— a eel ‘ ¢ Mt £ y - ° te : i a tro, ~~ : ” " ee ow : Py Y “ VOL. 71 PART 1. | JULY 25, 1947 TRANSACTIONS OF THE ROYAL SOCIETY OF SOUTH AUSTRALIA INCORPORATED ~ ee ee. eee a TT a LL lee ee ADELAIDE PUBLISHED AND SOLD AT THE SOCIETY'S ROOMS — KINTORE AVENUE, ADELAIDE Price - - Fifteen Shillings Registered at the General Post Office, Adelaide, for transmission by post as a periodical VOL. 71 — 1947 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 CONTENTS Womerstey, H., and Konts, G. M.: New Genera and Species of Trombiculidae from the Pacific Islands Me ts an et wh ¥ A oe Jounston, T. H., and Epmonns, S. J.: Australian Acanthocephala No. 5 Brack, J. M.: Additions to the Flora of South Australia, No. 44 .. Jounston, T. H., and Mawson, P. M.: Some Nematodes from Australian Lizards .. Turner, A. J.: A Revis:‘on of the Australian Phycitidae, Part I .. Mites, K. R.: Pre-Cambrian Granites and Granitisation, with special reference to Western Australia and South Australia Prescott, J. A., and Lane-Poote, C. E.: The Climatology of the introduction of Pines of the Mediterranean Environment to Australia .. Crocker, R. L., and Woop, J. G.: Some Historical Influences on the Development of the South Australian Vegetation Communities and their bearing on Concepts and Classification in Ecology Bursinpce, Nancy T.: Key to the South Australian Species of Eucalyptus L’Herit Strepnens, C. G,: Functional Synthesis in Pedogenesis Fintayson, H. H.: On the Weights of some Australian Mammals Wirson, Ariran F.: The Charnockitic and Associated Rocks of North-Western South Australia. Part IJ, The Musgrave Ranges—an Introductory Account .. Spricc, Rec. C.: Early Cambrian (?) Jellyfishes from the Flinders Ranges, South Australia Evans, J. W.: Some New Eurymelids from Australia and New Guinea (Homoptera, Jassoidea) Womerstey, H, B. S.: The Marine Algae of Kangaroo Island. J. A General Account of the Algal Ecology Keruinc, L. C. P.: Pythium debaryanum and Related Species in South Australia Mawson, D.: The Adelaide Series as Developed along the Western Margin of the Fiinders Ranges Lancrorp-SmitH, T.: The Geo'togy of the Jamestown District, South Australia .. Sprica, Rre. C.: Submarine Canyons of the New Guinea and South Australian Coasts .. Womerstey, H. B. S.: The Marine Algae of Kangaroo Istand. I. A General Account Western Australia Ornen, I. L.: Notes on the Genera Lychnothamnus and Lamprothamnium (Characeac) Jounston, T. Harvey, and Beckwitu, Anne C.: Larval Trematodes from Australian Freshwater Molluses. Part XII *3 aA a ss x3 na ae Page 67 324 CECIL THOMAS MADIGAN, M.A., B.E., D.Se., PGS, TRANSACTIONS OF THE ROYAL SOCIETY OF SOUTH AUSTRALIA INCORPORATED OBITUARY NOTICE CECIL THOMAS MADIGAN Cecil Thomas Madigan, M.A., B.E., D.Sc., F.G.S., a leading Australian geologist and geographer, a member of this Society for the past 25 years, and a Past President, died at the early age of 57 years on 14 January 1947, Curtailment of his span of life was doubtless the result of heart strain result- ing from exceptional exertions in the field of sport, exploration and war. His earlier education was received at the Adelaide High School, Prince Alfred College and the University of Adelaide, where he graduated in Mining Engineering in 1910. As a Rhodes Scholar from South Australia, he then pro- ceeded to Magdalen College, Oxford, to continue with geological studies, which were, however, to be postponed, for late in 1911 he joined the Scientific Staff of the Australasian Antarctic Expedition of 1911-14. Dr. Madigan’s record during more than two years in Antarctica was out- standing. It includes leadership of a sledging party which reconnoitred the ice plateau in winter time under record adverse climatic conditions. In the summer of 1912-13 he successfully led a sledge journey, charting a great length of new coast line of the territory thereafter known as King George Land. On return to winter quarters, he was selected to be in charge of all land operations pending the return of the Expedition Leader, who at that time had failed to return from a journey across the plateau ice. His graphic account of the summer sledge journey forms a section of the popular story of the Expedition, published as “The Home of the Blizzard.” He was chiefly responsible for the meteorological record of the Cape Denison Station, which is published as an important section of the Expedition’s Scientific Reports issued from the Government Printing Office, Sydney. On the advent of war when again in England in 1914, he joined the Royal Engineers and served throughout the whole period with rank of Captain. Later, having finally graduated at Oxford, he was appointed Assistant Government Geologist in the Soudan, This post he relinquished in 1922 to join the staff of the University of Adelaide. With the outbreak of the Second World War, Dr. Madigan was appointed chief instructor in the School of Military Field Engineering (Liverpool, N.S.W.), attaining the rank of Lieut.-Colonel. During his long term as lecturer in Geology, he found time to execute a number of important geological investigations dealing with problems relating to South Australia and to Central Australia, It is in the latter field that his work has received most recognition. His explorations in the MacDonnell Ranges and neighbourhood, and in the Simpson Desert, have established him as the leading authority on the Geography and Geology of Central Australia, For this work he was the recipient of wide recognition in scientific circles both in Australia and abroad. He is also well remembered for his zeal in furthering the good work of the Legacy Club and the University Graduates’ Association. His published scientific works additional to those dealing with his Antarctic activities mentioned above are the following :— 2 “A Description of some Old Towers in the Red Sea Province, North of Port Soudan.” Soudan Notes and Records, 5 (1922), 78-82. “The Geology of the Fleurieu Peninsula, Part I — The Coast from Sellick’s Hill to Victor Harbour.” Proc, Roy. Soc. S. Aust., 49 (1925), 198-212. “Organic Remains from below the Archaeocyathinae Limestones at Myponga Jetty, South Australia.” Trans. Roy. Soc. S. Aust., 50 (1926), 32-33. “The Geology of the Willunga Scarp.” Proc. Roy. Soc. 5. Aust., 51 (1927), 398-409. A.N.Z.A.A.S. Reports of the Glacial Research Committee (with D. Mawson), 19 (1928), 97-99. “Preliminary Notes on New Evidence as to the Age of Formations on the North Coast of Kangaroo Island.” Trans. Roy. Soc. S. Aust., 52 (1928), 213-214. “An Aerial Reconnaissance into the South-Eastern Portion of Central Australia.” Proc. Roy. Geog. Soc. (5S. Aust. Branch), 30 (1929), 83-108. ‘Lake Eyre, South Australia.” Geog. Jour., 76 (1930), 215-240. “Pre-Ordovician Rocks of the MacDonnell Ranges, Central Australia” (with D. Mawson). Qld. Jour. Geog. Soc., 86 (1930), 415-428. “Geology of the Western MacDonnell Ranges, Central Australia.” Qld. Jour. Geog. Soc., 88 (1932), 672-711. “The Geology of the Eastern MacDonnell Ranges, Central Australia.” Trans, Roy. Soc. S. Aust., 56 (1932), 71-117. “The Physiography of the Western MacDonnell Ranges, Central Australia.” Geog. Jour., 78 (1932), 417-433, A.N.Z.A.A.S.—Reports of the Glacial Phenomena Committee, 21 (1932), 464. “The Geology of the MacDonnell Ranges and Neighbourhood, Central Australia.” Reports A.N.Z.A.A.S,, 21 (1933), 75-86. “The Australian Sand-ridge Deserts.” Geog. Rev. (1936), 26, (2), 205-227. “Central Australia.” Oxford University Press, 1936 and (2nd ed.) 1946. S. Aust. Royal Society Centenary Address: “The Past, Present and Future of the Society, and its Relation to the Welfare and Progress of the State.” Proc. Roy. Soc. S. Aust., 60 (1936), I-XV “Additions to the Geology of Central Australia.” A.N.Z.A.A.S. Reports, 23 (1937), 89-92. “A Review of the Arid Regions of Australia and their Economic Potentialities.” A.N.Z.A.A.S, Reports—Presidential Address, Section P, 23 (1937), 375 “The Boxhole Crater and the Huckitta Meteorite,” Proc. Roy. Soc. S. Aust., 61 (1937), 187-190. “The Huckitta Meteorite, Central Australia” (with A. R. Alderman). Min. Mag., (1939), 25, (165), 353-371. “The Boxhole Meteoritic Iron, Central Australia” (with A. R, Alderman). Min. Mag., (1940), 25, (168), 481-486 “Simpson Desert Expedition, 1939: Scientific Reports—Introduction, Narrative, pseu and Meteorology.” Proc, Roy, Soc. S. Aust., 69 (1945), 118-139. “Simpson Desert Expedition, 1939: Scientific Reports— No. 6— Geology: The Sand Formations.” Proc. Roy. Soc. S. Aust., 70 (1946), 45-63. Adelaide, 14 July 1947 D. M. NEW GENERA AND SPECIES OF TROMBICULIDAE FROM THE PACIFIC ISLANDS By H. WOMERSLEY AND G.M.KOHLS Summary Description of Larvae-Shape ovate. Length (engorged) to 450, width to 300. Scutum pentagonal, as figured, with the anterior margin lightly sinuous, posterior angle rounded. Sensillae ciliated on distal half, with the sensillae bases nearer to each other than to PL and placed slightly anterior of line of PL. Eyes 2+2, on distinct ocular shields, the posterior eyes the smaller. Chelicerae not serrated on inner (dorsal) edge, with the usual small apical tricuspid cap forming the apical tooth, a small subapical dorsal tooth, and a small subapical ventral tooth. Galeal setae nude. Palpi stout, tibial claw bifurcate; seta on palpal femur 2-3-branched; on genu nude, on tibia only the ventral seta branched; tarsi short, with basal and subapical rods, and five or six ciliated setae, one of which is much stronger than the rest and over-reaches the tip of the tibial claw by the claw’s length. TRANSACTIONS OF THE ROYAL SOCIETY OF SOUTH AUSTRALIA INCORPORATED NEW GENERA AND SPECIES OF TROMBICULIDAE FROM THE PACIFIC ISLANDS By H. Womersrey ! and G, M. Karis @ [Read 10 October 1946] Eutrombicula gymnodactyla n. sp. Tig, 1A-F Description ef Lurvae—Shape ovate. Length (engorged) to 4504, width to $00». Scutum pentagonal, as figured, wiih the amterior margin lightly sinuous, posterior angle rounded, Sensillae ciliated on distal half, with the sensillae bases neurer to each other than to PL and placed slightly anterior of linc of Pl. Eyes [\¢ \ Vie. 1 Fotrambiewia (Aseeriseis) gxmpoitactyle 1. ep. A, dorsal seutim (20900); B, palp in dorsal view (x 8601; -C, tip of chelicera (x 860); D, galeal seta; MT, tarsus and metatarsus LLL (x 450); J¥, dorsal seta ( 860)- 2-+-2, on distinct ocular shields, the posterior cyes the sivaller. Chelicerae not serrated on inner (dorsal) edge, with the usual small apices) tricuspid cap form- ing the apical tovth, a small subapical dorsal tooth, and a simall subapical ventral tooth. Galeal setae nuce. Palpi stout, tibial claw biturcate: seta on palpal femur 2-3-branched: on genu nude, an tibia onty the yeutral Seta branched+ tarsi shart, with basal and subapical rods, and frye or stx citated setae, one of which is much stronger than the rest and over-reaches ihe tip of the tibial claw by the claw's length. Dorsal setac 24 in number, to 3G) long, ciliated auc arranged 2,6,6.4.4.2. Ventrally with a pair of branched setae on guathosoma, one on cach coxa, a pair hetween coxae T and between coxae JIT, thereafter uncertain as all specimens ©) Entomologist, South Australian Museum, ©) Major $9.C., Usited States af America Typhus Coiimission, naw Enlomolovist, TLS. Pulidic Tea'th Service, Rocky Moutta’n Lahoratery, Hamuiltun, Montana. Traus. Roy, Sou. $. Aust. 74 C2), 25 July i947 4 were heavily infested with sporozoa but approximately to 204 in length, Legs: I 2104 long, 11 195. U1 210p; tarsi land Tb with the usual dorsal sensory rods; tarst ILL without any lone mudde set. The Standard Data (sce Woinersley and Tfeaslip 1943) in microns ior the type and three paratypes, Standard ‘Vheoretion! Observed Coelh. of Monit Deviation Hansa” Range Variation AW ... 45-75220-75 1-50 +0156 44+25 -30-25 450-15 +0 are BW... 60-0 ‘No variation recorded SB... Wa =0-87 1-7340°61 Il-3 —21-7 15-90-18 -0 1-5 ASB .. 1-0 «No variation recorded PSP | 2ie2540+75 1-50-0+53 12-75 23-75 210-240 Gr SD, 4)-85-40°75 1-5020+53 36775-4373 39--42-0 36 AP | U5 20°87 1-7GaU-61 Le 3 24-7 10-210 8-8 AM 24-0 No variation recorded AL... 25+5 20°87 1:72:0-61 20-3 37 24-0 27-0 6$ hy re 2'- No variation recorded Sens. . 47-0 #1:°9 L-73aG 71 41-8 —-52:2 Ad +O] 37 Loe—tTis species iy desertbed from four specimens from Gatadactynts lowisiedensis fron: Sansapor, Duteh New Guinea (G. ME. K-, Sansapor, No. 3, 26 Aug. 1944), Renarks—In the pentagonal seutum th’'s species would fall into Sig Thor's enbgenus Penlaqorel/a 1925, but the bifureate palpal claw puts it into Ewing’s Entrombicula 1938, and the number of dorsal setae info -Lscariscus wing 1943, We conentr, however, wilh Michener 1946 (Annais, Ent, Soc. Ainerica, 29, (1), 101118, in regarding vlscarisexs as a synonym of Lutrowbicnia. The type and one peraivpe iv the South Anstralaa Museum, amd two para- typea in the U.S, National Museum. Genus Novstrembicula noy. Larvae—As in Trambicula Berl s1., with the dorsal seutum ftrnished with paired flamentous scnsitlae wil the usual five selac (one AM, two AL, aud two PL}. but produced posteriorly and tongue-lke, sa as to inehide, in addition to the above, same of the median sctee of some of the cersal rows as it the gerus Gahatirpia. * . Genotype Mevotrembicula owigusis mm. sp. Nevotrombicule awiensis 1 sy. Digs 2 A-bT Desciiplion of Lorvas—Shape eubrotund. Size sinall; lenge (unfed) 253 p, width 180). Dorsal seutum large, tongue-shaped and produced posteriorly to cake in the tywo anedian setae of the third dorsal row, in addition ta the AM, AL and PL setee; scusiflaé long ard flamentous with cvisiions om the distal half, bases wide apart and about midway between nes of AL and DL. Eves 2+ 2, os distinet ocular shields placed close ta the lateral seutal margin; posterior eyes the smaller. Cheliccra with only the nsual! apical tricussid cap, but on the immer margin be’ew the cap with a rather prominent subajical angle, Gateal setae long avd nude, Palni rather slender, with relher long slender and |Hfureaic ubial clay, the dorsal prong of which is small and ficistiner; palpal fenry ancl genu with a jone sle:vler, shortly citiared setaz all three setae on thin long and apparently amide; tarsi moderately loti and conical, with a tong sulvzacal sensory vod aud five of six louse eilinted setae, ns a) i . =, t tat Dorsal setae rather thick, slightly tapering, shortly ciliated and of two sizes, arranged 2.4.4[2].4.2.2; the second row with the medial pair just off the scutum and to 57 » long; the inner members of the third raw are on the scutum and are view; B, veniral 860); E, tip of Big 2 Novotrembicula owiensis nog. wid nosp. A, corsa! 1 view: C, dorsal scutum (+500); DTD, palp in dorsal view (x chelicera (x 860); F, waleal seta; G, dorsal seta (x 860); H, tarsus and meta- tarsus IID (x 450), 6 short to 24. Ventrally with a pair of branched or ciliated setae on the gnatho- soma, one ou each coxa, a pair between coxac T and between coxae [IT and there- after 6.2.2.2.2. to 24 long. All three pairs of coxae totiching. Legs: I 255 long, 11 240, JIE 273; tarsi I and If with the tsital sensory rod-like seta, LEL without any long nude seta, The Standard Data in microns for type aud two paratypes are: Standard Thenrelical Ohset'yed Coeff, of Meun Deviution Range Renga Variation AW |... 89-3120 2-NS+0+85 83+1-> 95-5 87-0— 91-0) 2-3 PW.) 85-31-33 2. 310-94 $8-—-102+2 94-0— 98-0 2:4 SB. = GT-0+1-00 1-7340°71 Sark 65-2 600 63+0 2:8 ASB, do-#i-0 L73s0°71 1W-8— 21-2 15:0— 18-0 10-9 PSB , 112-0 No variation recorded SD, 1280-41-00 L+73A0-71 122+8—133-2 127+(}-130+1) 1+3 A-2 .. 20-7403] 0340-22 10-1— 23-3 20-0— 21-0 2-6 AM , 40-3=0-88 1-3340-62 aa7— 44-7 59-G— 42-1) 3-8 PT cays 50-0) No variation recorded PLE ot 43-0 No variation recorded Sens... 67-0221-0 173071 GL:8— 72+2 66-G— 69-0 2-6 Lec-—Six specimens collected from the soil at the base of a large tree ot Owi Island, North Coast of Dutch New Guinea, 15 Aug. 1944 (G. M. K, Owi 22C.), Nemarks—But for the prolongation of the dorsal scutum to include some of the medial dorsal setae, this interesting species would fall into Trombicida $1. The form of the seutum, however. is parallel to that found in the genus Guhrliepia of the group of genera with clavate sensillac and justtties the ereetion of a new genus. The type and two paratypes in the South Australian Museum, three pata- types in the collection of the Rocky Mountain Lahoratory, Schongastia philipi n. sp. Fig. 3 A-G PA i of Larvor — Shape an elongate oyal, Length (unfed) 2364, width 182. Scutum more or less hexagonal, with the posterio? muirein 3-sided and deen hehind line of PL with a slight concavity medially; AM seta very much es than AT. and PL. AL the longest; sensillae mote capitate than clavate i their bases in dine with PL. Eyes 2+ 2, apparently oot on ocular shields, the posterior eyes the smaller. Chelicetae not unusually slender, with apical tricuspid cap and on the inner (dorsal) margin with three to four strong and some smaller teeth. Galeal setae nude. Palpi stout, with trifureate tibial claw; seta on palpal temur ciated or branched; on genn nile: on tibia dorsal and lateral nice, ventral branched; tarzts short, with basal and suhapieal sensory tads and several ciliated setae. . Dorsal setae 36-40 » long, ciliated. 28 io number and arranged 2.8.6.6.4.2 Ventrilly with paired br anched setae on gnathosoma, a ciliated scta on each COXA, a pair hetween coxae I and between coxae TIT, and therealter 4,6.6.6.2.2. to 23 fe lone. Legs: I 256» long, TI 228 ps VT 250 yj tarsi Land If with dorsal sensory rod, TTF with a long nude sela arising subbasally and dorsally. 7 The Standard Data for type and four paratypes are: AW... PW wns SB ASB ... PSB... 0 ne A-P ... AM AL Ph Sens... Standard Theoretical Observed Coelf, of Mean Deviation Range Range Variation 48-61-12 2-5140-80 41-1 36-1 45-0 51-0 st 69-6-20°40) (89-90-28 66-9--F2°3 68°0-70-0 1L+3 18-40-40 0-890-28 15-7—21+1 15+0-20-+0 498 24-0 No yariation recorded 21-0 No variation recorded 43-0 No variation recorded 23+ 60-40 0-892-0-28 209-2643 22-C—24-(0) a8 22:240-73 1-4a-0+52 17-+3-27°1 21-0-24-U 7-4 59-2+1)-49 1-0440-34 §5°9--62°+5 58-0—60-0 1-8 426-1712 2°3140-79 35-1—30+1 39-U—45-0 5-9 30-0 with head 15/18 Vig. 3 Schingastia philipi n. sp. scutum (x S00}; D, palp in dorsal view (x 860); E, tip of chelicera (x 860}5 I’, galeal seta; G, tarsus and metatarstis [II (450). A, dorsal view; B, ventral view; C, dorsal Loc—Ty¥pe and nine paratypes from a Vizard Letolepisina alpertisti from Goodenough Isiand, New Guinea. 17 Jan, 1944 (3. M. KL, No. 486). Memerks—The dentition of the chelicorae in this species ig somewhat different trota the many small tecth gound in other species of Schdngastio, but this fs not sufficient to warrant generic separation, The type and four paratypes in the South Australian Masenm, and one para- type to Ls deposited in the Eritish Museum; one paratype to be deposited in the U.S, National Musetm and in the Rocky Bfiotaquain Labertiory. Gents Oeneschongactia nov. Alked to SelOngusiis but the cholicerae are short, curved aui stumpy with the apex divided jute tvoa terminal Wiuny and thick teeth. with a pair (sometimes three), of rather smaller teeth subapically. Valpal claw shorc and stout, tri- fureate. Wledian tarsal chaw (empocdium) longer than the laterals but equally thick. Sensillae clavate or eapitate. OQenoschongastia cana n.sp, Tig. 4.A-G Leseription of Lartae—Shape an clongate oval, Lenath (infed) 252), wilth 162.2. Darsal scutwm rouehly hexagonal with indistinet and fine pitting; with the usual five nosreal ciFated setae , nf which AL are the lomeest and AM the shortest, AM wich shart branches, AL and PL with long outstanding branches; sensillaé broadty clavate, the head indistinetly and very shortly setulose, sensillae bases about m lew with Pl; anterior seittcl matyin convex, posterfor laterally angular, Thyes 2-+ 2, large, on well-developed ocular shields and closely adjacent to seutum, posterior eyes the enaller. Chelicerae of peculiar form, short, stow! and curved, apex truncate andl divided into two sirony, broad and blunt teeth, and two (sometimes three) smaller teeth placed subap2eally on the inner edge. {anleal setae imide. Palpi stout, tibial elow short, stout and trifureate; femur with a lone atrong ela with long branches; genu with a nude seta; tibia willt the dorsal and Taleral ectae mace, ventral branched: tarsi shart with basal and subapical sensory rods and three or four ciliated or branched sctac. Dorsal setae strange with strong ciliations, 36 im number ani arranged 2.8.10.6.6.4., t0 40-309 » Jong. Ventrally with the ustal pair of branched setae of enathosoia, a single sela. on each coxa, a paly between coxee T and between coxae ITL, and there- after 6.6.64.2, mere gender and with longer cations than the dorsal setac, to 30 tous. Lees: 1324p jong, 11 252 p. TEL S80 p: tarsi T and TT with the usuat dorsal mod-like setae, tarsi [LE with a loug outstanding nude seta*; median clay (empndium) longer than bar as thick a) the laterals. The Standard Data tn wictoms derived frem 22 specimens are: Standard Th-ore/ ica! Ohserved Coeff. at Mean Devintion Range Range Variation AW 62°8540-56 2H5+0+-40 FOS 69-0- 69-0 4.2 LAY 787 &)7F1 3:34+40-50 7 Foo aU 4-2 £6 — 34-854-0-49 2-3020-355 “75 D0+0--39- 0) 7-2 ASD... 28:9 0-35 Leedck0- 25 3 27th 22-0 5:3 3 She§ ees Peqeet he D1 370 24-)—27 +0) 5:8 Pe) 2-8 se()-45 2-ll4+f)+32 v1 4$8-0—57-0) 4-U AAS 29-2 m1) 32 1-30-1023 7 27-Y—F3+0 5-2 JA $2+452e0 46 2:1540°+32 9 7 -0-36-0 G6 Al. 74-3 &)-50 2a3tb0- 45 §1-3 7 D—-B0-0 3-0 PE ... 89s§ seDS9 1:8-f-F0' 28 Hel SS -fa-u a-l Seng... 460 with head 20/24 * Lniertunately omitted from fig. 4G. v7] Loc—The type and 72 paratypes from the mound of a brush turkey, Dobo- dura, New Guinea, 18 May if44 (G. M. IK, No, 325). The type and 22 paratype deposited in the South Avstralian Museum, 35 paratypes in the collection of the Roeky Mountain Laboratory, five paratypes m the U.S, National Museum and five in the British Museum. Yenierks—This genus and species in the pecttliar and characteristic chelicerae and the palpal claw is very distinct from Schdngasiia, Neoschongasta and Ascusehongasia, FR =. i ( ‘ A, dorsal view; B, dorsal cheticera (x 860); arsits LIT tx)450). Tig. 4 Ovnoschdngasha cure nog. pid nosp seutum (x500); C, palp ia dorsai view (x860); D, £, galeal seta; T, dorsal seta (x 860); G, tarsus and smetat 10 Ascoschongastia “) uromys n. sp. Fig, 5A-F Description of Larvae —Shape oval. Tength (amengorged) 300», width 200. Scutum roughly rectangular with fairly well produced posterior margin. AM short and normally ciliated, AL and PT lanceolate or foliate, to 16 4 wide, Tig. 5 Aseasehangestia wromjyys uesp. B, ventral view: €, dorsal scutum (xf00); D, tip of chelicera {x 860); Ic. waleal sela; Fy, darsal ait yertral s with longitudinal rews af strong dentures. Sensillae glohose, apparently nude, with their bases only slight!y in advance of Hne of [L. bwes 2-- 2, .0n distinct oculer shields; posterior the smaller, Chekeerae with only the usual apical tricuspid @) Ewing (Proc. Biol Soe. Washineton, 53, 1940, p. 70-71) : that the genus Paraschdngastia Wom, is syvuonymous with Neoseliire (genotype WW. aiericave), aml for the remainder of the species of LV conspecifie with emericdua lias proposed the name of slacos % Gater as genotype. The choeiee of this Species at genotype is, however, somewhat mil+ fortunate, as it is a rather aberrant species with PL distinetly off the seutum. recently shaw Ha Esving 1929 coschougusha nat IL eap. Galeal setae nude. Palpi fairly stoyt, with bifurcate tibial claw. Setae on palpal femur, genu and tarsi apparently all nine. Dorsal. sctac all foliate or lanceolate with strong dentures (ci. fig. 5D), to 64 long and 16m wide, 28 in number, and arranged 2,6.6,6,6.2. Ventrally with usual pair oF branched setae on gnathosoma, a single ciliated seta om each coxa, a pair between coxac I and between coxaeé IL, and thereafter 5.6.6, anterior of atius, and of which the outer two of the second six approach the doreal setae in form; then posterior of anus 42, foliate and dentate as on dorsimy the amlerior non-ftoliate ventral setac to 20 2 long. Legs: 1270p dong, if 220», 117 270; tarsi Land I with the usual dorsal sensory rod, FIL without any long nude seta. The Standard Data derived from the type and seven paratypes are as follows: Staudare Thearetion! Observe l Coeff. of Mean Syiatian Ranes Raines Variation NW 2. 50+6#1-74 4.02+1-23 35-B—fae 4 £5-0—-60+4 9-7 PAY, 09-S260-71 2-000-50 h3*5—73+5 t6-0°-75-0 2-9 Si dik +E0 1-41+0735 I4-8-- 23-2 18+0—21-0 7-4 ASE. 21-0 No varintiun recorded PSB. 18-0 Wo variation recorded SDs 34-0 No variation recorded A-P . 21+G0-42 PT 30 18-0 25-2 1-0--24-0 5°5 AM (2. 1G:520-67 164047 11-G—21-4 =3-0-18-0) 9-9 AL ww 40-50-67 1-64220 +47 44-6- 54-4 ++8-U—31+0 GA PL on. | GO O0+57 1-60+0-40 ABe2 fide d 47-063 +0 27 Seis... 28-1) with head 17/17. Only ore determination Loc —The type and. seven paratypes from Crore lainington from Doho- dura, New Guinea, 16 May 1044 (No. 315, G. M, K-.). The type and two paratypes in the South Anstratian Museum, three para- types in the Rocky Mountain Lahoratery, one paratype each in the U.S. National Museum, aid the British Museum CN. 1T,). Remarks—This species falls into the small gronp with [oliate seutat and dorsal setae comprising 4. Jrasiate (Gater), aneeullechi (Wom.), foliata (Gunther) and another species at present unpublishes, Tt is close to wicenllochi but car. be disUugnished by the differcat mumber and arrangement of the dorsal aid ventral setae, and by the centationa of the sctue, althatw's: stron, being nowhere so large as in aicerdocin. Ascosshongustia echymipeta n, si. Tig. GA-C Desrviptien of Larvae—Shape oval, Length (engorged) te 460», width to 350, Seuium toughb> reevangular, rather small, and faivly long, with the posterior juargin yery shallow helen] PL. awl ightly sineetis; AM the shortest, AL the longest but omy a fit: longer than PL, Sensil'ae giahose with short distinet ciliatious; the sensillac bases ebout midway hutweeu AT. and PL. byes 2-|-2, on distinct oeular shielde antl well separate Trem semtal margin; the posterior eyes very much the smailer. Chelicerne with only the apical trienspid cap. Galgal setae nnde, Palpi stout, tibial claw trifurcate; setae on paypal femur and gent iranched, on tibia all three mide, DS 34-48 in number, arranged 2.8.6.6.64.5(2),2(0), to 30¢ long. Vertrally with a pair of branched sttac om 12 guathosoma, a single ciliated setu on each coxa, a pair between coxae I and betiwecn coxae UI, and thereaiter ca. 6.66444.2. to 26 long, Lees: 1 213. lone, h 186, LU 220 pe 5 tarst I aucl il witl: the usual dorsal sensory rod, Iti without any long nue sein. Tue Standard Data from the type and 12 paratypes and threo other epeci- DI eS Seundard Observed Caelt. of Mean Deviatiog Range Variation . fhe jokthe 2) 1-1620+20 Hdd 45h 434-484) B65 paw o., | GleSckG- BG 2:12+0-87 55-167 4 57 -0-—-06-0 34 Noo a 2ide toe (3 1-17-4021 Tos!) 2o-4 is Gailey 5-7 21-0 No yeviation recorded 15-9 iion recorded 36-0 No variation recorded nor SL GAO 149+ +25 DGB aueg— ies AF SKS ZOE O35 1+45240226 a? 3 c BG Sofs Ga BY 219720 +42 ou a3 ( el Low. ©6442 Ge 34 MONE date Feb, i-4 ead i7/ fh — lscoschdngastia echinipera a.sp. A, dorsal seutum (x 500); By tin ny of chelicera (2 809); C, pulp in doraat view (x 860), Tg. suc—The type and 12 panatye es from Echaiitpera ceckerelli from Dobu- aura, ners Guinea, 29 Nov. 1943 (No. 46, G. MM. IS.), and another paratype from ganie host, locality and date (No. 48, G. M, 48.) ; also two other specimens from sun’ Lost and locality, 8 Dec. 1943 (No. 77, G. M. K.) 10 Dee, 1943 (No, 81, CG. 30 K.), ike type and four paratypes in the Seuth Australian Muscum, two para- types each in the US. Nat. Miseam aud the British Museum CN. 41.) : reniaim- ing epecimets in the Rocky Mountain Laboratory. Rewnarks—in a new key to the species of AlecoschGugastia in preparation, echyinivera will run down close to coorengense Jlirst and innisfailensis Wom. and Heasp.. but is casiiy distinguished ly the seutum and the mumiber and arrange- P.. pt } 3 § ment of the DS. AUSTRALIAN ACANTHOCEPHALA By T. HARVEY JOHNSTON AND S.J.EDMONDS Summary Parasites of this species were found in the fish Callionymus calauropomus, caught in St. Vincent Gulf, South Australia. Five of six fish examined contained this parasite, and in four of them they were abundant. The intestines of all these fish contained crustacean material, especially amphipods and cyprids. The collections examined contained both males and females, the latter being slightly longer and broader than the males. In all the specimens examined the proboscis was protruded, but in no case was the copulatory bursa of the male everted. 13 AUSTEALIAN ACANTHOCEPHALA faa, 5 Ly T. Lisevey Joursrow and S.J. Romenes, Department of Zovlogy, Universtiy of Adelaide [Read 10 Octoher 1946] livrorecninoruyscnus ALALotis Yamigati 1939 (Fig, 1-9) Parasites of this species were found iu the fish Cilfiowvmus cooanrepenues, caught mi St. Vincent Gulf, Sovth Australis, Dive uf six lish examine! con- iained tis parasite. and in four of ern they were abundant. The intestines of all these fish contained crusineein maceria!l, especialy amphipods and cyprids. The collections examine! contained bot males and females, te letter heing slightly longer and broader then the mates, In all the specimens examined the proboscis was protruded, but In no case was the copolatory bersa of the male evertesh Vor Teneth of the tale is 1-4 10 2°7 mim, and the female 1°6 ta 2-8 mm, The meaxiniing with of che tusle ia O45 tn G70 wim,, and the female G50 to 0-91 win. The beedly in both sexes is enrved ventrally anc devoid of spmes. The préboscis is globular ta spherical in shape aud ig attached to the trumk ventre- terminally (fiy. 1). The proboscis of the male is 0-10 to 0-14 mim. iong anil Q-i0 te G-15 wim, in tts widest part. “Vhe correspouline méasurements in the femate are 0-12 ta O16 mm. The acck portion of the proboscis is very short, The srohoscis is armed with 25 hooks arranged in (en fongitudinal rows, consist- ina af fiye rows each of three hocks alternating will fwe rows each of two hooks (fy. 4.55. The lengths of the hooks, neasured along the curve from the pout ef exivasion to the tip of the hook, are shown in the following table -— Male Anieriot Middle Posterior Row of three .... ace me 8 ta 78-1 jh 3U-35 jf 21-28 fe Row of two, ‘ne ae ot 71-89 is 23~32 14 Temiale Antesine halle Pustertur Row of three wi va BI-LO B42 28-82 py Row af twa... +. Ate me 7 29-33 14 The proboscis sheath is bwb-like and hr the imale measures G12 to 0-22 mm. Tong and OTL to O-14 mn. in its widest part. ‘The corresponding measurements of the femate are 0712 to 0-20 mim. and O11 to O16 mm, ‘Phe sheath, which is inserted wt the buse uf the prohoseis, is douhle-walle!. “Ube otaximuwm thickness of each Jayer in both sexes is 1) to lay, A spindle-shaped cangtion is strated at the pesterior end of the sheath, and the retinatula urise from the sie walls at about this Tevel, A strongly deveaped retractor is present in hoth soxes. The lenmisei aye short, elodt and eyindrical, and contain a well-developed tnewnar system. The hypodermis is thick aud the lacunae of the body wall anastomoses irecly. Ail: system—There ure two sphéricnl to oval-shaped testes jtuced one behing the other bul usualy pressed close together. They lie in the anterior part slighty larger than the posterior, the dinien- of the worm. The anterior testis is sions of the former being O21 to 0-28 mm. Jong and €-20 to 0-26 mm. wide; and of the posterior 0°20 to 0-22 mim. Jong and O18 to O25 wim, wide. Two yaa efrerentia uiile near the anterver end of Lhe Saeiftigea’s pouch ty forin a commit duct, which in most specinens is swollen at its hase te lerm a seminal vesicle, This terminates tn a penis which projects into (he atrium of the bursa Trang. Roy. Sac. 5S, Aust. 72 C1), 25 July 1947 4 and which is enclosed in a capsule or génital papilla. There are six cement glands which are elliptical to pyriform, In most cases they lic pressed closely together, The ducts of the six glands unite to [orm two lateral ducts, which join at their buses to form a U-shaped cement reservoir. Two well-developed diverticula project anteriorly from the bursa. The genital pare is terminal. Fig. 1-9—Aypoeeliinorhyeechus alacolsts: _ Aer Female systeim—The general anatomy of the female system is shown in fig. 6. In a typical spectmen the nterine bell ig O-15 mm. long and is separated from the ulerus by a tiarcaw constriction, O't2 vu, in length. The uterus in the same specimen is 0-35 mm. long and 0: 07 mm. wide at the anterior end, Some of the female specimens contain floating ovaries, while the others bave both ovaries and eges. Mature eges, when mounted in methyl salicylate, measure 50 to 54 long and 13 to 14 wide, The polar extrusions of the middle shell are. well developed. A mamber of fennies bear a copulatory cap at their posterior extrenity, Systematic position—We consider that this parasite belongs to the species Ifypoechinorhynchis aletcapis, deseribed by Yamaguti (1939, 325), from a Japanese fish, Our measurements agree very closely with those given by him. The hooks of our specimens, however, seem slightly longer, aiid the testes some- what smaller than in his material We regard Mypoccimorhynuchus as a yalid genus of the family Echinorhynenidae. The form of the probosets, as. well as the shape, number and arrangement of its hooks, are sttipeestive of those of Neocclitrorhynchus, but the characters of the cement glinds. are qanle different, Although the parasites described hy Yamaguli were taken trom Alacups flinthus, he stated that a single immature Temale spechnen was collected from Cullionuyinus altigelis, Pararhadinorhynchus mugilis n. gen., n. sp, (Pix, 10-22 ‘Thus species necurs in the mullet, Mugil cophalvs. Five of six fish examine irom Port Wihinga in March, 1939, were parasitised, in ome case heavily. The intestine of all the fish contained much plant debris with occasional mollises and small crustaceans. Two other fish taken at American River, Kangaroo istand, by Mr, TT. M, Coaper in January, 1945, also contained the same specics of parasite im considerable mumbers. In the stomach of these fish were gastropods and maumerous crustaceans (prawns, anmhlipods and copepods). Both imate and female specimens were collected. The worms ate long ard eylidrical, the ferale being longer and slightly broader than the male. Both sexes are devoid of hotly spines, Th 2 length of the inale ranges from 3-1 do 11-4 mm., aud the wmuniun width from 0-23 to O61 mm. The feimates are from 3° ta 19*2 ton, Jong and from 0-22 to 0-69 mm. wide. The posterior hali or third of the female is twisted in most cases into two or three spirallike convolutions fig. 19). Thig may be duce to the fixing processes, All incastretictits were mare on anunalé cleared in nicthyt salicylate. Although the colicetion consists of a considerable namber uf both sexes, in vnly two is the proboseis fully extcoded. It is therefore dificnt ta give a range of values for the Jength mvt breadth oF that organ, Whe proboscis ia most adult specimens is aoout OO nut. lone aud 0-2 nun. wile in the breadest part I appears Lo taper clightiy tewaréls the hase, It bears 18 loneiadinal rows of hooks, midst ef which are firntly attached hy rooting processes to the cuticle. Each row consists af 16 to 17 hooks. The form of the proboscis [s shown in fig 10, and TSCRIPTION GF FIG. 1-9 1, male; 2, female; 3, proftoscis; 4, 5. rows of hooks: @ female organs; 7, exer; 8, ToS. anale, through cement @lands; 9, TS. male, through cement duets. bh, lourta, Lin, brain; bw, bady wally ev, conitlatery cay od, counent duct: ef, comgulated body Mid; ce, cement gland; er, cement reserveirs hit eaewatory diet: =, sanelion; fo, woiital opening: gy, genital papilla: tf. earl aae a, Tacuuas lw. ligament; lin, loins tudinal mnsele; lo, lateral openingeof uterine bells ny nielens: BY, probasrig; ps, grohuscis shell; enh vetractar muscle: Sp, SeelFhigen’s peureliz sph, sphigcter; 1 testis: ny uteras; nb, aterine bell: yh, vagritio) lath; vd, vue deferens; ve, vesierta seatinalis, 16 View, 10 Pararhedpweyrciag agucles UM, pate fram an mir ley 14, 15, 1.5. titeveh cement dects: Wi PWS. through reginn of ejaculatory duct; 17, T.S. reeion of esnution of izle. or, middle anid posterior regions al pro ; 12 nosterior end of VS. anteciur enil of aale; 14, T.S. through region of cement glands, 17 the sive ard fur of the hocks in fie, 11. Vhere is a slieht nech region, he proboscis sheath is double-wailed Land mieasives From Obl ia 1:3 tr. long, and from O-£2 to O20 min. wide. The thickness of earth wall ig about 0°02 nan, WA brain is situated towards the base of the praboscrs sheath, The lacuuar yystem of the Sody wrall shows tio well-developed tor niin ‘nal lacunae, fromm which anastomasing chamuels arise (fe. 22), Nunerous small nirclel ate foutid in the borly wall, The Temnisct are about 0-8 mm. lang al exiend wadally as far as the posterior portion ot the prahaseis sient, Transverse: acetivnas of dhe lenimiiset shaw that they ate fat and that pyvo lacerat canals und a mmuber at largs 5 are po preset im these structures (fig, La). Male systent—There are two elougute testes which No closé together. one behind the ather, in the posterior half Gr rnd of Wie anual The anterior lestis mecsires from 0°28 to 1-1 mm tone and trom O08 to O24 mim, wide. and the posterior 9-27 to 1+) inn. long and O-58 ta 0°23 ing. wide. There are tivo Tome, narreyy cotenl glands whieh reaee iin lengrts from O-fS to 2-5 mou. and is most specimens they are swollen posterkaly. The duets frou these glands turn tivo long cement reservoirs which usally ero constticterl in cne or two plaves iowurds theiy posterior extremities. A long Sacituren's pouch lies hetwean the two eement ducts and reaches forware as far as tite distal euds of the cement stats. The vas deferens swells slightly towards jis postertar part to form a seminal vesicle. There is an ejaculatory Guet and a weil- developed bursa which pears rays. In none of the species examined was the bursa everted. The male apertur: is terminal and 1s cuirrounded by manerous célis probably constituting a ganglion, Aemal’ systert—The structure and arraweenanit of the female y ’ Pp My oblique; whitish sprinkled with fuscous; markings fuscous; a line from base of costa to two-thirds dorsum; a postmedian square uniting this with costa; a very fine line from costa near apex to two-thirds dorsum; some terminal suffusion; a terminal series of dots; cilia {uscous. Hindwings and cilia pale grey. New South Wales: Broken Hill in March; one specimen. CATEREMNA LEUCARMA Meyr. Proc. Linn. Soc, N.S.W., 1880, 230. Brisbane, Cunnamulla, Sydney, Birchip. CATEREMNA ALBICOsTALIS Luc. Proc, Roy. Soc. Qld., 1891, 93. Cairns, Atherton, Townsville, Mackay, Bundaberg, Brisbane, Stradbroke Island, Lismore. CATEREMNA METALLOPA I.ow, Proc. Linn, Soc. N.S.W., 1898, 46. Mackay. Cateremna melanomita n. sp. peAayopcros, With blackish lines. $. 18mm. Head and thorax grey. Palpi grey; second joint with post- median whitish ring. Antennae grey. Abdomen fuseous; extreme base of dorsum and tuft whitish. Forewings dilated posteriorly, costa slightly arched, apex rectangular, termen almost straight, slightly oblique; grey densely sprinkled with fuscous to middle, thence slightly; a sinuate blackish line from midcosta to mid-dorsuin; a similar doubly sinuate subterminal line; a terminal series of blackish dots; cilia grey Huindwings and cilia grey. North Queensland: Mackay in October; one specimen. CATEREMNA PAMPHAES Tutn, Proc, Roy. Soc. Qld., 1904, 47. Darwin, Townsville. CATEREMNA HNEMIBAPHES Turn. Proc. Roy. Soc. QOld., 1904, 47. Sea Lake, Hobart, Quairading, Perth, CATEREMNA ATOoDEcTA Turn, Proc. Roy. Soe. Qid., 1903, 129, Brisbane, Scone, Sydney. CATEREMNA THERMOCHEOA Low. Trans. Roy. Soe, S. Aust., 1896, 160. Darwin, Brishane, Cardiit, W. Aust. CATEREMNA ODONTOSEMA Turn. Proce. Roy. Soc, Qld., 1912, 126, Cairns, Imbil. Gen. Tynocuares Meyr. Ent. Mo. Mag., 1883, 256. Palpi ascending, recurved. Forewings with 4 and 5 stalked, 8 and 9 stalked. Hindwings with cell short (one-fifth to one-fourth), 2 from near angle, 3 and 4 stalked, 5 from angle. 50 TYLOCHARES COSMIELLA Meyr. Proc. Linn. Soe. N.S.W., 1878, 212. Duaringa, Brisbane, Sydney, Moruya, Broken Hill, Melbourne, Dirchip, Murtoa, Wirrabara, Mount Liebig, Perth, Rottnest Island. Tylochares epaxia n. sp. eraéeos, handsome. 2. 23mm. Head brown; face whitish Palni and antennae grey, Thorax grey; patagia whitish, Abdomen ochreous with a series of central inscous dats. Forewmgs with costa almost straight, apex subrectangular, termen slightly rounded, scarcely oblique; grey, a well-marked straight white sub-basal line, followed by a broad fuscous transverse fascia suffused posteriarly; an out- wardly curved stender whitish subterminzl line, indented beneath costa and above dorsum, edged posteriorly by a series of confluent fuscous dots, an interrtipted fuscous terminal line preceded by whitish suffusion; cilia grey, apices whitish. Hindwings bright ochreous; cilia grey, apices white, North Queensland: Lake Darrine, Atherton Tableland, in January (E. J. Dumigan) ; one specimen. Tylchares prays n. sp. murs, gertle. ©. 24 mm. Head, antennae, and thorax grey, Paipi grey; apiees of second and terminal joints white. Forewings dilated posteriorly, costa rather strongly arched, apex obtitse, termen slightly rounded, slightly oblique; grey sprinkled with whitish; a fuscous basal patch containing an oblique outwardly curved hlackisii transverse line; closely following this a slender outwardly ciirved grey line irom one-fourth costa to two-niths dorsum; a grey median dot above middie and another heneath two-thirds costa; a sufltised grey spot above three-fifths dorstim; a broadly suffused grey submarginal line not reaching turnus; a terminal series of grey or fuscous clots; cilia grey; apices white. Hindwings pale grey; cilia white with a grey median line. QOucensland: Bunya Mountains in January; Stanthorpe in March. Two specimens. TYLOCHARES SscertucHa Turn, Proc. Roy. Soc. Qid., 1903, 130, Stanthorpe, Gisborne. Tyjochares gypsotypa n. sp. Anforuros, with white marking, 8, 2. 18-21 mm. Lead, antennae, and thorax grey. Valpi long, much exceeding vertex, in male 2 and a hali, in iemale +; grey, towards base white. Abdomen ochreous, towards base grey, Torewings narrow, costa slightly arched, apex obtuse, termen obliquely ronnded; grey; a broad white stripe, sprinkled with grey, narrow ac cach end, from base to apex; sometimes a setics of munute fuscous terminal dots; cilia grey. Tlindwings pale grey with darker terminal lime; cilia white. North Queensland: Cape York in October and Noyeniber (\W. Ji. Barnard) ; two specimens, Type in Queensland Museum. Tylocheres anaxia n. sp. dvaévos, of little worth. 2, 9. 16-17 mm. Head, palpi, antennae, thorax, and abdomen erey. Forewings dilated posteriorly, costa slightly arched, apex obtuse, termen slightly il oblique; grey; a fine whitish transverse line at one-third, angled outwards in middle, preceded and followed by more or less fuscous suffusion; short fuscous streaks On veins i terminal area; a terminal series of fuscous dots; cilia grey- whitish. Hindwings and cilia grey-whitish, North Queensland: Cape York in November, Dunk Island in May; three specimens, Tylochares paucinotata n, sp. panctotatus, scantily marked. 2. 16 mm, THead and thorax fuscous-brown; face pale brownish. Anteunae fuscons, (Abdomen missing.) Forewings narrow, dilaled posteriorly, apex rounded, termen obliquely rounded; greyish-brown with some fuscous dots; three placed transversely al one-third; a median dot; an outwardly curved stib- termina! line of dots; cilia grey, Hindwings grey-whitisl ; cilia whitish with a faint sub-basal grey line, Cape York in Octoher (W. B. Barnard); one specinien, TYLOcHARES EREMONOMA Turn, Proc, Ray, Soc, Qld., 1912, 125, Adavale, TYLOCHARES PROLEUCA Low. Trans. Roy. Soe. S. Aust., 1903, 58. Srisbane, Deniliquin, TYLOCHARES HEMICHUIONEA Turn. Proc. Rov, Soe. Old., 1912, 226, Caloundra, Brisbane, Tylochares chicnopleura n- sp. xlovorAevpas. With snow-white costa, é, @. 16-20 mm. Tlead.palpi, antennac, anc} thorax fuscous. Abdomen pale grey. Forewings narrow, costa straight, apex rounded; fuscous; a snaw- white costal stripe fram near base to near apex; costal edge fuscous: cilia white, on tarnus grey, but sumetimes wholly grey. Hindwings and cilia pale grey. (Jueensland: Watwick in December, from laryae feeding in Acacia galls; two speciniens, Tylochares endophaga nz. sp, ‘vaadayos, feeding internally. @. 21 mm. Head, palpi, and antennae grey. Palpi not excceding vertex; grey, towards base white. Abdomen with basal half ocfireous, terminal half erey, Forewings with costa slightly arched, apex rectangular, termen strajelit, not oblique; grey ; a white stripe, sprinkled with grey, from base to near apex; cilia srev. Ilindwings grey wlutish; cilia whitish. Queensland: Bribie Island, near Culoundra, in October; one specimen froin larya feeding in an coca gall. Tylochates pastopleura n. sp. martcaTAcupes, With eprinkled costa, 4 2, 9. 15-16 mm. Head, antennac, and thorax grey, Palpi fuscous. Abdonicn whitish, in femate slightly orhreous-tinged. Forewings narrow, costa slightly ached, apex rounded; grey; a costal stripe from base to apex, over one- third breadth of wing, sprinkled’ with fuscous; a terminal series of minute fus- cous dots; cilia whitish cr grey-whitish, Hindwings with cell one-third ; grey- whitish; cilia whitish, North Queensland: Cape Yorly in Ociober and November (W. B. Barnard) ; two specinens. Type in Queensland Museum. n 2 ty TYLOCHARES GONIOSTICHA Tt, Trans. Roy. Soc. S. Aust., 1915, 803. Musgrave Range. Gen. PemeeitiA Hb. Verz., 369. Palpi ascending, recurved. Forewings with 2 from well above angle, 5 separate, 8 and 9 stalked. Hindwings with cell short (one-fifth), 2 from angle, 3 and 4 stalked, 5 absent. arate OPIMELLA Meyr. Proc. Roy. Sac. N.S.\W., 1878, 20 Brisbane, Mount Tamborine, Siasithose: Miles. Miulmerran. rend an CANILINEA Meyr. Proc. Linn, Suc, N.S.W., 1878, 209, Brisbane, Toowuonbhe, Carnarvon Rance, Marrurundi, Sydney, Goulburn. Eatoomka. PEMPELIA HEMICIILAENA Meyr, Trans. Ent. Soc., 1887, 260, Victoria, PEMPELIA MicRocosMA Low. Tratis. Roy. Soc. S. Aust., 1893, 166. Gen, Trissonca Meyr. Proce. Linn. Soc. N-S.W., 1882, 138. Paipi ascendiig, tectrved. Norewings with 5 separate, approximated at origin, § and 9 staked. Hindwings with ceil shorl (one-fifth to one-third), 2 from angle, 3 and 4 statked, 5 absent. Trissonca clytepa in sp. kAuromos, Noble. $, 9. 14-17 mm. Head fuscous; lower edge of face whitish. Palpi and e thorax fuscous. Antennae grey. Abdomen ochreous. Tees fuscous with whitish rings; SBOSHAEAE pair whit-sh. Fores NY igh: rs tather narrow. posteriorly dilated, costa d, termen t scareery obliqee; a ftuseois led 1 r blackish Tne posteriorly from two-fifths costa to mid- dorsum, atigied outwards beneath costa and again above muddle; disc bevond tins grey or partly grey-w hitish, with some fusseus adiwixttire; an outwardly curved ilackish subterminal tine; a sleriicr blackish submargival line preeeced Ly more of less white sulfusion; cilia prey, apices whitish, North Queensland: Kuranda (f. P. Dodd) ; ; two specimens. yates A finuted by a sleiiie TRISSONCA TANTUEMIS Meyr. Trans. Ent. Soc., LE97, ry epiterpies Tura., Pros. Rov. Soe, Old., 104, 42 Darwin, Cai thestot, Towisyille, Bowen, Paingcta, Yeppoon, Hfshariey Toow bomba, C Cc SSUNCA Menperitonsa Law. Trans. Roy. Soc, S. Say 1903, 37. Cooktawn, Towrsville, Nambour, Priskanc, Tweed Heads, TRIESONCA MESACTELLA Meyr, Proe. Linn. Soc. N.S.W., 1879, 225. Sydney 53 Gen. IlyroceypHta Rag. Bul. Soc. Mnt. Fr, 1890, 119. Paipi long, porrect, Maxillary palpi in male penicillate. Forewings with 4and 5 st alleed. 8 and 9 stalked. Tiindwings with cell short, 2 from angle, 3 and 4 stalked, 5 absent. Hypogryphia amictodes n. sp. duixtweys, Untiarked, é, 9. 22-23 mm, Head and thorax grevish-hrosu, Palpi 4; grey, near base wlitish, Antennae and abdomen grey. Forewings narrow, posteriorly dilated, costa siraight to middie, thence gentiy arched. epex rounded, termen obliquely rounded; greyish-browtn; ciliz concolorovs. Ilindwings pale ochreous- grey; chia whitish, Queensland: Duyaringa in December; Toowoomba in October (W. B, Bar- hard) 3 two speecmens. Type in Qieenstard Museum, Gen. Ancyzos!s Zel Tsis, 1839, 178. Palpi porrect. Manitary palpi fiiiorm. Porewings with 5 separate, 8 andl 9 stalked. Hindwings with cell short, 2 fram angle, 3 and + stalked. 5 absent. Type, 4. cinitamonea Dup., from Turope. ANCYLOSIS LAPSALIS. Wik, 29, 829, é, %. 14-16 mm. Head, palpi, antennae and thorax fuscous. Abdomen pale grey. Dorewings with costa slightly arched, apex routdl-pointed, termen straight, slightly oblique; pale ochreovs-grey; a fuscous de~ below middle at one- third ; yanartow dark uscous terminal fascia; eilin grey; apeces whitish. North Queensland: Townsville, Queenslanil: Yeppoen, Also from Ceylon, AwneyLosis ecrirAscra Hinps. Rom. Mem, 8, 193, a} 2 22-25 nim. Tlead and thorax whitish-ocshreois; patagia and legulae pinkish, Palpi @; lower edee yess e Aiitennae pate grev. Abdomen grey- whitish. Torewings with cotta ¢ r tu imide, theme arched, apex pointed, termen very obliquely rounnled sw h more ar jess piesish-tinged with slight inkling; a median line from hase jo two- hirds ar more, edeed ales with beneath wit blackish scales; a termina) series of bieckie dots; cilia winitish, fiindwinks gray-whitish: cia whitish. Norh Queevsinud: Towusville, Qucensland: Peek Downs, Gayodah, Mam- hiowr, Brishanc, Dalby, lnjtine. New South Wales: Scone. fciastiyes, saiphur-lined. os 22-24 mm. Jicad ant thorax ochreots-srey, Valoi 6; grey, lower etige exoept termimd jomt whitish. Antennae grey. “\ritoimen Fuseents. tore- swith costa genily arched, apex obtuse, termer ob iaely rounded; grey, in dite suffused with whitish; ia fomale a fite pale yellow Hue frem bate to apex; near apex, a broader stibmedian Hise trom base to terimen; a sinivier subdorsal line nich narrower exoest near ly a subcostal fiecons dot at one-third and another at end of cel; a terminal series of blackish dois; clin grey. Liindwings grey; ciliu grey, towards corsum whitish, Queensland: Tijune in Ociobcr (W. SB. Barnard); two specimens. Tyve Quecnslonul Musenm. PRE-CAMBRIAN GRANITES AND GRANITISATION, WITH SPECIAL REFERENCE TO WESTERN AUSTRALIA AND SOUTH AUSTRALIA BY KEITH R. MILES, DEPARTMENT OF MINES, ADELAIDE Summary A review of the Australian Pre-Cambrian Succession reveals occurrences of granites and rocks. of granitic appearance encountered over very considerable areas and apparently involving vast periods of geological time. With the changing modern concepts of the origin of granite and of granitic- looking rocks, all evidence concerning the relationships of these rocks in both time and space, both with one another and with other adjacent geological formations, assumes a new significance, from the point of view of both historical geology and the specialized problems of petrogenesis. It is time that such evidence should be brought forth once more and critically re-examined in this new light. Periodical stocktakings are always valuable, and, to the scientific worker, can prove both salutary and encouraging. 54 PRE-CAMBRIAN GRANITES AND GRANITISATION, WITH SPECIAL REFERENCE TO WESTERN AUSTRALIA AND SOUTH AUSTRALIA By Reitur R. Mares, Department of Mines, Adelaide [Read 14 November 1946] PLate [ CONTENTS Tage TR TRADUCTION A : gd re a me 64+ AA Qeiotn of Gir siete Nie WILKS C OTeTPTS a ws a ea OS GGRANITTSATION is 2 4, a. AP a Se BS GRANITIEAMION AND THE Pry Cadnyegas SUUCESSTUN 4. . oe oe Wrartrn ASTRALIA 45 rf é> i wd re 3. 58 Sorin AUSTRALE _ _ } rs _ wa ope 60) Teuxyane Con sineienioss ae ts aS ave ee ts Be CONCLUSICN .. _ L. .