VOL. 91 DECEMBER, 1967 TRANSACTIONS OF THE ROYAL SOCIETY OF SOUTH AUSTRALIA INCORPORATED ADELAIDE PUBLISHED AND SOLD AT THE SOCIETY'S ROOMS KINTORE AVENUE, ADELAIDE Price: Six Dollars Thirty Cents A REVISION OF THE STARFISH GENUS UNIOPHORA (ASTEROIDEA; ASTERITDAE) BY S. A. SHEPHERD* Summary The genus Uniophora which is widely distributed in temperate Australian seas is reviewed. Uniophora globifera, U. fungifera, U. sinusoida, U. multispina, U. obesa, U. uniserialis are placed in the synonymy of U. granifera, and U. gymnonota is considered to be a synonym of U. nuda. Population samples from Spencer and St. Vincent Gulfs, South Australia, are analysed and statistical differences noted. Specimens from New South Wales, Victoria and Tasmania are compared and observations are made upon the ecology and distribution of the South Australian forms. A REVISION OF THE STARFISH GENUS UNIOPHORA (Asteroidea; Asteriidae) by S, A. SHerpHEeRD® [Read 14 July 1966] SUMMARY The genus Uniophora which is widely distributed in temperate Australian seas is reviewed, Untophora globifera, U. fungifera, U. sinusoida, U. multiypina, U. obesa, U. uniserialiy are placed in the synonymy of U. granifera, and U. Eymnonota is considered to be a synonym of U, nuda. Population samples from Spencer and St. Vincent Gulfs, South Anstralia, are analysed and. statistical differences noted, Specimens from New South Wales, Victoria and Tasmania are compared and observations are made upon the ecology and distribution of the South Australian forms. INTRODUCTION The genus Uniophora is a prominent and distinctive representative of the seastars of temperate Australian waters having a distribution around the coast of the southerly half of the continent from Collaray Reef near Sydney on the east coast to Lancelin Island in Western Australia, The members of the genus are extremely variable and have long been a source of perplexity to taxonomists. Lamarck (1816) described Asterias grani- fera from Tasmania and Gray (1841) created the genus Uniophora naming U, globifera from New South Wales. Perrier (1875) described Asterias nuda from Pt. Lincoln, South Australia, Asterias fungifera from “Nouvelle Hollande” and Asterias sinusoida from Hobart, Tasmania. The species of Lamarck and Perrier were all subsequently placed in the genus Uniophora. In 1923 Clark added U. dyscrita from Western Australia and in 1928 U. gymmnonota, U. muitispina, U. obesa and U, wniserialis all from Spencer or St. Vincent Gulfs, South Aus- tralia. Subsequent shore collecting (Cotton and Godfrey, 1942) showed that of the ten species, certainly eight and probably nine were to be found in the gulfs of South Australia. This paper is largely the result of collecting by the author with SCUBA diving apparatus over a number of years. The examination of the considerable material so gathered shows that the supposed speciation of the genus in the South Australian gulfs does not exist, and it is believed that there are only two species in South Australian seas. The collections of the Australian Museum, the National Museum of Victoria and the Western Australian Museum have been examined; the forms from Victoria, New South Wales and Tasmania are vonsidered conspecific with U. granifera but the Western Australian form U. dyscrita is regarded as specific- ally distinct. Differences between the several forms are discussed and spine *c/- South Australian Museum, North Terrace, Adelaide, South Australia. Trans. Roy. Soc. §. Aust, (1967), Vol. 91, & 5S, A. SHEPHERD counts are used lo determine statistical differences between population samples. This method is belicyed to be noyel in the study of astervids. Geographical and ecological notes are also given and a map showing the distributions of the various forms in the gulf regions is added, In this paper the following abbreviations are used; A.M., Australian Musemm; N.M.Y., National Museum of Victoria; S.A.M., South Australian Museum. W.A.M,, Western Australian Museum, UNIOPHORA GRANIFERA Lamarck Asterias granifera Lamarck, 1516, p, 560, Uniophora globifera Gray, 1840, p, 288, Astevias fungifera Perrier, 1875, p. 337, Asterias sinusoida Perrier, 1875, p. 338. Uniophora sinusoida Clark, 1928, p. 411. Uniophora obesa Clark, 1928, p. 409, Uniophora uniserialis Clark, 1928, p. 415. Uniophora. multispina Clark, 1928, p, 407. Uniophora multispina multispina Stach, 1938, p. 832. Uniophora multispina uniserialis Stach, 1938, p. 332, Material Examined, “Granifera” group. S.A.M. (117 specimens) KI174-6, K191, K597-8, K600, K629-K641. A.M. (16 specimens) from N.S,W. ].127-8, J.l44, J.7444, J.6884, ],7008. From Tas, J141, J.143, |.168-9, ].5410-5414. N.M.V., 5 specimens labelled Port Phillip Survey, “Multispina” yroup (98 speei- tnens) S.A.M. K181-38, KI85, K187-9, KI90, KI93. K520-1, R601-2, K605, K626, K635, K643-651. In 1875 Perrier (1875, p. 342) noted the puzzling diversity of the specimens befure him and considered them to comprise four distinet species. Later Fisher (1923, p. 597) referred Gray's. globifera to the species granifera. Tt is now clear that there is only one species which shows considerable morphological variation both between and within its component populations. The differences in mor- pholowy relate principally to the frequency, size and regularity of arrangement of the dorsal spines. The populations examined fall into two groups cach of which has its own preferred habitat and. finds a distinctive phenotypic expression. One group is called “granifera” as specimens of it have predominantly the chariac- teristics. formerly ascribed to U. granifera and the other group is termed “multi- spina” to indicate a similar relationship with the former species U, mudtispina, “Granifera” Group This group is polymoerphic and falls into three more or less couspicuuns forms. Ifistorically these forms have been treated as three species, namely, Uniphora granifera, U. sinusolda and U. fungifera. A given population of the eroup may contain representatives of all three forms but usually there is a preponderance of one form with a few representatives of one or both of the other forms. Where two of the forms. are represented in a population there are usually numbers of specimens of intermediate form, and on ocvasions speci- mens of one form have heen seen with a regenerated arm of another form, The form first deseribed as U. globifera is most commonly enconntcred anil is characterised by swollen and capitate carinal spines. A typical specimen is figured by Clark (1923, p, 404). The carinal series of plates carries np te 24 enlarged capitate spines, frequently placed singly at the apex of each zizzug formed by the outline of the plates. Interspersed between these enlarged spines are a variable number of much smaller capitate spines either singly or in small “tiups. The dorso-lateral plates have numerous spines of varying size usually “ REVISION OF “VHE STARFISH GENUS UAIOPHOHA a about as large as the smaller spines of the carinal series and these ave often irregularly placed. Here and there the spines of the superd-marginal plates are as large as thase along the carinal lines. Specimens from New South Wales which have this form are remarkable [or the size of their enlarged capitate carinal spines which are larger than thase seen in South Australian specimens. They also differ from the latter in possessing a larger number of smaller capitate spinelets dorsally and on the intermaryinal plates, The actino-lateral spines are quite variable; sometimes they are typical but at others more like those in “multispina” group. The specimens were all taken apparently on rocky bottoms and are cither from Long Reef, Collaroy, or Sydney Iarbour. A second form which was formerly designated U. sinusvida possesses a markedly zigzag carinal scries of plates bearing spines which are capitate but not nearly as conspicuous as the globose spines of the previous form, In some specimens it is dificult to detect a carinal series of plates at all and dorsally the spines appear to form the margins of a series of discontinuous polygonal papular areas. The figures of Clurk (1928, p, 412) and Stach (1938, p. 332) are quite typical of this kind of specimen. A number of such specimens haye been examined from Port Phillip Bay, Victoria and have less conspicuous dorsal spines than the South Anstralian specimens while several rather resemble those in “multispina” group, Specimens from the Austvalian Museum taken in Tasmania are placed in this group with sume hesitation, The arrangement of the spines awd the general appearance of these specimens are typically sinuyoida-like but the dorsal spines of some speci- mens, particularly those taken in 1)’Entrecasteanx Channel, Tasmania, are fine and pointed as is sometimes the case in “multispina” group, ly same localities there commonly appears a third form which is remurkable for the proliferation of spines dorsally. Instead of enlarged capitate spines aceurring, singly or in groups of two or three along the carinal series, there are groups of 10 to 20 small and closely appressed, capilate spines, often with the central cue much larger than the surrounding ones. These groups of spines are usually contiguous. and together appear to form am almost Hat dorsal surface. In othey respects the specimens do not differ significantly from those previously mentioned, These specimens conform to such » striking degree with Perrier’s deseription of Asterias fungifera (1875) that nv doubt this form, or somethrig very close to it, is that which is so described. Jt is to be nated also that Fisher (1926, p, 198) who compared the type specimens of Asterias fungifera and Asterias yinusoida thought the Jatter to he a “slenderer and stouter spined exantple of the same species”, 1! have only ohserved “granitera” populations an reels and rocky bottor, Stach (1938) however, reports several specimens from Posidonia beds in the Sir Joseph Banks Croup, Spencer Gulf, South Australia, but it is possible that these specimens came froma granitic reef nearby, Some of the specimens from Port Phillip Bay, Victoria, and also from D’Entrecasteaux Channel, Tasmania, alsu appear to occur on sandy bottom but as indicated previously these specimens are atypical. The form ranges fram lnw water down to about 20 m in depth but is found in greatest wbundance in depths between 5 and 10 m; in fayourable con- ditions where the wave action is moderate but not severe two or three specimens may oceur to the square metre, The favourite dict is a small tunicate which abounds on shallow reefs. Very frequently the seastar is found with its rays encircling the Uinivate and 6 &. A. SHEPHERD protuding its everted stomach into the animal's oral aperture. low it achieves this is not known. Its tube feet clearly could not exert the same foree on the tunicate as they could upon the shells of a bivalve. In the latter case the force exerted by each tube-foot on the shell would be cumulative; whereas in the former case, the aperture appears to be controlled by musculature immediately surponn alg it and the seastar could bring very few of its tube feet to bear upon these muscles, and hence would exert little farce on them, It is possible that the animal discharges a stomach enzyme or toxin which paralyses or narcotizes its prey so enabling it to vain entry, Certainly when the starfish is removed the tunicate is unusually sluggish in closing its aperture. To a lesser degree the seastar has been observed toa feed on reef-dwelling univalves and bivalves and on various encrusting ascidians, Specimens taken from shallow water are usually brick-red in colour but occasional ones are seen in which the papular areas are a dark blue. Specimens from deeper water between 15 and 20 m are usually more brightly coloured, ranging from bright red to orange. Orally specimens are red to orange. “Multispina” Group Specimens from thesé populations are distinguishable from those of “yrani- fera” Populations by the comparatively regular arrangement of the carinal and dorse-lateral plates and the spines on them. The carinal series carries from | to 4 rather small capitate spines to each plate and the series forms a straight linc medially on the dorsal surface with a tendency in some specimens to zigzag distally. On each side of this series and parallel with it are two regular dorso- lateral series carrying similar but smaller spines. In some populations the latter series bear spines only sporadically and the sipero-marginal series bear them ieguiarly, The spines themselves vary considerably from being capitate or bluntly pointed to being fine and sharp. Where “multispina” and “granifera” populations intergrade, as is not uncommonly the case, the characteristic dif- ferences tend to become obscured, The forms described by Clark as U. obesa and U, uniserivlis are beth well within the range of variability shown by “multispina” races and there can be no doubt as to their synonymy. The figures of Clark (1928, pp. 407, 410 and 412) indicate the remarkable diversity existing in the appearance of specimens. “Multispina” populations are abundant on the angiosperm beds of Cyma- docie antarctica Endlicher, Posidonia australis Hooker, and Zostera mueller Irmisch characteristic of the coastal margins of the South Australian Gulfs par- ticularly in the less exposed regions. ‘Their principal diet is bivalves and uni- valves which are abundant in the sand among the fibrous angiosperm roots. Mast specimens are dark red in colour but some are lighter, tending to become fawn, This form is commonly found down to 10 tm but occasionally specimens hay¥e been taken from deeper water and there is an unusual record of 3 specimens from 30 m taken by Mr. BR. C. Sprigy off Carickalinga Head in St. Vincent Gulf, South Australia, UNIOPHORA NUDA Perrier Asterias nuda Perrier, 1875, p. 335. Uniephore gymnonota Clark, 1925, p. 405. Material Examined. S.A.M, (80 specimens) KI78-9, K522, K599, K636, K642, KA52, Perrier’s type was stated to have come from “Port Lincoln, (detroit de Torres)". A series of specimens taken off Cape Donnington just outside Pt, A REVISION OF TIT STARFISH GENUS TINIOPIORA 7 Lincoln Harbour. South Australia conforms exactly with Perrier’s description and vunfirms Clark's view (1928, p, 417) of the place of origin of the specivs, The species is distinguishable from the “multispina” group by the very small number and size of its spines, and many specimens, save for the adambulacrals, are quite denuded of spines. [t is without exception, covered by a thick skin which tends to conceal the existing spines but does not obscure the praminent plates forming the carinal, dorso-lateral and marginal series. Examination of Clark’s type of U. gymnonota shows no significant ditterences between it and the specimens from Cape Donnington. Clark purported lo clis- tinguish (7. nude from U. gymnonota on the ground that the lateer did not have any pedicellariae corresponding to those observed by Perrier in the ambulacral groove and which he described in these terms “droit, court, large, en forme de triangle”, ,... However there are similar pedicellariac in Clark's type and in many other specimens from the north coast of Kangaroo Island and Spencer Gulf, South Austialia and it is concluded that Clark’s differentiation was in error. ‘This species has a greater depth range and diversity of habitat than LU’ grani- fert. The specimens from Cape Donnington were taken in 14 m trom thick beds of the hammer oyster Malleus meridianus Cotton. Several specimens were observed feeding on the oyster and this no doubt constituted their principal diet. In Nepean Bay near Kingscote, South Australia in 5-8 m there ure Marge colonies inhabiting the beds of the angiosperms Zostera sp, and Halophila ovalis (KR. Brown) Hooker, There it feeds on bivalves and on small tunicates which find a ready attachment to debris and rocks scattered on the bottom, It also occurs in deeper water off the north coast of Kangaroo [sland and in Backstairs Passage, South Australia, in 20 to 40 m on rocky or shellyrit bottoms, I), colour the spocies varies from reddish brown to fawn, the specimens from deeper water being invariably the lighter in colour. UNIOPHORA DYSCRITA Clark Univphora dysorita Clark. 1923, p, 244. A total of 15 specimens have been examined from the private collection of Dr. E. FP. Hodgkin all of which are referred to this species. There are 5 jnvenile specimens numbered U.D. 9-15 from deep water (120 to 200 m) and the remain- er are from shallow water. The specimens show considerable diversity anc one or (wo approach U, granifera (sinusoida form) in appearance; but in the main the species is quite distinctive, In four specimens numbered U.D, 1-4 froin Rockingham and Woodman’s Point near Fremantle, Western Australia, the spines on the carinal series are stout at the base and conieally pointed. These spines are often 2-3 mn long and 0:5-0°7 mm in diameter at their base, and number from 25-30 along the earinal line. There are also 20 or more much smaller spinelets along the same line. Elsewhere on the dorsal surface there are many fine spinelets up ty 0-35 mm long on the plates. One of these (U.D, 1) is shown in Plate 1, These specimens conform to Clark's deseription execpt that the carinal series is mare evident than appears to be so in the type. In four other specimens numbered U.D, 5-8 from Woodman’'s Point, Cockburn Sound and Carnac Island, Wester Australia, the dorsal spines on the carinal series and elsewhere are predominantly capitate and globose, A few of the spines are rather bluntly pointed, Apparently there is considerable variation in colour in the species. The colour dorsally ranges from dark red to fawn and in some specimens the papular areas ave a ereenish-brown. Below the colour is pranze. Most of the specimens were taken on rocky hottom but several are reported from muddy or sandy bottom. 5 5. A. SHEPHERD ‘Two specimens uumbered U.D. 14 and 15 present some difficulty by reason of their extraordinary spinulation, One of these, U.D. 14 is shown in Plate 1. They were both taken at the Naval Base, Cockburn Sound, Western Australia. Thwse specimens differ from those described earlier in having few stout spines but many more fine bluntly pointed spinclets often about 1 mm long and 0-3-0-6 mm in diameter, some of which are slightly flattened at their distal ends aml somewhat chisel-like. The carinal plates are very sinuous and obscure and with the dorso-lateral plates enclose large polygonal arcas devoid of spines. The spines on the carinal plates number from 80-100 on each arm, The marginal spines and actino-lateral spies are not as flattened or as chisel-like as the specimens described above. A few of the actino-lateral spines are divided inta fvo or three short branches but this is nota distinctive feature of the species us Clark asserts (1928, p. 417). It is not considered that the differences in these two specimens are of sufficient magnitude to justify separation and in view of the cliversity shown by the other specimen, they are believed to represent at most a variant form. They were reported to have been taken at 3m depth on bite from which it may be inferred that their habitat is a sandy or muddy suittorn.. The juvenile specimens, the largest of which has R/r = 22/4 mm were all dredged off Rottnest Island, Lancelin Island or Cape Naturaliste, Western Aus- tralia, It is noteworthy that in one specimen (U.D. 9) the innermost series of actinal plates nearest the adambulacral armature carries 3 spines on each plate. In the other 4 specimens there are only 2 t each such plate. Otherwise there are no notable differences other than those une would expect Lrom juvenile specimens, The absence of adult specimens from deep water off Western Australia as for South Australia lends credence to the view that this is a shallow water species and that the specimens from deep water are expatriates which having settled there as larvae, do not find favourable conditions for survival to maturity. The adnit specimens show considerable variation in the character o! the spines but there is no doubt that they are conspecific. Having regard tu their diversity which at one extreme approaches U. granifera “granifera” group there exists some doubt whether this form should be given specific status. Taken individually, none of the characters nsed by Clark (1923, p. 244 and p. 246) in eteating the species ure reliable and each of them iy shown at least in some degree by one or other of the South Australian forms. On the other hand the overall appearance of the majority of specimens is unlike that of any seen from South Australia or elsewhere. The occurrence in some specimens of larger and relutively long pointed spines on the rays and disk is distitetive and the clusters of fine and pointed spinclets on the carinal and dorsu-lateral plates in uthers are umiygne in the genus. On the whole the species is considered valid. An endemic distribution in the temperate waters of the lower west coast of Western Austral is indicated from the preserit material, No specitneny have been cecorded from the Great Australian Bight, From the brief wotes accom- panying the specimens the species shows some variability in habitat and may occ on rock, sand, or muddy bottom in depths ta 20 m. Distribution in the South Australian Gulfs Uniophora granifera is seldom found on exposed coastlines; it prefers pro- tected waters and occurs predominantly along the coustal fringe of gulfs and hays dows to wbont 15m, Occasionally juveniles with R np ta 10mm have heen taken in deeper water of 40-50 m but this environment is apparently unsuitable as larger specimens have never been taken al such depth despite extensive collecting and dredging, A REVISION OF THE STARFISH CENUS UNIOPHORA 9 Uniophora nuda is found in the protected parts of more open waters but so far has not been taken in St, Vincent Gulf. Figure 1 indicates the distribution of the species in the Gulfs, Populations of “granifera” group occur in abundance on the reefs of the eastern coastal margin of St. Vincent Gulf southwards from Glenelg. These reefs are not continuous and colonies of the seastar are geographically separated by LEGEND UNIOPHORA GRANIFERA "granitera’ group * Common type. globifera sinusoida fungifera >7~mw "multispina’ group rs UNIOPHORA NUDA ao qx O/Pt. Hughes Spencer Gulf Sir Sa Joseph Banks Vincent Kangaroo = Island Fig, 1. The distribution of the species in Spencer and St. Vincent Gulfs showing the occurrence of the dominant forms. 1) S. A, SHEPITERD stretches of sandy bottom, ‘There are not many representatives of the “sinusoida® form here; the “fungifera” form is dominant at Glenelg, in Holdfast Bay and Halletts Cove but further south the “globifera” form is more common. The population samples are not sufficient to assess accurately the proportions in whieh the three forms occur. It is noted however that the specimens of the dominant form from any one locality are fairly homogeneous and that there are minor hut tecngnizable morphological differences when a population sample from one place is cumpared with onc from an adjacent colony, The inter-population differences do not appear to follow any pattert of variation and no clines have heen detected. The distribution wf populations of “muiltispinw” group corresponds with the extent of the beds of Posidonia and Cymaducia on the coastal margins of St. Vineent Gulf, Tt is found along much of the westcrn margin of the Gulf but on the eastern margin it is common only to the north af Holdfast Bay; specimens are seldom taken to the south. Some intergradation between U. nuda and “mullispina” forms of U. granifera has been observed on the north coast of Kangaroo Island near Cape Marsden and about Eastern Cove (Clark, 1928, p. 411) but its extent is not known, ‘Ihe evidence does not suggest that it is at all frequent er widespread. No inter- gradation has heen observed anywhere between U, nuda wud any forms of “granifera” group even where the two species are found in the same geographical arva, although in different ecological niches. The distribution in Spencer Gulf is still sketchily known. Despite extensive diving no specimens have been taken around the Gambier Islands, the Neptunes, or Thistle Island but they seem to be common enough in the quieter waters further north, In the Sir Joseph Banks Group it appears that Uniophara granifera occurs commonly both with the “granifera” and “multispina” kind of spinulation (Stach, 1938, p. 329), and WU. nuda has also been dredyed in deeper water off Spilsby Island. No specimens of U. vranifera have been taken west of Spencer Culf bul as little collecting has been done an the west coast of Eyre Peninsula its discovery in suitable localities along that coast would not be surprising. As vet only UU, granifera “granifera” group has heen taken off the coasts of New South Wales, Victoria and Tasmania, Differences in Uniophora Colonies The forms differ basically from each other in the number and nature of the spines, notably those dorsally on the rays, Spine counts therefore seem to offer the best method of presenting the differences statistically despite the irregularity ot arrangement of the spines. The spines along the plates of the carinal series from the base of the ray to the tip were counted and the results are expressed in spines per cm lo compensate for size variations. Because of the unpredictable local occurrence of colonies there ure not many collections which ure sufficiently comprehensive to give reliable statistical definitions of the several populations. For the sake of completeness even smull series are analysed so that the data in the following table and Figure 2 comprise the figures for all series of Ave or more specimens. The purpose of the statistical record is firstly tu show the close relationships between the several forms and sceandly to determine the yariability within a population and so complete the empirical description of the material, Analysis of the larger pupulation samples = thal the spine-counts follow a normal distribution curve and 2 comparison of population samples indicates 2 gradual A REVISION OF THE STARFISH GENUS UNIOPHORA ll TABLE 1 Geographical occurrence of Uniophora cclonics sampled. Mean size Form Locality + Depth (in metres) No. of | of sample specimens | (R) in cm. (1) U. nuda Cape Donnington—lim. 5 8-3 (2) a Sir Joseph Banks Group—30m. 6 4:3 (3) " Kingscote—Nepean Bay—5m. 57 4-5 (4) 5 . ” Backstairs Passage—30m, 5 8-8 5) U. granifera (“multispina” group) Edithburgh—im. 28 4:5 (6) se Outer Harbour—5m. 10 47 (7) 3 Semaphore—10m. 17 10-2 (8) 33 Pt. Turton—7m. 14 6°8 (9) U. grantfera (‘‘granifera,”’ group) Glenelg Breakwatcr—5m. 7 7-5 (10) 9% Halletts Cove—7m. 7 6-0 (11) ba Pt. Noarlunga—7m. 34 4-6 (12) + Pt. Willunga—8m. 21 4-0 (13} ” Snapper Rock off Cape Donnington—im. 6 6-4 (14) 3 Pt. Hughes—5m. 6 4-6 U. granifera (multispina gp.d oO 2 4 6 8 10 12 14 16 16 20 22 a4 26 Spines per cm. Fig. 2. The graph is a summary of Uniophora carinal spine-counts. In cach case the single line represents the range, the rectangle in block one Standard Deviation on either side of the mean, and the vertical line the mean. 1z S. A. SHEPHERD increase in spine counts from U, nuda to U, granifera. Despite the slight overlap between U. nuda and U. granifera it is considered that the combination of the differences between the two is sufficient to justify specific differentiation. Table 2 summarises the characters which are useful in discriminating between the forms. It is evident at once, that there are no diagnostic characters which can he relied upon as absolute determinants; the characters are all variable and can be used only in conjunction with others to distinguish the forms. Even then the degree of overlap may in some cases cause difficulty. Another problem is that Table 2 is based primarily on the South Australian material so that when the specimens from other States are examined it is found that the character com- binations are not quite the sare and intermediate degrees of relationship exist. Other characters are referred to in the literature which have been used from time to time as determinants, for example the presence of pedicellariae and pebhled areas in the intermarginal plates, but these are now ignored as they have not been found to be of any significance. TABLE 2 Major differences between Uniprora species; | | UU gpremifera di, dyserita. CT. render | | “eranitera’* op. “multispina” op. Carinal series. Ziy-war Straight, some- | Zig-cag sometimes | Straight, some- ic. central series times 7ig-2ag irregular times: Zig-zag of plates dor- distally i distally anally on rays ! | Carinal spines | Capitate or [neonspiewous Capitate, singly or | Capitate some- Frequency | blantly pointed | fine sane In ganips tines blantly (per em.) 9-15 pointer $-25 pointed singly or | 0-3 Mm groups: 4-13 Dorso-latoral Jrregularly Absent Trregularly Form Lor 2 spires arranged, often arranged, or repular series each in groups, | arranged to form | side of carinals Smaller and pelygenoal papular finer than | debs varinals Supero- Capitate ur | Rare or absent | Capitate Slightly capitate. marginal spines | fatrened Often in groups (orminally j of 2-3 Actino-lateral Fresent | Rare br absont | Present Present spines | The considerable increase in spine numbers per cm along the carinal line in “granifera” group as shown in Figure 2 is due partly to the exaggerated development of the characteristic zigzag carinal plates and partly to the increase in spines on the plates. Of the series from St. Vincent Gulf it is noteworthy that the specimens from Port Noarlunga sonthwards are generally stnaller and have fewer and more capitate spines than those further north. The series from Halletts Cove and Glenelsy Breakwater shows some intergradation with “multi- spina” group, A REVISION OF THE STARFISH GENUS UNIOPHORA 1 Table 1 also shows the variation in the mean size of adult specimens from the populations sampled, The reason for the diversity in size is not known. It does not appear to be related directly to age as there is no difference in maximum size of specimens collected in one locality at different times of the year and in different years. The seastars are sedentary in habit and it is probable that the Ope and quantity of food available is an important factor. It is noted for example that the specimens of U. nuda from the rich hammer oyster beds off Cape Don- nington are considerably larger than the specimens of the samie species from the quiet waters of Nepean Bay, Kangaroo Island. U. dyscrita UU. nudo U. granifera granifera gp. multispina gp. Fig. 8, Lateral view of prominent carinal spines on rays. CONCLUSION The greater amount of material now available reveals that in the South Australian Gulfs there are two species of Uniophora, One of these U. eranifera contains two component groups of populations, each of which has some pecu- liarities of behaviour, distribution and morphology. The populations called “granifera” are found on rocky bottom along coasts of moderate wave action whereas “multispina” populations oceur on sandy bottom and in calmer water, Specimens having characteristics of both “granifera” and “multispina” have been taken on bottoms of.a mixed character, The interesting problem is raised whether the differences in morphology are genetically or ecologically determined, but this cannot be resolved without experimental work, The existing evidence suggests that U. uucde is a distinct species but this view may require reconsideration should intermediate populations be discoyered, Whatever the case is, the two species have succeeded in occupying a varicty of habitats and so been able to fourish in the Gulf regions in an abnndance, which is without rival among the asteroids, There is nu evidence of the species achieving the samc success elsewhere on the Australian coast. Uniophora dyscrita must be retained as a valid species for the present. There is no record of the species from the Great Australian Bight, It is quite likely that the exposed coasts and long stretches of sandy shore of that region constitute a nalural barrier sufficient to prevent gene flow between the South Australian and Western Australian species. (See Womersley and Edinonds, 1958, p. 221, and Jennings, 1963, p. 45.) In view of the known ecological requirements of the specics much of this coast would not provide favourable conditions for survival. The differences already visible indicate genctic variation and hence the evolution af separate species. i4 5S. A, SHEPHERD ACKNOWLEDGMENTS I am grateful to members of the Underwater Research Group who assisted me in collecting specimens from time to time, to Mr. R. C. Sprigg who generously made available the research vessel Saori for numerous field excursions, Dr. E. P. Hodgkin and Mr, I. M. Thomas read the manuscript and offered valued advice. Finally I am grateful to Miss E. Pope and Miss J. H. McPherson who sent to me the collections of the Australian Museum and the National Museum of Victoria respectively, and to Dr, E, P. Hodgkin who sent his personal collection. REFERENCES Crark, H. L., 1923. Journ. Linn. Soc. (Lond.) Zool., 35, pp. 244-246. Crarx, H. L., 1928. Rec. §.A. Museum, Vol. Ill, pp. 403-417. Crark, H. L., 1946. The Echinoderm Fauna of Australia, pp, 158-161. Corton, B. C., and Goprrey, F. K., 1942. Rec, S.A, Museum, Vol. VII, p. 193. Fisuer, W. K,, 1923. Ann. Mag. Nat. Hist. Ser., 9 Vol., 12, p. 597, Fisner, W. K., 1926. Ann, Mag. Nat. Hist. Ser., 9 Vol., 18, p. 198. Gray, J. E., 1840. Ann. Mag. Nat. Hist., 6, p. 288. Jennincs, J. N., 1963. Trans. Roy. Soc. S.Aust., 87, pp, 41-62. Lamarck, J. B. P., 1816. Anim, s. Vert., ii, p. 560. Perrier, E., 1875. Arch. Zool. Exp., iv, pp. 335-345. Sracu, L. W., 1938. Proc. Roy. Soc. Victoria, 50 (N.S.), pp. 329-332, Womers.ey, H, B, S., and Epmonps, S. J., 1958. Aust. J. Mar. Freshw. Res., 9, pp. 217-260. S, A. SHEPHERD (2) Dorsal view of specimen of Uni- ophora dyscrita numbered U.D, 14 R = 55 mm. Collected Stn. 61 Naval Base, Cockburn Sound, W.A., among Pinna 10 ft. deep by Marine Gp W.A. Naturalists 10/4/60. (Photo E, P. Hodgkin. ) PLATE 1 Pirate | (1) Dorsal view of specimen of Uni- ophora dyscrita numbered U.D. 1 R = 54 mm. Collected R. Howlett on muddy. sandy bottom, Depth 25 ft. Palm Beach Jetty 3/3/60. Rockingham, W.A. (Photo E. P. Hodgkin. ) PRELIMINARY PALAEOMAGNETIC RESULTS FROM THE ADELAIDE SYSTEM AND CAMBRIAN OF SOUTH AUSTRALIA BY J. C. BRIDEN* Summary The natural remanent magnetization (NRM) of some sediments from the Adelaide System and Cambrian of South Australia is reported. In some of the formations, which were studied, notably the Cambrian sequence on Kangaroo Island and the Billy Creek Formation in the Flinders Ranges, directions were consistent with an NRM of Mesozoic or early Tertiary age. It is suggested the NRM in the Cambrian of Kangaroo Island was acquired during a period of slightly elevated rock temperatures and stabilised by cooling to near present day temperatures. In the Marinoan and Cambrian sediments of the Flinders Ranges, the secondary NRM may be due to a surface effect, which resulted in increased grain size of the magnetic fraction. PRELIMINARY PALAEOMAGNETIC RESULTS FROM THE ADELAIDE SYSTEM AND CAMBRIAN OF SOUTH AUSTRALIA by J. C. Bruwen® (Communicated by B. Daily) [Read 12 September 1967] SUMMARY The natural remanent magnetization {NAM) of some sediments from the Adeluide System and Cambrian of South Australia is reported. In some of the formations which were studied. notably the Cambrian sequence on Kangaroo Island and the Billy Creek Formation in the Flinders Ranges, directions were consistent with an NRM of Mesozoic or early Tertiary age. It is suggosted the NRM in the Cambrian of Kangaroo Island «yas acquired during a peried of slightly elevated rock temperatures and stabilised by cooling to neur present day temperatures, In the Marinoan and Cambrian sediments of the Mlinders Ranges, the secondary NRM may be due to a surface effect. which restted in increased train sizo of the magnetic fraction. INTRODUCTION The natural remanent magnetization (NRM) of some sediments from the Adelaide System and Cambrian of South Australia is reported. In some of the formalions which were sludied, notably the Cambrian sequence on Kangaroo Island, and the Billy Creck Formation in the Flinders Ranges, directions were consistent with an NRM of Mesozoic or early Tertiary age. It is suggested that the NRM in the Cambrian of Kangaroo Island was acquired during a period of slightly elevated rock temperatures, possibly connected with early Tertiary igneous activity, and was stabilised by cooling to near present-day temperatures, In the Marinoan and Cambrian sediments of the Flinders Ranges, the secoudary NRM may he duc to a surface effect which resulted in increased grainsize of the magnetic fraction. Results from other formations, in which NRM is weak in intensity and scattercd in direction, are reported in brief, because consistency of direction was evident in isolated portions of the sequences; although their significance cannot he assessed with such limited data, they may be important in the light of further work which is in progress in the State, SURFACE COLLECTIONS The procedure in the field was to collect from | to 6 samples from each site, Up to 4 specimens were cut from each sample, in which direction and intensity of NRM were measured in the laboratory using an astatie magnetometer, Direction is quoted in terms of declination (D) and inclination (1, regarded as positive downward from the horizontal) relative tu two sets of axes: (1) relative Trans. Roy. Soc. S. Aust, (1967), Vol, 91. 18 J. C. BRIDEN the folding of the beds on the assumption that each phase of Folding has been purely rotational (“corrected”), Directions are analysed by the method of Fisher (1953). Stability of NRM is assessed (1) by comparison of the uncorrected roean direction with the present field, (2) by comparing the estimate of pre- cision (k) of grouping of directions before and after correction for folding (if corrected directions are more precise, then NRM may be presumed to be older than the folding), and (3) by interpreting the thermal demagnetization charac- teristics of specimens in the laboratory. VABLE 1 Sammary of palacomagnetic measurements in the Adelaide Syste. Series Formation Locality of | Comments Samples | Pound Quartzito Railway cuttings 14 | Weak and widely equivalent through South scattered NAM with ABC Range Hummocks Ranges 3 negative inclinationa Quartzite equivalent relative tu the Upper Glacial 5 present horizontal Sequence’ MARINOAN ——_—- - — Sandatones and Hallett Cove 35 3 sites near middle of arkoses (14 sites) aAquence consistent. Correeled mean after cleaning at 400°C D191, i = +34", k-5 Sandstones Reynetla Quarries 10 4 sites consistent, (5 sites) initial corrected mean B=? T= 411", | hk = 35 Brighton Limestone | Heynella Quarries 3 Weak and widely Happy Valley z scattered NRM with ve THN Satie , -H: & neyative (upward) stumiran | Papley Hill Glstos, | Tepler Hilt j | inclinations rlative to the prosont Sturt Tillite Loeckler Road Quarry 1 horizontal TORRENSIAN | Stonyfcll Quartzite Greenhill Quarries 2 Nob detectablo NRM Adelaide System: Results. are summarised in Table 1, portions. of the sections were NRMs consistent. Only in isolated In a small face at the east end of Reynella Quarries, 4 sites in about 5 m of the lowest Marinoun sandstones have a mean. direction D = 7°, T = 11° after correction with k = 32. A fifth site, only a few metres away in another face, was not in agreement, The coastal section north and south of Hallett Cove was sampled thoroughly, but only at three sites near the northern end of the cove were directions consistent. The increase in precision after thermal cleaning at 400°C and correction for folding (D = 191°, f= +39°, k= 5) compared with the uncorrected mean direction of total NRM (k = 1) suggests that a relic of stable NRM which is older than the folding may have been preserved, PALAEOMAGNETIC RESULTS. ADELAIDE SYSTEM AND CAMBRIAN OF S.A. 19 Cambrian: Twenty-three sites were sampled in the Cambrian sequence on the north coast of Kangaroo Island (Duily, 1957). The distinction between the Emu Bay Shale and the equivalent of the White Point Conglomerate near Emu Ray depends on the recognition of a Redlichia horizon which also occurs in the Vlinders Ranges, not far below the base of the Billy Creek Formation. Because the author did not find the Redlichia horizon, 3 sites included with the White Point Conglomerate might strictly belong to. the overlying formation, In_ this context, the distinction is not important because the palacomagnetism vf the Kangaroo Island Group does not vary from formation to formation. All site mean directions except one (near Point Marsden) were significant with 95 per cent probability. Before correction for folding all inclinations were steep, and at all except three sites were directed upwards (in the same sense as the present field, and referred to as “normal’). At 2 sites, however, in the Emu Bay Shale in its eastern section, steep downward inclinations were observed (called “reversed"), and at a third site, nearby, both normal and reversed samples were found. The mean direction after adjustment for the reversed directions, is slightly steeper than the present field (Figure 1) and is more preciscly defined than when correction is made for folding (Table 2). TABLE 2 Analysis of NRM directions in Cambrian stdiments fon Kangaroo Taland. | | t pale 8B gn | 7 & a uncorrected 23 3AT7 75 UT-5 6 agW Gan carrectent 22 289 =73 7-9 IL —_—_ — Bis the wumber Hf significant sites (05 per cent prohahility): | is the semi-angle of the 95 per cent cone of confidence; (4, A) ia the palacumugnatie pole, Initial directions of NRM in the Billy Creek Formation are steeply upwards (Figure 2.) Upon thermal demagnetization, direction changes at most sites (at the north of Lake Torrens, in the Copley-Mount Scott district and between Wirealpa and Martin Well) are irregular, but tend in general towards a southwesterly direction, At Balcoracana Creck (Figure 3) directions change systematically until, after heating to 500°C and correcting for folding, the mean is horizontal and southwesterly, which is in agreement with that found in other Cambrian rocks in Australia by Irving and Green (1958). This suggests that, although the dominant component of NRM is secondary, in at Teast part of the formation a relic of more stable NRM is preserved, and it is possible that this is of Cambrian age. BORECORE SAMPLES Measurements of NRM of borecore samples are limited hy lack of azimuthal orientation: inclination but not declination can be measured. Nevertheless such measurements are useful for estimating inclination in areas where rocks at the surface have been remagnetized, since it is possible that primary magnetization might have been preserved at depth. Briden and Ward (1966) devised a method of estimating mean inclination and Fisherian precision from a collection of borecores, which has heen nsed in aualvsing results from South Australia and the Northern ‘Territory (Table 3), A vertical mark was made on each sample, and twa discs were cut from each, so that consistency of NRM direction within a sample could be tested; when differences cxcecded 30°, results were discarded. 