliiola district, near Davos (Switzerland), Imt here it is be*lien'eMl that tin* formation of the andaiusite' helonm'd to a late*]' { base ol‘ me't:imor]diism afteu' a ra]nd falling'- otf ol' shearing sti'ess while tein])e'ratures still I'emained high, l-lxamph's of the growtii of kyanite' in psenelonior[)hs afte'r chiastolite* as a re'sidt of legional metamoi'piiisiu su|ierim])os(‘el e)n normal theumial medamorphism of an are-a of original politic sediments are recorded by C. K. Tilley from Koss-shire-, Scot- land (1). The* pi'e'smu'e* of such a high graele inedameirphie' minei'al as kyanite in. rocks helie'ved to form ]mrt e)f the Older Gi’eamstone Se'vies is of partiemlar interest in set fai* as it shov\s that in some portions of the Ht. Mai'gai'(*t Cold- held this Serie's lias suffered a higher grade of melarnorpliisiii than had in-eviously licen recognised ihe're*. In these leAcalities tlu' rocks ha\'i* c\’idcntly' been suhjecte'd to regional ]n’essnre‘S and temperatures comiiaiable with tho^e* Avhicli must have existed in some of the moi'e advanced metamorphic areas of the State, e.g.. the Cluttering Valley ami tlu* Clackline- York districts. Igneous greenstones, helievial to ln*ne Ikm*!! chiclly hi\as (■ontemporaueous with the sediments, iirobahiy form the Inilk of the Older Oreenstone Series in most parts of the Mt. Margaret Ooldhehl. In some places well preserved striu'tural foatnn*s indicate that thest* igneous giaH*nst)UU‘s ha\'(‘ suffered \ery' little metamorphism of any kind but in the localities described above the ])reseneo of kyanite in associated mota-sedimentary beds sIioav that here at least they havt* been subject^al to high stri'sses and t(*mp(‘ratures. IIoAveveiv their composition is such that they shoAV no striking e\id(*nce of th(*se high Some Kyanite-bearing Rocks from the Eastern Goldfields, W estern Australia. 23 grade nietamorphie eoiiditions which have produced in them some internal recrystallisation often with development of schistosity, but without any marked mineralogical reconstitution. Finally the signihcant association of both the Camel Humps and the Mt. Leonora rocks with horizons of banded ferruginous (piartzite should be once more noted. They may reiiresent merely more argillaceous and non- ferruginou.s zones in such handed quartzites, or possibly portions of beds in which the iron content was largely leached away before metamorphism. It is interesting' to note; that such minerals as andalusite and sillimanite have been recorded a.s contact alteration products of banded ironstones near Bulawayo and Salisbury in S{)uthern Rhodesia. (21). Mt. Kenneth and QuinKs Rocks. The Mt. Kenneth and Quinn^s metamorphi('s bear a striking similarity to each other in their chemical composition. The principal mineral differences are in the aluminium silicates, the Quinn’s rocks containing i)redominant andalusite, whilst of the Ml. Kenneth types, one specimen (C) carried abundant kyanite while the other (D) has suffered but weak dynamic meta- morphism with r-ecrystallisation of white mica under inapjn'eciable tem- peratures, and no alumino-silicate has been formed. These rocks differ from the Alt. Leonora and Camel Humps types mainly in their lower alumina content and greater pi’oportions of ferrous iron, magnesia, and combined water. These ditferences are i-efleeted in the much lower ])ercentag’e of the aluminium silicate minerals and the relHtiv(‘ abundance of such minerals as chlorite, muscovite, and to a lesser extent, biotite. A noticeable featurf^ of all these rocks is their remarkably low lime content. The Mt. Kenneth and Quinn’s rocks both have the general appearance, texture and mineral compositioiq of original medium-fine grained, impure argillaceous grits, which since compaction have suffered some crushing under shearing pressures, and with increasing temperatiu’es have reached an early stage of recrystallisation. In the case of one of the Mt. Kenneth rocks kyanite has been j)roduced as a result of a continuation of shearing stress at the increased temperature. This stress must have been largely absent during the gi’owth of andalusite in the Quinn’s rocks but some retrogj’essive altera- tion of this mineral to chlorite and muscovite in specimen F suggests a later increased pressure during cooling. SUMMARY. During the 1939 field season of the Geological Survey, an oc(mn'encc of kyanite-bearing quartz schist, a type hitherto unrecorded from the district, was noted amongst the hills of the Camel Humps, north of lan'ei'ton, Mt. Margaret Goldfield. This rock is a fairly high-grade regionally metamor- j)hosed sediimmt which forms portion of an horizon of banded ferruginous quartzite. In bf)th its field occurrence and its petrography, it is closely comparabh* with an occurrence of metamorphosed sediments at ^Mt. Leonora, Mt. Margaret Goldfield. Mt. I.eonora voek is i UHHlominantly an andalusite-hearing (piartzite or (iuartz schist in which the andalusite has I'eached only an early stage of crys- talloblasti(' development. Although some siiecimens contain minor quantities of the higher grade stress minei’als kyanite and sillimanite, mineral eomposion 24 : Keith l\. Miles. indicates that, unlike th(‘ Camel Humi)'s type^ shearing stress was largely inelTective (luring rec-rystallisation and when at all ap[)reciable occurred only in the closing stages. The miueralogical compositions point to closely similar chemical compositions J'or th(‘ Camel Humps and Mt, Leonora rocks. Two other interesting varieties of meta-sedinumts fi'om Mt. Kcmneth in the Valgoo Coldtield, and from Quinn’s in the Murchison Goldfield are kyanite- bearing (luarlz-cliloi'lte schists and andalusile-bearhig (|uart/-chiorite schists respectively, the lattm’ containing incipient kyanite. Comparisons of tin* chemical analyses of these with the Mt. la‘onora rock indicate that the two formej* (consisted oi'iginaily of \{‘ry similai- sedinumtary mat(U'ial — argilhu-eous sands or line grits— and that whih* this may hav(‘ differed in minor details from the original Mt. Leonora (and Camel Humps) sediments, all these rocks Avere of ess(mtially similar chemical composition. The existing miueralogical differences have been i)rodu('ed ]Aartly by slight chemical variations in the original sediments, and partly as the result of differences in thf‘ type ami degree of metamoi'phisrn which they have suffered. Tile (liscovtM’v of a kyanite rock associated with the (Jidcu* Greenstone Series in the Mt. Margaixf Goldfield is of (-onsiderable interest in that it provides e\'idence of the existem-e in this district of an hitherto unsuspected high grade of I'egioual metamoriiliism comi>arable with that found in such plact's as tlu' (dHtt(*ring Valley and the '^'ilgarn Goldihdd. ACKNtnVLKDGMENTS. The writ(‘r gratcd'ully acknowledges his indebtedness to Dr. 1C T. Crider of the Depai'tmenl of Geology, Ciiivci'sity of Western Australia, for the inspiration of his (mthusiasti(' inter(‘sl and in\’aluablc jiractical assistance', and for adA'ice in distnissions and I'evision of the text of this jiaper. In additiem he wishes to thank Profe'ssor E. de (C Clarke, Avho made available hand-sp(‘cim(*ns and sections from th<‘ I 'uiv(‘i'sity G(‘m'ral (\)lh*ctiou, and Mr. H. J. Smith for *])ra('tical help Avilh tiu' pliotogvaphs. Einally, acknowledg- ments are also due to tlu' Govcu'iinumt Geologist for permission to mak(' us(' of s])ecim(ms and thin slict's from tlic Geological Sur\'(‘y collections, and to publisii this papei' and the acemnipanying maj), and also for leave to reprodin-e a photograph (Fig. Ik) fi‘om the exc-ellent Survey collection. G(‘ological Snrv(‘y of Western Australia, I^erth. HEFEKENC^ES. ( 1 ) Tilley, C. E.; ‘^Tlie Eole of Kyanite in the TTornfels Zone of the Cam ('liuhiiieiig (iraiiite, Koss-sliire. ^ ’ Min. Maff. 1935, 24, p. 92. (2) “ Metaiii()i‘]ihic Zones hi tlie Sonthei-n Highlands of Scotland.'^ Qtiart. Journ. OroK Soc. 1925, 81. i»)j. 100 - 1 ( 1 . (3) Harrow, G. ; ‘‘The Geology of Lower Deeside and the Southern Highland Border.' ’ PrOr. Gcol. A.s‘soc. London. 1912, 23, pp. 274-90. (4) Harker, A.; “ Metainorphistn. ’ ' London. 1932. (5) Simpson, E. S.: “Contributions to the ^liueralogy of W.A. Ser. 1 (4).^' Jour. Loll. Soc. W. AuM. 1925-6, 12, pp. 62-6. ((i) ‘‘The Oecurreni'e ()f Andalnsite, Kyanite, Sillinianite and Staiirolite in the Cliittering \'allev.'’ Jour. Pay. Soc. II'. Anst. 1931-2, 18, p. 75. (7) ‘ ‘ Kew Mineral Records — Kyanite, Middle Cliittering Valley." Ann. Pept. Chew. Branch Dept. Mines JV.A. Perth. 'l934, p. 78. (S') “ Coiitril)iitions to the Mineralogy of ^^\A. Ser. 9." Jovrn. Poy. Soc. 7V. Au.st. 1935-6, 22, ])p. 10-13. Some Ivyanite-bearikg Rocks from the Eastern Goldfields, AVestern Australia. 25 (9) ‘‘New Alineral RecMirds — Kyaiiite, AJilly Alilly Station, Murdiisoii River.’’ Ap/i. Prog. ihp(. GeoL Surv, fJ'.A. 191 o, p. 137. (10) “ [\v;uiite .Deposit at Siiiitlifiekl/ ’ Afi/t. Prog. Pept. (jfuL Aarv. If .A. 'for 1939, p. 10. (11) Miles, K. R.; “The Geology and Pliysiogra [diy of the Lower Cluttering Area.” Jour. IPn/. Soc\ IWAiuA. 1937-S, 24, j.p. 13-12. (12) Ellis, IL A.: “A Kvanite Dejiosii 10 miles S.W. of F^ridgetown. ' ^ Ann. Prog. RepL Gool. Surv. ff'.A. for 1939, pji. 9 10. (13) Carroll, D.; “ Kei)ort on Latorite Speeiniens from Kyauite Locality, S.A'. of Bridgetown.” Anu. IJ-og. Repi. Geol. Surv. W.A. for 1939, p. 11. (14) Talbot, 11. W. B. : “A Geological Reconnaissance in the Sontliern Portion of the A'algoo Goldfield.” Ann. Prog. Repf. Geol. Surv. fV.A. for 1919, pp. 9-10. (15) Jackson, C. E. W: “Geology and Auriferous Deposits of Leonora, Mt. Margaret Goldfield.” Geol. Surv. 11. A, Pull. 13, 1904, |)ji. lS-20. (10) MaitlaJid, A. G. : “Notes j)ii tlie Geology of Leonora. An'n. Prog. Repf. f^eof. Surv. Jl'.A. for 1909, p. 11. (17) Clarke, E. de C. : ‘“I'he Field Geology and Broader Mining Features of the Leonora-Duketon District.” Geol. Surv. H'.A. Bull. 84, 1925, pp. 23-4. (18) Gibson, G.: “The Geology ami Mineral R’esources of a part of the Murchison Goldtieid.” Geol. Surv. IV.A. Bull. 14, 1904, p. 40. (19) Feldtniaun, E. R.; “The Mining Centres of Quinn’s and Jasper Hill, Mur- chison Goldfield.’’ Geol. Surv. IV. A. Bull. 80, 1921, pp. 13-19. (20) Bastin, E. S. : “Chemical Composition as a Criterion for Identifying Metauiorpliosed Sediments.” Journ. Geol. 1909, 17, p. 445. (21) MenneJl, F. P.: “The Geological Structure of Southern Rhodesia.” Quart. Journ. Geol. Soc. 1910, 66, p. 357. / i < ( i 1 j 4 - . .. >> •r. ?• t. .#■•*;, •■'•A- /»■ .1 » lv^’4' X, i* s." ,**■ p t- -S'.. v.*f The Contact Between the Granitic Kocks and the Caedijp Series AT Armadale. 27 5 .— THE CONTACT BETWEEN THE GRANITIC ROCKS AND THE CARDUP SERIES AT ARMADALE. By Rex T. Prider, Ph.D., B.Sc., F.G.S. Read 13th August, 1940; Published 28th November, 1941. CONTENTS. Page I. Introduction . . . . . . • . . . • • • • • • II. Geology . . . . . . . . • ■ • ■ • • ■ • • • ^9 (1) The held distribution and age relations of the rocks 29 (2) The geological structure . . - - ■ - • • • ■ 31 III. Petrology . * . . ■ - • • • • • • • ■ ■ 34 (A) The older hybrid gneisses and the associated xenoliths 34 (B) The younger granite (aplogranite) . . . . . . 37 (C) The fine banded granitic gneisses . . . . . . • • 37 (D) The Cardup Series . . . . . . - ■ ■ ■ - ■ 38 (i) The sandstones . . . . • . ■ • • • 38 (ii) The slates . . . . . . . . ■ • . ■ 38 (iii) Contact nietamorphism of the sediments . . 40 (E) The basic intrnsive rocks of post Cardup age . . • ■ 43 (F) The quartz veins . . . . . . ■ ■ - • ■ ■ 48 IV. Conclusions . . . . . . - . ■ • ■ ■ ■ ■ ■ • (a) The age of the Cardup Series . . . . . . - - 51 (b) The Barling ^ ' Fault Scarp .. .. .. .. 52 (c) Geological history of the Area . . . . . . - ■ 52 V. Acknowledgments . . . . . . . . - ■ ■ • - • 53 VI. List of References . . • • ■ . ■ - • • • ■ 53 I.— INTRODUCTION. The Darling Scarp (Jutson, 1934, p. 84) is a prominent physiographic feature of the country near Perth. Tt forms the western edge of the Western Australian Pre-Cambrian plateau, which stands about 1,000 feet above sea level, and which is made np mainly of granitic roc*ks with later doleritie intrusions. West from this scarp and extending to the coast is a compara- tively flat coastal plain underlain by Cainozoic and Recent sediments and Bordered along the coast by Recent sand dunes (Clarke, 1926). On the western face of the scarp a series of steeply dipping slaty sediments Avhich form a narrow band beUveen the granitic rocks and the younger horizontal sediments of the coastal pdain is developed in a number of places extending from Gosnells and Kelmscott in the north to at least as far south as Mundijong (Honman, 1912 and Esson, 1927). These slaty sediments have l)een referred to as the Cardup Series (Clarke, 1930, table, pi. 187) and they are considered by both Clai*ke (loc. eit.) and Forman (1937, p. xxiv.-xxv.) to he contemporaneous with the Jinpierding Series of "V ilgarn (early Pre-Cambrian) age and therefore older than the granites and granitic ggieisses of the Darling Range. Forman (1937, p. xxiv.) says that the structures in the Cardup Series 1x4 ween Kelmscott and Miindi,iong confoini to the local structures in the adjoining gneisses and suggests that the gneisses ure, in part, of the same age as the Cardup Series and owe their origin to 28 Hex T. Prider the alteration of these bedimentary rocks by a granitic magma, either the same or an earlier magma to that which provided the massive granites of Geological skctcli ina|) of the country between TTpper Swan and Mundijong and extending to the coast, showing localilies mentioned in the text. The present investigation was undertaken to determine if possible,, whether this ‘^granitisation” of the Cardup Series has actually taken place and to determine the relative age of the granitic gi.'eisses and the slaty sedi- ments of the Cardup Series. I am indebted to Mr. H. A. Ellis, Assistant Government Geologist of Western Australia, for suggesting that the possi- bility of granitisation of the Cardup sediments should be investigated. Tlie area examined in detail is situated on the Darling Scarp just south of the Xarrogiii Inn at Armadale, some 19 miles south-east fi'om Perth (fig. 1). It measures only about mile x % mile, but outcrops are fairly abundant; fresh examples of both the sedimentary and granitic rocks are obtainable in the two quarries; and most of the rock that have been The Contact Between the Granitic Rocks and the Cardup Series AT Armadale. 29 noted elsewhere in the vicinity of the Darlino- Scarp are developed. This small area is therefore perhaps the most suitable place to commence a de- tailed study of the petrology and structure of the Scar]). 11. GEOLOGY. (1) The fiehl dlstribaiion nnd age relationH of the rocks. The eastern half of the area (Plate 1) is occupied by granitic gneisses in which there a])pear to be two distinct grou]>s: (i) The tine banded granitic gneisses forming the western edge of the granitic rocks. These form a band apj)roximately 8 chains wide Avhich trends 350'^ parallel to the contact of the granitic rocks and the Cardu]) sedi- ments. The foliation in those gneisses strikes ])ara]]e! to the gneiss-sediment contact and dips at 50° to 00° towards the east. These gneisses apjiear to be uniform and granitic in character throughout the whole band and no- where were they found to carry .xenolithic inclusions as do tlie hybrid gneisses farther to the east. Contortion of the gneissi(' l)anding is visible in places and in other })laces tin* banding may ])e almost invisible, the rocks there being indistinguishable from massive granites. (ii) The hybrid gneisses which occupy all the area to the east of (i). These gneisses have a general strike of 25° and dip at angles of 40° to 60° to the east. They are l)est exposed in the Roads Board (piarry in the north- eastern corner of the area. The rocks in this (piarry and their titdd relations will be described in some detail as it is clear here that there have been two distinct periods of granite intrusion, the twidence for which periods of granite intrusion in the Pre-Cambrian history of Westei'n Australia has been accumulating during recent years (Miles, 1038, p. 36; Prider, 1938, p. 101, and 1939", p. 93; Ellis, 1939, p. 91 ). The rocks developed and their ti(‘]d relations as seen in this quarry are (a) Dark coloured biotite-epidote-hornblende hornfelses. These rocks are well developed on the southern wall of the quarry where they ap])ear to form a flat lying band in a more granitic gneiss (b) which is crowded with xenoliths of this dark green hornfels. On the westtum Avail of the quarry, just inside the entranct' the biotite-epidote-hornblende hoiTifels occurs as niunerous xenoliths U]) to 12 inches diameter accompanied by xenoliths of coarsely granular quartz ( ‘F quartzite) in a hybridised gneiss (b) (plate 2, fig. 1). This dark coloured hornfels (together with the (piartz xenoliths) is clearly the oldest rock exposed in the quarry. (b) Mesocratic hybrid augen gneisses. These form the base in Avhich the xenoliths (described above) occur. They are well banded coarsely granu- lar rocks, often exhibiting augim structures and are composed of quartz and felspar with thin lenticular dark coloured bands, Avhich under the micro- scope arc seen to Ix^ remnants of the biotite-e]')idote-hornblende hornfels and they appear to have resulted from the intrusion of granitic material into the hiotitf^-t'pidote-hornblende rocks (see under ]>etroIogy) — this rock thus represents the first period of granite intrusion (or granitisation). The best exiiosuivs of this rock are to be seen on both faces of the (juarry just inside the eiilrance. (e) A])]ograiiite. This rock, a leucocratic medium grained massive to A’eiw slightly gneissic granite with little or no forro-magnesian constituents (with the exception of occasional biotite-epidoto-hoiTd)l(mds clots ])icked up 30 Kex T. Pkider. from the older rocks) is exposed over the g-reater jiart of the north-west and south-west walls of tlie quarry. It is clearly intrushe into the hybrid gneiss (b above) which, in i)laces, occurs in the form of large irregularly oriented angular blocks suspended in the aplogranite which truncates the banding of the gneiss. The aplogranite is therefore definitely a later in- trusion and I'cpresents the second period of granite intrusion. (d) Dolerite. A [)ost-aplogranite dolerite dyke is present near the north-eastern edge of the main (piarry and is younger than all the above mentioned rocks which it cuts indiscriminately. The exposiu’cs in this Koads Board quaiTV therefore give us consider- able information regarding the ago relations within the eastern group of gneisses of the Aimadale aiX'a. The sequence is: — Younger. (4) Dolerite dykes. (3) Massive aplogranit(’s. (2) Hybrid augen gneisses. Older. (1) Biotite-ei^idote-hornblende bornfelses and (juartz xeiio- liths in the hybrid gneiss. The field relations of the eastern group of gneisses to the western group are unknown. The eastern group shows the highest degree of hybridisation in the north-eastern part of the area but continue to be hybridised to some extent even at the south-eastm’u corner. A number of well defined quartz veins which all have the same strike of approximately 305° and dips of 85° to 90° to the north-east are seen to cut through both the hybrid gneisses and the fine banded granitic gneisses. Two of these quartz veins (one of them some three or four feet wide) were traced right to the contact of the gneiss and tlie Cardup sediments but cai'e- ful search sliows that they do not extend ]»eyond the contact and they appear to be of inv-Canlup age, probably representing an ultra-acid differentiate from tlie aplogranite magma. If this he the ease the position of the Cardup Series in the Pre-Cambrian siicc(‘ssiou is lixed as younger than the granite- but older than the (luartz dolerite intrusions, which are considered by all observers to have taken place during late Pre-Cambrian times and which represent the youngest ro('ks in the Western Australian Pre-Cambrian shield. (Clarke, 1930; Forman, 1937.) Quartz veins striking almost due north also occur but are not so well developed as those which strike north-west. The main imunber of the north striking group forms a lu’omitient quartz blow on the north side of the Bedfordale Road near the southern lioimdary of the ai-ea ( Idate 1). The ^‘blow” is lenticular and a])pears to be surrounded by granitic gneiss. There are some small miueworkiiigs on this blow in the form of a shallow incline shaft ami an adit — observations made on the white quaifz from the surface and the softer more micaceous rock below are described in the section on petrology. The surface (juartz in jilaces shows the presence of irregular' strings of line grained dark coloured roek, not uidike slate in appearance. Extending north from this (luartz blow along the same strike there is a band where no outcrops are to be found but wlim'e occasional small l)oulders of the dark coloured tine grained slaty rock are noticeable. There is no doubt that this band of slaty roek extends north to at least as far as the centre of the mai'tped area (Plate 1), Further consideration of the origin of this rock is given in the petrology section of this paper. The Contact Between the Cranitic Bocks and the (\vrdi'p Series AT Arhadale. 3T The westei'u stri|) of the area is oeeupied by the sedimentary roeks of the Cardup Series. These rocks strike parallel to the o-ranite gneiss- sediment contact and dip (on the average) 60° to the west. Ti-averslng this series in a westerly direction from the contact the succession is (luartzite (with fine cherty bands), sandy slate, tine white slate, dark greyish to blai'k caibonaceous slale^ and tine whit(‘ slate (lliese I'ocks ar(‘ leferred to as slates but they are little more than shales with a vtu’y jujorly developed fracture cleavage). A few exam})les of graded bedding were noted in tlie section exposed along the Bedfoi’dah' Hoad at the south end of tlie area and reailings taken here indicate that the, older beds lie to the east, i.e., that the sequence is normal and not overturned. Tlie actual contact with the granitic gneisses is only \'isibh‘ in one jilace (on the roa^l at tlie south end of the ar<‘a) but it can be Jixed within several yards over the remainder of the area and the succession of beds is everywhere tlie same. The Cardup sediments are best exposed in the quarry at the south-western corner of the area (whex'e the slates have been quarried for brickmakiiig) and are seen t<» be slightly drag- folded and traversed by numerous minor faults with displacements of several inches. The nature of this minor faulting and drag-folding will be more fully described in the following section dealing with the structure of the area. The Cardup Series is cut by a number of doleritic dykes which may be traced from the granitic rocks into the neighbouring sediments. From the geological plan of the ai’ea (Plate 1) it will he seen that the dolerite dykes appear to have been intriuhni along a definite set of fractures which extend from the gi’anilic rocks into the sediments. In one place (on the Avest wall of the slate quarry) gi'oenstone has been intruded in the form of a sill in the sediments and has produced sliglit contact alteration of the slaty roedvs. No such contact effects Avere noticed along the gneiss-sediment contact. From the field occurrence the chronological order of the roeks exposed in the area maiJjied ai>pears to l)o; — Youngest. (8) Minute barite veinlets in the dolerite on the east face of the slate quarrA\ (7) Dolerite dykes. (6) Plpidiorite sill in Cardup slates. (5) Cardup sediments ((juartzites, sandy slate, white slates, black carbonaceous slates, Avhite slates). (4) Qiuu’tz veins in gneisses. (3) A|)logTanite intrusions. 4 Fine banded granitic > gneisses, exact relation to (2) Hybrid augen gneisses. ) (2) and (3) unknown. Oldest. (1) Hiotite-epidote-horiiblendo hornfels and quartz (? (qiartzite) xenoliths hi the hybrid ang(Mi gneisses. (2) The Geological Structure. Mapi)ing of all of tlie axailable minor structures of the exposed rocks has yielded a considerable amount of information regarding th(‘ structure of the area (and of the Darling Scarji generally). The hybridised gneisses of the eastern section shoAV minor folding in a nuudier of places but no con- stancy in the orientation of these structures could l)o found, as Avill be readily understood after an examination of the exposures in the Roads. Board quarry, where the gneisses are seen to be in xenolitlhc lilocks, irregu- Rex T. Prider. :J2 lady (>ricnt<‘(l, in the intrusive a|)l()i»'ranit<‘. No attempts have therefore been made to intei’pret the minor dra;^- folded stnud-iires seen in these rocks. The drag-folds in the fine banded granitic gneisses are more constant in character and indicate that the easterly dipping rocks of this l»Jind form the eastern limb of a normal anticdine witli n pitch to the south varying from 0° to All outcro])s in this band were carefully examined for these drag-folded structunis and wherever visible they invariably indicated that this band formed the east limb of a normal anticline. Ttik Contact Between tiie Ghanitic I^ocks and the Cardtp Series AT AliilADAEK. Minor structures (other thiui the g'raded belaces on the Bedfordale Road) observed in the Cardup sediuuaits were visible only in the slate ed. A detailed survey of the (juarry was niad(‘. tor dragfolding', minor faulting and fracture cleavage and the results are shown in (igiire 2 (plan and set*- tion of the slate quarry), d'he drag-folds indi('ate that the beds to the west have moved up and over the beds to the east, i.e., that the slates in this quarry form the westei’u liiidi of a normal anticline. A close examination across the black carbonaci'ous slate band at the south end sliows that the dragfolds are directed in the same direc'tiou right throughout tiu* band thus precluding the possibility of isoclinal folding in tliese rocks. All of the dragfolds observed pitch to the smith at 10° to 15°. The bedding planes of the dragfolded slates show a, distinct set of sli<'kensides (Ilarnisch) which strike K. to W. normal to the axes of the dragfolds. Tests mad** of the roughness of the b(?dding surface in this direction indicate that the beds to the west have moved up and over the beds to the east. The fracture cleavage is rath(*v poorly developed in these rocks but where present is (-onstant in strike (])arallel to the bedding) and di]) (almost vertical) throughout- the quarry. In the northern wall of the (piarry there are a number of small faults de\eloped which ai'e generally ])arallel to the fracture clea-vage and along which movements u]) to 6 inches have Ixm'U observed. AH of these small faults (which an^ either vertical or di]) very steej'jly to the west) are of nnersed (diaracter and indicate uplhrusting fi’om the west (as deduced from the curvature, due to dragging, of the bedding planes in the vicinity of th*' faults). It will be seen that all of these minor structures ar*^ *dosely related and almost certainly developed by the sanu’ movenu'iit, viz., a moviummt causing the westeiTi beds to be moved uj) and over the beds to the east. Tlu'ri* are therefore two possible interju'etations p*>ssible for these minor strnctma's: — (i) Tliat these beds form tin* westcwn limi) of an anticlim* which ])it('h(‘S flatly To tlu* south or (ii) That there has been upthrustiug from the west, the sediments hav- ing been upthrust over the graniti*' gneisses lying to the east. Considering the striudural IVatures ol’ the Cardup sedinu'iits in con- junctioii with the structures in tlu* band of tint* grained granitic gneisses adjoining tlieni lo tin* east we s('e that they ap|)ea.r to bt* to some extent complementary, the gneisst's forming tlu^ east Hmh ot an aidicliiu', tlu* s(*di- ments forming tlu* wt'st limb ot flu* same anticline. Foi'inan (1057, p. xxiv.) considi*rs tliai tlu* gn(‘iss('s may I'epresent jiorlions of tlu* (knrdiip sediments which iun** sudei’ed gi'anitisation and it has I>e(*n sugg<‘sted to me (in dis- cussions with Mr. n. A. Fllis) that in the Armadale Ar(‘a tlu* gneiss<*s and sediments form part, of a singit* siriictural unit (an an1i(‘linal told ) wtiich at the tinu* of folding *'onsist**d (*ntirely of sediuu'iitary ro(dlaces remnants of untwinned ( ?) plagioe'lase (with the I'efringence ap])roximately the same as tliat of canada balsam) ('rowded with sericitic inclusions are present. This rock appears to have resulted from the alteration of a basic plagioclase rock which must have been closely associated with the rocks from which tiie next grouj) of xenoliths were derived. (iii) JTornhlcnde-epidote-hiotiic hornfels .coiolitlis. XeTioliths of this type, which are dark grtHUiish to l)lack in colour, are the most abundaid typ(‘’and liave exerted (considerable control on the char- acter of the hybrid gm>iss, baving conlril)uted most of the f(*rromagnesian content of that rock. Rocks of this ty{)e a]>i)('ar to be ratluu' constant in cliaracter wherever noted and are best developed in the Roads Board quaiTv. They are fiJie even grained, Avilh no trace of any dii'ccted structure, the structure being coarse hornfelsic. Tuider the mi('roscope the rock is seen to consist of a (h'cussate aggregat(‘ of liiotite and hornblende associated with granular aggregates of epidote. Irregular grains of magnetitf' ar(‘ scattert'd uniformly throughout the rock, and a]mtit(‘, in stout euhedra to 0,.“i mm. diameter, is abundant. The biotite and hornblende are closely associated. The biotite is a bi*ownish-green, ])ractically uniaxial variety with jS = 1.637 and in all of the slices examined is considerably in excess of the bhiish-green amphibole The Contact Between the (Jkanitic I^ocks and the AT Akajadaee. CARuri’ Series' ;i5 which oecui's in well sliaixul prisins with il’iVLi'ular tcrniinnlioiis, towards which the biotite is idiol)lastic. The anipliibole has pleochroism, X brownish- g:reen, Y biajwnish f^veeti, Z ^T(‘en (slightly bliiisli) and absorption X ■' Y Z. The (‘xtinction Z A c is 18° and the oi>ticnI chai‘a('ter -ive. In several s])ecinu‘ns a. blLU‘-gre(‘n anii>hibole with X light y(’llow-gTeeji, Y olive green, Z liluish-giHaai, ahsorjition X < Y Z and Z A c •= 17°, is idioblastic lowai'ds the biotite. It (X'ciirs in elongated prisins, often in clusters with the long axes of individual pi’isnis snbparallt*]. Th(‘ e|»i(lole occurs in ])atches with an aggi’egate structure, made up of a mesh of tiny euhedral prisms, with which is associated some brownish- green biotite (in nuudi smaller Hakes than in the biotite-hoi-nlilende ai'cas). The epidote is practically colourless and brightly ixdarising pistachiie, although some tine granular zoisite a)>})ears to be pr(‘seut in the granular aggregates. These el>iotash. F^’rom the texture of the rock it a]) pears to have been originally a medium grained basic igneous rock such as a gabhro although the PeO/MgO ratio is rather high. It must be borne iu mind, however, that this rock has suffered considerable change, the original pyroxene having gone over to biotite, involving a loss of silica, magnesia, and lime and an addition of potash and alumina — the chemi(*al evidence then supports the -suggestion that these rocks were originally basic igneous types. There has been a marked concentration and hxing of in these xenoliths. No rocks have been noted amongst the Cardup Series which (‘ould possibly be related in any way to the xenolithic masses in the hybrid gneisses. The hffhridised gneiss itself is a medium grained, well handed mesocratie rock. The dark hands which produce the gneissie appearance are composed largely of material similar to the basic xenoliths described above. Tnter- banded with this material are leucoeratic bands of pale tlesh coloured felspar closely associated with quartz and a pale greenish slightly epidotised fel- spar — in })laees the flesh coloured felspar is present as augen. Tjie Contact Between THE (jRAxrnc Kocks and the (’ardup Series AT AKilADALE. :]7 Under tiie nn('ros<' 0 ])e the most proiiouiuT'd feiiluri' is the i)r(*si*neo of dark cdots, elong’nt(‘d i)arall(‘l to the l)nnding‘, whieh are made u]) ot' brownish green biotite, eidiedi'al apatite, s])h(m(‘, magnetite, and rare blue-grecm liorn- bleiide (fig. 3A). These (-lots are undoubtedly deriveal fi'oni the basic xenoliths wliieh are so abundantly seatler(‘d tliroiigh this gneiss (plate 2^ fig. 1). The other constituents are eonsitlerably (*pidotised plagicxdase (most abundant), clear microcline' and (juartz. Tlu' <|uartz is all slightly strained. Both quartz and miciHxdine hav(‘ emdosial the ])lagiooth silica and soda to the original basic rock. As mentioned above some of the {piartz of the hybrid gneiss may be xenolithi(' although the greatei* part, as seen from its relations with the introdneed microeline, has undoul)todly been introdueed during the graiiitisation in-oecss. Head (192(i, p. Ifir)) has desciabed 11u‘ process of hybridisation of a gi'anitic magma l)y nltrabasic mat<'rial and his nanarks api)ly etpially well to the Armadale hybridised roc'ks. The basic xenoliths, in tinur high biotite content, are clearly hybrid typos laqna'senting basic igneous nuitfu’ial which has been [)ermealed with potassic solutions from an intrusive granite. ’’I’his gi’anite, wliitdi re[)resents the first period of granite intrusion in this area, was itself consid(*rably hyl>ridis(‘d by the incorporation of this basic hybrid and is now represented by the mesocratic hybridised gneiss described above. (B) The ‘ffounger granite { A])Iogra}ii(e) . The aplogranites are leucocratic, medium to coarse grained (in ])la('es ]>egmaiilie) rocks composeale greenisli to colourless, not sensibly pleochroic, y = 1.644). The origin of these peculiar cherty bauds and lenticles is obscure but they appear to be original serlimentary structures which have later been contact altered. The most probable origin that suggests itself is that they rei)rGsent small clay balls whidi have been flattened during the earth movements to Avhich these rocks have been subjected. The presence of (?) actinolite suggests that these balls may have been to some extent calcareous. The coarser gritty parts of these greenish (juartzites are similar to the Avhite (puirtzites with the addition of granular epidote which often occurs as angular grains moulded around the larger detrital quartz grains indicat- ing that the ei)idote has b(‘en introduced after the formation of the original gritty sedinumt and was j)robably derived from the quartz dolerite intru- sions which cut througii the sediments in this locality. (ii) The slates. Two distinct types of slate are developed — their held relations have been discussed above. (a) The black slates: these are finely bedded, dark grey to black in colour, different bands having varying content of graphitic material. The dominant constituent is fine fibrous sericite together with extremely fine granular weakly birefringent (?) quartz and black graphitic material. An interesting constituent is pale brownish green tourmaline, Avhich occurs in The Contact Between the Gkanitic Rocks and the Cardcp Series AT Armadale. 39 minute idieblastip prisms to O.Oo mm. in length scattered sparsely, Avith random orientation, throughout the rock. It is strongly pleochroic, oj brownish green, e colourless, and the refractive indices e = 1.621, oj ^ 1.649, ai»pear to indicate a membei of the dravite-schorl group. (b) The white dates: these rocks are made u]) almost (.mtirely of fine fibrous serieite with a much smaller amount of [>ale broAvnish flaky biotite and minute clear grains of (piartz. Minute idioblastic tourmaline prisms, similar to those in the black slates are distributed sparsely throughout the rocks (all of the slaty rocks examined contained this tourmaline none of which is detrital). Narrow darker coloured bands (up to 1 inch Avide) occur in the Avhite slates in places and these consist predominantly of pale broAvnish biotite with serieite in minor amount. The grain in these narroAV bands is considerably coarser but this is largely an original feature and is not clue in the main to recrystallisation as the quartz grains of such bands are markedly larger than in the Avhite slate. The minute tourmaline prisms appear to be more abundant in these bands. In some i)]aces near the Avestern Avail oC the slate quarry, close to the albite-epidiorite, the Avhite slate has a slightly knott<‘d structure, the “knots’’ usually being small rectangular shaped j)latc's (up to 4mm. x 4mm. x 1mm.) which consist of kaolin Avith small d any relicts of this mineral. Their true character must thenTore remain unknoAvn but they ajipear to be the result of contact nufamorphism of the slate by the nearby basic intrusion. Joint jdanes in the Avhite slates near the south-Avest corner of the slate quarry are coated Avith ])ale yelloAV earthy jarosite. Other than the o('casioiial thin dark bands and rare examples of the knotted slate, this group of Avhite slates appears to be rather uniform in character. An analysis of a specimen from the AA^est side of the slate quarry yielded the result shoAvn in table II. Table IT. 'SiO., 1. 65-22 2. 61 -63 3. 58*38 4. 50-10 5. 50-00 ai.,6, 16-71 16-33 15-47 25-12 24-14 FcoO, 1-93 4-10 4-03 5-12 2-79 FeO 3-23 2-71 2-46 1-52 4-67 MgO 2-87 2-92 2*45 3-93 4-16 •CaO 005 0-50 3-12 0-35 0-07 Na„0 0-76 -26 1*31 0-05 MO K.,0 H'0| 5-98 5-54 3-25 6-93 8-63 2- 12 3-24 3-68 6-82* 3-06 H.O— 016 0-31 1-34 . . 0-24 TiO., 0-35 0-68 0-65 0-50 0-50 €0., 0-05 0-41 2-64 0-07 P.,0, 0-08 0-16 0-17 . . 0-11 MnO 0-03 0-09 Tr. . . 0-04 BaO 0-1 1 O-OB 0-05 015 FeS., Nil O-OI . . . Cr.,0., 0-01 . . . 0-01 V.,0.. 0-03 ... 004 SO, 0-11 « « « 0-65 . . . 0-15 Ora])liite 0-05 Nil O-Slt 0-07 NaCl Nil ((Others) ... 0-06 ... 99-85 99-98 100-46 100-44 100-00 Analyst : C. R. Le Alesurier. * [gnition loss on material dried at IIO^C. t Organic material. 40 \ih:x T. Pridkr. 1.. sl;it(‘ west side <4 slnte (|ii;nry, Arniaand ol greenish mor(‘ sandy slate, ddiis hand swings towards the ea-st at tlu* south {‘lid ol Ihe (piarry. This nn-k is distinctly coarser lextiin'd and (comparatively larg(* rounded (piarlz gi’ains ai‘e abundant — the (^nly other (const it iieiit ol note is a gr('('nish lirown hiotitx* which app(*ars to have de\(*lopcd during ('ontaet metamorphism ol this band bv tlie basi(' ijitrusive. S(*paration oi llu* h(*avy min(*rals yiehh'd only a IV'W miudi woiai and rounded /ii('ons. A| tlu* south ('lul ol tlu* (piarry this gi'(‘enisli ro('k (‘ludoses several rounded lioiilders ol granitic material which consist lai’g(*Iy of a graiiophyri(C in((‘j'grow(h oi (piart/, and acid plagioelase, T!u*se “boulders’^ appear to be sonu'wiial rounded and some doubt exists wheilier th(\v are boulders or irregular g -anopliyrie intrusion.^, as tlu'.v are lrav('rsed by nan-ow (%dnch wide) v(*inh*(s ot (piarlz whiidi app(‘ai' to ])ass through both tlie boulders and llu* (‘ludosing green sandy s('(Iim(‘nt. I incline to llu* view (hat they ar(* honidei’s and that llu* associated v(‘inl(‘ts aia* similar to the (piartz v(*ins d(*V(’lop(Ml in (lio (hirdup Series (whi('h are descri])(‘(i iiiuh'r “(|iiartz veins'^' I)(*1()W ) . (iii) ( oiitaet vtefidHorpliisni oj the ('avdep sediteoits. Donsi(h‘ring first the an'iuuM'ous ro(dcs (-lose to tlu* I'ontaet with the gi-anitie ro(-ks— (he main eonlaet metamorphie (‘ffeets have h(‘('n (he intro- (Inetion of (‘pidoti* and llu* (lev(*lopm(‘nt in certain narrow hands of a(-i(-ular (f) aetinolib* the latl('r ari* in radiating aggr(*gal('s and appear to iM'snlt from (he liu'i’iual nu*tamor]>iusm of a slightly eaIcar(*oiis rock hut wh(*llu'r this alteration is due to intrusion by (lu^ granite oi* to intrusion by the (]uartz (lo!(‘rit(' dyk(‘s is not ch'ar tlu* introdnetion of (he ejiidote was most likely efr('('(('d during the intrusion of tlu* doleritif^ rocks. In the slat(*s lying farther to the w(‘s( tlu*re lias he(*]i some new mimu'al d(*v('!opmen( : — (.1) The (levelopnu'nt of the kaolinic knots whi(di seems to be closely relat(‘d to tlu* gi’oenstoiu* intrusions. The Contact Between tile (Jiunitic Hocks and the Cardcl^ Seiues AT ARilADALE. 41 (2) Tlio pros{'iK'(' of idioblaslic Itmniuiliao in all tlu* slaty rocks of this firea. is intcrcstia^;' — tliis mineral may have been introdueed from j>'i*aiiitio intrusions, from the basic intrnsiv(\s (AjiTell, 11)39, p. 333, lias describtHl the development of dravitt^ in adinoles ainl (‘onsiders that the boron was intro- duced from nearby albiti' dohu’ites) oi' may have developed from the boron contained in the oripnal unmetainoriihosed sediments. (loldschmidt and l^eters (1932) ha\'e shown that the boron content of clay sedinumis is oftim sufficient to bring- about crystallisation of tourmaline when thest' s(*diments suffer dynamic metamor])hism and tliat it is not rn‘cessary to look to later acid intrusives for the origin of Hu* boron. The absence of tourmaline from the rocks immediately adjacimt to tlu* granite and its (auavrrence in the slates some distance away seems to suggest that it was not d(*rived from that source. It is possible tluai that tlu' tourmaline was introduced from the basic igiu'ous rocks but niori* probable that it r(*sulted from tlu* crystallisa- tion of original constituents of the argillaceous sediments from which the slates wert* ih'rived. (3) The devt'lopment of biotite aiul (diloritoid. Biotite has undoubtedly been formed dui’ing tbe iiKlamorphism of the slaty rocks as it is ol‘t(‘n W(*ll developed on the fradure (deacage sui’fac'es. It is well dt*\’eloi ed in small ])ale brownish Hakes in the slates from tlie slal(' (juarry which an* (dost* to greenstom* intrusions but is alisent I'aidher to tlu* north (akmg the Hunbuiy Hoad) where tlu* slates ar(‘ some distance from the intrusive dykes, d’his ]>oints to llu* de\(']opmenl of tlu* Inotite lu‘iug diu* to contacd alt(*ration c(m- secpu'nt upon the intrusion of the hasi<‘ rocks. Specinu'ns from tlie liaulage tunnel h‘ading out from the south-w<*st ('oriu*r of the slatf* (|uari‘y alforil information reganling the contact alteration (d‘ the (’arlasts, at lirst rare and tlu*n hi‘(*oming more abumlant towai’ds tlu* west. Tlie diloritoid porjibyroblasts are well formed })ul have been r(*]ilaced by penniiiit** with exlr(*mely weak biri'f'ringt'iuu* (almost isotropic' willi very weak ultra bliu* inter- ter(‘nc(‘ colours), ])osili\'e (‘longalion and marked jilemdiroism X pale yellow green, V - / d(*(*]i gi-ei'ii. At lirst sight tlu'se small porphyrohlasts appc'ar to lu* eddoritoid but tlu* pleoc'hroism and m-ieiilation are those of pcmniiiite. 'Tlu'rc* is little doubt, however, that they were orig-innlly (diloi'iloid. This is tlu* only jilacc* where this mineral has l)(‘en I'onml in the area and it is of interest as it sc'em.s to tix the ]U‘riod of intrusion of tlie chloriti'-alhif (' (‘pidiorite as more or less (‘onlemporam'ons with the earth movenu'nis affect- ing the (karduj) Sc'ries. (diloi’itoid is genei'ally regarded as a sti’{*ss mineral— in the prc'sent instance* it is (dt'arly rdat(‘d in scum* way to the epidiorite sill as it dc'c-n'iist's in amount away from that body. * All lueasureinentN aio iroiii liu* east end of the tiimu l uiul luive ]je( n iiuasured along tile north wall. 42 Hex T. Prider. It api)ears most probable that it developed during the earth move- ments wliieh affected the Cardup Stndes while these rocks were- still at elevated temperatures following the intrusion of the green- stone sill. From 40' to 53' m—hlotite-qitartz hornfels — these rocks are hornfelsed argillac(‘ous sandstones. They are tine grained, dark green in colour and under tin* microscoi)e are seen to be made up of rounded quartz grains witli interstitial Oaky pale brown biotite with some pale green chlorite and sericite. From 53' to 00' in — rhlorite-albite rpirliorite sill {described in a later section) with a narrow intrusion of a microporphyritic epidiorite- between 57' and 58'. Fkuu f)0' to 00' Ill- contact altered sanchj slate similar to that be- tween 49' and 53' excejit that it is much Oner grained and apparently less altered, the biotite being in much smaller flakes. Fi’om 09' to 100' in — quart^-hiotite-aetinolite hornfels similar to the actinolite bearing hornfelses developed in the arenaceous sediments nearer the granite contact (described in the section dealing with the Cardup sandstones). From 109' to 311' in — uroHiised qaartz-dolerite. At 311' there are several small irregular bodies ( ? veins) of coarse grained quartz, (which under the microscope shows considerable cataclasis). From 311' to the west end — greenstone, i>robably the ehlorite-albite eiiidiorite but the rocks here become too weathered for exact deter- mination. ■CjU'^L n^ure 3- SiatG Qunrrij ^ n weathered albitc epidiorire Cv)- Scale 0 JO , '00 Peet. LEGEND AS IN 8 (beio«) Wesf riSUR£ 3 -3 • ' j ■ P ' P'lh! r ; 1' ' '' ' S. 1 - ‘ \ ^ A A A, A A yyest ' >> .-**.*\a A ^ A A A Scale Figure 4. 5 0 rc£r A, Geological sketch plan of the tunnel leading out from the south-west corner of the slate quarry, Armadale. B. Enlargement of geological plan of east end of the tunnel from the slate quarry, showing the relations of the Cardup sediments, the clilorite-albite' epidiorite sill and the later quartz dolerite dyke (note the narrow intrusion of microporphvritic epidiorite into the clilorite-alVdte epidiorite- sill). The Contact Between the Granitic Kocks and the Cardup Series AT Armadale. 43 The seGioii exposed in this tnnuel irulicMtes that the Cardup sediments have been eoiitaet metaniori>hosed over limited distanees by both the ehlorite- albite ei)idioi'ite and the later quartz-dolerites. It also indicates that the development of the epidote-aetiiujlite-quart/ite noted near the ^Tanile contact is a contact effeid due to the quai-tz-dolerites and not to the gTanitic rocks. (E). The bmic of p<)!hyrilie epidiorite, which is descrilied below. The chlorite-albite epidiorite is of interest because it can undoul)tedly b(^ correlated with the porphyritic alliite epidiorite (referred to as albite por])hyrite by Esson, 1927, p. (i) which occurs at various localiti(*s (Wun- gong', Cai'dup, Whitby Falls and Mundijong) extending some 15 miles to the south along the Darling Scarp. The rock at Armadale is not por[)hyritic but its texture and mineralogical composition (both of Avhich ditTer con- siderably from the normal (juartz-dolerites) are similai- to those of the rocks from the areas to the south. The Ai’maale greenish prochlorite, very slightly pleocliroic and almost isotropic (with very weak anomalous colf)urs) — the optic character is neutral and the refractive index 1.633 rk ,002. This mineral agrees with the chlorites of the low grade epidiorites desf-rihed liy Wiseman (1934, ]n 360.) Tile albite is fresh (althongli slightly dustc'd with inclusions and penetrated by acicular amphibole) and thus differs considerably from the epidotised plagioclase of the other basic igneous rocks of the area. The analysis of this rock is recorded below : — Table ITT. SiO., ... 47-08 Norm. AlA- 14-78 Fe.,0, 1-32 Or 0-12 FeO ... 13-80 Ah 20-20 MsO ... 7-45 An 18-35 CaO ... 4-24 Na.,0 ... 3-10 C 1-84 K.O ... 1-03 H.,0 H- 3-89 hv 20-49 H,0 — 0-22 oi' 9-47 TiO., ... 2-49 P.0, ... 0-41 mg 1-86 MnO ... 0-31 il 4-71 BaO ... Nil ap 1-01 Cr.A ... Nil py 0-14 FeS., ... 0-14 100-20 Analyst : K. T. Prider. Chlorite-albite epidiorite (20532), 78 feet from east end of tunnel, Slate quarry, Armadale. 44 Rex T. Prider. The most outstanding featiiros of this atialysis are the low CaO and the considerable Na.O content which indicate that the felspar prior to meta- morphism was moderately rich in the albite molecule and that the rock was approa.ching a spilitic type. It certainly differs consideral)ly from the later quartz-dolerites which contain a much higher proportion of lime (see Table IV.). An unusual feature is Uie excess of alumina and the appearance of corundum in the norm — this excess of alumina is no doubt contained in the chlorite and its i)r(^sence may be due to slight contamination of the epidiorite by the aluminous slates which is has intruded. Pignrc i5. The basic inlnisives. A. Uralitised ({uartz dolerite, sliowing subopliitic relation of uralite and turbid ])lagioclase (dotted), plates of leucoxene after ilmenite and angular areas of cud piiase quartz with rods of apatite. B. Albite epidiorite, showing clear laths of albite in a ground of biotite, uralitic ainpliibole and tiirlnd epidotic material. The field relations of Ibis rock are sliown in figure 4. Consideration was given, on a pre\'ious page, to the contact metamorpliism effected by this sill on the (hirdup sediments, and, in the develoiunent of chloritoid, it was considei'ed that tiu' jiost Cardup earth movements took jilaee shortly after the intrusion of the chlorite-alhite epidiorite. This rock then is a low grade ('pidiorite showing a grade of metnmorphism com})aral)le with that of the associated sediments. The (|uartz-dol(*rite group (to be described presently) appi'ars to belong to a lnt(‘r perital of intnxsion, as a niicro- porphyritie epidiorite (desci'ibed under (iii) below) which is considered to be genetically related to the (piartz-dolerites is intrusive into tlu‘ chlorite-alhite epidiorite'. (ii) Uralilised qi((irtz-(loIentes. bocks ot this grou}) form the bulk of the post Cardu}) infrusions and all are v(*ry unitoriu in character. Variations in granularity dependent on the position in the dyke are* noticeable but the mineralogical composition remains eoiistnnt with the exception that the end phase quartz and micro- pegmatite are somewhat more abundant in the central ])arts of the dykes than on the edges. The Contact Between the Granitic Rocks and the Cardup Series AT Armadale. 45- Tlie characteristic ophitic lo sub-ophitic texture of the quartz-dolerites is always developed, although somewhat obscured by uralitisation and epi- dotisation (ftgure 5A). A typical s])eciinen from the large dyke striking south-east from the Roads Board quarry is a greenish, medium even grained rock, which consists of pale green uralite, epidote, zoisite, leiicoxene, quartz and minor accessories including a little brown hornblende, biotite and apatite. A relict ophitic relation is visible between the uralitie amphibole plates and the zoisitic aggregates replacing plagioelase prisms, the only other notice- able textural feature being the angular patches of (piartz and micropegmatite (in which the felspar is represeiitetl by turbid zoisitic aggregates). The uralite is the most abundant constituent, occurring in plates up to 1 mm. diameter. It is a pale greenish tibrous variety with weak pleo- chroism, X very pale greenish (almost colourless), Y pale green, Z pale green, absorption X Y = Z, extinction Z /\ e = 16° and (— ) 2V larae. It is often dusted with magnetite inclusions. Xo relict pyroxene is visible, nor has any been noted in any of the basic intrusive rocks of this area. In the vicinity of the quartz and micropegmatite areas the uralite shows a further change to pale green, weakly pleochroic, optically positive chlorite and more rarely to brownish hornblende. The plagioelase is represented by granular aggregates of zoisite, epidote and line granular plagioelase. Ilmenite has been replaced entirely by fine granular leucoxene which occurs in areas up to 1^^ mm. diameter in which the skeletal structure of the original ilmenite is preserved. Occa- sionally the leucoxene is recrystallised to sphene, which occurs in rounded grains with central iron ore inclusions. The quartz and micropegmatite areas in this rock, up to V 2 Rhu. dia- meter, are similar to those of the normal quartz-dolerites except that the felspar is replaced by fine granular zoisite. These quartz and micropegma- tite areas are jxmetrated by slender prisms of a]iatite. In specimens from the centre of this same dyke the micropegmatite areas vary in size to iy 2 mm. diameter and the presence of brownish hornblende in its vicinity is more noticeable. 1. Table IY. 2. 3. 4. SiO^ ... 48-83 49-13 40-22 50 • 52 Al-A ... ... 13-46 13-13 12-62 13-76 Fe,A ... 2-88 3-65 3-16 3-87 FeO 10-29 8-95 11-09 8-50 MgO ... 8-03 7-64 6-42 5-42 CaO ... 11*95 H-84 10-59 9-09 XaoO ... 1*32 1-72 1-86 2-42 K.>6 0-50 0-16 0-30 0-96 H..O+ ... 1-39 1-72 2*24 1-51 H.>0— ... 0*08 0-04 0-12 0-76 CO., • • • • « • • • • Nil 0-58 TiO", 1*24 1-27 2-00 2-39 P.,0, ... 0-07 014 0-09 0*26 MnO ... 0*18 0-15 0-33 0-16 BaO Nil Nil 0-03 FeS., Tr. 0-45 0-22 0-06 (S) Cr.,0., ... 0-08 ... Tr. Tr. v.;o, ... ... Tr. 0-05 100-30 99-99 100-26 100-34 Analyst : R. T. Prider. R. T. Prider _ 46 Rex T. Prider. Table TV. — continued. 1 . 2 . 3. 4. Norms. Q Or Ab An di hy 1-86 3-48 4-14 2-78 1-11 1-78 11-00 14-15 15-74 29-19 27-52 25-18 23-86 24-88 22-15 23-07 18-60 20-12 mt cm il ap py Tralitised 4-18 0-22 2-28 0-34 5-34 2-43 0-34 0*45 4-58 3-80 0-34 0-22 Board quarry, Armadale. (Prider, 1938, p. 95.) (Clarke qiiart'/.-dolerite (20464), near Roads Fralitiped quartz doierite, Toodyay, Western Australia. Epidiorite (uralilised (piartz doierite), Bickley, Western Australia. and Williams, 1926, \). 173). Quartz doierite (average of 6 analyses), Wdun Sill, Northern England. (Holmes and Harwood, 1928, p. 530.) These rocks are therefore best described as eomj.detely uralitised qnartz- dolerites. An analysis of the typical specimen described above is given in Table W. wlu're it is compared with other nralitised qnartz-dolerites from Western Australia and with the average composition of the Whin Sill. It will be seen that the analyses of the throe Western Australian (luartz dolerites are very simihuy So far as i>s knoAvn, these roeks all belong to the same period and are all intrusive into the granitic rocks of the Darling- Range. In comparison with the normal fiuart/ dolerites (as exemplified by the Whin Sill roeks), they are somewhat richer in iron, lime and mag- nesia and coiTospondingly slightly poorer in silica and soda. (iii) Mlcroporpliyritic epidiorite. The occurrence of this rock as a narrow intimsion into the chloritc- albite epidiorite has been noted above. The rock is greenish and extremely fine grained and has a mici'oporphyritic texture with equidimensional pheno- crysts of uralite (? after pyroxene) u\) to Vo mm. diameter (often aggre- gated to form glomerophenocrysts), and tiny plagioclase laths to Yz mm. long, in a fine grained ground of plagioclase, epidote, /oisite, uralite and pale brown i.sh hiotite. The fibrous uralite of the ],)henocrysts is very pale in colour with similar pleochroism to that in the (piartz dolerites, and it shows i>alc,hy alteration to hiotite. No original i)yrox(uie is visible. This type appeal's to be closely i-elated to llie quartz dolerites, of which it is a fine grained p(>r[)hyritic re|>rescntative. (iv ) Biotite epidioriten. This ty|)e of rock has lu'en noted in two places only- — as a nai’row dyke (four feet wide) intrusive into the hybridised gneiss in the Roads Board quarry ami as a. narrow dyke (10 inches wide) intrusive into the slates in the Slate (piarry. The occurrence in the Roads T3oard quarry appears to be an offshoot I'rom the larger nralitised quartz doierite to the south-east (descril)ed above) but this is by no means certain. It is definitely intrusive into the liyl)i‘id gneiss, for it truncates the banding sharply and cannot be one of the basic xunioliths (which it resembles minoralogically). Tt is a due even grained gia^enish rock showing little, if any, sign of schistose structure. ruder the microsco]ie it is seen to consist of a fine allot riomorphic granular aggregate of brownish-green biotite, plagioclase, epidote (and zoisite) and pale green amphibole with accessory leucoxene (after small ilimmite grains) and (piartz in small angular grains. The Contact Between tfte Granitic Rocks and the Cardup Series AT Armadale. 47 The biolite is a brownish tureen variety witli pleochroism X pale yellow- green, Y = Z deei) brownish-green, and absorption X Y := Z. The axial angle is very small (almost nniaxial) and j8 ~ 1.628 ± .002. In parts of the rock it oeeurs in clotted aggregates with a decussate structure, appearing to be either xenolithic or replacing ferromagnesian phenocrysts. The ani]dhbo]e is a pale greenish, slightly pleoehroie uralite with ,Z /\ c — 16°. The plagioclase has mostly been replaced by granular epidote and zoisite, but rare small laths of iintwinned albitic plagioclase remain. This rock, in its high biotite content, differs considerably from the uralitised (luartz dolerites of the area. An examination of specimens col- lected from across the dyke showed that the edges are very rich in biotite, which is present to the exclusion of the amphibole, the rock consisting of biotite, epidote (and zoisite) and minor amounts of Ieucox(‘ne, albite and quartz. The biotite here is in flakes to 0.25 mm. diameter, which is large compared with Oakes in the central parts of the dyke, and is often aggre- gated into clots. At three inches in from the edge the structure i>s the same with clotted biotites in a ground of biotite, epidote, zoisite and amphibole with accessory leucoxene and quai’tz. In the centre of the dyke the grain is slightly coai-ser and the main ferromagnesian is the pale bluish-green amphibole which is in excess of the brown-green biotite. Angular grains of quartz are also more abundant. The mineralogical variation throughout the dyke may be shoAvn by the following estimates of the mineralogical com- position of the different rocks examined (figures quoted are volume- percentages) : — Edge of dyke. Biotite ... ... 65 Amphibole ... ... Nil Epidote zoisite ... 30 Leucoxene ... ... 1 Albite -j- quartz ... 4 3in. in from edge. Centre. 50 22 15 40 3(^ 30 1 1 4 7 An analysis of the siieeimen from 3 inches from the edge of the dyke yielded the result shown in Table V. It will be seen from this table that chemically this rock appears ti) be more closely relate ... 0-03 NiJ 0-08 Fe«.> n.d. 0-14 n.d. 100-59 100-26 100-30 Analyst : R. T. Prider. R. T. Prider. R. T. Prider. 1, Biotite epidiorite (20188) three iiudies in from edge of dyke, Roads Board quarry, Armadale. 2. Chlorite-albito epidiorite, Armadale. (Quoted from Table III.) 3. Pralitised <|uartz-d()lerite, Armadale. (Quoted from Table IV.) of am])hibo!(' I’rom the edget ^ of this dyke seems to indicate that it has been entirely r(q>laced by biotitt and it seems most itrobable that end phase l>otassic solutions have been active in these narrow dykes There is no defijiite evideiU'c to say to which of the other type.s of basic intrusive this type is related but the chemical data indicate that it is most pro])ablv related to the chlorite-albite epidiorite. The ro('k from the narrow dyke in the slate quarry is very similar to that occurring on the edges of the dyke described above with the exception that it is soTiunvhat sc'histoso and very rich in biotite. (F ). — The quartz veim^. In the g’ranitie f^aioisses there are two sets of (luartz veins as described in the first section of this paper. So far as can be seen by microscopical examination these veins are similar — they apjieai’ to be replacement bodies in slH*ar zones in the ^’Tanitie i‘o('ks, similar to those notcal in the Darlington Area (CHarke and Williams, 192(), p. 174), Specimens from the north-west striking veins dOO yards south-west from the Narrogin Inn, show that in ])laces it is a (piartz seidcite rock whieh in places has small ironstained cubic ca\'ities i‘rom which ]>yrite has weathered. Om* si)ecim0 ... 0-09 1-03 0-06 0-76 KoO... ... 10-25 10-17 6-79 5-98 H.,0+ ... 4-42 "Loss on 2-93 2-12 H.O~ ... 0-06 f ^ ignition. 004 0-16 Cbo ... ... 0-03 • » • 0-02 0-05 Tib.2 ... 0-85 0-33 0-56 0-35 Poo; ... Nil' • • . Nil 0-08 MnO ... 0-12 • • • 0-07 0-03 BaO ... 1-01 . . . 0-67 0-11 Cr,0, ... 0*02 0-02 0-01 VoO, ... 0-13 « • « 0-08 0-03 SO, ... 0-07 005 O-ll 0-05 (Graphite) 100-05 100 14 100-00 99-85 Analyst H. P. Rowledge. C’. R. Le IMesurier. 1. Sericite schist (19481), (dark patches xenolithic in quartz vein, from which all vein quartz has been removed), near Bedfordale Road, Armadale, Anal. IT. P, Rowledge. 2. Muscovite, Bamle, Norway. {Amer. Jour. 8c., vol. 24^ p. 259, 1959). 3. Analysis 1 with 8iO^ made up to 65.22% and the remaining oxides recalculated to sum to 100. 