- ot, ot wee OA AL RNOWLEMI MENTS a 1} ti bs + wee (OS INTRODUCTION J review of the Australiau Pre-Cambrian Succession reveals oceurrences. of granites and rocks of granitic appearance encountered over very considerable areas and apparcutly involving vast periods of geological time, With the chang- ing woderti concepls of the origin of granite and of granitic-looking rocks, all evidetice concerning the relauionships of these rocks in both time and space, both with ene atiother and with other adjacent geological foruutions, assumes a new significance, from the point of view of both historical geology and the specialized problems of petrogenesis. Tt is tame that such evidence should be brought forth once more and critically re-examined in this new light. Periodical stocktakings are always valuable, and, to the scientilic worker, can prove both salurary and encouraging, ORIGIN OF GRANITE — MODERN CONCEPTS Professor H. 11. Read in two memorable presidential addresses to the British Geaiogists Association, entitled “Meditations an Granite, Parts I and IT’ (Read 1943, 1944), has p'aced all English-speaking geologists forever deeply in his debt for a ciear presentition of the” facts of the problem of the origin of gratite and of the modern trend of thought, which are thrown into proper perspective by im exceedingly Waminaling historical background. This historical review, like all rood seiciice, hus u truly international flavour and follows the argaments and anttoverstes Of the British, Vrench and lennoskandian wasters from the ais 18th ceniury wp te the present day. Tt has probably come as a shock to many Austrahiin gculogis ts to learn that ideas very similar to the present concept ion of “granitisation” Were first put forward by the Frenchman (Ami Bond) in 1824, Ane ‘that the thesis lias beer steadfastly deve! loped andl claburated by most of the great Hrench geotogists from that vine onward. Such ideas were diatuetrically appoe ‘ed ta the principles of igneous gevlogy laid down by leading petrolegists of the German School, under whose far-reacliig influciice British and American thovigut, intortiunately, remained clouded until well into the 20th century. The nid concept of “ieneous” reek based on the fundamental thee-told classification of rocks inta Igneous, Aqueous or Sediruentary, and Metamorphic, the first rung of the ladder to which most of cur first student steps were guided, Trang, Roy. Soe, 5, Auet,, 71 (1), 23 July 1947 mn on is proving untenable in certain important respects and can no longer Sully sustain us on our upward climb. This fact has been amply demonstrated by Read in his review of French and Vennoskandian literature and sipported hy the latest reseqrehes of many workers in both Britain and America. [t has hecome increas- ingly apparent that there is a Tahdamnent: u genetic difference between the we main groups of so-called “igneous” rocks, ie., basalt and granites, and that cone trary to past beliefs an understanding of the origin, mechanies of emplacement, trend of differentiation. by-products and so on of the first group in general con- tributes very little towards ihe understanding of the same features of the latter group, W, ©) Kennedy (in Kennedy and Anderson, 1938), discussing this pr obtent, hns advanced the idea of two apparently iudep2udent expression; of “magmatic” activiry called Volcanic and Platonic Associations—the former con- sidered to be derived from a universal basaltic magina, which has originated trom the remelting of a basaltic éarth shell, the intermediate layer. The latter are considered to be derived from a primary wiiversal granodioritic parent magna which las developed by remeltiug of the so-called “pranitic’ layer within orogenic zones, where tectonic thickening at the earth's crtist has brought the base of the granite within the range of meling. Tinportatit ditterences in rhe mude of irruption in the two cases have been cited. The granite aud granodiorite barholiths appear lo penetraic slowly upwards, accompattied by a wave of granitisation aud migmatisarion of the country tocks, until arrested by some unknown form of pre- sure balance akin to hydrostatic contr) before they reach the surface. “Vhe ascetit of basaltic magma, however, is belicvedl to be directly towards the surface by wey of a system of relatively narrow dyke-like fissures with no large intercrustal reservoirs being forme. “The magnia is then erher extended as laya flows or furms injected hodies ef various sizes, such as sills or Jacecliths, which may themselves represent yolegaie reseryoirs. No large scale migmatisation or nietasomatic replacement of commlry rocks is ever performed by stich basic magmas, whose ditferentiation and subsequent evolution ave controlled largely by fractional crystallisation processes. Wirhowt necessarily conceding the validity of all of Kennedy's conclusions in their entirety, i! is apparent that there are certain maja irreecnneilable differ- chees in the mode of oecurrence of these two groups of rocks pointing to two quite distinct fomis of so-calied “igneous” activity. This is nowhere more clearly demonstrable than froemt studies of the Archaean grantic rocks throughout the world. from which it is also becoming more and more clear that the division between “Teneous” and “Metamorphic” ith the old three fold classification, fe, as far as Ketedy's “photonic aysoviation” of granitic rovks amd the hogh-grute mectano- phic rales Are Come rnd, is nwre apparent than real, ned that it haany cases the distinetton between ortho- and para-gneiss completely breaks down, J+ is Iron the observation of instances iWhusivaling this last faer ape by dedyerionts therefrom that we have reached this moderty viewpoint on the origin of pratiite. GRAN LTISATION The pith of this viewpoint can prebably best be summed up in ihe word “oranilisaiion,’ a word which aceording ta Grout (1941) has been loosely used by writers fer a number of years and whieh has pate chet defined by Read (194-4), retaf an analysis of definitions fram many sources, as “the precess by Which solid toeky ave converted to racks ol granitic obs meter without passing through a magmatie stage.” Some French writers appare ntly wee the ler felspallisaiion as synonymous with granitisation, Gramtisnction of country rock is dependent aqpon the sntrodtuction and peta sousatising action of some farn. of very active emanalions-—-zasemis, (uid or both— whose origin is ver ahsewre Those chanatins, whether vapours of solutiniys, i6 are apparently extremely active and tetuotis atid must be at least heated above the critical poimt of water (365° C. at aeproximately 200 atniospheres}, Vhe mdchamics of introduction ot these eraninsing emanations have bee discussed in considerable detail by Prench and Ienuoskausdian authorities. Apparently two chief processes are recognived—lit par lit tjection ar preferential replacenietit anda form of bodily chemical replacement througly “soaking up” or “hnbibiuion” (to use the French term) without distorden or diy sul rheseiy ve of the -countyy reek, and migration of material by what hits heen termed “oil spot? meeharism, dn this respect C. E. Wegmann (7931, 1935, 1938) has emphasise? the importauce af intergranuiar films iy the utolecular replacement and ovgratou of material and aimitur reactious. One of the most puzzling vatives of granitisation is why and how the final resint of the permeation and replacement process is apparentiy always the forma- lion of granite Or granitic rock, irrespective of the composition of the country rock attacked, This bas been exp!nined in the concept of migration and advancing “fronts,” developed by Weemann (ap. cit-), Backhmd (1938), teynolds (1943, 1944) and others. Material from the gramlising emanations is reyarded by them as selectively replacme the components of the pre-existing rock, adding here, taking away there; so as to leave natsicistang material granitic in composition, ‘Tlie displaced material 1s driven forward with outgoing emanations and fixed in an outer zone ar “Tronr.” The geochenical relationships invotyed in granitisarion have been expressed in mathematical terins by Holmes (1945) as: “Granite = pre-existing rock phty added material (A) introduced by aiid abstracted from incoming emanations (A -|- x) aims displaced iaterial (8B) driven forward with outgoing emanations (B+ x). The remarkable feature about these entanations (.A\—- x) is that (*) apparently Jeaves ne trace, fram which we may deduce that they camot represent ordinary twasma. Qualitative geuchemical studies of ine successive stages Of gramtisation in the field recently carried out hy Dr, Doris L. Reynolds on the Newry igneous Complex (1943, 1944) have demonstrated that the minimum iroductions (A) were Na, Ca and Si, and displaced matertals (B) eventually carried forward were Al, le. Me, Ik, H, Ti, P, Mn. These (predominantly basic) displaced inaterials were fotind to have been fixed in an aureole of “igneous looking’ basic ancl wltra- busie tucks, and it is cone uted from these stwdies that before any givei mass of comity reck was granitised ir passed through a preltuiuary stage of basification, In 2 very recent paper (146) (abstract only availabie at tiny of writin) Dr, Reyrolts has evtarved om this subject and cemonsirated that the eeochemical changes leading lo pranitisation in rocky of af) iypes invariably includes an iniial entieluucnt in femic constituents and alkatis (hasifeation,” “dest: ann fol- lowed hy @ stage of felspathisation or gramitisatiant proper. Details of the stages in mielisnmatic alteration ef petitic, seni-pelec, peammitic and hasic 1eHLous Waehes Ape eIVven, prev iding valuable criteria Ter the recogmilion of the provenance Of metal tftis ov “enclaves” in eennite. 1) vonsidering the ariel grauite perhaps the most satisfying feature of the “oplacement metamorph: sm” " hypothesis is im connection with the space jwollon, particularly as it as ers le enormotis regionally «neissuse granitic masses Of tae Archaean sivells and ancient continental blocks turned the muwypor hatholitlis usually torming cores tu folded ranges uf periods later than Archaen- vole. In both these cases, if the grunite is cons sidered to he strictly intrusive, that fe. viel to the pre-existing rocks, then the sprace factor renders this view tuiluindde. There is no evidence Of Gisplacemnent of country rocks on even the sintlles, fraction of The seale required, nor has there yet been imagined any mechamen of doming, subsidence, block faulting, stoping, ete., that conid acco ard jor the entrance of such volumes of usagima tovelyed. ATL the evidence, om the uther had, pomis Lo replacement with little or no bulk changes in volume. ITow- ever, us far as minor granitic intrusipns stich as dykes, silly and veins of any age are concerned, the actual injection of inuterial as liquid, ie, magma, tule places of weakness would appear to be an eqitally losical explanation. Tinully, attention may be drawi to the division of granites into the three catesories chtimerated ubove, viz.: (a) we Ay he ba s4 - &9 ReEreRnNcEs .. .. = a “ iy 4 89 Sovrers or PXroamMarion 4 a Hy ny 24 oe 3 89 INPRODUCTION Apert from its native grasslands and forests. the continent of Australia originally offered so little in the way of plants of economic value to man that plant introduction has always heen of great importance in the development of the country. The botanic gardens, early established in the capital cities of the colonies and at Darwin, were originally essentially acclimatisation centres and testing grounds for introduced species of plants, althongh they have generally lost their importance in this regard and this function has been repiaced by the plant mtroduc- tion activities of the several State Departments of Agriculture and the Council for Seientific and Industrial Research. ©) Director, Waite Agricultural Research Institute, ©) Formerly Inspector-General ot Forests and Acting Principal Australian Forestry School. Trans, Roy. Sec. S, Austs, 71 (1), 25 July 1947 E 68 In the early years of colonisation there nist have been extensive trial and error in these activities, but the exigencies of quarantine and the exhaustion of the more obvious introductions have made it increasingly necessary to place plant introduction oa more logical footing. Some highly desirable tmtroductions such as the sony bean have proved exe cedingly dificult, while there is no dotibt that athers have failed through a lack of knowledge of suil and clintatic requirements, In 2 reeent discussion on ecology anc the study af chinate reported in Nature by Day) (1946) emphasis was p’aced on the ‘ion for the growth-cycles of introduced spesics ta be in phase with the annual seasonal cycle, “Two cxarmples ave given of such lack of harmony in Beftain. The European larch begins its growth trmch too early, while Lhe Corsican pine continues growth utiti) itch too Tate: Jeeattse of this they suffer seriously from frost im many situations, This tends markedly to restrict the arcas within whieh they cau, be grown stieeesstully, It is the puipose of the present contr “pation to review the climatic reyuire- ments of @ mtibey a species of the genus Pinus which thrive in the shinee rancan tegion or in regios having a sivailer CANE sueh as See eee aud t determine kow far such climates are reproduced in southern Anstraha- In the light of actual experience with these speciés in Australia, the abi ss imay re aye! some general prine’ples which may be of service in such plant introdnction work, The species chosen are: Pinus radiata D. Don, the Moncerey pine, from California. Pinus catariasis C. Smith, the Canary [stud pine. Paes i itlasier Aiton, the clister pine, of southern France, Portugal and Pinas vigra Arnold ov Pinus lericio Poirot. the black or Corsican piac, also of the Mediterranean region, Pinus falepensis Miller, the Meppo pine of the eastern Medirgrrancun. he Monterey and Canary Island pines hays very restricted mative havitals. so that the climatic conditions of these environments can be very narrowly cetued, The other pines ave more widely distribases, and the associated climates are there- fore ms quite so easily cerermmed, CLIMATIC CONSIRE RATIONS lo dawing with perennial plats, 1emperuure ar! anoisture condiiuys throughout the year are of inpertinee, “Poterance to low or high temperatures aw be import. and greater erupaasis has therefore buen placec in this siudy on temperate, Rainfal is rot, however, cotip-otely overlouked In dealing with tauperatures, use has been nade of the convene method af wave fourm gnalesis adepred by Preseatt (12), whieh cuabies the iwelve mean monthhy temperdicures Lo be reduced to Use tlires values oF annuai mean, amphtinie BESCRIPTIGCN Gr FIG. 1 Maps of the Mediter (a) Natural distvibution of flying (h) Mean annual temperature, (ce) Mesn anoual amplitude of temperatirre. Py wieder, Fo ape. fd) Temperattre chase expressed in ters of Ing fv days behind solar raliation, (0 Sea alsa Vay (1845). i Ae] Peanariensis .~_ | ROR Apinaster. | EY Phalenensis. MEAN ANNUAL | TEMPEAATUAR "Fr-— TEMPERATURE | ; AMPLITUDE “F.f-~rg_! jt Wh, eet 0 id re a ‘ ' ‘ ' " Jf Big, 2 Map of the Mediterranean region, showieg mean annual rainfall in inches based on data of Kendrew (1927), Brooks (1932), and Terhertsoh (101). and phase. It is assumed that there is some correlation between daily ampitudes and annual amplitudes. Of imporrange in the ease GF moisture conditiuns are the total amount of effective or influemial rainfall and its distribution in time. In the Mediterranean climate: under consideration, the rainfall has a marked winter incidence and the ienath of the rainy season becomes an important index. To secure identical combinations of all these factors is wot always an easy matter, but it 15 essential to secure identical combinatians of at least mean autual temperature and ampli- tude with the appropriate seasonal rainfall One dilfculty 1s to assess the con- tribution of coud and fog to moisture supply. This ts particularly so in the case of the Canary Island pine and the Monterey pine. it is probable that mm these cascs, in the Australian environment, somewhat higher rainfall should be allowed than is recorded for the native habitats. Mazis of the Mediterraucan region, iiustrating the geographical distribution of four of these pines, tugether with temperature characteristics and annual rain- fall, are given in fig, 1 and 2. Pinus rApara (Monterey pine) Pinus radiate is a native of California and shares with Cupressus macracarpa the very restricted habitat of the Monrerey peninsula. Leimg essentially a costal species nccurring near sea level the temperature conditions can be readily ascer- tained without iterpolatian, and the raimfail conditions are simularly reasonably well known. The other habitats of this pine ate Ano Nucyo Point, San Simeon say (Cambria), two of the Santa Barbara Islands and the Island of Guadalupe off the enast of Southern California, The climatic conditions on these islands have nut been ascertained, bttt they are likely to include low amplitudes and a late phase. Atterition has necessarily been restricted to fhe occurrences on the inainiand, lo Table [ are given the essential data Tor Californian localities near the zone of natural occurrence of this pme. Santa Barbara and San Luis Obispo have late phases associated prestimably wiih greater oreanic influences. All the ainphindes arc quite low. The annual rainfall on the Monterey peninsula has been estimated hy Mason (1934) to range from 188 inches at Carmel to 23°7 inches on Htckle- berry Hill, with almost datly summer fos. 71 TABLE | Temperature and Rainfall Characteristics of Locavilies in ar near the Calilornian Zone of Occurrence of Pinus radinia Temperaturé Characteristics Rainfall Characteristics Phase, leg lel nd solar Length () Anntigl Mean Amplitude radiation Annual Seasonal of season Locality iad oF: days ins, ins. qmonths San Luis Obispo S8°5 6-2 49 21-9 21 6-7 Santa Barbara * on 398 0-3 50 18-9 18 6-7 Santa Cruz Ba ai, a +3 fed 36 270 25 7-8 Del Monte _.. ct . 56-4 O-0 35 15-4 14 5-6 For practical purposes, therefore, may be sought zones in southern Australia having the following temperature characteristics: Mean annual temperature .. reat pits .. 56-60° F. Amplitude 5+5-7-0° F, Phase... bigs vue ese ite dave a. 38-50-days Parallel zones (fig. 3 and 4) are found to occur in the sotith-west of \Weszern Australia atid in the “coastal regions of South Anstratia, Victoria and northeri Tasmania. Iangaroo Island al the islands in Bass Strait also come within the zone. The climate is thus essentially maritime, Rainfall conditions within this zone are, however, anuch more favourabie than in most of the Californian localities, and it tuay be assumed that adjacent zones with anual rainialls of 25 to 39 inches may come within the fayour able zone , providing soil conditions are suitable. tt is known from Australian evperience that this species is rather exact- ing im its soil requirements, and througheeart this discussion il must be bere im mind thet edaphic seideth are deliberately excluded from consideration. It is of interest to list Australian stations which offer pavalicls to these Californian stations. TABLE 2 Temperature ond Rainfall Characteristics of Australian Localities haying ‘Temperature Reaimes similar to those of Californian Stations in of near the Habitat of Pies radfeta. Mea ania Mean annuat Length of fempersuture Amplitude Phase rainfall wet seusun ‘Leealily er, oF, days ins. months. Eddystone Point, Tas. = 2 366 fy-3 42 29-4 12:0 Curme, King Island, Tas. |. ; 55-7 3-3 A7 34° 12-0 Albany, W. Aust... = 599 O35 45 37+2 11-4 Eclipse Island, W. Aust. .... we 8985 af h2 32-7 2-0 Karridate, W. Aust. sees wee SOR amie) 44] 478 - 98 Cape Borda, §. Aust. hae we 5876 fit AL 2463 vas) Kingscote, S, Aust. ... . 80-4 6-6 41 Wed rere! Cape Northumberland, S. Aust... 30°9 td 38 2045 6 Robe, S. Aust. vee O49 6-5 37 24-7 9-3 Mount Burr Forest. 8. Ast: 56+5 a7 56 30-7 9-4 () ‘Lhe mean length of season in this and the following tables is based of inean monthly values ior rainiall and probable evaporation. It would be preferalsie to use the meat of the lengths of individual seasons, but this iafovmation was avaiable in erly a ad cases. ‘The estimates are likely to be soumewhat luo high. perlcularly for the lighter values, Tt will be seen that the South Aus- tralian stations approach most closely the Californian ones in respect both of temperature and rainfall. The close parallel herween Kingscote, Kangaroo Island and Santa Barbara may be noted as one example. Vhe Atistralian stations have also longer rainfall seasons and a greater certainly of stunmer rainfall. This may well com- petsate for the absence of stmiuer fogs in tie Astralan environment. Pinus CANARIENSIS (Canary Island pine) This pine, like the Monterey pine, oeewtrs in a rather restricted zone, but ag titude plava a part in the de- termination of its habitat, it has proved ligcessury to interpolate same of the climatic information available in the standard textbooks of climatology and the report of Byles (1932). It has heen assumed that this pine finds tis optimum, temperature conditions al altitudes between 2,000 feet amd 6,000 Teet, with 4,000 feer as a probable ideal. Although the coastal stations of the Canary Islands have a mari- thie climate similar to that of coastal Calivarnia, there appears to he a sig- nificant increase in the annual tem- peratire rauge with increasing height above sea level. No part of Austraha shows the sharp changes 1n topography characteristic of the Canary Islands. ivies assumes a probable annual rain- tall of between 12 and 30 inches. ‘he lensth of the raitty season cannat be precis’y deteymined from the avail- alike data, but it appears to be in the reigibourhood of eight nionths, with a vreuter certainty than in s\ustralia rain in the = sunimer The temperature data which are appraptiate to the dtscussion are Table 3. absentee of mcutihs, sivesit li P. canariensis. Tig. 3 Meps of the southowest part of Australia, illustrating localities Where temperature con- ditions are similar with respect to mean, amplitade and phase to those of the zones of native occurrence of P. vadiata and P. canaviensis and the Portuguese habitat of P. pitasier, Whe numbered lines are isohyets of rainfall in inches. The line of circles parallels temperature conditions with respect to appropriate combinations of mean aid ainphtude in the ease of P. cenariensis. 73 TABLE 3 Tempe-ature Characteristics in the Neighbourhood of the Zone of native Occurrence nt Pins cannricusis, Leight above sea level Aunnal mean feet oF, Santa Cruz. ‘Venerifte ; a 130 67 +4 [uerto de Orotova, Teneriffe sn 70 67°3 Guiruar, Tenerife an} an 366 64-3 (2) Vilaflor, Teneriffe —.. PF we F400. 54°5 Las Canadas, Teneritic on 2 6,357 48-9 Las Palmas, Grand Canary . 30 67°8 Funchal, Madeira... . 80 65-0 Agadir, Morocco cess _ . — h3°2 Mogodor, Moraceo _... : 4 30 03°7 Phase, lag behind Amplitude solar radiation co F, days G3 53 cg) 57 75 63 9-8 41 11-4 32 54 59 6+3 a) 7:5 56 5+6 40 From the data available it may be assumed that the temperature requirements of Pinus canariensis approximate to the following conditions on Teneriffe: Mean Amplttnde Phase TF wr, days At 2,000 feet ang chs . 606 Bg 53 At 4,000 feet es . wae 8B 9-€ 44 At 6,000 feet 6 ext ... “SL0 11-C 36 P.radiata. Fig. 4 Map of south-eastern Australia, illistrating localities where teniperaiure cohditions are similar with respeet to mean, amplitide an phase to those prevailing in the California home of P. radiata. ‘Lue areas are shaded and include Kangaroo Island, the islands of Bass Strait, certain coastal repions af South Australia and Victoria and the north-cast corner of Tasmania. ‘Vhe numbered lines are the isohyets with antiual rainfall expressed in inches, Tt is possible to obtain appropriate combinations of mean annual empcrature and amplitude over the whole range. and these cambinations haye been plotted on the map of Australia and are illustrated in the maps of fig. 3 and 5, The nearest approach fo similar conditions occurs near the coast at the extreme western and eastern geographical linnts in Anstralia. Stirling West in South Australia, Terang in Vicloria, and |aun- eeston in Tasinania, closely approach in some respects the mea climatic habitat. Some selected yaltes ot this kind are given in ‘Vablé 4. wm () Calculated from partly interpolated data. 74 TAnLe 4 Temperature and Rainfall Characteristics of Australian Localitica showing some Features similar to those of the Zone of mative Occurrence of the Canary Island Pine. Temperature Phage, lag Annual hehind salar Annu Length of meat Amplitude radistion — rainiall wet season oR °F days ticles months Cape Nuturaliste, W. Aust. _. 61:3 6+7 $1 33*4 8-0 Busseiton, W. Aust. ..., _ _. 60-6 B85 39 32°0 §-0 Bombala, N.S.W. = i. 27 11-0 28 238 12-0 Gabo Island, Vict. . fu wa 591 69 45 37°3 12-0 Launceston, Tas. aye ew . §4-9 Q-4 3l 2871 12-() Terenyg, Vict. .... es ; _. 459-5 8-5 36 28°9 12-0 Stirling West, & Aust. , cae DOT G3 ot 46-8 11-4 Tn most of these cases rainfall may well be excessive and more favourable conditions may possibly be found in semewhat crier zones adjacent Lo the appro- priate température Zone, PINUS TINASTER (Clister pine) This pine occurs naturally and is cultivated successfully on the Attantic coasts of southern France and Portugal. Certain strains are characteristic of Mediterranean France. Spain, Italy and Corsiea, ‘The species also occurs in Morocco, in fairly humid environments in the mountains up ta a height of 7,009 feet, mainly im the middle Atlas and in certain localities in the Rif, It is rare in the Grand Atlas. It docs not generally occur east of lialy or Tunisia, but is said to occur on the coastal islands of Dalmatia, ‘The African occurrences are not regartied as jumportant, but they afford evidence at least o7 the climatic range of the species. On the whole, the cluster pine therefore belongs essentiatiy to the Western Mediterranan. It is regarded as beine much less drought resistant than the Aleppo pine, Pais halepensis, wuich is much more widely distributed but which does not occur im Portugal, On the basis of its distribution two races are recognised: maria and mesoguensis, he former race belongs to the Allantic coast, the second to the Mediterrancun regions and probably also ta Moreceo and Algeria. The Moroccan strain is said to be more tolerant of calcareous soils than are the principal races. The temperature and rainfall characteristics of typical localities are given in Table 5. The wettest month in these regions is October; antiwin and not winter is the wettest season. TABLE 5 Temperature and Rainfall Characteristics of Localities it or near the Zone of mative Occtrrence of Pinus pinaster, Temperature Phase, lay Annual behind solar Ati Ociaher Length of mean Ampliwde radiation rainfall rainfall rainy season Laorality or Ps days inches inches months Atlanuie Enyironment--— Bordeaux, Trance 54-1 13-9 3 307 36 12 Ayeachon, France 5567 13:7 al 34-1 4-5 2 Biarritz, France ... 507 115 37 — _— _— Lisbon, Portugal 59-3 10-3 38 29-7 34] 9 Oporto, Puartugal 57-4 99 35 = om, = Coimbra, Portigal 58-5 10+5 34 —/ — — Santiago, Spain 54°69 10-0 37 62-0 6-7 12 Meditertanean Environment— Var, France wee DOCS 13-8 35 39-3 6-4 il Nice, Franee thls .. 59-4 13-8 35 30-9 5°S ia Toulon, France ... ae a8-9 14-2 33 29-1 4-6 10 Marseiiles, France an 3688 14-4 32 22-6 4-0 9 Tle dé Levant, France .... 58-6 13:2 34 - — — Montpellier, France ... 56-1 15+8 30 30-9 4-1 11 Can Corse, Corsica a BORD 13-1 39 27-0 4-2 — Can Pertusata, Corsica... 60°6 12-4 42 22-4 2-9 —= Jles Sanguinaires, Corsica 61+7 13-0) 45 216 3-0 — Ajaccio, Corsica... G14 12-3 39 26:4 3+8 9 Bastelica, Corsica oo — — 466 44 11 Corte, Corsica... lea — — 34+5 3:8 16 Genoa, Italy vs a, ae 15-0 35 52-0 7+8 12 Sessari, Sardinia wa =5O9 4-6 37 = 3-4 — In Corsiva, the zone of 7. pinaster occurs ai altitudes of from 1,390 to 3,500 feel, when it gives place to P, lericio (P. migra). It is probab'e that originally 7’, pinaster: extended to sea level. Corte is uear the average for the Corsican Habitat with an annual + raiifall of 34-5 inches, a minimum monthly rain- fall of O-5 inches in July and a maximum incnthly ra: infail of 5:6 inches in November. An estimate of conditions in Morocco, near the north “Aftican Hinit of its occurrence, is possilile from data available for Dayet Ashlef. Estimates for the temperature characteristics of the Corgicaii and Moroczan habitats are given in Table 6. It is diffienlt to predict the correlation between amplitude and phase and altitude without some ioral data, but the data for Sessari 1 Sardinia at an altitude of 730 feet is of some assistance in the case af Carsica, TARLE 6 Ustimates of Temperature Characteristics of Corsican and Moroccan Habitats of Poays prvester. Catsiva Moroveo Sea fevel Monn range Upper limit Mean xnnual tenmperature Gl’ PF, 5a° Re, 51° FL 33° F, Amplitude - - - 13° F, 15° F, 16° FB, 16° F. Phase - - - 41 days 35 days 33 days 40 days ‘Che fotlowing general limits may he suggested for the temperature cherac- teristics of the zone of P. piiasler: Mean annual temperacire te, yes F. Aniphtude , an ee vile -15° P Phase. Se a 3Y dayg It is possible further to separate the eerie aba Mediterranean zone from ihe Portuguese zone in order to define the temperature couditions more narrewly. Vor the Mediterranean zone these 1 wnay he roughly dened by tive following limits and their linear interpolations: Mean annuzl temperature 54° Ty 61? F. Aniplitude “i i — 14°F. 13° F. Phase cue _ . 80 dars 41 days For the Portuguese zone the conditions can be more narrowly defined : Mean annual temperahire - — 39° Fb, Amplitude oe sii re aw 10° R, Phase Ait . ne me . 86 days ‘Thess latter conclitions, with mxlactuies annual rainfall, are closely reproduced by conditions at Collie, Bridgetown and Donnybrook in Western Austraiia, rela- 76 tively near to the coastal plantations of this species of the State lorest service. There iy an approach to these conditions m eastern Gippsland and the adjacent coastal regions of New South Wales, The relevant data for these Australian stations are given in Table 7, where the inetith of April corresponds ia October in the northern hemisphere, Tanne 7 Climatic Data for some Australian Stations cortespendive in Temperature Conditions to Portuguese Hahitats for P. pinaster. Temperawiure Riki tall Mean Wettest Length aunual Amplitude Phase Annual April month of s¢asun Loeality cae OF Pit: days Inches inches iiehes miunths Weatern Australia— Collie ten . O94 10-4 435 39-4 2:0 7-5 5-0 Donnybrook 60-6 8 39 41-1 17 73 8-2 Bridgetown ven BBA7 a7 37 34-2 1-7 59 81 Victoria and New Sont h Wales— Gabo Island _. 591 (9 45 37-3 3-1 4-5 12-() Mallra wa B73 §-3 3 23-5 1-4 2:3 12-0 Sale a au BPS Oey 31 23°9 1-7 2:5 12-0 Bega AR ve alhD 10-2 2Y 33+3 2-2 3-9 12-0 Bodalla : zat ABS, as 33 36-1 2-8 4-0 12-0 Moruya Heads ., 60-8 8-3 36 34-9 31 +0 12-0) The raintall cistri- bittion in the eastern Austtevan stations camo be said to approsch at all the conditions in Portugal, notably in the nbsciice of a true drought perigd in summer. Parallel conditions are therefore restricted to \Western Australia. For the Mediter- yanean zomtes there is ray wider range of parallel temperatiure conditions, proyreliig only mean ard ampii- td talcet1 imta coisheration, There is At approach to equiva emt phase con- ditiens in northern P_conariensis Viclovia and on the — western tide of the Fig. 5 Souther tableland of Map of south ea Bipin Australia, ilastratine localities where teniperalure conditions with respect to meau, atnplitude New Fateh Y ales, of and phase are sinilar to those prevailing in the habitat of whieh Purrvinjuek and P, catericisis in the Canary Islands. The arcas are shaded Acvlate n ray be taken and include areas in eastern Victoria and the south-east of fe rentesentative, The New South Wales. The line of circles indicates parallel ae cea Te conditions with respect to mean and amplitude only, The approoriate esmbina- numbered dines ure isohyets with annual rainiall expreesed toms Of mean and in inches. 7 amplitude in temperature oceur from the north of New Fngland in an arc reach- ing as far as Rutherglen in Vietoria, The conditions at Adclong ahd Burrinjuck are given helow in Table 8. Tarte & Climatic Data [or two Staliows in New South Wales corresponding in Température Conditions tu the Mediterranean Habitat of £. pinaster. Temperature Rainfall Mean Wettest Length Lopaiity wefiual Amp itude hase Annitat Anvil muuth of season Adelong _ oy oD 13-3 3 29-3 1:9 4-0 {2-0 Surrinjuct a B80 13-0 Ai) 33+K 2] 4-3 12-0 The conditions of evaporation at both these centres is such that near-drought conditions prevail during three summer months. It is probable that appropriate conditions occur near these localities at higher elevations. The information con- cerning the climatic conditions under discussion are illustrated in fig. 3 and 6, Pinus NIGHA (Black pine, Corsican pine, Austrian pine) The climatic condt tions under which the black pine grows. 1n its natural habitats are not so easily defined owing to the paucity of crect meteoralogi- cal data, associated with the fact that these habitats are at some elevation rang- ing from approxi- mately 600 feet in southern France and Australia to 6,000 feet in Cyprus and Ana- toha. No climatic data have actually been se- cured relating directly ta one of the known liabitats of this pine, but rainfall reqire- ments are generally P pinaster. Fig. 6 Map of south-eastern Australia, jilastrating localities where teniperatute conditions with respect to mean, amplitude : . and phase ate similar to those prévailing in the native regarded as high, in hahitats of P. pinaster. ‘These areas are shaded. The area the neighbourhood of macked M is just worth-cast of Canberra and reproduces 50 tncties per annum. the temperature conditions of the Mediterranean habitats. * f The arca macked P, near Adelaide, reproduces the Portu- Tre main centres of guese conditions more closely. The lines of circles distribuiton are wide- indicate areas where the Portmguesc conditions ure spread but obviously parallcled with respect. ouly ta mivan and. amplitude, The localised, They include areas enclosed in heavily dotted lines indicate aveas where i fn : the Mediterranean conditions are similarly reprodtced, southern France, ut The numbered lines are isohyeéts cf meat annual rainfall cluding the Cevennes expressed in inches, and Pyrenees ; Corsica, 78 Sicily and Calabria, Austria and Dalmatia, and the region of the Taurus moun- tains, incl ding Cyprus and localities in Syria, The species also oceurs in the Crimea and itr the Caucastis region and in the Balkans generally. Ai isolated occurrence in the Rif of Morocco at 5,000 feet links Africa with the Spanish habitat. In order to secure an approximation to the temperature conditions of these habitats, correlations were graphically established between altitude, mean tem- perature, phase and amplitude. Phase tends to be late both at sea level and at very high altitudes, and to be generally correlated with amplitude for geographi- cally grouped localities. On the basis of these correlations the following estimates of temperature conditions appropriate to Pinus nigra have been obtained : TABLE 9 Temperature Characteristics associated with Habitats of P. nigra. Mean Atuplitude Phase oR oF, days Mean of all estimates si uke ae Nolopale Taujekalo in the lewer Murrey in the lower Murray whieh had developed a keel-as a resuit of a suilden increase ji temperature aud/or salinity (Cotton 1935), and (4) the evidence Lor warmer seas in very recent times in South Ausiralia, and the suggestions that these may have been coincident with the period of greater aridity (Crocker 1946), Acklitiunal evidence is to We had ja the oceurrence of extenstye sypsilm deposits, as those ia Ue upper South-Last and near Menimeie (Jack 1921), and gypsum in buried profiles neur Dinduleer (Siephens ct af 1945) where coluliticns are tou nmi for thete tormacion ditaws and is ulso provided hy the necurrcuce of Inssil caleareeus soll havizons wider climatic conditions i which Time prt acermiation would not he expected to orenur we present. Sieh fos) TD horizons are to he louie in che “travertine” of the fordertown district, tin Western Austr: sqianeté sails occurring aw Roeky Cully, present manual rainfall over 30 inches, padiguie gromeér aridity (Stephens 1946) than at prescin. A cousiderable body of independent card, tach of whieh is of a pedelogical nature, has been advanced theretore in suggest greater avidity in the past, and most of it appears “yalid.’ On the other land, some of the evidence as the fact that the cune sysiems are aw vegetated (Mills 1939, Crocker 1941) is of little or no valne m ivself, iyidence wi reduction of the elective arca al “active type" duties mm south-western Oneensland, as advanced by Whitehouse (1940) and aecepled by Browne (1943), is alsa of dunbtful value, for the “eal? low, conipletcly-vegemted dunes which occur as outhors to che Simpeon Desert eit be matched by similar low yesetated dines within the desert proper. Wille no suggestions of the present being wore arid than the peried Of masxhvgu dune lnutdinzes have been advanced, it las beew stated ihat certain buried profi’es, as those ia the alluviuun areas of the Noarlunga-Sellicks district of Soitth Australia, ey inchieate preacer podsolisation than the super-impesed present suil aud miter fying red soils (Crocker 1946). Ji is very difienl: to assess the age ui these profiles. however. They conld incicnie, though at presenr ip is nectesary to proceed cautiously, thal Here has heen a yery pevent decline following post-artd wetter conditions, Whitehouse (1940) tends to such a conclusion ow the abundance of aboriginal artifaets in the far west uf Queensland, in regions where there js iow to peruavent water. While further evidence is still desiruble on some aspects, it would seen that (1) preceding the onset of aridity climatic conditions were itch more humic than at present; (2) subsequerit to the maxiniuwim aridity there was an increase in ruin- fall, though not of a high order, and indeed very snail by comparison with the preceding deerease—orherwise gypsum and He accunmtifations of the driver period wonld Eave been entirely removed i solution; (3) there muy have heen a yery recent minGr decline in rainfall, but considerably mute paleopedolowicid resuprel will be needed to further elucidate this. by short, climatic conditions today do not Ciffer wreath: (rent those when dine-building was ar is maximum, but rainfall is upparcnthy sightly higher. The discussion eq far has been concerted chiefly with rainfall as an index of clinate, Teniperatire effects are much more difficult to deline. It is apparent that the temperatures of the late Pleistocene, which coivided with the lust preat glawation, were low, so that in addition ia high rainfall precediig tke initigdon ef aridity, we can deduce colder conditions. The extinction in southerts Atistralia ot a suite of surface-dwelling narine species doimmnated by Avadara lrapesia, which are uow to be found in many locaitics alone the southern coast as sub-lossils. is of parricular interest. These species are sill living in mere northern (warner) waters. and it je considered ther very recent extinetion was die to a sudden decline in temperature, Lt is 18S also certain that the suite was present in southern waters for a very limited time. The suggestion has been mude that the briet stay of these species was coincidental with and consequent upon the high Lemperatures associated with aridity (Crocker 1946), It has not been suggested at what stage in the development of the desicca- ton madara and its associates became established along the southern Australian coast, hut it is likely that they persisted beyond the period of maxinnun cdiunc- building (Crocker arid Cotton 1946). There is strong evidence, therefore, to indicate that higher temperatures than those prevailing at present oceurred in Suuth Australia a very short time ago, and these were probably coincident with the lower rainfall of the arid period, Fig, 3 Tie recorded distribution of dune sysiems and dune sheets in Australia, (Chiefly aiter Madigan, Hills, Prescott and Crocler.) 3. GewntraL Errects or Agiwiry Loss of soil stability in South Australia and surrounding regions ov a erand scale resulted from the sudden modification of clirnate, and it led to the build-up of the extensive systems of avolinn deposits, dunes, dime sucets, etc., at most Of cur major and lesser ergs (fig, 3), Stich widesprea:dl wind erosion could onty have occurred following the wholesale destreetion of the local flora, It is apparent that for this to have happetied the aridily must have beot wot only extremely severe, but its onset must have been particularly sudden. Orierwise the vegetation would have been able to nraintain soil stability by simpie migration, It was obviously unable to do this. Jt is interesting, to picture what probably happened. A catastrophic’? decline in. rainfall which initiated the aridity placed such a stress on the pre- arid flora thet over the greater part of the State and beyond it was almost com- pletely wiped out. Aw carly replacement of rhis ofl flura by a xeromarpnic ome capable of withstanding the aridity, that is capable of migration anc ecesis under the extreme conditions, was not possible, This was chiefly because the most O) From the biological viewpoint. 144 desiccated region today, as ior exarnple, the Lake Myre Basin, were, prior to ahe anset of aridity, so moist that centres for the rapid dispersal of the diaspores of the required arid element mast have been extremely few, and perhaps alinost non- existent. As the remuants of the more humid Hora disappeared and disintegrated under the extreme desiccation, both wind and water erasion were greatly accelerated, The eroding power of occasional heavy rains, which no doube still oceurrerl, but with probabiy greatly reduced frequency. was very high and carried considerable quantities of silt and sand into the drainage basins. ‘Uhis, in the absence of a protective vegelative cover, was transported and winnowed by the winds to initiate dune systems, In other parts the fluviatile agencies pluyed a sma'ler and snialler part, until finally, at the other extreme, dune systems of a purely deflationary origin on which fluviatile agencies were uniniportant became piled up by wind (Crocker 1946) playing on soils which had completely lost their stability. This loss of soil stability was not general (sve fig, 4), nor did it proceed Wy the same degree or with the same lapidity everywhere. Certain soils were Anich more linhle to complete loss ol stabihty than others, depending, no doubt, Upon a comiplex of factors. In South Australia soils which were derived from highly ecaleareous parent material or which contained large quanhues of Lime in the profile were purticularly tngtable, As a result the A horizons of the soils developed on the line Peisiacene calcareous dunes, those of our present mallee regions, Which lad received large quantities of coleimmt as loess, and those derives from calcareous rocks, as the Tertiary Jimestanes af the Nillarbor Plains, were largely stripped, transported and resorted to fort ergs. Typical samples are the Nutarbor-Spencer , 1 ; 5 dune system, and those af the Sonth- Portions of Soitth Australia where loss of sail ; aha 3 , stability was cither general or frequent during Fast wu) “mallee” regions. the: inid-Kecent. aridity, LL ods possible that strong winds greatly assisted the dune-huilding, bil such a possibility may be dismissed [or wanl of any evidence one way or the uther. Althou vl uollung can be said of the Wtensity of arid winds, the general wind pattern was apiparcutly very simular to thal today, hecanse the dune systews throughout Australia, which are entirely or whnust entirely of the sandridge or se/f iype, show a consistent orientation willt the wid regimes of the present (Madigan 1936, ilills 1939). One consequence of the fact that soils varied in their erodibility and that possibilities for the maintenance of vegetative cover, despite the adverse climatic trends, varied in different localities (see later), was that the dunes were not all iniljated at the same ume. Por example, although resultant from the same arid peril, the time of dune-huilding initiation and maxima may not have coincided exactly in different localities, The Murray mallee systems may have been slightly im advance Gr behind thage, say, in the lower South-Tast. Fig. 4 4, CONTRACTION oF THE FLozA, awy Surytyar Poct Although over the greater part of South Australia and the adjoining hititer- land regions the old vegetation was largely destroyed and countless species must 105 have been entirely extinguished, remnants of the pre-arid flora managed Lo survive the desiccation in especially favoured situations. The pre-arid topography was alniost iWentical with that of today, except for the purely superficial sand deposits, Jt is obvious, therefore, that the rainfall gradients were similarly zoned. This pattern is determined chiefly by lautude and distance from the coast, but is modified by other factors, most important of which are the influences of mountain ratiges and hills. Lt is apparent that when the severe climate stresses associafed with the siden onset of desiccation were imposed, the mesic flora’s sole defence was a migration towards wetter conditions. AUSTRALIA Fig. 3 Principal refuges during the Great Australian Arid Period. Successinl migrations would oniy be possible where the rainfall gradients were steep—that is where the distances between climatic horizans were small and where the propagule dispersion capacity, the establishment capacity and the ecesis capacity of the species were rapid erough. Reference to the present rainfall distribution, and comparison with a topo- graphic map, indicate quite clearly that the possibilities for successful migration were very limited and could occur only adjacent to the maser ranges. The ranges were irnportant regions of survival for another reason even where their influence on rainfall was not so great, for within thew occurred the greatest diversity of microclimates. In the ranges and hills were the larges- number of ecological 106 niches, the greatest variation in habitats, to be offered the retreating vegetation. They were, therefore, the principal refuyes af the relic flora which survived the stress Hf aridity, Other refuges of less impor ance were the major drainate lines, any of the lakes which, although reduced in area, did not entirely dry up, and other centres more humid than surrounding areas, The principal refuges of the arid periad ix South Australia and adjacent regions are shown on the aceainpanying map (fig, 5). The most nriportant were, no doubt, those of the Mount Putty -Flinders Range system. Others of note were Kaigarou Island, the ratiges of easter Eyre Penirisu! a, the Gawler Ranges, fhe Muscraves, the Macdonnell-James Rane system (Central Anst_.), the Gram- pians ( Vict.) aud the Murray drainage system, Atihough these were the main centres of survival, the vegetative cover within them was very restricted, This is evidence?! by the preseuee of eliubing dunes in the Miltalic region of Eyre Peninsw. These duties are wutliers of, but more or less cantimuos with, those of the Nullarbor-Spencer Gune system, and cover portion ot the ranges of central County Jervois. Similar evideree is provided by the aeolian sands of portion of the Mount Loity Range system, as for example, those that are su eonspievtous in parts of the Bremer Valley. V. DEVELOPMENT OF PRESENT COMMUNITIES The muimerots ecological niches which acted as centres of survival were also the foval points from which commenced the migrations leatite ta the establish- ment of the present-day communities. No dontit the climatic duress associated wilh the desecation resulted in reat modifications in the flora. In all likelihood the isolation and continued sires of existing under conditions anoreaching or at the Tintits of tolerance tof most species, fayertted a large number of sib-species atudstruns relatively unitportant previously, and restilted tn endemism of varying deerecs, It is interesting to trace the major aspects of this development and to atiompi a defnitivn of seme of the migratory routes and the orinciples involved. 1. Micration ann REcCoOLONISATIGN Mtthongh the eeverity of the andi rested in such a rastriction in aren nf species en the destruction of the pre-arid veeetation on a grand senle, it fs probable that a claw recolonisation was besun hy a few droveht resistant and perhaps some new genotypes. before any iupravement in rainfall took place, Any sich yieration, hawevor, must Baye heen slow and limited to a few previously unimpartant species with pour disired| cupacitics, for atherwise the Ines af gal stability sie mot have been em widesntead nr Nave proces ted so far. le is obvious that the greatest number of species surviving in the relatively few refuges were far from heing under optim conditions and conmetition for the available ecological habitats was freat. Jt was not wt] an intmrovement in chumtie cun- ditims. which prohably included an increased rainfall and slivhtty Thwer tem- puratures, resnlied 3 in a relouse of hath the preseure of competition atid the jres- wate Of an acleeree climate that re-rolenisetion could) be very effective. The speed at which re rolanisatien proc veded uid the pattern it folirawed won'd depend an many fagtars, tn tle titimate analysis if wonbl he governed hy the ability of the individual cpccins ans is s hiotypes to produce and disseminate propaglas. The sped with which species could exploit the oppertiupities of an extenlel elimatic habitat range wold depend, therefore, on their especial inechbatisias tor dspersal. Althnush most plant seogranhers believe thet Tong distance dispersal ts of infrequent oseyrrence awing to the effective harrier of éestallished comiiainities, such a deterrent would not be ppurative tm this case heennse the area being colonised was viritally a bare ote, Tt is likely, therefore, 17 that will the relaxation af climatic stress the species with capacity for wide dispersal of gernvules rapidly became widespread anil, especially away From tic centres of survival, played the dominant role iv the early stages uf Ue seres. 2. Mreratory Raves Dissenmnation cf propagules is, of course, only the first stage in the process of re-colonisation. Tt ig necessary for them to germinate and hecome established in the new sites. Only germules which come to rest it saitable habitats, that is habitats within the range of rolerance of the ecotvpe, can hope to survive awl ecesise. Within the climatic amplitude of the biotypes, stecessful establishinent of species in the invading flora would be governed, therefore, principally hy olher enviranmertal factors. In the relative absenre of commpetitiun in the initial stages. there can be little douht that the distrilmtion of she early elements was determined within their climatic toler ranees, chiely by the edaphic environment. Soil conditions aver and ahove the effect of chance distribution uf the initial propagules, were umloublediy the major ecological agents which modified the checkerboard of the invading vegetation, and indeed. imposed upon it a special pattern. Athouch the underlying causes and the modes of dispersal are fondamental ta distribition, they would have had httle influence in just where the propagules initially came to rest. This was dependent upon the reiationshiw mi the parent plant to the agents of dispersion, physiograply, ete., thet is, chiefly dependent ypon imetcorvlogical factars, waterways atl pliysiagraphic conditions. he cxaet site ay which a propagnic lodges and germingtes is, as has been pointed ont by Cain (1944), entirely due to chanee in the sense thet there ig no conscious cO- operation hetween the geramle and rhe complex of dis aper sal agencies. Fallow- ing disperzal, germination, and in suitable habitats ecesis, aggregation and re- dispersal ot species no donk eveurred, aftd very soon the factor of competition becime Operative, iniposiny a general, shouch variable, brake upon the rate of nugration, Nearer the centres of survival the harricr o£ corapelit: on would obviously be imposed earlier because of the greater densily of propagtnes, With the release of climatic pressure on the relic fora, re-colonisatiog, no Combt, commenced fram most refuses, Tr is appirens. therefore, that put.artial nigration could have proceeded in many directions at ance. The extent and speed at which these movements tack place was goverued, apart from these fectors resi.bng within the species itself like dispersal capacity, chiefly hy the soil mosaic, as menhoyed previously, This is imeed a priteipie of eynamic plant gcorraptiy (Greason 1923). lt has heen stressed by (nai (1944) in saying that succeserul migration deptuds upon the aecurrenee of a “continuity of habitats which are within the Epa aight mf Jie thipraling spaces.” The “egolagical amplivde” of a anecies has its scat cw ita genetic goubL. vtion, and stric Uy, there: fore, within the species iteclf, hut the ocstittence ef sairnble habitats within this rang? is chichy » Tuneton of peologhal ard peru, ‘opical | uistory im owhieR vovetas fan phos ao important, int winor, rove, Ibis almost cer-ain, therefore, that vine eg migratory routes in the development of rhe present South Australi dom have bern from the suryival eetiires aiuag series oF closely revatie) spits. The extengive dune systems, muy at which are Hnked directly wits survival cantres ihe Spencer-Vincent evsten, Maurras oiiliee dimes, Simp Deseri, ele—would, [or exanple, have been adiniable rorites for re-colonisition, and were, 19 doubt, some of the niet important micratury tracts. Others were the gibber downs, the imallve.