20 J. C. BRIDEN TABLE 3 Analysis of inclination of NRM in borecores, B is the number of consistent samples; A, is the palacolatitude. —e—,;,CaKX—————— eee Borehole and location Age B I k: Ay Upper BMR/GRG No. 7 (136-1°E, 22-3°S) Cambrian 8 —15] 85 9 Upper BMR/GRG No. 5 (135-2°E, 21-+3°8) Cambrian 8 —42] 21 24 Lower Minlaton No, 1 (137-7°E, 34-8°S) Cambrian 13 —55 5 35 1340-1750 ft. | Upper 21 | —85| 25 | 80 Wilkatana No. 1 (138-0°E, 32-2°S) 1740-1845 ft, Marinoan 5 —4] 20 24 Lower 23 —36 6 20 Clarence River No, 1 (136-5°E, 31-1°S8) Marinoan eee CO ee Fig. 1. Site mean directions and thermal demagnetization of NRM in Cambrian sediments from Kangaroo Island. (a) Stereographic projections of uncorrected (left) and correc- ted {right) directions. At one site where polarity is mixed, the means of both normal and reversely magnetized samples are shown, and are joined by the broken line. The mean NRM direction is denoted by a star, with the circle of 95 per cent confidence around it, The directions of the present and axial dipole fields are denoted by P and D, (b) Thermal de- magnetization curves of samples KI5, KI19, KI57;_ intensities (m) are normalized by divid- ing by the total NRM intensity (M,). Open symbols denote upward and solid symbols downward inclinations. PALAEOMAGNETIC RESULTS, ADELAIDE SYSTEM AND CAMBRIAN OF 5.A. 21 In the Wilkatana No. 1 bore, there is an abrupt change in inclination at about 1,750 ft, so the groups above and below that level have been analysed separately. In Clarence River No. 1, inclinations are in two groups (Figure 4) with moderate positive and negative inclinations respectively. This suggests not merely a large scatter of directions, but rather that some of the cores have been marked the wrong way up. (An alternative explanation in terms of reversals of NRM is not considered plausible because there is no discernible stratigraphic pattern in the distribution of the two groups.) From the mean inclinations, the palacolatitude (A,) of each borehole is calculated on the assumption of a geocentric dipole field from the relation tani, = 2tana, and the results are compared on a map of palacomagnetic poles (Figure 5) with results from other Australian rocks, 1.2- 1.0 M 57 M ° 08 19 0.6 0.4 5 0.2 100 200 300 400 500 Temp, °C DISCUSSION OF THE ORIGIN OF SECONDARY NRMs Thermal demagnetization curves of the Billy Creek Formation and Emu Bay Shale (Figures 1 and 3) are similar to those of ancient secondary magnetiza- tions in the Bloomsburg red beds ({rving and Opdyke, 1965) and the Old Red Sandstone (Chamalaun, 1964). The hypothesis which is proposed for their origin is that they were acquired in the Earth’s field over a long period (say 10% years) after which the effective relaxation times of the magnetic grains were increased. This has the effect of “freczing-in” the NRM so that it may be stable over millions of years at normal temperatures, Increase in relaxation time may bo hw J. C. BRIDEN uncorrected Fig. 2. Thermal cleaning of Billy Creek Formation, Initial directions. and directions after heating to 200°C and 400°C, both before and after correction for folding, Sites are NEW (northeast of Wirrealpa); YC (Yarra Wurta Creek); LT (edge of Lake Torrens); Cl and C3 (Copley-Mount Scott); Bl and B2 (Balcoracana Creek); TMC (Ten Mile Creek) and E (Eregunda Creek). Open symbols denote upward and solid symbols downward inclinations. PALAEOMAGNETIC RESULTS, ADELAIDE SYSTEM AND CAMBRIAN OF S.A, 23 N Vig. 3, Thermal demagnetiza- tion of two specimens from Balcoracana Creek, in steps, at room temperature, 100°C, 200°C, 300°C, 400°C, 450°C, and 500°C. The stereogram (above) shows the change of uncorrected directions (circles) with progressive demagnetiza- tion. in the sequence indicated by the arrows, and (squares) after demagnetization to 500°C and correction for the tilt of the beds, which is also shown. The demagnetization curves (below) show vectorial mean intensity of the two specimens. Open symbols denote upward M x10°%e.m.u./gm and solid symbols downward 2 inclinations. Tere) 300 500 °C be due to fall in temperature, (due to uplift during folding, or at the close of a phase of igneous activity), change in grainsize, or chemical change. For the Bloomsburg Formation and the Old Red Sandstone, fall in temperature has been proposed. On Kangaroo Island, Lower Tertiary basalts outcrop on the low hills inland from the palaeomagnetic sampling sites. It is not unreasonable to suppose that they were formerly more extensive, and that temperatures in the Cambrian rocks were elevated at the time of their extrusion. The mean direction of NRM in these rocks is similar to that found in early Tertiary rocks in Australia (Irving, 1964) and hence the NIM may be the same age as the basalts. The most reasonable interpretation is that only the reversed rocks retain their NRM from that time, and that the normal rocks have NRM of recent origin. a4 J. C. BRIDEN UP ae 4 iyplieitions My, NRM in arence River No. 1 (Woomera HORIZONTAI: bore, The inclinations in borecore samples are, for the most part, not inconsistent with NRM of comparable age to their host rocks. But in the Wilkatana No. | bore, the upper part of the sequence appears to have been remagnetized in a direction steeper than the present field. This may be due to a surface effect operative in the Mesozoic or early Tertiary (Figure 5), and it is possible that this was a regional cffect which involved the Billy Creek Formation also. There is no geological evidence of elevated temperatures in these rocks, and it may be that chemical or grainsize changes were responsible for stabilising the NRM. The acquisition of viscous magnetization (which increases with the time of application of a magnetic field) at various temperatures by specimens of Billy Creck Formation and Emu Bay Shale has been investigated in the laboratory and is described by Briden (1965). ACKNOWLEDGMENTS The work was carried out at the Australian National University, under the guidance of Dr. E. Irving, to whom I am indebted for much advice. The Director of the Geological Survey of South Australia allawed me access to unpublished maps and information, and to the collection of cores. I wish to thank officers and former officers of the survey (especially Messrs, B. P, Thomson, B. P. Webb and A. R, Crawford) for their co-operation and for guidance in the field, and Mr. R. Dalgarno, who collected the samples from Balcoracana Creek. PALAEOMAGNETIC RESULTS, ADELAIDE SYSTEM AND CAMBRIAN OF S.A, 25 Dr. B. Daily gave me the benefit of his detailed knowledge of the Kangaroo Island Group. The Director, Bureau of Mineral Resources, Geology and Geophysics, Canberra, allowed me access to the cores from the Georgina basin, Fig. 5. Loci of palaeomagnetic poles. calculated from borecore measurements, by drawing cireles of radius },, centred on. the drilling sites. 1; Clarence River. 2; Wilkatana (a = upper. b = lower). 3: Minlaton. 4: GRC5. 5: GRG7. The curve is the approximate polar wandcring curve for Australia. constructed from the A-¢roup data of Irving (1964); dashes for late Precambrian to Devonian, dots tor the Carboniferous, solid line for the Mesozoic and dot-dash for the Cainozoic. REFERENCES Brinew, J. G., 1965. Ancient secondary magnetizations in rocks. J, Geophys, Res.. 70, pp. 5205-5221, Brien, J. C.. and Warp. M, A., 1966, Analysis of magnetic inclination in borecores. Pure and Appl. Geophys., 63, pp. 133-152. CHAMALAUN, F. IT,. 1964. Origin of the secondary magnetization in the Old Red Sandstoncs of the Anglo-Welsh cnvette. J, Geophys. Res.. 69, pp. 4327-4338. Daiwy, B., 1957, The Cambrian in South Australia, Bureau of Mineral Resources, Bulletin No. 49, pp. 91-147, Carberra. Fister, KR, A.. 1953. Dispersion of a Sphere. Proc. Roy. Soe. London, A, 217, pp. 295-305, Invinc, E., 1964. Paleomagnetism, Wiley, New York. TRvING, ie and Grrex, R., 1958, Polar movement relative to Australia. Geophys. J., 1, pp. 64-72, Invinc, E., and Orpyxe. N. D., 1963. The palcomagnetisrn of the Bloomsburg red heds and its possible application to the leclonic history of the Appalachians. Geophys, J., 9. pp. 153-167. ON THE RADIOACTIVITY AND RELATED FLUORESCENT PROPERTIES OF SEDIMENTARY AUSTRALIAN ZIRCONS BY I, A. MUMME* Summary The natural radioactivity of sedimentary zircons from various placer deposits in Australia was investigated with a gamma-ray spectrometer. The radioactivity of the zircon samples was generally weak, and was shown to be due to uranium series; however, in one sample traces of thorium series were also detected. Under ultraviolet light many of the colourless to lemon brownish zircons fluoresced with a brilliant yellow colour which was attributed to the presence of uranium in solid solution. ON THE RADIOACTIVITY AND RELATED FLUORESCENT PROPERTIES OF SEDIMENTARY AUSTRALIAN ZIRCONS by I. A. Musme* [Read 11 May 1967] INTRODUCTION The natural radioactivity of sedimentary zircons from yarious placer deposits in Australia was investigated with a gamma-ray spectrometer, The radioactivity of the zircon samples was generally weak, and was shown to be duc to uranium series; however, in one sample traces of thorium series were also detected. Under ultraviolet light many of the colourless to lemon brownish zircons fluoresced. with a brilliant yellow colonr which was attributed to the presence of uranium in solid solution. OCCURRENCE AND NATURE OF RADIOACTIVITY OF ZIRCONS Zircon is widely distributed in Australia as an accessory constituent of igneous and metamorphic rocks being especially common in the granitic rocks and only oecurring sparingly in the basic rocks. It is one of the earliest minerals to crystallize out of a magma. It is apparently more abundant in plutonic than voleanic rocks. By disintegration of these rock types it is concentrated by the action of water in sedimentary deposits along with other characteristic minerals including magnetite, garnet, ilmenite, corundum. As the mineral zircon (ZrSiO,) may contain uranium and thorium as sub- stitution solid solutions up to 0-1 per cent and even higher amounts as crystal inclusions, the radioactivity of sedimentary zircons from various placer deposits in Australia was measured with a gamma-ray spectrometer, and the uranium and thorium contents estimated by calibration with standard samples. THE GAMMA-RAY SPECTROMETER The gamma-ray spectrometer offered a rapid and accurate method of analysis of the uranium and thorium contents of the zircon concentrates. The spectrometer employed an electronic means of differentiating between the energies of the various intercepted gamma-rays. Essentially the gamma-ray spectrometer consisted of a scintillation detector which converted an intercepted gamma-ray to an electrical impulse, the voltage of which was proportional to the energy of the absorbed gamma-ray. * Australian Atomic Energy Commission. Lucas Heights, Sydney, N.S.W. Trans. Roy. Sow, $. Aust. (1967), Vol. 91. 28 L A. MUMME This impulse was amplified and passed through an electronic gating circuit which allowed only pulses of a specific yoltage to actuate a scaler, Gamma-rays emitted by radioactive nuclides are of characteristic energies and the intensity of the gamma-radiation emitted by a specific nuclide is pro- portional to the quantity of the nuclide present in the sample. The gamma-ray spectrum graphs were obtained by sweeping the spectrum with a pulse height analyser, The individual samples of zircon concentrates were weighed and packed in perspex cups and counted close to the face of a thallium activated sodium iodide scintillation crystal which was shielded by a lead castle. RESULTS OF THE RADIOMETRIC INVESTIGATIONS An examination of the gamma-ray spectra for the various zircon samples investigated showed that the uranium series were in equilibrium in each case, This is to be expected as the zircons resist alteration to a marked degree, and also the age of the zircons are greater than 3-5 x 10° years—the time required for radioactive equilibrium to be set up by the contained uranium. The results of the radiometric measurements are presented in Table 1, TABLE | Uranium content Thorium content Locality | (% U,O>%) (% ThOs) 1, Rocky River, Uralla (15662) NSW. 0-101 0-01 2. Anakie, Queensland (D21364) 0-087 < 0-001 3. Strangway's Range, Central Australia (D401) {)-0021 <0-00L 4. Hanging Rock Head of Dunean’s Creek, 4 miles along Niangula Read (040400) NSW. 0-0108 <0 001 5. Inverell (D16261) N.8.W. 0-0124 iY 138° 1-2’ Murray Bridge 46°65 79-7376 35° 15-5" 1g9° 27-1’ Tailem Bend 57 979° TA25 35° 17-7" 139° 49-1' Moorlands 2-4 79-7795 35° BB+! 138° 40-4’ Coomandook 33-3 B79°7977 $a° 34-7" 139° 44-6" Kiki 85°38 179-8034 33° 42-4" 139° 50-2" Coonalpyn 66-4) \ 979-8058 35° 49-3" 139° 55: 5° Culburra 57-3 979-813) Bae 54-4! 140° 0-5’ Tintinara fad -0 N79 8152 35° 57+" 140° 10-5! Coonibo §7°3 979-8205 36° G+2’ 140° 20-2° Keith { 94-3 979-8566 86° 12-6" Taq? 40+ 1" Wirresy | 202-3 979-8183 36° 18 +8! 140? 452° Bordertown 261-7 979-8347 TABLE 2 Elevation Topographic Reduced Theoretical Bouguer Correction Correction Gravity Gravily Anomaly (milligals) (millizals) (gals) (gals) (gals) 6-45 0-19 979-7304 979-7394 —0- 0090 96-78 1:30 79-7429 979-7469 —f)- 0040 200+40 eld 979-7342 979-7441 —0- 0099 82-21 78 979-7467 O79- 7472 —() 0008 74°32 O85 Y70- 7434 970-7498 —U- 0064 2°79 OT 979-7405 979-7562 = 0187 3-34 0-05 974-7659 979-7682 —0- 0023 1:94 O04 979-7815 979-7713 0-1102 2-00 O01 979+7997 9797871 0-0126 p:12 0-01 979-8088 979-7955 00130 a7 D6 0-01 979-8098 979-8065 " 000383 3-44 O02 979° 8165 979-8 1ti4 rQo01 3-24 0-02 970-8184 979-S$223 —0 00389 h- 2d 02 OTU-S847 YTY-4381 —()- 0034 4:66 ()- 2 9$79- 8623 Y79-8388 0-0235 1214 1-2 979- $304 979-8485 —G 0184 15°70 {)+02 979- R504 979+ ROST —0- 0083 34 I, A, MUMME INTERPRETATION OF RESULTS The regional Bouguer anomaly shows a gravity low at Murray Bridge which is possibly due to the presence of the Murray Bridge granite, which has a rela- tively lower density than the intruded country rock, or due to structural effects (Mumme, 1963). A large regional gravity anomaly occurs between Murray Bridge and Coombe with a maximuin value occurring at Kiki. This is probably due to basement structures associated with igneous intrusions. Another impor- tant feature of the plan is a sharp gravity anomaly centred on the township of Keith. It is of interest to note that the gravity highs along this traverse roughly correspond to magnetic highs (O’Driscoll, 1960) f Fig, 7), and suggest that they are due to basement highs in the Archaean metasediments whereas the gravity lows are due to granitic intrusions and possibly increased depths of the surface Tertiary sediments. On the basis of an elementary concept of isostasy, the following equation holds; Bouguer Anomalies equal —gravitational attraction of the topography —constant C {Marshall and Narain, 1954). Assuming that the topography could be regarded as an infinite, level plane at the altitude of the gravity station, we lave; _ Bouguer anomalies equal —0-034 h —constant C. The observed regression curve of Bougucr anomaly against elevation h for the area investigated (Fig, 1) suggests that the following relationship holds: Bouguer anomalies equal —0:0375 h —constant C. This equation does not differ very much from the theoretical equation and this then suggests that the general area is isostatically compensated in a regional Sense. The average Bouguer anomaly obtained for this traverse was —0-00112 gals. and the average elevation approximately 456 ft, The Bouguer anomaly values were then applied in yarious equations and by graphical methods relating elevation and crustal thickness, and gravity anomaly and crustal! thickness with the following results: (1) Applying the equation relating gravity anomaly and crustal thickness used by the Russian and Chinese scismologists, namely H = 35(1 + tan h 90-0037 Ag) where H is the crustal thickness and Ag the gravity anomaly, we obtain H = 35(1 + tan h 0-0037(—1:1)) = 35(1 — 0-004L) = 35 kilometres. CRUSTAL THICKNESS MEASUREMENTS. SOUTIL EAST OF S.A, 35 (2) Applying the equation relating gravity anomaly and elevation used by the Russian and Chinese seismologists, namely II = 38 tan h(0-38 Ah — 0-18) + 38 where H is the crustal thickness and Ah is the elevation H = 38 tan h(0-38 x 0:16 — 0-18) + 35 83 tan h(—0-12) + 35 —3-96 + 38 = 34 kilometres. (3) Applying Andreev’s formula, namely H = —0-1 Ag + 30 = —(0-1)(—1-1) + 30 = 80 kilometres. (4) Applying Woollard’s graphical method relating depth of the Mohorovicic discontinuity as a function of the Bouguer anomaly, we obtain a value of the crustal thickness of 32 kilometres. (5) Applying Woollard’s graphical method relating depth of the Mohoro- vicic discontinuity as a function of elevation, we obtain a value of the crustal thickness of 32 kilometres. From these various methods an average value of 33 kilometres for the crustal thickness is obtained. Now according to Cooke (1962) the average crustal thickness for a region can be expressed in the form; H = —62-9 + 0-485 Ag + 15-65 where is the mean crustal velocity, Ag is the average Bouguer anomaly and H is the crustal thickness. Applying the values of 33 kilometres, and —1-1 milligals for HW and Ag, we obtain a value for of 6-16 kms/sec. I I REFERENCES Cooke, 1962. Geophysics, 27 (6), pp. 892-897. Dootey, J. C., end Wriuiams, L. W., 1960. Absolute gravity value in Adelaide. Aust. J. Sci, 23, p. 17. Hammer, Sicmunp, 1939, Geophysics, 4 (3), p. 184. Hritanp, C., 1964. Exploration geophysical. MARSHALL. and Nanam, 1954. Regional gravity investigations in the Eastern and Central Commonwealth. U, of Sydney, Dept. of Geology and Ceophysics. Memoir 1954/2. Mumm, I. A.. 1960. Absolute gravity determinations in Adelaide and recommendations for a new permanent gravity base station. Aust. J. Sci., 23, p. 350. Mumnrte, en 1963. Tectonic Evolution of the Mount Lofty Range Region. Aust. Jour, Sci. 26 (5). AN AEROLITE FROM COCKBURN, SOUTH AUSTRALIA BY J. E. JOHNSON* AND D. H. MCCOLLT Summary A small aerolite, found in deflated sandy country near Cockburn, on the New South Wales - South Australian border, is described and classified as an olivine-hypersthene chondrite. AN AEROLITE FROM COCKBURN, SOUTH AUSTRALIA by J. E. Jounson® axn D, H. McCatrf [Read 18 April 1967] SUMMARY A small nerolite, found in deflated sandy country near Cockburn, on the New South Wales-South Australian border, is described and classited as an olivine—hypersthene chondrite. This meteorite was found during 1952 by the senior author on an aboriginal campsite on Ophara Blocks Station (now abandoned), at a point where the Cockburn-Egebek road crosses Ophara Creek. This is just inside New South Wales (latitude 32°8°S.. longitude 141°4'E.) and is approximately six miles southeast of Cockburn, South Australia. Hence, in accordance with the estab- lished practice of naming meteorites after the nearest landmark te appear on current survey maps, it has been called the Cockburn meteorite. The localit is typical of many such aboriginal campsites, located in a region of low aad: hills which in places haye been wind eroded, leaving a residuum of australites, aboriginal artifacts, and indeterminate chaleedonic flakes probably also originat- ing from aboriginal occupation. DESCRIPTION The meteorite has the form of a bipyramid based upon what was probably a near-trapezoid shape, from which one corner has been recently broken away, probably by aboriginals, so as to remove an estimated 20 per cent of the mass of the meteorite, The length along the long diagonal of the trapezoid is 2-78 centimetres, the breadth when unbroken, across the short diagonal is estimated to have been 2:0 centimetres, and the thickness between the apices of the pyramids is 1-48 centimetres, The faces of the steeper pyramid are all slightly concave, in contradistinc- tion to those of the fatter pyramid which are convex, showing that the latter was the anterior surface and the former the posterior during oriented transit through the atmosphere. The mass of the meteorite, prior to study, was 10-13 grams and when entire it would have been little more than 12 grams, The overall specific gravity is 3-47 which, although within the range for such meteorites, is rather low indicating the weathering, that has taken place. The surface bears the weathered remains of a fusion crust, now almost entircly converted to limonite. The thickness of this crust varies from ()-1 to 1-0 millimetres, and is thickest over the convex anterior surface. Extent and depth of weathering is further indicated by scattered warty excrescences of deste Llu d ER” Scale Re! Pd Oe LR Se eS ee eer BOS ® Technical Assistanl, Mines Dept. of South Australia, } Adelaide University, Geology Department. Trans. Roy. Soc. 8. Aust. (1967), Val. 91. 58 J. E. JOHNSON ano D. HW. McCOLL limonite up to 1-3 millimetres in diameter, through the fusion crust, These seem where observable to coincide with underlying grains of troilite and/or nickel-iron. MICROSCOPIC Transmitted Light: The meteorite is composed of an aggregate of grains of olivine, orthopyroxene, plagioclase and opaques, The chondritic texture is present although much obscured by. brecciation and recrystallisation. Weather- ing has also stained the more permeable parts with limonite. Olivine: Is present as clear colourless grains to the extent of 30 to 40 per cent of the total. The composition of it estimated by X-ray diffraction measure- ment of the 130 d-spacing is 25 = 3 mol, per cent of FesSiOx. Orthopyroxene: Occurs as slightly turbid, often yellow stained grains com- prising approximately 40 per cent of the total. The value of 2V was estimated only with difficulty as between 60° and 70°, corresponding to 25 to 30 mol. per cent. FeSiOs, Plagioclase: Constitutes no more than 5 per cent of the meteorite. occur- ring as colourless, clear, untwirmed crystals, filling spaces between the other silicates, It has a very low refractive index, some orientations being less than balsam, indicating it to be in the albite-oligoclase range. Reflected Light: The opaque minerals are shown to be nickel-iron, troilite, minor iron oxides (including chromite), und veins of secondary gocthite penetrat- ing along zones of weathering. The nickel-iron shows regular graphic inter- growths of kamacite and taenite (ie. plessita), indicating a nickel content of 10-15 per cent, which is typical of this class of meteorite. From this section, which covered the entire cut face of the specimen, a micrometric estimation was made of the percentage composition, counting at 20 micron intervals. This. gave the following results. Calculated Percentage Percentage (Area) by Weight Silicates et # 89.14 §2.5 Nickel-iron w! as 3.87 83 Troilite _.. x a 5,45 7.1 Iron oxides ~ a 1.54 2.1 Total... 100.0 CONCLUSION All evidence obtained concerning the composition of this meteorite places it among the olivine-hypersthene chondrites (Classification of Mason, 1962),* the most common classification of the acrolites. That this stone may be related to the Silverton meteorite,? is a possibility, as they are similar in composition and appearance. and were found abont twenty miles apart, They differ only in degree of weathering and cohesion of consti- tuent minerals, which effects may be related and consequent upon the sixty years which intervened between their discoveries. AEROLITE FROM COCKBURN, S.A. 39 ACKNOWLEDGMENTS The authors wish to thank Dr. A. W. G. Whittle of the Adelaide University, Economic Geology Department, for assistance in preparation of the microphoto- graphs, and Mr. J. H. Biddle of the Adelaide University, Geology Department, who carried out the X-ray diffraction measurements on the olivine. The senior author gratefully acknowledges permission granted by the Director of Mines for the publication of his observations on this meteorite. REFERENCES. 1, Deer, W. A., Howie, R, A., and Zussman, J., 1962. The Rock Forming Minerals, vol. 1, p.4 2. Mason, B., 1962. Meteorites, p. 91. Spencer, L. J., 1934. A New Meteoric Stone from Silverton, New South Wales. Min. Mag., vol. 23, pp. 569-572. oe J. E. Josnson anp D, H. McCoun PLATE 1 > C A. General view of Cockburn Meteorite. B. Thin section in polarised light to show chondrule and silicate texture (x60). C. Polished section showing troilite (grey) and plessitic structure in the nickel-iron (white) (x 250), PARACANTHORHYNCAUS GALAXIASUS, A NEW GENUS AND SPECIES OF ACANTHOCEPHALA FROM A FISH. AUSTRALIAN ACANTHOCEPHALA NO. 12. BY S. J. EDMONDS* Summary About 40 specimens of a fish Galaxias attenuatus (Jenyns) that were collected in some fresh water streams and pools on the property of Mr. B. S. Hyde near Port Lincoln, South Australia, were brought to me for identification by Dr. P. G. Martin of the University of Adelaide. The fish were found to be heavily infested with acanthocephalans which differ from previously described genera and species. If one uses Golvan's key (Golvan, 1960b: 713) they fall into the class Palaeacanthocephala and the family Cavisomidae. PARACANTHORHYNCHUS GALAXIASUS, A NEW GENUS AND SPECIES OF ACANTHOCEPHALA FROM A FISH. AUSTRALIAN ACANTHOCEPHALA No, 12 by S. J. Epmonps* [Read 13 April 1967] About 40 specimens of a fish Galaxias attenuatus (Jenyns) that were col- lected in some fresh water streams and pools on the property of Mr. B. S. Hyde near Port Lincoln, South Australia, were brought to me for identification by Dr. P, G. Martin of the University of Adelaide. The fish were found to be heavily infested with acanthocephalans which differ from previously described genera and species. If one uses Golvan’s key (Golvan, 1960b; 713) they fall into the class Palaeacanthocephala and the family Cavisomidae. Paracanthorhynchus n,¢. Diagnosis: Acanthocephala with characteristics of the subfamily Vanclea- veinae Golvan, 1960a. Parasitic in small intestine of fresh water fish. Body small, Trunk subcylindrical or fusiform. Body spines restricted to a small triangular area on the anterior and ventral surface of the trunk. Arrangement of body spines same in both sexes, Introvert of short to moderate length, cylin- drical and with hooks that are not differentiated dorso-ventrally, Hooks with simple rooting processes. Sheath double-walled and cerebral ganglion placed in its middle. Lemnisci as long as or a little Jonger than the sheath. Male organs occupy posterior half or two-thirds of the trunk, Testes ellipsoidal and placed behind each other, Four cement glands, short and pyriform. Eggs slender with polar prolongations of the middle shell. Type species: Paracan- thorhynchus galaxiasus, Paracanthorhynchus galaxiasus n.g., 0.8.p.: figs. 1-6 Small slender worms, ‘Trunk subeylindrical with maximum width in the anterior third of worm. Body of preserved specimens may be straight hut body of living specimens usually slightly S-shaped. Female noticeably Jarger than male and the posterior part of its trunk mere slender. Trunk: Length of male 2-7-4°2 mm and maximum width 0-5-0-6 mm; length of female 4:5-8-0 mm and maximum width 0-5-0-9 mm, Srmiall triangular area of body spines on anterior ventral region of trunk of both sexes, the spina- tion scarcely extending to the dorsal surface of the trunk. Spines comparatively large, 0-2-0°3 mm long. No genital spines. Introyert: Cylindrical and not long, Length of armed section in male 0-32- 0-36 mom and maximum width 0-12-0-15 mm. Corresponding measurements. in female are 0.32-0.38 mm and 0:13-0:18 mm. Unarmed truncated collar or neck * Department of Zoology, University of Adelaide. Trans. Roy. Soc. S, Aust, (1967), Vol. 91, S. J. EDMONDS lay | | | \ YU \ W Figs. 1-6. Paracanthorhynchus galaxiasus. 1 male, 2 female, 3 introvert, 4 some hooks from the introvert, 5 body spines, 6 eggs. PARACANTHORHYNCHUS GALAXIASUS wn.st. 43 0-10-0-15 mm long, Armed with 12 rows of 7 hooks per row, the size and shape of some of the hooks is shown in Fig. 4. Last hook of each row is smallest. No marked difference observed in size of hooks on dorsal and ventral surfaces of introvert, Sheath: Arises just posterior to introvert hooks. Double walled, Length about 0-6-0-8 mm and maximum width 0:19-0:23 mm. Ganglion present in mid-region of sheath. Lemnisci: Rather stout, about one to one and a half times as long as sheath, Male structures: Testes, ellipsoidal, 0-35-0-45 mm long and in tandem. Cement glands: Four, club-shaped and pressed closely together. Male aperture: Appears to be subterminal in some but terminal in fully extended specimens, Female structures: Uterine bell about 0-4 mm long and uterus and vagina about 2-0-2:5 mm long, Eggs; Slender. Outermost covering very thin and collapses easily 55-67 » long and 7-11 » wide with polar prolongations of the middle shell. Type host: Galaxias attenuatus (Jenyns). Type locality; Coomunga, near Port Lincoln, South Australia, Holotype and paratypes: Australian Museum, Sydney. Manter (1955: 67) reported the presence of acanthocephalans of the genus Acanthocephalus from Galaxias attenuatus in New Zealand. Parasites of this genus possess six cements glands and lack body spines. Consequently Manter’s specimens must be different from the South Australian specimens. REFERENCES Gotvan, Y.-J., 19602. Le Phylum des Acanthocephala (troisiéme note), Arm. Parasitol. 35 (3), pp. 350-356. Gotvan, Y.-J. 1960b. Le Phylum des Acanthocephala (troisitme note), Ann. Parasitol. 35 (5-6), pp. 713-723. Manven, H. W,. 1955. Parasitological Reviews. The Zoogeography of Trematodes of Marine Fishes. Exp. Parasitol., 4 (1), pp. 62-86. SUBDIVISION AND STRUCTURE OF THE PRECAMBRIAN (WILLYAMA COMPLEX AND ADELAIDE SYSTEM), WEEKEROO, SOUTH AUSTRALIA BY J. L. TALBOT* Summary The basement complex at Weekeroo can be subdivided into 6 major units of schists and gneisses. Only one of these units shows undoubted sedimentary features although other of the units may be sedimentary in origin. The basement complex has been deformed at least twice prior to the deposition of younger Precambrian sediments. High grade metamorphism with associated pegmatites and granites accompanied the first phase of deformation and a lower grade of metamorphism appears to have been imposed in the second phase of deformation. The younger Precambrian sediments consist of the Burra Group overlain unconformably by the glacial Umberatana Group. The sediments and the underlying basement were deformed in the Palaeozoic orogeny and metamorphosed to the biotite grade. SUBDIVISION AND STRUCTURE OF THE PRECAMBRIAN (WILLYAMA COMPLEX AND ADELAIDE SYSTEM), WEEKEROO, SOUTH AUSTRALIA by J. L. Tausot* [Read 13 April 1967] SUMMARY The basement complex at Weckeroo can be subdivided into 6 major units of schists and gneisses. Only one of these units shows undoubted sedimentary features although other of the units may he sedimentary in origin. The basement complex has been deformed at least twice prior to the. deposition of younger Precambrian sediments. High grade metamorphism with assuciuted peematites and granites accompanied the frst phase af deformation and a lower grade of Tavtamorphism appears to Have been iniposed in the second phase of deformation. The younger Precambrian sediments consist of the Burra Group. overlain uncon- formably by the glacial Umberatana Group, The sediments and the underlying basement were deformed in the Palaeozoic orogeny and metamorphosed to the hiotite prade, INTRODUCTION In the Olary region of South Australia two distinct sequences of Precambrian rocks are exposed (Fig. 1), an older basement of schists and gneisses named the Willyama Complex (Mawson, 1912) and younger Precambrian sediments referred to the Torrensian and Sturtian Scries of the Adelaide System (Campana and King, 1958), The Olary Province (Campana, 1956; Glaessner and Parkin, 1958) is one of a number of areas within the Adelaide geosyncline in which the basement is exposed, The age of these basement outcrops is unknown but granitic rocks within them have ages ranging from 1,700-1,400 million years (Compston ef. al., 1966), This paper describes the geology of a small area in the Olary Province. Detailed mapping was undertaken to investigate the relationships between the basement and overlying sedimentary mantle, The results of this mapping are summarised in Fig. 2 which shows the distribution of rock units in the area studied, A considerable modification of the rock units and structure reported by Campana and King (op, cit.), is suggested as a basis for extension of the mapping to other areas. General Relationships The distribution of rock types in Fig. 2 shows the younger Precambrian Burra Group resting with marked unconformity on a folded sequence of schists and gneisses. The structure in the Willyama Complex is dominated by a large antiform® in the central outcrop and a series of antifurms in the western outcrop. On the two simplifying assumptions that the units recognised in the Willyama Complex reflect original sedimentary units and that the ventral antiform is in A ih A Ee Sn I REA RN RS RSS RRS AARNE REY SNES * Department of Geology, University of Adelaide. Present address; Department of Geology. Lake ead University, Port Arthur, Ontario, Canada, *"Campana and King interpreted this ventral untiform as a syncline and hence concluded that the gneisses were younger than the schists. (Antiform is a duseriptive term for an arch- like structure. It does not contain the stratigraphic implications of anticline, ) Trans. Roy. Suc, S. Aust, (1967), Vol. 91. 46 J. L. TALBOT fact an anticline, a sucevssion can be erected for the Willyama Complex in which the “leucogneiss” is the lowest and the “bedded mica schist” the highest recog- nisable unit, Six mappable units have been recognised, their relationships being shown in the map and legend of Vig. 2. A discussion of the validity of the assumptions and hence the origin and significance of the rack units is given in the final section of this paper, Fig. 1, Locality Map. Area mapped outlined in solid black, north ef Mannahill. Micrases Cua foe _ i | rOUNGEH (UTE CAMA AN ee eo The younger sequence occurs in four prominent tongues protruding into the outcrop of the Willyama Complex. The structure of the tongues is complex but the north-south trends are in marked contrast to the ENE trends of the Olary Are in this general region. Explanations for this anomaly may he sought in the relationships to the basement structures but such a discussion is beyond the scope of this paper. The Willyama Complex The six nnits of the Willyama Complex consist of 4 lower units which are essentially gneissic and 2 upper units which are schistose. Each unit is quite variable but an attempt is made to describe the essential characteristies by which the unit may be recognised in other arcas. The lowest unit is described first. Layered and Foliated Leucogneiss The lowermost structural unit is exposed in the core of the antiformal strue- ture which dominates the southern central outcrop of the Willyama Complex. The unit is composed of a Jower sequence of massive to macro layered leuco- gneisses, granofels® and migmatites, a middle group of more migmatitic gneisses and an upper series of layered gneisses. These layered gneisses show remark- ably planar layering 1-10 cm thick and form a distinctive marker horizon. The base of the unit is not exposed. The mineralogical composition of the leucogneisses is simple, consisting essentially of acid plagioclase with less quartz and with accessory lo minor pale- olive biotite. Muscovite may also occur in minor amounts and some rocks with a granitic appearance also contain microcline. The pegmatitic segregations in the migmatites also show abundant microcline. The preferred oricntation of micas is normally parallel to the layering but in the cores of some small folds is parallel to the axial surface of the folds. The layering in the upper part consists of alternations of plagioclase-quarty-minor biotite layers with coarser layers of plagioclase-biotite with minor muscovite, (Plate 1, Fig. 1.) * Granofels, Medium to coarse metamorphic rock with no obvious foliation (Goldsmith, 1959), WiiiraM4s COMPLEN Seddid mica sehist AilWyiurm YOUNGER PAECAMSSIAN an Umberatone Group F = ; ara Gurra Grow (GNEOQUS FOCwS ey Afi phfonbe . a —1} Follated qrente Mica schist a Bedding Faotiatian Laythsd ones ‘ Eroailei crass Smull tovel TT om Cleavage Mignal tic genial aall — Prue (ones Leyeyane 53 Fig. 2. Simplified gealogic wap of yeirtivn wl Weekerar und Mtovialpa Stations, Sputh Australia, The houndarles of the area are Lonpitudes 139° 455' to 139° 59°E ond Luatitides 32* 10° to 32° 16'S. 45 J, L. TALBOT Migmatitic Schist Structurally overlying the layered gneiss is a quarts-feldspar-mica schist crowded with pegmatitic schlieren. ‘The pegmatites are commonly only a few centimetres wide, are lensoid and generally parallel to the foliation. Small isoclinal folds in the pegmatites are common eyen in the cores of the large folds, and a crenulation cleavage is developed parallel to their axial surfaces. Compositional layering within the migmatitic schists is comparatively rare aud where observed is thin and parallel to the cleavage, There are, however, two bands of cale-silicate schists about 25 m thick which show well developed layering. The layering in these schists conforms to the large scale structures, the layering commonly being at a high angle lo the cleavage. The relationship between the structures shown by the cale-silicate schists and the migmatitic schists is shown schematically in Fig. 3. The layering in the migmatitic schists is said to be transposed (e.g. Turner and Weiss, 1963, p. 94). Fig. 3. Sketch showing the relationship between the transposed migmatitic schist unit (linvd) and gneiss band (stippled ). The uppermost part of the migmatitic schist unit consists of a distinctive layered biotite-gneiss 20 m thick, In thin section the migmatitic schists show coarse relics of quartz, microcline, muscovite and biotite in a highly foliated sericitic matrix. Plagioclase appears to be uncommon in these schists but the presence of large amounts of sericite may indicate that any original plagioclase may have been completely replaced. The cale-silicate schists vary from layered two-mica schists to dark-brown biotite schists with up to 80% cpidote. A blne-green hornblende and calcite occur in some sections. Granitoid Gneiss Granitoid gneisses occur in the central part of the area and along most of the eastern border, A wide varicty of rock types have been included in this unit: weakly foliated granofels, gneissic migmatites, layered leucogneiss and some minor schists. The lack of characteristic markers within this unit makes correlations across faults and the delineation of macroscopic structures extremely difficult. The gneisses along the eastern margin are particularly unsatisfactory in these respects. A number of mappable sub-units can be distinguished in this area but they are sufficiently repetitious to make it impossible to decide whether a major structure corresponding to the central antiform is present in this region. Similarly a definite correlation between these gneisses and those of the central region must be regarded as tentative. An extension of the mapping to the east may solve some of these problems, SUBDIVISION AND STRUCTURE OF THE PRECAMBRIAN. WREKEROO, SA. 49 The mineralogy of the yneisses is relatively simple, quartz and plagioclase (Au 15) being the most important constituents with smaJl amounts of biotite and/or muscovite. Microcline occurs in some specimens and sphene and opaques are ubiquitous accessories. Layered gneisses and schists In contrast with the non-layered nature of the previous unit the next highest unit is layered on all scales. Macroscopic layering is clearly visible on aerial photographs and is an expression of alternations of |eucocratic poorly layered ancisses. with layered mica-gneisses and schists. The rocks are also well layered on a small seale antl where sufficient biotite is present a foliation parallel to the lithalogie layering is well developed. In a number of localities the more Jencn- cratic phases of the gneisses grade laterally into dense, unlayered “aplitic” and pesmatitic rocks, Campana and King (1958, p. 30) interpreted these changes as a result of “blastic recrystallization” of pre-existing feldspathic rock. The leucogneisses are composed of varying proportions of quartz and plagio- clase with small amounts of K feldspar in some specimens. Trains of accessory sphene, zircon and apatite commanly outline the only layering visible in the tocks. Variable amounts of mica mav be present in the rocks and forms an essential constituent of the schistose phases. Miva Schist and Bedded Mica Sehist The four units described above are predominantly encissic with some granitic and minor schistose phases. Micas do not average above 20% in these units and normally are considerably lower. In marked contrast to these gneisses is a group of brown and grey mica-schists whicli crop out in the northern and western part of the mapped Willyama Complex, These schists have been referred to as the Weekervo-Schists by Campana and King (1958). In this study these schists have been subdivided into a lower “Mica-Schist” unit and an upper “Bedded Mica-Schist” and wre described separately. Campana and king (1958) rvcug- nised an Ethiudna Cale-Silicate Group between the gneiss units already described and these seltist units. The present author docs not recognise this as valid regional unit for reasons to be discussed later. Mieu Schists The contact of the Weekeroo Schists with the structurally lower gneisses is gradational. The thickness and number of Jeucogneiss bands decreases: rapidly over a distance of 50 m to a dominant rock type of silky muscovite-rich schist, Most of the “Mica Schist” outcrops are monotonous unlayered quartz muscovite schists with muscovite commonly making ip over 50% of the rock, Some of the outcrops contain abundant layers of quartz-rich schist which may be transposed sedimentary layers, These layers ave however never sufficiently thick or persis- tent to outline large scale structures, ‘The most common mineral assemblage of the schists is quartz-muscovite with minur feldspar, biotite and/or chlorite, Garnet occurs in some specimens and a light brown to blue pleochroic tourmaline is a common accessory, Some localities show large white porphyroblasts with a tetragonal outline. Although these porphyroblasts are now predominantly compased vf a white mica, relict andulusite occurs in some, Rarely, slaurolite and chloritoid oceur within the porphyroblasts and in one specimen sillimanite vecurred with them, No mutual replacement textures were obscryed between these three minerals althoygh all three appear to be replacing, the andalusite. 