VOL. 72—1949 TRANSACTIONS OF THE ROYAL SOCIETY OF SOUTH AUSTRALIA INCORPORATED ADELAIDE PUBLISHED AND SOLD AT THE SOCIETY’S ROOMS KINTORE AVENUE, ADELAIDE Registered at the General Post Office, Adelaide, for transmission by post as a periodical ROYAL SOCIETY OF SOUTH AUSTRALIA (INCORPORATED) OFFICERS FOR 1948-49 Patron HIS EXCELLENCY LIEUTENANT-GENERAL SIR WILLOUGHBY NORRIE, K.C.M.G., C.B., D.S.O., M.C, President N. B. TINDALE, B.Sc. Vice-Presidents D. C. SWAN, M.Sc. A. W. KLEEMAN, M.Sc. Hon. Secretary Hon. Treasurer J. M. THOMAS, M.Sc. H. M. HALE Hon. Editor I. G. SYMONS Hon. Librarian D. B. ADAM, B.Agr.Sc. Members of Council R. S. BURDON, D.Sc. B. C. COTTON S. B. DICKINSON, MSc. T. J. MARSHALL, M.Agr.Sci., Ph.D. A. G. STRICKLAND, M.Agr.Sci. J. K. TAYLOR, B.A., MSc. Hon. Auditors F. M. ANGEL N. S. ANGEL lt fn ee Efe we > RAP AG ee ; Bp. nn — : - flee ff oer elge AS _~ Earpiey, C. M.: SimpsonDesert- Expedition 1939=-Scientific Reports No. 7—Botany a Pt I. The Phytogeography of Some Important Sandridge Deserts compared with ’ oy _~_* -that-of the Simpson Desert .. a ens io os Hedin i ae tenigs h | Corton, B, C.: Southern Australian Gastropoda, Parf IIT .. .. .. -.. ..- 30 —.. Jessup, R. W.; A> Vegetation and Pasture Survey of Counties Eyre, Burra and : Kimberley, South Australia ., re ay oe . ss > = Ae oe 6] Jounston, T. H, and Esowps, S. J,: Australian Acanthocephala No.7... 69 - Jounston, T. H., and Crarx, H, G. : Cestodes from Australian Birds, I. Pelicans © .. 77 WomersLey, H.: The Genus Tragardhula Berlese 1912 (Acarina Trombiculidae) a ee} Srecu, R, L., and Perry, R. A.; Plant Ecology of Part of the Mount Lofty Ranges (1) 91 Cresrin, I: Indo-Pacific Influences in Avstralian Tertiary Foraminifera] Assemblages (1) rae Cr f; me * - i: a “i Pe if rw Womerstey, H. B. S.: The Marine ‘Algae of Kangaroo Island. II. The Pennington Bay Region ., es .. Re SF + f es - es s. 143 Epmonps, S. J.: The Commoner Species of Animals and their Distribution on an Intertidal Platform at Pennington Bay, Kangaroo Tsland, South Australia .. 167 : Fr - : “PART. IL Witson, A. F.: The Charnockitic and Associated Rocks of North-Western South Aus- tralia. II, Dolerites from the Musgrave and Everard Ranges oe “x eae 4-1 Hossrevp, P. S.: The Stratigraphy of the Aitape Skull and its Significance... 201 Lovertpcr, A.: On Some Reptiles and Amphibians from the Northern Territory .. 208 Boomsma, C, D.: The Ecology of the Western Clare Hills, South Australia ., + 216 Boomsma, C. D.: Nomenclaturé of Eucalypts, with Special Reference to Taxonomic Problems tin. South Australia , Fs ES ts ta = i prt Zal Warts, A. W.: The Geology of the Boolcoomata Granite .. at 36 + +e 228 Mawson, D.: Sturtian Tillite of Motint Jacob and Mount Warren Hastings, North Flinders Ranges 2 A es Py x — * so ia wo 244 Hossretp, P. S.: The Significance of the Occurretice of Fossil Fruits in the Barossa Senkungsfeld, South Australia - zi am a a v 3 ~ er 2 Lancrorb-Smiru, T.: The Geomorphology of Courity Victoria, South Australia ere Mawson, D. and Sxentr, E, R.: Purple Slates of the Adelaide System .. op Peed, BALANCE SHEET Si 281 List or Fetrows = = ia - = uF ha sf Ms ea 23) gee 286 InDEX [the —_~ VOL. 72 PART 1 : AUGUST 23 1948 aN he ih s/ ; © TRANSACTIONS OF THE ROYAL SOCIETY OF SOUTH AUSTRALIA INCORPORATED ADELAIDE PUBLISHED AND SOLD AT-THE SOCIETY’S ROOMS KINTORE AVENUE, ADELAIDE Price - Si Fifteen Shillings Registered at the General Post Office, Adelaide, for transmission by post as a periodical THE SIMPSON DESERT EXPEDITION, 1939 —- SCIENTIFIC REPORTS NO. 7 - BOTANY PT. THE PHYTOGEOGRAPHY OF SOME IMPORTANT SANDRIDGE DESERTS COMPARED WITH THAT OF THE SIMPSON DESERT By C. M. EARDLEY Summary In the course of determining the plants collected on the Simpson Desert Expedition, 1939 (Eardley (36) ), considerable interest was aroused in the question of how the Simpson Desert compared in aridity with other great deserts of the world. 1 THE SIMPSON DESERT EXPEDITION, 1939 — SCIENTIFIC REPORTS No, 7 — BOTANY PT, I. THE PHYTOGEOGRAPHY OF SOME IMPORTANT SANDRIDGE DESERTS COMPARED WITH THAT OF THE SIMPSON DESERT By C, M, Earpiey * [Read 8 April 1948] With Map CONTENTS Introduction . aides vt ef 1 (A) Libyan Desert (North Africa) : Ls Lash - 4 (B) Takla Makan (Central Asia) .... te a sles val en my (C) Rub‘ al Khali (Southern Arabia) 8 (D) Western Sahara (North Africa) . - nu tt (E) Simpson Desert (Australia) th ile eet as nt ifs ey Se, (F) Kara Kum (Western Asia) tS. hen wink . _ _ wee 15 (G) Chihuahua (North America)... ee cs dea tn, vee a. 20 (H) Thar Desert (North-west India) at ae hae vr cmt a 21 Summary ,,,, rey Pre us sess wt we _ ten sai! a 24 Acknowledgment . _ a3 ah ete eet vag epee hs ain BS Appendix—Comparative Tables T-TV iit ih, in al Lis a+ 26-29 INTRODUCTION In the course of determining the plants collected on the Simpson Desert Expedition, 1939 (Eardley (36) ), considerable interest was aroused in the ques: tion of how the Simpson Desert compared in aridity with other preat deserts of the world. Information was now available on the vegetative cover of the Simpson Desert, so a search was made for similar litcrature ahout seven other deserts which con- tained parallel sandridges, As far as possible, meteorological data and relevant general descriptions were also sought, but the emphasis was on the vegetation, and the results are now presented in this study. It is felt that a useful measure of comparative aridity has been obtained, (A) LIBYAN DESERT The eastern half of the Sahara is now generally referred to as the Libyan Desert. Gatttier (5) stresses the difference between the Occidental and the Oriental Sahara, saying that in essence it is due to “the extreme aridity of the Libyan Desert, incomparably greater than that of the Occidental desert’? (p. 105, op. cit,). * Herbarium of the University of Adelaide. Authorities for extra-Australian plant names have only been given in this paper when supplied in the source yuuted., Traus. Roy, Soc, S. Aust., 72, (1), 23rd August, 1948 rept ef Capricorn SAND-RIDGE DESERTS 1. Libyan Erg 3. Rub! al Khali 5. Simpson Desert 7. Chihuahua 2. ‘Pakiu Makan 4. WesternSahara 6, Kara Kum and Kisil] Kum 8. Thar Deseri The Libyan Desert consists mainly of flat expanses covered with a stony surface or a sand sheet; it also includes areas of various sizes where sand- dunes occur. These dunes are distinguished by their tendency to form long parallel lines whose regional orientation depends upon the direction of the prevailing winds in the Jocality, It is these dume-areas which are of sa much interest to the present study. By far the largest of them is the Great Libyan Erg which Gautier considers “probably the most imposing mass of dunes on the whole surface of the earth” (p. 103. op. cit.), with nothing comparable to it in the Oceidental Sahara. The smaller dimne-areas will here be neglected in order to concentrate attention on the Great Libyan Erg, which Rohlfs (8) called the “Great Sand Sea.” Knowledge of this region, even of its extent, has been almost entirely lacking until about 1930, It is the most inhospitable part of the barren I ibyan Desert. A summary of recent explorations in the Libyan Desert is given in Bagnold’s popular book, “Libyan Sands” (2). His own desert journeys by motor car were most extensive and included excursions into the Great Sand Sea im 1929 and 1930 (1). The only published account then available was that of Rohlis’ historic journey of 1874 (8). The Great Sand Sea lies against the western border of Egypt. Jt forms an area roughly rectangular and about 300 miles long by 100 miles wide; the long axis lies N.N.W. by S.S.E., and this is also the direction of alignment of the long parallel dunes (just as in the Simpson Desert, as it happens), There is a narrow plateau between the dunes and the coast in the north, and the oasis af Siwa is situated at the base of the southern escarpment of this plateau, close to the northern dunes. Another formidable barrier, the eastern escarpment of the ereat Gilf Kebir Plateau, lies to the south of the Sand Sea; it is therefore very difficult to leave Egypt by the western frontier except in the extreme north and perhaps the extreme south. Among the names famous in the exploration of the Libyan Desert, only a very few can add to our ktiowledge of the interior of the Sand Sea, These are Ruhlis (8), de Lancey-Forth (4), Bagnold (1 and 2), Wingate (11), P. A, Clayton and Lady Clayton East Clayton (about 1933). 3 Rotilfs attempted to cross the dunes from east to west near the southern end with a well-equipped caravan of seventeen camels; he very soon (ound himself faced with the close, parallel sandridges over 100 metres high and from two to faur kilometres apart (much higher and broader than the Simpson Desctt dunes) He came to the conclusion that his camels would be exhausted after even four days of travel across stich dunes, anil regretfully turned his direction north-north- west for 300 miles along the lanes between the dunes to Siwa; or this occasion he travelled 420 miles in 36 days. Unfortunately, the original accounts of Rohlfs’ journey have not been seen by the present writer, but it has been gleaned at second-hand (from Bagriold) that Roklfs saw no plants in the whole of the 300- mile journey in the duns execpt a little grass an nearing Siwa, Such an absence of plant life is highly unusual, even for a desert, though scen in other parts of the Libyan Desert, as dleseribed by MacDougal (7) and Thomas (10). De Lancey-Forth, formerly commanding the Frontier Camel Corps, was inspired in his journeys by the search for the legendary oasis, Zerzura, as were many other explorers. His journeys into the Sand Sea were ky camel. Tn the winters of 192] and 1924 he went about 100 and 200 miles reayeerively, south of Siwa, the second journey being an extension of the first along the dite janes, This route lay some 50 miles west of Roblfs’ route, bur the western edge of the duges was not seen. Jm one or two sandy valleys he found add, grecn bushes (un-naned), too biller fur camels to eat, but there was no other vegetation, In 1922-23, de Lancey-Parth investigated the south-eastern edge of the dunes and penctrated them ag far as Rohlfs had done hefore turning north-north-west. The only vegetation found was in a valley thinly grassed for a wile or two. Tt is probable that this valley lay on the track of a vaguely known and very difficult Arab caravan voute across the southern dunes of the Sand Sea, connecting Kufra on the west with Dakhla Oasis on the east; Bagnvld also saw this track in his journey presently to be described, De Laneey-Forth’s Journeys were not puh- lished until 1930 (4), Bagnold’s travels, unlike these, were all by motor-car, He was among the pionecrs who demonstrated the accessibility of big dunes to light motor-cats. and heeame very expert at negotiating them P, A. Clayton also made extensive desert journeys hy car. Tt may be argued that there is less chance of seeing any vegeta- Gum present when travelling by car, but the fact remains that, of the accounts sten, Bagnold’s expeditions alone collected and named *he ane of two plant species foiind in the Sand Sea. Bagnold attacked the dimes from Ain Dalla, making virtually three trips into them in 1929 and 1930. Ain Dalla is a small nasis about half-way along their eastern mitegin. He penetrated further westward than Rohlfs of de Lancey-Forth {1922}, and found the mysterious caravan route between Kufra and Egypt, but he wae unable to follow it ont of the dunes on the west side and had to make his escape southwards along the lanes between then. This purt of his journey took perhaps a fortnight, but he had travelled 700 miles in the dunes and seen note of the urea than any of his predecessors, having crossed a great many of the dunes at might angles in both directions. W. 7B. K, Shaw went with Bagnold as botanist. His collection of 21 species in all was identified at Kew, and his published notes on the vegetation of the actual San! Sex (6, 1931) are most helpful. The only species he saw there was Ephedra alate yar. Deciisnei Stapf., usually as single, isolated bushes 18” high in pure sand, Fle also collected a piece af wood from the remains of a dead shrub in the heart of the Sand Sea; it was. identified as another species of [phedra. This genus contains several widespread desert plants and belongs to a group 4 represented in all the continents except Australia, Along the rarely used camel route across the dunes from Kufra, he saw the shriyelled remains of plants for some hundreds of yards, in places. They were nol worth collecting, but he thought they might have been a species of the grass virtstida, Then, for great distances. he saw no plants at all save for rare tufts of the same small, withered grass, Within sume miles of this route are the following two entries on Bagnold’s map- "40 green bushes” and °200 green bushes” (1), The other plants on Shaw’s list were collected outside the Sand Sca and near wells. He also went with Bagnvid in 1932 on a long round trip of 6,000 miles jrom Cairo and back, to the southern part of the Libyan Desert towards the Sudan, and collected a further 27 species, again identified at Kew and published with his notes (6, 1934), For further information on the desert plants hetween the Nile Valley and the Sand Sea, the accontits of the botanists MacDougal (7) and Ilamshaw Thoinas (10) may be consulied, Both emphasise the absence of plant life for stretches of many miles and the poverty in number of species. It is of interest to note that several of the genera quoted by them are also characteristic of the Kara Kum Desert, eg., Aristida, Caliganum, Salsola, Haloxylon, Ephedra, and Tamarix which is important in all the Old World Deserts. In 1933 Orde Wingate (also in search of Zeraura) actually crossed the dune belt from east te west about 130 miles, and returned on his own tracks; he tra- yelled with camels on lat. 264-27° N,, crossing two of Bagnold’s tracks and alsa Rohlis’, Wingate was m the dtines 35 days, Ile saw “several brilliant green shrubs, usually halfway up a dune. They were very bitter and only eaten sparingly by the camels (¢.f., de Lancey-Forth). He also recorded a couple ol pieces of dead tamarisk trunks. It is not known whether the green bushes of Wingale and de Lancey-Forth were Ephedra, ax in Bagnold’s case. Only allusions to the journeys of the Claytons have been seen by the present writer, and published accounts were sought in vain. Daring the recent military campaigns in northern Africa (1940-3), certain small mobile units operated far south of the coastal strip on special missichs ; they traversed the desert by all modern means of transport, and there was some travelling in the Sand Sea with jeeps, Bagnold pioneered the formation of such iinits, and some popular accounts of their work have been published, One of these, (a) is written by W. B. K. Shaw, the botanist of earlier expeditians with Bagnold; another is by Malcolm James (b). however nothing is added ti the precisé botanical knowledge of the Great Sand Sea, To attempt a formal account of the meteorology of the Sand Sea ts nt pos- sible; Rohlfs probably recocds some useful information, and we know that he experienced rain there in 1874 (his book was unayailable to the writer). Bag- nald, from hig extensive knowledge, states that the amount of raintall throughout the Desert is ill-known but conspicuously sporadic, and absent for perhaps eight years ala time. Engler (25) and Gautier (5) say the same thing, mentioning heavy local storms once in twelve years, and giving an apparent average of about 3:9" annually in the French Sahara. Another factor is the large diurnal range of temperature in the erg or dune desert. Also, the air is very dry, with relative humidity usually under 20%, and evaporation from a free water surface is quoted as being 4.2. or more, yearly (MacDougal). The temperature drops quickly at night in the dunes, and in winter there may be nocturnal frosts, but nothing fa’) "Long Range Desert Group,’ Collins, London, 1945 (iy) “Born Of the Desert," Collins, London, 1945 5 to approach the winter cold of the Takla Makan, Dry winds are an impottant feature of the Lihyan climate. Human inhabitants are few in umber and con- fined to the larger of the oases; this is in contrast tu the western Sahara, where there are small, nomadic tribes of camel herders. To sum up, the Great Sand Sea is quite uninhabited, scarcely visited, and very difficult for lrayelling, With stall exceptions the large parallel sandridges are bare of vegetation. Amoug the few plants which have been seen there and identified are Ephedra alota var, Decaisnex Stapf, and possibly another species of Ephedra, together with grass fragments, perhaps Aristida sp., aud remains of supposed tamarisk (11) trunks. These bare parallel dunes are very much more arid than the Simpson Desert, with its comparatively good cover of vegetation between the dunes at least; the Libyan dunes are also much higher. REFERENCES 1 Bacworn, R. A, 1931 Journcys in the Libyan Desert. 1929 and 1930 Geog. Jour., 78, 13, 524, 576 et seq. Bacnoitn, R. A. 1935 Libyan Sands, London 3 Bari, Joun 1927 Problems af the Libyan Desert. Geog, Jour., 70, 21, 96, 105, 209, 512 et seq, 4 Yortu, a B, pe Lancey 1930 The Zerzura Problem. Geog. Jour,, 75, 48-5i 5 Gaurizr, E. F. 1935 Sahara; The Great Desert. New York 6 Kew Burr. Misc. Invormy. 1931 The Flora of the Libyan Desert (Plants collected by W, B. K. Shaw), 4, 161; 1934, 7, 281 7 MacDovear, D. T, 1915 Desert Ecology in North America and North Airica, J, Ecol., 3, 42-55 (notices) 8 Ronnrs, G. 1875 Drei Monate in der libyschen Waste. Cassel, (Not seeti 9 Scie O. 1926 Die aegyptisch-arabische Wiiste. Introduction to “Vegetationsbilder”, Rethe 17, Heit 5/6. (Edited by Karsten, G., and Schenck, LL, et al.) Jena 10. Tomas, H. HamsHAw 1921 Some Observations on Plants in the Libyan Desert. J. Ecol, 9, 75-89 11 Wineate, Orne, 1934 Tn Search of Zerzura. Geog. Jour., 83, 280-308 bo B. THE TAKLA MAKAN DESERT The Talda Makan is the most terrible of all the five or six Asiatic desert regions. alid probably comparable only with the worst parts of the Libyan and Sahara Deserts. It lies in Central Asia in the southern patt of Sin-kiang, the westerh-most province of China. ‘This elevated depression, also known as Kash- garia or astern Turkestan, is almost surrounded by some of the highest moun- tains in the world. On the north are the Tian Shan Mountains, un the south the mountains and plateau of Tibet, and on the west the Pamirs; to the east lies the Wandering Lake, Lop-nor, and the Takla Makan extends into the Gobi Desert. These mountain batriers are considered by Popov (16) to account for the poverty of the vegetation of Kashgatia, so [ar as species numbers go, compared with the richness westward in the deserts af Asia Media; however, the inhos- pitable sund-dtunes of the Takia Mukan are hardly in need of this explanation for their lack of plant life. Sir Aurel Stein (18) estimates the size of this desert as about ‘900 miles long from cast ta west, and 300 miles in its broadest north-south 6 extent. It lies between long. 78°-93° E. and lat. 37°-41° N. The River Tarim, lowing from west to east, bounds the desert almost completely mm the north; three roughly equidistant smaller rivers flow across from the southern edge of the desert towards the Tarim, Only two teach it now, thongh in the past less arid conditions have prevailed. The other rivers coming down from the moun- tains sao lose themselves in the desert, Clur best accounts of rhe Takla Makan are those of the Scandanavian explorer Syen Herin, and the British archaeologist Sir Aurel Stein. Of the relevant Russian literature we have only seen a botanical paper by Popoy (16), Stein’s published maps (18) are excellent, including details of the veyetalion and terrain, [edin’s journeys were probably more numerous ard penetrated the heart af the desert, while Steity was chiefly concerned in visiting buried cities and ruins; we gather from Medin (14) that it was scatcely worth while making a botanical collection because of the sparsetess of plant life, The ‘Takia Makan depression is actually about 3,000 feet above sea level; this, combined with the high latitude (37°-41° N.), gives it Far colder winters thas any of the other deserts under consideration, Prosty ceriainly occur on winter nights in the Sahara, and in the Simpson Desert too, but the mean January (winter) temperature for the Sahara is everywhere abuve 50° F, as quoted by Kendrew (15), In Central Asia, the mean for January 1s much below freezing- point, small rivers are frozen all the winter, and the temperature rarely rises above {reezingepoint even at noo. Scanty records from Kashgar and Yarkand at the western end of the desert can guide us, together with observations made hy Hedin on his journeys and quoted by Kendrew. He recorded a minimum of -25" f, in January in the middle of the Takla Makan, and on 2 January i maximum of only 8° F. At Tarini Jangikol (alt. 2,890°) at the eastern end, Hedin observed the following mean temperatures during his stay— February 17° March 40° F, April 55° F and May 69° F. The stnumer temperatures are not correspondingly lower. In the Sahara lhe difference between the mean summer and winter temperatures is of the order of 35° F.5 at Kashgar it is 60" F., a tremendous range. The daily range, as far as records show, is in the neighbour- hood of 30° F., as in the Sahara, “The mean summer temperatute for July ig over 80°F. for Kashgar and Yarkand, compared with 90-100°T. for Saharan stalions, and there are maxima of over 100° F., according to Kendrew (15). ; Rain falls mainly in summer, but is augmented by melting winter snow; & mean anmial precipitation of 3-5 inches has been recorded at Kashgar; in the heart af the desert it is probably less aud very crratic. The mean relative hinuidity is gencrally low, and the high mountain harriers are responsible for extluding the rain-bearing winds, The figures just given (chiefly from Kendrew) for mean values in Central Asia are from records of only 1-3 years, (Refer also to ‘Table I in the Appendix.) There is no doubt that Hedin's journeys crossed the waterless desert dunes which are of so imtich interest to. us. The most arduous of these journeys was in spring 1896 from the western end, 120 miles castward to the first river flaw- ing north (the Khotan-daria) ; on this occasion he lost two out of five men and seven out of eight camels (12). The chief difficulties were the sandstorms, the heat, the slowness of travel over the high, compound, steep sand-dunes or “dawans", sometimes as much as. 200-300 fect high, Stein (18) later attempted a erussing in this region with a better prepared caravan, but he was forced to turn back; he vouches for the height of the sandridges, stating that the camels situply became exhausted crossing them, These dunes are practically paraltel, and continuous enough to make detours around them out of the yuestion, Some 7 of the dunes of the Sahara and Libyan Deserts are of this type and magnitude. In the dunes at the edge of the deser! there were occasional shrivelled reeds and grass tussocks and solilary poplars; the last plants seen were tamarisks, Im the desert itself on the fine, vellow, shifting sand of the dunes there was scarcely any vegetation, Onee or twice plants were seen on these dunes, a solitary tamarisk (Tamarix elongata) here and there, reeds (“Kamish’” Lasiogrostis Splendens ) m 2 hollow, and two or three icolated poplats. LEyentually Hedin and the remnants of his party reached the river with its dense groves of tamarisks and poplars (““Vograks”) in the sandy sail. The tamarisk out in the desert commonly Brows as a shrub on top of a conical ound of sand held by its reots, and these “tamarisk cones” are mentioned and figured hy both Stein and Hedin; the rare poplars and tamarisks found are often dead. Bagnold (2) describes tall tamarisk cones or “terabil” in the desert of southern Egypt, and these tamarisks seem to he the hardiest of all the vegetation. Along the River Tarim, the small ‘saline soil Saxaul” tree, Haloxvlon (or Anabasis) slounadendran (= Arthrophyton Haloxylan) occurs in place of tamarisks, Hedin (13) alsv crossed the eastern end af the desert from north to suuth, in the cold of winter. The sand-dunes here are still parallel, but this time he travelled almost in the same direction, occasionally crossing the main dunes at an angle, There were smaller transverse ridges which he had to climb, dividing the valley bottoms into a series of depressions or “bayirs’. The vegetation was as already described, rare patches af “Kamish” 8-9" high beitig seen. Ephemerals are absent from the Tall Makan. Popov (16) states that the saimé scantiness of vegetation prevails in the Sahara, but that there is compara- tively rich development of psannmophytes in the deserts of Asia Media on the western side of the Pantirs, 7¢., in the Kara Kum and Risil Kin. Ile supplies the following species names for the common names used by the explorers: Desert poplar or tograk = Populus euphratica Oliv, and P. pruinasa; Tamarishs = Tomerix Pallas, T. hispida, T. raimosissinia Ledeb. (forming cones), and others ; Reeds = Lasingrostis splendens (Gramincac)—the Index RKewensis has ‘Lasiagrostis’. Saline spots occur between the dunes, where the peripheral streanis penetrate the degert underground, and here halophytes like /ycitwm, Nilraria and Hadlostachys ave found. Other plants occurring, probahly im the belt of riverine follest veretation, are Phragnites communis (the common reed), Uheyerhise inmpflata and Alhayi Kirgisorun. Popoy says that the only psammuphytes he occasionally found in Kashgaria were Agriophyllun and Corispermium (both Chenopodiaceae}, which are annual sand halophytes. {n brief, then, the Yakla Makan is characterised by very large sandridges almost as devoid of plants as is the Great Libyan Sand Sea. Tis altitude (a platean over 3,000’ high) is equalled by only one oi the other deserts in the present study, the Chihuahua (Mexico); and the Jatitude is also higher than all except the neighbouring Kara Kum Desert, These two factors give the Takla Makan colder winters than the other deserts, though its summers are still very hot; the ahnual temperature range is therefore greater than in most other deserts. The Tew plants or their remains, seen only on tare occasions during a journey in the dunes, are poor, shrubby specimens of isolated Tomariv or Populus, and a smal) reed-like grass (Lasioyrostis), The lamarisks usually grow out of the top of a cone of sand held hy their roots. The species names are quoted here just 8 as given by Hedin and Popov, and it 1s not considered necessary in the present study to go into the relationship of the species or their synonymy, No ephemeral plants occur, and the desert supports no human inhabitants, The Simpson Desert, though uninhabited, does not approach the Takla Makan in barrenness or ii extent. REFERENCES 12 Ilemwn, Sven A, 1898 Through Asia, 2 vols, J-ondon 3 Hen, Sven A. 1903 Central Asia and Tibet, towards the Holy City ol Lassa. Pt, 1. London 14 Heo, Sven A. 1904-1907 Scientific Results of a Journey in Central Asia. 7 vols, Stockholm (5 Kexprew, W.G. 1927 The Climates of the Continents. 2nd Edn, Oxtord 16 Puroy, M.G. 1931 Between Mongolia and Iran. Bull, Appl. Bet., Lenm- grad, 26 (3-5), 45-84 (71-84 English Summary ) 17 Srery, Str M, Auren 1907 Ancient Khotan. 2 vols, Oxford 18 Svery, Sie M, Aveer 1928 Innermost Asia. 4 vols. and maps. Oxtord Cc. RUB’ AL KOA Southern Arabia contains the worst portion of the Arabia Sand Desert; it is kmown as the Rub’ al Khali or Empty Quarter aud is uninhabited. This remained one of the unexplored regions of the earth until quite recently, when it was crossed from south to north by Bertram Thomas (21) in 1931; a year later Philby (19) traversed the south-western part from east to west. Both these journeys were camel journeys of well over 300 miles, each through a varied desert terrain including high, bare parallel sand-dunes. Thomas, contrary to expecta- tion, [ound water, but it was saline; Philby found none. Vrom the accounts of these two experienced Arabian travellers are obtained our most adequate picture of the vegetation, and it is considered sufficiently representative. Both published good maps, especially Philby (19). The planrs seen were scarce, but were such an important part of the economy of the expeditions, providing the sole camel food and firewood, that they were continually mentioned in the narratives by their Arabic names, Some isolated vegetation was usually in sight, though often dead, and it was seldom necessary to make camp where none existed. Philby (20) states “from all accounts the Libya: Desett is worse off for vegetation than the Empty Quarter under the influence oi a long drought”, In the Libyan Desert fodder for the camels is catried with the caravan, but in Arabia the camels have only the forage they can find om the journey, and the desert crossings of Thomas and Philby were both made under these conditions, which would be quite impossible in the Taibyan Desert. ‘This gives a measure of aridity of the Rub‘ al Khali; the Arabs accomn- panying Philby often stated that, in their judgment, a given arca luoked as if it had had no fain for four or five years or longer, and this was very probably the case. There is little doubt which are the important plants of the region—an Arabic— but their exact botanical identity is more uncertain. Thomas (22) gives a few botanical equivalents, and Thilby (20) quite a long list in an appendix, On comparing these and weighing the evidence, the following information ts obtained. Hadh—probably Salsola sp. (Chenopodiaceae), according to Thomas (22). A saline, sage-coloured bush, the hardiest of all the desert plants. A good camel pasture, but too small lor firewood. Widespread. Philby (20) gives 9 “Hadh" as Cornuleca mouocantha Del, (Chenopadiaceac), on the authority of Rohlfs (8); the explanation may be that this name applies in North Africa where Rohlfs travelled. Abal—Calgonum sp, (Polygonaceae). A large shrub often forming thickets. Tt also is very hardy and widespread. The wood is a good fuel, and large enough for making titensils and small structures, These two appear the most important, Philby adds two more as being the further staple plants of the southern sands, viz.: Alyua—-Diplerygium glaucum Decne. (Cruciferae), Andab—Cyperus conglowmeratus Ratth, (Cyperaceae). A small sedge. Then two slightly less hardy: Rirkan—lagonia gluttnosa Del. (Zygaphylaceae). Zaht—Tribulus sp. A shrub, probably small, (Zygophyllaceae), Under more favonrable conditions at the cdpge of the desert occur several species of Acaci and grasses of the genus Aristida, There are dlso salt plains carrying samphires (Irthrocrenmm spp, and rélated genera), The plants showed distinct zonation, and the sands were usually better vegetated than the gravel plains. Philby (20) indicates that an ephemeral vegeta- tion occurred after rains, but suv little of it; he just mentions Sadan (Newrada Rosaceae) and Halam (7?) as tiny plants. His scientific “collections” (“plants” not specified) went ta the British Museum, and sore, at least, of his botanical information came from that institution, The writer could not discover where the plants named by Thomas (22) were determined. The Rub’ al Khali, then, is much less arid than the Libyan Desert or the Takla Makan, as evidenced by sume vegetation (dead or alive) being always in sight; this vegetation is sufficient to sustain a few travelling camels, and to pro- vide firewood. By contrast, camel fodder must be carricd with the caravan in the Libyan Desert. Two shrubby plant species are the most important in the Desert—“Hadh" and “Abal" (probably Salsola sp. and Cualligonum sp. respec- tively )—whilst a few others, including a sedge, a crucifer and some Zygo- phyllaceac also occur commonly in places; ephemerals grow after rain, and many other plants of less importance are present, The general se of Arahic common names by both Philby and Thomas would have made precise identifications diffi- cult, unless good collections were also available, The Rub‘ al Khali is uninhabited for the most part. The dune regions with their vegetated valleys, and often slopes also, must bear some resemblance to those of the Simpson Desert, but apparently extensive arcas deyoid of vegetation do occur, particularly on the gravel plains. REFERENCES 19° Paruy, AH. Sr. Jous 1933 The Rub! al Khali, Geog. Jour., 81, 1-26 20 Putcny, H. Sr, fonw 1933 The Empty Quarter. London 21 Tuomas, Bertram 1931 A Camel Journey across the Ruh' al Khali. Geog. Jour., 78, 209-242 22 Tromas, Bertram 1932 Arabia Telix. London 23) Twrtcuunt, K.S. 1944 Water Resources of Saudi Arahia. Geog. Review, 34, (3), 365-386 D, WESTERN SAHARA The western part of the Sahara is better known than the Great Sand Sea. but it has been more difficult to obtain precise mformation about the Hora of the ergs. as distinct from the flat, stony desert (hamada, gsevir and reg) and the wadis or dry watercourst's; the wadis, especially, are the must favourable plant if habitat, The ergs cover only a small percentage of the total desert area, they are more accesstble thai the Libyan Erg and may contain sand-free passages a few injles wide, through which it is easy to travel. Most of the knowledge of the Algerian Sahara is dite to the military and scientific surveys of the Kyench during the last 60-70 years, and the present in- gecessibility of niuch of the releyant French literature has been a great clis- adyantage in this study. Among the more important sources of information avail- able were Gautier (28), Engler (25) and Maire (30); the scientific reports of the Voureau-Lamy Expedition 1898-1900 were unfortunately among those nol seen (27), The Algerian ergs aré not quite so large as the Great Sand. Sea, nor so regular and unbroken; moreover, wells anc pasturages may be present. The Edeyen Ere in Fezzan is 2 humid erg and habitable, with permanent lakes which are usually salt or brackish but in some cases iresh, The Great Orienta) Erg and the Great Occidental Erp are drier, but the Occidental is furrowed by long lines of verdure which seem to follow underground wadis, Other ergs without obvious watercourses, like the Erg-er-Rawi atid the [enidi lrg, have plentiful under- gtoutid water and associated pasturage, The most inhospitable of the great Algerian ergs ts the Esh-Shesh Erg, where wells are very few. From Gautier’s remarks, it seems likely thal there have been rare “good seasuns” when the natives could find same grazing in it for thvir Mocks; he stated that it is “accessible at a pinch”. This was before the recent successfil journeys of Bagnold and others inte the Great Sand Sea, when that ere was still believed inaccessible, Monnd, in 1937 (31), casually mentions that he crossed the El Shesh Erg when on a camel trip and gives a photograph wy a mass of bare sand-dune summits, but no detailed account. ‘The area covercd by the Oriental and the Occidental Ergs is, in each case. rowehly 100 x 200. miles, and there are, of course, other relatively insignificant occurrences of dunes and unconsolidated sand in the West; but it is really difficult to olftain exact information about the vegetation of a giver erg or dtine-field- Generally speaking, sands are capable of better water relationships for plant roots than the various types of stony desert. but there must he ayailable the minimuin Water requitement cither as a phivial or shallow subterranean supply. Tt is enlightening to repeat the general ttpression of Gautier who las known the Sahara long and intimately. He says that the emphasis is on the absence of Plands, and a “pasturage"” is a relatively luxuriant area Where there are tufts of plants at least fifty or sixty yards apart. This makes grazing “an extremely ambulatory exercise”: moreover, these pasturages are tsolated from one another hy a camel journey uf matty hours and often many days; even distances of the order nf 100 miles. Local rain-storms cause the occasional appearance of short lived ephemeral vegetation in normally hare regions; even so, the plants ate very sparse. Apparently the root system of each must he able to exploit a great deal oi soil, without competition from its neighbours, in order to obtain sufficient moisture. This type of vegetation is “ashab" pasture and consists of a few different species, chiefly Sazignya longistyla (Cruciferae). a rosette plant with purple- flowering stems from 4 to 18 inches high. In the Australian semt-desert arcas, and probably the deserts also, ephemeral cruciferous plants of similar habic are yery common after winter rains, Such plants were found in numbers and relished hy the camels on Madigan’s 1939 Simpson Desert Expedition, on the eastern margin of the desert where rain had fallen. We have not the long acquaintance with the Simpson Desert which is necessary to speak authoritatively about the ephemeral vegetation appearing only after rains; it ia probable that these desert i rains are not so infrequent in Australia as they are in the Sahara, To return w the discussion of the ephemeral species of the Sahara, commonly accompanying Savignya longistyla and its allied species (see list helaw), there is another important crucifer, Moricandia arvensis, together with Plantaga ogdta and Launea resedifolla, Tamarisk “cones”, like those of the Takla~-Makan Desert, hardly figure in descriptions of the dunes of the Sahara, though Bagnold (2) does describe a small group of high ones up to 50 feet called “terabil”, Tamarix mannifera, in the flat sand-desert of southern Egypt. But several shrubs, including tamarisks, cause the accumulation of sand mounds in the western Sahara and stil] survive, é.9., Nitraria, Limoniastrum and Aristida pungens. A short list has heen appended of the important species present among the few platts of the Great Ergs; more crucifcrous species might be included. In the mass they occasionally produce quite a floral spectacle, and most nf them are favourite fodder plants; indeed, almost any plant provides fodder for camels. It is considered that the nomad tribes and their grazing camels have no small destructive effect on the vegetation of the Sahara (Chevalier). Qne very helpful source for this list was Engler (25). Maire (30) is probably the best modern authority on the systematic botany of the Sahara and gives a very good outline of past work on the subject, as well as a flora of the region; but it was disappointing not to see the third part of his study, which he planned to contain an account of the vegetation from the geographical and ecologi- cal points of view, Important PLaxts ov tits Westerns Satara Ow Dowes Aristida pungens Desf, “drinn” (Gramineae), Widespread here and in the other Old World deserts. A grass of moderate size with long, shallow rhizomes, valued as a fodder and sand-hinder. A. plunosa and other species also occur. Vig, Veg.-bilder, 6 Reihe, Heft iv, Taf. 24 (29). Cornmulaca monacantha Del. “had”, (Chenopodiaceac.) A spreading shrub, fayourite camel fadder. Calligonum comosum L’Hér, “arisch”. (Polygonaceae.) A broom-like switch- shrul or small tree; important as camel fodder, Also C. azel Maire. A noteworthy gens of Old World desert plants. Fig, Veg.-bilder, 20 Reihe, Heft i, Taf. 3c, (29). Savignya parviflora Webb (includes S. aegyptiaca and S. longistyla, Cruci- ferae). Polymorphic, short-lived, ephemeral herbs following rain; often the chief constituent of “ashab” pastures. Old World deserts; valued as caine] fodder. Pig. Veg.-hilder, 20 Reihe, Heft i, Taf. 3c. (29). Moricandia arvensis DC (Cruciferae), Another important “ashab" ephemeral with the qualities of Sazignya. Malcolmia aegyphaca Spreng. (Cruciferae). A widespread dune anrmal. Camel fodder. Leunva resedifolia O, Kantze (Compositae). Perennial. Plantage ovata (Plantaginaceac). A small herb of the “ashab’. Ephedra alata Dec. and other species (Gnetaceac). Mostly leafless switch hishes. Ox Frarrer SaAnpy AREAS Tamarfta spp. Tamarisk (Taniaricaceae). Small irees, especially along wadis, Including T. aphyila Karst. “athel’*. Limontastirum guyonianim Coss, et Dur. ( Plumbaginaceae). A shruh, often large and half-buned, Only the flowers acceptable as fodder. Retama ractam Webb (Leguminosae). A leafless switch-shrub of the Old 12 World deserts, Only the flowers taken for fodder. Fig, Veg.-bilder, 6 Reihe, Welt iv, Tat, 23 (29). Euphorbia guyoniona Boiss, et Rent. (Euphorbiaceae), A leatless switch shrub, Fig. Veg.-bilder loc. cit. Matthiola livida DC (Cruciferae). Small, herbaceous sand ephemeral. Traganum nuda Del. (Chenopodiuceae). Shrub, Nitraria reiusa Asch. (= N. tridentata Zygophyllaceac), Shrub. The dunes of the Western Sahara, then, are predominantly bare. Ths most noteworthy of the sparse vegetation are the rhizomatous, creeping grass Aristida pungens, various species of desert shrubs often broom-hke in habit, and a selection af ephemeraly (chiefly Cruciferae) of much importance after rain. These lalter attract ihe nomad tribes im certain seasons. The Simpson Desert is more denscly and variously vegetated than-the driest of the Saharan ergs, and its dunes do not attain the size of the largest in the Sahara. But in both cases ihe most widespread species is a grass, /lristida for the Sahara, and Triodia for the Simpson Desert. Further study of the ephemeral flora of the Simpson Desert is needed. REFERENCES 24 Cannon, W. A. 1913 Botanical Features of the Algerian Sahara. Carnegie Inst, Pub. No, 178. Washington D.C. 25 Enever, Avorr 1910 Die Pflanzenwelt Afrikas. Bd. I. (Die Vegetation der Erde, 1x.) Leipzig 26 Fourrav, f. 1896 Un voyage dans le S. tunisien et dans Je 5. algérieh région du Grand Erg. Paris. Sec. Géogr. Comptes rendues, 99-102 27 Foureau, F. 1905 Documents scientifiques de Ia Mission saharienne, 1, 400-413 (Mission Fourean-Lamy), Paris, (Not seen) 28 Gatrisr, E. F. 1935 Sahara. The Great Desert. (English ed., translated by Mayhew.) New York 29 Karsten, G. and Sctrencx, II. 1903 Vegetationsbilder. Reihe 6, Heft iv; and Reihe 20, Heit i, Jena 30 Marre, R, 1933 Etudes sur la Flore et la Végétation du Sahara central. Mém. Soc. d’ Hist. Nat. de l’ Afrique du Nord. (Pt. 3 not seen) 31 Monon, Tw. 1937 New Journey to the Western Sahara, 1935-36, Geog- Jour., 89, 152-156 32 Scuweiyrurtn, G, 1913 Pflanzen, in E. Hartert’s Expedition to the Cen- ital Western Sahara, Novilates zoologicae, 20, 145-163 33 Sytanrz, H. L. and Marsut, C. F. 1923 The Vegetation and Soils at Africa. With maps. New York E. TIJE SIMPSON DESERT This desert lies almost in the centre of the Australian continent, in the three States af South Australia, Queensland and Northern Territory, It is a uniform area of continuous, parallel sandridges, running north-north-west and south-south- east and forms part of what is known as the Arunta Desert. Two or three water- courses enter it, and there are scattered shallow claypatis occasionally filled by rain, but no permanent water except Anacoora Bore on the western margin, and sotie springs in the south (38). Tt has been explored and described recently in considerable detail, both from the air and on the ground, by Madigan (37, 38, 39, 40); it is therefore unnecessary to do more than mention the most important points here, but readers should consult these reicrences and maps. 13 Madigan (38) reviewed journeys by himself and by earlier explorers around the margins of (he desert. including his aerial reconnaissance in 1929. He was also the leader of the scientific expedition in 1939, which made the frst ground crossing of the centre of the desert. and travelled some 200 miles, from west te egst, by camel across the parallel dunes. The series of Scientific Reports of this expedition may be found in the Trans, Roy. Soe. S. Aust., where they were published between 1944 and 1948. (See especially the “Catalogue ot Plants” (36) and “Soils and Vegetation” (35), Reports No. 7 and 8), Colson (34). inv 1936, had crossed the southern part of the desert by camel, practically along the 26th parallel of latitude, No other crossings have been tecorded, and the aboriginals, though nomadic, only visit the deser{ margins, The dunes vary in height from abottt 30 feet to 100 feet above the level of the broad passages between them, and -hey are spaced on the ayerage at four per mile but may be closer when small, They are strictly parallel and apparently wobroken for lengths up tn 200 miles; the eastern slope is steeper than the western, and they consist of fine, red sand. The upper slopes ol the ridge are steep and usually bare and uncunsolidated, with # sharp erest of live sand, while the lower slopes and inter-dune passages may be well vegetated with low, shrubliy plants, clucfy spiny clumps of the grass “spinifex” (Triodia Busedowit TE. Pritzel), so characteristic a genus of the Aus- tralian sandy deserts. Occasionally larger sliuhs are also present, particularly several species of the genera Acacia, Cassia and Crotalaria (Leguminosac), Grevillea and {okea (Proteaceae). and Evemophila (Myoporaceae). The last three peneérva are particularly Australian in their affinities, whilst Acacia is repre- sented only by the peculiarly Australian phyllodineous section of the genus; other Aticia species are known as desert plants elsewhere. The shrubs Sida and Thibiscus (Malvaceae) should be added to the list, and also another shrubby grassy Important at times on the dunes, vis.. “Cane Grass”, Zygochloa poradoxra (R. Br.) 5S. T. Blake 1941 (better known as Spinifeas paradoaus (RK. Br} Benth.), The habitat differences hetween this species and Vriodia have been explained by Crocker (33); briefly, Zygochlne is found on the unstable sand of the dine crests, which are not always bare, while 2'riodia is restricted to the lower slopes and corridors between the dunes. The small, shrubby Chenopodiaceae, so common m the semi-deserts of Aus- tralia arid other parts of the world, occur to some extent, but the family Legu- minosae is more important among the dunes of the Simpson Desert. Euculypts are rare and usually associated with the few watercourses which are defined, nol by water, but by a line of verdure; only two species of Aucalypius were found. These were, firstly, small trees of E, coolabei Blakely and Jacobs (formerly included with &. aecrotheca TV. Mouell.), “Desert Box”, and secondly E. pyrephore, Benth, “Bloodwood”, up to 12 m, high, Some other small plants occurred in the desert. and they are listed in the present wriler’s Botanical Report (36) as “Plants found in the desert proper” It is believed thar there must sometimes be present an ephemeral flora similar to that seen on the River Diamantina to the east after good rains, but it was mot eonspicnous in che winter of 1939 when Marigan’s crossing was made, though he does record great mats of the succulent camel fodder “Munyeroo” (various Portulacaceae—in this case Porlufaca intraterranea J. M. Black) at camp 11 around claypans of walter near the middle of the desert. .A few small ephemeral types of Cruelicrac, Goadeniaceae and Compositae .were found which might be more abundant in oceasional good seasons. Compared with some, but not all, other important deserts of the warld, the Simpson Desert is well covered with vegeta- ion, though Madigan (39) found both caniel food and firewood very searce in 14 the centre of the desert. Between the dunes and on their lower slopes, the small shrubs (chiefly Triodie) are of the order of a few yards apart; many may be dead, but they do persist, holding the sand and forming a seed bed. Tvriodta is a perennial rhizomatous grass, often with a peripheral habit of growth, the centre of the bush dying and disintegrating, leaving a hollow ring. Many T'riodia species, like, 7. Basedowii, have pungent, hard, needle-like leaves of little use for grazing. Few precise records of the climate of the Simpson Desert are to be had. However, together with the Lake Eyre region to the south, it forms the driest part of the Australian continent, receiving less than 5’ average annual rainfall; seven South Australian stations are below this figure. Evaporation is high, and the Meyer Ratio P/s.d, relating it to rainfall (refer to stction on Thar Desert), does not exceed 2 for any month of the year (Prescott (41) ). Other sources of information on the climate are the Commonwealth Climatological Atlas of Aus- tralia (42) and the reports. of the 1939 Expedition, especially Madigan’s geological report (40), and the meteorological log uf his journey (39) from May to August. Crocker (35) gives data on the great variability of the rainfall, which has no particular seasonal periodicity. Madigan's obseryations in the winter of 1939 may be summarised as fol- lows :—‘Weather brilliant three-quarters of the time, with warm days to frosty nights. Temperature range 23°-78° F,, Min, temp. at grass 23° F., Mean min. at gtass 8°5 F. Average max. temp. 70°F. Rain, six rainy periods in ten weeks, mostly very light, In 5 days, 14” fell im the desert ; and in 6 days, 2” fell at the Mulligan—this was exceptionally wet. Wind did not exceed 20 miles per hour in the desert.” From Commonwealth sources—summer temperatures up to 110° F. or over, with nights at 60°-70° I’. Raimfall—never as low as zero, but varies from less than 2” up to #2; and droughts of less than 2” may last for three or four years. At Charlotte Waters, on the western margin, the total annual amounts in inches from 1895-1915 were as follows — 10-40, 2°84, 7:16 5°23, 2°74, 3:79, 3:22, 2:40, 9°67, 3°78, 3°94 2-11, 5°43, 12-31, 2°75, 6-81, 4°58, 2°73, 4°37, 9°06, 1°43. The lowest recorded is 0°43 at Kanawava on the Cooper in 1896. Some further data supplied by Madigan are quoted below. Simeson Desert Climate — From Chim AvoLoGicAL Atias oy AusreAtiA (W. S. Watt). Temperature Jan. Feb, Mar. Ap. May Jun. Jaly. Aug, Sept. Oot Noy. Dee Mean daily max. 100 100 95 8&6 76 7O 70 75 82 90 6 49 Mean niax. for year 86° F. Mean daily miin. 73 73 67 58 50 45 43 46 52 60 6? 71 Mean min. for year 59° B. Mean daily - - 86 68606 6872 GR OGH—iO—C GHC StsC*OL 80 Mean annual tetup. 73° F, Relative Humidity: 9 aan. 31 35? 45 28 (<34) Meau rel. him, at 9 am, for year 38? (<41)) 3pm. <2) 25 30 <20 Mean rel. hum. at 3 pm. for year 22. 30° C. (86° F.), The avnual rainfall has two jmaxima, one in spring and a lesser one in auttimm; there are no tains in summer for at least four months, sometimes cight. Much of the Tura sand-dune desert has an annual rainfall of less than 4’. Some figures are set out in the Appendix (Table I) giving rainiall and tempera- iure values. The Sanps and tis VEGETATION According to Doubiansky (43), the portion of the south-eastern Kara Kum mapped by him represents abour one-sixth of the total trans-Caspian Kara Kun and is made up as follows — (1) One sixth consists of a barchan range of bare sand bordering the River Oxus in a band 10-50 km. wide (6-31 miles), covering an area of about 9,000 sq. km. (= 3420 sq. miles). Repetek seems to be in this region. (2) One-sixth consists of sandy lily carrving @ growth of shrubby psammo- phytes and saxaul forests. (3) One-quarter consists of sandy ridges richly covered with vegetation. The rest of the area comprises sandy-clayey plains wnd foothills and does nut cornice us here; the regions just sct aut will now be described in more detail. (1) The berehan range is made up of parallel “barchan rows” running rotighly north-east anc south-west, generally at an angle to the river. The sand is probably derived mainly (rom fiver valley alluvium and forms at once into these barchan rows, hardiy passing at all through the stage of the companent sing:e crescentic barchans. The barchan rows are not very high, say 3-10 metres or even 18 metres (60 feet), they have vertically sintious crests and are up ito several himdred metres long, with a basal width of 100 metres or less and a valley pf the same width between. Work during four years at Repetek has shown that the crests of the barchan rows have a seasonal movement transverse ta their length; in summer they moye sonth-egst 15-20 metres and in winter back again under the influence af the pre- yvailing winds (which are not vecessarily at right angles to the barchan row); the net change of position, if any, is very smail; the side on which the steep or slip-slope occurs is thus regularly reversed. Plants establish themselyes on the more stable lower slopes and hollows between the rows, where a sparse growth nt the hardy endemic pioneer grass Aristida pennata Trin. var. Karelini Trin, et 18 Rupr. “seline* is almost the sole species found; it is a coarse grass forming shrubs three feet high, of sume fodder value, and able to survive much burying and uncovering by the sand. Doubiansky estimates that it covers at least 109% of the barchan sand area, This is (he psammophyte that Popov (47) expected in vain to see, above all others, in the Takla Makan, (C4, Triodi Basedowit, Simpr son Desert. } In the hollows between the barchans occur the annual sand halophytes — Agriophyilum latifolium F. et M, (Chenopodiaceae); Corispermum hyssupi- folium L. (Chenopodiaceae) ; Horanimowia ulivina FL et M. (Chenopodiaceae), a sinall spiny shrub; and Simirnewia turkestana Bge. (Leguminosae), a shrub occa- sionally stet: here, (2) Under suitable condilions, the hbarchans are stabilised into sandhills by pioneer shrubs, many of them switch bushes, First in importance is the Aristida sp. mentioned, then Calligonum spp. (C. Caput Medusae, Schrenk, C. arborescens Litw., C, elatum Litw.). These are all leafless Polygonaccae, some of them endemic. Further species arc also importatit:—Ammodendron Conall Bge. (Leguminosae), a shrub or tree; the whole genus consists of psammophytes endemic to Turkestan; Ardcia sp, (Leguminosae), arhorescenit; fremosparton flaccidum Litw. (Leguminosae), a leafless tree or shrub, the whele geniis is endemic to the sands of Turkestan, It ig not to be supposed that these first-line pioneers suffer no casitalties, for the colonisation of these wind-swept sands is often full of sethacks, Less resistant plants. following on these are :— Calligonum eriopodum Bge. (Polygonaceae), arborescent; tstragalus ammoden- droan (Leguminosae), semi-shryh; Aristida pennata Trin, var, minor Litw. (Gramineae); Salsola Richteri Kar. (Chenopodiaceae), arhorescent; Carex physodes M. B, (Cyperaceae), a low endemic sand-sedge forming a thin cover between the bushes, it is the most widespread plant in the sands of Turkmenistan and important for forage; as it covers the sand, the carliest pioncers disappear entirely; this plant does not survive in drifting sand. The third stage follows in which, among other species, Calliganum setosin Litw.. (Polygonaceae) is the most widespread shrub, and Arthrophatum urbar- escens Litw, the Sand Saxaul (Chenopodiaceae), forming tall, leafless clumps, is the most important, Under the influence of humus from this cover of vegetation, the sand becomes consolidated and saline in the surface layers and the climax forest community of these sands is able to develop, This is 4rthrophytim AHaloxrylon Litw. (Haloxylon Ammedendron Chenopodiaceae) or Saline Soil Saxaul. It develops first in the hollows between the sand ridges as an almost pure community af open structure, The Saline Soil Saxaul is a small tree 12’-20’ high, greatly valued in Turkes- tan as a source of fuel. It grows nowhere else in the world and is really a very remarkable forest formation, capable of developing in pure sandy or salty deserts with an annual rainfall of only 4 inches or even less. The forest occtipies small and scattered areas and has, of cotrse, heen over-cxploited in parts, especially along the Middle Asiatic Railroad. Regeneration from seed is not very easy; a denuded Saxaul forest probably reverts to hare barchan sands. Aphedra spp. (Gnetaceae) are small shrubs characteristic of the pioneer stage, but not abundant. Studies by Petrov (quoted hy Seifriz (48) ) of the root systems of Kara Kum psammophytes, establish the fact that the hardiest of the pioneers of moving sands, Aristida, Eremosparton and Aimimodendron, exploit only the “sub-super- ficial” moisture horizon with an extensive horizontal spread of roots in the layer uf sand 40-150 cm. helow the surface; there is driet sand both above and below 19 this. The plants following these in the succession soom develop roots which also tap the ground-water, found at a fairly constant depth of J-4 metres, in the barchan valleys} this is unusually shallow for a desert and probably goes far to explain the relative fertility of the Kara Kum sands. Besides the important plants named, there is quite a rich eplivimeral spring vegetation (Seifriz (48) ). (3) The most widespread formation in the South-eastern Kara JXam is that af the saudy ridges covered with vegetation. They are up to 50m, high {170’) and 3-5 km, long (2-3 miles), and run chiefly north and south or north-north-east and south-south-west; the eastern slope is the steeper and there are broad- bottomed valleys between. Secondary bare areas are to be found on the tops of these ridges, here and there. It seems unlikely that these ridges are harchan rows of the type just described, merely stabilised by vegetation; they are on a much larger scale and probably have a different history. Only on the occasional bare arcas are found a few of the more hardy psammophytes mentioned earlier. The steeper slopes have the shrubs Calligonim and Astragalus with an herbaceous cover of Curex physodes. On the lower slopes and valley bottoms these die out at the invasion of a cover of annual prasses which may yield a large amount of forage in the occasional good seasons, atid at other times leave the ground bare. Tt seems better, however. to have the more reliable growth of sand-sedge (Carex physodes), even though it is a poorer fodder, with some of the psammophyte shrubs to yield frewood. This can be ensured by checking the natural plant succession by controlled sheep pasturing, It will be remembered that trees or stunted trees of the genera Tamarix and Pobplis comprise the bulk of the extremely sparse vegetation of the Takla Makan Desert; so far these have not heen mentioned as important in the Kara Kum. However, the same species do occtir naturally along the Oxus and elsewhere, atid they are also used for planting to control the moving barchan sands beside the settled valley of the Oxus. Paulsen (46) mentions tamarisks on knulls in the “Hummock Desert”. Gur final impression of the south-eastern Kara Kain is that it is one of the more fertile deserts with a richly developed endemic flora (22% of the species, according to Paulsen) of well-adapted psammophytes, and this ts linked with yery favourable soil-water relationships established in spite of low rainfall. We consider it better vegetated and’ more fertile than the Simpson Desert, REFERENCES 43 DovunrAxsey, V. 1928 The Sand Desert South-eastern Kara Kun. Bull. Appl. Bot., Leningrad, 19, (4), 225-285 (English Summary) 15) Kenorew, WG. 1927 The Climates of the Continents. 2ud edn, Oxford 44 Koppen, W. 1931 Grundriss der Klitnakunde. Berlin 45 Ornov, B. P. 1928 Ecological Conditions in South-west Trams-Caspian Kara Kum. Bull, Appl. Bot., Leningrad, 29, (4), 359-401 (English Summary ) 46 Pautsen, OG. 1912 Trans-Caspian Vegetation. Copenhagen. (Review only seen, in J. Ecol, 1, 1913, 133-137) 47 Poroy, M.G. 1931 Between Mongolia and Ivan. Bull. Appl. Bot., Lenin- grad, 26, (3-5), 45-84 (71-84 English Summary) 48 Serreiz, W. 1932 Sketches of the Vegetation of some Southern Provinces of Soviet Russia. J. Ecol., 20, 69-77 49 Vassinizv, 1. M, 1930 Ueber den Wasserhaushalt von Pianzen der Sand- wiisten, Bull, Appl. Bot. Leningrad, 25, (3), 267-272 (English Summary ) 26 G. THE CHIHUAHUA DESERT The western and south-western part of North America contains the most important of the deserts of that continent, vis., the Great Basin Desert, the Mohave Desert, the Sonora Desert and the Chihuahua Desert. The last, which lies chiefly in the central platcau of Mexico and the State of New Mexico, will be our chief concern, because i} has a sandy reginn about 100 miles square (54), soine 30 miles south of Et Paso on the border. This includes an area of parallel dunes Iscally called Los Medanos, which ran north-west aud south-east for about 20 miles, They ate white or yellow in colour and nearly bare of vegetation, There is a raad through them, which oiten becomes entirely obliterated by blown sand, andl also a railroad traversing dunes about 40 feet high. The sand aud dunes sutrotind soine large lake beds belonging to an internal drainage systent in mortkern-most Chihuahua, and from these they are probably derived; the dunes are active and largely unstabilized, They are from 50 to 300 feet hish (53), and, as with most stich inhospitable places, are not very well known to botanists. Shreve (53) reports that the vegetation is mainly shrubby, sparse amd irregtilar, giving as the characteristic pants the tollowiyy:— Poliomitha tmeana (Lubiatac), a woody bush; Yucca ¢lata (= Y. rddiosa, Liliaceae), a plant with a large sani-bindiug rool- system and a stem which lengthens on silung up; Prosopis chilensis (= P. julifora, Leguminosac); Mesquite or Algaroba Bean- AV widespread stirub or tres in southern LUS.A, Arlanisia. flifolia (Compositae); Sand sage brush, a small shrub; Dolea scoparia (Leguminosae) ; and FHymenoclea monogyra (Conipositac ), shrub. To these, MacDougal (52) adds—a shrubby Senecio {Compositae), Chryse- thamuus (Compositae); two frequently occurring perennial grasses Sporobolus eryplandrus and an Andropegen,; and the remnants of annual plants: He stresses the doninance of Poliomintha. He states, atso, that the valley bottom near Samalayuea, an oasis on the rai way in the dtines 30 miles south of El Paso, has yegetation typical of the mesquite plains, vis., Prosopis juliflora Mesquite Tree (Leguminosae) ; Zisyphus (Rhamna- ceae); Aoeberlinia spinosa (Capparidaceae); Atripler canescens (Chenopodia- ceac) and ah annual Croton (Luphorbiaceac). The valley bottom referred to is prabably not an mterdune hollow. Chemical analyses of the dune-sand are also given, li is not easy to find incteorological records far this precise area; annual rainfall figures for the various stations in the neighbouring desert range from just under 3” to about 16” per annum, and the annual evaporation values quoted by MacDougal (52) are tnany times greater. FI Paso (alt. 3,760"), on the Rio Grande, is the nearest available statiow and has a rainfall of 9°5” with the maxi nitim in Jilly; July is also the hottest month with a mean tenmperalure of B0°5° ¥., January is the coldest month with a mean of 44° l.; the means of scyeral yearly extremes are 104° F. and 12° F.; the daily range of temperature ts quite high anl has a value between 30°75 and 38°5° I. for the whole year, as is usual in deserts. Shreve (54) has published a map showing the mean annual rainfall of Northern Mexico, in which the dune area under consideration falls into the zone receiving 9-200 mm. (0-8”). Judging from his tables showing the length of rain- less periods experienced at the recording stations, some of which are in negiph- 21 bouring deserts, it must be quite unlikely that so much as a year passes without rain; this is in strong contrast to the Sahara and Libyan Deserts, It is difficult to assess the degree vot aridity of this desert in comparison with others, since its much smaller size lessens its importance. The sandridge portions of it may be less vegetated than most of the Kara Kum or the Simpsotr Desert, but it seems more correct to rate it less arid than these in the present state of our knowledge. Jn comparison witht the Thar Desert in India, the Chihuahua, we believe, is more harren of plants, though the Thar is larger and more important. The criterion of hiuan habitation as a meastire ot aridity may break down where relatively small areas like the Chihuahua are concerned, because their lack of development is not such a serious economic loss to the country as the neglect of larger areas would he. REFERENCES 50 Brnsown, L.. and Darrow, BR, A. 1944 A Manual of South-western Desert Trees and Shrubs. Univ. Arizona Biol, Sei. Bull. No. 6, Tucson 51 Harsupercer, J. W. 1911 Phytogeographic Survey of North America, 645. New York 44 Koppen, W. 1931) Grundriss der Kitmakunde. Gerlin 52 MacDoucar, 1). T. 1908 Botanical Features of North American Deserts. Carnegie Publication, No. 99, Washington 53 Suseve, F. 1942 The Desert Vegetation of North America, Hot. Rev., 8 (4), 238 St Suereve, P. 1944 Rainfall of Northern Mexico, Ecolagy, 25, (1), 105- ill H. THE THAR DESERT This region is described hy Blanford (58) as that part of the great Indian Desert which has dense areas of parallel sand-ridges. The great Indian Desert iself, he states, is by far the miost important accumulation of blown sand in Tha; it is situated in Sind and Rajputana and extends roughly from Karachi and the Rann of Cutch, north-east almost as far as Delhi, and is bounded on the north-west. by the alnvinm of the River Indus and its tributaries, anid on the south-east by the Aravalli Hills. The sandridges of the Thar form two important groups converging on the Rann of Cutch, one group extending due north from the Rann and haying the indus on its west and north, the other runuing north-east front the Rann with the Riyer Luni on its south-east side. The latter area of sandridges consists of smaller and more irregular cures, The dunes near the Rann are highest of all and are quoted by Ilanford as having an elevation of 400-500°; the areas covered by dunes. are of the order of 30 miles wide. The other parts of the Indian Desert have more isolated and irregular sandhills. As these long, parallel dunes of the Thar Desert ate considered by Madigan (37) to be closely comparable to those of the Simpson Desert, though of lesser extunt, it stems worth while to examine their vegetation, The fullest account available was that of Blatter and Ilallberg (55), who made a journey in the eastern Thar in October-November 1917, through the lesser of the two mai sandridge areas, Their rotite formed a square, each side measuring about 100 miles. At the corners of this square were Jodhpur, Bap (just north of Phalodi), Jaisalmer and Barmer, There is a railway between Rammer, Jodhpur and Phalodi. The other part of ithe journey was made on camels. 22 A. vegetation map of India has been prepared by Schiokalsky (59) in 1932, in conmmeéction with her work on the soils of that country and her soil map. Accord- ing to this, the route of Blatter and Hallberg lay mostly in the types designated by her as (1) “Arid sandy desert and barren land,” and (2) “sendy desert wits thorn’; about one-quarter of the route lay in type (3) “steppe desert with thara”, The “thorn” formation is not described in detail, We may ussume, then, that latter and) Hallberg really travelled through some of the most arid sandy country in the Thar Desert and that the vegetation they describe is relevant to the present study. Vaughan Cornish (36) states that the region is well surveyed and that ell the plans slow these parallel sandridges. Madigan (37) reproduces a similar survey map with parallel ridges for country adjacent io the Simpson Desert. It seems unlikely thal country as arid and uninhabited as the Simpson Deser| would come within routine land stirvey work, and it is inferred that the Thar is a less im- hospitable region than the Simpson Desert. Meteorological data for comparison are scanty, but it is clear that though the Thar Desert is without streams and has a most erratic rainfall, it still supports sparse settlement with scattered villages aiid aulmerons flocks and herds of camels, goats, cattle and sheep; there are also shrubs and bushes and occasional small irees subject to grazing. The vegetation, therefore, is probably more comparable with the Australian semi-desert areas which also have parallel sand-dunes in a milder way- Blatter and Hallberg discuss the climatic data available in the arca visited by them, They quote a year with no rain at all recorded for certain stations, and another occasion on which 10” fell in one day at Jodhpur; 1917 (the year of their visit) was tinusually wet with totals ranging From 20-40” in different localities, as compared with average values, caleulated over 40 years, of 7-13". The mean nunber of rainy days is from 13-18 per annum; 1917, of course, was much higher. ‘The wet season is during the summer monsoon from Jute to Sep- tember inclusive; the cold season is from November to March and characterised by extreme temperature variations, often falling below freezing point at night, The relative humidity is always low, atid vety hot and violent winds with sand- stoims are frequent in April, May and June, In Australia, the Simpson Desert is in an area of the continent within the 5” atmual isohyet (42) and with a high variability. Prescott (41), in his analysis of the climates of the Australian deserts, uses the Meyer ratio—P/s.d.—, cal- culated month by month, to define areas of successive aridity; the area having lowest values coincides with the Simpson Desert region where it is between 1 and 2 for the months of greafest valiie, never higher; in the major part of the surround- ing truly desert region this monthly value rises to a maximum of 4 or 5, The monthly values for P/s.d,, calculated for Pachbhadra from data given by Blatter and Hallberg in connection with their journey in the Thar Desert, ate as follows :— Jan. - - O84 May - 0:58 Sept. - - 448 Feb. - - 0°35 June - - Wi Oet, - - O08 March - O17 July - - 8:08 Nov. - - O12 April - 0-09 Aug. - - 10:03 Dec, - - O20 Note the high values for July and August. Hosking (37) has prepared a map showing the values for the Meyer ratio (calculated from mean annual precipitation and mean aanual satucation deficir) over the whole of India. The lowest values occurred over the Indian Desert and here he analysed a small area in more detail,@) using the monthly instead of 2 ) This more detailed map is unpublished, and quoted by kind permission of the author. 23 annyal values. for the Meyer ratio, as we have done, and drawing the isologs of the maximum monthly value occurring in the year, as Prescott has done for Atis- tralia. Hosking found that there was not a close correlation between the lowest values of the Meyer ratio and the distribution of the sand-dunes of the Thar Desert. There was au area below 2, more than 100 miles across, on the right barik of the Indus around Sukkor, whereas the Thar Desert lies on the left bank of this river. The values for the desert itself are from 3 to 6+ (Ilosking) and as high as 10 (our own dala, table above). As far as this criterion of deserl chmate goes, not more than the north-western half of the Thar is comparable with the desert regions covering most of the centre of Australia, and none of it ts quite so arid as the Simpson Desert, unless the steppe west of Sulkor be included in its borders. In view of this, comparison of the vegetation of the two sandridge areas loses Some of its interest, However, a short description of the plants recorded by Blatter and Hallberg will be given here, and it becomes significant to note that their visit touk place in October-November just after the usual period of heavy rains, and m a phenomenally wet season. They may therefore have seen the vegetation in a better state than usual (as. did Madigan on the Diamantina), though no ephemerals were found and few atmnals. With this in mind, an attempt will be made to extract from their results what plants actually grow on the sandhills of the Thar Desert. It appears that whenever the dunes are suft- ciently stabilized, they can support quite a dense caver of vegetation. Most of the plants are deep-rooted shrubby perennials and are listed below. After this examination, one 1s drawn to the conclusion that the environment of the south-eastern Thar Desert, at least, is by no means as arid as that of the Simpson Desert. PLANts GRowinG on tHe Dunes of tne Trar Desert (ret. 55) x = Widespread and important plants < Calotropis procera R.Br. (Asclepiadaceae)—One of the earliest and most hardy colonisers of hare dune slopes; a quick-growing coarse shrub of catholic taste, JJunes may have a pure society of this plant. x Indigofera argentea Burm. (Leguininosac)—Another typical dune pioneer often forming a pure socicty; it has a low ctown of horizontal branches near the sand surface. x Crotalaria burhia Hamilt. (Leguminosa¢)—The most abundant plant of the region, following close after the above two pioneers and replacing them; shrubby. x Leptadenia spartium Wight (Asclepiadaceae)—A common dune shrub, usually a secondaty arrival and rarely dominant, Acrua pseido-lojnentosa Blatt. and Hallb,, and A. tomentosa Forsk. (Amaran- taceac), These grey-white shrubs grow with Crotalaria at the edge of dunes and on sandy flats, which probably recall the Australian bluehush and saltbush plains in appearance, x Calligoniun. polyganoides L. ( Polygonaceae)—A large bush up to 5-6’ diameter, occurring chiefly on sand, but adaptable, showing a preference for dune crests which it may monopolise; roots often exposed. x Cyperus arenarius Retz. (Cyperaceae)—-Establishes itself in thick patches on the lee slope of dunes; this and other Cyperactae are the most effective sand stabilisers im the region, on account of their creeping rhizomes, x Panicum turgidwm Forsk, (Gramineae)—An early coloniser with the shrubs. 4 . ¥ “ 24 Citrullus colocynthis Schrad, (Cucurbitaceae)—An early coloniser. Farsetia jacquemontii Hook. f. et Th. (Cruciferae)—An early coloniser. Penmisetum cenchroides Rich, (Gramincae)—A common sand and dune grass. Eleusine sp. (Gramineae )—An early coloniser, Capparis decidua Vax (Capparidaceae)—Occurs on Hat sand with Aerua; a shrub or tree. Among the commonest sand grasses wete Lvagrosis tremula Hochst., EZ, ciliaris Link, and other species; Cenchrits catharticus Del. and Pennisetum prienrtt Kunth. Breweria latifolia Benth. (Convolvulaceae)—A typical sand shrub. Small Convoelvulus spp. and Polvgala spp, occur between the bushes, The following may be mentioned among other plants occurring — Lyctum berbarina L (Solanaceae)—A shrib at umes conimoii on dines. Boerkavia diffusa L, (Nyetaginaceac)—A sniall trailing plant also throughont Australia. Aristida, two species (Granineac), Haloxylon salicornicum Bunge (Chenopodiaceae )—Abundant locally, but pre- fers gravel: grows in sand-holding clumps. Cassia obovata Collad, (Leguminosae)—Locally abundant on dunes. Zisyphus volundifolia Lam. (Rhamnaceae)—A prickly shrub or tree. Cistauche tubwlosa Wight (Orobanchaceac)—A root-patasite growing ot) various shrubs- Tamarix articulata Vahl. (Tamaricaceae)—Collected, but apparently not common, (Synonyms TP. epliylla, L.. YT. oricntalis, Vorsk., T. articulate Vahl). This is the evergreen Athel Tree now widely used for dry, sandy areas in South Australia. REFERENCES 55 Brarrer, E., and Hatreera, F. 1918-21 The Flora of the Indian Desert (Jodhpur and Jaisalmer), Journ, Bombay Nat. Hist. Suc., 26, 218, 525, 811, 968 et seq.; 27, 40, 270, 506 et seq. 56 CornisH, VaucHAn 1897 On the Formation of Sand-dunes, Geog, Journ. 9, 292 et seq, 57 Hosxine, J..S. 1936 The Ratio of Precipitation to Saturation Deficiency of the Atmosphere in India. Current Sci., 5, (8), 422-423 37 Maviaan, C. T. 1936 The Australian Sandridge Deserts, Geog, Rev. 26, 205-227 58 Mepricorr, HW. B., and Brawrorp, W. T. 1879 A Manual of the Geology of India. Ist edn., Caleutta. See Pt. I, p. 436, et seq. 41 Pruscorr, J. A. 1936 The Climatic Control of the Australian Deserts. Trans. Roy. Soc. $, Aust., 60, 93-95. 59 ScuoKxanscy, Z. J. 1932 The Natural Conditions of Soil Formation in India. Dokuchaiev Inst. Soil Science. Acad, Sei, U.S.S.R. Leningrad. 60 Tayvitor, T, GrirritH 1918 ‘The Australian Environment, Melbourne SUMMARY In order to assess the comparative aridity of the Simpson Desert in relation to other important sand-ridge deserts of the world, a study has been made of the veretation of seven of them. It is considered that vegetation is ihe best climatic index available for regions where quantitative meteorological data are generally so few, The results of this study are briefly summarised in the appended Tables I, 1, III, and IV, in which the deserts in question have been placed in order of aridity. Some doubt is still felt about the correct position of the Mexican 25 Chihyahva Desert in this list; it may be more arid than the Thar, but it ts cer- tainly less important on account of its much smaller size. Approximate geo- graphical position, altitude, and size of the deserls are given in Table I]. The degree of plant cover, the dominatits and the state of habitation are summarised in Table [1I. Table IV lists noteworthy genera in each desert. It is clear that the first four deserts in this list form a very arid and barren group, and that the Jast four, including the Simpson Desert, are much better coyered ; some of them, like the Thar, have much of the character of a steppe. A comparative study of the availability of tuderground water would go far to explain this; it is also certain that some of these deseris enjoy a far more regular rainfall than others in which dry periods of a iew years are the rule, ‘The stretches of absolutely bare dunes in the Great Libyan Trg and the Talla Makan seem to be unique, ever. among the deserts of the Old World. The plants actually present are mainly perennial shrubs and psanyuophytes, occasionally reinforced by ephemerals; the Tawiarix and Populus of the Takla Mukun are tree types of stunted growth, while the Kara Kum achieves regular forests of the small saxaul trees. Each of the regions shows considerable indi- viduality in its flora, particularly the Mexican and Atistralian areas, ag one might _ expect; there is a much closer degree of similarity between the more cortigious deserts of the Old World. Attention nmust be drawn to the importance of a shrubby grass in three ef the deserts—Triadia in the Simpson Desert, and two species of Aristida, respectively, in the Western Sahara and Kara Kim, For a detailed account of the vegetation of the Simpson Desert, the puper i this series by R. L. Croeker (35) should he consulted, as well as the list of species collected on Madigan’s expedition as reported upon by the writer (36). As far as this study is concerned, special habitats like wadis and water- courses have not been considered. As they have a denser and often distinctive vegetation, attention has been fucused on the dune areas. Where possible, meteorological data have been included, but they have proved quite inadequate for comparative purposes, except in the case of the Thar Desert. It is interesting to note that before 1925 the Libyan ire, the Rub‘ al Khali and the Simpson Desert were practically unexplored. Our knowlege of the Takla Makan and much of the Western Sahara gots hack only to about 1860, thowgh the primitive method of camel travel was used in all the recent snrface explorations, except in the Jabyan Erg where Bagnold and Clayton used motor cars. In conclusion, it muy be inferred that the Simpson Desert, the worst of the Australian sandridge deserts, conrpares in size with the most important dune areas of the earth and, though arid, is exceeded in this respect by the Libyan Erg, Takla Makan, Rub’ al Khali and Western Sahara, but not by the Kara Kum, Chihuahua and Thar deserts. ACKNOWLEDGMENT A. debt is due to the late Dr. C. T. Madigan, who tread the almost completed paper before lis death and made important suggestions for improvement. The section on the Simpson Desert was nat completed at that time and he contributed much metcorological data for ik. 26 a0 WS" E youeyw ur “xe ("ur 1Z) 2-8 ypjepy ur “xe wird Splodat sieac ¢ “Wd 0-Z41— FZ —asuey (Ud 6E+6) Let yaaeyy ol “xeur (ura OF) 42-9 Judy ur ‘xeut (rd 6) °F Widy ur ‘xe were Ppitoszaz sivat | AckOT “ety A.c8 Ase “une Hold A[uo spaczaa savak ¢-7 Ac03 (9.08) 4.98 JOYS Sps0dezy (9.08) Ao@zl “XE PodArIIsSdO (9,0€) 4.98 48S (90-62) Tob8 (9.0%) T.r0T ‘Xe “AV (Dof9Z) 41.08 ATof8 A.cé (D0$-0) Tog? Un “AW (Dof:L—) ToS (D.8-1) Hose (9.92) A.St- “ULL “AY. (9c6-S—) Aol Acti (414195) Lz6 ‘uel (9.