4. White shale, Armadale, Western Australia (quoted from Table II.). that it is very siinilai* to the analysis of muscovite. A noticeable feature of the analysis is the high baryta content, and the presence of very little SO 3 indicates that it is not in the foian of barite but must l>e in the mica.. No barite was visible in the thin sections examined. Floaters of this sericitic material can be traced for some distance in a northerly direction and this appears to be a shear zon(‘ in the granitic rocks along which I'eplacement quartz veins have been formed. The other possibility is that (his band of sericitic material is an infolded portion of the Cardu]i slatt^s along which quartz has been introduced. Comparing the analysis of the sericitic material with that of the white slate (table VI, col. 4) the main diftereiice is seen in the amount of SiO.. Jf silica is added to the sericite schist to bring it to 65.22% and the remaining oxides recalculated to sum to 100 (col. 3) there is seen to be a close agreement, the mai?i differences being in llu' higher FeO and MgO in the slate (due to chlorite), the different pi'oportions of the alkalies and the higher BaO content of the sericite schist (this seems to be a significant feature as the BaO in the sericite schist must he in the mica— in this connection it is interesting to note the comparatively high BaO content of the hybridised gneiss in tabh* I.). It must be noted that the products of sericitisation of alkali felspar are similar to those accumulating in certain fine grained argillaceous sediments (especially illitic clays) so that a comparison of these analyses, wdiich are both of highly sericitic rocks, does not convey any real information regarding the origin of the sericite 50 Kex T. Pridek. seJiist. The prosciu-e of similar tjuartz-scririlo rocks associated with the north-west system of (juartz veins, whi('h in view of their disposition in the field, cannot be regarded as infolded poi'tions of the Cardup slates, would ai)pear to indicate that botli gr()iu)s ol veins arc* of similar origin, i.e., replacements along shear zones. This is supt)orted by the absence in the- vicinity of the <]uartz blow at the south end of the aiea of other remnants of tlie (hirdLi]> seric's, for if the slates were infolded then a considerable ])ortion of the more ar-fmaceous sediments underlying them must also hav(' been infolded— no (race of these remains and the evidence is oxa'rwhelmingly against the possibility that this band represents nxi infohh'd portion of the (hirdiip slates. A similar (juartz vein in the granitic gneiss near the contact with the sediments of tlie (lardup S('ries has been noted near Kelms(mtt. This occur- rence is exposiul in a small road cutting on a rone one of the basic intriisives. This o('(nirrcnec is inteivsting sinc(‘ it throws some light on tli(‘ origin of th(‘ harite-tluor veins in the (hirdui> Series at Cardup (Clarke, 1930, map on i>age 199) — it indi('ates that the barite vOns are later than the basic intrusives and were ihend’or*' i)rohahIy derived from the greenstone magma rather than tiie granitt* magma. It is interesting to note here that Sweet (1930, p. 258) considers that the barite ami fluor bearing veins of the iiorlli of Bngland arc* genetically related to tin* (juariz dolerites of the Whin Sill. On the south siorphyrite of Esson) which is similar to tlu' alhitc cpidiorilc from Armadale. The suggestion is put foiuvard Inwe that these silver-l(*ad veins and barite-lluor veins are genetically related to the liasic magma ratlu'r than to the granitic- magma. TV. CONCLUSIONS. (a) The age of the (lardup Series. The evidence* in'(*scnt(‘d by tlie (piartz veins (both in tlu'ir tield relations and petrology) indicat(*s that the* granitic* rocks are o]d(*r than the Cardup sediments. This conclusion is sup])orted hy a number of otlic'r facts — (i) that the Cardup Series is a normal erosion seciuenc'e and is constant in character along the strike; (ii) that nowhere have the* “.uratiil ising'^ solu- tions been seen to traverses the basal f)cds of tlie seri(*s nor have any aimphyscs (sncdi as p(*gmatites or (|uartz veins) heem seen to jiass into the sedimc'nts; (iii) that tliere are no remnants (x(*nolitlis) of the sc'dimentary series in the gneisses, the only rc^mnants being of oldc*r basic igneous rocks; (iv) that no variation in character of the gneiss across the strike, sueh as would be expected if a series of varying lithology were granitisc'd, has h(*(*n noted, and (v) the slight eontacd metamorphic cdleGs noted in the Cardup Series ai’c* due to the* basic- intrusions. Indeed, the only evidence that has Hex T. Prider. found which in any way favours a pro-gTanite age for the Cardup Series, lies in the pi'estuna* in th(‘ slates of small amounts of idiohlastic tourmalijie, which as described al)ov<*, may equally well be ex])Iained as due to crystallisation of components of the original sediment during the earth movements whi('h tilted the Cardui> Series. The position of the Cardup Series in the Pre-Cambrian succession of Western Australia (as deduced from all the evidence* available in this area) is therefort* later than the tinal granite intrusions and earlier than the basic igneous inti’usions of the Xullagine period (late ifre-Cambi'ian ). I have res, may he summarised as follow's: — 1. Period of granitisation during w'hich ])re-existing basic I'ocks were permeated by granitic e*manalions and the hyl)rid gneisses pro- duced. 2. Perioel e)f granite intrusion (aplogranite), d. Eai'th movements, causing development of shear zones and joint ]>attern in the granitic gneisses, follow'ed by the formation of the quartz veins in the gneisses. 4. Deposition of the (birdu]) s(*diments. 5. Formation of an (*xteiisive downwai’p parallel to the present Darling Scar}.) and some distance to Ihe west, involving upthrusting along the eastern margin, development of fracture cleavage in the slates and tilting of the sediments to the west. 6. Intrusion of alhite epidiorite sill into Cardup Series, probably con- temporaneous w'ith (5). The Coxtact Between the Granitic Rocks and the Cardui^ Series AT Armadale. 53 7. Intrusion of the quartz dolerite series of dykes consequent upon earth movements affecting the Cardup Series. 8. Formation of barite veins as end phase effects of the basic intrusions (galena-sphalerite-quartz veins also probably belong to this period). End of Pre-Cambrian. 9. DoAvnwarping continuing and sediments being constantly deposited in the depression — the only evidence regarding this long period of geological time being the two thousand feet (at least) of Cainozoie sediments which underlie the coastal plain and cover up all inter- vening formations. Block faulting may have taken place (as sug- gested by Jutson, 1935, p. 469) during the uplift of the Western Australian plateau in late Miocene times but differential erosion of hard rocks of the plateau and the soft rocks to the west of the scarji seems capable of explaining the present physiographic features. V. ACKNOWLEDGMENTS. T desire to express my thanks to Mr. H. A. Ellis who suggested that the possibility of granitisation of the Cardup Series should be investigated and who was instrumental in having analyses of a number of the rocks made in the Government Chemical Laboratory. My thanks are due also to Mr. H. Bowley, Government Analyst and Mineralogist, for permission to publish these analyses. Chemical analyses made by myself in the Department of Geology were made possible by funds from the Commonwealth Research Grant and this assistance is gratefully acknowledged. I also desire to thank the Senior Students of the Lhn'versity Geology Department (Messrs. Crab- tree, Davis, Elvish, O’Driscoll, Telford and Thomson) who, under my super- vision, made the survey of the area as part of their field course. Finally I wish to thank Professor E. de C. Clarke for assistance during the revision of the text. VI. LIST OF REFERENCES. Agrell, S. O., 193It: ‘‘The Adinoles of Dinas Head, Cornwall,” Min. Mag. XXV., pp. 305-337, 1939. y . Campbell, W. D., 1910: “Report on an excursion to Armadale,” Jour. Nat. Hist. oncH of Jfest. Austr.^ Vlcxl. III., No. 1, pp. 28-29. Clarke, E. de C. — 1926: “The Geology and Physiography of the Neighbourhood of Perth, Western Australia.” A.A.A.Sc. Handbook for Pertli Meeting, 1926, pp. 23-30. 1930: “The Pre-Cambrian Succession in some parts of Western Australia ” Kept. A.N.Z.A.A.Sc. for 1930, pp. 155-192. and Williams, F. A., 1926: “The Geology and Physiograjihy of jmrts of the Darling Range, near Perth,” Jour. Hoy. Soe. W.A., XII., pp. 161-178. Ellis H, A., 1939: “The Geology of the Yilgarn Goldfield, South of the Great Eastern Railway,” Geol. Surv. W.A. Bull 97. Essoj^ A. G. D., 1927: “The silver-lead deposits at Muudijong,” Oeol. Stirv. W.A. Ann. Bept. for 1926, pp. 4-9. Forinan, E G., 1937: “A Contribution to our Knowdedge of the Pre-Cambrian Simcession in Some Parts of Western Australia,” Jour. Hoy. Soc. W.A., XXIII., pp. xvii-xxvii. Goldschmidt, V. M., and Peters, CL, 1932: “Znr Cheiir'e dos Hors IT,” Nachnohten von der Gesellschaft der Wissen, Mat. Phv.S., 1912: ‘ * The Extension of the Kelmscott Clay Deposit,’^ Geol. HnriK W.A. Bull. No. 48, pp. 63-65. Jutson, J. T. — 1934: “The Physiogra])hy (Geomorphology) of Western Australia,’’ GeoL. Surv. W.A. Bull 95. 1935: Report of the Australian and New Zealand Assoc. Adv. Sc., Vol. 22 (l^lelbourne, 1935). MacGregor, M,, and Wilson, G., 1939: ^'On Granitisation and associated pro- cesses, ’ ’ Geol. Mag., Vol. 76, pp. 193-215. Miles, Tv. R.. 1938: “The Geology and Physiogra^pliy of the Lower Chittering Area,” Jour. Boy. Soo. Vol. XXIV., pp. 13-41. Prider, R. T. — 193'4: “ Tlie Geology and Pliysiography of the Jiinperding Area,” Jour. Hoy. Hoc. W,A., Vol. XX^, pp. 1-16. 1938: “The Petrology of part of the Toodyay District, W'^estern Australia.”' Ilnjiublished dissertation for the Ph'.D., Cambridge, 1938. 1939: “The Petrology of part of the Toodyay District, Western Australia,” Ah.nt. Bins. TJniv. Canihridge for 1937-8, p. 93. Read, II. H., Pliemister, J., and Ross, G., 1926: “The Geology of Strath Oykell and Lower Lo(di Shin,” Geol. Hurv. Scot. Explanation of sheet 102. Saint-Smith, E. C., 1912: “A Geological Reconnaissance of a portion of the South-West Division of Western Australia,” Geol. Surv. West. Aust. Btdl. 44. Simpson, E. S., 1931: “Contributions to the Mineralogy of Western Australia,, Series VT.,” Jour. Hoy. Hoc, W.A., Vol. XVII, pp. L37-149. Sweet, J. M., 1930: “Notes on British Barytes,” Min. Mag., Vol. 22, pp. 257-270.- Wiseman, J. D. H., 1934: “ Tlie Central and South-West Highlands Epidiorites: a Study in Progressive Metamorphism,” Quart. Jour. Geol. Hoc., Vol. xc.,. pp. 354-417. EXPLANATION OP PLATE 2. Pigure 1 — - Hybrid gneiss from tlie Roads Board quarry, Armadale, shorving xenolithic character of the gneiss. The xenolitlis are of biotite-hornblende-epidote liornfels (dark coloured') and epidotic material (light coloured). The banded character of some of this gneiss is seen in the top left hand side of the boulder. Quartz xeuoliths are present (the coin is lying on one) but they are indistinct in the photo. The clinometer gives the scale. Pigure 2. — Pliotomicrographs of quartz veins. A. Sericitic rock from shear zone in gneiss, with secondary veinlets of quartz (from large quartz blow near the Bedfordale Road). Ordinary light x 40. B. Strained quartz from the same locality as (A). The entire field (and almost ad of the slide from which this photo was taken) is part of the same crystal. The strain pattern is interesting, consisting of subparallel lines along which actual granulation has taken place arranged obliquely to the direction of “slicing” or gliding shown by the strain shadows. Nicols crossed x 40. C. “Vein” quartz from the slate quarry. Occurs as masses in the slates. Shows unstrained character and even granular mosaic structure. Nicols crossed x 40. TiTK COiVTACT I^byrWEKX THE (iKAMTIC K0CK8 AND THE ('’/AIH>UE SeIUES AT Aemahale. o 5 Figure 2 Further Permian CokaIaS fro.m Western Australia. o7 4.--FURTHEH PERMIAN CORALS FROM WESTERN AUSTRALIA. Bv Dorothy Hill, M.Hl*., Ph.D. Communicated by (’urt Teicliert, Pli. 1). Read 13th August, 1940; Published Otli FobruRiy, 1942. The corals described in this paper are from the Cynthaxonia faunas in the Permian of \\'estern Australia. Tliese faunas are : — :NORTH-WE8T.i KENNEDY SERIES. WAN I ) AGEE SE KIES [Jnoprodurtus stage: ? I^junjphyllum reidi Hill Fai'OMites^ sp. Thamnojtom iinnien.^a Hill Thant Hopora a IT. wnnttioiwmvs (Etheridge) Im m el I i brati ch .s t agi ‘ . ( 'alceoVispomjhi stage AUotriophyduni. sj). EuryphylUott reidi Hill ? E/U'nphyfhfni. sp. Verbeekiella mersa, sp. nov. ... Thant no pora aff. niannionen^^iH (Etlieridgt') CTNDI.EGO SEIBES:— Jderophylintn, .sp. Verbeekiella mersa, sp. aoy lil’LGADOO SERIES. WOOKA.MEL SERIES. CAlJATHAlGiA SERIES:-- “ A}nple.rif>i " pustulosiis Hudle.stca EtoyphyUn in m i nnfnm, Hill. PleropJtyUnnt anstrale Hinde. l'achyia.snia, .sp. \'erl)eeki(d{a (albofi (llosking) ... Thamnopora insculpta, sp. nov ? Thantnopora. sp. Cladochonus strialus, s]). nov. LYONS SERIES (G LAUI .A LS). Page ()1 ()4 ().') ()() ()0 (U (>2 93 ()() (y2 ()3 ()4 97 9S ?(► KIMIH]KLEY.^ ERSKINE SERIES. LIVERINGA SERIES Tachyiastna denstini Hill. NO( )N< AN BA 1 1 SE RI ES Thaninnpora ntannioneihsi'S { Et lu‘ri(lg(9. ( 'ladorho/tuM nirhuLsoui (Etheridge) ... ... ... ... ... ... (jp {Fafreoli’^ponyia occurs with the* two latter). TIia.n>nopora iinntenm Hill fumi Uhristmas Ok. homcst(^ad niay hav(^ come from this series. NORA NORA SERIES. POOLE RANGE OK GLAOIAJ. SERIES. ‘ The succession liere given is tiuit given l)y Teicliert (U)39, ]). h) and i>ersonal coninnmication. * The succession given here is alter Wade in Clarke (J038, ]). i 140/41. 58 Dorothy Hill. IRWIN RIVER.^ Page UNFOSSILIFEROUS WHITE SHALES AND SANDSTONES. UPPER MARINE SERIES. COAL MEASURES. FOSSIL (TJFF BEDS:— “ Am.plexu^,'" sp. Kmi/phylium irizimalnm Hill. Oerthia .wlraia (Hind(“). Flerophylknn australe llinde. SHALES WITH LIMESTONFS WITH Metalegoceras. GLACIAL BEDS. Species to whicli no page reference has been given in the above lists have already been described (Hill, 1937a). The Cyathaxonia Fauna of the Palaeozoic consists of small, solitary Rugosa without dissepiments, and of Cladochonids, F'avositids, and Palaeacids. It ap|)ear,s very important in the Artinskian of the Urals and Timor, and the Middle J’ermian of TiiiKjr, and it is the only coral fauna known from the Kamilanji Series (mostly I^ermian, but j)ossi.bly in part Uralian) of Australia. Its occurrence in the Loner Carboniferous has already been summarised (Hudson, 1935 ; Ifill, 1938a, p. 5). Species belonging to it have recently (Dobrolyiibova, 1939) been scovian and Uralian of Russia, so that its continuous existeic'e from the Lower Tournaisian to the Middle Permiaji is proved. F]viecies from both tlie Wandagee Stage and the Callytharra Stage, however, are comparable with Aitlnskian or Middle Permian- Basleo species, ratiier tlian vvilli Uralian or Uj)])er Permian species, and Eury- phylluiii’ is so far known, outside Australia, only in Wtinskian beds. In the absence of species ct)mmon to the Western Anstralian faunas and the 'Timor or Russian faunas, cannot at present indicate the ])recise ago of our faiinas. Two S])ecies from tlie W'antlagee Stage, Euryphyllurn reidi (Ifill) and Titanmo- para immeytsa- (Flill), are associated at Uastle. Creek, Tlicijdorc, Queensland, at an unknown horizon in the Bowen se([uence. Cladochoniis nicholsoni (F^theridge) from the Nooncani^ah series, occurs in the Condamine F'^ault Block and in the Coral Stage (,)f the Bowmen Alarino Series in the Springsiire District, Queensland. ZOANTKAKIA MADKEPORARIA RUGOSA AMPl.EXIMORPHS. Am])l6ximorphs ; Hill, 194(1, p. These simjde, (*.ylindrical, or fast-iculate Rugose corals with short, e([ual lamellar se])ta, and complete tahiilae, and without dissepiments, are common in the Silurian, Devonian, Carboniferous, and Permian, and ]:)robably represent the endiJoints of trends of simplification in many stocks. In the Permian of =* Succesdon after Tek-iiert (1939, p. G), and Clarke (I93S. p. 429). Further Permiax Corals fro:m Western Australia. 59 Timor a group occurs which differs from the earlier ampleximorph.s in ha^dng dilated septa usually swollen a little axially, and continuous vertically thruj*y — Middalya road,, W^A., is probably a member of tliis species, although it is curved rather than erect, and cylindrical rather than turbinate. It is flattened in the plane of the cardinal septum, j»robahl 3 ^ by crushing, as the ta])ulae seen in a \^ertical section are somewhat shattered. There are 33 septa at a diameter of 15mm. x 1 8mm., extending a little over liaif way to the axis, sub-etjual, some with sJiarp axial edges, and some with swollen axial edges, all dilated so as to be almost in contact laterally. The cardinal septum is shorten’ than the rest and is on the longest side of the eorallum. No minor septa are developed. The tabulae are complete, shallow domes. Etlieridge has ah’eady mentioned tins species fix^m W illiambury Station, Minilya Hiver. “ Amplexus ” s]). (Plate II., Hg. 1.) Material : One specimen from the Permian of Fossil Cliff, Irwin River, in the collection of the UniversiU^ of Western Australia. Description : Tlie specimen is an obIi.|uelv broken fragment (iO mm long, and 32 mm in diameter. There are alKuit 59 sligiitly dilated major se|)ta, continuou.s vertically for about 5 mm. from the ei)it liecu. and then extending for a short distance over the Uf)per flat siadaces 4, pi. i, figs. 22a, 22b, 23. Chihsia Limestone Chihsiashan, Central Cliina (= Artinskian). IJifignosis : Simple Rugose corals, typically curved, and bearing scattered spines. The septa of the counter (juadrants and sometimes the alar and first meta-se])ta are grouped in a narrow crescentic area, embracing the counter side of tile (‘orallum ; the remaining septa of the cardinal ijuadrants are directed towards a point or points on its inner side, which is roughly midway between the nxis of the corallum and the eyntheca. The septa I^ecome amfdex- •oid in late stages. Tabulae are usually far apart, com])lete and obIi(jue, with a downtiirned iiordor of the same width as the crescent ; they sIo]:)e down- wards from the convex to the concave side of the corallum. There are no disse])iments, and Tuinor se[)ta are develofied very late or not at all. Retuftrh's : The genus is known in the Tournaisian of Belgium, the Dinant- ian and Lower Xamurian of Scotland, the Artinskian Chihsia limestone of China, and the U])|)er Permian of Ujoulfa in Armenia, Possible synonyms of the genus are discussed in tiie authors " (^arlioniferous Rugosa of Scotland, Part III.’ , in course of [uiblication by the Palaeontfigraphical Socletv of -London. ALLOTROPIOPHYLLUM, sp. (l>latc 1., fig. 2.) Materud : One specimen in tlie collection of tlie University of Western Australia, from the lowest horizon with large sjjecies of CUdceolispongUp mid of syncline west of Coolkilya Pool, Minilya River. Description : Tlie specimen is broadly trochoid, with a talon, and a very 0; 1937, a, ]>. 50; 1938, [). 25. Genotype : EiiryphyUnm reirU Hill, 1937, ]». 150 ; 1937, a, p. 50 ; 1938, p. 25 ; Permian (? Artinskian) Htiwen Marine series of Queensland. Diagnosis : Siin|>le, turbinate to eeratoid Rugose corals, erect excej)t at the tip, which is turned aside ; with woll-inarkod interseptaJ ridges, and ty|)ically an obii(|ue calical floor. The major se|)ta, which arc never carinate or seiTate, exttaid to the axis and are pinnately grouped about a long, closed fossida, bisected by a long canlinal septuui on tlio concave side of the cornl- lurn ; alar fossulae are [H'esent- Tijo septa are dilated, and at hrs-t arc laterally contiguous throughout, l>ut iluring ontogeny dilatation decreases in a widen- ing zone midway betw(>en tlie j^eriphery and the axis, leaving a w ide peripheral stereozone, and an axial structviro hu’med by the conjoined specimens from the Oleio- thyridina-Calceolispongia liorizon, all from the up 2 :)(U' CalceoHspongia stage, west limb of the same syncline. The specimens on the w hole arc rather larger than those from the Springsure ari^a, but thosii from the first named horizon are identical in size with Queensland S])oeimcns From Castk^ ('rct^k, Th(>odore. The species may prove divisible. I havti not yet mafle a tietailed study of zaphrentimorphs from the Upper Marine of Now South Wales, but from Etheridge’s figures none aj^ipears to be K. reidi^ tliough Z. grcgoriana is possibly a member of the genus. Thirteen sjjecimens from the highest CalceoHspongia horizon, at th(^ toj> of the Wandagee series, are identified as Eiiryph.ylluni reidi Hill. They are of the morphology of that sub-group from Castle Creek, Queensland, and from the locality 3 3, in tlio middle or ii 2 )])er parts oF the CalceoHspongia zone in the Wandagee Series, west or south-w'est of ('oolkliya Pool, Minilya Riv^er. The localities of these 13 thirteen specimens are : — south-w'est side of Wandagei^ Hill, neai- montli of South-West Creek ; near south-east corner of Wandagee Hdl, at crossing of Wandagee Wool- shed — Middalya road over highest CalceoHspongia horizon ; north-east corner of Wandagee Hill. Dorothy Hill. 62 Two spoeimens from the uppermost Pseudogastrioceras horizon of Lino- jyroduclus-Fenestella stage of Wandagee series, are identified as Euryphyllum reidi Hill, but they may represent a variant from the morphology characteristic of the Calceolispongia zone, for their septa are without any curvature. Never- theless, a similar morphology is known in specimens from Castle Creek, Queens- land. ? EURYPHYLLUM, sp. (Plato k, fig. 4.) Alatervd : One specimen from locality H 4 E, in the Collection of the University of Western Australia from the lowest horizon with a species of large Ccdceolispongia, middle Calceolispongia stage, east limb of syncline west of Coolkilya Pool, Minitya River. Desmpl/ori : The specimen is small and troclioid, the diameter at the upper edge. of the calice being 18 mm, with a deep calice and strong longitudinal ribbing. (")no side of the corallum, about an alar fossula, is excavate, but whetlior from injury or attaclmient it is impossible to say. Major septa only are developed, aufl they are considerably y tlu> outer edges of t(ie laU^ral laiiioJIao ; it is always joined to th(^ eoimter S(5j)tuiu, hut is not a eontiniiation of the countt.ir 80 ])tinn. Tlio tabulae are thin, coinplot(i and in(jonij)lote, arid domed. No diss('ijimf‘nts are dtweloped. Re'rns s}ioul{l be placf^d in Lopho- phylhi’iH- or JA)p}tophijll/ie Series) of tlu^ North-West, chiefly in the general abs(aice of spaens in its axial stT-uctur(> ans not only iii those larger dimoiisious, but also in having Jieai’ the (UiJice a clisiophylloid i*ather than a dibunophylloid sec- tion of tlie axial column, hisb^ad of there being a maximum of nine laTnamai-k, ISH), ]>. 204 ; for references, etc., see .lones, 193(3, p. 2, Gen,olectoti/pe (see Fdwards and ITaime, 1850, p. Ix) .■ F. gothlandicus Lamarck, 181(3, p. 20(3, Silurian, Gotland. Diagnosis : Cerioid tabulate c^oiuls forming massive or ramose colonies, in wliicli the contiguous corallites liavo thin or moderately thin walls, spiiiose or ohsolc't(' septa, and complete or mainly complete horizontal tabulae. Further Pekmiax Corals fho.^i \V ester x A listraiaa. ()5 Remarlcs : Tho genus is widespread, from the Upper Ordovician to the end of tho Devonian ; forms like Favosites in the Carboniferous have usuaJly been referred to Emmonsia, but Gerth (1921) has des(*ribed as Favosites tliree Permian species from Mandco and tho Middle Permian of Basloo, Timor. FAVOSITES, sp. (Plato II., fig. 5.) Material : Two specimens from tlio base of the lAnoproduchis stage of the Wandageo series, centre of synclino west of Coolkilya Pool, Minilya River (one near Station 4A, t'.o other 12 chains from Station f of 1939 survey). Descri'ptioyi : The corallum is tuberose, 30 to 40 nmi. in diameter, and 50 mm, or more long. The individual corallit(>s diverge outwards from the axis, each being about 1*5 to 2mm. in diameter, though smaller ones occur. Owing to the ])rocesscs of fossilisation, lautlua* spi^cimen is wortli sectioning. Neither mural jioros nor septal spines wi're provi'd, but one surface suggests a single vortical row of pores per wall. Tabulae ari^ thin, slightly domed, and rather distant, 3 in 2 mm. Renmrks : The specimens ilo not appear to be conspi^eific with the species desci'ibed by Gertli from Timor, or with any other species known to me. Genus THAMNOPORA Stoiiiinger. Thamno'pora Steininger, 1831, p. 10 ; 1834, p. 337 ; for references, etc., see Kill, 1937 a, p. 56. Genoleciotype : Thanmopora madreporacea Stoininger Alveolites ccrvi- cornia de Blainville, 1830. Diagnosis : Ramose Tabulate corals in which the cylindrical branches may be flattened and coalesced ; the corallites are typi(?ally polygonal, they diverge from the axis of the branch and usually o]>(^n normal to the surface ; the corallite walls are dilated throughout, and llie dilatation increases dLstally ; typically the growth lamination in the scleronchyme of the wall is obvious, while its tibrous nature is not ; septal spim^ are usually obsolotix and mural pores are large. Range : Silurian, Devonian, Permian, and rare in the Trias. Thamnopora immensa Hill (Plate 1., fig. 8 ; Plate II., tig. 6). Tham.noporo. immensa Hill, 1937a, p. 58, pi. i,, tigs. 21, 22, text-fig. 9 ; Permian, two miles East of Christma.s (’reek Homestead, south of Rough Range, Kimberley. Holo'ype : H 25, Geological Survey of Western Australia. Diagnosis : Large Thamnopora, in which the corallites have calices of t\vo sizes opening at right angles to the surface of a brancli, excessively dilated walls, thin tabulae, anti nmnerous large, ngular mural pores, frecpiently further excavated by boring organisms. Remarks : Additional specimens, from Btiolgadoo Pool, Minilya River, and from Coolkilya Flat, south of the Minilya River, probably from th(> lower part of the Linoprodnctus stage of the Wandagee sei-ies, were sent by Dr. Teichert. They have allow^xl an expansion of tho diagnosis given in 1937. The specimen from Boolgadoo is nnweathered, and large corallites are seen between smaller ones on the calieal svirface. The calices show' eight or nine coarse radial ridges, somewhat as in Striatopora Hall ; the ridges are without 66 Dorothy Hill. spines. In all specimens the corallites open at right angles to the surface, thus differing from T. ^narmionensis, where they open obliquely. In the M 16 specimen, dilatation is not so excessive as in the others, and thin tabulae can be seen. Individual septal spines or trabeculae have not been dis- tinguished, though the sclerenchyme is fibrous. Etheridge may have included some specimens of T. iinmensa in his T. marmionensis, as his description of the latter covers coralla with corallites opening at right angles to the direc- tion of growth as well as those with corallites opening obliquely. The only two syntypes of T. marmionensis which I have seen, however, had their coral- lites opening obliquely. T. immensa or a very closely similar form occurs in association with Euryphyllum reidi at Castle Creek, Theodore, in an unknown horizon in the Queensland Bowen succession. Thamnopora marmionensis (Etheridge). Favosites marmionensis Etheridge, 1914, p, 13, pi. i., fig. 1 ; pi. ii., figs 2-4 ; pi. viii., fig. 2. Permian, Mt. Marmion, Kiml.»erley District, Western Australia. Diarpiosis : Thamnopora forming largo lobate masses, with corail ite of two sizes opening obliquely to the surface, calices fi*equently with lower lip semi-circular, and with corallite walls becoming very thick distally ; with large irregular poi‘os, frecpient tabulae, and without septal spines. Remarks : The species differs from the eastern Australian T. wilkinsoni (Etheridge) in being lobate rather than ramose, and in having the calices closer, while the projection of the calical rim is low compared with that of T. wilkinsoni. Thamnopora aff. marmionensis (Etheridge) (Plate II., fig. 7). Material : A fragment in limonito (g) from the lower part of the Lino- prodiictns stage somewhat east of the Heliocoprion locality of Coolkilya Flat ; only the calical surface can be studied, and tins shows corallites of two sizes, most of the openings have one-half of their outline semicircular, repre>senting the lower lip, which, however, does not project, and the upper half polygonal, two or three short edges meeting at angles. This specimen is probably T. marmionensis, but the internal structure is obscured. A small lobate fragment (i) from high in the Calceolispongia stage at Station 25A, “ Upper worm track ” horizon north of Minilya River, witli thin walls and oblique calicos, which is probably T. marniionensis, but the calices are not clearly distinguishable into two sizes, and vary between 1 mm. and 1 • 5 mm. A worn, lobate fragment (b) from the highest beds exposed north-west of Station 23, in the lower part of the Linoproductiis stage, Minilya River, \vhich is probably T. martnionensis, although the corallite walls are thicker than in the sj-mtypes, being 1 mm. nr nearly 1 mm. throughout, and the septa appear to have donticulations resembling spines on their inner edges. A cylindrical fragment (d) from the lower part of the Linoproductus stage near Station 4A, north of Minilya River, whose internal structure is similar to that of (b). A cylindrical fragment (c) from the same horizon at Station 4A, north of Minilya River, closely resembling (b) and (d). All these are from tlie Wandagee beds. Remarks : One cannot be certain that these specimens, which are prob- ably conspecific, are T. marmionensis, because of their fragmentary nature, and because our knowledge of syntypes is very limited ; but the small, oblique corallites suggest that they are. I have seen nothing similar from eastern Australia. Furthe({ Permian Corals from Western Australia. ()7 Thamnopora insculpta sp. uov. {Plate I., figs 9 a-d ; Plate II., figs. 8 and 9). Type material : Eleven fragments, possibly from one specimen, from the Callytharra beds near Callytharra Springs, W.A., collected by C. Toichert. Diagnosis : Slender l>ranching Thamnopora with the corallitos opening a little obliquely to the surface, dilatation of the walls increasing greatly towards the surface, so that the openings are distant, and sunken into the unridgod, faintly tuberculated wall tissue. Description : The diam(?t(‘r of the cylindrical branching fragments varies between 3 mm. and 10 mm. The calical oponings are unequal, the largest being P5 mm,, some being 1 rnm., and a few 0*5 mrn. in diameter. They are unetiually spaced, the smaller being in the angles b(‘tweon the larger, which tend to be arranged in voi’tical rows. The ojienings are occasionally surrounded by a raised rirn, especially n(!ar the growing tips of the branches, its height being gn^atest on the under side of the opening, but in the older branches there is usually only a very faint trace of this rim, and that on the xmder side of the opening. Potwixm tlui openings there is instead from (0 5 to 3 mm. of dense sclonaichymo, which is lightly tuborculate, and no trace ■of the junction between the walls of lu'ighbouring corallitos can be seen. Faint septal ridges are obse vahle in one or two calicos, twelve being countes are almost vertical in the axial parts of the brancJi, and very small, being from 0*1 to 0-5 ixim. in diameter, with but slightly thickened walls. Outside this axial part of diameter about 2 mrn., tlie corallib^s bend rather rapidly outwards, opening to the surface at k^ss than 30^ to the hori- zontal. As they proceed to the surfaeo they increase in diameter, l>ut the greatest j>art of this incn'ase is duo to an increase in the thickness of the walls, the lumen never becoming wider than 1*5 mm. The walls may bo very thick, from the ecies with lamellar septa should be placed in a separate genus with the eastern Australian Trackypora wilkinsoni Etheridge, to wliicli it is very close. T. wilkinsoni has the twelve lamellar septa much more clearly developed, and in it the ■outer boundaries of the individual corallites are distinguishable by ridges on the surface. T. xvilkUisoni occurs in the Upper Marine of Mulbring, and of Ellalong, N.S.W., in the Coral Stage of the Bow'on Marino of the Spring- sure Basin, Queensland, and in the Condamiiio Fault block, near Silvorwood, Queensland. I have not .seen it from Western Australia, hut T, insculpta appears to mo to bo a related form. Favosites permica Gerth (1921, p. 101, pi. cxlix., figs. 1, 2, 3, pi. cl., fig. 1) from the Basloo Beds, also has twelve septa, lamellar in the upper part of the calico, but this form is encrusting. Its septa consists of spines fused together at their bases, forming a series of stripes on tho calical floor, but with separate spine ends in the lumen. Twelve septa, sometimes acanthine, but frequently with the spines fused at their 08 Dorothy Hill. bases to form lamellar septa, in the form of stripes on the corallite walls,, appears to be the higlujst numb(U’ possible in PscMdofavosites sttjlijer var. septosa (tertli (1921, p. 104, f)I. cxlviii., tip^s. 7, 8) from Dasleo, and this num- ber is usually attaimsl. Dut this ^(nms is without tabulae. ? Thamnopora s]). (Plate I., fig. 10; Plate II., fig. 10). Ma/erif(l: One s})ecinion, c-olleeted by C. Teiehert, from Callytharra opri ngs. J)esGnpff.(m : The (*orallum is partly encrusting, extc^nsi orm but of crum])le(l growth, with stumpy branches on tlui upp(u* surface, which has been f>itfed by rain weatlua'ing. Its si/*e is fiO x 00 x 15 iutii., tlu‘- branches IxOng mostly bi-oken oil at a heighl of 5 to 10 mm., and Ix'ing alxntt 8 mm. in diaiTiietei’. 1 h(.> corallites an* uiHXjuat, up to 2 or 2*5 mm. in diauK'ter, 'Tlio walls of tlio corallitfss in the brancluis an; nnu-h dilatinl, but those in the basal layiu's and in tlie paits iit which the branches ar(> se4> are thin. The walls of th(‘- corallit(!S in tfn> ax<^s of tJu^ branches are also thin. Wluai ana- lysf (xich s<>pfinn are dilatixl ajid in contact so that laiiKOlar so])ta a,r(^ formed, fiom the axial (xlges of whicli denticulations may never- th(iless arise ; and tJie dilaflow. Diagnosis : l^ifurcating, free branchtis, 4*5 to 7 mm. thick, with success- ive corallilt's on o]i})osite sides of a branch, tlie diameter at the tnlge of tho ealiet' being constantly 4*5 mm. Deseri piioif- (of specimens from Halmaningarra) : Tlu^ diarntfi-er of the fnH> branclu's, memsun'd at- right angles to the plants of the t^alie.t^s, is from 4*5 mm. to 7 mm., most fragments having approximatt'ly the sarnie diamett'c top and bottom, t'xcept at the poi7it- where hifui'cation txuMirs ; a l)ran('h of 7 mm. avt^i-age dianu>ter f‘X[)ande.d to 8-5 mm. at sueh a point. 'The tmgh^ he- tweem tlu‘ products of bifurcation is about 3t) ; the distaiuM^ hiUwcMni bifuren- tions varies, anoiit 45 . 'Pin' calic(ts alternat-e on ■opposite sides of a braneli, each hra-ueh usually havitig t-wo vei'tiea.l rows of calic(*s, though Just Ix'foi'tt hifuntation tlutre ma->' he irr(tgularily. tkich ■calice arises from the u})per part of tlie Ijase of tint out' h(4ov\-. The calical opiaiing extemls inwards with a very gradual and r(\gular decri'use in dianuder for at l(tast 4 mm., and thent ant signs of soptai ridgi's in its ujipi'r [larts. EtlxM'idgr's thin section shows ntticular tissue lining its lowei' parts. Banarks : Tlu' re})tant parts of tho corallum from whieli tln'sit fi-ee hrauclu'S H,ro pi’esumed to lia,ve arisen, ar(t as yet unknown. In tin' ( dndamiin' fault block, near Silverwood, Qmtenslaud, and in the (k)ral Stag(t of tint S])ring- sur(‘ District, Quci'iislaiid, fragnxtnts \A'lios(t dinu'nsions cori'c'spcjud to those of th(' Halmaningarra .s]xx*im(tns c'xectpt that the calical (slges may h«t 5 mm. in dianx'h'r rather than 4*5 mm. occur, assoeiated with otlif>r ( 'ladochonid fragments t>f diffi'ntut proiiortions, but wlii<4i yt't may prove fo h(t t his species, and with Thanmopora ivilklnsonl ( Id-heridg(t). In vi(‘v\’ of the ahundaiice of , specimens of o, Timor, are e-vt^n larger than this Australian sjxx*i('s, but their dirmaisious are attaiiuMl l)v a specimen in the lTii\ (‘rsity of (^mxnisland fi'om Cressbrook Cn^ek, mxir Esk. 70 Dorothy Hill. Cladochonus striatus sp. nov. (Plate IL, fig. 11). Monilopora ? nicholsoni Etheridge, Hill, 1937, a, p. 59, pi. i., figs. 23, 24, text-figs. 10, 11, Permian, Callytharra, W.A. Holotype : SpcciTiion figured Hill, loc, cit., text-fig. 10, in the Collection of the Univ'ersity of Western Australia, from the Permian of the creek half a mile west of Callytliarra Spring, Wooramel River, W.A. Diagnosis : Slender Cladochomis^ with eorallites widest at the calical rim, which is 3 or 4 mm. in diameter, and with 18-20 faint longitudinal stria- tions on the epitheca. Rctnarks : Tlio specimens previously described were reptant portions of eoralla, but 1 now Inive fragments of erect branches from the Callytharra beds near ( ^allythaj’ra Springs. The species has much slenderer pi’oportiona than the later C. nicholsoni, and the longitudinal striation on its opitheca has not been noted in any other spin-ies. AClvNOWLEDGMENTS. This work has l)oen ('arried out while the author hold a Research Fellow- ship within the University of Queensland fijranced by Commonwealth Funds through the Council for Scientific and Industrial Rt^search. She is indebted for faciliti(^s for study at the Uni\'orsity of Queensland to Prof. H. C. Richards, D.Sc. The specimens wore collected partly by Dr. C. Teichert and partly by Mr. H. Coley. They \ver(^ loaned by Dr. C. Teichert of the University of Western Australia, who has given the stage and bed locations. BIBLIOGRAPHY. Clii, S., 19:J7 : Oil Simple Corals from the Permian of Yungsin. Kiangsi. BulL (P'ol. Sor. ('hind: W'll., f»p. S3-10S, pis. i-iv. Clarke, E. (h' I93S : Cori'(‘latit)n of the (^u'hotiifennis and Permian Formations of Aus- tralia, 11., \Vest(M-n Australia. Hepf. A.X.Z. d,s>'. Aflv. Sci., XXlll., i)p. 420-430. Dobrolyuhova. T. A., 1930 : Hugosa (7)rais of the Midrilt* and Fpper (krboniferoiis and ik*rmian of Ihe North Crab AcrnL N'r/. .S.aS.R., Trans. Polar Poniiti. Fasc. 28, pp. 77-158, 81 text-figs. Etheridge, IL : 1914: Yestern Australian ( 'nrhoniferoiis Fossils, {•hielly from Blount Alarmion, Len- nanl Kivei', West Ximl)erley. BalL . ti5-l47, pis. exlv-cl. Stuttgart. Graban, A. W'., 1922 and 1928 : Palaeozoic Corals of C4una, Part 1.. Tctrascjdata. PalaeonL Sinica (B), II., Fasc. I (1922), 70 pp.. I pi. ; Fasc. 2 (1928), 175 pp., 0 pis. Horitsch, F. : 1930 : Korailen dt'r Aloskaiier-, Gshcl- iind Schwagerinej^-Stufo der carnischen Alpen. Ihilaeontogr. S/adgarl, I^XXXIU.(A), pp. 99-102, ])ls. xiv — xviii, 50 text- figs. 1937 : Bugose Korailen aus dem Salt Rangt*, aus Timor uiul ans Djonlfa mit Bemer- kungcii iiber die Stratigraphic dcs Perms. Sitz. AM. WUs. ]\'ien Alath.- uatnrw. Klasse Abt. 1., CXb\ I., pp. l-IO, pis. i. ii. Hill, I). : 1937 : Type Specimens of Palaeozoic ('orals from Xew South Wales in W. B. Clarke’s First ('ollection, and in the Strzelecki Collection. Geol. Mag. London, I.XXIW, ])]). 145-153, text-tigs. !-9. 1937, a : The Permian Corals of Western Australia.. Jour.., UU4.,- XXIII.„ pp. 43-03, pi. i, tcxt-ligs. 1-U. Further Permtah Corals from Western- Australia. 71 1938 : Euryphyllum : A new Genus of Permian Zaphrentoid Rugose Corals. Proc, Boy. )Soc. Queensl., XJjIX,. pp. 23-28, pi. i. 1938, a : The Carboniferous Rugose Corals of Scotland, Part I. Palaeontogr. Soc. Monogr,, pp, 1-78, pis. i, ii. 1940 : The Silurian Rugose Corals of the Yass-Bowning District, Xew South Wales, Proc. Linn. Boc. N.B.W., LXV., pp. 388-420, pis. xi-xiii. Hill, D., and Smyth, L. B., 1938 : On the Identity of Monilopora Nicholson and Etheridge^ 1879, with CladocLonus M’Coy, 1847. Proc. Roy. Irish Acad., XLV. (B), pp. 125-138, pis. xxii-xxiii. Hudson, R. G. S. : 1928 : On the Low'er Carboniferous Corals — Cravenia rhytoides and Cravenia tela, gen, et spp. n. Proc. Leeds Philos. Soc. (Sci.) I., pp. 252-257, pi. i. 1935 : On the Lower Carboniferous Corals : Rhopalolas?na, gen. nov„ and CryptophyU lam, Carr. Proc. Yorks. Geol. Soc., XXIII., pp. 90-102, pis. iv, v. Jones, 0. A., 1936 : The Controlling Effect of Environment upon the Corallum in Favo- sites ; with a Revision of sojne Massive Species. Ann. Mag. Nat. Hist. (10), XVII., pp. 1-24, pis. i-iii. M’Coy, F. : 1844 : A Synopsis of the Characters of the Carboniferous Limestone Fossils of Ireland, pp. VIII, 207, 29 pis. 1847 : On the Fossil Botany and Zoology of the Rocks associated with the Coal of Australia. Ann. Mag. Nat. Hist., XX., pp. 145-157, 220-236, 298-312, pis. ix-xvii. Nicholson, H. A„ and Etheridge, (Jr.) R., '1879 : On the Microscopic Structure of three Species of the Genus Cladodionus M'Coy. Geol. Mag. London (2), VI., pp. 289-296, pi. vii. Fenecke, K. A., 1908 : Uber ein neue Ivorallengattung aus der Permformation von Timor, in R. D. M. Verbeek, Rapport sur les Moluques.. Batavia. Jaarh. v.h. Mijnwezen, xxxvii, pp. 672-674, text-figs. Soshkina, E., 1928 : The Lower Permian Corals of the Western Parts of the Northern Urals. Bull. Soc. Nat. Moscoti (Geol.) VI., (NS), XXXVI. Teichert, C., 1939 : Recent Research Work on the Permian of Western Australia. Austral. Jour. Sci., II., pp. 5-7. 72 Dorothy Hill. EXPLANATION OF PLATES. Plate I. Permian Corals from Western Australia. All specimens are in the Collection of the University of Western Australia. All figures by 2 diameters, approximately. iig. 1 A mjyleTUs'' pustulomifi Jludleston. Transverse section. Upper part of Cally- tharra Series near Trig. Station K;')2, Kennedy Range, near Williambury- Middalya road. No. 4-1071. Fig. 2. AUotro'piO'phylhan, sj). Transverse section, ^liddle part of the Calceolieponyia Stage in the ^^'andagee Series, near Coolkilya Pool, ^linilya R. No. 41072. Fig. 3. Eurypbylln7)i reidi Hill. Transverse section. Upper part of the Calceolispongia Stage in the W'andagee Series, north side, Minilya R., of Coolkilya Pool. No. 4107‘L Fig. 4. ? EuryphylUnn, sj). Transverse section. Locality and horizon as for fig. 2. No. 41074. Fig. 5. Verbeekiella mersa. sp. nov. Transverse secticai of holotype. Lower part of the ('alreolif^pontjUi Stage in the \\'andagee Series, north side, Minilya R., W. of Coolkilya Pool. No. 4107."). Fig. (). Verbeekiella mersa. sj). nov. Vortical section. Upj)or ])art of the Calceolis- Stage in the Wandagee Series, north side Alinilva R., W. of Coolkilya Pool. No. 41070. Fig. 7. Verheekia ialbotl (Hosking). Transverse section. Callytharra Series, Cally- tharra vSprings. No. 41077. Fig. ,8. Thamyiopora itnmenm Hill. X’ertieal section. Rnlgadoo Pool, Minilva R. No. 41069 ( 20204). Fig. 9. Thamnopora insciilpta, sp. nov. Type material. Callytharra Series near Cal- lytharra S])nngs, a, h, traivsverse section ; e, tangential and d, median vertical section. No. 41078 a, b. e, d. Fig. 10. Thayyinojyora sj). Section. Callytharra Stage, Callytharra Springs. No. 41079. Frirni KK* Pkkmian Cok’als fko.m Wkstkkn Ausai^ALiA. Plate J. i 140/41. 74 Dorothy Hill. Plate 11. Permian Corals from Western Australia. All figures natural size, approximately. Fig. 1. Amplexus"' sp. Irwin River, Fossil Cliff. No. 20212. Fig. 2. Euryphyllum reidi Hill. IMicldle or part of Calceolispongia Stage in the Wand- agee Series, north side of Hinilya R., of Coolkilya Pool. No. 20213. Fig. 3. Phrophjllum sp. Upper Cuiidlego Stage, fossiliferous horizon, above Coolkilya Pool, Minilya R. No. 20214. lig. 4. Verbeekiella mersa, sp. nov. Specimen vertically split. Shows complexity of columella. Upper part of the Calceolispongia Stage of the Wandat^ee Series, north side of Minilya R., W. of Coolkilya Pool. No. 20215. Fig. o, Favosites^ sp._ Lower Linoprodudm Stage of the Wandagee Series off, 12 chains from Station f, north of ^Minilya R. No. 2(»2K). Pig. 6. Thamnopora imme.nsa Hill. Shows calices of two sizes. Bulgadoo Pool Minilya H. No. 41069. ^ iig. /. Thcunnopovd aff. viciTftiio^ieusis (Etheridge). Lower TAuopyoductus stage of the Wandagee Series, N.W. of Station 2:3, Minilya R. No. 20217. Figs. 8, 9. Thamnopora insculpta, sp. nov. Type material. Callytharra Stage near Cal- lytharra Springs. 9 is the young part of a branch, and shows sharp ridges between .some of the caliees. Nos. 20218, 20219. Phg. 10. Thamnopora sp. Callytharra Stage, Callytharra Springs. No. 20211. hig. 11. Cladoehonus striatus, sp. nov. 'Ihe longitudinal striation characteristic of the species does not show up \ery well. No. 20220. Further Permian Corals from Western Australia 75 Plate If. New forms of Australian Birds. 77 5.— NEW FORMS OF AUSTRALIAN BIRDS. By Gregory Mathem'S, C.B.E. Communioated by L. (ilaiiert. Read 10th September, 1940 ; Published 12th February, 1942. While examining the collection in the Perth Museum with the Curator Mr. L. Glauert, to whom my thanks are here tendered, 1 found the folk -wing new subspecies. Colluricincia brunnea julietae sub. sp. nov. The form differs from typical birds in its smaller size, wing 116 mm. ; tail 95 ; culmen 20 ; tarsus 31. Type in the Perth Museum, No. A4004, a female collected by Mr. O. H. Lipfert on the Canning Stock Route, Sturt Creek, on 6th January, 1931. Typical birds measure wing 133 mm. ; tail 1 10 ; culmen 20 ; tarsus 31, Remarks : Apparently this smaller form inhabits the country far inland. The rather large series of o\w fifty skins of Melifhaga virescevs in tbe Perth Museum sh()ws that we must admit the subs].’e( ie^'., M. v. vn rcJALd ? and new forms. M. V. MCHCHISONl AND NKW FORMS. 1. Meliphaga virescens glauerti sub. sp. nov. Differs from typical birds in being paler above and in having the stripes on the feathers on the breast not so noticeable. Type in the Perth Museum, No. 6912, a male, collected by '^h\ C. P. Conigrave at Ebano in October, 1904. Remarks : Six specimens examined. 2. M. V. lipferti sub. sp. nov. Differs from other form of M. virescens in being distinctly })aler above and below. Type in the Perth Museum, N(t. A4077, a male, c(Jlected by Mr. O. Ff. Lipfert at Well 33 on the Canning Stock Route on 13th October, 1930. Remarks : Six specimens in the Perth Museum and six in x\delaide examined. Two skins from Marble Bar are considered identical. 3. M. V. lewisi. Differs from M. v. lipferti in having the tliroat whiter and the feathers on the chest britwn, foi'ming an indistinct band ; the feathers on the under- surface being not so heavily streaked. Type in the Perth Museum, No. 7729, a male, collected liy Mr. John T. Tunny on Lewis Island, L)ampicr Archijjelago, on 27tlx June, 1901. Remarks : The birds from Barrow Island seem identical. Sericornis maeulata. Examination of the material in the Adelaide and Perth Museums shows that two new forms occur. 78 Gregory Mathews, 1. Sericornis maculata condom sub. sp, nov. Differs from S. ni. nufculata in lacking f he yellowisli tinge on the belly, this part being almost uniform brown like the hanks ; the ^.})otting on the chest is not so prouomu^cil as in maculata. Type in the Adelahh' Museum, No. HIM ID, a female, collected at Ho2:3e- toun in Western Australia by Mi*. fJohn T. 'runny on 12th July, HIGH. llemarkH : This bird was only found in dense slii'ub and was mostly in pairs. This form should lie c.oin]>ared wit h the form from Mondrain Island. 2. Sericornis maculata mondraini sub. sp, nov. Differs from S. m. )}WA'ulai(( from Albany in its larger size and in lacking the dark brown flanks anrl the yellowisii tinge on the belly, this part being whitish. W'ing 58 mm. ; tail 50; ciilmen 10; tarsus 23. T\ pe in t he Jk^rt h Museum, N(). 7125, a male, (collected on Mondrain island, Re(>her<'he Archipelago, on the south-east coast of Western Australia, on 27th October, VM)4, by Mr. John T. 'runny. Remarks : This form should he compared with condoni. It was shot in dense scrnb. Tlwee specimens examined. Aegintha temporalis cardwelli sub. sp. nov. Differs from A, 1. mncfiUkimtifi in lacking tlie golden yellows in the back and in having the under surface more uniform grey, especially on thetliroatand head. Tyf)e in the Queensland Museum, collected at C'ardwell, North Queens- land, by Kendal Broadbent. Cervinipitta kimbleyensis sp. nov. Head black, this colour extending fronx the loAver base of the lower man- dible, above the eye and inchuling t he ]y a broad black fringe to the ti}> ; total length 205 to 210 mm. ; wing 121-120 ; tail 40-45 ; tarsus 41-42 ; culmon 23. This bird differs Iroin PiUa 'megarftyncha Schlogel iii having the crown of the head black, and iii its smaller bill. TJie tyj)e is an unsexed adult in the Bertii Museum, Western Australia, No. A3767, collected in f-he Derby Distiict of NortJi-Western Australia. An- other example of tlie same locality is a mummy, but in good condition. The bill was black and the legs and feet in these examples are light and in life were probably fleshy. This bird may be a subspecies of Cervini'pUta 'tnoluccefisis of Muller, 1776. Some Basalts from the North Kimberley, W'estern Australia. 79 6.— SOME BASALTS FROM THE NORTH KIMBERLEY, WESTERN AUSTRALIA. By A. B. Edwards, Ph.D., D.I.C. Read 12th November, 1940 ; PublisluHl loth May, 1942. CONTEXTS. Introdiution ... General Geology Petrology : t\vo-]iyroxene doieiltes, two-])yroxene andesine-basalts, ande.sine-l)a.salts, niioro(-rystalline-andeKine-l)a«alts, hornblemle- oligoelase-l)asalts, tuff Diseussioii l..ist of Kef(‘ren(‘es ... Illustrations ... Page 70 81 82 8 (> 89 9(M)3 INTRODLATIOX. The following notes are intended to supplement our meagre knowledge concerning the petrology of the basalts of the North Kimberley region of Western Australia (Fig. 1). They are based on an examination of two small collections, one of which was made by Mi^ssrs. A. Gibb Maitland and C. Cl. Gibson, in 1901, when they were attached to the Brockman Exploring Ex- l^edition (Brockman and Crossland,. 1901), and the other by Mi*. W. ^^'^Fitz- gerald, who accompanied the Crossland b^xjXHlition in ioOo. (Fitzgerald, Figure 1. Locality Map, showing the position of the North Kimberley (shaded). Maitland and Gibson explored the country in the neighboiirliood of Wynd- ham ; the gorge of the Isdell River ; the land to the west, north-west and north of the S^\Tinott Tableland, including the Synnott Creek, the Charnley i 142/42. 80 B Edwards A. River Gorge, and the Calder River Goi’ge ; the vicinity of Moimt Kitchener^ Mount Lyell, and Mount Trevor ; the upper reaches of the Prince Regent River, and Mormt Hann ; and some of the hills overlooking Napier Broome Bay {Fig. 2). 2, Geologioal Sketch Map of the North Kimberley, showing the localities from which the specimens were obtained. Some Basalts from the North Kimberley, AV'estern Australia. 81 Fitzgerald oxamiiied tlio country along tlie ]\lay, Loiiiiard, Barker, Ad- cock, Throsscll, Upper Fitzroy, Hami, Barnett, Isdell, Sprigg, Lower Charn- and Lower Calder Rivers. Tlie localities from which the vS])ecim(>ns wen' obtained are shown in Figure 2, wliich is copied from Easton's (1922) geological map of the region, as re- printed, to accompany Jutson's Physiography of \V^‘storn Australia ” (Jut- son, 1934). No information is a,\ ailable as to tlie iield occurrences of most of the specimens, and, as will b(‘ indicat(>d, this is a matter to which any future expedition might give attention. The collection is the ]iroperty f»f the (leological Survey of Western Aus- tralia, and my thanks are due to Mr. H. A. Ellis, wlio was Acting Govern- ment Geologist in 1939, and who, at the instance of J^rofossor E. de G. Clarke, not only gave me perjnissiou to examine it, but wint, personally, to much trouble to put it in order and to collect all tlie lit>ld notes regarding the speci- mens which could be found in the otficial records. Mr. Ellis also allowed me to make use of a manuscript report on thes(' and other rocks from the North Kimberley, which w'as written many yeai‘S ago by H. A, Fai'ijuharson. The present study lias been carri('d out in the (hnilogy Department, University of ^lelbourne, by kind p('rmission of Pi-oh'.ssor E, W. Skeats. (JENE]L\L GEOLOGY. The North Kimberley District of VW'storn Austi-alia lies between longi- tudes 124° and 128° and bc-tween latitudes 14° and 18° (Fig. 1), and is an inaccessible, and therefore little knowTi, part of x-\ustralia. Our knowledge of its geology is based chiefly on the accounts and collections of the parties led by the surveyors Hardman (1S85), Brockman and Crossland (1901), Fitz- gerald (1907] and Easton (1922). Over the greater portion of this rc^gion the outcropping rocks consist of lava flows interb(’dded whth massive sandstones of Nullagino age, which arc either horizontally bedded, or only gently folded. The sandstones appear to overlie the lava flows to a largo extent, but Maitland (1902) indicates tJiat some of the igneous rocks occur as sills. He n*fors to tlnun ]>ri(>fly as "a series of b(‘dded and intrusive igiu'oiis rocks, the pre^'ailing types being andesite, , and rliabaso,’’ sometiuK'S a(;companied l)y beds of \'olcanic ash and breccias. In plaf;es ho rc^ports that “tlie sandstones are sometimes altta-ed into hard compact (juartzite, portions of which hav(^ been caught up in the body of igneous rock,'’ while some sections “intlicate (juite clearly that the igneous rocks have (in these localities) found an easy passage along the V->od- ding planes of the sedinamtary bi'ds and .... occur in tlu^ form of sills. According to Jutson (1934), the North Kimberley is an extensive plateau whicli is undergoing rapid dissection by a nnml^er of str(‘ams. These streams are arranged radially about Mount Hann (2,800 feet), which lies near tho centre of the plateau, and rises about 800 feet above it. Th(.i “ranges” marked on most maps of tins I'lgion ar<’^ really tlu^ ritlges or jilatean rem- nants forming the interfluves betweim the deep narrow gorg(‘s tieeupied by those streams. In their lower tracts the river \*alleys broaden, but they retain their steep w^alls througliout their courses. The coastal region has undergone subiufagence, giving rise to an extnanely broken coastline of tho rias typt', the harbour.s being flanked by precipitous cliffs, often several hun- dreds of feet high. 82 A. H. Howards. PETROLOGY. In the liand specimen tlie basalts can be divided into two distinct vari- eties, a ^roup of relati\'ely eoarse-grainerl rocks, and a group of extremely fine-graine< I rocks. Examination of thin sections under the microscope shows that the coarse-grained rocks are a uniform group of ophitic two- pyroxene dolerites. The fine-grained rocks, on the other hand, comprise basalts of several varieties that can be distinguished mineralogicaily and texturally. 1 . Two-pi/ro.ren e dolerites. [87.'>(>J,* from Camp F. li. .SI. Cbanilcy diver. 137,71], from Camp F.Jk 32, Charnley Hiver (analvsed), ^ [37-73]. from the Isdell River (analvsed), [37dOj, from Camj) F.H. 32. Syiinutl Creek, Chai-tilev Uiver (analysed). |37h2J, from a large hill l)etweeti CarnpH F.B. 37 and 38. [37(i7], from a "pocket" near Camp F.B. 47. j3/(i9], from a iiill near Crosslaml s \Vart. Rriiice Regent River. [3771], from 3 miles south-east of Mt. Lyell (analysed). [377.7], from Cam]) F.B. 73. [7234], dyke in granite, from the right-hand branch of Barker River. (72.7')], dyke in th(' Lennaid Ihver, 10 miles below Mt. Eliza. Similar, but with granophyric intergi-owths of quartz and felspar : — 137.74], from Camp C. 5. fsdell River. |37t)S], from gorge l)elovv F.B. .72 (analysed). 1 3774 ], from 4 miles .south-west of F.B. 00 (analy.sed). Jhis group comprises 14 of tin* 35 specimens forming the collection, and includes the freshest specimens. As will be seen from Figure 2, the localities from wliich the specimens come are widely scattered over the North Kimberley, so that there can be little doubt that this variety of basalt is one of the wide- spread types of the region. Iwo sub-varieties can be recognised, a slightly more acid gi’oup with inter- sertal patches of granophyric (juartz and orthoclase (quartz-dolerites), and a more basic one in whicli siteh granopliyric intergrowdhs are lacking. Apart from this minoi difference, the rocks appear to be identical. Tliey are coarse- grained (Plato I, Fig. 1), ami tionsist essentially of pyroxene and basic plagio- clase, with some ddoritized biotite, an^l iron ore. Tlie pyroxene is the coarsest- grained constituent, and forms crystals about 1 to 2 mm. across, wliich are in ophitic relation with tliv felspar laths. Tw'o varieties of jyvroxiuie are present, in intimate assodation. One shows a (-r ) 2\ of about 50'^, so tliat it is a diopsidic augite, wIiMe the other show^s a (-r) 2\ of about so that it is a pigeonito. The extinction angles show- a general maximum of 35'^, with an occasional 4tF. LTsing the data givem by Deer and Wager (1938), this suggests that the compositions of the two pyroxenes are a]7proximatdy as follow^s : — Augite d MgSiOg 55 l^FeSi 03 = 10 f CaSiOg = 15 Pigeonite d Mg8i03 --- 75 l^FYSiOg =:^ 10 1 Numbers in brackets [ ] refer to specimens in the collection of the Geo- logical Survey of 7V.A. Some Basalts from the North Kimberley, Western Ax’Stralia. 83 The plagioclas^ occTirs as laths which range up to 1 inni. iii lengtli, when they appear to be porphyritio, but they average about 0*3 nun. They sliow extinction angles as higli as 33" in the symnietrical zone, and are optically positive, so that they are labradorite, of a composition about Al')40. Small areas of chlorite, pleochroic from green to jiale straw, are of fr(‘quent occui*- renee. They show straight extinction parallel to a single, M'idely-spaced, hut perfect cleavage, so that tlioy appear to i*epres('nt altercnl biotite. Sometimes the chlorite is fibrous, with a plumose structure ; and sonK'times, as in [375()J, it appears to bo pseudomorphous after small crystals of olivine. The iron ore occurs as sporadic coarse, irregular shaped grains of ihnenite and ilmenite- magnetito intergroMths, moulded on the pyroxene, anidote lias attacked the pyroxene, and converted it to a soda-am|)hibole, presumably glaucophane, which is intensely pleochroic, with X = pale violet, V = deo]) violet, Z tieep blue-green. It shows a good clt^avage paralk'l to (HO), and extinguishes at 30° on this cleavage. Reumaats of tlie original pyroxene are sometimes enclosed wdlhin the glaucophaue. S) suggests that the glaucopiiane is rich in iron, ami [lossibly magnesia. It seems probable that the augite of the original roek has altemd, in part at least to ejiidote, while the pigeonite has changed to glaucopiiane. A little pyrite has Ix^en introduced, together with \nins and patches of calcito. When the chemi('al analysts of tht^ lock is eom]3ai'ed witli the analyses of the h^ss altered rocks of Table I., it is seen tliat there is a considerable general I’esem- Filance, as would be expected, but that K^O, MgO and TiO^ have be(ai loaclied out to a eonsideraljle degree iluring the alteration, w hile Na.^O lias been re- moved in lessia’ ilegree. On tlie other hand. SiO^, f’aO and (’().> have been introduced. 84 A. r>. I^DWARDS. (JIaiK'ophaiie is n^conlf’d <*lnaf]y from motamorpliie rocks, and its occurrence as an alteration ])roduct of a doleritn is unusual, particularly since, the doleritcs in this region are for tlu^ Ttiost part fr(‘sli and un-metamor]jhos(Kl. The presence of pyrite and carhonatt‘s in the rock suggests that it lias undergone local ni(dasonuitism, or possibly autopnemnatolysis. 2. l''wo-p}jrox'ene itmivshti -imsalt.s. [.‘1749t from I mile imrtlcwi'st of ('amp F.l>. :2S. junctio!! of Alainite (’n^ck and (’liamlcy Ihvrr (analysed). |:^7(il|, from the CahU'c Pdver, d miles west of F.B. 34. j37(>3|, fnmi the Harding Ihinge, S(M> fia'l 1k4ow the summit, and 1 mile from K.P>. 43. 137(m]. from the !lai' 0*d5mm. wide, and show a tendency to foi-m stellat(‘ (dustiM's in tin* dark glassy has(a 4'he pyroxene sometinw^s occurs in radiating s}i(>a,vi‘s. Two xariidies arc pi(‘S(‘nt. Ono has ( ]- ) 2\' — , so tluit it is a pigeoiiite. The oHier has ( f ) 2V' greater than 45'^. and is pi‘oha()l\' aiigite. In most of tlu^ speeinuais tlu^ pyroxeiu' latlis were just b(‘ginniug to erystalli/.(^ wlu'ii the rock was chilhsl, and indistiiud trichytes of pyrox(‘ne can madt^ out in 1h(^ dark glassy l)as(‘. Thesis trichyt('S Paid to a raosif ion about AlidO. In ('ach scx’.tioii there are one or two niiero|)h(‘nocrysts of alt<‘r('(i oli\'ine, which i‘(>tain tludr idiomorphie outlint* to sonue (‘xtent (Plate I., Fig. 2). ddie atamdant glass is gn'yish and opacjue, hut urubu* liigh maguilicatiou it resolves into a ufdwork of [>yrox('U(‘ microlites. Iron orc^s ane gtaierally absent. Tcextiirally thc'se rocks lia-ve considerabh' attinity with tholeiites, but mineralogieally tli(\y aine I’chifed to the andesites. A elu>mieal analysis of [3749] (d’ahh' II., No. I) iiulicates that tlu* rocks an^ liasaltic in com])osition. hilt midwfiy, as it W('re. betwfeen th(e thoh'iities ])i*oper and the andesites. The Al.iO;p Na^O and K.>() conhuits ar(’ too high for tholeiit('s, just as the amk^sine hdspar is too acid. Tlu>s(‘ ftmtures also distinguish this group of basalts from Ihe t wo-pyrox. 57. |3779|, from 3 miles math of F.P>. 79. at .\apier Broome Bay (analy.sed). 13777], an epidositized speeiiium from tlu* same locality as j377(>| (analysed). j377S|. from the same locality as |377ti|- [724()j. from tlu* Harm River, |7271], a dyke, from near Mt. .lo.seph. The sp(K*im 0 ns eom]irising this group are coarser grained than the basalts of Group 2. and contain less glass (Platt' 2., Fig. 3). The felspar is a more acid andesine, about Ab-,,, occurring in laths that He at all angles to ono an- Some Basalts from the North Kimberley, Western Australia. 85 ■other, and are often partially micaeisod. They are associated with laths of clinopyroxone, so altered to chlorite that the composition of the pyroxene camiot be gauged. These laths are set in a base of dark glass with which are associated numerous areas of a(>pIe-groen chlorite, showing ultra-blue polariza- tion colours. Tt is this chlorite, pi-esumabh% which accounts for the high MgO content in the chemical analysis of [3770] (Table IT., No. 2). In some of the specimens, notably [3777], the felspar has been oxtonsively altered to epidoto, which occurs in irn'giilar areas through the sections. An analysis of this extensively altered specimen (Table II., No. 3) sliows that SiO.