5) J. L. ‘TALBOT Bedded Mica Sehists To the north of the “Mica Schists” ave a sequence of schists and granofels which show macrosopically folded layering visible on acrial photographs. The macroscopic layering is parallel ta a small scufe layering which in places shows ernss-bedded ripple structures (Plate 1, Fig, 2). There appears to be little dowht therefore that the layering in this unit represents bedding. Consistent fureset truncations indicate that the bedding and strictures north of the western wiaphibolite are right way up, It is probable therefore that the “Bedded Mica Schist" unit is younger than the “Mica Schist™ unit unless there is some unsuspected structural complication, The mineral assemblages of the bedded mica schists are similar to those in the "Mica Schist" unit, Carnet and andalusite are however rare. The rocks are finer grained and contain more quartz. Older Precambrian Metamorphism and Igneous Activity There are a number of metamorphic episodes affecting the Willyama Cam- plex anc there are considerable problems in outlining a metamorphic history. The general appearance of the rocks, especially the gneisses, suggests the meta- morphic grade was quite high, Abundant migmatites, pegmatites and foliated granites are suggestive of at least lower amphibolite facies. Relicts of sillimanite and andalusite in the schists confirm this impression. Most of the rocks however contain assemblages which are consistent with the greenschist facies, that is, muscovite-chlorite-garnet in the schists and albite- niuscovite in the gneisses, Tt appears therefore that an earlier phase of higher grade metamorphism has been overprinted by a later phase of lower grade metamorphism. Structural evidence supports this hypothesis, and siuggests the second phase of metamorphism occurred before the deposition of the Adclaide System rocks. An early phase of tectonisin resulted in an essentially planar er at least isoclinal attitude of layering with « foliation parallel to this layering. The large develop- ment of migmatites and pegmatites also appear to be associated with this phase of tectonism. The pegmatites which are commonly parallel to the foliation vary in thick- ness from a few centimetres to over LOO metres. The distribution of pegmatites shows a high degree of correlation with rock type, pegmatites being most abundant in the migmatitic schist and layered gneiss units and somewhat less ahuntlant in the mica schist unit. On a smaller scale bands of mica schist und cale-silicate schist within the migmatitic sehist unit are remarkably free of pegmatite although the surrounding schists are crowded with pegmatites. Such field occurrences suggest an intimale genetic rvlalionship between the formation of the peginatites and metamorphism rather than an introduction af material from outside. Such processes are typical of the middle of upper amphibolite facies rather than greenschist facies. The second phase of metamorphism is related Lo the sevond pluse of defor- imation of the rocks in the Willyama Complex. This phase of folding has resulted in the general outlines of the large scale structures now visible in the Willyama Complex. The pegmatites are folded conformably with the lithologic layering. During this phase of deformation most of the rocks were recrystallized and the present mineral assemblages established, Most direct evidence of the previous higher grade of metamorphism appears to have been obliterated during this tectanie episode. SUBDIVISION AND STRUCTURE OF THLE PRECAMBRIAN. WEKEREROO. S.A SI The granites marked on the map (Fig, 2.) ave [pliated to some degree tut differ from the granitic gneisses in their lack of marked compositional variation across the bodies. The granites are composed of varying proportions of quartz, microcline, albite, muscovite and hiotite. A weak preferred orientation of the micas defines a poor foliation in some specimens. The texture is coarse-grained hypidiomorphi¢ and the minerals show abus- dant strain shadows in thin section, These textural features contrast strongly with the unstrained granular texture of most of the gneisses. Amphibolites are quite comipon fn the Willyvama Complex, the more immpar- tant bodies being indicated in Fig. 2. Large amphibolites occur within the schist sequences at several localities. The rocks are massive amphibole-slbite-epidote rocks with no well pronounced foliation. Although metamorphosed, relict igneous textures are common. It seems clear therefore that they represent altered basic igneous rocks. Some structoral features are suggestive of a volcanic origin (janes, Talbot and McBriar, 1962) but more recent work has brought to light intrusive contacts and breccias and the racks are more reasonably interpreted as dolerite plugs. Albitization of the surrounding country rock is also a characteristic feature af these amphibolites. Campana and King (op. cit.) interpreted these amphibolites as sedimentary and equivalent to the Ethiudna Cale-Silicate Group. The present author, how- ever, can find no evidence of sedimentary amphibolites in the Weekeroo area although other types of cale-silicate occur within the migmatitic schist unit. An- other group of dark green amphibolites occur as long, narrow nearly vertical dykes generally with a north-south trend. In a nimber of localities the dykes are folded on a large scale and in many places the margins of the dykes are now chlorite schists with the foliation parallel ta the dyke margins, In some instances the dykes follow faults. None of these dykes cut the Adelaide System rocks and «a number of them are cut of by the unconformity.” The dykes, however, cut across the large folds in the Willyama Complex and are therefore later than the sevond phase of deformation. It is not known whether the dykes are related tu the large amphibolite bodies but the generally massive nature of the latter Suggests they were formed after the metamorphism associated with the second deformation. The Younger Precanibrian Metasediments Lying with marked unconfonnity upon fhe Willyama Complex are meta- morphosed sediments of the Adelaide System. Two sequences are represented, x lower group believed to be the equivalent of the Burra Group (Torrensian) and an upper group which is correlated with the Umberatana Group of Sturtian age (for a discussion of correlations within the Adelaide Geosyncline sce Thomsun, 1966). The original sedimentary nature of the younger sequence of rocks fs readily apparent. The rocks are well-layered and show an abundance of small scale sedimentary features such as ripple marks, cross bedding and convolute lamina- tions, and in the basal beds rare suncracks. Facies chanyes in manv horizons gan be distinguished easily and original variations in the quartzites and ¢lactal horizons are particularly prominent, The macroscopic structures are relatively simple and individual marker horizons can be followed for comparatively large distances. * 4 single example of « light reen uralitized dolerite is seen cutting the folded Burra Growp in the northeast tongue. This dyke has opt becn distinguished on the map froni the Preeambrian amphibolite dykes but it is clearly later. Jj. L. TALBOT or it Many of the original sedimentary features, however, have been modified by deformation and the overall simplicity of the large scale structures is some- what misleading. Bedding in the schist units is largely subparallel to the foliation and is heavily earsHcinel (Plate 1, Fig..3). Thinning of the schist units is also readily apparent (Plate 1, Fig. 4) and so calculated sedimentary thicknesses are unreliable, Tectonic thinning or thickening has not been. allowed for in con- struction of the fence diagram in Fig. 4 so the thicknesses can only be regarded as relative, Burra Group A generalised stratigraphic column for the Burra Group in the northeastern seat is shown in Fig, 4. The relative positions of prominent marker beds in the other three outcrop areas is also shown. Detailed columns could not be con- structed in these other areas on account of poor outcrops. No attempt has been made to subdivide the sequence into formations as the dominant rock type. mica schist, is similar throughout the sequence, Fig. 4, Stratigraphic column for the Burra Gronp of the northeast tongue (right hand pillar) and fence diagram to show variations over the whole area. The other three uprights represent sections inersured in the other three tongues of the Burra Group. The numbered marker beds are:— 1. Basal conglomerate 2. Majar dolo- mite 3, White quartzite 4, Cross bedded silty sandstone 5. Sandstone &, Tillite 7, Festoon ripple sand- stone, Markers 1-4 are in the Burra Group and 5-7 in the Um- beratana Group. The contact between the Burra Group and the Willyama Complex is well exposed over most of the arca mapped and along all western contacts of the Burra Group tongues is an unsheared sedimentary contact. The rocks lying directly above the unconformity are for the most part meta- morphosed conglomerates. In a few localities impure quartzites rest on the Willyama rocks. The conglomerate is quite variable in particle size, composition and thickness, and these variations show a correlation with variations in the underlying Willyaia Complex. The maximum thickness of the conglomerate is about 15 m and the maximnm pebble size about 10 cm diameter. In general the thicker, coarser conglomerutes are found in contact with the more schistose rock types of the Willyama Complex; conglomerate may be absent altogether in contact with the more massive gneisses, These differences appcar to represent lows and highs respectiycly in the unconformity surface, The composition of the basal beds varies sympathetically with the rocks in the underlying Willyama Complex. In all the outcrops pebbles of vein quartz are prominent. Adjacent to the gneisses, pebbles of quartz-feldspar gneiss are very common and pebbles of quartz-feldspar-biotite gneiss are also. present, Adjacent to the schists, gneiss pebbles arc less common and vein quartz pebbles SUBDIVISION AND STRUCTURE GF THE PRECAMBRIAN, WEEKEHOO, 5A Bib predominate, but as might be expected schist pebbles are not abundant. Pebbles of amphibolite and albite rock are found in the conglomerate resting an the amphibolite and in some exposures epidote (from the amphibolite) is abundant in the matrix. The conglomerate is interpreted asa reworked regolith, The correlation of pebble type with underlying rock type in the Willyaraa Complex suggests that there has been little transportation of the pebbles. The higher beds of the basal unit consist of fairly uniform impure quattzites, metagreywackes, minor conglomerates and interlayered mica schists. The quart- zites commonly show cross-bedding, heavy mineral laminations, and ripple marks; suncracks were observed in two localities. Above this basal sequence the rocks are predominantly mica schists. The most common mineral assemblage is quartz biotite-chlorite (muscovite) and quartz-biotite-actinolite (dolomite, chlorite), Most of the rocks have a@ pro- nounced preferred orientation of biotites although in many specimens a schistosity is not obvious on casual inspection, Tn the lower part of the sequence, bands of dolomite marble are commun, and sedimentary magnesite occurs in some horizons. The must common assem- blaves ave: quartz-dolomite, quartz-dolomitc-talc, The assemblage quartz- dolomite-tremolite occurs in a number of localities. Metasandstones are also common throughout the sequence, two prominent horizons being a white ripple marked quartzite in the middle of the sequence ani a fner grained impure sandstone near the top of the sequence. In the two southern tongues a light grey quartzite with festoon cross-bedding occurs directly beneath the Umberatana “tillite”. Umberatuna Group Rooks of the Umberatana Group rest with apparent conformity on the Burra Group over much of the area but a distinct angular unconformity vf up to 40° occurs in the northwestern part of the area, In this latter region the Umberatana Group overlaps all parts of the Burra Group and rests directly on the Willyama Complex, Where an unconformity cannot be recognised the base of the Umberatana Group has been placed at tho first appearance of large houlders in the schists, The incoming of large boulders also corresponds to a general change in the appearance of the mica schists, 1n contrast with the schists of the Burra Group which are soft and show a good schistosity, the schists of the Umberstana Group are hard dense brown rocks commonly with only a poorly developed schistasity. The basal unit of the Umberatana Group, interpreted as a tillite by Campana and King (op. cif.) is a quartzrich biotite granofels with large boulders and with hedding completely absent in most outcrops. The horizon is very variable, containing numerous large boulders in some parts and being almost free of boulders in other parts. The boulders range in size up to 50 em diameter and are most commonly well-rounded, althongh boulders less than 5 cm diameter tend to be sub-angular. The composition of the boulders is very variable. in some outcrops boulders from the Willyama Complex are dominant, whist im others dolomites and quartzites from the Burra Group predominate. The matrix of the “tillites’* is a characteristic dark grey structureless fine-grained biotite * The sedimentary term. “tillite” is used to describe this rock as no sutistactory meta- morphic term is available, 7 54 J. i. TALBOT granofels which no longer shows sedimentary characteristics. Orginal grains larger than 0-1 mm are however still preserved, These grains, mostly of quartz but also of aggregates of dolomite or quartz (and feldspar) vary from rounded to sub-angular and shew a wide variation in grain size. The reconstituted matrix in which these fragments occur has a gtain size varying from 0-02-0-05 mm and consists essentially of quartz, olive-green biotite and carbonate, with accessory muscovite and granular opaqnes. The rocks above the basal “tillite” are dominantly fine-grained quartz-rich mica schists and granofels with minor dolomite marbles and quartzites. Scapolite occurs as porphyroblasts up to 0°5 cm in diameter in some horizons in the mica schists in the northern part of the area. Scattered boulders (most commonly granite) up to one metre in diameter occur in all parts of the sequence but are wacommon in the upper part. Sedimentary structures, such as convolute laminations and small scale crass-hedding, are common in the quartz-rich mica schists. In contrast to rocks of the Burra Group, biotite shows a rather poor pre- ferred oricntation in the Umberatana Croup rocks, with a consequent lack uf foliation in most outcrops. This lack of foliation is net regarded as a fundamental metamorphic difference between the two groups but rather due to differences in lithology, Palagozoic Metamorphiym and Igneous Acticity The mineral assemblages in the Adelaide System rocks are indicative of biotite grade (middle greenschist facies) of metamorphism. No granitic rocks or pegmatites are found within the Adelaide System rovks and the only evidence of large scale metamorphic differentiation is the presence of quartz veins in some localities. Quartz in pebbles and in quartzites show strong preferred orientatiuns which are related symmetrically to the axial plane schistosity (Talbot, 1962). The metamorphism and large scale folding and deformation appear therefore to be related and correlation with the major orogeny which affected the whole of the Adelaide Geosyncline, indicates it to be Palaeozoic in age.* A single light green dvke occurs cutting the folded Burra Group NNW from the Weekerou Station shearing shed. The rock is naw largely an assemblage of albite, chlovite, epidute and magnetite, showing relict lathes of plagioclase. It therefore appears that this dyke was intruded after the peak of the Palaeozoic metamorphism and certainly after Folding had ceased. Structural History A detailed analvsis of the structural geology is beyond the scope of this paper and only a brief outline of the structural history will be presented here. As noted previously the overall distribution of rock types within the Willy- ama Complex is largely the result of the second recognizable tectonic event, This period of deformation affected rocks which already possessed a well defined schistosity and which were already high grade metamorphic rocks, that is, gneisses and migmatites, Large scalc folds were formed with a new foliation, a crenulation cleavage, parallel ta the axial surface. The trend of this steep foliation is roughly NE to ENE and the associated large scale folds commonly plunye ut moderate angles to the NE although a wide range of plunises has been observed. The majority of the amphibolite dykes were intruded after this phase of deformation and it is alse pessible that the large amphibolite bodies were also formed at this time. * Latest information suggests the age of the metamarphism of the Adelaide Geosyncine rocks is 490 m, years, (Campston et. al., 1966.) SUBDLVISION AND STRUCTURE GF THR PRECAMBRIAN. WEEKKEKOO, S.A. 35 The younger Precambrian sequence was deposited on the already multiple deformed Willvama Complex. Some movements must have occurred after the. deposition of the Burra Group but the next major tectonic episode is the lower Palaeozoic orogeny which affected the whole Adelaide Geosyncline. ‘he major trends of this Palacozoic folding in the Olary region are ENE-WSW but highly divergent trends ate apparent in the vicinity of the areas of basement outcrop. Only minor inodifications of the fold patterns in the Willyama Complex have vecurred during the Palaeozoic orogeny but a number of large vertical slip faults have altered the distribution of rock types. Nomenclature and correlation The area mapped appears to be critical to a discussion Of the stratigraphic relationships in both the older and younger Precambrian, Cumpana and King (op. eit.) recognized Four main groups of rocks. in the Willyama Complex, namely “Archaean Mctusediments As”; “Feldsputhised Archagan Metasediments AF; “Migmatites and Granite Gneisses Am” and “Anatectic Granites. Granodiorites and Granulites Ag”. The last group docs not uceur in the areca mapped anil is not discussed further, Campana envisaged the first three units as representing different stages of granitization vf similar sediments and his map shows cores of Am surrounded by zones of Af and finally As, although a number of important exceptions do occur (e.g. near Old Boolcoomata Homestead ). The “metasediments” were divided into a lower group (Weekeroa Schists) a Middle Group (Ethindna Cale-silicate Croup) and an Upper Croup (Qutalpa Arkosic quartzites passing to granite gneisses)" (Campana and King, 1938, p. 21), The Ethiudna Group is said to occupy “a well defined stratigraphic horizon within the schistose succession”. The relationship between the “Weekeroo Schists” and the “Outalpa quartzites” can be seen in a number of localities, eg. Weekcroo, Old Boolcoomata and Ameroo Hill, where schists are in contact with eneisses, the wantuct being: parallel to the layering in the gneisses. The middle or Ethiudna Group is shown by Campana to lie within the schist sequence in all localities shown on the map except in the type locality of Ethiudna Mines. In this locality, near Plumbago Ilead Station, the group consists. of cale-silicates untt dolomites in an isolated gneissic terrain whose “stratigraphic” position is unknown. ‘The “stratigraphic” position of the Ethiudna Group as the Middle Croup of metasediments is however determined by the position of amphibolites at Weekeroo, Amuroo Lill and Old Boolcoomata, In none of these localities do the rocks resernble the Ethiudna rocks and in fact in two of them the cale-silicates are intrusive igneous rocks. In addition, cale-silicates occur in other parts of the section (e.g. in the migmatitie schist unit at Weekeroo) and it is considercd unwise to make long range correlations of this sort in gneissic terrains, It is therefore proposed that use of the term Ethiudna Calc-silicate Group us a regional unit be discontinued. The remaining rock types in the Weekerou section were grouped by Cam- pana and King inlo “Migmatite and Cranitic Gneiss unit Am” with the genetic implication that they are more altered than the “Archacan Metasediments As”. These have heen subdivided in this study into 3 units which are recognised as part of the overall sequence and not as more metasumatized rocks, * On Campana’s map of the Weekeroo region this unit is inched with the: “Miematites and Granitiy Gneisses Am”, imstead of with the “Metasediments As", This ix ¢learly a dratting error. 1 Plate TALBO'U L. 5G J. Li. ‘TALBOT A comparison of Campana and King’s units and the units proposed for this study is shown in Table 1. The sequence proposed for this paper is not neces- sarily a sedimentary sequence althougli many of the rocks are sedimentary in origin. The origin of the layering and of the boundaries of the gross lithologic units is of special importance to this problem. In all the schist units except the hedded mica schist unit, layering, whether originally sedimentary or not, is now heavily transposed parallel to the second generation cleavage. Hence strati- graphic relationships within these units have been completely destroyed, Never- theless, the gross relationship between the major lithologic units has not been profouudly altered by this second phase of deformation which resulted in the large scale folds now visible in the Willyama Complex. TABLE 1 Wallies Conepler Sequence Sequence of Informal secuence | Carmmpana & King (1958)* inverted uacd in this Paper r | Bedded mica schist. “Archaea” Weekerog schist | | Mies. Schist Meiasediments 4 | Ethiudna Cale-Silicate Group | {As} L Outalpa Quartzitest | Layered gneiss r Granituid gneiss Migmatites und 4 Unditterentiatedt Miguiatitie schist Granite-Gneisses (Ant) L N Leucogneiss * Campana's sequenre has been inverted as the evidence from the presint study suggests that the selusts are the youngest part of the Willyaria. Complex sequence, + This umt is called “Arkosie Quartzites, Passing to Granite-CGneisses” in the text (Campana & King 1958 p.18) but is referred to hy name in the map legend and also in Cumpana (1957). ¢ This unit was not recognised by Campana ass “stratigraphic unit but aa rocks ma particular alate of alteration, In contrast to the schist units the more massive gneissic units show relatively little transposition parallel to the new cleavage and even where folded show a foliation parallel to visible layering in the rocks, (Exceptionally a new foliation is developed in some rocks parallel to the axial surface uf smull folds.) In the more massive gueisses the foliation is discontinuous or defined by a preferred orientation of micas. In the layered gneisses foliation is well developed in the mica rich portions and is invariably parallel to the layering, In the quartzo- feldspathic layers trains of accessory minerals such as zircon and sphene are observed parallel to the gross lithological layering, The evidence suggests that although the foliation is parallel to layering the presence of layering is not a necessary condition for the formation of the foliation, Jt seems unlikely therefore that the foliation represents a simple “bedding plane foliation” imposed on relatively undisturbed planar bedding. It is more likely that the gneissic units have been heavily deformed and that the parallelism of foliation and layering are the result of large scale isoclinal folds with possibly large amounts of transposition on a small scale. This phase of SUBDIVISION AND STRUCTURE OF ‘TILE PRECAMBRIAN, WEIEKEHOO. S.A, 57 deformation resultect in an essentially planar attitude of layering and foliation and was accompanied by high grade metamorphism and the formation of peymatites. This planar layering was then folded by the second phase of deformation to give the observed major folds, The suceession of units in the Willyama Complex may not therefore have any stratigraphic significance (with exception of the bedded mica schist. This init appears anomalous in the general contest of the Willyama Complex and it was entertained that there could be an uncontornnity between it and the remain- ing units, No direct evidence could be found to support this suggestion ), The present study confirms the recognized stratigraphic relationships of the uverlying Adelaide System. Campana emphasized the transgressive nature of the Umberatana Croup and in the Weekeroo area an angular unconformity between the Burra and Umberatana Group is now well established. Recent mapping by the 5,A, Mines Department has demonstrated this unconformity in many other parts of South Australia, Differences of interpretation however arise with structure and metamorphism. The differences in structural interpretation are of limited interest but the failure by Campana to recognise the metamorphism in the Adclaide System rocks may lead to an incorrect assessment of the economic »otentialities of the younger group of rocks. Kecognition that the rocks have seen Metamorphosed to the biotite grade also removes some of the limitations to the tectonic interpretation of the area, and also imposes strict limitations to age determinations in the Willyama Complex, REFERENCES Casmana, By 1956. Granites, orogenies and iineral genesis in the Olary Province (South Australiu). Jour, Geol. Soe, Aust, 4. pp. 1-12. Campana, B., abd Kawa, D.. 1958... Regional Geology ancl Mineral Resouces of the Olary Provinve, Geol, Surv, S Anst. Ball, 84, 133 pp. Comms'ros, W., Craawron, A. K,, and Bormann, V. M., 1966. A Radiometric Mstimuate of the Duration of Sedimentation im the Adelaide Geosyneline, South Australia, Jone, Geol. Soc, Avst., 13, pp. 229-276, Gruaesyver, M, F., and Panes, L. W. (dy) (958. The Genlngy of South Australia. Jonr, Geol. Soc. Aust, 5 (2). 163 pp. Co tosmin Ry 1959. Granofels. a new metamorphic rock name, Jour, Geol, 67, pp. LOO-1 10, Jonws, J.B. Tatuor, J. L., and MoBnian, E. M.. 1962. A suite of voleanie rocks with apilitic affinities from the Archaean of South Australia, Aust, Jour, Sei, 29, p. 356, Mawson, DM, 1912, Geological Investigations in the Broken Hill Area, Mem, Roy, Soc, 5. Aust. 2, pp. 24 1-310, Tatsor. J, L.. 1962, A study of the structural and inetamorphic relationships between Older and Younger Preeambrion rocks in the Mount Lofty-Olary Are, South Australia. Unytrbl. Thesis, U. of Adelaide, 171 pp. Thomson. B, P,, 1966. The Lower Boundary of the Adelaide System and Older Basement Relationships in South Australia, Jour, Geol. Soe, Aust. 13. pp. 203-228, Yourner, F, J. and Wess, L. E., 1963, Structural Analysis of Metunorphic ‘Tectonites, MeGraw-Lhill. EXPLANATION OF PLATE Phare | Mig, 1. Banding in yneiss tear Lop of Jeuvogneiss amit, Lal, 32° 13-7'S Long, 138° 5608's, Vig. 2. ‘Vruneated laminations and scour anc fl strneture in uranofels of the bedded mica schist unit. Lat, 32° 12-65 Long. 139°" 54°, Fie, 8. Transposed layering in the wppermost beds of the Burra Groop. Arrows point to Fold closures, Lat, 32° 10-5’S Long, 139° 58-1'F. Fiv. 4, Bulk strain of the Umbpratana Tillite as indicated by deformed dolomite: erratics. 100 metves N of locality of Plate 1, Fig, 3. PHERETIMA TUMULIFACIENS (OLIGOCHAETA, MEGASCOLECIDAE) A NEW SPECIES OF EARTHWORM FROM THE SEPIK DISTRICT, NEW GUINEA BY K. E. LEE* Summary Pheretima tumulifaciens, a new species of earthworm, belonging to the family Megascolecidae, is described. It was found in the Sepik River Valley, northern New Guinea, where it is associated with unusual forms of micro relief that have been attributed to the earthworms’ burrowing and casting. The species is most closely related to P. sangirensis (Michaelsen), from Indonesia, but in some respects resembles Megascolex. The inadequacy of generic distinctions and of the criteria for distinction of groups of genera are discussed. PHERETIMA TUMULIFACIENS (OLIGOCHAETA, MEGASCOLECIDAE) A NEW SPECIES OF EARTHWORM FROM THE SEPIK DISTRICT, NEW GUINEA by K. E. Lee’ [Read 8 June 1967] SUMMARY Pheretima tumulifaciens, a new species of earthworm, belonging to the fanily Megaseolevidae, is deseribed. Tt was fund in the Sepik River Valley, northern New Cuinea, where it is associated with unusnal forms of micro- relief that have been attributed to the earthworms’ burrawing and casting. The species ig inost closely reluted to P, sdngirensis (Michaelsen), from Indonesia, hut in sone respects resembles Megascolex. The inadequacy of generic distinc- tions and of the eriteria for distinction of groups of genera are discussed. INTRODUCTION Haantjens (1965) recorded micro-relief in the form of irregular pits, trenches, mounds and ridges, not unlike some periglacial phenomena, from the humid tropical grass-covered plains bordering the Sepik River valley, south of Wewak, in New Guinea. Populations of a large earthworm species were always associated with the micro-relief and Haantjens (1965) considered that they were responsible for the formation of the micro-reliet features. In October, 1965, the author visited four sites in the vicinity of Yangoru, where Haantjeus had observed micro-relief features, to collect carthworms and examine the relationship between thcir activities and the micro-relicf features attributed to them, At each of the sites, large earthworms belonging to an undescribed species of Pheretima Kinberg were collected. The earthworms were Closely associated with the micro-relief features, as recorded by Haantjens, but it seemed unlikely that they were directly responsible for the micro-relief, A discussion of the micro-relief features and their possible origin will be pub- lished elsewhere (Lee (in press)). The large earthworm associated with the micro-relief features has heen named Pheretima tumulifaciens, and is described below. Pheretima tumulifaciens n.sp. Collection Data. (1) About 14 mi S of Kworo Village, 20 mi SW of Yangoru, District, New Guinea; 0-33 em in soil under kunai grassland; 2 clitellates; K.E.L. 29/10/65. (2) About # mi SW of Watibi no. 2 Village, 15 mi SW of Yangoru; 0-30 em in soil under secondary forest; | aclitellate; K.E.L. 22/10/65. (3) About 1 mi N of Watibi no. 2 Village, 14 mi SW of Yangoru; 0-53 cm in soil at edye of deep trenches, under kunai press ans 1 clitellate; K,E.L. 22/10/65. (4) About 1 mi SW of Haripmor Village, 15 mi SE of Yangoru; 0-45 cm in and under low mounds and ridges covered by kunai grassland; 6 clitellates; K.E.L. 93/10/65. * Division of Soils, C.S.1.R.0., Adelaide. Trans. Roy. Soc, §. Aust, (1967), Vol. 91. au K. E. LEE Type Material. Holotype and two syntypes lodged am South Australian Museum. Dimensions. Mature specimens are 500-600 mm in length, 9-10 mm in diameter (in ethyl alcohol) and have 550-600 segments. Detailed measurements of the holotype are: Length 590 mm; segments 587; diameter 10-5 mm at_vii, 9-5 mm at xiii, 9-5 mm at clitellum, tapering gradually tu 6 1mm at about half length and fairly uniform to posterior end. Posterior to iii the segments have secondary annulations, two to five per segment. Colour, Pale greyish-brown with pale brownish-red clitellum. Chaelae, Absent on i, otherwise ahout 140 per segment, evenly spaced on a slight ridge around cach segment, nearer the anterior than the posterior margin of the segraents; without a mid-dorsal gap, but with a small mid-ventral gap, about twice the nermal chactal interspace; a few chaetae missing immediately adjacent to each male pore, oa xviii, and across the mid-ventral line adjacent ta the female pores, on xiv. Prostomium prolohous. First dorsal pore at 10/11, Cliteluunm covering entire body surface from slightly posterior to the chaetal ring on xiii to the line of the chactal ring on xvil, Spermathecal pores: one pair. at 7/8, each on a prominent, transversely elongate papilla, about 2 mm x 1 mm, the pores each about 8 mm from the ventral mid-line, Female pores: one pair of smal) pores on xiv, about 0-75 mm apart. one each side of the ventral mid-line, surrounded by a pale elliptical area, in line with the chaetal ring. Male pores: ventral on xviii, one on each side, each about 3-5 mm from ventral mid-line, in line with chaetal ring, not on papillae; each pore medial to a deep longitudinal groove, oyerhung by a prominent ventro-lateral ridge that runs longitudinally from xvii to xix (Iigs. 1, 2), Tubercula pubertatis: two pairs of prominent, flattened papillae, a pair on xvii and 4 pair on xix, each papilla about 3 mm diameter, forming, together with the prominent ventro-lateral ridges on xviii, the edges of a deep depression that becupies the ventral surface of xviii (Figs. 1, 2). Sepla v/vi, vi/vii, vii/vili and viii/ix are thickened and strongly muscular. Thick muscle fibres arise from the body wall in segments vi-ix and rm forward to an insertion on the posterior surfaces of the muscular septa; some of the fibres are atlached anteriorly to the pharynx and not to the septa, Similar muscles have been observed previously in many earthworms, especially in large species; their purpose is apparently to produce unusually powerful contractions of the anterior segments, probably to facilitate burrowing and provide the traction necessary to move the large bulk of the carthworm'’s body. In most species of Pheretina, septa viii/ix or ix/x, or both of them, are absent or greatly reduced in size. This is not so in P. tremilifaciens. The pharynx occupies most of i-iv, and has a diffuse glandular coating, with many fine muscle fibres attached to the body wall. In v-vii the alimentary canal is a thin-walled tube with large finely lobate salivary glands lying, beside it and opening by narrow ducts into the pharynx. Giszard strongly muscular, in viii. Ocesophagus extends from ix to xv where it expands abruptly into the intestine, close to the anterior end of xv, There is no typhlosole. Many species of Pheretima have a pair of rounded or conical intestinal caeca, arising in xvvi PHERETIMA TUMULIFACIENS, NEW EARTHWORM SPECIES Fig. 4 vibv. Fig. 1. Pheretima tumulifaciens, ventral aspect, segments xiii-xx. Fig. 2. P, tumulifaciens, diagrammatic transverse section at segment xviii, to show position of male pores in relation to ventro-lateral ridges. Fig. 3. P. tumulifaciens, arrangement of hearts of segment xiii in relation to dorsal and ventral blood vessels (diagrammatic ). Fig. 4. P. tumulifaciens, left prostate, medial aspect. Fig. 5. P. tumulifaciens, left spermatheca, medial aspect. ABBREVIATIONS cl, = clitellum; d-b.v. = dorsal blood vessel; div. = spermathecal diverticulum; f.p. = female pore; h, = heart; lr, = ventro-lateral ridge; m-p. = male pore: pd. = prostatic duct; uct; p.gl. = prostatic gland; sac = spermathecal sac; sp.d. = spermathccal sp.p. — sperma- thecal pore; t.p, = tuberculum pubertatis; v.b.v, = ventral blood vessel. fi2 KE. LEE ov thereahouts, and projecting ferwurds, sometimes through several segments, usually pressed tightly against the ventro-lateral ov lateral aspects of the intestine. There are no such caeca in P. tasmelifactens. Dorsal Wood vessel unpaired; it can be traced forward to the posterior end of the pharynx, where it divides into several branches, Stout commissural vessels (hearts) in x-xiii, one pair in each segment, connecting the dorsal and ventral blood vessels; the two vessels of cach pair have a short transverse con- nection across the oesophazus, independent of the connection of the two vessels to the dorsal vessel (Big. 3), There is no independent supra-intestinal vessel connecting the hearts lonyitudinally. In vii-ix similar lateral vessels arise from the dorsal vessels, but divide into a oumber of brauches leading to segmental organs aed are not “hearts”. The testes are in x, xi; those of xi are in testis sacs that occupy much of the coelomic cavity in the segment, Ovaries in xiii, One pair of small, racemose resiculae seminales, in ix. Prostates in xviii, one on cach side; laminar glandular portion with branching duets, all discharging into a short, stout, strongly muscular prostatic duct (Viv. 4). Spermathecue in viii, one pair, each discharging through a thick, rauscular duct at the anterior murgia of the segment; the sac is cylindrical, a little wider than the duct, projecting backwards into the segment and set of from the line of the duet; a cluster of sinall diverticula opens into the duct close tu its junction with the sac (Fig. 5), Micronephridia arc small and numerous in cach segment, irregularly scattered over the peritoneum, DISCUSSION P. tumulifaciens differs from most species of Pheretima in having well- developed, strongly muscular septa in the segments adjacent to the gizzard, a characteristic shared with many large megascolecid earthworms, especially those which, like P. tumulifaciens, make extensive burrow svstems going Lh to considerable depths. In most species of Pheretima the anterior septa are not thickened and septa viii/ix or ix/x or both of these are absent, or much reduced, The absence of intestinal caeca also distinguishes P. tumulifaciens from many other species of the genus. The only character that decisively places this species in Pheretima is the position of the gizzard, in viii. If the gizzard were in v, vi, or vil, it would be placed in the genus Megascoelex, There is need for a thorough re-examination of the criteria on which generic distinctions in megascolecid earthworms are based. Revised groupings of genera have been proposed by Gates (1959), Lee (1959) and Omodeo (1958). The proposed new groupings are hdsed on slightly different bases, but none of the proposals has contributed greatly to the improvement of the state of confusion that has developed over the years since Michaelsen (1900) and Stephenson (1930). defined the genera amd grouped them into sub-families and families on the basis of a supposed “family tree" of generic relationships, derived from a study of only a few anatomical characters. Sims (1966) has attempted to clarify the relationships between genera, using a computer tu assess the affinities of 30 species, belonging to 29 megascolecid genera, on the basis of 43 attributes of each species. The results of his study favoured the scheme of classification proposed hy Gates (1959) for taxonomic categorics above the leyel of genera, Computer tech- niques may take a useful contribution to megascolecid taxonomy, as they make it possible to assess relationships on the basis of a larger number of attributes than can be handled by traditional taxonomic methods. However, if computer techniques are to be used, it is at the Jevel of definition of genera that they can ephaely contribute most, and Sims’ study assumes that the zenera are adequately efined. PHERETIMA TUMULIFACIENS, NEW EARTHWORM SPECIES 63 P. tumulifuciens shows closest affinities to P. sangirensis (Michaelsen), an Indonesian species. Similaritics and differences between the two species are summarised below. P. tumutlifaciens P. sangirensis Spermathecal pores 1 pair, at 7/8 1 pair, at 7/8 Female pores paired, close to Ventral unpaired, median mid-line ventral Septum viii/ix present, muscular present, membranous First intestinal segment XV xv Intestinal caeca absent sometimes present, very short Vesiculae seminales 1 pair, in ix 2 pairs, in xi, xii Spermathecal diverticula cluster of small diver- 1 ellipsoidal ticula diverticulum REFERENCES Gates, G. E., 1959. On a Taxonomic Puzzle and the Classification of Earthworms. Bull. Mus, Comp. Zool., Harv., 121, pp. 229-61, Haantyens, H. A.. 1965. Morphology and Origin of Patterned Ground in a IIumid Tropical Lowland Area, New Guinea, Aust. J. Soil Res., 3, pp. 111-29, Lez, K. E., 1959, The Earthworm Fauna of New Zealand. N.Z. Dep. Sci. Industr. Res. Bull., 130, pp. 1-486. Govt. Printer, Wellington. Ler, K, E. {in press). Micro-relicf Features in a Humid Tropical Lowland Area, New Guinea, and their Relation to Earthworm Activity. Aust. J. Soil Res. MicHAELseN, 1900. “Oligochaeta.” Das Tierreich, vol. 10, Berlin. Omonro, P., 1958, Le réserve naturella integralé du Mont Nimba. I. Oligochétes. Mem. Inst. Fr, Afr. Noire. 53, pp. 9-109. Sims, R, W,, 1966, The Classification of the Megascolecoid Earthworms: an Investigation of Oligochaete systematics by Computer Techniques. Proc. Linn. Soc, Lond., 177 (2), pp. 125-41. SrepueNnson, J.. 1930. “The Oligochaeta.” Clarendon Press. Oxford. PERMIAN DEPOSITS OF SOUTH AUSTRALIA AND THEIR FAUNA* BY N. H. LUDBROOK} Summary Lower Permian sediments occur in all the main sedimentary basins of South Australia. The Cape Jervis Beds are described, and the Lake Phillipson Beds and Stuart Range Beds formally named. In general, the formations follow a sequence of boulder clay overlain by sands and clays mostly of marine origin, with a widely distributed fauna of arenaceous foraminifera of which Hyperammina, Ammodiscus, and Hemidiscus are the most common. Twenty-three species, of which four are new, are recorded or described. The most varied fauna occurs in the Stuart Range Beds from which the gastropod Peruvispira was recovered. PERMIAN DEPOSITS OF SOUTH AUSTRALIA AND THEIR FAUNA* by N. H, Lupsrooxt [Read 8 June 1967] SUMMARY Lower Permian sediments ocenr in all the main sedimentary basins of South Australia, The Cape Jervis Beds are described, and the Lake Phillipson Beds and Stuart Range Beds formally named. In general, the formations fallow a sequence of boulder clay overlain by sands and clays mostly of marine orixin, with a widely distributed fauna of arenaceous foraminifera of which Fyper- ammina, Ammodiseus, and Hemidiscus are the most common, ‘l'wenty-three species, of which four are new, are recorded or described. The most varied fama occurs in the Stuart Range Beds from which the gastropod Peruvispira was recovered, INTRODUCTION As recently as 1955, Permian sedimentation in South Australia was 50 im- perfectly known as, on a map of Australia showing regions of Permian sedimen- tation (Hill, 1955), to be restricted to glacial deposits in a southeast-northwest trough south of Adelaide, extending from Encountcr Bay to western Yorke Peninsula. While this is still recognized as the main area of outcropping Permian glacigene sediments, Lower Permian sands and clays have been proved by exploratory drilling to occur widely in all the main sedimentary basins. More- over, as reservoir beds for nafural gas accumulation in the Great Artesian Basin, Permian rocks have assumed an economic importance not foreseen a decade ago (Sprigg, 1966). Until foraminifera were discovered in the Permian section of Minlaton Stratigraphic Bore (Ludbrook, 1956), the depositional environment of most of the Permian sediments in South Australia was inisunderstood, It was generally believed that the sandy nature of most of the deposits indicated a Auvio-glacial origin, and that “Huvio-glacial depositions in ponded waters are but rarely met with in our Permo-Cayboniferous strata and marine depositions of this age are entirely absent” (Mawson, 1926). The ubiquity of foraminifera in the Lower Permian in all the main sedimentary basins, except that part of the Great Artesian Basin, including the Cooper's Creek Sub-basin, occurring in the northeast of the State, disproves this helief, Alter initial over-deepening by ice, marine incursions in fjords or narrow troughs took place in two main and widely-separated areas following the general southeast-northwest trend. The southeastern area includes the Murray and St. Vincent Basins and the northwestern area the part of the Great Artesian Basin lying | between the Peake and Denison Ranges, Stuart Range, and Margaret Creek. * Published with the permission of the Director of Mines. } Senior Palaeoutologist, Geological Survey. of South Australia. Trans. Roy. Soc. S. Aust. (1967), Vol. 91, 66 Nv. TI. LUDBROOK Localities from which Permian marine sediments have been proved in surface sections or by drilling are shown in Fig. 1. All material and figured specimens are housed in the collections of the Geological Survey of South Australia, except those lent by the Geology Depart- ment, University of Western Australia (UWAGD). 