— oalebeechwood. Consociatians (one dominant only). ( Ouercetum vaboris - — - damp clays and Joams: onk ) neutral to alkaline Ouerceluat sessilijlarde - (a) savannah woodlands-podsols - (b) with heath-peaty podsols Fagetuim calciculum - — - rendzina pH 7-5-8-0 beech Fagetint rubosim — - - brown earth 4:5 -7°7 Fugetume ericetosum - - podsol 35-455 cersocies (seral stayes) are; btrchwood ————> oalkkewood ashwaod = ————> beechwooil Vhis unit of Tansley (the association) is too broad a unit of classification for practical purposes. Indeed, the cxample ubove secs to violate the very idea of an “association” of plants, for the yroupings of associated plants are tofally uniike in the different communities—some consociations have no species in common with each other! It also violates ‘l'ansley’s definition “constant habitat,” for, although climatic factors remem relatively constant in the above association, other equally important factors. ms. soil types, nutrients, pll and water relation- ships are widely different. The system really breaks down owing ta the wide patentia! environment of the dominant species—much wider than that of any of the species associated with it. Suppose we take a specific case, and apply Tanslcy’s classification to the conunmiiiies in South Australia dominated hy &Aucelyplus Baxter’, the Dbrowi strinsybark. The following well-defined cammunities are readily distinguished: E, Basteri-E. Huberiana — - - wet podsuls Fé. Baxtert-E. oblique - - - normal podsals EL Baxteri * - 2 dry and shallow podsols A. Baxtert-E, diversifolia - - siliceous sands and resinual podsols ever ironstone > lateritic soils (residual podsols ) EB. Baxteri-E. cosmophylla The rainfall over the whole gronp of communities varies trom 40 ta 14 inches per annum. The floristic make-up aud species-Irequency of cach of the above comugnrines is quite distinctive; some of the communities have few species in common except £. Baxter’, The first cummunity is a savannah woodland, the second and third dry scleraphyil forests, the fourth a mallée scrub, and the fifth a treeless macchia. There ig no evidenee whatever of any successional trends between then 125 Furthermore, in the Suuth-Rast of South Atistralia, ranges of residual podsols occur purallel to the coast and at right augles to the raivall isohycts, From 30 inches in the south ta 14 inches in ihe north, Progressing northwards, ofc fiads a gradual dropping out of sume species and their replacement by others, E. Bovteri continues as a dominant throughout; in the wetter areas as a forest tree, (hen ne a stuntes small tree, them as a slirub less than 3 feet high. Vinally it disappears, but most of the plants associated with tt in the stunted phase con- fine on ats heath or wallee-heath, obviously related to the former commmnity. Tt is clewe that HK. Bavteri ts a species with a wide potential environment, mich wider than that of most of the other plates associated with it, To celine a community based on the distribution of FE. Kaxteri alone is Lo ignore the asso- ciated species and simply to define the area ocenpicd by the cucalypt species. Relatively few species are distributed purely at random, and associations, under whatever system, are delermited subjectively. In defining ai association we take a middle course between the British-A\merican and the Scandinavian- Swiss schools. We define an association as a conslenl association (ie., growitg tog ther } af dominant specits recurring in siptilar habitals, The term dominant SPreies relers not only to the tree species, but also to the dominant shrub and/or herb specie s which give the characteristic look or Facies ta the community, It is similar in practice to the * scongogiation. "of Tanstey, but without the iuplications inherent in lis definition, This ts the grotiping ot greatest tse in Vegetation studies in Australia, though not neve sesaiiesly the most furduamental unit. fi prac- lice we bave found associations to be closely cor related with soil types; ar it on different soil types, then some comperisating Jactor can normaily be found, eg, water relalions, nutrients, ete. The assveiation iy made up of smaller units and may be grouped into larger ones. The snialler unita ate the type and the society. Type we define as « local chanyr ia the dominants ef the upper stration of an assoctation which is accon- panied by little or ne chanye in the other dancimants, A Suctety is 2 local change qi the dontinants of tie lower stratuns, ‘he coticept of type has beet used fora long time in forest practice. Appar- ently if was first used by Graves (1899), who ¢ tated “the same type of forest will tend x0 be produced on ‘the same classes of situation Keser suil in a specified region. There will be yariations within the type, but th sharacterisuic features will remain constant, that is the predomimant specits, Retisity habit. ot trees, character of undergrowth, etc, If a portion af the forest is destroyed by fire, wind or wther- wise, the lype may for the time being be changed, but if lefe undisturbe| will rever= to the original form, provieled the condition of the soil is not changed.” This definition, i essence, is sil retained in the Glossary of rechnical terms of the Society of American loresters (1944), where 1 1s added: “The tert suggests repetitions of the same characters wordler similar conditions.” Pryor (1939) first used the term in ecological liverature in Australia, nsing “onant titative floristic uiiformity of the domtnants” as the criterion, Pidgeon (1942) delined forest type as “a forest stand which has, wherever it necurs, the suine floristic composition of dominants, ancl which develops in essentially similar habitats.” Tn both definitious the term “dominants” refers to the dominant tree species. Hoth the above workers have sed this unit with advantage. for it is a reality in the field, “Phe anthors of the present paper disagtee with their definition of fores* type since it neglects the associated species, and in this way departs some what even from the original concept. ITowever, the difference is one of definition vather than of fact, for Pryor (lec. cit.) states: “In an area with very few dominant species, eg., jarrah region of Western Australia, the lowest order com- munity (ie., type) mitst be determined by some criterion additional to the floristic composition of dominants, ¢.7,, the floristic camposition of the lower strata of the community.” and again, “ihe vegetation type corresponds approximately with the Association of Braun-Bliunquet and the Faciation of Clements.” The essence of Braun-Blanquer’s association is “uniform floristic eonposition’—not of dominants only. We prefer, therefare, the definition of type as described above, In practice, there is no conflict between different Austratian workers in actual delimitation of types in the field. The corwept of type has been used in studying the forests of the mountainuus areas of New South Wales and South Anstralia (Pryor 1939; Pidgeon 1942; Boomsma 1946), In these areas the need for such a unit aries from the extreme sensitivity of many Lycelyptas species ta changes in the miero-habital; changes which in other parts ef the world cause allerations in the shrub or herb layers only. Ir is clear that some eucalypt species have a wore restricted potential enyiranment than have the shrub species with which they are associated, The type is a mote fundamental unit than the association as dutined above, alia jt will be clear that an association as defined by ts is a collection of types In a forest association, the floristies Of the ussociated species of the undergrowth remain the same, though there may be local changes in the tree species. In wailing associations into larger groups any characteristic of the association may surve as a basis for classification, “lhese classifeations abstract some characterislies of the association. Tl is clear that several alternative classifications can be made; they ate not necessarily mutually exclisive bul are complementary. and any may shed sume light on vegetation problems, The Freneh-Swiss school groups asvacintions inte larger units on the basis. of a common flora, although (his sometimes leads to grouping together of associations which are not nearly aiied. Using structure ani lite form as erieria, associa inns may he grouped into such units as selerophy!l forest, saytmmah woedlarid, savani- nah, cle., in any partionlar arca. These are the Forinatious of Tumsley and are groupitizs of greay convenience, although their use does not always lead to groupitte of nearly related associations, e.., the mallee-heath and heath of the upper South-East (Jessup 1946), nearly related im flovisties and habitat would he separated on this basis. Tn Sonth Australia we have used one feature of the habitat, vis., alMed soils. associated with velated flaristic caarposition ef specles as a basis for Cassification. Within definite climatic Jimuits, associations on nearly related soil types and witlt related flofistie composition are grouped cogether as an Mdaplic Caniples. striking cxample of an Exfaphic Complex is the associations Tout on podsolizerl soils in the Mount Lofty Ranges within the climatic sone linuted by the 25-50" unniual isoyet. Detailed work im pragress (Specht M46) shows that on these soils wellalefned aysoviations oceur with the followmyr eucalypts as domiiants; fs. oblique, BE. Bacteri, FE. cosmephylia, b, fascientoso, ff. Weucerylon, The podsols inelude normal, immature and residgal podsols, podsals on deep sands wl “vrey-broywn" podsels; they vary in their natrient and water retitions. Floristic lists for each of these eummilthities show that with few exceptiona the sue spucies occur in cach of them; their frequencies, however, differ greatly in the different assuciations, btit are constant for any one community associaied with & particular eucalypt. Shrub species dominant in one association may he only vecasinnal or Tate ii another, Ub js possible that licre we sce the selective effcet of environment at work on a relatively volng species-population. Selection has procveded to such an extent that the separate associations can readily be dis tinguished, hut the associations themselves bave unity in Horistic composition and in habitat in the Hdaphic Complex. 127 The greater portians of South Australia possess a heritage from the past, in thal many soils retain certain morphological features which are “fossil,” he. not Eormed in equilibrium with present climate, but relies of former, more or less irreversible, soil processes. The result of Recent aridity has heen a stripping and resorting ef the upper horizons of many of the old soils, and the building up and exposare of “new? soi] material, which ts naw re-subjected to pédogenic processes The result is a very diverse pedogenic history for many soils. 1a this complicated yaricty of habitats the Edaphic Complex is of yalue. For example, in the upper South-Hast, on neutral to alkaline sands, Jessup (1946) lias shown that a mallee sernb—cthe A. dizersifolia-l. augutosa association—oceurs om crests of hillss on shallaw sands-al the same type, anc on well-drained adjacent cand qiain or heath, the Casumina pusilla-Nunthorrheca aysfralis associatior ocenrs, The Bill slopes is an ecole or transition region wilh dominants from hath esgociutions present and wilh EF, Jeptopinylla more prominent--this is the well knew) mallve-heeth. In the twa assoriations the agsociajed species are quite dis- tinct. On utidulating plaing in this arca the vegetation ct first sight is extremely complex, yarving with every rise, and its edaphic relattouships only became clear when the well-detined sandhill-plain region was studied, ‘Such areas can often best be mapped as a Complex and sorting oui of associations left to detuiled study. Sinilarly, om the laterite and assuciated soils on Kangaroo Island, the fol- lowing agsociations can be distinguished : E. Baxtert-E, cosmophylla ~ - on Jaterite soils Ky remata - > “ - on heavy or massive laterite i diversifolia-E. cosmophytl om laterite soils with some siliceous sands om siliceous sands (A horizon of the laterite) ' t & Basleri-E. diversifolia =~ "Vhe matlee-broomlush complex in the South-East, Kangatoo Island and Eyre Peninsula (Crocker 1946, Jessup 1946, Wood 1937) 1s a good example of @ complex of associations closely allied, but showing some marked floristic differ- etices owing to endemism on Kangaroo Island and Eyre Peninstla, and to differing selection pressures im centres of survival in the three regions during the aridity. The vegetation on the immature soils derived from the Hawkesbury Sand- stone in New South Wales, and described hy Pidgeon (1942), is an Fdaphie Complex, In the associations described above there is no evicence of succession or of invasion, but only of transition. Indeed, selectivity af enyironment—espceially of climatic factors—is particularly marked in styeral areas of South Australia where a relatively uniform soil extends over'a wide range of present-day climatic condi- tions, Examples are the ranges in the South-Past. the restduai podsols on Kan- garoo Island and the mallee soils. [ere we may follow the slow dropping out ank| replacement of species by others as rainfall decreases, till at the two extremes are found commiinitics with few or no speeics in common. Communities which in Tanslvy's system are regarded as seral. ic, those which show allogehic succession, may also be growped as an edaphic complex, The commutitics on steep hillsides and skeletal soils, as well as on mature soits on tops of ridges, and on gentler slopes in the Mount Toviy Ranges, may he cited as am example, In each case the habitats are different and selective, so are the floristies, although they are allied. The hillside comimunities are stable, and a uniferm teritinal community is onty theoretically possible by a levelling down ot all physiographic units to the uniform soil typ¢. These examples might be termed edaphic successions; but in any case it is desirable to separate them from biotically induced successions. 128 ‘The examples given of the past listery of South Atistratian soils and yegeta- tion and the impossibility of reconstilutig the past, illustrate the dangers of dommatisiny about successions of this kind. ALL our experience indicates that succession should nat he wsed in any scheme of classilicaiion of communities, This does not imply that biotic stuccession does not occur, It is obvious on sand dunes, and in swamps generally wlere the plants themselves cause changes in the liabitit to produce a special soil carrying, at cquiaibrinm, a terminal com- mninity. Within associations as defined by us biotic sucerssidnal changes occur following fire, iclling. grazmg, cte., hut there is no evidence al seri changes, only a selectivity of the habitat, This is well illustrated in the lower South-East, where the following communities in the same ciimatic zone ate clear-cut both in floristics and habitat and with na evidence of seral staves between or leading to then. I. lisylert - - Sélerophyil forest - - on residual padsols Nuntliorrhoca-Hakea Heath — - - - - wet acid podsols Ji, cuinaldilensis = Savaunah woodiand - - meadow podsals intermediate meadow porlsol-rendzina GCahufe-Cladiwie - Siivannah - “ - - rendzina Alelelonca pubescens Savaniah woodland ~ térra tassa BE, ovata ~ - - Savannah woodland Much oi the lack of succession is doubtless due to the low degree of imtegra- tion of the open and ligt-demanding conimtaities of Eucalyptus as well ag to the historical factors diseyssedl previously. Invasion t more abyious in highly integr aied enmniunities and cera dainty oseurs if Australia where rain forést impinges on Fucalypt forest. bat the lack of suecession in our forests generally prolubits iis we a8 a characicristice in any scheme af elassileation: . VOT. DISCUSSION AND CONCLUSIONS OF the elements which make up the Australian flora the Australian element is the most prom iment in the southern regions, As has been iudicated its centre of origin is ohsetive. Imi in all likelihood: there were Many evntres. [ts present giv supeests a snutlie fs urigin, hut it is more Uhkely that the prin jury Genenis developed wher Auiistia generally had a tumperate and more unitocur elisa, 71 Ferbert nen has suegested ihat the yenws Excalyptns successfiiy established jtse! ‘ uniler a feraperate climes with abutmlant rainfall. li is perhips siznifieant tha 4 Sloat the tivaston rovtes af the Tadu-Melanesian element there has been practically nO byWwu-Wwely traffic, This lack of reriprority Was early noted by p Ploaber (1850), and it does SUgLOSE TNYisiuTs following with- drawal of the Anstraliatn elginent to the sous. Tu inty ease, the evidence is very st ronigly in voter of semuathings in the mature af a pan-Australian fiora int the carly Tertiary—at least over the sovthern half of Austra ha, Stich a pdssibiiity is miade more likely by the edaphic and climatic wifurmily which existed, for at this time the cortinent was reduced fo un alntosi perieet peiepluki. This uniformity was broken by marine trans rem sions, yoleanic uctivity and earth- mayenients which reached their mesxivum iu the lee Terti lity, thowsh cach, Ain especiaily the Teter, carried om inte the Quateriar Vv ia seme extent. J fabitar ‘liversity wits iurther } jiereased in the Pleistocene by the generally hich raitifall causing active crosion of the newly-fornied divides, are] ihe epi reading of large alluvial depasits. In the late Pigiglacene other factors such as the deposition ar large quantities of calcarcous loess ayer Southern Ausiralia led to a modification 129 of the edaplic conditions, while jn the Quaternary generally, large pseilations in the Mura occurred as a response to changing climatic conditions. OF these clituatic changes, the mos( significant from the viewpoint of the dis- tribution oF the present vegetation communities kas been a post-Pleistocenc period al aridity, which followed plavial cordifinns, The desiccation was so severe and sudden (hat it resulted in a considerable portion of the pre-arid flora being entirely wiped out, The surviving remnants were isolated in numerotis refuges, ios) imporiaay Of which were those Ta yuhs where habizat diversity, espeolaly eliniatic diversity, was sentient, The presert- day plant commmnities are the result af re-colomisation of vist, viebaally bare, areas, especially in the atid regions. This expansion uchicved ils greatest sti ans with a sight release of the clinvatic presspre, “The Auetralian Nera today can- (aims novery large pre-arid relic clenien:, Despite the fevourabh: concitions tay biotypical ciiferentiation with reunion following isolation and the possibilities ot later invasions (greatly favoured by the law level of competition), it is likely that this relic portion je still rhe Targer. Ln a broad way the dais of the arid regions isa blending of the Indo-Melanesivn element fronr the nests and north-wast wit the Awstraiian element from ihe santh-west. south om sauth-east, which las ovcurred subsequent. to the Great Australien Arid Period. It can he looked on, theretore, as a large ecotonal region, This htendiue of the two elements is well Mustrated by the floristic analyses of Murray (1931) and Gardner (1942), There is evidence for 4 post-leislncene xerothermis period in many places in the world (see Huntington and Visuer 1922), bul nowhere does it appear Lo have been sG severe, or to have had sech far-reaeling consequences as im Ats- tralia, nialess perhaps in (he sonthern Mediterranean reyion, Tn view of the close relationship between the accurrence of maximum arility, warmer seas and higher seas (10-20 feet) in Australia, and the evidence for a post-glacial worldwide (eustatic) rise in sea level of this arder, it is mot wolikely that aridtty was mueh more general at this tine than has beet imaging). Jansen (1924), frou saidies on the Rear iver delta, Gritish Colunbia, has concluded that present tempera- tures hay¢ not persisted for more tan 4,000-5,000 yens, the mird-post glacial being warmer. Although Anstralia largely liex in the Istitudes where dhe Ingh pressures tend to favour aridity, it is likely that the widespread edfects of the desiceation were dic to a conibination of Factors. OF theae the general lack of relief (neat-pencplanation) of the greater part of the continent, large areas of laterised sos which in many areas probably preserved up to this tinve much of their easily-atripped JA horieons, the ocenrrence over much of Australia af great quantities of alluywam and detrital meterial, anil of certain lime-rich soils in southecn Australia which were liable lo casy loss of stability, together with the presence af a very mesic fora, are wiost imporiant aad are stfiicient to explain the results. One fact which is apparent From the analysis ef the cevelopment of the south Australian plant communitics is that the asgectations themselyes are very young. Their distribution has been determined within a climatic zone chicily by c:laphie conditious, im that these have been responsible toy the sequence of related haditats, which have made ap the migratory routes aud nave largely determined the speed and direction of migration. As already mentioned the distr;budan patterns have heen greatly influenced, however, by other factors such as individual dispersal capacities, chance dispersals, opposing migrations, location of survival céitres, varying degrees of biotypical differentiation (including speciation), barriers, ete. It ig nol surprising that the final species-aggrcrates (associations. etc.) arc not made up of units with identical lolerances. Nov is it surprising that an attempt to inter-relate these communities 136 (Wood 1937) as seral stages in the accepred sense of Clements (Clements 1916) shold tive been unsuccessiul. The cegeee of stability of vegetation will be governed hy the historical sequences, and these will vary greatly froin woutry ty coluintry and within a country. In some respects the respente of the flora to aridity in Australia was very similar to sume of the fora Auetuations of the Pleistocene Ice-ages in ie northern hemisphere, Uhese, bke cur aridity, wiped out the yereution over laree areas, but jef{ centres of survival, Glaciation, however, was a mueh mort gradual Phenomenon, permitting a considerable amount of adaptive differentiation 14 proceed, More important still was the [aet that mast species bad a sufficiently wide amplitude and a fast erough migration capacily to keep ahead of the ice in ils advance, and to follow it fairly closely in retreat, In this way wholesale destruction wes avoided and the pattern oi re-columisation has ween, in fact, entirely different. Chances Jor unlimited fybridisation and/or cousulidation of biotypical differentiation did nat ovcur io the same dewree. The time factor, tun, Las heew so very differest. The lastorieal sequences aud their COTSEMENCEeS Sey sufficient 10 explain the specific sibility of the north Europenn and Amertean floras by eonmparisnn with ihe yroat specific complexity of the \ustratiaw Hora. The ordinal and specific peculiarity of the flora of South-west Austratia and the ‘high percentage of endemies (70-80%) is considered to be adequately explained by a consideration of the palcobutanieal evidence in telaiion to geological and chmatic history. A pan-Anstralian flova probably existed in the vatly Tertiary. A southerly elimatic-induced contraction in the Miocene coincided with the inundation of large portions of southern Australia, and effectively isolated Noristically south-east and eourhewest Australia, The sea had retreated altiost lo its present position by (he end of the Pliocene, Despite certain expansions of the flora since that time, a large measure of, if not complete, isolation of the flura of the south-west has subsequently been maintained. The barriers to thigration have heen chiefly edaphic and ctimatic, though the [oundering of the gulls of South Australia was also sitnificant. The richness in endemics of the flora of south-west Australia is the most interestine feature geographically in the flora of the continent. ly was mentioned in discussing this matter earlier that the Dora isolated in the sunth-west by the las Pertcury seas was a peneplain flora, and dhat before or sitve then it must have become “adapted” to laterisie soils because they were at some stage af the Tertiary 2 widespread feature of the peneplains. Western Australia largely avoided the jute ‘Pevtiary-Quaternary upliits whieh resulted in the eastern divides aud the Mouyt Lotty-Flinders system, and kas ven both prior and subsequent to this, one of the wost stable parts of the Commonwealth, Vhe Teteritic soils are preserved there to a betrer degrce than anywhere else in Australia, This emphasises the mupurtance of edaphic barriers in mainwaining isolation after the removal af other physical barriers on the fetreat of the Ter lary seas. Ln addition to the edaphic barriers mentioned earlier is that caused lw the develapment, with accession af calearcous loess during the late Pleistocene, of a larke area of pedocalcic sors lying right across the migratory routes. ‘This alone would prove a severe and extensive obstacle to late migrations, That the Australian element jn the sonth-west of Western) Australia ds one seleeted hy lateritic soil can hardly be doubted. According to Gardner (loc. cit.) those parts of the suuthayest richest im the purely Ansiralian element are the sand heaths. “Phe floristic wealih of these heaths is probably unparalleled anywhere else oar earth,” he says. Tf js significart that these are in the region of Australia where parts of the old laterilic profiles ate preserve, but more significant, these sand livaths are some ot the few regions where the old A barizang of ie laterite. 131 trineated more or less, and perhaps suffering some local rearrangement, are still presetved.()) The sand, which is underlain by laterite, is portion of this old A horizon. The floristic stability of the Mora of the south-west is paralleled, there- fore, by a great measure ot edaphic stability, Because of the narrow climatic zonation, climatic stability, ur at Jeast the provision of a wide range of climatic habitats within a short distance, Las been maititained also for a long time. From the stirvey that has been made in this paper of geological Instory, and its consequences on the Australian, and especially the Setth Australian flora, it is apparent that these geological [actors haye played an imporrant part in the development of many groups of plants. The foregoing considerations, both on the Australian flora and the develop- ment of the vegetation af South Australia, are haved on only a very impertect understanding of Tertiary and Quaternary geological and climatic history, in addition to a very restricted paleohatanical Literature. Their tentative nature must, therefore, he stressed. Tlowever, although many modifications to the con- clusions ate to be expected, the present agreement ammngst geologists on many of the basic facts in the Australian posi-Mesoznic record, extended ecological and peolosica) knowledge, and the establishment of many new principles in plant geography are sufficient justification for um attempt of this type. Further improved interpretations will depen! largely on advanees in Australian geology, pedolivy, paleohotany (including paleoevology), cvingenctics and phylogeny. This analysis of the time factor in ecology with ils insistence on the selective effect of edaphic factors and migrating plant poprlalions has strengthened the view put furward by one of us (J. G. W. 1939) and amplified here, wvis., that the basis underlying ecology is a physiological one, and that in the case of species growing naturally together the potential environment of the individual species overlaps the actual environment. In no case is fhere evidence of succession or progression to a climatic cliniax, In the system of classification of comruunities proposed, the unit (the association) is based upon constant eésociation of dominant species in which the term dominant refers not only to tree species, but lo characteristic species of other layers. The concept of edaphic cumplen, first propased as a unit of convenience, is shown by practical experience to be especially valuable in view of the complex soil mosaic consequent on the previous history of the area considered. TX. SUMMARY 1. Au analysis is made of some of the factors affecting the development of the Australian flora in the light of an improved knowledge of the geological, pedological and ernlogical pattern of Australia, ancl in relution to the more recent paeobotanical literature, 2. Evidence for a mid-Recent period of aridity in Australia is brought together and summarised, The significance of this desiccation on the disiihytion pattern of species and communities in South Australia is disctissed in some detail, The presence of a large pre-arid relic clement in the flora is postulated, and reasons given to explain the presetice of a large number of “species complexes.” 3. A namber of Australian plant geographical prohlems are considered, These include the origin of the Australian elemenl, and the high degree of endemism in the santh-west of Western Ausiralia. 4. LEeological concepts are discussed in the light ot the time factor inyolyed, The hases underlying systems of classification are discussed and crolugical units are defined. = —————————————————————— 4. G, Stephens, Bull, 206, CSA.R. (Austt, in press 132 ACKNOWLEDGMENTS The authors wish to acknowledge information on the distribution of certain species provided by N. C. W. Beadle, C. A, Gardner, and 7, W. Whitehouse. REFERENCES Antevs, E, 1938 Post-Pluvial Climatic Variations in the South-West. Bull. Amer. Met. Soc., 19, 190-193 Baas Becxino, L. G. M. 1934 Geobiologie. The Hague Boomsma, C. D. 1946 The Vegetation of the Southern Flinders Ranges, South Australia. Trans. Roy. Soc. S. Aust... 70, (2), 259-276 Brett, R.G, 1946 Some Aspects of Hybridisation in Relation to the Taxonomy of Lucalyptus. Unpublished paper, Section M, Aust., N.Z. Assoc. Ady. Science Brooks, C.E, P, 1926 Climate through the Ages, J.ondon Brovcuton, A, C. 1921 Notes on the Geology of the Moorlands (5.A.) Brown Coal Deposits. Trans. Ray. Soc, S, Aust., 45, 248-253 Browne, W. R. 1945. An atternpted post-Tertiary Chronology for Australia, Pres. Add. Linn. Soc. N.S.W., 70 Bevan, W. H., and Jones, O. A, 1945 The Geological Histury of Queens- land. Univ. of Old. Papers, 2 n.s., (12) Bryan, W. H., and Wiirruouse, F.W. 1926 Later Paleagraphy of Oueets- land. Proc. Roy. Soc. Old., 38, 111 Cain, A. S. 1944 Foundations of Plant Geography. Harper and Brothers, New York CampBAce, R, H. 1913 Development and Distribution of the Genus Eucalyptits, Pres. Add. Ray. Soc. N.S.W., 47, 18-58 CHapMAN, F. 1937 Descriptions of Tertiary Plant Remains from Central Aus- tralia and other Localities. Trans. Roy. Soc. S. Aust., 61, 1-16 CHarman, F. 1921 A Sketch of the Geological History of Australian Plants. The Cainozoic Flora, Vict. Nat., 37, 116-119; 127-133 CLements, F. FE. 1916 Plant Succession. Carnegie Insi., Wash. Publ. 242 Corron, B. C. 1935 Recent Australian Viviparidae and a fossil species. Ree. S. Aust. Mus., 5, (3), 339-344 Cowres, H. C, 1901 The Physiographic Ecology of Chicago and vicinity: a Study of the Origin and Classification of Plant Societies. Bot. Gaz.. 31, 73-108, 145-182 Crocker, R. L, 1941 Notes on the Geology of South-East South Australia. with reference to late Climatic Llistory. Trans. Roy. Soc. S. Aust., 65, 103-107 Crocker, R, L, 1944 The Soils and Vegetation of Lower South-East South Australia. Trans, Roy. Soc. S, Aust,, 68, (1), 144-172 133 Crocker, R. L. 1946 (a) Post-Miocene Climatic and Geologic History and its eerenee Bs ie Genesis of the Major Soil Types of South Australia. Bull, 193, C.S.LR. (Aust. ) Crocker, R, L. 1946 (b) The Swils and Vegetation of the Simpson Desert and its Borders. Trans, Roy. Suc, S. Aust., 70, (2) Crocker, R. L, 1946 (c) An introduction to the Soils and Vegetation of Eyre Peninsula, South Australia, Trans, Roy, Soc. 5, Atist., 70, (1), 83-107 Crocker, R. L., and Earpuey, C.M, 1939 A South Australian Sphagnum Bog. Trans. Roy. Soc. S. Aust., 63, (2), 210-214 Crocker, R. L., and Corrox, B.C. 1946 Some Raisec Beaches th the Lower South-East of South Australia. Trans. Roy. Soc, S. Aust., 70, (2), 64-82 Dartincron, C.D. 1940 Taxonomic Species and Genetic Systems, “The New Systematics,” ed J. S. Huxley, pp, 137-160 Davin, T. E. 1932 Explanatory Notes to acccmpany a new Geological Map of Australia. Aust, Med. Pub. Co., Sydney Diropzuansxy, T, 1941 Speciation as a Stage in Evolutionary Divergence. Amer. Nat., 74, 312-321 Diets, L, 1906 Die Pllanzenwelt von West-Australien. Wilhelm Engleman, Leipzig Diver, C. 1940 The Problem: of closely related Species living in the same Area. “The New Systematics,” pp. 303-328 Earpiey, C. M. 1943 An Ecological Study of the Vegetation of Eight Mile Creek Swamp, a natutal South Australian Coastal Fen Fortuation. Trans. Roy Soc, S. Aust,, 67, (2), 200-223 Fenner, C. 1921 The Craters and Lakes of Mount Gambier, South Australia. Trans. Roy. Soc. S. Aust., 45, 169-205 Fenner, C. 1930 The Major Structural and Physiographic Features of South Australia, Trans. Roy. Soc. S, Aust., 54, 1-36 Garpner, C. A, 1942 The Vegetation of Western Anstralia. Pres. Add. Jour. Roy, Soc, W. Aust., 28 Ginns, L. S. 1921 Notes on the Phytogeography and Flora of the Mountain Summit Plateatix of Tasmania, Jour, Ecol., 8, 1-17, 89-117 Greason, H. A. 1923 The Vegetational History of the Middle West. Ann. Assoc. Amer. Geog,, 12, 39-85 Goopwrn, Rk. H. 1937 The Cytogenetics of two Species of Solidago and its Bearing on their Polymorphy in Nature. Amer, Jour, Bot., 24, 425-432 Hanson, G. 1934 The Bear River Delia, British Columbia, and its Significance regarding Fleistocene and Recent Glaciation. Trans. Roy. Soc, Canada, 28, (4), 179-185 Heprey, C. 1915 Pres. Add. Ruy. Soc. N.5.W., 49, 1-76 134 Herbert, D. A, 1928 The Major Factors in the Present Distribution of the Genus Eucalyptus. Proc. Roy, Soe. Qld., 40, 165-193 Hitis, E. S. 1939 The Physiography of North-western Victoria. Proce, Roy. Soc, Viet., 51, (2), 297-320 Hooxer, J, D. 1860 The Botany of the Antarctic Voyage of ILM. Discovery Ships “Erebus” and “Terror” Introductory Essay to Flora of Tas- tatua. Lovell Reeve, Lond, Howenin, W. 1913 Pres, Add. Sect. C., Aust-N.Z. Assoc Ad. Science, 14, 148-178 Howenin, W. 1929 The Geology of South Australia, Adelaide Huntineron, E., and Visuer, S.C. 1922 Climatic Changes, their Nature and Causes: Yale University Press Huxtty, J.C. 1940 The New Systematics, Oxford University Press Jack, R.L, 1921 The Salt and Gypstum Resources of South Australia. Bull, 8, S. Aust. Geol. Survey Jessup, R. W. 1946 The Ecology of the Area adjacent to Lakes Alexandrina and Albert. Trans. Roy, Soc. S. Aust., 70, (1), 3-34 Manican, C. T. 1936 The Australian Sandridge Deserts. Geogr. Review, 26, (2), 205-227 Manican, C. T, 1946 Simpson Desert Expedition, 1939, The Sand Forma- tions. Trans. Roy, Soc. S. Aust., 70, (2), 45-63 Mawson, D. 1942 The Structural Characters of the Flinders Ranges. Trans. Roy. Soe. 5. Aust., 66, (2), 262-272 Mawson, D., and Cuarman, Ff. 1922 The Tertiary Brown Coal Bearing Beds of Moorlands. Trans, Roy. Soc. S, Aust., 46, 131-147 Mutter, H, J. 1940 Bearings of the Drosophila Work on Systematics, “The New Systematics,” 185-268 Ostvarp, H. 1923 Die Vegetation des Hochmeeres Komosse. Uppsala. Pipceon, I. M. 1942 (M.S.) Ecological Studies in New South Wales, D.Sc. thesis, University of Sydney. Pipczon. I. M.,.and Asuny, E. 1942 A new Quantitative Method of Analysis of Plant Communities. Aust. Jour. Se, 5, 19 Prescott, J. A. 1931 The Soils of Australia in Relation to Vegetation and Climate. C.S.1.R. (Aust.), Bull. 52 Prescott, J. A. 1944 A Soil Map of Australia, C.S.L.R. (Aust.), Bull. 177 Pryor, L, D, 1939 The Vegetation of the Australian Cupital Territory. M.Sc. thesis, University of Adelaide Rriptey, H. N, 1925 A Short Note on Endemic Plants. Jour. Bot., 63, 182- 183 Riptey, H. N. 1930 The Dispersal of Plants throughout the World. Lovell Reeve, Kent 135 Setcuett, W. A. 1935 Pacific Insular Floras and Pacific Paleogeography. Amer. Nat., 69, (725), 289-310 Sxorrsserc, C. 1938 Geographical Isolation as a Factor in Species Formation, and in Relation to Certain Insular Floras, Linn, Soc. London. Proc. Sess., 150, 286-293 Sxotrsperc, C. 1939 Remarks on the Hawaiian llora. Linn. Soc. London. Proc. Sess,, 151, 181-186 Svecirr, R. 1946 Ecology of Part of the Mount Lofty Ranges (1), Thesis: University of Adelaide Spencer, W. B. 1921 Pres. Add. Aust. N.Z. Assoc. Ad. Sc. Hobart Meeting Spricc, R. C. 1942 The Geology of the Eden-Moana I'ault Block. Proc. Roy. Soc. S. Aust., 66, (2), 185-214 Srricc, R. C. 1946 Personal communication SrepHens, C. G, ef al 1945. A Soil Land-use and Erosion Survey of County Victoria, South Australia. Bull. 188, C.S.UR. ( Aust.) Sveprrens, C. G. 1946 Personal communication Tansiey, A. G. 1939 The British Isles and their Vegetation. Cambridge Uni- versity Press Tatr, R. 1879 Pres, Add. Trang. Adel. Phil, Sec., 2 Tate, R. 1884 Post-Miocene Climate in South Australia. Trans, Roy. Soc. S. Aust., 8, 49-59 Tuomas, H. H. 1936 Paleobotany and the Origin of the Angiosperms. Bot. Review, 2, 397-418 Timoreer-Ressovsky, N. W. 1940 Mnutations and Geographical Variation. In “The New Systematics,” pp. 73-135 ‘Turritt, W. B. 1940 Experimental and Synthetic Plant Taxonomy. In “The New Systematics,” pp. 47-71 Wako, A, B. 1919 Mesozoic Floras of Queensland. Qld. Geol. Surv, Pub, 263 Watkom, A. B. 1919 Queensland Fossil Floras. Proc, Roy. Soc. Qld., 31, Warp, L. K. 1926 The Plan of the Earth and its Origin. Proc, Roy. Geog. Soc. S. Aust., 28, 171-199 Warp, L. K. 1941 Bulletin 19, Dept. of Mines, S. Aust. Wuttrnouse, F. W. 1940 Studies on the lute Geological History of Queens- land. University of Queensland Papers, Geol., 2, ns., (1) Wits, J, C. 1922 Age and Area. Cambridge University Press Wiis, J, C. 1940 The Course of Evolution. Cambridge University Press Woop, J. G. 1930 An Analysis of the Vegetation of Kangaroo Island and the adjacent Peninsulas. Trans. Roy. Soc. S. Aust., 54, 105-139 136 Woop, J. G. 1937 The Vegetation of South Australia. Government Printer, Adelaide Woon, J. G. 139 Ecological Concepts and Nomenclature. Trans. Roy. Soe. S. Aust., 63, (2), 215-223 Woop, J. G., and BAAs Becxine, L. G. M. 1937 Notes on Convergence and Identity. Blumea, 2, 329-338 Woops, J. E. T. 1866 Report on the Geology and Mineralogy of the Sotth- Eastern District. Government Printer, Adelaide Wortuincton, E. B, 1940 Geographical Differentiation in Fresh Waters. In “The New Systematics,” pp. 287-302 WricuHt, SEwett 1940 The Statistical Consequences of Mendelian Heredity in Relation to Speciation, In “The New Systematics,” pp. 161-184 ZeUNER, Ff. E. 1945 The Pleistocene Period—its Climate, Chronology and Faunal Suecessions. Rov. Society, London : ee | pee ree ¢ VOL. 71 PART 2 DECEMBER 22° 1947 , ores TRANSACTIONS OF 7 ee ‘THE ROYAL SOCIETY z: OF SOUTH AUSTRALIA INCORPORATED ADELAIDE PUBLISHED AND SOLD AT THE SOCIETY’S ROOMS KINTORE_ AVENUE, ADELAIDE Price - - Fifteen. Shillings Registered at the General Post Office, Adelaide, for transmission by post as a periodical KEY TO THE SOUTH AUSTRALIAN SPECIES OF EUCALYPTUS L’HERIT. By NANCY T. BURBIDGE, M.SC. Summary The publication, in 1934, of Blakely’s “Key to the Eucalypts” marked a definite advance in our knowledge of the genus, since the book made a very useful companion to Maiden’s rather bulky “Critical Revision of the Genus Eucalyptus.” It was a pity that Blakely’s work, which was the result of years of patient and intensive labour, appeared under such misnomer, since identification of specimens with the aid of his “key” is difficult even when one has acquired a certain familiarity with the genus. This is particularly so when dealing with some of the members of such a complex as that around E. odorata or E. oleosa. Owing to the number of new species described by Blakely and the modification of some of the older ones, the key in Black’s Flora of South Australia is no longer adequate and a new one is needed. 13? KEY TO THE SOUTH AUSTRALIAN SPECIES OF EUCALYPTUS L’HERIT, Ry Nancy T. Buruiogu, M.Se. Waile Agricultural Research Institute, Adelaide {Read 10 April 1947] The publication, in 1934, of Blakely’s “Key to the Eucalypts” marked a definite advance in out knowledge of the genus, since the book made a very useful com- panion to Maiden’s rather bulky *'Critical Revision of the Genus Eucalyptus.” It was a pity that Blakely’s work, which was the result of years of patient and intensive labour, appeared under such misnomer, since identification of specimens with the aid of his “key’t is difficult even when one has accired a certain familiarity with the genus. This is purticularly so when dealing with some of the embers of stich a complex as that around /. odarata or £. olcosa. Owing to the number of new species described by Blakely and the modification of some of the older ones, the key in Black's Flora of South Australia is no longer adequate and a new one is needed. Since Blakely’s death no botaiist hag taken over his eucalyptological work, which is fo he regretted when the economic and botanical importance of the genus is considered, The whole genus is badly in need of integrated research by taxonorists, ecologists and geneticists, and hence this paper merély tri¢s to clarify the present canfusion so that a key to the South Australian species can be construeted, The genus Eucalyptus meludes aii unusually high percentage of variable and unstable species which, for want of a more definite, term. can be called poly- morphic, Hybridisation is known to occur in the field, and evidence concerning ily extent in Tasmania is being accumulated by Rrett (5). Under natural cordi- tions, and with stich slow growing and long-lived species, it must be a long period before the resulis of any cross can be stabilised. It is possible that complete stability is never attained, ahd the plants representing various stages may be found close together in one locality or scattered throughout a region according ta UWteir adaptability to different habitat conditions. Possibly this explains the existence of some of the species complexes. But, whatever the cause of palymorphy may be, it is the rusulting variants which must be deal with in a key. Whien there is a series of related forms. the stalus given to any particular variant depends partly upon the amount of material available for comparison. If specimens A,, C, and E. are found they may all apparently deserve specilic rank. Add D, and they may be reduced to varieties. Add B. and Fy anu it becomes very difficult to sepatate them at all under any reliable character. On the other hand, the larger and more polymurphic a species. beconies, the harder it is to delineate i sharply from other related species, Llence a balanee must be struck, and since the meaning of the word “species” is, under these circummtunces, tnore that) ustially uncertain, one is foreed to take a somewhat yliilarian view and seek the most workable arrangemen! for practical purposes. This work is not intended as a complele revision of the local species which would require some years of study hoih in the herbarium and in the field, as well as a fuller knowledge of Eucalyprs occurring in other parts of Australia. The writer is in full ayreement with Ferdinand von Mueller when he states, "to assign to each species ifs proper place involves the study of all allied congeners. and these are often... . dispersed at wide distances in Australia” (9), At preserit there are large gaps in localities whence specimens have been oblained, Until the areas around the Great Australian Bight and to the north of the Nullarbor Plain have been studied, relationships with Western Australian species cannot Trans, Roy, Soc, S. Aust., 71, (2), 1 Ducrmber 1947 138 be solyed. Central Australia is still, botanically speaking, terra incognita and much remains to be learnt concerning the southern species whielr extend north and the tropical forms which occur over the boréer of Sottth Australia. In the east, alfmities with species in New Sonth Wides and Victoria are better understood, but knowledge is far fram compleie, DISCUSSION OF CHARACTERS USED IN THE KEY Jn his preface Blakely states thal the original intention of Maiden was to design a key based on the nature of the cotyledonary and juvenile foliage. Owing ra incomplete data this proved impossible and Blakely used the anther type in his primary division of the genus, The shape and number ot juvenile leaves is hilely to be of real importanee when relationships are bem traced, but in a key such a character would be difficult to employ since the necessary material is nat always available to the collector. In basing bis key om the anthers Blakely followed Bentham (1), but, anfortunately, great familiarity with his herbarium material antl i certain lack of field experience led to subdivision into a large number of groups based on sight differences in shape. The result is confusing both to the novice and to the experienced botanist. The original plan as designed by Bentham was reasonably sound, though until adequate anther material of all species is avail- able the value of this character cannot be fully estimaied. Plakely's subsections are so difficult to recognise that anthers have been emitted from this key, except where several features of the organ cun Le contrasted with corresponding characters in other species, Bentham’s secondary divisions were largely based on the character of the fruit, and in the preparation of this paper more and more reliance came to be placed on this feature, particularly om the nature of the dise and the type of the valves. Correlation between anther type and fruit character was imperfect, though it exists to some extent. This issue is probably confused beeause of the number of lines of development which are tepresenied amung the species. The top of the fruit is formed of four zones. The outer rim represetits the abscissian line left by the fallen operculum and is known as the calycre ving, the ext is the staminal ring, then the so catled disc whose ontogeny has not yet heen described. It is not certain that this term has been consistently employed by botanists. Inside the cise is the top of the ovary which ruptures into the valpes. The variations produced im these zones in different species are considerable. The calycine ring is rarely important, though it is wide and conspicuous in the Ruwers and fruits of &. pyrifornis and its affinities. This fact has been employed in the key, li some cases the staminal ring is a narrow clevated ring in the Hower, but it usually hecomes inconspicuous in the frat, In others, eg, EB. lencoxvion, the ring ts comparatively wide, but thin and projecting over the disc. In this species the ritig etther falls away during maturation of the fruit or ic remains partially” attached on one side of the fruit In E. intertexta, and also in ZL. calciculirix, which lies between £. lrucox\lon and ££. adorala there may also be a distinet Stamiinal ring in the maturing irvit, but usually it is not deciduous and it may hreak up into a number of small pieces attached aromd the rim of the fruit, In fresh flowers and young fruits the dise ig more or less continuous with the top of the ovary. As the fruit matures or, in collected specimens, dries the dis- tinction between the two paris becomes increasingly marked until, with the open- ing of the valves, they appear quite unrelated. The dise itself may be broad az in £. Martert, a narrow ridge as in E. wnerussula, a ring as in 4. aleasa or E. lepto- fiivila, or it may be little more than a Nning to the top of the tube when the capsule ts sunken as in A. fasciculosa ur A, odorata. When broud the disc may be domed, i.@., convex as in 2. camualdulessiy anc 2. wiminalis; Ral as iu EZ witeen ar onancave, 1.e., oblique as in &, ebliqua and EF, lencoxrylon, 139 The valves may be short and triangulur as in E, caialdulensis in which case they represent Lhe top of the ovary, or they may have sharp points formed by the rupturing of the persistent hase of the stvle asim 2, mevassata and £. dunwsa, or the sharp points may be long and coherent at the apex due ic the persistence at the entire style as in E. vleosa. La the latter case the valves are brittle ant may break off short, confusing the inexperienced student. Just which type can be regatded as the most primitive is a question demand- ing a wider and more extensive study and correlation with the affinities demon- strated by such characters as juvenile foliage and anther shape, Owing to South Australia's central position, geographically, the local Eucalypts include species vepresenting a number of different lines of development, Consequently much speculation here concerning phylogeny within the geuus would be ont of place. In specimens bearing immature ftnit the valves open during drying and release the seed. Sucli capsules may appear Eully developed and will give a false impression uf the size normal for Lhe species. In plants growing im arid habitats a lack of soil moisture or a sudden hor spell at the crucial time may induce a similar dehiscence of immature fruit. Another probable cause of subnormal size and shape in fruit, apatt from insect or mechanical damage, is when no seeds mature, This is suspected when Ji. Morrisii, which is supposed to have a high convex disc, bears fruits with flat ones (pl. i, lig. 4 b-c). The word “striate” has caused sorne colifusinn ih the literature. Blakely used it rather freely whenever the bud or fruit was slightly marked by longitudinal lines. Others have allowed the term to cover fruits which Blakely would have called ribbed. In this paper the word is used when longitudinal ridges or lines are apparent owing to the shrinkage of nor-sclerised tissnes, e.g. H. tuerdssata var. costuta, while “ribbed” has been tised when the ridges are deep enongh to have been evident when the specimen was fresh, e.g.. E. pyrifermis and E. inerassala var. angulosa, Ina Tew cases it is difficult to draw a line between the two condi- tions, but in general the above use of the terms can be quite consistent. With reference to distyibution, it may be stated bere that Blakely's lists al localities were compiled from the material in the Sydney Herbarium. They are, therefore, incomplete in some cascs, ¢.g., 2. elacophera, aud incorrect in others, e.g., E. gontocalya, The latter reference was based on @ specimen which is, in the writer's opinion, ZL, claeophara. Short deseriptions of the spectes have been included to make the use of the key easier. Field details such as bark, wood colouration and juvenile foliage have been omitted. ‘The writer carmor claim a wide knowledge of the first two, and a! have been covered in Dlakely’s descriptions. Key To THE SPECIES (N.B—tThis key is based on appearance of specimens when dry. Ribbing or wrinkling of bres and fruits is always lese evident in fresh material, ) A, Staminal ring Jobed, Bais tetragennus, Teaves oposite ~- 1. endesmipides AA. Steuminal ring. not Iobed $8. Operculum thick, leaving avhrnad tim cutside the staminal ring, Dise with an erect citizge atound the pattiglly sufken valves. ©. Buds nericellate, calyx tube ribbed, D. Vroit more than 4 en. i (ameter = ches a re 2 pyriforntis DD), Fruit less than 4 em. im diameter : 1 & pachy phyla CC, Buds sessile, calyx smooth, D. Buds glancous veel re aide _ lee wee ae AL oxwnilra DD. Buds net clancons oa chit =f a, 8 Old fieldii BB. Opereulum uct thiele. C. Dise broad and conver ar flat around the short broad valves. T. Valers either markedly exsert and contributing te the lemsth o) the truit ar in the orifiee. ; 140 E. Peduneles and pedicels slender, Operculum rostrate or conical ie Hi aut uA “3 an: Hy. Len EE. Peduncles and also pedicels when present not slender, usually short and stout. F, Mallees or small stunted trees. G. Species occtirring in arid interior. H, QOperculum conical to restrate. qtiite sessile, I. Operculum acuminate to beaked. Bud glaucous .. fl. Opereulum obtuse. Buds not glaucous .... HH. Operculum hemispherical. Fruit pedicellate GG, Coastal or subcoastal species, H. Operculum hemispherical or hemispherical-conical. Buds and fruits pedicellate HH. Operculum conical to rostrate. very shortly pedicellate or sessile FF, Trees occurring in higher rainfall areas, G. Fruit about 10 x 10 mm., markedly sessile (except var.), usually in globose clusters, Operculum hemi- spherical or short and very obtuse 3 5 Fruit almost ot Buds and fruits GG. Fruit smaller than above. Operculim conical, TT, Juvenile leaves avate-orhicular, glaucous. Buds and fruits more or less glaucous cert thn ia HH. Juvenile foligge broad or narrow lanceolate, pale green, Ruds and fruits not glaucous. T. Umbels with three flowers ee Mae sey Il, Umbels with three or more flowers occurring on same btanch. J. Umbels on leafless branches or racemes JJ. Umbels all axillary DD. E. Operculum conical, about as long as. the tube. femi-longitudinal, Bark rough 2 mn sm whe EE. Operculum shorter than tube, hemispherical. Venation longitudinal. Bark smooth wh ne a t, CC. Dise natrow and forming aring or rim on the top of the fruit or, if broad, concave and extending obliquely into the capsule of forming a lining inside the tap of the fruit. D. Dise forming a ridge ar concave and oblique, E. Qperculum hemispherical -apicu'ate, Dise strongly oblique mm - vig bn wh ne Valves enclosed and more or less concealed by the disc. Venation El. QOperculim conical or rostrate (except F. concinna), G. Staminal ting obscure or if evident on fruit then not broad and extending over the disc. YW. Flowers in axillary umbels, IT Stamens all fertile Onter ones not markedly fonger than inner, Buils without conspicuous oil glands. J. Anthers versatile, opening widely to base with cells parallel and side by side, with a con- spicwous gland on back. KK. Trees in high rainfall country. L. Buds not angular at base. Fruit turbinate LL. Buds angular ut base. Fruit cylindrical .. KK. Mailees. L, Buds and fruits sessile on very short but not flattened peduncles, bicostate or un- tibbed. "Top of ovary sot ridged between valves when dimmature. Valves in orifice . 