^, MgO, Na20, KoO and TiOo have beiai leaclu'd out of it, while the FeO has been oxi- dised, and a large (quantity of OaO has been introduced. In the absence of an analysis, t hese rocks y ould be regarded as basic pyrox- ene-andesites, but their chemistry indicat(‘s their basaltic charat^ter, and their affinity with the other rocks from this region. Specimen [3772 1 appears to be intermediate between these audesine-basa.lts and the two-pyroxene andesine-basalts of Group 2. The pyroxene in it forms colourless microphono- crysts, with (-h) 2V greater than 45‘^ so that it approaches augitc in composi- tion. The plagioclase is too mucli alttM’ed to secondary mica for its composition to be determined, but it shows a pronounced stellate arrangoim'ut, and the interstitial spaces are filled with a dark glass containing trichytic crystals of pyroxene, in [7240] the groundmass consists of a lattice woi’k of cijioritized pyroxene inicrolitos in a flark glass. In piac(>s this structure is replaced by granular calcite. Another specimen difficult to classify, but having affinities with this group is 17271], a dyke rock from near Mount Joseph. It consists of small stumpy laths of much altered andesine ophitically intergroYTi with areas of partially altered pyroxene, which shows a large optic axial angle, and appears to be augite. The glassy base is much altered, and patches of chlorite occur scattered throughout the rock. Associated with the chlorite are small areas of glaueophane, intensely ]ileoc]iroic from deep-green to pale-brown, pale violet, or nearly colourless. 4. Microcrifslalline andesine-bamltt-i. f37.>9], from the Synnott Tableland, north-east of Camp F.B. 32. 17230], a dyke at the base of Mt. Brennan, intruding Xullagiiie sandstone (Fitz- gerald). (Barqnhai'son suggests it mav l)e an interbedded lava or a sill.) [7240], from th(‘ Hann River, near its junction with the Barnett Riv^er. As will be seen from Figure 2, two of these rocks come from tlie south- east portion of the region, away from the main ai'oas of basalt, as ma}}ped by Easton, while tlie third occurs in the south-central part, associated with other basaltic ty])es, [7240] consists of minute laths of plagiocdase accompanied by small grains of iron ore, in an abundant groundmass of glassy material that has been altered to pah'-green chloritt;. Tlu; felspar laths lie at all angles to one another, with no SLigg(vstion of parallelism, and show extinction anghvs up to 12° in the symmetrical zone wdien twinned, and almost straight extinction whmi micro- litic, indicating that they coTisist of acid andesine, of a composition about Ab-Q. Small patches of (*alcite occur throughout the rock, in the glass ; anil calcite fills a iuiml)er of small amygdules in the rock, when it is associated with small amounts of radially fibrous chlorite and finely granular cpiartz. Similar granules of quartz occur through the rock in the vicinity of the amyg- dules. Oi'casionally an amygdule is coniydetely filled with chlorite. so A. B. Edwards. Ill [7230] tlio plagioi'laso latlis poaraer-grained, thoiigli still small, and dospitft partial kaoliiiisation, show extinction angles np to 12^ in the symmetrical zone, indicating that they consist of acid andesine (Ahyo). The amygdnles in this rock are tilletl (i) with chlorite ; or (ii) with quartz and unt^^ inned felspai’, stained with iron oxide, and sometimes accompanied by , chlorite ; or (iii) witli (piartz and chlorite. In other respects, however, the rock is identical with [7240]. The same texture is prt^served in [3759] from the Synnott Tableland, but in this rock the felspar is so altered to mica and kaolin material that its com- position camiot bo made out. In addition the rock, which contains numerous amygdulos of chlorite, is stained deeply with limonite. In the absence of a chemical analysis, these rocks might be regarded as basic andesites, but they have been classified as andesine-basalts in view of their general resemblance to the other rocks of the region, which when analysed prove to be baselts. 5. Hornblende-oUgoclase-basalL [7248], from, the Hann River, near its junction with the Fitzroy River (analysed). This variety, which is represented by a single specimen from the south- eastern part of the area, is a micro -crystalline rock consisting of laths of plagioclase and idiomorphic micro-phenocrysts of pyroxene, set in a dark mesostasis of groundinass which has partially crystallized as radiating sheaves of a grey-bro\ni hornblende tluit shows almost straight extinction. The plagioclase is largely altered to secondar\' mica, hut lamellar tvdnning can be made out in some t-rystals, and gives extinctioir angles up to 10*^ in the sym- metrical zone, so that the felspar apjjears to bo basic oligoclase of a composition Abyo-Abyr^. T'ho |)yroxene i^ completely altered to chlorite, so that its com- position (Uinuot be determined. In places it is ophitic towai'ds the ];)lagioclase. The hornhleinle fibres are sometimes interleavetl w ith green chloritic material, representing original glass, anil less frequently with line felspar laths. In view of the acid nature of the plagioclase, and the abundant hornblenile, the rock resembles an andesite, hut a chemical analysis (Table 11, Xo. 4) indicates that it is a basalt. The fine texture of the rock is shown in Plate 2, Fig. 4. 6. Volcaniv Tuff. [37or>], from the Synnott Tableland, north-east of Camp F.B. 32. This specimen is from a much weathered rock consisting of angular fragments, up to 1 cm, across, of highly chloritized material which contains a few small latlis of altered felspar, cemented together by material consisting chiefly of limonite, chlorite, and fine mosaics of secondary (juartz. The rock is almost certainly a w'eathered tuff. DISCUSSlOX, Tlie oufcro]5s of basaltic rocks iu the North Ivimherley region cover an area estimated by Easton (1922) as about (>,250 square miles. The inadequacy of any generalisations concerning these rocks, based on an examination of a collection of only 35 specimens is manifest. However, the s{)ecimens dis- play a marked uniformity in their general petrological character, wdiieh, coupled with the wide scatter of the localities from which they were obtained, justifies some suggestion as to the probable petrological character of the North Kimberley basalts as a whole With the excei>tion of [3771] (Table 1 Some Basalts from the North Kimberley, Western Australia. 87 No. 4) all the inialterc([ specimens that have lieen analysed are eitlier .saturated with res]:tect to silica, or nearly so ; and cJiemicaily, they have many features in common with the tlioleiitic basalts (Kennedy, 1933) or plateau-basalts (Wash- ington, 1922) as distintd from the olivine-basalts (Kennedy, 1933), as is shown by Table 111, in Avhieh the averaged analyses of the more or less unaltered North Kimberley rocks are compared with averaged analyses of the East Kimberley basalts, the tlioleiites of soutli-western Western Australia, and the theoretical tholeiite magma type. This resemblance is particular! \' marked witli the North Tvimberley dolerites. The basalts tend to be rather more an- desitic in c.om|)osition. Mineralogi{'ally, also, the North l\imberley rocks are related to the tlioleiites and their derivatives. They tend to form a distinctly (*alc-alkaline suite, grading towards andesites, rather than towards the more alkaline types which accompany olivine-basalts. As indicated, the North Kimberley rocks resemble, in theii* cliemical com- position, those of East Kimberley (Edwards and C’larke, 1940), but the varieties of basalt known to occur in the two regions cannot be matched. This is not surprising if the Noi'th Kimberley rocks are of Nullagine age, while the East Kimberley basalts are Cambrian, or even |)ost-C^ambrian in part. The rela- tive abundance of dolerite specimens in the collection suggests that this rock is the most mdespread in the region, and, since it is also the most basic found there, it may approximate to the composition of the parent magma. AVhether it bears such a relation to the basalts, also, must remain doubtful, for reasons given below. As pointed out liy Farquharson (in. his unpublished manuscript), the dolerites from North Kimberley bear considerable resemblance to the dolerites described by Maitland (1909), and by Talbot and Farquharson (1 920), from the drainage basin of the Ashburton, Cascoyne, ami Oakover Rivers, and the Hammersley-Ophthalmia Plateau, in the North-West Division. The sig- nificance of this resemblance is that the latter dolerites occur largely in the form of sills, up to 300 feet thick, intrue so, but, since there are no details as to the mode of occia*ren<'e of the specimens in the collections examined, except an occasional description of a specimen as “ from a dyke ” (and these occur in four of the five grou|)s established above), and in the case of one dolerite specimen [3771J a note ("laccolith ?”), it is a matter that must be left for future exploreis in the North Kimberley to decide. 88 A. 13. Edwards. I SiOo 31-13 ALO, 13-16 Fe.O. 2-0() FeO 11-52 MgO 5-88 CaO 9-70 Na^O 1-40 k,6 1-14 H.O+ 6-4] H.,0— 0-30 CO, tr TiO., 1-19 PoO, 0-01 MnO 0-19 FeS., n.d. 100-17 TABLE 1. o 3 4 50-70 50-98 40-90 14-02 11-00 14-12 0-90 5-72 0-07 12-70 10-07 12-30 3-06 5-88 6-90 8-06 8-09 10-92 3-01 2-75 2-83 1-37 2-11 1-07 0-20 0-28 1-24 1-33 I • 73 0-22 nil 1-S4 0-70 1-88 0-90 0-03 0-29 0-90 0-43 tr 99-94 101-00 99-02 5 0 A 54-44 50-34 49-42 7-72 12-80 14-95 8-22 4-11 1-38 11-20 8-98 10-76 4-00 1 ■ 10 0-16 0-08 12-90 9-85 3-14 Ml 2-70 1-91 tr 0-72 0-20 0-00 0-77 2-01 0-62 0-09 nil 0-08 nil I -48 0-34 1-95 . . . . • . 0-55 0-22 0-36 0-47 0-1 0 O-IO 0-26 101-38 99-08 100-03 Explanation. 1. [3774], tw(»*pyn)xi‘n(‘ (iiiartz-dolcrite. from 4 milos .soutJi-west of F.IL <>6. Analyst— A. 13. Edwards. 2. [3768], two-yjyvoxeno q iiartz-dolcrite, from ttio below F.13. .“>2, north of Mt Shadford, Fpper Frinee Regent River.' Analyst— Q. G. Gibson. Bull \u‘ (reol. Sun:. IF. A. 1. [37.'>IJ, two-pyroxene dolerite, fi-om Camp, F.B. 32. fSynnott Creek, Charnley River Analyst— C. C. AVilson. Bull. 07. GVoL Surv. IF. A. 4. [3771], Iwo-pyroxene dolcrit(% from 3 miles south-east oi Mt. Lvell, Glenelo- River {laccolith). Analyst- G. Gibscni. Bull. 67, (ieol. Sure. IF. A. 5. [3733], alt(‘red two-pyroxene ilolerit(‘, from the Isdell River. Analyst— C. G Gibson Ball. 67, (reol. Snrr. IF. A. 6. [376(1], Epidote-glaucophane rock (a]tere ... aiA- Fe..()., FeO ... MffO ... CaO ... NaAl ... KgO ... H,0 + H.,0— c6.> ... TiG", ... p,o: ... MnO ... FeS., ... TABLE JI. 1. o 3. 4. 51-40 51-04 45-59 49-50 16-38 13-00 13-54 17-29 1 - 20 2-08 9-39 3-95 12-78 8-40 4-87 7-78 2-30 7-94 3-01 7-24 7-32 5-68 20-19 7-75 3-14 4-30 0-57 3-23 1-72 1-83 0-00 0-67 0-12 0-30 0-28 0-27 1-99 2-64 0-68 Ml nil nil tr. 1 -00 1-30 0-08 0-87 0-01 tr. 0-34 0-37 0- 13 nil 0-17 n.d. 99-47 99-37 99-40 99-82 Evpla, nation. 1. [3749], two-pyroxene andesine-basalt, from 1 mile north-west of Camp, F.B. 28, junc- tion of Maurice Creek and Charnley River. Analyst— G. G. Gibson. Bull. 67, Geol. Surv. TF.A. 2. [3770], andesine-basalt, from 3 miles north of F.B. 79, Xapier Broome Bav. Ayialyst — C. G. Gibson. Bull. 67, Geol. S'urv. IF. A. 3. [3777], epidositized andesine-basalt, from the same locality as 3776. Analyst — C. C. Wilson. Bull. 67. Geol. Svrr. IF. A. 4. [7248], hornblende andesine-basalt, from the Hann River, near its junction with the Fitzroy River. Analyst — A. B. E(hva?*ds. iSoME Basalts from the North Nimberley, Westerist Australia. 89 TABLE IIT. 1 . 2 3. A. B. C. SiO, ... 50-3 49-9 50-5 52-2 50-5 50 A1,03 ... 14-5 13-6 15-8 14-5 14-8 13 PeO, Fe.^Og ... ... 13-2 14-0 12*1 11-2 11-5 13 MgO . . 5*7 5-0 5-8 5-0 0-0 5 CaO ... 8-3 9-3 0-0 7-3 10-9 10 Na^O 2-8 2-5 3-0 2-9 2-9 2-8 K 2 O ... 1-5 1-0 1-4 2-0 0-5 1-2 1. Average of North Kimberley basalts and dolerites (7 analyses). 2. Average of North Kimberley dolerite (4 analyses). 3. Average of North Kimberley basalt (3 analyses). A. Average East Kimberley basalt (7 analyses), A. B. Edwards and E. de C Clarke Jov)\ Roy. Roc. W.A., xxvi, 11)39-40, p. 93. ’ B. Average Tertiary tholeiite from south-w<‘stern Western Australia {3 analyses). A B Edwards, Jour. Roy. Soc. W.A., xxiw I937-3K. p. 7. C. Tholeiitie Magma Typt'. W". Q. Kemusly, Anicr. Jour. Sci., 8er. ;7, 25. 1933, p. 239. REFERENCES. Brockman, E. and Crossland, C., 1901 : Report on the Exploration of the North-West Kimberley. Perth : By Authority. Deer, W. A., and Wager, L. H., 1938 ; Two new Pyroxenes ineluded in the system clino- enstatite, clinoferrosillite, dio})side, and hedenbergite. Mineral May., xxv. j). 22. Easton, W. U., 1922 : Report on the North Kimberley Distrh't of WT'Stern Australia. Puh- lication No. 3, Deparlmeml of the North- West. Edwards, A. B., and Clarke. E. de C., 1940: Some Cambrian basalts from the East Kim- berley, W^estorn Australia. Jour. Roy. Roc. xxvi, pp. 77-94. Farqnharson, R. A.: Petrological mjtes on rock specimens from the Kimberlev Division. Unpublished manuscript rejmrt to Geol. 8urv. West. Aust. Fitzgerald, WL V., 1907 : Reports on Portions of the Kimberlevs, 1905-(i. Parliamentary Paper, West. Aust., No. 19. “ Hardman, E. 1., 1885: Report on the Geology of the Kimberley District, WUstern Aus- tralia. Perth : By A uthority. Jutson, J. T., 1934 : The Physiography of Western Australia, 2nd edit. Bull. 95, Oeol. Surv. West. Aust., pp. 34-40. Kt'nnedy, W. Q„ 1933 : Trends of differentiation in basaltic magmas. Amer. Jour. Sci 25, pp. 239-256. Maitland, A. Gibb, 1902; Ann. Prog. Rept. Oeol. Sure. Il'e^b Aust. for 1901, p. 9. Maitland, A. Gibb, ami 3 homson, d , A., 1909 : The Gascoyne, Ashburton, and WVst Pilbara Goldfiekl. Bull. 33, Oeol. Snrr. West. Aust., ])j). 163-9. Talbot, Wk E., and Farquharson, R. A., 1920: The Geology and Mineral Resources of the North-W'est, Central, and Eastern Divisions. Bull. 83, Oeol. Surv. West. Aust, pp. 142-5, 190-2. 195-214. Washington, H. 8., 1922 : The Deccan Traps and other Plateau Basalts. Hull. Oeol. Soc. Amer., 33, pp. 765-804. 90 A. J3. Edwards. Plate I. tig. 1. Alierophotograph of typical two-pyroxene dolerite. In ordinary light. Xlo. lig. 2. T\vo-i)yroxene andesine-basalt, in ordinary light, showing the pyroxene laths in the glassy base, and an olivine niicrophenoeryst. Xl5. SoMK Basalts from thk North Kimberley, Western Ai s'I’kaiha. 91 Plate J. 92 A. B. Edwards. Plate IT. Pig. 3. Andesine-basalt, in ordinary light, showing patches of chlorite, x 15. Fig. 4. Hornblende-oligodase-hasalt, in ordinary light, showing its extremely fine-grained texture, x 15. Some Basalts from the Xohth Kimberley, Western Australia 93 Plato II ■ t 5, t i ■! i J f 'i • ■]' i I i 95 Note on Fossil Cokals from Langley Park I^oke, PEiiTir. T.^NOTE ON EOSSIL CORALS FROM LANGLEY PARK BORE, PERTH. By John W. Wells, Ohio State University, Cohnnbiis, U.S.A. Communicated by Dorothy Carroll. Read : 12th November, 1940 ; Published : 15tli May, 1942. INTRODUCTORY NOTE. Through the kindness of Dr. Dorothy Carroll of the Department of ecol- ogy, University of Western Australia, I have been enabled to examine the specimens of shiny hexacoi-als obtained from the Langley Park Jlore at Pertli. All the spociinons came from a depth bet\A'(H'n 42S and 440 feet and otauirred in a “soft sandstone or unconsolidated grey sand, *nediuin grained, with a little clay.” Two species are repi't'sented in the colUnJ/ion, one belonging to Tremafotrochns latvroplmus Dtmnant, tlie oth^r jirobably a now species of oculinid but not specifically named at this time. Tlio fact that the first specaes jiroviously has been found in Balcombian sti-ata at Nhelforfl, Muddy Creek (lower), “ Fishing Point,” and “ Lower Moorabool,” and in Janjukian strata at Spring Creek, (-ape Otway, and Lake AloxHTidrina (Janjukian ?), Victoria (Dennant ami Kitson,* p. 135), miglit b(s taken to indicate BaJeomb- ian or Janjukian ago for some of the beds in the Langley Park Boro, but tlie known long range of some of the corals of the Australian Tertiary deposits (a few species even ranging from Balcombian to Recent) makes any such assumption based U 2 )on the corals unwarranted at this time. Very careful studios of the Australian Tertiary coral species and their stj-atigraphic dis- tribution are d(!sidorata, practically nothing having been done since Dennant’s work at the turn of the last century. DESCRIPTIONS. Family CARYOPHYLLIIDAE. Subfamily TURBINOLIINAE. Gcnvis TREMATOTROCHUS Tonison Woods 1S77. Trematotrochus latcroplemw Doanant 1899. Tm/iH, Proc. Hoi/. Hoc. .Sonllt Aaslrtdia. xxiii 282, pi. 9, figs 2u, b. The single, slightly worn spet-imon idtaitified with this species agrees closely with Donnant’s excellent description of tlie types, excejit for its size, which is smaller. Typical specimens, accoi-ding to Dennant, measure 5 mm. ill height, and 2x3 mm. across the (?ompi*ossetl calico. JIks present specimen, probably immature, is 2*5 mm. in height, tlie calico I x 2 mm. In all other respects there are no differences. 13ie septal aiTangeuKHit of the specimen is peculiar to this and ouo other species of the genus, that is, there art) thro(! complete cycles of septa (24) with the fourth cycle dev’^eloped only in those systems on either side of the ends of the longer axis of the ealiee, giving a total of 32 septa. * J. Dennant and A. E. Kitson, Catalogue of tlie Described Species of Fossils (except Bryozoa and Forarainifera) in the Cainozoic Fauna, of Victoria, Soiitli Australia, and Tasmania. Hec. Geol. Survey Victoria, Vol. 1, IT. 2, p. 89, 1902-6. i 143/41 90 JOHJN W. \A'ELJLS, OcouTvencQ- : Typo locality, Halcoiiibian at Shelford, Victoria ; and at Muddy Crook, Victoria. Langloy Park Bore, Forth, Western Australia, bo- twoen 428 and 440 foot. (No. 19,999, University of Western Australia, Dept, of Geology). Family OCULINIDAE. Subfamily OCULININAE. Goniis OCULINA Lamarck 1810. Oculina ? sp. Several small fragments may pertain to this genus, ranging from Upper Cretaceous to Recent especially in Europe and North America, but all are badly worn and certain essential structural details lacking, so that it is un- desirable to make them thci types of a now species. Thoi*e are, however, no species now known fj’om the Australian I’ortiarios that remotely resemble these pieces, and there is no point in comparing them with forms occurring elsGV'liere until a more precise generic assignment is possible. The corallum appears to have been small and dendroid, the corallites about 1-5 to 1-75 mm. in diameter, rather sliort and branching mostly in one plane simultaneously on either side of the ]->areTit nearly at right angles. ISepta strongly spinose laterally, in three compkdt' cycles (24), tlio first, two equal and extending to the columella, the third little more than rudimentary. Columella small, composed of 4 to 0 twisted trabecular strands. Dissexn- ments feeble. Externally the corallittis art; covered AA’ith small faint costal granulations, arranged more or less in rows corresi)onding to the septa. The })rincix3al obstacles to generic identification lie hi the absence of we]l-])r{'servod calicos showing the disj)ositioii of the pali, traces of which are lAresent, and the lac'k of information concerning the mode of colony-forma- tion. If x)ali are ])rescnt in one inx^gular crown b{'fore the first two cycles, and this is (|uit6 likely, the gtaius may be either Oculina or ArchoheUa, de- pendiiig on the presence or absence of a persistent axial corallite ; if there is but one ci’own before the second cycle it may be SclcrJielia, a genus known only from two Recent species. (No. 20,000, University of A\’estern Australia, Dept, of Geology). 97 Foraminifera and a TubicoloT'S W'okm from the Permian of 'jhe North-West Division of Western Australia. 8 .— foraminifera and a tituooi.ous worm FROM THE PERMIAN OF THF NORTH-WEST DIVISION OF WESTERN AUSTRALIA. Hy Walter tl. Parr. Road: Utth D(T*cinbDr, 1949; PuJilishod : 15tli APiy, 1942. INTRODUCTION. 'Thin pajior f>res('nts t\u) rosiilts of tlio oxaniination of a iiuidIkh- of rock samples and otluM* material (containing foraminifera from the I*ormian of the North-West Di\'ision of Wi^stiwn Anstralia. I'hc s])(tcim('ns w'l^re coll(K*t('d by Mr. Htairy (Viley and Dr. U. 'I’cicliert and. witli two exceptions wlii(cli will be noted lati'r, are all from beds forming pai't of tlie Wandagei^ Stagic of Dr. H. (1. Raggatt (UKJh) and latter authors. Foraminifera ha.\-e not Ixcon prtw'iously i'ecord(‘d from beds of tfac W'an- dagee Stagi* and, at first glanet^, the liard, brownish, impur(c limestoiK's and calcareous, tine-grained sandstones, in placi‘s highly h'l’rugiiHiUs, from these btnls apptTir indaA’ourable to theii* occui‘r(‘nc(L lA»rtunat<*ly a- eolkudion of tht^ largtM* fossils, pr(‘sented to the Must'nm of the Nuetorian Min<*s Depart- ment in 1 9‘hS by Mr. Htairy (V)l(cy, include(l a, speeinaai on the weatlasred surfatce of which exampkes of a. largi^ foraminih'r, now identiiitMl as ////per- finimina coleyi^ sp. nov,, were r(^(‘Ognis(‘d. Mr. ('okiy and Dil 4\4chert after- wards forwarded many samples trom which the foraminifea’a now' desicribed were obtained. Tlu'i calcareous nature of th(5 rock in most tcasf's enables the sandy-shelled foraminifea’a., which constitute almost thi^ whok^ of the fauna., to be freed from the surrounding mati*ix by imtiuvrsing the sani])I(' in wiatk hydrochloric acid. The results of this treatment have bfMni surprisingly successful, for, although the riumixa* of spoci<^s is not large, tln^ nuinbia*, size and piM^serv’ation of tlu^ specimens is remarkabkL Other sp(T*im<’ns were identified on the surface of those sampk\s which (•oiild not 1)(^ treatcul with acid and a number were obtairuMl from wt'uthenid sandy shale. 'Tw(4ve genera and fifttnai species of forarninihua are^ re(U)rded, of whicli twelvt^ spocies are di^scrilxM as new. Th(m> is also what is bifievcul to he a ncAv tubicolous Nvorm. DFSCRIPTION OF 'fHF vSAMPLFS. In a recent paper, Dr. Ttnchert (1939) has dealt with the stratigraphy of the Permo-Carboniferous soijuonco of the Nortli-Wc^st Division of W'estern Australia. The sefpRMioe has biHui divided by Raggatt (1931)) and others into the following stages, wliieh are given in doscend/ng ordiu’ : MMndagc'o Stage KenncMly Stage. Pyro Stage. Wooramel Stage. Callytliarra Stage. Ij^'ons Stage. 1 144 / 41 . 98 Walter J. Parr. As a result of his work with Messrs. H. G, Higgins and E. Utting on the Wandagee beds, Dr. Teichert subdivides the Wandagee Stage as follows- (from above) : — Penestella-Helicoprion bods. Lainelli branch beds. CalceoUspongia-Strophalosia beds. Calceolispiongia -A ti losteges beds. Lingula beds. On the presence of the shark Helicoprion and the crinoid Calceolispongia,. he suggests an Artinskian rattier than a Uralian age for the beds. In a later personal communication, dated 18th May, 1940, Dr. Teichert states that, on a visit to Wandagee Station in 1940, he foimd an horizon higher than any previously met with, with the result that the total thick- ness of the Wandagee beds is increased from the 2,000 feet given in his paper to approximately 2,500 feet. With the exceptions noted later, the samples- examined by the w riter were from the new horizon and from the Calceolispon- gia (the G.-Slrophalosla and the G.-Aulostcges bods have not been different- iated) and the JAng'iila beds. As a result of the study of a large ((uantity of material, it w*as found that each of those beds was characterised by the presence of certain species wEicli occ'urred in practtically every sample from that horizon. It is therf'fore not necessary to give particulars of moi’c than some representative samples from each of the beds. Before dealing wn'th these in detail, it may be stated that the residiujs left- from the samples after treatment with hydrochloric acid consist almost wholly of foraminifera, ferruginous mud, fine angular (|uartz. grains, and frecjuently numt;rous flakes of mica. Evidence that silieification of some of the calcareous fossils has occurred is provided by the presence of sili- ceous replacements of mollus(!an shells in the residues. Pppermost beds (uiuuuncfl)— about 209 feet above Fenestella beds: (1) Highly ferruginous fine-grained sandstone from Nalbia Paddock, about 110 cliains due east of Trig. Station, Wandagee HilL (Coll. C. Teichert.) F orain i i lif era — Amniodiscus wandageeensis, sp. nov. common, exposed on weathered surface of rock. (2) Highly ferruginous fine-grained sandstone from Coolkilyia Flat, approximately I mile south of Homestead-Garden Road and I mile east of Shed-Ontcamp tele|)hone line. (Coll. H. Coley.) Fo r am ini f era — Ammodiscus wandageeensis, sj). nov. common, ex])osed on weathered surface of rock. Calceolispongi(t beds, (3) South of Minilya River. Light brown impiire limestone, near top of Calceolispongia-Aulosteges beds. (Coll. C. Teichert.) The residues after treatment with HCl consist principally of countless fragments of Hyperammina coleyi, sp. nov. F orarnini f era — Hyperammina coleyi, sp. nov. abundant. Hyperamminoides acicula, sp. nov. frequent. VOKAMINIKKRA AND A 'I’ll lUdOLO OS \V^)RM RROM THR PKKMIAN OK 'I'Hl': NoK'rn-W'KST 1 )i\'ision ok W'kstiorn Ats'I'rai^i a. \h} (4) Minilya Koiwi, (’oolkilyia h'!al,oast’ limh of syiutliiu' nortli of W'aud- ago(> Hill. Light hrowji impuri' liinestouo. ((.’oil. C. 'IVichi'rt .) Koi'juniiiifi'i’a Tolypanimina undulata, sp. nov. \ory coimDojt. Hyperammina coleyi, sp. nov. fnaiuoiit. Hyperamminoides aciciila, sp. nov. v(r‘v cinninoii. (5) Hurua I’iuldock, Waiidiigi'o Stiition, on W'andagoi' M ia, Mia Koaxl, a.hout- :{ niikw in .V from liurna Hurtui Mill. 'r<’'ii*la'rl .) I )i‘. 'Toirlun't st^ali's (lia-t tlit' rock (a. light, hrown, donwe, itnpur*o liino.st.oiu‘) coot a, ins fi'agmont.s of a. sptvios of (JalceolispotKjiit and r(>mains of (Uc'iofhifridimi frotu which lu' con<‘ludos an approximately Middle \\'anda.g('o ag(i i>f l lu' h(‘ds. I fo adds I hai t.lio out.crop is (piit-i' isolaUnl hy majoi* fa.nll.s ami is sitauitiMl more t-luin K) miles noi‘(.h-ea,st of the lypi’i ar('a of tlu' \Vandag(M5 beds (\Va,nda,ge(^ Ifill). 'Tla^ hu’aminifera. inchid(' two sp<^cloK which liav(^ othej*wisi> occurr(‘d onl>' in the l/nujidu IkhIs. Koraminifera Ammodiscus nitidus, sp. nu\. fre(iii('ni. Tohf{i(ininih\ii sp. r*ar(‘. Hyperammina coleyi, sp. nov. (me hrokcai exampkv 1 1 i/j)er(itnftiinol(lc,H sp. a. fragnuMit. Tlmnnumuitf /Kipllhtfn If. B. I5r*ady. rart*. LltHfulit l)eds. (()) South side of Minilya li*, ix'ar Coolkilyia iN)ol. W’andagoii Station. Brown impur(^ linu^stoiu^ ((’oil. II. (’oley.) 'riiis material is very rich in for'aminiha-a. a.Jid iiichak^d a worm tnlx^ const.ruct(xl of hrokc'ii spong<^ spicuk^s. Boiaminihau — Ammodiscus wandageeensis, sp. jlov. vspin and Bai‘i‘. rare. Reophax subasper, sp. nov. ranv Reophax tricameratus, sp. nov. rana Trochammina subobtusa, sp. nov. nu*e. (7) Sanui locality as No. C). Bi’own impiin! lim(‘slone. ((’oil. ('. d’lachert.) k'or*aminif(a'a ar(! vei'y common and ijjchidf^ nunaa’ous ]). nov. common. Hyperammina (?) rudis, sp. nov, rare. Hyperamminoides acicula, sp. nov. common, Reophax subasper, sp. nov. rare Probal>l\- from Lingula beds. (9) Extreme south-east corner of Coolkilyia Paddock. (Coll. H. Coley.) These specimens differ from any of the othei’s recorded in having natura11\' weathered out of sandy shales. They are generally fragmentary and distorted, as is so frequently the case whoii foraminifera are found in shales. The weathered material in which they occur is a cinnamon-coloured fine sand. Foraminifera — Ammodiscus wandageeensis, sp. nov. rare. Tolypammina undulata, sp. nov. rare. Hyperammina coleyi, sp. nov. freqiient. Hyperamminoides acicula, sp. nov. common (specimens usually flattened, or transversely ribbed through pressure). Crithionina teicherti, s]>. nov. frequent (specimens generally colla])sed). Reophax tricameratus, s]i. nov. rare. In view of their interest, the following specimens are also included, although they are from beds below the AVa.ndagee Stage : — (10) Brownish-green thin-bedded micaceous sandstone from sandstone outcrops in the Cascoyne River at Jhnba Jimba Homestead. (Coll. C. Teichort.) Dr. Teicliert states that this is older than the Wandagee series of the Minilya district and is placed by Raggatt (1930, ]). 135) in the Byro, F oramii lif era — Hyperamminoides acicula, sp. nov. 6 selected specimens. (11) Ferruginous, finely-sandy shale from Gastropod horizon of Cundlego Series, about 1 mile west of Cundlego Well, Minilya Kiver, (Coll. C. Teicliert.) Foraminifera — Hyperamminoides, sp. cf. acicula, sp. nov. very rare frag- ments. Foraminifera and a TuBicoEors W'orm from the Permian of the North-West Division of Western Australia. 101 TABLE SHOWING SPECIES OF FOK.WHINIFEHA OCCI BRING IN BEDS OF THE WANDAGEE STAGE AND THEIR DISTRIBFTION. (This is Ijaspd on Samples Nos. 1-4 and 6-8.) Species. Uppermost (unnamed) horizon. Calceoli- spomjia beds. Lingula beds. 1 Ammodiscus wandageeensis, sp. nu\^ X X X X X X X X X X X X X X X X Ammodiscus nitidus, sp. nov Glomospira adhaerens, sp. nov Tolypammina undulata, sp. nov X Hyperammina coleyi, sp. nov X Hyperammina (?) rudis, sp. nov. . . Hyperamminoides acicula, sp. nov. . X Psammosphaera pusilla, sp. nov - Thuramminu fxtpiUata H. B. Brady Crithionina teicherti, sp. nov CalcitorneUa stephensi (Ilowohin) ... A mmobandik.s woolnouqhi Crespin and Parr Reophax subasper, sp. nov. — — Reophax tricameratus, .sp. nov Trochammina subobtusa, sp, nov. - The foraminifera, witli the exception of Calcitoniella stephcnsi, are all species coiLstmcting theii' sliells of extraneons material, nsnally (piartz grains, cemented together by the animal. As the .species are iniarly all new, a clo.se com arison with faunas elsewhere is not po.ssible. Tw o of' the .specie.s, Cal- oitornella stephensi anil Amitiobaculifes iroolrioiic/hi, have lioon dascribetl from the Permian of Eastern Australia, d’lie genera, with the o.xception of Hyper- amminoides and CalcitorneUa, all occur living at the present day. Cushman (1933, pp. 161, SI) states that (’filcitornella occurs in the Peim.sylvanian and the Permian, but that Hypcramminoides is known only from the Pennsylvanian. Cnshman and Waters, however, described at least 'one species of Hyperani- mmotdes, H. glabra, from tho upper part of the Clraham formation of Texa.s, U.S.A. Raggatt and Fleiohor (1937, p. 179) state that, while this was origin- ally placed in the Pennsylvanian, it is now, they undm-stand, considered to be Lower Permian. Arenaceous foraminiforal faunas of tho typo occurring in tlio Wandagee beds are in present day seas charactertstio of cold water, the genus Hyperam- mina, cvhich is .so common at W andagee, is a.s a Recent genus widely distributed, but is most abunilantly represented in cool w'aters, tho temperature of the ivater having more control than the depth. Tho related, but now extinct Hyperamrmnoides probably occurred under .similar conditions. Crithionina IS also particularly characteristic of cold wat(!r. Tho impure limc'stones and calcareous fmo-gniined sandstones in which these and the other foraminifera occur in tho Wandagee beds indicate .seas of no great depth, so it can be assumrTl that they represent a cool, moderately- shallow water facies. The types of tlu' uew species described ami other figured specimens have been deposited in the collection of the Oeology Department, University of Western Australia. Examples of most of tho si)ecics will also be lodged in the 102 Walter J. Parr. Commonwealth Palaeontological Collection, Canberra, the Western Australian Museum, Perth, the Heron-AJlen and Earland Collection in the British Mriseum, the Cushman Collection, the New Zealand Geological Surve}^ Museum, and in the writer’s own collection. ACKNOWLB DGMBNTS. The writer gratefully ackjiowledges his indebtedness to Mr. Henry Coley and to Dr. C'lirt Teichert for the material examined. Ho also thanks Dr. Teichert for his continued assistance in the preparation of the paper. Miss Irene Crespin, B.A., the CoTumonwealth Palaeontologist, has kindly furnished copies of literatiire M'liicli we 'e uot available in Melbourne and also supplied specimens fi*om the Cominonwealth Collection for comparison. The photo- graphs illustrating several of tht‘ species have, through the courtesy of Pro- fessor M \y. 8k(^ats, been taktai by ^Tr. J. S. Mann in the Geology Department of the Hni\‘ersity of Melbourrui. DESCRIPTION OF THE SPECIES. Phylum PROTOZOA. Order FORAMINIFERA. FAMILY AMMODISCIDAE. SUB-FAMILY AMMODISCINAE. Genus AMMODISCUS Hcnss, 1861. Ammolissus wanla^eesnsis, sp. now Plato IF., fig. 1. Test large, free, planospiral, composed of a small globular prolocidus, followed by a long undivided tubular chamber closely coiled in a single plane, the tube fairly thick-walled, almost circular in section and slowly increasing in diameter as it lengtheins ; number of whorls usually 6 or 7 ; spiral suture strongly depressed ; wall coarsely arenaceous with a rough surface and little visible cement : aperture formed by the rounded open end of the chamber. Diameter of test usually about 6 mm. ; thickness, 0.6 nun. Holotype from SaTupk^ 6 {Jjingida beds). The exceptionally !arg(' si/.(‘ (d' tin’s species (6 mm.) makes it a conspicuous object wherever it occurs. Palaeozoic snecies A mmodisc}iS are usually small and the writer is not ac()uaint(Ml with any other with a diameter of more than 1 mm. A. tvandageecnsis inay bo compared with A. ser:»iconst rictus, var. regidaris Waters, from the l^ennt-yK anian of Oklahoma and Texas, U.S.A. The American form is less than oiu'-sixth of the size of the Western Austialian species, which also difl'ors in its use of broken sponge spicules with tjuartz sand to fonn its test, and in its proportionately smaller proloculus. It is sel- dom that speci(*s of Ann-.odiscus inemporate spicules in the shell wall, but in A, wandugcccnsis they form a large part of the material iis('d. A peculiar feature of many of the specimens is that the spicules ha\'e been leached out leaving numbers of short cylindrical cavities in varying planes in the shell wall. The occurrence of siliceous replacements of molluscan shells has already been mentioned ; possibly the silicia ^vas derived from these spicules. Foraminifera and a TtiBicoLOtis Worm from the Permian of the North-West Division of Western Australia. 103 Ammodiscus nitidus, sp. nov. Plate T., figs, la, h. Test small, free, planospiral, composed of a minute proloculus and an ■elongate tubular chamber, almost semi-csrcular in section and slowly increasing in diameter, often transversely ridged, number of whorls usually 7 or 8, each whorl overlapping to a considerable extent its precedessor ; spiral sutiire only slightly depressed ; wall thin for the genus, composed of very small f[uart>! grains with little visible cement ; shell suiTace comparatively smooth ; aper- ture semicircular, formerin. but very small and delicate, and it apjjears that these represent the early stages of the larger exam]des. Sometimes they are slightly curved in the' earliest portion, but, with one exception, they are also incomplete. The excep- tion, which is figm'ed, shows a minute ovoid proloculus and is possiblv the megalosphenc form of the s]jecies. Another of these specimens, whicli is. also figured with the same magnification, is extremely slender and apparently represents the micros pi leric form. 106 Walter J. Parr. Several speeies of Hypemmmhioides have been described from tlie Penn- sylvanian of U.S.A, Of tliese, H. proleus (Cusliman and Waters) described (Cushman and Waters, 1928, p. 36, i)l. iv, figs. 5, 6) from Texas, most nearly resembles H. ac.ictiUu but has a different outline, particularly in the early stages, and attains a lengtli of only 2 • 25 mm. Prior to the receipt of the Wandagee material by the wTiter, Miss Irene Crespin, B.A., the Commonwealth Palaeontologist, had identified the genus H yperamtninoides in beds, [)robably of the same age as the U’andagee beds, in the Permian of New South Whales, and also in material collected by Dr. Arthur Wade and now in the collection of tlie Cieology Depanment of the University of Western Australia, from the West ICimberley District, C!rant Range, south section, just north of Hill 6. The Grant Range specimen has been kindly loanetl to me and the species proves to be H. adcnla. The ex- amples of Hyperainmnioides from New South W'ales are from shales and are not sufficiently well preserved to enable a satisfactory specific determination. They liave been recorded and figured, in a recent paper by Miss Crespin and the present vTiter (1941. p. 301, pi. xii, figs. 4, 5), as H. cf. proteus (Cusliman and Waters). Better specimens may prove them to belong to the [iresent species. Dr. Teichert has forwarded six selected specimens of a species of Hyper- arnminoides in thin-bedded micaceous sandstone from beds of the Byro Stage on the Gascoyne River, at Jimba Jimba Homestead. These are referred to H. acicula. They are all broken anil must hav'e been of unusual size when complete, as one specimen in its present state is 16 mm. in length. FAMILY SACCAMMINIDAE. SL B-FAMTLY PSAMMOSPHAERINAE. Genius PSAMMOSPHAERA F. K. Sehnlze. 1875. Psammosphaera pusilla, sp. nov. Plate I., figs. 6, 7. Test very small, spherical, consisting of a single chamber ; wall com- paratively thick, formed of fine ipiartz grains, very firmly cemented, the cement apparently siliceous ; surface smoothly finished. Diameter of holotype (an average specimen), 0-45 mm. Holotypo from Samjde 6 (Lingxda beds). There are twelve examples of this species. The strengtli of the shell wall is, for such a small organism, remarkable. Considerable pressure with the point of a needle is necessary before the test can be broken. P. cava, described by Moreman (1930, p. 48, pi, vi., fig. 12) from the Lower Palaeozoic of U.S.A., resembles the present species in external characters but has a tliin wall ; its diameter is 0-5 mm. Moreman states that P, cava was very abundant in the Chinmey Hill limestone. He remarks that it is very close to P. papyracea (Cushman), but that the latter has a thinner wall and is about twice as large. Foraminifera and a Tt^bicolous Worm from the Permian of the North-West Division of Western Australia. SUD-FAMJLV SACCAMMININAE. (Fnuis THURAMMINA H. K. Brady, 1879. Thurammina papillata H. B. Brady. Plato r., fi^. 8. TJmrmnmina papillata H. B. Brady, 1879, [>. 45, p\. v., figs. 4-8 ; 1884, p. 321, pi. xxxvi., figs. 7-18. Cliapman and Ilowc-hin, 1905, p. 9, pi. it., fig. 13. Heroii-Alleii and Farland, 1917, pp. 530 537. Moronian, 1930, p, 51, pi. V., fig. 13. Test spherical, consisting of a single iindi\ ie mort^ coa-rst'ly built geiuM’ally than Ht'cent specimens and also vary mort^ in form than do the s|)eci rntais found in any single Recent (Irfaiging. 1 teron-Alh'n and Farland [loc. rit.) havty howtwer, shown that 7\ papHlaia sai'ies ('onsiderably liotli in sluipc'. and in tlu> si/.e^ of the matoi’iaJ used in tla^ (‘oustruction of tlu* sludl wall. Tlu'* WandagtH^ s])eci- mens appear to fall within tiu' limits of thes(^ variations and are accordingly referred to Brady's sp('ci(‘s. One WandagtH' sjiecinuui has th(^ surface i)roJ(>ctions (extended in tlio form of comparatively long tubes, at the end of wliicli the aperturt's ans situated. Brady has figurtal a similar Ih^cent s[)tH*im(ai in the Ohalhaiger Report (j)l. xxxvi., fig. 14). While T. papillata is best known as a Recent form, it has been recorded by Chapman and Ilowchin from tlu' Permian of New South Wales and by Morernan from the Silurian .of U.S.A. FAMILY ASTRORHIZIDAE. Genus CRITHIONINA Goes, IS94. Crithionina teicherti, sj). uov. Plato L, figs. 9, 19. Test free, comparatively larg(a spherical, thick-walled, but variable in this respect, central cavity large, conneid-ed with tiui outside surface by numeious mod('ratt4y-sized pits which extend iiregulaiiy through the thickness of the shell wall and roach the exterior surface through minute o]ienings ; wall comjiostxl of line sandy material fairly w('ll cemtuited, with a thin compact surfaces layer. Dianu'ter of holotype, 2* 2 mm., other s])ecim(nis an* slightly larger. Holotype from Sample 7 (Ijltujula- beds). As far as I am aware, tlam^ is oidy one previous record, tliat of Moreman from the Silurian of U.S.A. , of this genus as a fossil. It is thorefor(^ specially intere^sting to moot with it in such numbi'rs in the l-*ermian of Australia. The specimens are oxipiisitely prescu’vixl and every detail of structure can Vie studied as readily as in the best Recent material. In the living condition, Crithionina is Ijost developed in cold waters, particularly in the North Atlantic, where it occurs in immense numbers. 108 Walter J. PAitR. FAMILY OPHTHALMIDIIDAE. SUli-FAMlLY CORNUSPIRINAE. Genus CALCITORNELLA Cushman and Waters, 1928. Calcitornella Stephens! (Howchin). Corrvuspim, sp. iiov. Jontis, 1882, p. (>. Nuhecalaria hccijuga l)(dranco, var. stephensi Howchin, 1894, p. 345, pi. xa,. xin sovoj'al internal c-asts fourul adherent to the tests of sandy species- after the calcareous inat(U'ial in the rock had been dissolv'od by treatment with hydrochIori(! acid. The casts show the arrangement of the tubular chamber found in C. ste]>k(msi, (J. Me/phensi was described from the Permian of Tasmania and has since been recorded from beds of this epoch in New South Wales, Western Aus- tralia, and Northern Territory. FAMILY LITUOLIDAE. SUH- FAMILY HAPLOPHRAGMIINAE. Goiius AMMOBACULITES (‘ushiuan, 1910. Ammobaculites woolnoughi Crt'sjrin and Parr. Plato L, fig. 11. ■ An}nK)b(tc}dites woolnoughl Cres|)in and Parr-, 1941, p. 304, pi. xii., figs, 2, 3. 'rhe.n? are two specinums whiclr apjx^ar to be I'iTerahle to this species. The typ(‘s ai’o from the Pcs’iniaii, Farley Koarl, 300 yards north-east of Farley Railway ‘Station, N(e.v Soutli Wales. The Wandagee sjreeimens, from the Lingula Ix^ds, are smaller (0*5 mm.) and built of (;oai*sei' material than those from New South Wales, hut an^ of similar form. Tire latter an^ from shales and this prohalrly actrounts for- tiu' finer-grained, rnon' smootlily finished test. FAMILY REOPHACIDAE. SUH-FAMILY REOPHACINAE. Genus REOPHAX Moiitfort, 1808. Reophax subasper, sp. nov. Plat(5 L, fig. 12. Test eonsisting of up to 7 cliambrrs, rapidly increasing in size as added, eai'ly cliambers indistinct, nsually ai'i'angfxl in a cui‘V(hI series, later chambers larger and more distiiu;! , in 7ioarly a straight lino ; wall built of coarse ((uartz grains, firmly cemented, snrfa<-o rough ; aperture a small opening between tliree or more sand grains at the (uid of the final chamber. Length of holotype, 1mm.; breadth, 0-4 mm. Holotype from Sample 8 (Lingidu Ix'ds). FoHAMINIFKRA and a FuhICOLOUS W'oRM FItOM rUF I*KRMIAN OF TIIF Nohth-Wkst Division of Wfstkkn Austrafia. iOi) Thoro aro niia^tcu'n (^xain[)lt>s of tliis s|)(‘cirs. 'riu> most (*los(‘ly r<'lato(l Palaeozoic species is probably h*. a^'ipcr ((’uslinmii and \Vat<>rs, ll)2d ; ap('rt ut‘(> baaninal. J^en^t'h of h()lotyp(\ 1 • 7 mm. ; bn'adth, I mm. Holo(yp(‘ from SampN^ (> (LimjnUi beds). Tliis spt'cies is rt'[U‘es(ait'ed by tlu^ holoty]x> from Sa.m[)U' (> and six cruslaHl specimens from the W(‘ather(>d sliale sainpl(‘ (No. D). It appcxirs to b<‘ diftorent from any previousIy-d(\s(!ribi^d form. PAMIIA TROCHAIVIMINIDAE. SUB KAM I LV TROCHAMMININAE. (;ciius TROCHAMMINA Pnrkt‘r nnd Jones, ISo!). Trochammina subobtusa. sp. nov. I’late I,, figs, l+u -r. Test snbglo})ose, trochoid, s])ire low, und<‘r sid(‘ uml)illcaJ(>, <‘ompos('d of three whorls ; chambors str'ongly inflated, with four in the last-fornnxl whorl, the chambers of which an^ so much larger t han IJios(‘ in the preccxling whorls that they form tlu' gi*ea,tei' })art of the test ; sutures distinct and d by six sjKH'imens, all of whi(*h are v(>ry v\(JI preserved. Trochaituninn o/vokm’u ('nsbman and Waters ( M)27a, 152, pi. xxvii., tigs. 4a c), descritxxl from tlio Ptainsy Ivanian of O’exas, U.S.A., shows a similar arrangiaruait of tlu^ (Ojambers, but- is nuuJi tlattoiu'd and spr(xiding and the wall is ratb(>r eoarsfJy arenaeeous. Phylum ANNELIDA. (lass (DAETOIDDA. Suh-(nass POFAC HAETA. ORDER C’RYPTOUElfHALA (SEDENJ’AKIA OR 1TilR( OBA). Amphictene U) permiana, si), nov. Plate IT., tig. 5. Tube tliin, (Jrcnlar in soiTion, tapering slightly, foinied of fragments of siliceous sponge spicules arrangtul coiuxaitrically in a single layer. I^engtli of holotype (imnmpiete specimen), It -5 mm. ; diameter, l-l mm. no Walter J. Parr. Holotypt^ from Sample 6 {Lingula beflw). This species is represented by a single specimen. Its identification has been a matter of some difficulty, but after an examination of the available literature, it appears that it is a tubicolous worm, forming a tube like the Re(ient Aniphictme auricoma (O. K. Miiller). No foraminifer known constructs a similar tost. A. auriconm occurs in tlie North Atlantic and normally constructs its long, thin-walled, tapering, slightly curved tube of sand grains. On muddy bottoms (McIntosh, 1894 and 1922), broken sponge spicules are used instead of the sand grains, the spicule fragments being so arranged that those in one row alternate with those in the next, like the bonding of bricks in a wall, to form the strongest tube possible. A minimum of cement is used to hold the spicules together and the r(!sultant tube is thin-walled and most neatly constructed and finished. Nothing showing a wall .structure like that found in the spicular tubes of A. auricoma ap])ears to have been recorded as a fossil and the present speci- men from the I^orTuian is tliorefore of unusual interest. While it also consists of a similar layer oF siaeular fragments, comparison with the figures of A, auricoma given by McIntosh (loc, ciL) and by Heron-Alien and Earland (1909, pi. XXXV., fig. 14) and with examples of the Hecent species for which I am indebted to my frimd, Mr. Arthur Earland, F.R.M.8.. show.s the fossil form to b(^ flistinct, being less neatly built and also smaller than A. auricoma. It should be add(Hl tliat, while this comyjarison with A. aiivicoina has been made, the two forms may bo unn4ated although apparcaitly similar ; the Pt^rmian species has accordingly b(jen doubtfully referred to the genus Amphictene. REFERENCES. Brady, H. B. : 1879 : Notes on some of the Retieulariaa Bhizopoda of the “ Challenger ” Expedition. Quart. Jovrn. Micr. Sci. (London), XIX., pp. 20-63, pis. iii-v. 1884: Report on the Foraminifera dredged hy H.M.vS. “Challenger” during the years 1873—1876. Rep. Voy, ChaUenger, /oology, vol. ix. Chapman, F., and l lowchin, 1905 : A Monograph of the t’oraminifcra of the Bermo- earl)oniferoiis Limestone of New Mouth Wales. Mem. deoL N.S.W., Pal. No, 14, pp. 1-22, pis. i-iv. Cha)>man, F., Howehin, W., and Parr, W. .L. 1934: A Revision of the Nomcnehiture of the Permian Foraminifera of New Sontli Wales. Pror. Boy. Sor. Vlr., XLVTl. (N.S.), pt. 1, pp. 175-189. Crespin, I., and Parr, W. J., 1941 : Arenaeeous Foraminifera from the Permian Rocks of New Mouth Wales. Jour. Proc. Boy. Bor. AC.SMf., LXXIV., pp. 300-311, pis. xii, xiii. Cushman, .R A., 1933 : Foraminifera. Their (dassifieation and Economic Use. Mecond Edition. Cushman, J. and Wate?'s, .1. A.: 1927 : Pennsylvanian Foraminifera from Michigan. Contr. Cushman Lab., III., pp. 107-110, pi. xxii, 1927a : An^naecous Pahuio/.oie Foraminifera from Texas. Ibid., III., pp. 146-153, })ls. xxvi, xxvii. 1928: Mome Foraminifera from the Pennsylvanian and Permian of Texas. Ibid.., IV., pp. 31-55, pis. iv-vii. Etheridge, R. (-Jun.) : 1907 : Descriptions of ('arbonih'rous Fossils from the Irwin River, collected hy Mr. C. F. V. Jackson, late Assistant CJovernment Geol()gist. West. Ansi. Ceol. Survey Bull. No. 27, p|). 26-37, pis. (Foraminifera on pi. x.) 1907a : Official Contrihutions to the Pa laeotitology of South Australia. Supplement to Parity. Paper No. 55 of 1906, p[>. I-2I, pis. i-xii. (Foraminifera on p. 13 and pi. xii.) Foraminifera and a Tubicodous Worm from the Permian of the North-West Division of Western Australia. Ill Heron-Alien, E. and Earlaiid, A. : 1009 : On a New Speeies of Techiiitella from the North Sea, etc. Jour. Quekett Micr, ('lub [2|, X., pp. 403-412, pis. xxxi-xxxv. 1917 : On some Poraminifcra from the North Sea, flredged by the Fisheries Cruiser Goldseeker” — ^V. On Thuraiumma papiUaia Brady ; A Study in Variation. 1917, pp. 530-0.77, pis. xxvi-xxx. Howchin, W., 1894 : On the Occurrence of Poraminifera in the Periuocarl>onifcrous Rocks of Tasmania. RepL Aust. Assocn. Afh\ aS'cl, Adelaide l\leeting (1893), pp. 344-348, pis. x, xi. Jones, T. R., 1882 : Catalogue of the Fossil Foraminifera in the British Museum (Natural History). London, 1882. McIntosh, W. C. : 1894 : On certain Homes or Tubes formed by Annelids. A7in. Mag. Nat. Hist. [6], xiii, pp. 1-18, text-figs. 1922 : A ilonograph of the British Marine Annfdids, vol. IV., pt. 1, London, Ray Society. Moreman, \\\ L., 1930 : Arenaceous Foraminifera from Ordovician and Silurian Lime- stones of Oklahoma. Jour. Pa?., IV., pp. 42-59, pis. v-vii. Paalzow, R., 1935 : Oie Foraminiferen im Zechsteiii des ostlichen Thuringen. Jahrb. Preuss. (Jed. Landesansiali, LVI., pj). 2fi-45, pis. iii-v. Raggatt, H. G., 1936 : Geology of the North-west Basin, Western Australia, with Par- ticular Reference to the Stratigraphy of the Permo-Carboniferons. Jour, and Proc, Roy. Noe. V.^ST.ir., LXX., pp. 100-174. Raggatt, H. G., and Fletcher, H. O., 1937 : A Contribution to the Permiau-Upper Carbon- iferous Problem and an Analysis of the Fauna of the Upjjer Palaeozoic (Permian) of Xorth-West Basin, Western Australia. Records Mas?. Mas., XX., pp. 150-184. Teichert, C., 1939 ; Recent Research Work in the Permian of Western Australia. Aust, Jour. Sci., II., No. 1, pp. 5-7. 112 Walter J. Parr. EXPLANATION OF PLATES, Platk I. Figs. Iff, ]b. Ammodiscus nitidus, sp. nov. Holotype. XoO. No. 20070. Fig. 2. Glomospira adhaerens, sp. nov. Holotype. x2r>. No. 20075. Fig. 3. Hyperammina (?) rudis, s}). nov. Holotype. x37. No. 20077. Figs. 4, 5. Hyperamminoides acicula, sp. nov. Early stages, apparently of microspheric form (Fig, 4) auH megalospherie form (Fig. OL Both X37. No, 20079. Figs. 0, 7. Psammosphaera pusilla, s]). nov. 0, holotype external view ; 7, internal aspect of another specimen. Both X37. No. 20080. Fig. 8. Tliiiram'inina papillata H. B. Brady. x43. No. 20081. Figs. 9, lo. Crithionina teicherti, sp. nov. 9, holotype, x 13 ; 10, internal aspect of broken specimen, showing wall structure, X13. No. 20074. Fig. 11. Ain'icohavalitfs U'oohiomjlt^ (Vespin and Parr. X50. No. 20009. Fig. 12. Reophax subasper, sp. nov. Holotype. x37. No. 20085. Fig. 13. Reophax tricameratus, sp. nov. Holotype. x20. No. 20084. Figs. I4ff !4c. Trochammina subobtusa, sp. no\^ Holotype. a, dorsal view ; b, ventral view ; c, edge view. Ail x75. No. 20083. Foraminifeea and a Tubicoloi:s Worm from the Permian of the North-West Division of Western At^straeia. 113 i 144/41 Plate I 114 Walter J. Parr. Plate IT. Fig. 1. Ammodiscus wandageeensis, sp. nov. Hoiotype. xl2. No. 20671. lig, 2, Tolypammina undulata, sp, nov. 4hree specimens adherent to a test of Ammo- discus wandageeensis. The hoiotype is the specimen just to the right of the extreme left hand sp(‘cimen. xi2. No. 20682. Fig. 2. Hyperammina coleyi, sp. nov. Holotyi>e. xll. No. 20676. Fig. 4. Hyperamminoides acicula, sp. nov. Hoiotype. xO.U No. 20678. Fig. 5. Amphictene permiana, sp. nov. Hoiotype. xI9. No. 20(i72. A’ :-T.i OF ViCTORJA Foraminifera and a Tubtcolotjs Worm from the Permian of the North-West Division of Western Atjstralia. 115 S Pi ATE 11. 117 The Seeds of Stkychnos Lucida, B.Br., and their Alk^veoid Content. 9. -THE SEEDS OE STRYCIINOS LUCWA, R.Br., AND THEIR ALKALOID CONTENT. By E. M. Watson. Read : 11th March, 1941 ; Published : loth May, 1942. I Strycfinos lUcichu R. Br., which is the only representative of the genus recorded in W'estern Australia, is recorded by Gardner (1) as being common in Central Kimberley. The author is indebted to Rev. J. R. B. Love, form- erly of Kunmunya Mission, for collecting the fruits used in this investigation. It is an erect shrub, b to 1 5 feet in height. It bears orange-yellow fruits which are globular or ovoid-glolmlar in shape, u}> to about^ 25 mms. m diameter and containing one or more seeds. Those available for this work were 12 to 15 mms. in diameter and were single seeded. The pericarp is brittle, hard and polished and is very sliglitly wrinkled and pitted ; it is about 0*5 mm. thick and is easily cracked. I’he [lulp, wliLch is milky-white in fresh fruits, dries to a dark tacky semi-solid which adheres to the seed. It is readily removed by soaking the seeds in cold water and then rubbing between the fingers. Tlie pulp has an intensely bitter taste and colour tests indicate tfiat it contains an ap]9rociable amount of brucine aiitl a much smaller amount of strychnine. The seeds are iypii^ally grey or greenisli-grey in coIoih, disc or Initton- sliaped, 10 to 13 mms. in diameter and about 3 mms. thick. They are gen- erally flat and regular in shape but ai'e occasLonally irregularly bent or oval in shape. The edge of the seed is rounded (or very occasionally acute) and shows a distinct raised inim’ojiyle wliicli is often connected by a ridge to the hilum. The testa is covered with ra.flially arranged, closely adpressed, silky, lignified hairs. TTie endosperm, which makes up practically the whole y Toakle (1930). Acacia aneura, its varieties and allied species, constitute the most important tree and shrub forms. The Malcolm area was surveyed geologically by Clarke (1925). Most of the (country studied during the two trips lay in the greenstone atv'as, all the transects being worketl in this type exce])t tliose across tjuartz “ blows.” According to the geological map ((.'lai*ke, 1925) tlu‘ area of spinifex sandplain seen on Glenorn lies over greenstone. Talbot ( 1920) considered that sandplain develop- ment could be eorrelatod wdtli tlie occurrence of granite. It is possible that the sand in this instance has been transported from gi'anite areas to the north and north-east. ECOLOGY. Melville recognises the following types of country :-- (!) Flood Plain, (2) Hills Scrub, (3) Shrub Steppe, (4) Sandplain. With the exception of the third section these hold for the Glenorn area. (1) Flood Plain. “ Flooded ” country is the local tcji’m used in referring to tlie vide flats across which di'ains the run off fi’om the low hills. Drainage is of two types. (a) Occasional well delined, but short, creeks which are shallow and nar- row. These soon lose their identity in (b). The course of the creeks can be followed, from some distance away, owing to the bright green foliage of the 120 NaN(’Y T. HTIRBII)(iF:. cr(M)k miin.4 {Eucalyptus caitialduicnsis), which are almost invariably associated with tluari. Other charact(M‘isti('. [)lants ar(5 jiimwood {Acacia, I-iurkiUi) and <;reek (‘assia {Cassia artetnisioklvs). 'riu3 botaiiit^al differences between (a) and (b) arts (juitts ma,rk(sd and arts ])ro})ably rtslattsd to the water supply factor. (b) W ide Hats where the wattsr (Itjws, as a sluset, atsross the almost level surfatMs. 'riiose Hats irichido most of the ttnil^a country that is pastorally valiarbks, all the btstter types of muljjca {Acacia amvura) bein^^ found in this habitat, 'riieso includt^ the “ Silvia’ bt^avtul ” nml^a and nn^diuni leavt^d forms. Curara bush {Acacia geuistoides) is characteristii^ of this ctnintiy tliough, where tlie creeks narge Into th(’ Hood j>lain pro])(‘r, curara may mingle with jarnwood for a shoi’t distance. Erciuophila LatmbH, E. Margarcthac and E. tnacidata o(^c\ir in the IndtiM' watcrcHl patcla^s. AH arc genmally n^ferred to as fuchsia buslies tlu)ugh this name is most commonly attached to E. uiaculata. This spiK’ies, though rega-rded \\'ith suspiiaon by some as a poison bush, is consid- ered quite useful f(‘ed on (llenoni. 'TIk' lirst named of tho three specios is the rriost vviilespread. Otlua* shrubs include Cassia erctnophila and, amongst the smallf'r foi’ins, Eurhylacfui (oua’ufosa, RJaujotlia. sp. AhutUmi crypiopcfalum. and A. olor(frpin}i. (Sne Transects III., IV'., Xll.) After the winter rains tfui flats carry an abundant crop of ephemerals, though at' other lina's the ground betwiH^ii trees and occasional shru])s is coinpl('t(Hy bai’e. '‘IHie ai’i^a covered by, and tho individual size of, those plants is strikingly related to tho amount of water available. By far the best dcvolopnu'nt. taints place on tin*, lower levels (3X(*ept where claypan form- ation, witli conseijueiU watia’-logging of thi' soil, has pi’i^ventod any growth whatever. Vh'mlxa’s of t he ( V)mp(>sit a«>, ( ’rncifera(>, C'lienopodiatH'ae and Amar- antaceae ar(> th(> most' important const itiaaits. 'IHio gi'owth resultiTig from sminner rains includi^s a number of annual gi’asses of which Aristida airnarlay Neurarhuc MitchetHana and Knneaiiogon cacrulcscrus aiM^ tho most- abundant. Tlu> clu(>f soil of tln^ Hood plain is a sandy-cla.y loam (Teakh', 193()) over- lying a silici'oiis hardpaii which rests on the country rock. 'Fhe surface of the hard[)an, wliicli is eluiraeleristic of the whok^ pastoral r(>gion, is variable so that' soil (l(q)1h may cfiangc from a few inches to two or throe feet within a few yai’ds. H'hi’oughout tlio Hood plaiii low I’idgcs of v’ariahio- extent and importance liave dmolopi'd. ][(^re sand ovi'i'lii^s th(3 hardpan to a eonsiderablo depth. HHicsi' an*, refern'd to us “ W'anila.rrii^ ” I'ountry and the voge?tatio?i differs markedly from that of the low«‘r Hats, 'riu* inulga nanains more or Ic^ss the Sana* though, so far as could i>e judged iu th(3 tun(‘ available, it was in a poorer condition, (’uiani is (unnpletely ahsimt and. is replaci'd by bogada bnsh Acacia hnu'hystachipi. It' may bi* noted her-e that this common name is used ill the Murchison ania to refen* to Acacia IhiophyUa^ while Acacia brachystachya is known as suga.r bi'oMu'r. A. liuophylla was not collected at (.Henorn. Amongst tlu' smaller shrubs poverty bush {Ercuurphila lcuct>phylla) hath. iiuiiua’ically and past'orally, imiierlanl as sIuh^) will f(‘(‘d on it when more palatahl(3 speck's disappi^ar during dry seasons. Aiiotlua* species, E. co}}tpacta also occurs hut is not so important on eitlua’ [)oint'. Both these' shrubs are (‘apable of st-a, titling (k)rmanl' for k)ng perioils. After I’ain th(*y respond with fresli growth within a short lime. Also the eoiulitions favouring gt'rminaiion tlo not stH'in (-0 l)e as exacting as in ttie case of .-Icao/a. aucura. In gooil seasons with satisfactory summer rains the soil carru>s an abundant growtJi of Wan- darrit' givissos which give the typt' of (‘ountry its name. The most important of tht'se art' Eragrostis vriopoda, E. setifolia and Erlachuc Hvlntsii, During Notes on the Vegetation of the North-Kastern Ooldfields. 121 di'y periofls these die back to tlieir woolly rlii/o^nos and become dormant so that the prround is bare between the shrubs. Less important ^jrassos are Danthonia blpdrtlta^ Nenrachne MitchelUanay Aristidd aretutria and Ermeapogon caerulescens. None of these latter grasses is restricted to any particular assoc- iation but they all prefer the deeper soils. (See Transects TT., VIII., XI.) During the August visit the poverty hushes had formtHl new growth and there was a number of young seedlings. The bogada had freshened up and some was in flower. The epheaneral g^’owth ^\'as very poor compared with that on the loamy flats. The following annuals wtu'e collected however : — Stenopeta- him pedicellare, S. robustum, Farietaria dehllis, HalorrJtngis odontoearpa, Velleia rosea, T’. cifcnopotarnica, Podolepis pallida, Cedotis midticmdis and C. hispidula. There was also a fair amount of Erodiuai cygnonim. If the VVandanle country be exc(^pt(Hl, the flood plain formation may be regarded as lying between the Hills ScTub anlatod to the soil and country rock and may be subdivided as follows : — (a) On greenstone outcrops and the n^sidual hills from the pre-existing plateau there is .a poor Acacia scrnb {Acacia Burkltd va‘.) which, iy low pro- portion of leaves to twigs. Henctr there is little top feed ” available for the sheep. Other trees, of which occasional specimens were soon, an^ kurrajong and sheoak (Casuarhia sp.). There arc few shrubs though poverty bush, bluebush and sagobnsh were scon. There was practically no e])li(!meral growth in this country in August. (3) yShrub Steppe. This country is locally Imowu as “ lake ” country. Providujg the rain has been sufliciently heavy the water draining from the higher ridges across the flats ultimately readies the so-called lakes. A portion of Lake Raesido extends across the south-western corner of the Glenorn property. The lake consists of loamy flats which may, after good rains, hold shallow pools of water from a few inches to a couple- of feet in depth. Mure fretpiently tlu' water is insuflicient to form surface pools. The soil from a dry flat was found to contain 1 ‘ 33*^0 NaCd. Loose sand, which sparkles with salt crystals, covtts tJie surface in some places and carries a sparse population of samphire. Outside tliis cen- tral area is a zone of very thick samphire (Artla'ocnenmm sp.) 1-2 feet liigh. All the samphire is valueless from a ])astoral point of view. It was not in 122 Nancy T. Burbidge. flower during either \ isit and was not studi(HL On its outer fringe the sam- phii'e association naages into tlu^ ef‘ [ilants of Fra)ik'(‘}iia fecunda and Frartkehia svtosa. Bassid d-'n'(n-ic