$-ZZ2—-) 4.37) [peyures jenuue UeEsTy Ayu 407 Ayenuef 107 ainjeigditay uBayy SLUASAAC NYMVW VINVL GNV WOM Vava AHL YOR VIVO SILVAITS AWOS "Uy UT [aAay vas aaoge epnywqyy uiseg Wey uy UexePY BPEL JO aspa ysoAy }iasap ay} ja a2pe ay} UO AIO JO JsaM SoTL NZ yoyoday JO ysaM-YINos say YOT jaasap aunp-pues ay} UL sIseO uy waynes yaesay pues Ws'sn ‘mxQ 2} Jo yynog — saypru 0S 3Osap eunp-purs JO daju39 UT yereday Jo "A'S SIU OST snxQ JaaARyT aye uC BAG [ely oy} JO saytu OZ UNM snxQ TANT ay} UO UOTENyTS XIGNaddy — J] FMV, (Maipuay) pupyle yz (Maipuayz) Jesyseyy (uaddgy) peqeypsy (“31puay ) AW, A.€9 ‘suoy N.$'8€ 3°T (Sysuerqnog ) yajyaday (uaddoy) PLM (uaddo x) snynN (Marpuay) YSAOAPURKITY-OF aq eyep jo adinas pe woes yasaq ENYENYyIY ay} Jo Ayplie Jo aoi#ap saneyar ayy jnoqe 79] [INS st }qnop swag y | (BIpuy 3saM-Y310N) Ss2] 40 {000° SoTU OFZ * OOF HoSL—~ Ho69 No6é~ Nove UL ‘H (091 ) (€S) aaoays vaae 3unp ,000°S — ,000'E Sayrat QOT * OOT Mol0I— M.90T N.d&— Nol€ pue pues ayy yenyenyiy “5 Saqit OSz X OOS (BIsy Uio}sa MM) S88] 40 ,000'T ‘Quofe Wny Ey 1.89 — oss NoSt—N.9€ Wny Sty pueuny wey (ueaipeyy) (uezipey) Sapa “bs Qoo'9g eeay (ueBIpEyy) (eyeajsny eau) 000'T OF [aAaT-eas WoL] SOUT NOE * OOZ AHobfl — HoSEl SaL2 7 Sof% qyaasaq] wosdung ~y Jajyeus S194 *(datyNer)) syvad raysty Teyuapio9g, pues [eWIIOH r —«sHPAD_ PU YON'9 03 surejUNOW ‘S812 years 94}, JO yova Joy (ediry YON) “yy 00ST #ojaq ATS0}y SeTTur QO¢2 * OOT | aeSE™ Moet | Nob€ —NoST BIEYES U19SIMA CL (€z) Ted | satu ‘bs Qgo'osT BV (erqery w1ayynos) 000'% — 002 Wo1z Saya OOS ¥ 00 Ho9S — Dosh N.€2~ No8l TeuM IF Gnu “OD (eIsy yeaqsD) 000’ eaoqy Sa] 006 X OOF H.£6 — AoBd Nol? Node UexeW, BRL ‘A (RaHFY 4340N) 0ST Mo1Dg saqrur 00f X 00T Ho8Z— A.SZ N.62— NobZ Sry weg *V (Aupire (9293) (stoqommey) | spngauay aPrane’y jO Japio azyeutxoidde uw) SPRY Pet HORISC GF ’ , | ' $}1989q7 (4ju0 ajeuxoidde saingy [Ty) Gassnosid SVANVY LUASAG JO AAVWWNS XIGNIddy = — J] aTAvy 28 Tarte IT] — Aprenprx SUMMARY OF DESERT VEGETATION Great Lisvaw Ere or SANn Sea: Paratlel sand-ridges complete’'y bare. Rare bushes of Ephedra to be seen. There ate inthcations that very localized ephemeral patches of Arisiida. may spring up alter rain, which falls possibly once in eight years. Entirely unitthabited. TaKLA MAKAN: Paral’el sand-ridges completely bare. Peripheral regious of the desert have solitary Tamarix, holding mounds or “cones” of sand, and almost as hardy is Populus, the “toerak’! ot desert poplar, These occur as miserable shrubs or a amass of dead uwigs; near the rivers they form dense groves, Rare patches of a small reed (Lasiogrosfts) occur Te the dunes, Ephemerals absent. Entirely uninhabited. Rus’ av Kriatr or Empty Quarrer: A varied terrain inctuding bare parallel sand-ridges. Stretcbes of, say, 20 miles without vegetation, either dead or living, are rare. The most important plants are the widespread “Ahal” (Calligonum) a large shrub, and “Fladh” (Salsofa) a stall shrub. A few other pants are frequeit and there is, with Jitcle doubf, an ephemeral vegetation after the infrequent rains. Penetrated on the borders by nomad camel herders in good seasons. WHSTERN SAHARA: A yatied region including some large dune-ateas or ergs. Some are humid and proyide some waier and grazing far flocks and herds, others are bare and dificult to. penetrate; veetation probably inchides the perennial “drinn” erass (Aristida pungens) and shrubs of the genera Cornulaca (Chenopodiaccae) and Calliqanum (Polygonaceae), After rains there may be a rich ephemeral vegefatian, chiefly of cructicrous herbs. Most of the ergs allaw of cecasional nomad camel srazing at the edges, at least. Srmpsan DESERT? The whale area covered with loug, parallel dunes, qitite well vexctated with shruhs and occasional small trees, Planis are thickest on the lower dunes and corridors between, where they are ustialy some feet apart. Lhe crests of the dunes ure, for the most part, relatively bare and unstable, though comp'etely covered dunes occur in patches. Trivdia Bascdowti, more or less alive, is quite the most conspicuous bush of the corridors and lower slopes (Porcupine Grass or falsely called “Spinifex’). True Spiutfer paradoyus (now Zyygochlow paradoxa) is the sparse shrubby dominant of the ridges. ‘There is an assortment of associated perennials, annuals and ephemerals, some attaining Jecal datninance. Uninhabited: water is lacking. Kara Kum: Chiefly saud-desert with bare, paral'el dunes bordering the tiver. They are unstabilized, but the large grass Aristida penneta var. thay occur on the Jower slopes, with some chenapndiaceous shrubs in the hellows. The barchans often become well covered with yegctation, forming shrnliby sandbilis and eyentually dwarf forests uf Saxanl trees, The yuuctation is varied and largely endetnic. An ephemeral flora oceurs fin spting, Shepherds and their flocks make sporadic excursions fram the casts and tiver valleys, and the dwarf forests yie'd firewood. CHIHUAHUA? The sna‘test of the deserts considered here, including same parallel dune areas atmost bare af vouetation: what occurs is shrubby, sparse and irrecular, Ephemeral vegetat on prohably present from time to fime. Degree of habitation nat known. The vegetatian seems lo be modified Mesquite (Prasejis) torrimtion, dominated by Palianintha CLabratee), Taar? Taree parailel sand-ridge areas are present, probably even better yegetated than the Mara Kum, with a variety of shrubs end some small trees. Sparsely inhahited. with ceattored yil’aees and numereds flocks anrl herds of camels, nats, cattle and sheep, There are commrehensive land sutvey mans. of the reo’on available, Important genera are Culotrehis, Crotalaria, Calliganun, Cyperus, etc. 29 Tarte LY -—— APpPenpdix Summary cf Chief Families and Genera represented A. Linyvan Ere Pphedra (Gnetaceae) ‘braces oi Gramineae B. Taxita Maxan Tamarix (Yamaricaceae) Populus (Salicaceae) Lastogrostis (Graminea) C. Rus’ at Kuarr Salsola (Chenopodiaceae) Cornulaca. (Chenopodiaceae) Calligonum (Polygonaceae) Dipterygium (Cruciferae) Cyperus (Cyperaceae) Fagenia (Zygophylaccac) Tribulus (Zygophyllaceae) D. Western SAHARA Aristida (Gramineae) Cornulaca (Chenopodiaceae) Calligonium (Polygonaceae) Savignya (Cruciferae) Moricandia (Cruciferae) Malcalmia (Cruciferae) Launea (Compesitac) Pianfaga (Plantagmaceac) Ephedra (Gnetaceae) E. Simrson Desert Triodia (Grainmeae) Zygochloa (Spinifex) (Gramineae) Acacia (Leguminosae) Cassia (Leguminosae) Crotalaria (Leguminosae) Hakeca (Proteaceae) Grevillea (Proteaceae) Lxemophila (Myoperaceae) Sida (Malvaceae) Hibiscus (Malvaceae) Portulaca (Portulacaceac) Salsala Rhag odia (Chenopodiaceae) Kochia . All present ta some Bassiit extent Atriplex Enchylaena FF. Tara Kum Aristida (Gramineae) Agrinphyllum (Chenopodiaceae) Arthrophytum (Saxaul) (Chenopodiaceae) Corispermusm (Chenopodiaceae) Foraninowia (Chenopadiaceae) Salsola (Chenopediaccae) Calligonum (Polygonaceae) Smirnovia (Leguminosae) Ammodendron (Leguminosae} Acacia (Leguminosae) Eremosparton (Leguminosae) Astragalus (Leguminosae) Carer (Cyperaceae) G. Carnuanus Drsert Poliemntha (Labiatae) Yucca (Liliaceae) Prosopis (Mesquite) (Leguminosae) Dalya (Leguminosae) Aymenoclea (Campositac) Senecio (Compositac) Chrysathamntius (Compositae) Sporobalus (Gramineae) Andrapegor. (Gramineae) H. Tuar Desert Calotropis (Asclepiadaccae) Lepladenia CAsclepiadaceae) Indigofera (Leguminosae) Crotalaria (Leguminosae) Aerua (Amarantaccac) Calligonum (Polygenaceae) Cyperus (Cyperaceae) Panicum (Gramineae) and others (Gramineae) SOUTHERN AUSTRALIAN GASTROPODA PART III By BERNARD C. COTTON Summary Shell pyriform, with flattened base; base spreading, with a sharp edge round the entire circumference; dorsal part of the margin, on the right side, anteriorly and posteriorly obscurely ribbed; the left margin edge medially wavy; dorsum sharply humped, the central portion of the dorsum in the holotype mostly white through the outer layer of shell being dissolved away with hydrochloric acid; normally the dorsum is light brown as shown by patches of the original outer layer left in other specimens which have also been treated with acid; margins and extremities calloused and spread; medium brown-coloured base becoming cream towards the lips at the aperture; aperture wide and curved; outer anterior lip declivous; aperture turns sharply left towards the posterior end; twenty-seven teeth covering the whole length of the aperture; teeth moderately developed, produced slightly across the base, rather coarse and widely spaced, brown with cream interspaces; fossula white, slightly concave, sulcus wide and shallow, neither denticulate; interior of shell, examined in a broken specimen, ivory white. Length 49 mm., width 35 mm., height 25 mm., animal unknown. Rit] SOUTHERN AUSTRALIAN GASTROPODA PART Ill Ry Bernarn C. Corton * [Read 8 April 1948] Two new species and one new gents of Gastropods from south-western Australia are described here Zoila rosselli, sp. nov. 1. I, fig. 2-6 Shell pyriform, with flattened base; base spreading, with a sharp edge round the entire circumference; dorsal part of the margin, on the right side, anteriorly and posteriorly obscurely ribbed; the left margin edge medially wavy; dorsum sharply humped, the central portion of the dorsum in the holotype mostly white through the outer layer of shell being dissolved away with hydrochloric acid; normally the dorsum is ight brown as shown by patches of the original outer layer left in other specimens which have also been treated with acid; margins and extremities calloysed and spread; medium brown-coloured base becoming eream towards the lips at the aperture; aperture wide and curved; outer anterior lip declivous; aperture turns sharply left towards the posterior end; twenty-seven teeth covering the whole Jength of the aperture; teeth moderatcly developed, pro- duced slightly across the base, rather coarse and widely spaced, brown with cream interspaces; fossula white, slightly concave, sulcus wide and shallow, neither denticulate; interior of shell, examined in a broken specimen, ivory white. Length 49 mm., width 35 mri., height 25 mm., animal unknown. Loc—F¥remaatle, W. Aust. Remarks—The species is constant in character, size and shape. Only three distinct species and four subspecies of the primitive cold water cowries belonging to the genus Zeila are known, and they are found in Western and South Aus- tralia. There are four Tertiary species in the Miocene of Victoria, 2. consobrina McCoy 1877, Z. platypyga McCoy 1876, 2. simplicior Schilder 1935 and Z. toxorhyncha Tate 1890. None of the Tertiary species resembles the Recent one here described. Probably Z. rasseiii is more closely allied ta Zoila decipiens Smith I880 described from North-Western Australia than to any other species, but if is distinct. Z. rosselli differs in the Aattened base with its spreading and sharply-edged margin and coarse spaced teeth of the columella and outer lip. Curiously, Z. rosselli has the general shape of Siphocypraca aus Linne 1758 Mediterranean, and of Bernaya tenlerei Cazenavette 1846 Arabia, and Bernuya fultoni Sowerby 1903 Natal, but these three species have well-rounded bases, and Bernaya is edentulaus. Syphocypraca has a complete set of spaced teeth ou the margins of the wide aperture, a feature which in some way recalls Z, rossellt. The holotype specimen D. 14220 is figured in five different positions, and another broken specimen belonging to the series is figured to shaw the interior structure, Mr, Harold Rossell, after whom the new species is named, took the four specimens before me at Fremantle from a beach near North Wharf just beyond the wall where all sorts of rubbish come ashore, The bucket dredges dump their contents straight out to sea opposite this little beach, and the shells may have come from the bottom of the harbour, He writes: “I remember small stones, some with shells stich as Turbo setosus stuck in them in a partly fossilised state... — | often picked up immature Cypraca scotti on this little beach, but of course scath were *South Australian Museum, Adelaide. Trans. Kay. Suc. S. Auats; 72, (1), 23rd August, 1948 31 and some actually had the remains of their animalg in... . If 1 remember, the time would be about October, after the equinoctial blows of September.” Mr. Rossell also remarks that “it must he a long time since | took them, . . . a! least thitty years ago, | remember it very well, atid the dorsum was pale brown, somewhat marked with tiny scratches and cleaned at once with the application of hydrochloric acid. 1 imagined then that it effected an improvement in the shell’s appearance, but | was just beginning to collect things, and had it been recently I would have known better.” The only specimens known are six taken by Mr. Rossell, four of them in the W. R. Steadman Collection. one of which, the holo- type, described, was donated by Mr. Steadman to the South Australian Museum. One specimen from Cottesloe is in the Australian Museum, actording to informa- Hon recetved, Aleyna acia sp. nov. Pl, f, fig. 7, 8 Shell small, conical, turbinifarm, solid, protoconch finely spirally Guar- rawed and blunt, of oné and a half whorls; first spire whorl conyex, closely lirate with seven close-set cords; third convex, bicarinate, upper carina at about the centre of the whorl, lowet immediately aboye the suture: ahout ten stout roundish axials from tlie suture cross ard form somewhat vertically compressed nodules on the upper carinae and on the intercaritial lirac, least prominent on the medial Jirae, more prominent on the lower carisa and largest on the upper; round the base are ten rather broad, rounded spirals: mouth nearly circular, labrum bevelled on, the inside, slightly effuse at the base, hollowed on its surface, without any spreading callus; columella with a prominent tooth which is actually a plait; a broad band of rosy tint winds round the base to the basal lip and extends between the tubercles of the lower carina and so ean be seen as spots just above the suture; in some speci- mens a narrow ted band appears between the main axials just below the stitute and in some a series of dashes extends from its inner margin to the base nearly to the aperture of the shell; the lip is round, smooth inside, with a slight gutter behind; the body whorl has seven or eight equidistant raund tibs of sizes equal to the interspaces; some have axial red lines eywal in width to the white interspaces, sometimes broken into articulated lines, especially about the periphery of the hody whorl, much less over the base: the first and second whorls may be red and the body whorl with seven or eight streaks of red with equally wide streaks of white from the suture to the periphery, and coral red base; some have the first three whorls red. the next two whorls, the ultimate and penuttinate, with about six zig-zag Lroad brown flames and rather narrower white interspaces extending to a brown hase; one form has the two carinations articulated with brown, the rest of the shell being light chestnut brown, the first and second whorls being rosy red, Dinmeter 1-8 mm.; height 3 mm. ‘ Lov.—W, Aust.: Rottnest (type), King George Sound, Mopetoun. S. Aust.: Cape Korda, 60 fathoms. Remarks—The species 13 bicarinate and more strongly sculptured than Alleyna australis Uedley 1907, from Mast Head Island, 17-20 fathoms; the columellar tooth is less developed, and the colour pattern is different. fIolotype: Reg, No. D, 14219, South Australian Museum. Austroliotia gen. noy, Genotype: /iatia botanica Medley 1915, New South Wales. Shell depressed, moderately heavily sculptured, aperture trumpel-shaped, not 32 strongly variced; umbilicus wide; operculum horny, multispiral but with faint traces of granules. Distribution—New South Wales, Victoria, South Australia, Western Aus- tralia, Tasmania, New Zealand. Remarks—Liotina Fischer 1885, genotype L. gervillei Defrance, a fossil, has becn tsed for this temperate Australasian genus with widely umbilicate depressed shells, with moderately variced aperture and simple, multispiral, horny, operculum, Liotina is more applicable to a well marked tropical and subtropical group of species, with extremely solid, not very depressed shells having a very narrow cylindrical perforation. The genotype, 4. botanica, typical of a southern Aus- tralian series, seems to be a temperate relative of the warm water Liotina, Austroliotia is separable from Munditia, in which the shell is more planorbid with a tendency to reduction of sculpture to knobs on the double keel, and having a very wide perspective umbilicus, lightly variced aperture and simple, horny, multispiral operculum. Species belonging to Austroliotia, besides the genotype, are A. australis Kiener, A. densilineata Tate, and others may also belong here. Key to Species oF AUSTROLIOTIA a. Spiral lirae defined be = ses sole = ne we australis aa. Spiral lirae weak ine Ws oe seve 23" E. leucoxylon red-brown earth, podsols, 19-23” Savannah skeletal C. stricta woodland skeletal, brown solonized 14—-10" E. odorata skeletal, brown solonized 14—19” L. wultflora-L. diva Savannah skeletal, red-brown earth, 103-18” browt! solonized, desert loam FE. anceps-E, dumosa browf solonized 12-144” E. brachycalyx-R. oleosa\ mallee scrub brown solonized, skeletal 103-12” E. alcosa-E. qrucilis brown solonized, skeletal §-104” M, flatycarpum-K. sedi- tree-shrub brown solotized, desert - x ee 1 ee, eT » 22” Drosera pygmaca- - = = ., oF Lomandra fibrata - - - - tam wt Sf Mf ff ob, odd id ok Oe! = Asclepias rotundifolia - - - = & he bb &@ & 3 — Iris germanica - - = - - x ae ee ee eh eo “a a Rumex acetosella - - - - = x o an 4 aa? Daviesia ulicina = - - ~ = Mame “A app aap oo e a we ee = Acacia rhetinodes - - - - ex & | &£ Fate & & & § ~ (2 Dillwynia floribunda = - - - rp hme ge ee ast PP it “e Cidcs i“ BE Thomasia petaloclayx - - - me 6 of RR oe tw eet _ Trifolium subterraneum - - ae ee a So Sa » 2l” (Mt, Barker strain) Rriza minor - - - - - PER + + ye Ee ye oe Oe Ue OS » 20” Velleia paradoxa - = - = ns CN ae eS old pees Rae ded aly = Bromus mollis - - - - a ee ee +f 3 2k” Geranium pilosuni 2 e - - x x i » of fF M of oe » 19” Geranium molle - += -— = ie a a Se So» Oe = » 22” Drosera whittakeri - - - OO £ 8 one woo 5 € 8 a Last Trifolium glomeratum See odd + e+ + oe we nm on & » 20" Stipa drummondii - - - a a » 19” Rosa canina - = - E - ck os Oe oe os Sah 2 4 «619 Acaena ovina - - - - oS ey ee hg ME CR: « 18” Halorrhagis ciata - - - - x x ¥ + wk ik kee » 17” Kennedya prostrata - - - 2 te He Ke or we ow hf oof oats » 17” Scirpus sp. - - - - - er A ah nae the ew 4G de ey eg _— Aira caryophyliea - - - - x * Fs we ie » 17" Briza maxima - - - = > x x x Ses eer » 17” Ranunculus lappaceus - - - See ee Se ee Se . o 8 _ Scleropoa rigida - - - - - «x KX + Ke es ee eh _— Cheilanthes tenuifolia = - - - sx MP ok cee ow o@ ye 4 7 » =«15" Casuarina stricta - - - - tPotd es) oy Fh oe me neh ob oe note wh » «14" Pimelia glauca - - - - “x &© © &© & e 2. 2» & 2 G » 14” Trifolium angustifolium - - - XxX YM Rw ww - ® » 16" Hypochaeris radicata - - - - x x ox x + eh ert be = y: » 18” Trifolium arvense - - - “x x xX x «x i - # » 15" Themeda australis - ~ - ‘xe x - Sy see ot é » 16" Bromus madritensis - = - - «x «* - Ox ye » 1?" Chenopodium carinatum - - - x x x* x*« ® . ‘i —— _ Lolium subulateun = ~ 4 - «x * ee et oy 3» 13” Moraea xerospatha - - - “x x x & x is % i y» «15 E 64 1) (2) (3) ¢4) (8) (6) (7) (8) (9) (10) C11) Limit of rainfall Onopordon acanthium - — - ~- =—- x x * K K 2s 2 6 6 4 wy 2" Helichrysum apiculatum - - = & 2 Me Se Be Ho » e 6f — Avena fatua - - - - - - x x xX xX * xX «© © «© © «© » 12” Anguillaria dioica - - - - - x x xX x * &% © © «© © © — Cynodon dactylon - - - a ee ee ee ee a a — Oxalis cernua - - - - ~ x x K xX XK X X © © © « » 14” Cryptostemma calendulaceum - ex K * © KX K KX ws mw + » =12” Vulpia myuros - - - - =“ * *£ x © NK 6+ € 6 wf ue — Lomandra dura - + - . a) x x, 3% x x - . x . , * ” 13” Acacia pyenantha - - - - = x x x x x +» =» w # «© 6 » 12” Bromus rigidus - - - - ~ ¢ x 8 KX * ¢ + & KH & A in Bs Erodium botrys <- - - - - xs Ss xX x* * X& & #8 =X @& 4 ». LO Hordeum leporinum - - - -~ «* % xX xX X X X © xX «= «6 » 12" Salvia verbenaca - - . - 7 ® xR RW KY eH KR © + » 14” Stipa eremophila - - - - =x x x x x «© « x «© & G » 13" Romulea rosea - - - - met: 3 x eG OR Ee Vee Gi oe 4+ yy aat Eucalyptus camaldulensis - - - x x x xX xX xX X¥ « x x * » 23” Stipa variabilis - - - - co x §S © X XX X X * KX + «8 — Danthonia semiannularis - - - x *« x* x x x* * x «© x x » 8" Oxalis carniculata - - - “kk x * & © *& «4 * & & » 13” Vittadinia triloba - - - - - x x x * © x x xs x xX & » 1a" Convolvulus erubescens - - -x © 4 B@ # * « & & B GF » 10” Enneapogon nigricans - - - - x x« x © * x x K &*& XK x — FErechthites quadridentata - - - x * «© &* X& «© «© X &® X X » 4 Wahlenbergia sp. - - - rew x kk! * 4% x 4) 3S ££ K& & » 12” Echium plantagineum - a - tt ix ‘c& £ £ FO * A a. 12" Sonchus oleraceus = - - - x © x* xX © + » x X © » 14" Solanum nigrum - ~ - - = = © x «* © X X& + « Kx * 2 » 14” Erodium cicutarimm - - - - x x * xX X 5» «© *£ X xX X _ Inula graveolens - = - - ~ =e x + KR MR KX YY we Mm XK B » 13" Exocarpus cupressiforinis - - ee ee ee ae a a = Lolium cylindricus - - - mie SM ma Wo yw & & le js a Calocephalus citreus - - - oy om & mo wo» te it of fF te = Eucalyptus leucoxylon yar pauperita . .« . , «© «© © «© © «a 4 et Ulex europaeus - pS . - a a Pe ee ee ee » 19” Plantago lanceolata - - - +#om & w& de tl te sf th Moe Pte = Clematis microphylla - - - Pio “e yt to YH ell ty 44 - Senecia lautus - - - ~ Sel oe ee nt ee ct = Bulbine bulbosa = - - - - - - & x tt ¢ Bie GG a) Acacia armata - - - - ok 8 £ R Be Bw ee Re » 14” Cryptandra amara var - - = =. 4 © & HF 46 w + 2 ee 14-23” Trifolium tomentosum - - - oe a a ee lower 13” Heliptcrum jessenii - - - 7 foe x ew xm © He e + 13-22” Dichopogon strictus - = ~ Se SC SS we HS «+ 4 of Be & lower 16” Cynara cardunculus - - - 7 ee ee ee ee rr — Foeniculum vulgare = - - - ee le a eS > “r Acacia brachybotrya = - ~ - jie oe at Se a. 4D GA Ay 11-21" Bursaria spinosa - - . - -~ 2. XX & x x * XX x «© 2 4 lower 11” Helichrysum rettsiim - - = Sif lf SH kb of ow HH ee —= Triodia irritans - - - - SE Me 3 of tt kl ee OA > Lomandra muitifiora - - - -.+ * * © ¢ € . & . . lower I0k” Aristida behriang - - - - - . © . oe # 2 ew ® wo ow » il” Eucalyptus Jeucoxvloin - - - ses sR S 4,5 SY ey Pf APL) » 18” Sherardia arvensis - - - a Se ae a » 14” 65 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) Limit of rainfall Callitris propinqua - - - 7 2 oo et ek okie «6 3 Yb 9-19" Medicago trunculata - = = - . x x x x x x x x « « fowerl3? Dianella revoluta - - - - Pi A ae hee ee Be cor ck oe oe «ae Rhagodia nutans = - - - - - » * xX X X xX *X X x x - y th” Trichinium spathulatum - - - + &® © X xX & X xX xX x es —_ Polygonum aviculare = - - ~ — § ee 2 EP: + x x « — Asphodelus fistillosuy - - - Bin Te Se ew M 3° Bb) ee be RB ad » 12” QOnopordon acaule - - - - i: i ie oe i a a ee 12" Carthamus lanatus - - - - * XX % BH OX KX & we x + » =~12" Dodonaea attenuata ~ - - - ) & # @ + = ww & x = Enchylaena tomentosa - - - 7 5 6% % XM x x x * S * x upper 21” Kochia_ brevifolia - - - ~ e A OKk Ok ox oe x 8-20" Medicago minima - - - - + x x & x © *% = x x» x lower 12" Euphorbia drunimondii - = - ft, (he x x X xX X x* x x wpper21” Medicago hispida - - - ~ + em * X © KX x x x x x ~~ Jlowerl3” Lepidium, hyssopifolium - - - 5 mM © x * © & «© x x » oO” Schismus barbatus - - = - + MH « ® & xX X X xX =X x upper 17” Salsola kali - - - - - - *8 © X& X* xX © =x X x x & 3 aoe Stipa elegantissima - - - 7 5 x x xX * © x x «© x x Iower 8” Marrubium yulgare = - - - - 4 © x xX x x x x x Xx » 11" Heliotropium curopacum — - - - x *¥ X x x x x x x upper23” Sida corrugata ~ - - - 7 1) © oe Rw = wf RF S & » 20" Scabiosa maritima - - - + - cai ot ocular : = Rhagodia parabolica = - - - See Do Be ee re 11-17" Atriplex semibaceatum e - - s+ © © «. © x & x. . 9-18” Nicotiana glauca - - - - - + # © KR © + K © & x & _ Callistemon terctifolius - - Se ye a os ee ly — Eucalyptus calcicultrix - - - 24 2 «2 +) ga, ry ft — Trichinium alopecuroideum - 9fe vost, Ae... oe: 64 a Acacia spinescens - - = Se See Bee ee ee lower 14” Lotus australis = - - - = 2 43 = yooh se ft 4 3 — Cassytha melantha - - - ~ e & 4 ee 4 & «x ta ft — Lycium ferocissimum = - - - =. - ew wT Rte t tke ws » 10” Rhagodia crassifolia - . = ob fy fh ath 3 le” —z et at 10-18” Eucalyptus odorata . - - -—- ¢ «© & x x XS x x x 14-21" Atriplex muelleri - 4 = - - x xX x « Ko ow. upper 17” Myoporum platycarpum + - oe oe! i a a so ars » 17" Bassia uniflora - - - - ~ 8 5 oe ® eM HM we KX KX K «x » 18” Cassia sturtii - - - - ee i upper 19” Kochia aphylla = - - - - slit bef fe ome a eR gy 11-19” Hakea feticoptera - - - - 4 Xo X + x x x x x upper 15” Acacia colletaides - - = = = © «© *® » x & x x 8-17" Cassia eremophila var. platypoda a a a a a a a n upper 14” Rhagodia gaudichaudiana - = >was A2 4 & x x xX x x »X ar AS" Erodium cygnorum - = - a x x » 16" Muchlenbeckia cunninghamii - ee a i a x * x = Lepidosperma sp. - - - - = Rp uw oe - SS ie — Xanthorrhoea quadrangulata = “uit + & wv & hve —_ Kochia tomentosa yar. humilis - se el elUlelU UG ‘ 7 — Salvia aethiopis - = 3 e -~ & 6 Xo. . - 4 _ 66 (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (13) Limit of rainfall Acacia ligulata - - - - Ree te & Nhae x & x - a Sisymbrium orientale = - ~ - j~ a OY he ee SE We om ‘ os Eucalyptus largiflorens = - Se a rr ee es ae. —* Cymbopogon exaltatus - - a a 38 » ¥ = Acacia acifiacea = - - - - Sg) eth ee te Te ae fe = Eremophila serrulata = - - - a a ae a) a — Cassia artemesioides - - > ee ae es ee a — Callitris glauca - - - - soi i a ws ae + = x L » =45” Codonocarpus pyramidalis = - - tie £ Q 27 FA" g sow ft — Eremophila alternifolia - - ms ey a ae e to ot oe OO — Eremophila oppositifolia - - SP ee ie ks Ce ee es » 10" Solanum ellipticumn - - - ceee er ete a CUE HE OE — Trichinium obovatum - - - a Og ee RE ets FAO V8! Sida petrophila - ~ - - ~ ts PRA es HE KE we S&S FS —- Chenopodium cristatum - - =P gg etal Pup bee a me - Xe ote. By » 13” Goodenia cycloptera - - - en Pk On) CO Pe tt SO 8-13" Cratystylis conocephala - - a Se-§ es me moe Ee oe 6 7-11" Beyeria leschenaullii = - - - A a ee era ee ne ee ae — LLepidium leptopctalum - - ee we ee «a o 2 '§ * sO _— Olearia pimelioides - - - A ee, Ce: ie re — Blennodia cardaminoides - - » . es Fe 2a Sees. So oat o- Centaura solstitialis - - - =~ OM nett te kecre Fy xa tt — Centaurea calcitrapa - ~ - deme cve eee se > A Ree 2 — Carrichtera annua - - - Sm D> bh te we ee Se Re A) — Bassia obliquicuspis a - Soe ae Be ee Be RR Mpper dl’ Sida intricata - - - - a ee ee ee Pa a a: ee: = Atriplex angulatum - - - oe yt se ot it oS OS wp] 2 ae TS » (12" Bassia paradoxa - - - = BO CA OT of Wwe BY oe ee sf Ad" Bassia brachyptera - - - “i & S18 2» © a S eR yy eE Atriplex limbatum - - - — BD up be le OF om cy oe OO — Babbagia acroptera - = - eee lee we oe OS Re mm ot » Tragus australianus - - - ~ ee ow & A ® *F = Paspalidium gracile - - - Pie Koay an | OF Se = Trnprostis dielsit - - - - Hit ag + 4 oe 5 + Se Op » i" Lotus australis var. parviflorus - ee a er Sa: a Cos a » UH" Chenopodium nitrariacewm (sinall leafed variety) - - - - - “oo. - . . 5 $ : . x x x ee Chenopodium vitrariaceuim (large leafed varicty) - - - - - = er er ee a Ge es eee swalnps Craspedia pleiocephala - - - ZS yt «6 w, +, wh ene at Pace 5 — Lepidium fasciculatum - - Slee lb et ole Lay we re’ 6 | EEE = Eremophila maculata - - - aller a ee ee ee ee ee — Marsilea drummondit - ~ ~ iy Se are aval otal com ©. fart) swamps water- Trigonella suavissima - - - —, + % & «w ¢ = B Peeters courses Lavatera plebeja - - - - 7 5 + te & 3 et are Le =~ Chloris truncata - - - - sg Paw ae TR wre = oS oa Zygophyllum iodocarpum = - - ee ee RS ee os 8 upper 10” Euphorbia eremophila - - - Ie. 0'. 4. pts £ 7 A BS xe 8 _—— Eremophila sturtii - - - > HS FP t+ GS “oF “els » 8" Bassia tricuspis = - - - - ae ee eer, a ed Ta — 68 (L) ¢2) €3) (4) (5) (6) (7) (8) (9) (10) (12) Limitof rainfall Stenopetalum lincare = - - - 2 oS @tD te Ste, le x x — Tribulus terrestris - = ~ Sr ee» Beek. Gy ote Bab be — Dactyloctenium radulans = - r Har ap ue we RD et ae ow or Ht =e ny Casuarina lepidophloia - - - Sh Smee OT te ae a ee de kh a 10” Kochia planifolia ~ - - - cl a ee Oe ee ee eee » 10” water- Erodiophyllum elderi = - - - eC CS length - - - 22 -24mm. 24mm. 242mm, 22mm. 20mm. 2Intun. 19mm, breadth ~ — 0-46-1-1 0-90 0-89 <4 fOr (a2 (8s Length of werm 4-4 4-9 deh 3-6 4-2 48 Maximum breadth 8 0-7 10 1-1 OF Oo Proboseis, length 0-65 0-80 0°67 0-75 ego Oat Testes, length - Q+4 O42 0-35 Testes, breadik - 0-2 02 Ord 73 13 Fig. 12-20 Gordiorhynchus hylae 12. portion of male; 13, adult female; 14, proboscis; 15, some proboscis hooks from adult female; 16, proboscis haoks showing changes in form of the rooting processes; 17, L.S, male, showing inter pseudo-segmentation; 18, T.S. male showing six cement glands; 19, vaginal complex; 20, exes. 74 The cysts from the crustacean may have one end slightly wider than the other aud both extremitics have a wide shallow depression, one of which contains the genital aperture. These larvae evert the proboscis if placed in tresh water. We have wot found these larvae in any othet animal living in the swamps. GoxninknyxcHts HYtag (Johnston) Johnston and Edmonds Fig. 10-20 Four female and two male specimens of this echinorhynch were found in the intestine of Poedargus sirigvides at Orroroo, South Australia, in May 1939, by Mr. J, T. Gray. The wortns are Jong, cylindrical and widest in the posterior region. Both sexes exhibit inner pseudo-segmentation of the type shown by Cordiorhyichus clitorideus Meyer (1930). The posterior extremity of the female ig much swollen and rounded, and near the female genital opening there is a small epidermal protuberance, The body of both sexes is smooth. The length of the males is 48-52 im., and the maximum width E:1-1-3 mim. The length of the females i; 56-81 mm., and the maximum breadth 14-18 weusttred just in front of the rounded bulb at the posterior extremity of the wortn- The diameter of this bulb is 1°9-2°5 mm, The proboscis is smal} in comparison with the length of the parasite. Its length is 0°55-0°65 mm.. and its maximum breadth 0°37-0'43 mm, Its width at the anterior extremity is 0°12-0-13 mm, Between the proboscis and the body there js a short tinarmed neck 0°10-0-15 nim. long and 0:30 mm. wide, In the largest fernale the total length of the proboscis and neck ig 0°75 mm, The proboscis is armed with 26-28 longitudinal rows, each of ten hooks, The first five hooks of each row are largest and possess well- developed posteriorly-directed rooting processes. These rooting processes of hooks six and seven are much reduced and an antcriorly-directed process appears. It hooks eight, nine and ten the processes are anteriorly directed. Fig. 15 and 16 show the general arrangement and size of some of the preboscis hooks. Occa- sionally a slight yariation in the shape and arrangement of the hooks in a row observable, e.g., fig. 15, hook 5a, ‘The proboscis sheath is double-walled and is inserted at the level of the tenth hook, ‘The point of insertion of the sheath, there- fore, does not divide the armed portion of the proboscis into two parts as is the case with the genera Centrorhynchus Lithe 1922 and Gordiarhynchus Meyer 1931. The width of the outer wall of the sheath is about 20 p, and its length 1+1-1-6 mm. fn one male specimen the sheath is yery much constricted near its middle. The lemnisci are about 3-4 mm. long. There is one long and well-developed lacunar vessel From which anastomosing vessels arise. The hody wall is thick and consists of aii Outer epidermal layer surrounditig layers of circular and longitudinal muscle tissue. No nuclei were noticed in it. Transverse and longitudinal sections show that in both sexes an inner pseudo-segmentation is present similar tu that gecurring in the female of G, clilorideus. The structure of the pseudo-segmentation of the nale is shown in fig, 17. Two tong and ellipsoidal testes of approximately equal length lie in tandem in the afiterior half af the worm, Their dimensions are length 1°2-2°2 mm., and maximum breadth 0-42-0°60 mm, There appear to be six tubular cement glands pressed closely together, Four of these arise near the posterior testes. The cement reservoirs are loug. The bursa bears two anteriorly directed diverticula and the male opening is terminal. All the females in our collection are densely packed with eggs and ovarial masses, and we have not been able to trace completely the female genitalia. The vaginal complex consists of three bulbs, and the genital opéting, marked by © swelling of the epidermis, is sub-terminal. As is the case in G, cliforideus the 75 ovarial masses develop in the larger, presumably dorsal, segments. Ripe eggs are ellipsoidal and when mounted in balsam are 54-60 long and 24-28 » wide. They are without polar prolongations. The nucleus in some of the eggs s€ems in be dumb-bell-shaped. Systematic Position The genus, Gordiorhynchus, as conceived by Meyer (1931, 120-22), consists of Centrorhynchinae with inner pseudo-segmentation in the female and with an appendix near the female aperture. The sub-family, Centrorhynchinae Meyer 1931, consists of these Polymorphidae in which the insertion of the proboscis sheath divides the proboscis into two parts. Both the male and female specimens in our material show inner pscudo-segmentation, and the females possess a struc- lure near the genital opening which seems to correspond to the appendix of G, elitorideus, Internal pseudo-segmentation in male specimens of the genus, Gordiorhynchus, has already been recorded in the case of G, falconiz Johnston and Rest (1943, 229). The proboscis of the parasites in our collection, however, is not divided intu two portions by the proboscis sheath, The double-walled sheath in each of our specimens arises at the level of, or just posterior to, the level of the tenth hook af cach row. We consider that the conception of the genus. Gordiorhynrhus, should he witened to include echinorhynehs with inner pseudo-segmentation in one or both sexes. Evcysted Kori From time to time during class dissections in Adelaide University white acanthocephalan cysts have been obtained [rom the mesentery of the frogs, Limnodynastes dorsalis and Hyla aurea, The presence of these cysts in H. aurea in New South Wales and H. caerulea in Brisbane has alrcady been reported by the senior author (Johnston 1912, 84-85; 1914, 83-84), who described the species (1914) as Echinorhynchus frylae, Meyer (1932, 252) placed the species amongst Acanthocephala incertae sedis. We have found that in many cases the proboscis of the encysted parasile can be made to evert if the cyst is freed from the mesen- tery and placed in fresh water. Fig. 10 shows a female specimen in the everted condition, The size of the proboscis and the shape, size and arrangement of the proboscis hooks show that this parasite is the encysted form of the worm which we have ubtained {rom Podargus striqoides. The shape and size of some of the proboscis hooks of the larvae are shown in fig. 11 and are drawn to the same scale as thost of the adult shown in fig. 15 and 16. The following measurements haye been made on the cysts and the everted larvae. Length of cyst 1-38- 1-70 imm., and maximum breadth 0°45-0-63 rim. Length of proboscis 0°58- 0°63 mm., and maxitoum width 0-35-0'40 mm., and armed with 28 longitudinal tows each of ten hooks. Length of Jarvae 2°1-3:0 mm, and maximum width 0°45-0°65 mm. It seems to us probable that the adult stage will be found in other nocturnal predatory birds, since Podurgus is not a common bird in the localities from which infected frogs were oblained, Travassog (1926, 43) and Meyer (1932, 117) muentioned that the larva of Cen/rorhynchus lummidulus occurred in certain frogs and snakes. LiTRRATURE Jonnston, T. H. 1910 On Australian avian entozoa. Jour. Proc. Roy. Sue N.S.W., 44, 84-122 Jownxstoxs, T, 11, 1912 Notes on some Entozoa. Proc. Roy, Soc Qld, 24, 63-YL 76 Jounston, T, H. 1914 Some new Queensland Endoparasites. Proc. Roy. Soc. QOld., 26, 76-84 Jounston, T. H., and Brest, E. W. 1943 Australian Acanthocephala No. 4. Trans. Roy. Soc. S. Aust., 67, 226-230 Jounston, T. H., and Deranp, E, W. 1929 Australian Acanthocephala No. 1. Trans. Roy. Soc. S. Aust., 53, 146-154 Jounston, T. H., and Epmonps, S. J. 1947 Australian Acanthocephala No. 6. Rec. S. Aust. Museum, 8, (4), 555-562 Marvat, L. 1905 Monographie des Acanthocephala des Oiseaux. Rey. Suisse Zool, 13 Meyer, A. 1930 Gordiorhynchus, ein neues Acanthocephalen Genus mit innerer ovarialer Pseudo-segmentierung. Zool. Jahrb., 60, (1), 457-470 Mever, A. 1932 Acanthocephala, in Bronn’s Klassen und Ordnungen des Tierreichs, Bd. IV, 2 Abt., 2 Buch., 62-73, 119-120, 120-122 Travassos, L. Contribuicoes, etc. Revisao dos Acanthocephalos brasilieros, H, Fam. Echinorhynchidae, ete. Mem. Instit. Osw. Cruz, 19, (1), 31-125 (Separate, pp.1-97) Van Creave, H. J. 1945 A new species of the Acanthocephalan Genus, Poly- morphus. Jour. Parasit., 31, 128-130 Vaw Cieave, H, J., and Srarrert, W. C. 1940 Acanthocephala of Wild Ducks. Trans. Amer. Micr. Soc., 59, 348-353 CESTODES FROM AUSTRALIAN BIRDS I. PELICANS By T. HARVEY JOHNSTON AND HELEN GOLDTHORP CLARK Summary Three new species of cestodes belonging to the genus Hymenolepis have been obtained from the only Australian species of pelican, Pelecanus conspicillatus Temm. Our present material was collected from eight birds, all from Tailem Bend, South Australia, Messrs. G. G., Fred., and Bryce Jaensch of that town, and Mr. L. Ellis, now of Murray Bridge, assisting us very generously. The material was obtained whilst we were engaged in other parasitological research work, carried out with the assistance of the Commonwealth Research Grant to the University of Adelaide. Types of the new species have been deposited in the South Australian Museum. 7 CESTODES FROM AUSTRALIAN BIRDS I. PELICANS Dy T. Warvey Jounston and Herex Gotpruorr CrAng * [Read 13 May 1948] Fig. 1-17 Three new species of cestodes helonging to the genus Hymenolepis have been obtained from the only Australian species of pelican, Pelecanus conspicillatus Temm. Our present material was collected from cight birds, all from Tailem Bend, South Australia, Messrs. G. G., Fred, and Bryce Jaensch of that town, and Mr. L. Ellis, now of Murray Bridge, assisting us very generously. The material was obtained whilst we were engaged in other parasitological research work, carried out with the assistance of the Commonwealth Research Grant to the University of Adelaide, Types of the new species. have been deposited in the South Australian Museum. Hymenolepis murrayensis n. sp. (Fig. 1-8) This cestode was found in four of the eight birds examined. Egg-bearing worms were 90-185 mm. long by 0-75-83 mm. in maximum breadth. Segments are broader than Jong, but in those which are gravid, the relative difference in dimensions iy less marked, The small scolex is distinctly marked off from the heck and measures O-t7--25 mm, in diameter. The rostellar sac is 0*08 mm. wide and extends back as far as the posterior margin of the suckers. The rostellum has 20-22 hooks of two sizes and differing in shape (fig. 2, 3), the larger being 0-02 and the smaller -016 mm. in total length (ie., the distance between two parallel lines placed one at each end of the hook). The hemispherical or ellipsoid suckers measure (08-1 by -1--12 mm, The unilateral genital pores lie in the middle of the edge of the segment. Elliptical calcareous corpuscles are scattered through the cortex. The ventral excretory canal of the poral side has a diameter ahout ten times that of the dorsal vessels. The yentral canal of the aporal side ig very much narrower than its fellow. The excretory ducts pass below the venital ducts, The testes develop before the ovary. Very early segments exhibit the out- lines of the three testes and the cirrus sac, and those with mature testes show an inimature ovary and yolle gland, while in later segments with a well-developed ovary the testes either are degetierating or have disappeared. One testis lies on the poral side of the segment, the other two on the aporal, one gland being more anterior and lateral than the other (fig, 4). In segments in which the ovary is just developing, the organs are about 0-11 mm, in diameter. The internal and external seminal vesicles can be seen most clearly in segments with a mature ovary and disintegrating testes. The external vesicula is retort-shaped and lies between the cirrus sac and the testes. Between the two vesicles is a coiled portion of the vas deferens, The internal vesicle occupies most of the length of the cirrts sac, and narrows gradually tq the ejaculatory duct. The long narrow cirrus sac, which has a well-developed muscular wall, ‘extends obliquely across three- fourths of the width of the segment and slightly under the overhanging part of the preceding segment. It measures 0'36-°37 by ‘08-09 min, in segments with *University af Adelaide. "Trans. Roy. Soc. S. Aust... 72, (1), 23rd August, 1948 78 doe =\orm 1+ *Tinm 3 -2mm Fig. 1-8, Hymenolepis murrayensis: 1, scolex; 2, 3, rostellar hooks; 4, segment with mature testes; 5, segment with mature ovary; 6, gravid segment; 7, 8, sezments from contracted strobilae. Fig, 2 and 3 to same scale; 4, 5,8 to same scale. esy, external seminal vesie’e: ex, excretary canal; gp, gemtal pore; isv, internal acminal vesicle; a, ovary; pr, prostate glands; rs, receptaculum seminis; t. testis; u, uterus; v, ‘vitellarium, 79 mattire testes, and 0°44--52 by :075--08 mm. in those with a mature ovary, There is a chitinized ring around the opening Gf the sac into the genital atrium, The ovary arises between the three testes and teaches its maxinium size (0°25-'3 mm. in diameter) when the testes arc degenerating (fig, 5). It has 15-20 well-marked lobes. The yolk gland is compact, The large chitinized seminal receptacle, 0+154 by +113 mm., lies ventrally in front of the ovary, and, together with the cirrns sac, persists in gravid segments, The vagina travels behind the cirrus sac and parallel with it from the receptaculum to the genital atrium. The utertis arises as two lobes, one on either side of the ovary, but the organ later fills the seyment and extends beyond the Jongitudinal excretory canals (fig. 6}. Eggs measure about 0-04 by “3 nim., the oncospheres 0-02 by ‘614 mm., and the hooklets 8 long, dn this species the ratio of length to breadth of segments varies consider- ably acearding to the degree of contraction, and correlated with this the testicular arrangement shows variation, In strubilae with very short contracted segments the three testes lie almost in a Straight line and the excretory canals are sinuous, but the number and sizes of the hooks agree with those of the more relaxer worms. Occasionally a strobila has some segments with a linear arrangetnent of the testes, whilst in others one aporal testes lies in front of the other, Some strobilae are more elongate, with segments squarish or even longer than beoatl, The only other species of Hymenolepis with 20-22 hooks, described from Pelecaniform birds is H. suedici Stossich, but the latter has hooks 0-03 mmm. long, and its very large cirrus sac extends below the overlapping part of the preceding segment tu its anterior aporal corner, and there is no aporal excretory canal (Fuhrmann, 1906, 749), Hy, fictitia Meggitt (1927 a) has 24 hooks, 34-39 j. and 48-52 win length. H. wagniuncinata Meggitt (1927 b) has more that ten hooks measuring 39 p, and H. parvicirrasa Meggitt (1927 b) has more than 14, measur- ing 43-48 », these two species being thus differentiated from H. murrayensis by the sizes of their hooks. In JZ, phalacrocorax Woodland (1929) originally described as unarmed, the testes lie outside the longitudinal excretory canals, According to Hughes’ key to species of the genus (1941), H, murrayensis would be placed near H. ficfitia. Hymenolepis jaenschi mh. Sp. (Fig, 9-13) This species was found in the eight pelicatss mcamined, Egg-bearing worms, measure 40-85 um. Tong and 62-67 num in inaximum breadth, with segments broader than lung. The scolex (fig. 9), 0-3-'39 mm), in diameter is sharply marked off from the neck. The rostellar sac, 0‘07 mm. wide, and 0-16--24 tuni. long, extends bavk to the posterior qargin of the suckers. The everted rostellunt is 0-22 by 025 nim, and has 14 hooks in two alternating series, the larger hooks being 0-028 and the smaller O-OL8 mm. long (fig 10,11). The suckers are abuut 0-2 mm. in eiamwter er O-14 by *18 mm, if elongate. The anilateral enital pores are at, or slightly in front of, the middle of the segment margin, The poral ventral longitudinal excretory canal is about 0°022--026 mm, ‘n diameter, while the aporal and (he two dorsal canals are very nartow. Calearenus corpuscles are abtindant. The testes and ovary appear at about the sane time in young seemichts, hut the former tend to persist in seements in which the developing uterus lias dis- placed the ovary. One testis is aporal, two lie in the extreme posteriol region uf the segment, the other aporal lestis lying in front uf the ovary, “F hey measure 0-064--08 mim, in diameter. The external seminal venicle is rounded o aval, 42nim jor 13 “3m ~~ nschis 9 seolex; 10, 11, rostellar hooks; 12, mature segment; 13, segment with developing uterts. Fie. 9-13—ITwwnenole p is fae Fig. 14-17—Ayoenolepis ellisi: 14, segment with mature testes ; 15, segments with mature ovaries; 16, gravid segmentt 17, segment from contracted strohila, Fig, 9, 12, 13 to same stale; fig, 10, 115 fiz. 14 and 17. 31 O-058--078 by -O5 mimi: the internal vesicle is elongate and may reach 0-05- O16 mm, The thin-walled citrus sac lies year the anterior margin of the seg- ment atid parallel with it; it measures 0-26-29 by O-05--06 mm., and extends almost to the aporal exerctory canal. There is a well-defined ring of small spines ‘round the opening of the cirrus sac into the atriuin, and the adjacent base of the cirrus is also provided with smatl spines for a distance of 0-05-10 imum, The eyary lies inv the midregion of the segment between the testes and slightly toward the aporal side, Jts two main lobes are subdivided ite a total of about 8-10 lobules, It measures 0°052--11. The large chitinized receptaculinn lics behind the ovary, near the two posterior testes and dorsally to the yolk ¢land: it Measures up to 0°073 by *054 min. in seements containing testes and uterus, The vagina Uravely forwards From the receptaculim and then parallel with the cirrus Sacto the atrium. In gravid segments, the bilobed sac-like ulerus fills the medulla and extends beyond the excretory ducts, It is about 0-4 min. long and 0-5 min, broad, Eggs measure O04 by O°3 min, and ihe ohcospheres 0-02 by 0-15 mm., With hooklets 8 long, We do not know of ay species of Hymenolepis with 14 hooks described from Pelecaniform birds. #7. magmuicinata Meggitt (1927) and FH. parcicirrosa Meggitt (1927b) have already heen anentioned above, but theit hooks ditfer in number atid size From those of H. jeenschi. Lf grouped according to Llughes’ key (1941). 1. jaeusehi would Approach Hl, fictitia Meggitt, a species with 24 hooks, 0-034--039 and 0°04--052 inn, long. Hymenolepis ellisi n.sp, (Fie. 14-17) The material consists of three fragments without scolices, collected in August 1942. They include mature and gravid segments and measure 60, 70 and 100 mnt. in length, with breadths of QO-72, 1-04 and +065 mm. Tespectively- Most segments are broader than long, put those which are gravid tend to be syuarish of even lounger than broad. The genital aperture is al about the middle of the margin, The three round or elliptical testes (*063-"08 mm.) develop before the uvary- One is poral. the other two aporal, one of the latter being anterior ie, and nearer the aporul edge of the segment than, the other, The external seminal yesicle lies aporally behind the end of the cirrus sac aud measures O-1-0-2 by ‘08--00 mm., or exceptionally 0-2 by 0-12 mm. The yas then continues as a narraw tube into the cirrus sac where it widens into a long internal seminal vesicle, 0-03-04 mm. wide and varying in length according to its contents. The long ejaculatory duet is coiled when the cirrus is retracted, The cirrus sac is large and thin-walled and my ctitve to form an are with its concavity directed posteriorly, or ij may appear ly project into the preceding segment: The organ measures 0-42--48 by “075-085 i. seginents with mature testes, and imay reach 0-5-*7 hy -O7—-11 mm. in gravid segments. Surrounding its opening into the large deep atrium isa ring bE small spines, and the series is continued on the base of the cirrus. These spines become detached readily. The greaily lohed ovaty lics in the middle of the section. It arises between the three testes, but the latter disappear by the time the ovary reaches its maximum size (O25 mm. in width), The slightly lobed yolk gland measures 0'1-+}3 by 07-09 mm, The thick-walled receptaculum lies between the ovary and cirrus sac, and measures 0109-11 by -145--16 tum. The wide vagina (average breadth -035, maximunr ‘O44 mm.) les behind and parallel with the cirrus sac, anil becomes slightly coiled as it passes ventrally to enter the receptaculum, The 82 uterus arises as two lobes, one on either side of the ovary, but when fully developed it forms a sac filling the medulla and extending beyond the excretory canals. Eggs measure about 0°04 by -03 mm., the oncospheres 0°02 by "015 mm., and the hooklets 7 in length. Gravid fragments collected on another occasion probably belong to this species because of the resemblance of the external seminal vesicle, receptaculum, wide vagina, deep atrium, spined cirrus, and ring of spines around the opening of the cirrus sac into the atrium. The maximum width of these segments was 0-09-1:3 mm., and the cirrtis sac measured 0:6-'75 by 0°8 mim. This species differs from H. murrayensis in its spiny cirrus, thin-walled cirrus sac and the ring of spines around the outer aperture of the latter. It differs from H. jaenschi in its much larger cirrus sac, the position of the receptacu- lum, the disappearance of the testes before the ovary reaches its maximum size. The lack of a scolex prevents further comparison. The very large citrus sac distinguishes it from all other Hymenolepis spp., from Pelecaniformes except H. medici Stossich. In his account of the latter, Fuhrmann (1906) did not give the measurement of the sac, but stated that it was very large, extending to the aporal anterior corner of the preceding segment or else bending into an arc in its own segment. The internal anatomy of the two species is rather similar, The finding of a scolex of H. ellisi should permit a decision as to the identity or otherwise of the two species. LITERATURE FuurmMann, O. 1906 Die Hymenolepisarten der Vogel, I. C. Bakt. Orig., 42, 730-755 Hucues, R. C. 1941 A Key to the Species af Tapeworms in Hymenolepis. Trans. Amer. Micr. Soc,, 60, 378-414 Mavuew, R. P. 1925 Studies on the Avian Species of the Cestode Family, Hymenolepididae, Illinois Biol. Monogr., 10, (1), 125 pp. Meceitt, F. J. 1927a On Cestodes collected in Burma. FParasitol., 19, 141- 153 Mecoitt, F. J. 1927b Report on a Collection of Cestodes, mainly from Egypt, Part II, Cyclophyllidea, Family Ilymenolepididae, Parasitol., 19, 420-448 Woon.anb, W. N. F. 1929 On some new Avian Cestodes from India. Para- sitol., 21, 168-179 THE GENUS TRAGARDHULA BERLESE 1912 (ACARINA, TROMBICULIDAE) By H. WOMERSLEY Summary In 1904 Tragardh (Results Swedish Zool. Exped. To Egypt and the White Nile 1901) described Trombidium niloticum from both larvae and adults found in apparent association and in large numbers on the leaves of an aquatic plant (Pistia stratiotes) on the White Nile (coll. Gebel Ahmed Aga 20 March 1901). 83 THE GENUS TRAGARDHULA BERLESE 1912 (ACARINA, TROMBICULIDAE} By If. Womersrey [Read 10 June 1948] Fig, 1 A-G, fig. 2A-J ia 1904 Trivardh (Results Swedish Zool. Exped. to Egypt and the White Nile 1901) described Trombidinm nileticum from both larvae and adults found in apparent association and in Jarge numbers on the leaves of an aquatic plant (Pishie stratiotes) on the White Nile (coll, Gehel Ahmed Aga 20 March 1901). For these larvae Oudemans 1911 (Entom. Ber., 3, (57), 123) erected the genus Slaxkaartia and later, in his monographie study af the larvae of the “Trom- bidiidae and Erythraeidae’™ (Zool. Jahrb., Suppl 14, 1912), he figured ancl described the larvae in great detail. “[ragardh’s adult material was studied by Berlese, who, in his monograph on the adult “Trombidiidae” (Redia 8, (1), p. 4, 1912) used the name Tragardhula as a subgenus of Trombicwla. Later, how- ever, in the same work (p. 96) he adopted Oudemans’ name Blankaartia also as a subgenus of Trogbicula. The adult genus Trombicula as understood at the preset time is characterised by the body being constricted medially, giving il a figure of 8 shape; by the elongate crista snetopica. with a stibposterior sensillary area {furnished with paired fila- mentous sensillae; eyes present or absent, but when present only one on each side of the crista and closely adjacent to, or away from, the sensillary urea, In Tragardhula. wilotica, however, they are placed well away from the crista and well in front of the sensillary area, as figured by Berlese (1912) and as seen in a specimen kindly given to me by the British Museum, which specimen was froin Pistia, im India, In the other known adult Trombicula species, either with eyes placed close to the sensillary area, or wanting, there are also other smaller features which enable them to be associated with some at least of the many larval genera which have been proposed. The larvae of the Trombiculidae are distinguished by having only a single anteriorly placed dorsal shield, of varying shape, from rectangular to pentagonal, hexagonal or tongue-like, and furnished normally with 1 ar 0 antero-median setae, 2 antero-lateral setae and 2 postero-latcral setac, as weil as a pair of sensillae which may be filamentous, clavate or globose. Additional sctac may occasionally oceur on those scuta which are more or less tongue-shaped, but these setae are usually situated hehind the posterolateral sctac. Eyes may he 1 or 2 on each side or absent. Coxae I are furnished with only a single seta, and there is a pir of setae between these coxac. ‘he legs are all furnished with a pair of simple claws and a longer and more slender empodium, The larva of Blankaatia nilotica (Trag.) has many characters which place it outside the family Trombiculidae as defined above, In the first place it has two antera-dorsal shields, the front one of which is large and reaches or overlaps the anterior margin of the hady and is furnished with 6 setaé in addition to the filamentous sensillae. ‘The second shield has a number of setae. The pair of setae, normally between coxae I have migrated onto the coxae so that cach carries 2 setae. Legs | and IT have only the paired claws, and no empodium, the claws being apically fureate. On leg IIL the claws have an apical nail-lke tip, «Entomologist, South Australian Museum. Trans. Roy, Soc. S, Aust., 72, (1), 23rd August, 194% 34 Thos, while the adult Tragardieula nilotiea (Trag.) belongs to the Trom- bictilidae, the larva certainly docs not, and the two stages described hy Tragardh under this name cannot be associated, ln 1936 Sig Thor (Zool, Anz. 114, 30) raised the genus Penlayanella tor those species of larval Trombidiids in which the dorsal scutum was pentagonal in shape, and the sensillac filamentous. As type he designated Trambidium ardcue ‘Trig. fram the White Nile. Other species were filirenhdlst (Ouds, 1910), lragardhi (Ouds. 1910), smuris (Ouds. 1910), acescutellaris (Walch 1922) and yorkin (Satnbon 1928). Tins genus has been rather ignored by most workers, but Willmann in 1947 (Das Tierreich, Lig, 71 b, Trombidiidae, p. 292) admits it, and meludes the addi- tional species desuleri (Methlagl. 1928) and cenlrepodis (Fwing 1928), Until now, 110 adults af any of these species have been known, but 1m this paper adnlts of P. acuscutellaris, both caught in the field and reared from larvae through several generations are described, and it 1s shown thal they agree fully with the subgenus Tregardhula Berlese. The subgenus Megatrembicula was erected in 1947 by Michener (Aun, LEntom. Soc. Amer., 39, 452) for four species of adult Trombiculids from Central America, The subgenotype was Trombicula allcei Ewing, and other inclocdesd species were 7. veluscot (Boshell and Kerr), 7) peruoiane (Ewing) ans! M. attenuata Michener. The essential character of the stibeenus was the position of the eyes, well away from the crista and well anterior of the sensillary arca, Michener was also successful in obtaining larvae of these species, except perwiiena, anc although he could only separate these with difficulty, he pointed out their great similarity tp Pentagonella of the Old World in their having a pentagonal scutum, In 1916 Tanaka (Zentbt. Bakt., Abt. 1, Orig.) reters ta, and on pl. 4, fig. 37, shows clearly, the position of the eyes as well away from the crista and in advance of the sensillary area ina nymph of Trombicula japonica, then relerred ta a sub- species of aulummnals (Shaw), which is not related in that the eyes in the latter species are entirely wanting. The following description of the adult of P. acuscutellaris fits clearly thost af Megairombicula and Tragardhula, and both Pentagonella and Megetrambicula therefore are synonymots with Vragardhula Berl. The facr that acuseutellaris is the only species of Pentagonolla from the Old World as yet known friwn the adult, but is not the actual genotype of Pentageneila, leaves a slight possibility that the adult of ardeag (the genotype) may nat he a Tvagdrdhule. However, it having been shown that Blankaartia nilatica (Ouds,) 1s nol related to Tragurdiuda nilotica (Berl), can any other Egvptian larval Yrombicala with pentagonal scutum be suggested as the possible larvae of Trg. wilolica? Of such larvae only two have been described from Egypt, J. ardeac and T. tregard/i (Ouds, 1910), both from Tragardh’s expedition. The first of these was Trem the legs of a heron, Ardea emercva, irom the White Nile, March 1901, the same locality and date as for Vrag. nilotica, While the association of this larva with the adult cannot be accepted until suecessful rearings have been inade, the habitat of the larvae on the legs of a wading bird, and the adults on the Jeaves of an aquatic plant (Piste) is highly suggestive. The second species of larvae, tragurdhi (Quds,) was frone the ears of a monkey, Cercopithecus, and thus fram such w host habitat cam surely. be disregarded, RS Genus Tracarniuna Berlese 1912 Tragardhula Ver), 1912, Redia, 8, 4; Bhurkaurtia Berl. 1912. Kedia, 8, 6 (ion Oudenmng 1911 larvae); Pentagonella sig Thor 1936, Zool. Anz, 114, 30; Megotrombicula Michener 1947, Ann, Entom, Soc, Amer,, 39, 442. Adults of typical Trombicule figure of 8 facies. Crista linear with sub- posterior sensillary area about as wide as long. Epistome rounded with fine teeth and 1 seta. \ecessory spines on palpal tihia arising from a slightly raised boss some distance from base of claw, Eyes large. 1-|- 1, well separated [rom erista and well in front of sensillary area, Sternal plate distinct and well defined, undivided, wider than long and lying between coxae I. Genotype: Trombidiaim niloticum Trig., adult. TRAGARDHULA actuscureraris (Walch 1922 ) Trambienla acuscutellaris Walch 1922, Kitasato Archiv. Exper. Med, 5, (3) 78; Gater, B..4, R., 1932, Parasitol, 24, 143-174; Metta, D. A., 1937. Ind. J. Med. Res,, 25, (2), 353-365; Philip, Woodward and Sullivan 1946. Amer. J. Trop. Med, 26, (2); Radford, C. B., 1946, Parasitol., 37, (1-2); Jayewickreme, S. H., 1947, Nature, 160, 578. Trombienla (Pentagoncla) acuseuicllaris, Sig Thor 1936, Zool, Anz., 114, 30; Womersley and Heaslip 1943, Trans, Rov. Soc, S, Aust... 67, (1), 78. Tragardhula acusculclaris, Willman 1947; Das Tierreich, Lig, 71b. This species is widely distributed in the Asialic-Pacific Region TL was originally described by Walch from Sumatra, aud has since been repurted from the Federated Malay States (Gater), India (Mehta), Maldive Islands (Radford), Ceylon (Jayewickreme), and | haye seen material collected by Lieut-Colt. C. R. Philip Irom the Philippines, and by Maj. R. N, MeCulloch from Borneo. Tn all casey the hosts of the larvae were species of Ruttus. The nymphal stage of this mite remamed unknown nntil 146 when Radford described very briefly the nymph reared from Jaryac. The adult retnained quite unknown and unrecognised in the field until last year when Jayewickreme reported in “Nauire” ihe successful rearing through several generations, using the recently established methods of feeding the nymphs and adults on the eggs of species of mosquitoes, Further, the adults can now be fairly casily recognised in the field, and laree numbers of larvae have been obtained [rom captured females. In the present work the adult male and fenalé are deseribed anet fignred for the first time, and the opportunity ts taken of redescribing and figuring the larva and nymph uiore iuily, For the material for this study | am yety much indebted to Mr. S. H. Jayewickreme who very willingly supplied me with a considerable amount of both reared and wild larvae and adults, “lhe redescription of the hyiph is from a specimen supplied by Dr. C. D, Radford. Redescription of Larvae, Lig. 1, A-P—Colour in life red. Shape oval. Lenetit nnengorged 260 . (excluding gnathosoma), width 195 4, Morsal shield (fig. 1, 1) pentagonal, with PW only very slighily greater than AW, and angle of convexity of posterior margin, 7... PW/PSU = 2-0 or thereubouts, Sensillae long and fila- mentous with ciliations ii distal two-thirds. Normal setae of setrun endl dorsal setae ciliated and tapering. Eyes 2— 2, on ocular shields, the posterior eyes the smaller, Palpi (fig. 1.12) stout; femur and genu furnished with a branched or long ciliated seta; tibia with the normal 3 sctac, of which ihe yentral has branches, the dorsal and lateral heing nude; paipal caav trifureate. Chelicerae (fig, 1,C) wita the usual apical tricuspid cap, and on the inner edge a subapical f orwardly directed tooth, Galeal setae nude, Dorsal setae 26 in mimiher, arranged 2.6.6,64.2 (fg, 1, AY. Veutral setae, excluding the single one tn each coxa, 24, arranged R264/44.2 (fig. 1, 1), Dorsal setae t9 70 p long, veritral to 504. Legs long-and 86 peers] ~~ 225 Ls =e ao8 see 36 ane! “qe° oO Qc uaF etek . cod ao Figs au.8 Sage x3 a PsA apres) a4 33 Sone ie Zz 1 3-5 Sens. 84-0 No variation recorded Redescription of Nymph, from a specimen reared from laryae by GC. D, Radford in the Maldive Islands. Colour in life bright red, Of typical Trombicwla facies. Length to 800 p, width across hysterosoma 400,, Crista linear with somewhat diamond-shaped ateola, and two filamentous nude sensillae, 146 « long, sensillac bases. 42 » apart- Eyes 1+1, well removed from sensiflary area and about half-way between extremes of etista, Chelicerae normal with fine inner serrations. Palpi with tibial claw strong, with 2 accessory spines but these on a slight boss and placed about midway between base of claw and articulation of palpal tarsts. Palpal tarsus slightly clavate, not reaching tip of claw. Legs: shorter than body, I 660 » lope, 1 460}, II] 470 p, IV 530 ,:; tarsus I less than twice as long as wide, 151 by 91, metatarsus 1 115 long. Dorsal setae numerous 40» long, uniform, arising from closely set tubercles, uniformly thick with strong ciliations. Genital discs 2 pairs. Description of Adult—Fig, 2A-J. No apparent difference in the size of males and females. Colour in Hfe deep red, Of typical Trombicula facies. Length to 1,800», width across hysterosoma 1,200. Crista (fig. 2A) littear with well-developed subposterior sensillary area, roughly diamond-shaped, with the sensillae bases at the ends of transverse dumbbell-shaped areola; crista 300 long and sensillae bases 65 apart; sensillae 195 « long, filamentous and nude; apex of crista forked. Epistome well developed, rounded conical with numerots teeth and with a single seta 71 «long. Eyes 1 + 1, large, situated well away from crista and midway of crista length. Chelicerae (fig. 2B) with finely-toothed inner edge, Palpi (fig. 2C and D) with strong tibial claw and 3 accessory spines which are distinctly away from base of claw, and another on outer edge opposite articula- tion of tarsus, and on the outer surface another strong spine. Legs normal, I 1275 p long, II 910», 111 9104, IV 1,235 py; tarsus I (fig. 21) 325 4 long by 163 » high, metatarsus | 247 » long; sterial shicld between epimera of legs I and 88 Fig. 2, A-J—TrayardAula acuscustellaris (Walch 1910) adu’t. A, crista, epistome and cyes; B, chelicera; C, palpal tibia and tarsus from above; D, ditto from below; E, coxae 1] and I and sternum; F, coxae III and IV with genitalia and anus showing relative positions; G, genitalia of male; H, penis; I, tarsus and metatarsus of leg I; J, posterior dorsa) seta. 89 If, markedly shorter than wide (fg. 2E). Genital orifice situated close into the epimera of legs ITT and IV (fig. 2), with 3 pairs of genital dises. In the male provided with a penis as figured (fg. G and IJ), the apical point of which is asymmetrical. Dorsal setae numerous, umiformly thick, with short ciliattons (fig, 27). The synonymy of Tragardhula milotica (Tray. 1904) is as follows: ?Trombtdium urdeae Tragardh 1904; Microtrombidium a, Ouds. 1910, M.a., Ouds. 1912; Yrombicula a. Ewing 1928; Pentagonella a, Sig. Thor 1936, PB. a. Willmann 1947 (larvae); Trombidiunt niloticum Trigirdh 1904; Trombicula (Tragardhula) n. Berlese 1912, p, 4; Trombicule (Blankuartia) n. Berl. 1912, p. 96 (non Blankaartia Ouds, 1911, larvae); Tragarihule n, Willmann 1947. The following table shows the adult er nymplal species which belong to the genus Tragardhula and the larval species which will possibly also be placed therein as their respective adults are discovered, Other species of Trombicula are known which in the larvae have a more or less pentagonal scutum but do not fit with any certainty into Tragardhula, for example 7°. aulwnalis of Europe, in which the adult has no eyes and belongs toa group for which the name Neotrombicula llirst is available. Aputts og Nympus Trag. uilotica (Trigardh 1904) Traq. atlenuatu (Michener 1947) Tray, acuscutillaris (Walch 1922) Trag. japonica (Tanaka 1916) Trag, Peruviane (Ewing 1926) Trag. alleei (Ewing 1926) : Trag velascot (Boshell and Kerr 1941) LARVAE Trag. alleet (Ewing 1926) Trag, [ahrenhelet (Ouds. 1910) Trag. celascot ( Boshell and Kerr 1941) = Trag, aris (Ouds. 1910) . Trag. ardeag (Tragirdh 1904) Trag. desaleri (Mcthlag!. 1928) Trag. tragardii (Quads. 1910) Trag. japonica (Tanaka 1916) Trag, yorkei (Sambon 1928) Trag, acuscutellaris (Walch 1922) Trag. centropodis (Ewing 1928) Trag. attenuata (Michener 1947) Tracarniuca ? (PENTAGONELLA) ARDEAR (Trag.) Prot. lvar Tragardh, whom | recently had ihe very great pleasure of meet: img in Stockholm, has very kindly loaned to me for study a number of slides of his (1901) Egyptian material. One slide contained 4 specimens of his ardeae. These have been remounted, and | am now able to give the followitig data to bring the description into line with recent studies ot the species of larval Trom- biculidac. Fig. 1,.G—Dorsal scutum pentagonal with PSB greater than ASB, and SB about in line or slightly in advance of PLL. Eyes 2+ 2, closely adjacent to postero- lateral angles. AM and AL about equal, PL much longer. Sensillac flamentous and ciliated on distal half. Chelicerae with the usual apical tricuspid cap only. Palpi stout, tibial claw trifurcate with the prongs subequal; setae on femur and genu with branches; on tibia ventral strongly branched or ciliated, lateral lightly branched, dorsal nude. Galeal sctae with 2 or 3 light branches. DS long, to 84%, arranged 2.6.4,8.6.4. Ventral setae rather shorter, arranged 2.2.6.6.24.2, Tarsus of lee TIT with a long nude seta; of leg 1 and IT with the usual dorsal rod-like sensilla, AW PW SB ASB PSB SD A-P AM AL PL Sens. Standard Data in microns from 4 specimens are: Mean 74-2 #14 87-5 +07 31°85 + 0°35 25°2 42-0 67:2 28-0 61-6 60-7 +£0:93 84-0 84-3 42:6 Standard Deviation 2-80 + 0:99 1-40 + 0-49 0-70 + 0-24 90 Theoretical Range 65°8 -—82-6 83-3 —91-+7 29°75 —33+95 No variation recorded No variation recorded No variation recorded No variation recorded No variation recorded 1-62 + 0°65 55°9 ~65°5 No variation recorded 4-51 + 1-84 70-8 —97°-8 Observed Range 72°8 —78°4 86-8 -89°6 30-8 — 32:2 58 +8 — 61-6 80-0-89-0 Coeff, of Variation 3°8 1+6 2-2 26 5:3 THE PLANT ECOLOGY OF PART OF THE MOUNT LOFTY RANGES (1) By R.L. SPECHT AND R. A. PERRY Summary This paper deals with the ecology of that part of the Mount Lofty Ranges between the Torrens Gorge in the north and Noarlunga in the south. THE PLANT ECOLOGY OF PART OF THE MOUNT LOFTY wb o&N Seem 91 By R. L. Srecut and R, A. Perry * [Read 10 June 1948] CoNTENTS INTRODUCTION... sla TOrocRAPHY nes GEGLOGY .. _- +. *3 CLIMATE .. ee = SoILs _ a! = “a ae on (1) Lateritic poiteuts ve 7 “ te a Sete (2) Padsols a's w. a x, AA e (3) Grey-brown podsols 4 bs i ae a (4) Calcimorphic soils Hs Pe rr av) + (5) Red-brown earths “s ts oe ate he (6) Ferrimorphic soils , +t ] . (7) Deep sands with neutral reaction (8) Miscellanecus. soils oF 2 VEGETATION ala _ ‘ ae A, Environmental Range of— ots rs _ (la) £. odorata = ot (1b) FE. odarata. “Whipstick Mallee" type ; E, leucoxylon rs us at E. vintinalis vi E. camaldulensis (syn BE. rastraia) (5) £E. obliqua E £. E. E . fasciculosa , cosmophylla . rubida ad “y 23 4: Fi (10) £, elaeophora Me Ss + _ -e (11) Casuarina stricta .- os a “9 a (12) Hybrids =“ 88 t zs a ee (13) Formations - _ te - Sa B. Classification of the Plant Communities (1) £. odorata association y (2) E, lewcoxylon — E. vinvinalis aésotiation (3) E. camaldulensis association wf + (4) Casuarina stricta association... i. (5) “Stringybark” edaphic complex a4 E. oblique association J oe E. obliqua— BE, Baxteri association E. Baxtert — E. cosmophylla association E. fascicilosa association .. «s + «fs E. rubida association Ds oe - a (6) Ecotones be - Rie mt (ff) Miscellancous communities SuMMARY 7 me @ fs ag és % ACKNOWLEDGMENTS REFERENCES APPENDICES _ us a ~- -s (1) Mechanical and chemical Gaatyses of sails (2) Comparative floristic lists of five major vegetational groups . Baxteri an 7 24 ne <2 RANGES (1) * Department of Botany, University of Adelaide, Trans. Roy. Soc. S. Aust.. 72, (1), 24rd August, 1948 92 INTRODUCTION This paper deals with the ecology of that part of the Mount Lofty Ranges between the Torrens Gorge in the north and Noarlunga in the south. The area surveyed by Specht lies between the Torrens Gorge in the north, grid ne 80 of the Adelaide and Echunga ordnance map in the east an! grid line 68 in the south, while Perry surveyed the arca from grid line H8 in the north to $2 in the south extending east from the coast te grid dine 68. The only previous ecological work on this area was a reconhaissance survey by Adamson and Osborn (1), As can be expected from reconnaissance Survey, sume generalisations were made which probably apply to part of the Mount Lofty Ranges but nat necessarily to the whole. Adamson anil Cshern did not publish any vegetation map, nor did they indicate in the lext the exact localities studied. Since then detailed work on the soils of the southern portion of the Hundred of Kuitpo, by Taylor and ()'Donnell (21) and on the geology by Sprigg (17, 18, 19) form a useful background. te this study of ecology. Considering the lmiited size but complexity of the area, it was deemed necéssary that the ecology should be approached from «a study of the autecology of the dominant tree species and of the formations. With this in view the distribution of the tree species and formations was mapped by projecting their limits on a contour map. As many factors as possible af the environment were examined ind attempts were made to cortelate the distribution of the species with the environment. The soils were surveyell by borings at each grid intersection on the ordnance survey maps, ie, at approximately 1,000 yard intervals. Because of the complexity of the environment the study of the vegetation could be regarded as a number of specially designed experiments in which possibly only one factor at a time was variable. For instance, it was found that soils of different nutrient status oceurred contiguously under the same climatic conditions, while soils of one graup extended [rom regions of low to high rainfall. Of course, any geheralisatiom can only be accepted on the understanding that it may not necessarily apply te other areas af the Mount Lotty Range. However, although other surveys of the ecology of this State have been of rather a broad nature in comparison, they have yielded conclusions roughly sinrilar lo some suggested in this survey. These have been indicated it their appropriate places within the text. It must be pointed out that in Healing with the trees whose roots tend ta penetrate to great depth, the early stages of growth are considered crucial in their development within the environment in which they are found. Ais most of the soils throughout the area are extremely pour most of the area has been left relatively undeveloped. It is only along the coastal parts in the south, the Adelaide Plains and fertile valleys and ridges of the Hills, that much agricultural development has taken place, Most of the soils al the coastal area are planted with vineyards and orchards, but some cereals are grown. The Adelaide Plain, althongh now closely settled, was unre an extensive cereal growing area, while the small fertile portions of the hills ave planted with orchards, vegetables and seme pastures. The savannah woodlands are used in their natural state for grazing: TOPOGRAPHY The area forms part of an original extensive peneplain which has been block-faulted during the Kosciuskan period (probably Pleistocene) and rubseqnently dissected by three major stream systems—namely fram north to south, the Rivers Torrens, Sturt, and Onkaparinga. 93 Creeks named First, Second, Thire, F outth, Fifth and Sixth Creek form the major tributaries of the River Torrens, and Cox's, Aldgate and Scott’s Creeks flow into the River Onkaparinga. Both the River Torrens and the Onkaparinga are part of the old river systems present belore the faulting cecurred. They must have been large enough to keep pace with the rise in level due to faulting and have cut meandering gorges up to 500 feet in depth. Smaller creeks such as Brownhill, Christie’s and Reynella Creeks aid the drainage of the area, In the southern section of the area three distinct faults are present. The land rises sharply from the gently rising Adelaide Plains along the Eden 30" a e4DELAIDE ST VINCENT GULF RAINF ALL. MAP LEGEND >Z200C" i5oo- 2000! 100-1500" 50o0-i0Q0" o-S500' ® RAIN GAUGE eNOARLUNGA 20" pel ALS Fig, 1 Rainfall and altitude rhoap. 94 fault (with a throw of 600 to 700 feet) on to the Eden-Moana Block, again at the Ochre Cove Fault Block and still again with a throw of 700 feet along the Willunga Fault. The original peneplain level gradually slopes to the south-west along these three fault blocks, the level of the Ochre Cove Fault Block in particular dropping from 1,200 feet north of Clarendon to 400 feet near Noarlunga. In the northern section, however, only the Eden Fault is prominent but the topography is complicated by the indistinct extremities of the Ochre Cove and Kitchener Faults, while the Burnside Fault occurring a little to the west of the Eden Fault aids in confusion. The monadnock structure of Mt, Lofty, which rises to 2,234 feet, dominates the topography. A general idea of the topography can be gained from the rainfall map on which contours for the 500 feet level and multiples thereof are indicated (Fig. 1). The dissection is influenced by the nature of the rock, the slates and phyllites giving more rounded and gentler sloping ridges than those developed in the less rapidly weathering quartzites. Plate V shows a photograph of a model made of the central section of the area. GEOLOGY Sprigg’s Reconnaissance Survey of the Geology of the Mt, Lofty Ranges (between the rivers Sturt and Torrens) (19) and the Geology of the Eden- Moana Fault Block (17) gives a detailed account of the geology of the area. The undermass rocks consist of a series of Pte-Cambrian sediments, called the Adelaide Series, which have been extensively faulted and folded. 8 h ADELAIDE pe. — —BELAIR i 4 ----STIRLING IN INCHES RAINFALL Fig. 2 Mean monthly rainfall at Ade‘aide, Belair and Stirting. 