6. camaldulensis & var. obtusa 4. pxqanilra . Morrisit . Bwarhona CONT 9, remota 10. diversifolia U1, Baxrtert & vat. pedicellata 12. rubida 13. ciminalis 13a. vininalis var. racemosa 13b. wintinalis var. Auberiana 14. vitrea 15. pouctflera 16. oblique & var, miegacarfa 17. owefa & var. grandiflora 18. elacophora 19. cosmophylla 141 LL. Buds and fruits pedicellate, or if sessile and bicostate then with flattened peduncles. Top of ovary usually ridged between valves in immature fruits. M. Valves deeply enclosed. Umbels deflexed MM. Valves in or near orifice. Umbels erect. N. Valves broad and triangular with short points. Fruit 10 x 10 mm. or larger. ©. Fruit smooth. Peduncle terete OO. Fruit striate. Peduncte terete or narrow but flatiened ... Eu git OOO, Fruit coarsely ribbed, Peduncle broad, flattened rH 70 ite NN, Valves narrow and more or less sub- ulate, Fruit smaller than above. QO. Buds and fruits short and fat, the latter broader than long and so sessile as to be almost imbedded in the end of the flattened peduncle Non OO. Buds and fruits not as above. P, Fruits sessile, striate, on thick peduncles wh sags * Ae PP. Fruits pedicellate, sometimes only shortly so. Q. Calyx tube 2-4-ribhbed and pass- ing into a flat pedicel. Peduncle flattened. Operculum —hemi- spherical-conical, obtuse “ny QQ. Calyx tube not 2-4-ribbed. R. Peduncles and pedicels not slender. Operculum = shorter than tube, S. Buds smooth or very minutely striate, Peduncle terete... ee ae A SS. Ruds with radiately ribbed or striate opercula. Peduncle angular or flattened st} Sie Peduncles and pedicels slen- er, S. QOperculum as long as the cupular calyx tube, obtuse .. SS. Operculum shorter than tube, hemispherical and flat- topped he _ soe jJ. Anthers more or less adnate to the top of the filaments; dehiscing by pores, slits or by broad cells opening back to back. Valves narrow and subulate. K. Filaments sharply reflexed in bud and with a bend when flowering, L. Buds and flowers pedicellate LL. Buds and flowers sessile a ty KK. Filaments crumpled in bud, with or without some reflexcd, not bent when flowering. L, Buds and fruits sessile. Leaves erect LL. Buds and fruits pedicellate. Leaves spreading. M. Fruit globose to sub-tirceolale. Opercu- lum mot wider than tube at junction. 20. pimpintana 21. tncrassata 2ia incrassata var. costata Z1b. incrassala var. angulosa 22. cunglobota 25. anceps 24, rugosa 25. dumosa 26. pileata 27. brachycalyx & var. chindoao 28. concinna 29, leptophylla 30, wncimata 31, cucorifolia 142 N. Leaves opposite, sessile or almost so. Plant usually very glaucous. Operculum rostrate ie -- . . NN. Leaves alternate, petiolate. O. Operculum with the same suriace texture as the tube. P. Operculum conical. Q. Leaves shining and glossy QQ. Leaves, buds and fruits glau- cous ... hs on ba ‘ PP. Operculum rostrate, Leaves dull and more or less glaucous .,.. iw OO. Operculum yery obtuse or flat- topped, the surface wrinkled and usually a different colour from that of the smooth tube ; MM. Fruit urceolate. Operculum smooth, wider than tube at junction Outer stamens sterile and longer than Ruds with translucent oil glands, J. Buds and fruits quadrangular IT. inner. JJ. Buds and fruits not quadrangular HH. Umbels paniculate, sometimes uinbels also present. I, Buds 7 x 5 mini, valves of fruit narrow, enclosed in orifice se so simple axillary II. Buds 4x 3 mm., yalyes broad and exsert Su GG. Stamina! ring broad (up to 2 mm.) and conspicuous after flowering, more or less deciduatis, H. Three flowers per umbel. I, Fruit smooth, J. Fruit 7-15 mm. lone He JJ. Fruit more than 15 mm. long If. Fruit subangular, 4-6-ribbed HH. Umbels with 6-7 mm. long three or more flowers. Fruit DD. tise Very inconspicuous and lining the rim of the calyx tube, E. Flowers in axillary ambels which may become clustered or terminal due to lack or loss of leaves. F. Leaves more than 0-8 cm, wide. G. Buds crumpled or wrinkled like a withered apple. Leayes lancedlate to ‘broad lanceolate, the marginal ficeve distant and the leaves often tri-ncrved at the ase, , Ah, —, aut box nig bts 1 GG. Buds more or less angular with angular pcdicels, not wrinkled like a withered apple. H. Fruits more than 4-5 x 4-5 mm. I, Buds sessile or on very short thick pedicels. Leaves lanceolate ot broad-lanceolate, rather thick, Mallee tty orth ston he ne Il. Buds pedicellate, often markedly so. Leaves narrow lanceolate or, in tree forms, broad au HH, Fruits 4-5 x 4-5 mm. Btids very shortly pedicel- late, slightly glaucuos. Leaves narrow lanceolate .. 32, Gillig & var. petiolaris 33. olcosa 33a, oleasa. var. Peeneri 33b. oleosa var. glauca 33c. oleasa var. angustifolia 34. Flocktoniae 35. calycogona & var, Stafforati 36. gracilis & var. erecta 37, intertexta & var, fruticosa 38. coolabah a9. leincosylon 39a. leucoaylon var. widcrocarpa 39b. leucorplon var. angulata 39c, leucoxrylow var. pauperite 40. calcicultrix 41, Lansdawneana 42. odorata 42a. odorata var. angustifolia 143 FE Leaves watrow Linear, less than 0-8 em. wide .... a. = 43, wiridis EF. Flowers in paticnlate umbels. I. Leaves alternate. j t. Fruit smooth ot with one or two striations, mot costate, H. Operculum as wide as the top of tube. T. Buds and fruits pedicellate. Trees, J. Opercutum conicil, x, KR. Fruit larger than $-7 x 4-5 tin. L. Leaves, buds and Irnits glaucous... oo) 6Ada. hem phloia var. adbicits LL. Leaves, huds and freits not glaucous... 44. homibhloin EK. Fruit 5-7 x +5 mm, (sometimes alimost my sessile) sans “ wie Sue apt AdD, Remiphleia var. mmieracarpa TJ. Operculurs hemispherical wi a, ws 45, laryiflorens Tl. Buds and fruits sessile ov very shurtly peuicellate (see also hemi phiéia yar, mécracarpu above). J. Opercilum conical, Fruit stbangular, up ts 10x 9 min, Leeves lanceblate en Al, Lansderneara JJ. Operevium hemispherical. Fruit 3-5 x 3-5 mm. Leaves broadly lanceolate wide as a 46. Behriana HH. Operculuin conical, narrower than tep of tube ... 47. faseivwlasa GG. Fruit iresolate, costate. Operciulum hemispherical- apiculate Sr 10d ta, Ao st ae aw 4B. cladocety.r FF, Leaves opposite te on eur nat set a 4, gamephylla NOTES ON THE SPECIES 1. EucAtyerus Fupesmiowes PF. Muell, in Fragm,, ti, 35 (1860). Dwarf mallee or small tree, Melure leaves opposite, shortly petiolate, lanceolate, subglaucous, 4-6 x 1-2:5 cm. Unibels axillary, three-flowered, peduncles terete. Buds clavate, tetragonous, obtuse, shortly pedicellate, 6x 4mm. Opferculum hemispherical shorter than tube, Stamens in four clusters aliernating with small calyx teeth. Fruit pedicellate, campanulate to urceolate, sinooth or slightly angled, 14 x 10 mm. (after Biakely)- This species, recorded for the Birksgate and Blyth Ranges in the extreme north-west of the State, has no close affmity with any other South Australian form. 2, Evcauyetes pyrivormrs Turez, in Bull. Soc. Nat. Mose., 22, pt. ii, 22 (1849), Mallee with strageling habit. Mulure leaves allernale, petiolate, broadly lanceolate, 6-15 x 24 cm., surtace dull. Uwibels axillary, 3-flawered, peduncles deflexed, thick and short. Buds ovoid or pyriform, shostly perlicellate, 5-6 x 3-4 an, Operciulin conical with a rostrate apex, very thick but narrowing to about 4 mm. thickness at the point of atlachment to the catyx tube, striate, twice as lony as tube. Tube wide and shallow with 7-9 coarse ribs and a few smaller ones. Svaiieis red or yellowish, Fruit broadly turbinate, more than half the height being due to the high disc. Staminal riug preseit as a line separating the calycine ring and disc, the latter erect and with a distinct ridge around the broad, thick, deitoid, partially sunken valves. fruit 4 x 7 cm. (pl. 1, fig. 2). This species, with FE. pechyphylia and E. Oldfieldu, is distinguished by the thickness of the operculum and by the ridged disc which forms a smal] crater- like valley around the valves, These characters are less distinct in E. Old fieldit, which forms a link with LZ. orynwtra, E. Ewartiana and FE. Marvrisis, It is known to occur along the transcontinental railway from Tarcoola westward, 144 3, Evcaryrtus pAcwypiyita F. Muell. in Jour, Linn. Soc., 3, 98 (1859). Mallee, Mature leaves alternate, petiolate, ovate or broad to narrow lanceolate, 6-10 x 2-5 em. Uombels 2-3-fowered, axillary, peduncles tercte, spreading not deflexed, 6-10 mm. long. Buds shortly pedicellate, ovoid-rostrate, 25x 20 min, Operculint conical-rostrate, striate to ribbed, about 2-3 mm. thick at point of attachment, 1-14 times as long as broad, Tube wide and shallow with 3-5 sharp but narrow ribs and with smaller ones hetween. Frits as in E, pyri- formis bat smaller, 15-20 x 20-30 mm,, ribs more rounded in older fruits (pl. i. fig. 3), Description of the bud was taken from Maiden (Crit, Rev, pl. 75, fig. 5-6). This species was collected by Professor J. 13, Cleland at the junetion of Boundary and Fraser Creeks on the border of South Australia and Northern Territory. 4. EvcaLyvprts oxyMmiTxa Blakely in Trans, Roy, Soc S, Aust., 60, 155 (3936). Glaucous mallee. Mafure (eaves alternate, petiolate, broadly lanceolate. thick, 5-7 x 2-4 cm. Uvwibels axillary, 3-7-flowered, peduncles short. Buds shortly pedicellate, ovoid or globular with a hooked-rostrate apex 10-12 = 7-8 mm., glaucous, = Opercudam hemispherical-rostrate or conical-rastrate, 2-3 times as long as wide shallow tube, Fruit pedicellate. glaucous, subglobesc, 11 x 11 mm. dise broad, semiconical, smaller than calycine portion (after Blakely) (pl. i, fig. 1). Frujting specimens used by Blakely have not been seen by the writer, The affinities of this species and also the identity of S. A. White’s Everard Range specunens (se¢ below) depends on whether the disc in this species is ridged around the valves or not. lis occurrence in this State is still in doubt. 5. Evcaryerus Otprizimt F. Muell. in Fragm., ii, 37 (1860), Mallee. Mafure leuves alternate, pdtiolate, narrow to broad lanceolate, 7-13 x 1°5-3 cm. Umbels axillary, 3-flowered, peduncles thick, sometimes very short, Buds globular to ovoid-rostrate, almost or quite sessile, 20 x 10-15 mm. Opercelum conical or rostrate, thick, longer than tube. J'ube smooth, wide and shallow. Fruit hemispherical to turbinate, 11 x 15 mm.; disc portion nearly as large as base, ridge and hollow around valves less evident than in H. pyrtfermus and Z. pachyphylla (mostly after Blakely). There is some doubt a3 to whether this species actiially occurs in Sotith Austraha. Plack (FI. S. Aust.. 418) mentions a specimen from Missionary Plains, MacDonnell Range under BE. Oldfieldit which has since been described by Blakely as , owymitra (see above), The specimen consists of buds and leaves without fruits. Of S. A. White's material mentioned by Black (/c.J, both have leaves and fruits but no buds. In his MacDonnell Range specimen the dis¢ is ridged though rather smaller than is illustrated for E. Drummondii by Maiden (Crit. Rev., pl. 74, fg. 4b), while im the Everard Range material the Eruits have a convex disc, reminiscent of that in FE. Ewartana. Pending collection south of the border the recording of FE. Oldfeldii for this State raust remain doubttul, 6. Eveatyprus cAMALDULENSIsS Dehn m Cat, Pl. Hort. Camald. ed. 2, 20 (1832). £. rostrata Schlecht. in Linnaea, 20, 655 (1847). Medium to large-sized tree. Mature leaves alternate, peticlate, lanceolate, 5-11 x 24 cm. Umibels axillary, 5-10-flowered, peduncles and pedicels slender and lerete, Buds ovoid or diamond-shaped, acute, pedicellate 6-10 x 3-5 mm- Operculym conical and acute or rostrate, longer than the wide shallow tube. 145 Fruit hemispherical or broadly cupular with the erect yalves at least as long as ithe wide base, 6-8 x 5-6 mim,, the valves broadly triangular and more or Jess ipcurved. This widespread species occurs throughout the State. It is associated with watercourses and creeks. 7, Evcaryprus Morrist: R, T, Baker in Proc, Linn, Soc, N,S.W,, 25, 312 (1900), Mallee or small tree, Mature leaves alternate, petiolate, lnear-lanceolate to lanceolate, 6-12 x 12 cm. Umbels axillary 3-6-fowered, peduncles semirerete or terete, 5-10 mim. long. Buds ovoid, sermisessile on short thick pedicels which may be slightly angulay as they pass into the calyx tube, 7-10 x 4-6 mm, with copious dark oil glands. Operculiam conical, yery obtuse, 14-2 times as long as tube, 7'ube obconical, about as long as broad, smooth or with one or two smal) ridges. Fruit truncate-turbinate or ovoid according tu the convexity of the wide disc, valves short and broadly triangular, exsert at the top of the disc (pl. i, fig, 4). With regard to the variation in the development of the dise, Maiden (Crit. Rev., pl. 135, fig. 6) figures the disc as continuous with the valyes. This is always more marked in fresh material than in dried. Matden’s fig, 6 was apparently taken from R. T. Paker’s original plate, but neither atithor states whether the specimen tised was fresh or dry, In material collected by the writer an excellent match with Maiden’s fig. 6a and 6b has been obtained. i inost of the South Australian specimens the buds and fruits are smaller than the dimen- sions given for New South Wales material. In this State the species is fuund in the Flinders Ranges, from Quom northwards. 8. Encaryprus Ewartidna Maiden in Jour. Roy. Soc. N.S.W.. 53, 111 (1919), Mallee. Mature leaves alternate, petiolate, lanceolate, 5-9 x 1°5-2-3 cm., surface dull. Umbels axillary, 3-7-Aowered, peduncles terete, up to 2 emt. long. Bus pedicellate, clavate or globular, 8-10 x 6-8 mm. Operculum very obtuse, being hemispherical or basally truricated-spherical, thick, 1-14 times as long as tube, smooth. Vube wide and shallow, half as long as wide. Fruit pedicellate, globose or clayate due to the very convex disc which provides nearly half the length, valves deltoid and very exsert, 8-10 x 8-12 mm. (pl. i, fig. 8), Despite the fact that Maiden (Crit. Rev., 44, 12D) staled that the anthers placed this species among the Macrantherae, Blakely included it with the Platyantherae, Blakely records this species for South Australia, but gives 4 Central Australian locality—Missionary Plains, MacDonnell Range. Hlack (3) states that the valves have long points (1p to'’4 mim. long), which break off carly. These have not been seen hy the writer, 9. EucaLyprus remota Blakely in Key to Euc., 197 (1934). Mallee, Mature leaves alternate, petiolate, lanceolate, to broadly lancenlate, slightly falcate resulting in better development of the base on one side than on the other, coriaceous, 6-!2 x 1°5-3 cm. Wuithels axillary, 3-8-flowered, peduncles semi- terete, 5-11 mm. long. Buds pedicellate, clavate, 4-6 x 4-5 mm, excluding the sub- angular pedicels. Operculum hemispherical or hemispherical-apiculate, about as. long as the tube. Tube obconical, passing into the pedicel ta form a long narrow funnel, surface wrinkled in dried material. #Hruit pedicellate, pyriform, slightly wrinkled when dried, flat-topped, 6-9 x 5-8 mm., dise slightly convex and rounded jo very slightly coneave but usually Hat, about 1 mm, bruad; valves shorr, in orifice or almost covered by dise (pl. ii, fg. 1), . 146 Blakely placed this species near GB. Consideniana and E. Sieberiana, which hoth belong to eastern New South Wales, Victoria and Tasmania, Llowever, to the writer the relationship with [. diversifelia, which occurs in the same localities as E. remota, is far more marked, especially in the fruits. The buds and slightly asymunétrical leaves suggest a link with Z. ebligua. E. remola is known ag 2 useful indicator plant for a soil type on Kangaroo Island, which is, so far, the only area in Which the species has been found, 10. Kucaryrtus piverssrotia Bonplaud. in Ml, fard. Malm., 35, t. 135 (1813). E. santalifolia F. Muctl, in Trans. Viet. Jnst., 1, 35 (1855), Latge Mallee, Mature leaves alternate, petiolate, linear-lauccolate to broadly laneegiate. Covbels axillary, 3-6-flowered, pediineles semiterete 4-7 mn. Jong. Bids shortly pedicellate or subsessile, shape of two cones attached hase to base, smooth or very slightly wrinkled when dry, 7-10 x 4-6 min. Opercitunt conical or conital-tosirate, about the same size and shape as the combined tube and pedical, 4-6 mm. Jong. Tube obconical, 3 mm. long, Fruit sessile or very shorily pedicellate, hemispherical to campanulate or abconical, 8-12 x 10-15 mnz., dise 1°5-2 mm. broad, slightly convex and tounded or fat; valves broad and triangular in orifice and slightly more exsert than in Z. remota, so that the short points protrude above the disc (pl. ii, fig. 3). When buds are lacking it iz noi easy to distinguish between Lf. diversifolie arid 2. vermota, but the fruits of the former are larger, have a wider dise and are always niore shartly pedicellate than those of the latter. This species occurs as a coastal or subcoastal mallee from the west coast ob Eyre Peninsula to the Victorian border, including Kangaroo Island, 11. Eucanyerus Baxtert (Denth,) Maiden and Makely in Crit. Rev., 70, 451 (1928). FE. sanrtalifelit yar. (?) Bartert Benth. in Fi. Austral. 3, 207 (1866), Small to medium-sized tree, Mature leaves allernate, petiolate, ovate to broadly lanceolate, asymmetrical at base. Umibels axillary, 5-9-flowered, peduncles stont 4-6 mm. long. Buds obovoid on short thick pedicels, 4-5 x 3-4 min. Operculum hemispherical, slightly shorter than the obconical or hemispherical hibe. Itrvit sessile, subglobular due to the convex disc which is 2-3 mm. wile, 9-12 x 10-13 inm., arranged in tightly packed globose clusters on the short stout peduncles, valves ineurved, in orifice or slightly exsert (pl. iil, fig. 1). lfa. FE, Baxrrrt yar. PRDICELILATA Maiden and Blakely in Crit, Rew., 70, 457 (1928), Ditters from the above in that the fruits are shorlly pedicellare. The opercula ave yertuense. The species occurs in the Mount Lofty Range and southwards to the South- East and the Victorian border, also Kangaroo Island. The variety is known to occur in the South-East and may be found in other parts of the above districts. 12. BucaLyrrus RUBIDA Deane and Maiden in Proc. Linn. Soc. N.S.W., 21, 156 (1899), _ Medium to large-sized tree. Mature leaves alternate, petiolate, lanceolate. Unibely axillary or lateral on leafless portions of the branchlets, 3-fowered, peduncles semiterete or slightly fattened, Auds shortly pedicellate or almusl sessile, ovoid or cylindrical, 5-7 x 3-5 tim. Opercului conical, obtuse. Tube cylindrical, slightly Jonger than or equal to the operculum, Fruif sessile or shortly pedicellate, obovoid or turbinate, 5-6-x 6-7 mm, ; dise concayt, valves deltoid and exsert. 147 The South Australian material seen was less glaucous than as described by Blakely. It appears to lie between /. elecophora and E. viminalis among our species. It occurs in the Mount T.ofty Range. 13. Eocacyprus vimrnauis Labill in Nov. Holl, wi, 12, t, 151 (1806). Large tree, Mature /eaves alternate, petiolate, linear-lanceolale to lanceolate, 10-20 x 1-2-3 cm, Umbels axillary, normally 3-flowered, peduncles subangular, 3-6 min, long. Buds obovoid with pointed apex and short usually thick pedicels, 6-10 x 3-5 mm. Opercilwan conical to conicalrostrate, equal to or slightly longer thai the obconical tube, at its broadest somewhat wider than tube. J'ruit sessile or very shortly pedicellate, turbinate to subetobular or pyriform, 6-8 x 6-9 mm. (including valves); dise convex, coniributing to the length of the fruit, aboul 1 mm. wide, valves exsert and erect, sburt and broad, Occurs in the Mount Lofty Range and southwards to Mount Gambier and the Victorian border. 13a. Fucatyprus viminaris var. kAcEMos* (T°. Mueil.) Blakely in Key In Fouc., 162 (1954). In this: variety the umbels are arranged on short leatless branches. Blakely states that the umbels are 3-flowered, but South Australian specimens agrecing excellently in other respects with the figure (Crit. Rev., pl. 118, fig. 9) have multi- flowered umbels. There is no rcferenee m the text of the Critical Revision to this variety, and Blakely’s note appears ta be the only description. The variety occurs inthe South-East of the State, 13b. Evcanyverus vias Ais var. Hunertawa (Nandin) N, ‘T. Burhidge stat, nov.; E. AunéxiAwaA Naudin in Second Mem,, 42 (1891), &. Huberiana was separated from £. tuminalis by Wakely, though Maiden (Crit. Rew, 28, 175) considered it merely a synonym and remarked that it was always possibie to find multifowered umbels on trees with mostly 3-Howered ones- E. Huberiana differs in timber value, rougher bark, form of growth and in the multilowered unibels. Recent specimens, collected mear Adelaide, showed some branchlets with 3-flowered umhels and some with four or more flowers, This supporis Maiden’s observation and there seems to be no character which can be used for differentiation at (he specific level and Ihe form is herein given varietal rank (pl. ii, fig, 2), Naudin’s species was based on a single tree growjug under cultivation at Nice. This is a practice which is open to criticism. Many Euca- lypts produce an atypical growth under horticultural conditions, and also rhere is no evidence concerning the origin or purity of the seed used. Distribution is the same as in A. wiuindlis, also on Mangaroo Island. is. Evearyetus vitrea R,'T. Raker in Proc. Linn, Soc, N.S,W,, 25, 303 (1900). Medium to large-sized tree. Jfalure leaves alternate, petivlate, narrow to broad lanceolate, the venation very oblique and approaching semilcngiludinal, the tail) nerve some distance from the edge and ircquently there are minor nerves outside, 10-15 x 15-2 cm. Umbels axillary, singly or it pairs or in short axillary panicles, 7-12-flowered, peduncles terete or semitercte, 7-10 mm, long, Buds on relalively stout pedicels which pass inta the tube without marked change, clavate, 5-6 x 4:3 mm. Operculwim hemispherical-apiculate or conical-acute, as long as tube. Tube obconical or cylindrical, /reit pyriform, shortly pedicellate, 5-6 x 6-7 mm. dise 1-2 mim. broad, convex or almost flat, covermg the short enclosed valves. In the south of the State. Specimens from Waitpinga, Kalangudoo, Mount Gambier and other South-East localities have been sten, 148 15. Eueatverus patcircora Sieber in Spreng, Syst. Cur. Post, iy, 195 (1827), Small to large tree. Malure leaves allernate, petiolate, lanceolate, venation longitudinal or almnst so, 8-14 x 13-3 cm. Uibels axillary, 5-12-floweredl, visually about 9, peduncles thick, terete or compressed 7-15 mm. long. Buds clavate, pedicellate 5-7 x 4-5 mim, Operculum hemispherical, obtuse, shorier than tube, Fruit shortly pedicellate, globose to pyriform 7-10 x 5-8 mm: disc flat bul rather prominent, extending over the enclosed valves (after Blakely). This ig a vare species in South Australia. No specimens were available during the preparation of the above description, It is reported to aceur in the Mount Gambier district. 16. Evearyprus optiqua L'Herit. in Sert, Ang. 18, 20 (1788). Large tree. Afatwre frawes alternate, petiolate, lanceolate, asymmetrical at base and slightly faleate, Umbels axillary or sometimes arranged in short racemes or panicles, peduncles terete or semitercte, 7-20 rin. long. Byds on long, relatively thick pedicels which pass without sharp increase in size into the calyx tube, 10-12 x 4-5 mm. with pedicel. Operculmn hemispherical-apiculate, slightly wider than tube at junction, shorter than obconical tube. Fruit pyriform ar ovoid, 7-9 x 7-4 mm; dise oblique, ie., concave, covering the short enclosed valves. In some South Australian specimens the dise is almost flat. Ii occurs from the Mount Lofty Range near Adelatde to the South-East, and is also found on Kangaroo Island. 16a. EUCALYPTUS OBLIQUA var, MEGACARPA Blakely in Key to Euc., 194 (1934), This can be distinguished from the above by its coarser buds and larger fruits, 10-12 x 10-12 tmm.. which are more definitely constricted at the orifice than in the typical form (pl. ii, fig. 2). The type material of the variety, which came from Millicent, has been seen at Sydney atid in the collection of Mr, J. M. Black. In other specimens, [rom Eight Mile Creek, in the Waite Institute Tlerharium, the buds are often clavate and swollen, a maliormation appatently due Lo some gall-lorming insect. 17. Kucacyprus ovata Lahilf, m Noy., Holl, pl. ii, 153 (1806). Large tree. Muaivre leaves alternale, petiolate, lanceolate to ovate-lanccolate or ovate, 7-l4 x 2-4em. Umbels axillary or on leafless portion of the branchlets, 4-8-flowered, peduncles terete or semiterete, 5-10 mm. long. Buds pedicellate, smooth, obconical-rostrate or clavate or like two cones attached base to hase, 7-10 x 46 mm. Operculuin conical, shortly acute or rostrate, shorter than or almost equal to the tube, sometimes with a sharp ridge marking the junction with the obconical tube. Frat! broadly obconical, 5-8 x 5-7 mm., the calycine and staminal rings forming a flat band or dark line around the ottside of the rim; disc rounded, half to one mm. broad, valves short and broad,in orifice (pl. iit, fig, 3). Oceurs in the Mount Lofty Range and south to Mount Gambier and to Victor Harbour, Also on Kangaroo Island. Ifa. EucaAyerus ovata var, GRANDTHLORA Maiden in Crit, Rev., 27, 146 (1916), Buds (15 x 9 mm.) and fruits (10 x 15 mm.) larger than above, the forme) with rostrate operculum and the above mentioned ridge, the latter broadly obconical and slightly constricted below the rim.. Kalangadoo and Mount Gambier, 18. EvecALyptTus ELAEOPHORA F. Muell in Fragm., iv, 52 (1864). Mediiun to large-sized tree, Mature leaties alternate, petiolate, lanceolate 16 broadly lanceolate, 10-22 x 1-5-3 cm. Unrbels axillary, sometimes in pairs, 3-7- flowered, peduncles flattened and angular, 3-12 mm, long, Buds sessile or sub- lay sessile on thick angular base almost indistinguishable from the tube, cylindrical, more or Jess bicostate or angular, 8-10 x 4-6 mm, Opercitlum conical, shorter than tube. Frit sessile or very shortly pedicellate, cylindrical or obconical with a thick more or less angilar base, otherwise smooth, 6-9 x 6-8 mm; disc less than 1 mm. broad, Forming a narrow rim around the short broad valves which are exsert or in the orifice (pl: iii, ig. 4). Blakely apparently had no South Australian material at Sydney except the poor specimen which was included under £, yontvealys, so reference to this State was ontitted on p, 146, but it was included in the list of species oceurring in South Australia on p. 289. _ Southern Flinders Range, Mount Lofty Range and southern parts of the State, includiny Kangarou Island, 19. Evearyrrus cosmopmycuaA I’, Mucll. in Trans, Vict. Inst., 1, 32 (1855). Varying from a small mallee-lke shruh to a medittm-sized tree, Matnre leaves alternate, petiolate, usually thick and coarse, lancealate to broadly Tanceo- late, 10-15 x 1-4 em, Usnbels axillary, 1-3-flowered, peduncles very short and sometimes lacking, Buds vbovaid, bicastate on very short flattened pedicels, 14-18 x 9-12 mm. Operculum: conical, conical-apiculate or rostrate, shorter than or eqnal to tube and sometimes wider. Filaments cream or pinkish coloured. Fruit hemispherical, turbinate or shortly cylindrical, more or tess bicostate, 10-17 x 12-18 mm., the calycine and staminal rings forming a sloping outer cdge to the rim, with the steeply oblique dise lining the inside around the 4-6 short, broad but acute valves which are just below the orifice (pl. iii, fig. 5), Blakely states that the umbels aré 3-6-fowered bul none with more than three flowers have been seen by the writer. Mueller’s original description was “tri-Horis.” Neither af Mueller’s variatians, included by Blakely (p. 103) are sufficiently distinct tn warrant separation, The figures. ia Critical Revision (pl. 91, fig. 4-5) show only slight deviation from the normal, and recent collect- ing does not indicate any real differentiation, Occurs in the Mount Lofty Range and south to Encounter Bay and ‘an- garoo Island. 20, Evucaryrrus PrmriniAna Maiden in Crit. Rey, 16,211 (1912). Mallee. Muture leaves alternate, petiolate, lancealate to broadly lanceolate, 6-9 x 2-4 em, Uimeels axillary, 3-8-flowered, peduncles terete or scmiterete, deflexed, Buds obconical or cylindrical, rostrate, pedicellate, striate, 20-30 x 10-14 mm. Operentim hemispherical-rostrate or conical tostrate, smooth of faintly striate, Frat pedicellate, cylindrical and gradually narrowing into the pedicel, striate, disc as described below for £. incrassata, valves with short points; deeply enclosed (pl. iii, fig. 6). Occurs on the eastern border of the Nullarbor Plain at Ooldea and Barton. 21, Kecatyyrus monassara Labill in Nov., Holl, pl. it, 12, t. 150 (1806), sens stricto, Mallee, Mature leaves alternate, petiolate, nartow to broadly lanceolate, 612 x 1'5-3.cm. Umbels axillary, 3-7-flowered, peduncles slender, terete or semi- terete and slightly angular, up to 3 cm. long, Buds cylindrical-ohovoid to ureeolate- rostrate, shortly pedicellate in the former case to long pedicellate (up to 6 mm.) in the latter, 10-18 x 4-7 mm, Operculum-conical to very trostrate with a more or less hooked apex, commonly somewhat wider than the calyx tiibe and shorter, smooth or with obscure fine wrinkles or with promiuent lines but not distitictly ribbed. Yvbe obconical or constricted in the middle and very urécolate, smooth to striate with fine lines or chscure nibs. Frith pedicellate or sessile, ureeolate or barrel-shaped being constricted at each end, smooth or slightly wrinkled when 130 mature, LO-12 x 9-10 mm.; the tube wider than the disc, so that the calycine ring forms a narrow but distinct platform around the erect rather sharp disc whose outer face is formed by the old staminal ring and whose inner is oblique to almost erect; valves with subulate points due to the persistence of the base of the style, when open either in orifice or, more asually, below and enclosed (pl. 1, fig. 8-9). When immature the top of the ovary is ridged between the unopened valves. This last character is found im the remaining varieties below, and also in the members of the dumosa complex, Occtits on Yorke Peninsula, in the Flinders Range, the Murray Mallee ane. south to the Upper Sourh-East. It has praved impossilie to make a satisfactory demarcation between this species und B. angulasa Sehaucr,, a wide range of inlermediates haying been enllected. The specimens have therefore heen grouped as follows. [i is probable that these represent mere points in a scries, and collectors are warned that inter- mediales may be fond. 2la. EucanyrTus wncrAssara var, costata WN. T. Burbidge stat. nov: E. costata Behr. et. Muell. in Miq. Ned. Kruidk, Archief., 4, 136 (1856). A coarser form than the above. Mature leaves fairly thick and coriaceous. Linbels un slender flattened pediincles (up to 3 cm. long), which broaden below the Aowers. Buds similar to the above. hut more definitely ribbed or striate, Opercntis as above, -Pviit on short, more or less angular pedicels, barrel- shaped or cylindrical, marked by striations or sharp narrow ribs, 10-18 x 9-12 mm,; dise as above, valves enclosed (pl. ii, fig. 10). ‘Chis is the FE, angulosa of various writers on South Australian ecology (10) (19). J. M. Black combines it with the following, as did Maiden. The striation or ribbing of both the bud and the frnit is far less pronounced in fresh material, The ribs ure revealed on drying, due to shrinkage of the less sclerised portions. Maiden (pl, 14, fig. 2) shows a fruit rather more coarsely ribbed thaq the majority ot specimens, but it can still be contrasted with that in his fig. 3, which is the form described below. Ocetirs on Eyre and Yorke Peninsulas and Kangaroo Island; alsa in the Murray Mallee and the Upper South-last, 2lb. Kucanyvrres Inceassat var. ANouLOSA Beith. im Fl, Atst,, iii, 231 (1866), pro parte; 4. angulosa. Schaner, in Walp. Repert., 1, 923 (1843) sensi stricta, A very coarse form with thick lanceolate leaves. Umbels 2-5 flowered, peruneles. thick, flattened, 10-20 mim. tong. Brds obovate or cylindrical with short pedicels, 20-25 x 10-15 mm, Operculum conical or shortly rostrate, striate or ribbed, shorter but not wider than tube. Y#be obeonical or cylindrical, coarsely and deeply ribbed by projecting hard ridges which continue into the slightly flattened and ribbed pedicel. J’raf almost cupular to cylindrical, coarsely ribbed by the abovementioned ridges, 15-20 x 14-18 mm.; dise and valves as in other forms above (pl. ii, fig. 11). This coarse form agrees with Maiden’s illustration (pl. 14. fig. I, 3, und 5), ie., with Western and South Austraiian inaterial, Tn this State it is mainly found on Eyre Peninsula. ‘ THE DIUMOSA COMPLEX This group of forms has always heen a difficult one. The type of EZ, dumasa was collected by Allan Cunningham in New South Wales. Specimens in the Sydney Terbanuin, collected at Wyatong., which is accepted as approximately the type locality, proved ta have cylindrical buds which were smooth ar almost sn, 151 and to haye a conical operculuin shorter than the calyx tube. This form is rare in South Australia but occurs along the eastern border and near the Murray River. In this State most of the material classed as E. dumasa has buds with ribbed conical upercula. ‘This agrees with the Western Australian species, E. pileata. There are intermediates in the eastern part of the State, but on the whole the two are reasonably distinct, There is also L. brachycalyx with an obtuse, conical op¢reulum which is striate rather than ribbed. On Kangaroo Leland there are a number of forms, “he most distinctive is E. conglobata, in which the buds are so sessile as to be flattened at the base and the calyx ite is wider than long, There is also Z. rugosa, where the bud has a flattened pedicel and the tube is 2-4-ribbed., Unitortunately, the whole group is complicated by the description of H, anceps. Maiden described this as a variety of HE. conglobatu, bm piaced it intermediate between /. conglobatu dnd E. dumosa. It differs from the former in that the buds and [riits, though sessile, are not basally flattened and the fruits are barrel- shaped and from E. dimosa in a relatively shorter calyx tube and in the lack of pedicels. Compared with true A. duimosa this diagnosis would hold, but in practice E. anceps lies between E. pileaia, F, rugosa and EB. conglobata, aril as such it includes a heterogeneous collection of variations, Presence or absence of pedicels is nota good diagnostic character, and sessile buds and fruits of L. rugosa wre difficult to place; the same applies to £. pileata, and apart from individualistic variations such as these there are probably a number of hybrids and ecotypes. Another species related to the group bit more casily recognised is EZ. concinna. An effort to reduce some of the forms to varictics proved unsatisfactory and they were retained as species to avoid adding to the existing confusion. 22. Fucanyerus concurezara (R. Br.) Maiden in Crit. Rev., 6, 273 (1922). Matlee. Mature [eazes alternate, peliolate, thick, narrow to hroad Janceolate, 5-14 x 1-3-2-5 em, Conbels axillary, 3-8-flowered, peduncles short, thick, more or less angular, broadened at the top to form a receptacle, Buds squat, broadly ovoid, flattened at the base and closely sessile, 8-10 x 6-8 mm. Operculium shorbly and broadty conical, obtuse or acute, smooth or shghtly striate, as long as or sightly longer than tuhe, Yube shortly cylindrical, bicostate, about twice as wide as long, Fruit cylindrical or hemispherical, 2-3-ribbed, 5-6 x% 8-10 mm_j disc narrow, valves with subulale points which project above the rim (pl, ait, fig. 7). Iéyre Peninsula and Kangaroo [stand, 23. Lucaryrrces ances (R, Br.) Blakely in Key to Euce., 118 (1934), Mallee. Juvenile Ivaves opposite, sessile, slightly lobed at base as to be almost stem clasping, ovale ty broadly lanceolate, Matire leaees alternate, petiolate, lanceolate, rather thick and rigid, 5-12 ~ 1-2-5 em. Unbels axillary, 3-6-Rowered, peduucles relatively thick, fattened or subangolar, 6-12 mm. long, Buds sessile or almost sa, cylindrical or ovoid, narrowed at both ends, 810 x 4-6 mm. Operculum conical, obtuse. acute or shortly rostrate, radiately striate, about the same length as the tube or slightly shartar, ruff sessile on thickened peduncles, cylindrical to barrel-shaped, striate, 7-10 x 6-8 mim; + disc marrow, valves in orifice, short but subulate (pl. ili, fig. 8). Eyre Peninsula, Yorke Peninsula, Murray Matlee, Upper South-East and Kangaroo Tsland, 24. Evcaryptus rvaoss (R. Be.) Blakely in Key ta Tue., 120 (1934), Malice. Motire loaves narrow to broadly lanceolate, 3-10 x 1-2 cin. Unibels axillary. 3-8-Alowered, peduneles stout, flaitened, broadened below flowers, becam- ing coarser during toaturalion of fruits, Buds pyriforn on short, stout, fattened, 1$2 3-4-angled pedicels, 6-7 (9-12 with pedicels) x 5-6 nm. Operculum hemispherical, flat-topped or very obtuse, smooth or slightly striate, a litthke narrower than the ribbed tube or almost equal, also shorter. /ruit pyriform or turbinate, ribbed with 2-4 main ribs two of which are broadened below into the flat pedicel, and some- tines with a few striations (pl. 1, fig. 10). Eyre Peninsula, Kangaroo Island, Encounter bay and the Upper South-East. 25. Eucaryerus pumosa A. Cunn. fide Schauer in Walp, Rep. Bot, Syst., 2, 925 (1843). Mallee. Mature leaves alternate, petiolate, lanceolate, 6-10 x 1-2 em, Uimbels axillary, 4-8-flowered, peduncles comparatively stout and semiterete. Buds smouth, eylindrical, shortly pedicellate. 7-12 x 3-3 mm. Operculun hemispherical to shortly conical, minutely striate or almost smooth, much shorter than the eyindrical tube, J’ruit cylindrical to campanulate, somewhat wrinkled or striate, shortly pedicellate, 6-10 x 5-7 mm.; dise narrow. valves with protruding subulate points (pl. iii, fg. 9). A New South Wales species which extends to the eastern border of this State and down into the Myrray Mallee. 26, Eucaryprus PiteaTa Blakely in Key to Euc,, 120 (1954), Mallee. Mature leaves alternate, petiolate, narrow to broad lanceolate, 55-10 x 1-3 cm. Uimbels axillary, 3-6-flowered, peduncles terete or semiterete, 6-15 wm. Jong. Buds pedicellate, obovoid or cylindrical, 8-14 x 4-5 mm. Qpereulyan hemispherical to acutely conical, radiately striate or ribbed, sometimes wider than tube, shorter than or equal to the cylindrical or obconical or cupular tiibe, which is smooth or striate. Jruil pericellate, cylindrical or obconical, striate or almost smooth, 7-10 x 6-8 mim.; dise narrow, valves sunken with protruding subulate points (pl. ai, fig, 11). Eyre Peninsula, Yorke Penitisula, Murray Mallee, Encounter Bay, and Kan- garoo Island. Forma, Obovgid pedicellate buds with almost smooth operculum. No fruits available, This form seems distinct enough to warrant mention, but without. fruits and further information it dacs tot scem strong enough to make a variety. It oecurs on Eyre Peninsula. 27. Tucanyrtus psrachycaLtys Blakely in Key to Euc,, 119 (1934). Mallee, Mature leaves alternate, petiolate, linear to narrow Tanceolate, shining, 5-10 x O'7-1'5 cia, Uvtbels axillary, 3-9-Howered. peduncles slender, slightly flattened, Buds avoid-clipsoidal, on slender pedicels, 7-9 x 4-3 mm. Opereuiuue conical, obtuse, about the saine size and shape as the cupular, striate tiibe, asually with irregular wrinkdes rather than striate, commonly slightly different m calour fram the tube. Frat hemispherical to cainpanulate or turbmate, sinooth or faintly striate, 6-7 x 5-6 mm.; disc narrow, less than 0'5 mm, wide; valves subulate and exsert except where the brittle points have been broken off (c.f. Blakely) (pl, ii, fig. 12). This species could be mistaken for #. oleosa vat. angustifalia, but it can be differentiated by the large obovate versatile anthers with a conspicuous gland and, in fruit, by the turbinate open capsule (nat globular as in LE. alevsa) and by the spreading rather than erect valves, Eyre Peninsula, Flinders Range and to the east theres! along the border of the shrub steppe and mallee scrub. 27a. EvcALypTus BRACHYCALY® var. CHINDOO Blakely Lo Buds somewhat sinaller thait those above, 5-6 x 3-4 mm. Operculwmn striate and very obtuse, slightly longer than substriate eupular tube, alse slightly narrower than (ube giying a marked “ege-in-ege-eup” appearance, Fruit as above, 154 It is doubtful whether this variety is a true one, The species itself 3s quite distinct and may be recognised by the slender peduncles anc pedicels to the wnbels, the cupular calyx tube about the same size and shape as the operculum and by the narrow shining leaves, Jt was untortunate that Blakely cottiradicted his owit description of the operculum. Distribution as above, 28. Eucatypres concinnaA Maiden and Blakely in Crit. Rev., 71, 49 (1929) ; E, ochrophylla Maiden and Blakely in Crit. Rev., 71, 50 (1929). Mallee, Mature leaves alternate, petiolate, lanccolate, broad or narrow, thick and glossy. Umbels axillary, 3-7-flawered. FPeduncles terete or semiterete, 612 min, long. Buds clavate, pyriform or cylindrical, 8-10 x 6-8 mn, on pedicels like the peduncles and 4-12 mm. long. Operculum hemispherical, very obtuse or flat-topped, about half as long as broad, more or legs striate, sometimes slightly broader than tube. Stamens red (type from Victoria Desert) or pale (Cleland’s specimens quoted under original description of FE, ochrophylla), Fruit pyntorm, almost turbinate or cylindrical, smooth or faintly striate, 6-10 x 6-10 mm; disc about 1 min. wide, rounded above with the dark line of the calycine ring outside, capsule slightly sunken but the subulate valves protruding and spreading or rarely coherent at top as in Z. olvosa (pl. ili, fig, 13). Cleland’s specimens of E. ochroplylla show a definite variation towards &, concinna, and the buds cannot be separated in any particular character, The type of &. concinna has very long pedicels (matched among L. ochrophylla material) and the buds are golden-brown, whereas in 2. ochrophylla they are greenish. The yellowish colour of the leaves in the latter speeics is not a reliable character. Occurs along the transcontinental railway at Ooldea and Immana, 29. Eucanyerus teeroritvie.a FL Mauell. in Miq. Ned, KRroidk. Archief., 4, 123 (1856), Mallee, Mature leaves alternate, petiolate, linear or narrow lanceolate, with numerous oil glands, 5-10 x U-6-1-5 cm, Umbels axillary, 3-8-flowered, peduncles 5-8 nim, long, terete or semiterete. Buds like two more or less equal cones attached base to hase, on a slightly angular pediecl as thick as. the peduncle, 5-7 x 3-4:nm. = Ofercoulwii conical, actite or obtuse, abottL as long as tube or shorter. Stamens sharply reflexed in bud and with a sharp bend in the filament when flowering. Fruit pediccllate, hemispherical to cupular, 4-5 x 4-5 mm; disc forming a flat or rounded riny at the top of the fruit; valves narrow, subulate and protruding through the orilice (broken off m old fruits) (pl. iv, fig. 12). Alone the transcontinental raihvay at Ooldea and Tarcoola, also livre and Yorke Peninsulas, Flinders Range, Murray Mallee and Kangaroo Island. 30, Evcaryprus uwcryAta Turez. Pull Soc, Nat. Mosc., 22, 23 (1849), Very close to the preceding species bul buds atid fruits somewhat coarser. The chief difference is in the bud, which is sessile or on a short thick pedicel indistinguishable from the tube, Operculiwn conicul, obtuse and shorter than the obconical or cylindrical tube 6-8 x 3-4 mm. J’ruit sessile or almost so, pyriform or barrel-shaped, 5-6 x 4-5 mm. (pl. iv, fig, 13). These two species were united under the latter name by Bentham in the Flora Australiensis, but were separated by Maiden. It would be difficult to separate them without buds, though the slender pedicels and yellow-red branchlets in E. lepiaphylla ate characteristic. Fyre Peninsula seems to mark the eastern limit of this Western Australian BDecies, ' 154 31. Eucaryrrus cnrortrotra DC., Prodr., 3, 220 (1828). Mallee. Malure leaves alternate, petiolate, narrow linear to linear lanceolate, 5-11 x 0-4-1 cm. [/mbels axillary, 6-15-flowered, peduncles short. Buds cyliny drical, shortly pedicellate or sessile, 7-12 x 3-5 mm, Operculwm smooth, cylin- drical to conical, obtuse, about the same length and shape as the tube. Fruit shortly pedicellate or sessile in densely packed semiglobular clusters, hemispherical, 5-6 x 7-8 mm.5 disc flat, about 1 mm. broad, around the partially exsert, subulate but short valves. Kangaroo Island and Encounter Bay. 32, Eucatyprus Gittir Maiden in Crit. Rey,, 15, 177 (1912). Glaucous mallee, Muaturve leaves opposite or almost so, glaucotis, sessile or shortly pedicellate, ovate, oblong or shortly lanceolate, 4-6 x 1-2°5 cm. LU mubets axillary, glaucous, 4-9-flowered, peduncles scmitercte, slender. Buds glaucous, ovoid or cylindrical-rostrate, pedicellate, 8-15 x 4-7 mm. Operculun conical, acuminate or rostrate, longer than the cylindrical tube. #rutt ovoid or globular, pedicellate, 5-8 x 5-8 mm.; disc very small, valves exscrt, long and subulate but usually broken off short. Northern Flinders Ranges. 32a. EucAryetus GiLtw var, PETIOLARIS Maiden in Jour, Roy. Soc. N.S.W,, 53, 59 (1919), Differs from the above only in that the leaves are definitely petiolate (up to 1 cm-_) and broadly lanceolate. It is doubtful whether this form is worth retaining as a variely, The type came from Wirrahara, but other specimens from the northern Flinders Range have been seen, 33. Eucayprus oLeosa I’, Muell. in Mig, Ned. Kruidk. Archief,, 4, 128 (1856), sensu stricto, Mallee or small tree. Mature leaves alternate, petiolate, narrow lancvolate, usually glossy, 5-I0 x 1-2 cm. Umibels axillary, 5-14-flowered, peduncles semi- terete, Huds pedicellate, ovoid or cylindrical, Operculum conical, longer than cupular or semiglobular tube. J’*rwit pediccilate, globular or clavate 5-9 x 5-9 mm.; disc nafrow, valves thin, long and subulate, very exsert, coheretit at the apex Lut breaking casily and henee often lacking from specitnens (pl. iv, fg. 4), A photograph and some fragments of the type material have been obtained through the courtesy of Dr, F. P, Jonker of the herbarium of the State University at Utrecht, Holland. It is mixed FA. wneinuta Turez. (buds and flowers) and &. oleosa F. Muell (immature fruits). The latter, which is regarded as the true type, was, according to the Melbourne Herbarium authorities, collected from the Murray Mallee, Whet inquiries concerning the type were first made it was hoped that the specimen would show the shape of the operculum, since this feature is variable and hag heen the basis of diiterentiation in several varictics, Unfor- tunately, this is still unseltled. It is believed, however, that the specific name was assoctited with a form having a conical operctilum, since Mueller’s species, £. sociths, published in the same paper. has a rostrate one. One of the most important and widespread mallee species, as can be seen by reference to Wood’s vegetation map (10), 33a, EvcALyrtus oLzosa var, Peenerr Blakely in Key to Euc., 270 (1934). Leaves, branchlets, buds and fruits more or less ashy-grey. Leaves oblong to oblong lanceolate. Operculum obtusely conical, twice as long as cupular inbe. This may not be a good variety, Blakely founded it on material from Ooldea, and specimens probably belonging to it have been receiyed from Fyre Peninsula atid Kangaroo Island, 135 33b. EUCALYPTUS OLEOSA var. GrAuCA Maiden in Jour. W. Aust, Nat, Hist. Soc. 3, 171 (1911); E. socialis F. Muell, in Miq. Ned. Krujdk, Archief., 4, 132 (1856); FE. transcontinentalis Maiden in Jour. Roy, Soc. NSW, 53, 58 (1919), Leaves lanceolate to broad lanceolate, usually dull or subglaucous. Buds cylindrical- or urecolate-rastrate. Operculum rostrate and longer than cylindrical ot urceolate tube. Fruit narrowed into orifice and sometimes with a tendency to he urceolate (pl. iv, fig. 1). The nomeniclature of this form was confused by Maiden, In Crit. Rew, 15, 167, he redeseribed var. glauca and listed £. soctalis F, Muell. as a synonyi. Later he quotes his description of E, trancontinentalis (Crit. Rev., 34, 268) with E. oleosa var, glauta asa synonym, but there iz no mention of £. socialis, Seth Ausiralian material matched with Maiden’s specimens in Sydney is identical with the type of E. sociulis [rom Mueller’s material in the Melbourne herbarium, It does not seem to be distinct enough for specific rank and forms a link between E. oleosa and E. Flocktoniae. It would appear that Blakely’s description of i. socialis was superfluous. This form occurs from the Nullarbor Plains eastwards across Eyre and Yorke Peninsulas, along the Flinders Range to the north and also east to the Murray Mallee, the Victorian border and Kangaroo Island. 33c, EucALyrTug OLEOSA var, ANGUSTIFOLIA Maiden in Crit, Rev, 39, 278 (1919), Leaves narrow to broad Ianceolate, asually glossy. Buds pedicellate, cylin- drical or ovate, Operculunt crumpled or wrinkled on the surface, cottical or cylindrical, very obtuse or flat-topped, different in colour from the tube, Fruit globular (pl, iv, lig, 2). When Maiden described this variety he gave #£. socialis F. Muell, as a synonym and figured (Crit. Rey, pl, 65, fig. 17 a-b) a Murray River specitnen collected by Mueller. This specimen is referred to as “the type,” and then as “a reptted type specimen from Mueller,” but it is not cleat whether he means it as a type of Z. secialis or of his var. angustifolia, There is, however, no mentiot of such a specimen under the original description of E. soctalis, so that B. socials is nota synonym of this yaricty. Blakely gave Mueller's species full ranking, but retained E. oleosa var. angustifolia and quoted a specimen from Pinnaroa, collected hy J. M. Black, as the type, Subsequent to the publication of the Key, Hlakely appears to have decided that E. oleusa yar. angustifalia should be given specific rank. There is a Jarge amount of material in the Sydney Herbarium which he separated out under the name E. lamprophylla. This. includes Black’s Pinnaroa specimen. The species was never published, it is a distinct variety but, without the characteristic buds, difficult to separate from the rest of E. oleosa. Since there is doubt concerning the bud shape in the original E. ofeosa, specific ranking for this form would hardly be advisable. Mueller’s specimen, mentioned by Maiden, is presumably the type- Distribution is the same as for E. vleosa seusw stricto. It is particularly common on Eyre Peninsula and Kangaroo Island. 