. wf're rare in normal edible shrub country. Trans('('t VII.) Wien seem in August the ])erennial shrubs wi'ri^ in good condition \\ith plenty of fri'sh growth ami a liea\-y taop of flowers. The pi'n^nnial slirub country is Ix'tti'r d(‘V('ioped on the north-i^astei'n side of th(‘ lake than on the sout li-western, w la'ro low ritlges carry wandarj-ie plants and tlie flats are normal flood tilain. This supports the theory regarding the tendency of these Iak(*s to migrate in a south-westerly direction. Of sjieeial interest in regard to this tfieory was the di.scovery of an irregular zone of mixed country carrying tht; plants of the flood plain association interpersed with tlie (xlible shrubs. If the lake has migrated, this zone may' well represent the intrusion of mulga and curura into th(^ siirub steppe. The peculiar balanced association of tlu* two tyj^es cannot bo explained by soil character or a diff(n’('nc(> in water siqiply. (St'C Transects V. and X.) Currant bush (ScaevoJa spinescens) was fairly common in this mixed country. This shrub, tliough reported as being able to survh'e heavy stocking (McTaggai-t, 193()), is not n^garded locally as a useful plant. (4) Sandplain. This typ(> of coimtry is not well (levelopi'tl on Ckaiorn. There are some small ar-('as of spinifex in the si'Ction of the station north of Mount Morgans. Therc^ does not seem to be any well defined zone of spinifex. The area seen covered about a square mile, and was sur-rounded by flood plain and hills scrub formations. The soil of the sandplain is lighter in colour than that that of tlu^ v*andari‘io areas. No information was obtained concerning the existence of hardpan in this country. The spinifex association is very distinct botanically. TrMia Basedowii is dominant. No other grass, mmiial or perennial, was seen. A low shrub form of AcacAa hvachydachgd- -smaller than the bogada of the wandarrie Notes on the A^eoetation of the Nohth-Kastkkn (Joldfields. 123 country — and an oceasi(.)nal specimen of a myrtaceous plant, wliich was not found in flower but is probably a species of Melaleuca, occur fairly coni' monly amongst tbo spinifex plants. The only tree form is an occasional group of mallee of which two si^ecies arc present. One of these is Eucalyptus pyriformis, but the other coiiM not be found in either dower or fruit and was not identified. (See Transec't T.) When seen in August the aca(*ia carried a few flowers, and the spinifex a few seed-heads, tliough there was practically no jiew growth. The ephomorals were almost completely absent. Where the pure spinifex merges into the mulga (flood plain association), the Acacia and mallee stop abruptly, but the spinifex mingles with the mulga formation for some distance. The plants of this transition zone, both the mulga and the spinifex, appear to be in bettor general condition than those in the purer associations. ECOLOGICAL TRAXiSEC'TS. Method. — The first transect was of the line type and ran for 200 yards. The remainder were strips 200 yards long by 24 feet ^^'ide, and were worked in squares of 24 feet sides. Soil depth was, in most cases, tested every 24 yards, i.e., every third square. The distance from the surfacf? to the siliceous hardpan was measured by a post hole auger. It was necessary to use water to soften the simbaked loam so no deep samples could be taken. Surface samples were obtained, however, and these have alreadv beim piiblishcHl (Teakle, 1938). Transect /. On spinifex sandplain north of Mount Morgans. This transect was of the line type. The following plants weii'. encountered in 200 yards ; — Triodia Basedoivii (spinifex) .... .... .... .... 103 Acacia brachystachya, .... .... .... .... .... 54 Melaleuca sp .... .... .... .... .... 2 Eucalyptus pyriforrnis (mallee) .... .... .... 2 Transect II. The transect ran from a sandy wandarrie ridge across the ('dge of a clay- pan. The first hundred yards ^vas marked by the number of saltbush j>lants. The rest of the trans(T*t was strongly influenced by the elaypan. 1st 100 2ml 100 Acaciri aneura. (iMulga) ; yards. yards. M(Hhum-Jeaved tree 5 1 Short -leaved pine typo 0 0 Dead tree.s 1 17 Atriplex hymenotheca (saltbush) .... 30 4 Waiiflarrio grass (dormant) 31 0 Ercufophila leucophylla (poverty bush) 3 0 Ehagodia sp. 2 1 Kochia sediiolia> (bluebush) 0 3 Koch in irillosa (glabrous type) .... 0 1 D{>a(l shrubs 3 30 124 Nancy T. Bttrbidge. Transects III. and IV. Tranyect III. was in line with biit in the opposite direction to Transect IV,, so that they give a strip 400 yards in length. All but 48 yards of this distance was in t;\ 7 )ical flood j:>Iain country, tlie mulga and curara bush when considered together averaging about one and a lialf plants per 64 scpiare yards. The ground was completely bare. Accordiirg to Melville the miilga association is a closed one. Fn the last 48 yards sand cov(a’('d the loam of the flood plain and there was an abrupt change into wandarrit^ contlitions. Tlie wandarrie grass in the list l)elow refers to tlu* doi-Tuant stumps only. It was impossible to tell liow many of these were capable of regrowth, so all were counted. This principh^ was adhered to in all transects. 1st 352 last 48 Acacia aneura : yds. yds. Medium-leaved tn'e 38 3 Narrow-leaved tr-ee with pine habit .... 5 0 Broad-leav(Hl inedibU' shrub .... 1 0 T)cad 42 1 Acacia gcnistoidcs (curara) ' 17 0 Acacia hrachystachya (bogada) .... 1 6 Eronophila IcHCOpJiylla (pox'erty ) 4 3 Ere/niopfiila loinji folia 1 0 Ercniophila sj>. . , 4 0 Abutilon otocarpnnt 2 0 Wandarrie grassi's .... 0 113 The following soil d('pths were recorded : — ■ ni.-IV. 12", 15", 11", 17", 28", 30", 11", 19", 10", 12", 17", 21", 19", 39", 15", 18", 14", 18". (The last three on wandarrie ridge.) Transects T. and X. These transects were worked parallel to one another and within a hundred yards. The ({uantity of edible shrubs, mixed with flood plain plants, is in marked contrast to the two previous transects. Wandarrie grass appeared ■wherever the surface became sandy. Blsewhere clay-loam. the soil was a sunbaked Acacia aneura : V. X. Medium-leaved tree 10 10 Broad-leaved tree 1 1 Broad-leaved shru]> 2 7 Dead 20 11 Acacia geyiistoides .... 5 8 KocJiia sedifolia (bluel)ush) 1 1 15 Kochia pyrainidalt (sagobush) 2 9 Kocliift villosa (glabrous type) .... 5 5 Afriple.r. Inpnenotheca (saltbush) .... 16 19 Hhagodia sp. .. . 16 6 I^ycium australe (waterbush) 2 9 Scaevola spinescens (curraiitbush) 8 8 I^olanuni orhiculaluni 8 0 Cassia eremophiht .... 6 4 Eremophila leucophylla (]K)verty) 1 0 W-andarrie grasses .... 123 57 Dead shrubs 29 30 Notes on the Vegetation of the North-Eastern Goldfields. 125 The following soil depths were recorded along the two transects : V. 13", 8", 8", 18", 21", 25", 27", 15", 20". X. 9", 9", 12", 14", 11", 12", 11", 11", 12". Tixmsects VL and IX. Botli these were worked in " lake ” countr\' and were within half a mile of each other. The transects lines were roiighiy parallel to the long axis of Lake Raeside so that both dealt with plants of the edible shrub zone only. The area chosen lay to the nortli of tlie lake. VL IX Kochia pjpxnnidata (sagobush) .... 3 95 Kochijf sedijolia (bluebush) 1 4 Atrlplejc hipnenotheca (saltbusli) 79 97 lihagodia sp. T 14 15 H])- II 29 30 Cratystyiis conocephahi (sage) 69 17 Lychan ausU'ale (w^aterbush) 13 88 EremopJiila miniata 0 3 Acacia aneiira 0 2 The following soil depths were recorded : — VL 28", 28", 31", <52", <52", 22", 14", 17", 18". IX. 10", 28", 21", 9", 13", 10", 9", 4", 4". Transect VII. This transect area was reported to have formerly carried edible slirubs. Most of the few shrubs left were dead, owing to drought or overstocking or possibly both. Atriplex Jrymenotheca 1 Atriplex halijnoides 6 Ifhayodia sp. 27 B a ssi a d. iva ri cata .... 6 Frankenia fexvnda 70 Frankenia sclosa 12 Cratystylis conocephala 6 Dead slu*ubs 104 (all very small) Xo soil dejjth tests were made in this area. Transects- VIII. a'tnl XL The wandarrie country was in poor condition, most of the perennials looking shru-elled and browned. Many of the poverty bushes lacked leaves M'hen seen in April but shewed fresh growth in August. It is often difficult to state with certainty w'hich plants are dead and which still capable of re- covering. The tw^o transects w'ere mapped within a hundred yards of each other and parallel. 126 Nancy T. Burbidge. Acacia aneura : VIII. XI Medium leaved tree 15 4 Narrow km-ved with pine habit 7 - 9 Dead 5 ■ 5 Acacia brachystachya 26 17 Eremophila leucophylla 56 63 Cassia eremophila' .... 0 4 Rhagodki sp. 2 11 Eolamint orbictdatinn 1 0 Wandarrie grass .... 328 380 Dead shrubs 10 15 The following soil depths were recorded : — VIIL 37% 371% 27% 28% 28% 36% 43% 42% 28% XT, <52", Co2", C52", <52", 39", 50", 18", 15", 26|". Transect XII. This was mapped in country carrying mulga of a broader leaved form. This type is known locally as Silver Leaved mulga. It retains in the adult leaf the remains of the bladdery hairs found on the young leaves of all varieties. The character is most marked in new growth and on the leaves of trees growing near water. ® The presence of a number of poverty bushes— a species ^A'hich in general prefers sandy situations— was the most interesthig feature of tlie transect. Acacia anenra : Silver leaved tree Medium leaved tree Dead Acacia genistoides .... 47 1 12 5 Eremoph Ha leucophylla 21 Eremophila sp. 2 Dead shrubs 7 The following soil depths were recorded : — ■ NIL 26% 17% 7i% 20% 8% 16% 12% 13% 9^% Transect XIII, The transect lino ran down a low slope below a quartz blow. The sod was littered vlth fragments of quartz. The slope ultimately became a samphire flat. Similar country nearby had formeily carried sage, sagobush and blue- bush but it had suffered badly in the bad seasons and practically all the plants were dead. The area selected did not represent the worst, though a very high proportion of the shr-ubs were dead. Kochia sedifoUa Korkia pyra }n idata Crafystylis cojiocephala Frajikenia. feennda Dead shrubs Eremophila scojmria (alive) .5 (dead) Hakea recurva 70 17 4 18 128 8 4 1 Notes on the Vegetation oe the North-Eastern Goldfields. 127 Soil deptlis wcu-o tested at distances of 48 yards along the transect. Tho quartz fragments in the soil of tliis transect and the following one made the work with tlie auger difficult. As a result full reliance cannot, unfortunately, be placed in the ligures recorded. XIII. 35", 35", 9", 6", 21". Trmisect XIV. The area chosen was the side of a stony myall ridge formed by a quartz blow. On the lower levels the ridge became a loamy flat which carried all the sagobush plants counted during this survey. As in the previous transect the great majority of the shrubs wore dead. Acacia aneura (medium leaved tree) Acacia sp. (myall) Cassia eremophila .... Kochia sedifolia Kochia pyramidata Dead shrubs The soil depth tests resulted as follows : — XIV. 13", 9", 9", 9", 35", 12", 7", 10", 12". 3 10 4 6 32 137 ACKNOWLEDGMENTS. Acknowledgments are made to Professor J. E. Nichols, formerly of the Institute of Agriculture, and to tho Pastoralists’ Association of Western Australia for their interest and assistance by obtaining a grant from the Pastoral Research Trust of Australia, to Messrs. A. L. R. Lefroy, C. Foulkes Taylor arid W. M. Oleland for their courtesy and hospitality at Glenorn and to tho Director of the Royal Rotanic Gardens, Kew, England, for facilities for ex- amination of the material collected. REFERENCES. Clarl^, E. (ie C., 192.) : The Field Geology and Broader Mining Features of the Leonora- Dukcton District. Iain, I = Spinifex sandplain. Marsileaceae: Marsilea hirsuta H. Br. ... Nardoo Aug. B Gramineae: Neurachtw. MiidielUana Xees Ap. Aug. . . . B, C Pasjiali3 • • • Aug. B, D A lopev u ntfi a^wiralis Nees Aug. B Eriachne pulchella Domin. ... ... , , , Aug. B Erim-.hne IldmMi Domin Broad-leaved Wan- Ap. Aug. . . . C (larrie Danthonia hipariita F. v. M. • • • • • . « • • Aug. B, C Enmapoyon caeruUscens (Gaud.) • • • « • • ... Ap. Aug. . . . B, C, D N.T.B. Triodia Bamlowii IMtzel Spinifex Aug. I Eragroslis eriopoda Benth. Narrow-leaved Wan- Ap. Aug. C darrie Erayrodis aetifolia Nees 33 33 Ap. Aug. . . . C Erayrofsfis leptocarpa Benth. , , , , , , , , , Ap. Aug. . . . B Erayrodis IHelsU Pjiger Ap. Aug. ... B, 1), G, Cyperaceae : Cy perns Iria L ... ... ... Ap A Scirpns mnlticulmis F. v. M. Ap. Aug. ... A Casuarinaceae : Cas^iarina sp. 1. Sheoak F M sp. II. 33 • • • • • • ... G Uriicac.eae : Parietaria debilis Forst ... Aug. C Proteaceae : Hakea recur va Heissn. Needle bush D, E Hakea lorea K. Br. Corkwood ... Aug. B Grevillea sp. (flowers only) • • * • • • • • • Aug. A Grevillea sp (follicles only) Water tree Ap B Santalaceae : Exocarpus aphylla K. Br. • • • • • • • • • Ap C, H Santahmi spicatum (R. Br.) D. C. Sandalwood Ap. Aug. ... A, B, H Santahmi lanceolatum R. Br. 33 • • • Ap. Aug. ... B, D Notes on the Vegetation oe the North-Eastern Goldfields. 129 Lorantha^eae : Loranthns gibhendus Tate Loranthus Murray/ F. v. M. et. Tate Loranthus Mitchellianus ... Loranthus penduhis Sieb ... Loranthus Quandong Lindl. Mistletoe on Acacia Ap. Aug. ... B aneura Mistletoe on curara Ap. Aug. ... B Mistletoe on curara Ap. Aug. ... B and mulga Mistletoe on willow Ap. Aug. ... B, G Mistletoe on Silver- Ap. Aug, ... B leaved inulfra -Polygonaceae : Muehlenheckia Cunnhiqhamii F. v. :\L Lignum Ap. Aug. •Chenopodiaceae : Rhagodia spp. Cheno'podium BlaHiauim, Aellen. Chenopodiurn myriocephalum Aellen Chenopodium cristaium. F. v. M. ... Chenopodmm m€.l.an(Karpuw. Aellen. Atriplejc hymenotheca Moq. Atriplex semilunaris Aellen. Atriplex halimoides Liiull. Atriplex spongiosum K. v. M. ... Atriplex nmtrmularia Lindl. Bassia sclerolaenoides F. v. ]\I. ... Bassia paxadoxa (R. Br.) F. v. M. Bassia dirarkata (K. Br.) F. v. M. Bassia eriacanlhn (F. v. M.) An- derson Bassia uniflora F* v. ^I. Bassia densiflora Fitz. Kochia (Jeorgei .Diels. Kochia Iriptera Benth. Kochia ayu-oena Diels KQchia- 2 )yramidata Btuith Kochia rillosa var. Kochia sedifolia F. v, M. Kochia glomerifolia F. v. ]\I. et. Tate Kochia carnosa (Moq.) Anderson Enchylaena tornentosa H. Br. Arthrocnemuyn sp. ... Balsola Kali. L. Amarantaceae : Trichinium obovatiirn Gaud Trichinium alopecuroUhum, Lindl. Trichmiuni heUpteroides F. v. M. Trichmitim corymbosum F. v. M. Ptilotus rillosifiorus F, v. M. Alternanthera deMtkulata R. Br. ... A izoaceae : Trianthema crystallina Vahl. Tetragonia. expansa Murr. ... Spinach Mesembryanthemum. aiistrale Soland Pigfaee Saltbush Annual Saltbush Old Man Saltbush Sago bush ... Bluebush ... Samphire . . . Roly poly ... Cotton bush Portulacaceae : Calandrinia balonensis Lindl Rcmunculaceae : Ranunculus peyitayidrus J. M. Black Parakeelya . . ... G, K Aug. ... B Aug. ... B Aug. ... B Aug. ... B Aug. ... a H Aug. ... B, G Ap. Aug. ... B, G, H Ap. Aug. ... B, a 11 Ap. ... ' ... G, H Aug. ... B Aug. ... B Ap. Aug. ... B, G, H Aug. ... B A]). Aug. ... B Aug. ... B Aug. ... B, G H Aug. ... B, C Aug. ... B Aug. ... B, E, G, Ap. Aug. ... B, C, H Ap. Aug. ... E, G, H Aug. ... E Aug. ... B Ap. Aug. ... B, C, G, • • . ... G Aug. ... B, G Aug. ... B, C Aug. ... B Aug. ... B, C Aug. ... B Aug. ... B Ap. Aug. ... B, C Ap. Aug. Aug. Aug. Aug. Aug. B, G B G B, C, G, H B 130 Nancy T, Burbidge. CrucAferde : Lepidhnn DrmmnoncHi The]]. Lepidiinn rolundmti (Dosv.) J). C. Lepidixw o.iiitrtclinrn fSpragiie . . . Stenopefalinn pedicellare Benth. ... StejiGpctalum rohusiion Endl. Menkm sphucrocarpa F. v. M. ... Menkea austmlifi (jchm. ... yienkea 'Allo.'^tda . 1 . M. IMack Capsdhi Andramna F. v. M. Crassulacme : Crassula colomfa Oistf. Piitosporacem : Pittospormn. philkjrae.oides J), C. Willow Legumiaoseae : Acacia genisioides Bentli. Acacia PftrkiUii Benth. ... Acacia ancara F. v. Af. Acacia hmchystachya Acacia Biirkittii var. Cassia Cimtelamiana Gaud. Cassia crcmophila A. Ounn. Cassia artcmisioidcs Gaud. Cassia SiaHii ]{. Hr Cassia de^oiata F. v. M. Cassia cardiospcrmn I'\ v. M. Brachyscma Chawbersii Benth. Davicsia aphylla H(‘nth ... Indigofara Oeorgei Pritzel Awainsonia microphylla A. Gray Swamsonia occidenialis F. v. M. Curara Jamwood ... !Mu]ga Bogada Creek Cassia Geraniaceae : Prodimn cygnonim Nees. ... Crowfoot Zygophyllaceac : Zygophyllmn iodocarpum F. v. .A[. Zygophylluni frulicolostim 1). C. ... Trihidgs asfrocarpus F. v. it. E (tphorhiarcae : Euphorbia australis Hoiss ... Hairy Baisaiu Euphorbia Drumuiondii Boiss ... Balsam Euphorbia crcmophila A. Cunn. ... Hock BalsaVu Stackhousiaccae : Stackhousia vimitim Sm. Sapindaceae : Dodonaea lohulata F. v. j\I. Hon}m« 5 h DodonaeafdifoliaKook... ,, Uhanuiaceae : Blackallia connata C. A . Gardii. yjalmceae : Abiitilou cryptopetalum, F. v. M. Ahutiloii otocarpum F. M. Plagianihus Ga.rdncri Bak. M, S. ... "" Sida corrugata var. ovata Sida calyxhymenia J. Gay. Sida ciliata X. T. Burbidge ms. ... Aug. ... B Aug. ... B Aug. ... B Aug. ... B, C Aug. ... B, C Aug. ... B Aug. ... B, G, H Aug. ... B Aug. ... B Aug. .. B Ap. Aug. .. . B, G Ap. Aug. ... B, H Aug. ... A Ap. Aug. ... A, B, C, 1 G, H Aug. ... G, I Aug. ... D Aug. ... D, E, F Aug. ... B, C, D, H Aug. ... A Aug. ... I), E Aug. ... I) Aug. ... 0 Aug. ... I Aug. ... B, C Aug. ... A, B, C Aug. ... B Alia:. ... B Aug. ... A, B, C, ]> Aug. ... B, I), H Aug. ... B, IX H Aug. ... B Aug. ... B Aug. ... B Aug. ... 1) Aug. ... ]) Aug. ... 1) Au". ... 1) Aug. ... In mulga-spiui fex zone Aug. ... B Aug. ... B Aug. ... B Aug. ... B Aug. ... B, H Aug. ... B, H Notes on the Vegetation of THE North-Eastern Goldfields. 131 JStercuUaceae : Hi'dchychiton (h'pyovii K. v. AI. ... Kurrcijong Frankeniaceae : Frankenia fecunda Sutninorliayos Frankenla seJosa Fitz. Thyjnelamceae : Pinelea microcepha/a R. Br. Pbnelea thesokU S. Aloore Myrtaceae • Eucalyptus pyrifot'tn/s Turcz. ... Alalloe Eucalyptus sp. Eucalyptus caniaiduleusis ... Crook Gum Halorrharja-ceae : Halondiayis odoutocarpa V. v. AI. Myriophyllunt. rhomholdcum N. T. Burhidgc tns. Aug. ... I), F Ap. Aug. ... G Ap. Aug. ... G ... Aug. ... B, G, H ••• Aug. ... B, G, H Aug. ... [ ... I Ap. ... A Aug. ... C Aug. ... In clay pan Uynbelli ferae : Didiscvs erlocxirpa Ikuith. Daucu.'< brarinafus Siob. ... Aug. Aug. Asclepiadaceae : Pentatropis Ken/peana Cogala Creeper ... Ap. Labiatae : Teucr'mm raceuiosum R. Br. ... ... 4,. Solanaceae : Solanuui orhiculatum Dun. Solanum Oldfieldil vai'. plicatile. Sp. AIoor(^ Solanuui aunblymenim Dun. Solanuui lasiophyllwm Dun. Nicotiana Gissei ... Eicotiana Murrayi F. v. M. et Tate Aug. Aug. Aug. Aug. Aug. Aug. B, D B, D B E E B, C, H D D Scroph'ulariaceae. : Peplidium Muelleri Benth. Myoporaccae : Eremophila rompm-ia H\i. Aloore ... Poverty bush EremophiJfi kumphylla Benth. ... „ Eremophila Maryarcthae Sp. Alooro Eremophtla oppositifolln. R. Br. ... Ere'ttiophtla uietaUk'orum Hp. yVoovt' Eremophila Latredm F. v. AI. ... Fuchsia bush Eremophila lairobei var. luberculafa „ Eremophila Panloni F. v. AI. ... Eremophila scoparki F. v. AI. Eremophila Fraseri F. v. AI. ... Fuchsia bush Eremophila longifolia F, v. AI. ... Eremophiki y^onnyH F. v. AI. ... Eremophila OldfieMU var. angusti- folia Sp. Alooro Eremophila uiaculata F. v. AI. ... Fuchsia bush Rubiaceae : Plectronia latifolia Benth. et Hook ... Aug. ... B ... Aug. ... C ... Aug. ... B, C ... Aug. ... B, D ... Aiig. ... lb E ... Aug. ... B . . . Aug. ... B ... Aug. ... B ... Aug. ... lb E ... Aug. ... I), E ... Aug. ... E ... Aug. ... B, D .. Aug. ... Ib E, .. Aug. ... E .. Aug. ... B Aug. ... B 132 Nancy T. Burbidge. Goodeniaceae : Velleia rosea sp. Moore ... ... ... ... ... Aug. Velleia cycnopotamica F. v. M. ... ... ... ... Aug Scaevola spinescens R. Br. ... Currant bush ... Aug.' ComposUae. ; Minuria leptophylla 1). C. ... ... Auo-. Calotis muHimulis Druce ... ... ... ... Aug. Caloiis hispid'ida R. v. M. ... ... ... ... Cratystylis conocephala (F. v. M.) Sage Auo-, Sp. Moore * Pieriyeroti eylindrlceps d.'Sl. Black ... Aug. lfelipte.rum spkndiduni Heins). ... Everlasting ... Auo-. Jlelipterum BnltH F. v. .\1. ... ... ... Jfelipfenim craspediokles Fitz. ... ... ... ... Au". Jlelipterum. Charsleyae F. v. AJ. ... ... ... ... AuS. Jlelipterum corymbosum Benth. ... ... ... ... Aug. Jlelichrysum roseum var. Daren^ ... Aua. portii Schoeuia cassiniana Stcetz. ... ... ... ... Myriocephaliis sp. ... Myriocephttlus (hicrinae F. v. M. ... ... ... Auo-. Arigiatilhiis l)riimuiomlii \io\\i\\. ... ... ... ... Aug. Calocephdlus BramisU Renth. ... ... ... ... Aug. Onaphalodes iiUgiuosnm A. Gray ... ... ... Aug Cephalipteruni Drammondii Benth. ... ... ... Aug. J^odolepis pallida Turcz. ... ... ... ... ... ^^jg JJrodiophyllmih acanthocephalwm Stapf. ... Aug. WWWWWtcWtdW tc tc' W td W W OOOcd SOo Bekaunite - -FKOivr Danbakagan, Western Australia. 133 11.— BERAUNUl^]- FIKIM DANDARAGAN, WE81MBiN Al^STRALlA. By C. R. Le Mesuriek, A. A. CM. Read 8th April, 1 941 : Ihibliwlied 20tli May, 1942. The occurrence of phosjjhatc minerals at Dandaraj^an was first officially reported by \\ . D. (/ampljoll iu MMXi (I). They coinprisi^fl tht; iron phosphates Dufrenite and Vi\'ianite and the aluminium phosphate Waveliite, besides beds rich in coprolites and apatised wood. In 1932 the late Hr. K. S. Simpson visited the area and colkictod sjiociniens of a new mineral, a hydrous basic jihosphate of |)otas.sium and aluminium, which was named Minyulite, after Min.\'ulo Well lu^ar which it was found (2). The subject of the ]iresent paper, Beraunite, a iiydrous ferric pliosphate with the formula 2b e^(OH );^( PO^).2.5H _.() was collected by Mi‘. K. Ftirman from a locality known as 1 he C'av'cs, ’ about 2^ miles N.F. of Vandan Hill, during a recent inspection of this area. The towaisfiip of Dandaragan lies 22 miles by road west of Moora, which is 106 miles north of Perth on the Midland Railw^ay. For the following description of the mode of occurrence of the Dandaragan Bei’aunite the author is indebted to Mi’. Forman, Government Geologist. The mineral is exposed on the face of a bluff about 50- 60 ft. higti, (Binposod m the lower 30-40 ft. of ferruginous sandstone overlaid by co])rolito bearing sandstone 3-5 ft. thick whicJi is in turn capp(‘d by what is probably greensand (glauconite bearing sand). The ferruginous sandstone is impregnated to a depth of from 2-3 ft. below the coprolite bed wath dark green dufrenite while the beraunite is confined to a nariow zone, not more than 3 inches in width in the upper surface immediately underlying the (coprolite beds. See section. Probable Creepsond S . (^o/>ea / C. C Perounde o Perruginoi/sSonrj The mineral occurs as radiating fibres with a silky lustre enclosed betw’eon w^alls of lustrous black limonite in irregular veins varying in width from 2 to 5 mm. The colour in mass varies from Ridgways 13’k, Russet to 2V"n\, dark olive. Hardness, 4-5. Specific gravity, 2*95. Under the microscope the fibres are transparent, wath opa(iue blotches due to adhering limonite, wliicli has penetrated between the finest fibres. 134 r. R. Le Mesurieh. Extinction is ditiicuJt to ascertain owing to the sheaf-liko form and extroino fineness of (lie fibres, Imt appears to be straight or of small amount. N in all directions is greater than that of metliylene iodide (1-74), and bire- fringence moderate. J’leocliroisni is strong from honey yellow to deep amber. 4’he material is readily soluble in acids and is decomposed by 5 *^'0 NaOH solution, in the latter case about 90 % of the is soluble. An analysis of the crushed matei'ial separat(Kl by floating in methylene iodide Sp. gr. 3*005, and sinking in bromoform Sp. gr. 2*85, then air dried and driefl to constant weight over HoSO^ and barium perchlorate gave the following results : — — Mols. Mol. ratio. f SiO ^ 1*88 insol. in HCl \Xot"SiO, . *03 ^AUO. •43 FOgO:i 54*11 338S 3*4 MnO Nil CaO Nil Sol, in HCl MgO Nil NaoO •58 KA) ■12 .PM, 27*24 1919 2 H.O 1- 15*60 8659 8*6 SO;: Nil Cl Nil 99*99 The excess mols. of F'e20;j and H^O over the ratio 3:2:8 may bo accounted for by intergrown limonite which it was impossible to eliminate. SUMMARY. A physical and chemical description is given of the mineral boraunite, a hydrous xdiosphato not jn’eviously reported from this State. (1) UeoL Surv. W,A. Bitll. 20, pp. 14-23. (2) Jour. Roij. Boo, W.A., 1932, XIX, pp. 13-16. Marine Copepoda from Western Australia. 135 12.— MARINE COPEPODA FROM WESTERN AUSTRALIA. I. LITTORAL HARPACTTCOIDS FROM ROTTNEST ISLAND. T3y a. G. Nicholes, Ph.D., University of Western Australia. Read 8tli April, 1941 ; Published 20th May, 1942, The species dealt with in the following pages wore collected from weed, attached to rocks on tho shore at liathurst Point, Rottnest Island. This island is a coastal limestone formation (actually a calcareous sandstone) lying some 12 miles west of Fnunantle, with its long axis running ajiproxi- mately east and west. A description of tho island is given by Glauert (1929), Bathurst Point lies at the nortli-east point of the island (Lighthouse Point in Glauert’s map). The material was collected during April, 1939. Fam. PELTIDIIDAE. Parapeltidium cristatum. P. cristatum Nicholls, 1941, p. 399, fig. 9. One specimen, an adult male, was taken and has been described in a recent paper dealing mth copepods from South Australia. lam. THALESTRIDAE. Eudactylopus australis. F. australis Nicholls, 1941, p. 410, fig. 15. This species, of which the female has already been described, is amongst those which are common to the coasts of South and Western Australia. Two specimens woyg taken here, one of each sex. The female (length, 1*68 mm.) shows a few minor points of dilloronce in tlie second antenna, first and fifth legs. The basal segment of the exopod of tho second antenna bears only one seta whereas there wore two in tho specimens from South Australia. In the tirst leg the endopod is somowliat more robust and relatively shorter, when compared with the exopodf than in ttio original material. Tho fifth leg bears an extra seta on the distal segment (fig. 1). Male : Length, 1*35 mm. Differs from tho female in the usual features and in the first leg which is rather more slender than in the female, resembling that of the female desciibed from South Austi'alia. The first antenna is 9- segmented, with a long sensory filament attached to the 4th segment ; there is the usual hinge between the 5tli anut having tlie distal sogmimt of the endo 2 >od longer ami moix* sl(md(n" ; distal segments of ondopods 2- 4 also longer than in littoralis (ef. ligs. 3 and 4), and seta formula for these legs as follows : endopod exopod P. 2 0 110 0 0 013 P- 3 0 no 0 0 013 [7. 4 0 121 0 0 013 Fiftli logs vmy like tlujse of llUoralis ; caudal rami nearly twice as long as wide, with a w'ell dovelo^jed inner denticulate ridg(j ; anal o[)orcukun unarmed. Male: J^ongth 0-90 mm. Differs from the fiunah^ in t)i(; usual sexual cliaractcrs and a fiiw minor points. First antenna G-st^gniontcd, suhrJdroneralc ; coxa of first logs with outer f)roj('ction ; endo 2 )ud of S(Mtond leg having the distal segment corisip. Rep., Pt. 1, No. 7, 227-307, Royal Society, Ja»ndon. Whiley, A., 1931.— Copepod Phenology. Observations l»asetl on tu'w material from (.’anatia aiul Bt'rmnda. Arch. Zool. Ilal., xvi, (iOI-017. 1935. — Haipacticoid (V>pe])o(la from Bermuda. 11. Ann. Mag. Mil. (10) xv, 50-100. Marine Copepoda from VA^estern Atxstralta. 139 Lefrends for Text Figures. Fig. 1. EudartylopuH auHtmlis, male and female. I'lg. 2. Phyllothalestris lata SJ). nov., female. The first anteima, logs 1, 2 and T) are moTti p°arts ‘Approximately one half that used in drawing the 140 A. G. Nicholls. Fig. 3. Orthopsyllus littoralis sp. nov., male and female. All the appendages are drawn to the same scale. Fig. 4. Orthopsyllus similis sp. nov., male and female. All the appendages are drawn to the same scale. Marine Copepoda from Western Australia. 141 PIG. 4 ^lATlONAL MUSEUM OF VICTORIA 143 New Leaf-Hoppers (Homoptera, Jassoidea) from \A^esterx Australta. 13.— NEW LEAF-HOPPERS (HOMOPTERA, JASSOIDEA) FROM WESTERN AUSTRALIA. By J. W. Evans, M.A., D.Sc., F.R.E.S. Read 8th April, 1941 ; Published 29th May, 1942. Communicated by C. F. H. Jenkins. Although the leaf-hoppers of Western Australia are not well known, sufficient material has been described from this region to arouse great interest. It is probable that the Eurymelidae, which is the dominant group of Australian leaf-hoppers, originated in Western Australia in Cretaceous times and to-day one sub-family, the Pogonoseopinae, which is mynnecophilous, is almost entirely confined to it. The most interesting genera of the Ipoinae, Cormiti^ 2 Jois Ev. and Bakeriola Ev. are Western Australian, as is also the most hand- some representative of the whole family, Eurymelops generosa (Stal). On this account it has been a privilege to have had an opportunity to examine the largt; collection of Western Australian Jassoids collected by Mr R. E. Turner in 1935 and 1936, and I am grateful to the Trustees of the British Museum and to Mr. W. E. China for having afforded me tins opportunity. Eight new genera arul fifty-one new species are described, and lists giv en of species that have previously been recorded from Western Australia or that occur in the pres(mt collection. The arrangement into families is based on the system proposed in an earlier 2 :)ublication (Ev'ans, 1939, b). T1k‘ sole new spei'ios described that was not collecteil by Mr. Turner is one which is represi^nted by a speeimen collected by Charles Darwin at King George’s Sound in 1836. All the types will eventually be returned to th© British Museum. EURYMELIDAE. The following species of Eurymelidae have previously been recorded from Western Australia. Those marked with a * are believed to be confined to th© State. Sub-family Eurymelinae ; Eurymdoj^s generosa*- (Stal), Eurymelops latijascia (Walk.), Eurnieloides walkeri* Dist., Eurynieloides punctata (Sign.), Euryrnelita terminalis (Walk.). Sub-family Ipoinae : Jpo (or pens* Jac., Ipoella norrisi* Ev., Stenipo swam* Ev., Ipoides cas^irinac Ev., Anncornutlpo lignosa (Walk.), Cornutipoides tri^rnis* Ev., Bakeriola pjrociirrens* (Jac.). Sub- family Pogonoseopinae : Bogonoscopus Icnis* (Jac.), /*. mynne.r* China, P. clarki* China, P. fuscus* Cliina, Lasioscopus acmaeops* (Jac,), Australoscopus whiteii China. Descriptions and figures of tlie Eurymelina(> listed above are given in Evans, 1933, of the Ipoinae in Evans, 1934 and 1939, c, and of th© Pogonoseopinae in China, 1926. IPOINAE. Anipo fusca sp. nov. (Plate I., fig. 14.) Length 3*5 mm. Head, ventral surfactL ant(vclypeus, lora and maxillary plates, pale yellowish-brown with a median chestnut-brown longitudinal stripe ; vertex anteriorly dark brown, posteriorly pale brown. Crown of even wddth throughout. Pronohon and Scntelhnn pale browau Tegnten colour- i 149/42. t This species also occurs in Victoria. 144 J. W. Kvans. less ljyaliiK*, clavus pale hyaMm^-browii, \ w ith brown ami white markings. Thorax, ventral surfac^e black. Himl tibia with a few spines in addition to a single spur, thus differing from tlie genotype'., A, porrigino-m (Sign.) in this cliaraederistic. Abdoaterij ventral siirfave pale brown. Mfile Oea Italia, ae'deagns as in Plate h, fig. 14. Tgpr ci' from l^e'rth. Anipo darwini sp. nov. (Plato I., fig. la.) Lragth 4*5 mm. Head, ventral surface, lora and maxillary plate's, pale whitish-brown ; ante-dypeus reddish-brown, fronto-clypeus and vtu'tex pale browm mc d' from King (leorge’s Sound (C. Darwin, 2/IS30). Anipo flavens sj). nov. (Plate I., fig. 12.) Leyigth 5 mm. (leneral coloration apricot, eyes grey. Tegmen hyaline, pale apricot. Alale Genitalia, aedeagus as in Plate I., fig. 12. Type & from Dedari, 1/3(1. Ipoella fulva sp. nov. (Plate I., fig. 13.) Length 5 mm. Head, ventral surface pale brownish-yellow wuth a largo dark brown T-shaped marking ; eyes reddish-brown. Pronotuni grey mottled with dark brown. Scutellum brown. Tegmen pale colourless-hyaline mottled with brown. Thorax, ventral surface dark brown. Abdomen, vtntral surface pak' brow'n. Male Geyiitalia, aed(^agus as in Plato 1. fig. 13. Type ale brownish-white ; vertex, including the crown, mottled with black and ])ale brown and with three longitudinal white stripes. Pronotum.. grey, mottled w'ith liglit and nds transversely from the costal border almost as far as tin' liind margin of the clax us, thence it bends towards tlie anal margin. Thorax and abdomen, ventral surface pale lu’owii. Male Genitalia, aedt'agus as in Plat(' 1., fig. lb. Type d" from Spai'gox ille, 1 '3b. Ipo speciosa sp. nov. (Plate I., figs. 10, 11.) Length (> mm. Head, ventral suifaee wbiite with pak^ brown markings on th(' antt'-clypc'us, lora, maxillary ))latos and fronto-clype\is anPuiorly, and blaek markings oii the vertex ; ('yt's Ijright red. l*ronotiim gi’ey, mottletl w ith \ <‘ry dark blown and with a luoad median longitudinal white* striiie. Scutelluin chestnut-brown witli ])ale brown markings, laterally pale brown. Tegmen New Leaf-Hoppers {Homoptera, Jassotdea) from . Western Australia. 145 whitisb-liyalino with dark brown markings and a broad wliito fascia extending from the costal to the claval bordei'. Tliora.\\ \ (‘ntral surface with wliitc' and brown markings. L<‘gs, tibiat^ \ery dark brown witli white and dark brown markings. Abdomen, ventral surface pale brown. Male Genitalia, sub-genital plate, paramere and aedeagus as in Plate L, figs. 10 and 11. Type cf from Dedari, 1 36. Stenipo grisea sp. nov. (Plate I., fig. 17.) Length 4 mm. Head, ventral surfa-<‘e grey anteriorly ; fi-onto-elypeus and vertex anteriorly, pinkish ; (wes black. Crown grey with brown markings. ProHotum pale gn^yish-brow ii. Peutellum pale reddish-brown. Tegnten hyaline- grey, veins pink. Thora.r, venti*al surface bla<*k, laterally pale biown. .46- domen, ventral surface and legs pal(> brown. Male Genitali((, aedeagus curved, narrowly cylindrical, paramere and sub-genital plate as in Plate I., fig. 17. Type c? from Dongarra, 10/35, BYTHOSCOPIDAE, The following species b(>longing to this family have been recorded from Western Australia : Eu nnos(:o]ms virldis Ev., Chinaella cadm.orei^ Kv., Hecalus hasedowi* Ev. and H. (dongatns* Ev. The two former are described and illustrated in Evans, 1936, the two latter in Evans, 1939, a. Of the three new' species d(‘scribed below’ one belongs to a genua closely related to Bytho- scoptis Cerin., whilst tw’o not only are not bythoscopoid in appearance but have dorsal ocelli ami l(‘drid-like heads. They are placed in the Bythoscopidae because of tlieir close affinity with Theniniatoseopits Kii‘k. The reason for placing the last named genus in this family has been {lisciissed previously (Evans, 1939, b). BYTHOSCOPINAE. Eurinoscopus transulcidus sp, nov. (Plate L, fig. 28.) Length 4 mm. Head yellow, eyes black, frontal and epicranial sutures*- indistinct. Crowai yellow, slightly w’uier in the centre than against the eyes. Pronotuni and Ecutelhini concolorous w’itli tlie crown. Tegmcn opacpie, pale green with ill-defined yellow and browai areas ; veins green, apex of t(>gmen pale hyaline-brown, llwrax and Abdomen, ventral surface and legs yellow'. Type ? from Dedari, 1 /36. PENTHIMIINAE, THAUMATOSCOPINI. PLATYSCOPUS gcii. nov. The head is produced and spatulato, the ventral surface concave, the- dorsal convex. The anhamae ai*e short and posterior to the eyes. There is a median longitudinal carina on the crown, and the ocelli which are on the crow’n lie midway between the carina and the sides of the liead. The pronotum is pai‘allel-sided and collardike and the tegmina overlap apically and have wide apijondices. The hind tibiae have four rows of spines ; a row’ of alternate long and short spines, a row of short strong spines separated by minute spines, a row of short spines and a row' of hair-liko -spines. 146 J. W. Evans Note ~lt is possible that in spite of the differences in coloration and the shape of the head that exists between the two insects described below, they are not distinct species but the two sexes of a single species. Platyscopus badius sp. nov. (Genotype). (Plate L, figs. 24, 25.) Length 7 mm. Head chestnut-brov\'n, eyes red. Pronotum chestnut-brown partially suffused with dark brown. Sentelliim chestnut-brown. Tegmsn pale hyaline-brown, apically smoky-grey, veins pale brown. Thorax and Abdomen, ventral surface and legs, brown. Last abdominal segment narrowly produced Tuediaily, emarginate laterally. Ovipositor sheath spinose. Type ? from Dedari, 1 736. Platyscopus coloratus sp. nov. (Plate L, figs. 20, 27.) Length 5*5 mm. Head, ventral surface dark brown. Crown chestnut- brown with a medium black stripe extending from the base to within one- third of the apex and two small brown markings close to the eyes on each side; eyes red. Pronoimn chestnut-brown. Scntellum black. Tegmen hyaline- brown, the costal border colourless ; apex and \^oins dark brown. Thorax and Abdomen, ventral surface black. Type c? from Dedari, 1/36. EUSCELIDAE, Kntettix norrisP f]\'. and Thamnofettix argentata- L\'. are tlie only pre- viously dc^scribed representative's of this family recorded from \\'estt>rn Aus- tralia. Botii are described and hgure of the row of strongest spines is mounted on an enlarged base and separated from its neighbour by three or four minute spines. The male genitalia hav'c wide flat sub-genital plates that narrow apically and narrowly producetl pygophores. Nkw IvEak-Hoppkks (Homoptera, Jassoioea) from W ESTEKN A USTR A LI A , 147 Occiplanocephalus ravus sp. nov. (Onotyjxj). (Plat(! II., IS.) Length O' 8 mm. IJead, vpntral surface palp p*(\yjsh-I)r()\vii witli ti‘aiis\ (as<5 brown mnsolo iinprossions on tlio fVonto-cIyppus, oyos dark blown, crown ^rity. PronotiifU grey mottled with Inown. Scutellntn with brown and black markingH, Tegtnen whitish-hyaline, veins brown ; vernation partially reticulate. Thorax, ventral surface grey with bi'own maikings. Hind tibia pak; brown, the bases of th arranged in pairs adja<^ent to each otlu>r, tiui spines of one of thes(i row ai'(i slightly larger than those of the otlan- and are s(^parated from each otlua- by sliort spiia^s. Kusceloficopus is close to ClradnUt 'Aott. Eusceloscopus yanchepensis sp. nov. {(kmotype). (Plate, II., figs. 15-17.) Length 4 mm. Head n‘.drow'n, eyes l>lack, onAVi hmI. (Vown of (wen width throughout. J*ronotM)n and Sentelluni riHldish-brow'n. 'Legmen hyaliruv brown but for the two ct'IIs adjoining tlu^ api)faalix which are smoky-hyaline ; vtiins pink. 'Thorax an' snffii.sed wiitli white Pronotum, antero-laterally white, medially grey. Seutellum wry dark brown, apicallv brownish-yellow. Tegmen, apical third pale brownish-hyaline ; proximal two-tliirds brou-n, but for the costal area distallv which is pink, and for irregular white areas. Thorax, ^•entral surface dark bro«-n. Abdmnen, ventral surface pale brown. Ti/pe ? from Dixlari, I '3(i. IDIOCERIDAE. Only two specie,s belonging to this family have been described previously from Western Australia. The\' are Idioceru,ale brown. Thorax anrl Ahdo- men, ventral surface apricot yellow. Type 9 from Merredin 12 ^35. Tumocerus grandis sp. nov. (Plat(' III., fig. 1.) Length 5-o mm. Head pale buff, eyes dark brown. Head wid('i- in the centre than against the (‘yes with two round black markings. Pronotum concolorous wuth the crown. Scutellum with two small brown spots lying on each side of the centre, muscle impr<‘ssions ilark brown. Tegmen, pale hyaline- brown, veins brown. TJforax anrl Abdotneu, ventral surface pale buff. Type 9 from Dedari, 1 'Sfi. Tumocerus glaucus sp. nov. (Plate III., fig. 1.5.) Ijength 5 mm. Head, ventral surface pale yellowish gn^en. Crown of even width throughout, narrowly pale yellowish-green against the eyes, medi- ally pink, with two marginal large black spots ; eyes red. Pronot^im, antero- laterally pale yellowish-green, medially pink. Scmtelliim, apricot. Tegynen, proximally, membrane and veins pale yellowish-gnam, distally hyaline-brown, veins dark brown. Thorax and Abdomen, ventral surface and legs, pale yellowish-green. Type (f from Dedari, I ;3ti. 150 J. W. Evans, GNATIA gen. nov. The ante-clypens is angularly swollen, the hind margin of the fronco- clypeus is distinct and the ocelli lie well away from the apices of the frontal sutures. The crown is of even width tlu’oughoxit and the tegmina are long and narrow and have wide appendices. The hind tibiae are short and have two spines mounted on enlarged bases in addition to several weak spines. Gnatia angustata sp nov. (Genotype) (Plate JIT., figs. 13, 14.) Length 5 mm. Head, v^entral surface, ante-clypeus pale brown, lora, maxillary plates and the vertex anteriorly whitish. Fronto-clypeus chestnut- brown mottled with dark brown. Crown medially dark grey, laterally pinkish ; eyes dark brown. Pronotuni antero-laterally pinkish-yellow, otherwise grey. Scutellmn pinkish-brown, inuscle impi*essions brown. Tegmen hyaline-grey with a dark brown fascia ; veins pale brown bordered with a double row of fine hairs. Thorax, ventral surface black. Abdomen, ventral surface yellow. Type 9 from Dedari, 1/36. Idiocerus luteus sp. nov. Length 4 mm. Head slightly convex, yellow ; eyes black. Proyiotum concolorous vdth the head. Scutelluni apricot. Tegmen, proximally bronze- hyaline-yellow, distally hyaline-brown ; veins yellow. Thorax and Ahdoynen, ventral surface yellow. Hind tibia pale green with two rows of dark brown spines. Type 9 from Dedai-i, 1 /36. Idiocerus coloratus sp. nov. Le}igih 4 mm. Head^ whitish, convex, eyes dark red ; ocelli yellow. Muscle impressions on fronto-clypeTis pale apricot. Crown wide, of t'ven width throughout. Pronotum grey. Scutellmn, bright orange-red. Tegmen proximally, and veins, whitish-hyaline ; distally and veins smoky-gn'V. and ventral surface, buff. Type $ from Dedari, 1/36. Idiocerus fucatus sp. nov. (Plate III., figs. 9, 10.) Length 3-5 Tnm. Head, ventral surface apricot, eyes red. Crown narrow, of even width throughout, the width of each eye equal to half the breadth of the crown. Pronotum anreyish-hyaline mottled with browa ; veins pink with wliite and brown bars and a series of evenly spaced white spots against the 'Costal border. Thorax, ventral surface mai’ked with a pattern of light and dark brown. Male Genitalia, pygophore more or less rectangular with an upturned apical process. Aedeagus as in Plat(‘ T.. fig. 4. T///>r from Dedari. Hackeriana translucens sp. nov. (Plate T., tig. 3.) Leyigih 7 mm. Head, ventml siirface greenish-yellow, ey(‘s r(‘d. Ocelli marginal, closer to the narrow apex of the head tlian to the eyes on each side, ('rown y)ale gn'enisleyellow with faint oval white markings, a central triangular area. flat, lati'rally declivous. Proy\ot>rm and Seufellnni concolorous witli tlie crown, 'fe'jtnen, colourless-hyaline, \'eins green. Thorftx and .46- ^loinen, \-entral surface yellow. Male (renitalia, paratneres spannei'-shaped, pygophores long aiid narrow, aedeagus as in Plate T.. fig. 3. T//pe d' from Ftur racoppin, 1 '3(i. Rhotidoides dongarrensis sp. nov. (Plate T., figs. 5, 6.) Length 1 mm. Head, ventral surfaci*. ante-clypeus, lora and maxillary plates mai'kod with a pattern of pale brown, dark brown, and black ; fronto- clypfMis and vertex, yellowish-white ; eyes concolorous witli the maxillary plates. Erowii slightly wider against the eyes than in the e.entn', marked with a pattern of ]iale aufl dark brown. ProneAnm. anterior margin in front of the wes, fiat, followed by a narrow deelivou.s area, the remainder and greater part. flat, brownish-griw with dar-k brown markings. Seutellmn pale brown. 'TegaiPii hyaline-colonrless. evenly mottled with dull greyisli-brown. Thorax, A'entral siu’faee with light and dark brown markings, begs, fore tibia pale brown with rlark brown markings, hind tibia pale brown, thi' bases of tlie spines brown. Abdomen, ventral surface, markt'd with liglit and dark brown mai'kiiigs. Alale Genitalift, pygophores im^gularly shaped and lacking ])ro- cess(\s. aedeagus and ]mram(‘re as in Plate T,, figs. 5 and 0. Tijpe d' from J^ongarra, 9 -35. NEWMANIANA gen. nov. Th(^ aute-olypeus is flal, and the fronto-elypeus convex anteriorly and depressed medially, sliglitly a.ntei'ior to tlie apex of tlu‘ head. The antennal ■depressions an* deep and continue anteriorly to the edge of the maxillary plati^s. Th(* frontal sutures divi^rge postts’iorly and terminati* at the ocelli, which are not visibli' eithci- in ventral or dorsal aspect. Tin* apical margin of the lH‘ad is mon' or less vortical anrl narrows nuHlially. The crown is flat anroihict*d and the eyes are large. The pronotum is slightly transversely coin'cx and the hind margin almost straight. Tlie scutellum is large and (S|uai in length to the combined length of the crown and pronotum. Th(' tegmina have mod(‘rately wide appendices and the \'enation and the armatun* of the hind tibiae are typical of tlie family. This giaius is close to I liieLeriana Kv. and Euprora Ev, Newmaniana viridis sp. nov. (Oenotype). {Plate r., figs. 7-9.) Length 7 mm. Head, ventral surface pale yellowish-green, posteriorly red, medially pink ; eyes red, crown pinkish-green. Pronotiun and Scmfellum yellowish-green. Tegtnen pale hyaline-green, veins pale green. Male Gemtalia with wide pygophores lac'king processes ; aedeagus and parameres as in Plato I., figs. 7 and 8. Tgpe d from Dongarra, 9 '35. New I.eak-Hoppehs (Homoptkua, Jassoidea) fkom W ks'I’erx Ai-stuaija. MACROPSIDAE. Four r('pres(‘utati\'ps of tliis family ha\'o boon (IcsoimIxmI j)roviously from Westf^rn Australia, 'Tlu^so aro Stenoscopus (/r/innNO/i(h’* Kv., M((cro/>-^hs occidcHtrdls'* |0\., .1/. lutvfts Fv., and Oaco/wAv HVf)})uU(s'^ \*W. (I08(i)- tho uow spo(*i{‘;-; (l<‘S('ri}H‘(l Ix'low, witli oiio ox(‘(‘ption, liavo boon placi'd in tlu^ i^onora Mdcrop.sis lj('\vis aud Onmp.'iis- Ibirm. apcordiiii; lo wb<‘tlu‘r llu' stri- ations ou th(‘ [)r()nolum an* obliijun or traiisvors(‘. It is, how(‘\-(U’. appn'ciat('d that they may not all b(‘ coimoiKM-ic with th(» gpnotypns of tin* two ^taiora, but tlir diff(‘T‘('ntia( iorj of furtli(‘r ^lan'ra is in- ad\isa]^l(‘ until moi’i* material is a\-ailal)l(‘ for study. STENOPSOIDES gen. nov. Tho Iu‘ad is slightly longer than wide and transvoi'S('ly eoiwox, a erown is not doN^eioped. The proTiotum which is oiiormously ant(*riorly produ(*ed, and wlheli fi'om ))elow a])p('ai's to b(‘ directly continuous with the face, luus a nn'dian x'ontral carina. The hind margin of tin* pronoturn is emarginatti and tho scutellum is large*. The* tegmina arc* apieally ataite* and ha\'(' ap- p(*ndic(\s that eontiniu* ai’oimd their apic(*s as far as the* costal bonh'rs. Th<^ liind wings are fully (hn-(‘lopt‘d. Stenopsoides turned sp. nov. ((JenotyjK*). (Plate' II., figs. 4 t).) Length 7 mm. Hcaf/, venitral surfac<> yelhnv with brown punctures, nyas rf'd. Pronotnni. “ veaitral " and dorsal surfaces yt'llowish-bi-own with brown punctures ; sides anrl apt‘x, dark brown. Sfnifcffunt yellow. I'cgdien pale liyaline-yollow. Thontx and AMome}}, v('ntral surface yellow. Tgpe V from Dedari, 1 /3t). Macropsis flexus sp. nov. (Folate* II., fig. 14.) Length 3-oinni. fiend buff witli minute pale brown puncturf's, ('yes r(*ddis}i-f)rown, oec'lli l>lack : musek’ imf)ressions greenish-yellow. ( ’rown \'isible from abo\'(' as a nai*row band of (’\<'n width. Pronot/rnt concolorous witli the lu'ad, slightly d(’cli\'ous but not hunified. Senfellnni concolorous with tile ])ronotum. Tegtnen colourli'ss-hyidiiu', \'('ins brown with white bars. Thorfii\ vi^ntral surface pak' brownish-yi'How, basi's of all the spines on tho liind tibia black. Ahdotne)}^ xaaitral surfac<* brown. Pgpe V from Dongarra, 10 35. Macropsis eburneus sp. nov. (Plato II., fig. 9.) Length 4-5 mm. fiend ivory with brown punctur(*s, fronto-ciypc'us nK'dially brown with d(‘ns(^ dark bi’ow'u punetun*s : muscle impressions smooth, yi'llow ish-hrowui. Crown visible only nai’rowly against the (‘yiss. l*ron<)tnm anteriorly cream, w ith d(*nse brown anti'riorly-direettHl striat ions, steeply de('livous, posttu'iorly gn'y with scatt('n*d brown punctures, Senfellnni orangtvbrown with black |)unctui'(‘s, Teginen byaliuc v\itb scattered small brown spots, tiKjstly on and along the sides of tin* veins. Thorn,i\ \'ejitral sin'laco black, <'dg(>d with l)rown. Lc^gs crc'am with brown markings. Af>- domen, v(‘ntral surface l)lack, the hind margin f)f each s('gm(‘nt ert'am. Type ? from Dedai-i, 1 3(i. 154 J. W. Evans Macropsis eitrinus sp. nov. Length 4 mm. Head, Fronotum, and ScMtellum, bright orange. Tegmen narrow apically, colourless-hyaline, veins pale brown. Thorax, ventral sur- face black. Legs orange, spurs on the hind tibia black. Abdomen, ventral surface black. Type ? from Dedari, 1 /36. Macropsis declivus sp. nov. (Plate IE., fig. 10.) Length 3*5 mm. Head, anterior to the ocelli orange-yellow with reddish- brown punctures, musch^, impressions pale orange ; posterior to the ocelli medially grey, latcu’ally pale pinkish-white with brown punctures ; eyes dark |•eddish-brown. PronoHim, anterior third steeply declivous and more or less at right angles to the r<^mainder, which is flat, antero-laterally orange, without iJunctures, the remainder greyish-brown with dense brown punctures. Scutellnm orange-brown with sparst^ brown punctures. Tegmen whitish- hyaline with large round brown spots lying principally on the sides of the veins ; veins pale brown. Thorax, ventral surface black. Abdomen, ventral surface orange-brown. Type 9 from Dedari, 1 /36. Macropsis flavomaculatus sp. nov. (Plate 11., fig. 7.) Length 4 mm. Head apricot mottled with whitish-yellow, eyes dark brown. Pronotom anteriorly declivous though not so steep as in M. deePruf^, greyish-white with apri<*ot punctures. Scutellum concoloi'ous with the prono- tum, with sparse orange-brown punctures. Tegmen colourless-hyaline mottled with pale coffee-brown : veins pale brown with white bars. Thorax ami Abdomen, ventral sui’fac*' and legs, pale orange-yellow. Type $ from Dedari, 1 3fi. Oncopsis aeneus sp. nov. (Plato 11., fig. 8.) Ley\gth 4 mm. Head, v(>ntral surface sordid yellow, eyes red. Crown narrow, visibl(> only against tin* eyes on each side. Pronoium golden-brown. ScMtcllarn yellowish-brown. Tegmen bron/.y-hyaline witli a longitudinal white streak lying along the first cubital vein proximally. Thorax, \'entral surface black. Abdomen, vent]*al surface yellowish-brown. Type S' from Yancliep, 1 1 38. Oncopsis gibbus s]n nov. (Plate IT., fig. 11.) Ijcngth 3-8 mm. Hv. Tartessus mundarensis sj». nov, (Plate 1., fig. 19.) Length 5 mm. Head, ventral surface, fronto-elypeus anteriorly faintly suffused with brovvui, sutures l)rown ; eyes black. Crown, fronto-elypeus yellow, vertex pale whitish-brown. Proytotum concolorous with the vertex. Scutellum anteriorly pale whitish-brown, apically lemon-yellow. Tegme)i pale liyaline-brown, apically smokv -brown ; v’cins dark brown. Tlmrax and Abdo- men, ventral surface black, legs yellow, bases of the spines on the hind tibia black. iV/r/Zc (fVn^Yr/hV/, aedeagus as in Plat(' L, fig. 18. Type ^ from Mundaring Weir, 2 /39. 'I. \V. Kvans. 156 Tartessus rugosus sp. now (Plate I., 20,) Lcngfh 7'5mni. fJead pale hrownisli-yellow ; fronto-clypcus anterior to the antennae, with brown transverst^ muscle impressions ; posterior to the antennae, and the vertex, rn^jose motti(\i witli pale and dark brown ; eyes dark brown. Crown, eonsistiiiK entirely of the vertex, pitted with lipli't and dark brown markings. Pmwtum and ScutHhim jjale brown with dark brown markings. Tegmen pale hyaline-brown, veins dark brown. Abdomen, \eutral surface marloMl with a pattern of light arul dark brown and black. Hind til>ia pale brown, bastes of iho s|)in<‘s dai'k browti. Male Genitalia, acideagus as in Plate I., fig. 19. Type from Yanchep, 11/35. Tartessus flavus sp. nov. (Plate I., fig. 21.) Lcngfh (> mm. Hea witli liglit brown and black markings. Legs, fore and middk' tibiae r(aldish-brown wath black markings, hind tibia pale brown, bases of the s|)ineH black. Male Genitalia, aedeagus as in Plate I., fig. 21. Type cT from Dedari, 1/36. Tartessus latus sp. nov, (Plate l.j fig. 23.) Length 6 mm. Head, ventral surface, fronto-elypous darker in colour than the rest oC tlie face with a transverse daik brown bar betw’eeu the ocelli and tlu> antennae ; eyes bla<'k. tVown consisting (Mitirely of tlu' \'crtcx, wider against tlc' ('y('s tium in the ecntrt*. rugos(\ PrrnnMam ])al(* brown. Seiitellnm daik In-oAcn with \-ery dark lu-own muscle impressions. Tvgmen pale liyaliiu'-brown, \(‘ins brown. Thorax and Abdomen, ventral surface, and legs ]ia(o brown. Male ihoTitalia, a(Mleagus as in Plate L, fig. 22. Type from Dedari, 1 3(). NIRVANIIDAE. Till' sp('eios describ('d below is the first representative of this Indo-Malayan group of l(>af-]iop]iers to be deserib(‘d from Western Australia. OCCINIRVANA gen. nov. The liead is ]irodue('d and spatulate, ventrally concave and dorsally convex. I lie anto-an appears to be more than a mere cross-vein. The hind tibiae have a row of evenly-spaced short strong spines set on enlargtid bases, a row of long strong spines^ anotlier of short spines and one of hair-like spines. Occinirvana eborea sp. nov . (Genotype). (Plate JI,, figs. 1-3.) Length nnn. Head, viaitral surface i\ ory, lora and ante-cl>'peus brownish-grey, eyes black. Grown pale ivory v ith a median longitudinal apricot band and two narrow sinuate lateral bands ; ocelli red. PronoVum ivory with a median apricot band and apricot mottlings antero-Iati ‘rally. Scutelluni ivory M'ith apricot muscle impressions and a faint broad median apricot band. Tegmen pale hyaline-brown but for the clavus which is v hite ; apically brown, and with a brown area at the fork of Cu , . Thorax, \ entral surface apricot anfl ivory. Ahdo)n€n, \^i'ntral surface i\nry ; last segment medially emarginate ; ovipositor sheath with proximal brown and apical black spines. Type ? from Perth, 3 3(5, on Casuarhia sp. STENOCOTIDAE. Stenocotis depressa Walk, and Smicrocotis solonionp Ew are the sole rt-pre- sentatives of this family so far rocordetl from Western Australia. (Exans, 1937, d.) LEDRIDAE. The following have been recorded from \^'estern Australia ; sub-family Ledrinae, Platyledra monsirosa'^ Pv. Sub-family Ceplialelinae, Auacephedeub^ siibreticulatus (Kirk.), A, minutus Ev., A. tains'^ Ev., ParadorydiiDtt niichaelseniA Jac. (probably belongs to Anacephaleus Ev.), N otocephalius hartraeyeri'^ Jac.,. Procephaleas hulhosa^ Ev,, (Je.phalelus punctatus'^ Ev. Jacobi’s, species, were described in 1909, Kirkaldy’s in 190(5 and the remainder in Evans, 1937, a, ami 1939, c. REKEREXCES. (hiiia, \V. E., 1920 — A new sii])-faTihly of the i5vth()seo[)icUie. Trans, Eni. Soc., Loiidon, 289. EvaiiH, .1. W., 1933 - A Hevisioa of the Eurymcliiii. Trans. Roy. Roc., R.. Ansf., 57 : 73. ,, 1934— A Revision of the liunnae. Trans. Roy. Ror., R. Ansi., 58: 149. 1936 — The Bythoseopirlae of Australia. Pap. Roy. Roc.^ Tas.. (il., ,, ,, 1937, a— Australian Leaf-Hoiijiors Pt. 4. Paji. Rog. Roc., Tas..,. 37. ,, „ 1937, b— Australian heaf-Ho])])ers Pt. 5. Pap. Roc., Tas..,. ~A, >> .. 1937, e— Australian heaf-Ho|)])ers P(. ii. Mem. Qneensland Mas., 1 1 (2) : 141). „ „ 1937, (I— Australian Leaf Hoiniers Pt. 7. Mem. fp/eeihdand Mn.^., II (2): 157. 5, 1939, a — Australian l.caf-Iloppers Pt. S. Pap. Roy. Roc., 'Fas., 1. ,, 1939, h — A eoni rihution to llu' Studv (»f the .lassoidcvi. Pap. Roy aSoc., Tas., 19. 1939, e — Sonu‘ new Australian Leaf Hoiiijcrs. TranN. Roy. aSoc., R. AnR.. 63 (1) : 44. -hicohi. A., 1909 — “Faun. S.t\'. Aust.J Miehaelsen u. H.artm(‘yer, ii. : 329. Kirkaidy, G. W., 190(L--Leaf Hoppers. Ball. IJatvait Rag. Ass. Eat., 1 (9). Myers, .J. G., 1928 — A note on Australian 'rypliloevhiuc Leaf Hopp(‘rs., Ball. Eat, /.V-s.,. 18 (I) : 311. J. W. Evans 158 Figure 1. 2 :i. 4 . 5. f>. 7. S. U. 10 . 11 . !2. la. 14. 15. 16. 17. 18. 19 . 20 . 21 . 22 . 2a.' 24 . 25 . 26 . 27 . 28 . Plate I. Putoniessa notOr head, ventral surface. Puprora widlen.si.% aedeagus. Hacker inna tranHlucens\ aedeagus. Patonlessa noUi., aedeagus. Jihofldolde^i fiounarrensis^ aedeagus. Photidoides dongarren/^is, paramere. S ewmammia vh'id\6\ [)aratnere. Newmaniann virdlis^ aedeagus. Nenmaniana cirldus, lu^ad, ventral surface. Ipo specioms sub-genital plate and paramere. Ipo sperlosa, aedeagus. Anipo flarens^ ae9 J) ?> 99 1 . 2 3 . 4 . o. >> 99 99 99 99 99 99 99 99 99 99 99 99 99 99 99 99 99 6 . 7 . 8. 9. 10 . 11 . li>. 13 . 14 . 15 . 16 . 17 . 18 . 19 . 20 . 21 . 22 . 23 . Plate II. Occinlrmna <^horea, head ventral surface. Occinirvana eborea, head and pronotum, dorsal surface. Ocdnirrami eborea^ tegmen. i^tenopsokles tumeric head and pronotum, ventral surface. i^te.7iopsoides tumeri. SteMOpsoides tumeri., tegmen. Macropsis flavouimucafaSy head and pronotum, dorsal surface. Onco 2 y^‘ih$ aeneufi, head and pi’onotum, dorsal surface. Macropsis eburneus, head and pronotum, dorsal surface. Mof.ropsls de.dirus, head and pronotum, dorsal surface. Oncopsis gihbufi, tegmen. Oncopsifi bU'olovaiii.Sy tegmen. Oncopf^is Jusfopunctaiujiy tegmen. Mmrop&is j}e.ru,% tegmen. iCnscelosmpns yanchejyensisy male genitalia. Eusceloseopus yanchepensis, tegmen. Eumeloscopm panchepensis, head and thorax, dorsal surface. Occiplanoceplmlus rm'uf<. Eu^^retosropiis tegmen. lyGltocephalus deGolorcbtus, head and thorax, dorsal surface. Dellocephaliis dedarensis, head and thorax, dorsal surface. -Oeltocephalus dedarensis, tegmen. Deltocephalus pidlatus, tegmen. New Leaf-Hoppers (Homoptera, Jassoidea) From Western Ap^stralia. 161 162 J. W. Evaks. Figure >> »> >> »> 1 . 2 . 3 . 4 . 5 . 6 . »» »» »♦ 99 99 99 99 99 8 . 9 . 10 . 11 . 12 . 13 . 14 . 15 . 16 . 17 . 18 _ 19 . 20 . Plate III. Tumocerus grandis, head and pronotuin, dorsal surface. Tumocerus meredinen-sis, head and pronotum, dorsal surface Tumocerus rri'eredinensis, hind tibia. Tamocerm vaTiu-% lieaecimcns which are of inestimable Aalue. I am also indebted to Dr. L. Diels of the Berlin Herbarium, for the generous gift oi‘ numerous type fragments of specimens Avhieh he, in company with Dr. E. Pritzcl, collected in AA^'estern Australia in 1901 and 1902. Finally I Avish to express my thanks to Dr. AAR E. Blackall who has assisted mo by placing at my disposal the whole of his herbarium, and for the donation to the State Ilerbarinm of all the undescribed specimens which form the tyjie material of many of tlie species herein described. State Herbarium, Perth, AY. A., March, 1941. j 332/42 166 Charles Austin Gardner. MOX( )COTVLKl)ONE.®. GKAMINE.ll Triodia Wiseana C. A. Gardn. sp. nov. Laxe caospiiojsa, innovatioues immorosac, euiinis iiumerosis e eaespite exscilis, rigidis, ads('en{lentil)us; paiu-it'olialis, vaginis stipremis breviter exsoi'tis; I'oliis ai’caialo-patonlibus fleinde liorizoiitaliter divavieatis, glaueis^ laevis, apioo eartdagiiK'o-puiigenlibiis, ora vaginuo et inarginibus inforioribus excopLis, rigidissiadsj tevefi-iiivolutis, basin versus exceptisj vaginis glabris, striatisj non viscosis, ab ora dense et longe albo-eiliatis; panieulis brevibusj paueisjjieidatisj spieiilis i(»nge peduneulatis, ereetis, jx'dimeulis seabris, tlexuosis; glmnis ('Iliptieo-laneeolatis, glabris, diiris, dorso convexis, laevis, crassis. ^Irinervis, apiee vix aeidis, glumis superioribus quam infei'iores vix longioi'ibiis; lenimatis d-o, earlilagineis, purpimds, glabris, dorso laevis^ breviter et sid)aequa!il('r tridentatis, denti})us ovato-triangularibus, aeutis, minute eiliolafis, nervis t) approxinuitis in faseiculis 3 quorum ncrvo centi’ale iX)so baso ad apiees loljormn ])ro(]ueto, lateralibus iiieompletis; 2>J>lea quam lemma, bi‘evior(', tmniiter pal).vraeea, oblonga, retusa, biearinata ; stamiiiibus 3. Planta U8({uc 40 cm. alta; lainimi folioi’uui iaferionuu 10 cm. longa^ superieriim 1.5-4 cm. kmga ; paiiicula usque 0 cm. longa ; gUimao 9 mm. longae; lemma vix 0 mm. louga; palea 3 mm. loiiga. Hal), in I'egione tropica ab moatibus Ilamersley, in rupestribus ad viam “Mount Margaret I'ass/’ eina 1000 m. alt., Hor. m. August, Gardner, 3129. Hane sp(‘ciem cuis est Australiae vero pro illustrissimo Franeiseo Iose[)lio Wise nomino, occidentalis Coneilio ])rovineia Agricolarcs res eonsulere eolonariasiiue, qui, zelo valdo memoriae digno, ad plautas (|uae regione fluvium Gas('oyne infer et Fort<‘seue vireseant invostigandas anno 1932, se mihi eomitem amabilissimum eonsoeiavit. Ai’linify l(j T. luirro.