95 190 ADELAIDE — — BELAIR -—- STIRLING IN TEMPERATURE Vig. 3 Mean monthly temperature at Adelaide, Belair and Stirling, Of importance to the ecology of the area are the metamorphosed rocks of the Barossian Complex, the sandstones, quartzites and phyllites of the Lower Adelaide Series (the easterly half of the area) and the slates, Sturtian tillite and flaggy slates and quartzites of the Upper Adelaide Series (the westerly half). Overlying the Adelaide Series on the Eden-Moana Fault Block from Blackwood to Happy Valley Reservoir are remnants of a lacustrine deposit of consolidated Oligocene sands on the old peneplain surface. In addition large areas of Miocene marine sediments, mainly marls and limestone, and limited areas of Pleistocene raised sea beaches occur on this block. Since the Pleistocene the Torrens River and Brownhill Creek systems have built up the large alluvial deposits of the Adelaide Plain with material dissected from the fault blocks to the east. In recent times the River Onkaparinga has developed an extensive alluvial flood plain which extends several miles inland from its mouth. CLIMATE The meteorological data available from stations within and adjacent to the area suggest that there is a close correlation between altitude and climate, Rainfall data indicate an increase in rainfall with altitude (sce rainfall map) rising from 17-94 inches per annum at Glenelg to 47-84 inches G 96 per annum on Mt. Lofty Summit, with a deterioration in rainfall to the north and in the rainshadow of Mt, Lofty Summit, This deterioration is caused by the fact that most rains are borne by south-west winds. In all cases February is the driest while June is the wettest month. The data available for Adelaide, Belair and Stirling West show that there is an increase in rainfall and humidity and a corresponding decrease in temperature with altitude (sce Figs. 2 and 3). l2 ADELAIDE —-— BELAIR I {' ——=- STIRLING (P/E FROM P/S.D) TRANSEAU RATIO Fig. 4 Mean monthly yaliies for Transeau ratio at Adelaide, Belair and Stirling. From the average humidity and average temperature, the Meyer and Transeau ratios have been calculated, both of which show an _ increase with altitude (see Fig. 4). From this data it is seen that Adelaide has an influential rain period (P/E greater than 1/3) for 6°8 months, Belair for 8:5 months and Stirling West for 10‘7 months with influential rainfalls of 15:9 inches, 26°3 inches and 45-8 inches respectively (Trumble 23). When Davidson's less accurate P/E greater than 1/2 is used, Adelaide lies within Davidson’s warm temperate semi-arid zone while Belair and Stirlmg West lie within his warm temperate semi-humid zone (10). However, it must be remembered that all the above data give a rough approximation only of the moisture relations within the soil, for the dissection of the country and the water-retaining capacity of the soils play a large part in producing microclimates. Much of the influential rains is lost to the soil as “run-off” into the creeks, which consequently are moister habitats, while a clay soil will tend to retain more moisture than a sandy soil in which most of the G7 water percolates through into the subsoil or to the underlying rock, Hence ‘under the same conditions of physiography and climate a deep sand will tend «© be drier than a shallow sand over a clay subsoil which im turn will tend to be drier than a soil with a predominantly clayey profile, Tt must also he remembered that Trumble (23) in proposing his influential rainfall period as those months with P/E greater than 1/3, based it on the time intetval wvet which the surface soil of the Waite Agricultural Research Institute tends tu be maintained above the wilting point for herbage plants. It does nat necessarily follow that the same factor of 1/3 holds for all other plants and seedlings. Iusolation has a marked influence in producing microclimates in dissected country; that is the micro-climates are also determined by aspect. Although no attempt has been made to demonstrate this for the Adelaide Hills, the southern sides of the ridges are apparently moister than the northern sides. This is largely due to the direct effect of insolation on evaporation—the sunny, warmer, north-facing slope being drier than the more shaded, cooler, south-facing slope. This difference in micro-climate may also be enhanced by the fact that most of the rain is borne from the south-west (this data is available only for Adelaide but should be similar for the rest of the area), hence giving a minor rain-shadow on the northerly sides of the ridge. Consequently, with the decrease in temperature with altitude, the figures fer Adelaide, Belair and Stirling West indicate that active growth of herbage plants (average monthly temperature Iess than 55° F) is restricted for 3, 4 and 6 of the winter months respectively, and moderate growth (less than 50° &.) for nil, 1 and 3 months respectively. (Trumble 23). Hours of sunlight and frosts appear to have little effect on the general vegetation, but may play a part in the agriculture of the area, From the data presented above it is evident that the area can be roughly divided into at least three main climatic regions, namely: (1) Country less than 500 feet altitiide with an influential rainfall period of 6 to 7 months. {2) Country 500 to 1,500 feet altitude with an influential rainfall period at 7 to 9 months. (3) Country greater than 1,500 feet altitude with an influental rainfall period of 9 to 11 months. with a Jarge number of micro-climates produced within each zone, SOILS In the seuthern two-thirds of the area a reconnaissance was made of the pedology by boring holes approximately 1,000 yards apart, while the remainder of the atca was mapped from observations made on profiles exposed in cuttings and by a superficial examination of the soil. As the sutvey Was on a broad scale no attempt has been made to map soil types, the survey being only sufficient to pive an outline of the sail groups present. Where the rocks of the Adelaide Serics or overmass sedinients ate exposed there is a marked corrclaiion between the geology and the soil while. in the south, soils derived from lateritic soils developed on the former peneplain surface tend to predominate, The soils can he grouped as follaws:— A. Soils derived from Plio-Pleistoverie lateritic soils; {1) Residual lateritic podsols. (2) Trimeated lateritic podsols, 98 (3) Deep siliceous sands with strongly acid reaction. (4) Lateritic podsols under conditions of poor drainage. B. Soils formed under the present environmental conditions: (1) Podsols. (2) Grey-brown podsols: (a) With high nutrient status, (b) With low nutrient status. (3) Calcimorphic soils: (a) Rendzinas. (b) Terra rossas. (c) Degraded rendzinas. (4) Red-brown earths. (5) Ferrimorphic soils (?). (6) Deep sands of neutral reaction. (7) Miscellaneous soils. RESIDUAL | TERTIARY LATERITIC DEEP SANDS SOILS Sala TRUNCATED p: 4 BAROSSIAN / L oReMIC ERR I MORP 3 COMPLEX SOILS uJ quarTzites \ < ‘ MOSTLY SKELETAL SKELETAL AND = t 1 iL SANDSTONES POOSALS a | 3 MATURE PODSOLS DEEP NELITRAL ! = SANDS Zz ree BASIE SLATES REO BAOWN GAaEr BROWN GAEY aS OWn < AND SHALES ; 4 A 7 AND STURTIAN EARTHS PODSOLS a L, PODSOLS TLLELITE ; to) (ica NUTRIENT) (Low KUTRIENT) TERRA ROS SAS ~RERRADED MEN ZINN AS e EIMESTONE. f ae , NO LIMESTONE z RENZINAS , Fa 1) s 20 2s 30 45 ao aS RA'NFALL 'N INCHES Fig. 5 Diagram showing the soils in relation to rainfall and parent material, / The relation of the soils to rainfall and parent material is illustrated in fig. 5. The parent materials are arranged from bottom to top in decreasing order of base content. Thus, generally, the soils at the top of the diagram have a lower nutrient status and a lower pH than those at the bottom. Vor the pur- pose of the diagram, the Tertiary lateritic podsols are regarded as the parent material from which the present-day lateritic soils have been derived. A. Sorts DERIVED rrom Puro-Preistocene Laterirtc Soiis These soils are reli¢s from a more pluvial climate in the Pliocene when they were developed aver most of the former peneplain surface. The profile developed during this period is that of a highly leached podsol in which a seasonally fuctuat- 99 ing water-table produced. concretions of ferruginous material near the junction at the sand and the clay horizons. Oiten ferruginisation appears as pockets through the clay and the parent tmaterial (Stephens 20). The lateritic conere- tions show great variation, In some cases they are massive while, in others, pisolitic gravel and nodules oceur. Over the Oligocene lacustrine deposits a ferruginised sandstone or a ferruginised conglomerate of water-worn quartz pebbles occurs in the position of the lateritic concretions in the typical lateritic profile, However, when faulting broke up the old land surface im the Pleistocene, dissection began and the lateritic profile was truncated lo varying degrees, Over most of the area; especially in the northern half, complete truncation, down to the parent rock has occurred. Eviderice of the former latetitic profile is. frequently seen) by interise ferruginisation of the rack on the tops of the ridges. These rocks thus exposed have developed soils in equilibrium with the present climate (see below). Lateritic soils in varying stages of truncation occur on the tops of the ridges which constitute the former penepiain level on the Eden-Moana, Ochre Cave and Willunga Fault Blocks. As shown in the section on topography, all these fault blacks ate tilted to the south-west, On the Eden—Moana Fault Block complete truncation of the lateritic profile occurs in its northern limit, At Belair the profile is truncated down to the lateritic gravel (a truncated laterite), the sand of the A horizon having drifted down the gradient to the south-west, or eroded down the valley. At Blackwood a little to the south a shallow sandy A horizon is present, this horizon becoming deeper as we move further to the south, while very deep sands (over six feet deep) occur near Happy Valley Reservoir, It appears that the sand of the A horizon has gradually drifted down the fault block, stripping the profile in the north and accumulating in the south. Of course, much of the sand will be washed away down the valleys, and there is a possibility that the sands were winnowed during the arid period proposed by Crocker (8)- Also there may have been a supplement of sand from the higher fault blocks. Hence, conseyuent to these movements, the results of niechanical analysis (see Appendix 1) show that the coarse sand: fine sand ratios of the sand and the clay horizons do not agree as found by Northcote for a residual laterite on Kan- garoo Island (12). Similarly on the Ochre Cove Fault Block, the truncated Jaterite is seen ai Cherry Gardens, while further to the south a sandy A horizon, which becomes progressively deeper, occurs, Near Blewett's Springs and McLaren Flat, to the south of the arca surveyed, deep tateritic sand acctimula- tions occur. On this fault block the laterite extends for at least 30 miles to the sotth, On the Willunga Fault Block only truncated laterite has beem observed as yel, his truncated laterite has been called Kuitpo gravelly sandy loam by Taylor and O'Donnell (21). The Bradbury kaolin mine is indicative of truncation of the deep lateritic profile to the pallid clay horizon. In recent times up to 12 inches of grey clayey soil of poor nutrient status has developed over the kaolin, For cofivenience the lateritized soils will be divided into four classes. the general profile characteristics of which are illustrated in hg. 6, 1. Residual lateritic podsols This soil is characterised by a coarse sandy surface which may vary from 8 to 27 inches in depth and which contains wbout 509 of ironstone gravel in ihe lower six inches. Quartz gravel often occurs with the ironstone, Owing to accunulations of organie maller near the surface the upper tew inches are nsualiy dark grey in colour, while (he remainder of the sandy horizons are light grey to 100 FROFILES CHARACTERISTIC OF SOILS DERIVED FRON LATERITIC FODSULS, RESIDUAL LATERITE TRUNCATED LATERITE dork erey Plt sand With orgenic matter ugh Deed 2 sandy loam with leteritic Erey to orange with gravel and organic matter yallow grey, red mottlings 7" orenge with 12 sandy elay-loam to cley red ae sand light grey loam with lateriti¢ grevel 1a" ay grey 23" eand + leteritic gravel + guertz clay sandy clay to clay age DEEP _LATERITIC SAND UNDER POOR DRAINAGE erey dark grey wand eand to sendy Louw red brown yellow grey clay + lroustone ko light red brown to grey and light grey light grey olay with red mottlings ironatone at depth Fig. 6 Soil profiles characteristic of soils derived from lateritic podsols. yellow-grey. Underlying the sandy surface horizon there is a mottled yellow- brown, yellow-grey and light grey clay which may contain some ironstone gravel in the upper few inches, The light grey colour usually increases with depth. Mechanical analyses show great variation in the coarse sand : fine sand ratio, possibly because of the movements of the A horizons as mentioned above. The restilts of the chemical analyses show the soils to be extremely poor in P,O, and nitrogen. Much of the organic matter is in a fibrous condition and does not break down readily to available forms of nitrogen. Throughout the profile the soil is acid in reaction (see Appendix I). 2. Truncated lateritic podsols These soils are similar in formation to the previous one, but there is little of the sandy A horizon present. The ironstone may be in large boulders or broken down to pisolitic gravel. 10] The profile cousists of a brown lo grey loamy suriace overlying at a depth of 1 to 7 inches a yellow-brown to red-brown clay contaming red mottlings. There are often large pieces of itonstone 6 to 12 inches in diameter near the surface of the soil. At about 12 inches the clay beconres mottled red-brown, yellow-brown and light-grey. The light grey colour increases with depth unul, below about 2 or 3 feet, it is by far the dominant colour with only a few red- brown inclusions in it. The results of chemical analysis show that these soils are slightly higher in nutrient status than the residual Jateritic podsol, The nitrogen figure is not a trae index of the available nitrogen, as most of it is present as undecomposed organic matter (see Appendix I). 3. Deep lateritic sands of acid reaction Within the area of residual tateritic podsols there are several areas of deep sands contaifing ironstone at depth. The profile consists of a grey sand over- lying a yellow-grey to light grey sand at variable depth. Near Blewett’s Springs these sands are predominantly coarse, the ratio nf coarse sand to fine sand being 8 or 10 to 1. The silt and clay fraction is never higher than 3-5% in the first 36 itiches. The high coarse sand : fine sand ratio suggests that these sands may have been winnowed by the winds of the arid period (8). The mechanical composition is remarkably constant with depth. These soils are the most acidic (pH 5°0-5:5) and the poorest in atitricnts of all the soils in the area (see Appendix I). 4, Lateritie podsols under swampy conditions In several small isolated positions within the range of distribution of the residual lateritic podsol these soils occur in flat and relatively low-lying situations. Under these circumstances they become waterloyged in winter, due to the imper- vious nature of the underlying clay. The profile is similar to the residtial laterite piodsols. B, Sols TORMED UNDER PRESENT ENVIRONMENTAL CONDITIONS Where complete truncation of the lateritic profile has occurred, soils have developed on the underlying rock in recent times. These soils show a close correlation with the parent material, podsols being developed over siliceous rocks, grey-brown podsols and ted-brown earths over argillaceous rocks, calcimarphic soils over calcateous rocks, and ferrimorphic soils over gneissic and schistose rocks. Chrnate has been responsible for the differentiation of some of the scil groups. The red-brown earth grades into the grey-brown podsol at the 25-inch isohyet, and this in turn is progressively more leached with the increase in influential rainfall until the nutrient status is so law that a definite sclerophyflous formation takes the place of the savannah formation about the 35-inch isolyet. In areas of impeded drainage deep degraded rendzinas occur. These grade into shallow rendzinas in drier conditions. Products of denudation tend to accumulate in some valleys forming a valley complex, while a saline alluvial flat hag formed near the mouth of the River Onkaparinga. 1. Puaésols Wherever siliccous rocks of low cation status, such as quarizites and sand- stones, ate exposed, podsols develop. As these rocks are very resistant to weathering, most of the area over which they occur is very rugged and bold m outline and skeletal podsols are developed (see fig. 7), However, in regions of rainfall greater than 35 inches per annum and on the tops of ridges the moisture 102 has been sufficient to severely leach the weathered material of such low cation status, and to produce a mature podsolic profile. Typically, the mature podsol shows a sandy A horizon containing floaters overlying a B horizon of sandy clay to clay (see fig. 7). On analysis these soils were seen to be low in P,O; and total nitrogen and acid in reaction (see Appendix I). PROFILES CHARACTERISTIC OF GREY BROWN PODSOLS AND PODSOLS GREY BROWN PODSELS DExP TEASE . SHALLOW PHASE dork grey 1" Be esate dark grey =n olay Lom with state and an dark brown an 3” organic matter derk brown yellow brown +4 clay loam +/- slate clay loam + alte red brown + /equartz 1s7 15" ig" slate clay + /- slate ag" decomposing slate FODSOLS DEEP PHASE SHALLOW PHASE dark send with organic matter ark grey ys = and quartzite durk grey 1" : sand with orgsnic matter dere grey td gs Pa eee Pe and quertzite light yellow * . dark to light brown send+ /+ quartzite +/~ grey ind o qgenkngts erange quartz 1} rsm 20" quartzite sandy clay to clay + /- quartette 48 decomposing quarteita Fig, 7 Soil profiles characteristic of grey-brown podsols and podsols. However, small areas of quartzites in the Middle and Upper Adelaide Series tend to give skeletal podsols of a slightly higher nutrient status than siliceous rocks of the ower Adelaide Series (see Appendix I), but these outcrops are apparently ecologically insignificant in their effect on the distribution of the tree stratum. 2. Grey-brown podsols In that section of the area with rainfall greater than 25 inches per annum, argillaceous rocks, such as slates, phyllites and Sturtian tillite have given rise to a profile typical of a grey-brown podsol. In the mature phase this soil shows an A horizon of a clay-loam sometimes containing floaters over a clayey B horizon 103 (sce fiy. 7). This mature profile is developed more extensively than in the case nf a pousol (see above} because of the faster rate of weathering of the argilla- ceous rocks, but it is scen especially in wetter regions (rainfall greater than 30 jnches per annum) and on the tops of ridges. In drier regions between 25 and 30 inches per aanum weathering is probably not so rapid, as erosion is nearer equilibrium with deromposition and only an immature form of the soil is developed (see fig. 7), Analyses show that io the drier extent of the grey-brown podsols the soil is relatively higher in P.O, and total nitrogen but grades into a soil about three times lower in P,Q. and total nitrogen approximately along the 35-inch isohyct (see Appendix 1). [It appears that this decrease in nutrients is caused by increased leaching with increased rainfail, This is supported by the fact that the savannah formation on the relatively high nutrient status grades imperceptibiy into. sclerophyllous formation on the relatively low nutrient grey-brown podsols with increase in rain- fall, In all cases the profile is acidic. However, the low nutrient grey-brown podsol is from three to four times higher in nutrient status than the podsol. As would be expected in sedimentary rocks, variations acctir within the slate, some tending to be more siliceous in composttion than others, These slates con- sequently give tise to a profile sandier than that described above. 3. Calcimorphi¢ soils The main area om which these soils are developed lies between the coast and the Ochre Cove Fault scarp, Thick beds of Miocene marine limestone occur in the southern half of this faull block, while cappings of travertitie limestone have overlain the Adelaide Series in the northern halt, and Miocene limestone in the southern half of the fault block. On these limestones and other limestone out- crops of the Adelaide Series alkaline calcimorphic soils have developed. Vhe travertine limestone probably had a lJoessal origi durmg the arid Pleistocene to Recent Periods as proposed by Cracker (8). With faulting in the Pliocene to Pleistocene, dissection of the original peneplain would occur, pro- ducing valleys and completely eroding most of the lateritic profile formed in the Pliveene. Later in the arid period toessal maller would be deposited aver the soil of the whole area and leached to varying degrees, depending on the influen- tial rainfall pattern. We can assume thar, in the Adelaide Hills, the pattern of annual rainfall must have been much the same as at present (see fig. 1), but af course lower in amount, Consequently the calciumt carbonate was entirely leached out in the higher rainfall regions, either then er under ihe present climate. However, in ihe moce arid region—now indicated by a rainfall less than 25 inches per annum—the ealcarcous matter was leached down to the underlying rock and there deposited as travertine, This travertine would be formed over the whole topography but subsequent erosion, especially in the present moister climate, has worn it away greatly on the slopes and completely in the valleys where the pnderlying rock is exposed. Of course, it cannot be dismissed that the calcareous slates and quartzite might have added to the travertine, but it is too widespread aver non- calcareous rocks, especially near Tapley’s Hull, for this to account entirely for its origin. One would expect to find travertine overlying laterite in some places. But, as yet. the only indications are the presence of ironstone within the traver- tine on Shepherd’s Hill and around the degraded rendzina on O'Halloran Hill (Sprigg 17). This laterite may be detrital lumps from higher altitudes. Stephens has also noticed a buried Jateritic profile about a mile north of Christie's Beach (private communication). 104 SOIL PROFILES RENDZINAS DEGRADED RENDZINAS dark grey an loam to clay loss dark groy grey brow: 6" 12 glayay loam + ¢aleium Bw carbonate particles travertine limestone grey to white br 38" underlying rock heavy ¢lay +much coleium carbonate DEEP NEUTRAL SANDS grey eh) eal saod with organic msetter Be Yellow grey to light grey sand sottled ferruginous sendatone Fig. 8 Soil profiles characteristic of rendzinas, degraded rendzinas and deep neutral sands. Over most of the travertine a dark greyish-brown loam has developed which shows very little differentiation into horizons. The depth of the soil may vary from 3 to 15 inches, but in many places the travertine is visible above the surface (fig. 8). The soils are well supplied with organic matter and have a fairly high nutrient status. These soils—rendzinas— have been cleared of vegetation and the land used extensively for cereal growing and vineyards. Interspersed with the rendzinas near Reynella are small areas of red soil of lighter texture, which resemble terra rossas. Soils allied to terra rossas have developed on the Torrens and Beaumont limestone outcrops in the northern half of the area, where they grade into the acid grey-brown podsols. Under conditions of retarded -drainage the limestone has undergone extreme decomposition, producing a deep degraded rendzina of heavy texture (fig. 8). There is a narrow fringe of these deep black soils at the boundary between the shallow rendzina soils and the grey-brown podsols. 10g Along the Eden Fault scarp these caleimworphic soils prade into the red- brown earths of the Adelaide Plains, while to the east of their extent they grade into soils allied to red-brown earths developed over slates. 4. Red-brown earths This soil is developed on the Adelaide Plains where the rainfall is less than 25 inches per annum, and appears to grade imperceptibly into grey-brown podsols along the Eden Fault scarp and into the calcimorphic soils in the south-west corner of the Plain. As no survey or analyses have been made over this area, the reader is referred to Piper (13) who gives a detailed description and analysis of the profile developed at the Waite Institute (the Belalie loam). In general the soil is relatively high in P,O, and nitrogen and becomes alkaline in the fi horizon, due to acctimulation of calcium carbonate, It is possihle that the calcium carbonate in the profile is of the same loessal origin as that proposed by Crocker to explain the origin of the mallee soils (8), the loess being deposited oyer the deep detrital soils of the Plain during {he arid Pleistocene to Recent Periods and subsequently leached, especially in the moister present climate, to form the red-brown earth profile, The degree of leaching, especially of the calcium carbonate, depends to a large extent on the rainfall, for at about the 25-inch isohyet there is a transition to allied grey-brown podsols which have no trace of calcium carbonate in the profile, 5. Ferrimorpiie soils (?) As these soils are of limited extent to the east of Mount Lofty Summit and along the Torrens Gorge, little data is available. They are developed on schistose and gneissic rocks of (he Barossian Complex and appear to be lacking in profile development. Superficially they are similar to the shallow form of the grey- brown podsol and in some cases are indistinguishable from it. From the sclero- phyllous nature of the undershrubs, they are apparently Jow in nutrient status. As large areas of these soils occur ta the north of the Torrens Gorge a more com- plete examination of these soils will be made by one of us in a subsequent paper. 6, Deep sands with neutral reaction These soils are of limited occurrence and occur mainly near Noarlunga and TYackham, The soil consists of a grey surface horizon containing some organic matter overlying a yellow-red or a yellow-grey sand at 3 to 8 inches. The under- lying C horizon is a mottled light grey, yellow and red ferruginous sandstone {see fig. 8). The soil is slightly alkaline throughout and has a poor nutrient status. The coarse sand figure is very high, which suggests that the sand from which the underlying sandstone was formed had been winnowed, 7. Miscellaneous soils Several other soils in addition to those described above occur only it small areas, (1) Along the drainage lines which have cut into the Willunga Fault Scarp near Kangarilla there is a soil which is waterlogged throughout the year. The profile consists of two inches of grey sand containing organic matter overlying a light grey clay, which is particularly heavy and sticky. A sclerophyllous formation dominated by E. leucoxylon is developed over this soil (see Appendix T), (2) Near Seaview a red sandy soil occurs. 18 inches of red coarse sand over- Kes ced clay, which becomes heavier and yellower at 24 inches (see Appendix 1), 106 TaBie [ UOHVIOOSSE Deo LIsHy “A puke wuowgErs -OSSE S1DUMiNE “FY-UOpLXoINay “sy ‘UO! -BISOSSE BPAOpO FF ysaaq BOOT “Sb 2Auge spusyxe ¢ .Sf— SE Wory sasingsia}E AA , SUOIPEIIOSSe Mayreg ‘q-onbyqo'gy Wey} «SUOAELL = IayEM Pue SMyEIS JUALINU 4o}eq JO fics uO ase. [EL UIT. SIQ] UTI suOtsod sap uO jospod iAGaq -4913 UO PURTpooOm = YyEuULAes apaazey uo purpoom [psqdorspog SPOS [RIAN|e. pue sjos -pod umoug-feas poysvey ATYRLY sjospod umoarq- aad payavay ATYSIET SPIEL PUR S[OSpOg uOE — SE ‘ed Bh -SE ‘ed ,8h- SE ‘ed weak — SE saspi yo sjoodse Amung Saqiaye, pue spospog ‘ed Op— ce sapsaye, penpisad tadasp uojtroonay "sf Jo Aqperadsa sayfiayey pure spospo,T S3aI]} aIEY “SpUMWAsINbas JayBAL MOT snes WaMu MoT 0} WENTpAyy ed OF ~~ gadojs pejapays doays JO worjowo.4 S}iseimber mayen Mop A194, jeynan yO pues snoaots daaq ed ce -02 ‘,2@ PAaqe spuojxd $(f08 ‘,2@ MOjeq sSasdInod1a};EM SoSUTs yp Runge pues spospod waosq-So17) ‘ed co -07 ‘syrRd 49}}0M. UL sfayjea pue sjadse Furdezy-yynvg (q) WOE > spadse Furaez-1yNVSg AL < Spadse suvory-ywonN (2) ‘SN}EIS JUALI|NU. YBIY JO S}105 spospod ussoig ‘wed Op-OE pue suoyEar sajea Aq parjosjuoD -Aaiz pue Saytsayey = payeounsy, ed (OP—SZ USZ MOlaq dAatsi19yXa !yAAYOSY OF eau Saysazey ‘S[OS syoadse BuUlsey-ysICN “f!OS TUTTE -pod uMorq-Aei3 ‘spospod ‘syjara sonnel IojEM Aq Apuo payjorusy UAMOIG-pat ‘syios — aRYydsqUTaTRD) ‘ed (Of -02 SHILA [tos a2u0e yy Weyaied YeuUeAES NOjALOINA] “q-pjpsopo “ay ~¢ WAydasayas SyDUAUeya “Tb ypAydoaayas wo[KxoIMa! “Yz ¢ 1PAydo -10]98 DpD4opo "FT q\Sydossjas wo/Cro9 nay y-pyDAOpO “| T spuyj;pooM | auUo}oaa HsLozHIsns YRUUEARS “quai xumna| pur “G]-DIDAU PO ‘57 [Aydoraqag SUOTPEOSSE BS9q} ‘uaamjoq $9u0} -o9a opue soddy “9 Uo EIDOSSE Sato} wind 3teqa; pues Dpigue “Tp HAqdosaypas UOIPELIOSSE 4s010F sseqsSurys nabygo yp | tAydossyag uopEInosse. Ynbyqo 4Sod05 SIAR ASUS qsexog "We 1 Aqdor3]95 - purypoos. nnd qnads-yaeq | uoiyeroosse py Lyd 0} qni9g -AduIys UMoIg -soo 'q-Mapeng “FZ [Aydo.sazag UOTPEIIOSSE Puy] poom wns yu | psopnniasuf “yy WAqdosapys xu oydepa yreqéaursyg norETDOsse _. PQILINS yvoays DNAPS DULEDNSD) TAysosaps WOTPELIOSSE purppoom wns pry Sisuapnpyouns -y YSULEAES wang wauryy | add} synurdirs “yg wns ang adA, woptvomay yy UOTJEIIOSSE SUyUt PuElpoom ult “Y-uopkxoray “a YeUUeAes | pueypoom lund yunusedda gy | UO!EIOOSsE NEnAOPA “7 YRUNEAES | AyuIMUIO-) vowed T 2WEN woUTUOT) Trans. Roy, Soc. S. Aust., Aucust, 1948 SPECHT AND PERRY BROWN es ADELAIDE RED BROWN EARTHS se GLENELG RED BROWN EARTHS ST. VINCENT GULF + sme fos ry ee ee ° es es ef/eis stew ew ee ’ . Cores Or ast binetts . hb ww oer © @ © * oh se & @ crete P— ac 9 peer KA “ 7 TF 64 <7 i. SOl. MAP ———— pr 4 —-LECEND — GREY BROWN PODSOL ST 2 a, ) PODSOL LATERITIC POOSOL RENDZINAS avo TERRA ROSSAS DECRADED RENDZINAS DEEP SANDS or NEUTRAL REACTION FERRIMORPHIC (?) SOILS SALINE ALLUVIUM EA = (Lil =] ‘ We \seaae \\ Bs KS CHA ASN — ir I 107 (3) Saline alluvial flats have been developed near the mouth of the Onka- paringa River. These consist mainly of variable Recent alluvium con- taining shells. Parts are subject to flooding on exceptionally high tides, as there is no part of the flats higher than 10 feet above sea level. RAINFALL w INCHES ny w) £ a1 9 ° 5 ° NUTRIENT STATUS LOW MEDIUM HIGH Solerophy]l Solerophy1) Sevannah Savannsh c Casuarina striate Casuarina atricta BE. odorats uw ot o et bad 2 5 * bt EX. loucoxylon. c w H E. leucoxylon x z e £ on E. vyiminalia a 4 B. viminslia — & 2 & w rod oa B. camaldulensies X, camsidulensis i! wv — 2&8 s e E. faacicnlosa 5 BE. fasciouloss & ry oo} oF ow > 8 z ae E. Baexters ar % E, obliqua E. oblique E, sosmephylla B, cosmophylla “Ulos Fo Sgpoedeo #ursyerea seiee MoT = 9 pus wnyped = _ydye = y iAwTTeA = A ‘yoedse oLequos | g (4oedey ULayyiog = Nf N z — I zt + ne EF, robide J. p —— Fig. 9 Diagram showing the relationship of the dominant Eucalypt species. with rainfall and nutrient status of the soils. THE VEGETATION It was found on reconnaissance that numerous changes in the species com- position of the vegetation occur over a very small area. Consequently it was 108 deemed advisable that the autecology of individual species should be investigated. A summary of the results obtained is shown in fig. 9, 10 and il. Following the discussion on the environmental range of the species, there is an endeavour to group the species into logical commutiitics, but of course any attempt is largely artificial and only for convenience. The nomenclature of the species follows that of Black (3) and Blakely (4). E.odorata OO [| SSRRSRS RRR SS as Ww ED} Diagram illastrating the distribution of various species of Eucalyptus in relation to soils and rainfall. The numbers on the figures refer to soils natned in Figure 5. Hatching from right to left (downwards) indicates presence on South-facing aspects only- Hatching from left to right (downwards) indicates presence on North-facing aspects only. Cross-hatchiug indicates presence on hoth aspects. A. EnyrronMENTAL RANGE 1A. Eucalyptus odorata (peppermint gunt) This species warrants further mvestigation in regard to its acceptance as 2 genotype, for, especially in its eastern limit, the peppermint gums exhibit a large variation in characteristic features. R. G. Brett (private communication) sugeests that the species is actually a hybrid polymorph between Eucalyptus odorata, Eucalyptus fasciculosa, and possibly a mallee with E. odorata the dominant member of the cross. However as all peppermint gums, which show the same wide range in morphology, agree with Blakely’s broad description of E, odorata, his description is accepted in this report. E. odorata occurs within the drier part of the area, the 30-inch isohyet forming approximately the upper limit of the species, while the coast forms a barrier to it reaching the lower limit of its potential range. The species appears to be uninfluenced by the texture, the pH, or the nutrient status of the soil, for it 109 flourishes equally well on red-brown earths, grey-brown podsols, podsols, residual and truncated Jaterites, and ovey the calcimorphic soils which occur within this rainfall range. It las been noted that towards ihe wetter limit of its distribution there is a marked tendency for the species to occur on the drier northern aspect only. Towards its drier limit the species is confined to soils of relatively high water-retaining capacity such as calcimorphic soils and red-brown earths, but as the rainfall increases, the species can spread over truncated laterites or residual laterites with a shallow A horizon, both of which though low in nutrient status have a clay horizon near the surface. However, the peppermint gums never occur on the deeper laterilic sands nor on the skeletal podsols developed over the thick quarizile of the Adelaide Series. These soils have a relatively low water-retaining capacity. These facts indicate that it is the moisture relations within the soil which are the major controlling factors in its distribution, This is illustrated im fig. 10, The potential range of the species in this area is far wider than that found in other parts of the State, for Crocker (7), Jessup (11) and Piper (13) have shown that the species is restricted, im the areas which they studied, to red-brown earths (sometimes becoming slightly podsolised) within rainfall limits of 15 to 25 inches per annum, On Eyre Peninsula (7), where the maximum rainfall is only 24 inches per annum, the species is restricted to the red-brown earths between 18 and 20 inches per annum, although podsols and residual podsols occur con- tiguous to it, Jessup (11) also deals with the drier limits (15 inches per annum) of the distribution of E. edorata on the eastern side of the Mount Lofty Ranges. Around Bordertown, in the Upper Sonth-East, the species flourishes on calci- morphic soils with a rainfall of 20 inches per annum. Hence it appears that in the drier limit of its range E. odorata is mainly confined ta alkaline soils of clayey nature, which have a high water-retaining capacity, whereas a wider range in regard Lo soils occurs towards its wetter limit, provided the higher rainfall and the lower water-retaining capacity of these soils compensate to satisfy the water requirements of the developing E, ederata. 1B. “Whipstick peppermint” ecotype of Eucalyptus odorata At the foot of Black Hill on a podsol developed over the Mount Lofty quartzite, two distinct areas of this “whipstick mallee" occur within the Z, fase#- culasa sclerophyll community. The tree appeats to approach Blakely's descrip- tion of E. odorata vat. angustifolia, although many of the trees exhibit fruits of the same size as the typical 2, odorata, As this ecotype of £. odorata is of restricted occurrence it can only be noted that the soil is a podsol, low in nutrient stats, and developed with a rainfall of approximately 25 inches per annum, 2. Encelyptus leucoxylon (blue gum) This species is largely controlled in its distribution by the moisture relations within the soil. It occurs withit the limits of the 25-inch and 40-inch rainfall isohyets. hut within this range it is markedly influenced hy aspect. In the drier limits the species occurs either mixed with FE. odorata or alone on the shady, moister, south-facing slope, but as the rainfall increases in about 30 inches per annum it occupies both slopes, while with further increase the species is confined ta the warmer, drier, north-facing slope of the ridges. This phencmenon is masked in the northern half of the arca where the rainfall increases rapidly along the scatp from 25 inches to 30 inches pev armum within three-quarters of a mile, 110 thus limiting E. /eucoaxylon to the north-facing slopes, whereas in the south the same increase occurs gradually aver two miles, Within this range occur grey- browti podsols with relatively high nutrient status, and residual and truncated laterites of relatively Jow nutrient status. However, although these soils differ im tiutnient status, their water-retaining capacities, apart from the shallow sandy horizon of the residual laterite, would be of the same-order. The fact that water relations play a large part in controlling the species distribution is further sub- slantiated by the appearance of £. Jevea@rylon on residual laterites in the fringing community around the deeper lateritic sands in the southern part of the area. Here E. fascieulasa js distributed over the deep sands of the ridges with E, leucoxylon occupying an area, often only a Tew chains wide, on the shallower sandy of the residual laterite, the clay horizon of which would tend to produce hetter water relations within the soil. In the wetter situations of the drainage lines and adjacerit flat areas, E. comaldulensts takes the place of F, leucoxylan. E, cumaldulensis also tends to supplant E. leucvaylon over the grey-brown podsals on wetter ridges along the Torrens Gorge, the Sturt River, and the River Onkaparinga. However, two exceptions to this general scheme occur. The podsols developed on the Mount Lofty quartzite act as a barrier to the distribution of the species. Wherever podsols occur, in whatever rainfall, there is a sharp division between the blue gums on the grey-brown podsols and the “stringybark complex” on the podsols. This is rather curious, for the pousols and laterites are of approximately the same acidity and nutrient status in P,Q, and nitrogen. Of course, most of the podsols are of a very skeletal nature, especially in thetr drier limit, and possess a low water-retaining capacity. This fact may be of value, but there is no gradual transition evident from blue gum to the “stringybark complex” as there ts with increase in rainfall and decrease in PO, atid mtrogen with leaching on the grey-brown podsols. This data suggests that some other factor is influencing its distribution as well as moisture relativn- ships. It may he inicro-nutrient deficiency on the podsnls, It may he bound up with competition between the trees of the “stringy-hark complex” and the “blue gums.” There is also a small occurrence of blue gums on deep Tateritic sands near Happy Valley Reservoir which seems an exception to the general scheme of water relations. Tig. 10 shows the general relationship existing between J. Jeucaxylou, soils, and rainfall. In other parts of the State water relations seem to be a pro- minent controlling factor. On the eastern side of the Mount Lofty Range Jessup (1L) has found ble gums occurring on red-brown earths with a rainiall as low as 13 inches per aunum. It should be noted that these soils would possess a relatively high water-retaining capacity. Boomsma (5), working on the Southern Flinders Ranges, also gives evidence which indicates that water relations play a large part in the species’ potential environment. He found blue guins along the watercourses where the annual rainfall was over 16 inches, on red-brown earths with am annual rainfall greater than 17 inches, and on podsols of lower water- retaining capacity with an anuual rainfall greater than 21 inches for sayanna! and grealer than 26 inches for selernphyll formations associated with it, 3. Eucalyptus viminalis (manna gun) This species occupies the moister, south-lacing aspect of the ridges with E. lewraxylon or E. comalidulensts aceupying the northern aspect between the 30- and 40-inch isohyets (sce above). In higher rainfall areas it lends to oceur along walercourses.. The density of the trees increases with rainfall front an oper woodland on the drier side to @ closed forest in its wetter limit. Althouyl this species occurs in small areas on truncated lateritic soils it ig largely confined Lo 111 the richer grey-brown podsols. This is probably due to higher water relation- ships within the grey-brown podsols, The distribution of FE, viminalis in relation to soils and rainfall is shown in fig, 10. 4. Eucalyptus camaldulensis (syn. E. rostrata) (river red gum) In the drier limit of the area this species is confined to the creek beds or to some distance on each side, but along the Rivers Torrens, Sturt and Onkaparinga, where the annual rainfall is greater than 27 inches, it extends over larger areas. The species is confined to grey-brown podsols on the slopes and ridges and alluvial soils in the valleys, both soils bemg rich in P,O, and nitrogen and having high water relations, Where the redgums extend over the ridges near the rivers they tend to occupy a similar habitat to that of E. leucoxylon on ridges along- side minor streams. The species does not extend into areas with rainfall greater than 35 inches per annum, Its soil and rainfall requirements are summarised in fig, 10. 5. Lucalyptus obliqua (stringybark) This species is a dwarf ecotype of the species growing in the eastern States, for it rarely grows higher than 80 feet. This stringybark occurs on a wide variety of soils, such as podsols, highly leached grey-brown podsols, lateritic soils, and ferrimorphic soils (all of which arc low in P,O, and nitrogen), provided the annual rainfall is greater than | Nee Ld OOK, oo] OX KO OT EET Efa sciculosa OE eee SS Oe Og Fig. 11 Diagratn illustrating the distribution of various species of Eucalyptus in relation to soils and rainfall, The numbers on the figures refer to soils named in Figure Hatching from right to left (downwards) indicates presence on South-facing aspects only. Hatching from left to right (downwards) indicates presence on North-facing aspects only. Cross hatching indicates presence on all aspects. 112 approximately 35 inches. lt extends up to the maximum rainfall for the area of 48 inches per annum on Mount Lofty Summit. Aspect plays a part in its dis- tribution towards its drier limit, for it tends to occupy the moister southern aspect. On the grey-brown podsols it tends to form either a pure commitunily or one mixed with Z. wiminalis on the southern aspect, while a pure community of KR, lexeoxvion occurs on the northern aspect., Similarly towards the drier limit of FE, obligna on the podsuls, E. obliqua tends to occur with &, Baxter on the southern aspect, while E, Barteri and FE. fasciculosa occur on the northern, The soil andl rainfall requirements of LE. obligua are illustrated in fig. 11. If we remember that a large perceritage of the rainfall is last to the soil by vun-off in the Adelaide Hills, the environmental ranges found by Crocker on the low undulating sand-dunes of the South-East (6), and by Raldwin and Crocker on the flat Jateritic peneplain of Kangaroo Island (2) are comparable to that shown above ior the Adelaide Hills. It should be noted here that on the flatter country of the other two localities the lower rainfall limit of the species may be as low as 24 inches per annum. 6. Eucalyptus Baxtert (brown stringybark) This species has.a restricted range of soil nutrients, occurring on only the very poor podsols, but with a wider moisture range than that of A. obliqgua {sce fig, 11}. This is especially evident where the padsols developed over the thick quartzite of the Adelaide Series occur in regions with rainfall as low as 30 inches per anntim such as on Black TIill, Marialta and Stonyfell Ridge. In these localities E. Baxteri tends to occur with E_ fasciculosa on the moister south- facing side of the ridge towards its drier limit, while E, fas¢icwlosa occupies the northern aspect alone. However, with increase in rainfall Z. Baxteri soon occupies: both aspects, while &. ebligua occurs with it at first on only the southern aspect but eventually at about the 35-inch isohyet on both aspects. As the rainfall increases still further H, Aaxtert tends to be more pteyalent on the skeletal quartzites than E. wbligua, and where the podsols are better developed there is a tendeney Jor E, Baxtvri to occur on only the drier aspects. Ilowever, although the environment appears to be suited for its growth, no specimen of #. Baxteri was found on the podsol on Acklands Jill. Taylor and O'Donnell (21) have noted &. Baxtert occurring on truncated laterite in the Hundred cf Kuitpo, and the authors have noted similar occurrences further down the Fleurieu Peninsula. Considering these occurrences, it at first seems rather surprising that B. Baxtert is not Found on the lateritic soils of the area, but on closer examination it is seen that these soils are mainly in regions with annual rainfall less than 35 inches, and this wetter limit is near the apparently anomalous area on Ackland’s Hill. Taking into consideration the effect of differences of topography on moisture relations within the soil between the Adelaide Hills, Kangaroo Island (lower rainfall limtt approximately 20 inches per annuni) and the South-East (lower limit about 22 inches per annum), the environmental range of H, Baxtert is similar in the three habitats, that is, the species is restricted to soils very low in P,O, and nitrogen such as podsols and laterites, and with a Jower range of moisture relations than E_ obligua, li, Eucalyptus fascteulosa (pink gum) This species occupies the dricst position in the “stringybark complex”. occurring in areas with rainfoll as low as 22 inches per annum and extending up to 40 inches per annum. In all the localities where this species has been fouttd the soil is acidic and Jow in P,O, and nitrogen, but the texture of the soils (all Ms phases of laterite podsols, highly leached grey-brown podsols and ferrimorphic soils), varies considerably. In its drier limit the species is confined to podsols and residual laterites with a deep sandy A horizon or deep Jateritic sands, the species maintaining its posi- tion only by the fact that these soils have a very low water-retaining capacity. When the A horizon becomes shallower, the water relations of the soil for the same climate improye, and EF. fuSciculosa gives way to E. leucoxrylon. As most of the country where tt occurs is gently undulating, aspect here plays little part. In the drier limit of its distribution on the podsol of Black Hill, Rocky Hill aid Stonyfell Ridge the species occurs pure on steep, skeletal, quartzite slapes ott all aspects. As the rainfall increases E. Barteri appears mixed with it on the southern aspect with pink gum pure on the northern aspect, and with further rainfall increases B. Baxteri occurs with it on both aspects. At about 35 inches per annum, as mentioned above, £. obliqua appears and the pink gum tends to he confined to the drier aspects, the size of the area it occupies becoming pro- gressively smaller until only a few scattered trees occasionally occur within the “stringybark coniplex” at about the 40-inch isohyet. The fact that the 40-inch isohyet is approximately the species’ upper limit explains. its absence on soils of low nutrient slatus between Stonyfcll Ridge and National Park, The soil and rainfall requirements of #, fascicylosa are illustrated in fig. 11. A\s this species occupics only small areas in the broad survey carried out by Crocker in the South-East (6) little comparison can he made. One author (Specht) however, has noted on the leached sands of the South-East, that as the rainfall decreases FE, fasciculosa tends ta become more prevalent and to supplant E. Baxteri. On the eastern side of the Mount Lofty Ranges Jessup (11) has shown that the species occurs over relatively rich red-hrown carths in a rainfall as low as 15 inches per annum. This seems anomalous compared with the environmental range found on the western side, 8. &. cosmophylla (scrub or cup gum) E, cosmophylla occurs in limited areas throughout the “stringybark com- plex” with rainfall between 35 inches and 45 inches per annum, principally on podsolic and lateritic soils and always on the sunny aspects of the ridges (see fig It). A pure community occurs on a sieep north-facing quartzite slope in Waterfall Gully and in the Brownhill Creek watershed. The species tends to be limited in extent because of competition with other taller Eucalypts which obstruct the sunlight. Thus £. cosmophylla tends to develop on very skeletal soils where the scarcity of soil restricts the number of other Eucalyptus species developing. The fact that the direct effect of insolation is a marked controlling factor is supported by the occurrence uf scrub gum on only the sunny aspects, to matter what the rainfall. Taking into account the difference in topography this environmental range corresponds to that found on Kangaroo Island by Baldwin and Crocker (2). On Kangaroo Island the direct effect of isolation does not apply, for the area is a lateritic peneplain, 9. Eucalyptus rubida (candle-bark or whife gum) This species occupies a considerable area around gully heads on highly leached grey-brown podsols with a rainfall greater than 45 inches per annum. Xs the annual ramfall decreases ta 35 inches the species is confined to a narrow belt of wet alluvial soi along the larger creeks, Its distribution in relation tu soil and rainfall is illustrated in fig. 10. 1l4 10, Eucalypius elaeophora (bastard box) As this species only makes its appearance in the extretne north of the area no attempt has been made to determine its etvironmental range, This will he dealt with in a later paper. In this area the bastard box occurs on podsols and ferrimorphic soils of low nutrietit statis, 11, Casuarina stricta (drooping sheoak) This species occurs scattered amongst FE. edorata and E. leucoxylon on various soils from the coast to about the 35-inch isohyet. On very steep cliffs such as occur in the Torrens Gorge, Morialta. Slape’s Gully, near the first water- fall in Waterfall Gully, the Sturt and Onkaparinga gorges, pure communities may be developed on either aspect, The sheoal occurs with Banksia near Noar- lunga and Hackham on deep siliceous sands of neutral reaction and of low water- retaining capacity with a rainfall 20 to 22 inches. From this evidence it appears that Caswerina stricta can exist in much drier habitats than E. odorata, which oftett occurs in contiguous habitats which are less steep or, as in the case of the deep sands, of higher water-retaining capacity. This evidence is supported by the occurrence of Casuarina stricta on shallow gneissic soils along the dry Palmer scatp on the east of the Mount Lofty Ranges, with a rainfall of 20 inches per annum. This stand extends. to the south, where it occurs on very shallow red- brown earths (with terra rossa affinities) around Lake Alexandrina (Jessup 11). Here the rainfall is about 15 inches per annum. On Eyre Peninsula Crocker (7) has found the species occurring over calcimorphic soils and skeletal gneissic soils from 14 to 22 inches per annum. Crocker (6) has also found Casuarina stricta associated with sandy terra rossas in the South-East, where the raimfall is as high as 28 inches per annum. These observations suggest that Casuarina stricta can exist over a wile tange af soils in respect of reaction, nutrient status and texture, but appears to be limited to those conditions in which the compensating moisture relations are low- 12. Hybrids Hybrids between different species of Eucalyptus have been found especially at the junction of the distribution of two species. These hybrids possess characters intermediate between the two parent species, makitig identification of some trees difficult. Such a case occurs between E, wminalis and Li, caimaldu- lensis giving rise to trees with anomalous E. ziminalis fruits, and between E, vimtnalis and E. rybida, thus forming stands of trees which gradually grade from one species into another, Hybrids also occur between E, viminalis and E, leucoxylon, E. obliqua and £, Baxteri, E, odoratw and &. lewcoxylon, and E. ramaldulensis and E. lexcoxylon. 13. Formations As is the case of the species of the first stratum, the undershrubs exist over a definite environmental range, some heing ecological wides, others having a limited range. In general there are two dehnite formations—the savannah woodland formation being developed over soils rich in P,O, and nitrogen, sucls as red-brown earths, calcimorphic soils and the slightly leached grey-brown podsols, while a sclerophyllous formation occurs on more heavily leached grey- brown podsols, podsols, laterttes and ferrimorphic soils which are lower in PO, and nitrogen. On the grey-brown podsols a transition between savannah and sclerophyll occurs. This is probably connected with increased leaching with increased rainfall. However, an abrupt change occurs at the junction between richer soils and podsols. Within the formations, variation within the under- 15 t Me ST WNCENT GULF FORMATIONS ORY SCLEROPrY LL i ] é r Espey BANKSIA ORNATA or DOMINANT 4 ied BANKSIA HARGNATA DOwINANT | SAVANNSH “urs o | 2 res Map showing the distribution of the formations. shrubs occuts with every microhabitat, some species being dominant m one locality, others in another, but throughout the formation the same composition usually occurs. Conospermum patens and Hypolaena fastigiata are restricted to the residual laterite on the Eden~ Moana fault block, while Banka ornata appears on the deep lateritic sands on the Ochre Cover fault block. In all cases the formations are developed independently of the dominant trees. 16 Some characteristic species (¢.g., Hitbertia acicularis, Olearia ramulosa, etc.) are ecological wides and occur throughout the savannah and sclerophyll, but extreme splitting would come into ecological classification if these species were taken into account, Further investigation, however, is wartranted into the aulecology of individual species of the lower strata. The distribution of the formations is indicated in fig. 12 and in pl, VI, B, CUASSIFICATION GF THE PLANT ComMMUNITIES In classifying plant communities the authors have adhered as closely is possible to the scheme proposed by Waod (9). An association is defined as any ttatucal group of specics occurring in similar habitats, When there is a modifica- tion of only the dotninant tree stratum of an association by a small change in environment, these stands are called vegetation types of the association, i, however, the associated undershrubs change with the microhabilat over a smail area while the tree stratum of the association remains the samc, these stands are termed societies of the association, The tem community is used for any assemblage of plants, All the communities appear to be climax communities, the only communities showing seral relationships occurring in the swamps and sandhills or as stages in the regeneration of forests after clearing and burning, The vegetation has been classified into nine associations, the relationships and environments of which are summarised in ‘able I. 1. Eucalyptus odorata association (see pl, VILL, fig, 3) Within the rainfall limit of 30 inches per annum a pure association of £. odorata savannah woodland occurs over red-brown earths, calcimerphic soils, aud grey-brown podsals. Over all this area E, odorata grows to a small tree less than 30 feet in height or may assume a “mallee” habit. Associated with it and sometimes codominant are Melaleucu pubescens (on the calcimorphic soils) and Casuarina stricta, both of which are small trees. Acacia pycnantha also forms a distinct second stratum with grasses such as Danthonia spp., Stipa spp. and Themeda australis, usually constituting the ground stratum, However, the savannah woodland has been invaded by a large number of introduced plants, which in many cases alter the facies of the association, ¢,y., olive (Olea europaea), hawthorn (Crataegus monogyna) or dog rose (Resa cantina), but usually the facies is little changed, for even though the invaders are numerous they are small, The land on which this association occurs is some of the best in the ares, Much of it has been cleared and is now being used for cereal growing, vineyards and grazing. A. floristic list is given in Appendix TI. 2. Eucalyptus leucoxylon— Eucalyptus viminalis association This association lies between the 30 and 35-inch isuhyects on grey-brown podsols and truncated faterites, LZ. leucoxylon occupying the drier northern aspeet af the ridges between the 30- and 35-inch isohyets, with £. wimtnalis on the moister southern aspects. The undershrubs are essentially the same as in the £, oderuta association, with Lfibbertie acicularis and Oleavia raniudesa more prevalent. The undershrubs show no variation at all with aspect. Jence as these two Eucalyptus species tend to occur mixed on the tops of the ridges and further south, it is best to look upon these pure stands with similar undershrubs, as E. leucoxylon and E, viminalis woodland types differentiated by mierohabitats within the E, leucorylon—E, viminalis association, Between the 25- ane 30-inch isohyels £. teucoxylon occupies the moister southert! aspect either alone ar mixed with Z. odvrate. E. odorata then oecurs i? on the drier northern aspect. These pure £. levcoxylan savannah woodland stands may be considered as an Fi. lewcoxylom type of the above association, for it occupies a similar microhabilat. However, although the mixed savannah stand of E, leucoxrylon and EF, odorata oceupies a similar place to the E, leucexylon type, it must he considered as an ecatone between the two associations. 3. Eucalyptus camalduiensis association (see pl. VIII, fig. 1) This association is well defined. As the distribution and the environmental! range of E, camaldulensis has been discussed above, there is no need for recapitu- lation. Along the creek beds Cyperus vaginatus, together with Juncus pallidus and J. polyanthemas, are the dominant plants. Occasionally Cylisus canariensis and Rubus fruticosus have invaded the association in the wetter regions, How- ever, away from the creeks the vegetation is of the same savannah type as decurs within the &. /eucosylon—E, vimirtalis association. In the watercourses amongst the lateritic sands Leptospernuim pubescens, Melaleuca decussata, Callistemon salignus, Acacia rhetinoides, and 41, verniciflua occur as well as the species men- tioned above. 4. Casuarina stricta association (see pl. 1X, hg. 4) This association occurs in limited areas of extremely low water relationships, such as steep rocky slopes and the deep siliecous sands of neutral reaction near Noarlunga and Hackham, where the rainfall is just over 20 inches per annum. On these deep siliceous sands Banksia inarginata and, in some places, Callitris propingua, are codominant. Here the association is definitely selerophyllous, the most importunt plants being Xanthorrhera seimiplana and Adenanthos terminalis. Other common shrubs are Jodonaea viscosa, Bursaria spinosa, Olearta ranvilosa, Exacorpus cupressifornus, Calythrix tetragona, Acacia pycnantha, A, armata, Bunksia ornata, Hibbertia stricta, Grevilledt lavandulacea, and Kennedya prostrata. Aseclepias rotundifolia and Solanum sodomecum have invaded the area, Scirpus nodosus and Lepidosperme carphoides ave fairly common, while Muchlenbeckia adpressa, Carpobrotus aequilaterale, and Adenanthos terminalis oceupy large areas of ground. Of the herbaceous plants Oenothera odorata and Salvia verbenaca are common, while the grasses Euncaupegon nigricans, Themeda australe and Brisa maxima occur sparingly. Loranthus Afiquelii sometimes infests Casuarina stricta. On steep slopes the association approaches a sclerophyllous community, Xanthorrhoea quadrangulata in particular being dominant. E.wxocerpus cupressi- formis, Bursaria spinosa, Dodonaca viscosa, Olearia ramulosa, Bdnksia marginata, Asclepias rotundifolia and Osteospermum moniliferum are often prevalent, together with ferns such as Cheilanthes tenttfoha and Pleurosorus rytifolius and the grasses Themedw australis, Stipa setacea and Danthonia semiannularts. Loranthus e¢xvcarpi occurs occasionally on Casuprina siricte and K-racarpus cupressifornis. From the nature of the Casuarima stricla association jt seems that it could be classed ag an extremely dry association of the “stringybark edaphie complex.” In some parts, however, it has closer affinities to a savannah woodland formation, while aceording to Wood (24) a savannah woodland formation of Casuarina stricta oceurs on rendzinas near Port Noarlunga, 5. “Stringybark edaphic complea” (see pl. Vill, fig. 2, 3,4, and 1X, fig, 1, 2, 3). On soils low in nutrient status, provided the water relalions do nat permit the development of &. edorata or EF. leucoxylon, there necur six species of Eucalyptus (namely E, fovcieulosa, E. Baxteri, EH, oblique, EB. cosmophylla, 118 £, rubida and &. elacophora) whose environmental ranges coincide with the habitats found within this section. All of these six species are extremely sen- sitive to small changes in environment. As shown above, the distribution of all is controlled by the moisture relationships within the soil and the nature and/or the PO, and nitrogen level of the soil, while EZ. cosmophylla is probably con- trolled by insolation. In dissected country, especially with a complex pedology such as occurs in the Adelaide Tiuls, a large number of micro-habitats occur every few chains. Each microhabitat is favourable for the development of a certain combination of Excalyptus species and undershrubs whose atitecology coincides with the con- ditions at that spot. Thus we have a large number of combinations of species produced, cach occupying only a limited area (which is often quite small but may be up to a couple of square miles in extent), Hence, within the area, it is possible to distinguish the combinations of Eucalyptus spectes which occur in similar habitats which are set out in Table IL. Taece IT QN PODSOLS AND LATERITES ON GREY-BROWN PODSOLS E, obliqua &, obligua E. obliqua—E. Baxteri E. obliquaé —L. fasciculosa EL. obliqua~E. Baxteri—E, fasciculosa Li, obliqua— E, rubida E. obliqua— FE, Baxteri—£. casmophylla E, fasciculosa — EB, cosmophylla FE. obliqua-£. Baxteri—E, cosmophylla- —E. fasciculosa &. fasciculosa £, rubida F, oblique — E. fasciculasa BE, obliqua-—E, fasciculosa— FE. cosmo- phylia BK, Baxtert—F., fasciculosa ON FERRIMORPHIC SOILS E. Baxteri—k, fasciculosa-E, cosmo- E, elaeophora phylla FE. elaeophora—F, fasciculosa E, Baxtert— FE, cosmophylla #. ¢laeophora~ E, obligua E, fasciculosa E, obligua E. cosmophylta In general the undershrubs constitute a sclerophyllous community of remarkably constant composition no matter what the combination of dominant trees, although the frequency of the individual species varies with the micro- habitats. However, these changes in the frequency of the species may occur within a single combination of Eucalyptus species as well as between combina- tions, thus giving little significance to the use of the tudershrubs in an attempt to classify the communities under discussion. A glance at the map of the distribution of the six Eucalyptus species in the area under discussion reveals the extreme difficulty of formulating a satisfactory system of classification of the combinations of these species. Considering the variability of the micro-habitats over a small area, it seems jrstifiable to consider areas containing these species where a more uniform habitat exists. Such a case is found on the more level topography of the lateritic peneplain of Kangaroo Island and the sand-dune ranges af the South-East. On Kangaroo Islanil E. Baxteri- E. cosmophylla sclerophyll association is present over a wide area with E. obhiqua, BE. Baxtert and £. cosmophylla occurring mixed in the wetter regions (2), In the South-Rast, E. Baxter sclerophyll association occurs over wide areas of a uniform edaphic condition with B. abligua occurring on the better soils with a rainfall greater tham 24 inches per annuin (6), We therefore see 119 that under a more uniform habitat the same species (both trees and undershrubs) will be present in the same combination and frequency over a wide area and can he classified as an association. On close examination of the Adelaide Hills the E&. obliqua-—E. Baxfert association of the South-East and the E. Baxter? - E. cosmophylla association of Kangaroo Island are apparent over small areas where an analogous environment is present, As K. fasctculosa occurs pure in large areas toward the drier limit of the complex it seems logical to class this as an association, H, obliqua also tends to occur pure over the highly leached grey-brown podsols. Definite areas of E. rubida occur along the watercourses in a rainfall of 35 to 45 inches per annum, and in larger areas on gréy-brown podsols in the wetter regions. Hence we could separate an E. ebligua association and an EF, rubide association, but for further evidence to support these last two associations we stall have to look to the eastern States. It is seen that within the “stringybark edaphic complex” there are possibly at least five associations, namely ;— E, obfigua association E, fasciculosa association £, obliqua—E. Baxteri association E. rubida association &, Baxtert Ff. cosmophylla association As only the southern limit of E. elaecophora occurs in the area, no conclusiott can be reached as yet as to its status in the “stringybark complex”, The other combinations may be regarded as forest types or ecotones of the above associations. The dominant tndetshrubs of this dry sclerophyllous edaphic complex vary from place to place independently of the iree species. Usually about six species (Acacia myrtifolia, Pultenaea daphnoides, Leptospermum scoparinm, Epacris impressa, Hakea rostrata, Nanthorrhoea semiplana) appear to be codominant, but local yariations in percentage [frequency occur. On the grey-brown podsols Daviesia corymbosa tends to become a codominant undershrub as well. On the deeper residual lateritic podsols near Blewitt’s Springs with a rainfall of 25 to 30 inches per annum Banksia ornate becomes a codominant undershrub, while under swampy conditions Banksia ornata is replaced by Callislemon spp. Often Ploridivm aquilinum and grasses have become dominant on the highly leached grey-brown podsols and Jaterites after clearing. A floristic list is given in Appetidix IT, Cytisus canariensis, Ulex europaeus and Rubus fruticosus have invaded the complex in places, especially along the roads. The moist shady valleys of Waterfall Gully tend to develop a peaty swamp vegetation with dense stands of Leptospermum pubescens, Gahnia trifida and Goedenia ovata, while on the edges occasional Utricularia dichatoma, Sprengelia fncarnata and the fern Schisaea fistufosa occur, It is within these swamps that specimens of Todea barbara occur. The swamps do not persist far down the creek but give way to moist regions which support dense mats of ferns such as Gleichenta circinate, Blechnum capense, B, disevlor, Adiantum acthiopicuim and Pteridium aquilinym, together with an occasional specimen of -4splenium flabellt- foliuns and Pléuresurus rutifolius. Similar swampy creeks are found near Longwood, In the creeks near Heathfield large quantities of .dporogetom distachyus (Cape pond weed) and Lemna minor can be found in spring. 6. Ecotencs (1) Evotones, in the sense of transition zones rather than tension belts, occur between all the associations, but especially along the junction of the “Stringyhark 13) edaphic complex” and other associations, as here we have seven Eucalypt species with overlapping environmental ranges. (2) E. odorata-E, leucoxylon—E. fasciculosa ecotone, LE. leucoxylon and F. odorata have overlapping environmental ranges between the 25” and 30” isohyets on both the grey-brown podsols arid Jaterites of the Eden-Moana Fault Block. Towards the drier hmit £. Jencarylon occupies the shady moister aspect of the ridges, whereas towards the wetter it spreads. out over both aspects. FE, odorata ig here unaffected by aspect and occurs mixed with E&, leucoxr on up to the 30” tsohyet. Such stands oceur on the Eden-Moana Fault Block from Happy Valley Reservoir to National Park, As shown above. the nutrient status of grey-brown podsols and laterites is different, the laterites being low whereas the grey-brawn podsols are relatively high, Consequently, although the Eucalyptus species are not affected, the undershrubs. are respectively those associated with selerophyllous and savannah communities. On the shallow lateritic residual podsols of the Ochre Cove Fault Block an E. leucoxylon scletophyllous com- imunily containing a. few isolated trees of E. fasciculasa merges into the £. fasci- culosa association of the “stringybark-edaphic complex” on the deeper lateritic residual podsols, This association contains an occasiotial tree of E. fewcoxylon, Throughout the areas of blue gum and peppermint gitm mentioned above, small areas of an E, wiminalis sclerophyllous community occur, The difficulty arises in classifying such stands, Ag the E, fencoxylon- £, odorata selerophyllous cormmunity has a constant structure over a considerable atea, one could look upon this as an association, with the EZ. leucoayion sclero- phyllous stand, such as occurs on the Ochre Cove Fault Block and at Tlappy Valley Reservoir, as a sclerophyllous woodland type within the association. However, #6 pointed out above, there is a marked change in the undershrubs independently ef the associated trees, so that E, lencorwlon—E. odorata savannah and FE. leucoxylon—F, odorata sclerophyllous woodlands occur contiguously over considerable areas. This change in the undershrubs is too vast to allow rhe stands to be classified as societies within the same association, The presence of patches of &. viminalis throughout, and occasional trees of L, fasciculosa in the stands on thé Ochre Cove Fault Block, provides further evidence that the stands do not constitute an association, It therefore seems better to regard the stands as an ecotone between the three associations, occurring at its limies, fe, the &, odvrate savannah woodland association, the £, lexcoaylon=E. wimenaliz savannah woodland association, and the E&. fesecic¢ulosa sclerophyll woodland jissoctation, The species composition of the savannah stands ts similar to that of the #, odorafa association. In the selerophyllous stands several undershruhs are co- dominant. On the Eden-Moana Fault Block Banksia marginale, Adenanthos terminalis and Kungea pomifera ace prevalent in the south on the deeper residual laterite whereas Xanthorrioca semiplana, Hokea rostrata and H. uftcina are prevalent in the north on truncated laterite. A list of the species present in the sclerophyll stand is given in the appendix. The “whipstick mallee” form of &, edorafa on the podsol at the foot of Black Hill could possibly be classed within this ecotone. 7. Méscellancous communes which oceupy only small areas (1) Coastal Sandhill communities At Port Noarlunga and Christie's Beach there are small areas uf sandhills on which distinct communities develop. A, At Port Noarlunga the dominant plants are three shrubs, Acacia lengi- folio, Myoporum insulare and Oleoria axillarts, which all grow to a height of 12) 4 to 6 feet. The most important associated plants are Calocephalus Brownii, Leucopogon parviflorus, Arctotis stoechadifolia, Muehlenbeckia udpressa, Cakile maritima, Carpobrotus aequilaterale, Rhagodia baccaia, Suacda australis, S pinifex hitsutus and Scirpus nodosus. Others which occtir much less frequently are Dianella revoluta, Scaevola crassifolia, Kennedya prostrata, Pimelea serpylltfolia, Lotus australis, Lugula campestris and the grass Layuris ovatus. It should be noted that a large proportion of these plants are prostrate creeping plants. All the plants are stu2ted and dwarfed owing to the poor and saline nature of the soil and the exposed position. B. At Christie's Beach there is a smaller area of dunes which are not as well stabilised as those at Port Noarlunga. The flota is not as well developed, the dominant and sometimes only plant being Ammophila arenaria, Other plants are infrequent, but the following do occur :—Acacia longifolia, Olearia axillaris, Dianella revoluta and Euphorbia terracina, (2) Coastal cliff communities With the exception of the two small areas of sandhills mentioned above the coastline is generally marked by cliffs, the vegetation of which, besides being extremely variable, 1s scattered and stunted owing to the exposed conditions and shallow soils. The chief shrubs present are Alyxia buxifolia, Myoporum insulare, Olearia ramulosa, Beyeria Leschenaultii, Nitraria Schoberi, Acacia Ligulata, A. longifoha, Eutaxia microphylla, Atriblex paludosum, Lixchylaena villosa, Rhagodia baccata, Pomaderris racemosa, Melaleuca decussata, and Wesiringia rigida. Other plants which commonly occur are Carpobrotus aequilaterale, Scaevola micracarpa, Myoporum parvifolium, Zygophyllum Billardieri, Kennedya rostrata, Dampiera rosmarinifolia, Dianella revoluta, Cakile maritima, Oenothera odorata, Asphodelus fistulosus and Pelleia parado-ra. (3) Coastal Swamps Near the mouth of the Onkaparinga river thete is a small area of saline swamp, some of which is subject to flooding at. high tides. This small area contains the best example of succession found during the survey, The sere can be illustrated as follows }— Arthrocnemum sp. Arthrocnemum. sp. - Salicornia sp. Salicornia sp,— Atriplex paludosum, | ‘ V Vo Alriplex paludasum — Nilraria Schober: The direction of the arrows indicates deeper watcr table, decreasing salinity and less frequent flooding, A. In the lowest areas which are subject to flooding at high tides Arthrocnemum sp. is the first plant to become established, The spreading habit of this plant leads to a gradual raising of the ground level due to collection of alluvial material, 122 B. On the slightly higher areas not subject to such frequent flooding Salicornia sp, becomes cadominant with Arthrocnemum, Associated with these two plants is Samolus repens, C. Where flooding is rare and only oceurs at exceptionally high tides Arthrocnemum. is not so important although still present. The dominants in this part are Salicornia sp, and Aériplex paludoswm. Associated with them are Somolus repens, Plagianthus microphyllus, Kochia oppositifolia, Frankenia pauci- flora, and Carpobrotus aequilaterale. D, The highest parts of the swamp are only Aooded by the infrequent fresh- water floods caused by overflowing of the Onkaparinga River. A more complete vegetative cover is present. The dominants are Nitraria Schoberi and Atriplex paludosum, and associated with them are Salicornia sp., Arikrocnemum sp., Enchylaena villosa, Suaeda australis, Rhagodia baccata, Plagionthus mucrophyllus, Wilsonia kuimilis, Carpobrotus acquilaterale, Frankenia pauciflora and Distichlis spicata. Juncus maritimus is common in some parts. SUMMARY This paper deals with the ecology of portion of the Mount Lofty Ranges between the River Torrens atid Noarlunga. Working on the premise that every plant has a certain potential environ- ment, the climate, soils and vegetation have been studied and attempts made to correlate the distribution of the dominant trees and association with the environ- ment. Both mechanical and chemical analyses have been made on soil samples typical of each of the soil groups present in the area, and suggestions are given as to the genesis of each soil. To add breadth to any conclusions drawn from the data collected, the environmental range of the dominant species is discussed with reference to their environment in other parts of South Australia which have been studied by other ecologists. ACKNOWLEDGMENTS It would be impossible here to satisfactorily acknowledge the great assistance which has been so willingly given in this research by a large number of people. We can sincerely say that their help has been greatly appreciated, In particular we would like to express our appreciation of the help and advice which has been s0 readily given by Professor J. G. Wood. REFERENCES (1) Avamson, R, S., and Oszorn, T,G. B. 1924 The Ecology of the Eucalyp- tus Forests of the Mount Lofty Ranges (S. Aust.), Trans, Roy Soc, 5. Aust. 48 (2) Barpwrn, J. G., and Crocker, R, L, 1941 Soils and Vegetation of por- tion of Kangaroo Island. Trans. Roy. Soc. S, Aust., 65, (1) (3) Buacx, J. M. 1921 Flora of 5, Aust. Govt. Printer, Adelaide (4) Buaxety, W. F. 1934 A Key to the Eucalypts. Sydney (5) Boomsma, C, D. 1946 The Vegetation of the Southern Flinders Ranges (S, Aust.) Trans. Roy. Soc. S. Aust., 70, (2) 5 Spr ; EREY Trans. Roy. Soc. S, Aust., Aucust, 1948 BRET, AD Ems “4 — la y aces Aropnens CHM) Xv, epee =<: ~) Sah \ ASS E> Y — y;) (3 - ss O $o8 Joa ‘va “ale S Ns ae Qe OUT CEN \e a TON x => ——"' r a _ > @ ADELAIDE é : oy \\ Ke = still U XY Wr en s " aes ra oO a a - a 006 Pn \ A AWW yi ! a Whe a \\ ll aN IN NN Sek SORA SN WY f 060: Ze0: = OFO- #50: Z50+ — 810 120: S10+ — sfo- — 9800: — +b00> — 9900- — — 9¢£00- Ss 60 0 1) 0 0 0 tr 0 tz 0 0 0-ss Lg Ep op 0°39 OS O-r Og £9 coo es BbT £86 962 od e-¢ 6L oes 09 g-9 0-6 TS S°6 (+9 £6 96T OS B82 BS bz O-TL OST 0-81 Il 8-Z0 Z:2I SEZ POL Set | OTE O25 OZL B29 | SST Oe Or9¢ £82 OZ Bee 2:99 cto) 6 9-6Z 0 GS % % % % % % % % % % %, % % ZI +81 8-0 0Z-Z1 ZI-9 0 9-01 OTN 8FZI eI +ro be-0e «= NZ- Vv cv IV a Vv V a Vv q Vv ) a Vv O8ZR59 OLR069 ceopes S6LZOTZ trZ0TZ 002099 ¥PuolrsC SPRY qunung u2]) Bu0T BRUNO N | AYOT IW AYOT “WL WH s.pueppy sTosdog ‘AMOJSIUMT IRIAN | “SYPIES MMOI-pat OF PITY «< OS G&S tS ofS ges ZS ass OL Set | O8 PB OB - Reo» = S40" — +20 t50- £90+ 61+ 980- 892+ | £20- "S80. F1l- . 610- 1z0: “5 — 120: — 20: zs0- £90: — — gr0- - 0 09 0 0 82 0 ge 0 0 0 0 0 - S29 ZT free 1-09) BLL Z9S = Se Leh OT G9 029 = Ort - I-21 = Z-0€ Z6z SLe S12 9-ZT OZ f-1Z £92 0-8 o8 eral - €-8l Zh Ze 9:02 Sehr ZrpZ —-FrBE O-TE €-OF €-Il $91 g-G¢ - TZ £8 zs SI E-bl 0-2 I+TL O-F 8-91 £51 O-fl 0-82 - % % % % % % % % % % % % Sr-ZI zI-0 9E-8Z 82-SI SI“ or =o o1-8 8-0 LZe-81 SI-0l «OI d Vv i) d V a Vv a Vv a a Vv OI8ZIZ 0£2069 062029 £24099 9g9R2s | AOALISO UMOJISWILINS yinig. isddp, | yar [[tquAorg Aayea Addvyy STIOG dallTy GNy STOSa0q NMONS-ATe) Hd uonarey = uadoity “ag Od [RABI Ary ws pues aug pues osiz05 ' 4 ~ = t t (saysm) wWydaq WOzIO TT aIUalopes pry Ayipeo oT t ' i Had wonseayy uasoiin “O’d Jaaessy ARID - aps pues aut pues as1%07) (seysut) yadacy WOZIIO FT aoadayad pissy AQYyeIo-T ‘sdeur a.ueupio Areqyur apreppy pue esunyIay ay) wo saryeso, ay} aad soduasayas pus ayy, "EDIE BY] UTYAL sjros jo yeatddy sayyord pros yo sasAyeur yeourayd pur peaueyr.eut jo synsay I XIGQNYddV 125 OE gS eg | Os FS 1-9 £:9 £5 9S 7 820° ocd: $90+ i — 290: 68F- 820: 280: * >" 9200- — — 210: == —s [Z0° 7 0 £8 < $-Z oT OZ SSS 009) SFO O-9T zcL = (0°6 - O-T OT 0-1 2:82 = GS-0 SG BZ etl LET * fel O21 g*8 257) 9-21 Orbt = G+ 9E 8-01 9-EP ‘ ee8 8-8 SRB 9-0 Lara ST 0-61 8-1 £ Ee . % % % % % % % % % Se-+1 FI -9 o-0 +f Le-81 B8I-Z «ZO S2-6 60 cw cv Tv fd ed 1a cv a cv 6220S 029059 089029 sfunds sjqunag suapser) Ai191] % +3) 2) suapiey A1I9yD = Ha WOT}IBIY RaZouIN - “ova - JaAein - AeyQ WS pues sul pues asivo7 (sayaut) y}doqy uodzIIOFy aYalayad piiry AVRIO'T SANVG OILMALW] daaq aANv SToOsd0g SILIWaLVT daLivonoal “ees z-9 £9 $9 90« = £0r 9f0- a — 800: — 0 0 0 It $69s o-F or 0-26 g-CL S+4 cL 0-01 $-EL O'zZe Cpe Sel SPL O-Ts oes c-02 % % % % +12 Iz-9f = 91-0 0Z-ZI a eV IV a PRI6LS Aayed Addepy -9 0-9 6:5 0f0- 1z0- 980° _ 800° — £9 vas 0 S-Z£1 SIT 0°38 G9 0-9 0-$ 0-61 0-61 $-9T 0-2¢ g*£9 $04 % % % - - H : : : x . Callitris propinqua- - - - - - M x x : C. tasmanica - - - - - - M : . . * x Themeda australis - - - - - Ch x x x x x Neurachne alopecuroides - - = - FH x . 