133 3 Vy NORTHERN | ————__,_ TERRITORY | QUEENSLAND a ene ei Boorthanna (Duff Creek) Stuart Range—> *eCoober Pedy Lake Phillipson “, sfo4oje s 5/641/6 Coorie Appa j Railway: LAKE ' TORRENS | Nullarbor, No.8 LAKE GA/RONER ore « | WALES Ls ho t \ NEW SOUTH Monash ; «Pon ess Minlaton a Renrerk Stansburyfs ADELAISE | Trowbridge Shoal 7 Clayton j Cape Jervis 4 a LOCALITIES WITH PERMIAN FORAMINIFERA— Second Valley MILES 50 Km 8¢ Fig. 1. Map showing localities with Permian foraminifera. ST. VINCENT BASIN AND ADJACENT AREAS Since Selwyn’s recognition in 1859 of glacial features in the valley of the Inman River, there has accumulated a considerable volume of descriptive litera- ture on the Permo-Carboniferous or Permian glaciation and glacial deposits in PERMIAN DEPOSITS OF S.A, AND FAUNA a the St. Vincent Basin and adjoining areas to the east on Fleurieu Peninsula and io the west on Yorke Peninsula. The papers of Howchin (1926) and Campana and Wilson (1955) may be consulted for historical reviews and comprehensive lists of references, Without designating any standard section, Howchin described most if not all of the known exposures of Permian deposits in the arca between Adelaide and Kangaroo Island. ‘The sequence at Hallett Cove he referred to as the “Hallett’s Cove beds” (1895, p. 65), while the glacigenes of Inman Valley were variously designated “Inman Valley Shale” (1895, p. 68) and the “Inman River Beds” (1926, p. 102). Unfortunately, neither of the names Hallett Cove nor Inman is now available for stratigraphic purposes as they have been applied to other units. The most complete ontcropping sequences described by Howchin are those at Hallett Cove, Cape Jervis (David and Howchin, 1897) and King’s Point, near Victor Harbor (1910a). Of these David and Howchin recognized the section at Cape Jervis as “the most extensive development of glacial till that has been up to the present observed in South Australia” (1897, p. 64). The section was iJlustrated hut not redescribed by Campana and Wilson (1953, 1955), The Cape Jervis exposure is here deseribed by R. B. Wilson and the writer as the holostratotype section of the Permian deposits of St. Vincent Basin and the adjoining areas: Cape Jervis Beds (Fig. 2). Type Section: Jervis 1-mile geological sheet, On coast, patenting for one mile (1-609 km.) northeast of Cape Jervis Lighthouse, Hundred of Yankalilla, sections 212, 211, 3009, 3010. Thickness; 95 feet (29-87 m.). The Cape Jervis Beds are unconformable upon easterly dipping greywackes of the Cambrian Kanmantoo Group. They are overlain by kunkar. The type section is described as follows:— Thickness Unit Lithology feet metres 6 Poor outcrop, grass covered slopes, probably clay as below, boulders on surface _ 23°5 7-2 5 Clay shele—grey-brown, fissile. gritty with scattered boulders, becoming red-grey motiled toward top a Ae 53 29-5 9-0 4 Sandstone—yellow-white, cross-bedded. with scattered cal- careous pebbles Ka . x ; “i 7 i 5 15 3. Sandstone—gritty sandstone and prit with boulders and thin interbeds of laminated fissile grey clay ‘m 7 be 13 3-9 2 Till-dark grey boulder till with clay-shale bands os bs 5 15 1 Till-sandy till with boulders of all sizes, principally Kan- mantoo yreywacke and Victor Tarbor Granite. ‘Thin asia (6 inches to I foot; 0-15— 0-3 m,) limestone bands... 22, 6-7 B "Total measured thickness of Cape Jervis Beds unconformubly overlying Cambrian Kanmantoo Group 2 os 98 29°8 The fossil locality number is 5/837/2 Gervis 2). The grey clay and over- lying red and grey gritty clay of bed 5 are tossiliferous, arenaceous foraminifera being present in samples CJ, 7 and C.J. 8. Abundant tests of Recurveides wilsoni Ludbrook sp. nov. accompanied in decreasing order of abundance by 68 N. H. LUDBROOK Hyperammina coleyi Parr, Ammovertella huwchini Ludbrook sp. noy,, Digitina recurcata Crespin and Parr, Ammodiscus oonahensis Crespin and Hyperammina acicula Parr. The Cape Jervis Beds are extensively developed on Jervis and Yankalilla l-mile sheets. WwW E SCALE O 40 80 FEET ie] 12:1 242 Metres Pig. 2. Sketch of type section of Cape Jervis Beds, Cape Jervis. A southerly view of the Cape Jervis section showing the Cape Jervis Beds (C.J.) overlying Kanmantoo greywacke (K) is shown on plate 1, figure 1. Fig, 2 illustrates. the lower part of the section to the sandstone ledge at 40 feet (12 m.) above the base; the interbedded nature of the sandstone and olay and the tendency for the boulders to he bedded in some parts of the section is also shown at the position of the hammer. Fig. 3 illustrates the dip ( probably depositional) and bedding of the same part of the section as figure 3, viewed south. Fig. 4 is that of a striated boulder on boulder clay at the most northerly end of the exposure. The Cape Jervis Beds are very subject to gully erosion which is illustrated in plate 2, figure 1 of the northern part of the exposure. The section at Cape Jervis provides confirmatory evidence of marine incur: sion in the Adclaide region following initial overdeepening by glaciation in the late Carboniferous to early Permian. Previously a tew broken fragments of Hyperammina. were recovered from clay collected by B. P. Thomson from Second Valley (locality Jervis 1, 5/837/1), but no other foraminifera had been observed in Permian glacigenes between Adelaide and Kangaroo Island. Parastratotype sections of the Cape Jervis Beds are exposed at Hallett Cove where the formation rests on striated pavements now declared a national reserve, and at King’s Point west of Rosetta Head, on Encounter Bay, where the sequence PERMIAN DEPOSITS OF 3.4. AND PAVNA 6y of moraine, glacial till, and sandstone described and illustrated by Howechin in 1910 is still undisturbed (plate 2. figure 3), At Selwvn's Rock (plate 2, figure 2) the southern bank of the Inman River has recently heen considerably eroded at the position of the large erratie in leoulder clay overlying the striated pavement. Exposures, however, nowhere attain any great thickness, and Permian sedi- mentation in the Adclaide region consists for the most part of infilling of over- deepened glacial valleys and basins as those of Fleurieu Peninsula (Campana and Wilson, 1953, 1955) and Kangaroo Island where a bore at Kingscote in 1909-1910 intersected 950 feet (289 m.) of typical Permian sequence before bottoming in slate. Leaf impressions were collected in 1953 by R. C. Sprigg, then of the Geological Survey of South Australia, in Permian sediments three miles (4-8 km.) southwest of Penneshaw, Their fragmentary nature preclu identification at the Australian Museum other than that they were possibly portions of equisitaean stems. In the St. Vincent Basin and adjoining areas only the Jower part of the Permian sequence uppears to be represented. The absence of coals and of good microflora-bearing clays demonstrates that the Artinskian and upper part uf the Sakmarian are missing in this area, On the other hand, the occurrence of Per- miun megaspores in the lower Tertiary of the St. Vincent Basin suggests that the ahsence of the upper part of the sequence may be due te erosian rather than t nun-ceposition, The Permian is 776 fect (236 m,} thick on Yorke Peninsula where it provides the clay pans for the salt lagoons which are exploited for commercial sult pro- duction. The sequence was completely intersected in Minlaton and Stansbury Bores (Ludbrook, 1966a). In Beach Petroleum Troubridge Shoal No, 1 Strati- raphic Well, 750 feet (225 m.) of Permian sediments were cut between S90 eet and the total depth of the well, Foraminifera cccurring sporadically hetiveen 1,010 and 1,550 feet (307-8 and 472-4 m.) permit close correlation with the marine interval in Stansbury Bore, Here also the upper Sakmarian- Artinskian earbonaccous sediments are absent. MUMNRAY BASIN Presumed Permian glacigene sediments were proved by drilling im the Murray Basin in South Australia at Renmark, Monash and Clayton. A.O.C, North Henmark No. 1 entered at 3,245 feet (989 m.) blue-grey boulder clay irregularly interbedded with calcareous sandstone in irregular lenticles and laminae. Slump structures were common, while erratic pebbles and grains of quart, granile, «rey metamorphic rocks and dark pink feldspar were variously scattered throughout the groundmass. Faceted pebbles were recovered al 3,431 feet (1,045 m.). At 3,788 feet (1,154 m.) (Core 15) there was a calcareous pyritic quartz sandstone, and at 3,864 feet (1,184 m.) (Core 16) blie-grey shale irregularly interbedded with lenses of calcarcous sandstone containing pink feldspar, biotite and pink garnet. The well bottomed in sandstone with scattered coal fragments and conglomerate bands, small slump structures and swirls of fine clay, Foraminifera occurred in almost all the cuttings between 3.410 and 4,000 feet (1,039 and 1,219 m.), but as none were recovered from cores their persistence does not necessarily indicate marine environment for the lower part of the interval. To N. J. TUDAROOK Similar sedinents with a few foraminifera were intersected in Beach Petre. leun Monash No, L Well between 2,930 and 3,820 feet (493 and 1,011 m.). ‘Ihe formation was conglomerutic at 3,285 feet (1,001 m.). A considerable thickness of at least 1,600 feet (487 m.) of Lower Permian sands and shale was intersected in W.G. Goyder Donna No, 1 Well drilled ai Clayton, 54 miles south-southwest of Milang, near the southwestern murgin of the Murray Basin. At 500 feet (152 m.) the well passed into brownish mud- stone with fine muscovite and some carbouaceous matter, This was a thin hed uverlying pyritic sand with subanyular to subrounded quarty. and pink garnct grains, grading to sandstone and sand with some interbeds of fine-grained cal- careous sandstones, The quartz grains were more or less rounded, with pitted surfaces; pink garnet was common und pyrite usually present with accessvries of muscovite, feldspar and lithic grains. A few foraminifera, cither Anwnodiscus or Hemidiscus, were present in most samples between 540 and 1,050 feet (164 ancl 320 m.). Without the evidence of foraminifera these sediments would be regarded as “fluvio-glacial”. In Donna No. 1 Well there is less evidence of boulder clay. while the sands for the most part are worn and of reasonably uniform size. The section therefore corresponds to the Lower Sakmarian. marine sediments in the Yorke Peninsula bores. Owing to drilling difficulties good bottom cores were not ubtained to prove the base of the sequence, Sediments of possible Permian age with a species of Trochammina not known from other Permian sediments in South Australia occur below 255 feet (77 mm.) in a water bore at Blackford on section 500, Hundred of Murrabinna. GREAT ARTESIAN BASIN Western Part — Lake Phillipson and Boorthanna Troughs Following the recognition of 2,000 feet of Permian sediments in Lake Phillipson Bore (Balme, 1957), the sequence of Lower Permian rocks inder- lving the western part of the Great Artesian Basin, was described by Ludbrook (1961), und the history of their deposition and geological sctting reviewed by Woptner (1964). The Lake Phillipson and Boorthanna Troughs between Lake Phillipson and the Peake and Denison Ranges contain the most complete suecession identified hy Balme as Sakmarian-Artinskian, consisting in upward sequence of a maximum of 2,974 feet of glacial till and boulder clay, marine mudstanes and siltstones, overlain by Glossopteris- and Gangamopteris-bearing Freshwater silts and clays with sandstone and coal interbeds now descrihed from the Arckaringa Sub-basin as the Mount Toonudina Beds (Freytag, 1965), All but the uppermost coul beds were recently intersected between 490 fect (149 m.) and the total depth of 2,048 feet (623 m.) in Stuart Rance No. 3 Hore, drilled by the South Australian Mines Department in search of additional under- ground water supplies for Coober Pedy Opalficld. The marine formatiun below the Mount Toondina Beds in this well provided the most significant fauna xo far recovered from the South Australian Permian, Stratigraphic, Wthopical, and palacontological details of this well are in press (Ludbrook, 1967). Except for the Monat Toondina Beds, the Permian sediments in the Lake Phillipson and Boorthanna Troughs have not been formally named. Balme (1964) informally used the names “Lake Phillipson Coal”, “Lake Phillipson Glacials”, in tabulating the stratigraphic pusitiun of pre-Tertiary micro-loras, and the names “Coober Pedy Beds” and “Stuart Range Beds” on a locality snap of Permian localities frou which palynological data were available. Tt is hore PERMIAN. DEPOSITS OF S.A, AND FAUNA 7 propused to formalize the names Lake Phillipson Beds for the glacigenes al the base of the sequence and Stuart Range Beds for the mudstones and silt- stones, marine in part at least, between the glacigenes and the Mount Toarnidina Beds, using Lake Phillipson Bore (Ludbrook, 1961) as the standard subsurface section. The Lake Phillipson Beds ave so named to avoid further duplication of stratigraphic names for the glacigene boulder clays or till at the base of the Permian sequence. It is recoguized that the boulder beds ocvur in outcreps on the Oodnadatta and Warrina 1:250,000 sheets from which a type section might he selected to comply with the Australian Code of Stratigraphic Nomenclature. The relationship of these auterops to the Lake Phillipson Beds or to the Crown Point Formation of the Northen Territory is not, however, firmly established. Scattered oeeurrences of Permian(?) tillite were briefly described and mapped in the Peake and Denison Ranges west of Mount Dutton, 45 miles (7-2 Kim.) north-northeast of Warrina, three miles (4-8 km.) northeast of Duff Creek Siding and three miles (4-8 km.) northeast of Box Creek Siding (Reyner, 1955). In the standard section, the Lake Phillipson Beds occurring between 2,337 feet (718 m.) and 3,140 feet (957 m.) overlie granite basement and consist of the sequence already described (Ludbrook, 1961) of 783 feet (238-7 m,) of grey boulder clay with some limestone and calcareous sandstone, The Lake Phillipson Beds were intersected in Stuart Range No, 3 Bore between 1,660 feet (505 m.) and the total depth 2,043 feet (623 m.) in which the formation consisted of 100 feet (80:4 m.) of pinkish sandstone with scat- tered coarse quartz grains, red feldspar, pyrite, biotite, pink garnet, granite pebbles and 283 fect (86:2 m.) of boulder conglomerate with granite and other boulders of heterogeneous rock types. The formation was also intersected in Coorie Appa anil Margaret Creek Bores (Ludbrook, 1961). The Stuart Runge Beds overlie the Lake Phillipson Beds. So tar they have not been recognized in outcrop, They are 1,613 feet (491-6 m.) thick between 744 feet (226-7 m.) and 2,337 feet (715 m.) and were described ( Ludhrook, 1961) as a sequence of 88 feet (26-8 m.) of bluce-grey mudstones with foramini- fera, overlain by 1,525 feet (464-8 m.} of sandstones and siltstones. Only the lower nit in Lake Phillipson Bore appears to contain foraminifera. The Stuart Range Beds are well distributed in the Coober Pedy-Stuart Range area of the Lake Phillipson Trough and in the Boorthanna Trough. ‘They were intersected between 910 feet (277 m.) and 1,660 feet (505 m.) in E. & W.S. Stuart Range No, 3 Bore where they contained an abundant and persistent fauna of foraminifera with rare molluscs, ostracodes and vertebrate remains. They are represented in samples of grey clay collected from spoil from three bores on Balta Baltana Station on Billa Kalina 1;250,000 sheet, Bulta and Ware l-mile sheets. The depths at which the Stuart Range Beds were intersected are net known accurately, but they are known to underlie the Lower Cretaceous Marree Formation, Jealities of the bores Balta Baltana No. 1 (Ware 5, 5/641/5), No, 2 (Ware 3, 5/641/3), No. 3 (Balta 6, 5/640/6) are shown on Map 8 Billa Kalina (Ludbrook, 1966b). All contained foraminifera which were abundant at locality 5/641/5. In the Boorthanna Trough the beds appear te attain their greatest thickness of 1,903 feet (580 m,) in Beorthanna (Duff Creek) Bore, thinning out to the south to 240 feet (75 m.) in Coorie Appa Bore and rapidly disappearing to the east of a line hetween Boorthanna, Anna Creek and Coorie Appa. 72 MN. , LUDRROOK it may be anticipated from seismic data (Moorcroft, 1964; Freytag, 1965) and the Permian isopach map of Kapel (1966) that over 3,000 feet (914 m, } of similar sediments occur in the Arckaringa Sub-hasin southwest of Oodnadutta and north of the Boorthanna Trough. Northern South Australia: Simpson Desert Lower Permian sections were intersected in two wells, Witvherrie No, 1 and Purni No. 1 of French Petroleurn Company (Australia) Pty. Ltd., located in the north of South Australia on the southern margin of the Simpson Desert. The sequence consisted of an upper unit 41 fect (12 m.) thick of dark shale with coal and a lower unit consisting of 290 feet (88 m.) of sandstone and shale and 241 fect (80 m.) of siltstone and shale with quartzite pebbles (Canaple and Smith, 1965). The upper unit which has been compared with the coal sequence in Lake Phillipson Bore presumably belongs to the Mount Toondina Beds, while the lower unit is regarded by Canaple and Smith as equivalent to the Crown Point Formation. Northeastern Sauth Australia: Cooper's Creek Sub-basin In the Permian of the Cooper's Creck Sub-busin tu the northeast of South Australia gas discoveries have been made on the Gidgealpa and Moomba struc tures. The sequence of three disconformably transgressive units has been named (Kapel, 1966) the Gidgealpa Formation. The Permian sediments were described by Greer (1965) as. sandstones, lignitic shales and coals which accumulated in alternating coastal swamps, brackish lagoonal and shallow-water marine environments, In Gidgealpa No. | Well, the Permian is “underlain by 400 feet of Permo-Carboniferous marine shales and sands”, Kapel (1966) subdivided the Gidgealpa Formation into three subunits; a lower of sandstone, shale and coal; a middle of shale, siltstone and minor coal; a top subunit of sandstone minor shale and coal. Wopfner (1966) divided the Formation into a lower Permian unit of conglomerate ant sedimentary breceia in black or dark grey silty matrix, and carbonaceous shale interbeds, and an upper Permian unit of light grey fine- to very coarse-grained sandstone, dark grey carbonaceons shale and coal seams. ‘the Formatiou therefore has yet to be formally defined. The highly esr- honaceous nature of most of the sequence is in contrast to the low carbon contett of the light grey claystones and sandstones containing foraminifera which occur over the boulder clays elsewhere in South Australia. Upper Permian carbonaceous siltstone, silty sandstone, conglomerate and arkosie crossbedded sandstone with coal and plant remains were intersected in Delhi-Frome-Santos Innamincka Nu, 1 Well in the northeast of the State. EUCLA BASIN Knowledge of Permian sediments in the Eucla Basin is limited to the single becurrence of Artinskian [ossiliferous clay in Nullarbor No, 8 (Yangoonabie) Bore (Harris and Ludbrook, 1966). FAUNA AND AGE OF THE PERMTAN DEPOSITS The fauna contained in the Cape Jervis Beds, the Stuart Range Beds, and their equivalents, is sparse, indicative of a restricted environment of low tem peratures or low salinity. With two exceptions, the faunal assemblage consists PERMIAN DEPOSITS OF S.A. AND FAUNA 73 only of 23 species of arenaceous foraminifera, of which species of Hyperammina, Ammodiscus, and Hemidiscus are the most common and ubiquitous. Hippo- erepinella biaperta Crespin and Reophax thomasi (Crespin) are distributed in small numbers in all the sedimentary basins; Lagenammina ampulla (Crespin ) and Glomospirella nyei Crespin are restricted to the Lake Phillipson trough; Thuramminoides sphaeroidalis Plummer, Ammovertellina (?) glomospiroides Ludbrook sp. nov., Glomospira adhaerens Parr, Tolypammina undulata Parr, Recurvoides wilsoni Ludbrook sp. nov., are well distributed frequently in abund- ance, As the foraminiferal fauna is almost entirely endemic it provides no ieliable basis for correlation. Thurammrinoides sphaeroidalis described from the American Pennsylvanian is known to have a range of Middle Silurian to Permian, while (?)Hyperammina expansa (Plummer) was described from the Middle Pennsylvanian Strawn Group of Texas. The most significant fossil so far recovered is the gastropod Peruvispira in Stuart Range No, 3 Well at 1,285-1,305 feet. This genus, known only from Lower to Middle Permian marine rocks, confirms the Permian age of the Stuart Range Beds determined by Balme (1957) from microforal evidence. In Stuart Range No. 3. Well the single specimen of Pertuvispira was associated with a fragment of ? Contlaria. Other sparse elements of the fauna include astra- cndes and vertebrate remains which have not been identified. SYSTEMATIC DESCRIPTIONS Order FORAMINIFERIDA Superfamily AaMopisCACRA Family ASTRORHIZIDAE Subfamily RaizaMMiINNAE Genus Hiepocrerinetca Heron-Allen and Earland, 1932 Hippocrepinella biaperta Crespin pl. 3, fig. 1 1958. Wippucrepinella biaperia Crespin, 37, pl. ¥. figs. 1-18. Description, Test free, broadly clangate, consisting of a single tube open at both ends, sides parallel, ends constricted, the greatest constriction towards the aperture which is surrounded by a more or less thickened lip, aboral end open, less thickened when it is preserved, but usually broken off;, wall thick, wrinkled, agglutinating, consisting of quartz grains in a siliccous matrix, Dimensions. Hypotype F£379, length 1-05, width 0-25 mm. Type Locality. Oonah, Tasmania; Quamby Mudstone. Material. Hypotype G.S.S.A. F£379;, 51 specimens. Distribution. Minlaton Bore, Stuart Range No. 3 Bore, Fossil localities Ware 5 (5/641/5) and Balta 6 (5/640/6), Nullarbor No. 8 Bore, A.O.C. North Renmark No. 1 Bore, Stratigraphic Range. Lower Permian. Subfamily HirpocREPINnINAE Genus HypexamMMina Brady, 1878 Hyperammina acicula (Parr) pl. 3, fiz. 7 1942, Hyperamminuidey acicula Parr, 105, pl. 1, figs. 4, 5; pl. 2, fig, 4, 1958. Hyperammina acicyla; Crespin, 44, pl. 5, fig, 11; pl. 6, fig. 7 (xynonyrny ), 74 N. Ui. LAUDBROOK Hyperammina acicula is a large tapering species with a thick smoothly finished wall with much cement. Fragments only have heen recovered in South Australia, but they compare with paratype material. Dimensions. Holotype length 11 mm., greatest diameter 1-3 mm., paratype on UWAGD slide 20767 length 11 mm., greatest diameter 1-2 mm.; the figured hypotype GSSA._ F£370 is only L mm. long. Material. UWAGD 20767 five paratypes, 20770 three paratypes, 20771 one paratype, 20772 four paratypes, 195 fragments in South Australian samples, Distribution. Stansbury and Minlaton ores; locality Balta 6 (5/640/6), Boorthanna, Coorie Appa and Stuart Range No, 3 Bores: Goyder Donna No. 1 Bore; Nullarbor No. § Bore. Stratigraphic Range. Lower Permian. Hyperatmina coleyi Parr pl. 3, firs. 89 1942. Hyperammina coleyi Parr, 104, pl. 2. fig. 3; 1958, Crespin. 46, pl. 6. figs. 5, 6; pl. 7, figs, 3-5; pl. 33, figs. 1-2. Hyperammina coleyi is a large species with a subglobular proloculus and a slender tubular chamber of lesser diameter than the proloculus. The wall is thick, firmly but not smoothly cemented as in H, acicula; the surface is rough. Dimensions, ‘The type series ranges in length to 9 mm.,, the proloculus diameter ()-5-l mm., tubular chamber diameter 0-5 mm. Figured hypotype GSSA F¥371 proloculus diameter 0-55, diameter of tuhe 0-45 mm. Material. UWACD 20766 three paratypes, 20769 one paratype, 20770 two paratypes, 20771 four paratypes, 20772 one paratype; 100 fragments in South Australian samples. Distribution. Cape Jervis; Stuart Range No. 8, Stansbury and Nullarbor No. 8 Bores. Stratigraphic Range. Lower Permian. (?) Hyperammina expansa (Plummer) ph. 3, fig, 10 1945. Wyperamminoides expansus Plinmer. 223, pl. 16, figs. 1-8, 1958. Hyperammina expansa: Crespin, 49, pl. 4, figs. 9-12, Short flaring specimens with a thick wall and apparently wide aperture in undamaged specimens belong to the species referred hy Crespin (1958, p. 49) to Hyperamminoides expansus Plummer, As the genus Hyperamminoides has been synonymized by Locblich and Tappan (1964) with Hippocrepina, and Conkin (1961, p. 261) has noted similarities between H. expansa and the micro- spheric form of Hyperammina casteri Conkin the specific identification and affinities of the Australian species must remain in doubt. Dimensions of Figured Specimen. Height 0-4, diameter 0-4 mm. Type Locality. Algerita, Texas, U.S,A,; Strawn Group. Middle Pennsy)- vanian. Material, Hypotype GSSA F372; 11 specimens. Distribution. Stansbury, Coorie Appa, Stuart Range No, 3 and AOC North Renmark No, 1 Bores. Stratigraphic Range. Middle Peunsylyanian; ?Lower Permian. -1 ct PERMIAN DEPOSITS OF S.A. AND FAUNA Hyperammina hebdenensis Crespin pl. 3. figs. 2-6 1958. Myperemmina hebdenensis Crespin, 52, pl. 6. figs. 8-12, Description. Test free, small, broad, straight, consisting of a globular proloculus followed by a nearly straight tubular second chamber; wall agglu- tinating, moderately thin, with fine quartz grains in much cement. ‘The cement is usually siliceous but some specimens have calcareous cement. Aperture a large circular opening at the end of the tube. Dimensions. Hypotype GSSA Ff375 length 0-75, proloculus diameter 0-25; hypotype GSSA F376 length 1-26, diameter 0-6 mm. Type Locality. Hebden, N,S,W., Mulbring Subgroup. Material. 526 fragments. Distribution. Ubiquitous in South Australian material. Stratigraphic Range. Lower Permian. Observations. Vt is not always easy to distinguish H, hebdenensis from H. coleyi when they usually occur as tubular fragments. In this study the two have been differentiated by the wall thickness with relatively little cement in coleyi; the smaller size, relative smoothness of the wall finish and the fact that the tube diameter is only a little less than that of the proloculus in H, hebdenensis. Family SACCAMMINIDAE Subfamily SaccamMMrmNinaE Genus LacenAMMinA Rhumbler, 1911 Layenammina ampulla (Crespin) pl, 3, fig. 14 1958, Pelosina ampulla Crespin, 42, pl. 2. figs. 1-3. Description. Test free, consisting of a single Hask-shaped globular chamber; wall agglutinating, consisting of very ine quartz grains in much siliceous cement; aperture usually oval, terminal, at the end of a short neck. Dimensions. Hypotype length 1-00, width 0-5 mm. Type Locality. Oonah, Tasmania; Quamby Mudstone. Material. Wypotype GSSA F£387, 7 specimens. Distribution. Locality Ware 5 (5/641/5), Stuart Range No. 3 Bore. Stratigraphic Range. Lower Permian. Observations. Conkin (1961, p. 248) noted the obscurity of the affinities of Proteonina Williamson, 1858, Lagenammina Rhumbler, 1911, and Seccammina Sars, 1869. Pelosinu ampulla Crespin possesses the diagnostic features of Lagenammina rather than those of Pelosina as it has been redefined by Loeblich and Tappan (1964, p. C200). Genus Saccammina M. Sars in Carpenter, 1869 Saccammina orca Ludbrook sp. nov. pl. 3, figs. 11-13 Description. Test free, large, solid, consisting of a single globular chamber of somawhat variable shape; it is commonly barrel-shaped with irregular annular constrictions but may be flatly spherical without apertural neck, or Hask-shaped with a broad neck; wall thick, consisting of an inner pseudochitinous layer and 76 N. JI. LUDBROOK an outer layer of quartz grains held together by calcareous cement; extcrior roughly finished; aperture oval, usually without neck, but may be at the end of a broad, short neck. Dimensions. Holotype F384 length 1-8, diameter 1-1; paratype Ff385 length 1-15, diameter 0°75; paratype Ff386 length 1-2, diameter 1-1 mm. , Type Locality. E, & W.S, Stuart Range No. 3 Bore, 1,305 feet 8 ins.-1,327 eet, Material, 368 specimens, Distribution. Stansbury Bore; locality Balta 6 (35/640/6), Boorthanna, Lake Phillipson, Stuart Range No, 3 and Nullarbor No. 8 Bores. Stratigraphic Range. Lower Permian, The specific name is taken from the Latin noun urca, a jar’ or pol. Genus SaccuutNeLya Crespin, 1958 Sacculinella australae Crespin One specimen doubttully referable to this species was recovered from A.O.C. North Renmark No. 1, 3,410-3,415 fect. Sacculinella Crespin has been synonyzn- ized with Brachysiphon Chapman by Loeblich and Tappan, hut there seems to be insufficient support for this view from the material available. Genus ThuramMinomes Plummer, 1945 Thuramminoides sphaeroidalis Plummer pl. 3, figs. 15, 16 1945. Thuramminoides sphaeroidalis Plummer, 218. pl. 15. figs. 4-10; Crespin, 1958. 49. pl. 3. figs. 9-11; pl. 31, figs. 1-2; Conkin, 1961, 243, pl. 17, figs. 1-10; pl. 18. figs, 1-4; pl, 26, figs. 1-3. Description. An amended description of the species. has been given by Conkin (1961). South Australian specimens are always preserved as deHated or flattened dises, The wall is finely arenaccous with agglutinated quartz grains cemeénted by siliceous material in some tests and culeareous material in others. Of 182 tesis examined 88 had calcareous cement; these all came from Stuart Range No, 3 Bore with the exception of one specimen from Ware 5, locality 5/641/5, and three from Stansbury Stratigraphic Bore. The development of protuberances on some of the tests is similar to that on Texas specimens, Dimensions, The species shows a considerable range in diamcter as shown by Conkin (1961, p, 247), with whose measurements (in mm.) those from the different South Australian localities may be compared, Locality 5/641/5 0-6 (1 specimen) Stansbury Stratigraphic Bore )-4-0-5 (14 specimens) Minlaton Stratigraphic Bore 0-4-0°7 (8 specimens ) Buorthanna Bore 4075 (10 specimens) Lake Fhillipson Bore -25-0-63 (4 specimens ) Stuart Range No, 3 Bore Q-5-2-25 (69 specimens) Nullarbor No. 8 Bore 0:5-1-0 (47 specimens) The South Australian specimens are on the whole of medium size. The largest specimens with diameter of 1-75 mm. or more are smooth specimens with pro- tuberances (pl. 3, fig. 16). Wall Siruciure. The problem of the cementing material and the possible replacement by silica of original calcite is not resolyed in the present study. PEHMIAN DEPOSITS GF S.A. AND FAUNA 77 Same tests of both Thuramminvides sphaeroidalis and Hyperammina hebdenensis have calcareous cement in certain samples. Both species occurring in Ware locality 5/641/5 and at depths of 1,225 to 1,367 feet in Stuart Range No. 3 bore have both calcareous and siliceous tests in the same sample, Tests of Thuram- minoides sphacroidalis have siliceous cement at depths of 740 to 1,225 feet and calcareous cement below 1,367 feet in Stuart Range No. 3 Bore, while all the tests in Minlaton, Boorthanna, Lake Phillipson, Stansbury, and Nullarbor No. 8 had siliceous cement. On the other hand, some tests of Hyperammina heb- denensis from Nullarbor No, 8 have calcareous cement, Australian Species of Vhuramminoides. Conkin_(1961, p, 243) has plaved Crithionina tetcherti Parr, which was transferred by Crespin to Thuranminoides, in synonymy with Thuramminoides sphaeroidalis. The two species are, in the opinion of the present writer, specifically and generically distinct from each other, Seven slides. containing some of Pars original type material were kindly lent by the Department of Geology. University of Western Australia, While the holotypes so far cannot be located slides 20768 and 20772 contain topotypes of Crithionina teicherti from Part's original material, Slide 20765 has seven mounted specimens which, with one exception 1-8 mm. in diameter, range from 2.0 to 2+5 mm. in diameter. They are spherical and uncrushed with a loosely cemented wall 0-2 mmm. thick. The otherwise smooth outer surface of the wall is marked irregularly by small openings of the irregular canals contained in the thick wall. The inner surface is strongly and irregularly punctured with larger openings. From the limited amount of material available the genus Oryetoderma Loe- blich and Tappan. 1961 (type species Crithionina rotundata Cushman) is to be preferred for Crithionina teicherti. he test wall is relatively thinner than illat of either the type species or of O. palacoxoica (Conkin), but the spongy wull texture may be seen where there has been limonitic infilling, and the smal] round or polygonal openings on the exterior surface are not characteristic of Thuramminoides. The wall texture differs also from that of South Australian specimens of Thuramminoides sphueroidalis, Oryctoderma teicherti has not se far been recognized in the South Australian Permian. Type Locality, Thuramaininoides sphaeroidalis was described trom the Middle Pennsylvanian lower Strawn Shale, east of San Saba, Texas, U.S.A, Material. Hypotypes GSSA Ff377, FfS78; 162 specimens. Distribution. The species is widely distributed in Stansbury, Minlatun, Boorthanna, Lake Phillipson, Stuart Range No. 3 and Nullarbor No. 4 Bores and locality Ware 5 (5/641/5). Stratigraphic Range, Middle Silurian te Permian. Family AMMODISCIDAE Subfamily AmMMODISCINAE Genus Ammonpiscus Reuss, L862 Ammodiscus oonahensis Crespin pl, 3, figs. 17, 18 1955. Ammorliscus conahensis Crespin, 69, pl. 12. figs. LO, TE, Description, Test free, depressed in the centre, consisting of a small pro- loeulus and lony undivided second chamber closely planispirally wound in ahout 6 yulutions with oecasional growth constrictions, sutures distinct, slightly de- pressed; chamber in last volution broad; aperture at end of second chamber, slightly arched. 78 NN, H. LUDBROOK Dimensions, Hypotype, greatest diameter 0-825, thickness 0-20 mm. Type locality. Oonah, Tasmania; Quamby Mudstone. Material. Hypotypes GSSA Ff402, 403: 215 specimens. Distribution. W.G. Goyder Donna No. 1, A.O.C, North Renmark No. 1, Stansbury, Minlaton, Beach Petroleum Troubridge Shoal No. 1, Boorthanna, Coorie Appa, Stuart Range No, 3 and Nullarbor No, 8 Bores: localities Balta 6 (5/640/6) and Ware 5 (5/641/5), Stratigraphic Range. Lower Permian, Genus Hemmiscus Schellwien, 1898 Hemidiscus balmei Ludbrovk sp. nov. pl. 3, figs, 19-231 Description, Test fairly large, irregularly discoidal, depressed in the centre. consisting of a proloculus and an undivided tubular second chamber at first planispirally wound like Amunodiscus then coiling unevenly so that later whorls lap over the earlier whorls on one side; there are about five volutions in a specimen of diameter 2-4 mm.; growth constrictions may occur at the former positions of the aperture; diameter of second chamber increasing very gradually; aiken distinct. slightly depressed; aperture at the open end of the tubular chamber. Dimensions. Holotype F399 diameter 1-3 mm.; paratype F400 diameter 0-7 mm.; paratype Ff401 diameter 2-4 mm. Type Locality. Stuart Range No. 3 Bore, 1.225 feet 3 ins.-1,246 feet depth. Material. Holotype GSSA Ff399; paratypes GSSA FF400, 401; 212 speci- mens, including 6] topotypes. Distribution. Common in W.G, Goyder Donna No. 1 Well between 600 and 2,073 feet; Stansbury, Beach Petroleum ‘Tronbridge Shoal Na. 1, Boorthawna, Coorie Appa, Stuart Range No. & and Nullarbor No, 8 Bores, Stratigraphic Range. Lower Permian. The specific name acknowledges the contribution made by Mr. B. E, Balme to understanding the South Australian Permian. Genus AMMOVERTELLINA Suleymanoy, 1959 Ammoyertellina(?) glomospiroides Ludbrook sp. nov, pl, 3. figs. 22-26 1988. Ammetertella inclusa Crespin, 72, pl. 18, fig 5 (non Cushman and Waters). Description. Test free, fairly small, consisting of a proloculus followed without any distinct separation by an undivided tubular second chamber of uniform diameter throughout; flatly streptospirally coiled in the initial stages then more or less planispiral; in the final stages the tube swings over and em- braces the earlier coils at Tight angles to its previous plane of coiling. Wall thick, agglitmaling, consisting of quartz grains well cemented in siliceous cement: aperture arched, at the end of the final coil. Dimensions, Holotype greatest diameter 1-0 mm., diameter of chamber 0-25 mm. Type Locality. Stuart Range No. 5 Bore, 1,265-1,285 feet. Material. Wolotype GSSA F396, paratypes CSSA F397, 398; 230 specimens. Distribution. Locality Ware 5 (5/641/5), Stansbury, Minlaton, Boorth- anna, Lake Phillipson, Stuart Range No, 3 and Nullarbor No. 8 Bores. PERMIAN DEPOSITS OF S.A. AND FAUNA 78 Stratigraphic Range. Lower Permian. Observations. This is the species identified by Crespin with Ammovertella inclusa (Cushman and Waters). Unlike A. inclusa the species is not attached nor when sectioncd does it show the planispirally coiled early portion figured by Conkin (1961, pl. 23, fig, 8) or the planispiral zig-zag type of coiling illus- trated by Ireland (1956, text fig, 5, 6-14). The type of coiling differs from that of Ammocertellina in that in the final stage the tubular chamber does not uncoil. The mode of coiling, however, is irregular when compared with that of Glomo- spira or Glomospirella. It appears to represent a style of coiling intermediate between Glomospira and Ammocertellina. In many respects the species is close to Clomospira articulosa Plummer, which coils in a haphazard, but compact, fashion, Genus GLtomospma Rzehak, 1885 Glomospira adhaerens Parr pl, 4. figs. 1-4 1942. Glomospiru udhaerens Parr, p. 103, pl. 1, fig. 2; 1958, Crespin. 70, pl. 13, figs. 6. 7, The South Australian specimens are attached to quartz grains, coiled in Glomospira fashion, the wall is thick, with agglutinated quartz grains im a siliceous matrix, Dimensions. Hypotype F391 diameter 0-75, diameter of tube 0-15 mm, near the end. Type Locality. “Lingula beds”, south side of Minilya River, near Coolkilya Pool, Wandagee Station; Quinnanie Shale. Material. Hypotypes GSSA Ff391, Ff392, Ff£593; 64 specimens. Distribution. Stuart Range No. 3 and Boorthanna Bores. Stratigraphic Range. Lower Permian, Genus GLomosrmeLLA Plummer, 1945 Glomospirella nyei Crespin ph 4, fig. 15 1958, Glomospirelle nyet Crespin, 70, pl. 13, figs, 1-5, Only the specimens from Lake Phillipson Bore are small and within the range of the type series described by Crespin. There seems to be no criterion but size to distinguish the specimens from Stuurt Range No. 3 Bore from G, nye, Dimensions. Hypotype diameter 1 mm., diamctcr of tube 0-15 mm, Type Locality. Cumdlego Crossing, Minilya River, Western Australia; Cundlega Formation. Material. Ilypotype GSSA F£394; 100 specimens. Distribution. Locality Ware 5 (5/641/5), Lake Phillipson and Stuart Range No. 3 Bores. Stratigraphic Range. Lower Permian. Subfamily ToLyPAMMININAE Genus TouyrpamMMina Rhumbler, 1895 Tolypammina undulata Parr pl, 4, figs. 6, 9 1942. Tolypammina undulata Parr, 104, pl. 2, fig, 2; 1958, Crespin, 72, pl. 19, figs. 7, 8. The South Australian specimens are all small, about 0-5 mm. diameter overall, commonly adherent to quartz grains or to tubes of Hyperammina hebdenensis, as noted by Crespin. &0 N. H. LUDBROOK Type Locality. “Lingula beds”, south side of Minilya River, near Coolkilya Pool, Wandagee Station, Western Australia; Quinnanie Shale. Material, Hypotypes GSSA F{380, Ff381; 184 specimens. Distribution. Locality Balta 6 (5/640/6), Stansbury, Minlaton, Boorthanna, Stuart Range No, 3 and Nullarbor No. 8 Bores, Stratigraphic Range, Lower Permian. Genus AmMMOVERTELLA Cushman, 192 Ammoyertella howchini Ludbrook sp. nov. pl. 4, figs. 6-8 Description. Test attached, consisting of a small proloculus followed by a long narrow tubular second chamber which at first winds back and forth in 7ig. zag fashion and later winds irregularly over the surface of attachment: wall agulu- tinated with fine angular quartz grains in a small amount of siliceous ecment; aperture at the open end of the tubular second chamber. Dimensions, Holotype Ff414 diameter 1 mm. x 0-8 mm; hypotype Ff415 diameter 0-65 x 0-625; hypotype F416 diameter 0-9 x 0-8 mm. Size of the tests is variable. The initial portion is usually about 0-5 mm, in diameter, but the lotal area covered by a single specimen may be as much as 2 mm. or more; diameter of tube 0-075 mm. Type Locality. Cape Jervis; Cape Jervis Beds. Material, The holotype Ff414, paratypes Ff415, 416, and 52 other specimens, Distribution, Stuart Range No. 3 and Stansbury Bores: 2 pebble 36 mm. in diameter from Stansbury Bore. is extensively covered with many encrusting individuals. Superfamily LrrvoLacea Family HORMOSINIDAE Subfamily [lonmosininaE Genus Reorpwax Montford, 1908 Reophax subasper Parr pl. 4, fiz. 15 1942. Reophax subasper Parr, 108, pl. 1, fig. 12; 1958, Crespin. 64, pl. 11, figs. 10, 11, South Australian specimens have calcareous cement. Dimensions. Hypotype length 0-8, width 0-4 mm. Type Locality. “Lingula beds”, east of Coolkilya Paddock, Wandagee Sta- tion, Quinnanic Shale. Material, A single paratype on UWAGD slide 20770; hvpotvpe GSSA F389, and 4 other specimens. Distribution. Stansbury and Stuart Range No, 3 Bores. Stratigraphic Range. Lower Permian, Reophax (?) thomasi (Crespin) pl. 4, figs, 16, 19. 20 1958. Lugtonia thomasi. Crespin. 65, pl, 7, figs. 6, 7. Description. Test free, straight, flaring slightly, consisting of as many as six chambers in a rectilinear series; sutures distinct, straight. depressed; wall ver thick, agglutinating, with close set fine quartz grains in a small amount of cake careous cement, leaving very reduced chamber cavities connected by necks; aperture circular, terminal, surrounded by a thickened lip, PERMIAN DEPOSITS OF S.A. AND FAUNA 81 Dimensions. Hypotype (incomplete) length 1-2, width 0-5 mm. Type Locality. 14 miles southwest of Nalbia Dam, Wandagce Station, Western Australia; Bulgadoo Shale. Material. Wypotypes Ff389, Ff440; 30 specimens. Distribution. A.O.C. North Renmark No. 1, Stansbury, Minlaton, Coorie Appa, Stuart Range No, 3, Nullarbor No, 8 Bores. Observations. Conkin (1961, p. 276) has discussed the anomalous position of this reophacid species in the genus Lugtonia Cummings, which was placed by its author in the Earlanditdac. Crespin (1958, p. 35) states that the Western Australian specimens of R. thomasi have siliceous cement; South Australian specimens have calcareous cement, In thin section no evidence can be seen of the domed chamber form of Lugtonia, although most of the tests are poorly pre- served and distorted. The thin section figured (pl. 2, fig. 20) shows the unustially thick wall and reduced chamber cavity, while the nature of the initial part of the test iy somewhat doubtful. For these reasons the geuus Reophax is used with caution until sections can be made of a wider range of undistorted specimens, Reophax tricameratus Parr pl. 4, fig. 14 Five specimens, some complete, occur in Nullarbor No. § Bore, 1,280-1,290 feet. The species is large; the wall of South Australian specimens consists of coarse quartz grains in calcareous cement, Dimensions, Hypotype length 2, diameter 1-4 mm. Type Locality. “Lingula beds”, south side of Minilya River, near Coolkilya Pool, Wandagee Station, Western Australia; Quinnanie Shale. Material. Paratype on UWAGD Slide 20772, hypotype CSSA FY390; five specimens. Distribution. Nullarbor No. 8 Bore, 1,280-1,290 feet. Stratigraphic Range. Lower Permian ( Artinskian ). Family LITUOLIDAE Subfamily ITapLorpuRAGMOIDINAE Genus Recurvomrs Harland, 1934 Recurvoides wilsoni Ludbrouk sp. noy. pl. 5, figs. 1-10, 15 Description, Test free, subovate to flatly globose, asymmetrical, strepto- spiral; first whorl is planispiral but later whorls in different planes; periphery more or less oval, lobulate, chambers mumerons, about seven in the final whorl increasing very gradually in size as added, inflated unless flattened during fossilization, with pseudochitinous lining; sutures distinct, depressed, nearly straight, wall coarsely arenaceous with coarse angular quartz grains in a siliceous matrix; aperture interia-areal apparently a small low arch, with a lip. Dimensions. Uolotype F403, greatest diameter 0-60, thickness 0-4 mm<¢ paratype Ff407, greatest diameter 1-25, thickness 1:0 mm.; paratypes are in the range of greatest diameter 0-75-1-25 mm. Type Locality. Cape Jervis, Cape Jervis Beds, Muterial. The holotype Ff405, paratypes Ff406-F(413, Ff441; 348 speci- mens. 52, NN. H. LUDBROOK Distribution. Cape Jervis, Balta 6 (5/640/6); Stansbury, Minlaton, Beach Petroleum Troubridge Shoal No, 1, Boorthanna, Stuart Range No. 3 Bores, Observations. The genus Recurvoides has not previously been recorded from rocks alder than Oligocene, The streptospiral test of Recurvoides wilsoni is very variable in shape and when fully developed is most closely related to the Recent R. turbinatus Brady, R. contortus Earland and &. trochamminiforme Héglund, The avis of coiling in R. wilsoni as in B. trochamminiforme continues to change throughout the development of the test. Immature specimens are planispiral and similarly coiled to the genus Adercotryma, Both &. contortus and R. trochamminiforme inhabit cold waters or fjords. The relatively abundant numbers of B. wilsoni at Cape Jervis in beds deposited in an over-deepened yalley or fjord appear to parallel the occurrence of R. trochamminéforme in Gullmar Fjord. R. contortus occurs off South Georgia and in Antarctic waters. Species of bath Adercotryma glameratum (Brady) and Recurvoides turbi- natus have been described from Arctic waters. Both species occur together at several stations off Greenland, at depths of between 31 and 91 metres and also off Humboldt Glacier at 201-2 metres (Loeblich and Tappin, 1953), Sectioning of the test shows (pl. 5, fig. 15) the interior of the chambers to be conspicuously lined with pseudochitinous material. FR, obsaletum (Goés) and R, higeinsi Bronnimann from the Crise and Lengua Formations of Trinidad have been described (Bronnimann, 1953, p. 98) as having the interior of the chambers coated with a brown or thin ?chitinous film. Specifie Name. The species is named for Mr. R. B. Wilson of Geosurveys of Australia Ltd. who collaborated with the writer in selecting and describing the type section of the Cape Jervis Beds, Subfamily Lrrvotinar Genus AmMopacutrrrs Cushman, 1910 Ammohaculites woolnoughi Crespin and Parr pl, 4, fiz, 17 ID41. -Ammokaculites woolmwughi Crespin and Parr, 304, pl. 12, fius, 2a. Th. 3a, bh; 1958. Crespin. 