^faclfi/a K. Hr., difteriiig in the Iiaiiy leaves, in the much shoj'ter culms, the much reduced and few-flowered and (lexuosely-branched )>ani<*le, glalirou^ glunu's, t'ce. It liears some resemblance to T. lo'agiceps J. iM. Hlat'k, but can ))e readily separated by tlie much shorter panicle, ftnver- flowercnl spikelets, th(‘ teeth of the lemma Avhich is never hairy, and in the hairs of' the leaves. The com])arative shape and size of the teeth or lobes in 1 vied a is not reliable to any extent, and varies considerablv in T, Basedoivii F. Fritzel. DICOT YLEDONE^]. CASUARTNACF^. Casuarina pinaster C. A. Gardn. sp. nov. Frutex 1-4 metralis altus, ramosissimus, ramis ]iatonto-divaricatis,. reliquis squainorum delapsitrum persisteiitilius indutis; ramulis confertis, divaricatis, (luatornato-vertieillatis, biarticulalis, niembro ramulorum infero in dentes (piatuor saepo marcescent(‘s desimmte, dentibus ovato-lanceolatis, apice setaceis, enudis; sn]>(*ro inarticulalo, elongaio, folium tetragonum imitante, internodiis albo-pubescentibus, demum glabris, a]dce pungento-acutis; internodiis amentigeris valde abbi'eviatis; ameiitis masculis in axillis dentium congestis, quaternis, brevibus, squamis subovatis, extus pubeseeiitibus; strobilo breviter ])ednncidato ovoideo; bractea crassa, e liasi lata dorso rufo-pilosula, in setam brevaem produefa; bracteolis exsertis, apice rotundo-obtusis, crassis^ CONTRIBUTIONES FlORAE AuSTRALTAE OCCIDENT AJLIS XI. 167 extus leviter coiivexis, lamina exteriore su])libei*o, in rostrum longi™ triquetra produeta, valvis exedonte; aehaenio riigroseeiite, ala hyalina obliqua. Frntex 1-4 ui. altus, habitus fere piiiiodei; ineinbrum foliae-eum 4-5 cm. loiigum; amenta usque ad 1 cm. longa; strobihis 2-2.5 longus ac latus. Hab. in distr. Stirling, prope Kukeritij in aronoso-lutosis, Gardner 1732; prope Xyabing, W. E. Blaekall 3094. Tbe aHinity of tliis species is witli C. acuan'a F. but difPers in the much larger size, the rigid divaricate brandies^ the length of the terminal (upper) internode and its distinctly tetragonal form, in the scale-leaves which ai'e long and niem]>ranon.s and terminate in tine sdaceous points, and ciliate on the margins; I he teeth of tlie male spike are much larger and acuminate, and the Hi>ikes shorter, more robust, and more coiitracted; the cones are larger (almost twice the size of those of C. acmiria), and the dorsal lamina is very much reduced, and ]>roduced into a stout triquetrous spine which only shortly exceeds the v'alvular ])ortion and iiiciirved at the a})ex. C. Prinsepiana C\ Andrews, in Jour. West. Aus. Xat. Hist. Soc. i. 43 (1904). The type specimen of this species, now in the possession of the State Herbarium, possesse.s no cones, the type consisting only of a small Iiranch with a young female inflorescence. A search for material in tlie herl)aria of the Royal Botanic Gardens, the Natural History Museum, South Kensington, and the Berlin Herbarimu proved futile in this respect, and we must assume that the type cones ai-e Jiot in existence. In 1936, Horace Brown brought me sjieci- mens of a ('asnarina from Wurarga, with several coiies, the bi'acteoles of which are tricnspidate, the two lateral points are side ])y sid(‘ very (dose to the extremity of lh{‘ vah e, witii a rounded thin valve afiex between tbein, and a third conical process from the dorsal surface. This api>eared to be an nndescriluul species, for wbicli I had ]M'oi>osed the name C. fricuspidata, hut still further material has been received from Mullewa, and a search for another Oasnarina which might prove to be tyjncal C. Prinsepiana has been un])T-oduciive. Andrews, in describing the bracteole of (’. Prinsepiana says ‘Ahalvos much exserted with ro?nided poinis, hack covered with long brown haii- below, glabrous in the fre(‘ part with two ])roniinent ridg(*s, o))tuse and extending beyond tbe apex of the valve; occasionally there is a third shorter protuher- anec between the other two, adnate lower down and ending in a short point.’’ This descriptioTi does not exactly api.dy to the Wurarga sixvdniens, but considering that these sharp ])oinls are so (‘asily damaged, and that the braiudilets of the two are imlistinguishable, it is ]ie)'haps best to ('onsider the Wurarga and Mullewa specimens recently collected as representing 0. Prinsejriaim. The aHinity of the s[)ecies is with (\ tfeatiralvis, tlie plant differing in being a tree wifh tlie habit and ap])earance of C. (jJaaea, biit the cones of C. aenliralris, except that tlu^ bT'acteoh's Itear a conical dorsal appen- dage, and instead of a simple terminal conical point, there are two collateral conical protuberances — not ‘Vidges’^ extending beyond the valvular portion. In C, aentiralvis there may occasionally be found a division of the apex of the bracteole jmint into two ])oints, but in C. Prinsepiana the two points are quite distinct, and the obtuse and thin apex of the valvular portion is plainly visible Ijctween the two. C. bicuspidata Di(ds non lienih. The specimens deserilied as C. bicuspidata l)y Diels, collected near Southern Cross, the type of whicli I saw in the Berlin Herbarium, is C. acutivalvis F. Miiell. C. bicuspidata as far as is known, is restricted to South Australia and Tasmania. 168 CHAin:.ES Austin Gardner. PROTEACEiE. Grevillea eriobotrya F. Muell, Fragm. x. 44 (1876). G. Victori A. Morrison, the type of which is in the Edinburgh Her- barium, is synonymous with the above. The species is common in yellow sand over an area extending from Kununoppin to Koorda, and the species is a shrub 3 or 4 metres in height with white flowers. The seeds are thick, glabrous and hemispherical in an orbicular white villous follicle. They are much like those of G. annuli f era F. Muell. Grevillea nana (7. A, Gardn. sp. nov. Section Hehegyne. Frutex humilis, ramis elongatis, decumbentibus; foliis pinnatis, elongatis^ rhachi flexuosa, subtus bisulcata, segmentis (pinnae) 7-9, anguste linearibus,. divaricatis, rectis vel curvatis apice pungentibus, glabris, supra convexis,. subtus bisulcatisj racemis terminalibus, secundis, longe pedunculatis,. laxifloris, rhachi dense albo-villoso ; pedicellis elongatis, albo-villosis; perianthio albido, extus dense albo-sericeo, sub limbo globoso revoluto; toro plus ruinusve recto; glandula hypogyna late semilunari, crassa, conspicua, a toro libera ; ovario sessili, albo-villoso ; stylo purpureo, glabro, stigmate obliquo. Foha usque 20 cm.^ longa, segmenta l.S-4 cm. longa; racemi 8 cm. lougi; pedunculi 8-9 cm. longi; pedieelii 8nmi. lougi; perianthium 6 mm. longumr stylus 1.8 cm. longus. Hab, in disti\ Avon inter Koorda et Beneubbin, in arenosis apertis, flor. m. Sept. TF. E. Blackall 3380. AlTmity to G. asparagoides Meissn., differing in the simply pinnate glabrous and much longer leaves, and the absence of a glandular pubescence. There is a certain similarity in the leaves to those of G, Ilookeriana Meissn.,. but they are pungently acute and different in other respects. Grevillea rufa C. A. Gardn. sp. nov. Section Plagiopoda. Frutex rigidus, (lin'iisus, semimetralis altus, ramis divaricatis, ramulis pubeseentibus; foliis cuneato-ovatis vel cuneato-obovatis, ])etiolatis, sinuato- dentatis, dentibus s])ino-pungentibus, remotis, conspieue reticulatis, supra scabris, subtus aiipresso-pubescentibus, basi leviter attenuatis, apice plusminusve triincatis; racemis terminalibus, rhachi breve; pedicellis rufo- villosis, perianthio suba:quilongis ; floribus rufo-purpurascentibus; perianthio dense villoso, intus glabro, sub limbo ovoid eo-globoso revoluto; toro valde obliquo, glandula liypogyiia inconspicua, toro concave adnata; ovario breviter sed distinete slipitato, dense rufo-villoso ; stylo vix exserto, appresso-liirsuto,. crasso, complanato. Frutex eireitei' 50 cm. altus; folia 4 cm. longa, 2.5 cm. lata; pedieelii 1-1.2' cm. longi; perinnthium ca. 1 cm, longnin; stylus 1,7 cm. longns. Hab. in distr. Coolgardie, ad fines meridionales prope Pallarup juxta Mount Short, in clivis glareosis, 11. m. Septem. 1930. Gardener. Affinity to G. inslgriis Kipp. ex. Meissn., but with an indumentum of + rufous hairs throughout, that of the flowers being a I’usty red. The leaves are cuneate at the base, and vary from obovate to ovate in outline, resembling somewhat those of G. agrifolia A. Cnmi. in shape and texture. This interest- ing and apparently very rare species is known only from the original locality where only two or three plants were found. CONTRIBUTIONES FLORAE AuSTRALIAE OCCIDENTALIS XI. 169 Greviilea Dielstana C. A. Gardn, sp, nov. Section Calotlvyrsus. Frutex ramosus, cireiter metralis altus, ramis ramiilisque intricato- divaricatis, glabds vel sparse appresso-pnbescentibus; foliis longo petiolatis, triternatis, iterum trifidis, segmentis linearibiis, divaricatisj angnlato-costatis, glabris, apice pungentibns; racemis paucifloris, secundis, rhachi glabra vel pauce pnbescenti, pediecllis gracilibus; periantliio aurautiaco vcl coccineo, sub limbo ovoideo-globoso revolnto, extus glabro, intus albo-tomeiitoso ; toro valde obliquo; glandula bjq^ogyna conspictia, crassiuseula, subpatolliformi; ovario glabro, stipitato; stylo elongate, glabro, viridi, disco oblique laterali, Frutex 70-90 cm. altus; folia 4-5 cm. longa, segmenta ultima 1-1.4 cm. longa; pedicelli 6-7 mm, longi; perianthium cireiter 8 mm. loiigum ; Ktylus 2.5 cm. longus. Hab. in distr. Irwin, in arenosis fruticulosis prope fluminem Murchison, Gardner 2590 (Typus); atque inter Indarra et Yuna, in locis similibus Gardner, Affinity to G. teretifolia Meissn., from Avhich it differs in the constantly trichotoraously divided leaves, larger flowers and fewer flowered racemes, the oblique torus and the saccate base of the perianth, as well as in the indumen- tum of the inner surface of the perianth-segments. Greviilea obliquistigma C. A. Gardn, sp. nov. Section Conogyne. Frutex erectus, 2-5 metralis altus, ramis ramulisque erectis, demiun glabris; foliis angustissime linearibns, integris, acuminatis, supra valde convexis et uninervis, snbtus profunde bisulcatis, glabris; racemis terminal- ibus, erectis, densifloris, cylindricis; rhachi glabra; floribus albis; pedicellis glabris; periantliio sub limbo recurvato, extus glabro, intus pilis brevibus adpressis veslito; toro recto; glandula hypogyna parva sed distineta, reniformi; ovario stipitato, glabro, styh) glabro, breviter exserto, stigma te brevi eonico, obliquo, basi niarginato margine membranaeeo plieato crenulato. Frutex 2-5 m. altus; folia 16-18 cm. longa, 1 mm, lata; raeemi 5 cm. (cum pedunculo 8 cm.) longi; pedicelli 4 mm. lougi; perianthium ca. 4 mm. louguin; stylus 5 mm. longus. Ilab. in distr. Austin, inter Pindar et Wurarga, in arenosis aridis apertis floi*. m. Septem. Gardner 2620. I have referred this species to the Section Conogyne, although the stigma is almost that of (I. trachijtheca F. MuelL, whi(di Bcntham places in the Sect. Lissostylisy and from Avhich it differs in the much longer leaves, in being glabrous, and in the presence of a liypogynous gland. It is also close to G. didymobotrya Meissn., and (7. neinatophylla F. Muell., but differs from both in Ihe distinctly bilateral leaves which are convex above and deeply silicate from the closidy revolute margins; in addition it differs from G. didymobotrya Meissn. in the presence of a liypogynous gland, in the glabrous exterior of the perianth and the much shorter and thicker style; from G. ncmafophylJa it differs also in the simple inflorescence, much longer pedicles and shorter style, while the stigma separates it from both. 170 Chaiiles Avstix Gardner. Grevillea candicans C, A. Garchi. sp. nov. Section Cycladenia. Fi’utex ereetusj 2 m. altus, vamis erectis, ramnlis serieeo-pubeseentibus ; foliis erectis, pinnatis, pinnis 3-5, angustissime linearibns, rigidis, supra eonv(‘.\is, uninervis, subtus bisuleatis, glabris, aciitis, apiculo xxstulato praeditis; raeemis plei‘um<|ne singulis, tonninalibxis, ereetis, plixsminusve Jaxitloris; podnnculis brevissime pilosis; pediccdlis brovissinxis patentibus, pilosis; ptu’ianthio albo, extn.s pilis breviljiis vestito, intixs g’labx'O basin exeeplo, sub linil)o ovoideo-globoso vevoluto; toro redo; glandula hypogyna conspicua, annixlitonni, ovario globose, glabro, longe stipitato, non viscoso; stylo albo, elongato, Jeviter i)Iano, disco oblique laterali. Frntex 2 ju. altus; folia usque 17 eiu, longa, ])uitiae usque 14 <'ni. lougae, 1 uuu. latae; rav-eini 9*M etn. longi; ])eriauthiuu] 1 eui. longuui, stylus 1.5 cm. loiigus. flab, in disti'. Irwin ad lines septentrionales, ab Galena se]>tentrioneni ciiM'u (>0 kin. versus, in locis areiiosis apertis, dor. in. Septeni. W. E. BlackaU 4718. Typus in M(n*b. Pertli. AKinity to G. lencopteris Altnssn., tliifering priiici])ally in the I'edueed inlioreseiniee (usually a siiiijile raceme or very rarely 2 together), the shortly hairy perianth, fewer Itnif-segimmts, axis of the raceme &c. STEANGEA Aleissu. In working ovin* some S])ecimens in tlu' Herlin lierliarium in August, 1937, 1 found a s] lecimen of Straiujca Jittsarls Aleissn. in fruit, which I immediately recognised as congeneric with a plant formerly known as Hdkca f stenocaypoicle:^ Y, Aliudl. ex Benth., and descu'ilied in this Journal xix. 79 (1933) under the name Diplopter^ .^tenoenrpoh'les ( P. Muell. ex Bfuith.) €. A. Gnrdn. Still later, in January, 1949, whilst colh'cting at (’oclvlcsholl Gully, I saw Styaiujea cifanehicarpa in both llowau’ and fruit, and through the courtesy of Mr. C. '\\ White, the Go\'ernm(*ni Botanist of Queensland, I secured fruiting material of S. linearfs. There is no doubt that the three species constitute a separate and distinct genus, but in no case have tnau ioxxs authors described the seed correctly. This is solitary and ixuidulous on a slender funich', and togetlu*r Avilh the fiinicle enclosed in a 2-s(n‘iate ondocarj); the inmu’ layei’ ol' the endocarp conforms to tlie shape (tf llie cavity of the follicle and tnUii'cly ss peidladly developed although ready U) split to tlu' base into tAA'o nnri'ow convex and ratlnu' thick valves .... with one single unrijte blackish seed Avhicli is ten lines long, 3 lines broad, lanceolat(', tapering at both ends, flat and membranous, bordered on one sidt^ Avith a sliglitly thickened neiwiform margin.” 3'his description refers to the endocarp of the fruit and not to the seed itself. Alueller, in describing the same s])eeies (S. linearis) refei’s to the ^SA'ings of the testa at the apex suddenly and obliquely obtuse, at the base gradually cusixidate; the keel of the investing membrane is seen at tlie margin of the fol]iclo.”t 171 COJSITRIBUTIONES FLORAE AuSTRAHAE OCCIDENTALIS XI. Druiiimoiulj ’wlieii describing the seed of S. cynuyichlvarpa speaks of ‘hi single seed about an inch long, shaped like the seed of the Ash, each seed being enclosed ])y three membranous coverings, the outer and inner being smooth and l)rown, not unlike the wings of the seed of Baicea atid Banksia in substance, Ihe middle membrane being of a light brown colour and very brittle.’^ Mueller, in describing the seed of Strangea cgnanchicar pa si)eaks of tAvo seeds, and proceeds to describe in some icem hoilzontalibus vel reflcxis in membrana disciformem amplam integram coalitis siccando nigresccnte. Planta 50 cm. alta; folia usque 1.5 cm. louga; periantliium fruetiferum 5.5 mm. diam., 4.5 mm. loiigum. Ilab. in distr. Austin inter vicos Meekatharra et Wiluna, in campis lutosis iiitrosis, fruet. m. July, etiam prope Leonora, fruct. m. Septem. Gardner. Affinity to K. pyrajuidaia differing in the slender erect habit, in the total absence of spinescent branchlets, in being almost entirely glabrous, with acute leaves and .smalleT' fruits with an undulate more or less reflexed wing. CONTRIBtTTIONKS FLORAE AuSTRAI.IAE OCCIOENTALIS XI. 173 AMARANTACE.^. Trichinium albiduni C. A. Gardn, sp. nov. Section Parviflora. Frutieulus 30-60 cm. altiis, eopiose rainosus, rnmulis erectis, teretibns, albo-tomeiitellis ; foliis alternis, crectisj liiiearibus, sessilibug, acutis, crassiiis- ciilisj toineuto adpresso brovi cineceo vestitis; spieis terminalibus, valde abbveviatisj ovoideis vel giobosis, erectis^ rhachi nlbo-tomentella ; bracteis latissimo-.siiboTbiciiIaribus, luscis, g'lal)ris, apiculatis; perianthii segnientis laneeolatodinearihus vei ovato-oblongis, apiee acutiSj glabris, eeteriim cxtns longe albo-pilogis ; iiitcrioribiis angustioribiis, omnibus brunneis vel viridibiis et hj'alino-mavginatis ; staininibus 5, in tnbum cnpnlnm ad basin connatiSy filamentis dilatatis; squamulis intrastaminibns nullisj ovano g'loboso, glabro, SGSsilq stylo olongato. Frutieulus 30-60 cm. altus; folia j folia 4-7 mm. louga; spica u.s*jue 1.5 em. X 1.3 ein.; braeteac (drea 3X4 mm.; perianthii segmenta 6.5 mm. loiiga. Hah. in distr. Austin inter IMeekatliarra et Wilunaj in lutosis nitrosis nudis, lior. m. Jnl. Gardner 2308. Nearest to T. arthrolasinm. (I'. Mucll.) Benth., differing in the smaller and shorter spikes, the smaller sessile leaves, and in the larger flowers with relatively shorter bracts. :mtmosace/E. Acacia daviesioider, C. A, Gardn. sp. nov. Section Continuee, Fruticnlns 30 cm. altus, ramis ramulisque ^-aide tortilibus, intricato- divaricatis, prominenter 10-sulcatis, costis resinoso-asperatis; phyllodiis Az decurrentibus, valde recurvis, tetragonis vel sub-pentagonis, prominenter 5- nervosis, rigidissimis, acriler pungentibus, iiervis scabris; catntulis pednneu- latis, solitariis, axillaribiis, iihyllodiis aeqnantibus vel brovioribus, ea. 8-10 floris; floi'ibus 4-meris; calyce tenni inaequaliter 4-5-lobato, glabro, lobis brevibiis, obtusis; petalis subliberis, calycem duplo superanfilms, ovatis vel ellipticis, acntis, crassis, marginil)us minute flmbriolatis ; ovario glabro. Frutieulus 30 cm. altus, 60 cm. latns; ]0iyllo(lia 5-7 mni. longa; pedunculi 5 mm. lougi; capitiiH 4 mm. dinm.; petala 2 mm. longa. Hab. ill distr. Irwin orientalis inter Latham et Buntine, in arenosis axiertis, 11. m. Stqitem. IF. E. Blackall 772 (Ty])us in Herb. Perth). Afhnity to A. voluhilis F. MiielL, and A. ataxipjiylla Benth., ditfering from th(‘ former in the pimgently acute phyllodia, more numerous ribs of the branches, shortly lobod glabrous calyx, and from the latter in the much shorter ])hyl]odia, absence of stipules, fewer-flowered heads, and the obtusely and unequally lobed calyx. Acacia Wiseana C. A. Gardn. sp. nov. Frutox 1-3 motralis altus, ramis ramulisque promhnn.-.er sulcatis^ viridibus, spinescentibus, intricato-di\ aricalis ; phyllodiis articulatis, perpaueis, caducis, linearibns, rectis, divai-icatis vel deflexis, acutis, vix pungentibus, unicostatis, ]flaiiis, marginibus norvo.^is ad medium marginis anterioris glanduligeris; nedunculis axillaribus, patentibus, tcmiibus, solitariis; 174 Charles Austin Gardner. capitulis globosis^ ca. 20-iioris; floribus 5-meris; braeteolis lineari-spathulatis ; scpalis lineari-spathulatis obtusis, sparse et breviter pubescentibus; petalis quam sepala dui^lo longioribus, oblanceolatis; acutis, levibus, eonnatis cleinde liberis. Frutex 1-3 m. altus; jdijllodia 1 cm. longa^ 1 mm. lata; pedunculi 1 cm. longi; ]}etahi 2 mm. longa. blab, in distr. Minilya jn’ope Wandagee, in frutieetis rii)ariis fluminis Minilya lutosis solum, flor. m. August. F- J. S. Wise et C. A. Gardner n. 3283. The affinity of this species is with A. si^ineseens Benti}., from which it differs in the smaller phyllodia Avhich are rarely present, the long peduncles and more numerous flowers, in the free sepals and proportionately longer calyx. A. spinescens is not recorded from Western Australia. Acacia anrea C, A. Gardn. s}). nov. Section 01 ig on e ii rce. Frutex; ramulis phyllodiis junioribus peduneulisque hirsulis; phyllodiis erectis vel patentibus, obovatis, obliquis, plurinerviis, nervis erassis, pro- minenter reticulato-ramosis, apice mucronulatis, basi abrupte contraelis, breA’iter ])etiolatis: stipulis ovatis, deciduis; pedunculis tlavido-hirsutis, axillaribus, solitariis, 2-ramosis, duobus l)racteis e basi ramorum secondarium ; bracteis ovatis, brunneis, concavis, hyalinis; capitulis globosis, ca. 3()-floris, densis, lloi'ibus b-meris; sepalis subliberis, oblongis vel oblongo-cuneatis, obtusis, extns dense pnb(\scentibus ; petalis liberis, quam sepala vix duj>lo longioribus, lineari-oblongis, basin versus attennatis, apice crassiusculis, aureo-pubescentibiis. Pliyllodiu usque 1 cm. loiiga, b mm. lata; pctioli 1 mm. loiigi; pediineuli commiiiii 1 mm. longi, jiediimnili sccoiidarii 6 mm. longi; capituli 4.5 mm. diam.; sepala ca. 1 mm. longa; petala 1.7 mni. longa. Ilab. in distr. Stirling, pi-ope Gnarniing, comm. W. E. BlackalL The affinity of this species is Avith A. Kmrpana Maiden and A. defiexa Maiden. It dilfers from the former in the compound inflorescence, the shape of the phyllodia, the fi’ee sepals and the petals; from A. defiexa it differs in the inflorescenee, the shape of the phyllodia and the broader calyx-lobes. Acacia Yorkrakinensis C. A. Gardn, sp. nov. Section Jnliflorre. frutex 2-,l meti'alis altus; I'amis ereclis vd patentibus, gracilibiis: ramulis angulatis, pruinosis, demum brunneo-purjnirascentibus, nitentibu.s; ])liyIiodiis erectis, oblongis vel oblongo-lanceolatis. paido falcato-curvatis. glauci.s, ainee obtusatis vel acutis, iiibu’dum obli(]ue mucronntis, basi breviter contractis, brevissime sed distincte petiolatis, rigidis, tenuiter longitudinaliter multi- striatis, costa cent rale A'ix (-onspicua, inarginibus ])romineutibus, erassis, pur p lira scent ibus ; st ijuilis nullis ; pedunculis axillarilms vel terminal ibus, binis vel raramiosis preeipue in i>artibus superioribus ; st)icis cylindraems, densis, (piam pediuu-uli aeiiuilongis \'el brevioribus; ealyce late cupuliformi, extus albo-pubes('enti, bi'eviter 5-lobato, loliis obtusis, ciliolatis, demum subliiieris; petalis liberis anguste oblanceolatis, acutis, calyce duplo longiori- bus; legumine non viso. Phyllodia 2-5 cm. (superiora minus), usque P2 cm. lata; pedunculi 1-1,3 cm. longi; spicae usque 1.7 cm.; petala 2.5 mm. longa. CoNTKiBUTioNES Klorae Ax:straliae Occidentaeis XI. 175 llab. in distr. Avon prope Yorkrakino, (-a. 18 kin. a Bun^iilla septen- trionaloin vorsns, in cainpis aronot^is aridis, flor. in Aug’. Gardner, Tho speoies is vory dose to A. signata Y. Muoll., ditlering in the shorter and very inudi liroader jihyllodia whieh are never linear, nor have they the aecuininate curved apex of the latter. It might bo considered as a variety of A. signaht but its haliit, and the much shorter and broader distinctly pruinose pliyllodin give it a very ditTerent ai)})(nii'ance, and these difTerences are quite as distinct as those separating many other sjiecies of this largo g-eiius. The pods when obtained may ]>ossibly exhibit furtlier differences, The.lohing of the calyx is variable. In one spedinen the calyx very deeply lobed, while in another from the same locality the lolies are short and almost hemispherical. The 'Western Australian species of Acacia are in urgent need of revision. CAERALPJXTACEdf. Labichea teretifolia C. A. Gardn. sp. nov. Frutex densus, sulnnetralis altus, ramis ramulis(iue appresso-])ubes- centibus, ])lusminusve int]’i(attis ; foliolis 3, digitatis, sessililms, liiieari- teretibus, supra canaliculatis, suhius ]irofunde unisulcatisj breviter pub<‘scenti- bus, aeriter pungentil)us : racemis terminalibus vel axillaribus. pam-ifloris; bracteis ovato-lanceolatis, cymbiformibus, acutis, pubescentibus; sepalis 5, exterioribus extus sericeo-iiilosis. inti’rioribus minoribus, glabris : jietalis flavis obovatis, obtusis vel retusis; staminilnis 2, antheris subsessilibus, una breviori (luam alt(‘ra, uterque obtusis; ovario sessili, aib()-\-iiloso, 2-()vulato; stylo glabro. Frutcx subiuetralis aitus; folia usque l.o ein. longa, \ix 1 lum. lata; pedicelli 4 nun. loagi; sepnla g a nun. longa; ])etala o.o min. longa; anthera major 5 mm., minor 3.75 mm. longa. Ilab. in distr. Trwiii, GO km. a llumine i\rurehison se])tt‘ntrioiia]em versus, in arenosis aridis in fruticetis, dor. iii. Augaist-Sept., Gardner 2575. Affinity to L. nitidtf Beiith., differing in the number of leaflets which is constantly tlirce, narrow-linear and snbterete, deeply sulcate underneath, in the much smalhu* flowers and tln^ 2-ovulat(‘ ovary. Labichea eremaea (\ A. G(frdu. sp. nov. hrutex nanus, diffusus \'el caespitosus, vix 50 cm. altus, ramulis albo- pubescentihus; foliolis 5, petiolo conuniuii brevi; stipulis parvis, subnlato- setaceis, per.sistentibus ; la minis ellipt ico-laiiceolatis, duris, valde conca\'is, maiginibiis iinolutis, crassis, brevitfU' petiolidati.s, in apieem ])ung'entem attenuali'^, basin versus angiislatis, ])i'ominenter uninerviis, marginibns crassis, utrimpu* sparse* pubescentibus; lif>ribns saepius binis in axillis snperioribiis, rariu.s solitariis; pedicellis (piam foliolis brevioribus; sepalis 5, exteiioribus glabris, ]>lanis; petalis obovatis, calycem subaequalibns : stamina 2, antluu'is inaequalibiis, snbsessilibus, nna breviori (juaiu altera utrisf|ue obtusis; ovario sessili, alboA‘iIloj-o. ovulis 2, stylo glabro. Frutex vix 50 cm. altus; foliola usque 1.5 cm. longa; pcdieclH 3-4 mm, loiigi; sepala 7 mm. longa; petala S mm. longa. Hah. in distr. Austin prope Anketell, a vico Sandstone ca. 25 km. ■occidentalem v(u*sns, in arenosis aridis aperfis, Gardner 2505. 176 Charles Austin Gardner. 1. L. punctata Benth. Key to the Western Australian species of Labichea. A. Anthers equal; undershrub Avith almost simple stems; leaves not pungently acute E. Anthers unequal; branched shrubs with pungently acute leaflets (or leaves). a. Leaves pinnate with an elon- gated rhachis b. Leaves digitate or simple. a. Sepals 4; petals 4; glab- rous shrub with mostly simple leaves 3-8 cm. long, or the leaflets 3 with the middle leaflet usually the longest, all rather broad 2. L. cassioides Gaud, 3. L. lanceolata Benth. Sepals 5; petals 4 or 5; branchlets and leaves pubescent; leaflets 3 or 5. less than 2 cm. long. T. Leaves narrow-ellip- tical or ('Iliptical-laneeo- late with involute mar- TI. Leaves, terete or very slightly com- pressed, deeply grooved underneath 4. L. eremaea C, A. Gardn. 5. L. teretifolia 0. A. Gardn. PAPILIONACE^. (rompholobium obcordatum Turcz. var. pachyphyllum C. A. Gardn. var. nov. Fruticulus ^ humilis divaricate ramosissimus, ramis ramulisque viseidissimis; foliis trifobatis, petiolo brevissimo; foliolis paivis lato-euneatis, retusis, glabris, crassis, supra convexis, viseosis, marginibus recurvis; stipulis minutissimis ; doribus solitariis, tenninalibus, pedieellatis; ealyce glabro, viscoso, profimde subaequaliter 5-Iobato lobis valvatis, aeutis; petalis quam calyx yix longioribus, flaA’is; vexillo orbiculato, emarginato, alis minoi’ibus, oblongis aeutis, earina obtusa; ovario subsessili, ovato, glabro, ovulis 4, stylo meurvato, subaequilongo, crassiusculo, stigmate minuto; legumine maturo depresso-gioboso, glabro, duro. Fruticulus circa 15 cm. altus; foliola 2-3 mm. longa et lata; pedicelli 3.5 mm. longi; calyx 6 mm. longus; vcxillum 6 mm. longiim, 7 mm. latum, Carina D.o mm. longa. Hab. in distr. Avon prope Yorkrakine, in frutieetis apertis arenosis, flor. m. Decern, et Jan. Gardner. CONTRIBUTIONES FlORAE AtJSTRALIAE OcCIDENTAEIS XI. 177 Mirbelia longifolia C. A. Gardn. sp. nov. Frutex ereetus, 1-2 m. altus, ramulis griseo-pubescentibus, angulatis; foliis alternis, linearibus, petiolatis, exstipulatis, erectis vei siibpatentibus, marginibus arete revolutis, breviter mucronatis, supra eonvexisj laevis, uninervis, subtus bisulcatis; raeemis terminalibus, elongatis, paucilioris, folia multo excedentibusj rJiaclie angulato, pubescent!; pedieellis alternis, brevibus; bracteolis dcciduis, verisimiliter setaeeis; calyee eampanulato, lol)is valde inaequalibuSy lobis supcrioribus inajoribus late truncatis, ad apiceni connatis^ retusis vel emarginatis, lobis inferioribus ovatis, acutis; jDctalis fiavis voxillo cuneato-orbifulare, alis dolabriformibus, carina obtusa; ovario breviter stipitato, ovoideoj glabro, 2-oviilato, stylo ereeto^ erasso, uneinato. Frutex 1-2 m. altus; folia usque 4.5 cui. longa; calyx 9 mm, longus; segmeiita postica 3.5 nim. louga; vcxillum 9 mm, loiigum; alac 7 mm. loiigae, ■Carina C.5 mm. long:xj ovarium cum stylo 7 mm, longum. Hab. in distr. Avon j^i’ojxe Morawa, in fruticetis glareosis, flor. m. Septem. Gardner 2f)71 (Typus), necnon circa oppidulum Mullewa, in arenosis lapidosis, 11. in. August, Gardner, Neai- j1/. taxifolkij but inter alia a much larger shrub with sparse and longer leaves, slender elongated racemes Avith fewer floAvers and an angular rhachis closely pubescent and not villous. Mirbelia taxifolia C. A. Gardn. sp. nov. Frutex ereetus, densus, ramis ramulisque cinereo-pubescentibus ; foliis sparsis, raro apicem ramorum versus oi)positis, linearibiis, breviter petiolatis, exstipulatis, erectis vel patentibus, marginibus arete revolutis, breviter mueronatis, supra sulcatis, viridibus, minute asperatis; raeemis densifloris, terminalibus, folia multo exeedentibus, rhache villosa; pedieellis brevibus, aparsis, bracteolis linearibus, dense ciliatis, acuminatis; calyce eampanulato, sericeo, 5-nervis, segmentis valde inaeqiialibus, lobis supcrioribus majoribus late truncatis; petalis aurantiacis, vexillo cuneato-orbieulare, alis dolabri- formibus, carina obtusa; ovario glabro, breviter stipitato, ovoideo, stylo ■erecto, glabro, crasso, a]>ice uneinato; ovulis 2. Frutex solemnitcr 1 m. altus; folia 1-1.7 cm. louga; calyx 9 mm. longus, tegmenta postica (labium superum) 4 mm. louga, 6 mm. lata; vcxillum 1.2 cm. ovarium cum stylo 8-9 mm, longum. Ilab. in distr. (’oolgardie projie Karalee, in fruticetis arenosis apertis, flor. m. Soptem. Gardner. 4875. Affinity to M. aotoides F. MuelL, M. longifolia C, A. Gardn. and M. densiflora C. A. Gardn. Mirbelia densiflora C, A. Gardn. sp. nov. Frutex ereetus, densus, ramis ramulisque erectis, cinereo-vel albo- pubescentibus; foliis sparsis, densis, linearibus, erectis, breviter petiolatis, exstipulatis, marginibus arete revolutis, breviter mueronatis, supra eanalicu- latis, viridibus, aspernti.s, subtus bisulcatis; raeemis terminalibus, densifloris, brevibus, folia vix exeedentibus, subspicatis. rhache villosa; floribus subsessilibus ; bracteolis linearibus, dense birsutis, acuminatis, quam flores breviorilius; calyce late eampanulato, sericeo, enerve, segmentis subaequalibus, lobis supcrioribus ad medium- connatis, omnibus acutis; petalis aurantiacis, vexillo cuneato-orbieulare, alis subaequalibus, dolabriformibus, carina breviore, obtusa ; ovario subsossili, ovoideo, 2-ovulato, glabro, stylo erecto, uneinato, crasso, stigmate capitate. 178 Chakles Austin Gardner. Frutex 30 eni. altus; folia usque 1,3 cm longa; calyx 0.5 mm. longiis, lobi 3.2 mm. longi; ^■exillum 10 mm. iongum ac latuni^ alae 8.5 mm. lougne, carina 7 mm.; OTariiim cum stylo 6.5 mm. Iongum. Hab. in distr. Coolgavdie ad fines meridionali Young Kiver, Gardner Jany. 1936. The three species described above may be differentiated by the foilo-\ving key : A. Calyx very unequal, the hvo upper segments united almost to the summit into a truncate or emarginate lip. a. Kacemo'^ slender, loose, the rhachis angular and shortly pubescent; leaves 4-4.5 cm. long; shrub 1-2 metres high. . . . . . . . . M. longifolia b. Kacenu^s dense; rhachis not angular, densely villous; h*aves 1-2 cm. long; shrnb less than a metre high . . . . . . . . . . M. taxifolia E. Calyx regular or almost so, the two ].>osterior lobes united to the middle; leaves 1-1.5 cm. long . . M. densiflora With the exception of M. ojcyclada F. MuelL, a species referred with some doubt to this genus, the three species described above differ from all otlier \Yes1eni Aiisfrnlian species in the number of the ovules which is con- stantly two. This brings the group close to 1/. reticuliita and 1/. aotoules, species restricted to Clueensland and Ncav South Wales. The habit of the plants is much like that of O.njlohiuni § Podolobium, but the ovary in each case is completely 2-eelled when young. The character upon which Mirhelia is established, namely the longitudinal division of the pod into two cells is not always readily observable in ilowering specimens. In 3/. dilaiata and 3f. florihimda it is at once npY>ar(mt, but in most of the other AVesteru Aus- tralian species the partial intrusion of the false dissepiment from either one or both sutures is (d’ten very imp(*rfect and easily overlooked. The develop- ment of this fnlso diss(q>iment reijuires further study. In 3/. racemosa Turez.,. for example, it is entirely absent, and was described as such hy Tin’czaninow and Eentham is in reality a development of cellular tissue characteristic of CalHstachys and some species ()f OiJCfjlohmmj e.g. O. graniticfini. 37. racewom is in fact a true Oaylohium, and ivas described by the author under the name Oxylohium BennettsH in this Journal. The type species has been matclied with the type number of Mirhelia racemosa at Kew, aiid the tAvo are identical. Oxylohium racemosum (Turez,) C. A. Gardn. comb. nov. {Mirhelia racemosa Turez.; Oxylohium BennettsH C. A. Gardn.) Mirbelia seorsifolia {P\ MuelL) C. A. Gardn. comb. nov. The ]Aod of this ])Iant is typical of Mirhelia. Alueller described the ovules as being tAVO in number, but in reality 4 and 6 ovules are commonly present. The sj)ecies had already been regarded as an anomalous Gastrolobium. — Gastrolohium seorsifoliim F. Aluell. CONTRIBXJTIONES FLORAE AuSTRALIAE OCCIDENTALIS XL 179 Gastrolobium pauciflorum C. A. Gardn. sp. nov. Section Axillares. Friitex divaricate rainosus, ramis teretibus, junioribus cano- pubeseentibiis; foliis oppositis. petiolatis, obovatis vel lato-oblaneeolatis, obtusis aeutis vel mucronatisj siibplanis, glabris, glaucds, rigidis, eoriaeeis, jxromineiitei’ pallide retieulatis; stipulis setaceis, fiiscis; lloribus 1-3, axillaribus, brevitev pcdieollatis; braeteolis parvis, rigidis, cuneatis, tridentatis, pul)cs(-(*nti})nSj pedicelli excodentibus; calyce dense sericeo- ])ube.seeiite, lobis qnnm tubus brevioribus, 2 su}.)erioribus truncatis vel emarginatis, bttioribus; petalis (juam calyx vix duplo longioribus, vexillo flavo, latissinio emarginato, alis vix brevioribuSj carina purpurascenti obtusissinia : ox ario longe stipitato, dense villoso, 2-ovulato, stylo faleato, lateraliter conipresso, glabro. Frutox forsan 50 cm. altus; folia usque 2 cm. longa, 8-10 mm. lata; petioli 3 mm. longi; pedicelli 2-3 mm. lougi; calyx 5 mm. lougus; vexillum 1 cm. loiigum ac latum. Hab. in dLsti-. Iiuvin, ca. 9 km. a Three Springs oecidentalem versus, in fruticetis aja-rtis areiiosis, fl. m. Septem. TF. E. BlacJcall 4895. Affinity to G. ohovatim Benth., differing in the flat, not complicate leaves, fewer- floAvered clusters without any evident peduneh', the shorter pedicels etc. The flowers are frofjiiently solitary in the axils. Gastrolobium Bennettsianum (\ A. Gardn. sp. nov. Section Racemosre. Frutex erectns vel ]mtens, 1 m. altus, ramis ramulisque anguiatis, einereo- puberulis; foliis ternis vel oppositis, breviter i)etiolatis, oblanceolato-com- plicatis, a.pice recurvatis, obtusis, pungento-mucronatis, eoriaeeis, glaucis, erectis; stipulis setaceis, firmis, petiolo aequilongis; racemis terminalibus vel axillaribus, folia multo excedentil)us, densi-floris, rhadie puberula; braeteolis ovato-laneeolatis, obtusis; calyce parce pubescente, segmentis 230stieis altius eonnatis; vexillo reniforme-orbiculare, alis obtusissimis carina aequilongis; ovario anguste-ovoideo, villoso, stipitato, stylo iiicurvo, latera- liter com])resso, inferne villoso, eeterum glabro. Folia 2 cm. longa; stijuilae fuscae, 2-2.5 mm. longae; racemi 4-5 cm. longi; pedicelli 1-1.5 mm. longi; ovarium 2 imn. (stipes 2.5 mm. cxcepta); ■stylus 1 mm. longus. In collibus glareosis regionis Encalifpii reduncae distr. Avon pro- l)rium. Adest ad Yorkrakine i>ro[>e Tanmiin meridiem versus ad usque AVagin, fl. m. Septem. Typus est North Bimgulla, Gardner vSept. 1936. Collogae assiduo in herhis venenatis investigandis, excellentissimo vii‘o Haroldo Gulielmo Bennetts, scieutiae veterinaris doctori, gratissimo dedicatnm. This species lias a close affinity to G. microcarimm Meissu., and G. flori- hnndum S. Moore. It differs from both, however, in the recurved apex of tlie leaf whi('h is furthermore distinctly pungently mucronate. In tliis respect it •closely resembles G. oboimtum Benth., with which it might easily he confused when not in flower or fruit, but in G. obovaium the leaf is always aenfe or acuminate. From G. niicrocarptim and G. oxylobioides Bentb, it differs in the complicate leaves, as Avell as in the shape of the leaf blade which does not taper to the apex. The loaf texture, and a cross section of the same is much like that of G. floribundutn, ])ut the shape, ns well as tlie floral characters ure quite different. 180 Charles Austin Gardner. The species is coiniMon throughout the area of its habitat which extends eastwards to Koudiniip and it appears to be constantly associated with gravelly hills and rises, usually in association with Eucalyptus reilunm var. elata. In coniinon with the other species of tlie Bacemosae it is highly toxic to both shee}> and cattle. Gastrolobium glaucum C, A. Gardn. sp. nov. yectioii Uacemosce. I rutex erectus vel diftusus, 50 dm. altus j raiiiis raniuliscjue vix angulatis, einereo- 2 )ubescentibus demum glabresceiitibus ; foliis ternis, sub- sessilibus aut brevissime petioJatis, late obovatis, coriaceis, consjHcue retinervis,. planiS; obtusis, glaucis, pungentomiuci-onatis, ad basin obtusis, costo centrale subtus \ aide proininente; stipulis setaceis, fuscis; raceiuis terminalibusj pluri- et deiisifioris, villosis, folia inulto excedentibus; bracteolis lanceolatis, deciduis; pedicellis villosis; calyee pubescente, se^gmentis posticis altius connatis, emarginatis, ceteris ovato-deltoideis ; vexillo reniforme-orbiculare,. alls acutis a el subobtusis, cai'ina excedentibus; ovario ovoideo. villoso, longe stipitato ; stylo com 2 )resso, glabro, ovulus 2. lolia^ us(juc 2 cm. luiiga ct 1.2 cm. lata; stiiutlae 4 mm. lougac; racemi S2>icifornii, 2-3 cm. longi; pcdicelli 1.5 mm. loiigi; calyx 4-5 mm. longiis, lobi I^ostici y\x 2 mm. loiigi, r('liidotis^ filamontis glaberriinis, iinearibus, petalis brevioribus; earpidiis lepidotis* Microcyhe paucijlora var. nnljlora 1). A. Herbert. I ruticulus 50 diu. altusj folia 4-5 mm. loiiga. Hab. in distr. Coolgardie, prope. Carrabin, in arenosis lutosis, ii m. Oct. Gardner; pr. Vellowdine, W. E. Blackall 11)37, pr. Hatter’s Hill, Hlackall 1867; in distr. Avon jiixta Ballidu in fruticetis, Gardner. Witli the liabit and app(‘arauce of a Microcyhe, with which it might easily be at first sight mistaken, this species, together Avith the following,, are distinguished from all other Western Australian st)ecies by the solitary sessile or subscssile flowers. Phebalium clavatum C, A, Gardn. sp. nov. Frutex erectus, eirciter 1 rn. altus, pluri-rainosus, rarnulis foliosis, sparse- glandixloso-tubercnlatis, dense argenteo-lepidotis; foliis late et obtuse clav'atis^ valde obtusis, petiolatis, dense lepidotis, glandula minuta terminale praeditis; floribus albis, solitariis, sessilibus; calyce pai’vo, acute 5-dentato, dentibu& erectis, tubo subaequalibus vel longioribus; pctalis leviter lateraliter imbricatis, extius denee lepidotis; staminibus subaequalibus, subexsertis, filamentis tenuibus; earpellis 5, ovoideis, obtusis, dense stellato-lepidotis. I rutox circ. metralis altus; folia 3-4 mm. loiiga; calyx 2.5 mm. longus ac latus; petala 4 mm. longa. Hab. in distr. Coolgardie, in arenoso-lutosis, prope Widgicmooltiia, flor. m. November. Gardner. MTJIKIAKTHA C, A. Gardn. nom. noA^ Mueller, in 1887, described under the name Chorilaena HasselUi, a plant collected by A. Y. Hassell from the west end of the Stirling Range. This was raised by me to generic rank in this journal xix., 83 (1933), und(‘r the name of Muiria HasselUi (F. Muell.) C. A. Gardn. A shoiT time before this, however, N. E, Brown bestowed the name Muiria on an Aizoaceous plant, so that a new name has to be found for the local genus, and the name Muiriantha is here projmsed. The description will be found under the reference quoted above. EUPHOKBIACEHtl. Euphorbia bobphthona C. A, Gardn. sp. nov. § EremopJiila. Herba perennis, radice crasso, contorto, albo-corticato ; (-aulibus dichotomo-divancatis, glabris, laetiferis; foliis lim'ari-Iauccolatis vel hhieari- bus, aeutis, remote d acute denticulatis^ in j)e1iolis brevibus attenuatis, inferioribus alteiniis, superioribus o})])ositis vel quasi o])])OHitis, omnibus plusminusve caducis; stipulis nullis; cyathiis axillaribus, solitariis vel binis, brevitor pedunculatis vel subsessilibus ; pedunoulis non articulatis; glandulis 4 vel. 5, latis, exappendiculatis ; ovario glabro, laevi, stylis hrevissimis, crassis, biramosis; fructibus subglobosis, trilobis, laevibus, seminibus eylindraceis, dense sclero- et albo-papillosis; carnnculo magno, pildforme, margine discoideo, sub somine constricto. 182 Charles Austin Gardner l-'lanta l(i-20 cm. alta; folia uscjue 3.5 cm. longa, 5 mm. lata; eyathium vix 2 mm. loiigum; stip(\s 4 mm.; fruetus 6 mm. longus; semina 4.5 mm. longa, 2.75 mm. lata^ carunciilus 1.7 mm. latus. Species ereiuaea Aiistraliao occidenlalis aream amplam occujjavit in loeis alluviis Intosis (([uos dicnnt ‘'Creeks-') e.g, Jimba Jimba atl lliiminem Gas- eoyne Kiver, dardner n. 3302 (Typiis) ; a Cue oecidentalem versus prope “Stock iC)ute/'‘ ill loeis similibus Gardnerj a La\^erton orientaiem versus WarluiJ'ton Hatige, E. de C. Clarice j frucl. m. July-Augiist. The affinity of tins species is with E. ciutioides Forst., from which it can be distinguished at once by its lower more widely and dichotomously branched habit, by its acute leaves^ and by the seeds which are densely co^'ered with large haj’d and white papilla-like excresences naiTowed towards the base, and, furthermore, the caruncle is almost hat-shaped, and its lower disciform end is separated from the seed by a short and narrow stalk or constriction. Euphorbia hravpM/nji/a is a cyanogenetic plant toxic to stock, especially hungry travelling stock, and has accounted for the death of thousands of -sheep and cattle in tin* Gascoyne and ifurehison districts, especially in the vicinity of Jindia Jimlia, the Lyons River, and in places north-west of Cue. Euphorbia ciutioides (Forst f.) C, A. Gardn. comb. nov. {Croto7h clutiokles Forst f. Prodr., 92 (1786).) (Enplhor}>ia htnnensi?i Sprtmg., FI. Hal. Mant. 42 (1807).) [E, Vieilardi Baill, Adansonia ii. 212 (1862).) (E, eremophila A. Cunn ex Hook, in Mitch. Trop. Austr. 348 (18-18).) Its range extends from the Island of Tanna in the New Hebrides to West- ern Australia, eml)racing New Zealand and the northern half of Australia. E. hodphtlwna is so far only known from W(‘stern Australia, but it appears prol)able that its range will be found to extend into Central Australia. E. ciutioides does not appear to be toxic to stock. Key to th(‘ si)(‘('ies of the Sect. Ereuiopliila occurring in Western Aust ralia. A. Seed smooth or covered with minute whitish tuliercles; caruncle em- bracing tlio end of the seed. a. Involuci’al glands llmbriate- ciliate, prominent ; seeds smootli; leaves acute .. F. Finlaysoni J. M. Blade. b. Involucral glands entire, broad, inconspicuous ; seeds granu- lar rugose OT- nearly smooth ; leaves mostly obtuse . . F ciutioides (Forst. /.) C. A. Gardn. T5. Seeds prominently schro-papillose with elongated club-shaped white excrescences ; caruncle separated from the seed by a thin but distinct stalk; iii- volucral glands exappendicu- late . . . . - . . . E. bobphthona C. A. Gardn. E. Stevenii Bail., has not yet been recorded from Western Australia. Tt belongs to the above Section. CONTRIBUTIONES FlORAE AtJSTRALIAE OcCIDENTALIS XI. 183 RH AMXACE/E. Cryptandra grandiflora C. A. Gardn, sp. uov. Fmtex ramosissimus, ramis divaricatis, ramulis iiovellis appresso- pubescentibusj denmm glabris; folds plusminusve subfascieiilatis, obloiigis vel oblongo-obovatis, petiolatis, basin versus sensiin augustatis, apiece obtusis^ mueronido rocurvo praeditis, iiovellis complicatis, dense sei‘iceo-[>uboseentibusj. adultis supra olivaeeis et glabroseentibus, subtus scrieeo- et a]>presso- pubeseentibus; stipnlis selaeeis vleiunni deeiduis; floribus albis, in pseudo- eapitulis teriniualibus aggregatis ; bracteis ovato-orbieularibus vel late obovatis, obtusis vel retusis, mueronatis, inarginibus eiliatis^ riuani tubum ealycis brevioribus ; braeteolis spathulato-obovatLs braeteis subaeqiuuitibus, toto dorso piloso; ealyeis tubo aiiguste caiiipanulato, extus paueis pilis tenuis sericeis adpressis, lobis angusto-triangularibas, acutis, intus glabris et prominenter earinatis; peialis unguieulatis, cucullatis; disco angusto, annulari minide jiubeseente; ovario vertice a]bo-i)iloso ; stylo crasso, quam calyx breviore. Fiutex forsan metralis altusj folia 6-15 iiini. loiiga, majoi'a in ramulis novellisj bracteae bracteolacque usque 5 mm. longucj calyx 6-7 mm. iojiguSj lobi eirc. 2.5 mm. longi; petala .75 mm. longa. Hab. in distr. Irwin ab Carnamah oeeidentalem versus, 11. m. Sept. E, M. Barli'er, per IF. E. BlacIwU. Aflinity to 0. hmeophracta, differing in the larger leaves, larger and shortly but distinctly pedicellate flowers Avhich are more numerous in the cluster, in the vestiture of the calyx which consists of fine silky hairs, and the longer and more acute calyx-lobes. BLACKALLIA O'. A. (ran(n. gen. noAv Calyx urceolatus vel ovoidems, tubo ultra discum longe pi'oducto, lobis 5, eonnatis, erectis, acutisj petala 5, breviter stijritata, in ore calvcis inserta,, eueullataj stamina 5, cum petalis inserta, hlamentis breviter liberisj antherae versatiles, oblongae, rimis 2, longitudinaliter dehiscentibus discus hypogynus fundo tubi calyeis adnatus, annularis, margine bbero, integro \el sinuatoj ovarium ovoideum vel obovoideum, fundo culycis rectum breviter attenuatuaq stigmate 3-lobo. Capsula ovoidea vel cylindracea, libera, coriacea, 3-valvis, eoccis erustaeeis; semina erecta, cylimlracea; fuuieulo crasso-turbinato vel subcupulato insidentia. Frutices ramosissimi, ramulis alternis, spinescentibus ; folia fasciculata, pax\a, stipulae fuscae, par\'ae, flores in I'amulos breves subraeemosi vel fascieulati, breviter pedicellatij bracteae imbricatae flores cinctae vel nullao- Amicissimo cIarissimo([ue viro, Gulielmo Fdwardo Blackall, scientiae medicinae doctori, novum hoc genus dedicoj qui socius humanissimus et collega de rebus Australiae occideutalis botanicis maxime eiaiditus, saepe saepius ])artes huius regionis incultas mecum peragrinavit. A genus closely related to (.h’yt)tandra, of which it has many of the characteristics, but the completely superioi' ovary and the small annular free-margined disc are characters of the Tribe Colletieae, of which Blackallia is typical except that the branchlets and leaves are alternate and not opposite. It is close to Disearia, differing in the erect + connate calyx-lobes, the capsular fruit, as well as in the arrangement of the branchlets and leaves. The calyx does not appear to spread its lobes, which in all the specimens examined are + connate and erect. 184 Charles Austin Gardner, 1. B. connata {C, A, Gardn.) C. A. Gardn. comb. nov. {Cryptandra connata C. A. Gardn. in Jour. Roy. Soc. AVest, Austr. xxiv. 80. (1928).) Additional localities are — Near Payne’s Find, W. E. Blackall; between Lawlers and Sandstone Gardner, 2. B, biloba C, A. Gardn. sp. nov. Frutex 30 cm. altus, ramis divarieatis, ramulis altei’nis spinescentibus; foliis fasciculatis, lineari-cuneatisj basi attenuatis, apiee bilobatis, lobis obtusis, glabriSj supra concavis vel plusminusve complicatis, glabris; floribus singulis vel fasciculatis e fasciculi foliorum, longe pedicellatis; braeteis nullis; calyee gabro, cylindricOj basi saccati, lobis deltoideis, acutis, ei’ectis, petalis subsGssilibus, eucullatis, calycis limbi semi-aequantibus ; disco tenuo glabro; ovario glabro, obovoideo, stylo recto, minute trilobate. Fnitex 30 cm. altus; folia iis<]iie 6 lunu loiiga, 1,5 nim. lata; peduiiculi 4 inui. loTigi; ealys 5 nun. longus, lobi vix 2 nun. Hab. in distr. Irwin inter ^■ieuln Northampton et Lynton, dor. m. Sept. W. E. Blackall 4517. Differs from B, connata in the long pedunculate flowers, the absence of bracts, narrower and entirely glabrous calyx, and in the shaj^e. of the leaves. MYRTACE^IC Eugenia Stokesii C. A. Gardn. sp. nov. Arbor 7-lU ni. alta, glabra, ramulis teretibus; foliis ovato-oblongis vel ellipticis, ])etiolatis, basin versus attenuatis, apiee ohtusis vel subaciitis, crassis, rigidis, glaucis, irregulariter iienninervis, vena intramarginali ab margine remota, nervis sceondariis retieulatis ; cymis axillaribus, pauci- (4-7-) doris; calyee canipanulato, lobis plusminusve inaequalibus, late hemis- phaerieis, persistentibus, marginibus tenuibus, tubo supra ovarium in disco concavo staminifero producto; petalis orbicularibus, albo-marginatis, delapsis singulis; staminibus (|uam 7)etala longioribus; ovulis cireiter 10. Arbor 7-10 ni. alta; folia 7-12 cm. loiiga, 5-0 cm. lata; i)etioli 4.0 mm. longi; pedicelli 3 mm. longi; calyx I cm. toiigns; petala 9 mm. longa, 10-11 mm. lata. Ilab. in regioiie tropica ad AVeber Range, ad dneni orientalem teritorii Australiae occidentalis, dor. ni. July, S. J. Stokes n. 51. Boundary Survey of AVestern Australia, 1937. Aflinity to E. pendens Dutliie, but with glaucous thick broader and obtuse leaves, the calyx-lobes not redexed after dowering but remaining erect, short pedicels and eampaniilate calyx. Among its Australian con- geners it is closest to E. myrtifolia Sims, but the leaves are much broader and with much coarser venation, and the calyx-lobes are very obtuse. Eucalyptus megacornuta ('. A. Gardn. sp. nov. (*5 Cornntae). Arbor 7 metralis alia, ramis erecto-patentibus, cortice laevi, einereo- brunnoo; foliis primariis non visu; foliis ordinariis alternis, oblongo- lanceolatis, concoloilbus, nitenti-vhldibus, petiolatis, acuminatis vgI obtusis, venis second ariis inconspicuis, vena intramarginali juxta margini; pednnculo elongate, pendulo, lorato, apicem versus incrassato, dores 2-3 sesislcs gerente; calyee elongato-campanulato, apicem versus dilatato, multo-costato; operculo cylindrieo, verrneosissimo, obtuso, quam calyx duplo longiore, basi abrupte dilatato; staminibus multi-seriatis, viridibus, erectis vel dexuosis, dlamentis CONTRIBUTIONES FLORAE AuSTRALIAE OCCIDENTALIS XI. 185 .quadrangulatiSj glandulosis; antheris loiigitudinaliter dehiseeiitibas ; friietu inagno, campaniilato ve! liemisphaerico-eampanulato, rugoso-costato, margine angiisto; disco heniisphaerico-dopresso, striato, valvis erassis, inflesis, leviter ■Rxsertis. Folia usque 8 cm. loiiga, 2.5 cm. lata (petiolus ] cm. lamina 7 em.)j pedunculus usque 4 cm. longns, 1.7 cm. latus; calyx 3 cm. loiigns, 2.2 cm. latus; operculum 5 cm. loiigum, ad basin 2 cm. latum; fructus 3.5 cm. lougus, 3.2 cnu latus. Flab, ill distv. Coidgardio ad tines australes prope Kaveiisthorpe septen- Irionem versus in clivis glareosis, flor. m. November, Gardner. This species is very close to E. Burdettiana Blakely, dilTering in its arborescent pi’oporlions, its widely branched habit, much larger buds and flowers, the markedly verrneo.se and broadly dilated operculum, the larger costate fruits, and the broadtu* leaves. E. Burdettiana is a small mallee inhabiting the quartzite hills of the south coast, between Flopetoun and the Fitzgerald River. It rarely exceeds 2 metres in height, and becomes smaller towards the western limits of its habitat. The fruits of E. Burdettiana are much smaller than those of E. megacornuta, smooth or unicostate, and almost .spherical. It is very close to E. cormito. Eucalyptus erythrandra Blakelg. Ill this Journal xix 88 (19.33) 1 described as a new variety of E. angulosa Schan (vnr. rohnsta), a jdant collected by 11. Steedman near Kundip. I liave since received, through the Conservator of Forests, specimens of this plant collected by ]\Irs. Daniells of Hopetonn which exhibit a perfect scries embracing on tlie one hand E. tetrapiera Tnrez., and E. angulosa Wchau. on the other. Amongst the intennediate forms is typical E. erythrandra^ which I consider to be a hybrid. Tlie evidence in favour of this theory is quite clear. Eucalyptus Brockwayi G. A. Gardn, sp. nov. Arbor 25 metralis alta, eortice laevi, albo vel salmoni-colorato, deciduo, ramulis acutanguHs; foliis alteruis, coriaceis, Fnleato-lanceolatis vel lineari- lanceolatis, concoloribus, utrinque nitentibus, punetatis, venis lateralibus peimato-patentilins, vena intramarginali a margine remota; umbellis axillari- bus vel lateralibus, solitariis, patentibus; peduncnlo leviter comprosso; pedicellis brevibus angulosis vel subteretibus, calyeem multo brevibus; tubo calycis paene In’ovicylindrico, ad basin obtusissimo; oporeulo hemisphaerieo vel late ovoidco; staminibus ante expansionem inflexis; stylo recto; fructu parvo, ocostato, nrceolaio-globoso, apice contracto, margine elevate; valvis lirofunde inclusis; seminilius fertilibns subovatis, eompressis, laevibus, nigrescentibus. Folia usque 10 cm. longa, 1.2 cm. 1:ita ; jiediiuculis 1 cm. longis; pedicellis 1 mm. longis; calycis tubus 4 mm. lougus; operculum 1.7 mm. longum; fructus 6 mm. lougus, 5 mm. (liametrus. Flab, in distr. Coolgardie, prope Norseman, in encalyptetis lutosis, G. E.iBroclmmy et ^7. A. Gardner, 15. Decern. 1940 n. 5598. Affinity to E. ovularis Maiden, differing in the venation of the leaves, the much shorter pedicels, the very obtuse operculum and globular-urceolate fruits. The fruit is much like that of E, Flocfctoniae Maiden, but the ■operculum is very different. It is probably also close to E. Cooperiana F. Mud!. An examination of the immature anthers indicates that the species 186 Charles Austin Gardner, may belong to the Flat jjanthe rue, but this is not certain. It is also not ahvays ]>ossible to clearly define the species of this Section, since in some there is a definite ai)proacli to the vS Macrardherae. The anthei's as seen are short, :md the dehiscence ai)pears to be by Avide lateral slits, but in the absence of Ihjwm-ing material I think it is perhaps best to group this species, together Avith E. ovalari^ and E. FJocldoniae in the ^ Flafijantherae^ The si)e('i(‘s, whii-ii attaijis a lieight of 80 feet, has a })erfectiy smooth salmon-coloured or AA'hitisli bark Avhich in fracture is blood-red and gummy. It is fairly high in tannin. The heartAAmod is reddish like that of E. salmonopliloia^ and the tAVO trees look much alike in the field. The species is named out of com})liment to (ieorge Ernest BrockAAuy, Divisional Forests Officer at Kalgooriie, aaOiosc collections of Eucalyptus in his district, and the intert'st he has tak(m in the trees of the area, haA'e added much to our knoAV- ledge of the genus Eucalyptus in eastern areas. Eucalyptus brachyphylla C. A. Gardn. sp. nov. .Vrbor parAu, cortice aspero jiersistente in parte tninci inferiore, deiude in ramis lacA'i et deciduo, I’amulis pi’uinosis, teretibus; foliis oppositis, breviter sed distincte petiolatis-; laminis ovatis vel ovato-lanceolatis, breviter acutis, basi obtusis, glaucis, ]>unctatis, venis lateralibus obseuris, vena intramarginale a marg'ine \uilde rf'inota ; innbcfilis paucifloris (5-7) axillaribus vel in paniciilas breA’es terminales et laterales collectis ; pedunculis tereti-eompressis, pedic(‘llos longitudine excedfuitibiis; alabastris ol)ovoideis, apiculatis vel rostrato-acutis, longitudine jiedicellorum vel longioribus ; calycis tul)o longi- tudine operculi vel longiore; fructibus <*ylindrai'eiK, ecostatis, truncatis, apici vix constrictis, inarginibus tenuis, <]uadriloculatis; valvis p>rofunde inclusis, deltoideis. Arbor altitudine 4 ni., cortex in laminis dnris secedens; folia plcraque 3-4 cm. longa, iiscjue 2.5 cm. lata; pednnculi 1 cin. lougi; pedicelli 3 mm. iongi; operculum (in statu immaturum) 2.5 mm. longum ; fructus 5 mm, longuSj 3.5 mm. latus. llab. in distr. Coolgardie, ]>rope Lake Cowan oceidentalom versus, juxta. Binya)'inyinjia, G. E. Brockwai/j duly, 1040. Near trt E. Kriis/^ana E. MuelL, Avith which it is associated, differing in the distinctly ptfiolate ovate to (O'ate-lanceolate rather acute and larger leaves, and in the long(*r and difi'erently ‘^Iiaped fruits which are not, or very slightly constricted at the summit. Eucalyptus Forniani C. A. Gardn. sp. nov. Arbor 10 m. alta, cortice cinoi’ascente, riigoso, in trunco ramisque persistente, ramis ultimis laeA’is; foliis primariis erectis, angusto-linearibus, erassis, fere seraiterelibus, glaueis A'el pruinosis, uncinato-acuminatis valde innudatis, utrinque subcanalicniatis ; foliis ordinariis angusto-linearibus, reel is a cI ](‘viter falcato-ciu'A-atis, (U-(>ctis, distincte petiolatis, apice uncinato- acuminatis, costa centvale utrimiuo impvossis, ceteris enerviis, Auridibus, glandidoso-pnnctnlatis; umbellis axillaribus, 5-8-floris; peduncnlo tenuo, apicem A'ersus incrassalo; ]H‘dicellis seraiteretibu.s, qnam alabastrum breviori- bus ; calyco liemispbaerico-turbinato, glandnloso, qnam operculum conieum acuminatum breviore; filameutis inflexis, antheris parvis, orbicAilaribus (matuvis non visibus) ; fructu globoso-hemis|)haorico, laevo, margine elevato,. disco concavo, valvis deltoideis, inclusis. CONTKIBUTIONES FlORAE AuSTRALIAE OcCIDENTALIS XI. 187 Folia primaria 3-3.5 cm, longa, .75 iiini. lata; folia ordinaria 5-7 cm. longa, 2 nun. lata, rare 3 mm. lata; pediinculus S mm. longiis; pedioelhis 3.5-4 mm. longus; calyx 2 mm. longus; operciilnm 3-4.5 mm. longum; fnictus 4.5 X 2.5 mm. Hal), in distr. Coolgardie ad titles boreali prope “Die Hardy Eange/^ a vico Southern Cross, 175 km. septentrionem versus in frntieetis apertis arenosis, F. G. Forv}cm et R, S. Matheson^ July, 1940. Althongii dowers have not been seen, this species can be delinitely assigned to the ^IHafyantherae. It is close to E. Kocliti IVIaiden & Blakely, but is a fairly large tree with the liark of E. longicornis F, MuFl.; the leaves however are narrower and conspicuously oil-dotted with a + impressed midrib, tlie peduncle is more slender and elongated, the calyx much shorter and hcmisphericnl-turbinate, and the operculum elongated ovoid-rostrate or + acutely conical with an erect beak which however is variable in length. The fruit is not as long as broad, and tlu' deltoid valves are not exserted. The primary leaves are quite unlike anything kjiown in the genus, being almost terete. This species ctuimiemorates the name of Francis Gloster Forman, Govern- ment Geologist of AVestern Australia, who brought me the first specimens of this ])lant. Eucalyptus orbifolia F. MacU. I have rcc(*ived fi'om Mr. G. E. Brockway, .Divisional Forests Ol'ficer at Kalgoorlie, through tl\e (k.>nservalor ol' For(^sts, (S. L. Kessell, Esq.), complete matei’ial of this si)ecies from tlu' Hosi>iLa] Ko('ks (West of Mulliue), and from Pigeon Rocks north of Bullfinch. (k)nsidei5iig th<^ inte)‘est that has surrounded this long lost species, of which only the small ty])e si>ecime)i in the Kew Hcu'harium was previously known, the re-discovei\v of the sp(*cies is of exceptional iuter(*st. The folloAving description is taken from i\Ir. BrockwayV si>ecimens : — Mallee, attaining a height of 25 feet, although usually about 18 feet high (Brockway) ; stems 5-() cm. thick, with a u'aiun red flaky striated thin persist- ent bark Avith decorticates leaving a i>ale green bark lietween the strijAS, heart- A\mod pale nal, closely grained. Bi-anclilcts reddisli-pruinose, terete. L(‘a.ves alternate, l)roadly obovate-oi'bicnlar, shortly tapei'ing at th(‘ base into a slender jx’tiolc of 1.5-2 cm., th(‘ lamina 3-7 cm. long ami al)oiit of ecpial breadth oi' even broadm*, obtuse to retuse, glaucous, rather ])rominent!y veined, the secondary veins roughly parallel but i'(miot(*, tlu‘ intraniarginal vein distant from the margin. Peduncles axillary or lateral, erect-spreading, straight, pruinose oi‘ reddish, slightly 2-ang]ed, slightly thickened towards the a])ex, 1.5-2. 5 cm. long, hearing an umbel of from 2 to 5 floAvers; pedicels thick, subter(‘te, 5-9 mm. long, divaricate; calyx-tube almost hemispherical, but shortly tapering into the ])(Mlicel, 5 mm. long, and 8 mm. In-oad; opercidum hemispherical-conical, not mvieh longer than the ealyx-tube, usually about as long, longitudinally striate, the striae not V(U'y conspicuous, glaucous, as well as the calyx and ]>edieel; stamens pale yellow, immei’ous, the filaments flexuose in the Inid, 5-G mm. long; anthers oA^ate, opoiing outwards in longi- tudinal slits, the giatjd .small, dorsal; style thick, conical, thicker tlian in any other species obserA'ed. Fruit liemis])herieal-campanulate, 1 cm. long. 1.7 cm. broad, smootli oi* striate, pruinose-glaucons, the disc broad and fiat, tlie valves subulate-deltoid, prominently exserted; seeds black, triangnlar-cuneate, not Avinged. E. Wehslen'anti Maiden, is very close to this si)ecies, and perhaps a varietv onlv. 188 Charles Austin Gardner. Baeckea leptospermoides G. A, Gardn. sp. nov. Frutox erectus^ metralis altus, giaber, rainulis strictisj erectis, teretibus; foliis erectis, petiolatis, oblongis, crassis, planisqtie, obtusis, basi breviter eontractis rotundis, peliolis tennibus; floribus solitariis, albis; pedicellis axillaribus, qiuim folia multo longioribiis; bracteolis 2, late ovatis, aciitisj earinatis, a medio peduiiculi inscrtis; calyee late cupuliformi, lobis quam tubum brevioribus, semiorbicnlaribiis, marginibus tenuibiis, albis ; pcdalis quam ealyeis tubum triple loiigioribus, orbicularibus, albis; stamiuibiis 25^ aniiulo unduJato disco margiuc sinuate iiiseilis, lobis disci extra stamines processus subulatos gerentibus, aliqiii staminibus regulariter dis])ositiSj subacquis, quam petala dui)lo vel triplo brevioribuSj Hlamentis teretibuSj tenuibus, ineurviSj antberis dorsifixis, longitudinaliter deliiscentibus, glandula albida, eonsjjicua, giobosa^ quam anthera dimidio vel Imeviore; ovaiio triloeulari, ovulis numerosis, plaeentam peltatam allixis, stylo brevi,- tilamentis aequante. Frutex 1 m. altiis; folia usque 5 min. ionga; calycis tubus 4.5 mm. lougus ac latus. lobi 1.5 mm.; petala 4 mm. diam. Ilab. in distr. Coolgardie prope Karalee, in fruticetis arenoso-glareosis, fior. m. Sept. Gardner. (Typus.) Prope Merrediiq E. H. Wilson et D. A,^ Herbert, Affinity to B. crassifolia var. icosandra, F. Muell, differing in the number of stamens and the iobed staminal disc which bears about 20 setae or subulate processes external to the stamens opposite each petal. In this latter respect it resembles B. Haideni Ewart and White. Astartea clavifolia C, A. Gardn. sp. nov. Frutex metralis altus, glabor^ mulli-ramosus, ramulis erectis, deusis;: foliis ovato-clavatis, leviter compressis, distincte petiolatis, apice basique obtusis; floribus axil]aril>us, podieellatis, calycis tube turbinato, laevo, lobis suborbicularibus; ])etalis quam lobi calycis fere triplo longioribus, orbiculari- bus, roseo-albis; phalangibus distinctis, circiter 8-antheras gerentibus, apicem versus atteiiuatis; filamentis brevibus, tenuibus, plusminusve ineurvis; antberis ovatis, longitudinaliter deliiscentibus; disco lato, eoncavo; ovario 2-loculato, ovulis in locnlis (i, in annulum lateraliter dispositis. Fj-utex jnctralis altus; folia 1.5-2..5 mm. longa, usque 1.