5 x *Ehrharta longifolia - - - - - “ . ‘ x " . Stipa variabilis - 7 = = = x x x x x S. setacea - - - - - - -~ #H x x x A *Oryzopsis miliacea = - - - - « x x , 4 *Aira caryophyliea - - - - - # x x *Avena fatua - - - - - - HA x x ‘ *Holcus lanatus - - - - - » HA . 4 x Danthonia spp. - -~ - - - - #H x x x x Enneapogon nigricans - - - - #H : 4 x . Briza maxima - - - - - -~ FA x x - x B. minor - - - “ ~ - — ott x x x . : Distichlis spicata - - - - ~ H x . . . . *Vuipia myuros - - - - - - FA x x + 4 ‘ *Bromus spp. - ~ - = = - #H x x a Fs : *Cynodon dactylon - - - - - HA x x ° ‘ . *Lolium spp. - - - - - - #H x x Fs 4 4 *Hordeum murinum =~ - > - - H x x - < Cyperus vaginatus = - - - - - H : : . x Scirpus nodosus - - - - - - G&G ’ . . x . Lepidosperma laterale - - - Ch : ’ x = x L, viscidum - - - - - - Ch ‘ . : . x L. semiteres = = - - - Ch , “4 x ‘ x L. carphoides = - = - = 2 £” 6h . i x x x Chorizandra enodes - -~ + = Ch : : : . x Hypolaena fastigiata - ~ - 7 H ‘ ‘ x a Juncus pallidus - - ms < i ~ #H x x x x x Dianella revoluta - - - - G x = x x x Reya umbellata - - . - - ~ G¢ x x x . x Anguillaria diotce - - - - A 1G: x x : x Lomandra dura - - - - - - Ch x x ' x Thysanotus Patersonii T. dichotomus = - - Caesia vittata - = Bulbine bulbosa - Dichopogon fimbriatus Bartlingia sessiliflora - Xanthorrhoea quadrangulata X. semiplana - 7 *Asphodelus fistulosus Hypoxis glabella ~ Calostemma purpureum *Sparaxis tricolor - *S. bulbifera - - *Romulea rosea - - *Homeria collina - * *H, miniata - - - Patersonia longiscapa - Dipodium punctatum Thelymitra aristata - T. antennifera - ~ Acianthus exsertus - Lyperanthus nigricans Eriochilus cucullatus - Caladenia dilatata - C. deformis = + Glossodia major - = Diuris maculata = Pterostylis nana - Casuarina stricta - C. striata - - “ C. Muelleriana - - Isopogon ceratophyllus Adenanthos terminalis Conospermum patens Persoonia juniperinum Hakea rostrata - ~ H. rugosa - - - H. ulicina - - - Banksia marginata - B. ornata - - - Grevillea favandulacea Exocarpus cupressiformis Loranthus Miquelii - L. exocarpi - - *Rumex crispus - - *R. acetosella - - 4 128 Life Form = agnaaAmAgaAAmMAMAAgRAAAANAaAEmAAAAA x HititiezzzeezeuzzZzee E odorata association (1) 4H 4 4 # E. lettcaxylon- E. viminalis assoc. x x eR KR KH x A . H A A OK Sclerophyll of E. odorata-E. leu- coxylon-E, fasci- culosa ecotone ~ A AR HA A Aw AK KR * A x Hw Casuarina stricta association edaphic complex Stringybark (3) YAH AWA RK RK WH A ww AA RA 4 A RR AA A A Hw HX Muehlenbeckia adpressa Carpobrotus aequilaterale Ranunculus lappaceus Cassytha glabella Drosera binata D. glanduligera D. Whittakeri D. Planchonii D. auriculata D. peltata - Crassula Siebariana Bursaria spinosa Marianthus bignoniaceus Cheiranthera lincaris Billatdiera cymosa *Rubus. fruticosus *Rosa canina *Crataegus monogyna Acaena ovina A. sanguisorbae Acacia continua A. armata - A. obliqua =~ A. spinescens A. verniciflua A, rhetinodes A. myrtifolia A. pycnantha A. rupicola - A. vomeriformis A. melanoxylon Daviesia corymbosa D. ulicina = - D.. brevifolia Eutaxia microphylla Pultenaeca daphnoides P. pedunculata P. largiflorens P. involucrata P. acerosa var. acicularis Dillwynia hispida TD. floribunda Platylobium obtusangulum *Ulex europaeus *Cytisus canariensis “Trifolium procumbens 129 Life Form AAAZZAAAZAAAAAZAAASSRS SSS ee Se MIAssseeeynaagynawyws E odorata association ~ (ay ~ ye 6 ' wx KOR 4 KA KR KR KH A A E. leucoxylon- E. viminalis assoc. Ho RR Sclerophyll of E, odorata-E, leu- coxylon-E., fasei- culosa ecotone Casuarina stricta association edaphic complex Stringybark =~ wn ~ AA KK KR KH RH HM HK KRW a HH KR A KR KH A AH RH OR MK KR KAW KK AK OW OR KY ‘ *T. subterraneum - *T. arvense - - *T, angustifolium - Psoralea patens - Swainsona oroboides var- Kennedya prostrata - Hardenbergia monophylla Glycine clandestina - *Erodium botrys - € *E. moschatum = - - *Oxalis corniculata - *O. cernua - - - *Linum marginale - ¥*L. gallicum - - Boronia caerulescens - Correa rubra - - Tetratheca pilosa ~ Adriana Klotzschii = - Stackhousia monogyna Dodonaea viscosa - Spyridium parvifohium S. spathulatum - - S. vexilliferum = - “ Cryptandra tomentosa *Malva parviflora - Hibbertia sericea - H. stricta - - - H. acicularis - - H. virgata - - *Hypericum perforatum Hybanthus floribundus Pimelia spathulata P. flava - - - FP. octophylla - - Lythrum Hyssopifolia Leptospermum scoparium L. pubescens - = L. myrsinoides - - Kunzea pomifera - Callistemon salignus - Melaleuca decussata ~- M. pubescens - - Eucalyptus obliqua - FE, Baxteri - - - E. odorata - - E. cosmophyla = - = hirsuta 130 Life Fovm BZRQZ22m2z222n4e42an444e4e2o4zz2zrrToasnrazroinen E odorata association E. leucoxylon- E. viminalis assoc Sclerophyll of E, odorata-E, leu- coxylon-E. fasci- culosa ecotone 3 K 1 a BE 22 PR ee 29 2c on nS Ow ane (4) (5) a : , : . ‘ x =x x ‘ x 5 ; : ¢ - 4 x x . x - x x é . x . x * x . =x « x x . x . x x . x *. x x * x . x x - x - x x . x . - =x . x > = yiminalis rubida = - - elacophora = - leucoxylon = - . fasciculosa - Calythrix tetragona *Oenothera odorata Halorrhagis tetragyna H. teucrioides = - Eryngium rostratum *Foeniculum vulgare Astroloma A. conostephioides Lissanthe strigosa t Leucopogon parviflorus L. hirsutus - L. virgatus - L. rufus = - Acrotriche serrulata A. fasciculiflora - Epacris impressa - Sprengelia incarnata *Anagallis arvensis *A.femina = - *Olea europaea - Logania fliniflolia *Erythraea centaurium *Asclepias rotundifolia *Convolvulus arvetisis *Echium plantagineum Teucrium racemosum *Lavandula Stoechas *Marrubium vulgare *Salvia verbenaca Prostanthera Behriana *Stachys arvensis *Solanum nigrum *S. sodomaeum = - *Lycium ferocissimum *Verbascum virgatum *Tinaria Elatine - *Bartschia latifolia *Plantago varia = - *P_ lanceolata - *P. coronopus - Opercularia scabrida humifusum 131 Life Form MM MM MM 2 a al Z2me2azdeeezeazgaaazaezaaazaanogniasr 420 > AAAHHH ZS SZ AZHZAZAs ie) om E, odorata association Cd (ee ~ AK E. viminalis assoc, E, leucoxylon- land bh ~ * HW 4” HW Sclerophyll of E, odorata-E, leu- coxylon-E, fasct- culosa ecotone . yn OA Casuarina stricta association Stringybark edaphic complex =~ wn nad HK KH HX WH KH A KM OR “a *Sherardia arvensis - *Scabiosa maritima = Wahlenbergia gracilis Lobelia gibbosa - es Goodenia geniculata - G. primulacea - - Velleia paradoxa - Scaevola microcarpa - Dampiera lavandulacea Brunonia australis - Stylidium graminifolium Brachycome exilis - Vittadinia triloba - Olearia grandiflora = - O. ramulosa - - Senecio hypoleucus = - *Osteospermum moniliferum Cryptostemma calendulaceum Helichrysum Baxteri H. scorpioides - - H. apiculatum = - - Ixodia achilleoides - *Inula graveolens - Craspedia uniflora - *Cynara cardunculus - *Centaurea calcitrapa - *Carthamus lanatus - *Tragopogon porrifolius *Cichorium Intybus - *Hypochoeris radicata *Picris echioides - - *Sonchus oleraceus - Megaphanaerophytes Mesophanaetrophytes Nanophanaerophytes Chamaephytes - Hemicryptophytes 132 Life Form Ch AeAzzntndamAzogggrz2ar RAUNKIAER’S Lire Form Kry (MM) (M) (N) (Ch) (H) oS o = aa of 2 ge aR ws £9 of eg Tu tH og Se ad piel q) (2) x =x . x x x x x x . x - x x x x x x x x x Pr x x x x x x x x x x x x x x x x Geophytes - Helophytes Therophytes Epiphytes Succulents Sclerophyll of E, odorata-E. leu- coxylon-E. fasci- culosa ecotone (3) Casuarina stricta association Stringybark edaphic complex | E. odorata 4 E. camaldulensis 7 E. fascicvulosa 2 E. teucexylon 5 E. obliqua 8 E. cosmophylla 3 EL yimngls . 6 E. Baxter 9 E. rubida PLATE V Meteraph of anndel ta seals showing Wastribution wt chier Knculypt species in Peon bye Lopoe ry piiw. Trans. Roy. Soc. S. Aunst., 1948 Vol. 72, Plate V Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. fw nd Bw dN EXPLANATIONS OF PLATES VI—IX Pirate VI Aerial photograph showing sclerophyll formations (dark areas) on podsols on Stonyfell Ridge as compared with savannah woodland on grey-brown podsols on Greenhill and Long Ridge to the South. Aerial photograph showing sclerophyll forest on podsols on Black Hill and Rocky Hill (Morialta area) compared with savannah woodland on grey-brown podsols near Montacute Road. Piate VII E. leucoxylen savannah woodland on the north-facing slope Elbow. 1 E. viminalis savannah woodland on the south-facing slope } mile above Devil’s Elbow. 3} mile above Devil’s E. leucoxylon sclerophyll association on truncated lateritic soils. E. leucoxylon—Banksia marginata association on the deep sandy soils near Happy Valley reservoir. Piate VIII E. camaldulensis savannah woodland near Blackwood, E, fasciculosa sclerophyll scrub on truncated lateritic soils near Chandler’s Hill. E. odorata sclerophyll association in National Park. E. Baxteri sclerophy!l forest near Mt. Lofty summit. PLaTe IX E. obliqua sclerophyll forest near Mt. Lofty summit. E. cosmophylla sclerophyll association in Waterfall Gully. E. rubida association on grey-brown podsols (low nutrient status) at Picadilly Valley. Casuarina stricta association on deep neutral sandy soils between Hackham and Morphett Vale. Vrans. Roy. Soc. S. Aust, L948 Vol. 72, Plate VI Fig, | Vil Plate i I~ Vol. LO48 Aust, Soc, &. Roy. Traus. Vol, 72, Plate VIII 1948 Roy. Soc, S. Aust., Trans. I “Sig r 72, Plate IX Trans, Rov, Soc. S. Aust, 1948 Vol. 72, Plate ig. 4 INDO-PACIFIC INFLUENCES IN AUSTRALIAN TERTIARY FORA MINIFERAL ASSEMBLAGES By IRENE CRESPIN Summary A recent examination of the microfaunal content of samples from numerous bores drilled in the Adelaide Plains by the Department of Mines, South Australia, has revealed some interesting information about the palaeogeography and stratigraphy of Tertiary sediments in Australia. Certain forminiferal assemblages and zonal foraminiferal species have been discovered in the Tertiary sub- surface deposits in the Adelaide Basin which indicate an extension to Southern Australia of Indo- Pacific conditions in Upper Middle Miocene and Lower Pliocene times. 133 INDO-PACIFIC INFLUENCES IN AUSTRALIAN TERTIARY FORAMINIFERAL ASSEMBLAGES '!! By Irene Cresrin {Read 10 June 1948] 1. INTRODUCTION A recent exantination af the microfaunal content of samples from numerous bores drilled in the Adelaide Plains by the Department of Mines, South Australia, has revealed some interesting information about the palaeogeugraphy and strati~ graphy of Tertiary sediments in Australia. Certain [nraminiferal assemblages and zorial foraminiferal species have been discavered in the Tertiary sub-surface deposits in the Advlaide Basin which indicate an extension to Southern Anstralia of Indo-Pacific conditions in Upper Middle Miocene and Lower Pliocene times. The work upon which the wriler has been engaged in recent years has involved investigation of foraminiferal faunas af Tertiary deposits nat only in Australia but in New Guinea, Papua, Timor, Java and Sumatra, With the dis- covery of Indo-Pacific assemblages in the sediments in the Adelaide Basin, time seems opportune to consider some observations that have been made and to attempt a correlation of the Australian deposits with those in the Indo-Pacific Region. Perhaps the most striking observation is the very large area over which the same foraminiferal assemblages and even the rock types are to be found in the Region. The area extends from Japan south through the Philippines and Sumatra, east through Jaya, the smaller islands. cf the Netherlands East Indies, New Guinea; Papua, the Solomons and New Ilebrides, and north to islands such as Guam, From Timor it extends south to north-western Austral, and from there across southern Western Australia to South Australia and north-western Victoria. Early in these mvyestigations it was realised that eastern Australian Tertiary and Recent faunas were very slightly influenced by the climatic and hathymetric conditions that prevailed in the Indo-Pacific Region from Tertiary times onward, and it is intended in this paper to indicate the hmuts of such influence, as indicated by the distribution of the larger foraiminifera. The sources from which information regarding the distribution of the larger foraminitera in the Australian Tertiaries has been drawn, are rock collections which have heen made during reconnaissance surveys by geologists attached to various companies in their search for oil in the North-West Basin of Western Australia, and by geologists of the Commonwealth Bureau of Mineral Resources from the Nullarbor Plains and Eucla sections in Western and South Australia, as well as samples from bores which have been drilled for water in the Adelaide metropolitan area by the South Australian Department of Mines, and in the Mallee and Wimmera areas in north-western and central western Victoria by the State Rivers and Water Supply Commission. [fl PUBLISHED AND UNPUBLISHED SOURCES OF INFORMATION Little published work is available on the problem of the stratigraphic position of the Tertiary foraminiferal rocks in Australia and their relationship with Indo- Pacific stratigraphy. Howchin (889) recorded Lepidacyclina under the name of Orbitolites from the limestones at Clifton Bank near Hamilton, Victoria. ©) Published by permissiun of the Director of the Bureau of Mineral Resources, Geolazy and Geophysics, Departinent of Supply and Development. Based oo u proper read before Section C of the A.N.Z.AVA.S, Verth, August 1947, Trans. Roy. Soc, S Aust., 72, (1), 23rd August, 1948 NOATHERN TERRITORY GREAT AUETRAUAH BIGHT AUSTRALIA SHOWING DISTRIBUTION OF MARINE TERNARY : ROCKS AND FORAMINIFERA tocAuTicS AFFECTEO fs ex eu, BY INDO=PAGIFIC INFLUENCED, yreace 4 us ca ET Chapman (1910), in his study of the Batesford limestone, referred the Lepidocyclinae to European species, and Crespin (1926) listed Ledipocyclina from a limestone at Green Gully, near Keilor, Victoria. The first record of larger foraminifera in north-west Australia was made by Chapman (1927), when he determined Lepidocyelina dilatata in a limestone submitted to him by F. G. Clapp from Exmouth Gulf. Later, in 1935, Chapman and Crespin determined Eocene foraiminifera from rocks from. the Giralia-Bullara area, North-West Division, Western Australia, collected by Messrs. Rudd and Condit. Species of Discocyclina, Asterocyclina and Pellafispira were recorded. In 1932 these two workers described Lepidocyclinae from bores in the Tertiary rocks of Victoria and suggested a correlation (based on the determination of Spiroclypeus in a limestone from Hamilton) of these deposits with “Stage e” in Java. Crespin later (1936) considered the specimen to be a section of a test of Lepidecyclina. Raggatt (1936) included a short account based on departmental reports by Chapman and Crespin, of the Tertiary rocks of the North-West Basin, Western Australia, in his geology of the area, 135 {fn a report to the Australian and New Zealand Association for the Advance- meat of Sesence in Auckland (1937), the writer summarised the occurrences of Tertiary marine rocks in north-west Australia and indicated their relationships with Indo-Pacific types, and at the same time Chapman reported on the discovery of Eocene rocks in that area. Crespm (1936) discussed the larger foraminifera in the Tertiary rocks of Victoria which were referred to the Lower Miocene, and suggested that the sub- genus Lepidocyelina indicated a younger age for the beds. Later, in 1941 and 1942, she made: detailed studies of the genera Cyeloclypeus and Lepidoryclina respectively in the Tertiary of Victoria and indicated their position in the Indo- Pacific stratigraphy. The subject was discussed further in her work on the Ter- tiary deposits of Gippsland, Victoria (1943). Glaessner (1943) brought up to date the various lincs of thought regarding Indo-Pacific Tertiary stratigraphy in his “Problems of Stratigraphic Correlation in the Indo-Pacific Region.’ Australia received little attention by Glaessner because of lack of published work. Information on the foraminiferal assemblages in South Australian Tettiary deposits was derived from Howchin and Part’s study of the foraminifera in the iietropolitan Abattoirs Bore, Adelaide (1938), and from Howchin'’s work on other bores in 1935 and 1936. Chapman’s report (1916) on the Mallee Hores in Victoria formed the basis for work in that area and in the Wimmera. Considerable evidence has been derived from’ unpublished departmental palaeontological reports since 1934, and these may be discussed under ewo head- ings: f(a) North-West Basin, Wester Australia. atid ([b) Southern Western Australia, South Australia and northavestern Victoria, (a) Norta-Wesr Basix, WESTERN AUSTRALIA Information has been derived from rock collections made by geologists dur- ing recornaissance surveys for various companies engaged in the search for oil. In 1934 D. Dale Condit and E, A. Rudd, on behalf of Oil Search Ltd., investigated the Exmouth Gulf area and collected limestones containing typical Eocene atid Miocene Indo-Pacific foraminifera. H. G. Rageatt carried out further work for Oil Search Ltd. in this arca in 1934, and in 1935 he was accom- panied by Washington Gray of the Commonwealth Oil Refineries Ltd, Both geologists made large collections of fossiliferous material, Im 1936 E, A, Rudd made a further reconnaissance of the Exmouth Gulf area for Oil Search Ltd. and extended his survey down to the coastal section at Red Bluff and Cape Cuvier, aud in 1941 geologists attached to Caliex (Australia) Oil Development, visited the Exmouth Gulf area and made enllections from sections in Rough Range and Cape Range area, (b) Soutuers Westexn Austratia, South AUSTRALTA AND NorrH-WEsrERN VICTORIA The greater part of the information on this region has been derived fom hores drilled for water in South Australia and north-western Victoria, and from collections of fossiliferous limestones male by geologists of the Bureau of Mineral Resotirces, Canberra, and students of the University of Adelaide. The State Rivers and Water Supply Commission ef Victoria has bately been drilling for water in the Wimmera, and since 1908 has had a drilling programme in the Mallee. In 1944, H, B. Owen of the Burean of Mineral Resources accompanied a party of geophysicists of the same organisation on a trip along the coastline of the Great Australian Bight from Port Lincoln jn South Australia to Eyre tu 136 Western Australia, and then across to Booanya and Balladonia. Owen collected from all available limestone outcrops between Ceduna and Norseman and made geological observations in this Tertiary Basin. Last year Mr, Kitig, a student of the Department of Geology, University of Adelaide, explored caves on the Nullarbor Plains and collected rocks from various localities in the vicinity of the caves. Limestones have also been examined by the writer from Cook and Ooldea on the Transcontinental Railway, HI. METHOD OF CORRELATION OF TERTIARY MARINE ROCKS IN THE [NDO-PACIFIC REGION The basis of correlation of the Tertiary marine rocks in the Indo-Pacific Region is the distribution of the larger foramimifera in them. Dutch palaeonto- logists found it difficult to recognise some of the European stages in the Indo- Pacific, and in 1927 Van der Vlerk and Umbgrove instituted a “letter” classifica- tion, Since that time a considerable amount of detailed work has been carried out on the Tertiary fautias in the Netherlands East Indies by Dutch palaeontolo- gists in collaboration with field geologists, with the result that a convenient classi- fication was put forward by the late Dr. Tan Sin Hok in 1939, and this can be applied readily to those ‘Tertiary rocks in Australia which come under the influence of Indo-Pacific conditions at the time of deposition, Tan's scheme, which has been modified by Glaessner (1943), is as follows: Upper Neogene (formerly “g” -“h”) Upper Middle Neogene ("E, *) Neogene - = > Lower Middle Neogene (“L, ’ -"£.,") (Lower Neogene (Aquitanian, “e”, included in Miocene) Oligocene (‘c?-"d", distinguishable from Palaeogene - - 4 ., Agquitanian only if reticulate Nummul- lites present) Eocene (“a’ - “b’?) This scheme is further modified in its application to Australia. It is con- sidered that the following close approximations can be made—the Lower Neogene ta the Lower Miocene, the Middle Neogene to the Middle to Upper Miocene, and the Upper Neogene to the Pliocene. Characteristic larger foraminifera for the stages are: Pliocene - ~ - Absence of larger zonal foraminifera Lipper Miocene (L, vy - Trybliolepidina rusten Middle Miocene (“E, ae oe Nephrolepidina, Miogypsina, Flosculinella, Cycloclypeus, Ausirotrillina lower M iocene (ery - - Lulepidina, Spiroclypeus, Nevalveolina Oligocene (‘'c”-“d’?) - Eulepidina and reticulate Numazelites Eocene (“s7-"b’") - Assilina, Nunimultes, Pellatispira, Disco- cjclina, Asterocyclina, Actinocyclina Rocks containing typical Eocene (“a"-"h”), Oligocene (“c”-“d") and Lower Miocene (“e’’) larger foraminifera are fotind only im north-western Aus- tralia. Zonal foraminifera af “f stage” are widely distributed not only in rocks in the North-West Basin but on the Nullarbor Plains and in the Eucla Basin, and in bures in the Adelaide Plains and north-western Victoria. ‘lypical “f,” rocks 137 with Tryblivlepidina rutlent have not yet been found. Pliocene deposits probably equivalent to ‘'g” occur along the north-western cuast, on the Nullarbor Plains, and im the Adelaide Basin. IV. APPLICATION OF THE “LETTER” CLASSIFICATION TO AUSTRALIAN MARINE TERTIARY DEPOSITS Localities from which typical zonal foraminifera of the Indo-Pacific Region have been recorded in Australia are listed in Table 1, which also gives the strati- graphic position of the various deposits. A. Rocenr (“a -“b") The only localities in Australia from which “a”-"b” foraminifera have been recorded are Giralia and Bullara south of the head of Exmouth Gulf. and at Red Rluff and Cape Cuvier in the coastal section south of the Exmouth Gulf area in the North-West Basin, Western Australia. Species of Discocyclina, Astero- cyclina, Actinocychina, Pellatispira and Numogprulites have been determined, The exact position of these beds in the Eocene stratigraphic sequence is uncertain, Pellatispira and Discocyclina disappear at the top of the Eocene, B. Oricocenr (“e"> “d*) Definite Oligocene limestones are also restricted to the North West Basin of Western Australia. Large species of Eulepidina (E, dilatata, E. papuaensis) large Cycloclypeus (C. eidae) and small reticulate nummulites occur in the lime- stones from the base of an open gorge north of Mount King, Cape Range, and from the base of the section at Badjirrajirra Creek (north fork) 4°7 miles from it; mouth on the Exmouth Gulf side of Cape Range. Chapman recorded #. dilatata from Exmouth limestones in 1927, He described E, pupuaensis from a limestone at Bootless Inlet, Papua (1914), which contained small nummulites (N. iatermedia) and which is now considered tu be of Oligocene age. C. Lowrr Mtocenr (“e’’) Similarly, limestones containing typical Lower Miocene zonal foraminifera have not been foutd south of the North-West Basin. Eulepidines typical of stage “e” CE. insilaenatais, LE. papuaensis) are present in limestone 22 titles south of Yardie Creek Station, Cape Range. Other zonal forms, Spiroclypenus tidvegunensis and Neoalveolina pyytiaed, together with small Nephroleptdina occur in rocks from Rough Range near Exmouth Gulf Homestead. Eulepindina, Spireclypeus and Neoalveolina are not known to range above the top of “stage ev”. D. Mivpin to Uerer Miocene (£") Rocks of “stage {” are widely distributed in the North-West Basin und accur as far south as the section across southern Australia from Roganya and Tialladonia in Western Australia tu Colona east of the head of the Great Aus- tralian Bight in South Atistralia, thenee to tear Adelaide and across to north- western Victoria. No typical “f,"" rocks have been found. It ig more satts- factory to subdivide the “stage f” rocks irr Australia into “f,-f," and “f, - £,”- Probably some of the limestones in the Badjirrajirra Creek section may he refer~ able to “f,", but “£,”" has not yet been recorded. “f,=£," rocks contain two assemblages of zowal foraminifera : (2) An upper one with Nephrolepidina with some species showing Tryblio- lapidina tendencies, Floseudinella, Cycloclypeus and Miogypsina. (1) A dower one with Nephrolepidina and Cycluclypeus, 138 Limestones containing Assemblage 1 occur in the Badjirrajirea Creek svc- tion, Cycloclypeus being the dominant form. This tock may be referable to “f,"'- Well preserved specimens of the echinoid Conoclypeus were found in thése beds, Assemblage 2 shows an intermingling of typical “f° and “f,” species and is fairly widely distributed in limestones in the Cape Range and Rough Range sections. “f,-£," rocks have a wide distribution in Australia, and two assemblages of foraminifera are recognisable in them; (1) Without Lepidacyclina but with Austrotrillina, Flosculinella, Margino- pora, Valvulina, Sorites, Peneraplis and Miliolidae, (2) With Lepidocyelina, also Flosculinella, Marginopora and Miliolidae. It is possible that these two assemblages are equivalent in age, the absence of Lepidocyctina in Assemblage 1 being due to change in facies rather than difference in age, Assemblage 2 is recorded from only two localities, one at Exmouth Gulf Station Outcamp in Rough Range, and the other near Yardie Creek Homestead, Cape Range in the North-West Basin. The assemblage is well known i the limestones in New Guinea and Papua and in parts of the Netherlands East Indies- Assemblage 1 occurs in rocks in widely separated areas, fram the North- West Basin, Western Australia, through the Nullarbor Plains to South Australia, thence to north-western Victuria, the same rock type octirting from North-West Basin to the Nullarbor Plains. The limestone is cryptocrystalline, the fora- minifera tests usually being stained black, The assemblage occurs in friable: limestones in bores in the Adelaide Plains and in the Mallee and Wimmera inl north-western Victoria. Rocks containing Indo-Pacific Pliocene assemblages are not common in the Australian ‘Tertiarics. Some of the coralline limestones in the Yardie Creek area, North-West Basin, have been placed in the Pliocene but no joraminitera are available to confirm this. The limestones near Minilya Station, North-West Basin, contain Margino- pora, Sorites, Peneraplis and Valvulina and in the absence of zonal Miocene farms ate referred to the Pliacene, An identical rock, both in lithology anil faunal content, occurs at Qoldea on the NuUarbor Plains and in the vicimty uf Weebabbie Caves near Eucla. Numerous casts of molhiscan shells, reierable to species recorded from the Adelaide Plains deposits are present in the rack, In the vicinity of Adelaide, extensive fossiliferous deposits known as the “Adelaidean” underlie the Adelaide Plains and outcrop at Ilallett's Cove, at Christie's Reach, Port Noarlunga and at Aldinga. The foraminifera genera Marginopora, Sorites, Peneroplis and Valyaltma torm a characteristic assemblage, bit the occurrence of zonal species of the Kalimnan stage of Victoria indicate a Lower Pliocene age for the beds. Vv. COMMENTS ON THE INDO-PACIFIC ZONAL FORAMINIFERA IN AUSTRALIAN TERTIARY DEPOSITS The zonal foraminifera in the Eocene (“a-b"), Oligocene (“c-d’") and Lower Miocene (“e’) rocks call for little comment. All species that have been determined are characteristic of the deposits of these ages throughout the Inco- Pacific, from Japan, the Philippines, Netherlands East Indies, New (Guinea, Papua and north-western Australia, but are not known from southern Australis. Tlowever, in the Middle to Upper Miocene rocks which do extend to southein Australia, several species vf zonal importance througboul the Indo-Pacific not anly occur there but were described from the area, 139 The most important zonal form. used in the correlation of “{ stage” rocks is Lepidoeyelina with its two subgenera Nephrolepidina and Trybliolepidina, The genus is found in the North-West Basta of Western Australia, but has not yet been recorded from limestones on the Nullarbor Plains nor in the vicimty of Adelaide and north-western Victoria, but it does occur in western, central southern and south-eastern Victoria. Lhe subgenera Nephrolepidina is the zonal furm for “{,-1,” beds and it has not been recorded south of Cape Cuvier. Similarly, Miogypsina, which is characteristic o£ the "“f,-f," assemblage, has not been found south of the North-West Basin, In “f,-£," beds, Nephrolepidina gradually gives way to Tryblolepidine, which is yery common at this horizon in the Indo-Pacific. On the Nullarbor Plains, from Boounya in Western Australia to Colona in South Australia, thence to the Adelaide Bures and Mallee and Wimmera Bores, 7rybliolepidina is replaced by an assemblage in which the most important zonal form is Austro- ivillina howchint, The associated zonal genus Flosculinella does not occur east of Colona, Austrotrillina howehint, of such importance in “f stage” foraminiferal faunas in the Indo-Pacific, was described by Schlumberger (1893) from Lepidocyclina limestones at Clifton Bank, near Hamilton, western Victoria. The species has been recorded from “e stage” but has its preales! development in "f stage’. The age of the beds from which .4. howchini was described has been based on the Lepidocyclina population, which is dominated by the subgenus Trybliolepidina and which is regarded as the equivalent of “f,~ 0", aud somewhere about the boundary of the Burdigalian and Vindobontan of the European stages (Crespin, 1942), Records indicate that 4. iowchint has not been found in the Indo-Pacific Region outstde Australia in rocks younger than "f,”. Other species which have been described from the Victorian Lepidacyclina limestones and which are found as far north as the North-West Basin ih Western Australia, are Gypsina howchini, Planorbulinella plana, and P_ inaequi- lateralis, Two other zonal species, Crespinella umbonifera and Calcarina verri- culate, were described from bores in the Adelaide Basin from the ustrorrillina Aowchini horizon (Howchin and Parr, 1938), C, werriculata is very common in the Lepidocyclina limestones at Batesford, Both species are fotmd m the North-West Basin, An interesting assemblage of recent warm water genera, Muarginopora, Sorites, Peneroplis, Miliolidae, oceuts in the “f,-4," rocks and in the Pliocene in that portion of the Australian Tertiaries which have been subjected to Indo- Pacific influences at the time of depasition. This assemblage is found in rocks of topmost Miovene to Recent age throughout the Indo-Pacific, age being deter- mined by the presence of zonal foraminifera. It is associated with dustrotrillina in “f,-f,” limestones at Cape Cuvier im north-western Australiz, in some of the limestones on the Nullarbor Plains, and in bores in the Adelaide Plains and in the Mallee and Wimmera. Marginepora rarely occurs in association with Lepidocyelina, duc apparently to the fondness of the former genus for quiet iropieal waters such as are fuund in the proximity of coral reefs. Lepidocycline is usually associated with bryozoa which thrive where currents are present. The Marginopors-Sorties-Peneroplis-Miliolidac assemblage occurs with zonal Lower Pliocene (Kalimnan) species in the “Adelaidean’’ of the Adelaide Basin. VI. POSITION OF THE MIDDLE TO UPPER MIOCENE AND PLIOCENE DEPOSITS OF SOUTHERN VICTORIA IN THE INDO-PACIFIC “LETTER” CLASSIFICATION Rroadly speaking, the majority of rhe bryozoal limestones in the Tertiary deposits of southern Victoria can be correlated with “f,-f,” stage’, based on 140 the presence of Trybliolepidind in the upper portion of the section. The writer, in her investigations on the genera Lepidocycline and Cycloclypeus in Victoria (1941, 1942), showed that the subgenus to which the Lepidocyclina belonged was Trybliolepidina, and that the species of these two genera were different trom those in known Indo-Pacitic deposits. She suggested that this difference hit species. was duc to “the presence of ati embayment in south-eastern Atistralia in which other species, but still with Indo-Pacific affinities, flourished.” Glaessner (1943) sums up the problem in the following statement: "Fauna! relations between the surprisingly uniform Tertiary of the Indo-Pacific Region and that of south-eastern Australia and New Zealand are limited and are either created hy “eturythermic” species which were able to cross the boundary of the tropical belt or by short-lived climatic or ecologic changes, creating a suitable environment for warm-water species and genera. In the Tertiary of Victoria the Bateslordian, containing a group of Lepidecyclina as well as Ausirotrillina and perhaps some other Indo-Pacific smaller foraminifera, forms a short-lived link with the Indo- Pacific Region.” Although 4Axsirorillina howchini was described from the Lepidecyctina lime- stone at Clifton Bank near Hamilton, it is not a commo y iorm in the Lepido- eyeling deposits in southern Victoria. It does not occur im the rich Lepidoryclina limestones at Batesford and Flinders in the central southern portion, nor in Brock's Quarry and in the mamerous bores from which th> Lepidocyclina horizon has been recorded in Gippsland, but it does occur sparingly in that horizon at Skinner's section, Mitchell River, near Bairnsdale, Gippsland, Tine specivs is well repre sented in the partion of the Mallee Bores which have been subjected to Indo- Pacific influence and where it is associated with Marginopora. The typical Indo-Pacific genera Miogypsina and Flosculinella have not been recorded trom Victoria, The Middle Miocene bryozoal limestones of south-eastern South Austraha and western Victoria contain no large zonal foraminifera, but some of the Time- stones are correlated with the “f,-f,” beds of Victoria by means wi smaller zonal species which are associated with the larger forms in the Lepidorycfina limestones, No zonal Indo-Pacific foraminifera are known From the Victorian Lower Pliocene—referred to as the Kalimnan—which suggests unsuitable ccolagic con- ditions during Pliocene times in Victoria. According to present knowledge, suit ahle conditions did not extend eastward beyond the vicinity of Adelaide. Vil. SUMMARY 1, Indo-Pacific influence in faunal assemblages in Australian Tertiary rocks extends from North-West Basin, Western Australia, to southern Western Aus- tralia, across the Nullarbor Plains to South Australia and north-western Victoria, 2. Correlation of the marine Tertisaries of Australia with occurrences clse- where in the Indo-Pacific Region is made by means of the larger foraminifera, which form the basis of the letter classification instituted by Dutch palaeontolo- gists in their work in the Netherlands East Indies, 3. Eocene (“a-b'), Oligocene (“c-d") and Lower Miocene (“e'’) rocks containing typical zonal foraminifera are not recorded south of the North-West Basin, Western Australia. 4, Middle to Upper Miocene (“f") zonal foraminifera are widely is- tributed in limestones in southern Western Australia, across the Nullarbor Plains, in the vicinity of Adelaide and in north-western Victoria, Even similar rock types have been recognised. 141 5. No larger zonal foraminifera of age yalue are present in the rocks of Pliocene age, local zonal species being age determinants. But associated with the local species are well-known Indo-Pacific recent forms. Rocks containing such assemblages occur in the North-West Basin, at Qoldea on the Nullarbor Plains and im the Adelaide Bore sections and in the coastal sections south of Adelaide, where they are referred to as the “Adclaidean” and gre Lower Pliocene. 6. The Lepidocyclina limestones of Victoria, which are considered to be equivalent to “f,-f,” stage indicate a short-lived link with the Indo-Pacific. The genus Lepidocyclina, which is represented by the subgentts Trybiialepidina, shows Indo-Pacific affinities, but the species are distinct. 7. The conclusion drawn from the above observations is that Indo-Pacific conditions extended down the coast of Western Australia, across the southern part of the State into South Australia and north-western Victoria in Middle Miocene to early Upper Miocene (“f.,-f,”) time, with a very limited extension into southern Victoria, Then followed a gradual recession of conditions. No Pliocene deposits containing Indo-Pacific forms are known east of the Adelaide Plains and the sections south of Adelaide along Gulf St. Vincent. According to Cotton, of the South Australian Museum, the examination of molluscan faunas in South Australia suggests that this recession has coritinued westwards during Pleistocene and Recent times. Vill. REFERENCES CuarmaAn, F, 1910 A Study of the Batesford Limestone. Proc. Roy. Sac. Vict., N.S. 22, (2), 263 Charman, F, 1914 Descriptions of a Limestone of Lower Miocene Age from Bootless Inlet, Papua. Journ. and Proc, Roy. Soc. N.S,W., 48, 281 Cuapman, F. 1927 Ona Limestone Containing Lepidocyclina and other Fora- minifera from Cape Range, Exmouth Gulf, Western Australia. Proc. Roy, Soc. Vict., N.S. 39, 125 CuArman, I, and Cresrix, I, 1930 Rare Voraminifera from Deep Borings in the Victorian Tertiaries—Victoriella gen. noy., Cycloclypeus com- mutus Martin, and Lepidocyclina hornenensis: Provale. Jbid,, N.S, 42, (2), 110 Cuarman, F., and Cresmn, I, 1935 Foraminiferal Limestone of Eocene Age from North-West Division, Western Australia, /bid., N.S. 48, (1), 55 Connpit, D. D, 1935 Oil Possibilities in North-West District Western Australia. Feon. Geol., 30, (8), 860 Cresrin, I. 1926 The Geology of Green Gully, Keilor, with Special Reference to the Fossiliferous Beds. Proc. Roy. Soc. Vict., N.S. 38, 101 Cresrin, I. 1936 The Larger Foraminifera from lhe Lower Miocene cf Vic- toria. Bur. Min. Res. Bull. 2, Pal. Ser. No. 2 Crrsetn, I. 1938 Tertiary Rocks in North-West Australia, Kept. AN.ZAAS., 23, (Auckland), 443 Crespin, T. 1938h The Occurrence of [.acazing and Biplanispira in the Man- dated Territory of New Guinea and a Lower Miocene Limestone from Ok Ti River, Papua, Bur. Min, Res., Bull, 3, Pal. Ser. 3 Crespin, I. 1939 Note on the Present Knowledge of the Tertiary Sequence m Papua and the Mandated Territory of New Guinea. Proc. 6th Pac. Sct. Cong., San Francisco, 529 142 Crespin, I, 1941 The Genus Cycloclypeus in Victoria. Proc, Ray. Soc. Vict., N.S. 53, (2), 301 Crespin, I. 1942 The Genus Lepidocyclina in Victoria. Jbid., 55, (2), 157 Crespin, I, 1943 The Stratigraphy of the Marine Tertiary Rocks in Gipps- land, Victoria. Bur, Min. Res. Bull. 9, Pal. Ser, 4 (Mimee) GertH, H. 1929 The Stratigraphical Distribution of the Larger Voraminifera in the Tertiary of Java. Proc, Fourth Pac. Sci, Cong., 591 Gertu, H. 1935 The Distribution and Evolution of the Larger Toraminifera in the Tertiary Sediments. Proc. Kon. Akad. van Wet. Amsterdam 38, (4), 1 Guarssner, M, F, 1942 Problems of Stratigraphic Correlation in the Indo- Pacific Region. Proc. Roy. Soc. Vict., N.S. 55, (1), 41 Heron-Atren, E., and EartAnp, W. 1924 The Miocene Foraminifera of the “Filter Quarry’, Moorabool River, Victoria, Australia. Jour. Roy. Mier, Soc., (2), 121-186 Howeurn, W. 1889 The Foraminifera of the Older Tertiary of Australia. (No. 1, Muddy Creek, Victoria.) Trans. Roy. Soc. S. Aust., 12, 1 Howcuin, W., and Parr, W. J. 1938 Notes on the Geological Features and Foraminiferal Fauna of the Metropolitan Abattoirs Bore, Adelaide, Ibid., 62, (2), 287 Parr, W. J. 1939 Foraminifera of the Pliocene of South-Eastern Australia. Min. and Geol. Journ., Vict., 1, (4), 65 Raccatr, H. G. 1936 Geology of the North-West Basin, Western Australia. Journ. Proc. Roy. Soc. N.S.W., 70, 100 Scuiumepercer, C, 1898 Note sur les Geres Trillina et Linderina, Bull. Soc. Geol, France, (3), 21, (2), 119-120 Sincteton, F. A, 1941 The Tertiary Geology of Australia. Proc. Roy. Soc. Vict., N.S, 53, (1), 1 Tan Stn Hox 1936 Over yerschillende palaeontologische criteria voor de geleding van het Tertiair, De, Ing. in Ned-Indie, 3, (4), 173 Taw Sr~w Hox 1939 Remarks on the “Letter Classification” of the East Indian Tertiary. Jbid., (55), 98 VaucHan, T. W. 1924 Criteria and Status of Correlation and Classification of Tertiary Deposits. Bull, Geol. Soc, Amer., 35, 677 Vierk, I, M., vAN pER 1928 Het genus Lepidocyclina in het Indo-Pacifische Gebied. Wetensch. Mededeel, 8 ViERK, I. M. vAN peR, and Umecroye, J. H. F, 1927 Tertiaire gildsfora- miniferen van Nederlandsch Ost-Indie. Jbid., 6 Taste 1 Correlation of Australian Tertiary Localities Based on Indo-Pacific Foraminiferal Assemblages De eT ee ee ee ee ee ee a Letter. Nullarbor Plains, i . Mallee and Classification Zonal “yar ° , . * Adelaide Basin, Wimmera, Age patD tc v Foraminifera North-West Basin, Western Australia Te Celene x S.A. North-Western 11 miles N.W, of Minilya-Waroora Marginopora boundary on coast road. Sorites Minilya Station } m W. of Gerardi Adelaide Bores Valiulina Bore. Ooldea “Adelaidean” Lower Pliocene ? “g" Peneroplis Minilya Station 3.25 m. S.E. of Chirrita Surface near Hallett Cove -- (absence of Minilya Station 3.25 m. S.E. of Chirrita Weebabbie Christie’s Beach zonal Miocene Bore, Caves Pt. Noarlunga forms) Mini'ya Station shallow well on track Aldinga Section : between Chirrita and Gerardi Bores. Yardie Creek, W. of Cape Range. “F 3" Trybliolepidina —_ -= _ = rutient E. flank, Giralia anticline where Bullara Mallee Bores track crosses first Jow hills. Adelaide Bores Nos. 2, 4, 9, 11 (without Approx, crest Chargoo Dome { m, from Bore No. 65 (near Murray- Lepidocyclina) Minilya-Waroora boundary fence. Booanya (600'— 620") ville) “EU Austrotrillina Limestone Range on Waroora road Balladonia Bore No. 58 Ph. Ballarang, Flosculinella 4.5m. N.W. of Waroora road crossing Madura (208'- 311’) Allot. 28. to Calcarina of Lyndon River on main coast road Top of Eucla Nathan Brewery | Campbell's Bore ; Crespinella from Minilya, Section (500’— 524’) Ph. Tunart ro Marginopora Trealla Hills, Cape Range, S. end of White Wells Kinnish Direk (387'= 406’) Valzilina Chargoo Anticline. Outstation No. 1 (280’- 318’)| Wimmera Bores Miliotidae Cape Cuvier, Coastal Section. Colona Kinnish Direk Dimboola No. 1 Upper Miocene Gully 4 m. N. of Red Bluff, Coastal No. 2 (265’- 365’)| (80'~ 131") Section. wen OT Terra Se — eT | Sara Leputocyclina (chiefly Yardie Creek Homestead, Cape Range. — — _— Middle Miocene Trybliolepidina) | Exmouth Gulf outcamp Rough Range. Flosculinella Marginopora Miliolidae Gorge W. of outcamp Exmouth Gulf Station, Rough Range. Open gorge N. of Mt. King, Cape “FQ” Nephrolepidina Range (top of section). Flosculinella E. of Yardie Creek Homestead, Cape to Miogypsina Range. — _ — Cycloclypeus Bed of Creek, 3.5 m. from Bullara on 1" road to Giralia, S. of head of Exmouth Gulf. Portion of Badjirrajirra Creek Section, Cape Range. Rough Range, near Exmouth Gulf Bulepidtia Station Homestead, Neoalvectina 22 m. S. of Yardie Creek, Cape Range. _ a I ewer “Milscent Mel! Spiroclypeus W. end of Cape Range, Yardie Creek Miogypsinoides Station. 2 Cycloclypens Gorge N. of Exmouth Gulf Station outcamp, Rough Range. Eulepidina Open gorge, N. of Mt. King, Cape Oligocene “c"— dd" with reticulate Range (base of Section). — —_ Aa Nummulites Badjirrajirra Creek 4.7 m. from mouth, Cape Range. Bed of Creek at track crossing 8.5— Nummulites 9 m. from Bullara on road to Giralia, Discocyclina S. of head of Exmouth Gulf. Eocene “a’—"'b” Pellatispira 10.5 m. from Bullara on road to Giralia. — — -_ Actinocyclina E, flank of Giralia anticline where track Asterocyclina to Bullara crosses the first low hills. Red Bluff Section most southerly ex- posure of yellow limestone, Cape Cuvier. THE MARINE ALGAE OF KANGAROO ISLAND Il. THE PENNINGTON BAY REGION By H. B.S. WOMERSLEY Summary The first paper of this series (Womersley 1947) gave a general description of the algal ecology of Kangaroo Island, together with an account of the more important environmental factors for the island as a whole, and a discussion of the terminology found most satisfactory in describing the algal ecology. 143 THE MARINE ALGAE OF KANGAROO ISLAND II. THE PENNINGTON BAY REGION By H. B.S. Womerstey* [Read 8 Jaly 1948] CoNTENTS INTRODUCTION = fe : Ro - on oe ne ne ENVIRONMENTAL Conarmriné 5 is am Py ZON ATION u's te ote on bs =:3 ah : ss ne Tue Arcar Aswcarions te = = 25 =n Description of the “main reci,” Penningitun Bay .. =~ a bs The Supra-littoral Zone — _. Prasiola community zs te vs vs : a ' Lichina community vt a Isolated rock-pool community =~. The Littoral zane... ~ -- an Rear littoral ‘ Batictabieis fe 1. Symploca hydnoides association i ws 2. Gelidium pusiliim association a4 3, Rivularia firma association .. Ls 7 it 4. Ectocarpus confervoides and Pylaiella futveseent seasonal associations .. Ni 2: 5. Entcromorpha acanthophora agadcdattast by AL +> 6. Ulva lactuca associatior; Littoral assoc‘ations : be as <4 S2 - yé 7. Polysiphotiia Fhutex association be 8. Cystophyllum muricatum association “; au 9, Jania fastigiata asscciation .. 10, Taurencia heteroclada association a+ 11. The Cystophora complex at bs ¥. (a) Cystophora uvifera (b) Cystophora subfarcinata {c) Cystophora siliquosa (ad) Cystophora brownii we i (e) Sargassum tmuriculatum as ze ray 12, Hormosira banksii association a A “i a 13. Cystophora—coralline association Intermediate communities AA re at Chance distribution of minor species on ther main ret a Flora of shaded, rear littoral pools .. 2.6 et: ¥y The Sub-littoral fringe .. _ “¢ +4 + 4 Cystophora intermedia association The Sub-littoral * * ; Zonation below the sub- littoral fringe ‘a a 4 re F'ora of deep outer pools on the reefs “. The deeper sub-littoral flora + + ¥. bs SEASONAL VaRIATION ty THE AtcaL Frora .. ch Ao as 33 et (a) Seasonal occurrence .. 4 Ae e* 2% (b) Seasonal development of stable specits te 1% +4 ty (c) Seasonal variation in reproduction 4 bg _ = . * Departinent of Botany, University of Adelaide. Trans. Roy. Soc. 5. Aust., 72, (1), 23rd August, 1948 144 Vamarien UNues Wave Aciinx ee Ay tn ai 7 7 -» 43 Parisiism anp Evirairtisat 2 ~ af an ’ AA : ‘ 14 VawricaL Distrmecrion om ReLAtON va Lrautr Ad wt An ae .. To4 SvuarMary = _ fe & tx is ' i t - hs AUK NUWLENGEM ENTS ‘, 8 = OU i: ' vs ‘ “1 ’ 16 REFERENCES «(> Men a le se ets te me Ls a: w 15 INTRODUCTION The first paper of this series. (Wemersley 1947) gave a general description of the algal ecology of Kangaroo Island. together with an account of the more neaportant environmental factors for the island as a whole, aid a discussion of the ternunolcey found most satisfactory in describing the algal ecology, This paper deals with one of the distinctive regions of the Kangaroo fsland coast, wts., the horizontal rock platforms of the Pennington Bay regioiy and other areas of the south coast of the island (see previous paper). The Pennington Ray reefs, because of their easy accessibility [rom the American River settle- ment, have been studied in some detail, The sare type of rock platform has heen briefly examined elsewhere along the coast—east of Vivonne Hay and at Sou'-West River—and their algal ecology wppears to be similar in a‘l cases, This is in accordance with the similarity in structure of the recfs and in the environ- mental conditions to which organisms inhabiting them are subject, These rock-platform reefs are classed as a distinet section of the Exposed Rocky-ceast Subformation around Kangaroo [sland (Womers!ey 1947). They are formed by wave action from consolidated calearcous sand-dunes af Recent to Plerstocete age, and alternate with areas of Precambrian rocks along the south coast. Pennington Bay is nut well defined. The western part of the bay ts formed of cliffs 80 ta 100 feet high overlying Preeambrian rocks. In the cetitral and vustern parts the sand-rock cliffs are lower, usually less than 50 feet high, form- ing small outereaps and headiands, separated by sandy beaches backed by sand- dunes (see pl, X), Further to the east the cliffs are more continuous aud tend ta be higher. From the cliff bases stretch ottt the horizontal wave-cut platforms, composed of the same calcareous sand rock as the cliffs. Exeept ina few cates where wave action has eroded small caves at the cliff base, the cliffs descend almost vertically to the Mat reef surface. Rock aboye high water level is usually sharply pitted and ridged, that on the reef itself less so but sul forming a hurd and rough surface. The characteristic feature of the reefs is their flat, horizontal surface and vertical drop off at the edge into decp water of 10 to 20 ject. Many reels are undercut in varying degrees, with occasional tunnels and blow holes up through the reef surface. The following account is based largely on the reefs occurring along some 1} miles of coast at Pennington Bay and to the east. The Precambrian rocks which form the western headland of the bay are vot included in the discussion, The most accessible rcef is also one of the largest, It is situated some 100 yards west of the track to Pennington Bay frem the American River — Hog Bay road, and almost die south trouta large sand hill known as Mount Thishe. This reef itlustrates well the main features of the reefs generaily. and must of the algal associations found in the region occur on it. The distribution of the algal associa- tions on this ‘imain reef? (pl, XI, fig. 1) will be used ay a basis for the following deseriptious, reference being made where necessary to other réefs, 145 Almost the whole surface of these rock platforms is dominated by algac, though moiluses and other animals are characteristic of many associations. In the rear littoral, on sloping and more exposed rock, several animal associations occur, often pire but sometimes ca-daminant with algae in restricted areas. It has praved must satisfactory to deal with the algal associations by themselves. although mentin ig made jn cases where algal-animal biocenoses oceur. A full accoynt of the fauna of the Pennington Bay reefs is given by idimonds (Edmonds 1948). ENVIRONMENTAL CONDITIONS A general accotmt of the environmental conditions of the Kangaroo Island coast has been given previously (Womersley 1947). The mote important environ- menial factors at Pennington Bay are summarised below. Tidal Range—Spring range, 24 feet approximately; neap range, 14 feet approximately, The effect of wind and the state of the sca greatly modify the lidal rise, often cansing low or high water when Icast expected. Roanglness—Conditions on outer paris of the reefs are always rough, becom- ing calmer further in as breakers passing uver the reefs gradually lose their Larce. Breakers are a constant feature of the Pennington Bay coastline, but with a north wind and cal, sea they are only 1 to 2 feet high (pl. X and XT), Such conditions, hawever, are nut common, and with a heavy swell breakers up to 8 or 10 feet high occur off the teefs (pl IX, fig. L, pt. 1).4) During calm conditions and a low tide, most ni the reefs may be exposed, or have few waves passing over them. Jtigh summer air temperatures may kill or damage large recf algae under such conditions. At the reac of the reels conditions are usually much caliner, bat most areas wf slopitig rock are washed over by srnall waves, and pools of still water occur only at low tide. Lentperature—The following table of sea temperatures on the main reef gives some idea of the yearly range. All figures are isolated readings, but the daily or weekly variation is small owing to dcep water close im shore, Taste I Sea temperatures on the main reef, Pennington Bay Jan, April May June July Supt. Nav. Temp. °C, - = 19°0 18-5 17°5 16-0 15-5 14-0 16-0 Temperatures on the main reef are usually within 1° C. of the sea tempera- ture off the reef (warmer in suinmer, warmer or cooler in winter depending on the air temperature). On rock platforms aboye the reef siurface, and usually ahove high water level, isolated pools may occtir where water termperatures reach 30°C. during summer. Such pools bear a characteristic but sparse algal com- munity (see later). The yearly range in sca temperature ig small (about 5:5° C.), but variation ja size and seasonal necurrence of a number of reef algae can best be attributed to this. Airy temporatures—Pew figures of air temperatures ure available. Some for Kingseote (on the north coast of Kangaroo Island) haye been given previously (Womersley 1947). The climate generally is insular and mild, Air tempera- tures of most impartanve in ihe algal ecology occur on hot suyrmer days, when they may reach 37°C. Ef combined with a norfh wind and low tide exposing the ( Part £ refers to the first paper of this series (Womersley 1947), 146 reef surtace, this may result in considerable damage to reei algae, particulariy species of Cystephora and [ermosira banksti (which oceurs on higher levels of the reefs). No harmful effect of exposure during low tides in winter montns has been observed, nor would be expected as air temperatures ave nut often ligher than sea temperatures {roni June to September, IWinds—Southerly to westerly winds are most frequcut on the south coast of Kangaroo Island. Easterlies often oceur, but oceasioual northerlies during hot summer weather arc the only winds of direct importance in the algal ecology (see under Temperatures). Chlorimty—tsolated readings (January 1946) have given a chiorimty of 19°5-19-6/,,. (salinity 35°2-35-4%/,,,) for water om the reefs. Little annual change would be expected. Tsolated pools above the reef surface are often more saline, particularly in summer (chlorinity 20-5-22:29/,,), and usually lower after heavy rain. wkalinity—Colorimetric methods have given a pl of 82 to 8-5 for water on the reef, ZON ATION The relatively small tides, and constant breakers, result in 10 marked hori: zontal zones such as described by numerous workers in regions of vreater tidal range. Algal zonalion does occur, however, and differerices im reef levels of only 2 or 3 inches frequently cause profound changes in the algal associations (see reef section, fig. 1), The greatly dissected nature of the coastline, and the variety and grading of the habitats on the reefs, tend to obscure any obvious zonation, but over the whole area of the coastline examined there occur numerous well-marked associations, characterised by fairly distinct environmental conditions, The average horizontal level of the reefs seems to correspond fairly closely to a tucan low tide level (neaps). Some areas, either on the outer edge or else where on the reefs, are a few inches higher than the rest, while shallow channels and pools. often occur on the reel surface. Such features are well shown on the main reef (see map and pl. X1, fig. 1), The reef surface is therefore classed as being in the litferal sone TTow- ever, Gwing to the form of the reefs, few of the algac on their flat surface are ever completely exposed: many occur in shallow pools or areas of water retained on the teef at low tide. In other slightly higher areas the growth is so dense that a2 considetuble amount of water is retained in the masses, and such areus arvé washed over by occasional wayes. Only at the rear of the reefs on sloping and vertical rock which is classed as the rear [ittoral are the algae exposed for any length of time at low tide (see reef section, fig. 1). The algal associations grouped together in the littoral zone are therefore stibject to a much smaller degree of exposure than are littoral associations on steeply sloping rock in calmer seas. This is a characteristic feature of horizontal rock platforms wherever they oceur, in contrast to {he conditions on stecply slop- ing coasts. Algal ecological literature contains very few accounts of similar rock platforms, which are however a distinctive feature of many parts of the southern Austratian coast. The littoral zone is considered ta imelude all algal associations at higher clevations, subject to wave action ar spray. These may oceur 2 or 3 feet above actual high tide level, but this is due entirely to the effect of wave splash and, in some areas, of shade. () For discussion of terminology see Worersley 1947. 17 The only algal community which can truly be classed at Supraliltaral is one of Prasiola, which is found only where penguin colonies occur, This habitat is alsa subject to fine, blown spray, bnt is more seini-terrestrial than marine. At Jeast ome mollusc association can be best considered ag supralittoral (Edmonds, this journal, p. 168). The edge of the reef, and ta 1 or 2 feet down the vertical side, is termed the sublittoral fringe (pl. XTV, fig. 4). A very distinctive association of algac veeurs on this part of the reefs, The imporiance of the sublittoral fringe zone i the broad aigal ecology of Kangaroo Island has been ouilined in the first paper of ths series, selow the edge of the reef, for at least 3 or 4 feet down {as tat as it is possible ta investigate), zonation qccurs, but the deeper sublittoral flora is known only from the assemblage found cast up and nol growing on the reefs, Dredging olf the shore is, unfortunately, quite impossible, THE ALGAT. ASSOCTATIONS Description or THE Mats Reev The shape and appearance of the main reef at Pennington Tay is shown in pl. XT, fig. 1 and by the map. It measures some 70 yards across and 75 yards from beach to cuter edge. Other rects ae Pennington Ray are shown in pl. EX, fie. 3, pt. I. On the noclh-west carner of the main teel are cliffs aboul 15 (eet hieh, with a lodge 2-3 feet wide at the base. This ledge forms the western side of a sandy pool, which is about 3 feet deep in the outer corer, and has a sinall sancy beach the rear, The amount of sand on the beach and in the pool varies greatly at different times. As a general tendency the beach and pool are heavily sanded up in summer (pl. XIV, fig. 1), but with considerable bare rock in winter (p. XL, hg. 3; also see map; this allows development of Ectocarpus aud Enteromorpha associations on the exposed rock in winier. The amount of sand present, how- ever, 18 largely dependant on the weather over the previous few weeks. The eastern part of the reef is backed by rocks whieh have fallen from the elilfs—the “Tailen vock region” (pl, XL, fig. Land 2). AL the north-ease corner of the reef ig a sinall rack platform whieh contains several reel pouls on the top, about 5 feet above the reef surface, A noticeable teature af rhe reed is the absence of louse rocks an the stirface ; a few which veenr i the fallen rock revion are aimost dry at low tide and too large to be removed by wave action, This results i the virtual absetice of the characteristic fauna, and to a lesser extent the flora, which inhabits the under- surface of loose racks (see Pape 1943), An important structural feature of the main reel is a ledge forming a drep af (12 inches, running in a curve throuch the centre of the reed to the south- cast corner (pl. NL, tig. 1; pl, XTV, fig. 1 and 2, andanap)}. This ledge, due to the wniost continual streaming of water over it, even at low tide, bears a distinct algal ussociation, The areas colonised densely by J7ermtosira on the otiter part of the recfs are slightly higher than the rest. Between the main area of J/ormosira and the ledge is a shallow channel. 6 to 10 inches below the eastern part of the reef, which widens out towards the sandy pool, [xcepi al an extremely low tide there is same water movement along this channel, A less well defined channel passes in from the attside edwe of the rec? between the two main areas of /Tarimosiva, The western side of the reef js fairly even and 6 to 12 inches lower than the eastern half, Psoyudoyjues? eydsowolspug ae SAPIOA BUD nd 12 20}93 Pyudsowouajug swos Shd¥ VOOLD3 148 ny w3adN3Ld35—-uSNONy ‘deur 008 ‘uomsed pue eyslosqo vidwey> 2S HEHUEG GIISCUIOL ems Hawns tuyjeuey TTT wnyedinWw wh}Auda sé J J spy mor 7) j2 jdAal sayem VHOHGOLSAD oA AO, “AEE, wopButuag ‘foes WLI

TT WATSAHEOLSAD epeDOJayey PlauaINE EH erpulssejans Picydoys4> FE eyeyhaued Proydayshs cos Pipswdajul evoydoish> fe Guasaid faye) ysulWcad PM WweEYys aaxiw 4d43ayu NIVA JHL 30 [ ahr spiek * £ T ' ° ee Gt & J1VDS a ec ik = (abpai) Via InAld VYOHAHOLSAD O3xIW ee Ayumwwos Pousiney VXISOWBOH ~Byoydoyss 7 -PusowWuar; {771'S) BPewsawt ANI TIVYOD — ¥YOHdOLSAS LSAD Set (a NOILDSAS SSOUYD 149 The outer edge of the main reel is greatly dissected, with numerous pools and pot-holes (pl. XV, fig. 3). This is the roughest area of the reef and bears the dense and distinctive sublittoral fringe association. The south-eastern corner is more cven, but slightly higher, with a greater degree of exposure between waves. Am association characterised by the articulated corallines, Corallina and Jania, together with specics of Cystaphora, is developed here. Owing to the position of the shallow channel, and slightly higher castern side of the reef, general water movement is across. (he reef to the western side where it streams off the reef. Very little debris is ever cast up at the rear of the main reef, but is deposited on beaches on either side. The suai reef has been described i9 some detail hecause it illustrates well the manner in which slight differences in height of the recf (sometimes only 2 or 3 inches), and the degree of toughness in different parts, control the uccurrence aud distribution of the algal agsociations (pL XAT, fiz. 2). This will be evident in discussing the associations, and also Irom the cross section nf the reef shown in fig. 1. The map shows the distribution of algal associations cn the main reef during Jannary (full map), and daring August- September 1946 (inset). These illustrate the summer and winter appearances of the main reef, but owing to the varying amouuts of sand covering paris of the rear fitioral in winter months, the distribution of rear littoral associations often differs from that shown for August-September 1946. At the western end of Pemington Bay, near the headland of U'recambrian rock, is a ceef Formed of three pr four horizontal lerraces, each 1 in 2 Teer vertically above the next and sloping steeply between eactt (pL XL, fig. 2). The sand rock of this reef is the same as elsewhere in the bay, and there seems no obviots reason for the deyeloyment of such a distinctive and different type of reef. This “western terraced reef” bears seyeral algal-animal communities which are absent or poorly developed on the typical rock platforms, The Mormosira — Ectocarpus / Pylaiclla -anemone community will be referred to later, and a yery prominent Modiolus (mollusc) association is described by Erlmonds. An asso- ciation of stunted Lauwrencia heferocleda oceurs at a low littoral level, The following account of the algal associations of the Pennington Bay region is based largely on their distribution on ihe main recf, Reference is made where necessary to their distribution on other reefs elsewhere in the Pennington Bay region and also in Vivorne Hay, but the main reef has proved to be very typical of this type of reef geusrally. THE Suprarrrronan ZONE Over most of the enast no algue are found in the supralittoral zone, bur i restricted areas tiyo poorly developed algal communities occur, and also a. sparse lichen community. PRaAstora COMMUNITY Prasiola sp, occurs ou rock or stones on sloping parts of the cliffs up to 25 feet above high water, hitt only where well developed penguin tracks pass op these cliffs. One stich locality is about } mile east of the amain reef, The com- munity as not prominent, and of infrequent occurrence. The ales, which is undetermined specifically, forms small preen patches, rarely more than 2 or 3 em. across, of tungled cylindrical flaments. Is habitat is subject to fine blown spray in considerable amotmt, bul it is best regarded as a semi-terrestrial alga rather than a marine one, Prasiola is best developed during the winter months, 150 The other two communities described below are not strictly supralittoral but wecur at a level above the main littoral communities, and are deseribed here as a matter of comyentence. LicHinA COMMUNITY Small black patches of the lichen Lichina confinis (Muell,) Ag. occasionally cecnr in sheltered areas in the fallen rock region of the main recf, and in similar habitats along most of the coast, On the south coast of Kangaroo [sland it rarely forms a distinct association, but ig much more prominent and covers con siderable areas of rock in calmer areas on the north coast. It occtirs at least partly above high tide level, its distribution being controlled by waye splash and shade, TSOLATED ROCK-POOL COMMUNITY On the small cleyared platform at the north-east corner of the main reef (see map) occur several isolated pools. These pools ate about 44 feet above the reel stirface, and are subject to wave splash only under tough conditions. Similar pools situated abave high tide level occur tafrequently in the Peminglon Bay region. Owing to their small size, habitat couditions m the pools are extreme; on hot summer days water temperatures may teach 35° C., falling to 20° C. or less at night; while winter temperatures are often lower than the sea lempera- tire. Salinity conditions are also variable. The flora of these pools is sparse and variable in its oecurrence, hut charac- seristic of the habitat. It ig usually beter develaped during the winter. The jolluwing species ure most Trequent: Pulysiphenia abscissa H. & H., PL fruter Ilarv., P: dasyoides Zan., Laurcncia heteroclada Hary., Eclacurpus confercordes (Roth.) Le Jol. (winter), Ceramiit minigtum Sule. and Cratroceras clogulalam Ag, All are species of wide habitat range, and in many cases they are common in littoral associations. Tus Latroyan Zoxn The associations of this zane utc conveniently divided into those occurring on sloping rock at the rear of the reefs, and those on the flat reel surfaee. Those in the tear littoral are more tolerant of exposure than the reef surface forms, and except at high tide ate exposed between each wave. The distribution of the Associations on the main reef during summer (January) and winter {(August- September) ig shown in the map, The cross section of the reef during Septem- ber (fig. 1} shows the yertical relationships of the associations berween Lhe shore and the outside edge of the rect. Rear littoral associations J, SvMPLOCA HYDNOIDES ASSOCTATION Synploce hydnoides itz., a hlamentous blue-green alga, forms dark, felt- like irregular patches, to 3 ar 4 en. across, on sloping and vertical ruck at the rear of the reefs. At high tide it is washed or splashed by each wave, but may he left exposed for several hours at low tide. The soft spongy mats of the alga retain water strongly. The height to which Syiploca reaches is controlled by the amount of wave splash, varying from 1 (o 4 or 5 feet above the reef surface. it is best developed in shaded areas. Although at times this association is very inconspicuons, 5. ‘yuiploca ftydnoides peeurs on rock bare of other algze. Lt is found in similar habitats on rock platforms elsewhere along the south cvast of Kangaroo Island. On one shaded rock in the iallen-rock region of the main reef a bincenose ol Synploce hydnoides and the barnacle Tetrachita purpurascens occits. 15! 2. GELIMIUM PUSILLUM ASSOCIATION This association at Pennington Bay is but a poor example of the G elidium pusillum (Stackh.) Le Jol. association on calmer coasts, stich as at Rocky Point or in the American River inlet. Conditions at Pennington Bay are mostly too rotigh, and it is restricted to thin patches ou vertical and sloping rock in locally sheltered places. On the main reef the cliff above the ledge forming the sandy pool hag thin scattered patches in shaded hollows; it also oceurs in the fallen rock region, mainly on the sides of the rocks. On exposed rock, Gefidiuim pusilla forms yellowish-brown anich branched thalli, Init when growing in pools the fronds are upright, less branched, ancl 1 to 3 em. high. Like Syinplocd, this is mot a conspicnous association, but it is of general Gecurrence i similar habitats on the sandrock reefs. In sotie areas it Is co- dominant with the serpulid worm Galcolwria cuespitasa in a biocenose. 3. RIvULARIA #IEMA ASSOULATION Rivuaria jira Womersiey forms dark blue-green gelatinous blobs up to 2 cm. across (masses to 4 or 5 ent, across) in parts of the rear littoral where there is constant wave splash. or on ledges ever which waves stream. Such habitats were exposed betweei waves at low tide, but ouly for very short periods. A high degree gi aeration seems to be of ruost importance, “The firm gelatinotis thalli can resist considerable exposnre during the oceasioval vary low tide and hot sunnier weather, Rivularia frie occurs aiong the south aud weal coasts of Kangaron Island, in the upper littoral, whatever the type of rock. The association varies greatly in its development frein place to place, for no obvious reason. At Pennington Bay rocks i the rear littoral may be densely covered with the hemispherical blobs (pl. X, fig. 2, pt. 1), with ne other algue present; in other places, and at different tintes, the rock may be bare. Development is nsually better during winter months, wul may be very poor in simmer. On the main reef, the ledge running in a curve through the centre of the reel to the south-east corner is usually dominated hy Rivwleria firma, This is a region of constantly streaming, broken water, and is colonised by a distinctive group of awac, Apart from Lt. frat, the coummonest are Hydroceleum glutinoswm (Ag) Gomont, Ricwlerin atta Roth, Pelysiphonia frutex (epiphytised by Calothrrx confersieula (Woth.) (Ag.). Hormasire banksit Dene, Léurencta hieteroclada, with scatterer! plants of [rengelia pluinosa Hary,, and Chimpia absolela Harv. The dominance of blue-green algae on this ledge is striking, but similar ledges haye not been found elsewhere in Pennington Bay, though they may occur. On the small “islands” on the ledge (sec map) and in more exposed parts af this association, traleoleria cacspitosa atten becomes co-doninant with the Nioularia. 4. Pcrocarpus CONPERVOIDES AWD PYLAIULDA. WULVESCESS SEASONAL ASSOCIATIONS Ectocarpus confervoides and Lyloiclla (Buchelotia) fuloescens Bornet occur fa the same habilat on slaping rock i the jower rear littoral, or sometimes. in pools in a similar position, at different Unves of the year, Eetocarpus occurs during winter months (April to November), Pylaiella during the summer (Septenibe: to May). Both species grow on sloping, well washed rock which is exposed between waves at low tide (fig, 1; pl. XU, fig. 3 and 4, and map). They reach their greatest size (to 7 or 8 cm, high) however where always covered, with 132 slight wave action. During summer the sloping rock at the rear of the sandy pool (main reef) is ustially buried under sand. preventing Pylaiella front developing there. In appearance both algae are very similar, forming brown tufts, but whereas Ectocarpus is abundantly branched, yioiella fulveseens is branched only at the hase and the tufts are less distinct. Seyiostphon lomentarius (Lyngb.) J. Ag. is frequently found in both asso- ciations, especially in winter whe it is heavily epiphytised by Fefocarpin confervoides. 5. ES TEROMORPITA ACANTHOPHORA ASSOCIATION Enteromorpha acanihaphora Kiitz, forms a striking and pure association in the rear littoral of the Pennington Bay reels, in very much the sate habitat as Ectocarpus and Pylatella. Both associations hawever are remarkably distinct, with tarely any intermixing, Why this is so is not clear, for rock on the same slope and subject to apparently identical conditions may be covered by two quite distinct areas of Eeclocarpus and Lnteromarpha. The association varies con- siderably in its development from time to time, heing best developed in late winter. Srnall scattered plants oceur in the inner channel area from time to time, Goth the Enleromorpha and Fetacarpits—Pylaiella associations occur where loose sancl is carried about by the waves. The algac appear ta be very tolerant to this sand movement, and may even survive short periods of burial in sand, 6. ULVA LACTUCA ASSOCIATION Ulva lactuca L. (£. rigida (Ag.) Je Jol.) often colonises areas slightly lower than Kaleromorpha, and scattered plants occur at the rear of many reefs. The habitat is Jess exposed than that of Luteramarfha. Vhe platits are variable in form, from broad expanded sheets to elongate undulate ribbons, and rarely more than 12 em. high. Littoral associations The associations described below as being in the littoral zone are not strictly so, as most of the algac are rarely exposed for any length of time. At low tide many of the larger species reach the water surface, and others are cevered by only u few inches of water, A few, stich as Hormosira, ate exposed at every low tide. On calmer coasts Cystophara sybfareinata, C, silignosa and atten C. brown are characteristic of the upper sublittoral, but with the exception af C. siliquesa few of the reef algae extend into the sublittoral. Prom most view- points it is convenient to regard the following associations as characteristic of the littoral zone, 7. POLYSIPIONIA FRUTEN \SS8OcIATION Polysiphonia frutex is common throighout the year on the jiedge and caliner areas of the main reef. On other reefs at Pennington Bay, imeluding one imme- diately west of the main reef, it forms a pure association on flat or sloping rock in areas of medium to slight wave action, The alu reaches a height of only 5 to 7 cm., but is dense cnongh to give the area a dark brown appearance, Common epiphytes are Polystphenia abscissa and Calothrix confernicola, 8. CysTorHYLLUW MURICATUM ASSOCIATION The area between the sandy pool and the ledge on the main reef (see map) 3s cavered by a pure assuciation of Cystophyllim nevrteatun (Ttrn,) J. Ag. Other algae are virtually absent from this area. Cystapiyllum also oceurs as seactered plants on rock in the sandy pool and near the fallen rock region. On other reefs it is not common. but usually pure when it does occur. ‘Trans. Roy. Soc. S. Aust., August, 1948 H. b. S. WomeRSLEY MAP of tHE MAIN REEF, PENNINGTON BAY CYSTOPHYLLUM —~-| CYSTOPHORA- ASSOCIATION CORALLINE ASSOCIATION 1X PYLAIELLA MIXED CYSTOPHORA ASSOCIATIONS ASSOCIATION ttttttt A le thd aoe ENTEROMORPHA HORMOSIRA oe 4 Oe i + + - “| ]6¥STOPHORA cae bt ee aaemeenennraaet : i-4% . ASSOCIATION ASSOCIATION DEEP OUTER POOLS UW Ss . i . UVIFERA oo oo so oo ee so eo oo o°o ° eoooomoe eo © 4PPROX LOW TIDE LEVEL YARDS SANDY BEACH SUMMER ASPECT APPROX, HIGH TIDE LEVEL WINTER ASPECT JANUARY AUGUST-SEPTEMBER 153 The alga reaches a height of 20 cm., developing best at the cdges of slightly deeper pools, Sphacelarie biradiala Asken. and Sphacelaria furcigcra Kuta. are common epiphytes on the lower stems during summer. Occasionally {he plants are alniost buried in sand, but appear to suffer litile damiage. The assuciation 15 always, but only just, submerged, The rocky bottom in the north-eastern half of the sandy pool is rarely covered by sand, and beavs a very mixed assemblage of specics. At low tide it is coveted by 1-3 feet of water, with fairly calm conditions. Cystophyliem wiuricatium is present here, though not well developed and in varying amounts throughout the year. Other species may be prominent at different times, developing rapidly and disappearing after two or three months, Such species are Liagora harveyaia Zeh., Melinthory tnmens J. Ag.? Champia obsoleta (winter), Dictyopterts acrostichoides (J, Ag.) Sehinite? Eetocarpus confervoides and Scytasiphon tomen- turiuy (both winter), Pyfeiella fulvescens (summer), Muellerena sp, (winter), with Lira lacluea and Enteromor pha acanthopltora in the shallower parts, This mixed group of species is also found on the eastern part of the main reef, in shatlower water near the fallen rock region, but is less well developed here, Most of the species are found in other associations on the reef, 9, JANIA FASTIGIATA ASSOCIATION Or several reefs in the Pennington Bay cezion (one immediately east of the main reer) and also on reefs in Vivonne Bay, Jania fastigiata Wary. colonises otherwise bare rock in the littoral zone. The association is nof a dense one, but distinctive in appearance, “he fronds (to 4 or 3 cm, high) are bleached white in swmnmer, but in many cases are heavily epiphytised by Coloflvix confercicolir, viving the association a hlackish-green appearance. The habitat ts usually of fairly ever rock, Just exposed at very low tides. 