34, Kucanyrrus FrocxTontae Maiden in Jour, Roy. Soc. N.5.W., 49, 316 (1915). Mallee or small tree. Mature leaves alternate, petiolate, lanceolate, 6-12 x 1+5-2°5 cm. Usmbels axillary, 3-7-flowered, peduncles semiterete, broadened at lop but nol flattened, Buds urceolate-rostrate, pedicullate, 12-17 x 5-8 mm. Operculium with broad base, hemispherical or conical and with a lofig beak, 136 up to twice as long as the urceolate tube and broader at the junction. Frutét yrcevolate, broad at base and constricted below the wide orifice; disc arrow, valves as in &, oleosa but very brittle and usually broken off and hence appearing enclosed, 8-10 x 7-2 mm.; pedicels angular (pl. iv, fig. 3). This species forms a link between EF. aleasa and the Western Australian species H, torquata. It occurs from the Nullarbor Plain across Eyre and Yorke Peninsulas. 35, Eucaryerus cArycocona Turcz. in Bull. Phys. Math. Acad, Petersb., 10, 338 (1852). Mallee. Mature leaves alternate, petiolate, linear to lanceolate, 5-10 x 0-7-2 cm. Unmobels axillary, 3-7-flowered, peduncles slender, terete or yemiterete. Buds clavate, quadrangular, pedicellate, 6-10 x 4-6 mm.; with numerous oil glands, Operculum shortly conical, obtuse, acute or apiculate, not angular, shorter than the angular tube. Frat cylindrical or urceolate, quadrangular but less sharply so than when in flower, smooth, pedicellate, 8-12 x 5-6 mm.: dise lining the top of the tube and obscured by the narrow persistent staminal ring; valves small and deeply enclosed (pl. iii, fig. 15). Eyre and Yorke Peninsula, Murray Mallee and upper South-East, 35a, EUCALYPTUS CALYCOGONA var. StArrorpi Blakely in Key to Euc., 265 (1934). A coarser form than the above. Mature Jeuues lanceolate to broadly lanceolate, thick, 6-12 x 1°5-2°5 em. Uwmibels as above. Buds 15 x 5-6 mm, including the pedicel, shape as above, Operculym conical acute, apex angular. Stamens white or pink. Fruit 14-16 x 7-9 mm, Lyre Peninsula and, according to Blakely, also near Encounter Bay. 36. Eucaryerus cracivis F, Muell, in Trans. Vict. Inst, 1, 35 (1855). Mallee or medium-sized tree, Mature lcaves alternate, petiolate, narrow linear to lanceolate, oil glands conspicuous as black dots, 4-9 x 0-6-1-5 cm. Umbels axillary, 4-12-flowered, peduncles terete or semiterete. Suds clavate (with pedicel), not angular, with numerous oil glands, pedicellate, 3-5 x 45 mm. (pedicel about 3 mm). Operculam hemispherical or patelliform, obtuse or apiculate, shorter than the obconical tube. Jirwit ovoid to pyriform, smooth, pedicellate, 5-6 x 3-3 mm.; disc lining the top of the tube or oblique, the rim formed by the narrow ridge of the staminal ring; valves short and broad, enclosed or very near the orifice (pl. iii, fig. 14). Eyre and Yorke Peninsulas, Flinders Range and south to the Clare hills, Murray Mallee and upper South-East. 36a. Eucaryprus GRAciLis var, rreeta Blakely in Key to Enc., 266 (1934), Leaves erect, narrow linear, shining, oi) glands yery numerous. This appears to be a dry country form. It is possible that it is not a very definite variety. Eyre Peninsula, Murray Mallee and the northern edge of the mallee scrub, 37, EvcaLyptus INTERTENTA R. T. Baker im Proc. Linn. Soc. N.SW., 25, 308 (1900). Small to fairly large tree. Jaltre leaves alternate, petiolate, linear to lanceo- late, 7-12 x 0-7-2 cm., bluish or stibglancous. Cowhels in terminal or subtenninal panicles or rarely solitary in upper axils, 46-fowered, peduncles slender, terete or sethiterete. Buds oboyoid, clavate on slender, subangular pedicels, 5-7 x 3-5 min. Operculwnt conical, acnte, somewhat narrower than the tube and shorter, Tube obconical and sometimes slightly constricted in to the orifice, with one oar two lines continuous with the angles on the pedicels. Fruit abovoid or pyriform, fet pedicellate 6-8 x 5-7 mm,; orifice contracted, the persistent staminal ring ¢om- mouly hiding the narrow isc; valves short, enclosed ot near orifice (pl. iv, fig. 14), Everard and Birksgate Ranges and south-eastwards to the Flinders Range, also west of Port Augusta, 37a. EUCALYPTUS JNTERTEXTA var. FRUTICOSA Blakely and Jacobs in Key to Euc., 168 (1934). Mallee. Mature leaves as above but rarely dull and bluish, more usually shining and yellowish-green when dried. Umibels as above, Operculirm shortly conical and sometimes apiculate, due to the point on the thin pre-operculum which persists till or almost until flowering. Fruit as above, This form is found on rock slopes in arid country, whereas the typical form oceurs low on the slopes adjacent to creeks hordered hy E, camaldulensis, It was based on a Northern Territory specimen but extends down to the northern Flinders Range atound Hawker and Quorn. 38 Eucatyerus coorakaH Blakely and Jacobs in Key to Euc., 245 (1934). Tree. Mature leaves alternate, petiolate, narrow to broad laneolate, 7-15 x 1-3 em. Umbels in short terminal or subterminal panicles, 3-6-Howered, peduncles slender, subangular. Buds ovoid, pedicellate, 3-5 x 3-4-mm. Opercuium conical, acute or apiculate, about the same length as tube or slightly longer, Lruil hemispherical or broadly turbinate, 3-4 x 4-5 mm.,, the rim of the tube thin and fragile with very thin disc; valves short, broad, obtuse, exsert and more or less incurved, The South Australian material of this species was formerly placed with E. microthecg F. Muell. The differences between the two appear to be slight. Tt occurs in the northern parts of the State near Lake Eyre and the northern Flinders Range. 39. BucALyprus LeucoxyLon F. Muell. in Trans. Vict. Inst., 1, 33 (1855), sensu Stricto. Small or medinmrsized tree. Mature leaves alternate, petiolate, narrow ta broad Iancevlate, 7-15 x 1-3 em. Umbels axillary or lateral on leafless portions of branchlets or in short panicles, 3-floweted (except variety), pedtncles terete or semiterete, Buds clavate, ovoid or cylindrical-rostrate, wrinkled when dry, pedicels normally lorig and flattened and angulat above as they pass into the calyx tube, sometimes short and semitercte. Opercitluim conical, acute or rostrate with a sometimes angular point, as Jong as or longer than the obeonical or turbinate tube, Stamens with translucent oi} glands in the filaments, especially the outer anantherotis ones, Staminal ring 1-2 mm. wide, conspicuous when the stamens fall, hiding the oblique dise, deciduous during the maturation of the fruit or remaining attached to one side of the rim. Fruit pyriform, ovoid or subglobular, on long terete or semiterete pedicels (except in short pedicelled forms); disc oblique but not covering the short broad enclosed valves. In the typical form the buds are 8-13 x 6-7 mm. and the fruits 10-15 x 8-13 nn, (pl. ii, fig. 7), This occurs in the southern Flinders Range, Mount Lofty Range, Murray lands and in the South-East and on Kangaroo Island, This species shows considerable variation and a large number of forms have been named. Both red- and white-Apwered individuals occur, bit are not restricted to any particular variety. Some plants are glaucous, but not reliably so, Many of the forms seem to grade into one another, but despite this the fullowing varieties are Included. 404. EvcanyPrus LEucoxYLon var. mMAcRocarFA J. F, Brown in For, Fi, S. Aust. (1882; &. Irucoxylon var. erythronenia F, Mucell, in Mig. Ned Krnidk, Archief,, 4, 127 (1854), 158 Buds 16-20 x 10 mm., pedicels very angular. Operculum rostrate. -Fryit, 15-22 x 15-18 mm, (pl. ii, fig. 6), There seems to be some room fur doubt concerning the legitimacy of Brown's uame. Maiden (Crit, Rev., 21, 91) accepts it because the variety includes both red- an? white-flowered forms. This point had been mentioned in Brown's deseription, where it is stated that “we have ventured to apply the name micero- carpa,,..as being more directly specific. and. in order to individuate them from the red-flowering sports of the true representative of the species Leucoxylon,” This latter fact is the more important as it means that Mueller’s description is definitely inadequate, being “filamentis sanguineis.” According to Maiden (Lc.) and Brown there is no doubt as to the synlonyimity of the two names, Maiden and Blakély have been followed in this paper because: (i) Mueller’s description is inadequate; (ii) the wide variation within the species means that varietal names represent little more than arbitary points in a series; and (iii) 50 litte can be gained in this case by a strict application of the rules of priority. To waive Mueller’s description hecause it is not diagnostic is an admittedly risky procedure, since what constitutes an adequate deseription has never been definitely stated and the great majority of the early descriptions leave much to be desired, Tlowever, in this case an argument can be made for either name, and therefore the status quo has been maintained. The variety was based on specimens from Eyre Peninsula. It also ocetirs in the Mount Lofty Range and on Kangaroo Island, Tntermediates occur in the Flinders and Mount Lofty Ranges. 39b. Evcatyrtus Levcoxyion var. AncutaTa Benth, in FI. Austral. 3, 210 (1866). The relation between this variety, the preceding and var. regulosd F. Muell, is nut clear (c.f. Crit, Rey,, 12, 91), Bertham’s description states that the calyx tube is distinctly angled. ln most large-fAuwered forms the pedicel is flattened as it passes into the tube and the two rihs thus formed, with two smaller ones between, pass upwards to be lust in the upper part of the tube. Blakely states that the fruits im this variety, which is not known to the writer, are subangular. Iu all material available the fruits are rounded, any angularity having heen lost during maturation. Maiden considered (1e:) var. angular was not the same as var. mucrocarpe, but gives no grounds for his opinion, Further collecting may elucidate this point, but the variety docs not seem to be a sound one, It is reported to have come from the Mount Lofty Range, 39c. EucALyprus LeucoxYLON var, PAUPERITA J. E. Brown in Bor. Fl, S. Aust. (1882) ; A. jugalis Naudin in Second Mem., 37 (1891). Umbels three or more flowered. Buds ovoid or pyriform, peduncles, and pedicels shorter than in above varieties, 5-7 x 3-5 mm. Operculum conical or conical-apiculate rather than rostrate. Frit globular turbinate to hemispherical, 6-7 per mm,, relatively wider and shallower than in other varieties (pl. ii, fig. 4-5), This form docs not seem to he sufficiently distinct to warrant specific rank, Some specimens are distirictive with more than three flowers per wmbel and short pedicelled pyriform buds. but these grade into normal but small fruited forms of 4. lencoxylon. Mueller’s original description of EB. leucoxylen stated “tri-raro-quingtte- floris.” This referred to his variety pluriflora which, as stated by Maiden, is E.. calcieulirix F. Muell, From Clare north into the Flinders Range, also in the Mount Lofty Range and on Eyre Peninsula. 159 40. Eucatyrrus cALcrcuLTerx F. Muell, in Mig. Ned, Kruidk. Archief., 4, 129 (1856). Mallee or small tree, Mualwre leaves alternate, petiolate, lanceolate ta broad lanceolate, the marginal nerve usually distant. Umbels axillary or clustered ou leafless portions of the branchlets,3-7-flowered, peduncles semitercte, 4-8 mm. long, Bury clavate on more or less angular pedicels 5-7 x 4-5 mim., the surface crumpled or wrinkled like that of a withered and shrunken apple. Operculum cortical, acute or obtuse, slightly shorter than the tube. Fruit pedicellate, some- times shortly so, pyriform or truncate-ohovoid, 5-8 x 5-8 mm.: dise narrow, very oblique or lining the top af the tuhe, yalves short and enclosed but not so deeply so as in £. odorata (pl. i, fig. 7), E. culcicultric occupies a position between E, odorata and ihe sitall fruited forms of E.lencorylon, and it is Sigtificant that, while based on Mueller’s F_ odorata var. calciculérix, it also includes his Z. lexco.rylow var. pluviflera. The species ig a well-marked one and can be distinguished by the characteristic wrinkling or erumpling of the surlace of the buds when dry, the usual retention of a staminal ring during the maturation of the fruit and hy the marginal nerve of the lanceolate leaf being distant fram the edge, J3lukely’s var, porosa based) on E, porosa TF. Muell, illustrates the tendency lowards E. lencoxylon and does not warrant separation. Var, obscura appears to have been well named, There are two specimens in the Sydney Herbarium named A. caleiculirta var, obscura, Both were collected by Maiden at Bundaleer Forest, March 1807, One specimen is E, fruticetorum, which as stated below ts Z odorata var. angustifolia, and this is also represented in the Melbourne Herbarium. The other appears to be that referred to as having “ihe general facies of EF. fascicnlosa” (Key to Euc., 225). This latter specimen probably belongs with L. hemiphloia var. microcarpi, Eyre Peninsula, Yorke Peninsula, flinders Range, Adclaide Plains, and Murray Mallee, THE ODORATA COMPLEX This, like ihe diosa complex, has been the centre of considerable confusion and the arrangement in the Key to the Eucalypls 1s untortunate. The true sittia- tion appears to be as follows: E. odorata, for which E, cajuputea F, Muell, is a synonym, is the tree form which is found around Adelaide and which becomes mallee-like as it goes north to Mount Remarkable and south to Fneounter Bay: E, odorata var. angustifolia is synonymous with E_ fruticetorum F, Muell, which is not distinct enough to warrant specific ranking. Tt appears to link E- odorata with ihe very narrow-leaved &. viridis R, T. Baker, ‘The type specimen af Mueller’s £, odorata var. erythrandra has been seen. As pointed out, in a private commuttication by a Melbourne authority, it is the same as 2. Lansdowmeana F, Muel!, and J. E. Brown, The latter is a distinct species. having normally sessile and coarser buds and {rvits in short panicles or clustered umbels. lakely’s descriplion of this variety is erronecus and appears to have heen based on a ted- flowered specimen of K. oderata. Tm the type specimens of E. adardlw (from Light’s Pass) and of . cajuputea (from the Fhnders Range) the filaments appear to have been red, though it is not always easy to be certain of this character when dealing with dried material. These specimens are in the Melbourne Herbarium and have been seen. Blakely’s Z. adurata var. refracia (Key to Enc. 226) is an abnormality. Similar reflexed stamens have heen observed in flowers of other species. Usually only some of the flowers on a branch are affected, T his variety cannot be retained. Var. macrocarpa (Le.) is £. Jansdouneana. 41. Evcanyrrus LanspowweAna F, Meell. and J, E, Brown in For. Fl., S. Aust, pt. dc (1890), £ odorata var. erythrundra F, Mitel, in. Mig, Ned. Kruidk, Archief,, 4, 129 (1856). 160 Mallee. Mature leaves alternate, petiolate, thicker and coarser than those of F£, odorata, lanceolate to broad lanceolate, 8-13 x 1-3°5 em, Umobels always clustered and usually in short panicles or the umbels grouped close together on a leafless portion of the stem, 3-8-flowered, peduncles semiterete, Budy sessile or on short thick angular pedicels difficult to distinguish fram the obconical and some- what anfgular tube. Operculum conical, acute or obtuse, shorter than tube. Jruzt sessile or shortly pedicellate, cylindrical or barrel-shaped, striate or bicostute, 8-10 x 7-9 mm: disc very narrow, lining the top of the tube, valves short, deeply enclosed (pl. iv, fig. 8). Southern Eyre Peninsula and Kangaroo Island, 42. EvcaLyrrus oporata Behr tind Schlecht, in Linnaea, 20, 547 and 567 (1847). sensu stricto; E. cajupulea F. Muell, in Mig. Ned Kruidk. Archief., 4, 126 (1856), Tree or mallee. Mature leaves alternate, petiolate, narrow to broad lanceolate (narrow in mallee forms), 6-14 x 1255 cm, Unmbels axillary or clustered on leafless portions of the branchlets, peduncles semiterete, 8-15 mm. long. Buds an angular pedicels 2-7 mm, long and passing without abrupt change into the slightly angular obconical tube, 6-8 x 3-5 mm,; dise lining the top of the tube, the rim thin, valves deeply enclosed, short (pl, iv, fig. 5). Eyre and Yorke Peninsulas, southern Flinders Range, Adelaide Plains, aa Lofty Range and south to Encounter Bay, Kangaroo Island and Murray Mallee. 42a. EUCALYPTUS OpoRATA var. ANGUSTIFOLIA Blakely in Key to Enc., 226 (1934); E. fruticetorum F. Muell, in Miq, Ned. Kruidk. Archief., 4, 131 (1856). Mallee. Mature leaves alternate, petiolate, narrow lanceolate, 5-14 x O'3-1'S cm., surface dull and subglaucous, especially when young though older leaves are like those of the above. Buils shortly pedicellate, smaller than in the above but sttbglaucous though similar in shape. srwit turbinate, cipular or pyriform with striations running down into the short angular pedicel (pl. iy, fig. 6). Port Lincoln, central Flinders Range between Quorn and Wirrabara. 43. Euecacyprus yirtpis R. T. Baker, in Proc, Linn, Soc, N.S.W., 25, 316 (1900), Mallee or small poorly-shaped tree. Mature leaves alternate, petiolate, very narrow, linear, 6-16 x 0°3-0:5 cm. Unmbely axillary, 3-S-flowered, peduncles semitercte, 5-7 mm. long. Buds cylindrical or ovate, on short pedicels, 5-7 x 3-5 mm. Operculwm shortly conical, smooth or slightly striate but not pinched 11 to form an angular pot, as long as the tube which is cylindrical and which passes into the pedicel more abruptly than in any other members of the odorata complex, Fruit pedicellate, hemispherical ta semi-globular, 4-6 x 4-5 min.; disc very narrow and lining the rim, valves short and enclosed (pl. iv, figr, 7). Flinders Range near Ilorrocks Pass and Quorn, 44, Eucalyptus HEMIPHLOIA F. Muell. ex. Benth. in Fl, Austral., 3, 216 (1866), sensu stricta, Tree. Mature leaves alternate, petiolate, lanceolate 10 broadly laricvolate, 7-t4 x 1-3°5 com. Usnbels in panicles or in paniclelike groups on leafless portions of the branchlets, 4-8-flowered. Buds cylindrical-clavate, pedicellate, 7-10 x 5-6 mm., pedicels subangular and sometimes the striations pass up on to the tube Operculum conical, smooth, acute or pointed, aboitt as lung as tube or shorter. Fruit pyriform urceolate or cylindrical, pedicellate, canstricted at the orihce. 6-8 x 4-6 mm,; dise small and oblique or lining the top of the tube, the old stathinal ring forming a narrow rini to the orifice; valves deeply enclosed (pl. iv, fig, 10). 161 Marden notes (Crit. Rev., 11, 14) that Mueller in his Census gave Pragm. ii, 62 (1860-61) as the reference for the description of this species, However, the remarks in this volume of the Fraugmenta hardly constitute a real description, and Blakely was tight in giving the reference as above. On the other hand, Blakely gives E. albens Mig. in Ned. Kruidk. Archicf., 4, 138 (1856), despite the fact that Mueller in the Eucalyptographia (also quoted by Maiden) points out that the name, which was a misprint [or £. pallens DC., was given without diagnosis, hence the correct reference should be 2. albens Mig, ex Renth., Fl, Aust., 3, 219 (1866). Blakely accepted E_ hemiphloia, FE. albens and E, microcarpu as distinct species, but the writer ig in agreement with Maiden when tre considered the last two io be varieties of the first, a decision he later revised. As noted by Blakely (Key to Euc., 236), “in Victoria it (i.e., E. microcarpa) seems to pass gradually into the typical form (te., E. hemiphloia).” This is typical of the problem facing the student of the genus. Specimens which appear absolutely distinct are found on more extensive collecting to grade into one another. &. albens differs in its usttally larger buds and fruits and in the glattcousness of the leaves, branchicts, buds and fruits. Jn no species of Hucalyptus does glaucousess prove to be a reliable character, and a study of the dimensions given for EH, hemuphioia and E, albens in the Key and hy Maiden (Crit. Rev., 58, 443) will show that there is a cerlain ovetlapping. Southern Flinders Range between Latira and Mount Remarkable. 44a. Eucaryetus HEMIPHLOIA var, MicrocarpA Maiden in Crit. Rev., 47, 207 (1921); E. microcarpa Maiden in Crit. Rev., 58, 483 (1923). Differs from the above mainly in its smaller buds and fruits which are pedicellate to sessile. Buds 6-7 x 3-4 mm. Opercilum has an angular tip (also seen in species of the odorata complex), Tube subangular, ruil cylindrical or slightly smaller at each end. 4-7 x 4-5 mm, (pl. iv, fig. 11). Distribution as above, 44), EucaLverus HeMIPHLOLA yar, aLBENS F. Muell. ex Maiden in Crit, Rev,, 11, 20 (1914) ; &. albens Miq. ex Benth. in Fl. Aust. 3, 219 (1866). Differs from the typical form in being glaucous on all parts and in the usually larger buds and fruits. Blakely gives the following dimensions; Buds 10-15 x 5-8 mm. Jirwt 9-12 x 8-10 mm, In most of the South Anstralian material these organs are smaller but still larger than the dimensions for HJtemiphloia sensi stricto (pl. iv, fig. 9). Distribution as above. 45, EucALYPTus LARGtFLORENS FL Muell. in Trans. Vict. Inst., 1, 54 (1855); E. bicolor (A. Curn.) Hovker in Mitchell’s Jour. Trop, Aust., 390 (1848). Tree. Mature leaves alternate, petiolate, lanceolate, surface dull and faintly binish, 7-13 x [-2 em. OUvmbels in awillary or terminal panicles, 3-8-lowered, peduncles slender, Buds obovoid or eclavale on slender pedicels, 3-5 x 2-4 mm. Operculum hemispherical or almost conical, shorter than tube. Fruté cyhtutrical or pyriform, the orifice slightly constricted, pediccllate, 3-4 x 3-4 mm.; disc very narrow, valves enclosed and short. Along the River Murray. Also occasionally in the Flinders Range. 46, Eucanverus Renrtawa F. Muell. in Trans. Vict. Inst, 1, 34 (1855). Mallee or small tree. Malure leaves alternate, petiolate, ovate or broadly lanceo- late, 7-12 % 1°5-5°5 cm, Uihels 3-6-Nlowered, im terminal panicles which may be open or short and campact. Bui sessile, cylindrical to Obovoid, 5-G x 2-3 mm. Opercuium hemispherical or shortly and obiusely conical. /’ruit sessile or almust 162 so, hemispherical or chovoid, 3-5 x 3-5 mm,: disc narrow, valves short, enclosed or just below the orifice. Blakely gives 5-7 mm. as the length of the fruit, but none as Jarge are present in South Australian specimens. Maiden figures (Crit. Rev., pl, 48, fig. 6) some larger fruits, but the disc is quite different from that in other material of the species atid the identity is therefore in doubt, Eyre Peninsula, southern Flinders Range, Murray Desert and Kangaroo Island. 47, EwcALYetus FASCcICcULOSA F,, Muell, it. Trans. Vict, Inst, 1, 34 (1855). Medium-sized tree. Mature leaves alternate, petiolate, lanceolate to broadly lanceolate, 8-18 x 1'5-3°5 cm. Unibels in axillary or terminal panicles. Buas clavate on stibangular pedicels, 4-6 x 3-4 min. (pedicels 3-5 mm. long), usually with iiumerous oil glands. Operculum shortly conical and narrower than top of tube. Fruit pediccllate, obconical, cylindrical or pyriform with a thin rim lined by the disc, valves short and enclosed (pl, i, fig. 6). Mount Lofty Range to Encounter Ray, Murray Bridge, Upper South-East. 48, Eucatyrtus crapvocanyx F, Muell, in Linnaes, 15, 388 (1852). Small to medium-sized tree. Mature leaves alternate, petiolate, lanceolate to broadly lanceolate, 8-14 x 1°5-3 em. Usnbels clustered on leafless portions of the stems or in panicles, peduncles terete. Bwds cylindrical, urceolate, pedicellate, 10 x 46 tum. Operculuim hemispherical, apiculate, much shorter than tube but sometimes a little wider, 3x.4-6 mm. Fruit ovoid or cylindrical-urceolate, having a constricted orifice, almost smooth or becoming rugose or costate on drying, 10-15 x 5-10 mm. ; disc forming a narrow rim, valves deeply enclosed (pl, i, fig. 5). Eyre Peninsula, southern Flinders Range and Kangareo Island. 49, Eucanuyprus GAMopiyria F, Muell. in Fragm., xi. 40 (1878). Glaucous mallee, Mature leaves opposite, sessile or amplexicaule, orbicular, lanceolate or cuneate when the pairs are joined, 3-8 x 1-2 cm, T/mbels poorly formed, singly or in short panicles, in the upper axils or terminal. Individual umbels with few flowers. Buds clavate or pyriform. pedicellave, §-7 x 4-6 mm. Operculum hemispherical, much shorter than the obconical or pyriform tube. Fruit campanulate, pyriform to almost urceolate, 8-11 x 6-8 mm.; rim sharp around the narrow oblique disc, valves short, broad and triangular, near orifice or enclosed. Blakely states that this is a very glaucous species, but our material was ouly just glaucous. This character is never very dependable, Near the north-western corner of the State. Species Not INcLipep In THE Key There is 2 group of tropical species which is known to extend inta South Australia. Matertal of these is by no means satistactory, since buds, fruits and leaves are rarely included on the same specimen aud notes regarding habit and bark character are usually lacking. Consequently there is a certain doubt concerning the true identity of the specimens, end the species have, therefore, been omitted from the key. However, an account of the South Australian Eucalypts would be incomplete without some reference to them. In his Flora of South Australia, Black includes FE. pyrophora Benth, and E, terminalis F, Muell. (p. 420), and in his additional notes (p. 694) it is stated that the White Wash Gum of Central Australia is 2, papuana I’. Muell., not &, terminalis F. Muell. Ina later paper, by J. B, Cleland, for which Black identi- fed the plants (G), there is a reference to EB. dichromophloia F, Muell,, which has 163 been recorded for the Mann and Musgrave Ranges by Black (3), and, in the same journal, a paper by Blakely (4) describes E. papuana var, aparrerinja, In the Key, Blakely lists H. ferruginea F. Muell, as South Australian bit gives a Central Australian locality. All these species come in the Corymbosae and all belong to the Macantherae. Without adequate material and field notes it is difficult to distinguish between some of them, Until further collecting is done it will be difficult to decide which of them should be recorded for this State. Another species of interest is A. gonyylocarpa Blakely (4), for which one South Australian specimen is recorded with the type description. ‘his came from “25 miles 8.5.W. of Mount Watson,” but consists of leaves only and, in Blakely’s Opinion, represents the juvenile slate of the species. This requires checking. E, trivalva Blakely, l.c., and &, orbifolia F. Muell. (4) (7) may also occur in this State. F. Kalangadooensis Maiden and Blakely is described as a probable hybrid. The type material appears to represent an unusually large form of FE. viminalis yar. Huberiana. Further collecting may determine the status of this form. ACKNOWLEDGMENTS The writer wishes to acknowledge the financial assistance given by the authorities of the University of Adelaide, which made possible a visit to the important collections in the National Ilerbaria at Melbourtie and Sydney. Reference must also be made to the helpful co-operation of the staffs of these institutions, both during the visits and at other times, Mr. J. M. Black and Professor J, B. Cleland kindly lent their valuable collec- tions, which include a number af type and other important specimens, The Eucalypts among the specimens. of the Tate and the Flder Exploring Expedition Collections were also consulted in the Herbarium of the Botany Department of the University of Adelaide. LITERATURE CITED (1) Benrnam, G, 1866 Flora Auistraliensis, 3 (2) Bracx, J, M. 1929 Flora of South Australia, Govt, Printer, Adclaide (3} Brack, J. M. 1934 Trans. Roy. Soc. S. Aust., 58, 168 (4) Rraxety, W. F. 1936 Trans. Roy. Soc. S. Aust., 60, 153 (5) Brerr,R.G, 1937 Papers and Proc, Ray. Sac. Tas., 75 (6) Cxetanp, J. B, 1936 Trans, Roy. Soc. S$, Aust., 60, 114 (7) Gagpwer, C. A, 1940-41 Jour, Roy. Soc. W. Aust., 27, 187 (8) Jessur, R. W. 1946 Trans. Roy. Soc. S. Aust., 70, 3 (9) Muetrer, F. vow 1879 Encalyptographia (10) Woon, J. G. 1937 The Vegetation of South Australia, Govt. Printer, Adelaide. Frequent reference is also made to Blakely’s “Key to the Eucalypts” and to Maiden’s “Critical Revision of the Genus Eucalyptus.” 164 Puate I . E, axymitra (from type in Tate Herb.) 5. E. cladocalyx E . pyriformis 6. E. fasciculosa 1 2 3, EB. pachyphylla 7. E, calcicultrix ; 4, E. Morrisii 8. E. Ewartiana (from portion of type in J, M, Black’s collection.) (All drawings natural size.) 1, 2. 3. 4. 5. 6. Ey E. E E Py, E remota obliqua var. megacarpa . diversifolia , leucoxylon var, pauperita leucaxylon var. pauperita . leucoxylon var. macrocarpa (All 7. E. leucoxylom sensi stricto & E. inerassata sensu stricto 9. EB. inecrassata sensu stricto 10. E. tucrassata var. costate 11. E. tnerassala var. angulosa drawings natural size.) {5b Pate IT 1. &. Baxteri 6. E. pimpiniana 11. E, pileata 2, E. viminalis var. T[uberiana 7. E. conglobata 12. E. trachycalyx 3. &. ovata & BE. anceps 13. 5. concinna 4. FE. elaeophora 9. E. dumasa 14. EF. gracilis 5. E. cosmophylla 10. &. rugosa 15. E. calycogona (All drawings natural size.) i. Lansdowneana 2. hemiphloia var. albens oleosa var. glauca eleosa var. angustifolia oleosa sensu stricto ND Gt bh ON bs the toy tb ay odorata var. angustifolia . hemiphloia sensu stricto . hemiphloia var. microcarpa « leptophylla » wncitata i. mteriexta (All drawings natural size.) FUNCTIONAL SYNTHESIS IN PEDOGENESIS By C. G. STEPHENS Summary Soil is composed of weathered rock material modified by additions of organic matter derived from vegetable and animal life. Despite the obvious association and contributions of rock, climate and organic matter, there have been in relatively modern times different genetical classifications of soils which have rested purely on geological, climatic or organic factors of soil formation. Largely because of the language difficulty these monogenetic classifications were but slightly influenced by the work and theories of Dokuchaiev until well after the beginning of this century, four or five decades since his earlier publications on the genesis of soils. There seems little doubt that Dokuchaiev was the first to fully realize the function of more than one factor in soil formation, and that the factors were not independent of each other. In this connection Neustruev (1927) states: “.,.1t should be noted that though Dokuchaiev had established such factors of soil formation as climate, parent rock, relief, vegetation and age of the country, he still, in the first place, emphasised the idea that it is their interaction which leads to definite conditions of soil formation.” Ototzky (1946) quote Dokuchaiev as follows: “Knowing all the physico-geographical elements of a certain locality, it is easy (!) to predict which soil covers it.” 168 FUNCTIONAL SYNTHESIS IN PEDOGENESIS By C, G. Srepuens * [Read 10 April 1947] Soil is composed of weathered rock material modified by additions of organic matter derived from vegetable and animal life. Despite the obvious association and contributions of rock, climate and organic matter, there have been in relatively modern times different genetical classifications of soils which have rested purely on geological, clirnatic or organic factors of soil formation. Largely because of the language difficulty these monogenetic classifications were but slightly influenced by the work and theories of Dokuchaiey until well after the beginning of this century, four or five decades since his earlier publications on the genesis of soils, There seems little doubt that Dokuchaiev was the first to fully realize the function of more than one factor in soil formation, and that the factors were not independent of each other. In this connection Neustruey (1927) states: “|... it should be noted that though Dokuchaiey had established such factors of soil formation as climate, parent rock, relief, vegetation and age of the country, he still, in the first place, emphasised the idea that it is their interaction which leads to definite conditions of soil formation.” Ototzky (1946) quotes Doku- chaiev as follows: “Knowing all the physico-geographical elements. of a certain locality, it is easy (!) to predict which soil covers it.” Dokuchaicv gave the formula P=f (K.0,G,V), in which P stands for soil (pochva), K for climate (klimat), O for organisms (organism), G for subsoils (gornaya poroda) and V for age (vosrast) of the soil. According to Neustruev it was Sibirtsey who first established the frequently “predominant role of climate,” “the idea of zonality of soils and zonal types of soil formation,’ Sibirtsev also distinguished ‘a division of intrazonal soils” to which, in addition, he appled the term semi-zonal, thus acknowledging their pattial dependence “on climate and their occurrence in definite zones beyond which they are found but rarely, and that in peculiar conditions.” A recent mathematical treatment of the role of factors in soil genesis is that given by Jenny (1941) in his book, “Factors of Soil Formation.” He expresses the relationship between soil properties (and hettce the soil which is an ensemble oi soil properties) and the environment by two equations: s=f’ (cl’,o’,r’,p,t), and s={f (cl, 0, r, p,t ----) in which the symbols cl’, o’, r’, p and t represent soil climate, soil organisms, soil relicf, patent material and time, and cl, o and r represent environmental climate, organisms and relief; s is any soil property, and the svil-forming factors represented by the symbols are postulated as independent variables, ne., “they are the independent variablee that define the soil system.” From the second of the above equations Jenny derives an equation of partial differentials, namely: ds ds i os is = (—) del + (— do + =| dr dcl/ oa, t.p, b do! cl, r, p, t. dr/ cl, 0, p,t. os ds + (=) dp + (—) at dp? clo, t. at/ clo, r. p. * Waite Agricultural Research Institute, University of Adclaide, Trans. Roy, Soc, S, Aust. 71, (2), 1 December, 1947 i69 The interpretation of this equation is that the tota! change of any soil pro- perty, and hence differences between soils, depends on and is a function of the sum of the changes of the soil-forming factors; also, assuming that it is possible to assign a numerical scale to each of the variables, the magnitudes of the partial differentials are true indices of the importance of the yarious factors, In reality it is not possible to derive the values of the partial differentials because of actual and uncontrollable variation in one or more of the remaining variables, and because it is extremely difficult, if not impossible, to assign a numerical scale to some of the variables, particularly o and p. In addition, cl, 0, r, and p are, or can be, multiple factors and yield groups of functions; for example, the ratio of precipitation to evaporation (P/E) and temperature (T) are largely used to characterise cl. Furthermore, the variables, despite their being assigned an independent status to facilitate the mathematical approach, are in nature not completely independent, relief for instance, as is well known to geomorphologists, being a function of parent material and climate, but having a partially independent relationship with soil. The treatment is thus to.a degree philosophical, rather than rigidly mathematical. Hence it is necessary to work ont the significance and role of the various soil-forming factors on a statistical, gtaphical, or diagrammatic basis. Jenny illustrates the function of each of the soil-forming factors by dis- cussing them in turn in their relationships to various soil characteristics. He does not attempt to show the effects of interaction or illustrate the variables operating together; rather his efforts are directed towards. separating and illustrating the effects of the individual variables. More recent work by Jenny (1946) illustrates the relationships of soil series and types mapped in vatious soil surveys to the individual soil-forming factors, particularly how various sequences of soils are related to changes in the value or composition of one of the variables. Ife postulates “five canonical functions of pedology” where four of the variables remain constant, thus: Climolunctions = - s=ft (cl) O.npt Biofunctions - - s=f (0) cl, r, p,t Topofunctions = - - s=f (r) cl, 0, p, t Lithofunctions = - - s=f (p) cl,o,r,t Chronofunctions. - - s=E (t) lo,T, p Designating an ensemble of s values by E(s) he writes: Soil = E(s) = [ (clo, r,p,t.-... ») and defines the various soil sequences as follows: Climosequences. - K(s) =f (cl) 9 5, pt Biosequences - E(s) =f (0) g, rq, Toposequences = - E(s) =f (r) cl, 0, pt Lithosequences - E(s) = f (p} cl,o,r,t Chronosequences E(s)=€(t) oo,r, Pp Toposequcnces include clinosequences and hydrosequences, the former cover- ang the effect of the slope factor (i) and the latter the effect of the water table factor (w) on soil formation. Each of the soil sequences is illustrated by reference to maps and diagrams relating to soil series defined in soil surveys. Examples quoted are as follows: c 170 Chronosequence: (Tujunga), Hanford, Greenfield, Ramono, Placentia, (San Joaquin) series. Lithasequence: Colma, Ilugo, Sweeney, Sheridan series. Toposequernces: See Clinosequence and Hydrosequence. Clinosequence? Sheridan sandy loam, shallow phase, Sheridan sandy loam, deep phase. Hydrosequence: Panoche clay, Oxalis clay, Levis clay. Climosequence: Panoche, Sorrento series. Biosequence: Parr, Octagon, Miami series, Jenny states that every homogeneous soil type or series belongs. theoretically to each of the five canonical functions and illustrates this by pointing out that the Panoche series is a recognisable meniber of more than one soil sequence. He thus indicates a method of recognising the rdle of more than one variable in determining the soil series or type. Recently, in Wilde’s “Forest Soils and Forest Growth” (1946), Doktuchaiev was quoted as follows: “The soil is a result of reactions and reciprocal influences of parent rock, climate, topography, plants, animals and age of the Jand.”” Wilde's formulation following this is Ss ={ (g.e, b) dt, in which $ is the soil, g is geological substratum, e is environmental influences, b is biological activity, and t 1s time. This equating of the soil as an integral of the soil-forming factors against time has a distinct appeal as a more precise formulation that that of Dokuchaiey or that of Jenny. In particular the value of some of the variables has changed with the passage of time: for instance certain soils largely owe their morphological characteristics to a climate operating in past geological time, ¢.g., lateritic soils were formed in Australia most probably in the Pliocene, and as they exist in “fossil” form today are not particularly affected by or responsive to the prevailing climate. Wilde apparently does not regard g, e and b as independent variables. The purpose of the paper presented here is to record a diagrammatic method of synthesis of the role of the soil-forming factors and to relate the components of the diagrams both to the independent variables of the Jenny equations and to the pattially dependent and partially independent variables which the soil- forming factors actually comprise in nature, Let the variables which determine the soil (S) be represented as follows: C forclimate © for érganisms R for shape of the land surface W for height and other features of the water-table P for patent material T for time In nature these are not completely independent variables and their known dependencies are sect out in Table I- Taste I A change in C may cause a change in O, R, W, P, S but not in T a ” $7 » chi ”? ” ” Cc, W, P, S ” ” ” T, R ” " ” R ” noo ” ” C; oO, WwW, P, S) ” on T ” a” a3 Ww ” ” ” ” uu O, Ey S ” "7 ” T, C, R ” oi 4 P w ” ” ” 2] O, R, W, 5 yy ” $9. T, Cc ” " ” T ” teow ” a G O, R, W, P t ” cay S rT) rot ” ” O, R, W, Pp but not in T, Cc 171 A complete list of the possible relationships between any two of the variables is given by the partial differentials in Table IT. Tasie II as as aS aS 0s as ac 80 eR aw a at ac 3C ac ac ac aC dS 80 aR aW aP oat a0 90 a0 40 80 90 a8 aC aR aW aP at aR dR OR aR oR OR as ac 380 aw aP art aw aw aW aWw aW aw a8 @C @0 aR aP aT dP oP oP oP oP aP 8s ac 80 OR aw aT oT oT oT oT oT oT 8s aC 80 OR aw aP Reference to Table I and a consideration of the lack of effect of changes in aT OT 8ST AT aT oT C, O, R, W, P and § on the change of T show that —-» —,—- »——» — & — as. ac. a0. aR. ow oP are of no consequence in soil formation and that T is a completely independent ac dC dR variable, Similarly —-, —, —, & —— do not operate, Hence the relationship ap a0 aw between a soil.and the soil forming factors is probably best expressed like Wilde’s equation as: sS= Ss (C, O, R, W, P) dT in which T is independent and C, O, R, W and P can have both dependent and independent status. The function of these soil-forming factors can be qualitatively (and some- times quantitatively) but precisely indicated both as independent and interacting variables by the diagrammatic and mathematical representation of the relation- ships of the soils mapped in soil surveys to the variations in the factors which characterise the environment of the area surveyed. Provided the environment is studied in the necessary detail during the course of a soil survey, it is possible to determine in some degree the relative importance of the different soil-forming factors in the genesis of the different soil series or types. This has been done in some soil surveys in Australia, with the result that it is now possible to present useful diagrammatic expressions of the dependence and interaction of the soil- 172 forming factors. The comparison of a soil map based strictly and entirely on ficid obseryation of soil morphology and distrilmtion with topographic, hydro- graphic, climatic, geological and vegetation niaps and a study of the Tertiary and Recent geomorphology of the same area are geverally most revealing regarding the functions of climate, arganisms, relief, water-table, parent material and time in soil formation, The construction of diagrams to illustrate the importance of the factors and the order in which they have operated calls for some ingenuity, but it has been found possible in a relatively simple arrangement to inlegrate the functions of the variables operating in the formation of different soil scries and types occurring in the same locality. Examples of three such diagrams relating to widely separated surveys of various sizes are illustrated in fig, 1,2 and 3. ‘Iwo of these diagrams; in a provisional and rudimentary form, and devoid of any mathematical interpretation, were published previously in the bulletins describing the relevant soil surveys; namely in: 1. CS.LR, (Australia) Bulletin No. 150, 1942, “The Soils of the Parishes of Longford, Cressy, and Lawrence, County Westmorland, Tasmania.” 2. CS.LR. (Australia) Bulletin No, 188, 1945, “A Soil, Land-Use and Erosion Survey of Part of County Victoria, South Australia.” The third diagram relates to the following: 3. Tranzactions of the Royal Sociely of South Australia, 67, (2), 1943, 191- 199, “The Pedology of a South Australian Fen.” In their present form the diagrams give qualitative expression to the variables of the Jenny equations and provide 4 means of recognition of his soil sequences; which are naturally segregated in different portions of the diagrams, They also confirin the crude additive character of the soil-forming factors expressed by his equation of partial differentials, Im addition they provide a framework or teference grid on which can be recognised the appropriate place at which the partial differentials of the partially dependent and interacting variables C, Q, R, W, and P operate in determining the course vf soil development, in relation to the independent variable T. Inspection of the diagrams shows how a “lateral analysis” relates the various steps in soil formation to the independent Jenny variables and how a “vertical (time) analysis’ reveals the dependent reaction of the natural yarighles and their ultimate control of soil formation. ‘The diagrams should provide a useful starting point in the quantitative evaluation of the factors C, R. W. and T, which lend themselves to such treatment. Fig. 1 is a pedogenetic diagram of the soils of the eastern portion of County Victoria in the wheat-growing and sheep-raising country of South Australia. It includes all of the soils mapped aver an area of about 600 square miles, shawing how each soil series is a synthesis of the functions of the variables dealt with above. The morphology of the soils, a soil map, a topographical map and diagram, a rainfull map and other climatic data and details of the gealogy and vegetation of the area were published in the soil survey bulletin mentioned pre- viously. In addition, in this diagram the contribution of calcareous Ipess ta the morphology of two of the soil series is shown; this following the recent work of Crocker (1946). In the first column on the right-hand side of the diagram is shown the relationship of the various steps of the diagram to the Jenny variables, Recogni- tion of his soil sequences is easy, for example the Caltowie, Canowie and Bunda- leer series comprising a climosequence. In addition, soils whose genesis is dominated by one or more variables cam be recognised, For example, both the ¥angya silty-loam and the Beetaloo series are formed on calcareous materials in 173 *YTTVEESOY BLOG 'VIMOLOIA LENNOD AO KOIDE NEALSVT EBL WO STICS TEL LO KVEDVIC OT LANEOdITS Laaerat § seTqQutres Temyes Fup{owrsAU; @Ty JO SdyysuCTyS Ter Og} AOYS BTeT{UsIETITP Terjzed euz wapant Ts Wee Tros uyzea Tres wUpBpUeLT Troe haba a ES woes Res] BAOHA~Ped | SOTTO wurepued | peperteq | Tee TS7s setzas aeprep gel 115 3 alt PE Ea cetae |_| spay [~~ totoe os eoyres STAOR TES og Py * WerssAus gnt26 puYTpoom / ywurEaes vmaniee ee pamiosi= cm ayuors wag eRgOA TROT enqdtteang | ByTONs TEE BywONT ter Saaenes Taare oe oe (wOysvyavA _ BOTSTTEOOT BOTATTWOOT SOTPTINIOL BOTITTEOOT BOTT VOOT BEYATTROOT MOFRFTMOOT SOFATTEIOT GOTFTWOOT STIFF TERT Tworydeszee)) Tete TY 10930K r9F 10 BUT TSS 480439K 101394 zetag TW ty twoo “4 * -- 203 Ii, Tae Non-Gnersstc Cuarnocnitns (1) The Mt. Waodraffe and Trudinger Pass Totecmediat and Basic Rocks .. L204 (2) The Upsan eins Giaii! tic “Rocks x ss wv te ts .. 206 (3) The Norite to the S.W. of Tjatjanja .. 4 _ “ 24 ,. 266 (4) The Ernabella Granudivritic Massif .. ty se - ot .. 206 Til. Pegmatites .. ac . _ a e te as . ts o 208 TV. Doalerites a . ele os ve ye . a AR “ .. 208 V. Mylonization Af. 33 it ie ta 4 bt 3 iG .. 209 C. Conciusion ann SUMMARY i « r + " 4 ia .. 209 D. ACKNOWLEDGMENTS .. an . ile = * Me . 2 .. 210 E. BieiiocrApHy ., + a ict ote + by “i a Ly ive 2D A. INTRODUCTION This paper is the first of a proposed series dealing with the geology of the Musgraye Ranges. Some features of the major rock types are outlined below, and other papers are in preparation which will amplify and supplement this intro- ductory account. Ultimately the petrogenesis of this new series of “Charnockitic” rocks will be treated. Previous Work A bibliography of previous workers in the Musgrave Ranges is appended. The two most notable contributions to our knowledge are those uf Basedow (1905) and Jack (1915). Basedow’s account covers the reconnaissance of the South * Geology Department, University of Adelaide. Trans, Ray, Soc. S. Aust., 71, (2), 1 December 1947 196 Australian Government North-West Prospecting Expedition of 1903 which penetrated further west than the Musgraves, into the Mann and Tomkingon Ranges and to the Western Australian border. j ack’s work was carried out in an abnormally dry season (1914), Gonsequently most of his time was spent 10 ihe south-east of the author's area. Notwithstanding, the accounts of both Basedow and Jack contain valiable obgervations, to which reierence will be made in later papers. Locanity axp TeeeaLy The Musgrave Ranges are the casternmost of a series of rugged east-west anonntain chains i the far North-West of South Avstraha. Phese rariyes begin in the east witli bold desert-red Sentinel Hill (180 miles \est Of the Adelaide to Alice Springs railway Hine at Finke, Northern Territory), [See locality plan, p. 199.] As one continues westward from Sentinel Hill the Miseraves grow in magnitude and area, but are never more than 30 niles fron orth to south, Tkey reach their culmination in the mountains in the vicinity’ of Mount Woodroffe, 5.000 feet above sea level. and 4.000 fect above the desert plain. This range continues west for aver 100 nules. but a spur passes west- north-west throwzh Opparinna and thence into the Petermann Ranges within the Northern Territory. ‘To the west of the Muggraves, the Mann and then the Tomkinson Ranges cottinuc almost into Western Australia. A characteristic feature of the Musgrave Ranges is the abrupt mater in whieh most of Uke mountains ise from the plain. Groups of precipitous istand-like hills punctuate the flat plains, which are bormded on either side by precipitous mountain ranges, Such rugged scenery creates a strong impression of some Norwegian ford. ‘This illusion is greatly accentuated by the lengthening shadows of evening. The main aren under consideration (see map) comprises the ranges from Ketimore Park (approximately longitude 132° 30’ E.), westward for 30 miles to Mount Woodroffe and Erlywanyawanya (approximately longitude 131° 40"). Latitide 267 15S. runs through the centre of, the ranges, Most detailed work, however, was done within a few miles of the Preshyterian Mission Station at Frnabetla, ‘Che period of the two field operations undertaken in the Musgraves was during the most exhausting time of the year—that of the Central Australian summer, The first occasion was during December 1943, January and February 1944, followed by a second in December 1945 and January 1946, these perinds heihg the only times available for the purpose, W ith Ernabella as the base of the survey, the ranges were investigated by horse, camel. and motor truck, but largely on foot, Transport difficulties and seasonal conditions limited the most intensive work to a tadius of 10 miles from the base. Considerable reconnaissance and some detailed work was carried out over a wide area, as far west as Mount Woodroffe and Erlywanyawanya, and Kenmore Park in the east. Positions uf camp sites and consequent geological work were often hampered by water difficulties. Concrentne MArs aAnp Metiions in 1892 Carruthers published a map of the north-west of South Australia, on a scale of £ inch to 1 mile, This appears to have been the basis of maps used by all later expeditions to the north-west. When a map of the Central and astern Musgraves was needed, existing maps, which had their uses for general purposes, were inadequate for more detailed work. [ence T was compelled to make my own maps, and used rough theodolite and compass methods for speed. ‘The aix-trigged points of Carnithers which fell within my area were incorporated, 197 with the exception of that of Mount Everard. The actual location of the hurler ig unknown to the author and to all at Ernabella. Nor do the natives know of any trigged mountain in the plotled posiiion of Mount Everard, The map of the Vicinity of Ernahella was prepared on a seale of 4 inches to 1 mite. The lacver arca was prepared on a seale of 2 inch to 1 mile, using some of Carruthers’ data when my own data and time were insufficient. Nruanerous patioraunc photographs were used to check as much detail ag possible, Some truck ntleages were incorporated. Owing to difficulty in location of parts of the track from Lrnabella to Ken- more Park, reliance bad to be placed om truck mileages and compass readings, Wath the truck “milcages" in need of calibration certain parts niist be regarded as sketch maps. This applies also to the area just north of Aluka, the viernily of the “pass” to the cast of Ombagunda, near Pig Tree Gully Seak and Unesan Dowrts. A pnémutic compass bias only hmited use in these ranges. The magnetic variation was found tu be 45° Lo the east of true north. The rocks of the area greatly affect the compass, so much so that great care must be taken a that instrument is to be used among the mountains. When on the plains, accurate readings can be obtained, The magnetic variation is tisually very crratice. Lven on the same hill | have noticed a difference of up to 5° ty 6° between readinys taken sitting and standing, although exactly im the same place, Bearings taken when one is enclosed in rugged ravines are often well-nigh useless, Mary hear- igs were found, on checking, tn be hopelessly inaceairate, he only way to take 4 reasonable compass reacing was found to be by standing in such a position that the interfering rack is equally distributed om all sides. Such being the case, a cumnpass suryey was olten found most tedious, Some type of sun-compass would dnubtless he easier to manage, Aboriginal names were used on the map where possibte. Care must be taken with these, however, for native placenames are rarely ag precise in meaning as our own. For instance, while pointing te Mount Spee, the author was told that th aborisinal name is “Inindi.” But it was later discovered that the mountuin has no particular native nae, but that *“Inind1” reters to the “region” af the rock hole situated at a considerable distance to the south-west. The probable explana- tion is that to the aboriginal the turmidable and barren Mount Spec has no need foria name, so when pestered by the white man for one, the nearest he can vive is “luoindi.”’ Similariv, “Tjatjanja” (ar “Tjatja’) probably refers to a rock hole the south of the mountain given that name. “‘Ltjinpiri” to ihe aboriginal iienotes the series af rock holes to the east of the imposing mouniain mass which the missionaries at Ernabella call Mount Itjinpirt. In the absence of suitable aboriginal ames several were coined. ‘The origin oft most of these will be obvious, but some need explanation. “Trudinger Pass” (snitable for harses het found aiffcult for camels) was named after Ronald Trudinger, friend and host at the Ernahella Mission, “Brock Pass" (probably negotiable with camels) was named after Richard Brock, who accompanied the author on the second expedition. “llenderson Rack Hole” is a fine water about which the aboriginals apparently know very littl because ij is just above a big cliff and away from normal hunting grounds, It was discovered while descending Mant Woodroffe and was named after Ronald Henderson, who skilfully drove the truck through the nearby difficult Brown's Pass. The large bare water catchments of the Musgrave Ranges make rock holes fairly common, but very Few are reliable. On the atcompanying map all reason- ably useful waters are shown, that is, those which were found with some water during the two dry seasons of the author's visit. 198 When vertical aerial photographs are taken'*? many parts of the accompany- ing map will, of course, need to be somewhat modified. However, it is considered that the map is a definite contribution to our present knowledge of the area, and little difficulty should be experienced in locating any position marked on the map. Therefore, apart from the geolory represented, the location of various features of the area (rock holes, passes, etc.) seems to justify its publication, Tae Term “CiwaArnockirn” The ranges coniprise a complex mixture of gneisses and ancient but later deep-seated non-gieissic igneotis tocks. Numeraus dolerite dvkes cut all of these earlier systems. Al rocks of the Musgrave Ranges are of Precambrian age exeepiing, of course, the thick mantle of Recent red desert sands and loans of the yulleys between the ranges. Most of the Musgrave Ranges rocks, both the gneissic and non-gneisste types, are hypersthene-bearing. Notable papers have been published im recent years trealing Chartiackites from several regions of the world, and it appears from the diverse origins which are claimed for these hypersthene-bearing igneous rocks, eneisses and granulites, that the term ‘“Charnockite’ has lost antich of what it was originally intended to convey. Ilowever, for the present, until petrogenctic relationships in the Musgrave Ranges have been more fully interpreted, these rocks may well be called “The Charnockites and associated rocks oi North-Western South Australia.” BRB. THE MAJOR ROCK TYPES I. Tr Gyrisses Loatnyarra To Kenarore PARK At Umbyarra Bore ancient gneisses were first noticed on the track west from Finke. Umbyarra is 64 iviles west of Winke in the Northern Territory. They outcrap as low hills annd remnants of horizontal sandstones of the late Mesozoic and early Cainozoic. At Utnhyarra some peginatites were obseryed, bat most of these are barren, One pegmatite conteins guod erystals of muscovite and some tourmaline. The general trend of the peeatites is N25° EF. To the west the ene'sses are more extensive, though they rarely form any more than low hills and mcanspicuous outeraps in the mulga serub, About 2 miles west uf Umbyvarra Bore a prominent but low range of hills, about a mile and a half to the worth of the track, was visited. This is probably typical of similar low lines of bills which protrude at intervals from the surround- ing’ flat country. The hills consist of a non-foliated biotite-bearing eranite; the flanks of the hills show excellent curtacts with the ancient gneisses, Nimmerous chunks of gheiss were noticed to be partly “assimilated” by the granite. A Few barren pegmatites are present. Siuilar granitic gnetsses continue to the west, but become more heavily injected by vast-west dolerite dykes towards “The Gap.” 122 miles [rom Finke, and approximately on the Sonth Austrahan-Northern Territory border. At The Gap” the gneisses show considerable crush, The doterites are similar to those of the Musgrave Ranges. Duscdow made observations concerning parts of this area of the “Ayers Ranges” (Hasedow 1905, p. 78). ©) Within a few days of presenting this paper to the Society a batch of 27 low. level oblique aerial photographs were sent ta the author by courtesy of the Australian Gtoxraphical Society, These were taken with a view to publication in “Walkabout,” the oraan of the society, bence the prime purpose of the photographs was not carto- graphic, They show the northern part of the Ranges near Mount Woodroffe, Trudinger Pass, Alalka, ‘Valjaritia and Wedge Hill. The photographs reasonably confirmed ihe accuracy Of the map in these regions. A few minor alterations were made. 199 At Beefwood Creek, 135 miles west of Finke and a few miles south of the Northern Territory border, an interesting pyroxene granite intrudes the acid gieisses. From Beefwood Creek the types of gneiss vary more frequently, The most common is granitic, but intermediate and basic types are plentiful. At Ken- more Park, nearly 180 miles west of Finke and just within the area covered by the map, the variety of gneisses is great. Garnet-bearing and/or hypersthene- i | | i 1 NORTHERN TERRITGRY | 1 ' AO" oe \ FINKE | | a ery oan - | Ve ~- {22785 Lae - be Fae k a! —- so ae ir ieee i Oe ener RO ia en hoe el agen ae BES" yy ysGnRne Res, ( ; Re t ¢ Ws . sy et” \ / A 4 ! { a os ! ss | —_-*gOODNADAT TA | a. | \ he. i i} Bae . Po | / | { MARRED i 1 | COLDEA | »TARCOOLA ) | } Hike ! SOUTH AUSTRALIA | i ! SHOWING POSITION OF ] | ERNABELLA 3 7 ae t , = ] Fig. 1 Locality Plan bearing gneisses are abundant. The general trend of all of these gneisses from Umbyarra to Kenmore Park is north to south. The dip varies from 60°-70° to the east or west. Tur Mayor AREA AND GENERAL I'EATURES OF TILE GNEISSES The area to the west of Kenmore Park was studicd in more detail. Irom the map, the importance in this arca of gneisses and granulites is readily seen. Most of these gneissie rocks are a dove-grey colour with streaks of dull dark brown pyroxene and occasional black shining amphibole. 200 The map shows that the gneisses commonly have a north south strike“) with a steep dip to the cast, Towards the north-western part of the area, however, the strike of the gneisses swings around from a few degrees west of north to take on an almost east-north-cast trend near Kuna-mpunja and Wardulka, The dips are such as to suggest that the western half of the area is part of the westerit limb of a south-pitching syncline, Anomalous dips and strikes are not uncommon, but most of these are close to junctions with the non-gneissic igneous rocks.‘ These may represent minor buckling, reversal af dip and possible overthrusting related to the origin of the igneous rocks, Further detailed work is needed in critical areas. The question of the origin of these giieissic rocks is difficult, and ir would be premature at this stage of the investigation of the rocks of this region to state final conclusions as to their petrogenesis, However, certain suggestions are put forward, and these are linked with the ideas of “palingciesis” and “granitiza- tion” considered as having acted on a grand scale (see p. 209). From the held evidenee ile author considers that the area consists of a series of ancient highly metamorphosed sediments whose eventual granitization has brought abotit most af the non-gneissic igneous rocks. However, in hand specimen, the sedimentary origin of the gneisses is usually not so obvious, The following observations on some of these gneisses are nut primarily con- cetned with the setting out of petrogenetic relations, ‘These will appear in sub- sequent papets. However, the order of the gneissic types is probably chronological: thal is, the racks first described are the oldest, (1) Gneisses. showing sedimentary origin. (2) Gneisses. not showing obvious sedimentary origin, (a) Acidic gneisses without flioresciny zircon, (b) More basic gneisses without fluorescing zircan, (3) Gneisses not showing obvious sedimentary origin, Mu showing fluorescing zircon. (1) Gyeisses suowInG SEDIMENTARY OrICIN, The gneisses showing an obvious sedimentary origin are not plentiful, The recognition of rocks as of sedimentary origin can often be difficult. In this area of charnockitic rocks, plutomie conditions resulting in palingenesis and granitization ure assumed to have been in force for considerable periods. Rocks of the arkase type, 1f present under such conditions, could be presented in such a form as io defy recognition of their origin. Argillaceous and highly calearedus racks probably could be more casily identifted, No rock was found whieh obviously came trom a highly calcareous rock. Wut there are some for which an argillaccous origin is suspected. ‘These areas are of limited extent: (1) part of the complex gneissic system to the south-west of Mount Cur- ruthers ; ©) The term “gneiss” is here used for rocks showing vatying degrees of banding. The word “gneiss” is preferred to “foliated rock,” since the banding is considered to represent the bedding planes of the original sediments, now pneissified. The ternis ‘din’ and “strike” are therefore used to describe this handing, No schists were noted, () The term “non-gneissic igneous rock” igs meant to convey the idea that rye major igneous rocks of this urea liave heen derived from sediments, the gneissic atruc- ture of which has become obliterated in the process of formation of the "igneous masses. 201 (2) possibly the greater part of the gneisses to the north-west of Pulpatjara Well; (3) small patches 1 mile to the noruh-west of Taljarit}, and 14 miles to the west-suuth-west of Talfaritja; (+) patches 1 mile worth-west of Top Springs; (5) patches up the Ernahelia Creek from Top Springs; (6) some of the gneisses to the north-east of Erlywanyawanya Roek Hole: (7} 23 miles due west of Mount Carruthers; (8) 1 mile north-east of Taljaritja Well. These rocks contain variable ainounts of at least Sour of the following minerals garnet, cordierite, sillimanite, biotite and a green spinel, The garnet is always visible in hand specimen es pink rounded crystals, usually about 2 mm. in diameter. The cordicrite in the field is not noticeable in most specimens, though the presence of a “greasy” dull yeliowish-brown wilh the garnct Icads one 10 suspect that mineral. (but “greasiness” is by no means necessarily indicative of cordierite, for most of the hypersthene-bearing rocks of this area show some “preasiness”.) Only in one locality, (7), has macroscopic sillimanite been observed. It occtirs as brown crystals up to 2 em. Jong, The dull dark green spinel shows very high ahsorption, so much so that with a cursory glance at 4 slide, one could quite easily overlook it, grouping it with the iron ores. It usually oecurs intergrown with iron ures, Sillimanite is commonly aligned within the spincl, but ar localities (7) amd (8) the sillimanite also occurs as plentiful euhedral crystals. In all of these rocks biotite is plentiful. However, in the normal gneisses and non-gneissic charnackites to be mentioned below, biotite occurs usually as a yery minor accessory, It should be stated thar the presetice of garvet in acidic rocks must not be taken as necessarily indicative of direct sedimentary origin in this region. Garnet has been nated in many other places. byl usually near the junctions of the main Frnabella non-gneissic “igneats’” mass and im the gneisses around certain “cupolas” to the east (é.g., Spinifex Hill), and often in mylonized zones, Some analytical data and petrographic descriptions have been assembled, but will be withheld tor ielusivi in later papers. However, it may be stated that in the analysis of an “injection enciss” over 5% corundum is indicated in the norm. The author considers that the rocks which show evidence of sedimentary origin are often greywackes of perhaps relatively high alumina content, and that later investigation may show that kurge tracts of the gneisses can claim palin- genetic rélationships. with arkosic rocks, (2) GNEISSES SHOWING NO Onviods SeniatrANrany Orta. (a) Acidic Gueisses withait Finoreseing Zircon. Many specimens representing over 00 localities have been carefully studied under ultra-violet light. Tn large munibers, a zircon has been revealed as Hiny points of orange light, Sometimes the crystals ate tip to 5 mm, but ustially of the order of 1 mm, in length. In most of the gneisses, however, no fluorescence was observed. and petrographic work has revealed that a somewhat different non- flavrescing zircon (?) exists in these rocks. The author hopes (hat it may be possible lo trace petrogenetic relationships by making us¢ of this zircon criterion. The areas represented by the Anorescing zircons are indicated in the sketch map, p. 202, 1n hand specimen, little indication is given of the presence of otherwise of the Auorescing zircon. Lowever, those with Auorescing zircon do seem to exhibit a somewhat more “igneous” appearance, E 202 The present indications are, however, that the gneisses without the fluorescing zircon are somewhat older than those with the fluorescing zircon. Investigation into this interesting phenomenon is at present only in the early stages. A number of modes have been established by the Rosiwal method, and same analytical work has been undertaken. (Details will appear later.) From this it was found that all gradations from granitic to granodioritic recks occur within this acid group. At average type (adamellitic) occurs at the little rock hole, 2 miles west of Top Springs. This group of gneissic and sometimes granulitic rocks are dove-grey and of a greasy lustre. The dark minerals are usually concentrated into rough, more or less continuous plates throughout the rock, and there are usually sume rude bands of feldspathic material. These strike north-south, with a fairly steep dip which varies from east to west. The rocks are usually fine-medium grained. Microscopically these gneisses and granulites usually show rather rounded yet REA CFFLUORESCINGGNEISS — Ee | longqus Rack 6 B Miley interlocking grains, with the ferromagnesians of slightly smaller dimensions than the salics. Some quarte yrains are particularly ragged and associated with myrmekite, while others protrude as smalf irregular “fhngers”’ through the rock. The plagioclases are tstially twinned on albite and pericline laws with composition about andesine (An.33). They ate variable in amount, but are commorly plentiful, Myrmckite is Giten assuciated. Quartz at times contains numerous non-orientated needles. Shadowy extinction is common. The “fingers” of quartz are mentioned above. The potash feldspar is usually abundant and represented by an orthoclase microperthite, but sometimes by a poorly twinned microcline microperthite. The proportion of the other minerals yaries greatly, but hypersthene is usually the cotamonest. The pleachroism is strong in rusty pinks and greenish-greys, with a slightly variable optic axial angle (negative) of about 80°. This pyroxene often shows considerable cracks filled with greenish alteration products. A pale green non-pleochroic motiochine pyroxene (probably diopside) is usually present, but hypersthene predominates. 201 Hornblende is sometimes relatively plentiful but is often absent. Pleochroism is usually yellowish-grey, light hrown, dirty green, with ZAc 23°-25°. Magnetile is often abundant and at times an important constituent. Secondary magnetite is commonly deyeloped ueat hypersthene. Biotite is asso- ciated in small amount with the iron ores, Pleochroism is pale straw-yellow to orange-brown. Zircons (?): present indications are that all of these crystals, which are non- Huorescent, are usually murky and tending to be brownish and much cracked- Apatite is commonly present. On chemical analysis, a lypical gneissic rock revealed the normal features of ati “igneous” rock of adamellitic composition, Such rocks could well be called acitlic giieissic charnockites. (b) Intermediate and more Basic Gueisses without Flucresciug Zircon. ‘These are best represented by some of the dark gneisses at Gilpin's Well, although patches of similar rocks occur near Kuna-mpunja, Brown's Pass, and elsewhere. In these rocks the gneissic structure is usually marked and ts revealed on the weathered surface by streaks of tiny greyish ragged quartz crystals, and shiny magnetite, On fresh surfaces, specituens are a bluish dark grey (contrast the “dove grey” of acitlic gneisses} with the typical greasy lustre of the other hypersthene-hearing types. Microscopically these rocks are gratioblastic, with an ayerage grain size of the plagioclase just under i mu. in diameter, and that of the ferro- magnesians about 0-5 mm. Modal proportions are Plagioclase 65%, Ilypersthene 14%, Diopside, 79, Quartz under 59%, and Hornblende, ete., 9% (Gilpin's Well). The plagioclase is usually a well-twinmed labradoritic antiperthite (An. 45) with plentiful opaque needle inclusions. The hypersthene is fresh and displays the usual strong pleochroism. The monuclinie pyroxene is pale bluish green-grey, non-pleochroic, and has ZAc = about 45°. Quartz usually oveurs as the finger-like protuberances so commonly seen in the more acidic gneisses. The mineral ofteti bas needle inclusions similar to those in the plagioclase. Mugnetite is usually quite plentiful in these rocks. The zircon (?) 1s non- fluorescing and of the murky type, noted above in the acidic gneisses. Some apatite and hintite are commonly present and associated with the magnetite. Potash feldspar and hornblende are very uncommon, Such rocks could well be cafled quagtz-dioritie gneissic charnockites. (3) GNuISSES SHOWING NO OBVIOUS SEMIMENTARY ORIGIN AND CONTATN- ING FLUORESCING ZIRCON. From the sketch map on p. 202 it can be seen that these gnheissic rocks appar to occupy mainly — (a) an east-west belt from south of Monnt Woodroffe through south Trudinger Pass and Brock Pass to the Inindi area; (b) the east-west Mount Cuthbert Range to the north; (c) the east-west belt from Mount Ferdinand to Ombagunda; and (d) many other minor localities. The strike of these gneisses is still the same north to south of the whole region, with the dip exceeding 45° east or west, excepting in the area near the Wardulica Rack Hole [Locality (b) ]. Here the gneisses take on a imore of an east-west trend. 204 As was suggested above (p, 201), it seems that these gneissic rocks are a little coarser (thongh still far from coarse-grained) and look more “Tpneous” than the gneisses, free irom the fluorescing zircon, As in the first group of gneisses, this gronp shows gradations from the highly acid to the granodioritic and quartz-dioritic types. The normal grain size is found to be about 1 mm. in diameter, with the ferromagnesians perhaps a little smaller. The rocks show the customary greasincss of all the hypersthene-bearing rocks of this region, In most, the typical doye-grey colour is evident, though some specitnens show more of a fawi-grey, owing, perhaps, to alteration, Orthoclase microperthite is usually plentiful, bat microcline microperthite is only oceasionally present. At times, an acid andesine is the dominant feldspar, This plagioclase is commonly. antiperthitic. A quick method for distinguishing the perthitic from antiperthitic feldspar was noted. This consists in viewing them tsder considerably reduced ilhimination. The antiperthitic particles ther appeared pale pink, while those of the perthite remained colourless. Quartz occurs as ragged tongues and scattered like millet-seed throughout the putash feldspar in varions rocks, Occasionally non-orientated ncedle- inclusions are developed in the mineral, The dominant ferromagnesian iniverals are a typical pleochroic hypersthene and a pale bluish-green non-pleachroic diopside. These are commonly arranged in rough bands which are separated by the salic-rich hands. Common hornblende has Leen noticed, but it is not typical of these gucisses, Iron ores are often plentiful and are conmmonly interlocked with the pyroxenes. Apatite oeurs in variable amount, hut mostly associated with the pyroxenes and iron ores, Seyeral crystals of zircon of the fluorescent type occur in all slides. The fluorescing zircon is clear and colourless under convergent light, whereas the non-fluorescing zircons (?) ar¢ almost always murkier and, under con- yergent light, usually show a very pale pinkish-fawn tint, II. Tre Non-Gyeisstc CHARNOCK!ITES Large areas of non-gnetssic rocks) are shown on the map. Most of these would usualy be called igneous rocks. Though these are younger than the gneissic rocks, the final period of crystallisation of the non-gheissic “igneous” rocks is probably not far removed from that of some of the gneissic types. At present these non-sneissic rocks may be divided into four divisions. In this scheme the oldest types are listed first, he four divisions are — (1) The Mount Woodroffe and Trudinger Dass Intermediate and Basic Thocks, (2) The Upsan Downs Granitic Rocks. (3) The Norite to the south-west of Tjatjanja. (4) The Ernahella Granodioritic Massif. (1) THe Motnt Wooprorre-Trupincer Pass Ixreraepiate anp Basic Rocrs. (a) The Mount Woodroffe rea, The map indicates that on all sides of the Mount gneissic rocks predominate, but near the summit considerable areas of quartz-dioritic and dioritic rocks occur. There are basic and ultrabasic patches in places. Directional characters were not noted in any of these charnockitic rocks on Mount Woodroffe. However, the strike and dip of the surrounding gneisses, and the comparison with the basic ©) See footaote (? on p. BOO 205 rocks which continue itt a north-easlerly direction to Trudinger Pass, suggest that these rocks probably comprise a large sill-like structure. This is more evident in the range between Brown's Pass and Landmark Hill. Directional characters were not noted in any of these rocks, bur suine may be expected when field work can be carried out on the cross-section of the apparent sill structure.‘ Recon- naisance only has been possible in this area up to the present. In hand-spechnen the rocks of Mount Woodroffe are medium-grained, con- sisting of pale amethyst grey and greasy bot shiny plagioclase, mottled with patches of dull black pyroxene, Twin striae are not easily seen in the plagioclase in band-specimen, bul Under the microscope albitetwins are plentifid. This antiperthitic feldspar (70%) 1s a tid-andesine. Hypersthene usually occurs as large tabular rounded crystals about 4 mm, x 3mm, It is very strongly pleochroic, Diailage is also pyesent. Chiartz, is noted in accessory amount. Potash feldspar hay nor yet been sceh except as the anti- perthitic lenses, Zircon is quite rare and of the non-fluoreseing type. Hypersthenites appear as patches np to an acre i” area on the sides of Mount Woodroffe, Another outcrop occurs in [rown's. Pass between the Woodroffe vocks and the basic rocks stretching north-east to Trudinger Pass, These ultra- basic rocks are commonly associated with magnetite and very coarse: dinritic and noritic rocks similar in appearance to those of the summit described ahove. The hypersthenites are fresh, dense homogeneous iedium-grained, greasy and very dark grey rocks, Micvoscopically, they are found to consist almost entirely of pyroxene, most of which is a strongly pleochroi¢e hypersthene. There is a little non-pleachroic pale green diallage. Plaginctase is usually present in extremely small amount, as tiny interstitial particles, Very little primary iron is present iv slides studied to date. bil in the field, solid boulders of magnetite are common as floaters near the hypersthenites. Directional features were not observed. (b) The Trudinger Pays Basie Rocks. As suggested above, these noritic rocks are linked with the Mount Wood- rofte rocks, ‘The outcrops in the Pass were noted during: a particularly arduous reconfiaisance on caniel, and time coyld not be spent trying to determine the yelation of these rocks to the gneisses. The grain size of this hasic suite varies greatly. Sume textures are those of the very coarse norites, others are doleritic, but the average is that of a medium grainedL norite. The composition varies somewhat from a charnockitic basic dtorite toa typical norite, In hand-specimen the diorites are similar to those of Mount Woodroffe, but a little darker. They are also comparable microscopically. In hand-specimen the norites, however, are coarse dull brownish-black heavy rocks composed mainly pi large dull grey plagioclases, poikilitically including numerous small preasy grains of pyroxene. Albite-twinming is not obvious 10 hand-specimen, There are several large clots of dill greasy dark brown pyroxene crystals. Microscopitally, these norites consist of over 6096 basic andesine with the test of the rock composed of hypersthene and diallage, whose propertics are similar to those of the Mount Woodroffe roeks. Tron ores are not plentiful. No fluorescing zircons occur in these rocks. OY The aerial photographs (see footnote, p. 198) suggest that there is very little @ucissic structure im these rocks. Some of that anggested in the pholagraphs of the north-west Hank of Mount Woodroffe may be linked with the profound mylonitization known to veeut along the northern flank of the Mount, 206 (2) Tne Ursan Downs Granitie Rocks, Litt'e work has been done in the Upsan Downs area. These medium-grained granites are characterised by large flesh-coloured feldspar crystals, Considerahle crush has taken place, and in some localities beautiful crush-augen gneisses are developed (¢.g., the Pass one mile to the north-east of Victor’s Well, and at Erlywanyawanya Rock Hole to the south-west uf Mount Woodroffe). These rocks contain no fluorescing zircon. At the tock hole in Brock Pass a granite vccurs among contorted and partly “assimilated” gneisses, This granite shows fluorescing zircon, thus, although show on the map by the symbol used for the Upsan Downs Granite, it may well repre- sent another epoch. (3) THe Norte to tre Soutti-west or TyAryAnya. These norites comprise a considerable belt on the spur two or three miles south-west af Tjatjanja. The rocks from this locality vary in grain size, ranging from coarse norile to micronorite. Normally the racks are mottled greasy, dark blue-grey medium-grained, and the abundance of a resplendent bronzy biotite is the most striking feature. | Noted by Jack (1915, p. 18) J. The plagioclase shows prominent albite-twin striae. Under the microscope, half of the rock appears to cotisist of an acid labradorite, It is often associated with the hypersthene poikilitically, The plagioclase is an antiperthite. Hypersthene and diallage are present in equal amount. Mapnetite is scarce, as also is soine imteftstitial orthoclase microperthite. No fluorescing zircon was noted, A small orthite-bearing pegmatite similar to that at Koli Koli (see p. 208) cuts these norites near the northern-most junction with the charnockitic granite af the Tjatjanja spur. This alone suggests that these basic rocks are older than thase of the Ernabella massif, (4) Tite ExnAperta Granonroritic MAssip, This rock type oceurs as a great meridional helt with Ernabelia near the centre. Though the belt is granitic around Tjatjanja ta the south, the massif, taken as a whole, is probably granodivritic, The map shows “cupolas” of similar inaterial in several other places. notabl+ among the gneisses :--- (a) South and north-east of Koli Koli rack hole to the west. (6) South of the Stony Jump Up to the east. fc) Bald Hill and Spimifex Hill further to the east. Tt is regretted that owing mainly to difficulty of terrain and climatic condi- tions, many of the junctions between the gneisses and the non-gneissic acid rocks could not always be studied as well as ane would desire. They have been located on the map, however, to facilitate future investigation. The best junctions were found on the western side of the massif inv the Mount Carruthers area near Ernahella, and at Alalka to the north. In these localities large sheets of fresh rock are exposed in the water-courses, and the study of xenoliths and other phenomena is facilitated thereby. The eastern junctions are not readily studied, Many of these occur beneath sandy valleys. The junctions in the Tjatjanja area are relatively broad zones, but xenoliths are not common. In the Mount Carruthers arce the trend of the xenoliths in the Ernabella massif is between N.25" EB, und N. 35" E. with an approximate dip of the rude xenolithic lenses of 70° to the east, Small aplitie veins are often found with 207 identical dip and strike of the xenoliths, but the pegmatite veins are almost always trending W.10° N. and dipping south 75°-80°, ‘The strike of the gneisses ts usually north-south, hence these xenoliths are usually at an angle of appraxi- mately 25° (o the gneisses, This may indicate some pre-crystallisation movement of the non-gueissic “igneous mass” relative to the gneisses, from which it was probably derived by granitisation. In the dlalka Rock Tole area, xenoliths are common. The trend is between North and N, 18° E., the usual being about N. 15° EZ, No evidence of aplitic or pegimatitic activity was found, In all localities where xenoliths could be studied, a great diversity of xeno- lithic rock types was apparent. The xenoliths which are most obvious are armphi- bolitic, but on closer investigation partly assimilated clongated fragments of gneisses of all types outlined above are found to be the more abundant, Xenoliths in the area south-west of Mount Carrtithers were traced to their probable equiva- lents in sifie in some instances, A common size for the lens-shaped xenoliths ts about 20 ems. x 5 ems, x 10 cms, Many larger and smaller ones were found, A feature af many of the acidie inclusions is the presence of biotite which has developed throughout the lens. Basic inclusions usually show a marked reaction rim of hornblende. It iy hoped later to publish some microscopic observations on the xenoliths. The conspicuons paucity of pegmatites, aplites and quartz veins mdicates the special “dry” conditions under which these post-gneiss hypersthene-bearing acidic rocks crystallised. The Racks of the Eraabella Massif, whether granitic or granodioritic are very similar in most of their characters, In hand-specimen they are medinm- grained, greasy bluish dark grey granitic rocks composed of bliish-grey feldspars of medium size, small blue-greyish glassy quartz irregularly scattered, and small ragged areas of greasy dark brown ferromagnesians. In several localities large phenoctysts of plagioclase are present (up lo 6 cms. x 2 cms,), and these with fhe rest of the plagiaclase usually shaw splendid albite-twin stnac. Phenocrysts ol the pyroxencs are found at times, Under the microscope these rocks are holocrystalline and more or less allotriomorphic granular. For the most part the ferromagnesians are of somew bal smaller dimensions than the. salic constituents (average of rock 1-3 mm, to 2 mm, long). The plagioclase (50% ) is a mid-andesine and well twinned, It is sometimes a little antiperthitic. Orthoclase microperthite (about 207%) contains rounded quartz inclusions (millet-seed type) i considerable number. T hese inclusions ate uncommon in the plagioclase, Ouarlz (15%) is ragged and interstitial except when included in the polash feldspar. Myrmekite is plemiful. ‘Ine tongued character of the qiartz (so common iu the gieisses) is an uncommon feature of these mon-gneissic charnockitie rocks. Preferred orientation of the dark minerals is not present io any marked degree cither in the field or in microslide, flypersthene is not so notably pleochroic as itt most gneisses and racks of the Mount Woodroffe area: A weakly pleochroic or non-pleochroie pale gryenish- grey monoclinic pyroxene (probably a species of diopside) is plentiful. There is no apparent age difference in the two pyroxencs- Apatite and magnetite arc plentiful. Hornblende occurs, but usually in small amount and has crystallised after the pyroxenes. Hornblende is plentiful, however, in the cupola phases, for instance, at Spimfex Hill and Bald Hill, Biotite oceurs as sinall faking crystals on some of the patches of magnetite, Zircon is plentiful and is of the fluorescent type, 208 Analytical and petrographic datu have been collected, but are being withheld for lhe present, Tt should be noted ttal this rock type fas all the characters of a “nositial igneous rock.” It is hoped to show later the palingenetic relation of this “igneous type” to the gucisses, Ill. 3 Peasxatirrs Pourly defined small pegmatites outcrop throughout the Ernabella plain among the charnockitic granodiorites. Only occasionally are they found among the gneissic rocks, The Musgrave Ranges, as a whole, are very poor in peg- matites. An ititensive search was made for these, and specitnens and notes have been taken of each one found, Aplites are much rarer. Mast pegmatites are usually only about one foot wide. and less than three or [our chains long. Their strike 1s almost always West 10° North, with a dip Sonth 75°-80°. In places dolerite dykes are found cutting across these pegmatites. The minerals noted are set otit in approximate order of abundance. Heematite is otten in well crystallised masses, Fiolite is usually much contorted, and ragzed books oceur up to 7 ems. in diameter, Large strcons are often found embedded. No omscovite has been found anywhere in the ranges. Hornblende—This mineral is the dominant mineral cf many pegimiatites, and has heen noted us Iurge black shining erystals up to 8 cms. in length. Garnet eccurs in some of the pegniatites ay dull greasy reddish-brown rounded subhedtal crystals up to 3 ems. across. Orthite (Allanite)—Oceasional small pockets of large tabular crystals were noted in a few of the pegmatites of ihe Ernabella plain. The mineral is jet black, and has no cleavage but a pitch-like conchoidal fracture. Specific gravity is about 3°5, and hardness roughly 6-5, Thongh too weak to affect the electroscope, a very Weak radio-activity was revealed after an exposure of wp to three weeks on photo- graphic plates. Throughput the ranges only a few pounds of the mineral were obtained, though a thorgugh search was made. Anulytical work is in progress to use this mineral in age-determuination. Zircow js present in certain pegmatites, as black or dark brawn euhedral crystals, up to 2 ems, long, Though very weakly radioactive (about the same as the orthite), they do not fluoresce under altra- violet light. This is remarkable, for the zircons of the parent rock, the charnockitic granodiorites of Ernabella fluorescence in orange. Preliminary iivestigalions with autoradiographs indicate that the fluorescence decreases regularly as the radigactivity and depth of dark brown increase. Apatite occurs as tiny etthedral pale blue crystals, but is rare, Yorrtmalixe was found in only a few cases, and is of the cominon bluck variety. Among ihe gneisses true pegmatites are tarcly found. At Koli Kol Rock Hole an imteresting but quite small hornblende pegmatite occurs, in which nail- hike erystals of orthite and an orange-iucrescthg euhedral resinous “zircon” occur. Similar pegmatites at Trawanja and Arkalanja show neither orthite nor zircon. Work is in progress on the pegmnatites. TV. Dornernrtre DyKes Cutting all of the other rock types, dut probably nor far removed im age from the Ernabella massif, is a large suite of dolerites. A common trend is a few degrees south of west, with a dip to the south of 60°-70°, Another set is present near Ernabella, with an approximate north-south strike and dip to the east nf 25°-40°. Similar dyke-rocks occur all the way to Umbyarra, about 140 miles ta the east, and according to Dr. Jack, far to the south-east into the Everard Ranges. and beyond. Dr, Jack observed (Jack, 19L5, pp. 15, 18)in these regimrs to the south-east, that the basic dykes cut the gncissic and non-gneissic rocks, but not 209 those of the Adelaide Series (Proterozoic—the “Cambrian” of the earlier geolo- gists), These dyke rocks, as scen in the Musgraves, consist essentially of labra- dorite lathes, with hypersthene, diallage and olivine combining to give the typical ophitic texture of dolerite. Specimens [rom about one hundred dykes have been collected for description, V. MyriowrrizAtion Most rock types of the Musgrave Ranges have been mylonitized in places. A large collection of material showing transition types lias been made for future work. Dseudo-tachylytes are well developed. Both gneissic and non-gneissic types may show a crushing and a rolling ont. They are often found to grade ihrough a crush zone of mylonitized augen- gneisses to dense halleflinta types. The gneisses so farmed are not to be confused with the gneisses outlined aboye. The crushed rocks are only in association with major fault zones. Thongh considerable arcas of crushed rocks may be of later origin, most of this mylonitization seems to have taken place after the pegmatites were emplaced, and is probably associated with the widespread epidotization noted with the joint- ing which led to the dolerite intrusions. The coincidence of the west-south-west trend of the dolerite dykes and major zones of miylonitization is evident in the following cases, Intense mylenitization (with pseudotichylytes), epidotization and basic dyke injection are plentiful throughout :— (1) the tectonic valley stretching from Upsan Downs through the valley just north of Tjatjanja and Mount Ferdinand to the Gilpin’s Well and Big Hill localities; (2) the tectonic valley from Taljaritja Well through Top Springs and along the creek south of One Tree Hill: and (3) the great tectonic valley running along Tietkin’s Creek, through Nalja- wara, Ernabella, Brock Dass and into the Oowallinna areca. There is also a coincidence of these same features with the weaker develop- ment of the north-south basic dykes in the Ernabella. area, Tt is apparent that the conditions which allowed typical mylonites ta form mist have been considerably different from those prevailing at the depths where the charnockites crystallised. Hypersthene is always absent from these mylonitized rocks, and hornblende and garnet with biotite and marked mortar structure are characteristic. C. CONCLUSION AND SUMMARY The Musgrave Ranges consist of a complex series of many rock types, all of Precambrian age. Present indications are that the chronological order of the major rock frotips seems to be the following —- 1. Guneisses showing sedimentary origin and represented by garnet, cordierite, spinel and sillimanite gneisses, and containing no fluorescitg zircon, This group is thought to represent the oldest rocks in this area, 2. HUypersthene-bearing Gneisses Sor showing obvious sedimentary origin but containing No fltcorescing zircon. 3. Huxpersthene-bearing Gneisses Not showing obvious sedimentary origin but containing a flworescing zircon. 2i¢ 4, Hypersthenc-bearing non-yneissic Intermediate and Basic Rocks of Mount Woodroffe and Trudinger Pass. . These haye no fluorescing’ zircon. 5. The Upsan Dawns Granitic Rocks. These have no fluorescing zircon, 6. Hypersthene-beuring non-gneissic Granites and Granadiorites of the Ernabella Massif. These “igneous” rocks contuin fluorescing zircon and xenoliths of the guicisses, 7, Pegmmatites of the Ernabella area. 8. Mylonitisation and Basic dyke injection. The hypersthene-bearing rocks may be called charnackites. The position m the above sequence of the numerous altered basic dykes and masses of several epochs has yet to be determined. The gneisses of group 2 are thought to represent, in part, the granitized equivalents of the ancient sediments. The gneisses of group 3, with their fluoresc- ing zircon, are tentatively considered to represent possibly a superimposed meta~ morphism (with the introduction of considerable new material) of some of the eatlier gneisses. This is well seen at Kuna-unpunja, where patches of the rela- tively coarsely-grained aad more granitic rocks of group 3 cut across acid, inter- mediate and basi¢ types of group 2. Similar directional features are always present in both groups, but those of gtoup 3 are almost always much less marked. A criterion suggested for the differentiation of the gneisses of group 2 and group 3, therefore, rests on the critical problem of the origin and introduction of the fluorescing zircon. The non-gneissic igneous rocks of group 6 may indicate a relatively mobile and somewhat contaminated equivalent of the gneisses of group 3, because these, as stated, contain plentiful fluorescing zircon and xenoliths of all types of gneiss, The rocks of group 4 and 5 contain no fluorescent zircon and are thought to have crystallised alter those of gronp 3, The ubiguity of hypersthene, the relative rarity of hormblende and biotite and the marked paucity of pegmatites signify the “dry” conditions under which these charnockitic rocks finally crystallised. D, ACKNOWLEDGMENTS I wish to express my indebtedness primarily to Professor Sir Douglas Mawson, and also to Messrs. A, W. Kleeman and H. E. E, Brock of the Geology Department of the University of Adelaide, for interest and advice in many matters. D. R. Bowes kindly made two chemical analyses. Without the willing co-operation of the Superintendent (the late Rey, J. Love), staff, and natives of the Presbyterian Mission Station at Ernabella, the field work during the two seasons would haye been impossible. Mr, Trudinger, in particular, rendered yaluable hospitality, Finally, I must thank R. Brock for his outstanding assist- ance while accompanying me on the second expedition to the area. E. BIBLIOGRAPHY Basenaw, 1H. 1905 “Geological Report on the Country Traversed by the South Auatralian Government North-West Prospecting Expedition, 1903,” Trans. Roy. Soc. S. Aust., 29 Basevow, H. 1915 “Journal of the Goyernment North-West Expedition (March to November 1903). Proc. Roy. Geog. Soc, Australasia, S. Aust, Branch, 15 Trans. Roy, Soc. S, Aust., 1947 Vol. 71, Plate it Fig. 1. Looking north-north-west from Naljawara, 34 miles east of Eruabella. Note how island-like hills of non-gneissic charnockitie gramodiorite punctuate the plain. Fig. 2. North-north-east approach to Trudinger Pass, taken two miles south-west of Kuna-unpunja, Most of the rocks are of gneissic charnockite, with norite in the pass. Tjakunja on left; Landmark Hill on right in distance. Vie. 3. Mount Spec, taken from northern junction of mica norite to south-west of Tjatjanja. Most rocks are acidic and intermediate gneissic charnockites. Fig, 4. Non-gneissic charnockitic granodiorite cut by parallel dolerite dykes. South-west tip of hills, two miles west of Palpatjara Well, Trans. Roy, Soc. §. Aust., 1947 Vol. 71, Plate Ill Fie. 1. Looking north fron Mount Carruthers, showing north-south gneissic charnockites in distance on left, Top Springs and Ernabella Creck gorge in centre in distance, and Mount [ijinpiri and other non-gneissic charnockitic granodiorite areas to the right. Fig. 2. Looking north from east side of Razor Hill, nine miles north-vorth-east of Ernabella. Razor Hill consists of non-gneissic charnockitic granodiorite; hills to right are gneissic and in part show obvious sedimentary origin, Trans, Ray. Soc. S, Aust., Lo4F Vol 7l, Plate IN big. 1, Looking east frown suniniteal Mount Woodroffe (over 3.00 feet, highest mountim in South Australia). Mount Spee is flat-toppec mountain on deft iA distance. A stunted species of “spinifex” (Triodia) is abundant. Fig. 2. Looking west from simimit of Mount Woodroffe. Foreground of noin- gueissic charfockitic quartz-diorite with gntissie wranitic charnoclite oy rivht Major vetetaton—sturnted spevies of Vriocdin. Trans. Roy. Soc. S. Aust., 71, (2), Dec. 1947 SENTINEL HILL 4 ’ MATUNGA HILL MT.CUTHBERT® ~ A, WILSON i : ‘ 4 MITCHELLS KNOB 5 +. * : + OR HIL we WEDGE HILL FIG-TREE GULLY RLH . Va ‘ Sse eer saRUNNING-WATER SOAK ‘ . . ’ 7 ¢ 7 4 \ : ONE TREE HILL2) | ; . , RH! > 1G ALL ra 1 MT.CARRUTHER , * GILPINS WELLea "oo. SEpe we . XA ~a.ie* > s ee Se ae ' aw . 7 : “S! ) KENMORE PARKA “Fs \We 6G > “MTFERDINAND « . ‘y oe *, . a ye ae * a | . eee ' MTWOODROFFE . - . ‘ ‘ \ TJATJANJA +t jt ' *. tw +H Tey F ne 7 USGR SA GEOLOGY & Centra aE astern MUS AVE. Res. SAUs [+] GNEISSIC CHARNOCKITE (OLDEST Rocks) fro STRIKE & DIP OF GNEISSES PAZ] CHARNOCKITIC DIORITE & NORITE / VERTICAL GN. + _ HORIZONTAL GNEISS E a HYPERSTHENITE a TRIG POINT OF CARRUTHERS [xx 9 GRANITE » UPSAN DOWNS PROMINENT HILLS FeeN%] BIOTITE NORITE eRH ROCK HOLE H+] CHARNOCKITIC GRANITE & GRANODIORITE [__] RECENT FixeED REO SAND f———] BASIC DYKES (Post CHARNOCKITE) *S, UPSAN DOWNS = Pr SCALE IN MILES 211 Brown, H. Y. L. 1890 “Report on Journey from Warrina to Masgrave Ranges.” Parl. Pap. No. 45 Carrutuers, J. 1892 “Triangulation of North-west Portion of South Australia.” Parl. Pap. 179 “Erper Exproration Expepition, Scientific Results of” 1893 Trans. Roy. Soc. of S, Aust., 16 Forrest, J, 1875 “Explorations in Australia.” London Gites, FE. 1872-1874 “Geographical Travels in Central Australia, 1872-1874” Gossr, W. C. 1874 “Reports and Diary of Mr. W. C, Gosse’s Expedition in 1873.” Parl. Pap. No. 48 Htsse, S. G. 1897 “Stock Route Expedition from South to West Australia.” Parl, Pap, No. 51 Jack, R. L. 1915 “The Geology and Prospects of the Region to the South of the Musgrave Ranges, and the Geology of the Western Portion of the Great Australian Artesian Basin.” Geological Survey of S. Aust.,” Bull. 5 Macrice, R. T, 1904 “Extracts from Journals.” Parl. Pap, No. 43 Murray, W,R, 1901 “Report and Map.” Parl. Pap. No. 148 We tts, L. A., and Grorcr, F. R. 1904 “Reports on Prospecting Operations in the Musgrave, Mann and Tomkinson Ranges.’ Parl. Pap. No. 54 EARLY CAMBRIAN (?) JELLYFISHES FROM THE FLINDERS RANGES, SOUTH AUSTRALIA By REG. C. SPRIGG Summary A richly fossiliferous horizon has been discovered within the massive Pound Quartzite formation which underlies the Cambrian Archaeocyathinae limestones in the Flinders Ranges, South Australia. The fossils occur as impressions on surfaces of flaggy quartzite. The five genera and species described are almost certainly all pelagic Coelenterates, and while several forms are referred to the class Scyphozoa, it is possible that one or more species may be more correctly assigned as Hydromedusae. The more problematical forms may prove to be pneumatophores or swimming bells. >t: —+= EARLY CAMBRIAN (?) JELLYFISHES FROM THE FLINDERS RANGES, SOUTH AUSTRALIA By Rec. C. Sprice * [Read 8 May 1947] Pirates V ro VITI ABSTRACT A tichly fossiliferous horizon has been discovered within the massive Pound Quartzite formation which underlies the Cambrian Archaeocyathinae limestones in the Finders Ranges, South Australia. The fossils eceur as impressions on surfaces of flaggy quartzite. The five genera and species described are almost certainly all pelagie Coclenterates. and while several forms are referred to the class Scyphozoa, it is possible that one or more species may be more correctly assigned as Hydromedusae. The more problematical forms may prove Lo he pneumatophores or swimming bells. INTRODUCTION In this paper a group of fossils recently discovered in the uppermost forma- tion of the Adelaide Series (Upper Proterozoic to Lower Cambrian) is discussed. The fossils were found on a rise approximately three hundred yards south-west of the principal south- mine workings at TEdiacara, near Beltana, South Australia, They ae a pserione ard uccurred as impressions in flaggy ( EDIAGARA / ait Paste Caakresr ° - Herrsaie, quariziic and are among the oldest , cirest records of animal life in the act world. SOUTH AWSTRALIA if the environmental associations of the forms have heen correctly interpreted there is good reason [0 consider all the forms pelagic and [ree swimming, They all appear to lack nurd parts and to represent animals of very varied affimitics. All are pro- sably Coelenterates and all may be ADELAIDE jeliyishes, although in at least two cases insufficient dctail is available to niake rcliable comparisons with any living or fossil animals (zis., Papilio- natu, Dickinsonia and ithe tnmamed circular form described in association with Beltanella). Just possibly the latter two forms may be floats or pneumato- phores of colonial coelenterates. One form (Ediacaria) is referred tentatively to either of the orders Semaeostomeae (Discomedusae) or Rhizostomeae of the class Seyphozoa, while another (Beltanella) may be referable to either of the classes Hydrotmedusae or Seyphozoa. * Assistant Government Geologist, South Australia. Trans, Roy. Sos. S. Aust., 71, (2), 1 December 1947 213 STRATIGRAPHICAL CONSIDERATIONS Fossil jellyfish as casts and impressiotis have heen described from the Lower Cambrian of New York, Sweden, Russia and Bohemia, and now fram Australia; from the mid-Cambrian of British Colunihia and Alabama; from the Silurian of Victoria and from the Permian uf saxany; from the jurassic of Solnhifen, Bavaria and from the Cretaceous of Texas. he identity of some of these fossils has heen questioned, and some are better referred to the Prohlematica. The stratigraphic range of jellyfishes is alinost certainly pre-Cambrian to present, As indicated above the fossil jellytishes described in this paper are very probably Lower Cambrian in age. The fossil impressions were found in the “Pound’ quartzite formation of the Upper Adelaide Series, ‘This massive forma- tion, which frequently measures several thousands of feet in thickness, imme- thately precedes massive Cambrian Plenspongia (= Archaeneyathinac) lime- stones, ‘The impressions occurred at a horizon approximately 100 feet strati- graphically below the base of the limestone, and approximately 600 feet below the first Pleospongia remains yet located at this localiry. Generally speaking the horizon of the Pleospongia is considered to be near the top of the Lower Cambrian (David, 1927). On this assumption the present fossils would have been living in about middle Lower Cambrian rine. The “Pound!” quartzite im which the fossils were found js considered tentatively (Mawson, 1939) to represent the base of Cambrian sedimentation in South Ans- tralia, although there is reason to believe that the dawn of the Cambrian nay eventually be taken back still further (Sprige, 1942), With perhaps the exception of Davill’s (1936) Tapwiai problematica, and certain anmelid tracks common in several quartzites of the Adelaide Series, the new fossils represent the oldest undoubted marine animals recorded in Australia (fg. 2). They provide the first reliable indication that the Pound Quartzite is uf marine origin. MODE OF OCCURRENCE AND PRESERVATION Considering the extremely perishable nature of jellyfishes and related soit- hodied coelenterates (many jellyfish contain 99% sea water), it is remarkable that any of them should have left traevs of their existence in the fossil state. It is obvious that very special conditions of hurial would be necessary for their pre- servation. Walcott (1898) suggests that in the ease of many United States Cambrian forms the medusae probably had the habit of living on a muddy bottom in great mumbers, Associated fossils suggest a shallow water envitonment. There appeared to be rapid burial and consolidation of the sediment, not by exposure hetween tides, but entirely beneath the water, Tn other cases, stich as at Solh- hofen, Bavaria (Lower Jurassic), medusae have been found in extremely fine- grained slales. Agassiz (1862) notes that the living Aurelia flavida, aller the spawning period, ts Freyuently seen iv lange numbers Aoating on the water. There has been a thickening of the tissues by am increased deposition of animal substance, The ©) Tt is to be noted that, contrary to popular conception, the South Australian Pleospongia ate not reef builders. Their spatial distribution within the enclosing sediment simulates “coral meadows” conditions, This explains why. in this case (and im many other cases) Archaevcyathinae do not occur right from the base of the so-called Archacocyathinae limestone, The Archaeos are usually restricted {a mare or less definite hotizens within ithe Limestone formation, and even in the most highly fossiliferous of these, actux! focsil remains form only a relatively small portion ef the rock. They are not massed remains as in modern coral reefs. 