75, pl. 14, figs. 10, 11 (synonymy). The three specimens from South Australia are poorly preserved and identi- fication is doubtful. One specimen was recovered from Minlaton Bore, 554-560 fect, and one from Stuart Range No, 3 Bore at depths of 1,205 feet 5 ins. - 1,235 feet 5 ins. and 1,265-1,285 Feet, Family TEXTULARIIDAE Subfamily Trexrunagusar Genus Textutania Défrance in de Blainyille, 1824 Textularia bookeri Crespin pl. 4 fic, 18 141. Texbularia eximia Crespin and Parr, 305, pl, 13, figs, Tac. Su-c: 1947, Crespin, pl. 2, figs. 23, 24 (non Bichwald). 1956, Textularia “exima Eichwald” of Orespin ard Parr, Cuuitnings, 214. 1058. Vertularia howkeri Crespiu, 77, pl, 15, fiys. 1-7. South Australian specimens are usually distorted during fossilization and the arrangement ot the carly chambers is difficult to determine. Dimensions. Hypotype Ff383, length 1-1, width 0-$ mm. PERMIAN DEPOSITS OF S.A, AND FAUNA a3 Type Locality. Four chains west of Farley Railway Station, Hunter River District, New Sonth Wales; Dalwood Group. Material. 40 specimens. Distribution. Locality Ware 5 (5/641/5), Stansbury, Boorthanna and Stuart Range No, 3 Bores. Stratigraphic Range. Lower Permian. Family ATAXOPHRAGMIIDAE Subfamily GLOBOTEXTULARINAE Genus Diertmxa Crespin and Parr, 1941 Digitina recurvata Crespin and Parr pl. 4, figs. 10-12 1941, Digitina recurvata Crespin and Parr, 306, pl. 13, figs. 9a, b, 10a, b: 1958, Crespin, 79, pl. 15, figs. 10, 11, Description, Test free, moderately elongate, flaring, of moderate size, cou- sisting of a minute trochospiral stage with about 4 chambers in the initial whorl, followed by a triserial portion with about 9 inflated globular chambers increasing, fairly rapidly in size, then becoming irregularly biserial and finally uniserial. Sutures distinct, depressed; wall coarsely agglutinated with coarse angular quartz grains in a thin matrix; aperture interio-areal, a very small arched slit near the base of the apertural face. In the figured hypotype (F{418) and two other specimens from Stuart Range No. 3 Bore, in which the uniserial stage has de- veloped, the aperture is subterminal, as it appears to be in the holotype (Crespin and Parr, pl. 13, figs, 9a-b). Dimensions. South Australian specimens are relatively short and broad, adult specimens from Stuart Range No, 3 Bore attaining a length of 1-0, width 0-7 mm; most Cape Jervis specimens come within the range length 0-4-0625, width 0°325-0-45 mm, Type Locality. Foot of Victoria Pass, Mitchell Highway; Capertee Group, Material. 29 specimens. Distribution. Cape Jervis, Stuart Range No, 3 Bore, 1,285-1,367 feet. Observations. The genus Digitina has been retained here only because no specimens of Mooreinella Cushman and Waters, 1928, are available for com- parison. Cummings (1956, p. 214) suggested the possibility of the synonymy of Digitina with Mooreinella which was rejected by Crespin (1958, p. 36) on the basis of apertural differences. Although the aperture is usually, as Crespin emphasizes, at the base of the final chamber, in fully adult specimens with a uniserial final chamber, the aperture is, like that of Afooreinella, subterminal. However, no description of Mooreinella mentions the triserial stage, through which Digitina passes so conspicuously that the immature specimens from Cape Jervis with poorly preserved initial portions appear to belong to Eggerella (pl. 4, figs. 11, 12). Phylum MOULLLUSCA Class GASTROPODA Genus Peruviserra Chronic, 1949 The single minute inconiplete specimen 1 mm. diameter (pl. 5, fig, 13) is not sufficiently well preserved for its specific identity to be established, but with its veneric range of Lower to Middle Permian, it is an important element im the fauna, The specimen was recovered from 1,285-1,805 feet in Stuart Range No. 3 Bore. &+ N. Tl. LUDBROOK Gastropod indet. A minute stemkern of a protoconch also occurred in Stuart Range No, 3 Bore at 1,327-1.349 fect. Phylum COELENTERATA Class SCYPITOZOA Genus ? Coxunanta Sowerby, 1891 The fragment figured (pl. 5, fig. 14) of a pyritized cast appears ta be that of a Conularia, from Stuart Range No. 3 Bore, 1,327-1,349 fect, OTHER ORGANIC REMAINS Two species of ostracodes (pl. 5, figs. 11, 12) were recovered from Stuart Range No. 3 Bore, 1,327-1,349 feet, together with vertebrate teeth and bones. ACKNOWLEDGMENTS The writer is indebted to Professor R. T. Prider, University of Western Australia, for the loan of the Parr slides, and to Miss Ruth Todd, of the United Stafes National Museum, Washington D.C., and Dr. G. R. J. Terpstra of the Bureau of Mineral Resources, Canbarss, for making comparative material avail able during visits to their institutions, Dr. J. M. Dickins of the Bureau of Mineral Resources, kindly supplied casts uf Peruvispira allundalensis. The assistance of Mr. W. K. Harris and Miss L, D. Linke of the Palaeontology Section, Geological Survey of South Australia, in photographing the specimens is grate- fully acknowledged, REFERENCES Baume, B. E., 1957. Upper Palaeozoic microfloras in sediments from the Lake Phillipson Rore, South Anstralia. Aust. J. Sei, 20 (2), pp. 61-62. Bator, B. E.. 1964. The palynological tecord of Australian Pre-Tertiary floras. Ancient Pacific Floras. Univ. Hawaii Press, pp. 49-80, Broxyimann, P. 1953. Arenaecous Foraminifera from the Oligo-Miovene of Trinidad. Contr, Cushinan. Fdn. foramin, Res, 4 (3), —pp. 87-L00. Campana, B.. and Winson, BR. B.. 1953. The geology of the Jervis and Yankalilla Military Sheets. Rep. Inyest. geol. Surv. S. Anst.. 3, pp. 1-26. pls. IX. Campana, B.. and Wiisow, R, 8. 1985. Tillites and related glacial topography of South Australia, Eclog, geol, Hely., 48 (1), pp. 1-30. pls. I-IV. Canapte, J., and Sanity, L., 1965. The Pre-Mesovoic Geology of the Western Great Artestaiy Basin, Australas. Petrol, Explor. Assoc. APEA J,, pp. [67-LL0. Comman, J. E,, 1961. Mississippian smaller Peraminifera of Kentucky, southern Incliana. northern Tennessee, and south-central Ohio, Bull. Amer, Paleant.. 43 (196), pp. 132-868, Crespin, [, 1958. Pertnian Foraminifera of Australia. Bull. Bur. Miner. Resour. Ceal. Geo phys. Aust., pp. 1-207. Crespin, 1.. anche Park, W. J. 194). Arenaceous Foraminifera from the Permian rocks of New South Wales. J. Proc. R, Soe. N.S.W,, LXXIV. pp. 300-311, pis. XIL-NTIT. Cummanes, R. IT. 1956. Revision of Upper Palacozoic textulariid Foraminifera. Micro- paleontology, 2 (3), pp. 201-242. Damp, T. W. E., and Nowcurs, W., 1897. Notes on the glacial features of the Tnmun Valley, Yankalilla and Cape Jervis District. ‘fYrans. R, Sov; 5, Aust, 21, pp, 61-67. Preytac, LB. 1965. Mount Toondinu Beds — Permian sediments in u probable: piercement structure. Trans. R, Soc. $, Aust.. 89, pp. 61-76, pls, L-LIT. ; CGnren, W. J., 1965. The Gidgealpa Cas Field. Australas, Petrol. Explor. Assoc. APEA J. 1p, G5-GH, Hanns, W._K.. amd Lupssoox. N. H., 1966, Occurrence of Permian sediments in the Fuela Basin. South Australia. Quart. veol, Notes, geol, Surv. S. Aust., 17. pp. 11-14. Thi, D., 1955. Contributions to the correlation and fyuia of the Permian ir Australia and New Zealand. J. geal, Soc. Aust, 2. pp. 84-107. ; Howes, W.. 1895. New facts bearing on the glacial features of Hallett’s Cove. ‘Trans. R, Sov. $, Aust, XIX. pp. 61-69. PERMIAN DEPOSITS OF S.A, AND FAUNA $3 Howcmx, W., 1910. The glacial_(Permo-Carbonifcrous) moraines of Rosetta Head and King’s Point. South Australia, Trans. R. Soc. S, Aust., XXXIV, pp. 1-12. pls, L-XVU. Hower, W., 1926, The Geology of Victor Harbour, Inman Valley and Yankalilla Districts with special reference to the great Inman Valley glacier of Permo-Carboniferous age. Trans, R. Soc. 8. Aust., L, pp. 89-119. Inccanp, H. A,, 1956. Upper Pennsylyanian arenaceous Foraminifera from Kansas. J. Paleont, 80 (4), pp. 831-864. Kaper, A., 1966, The Cooper's Creek Basin, Australas. Oil Gas J., 12 (9), pp. 24-30, Lorpiics, A. R,. and Tappan, IL, 1953. Studies of Arctic Foraminifera, Smithson. Misc. Coll, 121 (7), pp. 1-150. Lorsuicu, A. R., and Tappan, H., 1964, Protista 2. Part C, Treatise on Invertebrate Paleontology. Geol. Soc, Amer. and Kansas Univ, Press. Luvunoox, N, H.. 1956, Permian Foraminifera in South Australia, Aust. J, Soi. 19 (4). pp. 161-162, LuparooK, N. H., 1961. Permian to Cretaceous subsurface stratigraphy between T,ake Phil- lipson and the Peake and Denison Ranges, Trans, K. Soc, 5. Aust., 85, pp. 67-80, Lupproox, N. H., 1966a,_ Minlaton and Stansbury stratigraphic bores. Appendix to Craw- fare, A. R,. The Geology of Yorke Peninsula. Bull. geol, Surv, S. Aust., 39, pp, 83-96, g. 1. Luynnoox, N. H., 1966b. Cretaceous biostratiyraphy of the Great Artesian Basin in South Australia. Bull, geol, Surv. $8. Aust., 40, pp. 1-223, pls. 1-28, figs, 1-34, maps 1-11. Lunprook, N. H., 1967, Stuart Runge No. 8 Bore, Cooher Pedy: Stratigraphy and Micro- palaeontology, Min, Rey. Adelaide, 122 (in press). Mawson, D., 1926, Varve shales associated with Permo-Carhonifcrous glacial strata of South Australia, ‘Irans, R. Soc, S, Aust., 50, pp. 160-162, Moorcrort, E,. 1964. Geophysical Investigation, Mt. Toondina Area. Quart. geol. Notes, geol. Surv. S. Aust., 12, pp. 4-6. Parr, W. J,, 1942. Foraminifera and a tubicolous worm from the Permian of the North-west Division of Western Australia. J. R. Soc, W. Aust.. XXVIL, pp. 97-115. Poumsen, H. J., 1945. Smaller Foraminifera in the Marble Falls, Smithwick and Lower Strawn Strata around the Llano Uplift in Texas. Uniy. Texas Public:, 4401, pp. 209-271. Reyxer, M. L.. 1955. The geology of the Peake and Denison Region, Rep, Invest. geol. Surv. S, Aust., 6, pp. 1-23, pls, J-TV. Spruce, KR. C., 1966, Palaeogeography of the Australian Permian in relation to oil search, Australas. Petrol. Explor. Assoc, APEA J., pp. 17-29, Worrner, H., 1964, Permian-Jurassic history of the western Great Artesian Basin. Trans. R. Soc. S. Aust., 88, pp, 117-128. Worrner, H., 1966. CGidgealpa Gas Field, South Australia. A case history. Australas. Oil Gas J., 12 (11), pp. 29-53. Figs. Fig. Figs. Fig, Figs, Fig. Figs. Figs. Figs. Figs. Figs. Figs. Figs. bo gah ne 4, 1-4. 5, 9. 6-8, N. H, LUDRROOK EXPLANATION OF PLATES Prate 1 Cape Jervis Beds (C.J.) overlying Kanmantoo greywacke (K) at south end af type section, View southerly. View upwards through units 2 and 3 of Cape Jervis Buds to sandstone ledge of viait 4, showing inter-bedded sandstone and clay of anit 2. and boulders in ayers. Southerly view of same part of section a4 Fig. 2, showing dip in lower part of section and bedding of boulders, Striated houlder on tillitie clay at northern end of type section, Prate 2 Gully erosion in Cape Jervis Beds, north end of type section, Erratic in boulder clay overlying Selwyn’s rock (lower left), Inman River, King’s Point, Encounter Bay: boulders on shore at foot of point overlain by boulder till, with sandstone at top right. Burrowings in sandstone at t6p of section, King’s Point, This part of the sequence is doubtlully of Permian age. Pirate 3 All Figures X 30 Hippocrepinella biaperta Crespin, F379, Minlaton 500-530 feet: Hyperammina hebdenensis Crespin, 2, F13T4a, with siliceous cement: 3, FF374h, with calcareous cement, both locality 5/641/5; 4. Ff373. with siliceous cement, Stansbury 845 feet; 5, F375, with calearcous cement, Nullarbor No. & 1290- 1300; 6, M376, with siliceons cement, North Renmark No, L 3755-3760 feet. Hyperammina acicula (Parr), £370, Stuart Range No, 3, 1205.5-1295.5 fect, Hyperammina coleyi Parr. F{371a: b, Cape Jervis CT8. ; Hyperammina expansa Plummer. Ft372, North Renmark No. 1, 3785-3790 feet. Saccamminu orca Imdbrook, 11. holotype Ff384; 12, F386, paratype both from Stuart Range No. 3, 1505 feet 8 ins, -1327 feet; 13, paratype F385, Stuurt Range No. 3, 1827-1349 fect. Fepenavtirtis ampulla (Crespin). F£387, Stuart Range No. 3. 1305 feet 8 ins. - 1827 feet. Thuramminoides spheeraidalixy Plummer, 15, F377, with siliceous cement, Stansbury 830 feet; 16, Ff378, with calcareous cement, Stuart Range No. 3, 1305. feet & ins. - 1327 feet, Ammodiseus oondhensis Crespin. 17, F402, North Renmark No. 1, 3900-3905 feet: 18, FI403, Stuart Range No, 8, 1020-1040 feet, Hemidiscus halmei Vudbrook, 19, holotype. FF399, Stuart Range No, 3. 1225 feet G ins.- 1246 feet; 20, paratype, FHO00, Nullarbor No. 8, 1320-1333 feet: 21. Ff401, Donna No. 1, 870-880 feet. Ammutvertellina glomaspiroides Ludbrook, 232, holotype, FI396, Stuart Range No. 3, 1265-1285 feet, the aperture eun be scen at the top of the Figure; 23. holotype, reverse side; 24. paratype, Ff397. Stansbury, 210-215 feet. 25. 26, two views of paratype F£398, Stansbury, 775 feet. Puiare 4 All Figures X 30 Glomospira adhaerens Parr, 1, 2, two views of Ff89T. Stuart Runge No, 3, 1225-1246 feet, the quartz grain to which the specimen is adhering occupies the central half of Fig. 2; 3, F£393, Stnart Runye No. 3. 1182-1205.5 feat: 4, Ff393, Stuart Range No, 3, 1367-1400 feet. wound around quartz grain. Tolypammina undulata Parr. 5, F381, Minlaton, 500-530 fect. a 1ering to Ryperammina hebdenensis; 9, FE380, Stansbury, 775 fect, adhering to «qnartz grain, Ammovertella howchini Ludbrook. 6, holotype. F#414, Cape Jervis; 7. para- type, Ff415, Cape Jervis, showing aperture at the end of the tube on the right- hand side of the Figure: 8. paratype, Hf416, Stuart Range Ko, 3, 1020-1040 feet. Figs. Figs. Fig. Fig. 10-12. 13. 14, 15. 16, 19, Le. 18, PERMIAN DEPOSITS OF S.A. AND FAUNA 87 Digitina recurvata Crespin and Parr. _ 10, Ff418, Stuart Range No. 3, 1305 feet 8 ins - 1327 feet; 11, 12, Ff417a, b, Cape Jervis. Glomospirella nyei Crespin. F£394, Stuart Range No. 3, 1265-1285 feet. Reophax tricameratus Parr. F£390, Nullarbor No. 8, 1280-1290 feet. Reophax subasper Parr. F388, Stuart Range No. 3, 1349-1367 feet, 20. Reophax thomasi (Crespin). 16, 19, Ff£389a, b, Stuart Range No. 3, 1040- 1060 feet; 20, Ff440, thin section, North Renmark No. 1, 3785-3790 feet. Ammobaculites woolnoughi Crespin and Parr. F£395, Stuart Range No. 3, 1265-1285 feet. Textularia bookeri Crespin, F£383, Stuart Range No. 3, 1060-1120 feet, PLATE 5 All Figures X 30 1-10, 15. Recurvoides wilsoni Ludbrook. 1, holotype, Ff405, Cape Jervis; 2, paratype, 11, 12. 13. 14. Ff407, Stuart Range No. 3, 1182-1205.5 feet, spiral view; 4, Ff407. apertural view, showing large quartz grain on apertural face; 4, Ff409; 7, Ff412; 9, Ff410: 10, Ff411, showing change in plane of coiling; 5, if406, showing apertural lip; 6, Ff408; 8, Ff413_ showing position of aperture; 15, Ff441, Stuart Range No. 3, 1205.5 - 1225.5 feet, thin section, showing pseudochitinous lining of early chambers. Ostracodes, Stuart Range No, 3, 1327-1349 feet. Peruvispira sp., M3180, Stuart Range No, 3, 1285-1305 feet. ? Conularia sp. Cr.3, Stuart Range No, 3, 1327-1349 feet, PLatrr | LupBRooK H. N. 3 o 4 FH. N. N. H. Luprroox a Puare 3 N. H. Lupsroox PLate 4 N. H. Lupsrook PLATE 5 FORAMINIFERA AND STRATIGRAPHY OF THE TYPE SECTION OF PORT WILLUNGA BEDS, ALDINGA BAY, SOUTH AUSTRALIA* BY J. M. LINDSAY} Summary The type section 20 miles (32 km) south of Adelaide has been reexamined. Comprising about 107 feet ( 32'/ m.) of bryozoal calcarenitic impure limestones, sands, silts, and clays, it is informally subdivided into three apparently conformable successions of beds, the middle interval characterized by horizons with spicular cherty nodules. Equivalents of these three intervals are present in the Willunga Bore W.B.1, and in bores in the Adelaide Plains Sub basin. The top of the Eocene Aldingan Stage is represented by the top of a hard grey marker bed at the base of the siliceous interval about 45 feet (14 m.) above the base of the formation. Planktonic and benthonic foraminifera indicate that the age of the section ranges from Upper Eocene to Oligocene. Four informal planktonic zones are recognized, using successive extinctions of Turborotalia aculeata (Jenkins), Globigerina linaperta (Finlay), Chiloguembelina cubensis (Palmer) and Guembelitria stavensis Bandy. These zones span Aldingan to lower Janjukian Stages, and can be related to a planktonic zonal scheme recently proposed for New Zealand, FORAMINIFERA AND STRATIGRAPHY OF THE TYPE SECTION OF PORT WILLUNGA BEDS, ALDINGA BAY, SOUTH AUSTRALIA* by J. M. Linpsart [Read 13 July 1967] SUMMARY The type section 20 miles {a2 km.) south of Adelaide bas been re- examined, Comprising about 107 feet (32% m.) of bryozoal calearenitic impure limestones, sands, silts, and clays, it is informally subdivided into three. ap- parently conformable successions. of beds, the middle interval characterized by horizons with spicular cherty nodules. Eqnivalents of these three intervals are present in the Willimga Bore W.B.L, aud in bores. in the Adelaide Plains Sub- basin, The top of the Eocene Aldingan Stave is represented by the top of a hard grey marker bed at the base of the siliceous interval about 45 feet (14 m.) above the base of the formation, Planktonic and benthonic foraminifera indicate that the age of the section ranges from Upper Eocene to Oligocene. Four informal planktonic zones are recognized, using snecessive extinctions of Turborotalia aculeata (Jenkins), Globierine linaperta (Finlay), Chiloguembelina cubensiy (Palnier) and Guembelitria stavensis Bandy.” These zones span Aldingan to lower Janjukian Stages, and can be related to a planktonic zonal scheme recently proposed for New Zealand. INTRODUCTION Rocks referred to Port Willunga Beds comprise a significant part of the Tertiary sequence in the St, Vincent Basin and its sub-basins. Investigation of the type exposure of Port Willunga Beds has therefore been a uecessary step towards an understanding of the place of the formation in sequences elsewhere, especially in the Adelaide Plains Sub-basin. The type séction of the formation at Aldinga Bay, 20 miles (32 km,) south of Adelaide (Fig. 1), forms part of the coastal exposure of the Cainozoic rocks in the Willunga Sub-hasin of the St. Vincent Basin. Following interest in the sequence over a period of at least 75 years by geologists and palaeontologists, it was mapped, described, and formally subdivided and named by Reynolds (1953). The Port Willunga Beds consist of a bryozoal calcarenitic series of variable hard and soft rocks including impure limestones, sands, silts, and clays, with an interval characterized by horizons of siliceous nodules, The beds, which are well exposed in low coastal cliffs around the central part of Aldinga Bav, have gentle southerly dips of up to 3° but are in part slightly folded and faulted (see Fig. 2). They are now estimated to be about 107 feet (32% in.) thick which is close to Reynolds's figure. Following Ludbrook (1936, p. 17), the basal bed is taken to be the thin gravelly sand (sample 119-66 of the present series) which overlies with minor disconformity the green fossiliferous clay * Published with the permission of the Director of Mines. } Palaeontologist, Geological Survey of South Australia. Trans, Roy, Soc. S. Aust. (1967), Vol. 91. 4 J. M. LINDSAY é - 7 as the top unit of Clinaman’s Gully Beds, The Port Willunga Beds are overlain by Pliscene Hallett Cove Sandstone with mild angular unconformity, bed (118-66) included by Reynolds in Port Wilhinga Beds, but now regarded REFERENCE Tor Walury Beds oyterap, Sevenpe riwtetims. Prams WIiKL LONG A Port Hoarlungs, s FINE Te tthe ey 4=n Fig. 1, Locality map and plan showing position of samples taken from stratotype Port Willunga Beds. METHODS Fig. 1 shows the localities at which samples were taken and Fig. 2 their stratigraphic positions. Included are samples. from a pit dug at the site of 129-66, and also tube samples from a “Wacker” hammer hole sunk from the bottom of the pit until stopped in hard sandstone, Washed and unwashed samples and microfaunal preparations, are held in the Palaeontology Section of the Geological Survey of South Australia. The photographs of planktonic foraminifera in Plate 1 were taken with a Leitz Laborlux microscope, using a Leitz Ultropak lighting unit in combination with a 6.5X objective and relicf condenser, Adox KB 12 film was used, and prints were made on Agfa Brovera paper. CORRELATION ACROSS ALDINGA CREEK The exposed Port Willunga Beds are divided into a northern and a southern portion by the sand-covered and alluvium-filled entrance to Aldinga Creek, as shown in Fig. 1, In order to consider the sequence as a whole, it is thercfore necessary to determine the stratigraphic relationship between the northern and southern sections across the distance of more than 800 feet (244 m.) which separates them, There are several reasons why this is not a simple matter, Firstly, some of the beds are observed to vary IJaterally in thickness and litho- logy. As early as 1878, Tate noted their “most diversified character — clays, limestones, and sands rapidly replacing one another in horizontal and vertical 95 A BEDS ' 4 c FORAMINIFERA AND STRATICRAPHY OF PT. WILLUN aWws SIWLIW Oz i) HO0d 4 aN 2 3 0 Did (| ANADOZ 4Yadadn J5y15 NVTVELSNY NYIMOPNY CE NvyoONIOdYV ES = al. JNOZ DINOLINWd — eraiagwenc! srseaqna EUSA BWD Bydadeny elildaiigeyg HIPSHIZE BYPIOJOQUAL fe oi > v 7 f c TEST RAeYS i é ue ._— te SBuBGR? BU 7 Sn rh Piswayzmanbinw 42 oF nS ' " ” ” (3-5 Frequent (6-10 PanLigaeng 1) Le SOARES Buk Wedio “Dp Waist: Very rara (! - 2 specimel Rare Common {1-25 Abundant (25+ wa SINRREOTEEWIN RUN NCEE) Beneaesxenraed SeAPV Ng aaaNNN JTdW¥S TWAWSINI TidWvS | i : WaaH9 YSNICIY 40 Hk va 5 tf ALNOS ASOTOHLIT ; AU deat pai Sood LAO stl VEG Shi PCH] WILLUNGA BEDS CHINAMANS GULLY BEDS’ @ SF 1NGON ALYSHD WLM TAU SIN I “Wea WONIOTW 40 NOILOSS DILYWAVHSVId GaAgANNS La Hen AAI WAI VONIAIY ssoedov TWAYAIN| TAN ys SQ3q VWONNITIM 1HOd 320 NOILOAS 3dAL 30 NO|LYTayYdOD dalsa5ens WRWNK Id WHS dD aS Od) STH13N SetAMOT A? TALI Sessa? Pls vet Cad THT waeut sisdevelngar) Sua rhaysa ka PLEA Beep hdagurs Se ae ID TM meet eadeul t) | ee OE Se — SS > Se | a fh a — ee—_—— —— o——Sal > > Pe tO Sees tee ——$>> er Foraminiferal log, lithology, and stratigraphy of stratotype Port Willunga Beds. Fig, 2. Hil J. M. LINDSAY sequence as shown in Fig. 2. It is likely that the concealed section has beon affected by one or more of these factors,” In addition, matching of the sections immediately north and south of Aldinga Creek is rendered more diffieult by the limited thicknesses available for examination. Glaessner (1951, p, 275), recording the measurement by Dolling in 1949 ot 25 feet (74 m.) of “polyzoal sands and clays” north of Aldinga Creek, and 87 feet (30 m.) of “polyzoal sandy marls” south of it, listed these thicknesses as consecutive and did not discuss the relationship between the two sections, Reynulds did not explicitly state the thickness of beds common to both exposed sections. Tn his Fig. 1, this thickness is apparently drawn as nine feet (2-7 m.). In his more detailed and definitive Fig, 2 it is shown to be little more than two feet (0'6 m.), However, in both of these representations the total thickness of the formation is 111% fect (34 m.). Glaessner and Wade (1958) produced from these data a composite section with the same total thickness. At the time that Ludhrook and Lindsay (1966) recorded their preliminary notes.on the range of Globigerina linaperta and the extent of the Aldingan Stage within the formation, the writer accepted Reynolds's correlation of beds either side of Aldinga Creek as shown in his Fix, 2.’ However, the ranges of Turboro- tulie aeuleata in the northern and southern sections suggested that more of the heds than he indicated in Fig, 2 are common to the two sections. Later in 1966, during 2 discussion with Mr, W. Stewart of the Geology Department, University of Adelaide, it became apparent that further measurc- iments should be made to check the thickness common to both sections. In a brief stadia survey, several marker beds near Aldinga Creek were traced by the writer, and the results are presented slightly diagrammatically in Fig, 2. The green clay with white limy nodules, which was used as‘a marker bed by Reynolds and from which his sample 4.21) and the writer's sample 182-66 vere taken, is matched with one of the green clays north of the creek as shown. Both beds contain the uppermost occurrence of Turborotalia aculeata in their respective sections, This correlation is considered to be reasonably consistent with the lithological, structural, and microfauna] data, but it is certainly desir- able that the depth to Chinaman’s Cully Beds should he proved south of Aldinga Creek hy drilling. The type section of Port Willunga Beds as now re-measured, is therefore approximately 26 feet (8 m.) thick north of Aldinga Creek, and 89 feet (27 m_) thick south of it, with nearly 8 feet (24 m.) of overlap between the secticms, yielding a total thickness of about 107 feet (324 m.). extension” Secondly, minor fulding and faulting sre apparent in parts of the LITITOLOGICAL. UNITS Reynolds, in his Fig. 2, distinguished 29 lithological units in the type section, His basal bed, a green fossiliferous clay, is now considered to he the top unit of Chinaman’s Gnily Beds, The remaining 28. units may be grouped into three lithological subdivisions which are recognizable elsewhere in the Willunga and Adelaide Plains Sub-basins. For the present, these three subdivisions ate simply referred to as the lower, middle, and upper sequences, but [ture work may show that they merit formal stratigraphic status. Numbering Heynolds’s 28 units from the bottom up, the lawer sequence contains units | to 12 and is by present measurement aud correlation about 43 FORAMINIFERA AND STRATIGRAPHY OF PT. WLLLUNGA BEDS a7 fect (13 m.) thick. It consists of a thin basal gravelly sand fallowed by cream cross-bedded bryozoal calearenitic sandy limestones, calcareous sand- stunes and sands, green-grey clays (often with white limy nodules) and fawn, brown, or pale grey, impure calcarenitic limestones, silts and marls, Tris variable succession extends up to the base of the first hard bed which has some silicvous vernentation — the “hard consolidated grey marly bed ta be seen in the caves below Port Willunga” (Reynolds, Fig. 2), The middle sequence consists of most of the remainder of the type section, up to and including the lowee part of unit 26. It is ahout 52 fect (16 m.) thick, und is characterized by the sporadic development of a hard spicular cherty phase typically oecurring as hands of fawn and grey fossiliferous cherty nodules in saftey [awn-te pale grey impore limestones, silts, and mmarls, all calearenitic and sandy, At the base of this interval is the hard marker bed noted above, which has same siliccous phase but ix mostly limestone to caleareous siltstone and sandstune. The base of the lowest bed with prominent siliceous nodules is from 4 ta 5 feet (1-3 ta 1-5 m.) above the top of the marker bed. ‘This middle, silievous, wterval in Port Willunga Beds is distinct from the alder siliceous beds developed in Rlanele Point Banded Marls, The two are separated by about 107 feet (32% m.) of Blanche Point Soft Marls, Chinaman’s Gully Beds and Aldingan Fort Willunga Beds in the exposures at Maslin and Aldinga Bays. The npper sequence comprises the upper part of unit 28 to the top ot the exposed seetion, [t is about 12 feet (34 m-} thick, consisting, of yellow-brown, fawn, and pale grey byrozoal beds including hard limestone bands, softer imoure himiestavies. silty sands, and at the top of the exposure cross-bedded calcarenitic Siinclstomes. Reynolds did not define the top of Port Willunga Beds bul simply nnted that the top of the formation is not reyealed in the type section. ‘Three and a half miles (5% kro.) inland, the Wilhinga Bore W.B.L. (Clacssner and Woudard, 1958: Ludbrouk, 1956; Lindsay, 1966) penetrated, above the stratigraphic level of the top of the type section, more than 160 fect (49 m,) of calcareans sands, sandstones, and santly limestones, which are evidently also Port Willunga Heds. They extend up beyond the level of Janjukian (Oligocene) beds at the top of the type section inte beds which are Longfordiun (Lower Miocene). Thr three lithological subdivisions of the type section can also he recognized in broad outline in the bore, despite changes there in lithotacies due to depesition under more marginal and restricted couditions. Calearenitic limestones, sands, and clay, which may be included within the scope of Port Willunga Beds, ure widespread in the St. Vincent Basin (Glaessner and Wade, 1958). Im the Adelaide Plains Sub-basin Miocene beds up to Bal- combian Stage, including the Manno Para Clay Member, are known trom bores (Lindsay and Shepherd, 1966), and in one locality the presence of Heterolepa victoriensis (Chapman, Parr and Collins) suggests Bairnsdalian Stave (Lindsay, 19635), It has been recently demonstrated (Lindsay. 1967) that equivalents nf the three lithological subdivisions of the type section are recognizable in deeper bores in the Adelaide Plains, The middle, siliccous, interval is well-developed al least as far north as Hundred of Dublin, but the lower interval is ditheult to separate from Blanche Point Soft Marls, and Chinaman’s Gully Beds usually camot be distinguished. The whole succession of Port Willingu Beds attains a maximum known thickness of 874 fect (267 m.) in the Croydon Bore, where, in the lower part of this succession, equivalents of the type section are 440 feet (125 m,) thick, Oe ]. M. 1A4NDSAY CORRELATION WITH AUSTRALIAN AND NEW ZEALAND STAGES After tracing the usage of the stage name Aldingan, Ludbrook and Lindsay redefined the term “in the restricted time-rock sense as representing the time iuterval required for the deposition at Aldinga and Maslin Bays of the Torta- chilla Limestone, the Blanche Point Marls, the Chinaman’s Gully Beds, and the Jower half of the Port Willunga Beds*. The upper boundary of the stage was drawn at a level of natural subdivision, invelving both an important micru- faunal event — the top of the range of Globigerina Tinaperta —and an important lithological development — the commencement of the middle, siliceous, interval of Port Willunga Beds, The end of the Aldingan Stage is now further defined as heing represented by the top of the hard grey marker bed which is the basal unit of the middle interyal as described above. By prescut measurement and correlation the boundary is at approximately 45 feet (14 m.) above the base of the formation. Raggatt and Crespin (1955), followed more recently by Carter (1964), have restricted the Janjukian Stage in a time-rock sense to xepresent the time interval required for the deposition of the Jan Jue Formation in the Bell's Head- land-Torquay area of southern Victoria, The most diagnostic planktunic event at present available to link the type sections of Jan Jue Formation and Port Willunga Beds, is the extinction of Chiloguembelina cubensis which takes place near the base of the former aud near the top of the latter, This event is a feature of Carter's Faunal Unit 4, in the lower episode of the Janjukian. Prom available data on the range of the species’ in both sequences it may be inferred that Jess than 20 feet (6 m.) of Port Willunga Beds, at the top of their type section, are early Janjukian. Most of the middle, siliceous, interval appears to be strictly pre-Janjukian and it is post-Aldingan us at present defined. Further description of the Jan Juc Formation and its planktonic foraminifera will be necessary to further clarify the relation between Aldingan and Janjukian Stages. Comparison of the ranges of several planktonic species in Port Willungs Beds with their ranges as recorded by Jenkins for New Zealand, suggests cor- relation with the Kajatan, Runangan, and Whainguroan Stages as shown in Fig. 2. AGE OF THE TYPE SECTION Eucene age was recently uscribed to the lower, Aldingan, part of the type section by Ludbrook and Lindsay, the writer having recognized in it a foramini- feral succession, apparently unaffected by reworking, containing Clobigerina imaperta and associated Eocene species. Although in carlicr years considered all of Eocene age by Tate (1879, 1899) and ‘late and Dennant (1896), the type section was more recently considered to be of Oligovene to Lower Mincene age. Tlw foraminiferal evidence now available suggests an Oligocene age for the post-Aldingan part of the section, PLANKTONIC ZONES Lithologics and microfannus suggest that stratotype Port Willunga Beds, as indeed most of the marine ‘Tertiary of the St. Vincent Basin, were deposited under conditions of somewhat restricted access from the open ocean, This eu- vironment does not favour the presence of the tropical or temperate planktonic foraminifera which have been chosen to diagnose planktonic zones in standard sequences of comparable age elsewhere. For cxample, of the zonal species of FORAMINIFERA AND STRATIGRAPITY OF PT, WILLUNGA BEDS 99 Blow and Banner (1962), only Globigerina ampliapertura Bolli has been found, Zonal species of Jenkins (1965) not yet enconntered_ inchide Glohoratalia in- conspicua Howe, and Clobigerina brevis Jenkins. Globigerapsis index index (Finlay), key species of the index zone or zouc 2 of Carter (1964, p. 46) fol- lowed by Wade (1964), occurs only very rarcly and as immature specimens. Globigerina angiporoides angiporoides Hornibrook, another of Jenkins’s zonal specics, althmgh very rare, is typical and persistent within the range of G. Linaperta s.str. Above this, however, only occasional doubtful specimens are preserit at the level of the angéporoides angipervides zone in New Zealand, and the zone is thus not suitable for local use. G.-evapertura Jenkins is present, but not well-developed in the type section, Jenkins, however, defined the lower boundary of his evapertura zone by the extinction of G. angiporoides angi- puroides, and as noted above this is nat a suitable criterion at Aldinga Bay. The linaperta zone of Carter (1964) followed by Wade 1964), was equated by them with Carter’s Faunal Unit 8, which is characterized by the microfauna ol the upper part of the Castle Cove Limestone, and the “Lower Glen Aire Clays”, containing “a form of Globigerina linaperta with swollen chambers” (Carter, 1958, p. 21). Examination of available material leaves little doubt that this is C. angiporoides ungiporoides, and not G. linaperta s.str, Jenkins records similar upward ranges for both G. linaperta and Globigerapsis index index in New Zealand. Faunal Unit 3, or the “linaperta zone” in this sense, has not been used in the present study due ta some uncertainty as to the planktonic content and sttatigraphic position of the beds used to define it. Jenkins defined a different zone of C. linaperta in the Upper Eocene of New Zealand, between the extinction of Globorolalia inconspicua and the initial appearance of Globigerina brevis, Neither of these species is known from Port Willunga Beds but his zone is adapted for local use as described below- The most useful characteristic of Carter's Faunal Unit 4, the final appear- ance of Chilozuembelina cubensis, is utilized as the upper limit of a zane of C, cubensis which follaws the zone uf G, Iinaperta, Guenrbelitria stavensis is asso- ciated with C. cubensis at the level of Faungl Unit 4 and ranges a little higher. This relationship is used to define a zone of GC. stavensis, which occurs at the top of the type section. Despite their usefulness at a certain stage in the development of Australian Tertiary stratigraphy, Carter's Faunal Units or Zones are for various reasons proving unsuitable or difficult to use (at least in the Eovene and Oligocene) as a framework for planktonic zonation. The recent work of Jenkins im New Zealand has provided the basis for an alternative approach and it is therefore proposed to define from stratotype Port Willunga Beds, informal local plank- tonic zones which may be nseful within the St. Vincent Busi, and may also be velated to zones elsewhere via more diverse planktonic sequences from southern Australiz and New Zealand, The Jowest such zone in the type section is that of Turborotalia_aculeata (= Globorotalia inconspicue aculeata Jenkins) which extends below Purt Wil- tunga Beds, and whose upper limit is marked hy the top of the range of the species, Jenkins ceenrded the same extinction level for h of the forms he regarded as subspecies of G. inconspicua. if 7. aculeata has a similar range in South Australia, the basal 22 feet (6-7 m.) of Port Willunga Beds containing the species, correlate with part of ihe Kaiatan Stage of New Zealand at the top of the zone of G. inconspicua. Species associated with 1’. aculeata in the aculeata zone at Aklinga Bay include Globigerapsis index tudex, Globigerina 100 J. M, LINDSAY lineperta, G. angiporovides angiporoides, G. anpliapertura, Turkorotalia incre- beseens { Bandy), Chiloguernbelina cubensis, and Cassiverinella sp. cf. C. chipe- lensts (Cushman and Ponton). Below Port Willunga Beds, the sequerice at Aldinga and Maslin Bays is not at present known to have other planktonic events suitable as a basis of zonation until Hantkenina alabamensis compressa Parr is encountered in the lower part of Blanche Point Transitional Marts, A zone of Clobigerina linaperta has its lower boundary defined by the top of the local range of Turhvrotalia aculeata and its wpper by the top of the rouge ol GC, linaperta s.str. This is comparable with the G. Iagperke zone of Jenkins in the upper Kaiatan and Runangan of New Zealand, but there he shows the final appearance of G. linwperta to be contained within the basal part of his brevis zane which is not yet recognized in Australia, Species associated with G. lina- verla in this adapted linaperta zone inelude Chiloguembelina cubensis, Casst- gerinella sp, ct ©, chipolensis, Turborotalia increbescens, Globigerina aniplia- pertuea, G. angiporoides angiporvides, and, at the top, Guembelitria stavensis apd Globigerina cuapertura, The zone of Gy linaperta is present in the upper- most part of the Aldinyan. The zone of Chiloguembelina eubensis has its lower boundary defined by the top of the local range of G. linaperta and its upper boundary by the final appearance of C. cubensis. Species associated with the zonal specivs include Guembelitria stavensis, Cassigerinella chipolensis, Globigerina enapertura, and very rare and doubtful specimens of G. engiporvides angiporoides. Also occur- ring towards the top ot the zone in the type section are Globigerina bulloides dOrbigny, G. sp. ef. G. ciperoensis cipéroensis Rolli, G. lablacrussata Jenkins, and Globorotaloides. testarugosa (Jenkins). A specimen of Globanomalinu sp. of, G, naguewichiensis (Myatliuk) was recovered from the top of the zone, The zone of C, cubensis is pust-Aldingan, and its uppermost part, at Jeast, is early Janjukian. ‘Ihe zone is likely to be synchronous with most of the brevis zone, the whole of the angiporoides angiporvides zone, and the basal part of the euapertura zone of Jenkins. The zone: of Guembelitria stavensis has its lower boundary defined by the top of the local range of Chilozuembelina cubensis, and its upper boundary by the final appearance of the zonal species, the latter event almost certainly occurring stratigraphically higher than the top of the type section of Port Wil- linga Beds, In the Willimga Bore W.B1, and generally in the Adelaide Plains Sub-basin, G. stavensis ranges up above C. cubensis (Lindsay, 1968, 1967, cited above). At the top of the type section the zunal species is associated with Cassigerinella chipolensis and Globieerina bulloides. The zone would no doubt eceupy ao interval in the lower part of Jenkins's evapertura zone (though Jenkins does not record G, staversis), in uppermost Whaingaroan and perhaps basal Duntroonian. FORAMINIFERA PLANKTONIC Species Throughout the type section, planktonic species are mostly small and re- stricted im variety, although at tmes abundant, No reworking is apparent, The local rauges of more significant species are plotted in Fig, 2, Globigerina praebulloides (many comparable with subsp, leroy! Blow and Banner), G, angustinmbilicata Bolli, G. officinalis Subbotina, and G, ouachittaensis Howe and Wallace, range through the sequence, comprising an association of stnall, apparently tolerant, species related to the Globigerina bulloides Lineave which was discussed by Wade. They often form the mest obvious and abundant FORAMISIFERA AND STRATIGRAPHY OF PT. WILLUNGA BEDS 1M] planktonic vornponent of the samples examined and appear to be typically presen in Upper Eocene and Oligocene sequences elsewhere. Because of their ong ranges, however, they are omitted from Fig. 2. Guembelitria stavensis Bandy Pl. 1, Fig. 1 1949, Cuembelitria stavensis Bandy, Bull. Am. Paleont., 82 (131), pp, 124-125, pl. 24, figs. aD, 1964, Guembelitria sp.; Wade, Micropaleontology 10 (3). pp, 286-287, pl, 1, figs. 12a-b. The South Australian species of Guembelifria which occurs in the upper- most Eocene and Oligocene, has been compared with the types of C. stavensis kindly loaned by the University of Indiana, At Port Willunga, the species com- reces its range about 7 feet (2 m.) below the extinction of Globigerine linaperta and continnes up to the top of the section beyond the highest occurrence of Jhileguembelina cubensis, this latter part of its range comprising the basal part of the zone of G. stavensis. The species has a wide distribution in the St, Vincent, Murray and Otway Basins, Chiloguembelina cubensis (Palmer) Vi. 1, Figs, 2-3 1934. Giimbelina cubensis Palmer, Mem, Soc, Cubana Hist. Nat, 8 (2), pp. 74-75, teat figs. 1-6 (fide Fllis and Messina, 1940 ef seq, ), 1957. Chilovwembeliza cubensis; Beckmann, Bull, U.S. Natu, Mus, 215, pp, 84, 89, pl. Qt, fiz. 21, text tig, 14 (5-8). (984, Chiloguembelina rosa, Wade, Micropaleontalogy, 10 (3). pp. 286-287, pl. 1, fi. 2. Beckmann (1957) commented on the stratigraphic usefulness of Chiloguem- helina, particularly in samples containing mainly a benthonic fauna, and his remarks are supported by the writer's experience of both Chiloguembelina and Guembelitria in the Upper Eocene and Oligocene of the Murray and St. Vincent Basins in South Australia, Reynolds (p. 129) was the first to note the presence of “Gumbelina” in type Port Willunga Beds, from sample A.114, 18 feet above the base of the formation (Carter, 1955, p. 25). It is now known that C. cubensis ranges almost through- out the section, appearing to approach extinction towards the top in basal Janjukian equivalents, where it becomes very rare and sporadic. The last- appearing specimen is figured. Resides Carter's record of the last appearance of the species itn Faunal Unit 4 low in the Jan Jue Formation at Bell’s Headland (Carter, 1964, p. 42, fiz. 14), Taylor (1966), records C, cubensis above G. linaperta trom Esso Gipps- land Shelf No. 1 Well in his Zonule | which he compares with Faunal Unit 4. Jenkins puts the extinction of C. cubensis within the basal part of his euapertura zone, high in the Whaingaroan of New Zealand. Ue has recently (1966) made this extinction the eleventh in a series of twenty-nine homotaxial datum planes ehosen hy hima tor the Pacific and Trinidad Tertiary. In Trinidad, C. cubensis makes its final appearance in the opima epima zone. The ooeurrence of this datum plane near the top of stratotype Port Willunga Beds is thus of consider- able importance. Although the planes are described by Jenkins as homotaxial and not necessarily isochronous, he accepts the extinction of C. cubensis as an Oligocene event. Globanomalina sp. cf. G, naguewichiensis (Myatliuk) PL 1, Figs. +5 1950, Globizerinella naguewichiensis, Myatliuk, Trad veces. aeft, nauchno-issled. geol-razy. Inst, (VNIGRT), Mikrofaunw $.8.8.8., sb. 4, vyp, 31, 251, pl. 4, figs. 4a-b (fide Ellis and Messina, 1940, et seq.). 