10, J.AURENCIA HPTEROCLADA ASSUCIATION Laurencio heteroclada occurs as a minor component of several reef associa- tions, and is widely distributed along the south coast of Kangaroo Island, On the lower level of the western terraced reef (pl. XL, fig. 2), and sinilar sloping roek in the low littoral, it forms a communiry whieh is doubtfully of association rank. The alga grows as a dense burl, up to 7 em. high, yellowish-brown in colour, of etirmed plants where it is subject to fairly heavy wave action, This habitat ts rately exposed, being ently slightiy higher thaw the sublitloral frmge, A stunted Sargassum, Cladosteplius verticillatus (Lightt,) Ags Cuslerpa brewmit Fadl... cu. simplicinsenla Ag. and Cheelomorpha dercwinkt Kitz. often oceur with it. ll. Vw Cysrorrora comPLex ‘The greater part of most of the Teunington Bay reefs is colonised hy species of Cystophora. Their extent on ihe main reet is shown by the nip. The mixe:i Cyslaphora association covers the area east of the ledge, consisting of numerous shallow pools retaining 4 to 10 itches of water at low fide, and also the channel area and must of the western part of the rect. Four species of Cystophora, C. uaifera (Ag) J. Ag. C subfarcinila (Mert. } J, Ag, © siliquosa J. Ag. and to a lesser extent CL brownit (Turn.) J. Ag.. tngether with Surgassum amurienatun J. Ag. form a complex of associations. Depending on small variations in depth of water and calmness, cach may form att almost pure association, or, as is more common, a nvxed association of almos! any cumbitation of two or more species (pl N, fig. 1, pt. Ly. The shallow pool area enst of the ledge on the mam reef is dominated by Cystophora sub. forcinali, C. siliquesa and C. uvifera, svith some Sargassont mariciulatwn. This is refereed to as a “mixed Castophora association”, Ags pire associations, 154 C. subfarctnata and C. siliguasa are most Frequent, The importance of Cyslophora in the algal ecology of Kangaroo Islatid is in many ways analagous to the importance of Enealypts in South Australian land ecology, Ti the previous paper it was shown that the algal formations anil sub- furmations around Kangaroo Island can be characterised by the presence or absence of species of Cyslophora. OF some 25 species in the genus, 17 veeur wround Kangaroo Islatid, (a) Cystophora wifera forms a pire association on flatter and calmer parts of Wie reefs. On the main reef it dominates an areca af more even rock running ont- wards towards the sotth-east corner, and also parts near the western edge (pl. XIV, fig, 3). Tt is densest where just covered at low tide, partly due to insufficient depth of water at low tide to allow other (and larger) spevies oi Cystophora to devetop. C. wiifera forms short, rather stunted plants, to 25 em. high, with one to several stems arising Irom the base. It shows remarkable seasonal variation in vesicle production, which seems to be atiributable only to seasonal variation in sea temperature (a range of about 55°C. on the reef), Cy awiifera has always heen deseribedl as bearing spherical metic vesicles on the main stem (Hatvev Phyc, Aus. pl 175). This is the form found in calmer waters or rarely cast wh from the sublittoral, Cir the Pennington Bay reefs. vesicles have never been found on the plants during January. My May, small clongate vesicles with a strongly develuped mucra occur in small numbers on most plants. These approach closely the vesicles of C. cephalornithos, which is only found in much calnier habitats. In September vesicles are wumerous, the older ones eing almost Spherical and with only @ stall mucro. The “typical” mutic vesicles are rarely seen, and hy mid-summer the yesicles are lost. Lt seems evident that besides the controlling effect of temperature in vesicle formation, the relatively rough conditions on the Pennington Bay recfs cause ihe juvenile forms of vesicle (elongated aud mucronate) to be largely retained. Sargassum miurieaiine 15 commionly. present in the gaine area as CL nvifera Sphacelavia furcigura, ant occasionally Sph, biradiata, grow epiplytienlly on the stems. (bh) Coslophora subfarcinata— Tins alga grows i slightly deeper water than C. ueifera, where wave action is greater but not extreme. Apart from Horinasira hanksit, Wis the most widespread species on the reefs. On the main reef it ts ime of the dominants of the ntixed Cyslophora association, and is particularly prominent iu pools more than 6 inches deep and in the channel area, where it is always in G to 12 inches of water (pl. XUV, fig. 3). Cystophora subfarcilata never produces vesicles on the Penningion Thay reeds, or elsewhere om the south coast, In sheltered parts of the north coast, however, vesicle lormation is commou. This is a clear case of ecological Forms uncer different degrees of wave action. (c) Cystophora siliquosa inhabits deeper pools than C. sucbfaremata. It forms a characttrisGe fringe around most of the deeper rock pools towards the outside of the reel, and also on the western side of the veel where water streams off. It cannot surviye strong direct wave action. Small plants occur over most of the mixed Cyslophora association on the mai rect, hut only in deeper rock pools dacs it attain WS anaxinnun length of 5 or 6 feet. Near the western edge of the reef, Antithamnion hanawieides (Sond) De Tam, Antithanniow sp. and Dasya spp. are common epiphytes on C. siliquosa. On other reefs, and rock off the edges of reefs, C. silignosn forms a pure association in 2 to 5 or more feet of water. 135 (ad) Cystophora browuit—This species is less common than the previous three, and does not oceur on the main reef. On recs east of Pennington Ray it ts sometimes codominant with C. subfercinata in 1-2 {vet of water (at low tide) towards the rear. Only rarety has it been seen pure. (e) Safgassuur murtculatum—This, the only nportant littoral species of Sar- gassnit, 8 common an most af the rock platforms, and shows remarkable scasonal development. On the main reef it occurs near the fallen tock region, on flatter areas with C. wafer, and in the channel, but iy detisest ow the western side oF the rect where it thrives in (he constantly streaming water, In the lalter area itis often dominant, bat elsewhere is usunily subdominant to the Cystophora. Doring suminer Sargassum ohuriculatum ig & short etunted plant rarely more than 10 or 12 em. high, and it can be easily distinguished (rom Cystoplora unifern only by is fattened stem. Growth commences during March or early April, and during wiriter the fertile fronds are developed, reaching a licight of 40 or 45 cm, By September small vesicles, almost identical with those of S. sowdert J. Ag., are produced, but by varly November the fruiting frond disintegrates, losing both receptacies atid vesicles. The stalk or the ok frond persists for some weeks, but by January ihe plants are redueed to a short stem bearing a few branches with crowded, almost terete leaves (such as figured by Agardh [1889] ). This sinking seasonal development must be due to lower water temperatures, though the annual range, as given previously, is small. Throughout the Cystophora associations scattered plants of Hormosira hauksii may occur chiefly on higher ridges (pl. X, fig. 1, pr. 1). Tiny dark blue-green thalli of Aaniarie atra are alsa common on bare rock in the associa- tio, An intermediate conunnnily helweer the mixed Cyslophora and Hormosira associations on the main reef will be discussed later. Several other algae ocetir irregularly in the Cystophera complex, varying greatly in their distribution and occurrence in different seasons. These will be dealt with later under “Chance distribution of minor species” (see p. 157). Although the species of Cystophera ate the largest algae grauwing on the reefs, much of the roek between the plants is feft bare, This is probably due im some ncasure ta temoval of young plants of other species by the fronds of the Cystophora as the latter are moved over the ro¢ék by waves. On one uccasiofi (April, 1947) the channel area had beet almost denuded of algae, otly the holdiasts remaining. Growth of new plants ia this area was poor, and it would seem that few species apare from those of Cystophora grow readily in arcas of shallow streaming water, Vet such a halter appears to be jtist as suitable as many where algal growth is heavy, stich as the sublithoral (ringe assuciation. When compared with the latter association the Cystephora: complex is poor in humber of species, though composed of lacger planis. 12. TlokMOsIRA RAW KSTI ApsocnATIOW. Formasiva hanksit (¥, siebert [Bory} Dene.) is more widespread than any other alga on the Pennington Lay reefs. It oceurs from the outer edge to the inner calmer areas, but only on higher parts of the reefs in fairly rough condi- tions does it form a dense and distinet association. On two areas on the outside of the charinel on the main reef, and a smaller area just east of the ledge near the Cysiophora-corallime association (sve map), it becomes very dense, to the exclusion of all other algae except a few epiphytes (pl, NLL, fig. 2 and 3). The areas of pure Hormosira are only a few inches above the rest of the teef, but are exposed hetween waves at low tide. sometimes for several minutes. On many reefs the Hormosira association is near the outer cdge or sides, Many reefs have the surfaces markedly ridged, and hear Sermosira only on the ridges where il is exposed at low tide (pl. X11L, fig. 1)- 154 The chains of bead-like vesicles (about 1 cm. across) of Horinosira provide am abundant reservoir of water for the alga to tolerate exposure and desiccation. No other alga has a similar structure, so it is not surprising that Moruresira alone can colonise these higher areas. The abundance of the plants in the pure areas, aud their large size compared with the same species elsewhere on the reef, would indicate however that some degree of exposure is beneficial to the growth of Hormosira. On the rare occasions of a hot north wind and very low tide, leaving the reef exposed for some hours, Horwrosira on higher areas may suffer consi:lerable damage, becomimg black and withered. Noatheia anoinala is a very common parasite, growing from the conceptacles. Other epiphyres are the black tuits of Polysiphonia nigrita Sender, and H qydia~ coleum (wagbvacewm Kitte. which forms small dark blue-green furry growths in the constrictions between the vesicles. The secend level of the western terraced teef (pl. XI, fig. 2) comprises numerous pools, 3 to 6 fect across and up to 1 foot deep. Waves enter the pools at medium aud high tides. Hoermostra banksti forms a dense fringe around the edges, and well-deyeloped plants occur on the betiom, ustiaily reaching to the sur- face (pl. XID fig. 4). Also on the bottom of the pools occur brown tutts of the filamentous species Polysiphoniua abscissa, Ectoctrpus confermoldes (witlter} and Pylaiela fulvesceus (sunmner), The rec anemone Aetinia tenebrosa is prominent in shaded hollows around the edges of the poois, just below and for a few inches above the water line. The development of this Hormosira-Actinia biocenose is due exstuitially to the specialised habitat—shallow pools expased at low tides and subject to wave influx at higher tides. Prom the algal viewpoint it is best considered ag a variant of the typical TJormosira association and has been seen only on the western terraced reef. 13. CysTOPIORA —CORALLINE ASSCICLATION The south-east earner of the main reef receives the full foree of breakers, though on the eastern side waves tend to surge ayer and along the wlye, This corner is shghtly higher than most of the reef, and bears @ closed and deuse asso- ciation (pl. XV, fig. 1). On higher, but rough and ennstantly wave-swept parts of other reefs the sane association is promiiteiit. Dominant are the dark brown, densely branched fronds of Cystophora paniculata (Tiirn.) J. Ag, often stunted lighter-brown fronds of C. subfarcinata. and two species of the articulated coralline algae, Small tufts and mats of Covallina cuvicri Lamx, cover most of the reck between the Jarger algae, while Jania fashigiata Hat, is heavily epiphytic on.C, subfarcinala (pl. XV. hig. Land 2). During winter Corailina cuveri (1. crispdta Lamx.) is a pale pmk, while Jue ts a brighter red, In sunnner both are bleached to a Taine pink or white, but on ail occasions the contrast of the brown fronds af Cystophere and the pink er wlnre of the corallines gives the association a distinctive appearance (pl. NV, fig. f and 2). Cystophora paniculata is well developed in the rough areas but absent from the rest of the reef. Though usually abundant, on oceasion it hus been almost absent (September 1946). On some reels it is tare and the association 1s dominated by the corallines. The degree of epipliytisin i this association is yery high. Most species can grow on either rock or the Cystophorw, Vhe rough warted stem of C. paniculata provides an excellent substrate for many species. Some epiphytes are seasonal in their aceurrence, 187 Species present Uiroughont the year include: Dassepsis clavigera Womers- ley, Polystphonia nigrita, P. dasyoides, Heterasiphonia gunniana (Maty.) Falk., Chavlomorpha darwin, Caulerpa brorwnti, C. simpliciuscula, Laurencie, hetera- cada and Autithomnion hanowioides. Less common ones are Zonaria sub- articuluta (Laimx.) Papenfuss, Parhydictvom paricuddtion Jo Ag. Sargassum hractcolasnin J. Ag, Si smericialdalumn, Siphophora chondrophylla (8, Rr.) ary, (in pure but small palebes). odd plants of Hormosira banksii, Ceranium nobile Jo Ag. C. miniatium, Mrenyelia plumase, Melugoniolithun charvides (Lamx.) W.v. B. Melebesia sp. and Calothiria confervicala, MVonospera clonyala ( Harv.) De Toni and Griffitisia monilis Hary, ave common epiphytes, forming bright red tults, but they only ocenr during the winter (April to November). Thuretia teres Hary. and Mychodea folioxa (Hurv.) J. Ag. are found mainly on the stenis of C. paniculata. INTERMEDIATE COMMUNITIES Mitermediate and gracing communities between the associations discussed above are not frequent, but as is expected on such reels variations in height and exposure belween different associations will result in 2 mixing of species. In addition, ather species may he suited by the intermediate habitats yel no: occur in either of the main associations. On the main reef, rock to the east and sonth-cast of the mixed Cystophora association bears 4 community dominated by Horniosira benksii, Cystophora wvifera and Lavrenciea heirroclada (which varies greatly in abundance during the year). Sargasswn mariculitwt also occurs. The change between this com- munity and the Cys/upiera—coralliue association takes place over 2 or 3 fect. it is essentially an interinediate community between the mixed Cystophera and fHormoxira associations, under conditions which ace suitable [or Lanrenciu heteroviaila. Near the eastern edge of the reef Hormosira hecomes dominant, The mixed ar variable assemblage of species found im shallow areas near the Fallen reek region and on vock in the sandy pool is described on page 153. The Hormosia—ancnume pool variant of ihe Hormesira association has been dealt with on p. 156. Letocarpus confervoides and Pyluielia fulvescens, which are subdominant in this community, dominate seasonal associatiots in the rear littoral, while Polysiphonia abscissa is ulso characteristic of more isolated rock pools (see p. 150). Coance Distrmcrion or Minor Species on tite Main Krer A niinber of algae ocetir on the main reef, but do not form a stable com- panent of any particular associagion, At diferent times during the year they may occtr on widely separated areas of the reef, usually in relatively small patches within Jarger fairly uniform halntats. Chance establishment of the species under temporarily suitable conditions, rapid development of individual plants and rela- tively small spread from the original area, followed within.a few months by death under changed conditions, seems to account for the observed distribution, The following species bave attracted attention. Liagora harpeyana grew near the fallen rock region and the eastern edge of the reef in January 1946. In January 1948 if was prominent on rock in the sandy paol, while in November 1947 it was found only on the outer western part of the channel area, On other occasions it hus been found confined to the ledge. Felininthora tumens €?) 1s best developed in the late winter, occurring on inner parts of the reci. In September 19H it wae partly buried by sand in the rear littoral, where it was exposed between waves. At orber times it has heen confined 158 to the sandy pool (January 1948) and a small area on the western side of the chatinel (November 1947). Plants exposed and buffeted by waves are consider ably stouter than those always covered. Chanipia obsoleta, during September 1946, was so prominent in shallow inner areas of the reef that it conld well he considered to dominate a community, -At other times it has been confined to the ledge or eastern edge of the reef, but often is vittually absent from the recf surface. Cladosiphon fila (Warv.) Kylin is often absent from the reei, bul a tow scattered plants may ocetir near the fallen rock region, and in November 1947 it was qilite dense over a few square yards of the mixed Cystephora association in the channel. Cladosiphon verinicuaris (J, Ag.) Kylin bas been found in outer pools on the main reef (November 1947), but is qitite common in a cove west of thy inain rect. Halapteris pseudospicata Sauy. and Phlveacaulon spectabile Reinke occur in pools in 2 or 3 feet of water, and occasionally on rock in the sauly pool. Cladostephus verticillatus is only rarely found in pools on the main reef, but on the sides of rock in the lower littoral and upper sublittoral in other parts of Pennington Bay it may be quite Gense. Cladophora valunioides Sonder may occur as scattered plants almost any- where on the reef, but usually only in a restricted area at any one time. Priors of Suavep, Rear Lrrrorat Poots A number of reefs in the castern part of the Pennington Bay region have small caves, up to & or 10 yards im length, at the base of the cliffs. At high tide waves ustially enter them, and pools often oceur at the entrance or just inside. Such pools ate shaded, relatively calm, and contain an assemblage of algae of sub- littoral affinities. The habitat is similar to sublittoral conditions in light and degree of roughness. Algae most frequently found in these pools are Apjohnia loetevirens Vary (stunted), Rhipiliopsis pellata (J. Ag.) A. & B.S. Gepp, Ecklonia radiata (Turn.) J. Ag. (Cin larger pools), Plocamiunt aigustiem (J. Ag.) H. & II, Phacelocarpus labillardieri ). Ag., Corallina cuvieri, Laurencia clata (Ag.) Harv, Haloplegma preissit Sonder, Spyridia oppasita Hary., Ballia scoparia Hary. In oue stall pool less than 2 feet long and 1 foot wide, under an overbanying rock in the fallen region of the main reef, the following small Clilorophyceae were tound: Jwhipiliopsis peltata (abundant), Hryepsis baculifera J. Ag.. Derbesia clavacforimis (J. Ag.) De Toni, Cawerpa sp, and aycheriy sp. These are all rave species at Pennington Bay. Other sublittoral species probably occur in such pools on other parts of the south coast of Kangaroo Island. Tre Superrrroran Friwce Zoxe (CYSTOVMMORA INTERMEDIA ASSOCIATION The sub-littoral fringe on the Pennington Bay reefs corresponds to the outer at side cdges and a foot or so below (pl. XTV, fig. 4, and pl. XV, fig. 3). The outer edge of some reels is probably a little above nican low waler mark, and on other reefs slightly below, but the conditions at the reef edge are uniform as even at low tide waves are constantly breaking on it, leaving it almost or just exposed for a few seconds betweeti waves, ‘Lhe outer edge is the roughest habitat on the reefs; the sides are less rough, but both are habitats of high aeration of the water, short but frequettt exposure between breakers, and heavy forces from wave action. 155 The sublittoral fringe bears the densest algal association found anywhere on the reefs, The rock is usually completely covered, and epiphytic growth is profuse. In wumber of species the assuciaticm is very rich, On au area of 4 or 5 square yards at the outside of the inain reef over 50 species have been collected + riast are small in size and often stunted, onlv the Tueales being of any bulk. The chiet requisite of an alga in this association is a strongly developed holdtast. The masses of Lromds, however, afford considerable protection for each plant, as wave forces can be exerted only from above and not from underneatl the plants, The association is dominated hy Cystophore intermedia J. Ag. The dark brown pinately branched fronds of this alga reach a length of 50 em., and give it characteristic appearance to the reef edge (pl XLV, fig 4, and pl XV, fig. 4), C_ gatermedia is remarkable for its inability to grow anywhere except in the roughest, well aerated places, No satisfactory explanauion of this is available at present, Although iis stems ace omy 3 or 4 mun. in diameter, they are extremely sltong. A very heavy pull is needed to remove a well-developed plant, and only m extremely rare occasions has this alen been fotd cast up on the beach. The algae of the sub-ittoral fringe Delong mainly to two types. The majority have elongate, often nich divided fronds which offer minimum Tesist- wee to waves; others Torm mi(-like masses on the crock which are also protected by larger bushier forma, Where waves pass along the reef edge (eastern side of iain reef ), ©. inter- inedia grows densely from the edge to 2 feet helow. In situations where there 15 heavy wave splash, caused hy projecting parts of the reef, it will vrow on the surluce; this is particularly so on the outer highly dissected part of the main reef and on the large rocks off the easter edge. However, where water streams off the western side of the reef, and breakers are of reduced foree, €. siliguosy replaces C, inferiirdia in the sublittoral {tinge to a large extent, Cystophora intermedia is often leavily epiphytised by other algae, whieh stow only front the conceptacles. The stem is too smooth and mueoid ly provide a hold for algal spores. Daring sum the small brown blobs of Corynaplilec eystophorae |, Ag. ave common on the upper fronds, while a species of Davw and Grovwania muelleri Lary, wceur in winter. Other larger alpae of the sublitroral fringe are Saryassune bractevlossi (with lerechaetivm sp. epiphylic on the leaves) (pl. XV, fig. 4) and oveasional pluits of Keklonia vadiaie, Seyvtothalia goryearpa (Turn) Grev.. Cystaphore sportioidcs (Turn) J. Ag, and i, penienlata, C. subfarcinata and ST iganisee do tit orcur below the reef edie except i calmer places. The following are the must imperlant of the smaller eperies, Nearky all are stunted in size wins to the rovgh Habitat, much lareer Lorie beitige cash my) from deeper water, Caulerpa browmi, C. sedoidcs (RB. Br.) Ag, CC stuphiciuseylas Codina pomoides qT, Ag. Fachydietyon Mematciediet uit, Lobos pita bicuspidata wlresch., Dictyopteris acrostichoides (?), Cyinaserus parieqains (lamx) J. Ag; Liagera harveyanu, Gelidium australe J, Ag., Meiayoniolithan charaidis, Corullina cuvicri, Mychodea foliose, Gigartinn sp. lridaca prolifera (J. Ag.) De Toni (rare), Chainpia obsolela, Hymenocladia polymorpha (1 Jarv.) J. Ag, L. consbersa (Harv) J. Ag. Guvenile state only), Memastone furedayae Hary... Phytime- phora imbricata, J. Ag., Sarconcnia dasyoides Darv., Ballja Scapuria, Calli- hamanion laricinuin Uary., Griffithsta avnilereticn UW. & TY. Lalapleyina preissii, HWerangelia plumosa, Chondria sp Laurencta heteroclade, L. robusta? L. elute =) M.S. name for an aPharently wn¢deseribed species_ 160 (with epiphytic Janczewskia tusmantca Falk.), Jeannercttia lobata H. & 4H. Dasyopsis clucigera, ITeterasiphonia gunaiana, Thuretia teres, In ardition the jallowing species are usually epiphytic on larger fucoids: Ceraminnt nobile, Anti- thamnion hanowloides, Crouania Muelleri, Monospora clongata and Griffithsia neonilis (both only in winter months), Mnellerena insignis (Harv.), De Toui. Polysiphonia nigrita and P. dasyoides, The large rocks off the eastern edge of the main reef (see map, and pl. UX, fg. 2, pt. L) provide a_ habitat particularly favourable to Laurencia, L. hetcrocluda, L. elata and &, robusta occurring in abundance. Al ntithanmyont hanowioides is often densely epiphytic on all three species. The bright green oi Coulerpa broqwuii and GC. siarplicinseula jroyides a striking coutract wmongst the deep red of the other species. Tur SepuirroraL ZONE ZONATION BELOW TILE SURLITTORAL FRINGE Study of zonation in the sublittoral is limited to 3 or 4 feet down the vertical sides of some reefs; ¢ven this can be observed only on very caln days with low tides. The following is, very broadly, the zonation down the eastern edge of the miain reef, To 2 feet below edge - - Cyslaphora tnterimedta From 14-24 feet below edge - IWrangelia clavigera Harv., Gelidiun australe and Seytothalia dorycarpa Irom 2-3 fect below edge - - Perithalia inermis (R. Br.) J. Ag., and usually below this Plocamun costa- tum (J. Ag.) H. & H. and Phacelo- carpus labillardieri Luureicia clafa may be prominent at about this level FLORA OF DEEP OUTER LOOLS ON THE RVEFS Seattered over most of the Pennington Bay reefs, particularly on the outer parts (often within the siblittoral fringe—pl XV, lig. 3), are rock pools up to 6 feet or more deep, with vertical or steeply sloping sides. The Mora of these pools is always submerged aml musi therefore be considered subiittoral, Light intensity is lower in the pools than on the reef surface, especinily Lor smaller species which are usually shaded by large fronds of Cystophora growing aronud the edge, Conditions within the pools are calmer than on the reef surface, Around the edge Cystaphora siliquosa is usually dominant. Were the fronds remain well submerged, and wich ample roont for their development viten reach a length of 2 metres. Sarqussune bracieolosum, and tu a lesser extent Scvlolfalta darycarpa and eklonia radiata, are common near the edge, Apart from the [ringing edge of Cystophora, the conspicuous feature of these pools is the bright green masses of Cuulerpa which cover the sides in patches up to a foot or more across, Carlerpa brownii, C. obcura Sond., C. Ayproides CR. Br.) Ag. var. mvelleri (Sond.) W. y¥. Bosse, C. longifolia Ag, C. siiplicius- culo, C, sedoides and sometimes C. scalpelliformis (R. Br.) Ag, are fonud, best deyeloped where shaded by other algae or the pool edge. The characieristic vegetative growth by means of surculi results in fairly pure patches of one species, but much of the rock is left bare. Other species irregularly distributed in these pools are Rhtpiliopsis peltata, Dirtyasphaeria sericca Harv., Apjoknia laetorvirens (often the basal part only), Halopteris pseudospicata, Phlorocarlan spectabile, Cladosrplion sermticiularis, Dictyopleris acrostichoides, Dilophus sp. an Gyrinosorus varicgatus. 162 THE DEEPER SUBLITTORAT. FLORA The following fist of species includes those found cast up but known not to grow on the reef surface. Other species in the list do grow in the littoral or sublitioral fringe zones, but are usualiy much larger when growing in the deeper sublittoral. All sublittoral species found in the Pennington Bay region are nat listed here, but a complete census and notes on the species known from Kan- garoo Island will he published as a later paper. The cast up species found at any one time depends greatly om the weather over the previous few days, and no rehable data as to the absence of a sublitcoral species at any time can be obtained, CuLoropiycese: Codtwn galeatim J. Ag. (conmnon), C. manillosuim Harv., C. spongiosum Wary, C. pomoides, Caulerpa harveyi W. vy. GB. Cun abseura, Can. Ayproides, Cau, ethelac W. vy. B. Apjohnia laectevirens, Clado- phora valonivides. PGLAKOPITYCENR: Girardyva sp. (on Posidania australis); Phlocacaulon spectalile, Faloptrris pseudospictta Sauv.; Dirtyota latifolia J, Ag., D. radicans Harv., Dictyopteris mucllert (Sead.) Schmivz, Dilophus fastigialus (Sand,) J. Ag, Zonaria subarticulota (syn, 2. turneriqna J. Ap), 2, crenula |, Ag, Chlenidophora microphylla (MWarw.) J. Ag. IToitucastiicius Suposns (NM Tr.) J. Ag, YW. spiralis J. Ag., Lobospira bteuspideta, Sporechuus scoparins Harv, S, comesns Ag. Encyotialia cliftoni Ttarv., Bellativ eriophorum Lary. Palyecriu sastericola (Flarv.) WKylin Cf), Merete anstralis Harv., Myriodesina calophylum J. Ag, M. integrifalia Harv. Sieracocens axtilaris Grev,, Carpo- glossum. conflucns (BR. Dry Kitz, Seaberia agerdiit Grev. (very common), Cyslophora raceimosa Uarv., C. peetinua (Grey, and Ag.) J, Ag., C. platvlobiume (Mert.) Jo Ae, ©. rofarta (Mert) j, Ag. C. greviller (Ag.) J. Ag, CL dumesa J. Ag, C. monijiferg J. Ag. Sutgassiie bracteolosion, S. cristatua J. Ag., SN. trichophyllunt J. Ag. S. decipiens (R. Br.) J. Ag., S. sendert J. Ag., S. varions Sond. RuupepHycear, Common: slsparagopsis orimuta Harv., Delisea hypreotdes Harv., Callophydits lambertit H. & H., Areschougia lowrencié Bary. Methrocloniiun mueller? Sond., Hypiea episcopaliy H. ard H,, Plocaminm costalum, P. nidificun (Hary.) J. Ag., P. mertensit (Grev.) Marv., P. preissianum Sond., Phicclo- carpus labillardicr’, Mychoded ronipressa Hary., M. carnésa WW. and H., Giviesac clay brown? Varv., Hynenecladia polvnorphe, Asmstithamnian aucroralun (f. Ag.) De Toni, Bulla scoparia, B. callitricha J. Ag, Coramium puberulin Sond, Masyphila preissii Sond. Luplilota articulata (J. Ag.) Selnntiz, Lasio- tralia farniosa {Uarv.) De Toni, Afeuaspora elongata, Muellerena insignis, Spougoclonium sp. Spwidie oppasita, Surcomenta dasyonles, Polysiphkaniqa mallar- die Trarv., RBrangaiavtolic custeatiy (Xe) Sehmitz., Dexedusya bulbachacte (Harv.) Falk., ebaensia pinnalifida Llarv., Aneuria fatifolie (Harv.) J. Ag. Osmnudaria pralifera Lams, Posy naccarioides Hary., Thuretia quereifolia Dene, Qeessowan: Gualsonia wuawata Uatv. crare), Delixea faudehra Mont. Brachryclodia marginata (Sol) Selinitz, Calloplwltiy coccinca Harv,, Gelinaria uleniden Sond., Thammnocloniint claviferion J. Ag., Pevssonnelia australis Sond. Metiayoniolithaw stelligekum (Lamx.) W. vy, By Coralinw pilifera ‘ants., clreschougia gruciiivioides (?), Thysavecladia laxa Sond, Erythractoniun wigustatun Sonc., Gleiophyilis burkevlee (Hary.) J, Ag. Rhodophyllis tewwifolia ((ary.) J, Ag, Plocommm anguston, Po leptaphylina BKiite., Phacelocarpus sessilis Harv., Stenaeludia harvesena J. Ag. slerelylus australis J. Ag., Giyartine disticha Sond., (ridaca australisica J. Ag. Champia affnis (A. and H.) J, Ag, 162 C. tasmanica Varv., Cloioderma tasiuiica Zan., Rhodymenta aysiralis (Soud.) Harv., Mymenveladia usnea J. Ag. Autithamnion dispar (Harv.) J. Ag. (rifithsia unlurctica, Corman Fsogo nits Hary., Crauania westita Mary., Ptile- cladia pulchra Sond., Spoitgocloniitm broynianum (Harv.) J. Ag. Wrangelio crassa HE, and H., HY’, nuyriophvlloides Hary., WW. princeps Harv., Sarcomenia delessertuidts Sond., Hypoglosszm mieradantum J. Ag. (PI, Claudia elegans LLamx., Polysiphonia datvee Renmb,, Cladurus clatus (Sond.) Falk., Coelocloniuin opuntivides (Harv.) J. Ag., Laurencia filifarnus 1 Mont. Jeannerettia lobeta, Protokiiisingia anstralesica (Mont.) Falk, Artesia kittcingioides Mary., Leno mandia marginata WL, and TH, L, mrelleri Suad., £. smithive (A, and F.) Falk, Trigenta yinbellata J. Ag, Halodic ayer rebustim Elary., Dasva villosa Harv, MH. ha five Harv, Heterosiphonia mtucitert (Sond.) De Toni. SHASUNATL VARTATION IN THE ALGAL FILORA (a) Seasonal ecculrence The great majority of the algae growing on the Pennington Bay rock plat- forms are present. thrpaghout the year. The larger brown algae, nearly all qnentbers ef the Fueales. are the most stabie species, and probably live tor several vears. The Hetocarpus confervoides and Pylaicila filvescens seasonal associations are the only ones limited in their oecnrrence tu definite periods of the year. Certain characteristic species within other associations, particularly the Cystophora—coralline and sublictoral fringe associations, are also scasoial. In both these asaciations Afanoparu clangiuts “and Crifjithsia moiiilis are found onty during whiter months, although the fornier ig accasionalls y east up frum the stb- littoral in stimmer. Neonastomu fercdwyce is strictly a summer species. A uuinber of epiphytes are alsa of seasonal uscurrence. Coryntophloce evstaphorac is epiphytic on the recyptuctes of C. ywiophora iuternedia and C, aTignuse during the summer, while a species of Dasve att Crouanio annelicri epiphytise the same species of Cyslaphora mamly durag winter months, Species dealt with under “Clauce distribution of minor species” are very irreatilar ip their oceurrence anil apperent!y ned restricted to any oe period. Information on seascnial ceeurrence of subtittoral species, derived from casr up plans, is aul reliable, bit definite coilectae records will be give in the census in a later paper, (b) Seasonal dewulopurchl of sluble species Many speeics on the reels aitain their maxinium development iz the late winter. This is probably associated with lower sea temperatires, The svasonal development of vesicles in Cystaphara weifera and of the fertile frands of Sargassum murteulatuin has been described on pages 134 and 155. Grher species of Sargossum cast ap from the sublittoral are best cleveluped in date winter when the fertile fronds are borne. In this state the plirts are imost easily torn off and cast up. The Cystophera—coralline associatien is better developed ia winter when the corallines are actively growing. Rivlaria firina and Enteromorpha aeailha- fhora attain their maximum size in late winter. (c) Seasonal variation in reproduction Many species have heen found bearing reproductive organs on every visit Lo Pennington Bay (at approximately 2-nonthly intervals during 1947), and it is probable that most species are not strictly limited to any one period of the year. 1d Whether aay monthly eyele in reprodnetion carrelited with the tidal cycle aceurs is not known, but is unlikely. as monthly udal variation on the south coast of Jcangaroo Island is small and its effect saininiused by constant breakers. During late winter (September) ucarly every specius an the reef bears vepro- ductive argans, usually abundantly, and this time seems to be the most favour- able period for reproduction, In September 1S46.it was difficult to find a speci- men of most af the small Khodophycvas—such as Champia obsoleta, Aluellerenn sp.. Monospora elongata and Liagora harecyena—not dearing reproductive cells, hut at other Gines (e.g., mid-stummer} most specimens are sterile. Oiten 80% ar 90% of the plauts of dipiobiontic Rhodophycere are tetrasporic, while very few plants are spenmatiigrel, Liberation of zoospores and gametes in Eaferoumerphe and Ula is probably dependant on suficieniiy fone exposure during very low tiles, Most of the tucoids are fovhle througheat the year, but the species of Saryassuint produce tertile fronds i winter and lose then in early stunimer. "VARIATION UNDER WAVE ACTION Stunting of algae under severe wave action ig a geueral feature of species which grow both in the sublittoral proper and the sulitloral fringe ur Cystophora—coralline associations. The roughest situations always show the densest algal grawth, hut reduction in the degree of branching Im some (eu, Hymenoclaia polyinerpita, Dasyapsis clavacra), and iicrease i branching frou the base givin compact tafted forms in ethers (er, Cladaphara walontaides ) are the most gureral qadaptions to withstand the wmechinien’ foree of the breakers. Custaphors intermedia, which grows best in rhe romuhest localities anc not on calmer coasts, is one notable exception to stunting under rough conditions. In Many Cnsts if ig eviden) that sige and pross external form are of litle use as specific characters. Normaily pinnate or well-bratiched forms may become simpie or nearly so on the edges of reefs, and all variations of form between the extremes eens in intermediate habitats, Other species may reinin VY >Z. Habit: the most abundant pyroxene; subhedral, subophitie or strongly ophitic crystals, often strongly zoned aid may form perfect Jink between augite an some pigeoniles (q.v-)} grain size very variable (1°3 mim. x 0-4 nim. to 0-1 mm. x O-l min.) ; refractive nidex, double refraction usually increase, whereas 2 V anil ZC decrease regularly toward the border zones of the subcalete atigites. Pigconites. y= 1°82; 2V (+ve) = 20°-0° (optic axial plane 2 (010) ): 7 = 1:730; 2V — 0°-30° (optic axial plane || (010) ) + twining extremely uncom mon; cvlourless, or moderately pleochroic in faint pinkish fawns and greenish tones sinular to the more ferriferons subcalcic augites. Ilabil: never abundant but occurring with one or more of the other pyoxenes (depending on the variety of pigeonite) and with or without olivine; late erystallisations often strongly ophitic. with other types subhedral and strongly zvued; grain size yery variable but usually somewhat smaller than the accompanying clinopyroxenes, Present data indicate several unnsnal occurrences and associations, atid, as far as can be ascertained, nothing quite comparable has been described from elsewhere, Digeonite occurs in seven dolerites, and some of the properties anil associates are set out in Table T. From this it is apparent that the mode of oceurrence of the pigeonite differs in all seven rocks. The probable significance of ihis is discussed i the section on the petrology. 2. OLIVINES Olivine oceurs in a large number of dykes. There seems to be no rile regarding its occurrence, as for instance, m certain of the Kartroo dolerites, where the mineral is present in all sills which do not contain orthopyroxene, (Walker and Poldervaart, 1942, 135). In the Musgrave and Everard Ranges, olivine is often found with an ortho- pyroxene which at times encloses: it. The olivine shows a remarkably constant composition (as deduced fron measurements of optic axial angle which varies fray, 2V (ye) = 88° to 2V (—ve) =85 (2V (—ve) = 87° is common), This suggests a range in com position from a forsterite (Pa 7%) to a chrysolite (Va 23%). The mineril is usually colourless, cracked but remarkably fresh and free from inclusions. In two rocks (7965 and 2460) there is a marked pleochroism: (X = blue-prey, Z= fawn grey; N>Z). Some microscopic particles, possibly iron ore, are orientated i117 the plane LZ, but are not thought to accotint for the marked pleachraism, Tn 7964, 2446, aud especially in 2449, the olivine contains plentiful inclusions of rot ore, The niveral wsually occurs as subhedral ervstals. about Q-6 mm. x OF mn, but crystals up to 1-6 min. x 1-6 mm, occur. In 7965 olivine of essentially the same composition (Mu 2196) is found enclased in augite, and also as sinall ophitic particles of late formation (0-1 mm_ x 0-1 mm.). In 7967 (and to less extent in 7968 and 7969) rare skeletal crystallisatians occur in association with an unusnal “plamose” pyroxene (see pl. XVII, a; andl description on page 194), (4) Minox Consrirurnts Jron ore is present in varying amounts in all rocks. This comstiluent appears to be mainly aiaguetife with some ideenite in places intimately associated. The grain size is yery variable, attaining 0-3 mni. x 0-3 mm,, but grams 0-3 mm. » ( Lisine the proposed classification of Deer and Wager (1939), (84 0:2 mm. are common. Ti many rocks the primary magnetite is often concentrates wiih the small patches of anorthoclase and secondary dusty iron ore. Chromite (2). A few grains of the dull black mineral is included m maz netite In most rocks. vile and Dyrrholite aeeur sparingly in most rocks, aud are associated witli the other iron ores. Amphibole, No primary amphibote is present but a strongly pleochrime variety occurs iu small amounts in the patches of deuteri¢ material in No, 7971. Here it bas been formed in part by the alleration of some of the pyroxene. X = yellow-green, Z = deep blue-green; ZAC =abotit 15°. Bivtite. ‘his amineval is present in almost all rocks as small flecks associated with the iron are and usually with deuteric anorthaclase, Some also occurs as Hecks in ail the earlier minerals (otherwise perfectly fresh) and apparently out of teach of later solutions. Common size is 0:3 mm. x O'2 mm. X= yellowish, ¥Y = Z== dark chestnut brown. Apatite vecurs only as long colourless rods in the deuteric anorthoclasy. In 7971 these rods are up to 2 mm. in length, Tron Spinel is rare but occurs as small deep green isotropic grains associated with magnetite, chrnmite and olivine, c.g., 7963. Quarts was noticed in only two rocks. Tn 7971 it comprises about a halt of the 3-4% of the acidic residue. If is distinguished from the anorthoclase and plagioclase by its clarity, higher refraction and double refraction. It gives i normal uniaxial (-+-ve) interference figure. In the other rock (7972, q.v.) a xenolith of quartz occurs strrounded by 2 strong rim oi subcalcic augite (the dyke injects acidic gneisses ), Chiloritic and Serpentinous and other alteration products are present in minor amounts. ‘The rocks, however, ate unusually fresh, IV, PETROGRAPHY For descriptive purposes the dolerites have been classified on the tollowiing features: (1) presence or otherwise of olivines (2) presence or otherwise of orthopyroxene ; (3) texture. A. O1avixn-nesring DOo-eRires (1) OLIVINE OvriroryRoxENE-BEARING DOLERITES : (a) those with coarse poikilitic texture (7964, 2446) ; (h) those with subophitic texture (7963, 2460, 2449) ; (c) those with ophitic levture (7960, 7961). (2) Onivini-seanine Doierites (ic, free from orthopyroxene) - (a) those with coarse poikilitie lexture (7968, 7969) ; (b) those with subophitic texture ( 7966) ; (c) those with ophitic texture (2453, 7905, 2450) ; (cd) those with an unusual “Phuwese” texture (7967). B. Dorerrtrs (ic. free from Olivitie) (1) OeriropyRoxeNE-BEARING DOLERITES: (a) those with ophitic texture (7970). (2) Norman Docerires (ic., free from both Olivine and Orthopyroxene +: (a) those with sub-ophitic texture (7972) 5 (b) those with aphilic texture (7971). 185 A, QOuivint-srarinGc DoLerrres (1) Ouwine Ontivopyroxene-pearine Dorerrres (a) Olivine Orthopyrexenc-bearing Gabbroic Rock with coarse Poikilitic Terk) |, No. 7964 is a remarkable fresh dark grey medium-graitied gabbroic rocks. It was collected from a large dyke on the new “Royal Mail” track (Feb. 1944), 14 miles east of De Rose Hill hut, In thm section the gabbro appears unusually well crystallised. The olivine. orthopyroxene and clinopyroxene form euhedral crystals which are set poilsiliti- cally in large tabular crystals of plagioclase up lo 6 mm x 2 mm. A stmmary of the main mineralogical data is included in Table IT, The Plagioclase is well twinned on the alhite and pericline and occasionally onthe Carlsbad laws. Its composition is unusually constant (An 65%) but gives way abruptly fo a clear anorthoclase. This mineral is in strong relief tu the labradorite which, though perteetly ftesh, is often a pale chestnut brown on recount of myriads of tiny Inchisiuns of iron ores (2). The anorthoclase has all refractive indices less than that of Canada balsam, but 2V (-ve) varies from 51° im one case to 59° in another (see fig. 2). Associated with the anorthoelase are the usual acicular apatite needles and biotite. The anorthoclase, which makes up 52% of the rock, forms clear-cut boundaries with the Inbradorite (sve pl. XVILd). The orlhapyroxene occurs as beautifully formed crystals. The cores are of a colourless. enstatite (OF 9%) and all gradations are seen fmta the strongly pleuchroic rims of bronaite (OF 16%). “Lhe orthopyroxene contains no inclusions af importance. The extinction is at times frregular and suggests the beginning of the lamellar intergrowth mainly, so well seen in 7963 (see p, 199), The monoclinic pyroxene is represented by non-pleochroie very pale green-grey augite, The borders of some of the erystals begin to show a weak pleochroism in greens and pinks, but it is the double refraction and extinction angle (ZAC = 36-37°) which serve to distinguish the ortho- from the elinopytoxene in recon- naissance, Magnietite 1s included in the olivine in two ways, wis., as particles of iran ore crystils scattered irregularly throiigh the mineral, and as myriads of tiny oficntated iron ore inclusions. 2. No, 2446 is a “dark-coloured rock of medium grain with the normal weathered surface’. It ocours as “a large east and west dyke 154 miles north of Mount Tlbilie” (which ts the most promitient hill in the Everard Ranges). The rock has heen briefly deseribed Ly Jack (1915, 18), and the dyke is figured in the text.‘*?) His deseription iy deficient in that the most abundant pyroxene (orthopyroxene) has been iwistuken for a clino-pyroxene, and the anorthoclase has been overlooked. The rock in thin section is very similar 1o No. 7964 deseribed above, The ctystals of pyroxene are mol quite so well crystallised, however, and there may be a little more clinopyroxene present than in 7964, Vhe optical properties of the minerals, however, are almost identical in the twa rocks, even though roek No. 2446 was collected 35 miles trom 7964, Both 7964 add 2446 contain minerals tore magnesia-rich than most of the dolerites. Even thongh they have apparently cooled for a longer period, lamellar intergrowth so ‘) This rock is strictly of eabbroic texture but is. treated With the dolerites because of its obvious telations, (ROL, Jack briefly described tocks from several basic dykes in the region of the iiverard and Musgrave Ranges. Vhese sippear an the report on his eeological reconusts- sante in the area (Jack 1915), The origitl thin sections described by Dr. Javk are still ivallable for examination, but unfortunately the hand specimens. are tissing. Las \ypical of many bronzites (ey., those of 7961, 7970, 7963, tte.) is absent. A “shadowy” extinction is, however, occasionally present. It is probable that at no time did early pigeonite form, (b) Ofivine-Orthopyraxene-bearing Dolerites with Suboplitic Texture. 1. No. 7963 is a fine-mectium lolo-crystalline very dark grey rock, It occurs as a steeply dipping east to west dyke cutting gneissic granitic charnockites near a tiny rock hole a furlong north of the track about 19 miles from Ernahella toward Kenmore Park in the eastern Musgrave Ranges, In thin section the rock appears as a temarkably fresh olivine-bronzite- bearing dolerite with only a very poor ophitic texture. It has apparently cooled fairly slowly and in fairly “dry” conditions, The results of @ chemical analysis and calculation of the norm and mode are set out in Table IIL, and a stmmiary of the main mineralogical data is included uy Table IL, Tr is apparent at once that the normative and modal minerals are at con- siderable variance: lt is stispected that this is largely due to a high alumina content of the clinopyroxene, "This carmot be readily proved, however, far the pyroxenes ure all so similar in general appearance and properties that it ts Loo difficult at present to attempt a nimeral separation with a yiew to analysis, Caleu- lations, based on estimated relative prapertions of the heayy iinerals and wark- ing back Frain optical cata of the plagioclase, olivine and orthopyroxene, have itlicated that the clinopyroxene probably contains about 109% Al,O,. Some ALO, however, may be held in the orstopyroxene or even in the olivine. Frankel (1943, 20) noted that some analyses (sitvilar to that under dis- cussion here) showed “little similarity between the norms and modes by volume, and still less between the former and mode by weight. Bowen (1928, 142) points out that the norm does not take into account any alumina which may be in the pyroxene, and thus false values for anorthite and consequently for pyroxene are obtained”. The relative percentages of MeO (notably hich for a dolerite), CaO and AIO, are unusual, which fact has made selection of comparable analyses diffi- cult. A iew dolerites from the adjoining region in Western -Ausiralia have been analysed but tiost of these are quartz-doletites. The olivine dolerites from that State are usually much Jower in MgO and also differ in other respects, No. 4, in Table 111, is an exaniple of one such dolerite. A search through the available literature has shown that the most comparable rocks are probably those deseribed hy Prankel from the Karroo, and Ferguson from Rhodesia. The analyses with their norms appear as Nos. 2 and 3.in Table S11, The modal percentage (by volinne) of pyroxene in 7963 is given as 45:68. The main pyroxene is a mon-pleoehroic faint yellowish-preen avgite with 2V (-- ve) — 48", ZAC=39" A few simple twins are present (|| (100) ). Commonly fan-like bunches of crystals oceur which presents an irregalar eXtinc- lion in places, rendering difficult the aceurate measurentent of ZV and ZAC. This extinction ts a type of “polarisation cross’? with the cross. indicating an extinction angle (Z/\C) of about 38°. The refractive index (y) = 1-708 002. It is probable thar this is the pyroxene which has the high alumina content discussed above. Next in importance is bronsite which comprises about one-quarter of the total pyroxene (45°6%). Ltis pleochroic in pinks and faint green-greys and has 2V (-ve) =71°, which suggests att approximate composition OF — 22%, The extinction is usually straight and ita double refraction ts lower than im the clina+ pytoxenes. 187 The mineral often encloses olivine (Va 18%) and is itself sometimes partly enclosed by the nan-pleochroic (or weakly pleachroic) clinopyroxene. A pately extinetion Lending to grade into a fiae lamellar intergrowth parallel 1a (OLO) is often seen in suitable sections. _ The proportion of the third pyroxene {a sithcalcic augite) js difficulty tu estimate. It probably makes. yp, however, between one-fifth and one-sixth of the total pyroxene, At first sight contusion could easily arise as the pleochroisin is almost exactly the same as that of the bronzite, Phe only certain method of distinction is that of measurement af the optic axial angle (2 V (-}-ve) = 40-457). although the large extinction angle (ZAC == 38°) and higher double refractian assisl to identify the clinopyroxene in reconnaissance.“) This pleochroic clitic- pytoxene is similar im some respects to that hr No, 7968, where it forms a con timmons link between the carly pon-pleochroie augites, weakly pleachroic suly calcic anyites aud more iron-rich pigeonites of later crystallisation, No pigeonite (2V(-+-ve)<40°) was noticed, although a careful search wa- made, “Nearest approach” to a late pigeonite is probably the pleochrote sub- caleie angite deserthed above, No “carly pigeonite can he expeeted owing LW the presence of the lamellar intergrowth in the bronzite, whieh intergrowths hav. been formed by an mversion (?) of a pigeonite during the cuoling process. [Tad the cooling heen accelerated some early pigeonite may have remained in metastable condition, as in the ophitie dulerite Nu, 2453, This rock covtamns much more offidne (han most of the other dolerites, It is remarkably fresh and clear, Of the 4% accessories, the ivon ores are the most importint, Ty order af importatice they are magnetite, ilmenite, chromite, and pyrite. About 0-5% uf brown dotite is present. Jt is asseciated with the iron, but no anerthectase was noticed. The plagioclase varies in composition from bytownite (An 829¢) to labra- dorite (An 6796). Some edges of the plagioclases are somewhat more sodic, The average is probably bytownite (An 729), The most common twin laws repre- sented are albite, Carlsbad and pericline, ‘The mineral is somewhat clouded, nor by alteration products Init by dusty inclusions of (2) magnetite (cf. 7964. 7961, etc). 2. No. 2199. Nos. 2449 and 2460 show the subophitie texture better than the more basic rock No. 7963. Nos. 2449 and 2460 were collected by R. L. Jack (1915, 19), and briefly described by him as) Nos. 14-32-48 and 14-3-23 respectively, No, 2449 is a “dark-coloured rock, fairly coarse-grained and weathering to a reddish surfiee”, The rack occurs in a “large dolerite dyke forming the cap” of a black hill two wiles west-north-west frum Mount Carmeena in the Evyerari| Ranges, where the dyke cuts granite. The dyke strikes approximately egst to west und dips about 35° nerth (a phulograph of the dyke is given facing page 18 af Dr. Jack's account). In ttit section it appears us ai olivine orthopyrexene-hearing dolerite, s\ sununary of the main mincralogical data is included in Table 11. The pleochroic niineral 1s not augite (except only on some border zones) but a bronzite (cf. Jack 1915, 179). The hiotite is always “associated with the magnetite” it is true, but in this rock is duc to a late inagmatic (deuteric) alteration of the mag- (% Johansen (1937, 3, 212) has eniphasised that orthopyraxenes cun have an inclined extinction in certain sectians, Great care had to he taken in several of the dolerite here described because of Ure presence of huil) an ortho- aud cling-pyroxene which showed similar pleochroism. netite and sometimes pyroxene. Colourless anorthoclase wich many coluurless needle-like inclusions of apatite comprise the matrix of the biotite. No quariz is present. Typical masses of anorthoclase have 2V (=ve) = 52° and all refractive indices below that of Canada balsaiu, A plot (1X) of the (010) and (001) cleavages is indicated in fig. 2, Another good area of the mmeral has 2V (-ve) 42°+ 2°. No twinning is shown even in (he highest magnification, The mineral has very low double refraction. A notable feature of the olivine (Fa21%) is the abundance of squat needles of (?) magnetite inclusions (especially in the cores), The mineral 1s non-pleochroic (contrast Nos, 2460 and 7965). The bronzite is more ophitic than the subealeie augite. The most ferriferous hronzites occur in this rock (bronzite, OF = 22). It occasionally shows a patchy extinction but no notable lamellar intergrowths. Neither early nor late pigeonite rs present, Though fresh, the feldspar is stained a dirly brown with minute dusty inclusions. As the outermost zones of the plagioclase are reached this staining is less marked. The composition varies gradually from cores approaching a bytownite of An 75% to rims of andesine (An449% or even less). There is often gradation from this last zone to the anurthoclase in such a way as tu suggest the latter's crystallisation [rom the residual liqtors. 3. No. 2460 is “a very finely crystalline dark grey rock weathering to a reddish surface”. It occurs as a “dyke in a swasip on track, 264 miles north 63" west from Moorilyania Hill", and is in the region between the Musgrave and Everard Ranges, Tn thin section it appears as an olivine orthopyroxene-bearing dolerite. Dr. Jack (1915, 19) makes mention of an earlier generation of augite which is distinctly pleochroic. On investigation this earlier generation “augite" proves to he a typical brensite cuinparable with that in No, 2449, It makes up approximately one-half of the total pyroxene. ‘The bronzite encloses olivine (Fa 12%) but is itself enclosed by the non- or weakly-pleochroic subealcic augite. It is readily distinguished from the clinupyroxene by its lower double refraction, pleochroism and comiton straight extinction, “The tuineral only oceasionally shows a patchy extinction. Neither carly nor late pigeantte was observed. Features of this rock are the protiounced development of dark brown dusty inclusions in the lwtownite and labraclorite, and the pleochroism of the offerte (¥a 12%) in Dive-greys and fawn-greys. No. 7965 is another rock showing 4 pleochrote olivine (q.v.)- (ec) Olivine Orlhopyroxcac-heering Dolerites with Ophitic Texture, Nos. 7960 and 7961. Dolerites belonging te this group are best represented hy Nos. 7960 and 7961, which were both taken from the prominent dyke which euls the charnockitic granodiorite on the west of the win valley between Erna- bella and Mownt Carrtithers in ihe eastern Museravé Ranges. The dyke itself is several miles fony and is probably cont'nuotts with a Similar dyke to the sonth of Ernabella (see main map, Wilson 1947). This dyke is the most important of a poorly developed set which strike more or less north and sotith with a shallow dip to the east. This particnlar dyke is only about 15 feet thick, Int as. never- theless, one of the largest dykes iu the eastern Musgrave Rarges. () Most of the doferite dyes of the Muswrave Rives strike south of ives, and dip steeply 60°-70° toward the soirtls, 189 In hand specimen the character of the dyke rock varies from a medi 10 coarse-grained very dark grey dolerite. Voth types appear retmarkably fresh. A suniinary of the main mineralogical data is tneluded in Table IT, ‘The finer phase (7960) was taken from near the bottom contact, whereas the coarser phase (796L) was collected near the middle of the narrow low dipping dyke. Lille ilitferentiation has taken placc. The coarser phase is apparently somewhat richer in fron and volatiles, as is revealed in both modal percetitages and optical properties of all three pyroxcties, and the presence of a little late pigeonite which ts absent from 7960. The plagio- clase is a slightly more sadic variety of labradorite. Bronzite commonly has excrescences of a subcalcic augite, which often show a peculiar fan-like o1 stellate form with shadowy extinction (cf. that seeu in 79603), In the modal percentages, an accessory amount Uf a soda feldspar is included with the plagioclase. The mineral is sisular to that deseribed from tocks num- hers 7964, 7966, etc., where it was cleterntiined as an anorthoclase, Ln this dyke the amount varies from about 2% in No. 7960 to about 4% in No. 7961. The anorthoclase is both interstitial and partly replacing the outer more sodic zones of the plagioclase, No quartz has been determined but long apatite (7) needles are plentiful as inclusions. .A pleochroie (yellow to brown) biotite and dutsiy magnetite are usually associated with these patches of anorthoclase, which is upparently mainly of deuteric origin. Primary magnetite and pyrite ate plentiful, but the pereentage of total iron ores 1s increased by the dull black material liberated om alteration of some of the olivine and pyroxene. The coarse phase (7961) contains much more bronzite than the finer phase (7960). This may represent the inversion tu bronzite of the early pigeonite (see p, 199), The presence o7 a continual gradation from the subcalcic angites into the late pigeonites seems in this rock lo be assisted by the somewhat higher concentration of late liynors (as indicated by the coarser grain size ancl increased percentage of detiteric anorthoclase), (2) Ontvine-pearing Donerirrs (1v., those free from orthopyroxene), (a) Olivine-bearing Dolerites with coarse Potkititic Texture. tL. No. 7968 is a dark grey medium-grained rock take fom the most southerly of several parallel dykes which cut the south-west tip of the large mass af charnockitic granodiorite situated about two miles west-north-west from Palpatjara, which is three fhiles south of Ernabella. The dyke, which strikes teward the west-south-west and dips 63° toward the south, is tignred ina pre- viotis paper ( Wilsorr 1947, pl. 11, fig, 4), Jn thm section this fresh rock appears holoctystalline with pyroxene and olivine mainly set poililitically in relatively large (imax, 4-6 mn x 1°2 mm; average 2°0 mm. x 1°0 mm.) crystals of labradorite, A summary of the main mineralogical data is included in Table II. The plagioclase (42-59%) is usually clear but occasionally has a pale brown dusty patch due to tiny inclusions. The mineral is so poorly twinned (mainly on the albite and pericline laws) that uo easier method of estunating its composi- Hien was available than to use the plotting the poles of the cleavages on a stereogram, Zoning is not prominent and the feldspar has a fairly constant composition of An 67% + 3% (ie., a basic labradorite). Neo anorthoclase was detected. The pyroxene (48-19%) occurs i two generations, The larger crystals (up ty 1'3 wim. x O4 mm,, but average 0-8 mm. x 0-6 mm) arc commonly subhedral- elear and colourless (but very pale brownish pink in some atttermast zones )- 196 The cores have 2V = 52° and ZAC= 36", but there is a gradual decrease towards the rims to pigeonite with 2V = 19°, ZAC = = 30°, optic plane || (010) (see fig, 3 (a)). The strange “polarisation cross extinction effects are com- monly present, The smaller set of pyroxene crystals (a late crystallisation) ave virinally colourless, although some show a very pale pink when parallel to X. They are ragyed or sphenoid in shape, and have an average grain size of (15 mm, x 0-05 mm. ‘The grains are too small, and extinetion tom erratic to do accurate optical determination om most grains “iu sitn”, but of those studied, many are pigconite with vartable wplic axial angle (one with 2V (+-ve) as low as 17°; optic plane || (010). ZAC could not be determined as there are neither suitable cleavages tor twinning, The olivine (669%) which contains about 249% tayalite molecule is fairly clear and colourless, but along the plentiful cracks secondary maymetite and serpentine have forined. Vhe mineral 4 is ul times subhedral but éommanty holey, ail on rare cecasions simulates sume of the “skeletal" olivine of rock No, 7967 (qve), Accessomies (2°8%) comprise mostly some primary magnetite, secondary magnetite and serpentine, with a little pyrite. Neither anorthoclase nor bronztie was observed. Though poikilitie and containing relatively coarse plaginclase, the name “dolerite” is preferred to “gabbro” because of (he grain size and charac- ters of the finer set of pyroxenes and its affinities with the main suite of dolerites. The apparent absence of both bronzite and early pigconite but the abundanee of ‘a magnesic clinopyroxenc¢ is interesting. Tt is possible that the conditions which allowed this poikilitic form (so uncommon in this avea) ur develop were unsuited to the formation of cither at orthopyroxene ur early pigeonite. 2. No. #969 is dark grey medium-grained rock occurriny as a smtal] east- west dyke cutting the gneisses near the junetion with the large charnockitic granodiorite mass to the east in the ercek beds one mile west of ‘Top Springs (5 miles north-north-west of Ernabella)). Tn thin section this rock shows similaritics to two widely separated ezst-to- west dykes (Nas. 7968 and 7967). “The plagioclase (uy to 6 wan, x 2 m0. average 2-2 mm. x OB min.) ts a ae basic labradorite and comprises appruxi mately 50% of the toek. lt cheloses poikilitically the pyroxenes and very similat to those deseribed wbove in No, 7968. except that the smaller set (average M15 mn. x 0°05 mm.) tend ina few places toward a crude development of the plumose texture so pronounced in No. 7967. The grains are very small but no senses was definitely proved (contras: 7968) although it is suspected. Neither bronzite nar early pigeonite was observed. Olivine (often in part decomposed) and inagnesia-rich clinopyroxenes ure plentiful (ef. 7968)- (b) Ohwvine-bearing Dolerites with Subophitie Texture. 1. Nu, 7966 is a dark grey dyke rock occurring ov the eastern Aank of Razor Lill, 11 miley north-north-cast of Krnabela. dt cits Whe chartivelitic franodiorite just inside its junction with the gneisses. in thin section the rork is seen to haye undergone a considerable amount of alteration by (mainly) detiteric solutions. This has resulted in a precipitation of dusty iron ore and formation of an amphibole, biotite and anorthoclase. A symmary of the main mineralogical data is included mm Table JI, The percetitage (58:1) for plagioclase includes 4-5% anorthoclase similar in optical properties and relations to that described elsewhere (¢@.g,, 7964). Usually the cores of the much zoned plagioclase are quite a dark shade of brown owing to myriads of tiny dtisty inclusions of sper ore (?), The more sodie outer tims show fewer dusty inclusions, and the 19% anorthoclase is perfectly clear apart from the usual few needle inclusions of apatite (7). The plagioclase often has a core with composition about An68%. the outer zones ranging through the average type (a labradorite An65%) wa sodic Andesine (An34%) or even less, as it gives way in many casts to the anorthoclase. ‘The pyroxene is a strongly pleachroic relatively ferriferous variety of sub- ralci angite and in places gives way to a pigconite (optic plane || (010), Plenti- ful clouds of magnetite (?) dust are often seen scattered at random through the pyroxene (contrast the regular zones of “dust” in the pyroxenes of No. 7970). The olivine (Fa= 19%) is plentiful (6°3%), but much has heen altered to serpenttinous material and iron ore (the original rock proper contained about 8% olivine). A. little of the pyroxene has often becn altered to a biotite and pale green amphibole, bat most of the deuteric products have been formed from the iron ores, Along a few lines of fracture abundant magnetite has been liberated frum the pyroxenus, ‘These fractures Occurred after the maim period of deuteric alteration, as rarely is the liberated iron ure surrounded by biotite, The most peculiar feature of this rock is the absence of bronzite hut the presenice (though rarely) of a small pigevnite core in some of the subhedral crystals of subcalcic augite, The pigeonite has a high 2V (27°-30°) with optic plane || (O10) and grades (with no break) into the surrounding subcalcic augile {2V = 30-40°). (e} Olivine-bearing Dolerites with Ophitie Texhore. All olivine dolerites yet examined which lack in bronzite and have a notable ophitic texture contain a pigeonite. 1. No. 2453 is a “very dark grey toek with reddish-brown weathered. sar- faces”. Lt occurs “7 miles north-west of Ilealy Springs” on the track between fndulkana and Moorilyanua Hill. ‘The rock has been briefly deseribed by Dr. Jack (1915, 19) as No. 14-3-9. The important mineral pigeomte is omitted from his description. his is not surprising, as both the subcalcic augite and pigeonite are very similar in all obvious optical properties and only easily distinguished by estimation of opti¢ axial angles. The subcalcic augite shows ophitic texture exceptionally well with the ragged plates averaging 25 min. x 15 mm. With this occitrs the pigeonite (4 (?) of the total pyroxene). This pigeonite differs from some af the pigeonites described in other rocks (c.g. 7970, 7961, 7968) in thar the optic axial plane is perpen- dictlar to (O10) and there is a marked break between the two mincrals when viewed under crossed nicols. Both minerals are ophitic but the pigeonite less so than its host, the subcaleic wegile, Fig. 3 (b) indicates the strange variwiion in optical properties from the pigeonite lo the rim of the enclosing clinopyroxene. ‘The pigeonite itself appears to have litte variation ( relatively) in optic axial angle (2V (--ve) = 14° to 22°; ZAC = 39°; optic axial plane 1 (010)). This probably indicates that the trend of the noti-pigeanitic climopyroxenes is upwards toward the “Hump” of Tess’ “conrse of crystallisation” curve (see p, 197, and fis. 4). fhe absence of bronsite is noteworthy, The conditions of cooling were apparently such that the carly pigeonite at no time was ablé to invert to ortho- Jwrosene, The olivine which is more magnesia-rich than usual (Fa 119) is not plenti- ful (2°2%). IL is usually somewhat “corroded” and ettclosed in the early pigeonite or other clinopyroxene. “The minwtal covtaing some dusty magnebte inclusions and js somewhat serpentitised, 194 The plujioclase (59°79) is not nearly so variable m composition as usual. ly js a labradorite (An62%). There is a small amount (049%) of a Lypical shorthaghase. A summary of the main mineralogical data is included in Table U. 2, No. 7965 is a dark grey rock taken from a prominent dyke at Vietoria downs Homestead, 65 miles east from Frnahella and 125 miles trom the Finke vailway siding in Northern ‘Territory. in thin section the rock presents a medium-grained ophitic texture. A suinmary of the main niineralogical data is included in Table IT, A striking feature of the rock is the pleochroic olivine (X = blue-grey, Z == fawn-grey). There are myriads of little inclusions of iron ore arranged 1 %, and the possibility of these causing the “pleochroism"” was investigated. It seems that the pleochroism is intrinsic. This is yncommon, especially in such a inagnesia-rich olivine (Fa 21%). The olivine began to erystallige first and continued to form (with much the sme composition, viz, Fa21%) to the end, where is oceurs with the late igeonite and subealcic augites as small sphenpid particles only Or1 mm, x O°1 mm. No. 2460 contains a pleochrote olivine (Fa 124), but ity associates are different ‘rom thuse in this rock. Augite and pigeonite together make up 325% (by volume) of the rock, Since both minerals exhibit the same pleochroism and both occur poikilitically the only easy method of distinguishing the two species is by an estimation of optic axial angle. The pigeonite usually uceurs in somewhat smaller grains, has a low uptic axial angle and very poor cleavage, therefore it was found tnpossible to definitely prove whether this species had its optic axial plane parallel or _pet- pendicular to (010). The work done suggests, however, that it is |] (010). There is probably a continuous series between the lower optic axial angle subcaleic augite and these pigeonites (cf. No. 7968}. This pigeonite usually occurs as small sphenoid erystals controlled by the poikilitic texture of the rock. The aigite varics widely in composition, as indicated by the fact that the sptical properties vary from 2V (-+-ve) = 46°, ZAC 42° through the mean of 2V (+ve) =42°, ZAC=4l°, to 2V (+ve) = 31°, ZAC=— 37". As stated above, there is probably a continuous series towards an almost “uniaxial” pigeonite, ‘There is also a hitile pigeowite of a different nature. Tt is enclosed by the augite, has 2V ( {-ve) = 19° with optic axial plane 1 (010), Thus this dolerite contains two pigeonites, The “early pigeonite’ shows a break with the atgites, Init the later type shows nu such break and probably forms a continuous series fram the subicalcie augites, No bronzite iz present; conditions of crystallisation were apparently such as ‘0 prohibit inversion uf the small amount of “early pigeonite”, The plagioclase is zoned with compositions cstimated at hytawnite (An729%) fur sume cores to basic andesine (An49%) for same rims. A basic labradorite ( An68% ) is common. 3. No, 2450. Dr. Jack (1915, 19) states that he collected this specimen fram a dyke six and a half miles $.60° W, from Moorilyanna [fill”, @e.. about 18 miles west from the locality of 2453. The getreral appearance of this rock resembles Nu. 2453. The oplntic texture, for imstanee, is similar to that of 2453. However, there are considerable aid significant mineralogical differences. A sunumary of the main mineralogical ‘lata is included in ‘Fable JI. As it Inost rocks oliie was the first ferromiagnesian inineral lo crysiallise, wud it continued for same Vime, Nu bronsite vor “early pigeonite” is present, i 193 and the next mifieral to appear in the dugite (2V (-+-ve) — 49°-45°) rimmce with subcaleic augite. The augite cores 2V = 49° with continuous gradation toward the rims of a subealcic augite with 2Y = 37°, On some edges (never enclosed) occurs a pigeanite, It has 2V (+e. ranging from 22° to 15° with optic axial plane perpendicular to (010) and hence usually shows a pronounced optical break with the main clinopyroxene, ‘I'hi- change from the subcalcic augile to the pigeonite seems always to take place about 2V=37°. oth pigeonite and subealcie augite have well-developed ophiti texture and seem to have come out together at the end as the Gnal crystallisation (sce fig. 3(c) ). In 7965 a “late” pigeonite crystallised with the last subcalcic augite, but int that rock the pigevnite has the optic axial plane parallel to (010) and there is probably @ continuous series between the two clinopyroxencs. Moreover, a plea- chroie olivine (Fa 2190) is found in 7965 crystallising right through to the ened uid occurs as small ophitic sphenoids with the two clinopyroxenes, In 2450 the variation in the composition of the clinopyroxene is shown bs the gradual decrease iti optic axial angle from 2V (-+-ve) =49° tw 37% It is probable that the trend is along the “course of crystallisation" (see p, 199), beginning near the top of the hump and ricar the poitt reached hy the clino- pyroxencs of rock No. 2453. Plagioclase is abundant (5596) and notably calcium-rich, Iv is well zoned with cares of bytownite (An82%) ta rims of andesine (An46%), with an average af bytawnite (about An75%). The plagioclase is much ¢learer thin usual, with only a few cloudy patches of itelusions, Anorthoclase is present but rare, and there is a little subhedral original mmugnelite as well as the secondary dusty iron ore. Some serpentinous materral is present as an alteration of some pyroxene and olivine. (d) Olizine-bearing Dolerites with an unusual “Phonese’? Tertnre. There are several fine-grained dark grey dyke rocks which cut the north-west slopes of the charnoekitic granodiorite hills 14. miles east-south-east of Ernabella. Similar dykes have been noted cutting similar rock near [yinpiri, 6 miles north of Ernabella. No. #967. A suinmary of the main mineralogical data ts included im Table II. In thin section the plagioclase appears as lathes which attain the size af 2 mm. x 0-05 mm., with many 1 mm, x 0-04 mm. (average probably abour 04mm, x 0-03 mm.). The lathes show no flow structure and are in some places criss-cross, in others intergrown with the “feathers'’ of pyroxeue. The plagio- clase {about 25% of the rock) is clear of the dusty inclusions so common in the dolerites of this area. The mineral is poorly twinned on Carlsbad and albite laws and imeéasurements of extinction angles indicated a basic lahradorice (An67-659), Zoning is not cominen attd anorthoclase rims are absent. Pyroxene makes up about 55% (by volume) of the rock. lt oceurs iu two forms. Up to two-thirds of the pyroxene form irregular and occasionally suh- hedral crystals up to 0-4 mn. x O'3 mm. with an average size uf O-2 uum x 0-2 mm. The mineral is a clear, non-pleochroie augite of fairly uniform cone position as indicated by 2V varying only from 45° to 47° and ZAC = 37 +27, The augite often shows an extinction similar to the ‘polarisation cross’ noted in the pyroxenes of other dolerites of the area (¢.g., 7963). The other form af pyroxene is that which helps to make the texture of thi: rack so distinctive, It occurs as feather-like and fan-like bundles of irregularly tapering rods. The bundles are on the average 0°6 mm. long % 0:2 mim. wide at the broad end and taper to OL mm. wide at the apex, although some attain I the dimensions of 1 min. x O'4 mm. x OZ mm. The pyroxene indivaenals then selves are usually about 0-5 mim. long x 0°02 mim. wide. These phiinase clnster- show ne preferred orientation, for sectiuits cut perpendicular and parallel to the conling surface shaw uo significant difference (see pl, XVIII (b) ). The mineral may be of almost the same composition as the main augile, but no deimite measurements could be made owing to the small size and Lhe presence of a wavy extinction, However, its double refraction is similar to that of the ‘ugites, it has a medium extinction angle, is non-pleochroic and 1¢ often seer “sprouting” from the larger generation of clinopyroxenes. There is, however, amall possibility that it is a “late pigeonite” similar to that found in No, 7968 and suspected tn Na, 7969, Olizine (Fail%) is abundant (between 18 and 20%) sil distinguished fram the clear pytoxene by having (1) a faint fawn tint (and a very weak pleochroism; X slightly darker than Z), (2) many cracks filled with dusty iro ore and irregular patches of brownish dust, (3) a somewhal higher double refrac- tion (2V (-+ve) = 88"), The most striking feature of the mineral is tts mode of crystallisation, Sometimes a solid subhedral crystal is seen, bot usually it otcurs 28 a strange type of zuned crystal. Orientated inclusions of a very fine niass of pyroxene atid plagioclase (sometimes suggesting coarsely devitrified glass) are often so abundant and specially arranged that the crystal appears skeletal. Plate XVIIT (a) shows a hexagonal form reminiscent of pseucdo- hexagonal twinning (as is cordierite) but the arms of the crystal extinguish together and have no properties of twins. In some cases crystals of pyroxene well beyond the borders of an olivine are seen in optical continuity with pyroxene some distance within the skeletal olivine. Fairly large individuals of clino- pyroxene have been enclosed by groups of olivine crystals in such a way as to suggest that some of the olivine crystallised after some of the clinopyroxene. In places olivine acctits as small granules and blunt ragged prisins intergrown with the phimiose pyroxene. The evidence indicates that though olivine began to crystallise first (as in all of these dolerites) it was soon followed by a clinopyroxene (but neither early pigeonite nor brouzite) which contitiued with the olivine for a relatively con- siderable length of time, While the last olivine was being formed the “plumose” generation of clino-pyroxene began its formatian and continued to the end, There is very little primary iron ore present. but most of the cracks in tle olivine are filled with a dusty iron ores Neither early pigconite nor bronzite is present, and no late pigeomie was proved, Anorihoclase was tot noticed, B, Donerrtes (Olivine Free) (1) OetHoryroxeve-pranine Doceni res, (a) Orthapyroxcue-bearing Dolerites with Ophitic Texture, 1. No #970 is a small cast-to-west dyke which intrudes gneisses near the junction with the chartiockitic granodiorite about three wiles wesl of Ernabella- ln hwnd spevimen the rock is similar to the other dykes, In thin section the rock appears fairly fresh. A summary of the main mineralogical data is included in Table 11. Most of the $7°6% of plugivclase \n the tock is a basic labradorite (An66%) but cores are often as basic as a bytownite (AnW7290) and there is commonly a continuous gradation toward a narrow oligoclase border (about Ang5%). A little deuteric anorthoclase (2V (-ve) = 52°) is presetit, but mo quartz was obsetyed, The central part of many feldspar crystals shows the dusty inclusions so commoat in the dyke rocks of this crea. 195 The pyrovenes present an interesting problem. it appears that both ortho- pyroxene and pigconite have crystallised at much the same time. No ev- dence was found which was sufficient to prove which mineral begai Lo erystallise first. The final pyroxenes to form are a mixture of subealeic augite with some hronzite and pigeonite. The orthopyroxene is similar to thar seen in No. 7961, and has a hronzite core (Ot 16%) which gives way gradually in the more normal Inanner toa bronzite rim (ahout OF 22%), The bronzite shows marked lamellar twinning (|/ (010) ). The course of crystallisation in the clinopyroxencs. hawever, is difficult ta follow. Some subhedral crystals arc found showing a rentarkable zoning, Occasionally the core ix enclosing an cistatite (OF 99%) aud is a prgeenite with 2V—ahout 20° and optic plane 4 °(010). "Phen follows « perfect gradation toward w uniaxial pigeonite (2V—O0°) into pigeonite with the optic plane || (Q10) (2Vs- 0" to 30°), Finally the gradation Icads (with no hiatus) to a subcalete angite with 2¥V = 40°. The double refraction regularly increases slawly from the core to the rim, but the extinction angle (ZAC) remains fairly constant (about 34°). {n such a zoned pyroxene abundant Nusty iron ore inclusions have been precipitated when the 2V reached about 32” (fig. 3(d) ). The remarkable feature of this zoning is the fact that bronzite. which shows a patchy extinction us well as a fine lamellar twinning (|| optic plane), is present in the same rock and may have crystallised at much the same time its the clinopyroxenes (see p. 197 for fuller discussion). The pyroxenes comprise 34°79 (by volume) of the rock with the bronzite. pigeonite and subcalcic angite represented in probably equal proportions. The texture of the rock is moderately ophitic, but a considerable number of the zoned clinopyroxenes have developed a subhedral form, The bronzite and soiie clinopyroxene have, in places, developed a notable bladed form and some- limes are secu in a coarse plumose Erin somewhat reminiscent of that in No. 7967. (2) Noxarsn Doverirrs (7.c., free from olivine and orthopyroxene). These rocks are similar in many ways to the olivine-bearing suite described above. The absence ot hoth olivine and orthopyroxene, however, aod the presenoe of somewhat more ferriferous pyroxenes and more detiteric acidic inaterial suggest that this group is probably a later differentiate fram the com- inon doleritic magma. Two types are recognised. (a) Normal Dolerites with Ophitic Texture, These are plentiful, especially in ihe eastern Musgrave Ranges. No. 7971 is a coarse-grained dark grey dolerite which outcrops prominently about 4-4 miles from Kenmore Park on the track to Ernabella. Lt cuts the hypersthene granu- lites and gueisses at a steep angle. In thin section this coarse doleritic rack appears temarkably fresh, except for certain denteric reactions. Mineralogical (lata are set out in Table 11, The plagioclase (49°0%) is very well zoned with a composition showing perfect gradations from cores of fabradorite (An66%) to rims of andesine (An349), The probable average is a labradorite (An62%). The plagioclases show the characteristic pale brown dusty inclusions. The modal figure of 49-0% includes about 49 of clear acid deuteric material composed of both quartz and anorthoclase, The deuteric solutions have attacked the pyroxenes in places yield- ing a strongly pleochroic amphibole (X = yellow-green, Y= green, Z— deep bine-green) with ZAC = 15", biotite (X — yellow, ¥=Z= very dark brown) magnetice and apatite. The amphibole comprises 3:8% of the rock, biotite 1°2% and about one- third of the 7% magnetite is a dull black secondary tron ore. [5 The pyroxene (37:09) is a pinkish subcalcic augiile somewhat more fer- rilerous toward the rim than at the core, The lowest optic axial angle observed was 2V (+ve) = 35° (and ZAC = 38°), which was the outermost rin of a crystal with a core with 2V (-|-ve) =40° (and ZA\C= 41°). Twinning on (010) is more plentiful in this pyroxene than in any other of this area. The donble refraction jis a little higher toward the rim but this is sot as marked as in No. 7972, Neither pigconite, bronzite, nor ohyine is present, (hb) Normal Dolerites with a Subophitic Texture. No. 7972 is a medium-grained dark grey dyke rock, and one of several which cut the gneisses forming the ridge between Gilpin’s Well and the track linking Ernabella and Kenmore: Park, and 20 miles from Frnabella, In thin section the texture is somewhat different from that of the other dykes yet studied. Thenocrysts of clinopyroxene wre present, as well as the smaller generation of srualler but more ferriferous pyroxene. Tt 15 in this later’ pyroxene that the subophitic fexinure is developed. The plagioclase, ti, occurs in two poorly defined generations, The larger set are comparable in size with the phenocrysts of pyroxene, They shaw very marked discontinuous “on- ing (pl. XVII (ce) ). The smaller set show the gradational zoning so cont mon in this region. Mineralogical data are set out in Table IT. Plagiaclase comprises 51°3% (this includes about 39 of deuteric quartz and anorthoclase). The average composition ef the mineral is that of an acid hytownite (An729%) but the cores of some crystals are as basic as an anorthite (An91%), with the borders of many crystals reaching almost to an oligoclase (An31%)}. The plagioclase is stained a fawn-brown (contrast the 4% deuteric material which is clear), The only pyroxene present is 4 suhealcic augite, The optical properties of the atiter zones of the phenocrysts and smaller crystals suggest a gradual iron enrichment ag the crystallisation proceeded (2V decreases from 39° to 36°; ZAC increases from 40° to 42°; double refraction increases from. core to rim). Two or three very small corroded crystals of a faint dull brown olivine up io 0-3 mm, x 0°15 mm, oceur as inclusions in the phevocysts of subcaleic augite. Beside the presence of quartz amongst the deuterie material; quartz oeciirs as ah interesting xenolith. The clear mineral is surrounded by a prominent rim ot subealcie augite. The lens of quartz anél pyroxene together is about 1°8 mm long x 1 mm. wide, with the actual core of quartz about O'S mm. x O-3 mm, The plentiful iron ores (5-6% ) comprise about half primary magnetite and the rest ts dull black dusty secondary iron ore. Neither “early” nor “late” pigeonite nor bronzite is present and otly two or three tiny adventitious crystal of olivine occur. V, PETROLOGY?" Tue general the minerals of each dolerite indicate a normal course of crystallt- sation with a trend toward a soda enrichment in the plagioclase and an iron (arul part calcium) enrichment in the pyroxenes (and olivines), phenontema ably dis- cussed hy several recent authors. In several ways, however, the mineral assemblage differs considerably from that of doléerites desctibed elsewhere, and associations obtain which have hilhertn been largely thought improbable. Some of the main unusual features will become apparent in a brief survey of the pigeonites of these rocks, (Table 1). Piyeonite makes its appearance in @) Qwing to lack of large sills no extended study of the course oi erystullisation of the duleritic magma could be undertaken. he specimens here described wore collected in the first instance to give an idea of the types of basic dyke rock in the urea, 197 styen dolerites and in each with its own peculiar associations, On plotting the compositions of the ferromagnesian minerals of the pigeoniie-bearing rocks in fig. 4, it is apparent (hat most of the atigites and subcalcic angites fall below the “course of crystallisation” curve of Hess (1941, 585, fig. 10), When the pyroxenes reach a Fs content of between 35% and 40% further anomalies arise. In three rocks (7968, 7965, and 7961) the “course” drops steeply and passes wver the border (2V = 30°) into the pigeonite zone, These pigeonites have properties merging perfectly into the subcalcic augites. None of this type, haw- ever, has been found to gradually transgress the line (2V == 0) inte the field of pigeonites with the optic axial plane £ (O10). Walker (1943, 518) has recorded sinilar eases and consillers such pigeonites are “ferriferous and hie just below ie Terro-augite field of the triangular diagram.’ He suggests that such pigeotiles (optic axial plane | (O10) have 2V tot much below 32°. In these racks, however, 2V as low as 13° was measured (in 7965) and in 7970 all grada- tions exist from ZV = 20" (1.010) through 2V = 0° ta 2V= 40° (|| (010) ), Ja 2/50, on the other band, the course of erystallisation proceeds to Ts content of 409% and the subcaleic augite is joined by w pigeonite with 2V = 20° and optic axial plane 1 (O10). The reason is unktiown for such a marked break hetween the two clinopyroxenes in 2450, whereas a perfect gradation exists in Nos, 7961, 7965, and especially in 7968. All are olivine dolerites and only No. 7961 contains any bronzire, In 7263 there is an “early as well as a “late” pigeonite (see above). The warly pigeomite in this rock, however, is rare, and is always found enclosed by ihe augite (2VW=-=46"). Its optic axial plane is 1 (010), and there is con- sequently @ sharp line of demarcation hetween the two pyroxenes, ‘I'bere is no bronzite in this rock, probally because the cooling was sufficiently rapid to pra- hihit the inversion of the early pigeonite, Olivine (Ta 21%) began to form early, continued throughout the cooling and is plentiful im the final crystallisation as sal! ophitic sphenoids (of much the same composition as the earlier material) ald in association with the late pigeonite and some subcatcic augite. In 2432 the crystallisation is interesting. Here (as in 7965) a marked break is uoled between the early pigeomite (2V—=22°-14°; optic plane 4 (010) ) and the other clinopyroxene. But the olivine (Fa 11%) is more iuaguesic. has already finished its main crystallisation and is found ag rounded aml partly corroded crystuls in the pigeonites (and elsewhere), It does not tppear as a later erystallisation. Moreover. the zoning of the clinopyroxene cher Huin pigeonite is unusial. The early pigeonite gives way abruptly to a subcalele augite (2V = 40°), which is gradually zoned till borders of augite are reached (2V~<46°). This suggests that this rock shows 4 course of erystallisa- tion comparable (though farther from “Wo") with that to the left of the majar “hump” of the curve of Hess. No, 7970 18.4 most interesting rock. Crystals of clinopyroxene occur showing a remarkable zoning. The core, which sometimes encloses and enstatite is < pigeonite (2V = about 20"; optic axial plane 1 (010)). Then follows a per- lect gradation through uniaxial pigennite (2V—0°) into pigeonite (2V—~ 0-40"; optic axial plane |! (010), and finally itoasubealcic augite (2V = 30°-40°). Hrongite (OF 16-22%) is pletitiful in che slide and in places is rimmed with sub- caleie augites (2V=35-40°). ‘The bronzite shows a fine twinning (|| optic arial plane (010) ), and a shadowy extinction which suggests some exsolution of lime. It is dificult to explain the presence and apparent stability of the early pigconite which has its optic axial plane 1 (O10) and rims enstatite, Mveasure- ay ae, ~ \ 2V252° 4010 A= Se 2V:30° ~~ — —_—_—, (a) No. 7968. Augite strongly zoned throwgh subcalcic augite to rim of pigeouite Gwith optic plane || (010)). (x 45). (c} No, 2450. Augite (2V — 48") zoned outward toward subcalcic atgite (2V = 37°). A break eccurs and pigeonite with optic plane orien- tated differently (i.e, 1(010)) forms tips on the main crystal. (x 28). 198 (b) No. 2453. arly pigeconite with optic plane 1 (010)) enclosed by subcalcic augite (2V = 40°), which is itseli zoned toward an augite rim (2V = 46°). The pigeonite is not as ophitic as the other clinopyroxene, (x 20). 2V=/0°Loo ~~ % oe mS. 2Ve27°L010 | 2Ve38" (d) No, 7970, Pigeonite (with optic plane 1 (010) yery strongly zoned toward uniaxial pigeonite, and thence into a pigeonite with a different oricntation (z¢., optic plane || (010) and 2V=30°. A gradation exists near the rim fram this pigeonite into a subcalcic augite (2V— 40). Magnetite dust was precipitated while subcalcic augite (2V = 32) was crystallising. (x 35), Pig. 3. Vigeonite Relations in Soni: Dolerites (see also fe. 4 and Table 1) 199 ments of refractive index have indicated that this zoning shows a cottse crystallisation fairly close to the diopside- clinoenstatite line and toward the extreme left of the two pyroxene field of Iless (1941, 586)- p! ar <> ee te 2s nok / ae f /PIGEONITE [iio] / COURSE OF CRYSTALLISATION BY HESS - Lae PIGEONITE [ | 010) EN Vig. 4 Trends in crystallisation, as shown by the Clinopyroxenes of certain Pigeonite-bearing Dolerites, compared with the course af crystalli- sation suggested by Hess [Pyroxenes roughly plotted aguinst y-curyes (Deer and Wager, 1938, p. 20, fig, 2) and approx, 2V curves (Poldervaart 1947 (a), p. 161, fig, 2 J. Tu the non-pigeonite-bearing dolerites (e.g., 7963) a bronzite often occurs which shows a patchy extinction or fine lamellar intergrowth, apparently identical to that described elsewhere in orthopyroxenes within the range OF 15-35 (Polder- vaart, 1947, 167, and pl. XIII, fig. 1). Hess seems to limit this phenomenon to arthopyroxenes with a composition about OF 30 or greater, In these dolerites th« first appearance of this apparent inversion [rom early pigeonite seems to be 1m a bronzite with OF =16 (as in 7970) and is found in all more ferriferous varieties to OF 22, In the coarse rock, No. 7964, much of the bronzite as mag- nesic as OF 12, shows a prominent patchy extinction. Owing to the narrowness of the dykes the strange zoning phenomena shown by the clinopyroxenes cannot be explained by a gravitational settling of phenn- erysts into deeper regions, especially in view of the fact that zoning is normal 1) both orthopyroxenes and plagioclases, REFERENCES BAsepow, H. 1905 Geol. Report of Couiitty traversed by S. Atist, Gavern- ment North-west Prosp, Exped., 1903. Trans. Roy. Soc. 5. Aust.. 19, 57 CAMPBELL, W. D. 1906 Geol. of Norseman District. W-. Aust. Geol. Surv. Bull, 21 Cnunona, K. 1933 Determination of Feldspars in Thin Section, london Deer, W. A., and Wacer, L. R. 1938 Two Pyroxenes included in Systeu: Clinoenstatite, Clinoferrosilite, Diopside and Hedenbergite, Min. Mas.. 25, 15-22 Derr, W. A., and Wacer, L. R. 1939 Olivines from the Skaergaard Intru- sion. Amer. Min., 24, 18-25 Emmons R, C. 1943 The Universal Stage. Geol. Soc. Amer, Mem. 8 nnn ee Tee vv" OLIVINE PIGEONITE - - - [Fa, 11] j [2V = 14—22°; 4 (010)] 7 att No Bronzi ave - + SUBCALCIC AUGITE ——> AUGITE o Btanats (2V = 40 —45°] [2V = 46°] III I Burin” AP, Stak ee ee ee ee as Ee eee OS ee oe AT [Fa. 21] PIGEONITE AUGITE ———> SUBCALCIC AUGITE + -»- + «= « it wero, | =o amas Scie ‘ PIGEONITE-~ - [2V = 30 —13°; || (010)] NN — ———— ee rerreoq_ mF OLIVINE - - - - - - = ft ae “ - =~ ~~ < - 4 (Fa. 13] AUGITE ——>SUBCALCIC AUGITE - - - - - - - . 2450 [2V = 49—45°] [2V = 45 —37°] No Bronzite PIGEONITE - wt le [2V = 22 —15°; 1 (010)] a ee ee ee ee ee OLIVINE - - - - - - - - - [Fa.24] AUGITE —— > SUBCALCIC AUGITE - - - - 2 ? 7968 [2Vv = 52 —45°] [2V = 45 —30°] No Bronzite « 5 a -[2V = 30 —19; || (010)] - PIGEONITE - SS OLIVINE - - . - - p [Fa. 19] 7966 PIGEONITE ——> SUBCALCIC AUGITE No Bronzite {[2V = 28; || (010)] [2V = 38 —40 —35; ||(010)] nn esos 00—0 SS OLIVINE . - Ps - 7 ap be - ? , - — - - “ [Fa. 12] BRONZITE - - - - - = oY wom £2 (OF. 17 — 22] 7961 SUBCALCIC AUGITE- - > = [2V = 40 —30] - - = PIGEONITE - [2Vv = 30—25; || (010)] ENSTATITE ——_——_> - BRONZITE- - . (OF, 9— 10] [OF. 16— 22] 7970 PIGEONITE ——> PIGEONITE ——> PIGEONITE | - - No Olivine (2V = 20 —0; 1 (010)] {[2v = 0] {2V = 0 —30; || (010)] - - - SUBCALCIC AUGITE - [2V = 30 —40; | (010)] Tage I. Order of Crystallisation of Ferromagnesian Minerals in certain Pigeonite-bearing Dolerites a ROCK NUMBER 7964 7963 2449 2460 7960 7961 7968 7966 2453 7965 2450 7967 7970 7971 7972 a NE OO SSS SR _ ———eEeE=E=—OOooeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeaeeeEeEeeeee—e—eamamaaBapEeEeEeEeEeEeEe=eEeEeeEeEe=eeeeeeEeeeeEe—_eeeeeeeeeeeeeee nn SS eS eee. PLAGIOCLASE 33-9 34:3 54-2 43 64-1 61:4 42-5 58-1 59-7 55+4 55 28 57°6 45-0 51-3 PYROXENE 50-2 45-6 29-7 20 23-4 30°7 48-1 19+3 35-2 32-5 30 49 34:7 37-0 41-3 OLIVINE 8-0 16:1 11°8 15 5:7 1-9 6°6 6:3 2:2 8-0 ‘ 12 20 — _—~ =: ACCESSORIES 7:9 4-0 4°3 4 6:8 6-0 2:8 16-3 2-9 4-1 3 3 7-7 18-0 7-4 TEXTURE POIKIL SUB-OPH. SUB-OPH. SUB-OPH. OPH. OPH. POIKIL. SUB-OPH. OPH. OPH. OPH. “PLUMOSE” OPH. OPH. SUB-OPH. een ee oe . PLAG. An. % 67 — 65 82 — 72 — 67 75 —72—44 72 70 — 68 — 28 72 — 62 — 30 67 65 — 34 62 72 — 68 — 49 82 — 77 —46 67 —65 72 — 66 — 50 66 —62—35 91—72—31 Ay. Diam. 4X 15 mm. 9X +2 *8 X .25 1X +4 9 X +2 1:4 X 33 2 Xl 1:4 X <6 1 xX +25 9X +2 1:3X +4 “4X +03 4X +5 12% +4 ase eg 2V 51 — 48 50 — 48 — 35 50 —44— 37 45 — 40 43 — 38 — 30 40—31 52— 47 — 31 30 — 40 — 35 40 — 46 46 — 42—31 49 — 45 — 37 46 30 — 40 40 — 35 39 — 34 AUGITES and ZAC 37 — 36 42 — 39 — 38 38 — 39 39 40 36 37 36 39 38 41 38 37 32 41— 38 40 SUBCALCIC R.1I. 1:695=y 1-701 = y 1692 =y 1-698 =y 1-708 = y 1:710= 1:705 = 1-710 = y 1-680 = y 1:724=y 1-695 = y 1-698 = y 1:705 = y 1-719 = y 1:724=y AUGITES (2V = 48) (2V = 45) (2V = 44) (2V = 43) (2V = 38) (2V = 36) (2V = 47) (2V = 43) (2V = 43) (2V = 40) (2V = 45) (2V = 46) (2V = 35) (2V = 38) (2V = 36) Pleoch Nil Nil or Nil Nil, to X = faint green Nil or weak X = faint rust pink X = faint pink X = faint pink Nil Nil X = faint pink Nil to X= faint fawn X = faint pink cf. bronzite X = pink cf, 7961 Y = pinkish Y = faint fawn Y=Z=colourless Y—=Z-=faint green Y=Z=faint green Y = faint pink fawn Y = faint fawn Z= faint green Z= yellow green Z= faint fawn grey Z= faint yellow fawn Z=fawn grey Av. Diam. 1-4 X -8 m.m, "6 X +3 6X +5 6X +4 “4X +4 +35 X +25 “5X +4 4X 3 2:5 X 1:5 1-6 X +6 ete. 1:2X1 +2 X +2, ete, +3 X +25, etc. 1:2X <8 *8 X +6, etc. PIGEONITES 2V 30 — 22 30 — 19 28 — 30 22— 14 30 — 16; 19 22 28 —0; 0—30 ; ZAC a — — — _ 37 30 39 39 — 38 29; — 38? _— 30 RIL. Pi =y 1-720 = y 1-720 = y 1-682 = y 1-730 =y y = 1-712 y=1-685 [2V=15 1(010)] — —_ Pleochr. (2V = 25) (2V = 22) (2V = 29) (2V = 16) (2V = 22, || (010)] (2V = 22) y =1-698 [2V=25 || (010)] Optic cf. augite X= Y=V. faint pink cf. augite X = Y = faint pink cf. augite Nil cf. augite plane Z = colourless Z=faint green , || (010) || (010) || (010) 4 (010) || (010); 4 (010) 4 (010) 1 (010; || (010) Av. Diam. 2X +1 ‘3 XK +2 3X 2 "6 X +2 “3X +2 “4X +2 4X 3 ORTHO- PYROXENES Comp. OF. 8— OF. 16 OF. 22 OF. 22 — OF. 26 OF. 18 —OF. 22. OF. 18 — OF. 22 OF. 9 — OF. 21 2vV + 83 to —81 —71 —72 to — 66 —77 to —71 —77 to —71 +84 to —73 Pleoch. Nil to X = pink X=rusty pink cf., 2449 cf. 7961 X = pink — — = —_ — — X = pink —_ — X =rusty pink Y = faint fawn Y = pale yellow S Y = faint green Y = faint green Z = faint green Z = faint green Z= faint green Z=pale green Z=pale green Av. Diam. 4X +3 mm. 5 XK +3 1X <6 4X +4 6X +3 8 X +4 OLIVINES Comp. Fa. 15 Fa. 18 Fa. 21 Fa. 12 Fa. 12 Fa, 12 Fa. 24 Fa. 19 Fa. 11 Fa. 21 Fa. 12 Fa. 11 — — — 2V — 88 — 87 — 86 + 89 + 89 +89 — 85 — 87 + 87 — 86 + 89 + 88 Av. Diam. 1 X +7 m.m. ‘SX +3 8X +6 8X +5 2X +3 “4X +3 6X +4 *25'X +2 5x +4 -3.X +25, ete. 6X +4 4X 3 ANORTHOCLASE % 5-2 Nil 2 2 2 4 Nil 4—5 0-4 Nil 0:2 Nil 2 3 2 2V —51 to —59 —42 to —52 —52 —52 —50 TABLE 3 Etc. Chemical Analyses 1 2 3 4 5 1 47-98 48-61 49-35-4780 = 49-27 Q - 0-90 0-82 063 1:23 2-13 OR 2:22 13:79 13-35 13-15. 14-30-1357 AB 0-78 1-72 096125247 AN 8-89 9-13 «967, «12-09-1278 wo DI jEN 0:16 0-14 O15 0-40 050 les 14-02 12-71 13-31 7-40 426 (*s HY : 11-27. 9-67, 90110017, 851 FS 1:77, «2:02.14 254 244 Pes Ne 0-39 0-44-1513 060142 FA 7-53 0-31 1:06 = 0941-87055 MT 1-16 0-02 0:22 0-09 0-44 0-22 IL 1-67 0-12 an 0-10 = 1-23 AP 0-34 0-05 = Nil 4 +. PY 0-12 0-19 0-24 = Nil oo ee CR 0-22 Norms 2 3 4 5 2 - 35 264 2-44 668 3-89 BH 1697 9-65 21-48-2044 26-05 27:42 25-58 21-96 9-12 7-20 ) 5-91 4-544 20-72 | 10677 2-59 2-21 | | 10-80-1794 5-02 | 24:01 4:75 8-76 10:54 7-47 | 17-05 = 510 406 2:55 1:39 1:86 3-71 1655 1919 22B 4-10 _ 0-24 _ 2-69 0-33 - L. w Nil = Nil = — a 0-06 0:22 0-24 034 Totals 100-64 100-19 99-85 100-38 99-69 Olivine Bronzite-bearing Dolerite, Rock Hole, 19 miles E. Erna- bella, Musgrave Ranges, Anal., A. F. Wilson. No. 7963. Olivine Dolerite, Kranskop, Orange Free State (Frankel, 1943). Olivine Dolerite, Filabusi Dist., S. Rhodesia (Ferguson, 1934). ete Norseman, W. Aus, Anal., E. S. Simpson (Campbell 1906, p. ; Quartz Dolerite, Mt. Holmes, W Aus., (Farquaharson, 1912, p. 49). Anal., Survey Lab, Modes* (by Volume) 1 2 Feldspar 34-3 32 Late Feldspar — Pyroxene 45+6 42 Olivine 16:1 14 Accessories 4-0 *Modes of 3, 4 and 5 not available. Traus Roy Soc. S. Aust., 1948 Vol. 72, Mate XVIIL Go No. a7. Olivine (a l1). showingr (bh) New Ar “SPlumaoase” pyroxene. in the dolerite wiairked skeletal and “hexaonal’ This is abtinedant Tnelu whith contains the “skeletal” olivine. The acicular crystals may be seen “sprouting” from larger clinopyrox- enes toward the base of the photo. Ordinary liwht, x 60, fori tinder erossed nicols. sions vaindy very fine pyroxene. x Al (ye LOA). - 2 " (c) No. 7972. Plagioclase, strongly (d) No, 7964, Clear anorthoclase (fig. zoned from Anorthite (An91) to 2) rimming labradorite (fresh Andesine (An35). Subcalcitc augite but with abundant dusty inclusions). is other mineral. Crossed nicols, Biotite, rimming magnetite, occurs x 40, at bottom and top left. Long apa- tite needle included in anorthoclase. Pyroxene at top is euhedral augite; and that on right and left is enstatite rimmed with bronzite. Ordinary light. x40. 200 arouAnarson, R. A. 1912 Petrological Contributions to Geology of Western Australia, W. Aust. Geol. Surv,, Bull. 45 Fercuson, J. C. 1934 The Geology of Country around Filabusi, Insizwa JDis- trict. Geol, Surv. S. Rhodes, Bull. 27, 64 FRANKEL, J. J. 1943 Studies on Karroo Dolerite: (2) Some Younger [ntru- sions of Olivine Rasaltic Dolerite. Trans. Geol. Soc. 5. Africa, 45, 1 linss, H. 1941 Pyroxenes of Common Mafic Magmas, Am. Min., 26, 515-535 and 573-594 Jacx, R.L. 1915 The Geology and Prospects of Region 10 South of Musgrave Ranges. Geol. Sury. of S. Aust., Bull. 5 Jonansen, A. 1937 Petrography. Chicago PotnervAsrt, A. 1947 (a) Subealeic ferroaugite from Mount Arthur, [asi Griqualand. Min. Mag., No, 198, 159 Powpervaart, A, 1947 (b) The Relationship of Orthopyroxene to Pigeomte- Min. Mag., No. 198, 164 Srretcr, V, 1892 The Country between Everard Range and Barrow Range, and between Barrow Range across the Victoria Desert to Fraser Range. Trans, Roy. Soc, S. Aust., 16, 74 Tannot, H. W. B., and CrArkr, E. de C. 1917 Geological Reconnaissance of Country between Laverton and the South Australian Border, W, Aust. Geol. Surv. Bull. 75 Tare, R., Warr, J. A., and Siti, J. 1896 Horn Lxpedition to Central Aus- tralia. Geological Report. Melbourne Tiromson, J. A. 1911 Rock Specimens from Central and Western: Australia, collected by Elder Scientific Exploring Expedition, 1891-1892. Trans. Roy. Soc. N.S.W., 45, 292-317 Tomita, T, 1934 Variations in Optical Properties, according to Chemical Com- position, in Pyroxenes of the Clinoenstatite-Clinohypersthene—Diopside- Hedenbergite System. Jour. Shanghai Sci, Inst., Sect. 3, 2, 41-58 Wacker, F. 1943 Note on Pyroxenes of Basaltic Magma, Am. Jnl, of Sci., 241, 518 Wanker, F., and Potnervaart, A, 1942 The Karrovo Dolerites of the Calvinia District. Trans. Geol. Soc, S. Africa, 44, 127 Witson, A. TF. 1947 Charnockitic and Associated Rocks of North-western South Australia, Part I: The Musgrave Ranges—an Introductory Account. Trans. Roy. Soc. S. Aust., 1948, pt. ti, 195-211 THE STRATIGRAPHY OF THE AITAPE SKULL AND ITS SIGNIFICANCE By PAuL S. HOSSFELD Summary During April 1929, the writer was engaged on a geological survey of the northern slopes of the Barida Range near Aitape, Northern New Guinea. A survey of the lower section of the Paniri Creek, on 26 April, resulted in the discovery of several human skull fragments. On the completion, at the end of 1929, of the geological surveys in New Guinea by the Anglo-Persian Oil Company, to which the writer was attached as a geologist of the Australian Commonwealth Government, the skull fragments were taken to Australia, and later deposited in the Australian Institute of Anatomy at Canberra. 201 THE STRATIGRAPHY OF THE AITAPE SKULL AND ITS SIGNIFICANCE By Paut S. Hossreip * | Read 9 September 1948] INTRODUCTION During April 1929, the writer was engaged on a geological survey of the northern slopes of the Barida Range near Aitape, Northern New Guinea. A survey of the lower section of the Paniti Creck, on 26 April, resulted in the discovery of several human skull fragments. On the completion, at the end of 1929, of the geological surveys in New Guinea by the Anglo-Persian Oil Com- pany, to which the writer was attached as a geologist of the Australian Common- wealth Government, the skull fragments were taken to Austtalia, and later deposited in the Australian Tnstitute of Anatomy at Canberra, In a report by the Anglo-Persian Oi] Company to the Commonwealth Government (Nason-Jones 1930) the discovery of the skull was referred to bricfly and the associated rocks and fossils described, Nothing more was done until Fenner (1941) described the fragments and included a brief description of the site as supplied by the finder, the present writer. The present paper has been written in order to describe fully the stratigraphy of the location and its possible significance to huinaty pre-history, The writer holds the opititon that the occurrence of human remains in Jleiste- cene sediments in New Guinea, and the investigation of the relation of these ileposits to the Pleistocene lee Age, may supply important evidence of the advent of Man in Australia The region which will be deseribed in some detail conrprises that part of ilorthers: New Guinta which Ties between the Dutch border on the west and the seitlement of Aitape on the east, and extends inland from the coast to the foothills of the Bewani aud Torricel Mountains, TOPOGRAPHY Within this region there exist three well-marked hill areas. The most westerly extends into Dutch New Guinea and includes in its northern portion the (Oenake-Bougainville Massif. The central arca extends inland from the Serr Hills on the coast to the relatively low hills between the Bewani and Torricelli Mountains, The most easterly area consists of the northern foothills of the Torricelli Mountuins and extends north towards Aitape, from which it is separated by a coastal strip of swampy lowland. Between these hill areas the country consists of large areas of alluvinni, con tuning a large proportion of sago swamps. The junction of the hill areas and alluvium is sharply defined and is probably vonsiderably less than 300 feet above sea-level. The hills rise steeply from the alluvium and attain heights of the order of 2,000 feet and over. The wholv region with its juvenile drainage atid sharp differentiation of hills and plains presents the appearance of an area from which the sea has receded recently. This view is supported by palacontological und other evidence. The western area of alluvium inchides the floodplain of the Newniayer River und its tributaries, while the eastern area contains the floodplains of the Bliri. Pier, Yalingt and Raila Rivers wnd smaller streams and their distributaries, * Geology Department, University of Adeltaidy, feuns Kea. Soc. &. Anst., 73, (2), 30 March 1945, 202 PART_OF_ NGRTHERN NEW GUINEA SCALE Q 2 Cates LEGEND RECENT —>F 7 ane — | ARV 7 PLEISTOCENE ALING! SERIES a maniliO GROUP WY AURELALE aes se o X SITE OF HUMAN SKULL FRAGMENTS BE WANI MTs. BASED ON MAP OF THE FIMSTH COAST ASTA APO.C.REPORT 51.420) 187 Fig. J GEOLOGY The three Inll areas consist of sediments belouging to the Lower Wanimo Group, Finsch Coast Series, the Upper and Lower Aitape Groups, and of the Oenake Series of igneous rocks. (Nason-Jones 1930.) The sediments range in age from Pliocene to Oligocene and possibly Late Eocene, while the igneous vacks are prohably of pre-Vertiary age. These Terliary sediments exhibit steep folding and have developed complex structures. lt is upon the eroded surfaces of these highly folded rocks that the vently dipping Pleistocene beds have been deposited. The Pleistocene beds have been divided into two grotps: (a) The Yalingi Series; (hb) The Upper Wanima Group, (a) Tar YALinar SERIES These consist of beds of coarse pebbles and boulders, cemented in part, with some interbedded thin argillaceous and areuaceous hands. They form a sub- horizontal or gently dipping unconforniahle capping on the older rocks and are developed chiefly in the Upper Nengo-Yalingi River area. Allbough subsequent 205 dissection bas broken the former continuity of this “sheet 1 ihany places, tt former extent asa fluviatile outwash depusit on a gently sloping relatively even surface can he recognised (Nason~Jones 1930), (b) Thr Upear WANIMO GRouP This consists of a calearcous and an argillaceous facies. Jn general, the leis af this group outcrop along the intersection of the hill arcas and the coastal allayial plains, and their inland continuations at elevations above sea-level vslimated in most instances not to execed 200 fect. The few isolated high level occurrences observed present anomalous features suggesting fault movements anil local elevation. The general picture is that of relatively thin deposits trmgmg ai uld coast line which is marked by the lower slopes of the hill areas deseribed above, Thu beds dip gently up to five or six degrecs away from these arcas and disappear under the alluvium of the plains. The soil cover and lusuriant yexetation limit the outcrops of these beds to the banks of streams. In these, the beds occur only near the junction of the hill areas and the plains. These beds exhibit no evidence of folding, the only earth movements in which they appear to have participate: being those due to faulting, which are local in character wherever observed, Asa rule, the calcareous facies ix developed wherever the rocks of the hill area consist of calcareous or igneotis material, and the argillaceous facies wherever the blue mudstones (bentonitic in part) of the Tower Wanimo and Finseh Coas, Series supplied the detrital material. The argillaceous facies in which the human remains were discovered consists of blue mudstones and siltstones interbedded with some more urenaceous beds and containing a few richly fossiliferous horizons. Current bedding was soted but is not a general feature. The beds are characterised throughout by the occurrence of broad zones containing partly carbonized driftwood and other hard plant remains distnbuted irregularly throuzh the zones, The abundance of this driftwood, the presence of currem bedding and lenticular beds, and the recognition of fossil shells of fresh snd brackish water and marine forms indicates the close proximity of a former shore line to a shallow sea, or alternatively, deltaic conditions. ‘The origin of these beds as deltaic sediments, except in restricted areas, is discounted hy their apparent continuity as deposits fringing the farmer coastline. The opinion that these beds originated under shallow water conditions in quiet backwaters. in which estuarine and marine deposits were tntcriningled, anv the author’s long association with anthropology, led to the close examination uf the beds for human and animal remains, when acl where this did not detract materially from the primary objective of the Survey, the geographical an geological mapping of the area for its petroleum potentialities. All such examina- tions of these Pleistocene deposits had therefore to be brief, and the author can- siders himself fortunate that despite the incompleteness of the examination, which should have oecupied months instead of the few hours that could be devoted tu it, the search was tewarded by the discovery of the human skull fragments. The examination of a section of these beds exposed ma cliff face forming the bank of the Paniri Creck just above its junction with the Kiyen Creek, showed a completely undisturbed layer of fossil shells dipping gently i a northerly direction, The close examination of as much material a5 was possible in the time available—about four hours—resnited in the discovery of the skull fragments and a carbonized fossil fruit resembling a coconut, and the collection of a number of shells and maternal containing Loraminiferal renmins. The skull fragments were abtaincd at a Tevel approximately four fect below the ereded upper surface of the formation, ot which a total thickness of 10 feet is exposed in this section. 204 The eroded upper surface of this formation is overlain by coarse gravel beds, twelve feet in thickness, which immediately underlie the subsoil, soil and primary forest cover.) SECTION OF BANK OF LOWER PANIRI CREEK o4 85 a4 5 " zo FEET ae | Tig. 2 The whole of this sequence is exposed in the cliff face forming the site of the discovery and is reproduced in the section included in the text. While a search of the site failed to disclose any other remains either human or animal, this search was by no means exhaustive. Further, the occurrence of oittcrops of this formation in numerous places, which could be given only a brief examination, as well as their recorded occurrence outside the areas surveyed by the author, suggest the necessity for their detailed exarnination for human and animal temains, an examination that was not within the scope of the petroleum survey during which the discovery was made. The skull fragments have beet described in detail by Penner (1941), and the associated fossils by the Commonwealth Palaeontologist, see Nason-Jones (1930), and by B. C. Cotton, see Fenner (1941). While there is some discrepancy in the two lists of species identified, due partly perhaps to the fact that two separate lots were examined, both wrilers agree that the fossil suite exhibits forms of fresh and brackish water and. matine species. The collection of fossils submitted to the Commonwealth Palaeontologist, whose identifications are quoted below (Nason-Jones 1930), included, it was ©) Through an error, for which the present writer is responsible, the information supplied to Fenner gave the thickness of the gravel beds as 6 ff. 205 believed, all the types represented, and was the main collection. The collection retained by the present writer and examined later by Cotton (Fenner 1941) con- tained, it was thought, only such types as were duplhieated., The following fossils, obtained from this site hy the present writer, are listed by Nason-Jones. (1930) PELECYPopa— Arca nodosa Puphio afl, alba ef, Eryeinra sp. Placenta mandirantfoncnsis (74STEROPEDA—— Néritina sp. Melama denisoniensis Melania woodwardt Melania seabra FoRAMINIFERA— Cibictdes praecinetus Icterostegita sp. Cristellaria orbicularis Operculinela venosa Cristellaria cultraia Operculina granulosa Cristellaria calcaratu Palystomella craticulata Lpamdes lumidus Owngueloculing lamarckiana Eponides procera Rotalia sp. nov. Epistomina eleguns Roatalta schroeteriana Globigerina triloba Rotalia papillosa Cotton (Fenner 1941) lists the following fossils in the lot examined by him, which too was collected at the site hy the present writer — “lrca (Tegillarca) granosa Linn. Cyclophorus sp. Telescoptum fuscum. Schumacher Melania ci. guncea Lea Papuine sp.; Land Shell AMelania cf. recta Lea Nerttina cornea Linn. Melanw cf. canalivulata. Reeve Neritina sonzerbianda Montrouzier Cyrene coarans Gmelin Laoma sp. The fossil fruit was submitted to Professor T, G, B, Osborn, at that tinie Professor of Botany at the University of Sydney. A report on the examination was forwarded to the writer and the following is a summary of Professor Osborn’s conclusions, “Morphologically this palm fruit closely resembles that of Cocos nuciferer, the coconut, Its drupaceous unilocular character places it close to Cocos in the Cocoineae section of the Paluiaceae, Its large size is also characteristic of the fruit of the coconut palm, Anatomically it also shows a strong resemblance to Cocos nucifera.” As stated above, the argillaceous deposits included in the Upper Wanimo Group occur on the lower slopes of the hill areas near their interscetion with the alluvial plains, and he unconiormably upon the eroded surfaces of the steeply folded rocks ranging from Pliocene to the older Tertiary in age. In the Barida Range area they were observed in all the streams draining that range, both on the northern and seaward slopes as well as on the southern slopes facing the Mene-Bliri River floodplain, and parallel to the valley sides af those two streams. They were recognised also ou the southern and inland slopes of the Oenake Range facing the Pual-Neumayer Plain, and to a less extent on the northern and seaward. slopes of that range. DISCUSSION A review of the literature dealing with the island of New Guinea indicates that argillaceous and calcareous deposits referable to this formation have been observed in many localities presenting similar topographic features. These 20 Meistocene deposits will he telerred ta in detail ina paper which is in course of preparation. Various observers are in agreement that these deposits and the enormous areas of juvenile drainage characteristic of the island of New Gajnea indicate recent emergeuce of the land and hence elevation. [t ig Leue thal mumerons localities are known where differential movement has taken place. How- ever, the accurrence at intervals around the island of coustal plains with thet? abrupt fransition to hill areas, and the wide distribution of shallow water depusiis Of muds and silt fringing the hill areas, would necessitate, if clevatiun of the lanel be postolated, the upward movement of the whole of the island of New Gries is a block. The writer is unable to aceept this hypothesis and believes that alternations in sea-level supply the correer explanation for the regional character af these lopographic and stratigraphic features. Alternatious i sea-level have been due to various causes. In discussing marine sediments and features of Pleistocene age, however, the effects of the generally accepted fluctuations in sea-level during the Pleistacene Ice Age merit first vonsileration, In the writer's opinion sich fluctuations can and da explain satisfactorily the topographic and stratigraphic Leatures discussed above. Tt is suggested as probable that the mpper limit of transgression of the beds of the Upper Wanimo Group, where the heights ahove sea-level af such lintits ire consistent over a sufficiently large area, marks the upper limit of transgression ot the Mieistocciie Ocean prior ta the commencement of the First Glacial Period, IE this is accepted, the first retreat marks the commeneement of the Pleisto: cene Ice Age, In the locality discussed in detail, fluctuations and subsequent advances and retreats. of the sea are maskerl by the extensive alluvial deposirs lIowever, the observed erosion surface on the upper surface of the argillaccons facies of the Upper Wantimo Group and deposition thereon of wrter-sorted gravels suggest the return of the sea to its original limits, and such return could perhaps he correlated with the First Interglacial Period, Tt might be argued that the postulated successive advances and recessions of sea-level during the Pleistocene Ice Age could account for the varied environment of the fossil tauna of the Panirt Creek deposit, The deposit, however, exhibits no evidence of any break in deposition from ils base where it rests unconformably on the Pinsch Coast Series, to its upper eroded surface on which lie the gravel and boulder beds. Further, although the fogstl shells and foraminifera exhibit a varied environment, all of them were obtained frant the narrow zone indicated in the accompanying section. It is obvious, therefore, that if atieeessive advances of the sea occurred, sea-level rose ta ifs amginal height oly once, at the stage during which the upper parts of the deposit were eroded, or alternatively, if a uumbey of advances of the sea occurred, all evidence of previous transgressions except that of the last two has been removed completely. If such were the case, then the deposits in question could have been lait down at a date stage of the Pleistocene. Further research may permit of a decision being made, but if the latler supposition should prove to be correct, then the writer’s view that these deposits date back to Early Pleistocene obviously caunol be mailained. The writer desires to pomt ont, however. that should it be possible to correlate the inital retreat of the sca due to the commencement of the Pleisto- cene Ice Age with the upper limits of the Waninio Group, the occurrence of tutian remains with Australoid affinities in these heds would prove the existence of Man m New Guinea poor to the cammencement of the [ce Age. Tt would follow that Man, even if he had no knowledge of sea travel, could have entered Australia as sonn as the increase in the amount vf glaciation had lowered sea-level sufficiently for hint ty crass the present Vorres Strait uver narrow and shallow 207 stretches of water. Even should further research place these beds higher in the Pleistocene sequence than postulated by the writer, their dating, 1f this be found possible, will have an important bearing on the date of the entry into Australia of Australoid Man, The emphasis placed on the argillaceous facies rather than on the calcareous facies of the Upper Wanimo Group is due to the greater amount of information supplied by the former and the difficulty in many instances of determining by field examination the age of the caleareous deposits. A stndy of both facies will be necessary if further work in the directions indicated is undertaken. The widespread occurrences of the argillaceous beds. however, and the pos- sible occurrence in them of animal and human remains, suggests the importance of a detailed examination of all the accessible deposits of this formation not only in New Guinea, but also the search for them in adjacent islands, SUMMARY This paper describes in detail the occurrence inland from Aitape, in Northern New Guinea, of fossil human skull fragments embedded in and contemporaneous with deposits of Pleistocene Age. It the writer’s opinion, it may be possible to correlate the upper beds of these deposits with the commencement of the Pleisto- cene Ice Age and to date them therefore as Early Pleistocene, but much more research is necessary before stich a correlation can be accepted. It is pointed out also that the widespread occurrence of such deposits in New Guinea (and probably in adjacent islands) offers a promising field for such research, Any positive evidence which would assist the dating of these deposits would affect materially the views held of the time of Australoid Man’s entry to Australia. REFERENCES Nason-Jones, J. 1930 Geology of the Finsch Coast Area, North-West New Guinea, in the Oil Explor. Work in Papua and New Guinea by the Anglo-Persian Oil Co., on behalf of the Comm, Govt. of Aust., 1920- 1929 Fenner, F, J. 1941 Ree. S. Aust., Mus., 6, 4 ON SOME REPTILES AND AMPHIBIANS FROM THE NORTHERN TERRITORY By ARTHUR LOVERIDGE Summary In herpetological literature of a century ago Port Darwin and Port Essington frequently appeared as type localities of considerable importance. Since those days relatively little has been added to our knowledge of the herpetofauna of Australia’s Northern Territory. It was, therefore, with considerable satisfaction that the Museum of Comparative Zoology at Harvard University received part of the collections made during 1944 and 1945 by Mr. T. R. Tovell of the Australian Imperial Force, particularly so as it contained half-a-dozen species unrepresented in the museum’s collection, of which one — Typhlops tovelli — had to be described as new. 208 CN SOME REPTILES AND AMPHIBIANS FROM THE NORTHERN TERRITORY By AgtHur Loyeripcn +) (Communicated by H. Womersley ) In herpetological literature Of a century ago Port Darwin and Port Essing- ton frequently appeared as type loculitics of considerable importance. Sihce those days relatively little has been added to our knowledge of the herpetofauna of Australia’s Northern Territory, It was, therefore, with considerable satisfaction that the Museum of Comparative Zoulogy at Harvard University received part of the collections invade during 1944 and 1945 hy Mr. T. R. Tovell of the Aus- tralian Imperial Force. particularly so as it contained halt-a-dozen species un- represented in the museum's collection, of which one—Typilops tovelli—had to he described as new. Unfortunately, the data accompanying the first consignment was not toa precise. Subsequently, Mr. Tovell kmdly supplied me with the following mformation about the localities which had originally been stuimarised as “near Darwin”. They are Batchelor, at about 60 miles south of Darwin; Werrima and Knuckey's Lagoon, about ¥ miles: Koonewarra, about 7 nvles; and Noonamah, ahout 24 miles south of Darwin. TyriLops rover. ( Loveridge) Typhlops lovelli toveridge, 1945, Proc, Biol. Soe. Washington, 58, LI1: Koonowarra Sports Ground, Northern Territory, Australia. 2 (MACLZ. 48844-5), Koonowarra Sports Ground, Midbody scale rows 20; snout rounded, nasal cleft proceeding from preocular. Diameters included in total length 36-40 times. Total length, 122 (118°5 + 3:5) min, TyPniLors GUENTITERL Peters Typhlops (Onychocephalus) quenthert Peters, 1865, Monatsh, Akad. Wiss. 4erlin, 259, pl. ——, fig. 1: Northern Australia. 1 (M.C.Z. 48843). Batchelor. Midhody scale-rows 18; snout rounded; nasal cleft proceeding trom second labial, Diameter 2°75 mm., included in total length 63 times. Total length 175 (172-5 + 2°5) mm. This blind shake appeais to be closely related to 7. wiedii Peters of Brisbane, Queenslatid. U1 would be interesting to know whether this black-tailed species carries the tail upraised Wke a false head, aiter the manner of the Asiatic Mulicora, the African Chilorhmophis, the American 4postolepis, ete. NaATRix MAIRN Marten (Gray) Tropidonotus mairii Gray, 1841, in Grey, Journ. Exped. West Atistralia, 2, 442: Australia. 11 hatehings (M,C.Z, 48851-61), Winnelite near Darwin, Midbody scale-rows 15; ventrais 136-146; anals 2; subcandals 56-61; upper labials 8, the third, fourth and fifth entering the orbit. except on right side of M.C.Z. 48858 where third and fourth are fused resulting in 7 labials; lower lahials 8, the first five in contact with the anterior chin shield; preoculars 1, ©) Museum of Coimparative Zoology, Cambridge, Mass., U.S.A. Heaney, Roy. Suc, S, Aust, 72, (2), 30 March 1949, wc 209 except ou left side of M.C.Z. 48859 where there are 2; postoculars 3, Total feneth about 187 (147 +40) mun These cleven young were taken from a batch of twelve eggs, one of which had already hatched, found about 20 April, 1.0, at the begining of the dry seasai, beneath a pile ef rubbish. When found some eggs were single, others slightly cemented together in twos or threes. They measured approximately 25 x 16 Wu, (T.R.T). _ Trinomials are necessary since the separation by Rrongersma (1948) af a well-defined race in Dutch New Guinea. CERBERUS RYNCIOVS AUSTRALIS (Gray) Hamalopsis australis Gray, 1842. Zool. Mise., 65: Port Essington, Norther Territory, Australia. 9 @ (M.C.Z_) 48846, 48862), Fariny Bay about 14 miles trom Darwin. Midbody scule-rows 23-25; ventrals I43-l44; anals 2: suhcaudals 44-51) nostril cleft in contact with second labial; wpper labials 8-10; separated fron orbit by suboculars; 3-4 lower lahials in contact with an anteriwr chin shiel. Larger ¥ (M.C.Z. 48862), 587 (486 + 101) mm. (ne was found lying at the bottom of a salt pan (T. R. T.). That Cerberus, and not Hurria, is the correct name for these water snakes was pointed out by Malevltu Smith {1930), and that rynchops, not rhynchops, was Sehneider'’s original spelling by Loveridge (1948) when describing a new race and providing a key to the gertus. ASPIDOMORPHUS CHRISTIEANUS (Fry) Pseudelaps christicanus Pry, 1915, Proc. Roy. Soc. Queensland, 27, 91, fig. 6: Fert Darwin, Northern Territory, Australia. 9 {M.C.Z. 48847), near Darwin. Midbady scale-rows 17; ventrals 195; anals 2; subeaudais 47> labials 7. third and fourth entering the orbit. Total length 350 (300 + 50) mim, This gravid ?, which carrics three eggs each measuring about 26 x G mm.. has more ventrals and fewer subcaudals than the topotype ¢ already in our collection. That Pstudclaps of Duméril is a synonym of Aspidomorphus has been shown by Brongersma (1934, Zool. Med. Mus. Leiden, 17, 224). © DEMANSIA PsAMMOPIIS (Schlegel) Elaps psummophis Schlegel, 1837, Phys, Serp.. 2,455: Australia. Elope- cephalus ornaticeps Macleay, 1878, Proc, Linn, Soc. N-S.W., 2, 221: Port Dar- win, Northern Territory, Australia, &? (M-C.Z. 48848-9), Batchelor and Rerrima. Midbody setale-rows 15; ventrals 180-181; anals 2: subsaudals 70 (2) -91 (3) pairs; upper labials 6, the third and fourth centering the orbit. Total length of ¢ (M.C.Z. 48848), 290 (252 4- 58) mm. The head of the young ¢ is black ahove and scarcely distinct from the deep black nuchal bar; the body is fawn, each seale with a somewhat paler edge. The head of the older 3 is oliye with the markings described by Macleay. I follow Kanghorn (1942, 118). who has had the advantage of examining much mere material, ii relegating ornaticeps, of which these specimens are almost topotypes, to the synonyiny. Llowever, the fame proposed by Macleay was Flapocephatus, not dilapognalitus as cited by Kinghorn. Kinghorn’s conclusion appears to haye been based largely on the highly variable colouration, known tr change with age, IJ would suggest the possibility af a northern race with more numerous suheaudals for which the name alfvacca Gray, 1842, would be available. If the Australian Museum's material could be sexed and arranged geographically to supplement that furnished by Boulenger (1896, 322-324), while ignoring Boulentger’s arrangement based on colour, the point might be settled. 210 DEMANSIA TEXTILIS NUCITALIs (Gunther) Pseudonaja. uuchalis Giinther, 1858, Cat. Snakes Brit. Mus., 3, 2272 Port Essington, Northern Territory, Australia. é@ (M.C,Z, 48850), near Noonamah. Midbody scale-vows 17; ventrals 197; anals 2; subeaudals 63+ pairs; upper labiats 6-7 (left and yight), the inird and fourth entering the orbit. Total lengti af &, 1130 (940 + 190+) nim. HETERONUTA BiNOEL Gray Heterovola binoci Gray, 1845, Cat. Liz. Brit. Mus., 174; Lloutman’s Abrolhios, Western Australia, 18 (M.C.Z. 48801-6), Batchelor or Kerrima. Dorsal tubercles keeled, m 12-16 rows, usually 14; preanal pores of eight tales 4-5. Largest @ (M.C.Z. 48801), 105 (48 4-57) nm. By day these geckos hide under any Object not of ti Or iron, the heat of which is too great during the noon hors (T. R. TL). DmeLopacTyLus stniceres ciaAris Boulenger Diplodactylus ciliaris Boulenger, 1885, Cat. |.iz. Tyrit. Miis., 1, 98, pl. viii, fig. 2: Port Darwin, Northern Territory, Australia. Juy, (M.C.Z. 48807), near Darwin, Dorsal tubercles flat, Forming 2 ill-defined rows; no pores, Length, 53 (32-+.21) mm, OFDURA RHOMBIFER Gray Oedura rhombifer Gray, (844, Zool, Erebus & Terror, Rept. pl. xvi, fig. 6: Australia, 3 (M.C.Z.), near Darwin. Dorsals granular, small; femoral pores 12-+ 12, being separated in preanil region by five scales; tail depressed, aval. Length 87 (43 +44) mm. The shape of the tail conflicts with Boulenger’s redescription and conforms to what has been noted by Kinghorn (1942, 120). Since the separation of the African geckos under the name of Afroecdura (Loveridge, 1944), the range of Oedwra is restricted to the Australian region, GRHYRA VARIEGATA AUSTRALIS Gray Gehyra australis Gray, 1845, Cat. Liz, Brit, Mus., 163: Port Essington and Swan River, Australia. 3d @ (M.C.Z, 48864-5), near Darwin, Dorsals granular, small; preanal pores 14 in male; scansors not separated by a median groove, Length ot ¢, 124 (62 +62) mm.; 2, 109 (53 4-56) mm, Taken in au old! building at MeMillan’s (T, R. T.). Gehyra (part), Gray, 1834, antedates the use of Peropus Wicemann, 1835, for this gens. DiporiPHoRA BILINEATA Gray Diporiphora bilineuta Gray, 1842, Zaal. Misc., 34: Port Essington, Norther Territory, Australia. 5 (M.C.Z. 48808-11), Batchelor or Berrima. Gular fold absent; preanal pores 2 in male; tail twice the length of head and body. Tength af @ (M.C.Z. 48808), 179 (32. 127) mm.; 2 (M.C.Z. 48809), 166 (55 -+ 111) mim. Timaua scrncomes sctxcorpes (Shaw) Lacera scintcoides Shaw, 1790, i White, Journ, Voyage N.S.W., App, 242, pl. ———: New South Wales. Juv. (M.C.Z. 48817), Berrima, 211 Midbody scale-rows 36; anterior temporal as Jong us interparietal; forelimb shorter than head and contained about twice in distance from axilla to groin. Length, 154 (102 + 52) mm. In vicw of its small size the proportions of this. skink are interesting for com- parison with those of the New Guinea race—T. s. yigas (Sehueider). LYGOSOMA (SPILENOMORPHUS) TAENIOLATUAD TAENIOLATUM (Shaw) Lacerta tacuiolola Shaw, 1790, in White, Journ. Voyage N.5.W.. App.. 245, pl. xxxii, fig. Ll: New South Wales, 3 (M.CZ. 48818-20), Hatchelor or Lerrima. Midbody scale-tows 24-26; prefrontals separated, AJL three are jmmature- Owing to the findings of Malcolm Smith (1937, 213), Sphenemorphus and Leiolopisma are yelegated with some wusvivings to their former status of sub- genera or, as Smith prefers to call them. “sections”. LYGOSOMA (SPUENOMORPIIUS) FiscireRr Doulenger Lygasonia fischert Boulenger, 1887, Car, Liz. Brit. Mus., 3, 2283; na. for L. mucelleri Fischer (preac.), 1882, Arch, Naturg., 295, pl. xvi, fig. 16-19: Nicol Bay, Western Australia. 2 (M.C.Z, 48821-2), Batchelor or Berrima. Midbody scaic-rows 30; prefrontais separated; colouring characteristic Jength of 9, 143 (49+ 94) mm. Eneysted nematodes are numcrous on ¢xternal surface of stomach. LycosomMa (SPHENGMORPHUS) ISULEPIS ISOLEPS. houlenger Lygosome isolepis Boulenger, 1887, Cat. Liz. Brit. Mus., 3, 234, pl, xv, fig. 1: Nicol Bay and Swan River, Western Australia. 2 (M.C.Z. 48823), Batchelor or Berrima, Midbody seale-rows 30; lamellae beneath fourth toe 23, Length of ¢, 149 (72--+- 77) mm., but tail-tip regenerated. Agrecing in all respects with the typical form rather than with L. 2. forrest Kinghorn (1932, 358), this gravid @ holds four eggs measuring about 12x 7 mnt. Lycosoma (LEIOLOPISMA) PECTORALE (De Vis) Heteropus pectoralis De Vis, 1885, Proc. Roy. Soc. Old,, 1, 169: Warro, Port Curtis, Queensland. 14 (M.C.Z, 48826-36), Batchelor or Berrima. Midbody scale-rows 26-32; lamellae bencath fourth toe 19-27, in one specimen there are 24 on the right and 27 on the Icit toe. Largest ¢ (M.C.Z. 48826), 118 (41 + 77) min, In addition there were 13 damaged examples from same series or “near Darwin”, which were not retained. All but iwo of them were of the strongly keeled pectorule type, two others with dark throats represent the synonym munudum De Vis (1885), their dorsal scales being almost smooth vet faintly tricarinate dorso-laterally, A. large specimen was recoyered [rom the stomach of a Linliy burtoms. In the axilla of another of these skims were some mites (Troubicnla sp.n.), for whose identification Tam indebted to Mr. H. Womersley of the South Australian Museum. “Che species will he described in Mr. Womersley’s forthcoming monograph. Lycosoma (LyGosomMa) PUNCTULAYEM Peters Lygosoma punctilatum Peters, 1871, Monatsh. Akad. Wiss. Berlin, 646, pl. ——, fig, 5: Port Bowen, Queensland. 212 1 (M.C.Z. 48866), Winnellic, near Darwin, Midbody scale-rows 20; digits 5; toes 5; lamellae beneath fourth toe 14, Length 118 (44-74) mm. In life brown with a coppery sheen (T. R. T.). Lyoasoma (LYcosoma) PUMiLIUM Boulenger Lygosome. pumilian Boulenger, 1887, Cat. iz, Grit. Mus,. 3, 323: Cape York, Queensland. 2 (M.C.Z. 48834-5), Batchelor or Rerrima. Midbody scale-raws 20; digits 5; toes 5; lamellae beneath fourth toe 18-L9- Larger measures 91 (41 + 50) mm. ‘The alleged difference in relative size of nostril and ear-opening between pumilium and punctulatwm is uot apparent. LL. pumilinm seers to be closely related to crassicaundnim which Malcolm Smith (1937, 322) refers to his new section Jetiscincus, ABLEPHARUS BOUTONIT METALLICUS Boulenger Ablepharis bextonii var. metallicus Bouwlenger, 1887, Cat. Liz. Brit. Mus. 3, 347: North Australia. 4 (M.C.Z. 48337, 48867), Batchelor or berrina, Midbody seale-rows 22-24; lamellae beneath fourth toe 17-20. Largest only measures 93. (38 -+- 55) wm. In life iridescent grey with black markings. Found on trees and yiosts, not wider logs or in grass. A very active skink (T. R. T.). ABLEPTIARUS LINEQOCELLATUS LINEOOCELLATUS Dumeril anil Bibron Ablepharus linco-ocellatus Duméril and Pibron, 1839, Erpéi, Gen., 5, 817: Australia. 5 (M.C.Z, 48838-42), Batchelor ar Berrima. Midbody scale-rows 24-26; lamellae beneath fourth toc 17-18; supranasa!s absent. All young, the smallest only 29+ (14+ 15+) nin, its tail-tip missing. This little snake-eyed skink disgorged a spider. DELMA FRASER! FRASERT Gray Delina fraseri Gray, 1831, Zool, Mise., 14; Western Australia. 4 (M.C.Z. 48812-5), Batchelor or Berrima, Snout as long as, or longer than, the distance between eye and cat; frontonasals in 2 pairs; fourth labial below eye; midbody scale-rows 16; anals 3, except in M.C.Z. 48814, where the wedge-shaped central scale fails io reach the anal border. Largest measures 75 mm. from stout to anus, fail missing. All taken beneath rocks; quite common in this area (T. Rk. T.), Liauis nurtonts Gray Lialis burtonis Gray. 1834, Proc, Zool. Soc, London, 134: New Sottil Wales, 2 (M.C.Z. 48816), Berrima. Rostral twice as broad as high; upper labials 14; preanal pores 4; colour form punctulata, Length of ¢, 260 (200-++ 60) mm., bit tail regenerating. The oviducts of this gravid 9 held two undeveloped eggs ineasuring about 20x 12mm. In its stomach is a skink (Lygosoma pectorale)} measuring 49 mm. from stioul to anus, CycLORANA AUSTRALIS (Gray) Altes australis Gray, 1842, Zool. Misc., 56: North coast of Australia, ir, Port Essington, Northern Territory, Australia, 2 (M.C.Z. 26002), MeMillans, near Darwin. 213 These juvenile forms are so shrivelled by itnmersion in strong formalin that theic habits might be described as “slender’’, ic, in this respect referable tu aboguttatus (Gunther) of Parker’s (1940, 16) key, which differentiates the twa species as follows: Zygomatic process of the squatosal heavily sculptured und forming a broad suture with the maxilla, Habitus stout ... ww eRstralis Zygomatic process not sculptured aid separated from the maxilla, wr oonly very narrowly in contact with it. Tlabitus slender . = salt 1% i f dy iu vee Ubguthadies Rot Parker (1940, 20) is mistaken in referring part of my (1935, 13) albogutiatus to the synonymy of australis and suggesting that the frog (M.C.Z. 11647) from Alexandra (rot Alexandria), Northeru Territory, is really an vastralis, Vt is true that the frog was received from the British Museum in $925 as 'Phractops tustralis’ (presumably identified by Boulenger), but both in zygo- matic structure and colour pattern it agrees with alboguttalus. That the British Museum skeleton of another frog taken at Alexandra by the same collector happens to be qusfralis is interesting, for Parker records both species as occurrilig at Port Denison, Queensland. Despite their close relationship the two spevies are quite distinct, The larger frog measured 45 mm. and was taken in a ditch of stagnant water about six miles north of Darwin, the smaller was im sand behind the beach at Lee Point about ten miles north of the town (TR ‘P.). LIM NODYNASTES CONVEXIUSCULUS (Macleay) Ranaster convexiusenlus Macleay, 1828, Proc Linn Suc. N-S.W., 2, 135,; Iatew, @c.. Binaturt River, Dutch New Guinea. 5 (M-C.Z. 26003-7), near Darwin. Vomerine teeth extending well beyond lateral borders of choanac; first and second fingers stibequal; inber metacarpal tubercle slightly longer than the second: asingle metatarsal Luberele which is not shoyel-shaped., Largest (M.C.Z. 26003). ineastites SO tim. Jn lite. Above, marbled with black aud grey: the spots someumes fAneh edeed with white. Below, white vermicnlated (with brown). (T. RB. T.) This species has heen recorded already from Darwin by Parker (1940, 54). As he has scen the type ot L. elivaceus De Vis, which he refers to the synonymy Gf ronveriusculus, it must be assuined that De Vis’ description of olizacrus as hayine Livo metatarsal tubercles is erroneous, Parker is quite correet in concluding my (1935, 19) L. sulmvai Steindachner is a composite, for both Queensland frows (M.CLZ, 3610. 2623) conform to his new definition of converinsculus. Ureronei, revcosa (Andersson) Pseudaphryne rugase Andersson, 1916, Svenska Vetensk-Aka. andl, 52, Na. 9, 31; pb i, fig. 4: Colossenm, southern Queensland. 1 (M.C.Z. 23991), Noanamah, Ii correctly ideutilied, this 18 mm, juvenile is the first example of rugosu ta he recorded from the Northern Territory. Also the first of its species in the Museum of Comparative Zoclogy, jor L (1935, 31) erred in making ringasa a sub- species of wiarmorafe aud the five frogs then referred to OU. on. rnyasa are simply marintorata. As the ranges are largely co-extensive, Parker (1940, 70) did not (eteet my moistake and the ciation un his p. 70 should he transferred to p. 69, CRINIA SIGNIFERA SIGNIFERA Girard Crinta (anidella) siguifera Girard, 1853, Proc, Acad. Nat. Sci. Philadelphia, 6, 421; “New Follatid,’ we, Australia, 214 8 (M.CZ, 35999-20000) Kiuckey’s Lagoon. All are juvenile, the largest measuring only 12 mim. Six of them were taken beneath a pandanus trunk (T. R. T.). This is the form to which Darwin frogs are referred by Parker (1940, 87), whose synonymizitg of my 1935 references is probably correct, for I utilised or stressed other characters in defining the species of this difficult genus which he hias so thoroughly revised. ANLA CAERULEA (Shaw) Jtana caerulea Shia, in White, Journ. Voyage NUS.W,, App, 248, pl, ——: New South Wales. 4 (M.C.Z, 25992-3), Berrima. Vomerine teeth betwee the posterior borders of the choanae, fron, which they ace well separated: head as long as, or shorter than, broad; snout once and a half as long ag eye; tytnpanuin two-thirds to seven-eighths the orbital diameter ; outer finger half webbed; heel of adpressed hind limb reaches the tympanum ar eye. Length af ¢ (M.C.Z. 25992), 72 mm,, of ¢, 74 mm. The largest was taken at night on rocky ground, the others between sheets al galvanised trou al Larrakeyah Barracks (T. R, T.), HYLA RUBELLA Gray Tyla rubella Gray, 1842, Zool. Misc., 57: Port Lssington, Northern ‘lerri- tory, Australia. @ (M.C.Z, 23998), Knuckey’s Lagoon. Vamerine teeth between the posterior borders of the choanae: head luinyger thai broad; snout once and a half as long as eye; tympanum two-thirds the orbital dianieter ; outer huger without web; heel of adpressed hind limb reaches shoulder. Leneth of gravid 9, 32 mm. HyLaA avRea (lesson) Rone aurea Lesson, 1830, Zool. m Duperrey, Voyage autour du Monde w. sar... La Coquille, 2, 60, plo vil, ig. 2: Macquarrie and Bathurst Rivers, New South Wales. 5 (M.C.Z. 25994-5}, Knuckey'’s Lagoon. Vomerine teeth between the choanae; bead as long as, or longer than, broad ; shout cnee aud a half as long as.cye; tympaniim three-quarters of, or equal to, the Webital diameter; outer finger withoul web. Too shrivelled toe be worth measuring. JIXLA NASUTA (Cray) Pelodvies wasula Gray. 1842, Zool. Mise.. 86: Port Essington, Northern ‘Teeritory, Australia. 2 (M.C.Z. 26001), near Darwin, Vomerine teeth belween the choenac; head much longer than broad; snont iWwide as loug as the eve; tympanum seven-eighths, or equal to, the orbital diameter; puter finger without web; heel of the adpressed hind lin) reaches beyond tip of sont. Lareer meastres 43 mm. In life greentsh-black with a broad brown stripe down centre of back. Found aneng leaves after burning of spear grass at McMillan's, about six mules from Darwin (TLR, FL). DIBLTOGRAPILY Rovnmncer, G. A, 1896 “Catalogte of the Snakes in the British Museum (Natural History),’ (ed. 2, London), 3, xii-+ 575, pls. i-x] 215 Kkincrorn, J. R. 1942 “Herpetological Notes, No. 4.” Rec. Australian Mus., 21, 118-121, fig. 1 Lovertpcr, A, 1935 “Australian Amphibia in the Museum of Comparative Zovlogy, Cambridge, Massachusetts.” Bull. Mus. Comp. Zool., 78, 1-62, pl. Parxer, H. W. 1940 “The Australian Frogs of the Family J.eptodactylidac.” Novit. Zool. (Tring), 42, 1-108, fig. 1-20, pl. i Sautn, M. A. 1937 “A Review of the Genus Lygosoma (Scincidae: Reptilia} and its Allies.” Rec. Indian Mus., 39, 213-234, fig. 1-5 THE ECOLOGY OF THE WESTERN CLARE HILLS, SOUTH AUSTRALIA By C. D. BOOMSMA Summary The object of this paper is to describe the ecology of the E. macrorryncha and related associations, since little has been published on its occurrence in South Australia. A description of a fibrous- barked eucalypt in a personal communication by Mr. C. B. Scarfe led to a field examination in the Clare district. The eucalypt was located and it proved to be E. macrorryncha, thus constituting a re- discovery, since Tate records it for this district in his notebook in 1889, which is the same as his Adelaide district in hi Flora of South Australia of 1890. Later, in 1907, Maiden (6) briefly acknowledged Tate in recording its occurrence for the Adelaide district, but after a field excursion he denied its presence in 1909 (7). Black (1), in 1926, followed suit by omitting to record it, but Blakely (2), in 1934, records it for the Adelaide district. Similarly, it is not recorded in the most recent work by Burbidge (3). 216 THE ECOLOGY OF THE WESTERN CLARE HILLS, SOUTH AUSTRALIA With Special Reference to the Disjunct Occutrence of E. macrorryncha (F.v.M.) By C, D, Boomsma * [Read October 14 1948] The object of this paper is ta describe the ecology of the i. inacrorryneha aivl related associations, since little has been published on its occurrence in Smuth Australia, A description of a fibrous-barked eucalypt in a personal commuimica- tion by Mr. ©, B. Searfe led to a field examination in the Clare district. “The eucalypt was located and it proved to be £. macrorryncha, thus constituting a 1¢- discovery, since Tate records it for this district in his notehook in 1889, which is the same as his Adelaide district in his Flora of South Australia of 1890. Later, in 1907, Maiden (6) briefly acknowledged Tate in recording its occurrence for the Adelaide district, but after a field excursion he denied its presence 1909 (7). Black (1), in 1926, followed suit by omitting to record it, but akely (2), in 1934, records it for the Adelaide district. Similarly, it is not recorded in the most recent work by Burbidge (3). The area comprises 40 square miles of range country, being the western hall of the Hundred of Clare, lying directly west of the main Adelaide to Clare road, E. macrorryacha is limited to the uplands east of the divide. See fig, 1 and 2. The divide has an clevation of 1,500 feet; a few peaks are higher, and it ts dissected by inany smal] valleys formed by the winter-flowing (ributaries. of the flutt River. Valley formation has been slow, so the topography is accordingly rugged and surface run-off is rapid. In the main. the bedrock is of Proterozoic age and is quartzite or sandstone with limited occurrence of argillaccous rocks. The general strike is north and south, as shown by the north-south direction uf the axis of the range. Inter mittent winter springs were obseryed which may indicate fault zones, otherwise mm) fants were observed. CLIMATE Rainfall—The rainfall is mainly derived from the normal winter rains due io the general west to east movement of the pressure systems, wich infrequent heavy summer rains due to tropical troughs. from the north, The mean annual rainfall in the tanges ig 27 inches, with 19 inches in the winter six months and 8 mches in the stummer six months. I is probable that this rugged topography produces Incal rain-shadows with rainfalls up to 30 inches per annum. High suminer iemperatures are frequent and severe frosts are common. The frosts may be experienced during a period of seven months of the year, In reneral. the cliniate is cold and wet in winter, watm and dry with occasional frost in early sunimer. SOILS Kield observations show that the soils are, in gencral, shallow; skeletal soils are frequent in oecurrence, and deeper transported sails are limited io the broader Hutt River valley. Slight podsolisatiou is general, as most of the sails have a grey sandy-loam surface horizon and up to 20 inches of slightly oxidised clays. in the lower horizon. Occasionally a red-brown clay is observed with ber rock at less than 36 inches. Trak. Roy, Soe. B. Ayet., 72, (2), ch Mace bh 1949, ba pes “I VEGETATION ASSOCIATION CLARE DISTRICT HO MILNE | SSaeee SSSSSS STS Feleophors EZ kconslelslensis TTT. feucorylon @ Cathiters progirgua BLYTH Cas, stricta c SY & Benksés marginals QO Eeverato NIWORT HAM Tic. | VEGETATION Mloristics—The nomenclature of eucalypts is after Blakely (2) and that of the other species after Black (1). The Horistic make-up of the understorey 4s continuous in all the forest types, being composed of occasional scattered trees of Casuarina stricla, Callitris propingua, Acacia pycnantha, and Exocarpus cupressi- formis; Bursaria spinosa, Acacia Wattsiana, Hibbertia stricta, and Lissanthe 218 sirigasa are the chief shrubs. The commonest geophyte is Tliysanolus dichotomus, and the characteristic ground cover of grasses consists of Themeda australts, Hunthonia spp. and Stipa spp. Leguminous shrubs are missing quite frequently from the understorey, which is unusual when there is a stringybark dominant. LF BLAD jece see ie eet A tie / a on | fF wie THE PECOEOLD DISTRIBUTION OF EF MATEORM HCA Felt THE Saye NM TALES AFTER CE CARTELS SOUTH AMSTPALTA AY ¢ OD Bete Nata Fig, 2 Alssoctations—Those present are all dominated by the eucalypts E. cameldi- leusis, E. leucoxylon, EF. claeopharw and E, macrorryncha. There was no {°, odorata, E. fasciculosa or mallee present. A stray E. caleicultrix was observed i the Clare Pienic Ground, adjacent to the main road. Stands with more than one eucalypt species as dominants are frequent, but . leucorylon is typical of ihe drier exposed portions, A, camaldulensis of the moister valleys with higher fertility, J2. mecrorrvacha on the wetter, cooler uplands, and E. claeuphora sup- planting L. camaidulensis on the higher fertility sites with greater exposure. As sh distinet boundaries between sites occur, il ig ual surprising that several eucalypt species are found competing for dominance, It is therefore expected that the ilegree of variation in the factors of the etvironment is greater than that tolera{ed hy any one species, The associations are ali savannah woodlatids with a variable number from whe to a hundred trees per acre. FE. lencoxylon association—This association occupies the greatest area, being w part of the continuous distribution along the coast side of the Mount Lotty Ranges from Adelaide to Quorn. In the area described it occurs on both the western abil eastern slopes as a dominant, with occasional inlying islands of ( usuering stricta and 2. elacophora, The ground cover is grassy, often bare on skeletal soils, and the dominant is only a medittm-sized tree, having a small crovuked! stem up to 30. inches in diameter. The number of trees per acre varies from under ten to a hundred, depending on the amount of clearing, grazing, buruing uid the size of the trees. &. camalduiensis assoviatian—This is typical of the valleys which have deep. fertile, transported soils. The trees may reach a gross size up to 48 inches in cHameter, heights up fo 80 fect are common, but regeneration is slower than wulisation and the present occurrence ig only a remnant of the original stand. ‘The understorey is typically grasses, with a few bulbs, and occasionally Casuarma stricta, Acacia pycnantha, Acacia rhetinodes aud Barsaria spinosa. Near Pen- wortham, Banksia marginal is in the onderstorey, 219 FE, elucophora assoviation—This specs is rarcly an exchisive dominant on areas greater than 5 acres in extent. It is found with £. Jencoarylon or E_ macrur- ryncha, and again the understorey is continuous wilh meighbouring associations. When it is found in vallevs the soils are less fertile and it 15 later replaced by F, camaldulensis lower down the valley where the soils are more fertile. FE. mocrorryncha. association — This association is now maiihy a savannall woodland, but the ridge-tup parts of Sections 364, 365, 436, 535 and 536 have Xanthorrhova guadrangulata and Lissanthe sirigasa im the understorey with w grassy ground cover. The absence of leguines and Proteaceae ig marked, anil this community can be classed as a gradation between savatinah woodland and dry sclerophyll forest, The understorey is compused of scattered shrubs of Bursurit spinosa and Acacia pycnaintha, with wv few herbs Hibbertia stricta, Avarit sanguisorbae and a grassy ground cover, The present facics of the EF. ieerorryacha association in this district 1s similar to that of the E. leucoxylon and E, camaldulensis association. The under storeys are continuous with the exception of the restricted occurrence of the Nanthorrhuca quadrangulata understorey already mentioned. IJowever, this understorey occurs in other woodland associations in other districts of South Australia, ‘The distribution of £. secrerryncha in the eastern States af Australia is after Carter (4). See fig. 2. Ecologists have described its field relations, notably Pryor (9) in 1939, who states that it reaches its best development as a consocia- tion on exposed upland sites of 2-3,000 feet altitude m the Australian Capital Territory. Associated encalypts are FE. Rossii, A, meculosa and I. cordiert. The understorey is considerably better developed there than in South Australia and includes the tall shrubs Pomaderris eliptica, Acacia falciformis and Exocarpus cupressiformis, Abundant small shrubs, mainly epacrids and legumes with Xanthorrheca australis ensure an almost continuous ground cover. In Vietaria. Ewart (5) in 1925 and Petrie et al (8) in 1929, deseribe it as the dommant of a consociation on exposed upland sites on the Silurian formation at altitudes of 2-3,000 feet. These occurrences are comparable with each other in the broades characteristics. The nearest recorded oecurrence of 2. macrarryncha to this area ics at Stawell, 300 miles to the east in Victorian, As an isolated duplicate, genetic origin is highly improbable; the restricted ocetirrence of this species requires u knowledge of the methods of colonisation and invasion of eucalypts sitice earls Tleistocene times. Similar disjunct occurrences have been recorded for E. hemiphioia, E, albens, L. niicrecarpa, LE. pauctflera, Ee. rybida, be. cladocely and Jf. evata, lt is highly improbable that the Hundred of Clare contains the only suntable site for £. macrerryncha in South Australia, and so the alternatives ace that thr is a relict of a previously wider oceurrence or that it occurs elsewhere and ha- not yet been observed or that it is a recent arrival and has net spread, The association is not vigorous since it has not extended its boundaries m sixty years but rather has contracted, due to death of mature trees and absenci of seedlings to niaintain dominance along the boundaries. To this exrent the fell observations favour the theory that it is a relict. Although occurrences. elsewhere in South Australia are possible, they are improbalsic. Ti is notewarthy that all the factors. which savour the “latis of at relict strongly argue against this species heme a now arrival, 220 SUMMARY The rediscovery of the occurrence of E. macrorryncha in the Clare Hills, which is the only recorded area in South Australia, has led to mapping the aecurrence with a description of the environment and community relations, The disjunct occurrence of E. macrorryncha 300 miles from the nearest recorded occurrence in the east at Stawell is difficult to explain. The explanation depends on the past climatic changes which may have destroyed connecting sites hetween the east, leaving this occurrence in the ranges as a relict. Other species with similar field occurrences are E. albens, E. hemiphloia and ff, microcarpa, 10 REFERENCES Brack, J. M. 1922-45 “Flora of South Australia.” Govt. Printer, Adelaide Brakety, W. F. 1934 “A Key to the Eucalypts,” Sydney Bursince, N. T. 1947 Trans. Roy. Soc. S. Aust., 71, (2) Carter, C. E. 1946 “The Distribution of the more important ‘Timber Trees of the Genus Eucalyptus.” Govt. Printer, Canberra Ewart, A. J. 1925 ‘Handbook of Forest Trees” Maen, J. Il. 1907 “Critical Revision of the Genus Eucalyptus” Maren, J. H. 1909 Trans. Roy Soc. S. Aust., 32, (1), 325 Perrie, A. H. K,, Jarrert, P. H., Patron, R. J. 1929 “Journal Ecology, 17, (2) Pryor, L. D. 1939 Handbook A.N.Z.A.A.S., Canberra Woop, J, G., and Crocxer, R. L. 1947 Trans. Roy. Soc. S. Aust., 71, (1) NOMENCLATURE OF EUCALYPTS WITH SPECIAL REFERENCE TO TAXONOMIC PROBLEMS IN SOUTH AUSTRALIA By C. D. BOOMSMA Summary The identification of eucalypts has always been a task requiring all the faculties of judgment, and even then, numerous contentious issues arise which persist without a satisfactory explanation. Such an issue is the identification and naming of specimens which have some characters common to at least two species; or a gradual variation of a single character from one specimen to another. In fact, variation is so general that the limits require definition, but to do so involves a major statistical examination on the four following variations: - Variations of a single organ during maturation; variations of an organ on a single tree; variations of an organ between trees, including (a) variations due to environmental causes and (b) variations due to genetic causes; variations between species. 221 NOMENCLATURE OF EUCALYPTS WITH SPECIAL REFERENCE TO TAXONOMIC PROBLEMS IN SOUTH AUSTRALIA Ry C.D. Boomsaa* | Read 14 October 1948 | The identification of euealypis has always. been a task requiring all the faculties of judgment, and even then, numerous contentious issues arise whicli persist without a satisfactary explanation, Such an issue is the identification an¢! naming of specimens which have some characters commion to at least tywo species: or a gradual variation of a single character from one specimen to another, fi fact, variation is 50 general that the limits require defnition, but to do so inyolve> a major statistical examination on the four following variations —Variations of a single organ during maturation; variations of an organ on a single tree; varia- tions of an organ between trees, ineluding (a) variations due to étrvironmental causes and (b) vatiations due to genetic causes; variations between specics. VARIATIONS OF A SINGLE OkGan DURING Maruration The main importance of the stage of maturity of an organ to those engage: it. the determination of species is the ability to recognise a given stage andl describe it correctly, Juvenile leaf characters, for ¢xample, afford a ready means for identifying some species, and this character may be used to separate or show alliances between species. But tates of muturation vary and even an individual season will determine the length of flowering period, character of nectar produc- tion and the beginning of shoot growth, su this class of variation is least under- stood. VARIATIONS OF AN ORGAN ON A SINGLE TREE Tt is known and accepted that variations occur, but just how great hay nut been reliably determined. It is to be expected that the variation within a tree would be Jess than between trees, and in the case of BE. obliqua this is borne out by 4 statistical examination of capsule dimensions carried out by the author, VARIATION (i; AN ORGAN BETWEEN TREES It is known that variations are great, but fhe limits have not been determined. The amount. of variation is apparenily least in endemics ta Sonth Australia, par- ticularly those restricted in area, for example Lf, cladoculyx, FH, cosmoephylla, E. cneorifolia, E. fasciculosa and FE. vemota. On the other hand species with a wide distribution, including areas at the limits of the specific ccological amplitude of the species, generally show a correspondingly wide variation in the character of any one organ. The causes appear to be partly environmental and partly penctic. Species which show wide yariatious include JE. olevsa, E. anceps, E, lewcaxylou, EK, odorata, E. Hubeviann, , Baxtert, E. obliqua, and £. lepto- phylla, The variations are diverse and different botanists have given varictal and even specifte rank to specimens showing characters which anly represent one stage in a séqtience of characters from one species to another, This confusion has been accentuated by the practice of taking a small sample of a tree and raising it to the cxalted position of a “type” with which all future specific determinations must be compared. In material showing wide variation, the patadox is reached that the more closely a single specimen is described for specific rank, the more * Forestry Department, Adelaide. ‘rans. Roy. Sec. S. Aust, 72, (2). 