214 of the umbrella and oral region drop off during this period. It can be seen that in this dried out condition the medusae, when stranded on beaches and covered by sediment, offer much better chance for fossilization, Caster (1945) notes that when Aurelia and other medusae are washed ashore and stranded during low tide in midsunmer, they quick'y dry oul on the surface. “Dehydration of the aqueous jelly brings out in surface relief embedded struc- tures, which in life would hardly be discernible, except by transparency, on the exterior, While the upper surface x is hardened. thus inhibiting de- N hydration, the surface in contact NY with the heach remains soft and NS 2 : ine, . N often turgtd as in life. Depending Vit laze « Crindstone ange on the texture of the sirand, the OQ p<". Feel #2 sanals fore ge fzive, imprint made br the surface may be faithful or mdistinci. The in- coming tide picks up the partially embedded jellies and carries them say tie hy old Se further ashore, and occasionally Perr cherty PO aa ae turns then over to embed them feet nnindr Hlaggy guarfzife, aga At (urn Ot tide with the erst- while downside up.” He notes iurther that extraordinarily rapid i iki a cna exclusively with buried pelagic forms. The fossils are impressions in flaggy sandstone quartzite, and the grooves of the impressions are ” epee | ts F E ry solidification of the entombing \ inatrix is not necessary, as firm 8 jellyfish can be dug from tidal sands y SSPE Chocolate shales one of today wherein they haye appar- Ng Feet sandstones i etitly remained buried for a much = OF fact, COGLELES GH Graneted longer time than tidal periodicity. N veas Checolete shales ad Sx the ‘Casé of the’ newly. dig R Seep Sernatstonés with covered fossils we are dealing & 1%) FOO foep Grey shales. 445 feck Butt- white sand share. |——- Agoréx. horizon of stained with ferruginous material HEAR, or possess a film of clayey material EL F040 Massive and Ft Th Toe S ys se Lg OO4 ee Porentamed mao ¢ enclosing sediment was origin- Jj ARCAAEOCVATIAIMALE ally a fine-graiied and well-sorted HA sund witich had aeeumulated near L] ‘ i t g PS > ee tS : the western margin of the vast S'S Nppo « A LYE SHO RS. Flinders sedimentary geosyneline a Ny Fee (Out eee porte Fe, The environment of entombment uy was that of intertidal flats or ot the S 8 strandline, My g Toughening of the surtace tissues S) ‘ of the fossil animals as described by Ne Agassiz and Caster seems almost pad certainly to have occurred prior to Fig. 2 h 2 . . rs ai , . burial, and judging from the attt- eeteliaey Senne peas tobias ee tude of the fossils in the field associate tMHations a PIIndercs anges he yt we pak . [modified after Sic D. Mawson 1939 (2) |, ste animals were mostly preserved showing approximate horizons of known with their ex-umbrella surfaces fossils. uppermost. als It is noted that the fossils were all found on “free” faces of faggy quartzite, and always on the upper surfaces of these slabs. (Beds dip at 10° to 20°). The fissility of the quartzite appears to have been controlled to some extent by the presence of clayey films, which might also have had significance in the preserva- tion of the fossil impressions. DESCRIPTION AND TENTATIVE CLASSIFICATION ® OF THE FOSSILS Class SCYPHOZOA Order (2) SAEMOSTOMEAE or (?) RHIZOSTOMEAE Genus. Ediacaria Sprigg, gen, nov. Genotype Ediacaria flindersi Sprigg, gen. et sp. nov, Pound Quartzite, Upper Adelaide Series. (Lower Camhrian), Ediacara, South Australia. Being monotypic the new genus shows the species characters described below. Generic characters include the hell-like manubrial structure and other structures of the central disc area. Comparisons between Edtacaria and other most closely related genera are given below, Ediacaria flindersi Sprigg, gen. et sp. nov. (Pl. I, fig. 1) Ilolotvype: No. T.1, Tate Museum Collection, Adelaide University, South Aus- tralia, Deseription—Medusa impression circtilar, radially syimmetrical; surface flattened, bet with radial and concentric features of low relief, Three concentric zones are clearly distinguishable, Inner Zone—(?) Manubrium bell-like, constricted near its jumetion centrally with the suh-imbrella surface and expanded distally. It lies aver sideways and is compressed laterally, Length 15 mm., and maximum width (flattened) 14 mm. At least three pendant lobate pouches extend 9 tu 11 mm, centrifygally from the base of the manubrium. Beyond these pouches the central zone is essentially smooth, although there is an incomplete concentric grooye half-way to the zone Wargin, Median. Zone—Suriacc smooth, somewhat inflated: zone delimited on inner and outer aspects by concentric grooves—one (or Lwo) on inher margin, and one deeper with associated minor and less regular grooves on the outer. Two well- marked tradial grooves are present, while indistitict radial striations are more numerous. External Zone—Surface flattened or only slightly comvex in transverse section with minor concentric undulations or flutings and numerous radial grooves or striae. In the annular segment representing three-fourths of the perimeter, at least 44 separate radial grooves can be recognised. Although somewhat irregular in themselves, they are distributed around the zone relatively evenly. Most diverge centrifugally, but some converge in this direction. The outer margin (perimeter of fossil) is fairly regular (ciretilac), and with one or two doubiful exceptions is devoid of marginal notches. A concentric groove les approximately 4 mm. in from the perimeter of the form. () Classification used in this work is ‘bused on that of Parker and Haswell 1940. 216 Dimensions—largest diameter 114 mm. Respective widths of inner, median and outer zoncs along greatest raditis 20 mm., 17 mm, and 25 mm, Discussions and Comparisons—The specimien is considered lo be the impres- sion of the sub-umbrella surface of a “dried out” jellyfish. Organs adjacent to the oral surface uf the original animal have coime to stand ott in relier, and the manubrium. stands out strongly, The central zone probably corresponds with the gastrovascular cavity, and external structures of the central disc region are supéet- imposed wpoEn It. The sub-triangular (?) manubrial structure has been so iuterpreted because of its apparent fusion centrally with the sub-wmnbrella surface, and because no other comparable structures are distributed radially about the centre. The flattened attitude of this manubhrial bell bears a superficial resemblance to the insert lobes of the central discs of Kivklondia (Caster) and Rhizostonutes (aekel). How- ever, the absence of more of these structures radially disposed about the centre largely contradicts this view, In life the manubrial structure would be suspended vertically from the central region. The shape of the month opening cannot he judged. Borgimatl Lespet er Orreeatge Carat SR Kestiel Canals ee r Srameach. ‘3 Cad Ore = vesculer cme sic) oe Wiarw brie eretent pouches (genital sees) B SYarracth ae Kediot Comat Weve ‘iietle apna? saeusth > Merokica Ce ee Ir gy Tee, obit ‘ we has Seerian, Fig, 3 Edijacaria fliudersi A, details of the subuimbreila surface of the reconstructed form; B, crass section through radial canals; C, actual profile section along a diameter of the fossil. The three pendant pouches extending radially from the base of the manubriwn are possibly gastro-gemital sacs in conection with radial canals, Judging from the distribittion of the three sacs preserved there were probably eight of these in the original ammal. Various concentric futings, with the exception of that adjacent the margin of the form, are referable to the circular muscles of the stb-umbrella. The epi- marginal groove is probably a circular canal, in which case the narrow flange beyond it would probably be a “hood.” The well-marked radial grooves of the median zone correspond with inter-, ad-, or per-radial canals, whereas the much finer, numerous radial striations of the outer zone and to a much lesser extent of the median zone suggest more minor canals, splitting and radiating towards the cireular canal. This may merely be shrinkage creasing, but in any cases such ercases would tend to follow such rela- tive weaknesses as the canal lines. The grooves are sub-parallel and tend to inerease in number centrifugally, a feature which is in keeping with the canal theary. 217 Two marginal notches can be interpreted; these are at intervals corresponding with the separation indicated by projection to the margin of the stronger radial canals of the median zone, It is noticeable that in each case deeper radial striations continuc to cach notch. The annular (?) hoad structure, where observed beyond one marginal notch, is indented in sympathy. This would support the view that the notches are regular marginal features, possihly originally enclosing sensory structures, Om the other hand it is noted that in other portions of the fossil where cantinuons sections o7 the margin are preserved, other notches are not apparent. This would suggest that the two notches noted may be accidental invaginations of the margin consequent upon deformation during burial. There are no indications of marginal tentacles, but in view of the coarse nature of the enclosing sediment and the probable delicate nature of such tentacles tf present, it is difficult to imagine that evidence of thent could have been pre- served. Furthermore, experience with modern jellyfish (Aurelia, etc.) indicates that in many cases the marginal tentacles drop off in the senescent stages. There appears little doubt that Ediacaria is a Seyphozoan. The form obviously had a datiened saucer or dise-Jike umbrella, and for this reason is referred to either of the orders Semaeostomeae (Discomedusae) or Rhizo- stameac, To decide further to which of these orders the form belongs, a detailed knowledge of the structure of the mouth and oral arms would be necessary. In view of the nature of the preservation af the specimen thig carmot be hoped for, However, it is noticeable that the manubrial structure ag interpreted is relatively simple—a fact which suggests correlation with the Semaeostomeac, or could per- ceivably mdicate an even simpler class relationship (2iz., Hydrozoan), The absence of marginal tentacles on the other hand ts a Rhizostomid character, but in view of the difficulties already pointed out evidence such as. this is primarily negative. It does seem, however, that no strong tentacles existed—a fact which would preclude membership of the order Trachymedusae of the class Trachylinac. Detailed comparisons with other fossils is exceedingly difficult in view of the lack of kuowledge of many critical features, but it is noticeable that closest (super- ficial) resemblance is perhaps with Rhizestomiles and Semacostomites (both Haekel) of the Upper Jurassic of Solnhofen, Bavaria. In these forms three concentric zones can be inferred, but otherwise there is little similarity in ayail- able detail of the central dise regions, King muscles are well developed in the outer portions of A/isostomites. as they are in Ediacaria. No obvious ring canal is pteserit in Rhizostomites as it is in Ndiagaria and Seviaeostomites, and whereas the margin of Scmacostomites is split up into 120-128 marginal jobes, such sub- division is not apparent in the other two forms. Ediacaria (7) sp, Sprigg (Pl, 1, fig. 2) Specimen No, T.2., Tate Mus. Coll, Adel, Univ., 5. Aust, Deseription—Impression fragmentary; original fetm apparently circular, but less than one-eighth of the perimeter is preserved, although the outline af the central disc region is almost complete. In all, three concentric zones can he made ont. Central Zone—Surface smooth except where badly preserved; slightly inflated; perimeter sub-circular. Median and Outer Zones—Zones poorly differentiated; median one js smooth and of variable width. Outer zone inflated somewhat towards outer margin, and traversed by numerous sub-paralle! radia! grooves which fade towards either margin. Exterior margin slightly lobate with indications of two minor notches. F 218 Discussion and Comparison—In view of the absence of structures in surface relief in the central and median zones of this fossil, it is suggested that here is a cast of the ab-oral or ex-urnbrella surface of a jellyfish. The poor state of pre- servation of the centremost portion of the fossil form prevents complete assess- ment of this fact, but nevertheless there appear to be no signs of mouth or pastro+ genital structures, The centrally inflated region appears to have collapsed over the stomachs of the animal and preserved, therefore, the roughly circular outline of this region. Tn the outer zone which, presumably, in the living animal was thin, the sub- parallel radial grooves are well preserved, These are very numerous and appear unbranched; they may represent radial canals or merely be shrinkage creases. Prominent tadial striations continue directly to the two minor marginal notches, and there are no signs of marginal tentacles, ‘Tentatively the fossil is considered to be an impression of the ab-oral surface of a species of Ediacatia. As with the holotype specimen, three concentric zones can be interpreted, the outermost of which is traversed by very numerous racial striations, Indefinite notches which are separated by approximately equal intervals appear marginally in both forms. Width relations of the respective zones of the fossils agree favourabiy. Class (?) SCYPHOZOA Genus Beltanella Sprigg, gen. nov. Genotype Beltanella gilesi Sprigg. gen. et sp. mov. Pound quartzite, Upper Adelaide Series (Lower Cambrian) Ediacara, South Australia. ‘ Being monotypic this new genus shares the species traits described below. Until more is known of the fossil, generic characters should include the octagonal arrangement of the circular (?) gonadial structures and their association with the radial canals, the presence of a well-developed delicate peripheral umbral or velar structure and the simple circular oral aperture. Beltanella gilesi Sprigg, gen. et. sp. nov. (Pl. ii, fig. 1) Vfolotype: No, T.3., Tate Mus. Coll., Adel. Univ., S. Aust. Description—Medusa impression circular. Umbrella rather flat, but falling away sharply near its onter margin. (?) Velariuim horizontal, depressed approxi- mately 4 mm. in relation to the flat ex-umbrella surface. Umbrella region sub- divided into two zones by a faint annular groove as follows, Inner Zone—Surface smooth, broken only by annular grooves respectively 5 and 12 mm. in diameter at the centre. Centremost area depressed very slightly. Ouler Zone—Surface dominantly flat, but slopes away steeply near outer margin of umbrella. This secondary (sloping) surface has. the form of a highly truncate cone whose apical angle is approximately 80 degrees. Zone characterised by the presence of circilar (?) gonadial strictures, approximately 10 mm. in diameter, These regular structures are arranged on either sides of the major radial canals in an octagonal pattern centrally within the zone. At Jeast four of these can be recognised and each possesses an inner concentric groove 3-4 mm. in diameter. Two paired radial grooves (? canals) are diametri- cally opposed, and a third set lies radially at right angles. The grooves pass inter- mediate hetween (7?) paired gonadial structures but do not continue into the inner 219 zone. The ex-umbrella surface is slightly lobate at the edge of the flat raised portion, but below where the conical surface meets the (?) velarium the margin is smooth. Fig. 4 Beltanella gilesi A, details of extmbrella surface and related embedded organs; B, transverse section through restored form; C, true profile section along diameter of form. Velarium—Structure marginal, obviously thin, well developed; undulose surface depressed ; undulations annular in plan, Dimensions—Maximum diameter of fossil 110 mni., minimum 97: widths along single radii of inner and outer zones and yelarium respectively 18-20, 21-23 and 10-14 mm. Discussion and Comparisons—The specimen is the cast of the ex-umbrella surface (ab-vral) of a jellyfish. The central zone evidently corresponds with the gastrovascular cavily as at its margin it gives off paired grooves which are interpreted as interradial canals. There are no signs of subdivision within the cavity and no indication of com- plicated manubrial structures. The simple circular grooves situated centrally may be oral structures, or possibly representative of a collapsed truncate gastric cone which occurs in some jellyfish to aid in the even distribution of food to various portions of the animal's stornach, The radial grooves of the outer zone are thought to be interracial canals, although why they should be paired is not known. There is no sigi) of branch canals from them, uor is there present any groove suggestive of a circular canal. The circular (?) gonadial structures which are distributed- evenly around the centre of this zone may be considered as paired in relation to the supposed radial canals. The central annular grooves of each gonadial structure may mark a genital operculum, The (?) velarium or peripheral umbral jelly is remarkably well preserved considering its obvious delicateness; its contained annular undulations may indicate ring muscles, The decision to place Beltanel{a within the Seyphozoa must be regarded only as very tentative, as the animal has many restrictive features characteristic of either certain orders of the Scyphozoa or of the Hydrozoa, For example, the simple mouth, the presence of a few unbranched radial canals and the association 220 of the gonads with the radial canals are Trachylinid (Hydrozoan) characters. On the other hand the flattened disc-shaped umbrella, its relatively large size, and the absence of large tentacles are more characteristic of the Scyphozoa. It appears very likely, therefore, that Bellanella is a. member of a group, ancestral to either or both the Seyphozoa and Hydrozoa of modern times. Indeed, this is to he expected considering the great age of the fossil. The author knows of no living or fossil jellyfish with which useful comparison of the foregoing specimen can be made. The fossil 1s tentatively classified with the Scyphozoa, mainly in view of the absence of the large tentacles which are typical of the Trachylinids. On the slab of rock carrying the holotype there are at least four additional circular structures (10-23 mm. in diarneter) which may represent juvenile forms. All display annular grooves, and in the largest specimen, the central portion (9 mm. in diameter) is raised relatively to the otiter zone. In the latter example (which is photographed with the holotype) there is very definite evidence of a velar structure, 2 mm. wide, similar to that of Beltanella, Still another possibly related form found at the same locality is the dise-like impression shown on p), ii, fig. 2 (specimen No. T4). This external cast is almost Fig. 5 True profile section of nha form figured on pl. uy, fig. 2. perfectly circular with a slightly raised rim near the external margin. The whole structure stands in relief a little above the face of the quartzite slab, and its surface is evenly convex, except in the marginal region, where it is raised some- what; it shows no recognisable anatomical structure or ornamentation. This smooth depressed (annular) arca (7-10 mm. wide) is faintly reminiscent of a velar structure, Maximum diameter is 104 mim., nat including the latter structure. The impression may be that of a simple discoid jellyfish or of a hydroid float. Class (7) HYDROZOA or (?) SCYPHOZOA Genus Cyclomedusa Sprigg, gen. nov. Genotype Cyclomedusa davidi Sprige, gen. et sp. nov. Pound Quartzite, Upper Adelaide Series (Lower Cambrian) Ediacara, South Australia, This genus also is monotypic, and in view of the lack of detail of most of the critical features of the form the generic characters are based tentatively om the sculpturing of the (?) ex-umbrella surface of the form. Cyclomedusa davidi Sprigg gen. et sp. nov. (Pl, iii, fig. 1) Holotype: No. T.5., Tate Mus. Coll,, Adel. Univ., S. Asst, Description—Impression circular, depressed, with concetitric undulations. Centra] portion raised, distinctly nodular. The whole form exhibits striking radial symmetry, Surface subdivided by at least seven annular grooves. For convenience three major zones are recognised. Inner Zone—Consists simply of the hemispherical nodular portion mentioned previously. Approximately 2-5 mm, in radial width and 1-5 mm. in height. Median Zone—Subdivided into two annular portions of low relief, the inner of which is traversed in turn by two annular grooves, while the outer is z21 ornamented by very numerous radial striations about 3 mm. in length. There are approximately 16 of these grooves in each quarter. A poorly developed annular groove traverses the outer portion. Outer Zone—Surface sculpture very similar to that of the median zone; the smooth inner portion is bounded by annular grooves and the outer is traversed by numerous poorly developed radia! striations averaging 4 mm. in length, The outer margin to the zone is very poorly preserved. Dimensions—Maximum diameter probably 50 mm. Average widths of the inner, median and outer zones along single radius 2-5, 11-0 and 11-0 mm, respectively, Discussions and Comparisons—The surface ornamentations, which stand in strong relief, appear to be of superficial significance, but some of the annular flutings may be related to ring muscles of the subumbrella of a jellyfish. No structures which can be attributed to a body cavity or to gonads are visible. A arpoe Fig: 6 de —— fenctio! Cyelamedusa davidi fi s ‘nf liars, . $ i. 4D EZ A, details of the surface ornamentation of the Ye restored form; Aisate 4 . B . B, ptofile section along a diameter af the i . Fretife Section ‘ i er sis OT ead say are ss tei, actual iussil. There is no evidence of extra-matginal structures, but as noted, the marginal region of the fossil is poorly preserved. , In view of the paucity of critical detail, it is only with hesitation that the fossil is provisionally included within the Coelenterata. The form is highly problematical and possibly only represents the restricted central portion of a larger animal. It does seem certain from its regularity and complexity, however, that the form is not a pseudafossil. Genus Dickinsonia Sprigg, gen. nov. jenotype Dickinsonia costata Sprigg gen. et sp. nov. Pound Quartzite, Upper Adelaide Series (Lower Cambrian) Ediacara, South Australia This genus which is monotypic, exhibits the species characters described below, Animal of ovoid form, (?) inflated aborally and possesses a marginal crenulate Mange. Median longitudinal furrow gives off very mumerous subradial grooves to the outer (crenulate) margin of the form, Dickinsonia costata gen. et sp. nov. (PL iii, fg. 2) Holotype: No. T.6., Tate Mus. Coll., Adel, Univ., South Australia. Description—Impression ovoid, bilaterally symmetrical, essentially flat. Median longitudinal furrow approximately 35 mm. long, gives off 80 to 90 radiat- ing or diverging grooves or costae (7?) alternatively to the outer margin of the fossil. Margin slightly crenulate, the notches corresponding with the intersection of the radiating grooves. Well developed concentric epi-marginal sulcus marks off a marginal Hange. Dimensions—Length 68 mm.; width 60 mm, Flange width variable, from 2 to 7 mm., due to distortion. 222 Discussions and Comparisons—The fossil is the impression of the (?) dorsal aspect of a bilaterally symmetrical animal of very doubtful affinities. During burial the animal was flattened and compressed slightly obliquely in a manner which suggests that it was strongly convex dorsally, The afiimal was symmetrical across both longitudinal and transverse planes. Radiating grooves or costae may represent chitinous rods or canals, while the epimarginal groove may represent a circular canal. No gastrogenital structures or appendages are apparent, It is exceedingly difficult to classify Dickinsonia on the little detail ayailable. The author knows of no related animal with which to establish relationships, and until new specimens with more detail are found little more can be suggested than that the animal is probably a coelenterate. Fig. 7 Dickivisonia costata Details of surface ornamentation of the restored form as viewer in elevation (A) and in plan (B). There are certain superficial features in which it resembles the problematical form Discophyllum peltatum (Hall), but these apparent resemblances soon dis- appear when detailed comparisons are made. Discaphyllwin is a rounded or oval flattened form with bars which radiate froni the centre of the form, and not from a longitudinal furrow as in Dickinsonia. Also there are fine concentric undulating lines that cross both the rays and the interspaces between them, Comparable structures do fottoccur in Dickinsonia. There is no flatige structure in Disco- phyllum, nor is the external margin crenulate, The ribs fade out before the margin is reached. (?) Dickinsonia sp. (PI. iv, fig. 1) Specimen No. T.7., Tate Mus. Coll., Adel. Univ., South Australia, Description—Impression fragmentary, radially costate; costae continue to the margin, Margin rounded, slightly crenulate in sympathy with the disposition of the costae. A slight bending of the costae, noticeable 7-8 mm. in from the margin of the fossil, indicates a tendency to flange formation, Discussions and Comparisons—From the little detail available for comparison of this specimen with the foregoing holotype the major similarities are concerned with the radiating costae, which in both forms are strongly developed and continue to the somewhat crenulate margin. In the holotype specimen the costae are interrupted by a very well marked concentric sulctis, whereas in the latter specimen there is only the faintest hint of interruption of the radial costae in a similar position, None of the costae in the second specimen appear to diverge as if to unite alternatively into.a central furrow as in the type form, The correlation of this fossil with Dickinsonia is quite tentatve and the form must be considered problematical. 228 Genus Papilionata Sprigg, gen. nov. Genotype Papilionata eyrei gen. et sp. nov. Pound Quartzite, Upper Adelaide Series (Lower Cambrian) Ediacara, South Australia, The gerius is monotypic and shares the species characters described below. Unfortunately, the yereric characters must be based on the shape of the fossil and its resiticted surface sculpture, The form is bilaterally symmetrical, papilionaceous with the amygdaloidal “wings” fused anteriorly. A marginal gtoove is present and the posterior margin of the wings ig met by sub-parallel or slightly radiating grooves. Papilionata eyrei Sprigg, gen. et sp. mov. (Pl. iv, fig. 2) Holotype: No. T.8.. Tate Mus. Coll., Adel. Univ., South Australia, Description—Impression bilaterally symmetrical, Left hand portion im- perfectly preserved and fragmenta]. Complete impression papilionaceaus, the “wings” being fused for approximately 30 mm, in the anterior aspect, The left hand wing is amygdalvidal in plan, curved convexly on the inner margin, and rounded and slightly lobate (posteriorly) on the lateral margin, From the posterior portion of this margin, radial grooves converge towards a central point near the inner margin, but fade out after 10 to 20 mm. The grooves are stronger and longer at the posterior extremity, and they are fot visible forward of the cetitre of this margin. The single notch on the lateral margin may mot have anatomical significance, being probably a crenulation due to the animal's assumed position of rest. The lateral margin is paralleled by a well-developed groove through which the radial costae continue uninterrupted. The groove diverges slightly from the margin at the anterior end. ). Fig. 20 Discotd Seyphozeun or Zooidal Hoat (x Trans, Roy. Soe, S. Aust., 147 Vol. 71, Vlate VII Thate hy KU. Mallips Mhioty diy te TS Phiten Tice. 2 Jicvkinsania costal Sprige Shelitly ducer than falaral size, Trans, Roy. Soe, S. Aust., 1947 Vol. 71, Plate VIIL Phote by K.P. Phillips Big. 1 Dtchinsonta (2) sh Capprax. x14) Photo hy K.P. Phillips Kiv, 2 Pupilionata eyrei Sprige (approx, x @). SOME NEW EURYMELIDS FROM AUSTRALIA AND NEW GUINEA (HOMOPTERA, JASSOIDEA) By J. W. EVANS Summary A collection of Homoptera recently sent me by Mr. D. C. Swan included representatives of eight species of Jassoids belonging to the family Eurymelidae, which he had collected in New Guinea during the war. Five of these species are described below. Of the remainder, one, Eurymeloides nigra Evans, is confined to New Guinea, and two, Eurymeloides punctata Sign, and E. adspersa Sign., both of which are well-known Australian species, are recorded from New Guinea for the first time. The host plants were species of Eucalyptus and Melaleuca. 225 SOME NEW EURYMELIDS FROM AUSTRALIA AND NEW GUINEA (HOMOPTERA, JASSOIDEA) By J. W. Evans* Communicated by D. C. Swan [Read 7 August 1947] A collection of Homoptera recently sent me by Mr. D. C. Swan included representatives of eight species of Jassoids belonging to the family Eurymelidae, which he had collected in New Guinea during the wat. Five of these species are described below. Of the remainder, one, Eurymeloides nigra Evans, is confined to New Guinea, and two, Eurymeloides punctaia Sign. and E, adspersa. Sign., both of which are well-known Austrilian species, are recorded from New Guinea for the first time. The host plants were species of Excalyptus and Melaleuca, Of the remaining four new species of Eurymelids described below, the most distinctive is [poella devisi, which was collected in North-west Australia by the late Dr. Consett Davis, whose tragic death in 1944 was an irreparable loss to entomology in Australia, Eurymeloides sogerensis sp. nov, Length, 5mm. Head, face evenly dark brown, but for the outer margins of the lora and maxillary plates, which are cream. Pronatwm brown mottled with pale greyish-brown, Scutellum dark brown. Tegimen dark brown, irregularly mottled with dark grey. Thorax, ventral surface and legs, dark brown. Male Genitalia, aedeagus as in fig. 1, A. Holotype, 8, from Sogeri, New Guinea (D. C, Swan, 5/45), in the British Museum. Paratypes in the South Australian Museum, Eurymeloides motuana sp. noy, Length, 7 mm. Head dark nigger or reddish-brown mottled with yellow ; ante-clypeus dark brown; lora and adjacent parts of the maxillary plates yellowish or pale brown, Pronotwi nigger or reddish-brown mottled with ivory. Scutellum, the angles black, centrally dark brown with oval yellowish markings. Tegmen black with two irregular white, transverse fasciae and with a white marking at the distal apex of each anal vein. Legs brown, the bases of the spines and the proximal tarsal segment of the hind tibia, white, Male Genitalia, aedeagus as in fig. 1, 0. Holotype, &, irom Port Moresby, New Guinea (D. C, Swan, 5/45), in the British Museum. Paratypes in the South Australian Museum, Eurymeloides nigrebrunnea sp. nov. Length, 7-2 mm. Head, maxillary plates pale brownish-yellow ; ante-clypetus lora and fronto-clypeus pale reddish-brown; vertex dark hrown mottled with apricot yellow. Pronotum and Scutcllwm chestnut brown or brown mottled with yellow. Tegmen black with two white fasciae, each with a large irregular biack marking, «anal veins distally white. Legs, femora pale brown, tibiae dark brown, the bases of the spines and the proximal tarsal segment, white. Male Genitalia, aedeaptts as in fig. 1, C. Hololype, 8, from Port George, North-west Australia (C, Davis, 5/43), in the British Museum. Paratypes in the South Australian Museum, * Imperial Institute of Entomology- Trans. Roy. Suc. 5. Aust,, 71, (2), 1 December 1947 226 Ipoides brunomaculata sp. nov. Length, 5 mm. Head, face, but for the vertex, yellowish-ivory with a median longitudinal dark brown stripe. Vertex pale brown irregularly mottled with chestnut brown, Crown and pronotum pale chestnut brown mottled with yellowish-ivory. Scutellum deep chestnut brown with two pale semi-circular markings. Tegmen pale hyaline brown with irregular white markings; veins white, brown in part. Legs very pale brown. Male Genvtala, aedeagus as in fig. 1, F. Holotype, &@, from Port Moresby, New Guinea (D. C. Swan, 5/45), in the British Museum, Paratypes in the South Australian Museunz. F G H Ipoides melaleucae sp. nov. Length, 5 mm. Head, face, niaxillary plates greyish, the rest of the face pale greyish-brown evenly mottled with dark brown. Crown, pronotum and scutellum, greyish, sparsely mottled with brown. Tcgmen very pale hyaline brown; yeins brown with white bars. Legs, femora dark brown, tibiae pale brown. Male Genitalia, aedeagus as in fig. 1, G. Holotype, &, from Sogeri, New Guinea (D. C. Swan, 5/45), in the British Museum, Paratypes in the South Australian Museum. Anipo pallescens sp. nov, Length, 5°5 mm. Head, face, maxillary plates, lora and fronto-clypeus laterally, ivory partially suffused with apricot; ante-clypeus, the fronto-clypeus 227 medially and the vertex, deep coffee brown with pale oval markings, Crown and Pronotum pale greyish-brown mottled laterally with dark brown. Sextellum, anterior lateral angles dark brown, the remainder pale brown, Tegmen pale hyaline brown with evenly distributed circular colourless hyaline areas. Thorax, ventral surface and legs marked with a pattern of pale and dark brown. Male Geniiaha, aedeagus as in fig, 1, F. Holotype, 8, from Sogeri, New Guinea (D. C. Swan, 5/45), in the British Museum, Paratypes in the South Australian Muscum., Ipoella davisi sp. nov. Length, 5 mm, Head, face almost fat; ante-clypeus natrowly depressed anteriorly, dark brown; fronto-clypeus medially dark brown, shading to pale brown, laterally ivory; lora brown adjacent to the arite-clypens, laterally ivory ; maxillary plates ivory anterior to the antennae, dark brown beneath the eyes; vertex chestntir and very dark brown, posteriorly ivory. Crown brown with irregular dark brow? markings. Pronotum buff mottled with brown. Scutéellum dark brown. Tegmen hyaline, the costal margin and the apex smoky brown, the remainder dark brown with a broad proxima! white fascia and white markings at the apex of the first anal vein. Some specimens have also a distal white fascia. Male Genitalia, aedeagus as in fig. 1, D. Holotype, 8, from Isdell River, Walcott Inlet, North-west Australia (C, Davis, 9/43 on Ficus), in the British Museum. Paratypes in the South Aus- tralian Museum. Bakeriola tasmaniensis sp. nov. Length, 5 mm. Head, face black sparsely mottled with brown but for the outer margins of the maxillary plates, which are white and the Jora, which are cream, partially suffused with pale brown. Crown of head and pronotum pale brown mottled with dark brown. Seutellun darle brown. Tegmen, clavus con- colorous with the pronotum, the remainder hyaline brown with circular and irregularly shaped colourless area, Legs dark brown but for the coxae, the tarsi, and both ends of the tibiae and femora, which are pale brown. Male Genitalia, aedeagus as in fig. 1, H, Holotype, 8 from Risdon, Tasmania (J. W, E., 4/39), in the British Museum. Paratypes in the South Australian Museum, Bakeriola tubra sp. nov. Length, 4-8 mm. General coloration dark brown. Head, face evenly convex, dark brown mottled with reddish-brown. Crown and pronotum irregularly mottled with dark brown and reddish-brown. Scutellum dark brown, mottled antero-medially with pale reddish-brown . 2'vgmen, hyaline dark and reddish- brown with oval white markings. Legs reddish-brown, hind tibia with three spurs and several spines, Holotype, é, from Moolooka, Queensland (4/44, E. F, Riek), in the British Museum. Paratype in the South Australian Museum, Trans. Roy. Soc, S. Aust., 71, (2), Dec. 1947 T. LANGFORD SMITH - GEOLOGICAL MAP PORTION OF HUNDREDS OF BELALIE, YANGYA, CALTOWIE, TARCOWIE, MANNANARIE AND WHYTE COUNTIES VICTORIA AND DALHOUSIE SOUTH AUSTRALIA + SCALE = 80 60 40 20 0 80 ig60 CHAINS MILES a ‘69 ¢ | afi tL ty ' I~ aN | [iI \ 1 POREST RESERVE sesserserenlT) Quartzile & Sandstone [====]H Slates & Shales (Tapley Hill Series) = =lc Slates &Phyllites with some Shales & thin beds of Dolomitic Limestone “4° 4°|G Tillite Series oe —--=|* Arkose Grit in Tillite Series JA Phylliles a Slates with some thin beds of Dolomitic Limestone -——F Sub-Glacial Quartzite Vv} Alluvium with no rock outcrops SIS JE Slates Phyllites « Hornfels T. LANGFORD-SMITH 1942 SlB Quartzite & Sandstone THE MARINE ALGAE OF KANGAROO ISLAND By J. W. EVANS Summary Kangaroo Island lies off the South Australian coast at the base of Gulf St. Vincent, being separated from Fleurieu Peninsula by Backstairs Passage (10 miles wide) and from Yorke Peninsula by Investigator Strait (about 26 miles wide). The Island is 90 miles long and up to 32 miles wide, narrowing to only ¥% mile wide between the American River inlet and the south coast (see fig. 1). The long axis of the island is approximately east-west; the island lies between latitude 35° 5’ S. and 35°34 7’ E. and 138°8’ E. 228 THE MARINE ALGAE OF KANGAROO ISLAND I. A GENERAL ACCOUNT OF THE ALGAL ECOLOGY By H. B.S. Womurscey * Puates IX to XILL [Read 7 August 1947] INTRODUCTION Kangaroo Island lies off the South Australian coast at the base of Gulf St. Vincent, being separated from Fleurieu Peninsula by Backstairs Passage (10 miles wide) and from Yorke Peninsula by Investigator Strait (about 26 miles wide); The Island is 90 miles long and up to 32 miles wide, narrowing to only 4 mile wide between the American River inlet and the south coast (see fig. 1). The long axis of the island is apptoximately east-west; the island hes between latitude 35°5’S. and 35°344°S, and between jongitude 136° 32’, and 138° 8 E. ‘The situation of Katgaroo Island in relation to the mainland, and the shape of the island itself, result in great variation in conditions of roughness along the coast, The exposed and rough south and west coasts contrast markedly with the calmer areas of the north coast, while the American River tidal inlet forms a distinct type of habitat not found elsewhere around the island, From the point of view of algal ecology Kangaroo Island offers a particularly satisfactory area for study, cspecially in ilustrating the control exerted on the algal flora by the degree of wave action. This paper is the first report on work carried out during the past four years. The aim has been, firstly, to give a general account of the intertidal algal ecology of a part of the Southern Australian coast, since there have heen no previous ecological studies of this region; secondly, to obtain as comprehensive a list as possible of the species present; and thirdly, ta carry out autecological studies of the more characteristic and dominant species. In this paper it is proposed to describe the more important environmental factors for the island as a whole; to discuss the terminology found most satis- factory; and to give a preliminary general account of the broader aspects of the algal ecology. In subsequent papers the more detailed ecology of characteristic regions will be dealt with, and a cetisus of the known species will be given. Floristic and ecological comparisons with other areas will also be left to later papers. By first presenting a general survey of the algal ecology of the whole island, it is hoped to give perspective to the later detailed descriptions of indi- vidual localities. The localities atound the island which have been studied are shown in fig, 4. Of these, Pennington Bay and American River have received most attention. It has been possible to pay only one or two short visits to the western end of the island, while the very rough nature of the country and lack of roads prevents visits to most of the intervening parts of the south and west coasts. This, and the restriction of field work to the University vacations, have limited examination af most localities other than American River and Pennington Bay to January. Seasonal changes in places other than American River and Pennington Bay are therefore unknown as yet, but the associations which occur in these places and are described in this paper are almost certainly present throughout the year. * Department of Botany, University of Adelaide. Trans. Ruy. Sor. 5. Aust, 71, (23, 1 December 1947 229 The areas studied at Western River and Middle River consist of the coast on cither side of the river mouth, These rivers rn only after heavy rain, al other times being blocked by a sand bar at the mouth, American River, however, is an extensive tidal inlet, consisting of a series of lagoons with wide tidal flats and i central channel, opening to the sea thrangh a mouth some 250 yards wide. The amount of fresh water entering the lagoons from small creeks is teghible, except for possible local effects after heavy rain, The conditions at all “River” localities, therefore, are truly marine. Previous Marine Ecovocican Stupies im AtsTRavia Australian phycology is based to a Jarge extent on the work of the early European algologists, Harvey, Agardh, Sonder and others, who described collec- tions sent to them from Australia. As a result, very little ecological information is available about the carly localities from which collections were made, THarvey (1854), however, gave short notes on the Western Australian coast, und ecological information about many species in his Phycologia Australica, More recently, Hedley (1915) has presented a very general survey of the New South Wales coast, and Lucas (1935) has described the algal ecology of Lord Howe Island. Pope’s (1943) survey of animal and algal life on a reef near Sydney is the only work of a detailed nature, but deals primarily with the animal ecology. The review of Chapmati (1946) gives a comprehensive list of references to algal ecology studies elsewhere in the world. Previous Recorps FROM KANGARGO Istann Previous records of marine algae from Kangaroo Island are very few. Cleland and Black (1941) listed 11 species collected from near the mouth of Sou" West River, and determined by A, H. S, Lucas, Lucas (1929), in his census, lists two species from the island, while Part [ of “The Seaweeds of South Australia” records Caulerpa hedleyi W. vy. Bossc, “dredged in some 8 fathoms off the coast.” Part II of the “Seaweeds of South Australia” contains several records, all incorporated by the present atithor. Two shott Teports on the dtivularias and a new species of Dasyopsis have also been published (Womersley 1946 a and b). The land vegetation of Kangaroo Island has réceived considerable attention. Tate (1883) gave the first general account of the Island, and more recently Wood (1930) has elucidated the relationships of the flora and shown a high degree of endemism, particularly in species confined to the western end, Baldwin and il (1941) have described yegetation commitnities in the central part of the Island. ENVIRONMENTAL CONDITIONS 1. THE Coasran GEoLogy The present study has shown that while the individual type of rock has little effect on ihe flora, the rock topography, in that it may result in different types of habitats, may be of considerable importance in determining the algal associations of an area. The only reasonably detailed geological map of Kangaroo Island is that of Wade (1915), which is followed in fig. 1, The backbone of the island consists of Pre-Cambrian schists and gneisses, overlain in the central part hy siliceous sand and laterite. On the west coast, eastern part of the north coast, and to a lesser extent on the east coast, these rocks form magnificent cliff scenery, The coast west of En Bay is of later age (Post-Cambrian) (Madigan 1928). 230 CSI6L 8PEAA LB) ‘puyysy Ooansuey Jo drum yayeys [eITBojoar) 1 "3 AMY wal waSqM pues | _pogrimeavas- owns sapp Spuojsawt7 saapjoog her rr IP| 9) 31¥3S Nal 3NI3IONY poe jermucieien ce WW NWIHALEN 38d : ——— Bugsauny pPorhjog ntebaye bbues suaspueshN Ssyaug & 3) UPL5 & Fajen ANa?au ua. VRE ANID3Y ¢ Urwquedtacg buh; 7 // ‘ % 9nd “Dhaaye ial /j /] “sz, *s aunvay UPD Dd & AVuIAVaL Ysddn NbivdWooavd Hy) r {| Pseg 481475 © aI ZysAPOY Uff MANTAM| Ob PUPS UAC] z Pe, G pasisa Gg), EHD abesstq $4199542eQ ‘ys J07eB) 1SaAU| 251 The most significant geological features of the south coast of Kangaroo Island are the outcrops of ancient rocks at the capes. These outcrops do not rise to any notable height ahove sea level, and mostly appear at the font of cliffs which fringe the shore. The extent of these outcrops of Pre-Cambrian rocks is shown in fig. 1. Two types of rocks, granile and quartzite or niica schists, form alternating patches along some of the coast. Between the areas of older rocks the coast consists either of sandy beaches backed by sand dunes or of sand- rock cliffs and horizontal reefs formed from older consolidated sand dunes, This rock weathers into very sharp edges and pinnacles, and by its variable hardness makes an irregular coast. The two different types of rock found on the south coast of Kangaroo Island, i¢., the flat rock platforms of consolidated sand-rack such as al Pennington Bay, and the harder more steeply sloping rocks suck as secn at Cape Willoughby, west of Vivonne Bay, and Cape Coudie, result in two very different habitats for algal and anima! growth, and appear to illustrate the conditions occurring along most of the south coast (cf., pl. ix, fig. 3, and pl. xii, fig. 1). Most of the American River inlet consists of extensive sandy or sandy-mud tidal flats, but in many areas, particularly in Pelican Lagoon, these stretch out from low cliffs of the same consolidated sand-rock as at Pennington Bay on the south coast, The northern coast of the island is composed mainly of ancient rocks sloping off into 3 to 12 or more feet of water, giving a relatively small area of rock exposed at low tide- 2. Decree or WAVE Action The degree of roughness of any locality is of prime importance in determining the algal associations present. The south and west coasts ate exposed without any protection to the Southern Qcean, and conditions are invariably rough. Heavy breakers are a constant feature of this coastline (see pl. ix, fig. 1). Passing along the north cnast, from west lo east, conditions become progressively calmer, owing to the shelter afforded by the mainland. Northwards from Cape Willoughby and past Hog Bay wave action is moderate, with breakers only in rough weather, American River and Pelican Lagoon, quite unlike the rest of the island, form an almost land-locked area where wave action is at a minimum. Although wave action is of great importance as an ecological factor, its measurement in any satisfactory way scems impossible, Average values of the forces from wave action to which algae ate subjected in any one locality are needed, These should be measured over short periods (when extrenies may necur), as well as over monthly and yearly periods, In the absence of any such measurements it is necessary, in some cases, to use the algae themselves as an index of the conditions. This has heen done in subdividing the Rocky Shore Formation into subtormations, depending on the presence or absence of Cystophora intermedia J. Ag. This alga is dominant in the sublittoral fringe on rough rocky coasts, but is replaced by other species of Cystophora on calmer coasts. 3. Tipes The tides around the Australian coast, including the main characteristics and ranges of the tides in the South Australian Gulf region, have been described briefly by R, W. Chapman (1938). The tides around Kangaroo Island are of the semi-citrnal type, wilh two maxima, one appreciably lower than the other, and two ininima during each 244-hour period. Vip. 2 shows the form of the spring and neap (“dodge’’) tides at American River, 232 SPRING TIDES ~_ wt 3 o 2 w ioe eal 9 “fg is. < \\ ee \ \ we V\ 6 \iog load ey « ’ NX’ sree" ry. N32 0.7 Pennington Bay = DODGE TIDES a) hg Mat QO z < o American R. Jetty we = b Muston, 5 . 7 i “~, See / 600 TIME (Hours) Fig. 2 Tide curves for spring and dodge tides in the American River Inlet and at Pennington Bay (spring tides only). The range of each tide curve and times of high and low water are comparable, but the heights given for each are arbitrary. The curves are derived from 24-hour surveys carried out at the American River Jetty, Muston Jetty, (3 miles south of American River Jetty), Pig Island in Pelican Lagoon, (2 miles East of Muston) and at Pennington Bay on the following dates: spring tides fan. 10-11, 1947; dodge tides Jan. 16-17, 1947, jOOO 233 South Australian tides vary so greatly from piace to place along the coast, both in their nature and times of high and low water, that it is necessary to obtain actual records from cach locality. The tides have been analysed at com- paratively few places, especially around Kangaroo Island, An automatic tide gauge recently established at Hog Bay should give most interesting results when records became available. The data given below for Kangaroo Island tides are derived from information made ayailable by the Seuth Australian Harbours Roartl and trom 24-hour surveys carried out at American River and Pentugton Bay. It will be evident that until accurate and more extensive tidal data are availabie, general limits and heights only can be given for the main algal zones. Tidal range around Kangaroo Island is small, Along ihe south and west coasts the spring range is about 24 feet, Passing eastwards along the north coast it imereases Ly 44 feet at Kinyscote, just over 4 feet at American River aud Hog Bay, 44 feet at Antechamber Bay, while a rise of G feet is recorded from Cape Willoughby from old data (but this is probably too high), The neap (or “dodge") tides probably have a range of about 14 feet on the south coast and 24 feet on the north. On the south and west coasts rhe small tidal range means that the wind and strength of the swell may exert nearly as great an effect as the tide itself, and little reliance can be placed on the tides alone. The most notable peculiarity of South Australian Vides is the “dodging’* tide, This is discussed by R. W, Chapman (1924). At Port Adelaide, where the effeet is most prominent, the water level may remain almast constant far 24 houl's or more at the neap periods, The cause is that during the neap period the sun and the moon, together with the other tide-producing forces, exert almost equal but opposite effects, one nullifying the other, It has heen suggested that the abnormally large effect of the sun is accounted for by the synchronising of the natural period of swing of the basin of water between Australia and Antarctica with the period of the tide-producing forces. At American River (see tide curves, fig. 2) the dodge effect scems to be present, though small, at the neap period. For abaut 6 hours the water level remains almost stationary, before the next rise or fall commences, Along much of the north coast this period of sweady water leyel seems to oeenr, but no data are available as yet apart from isolated surveys at American Kiver. During winter the mean sca level at Port Adelaide is from 4 inches to 6 inches higher than in summer. This applies also to Kangaroo Islatid, with con- sequently higher tides during the winter months. With heavy west tu north weather daring winter very high tides often oceur along the north coast. This is due to Inyestigalor Strait being about 24 times as wide as Backstairs Passage, with consequent building up of the water mass in the area north of Kangaroo Island under the influence of westerly weather, The higher sea level during winter is of considerable importance on the south cuast, where the inercase i8 large compared with the tidal range. The level of the horizontal rack plat forms of the Pennington Bay region appears to correspond approximately with an average Jow neap tide level in summer. While north winds and low tides oecasion- ally leave much of the reefs exposed in summer, with consequent drying and desiccation of the algae, this rarely, if ever, occurs in winter, Apart from allowing a heavier growth in winter, this is also one of the factors controlling seasona] changes on the reefs, Similar considerations may apply to a less extent in che American River inlet, 4. Currents The surface current flows from west to cast across ile Great Australian Bight, passes along both sides of Kangaroo Island and on towards Tasmania G 234 (see “Australia Pilot,” 1, 24). Eastward from the Bight the coastal current 1s strongest in the period May to July, with an average rate of 7 miles per day off the coast between Kangaroo Island and Cape Northumberland, During February to April and August to October it is weaker, averaging 3 miles per day, while from November to Jantiary it averages less than 2 miles per day. The currents in Backstairs Passage are largely tidal, reaching a speed of 23 knots. 