^ min. lata; petioli .5 mm. loiigi; pedicelli 2 mm. longi, a medio articulati; bracteoli 1.3 mm. longi; calycis tubus 2.5 mm. longuSj lobi 1 mm. longi; petala 2.5 mm. diam. Ilab. in distr. Avon prope ]\laya, in lutosis areiiosis, flor. m. Septem.- Gardner 2704. Affinity to A. chmdata Turez. and A. fascicnlaris (Labill.) D. C.,- differing in tin* 2-celled ovary, the much shorter and broader clavate leaves^ which are n(*ver clustered, as well as in other minor features. CALYTHROPSIS C. A. Gardn. gen. nov. Myi ’tacece-Chamaelaiiciece. Calycis tubus obovoideus, 4-gonus, ovario adnatiis, segmonta vel sepala nulla; petala 4, integi*a, patentia, decidua. Stamina indefinita, numerosa, pluriseriata, libera; filamenta glabra, filiformes; antherae versatiles, loculis parallelis, longitudinaliter debiscentibns, connective glandula parva, globosa^ appendiculato. Ovarium infevum, calycis tubo subaequale, 1-loculare; ovula COoN’TKlBUTIONES FLORAE AuSTKALIAE OcC iDENTALlS XI. 189 erecta, ijiacenta filitorini e fundo ovarii oriunda et usque ad apieem extensa, collateraliter afiixa; stylus filiformis, glaber, stigmate i^arvo, capitato. Fructus siccus, iiidehiscens, calycis tubo adnatusj semen solitariuin. Erutex cricoideus. Folia sessilia, s^^arsa, plusminusve densa, parva, integerrinui, marginibus liyalinis. Stipulae nullae. Flores sessiles, in eapitula terminalia conferti singuli in axillam bracteae vel folii floralis. Bracteolae 2, subfoliaceae vel ad margines scariosae, persistentes, basi eonnatae, rigidulae, parte libera imbricatae tubumque calycis iucludentes. S 2 :>ecies 1. Australiae occidentalis incola. The ailinity is with CaJythrix and Lliotzlcya "vvith similar ovules and jDlacentation, while the plant has much the aspect of Calythrix aurea Lindl. It dihers from Calythrix and Lhotzkya however, in the complete absence of any sepals or calyx-lobes, in the number of petals, and in the sti'ongly 4-ribbed calyx-tube which is quite sessile and constricted at the a'pGX, The calyx tube is almost that of ActinodiuMy while the bracteoles are those of some species of Calythrix. C. aurea C. A. Gardn. sp. nov. Fruticulus metralis altus, glaber; ramis erectis, ramulis inf erne crebo cieatriciferis, superne dense foliosis; foliis erectis, oblongo-lanceolatis, breviter petiolatis, apice aeutis A^el mueronatis, su])ra concavis, subtus carinatis, Tuiinervis, glandulis porpaucis inspersis, marginibus ciliolatis ; floribus sessilibuH in capitiilis ad apieem ramulorum, foliis doralibus vel braetcis foliis subsimiiibus sed brevioribiis; braeteolis ad medium usque connatis, ealyceni excedcntibus, obtusissimis vel breviter muci'onatis, marginibus liyalinis; ca]3’cis tubo obovoideo, fortissimo 4-gono, suj^erne constrieto, glabro; petalis ilavis, oblongo-ovatis, imbricatis, qnam calycis tubus inulto longioribus; staminibus numerosis, (? fuscis) ; stylo quani petala stamincs broviore, ai3ice minutissime capitato, glabro. Folia 0-7 nnn. Joiiga; cniyeis tubus uini. longus; potnla 7-s nun. longa; fructus imjierfectns quani braeteoli breviori. Hal), ill distr. Irwin, ad ripas fluvii Arrowsmith, tlor. m. Septem. ir. E. Blachall 4449 (Typus in Herb. Perth). Although ripe fruits have not been observed, the ovaiy after flowering becomes cylindrically elongated with a short constricted portion below the summit, but retains the four strong ribs of the llowering calyx-tube. I have described the ovules as being 2 in number, and they are the same as those which I have seen in specimens of Calythrix and Lhotzhya as far as I have examined sj;)eeimens of those genera. These ovules, which are erect near the middle of a filiform placenta, appear to be almost or quite separate when viewed from one side, but from the 02 :» 2 iosite side they appear to be fused to about the middle. It is iiossible that the^^ represent a single lobed ovule. Darwinia Meeboldii C. A. Gardn. sp. nov. Frutex rigidus, erectus; 2-3 metralis altus, glaber; cortice allio-suberosa ; foliis densis, decussatis, erectis vel subpatentibus, lineari-oblongis, sessilibus, basi breviter contractis, apice obtusis, non evidenter nervosis, planis vel marginibus Icviter recurvis; cai^itulis terminalibus, magnis, nutantibiiis; bracteis involuerantibns pluri-seriatis, ovatis vel oblongo-lanceolatis vel lanceolatis, aeutis vel acuminatis, exterioribus bi’evioribus, viridibus, 190 Chakles Austin Gakdneb. intoi'ioribus praesortim siirsuin i)urpiu'eis^ oiunibus iiitegris; braetoolis liiicaribuSj tcnuibus, integrisj juaitisj vhIcIg concavis vcl cymbiformibuSj quam calyx vix loiigioribus ; calycis tubo cylindrico, duro, crustacoo, laevo et iiitenta, niinnte foveolatoj sursiuii ])apilloso, lobis erccti^, duris, erassis^ a])ieein acuti^j sursuin atro-violaccis j potalis quain ealycis lobi vix loiigioribuSj triaiig'ulavi-lancoolatisj ac'utisj iiitegrisj stamiiiibus brevibusj stanunodiis siibulatiSj (jiuuu filaniaiita bvcvioribusj stylo eloiig'atoj per antbesiii rLiguloso- coiitraetOj sub a2)ir:e barbato, deiuimi elongate. Fnitex 3 m. attiagens; folia usque 9 iniu. lojiga, 1.5-2 mm. lata; capitula 3 cm. longa; bracteac intcriores 3 cm. longae; flos 11 mm. lougus. Ilab. in dislr. Rtirling pro]>e Craubrook. A. 3Ieebold; a pede Peak Donnelly in elivis lapidosis, 11. ni. Noveni., W. Forteoiis et C. A. Gardner. A [Unity to D. leiostyla (Turez.) Domin, differing in the larger size and more robust habit, the broader and longer leaves, the much larger and narrower amunuiate bi'acts of the involucre and the larger floAvers. The two species look entirely different in the field. Verticordia Etheliana C. A. Gardn. sp. iiov. Section Catocohjptra. Fruticulus semimetralis altus, rainis patentibus; folds breviter potiolatis, erecto-patentibus, orbiculari-obovatis, obtusis, subglaucis, imbricatis, concavis;. floribus axilla ribus, paucis, inagnis, purpintuseentibus, braeteolis amplis^ hyalinis, subpersistentibus ; calycis tubo turbinatoj 5-costato, segmentis primariis in lobos penniforines ll-incisis, lobis accessoriis hyalinis, refiexis sed non a bn si ereetis, orbicularibus, ciliatis, appendicibus herbaceis inter tubi costas conspicuiKS, aciitis, subliberis, lineari-triangularis; petalis tubo stamineo adnatis, oblongo-cordatis, ad apicein versus longe ciliatis; staminibus cum staminodiis basi in tnbnin stamineum coalitis, antheris ]iai'vis, oblongis; staminodiis subulatis, stylo sub apice barbato. Folia 4-5 mm. loiiga; pedicelli usque 8 mm. loiigi; tubus calcyiuus 5 mm., loiigus; segmonta primaria 7 turn, loiiga, aecessoria 5 mm. louga ; pctala 7 mm. longa; stamiiiodia stamimupie 3 mm. louga. Hab. in dislr. Irwin, inter Yaringa et Northampton, fl. m. Septem.- TF. E. Bhudall 4724. Novaui haiU' nominando pulcherimamque speciem Ethel commemoro, quae, et ojiennn et explorationum socia devotissima, una cum dnetore clarissimo (lulielnio Edwardo Blackall, viro suo, specimenibus turn colligendis turn ordinandis enixe laboravit. Affinity to V. chriisostachifa Meissn., from which it differs in the much larger crimson or scarltd flowers, the longer pedicels and larger hei’bactmus appendage's, as well as in the much shorter inflorescence. Neither in this, nor in any other species of the Sect. Caiocahjptra investigated are the accessory' calyx-segments reflexed u])\vards on the calyx-tube. VERBENACEAE. Pityrodia maeulata (\ A. Gardn. sp. nov. Section Eupiiyrodia. Suffrutex 50 cm. altns; ramis ereclis, strictis, teretibus, pauce glandidoso- pilosis ; foliis distantibus, scssilibus, basi anriculato-dilatatis, ellipticis, tenuibus, .dentibus iitroque eirciter 10, aeutis, venis pinnatis, grosse anastomosis, ]>auce glanduloso-pilosis vel subglabris; floribus in axillis CONTKIBXJTIONES FLORAE AuSTKALIAE OCCIDENTALIS XI. 191 foliorum sni^cj’ionim, in raeemis 2-I)-0oribus dispositis, bracteolis hezdnieeis, ovatis, setaeeo-donticulatis, persisteiitil)ns, fere glabris; pedicellis gracilibus, apicem versus stelJato toinontosis ; ealycis segmentis subliberis, oblongo- lanceolatiSj aoulis^ extus laxo stellato-hirsutis, intus glabris; corolla lilacina, e basi sensiin ampliata, limbi lobis 2 superioribus altius connatis, 3 inferioribus ovato-orbicu]aril)us j staininibus inclusis, basin versus tubo insertis, antheris omnibus l)i’eviter appendieulatis ; ovario apice tomentoao, stylo g’labro, bifido. Folia 2 cm. loiiga ot 8 inni. lata; vacenii '2.~i <'m. Joiigij calyx 6 nnii. loiigus; corolla 1.3 cm. loiiga. Hab. in distr. Avon^ ]>rope r3allidu, in arenosis apertis, flor. m. Septem. Gardner 2708. Affinity to P, dilaiata (F. Muell.) Benth., differing’ in the ovate dentate thin subglabrous leaves which are not bullate, the abbreviated axillary racemes, ovate bvacteoles, oblong-lanceolate calyx-segments, and in the indumentum. The corolla is conspicuously brown spotted in the throat. The flowers have been damaged by insects due to early neglect, and thus the .shape of the corolla lobes has been described from undeveloped flowers. Dicrastyles fiexiiosa i Price) C. A, Gardn. comb. nov. xV. specimen collected by \V. F. Bla('kall between Menxies and Broad Arrow is id<‘ntical Avith luce’s plant -delloideis, subacutis corolhe tubo calycem aux excedenb*, (‘xtus albo tomentoso, i.utiis glabri, baso excepto, sub insertione filamentorum barbato, promijumter atro-violaceo striato, lobis intus albo-tomentosis ; stamini})us o, omnil)us i‘ertilibus, in tubum corollae insertis, tilamentis br(*vil)us, ad basin baiBatis; disco glabro, crasso; ovario ovoideo, glnbi-t), loculis 2, 4-5~o\'ulatis : stylo (dongato, glabro. Frutex usque 3 n\. altus; folia nsipie cm. loiiga, msijuv 1.8 cm. lata; calyx 6 mjii, loiigus; lobi 3 mm. loiigi; coi’olla 8 mm. loiiga ot lata. Hal), in distr. Ti’Avin prope Indarra, in arenosis a})ertis, II. m. Se])t. Gardner 2tib() (Typns) ; atipic septentrionalcm versus tinmen Murchison Elver transgreditiir, ubi in arenosis apertis adest tlov. m. October. Affinity to A. Blackii, differing in the close hoary tomentum of the branchlets and leaves, the shape and size of the leaves, the smaller Avhitish (not dee]) violet) OoAvers with acute corolla lobes &c. Tt lias the indumentum of A. Walcottii, from Avhich it differs in stature, the subsessile floAvers, and from both snecies in the distinctly tubular corolla, the shape of Avliich is that rif Myoi^ormn. 192 Charles Austin Gardner. PEDALIACE.^. JosepLinia Eugeniae F. MuelL This species is recorded from Tambrey Station in the vicinity of the Hamersley Range, where it is reported as spreading and becoming a trouble- some weed. ACANTHACE^. Dicladanthera Forrestii F. Muell, An undershrub almost a metre high, with pale pink or white flowers.. It grows in stony watercourses along the Upper Hardie Eiver near Mount. Samson; fl. m. August. Gardner 3168. Ruellia primulacea F. Muell. A very attractive species which is either a small shrub, or semi-seandent and attaining a height of several feet; the flowers are a rich violet in colour. Barrabiddy Creek, Minilya River, Gardner 3020; 90 kilometres northi of Carnarvon, (7. li. Meadly, This is the flrst record of this species from Western Australia. MYOPORACE.Tl. Eremophila lachnocalyx C. A. Gardn. sp. nov. Section Eriocah/x. Frutex eroctus 2 m. altus, ramis ramulisque dense albo-tomentosis,- deninm glabris, relitjuis foliorum evanidorum induiis, sursum dense foliatis; foliis parvis, imbricatis, patentibus vel reflexis, orbieulari-obovatis vel ellipticis, basin versus contractis, late sessilibus, obtusis, utrinque eano- tomento.sis; (loribus coiispicuis, axillaribus; pedunculis solitariis, tenuibus, quam folia paulo longioribus, dense floecoso-tomentosis; ealycis sogmentis aeqiialibus, linearibus, aeutis vel acuminatis, intus glabris, extus dense floccoso-tomeiitosis; corolla basi vix constricta, campanulata, calycem duplo superante, intmise violacea, lobis subaequalibus, ovatis, subacutis, 2 superioribus alte connatis, mediano (|uam cetet’i paulum longiore; staminibus inclnsis ; ovario lanoso, stylo plusminusve glabro. Folia 4-5 mm. longa^ 3-3.5 mm. lata; podiinculi 7-8 mm. longi; sepala 1 cm. longa, 1 mm. lata; corolla vix 2 cm. longa. Hal), in distr. Austin prope Paynesville, in la]fldosis lutosis, fl. m. August, Gardner 2515; ])r. Cue, in rupesfribus lutosis, Gardner July, 1931; Xannine, lutoso-nitrosis, Gardner ^ C. A^idretvs; Mount Gascoigne, E. S.. Slinipson Septem. 1927. Aflhiity to E. Mackinlayi F. !Mucl]., and E. sfrongylopJiylla F. Muell., differing in the smaller leaves, the corolla not constricted above the ovary but gradually Avidened into a campanulate base, the longer acute corolla lobes, much narroAA'or and more deeply divided calyx' covered Avith a long floccose wool. Tliis AAdiite aa'ooI is sometimes suffused with violet, and alwavs forms a subglobular Avoolly mass Avhich quite conceals the calyx. The specimen from Mount Gascoigne has more acute and erect leaA^es, but is otherAvise typical: it may perhaps constitute a distinct variety. CONTKIBUTIONES FlOKAJE AuSTRALIAE OcCIDENTALIS XI. 19 :^ Eremophila inflata G. A. Gardn. sp. nov. Section Eremocosmos. Frutex gracilis, erectis, glaber (floribus exeeptis), ramis ramulisque virgatis, ramulis pauce resinoso-tubereulatis ; foliis lineari-oblaneeolatis, obtusis, nonmuiquam breviter uneinatis, basin versus attenuatis vel breviter petiolatis, siecanclo nigrcseentibus, costa centralc solum eonspieua; floribus axillaribus, solitariis, graciliter pedunculatis; pedunculis glabris, patentibus vel recurvis, sub calyee clav^atim incrassatis; calycis seginentis oblongo- elliptieis, solutis, obtusis, tenuibus, valde reticulato-nervosis, marginibus et nervis ciliolatis, post anthesin deflexis; corolla pallide violacea, glabra, tubo campanulato, basi inflato, sursum contracto, plusminusve unceolato, extus parce piloso, intus glabro, lobis brevioribus, obtusis, subaequalibus ; staminibus inclusis; ovario longe hirsute, stylo glabro, uncinato. Frutex 2 in. altus; folia usque 3 cm. longa, 3.5 nmi. lata; sepala 3.5 nim. longa, 1.6 mm. lata; corolla 7.5 mm. longa. Hab. in distr. Coolgardie prope Mount Holland, juxta Lake Cronin, leg. H. Steedman, Decern. 1929. A very distinct species with afTinity to E. Stiirtii R. Br., but with several differences. The reflexed calyx-segments and the much dilated base of the corolla, as well as the very shoi-t and obtuse corolla-lobes are features peculiar to the species within the ^Eremocosmos. Eremophila Macmillaniana C. A. Gardn. sp. nov. Section Platychilus Frutex robustus, 2 metralis altus, rigidus, erectus; ramulis erecto- patentibus, reliquis foliorum delapsorum persistentibus indutis, cano- tomentosis, sursum foliatis; foliis congestis, patentibus, lato-spathulatis vel spathulato-obovatis, basin versus attenuatis, breviter petiolatis, junioribus albo-tomentosis demum glabris et argenteo-nitentibus, apice abrupte acutis et recurvis nonnunquam eomplicalis; floribus axillaribus vel terminalibus,. solitariis; pedunculis erectis vel leviter ]mtentibus, sursum comjflanato- clavatis, albo-tomentosis, angiilaribus ; calycis segmentis magnis, oblanceolato- obovatis, inaequalibiis, fere liberis, acutis vel mueronatis, extus sparse tomentosis, demum glabris, ])rominenter reticulato-nervosis, post anthesin leviter am]flifieatis, ad basin vix imbricatis; corolla atro-purpurea, extus fere glabra, tul )0 supra ovarium constricto, deinde sensim ampliato, plusminusve recto, intus prope basin dense lanoso, lobis subacqualil)us, 2 superioribus angusto-ovatis, acutis, alte eonnatis, lobis inferioribus ovatis, obtusis vel mueronatis; staminibus breviter exsertis; ovario longe cylindrico, minute aureo-granuloso, stylo elongato, glabro, Frutex 2 metralis altus; folia 3 cm. longa, 1 cm. lata; calycis segmentis usque 2.6 cm. longi, 8 mm. lati; corolla usque 3.3 cm. longa, lobis 8 mm. longis. Hab. in distr. Austin, ab lato orientale montium Barloweerie, in clivis lapidosis sterilibus, fl. m. Jul-Aug. Gardner 2533; atquo ad vicum Cue, in eollibus lapidosis. Quam speeiem honoris causa pro illustrissimo Roberto Furze Macmillan Equite, nomino; qui vero, multos per annos Judicatus in Australia oecidontali primatum tonens, hand minore ingenio quam url)anitate artium simul ae seientiarum hac in republica fautor cxcellebat. 194 Charles Austin Gardner. Affinity to E, Evaseri F. Muell., and E. spathulata W. V. Fitzg., differing from l)oth in the shape of the calyx-segments and in the corolla. The calyx is almost that of the sjiecies of § EveTHocosuios. There is also some resem- blance to E. Ereclhign F. Muell., but the leaves are broader, the calyx-segments lai’ger and of different shape, and the stamens exserted. Eremophila spectabilis (7. A, Gardn. sp. nov. Platychilus) hrutex 1-2 metralis altus; ramis patentibus, inferioribus pendulis, ramulis foliaceis, resinosis et breviter pilosis: foliis linearibus vel lineari- la.nc(‘olatis, remote dentieulatis, acuminatis, basin versus attenuatis vix petiolajis, viridibus et glaucis, breviter ])ilosis, supra subtiliter canaliculatis ; l)edunciilis axillaribus, solitariis, elongatis, sigmoideo-pateiitibus, apiceni \ersu8 augulatis et valde iiKU’a-ssatis ; calycis segmentis valde imbricatis, inaepuidibus, segnuuito exteriore unilateraliter decurrenti^, ovato-aeuminatis, celefis angiisto-lanceolatis, oniniljus minute hispidulis et viscidis; corolla inl(ms(‘ violacea, subglabi’a, tubo cylindrico, supra ovarium leviter constricto, deindi' ampliato, lobis o\'atis vel obovatis, acutis, 2 superioribus alte connatis, melant is closely related to E. Ereelingii, from which it diOers in tin* sirssile denticulate leaves, much longer sigmoid jieduncles and acute corolla lobes, as well as in the calyx. It also has some points of resem- blance with h, (hlvs'n, differing in the distinctly and somewhat coarsely den- ticulate leaves, shorter and thicker peduncle, imbricate calyx-segments, acute corolla lobes (4c. E. 1' reeJingii, E. ^})ectabilE, E. Gilesii and FL folio^F^ima have all the same characteristic indumentum of short stiff curled white hairs accompanied by some viscidity, but the markc'd decurrence of the right hand niargin of tin' outer sepals of F]. specta})}lis is a feature not olu:('rved in the other s]>ecies. Eremophila miniala (\ A. ikmlu. sp. iiov. Idalychilus.) Ai'bor ])ar\'a vel frutex elatus, corti'X iiigro-suberosa ; ramis rigidis, 3'ectaiigulariter di\ai*icatis, crassis; foliis apice ranudorum versus coiigcstis, lato-linearibus v(4 liiieari-oblougis, sessililjus, subpiaiiis, pubescentibus, apice acutis vel acumiiiatis, costa ('(uitrale solum consjucue; peduneulis solitariis, sigmoi(h'(>-patenti])us, pubesc(‘ntil)us ; calycis segmentis \‘aide ini])ricatis, liheris, ituuHiualibus, lato-ovatis vel suhrotundis u.^tjue elliplicis, mucronatls, TOsfMs, i'eti('ulatis, post anthesin ampliti('atis et pi'ominenter venosis, puberulis ^'el glabris; corolla miniata. sparse et bn'viter pilosa, intus glabra, fauce bruniKM) macuiala; tubo basi cylindih'o, sui)ra ovarium sensim ampliato, lobis 2 sup{‘i*ioribus obovatis, acutis, alte connatis, inferioribus ovatis, acutis: staininibus exs(u*ti,s; ovario styloque glabro. I’laiita usque 5 m. alta; folia 2-2.5 cm. longa, 2-4 mm. lata; pednnculi 1.5 cm. longi; Hei)ala 1.5 cm. longa et 1.3 cm. lata, usque 2 cm. longa et 8 mm. lata: corolla 3.3 cm. longa; limbi lobi 1.2 cm. longi. CONTRIBUTIONES FLORAE AxtSTRALIAE OcCIDENTALIS XI. 195- Hab. in distr. Avon^ in frutieetis apertis lutoso-nitrosis prope. Co-vv- cowing', flor. m. Septein. Gardner 2740 (Typiis); in distr. Coolgardie^ ad Penny^s Find prope Kurnalpi, J. H. Franks fnict. m. Decern., prope Broad Arrow, IF. E. Blackall 2144, iior. m. Septeui. Apart from the shape of the corolla and its lobes, this species might,, from the description, be mistaken for E, Duttoni F. Muell. It belongs, however, to the Section Platychilus, and should be placed next to E. viscida EndL, from which it differs in several features, such as the shape and size of the leaves which are not viscid, the size and lobing of the corolla, etc. E. viscida EndL, previously known only from the single specimen in the- Vienna Plerbarium, has been collected by the author from Ballidu in granitic soil, and by F. E. Victor from Kununoppin. It is a handsome shrub of 3 to 4 metres in height, with white flowers, the throat spotted with violet. The calyx segments enlarge after flowering, but become membranous, and are elliptical to almost obovate in shape. Eremophila virens G. A. Gardn. sp. nov. Section Stenochilus. Frutex erectus, virgatus, 3-5 metralis altus; ramulis erectis, viscidis; folds alternis, patentibus, petiolatis, planis vel complicatis, oblongo- laneeolatis, laneeolatis vel ovato-lanceolatis, longe petiolatis, viscidis, vireseen- tibus, glabris, basi inaequaliter abrupte coiitractis, apice aeutis vel acuminatis; floribus viridibus; pedunculis solitariis vel binis, axillaribus, elongatis, tenuibus, sigmoideo-incurvatis, quam calyx multo longioribus; calyeis segmentis liberis, imbricatis, exterioribus ovatis, interioiibus oblongo-obovatis vel oblanceolatis, intus lanosis, omnibus post anthesin amplifleatis ; corolla viridi, extus albo-tomentosa, sed ]mrfeeta non visu, tubo suj^ra ovarium con- tracto sursum leviter ampliato, sub limbo attenuato, limbi lobis aeutis; staminibus exsertis; ovario subgloboso, glabro, stylo glabro. Frutex 3-5 m. altus; folia 7,2 cm. longa vel minus, usque -.6 cm. lata, petiolus 1.5 cm. longus; pcdimeuli usque 2 cm. longi, graciles; calyx (> mm. longus, post anthesin 1.3 cm. longus; corolla 1.3 cm. longa. Hab. in distr. Coolgardie prope Campion, in frutieetis gi'anitieis, deflor. m. Septem. Gardner 2765. Affinity to E. serridata (A. Cum.) Druce, but with much larger and ([uite entire differently shaped leaves, the corolla woolly-tomentose outside. The habit of the two is also quite different. When collected, the specimens were past flowering, and but few somewhat damaged flowers Avere obtained, Avhile one or tAvo Avere immature. The structure of the corolla is hoAvever, entirely that of § Stenocliihis, Avith four small deltoid teeth forming the upper lij), and a much more deeply separated loAA'Or lip formed from a single ovate-oblong lobe. LABIAT.35. Hemigenia divaricata C. A. Gardn. sp. nov. Section Biplanthera. Frutex divaricato-ramosus, ramulis oi^iAositis, rectaugularitcr divaricatis, teretibus; foliis oppositis, iu ramulis juiiioribus saepissime fasciculato- oppositis, linearibus vel lineari-spathnlatis, sessilibus, concavis, vix aeutis, glabris, apice saepissime recurvatis; floi-ibus axillaribus, solitariis, brevi- pedunculatis a^'cI subsessilibus, supra medium bibracteolatis : bracteolis parvi.s. 190 Charles Austin Gardner. lineari-laiiceolatis, acutis, eiliatis, quam calyx multo brevioribus, deciduis; calycis tubo caiapaimlato, laxe hirsutOj lobis subaequalibus, ovato-lanceolatis, acuniinatis, eoiispicuc costatis, tubo subaequantibus ; corolla violaeea, extus laxe liirsuta, intus glabra, tubo e basi cylindrico sub limbo ampliato, limbi lobis emarginatis vel bilobis, labii superioris lobis 2 ovatis, labii inferioris lobis 3, quam superius longioi'ibus; aiitberarum superiorum loculo altero sterili in appendiculam glabram produeto, antheraruni inferiorum altero globoso sterilesc elite. Priitex 1-1.5 111 . altus; folia 4-0 mm. longa.; pedimculi cire. 1 mm. loiigi; bracteolae 2.5 mm. loiigaej calycis tubus 4 mm. longus, lobi usque 0 nun. loiigi. Hab. in distr. Austin prope Pindar occidentalem versus in locis graniticis apertis, tlor. m. Septem, Gardner 2024. This species is most closely related to H. diplcmtliera F. MuelL, differ- ing principally in tlie broader hirsute eeostate calyx-tube and the hairy corolla. Hemigenia coccinea C. A. Gardn. sp. nov. Section EiiheTuigenia. 4rutex divaricaliis, us(jue 20 cm. altus, pilis appressis vestitus; ramulis intricato-divaricatis, teretibus; foliis oppositis, erectis vel leviter patentibus, linoari-oblongis, valde concavis, sessilibus, obtusis, apice leviter recurvatis, supra glabris, subtus approsso-pilosis; (Joribus uxillaribus, longe pedunculatis ; peduncnlis medio bibracteolatis ; bracteolis lineari-setaceis, deciduis, calycis tubo cylindrico-canqianulato, jiarce piloso, 1 0-costato, lol)is ti'iangulari- lanceolatia, acuniinatis, subaequalibus, quam tubus longioribus ; corolla coccinea, extus fere glabra, intus glabra, basin excepta, tubo e basi angusto- cylindrico, sensim ampliato; labii superioris lobis 2 ovato, oblongis, truncatis, supra medium connatis, labii inferioris lobis 3, quam superius longioribus, lobo mediano emarginato, latei’alibus oblongis brevioribus, omnibus serrato- dentatis; anthorarum loculo altero sterili, in appendiculam glabram iiroducto. 1‘1’utex 20 cm. nltiis; folia 7-S mm. longa: pedumnili (i-7 mm. loiigi; tubus calycis 2. .5 mm. loiigus, lobi 3 mm. loiigi; corolla l.S cm. longa. Hab. in distr. Irwin ad fines orientalcm versus, prope Canna, in lutosis graniticis, tlor m. Septem. Gardner 2661. Amongst the sjiecies of the Sect. Eul\emigenia this sj'ocies has its closest affinity to 77. linmiUs Benth., differing in the narrower deeply concave leaves, the longer peduncles and narrower hracteoles, the much longer corolla, and in liaving all the anther appendages glabrous. It is a low intricately divari- cate shrub with the aspect of 77. divarieata C. A. Gardn. but has scarlet bowers Avith a much longer corolla-tube. The ditt'erencos between the Sections Diplanthera and Euheyyugenia are not. entirol.v satisfactory. Prostanthera magnifica C, A. Gardn. sp. nov. hrulex 1-2 nieti’alis altus, fere glaber, ramis ramulisque divaricato- patt'utibus, glabris vel breviter crispo-liirsutis; foliis ellipticis vel obovatis, crassis, integris vel lunice donticulntis, basin versus attenuatis, in petiolo hrevi; iloribus magnis, in axillis supremia spiels foliatis formantibus; peduncnlis erectis, sub calyco bibracteolatis; bracteolis lineari-lanceolatis, demum deciduis; ealyce roseo, tubo glabro, brevi, striato, labio supero integro, magno, rotundo-ovato, quam corolla vix hreviorc, elegantcr nervoso, obtuso, labio infero ovato vel oblongo-ovato, quam superius du]'>]o hreviore; eorolhi ])urpurea vel coccinea, extus breviter albo-pilosa, intus farinoso- CONTRIBUTIONES FlOHAE AuSTRALIAE OcCIDENTALI« XI 197 pui)escoiit(s sRbiUHjiuilitor o-lobata, lobis (loiisa (-iliatis; lal)i() ' .su))(‘r() late rotundato, emar^inato, ^■(*l lobis 2 brindbus auetis; labii iiifei’i ]oi)is iatera]i])u>i ovatis, subaeutis basi subeordutis, lobo mediano (*inai‘0'iiiato-l)ilobo ; antlu‘raruni calcare lon^itudiiieui loeidi duplo exeedeiit(‘ ; stylo glabro. Friitex 1-2 in. altus; folia usque 2.-') cm. i()uj;a, 1 cm. lata; pedum-uli 4-7 mm. lon^'i; liracteoli 4-7 mm. lon^i; calyx usijue 2.4 cm. lonf>us, labium siipeinm 1.7 cm. latum, infenim 1 (un. loiiguni, (i mm. latum; corolla 2 cm. longa, limbi lobi circiter o.4 mm. longi, Ilab. in liistr. Irwin, ad vicnni Mnllewa MO kni. occidentabun xt'rsus, tlor. rn. Sejitein. II . E. Hlachifll 2783: ("ry])iis) ; atipie -15 km. ab orientabmi versus Ajana, Mrs. ]). L. ('arson, Oct. 1940. var. asperata (\ A. (iardn. var. iiov. loliis miii(>]‘ibiis, asianatis, mimit(‘ a.s})erat()-d(*ntiiciilal is ; lloribus coccineis, ininoribiis. Ad sumnum iiiontem ('hnrchinan in rupestiubus graniticis, llor. in. Septem. 1!'. E. Bhu‘h-all 3458. This speides is Aery distinct. It belong's to the Section E npostanfhera , but has no cleai' ahinities. The rich [unq)le or scarlet coloui' of (he calyx, and the scarlet or puiiih' ('orolla, the size of tin* (lowers, small statun* and floidfei’ons habit render it a most attractive spcmies for horticultural dc'velop- numt. OOODENIAC'H.T]. Gocdenia Hilliana ('. .1. (iardn. sp. nov. ( Phigoodenia-Hracleolatae-Racemosai*. ) llerba |ierennis vel suffrutex, erecta \-el ramosa, 45 cm. alta; ramis teretibus, crassis, dens(‘ et lirevitiu* g'landiiloso-pubescfud ibus ; foliis herbacOs, erectis, planis, ohlongo-obovatis, glanduioso-iiilosis, basin versus atlenuatis, vix ])etiolatis, a pice ohtusis, marglne dentatis; raceinis densillovis, erectis, terminalibus, foliis multo excfulentibus ; pedicellis basi bractiudis, sub calyci' bracteolatis ; bracteis loliaceis, erectis, oblong'o-obo\’a1 is, inb'gris, glanduloso- pubescentibns, (|uam jU'dicelli longioribus, ajiice obtusis, basin versus attenuatis : bracteolis similibus bractius sed minoribus angusl iorihuspue, g'landu]oso^]>ubescentibus ; corolla caerideo-\ iolacea, ext us sparst* hirsuta, intus glabra, loins 3 intmuoribus su])ra nuMluim coniiatis, 2 supei'ioribus protimde divisis, aiiice acutis, omnibus alatis, alls loiiorum ae(|uilatis; staminibus liberis, antheris iiuam lilamenta dupio brevioi'ibus, oblongo- lim^aribus; stylo s[iarse piloso, qiiam stamina subduplo longiore, indusio compresso-cupulato, margine bre\'it(‘r (*t densi* ciliato; o\'ario 2-loculari, dissepimeido incompleto, ovulis 3-5; capsula oblonga, quadrivaivata, semini- bus angusti' marginatis. I lanla 15 cm. alta vel ujiiius; folia usipic 5.5 cm. longa, t>-7 mm. lata; 1 -iceud us(jre 10 cm. lougi ; pcdicelli 3 mm. loiigi; calvx 7 mm. longus, tubo 2 mm. loHgo; segm ntis 5 mm. lougis; corolla 1.4 cm. loiiga, lobi cum alls 4 mm. lati. Ilab. in distr. Ii’win, in collibus glareosis ad liuvium ‘Midi River,” (lor. A fruct. ni. F(‘by. 1941. (rardner. Ilaiu' sp(‘ci(*s viro celeberrimo praeclarissimo(|U(‘ Arthur \V. Hill, (uiuili, Ilorti Botanici Regii Kewensis J)ir(‘ctori nominavi, <]ui peregra\'it Australian! occidentahnn in anno 1927, ipiique s]>ecimena tanta contulit hujiis regionis. j 332/42 198 Charles Austin (Gardner. A very (lislinct and handsome specdes near(‘st to G. quadrilocularis K. Br.j l>ut differing in the whole plant being densely g’landular-pnbeseent throughout, in the long leaty ealyx-lobos and the e<|ually winged segnumts of the zygomorphie corolla. The lateness of its flowering season, together with its floriterous nature make it a speeies worthy of horticidtural iIev(*lop- ment. STVLIDIACE.K Stylidium gaiioides C. A. (kmln. sp. nov. (Tolypangiuni — Vertieillata?) IVrennis, omnino glabra, caulibus teretibusj tlexuosis, diffusis, probabiliter scandeiitibus, paree ramosis; folds in vei'tieillis saepissim<‘ circiter 8-9, inter sese distantibus congestis, oblaneeolatis, a]>icein versus abrupte eontractis et acutis, basin versus sensim attenuatis, ])lanis vel marginibus leviter recurvatis, exterioribus s(|uainis redaetis; raeeinis terminalibus, laxis, paucidoris, longe pedicellatis; bracteis lineari-lanc(‘oiatis, aeutis, basi ultra insertionem in piau cessurn lirevcun }>i*oduetis ; pedic(*llis calyceni semper superantibus, diu)biis braeteolis bracteis similibus sed niinoribus instruetis; calycis tubo ovoideo, lobis oblongis, tubo aeijuilongis, obtusis vel vix acutis, minute ciliatis, fere aequalibus; corolla lutea, tubo breoi, segnumtis late oblongis, integris, basin \ersus margini})us glandnloso-ciliatis, subae(|ua]ibus ; faucis appendiculis 8^ anterioribus 2 late o\'atis, integris, valde conspicuis, ])ost(U’ioribus 6 multo minorilius, subulatis; labello ovato, seiwato-denticulato, ajipendiculis l)asalibus ereetis, oblongo-ovatis, tenuibus, integris. Idanta 30-.5!) cm. alta; folia usque 4 cm. loiiga, 4.5 mm. lata; racemi 7 cm. longL; iKMlicelli iisipie 1.3 cm. longi; calycis tubus 5 mm. lougus, segmcuta 4-5 mm. loiiga, 1,5 mm. lata; curollae tubus vix 3 miu, longus, lobi circiter 7-8 mm. longi, 4 mm. lati. ilab. in distr. Eyre, a sumniam montium East .Mount Barren, inter rupestres, llor. m. Xovein. Gardner 2955. Affinity to S\ verfieillatnni F. Muell., but with larger differently shaped leaves, the inflorescence reduced to a ra(-eme with glabrous pedieles and calyx, and vith more throat-appendages, the eolour of the flowers is also different, Stylidium laciniatum C. A. Gardn, sp. nov. ( Tolypangimn — Jnncea‘) Perennis, innovationes j^rjiauices, adscendentes ant siinplices; foliis omnibus radicalibus, dense I'osulatim eonfertis, parvis, obovatis vel late spathulatis, acutis, suhtus plusminusve carinatis, margiinbus scariosis, sub anthesin evanesceiitihns ; sea pis voinbilibus, glabris, 3 inetralis altis, plus- ininus\'e robustis \‘el flexiiosis, efoliatis, raro ]>artibtis superior] bus bracteis perpaucis lanceolato-liueariljus acutis basi (-alcaratis instruetis; floribus in racemis tenninalibus dispositis, pediceilis caly('cni aequilongis; bracteis siinnl marginibus hyalinis vel subseariosis laneeolatis, basi ultra insertionem in pi’ocessum angustum curvato-patenteni productis; braeteolis bracteis similibus sed niinoribus; tubo calycis parce glanduloso-pubesccmte, oblongo, lobis tnbi (piam ovarium vix longiorihus, valde inaeiiualibus, posteiaoribus anguste lineari-subuiatis, 2 anterioribus longiorihus, aiiguste liiu*ari-lanc(‘olatis, omnibus valde acumiuatis et margine scarioso-limbatis : corolla rosacea, tubo lobis calycis posterioribus subaequilonge, lobis extus parce glanduloso- l)ubescentibus, ovatis vel obovatis, omnibus profunde incisis vel laciniatis, posticis qiiam laterales longiorihus. a sinu lato separatis, dentibus linearibus CONTRIBUTIONES FlOKAE AitsTHALIAE OCCIDENTALIS XI. 199 vel liiieari-laiu'eolatLs ; i'aueis ai)i)eiuli('ulis (luohus minutis, sul)ulali.s, e basi posticum seg'nifiitorum oi’tis; labello ovat(», obtiiso, iinieroiiato, exappen- (liculato; ovaj’io (>l)]()iigo, ovnlis niiiiierosiH a i)lacajita l)asalia lilx'ra. Folia 5-6 t-m. longa; Sfa})iis usque 5 ui. altus; braateae 1 eiu. longae; bracteolae 5 nuu. lougae; podicelli 5 uiiii. Joiigi; calyris tubus (5 inni. longus; soguieutis [K)sti('is 4 iiim.j aiitieis 7 mui. loiigis; ooroHa 1.- oni. loiiga, lobis + 8 111111 . longis. Hab. ill (listr. Warren prope iluviuni Franklaiub in lutosis iialudis huniosis, flor. in. -Jan.-Fel)!*. Edirard (nn'dm'r Jan. 19J() (Typiis); at(]uo pro[)e Doiiinark. Juxta “lleirw Hole,” in loeis siniilibusj C. E, Lane-]^oole n. 324, Febv. 1923. A very distinct species allied to *S'. jiinceiim IF Br., but with a ditl'ereut habit, loose racemose inlioreseence and deeiily dentate corolla lobes, as well as in the sliape of the capsule. 200 C'hakles Austin Gardner. L^LATK I. >StranjiC'a (•yiiaiU'lii(‘ar])a (Meissn.), F. ^Fuell. A — FTal)it, sliowiiif^ tldwei's and fruits. H — l^ractealos. (' — Feriaiitli. D— Periaiitli in section. E — Toi'us and ])istil. F — Section of folIicUs slio^vinj^ ])osition of ovnie, with the inner layer of the eitdocarp ])artially I’einoved, tunl the outer inenibranons layer placed to one side. (1 — Sindion of follicle, showing relative ])OSitioii of inner layers of fruit. H — [^[>])er part of endocai’p, showing attachment of inner inul outer layers, with funicle and l)ase of ovnle. or CoNTRiBUTioNES FLORAE Australiap: Occidentalis XL 201 202 Charles Austin Gardner. A— Habit J’erianth PLATE IT. Strangea stenocarpoidOvS (E. Muell, ex P>entli.), ('. A. Gardn. 4 - , showing flowei's and fruits. IT — Tntioreseenee. C, 1), I], segments. G — I’'lower-l)uds. H — l^istil. I — Endo(.'arp. h thvougli the endoc-av]), showing ovule in position. and F— ; — yeetion CONTKIBI'TIONES .FlORAE Ai’STRALIAE OcCIDENTALIS XI. 203 204 Charles Austin Gardner. IM.ATE TIT. Euphorbia bobphthoiia, C. A. (Jardii. Elant (half nat. size). E — Leaf. (’ — Hraiu-lilet with cyathia. E — luvohu-ral ^land. E— Seed. G— Rase of seed with earuiiele. H ])ersisteat axis of fruit. 1) — (/yathiiun. — Reduiude, a CoNTRiBUTiONES Florae Australiae Occidentalis XL 205 206 Chaklks At’stix Gardner. PI. ATE rv. Eupliorbin cliitioides (Eorst. t.), i\ A. Gardn. A— T> — Leaves. C’ — C-vatluuin. 1) — Seed. CONTHIBT’TIONES FlORAK AusTKALIAE OOCIDENTALIS XI. 207 CONTRIBl'TIONTES FlOHAE AuSTRALtAE OCCIDEXTALIS XT. 209 .Acacia „ aurea n. sp. „ daviesioides n. sp. „ Wiseana n. sp. „ Yorkrakiiieiisis u. sj). ACANTHACPkdC AMARANTACPk^i Anthotroche „ luyoporoides n. sj). Astartca „ clavifolia n. sp. Baeckea „ leptospcnuoidcs n. sj). ... Blackallia n. gen. „ ))il()!)a u. sp. ... „ connata ii. c()!ul». €AP:8Aid*lXIA('E.R (’alythr()[)sis u. gen. „ aurea n. sp. CASUARINAOPk'P; OasTiarina ... „ acutivalvis E. .Alueil. ... „ l)icus[adata Heutii. „ (’. ])iuaster n. sp. „ Pnnse})iana (’. Andr.... ohp:n()i>()1)iace.k (>yj)taiulra „ grandirtora n. sp. ... Darwinia „ Meeboldii u. sp. Dicladanthera „ P'om'stii V. Miiell. Dicrastyles ... „ fiexuosa n. comb. ... Kromo])hila „ intlata n. sp. ,, lachnocalyx n. sp. . „ Macmillaniaiia n. s ,, miniata ri. sp. „ spectabilis n. sp. . „ \ irens n. sp. „ \iscida Pkidl. Eucalyptus „ lirachypliylla ii. sp. . „ Brockwayi n. sp. . „ cii/lhmnda Blakely . „ Eorjuani n. sp. „ megaconuita n. sp. . „ orbifolia V. .Muell. . Pkigenia „ Stukedi u. sp. ... PIUPHORBIAC'E.B Pkiphorbia ... „ bod[)lithona n. sp. . „ chitiokles n. comb. . „ erewophila A. Onnn. . „ Fitilaysoni .1. M. Blae „ Stevenii Bail. INDEX. Rage 173 174 173 173 174 192 173 191 191 188 188 188 188 (lastrolobium Bennettsianum n. sp. glaucum n. sp. ... ,, paucillorum n. sp. ,, ’^eorsifolium (Tom])liolnbiiiin .. obeon latum 'Furez. var. ]>acbypliyl- him GOUDENIAFE.E Goodenia Hilliana n. sp. GR a:\itne.e Gn'villea 183 184 184 177) 188 189 iim 16b 167 candicans n. sp. Dielsiana n. sp. eriobotrya F. .Muell. „ nana ii. s]). ,, obli(juistignia n. sp. „ nifa n. sp. Hemi<£enia ... .. co(‘cinea n. sp. di vai’icata n. sp. 167 1 66 167 * 172 1 83 , 183 I 189 ! 189 192 192 191 J 191 i 192 193 ; 192 193 194 194 195 : 195 184 j 1 86 185 185 186 184 187 184 184 181 181 181 182 182 182 Josephinia. ... Eugeniae F. .Miu'll. Kochia ,, thesioides n. sp. eabiat.r Labiehea eremaai n. sp. teretifoiia n. .sp. .\1L\10SA('E.E Aliri)elia .. deusifiora n. sp. longifoiia n. sp. ,, seorsifolia n. coml». ,, taxifolia n. sp. ... .Muiriantha gen. nov. Hassellii n. comb. MMrrAFE.E .mvgrohace.e G-xylobiiim „ Gard racemosum n. com raimlioxacea: REI)AL1A('1GE Rhebalium ... anil)iguum n. sp. clavatum n. sp. ,, lineare n. sj). Rityrodia jle.vi.Uhsa Price ... maciilata n. sp. Rrostantliera magniHca n. sp. 182 RK()4'KA(‘E-E Rage 179 179 180 179 178 176 176 197 197 166 168 170 169 168 168 169 ! 68 195 196 I 95 192 192 172 172 195 175 175 175 173 177 177 177 178 177 181 181 184 192 178 178 178 176 192 180 180 181 180 190 191 190 196 196 168 210 Charles Austin' (1ari)np]il INDEX Pago HHAMXACE/E 183 Kiiellia 192 „ primulatra F. .MueJL ... 192 RUTACE.^ 180 SOl.AXACE/E 191 Strangca ... ... ... ... 170 „ cvnanchicarpa (.Meissn.) - F. Mucll. 171 „ stenocarpoidcs (F. -Muell.) C. A. Gavdii. 171 oiithiued. Rage- STYLIDIACE.E ... 198 .Stylidiinn ... 198 gali()i. 198 ,, laciniatiim n. .sp. 198 'rrichiniiau ... 173 alhiduin n. sp. ... 173 Ti'iodia ... 100 „ Wisoana n. sp. ... 160 VKHBEXACHAE 190 Vcrticordia 190 ,, EtheJiana 190 Mineralogy of the Donnybrook Sandstones, 211 Western Australia. 15.— MINEEALOG-Y OF THE DONNYBROOK SANDSTONES, WESTERN AUSTRALIA. By JlOROTHv Carroll. Read 13th May, 1941; Published 7th August, 1942. INTRODUCTION. Sandstones, Permian or Triassie in age (Maitland, 1939), outcrop along the w-estern edge of the Darling Plateau in the vicinity of Donnybrook (Lat. 33° 10' S. Long. 116° 10' E.) about 130 miles south of Perth, Western Australia. It has been supposed on lithological grounds that these sandstones, known as the Donnybrook sandstone (Saint-8mith, 1912), are a remnant of the Collie Coal Measui'e Series Avhich are Permian (Maitland, 1939, p. 185), but there is as yet no sut)porting palaeontological evidenc<‘ although farmers in the district have discovered fossil footprints, not yet identified, in the u])per part ot tlie tormation. Beyond some lithological resemblance to the (Mai Measure series at (k>llie and the occurrence of some poorly developed thin beds of coal, there is nothing to show that the Donnybrook sandstone Jormation is of the same age as the Collie beds; however, both ajipear to be of estuaiine or lacustrine oi’igin and to have l)eeu laid down on an uneA'en, eioded surtace of the i ^re-Cambrian complex. There is apparently no con- tinuity betwc'en the two formations. The Donnybrook sandstone formation varies from coarse, unevenly graded and bedd(‘d sandstone with pebble bands near the base, to fine-grained, e\enly bedded, ripple-markcfl sandstones with clay t>artings towards the top. Ihe beds dip at about 4° to the south-west. The thickness is not definitely known, but Imres ha\-e penetrated it to a depth of 200 feet (Maitland n 183). ' ^ Ihe Donnybrook sandstone J.orms fiat-topped hills, covered bv laterised sandy soils, to the M'est, north and south of Donnybrook. Laterih^ (duricrust) above the sandstones is distinguishable from that formed from the Pre- Cambrian rocks by its abundant (]uartz grains. i be disti ibut.ion ol the Donnybrook sandstone, is shown in Figure 1 where the formation is seen to overlap tlie edge of the Pre-Cambrian shield, indicating that the sandstones were in part laid down on an eroded surface of Pre-Cambilan rocks (Forman, 1936, p. 4). The Pre-Cambrian rocks immediately to the north-west of Donnybrook are at a much higher level than the sandstones. A range of sandy hills, known to be underlain in some ])arts by sandstone, occurs to the south-south-west of Figure 1, suggesting a continuation of the formation in this direction. The contact of the sandstone with the Pre-C ambrian just west of Nannup, about 30 miles south of Donny- hrook, is similar to that at Donnybrook. Economically these sandstones are of importance, tor the fimu’, evenly-laalded types jirovide an ensily- worked, durable building stone. In 189/ gold was dlscoATU’ed in the Donnybrook sandstones which were worked until 1906, yielding about 850 ounces of gold from 1,650 tons. The gold is thought to have been derived from small (luaiitities in the surrounding Pre-Cambrian rocks (Maitland, [). 181) and to have been deposited from solution in the sandstones. An unsuccessful attempt was made about ten years ago to recover gold from the Pre-Cambrian rocks near where the old mines are situated in the sandstones. j 327/42. 212 Dorothy Carroll Various rej)ortSj with tin* disti'ihutioii of (lie Ihmnyhrook sand'- Minp:ralogy of the Donnybrook Sandstonp::s, Western Aesthalia. 213 ot* finding workable ('oal seaiUH in the district, and with the oceiirrenee of gold have been summarised ])y (Jibb Maitland (lt)39) but actually there is little detailed knowledge either ol’ the sandstones or oi‘ the I’re.-Cambrian complex, whi(di lies to tile east, north and soulli of tin* sandstones (Figure 1). Tile Fre-( 'ambrian is represimted predominantly by gneisses at Donny- bi’ook, and, in the valley ot' the Freston Fiver, tlu'i’e an* numerous iiegmatitc* dykes some ol' which i-arry lai’gi* crystids of beryl and tourmaline; garneti- f'erous gn(‘iss(*s, associated witii mica schists, an* also common. (tranite occui's fartliei- to tin* east. The soils oNci'lying the l-lonn^du’ook sandstones are gmierally sandy, torming sand-plains. In phu-es, howe\'(*r, the ciay in some parts of the forma- tion is suffii'icnt to give a loamy soil with a ('lay sub-soil evidently well suited to the reiiuinauents of Jarrah [Kn(uihjitl.H>< )nitr(jiKi(ta) which provides timb(‘r for the mills near Donnybrook and Nannu]). Tlu' sand-plain, lying as it does in a region of .'f)-40 inch annual i-aini'all may constitute a, distinct type jn the sand-j>lains of Western Australia, and the investigation given below was begun several years ago when the writer was examining the heavy minerals ot sandplain soils gfuierally. Tin* iH'sults 1‘or sonn* otln*!' types of sand- plain hav(' already Inam publislu'd. MlXFHALOdV OF TIIF DONNYBROOK SANDSTONF8. The specimens (*xamined fall into two groups, actual ro('k specimens, and soils d(‘rived from the W('atlu*ring of tin*, sandstoin‘s. 1die sandstones were examin(*d by thin section as well as by ('oiua'iitrating tin* hea\'y minerals. The sandstoiu's were ci’ushed until a larg(* (|nantity of the material passed an 85 F>.S. sie\’(* (approximately 70 l.M.iM.), the linest }>articles were wasln'd out, tin* sample dried, and about lO to 15 grams were* separated in bromoform. Tin* soils ^yvr^' si(*ved in a similai' way and the ('lay grade* of material wusIkmI out before bromol'oi’m separation. A few ot‘ the sands were treated with IICl io remo\'e feri'uginotis coatings tVom the grains, ])ut most of tln'iu wvw white or almost white in the natural staff; so that acid treatment was untn'cessary. Tin* heavy residues were not Wf'ighed so that Jio percentage ligures art* availabh*, but it was found that the residues Avert* fairly large and would ha\-e, in all })robability, l)(*en at least 1 pf*r cent, of the sand separated. Beforf* momiting, inagn(‘tit(* AA^as r(*moA'ed from each r<‘sidue. In thin sf'ction tin* Dojinybi'ook sandstone is s(*('n to be a medium to tine-grained rot-k ('onsisting of al)oui 75 pt'r ('('iit. (piart/; tin; r(*mainder is felspar, both oHgoclasf* and inicrot-lim*, kaolitiite or other (day mineral sup- plying tin* binding. Tin* (piart/ grain*; an* rounded to suh-angidar in shape and oft('n show re-growth at. tin* edges; such grains W(*r(* also I'ound in some of the soils directly (h'l'ived from the sandstones. The fels[)ars in the sand- stones are someAA'liat kaolinised. Tin* heavy mirnwa] ass('ml)lag(*s contain the following minerals: — mag- netite, ilmenite, leu('ox(*ne, zircon, tourmaline, rutile, splu*ne, epidote, zoisite, garnet, amphibole, chlorite, anatase, nionazite, kyanitc*, sillimanite, staurolite, spinel, and muscoA’it(‘. The suites are rather similar throughout as is shoAvn in Table 1, Avhere the relative abundance is indicated. Notes on Individual Minerals. Magnetite, ilmenite, and leucoxene require no comment except to state that all are rather abundant in the residues, particularly magnetite and ilmenite. Heavy minerals in the Donnyhrooh Sandstones and Derived Soils. 214 Dorothy Carroll. * 1 1 I ' 1 ■ i ■ l--i Cl < < - ! ■ t f ■ • 1 1 ' i - + Cl , ! 1 1 * -1 - + 1 1 znznznzn CO Cl A’ ■ 1 - 1 -b zn Cl Cl mewliat ronnd<'d, colourless grains is one of the ininor constitinmts, tor it occurs as odd grains only in four of the residues. Carnetiferous giieiss('s weiv oA'idtuilIy not present to any appreciable extent in tlu‘ ei’od<‘d tei'raiu, and (lie scant\’ gaiJU'l. in these I'esidues could have been jjassed on from a pre-existing sediment. Kyainte, siMiniaiiUej and sfau/rojite : Tlu‘s(! metamor])hic mi]H*rals oc('ur in the majority of tin- specimens, kyauil(‘ Ixung the. most [)vomiuent. The kyanite grains are usually well-worn, some b(*ing almost I'ouiul, but others are merely binkeii fragments with “sb'ps^’ due to Innaking along cleavages. 8tauroliU‘ occurs in nunn than halt the residues as bright, brownish-yellow, strongly pleochroic grains, most of wliich are either angulai- or sub-angular. The pinsence of kyanil<‘, sillimanite, and staurolite indicates that at some time Pre, -Cambrian meta-sediments contrilmted materia! to the Donnybrook sand- stones, but in vicAV of the scarcity of these minerals, Avith the ('xet'jdion of kyanite, it is unlikely that llu'y wf're derived direiTly, i.e, thes(* minerals iiave come to lh(' Donnyhi'ook sandstones via soim* ('arliei' sediment. Xo sources for thes(' minerals are known in the Donnybrook district, but, as mentioiu'd iire- vioiisly, there is little detailed knowledge of th(‘ Pre-Cambrian complex in the vicinity. Xpmc/: Odd grains of gr<-en spinel w(*r(> found in four of tlie residues. Spinel is commonly present in residues derived from the Ib’o-Chnubrian in the soutli-A\cstei‘u pai‘t of \A estcrn Australia. These s])in{‘] gi'ains are W(!ll- worn, but otiun’wise arc typical. Mnscovifc: With ty])i(‘a] i>Iaty habit is only present in one residiun SIGNIFICANCE OF DETKITAL HEAVY MIXERAI.S DONNYBROOK SANDSTONES. IN THE It lias be(m staled {Saint-Smith, D)]L>, p. ^21) that the Donnybrook sandstone has hvau derived more or less directly, witli little transpoif, from tile weathering ef tin; PrivCanibrian rocks upon whifdi it rests. The liea\'y lesidues show that this is ratlier unlikely, foi' tlie nuiun'ais wliicli are kuoAVu to ha present in the underlying and suri'ounding Prc'-t kniibrhm rocks are pooi-ly re])resented in the r(*sidnes separated from tlu* sandstones. Moreover, a feature wliich was not known when Saiut-Smitli’s statement was made, is that most of the heavy miiU'ral grains lun'(‘ been lonmled by alu'asion during transport. Again, altlumgh there is a considerable numlier of minerals in the residues, xircon is the only one of flu’ non-opa(fue minerals Avhich is pres(‘nt in large quantities, and this suggests that part at least, of the forma- tion Avas (lt'ri\a*d from a pr(*-existiug s(Hlinu'ntary series. The assemblages ari* dominantly granitic as shoAvn by zircon, tourmaline, si)hene, and monazite. IMinei'als derived fi’om greenstones occur very sparingly, as do the metamorjihie minerals (tf Avhich the most conspicuous is kyanite, the most resistant of these to Aveathrring and alirasion. Andahr site, Avhich generally aeeompainh's these motamorphic minerals, is absent. Ihis suggests some transi>ort of the material, from the place of origin, for andahisite does not seem to he able to survive much abrasion. The almost complete absence of garnet and the scarcity of tourmaline in most of the residues is conclusive evidence that the source Avas not in the Donnybrook district, for garnetiferous and tourmaliue-bc'aring rocks are plentiful in the Preston Valley^ but these ro<'ks Aven* (wideiilly not uncovered Avhen the a/md.. Mineralogy of the Donnybrook Sandstones W’^ESTE RN Au STB A L T A . 217 stonc-s \v(’iv boini'' laid down. The scju'city oi‘ ainpliibolc, wliit-h makes up a very minor part of the r(‘sidu(*s, sug-^'('sts deposition some distance from lh(‘ source and also jtossible re-workino* of a seosited in a separate basin, ("Woodward, 1894, p. 548). Examination ol‘ the heavy minerals may be expected to indi- cate. Avhi(']i of these th(H)ries is correct. If the g-raben theory is supportc'd, then the lieavy minei'als may alfoi'd a nu'ans of correlation between the Donnybrook sandstones and the Collie Coal Measures. If the scpai-ate basin theory seems more probable, then it is unlikely tliat the delrital heavy minerals will be helpful in the correlation of the coal measures with any l)art of the Donnybrook sandstones. Saint-Smith (1912 p. 22) has suggested that the I)onny})rook sandstones are st ratigrai)hically above the. Coal Measures; the examination of the detritals in the Donnybrook sandstones suggests the possdiility that part of the ( oal i\Ieasur{' sandstones was eroded to form tlie Donnybrook sandstones. I’ew (Hd crops, Iat<'ritic cappings, and heavy timber make held work dirhcnlt. Also, over a considerable part of the area when* Coal ^Measures or Donnyl)rook sandsttuie occur, more rec(*id lake beds fuitlu'r obscure, slratigraphical details (Woolnougli, 1916, 9. v.) Tlie tact that tlie ]at(*st dissection ot tin* plateau has uncovered and eroded large hills in the Pre-( kimbilan along the Preston Valley, and that these hills are higher than most of the Donnybrook sandstone outcrops, suggests that the lake or estuary in whi('Ii they wer<‘ laid down had an irregular floor caused l)y pre\'ious erosion, and tliat sonu* of these older surfaces are now being once again eroded, possibly because of a general uj)lift of this part of tiio country. Tlie Pre-Cambi'iau rocks on the noilhern side of the Preston Valley may have l)een a land surface which effectiv(*ly blockeetrology initiated by the Commonwealth Pesearclt Giauit to tlie University of V>sterii Australia, and is hei'elyv gratefully acknowh'dged. M} thanks are due to Professor E. de Ck Clarke for reading through this paper and for many helpful suggevstions. KEEEKEN'CIES. Forman, F. G., 1936: Inspection of Gold Prospecting at Donnybrook: Ann. Frog. Report Oeol. Snrrei/ TV. Aust. for 1935, p. 4. Maitland, A. Gihb^ 1939: The Donnyljrook Sandstone Formation and its Associates: Jonr. Royal Society TV. Aust., Vol. XXV., p. 177. 218 Dobothy Carroll. 8aint-8iiiith, E. (^, 1912: A Geological Eceoinuussaiiee of a Portion of the South- West Division of Western Australia: Geol. Survey JV. Aust., Bull. 44. Woodwai-d, H. P., 1894: Xotes on the Geology of Western Australia: Geol. Maa. (Dec. IV.), Vol. I., p. 545. Woolnough, W. G., 1916: Report of the Royal Commission on the Collie Coal Industry, Perth, By Authority, pp. v.-x. The Ooean Hir-ds of q^he Perth I^eachEs. 219 l().-~TIIK OCKAK I^lliDS OF TIIF PFHTII PLEACHES. By L. (jlauekt.* Hoad lOth Juno, 1941; Puhlishod lltli August, 1942. Tlio study ol' tin* (u-oan biivls oast away on out* looal boat-lies may be said lo have oommonood in the wintor of 19.15, whon \V. B. Alexander, M.A., then on the stall' of the. ALusoiuu, eollootod a numboi* of Prions and g Petrel at Pottosloe Beaoh. Two years later, the late F. L. ^slronaeh, a retired civil s(‘rvant living* at North Cotlesloe, htwame interesttal, and during the next lew years was partly restionsible for tlu^ rapid growth of the Miiseunds collection of Procellariiformes, his bag on one occasion consisting of no less lha.n 29 specimens, rein'esenting six species, and ranging from a. yellow-nosed Albatross or Mollymawk lo the white-faced Storm Petrel. On my I’eturn trom active servict* in 1920 I look u]) residence near the coast and IVom then onwards regularly patrolled the ha-al beaches in the winter time, now and again securing siiecimens ttf unnsual interest. To mv lasting regret I failed lo riadise the ini]K)rtance of a tlock of o\'ei* 20 Prions which had come to grief against the wall of a dressing shed at North Pottesloe in 1921. My successes induced J)r. A. L. Servimly to become interested, and he in turn influenced the veteran collector, H. Lawson Whitlock, then living in Perth, to i)artici})ate in the search for “Ocean Derelicts.” Unfortunately iittl(‘ ot the material Ihese expert collectors secured found its way into the Museum, although details inihlished in the “Lniu” iiave made ]>ossil)le the incorporation of tlu*ir ri‘sults in this pa]>er. 1 he material obtained ditlenMi from year to year, no doubt owing to the direction, intensity and duration ol the gales. In this connection it 'may he noted that I^rioiis, which were v(‘ry pleiititul 20 years ago, have been scarce during the last year oi* two, whilst the (kipe Pigeon or Pintado Petrel, liist recordi'd in P120 and snl)se(|uently a gi'eat rarity for several years, was undouhtedly the (ammionesl, victim of re<'ent winter storms, if \vc except the much larger (Jiant Petrel or Nelly, whii-h appears in numhers every year. The area covei'cd by this paper ranges from Safety Bay, soulii of Pt, Peron to North Biaich, a distance of about JO miles. Birds haunting coastal waters or tin* shore are not included. A key lias been appended to assist in the identification of the species dealt with in tliis iiaper. It has a purely local value and so would he of little use on our south (-oast or in the Kastern States where a dilTej*ent avifauna occurs, although some forms are found })oth lu*re and there. Pygoscelis adelige (Hond)r. A Jac(j.). Adelie Ibmguin. 1 he Museum possesses a s]>e(4men, A4819, caught neai* Pity Beach on Kaster Monday, 19J7. As sonic Japanese whal<‘rs had called at Fremantle a short time before it is most proliahlc that I he Iiird was an escaped pet. Eudyptes crestatus moseleyi {Math. A Iredale). Hock Hopper. A specimen captured alive on Hotlnest Island in 1909 is the only local I'ecord, although hardly a year passes without one or more living birds coming ashore somewhere in the lower South-West. *CoinimiincHt<*(l by ])ernnssi hotogTaplied bei'orc it rocoveivd its stronuth and flew oil' to sea liad a wliite ehiii. TJie ^luseuiii idso bits a wliite sjiei'iiuen, A4Sd7 {male), bound at Busselton in dune, lOdT, and Ji breeding' liird with whitish heiul and neck, A507{) (nude), eauyht at Aiistralind by AIi*. K. Miitti'rs in Septembtu', 1939, A description of this bii’d is ^i\'en in an appcmdix. iVnother bii'd, AolSd, with dark (-rown, ^Tey face ;ind white chin and throat was received on 25th S(‘|)tendH‘r, 1949, from Mr. A. X. Xewman of (.’ape L9 (male) Leighton, September 24lh, 19.35, Kevin Jeffrey. Tin' bird’s presiaice is also noted by F. Lawson Whitlock and Dr. I). L. Serventy. The bird is rare in collections. The Blu(' Ih'trel n'sembles the Prions from which it can I>e distinguished by its larger si/e and the white' tips to tin* tail feather. There is also an absence of the pe'ctinations so highly deAX'Ioped in the uppei- mandible of the AAJiale bird. (Temis PACHYPTILA. rin^ Prions havi' long been I'ecognised ns a m(»st puzzling group of birds, superficially, except for the shape ainl striudure of the bill, they are A’cry much alike, forming a series Avhich might logically be included in one species as was sugge.sted by Loomis. IIoAvever, tlie study of liirds from different nesting sites suggests that the matter is not so simph*. 34i(. most recent rcA'icAver, H. A, Falla (li)40) rt'cogui ;es tin' following species and sub- species fivim localitii'S in tlu' Soutlu'rn Indian Oci'an from Avhich tin' birds on our local beaches may be presumed to have trnA’elied although there is- ahvays the possibility that birds from otlu'r nesting sit(‘s are presi'iif as they are knoAvn to range far afield. Vachyptila riitata wacgilU/vraiji (Alathews). St. Paul Island. PacliuptiJa salvini salrini (MathcAvs). Marion Island. Pachyptila salvini rrozrti. (AlathcAvs). Prozet Islands. Pachyptfla {llrterojivion) destflata desolata ((Tinelin). Kerguelen, Macquarie Islands. Antarctica (Cape Denison). Pachyptila {TTeteroprion) helcheri (MatheAvs). Kerguelen, Falkland Islands. Pachyptila {Pseudoprion) iurtnr fallai (Oliver), Southern Indian Ocean, Pachyptila {PsenAoprion) frassirostris eatoni (Mathew's). ? Kerguelen. 222 L. Glauert. One hundred and thirty-Hve of these little birds have reached the Museum Iroin local beaches sine<' 1912 suo-iiesting that along the extensive coast line of South-Western Australia thousands must be cast up every year. The great majority of our material falls within the limits of P.salvwi and broad-billel91 (h)ttesh.e, ()ctol)(*r 8th, 19d0, A5192 Cottesloe. (IVmale). Oedober Htli, 19d0. The islands off' Frenianlle* appear lo he the soiitiieni limit of Hie bird's breeding range on tlie Avest coast of Australia. Pufflnus assimilis glauerti (Math.).^^ Little Shearwater. This little bird every year falls a victim to winter gales. It is knoAvn to breed u|)on the Iloutmaids Abrolhos, and is iirobably the foi-m reeorded as breeding on Rottnest Island by Angus Robinson. Twenty-five local speci- mens htiAe i(‘a(died the Mns(‘iim siiu'e 1912, a au'i'A' liigli figuri* AA'lum it is lecalled that loeal s])eei(‘s rar(‘ly sucemnhefl r specimens from Cottesloe in the Museum are A1767 (female) May 2J)th, 1920, A2588 August 25th, 1923, and A5042 (female) August 21st, 1939; A50()4 (female) was found at North fiemantle, August 28th, 1939, In ai)orted by iiiateiial. the male A2(90 in all its nieasiircnients, e.xeejit tlu' (ail, is close, to the type of talla .s suh-spindes from Koval Sound, Kerguelen, it also has the^“less defined pale edging on the greater wing eoverls." The female A2791 is somewhat larger and has well (h'fined whitish edges to some of the greater wing eoverts as in the (y]iieal form. The above name is therefore given with some hesitatiun. Pelagcdroma marina dulciae (Math.). White-faeed Storm Petrel. Kow and again speciniems are found on local heaehes after stormy weather. The Mii.scuni reeords are: AlfiSl, At (>44, At(i5(t, A29()0 and Aois’fi from Cotte.sloe and A2948 from Fremantle. The liii-d is known to breed on islets off the coast. Phaethon rubricauda westralis (Math.). Bed-tailed Ti-opic Bird. This northern ])ird at times visits the Perth area. On December 30th, 1928, a couple flew so low in a southerly direction at Cottesloe that their characteristic tad feathers could be recognised with the naked (\y(*. Several *PterQ(lrmna, brfivirn.s-fris f Lesson) auct. 224 L. Glauert. others liave been S(;en since that date. Tlie ^luseimi has a specimen A2754 shot at (lin^'in and a second A5092 cau^'ht whilst incubating its eggs on the beach at Busselton. I have also had a repovt tliat a bird had been seen dying low at Albany “live or six velars ago.” Morus serrator serrator (Gray). Australian gannet. A few individuals make their appearance in Gage Road and Cockburn Sound every winter. The local specimens in the collection are A41G8 found on the Fremantle Railway Bridge in June, 1932, and A48ol fi’om North Fremantle, July, 1937. There is also a mounted specimen C461 found at Cottesloe in 1896. Fregata minor minor ((tmelin). Greater Frigate Bird. A specimeiij A1252 caught alive on -Mount's Bay Road on May 4th, 1917, is lh(‘ only evidence' of tlu' presence' of the species as a wanderer near l\'rth. Catharacta skua lonnbergi (Math.). Brown Skua. Two or three birds can usually be seen off Fremantle in the winter months. The bird might be mistaken for a large mutton bird but can always be distinguished by its heavy body and more laboured flight. When closer the white jtatch on the Avings ads as a good identification mark. It is a scavenger and usually hunts its own food though it may liarass other sea birds but not to the same extent as the Arctic or Barasitic Skua. The !