30 March 1949, , 222 fikely is it that an identification key will be unwarkable, Tn henth Australia, al least half the total number of species exhibit such wide variation in characters jhat the existence of intermediate forms between two species must he accepted, (a) (urtations due lo environmental causes, Generally a species poscveses a wider potential environment than és exhibited in its natural distribution; (oy example, [, cladocalys can be grown on solonized brown suils in the 12” mean annual rainfall of Loxton, although its best natural occurrence is i the Plinders Ranges ow podsols with at least 25% mean annual rainfall. Wurthermore, deserip- tions of characteristic environments for a species ate fragmentary and ther limits of both potential and natitral environments ltive to be defined, so that to describe vurlations ul character in argans due to environmental eatises can only be bused on speculation, However, variations in ecological character duc ta environmental causcs. Le, stent form, frequency of occurrence and exclusiveness are frequently encountered Such variations aré common to all plum species and need mot be delailed here other than to remark that variatious are correlated with the diversity of enyiren- ments occupied hy any one species, so that ©, leucoxylan and L. vlewsu, bw widely occurring species, show many variations die to environmental causes. tt would be expected that variations in ecological character would be at a miniomne in large areas of constant etvironmental character. Soa far, no such large ares has been encountered in South Australia, and it is unlikely that such an ares exists in the tree zone. (b) Furiations due to genetic causes, This question is intimately rele teil to that discussed in the Following section. Maiden (4) in 1922 summarised the evidence for genetic variation, and slated that intermediate forms occurred between many eucalypt species. Furthermore, he described several plaiits as natural hybrids. In South Australia it is probable that E, juyaly (Naudin) is of hybrid origin, while 2. Nalangadoosnsis (M) and kK. Melntyrensis (M) are also pro- bahly hybrids. Tn 1946, Brett (2) showed that eradations arise by selfing of the F1 generation and occasional back-crosses. to the original parents. Some members of such a swarn of individuals will be more suited to the site than others and so will produce a stand of trees with a wide (but determinate) range ol characters. As time progresses, fixation of characters inay oceir to give rise to a hybrid-polymorph, ‘The stages from the hybricd-swarm to the hybrid polymorph are conjectural, but Geld observations support such a view, lt is possible, then, that polymorphs may produce new biotypes, 7.c., collections of individuals breeding true for characters of taxonomic importance, and with a narrow determinale yarialion of characters, No attempt has yct been made to rlefine a species on such a basis, so it is a matter uf opinion when the amount of variation shown by a polyinorph is sufficient to be accepted as a species or bia- type, Tt shoult be noted that no two imdividuals arc identical in genetic character, which explains the difficulty encountered in identification of material collected from areas where hybrid swarms occur. VARIATIONS OF AN Oroaw BETWEEN SeECIES Interspecific variation is probably o7 genetic origin and the slady of grada- tions between species is greutly facilitated by the theory of hyhridisin, and the sexregation which is continwing in some cases ta the present day, Luecalypt species in South Ausiralia do not all show segregation, but the probability of observing genetic variation is highest amongst the widely oceurring species. Observations suggesting genetic relationships belween species are sel oul in the following notes, but first the scope of specitic sequences in juvenile and mature leaves, buds and capsules, illustrated in the figures, should be examined, The nomenclature is aller Blakely (1). 223 tig. 1-4 show a specific sequence in the juvenile loaf shapes of JE Mfuberinit, lig. &-? show an interspecific sequence between HE. grmialis (fig. 5), J Huberigny (lig. 7). and B, leneaaylon (hig, 9), hig. 1-4 show a specie sequence iu juvenile leaf shape in BE. Huberiana; lig. 11-18 show a portion af the mature leaf-shape sequence in E. trauscontinenta- His; fig. 23, 24, 27, 28 and 32 show sequences of bud characters, while fg, 25 and 26 show the range of sizes and shapes encountered in E, lewcosy/on capsules. Fig. 40 shows the diversity of capsule characters in E. Bascteri, aud fig. 39 shows a large- fruited form of /2. Luberiana, Evidence of interspcofie sequences may be had from two sources. One is the published data (the mos; recent being that of Burbidge (3) ), and the other ay be obtained from the field. From the first souree there is general agreement that many difficulties in uomenclature are due to mlerspecilic genetic variation, win) trumerous exaniples are given by the authors already referred to, Only a few of the sequences observed im the field have been illustrated and the most familiar will be discussed, From the small area of ten acres in the Hundred of Coonarie, specimens comprising fig. 29 were obtained, Fig, 29a has buds that compare favourably with those of f. leptoplyla, but its leaves are up ty ter centimeires long and two em, wide and agree with those of the nearby kL’. wleasa, Fig. 29b has buds about twiee the size of those of 29a, and agree with those af a type specimen of F. socilis in the National Herbarium, Melbourne. The elongated conical operculum of fig. 29¢ and the sinal! waisted type of fig. 29d are steps to fig, 29e, which is similar to the operculum of L. transcenti- nentalis. lf the examples do not represent species but varieties, the position is i? 1 Sy Fig, 1148 show a specific sequence in mature leaf shape of &. transcnntinentatis. Vie, 19-22 show an interspecific sequence im mature wal shane and capsives beareen E. THuboriana (fig. 19) and 7. ome (fig. 22). Vig. 20 and 2) are intermeciate forms. unimproved and the acceptance of numerous forms would he misleading. In this occurrence the field suggests that this is a hybrid swarm as variations were also collected of I. diversifolia, FE. rugosa and E. leptophylla. Interspecific seqtietices lave also been observed in E. odorata, but these are not quite so obvious. so that difficulty 1 experienced in separating E. raleiculirix, E, odorata, E.Lansdowneana and E, jugalis. Vig. 33-35 show the similarily of the capsules, E, viminalis has often been observed in an interspecific sequence with E. leucoxylon, as seen in fig. 5-9, which include a few steps of the juvenile leaf shape sequence between the two species, On the same sinall area, specimens were collected with triphylly and alternate leaves. See fig, 42 and 43, E. leucoxslon is widely scattered and is concerned in matty interspecific sequences. Fig. 5-9 show an interspecific sequence between F. feucoxylon and E. Huberiana, E. leucoaxylon is closely associated with E. jugalis in the Yatina district and specimens occur which are difficult to separate into species, Other examples of genetic variation occur in LE, Jewcoxrylon, such as the sequence between E. leucoxylon and F. calctculfrix near Cape Jervis, and specimens with more than three buds per umbel from the Meningie district. See fig. 41. bo Nh on #3 , (, t, T | a 5 0 °..%9 9 9 @ 24 Pig. 23-24 show portion of the specific sequences in operculum shapes of E, oleosa (fig, 23) and £. lranscontinentalis (fig. 24), Fig. 25-26 show specific sequences in size and shape of capsules of E. leucorylon. Fig. 27-28 show specific sequences m size and shape in the buds of &, cosmo phylia. Pig. 30-31 show an interspecific similarity between buds of E. bicolor and E. interiexta respectively. Fig. 32 shows a specific sequence in operculum shape of buds of EB. camaldulensis. Fig. 33-35 shaw the interspecific similarity of capsules of EB. odorata, E. jugalis and E. micro- carpa respectively. Ig. 36-38 show an interspecific similarity between capsules of E. viminalis, E. rubida and E. Hubervana respectively. Fig. 39 shows a large capsule form of &. Huberiana. Mig. 40 shows the diversity of character in capsules of E. Basxteri. Examples from other species such as A. jugalis, E. elaeophora and E. Hube- riana have been collected and not figured. It is evident that these sequences indicate relationships between species, and from the nature of the sequence and from authoritics already referred to, Table I has been compiled. The purpose of compiling the table is to indicate the species relationship in South Australia. The groups are naturally subdivided, so that there is little chance of observing an interspecific sequence between species of two groups and the order of each species is to some extent arbitrary, but the probability of two adjacent species forming sequences is higher than in those distantly connected. It is to be noted that the table contains some species that are of reputed hybrid origin, such as E. jugalis, E. unialata, E. Huberiana, E. vitrea and E. anceps, Interspecific 290. Fig. 29 shows some genetic variants of £2. oleosa. Fig. 41 shows an umbel of &. lencaxylon containing more than three buds. Fig. 42 shows tripbylly in &. zdmanalis. Fie. 43 shows alternate juvenile leat arrangement of £. wlpatitallis. Fig. 44 shows an umbel of buds of 2, cosmophylla containing more than three buds. sequences help to emphasise the relationships of species within each group, so that it ig proposed that a study of these sequences will be of considerable use in the identification of species. SUMMARY The identification of species of the large genus Eucalyptus (more than 500 species) is complicated by the occurrence of many forms that are atypical. Some of these [orms are genetic variants and are of bybrid origin. A selection of examples that occur in South Australia is figured and some interspecific sequences are discussed, These sequences ate of use in confirming species relationships in groups as an uid to identification. Interspecific sequences are frequently observed in widespread species, but not in endemics. REFERENCES 1 Buaxety, W.F. 1934 “Key to the Eucalypts” 2 Beer, R. G. 1946 “Some Aspects of Hybridisation in relation to Taxo- nomy in Eucalypts”. Unpublished paper. Section M., Aust, N.Z. Adv- Science Burner, N. T. 1947 Trans. Roy. Soc. S. Aust., 71, (2) 4 Matwen, J. If. 1922 “Critical Revision of the Genus Eucalyptus” ive) 227 Taste I Showing the Taxonomic Relationships in the Genus Eucalyptus in South Australia ajsapin ts Ed spas Sptimntnndzengtntonteyabtih vitrea Spalantei OTA cies scscuscmtacetvsterue de Beuzevillei x Soukiseqnsurapseratussnpetecessrnatogapstcane niphophils x camaldulensis.... cece rudis x e yngitnssteszeptesiect cabeatedircechvstveap® umbellata x Wie Bs vassli le eax peungdgoshenlnnondst Morrisii Searle Sheba fone plate odaettbe uke pebtie ATCEP Ss. cccrtecessesaunsesponyesy sve ates Peidel Meideedbe gered rugosa PUUIMIOSA..ssecsesssessseaseesecrseeen pileata...... ines Drachycalyx....... conglohata Sicates destinbedseraonuceiigesntonsi Depots: INCTASSAtA.... sere eeeeees tetraptera x fe opahe sae anna sul atta AM yea nba) Blaxlandi x PBAXtOT le csscccecssecseeesesereeeeees capitellata Xvieccuccsesseseeees Jsfvmtgange a ODHGUA q Sy vele nti day enh rprateastct WES Sebuayad otssa te ddasets Sis eot Seibel diversifolia......... remota letter ieee ee Sideroxylon fruticetorum....... viridis TeUCOXYON..sssecsessecnseneeas calcicultrix.... ee Od Orta... csecsenees microcarpa,........ hemiphloia Fs atts dsueteetioha Puce sien jligalis Lansdowneana albens elaeophora ; las sresdtsataiedih cruasrsceetsanesoatsean es Huberiana........... sees VINUNALIS.. ce ceeeeees rubida... ee Dalrympleana OVALA acscecceeeenerteeeres unialata xX... globulus x........ bicostata x YY. gM age ota teToa TH iear cae dl cneorifolia EOLCOSA..cacseesssesseresoens won transcontinentalis........ Gillii....... Sebeiand .... Flocktoniae x denotes interstate species THE GEOLOGY OF THE BOOLCOOMATA GRANITE By A. W. WHITTLE Summary Old Boolcoomata Station is situated approximately 280 miles north-east of Adelaide, and 12 miles north of Olary, which is on the railway line to Broken Hill. A very great portion of the area of the Station is occupied by interesting granitic rocks and associated gneisses. These rock types occur amongst a normal folded sedimentary series of argillaceous and quartzitic rocks. 228 THE GEOLOGY OF THE BOOLCOOMATA GRANITE By A. W. WuiTTLe * [Read 14 October 1948] 1. Inrropection axp Previcus Work ai he els \> ts -. 228 Il, THe prtncipan GRANITE Pose a's 4 4. _ “e t, .. 229 JIL Tur ArkAs occur sy GNEISstc Rock .. bi 4, ‘. “8 .. Bde TV. Tur Country Rocks ., an os be 44 - . - .. 24 V. Srraucrcre ge ¥ a ty ts oe ies ste x3 .. 235 VIL MertramorPHism “ dle ws ~ e . 236 VIL lIewrous Injection IN THE CounTtay Rocks = ti ee .. 236 VIII. Mrinerarization CUeNtTRes o “a fe A. ia rs a .. 238 TX. Fearcres or tHE Contact, ‘4 +4 wo a Vv o> .. 238 X. Grawrrization PHENOMENA -, a2, i e's os “4 a .. 240 XI. Conchusioxs f. ir +4 a an .. 44 28 24 .. 242 1. INTRODUCTION Old Boolcoomata Station is situated approximately 280 miles north-east of Adelaide, and 12 miles north of Olary, which is on the railway line to Broken Hill. A very great portion of the area of the Station is occupied by interesting granitic rocks and associated gneisses. These rock types occur amongst a normal folded sedimentary series of argillaceaus and quartzitic rocks. Previous Work The first important contribution to the geological study of the north-eastern portion of South Australia was that published in 1912 by Sir Douglas Mawson, Professor of Geology at the University of Adelaide. This work, which covered a large area including the district in which the present work was done, was pub- lished a5 a Royal Society Memoir. Subsequently Sir Douglas Mawson has con- tinued his study of this region, including much detailed field work and petro- logical analysis in the Olary district, A list of investigations by other workers is given in the bibliygraphy. The present investigation was carried out to determine the nature of the Old Boolcoomata Granite and its relation to the sediments which it intrudes, An attempt was also made tu discover some data which might suggest whether this granite is the result of an intrusion or of the process known as gfanitization. The area of the hatholith examined measures three miles east and west, by two miles north and south, T.ow rounded hills of igneous and gneissic: rocks, with a few large bare granite hills, make up the landscape. Traverses were made across it in a north-south direction at quarter mile intervals, to pick up the dis- tribution of rock types and to ascertain the position of conlacts between thei. Subsequently, several of these contacts were followed to prove continuity between the positions located by traversing. Although there is a great diversity of rock types in the batholith, they may he coriveniently considered in three main groups. These are (1) a medium to coarse-grained muscovite granite, (2) a finer-grained, somewhat gneissic por- phyritic felspar granite, (3) granitic rocks in which the presence of injection gneisses is significant. The distribution of these groups may be seen on the folded map. * Mines. Department. Trans. Ray. Soe. S. Atist,, 72, (2),.30 March 1949. 220 I. THE PRINCIPAL GRANITE PHASE There are three areas in which this is the dominant reck type, two of which are partly made up of patchy developments of the porphyritic felspar granite. The principal granite area extends three-quarters of a mile north from the station house and is then replaced by gricissic rocks, To the west it runs against alluvium and the contact with the slates, while its greatest extent is eastward, where it runs uninterrupted down towards Binberrie 1ill. Throughout this area, medium to coarse-grained felspar-rich granite is the dominatit tock, The rock js mostly massive, hut becomes slightly gneissic in places, particularly towards the northern border of the belt. Jointing is well developed in twa major directions at tight angles, while a minor third direction is sometimes present, The joint planes usually have steep dips. These features produce excellent “tor structures with occasional balanced boulders. Sheeting structure occurs in large outcrops where a series of fractures have formed separated hy several fect and lying parallel to the rock surfaces. In a Jarge exposure, such as the slope of a hillside, a series of steps develops as over- lying shects are weathered away, leaving the immediately underlying pavement strewn with decaying granite boulders, ‘These pavements are large, smooth and gently rounded, and have resulted from the expansion of felspars undergoing kaolinization as well as by the alternate heating and cooliug of the granite by day and night. Erosion of the area is greatly assisted by the joint system, becarisé these fractures start streamilets running along well-established channels which are obviously scoured-out joint fissures. Similarly, the major streams which drain the granite hills trend roughly parallel to one or other of the major jointing directions. Pegmatite veins, up to two or three feet wide, im general follow the N.W.-S.E. direction of jomting, which suggests that the fissures ate pre-pegmatite, developed in the early stages of the formation of the batholith, providing an easy means of escape for the pegmalitic materials developed in the late stages of con- solidation of the granite. Lrobably these fractures, carrying pegnmaties up ta three fect wide, were enlarged by remelting along their margins, because there is a gradation in grain size between the pegmatites and the enclosing granite. This, however, may be the result of a chilling of the pegmatite along its margins. The granite breccia (referred to later) trends parallel to the E.-W, joint system and may therefore be related 10 the jointing, Naturally, accumulated stresses would most readily he relieved along an already well-established fracture system. Movement has occurred along some planes of jointing, shown hy the local faulting of minor and major pegmatites. These joints post-date the pegma- tites they offset, and helong to a different system from those along which the pegmatites moved. These joints may be related to the brecciation, The joints change direction toward the western portion of the main granitic belt, and the associated pegmatites swing round to an almast westerly trend. These lincal pegmatites are very abundant in the granite, half a mile N-E. of the homestead. They are narrow. but extend many hundreds af yards east and west. East and west of this locality, the lineal pegmatites are replaced by discon- tinuous wider irregular pernatites which show no structural control. The peg- matites are coarse-grained, rich in felspar, and poor in micas and accessories, although tourmaline is abun¢lant in some, An interesting feature of this granitic belt is the granite breccia, made up of angular granite fragments up to several feet in size, sct in a fine-grained dark matrix largely composed of crushed clayey ficlspars, quartz and mica. The large 230 broken boulders are all of the same type, viz., a granite identical with that making up the major granitic zones. There are occasional slate or quartzite boulders within the breccia, presumably, broken xenolithic material, In places there are denuded surfaces resembling crazy pavements where the nature of the breccia can be closely sttdied (pl. xx, fig. 7). Where angular boulders are absent, larger grains in the matrix are streaked out into parallel rows set in finer mylonitic Paraliel ore ih y \ of Fracture inl: +3)! Granite. FRACTURE PATTERN IN GRANITE NEAR CONTACT Fiz, 1 Fracture pattern in granite near its contact with the granite lreccia. Ke Matrix rich in ¢'|Felspar Quartz a land Muscovite Fracture) \ 3 Lines, ips FEET.S T= 20 FEET. FRACTURED GRANITE PASSING INTO BRECCIATED GRANITE Fig. 2 Fractured granite passing into brecciated grate. material, producing a “pseudo-hedding” effect, by the freak “sorting” of grains. Most of the breccia carries angular boulders but there is much variation, this between wide hands containing granite boulders there are these “bedded” zones with a fine-grained matrix set with parallel rows of rolled-out rounded granite or felspar grains which are porphyroclastie remnants of pre-existent granite (pl. xx, fig. 2). 231 "2,RIS PIFAL]UI UB UO JEpUOI d}MINIH ayi Jo WoOysoOg {ANT evel Wdy piorrosheh WAIANTIY eta eS CALL a3zilinvas — « —= aad | (AS tn Poesy oe JLwWdsS G3AuYA es DAIZLYWWAD POT SUINWHS DLAI GINIVES 35uvO2 Cr ID D4ue OINYIWLAS ne viozaua Ee oty = -SLINYYD BYTADNY ON3931 T5¥iNGO. ZLINVES 3Hi- id NY G20uy Ia SLUNVES WINWOOD 10GE San 20 bdat ov 5 ony —-——~—"NOVLWITOS 49 Id ONY FHNLS Ory oe DNIDGI 40 did ONY aNI¥LS a JOVAVITD 40 did GNW INidAs (on 03 48) SY VENOB Fw31907039 "10248 aS 8G) AEVaNNOS WI9071019 a ae SN93$ OL 3DN2H433U 232 The main belt of breecta occtirs along the southern border of the granite near the homestead and is more or less continuous for about a quarter of a mile northwards, while beyond this point separated discontinuous bands up toe a hum- dred feet in width continue fer a further half mile north At either end these bands pass through fractured granite (fig. 2 and pl, xxi, fig. 1) into tinbroken granite, The fractured granite is a stage in the formation of the breccia, and i{ is always closely associated with the brecciated granite. The drawings (fig. 1, 2) illustrate how the granite is thoroughly traversed by intersecting cracks which split i up into angular fragments. These cracks are at first quite narrow. but towards a zone of brecciation the fissures widen and became filled with finely crushed material, as the rock grades igto the true breccia. The major mass of breccialed granite runs into alluvinnt near the homestead, while eastwards it is gradually obscured amongst the granite. The breceia often carries tillite, slate and quartzite, which have not been granitized, forming local bands of angvilar boulders, set in a fine-grained chlorite- rich base runting parallel with the general trend of the main granitic breccia. -4 = ‘ _* - - + se fae ™.,§ es | SP ee ee ‘ tant Ble ga! Meek lees A \o T.+ MEDIUM” COARSE-GRAINED- Nowra oat VE GRANITE ete ye ve ae (Rr ae. a aoe -~- + + aye ~ ~SFing Grained Brecciated EST eg tay: Feet 5. f 5 SCALE 4p 15 20 fErr PEGMATITE WEINS OFFSET BY NARROW SHEAR ZONES Vig. 4 Peematitie yeins offset by narrow shear zones, The map (fig. 3) showing porlinn of the brecciated area om a scale ol 100 fect to the inch, provides an indication of the general nature of the broken zone. The angular granite boulder breccia occupies the greater portion of tlie broken zone, while the general direction of the streaked and “bedded” portions is shown following the trend of the zone as a whole. An area of irregularly altered slate and tillite is shown with a doubtiul boundary due to surface rubble obseuring the contact. Lens-shaped granite bodies occur in this rock. Several lines of evidence support the concept of this peculiar formation as a breccia. The angular shape of the boulders and the nature of the matrix are typical of brecciation. Further, the angular fragments. and the matrix materials are of the same type as the massive granite in the vicinity. The closely asso- ciated fractured granite represents partial brecciation. The best evidence is that of shearing in the vicinity of the broken granite, where there are long narrow zones of mylonitized granite, two or three feet wide, in which the rock is reduced to an extremely fine-grained chlorite-nch 283 sheared mass, with flattened elongated remnants drawn owt parallel to the fine banding in the crushed matrix. These shears, which are several hundred feet long, are parallel to the breeciated bands. Several of these are plotied on the map- There ate numerous examples of pegmatites, faulted by joint fissures, which are filled with fine-grained and angular granitic material (similar to the matrix of the main breccia)derived from shearing of the walls of the fissure by the movement which offset the pegmatites. These minor shear zones showing a dis- placement of a foot or so extend parallel to the trend of the breccia (fig. 4), On the other hand, no evidence of displacement is visible in the country rock west of the point where the breccia runs out against the allivial flat, while eastward the breccia disappears amongst the granite and can nowhere be traced. Thus, when its great width is taken into account, it is odd that, if this is really a breccia, it should have such a short longitudinal extent, There are two other areas in which purely granitic rocks dominate; in both, however, there are patches of porphyritic felspar granite. Roth have similar features and cau be described under one heading. The grain size is irregular in these rocks, and biotite occurs as well as muscovite, thus it may perhaps be deduced that they are a little less completely granitized than the main granitic mass, Generally it is a medium-grained granite, but fine-grained or coarse- grained types occur in places. Although in the main massive, It sometimes becomes gueissic in greater or lesser degree, and may locally puss into injection gneiss ot carry xenoliths oriented parallel to its gneissic structure, There is a special tendency for it to break down into numerous cuboidal blucks up to 6 fort in dimensions rather than to form massive granite tors like the main granite. Thts ig probably due to a more closely spaced jointing system, and the more irregular grain size giving great unevenness during expansion, The granite boundaries afe either well defined against the gricissic rocks, or grade hy the addition of xenoliths into injection gneiss, Lath-shaped felspar phenocrysts up to $" long and 4” wide, oriented parallel to the direction of the gneissic structure, are cnn- mon in these gtatiites. Toward the north-east these granites continue beyond the area mapped, while to the west they grade into gneissic rocks toward the border af the “batholith". There is no regularity in the distribution of the porphyritic phase in this gramte, the strongest development being in the northerly belt of granite where porphyritic rock occupies a ridge for nearly a mile in an east-west direction. Irregular masses of pegmatite occur but are not strongly developed in either of these two regions. itll. THE AREAS OCCUPIED BY GNEISSIC ROCKS These accupy a very large area. Throughout there is am abundance of granitic rocks ranging from fine- ta coarse-grained types, which form, as it were, 4 base in which the gneissic material occurs. Thus, although the map shows large areas of these gneisses, these are not to be considered as made up wholly of this rock type, but rather that in these areas the occurrence of gneiss amongst the granites is the significant feature. Pegmatites of an irregular nature, both large and small, occur in these areas. It is usual for granitic and pegmatitic material to occur in lit-par-lit fashion along planes of schistosity in the gneiss, producing banded injection-gneiss. The bands of igneous rocks are closely spaced, varying in thickness up ta an inch. Schistose or slaty rocks have been converted, in general, to dark biotite-muscovite- rich gneisses in which the black biotite bands contrast strongly with the light- coloured quartzo-felspathic ones. 234 Quartzites are not very much altered, although in places there is a partial loss of sedimentary banding, while elsewhere this is preserved by the introduetion of granitic material along the bedding. planes, Caleareous rocks which have been inclided in the granitic and eneissic masses have a rec-hrown earthy appearance. They are massive, extremely fine- grained and show no alteration, unless their highly ferruginous nature is due to addition of iron oxides. The distribution of the gneisses within the batholith may be seen on the map. There is a change toward migmiatite well within the “batholith’. This is poorly deyeloped in the border zone and narrow belts of gneiss, but becomes important and at times dominant well within the batholith, Excellent examples of the grada- tion of injection gneiss to migmalite are common. The richly biotitic injection- gneiss gradually loses its dark bands, while the granitic bands expand and finally take up most of the rock, leaving only faint wisps of dark micaccous minerals winding in contorted lines through the granitic mass. Complete absorption of the sedimentary material is not common, for in most cases contorted, discontinu- ous faint wispy remnants remaim. Where the gneiss was coarse-grained the final product is a coarse-grained granitic rock or migmatile, The outcrops of gneiss aré different from those of the granite. There are no “tor” structures or pavements formed by sheeting, but instead these rocks form low outerops frequently running like blades or low rough walls for hundreds of yards, in parallel rows, Jointing is present only in the strongly granitic portions of the gneiss, thus it is not significant. Targe bodies of basic rock occur in the gneisses. The rock is-a fine-grained uralitized dolerite carrying epidote, and often it is associated with quartz reefs. Sunilar associations of basic rock with quartz are also met with outside in the country cock. The larger basics are marked on the map, but smaller occurrences are numerous. lV. THE COUNTRY ROCKS There are three main types, ws, slate, tillite and quartzite, and of these the slates have the greatest development. The tillites are distributed as one large formation, several smailer ones, and as minor intercalations in the slates, The quartzites occur similarly as one major band, and as groups of narrower bands. The former is, in the main, a thick massive quartzite with associated bedded qnirtzites, and it is interesting ta note that quire large ertatics are to be found in it m places, The several groups of thin quartzites are arkosic, presumably Nuvioglacial horizons in the slaty and tillitic Keds. Fracture cleavage oceurs in the massive quartzites near fold axes, although in argillaceaus quartzites there is a weak cleavage nearly always present. Jointing is strong mm the massive quartzites, The slates are strongly silicified and are of chocolate colour. In the mai! they are varved slates with alternate fine sandy and clayey bands numbering from 6-12 to the inch and in them cleavage is usually very well developed, but bedding is only obvious when the rock is weathered sufficiently to bring out the lithological differences. It is not always safc to nse the varving as bedding because of ite cross-hedded and irresular nature, Here and there the slates carry isolated erratics up to 10” in dimensions. Frequently the slates are without structure. even cleavage disappears, whet they resemble dark hiack-brown horufels, Occa- stonal intercalated earthy limestones occur in the slates which are discontinuous along the strike and are of no use as markers. Near the fluvioglacial sandstones 235 the slates change hy enlargement of the sandy component of the varying, and pass into arkoses. These arkosic horizons are made up of several bands of quartzilic rock up to 50 feet thick, separated by slate beds of sitnilar widths. The major tillitic horizom will be considered later, while the other tillitac beds are insignificant and may be passed over, There is the usual assortment of rock types as erratics, although granitic, gneissic, and pegmatite material dominates. Boulders range from small ones an inch or two in size tu larger ones nieasuring a foot or more in diameter. The base of this fillite is slaty or phyllitic, rich in miieacvous and chloritic material with a well-developed cleavage. V. STRUCTURE (a) Wiriin THE BaTHouri The structure is straightforward, and although locally confused, these small areas may be ignored in considering the general features. The folded map shows the regularity of the strike of the foliation in the eneisses, and in the granites, when preseut, The dips are steep to vertical with a tendency to slope slightly southward towards the contact. There is a tendency for one direction of jointing to follow the foliation, and for one or two others to develop as nearly as possible at right angles to it. (b) Ix tHe Country Rocks Great difficulty occurred in deciphering structures in those areas where alteratiun and injection by pegmatites occurred at a maximum. In the southern and western portion of the area mapping was sin\ple, but elyewhere there is s0 much alteration with the production of gneiss that bedding is obscured or even obliterated. The best marker beds available were plotted, thus a general idea of the structure was obtained, but there are many local dis- turbanees and discontinuities. On the edge of the altered area a syncline occurs with steeply-dipping over- turned limbs to the south, becoming normal with flatter dips to the north. Beds in the northern limb of this fold are strongly gneissic and carry numerous pegma~ tites, while the southern limb is made up of normal quarizitic and slaty rocks. Air photos suggest the continuity of this synclinal axis eastward beyond the zrea mapped. North of this syncline an irregular, indefinite anticline occurs, Bedding 1s ohseure atd indefinite here, but there is a general suggestion of an anticlinal structure. Pegmatitic injection accompanied hy intense ptygmatic folding is at its greatest development in thts region, and it is possible that these beds were converted to-a sen-plastic state during the falding, hence their unusually irregular puture (pl. xxi, fig. 2,3). The truncation of beds and their disappearance into masses of pegmiatitic injection gneiss is a common feature. Air photos indicate the presence of a fold axis extending eastward as a continuation of the anti- clinal axis through these altered rocks, The greatest density of pegmatites, the abundance of injection-gneiss and the irregular discontinuous structural features of the rocks, suggest, if one favours a granitization theory of origin, that the country in this vicinity is in an ailvaneed stage of conversion to granitic rock, and the area is thus comparable to the belts of gneissic rock within the batholith. If this is true, then in this Incality the “granitic front” is moving in along an anticlinal axis. 216 VI METAMORPHISM The study of metamorphic stages is difficult because im the places where change has occurred there is a great amount of pegmatite which would be expected to introduce complications due to introduction of material with great effects on the already stressed country rack, On the western side of the area where iolding is absent the sediments are normal varved slates with occasional interbedded quartzites, Towards the east. the alteration of the slates becomes progressively greater, thus half a mile west of the homestead the slates became garnetiferous ninscovite-quarte schist, while interbedded quartzites develop epidvi¢, actinolite and occasional red alimandine garnets. Amongst the injection-gneisses in the greatly altered country a milé south- east of the homestead there are gneisses: free, or nearly so, of felspar. They differ from the injection-gneiss by the abserice of thin quartzo-felspathic banding and may thercfore he considered as products of stress rather than injection and as representative of local advanced regional metamorphism, high-grade xnotted crystalline schists are associabed with these gneisses. The more arenaccous. beds associated with these occur as quartz mica-gneisses with epidote. These beds in the lesser metamorphosed areas are qeartz-smica-schists rather than the more common mica-schists which were yarved slates. Pure quartzites amongst the greatly altered rocks are converted to epidote- rich formations, in which bands up te 12” wide, rich im green epidote, traverse the rock parallel to the bedding. Occasionally epidote bands traverse the rock ai an angle to its bedding, filling joint-fissures, thus suggesting introduction of epidote-forming material from an outside source, ‘The presence of a little graphitic mica-schist is noted on the map showing mineral occurrences. This has a gritty quartzitic base, with argillaceous and graphitic material making up the major portion of the rock, Muscovite kyanite-schist occurs with kyanite up to an incl long, but it is restricted to certain horizons, The kyanite crystals are arranged. parallel to the schistosity and their presence indicales am advanced stage in the metamorphism. Some of the high-grade knotted mica-schists contain smali black prisms of tour- maine forming a brownish red mica-tourmaline knotted schist. Although garnets occut in the mica schists they are not found in the gneissic rocks, Lime silicates, such as sphene, zoisite and epidote, occur in the schists and gneisses. Thin impure siliceous limestones frequently occur amongst the varved slates and arkosic quartziles, thus it is not surprising to find these minerals developed in relatively large quantity in favourable places. Incipient growths of sillimanite occur in some of the gneisses, VIT, [IGNEOUS INJECTION IN COUNTRY ROCK PrGMATITic AND GRANITiIc Prasss Pegmatites are extremely plentiful, while there are a few = ee, SEH, vecnabinesiee\\ phy st eps Baie Ns By ish Bison ape) Pee ee MINES _BOOLC TA__GRANITE_ LOCATION OF MINERAL GCCURRENCES FOR tCeEceEne SEE FOLPER MAP Fig, 5 The principal mineral occurrences of the grea. 240) micas, Simiilariy, calcareous and tilhlic bands are without change right up to the contact. Although the line ol contact between the country and the granite rock may be mapped as a regular line, it has embavments and other irregulatities extending either side of a nican contact line, but they cannot he shown on the seale used. An unusually irregular portion of the contact is illustrated in fig, 3. The contact to the north-east is well defined with strong mylonite zones developed dlong it, The country rock against the mast easterly portion of the contact shows an enrichment in felspar, producing m grey-white fine-grained granntkir rock, X. GRANITIZATION PHENOMENA Evidence to stipport a granitization theory of origin is to be gathered from (1) the major folded tillite horizon an the south, and (2) from the batholith itself. The tilite forms excellent exposures where it is folded in the southern por- lion of the area, The voek, where well granitized, has a dense matrix of pink orthocluse, black biotite, muscovite aud quartz, in which erratics with indistinet outlines are set (plo xx, fig, 3). This matrix was formerly the normal phyllitic ar slaty type which is typical of tillites in this country. The stage of alteration of the rock is evidenced by its erratics, the outlines of which become faint in slrongly greuitized rock, while the original shape and lithological character are retained in unaltered portions of the tillite sull retaining a slaty base. Argilla- ceous erraties are dark and harnfclsic, with none of the original slaty cleavage. Quartzite erratics are unchanged except for re-crystallization which has occurred in some, producing a dense stvutcuire which is so coutinuous. that individual prains cannot be distinguished. Granitic boulders appear fresh and unaltered, probably due to recrystalliza- tion with the reconversion of secondary minerals to the original primiry ones, which would readily occur under granitizing conditions. Gneissic rocks appear fresh aid reerystallized with a strong gneissic structure, which may be the result of reerystalization of coniponents such as felspars, quartz, and micas, to larger individuals. The granitic nature of the rock as a whole is manifest by ils general appear- ance in Outcrap, namely, in the development of right-angular joint sets, and sheet jointing parallel to the surfaces. The granitization is nat homageneous through- wut, for unaltered tilltie may be in contact with strongly or partially altered material. Where granitization has been weak the illite retains its slaty-phyllitic hase, its grey colour, and its typical form of bladed cleaved outcrops. These weather relatively easily so that erratics lie scattered avout, and cavernous holes, he spaces once occupied hy erratics, appear in the outcrops. This contrasts strongly with adjacent rounded granitic-looking outerops with no locsened anil weathered erraties lying about, The solidity of the altered (illite is exhibited where, in the anticlinal Told, ic forms a great ridge covered with rounded platy outerops and granitic tors. Eancs af fiuvieglacials within the tillite take on an apltic appearance where the grajitizing effects are greatest. When weathered extensively, the granilized tillite looks like any decayed granite, The rectangular joints become enlarged and the rock Wecomes syollen and friable by internal expausion of felspars as these alter to vlays, finally leaving a sandy tmicaceous residue of soil surrcynding the outerops, : North-westward from the main anticline in tillite, a long ridge of altered tilite makes a strong topographical feanire, Ieregularly granilized (lite and Ihivio-glacials extend the (nll length of the ridge, but taward the creck at the north-west ond of the ridge only odd bands of well gtanitized rocke remain amongst onaffected tillites, {luvivglacials and varved slates. 241 A few hundred yards to the sorth of the granitized tillite here is a small patch of granitized country rock, Locally the structure of the varved slates heooumes unrecognisable, but in general they appear to “flow around" this small altered area, Various rock types occur within, including am abundance of injection-gneiss with granites and pegmatites, Quartzites, having wellives. Tn many cases they form irregularly-shaped bodies which are not truly dyke pe sill-like. The structure of the gneisses stugests replacement rather than intrnsion. ‘(here ts a general nodisturbed schistosity throughout the “batholith” sensibly parallel to that of the adjacent country rock close to the margin of the granite. Within the batholith there ts evidenee of two stages of the complete eranitiza- Lt prucess. The areas rich in mjection gneiss may represent the preliininary desilicalian stag brought ahout by telspathization, plus some degree of hasification. The greater part of these giieisses are felspar-rich because of numerous thin granitic bands within them, the introduced felspars being mainly microcline. perthite and oligoclase. Vhis indicates a considerable entichmerit in the alkalies potash and soda, tic. a desilication uf pre-cxistent coiutry rock. Furthermore, these gneisses wre niten rich in biotite and muscovite which represetits an increase in K,O, MeO and FeQ, for there 18 insuflicient ai these bases in the country rock to permit the direct production of these minc¢rals in the gneisses without outside wugmentation. These added constituents may have been eliminated from country already com- letely granitized, i.c., [rem country tow occupied by granite or porphyritic granite and may be regarded as temporarily concentrated in the gneisses (which are country rocks in the preliminary desilication stages). The next step would thecefore be the elimination of the MgO, FeO and excess K,O from the 2AZ gneisses into the surrounding country rock and the conversion of the gneisses into granitic-looking rocks. Apatite (representing P,O.), magnetite, and other iron ores (representing Pel)) are relatively abundant in the gneisses and may represent some added degree of basification. There is a similarity between the irregular granitization within the “batholith” and the gramitized tillite further south in which are found all gradations from completely unaltered tillite to crysialling granitic muterial in which the original etratics are scarcely discernible, Heuce iL would appear that granitization as a whole and the preliminary desilicavion process has operated selectively in tayvour- able places and on favourable rock, thus explaining why the tillite (as a whole) is granitized while beds above and below are not. THE POSSIBLE EXISTENCE OF “WRoNTS” Outside the batholith there is further evidence of granitization afforded by coticentrations of certain materials as mincrals in country rock adjacent to the hatholith. On the hill opposite the homestead, sphene, garnet, hornblende and epidote are abundant and may represent FeO, MgO, TiO, expelled from the batholithic area, while the occurrence of the allile magnetite rocks nearby suggests local conceniration of ntigrating Na,OQ and FeO. The quartzites have locally been altered to quartz-biotite-epidote-gneisses and, considering the large quantity of epidote present in them as well as the abundanee of biotite, sonie MgO and FeO must have been ijatroduced into these rocks from an outside source, vis., from country which has been converted tw granite. The occurrence of copper and iron adjacent to the batholith may also be sig- nifieant, having been driven forth from the granitized area and accumulated in these places, The baryltes which occurs clase by may have resulted from the deposition iu Javourable places of minor disseminated quantities of barium coam- pounds expelled irom: the gramilized sediments. The schists, gneisses and quartaites m which these mimeérals occur are in advanced stages of felspathization, manifest by numerous felspar-rich pegmatites ationgst them and the abundance of telspar and biotite in the wneigses themeelves, Hence it would appear that in the caslerm portion of the area where the country rock is strongiy folded and theroughly peritcated by pegmatites, and where the greatest variety of miuerals bearing basic constitucnts are concentrated, that a new centre of granitizalion is being established outside the batholith, pro- ducing rocks comparable to those of the gneissie belts within the batholith, Xf. CONCLUSIONS There is much evidence which suggests that the granite masses of Boolcoo- math are the result of @ process Of granitizacion rather than of intrusion. Similarly the alteration of the bordering country rocks may not be entirely due to -straight-out processes of regional metamorphism, bat inay be partly or even wholly duc to metasomatising effcets of cmanations driven forth from an adjacent area undergoing changes which eventually converted it to granitic rock, However, before indisputable conclusions cam be reached on these probletuis, more detailed work must be done. This particularly includes chemical work, in order to he able more accurately to state the relative gains and losses of substances in the several areas. Such conclusions as have heen reached here by field observa- tions and by the examination of thin sections, can be only regarded as tentative, otherwise false impressions may be gained. 243 Tt is hoped that they will be a useful guide to those who will probe more deeply into the origin of this granite mass and its possible relationship to the widespread. mineral occurrences in the Olary district. ACKNOWLEDGMENTS The author is entirely indebted to the Mines Department of South Australia for the time and facilities made available to him to carry out this study, and to Mr. S. B. Dickinson for help and advice on the work in general. BIBLIOGRAPHY Axperman, A. R. 1925 The Vanadium Content of certain Titaniferous Jron Ores of South Australia. Trans, Roy. Soc. S. Aust., 49, 88-90 Benson, W. N, 1909 Trans. Roy. Soc. S. Aust., 33, 132, 110 Dickinson, S. B. 1942 Beryl and Andalusite Occurrences north of Olary. Geol. Surv. 5. Aust. Review, No. 76, 80-81 Kvueeman, A, W. 1944 On the Analysis of Beryl from Rooleoomata, S. Aust. Trans. Roy. Soc. S. Aust., 68, 122-124 Mawson, D. 1912 Geological Investigations in the Broken IIH Area. Me- moirs of the Roy. Soc, of S. Aust., 2, pt. iv Mawson, D. Bad Amazonstone from Mulga Hill. Trans. Roy. Soc. S. Aust.. 31, 123 Mawson, D. 1916 Beryls of the Boolcoomata llills and Loadstone near Woman-in-White Mine, Boolcoomata. Trans, Roy, Soc. S. Aust. 40, 262, 263 Mawson, D. 1944 The Nature and Occurrence of Uraniferous Mineral Deposits in S. Aust. Trans. Roy. Soc. S. Aust., 68, 334-357 Mansrietp, L. 1947 Mulga Hill Barytes Occurrences. Geol. Surv. S. Aust, Rev., No. 87 Spricc, R. C. 1944 Uranium Occurrence near Amerco Hill, North of Olary. Geol, Surv. of S. Aust. Rev., No. 81, 91 Wootnoven, W. G. 1904 Petrographice Description of some Varieties of Granite near Olary, S. Aust. Trans. Roy. Sor. S, Aust., 28, 181-192 GEOCLOSGICAL PLAN BOOLCOOMATA GRANITE _AND ASSOCIATED GRANITIZED SEDIMENTS ee hd SCHIST WITH WINOR QUARTEIFES ___ — | GRANITIZED TILLITE ANG SLATE OQUARTZITIC SLATES WITH MINGA QUARTZITES AND TILLITES . terse HIGH GRADE SCHISTS AND INJECTION GNEISS PARTLY GRANITIZED SEDIMENT. —— EPIDOTIZED QUARTZITE LOW-MEDIUM GRADE ScHisTs_-——- | COUNTRY ROCK COMPLETELY GRANITIZED_|= COMPLETELY GRANITIZED COUNTRY ROCK WITH PORPHYAYTIC STRUCTURE... byte" as BRECCIATED ZOWES-.-~-..------- Rea Nt &y wy Br te ARAN MANS Nh AN eed oe Ne] PEGMATITIC (INTRUSIVE Oya ht i, ere 4] gti Met pty PSA MELAS CS cee ~~ 7 ‘L ee ak aaa Hn Vd seca) Ne eX ey De Basic tavhiteite gence nl ts NAW A Ps ad (VA eli ARS MRA ee a GRANITIC WORN QUARTZ REEFS. REFERENCE TO SIGNS GEOLOGICAL BOUNDARY OBSERVED... -- oe GEOLOGICAL BOUNDARY INFERRED STRIKE AND DIP OF FOLIATION IN GNEISS Pion STRIKE AND DIP OF GEDDING__.--~ ~~ Pon STRIKE AND DIP OF CLEAVAGE __ Adina STRIKE AND DIP OF LINEAR PEGMATITES__ Zw STRIKE AND DIP OF LOCAL JOINTING =..." CHAINS. 50 3 siLes. STRIKE OF LINES OF LOCAL SHEARING... “"* ANTICLINAL AXIS_..---------------. SYNCLINAL AXIS. WATERCOURSE....-.--------—-== ee GEOLOGICAL MAP OF AREA KN , Plate ms ~ Vol, . L948 Aust p S oc. ans. Roy. $ “re + “SyYRoOp Siu. Uy daeqs aynb sf Joewod ayy “punosmaa0f ayy UL 9zeTS SMYTH PUR ayTZ1aeENh YseEp IL puttosayorq ayy sordnsa0 JUNE OISSUS Pasnopos-jysry~ “ORO ayuUeEG IY | + “RIA INE UL OTUeIS sl XPyeM yoo0s AUT "SAULJNO JOUNStpur spay}? FO asnBIaq AQproyyIP YPM A]uO upas aq ABLE Sayesda B41 GOYA Ut ‘OW pazyiesry g Sry 3p *SMO1 [a[pRaEd ul pasguet4e azis Ub Soypur [RAIS spuoUIsesp aedspay 40 ayuURIs popunos aie SpuaMsEsy papnpaur ayy, ‘olurad payetooiq ul ,surppaq opnasg,, Z ‘Bie “e1O9a1g ayuess ayy JO aanjonays Syt Suniqwyxo ‘proysemopy ay} teau Juawaared Yyooy Tad Trans. Ri _ = in = ~ S. Aust, 1948 Vol. 72, Plate XNI Fig. 1 Fracture in erate bordering the hrecciated granite. The fractures are filled with richly chloritic and biotrtic pulverised granite. in wv wn s Y a s a oar = a oo o v oy) — ie z == =. Nak of osu 3 = 6 we ara! WKF a - 5 = om eo 1A =| so + Ng 1 Ps 375 =o sh & eh we Ob a ere 7 Sp iS me mB Sos = oS us a = a: 2e rs m ap = fein FE ws =, = SDs, £5 Aes a = v7 C 4 ‘S g = 5 oo ee A, C STURTIAN TILLITE OF MOUNT JACOB AND MOUNT WARREN HASTINGS NORTH FLINDERS RANGES By D. MAWSON Summary Since my earlier contribution (Mawson, 1934) on the general geology of this area, the broader features of the Proterozoic stratigraphy of the Flinders Ranges have been solved by a succession of reconnaissances extending over wide areas during the past twelve years. What were then referred to as the “Munyallina beds”, I hae since been ablt to correlate satisfactorily with defined horizons in the Proterozoic succession of the Adelaide System. As a consequence, this paper is being written with that object in view as well as to supply, as a result of a further examination of the region executed in the year 1939, a detailed cross-section through the glacigene beds. 244 STURTIAN TILLITE OF MOUNT JACOB AND MOUNT WARREN HASTINGS NORTH FLINDERS RANGES By D. Mawson * [Read 11 November 1948] lotreductory Remarks stu 1ifs ait tiax pus ws a ae ne 244 Cross-Section of the Glacigene Sediments of Mount Jacob ... - us 293 245 Cross-Section of the Glacigene Sediments of Mount Warren Hastings tab 247 ‘Vhe Post-Glacial Record within the area under consideration Au sats if 248 fhe Glacial Record .... ee sha cosh i, ct i 14 tym hase 249 The Pre-Glacial Record ‘ ey by wt ~ rash ty: P ay 250 Description of Plates - ws is ee G3 Li a a. 251 References mt nial : oy} bie set bes os ne -_ in 251 Since my earlier contribution (Mawson, 1934) on the general geology of this area, the broader features of the Proterozoic stratigraphy of the Flinders Ranges have been solved by a succession of reconnaissances extending over wide areas during the past twelve years. What were then referred to as the “Munyallina beds”, I have since been able to correlate satisfactorily with defined horizons in the Proterozoic succession of the Adclaide System. As a consequence, this paper is being written with that object in view as well as to supply, as a result of a further examination of the region executed in the year 1939, a detailed cross- section through the glacigene beds. x ARKAROOLA STATION \ ! SCALE a MT Warren Hastings : ed MCE ! | WOODNAMOKA WELL : ! AMI. Jacob MAC ing / 2 WOOLTANA H.5. Fig, 1 A very important matter left in doubt when my carlier contribution was published was the existence or otherwise of a fault along the face of the massive (ave Limestone formation. The possibility of a major fault in that location was * Geology Department, University of Adelaide. Trans. Roy. Soc. S. Aust., 72, (2), 30 March 1949, 245 discussed (Mawson, 1934, 188) but finally discounted, f#lowever, our later investigation, and sttatigraphical considerations, show this to be a major fault line, thrawing down some thousands of feet the block to the east thereaf; namely, the country between the Cave Limestone belt and Mount Jacob. Thus it is now demonstrated that the glacigene beds of the Mount Jacob Range are repeated by faulting at Mount Warren Hastings. _ Detailed cross-sections of the beds im these two localities are graphically displayed in fig. 2 herewith. The lines of section are marked on the locality map, hig. 1 CROSS-SECTION OF THE GLACIGENE SEDIMENTS OF MOUNT JACOB The mail coach track from Wooltana homestead to Paralana follows nort)), for the first four miles, along the fovr of the Mount Jacob Range, with the gentiv falling Lake Frome Plains extending beyond the horizon to the east, From this track, at a point about two miles nerth of Wooltana homestead, the geological section to he detailed below was run to the west over the Mount Jacob Range. traversing the complete thickness of the glicigene beds. These are clearly Sturtian in age. Herewith are descriptive notes dealing with each of the divisions recognised in the plotted cross-section (fig. 2), stated in ascending order from below upwards. 1, A few hundred yards across the outerop of poorly exposed pre-glacial sediments, amongst whielt slates and dolomites are conspicuous. Minor intrusive bodies of diabasé have heen reported (Mawson, 1926). 2. A belt of melaphyre. in places amygdaloidal. This may have been a laya flow in Proterozoic times. 3. The above melaphyre is overlain by a reddish sandstone, of some con- siderable thickness, dipping moderately steeply to the west. It appar- ently post-dates the melaphyre. In sev¥eral places it is greatly disturbed and shattered, suggesting that it is traversed by fault lines. {Late Tertiary) connected with the downthrow of the Lake Frome Plains to the east. Dip variable 30° t0 350° ta W; strike N. 2° K, 4. Melaphyre with some intercalations of a consolidated tuffaceaus nature form the lower face of the range. The relation of the red sandstone ta this basalt was not satisfactorily determined, the locality being dts- turbed by faulting. Llowever, westward frou that point, the sue- cession of beds is undisturbed by jaulting. This thick basaltic formation appears to be mainly in the nature of a lava flow and was originally scoriaceous in part, more especially at the upper and lower limits. Former stcambholes are iiow filled by secondarily introduced miterals, resulting in some locations of a truly perfect and striking example of an amygdaloidal melaphyre. The total thickness of this basaltic formation was found toe be 590 feet. Tis upper surface is irregular and had evidently been subjected in places to some crosion prior to deposition upon it of the overlying glaciyenc sediments, 5. Dark grey tu greenish-grey breccia. Near the base there are plentiful angular pieces and boulders of the underlying basalt, also abundant fragments of limestone and dolonite, anc some shale, quartzite and reef quartz, This is now interpreted as a glacigeie breccia. At about 300 feet aboye the base some bands exhthit evidence of water “I 16. 17. 246 True thick- Wess mm feel sorting. Above this level the tllite is of a more sandy nature; in other ways also there is indicated to same degree the effects of water sorting and elutriation. Erratics up to 2 feet in drameter were observed, In the upper part of this division, limescone erratics are scarce; of these quartzite is common, basaltic rocks are plentiful, and occasional erralics Of quartz porphyry were noted .,,. ~s0 rote ne “17 Tillite with mich fine base in whick are distributed erratic boulders up to 18 inches in diameter. Jaspero'd quartzite is the more abundant rock type represented among the erratics. Only a small percentage of the erratics shaw faceting : striae are rarely seen. A well striated example was ohserved partly embedded in the tillite at about 200 feet above the base of this division, At about 300 yards above. the base there are some uarraw glacin-fluvial sand bands intercalated in this tillite, At higher horizons, erratics exhibiting striations are more frequent. These etratics are still mainly quartzite, some of a white variety, others the jasperaid lvpe, Erraties of «quartz porphyry, con- siderably kaolinized, are less frequent ... ei put beh tie Glacio-fluvial sediments allernating with tillite, (a) Band of sandstone. 1 ft. in thickness. (b) Above is tillite, 6 ft. thick. (ec) Dand of sandstone, 1 ft. thiek. (d) Tillite, 34 ft. thick. (e) Sandstone in part exhibitmg 4-inch Jaminations, At the base ts a section of true varves; alternating laminae of red argillite and white sand. Thickness, 93 ft, Dig, 33° to W.; strike, N.5° L.. Typical tillite in which some erraties of an unusial red-calouredd porphyry were observed .... Se , : ie ne Commences helow as a sundy phase of tillité bitt passes wphdtd into a normal willie i which) small crratcs are studded through an abundant rock-flour base .... tee sate bi: er eit whee Glacio-fluvial, gritty sand-rock with some pebble erratics : Commences below as characteristic tillite but quickly changes above toa sandy phase. Near top are abundant pebbles ina very sandy base Glaciv-fluyial sandstone 2 feet thick below. followed by LO feet of sandy tillite, capped by 20 fect of glacto-fluvial sandstone we ats Somewhat sandy tillite with plentifil erratics Sandy tillite, less resistant ta weathering «.. no ews tart ces Characteristic tillite in which, towards the upper limit, along the stinmit of the Range, there are exceptionally large etratics up to 6 ft. in diameter; most of these are of quartzite, and many are well striated a wel hla ake * Tilhte with plentiful large crratics a a, post ett ways Characteristic tillite with many latge erratics exhibiting striated faces. Ati nustially good example of an ancient, typical tillite. Among the largest crratics are chocolate-coloured quartz-feldspar porphyry and quartzite, one of the latter measuring 5 ft. by 5 fu by 4 ft. This is the topmost of the glacigene beds and where erosion has stripped the 500 247 ‘True thick- mesa in feet overlying beds there is exposed the original surface of the old tillite formation. Thus can be observed of the old suriace a concentration of -erratics standing in relief by removal of the finer-grained elements ef the old botlder clay, evidently the reswit of strface wash. See pl. xxii, fig. 2... ue roa xf, ae p..t Bice ae 7) Yotal glacigene formation .... 2,547 18. Immedtutety above the tilhte ts glucio-lacustrine mud and an irregular band of limestone up to 3 ft. in thickness, followed by blue-grey shales, dipping 23° to the west. Strike N.8° E. (see pl. xxiii, fig, 1)- These post-glacial shales are, in part, finely laminated and correspond to the Tapley Hill shales of the Sturt locality near Adelaide, Ags they pass upwards, bands obviously somewhat calcareous make their appearance, and finally, at about 1,200 feet stratigraphicatly aboye the base, a thick limestoue formation with some cryptozoonic mottling is tet with, The general upward succession of strata to the west of the Mount Jacob Range ullite figured in the diagram herewith (fig. 2) and extending as. far as the vicinity of the Cave Limestone is well set out in the section on page 190 of my carlier reconnaissance (Mawson 1934). Note, howeyer, that a fault extends in a general north to south direction along the Cave Limestone belt throwing down the Mount Jacob range black. THE MOUNT WARREN HASTINGS BELT OF GLACIGENE BEDS This belt 35 a repetition by faulting of the Mount Jacoh tillite. Some varia- tion im details of the succeéssioi atid in total thickness is to be noted, but this may be expected as anticipated irregularities inherent in glacial sediments deposited al points several miles apart. A matter calling for special remark is the fact that whereas these glacial depositions at Mount Jacob rest upon basaltic effusive debris, there is an absence of such at the western outcrop, the glacial beds there lying directly on top of the undisturbed dolomite-magnesite-bearing series. This line of section is marked yn the locality map, fig. 1, Notes relating io the numbered divisions appearing on the diagraramatic cross-section (fig. 2) are given below, True thick. ness in feet 1. Preglacial shales and dolomites with some bands of sedimentary mug- nesite. These beds strike N,12° ©. and dip to the west, ranging between 28° and 34° (see pl. xxii, fig. 1). 2. illite with abundant dolomite erratics, OQuartzite erratics up to 3 feet diameter are also abundant and not infrequently show ¢lacial striae. A fine-grained grey granite is also represented among the erratics (see pl. xxii, fig. 2) “its ron sane st ves ai 50 Glacio-fluvial sandstone, siltstone and rock-flour beds with some erratics. Included are occasional bands a foot or two in thickness of typical tillite ages 340 st» vas Leg re sts a. 626 4. Typical tillite at base of this section with quartzite erratics up to 24 feet long; some large ones of grey granite and gneiss, At 50 feet abhave the base the tillite becomes more sandy until finally a ¢lacio- fluvial arkose; later, appruaching the top of this division, it reverts in nature to a sandy tillite .... tem dae abs bits un «= BA a) y) Zz < ia jaa) s . Tillite with well glaciated ertatics nites we his na ste 0. Glacio-Auvial silts and muds, containing large erratics in the upper sechou hed fast wee ie tase shed abe sate nie o4 7. A glacial nud base with plentiful very large erratics up to 2 feet diameter, Nearly all erraties are quartzite; one is a quartzite con- elomerate (pebbles in it also quartzite) -... oct Jase shi we 146 S. ‘illite and glaciotacustrine mud shales with plentiful latge erraties; one is 4 course granite fief ) - . ae we «218 & Typical tillite with abundant striated erraticy io 3 Jeet in diameter ; granite with quartzite common, Sonie erratics are greatly weathered basaltic rocks ‘ie yes sive ari sate ore ack we 390 10. Glaciottuvial sandstone —.... dt fa sere odes am sad 32 11. Tillite oie en at &l 12. Glacio-fluvial sandstone... ess fis ves eee oye Mee 75 13. Characteristic tillite crowded with erratics eT sete sits pa aE i4. A sand phase tillite seal ae ast fiat ois awe bY2 15, Typical tillite, rich in large erratics, many of which are striated wa «= 704 Total glacigene formation .... 3,098 16, Post-glacia! slates, strikiug N.12° E. and dipping west 33°, Several hundred feet ahove the iipper surface of the foregoing glacigenc beds, intercalated in this slate formation are thin hauds of limestone, then more slates and finally a thick formation of crystalline limestone at possibly 800. feet ahove the base. Dip here 36° to W. THE POST-GLACIAT. RECORD WITHIN THE AREA UNDER CONSIDERATION These outcropping belts of glacigene accumulations are each followed by a corresponding succession of post-glacial sedimenis, Namntely laminated rock-flour silts below, passing upward into flaggy beds, among which are occasional bands of a somewhat calearcous nature; these in turn finally lead to massive limestones. This also, it will be noted, is the order of succession above the Sturtian Tillite near Adelaide, which locality is some 330 miles further south. Tere, however, the laminated slates formation is much less thick, while the limestone develop- ment ts vastly greater and more varied. On my second visit to the Mount Jacob area, T did not re-measure the post- wlactal succession ahove the Mount Jacob belt; such data as appear in my eatlier paper (Mawson, 1934), though hurriedly secured, should be a satisfactory record up to the faulted area where chocolale shales appear; refer to item 10 of that record. Summarized, the overlying beds there stated include from helow upward about 1,250 feet of laminated slates and flaggy calcareous slates followed by over 2,000 feet of beds highly calcareous. These latter are massive limestone below, in part siliceous and dolomitic, passing above to flaggy, banded limestones with ussociated argillites, which bear the characteristic features of shallow water accumulations. They are rich in oolitic and pisolitic structures, as well as exhibiting several stromatolithic forms corresponding to types often referred to 24 records of calearcous algae, These latter can be seen also in the face of the Nepowie Rampart a little to the north of Baleanoona homestead. z49 The argillites and limestones of this belt are surprisingly (ree from any fori of advanced metamorphism such as has affected their equivalents in the Umberatana-Yudanamatana region sume 20 miles tu the west-north-west, THE GLACIAL RECORD Tt will be observed that the measured thickness of the glacigene belt at Mount Jacoh is only 2.447 feet, while at Mount Warren Hastings it is 3,098 feet. Perusal of the succession suggests that there is some indication of a rough corre- pondence in the deposits of the two localities, if read downwards from the tops in cach case. This suggests that the underlying surface upon which the glacial sediments were deposited stood at a higher level at ihe Mount Jacob locatiory than was the case further to the west; perliaps a result of the volcanic accumulations in the former locality. Details already listed of the changing nature of the glacigenc accumulations illustrate the interbedding of true tillites with glacio-fluvial deposits. The latter range from true varved sediments and well-graded sandstune, to water deposited arenaccous and argillaccous beds with occasional embedded erratics. The most massive tillites met with constitute the later depositions; there the erralics reach unusually large dimensions, Dr. Woolnough (1926) mentions having observed, in the more southerly extension of the Mount Jacob belt, a vitreous quartzite erratic some 9 feet in length. The glacio-fluvial sediments embedded in the glacigene successiom are nel anywhere of great thickness, so that they cannot be construed as necessarily inferring ati interglacial climatic break; they may haye resulted from purely local conditions. There is therefore in this arca nothing corresponding to the thiclc interglacial Wéillyerpa quartzite formation of the Bibliando Dome, which is distant little more than 100 miles to the south, As these glacial beds are all of one ¢poch, and as they are followed by a sedimentary sequence corresponding with the post-Sturtian succession, we conclude that the Mount Jacob tillites are of the Sturtian epoch, not of the earlier Bibliando (Mawson, 1948) period. Erranes embodied in the glacigene deposits include some granites anil por'phyries, so closely similar to types met with in sit in the Mount Painter com- plex that it seems certain that they were in fact, derived therefrom. These are mainly lioth grey and brown acid porphyries and both light grey and reddish granite. less often gneiss and schist are encountered. One quite remarkable erratic recorded is itself a tilloid rock, of which there is Tittle doubt that it is actually a ullite. Thus the existence of an earlier glacia- tion appears to be evidenced. Cormmonest of all are quartzites, and of then one type appears to be identical with the rock constituting the great quartzite formation overlying the central older Precambrian complex in certain localities, such as along the Arkaroola near the junction of Radimn Creek. Low down in the succession erratics of dufomite and limestone are rematk- ably abundant; a common type closciy resembles a limestone of the underlying SeVICS, In the Mount Jacob area the lowest part of the glacigene belt is remarkably rich in fragmenis' of the underlying basic igneous rocks. This leaves no doubt but that the ice responsible for the accumulation overrode areas of that basalt, These facts are evidence of the existence of jiee-covered land in the vicinity of Mount Painter during the Proterozoic glaciation, Jt is not necessary to look 250 ia the Gawler Ranges, as suggested liy Howchin, for the source of erratics of red to brown acid porphyry, for occurrences. of the kind do oceur in the neigh- bouring Mount Painter complex. THE PRE-GLACTAL. RECORD ‘The sediments underlying the glacial formation are arenaceous and argille- ccous flaggy beds with much interbedded dolomite and magnesite. This corre- “ponds to the magnesitiferous formation met with beneath the Proterazoic illite in many other parts af South Australia, for example with that recorded (Mawson, 1947) west of Coptey and at Mundallio Creek, The extensive development of sedimentary deposits of pellur magnesite (see Mawson and Dall- witz 1945, p. 23) below the Mount Warren Liastings belt of tillite is specially noteworthy, We did not meet obvious magnesite beds below the Mount Jacoh tillite outcrop, doubtless owing to the very narrow bell exposed to view, The pre-glacial basaltic igneous acuivity (Mawson, 1926) on a consitlerable xeale exhibited below the Mount Jacob helt and its absence in the section 8 miles to the west is evidence of its local nature, NE. COPDIERITE ~ PLEQUASTE— CORUNDARA = SCHIST ETC) re NAL Al ‘ “te ADELAIDE SYSTEM Pig. 4 By following the Wywyana@?) dowr to its jluiction with the Arkaroola, thence to the junction of Radium Creck, a complete cross-section of all under- lying sedimentary formations between the tillite and the old Mount Painter complex is traversed (sec sketch section, ig. 3). [ivst in the downward sequence comes ihe imagnesitiferous dolomite-bcarnig series, then a massive quartzite formation which shonld correspond with the Emetogo Quartzite (Mawson, 1947) of the wearer Flinders Ranges. Next comes an older succession: argillites and limestone, in part richly dolonvilic and sideritic, overlying a basal quartaite. Asso- ciated with this section of the record is rather abundant basalt mt part, at least, intrusive, Beneath this older group of sediments are the granites, gneisses and granulites of the Mount Painter canplex, The oldest series of sediments metitiohed above is referred to in the sketch- section as the Arkaroola Series. It is everywhere greatly inctamorphosed, [n come of its course the quartzite is rendered almost “vitreous”. The calcwrenus formation has been converted to actinolite-phlogopite-siderite schists and marbles. Calearcous slates haye been changed io scapolite-rich schists, We have met a corresponding sttecession in other localities when passing into the Mount Painter complex. In some other areas, this basal quartzite has heen feldspathised and otherwise incorporated into the underlying gneissic system. Roof pendants (Mawson 1923, p. 376) in the red granite cast of Mount Pitts, composed of corundum, pleotiaste and cordierite schists, appear to be highly metamorphosed fragments of this series, ~) "This creek iy spelt variously on maps in circulation. However, the speiling as bere given is that advised as correct nsage by Mr. Greenwood, proprietar af that area of country, 251 REFERENCES Mawson, D. 1923 Igneous Rocks of the Mount Painter Belt. Trans. Roy. Soc. S. Aust., 47, 376 Mawson, D. 1926 The Wooltana Basic Igneous Belt. Trans. Roy. Soc. S. Aust, 50, 192 Mawson. D. 1934 The Munyallina Beds, A Late Proterozoic Vormation. Trans, Roy. Soc. S. Aust., 58, 187-196 Mawson, D. 1944 The Nature and Occurrence of Uraniferous Mineral Deposits in South Australia. Trans. Roy. Soc. S, Aust., 68, 354-357 Mawson, D. 1947 The Adelaide Series as Developed along the western Margin of the Flinders Ranges. Trans. Roy. Soc. S. Aust., 71, 259-280 Mawson, D. 1948 The Proterozoic Glacial Record of the Bibliande Dome, Proc, Roy, Soc. N.S.W. Mawson, D., and Dattwirz, W. B. The Soda-Rich Leticogranite Cupolas of Umberatana. Trans. Rov, Soc, 5, Aust., 69, 22-49 Wootnouvén, W. G. 1926 The Geology of the Flinders Ranges, South Aus- tralia, in Neighbourhood of Wooltana Station. Proc, Roy. Soc, N.S.W., 60, 283 DESCRIPTION OF PLATES PLatTE xxii Fig. 1—General view of the outcropping magnesite-dolomite-bearing series under- lving the glacigene sediments. View looking south-west from a point about a mile north-west of Mcleish’s Well Thotre by R. H. Jones Fig. 2—General view of the lower part of the outcropping glacigene sediments where they overlic the magnesite-dolomite-bearing series about 14 miles west-north- west of Meclcish's Well. The base of the glacigene beds is the tilite knob showing as a dark patch near the left margin of the picture. Photo by R. H. Jones PLATE XXIil Fig. 1—The top limit of the Ullite, where is appears as a cliff face capped by a thin band of limestone followed by a thick series of laminated shales. Photographed in creek bed on the western side of the Mount Jacob Range at western end of section. Photo by R. H, Jones Fig. 2—The top limit of tillite on the western side of the Mount Jacob Range and west end of section. Here the thin band of limestone and slates capping the tillite has heen eroded back to reveal a fossil moraine surface studded with large erratics. Photo by R. H. Jones Trans. Roy, Soe. S. Aust.. 1948 Vol, 72, Plate NNT Fig. 1 Dolomitic series underlying the ghicigene belt of Mount Warren Hastings. Vi, 2 Lower portion of the Mount Warren Llastings glacigene series Trans, Rov. Soc, S. Aust, 1948 Vol, 72, Plate NNITI} Fig. 1 Glacio-Auvial sediments overlying the upper limit of the Mount Jacoh tillite. 2 Fig, The original surinace features of the Proterozoic tillite uncovered by erosion of overlying elacio-Auvial sedunents. THE SIGNIFICANCE OF THE OCCURRENCE OF FOSSIL FRUITS IN THE BAROSSA SENKUNGSFELD, SOUTH AUSTRALIA By PAut S. HOSSFELD Summary In a paper read before this Society in 1935 (Hossfeld, 1935) the writer, on page 51, referred briefly to the occurrence of lignitized fossil fruits below the present surface of the Tanunda Plain. These fruits were obtained at a depth of 320 feet below the surface from a bore just over two miles from Tanunda, in a direction a little east of south, on Section 650, Hundred of Moorooroo, South Australia, on the property of Mr. H. A. Lindner, who made the specimens available to the writer. 252 THE SIGNIFICANCE OF THE OCCURRENCE OF FOSSIL FRUITS IN THE BAROSSA SENKUNGSFELD, SOUTH AUSTRALIA By Paus $. Hossretp [Read 11 November 1948] In a paper read before this Society in 1935 (Hossfeld, 1935) the writer, ou page 51, referred briefly to the occurrence of lignitized fossil irnits below the present surface of the Tanunda Plain, These fruits were obtained at a depth of 320 feet below the surface from a bore just over two miles from Tanunda, in a direction a little east of south, on Section 650, Hundred of Moorooroa, South Australia, on the property of Mr. H. A. Lindner, who made the specimens avail- able to the writer. FOSSIL FRITS KITCHENER Trans. Roy. Soc. S. Aust., 72, (2), 30 March 1949, 253 They have been described and identified by Miss Helen T, Paterson, B.A., who is responsible also for the sections ahd. drawings. The descriptions and illustrations are incorporated in the present paper. A plan and a section are included in order to. demonstrate the writer's con- clusions regarding the structure of the area. The area described in this paper consists of the southern part of the Tanunda Plain, the lowest part of the area, and the higher areas to the east and west. The eastern section includes part of the Barossa Ranges in the vicinity vf Mount Jatchener, and the westeri section consists of part of the Gomersal Plateau. As will be noted from the section the area is one of diversified relief, pro- duced by the differential movements of [ault blocks and their subsequent dissec- tion and erosion. The area appears to have formed a part of (he extensive Aus- tralian Pre-Miocene Peneplain. Warping and block-fanlting, commencing probably in the Larly Miocene in soine areas, and corititing episodically until Recent times, have dismembered the region and destroyed much of the pre-existing drainage. The trough faulting which took place in this area resulted finally in the development of the Mount Kitchener Horst, descending by a series of step-fatlts to the lowest part of the fault basin on which Tanunda is situated, and ascending on the western side hy two or more step-faults to the comparatively low Goinersal Plateau. The maximum vertical movement known is in the vicinity of 1,400 feet, the height of Mount Kitchener being 1,965 feet, and the fossil fruit deposit approximately 560 feet above sea level. The Gomersal Plateau, before recent dissection partially destroyed it, appears to have had an clevation of between 850 and 900 fect above sea level. Mount Karceewen PLIOCENE FOSSIL ER Ay FRUITS GAWLER X WARCOOTA SERIES HORIZONTAL 1} JAGH = J] MILE SCALES | VERTICAL 1 INCH= 2000 FEET FIG. All of the above fault platforms, the postulated positions of which are marked on the plan, ate covered to some extent by alluyial deposits, which mask some of them completely, Thus ihe existence of the fault block an which the fossil fruits oecur, and which will be referred to as ihe Kabminye Block, could not be proyed nor its depth determined tuntil the bore supplied the necessary evidence. The fertile plain on which the towns of Nuriootpa and Tanunda and a number of villages are situated, owes its origin to the filling of a fault trough, chiefly by alluvial deposits from the Barossa Ranges which form its steep eastern margin. “ED wht > (7. 1 A 19. rst FIc. 3. Phymatocaryon Mackayi—lig, 1, Whole [ruit presenting sutural edge. Fig. 2, Valves separated, external view, Fig, 3. Valve, imner side. Fig. 4, Whole fruit showing its sumunit. Concotheca turgida—Fig. 5, Whole fruit. Fig. 6, Whole fruit showing sutural side. Fig. 7-9, Valves separated, with shrivelled seed. Fig. 10-12, whole fruits. Fig. 13- 14, Valves external, internal. Pleioclinis Couchmanti—Fig, 15, Side view of fruit. Fig. 16, Fruit seen from below. Fig. 17-18, Showing valves, Fig. 19, Transverse section showing seed. Fig. 20, Fruit seen from above. 