5. TEMPERATURES Observations taken around the coast of the island give the following results for sea temperature (Table 1). No data from other sources are available, but the figures agree well with the temperature isotherms given by Sverdrup et al (1942). TABLE I Sea Temperatures around the Coast of Kangaroo Island South coast—Summer (Jan.) - . inshore 19-20° C,, offshore probably 18° C. Winter (early June) inshore 16° C. (late July) inshore 13+5° C, (Sept.) - inshore 14° C. { ' inshore 20-21" C., offshore 19-20° C, inshore 11-13° C,, depending on depth and + air temperature; offshore 13-14° C, North coast—Summer (Jan.) - Winter (Jutre) - 4 1 From this table it is evident that the yearly range of sea temperature on the south coast is small, being about 4° C, offshore and 5 to 6° C. on reefs, The range is greater on the north coast and depends greatly on the depth of water and degree of roughness, since the calmer water is affected much more by air tetn- peratures. On the tidal fats at American River temperatures as high as 32° C, have been recorded during stmmer in 6-12’ of water, and as low as 10° C. in winter, Algae on the flats must be able to withstand a far greater range in tem- perature than south coast forms. In isolated rock pools at Vivonne Bay and along the north coast temperatures of up to 30-34° C, are frequent in summer. Atr TEMPERATURES _ The climate of Kangaroo Island is fairly uniform. Some data for Kingscote are given in Table If. The humidity figures probably give little indication of the humidity near algae exposed at low tide. Air temperatures are of greatest importance when a hot day (sometimes 35- 38°C. in summer) coincides with a low tide. Under such conditions algae on the Pennington Bay reefs may be almost or quite exposed for several hours and considerable damage may result. Taste It Air Temperature and Humidity Data for Kingscote (From data made available by the South Australian Weather Bureau) All readings were taken at 50 fect above sea level, over a period of 17 years. : Jan. Feb, Mar. April May June July Avg. Sept. Oct. Nov. Deér. Mealy Mean Max. Temp.°C, 22-3 22-7 21-4 19-3 17-7 14-9 14-2 14-5 15-8 17-8 19-8 21-4 18-4 Mean Min, Temp.°C. 14-5 15:6 14-2 12:35 11:0 96 BF 8S YS 105 12:1 13-8 11-7 Mean Rel. Humidity 70 74 75 76 80 83 82 RO 79 75 72 72 76 235 6, SALINITY Chlorinity of sea water on the south and north coasts is within the range 19'6-19-99/, (salinity 35°4-35°99/,.). North coast values are usually slightly higher than those from the sonth coast. At Pelican Lagoon chlorinity in sum- mer reaches 20°5°/,,, (salinity 37°0°/,,,), while in isolated rock pools (some with a heavy growth of Auteromorpha) summer chlorinity figures of 24-0°/,, have been ubtained, After three days of heavy rain (January 1946) chlorinity on the tidal flats at American River decreased to 17°/,,,,; such conditions, however, are very exceptional. Normal sea salinity around Kangaroo Island is high compared with other regions (¢.g., 34°9°/,, salinity near Sydney (Pope 1943) ). 7. ProsPHate AND NITRATE Estimations of these two major nitrients are as yet tuo few in number for any general conclusions ta be reached. It appears, however, that nitrate is often extremely low (less than 1 part per 10°), while phosphate is rather variable. Phosphate figures of 14 and 23 p.p, 10° have been obtained from the south coast, and values between 2 and 60 pp. 10° from the American River inlet, Isolated high figures. obtained at American River are probably due to the large bird population. 8. ALKALINITY The pH of water (by colorimetric methods) at Pennington Bay is about 8°2-8'3, while at Pelican Lagoon figures of 8*1 have been obtained. 9, Dussorvep OxYGEN The constantly boisterous seas on the south and west coasts result in water supersaturated with oxygen, The water on reefs at Pennington Bay is usually about 110% saturated with oxygen. Oxygen figures in shallow water at the American River inlet in summer show high supersaturation during the day (120- 150%, rarely as high as 250%), dropping to 50-70% saturation at night (un extreme of 10% saturation has been recorded). Such large ranges are due to the heavy growth of algae in the calm shallow water. For the most part it is unlikely that oxygen content of the water is of importance in the algal ecology. The Winkler method was used in all estimations. 10, Licut No attempt has been made to measure light intensities at different depths, but correlation of shaded littoral areas with communities of sub-littoral algae has been observed at Pennington Bay and Vivonne Bay. On the south coast, with constantly broken water, light penetration will be less than in calmer waters off the north coast. At American River the large amount of silt carried in the tidal current reduces light penetration and may influence algal distribution. TERMINOLOGY At the present stage there is little uniformity in nomenclature used in marine algal ecology. Chapman (1946) reviews opinions expressed about terminology, and advocates adoption of the terms used in land ecology. However, atid ecolo- gists are far from agreement on their terminology, and until many more marine ecological studies in different parts of the world have been carried out, uniform and satisfactory meanings of the terms camot be expected. This applies par- ticularly to the Australian coasts. As in the past, each worker must use the terminology which best suits his locality and his own concepts 236 The definitions adopted here have been found satisfactory in describing the algal vegetation of the Kangaroo Island coasts. Only further studies in other regions of the Southern Australian coast will show to what extent the concepts need to be modified, and their usage at present makes no pretension to be final. AsSOciATION AND ComMMUNITY The concept of an association is fundamental to all ecological work, yet many different meanings have been applied to the term, The association is tised here in the sense of a grouping of organisms distinct in species composition and facies from another grouping. It is composed of a dominant or dominants usually accompanied by other species whose presence is determined by responses to factors similar to those influencing the dominants (see Rees 1935). This concept is to some extent subjective, but experience shows (hat most associations are objective entities. Studies over a long stretch of coast are usually necessary betore the associations present can be determined, Intensive work on small areas often results in variations of one basic association being considered as separate associations. On the other hand, an association may be scattcred in its occurrence and cover areas of only a few square feet of rock, yet may be typical of that particular habitat, and pure and well defined in its occurrence. This is especially true of irregular and dissected coastlines, When associations oceur during certain periods af the year only, they sre classed as “seasonal associations,"' The term “community” is commonly tised with the same meaning as associa tion, but often in a more general sense. It is applied in this and following papers when the status of the algal grouping has hot been satistactarily established, T'oRMATIONS Apart from classifying the associations in their zones (see Jater), the only other grouping used is the formation. This is applied to the principal types of marine vegetation, much as it was used by Cotton (2912) for Clare Island ard Rees (1935) for Lough Ine. Cotion’s formations were based on the substrate and environment for algal growth, and comprised the fullowing : 1. Racky shore Formation. 2 Sand and Sandy-mud Formation. 3. Salt-marsft Formation. lle also distinguished: 4. Vegetation of river nsouths. 5. Vegetation of brackish bays, OE these, the Rocky Coast (“Shore’! of Cotton) aud Sand and Sandy-mud Formations are found around Kangaroo Island, and they are real and natural entities. Rees’ formations are based to a larger extent than Cotton's on the degree of Wave action, but this is nearly always closely associated with the nature of the coast, and there is little difference between the formations of Rees and Cotton, The distribution of algae around Kangaroo Island shows that the degree of wave action is the most important environmental factor, as was emphasises! by Rees. Use of the substrate and environment as criteria for “formation” is criticised hy Chapman (1946, p. 658), who advocates following the practice of naming land plant formations on the dominant species. Towever, Tansley (1940). 10 whon Chapman refers, describes Salt-marsh and Sand-dune Formations, and the naming of formations on the type of plant, while conveying at the same time some- thing of the natute of the environment, is common in land ecology (e.g,, mallee, 237 savannah woodlatd, forest formativns), Naming of the formation on the dominant plant or amunal species is quite impracticahle in many cases, snch as with the Sand and Sandy-mud Formation, In any case the formation is an abstraction. Of the four chief characteristics of an association, wiz., floristic composition, life-form, structure and habitat, ‘Tansley uses two only (life-form and structure) as a basis for uniting associations into formations, Tt is just as logical to use either habitat or floristic composition as a crilerion for such higher grouping (Crocker and Wood 1947). Some algal formations in Tansley’s sense can be readily determined by life form, ag., a blue-green formation and coralline-mat formation, On Kangaran Island these formations are formed essentially of a single association, each delimited by definite environmental conditions, with the former occurring above the latter. It is evident that when the distinctive zones of algae around a coust- line have very different life-forms. one zone mus! be chosen on which to base any classification: around Kangarva Island this is the upper sub-ittoral zone, To regard each zone as a distinct iormation is clearly not justified. lt is well not to ose sight of the fundamental principle underlying the existence of communities, namely, that certain species live together in a particular situation because they have been selected by that environment, ie, all of the species have the same habitat requirements for growth, This gives the basis for sicfining associations. Within any one tidal zone, different associations may occur depending tupon lucal variations. These associations often have similar life forms, and possess unity in the fact that they have certain habitat requirements in common. Such “habitat zones” are realities atid are the natural units of higher grade than the association. In practice tide level, degree of exposure to wave action and nature oi the substratum are the chief habitat factors. Using he latter two criteria a large unit (formation) is obtained ; depending on the degree of waye action, the forma- Hott inay be divided into sub-formations; and by using tide levels zonation 1s obtained. Using habitat factors as criteria does present us with realities common in all parts of the world, and in this lies a real hope of achieving some degree of uniformity in marine ecological nomenclature, The formations of Cotton and Rees, or a combination of them, would be of world-wide occurrence, and sub- divisions could well express the characteristic algal groups of the geographical regions, ZONATION The ncetvrence of marine algae in distinct zones between and often below tide levels is @ listinctiye feature of rocky coasis, though nrore prominent where ihe tidal range is large. The tidal range around Kangaroo Island is small (between 2 and 44 feet), but zonation is always present and often tarked. In delimiting the zones around Kangaroo Island absence of accurate tidal data is a limiting factor. Until such data become available, the position of the zones of alwae in relation to tide levels can be given only approximately, and what appear to be critical levels only from subjective observations, ‘Thus the two main littoral associations of rocky cousts are referred to as being in the “upper littoral’ and “lower littoral)! but the relation of these to tide levels cannot be given. The upper littoral] zone of blue-green algae probably does nol extend to high water mark of spring tides, except when influenced by splash effects. As long as the occurrence of algae and animals on the shore ts referred to as “zonation"—a word which js far toa well established to be dropped—there 238 seéins 110 justification for replacing the term “zone” by “belt.” as is advocated by Chapman, It has been possible to relate the algal zones to measurements of tide levels only in the Ainerican River inlet. Here the stationary low water level of dodge tides apjrears to be of most importance, marking the separation of the Hurmosivet zone from a zone of red algae (Hypnea- Ci entroceras~Spyridia) which is nearly always covered. This level is very little higher than the low water level of neap tides (see graphs, fg. 2). On the south coast the low water mark of neap of dodge tides in summer appears to correspond closely with the surface level of the flat rock-platforms, and this marks a distinctive change in the algal flora. The higher mean sea level in winter may cause an elevation of the littoral flora, but the lower summer level will be the limiting factor at least for the more perma- nent algae, The littoral zone is therefore considered as ranging from the stationary low of dodge tides, or the low water mark of neap tides to the upper limit of the algal vegetation, Accurate fixation of this level will have to await detailed tidal information. The term “supralittaral” is often applied to the zone ahove high water level of spring tides. Alternative names are the “splash” or “spray” zones. Cotton (1912) has given good reasons for rejecting this term. and investigations around Kangaroo Island support the view that algal vegetation above actual high water level is simply an upward extension of the upper littoral algae under the influence of shade and wave-splash, One exception to this lies in the occurrence of Prastola during winter at Pennington Bay and on Shag Rock in Pelican Lagoon, well above the arca splashed by waves. This alga is subject to fine blown spray, but is as much terrestrial as marine. In both localities # occurs only where penguin and shag excrement is present. The lichen Lichina may occur in small patches in and above the splash area, and the mollusc Melaraphe untfasciata. extends many feet above high water mark. Apart from these associations the term supralittoral is of little use in describing the algal ecalogy of Kangaroo Island. Below the littoral is the sublittoral, which extends down to the limit of algal vegetation. The upper limit of the sublittoral. particularly on exposed rocky coasts, bears a distinctive algal flora, and this area, between low water mark of neap tides and extreme low water of spring tides, has been termed the “sublittoral fringe” by Stephenson (1939). The sublittoral fringe on rough coasts is exposed during the suck back between waves at low tide, and the short but frequent periods of exposure to air are probably of importance in determining the algal flora present.. On the south and west coasts of the island this zone is dominated by Cystophora infermedia, which is strictly confined to the Tegion exposed between waves at low tide, On calmer rocky coasts other species of Cyvtophara are dominant, but these extend to 6 or more feet below low water, At American River also the flora just below low water is not so distinctive, and is better referred to as “upper sublittoral." The sublittoral fringe must be regarded simply as a useful division of the sublittoral in certain areas, such as the rough coasts of Kangaroo Island. A GENERAL ACCOUNT OF THE ALGAI, ECOLOGY This account is of a general nature only, While it is derived from the study of loealitics illustrating most of the coast and appears at the present stage well founded, it makes ho pretensions to be final, and modification may be necessary as other coastal areas are visited. Descriptions of the typical localities will be given in [ater papers, ‘4Jsvo3 ayy JGequE sBeys JO summuad asayM Ajuo punoj St Jaany weossury pue Aeq VoBuTUUE F¥ LO!EPOSse pose4g BT, MOLRIDOSSR FELL-aUy[esOI ayi WO pasodwt -jadns WMOYs ase Sotmads BuyAuedurosve Juautmosd ysous ay} auc jesojyy FeMo] Puy Ut {hq ‘suotesosse urew ata sastfoquiAs Aunjop Jo Suipeys “puesy ooesuey JO 7sBOI aijt punose parpnys saiqyeso] ayy ye UoOLenoZ jeBpE seq OUT § ‘3a Pont Ig eyeipes f= = =| sI]P44She IAN) dds — eipsWwieyUr (i you F B10) 495 [a5 snjouoy Pscafdoysh proydoyshsy = PULP IOD 1 (aonis' 19 WHOLLITENS Wadd WHoOlin W3M07 WNP lao Wun pijac MALL eiypfsisog | enppdujsog WnipUyrelds [ood eroydayshy eyd rowojaua! (Jayuim) eloiseid heg seg {iad Selah | Keg Jaaty Joan keg aipnop hog Aeg daqueyrayuy' Boy fyroy | ueoi4auy nuw3 FPP Us2aysaM 1s9mM adey auuosL, | UoybuwUdg 239 WHOL LA waddn IS 2 2 tS {apeygs 40 Aeu os) uolsuarx3 yes0}3179 saddy er auney S100d A50uU (uaiuim) Wudns PydueWOIsU3) Bose piveOLy) apig 183 Aqybnoyy! adez 240 Vig. 3 shows the basic zonation at the localities studied. Two main regions are clearly defined: (1) The American River tidal inlet where species of Cystophora (or other large brown algac) are almost completely absent from the upper suh- littoral zone. Horinosira banksii Dene., Celidivint pusillam (Stackh.) Le Jol and Bostrychia simpliciuseula Harv. dominate the basic littoral zones from lower to upper littoral, (2) The rest of the coast of the island where species of Cysfophora (or Eckloma radiata (Turn.) J. Ag.) are dominant in the upper sub- littoral zone, Coraltine-nat and blue-green algae here form the two characteristic zones of the littoral. Hence, depending on the presence or absence of species of C'vstaphora, the coast may be divided into two formations named from their characteristic habitat, the “Sand or Sandy-mud (Flat) Formation” (American River jnlet) and the “Rocky Coast Formation” (see fig. 4). The naming of these formations os rhe habitat has been discussed under “Terminology,” The chief differences between the two formations are: (1) The difference in species compositivn, Species common to. both forma- tions are rare, and if cominon they usually differ greatly in relative abundance, The Gelidium pusillum association, well developed and pro- minent im Pelican Lagoon, is present, but poorly developed at Pennington Bay, and fragments may be found elsewhere along the coast, Hermosira banksit forms well-developed associations in both formations, but ibe ecological forms in cach are yery distinet (sce pl. G, fig. 4, and pl. E, hg. 4). In number af species the Rocky Coast Formation ig siuch richer than the Sahd and Sandy-mud Flat Formation, while the size of the algae 1s usually greater in the former. (2) Methods of attachment. A wide, expanded, holdfast dist is charac- (A — teristic of all the larger algae in the Rocky Coast Formation. Iu rough places this dise is extremely strong. Other methods of attachment are found in the littoral zone; viz., ramifying and densely matted attaching filaments of the coralline mat association, and the gelatinous adhesive thalli of the blue-green association of the upper littoral. In the American River inlet the expanded holdfast disc is rarely found, and the adhesive thalli of blue-green algae are almost completely absent from the upper littoral. The chief mode of attachment is hy thizoidal filaments (a notable exception being Hormosira), Attach- ment of filaments by means of a basal cell or cells occurs in both formations. The growth substrahim. On rocky coasts the vast majority of algue grow on rock, while a few occur as epiphytes where the growth is dense. Many epiphytes can grow equally well on tock or on other algae, but a few are limited to particular hosts (¢.g., Notheia on Hormosira). At American River most of the tidal flats are colonised by the marine Angiosperms Posidonia australis Hook, and Zostera wtuelleri lrmisch, and these bear a profusion of epiphytic algae. Apart from Bostrychia, Gelidium and Porphyra, which occur on rock in the upper littoral, and to some extent Horimosira, all others are either epiphytes or grow on shells anc small stones in the sand and mud. No macru- scopic algac are able to grow directly in or on the sand or mud, although they may often be partly buried in mud; growth in sand or mud is, how- ever, characteristic of the marine Angiosperms. 241 In discussing the Rocky Coast Formation, no account is taken of stretches of sandy beach between rocky sections of the coast. Rees considered such areas as a separate subtormation, but devoid of algae. No macroscopic forms are found on such beaches, and wherever rocks occur the typical algae of the Rocky Coast Formatiori are found. Within the Rocky Coast Formation conditions of roughness vary from very rough to moderately calm, and the coast may be divided on the presence or absence of Cystophora intermedia. This brown alga is found ouly in conditions of Lairly strong to very strong wave action, and never occurs on calm coasts, It grows best under the constant action of heavy breakers, and has by far the strongest thallus for its thickness of any Kangaroo Island alga. Cystophora intermedia is dominant in the sublittoral fringe zone fromr Cape Willoughby along the south and west coasts, and along the north coast to between Western River and Middle River, where it is replaced by other species of Cyslaphora, . silyy Exposed Becky Coast Sub-formation. Rarhy Sheltered Rocay Coast pron y G a ast Veg y Sub-formadun. Formetion, Sand and Sandy-mud Formatven, Fig. 4 The Algal Formations and Subformations arownd the Kangaroo [sland coast, Aréas in the Bay of Shoals and Western Cove not yet stUdied are left unclassified, Only the localities actually examincd are shown on the map, On the north coast between Cape Willoughby and Middle River (excepting the Ameticans River inlet) other species of Cystophara (C. subfarcinala ( Mert.) J. Ag., C. siliquosa J. Ag., and others) or Ecklonia radiata, and in some places the red alga Acrotylus australis J. Ag., are dominant in the upper sublittoral zone. These species require constant water movement, but sudden rough weather will remove many plants from the rock. The outer edge af the Pennington Bay rock platforms is very rough, with calmer conditions nearer in and at the rear of the reefs, where Cystaphora subfarciiata and C. siliquosa are very common. Along the whole sauth coast, however, C. inlermedia is dominant in the sublittoral fringe, and although fairly common, the species characteristic of more sheltered coasts occur only where conditions ate Jocally Jess rough, The Rocky Coast Formation is therefore divided as follows (see fig. 4): 242 (1) Tue Exrosep Rocxy Coast Suprormation: from Cape Willoughby along the south, west and north coasts to between Western Riyer and Middle River. The area is characterised by the presence of Cystophora intermedia in the sublittoral fringe. (2) Tue SuELTereD Rocky Coast Suprormation: found along the north coast between Cape Willoughby and Middle River, excluding the Ameri- can River inlet, Characterised by other species of Cystophora, Ecklonta radiata, and in some ateas Acrotylus australis, m the upper sublittoral, Sargassum spp, may also occur in some areas. Cotton found it necessary to divide his Rocky Shore Formation into Exposed and Sheltered Series. This appears to correspond closely as far as environment and status go with the two subformations of the Rocky Coast Formation on Kangaroo Island. Although referred ta as “sheltered,” the degree of shelter in this sub- formation is very much less than in the Sand-and Sandy-mud Formation. The latter is developed mainly in almost land-locked areas, whereas the Rocky Coast Formation is always found on open coasts. Within the exposed Rocky Coast Formation two distinctive types of habitat occur, dependent om the geology of the coast (see “Coastal Geology,” under “Environment”), These are the horizontal sand rock reefs, actually wave cut platforms, occurring along much of the south coast (pl. ix, fg. 3), and the steeply sloping rocky areas occupying the rest of the coast (pl. xii, fg, 1), A brief description of the main associations in these two areas is given below, but detailed reports will be left till later papers. A, THE ROCKY COAST FORMATION I. THE EXPOSED ROCKY COAST SUBFORMATION (a) The Pennington Bay Rock Platforms The type of horizontal waye-cut platform (pl. ix, fig. 3) found at Penning~ ton Bay occurs along much of the south coast of Kangaroo Island (see fig. 1)- The reefs which have been studied in detail at Pennington Bay are probably representative of this type of algal habitat, arid a detailed account will be given in another paper, The following are the main associations found in the Penning- ton Bay area: {1) Tse Litrorat Zone. Rear LrrroxaL Assoctations—These occur on the vertical or sloping rock backing the reefs, usually at a higher elevation that the reef itself. The associations are exposed at low tide, but washed or splashed continuously at medium and high tides.. 1, Rivularia firma association (pl. x, fig. 2), This alga favours areas where wave splash is moderate or else there is constantly running water, For further notes sce Womersley (19404). 2. Symploca hydnoides association: forming scattered patches in shaded hollows of vertical or sloping rocks, 3. Gelidium pusilluim association: common, but usually poorly developed, 4, Enteromorpha association: forming bright green, usually pure areas on sloping well-washed rack. 5. Ectocarpus confervoides and Pylaiella seasonal associations. These form brown mats and tufts on well-washed sloping rock, Ectocarpus occurring during winter and Pylaiella mainly in summer. 245 Lirroran Associations (on the flat reef surface). 6. The Cystophora complex. Four species of Cystophora, C. subfarcinata, C. siliquosa, C. uaifera (Ag.) J. Ag. and C, brownii (Turn,) J, Ag,, together with Sargassum muriculatum J. Ag., form a complex of associa~ tions on the flat reef surface, always where they are submerged. Each species may form a pure association or occur mixed with one or more of the others, depending on the depth of water at low tide (pl. x, fig. 1), This complex covers the larger part of the reef sutface. 7. Hormosira banksti association, This is a well-marked association on slightly higher and therefore more exposed parts of the flat reef surface. On higher areas it is pure and dense (pl. ix, fig. 4, and pl. x, fig, 4); in other places it may become mixed with species of Cystophora. Notheia enomala Bail, et Harv. is always found growing from the conceptacles, & Cystophyllum muricatum association. A pure area of this alga occurs in well-washed, relatively calm, and rather sandy parts of the main reef, 9. Laurencia heteroclada association. This occurs in fairly rough places, where it forms a dense mat of stunted piants. 10, Cystopkora—coralline association, A distinctive association found on the rougher parts of the reefs, consisting of Corallina curieri Lamour. on rock and Jania fastigiata Harv, on species of Cystophora; a dense and rich association. (2) THe Supcirrorat Frince 11. Cystophora intermedia association. This brown alga dominates the outer edges of the reefs in the roughest conditions. It is an extremely rich association of small, often stunted species (over 50 have been recorded from an area of a few square yards), completely covering the rock (pl. ix, fig. 2). The stiblittoral assemblage will not be dealt with here, but it is very sitnilar to that listed for Viyonne Bay (see p. 244). The coast at Pennington Bay js very rich in number of species; over an area of 4 mile more than 220 species have been recorded, and many more, no doubt, remain to be found. (b) Steeply Sloping Coasts Three main zones, each coniprising one association and in certain localities others, uccur in the intertidal areas of the more steeply sloping parts of the south and west coasts. Heavy wave splash or regular passage of breakers up sloping rocks results in considerable upward extension of these zones. 1. Rivunarra-Isactrs Associarion of the upper littoral, This association is composed of Hiznilaria firma Womersley, R. atra Roth. and Jsactts plana (Hatv.) Thuret, all forming scattered, dark blue-green pela- tinous thalli on otherwise bare rock. In some areas they are very well developed (especially R. firma), in other places they are almost absent. Degree of wave action is the determining factor, but the association is often poorly developed where least expecied. 2, CORALLINE-MAT ASSOCTATION. The lower littoral, between the blue-green algae and the sub-littoral fringe, usually consists of a dense mat of stunted Fania fastigiata and/or Corallina (pro- hably C. cuateri) (pl. xi, fig. 4). This mat is 1-3 em: in thickness, pinkish-white in colour, and forms a continuous covering-on much of the rock. Where breakers run well up sloping rocks it may reach a height of 5 or 6 feet, At Vivonne Bay 244 (on the south side of Ellen Point), Dasyopsis clavigera Womersley, and a small stout Lawencia are prominent amongst the corallincs. This Jania-Dasyopsis- Laurencia variant is probably general in many arcas of the south and west coasts, but rather than being a distinct association it consists of the addition of the latter {vo algae to the baste coralline-mat association, 3. C¥sTOPHORA INTERMEDIA Association of the sublitoral irttige. This brown alga forms a Striking sublittoral fringe zone on sloping rocks af the south, west and north-western coasts of Kangaroo Island, The upper edge of the ussociation is olten very sharply limited, as shown in pl. xi, fig. 2, the zone appearing as a dark band stretching along vertical rock at Cape Willoughby. In situations where waves pass along the rock, rather than breaking against it, the coralline-mat and blue-green zones may be poorly developed, but the sharpness of the upper limit to Cystuphora intermedia can be sven from pl. x, fg. 3. Where waves break heavily on rocks the upper edge 1s less well defined, and the coralline- mat often merges with Cystophora trtermedia (pl. xi, fig. 4). The dark-brown pinnate fronds of Cystophora intermedia reach a length of 40 to 45 em, The stems are extremely strong, and only very rarely are fronds found cast up. A common epiphyte is Corynephloea cystophoree J. Ag. THe SUBLITTORAL, Study of the sublittoral flora is restricted to the algae cast up but not known to occur in the intertidal area, The following list includes the commonest forms af the sublittoral assemblage of the south coast, but comprises omly a small fraction of the total, CiLoropiyceaE—Caulerpa harveyi F. v. M.; C. obsew‘a Sonder; C. vesiculifera Harvey: Codium goleatim J. Ag.; C. masnllosiun Marvey; C. pomoides J. Ag. PicseoruyckAr—Phloeocaulon spectabile Reinke; Dictyeta latifolia J. Ag.; Zanaria turnoriana J. Ag.; Sporochnus scoparius Harvey; S. comosus C, Ag.: Bellotia eriophorum Harv.; Bacyathalia cliftont Harvey; Perithatia inermus (R. Br.) J. Ag.; Keklonia radiata (Turn.) J. Ag.; Scytothatia dorycarpa (Turn.) Grev,; Sierecoccus axillaris Greville: Scaberia agardit Grey.; Myriodesma quercifoliun (Bory) J, Ag.; Carpoglossam confluens (R. Br) Kiitz.; Cvstophura monilifera |. Ag.; C. dumosa J. Ag. C. reforta ( Mert.) J. Ag: ©. racemosa Hary.; C, platylobtum (Mert.) J. Ag., ©. spartioides {Turn.) J. Agi; C. siliquesa J, Ag.; C. paniculata (Turn,) J. Ag.; Sargas- “a varias Souder; S. sonderi |. Ag. S. trichaphylliam J, Ag. ; S. cristatiem _ Ag. Ryovopuycear—Asparagopsis ermata Uarvey; Mychodea compressa Harvey; Hypnea episeopalis 11, & H.; Delisew elegans C. Ag., Phacelocarpus lubillar- dieri J. Ag.; Plocamium nidificum (iarv.) J. Ag.; P. preissianwin Sonder; P_ coslatum (J. Ag.) H. & H.; Avmenocladia polymorpha (Harv.) J. Agi; Antithamnion wmucronatum (J. Ag.) De Toni; Afonospora clongata (Harv, ) De Toni; Cereminm puberulum Sonder; Lasiothalia formosa (Warv.) De Toni; Spongocloniwm sp.; Spyridia opposita Hary.; Sarcomenio dasyoides Harv.; Nitophylhen curdieanin YWary.; Amansia pinnatifida Harvey; Lenormandia spectabilis Sonder; Osmunduria prolifera Lamour. ; Thuretia unercifolia Dene. Oruer ComMuNItiEsS oF STEEPLY Siorinc Coasts An Enteromorpha association occurs on ruck well above normal wave-splash at Cape Coudie. It is dependent on the presence of fresh water percoiating through the upper limestone stratum and ruming down oyer the harder ancient rocks forming the base of the cliffs. “The species has not been determined, but it occurs in dense, pure masses on otherwise bare rock. Enteromorpha associations dependent on the presence of fresh water have been recorded by numerous other authors (see Cotton 1912), A Splachnidiwm rugosum association is found in the upper littoral at Cape Willoughby, usually at a higher level than the Iue-green zone. The Cape is composed of granite boulders, and where waves break heavily, leaving the rock exposed hetween waves, Splachmidivm forms a pure association of short, tufted plants (sce pl, xi, fig. 3), At Vivanne Bay, on gneissie rock, Splacknidivns rugoasum: (1...) Grev. is often common on wave-splashed rock, but may merge with Rivilaria firma, Helnin- thora tumens J, Ag. and Polysiphonia dasyoides Zan, are also characteristic of this region during January. On other types of rock Splachnidium is very rare. If further studies show thar it is restricted to granite or gneissic rock, this will be one of the very iew cases known from Kangaroo Island of the type of tock influencing algal distribution. A marked feature of the south side of Ellen Point, Vivonite Hay, is the occurrence of at least five species of lithothamnia, Elsewhere om the island they are rare. Two distinct species (generic determination has not yet been possible) form pure but localised communities in the littoral zone in what are apparently rither specialised habitats. They grow as crustose thalli forming small irregular branches. The coast at Ellen Point consists af fossiliferons caleareous limestone over- lying the hard gmeissic base. Wearing hack of the softer limestone has resulted in many rock pools, from very small to over 20 yards across, being left in the harder base (pl. xi, fig. 1), Most of these pools are subject to wave itflux only at high tide, and during summer their water temperature fs cot- siderably higher than that of the sea (up to 28° C. when xea temperature is ree C.). . home ot these pools bear distinctive algal conmmunities; others, where con- fitions are apparently too severe, are devoid of growth. One pool, shown in the foreground in pl. xi, fig, 1, contains a Lavrencig— lithothamnion community. The lithothammion Lorms scattered, irregular pinkish masses (to 10 em, across and 3 cm, thick), while Lawrencia heteroclada Harv, gtows on the rack or the lithothamnion and is heavily epiphytised by Céeraminm miniiatin. Suhr. and Polysiphonia abscissa. Harv, In the rear pool of the two shown in pl. xi, fig. 1, the end shaded by the cliff hears a community of red algae which are normally sublittoral forms, Dietymenid tridens Grev. and RBornetii sp. are the commonest, while in another shaded aren of the same poal Lyngbya meajuscula (Dillw.) Harv, forms a pure community in January. The effect of continual shade is evident in both cases. In another pool, at a lower level and snbject to wave influx except at low tide, species of Cysfophora are dominant. One corner, however, is shaded by overhanging rock, and here Ecklonia radiata, Scyhothealia doryearpa, M yriodesmut fotifola Varv, var, duriuscula J. Ag. (with epiphytic Sphacclaria tribulvides Menegh.) and Gelidium onstrale J. Ag. are prominent, All these are Hormally upper sublittoral forms, ; Tn some of the pools minor cotimuinities of Bryopsis plumosa (Huds) C. Ag. Bryopsis baciulifera J. Ag.; Derbesia sp, two lithothamnia, and two species of coral occur. This assemblage shows more relationship to the flora af tropical waters, and is almost certainly due to the higher temperatures main- tained in these pools during summer. 246 To deal adequately with the complex nature of the littoral zone at Ellen Point requires detailed mapping of the greatly dissected coastline. The variation in minor habitats is almost without limit, and similar complex areas probably oceur along other parts of the south coast. However, the basic zonation of bluc- green, coralline-mat and Cystophora intermedia zones is found on all rocks directly exposed to the sea. Ir. THE SHELTERED ROCKY COAST SUBFORMATION Wave action on the coast included under this subformation is from moderate to slight (see pl. xii, fig. 2 and 4). In fine weather waves gently Jap the shore, while breakers a few feet high occur in rough weather. Some degree of water movement is always present, whereas in the American River inlet conditions are more often than not a dead calm on the tidal flats, Littoral zonation is basically similar to that om exposed coasts, comprising blue-green and coralline-mat associations. 1, Tse Uprer Littorar Zone of BLUE-GREEN ALGAE. On the eastern end of the island, and at Middle River, Rivularia firma 1s dominant, accompanied by Rivularia atra, [sactis plana and sometimes Symploca hydnoides Kutz, In calmer areas KR. firma disappears and J. atra and /sactis become dominant, From Middle River to Stokes Bay (and probably further east) Brachytrichia quoyi (Ag.) B. & F. is prominent in January. In some places where wave-splash is absent this blue-green zone may be very inconspicuous, The gelatinous thalli occtir scattered singly or in patches on other- wise bare rock. No other algae normally occur in this area of the upper littoral. Blue-green algae are absent at Rocky Point, where the substrate is consolidated sand rock and the littoral zonation is closely allied to that found in Pelican Lagont. A community of Nemalion helminthoides ( Velley) Batt. occtrs on rocks on the east side of the beach at Middie River, in the mid-littoral, while on the west side of the beach Cladosiphon. filum (Harv.) Kylin is common below the blue- green zone. 2. CORALLINE-MAT ASSOCIATION. This is ustially well developed in the lower littoral, often forming a closed community with a well-defined upper edge (see pi. xii, fig. 4, at Emu Bay). Janta fastigiata and fragments of C erallina are the tain constituents, but Dasyopsrs elavigera and the Laurencia of the south coast association are abscut. Gelidium prustllusn is often, in calmer localities, am integral part of the mat, while other species commonly present are; Mrangehia plumosa Harv,, Pachydictyon pant- culatum J. Ag., Zonaria turneriana J. Ag., Amphiroa charoides Lamour, Laurencia heteroclada Harv., Ceramiuim miniatum and Polysiphonia spp. At Rocky Point Gelidiwn pusilluim has become completely dominant, with enly fragments of coralline left amongst the Horimosira association at a lower Jevel. Above this is a zone of Bostrychta, similar but often better defined than in Pelican Lagoon, 3. Tue Uprer Suscirrorar. ZONE. This region, on sheltered rocky coasts, is characterised by the dominance of fairly large brown algae, forming several associations in different localities. Cystophora association. Three species, C. subfarcinata, C. polycystidia Aresch., C. siliquosa, and to a lesser extent C. spartioides, extend from low water mark to a depth of 6 or $ feet, They may reach a length of 14 metres, and at luw tide in some localities 247 the fronds float at or below water surface, giving the zone a distinctive appearance. [ C. subfarcinata is fotind throughout the subformation, but in calmer areas (e.g., Rocky Point) forms numerous vesicles and is more branched. C. siliquosa and C. sparttoides are restricted to slightly rougher parts, while C. polycystidia becomes dominant in calmer regions. In shallow water and locally calm places Cystaphyllum muricatum forms a distinct community, Although C. polycystidia is characteristic of the north coast of Kangaroo Island, it does occur in locally sheltered places on the south coast, On the east side of Ballast Head (north of American River) Sargassum sp. dominates the upper stblittoral. Ecklonia radiata association. At Hog Bay and Rocky Point, where wave action is slight, Ecklonia radiata forms a distinct sublittoral fringe, accompanied by some C. snbfarcinata. Under locally suitable conditions elsewhere around the island it may be found; at Cape Coudie a protected channel bears a dense fringe of Ecklania. Acrotylus australis association, At Middle River and the east side of Cape Willoughby dense and pure patches of the red alga Acrotylus australis occiir in the upper sublittoral. The dark brown dichotomous fronds, from 10 to 20 em. high, completely cover the rock, forming a distinct association. At both localities conditions of wave action are very similar, and the general algal ecology is almost identical Caulerpa brown Endl. often forms dense bright green mats within a few feet of low water mark, The following species are commonly cast up within the Sheltered Rocky Coast Subformation; Codiuim spongiosum Harv., C. pomoides, Cladostephus verticillatus (Light£.) Ag., Halopteris pseudospicata Sauv., Sargassum sp. (small stunted plants), Cystophora batryocystis Sonder, C. grevilles (Ag.) J, Ag. C. monilifera J. Ag. Amphiroa charoides, Dictymenia harveyane Sonder, Laurencia, sp. In Eastern Cove, atid probably in Western Cove, at least four distinet asso- ciatiuns wecur in deep waler, (1) Posidonia australis, known as the ‘sea grass” or “tape weed,” forms extensive meadows on a sandy bottom in from 1 to 6 or 7 fathoms of water, (2) Scaberia agardhii Grev. occurs on a rocky bottom in from 4 to 3 or even 5 fathoms below low water. (3) Cystophora monilifera occurs in from 1 to 7 fathoms, (4) Chiracanthia arborea (Iarv.) Falk, forms dense masses, especially in winter, in 1 to 24 fathoms. The shore in the Bay of Shoals and Western Cove is sandy and muddy, with few rocky areas. Tidal flats, however, ate not formed to any extent. ‘This type of habitat is intermediate between the Sheltered Rocky Coast Subfermation and the Sand or Sandy-mud Formation, with closer affinities to the latter. Until the area has beett more thoroughly investigated, no classification will be attempted. Along the shore of the Bay of Shoals Zostera muelleri is common, and amongst it, on old shells, occurs the green alga Acetabularia peniculus BR. Br. (probably a wititer form). In deepet water Posidonia australis is dominant. Rock Poon AssocraTIons, At Western River and Middle River rock pools are a feature of the coast. They are mostly small, from 1 to 10 feet across and to 2 or 3 feet deep. During summer the temperature in smaller pools (containing Hnteromorphe) reaches 248 35°C. The conditions in any pool depend on its size, height above sea level, and general situation; the environment of smaller pools duriig summer is extreme in both temperature and salinity conditions. ‘Two types of pools occur: (1) EntixomorrHa Poon Association. This association occtirs in the smaller and higher pools where coriditions are extreme and very variable. &. lingtulata J. Ag, and E. tnlestinalis (L.) Link, form a dense fringe around the edge. In summer exposure on the water surface often kills and bleaches the upper plants. The toleration these algae have for high temperatures is shown by their active oxygen liberation under temperatures of 30-35° C, (2) Hormosira -CrstorHorA Poor, Assocrarron. In larger and lower pools, where waves enter more frequently ahd emperalures are therefore lower, Hormosira banksiit forms a dense fringe around the edge, at or just below water level (see pl. xii, fig. 3), while Cystophora subforcinata, C. palycystidia, C, siliquosa, C. brownii, and often Cystophyllum meuricatieu and Surgassin sp. grow on the lower sides and bottom. In many pools along the north coast, particularly those with a sandy bottom, the only algal growth consists of small mats of Gelidium pusillten and tragments of Corallina and Jania, B. THE SAND AND SANDY-MUD FORMATION The American River tidal inlet comprises several large lagoons with wide tidal fats and a central channel, opening into Eastern Cove through a narrow neck. Conditions are very calm, particularly on the tidal flats where large beds of Pos:dona and other weed tend to minimise wave action. Tidal range is just over 4 feet, decreasing only 2” or 3” from American River to Pelican Lagoon (see fig. 2), The fast scouring currents during spring tides, together with the sandy bottom, prevent algal growth in the channel proper. The temperature range on the flats is large, for at low tide less than a foot, and often only 1 or 2” af water covers the algae. Winter temperatures reach as low as 10° C., summer up to 32° C. on the flats, In Pelican Lagoon the tidal flats usually extend out from low ¢liffs of cal- careous sand-rock (similar to the coast at Peuningion Bay) (pl. xiii, fig. 2), but sandy beaches are frequent, especially between Muston and American River jetty, South of Muston samphire swamps cover several miles of the shore, An important characteristic of this formation is the large quantity af move- able sand and mud. The fast tidal currents carry suspended mud, which algac on the flats must be able to tolerate. The characteristic colour of Rhodophyccae at “American River is a dirty brown, very different from the red of clean water forins at Pennington Bay. The basic zonation in Pelican Lagoon is shown in fig. 5. Where a sandy beach occurs, [Yormosira or Zostera comprise the upper zone. Over most of the flats the area colonised by Hormioyira, Zostera and Posidonia is much greater than shown in the figure, but the sequence of zones and their positions in relation to tide levels applies generally. Microscopic algae, particularly diatoms, are usually present as epiphytes on the larger algae, but identification has not been attempted. On Shag Rock (a small island at the entrance of Pelican Lagoon), and probably elsewhere where shag colonies occur, a filamentous form of Prasiola covers rocks affected by the bird droppings, This appears te be a winter asgocia- tion only, ovcurring seyeral feet above high tide level. 249 LitTorAL ASSOCIATIONS 1. BostrycutA AnD GreLIDIUM ASSUCIATIONS. These two associations will be considered together, as they occur in very similar habitats and sometimes become mixed, At Rocky Point, however, cach association forms a distinct zone, the Gelidiuwm belaw the Bostryciia. Roth algae form dark reddish-brown dense mats, up ta 1 em. thick, covermg the rock from about mid to just below high water level of spring tides. They are restricted to shaded areas of rock (see pl. xin, fig, 2, and fig. 5), With continual shade, the mats retain sufficient water during the periods of exposure. Rock not shaded in this zone is usttally bare of macroscopic algae. Sh | “ a4, he ae te ire 5wo Eo. thes Lwsl----—-~=—~~——~—~~—~—~——~—~—-—-—*> Mii pe >= -| > bik yer ry, WM) 2 G4 ’ 3 ty | Gracilaria (on mud) -Sr Porphyra wht) Whe Bost fostrchie Bais ane) o™ Harasin a LL ee ee aidan la Bare Zone dormosira Hypnea~Centroceras- Zostera Posidonia di (molluscs) Spyridia Vig. 5 Typical algal zcnation on Pig Island, Pelican Lagoon, The rel lation of each zone 1o tide levels is approximately true; but in many areas the tidal fats are much wider and the Marmosira, Zostera and Po: sidoitue assuciations occupy far greater areas than indicated. H.W.S., bigh water springs; 11..W S,, low water springs; FLW.N,, high water. neapa: S.W.D., stationary water dodges. As a general rule, Gelidium pusillum covers lower and more exposed rock than Bostrychia (B. simplictuscula, with some B. mixta H. & H.). The roof and sides of small wave-cut caves are the rnost suitable habitat for Bostrychia. During winter a community of Porphyra umbilicalis (L.) J, Ag. occurs in the lower Gelidium zone, but on more exposed rocks. The red-purple fronds reach a lefigth of only 8 to 12 cm,, and Tie as flat sheets on the rock at low tide. From the base of the low cliffs to the start of the Hormosira association (a distance of up to 20 yards) is a zone bare of macroscopic algal growth. The area is left quite dry and fully exposed to the sun at low tide. Such conditions are too severe for any alga, but the molluscs Bembicrum melanostama and Modtolus areolatus are common. H bho wn co) 2. Jlormosira Association, Hormosira banksii forms an extensive and couspicuous association in the low littoral, from abour 1 fool above the dodge low to atv inch or two below. Each plant erows either on the cock, or, if the substrate is sandy or muddy, on the partially buried bivalve Brachyodentes crosus. The spherical water contain- ing receptacles of Hormosira enable it to xrow where it is always exposed at low tide (pl. xa, fig. 4). The form of Hormosira banksit at Ainerican River is labillardicri Harv. Tt differs from the south coast fori in being more branched, and having larger aud more spherical vesicles. In both localities it grows only where it is exposed for a period each day; this exposure must in some way be essential ta the growth of the alga. During winter Ectocarpus confervoides (Roth.) Le Jol, is a comman epiphyte on Hormesira at American River, : 3, Utva Association, Ulva factuca is always prominent at about low tide level, in the lower Hormosivu and upper Aypuea- Ceatroceras - Spyridia associations. On the flats north of the American River jetty it fotms a green band along the shore, super- imposed on the Zoslera. association. In other areas it frequently becomes dominant to form a distinct association. Forms of Enteromorpha prolifera J. Ag., FE, clothrata Roth. and &, bulbosa Kutz. are common in the lower littoral and upper sublitteral, sometimes forming communities. SupLittoran ASSOCIATIONS 4. Hyrrea -Centrocerass - SpvemraA AssociAtron. This association extends from about low dodge level to 1 foot or slightly more below. The dominant algae are brown-colovred Rhodophyceae, and where dense give the zone an even brown appearance. Hypnea miusciformis (Wult.) Lameour,,. Cenivoceras clavulatun Ag, and Spyridia bviannilata J. Ag. occur in varying but often about equal proportions. They are all slender forms, rarely more than 25 em, long, and are just expused at very low tides. In some places Gracilaria confervoides (L.) Grev. is common; on muddy patches around rhe small islands in Pelican Lagoon it forms dense but Ipcalised communities. The hase of each plant is buricd in mud but actually attached to a molluse, Other algae occur in the lower part of this association, and in somewhat deeper water. The distinctive feature, however, is the virtual ahsence af any species of Cystophora or other large brown alga. (The one exception is Surqassin bifyrnie, which occasionally grows of rock on the sides of the channel where there ig more water movement, but rarely in the tipper sublittoral.) The commonest speci¢s are: Canlerpa remotifolia Sonder; C. simpliciuscula; Codiuin muelleri utz., a lithothaminion forming spherical nodules to 5 em. across; Corallina sy. (hemispherical tufts 20 cm. across), and more rarely Crstophylluam muricatynt, 5. ZoOsreRA MUELLERT ASSOCIATION, Zostera muellert Lorms a pure atid dense association over large areas: of the tidal flats (pl. xii, fig. 3), from low water level to 2 feet below, m some places to uy nach as 6 feet belaw, This angiosperm spreads, almost entirely by means of runners which anchor the plant in the mud and produce narrow leaves (to 25 cm. Jong). dt seems to prefer a muddy substratum. Epiphytes on the leaves are common (pl. xii, fig, 3), especially Centroceras clawulatum, 251 Cladophora ceratina WKitiz., Ceramiwn sp, and Nimenularia polyotis (Ag.) Bb, & I. Amongst the Zostere, on sinall sandy patches, Chondria dasyphila (Grev.) C. Ag., Spyridia biunnulata, Polysiphaiia patersonis Sonder and Graciliaria confervoides oceur. Very rarely plants of Cystephyllum muri- catum and Cystophera cephatornillas (Lab.) J. Ag. may be tound. G. PesinontA austrapis AssOcraTIoN, Posidonia australis colonises deeper parts of the tidal flats and the chantel edges, [rom 1 to 10 feet below extreme low water, with occasional plants to 14 feer, In Eastern Cove it extends to about 7 fathoms. The long strap-like leaves have a distinctive appearanes ay they just reach the water surtace on the flats. at Juw tide, The association is dense and pure. but the rough leaves hear a wealth of epiphytes. On sntall pieces of leaf JO species are afien present. The most important ones are: Mivularia polyotis, Ectocdrpus sp.,