^^UKeuln ])ossess(^ a single specimen A2366 secured out at sea. Stercorarius parasiticus (Linne). Arctic Skua. The Ardic or Ibn’asitic Skua wliich breeds in the northern hemisphere is a regidar summer Ausitor to our short's and can often be seen pursuing the twisting and turning Sih'ergull or Tern until the rt'ct'utly caught fish is disgorged. Both light and dark phases are represented every year. A character of the species is the somcAvhat elongated not twisted central tail feathers. F. Lawson Whitlock records that on one oct-asion he found an exhausted l>ird on the beach at Cottesloe, Avhich, however, managed to elude captiu'e. Stercorarius pomarinus (Temminck). Pomarine Skua. The Pomarine Skua is another northern visitor to Gage Roads in sum- mer time. It is smaller than the Brown Skua and somewliat larger than the Arctic Skua from which it can be distinguished at close quarters by the curious twist in the someAvhat elongated central tail feathers. Light and dark phases occur. Observations made from the “Zephyr'^ when travelling to or from Rottnest indicate that it is rarer than the Parasitic Skua. Thk Ocean Birds of the Perth Beaches 225 Sterna dougallii gracilis (Oould). Hoscatc Tern, The Roseate Tern breeds upon llie Houtmaids Abrolhos. North-west gales at times l)ring it soiitli, as is shown by the single specimen, A5029, in the Musuem collection. The bird was found dead at Hcarborougli in Jiilv. 1939. F. Lawson liitlock also records its presence at Cottesloe. Sterna fuscata glauerti (Math.). Sooty Tern. The Sooty Tern breeds on the Tloutman^s Abrolhos. A specimen, A1511, obtained on the Swan River in December, 1917, is the only local material in the Museum. This l)ird arrives at its nesting site in Septembei', leaving again about April. Anous tenuirostris melanops (Oonld), Lesser Noddy. This bird, which nests in great numbers on the Houtmaids Abrolhos, is now and again blown into local waters and cast ashore dead or alive after severe north-westerly gales. Specimens obtained in the vi89, A2696, A2789, A4805, A4996, A500L A5065, and A5()70. F. Lawson Whitlock also records the species from Cottesloe. APPENDIX. l)K8('RTIMIOX OF AX A])V\/r MACUONECTES GIGANTFAIH FPOM AFMTRALTXD, XFAR HUXBURV. This specimen, z\o076, is of interest as it is the first re]>resentative of the mature or brownish-grey colour phase with white head and neck to reach the Museum. This bird belongs to the “intermediate j)hase with white head and neck.’' The back from the /one of transition at the neck to the tail is uniform greyish or dark-brownish grey accordiJig to the lighting, the feathers all having blackish shatts and in most cases lighter edges. The tail is lighter with less brown and the under tail coverts whitish. On the head the forehead is white, the crowti mottled, some ot the feathers being grey at ttie tij>, the hind neck pale almost wliite, sides of the lace white with a few pale grey leathers on the cheeks and near the ga[)e, chin and throat white, the feathers in these parts being white to the base. The under paifs are mottled greyish, the individual leathers being either uniform greyish or becoming darker tow^ards the tip Avith, in many instances, lirownish-grey edges, jirobably due to wear. In all cases the shafts are light, almost white. Feathers on the legs greyish, almost wliite toAcards the base. In general ajApea ranee the bird seems to resemble those illustrated by Falla (B.A.N.Z. Antarctic Research Expedition, Vol. If., Birds, figure 133, right foreground, C'aroline Coat, Macquarie Island). The soft parts of the bird \\4ien received were: — Iris dai'k, feet reddish ^late, webs brownish-grey, chnvs horn, bill Avhitish horn Avith limy incrusta- tion at the base of the u]>per mandible. Measurements taken : — Length 932, wing 520, tail 21 (i, tarsus 191, culimm 99, Avidth 33.5, depth 38.5. S(‘X — male fO.II.L.) 226 L. Glauert. KKV TO THE SPE(MES. I. Flightless birds ... ... ... ... 2 Active hying birds ... ... ... •>. Large birds, white ring round eye ... Pyi/oceli^ adeliae (Hombr. & Jacq.) Aledium sized birds with yellow crests cre-Hatus n/,oselei/l (M.& J.) »Smail birds yellow ere.sts absent ... 3. Nostrils at the end of a longer or shorter tube ... ... ' Nostrils normal ... 4. Size large (Albatrosses and giant petrel) Size smaller (Petrels, Prions, Storm pet- rels) o. Tube extending almost to the tip, bill heavy Marronecfes gif/anfeus (Gmelin) Na-sal tubes short ... ... ... 6. Size large, wing spread to 12 ft. or so ... Diomedea exulans L. Siz(‘ smaller, wing spread to H ft. or so ... 7 7. ('nlminieorn* and laterieorns in contact behind th<^ nostrils ... ... ... Diomedea welanophrijs (Temm ) ( uliniineorn and laterieorns aot in con- tact behind the nostrils ... ... 8 8. (Adminieorn broad, rounded basally ... ( ulniinicorn narrow, bluntly pointed basally ... 1). Blue-grey above, white below Blackish abovt', white below Dark brown or (lark grey above and below 10. Tail broadly tipped with white d’ail broadly tipfied with black 1 I. Black tip to tail w ider (40 mm.), liill less than 25 mm. in length Black tip to tail narrower (to 30 min.), lull more than 25 mm. in length 12. Widtli of hill lOi to 124 mm. ... Width of hill 8 to 1 1 mm. 13. Eamelhie in u])p('r bill well (U'velojjed, visible when the hill is closed Lamellae rudimentary not vi.sible when the bill is closed 14. - Width of bill 18 mm. or tnore^ length 33 mm. and more Width of l)ill 17 mm. or /eer bill fused more or less completely Size lax’ger. brown w ith Avhite wing patch Size smaller, usually dark above and paler below t ... Central tail feathens elongated, twisted Central tail feathers elongated, ?iot twisted ... Bill long, slender, hooketl at the tip . . . Bill shorter, stouter, not hooked at tip Size larger Size smaller Pterodroma macroptera (Smitli) Pierodroma lugens (Kuhl.) Caiharadn dua (Brunn) Stercornrius pouiari nu.‘s (Temm.) Sicrcorarius parasiticus (L.) Frerjata minor (Gmeliji) Middle tail feathers elongated, narrow', red Middle tail feathers, not elongated, not narrow', not retl Dark with whitish head Dark upper parks, whitish below' Upper parts pale grey, under parts with rosy sufiusion in life ... Phaethon ruhricauda (Boddaert) Moms serrator (Gray) ienuirostris (Temm.) Sterna fuscata (L.) Sterna domjaUii (ilontagu) 2.8 25 24 20 27 28 * Ihp u|*pc-r Mlk of AIliafroK^es, Mollymawks. Prions and Petrels conskt of tliree the ciil .nin eorn teniumuuv- in the n.-nl (.r dertnuu and Hanked on either side by a latericornl The ot these plates i> iiu])ortaiit. in 1 ). exulans and D. mfianophiya the cnhuinkoni is in contaet w?th thp iten-orns behind the nostrils whilst in D, chlororhyvcho^ Ldl). is ted ^ bhinth c uinunK orn is broad and rounded basally but iu />. rMororhynchu. it is narrow ind t A darker phase in the spei ies of Sfercororim also oeeiu-' ull our coast. B-IBLIOGBAPIIY. Alexander. M. B.: Ibrds (h’stroycd ]»y Storms. Emn, Vol, XV., Jan.. UK), p 1!).';. Alexander, \V. B. : erocellaniforjiu's in Westeni Australia. Evui, Vol XVII July, 1917, ]). 4U. Alexander, W. B. ; Crested Peiignins in ■\Vcstern Australia. Kmu, Vol XIX April, 1920, p. 295. Alexander, M. B.; Australian .species of Tubinares. Emu, \'ol. XX Julv and Oct., 1920, pp. 14 and 66. Alexaudm-, M. B, : The Birds of the Swan Eiver District. Emu, Vol. XX Jan 1921, p. 149. ’ Alexandei;, \\ )h; Oliservatious and Kecords of Australian Sea Birds. Emu, ^o]. XXJ., April, 1922, p. 261. Anon.: X>w Bii'd for Australia. Emu, Vol. XXVI., Get., 1926, p, 136. C’ampbell, A. J.; X^ests and Eggs of Australiau Birds, 1900. The Kerenguins. Emu, Vol. XXV., July, 1925, p. 1. Fiil])atrick, A. 0.: Birds of Pottnest Island. Enott, Vol. XXXII., July, 1932, p. 80. MacCillivray, W.: A'otes on Sea Birds. Emu, Vol. XIX., Jan., 1920, p. 102. ?\TacCil]ivraY, V.: Birds from a Coastal Boat. Emu, Vol. XXVII., Oct., 1927. p. 92. Mallu'ws, G. i!.: I'l’.e Birds of -Australia, \'ol. I!., 1912-1913. Alatliews, G. il.: Austral Avian Kecord, Vol. III., liGO, p, 55. Mathews, G. M. : A check-list of the Order Procellariiformes. E‘OviUiti\s Zoo- logicm, Vol. XXXIX., 1934, p. 151. ?.i;';lhevvs, G. AT.: Pemarks on Prions. Emu, VoP XXXVII., Oct., 1937, p. 118. Ma hews, G. A1.: Pachyptila or Prions. Emu, Vol. XXXVIl., April, 1938, p. 280. Mathews, G. Al. and Iredaie, T.: A Alanual of the Birds of Australia, 1921. Pohinson, A.: Little iSliearwater breeding on Pottnest Island. Emu, A-ml. XXXT\'., A})ril, 1935, p. 314. Serventy, B. L. : Votes on Western Australian Petrels. Emu, Vol. XXXA^II., July, 1937, p. 50. Sorventy, I). Ti,: Birds of the Islands off Fremantle, AVestern Australia. Emu, Vb)i. XXXVII., April, 1938, p. 265. Whitlock, F. L.: Ocean Derelicts. Emu, Vol. XXYI., Jan., 1927, p. 154. \V;i llcck, !■'. 1 j. : Further Notes on Ocean Derelicts. Emu, A^ol. XXX., April 1931, p. 203. A\']iitlock, F. L, : AUmtrosses and Petrels in the South AVest (W.A.). Emu, Vol. XXXA^I., July, 1930. p. 14. X-Ray Analysis of the Gingin Upper Greensand. 229 No. 17.~X-RAY ANALYSIS (BY THE POWDER METHOD) AND MICROSCOPIC EXAMINATION OP THE PRODUCTS OF WEATHERING OF THE GINGIN UPPER GREENSAND. By W. F. Cole. Read 10th June, 1941 ; Published I7th August, 1942. INTRODUCTION. Available geological literature has reveah'd to the author a scarcity of published data on the weathering products of glauconite or glauconitic green- sand. It was first proposed by Glinka (1) that with the loss of lime, potash, magnesia and ferric iron, glauconite was replaced by ferruginous clay. Cay- oux (2) was however of th(‘ opinion that it alters to limonite. In recent years Denison, Fry and Gile (3) have shown that muscovite and biotite when under- going weathering in the soil, pass by (continuous variations from true micas through stages corresponding to glauconite and illite in composition. The weathering of the micas was accomj)anied by an iiunvase in water content, osited on aluminium foil of thickness 25 p. The nuTliod of [)re])aration used is descrilaxl by Wood (8). With the filh'r us('d in the cours<^ of this work no trace of the component appeared even after tlu' longest (exposures. Suitable jjliotographs with Fe Ko were obtained afh'r 16 hours exposure Avith 4-5 mA tube current at 45-50 K.Y. Films Avith unfiltered Cu radiation could be obtained AA'ith shortfn- exposures. When Cu radiation Avas used the manganese foil Avas still retained as a A\ indoAV for the X-ray tube. X-Ray Analysis of the Gingin Upper Greensand. 233 For other experimental details regarding mounting of powders, films, etc., reference may be made to a previous publication (4). ACC'URAC'Y OF RESULTS. In tliis work the determination of d /n by a graphical method pi’eviously' described (4) was not resorted to. 1’he intei-planar spacings v ere determined by a direct calculation from the Itragg equation nX - 2d sin 0. This was rontit'red possible because quartz present in the coai'se fraction of each sample produced characteristic blotchy lines whicli could easily be recognised and used for cali- bration. I’ho lilotehy characti'r of the lines was a.s.sociatefl with the jjarf icle size. It was found that r (the radius of curvatiire of the film) was constant along a film but varied from film to lihn. This variation was small and its extreme range was not moi-e than 0 • 5 per cent. Accuracy of Results for the Coarse Fractioyis. — -Values of r calculated on the basis of the three t[uartz lines of interplanar spacing 3*34, U814 and 1*538 A were found not to vary by more than 0*25 per cent. Such a. variation lies within the limits imposed by inability to measure any 2,5 (the separation of corresponding diffi*action lines along the film in the etpiatorial plane) to any greater accuracy than 0* 1 mm. Hue then to an error in 2s of 0* 1 mm. in lines used for calibration an error of 0 • 25 per cent, aiul 0 • 09 per cent, is possible for any value of tl n calculated ovta- tlie range 10 A to 1 A. On the assumption that there is no error in r involved in calibration measurements and that r is uniform across the film, the error in d /n associated with an error of 0* 1 mm. in 2s over the range from 10 A to 1 A varies from l’]7 per cent, to 0*075 per cent. Hence the total maximum percentage error, assuming (a) uniform radius along a film and (b) d(2.9) 0* 1 jnm. in calibration and in measurement of an unknown line, is 1*42 p*(H A. Tliis is due to variations in 2s of 0*(d Tum., corres- ponding to the maximum en'or (‘onsicU'n'd possible in measurements of 2s for all but the \ ery I’aintest liiu's. Hence, on this basis, it is considered that the assumyition of a constant radius of 2.798 cms. yields results for the finer frac- tions wliich are no less inaccurate +han those for the coars(T fractions. X-RAY DATA. In Tabh^ I. are tabulated all obs(‘r\'e (H) y 13-95 ms 15-0* U y 12-75 G s 10-12* G y 0-43 G(M) s 10-12 G A'W 9-56* GM m 10-02 .K vw 7-20 U w 5 • 43 U vw 5-47 G in 4-94 G VW 4-99 (JM vw 4-93 G(M) r 4-50 G(IVI) s r 4-48 G(M) m 4 - 45 GM r 4 • 53 GM w-m 4-47 QGo 4 •26 -...s' QGo ms/ 4-21 Q = J 4-20 Cg 1 4-09t Cg m(d) L 4-12 Cg vw 4-18 Cg L 4-07 Cg n 4-07 .... f 3-80 GCg ms 3 • 69 GCg vw 3 65 Cgf.T VV' 3-70 G('g m/ 3-68 s 3-34 QG w 3-34 QG m 3-33 QG y s'- 34 Q(J .... t s 3-34 G w(d) 3-08J G vw 3 • 09 G W 3-07J O w 2-<)7. G AW 2-95'! GM VW 2-97:i G w 2-84 HGoG vvv 2-73 lacement is largely restricted to mutual substitution of A1 and 8i in [Si 2 ] positions. Tla^ characteristic large spacing for all members except hydrous halloysite occurs at apyjroximatoly 7 A. Hytlrous halloysite, which readily reverts to halloysite by the loss of two molecules of water, has a basal spadng of 10*3 A. The patterns of alt members of this group are destroyed by heating to r>00®C. (2) Monhnorillonite Group.^ — The montmorillonite group includes the fol- lowing minerals ; montmorillonite, saponite, nontronite and beidellite. Mont- morillonite has the ideal formula [Alg] [Si4]Oio(OH)2.xH20 in which extensive isomorphous replacement can take place. Substitution of [AI 2 ] by [Mgg] gives saponite [Mg 3 l[Si 4 ]Ojo(OH) 2 .xH 20 ; substitution of [Al 2 ]“by [¥o^] gives nontronite fKo 2 ] > substitution of Si by A1 in position together with the replacement of O by (OH) or the replacement of A1 in [Alg] position by other ions gives beidellite [AI 2 ] ri^i 3 Al] 09 (OH) 3 .xH 20 in which the Si02 ; I'atio lies close to 3 ; 1. The characteristic* large spacing for air dried Tnatc'rial of all ineinbc^rs occurs at 14-lo A. Members of this group show rc'vcrsible lattice shrinkage and expansion acco 7 *ding to tlu'ir water content(ll). Ujjon heating to o00°C. the 14-15A basal spacitig shrinks to 10 A, (3) Mtca Group.- — In this group no subdivision is at ))resent recognised. Gruner (12) however, has shown that the structures of glauconite and mica are almost ideiitical. Mica has the ideal formula KfAl 2 ] [Si 3 Al]Oj 9 (OH )2 in which extcnisive isomorphous replacement can take jilace. Replac^ejnent of K by H 2 O accomfianied hy substitution of Si for A1 in tetrahedral co-ordination or (OH) for O, together with Mg and Fe replacing [AI 2 ] Avith octaliodral eo- ordinatioii, results in a mineral of the glauconite type. The charactcaistic basal spacing for the mica group occurs at 10 A, No {'hango in pattern is produced by heating members of this group to 500T, As the minerals Avithin a group cannot readily be distinguished in the diffraction ]3attern of a soil colloid, the terms “■ kaolinite ” and ” montmorillon- ite ” A\ ill be used, for brcAuty, in the remainder of this paper to designate a mineral bcdonging to the kaolinite group and montmorillonite group respec- tively. In the draAving up of Table I. an attempt has been made to assign a pos- sible origin to all tlie obser\a*d lines. For montmorillonite and kaolinite the data published by Kelley et al. (13) Avas used. This data includes only spacings to whudi Miller indices related to a definite unit cell could be assignetl. The kaolinite spacings were those tabulated by Gruner (14) with 7’espect to the monoclinic unit cell a ^ 5*14 A, b = 8*90 A, c 14-51 A, /S 100° 12'. The montmorillonite spacings Avere selected from the data ofMaegdofrau and Hofmann (15) Avho referred the mineral structure to an orthorhombic cell AAuth dimensions a ^ 5*18 A, b = 8*97 A, AA'ith c variable according to the degree X-R-ay Analysis of the Gingin Upper Greensand. 237 of hydration. The data relating to glauconite was taken from Gruner (12), who, by refemng the mineral to the mica structure, was able to assign Miller indices to most of Ids recorded lines. Quartx data was taken from Favejee (16) who has tabulated estimated interplanar spacings relating to the unit cell a = 4-903 A, c - 5*393 A. Data supplied by Ifanawalt, Riim and t revel (17) was used when considering the possibility of the presence of the oxides and hydroxides of Fc and Al. In preparing Table I. the ])rinciple was eiujiloyed of only assigning an origin to a lino \vhen on intensity considerations the line was likely to bo contributed to by the mineral in (juestion. Consideral>lo difficulty was encountered in recording 10 A diffraction lines because a pronounced central scattering extended out to about this spacing. Measurements of spacings in this region consequently vary considerably. Films taken with Cu radiation, however, often gave clear evidence of the existence of a 10 A line. (1) Upper Greensand, 1 he original observations of early workers on the high glauconite (’ontent of this sand have been confirmed. A comparison between the fine and coarse fractions seems to indicate that there is a higher glauconite content in the coarse fraction. I'he agreement between the glauconite spacings listed by Gruner and the spacings recorded for the coarse fraction of tlio uiijier groi'nsaiul is sufficiently good to i:»ermit of a calculation of the unit cell dimensions of the Gingin glauconite assuming the mica structure of glauconite and the indices assigned by Gruner to the various spacings. In this calculation the following lines and corresponding indices were used : 10-12 A (002) ; 2-o8 A (2o¥) • 1*510 A (000) ; 1*305 A (400). Table 11. IXIT CELL 1)IMEN8I()NS OF GLAITOMTE. Gingin glauconite. Gruner’s average of six sam])les. ('o 20-21 1 o ba 9 -00 9-07 ^0 5-24 5-24 94° 58^ 95° m' Simpson (18) has noted the presence of dufrenite (a bright green basic ferric phosphate Fe 2 ( 0 Pr) 3 P 04 ) in the upper greensand but in the absen(‘e of any X-ray data, known to the author, with resp(a*t to this miruu-al and in fhe absence also of reasonably pure specimens fVoni which X-ray data may be obtained, no conclusion has been reached as regards its presence in the sample examined in this study. The jirestnice of (|uartz: lias beta) prex'iouslv refernHl to. (2) Subsoil. The most noticeabk' differences between this pattern and tlu' jirevious one ai(> the existence of a 15 A and a 7 A spacing. The 15 A spacing shrinks to 10 A and the 7 A spacing disappears when the sample is licated to 500°C. •238 W. F. Cole. These lines, it is considered, are the characteristic basal spacings of mont- morillonite and kaolinite. The 10 A glauconite spacing is greatly reduced in intensity. In comparing the two fractions of this sample it appears that the ■coarse contains more quartz and the fine contains more kaolinite. <3) Soil. In the coarse fraction quartz is the only constituent that can be definitely Tecognised. In the fine fraction however, in addition to the quartz pattern, there is a number of lines, including a 7 A line, which disappear when the sam- ple is heated to 500°C. These lines, it is considered, are due to kaolinite. In both fractions there is a little haematite and /or goethite and a possible trace of glauconite. Additional effects of Heat Treatment. Several changes, in addition to those previously noted, occur in the X-ray patterns of the heat treated samples. In most unheated samples a broad area of strong intensity extends from 4* 50-4* 10 A, which after heat treatment, is resolved into two lines (4-5 and 4-1 A). In the greensand and soil samples the line vtth the larger spacing is the more intense whereas in the svibsoil ■sample it is tlie l(;ss intense. The line with the smaller spacing is due to small amounts of candle grease which lined the walls of the celluloid containers in whicli the samples were mounted for X-ray analysis. The method employed for the preparation of celluloid containers using candle grease ha.s been ])r(‘\'iously describ(‘d (4). randlo grease is characterised b}^ two strong lines which occur at 4* 12 anarticles, but an occasional particle is sufficiently well ori(nit('d to give a biaxial figure of negative optical character and of small axial angle. {(I) Ihirticles distinguisfM'd from (c) mei'ely by the fact that they are only very weakly liinTringent and their refractive indices' vary from 1*535- d-555. Tlieir colour is light yellow-green. TJie Snhsotl.- Tiu' most notiiu'able featun' of tlu^ subsoil is the large amount of crystalline material of which only a small amount is ijuartz and glauconite. The (piartz as before forms subangular grains <*oated with other minerals. 1 he glauconite is recognised by its possessing similar optical proper- ties to the glauconite in the upper greensand. The refractive indices are, how- ever, a little lower, the range being from 1 -58- -I •()(). The remaimltn* of the sample is made up of the particles a — d considered in the upprr greensand section. 240 W. F. Cole The SolL 4’h(‘ most noticcahh' featinv of tiio soil sample is the small amount of crystHlliiu' material of whieh fiuart/. is the main constituent. This- mineral is h<‘avily coatial with iron oxides. Oc.casional grains of glauconite and the alteration t)i’oducls previously described (of which some of the brown grains si)ow traces of a, splierulitie cross) occur but the main fraction of the sample consists of opa(|iie grains yellow-brown in reflected light. Reddish- brown iron oxid(‘ grains ai‘(‘ frc(]uent. The Fine Fnietions. In the det(^rmination of the opti(*al properties of the 2 ^ fractions use was made of tlie fact (HI) that indtvidunl particles existing iJi a soil colloid suspt'Tision tend to orient themsc'lv'es after drying into aggregates which possess uniform optical properties. This takes place even wht'n two or more- clay minerals anj 2 )resent so that the o 2 )tical projaa-ties of the aggregates- depend upon the pro])orlion of the different constituents forming the aggi’e- gates. Jn the soil colloids oxamiiuMl in this study there was no marked tendency for the aggi'ogates to show a crystallographic orientation of the individual particles. No aggregatf^s were found sufficiently well oriented to give an intorf(n*ene(i ligia*e from whicii the oj3tioal character could bo determined. However, enougli oriented aggregates were found to determine tlie oj 3 tical constants. I he uj> 2 >er gn'ensand fratdiou wlien viewed beneath the microscope was seen to bo in the form of light gi-een aggregates up to 50 /x and greater in di- amet(*i‘. About 2i) p(‘t’ cent, of these aggi'cgates show fairly uniform crystallo- grapfiic oi’ientatiou of individual compoiieiits and for these y — a ^ 0*01.. Tlie remnind('r of tin* aggregates are non-liirefringent. I he subsoil fi*action whcai vknved b<‘neath tlie mieroseopo was S('en to be in tlie form of light yellowish-brown aggregates uj) to 50 fi and greater in di- auK'ter. I lu're is aljout the same jiroportion of bindVingfnit material as in the uppi'r gretaisfuid sampk*. The l)ir(dringetLco as measured appears to be about 0-005. I’lu's figm*(‘ is probably low due to tlie fact that tlie colloid particles ]3i*ol3ably coat('d witli iron oxides or hydroxides whicli obscure tlu' birt'fring(ni( (‘, I 1k^ soil fraction when viewed beneath the microscope was seen to bo in tlie form of dark lirown non-bir('fi’ing('nt particles up to 50 jx and greater in dianu'tcu*. The t)pti(',al jirojierties of aggn'gates of the soil colloid fractions from the upper grec'nsanti, subsoil, and soil are summarised in Table ITT. Tahle Til. OPTICAL DATA RELATIN(; TO THE FINE FRACTIONS. Sample. a ± 0-002 Mean Refractive Index + 0-005 r + 0-002 y — a Ippcr greensand ... 1-018 1-028 0-010 Subsoil l-OOO 1-005 0-005 Soil ... 1-020 ... X-Ray Analysis of the Gingin Upper Greensand. 241 DISCUSSION. Coarse Fraclions.—X-ray evidence indicates that in the coar.se fractions the glauconite content decreases and the <|uartz content increa.ses in pa.ssing from greensand to soil. The microscopic e.vamination confirms this. X-ray evidence also indicates the pre.spnce of montmorilloniti' and a little kaolinite in the subsoil, a possible trace of montmorillonite in the greensand and of kaolinite in the soil. In view of the fact that the sample.s e.vaminetl contain & fibrous mineral with .sphernlitic structure which lias not been iihmtifiod and which readily alters in a characteristic manner to give a variety of particles .some of which have optical proportic.s .similar to clay minerals it has not been possible to verify beneath the mioro.scope the pre.sence of any clav minerals in these fractions. Fine FracHons.^Fov the fine fraction.s X-ray evid.>nce indicates that ‘glauconite is the main constituent of the uppcT green.sand, montinorillonito the main constituent of the subsoil, and kaolinite the main constituent of the soil. Optical data relating to pure clay minei’als havi' been summari.sed by the author elsewhere (19). Members of the montmorillonite and mica groups have similar optical properties which are markedly diftei'ent from the •optical properties of members of the kaolinite grouj). In Table III. it i.s seen that the upper greensand and subsoil have .similar optical propei-tie.s, both sample-s being more birefringent than tlu' soil sample which is non- birefringent. The non -birefringent character of the soil sarnjile could he due to the presence of kaolinite or to the presence of coatings of iron oxi.les ■or hydroxides on the soil colloid particles which tend to mask the hirefringence. It is believed that both play a part. It must he realised that the ojjtical properties of aggregates of the fine fractions dt^jjend on all the con.stituents present, so that only broad generalisations may be drawn when the main •constituents alone are considered. Generally speaking the o])tical properties ■of the fine fractions are consistent with the composition as determined bv X-ray analysis. CONCLUSIONS. With regard to the composition of the .sample.s X-rav data lead to the ■conclusions summarised in Table IV. Some modilk-ations, the extent of which it IS difficult to estimate, may be necessary in the light of the following (a) The relation between relative intensities of lines and relative (piantity of corresponding; diifraoting^ niatt'rial. All estimates of relativ<^- quantities of the ditfenml constituents are based on visual obserA-a- tion of line intensities. ^rh('S(' estimated relative (juantities may correctly indicate tlie actual relative (|uuntities only if the ratio of the intensities of patterns of differimt constitiK'uts' is ecpial to the ratio of the proportions of tlie constituents. This may not l)e true and may (lepart widely fn)m tlie truth if quartz is oim of the con- stituents since the diffracting power of <)uartz is markedly greatca- than that of any clay mineral (9). (b) Microscopic evidence. From Table IV. the following generalisations as regards mineral changes an passing from greensand to soil may be drawn : — (а) The content of glauconite decreases. (б) The content of kaolinite, cpiartz and oxides and hydroxides of iron increases. ^c) The montmorillonite content is concentrated in the sul)soiI. 242 W. F. Cole. It is hence suggested that in the weathering of glauconitic greensand the- glauconite alters firstly to a clay of the montmorillonite group which later is replaced by a clay of the kaolinite group together with free quartz and haema- tite and /or goethite. The formation of kaolinite through the splitting off of silica from montmorillonite has been suggested by Kelley et al. (13) as an explanation of the occurrence of kaolinite clay overlying montmorillonite clay in the Suscjuehanna soil colloids. It is quite logical to expect on this hypothesis that kaolinite would be present with montmorillonite in the subsoil fraction but it must also be borne in mind that contamination of the subsoil fraction with the overlying soil would give kaolinite in the subsoil fraction. Such a contamination of the subsoil fraction is considered to be negligible. Table IV. COMPOSITION OF THE THREE SAMPLES. (From X-ray evidence only). Mineral. C})per Greensand. Subsoil. Soil, Coarse. Fine. Coarse. Fine. Coarse. Fine. Glauconite much Much Little Little Possible trace Possible trace Quartz Little Very little Much Very little ■\'ery much Little Montmorillonite . . . Possible trace Possible trace Much Much ... ... Kaolinite ... ... Little — 9 ^ Little Possible trace Much Haematite and/or Goethite Tery little ... Possible trace Little Little Xote . — An arrow ( — >•) indicates an inereawing amount of kaolinite. ACKNOWLEDGMENTS. The investigation was carried out in the Physics Department of the Univwsity of Western Australia during tluj tenure of a Haekett Research Stu(l('ntship and a Commonwealth researcli grant. The author wishes to express his thanks to tlie University of Wevstoru Australia for the former and to the Council for Scicnt ilic and Industrial Research for the latter. The X-ray equijnnent was originally providetl by a Commonwealth gi-ant. Tlie author greatly appn'ciatc^s llie facilities that have been made available through the Council foi- Scientilic and Industrial Research for carrying out the work. The |.u’ohi('m was suggested by Professor E. de C, Clarke and Dr. D. Carroll of the De])artment of (Geology of the University of Western Axistralia to whom the author is inde))ted for the pro\ ision of samples and apparatus and for many helpful discussions. The work was carried out under the supervision of Mr. J. Sliearer of the De])artment of Physics of tlie Lmiversity of Western Australia to whom the authoi''s tlianks are due for the checking of the X-ray data and conclusions and for \'aluable advice and criticism w’hi«‘h has alw'ays been freely given. X-Ray Analysis of the Gingin Upper Greensand. 243: The author wishes to express his thanks to the University of Western Australia Text Books Board and to the authors of “ Junior Geology ” for per- mission to reproduce Fig. I. To the Text Books Board the author is also- very greatful for the loan of the block of this figure. To Miss M. Bowmen the author’s thanks are due for the drawing of Fig. 2. BIBLIOGRAPHY. (1) Glinka, K. : “ Her Glaukonit, seine Enstehung, sein chemischer Bestand und seine- Verwitterung,” Saint- Petersbourg, 1896, p. 114. (2) Cayeux, L. : “ Les Roehes Sedimentaires, Paris," 1916. (3) Denison, 1. A., Fry, W. H., and Gilo, L. : “ Alteration of Muscovite and Biotite in the Soil.” ILS. Dept, of Agric. Tech. Bull. 128, 1929. (4) Shearer, J., and Cole, W. F. : “ Analysis of Soil Colloids by X-ray Diffraction Methods Jour. Ro7j. Soc. W. Ausf. 1939-40, XXVI., pp. 121-131. (5) Clarke, E. de C., and Hanrahau, L. F. : “Junior Geology,” Perth, 1941. (6) Feldtmann, F. R. : “ The Glauconite Deposits at Gingin, South-West Division.”^ Ann. Prog. Kept. Geol. Snrv. West. Ansi. 1933, p. 6. (7) Hosking, J., and Greaves, G. A. : “A Soil Survey of an area at Gingin, Western Aus- tralia.” Jour. Boy. Soc. W. Ausf. 1 930-36, XXII., pp. 71-112. (8) Wood, W. A. : “ Note on the elimination of the d wave length from the characteristie radiation of iron.” Proc. Physic. Soc. 1931, 43, pp. 275-278. (9) Shearer, J., and Cole, W. F. : “ X-ray analysis of some Tasmanian Soil Colloids.’^ Jour. Boy. Soc. W. Ausf. 1939-40, XX\T., pp. 133-137. (10) Hendricks, S. B., and Alexander, L. T. : “ Minerals pre.sent in Soil Colloids : I. Des- cription and Methods for Identification.” Soil Sci. 1939, 48, pp. 257-267. (11) Hofmann, U., Endel, K., and Wilm, D. ; “ Kristallstruktur und quellung von Mont- morillonit.” Zeit. f. Krist. 1935, 86, p. 340, (12) Gruncr, J. W. : “ The Structural Relationships of Glauconite and Mica.” Amer Min. 1935, 20, p. 699. (13) Kelley, W. P., tVoodford, A. O., Dore, W. H., and Brown, S. M. : “Comparative Study of the Colloids of a Cecil and a Susquehanna Soil Profile.” Soil Sci 1939, 47, pp. 175-193. (14) Gruner, J. W. : “ Crystal Structure of Kaolinite.” Zeit. f. Krist. 1932, 83, pp. 75-88. (15) Maegdcfrau, E., and Hofmann, U. : “ Die Kristallstruktur des MontmoriDonits Zeit. /. Krist. 1937, 98, pp. 299-323. (16) Favejee, J. Ch. L. : “ Zur Methodik der Rontgenographischen Boden-forschuns Zeit f. Krist, 1939, 100, p. 425. ^ (17) Hanawalt, J. D., Rinn, H. W., and Ftvvel, L. K. : “Chemical Analysis by X-ray Diffraction.” Iml. & Kng. Chem., Anal. Kd. 1938, 10, pp. 457-512. (18) Simpson, E. S. : “Contribution to the Mineiulogy of Western Australia ” Jour Boy. Soc. W. Aust. 1936-37, XXII T, pp. ' 17-35. (19) Cole, W. F. : “ Results of a Microscopic Study of some Soil Colloids.” Jowr Boy Soc. W. Aust. 1940-41, XXVH.. pp. 1-7. Quo/ty^ S^reom Pits'- 'rf“’A ^ounhtn Quarr^/p^ 'S^ove/piA f^ocK Quarries reor\(w*^Y Laferi fi GEOLOGICAL C05NELL5 AREA RrrrRCNcc LOCALITY PL AN Geolofgicol Boundaries Ed^a of Later I he Rubble SfriKe S Dip Quarr.es (Z Form Lines Local’ion L.ines \ Lower ChiMenn, Areo I floho^ny '• CreeK IV le^ Scolc of Chains '? LEGEND Go^neKs /Area Sand ^ Obscured Lateril'e d BreoKaway Epidio^i^e HferphyriMc EpidioriLe Cordup Series Quciriz 'Blow" A Veins G rani ^e ChloriLic Scl^is^ KelmscoH Armodqle t Wongon< -T* Cordup i Area Scoforfh Lu ferile JreaKa wuy C DS. Davis s f \ / * ^ 1 7 ■■ / * \\ 1 \ / i 1 \K M * v/ \\ ' \ N - - - ^ 3 t 1 % 1 \ \ s \ 1 \ 1 \ 1 \ X .-r. L L L L L L L L L L L L ^.yp L L L L L t L L L L L ' L L L L L L' L L L L L L L L t_ / \ -~.L L / L Rifue Range. L® L StaTG? PICK WOQK© OvjAPPV 'A . V. ' MZ-i " , L L AT ^ l_ \L T L ' "1 ^.,X- .L-' X L ''L ' X L _L L L "l L j/ L L \L^ L L L L L L/ L L L \ l - L L lllux’llll^ L L L / L L L L L " L ft, L L L J. i: i L J I L L L L L L L L 1 L L L _ L-" ' L L L L L L L L L L L L ;L L L "L L L L ^ L /' L L L; jX L L 11 L L X L L X L ■ L L L/ L L L®°° L L L / L L ' L L L L L /'L L L L L L / L L L L L L L 1 L L L L L L / L L L L L ' L L ' L/' L L L L L L L I L L L .0 L L 8 : Malkup Locality Plan i x wjinotpoiwc \Lpw^CH.rTni.NC L.i X L. L L ^50 L L ^ L ^ $/ ^ ) L. Arkaoall W >« 4 £OMV 5 - A A A A / A/ A 1 !V^ i *\> / . 4*7 A / A. ^ Se/y >TPC/^ v*70^ > ►io- 4 ^" / A Ais: A A A A / r/o ou fcrofo A ✓ A #/ h> ^ A .*n!r^« A 350 ', A/o oc//c)z>lb M? OutcTO'pS £::^^r/Pc So//. rA ^5 The Geology and Physiography of the Gosnells Area. 245 18 .-THE GEOLOGY AND PHYSIOGEAPHY OP THE GOSNELLS AREA By C. E. S. Davis, B.Sc. (Hons.). Read 10th June, 1941; Published 2nd October, 1942. CONTENTS. I. Introduction ... II. Physiography : A. General Features B. Darling Scarp C. Streams D. Remnants of the Peneplain E. Effect of Geology on Topography F. Springs III. Structure and Distribution of the Rocks : A. Xenoliths B. Granite C. Acid Intrusives into Granite 1). Cardup Series E. Basic Dykes ... F. Edge of the Pre-Cambrian Rocks G. Laterite : (1) High Level (2) Low Level H. Later Superficial Deposits ... IV. Petrology : A. Xenoliths B. Granite C. Acid Intrusives : (1) Pegmatite and Aplite (2) Quartz Masses D. Cardup Series ; (1) Arkose and Slate (2) Sandstone (3) Metamorphism E. Basic Intrusives ; (1) Uralitised Quartz Dolerite (Epidiorite) (2) Biotite Epidiorite (3) Biotite-Epidote Veins (4) Porphyritic Epidiorite V. Geological History ... VI. Acknowledgments VII. Bibliography ... Page 245 246 246 247 247 247 247 248 248 248 249 250 251 251 252 252 252 253 256 256 257 258 259 259 261 262 262 262 263 263 1. INTRODUCTION. The Gosnells Area lies 15 miles south-east from Perth, on the long, straight Darling Scarp which separates the loAV-lying plain (on which Pertli stands) from the level Darling plateau whose surface is about 1,000 feet above sea level. The Darling Scarp here forms the western boundary of the Pre-Cambrian rocks which outcrop over so much of Western Australia, j 331/42 246 C. E. S. Davis. In tli(* (‘astorri part of the eountry shown on the locality ])lan, the main rock type is ^’ranitic, covered in the higher country by a ca[>ping of laterite. The granite is cut by quartz masses (mainly thin veins^ but there is also a huge (jiuutz blow at Gosiiells) and by basic dykes. On the western margin of the granite, stee))ly-dipping, slightly metamorphosed Pre-Cambrian sediments of tlu* Cardup Series outcrop from (Josnells intermittently south- wards. The Cardup Series and granitic, rocks close to the contact at Armadale were studied in l!)30 (Prider, 1941), and, in 1940, areas were map])ed at Gosnells and at Wougong-Chirdu]), respis-tively north and south of Armadale. At the end of the description of the Wongong-Cardui> area (Thomson, 1941) the knowledgt' gaimal and the problems arising from the study of aU three areas have Ixarn summarised. The (lOsnells Area is a strip of th(‘ scarj) about 4 miles long ami 1% miles wide. Particular altention was i)aid to the northern part where the Cardup Series and quartz blow crop out, and the granite, basic dykes and quartz are exposed in (piarries. The coiiidiy was map])ed entirely by pace- and compass-traversing by the writer working alone. These traverses were tied ire(|uently to th(» Lands and Survey l)artment’s pegs. .Form lines were dravm from levels obtained by aneroid baromCer leadings, woi'king from Gosnells railwav station as datum. IT. PHVSIOGPAPHY. A. (reneral Features. East of the i)resent foot of the Darling Scarp, there was originally a laterite-covered iiemqilain (Jutson, 1934, p. 291). This ))eneplain was later elevated and youtld’ul, westerly-tlowiiig streams have now remo\ed the laterite and ex]>osed tlu' underlying pre-Cambrian rocks along a belt, gener- ally from hall a mile to two miles wide. Pre-Ckimbrian rocks are exposed fartlier east, only in the valleys of the larger streams and in some moiuuhiocks rising above the plateau. B. The Barling Scarp. As Jutson concludes (1934, p. 87), the elevation of the Darling peiu'plain was effected by faulting, but the present face of the Darling Scarp is to the east of the actual fault-plane. Ko trace of the fault Avas detect(‘d in the Gosnells area. W oolnough (1918, ]>. .399) believes tltere were two periods of uplift of the penei)lain. After tin* first movements, mature valleys Avere formed which now occu|)y so much of the inland “wheat-belt’^ country: and later movemmds elevated the ])ene[)lain io its present height. The topography of parts of the Gosnells area is comi)osite, and supports Wooluough’s belief. This is especially marked in the valley of Ellis Brook, which is mature above a height of 929 feet (the lop of the Rixty Foot Falls). At and just l)elow the falls, the streuTU bed en considei'ably. Fpstream are some meanders, now slightly entrenched. In many other ]daees, the scarj) (lattens a])ove a height of about 500 feet. Clarke and AVilliams (1929, p. 167) note that terraces at a height of 450 feet in the Helena valley probably record a pause in ui)lift along the Darling fault. The Geology and Physiography of the Gosnells Area. 247 As maturo valleys were ])Vodueed in the first cycles more time must have passed between the two movements than has passed since the last uplift. But until we have more quantitative data of the rate of erosion of the local granite, we cannot suggest the absolute age of either movement. C. Streams. None of the streams draining the area can be described as the ‘bnajoP^ stream. Most of them are short, flow due west down the scarp and may be explained as consequent on uplift of the peneplain. Jutson (1934, p. 1(19 and tlg'ure p. 171) states that the initial drainage of the elevated peneplain was to the south-east. Traces of a south-easterly ry steeply.^ (Quartz veins, about two feet wide, may be traced from the granite into tlie schist. They resemble other quartz veins from the area, whhdi, it will be shown, were formed at a late stage in the cooling of the granite. Thus, the schist existed before the granite had completely cooled, and it is probably a xenolitJi in sj)iu* of its dyke-like outcrop. Although large number of xenoliths was noted near Gosnells, they are actually imsigniticant in bulk compared with the host-rock, and, except £oi the chloritic schist, these small masses are not shown on the uiap. In- numerable xenoliths are found at Armadale (Prider, 1941, p. 29) and small xenolithie fragments occur in the granite as far north as Stathands (Clarke and Williams, 192(), p. 1()9). B. Granite. The granite of the area is nearly all massive, and even in the better exposures in the (piarries no flow structures were detected (although these may have been largely obscured on the stained joint-surfaces forming the walls of the quarries). Faint flow-layers (and, in two places, flow-lines) are found in isolated outcrops and boulders in the extreme south and in the north-east. The strike of the flow-layers is east of north, but very variable, and their dip is to the east at 40° to 70°. It is imiiossible yet to say whether the massive granite grades into or intrudes the more gneissie granite of the south of the area. Mapping and detailed ])etrological examination of the country immediately to the south may reveal whether the granites are the same or dilferent. C. Acid I^itrusives into Granite. Dykes of aplite and pegmatite cut the granite, but they are too short and thin (none was over two feet wide) to be noted on the map. (Quartz veins have also been found, and they intrude the pegmatites. They usually outcrop strongly, and many of them are large enough to be shown on the map. Near most (luartz veins, tin* granite is altered and has been weathered away. The edge of the quartz ‘‘blow” shown on the map is the edge of a zone of decomposed granite throughout which quartz veins occur. Thin quartz veins (from an inch to three feet wide) have a general north-south trend (of 22 directions measured, 14 lay between 340° and 40°). Such thin quartz veins, however, rarely persist for over a chain. A larger The CJ-bology and Physiography of the Gosnells Area. 249 (jUfirtz vein, nearly 100 feet wide, forms the crest of the steep northern slope of stream A and crops out over a distance of about half a mile on a course just north of east. A lar^e family of veins makes up the quartz “blow.’’ It may be traced for 1% miles till it is lost under the laterite cover to the east. Small pebbles of milky, sub-ang-ular quartz in laterite nearly a mile farther east are probably derived from an extension of the “blow.” Near its southern end, the quartz body consists of four or five parallel veins (each about one chain wide), spread over a width of 20 chains. Be- tween these veins, the underlying rock is obscured by soil and by quartz talus. In (juarries and cliffs in the ^0>low,” however, schistose quartz sericite rock and thin quartz \^eins are exposed between the strongly outcropping veins. The “blow” runs slightly obliquely to the course of individual veins. The latter trend about 80° south of Ellis Brook, and between 50° and 90° farther north. Vertical major joints are common in quarries and cliffs, and strike 135° near the White Bock and Mountain Quarries and 65° north of Ellis Brook. Sericite schist and (iuartz are roughly banded near the :Moun- tain QuaiTy: the sti'ike and dip of these bands (strike 20°, dij) west at 00°) may i:)ossibly be the strike and dip of some of the veins. Tin* l)anding may be thought to be a relict sedimentary structure, and the quartz “blow” to be a huge xenolith of sedimentary rock in which (juartzose bands now appear as quartz veins. But there is w further sign of sedimentary structures even near the centre of the mass, and petrologically, both in hand s})ecimen and in thin section, the quartz of the “blow” resembles that of the smaller veins, and is thus igneous in origin. The acid igneous solutions from which the quartz veins are derived rose along planes of shearing and faulting in the granite. But, as the granite is structureless, there is no apparent explanation why these directions of weak- ness trend predominantly slightly east of north. Many of the <|uartz veins have been examined for gold. The large vein just north of stream A has received particular attention. Five costeens have been dug in it, and Messrs. Boss and Son, in 1909, put in an adit-crosscut 1/5 feet long and a shaft 2/ feet deep. The State Mining Engineer, A, Montgomery, visited the prospecting allotment and reported (1910, p. 124) that “the reef is very poor indeed in gold and (juite un[)ayable at present.” The cross-cut was (‘xtended 40 feet, but mining Avas then abandoned and nothing has since b(‘en done. The (juartz of the “bloAv” Avas recently quarried as an aggregate for concrete and for bitumen roads, but owing, it is said, to the heavy AA’ear of the rock-crushing machines, it has not been Avorked since 1929. 1). Cardup Senes. This series outcrojis west of the granite, near Ellis Brook. It strikes nearly due north, parallel to its contact Avith the granite and dii)s Avest at (usually) 50° to 55°. The dip steepens to vertical near the barite pits, but this is due to buckling ])y the nearby basic intrusion. Faint jointing-^ practi- cally normal to the strike, is deve1o})ed in the slate's, but no fracture cleavage or dragfolding Avas noted. 250 C. E. S. Davis. Outcrops urc poor, and sandstone rubble often obscures slate and cons-lomerate beds. There are thin bands and lenses (from an inch to six inches thick) of arkoso in slate and bands of sandstone in eoin>foinerate so that the succession deduced from exposures in pits just north of Ellis Brook has been sh<>hlly generalised. It is, in descending order:— llematitic Sandstone ? (M'estern bound obscured.) Sandstone with Cherty lenses 30 feet. Slate . . 20 feet. Conglomerate 24 feet. Vughy Sandstone (with barite) 8 feet. Basnl Conglomerate 15 feet. Although coarser-grained than those which have been worked at Armadale and in the Wongong-Cardup Area, the Gosnells slate is of excellent quality for brickmaking. Shallow trenches have been dug to jirove its extent and it is (‘xpected that, although tlu' slate crops out oven' such a small area, it will bo workod in Iho near future. The eontnet. between stnliments and ,u']‘anite is nowhere oxi>osed, but it may be mapped within five feet on the hill just nortli of Ellis Flrook, and within JO yards in several otlun- |)laees. No i»-ranite apophyses are i'ound in the sediments, and small (juartz veins in the granite may continue riu'ht to the contact but do not intrud(‘ the sediments, j’nst as has been fotuid at Aiimulale (1 iid(*r, 1041, j). ,30). This indicates that the Cardup scu'ies is younger than the granite and (piartz veins, a coiieliision which is consistent witli their very low grade of metamorphism. E. Basic Dykes. All previously mentioned rocks lun e been intruded by epidiorite dykes. Most of these dykes trend slightly west of north Avith dips (where nieasur- al)ie) to the east at 60°. A Lew trend north-west, or very rai'oly, due west. Their width vai'ies from five inches to about five chains — it is usually about a chain. Ln the (-entre of the wider dykes expos(‘d in the fpiarries, ^dadde)-" joints (Balk, 1937, p. 97) are developed jierjiendicular to the dyke walls, but their marginal three feet is s('histose and contains segregations and veins of quartz, pyrite, calcite and epidote. Epidote veins may thread the epidiorites and also enter the surrounding granite. I bin, irregular, basic veins (nowlu're move than two feet wide) have been exposc'd in the AVhite Rock (^luarry. They occupy joint-cracks in the gi unite and H'semble the schistose iiiarg*ins ol the dykes in niinei'alogical composition. A porphyritic basic rock (porphyritie epidiorite) intrudes the Cardup sediments. The dumps of two shalloAv pits dug in this rock contain small masses of barite, up to about four ])ounds in weight. Nothing is now visible to indicate further the mode of origin of the mineral. Non-porphyritic epidiorite dykes intrude the Cardu]) series at Armadale (Frider, 1941, p. 43), but no such intrusions were found over the small area where the sediments cro]) out at Gosnells. The Geology and Physiography of the Gosnells Area. 251 F. Edge of the Pre-Cambrian Rocks. The western boundary of the Pre-Carnbrian roeks cannot be drawn accurately. Some small, isolated masses of j^i-anite ar(' found several chains west of the edge of the continuously outcropping granite. These may be outcro]>s, and may be boulders. Granite fragments are found in a well, 20 feet dee[», 200 yards W.N.W. oi' the jMountain Quarry, far from any granitf* outcrop and 100 yards west of the nearest quartz outcrop. Near the 8ealorth Home, a (|uartz mass extends far west of otluu' outcrops. But in spite of these irregularities, and other irregidarities due to laterite in the south, the boundary runs faii'ly straight in a north-soutli direction. Outcrops of (lie Cardup series are very discontinuous. They are found next at Kelmscolt, dVo miles farther south (Honman, 1912, ]>. 63). They may exist, however', under the laterite rubble which is so common in the south of the Gosnells area. Tei’tiary beds, found in Irores near Perth, are probably developed west of the Pi*e-Cambriaii rocks. A sulr-artesian boi*e {dejdh 172 feet) was put down near' Gosnells in 1872 Its exact location cannot be ascertained — it is mentioned by Brown (1873, p. 10), who staled that (among other things) fragments of lignite had been reportral from the borr*. Evidently the bore Avas put doAvn entirely in tlu' youngm' rocks west of tlu' P]*e-(bunbrian : Sini 2 )son (1916, p. 173) called them ‘^Mesozoic sandstones and shales.^’ G. Jjaterite. As the laterite has been studied maiidy iir the field, it is best to discuss it at length now. ITiere are two distinct types: liigh- and low-level laterite. 1 . Jligli-LereJ Laterite. The edge of this capping lies at a height of belAveen 709 and 850 feet aboA'e sea level. Although in places it is marked by a ‘‘breakaway” — a scarp (up to 30 feet high) top]jed by a few feet of solid pisolitic laterite, — more often the edge of the latru’ite is covered by a. feAV feet of sand and rubble so that it cannot l)e n]ai)ped with certainty. Furthermore, laterite is found in places at a level below that of the breaka.Avay only a short distanc(* away. North cd' stream A, there is a small out('rop, .30 feet below the scaiq), and north of Wright Brook, a ridge; about 50 feed, below the breakaAvay, projects for Vi mile. OTiese outcro])s are separatee! from the* main plateau by rubble which in olhew places extends for 200 fe^et bedow the breaka\Amy. Detritus obscures most granitic outci'ops over the whole scarp east of the SeHfe)rth Home, but it is now being cut thi'ough by a stream. A pit has been elug a chain away from and a few feet above the edge of the laterite outcrop, ea^t ol* tlie White Hock Quarry. It- passes through about 30 feet of laterite, underlain by pink and white clay. Laterite must therefore extend below the level of the breakaAvay, and, as has just been mentioned, its boundary is likely to be obscure, especially where there is no breakaway. As the laterite capping formed (Woolnougli, 1918) on the surface of the peneplain, differences of laterite level are due to initial relief of thd peneplain. The thickne.ss of the capping probably rarely exceeds the 30 feet found in the pit (above). In places, it is much thinner, for epidiorite fragments have been found well above the base of the laterite. 252 C. E. S. Davis. 2. Low-Level LaterUe. Ntnu- the foot of the Darling searp, laterite crops out in places, west of all outcrops of Pre-Cainbriaii rocks. In a gravelpit just west of the, Mountain Quarry, the deposit grades from ill consolidated detrital material mixed "with large quartz boulders to more homogeneous laterite. Small '^breakaways” in laterite ha\'e l)(‘eu noteil (e.g., at a height of 300 feet near the south-west corner of the area, and at a height of 180 feet, west of the White Hock Quarry). The laterite may not be all detrital in origin. Near Ellis Brook, just west of the porphyritic e])idiorite, is a red, clayey deposit with incipient pisolitic structurr^ — perhai)s a partially formed laterite. The heavy minerals of this clay and of the porphyritic ejiidiorite were almost exclusively magnetite, so it may be suggested that here the por[)hyritio epidiorite (or some other rock of favourabh' composition and texture) is being laterised m situ. But all that can definitely be asserted is that, near Gosnells at least, the low-level laterite* is too discontinuous to mark the posi- tion of a step-faulted block, as suggested by AVoolnough (1919, p. 16}. H. Later Stiperfcial Deposits. These include silt, sand and talus slopes. A little silt has been de]»osited in the valleys of the larger streams, but it is not shown on tl \3 map. Sand occurs over the laterite on the gently-sloping ])lateau and over low-level laterite near the plain. Except for the lateritic rubble which covers such a lot of the area, talus slopes are small. TV. PETROLOGY. A. XenoUths. Although in the field these are much darker, some contain only a little mor(‘ biotite than does the surrounding granite, so that the two rock-types are very similar in section. Usually, the biotite is a green variety (N = light yellow Y — Z = dark green) Avith inclusions of opaque magnetite (?) aligned parallel to c. In particularly biotitic rocks, recognisable felspar crystals are rare, their ])lace being taken by masses of finely granular epidote. E])idote may he found, too, as coarse, turbid and red stained grannies associated Avitli biotite aggregates. Quaifz is subordinate to biotite and epidote. Sphene (in the form of scattered granules) and apatite are rare. Most of the xenoliths are these biotitic rocks more or less digest(‘d by the ]dagioclase- and quartz-rich granite. Potash fels|>ar is found in veiy few of the xenoliths. Plagd<>f'b^^^* i^^ 'saussuritised and sericitised to a varying extent; and some individuals con- tain small columnar crystals of zoisite. Extinction angles measured on albite twdii lamellae indicate that the felspar averages Ah,.., Avith a range in com- position from Ab^„ to Ab^,.. In rocks containing heavily sericitised felsi^ar, the biotite has recrystallised to coarser green flakes. Sometimes, this process is incomplete and leaves the biotite intergrown Avith muscovite or with colourless chlorite (?). Occasional specimens contain a. broAvn biotite (X = light yellow, Y = Z = deep copper-broAvn). Such a rock (with granular epidote marking the place of felspar) is found at Statham’s. Clarke and Williams (1926, p. 169) noted that such ^finotitic .segi’egations” exist, but they did not describe t hem. The Geology and Physiography or the Gosnells Area. 25 :^ The Gosnells xenoliths resemble the hornblencle-biotite-zoisite hornf'els xenoliths found at Aimadale (Pridev, 1941). They are, however, much more digested by the granite magma. Hornblende has disappeared, and apatite is much rarei’. Quartz (and plagioclase in the more assimilated rocks) have been introduced, and epidote has recrystallised to coarser granules. A rather different type of mateiial is the chloritic schist, ^vhicb resembles one, but only one xenolith, near the foot of the scarp north of Ellis Brook. The former is green, fine-grained and finely banded (about 15 bands to the centimetre). The constituent minerals found in thin section are quartz, chlorite, magnetite, muscovite, and a little (’pidote and a])atite. Alternate bands are rich in <|uartz and in chlorite and magnetite, and the minerals are usually elongated parallel to the handing. The ((uartz veinlets seen in the field are found, on a small scale, in a thin section. As a result of weathering, hematite has been formed from the magnetite and has dis- coloured the rock, and eonse(punirly it was imi>ossible to compare specimens of the sichist from both sides of sti’cam A. An epifliorite dyke alongside the schist contains abundant green chlorite, but the schist re.sembi(*s most closely a dark, massive xenolith found near Ellis Brook. The xenolith is made np magnetite, biotite, apatite, sericitised felspar and mnscovite; a higher grade assemblage than occurs in the schist. Sericife is abundant in the schist, to the exclusion of felspar, of which it is the alteration j)roduct: apatite is less common; biotite has been con- verted to chlorite and quartz has i)een introduced as bands and veinlets. In both rocks, the sericite has been partially recrystallised to small Hakes of muscovite. From its Held occurrence the schist is considered to be a xenolith, jmrmcated ])y siliceous solutions derived at a late stage of cooling of the gi'anite. Xear ijuartz veins in the area, granite is genei^ally altered to a schistose aggregate of low-grade minerals: similar changes in a xenolith have produced the schist. B. (Granite. A hand-specimen of this I’ock is medium, even-gTained and (uther light coloured, or tinted red by weathering. The minerals recognisable arc limpid (iuartz; i)ink, greenish or white f('Ispar and Hakes of biotite. The texture is allotriomorphic granular (or occasionally poikilitic) and the minerals found in thin sections are quartz, a plagioclase near alliite and some microcline with dark minerals (biotite, chlorite ami epidote) very subordinate. Table I. shows the range of mineralogical compo^ition of the granite. Plagioclase in all the racks sectioned is of }>ositive oi)tical character, and in most of them the maximum extinction angle on the albite twin lamellae is 12° to 15°, indicating a coni})osition between Al).,, and Ah,,.. The average composition is then the same as that ()f plagioclase in the xenoliths, but its. range is narrower. The ])lagiociase in some of the granites is saussuritised. In most, how- ever, it is sericitised, and it has been found in all stages of alteration from being perfectly fi'esh to being completely laqilaced by sericite. These altera- tions must be primary; they are too widespread and independent of jointing and other means of ingress of water to be regarded ns due to weathering. Plagioclase in lh(» peumatite aiid a]dite dykes is remarkably fresh. In a 254 (I 1^]. S. Davis few f’l-Miiilcs, loi), plMKiiiclasc is only sliylilly sericitiscd, and in llu-sc n.f-ks as nnndi sanssnrilisfdion lias lakcn place as scricitisalion. These rocks are rich in niicrocline, and under llie microscope several id' lliein shown si»-ns (d' ^n-nshin,!' and e\'en oi' I'anllin^’ cd' small displacement. 'rA«i,K I. Mirroinelrir. An(iIi/.9(>s of (hsuells ( Figures are voliinu' ]>ereeutages.) 1. 2 3. 4. (Quartz 37-9 35- 1 .34 -tJ 37-4 Rlagioelase ... 47-7 4(>-7 37-S 19-7 38-7 Mieroeline ... 10-4 13-1 24-4 Muscovite ... !■() 1 -S 01 2-8 lilotite !•(> 0-2 0-5 Chlorite 0-S 1-2 2(i Apatite. Kpidot(‘ 0-1 1-7 1 -0 1 . (iranite centaininK sericitised plaf-ioclase. Read Hoanl Quarry (for analysl.s see. lalue il., eoluiun I) s])eoiin(‘ii ‘>0 ViHt ('oniaining saiissuril ised niid seriei(ise<) ]>lagi()elase. SiK'oimen (Jranite eoiilaining clear felspar and a erusli l>r('eeia. Tho analysis is of the nnerusluHl part. Tho enisfi l>reeeia oeeiijues 14 per cent, of the scetion. Six'einicn 20 , 70 (). ^ 4. (iranite: gneissie wit li slightly tiirl.id hut unserieitised plagioelase. Speei- nien 2(>,7d4. ‘ * 'riu^ numbers rehu* to eataloguee])artment. University of Western Australia. 'rAHiao 11. Avafi/scs of Darling Dauge Dmnites and oj (■erla.in “ Arm/ye” Types. I. 2 3. 4. 5. SiO., 70-01 73-49 73 -Jit! 73-30 75- 99 430.. 0-20 0-14 0 04 Oil 0-09 Al..().j 13-14 14-24 13-88 12-33 13-14 ... I -08 0-88 0-84 2-58 0-93 FeO 1 -39 0-92 0-93 1-28 0-24 MnO . . 0-01 tr. 0- ir> 0-02 MgO 0-02 0-43 O-f)! 0-20 0-08 ('aO . . 0-2() 1-84 1-09 0-40 0-02 Na^O ... 3-90 3-80 3-22 4-55 3-92 K.>() 4-00 3-42 5-07 4-20 4-70 11., O i 0-28 0-55 0- IS\ r 0-27 H„() 0-.30 0-08 0-11/ 0-8tW 0-08 CO.. 0-07 I‘.,()r . . . 0-07 O-OI 0-07 0-05 Hat) 0-08 0-07 0-09 0-03 lA'So 0-01 0-05 ... 99-80 100-01 100- 19 100-00 100 -09 'fhe alkalis were (h'termined in a granite from Statluim’s by V. It. Le Mesurier in 11)21) ; giving Na^O 4-25 : K.,0 2-:iI ]H'r eeut.* 1. Granite; Koad Hoanl Quarry, (to.snells. Analyst: C. K. S. Davis. 2. Granite; (Jovernment Quarry, Hoya. Analyst: ,1. N. A. (Sraee.* 2. Granite: Mahogany Ui'oek. Analyst: A. d. Robertson (Simpson, lOlb, p. IS). _ ‘ 4. Alkaline (Jranite : Average of 10 analyses (Daly, IDd.’J, p. 10). 5, Kalialaskil(‘ : Average of six analyses (dohaniisen, 1932, p. 49). * Rublished by kind permission of the Government jMineralogist and Analyst. The CrEOLOCiY and Physicx.