255 The down-fauiting and warping which produced this fault trough extend much further, and although not marked by the deposition of the deep: alluvium characteristic of the Nuriootpa-Tantinda Plain, these extensions are so obviously part of the same system of crustal deformation that they have been grouped by the writer under the term “Barossa Senkungsfeld” and will he described aud referred to as such in a paper which is being prepared. The following descriptions, identification, scetions aud illustrations have been supplied by Miss Helen T. Paterson. Tn classifying the specimens Miss Paterson retained the nomenclature uses by Baron von Mueller, but as his work had not been revised according to the present knowledge of our flora the affinity to the existing genera has been shown where possible. NESCRIPTION OF FOSSIL FRUITS FROM SECTION 450, HUNDRED OF MOOROOROO, SOUTE AUSTRALIA Family SAPINDACEAE Prywatocarvan Macnavi Mueller Phymatocaryon Mackay, TY. v, Mueller, 1874, p. 11, 12, pl, it, fig. 1-15, Two specimens of an oval drupaceous Ituit, somewhat pointed, with an irregular muricated putamen, 3-valved. The furrows run in symmetrical form from hase to apex. One specimen is very compressed as though subjected ty pressure. These fruits resemble the living forms of Llacocarpus, for example. FE. grandis, and correspond to those described by Baron yon Mueller, 1874-1883, in his Observations of New Vegetable Fossils of the Auriferous Drifts, as Phymatocaryon Mackayi, and by Henry Deane in his Tertiary Fossil Fruits from Deep Lead, Fuster, South Gippsland, 1923, p, 490, pl. bx, fig. 15-16. Several fruits of this genus are also described by Ettingshausen, Tertiary Mora of Australia, 1888, p. 157, pl. xiv, fig. 3, 4, 5, Sa; p. 63, pl. vi, ig. 9-12 These were collected from Beaconsheld, Tasniania; apd Elsmore, New South Wales. ConcorHecs TuRcIpA Mueller Concothece turguda Wy. Mueller, 1874, p, 42, pl. x, fig. 6-12; Deane, 1923, p. 491, pl. lx, fig. 11, 17-21. Several small capsular fruits occur, globular in shape, cavity deep and roumi. Size circ. 3” long. Valves shaw an acute line at the edge and a partial aril or prominence. These fruits resemble those found at Nintingbool in the Older Plio- cene drift and also at Tangil, as described by Mueller under Concotheca turgida In his Tertiary Vossil Pruits fram Deep Lend, Foster, Deane identifies several fruits similar to C. turgida as belonging to the family Sapindacese, the genus Alectryon (Nephelium 1. y, M.) of the preseni-day Hora suggests some afhnity. PLeiocrinis Covcuwaniu Mueller Pleioclinis Couchmowi FP, v. Muclier, 1883, p, 19, pl. xix, fig, 1-11. Observations—A fruit resembling a nutmeg in appearance, 7-valved, 1” long, wide, ovate. The outaide tubercular and of woody hardness, ane seed only, not well developed, oblique with a minute apex and a smooth thin testa. This fruit is distinguished from the genus Penlenne by the number of its valves, Similar specimens have been found at Maddon and Nintinghool. The present-day genus Pletugyitiner, Engler, shows the same plurality af valves- ae a 3]. Penteane trachyclints—Fig, 21, Front view of fruit. Fig. 22, End view. Fig. 23, Section showing valves and seeds. Rhytidotheca Lynchii—Fig, 24, Whole fruit. Fig. 25-26, Valves and seeds. Fig. 27-30, Valves of smaller fruit, external, internal, side views. Spondylostrobus Smythii — Fig. 31, Whole fruit. Fig. 32, Fruit seen from above. Fig, 33, Section showing seeds. 257 PENTEUNE TRACHYCLINIS Mueller Pentoune trachyclinis F, y. Mueller, 1874, p. 22, pl. viii, fig. 10-17. Observations—a large fruit 14 long, 1” wide, broadly ovate; consists of five valves to the base. These are thick and woody, seeds not developed. Similar fruits have been found at Smythe’s Creek and in the Tertiary traver- tine at Geilston Bay, Tasmania, Henry Deane in his Fossil Fruits from a Deep Lead, Foster, compares fruits found there with P. Clarket, F. v. M. (1923, p. 490, pl. Ix, fig. 1, 2, 3) and the present-day Owenta venosa, but points out the difference in the number of valves, only one out of three being 5-valved. These fruits show a similarity in shape and number of yalyes to the family Meliaceae, which is not represented in Victoria at present. Family MELIACEAE Raytinotneca Lyncoi Mueller Rhytidotheca Lynchit F. vy. Mueller, 1874, p. 15, pl, iv, fig. 1-8, Observalions—Several boat-shaped valves occur, c. 2?” long and 4” wide, with roughened tubercular strfacé. These resemble the genus Flindersia, but its wing-shaped seed in a decomposed state has not been preserved, In F, v. Mueller’s Fossil] Fruits a similar type of fruit is figured as Rhytido- theca Lynchi. This species shows an affinity to the present-day Plindersia. Henry Deane describes and refers two fruits from a Deep Lead, Foster, to RK. Lyncha, but stresses their resemblance to Flimdersia maculosa. These valves are of a smaller variety. Two complete, oblong, capstilar fruits, one 13” long, 4 wide, have each five roughened valves, not echinate, and contain a woody placetita with flat seeds on either side. These fruits may be referred to the same genus, Family CUPRESSINEAE SPONDYLOsTROBUS Smyth Mueller Spondylostrobus Smythii F. vy. Mueller, 1874, pp. 8-9, pl. i, fig, 1-8; Ibtd, 1883, p. 13-14, fig. 5A, 5B Observations—A large fruit, 14” long, 3” broad, consisting of four valves with wide ridges at the apex, becoming convex and narrow at the base. Seeds two, free from the yalve. Similar fruits have been found at Haddon, Victoria; Orange, New South Wales; and Launceston, Tasmania. There is no evidence of this type in the present flora; the nearest affinity is found in the coniier Callitris and its sub-genus Frenela; these plants show a similar verticillus of fruit valves. Discusston A review of the available literature stiggests that while these fossils appear to be of Pliocene Age, their position within the Pliocene catnot be determined at present Further collections and investigations might modify the above age determina- tion. Whatever the final verdict as to their age may be, these fossils supply evidence of the beginning of downwarping and possibly faulting, in the area. The dismemberment of the peneplain in this area appears to have begun by crustal warping, which eventually culminated in the development of block-laulting in 258 portions of the warped areas, Further to the north, warping produced monoclinal flexures which can still be recognised. If the fossils are definitely of Pliocene Age then the movements must be at least as old as the Pliocene. These movements have continued episodically until recent times and may not have ceased entirely; alluvial, lacustrine and fluviatile deposits covering the Pleistocene-Recent Period occur in a number of areas in the Northern Mount Lofty Ranges. There is reason to believe that other deposits similar to the fossil fruit occurrence in the Kabminye Block occur, not only beneath the Nuriootpa-T anunda Plain, but possibly also in other similar localities such as the Dutton Plain, the Mount Crawford Plain and others. Although the existence of lignite of Pliocene Age is a distinct possibility, it appears likely that, should such deposits exist, they would be comparatively small, and probably uneconomic because of other adverse factors. Acknowledgments—The writer is indebted to Mr. H, A. Lindner, of Kabminye, for the loan of the fossils; to Mr. H. B. Lindner for his assistance to the writer in many ways; to Miss Helen T. Paterson for the identification and description of the fossils; and to Mr. W. Riedel for assistance in the age determination of the fossils. REFERENCE Hossretp, P. S. 1934 “The Geology of Part of the North Mount Lofty Ranges. Trans. Roy. Soc. S. Aust., 59 THE GEOMORPHOLOGY OF COUNTY VICTORIA, SOUTH AUSTRALIA By T. LANGFORD-SMITH Summary The geomorphology of County Victoria is analysed on the basis of three distinct zones, known respectively as the Pirie Plains, the Central Highlands, and the North-Eastern Plateau. An attempt is made to interpret certain structural and landform phenomena within these zones, particular attention being paid to the sand dunes of the Pirie Plains and the ridge-and-valley structure of the Central Highlands. 259 THE GEOMORPHOLOGY OF COUNTY VICTORIA, SOUTH AUSTRALIA By T. Lancrorp-Smiru * [Read 11 November 1948 | CONTENTS Page ABSTRACT a - Lg i. a oe . % 2 f a “ .. 259 INTRODUCTION a te “— J, ats to ‘4 ts ise ct .. 261 PHYSIOGRAPHIC ZONES af .. be 4 a 4, 7 Es we .- 261 A. Pirie Plains ‘. = a y; rw ue > be ‘3 .. 261 (i) Shoreline . a os _ oe eo ~~ = .. 261 (ti) Tidal Flats . te ce e as _ ze _ .. 262 (iii) Broughton Plains ve 23 ~ a ot .. ae -. 262 (iy) Broughton River Flats ve i) a4 4 a tf .. 262 (v) Sand Dune Belt ,, “y a Pr . - - ie _. 263 (vi) Napperhy Conoplane ., Aes 4: os os ze a 1. 264 (vii) Napperby Pediment .. a ar ry ss a 4. -. 264 (viii) Milcowie Bleck ,. ws “s ry Eas * i + -. 265 B. Central Highlands - se 3s we - 4 an ie .. 265 (3) General .. is ra bs ae 1s al _ 265 (ii) Ridge-and-Valley Structure 265 (ii) South Flinders Fault-Line Scarp a = sf “ .. 266 C. North-Eastern Plateau .. as be ‘ ie 44 ts ae .. 266 TERTIARY GRAVELS in at .- le es we os “ _ _. 267 Tectontc MoavEMENTS +. iw “4 .. 7 =} 4, ate i .. 268 Main Tectonic Associations .. An a J we =! ye .. 268 Age of Movements... os zs Ra 7. H us is es .. 268 “Dotible Planation"” Hypothesis an Bi. ds wre _ zt b .. 269 Dratnace SYSTEMS .. 3 $4 14 Ae se 4 aie ras als .. 270 General .. t3 34 x an 06 oy no .. ke re i arr “Stream Capture’ Theory .. “4 + tt re ta : - .. 270 Broughton River System nes Kile +] "s aif is A .. 272 ACKNOWLEDGMENTS - ict ze an ve ds om ote Ss .. 274 REFERENCES .. qs nA - fF he 3s qa af wi se .. 274 ABSTRACT 1, The geomorphology of County Victoria is analysed on the basis of three distinct zones, known respectively as the Pirie Plains, the Central Highlands, and the North-Eastern Plateau. An attempi is made to interpret certain structural and landform phenomena within these zones, particular attention being paid to the sand dunes of the Pirie Plains and the ridge-and-valley structure of the Central Highlands. 2. Attention is drawn to outcrops of Tertiary gravel within the County, and the probable relation of this gravel to Tertiary and post-Tertiary topography and drainage patterns. 3. An attempt is made to correlate tectonic movements that haye taken place since the Tertiary. 4. The drainage systems within the County are discussed, and evidence is submitted in support of a “stream capture” theory which would account for anomalies in the course of many of the existing streams. * Department of Post-War Reconstruction, Canberra, A.C.T Trans. Roy. Soc. S. Aust., 72, (2), 30 March 1949, souoz wWyduadons«yg ple wsayjeg aseutig] § “A rey ) = ‘aIMowoRooNg . 938 : PAN Be Sc 3 tu f a ay ff! , 4 oh o a Me ( g ponies weehycea., a a! i _A\sNBo, =m ' = » ral wns aD = Wy LVaa { \ JaHmeG 32 N JB gy & anws [eS 9) we btn, Je ~~ Ht : 2 + § S \y ty A > — 7 ¢ LOS LIS \ ~ i.) [> shea ~ E fr < NM -* big e — a J ae /( iy f = Sa >, vy fq R! { 3 AN ta yeM dss = * ; Sr H EN aynat { f\ bee Bo nner ERM UYrS [enema ae wi Ss uve} SNIVIE omer ba : ba? Ay Saket BLHERON Me Vy noni 2 Re, | eo ‘ Si ‘ ‘t <,_ \ f 4 | iv add. 4 \ i 4 + W'S‘WIHOLOIA ALNNOD /SHNOZ DIHdVYOOISAHd GNV NYaLlvd JOvNIvUC 261 INTRODUCTION This is the second paper on the Geography and Geology of County Vic: toria.“) It is an attempt to analyse the physiography of the area, and then to interpret, on a geomorphic basis, certain structural and landform phenomena. Air photos to a scale of four inches to a mile were available for the eastern half of the County, and these were used extensively during the field investigations, and also for mapping details of stream patterns. A contour map of the County was compiled from the photographs with the aid of engineering levels and some aneroid heights, and this has since been published in C.S.I.R. Bulletin No. 188, 1945. The drainage systems in fig. 1 are based on this map. PHYSIOGRAPHIC ZONES There are three major physiographic zones in County Victoria: the Pirie Plains, the Central Highlands, and the North-Eastern Plateau (see fig. 1), OF these, the Central Highlands is by far the largest and occupies all the central part of the County. Only a small part of the North-Fastern Plateau (mostly confined to the Hundred of Whyte) is included in Cotinty Victoria, but this zone extends over a considerable area otstside tlhe County boundaries. The Pirie Plains include the regions between Spencer Gulf and the Central Highlands. The three zones are discussed in detail as follows: A. Toe Pirte PLarys These are defined as the low-lying coastal areas which are bordered on the cast by the highlands and on the west by Spencer Gulf* The parts of this zutie in the neighbourhood of the town of Port Pirie have been described in some detail by Martin (1939) in a regional study of the Port Pirie district, Martin has adopted a subdivision of the area which lus bean extended in ihe present study to include the whole of the Pirie Plains, The writer is of the opinion that Mattin’s classification would be difficult to improve, and the only aniendments suggested are the insertion ef an additional very narrow subdivision, the “Nap- perby Peditnent,” and the use of the spelling “plane” in preference to “plain” in “Napperby Conoplain.” Martin's subdivision “Pirie Plains” has been changed to “Broughton Plains” to avoid confusion with the larger physiographic zone discussed in the present paper. With these slight amendments, the classification reads as follows (sce fig. 1) (1) Shoreline (v) Sand Dune Belt (it) Tidal Flats (vi) Napperby Conoplane (i) Broughton Plains (vii) Napperby Pediment (iv) Broughton River Flats (viii) Mileowie Block (i) SHORELINE Within the limits of County Victoria the constline is approximately arcuate. Although it ts not broken by any major irregularities, there is an intricate network of small inlets and tidal waterways. Mangroyes flourish, and give to mtich of the cuastul area the appearance of a tidal maigrove woodland. The coastal plain adjacent to the shoreline is very fat, and the sea very shallow. There is thus an extensive tidal area. ©) The earlier paper was entitled “The Geology of the Jamestown District, South Australia,” Trans. Roy. Soc, 8. Aust, 71 (2), 261-205, 1947. ‘The ficld worl for both papers was undertaken by the writer in 1941 aud 1942 while a inember of a C.S.1.R. Soil Survey Party. * See pl xxiv, fig. 1. 262 The larger inlets are Port Pirie Creek, Fisherman Creck, and Port Davis Creek, Fenner (1931) considers that Port Pirie Creek was once the mouth of the Broughton River, and Martin thinks that Fisherman Creek and Port Davis Creek are also the abandoned estuaries of former rivers, Martin (1942) has noted the presence of submerged off-shore bars in the neighbourhood of the Fisherman Creek inlet. These bars are several feet below sea level, The characteristics of a shoreline of emergence as outlined by Johnson (1919) correspond very closely with those of the shoreline under discussion. Examination of the dune ridges inland shows that considerable emergence must have already taken place. However, the present existence of off-shore bars is indicative of the initial stages of a period of emergence. It is therefore probable that the whole process of emergence has not been continuous, bit bas occurred in stages which have included minor static periods or even briet periods of sub- mergence. (ii) Tirpau Frats Martin has defined this area as “that lying belween sea level and the western Innit of the red sandy soils on the (Broughton) Plain to the east.” Phe whole of the area is not continually subjece to tidal actiony the higher parts are only occasionally flooded, following abnormally strong winds and high tides. Swamps are a feature of the area, The soils are saline and are heavy in texture, cracking extensively when dry. In times af heavy rain the area receives additions of clay and silt which are carried down i suspension from the conoplane which lies eastward. The tidal flats are characterised by ihe growth of saltbush and samphire, zens (in) THe Brovcntow Piatns These are hounded on the cast by the sand dune belt, and on the west by the tidal flats. The plains are divided by the Broughton River Flats into narthern and a southern sector. The Broughton Plains are found cnly below the LOO-ft, contour. They are not perfectly Hat, but include a number of shallow depressions which become swamps after rain. Martin has indicated that these depressions are due to the action of streams which are now extinc!. IIe has found Lurther evidence of past stream action neat the town of Port Vizie, If the Broughton River once flowed to the sea via Port Pirie Creek, it mus! have since passed through a period during which it wandered across various parts of the Plains, and in this case could have been responsible for many of the stream relics seen today. (iv) THe BroucntTon River FLats These are the flood plains of the Broughton River, They cover an area some 5 miles in width and 16 miles in length, stretching [rom just west of the town of Crystal Brook to the Tidal Flats near the coast, There is a rather abrupt break of slope in both the Broughton River and Crystal Brook, as each of these streams emerges from the hills of the hinterland on to the Pirie Plains, and in consequence there has heen considerable deposition of suspended material. During floods a large part of the Broughton Flats is covered by a sheet of water, which in many places does not completely drain away or evaporate for some days. With each flooding’a fresh layer of silt and clay is deposited. The Mood plains are almost completely flat, falling into the slope category of O° te 0° 30’. 2635 (v) Tae Sano Dune Ber This consists of a series of undulating dunes, which show a tendency to ran in a north-south direction, forming a succession of ridges approximately parallel toa the coastline. The north-south trend is well defined in the northern sector, ear Port Pirie and Warnertown, In the Hundred of Wandearah to the south the dunes are more irregular, although the north-south tendency still persists; they are here more gently undulating than to the north, The dunes rise to a maximum of about 200 feet above sea level in the south: castern corner of the Hundred of Napperby. In general, however, it is mnusual jor the height to exceed 100 feet ahove sea level. Throughout the area, the crests of the dunes are rarely more than 50 feet above the neighbouring swales; 20 to 30 fect is more usual. Beeause of the amount. of loose sand that is blowing about at the present tiie as a result of recent erosion, there is a common inipression that the whole dune system is unstable, and that the dunes themselves are migrating, On investt- gation, however, the majority of the dunes were found to be supporting a stable growth of saltbush, bluebush and acacias, with some mallee. The degree of erosion to which they had been subjected was rarely greater than that required in expose the roots of the acacias. This in itself would indicate that although tigration may have taken place in past geological time, there has not been any recent mass movement, Additional evidence was afforded by the profiles of the dunes. Whetiever these were examined they were found to conform ta those of mature, fixed soils. Grey-brown to brown mallee sand (Wandearah series) was the dominant soil type. Usually there was well-developed limestone rubble in the subsoil, ancl in some cases a layer of shect travertine, Features such as these are characteristic only of fixed dunes. The impression that the dunes are migrating is caused by erosion of portion mf the topsoil, which is almost pure sand, This erosion has been instigated hy gazing and, if: some eases, by cropping which has disturbed the surface and ittterFered with the narmal vegetation cover. It has required very little removal vf topsoil to build up large masses of loose sand, which at present are continually ivitting. I parts these sand drifts have caused considerable obstruction to roatt transport, especially in the Hundred of Wandearah, In ihe past there has been some tendency to assume that wind is the only eroding element which seriously affects regions such as the Sand Dune Belt. However, there is a growing inclination to place far greater significance than tormerly on the relative mportance of water erosion in arid and semi-arid areas. Although the greater part of Port Pirie’s rain occurs durmg the winter as a result of the southern low pressure systems, nearly one-third of the yearly total of 13” falls in the period October to March, and much of this in the form of thunder- storms and rain of high intensity. Once soils of light texture such as those in the area concerned have been cleared of their natural cover, they react readily ja the erosive effects of heavy rainfall. The compactness of ihe upper horizon is vonsiderably reduced, and even if there js no actual transportation of the soil it Pacomnee extremely susceptible to wind sorting when dry and commences to rift, Reports from the Ifundreds of Pirie and Wandearah describe abnormal stream How phenomena in some water-courses after heavy rain, for it would appear that the direction of flow is not always constant. Since the gradient of the water-courses in these areas is yery slight, any drifting of loose material along the beds of the streams is likely to have a damming effect in the event of heavy rain, resulting in the ponding of water. In such a case, unless the water breaks 264 ihrough the obstruction and resumes ite fortner course, it will eventually build up until it finds a fresh outlet upstream, and a new stream system will come into being, The trend of the sand dunes parallel to the coast, their “fixed” character, and their relation to the shoreline all indicate that they are old shore ridges, vepresenting successive stages in the history of a prograding shoreline, The probability Uiat they came into being following emergence of the coastal area has heen noted by Martin, who also mentions a view which has been expressed to the effect that the dunes came into existence following man’s settlement of the area. The latter hypothesis would be completely untenable in view of evidence available concerning the fixed nature of the dunes, although there is little doubt that most of the sand at present drifting is the result of erosion which followed varly human activity. fvi) Tae Napreersy CoNoOPLANE The Napperby Conoplane and Pediment are transitional zones between the Pirie Plains and the South Flinders Range. Martin has included in his Cono- plane the areas approximately between the 200’ and 500’ contours. The writer has inserted a “Pediment Plane” in the upper limits of the zone, and this has reduced the upper boundary of the Conoplane to about 400’, The Conoplane consists of a iairly regular series of parallel alluvial cones, composed uf material brought down from the steep western scarp of the South linders Range by short fast-flowing consequent streams. The longitudinal profile of cach of these streams shows an abrupt break of slope at the foot of the Pedi- ment Plane, where a vast quantity of coarse rock miaterial and sediments has been deposited. “Che present streams are not entirely responsible for building up the higher parts of the cones, for here there is a corisiderable accumulation of talus and colluvial material which has been brought down directly from the searp and carried across the Pediment Plane. Further down the alluvial material lecames progressively finer in texture, and where the cones fan out on the Pirie Plains it consists almost entirely of sill and clay. Each heavy rain results in the deposition of a fresh layer of snaterial, and roads across the Iower parts of the Conoplane are frequently impassable aiter raimas a result of freshly deposited sediments. In the lower part of the alluvial cones the streams lose the well-defined channels which are characteristic of their upper courses, and numerous distribtu- laries appear. (vii) ‘Tro Narversy PepiIMent This subdivision is mainly of academic interest, for it is very much smaller it arca than the others. It is found wt the base of the western searp of the Flinders Range in the Napperby region and occurs as a narrow, steeply sloping plane, which grades into the Conoplane in its lower limits (see pl. xxiv, fig. 2). The width is variable, ranging from a quarter of a mile to two miles. The angle of slope is also somewhat variable, hut averages about 6°. A useful section through the plane is shown by the Napperby Creek, ‘This exposes bedrock in many places, particularly in the upper limits. It is significant that the Pediment Plane is too steep to permit accumulation vf allavial and colluvial material, and yet on the whole is not subject to down- ward degradation like the tmotritain scarp. At the same time it is underlain by bedrock which is part of ihe range system. It is therefore probable that the plane is dominantly one of “lateral planation,” corresponding to Johnson’s definition (1932) of a pediment plane, 265 From evidence of shatter zones in the bedrock seen im the creck beds near the lower limits of the Pediment, it would appear that the junction between the Pediment and the Conoplane may represent the approximate position of ahe original line of fault, and that the mountain scarp has retreated castwards a dis- tance equivalent to the width of the Pediment—that is, between a quarter of a mile and two miles. (wil) Tae Mircowte Brock Like the Napperby Pediment, the Milcowie Block is transitional between the Pirie Plains and the Central Highlands. Jt is a relatively small area between Crystal Brook and the southern extremity of the South Flinders Range. Martin has noted that its western boundary coincides approximately with the 450! con- tour, and is represented by a low but well-marked scarp. The Crystal Brouk- Hughes Gap road approximately constitutes the eastern boundary, The maximum height of the area slightly exceeds 600 feet above sea level, the average being about 500. feet. Where the Sand Dune Belt adjoins the Milcowie Block, it is appreciably higher than the remainder of the Gelt, and it is possible that the uplift which prodtced the Block was responsible for the extra elevation of the sand dunes in this area. Lf this he so, the Milcowie Block must have been uplifted subsequently to the major faulting responsible for the adjoining South Flinders horst, for the sand dunes are tnore recent im origin than the fault scarp. B. Tue Centrar, Hicaranps (i) Generac luland from the Pirie Plains ts an upland zone consisting of a complex series of alternating ridges and valleys. Except for a small atea in the vicinity of the Bundaleer Reservoir, these ridges and valleys are parallel and have a remarkably consistent north-south trend. Three high ranges are included in this “ridge-and-yalley" area, They are the Sauth Flinders Range, the Campbell Range, and the Browne Hill Range: both of the latter ate technically within the North Mount Lolty Ratuge system- The highest of the ranges is the South Flimders. which reaches a maximum of 2,328 fevt at the Bhat. Bordering the northern part of the Browne ITill Range in County Vietaria. the upland area gradually toses its “ridese-and-yalley” structure and merges inte a plaienu area. This has been termed the “North-Eastern Plateau” for the pur- poses of the present study, and is discussed later under this heacling. (it) Runce-anp-VaLLtey StRuctTurr ln the past there has been no detailed investigation of this structure, whiel is the most prominent physiographic feature of County Victoria (see pl. soxiv. fig. 3:and 4). Fenner (1931), commenting on the popular view that the ridge-and-valle; structure has been due to tilted blocks, draws attention to the possibility of differential erosion along the strike of hard heds as an explanation. He notes that in any case the true origin of the structure has yet to be determined, Tilted blocks have played an important part in producing mtich of the structure of the Mount Lofty Ranges to the south, but except for a relatively small area in the vicinity of the South Flinders horst, the structural geology of County Victoria is notably different from that of ranges near Adelaide, Over a large part of the vastern portion of the County, the writer’s investigations (Langford-Smith 1947) have shown conclusively that the tilted block theory is not tenable, Here there has been very little major faulting, and the country is folded into a series of synclines and anticlines. Some of the present yalleys conform to the original 266 synclines. Examples of these may be found in the wide valley containing the township of Caltowie, and that containing the headwaters of Belalie and Fresh- water Creeks. Other valleys ate eroded anticlines, an example being the broad valley in which Jamestown is situated- In almost all instances, the crests of the ridges are strengthened by outcrops of highly resistant quartzite. These are particularly prominent im the eastern and western borders to the Jamestown valley, which are represented respectively by the Browne Hill and Campbell Ranges. These ranges are the upfolded margins of the adjaining synclinal valleys, and the quartzite which outcrops pro- minently in the ridges is associated with the {illite series and its subglacial heds, Differential erosion has therefore played an important patt in the development of the existing ridge-and-valley structure. The folding which produced the original synclines tonk place during one of the great Palaeozoic mountain-building periods (Mawson, 1942). Geological sections of the Jamestown district (Langford-Sumth, 1947) indicate the vast amount of erosion that must subsequently have occurred to produce the Tertiary pencplane. In the ridge-and-valley areas of the Central Highlands the mai features uf this Tertiary landscape have been preserved io the present day. Sonmw estimate of the minor degree of modification which has taken place caw be made by 2 study of the Tertiary gravels and laterites which occur as residuals. -aeey (iii) Souty Fiinpers Pavurt-Ling Scaapr The western scarp of the South Flinders Range is a particularly prominent feature. The scarp is very precipitous, and in the vicinity of the [luff it rises 2,000 feet within two miles, attaining a maximum height of 2,528 feet above sea level. The line of the scarp is partictilarly straight, running almost due nerth and south. Evidence from gorges and creek beds indicates that the position of the Tantt or serics of fanlts which have been responsible for this elevated block is com- paratively cluse to the present scarp, and tho latter must therefore be classifier as recent. As was noted earlier, it is probable that the original line of fami! cortesponds to the lower litnit of the Pediment Plane, This was shown to be the ease in the Port Germein atea, some 10 miles to the north, but within County Victoria conclusive evidence could sot be obtained iv the Lime available in the field. Apurt from the main line of fault, there has been considerable minor faull ing, evidenced by shatter zones along the base of the clilfs, C. Tur Nortu-Eastern PLATEAU From the physiographic point of view, the North-Eastern Plateay does not have the same interest as the other two zones in County Victoria, Also, although the platen extends for some considerable distance beyond the County bowndarics. the part within the Connty itself is small, being largely coufined to the Tundred of Whyte. For these reasons, only brief mention is made of the zone in the present study. These is no distinct boundary between the Central Tighlands and the North- Eastern Plateau. The typical “ridge-qnd-valley” structure of the former fs gradually replaced by a more irregular pattern, and the relative relief between ridges and valleys becomes much smaller. In the far north-east corner oF the County there are wide areas of comparatively level land, The altitude of the Plateat is for the must part between 1,700 and 2,100 feet above sea level, The zone is dominantly am area of internal drawage. There ure no very distinct drainage lines, and the streams which do exist flow only after rain. ‘This is largely hecause of the low rainfall, which is less than 16% for the 267 whole zone, and less than 14” in the north-east corner of the County, Near the township of Whyte-Yarcowie are two playa lakes which contain large quantities of water after rain In the summer months most of this water evaporates, leav- ing 4 dry salty surface, The soils in many parts of the Plateau zone have saline tendencies, and there are appreciable areas of halomorphic soil (Pirie series). ~The hills and ridges within the zone are more rocky than in the Central High- lands. They have little soil or vegetation cover and present a very barren appearance. TERTIARY GRAVELS There are a number of relics of Tertiary gravel in County Victoria. For the most part these outcrop as résiduals on low, rocky ridges within the broad north-south valleys of the Central Highlands. Howchin (1931-33) in his “dead-river” hypothesis claimed that these grayels represent the courses of old north-south rivers which flowed during the late Tertiary, and which were truncated by the Peterborongh-Olary upwarp. He considered that the presence of gravels throughout the valleys of the Mount Lofty Rauges inferred that a major nver system once flowed right through these parts to the southern coast. As far as the Mount Lofty Ranges near Adelaide are concerned, the argu- iments against this theory seem fairly conclusive (Fenner 1939, Sprigg 1945-46}. Ilowever, the present writer is inclmed to the view that althongh Hawchin’s hypothesis is far too sweeping in its claims, it may have elements of truth in the case of the northern areas in the neighbourhood of County Victorin. From study of the “ridge-and-valley” structure of the County Vicluria region, it would seem that there must have been sume form of north-south drainage prior to the Petérborough-Olary upwarp, This drainage may have been related to the present Lake Torrens area, but there is no evidence tu the effect that it was part of the great north-south river system of the Tertiary which was later ponded back to form a central Australian lake (David's “Lake Dieri’’), However, even a small local drainage system would have left zelics of stream action, and it is not unreasonable ta suppose that same of these relics would still be preserved. Some of the gravels trom the low residual ridges within the broad valleys of the Central Highlands are typical of present stream conglomerates (see pl. xxv, fig. 1), vxcept that they have been somewhat silicified. The pebbles contained in these specunens are well rounded atid have certainly been stream-worn, Gravels sue as these are quite distinct from other gravels which occur on sitmlar residual sites (in fact, sometimes on adjacent sites), and which contain much angular material. “The writer has followed outcrops of the old conglomezate for cou- siderable distances through the valleys near Jamestown, and it is significant that the outerops followed a line which would be the logical course of earlier longi- tudinal streams, [ct was also noted that the height of these residual outcrops above the present baseline in the valleys wicreases from east to west. Tor example, a typical gravel outerop in the Ifundred of Whyte is 30 feet above the present stream bed, A similar outerop in the Hundred of Belatie is 130 feet above the valley baseline. This is concurrent with the rainfall, which also imcreases. very markedly from east to west, The inference is that erosion during Pleistocene and Recent times has been proportional to the incidence of the rainfall, ‘ Fenner (1939), from his examination of ancient gravels in the Adelaide teginn, concludes that they consist of re-sorted fault-apron material and relics from streams which meandered over the pre-Miocene peneplane. He considers 268 that the deposits have largely “been formed by streams which cut back into the scarp faces and flowed at right angles to the supposed directions of the hypo- thetical ancient streams’ He therefore believes that jhe dead river theory “ts in the main an untenable interpretation af the facts.” However, in County Victoria, the location of the “conglomerate” type of gravels on tesidual sites in the centre of broad synclinal and eroded anticlinal valleys shows that they could not be resorted fault-apron material. Also, the definite longitttdinal pattern of the outcrops down the central parts of the valleys would indicate that they were formed im situ by streams which flowed imme- diately prior to the Peterborough-Olary upwarp. It is therefore contended that a modified version of Howchin’s hypothesis may be applicable ta the County Victoria region. It is considered that the more angular silicified gravels found on residual sites both within the broad yalleys ard in some cases on the adjoining valley slopes are relics of Tertiary laterisatiou, and that the silicified “conglomerate” type of gravel has also been subject to laterisation. Bryan (1939) and White- house (1940) have shown how laterisation can be responsible for the silicifica- tion of sediments. They believe that the “billy’’ deposits of Queensland were formed during wet, warm periods in the Pliocene, In view of the fact that laterites are formed largely in areas of impeded drainage, it would not be incongrtious to find pure laterites adjacent te “laterised™ ald stream gravels in sites which must have represented the Tertiary valley floor. Since most of the ridges iit Courity Victoria have a resistant core of silicified quartzite, it is unlikely that they have changed in form very noticeably since pre- uplift time. Therefore it should not be difficult to make an approximate recon- struction of the late Tertiary topography by mapping all the existing residuals of laterite or laterised sediments, particularly in the Central Highlands where there has been little or no major Tertiary or post-Tertiary faulting, TECTONIC MOVEMENTS. Matx TreeTtomic Associations County Victoria is an area transitional betwee the North Mount Lofty and the South Flinders Ranges. Largely as a result of this it exbibils structural features witich can be related to fotir signiheant tectunic niovements; the Peter- horough-CMary upwarp, the Flinders horst movements, the Mount Lofty horst movenients, and the periodic uplifts of Pleistocene tu Recent time. First, there is the high plateay country fo the central north and norrh-cast, which isa result at the Peterborough-Olary upwarp; secandly, the South Flinders Range in the Port Pirie area, which is part of the Flinders horst; thirdly, the mountaing and ridges of the Central Highlands which consmate the ceniral and southern, parts of the County, and which although related to the Peterborough-Olary upwarp are also associated with the Mount Tally horst movements; and finally, the coastal plains which exhibit evidence of nphtt from Pleistocene to Recent times. The Gulf area to the west of the County, which constitutes part of the Spencer-Vincent Sunkland, is a result of the Flinders and Mount Lofty horst movements. Ace of MovEMENtTS There has been considerable ¢peculation in regard to the nature aud age of these tectonic movements. The 1939 A.N.Z.A.A.S. committee an “Stmictival and Land Forms” agreed upon a tentative correlation for the major Tertiary and post-Tertiary movements in South Australia. This involyes a gentle uplift period 269 at the conclusion of the Miocene, accelerated uplift in the Pliocene followed by a period of stillstand, and then further differential movement in the Pleistovene continuing in special cases into Recent times. Since 1939 there has been a tendency to favour a more recent age for these inayements. It has been suggested (Brown 1945) that the period of accelerated uplift may have been the early or middle Pleistocene instead of the Phocene, Rrown quotes Fenmer’s yiew (1931) that the Peterborough-Olary upwarp pre- ‘ited the more violent Flinders and Mount Lofty horst movements, and remarks that the upwarp may have begun in the late Pliocene. Mawson (1942) has suggested that the Flinders and Mount Lofty horsts may be the result of two distinct movements, the former being the carlier. The Central Highlands of County Victoria, with the exception of the South Flinders Range near Port Pirie, are technically part of the North Mount Lofty Ranges, However, from the geomorphic point of view they do not appear to be closely associated with either the Flinders or Mount Lofty horsts, and are more directly related to the Feterhorough-Olary upwarp. The writer’s. stratigraphical investigations (1947) of the Central Highlands im the Jamestown area showed that there had not been any appreciable block faulting in this region, and there was no evidence of major post-Lertiary tectonic activity other than simple uplift. Since this area. merges into the Peterborough-Olary ridge, a large part of the uplift must have been due to the Olary upwarp movements. However, the area may have been further \plifted at the time of the Flinders and Mount Lofty horst movements, Evidence of Recent uplift in County Victoria, which was discussed in an earlicr section, is common to @ large part of the South Australian coastline. Crocker (1946) has discussed in some detail the evidence of uphlt in the South- Eastern districts and on Yorke Peninsula, as indicated by sand dunes and raised beaches. The following is submitted as a tentative correlation of tectonic movements in County Victoria from the Tertiary onwards, The table is based on the writer’s observations within the County, co-ordinated with recent research elsewhere in the Staie. Brown's suggestions regarding the age of the main movements have been adopted provisionally. Late Pltocene Gentle wupwarp of the Peterhorough-Olary ridge, accom- (or early panied by uplift of the adjoming peneplane area at present Pleistocene ?} represented in County Victoria by the Central Highlands (with the exception of the South Funders Range), inurly to (2) Major upift, producing the Flinders, horst, which included Middle the extremity of the South Plinders Range near Port Virie. Pleistocene Possibly further uplizt ef the Central Mighlands, (lL) Renewed major uplift, producing the Mount Lofty horst, and probably associated with this, renewed gentle uplift of the Central Tighlands. Middle Continued periods of mitor uplift, reflected by the sand-dune Pleistocene system on the Port Pirie Plains. Evidence trom the Port to Recent irie areca indicates that these movements are still in progress. “Youre Prawation” iHypornesrs Fenner (1931), following a suggestion by Douglas Johnson, drew attention ta the possibility of a “double planation” hypothesis to explain certain existing phenomena in Ue Mount Lofty Ranges, This hypothesis involves two periods 270 uf major uplift, each followed by a period of peneplanation, Fenner considered that there may have been a prolonged period of stillstand between the two uplift periods, during which much of the matare landscape of the Ranges (atich as the Upper Torrens valley) was produced. It was noted in the A.N.Z.A.A.S. Report of 1939 that supporting evidence for this theory had not been forthcoming, and subsequently certain workers, notably Sprigg (1945), tended to discount the hypothesis. Iowever, as a result of some recent work, Sprigg (1946) is inclined to reconsider the question. He notes that laterite occurs in many parts of the Mount Lofty Ranges and stresses the necessity of impeded drainage conditions as one of the pre-requisites of lateri- sation. Conditions such as these could have eccurred during a period of still» stand between two major uphifts. Further details concerning the age of the laterites in the Mount Lofty Ranges will not only help to clarify the age and nature of the uplift movements in that area, but by correlation with the laterites in the County Victoria region may throw iurther light on the inter-relation between the Mount Lofty horst movements, the Flinders horst movements, and the Olary upwarp. DRAINAGE SYSTEMS GENERAL The drainage divide represented by the Peterborough-Olary ndge is just outside the northern limits of County Victoria. This divide constitutes the houridary between exoreic and endoreic streams. North of the divide the whole of the drainage 1s endoreic. For instance, to ihe north-west, drainage is by way of the Willochra system to Lake Torrens; directly north is a region of local internal drainage, while to the north-cast the streams spread out and lose their identity ow the Walloway Plains, South of the divide, as the generul slope of the County is from north to south, all the main streams flow in a southerly circction, They are exoreie, for they finally join the Broughton River to enter the sea near Port Pirie. The accompanying map (fig. 1) illustrates the general drainage patteri of the area. “STREAM Caprure” THEORY As noted in a previous section, it watld appeat that the general structure of the country immediately prior to the Olary upwarp was very similar to that at present jn existence. It is probable at this early stage that the north-south drainage was much more extensively developed than at the present day, and that the upwarp truncated this system, building up the divide between the present north-llowing and south-flowing streams, However, it is apparent that the present streams are not completely governed by the north-south structural characterist.cs. At intervals they make abrupt right- angled bends to fow westward through gaps in the adjoining ridge. These gaps are often in the form of narrow gorges, the walls of which rise steeply on either side. The gradient of a stream in the gorge section is greatly increased, and rapids and even small waterfalls are common. The slream emerges from the gorge to jain another southward flowing drainage system. ‘The streams therefore cannot be regarded as having attained profiles of equilibrium, as the grades reached are separaied by short ungraded sections. This type of stream pattern is characceristic of practically the whole of the Maunt Lofty and Flinders Ranges and bas long been a topic for speculation. Dr, Charles Fenner, in a personal communication ta the writer in 1942, stressed 271 the point that the phenomenon is a gearral one, and therefore that the explanation for it must also be general and not restricted to any one locality. The writer believes that some previous theories have suffered from the weakness that the data used may not have been obtained from a sufficiently wide area. In the past, all the detailed physiopraphic work in connection with the subject has been restricted to the Mount Lofty Ranges in the Adelaide Region, and the writer feels that the tectonic relationships peculiar to this area have at times been allowed la Qutweigh other geomarphic factors. In the Adelaide area the Ranges have been extensively block-faulted, whereas in County Victoria (with the exception ef the Sowth Flinders Range) block-faulting 1s almost non-existent. Near Adelaide the north-south valleys are to a large degree the result of this faulting, whereas in County Victoria they are due to the presenee of synclines and eroded unticlines in a peneplane which has been gently uplifted. In County Victoria, therefore—and particularly in the Jamestown district where the writer completed a detailed stratigraphical survey—any hypothesis which ineludes block-faulting cannot apply. Theories which regard the east- west sections of the streams as antecedent, haying cut through fault-block edges fring a period of uplift, are theretoré untenable. AAs a result of field investigation near Jamestown (Langford-Simith, 1942) the writer came to the couclusion that (he explanation conld be found in streanr capture on an extensive scale. More recently Sprigg (1945) has expressed the same opinion in regard lo sireams it) the Mowit Lofty Ranges in the Adelaide area, and it is submitted by the writer that in view of the conclusions of these entirely separate investigations. the general theury of stream capture should reecive serious consideration. The Jamestown district affords typical examples of the stream pattern under consideration, Therefore, in view of the fact that the structural geology of the district is known in some detail, it js considered eminently suitable fer investiga- tion i regard to the “stream capture” theory. The accompanying block diagram (fig. 2) which is based on aerial photn- graphs, illustrates the main physiographic features of the district, It may be noted that the Belalie Creek flows ina southerly direction down the valley to thr west of the Browne Hill Range, before swinging abruptly to the west and enter- ing a gorge in a north-south ridge. The creek has a Jow gradient in the north- south valley, but this increases considerably in the gorge section where there ary tumerous rapids, and in one place (“The Cups") a stnall waterfall. After leav- ing the gorge, Gelalie Creek is joitted hy a small north-south creck at Jamestown, and shortly afterwards swings south again into the Baderloo Creck valley, where it enters another low gradient stage, The diagram shows how I'reshwater Creek drains the valley west of the Browne Ilil! Range below the point where Belalic Creck swings west into the Jamestown valley, The ridge containing the Belalie Creek gorge is a watershed four numerous small consequent streams, which flow to both east and west. All these streams show very active headward erosion and many of them have cut well back into the ridge. It is of great significance that headward erosion is especially active ou streams flowing to the west of the divide. ‘This is because the floor uf the western valley is lower than that ta the cast, and in general stream profiles ate steeper. One westward-fiowing consequent stream (marked A on the diagrain) has eroded its way back completely through the ridge, being assisted to same extent by the presence of a very small transverse fault. It is now commencing tu cut its way into the adjoining valley floor, and its headwater zone has already capiured some of the water from this valley. Before long, if headward ernsion continues, it will capture the headwaters of Belalie Creek, 272 In the light of the foregoing evidence it is probable that Belahie Creek once flowed right down the north-south valley and di not swing westwards through the gorge near Jamestown. At this stage, Belalie Creek and Freshwater Creek would have beet one continuous stream. Wigs, 2 Block diagram of draimage pallern wear Jamestown; stream capture at A. This is by no micans an isolated case, and throughout the Counly east-west streams may be found in all stages of headward erosion right up to the point af capturing a north-south stream. It is not known to what extent small transyerse faults have accelerated headward crosion in consequent streams by creating zoties of weakness, although in a few instances this is known to have been a contribut ing factor, as in the example cited, Brouguron River System Although most of its tributaries flow from or through County Victoria, the upper and middle reaches of the Broughton itself are south of the County houn- dary, and consequently could not be investigated thoroughly in the time available in the ficld. The lower reaches of the river are well within the County and were examined in rather more detail, Unlike most of its tributaries, the Broughton flows across the grain of the country in a general east-west direction. It has its source in the high country south of the Booborowie Flats. Flowing approximately westward to Spalding, 273 it is met by Deep Creek and Freshwater Creek flowing from the serth, and the Hill River from the south. A few miles south-west of Spalding, after being joined by the Hutt River flowing from the south, it dives through a gorge in a north-south ridge. Shortly after emerging from this gorge it is joined by the Bundaleer Creek from the north. After passing through Red Hill it is joined by Rocky River, and then flows on to the Pirie Mains where it is -met by the Crystal Brook. The course of the Broughton River over the Pirie Plains is through floud plains of its own making. An important feature of the upper parts of this sectivt is the deep channel through which the river is at present Howing, The height of the banks, which are very steep-sided, averages about 20. feet above the stream bed (see pl. xxv, fig. 2). This down-cutting is evidence of an endeayour by the stream to regrade its course after comparatively recent uplift. The course of the Broughton River neaver the coastline is not clearly defined, and here are a number of distributaries. Meandering channels are a feature of this part. Tle most prominent channel is known as Deep Creek, which contitiues to the tidal Port Davis Creek, A detailed description of most of the Broughton tributaries has already been) recorded (Langford-Smith, 1942), and it is not proposed tu discuss (hese at length in the present paper. However, certain features which are common to ther all have particular geomorphic significance. One of these is related to the stream anomalies which have already heen expounded in the disctission on “Stream Capture”. Because of these anomalies, each stream is subdivided inta series oi genily graded reaches separated by shurt sections of steep gradient, Therefore, the streams (with the exception of the lower Broughton) do not have profiles of equilibrium. At each of the east-west sections of steep gradient the streant is dowa-cutting rapidly in an attempt to propagate upstream the hreak in its profile, In some cases this has already resulted in down-cutting ih the north-south sec tions, and this will tend to continue until the whole stream system has attaimed| a profile of equilibrium, Douglas Johnson (1938) has shown that a stream profile can only be used as an indicator of change in level when the stream itself had a profile of equilibrium prior ta the uplift, In other words, the only stream int County Victoria which can be sed for this purpose is the lower Broughton. However, a significant feature of nearly all the tributaries is the presence of twa liver lerraces, one aboye the other, The higher terrace represents a ‘fossil flood- plain originating during an earlier period. It would appear that this floud-plain is inJicalive of an earlier period of heavier rainfall, for as already puinted out the streams are not in equilibrium, and the almost universal down-cutting through old alluvium cannot have been caused by isostatic change. The presence of old alluvium above the present flood-plains of streams has bev noted in many paris of castern Australia, and Whitehouse in particular has assumed thal it is indica- tive of higher rainfall periods in the Pleistocene. The time available for field examination of the middle and upper reaches of the Broughton River was limited, and it was not possible to consider in detail the question of its origin. However, a few significant paints are discussed briefly, The most prominent feature of the Broughton is that i fluws in a general east-west direction, which is apparently right across the grain of the country, This fact has led Fenner (1931) to suggest that it is am antecedent stream. Howcehin has proposed that it came into being after the Olary upwarp period, as a result uf extensive ponding in the luw areas to the south of County Victoria. However, it is important io note that over the course followed hy the Broughton, the “ridge-and-valley” structure js not as strongly developed as in areas either to the north or the south, and the river Rows through very few gorges in narth- 274 south ridges. If the river capture theory is substantiated im connection with stream anomalies in the Broughton tributaries, it wotld not be unreasonable to suggest that headward erosion of consequent streams may have been responsible for the few “break-throughs” in the present course of the mam stream. During the period of uplift there may have been a certain amount of local ponding which could have hastened the “breaking-through” process in the case of individual tidges. In the event of ponding, the degree of headward erosion required on the part of a consequent stream would have been reduced. ACKNOWLEDGMENTS The author wishes to acknowledge his mdebtednegs to Professor J. A. Prescott who provided the opportunity for carrying out the work. Dr, Charles Femier and Sir Douglas Mawson proffered helpful advice and criticism during the course of the investigations, and the encouragement of Mr, C. G. Stephens of the Soils Division, C.S.1.R., was much appreciated. Acknowledgments are also due to Mr. P. D, Looper of the Soils Division, S.LR., and Mr. 11. E. Smith of the Regional Planning Division, Department of ost-War Resanstruction, for their assistance with the drafting of the maps. REBPERENCES Browne, W. R. 1945 “An Attempted Post-Tertiary Chronology for Australia” (Presidential Address). Proc. Linn. Soc, N.S.W., 70, (1-2), v-xxv Bryan, W.H, 1939 “The Red Earth Residuals and their Significance in South- Tvastern Queensland.” Proc, Roy, Soc. Old., 50, 23-32 “COMMITTEE on the Structural and Land V’orms of Australia and New Zealand,” 1939. Report of A.N.Z.A.A.S,, 24, 390-408 Crocker, R, L, 1946 “Some Raised Beaches of the Lower South-East of South Australia and their Significance.” ‘Trans, Roy. Soc. S, Aust. 70, (1), 64-82 Crocker, R. L. 1946 “Notes on a Recent Raised Beach at Point Brown, Yorke Peninsula, South Australia.” “Frans. Rov. Soc. S, Aust., 70, (1), 108-109 Fenner, C. 3931 “South Australia: A Geographical Study.” Whitcombe and Tombs Ltd., Melbourne Fenner, C. 1939. “The Significance of the Topography of Anstey Hill, South Australia.” Trans. Roy. Soc. S. Atist., 63, (1), 79-87 Howcnin, W. 1931-1953 “The Deal Rivers of South Australia.” Trans. Roy. Soc, S. Aust., 55, and 57 Howenrn, W. 1929 “The Geology of South Australia.” Adelaide Jounson, D. W. 1919 “Shore Processes and Shoreline Development.” John Wiley & Sons Inc., New York Jounson, D. W, 1932 “Rock Planes of Arid Regions.” Geos. Rev., 22 Jounson, D. W, 1938 “Stream Profiles as Evidence of Enstatie Changes of Sea Level.” Jour. Geomorphology, 1 LANGFORD-SMITH, T. 1942 Thesis for the Degree of M.Sc. Adelaide Uni- versity Lancrorp-Smiru, T. 1947 “The Geology of ithe Jamestown District, South Australia.” Trans. Roy. Soc. S. Aust., 71, (2), 281-295 Maxtin, F. ©, 1939 “Geographical Observations upotr Port Pirie and Dis- trict." Proc, Roy. Geog, Sov. S. Aust, 40, 43-68 C, p 275 Martin, F.C. 1942 Personally communicated Mawson, D. 1942 “The Structural Character of the Flinders Ranges.” Trans. Roy. Soc. 5. Aust., 66, (2), 262-272 Spricc, R. C. 1945 “Some Aspects of the Geomorphology of Portion of the Mount Lofty Ranges.” Trans. Roy. Soc. S. Aust., 69, (2), 277-302 Spricc, R. C. 1946 “Reconnaissance Geological Survey of Portion of the Western Escarpment of the Mount Lofty Ranges.” Trans. Roy. Soc. S. Aust., 70, (2), 313-347 Whirenouse, F. W. 1940 “Studies in the Tate Geological History of Queens- land.” Univ. of Queensland Paper ‘anqonays “ULMOysaTURf IPou = i ‘> vi VA[[PA-puv-ISpLt BuyRsisnye ‘asuryy ppeqdwuey ayy ASpPLe yNos-1pjiou TeodAy e Suoje AuLpooT S SpleMO] OISplLt otueS 9} UOI, JsoM SULyOOo'] ¢ Buy Es b Sty aj = ~ “ ‘punomyoeq ayy ut darag “davog siopuryy yNog sy SPsBMay SAAPUIP YA WNOS ayy LM ‘yuautpag Aqiadden oy, SILA [PLL oy Worz yseo Sugpool ‘suw[_q lq Puy, cSt 1 31y Soc. S, Aust. 1048 Roy. ‘Trans. “I bo Tras. Roy. Soc. S. Aust, 1948 Vol. 72, Plate XXV Fig. 1 Gravels from a residual site. Note rounded, stream-worn pebbles. Kip, 2 Broughton River on the Pirie Plains south of Crystal Brook, showing deeply incised channel. PURPLE SLATES OF THE ADELAIDE SYSTEM By D. MAWSON AND E. R. SEGNIT Summary The nature and genesis of the shales and slates of chocolate to purple colour occurring at several distinct horizons in the post-glacial Upper-Proterozoic strata of the Adelaide System is a matter of interest. Certain features have suggested that in part they may be of tuffaceous origin. There are several factors indicating this probability, such as the unleached nature of the shale, a considerable proportion of the coarser particles being recognisable feldspar. The alkali and lime content of the rock is high. Further, we have observed that these shales in some areas at least have a notable content of barium and even copper; veins of barytes are a common associate and the occurrence of blue and green copper stains between the laminae of the shale has been observed in a number of localities. Also, among embedded rock fragments at one horizon in a belt of purplish-brown shale, fragments of a soda-rich basaltic rock were found to be common. 276 PURPLE SLATES OF THE ADELAIDE SYSTEM By D, Mawson and E. R. Secnrt * {Read 11 November 1948} The nature and genesis of the shales and slates of chocolate to purple colour occurring at several distinct horizons in the post-glacial Upper-Proterozoic strata af the Adelaide System is a matler of interest. Certain features have suggested ihat in part they may be of tuffaceous origin, There are several factors indicat- ing this probability, such as the unleached nature of the shale, a considerable proportion of the coarser patticles being recognisable feldspar. The alkali and lime content of the rock is high, Further, we have observed that these shales in some areas at least have a notable content of barium and even copper; veins ot barytes are a common associate and the occurrence af blue and green copper stains between the laminae of the shale has been observed in a number of localities, Also, among embedded rock fragments at one horizon in a belt of putplish-brawn shale, fragments of a soda-rich basaltic rock were found to be common, However, elsewhere, throughout 4 considerable thickness of strata, these shales are very fine and uniform in grain anc. where they grade into arenites these latter are domifantly composed of quartz particles, but with a notable contribution of feldspar. In order to probe further the origin of these shales, material collected close to the old Motint Deception Station homestead on the eastern slopes of Mount Deception, 11 miles north-west of Peltana, has been analysed hy one of us and atherwisc more fully examined, with the results detailed here- with, The horizon in the Adelaide System from which these specimens now sub- jected to detailed examination were oblaimed is the thick purplish-brown belt below the Cambrian Pound Quartzite, referred ta by Howchin as the Purple Slates. At one place in the neighbourhood of the spot where the specimens were obtained malachite stains were observed in the bediling planes of the shales. Of the two specimens examined, one of very fine grain represents the more normal type of this rack, while the other represents a coarser than usual band of the shale. The finer grained variety [8126], whew examined in the hand specimen, is observed to be of a dark vinaceous red colour (Ridgway) and a faintly dis- tinguishable fine sedimentary banding. It is soft, remarkably even-grained and breaks with a smooth sub-conchoidal fracture. The microscope slide reveals. that it is principally composed of grains of suarta, altered particles of feldspar and minute flecks of mica, The quartz grains are clear and frequently very angular; they tend to be concentrated in the lighter bands of the shale. The mica is clear and colourless in tiny flakes so thin as to give only red and yellow interference colours, The feldspars have evidently been largely decomposed to a featureless mass, possibly mainly composed of the kaolinite group of minerals, It is this matrix that is stained purplish-brown by the ferric oxide content of the shale, thus imparting that colour to the reck as a whole. Calcite is distinguishable, distributed throughout in irregular grains. It is stifficiently abundant to link up with the 5% of carbon-dioxide found to be contained in the rock. Also, there is present a small amount of chlorite, haematite, and a highly refracting mineral, possibly zircon, * Geology Department, University of Adelaide. Trims, Roy, Soc, 5. Aust., 72, (2), 30 March 1949 277 The second [8127], coarser-grained specimen is a harder, darker purple rock In this the mineral assemblage is similar to the preceding but the particles are larger and more determinable. There is much quartz of larger grain-size and frequently very angular; also fresh feldspar in grains of similar size and shape. These mineral grains are cemented in an abundant matrix of dark-red iron oxide with which is an assdciation of mica fragments and serpentinous material, The more obvious feldspar is twinned plagioclase, the majority of which is charac- terised by a low R.J, and an extinction angle on albite twins of 12°-14°, indicating a rather sodic variety. Microcline and orthoclase are also present, ‘The un- twinned feldspar is only with difficulty distinguished from the quartz when all ig embedded in the dark matrix. Though very fine-grained this rock probabls comes within the silt grade. An analysis of the fther-grained variety is svated herewith, and included 7m the table are analyses of other releyant shales and of a red micro-granite, Publications dealing with the origin of red rocks are camiparatively few. 1: is generally agreed, however, that the red colour is due to the présence of ferric oxide, either hydrated as sume form of limonite or anhydrous as hematite (Dorsey 1926, and Raymond 1942). Thus arises the question of the derivation af the ferric oxide. Raymond (1927) deals at length with this matter, stressing the widespread development of highly ferruginous laterites under warm. moist conditions as an abundant source of Terric oxide. Te also discusses the develup- ment of ferric oxide in atid regions, as is evidenced by the red colour of sands and rocks in such areas, Recently Lord Rayleigh (1946) hay investigated the cause of the red colour of certain sandstones and finds that if chalybeate water is evaporated on the surface of a solid body, eg., a sand grain, a red coating results. The same rei colouration develops when evaporation takes place in air from an undisturhed liquid surface. If on the other hand evaporation of chalybeate waters tales place in an agitated liquid, as in a stream in the open, the iron separates as a light yellow precipitate, Tt is thus obvious that sand particles moistened by waters containing some ferrous bicarbonate in solution and dried out will be reddened, rom this i would appear also that argillaceous deposits in chalybeata waters evaporating in shallow playa lakes would be reddened. So red shales and sandstones can be accumulated either dm sity as terrestrial deposits or can be derived from the transport of such peroxidised materials and their sabsequent deposition under non-reducing conditions, This also waultd normally imply a terrestrial origin, but in special cases, where organic life is much subordinated or non-existent, deposition might conceivably take place as shallow-water marine deposits adjacent to large land masses. In our case the red-brown colour is given to the shale by the presence of the remarkably small amount of 1°94% of ferric oxide, which amount is unusually: low for a shale and particulatly so for red varieties (sce table of analyses), It is interesting to note that certain bright-red to chocolate-coloured igneous rocks may contain extremely little ferric oxide, as for example the porphyritic micro-granite [5778] (Mawson and Segnit 1946), containing only 0-929. Further, it is possible for a green slate to contain the high proportion of 3-48% ferric oxide (see table of analyses), In this latter case jt must be present in 2 combination with other constituents, 278 I If Il lV Vv VI Vil SiO, - = = = $830 38-38 60-95 © 56-49 864 38-98 — 72-50 TO, + SF oe } fra 0-65 O86 O48 0-56 2-05 4-08 ALO, - = w=} WS-A¥ Wel5 11-59 11-350 28-00 14-02 Fe,O, es 5 + 194 4-03 5-160 3-48 3-07) 14589 0-92 FeO - - - = 2-80) 2546 254 1420 73 GOR 1-24 MgO - - - - 387 2-45 3220 G43 2400230023 CaQ = = = = 4-31 9-12 0-71 5:1] 3:50 15 0°77 Na,O 30S -& 1-81 1-31. 1-500 08320 ASL O08 347 ROe = = = 316 3-28 3-01 377) 2:10 O18 5 +24 H,O+ a 3-96 sens 5-08 2-82 i 10-38 = O73 H,O— co President, 1942-43. 1942, *Coormx, IT, M., SL ‘Hustings Street, Glenelg, S.A. O44, Consist, Mevitee, State Bark, Pirie Street, Adctaide, 1929, *Corron, B. GC, S.A. Museum, Adelaide—Couneil, (943-40, 194e-, 4924, ne Crespromy, Sir CT, 0, 5,0, M.D, PARA, 219 North Terrace, Adelaide. 1937. *Crockem, RL. D.Sc, Waite Inevitute (Privute Mail Bag), Adcluide—Seeretary, 043-45: Connei, 145-47. 4 “Drenivsox, 8, 7, Mise, Govt. Geologist, Mines. Beparument, Flinders Sireet, Adelaide —Couneil, 1946-. 1930. Dix, E. V., Hospitals Department, Rundle Street, Adelaide, S.A. 1944. LGwstong, 5. M. L, MLB, B.S., 124 Payneham Road, St. Peters, Adetaide. 1931. Dwyer, J. M. MB, B.S. 105 Port Road, Tindmarsh, S.A. 1933. Beare Miss C. M., B.Se., Waite Institute (Priyate Mail Bag), Adelaide—Couneil, 943-46, 1945, *EpMonns, S. J., B.A, M.Sc, 56 Fisher Terrace, Mile End, S.A, 1902, *Engursr, A. G., 19 Farrell Street, Glenelg, S.A. 1044, Ferres, Miss H, M., M.Se., 8 Taylor’s. Road, Mitcham, 5.A. 1927, *Winuayson, H. H., 305 Ward Street, North Adelaide—Conncil, 1037-40, 1023. *Fry, H, K, D.S.0, M.D. B.S, B.Sc, F.RACLP., Tawat Hall, Adelaide—Couneil, 1933-37 ; PGce-President, 1937-38, 1939-40; President, 1938-39, 1932, *Grason, E. S. H., B,Sc., 207 Cross Roads, Clarence Garriens, Adelaiue, 1935. *Giastoxaury, J, O. G, BA, M.Sc, Dip.ed, Beatty Lerrace, Murray Bridge, S.A, 1927. Gonrreny, F, K,, Box 951H, G.P.O., Adelaide, 1935. {Gotosack, H., Coromandel Valley, 5.A. 1939, Goom, J. R., B.Agr.Se., P.O, Box 180, Whyalla, S.A. 1925. jGosse, Sry James H,, Gilbert House, Gilbert Place, Adclaide, 1410, *#Grant, Prov. Sr Keene, M.Se., F.1.P., 56 Kourth Avenue, St. Peters, SA 1930. Gray, J. T., Orroroo, S.A. 3933. Greaves, 1T., 12 Edward Street, Gtynde, §.A, 1904. Grorrn, H. B., Dynrobit: Read, Brighton, S.A. 1448. Gross, G. F., South Australiun Museum, Adelaide. 1944. Guery, D. j., B.Sc, Minerai Resources Survey, Canberra, A,C.T. 1922. *Ha.e, H. M, Director; S.A. Museum, Adelaide—Verca Medal, 1946} Cowieil, 1931-34} Vice-President, 1934-36, 1937-38; President, 1936-37; Treasurer, 1938- 149. Want, D, R.. Mern Merna, via Quorn, S.A, 1945, *Harpy, Mrs. J, E. (nee A. C. Beckwith), M,Se., Box 62, Smithton, Tas, 1944, Haners, J, BR, B.Sc,, 94 Arctier Street, North Adelaide, S.A. i947. Hennerson, D. L. W., Craigsiane, Woodford, N.S.W- 1944, Hergror, R, I., B.Agr.Sc., Soil Conservator, Dept. of Agriculture, §.A. 1949. Hotvoway, B. W., B.Sc,, 33 Kyre Avenue, Kingswood, S.A. 1924. *Hossrrzn, P.'S., MSc. 132 Fisher Street, Fullarton, SA. i944. Humuore, D. S$. W., 238 Payneham Road, Payneham, 5.A. 1947. Hurton, J T.. B-Se, 13 Sherbourne Road, Medindie Gardens, S.A, 1928. Irourn, P., Kurralta, Burnside, S.A, 1942, Jenxins, C. F, H., Department of Agriculture, St, George’s Terrace, Perth, WA. 1918. *Jenntson, Rev. J. ©, 7 Frew Strect. Hullarion, Adelaide. 1945. *Jussur, R. W., BSc, 3 Alma Road, Fullarton, SiA. 1910. *Jounsow, E, Ay M.D. M.R-CS., “farni Warra,” Port Noarlunga, SA, 1921, *Jounsron, Pror. T. Hy. M,A,, D.Sc... University of Adelaide—lVerco Mert, 1935; Council, 1926-28, 1940-) Vice-President, 1928-31; President, 1931-32) Secretary, 1938-40; Rep. Kaun and Flava Board, 1932-39; Editor, 1943-45, 1939, }KHAR#HAR, H. M., Ph.D, M.B,, F.R.GS,, Khakar Buildings, CP. Tanke Road, Bom- bay, India, . 1933, KiReMAN, A. W., M.Se., University of Adelaide—Secretary, 1945-485 VicePresi- dent, 1948-, ; 1927, Lenpon, G. A, M.D.. BS., F.R.C.P., A.M-P. Building, King Willian Street, Adelaide, 1948. Laretan, T. R.N., N.D-H. (N,Z,), Director, Botanic Gardens, Adclaide, 1949. Lower, H, I. 7 Avenue Road, Highgate, SA. 1931. *Lupremic, Mas. W. Ve (nee N, H. Woods), M.A,, Flimatta Street, Reid, ACCT. ’ Thate a! 384 Eleedan, 1048. 1938. 1932. 1939, 1929, 1905, 1920, 1943, yg, + 139, 1425. 1933. 1938, 1940, 1936, 10-4, 1944, 1944. 1945, 1930. 1947, 1913, 4937. 1945, 1029, TW2h. 1948. 1847, 1925, 1926. 1945, 1944. 17, 1948, 1947, 1946, 1945. 19444, 1943. 1946. 1924. 1925, 1936. 1945, 1934, 1924. 141, 1941, 1936. 1947, 1936, 1947. McCuuiscn, R. N., MBE, BSc. (Oxon), TAgt.Sei. (Syd), Raseworlhy Agricul tural College, S.A. Maprern, C. B., B.DS., N.N-Sc., Sheil House, North Terrace, Artelaite Mann, EF. A., C/o Batik of Adelaide, Adelaide. Marsuaty, T. J. M-Agr.Se, Ph,, Waite Institute (Private Mail Bag), Afetside-- Council, 1LO48-. Manrin, F. C., M.A., Techmeal High School, Thebarton, S.A, *MAwson, Prov. Siz ‘Daucus, O.B.E., D.Sc. BE, F. RS., University of Adelaide— Merca ” Medal, 1931; President, 1024- 25, 1944-45; Fice -President, 1923-24, 1925-26; Couneil, 1941-43. MAYo, THE Hon, Mr, Justice, LL.B, K.C., Supreme Court, Adelaide. McGartny, Miss D. F., B.A., B.Sc,, 70 Halton Terrace, Kensington Park, ‘Mines, K. R. D.Sc. FIGS, Mines Depattment, Finders Street, Adelaide. Mincuam, V. H., Hammond, §.A, ; 7Miteyerr, Prop. Siz W.. K.CMG., M.A., D.Se., Fitzroy Ter. Prospect, SA- Mirenert, Pror. M. L., M.Sc, University, Adelaide. Moornouse, F. W., M.Sc. Chief Inspector of Fisheries, Flinders Street, Adelaide. Morrioce, J. A. T., 37 Currie Street, Adelaide. *Mountrorn, C. P., 25 First Avenue, St, Peters, Adelaide, Munrerr, J. W., Engineer‘ng and Water Supply Dept. Port Road, Thebarton, 5A. Neat-Sauira, C. A. B,Agr. Sci, 16 Gooreen Street. Reid, Canberra, ACT, Ninses, A. R., B.A., 62 Sheftield Street, Malvern, S.A. *NorrHoore, K. FL, BAge Sc, ATLAS. 6 Charles Street, Porestville 5.4. Ockenven, a. P., c/o Finders Street Practising School, Finders Street, Adelaide. *0vaer, L. b.. 65 fait Avenne, St. Peters, S.A, +Qsnorn, Prov. ‘T. G. B. DSc. Department of Botany, Oxford, Magland—Cuwned, 1915-20, 1922 OA; Presidetit, 1925-265 Wico-E'vesidest, 1924-25, 1926-27, *Parkiv, L, W.. R8c., c/a Mines Department, Flinders Street. Adelaide. Patrrson, G., 68 Partridge Street, Glenelg, S.A. Patt, oat G.. MLA. Se. 10 Milton Avenue, Fullarten, $.A. *Piprr, C. D.Sc., Waite Tnstitute (Private Mail Bar), Adclaide—Couneil, 1941-435 Vie Sine siden, 1943-45, 1945-47 » Presitont, 1945-46. Pownie, J. K., B.Se., CSIRO., Davision af Biochemistry. University, Adelaide. PoynToON, J). 0, M. iD., M.A,, ChR.. MRCS. LR.CP., University of Adélaide. *Prescory, Prom J, Aw ©. BE, D.Sc, ALC, Waite Institute (Private Mail Bag), Adelaido—lorea Medal, 1938: Council, 1927- 30, 1935-30; Fice-Presitent, 1930-32: President, 1932-33, Price, A. G., C.M.G:, M.A.. Litt.D., F-R.G.S., 46 Pennington'Terrace, North Adelaide, Pryor, L, D., M.Se., Dip.For,, 32 La Perouse Street, Griffith, N.S.W. Rremman, 0. §., B.AgrSe., CS.LR.O,, Divisinn of Nutrition, Adelaide. toro, W, R., B.Se, St. Mark's Colleze, Pennington Terrace, North Adelaide. Rimes, G. D., 24 Wiinstan Avenue, Clarence Gardens, S.A, Rix, C. EB. 42 Waymouth Aveuve, Glandore, SA. Rovrysos, E. G., BSe., 42 Riverside Drive, Sudbury. Ontario, Canada. Rrmii., J. R., Ohl Penola Estate, Penola, S.A *Sanpars, Miss D, F, MSc,, University of Queensland, Brishane, Qheens and Scnneiper, M., M.B., B.S. 175 North Ter., Adelaide: aSrenit, ©. RR. M.Se,, C.S.LR.0., Division of Sidusirial Chemistry, Box 4351, G.P.O0 oles ier te Victoria. *Srenir, R. M.A, B.S. Engineering end Water Supply Depertment, Victoriz Square, Atclide. Seren, 1930-35: Cremeil, 1947-38; Fice-President, 1938-39, 1940-415 President, 1939-40. *SHEARD, H, Pert Elliot; S.A. *Surann, K,, Fisheries Resvareh Mhiy. C.S.1.RO 178200 Chihuahua Desert, Phyrcogranhy of .. f 20 Clark, Helen G. T. H. Fohnatecr anu ; Cestodes from Australian Birds, T Pelicans -. 77-82 Cuncotheca furgide. bh 255 Cotton, B, C.; South Australian Gastropoda, part [Th .. -- 3032 Crespin, 1; Indo-Pacific Influetices in Australian Tertiary Foura- minifera! Assemblages .. - 193-142 Deserts, Phytogeography of Sand _ ridge-; C. M. Eardley 1-29 Dolcrites from the Musgrave atid Tiverard Ranges; A. I, Wilson 178-200 Iard'ey, C, ML; Phytogeography of sotne Important Sandridge Deserts cotnpared with that of the Simpson Desert .. * 1-29 Feology of the Western Clare Tills; C, D. Boomsina » 26-220 | | | Edmonds, §. J.; The Commoner Species of Animals and. theit Distribution on an Intertidal Platform at Pennington Hay, Kangaroo Island -. 167-177 T. Ho Johnston and: Aus tralian Acanthocephala, ‘No. 7 60476 Eucalypts, Nomenclature of, C. D. Boomsma . . 221-227 Eucalyptus amacrorryncha, “age hia occurrence of 219 Foraminifera: Australian Tertiary Foraiminiferal Assemblages ; T. Crespin . 143-142 Fossil Friits; P, S. Hossfeld . 252-258 Geomurphology of County Vicluria; T. Langford-Simith 2. 259-275 Granitization; A. W. Whittle 242 Hossfeld, P. S.: Significance of the Occurrence of Fossil Fruits in the Barossa Senkungsfeld 4, Stratigraphy of the : Aitebe Skull t . 201-207 TIymenolepis sagan tibertsis ra 77 H. jaenscht 79, A. vilist 81 252-258 Indo-Pacific Influences in Austra- lian Tertiary Foraminiferal Assemblages; 1. Crespin 134-142 Jacob, Mt., Geology of; D. Mawson 245 Jessup, R. W.; A Vewetation and Pasture Survey of Counties Eyre, Burra and Kimberley .. 33-68 Johnston, T. H. and S. J. Edmonds; Australian Acanthocepha'a, No. 7 ot oe Yr «O76 and Helen G. Clark; Ces fodes from Australian Birds, I Pelicans ie 77-82 Kara Kum Desert Pliytogeouraphy of .- on 15 Langford-Smith, T.; Geomarpho- logy of County Victoria 259-275 Libyan Desert, Phytogcography of 1 Loveridge, Arthur; On Some Rep- tiles and Amphibians from the Northern Territory .. 208-215 Marine Algae of Kangaroo Island, II The Pennington Bay Region ; H. B. S. Womersley : Marine Animals, their Distribution on an Intertidal Platform at Pennington Bay, Kangarow Island; S, J. Edmonds Mawson, D.; Sturtian Tillite of North Flinders Range Mawson, D. and E. R. Segnit; Purple Slates of the Adelaide System : te Penteuna trachyclints Perry, R. A. R. L. Specht ad: Plant Ecology of Part of the Mount Lofty Ranges, I Phymatocaryou Mackayi Phytogeography of Sandridge Deserts; C. M. Eardley F Plant Fcology of Part of the Mount Lofty Ranges, 1; R. L, Specht and R. A. Perry Pleiockinis Couchmuanti Polymerphus bisturac an . Purple Slates of the Adelaide System; D. Mawson and E.R. Seenit f i i Reptiles and Amphibians from the Northern Tereifery: A. Lovye- ridge : =“ ' Rhytidotheca Lyne hit ‘ oe Rub’ al Khali, Phytogeos raphy of Segnit, I. 1. D. Mawson and; Purp-e S‘ates of the Adelaide System : Simpson Desert, Phi toxcography vf; C. M. fardley . 143-166 . 167-177 . 244-251 . 276-280 . 208-215 257 8 . 276-280 Southern Australian Gastropoda, part III; B. C. Cotton Specht, R. L. and R. A, Perry; Plant Ecotogy of Part of the Mount Lolty Ranges, T a Spondylostrobus Smythit Sturtian Tillite, North Flinders Range oe : oe Takla Makan Desert, Phytogen graphy of Thar Desert, Phytiizeograptly of. Tragardhula Berlese 1912 (Actin, Trombiculidae) ; H. Womerstey Vegetation and Pasture Survey of Counties Eyre, Burra and Kim- berley; R. W. Jessup .. Warren Hastings, Mt, Geology af Western Sahara, Phytogeography of Whittle, A.; Gedtbay ‘ot the Bool- coomata, Granite P Wilson, A. F.; The Charnockitic and Associated Rocks of North- Western Sonth Australia; II, Dolerites from the Mus- grave and Everard Ranges Womersley, H.; The Genus Tragardhula Berlese 1912 (Acarina, Trambiculidac) Wamersley, H. B. S.; Marine Algae of Katgarou Isiand; II, The Peminaton Bay Re- sion = a Zoatle rosscilt 33408 247 9 , 228-243 . 178-200) 83-90 143-160 3u