102, J. M. LINDSAY 1982. Vee daheucbuertag naguewichiensts; Blow and Banner (University Press, Cambridge). p. id. Loeblich and Tappan (1964, p, © 665) are followed in regarding Pseudo- hastigerine as a synonym of Globanomalina. A solitary small planispiral speci- men with six chambers im the final whorl was recovered from the top of the cubensis zone, It is Glohanomalina, apparently less akin to G. micra (Cole) than to G, naguewichiensix (fide Ellis and Messina, 1940, et seq.). The extinction of Myatliuk’s species, as the last-surviving species of Globanomalina, marks the lower boundary of the hasel Neogene zone N.1 of Banner and Blow (1965 }, Too much importance cannot be attached to a single specimen, bul the vccur- rence does provide some support for the widely-held view that the extinction of G. cubensis oceurred in the Palaeogene and Oligocene. Cassigerinclla chipolensis (Cushman and Pontor)) Pl. 1, Fig, 6 1932. Cassidulina chipolensis Cushman and Ponton, Bull, Fla, St geol, Surv. No, 9, q. 98, pl. 15, figs, 2a-c (fide Fllis and Messina, 1940 ef seq.) W683. Cassigerinella chinolensis; Blow and Banner (University Press, Cambridge), pp, 81, 83, pl. XV (M,N) (synonymy). Ae. poseerinete chipolensis; Wade, Micropaleonlology, 10 (3), pp. 286-287, pl, 1 ig. 22, Specimens with well-developed biserial enrolling and prominent aperture, occur through the cubensis and stavensis zones, The species ranges as ligh as the Baleombian Munneo Para Clay in the Adelaide Plains Sub-basin, and up to Bairnsdalian Pata Limestone in the Murray Basin of South Australia, Cassigerinella sp, cf. C, chipolensis PI. 1, Figs. 7-9 Blow and Banner did not find any forms referable to the genus Cassigerinella in the Eocene of the Lindi area, Tanganyika. However, in stratotype Port Willunga Beds, forms from the upper part of the aculeata zone, and the linaperta zone, are referable to Cassigerinella and are compared with C. chipolensis. Eleven such specimens have been recovered. They usually haye a distinct planispiral early stage, but also display the biserial enrolling of € rassigerinelia, In some examples the aperture tends to be more restricted and slit-like than is the case with C. chipolensis, but the earliest form (pl. 1, figs, 7-8) hag a alis- tinctly open oval aperture. The lowest sample south of Aldinga Creek with C. sp. ct. C. chipolensis (155-66) also contains immature but characteristic Clobi- gerapsis index index besides Turborotalia aculeala, The earliest C. sp. cf. C. chipolensis in the type section is from sample 123-66 north of Aldinga Crévk, and is associated with the latest Maslinella chapmani Glacssner and Wade. Todd (1966, p. 14) hus recently discussed the possibility thal Cassizerinella occurs in the Eocene. The evidence from Port Willunga Beds seems to confirm that it does, Turborotalia aculeata (Jenkins) Pl, 1, Figs. 10-11 1963. Globorotalia inconspicua aeuleuta Jenkins, N-Z. 1, gool. geophys., 8 (8), pp. 1118, 1120, fig. 13, nos. 129-125. Tn the present study, nou-keeled Globorotaliinac are assigned to Turboro- talia following Loeblich and Tappan (1964, p. C 668). This distinctive finely spinose species described from the Bortonian and Kaiatan Stages of New Zealand, is present, at times abundantly, in the basal 22 feet (6% m.) of the type section, in the aculeata zone, Its differences from Cloborotalia inconspicue Howe, noted by Jenkins, appear to sulfice for its transfer from a subspecios of FORAMINIFERA AND STRATIGHKAPHY OF PT. WILLUNGA BEDS 4U3 G. inconspicua to a distinct species in Turborotalia. Both north and south of Aldinga Creek, the highest oceurrence of JT. aculeata is in a green clay with white limy nodules, sipporting the equivalence of these two beds as shown in Fig, 2. The species is known from the Eocene of the St. Vincent, Murray, and Otway Basins of southern Australia. Turborotalia increbescens (Bandy) Pl 1, Fig. 12 1949. Plodigetine increbescens Bandy, Bull, Am, Paleont, 32 (131), pp, 120-131, pl 2% s. dat. 1962. Globorotalia (Turborotalia) inerebeseens; Blow and Banner (University Press, Cam- bridge), pp. 118-119, pls. XUN (T-V). XVII (D, K), tig, 9 (xiii-xv), 15683. Racbaronlie increbescens; Reed, Bull. Am. Paleout,, 49 (220), pp. 86-87, pl 14, vs. 11-13. The species has been formd only in two samples straddling the top of the aculeata zone north of Aldinga Creek. This is somewhat higher than the range recorded by Jenkins from New Zealand, but is within the lower part of the range recorded by Blow and Banner from Lindi (op. cit., Fig. 20). Turborotalia opima continuosa (Blow) Pl. L. Figs. 13-14 1959. Globorolalia opima ciintinuosa Blow; Bull. Ain. Paleont., 39 (178), pp. 218-219, pl. 19. figs. 125a-c, 1963. Cloborotalia continwosa; Jenkins, N.Z. J. gcol. geophys.. § (G6), fig. 2 (range chart), Only two examples have been recognized, both from sample 145-66 at the top of the cubensis zone, from a level near the bottom of the range of the species as recorded for New Zealand by Jenkins. The specimen figured agrees closely with Blow’s diagnosis of the subspecies, which by the criteria of Loeblich and Tappan belongs to Turborotalia rather than to Cloborotalia, Globigerina ampliapertura Bolli Pl. 1, Fig. 17 1057. Globigerina anpliapertura Bolli, Bull. U.S. Nati, Mus, 215, p. 108. pl. 23. figs. 4u-7b, 1962, Globigerina amplianeriure anpliapertuya; Alow anc Banner (University Press, Cam- bridge). pp, 83-84. pls. XT (A-D), XVIL (C), fig. 12b. The species has heen seen only in the Aldingan portion of the type section, mostly in the /inaperta zone, where it is small with a relatively high arched aperture, This is comparable with the basal part of the range of the species in New Zealand as recorded by Jenkins. Globigerina angiporoides angiporoides Hornibraok PL 1, Fig. 25 1965, Globigerina angiporoides Hornibrook, N.Z. J. geal. geaphys., 8 (5), pp. 834-838, figs. 1 (ai), 2 (synonymy). ; 1965, Glubizerina angiperoides angiporvides: Jenkins, NvZ. J. geol, geophys,. & (6), pp. 1092, 1096, fig, 3. (range chart). Through the aculcata zone this form oceurs typically and consistently but mostly very rarely. Tt ranges into the fingperta zone and occasional doubtful examples are present nearly le the top of the evbensis zone. Taylor (op, cit.) recorded this species (as G. angipora Stache) only from his Zonule K of upper- most Eocene age in Esso Gippsland Shelf No. 1 Well, associated with C, Tina- peria. G. angiporoides angiporoides is abundant in the “Lower Clen Aire Clays”, and as was noted above, there is litte doubt that it is the form referred to by Carter as “G. linaperta with swollen chambers”, the characteristic species of his Fauna) Unit 3 or “Jinaperta zone”, 14 J. M, LINDSAY Globigerina bulloides dOrbigny Pl 1, Fig. 16 1941, Globigerina bulloides @Orbigny; Cushman, Conir, Cushman Lab. foramin. Res., 17 (2). pp. 38-39, pl. 10, figs, 1-13 (synonyny ). 1959. Glabigoring bulloides; Blow, Bull. Am, Paleont., 8% (178), pp. 175-176, pl. 9, igs, GSa-e, There has been some disagreement among micropalacontologists over the age of earliest G. bullvides, varying for example from middle Miocene (Blow, 1959, p. 175) to Upper Eocene (Wade, op. cit, p. 278), Jenkins recorded earliest G. bulloides in New Zealand from middle Whaingaroan Stage ( Oligo- cene). and comparable with this, in type Port Willunga Beds, forms attributable to G. bulloides enter high in the cubensis zone and are prominent within the Janjukian interval, Globigerina sp. cf. G, ciperoensis cipcroensis Bolli Pl. 1, Fis, 18 1954, Globizerine ciperaensis Bolli, Coutr. Cushman Pdn. foramit, Res,, 5 (1), pp, 1-2, text fizs, 3, Sa. 4, 4a-b, 5, 5a-b. 6, 1957. Globigerina ciperoensis ciperoensis Boll, Bull, U.S. Nato. Mus,, 215, p. 109, pl. 23, figs. 10a-h, At a similar level to that recorded by Jenkins for New Zealand, the ciperoen- sig form is emerging as an offshoot from the G. qngustiumbilicata population at the top of the cubensis zone in Port Willunga Beds, as it devclops a more highly trochospiral, five-chambered whorl, a slightly hispid test, and 4 more open centrally-situated umbilical aperture tending to Jose its lip. The specimeii most similar to G. ciperoensis ciperoensis is figured, but its umbilicus is still relatively small, and an apertural lip is still slightly developed. Globigerina euapertura Jenkins Pl. 1, Fig. 15 1960. Globigerina evapertura Jenkins. Micropaleontology, @ (4). p, 351, pl. 1, figs. Sa-c. 1962. Globizerina ampliapertura enapertura, Blow and Banner (University Press, Clagnu- bridge). p. 84, pl. XT (E+G), Commencing its range near the top of the Aldingan, G. euapertura oceurs most frequently in the Janjukian interval. The rather low, rimmed, widely- arched aperture and depressed final chamber, are distinctive features. Blow aud Banner discussed euapertura as a subspecies of G. ampliapertura, and showed its emergence from that lineage near the top of their turritilina furri- tilina zone (uppermost Eocene) at Lindi. Similarly in New Zealand, Jenkins records the commencement of the range of G. euapertura near the Runangan- Whainyaroun boundary, At Port Willunga, sample 147-66 from just below the top of the Aldingan Stage contained the specimens figured of the carliest definite C. euapertura and the latest G. ampliapertura seen. Globigerina labiacrassata Jenkins Pl, 1, Figs, 20-22 1965, Globigerina labincrassata Jenkins, N.Z. J, geol. geaphys,, 8 (6), pp. 1102, 1104, 11086, fig: 8 nos, 64-71, A few examples of the species have been recovered from samples 9-67 and 13-67 in the upper part of the cubensis zone, ata level equivalent to the middle of its range in New Zealand. The forms have moderately thickened apertural rims, and variable size and height of aperture, as compared with the morc typical figured specimen from the Otway Basin, FORAMINIFEBA AND STRATIGRAPITY OF PT. WILLUNGA BEDS 105. Globigerina linaperta Finlay Pl, 1. Figs, 23-24 1939, Glahigerina linaperta Finlay. Trans, R. Soc. N,Z,, 69 (1), mp. 125. ph 15, Ags. £4-57. 1962, Globizerina linaperta linapertay Blow and Banner (University Press: Cambridge), pp. 35-87, pl, XI (TL) (synonymy). The apparent environmental tolerance of this species makes it stratigraplie- ally important in Eocene correlation, Blow aud Banner demonstrated its extinction at Lindi at the top of their furritilina turritilina zone, in beds sHll Eocene on the evidence of diagnostic larger foraminifera such as Discocyclina sp, For New Zealand, Jenkins recorded both G, linaperta and Globigerapsix index index as haying become extinct at the top of the Runangan, and accepted this as the Hocenc-Oligocene boundary, McTavish (1966, p. 16) maintained that G. linaperta “persisted into the Oligocene in Australia”, while acknowledging, its geveral extinction at the close of the Eocene, However. it seems likely from present studies that Australian records ot G. linaperta from the Oligocene refer vither to G. angiporoides angiporoides or to forms not conspecific with G. linaperta but comparable with it in some aspects, The specimens figured from Port Willunga Beds show the diagnostic features of the species. and in particular the specimen from sample 148-66, the uppermost recovered, is very comparable with New Zealand material examined, Globorotalsides testarugosa (Jenkins) PY, 1, Fig. 26 1060. Globorotalia testaruzosa Jenkins, Micropaleontology, 6 (4), p. S68. pl. 5. fin Baec. 1085. Cloborotaluides testarugesa: Jenkins. N.Z. J, geol, ycopbys. 8 (6), p. 1092, fig, 2 (range chart), Good examples are present very rarely at the top of the cubensis zone, equivalent to the middle of the range recorded by Jenkins from New Zealancl, He described the species from the hase of the Lakes Entrance Oil Shaft in Victoria at the top of its range. The form has also been recorded by Taylor (op. cit.) from Esso Gippsland Shelf No. 1 Well. The figured specimen shows the characteristic coarsely pitted wall and relatively straight tangential sutures on the spiral side. Globigerapsis index index (Finlay) PI. 1. Fig, 19 1930. Globigerinoides index inlay. Trans. R. Soc. NZ, 69 (1), p. 125, pl. 14. figs, 85-88. 1957. Globigerapsis index; Bolli, Bull. U.S. Natu, Mas, 215, p, 165, pl. 36, figs, 14a-J8b, 1965. Globigerapsis index index; Jenkins, N.Z, J. geol. genphys., 8 (6), pp. LO91, 1094. dig. 2 (range chart). Immature examples with wide single apertures, as shown in the figured specimen, occnr very rarely within the aculeata zone, Jenkins does not record Globigerapsis tropicalis Blow and Banner from New Zealand and ranges hoth G. index index and Globigerinu linaperta to the top of the Runangan. G, index index is evidently more affected by adverse environment than is G. linaperta and its usability as a zonal indicator is thereby lessened. However, its ovva- sional presence within the bottom 16 feet (5 m.) of the type section of Port Willunga Beds confirms the Upper Eocene age of this interval. BeENTIIONIC SPECIES The benthonic foraminifere have not been studied in detail and many of the species are as yet undescrihed. A few of more immediate stratigraphic signifie- ance will be noted here. 1G J. M. LINDSAY An interesting assemblage is prescrit as a wumerically minur constituent of the microfauna in the hasal 20 feet (6 m,) of the type section, within the Upper Encene aculeaia zone, The members of this assemblage are: Crespinina Kings- evlensis Wade, Linderina sp. Hulkyardia sp, cf. 11. bartrumé Parr. Maslinella ehapmani Claessner and Wade, Reussella finlayi Dorreen, and a genus close to Bolivinella, Their sanges within the type section are plotted in Fig. 2. All South Australian specimens of Crespinina kingscotensis recorded by Waile in her description of the species (1955), came from Eocene beds, It has in addition been recorded by Ludbrook from Buccleuch 4, and B in the Murray Basin (1961, Table X), and at Aldinga Bay (1956, p. 17) from as high as basal Fort Willunga Beds, In South Australia it is thus only known as an Eocene species, Linderina sp., and Halkyardia sp. cf. H. bartrymi, have been recorded by Ludhrook, and the latter figured, from Buccleuch A (1961). She has also noted them (1963, pp. 8, 9) from other calcareous sediments in the U pper Eocene of South Australia, Maslinelle. chapmani is ouly known from the Eocene, Glaessnet and Wade (1059) described South Australian occurrences from Upper Eocene beds at Maslin Bay, Port Noarhinga, Kimyscote (Kangaroo Island), and Moorlands, Ludbrook noted the species from Buecleuch A in the Murray Rasin (1961), and in 1963, in Middle and Upper Eocene microfaunas from South Australia, inelud- ing the Eucla Basin. A few examples have now been recovered from. the weuleata zone in type Port Willumga Beds north of Aldinga Creck. A genus closcly related to Bolivinella was recorded and figured by Lud- brook (1961) frown Upper Eocene Ruecleneh A in the Murray Basin, and noted by ber (p. 86) to occur assoriated with Mantkenina alubamensis compresse in Blanche Point Transitional Marls at Maslin Bay. A fow glanconitie internal tasts of the form have now also heen found at the top of Chinaman’s Gully Beds in sample 118-66, and typical examples oceur Jaw inv Port Willunga Beds. In a few of the specimens some coiling of the initial chambers has been ob- served. This species is only known from the Rocene of South Australis. As with the planktonics, these henthonies do not appear to have been reworked from older beds, Occasional specimens show. slight wear, iuter- preted us due tothe kind of contemporaneous abrasion which night he expected in beds such as the basal cross-hedded bryozoal sands. The assemblage noted above thus supports the planktonic evidence far an assignment of Upper Kocene age to the aculeata zone in type Part Willunga Beds, Cihicides pseudaconvexus Parr, 1938, rangus throughout the section but vevurs only oecasionally above the aculeela zone, the uppermust example heing Inund at the lop of the enbensiy zone. The species was reported by Carter to be present as high as Paunal Unit 4 (1958, Table 8; 1964, Table 3), Althongh mifiohds arc scarce in the type section, a single good example of Maysilinu torquayensis (Chapman, 1922), was found at the top of the cubensis zone, within the Janjukian interwal. A comparable finely striate siliceous internal cast was recovered frum low in the cubensis yone, This is a characteristic species of the Jan Jue Formation and its South Australian correlatives. A specimen from the Ettrick Formation has been figured by Ludhrook (1961, Pl I, big, 3). Placed for the present in Notorvtalia, a species which is probahly mew appears in the Janjukian near the top of the eybensis zone, It ranges up into PORAMINIFERA AND STRATIGRAPHY OF PT. WILLUNGA BEDS LaF the stacvensis zone, and elsewhere into the basal Miocene of the Murray aod St. Vincent Basins. More robust, and somewhat Jarger Uian Notorotalia howehini (Chapman, Parr and Collins), it has a more or less pronvinent but flush um- hilical plug. It is quite distinct from Porosorotalia crassimura (Carter), which is present throughout the type section, Pseudopolymorphina rutile parri Cushman and Ozawa, 1930, has only been seen trom the top of the ctibensis zone, It was deserihed from the Jan Juc Formation and occurs sparsely in Janjukian correlatives in the Murray aud St. Vincent Basins. Sherbornina atkinsoni Chapman occurs throughout the type section, pre- ferring calcareous sandy facies, but S$. cuneimarginafa Wade is not present, Notably absent from even the Upper Eocene part of Port Willunga Beds are Asterigerinella aclelaidensis (Howchin) and a distinctive, striate, Pseudo- polymorphina sp. (of Ludbraok, 1961, Pl. 1, Fig. 1), In the St. Vincent Busin the former is not known from above Blanche Point Marls and the latter ranges no higher than the Blanche Point Transitional Marls Member. Victoriella conoidea { Rutten) has not been seen in the type section, and in any case use of the Victoriella conoidea zone (Carter, op. cit; “V. plecte” zone of Glaessner, op. cit.) would be mappropriate for a scheme of planktonic zonation. The cubensis and stavensis zones vt the present scheme are used instead for all but the uppermost part of the zone of Victoriella conoidea, The species has been found by the writer recently for the first time in the Adelaide Plains in the Croydon Bore where it occurs in Port Willunga Beds at 1,040- 1,045 fect (817-319 m.) in pale brownish-prey limestone, 25 feet (7-6 m.) above the top of the stavensis zone, and associated in an upper Janjukian microfauna with Cibicides psuedoconvexus, Clubigerina sp, cf. G. angulisuturalis Bolli, Massilina torquayensis, Sherbornina atkinsoni and Sherborning cuneimarginata. This is above the level of the type section of Port Willunga Beds, and must be close ta the Oligocene-Miocene boundary. ACKNOWLEDGMENTS Grateful acknowledgment is made of guidance by Dr, N. H. Lndbrock, Senior Palaeontologist, helpful discussions with Mr. W. K. Harris, Palynologist, and technical assistance in photography by Mr, Harris and Miss L. Linke. Laboratory Assistant. The drafting was done by Mr, D, C, Sutton, of the De- partment of Mines Drawing Office, Dr. Iudbrock and Mr. Harris read and criticized the text. REFERENCES Bannna, F, T. abd Brow, W. O. 1965, Progress iv the Planktonic Forasniniferal Bio- stratigraphy of the Neogene. Nature, Lond.. 208 (5016). pp. 1164-1166. Becxmamn, J, P., 1957, Chiloguembelina Loeblich and Tappan and Related Foraminifera from the Lower ‘Tertiary of Trinidad, B.W.I. Bull, U.S, Natn, Mus,, 215, pp. 82-46, Buow, W. H., 1959. Age. Correlation, and Biostratigraphy of the Upper Tocuyo (Sau Lorenzo) and Pozon Formations, Eastern Waledn, Venezuela. Bull, Am, Paleont., 39 (178), pp. 59-251, pls. 6-19, Biow, W. H.. and Banner, F, T,, 1962. The Mid-Tertiary (Upper Eocene to Aquitanian) Globigerinaceae. Part 2 in Eames, F. E,, et al., Fundamentals of Mid-Tertiary Stratl- graphical Correlation, viii + 163 pp., pls. LXVIT (University Press, Cambridge}, Canrer, A. N., 1958. Tertiary Foraminifera from the Aire District, Victoria. Bnill. Geal. Surv, Vie., 55, 76 pp,, pls. 1-10. Canren, A. N., 1964. ‘Tertiary Foraminifera from Gippsland, Victoria, and Ther Steati- graphical Significance. Mem. geol, Surv. Vict., 23, 154 pp. pls. 1-17. Crtaeman, F,, 1922. Report of an Examination of Material Obtained fron a Bore at ‘Torquay. 108 J. M, LINDSAY Kev. geol, Surv, Viet. IV (8), pp. 315-324, pl. LL. Eras, B. F., and Messiwa, ANcrnina, 1940. ef seq. Catalogue of Foraminiferuz. Am. Museum Nat. History. Guarssnen, M, F.. 1951, ‘Three Forantiniferal Zones in the Tertiary of Australia, Geol. _ Mag. LXXXVIT (4), pp. 273-283. GiArssNen, M, F., and Wave, Many, 1958, The St. Vincent Basin, 115-126, f@ Gnansssrn tye? and Fastin, L. W. (ed_), Vhe Geology of South Australia, J. weal. Soe, Aust, 5 +), pp. 1-163. Guagssnek, M. F., and Wane. Many, 1959, Revision of the Forammiferal Family Vic- toriellice, Micropaleantolozy, 5 (2), pp. 193-212, Jexxins, D. C., 1965, Planktonie Foraminiteral Zones and New ‘Taxa from the Danian to Lower Miorene of New Zealand, N,Z. J. geol. geophys.. 8 (6), pp, 1088-1126, Jexsrys, D. G.. 1066, Plauktonic Forsminiferal Datum Plancs in the Pacific and Trinidad Tertiary, N.Z, J, yeol. geophys., 9 (4), pp, 424-427- Luspsay, J, M., 1965. Immanuel College, Camden Park, Bore No. 1, Stratigraphy and Micro- paltcontology, Geol, Sury. S. Aust. Pal. Rep, 14/65 (unpmblisbed ), ; Linnsay, J. M., 1966, Stratigraphy ancl Micropalacontolopgy of the Willanga Bore W.B. 1 — a Re-examination, Ceol. Sury. $, Aust. Pal. Rep, 8/66 (unpublished ). Linpsay, J. M., 1967. Adelaide Plains Sub-basin Stratigraphy and Micropylaeontolouy, Summary Report No. 4, Geol. Surv. S$. Aust. Pal Rep, 1/67 (unpublished), Linpsay, ], M., and Sueparep, B. G., 1986. Munno Para Clay Member. Quart. geol. Notes, geol, Surv. S, Aust, Nu. 19. hily, pp. 7-11. Tornutcn, A. R., and ‘Tappan, TY. with others. 1964. Treatise on Invertebrate Paleontology. Lt. C. Protista 2. Sarcodina, chielly “Theeamovbians” and Waraminiferu, 1, 2) axxi, C1-900, (‘The Geological Society of America and The University of Kansas Press, ) Lvpskoox, N. H., 1956. Supplementary Note on Willunga Basin Sediments. Rep, livest, weol. Surv. S, Aust. No. 8, Appendix, pp. 15-18, Lunvuoox, N. H.. 1961. Stratigraphy of the Murray Basin in South Australia, Bull. seal, Surv. &. Anst., 386, 96 pp, pls. 1-VIIT. Lupwnoor, N. H.. 1963. Correlation of the ‘Tertiary Rocks of South Australia. "Trans. H. Soc. $. Aust.. 87, pp. 5-15. Lupunoox, N. Hard Linosay, J. M, (966, The Aldingan Stage, Quart. geol. Notes geal, Surv, 5. Aust.. No. 19, July, 1, 2. MeTavist, R, A,, 1966. Planktonie Toraminifera from the Malaita Group, British Solomon Islands, Micropaleontology, 12 (1), pp. 1-36. Pan, W. J., 1959. Upper Eocene Foraminifera from Deep Borings in King’s Park, Perth, Western Australia. J. R. Sec, West. Aust. XXTV, pp, 69-101. Kaccair, HW. G., ond Crespixn, 1. 1955, Stratigraphy of Tertiary Rocks between ‘Torquay and Eastarn View, Victoria, Proc. R, Sus. Vict, (N.8.). 67 (1), p. 75-143. pls. FV-VUl Reynowps, M. A,, 1953, The Cainozoie Succession of Maslin and Aldingn Rays, South Australia. Trans. R, Soc, $. Aust., 76, pp, 114-140. Tater, R. 1878. Notes on the Correlation of the Coral-bearing Strata of South Australia, with a List of Fossil Corals Occurring in the Colony, Trans. Proc. Rep. phil. Sue. Adel. for 1877-1878. pp, 120-133. Tare, R., 1879. The Anniversury Address of the President. Trans, Prov. Kept. phil, Soc. Adel, for 1878-1879, xxxix-lyxv. ; ‘ Tak, R., 1899. On Some Older Tertiary Kossily of Uncertain Age front the Murrav Desert, Trans. R, Soc. S. Aust, 23 (1), pp, 102-111. pl. 1. Tate, R. and Dennant, J.. 1896. Correlation of the Marine 'Cortiariog of Australia. Prt fil, South Australia and Tasmania. Trans. R. Sov. S. Aust. 20, pp, 115-148. Taston, D. J, 1966. Essa Gippsland Shelf No, 1, the Mid-Tertiary Foraminiferal Sequence, Appendix 3, 31-46. In Publs. Petrol. Search Subsidy Acts Aust, No, 76, Town, Rute, 1866, Smaller Foraminifera from Cuam. Prot. Pap, U.S. geol, Surv. 403-1, 41 pp., . L19, Wanr be Peete A New Genus of the Chapmanininue from Southern Anstralia, Contr, Cushinan Fdn. foramin, Res., VI_(1), pp. 45-49, pl. 8. Wap, Many, 1964, Applicstion of the lincuge concept to biostralivruphic zoning based ou planktonic foraminifera. Micropaleontolagy, 10 (3), pp. 273-290, Mis. Vig. Fig. FORAMINIFERA AND STRATIGRAPHY OF PT, WILLUNGA BEDS 109 . 17, 20-22. 23-24, 25. 26, EXPLANATION OF PLATE Puate 1 (all figures X 110) All from type section of Port Willunga Beds except Pig, 22. Guembelitria stavensis Bandy. Wypotype Ff419, sample 145-66, top of eiibensis zone, Oligocene. Chiloguembelina cubensis (Palmer). Hypotype I'f420, sample 145-66, as above, 2, Side view. 3. Oblique view showing aperture. Globanomalina sp. cf. G. naguewichiensis (Myatlink). Hypotype Ff421. sample 145-66. as above. 4. Side view. 5. Apertural view. Cassigerinella chipolensis (Cushman and Ponton). Hypotype Ff422, sample 144-66, high in cubensis zone, Oligocene. Side view, showing aperture. Cassigerinella sp. of. C. chipolensiy (Cushman and Ponton), 7, $. Hypotype Ff423. sample 123-66. genleata zone, Upper Eocene, 7, Apertural side view. 8, Oblique apertural view. 9. Hypotype F424. sample 126-66, top of aculeata vone, Upper Eocene. Side view. Turborotalia aculeata (Jenkins), Hypotype F425, sample 163-66, aculeata zone, Upper Eocene, 10. Umbilical view. 11, Spiral view. Turborotalia inerebescens (Bandy). Hypotype Ff426, sample 127-66. basal linaperta zone, Upper Eocene. Umbilical view, Turborotalia opima continuosa (Blow). WHypotype Ff427, sample 145-66, top of cubensis zone, Oligocene. 13. Side view showing the “comma-shaped aperture” noted by Blow. 14, Umbilical view. Globigerina euapertura Jenkins. Uypotype F428, sample 147-66, near top of linaperta zone, uppermost Eocene. Umbilical view. Globigerina bulloides C@Orbigny. Wypotype PF429, sample 145-66, top of cubensis zone, Oligocene. Umbilical view, Globigerina ampliapertura Bolli, Hypotype Ff430. sample 147-66, near top of linaperta zone, uppermost Eocene, Umbilical view, Globigerina sp. cl. G. ciperaensis cineroensis Bolli. Hypotype Ff431, sample 145-66, top of cubensis zone. Oligocene. Umbilical view. Globigerapsis index index (Finlay). Hypotype Ff432, sample 161-66, aeuleata zone, Upper Eocene, Umbilical view showing single aperture of immature specimen, Globigerina labiaerassata Jenkins, 20. Hypotype F433, sample 9-67, high in cubensis zone. Oligocene. Umbilical view showing relatively small aperture with thick rim. 21. Hypotype F434, sample 13-67, cubensis zone, Oligocene. Umbilical view showing larger aperture and thick rim, 22, Hypotype F435, Oil Development N.L. Mount Salt Structure Hole No, 3, 560-570 feet, Gambier Limestone, Oligocene, Otway Basin, 10 miles (16 km.) south-west of Mount Gambier. Umbilical view showing typical apertural features, Globigerina linaperta Finlay. 23. Hypotype F436, sample 120-66, aculeata zone, Upper Eocene. Umbilical view, 24, Hypotvype Ff437_ sample 145-66, top of linaperta zone, uppermost Eocene. Umbilical view. Globigerina angiporvides angiporoides Hornibrook. Uypotype Ff438, sample 21E-64, high in aculeata zone, Upper Eocene, Umbilical view. Globorotaloides testarugosa (Jenkins). Hypotype F439, sumple 145-66, top of cubensis zone, Oligocene. Spiral view. J], M. Linpsay Phare 1 A NEW SPECIES OF EREMOPHILA FROM SOUTH AUSTRALIA BY ELIZABETH A. SHAW* Summary In this paper is described Eremophila hillii, a species so far known only from two collections made near Ooldea, South Australia. It belongs to the section Stenochilus (R.Br.) Benth., but differs markedly from the other species placed herein by its small obovate to suborbicular crenate leaves. A NEW SPECIES OF EREMOPHILA FROM SOUTH AUSTRALIA by ExizAserH A, SHAwW* [Read 13 July 1967] SUMMARY In this paper is described Eremophile hillii, a species so far known only from two collections made near Onldea. South Australia. Tt belongs to the section Stenochilus (R.Br.)Benth., but differs markedly fom the other specics placed herein by its small obovate to suborbicular crenate leaves. Eremophila hillii E. A, Shaw, sp, noy. Section Stenochilus { R.Br.) Benth., Fl.Austr.d(1870)27. Frutex usque ad 70 em altus, intricate ramosus, habitu cumulatus, pilis erectis, albidis vel griseis, mnltiramosis vel ste)latis vel + plumosis vestitus. Rami plerumque cicatricibus foliorum delapsorum praediti; iuniores dense tomentosi sed vestutiores glabrescentes. Folia matura ca. 7-10 mm x 4-8 mm, dense tomentosa, spiraliter disposita ut videlur, versus apices ramorum dense conterta, obovata ad obirullata vel suborbicularis, sessilia yel in petiolum brevissimum anguslata, pleruamque ad apicem rotundata vel subacutata raro acutala, raro integra plerumque crenata. Flores singulalim in axillis foliorum summorum disposili; pedicelli ca. 2-3 mm longi dense tomentosi. Sepala 5 imbricata, post anthesin probabiliter non aucta; exteriora 3 ca, 5-2-7-0 mm longa x 2-5-3:4 mm lata, ovata ad deltata ad apicem rotundata vel subacutata, in superficiebus ambabus dense tomentosa; interiors 2, ca. 4:5-5-5 mm longa x 1-8-2-6 mm lata, plerumque ovato-lanceolata subacutata, extus dense tomentosa sed intus basem versus glabrescentia, Corolla ca. 3-0-3-5 cm longa in vivo rubra, in sicco cinnabarina ad lateritia vel interdum flavida, in statibus ambabus intus pallidior; extus pilis brevibus glandulosis sparsis aequaliter praedita, intrinsecus pilis alandulosis irregulatim sparsis. utringue base versus (sub seamentis calycis) glabrescens; supra ovarium corolla subglobosa cxpansa deinde parum. contracta el supra denno ampliata ct leviter arcuata; {obi labii superioris 4 vonsimiles longitudine, ca. 2-0-3-5 mm longi, triangulares (ci laterales interdum oblique triangulares) acuti; labium inferius ca. 12-16 mm longum, loratum reflexumque, plerumque subacutum, Stamina 4, inter se non valde diversa, ca. 3:0-4-5 mm, longe exserta, filamentis complanatis et pilis glandulosis aspersis praecipue ad basem vestitis, Otarium ca. 2°3-3:5 mm longum, glabrum oyoideum, biloculare ovulis 2 in quoque loculo; stylus ca, 3:8-4-5 cm longus gracilis ct glaber, I'ruectus ignotus. Semina ignota, Tlonoryeus: ca. 3 km north of Ooldea; 209.1960; F.Wilson 1795. Sandhills. Rounded bush 2 lt, high, Hower rec. Leaves whitish. — AD 96131130! Shrub to ca, TO em in height, intricately branched and mounded in habit, covered with white or pale grey many-branched, stellate or = plnmose erect * State Herbarium of South Australia, Adelaide. Traus. Roy, Soc, §. Aust. (1967), Voi. 91. 112 ELIZABETH A, SHAW Fig. 1. Eremophila hillii W. A. Shaw. A, dowering branch; B, flower with corolla remoyed; C, entire flower—(ill ex Wilson 1793). NEW SPRCIES OF EREMOPHILA 113 hairs. Branches usvally with rather conspicious leaf scars, the younger ones densely tomentose, but the older ones glabrescent. Leaves ca. 7-10 mm x 4-5 iwmm when mature, densely tomentose, apparently in a spiral sequence, very densely crowded towards the branch tips, chovate to obtrullate to suborbicular, apex usually rounded to subacute, rarely acute, sessile or tapering into a very short petiole, rarely entire usually crenate; midrib not visible above, shghtly raised abaxially. Flowers single, in the axils of upper leaves; pedicels ca. 23 mm long, densely tomentose: Calyx segments 5, quineuncial, probably not en- larged atter flowering; 3 outer sepals ca, 5+2-7-0 mm Tong x 2-5-3-4 mm_ wide, ovate to deltate, at apex rounded to subacute, densely pubescent on both sur- faces: 3 inner sepals shorter and narrower, ca. 4-5-5-5 mm long x 1-8-2-6 mm wide, usually ovate-lanceolate, subacute, abaxially densely pubescent, but on the inside glabreseent towards the hase. Corolla ca, 3-0-3-5 em long, in civo bright red, in sicco deep pink to brick-red or sometimes yellowish, in both states aler on the interior; exterior with sparsely and evenly scattered short glandular aits, glabrescent towards the hase (under calyx), interior with scattered glan- dular hairs, glabrescent towards the base; corolla expanded over the ovary, then contracted and again expanded and slightly fared and curved above, 5-lobed: the 4 lobes of the upper lip all + the same length, ca, 2:0-3-5 tm long, triangular (the lateral ones sometimes obliquely so) and acute; the lower lip ca, 12-16 mm long, deeply cut into the tube, ligulate and usually subacute, reflexed. Stamens 4, all + the same length, ca. 3-(-4-5 mm, much exserted, slightly fattened and bearing scattered glandular hairs, especially near the base. Ovary ca, 2-5-3'5 min long, glabrous, + ovoid, bilacular with 2 ovules per locule; style ca. 3-8-4-5 em long, slender and glabrous, Fruit and seeds unknown. South Australia: ca. 5 km. north of Ooldea; 20,9,1980; P.Wilson 1795. Sandhills, Rounded bush 3 ft. high, ower red, Leaves whitish. — AD 96131130 (holotyjpus): 2 km south of Ooldes; 21.9.1960; D. J. E. Whibley 753 — AD 96104227. Distribution: FE. hillit is so far kuown only from these two cullections made at Ooldea on the Transcontinental rail line, When these specimens were found during a routine sorting of Myoporaceae I thought it possible that they had somehow been introduced from Western Australia, but they belong to none nf the described species from that State. It is likely that E. hillit may be found ta ae cons af the rail line; this is an area in which very few collectors have worked. Discussion: E. hillit belongs to Sect. Stenochilus of Eremophila, a rather small group; Kriinzlin {1929} who treated it as a yeuus, Slenochilus R.Br., dis- tinct from Eremophila, included only 15 species. ‘They are well distinguished from other species of Eremopliila (s.1.) by having the lowest segment of the 5-lobed perianth cut much more deeply into the tube (4-6 times) than do the uthers so as to form a narrow tongue which is usually reflexed. 2k. hillii differs most markedly from all these 15 species cited by Kriinzlin by having small obo- vate to suborbicular crenate leaves. So far as [ know, there has been described since 1929 only ane species of Eremophila belonging te this section. This species, E, vireny C. A, Gardner (1942) from Western Australia, differs from F. hillii in several respects, for example, height (3-5 m), leaf shape and size (generally laneeolate and up to 7-3 cm long), and corolla colour and size (green, 1-3 cm long), E- maggilttc- rayi J.M.Black (1926), included umler section Stenochilus in the key to Eremo- wile in Black's Flora of South Australia (1957), was not known to Kriinzlin, (at differs from E, hillii in having much longer (1-5-6 cm) narrow-lanceolate leaves, smaller foral organs, and calyx segments valvate rather than quineuncial. 114 ELIZABETH A. SHAW E, hillii seems to be most closely related to E, glabra (R.Br.) Ostenf,; in its usually accepted delimitation E. glabra includes quite a wide range of forms, many of which have a tomentum like that of E. hillii and flowers practically indistinguishable from those of the Ooldea plants. However, plants of E. glabra generally have leaves which are * lanceolate and 1-5-5 cm Jong; I have seen none which approach the crenate, obovate to suborbicular leaves of E. hillii and I believe that the recognition of the latter as a distinct species is warranted. This species is named in honour of Mr. Ronald Hill of the Botanic Garden, Adelaide, who has a wide knowledge of this genus and a keen interest in it. ACKNOWLEDGMENTS I wish to thank Mr. R. Hill for providing information about the appearance of these plants in the field and Mr. L. Dutkiewicz for preparing the illustrations. REFERENCES Bentiram, G., 1870: Flora Australiensis 5. London. Brack, J. ie 1957: Flora of South Australia, ed. 2, vol. 4 (revised by E. L. Robertson} Adelaide, KrAxzuin, F., 1929: Beitriige zur Kenntnis der Familie der Myoporinae R.Br. Feddes Reper- torium. Beihefte. Bd. 54. A CONTRIBUTION TO OUR KNOWLEDGE OF SOME MAMMALS FROM INLAND AUSTRALIA BY C. M. PHILPOTT* AND D, R. SMYTH*£ Summary We record the positive results of 24 weeks of field work on the mammals of northern South Australia and adjoining areas. Localities, notes on the natural history and some discussion of the taxonomy of the species located are given. We found the rare rabbit bandicoot, Macrotis lagotis Reid, in two areas. The marsupial mouse, Sminthopsis crassicaudata (Gould), was moderately common near Birdsville and Oodnadatta. There is evidence that it is more active in the early part of nights without moonlight. Two other rarer species of Sminthopsis and Antechinomys spenceri Thomas were taken as solitary specimens. We located the two rock wallabies Petrogale lateralis Gould and P. xanthopus Gray. Near Oodnadatta, we found a native rodent, either Pseudomys (Pseudomys) minnie Troughton or Ps. (Ps.) rawlinnae Troughton. Two groups of specimens of Pseudomys (sub-genus Leggadina) were obtained, but not specifically identified. Ps. (Leggadina) hermannsburgensis Waite and the northern hopping mouse, Notomys alexis Thomas, were moderately plentiful in and near northwestern South Australia. N. cervinus (Gould) and a solitary specimen of WN. fuscus (Wood-Jones) from in and near south-west Queensland, were recorded. A CONTRIBUTION TO OUR KNOWLEDGE OF SOME MAMMALS FROM INLAND AUSTRALIA by C. M. Prmprorr*f ann D, R. Smvru®t [Read 10 August 1967] SUMMARY We record the positive results of 24 weeks of feld work on the mammals of northern South Anstralia and adjoining areas, Localities. notes on the natural history und seme discussion of the taxonomy of the species located are given. We tonnd the rare rabbit bandicoot, Maerofis lagotis Reid, in twa areas. The marsupial mouse, Sminthopsiy crassicaudata (Gould), was moderately common near Birdsville and Oodnadatta. There is evidence that it is more autive in the carly part of nights without moonlight. Two other rarer species of Sminthopsts and Antechinumys spenceri Thomas were taken ag solitary speci- mens.. We located the two rock wallabies Petrogale lateralis Gould and P. sxanthopus. Gray, . Near Oodnadatta. we found a native rodent, either Pseudamys ( Pseudomys) minwe Troughton or Py. ( Ps.) rawlinnae Troughton. Two groups of specimens of Psendomys (sub-genus Leyeadina) were obtained, but not specifically identi- fied. Ps. (Legzadina) hermannsburgensis Waite and the northern hopping mouse, Notomys alexis Thomas, were moderately plentiful in anc near north- western South Australia. N. cervinus (Gould) and a solitary specimen of N. fuscus (Wood-Jones ) from in and near south-west Queensland, were recorded, INTRODUCTION As early as 1925, Wood-Jones documented the alarming decline of popula- tions of many South Australian mammals (Wood-Jones, 1923-5). Finlayson (1961) surveyed the Central Australian mammal fauna, and his conclusions showed that this decline was continuing. Calaby (1963) suggested that six marsupials and several native rodents from Australia were already extinct, and that as many as fifty species were either close to extinction, or so little known that no definite statement could be made, It is generally agreed that if many of Australia’s mammals are to be kept from extinction, urgent measures to conserve them must be undertaken. It is also agreed that knowledge of their distribution and natural history is a pre- requisite to their conservation. With the aim of contributing to this knowledge, we surveyed the mammal fauna of parts of arid Australia. In particular, we sought information on the following seven rare marsupials: Myrmecohius fasciatus rufus Wood-Jones,. Macrotis lagatis Reid, M, leucura (Thomas). Chaeropus ecaudatus (Ogilby), Bettongia pencillata Cray, Caloprynmus campestris (Gould), and Onychogalea lunata (Gould), We also obtained information about other mammals during the search for these seven rare species. * Royal Zoolovical Society of South Australia, Frome Road, Adelaide, South Atistralia, + Present address: P.O, Box 1, Aldgate, South Australia, t Present address: Department of Genetics, Australian National University, Canberra, Australian Capital Territory. Trans. Roy. Soc. §, Aust. (1967), Vol. 91. tle CM. PMMNPOTY axp D, BR. SMYTIE METHODS The areas involyed in this survey are vast, and our time was limited, This forced up to rely particularly on information from local people. The aboriginal people, and in particular the older men from the large reserves in the north-west of South Australia and Central Western Australia, gave us valuable informution, Several rodents, and the fattailed marsupial mouse, Sminthopsis crassi- cantluta (Gould), were readily found and caught at night. The procedure we adopted was to drive along roads or over other accessile country at abont 25 miles per hour, with the lights of our truck on low beam. Small mammals within about 30 yards (27 metres) of the truck could often be seen. They usually ran away from the lights, but when followed with a spotlight trained on them, they often sought refuge in depressions or hehind grass tussocks. Here they would crouch motionless, and could readily be approached on foot. They were then caught with a net or by hand. This method was particularly effective for S. crassicaudata, which cannot run quickly. Observation of sign in the form of burraws, tracks and droppings is an obvious method of locating some species, The form of a burrow is often a clue to the species: which made it. "Kuns” of the northern hopping mouse, Notomys alexis Thomas, and the sandy inland mouse, Pseudomys (Leggadina ) hermannshurgensis Waite, atten led to burrows. Aboriginals are particularly adcpt at digging these out, This is a practical method for obtaining alive such burrowing species as the rabbit handicoot, Macrotis lagotis Reid. Several species of Notomys, Pseudomys, and Sminthapsis were taken in this way, We did not use traps. However, it is possible that the setting of a large number of traps on many nights may have revealed the presence of rarer, more cryptic species. The older aboriginals of the Pitjantjatjara tribe (from the Blackstone Range to the Musgrave and Everard Ranges), and the Nga:nyatjara tribe (from the vicinity of the Wacburton Ranges}, have specific names for the mammals of their country. We found that the use of these names amongst thern created immediate interest, and helped us obtain information on localities and the natural history of many species. The aboriginal riames of some of the species considered below are recorded. They are spelt