-kaphy of the (Josnells Area. Potash felspar is always nuieli fresher than the plaj^'ioclase, and any alteration is by kaolinisation, not serieiiisatioii. As it always sliows f*'ri(l- iron twinning', it lias been deseribial as iuiero(dine. No orthoelasc' was recog'iiised in any section of a ^Tanite, Occasionally, thin striiy^'s of clear -quartz tliread the potash fidspar, forniinjj' an injection iuicrope<>niat ite. Quartz is found as ('tear ii'rains which always show wavy (extinction. Next to f{‘ls}>ar, it is the dominant luiiuu'al. I^iotite has the ]d('ochroic sclienu' N = light yellow, = Z = deep brown, but in many ro('ks it has changed to the gr('en \ai'iety which may bo interlaininated with cohmriess chlorite (^) and miiscovitt'. Tin* giaum biotite, too, may la* replaced and pseudomorphed bv W(‘aPly birefringent gi'een ('hlorit(‘. Kpidote, if iiresent, usually forms coarsi* turbid graindes associated with biotite, as it does in the xenoiiths. Rocks containing this 1yp(‘ of cjiidote are, then, slightly hybridised. Epidote in th(‘ veins emanating from the ■epidioritf's is colourless to (-anary yellow and chair. Short stumjis of ajiatite ari^ rare ('onstitiumts of the gi'anitiv They are larger and more ])rominent m completfdy sei'icitis(Hl rocks. Occasionally sericite lias partially recrystallised to muscovite, which is ■develop(xl both as anhedi'a within the iiarent tilagioclase and as laths Ixdween the ]>lagioclase grains. An analysis was made of an albite-rich granite in which tin* albite was moderately sericitised (it is estimated that alxuit 20 jier cent, of tlu' albite has been replaced by sericite). The result is sliown in Table II. in which it is compared with other granites analysed from th(» Darling Ranges near Perth, and with two ‘bi\'erage^^ types. The main point of difference between the Oosnells graniti' and the other two granit(?s near Perth which have been analysed is its extreme pooi-ness in magnesia and lime. It is slightly poorer in alumina, and slightly richer in silica, titania and both iron oxid(*s. The total alkalis in all three rocks ai'c fairly constant, although tlnwt; is consid(*rabh' \ariation in tin* rela(iv(* ])ro- portions of potash and soda. 01' Daly’s average rock ty[)es (1933, p. !)) Ihe Oosnells granite resembles most closely the alkaline granite (Table II., column 3), but is (h'cidefUy low in ferric oxide, a little low in magnesia, linu* and tlu' alkalis; but ricluu’ in silica and alumina. Johannstm-s ‘‘kalialaskite” (Table II., column b) has, like the Oosiudls rock, a low magnesia and linu*. content. From tin* mode of the (losiu'lls granil(* (dkibh* I., (*olumn 1), its alkali content can be calculated to be NTnO KjO 2.0%. The excess ]>otash found in the analysis must be dm* to the sericite* Hakes, and to a little potash- fels])ar in solid solution in the plagioclase. The ])Iagioclase of tin* gi'anite lu'obabty crystallised as a solid seHution containing up to 30% of |)otash r(*lsi>ai‘. The solubility of the two diminislu's rapidly with fail in t(‘mperatur(* (»Johannsen, 1932, p. 141), and the excess potash-felsjiar normally separates as antiperthite. In tlu* (losnells granite, sericite is found instead of p(‘rthitic inlergrowths : the potash-content of the plagioclase se])ai'at('d as sericite, which is stal)le at a low tem])erature and ((ispecially) under conditions of stress. Some I'ocks contain unsericitised plagioclase and much })otash-fels])ar, and many of th(‘S(* microcline-rich 25(i C. K. 8. Davis rocks show signs ol:' crushing, Tlio plagioclaso probably rccrystallised under the stress, and later, smaller stresses jirodnced the fauiting now visible. In ])egmatite and ajilile dykes, which crystallise at a lower temperature than the jiarent gi'anite, ori.ginal plagioclase was comparatively free from iiotash- felspar, and little .sericite has since formed. C. Acid Intrusives. 1. Pegmatite mid Aplite, Tlieso two typos avo often associated, and in such instances the pegmatite occupies the border and apiite irregular ])atches in the centre of the intiu.sioii. C,oais(* ('rystals scatter(‘d nt^ar the edge of an apiite dyke are seen in section to be resoi’]K‘d. Emmons (1940, pp. 5-0) points out that the outstanding diffenmce between pegmatite and apiite is textural: if the volatile' constituents escape from a li(|uid wliidi is crystallising as a pegmatite, the residuum will solidify as a tine-grained ai)lite. rile pegmatites consist of plagioclase, microcline, microperthite or giajihic intergi'outh ot (luail/ and microcliiu*, together Avith fpiartz and small ])ooks of muscovite. The grain size of the pegmatite is up to 10 cm., but that of the aplites is only 0.3 mm. The plagioclase in lioth typ(‘s ot roi-k, as mentioned preA’iously, has been remarkably little sericitised. Microcline is usually subordinate to plagioclase. (jjuartz, though fairly common, is not as abundant as it is in the granite. Out' ajilite contained a few rods of muscovite and a few laths of tourmaline (pleochroic scheme X = deep blue, Z ^ colourless) but no dark minerals. 2. Quorfz As described in the section on their held characters, (piartz masses may be from an inch to o^au’ 20 chains Avide. In spite of this tremendous difference in their size, all such masses appear to be of similar character. The central ] arts consist mainly of hue-grained massi\-e white puartz. In thin section, th(‘y are made up of irregular interlocking grains of (piartz of A-ery uneven texture (in one section the grain size varied from 0.03 mm. to 0.5 mm. and in another from 0.2 mm. to 1 mm.) Fndulose eN'tinction is very common, and is t'sp(*cially notic(‘able in the larger grains. Some of the (piartz from near tht' cmitre of tlu* “blow’’ is friable and “sugary ’ and contains numerous small tlakes of sericite. This sericite content iiH'reases towards the edge of the mass, but nowhere has much sericite recrystallised to muscovite. On the w(\st sale of the Mountain Quarry, sericitic and (piartzose bands are int(*rleave(l. In otinw iilai'es imar the edge of its out('rop, 1h(‘ (piartz “blow” is made up of a stockAvork of tiny ()iiartz veins (about 1 cm. wide) in a mi('ac(’Ous matrix. This is especially clear on weathered surfaces. “Shalv bands” noted in Aveathered rock near tin* edge of the “blow” consist entirely of an aggregate of small flakes of sericite. The granite is generally Aveatiiered aAvay for some 50 yards from the edge of tlu' (piartz outcrops. Where granite does outi'ro]) in the immediate vicinity of the larger quartz masses, it contains stockAvorks of (piartz, flanked by crushed and silicified granite containing bands rich in sericite. Most of the thinner quartz veins are surrounded by flaky, dark, weathered rock. bericite and subordinate green biotite haA e been introduced in cracks through the original felspars in apparently massiA’e granite threaded by The Geology and Physiography of the Gosnells Area. 257 quartz voinlots. Oec-asionaliy the twin lamellae of the feis])ars are bent; otherwise they are not altered. Quartz, as it is not unduly strained, must have been introdneed after the stresses operated to deform the felspar. Other specimens of g'ranit(% which can he seen in hand-siiecimen to be altered, contain felspar in various stages of replacement by s('ridte. So rock types are known, in the waihrocks and in the tiuartz masses, whicli are transitional between granite and pure massive (]uartz. Jt has been held that the (piartz “l>low’' is merely recrystaliised (|uartz- ite xenoiithic in the granite. But there is no trace of bedding in the “blow"’ (except possibly the sericitic bands found in the Mountain Quarry). It is unlikely that such a large s(ade structure would be obliterated elsewhere. Recrystallisation, too, shonhl j)roduce a ma^s of even-grained (juartz and muscovite, ajid no stockworks of (]Uartz threading the surrounding granite. Thin sections from the ^‘blow’’ and from narrow (juai'tz veins resemble one another. Both contain e\idence of replacement, and because so many transitional rocks are found, it is concluded that all these masses (including the wilaceinent of original and secondary minerals by (piartz. Piirnival concludes ([). 859) that the hydrothermal solutions passed along faults of great persi.stence and disi)lacement along AA^hich the ro('ks were severely fractured over Avidths of up to 1,000 feet. Similar faulting in the massive granite near Gosnells, although it cannot yet be proved, is luwertheiess possil)le. The (piartz Avas i)robabIy derived from ■end-stage solutions from the granite, as no other acid igneous rock has been found in the district. The long (piartz vein north of stream A contains small scattered crystals ■of pyrite Avhich weather easily to make the I'ock appear \-ughy and to stain it reddish-brown, gi'een and yelloAV. Iron oxide has been deposited ])y ])erco~ lating Avuiter in many ])lae(\s in the adit, AA’hieh cuts through the AX'in 90 feet beloAV the surface. Veins 1 (in. wide, rich in blu(‘-grey tourmaline, cut the (juartz of the White Rock and Mountain Quarries. The quartz of these veins is eolourh'ss, limpid and coarse-grained (the grain size is up to 1 cm.) Imt excessively strained. The tourmaline is in the form of wisjis or rods, aggregating to an irregular Hlirous mass Avith pleochroic sclieme X — deej) blue, Z = jmle broAvn. The r('fracti\’e index e = 1.628 ± 0.003 indicating a tourmaline about midway b('tA\T*en (Iravite and schorlite (Winchell, 1!)27, )>. 246). Simpson (1931, p. 141) describes the oecuiTcnce of dravite in many places near Perth, and mentions that “In addition scliorl has Ixnm det(‘cted in small quantities at Gosnells and Cardup.” D. Cardiip Serieft. 1. Arl-ose and Slate. Being intimately associated in the field, the two are discussed together. Arkosic hands and lenses in the slate contain pebbles of (piartz up to 5 cm. long, set in a mass of smaller grains of rounded quartz, more turbid felspar and (in Aveathered specimens) white kaoHuite (?). Some slaty mat(‘rial is found in the arkosic patches. 258 C. E. S. Davis. The ('onstitnonts present in seetions are siibangnlar (piartz (grain size 0.1 niin. to 2 nun.), subordinate fresh inieroeline and a little slightly seri- eitised plagioelast', and rare aggregates of a flaky mineral of low bire- fringence' which may be kaolijiite, all set in a sericite cement. In weathered spc'cimens, plagioclasc' and sericite ai’o absent, but the clay constitiu'iit (a turbid, earthy mass) is abundant. The slate is a fiiu*-grained finely bedded micaceous rock. Fresh speci- mens are grey, and slightly weathered ones stained pink. It consists of sericite ami oval sliaped quartz grains about 0.05 mm. long, and a few small, rod-lik(* idioblasts of toiu'malim' iji caimnis stages of develoimu'nt (the- largest noted was 0.2 mm. by 0.1 mm.). The pJeochroic scheme of tin- tour- maline is X — deep blue-green, Z — very ])ale blue. H('dding of the slate is conspicuous and marked by orientation of tho sericite and by iron-staining. Although hand specimens of weathered slate' and of "shaly bands” from near the edge of the quartz “blow” resemble each othei', thc'y art' ^’('ry diftei'ent in seetion. “Redding’” is irregular in the •‘shaly bands,” flu; Hericit(' is haphazardly oriented and not associated Avith rounch'd quartz grains, ami (juartz occurs only in c'oins conqmsc'd of inter- locking angular grains. 2. Sandstone. Rocks classed as “sandstones’* grade from sandy slates to porous- (juartzose grits reseml)ling the arkoses. Bedding, though clear in the finer grained rocrks, may not he noticed in the coarser ones. The dominant constituent is qiiartz, in rounded grains up to 5 mm. long with wavy extinction. Seri(dto, present in small quantities as a cement in all tli(‘ sandstones, is as common as (juartz in the sandy shales. Felspar' and tourmaline are rai'e, but rounded zircon ])lentiful. A f(;w ATins, consisting entirtdy of an interlocking mosaic of quartz,, thread the lower sandstone. Their grain size varies from about 0.1 to 0.5 mm. and in one specimen the grains of the vein (piartz Aveiv* not<‘d to be Im-ger than those of the surronnding sandstone. Rti-ain shadows may be noticed in the larg(‘r grains. Cherty lenses in the mi[ bni'itc, is pro])ably tlie source of the barite found in the sand- stone. ft is suggested that Ijarito-ricli solutions dissoU’ed some material out of* the sandstoiu', forming tin* vughs Avhicb they later iiicrusled AAuth quartz and barile; this suggostioii may be tested by (inding out (at Armadale and soutliAvards) wh(*ther Ihe vugby sandstone is always baritic. The (Geology and Physicgraphy of the Gosnells Area. 259- Metamorph ism, Metamorphk' eftects in the Gosnells sediments are very slight. Tonr~ maline has developed in the slates, and serieite may be recrystallised. In the sandstones, a little quartz rccrystallised at some stag’e as cherty lenses and as (juartz veins. But hematite has not been changed to magnetite, and no biotite has developed. Only slig’ht jointing has been produced. Argil- laceous sediments are very readih^ affectedi by lise of temperature, and the very slig’ht metamorphism of these rocks indicates that they were deposited after the granite Jmd cooled. E . Bas ic In trn si ves. 1. TJraJitised Quartz Dolerite (EpicUorlte) . Hand sneeimens of this rock are grey to greenish-black in colour and are line-grained near the edge of the dykes, but medium-grained near the centre. They consist of white felspar set in a darker matte of amjdiibole. Essential minerals ])resent in thin section are felspar, arnphibole, ei)idote and ilmenite. iVIost of the arnphibole is a pale variety consisting of llaky aggregates or irregular plates, and is ('vidently a secondaiy mineral. It is referred to as ‘Walite.’^ A litth; Y>Gniary brown arnphibole is present in most rocks, and bluish, more euhedral arnphibole is produced by recrystal- lisation of the uralit'*. Both these latter types of amj)hibole are referred to as “hornblende.^’ In similar basic dykes north of Gosnells (e.g., in the Lower Cluttering area (Miles, 1938, p. 29)), uralite is derived from ])yroxene. The author has seen a section of a dyke near Dailington in which pyroxene has partly altered to a ]>ale green uralitic aggregate which farther from the parent mineral, has recrystallised to a. blue-gTcen type of hornblende. Although no relii's of the primary, high-grad(‘ parent mineral of the uralite have been found in the Gosnells area, it is believed that this uralite, too, is derived from pyroxene. Holmes (1928, p. 92) defines “epidiorite” as “A doleritic or basaltic rock in which the augite has suffered alteration to hornblende so that the rock (mineralogically) approaches the composition of a diorite.’' The writer has added the word mineralogically as, chemically, such alteration does not produce a more acid rock. Basic dyk('s neai* Perth with pyroxene entirely replaced by horn])lende have been desciibed in ]>i-evious literature as “epidiorite.” This has, how- ever, become a fhdd- and a sack-name and moreover has often been altered to “diorite,” even in g(M)logica] pu])lieations. It is best, therefore, to give 1o the rocks a name indicating their genetic relationships, and they are here termed uralitised qmu-tz dolerites. ITalitisation must have been deuteric, for the Cardup sediments are so very low-grade that there could have been no regional metamorphism to effect the change. Fresh primary felspars are clouded and coloured brown, but this colouration is far less marked than it is in dolerites from Parlington and Lower Cluttering. MacGi-egor (1931) suggested that, were a basic intrusive- 260 C. E. S. Davis. heated at some period after its consolidation, minute traces of impurities, of which iron oxide is the commonest, would separate, resulting in o'rey- nnd brown-clouding, lie contends (p. 537) that: — “it seems necessary to i)rove that (any igneous rock with clouded fe]s[)ars) . . . can never have been subjected to regional- or contact- thermal metamorphism before the clouding can be regarded either as an original teahna' ot the felspar or as a deuteric effect that arose at a late stage in the consolidation period.” In the (rosnells area, there are no later inti’usioris to etfect contact nu'tamorphism and, as has just been shown, no regional metamori>hLsm can have taken i)lace. (bonding, Ihen, was producml either at a primary or at a deut(n*ic stage. He('rystalliseidote are scattered through the dykes exposed in 1he (marries in the (iosnells area. The dolerite surrounding a nodule, sphcrulitic in structure, was totally recrystallised to a non-porphyritic medium-grained (grain si/e 2 mm.) ophitic intergrowth of clear albite felsi)ar and pale green hornblende. The albit(‘ contains a few ])ale green inclusions and both albite and hornblende are v(u-y similar in this rock and in a fresh porphyritic epidiorite (4. below) from fartlmr south. The pleo- chroic scheme of the hornblende is X = pale yellow; \ — Z = pale green, and its refractive index ^ is 1.641, compared with 1.657 in primary brown hor-nblende from the Bickley Brook Reservoir Quarry. AND PtlYSLOCAlAPaY OF THE OOSNELLS AkeA 201 A little of this hoinbleiule has been converted to biotite. However, aiiothei- recrystallised specimen (about three feet from the vdgv of a dyke/ ■contains much biotit(‘ so that it resembles biotite e]>idiorites, described in the next section. A few V(‘ins of calcite, up to 3 cm. wide, occupy ])rominent joint cracks in the dykes. Some pyrite, too, has crystallised with the calcite in and near the veins. The rock surrounding- calcite veins is very low Ki’ade: it consists of pale green chlorite, (luartz and turbid patch(‘s of calcite. Occurrence of t;^dcm/.~The author has seen two loose spe<*ime 7 is of galena from Gosnelts, l)ut, in spite of a long- seai-ch, none itj situ. One, from a dump at the Blue Hock (tjuany, contains a mass of galena about 1 cm. in diameter, set in coarse, strained and slightly granulated milky (luart/. According to (luarrymen, it came from a mass of ([uartz adhering to tlu* granite at the dyke contact on the west wall of the north (|uarrv, but no galena-bearing quartz could be found in situ. The other specimen, from the White Hot k (^)uarry, was collected by S. E. Terrill early in 1935. Now that so much fuj-ther quarrying has betn'i earned out, it was inipossible to i*ecognise tin* dyke from which it came. His spe('imen is a hue-grained, rather schistose, dark rock, with fine dis- seminated grains of galena and pyrite. It eontains a Iilth‘ residual green hornblende and biotite, but the main ferro-magnesinn is ehlorite. Other minerals present are clear recrystallised albile, very subordinate epidote and a quartz vein containing some calcite. Qliis occurrence ot galena at. Gosnells, althongb on a veiy small scale, is interesting because the galena is d(>finitely associated with the uralitiscd quartz dolerite or one of its derivatives. It throws light on the genesis of the galona-sphalerite-(iuartz veins of Armadale (Prider, 1941, [)? 51) and silver-lead deposits at Mundijong (Esson, 1927), and conhrms Vridcr’s sug- gestion that they are genetically related to the basic rather than to the granitic magma. 2. Bio t ite E puli o r Ue. A dyke, 19 feet wide, exposed only in the sonth(*rn coi-ner of the White Rock Quarry, is the sole rock found of this type. It contains many calcite veins and numei-ous small brown rounded biotitic patches. A section cut from the centre of the dyke shows that the }m\c liiue-green liornblende has been largely altered to brown biotite— the brown palelies consist of a decus- sate aggregate of biotite flakes, Recrystallised albit(‘ is common. Near its edges, the dyke consists of a line-grained mass of green biotite Avith sub- ordinate epidote and leiu^oxene, and clear felspar is rare. These rocks are similar to those described by Prider (1941, |>. 46) from Armadale which, being chemically very rich in potash and poor in lime, re- semble the chlorite-albite epidiorite and, probably, the porphyritic epidiorite (4. below). But the Gosnells rock seems to be a derivative of the iiralitised quartz dolerites, altered by end-})hase ])otassic solutions. A narrow epidotit' biotite epidiorite dyke (dve inclu^s wide) in the White Hock Quarry is Hnnly welded to tlie granite. It consists of epidote and (especially near its margin) of green-brown biotite. Microscopie veins of biotite and chlorite penetrate the granite for at least Vo cm. both between and through its constituent crystals, and cause a bending of the albite twin lamellae of the felspar. In composition, this dyke is intermediate between the biotite epidiorites and the biotite-epidote veins. j 331/42 262 C. Fj. S. Davjs 3. Biotite-Epidote These thin veins (maximum width two feet) are, unlike the dykes irreKu lar and not persistent. ’ The ^-eln material is aphanitic, green or black and flaky, evidently eon- taining abundant biotite, Indeed, it consists of green biotite (largely con- verted to chlorite) and very pale green epidote. Granite inclusions in the vein are cut by stringers of biotite and epidote. Albite twin lamellae of the felsiiar in the xeuolithic rock are bent and fractured, but thi' felspar is clear and nnsericiti.sed and may have recrystalliscd before the.se deforming move- ments took place. 4. Porpln/ritie Epidioritr. This type is found only near Ellis Brook, where it intrudes the Cardiip sediments. Porphyritii' epidiorites occur at various jilaces (Wongong, Car- diip, Muudijong) farther south, and a similar non-])orphvritic rock at Arma- dale (Prider, 1!I41, p. 48). It is a ilark, weathered rock made uji of laths of felsiiar, up to 5 mm. long, set in a line-grained ground. Prom a section, it is seen that felspar is developed as stout prisms (of all sizes from i /2 mm- to 5 mm.) which form a coarse network. The felspar is near albite, contains abundant ehloritic in- clusions and IS little altered except for slight kaolini.sation. The ferromag- uesiaii constituent is now green to brown biotite, forming decinssate aggre- pte.s, tilling the interstices between the network of albite prisms. Magnetite is fairly coiainoii. ^ Small iuiin^s of barite have been found near a pit in the porph^Titie epidiorite. The mineral is associated with veins of strained quartz, and is largely massive (g:rain size 0.3 mm.), but is also developed as tabular cry- stals up to 5 cm. long’. As this is the sole occurrence of barite in association with any igneous rock in the area, the porphyritic epidiorite is probably the source of the barite in the (^ai’dup sandstones at Gosnells. V. GEOLOGICAL HISTORY. There is no record in this area of any event prior to the emplacement of a granite batholith in middle Proterozoic time (Clarke, 1930, p. 160),. although a little to the south large masses of earlier, more basic rocks are preserved (Prider, 1941). These have been largely digested by the granite* at Gosnells, and only a very small (juantity of them is found as xenoliths.. End-liquids circulating in Joints tormed in the cooling* mass crystallised as dykes of pegmatite and aplite. Stresses at a later stage fractured the granite over a wide zone, and quartz was deposited from hydrothermal solu- tions over this zone and in many smaller fractures. There followed a period of erosion and deposition of a normal sequence of sediments (the Cardup series) which have undergone very little anamor- phism. Although only a small thickness of sediments is exj^tosed at Gosnells, - several hundred feet are exi>osed farther south, and a considerable thickness may underly the surface rubble even at Gosnells. All pre-existing rocks were invaded by basic dykes of (Clarke, 1930,. table p, 187) late pre-Cambrian age. These dykes now contain the lower- grade mineral hornblende instead of jiyroxene. In a porphyritic basic dyke,, a segregation of barite was formed. At the same time, probably, baritie solutions entered the more porous basal beds of the Cardup series. The Geology and Physiography of the Gosnells Area. 263 Since pre-Cambrian times, a great thickness of sediments has been laid down in the trough west of the Gosnells area. In the Tertiary (Woolnough, 1918) a laterite capping was develo]>ed over the peneplained surface of the pre-Cambriau rocks. The peneplain was uplifted (in af least two stages) and a small amount of subsequent erosion produced the present topography. VT. ACKNOWLEDGMENTS. The author wishes firstly to thank the Senate of the University of Western Australia, who granted him a Hackett Scholarship for the purpose of examining the area. He thanks the officers of the Lands and Surveys De- partment for making available data for the compilation of the map. To Pi’ofossor E. de C. Clarke and Dr. R. T. Pilder he expresses his deep grati- tude for many helpful suggestions in the course of the work, and for assist- ance in the revision of the text. VTT. BIBLIOGRAPHY. Balk, R., 1937: ''The Structural Interpretation of Igneous Rocks. Geol. Soe. Am. Memoir 6. Brown, H. Y. L,, 1873: "General Report on a Geological Exploration of that Portion of the Colony of Western Austriiliu lying Soiitliward of the Murchison River and Westward of Espcrance Paper No. 1, Azotes and Proceedings, Legislative Council, W.A., for 1873, pp. 1-20. Clarke, E. de C., 1930: "The Pre-Cambrian Succession in some parts of Western Australia." Mepori A.X.Z.A.A. Sc. for 1930, ])p. lo5-192: and Williams, P. A., 1920: " Tlie Geology and Physiography of ])aits of the Darling Range near Perth." Jour. Roy. Soc. W.A.y XIL, pp. 161-178, Daly, R. A., 1933: "Igneous Rocks and the Depths of the Earth." New York. Emmons, R. C., 1940: "The Contribution of Differential Pressures to Magmatic Differentiation." Amer. Jour. Sci., Vol. 238, pp. 1-21. Esson, A. D. G., 1927: "The Silver-lead Deposits at Mundijong. " Geol. Surv. W.A. Aun. Rept. for 1926, pp. 4-9. Eurnival, G. M., 1935: "The Large Quartz AYiiis of Great Bear Lake, Canada." Ec, Geol., XXXI., pp. 843-859. Holmes, A., 1928: "The Nomenclature of Petrology." Loudon. Honman, C. S., 1912: "The Extension of the Kelmscott Clay Deposit." Geol. Surv. W.A., Bull 48, pp. 63-65. Johainisen, A., 1932: "Petrography, Volume II." Chicago. Jutson, d. T., 1934: "The Physiography (Geomorplioiogy) of Westerji Aus- tralia." Geol. Surv. TV. A., Bull 95. MacGregor, A. G., 1931: '‘Clouded Felspars and Thermal Metamorphism." Min. Mag. XX IT., pp. 524-538. Miles, Keith R., 1938: "The Geology and Physiogra})hy of tlie Lower Chittering Area." Jour. Roy. Soc. W.A., XXTV., pp. 13-41. Montijomery, A., 1910: " Rei)ort on a Quartz Roof in the Darling Range near Gosnells." Ann. Rept. Depf. Mines for 1909, pp. 123-125. Prider, Rex T., 1941: "The Contact between the Granitic Rocks and the Cardup Series at Armadale." Jour. Roy. Soc. W.A., XXA’TI., pp. 27-55. Simpson, K. S. — 1910: "Analyses of Western Australian Rocks, Aleteorites and Xatural Waters." Geol. Surv. TV. A., Bull. 07. 1931: "('ontributions to the Mineralogy of Western Australia— Series VL (4) Dravite, Darling Ranges, S.W. Division." Jour. Roy. Soc WA XVI r., pp. 141-143. - • V Thomson, B. P., 1941; "The Geology and Physiography of tlie Wongon-g-Cardnp Area." Jour. Roy, Soc. W.A., XXVIT. pp. 26.‘)-28-3. c- 264 C. K. 8. Davis. Wincliell, A. N., 192/: “Elements of Optical Mineraloj^y/ ' Volume II., New I Ol’K. Wooliiougli, W. G. — 1918: "'The Darling Peneplain of Western Australia.’' Jour. TiOii. Soc N.S.W., Lll.y pp. 385-396. 1919: “The Physiographic Elements of the Swan Coastal Plain.” Jour, IRoy. Soc. W.A., \.y pp. 15-19. By Authority: Robert H Miller, Acting Government Printer, Perth. The Geology and Physiography of tbk VVongong-Cardup Auea. 265 19 -THE GEOLOGY AND PHYSIOGEAPHY OF THE WONGONG-CARDUP AREA By Brendan P. Thomson, B.Se. (Hons.). Read 10th June, 1911; Pul)lisibed 5th October, 1942. CONTENTS. I. Introdnction ... 11. Physiography : A. General Relief B. Drainage III. Structural Geology and Field Distribution of the Rocks ; A. Granite and Gneiss ... B. Cardup Series C. Basic Intrusives D. Later Rocks ... Page 265 266 266 268 269 270 270 IV. Petrography : A. Aplogranites, Gneisses and associated Basic Xenoliths B. Cardup Series C. Basic Intrusives of Post-Cardup Age D. Quartz Veins and Barite ... E. tjater Rocks ... 271 274 275 276 278 V. Conclusions : A. Age of the Cardup Scries and its relation to the Granitic Rocks 279 B. Darling Scarp 279 G. Comparison of the Wongong-Cardup, Armadale and Gosnells Areas 280 VI. Geological History of the Area ... ... ... ... ... ... 281 VII. Acknowledgments 282 VIII. List of References ... ... ... ... ... ... ... ... 282 I. INTRODrCTION. The Wongong'-Cardup area lies about 20 miles south-south-east of Perth. It extends for four and a half miles along the Darling Scarp, -which forms the west(>rn edge of the Darling Peneplain (Jutson, 1934, p. 81), from Cardup I>rook in the south, to half a mile north ol‘ Wongong Brook and it eovc'rs about six square miles. The rocks, bounded on the west by the sediment '.ry Cardup series, C)l bond in tho Wonj^'on^' at'tor it (nitors the aren suii'^^ests tlu' larger rpature observed in ihe (dinning- Kiver ((Tarke and Williams, 192(), ]>. Ui-t) and likewise* may lie a reiie- ot' an old nor( h-soiith drainage ehannel. This is suggested by tlie alignment of tribu(ari(‘s witii the* straight eastern side' of the bend and with the early-mature* valley of a sonth-llowing tributary ol^ the J^ee'iiyup Brook, a little over a. mile* to the south. West of tlii;. b(*mP the AVongong Hows for a (|narter of a mile in a vewy juveaiile vall/\ with walls rising steeply to 5t)t) feet above the stivam. Here the* trihutaries are she^rt, stee*p and inseepuait. n(.>wnstre*am the* brook swings we*st-north-west into a broaeler valley: at the bend, a sliort e-haiim*! has b(*e'n abandone'el. The course is genily winding and e'onlains a little alluvium; it ree-e'ive's se'veral tributarie's and then passes through the foothill zone onto the ]>laiu. 2. The BeeugHp and Cardop Brooks, — These* have yemng \'alleys whie'h lack the* gorge-likt* cliarae-ti't istics of the Wongong ami posse'ss st(‘e'pe*r gi-aeles. These dirrer<*nc(*s nmy be- (‘Xphnne'd by cemsidering the* ipipe-r porliem of the* Wongong, east of the ar(*a. This can be ivgareled, like* tin* upper llede'iia and ( anniug I\i\'ers, as another example* oi a e'onse'eiue'ut semth-(*asli*rly bowing stream wliiedi elevelojied afte*r the uplift of the Darling Pe*neplaiii, and w’Hs ('a])ture'd and revei’se'd by a more active* we'stward flowing e‘e)useiim‘nl stream (dutsoii, UIdd, p. Kid). Probably the (hre'e b]-oe)ks we*re* de'veloix'd ('■({ually as we'stward (lowing conse(]ue‘nts, Imt the* young Wongong, bee-ause^ it had a shorter distan<*(' io advaiu'e by lu'aehvard e're)sion, reaclu'd flu* semth- east stream. This le‘d to auginente‘d How ami de'e'peming e)f the* vall(*y. Mean- while the* Beenyup and (dardiip Brooks el(*veh)p(*d I libutarie's, hut la'taiiu'd re*lati\'(‘ly st(‘e]'> graeles. 3. Minor tribiifaries io the main stream comprise*: — (a) I nceu’tain i'(*li(-s of a mature norlh-se>utli system. (b) Shen-t st(*(‘p-grad('d minor stivams whie-h ge-m-rally enter at I'ight angles to the* main str(*am. (c) Lai’ge*}* tributai’i(‘s which temd to be de*ve'lo|)ed m‘ar the scarp;. the* N'alle'vs are* ye)uthful, but some* bre)!ulen in tlie u]>j>er parts. Small wateidalls oi‘te‘n etccui* at an eIe*vation of dtIO to 450 f(*et. 4. Independent sear}) streams. — The*se* have* steep graele*s and He>w west. Ttxamph's are* the* twe) creeks iu)i-th e>f tlie* Byfeerel Brickworks ((luarry. 5. Adjastment of fo})ogra})hif to geologip — Adjustment eif the drainage to the major i-ock sti-ucture* is not ap|)ar(*nl,. The* course's of the* thre*e main brooks form arcs, slightly e-oncave^ lei the north which cut- ae-ross the* strike* of the gneiss and the* s(*dim(*n(s. Two tributarie's of tho Beenyup Bre>ok appear to have* be-en iiiHue'iice'd iiy (piartz blows trending north-W(*st, Between the Wongong and Beenyup Broeiks, stre'ams have de‘ei)ly eunbayed the* late'i’ite capping by he*aelward erosion, le*a\'ing a iiumbe*!- of largev spurs, some* ove'rlying large e])idie>rite dyke's. Thc'se* spurs represe'iit tlie* most westerly extension of (he high leve*] latei'ite. The sharp edge* of the latei-ite disa]>pears where the* matui’e lu'arls of the creeks approach it and laterite pe'bbles and sand slo])o gently u)) to the laterite* love*!. Tn such places soaks may de*veh)p, the^ water appare'iitly coming from under the> laterite cap|)ing. 238 JiRENDAN P, Thomson. TIL STRUCTURAL GEOT.OGY AND FIELD DISTRIBUTION OE THE ROCKS (seo Plate I.). A. Granite anei Gneising steeply Avest. LoAAT'r (1) Grits and Sandstone — In bands up to 18 imdies thick (some are baritic) Avhich allernat(* Avith Sandp Shale over several chains. Graded b('dding in the sandy bands indicates that the b(*ds are normal and not over-turned. Th(*se b(*ds lie uuconformnbly on the hybrinarry is found diitpitig w(*st at ()5°. In the Cardti]) (Quarry at least three epidioi’ite dykes intrude tlie slat(*K, which a])i)ear to dip almost vertically. The bedding of the slates is obscnnal hy dosely s])ac(‘d fracture ]>la.nes, cans<‘(] probably by the intrusion of the epidiorite dyk('s, because 5 chains west of the large dyke on the west side of the dardnp Quarry the dip of tin' slat(' rcdurns to tlie normal 60° west. I he wider ('xposnres of slate in the south<‘rn i)arts of the area may best be exi)lained as a more complete scHimmce of tin* Cardtip Series whicii has Ix'en protected from erosion by (piarl;; blows and basic rook. This epidiorite outcrops south of the area, at Whitby Fails and Mundijong (Esson, 1920). Xo reliable' field evidence can be obtained in the Wongong-Cardup area concerning tin* relationship between this porphyritic epidiorite and the uralitised (juartz dolerite. I). Later Bocl's. The later rocks are : — (1) Ijaterite.~Vh\s~ forms a ca])])ing over most of the area above BOO teet elevation. Baterite boulders and ])ebbles occur on the lower levels ;at the foot of flu* scarp. (2) Allwmal deposits . — These occur along the courses of Wongotig and Cardup Brooks. Silt and sand (-over much of the plain (sege pae 266) Iving west of the scar|>. (d) Talus slopes . — ’These are formed on the steeper sides of the valleys, and are composed of aplogranite, gneiss and dokwite. The Geology and Physiography of the \''’ongong-Cardup Area. 271 lY. PETROGRAPTIY. A. Aployranites, Gneisses and Associated Basic XenolitJis, Tho basic xenolitlis include hybridised varieties of the biotite-horn- blende hornfels found at Armadale (Pridcr, 1941) and hornblende schist xenoliths. 1. Biotite-hornhlende Koryifelses. — Xenoliths of these rocks are common in the hybrid g-neiss. They form rounded masses up to four feet long. In the granitic phases of the gneiss the xenoliths form narrow sheets intercalated * ‘ lit-par-lit ’ with line-grained granite. The rocks are dark grey to black in colour, and generally have a fine even-grained hornt'elsic texture. In thin section, clots of decussate greeTiish biotite are associated with saussuritised plagioclase. Subhedral hornblende may be present with synantectic reaction rims of biotite or chlorite. Quartz is rare as small clear interstitial areas. Apatite prisms and euhedral sphene, as crystals up to 1 mm. long, are chai-acteristic accessories together -with pyrite and magnetite and more rarely, pink euhedral zircon up to (Ko mm. long. The felspar crystals are crowded Avilh epidote grains and small clinozoisite prisms. Lamellar tAvinning is common and extinction angles indicate all)ite-()Iigoclase. Biotite varies from a brownish variety with X = pale yellow to colour* less ; Y = Z = yellow-brown ; X < Y = Z, to a greenisli chloritised variety with X = pale green-yellow ; Y = Z = yellow-brown ; X('gmatitic and contain mierocline crystals up to 2 inches long. Whetluw this pegmatitic phase is derived from the early granite magma that formed the hybjid gneisses or from the latter aplogranite magma, cannot at ]>resent be stated. Garnet-Muscovite yipUte. — A vein of this rock 13 inches wide occurs in the gneiss in the extreme east of the area near the Wongong Brook. The aplite is pale pink in colour and varies in grain from a fine-grained saccharoidal to a more pegmatitic phase with small mierocline ])henocrysts and muscovite plates up to 12 mm. aci’oss. The texture in thin section is allotriomorpliic. The minerals present are fresh albite, unstrained quartz allotriomorphs, perthitic mierocline, rare large muscovite flakes and garnets. The gaimets form small pale pink euhedra up to 0.5 mm. diam. Acces- sories are a little chlorite and epidote. Sonu'what similar garnet aplites occur at Jimperding (Prider, 1934, p. 10) and Malku]i (Cole and Gloe, 1940, p. .160), whei-e they are related to granite intrusions of post-gneiss age. The origin of this garnet-muscovite aplit<‘ is not certain, but Ix^ause of its remarkable freedom from alteration, it is most pi'obably a phase of the aplogranite. 6. Relation between the Aplogranite and Hybrid Gneiss. — In the M on- gong-Cai’du]t Area no evidence was found of two tyiies of gneiss striking discordantly to each other, as do the fine-banded and hybrieriod, represented by the aplogranite. In the Wongong- Cardup Area, tivo periods of granite intrusion are indicated by — (a) Angular xenoliths of gneiss in aplogranite, best ex]> 08 ed on the soiuh side of the Wongong Brook gorge. (b) The truncation of gneissic banding by masses of aplogranite. 274 i3KENDAN P. Thomson. B. Cardiip Series. The series may be subdivided into 1. Basal sandy beds. 2. Slates. 1. The basal sandy beds eomi)vise sandstone, epidotie f|uart.'dte, and interbedded sandy slates. (a) iS'awd.strtMe.— This may vary eonsiderably in grain size. The coarser types, Avliieh can be classed as grits, contain sub-rounded quartz grains up to la mm. long. The matrix is often argillaceous and the (piartz grains show impert'eet graded bedding. In finer-grained sandstone on the western side of the Brickworks Quarry ripple-marking apjjears to be present. \ughs, which may liave originated by the leaching out of calcite, are common in the grits. These vughs are lined with smalf quartz crystals, and some contain white or colourless barite. At Cardup a small amount of purple fluorite is associated with the barite, a little pyrite and chalcopyrite are also present. White calcite occurs in some unweatliered stiecimens. hi thin section: — The ty])ical sandstone contains rounded to sul)-angular quartz grains interlocking loosely in a ground mass of micro-crystalline quartz and serieite. The quartz shows undulose extinction. Rounded pink zireons occur sparsely, but are rarely absent. In one section several small idioblastic crystals of tourmaline were found. The baritic sandstone has a gi*anular structure Avhich is obscui’ed l)y a mosaic of later vein-(]uartz. Angular cloudy felspar fragments (mostly micioeline) are common. Barite forms irregular grains and strings. Calcite and roumled pink zircons are present. (b) hpidolie Qitartzite , — This rock has been shown by Prider (1941, p. 40) to be sandstone which has undergone contact metamorphism, due to the intrusion of epidiorite. This is supported south of AYongong Brook, where an outcrop of the quartzite grades into sandstone and is associated with fine- grained epidiorite. The quartzite varies from a grey-green even-grained rock with a con- choidal tracture to a banded rock in which sandy and cherty bands alternate. Some of the sandy bands show grade-! bedding. Pyrite and marcasite are present in small amounts. In thin section clear rounded quartz grains are seen to be rimmed with a plexus of small pale green needles of an optically negative biaxial mineral with oblique extinction (probably actinolite). The quartz is strained. Epidote is present as small interstitial grains or as yellow-green clots. The felspar foiuns slightly clouded subhedral crystals, either of microcline or plagioclase. Small rounded zircons are common. The cherty bands contain inicrocrystalline quartz which is streaked with nearly isotropic white material and is spotted with epidote grains, 2. Slates : — Black and white slates, similar to those found at Armadale, are exposed at the Brickw’orks Quarry and at Cardup. On the north bank of the Won- gong Brook, white and grey sandy slate outcrop, but no black slate is ex- posed. Phe Geology and Physiography of the W’ongong-C^ardxh^ Area (i) ^‘Black"' Slate. — This is a black to dark-grey coloured rock, in which ale yellow-green ; Y Z brown-green X < Y = Z. I ralite forms sparse patches Avith biotite wdiich is apjiarently derived from it. 3. Porphyntic chloritc-albite epidiorite. — Aliiieralogically, this rock corresponds to the albit(' epidiorite of Armadale ( Prider, 1941). Hand specimens show yelloAV subhedral albite phenocrysfs (up to o cm. long), in a dark tine-grained ground mass. In the (^ardup specimens the phenocrysts are grouped radially, forming ‘A’osettes” up to 5 cm. in diameter. In thin section, albite phenocrysts are set in a ground mass of albite laths, uralite, biotite and chlorite with a little interstitial quartz and leucoxene. The albite is cloudy and contains ei>idote. The phenocrysts are rimmed and penetrated by the chloritic ground mass. I'ralite is rare and is a soda-rich variety with X < Y Z, X — pale yelloAv; Y — pale olive-green; Z = bright green-blue, Z /\ c = 17°. In the ground mass are found aggregates of deep-gi’een chlorite or decussate brown l)iotite, each containing several euliedral epidote crystals. D. Quarts Veins and Barite. The (piartz veins may be subdivided into — 1. Quartz veins in the granite gneiss. 2. Quartz veins in the Cardup Series. The Geology and Pjiyskxjhai’hv of 'phe \\'on(U)N(j-Gakui f Area. 1. Quartz veins in the granite gneiss. Tiiei'e are two types: — (a.) The small north-south striking vein south of Heenyup Ilrook. The vein averages five feet in width and is iar is |)r('s('nt. The (u'vslals are pi'ol)ably of microcliiK', they are allot riomorphic and slightly rounears to be two parallel series of tjuariz blows, striking north and south in tlu' slatt's at (kirdup: (i) Tin* westerly series which is tlu* more strongly d(‘V(do)>(‘d. Tliese average 5 fetd in width and outertq) continuously for distances of more than 10 chains. A section of out* of the blows ol’ this serit's is to be semi in the east wall of tin* main Cardup (luarry and it apjiears to Ih‘ mad(‘ uj) of bullxuis quartz hmses up to 4 feet wide, dipping at 85° to tin* west. These hmses app(*ar to conform (in strike and dip) to tlu‘ slate which has Ixhui distortisl, so that the bedding now 1‘ollows around the (“dge of th<* (luartz hmses. (ii) Th(‘ easterly series which outcrops in a north striking zon(‘ s(»m<^ 5 or () chains to llu* east ol' (i). 44iese veins ari' lenticular and tiie lenses are more elongated than in (i). They conform to tlu* dip and strike of the slate. Under tlu* uiicroscop**, tlu* (piartz in (i) and (ii) is pi'ai-tically identical as regards intensity of shearing. Weathered outcro[)s may lu* massiva* or ■show a pseudo gneissic structure caust'd by line IVrruginous str(*aks. This structun* is c«l by later (piartz \'eiidets, up to oiu* inch wide, in which vughs ’Occur. The (piart/ is int(‘ns(*ly slu'ared and !arg(* crystals (including those of' lat(‘r v(*inh*ts) show slicing and giauiulation. Thin shreds of gr(*(*nish ph'ochroic chlorite (?) arc; arranged ])arallel to the lines of granulation. Twisted i>lates of muscovit(* occur s})ars('ly and iron ore in subhedral grains is common. 278 I^liekdan P. Thomson. (b) “(’oiitnct’’ Quartz Veins. These occur at several places on the gTanit(‘-»'iieiss-Car(lui> Series contact. One is well develojjed south of the Brickworks Quarry aiul it closely resembles the noi'th-soutli type of blow of Ai'inadale. It contains sericite bands and the quartz shows cataclastic structures. (e) Vein ([iiartz in the baritic sandstone. “The (luartz-barite veiiP^' (ClarkOj 19.30, ]). KiO) at Cardiip is sandstone which is veined and i)artly replaced by later quartz and barite. In lliin section the Aadnlets are setm to consist of clear quartz forming a inozaic of slightly straiiUHl crystals. (d) Narrow \(‘ins of vitreons quartz, up to one inch wide, occur in the black slate of the Brickworks QuariA'. The ([uartz is limpid and shows undulose extimdion. Origin of the quartz veins in the Cardup Series : — In the light of other eviflence, indicating that the Cardup Series is ]>ost granite in age, it is sugge.st(‘d that these veins are the residt ot' a period of vein formation that followtal tlu' intrusion of the epidiorite. The veins are possibly the' result, of difterentiat ion of the basii- magma. Coulson (1983, ]). 114) suggests sudi an onlgin for the quartz reefs associat(Ml with f'ertain Indian barite dejiosits. Although the age relation- ship betwetm the epidiorite and (fuartz veins in the Cardup Series has not. been determined \u the Wongong-Cardup area, post-e])idiorite quartz veins, are known to oc('ur at localitic's iu*ar the arc^a. Prider (1911, p. 42) ineiitions tliat at Armadale, in the tunnel west of the slate quarry, vein-lik(‘ bodies of ((iiartz that show jmouounced cataclastic structures occur in uralitiseossil)Iy the calcite, apjiear to be the result of one period of mineralisation. Because of tlu* irregular distribution and the abs(*nce of banded veins,, the paragenesis cannot be worked out. Nevertheless, sucli an association of minerals indicates a hydrothermal origin for the barite (Laurence, 1939, l>p. 190-290). Prider (1941) suggested that the silv<‘r-lead bearing veins of Mundijong and the barite and (luorite are l)oth r(*la(('d to the albite epidiorite magma. Coulson (1933, p. 115) considers that the barite of certain Indian deposits are derived from a somewhat simitar basic magma. AVhatever their source, the hot bai’ytic solutions appear to have ascended into the jAorous basal sandy beds of the Cardup Series iii many ])laces. Solutions containing barium ions have also migrated along the fracture planes- of the slates, where barite has l)een deposited (]">. 275). E. Later Bocks. 1. Lateri{e.~V\\Q. distribution of this rock has been desciibed (p. 270). The high level laterite is the common pisolitic variety, and it does not appear to vary mai'kedly in composition, whether developed over epidiorite dykes or granite gneiss. The (ih:oLOGY and Physiochiaphy of the W'^ngong-Cardup Area. 279 On the .slo})es below the Interiie, frag'nicnts of a dark-red fine-grained limonite rock are found in a few restricted areas. Thin section lana^aJs a fine-grained inati'ix of limonite with a small amount of interstitial (jiiartz. This rock may be e(iuivalent to the ferruginous grit found below the laterite of other Darling Range areas (Miles, 1938, p. 32). On the lower levels laterite occurs as boulders of the ]>isolitic variety or as unconsolidated deposits of limonite pebbles with sand or ferruginous clay. 2. AUtivi'um . — The alhnium in the Wongong Brook is sandy. At the bend west of the gorge, large l)oulders of conglomeratic rock lie in the aban- doned stream bed. This rock contains pebbles and boulders of granite and gneiss Avhich have been ceuumled by clay, in the banks of the stream. V. CONCLUSIONS. A. Af/e of the Cardup Series wfid its relation to the (jranifie rock's. The structure of tlie gtUMss can be broadly interprded as due to folding which took j)laee before Ihe formation of tlu* Cardu[> Series. Further facts similar to those recorded by Prider (1941) support the conclusion that Ihe gneiss and granile are older lhan the Cardnp Series. They are (i) No (|uart-/ or pegmalitt* veins have been obsen'ed to pass from the granile and gneiss into the (hualup Series. (ii) The Cardup Series maintains a normal (U’osion sequence and the basal lieds must rest iqjon the eroded grnnit(^ gneisses Avhich strike obliquely to th(^ Cardup Series. (iii) Xenoliths of the sediments are alisent from the gneiss. (iv) The gneiss does not vary in character across the strike and therefore does not appear to be a granitised ])hase of the Cardup Series. (v) Contact metamorjihic effects in the sedinunts are slight and are due to basic intrusions. (\i) The wides]Ar(‘ad occurrence of small idioblastic crystals of tour- maline in t he slates can be readily attributcal to original Iioron in the sediments rather than to pneumatolytic elfects of a granite intrusion. (vii) Hounded zircons are found in the basal sandy beds of the sedi- ments, whereas mim‘ralogi('alIy similar zircons occurring in the gneiss and basic xenoliths are unrounded. 'This points to the possible derivation of the Cardup Series (in part at least) from the gneiss. (viii) Detrital microcline, most reasonably regardt'd as derived from the aplogranites and gneisses, occurs in the Cardup sediments. B. Dorlhig Scar}). Prider, (1941), considered that ihe Darling Scarp was produced by the differential erosion ol‘ a monoclinal fold, and Avas not luamarily a fault structure as has been suggested by most geologist-s (Jutsoii. 1934, p. 87). He based his conclusions on the minor structures in the slates at Armadale, AA'hich indicate an upflirnst from the Avest, the result of a dowinvarp of sediments to the west of the area. In the Wongong-Cardu]> area, similar structures are found in the slates of the State Brickworks Quarry. It is true that at Cardup, slate further Avest of the granite gneiss Cardup Series contact dips almost vertically Avhieh is hard to reconcile Avith the u])thrust, but this steepening of dip may be attril)uted to the deformation produced by the nearby basic dykes. 280 Bkendan P. Thomson. The ^Toothill zone’' (p. 266), which is partly underlain by the Cardiip Series, can be attributed to the effect of differential ei’osion between the sedi- ments and the granite gneiss. C. Comparison of the Wongonu-Cardap, Armadale and Gosnchs Areas." All three areas are situated on the Darling Scarp, on an igneous meta- morphic complex, flanked on the west by the Cardup sedimentary series which strikes almost due north and dips steei)ly to the Avest. Fig. 1. There are as yet insuflicient data to give a complete structural pi{‘ture of the igneous-metainorphic complex. The knoA\m facts from the three areas examined (which are illustrated in 6g\ 1) art': — (1) The. predominant I'ock in the areas examined is a hybridised gneiss which is best developed in the Wongong-Cardup and Ai’inadale areas. In the Wongong-Cardu]) area the gneiss is consitlerably folded and the strike varies from N.W. to N.N.F. while in the Armadale area the general strike is almost due north. At Gosnells the granitic rocks are generally nia.ssive without any gneis.sie structures. Post-gneiss massive granites occur at both Armadale and Wongong-Cardup Avhere they form only a minor ])art of the complex — these granitic rocks (aplogranites) may possibly be offshoots from the main Darling Range massive granite which lies to the north of Gosnells. *This section was written in collaboration with Dr. R. T. Prider and Mr. C. E. S Davis. The (iEOLouv and PHVsrooKAPHY OF THE Wonoono-C'ahdx’P Area. 281 (2) Quartz x’cins aio tlev’eloped in all three areas, tlie trend being N.N.W. at Cared in all thr(‘(‘ areas and ihe succession, lithology, grade of metamorixhism and relation of the Series to the granitic rocks and to the lat('r basic intrusives are similar throughout . Post-(’ardup basic intrusives belong to two main types, (i) albite epidiorite and (ii) iiralitised (juartz dolerites, both of which are developed in all three areas. They haA'e been assigrual to two ages — the earlier albite epidiorites having accommodated thoraselves to the earth movements affecting the Cardu}» Series, while the later (piartz dolerites have been comparatively little altered. Although the .signihcance of the various structural [>atterns of these three areas (as seen in the general trend of the gneissic struclui’es, (piartz veins and basic, dykes) is not as yet completely understood, it is the authors* opinion that the Barling Range complex consists of an older hybridised granitic gneiss intruded by a later batholithie granite extending over the area behveen Gosnells (on the south) and Lower Cluttering (on the north). Offshoots (aplogi-anite) from this granite mass have penetrated the older hybridised gneisses lying to the south of Gosnells. The Cardup sedimentary Series is definitely of later age than the granitic rocks and the various basic intrusions are younger still. VI. GEOLOGICAL HISTORY OF THE AREA. (1) Period of granitisation, during whieh the hybrid gneisses were formed from older basic rocks. (2) Folding of the hybrid gneiss (by pressures directed from the N.N.W. and S.S.F.), probably closely followed by, or contemporaneous with, (3). (3) Aplogranite intrusions. (4) Earth movements, forming shear zones and joints in the gneiss, lollowed by the formation of quartz veins, whieh may represent the end- phase of the aplogranite magma. (5) Erosion of the granite and gneiss and deposition of the Cardup Series. (6) DoAvnwarp to the west of the area, causing upthrust of the Cardup Series toxvards the east over the granite gneiss. (7) Intrusion of the epidiorites. j 3.30/42 282 BitKNDAN V. Thomson. (8) Fornialion of ]iost-rardiip (juart/ veins and the introduetion of ])arytic solutions. Both of these are the end-phases of the (‘pidiorites. Fnd of the Bre-(,^ainbvian. (0) Deposition of sedinnoits \V(‘st of llie area, leading to further dowinvarjjing whieh may have pi'odnced strain etfeets in tin* (puirt/ veins in tlu‘ slates, (Id) Continued erosion leadin^^ to i>eneplanation and laterite for- mation {Wo(tlnon^h, 1018, p. 385). (11) Uplift of the penejdain in late Mioeene times (Jiitson, 1034, ])a”:c 205). and pi'ohable formation of the Darlinj^f Fault. (12) Further erp. 283-305, and pp. 353-375. I'larke, K. de and Williains, F. A., 192(i: ^‘Tlie (Jcology and Physiograi)liy of Parts of tlie Darling Range near l*erih.'^ Jo^ir. *S'oc. JV.A.j Vol XIT., j)p. 101 17S. Clarke, !'k do 1930: *^Tlie Pro-Cainbrian Snccession in some parts of Western Auslralia.'^ lirpf. Anst. and N.Z. Asmoc, Adv. Sri., 1930, pp. 155-192. (.'larke, F. de C., 193S: “Middle and West Australia.” HrgH>n. 114. Fsson, A. C. I)., 1922: “The Basic, Rocks of the Wongong Dam-Site.” CcoL Snrv. if\A. A/ru. h’cporl, 1922. Fsson, A. (i. 1)., 1920: “Tlie Silver-Tjead Dei)Osits at Alnndijong. Gcol, Surv. fl\A. Amt. Hcporl, 1920, ]»p. 4-9. Forman, F. (»., 1937: “A <5)nt ribntion to our Knowledge of the Pre-Cambrian Siuu'cssion in some Parts of Western Australia.” Jour. Fog. *Sec., IV. A. ^ Vnl. XXUl, p. xxiv. Ilonman, i'. S., Iitl2: “ TIte I'ixU'usion of the Kelms('ott Clay Deposit.” Geol. Siirv. iV.A. Gull. Xo. 48, p. 03. .tutson, .1. T., 1934: “The Ph vsiograjiliy ( ( ieoniorphologv) of Western Aus- tralia.” Geol. Surr. IV.A. Gull. 95. Laurence, It. A., 1939: “The Origin of th(‘ Sweetwater, Tennessee, Barite De])osits.” Ecou. Geol., Vol. XXXIV., pj). 190-200. Miles, K. R., 1938: “The Ceologv and Physiogra])hv of the Lower Cluttering Area/’ Jour. Fog. Sor. IV.A.', Vol. XX’lV, pp, 13-41. The Geology and Physiography or the Wongong-Cardup Area. 283 MacGregor, A. G., 1931: “(’louded Felspars and Tlierjiial Metamorphism, ^ ’ Min. Mafj., Vol. XXir., p. .VJ8. Prider, K. T., 1934: ‘‘The (''eologv and Physiography of the Jimperding Area.” Jour. Hoi). Hoc. fr./L. Vol. XX, pp. l-l'6. Prider, K. T,, 1941: “The Oardup Series — Granite Contact at Armadale.” Jour. Hoif, So?. W.J., Vol XXVII., i^p. 27-55. Woolnougli, W. G., 1918: “The Physiographic Elements of the Swan Coastal Plain.” Jour. Hoy. Hoc. W.A., Vol. V., pp. 15-19. By Authority: Robert H. Miller, Acting Government Printer, Perth, IX. CF.NKHAL INDKX. Aegintha temporalis cardwelli, »uh. sp. wov. Allotropiopliyllum A m moharulik''^ tronl mmtjh i Ammodiscus nitidus, j^p. nov „ wandageeensis, s]). nov. ... Amjthldscnpsi/^ InrsHhts ... Amphictene ? permiana« s]). nov. Ampiexu<'( pu.'^fulosuft Arnin(lal(\ (’!itjdaragan ,, iund_\sis of occurrtMKH' Birds, .\nslralian, tinw forms of ()c<'au, of IVrlh heaclu's Burhidgn, Nancy 4’. (\iU'itnnn‘Ha Mfephoisl Cardnp Scuacs • • .< ag(‘ of sandstotu's slat(*s Carroll, I'orothy ... Cervinipitta kimberleyensis, sp. nov ( 'ladoclinims • • HU'lloinOHl ,, striatus, sp. nov. ... (’lav .Minerals, Optical prop('rl i('s of ... Cole, \W 1C ('olloids. Soil. Microscopic stnd\’ of Colluricincla brunnea jiilietae, suh. sp. nov Contrilnit ions Clorac Anst?'aiia(' Occidcntaii.s XI. I •• .. .. .. ind(',x ... Cop<‘|)oda. Marine, from "W’.A. ... ... ... (Vti'als, ('iiailKixonia faumis in I’crmian, T\\A. " .. .. ,. list of •• .. ,. ,, sncccMHi(ui of .. Co.ssil from l.angicv I'ark Bore. IN'rlli lurtJu'r Bcrinian from \V..'\. Crithionina teicherti, sp. nov ('ynlhaxain'fi faanais 7‘k 01. 02 .. SS. SO 50 51 52 122 124 122 77 210 no I os 2S 270 2S 2S 7S OS (iO ... 70. 74 2 ... I. 220 I 77 100 200 cn r)7 .77 r»7 or> .".7 ... 107. I 12 57 Darlinjr ‘‘ Caidt Scarp Davis, C. 1C S. I )onny hrook Sand.stoncs, M iiua’alogy of Kdvvards, A. B>. I'judtiriyiapun an.sfra!.(,s ICiryphylfum ri'.idi Kvans, .1. \V. 245 211 70 ... 12.5. 120 01 01, 72, 174 I 12 lOivosito.s Flora, 7io\v species dcstn'ilKsl Foraininifera, IVrinian. N.W. Division* f W.A. 01, 75 I'-O 07 X. General Index. Gardner, C. A. Geology and IMiysiography of the Gosnells Area „ „ „ „ 'Wongong-Cardup Area Gingin Upper Greensand, Microscopic Examination of ,, „ ,, Analysis of Glauert, L. Glomospira adhacrens, sp. nov. Gosnells Area, Geological History „ „ Geology and l^hysiography „ IVtrology ,, ,, Structure and Distribution of Rocks Granitic rocks at Armadale i*age 109 245 205 229 229 219 103, 112 202 245 252 248 27 Hill, Dorothy ... ... ... ... ... ... ... ... ... ... 57 Homoptera. dassoidea from AY. A. ... ... ... ... ... ... ... 143 51 new species and eight new genera described 143 Hyperammina coleyi, s]). nov 104, 114 „ (?) rudis, sp, nov 105, 112 Hyperamminoides acicuia, sp. nov 105, 112, 114 Kyanitc, lu^aring rocks ... ,, in crystalline schists ... „ in W.A., field occurrence „ „ properties of ,, localities found in 5V.A. 9 9 10 10 10 Le iMesuri(;i’, ('. H. ... ... ... ... ... ... ... ... ... 133 Jveaf Hopptu’s {Homoptera, Jassolde.a) from W.A. ... ... ... ... ... 143 ,, ,, 51 new species and eight new genera described 143 Macnuuatcs giganieus ... ... ... ... ... ... ... ... ... 225 Marini' (oju'poda iVom W.A. ... ... ... ... ... ... ... ... 135 Mathews, Gregory ... ... ... ... ... ... ... ... ... 77 Meliphaga virescens giauerti, sul». sj). nov 77 „ „ lipferti, sul>. sp. nov. ... ... ... ... ... ... 77 ,, ,, lewisi ... ... ... ... ... ... ... ... 77 jov.s.seaumei ... ... ... ... ... ... ... ... ... 138 iMicroscopic E.xandnation of Gingin ri)per Greensand ... ... ... ... 238 ,, Study oi Soil (’olloids ... ... ... ... ... ... ... 1 Miles, K. H. ‘ 9 Mineralogy id' Donny brook Sandstones ... ... ... ... ... ... 211 Nicholls, A. (j. ... ... ... ... ... ... ... ... ... ... 135 Nortii Eastern Goldfields, Notes on Vegetation ... ... ... ... ... 119 ,, ,, species vegetation collected ... ... ... ... 128 Ocean Birds of Berth Beaches ... ... ... ... ... ... ... -19 Oculina ... ... ... ... ... ... ... ... ... ... ... 90 Orthopsyllus littoralis, sp. nov. 130. 141 „ similis, sp. nov 137. 141 Para ppJHdiuv) rri, datum ... ... ... ... ... ... ... ••• ••• PaiT, W. .1. 97 Permian (brals from W.A., Further ... ... ... ... ... ... ^7 ,, Eoraminifera, N.W. Division of W.A. ... ... ... ... 97 „ ,, stratigraphical sequence ... ... ... ... ••• 97 IVrtfi Beaches. Ocean Birds of ... ... ... ... ... ... -19 Phyllothalestris lata, sp. nov 135. 139 Plerophvlhim ... ... ... ... ... ... ... ... ... ... 02 Prider. '\i. T :^7 Psammosphaera pusilla, sp. nov. lOh, 112 Reophax subasper, sp. nov. „ tricameratus, sp. nov. Sandstuni's, Donny brook. Mineralogy of Sericornis condeni, sub. sj). nov. .. nKf.rafaJa ,, niondraini, sub. sp. nov. 108, 112 109. 112 211 78 77 78 GENERATi Index. XI. Soil Colloids, aggregation of >» composition of some >» kaolinite ” ■> mica ... ?? microscopic study of 5, montmorillonite ... j» o])tical pro^ierties of Htrychno^i hu-ida, alkaloid content of seeds ... Thamno|)ora .. immensa Mn.iiruoneiisis off nuirrnlonensis ... ■j insculpta, sp. nov. Thom])son. 1C 1*. Thuramniina pnpiJlata Toiypammina undulata, sp. nov. ... T re.inatotrofliu.fi iaterophnns Trochammina subobtusa, sp. nov. "* fufucolous "Worm from Permian of N.'\V. Division of W. A. Wgctation. North-Eastern Goldfields, Notes on . ,, •' species collected \ erl)eekieJla ,, mersa, sp. no^’. ,, talhoti X-ray Analysis of Gingin L ])per Greensand data of soil colloids ’\\hitson, E. \\ ongong-Cardii]) Area, Geology and Physiography of ” ” ■’ ••• Comjjarison \vith Armadale and Gosnells ■’ ” Geological history ” ” ” l^etrograpliy ... - . •* , •’ Structure and distrihution of rocks -v onn, tu;/iruIo.!s, from Permian >f NAV. Divisions of W. A. phivaitimorjjhs ... Areas Page 1 4 3 117 65 65 , 72 , 74 66 ... 66 , 74 67 , 72 , 74 265 ... 107 , 112 ... 104 , 114 95 ... 109 , 112 97 , 109 , 114 119 129 63 63 , 72 , 74 ... 64 , 72 229 233 117 265 279 280 281 271 268 ... 97 , 109 60 liy Authority: Robert H. Miller, Government Printer, Perth. W to 05 o / /FN ^ — [ & ii!’*. •:, '''I''' 'i ! ! I ■‘.''-LV r^J^; !. I '• ,11 1 *i. f,'.* f* ,i'-!'i''i .'i'- '■ ' ■: I -'-■ •■ :r,' t' '■ -i' ' ■'• ;. r ,.;,L i;;j^ t) .i^t: ^ 1 ■ i' iix ^ ir ' ' - ; i ■-^.. -u. '.v ■ ■ • ; r 5 ^ ^ :J A ^ J