phonetically following Douglas (1964). Measurements of lengths of head and body, tail, hind-foot (pes), and ear vf all specimens were taken using the methods described in Wood-Jones (1923). All ead specimens secured on this survey are preserved in the South Australian Museurn under the registered numbers used below. RESULTS Between April and December, 1966, we spent 24 weeks in the field. F ig. 1 shows the route we followed. Special attention was given to the following uteas: near Birdsyille (3 weeks), south of the Blackstone, Manu and Musgrave Ranges (3 weeks), and near the Warburton Range (3 weeks). We obtained intormation on the distribution and natural histury of the following species of mammals. Where no mention is made of a Species pre- viousty recorded from inland Australia, it may be assumed that no positive records of it were abtained, Order Onnrmopecenta Tachyglossus acileatus (Shaw and Nodder), Echidna Aboriginal Names; t{irilya—Pitjantjatjara; tjilkamata—~Nyga:nyatjara, SOME MAMMALS FROM INLAND AUSTRALIA 117 Field Notes. We made no serious effort to locate the moderately plentiful echidna. Aboriginals at four settlements (Fregon, Ermabella, Musgrave Park and Warburton), reported that it was common in nearby pOey hills. Residents of two stations cast and west of Lake Eyre reported seeing indisputable signs of Tachyglossus recently. We obscryed tracks and faeces at Mt. Lindsay, 129°54’E, 27°02/S. GLENGYLE Hs. N.T. . BIRDSVICL MISSION M ys evEzaRo fa, s - Pig Ne ee cREER BORE ODNADATTA KILOMETRES oo AUTHOURS’ ROUTE SOUTH AUSTRALIA Fig. 1. Map of South Australia and adjoining areas, showing the survey route followed by the authors, and the principal place names mentioned in the text. Order DIDELPHIA Family DASYURIDAE Sminthopsis crassicaudata (Gould), fat-tailed marsupial mouse (Plate 1, Figure 1) Aboriginal Name: Papalakuntalkuntalpa—Pitjantjatjara (probably a com- pound name). Localities. (i) Within 15 miles (24 km) of Birdsville, SW. Queensland; 139°21’E, 25°54’S: June 6-23, 1966; 7 females ( 9 ) and 12 males (¢);1¢ now South Australian Museum registered number M 6398. (ii) Within 5 miles (8 118 C. M. PHILPOTT anv D. R. SMYTH km) of Glengyle Homestead, SW. Queensland: 189°36’E, 24°47'S; June 11-19, 1966; 3¢ and 48. (iii) Wire Creck Bore, 20 miles (32 km) N. of Oodnadatta, N. South Australia; 135°28’E, 27°16’S, July 9-August 16; 8 9 and 4 4:1 ¢ now M 6399, Field Notes. Most of the marsupial mice were caught at night using a spot- light. At Birdsville and Glengyle, they were found on open gibber plains and on sand-ridges and the flats between these ridges. A sand ridge where many individuals were caught is shown in Plate 1, Fig. 2. At Oodnadatta, we caught them on a gibber-strewn and deeply cracked flat shown in Plate 3, Fig. 2. Occa- sionally a marsupial mouse, secn in the spotlight, would seek refuge in a well- used burrow Jess than 12 inches (30 cm) long and 4 inches (10 em) deep. We obtained some indication of the most active period of S. crassicaudata. Fig. 2 is a histogram showing the number of marsupial mice caught per hour of searching, against the number of hours elapsed since sunset. To draw this graph, we considered only those nights when one or more mice were caught. The accurate time of sunset in astronomical tables for all these dates was deter. mined, and all times were converted to local time. We did not search earlier than one hour after sunsct or later than nine hours. END OF ASTRONOMICAL TWILIGHT — Ls) MICE CAUGHT PER HOUR OF SEARCHING o t 2 3 4 5 6 7 8 9 10 HOURS AFTER SUNSET Fig. 2. Histogram of the number of fat-tailed marsupial mice (Sminthopsis crassicaudata) caught per hour of searching with a spotlight, against the number of hours elapsed since sunset. No searches were made earlier than one hour after sunset and Jater than nine hours. Each interval includes at least three and a half hours of searching. SOME MAMMALS FROM [NLAND AUSTRALIA 119 Fig, 2 shows a trend of decreasing numbers of mice being caught per hour of searching with increasing hos after sunset, This suggests that marsupial mice were most active in tho early evening. The end of astronomical twilight, and the beginning of complete darkness was about 1 hour 25 minutes after sunset. Although we searched on several agcasions hefore this time, we saw tio active marsupial mice, We caught S. erassicaudata on calm and on windy nights, and on clear and on cloudy nights, Some association of success in locating mice and the presence of moonlight is suggested by the following figures. Only twenty-two per cent of our successful searching time was carried out under moonlight (7 ont of 32% hours), Also, during these moonlight searches, we only caught eleven per eent of the total mice secured by this method (4 out of 37). All mice were weighed alive within 24 hours of capture. There were no significant weight differences between localities, The mean weight of 19 males was 13-2 + 0-18 em, with a range of 10% to 17 gm, For 18 females, the mean was 13-9 + 0-34 gm, and the range from 10K to 19 gm. There was no signifi- cant difference between the weights of the sexes (tj; ~ 1:02, 308 << P< 40%). None of the females had pouch-young, A photograph of the tracks left by a live captive is given on Plate 1, Fig. 8. The mouse was running quickly for cover. ‘he prints left by the two front foet aro the last two in cach series in the photograph. ‘The left and right front fect do not regularly alternate as the leading foot, ulthough this does occur in Fig, 3. Sminthopsis larapinta Spencer, stripe head marsupial mouse Locality, Wire Creck Bore, 20 miles (32 km) N. of Oodnadatta, N, South Australia; 185°28’R, 27°16'S; July 7, 1966, 1¢. Field Notes. ‘This marsupial mouse was caught under similar circumstanecs tithe S. crassicaudata individuals from locality (iii) shown on Plate 3, Fig. 2, and considered above. It lived sympatrically with them, and a male and female S. crassicauclala were caught close by within 35 minutes of the time of capture of this specimen. Jts weight on capture was 2] gm. Sminthopsis species, unidentified Locality. Fregon, NW. South Australia; 132°02’E, 26°47'S;, July 16, 1986; 1 ¢; M6366. Field Notes. We dug this animal from a shallow burrow on a sand-ridge. The burrow resembled those of Pseudomys (Leggadina) hermannsburgensis Waite, a small rodent which was common in the area. Aboriginals from the settlement at Fregon did not haye a specific name for this species. They gave us oly Mie word “mingkiri”, a word used generally for all sinall mice. This suggested that the animal was rare or particularly cryptic. Taxonomy. This marsupial mouse was smaller and more delicately built than. §. eressicaudata and S. larapinta. The flesh dimensions (given in mm) were: Lead and body 70, tail 85, hind-foot 13, and car 16, Its weight on capture was 10% am. ‘The skull dimensions (given in mm) were: greatest length 20-9, basal length 20-4, zygomatic breadth 11-5, interorbilal 4-3, braincase breadth 8-9, palate 11-0, nasals 7-4 x 2-0, bulla 5°7, anterior palatal foramina 1-5, P4 approx, 2x P; and Ps, Carcful examination of the pes and manus showed that the sole-pads had an obvious longitudinal row of enlarged granules at their apex. Troughton (1964) claimed that this was a diagnostic feature of the murina “complex”. He 120 C. M. PINLPOTT anp D, K, SMYTH stated, however, “The fact remains that hecause of the acknowledged lack of differential cranial characters it is impossible to provide objective diagnoses of the typical and allied forms of murina al present,” Positive identification of this interesting specimen will have to await further collections and comparisons, Antechinomys spenceri Thomas, western hopping marsnpial mouse Locality, 16 miles (26 km) W. of Glengyle Homestead, SW. Queensland; 139°22’E, 24°48°S; June 14, 1966; 1 ¢; M 6387, Field Notes. We caught this marsupial mouse while spotlighting at night on an undulating gibber plain. It did not have pouch-young, and weighed 19-5 gm on capture. The native rodent Notomys cervinus (Gould) was caught eluse by using the same method and it is probable that these two superficially similar species are sympatric. Family NOTORYCTIDAE Notoryctes typhlops Stirling, marsupial mole Aborivinal. Names: itjaritjari—Pitjantjatjara; yitarutju—Nega :nyatjara. Field Notes. We were told that specimens of this distinctive species haye been obtained within the last five years in sand-ridge areas on Everard Park Station, Musgrave Park Station (No. 16 bore), Curtain Springs Station (where it may be more common) and west of Warburton Mission. Aboriginals. from Fregon and Warburton said that it is not seen in winter. However, others dis- puted this claim. Most suggested that this burrowing species comes to the surface after rain, It is not eaten by aburiginals. It seems likely that the mar- supial mole, by reason of its small size and cryptic behaviour, is more common than is indicated by a general survey. Family PERAMELIDAE Macrotis lagotis Reid, rabbit bandicaot (Plate 2, Figure 1) Aboriginal Names: talku, ninu—Pitjantjatjara; ninu, matura—N va:nyatjara. Localities. (i) 12 miles (19 km) SW, of Mt. Aloysius, Blackstone Rane, Central Western Australia; 128°27’E, 26°06’S; November 9 and 14, 1966; 2 2; 1 2 now M6340. (ii) 7 miles (11 km) N. of Warburton Mission, Warburton Range, Central Western Australia; 126°34’E, 26°02’; December 7 and 8, 1966; 2 9;1 2 now M 6397. Field. Notes. The results of a field study of a population from locality (ii) will be reported elsewhere. This species was collected near Birdsville during 1957-59 (Mack, 1961 }. We were told that it had not been seen there for about five years, Tlowever, we obtained reports of sightings of a mammal which was probably M. lagotts within the last two years from localities west of Glengyle Homestead, The only area we visited where it seemed tu be moderately plentiful was near Warburton Mission. None of the females had pouch-young. Taxonomy. ‘The measurements made on live specimens as soon as prac- ticable after capture are given in Table 1. General appearance and the above measurements suggest that these specimens belong to the type subspecies M. lagotis lagotis Reid. SOME MAMMALS FROM INLAND AUSTRALIA 121 TABLE 1 Flesh measurements of four female rabbit bandieoots, Macratiz layotia, from Central Western Australia Juvenile 9, Adult ¥*, Adult 9, Adult 2 Blackstone Ra. | Blackstone Ra. | Warburton Ra. | Warburton Ra. Date of capture 9, XT.66 14.XT.66 7.X1T.66 8.XII.66 Wt. on capture (gm) 400 700 960 8m Head and body (mm) 310 320 330 320 Tail (mm) 180 190 230 220 Ear (mm) 70 85 90 85 Hind-foot (mm) 78 88 96 92 * M 6340 + M 6397 Family MACROPODIDAE Petrogale lateralis Gould, black-lanked rock wallaby (Plate 2, Figure 2) Aboriginal Name: waru—Pitjantjatjara, Nga:nyatjara. Locality. Alkara, 90 miles (145 km) SW. of Mt. Woodroffe, Musgrave Ranges, NW. South Australia; 130°30'E, 27°04’S; July 27, 1966; 1 4 M 6388 (skull only). Field Notes. We observed two or more rock wallabies on a rocky hill of tumbled, granite-like boulders shown on Plate 2, Fig. 3. They emerged from their Jairs at about mid-afternoon. The testimony of aboriginals that this wallaby does not drink regularly was borne out by the absence of permanent water nearby. Herbs and shrubs growing on the slopes, and which were possible food species, included buckbush (Salsola kali), wild geranium (Erodium cyg- Nora: several perennial grasses, and the shrubs Sida virgata and Ptilotus obovatus. Occasionally, aboriginals living on settlements from Fregon to Warburton Mission catch and eat rock wallabies, and suggest that it is one of the more plentiful mammals of the area. The wallaby weighed 4-590 kg on capture, and its body dimensions (in mm) were: head and body 525, tail 540, hind-foot 142 and ear 66. Petrogale xanthopus Gray, yellow-footed rock wallaby Field Notes. We observed these wallabies on April 10 to 16, 1966, at three localities in the Northern Flinders Ranges between 138°40’ and 139°20’E longi- tude, and 30°20’S and 30°40’S latitude, In two areas, we watched them feeding on the green herbs in creek beds at dusk and at dawn, One female had a large jocy. Order MonopDELPHIA Family MURIDAE Psendomys (Pseudomys) species, unidentified native rodent (Plate 3, Figure 1) Locality, Wire Creek Bore, 20 miles (32 km) N. of Oodnadatta, N. South Australia; 185°28’E, 27°16’S; July 9 and August 16, 1966; 4 9,3 3 and 8 juveniles (J); 1 @ now M 6341. 123 C, M. PHILPO'T axp D, R, SMYTH Field Notes. ‘These native rodents were living on an open gibber flat, with silver saltbush (Atriplex rhagodioides) heing the only perennial plaut. The habitat is shown on Plate 3, Fig, 2, We dug eight of them from burrows built under or near the saltbushes, “These burrows were less than 2 feet (60 cm) long, and one is shown on Plate 3, Fig. 3. A male, a female and three juveniles were dug from one burrow; another contained a male and a female, whilst a third contained a solitary female. A male and a female were caught with a spotlight about 4% hours after sunset on different nights. One of the females captured on July 9 produced a litter of 3 on about July 12. She was paired on October 21, and produced another litter of 8 on December 2. On January 3, 1967, she again produced young; this time there were 4 offspring. Taxonomy. Tables 2 and 3 present flesh and skull dimensions of three of these native rodents, and of types of Ps. minnie Troughton and Ps. rawlinnae Troughton taken from Troughton (1932), TABLE 2 Flesh dimensions of three adult: females ofan unidentified Pseudomys (Pacridomys) species from. near Oodnadatta, N. South Australia, and of type specimens of Ps. minnie and Py.rawlinnae. i @, MB341 | 9, M6342* | 9, M6343") G, allotype 3, holotyps Psominnie Ps. rawlinnne Wt, ont capture (gm) 33 = ba —— = Head and body (mm) EH) 8 101 117 113 Tail (mm) fn 85 93 111 oP Ear (mm) 23 21 22 23 214 Hind-foot (mm) 26 24 27 27 2Tt * Born in captivity, and measured when 6 months old. TABLE 3 Skull dimensions (in pin x 10) of three adult fernales of an unidentified Pseadomys (Pseudamys) species from near Oodnadalta, N, South Australia, and of type specimens of Ps. minnie and Pa, rawlinnac. i | 9, M6p4L 2, M6a42* = | Oo, M6343* Pallotype | ¢ holotype Ps. minnie Pa. rawlinnae Greatest ieneth 29-5 28-4 | 20-4 307 BO-8 Baza! length 26-6 25-1 | 28-8 25-0 — Zypomatic breadth | 15-4. 14-6 11-5 16-8 1u-L Brainease bread 14-1 12-9 13-7 14°3 | 13-7 dnterorbital breadth 4-6 a-8 4s3 4-0 ars Nasals length JL-4 LO-8 11-2 1-1 12) Nasals hreadth 3-3 3-0 neh 3-0 2g Palatal lungth 16-6 TA-4 16-8 14-0 14:5 Anterior palatal foramina 6-8 +7 67 7-3 TO Bulla, a2 4-3 5-0 — Upper molar series Gt 5-3 523 a-$ 5-a Breadth M1 1-8 1-9 1-8 1:8 * Born in captivity, and measured when 6 months old. We have used Tate (1951) to assist in the following deductions, The absence of accessory cusps anterior to the first Joph of ml, and the large size of the species indicate that it may be assigned to the sub-genus Pseudomys Gray. SOME MAMMALS FROM INLAND AUSTRALIA 123 The above tables show that the dimensions of our series are similar to those given by Troughton (1932) for Ps. minnie and Ps. rawlinnae. The coat also appears to be similar. Finlayson (tonite) noted that these species were similar jn some features, but that the upper molar row, anterior palatal foramina, and palatal length were shorter in Ps. rawlinnae. These differences are slight in Table 3, but they may indicate that our series are closer to Ps. rawlinnae. Trough- ton (1932), in his description of Ps. rawlinnae, made no mention of a black caleaneal patch which is well marked on our specimens, However, Finlayson (1939b) reported that the patch was present in the specimens he called Ps. vawlinnae. J. A. Mahoney of Sydney has examined these specimens. and_has indicated that he considered the sum total of characters to be closer to descriptiuns of Ps. minnie. However. he also noted that recorded descriptions of the two species are little different, and that they may perhaps be shown after further study to be canspecitic. Pseudomys (Leggadina) species (group 1), unidentified native rodent (Plate 4, Figure 1) Localities, (i) 30 miles (48 km) N. of Birdsville, SW, Queensland; 139°28’E, 25°23'S; June 12, 1966, 22: M 6350-1, (ii) Wire Creek Bore, 20 miles (32 km) N, of Oodnadatta, N, South Australia; 135°28’E, 27°16’S; July 9, 1966, 1 3; M 6352. Field Notes. The two specimens from locality (i) were obtained by spot- lightiny: about 7 hours after sunset. The single specimen from locality (ii) was taken rom a shallow burrow, which was without a nest, and which had been freshly dug after rain. The habitat at locality (fi) is shown on Plate 3, Fig. 2. A series of tracks of a live captive from Birdsville are shown on Plate 4, Fig. 2. The mouse was running for cover, Taxonomy. Tables 4 and 5 contain skull dimensions of these three native rodents, and also of types of Ps. forresti Thomas, Ps, waitei Troughton and Ps. messoria Thomas taken from the original descriptions, TABLE 4 Flesh dimensions of three adult males of on unidentified Pendomiys (Leggadina) species (group 1), and of type specimens of Ps. forresti, Ps. watei and Py, measoria. ls. M6350" | J, MG341* | 3, Ma3s2t| Gtype holotype, 3, type \ Ps. forresti| Ps. welted [Pavnessoria Head and body (mm) 80 76 83 a7 Tail (mim) 56 60 oo ad Kar (mm) 12 Il 12 13 Hind-fuot (mom) 18 17 164 7 * From 30 miles N. Birdsville, 5.W. Queensland. ft From near Godnadatia, N. South Australia; weighed 18-5 gm on capture. The general characteristics of the skulls of these specimens fit Thomas (1910) definition of the subgenus Leggadina. The specimens readily fit into the forresti-vaitei-messoria group of Tate (1951), who provides a key to the groups of Leggadina, The measurements above indicate that our specimens are considerably smaller than Ps, forresti, but larger than Ps, messoria. Flesh dimen- sions match those of Ps. waitei, but there are a number of differences, notably the palatal and nasal lengths of the skulls. Coat colours of all three species in 124 C. M. PIILPOTT anv D, RK. SMYTH TABLE 6 Slrull dimensions of three adult mules of an unidentified Psewdomya (Leggadina) species (group 1), ant of typy specimens of Ps. forresti, Ps.waitel, and Fe. messoria. Ss SSS 3, M6350" | 3, MG351*|3, M6352t] }, type |g. holotypel gy type Ps. forresti| Ps. uxeites |Ps.mesaaria Geeatest length 23-1 24-4 25:1 25-0 22-6 2a+5 Basal length 20-6. 217 21-5 — — — Zygomatic breadth — 13-3 Lb. 1 13:5. 12-0 13-6 Rraincase breadth 11-3 11-7 11:6 — Shel 11-2 {nterorbital broadth Hei de6 ob: 7 oeé 3-5 3-6 Nasals Jength RS 8-3 h-4 8:5 7-6 8-0 Nuasals breadth 2+ 2-8 Beh = 2-3 — Pulatal length 12°9 134 13-1 13-00 6] Jing 11-5 Anterior palatal foramina 5rl aes 4-9 5-5 3-3 4-0 Bulla — 4-2 42 — — 4-1 Upper molar series 405 405 4-4 4-4 4-1 4-5 Breadth M1 1:5 15 1l+4 — =~ — | ee ee es See * From 31) miles N. Birdsyilly, &.W. Queensland. $+ Frorn near Oodnadatta, N. South Anatralia. the group, and of our specimens are apparently similar, However, Thomas (1910) said that the hairs on the belly of Ps, messoria were slaty grey, whereas tliose of the above series are pure white to the base, Pseudomys (Leggadina) species (group 2), unidentified native rodent. Aboriginal Name: ilpalya—Pitpantjatjara. Localities—(i) 20 miles (82 km) SSE, of Mt. Aloysius, Blackstone Range, Central Western Australia: 128°44°E, 26°17’S; November 15, 1966; 1 9 and 3]: M 6344. (ii) 26 miles (42 km) SE. of Mt, Aloysius, Blackstone Range, Central Western Anstralia; 128°52’E, 26°18’S; November 15, 1966; 4 2.14 and 11]; M 6345-9, Field Notes, We dug these mice from burrows in two loamy plains near rocky hills shown on Plate 5, Figs, 3. and 4, There were many small saltbushes, grasses and ephemeral herbs on both plains, hut no other perennial plants. The burrows were about 6 inches (15 em) deep and 15 inches (87 cm) long. A nest of grass and fowers of an everlasting daisy was found in the deepest part, One or more blind tunnels led from the nest. Unlike the unidentified Pseundomys (Pseudomys) species, adult males and females were not found in the same burrow, Two burrows contained solitary mice; one a malc, the other a pregnant female. Four burrows each contained a lactating female and her young. ‘There were two litters of three and two of four mice. Living sympatrically with these rodents was the house mouse, Mus musculus Linne. We dug out two pregnant house mice from burrows near to and closely resembling those of the native mouse, Pitjantjatjara_ elders from Musgrave Park unhesitatingly called the native mice “ilpalya’. Finlayson (1961) reported that this was u Pitjantjatjara name for a Notomys species “close to alexis but not specifically identified”. Taxonomy, Tables 6 and 7 give measurements of individuals in the flesh and of skulls of this series. We have considered this group of specimens separately from the preceding group of Pseudomys. Closer study may show them to be conspecific. As with Pseudomys (Leggadina) species ( group 1), they fall readily into Thomas’ (1910) subgenus Leggadina, and Tate’s (1951) forresti-waitei-messoria group. SOME MAMMALS FROM INLAND AUSTRALIA 125 TABLE 6 Flesh dimensions of six adult specimens of an unidentified Pseudomys (Leggadina) species (group 2), from SE. of the Blackstone Range, Central Western Australia, All females were lactating. i Y 9, M6344 | 9, M6345 | J, M6346 | 9, M6347 | 9, M6348 | 9, M6349 Wt. on capture (zm) 25 29 18 wu 29 214 Head and body (mm) &8 101) 380 83 \ 90 $1 Tail (mm) 59 62 58 59 69 56 Er (mm) 45 14 44 14 14 12 Hind-foot (mm) 17 17 16 16 18 16 ee a ood * Pregnant. TABLE 7 Skull dimensions of five adult female specimens of an unidentified Pseudamys ( Legquelina) species (group 2) from SE. of the Blackstone Runges, Central Western Australia, i @, M6544 O, M6345 &, M6347 | 2, M6348 2, M6349 Greatest length 23-3 24-3 | 88-4 23:8 224 Basal length 10-9 212 21-1 20+3 19+2 Zygomatie hreadth 12-0 1304 12-2 12-5 119 Bruincase breadth Tet 1-7 11-0 11-0 10-8 Interorhitel Breadth 3-6 3:3 3+9 3-5 3:3 Nasals length ve) $3 7-9 8-0 7-9 Nasals breackth 2-2 2-4 22 2+4 2-3 Palatal length 12/0 12-8 12-0 12°35 11-5 Anterior palatal foramina Bel 5-4 5-1 H-1 4-7 Bulla 4:2 4-1 4-1 3-9 4:0 Upper molar series 4-7 4-8 465 4-4 4-4 Breadth of M1 1+6 1-6 16 1:5 1+5 J. A. Mahoney has examined a sample of the two preceding unidentified groups, He placed both of them in the forresti-waitei-messuria group, and, believing that its members are conspecific, tentatively called all our specimens Pseudomys (Leggadina) forresti Thomas. The relationships of what we have called Pseudomys {Leggadina) species (group 1) and (group 2), and other native rodents will have to await further collections, and study of new specimens and the types before any definile state- ment can be made. Pseudomys (Leggadina) hermannsburgensis Waite, sandy inland mouse. (Plate 4, Figure 4) Aboriginal Name: mingkiri — Pitjantjatjara, Nga:nyatjara (now used for all small mice). Localities. (i) 15 miles (24 km) E. of Mt. Ilbillee, Everard Ranges, NW. South Australia; 132°43’E, 27°01’S; July 18, 1966; 3 ¢ and 3 ¢. (ii) Fregon, NW. South Australia; 132°02’E, 26°47'S; July 16, 1966; 7 9 and 7a. (iti) Bowden Hill, 80 miles (129 km) SW. of Mt. Woodroffe, Masgtave Ranges, NW. South Australia; 130°41'E, 26°58’S; July 23, 1966; 1 29 and 1 @; 1 9 now M 6354. (iv) Alkara, 90 miles (145 km) SW. of Mt, Woodroffe, Musgrave Ranges, NW. South Australia; 190°30’E, 27°04’S; July 27, 1966; 3 ¢ and 3 3; 1 © now M 6353. (v) 42 miles (68 km) SW. of Mf. Aloysius, Blackstone Ranges, Central Western Australia; 128°06’E, 26°25’S: November 12, 1966; 2 9; 1 3 now M 6355, 126 C. M. PHILPOTT ann D. R, SMYTH Field Notes, This species was one of the commonest mammals between 128° and 132°F longitude and 26° to 27°S latitude. All specimens were taken from burrows in sandy-loam soil, often in mulga (Acacia aneura) scrub, The burrows were characterised by the absence of a large mound of soil neat the mouth, as illustrated by Plate 4, Fig. 3. From one to four mice, with sex ratios amongst them variable, were taken from single burrows. One female from Fregon and two from locality (v) were pregnant, and a litter of four juveniles was secn at Fregon, The mean weight of 17 adult males which were weighed less than 24 hours after capture was 11-1 + 0-4 gm with a range of 8 to 13 gm. For 9 adult, non- pregnant females, the mean weight was 10-3 + I-4 gm, with a range of 9 to 13! em, Taxonomy. Flesh and skull dimensions of the three specimens in the South Australian Museum (M 6353-5) all fall within the limits given by Finlayson (1841) for a large series of Ps, hermannsburgensis. Habits and coat wlour which he described for this species are closely similar to those of our series, Notomys cereinus (Gould), fawn hopping mouse, (Plate 5, Figure 1) Localities. (i) From 18 miles (29 km) S. to 70 miles (113 km) N, of Birds- ville, SW. Queensland; 139°21'E, 25°54’S; fume 11-23, 1966; 3 ° and 7 @. (ii) 5 miles (8 km) S. of Glengyle Homestead, SW, Queensland; 139°36’E, Seine 9 8 and 9 &; 10 specimens from both localities now M 6357-65 and M 6400, Field Notes. This species was moderately plentiful, and all specimens were taken while spotlighting at night on bare claypans and flats illustrated on Plate 1, Fig. 2. It was seen on the same nights and in the same areas as Sminthopsis crassicaudala and Antechinomys spenceri. All specimens were taken between 3 and 8 hours after sunset, Taxonomy. ‘There was no gular pouch in any of the specimens but a marked groove was present on the anterior surface of the upper incisors of 9 skulls examined from individuals from both localities (M 6357-65). Finlayson (1959) stated that these are characteristics of N. cervinus, and that they distinguish i from N. fusciis (Wood-Jones ). Notomys fuscus (Wood-Jones), Wood-Jones’ hopping mouse. Locality. Pandie Pandie Homestead, Diamantina R,, NE. South Australia: 139°23'E, 26°07'S; summer of 1964-65; 1 ¢: M 6356. — Field Notes, This specimen was obtained by Mrs, L. Murton, who reported thal similar hopping mice were often seen during the summer of 1964-65; we found no live specimens while in the area. Netomys alexis Thomas, northern hopping mouse, (Plato 5, Figure 2) Aboriginal Name: tarkawara—Pitjantjatjara, Nga:nyatjara. Yocalities, (i) Turner’s Well, 22 miles (35 km) NE, of Mt. Woodroffe, Musgrave Range. NW. South Australia; 132°00’F, 26°07'S: July 19, 29, 1966: 3 9.4 ¢ and 6J; 1 ¢ and 4] now M 6367, (ii) 14 miles (23 km) WSW. of Mt. Aloysius, Blackstone Range, Central Western Australia; 128°21’E, 26°10°S November 1, 1966; 2 9 and 15] (including 8 2 and 7 3 ); M 6369-85, * ‘ SOME MAMMALS PROM INLAND AUSTRALIS iat Field Notes, This species was common ever much of the area where we found that Pseucdomys (Legeadina ) hermannsburgensis was plentiful. However, we obtained roost N. alexis from more sandy areas, with less mulga serub. All specimens were dug from burrows, From two to five juveniles were taken from six single burrows; in four of these burrows lactating females were also taken. The burrows had a vertical entrance shalt, about 6 to 18 inches (15 to 43 cm) deep, and with na soil at the mouth. Horizontal “drives”, with occasional branches extended from the base of the shaft to a maximum of about 6 feet (1-8 m). At the end of the “drive”, a vertical escape shaft was often dug stopping about an inch (2:5 em) below the surtace. At levality (ii), empty shells of the quandong, Santelum acuminatum, were found near the etitrance and along the “drive” vt oue N. alexis burrow, The shells had bees opencd by removal of a small circular segment of the hard outer coat. DISCUSSION Of the seven rare marsupials searched [or specifically, we located only Macrotis lagotis, Because of the large fluctuations in numbers which o¢cur in many desert-living species, and because we searched only a small proportion of suitable habitats, it is by no means certain that any of the others are extinct. However, only elderly aboriginal men had any knowledge of some of the species, and this knowledge was obtained when they were young men, For example, Myrmecobius fasciatus is spoken of as having gone for many years from the Everard and Bedford Ranges. However, we should note that aboriginal people everywhere are much less mobile aud dependent on vative mammals for food now than they were: in Lhe past. None of the 18 female Sminthopsis crassicaudata caught during June, July and August had pouch-young though some of tle were toa young for breeding. Conditions during this period were excellent, with recent rains having increased plant and insect life. Martin (1965) recorded breeding of this species from july to April under optimal conditions in the laboratory, It is therefore possible that breeding in the wild is to some extent regular and seasonal, unlike the fairly regular breeding noted under laboratory conditions, Nearly all of the females of the unidentified Pseudomys (Psendomys) species, Pseidomys (Leazgadine) species (group 1), and of Notomys alexis were lactating, or were pregnant. Recent rains had resulted in good growth of herbs, swhich had qwaltured and shed their seed, This large-scale, synchronous breeding of native rodents and consequent population increase. can perhaps explain to some extent the taxanomic difficiilty ot this group, ‘That there are large fluctuations in numbers of many native rodents has often been noted, especially m arid regions such as the Lake Eyre Basin (e.g, Finlayson, 1939a). A large increase in variability has been observed wher a large increase in the numbers of several animals has occurred. Ford and ford (1930), working with the butterfly Melitaca aurinia, were one uf the first to draw attention to this lurge increase in variability. The presence of much variation in rapidly increasing populations of rodents would tend to decrease the number of differential characters which could be used to separate closely related species. Also, variability of the one species from different areas would probably be large. This is because populations from different localities would perhaps be derived from isolated small numbers of individuals present before the large increase. 128 C. M, PHILPOTT axn TD. KH. SMYTIT Finlayson (1961) considered that pastoral exploitation, and the introduced rabbit, fox and domestic cat were important factors influencing the abundance and range of Central Australian species. In order to conserve at least some of these species, they must first be located. Then, the influence of the above factors should be analysed. Finally, suitable areas should be proclaimed as reserves, and some contrel of the environment attempted in order to minimise these deleterious factors. We hope our study has contributed ta the first step in such a programme ol urgent conseryation, ACKNOWLEDGMENTS This study was carried out as part of the scientific work of the Royal Zoo- logical Society of South Australia Inc. and was made possible by a generous grant from the Nuffield Foundation te Dr. P. S. Watts, its president. We are extremely grateful to Professor H. G. Andrewartha, of the Depart- ment of Zoology, University of Adclaide, for many discussions and useful guicd- auce throughout the year, We also thank Dr. S. Barker for much encouragement and interest. The Council and staff of the Royal Zoological Society of South Australia, and especially Dr. P. S, Watts and Mr. W. E, Lancestcr, made our task lighter with unlimited help. We received much help and cooperation from the various State Fauna Authorities, and especially the South Australian Department of Fisheries and Fauna Conservation. We are appreciative of help and under- standing fram the South Australian Department of Ahoriginal Affairs and the West Australian Department of Native Welfare. The South Australian Museum provided us with unlimited facilities for which we are most thankful. In the field, we received nothing but kind hospitality and cooperation from many station residents. Dr. and Mrs, G, C. Gregory of Leigh Creek township were especially helpful, The staff at Fregon, Ernabella, Musgrave Park an Warburton Ranges made our stay in their area both fruitful and enjoyable, Finally, we have spectal thanks for Mr. T. Dodd, and those people from the Pitjantjatjara and Nya:nyatjara tribes whose help was entirely insttumental in dur success while working in their lands. REFERENCES Cimon: H., 1963. Australia’s Threatened Mammals. Wildlife, Brisbane, 1; No, 1, pp, 5-18, Deveias, W. H., 1984, An Introduction to the Western Desert Lapguage, Oceania Lin guistic Monogr, No, 4 (revised). (University of Sydney: Sydney, } Wriayson, H, H., 1939a. On Mammals from the Lake Eyre Basin: Part V. General Re- marks on the Increase of Murids and their Population Movements in the Lake Eyre Basin during the Years 1930-36. Trans. R. Soc. 8, Aust., 63, pp. 348-3473_ Myiarson, H. H.. 1938h. Records and Descriptions of Muridae from Onldea, South Ans tralia. Trans. H. Soc, 5. Aust. 63, pp. 354-364. Frt.ayson, H. H,, 1941, On Central Australian Mamynals; Part H, The Muvidae (con- tinued). ‘Trans, R. Soc. S$. Aust., 65, pp. 215-232, Fuxtayson, HW. H., 1960. Nomenclature of Notomys (Muridae) jn the Lake Eyre Basil, Trans, R. Soe. $. Aust., 83, pp, 79-82, Foxcayson, HW. 11, 1961, On Central Australian Mamunals: Part IV, The Distribution and Status of Central Anstralion Species. Rec, S. Aust. Mus., 14, pp, 141-191, Fouv, H. D., and Forn, E. B., 1930. Fluctuation in Numbers, and its Influence an Variation, in Melitaea aurinia, Rott, (Lepidoptera). Trans. R. ent. Sac, Lond., 78, pp, 345-952. Mack, G,, 1961. Mamuials from South-western Queensland. Mem, Od. Mus., 15, pp, 215-229. SOME MAMMALS FROM INLAND AUSTRALIA 129 Martin, P. G., 1965. The Potentialities of the Fat-tailed Marsupial Mouse, Sminthopsis crassicaudata (Gould), as a Laboratory Animal. Aust. J. Zool., 13, pp. 559-562. Tate, G. H. A., 1951. The Rodents of Australia and New Guinea. Bull. Amer. Mus. nat. Hist., 97, pp. 183-430. Tuomas, O., 1910. The Generic Arrangement of the Australian Murines Hitherto Referred to as Mus. Ann. Mag. nat. Hist., 8, pp. 603-607. Troucuton, E., 1932. On Five New Rats of the Genus Pseudomys. Rec. Aust. Mus., 18, pp. 287-294. Troucuton, E., 1964. A Review of the Marsupial Genus Sminthopsis (Phascogalinae), and Diagnoses of New Forms. Proc. Linn. Soc. N.S.W., 89, pp. 307-321. Woop-Jonges, F., 1923-5. “The Mammals of South Australia, Parts I, II and III.” (Govern- ment Printer: Adelaide. ) C. M. Puiteorr ann D. R. Smyrin PLATE 1 Fig. 1. Male Sminthopsis crassicaudata from near Birdsville, approx. life-size. Fig. 2. Looking along a stabilized sand-ridge 3 miles east of Birdsville where we caught 6 S. crassicaudata. Notomys cervinus was also living nearby. Fig. 3. Tracks of a captive male S. crassicaudata from Glengyle. Direction of travel is from bottom to top, and the scale is in inches and centimetres (see text). C. M. Puteeotr ANp D, R. SMyTH PLATE 23 Fig. 1. Immature female rabbit bandicoot (Macrotis lagotis) from south of the Blackstone Ranges, approx. 4 life-size, Fig. 2, A male black-flanked rock wallaby (Petrogale lateralis) taken in the wild at Alkara. south of the Musgrave Ranges, Fig. 3. Alkara, 90 miles south-west of Mt. Woodroffe. We observed P. lateralis on the darker rock on the left of the hill. We caught Pseudomys hermannsburgensis on the sand ridge in the foreground, C. M. Puitpotr ann D. R. SaiviTn PLATE 3 Fig. 1. An adult female Psendomys (Ps.) minnie or rawlinnae, obtained as a juvenile from Wire Creek Bore, approx. 0-65 x life-size. Each division of the scale is 1 centimetre. Fig, 2. Habitat at Wire Creek Bore 20 miles north of Oodnadatta. The shrub in the fore- ground is the silver saltbush, Afriplex rhagodioides. On this plain we caught Sminthopsis crassicaudata, 8. larapinta, Ps. minnie or rawlinnae, and an unidentified species of Ps, (sub-genus Leggadina). Fig. 3. A burrow of Ps. minnie or rawlinnae under a silyer saltbush at Wire Creek Bore. The mound at the entrance is approx. 9 inches wide. |. M. Parueorr Ann D. R. Smyrin PLATE 4 Fig. 1. An unidentified male Pseudomys (Leggadina) species from 30 miles north of Birds- ville, approx, 0-6 x life-size, Vig. 2. Tracks of the Ps. (Leggadina) species shown in Fig. 1. Direction of travel is from bottom to top, and the scale is in inches and centimetres (see text). fig. 3. Entrance to a Pseudomys hermannsburgensis burrow near Fregon. Fig. 4. A male Pseudomys (Leggadina) hermannsburgensis from Alkara, south of the Musgrave Ranges. C. M. Puitporr Anp D, R. SmyrH PLATE 5 Fig, 1. A male Notomys cervinus from near Birdsville, approx. 0-6 x life-size. Fig, 2. A male Notomys alexis from Turners Well, near Ernabella, approx. 0-6 x life-size. Fig. 3. Habitat of an unidentified Ps. (Leggadina) species (group 2). 20 miles south- south-east of Mt. Aloysius. Fig, 4. Entrance to a burrow of Ps. (Leggadina) species (group 2), at the locality shown in Fig. 6, The burrow contained an adult male. FOSSIL ELASMOBRANCH TEETH OF SOUTH AUSTRALIA AND THEIR STRATIGRAPHIC DISTRIBUTION BY N. S. PLEDGE* Summary Several hundred fossil elasmobranch teeth, collected from Tertiary formations in South Australia, have been identified to species level in most cases. Twenty-five species have been recognized, including the new species Carcharias maslinensis Pledge, and three forms are identifiable only to genus level. The majority of species belong in the Infra-class Osteodonta, and some absentees are noted. Diagnoses for the various genera and species are given, and a guide to their salient features. When the occurrences are arranged stratigraphically, they fall into two main time spans, in Middle to Upper Eocene and Lower to Middle Miocene, when conditions were most suitable for preservation. It is concluded that the establishment of zones in Australia, based on assemblages of elasmobranch teeth, is not yet possible. FOSSIL ELASMOBRANCH TEETH OF SOUTH AUSTRALIA AND THEIR STRATIGRAPHIC DISTRIBUTION by N. S. Prepc* [Read 14 September 1967] SUMMARY Several hundred fossil elasmobranch teeth. collected from Tertiary forma- tions in South Australia. have been identified to species Jevel in most cases. Twenty-five species have been recognized, inchiding the new species Carcharias maslinensis Pledge, and three forms are identifiable only to genus leyel, The majority of species belong in the Infra-class Osteodonta, and some absentecs are noted, Diagnoses for the various genera and species are given and a guide to their salient features. When the ocettrences are arranged stratigraphically. they fall into two main time spans, in Middle to Upper Focene and Lower to Middle Miocene, when conditions were most suitable for preservation. It is concluded that the establishment of zones in Australia, based on assemblages of elasmobranch teeth, is not yet possible. INTRODUCTION The Elasmobranchii have a history reaching back to the Devonian Period, but many major groups became extinct before or during the Mesozoic, Although marine sequences of Cretaceous age are known in South Australia, no shark remains have been recorded from them yet. All the teeth described herein are from Tertiary deposits. Apart from a brief note by McCoy (1875), Chapman (1913, 1914, 1917), Chapman and Pritchard (1904, 1907), and Chapman and Cudmore (1924) have been the only ones to write on the Tertiary fishes of Australia, Most of their material was from Victoria; only rarely were South Australian specimens mentioned. The first work of any importance concerning fossil fishes was by Louis Agassiz, whose “Recherches sur les Poissons fossiles” was published at Nen- chatel between 18384 and 1843. Other papers soon followed in Europe and America, and a few workers studied the fossil fishes of Indo-Pacific region. Very recently, L, S$. Glikman (1964), of the Academy of Science of the U.S.S.R., published a volume wherein assemblages of sharks’ teeth are used to define twelve zones in the Russian Paleogene, The present work was undettaken to determine whether similar zones could be established for the Australian Tertiaries. MATERIAL The paleontological collection of the Department of Geology, University of Adelaide, includes several hundred fossil elasmobranch teeth, mostly from local Tertiary formations. A large number of teeth in the collection of the Geological auvey, of South Australia, and several small private collections, were also studied. * University of Adelaide. Present address: Geolugy Dept., University of Wyoming, U.S.A. Trans. Roy, Soc, S. Aust. (1967), Vol. 91. 16 \. 8, PLEDGE DEPOSITORIES Teeth held in the Department of Geology, University of Adclaide, ave catalogued with the prefix “i”. Those in the collection of the Geological Survey are recorded in the Verte- brate Catalogue, and have timbers with the prefix “V", e.g, V35. NOMENCLATURE Many elasmobranch genus names were established by naturalists of the eatly 19th century, such as Rafinesque (1810), de Blainville (1816) and Cuvier (1817). Cuvier is best known, but many of the genera he established were synonyms. Agassiz propagated these synonyms in the literature, and it is only in yecent years that the original names have come into more general usage for the fossil species. Some authors, such as Glikman (1964), have split these “classical” genera, but it is difficult to see their reasons for so doing, As genera and species of fossil sharks are established only on the character of the teeth, it is obvious that both “splitting” and “lumping” may occur. Com- parison with the teeth nf living sharks can reduce this risk somewhat, according lo the particular genus of shark, but not remove it, Many species are therefore only morpho-species, and may represent several different biospecies, or con- versely one biospecies may be represented by several morpho-species. As far as possible, the teeth described herein are compared with species originally de- scribed from Europe or America, on the grounds that most species of larger sharks have very wide distributions, and the number of parameters by which sharks’ teeth are identified are not sufficient to establish geographic species for similar forms from different parts of the world. CLASSIFICATION Several classifications of sharks have been proposed and these vary widely. Most have a strictly zoological approach, using characters, rarely, if ever, pre- served in fossils. Nicholson and I,ydekker (1889) used a major division based on the structure of the vertebrae, Tectospondyli and Asterospondyli, while Bertin and Arambourg (1958) had major Aivisions, Pleurotremata and Eypo- tremata, based on gill position, i¢. more or less on body form. Smaller divisions ate based on morphological and physiological eriteria in which tooth form is rarely included. Families and orders of fossil sharks’ teeth are therefore hased on living forms. However, the major division in fossil sharks is one of the basic tooth: structure, and this has been utilised by Glikman (1964) in his classifica- tion of extinct and living clasmohranchs, which is marked by a distinction hetween orthodont and osteodont dentition. Orthodont teeth (Text-fig. la) have a pulp cavity surrounded by artho- dentine, and such teeth are found in the majority of living and fossil elasmo- branchs, including one order of “true” sharks, the Carcharhinidae (whalers, topes, tigers and Fathmerhead sharks}, also the dugfishes, rays and skates, and those primitive sharks: the hexunchids (notidanids) and heterodontids. The Osteodonta have teeth (Test-fig. Ib) whose compasition, including the root, is a vascular bone-like form of apatite called osteodentine. Only “trac” sharks are included in the group: nurse and sand sharks, bulldog sharks, por- heagles, makos and blue pointers, threshers, and white pointers. This division raises the question: which feature is of greater phylogenetic importance — a basic tooth structure, a basic vertebral structure, ar overall hedy shape and anatomical layout? FOSSIL ELASMOBRANCH TEETH OF S.A. 137 sary AAD) Fig. 1. a. Orthodont tooth (Carcharhinus), b. Osteodont tooth (Carcharias), e; Fnamel (vitrodentine), or: orthodentine, os: osteodentine, ps: pseudodentine, p: pulp cavity (diagrammatic sections, after Bertin and Arambourg, 1958). Fig. 2. a. Profile, Carcharias macrotus. b. Inner face, ditto. c. Lamna_apiculata, inner face. d: lateral denticles, e: entire margin, 1: root lobe. m: median boss, n: nutritive pit, §: striations. Fig. 3. a. Serrated margins, e.g. Galeoverdo. bb, Serrated margins, e.g. Hemipristis, c, Den- tate marftins, e.g. Carcharodon, Fig. 4. Carcharias arenarius Ogilby. a. Upper jaw. b. Lower jaw. Anterior at left, ot Fig. 5. Istiropsis mako (Whitley). a. Upper jaw. b. Lower jaw. Anterior at right. TAs N. §, PLEDGE However, the tossil record furnishes mainly the teeth, and the vertebrae found are difficult to ascribe to families. The use of the divisions Tectospondyli aod Asterospondyli has been generally discontinued, and it is therefore felt that the use of basic tooth structure as a criterion for classification is defensible. TOOTH MORPHOLOGY AND VARIABILITY (‘Text-figs. 1-5) Sharks’ teeth consist of two main parts —the crows or biting, surface, and the root. The crown inay be solid (ie. osteodont) or hollow (orthodont), The crown consists of the main cusp or cutting blade, which may possess some accessory cusps, In osteodont sharks, these are paired lateral denticles or cusps; but in some orthodont sharks the tecth are asymmetrical, and accessory cusps Occur only or mainly one the posterior side, e.g. “Notidanus” and Geleo- cerdo. The crown is often compressed into a blade, with the labial (outer) face being more planar than the lingual (inner) Face. The margins may be sharp cutting edges or quite blunt. Unless they are interrupted by dentes or serrae, they are termed “entire”. Dentes are squarish “teeth” produced by invaginations of the edge, and serrae are saw-like points; intermediate shapes also occur, Paired Jateral denticles may take a variety of forms, even within the one species: sharp-pointed cones or barely discernible tubercles, straight or curved, broad sharp-edged compressed cones, or wedge-shaped blades. They may also bewr serrac or dentes. Ornamentation of the enamel is often present: plications of the enamel of the outer face, and longitudinal striations on the inner face. The root is of bone-like material. and is usually bifurcated into two more or less distinct lobes which can take a variely of forms, according to position in the jaw. At the middle of the inter side of the root there is often a “median boss" which bears a yertical groove containing the “nutritive foramen”. This is inainly seen in osteodont teeth, but some carcharhinids also have a nutritive pere. Within any one species of shark, variation in tooth form is seen; it may be slight, involving only an increasing obliquity in tha more distal teeth, or radical. as in the hexanchids, where the difference between teeth of upper and lower jaws is extreme. Some generalisations may be made, although exceptions can be found. i. Tecth from the upper jaw are broader than their lower jaw equivalents. 2. Upper tceth show more inclination than the equivalent lower tecth, which are normally erect. 3. ‘There is a gradual decrease in the lungth-breadth ratio of the crowns as one proceeds backwards in the jaw. 4. Tn sharks having “eye” teeth (which are considerably smaller than their immediate neighbours), the anterior teeth are considerably more elongate than those lateral teeth immediately followmg the eye teeth, There may be an increase in the number of latcral cusps on teeth in the distal parts of the jaws; the cusps also become broader and relatively coursér. 6. Small symphysial teeth may occur at the symphysis of either jaw, The living sharks, Carcharias arenarius and Isuropsiy mako ( Text-figs. 4, 5), found in South Australian waters, show many of these features, Others, such as Galeocerdo and Carcharadon, show little variation within the jaw, save inelina- tion and crown proportions. a) FOSSIL ELASMOBRANCHL TEETH OF S.A. 189 Measurements are taken along three approximately perpendicular directions. Generally, crown height only can be measured due to damage to the root, and this is taken in the plane of the outer face, from apex to base, in a line per- pendicular to the base. Width is measured at the base of the crown, where it naturally scparates from the root, Thickness is often difficult to measure be- cause of the configuration of the tooth; it is taken in a line normal to the outer face, as close to the base as possible. TAXONOMIC CHARACTERS The characters that define genera of sharks’ teeth are in some respects quite vague, yet in conjunction they serve to separate the groups fairly well. Features used are: Overall shape and proportions. Presence or absence of serrated margins, Presence or absence of latera! cusps. Shape and number of lateral cusps, if present. Shape of the roots. Specific characters are the same as those listed above, with the addition of: 6. Presence or absence of striations on the lingual face. 7. Presence or absence of plicae in the enamel of the labial face. 8. Contortion or inclination of the crown. In addition to just the presence or absence of these features, specific deter- minations also depend considerably on the degree to which a character is pre- sent. This can, and undoubtedly does, lead to some confusion in identification from inadequately illustrated descriptions, SYSTEMATICS The classification used herein is given below, and is adapted from those of Glikman (1964) and Bertin and Arambourg (1938), Superclass Pisces Class Chondrichthyces Subclass Elasmobranchii Infraclass Orthodonta Order Hexanchida Family “Notidanidae” (Hexanchidae) Order Heterodontida Family Heterodontidae Order Squatinida Suborder Rajoidei Families including Pristidae, Rajidae, Myliobatidac, Dasyatidae Suborder Squatinoidei Pristiophoroidei Orectoloboidei Squaloidei Order Carcharhinida Family Carcharhinidae Sphyrnidae Infraclass Ostcodonta Order Carchariida Family Carchariidae (= Odontaspidae) Isuridae GU GA po po 140 N, $5. PLEDGE DESCRIPTION OF SPECIES Infraclass Orthodonta Family “Notidanidae” Genus “Norwanus” Cuyier, 1817 Diagnosis. Teeth of upper jaw with prominent prime cusp and one or more small secondary cusps on distal side. Lower teeth very wide with numerous secondary cusps giving a saw-like form. “Notidanus” Cuvier includes Iexanchus and Heptranchias Rafinesque, 1810, and Notorhynchus Ayres, 1855, as there is no apparent generic difference between the teeth of these living genera. “Notidanus” serratissimus. Agassiz (Plate 1, Figs. 1-3) Notidanus serratissimus Agassiz, 1843, p, 222, pl. 36, figs. 4, 5, Woodward, 1886, p. 216, pl. 6, figs, 23-26, Diagnosis, Teeth small, five to ten cones with blunt apices; principal cone ouly slightly larger than immediate neighbour, but more robust. Anterior edge produced downwards and denticulated on lower half. Observations. Woodward's figures 24 and 25 (1886) most closely agree with two teeth at hand, although one of these has eleven cones—more than pre- scribed by the diagnosis. In all other respects, including gcologic age, they agree with the figured material. Another specimen, however, from the Naracoorte Bore, exhibits a somewhat different form. The principal cusp is very broad and less acute, and about twice as high as the following cusps which are about the same height. It also has eleven cusps. Four fragments (Pl. 1, Fig, 3) found in close proximity at Blanche Point are tentatively assigned to this species, although they show some variation in the coarseness and style of serration. Occurrence. Blanche Point Marl, at Port Noarlunga and Blanche Point. Knight Formation equivalents, Naracoorte Bore, 426 ft. Age. Middle and Upper Eocene. Material Examined. ‘Three teeth and several fragments. Family PRISTIDAE Genus Pristrs Latham, 1794 Diagnosis. Rostrum of skull long smooth blade with rounded end and parallel sides. Rostral teeth triangular, compressed dorsoventrally, faintly striated lengthwise, occur regularly along rostrum. Pristis cudmorei Chapman (Plate 1, Fig. 6) Pristis cudmorei Chapmun, 1917, p. 139, pl. 9, fig. 7. Diagnosis. Rostral teeth small, elongate, triangular, slightly inclined. An- terior margin sharp, finely denticulated to apex. Posterior edge thick and blunt, Both faces yery finely and irregularly striated for half of height. FOSSIL ELASMOBRANCEHE TEETH OF 5.A,. 141 Observations. ‘The three specimens at hand agree closely with Chapman's description of material of Upper Miocene age, but full equivalence of this neritic form is doubted because of the large time gap between the Victorian and South Australian occurrences. Occurrence, Blanche Point Marl, Maslin Bay. Age, Upper Eocene. Carter's Unit 2. Material Examined. Three specimens. Family MYLIOBATIDAE Genus Mytiopatis Cuvier, 1817 Diagnosis. Dental pavement formed by regular interlocking of hexagonal tooth plates. Tooth plates with comb-like base—numerous transverse parallel blades on articulating surface. Median plates very broad, lateral plates regular hexagons. Myliobatis spp. (Plate 1, Figs, 4, 5) Observations. Two pieces from the Blanche Point Marl, and six pieces of different species from Naracoorte are the only South Australian specimens seen by the author. A fragment of caudal spine (PI. 1, fig. 5) is also referred to this genus, but none of the material has been identified to species level. Qceurrence. Blanche Point Marl, Maslin Bay; Naracoorte E, & WS. Bore No. 5, 426 ft., Knight Group equivalents, Age. Middle and Upper Eocene. Material Examined. Spine fragment from Blanche Point, 6 tooth plate fragments from Naracoorte bore, 2 from Blanche Point. Family CARCHARHINIDAE Genus Cancearuinus Blainville, 1816 Diagnosis. Teeth small, stout; lower teeth narrow, erect; upper teeth broader, inclined. No lateral denticles. Base widely expanded, Cutting margins often finely serrated. Root nearly straight, with central notch, Carcharhinus of. collatus Eastman, 1904 (Plate 1, Fig, 7) Carcharias collata Eastman, 1904, p. 85, pl 32, fig. 3. Chapman and Cudmore, 1924, p. 118, pl. 9, figs, 13-16, Diagnosis. Robust Carcharhinus teeth with cutting edges finely serrated on basal prolongations, very finely on cusp, entire apex, Inner face strongly convex, Enamel of outer face extends much Jower than on inner face. Root elongated, large, symmetrical, Observations. The four specimens at hand agree with the figure of C, collatus given by Chapman and Cudmore (1924), and also with Eastman’s figures of his types from the Miocene deposits in Chesapeake Bay, Maryland, although one specimen is more oblique, and another more robust. Occurrence. The teeth all came from 255 ft, depth in the Marion No. 2 Bore, Sturt River Valley, Adelaide Plains, ‘This depth is in or near the Munno Para Clay member of the Port Willunga Beds, as defined by Lindsay and Shepherd (1966). Age. Lower to Middle Miocene: Batesfordian-Balcombian. Material Examined. Four specimens, 142 N. 8, PLEDGE Carcharhinus (Prionodon) aculeatus (Davis) (Plate 1, Fig. 8) Caleocerdo aculeatus Davis, 1888, p, 8, pl. 1, figs. 1-3, Carcharias (Prionodon) ucyleatus Woodward, 1889. p. 440. Chapman and Cudmore, 1924, p. 119, pl. 9, figs. 19, 20. Diagnosis. Carcharhinus teeth with whole cutting margin serrated, rather coarsely on lateral prolongations, finely on cusp, coarsest in notch; apex almost entire. Observations, This species is represented by a solitary tooth from the River Murray Cliffs. It is similar to specimens of C. collatus at hand, but the serra- tions are considerably coarser and the tooth is slightly larger. The cusp shows a distinct inclination and the anterior notch is shallow. “The tooth compares favourably with the figures of Chapman and Cudmore (1924). Oceurrence. River Murray Cliffs, locality uncertain, preseryation suggests Cadell Marl Lens, Davis’ material came from the Oamaru area (?L. Miocene) and from the L. Pliocene “Awatere Series”. Age. Lower Miocene. Material Examined. One specimen, Wet: wel ‘ Genus Hemurrusris Agassiz, 1843 Diagnosis, Principal tecth large, triangular, compressed; cutting edges coarsely serrated almost to apex. Lower anterior teeth slender, subulate, in- wardly curved, only a few minute serrae at base. Hemipristis serra Agassiz (Plate 1, Fig. 9) Hemipristis serra Agassiz. 1843, p. 237, pl. 27, figs, 18-30. Woodward, 1889, p, 449, Chapman and Cudmore. 1924, p, £15, pl. 9, fig: 12. Diagnosis, Marginal serrations in broad upper teeth large, extending almost to apex. Margins of anterior lower teeth very sharp distally, Toner face of root with deeply cleft median boss. Observations. This species, while couimon in the Tertiaries of Europe, America and Indonesia, is quite rare in Australia, and only twe specimens are known to the author, one found by Cudmore (Chapman and Cudmore, 1924) and the other very recently by Mr. R. J. F. Jenkins. The latter specimen is slightly smaller, but broader than that figured by Chapman and Cudmore. The marginal serrations are coarse, very sharp, and stop about 5 mm below the apex. The crown is inflated, but the external face is depressed at the base; the margins are twisted, convex outwards. Most of the root is missing. Occurrence, Cliff, left bank of River Murray, one mile below Pelican Point, near Morgan, South Australia. Cadell Mar] Lens in the Morgan Lime- stone, Age. Lower Miocene - Batesfordian, Material Examined, One specimen (F17288). Genus CarcHarowres Ameghino, 1901 Diagnosis. Teeth similar to Lamna, but with denticulated margin as in Carcharodon. Lateral denticles compressed, dentate. Orthodont. FOSSIL ELASMOBRANCH TEETH OF S.A, 143 Carcharoides cf. totuserratus: Ameghino (Plate 1, Fig, 11) Carcharnides tatuserratus Ameghino, 1906, p. 183 (footnote), fig. 50. Chapman and Cudmore, 1924, p. 121. Diagnosis. Teeth large; inclined crowns. Large pointed lateral cusps, all margins denticulated., Description of Specimen. Crown elongate triangular, acute, slightly oblique. Margin irregularly dentate, dentes wide and separated by shallow grooves, Outer faec almost Hat, smooth, inner face convex, smooth, Observations. The sole specimen at hand consists only of the enamel shell of the crown, and no root or lateral denticles are present. It is somewhat smaller than Ameghino's type, and less oblique, and shows a flaring of the base that is difficult to visualize in C. totuserratus, It is also similar to C. tenuidens Chapman; Chapman records both forms from Janjukian deposits at Waurn Ponds, Victoria; Ameghino's material, from the Patagonian Series, is now re- garded as being of Lower Miocene age. Oceurrence. Road cutting near Strathalbyn cemetery, Mannum Formation equivalents. Age. Lower Miocene. Material Examined. One specimen. Genus Gateocerpo Miiller and Henle, 1838 Diagnosis. Teeth serrated on both edges. Apex sharply inclined, deep notch posteriorly, coarse serrations below notch, Galeocerdo cf. aduncus Agassiz (Plate 1, Fig. 10) Galeacerda aduneus Agassiz, 1843, p. 231, pl. 26, figs. 24-28, Woodward, 1889, p. 444, Chapman and Gudmnore: 1924, p. 117. Leriche, 1957, p, 38, pl. 45, figs, 18-21, Diagnosis. Anterior margin arched; apex short, broad, sharply inclined, very similar to existing Galeocerdo cuviert but smaller, Observations. There is only one South Australian specimen of this species at hand. At first glance it seems referable to G. lutiélens because of its rela- tively great width and low crown. However, the anterior margin is distinctly arched, and finely denticulate. On reference to a jaw of the living G. cuviert, it would appear that the tooth came from a posterior position, where the crown is relatively low. Another tooth (Plate 1, Fig. 10b), of presumed Miocene age, and from an uncertain locality, is at hand. It differs from the above in having an elon- gated apex reaching almost to the posterior extremity of the tooth. The anterior edwe shows a sharp bend about halfway along its length, and the two halves are straight. As such, it does not fit any species known to the author, but no attempt has been made to erect a new species on it at this stage. Occurrence. Exact locality unknown, but from the River Murray cliffs, probably at Morgan. Age. Lower Miocene. Material. Examined. Two teeth. 144 N, 5. PLEDGE Family TRIAKIDAE Genus Musterus Linck, 1790 Diagnosis, Tecth small, similar, rhomboidal occlusal surface, arranged in oblique rows to form wide curved dental pavement on jaws; many in function at same time: Fig. 6 Fig. 6. Comparison of Mustelus antarcticus and PMustelus sp, . M. antarcticus, occlusal surface, Recent. . PMustelus sp., ditto (17287), U, Eocene. M. antarcticus, postero-occlusa) aspect showing length, . PMustelus sp., ditto. (x 5 approx.) a ec b d cf, Mustelus sp. (Text-fig, 6) Description. A single, small, subrhomboidal tooth was found by Mr. R. J, F. Jenkins at Blanche Point, It is similar in form to a tooth from the living gummy shark Mustelus antarcticus Gunther, but differs in that the posterior corner of the rhomb is truncated, while the anterior truncation is wider. The gently arched occlusal surface of the tooth is granulate, but there is a distinct smooth triangular area at the posterior corner (in the living M, antarc- ticus the whole surface is granulate, more coarsely on the posterior half). The tooth is not as deep as M; antarcticus, and its articulating process is obscured or missing. Observations. This is the first report of any tooth of this form from Aus- tralian Tertiaries, although the Triakidae have a history reaching back to the Cretaceous, and Mustelus itself has been recorded from Oligocene deposits (Arambourg and Bertin, 1958, p. 2036). If the present determination is correct, the genus will thus have its strati- graphic range extended back to the Upper Eocene. Occurrence. Blanche Point Marl, Maslin Bay. Age. Upper Eocene, Aldingan Stage, Carter’s Unit 2. Muterial Examined. Oue specimen, plus example from hving gummy shark. Infraclass OsrEoponTA Family CARCHARIIDAE (= ODONTASPIDAE) Genus CarcHanAs Rafinesque, 1810 (syn. Opontasris Agassiz, 1843) Diagnosis. Teeth with high, narrow, compressed crown; with one or two pairs of lateral denticles, generally sharp pointed. Anterior teeth very high crowned, large and slender, with much produced bifurcated root. Similar to Cretaceous Scapanorhynchus and some forms of Lamna, Cutting edges entire, beginning several millimetres above base of crown, FOSSIL ELASMOBRANCH TEETH OF S,A, 143 Carcharias macrotus (Agassiz) (Plate 3, Figs. 1-8) Lemna eleguns Agassiz, 1843, p. 289, pl. 35, figs. 1-7, pl 37a, figs. 48, 59. Otedus macrotus Agassiz, 1843, p. 273, pl. 32. figs. 29-81 Odontaspis elegans Woodward, 1889, p, 361. Q. macrota Woodward, 1899, p. 9. pl. 1, figs. 19, 29, O. mactota striata White, 1931, p, 58, figs. 45-74. Diagnosis. High narrow crown, slightly curved; outer face smooth, slightly convex; inner face convex but medially flattened, longitudinally striated, striae rather irregular, Single pair of lateral denticles. Prominent median hoss with nutritive pore. Root lobes long, generally pointed. Observations. Teeth of this species show a variety of forms, depending on position in the jaw, and also, to some extent, on age. The denticles show a number of different forms, from fine curved cones to broad compressed cones, to chisel-like blades. Striations vary in continuity und degree, from strongly incised groves to barely discernible narrow facets. Many specimens lack the root and/or lateral cusps, making identification rather uncertain. Several bio- species may be represented. According to White (1931, p. 62) the forms “Odontapsis elegans” (Ag) and O, macrota are the same species, “O, elegans” being juvenile to O, macrota. Two other forms have heen separated in the South Australian material. One is represented by six teeth (V49) from the Naracoorte bore assemblage (Plate 3, Fig. 7), They differ from most specimens in having a quite robust habit with short narrow crowns and strong, thick, widely bifurcated rocts, AI show the typical C, macrotus striations. Three of them are from the upper jaw, having shorter, inclined crowns. All bear short lateral cusps, and fwo of the lower teeth also have a second, minute, inner pair- The other aberrant form (Plate 3, Fig. §) is represented by three teeth {V53) from the Pecl’s Bore assemblage, which differ in size. preservation and roat form from the numerous specimens of C. macrotus also present. Al! three bear strong striations, and lack lateral cusps, One is a compressed symphysial tooth, and the other two have roots with widely divergent, Hattened, round- ended lobes, Not enough is known of these two forms ta put them in other specics. Occurrence, Basal member of the Tortachilla Limestone, upwards ta the lower part of the Port Willunga Beds—Maslin and Aldinga Bays. Peel’s Bore near Robe (390 ft.), and an aberrant form from the E. & W.S. Bore No. 3, at Naracoorte (426 ft.). Alsa from fhe Strathalbyn cemetery road cutting, Morgan type section, and Dry Creek Sands (Tenant's Bore, Salisbury). Age. Common in Middle and Upper Eocene, and sporadically to Pliocene, in South Australia, Material Examined. Approximately 50 specimens from Maslin and Aldinga Bays in A.U, catalogue; about 100 from Peel's Bore and otber places, held in the Geological Survey collection, including the aberrant forms (V49 and 53). Carcharias contortidens (Agassiz) (Plate 3, Fig, 1L) Lamna (Odontaspis) contartidens Agassiz, 1843, p. 294, pl. 37a, figs. 17-23, Qdontuspis contortidens Woodward, 1889, p, 366, Chapman and Cudmore, 1924, p. 122. Diagnosis, Teeth slender; delicate longitudinal striae on inner face, Sig- moidal curvature of anterior teeth and median boss of root more pronaunced than in C. macrotus. lt N. S, PLEDGE Observations. A large number of teeth from Peel's Bore were separated from C. macrotus hy their greater slenderness and sinuosity, The striations are often finer and more uumerons than in C. macrotns and extend nearer to the apex, Miocene specimens from the Marion No. 2 Bore are similar to, but cousiderably smaller than those from Peel’s Bore. Occurrence. Peels Bore (390 ft,), near Robe; Mario No. 2 Bore (255 £t.), Sturt River Valley; Blanche Point Marl, Aldinga Bay; and Dry Creek Sands in the Abattoirs Bore. Age, Middle to Upper Eocene, Miocene, Pliocene, win rm Examined, About 50 specimens, mainly from Peel's Bore (VL4, Carcharias maslinensis sp, nov. (Plate 2, Figs. 1-8) Diagnosis. Teeth with very slender crown, circular cross-section al base. Anterior teeth with very sinuous elongate crowns, twisted, Anterior cutting cdge more prominent. Regular, deeply-incised,, longitudinal striatious. on imner face, almost to apex. Pair of minute, conical lateral denticles, well separated from crown, Root of anterior teeth with strong median boss. Lateral and distal teeth straighter, erect, having roots with widely divergent Hattened Iwhes. Description of Types Holotype (F17260) (Pl. 2, Figs. 1, 2). A typical anterior tooth of Car charias, but distinguished by extreme slenderness, prominent sinuosity and axial twist. Jt is almost circular in section at the hase, where the cutting edees are nonexistent, but becomes semicircular towards the apex, with the outer face being almost flat. Anterior cutting edgé begins about two millimetres above base and is quite sharp; posterior culting edve begins aboul 5 uum above hase but is less prominent and rather blunt. The outer face is smooth except for a small sub-median plica, the inner face is strongly convex and longitudinally striated to within 3 mm of the apex. Striae are fine, regular, almost parallel, bifureating towards the apex. A very small, curved-conical lateral cusp js preserved on the anterior side of the crown, some distance beluw the base, The root is strongly bitid with narrow lobes at about 40°, separated by a prominent median boss with a deep nutritive pil at the apex. Dimensions. Total height 20-8 mm; crown height (alony outer face) LAS mm, inward displavement of apex ce. 10 mm; base width 4°35 nun; base thiek- ness 4-0 mm, Paralypes (a) Anterior teeth (V35) (Pl. 2, fig.3). Similar to holotype, but most have lost the lateral cusps. (b) Lateral teeth (V1S) (PL 2, figs, 4a, 4b). A number of teeth found in the same Naracoorte bore assemblage are deemed to represent the Iateral and posterior teeth of C. muastinensis sp. nov. because they are similar ta the anterior teeth in slenderness, stviation, and lateral cusps. They ditter in being evect, with equal margins, a relatively short crown, and widely-bifurcating, Ilat- lobed roots Two groups cau be separated; one, having relatively wide main cusps, presumabh- represents teeth from the upper jaw, the other, lower jaw teeth, This. conclusion is. based on observations on the variation of teeth in the jaws of the living Carcharias arenarius Ogilby, which observations also explain the differences in roal form and the shortness of the crown in the lateral teeth, FOSSIL ELASMOBRANCIL TEETII OF S.A. 147 Judging by the gradation in size and form of both lateral and anterior tecth and the preservation, it is considered that they might represent one individual, Occurrence, E, & W.S. Bore No, 5, Naracoorte (426 f[t.). Depositories of Types. Holotype: Dept. of Geology, University of Adelaide, F17260. Paratypes; Geological Suryey of South Australia, V15, V35. Observations. The teeth differ from Lamne attenuata Davis, which Chap- man reported (1918) from the Lower Aldingan Beds, by possessing striations on the inner face, and a twisted sigmoidal shape of the crown, Chapman and Pritchard (1904) describe Odonlaspis dttenuata (Davis) as having a striated inner face, but none of the material seen in the Victorian National Museum agrees with the specimens of C. maslinensis. The lateral tecth of C. maslinensis are similar to yery smal] specimens of C. macrotts. TABLE 1. DIMENSIONS OF TYPES Anteriars Upper Laterals Lower Laterals height | width | thickness height with, | thickness] hei hit width | thickness is holotype paratyped paratypes 16°8 4-5 4-0 $3 4e8 2-4 1-7 3°09 2:3 72 4+] 2-0 79 340 2+5 paralypes 16-1 4-9 4-1 6-1 3:3 2-0 7-0 2-4 Be, 12+1 Bb 2d 648 3-4 2-0 5-8 #0 1-7 12-1 h-6 2:9 7-0 3:7 23 13 2-8 2-L iO a+4 2-5 6-5 2-8 2-1 5-8 24 1-7 9-9 he | 2-3 Ad) 27 1-8 T- a0 2-3 12-6 3-0 2-5 5-1 2-9 1:7 yao | 3:0 2-0 Red 2-8 2rd 4 33 1-5 fo 2:9 1:6 a-6 2-3 1:7 7+6 4-5 23 C. maslinensis is commonly found in the Eocene deposits of South Australia, and appears to have undergone some evolutionary change, as specimens from the Blanche Point Marl and Port Willunga Beds (lower part) are stouter than those from the older Tortachilla Limestone and Knight Formation. It is a common constituent of the assemblage of crowns in the basal grit member of the Tortachilla Limestone, but most identified specimens are of anterior teeth, for it is difficult to distinguish the lateral tecth on crowns alone, Three poorly pre- served specimens from the Strathalbyn assemblage could page 28 be reworked from underlying Eocene strata during the Miocene transgression. Occurrence. Peel's Bore, 390 ft, Co. Rohe, Hd, Ross, Sec, 19; E. & WS. Bore No. 5, Naracoorte, 426 ft: a farm bore near Lucindale, 200 ft; a farm bore near Kingston, 715 ft; basal grit of Tortachilla Limestone at Maslin Bay, basal part of Muloowurtie Formation, Yorke Peninsula, Blanche Point Marl; lower part of Port Willunga Beds; and Strathalbyn (cemetery road cutting) Manoum Formation equivalent, Age. As recorded so far, C. maslinensis ranges from Middle to Upper Eocene, and possibly to Lower Miocene, unless reworked. Material Examined. About 50 specimens from south-eastern bores, includ- ing the types F17260, V35, and V18. Another 50 crown enamels from the Tortachilla Limestone, and several teeth from other levels. 148 N. 5, PLEDGE Carcharias cf. cuspidatus (Agassiz) (Plate 3, Figs. 5, 18, 13) Lamna cuspidata Agassiz, 1843. p. 290, pl, 37a. figs, 43-30, Odontaspis cuspidata Woodward, 1889, p. 125. White, 1931, p. 53, figs. 19-44, Diagnosis, Teeth very similar to C. macrotus oxcept for absence of striae on inner face, Sometimes more robust than corresponding teeth of C. macrotus, but smaller. One pair of small, pointed, lateral cusps. Nutritive foramen in deep cleft. Observations. A large number of teeth, many from Peel’s Bore, fit in this diagnosis, but most are wanting roots and/or lateral cusps. Those which are complete come within the range of variation for the species as illustrated in White's figures, but altogether they make up a rather heterogeneous collection which may represent several species. It has not heen possible to separate them into more satisfactory groups. Occurrence. Peels Bore, 390 ft. (Knight Formation); basal part of Tor- tachilla Limestone, Blanche Point Marl, Port Willunga Beds in the Willunga Basin; and Strathalbyn (Mannum Formation equivalent). Age. Middle Kocene to Lower Miocene, in South Australia, Material Examined. Almost 100 specimens, including about 50 from Peel's Bore, and many crown enamels from the Tortachilla Limestone, and Strathalbyn, Carcharias cf. rutoti (Winkler) (Plate 3, Pig, 14) Otodus futoti Winkler, 1876, p. 6. pl, 1. figs. 3, 4, Odontaspis rutati Woodward, 1889, -:p. 361. Chapman and Cudmore. 1934, p, 125, pl, 10, fiz, 97, White, 1931, p. 49, fizs. 4-12. _ Diagnosis, Teeth small; robust but not very high, with faces smooth, but there may be vertical plications on outer face. At least two pairs of sharp lateral cusps—outer ones insignificant. Root with prominent median boss. Outer base-line of crown re-entrant. Observations. Small specimens similar to C. rufoti have been recovered from the Blanche Point Marl, where the appurently casily detached lateral cusps are preserved. No specimens have been seen having three pairs of lateral cusps as in Winkler’s description, but from White’s figures, such occurrences are un- common—only one of his nine figures shows three pairs of cusps. One specimen ascribable to the species was found in the Peel's Bore assemblage. On the specimens available, the main cusp is more or less inclined, fairly broad, and relatively thin. The denticles are conical, sometimes compressed, and sometimes incurved. towards the main cusp. Chapman and Pritchard (1904) recorded similar teeth from Victoria as “Lamna bronni’, which is a Cretaceous form, and it has been suggested that C. rutoti is the correct determination, Occurrence, Blanche Pomt Marl, Blanche Point, Maslin Bay, Knight Formation, Peel's Bore (390 ft.). Age, Middle and Upper Eocene. Material Examined. Two teeth from Blanche Point, one from Peel's Bore. FOSSIL ELASMOBRANCH TEETH OF §&.A. 149 Carcharias dubius (Agassiz) (Plate 3, Fig. 10) Luning dubia Agassiz, 1843, p. 295. pl. 37a, figs. 24-26. Odontaspiy dubia Clikman, 1964, pl. 25, figs. 12, 13. 16, 17, 19» pl 27, Ags, 18-23. Diagnosis. Teeth very similar to C. contortidens, but lack striations on inner face. Crown rounded, subeylindrical at base, not flattened towards, apex. Single pair of lateral cusps: cylindrical, clongate, very acute. Root and lateral cusps rarely preserved. Observations. A number of teeth having similar characters to this form were found in the Peel’s Bore Assemblage. The crowns are elongate and acu- minate, with little or no. cutting edge on the lower part. Width is only slightly greater than thickness. Rovt large, prominent median boss, lobes at about-90°. The teeth show signs of being polished, but not of excessive wear, They are similar to those figured by Glikman (1964) except that the root is not as broad. Occurrence. Peel's Bore (390 ft.), County Robe, Hd. Ross, Sec. 19. Age. Middle Eocene. Material Examined. Five teetly from Peel's Bore. Carcharias sp. indet. Observations. Five entirely unworn tecth, having some resemblance to C. dubius and C, cuspidatus, but differing in detail, were found in the Naracoorte Bore Assemblage. The teeth are small, having narrow erect crowns and a pair of large conical lateral cusps. They differ from C. dubius in having sharp cutting edges, also on the denticles, and widely divergent root lobes, The median boss is pro- minent. One tooth is from an anterior central position, while another is posterior and is developing a second pair of denticles, Qceurrence. E. & W\S. Bore Na. 5, Naracoorte (426 ft), Knight Formation equivalent. Age. Middle Eocene. Material Examined. Five teeth from the Naracoorte Bore. Carcharias cf. ensiculatus (Davis) (Plate 3, Fig. 15) Lamna ensiculaia Davis, 1858, p, 18, pl. 3, figs. 6, 7. Diagnosis. Teeth small, strong, erect, Crown enamel divides just above hase, at point marked by constriction, and extends over upper part of root lobes. Faint indication of lateral cusps. Observations, This form is represented here by a single specimen which well illustrates the peculiar flaring and bifurcation of the lower part of the enamel. No lateral cusps can be seen, The root is strong and deeply bifurcated at about 90°; the inner side is flattened and bears a prominent nutritive pit, However, the root is not as large as in the figures of Davis. His material came from the “Oamaru Formation” (Oligocene) of New Zealand, which is consider- ably younger than this specimen, Occurrence. E. & W.S. Bore No. 5, Naracoorte (426 ft.). Age. Middle Eocene, Material Examined: One specimen. 150 N. $. PLEDGE Genus CancuAnroLamNa Hora, 1939 Diagnosis, Teeth having form of Lamna and bearing close resemblance to Carcharaides, but being distinguished hy osteodont structure, less developed lateral cusps, finely denticulated margin, and bhint apex. Carchariolamna cf. heroni Hora (Plate 3, Fig. 16) Carchariolamna heront Wora. 1939, p, 203-203, pl. 13, figs. 1-4. Diagnosis. Teeth similar to Carcharias or narrow specimens of Lamna sp., erect, broad base, apex blunt; margins finely denticulate, from apex almost to base. Jateral denticles small, blunt, incipient. Root broad. Observations. The specimen at hand consists only of the crown of a taoth, It agrees with Hora’s description and figures, cxcept that the apex is more acuminate, and no flaring is seen at the base. The margin is well preserved, and differs from that of Carcharoides in the fineness of the dentes, Occurrence. Marion No, 2 Bore, 225 ft., Sturt River Valley. Another tooth of this form was found by Howchin in a quarry at Mt. Gambier. Age. Lower to Middle Miocene (Batesfordian-Balcombian). Material Examined. Two specimens. Family ISURIDAE Genus Lamma Cuvier, 1817 Diagnosis. Teeth as for Carcharias Rafinesque, but principal cusps broader, and lateral cusps larger and stouter, Cutting edges do not extend to base of crown, Lamna obliqua (Agassiz) (Plate 4, Fig. 1) Otodus obliquus Agassiz, 1845, p. 267. pl. 31, 36, figs, 23-27 Lamna oblique. Woodward, 1889, p, 404, White, 1931, p. 46, fig. 3. Diagnosis. Teeth robust, large; crown moderately compressed, acuminate, one pair of broad Jateral cusps, sometimes a second smaller pair in distal teeth, Both faces smooth, Rarcly, cutting cdges faintly denticulated, Prominent median boss, nutritive foramen not in groove. Observations. This species is rare in Australian Terltiarics, and only two specimens have been available. The larger specimen is a distal tooth having a short crown and two pairs of stout lateral cusps. The root is stout and widely bifurcated with Daintel: lobes. The other specimen, from Peel's Bore, is small, more oblique, with one pair of small denticles. Qccurrence. The larger specimen is part of an old collection and is labelled “Miocene,? Deepwell, Murray Scrub”. However, the preservation is similar to that of teeth from the Knight Formation of the Murray Basin, The smaller specimen is from Peel’s Bore, near Robe, Age, Middle or Upper Eocene. Material Examined. Two specimens. FOSSIL ELASMOBRANCH TEETH OF S.A, 131 Lamna cf. apiculata (Agassiz) (Plate 4, Fig. 2) Otedus apiculatus Agassiz, 1843, p. 275, pl. 37, figs. 32-35, Lanma apiculate Chapman, 1918. p. 13, pl. 5, figs. 17-20. pl. 6, fig. 4, Chapman and Cudimore, 1924, p, 126, pl, 10, figs, 28, 29. Diagnosis. Moderate size, triangular, relatively thin, more or less oblique; outer face with some small vertical plicae at base; pair of short, blunt, biconvex luteral cusps. Root strong; widely divergent, pointed lobes. Detached crowns very similar to Isurus hastalis, but have greater median thickness. Observations. Only a few of the specimens at hand still retain the lateral cusps; identification of others is uncertain. Many specimens are small, and may represent a different species, as they occur in only one horizon—the basal part of the Tortachilla Limestone. The form occurs. sporadically through the Tertiary. Occurrence. Basal member of Tortachilla Limestone, Maslin Bay; Blanehe Pojnt Marl at Aldinga Bay, Port Noarlunga and Moana; Strathalbyn; Morgan Limestone Lype section (Cadell Marl Lens); Morgan township; base of T.oxton Sands, Waiketie. Age. Upper Eocene to Pliocene. Material Examined. Fourteen specimens. Lamna cf. erassidens Agassiz (Plate 4, Fig. 4) Lamnea. crassidens Agassiz, 1843, p. 292. pl. 33, figs, 8-21, Odontaspis (?) crassidens Woodward, 1889, p. 373. Lamna cressidens Chapman and Cudmore, 1924, p, 126, Diagnosis. Teeth robust, large, rather spatulate outline; outer face gently convex; inter face strongly convex, medially depressed. Root strong, long widely divergent lobes. Crowns yery similar to Istwus desorii, Observation. A single tooth fits this description. The crown is erect, and spatulate, with the inner face strongly convex but medially depressed. It is placed in Lamna because the cutting edges do not extend to the base, and it is tuo robust to be put in Carcharias, There is no sign of lateral cusps, The root has widely divergent lobes. Occurrence. River Mnrray cliffs, probably near Morgan. Ave, “Lower Murravian” ie, Lower Miocene. Material Examined. One specimen. Lamna vincenti (Winkler) (Plate 4, Fiz, 3) Lama compressa Agassiz (pars). Agassiz, 1843. p. 290, pl. 37a, figs. 41-42. Lamnea vincenti (Winkler), Woodward, 1899. p. 10. pl. J, Ags, 21, 22. Leriche, 1905, p. 125-127, pl 6, fus. 36-51. Leriche. 1936. p. 390, pl, 2G. figs, 14, 15. Diagnosis. Teeth strongly compressed, crown expanded at base, inner face smooth: upper teeth oblique, lowers erect; one pair of Jateral cusps, moderate size, well separated from crown, divergent; small outer pair of cusps in more distal teeth. Observations. Two undoubted specimens of this species were found in the Blanche Point Marl by Mr, R. |. F. Jenkins, and several incomplete teeth may also be referable to the species. Tt is distinguished from the fourm C, cf. ruloli, whicli is also present, by the compressed crown and straight outer base-line. Occurrence. Blanche Point Marl, Blanche Point Age. Upper Eocene, Carter's Unit 2. Material Examined. Two complete teeth and several crowns. 152 N. 8. PLEDGE Genus [surus Rafinesque, 1810 Diagnosis. ‘Teeth generally large: crown compressed, triangwar, acute, erect or slightly oblique, margins entire, extend right to base; no lateral cusps. Root thick; short, rounded lobes. Tsurus cf. hastalis (Agassiz) (Plate. 4, Fig. 5) Oxyrhina hastalis Agassiz, 1843, p, 227, pl. 34, figs, 3-13, 15-17. Woodward. 1889, p, 385. Istetus hastalis Ishiwara, 1921, p, 62, pl, 10, figs, 1-32. Chapman and Cudmore, 1924, p, 129. pl. 10, fiz. 30. Oxythina hustalis Leriche, 1957, p. 27, pl. 3, figs. 1-8. Diagnosis, Teeth large, broad, compressed; outer face flat or concave, sometimes with plicac; root short, almost straight, blunt-ended, Anterior teeth very large, triangular, thin erect; upper laterial teeth inclined. Lower teeth thicker, more erect. Observations. This species has been the subject of much “s liting and “lumping”. Agassiz’s species Oxyrhina hastalis, O. triganodon, and O. plicatilis are now commonly included in the synonymy, and Ishiwara (1921), for instance, has included a number of Jordan’s Californian species, As a result, Isurus hastalis consists of a number of forms, and allocation to the species is sometimes uncertain. Reference to the teeth of the living Isuropsis mako (Whitley) (= Isurus glaucus (Miller and Henle) ) shows that such a wide range of forms is to be expected, Few undoubted specimens of J. hastalis arc al hand from South Australia, although they are abundant in the Victorian Miocene. The teeth show a similar range in form to those figured hy Ishiwara (1921), but they are often con- siderably smaller. They also have a long time range, and may therefore represent several species. Some specimens assigned here may be incomplete teeth of Lamna apiculata, Occurrence, Blanche Point Marl, and lower Port Willunga Beds in Aldinga Bay (uncertain identification); Strathalbyn cemetery road cutting; Cadell Marl Lens, Morgan type section; Morgan township; Mannum; Mt. Gambier, Age. (?) Upper Eocene to Middle Mioccne. Material Examined, Eight good specimens and numerous enamel fragments from Strathalbyn. Tsurus desorii (Agassiz) (Plate 4, Fig. 7) Oxyrhina desorii Agassiz, 1843, p. 202, pl. 37, figs, 8-13, Woodward, 1889, p. 382. Tsurus desorit Chapman and Cudmore, 1924, p. 132, pl. 10. fi, 32. Oxyrhina desorit Leriche, 1957. p. 26, pl. 44. fixs. 18-23. Diagnosis, Teeth with crowns narrower than I. hastalis, broader than T. retroflexus, sigmoidal curvature in upper anterior teeth. Outer face nearly flat, inner very convex, but medially depressed, Observations. The few specimens available are generally in a poor state of preservation, consisting of detached crowns only, There are few really distinctive characters, but the teeth fit roughly between the limits set. However, there 2 ap to be a tendency for the forms hastalis, desorii aud retriflexus to interyrade, and these determinations must therefore he regarded cautiously. FOSSIL ELASMOBRANCH TEETH! OF S.A, 153 Occurrence. The better preserved teeth are of uncertain origin, but appear to have been collected from the Murray Cliffs—possibly from the Cadell Marl Lens near Morgan. Others, consisting of enamiel only, were collected from the road cutting near the Strathalbyn cemetery, ic. Mannum Formation equivalent. A tooth ascribable to I, desorit was also found in the Peel’s Bore Assemblage. Age. Middle Eocene (?), Lower Miocene, Material Examined. Hight specimens. Isurus retroflexus (Agassiz) (Plate 4, Figs, 8, 9) Onyrhina retroflexa Agassiz, 1843. p. 281, pl. 33, fig. 10. Woodward, 1889, p, 389. Isurus retroflecus Chapman and Cudmore, 1924, p, 130, pl. 10, fig. 31. Oxyrhina retraflexa Leriche, 1957, p. 30, pl, 45, figs. 9-11. Diagnosis. Isurus teeth, rather short, robust, narrower than J, desorii. An- terior crowns strongly incurved. Inner face very convex, not depressed; outer face slightly convex in lower teeth, flat in upper teeth. Lateral teeth curved backwards to angle of jaws. Observations. Teeth of this form are very similar to those of I, desori, but are marked by greater incurvature, and by convex outer faces. One specimen of an upper anterior tooth is almost complete, and shows a basal expansion of the crown enamel onto the root. Occurrence. One specimen, a lower anterior, was found in limestone at Myponga in a posthole at the Myponga cheese factory. The others are from less well defined localities: Aldinga Cliffs, Murray Cliffs, according to their labels, Ave. Upper Eocene (?)—Oligocene, Lower Miocene. Material Examined. Seven specimens. P Isurus sp. (Plate 4, Fig. 10) Description. A number of similar teeth, represented only by the enamel of their crowns, have been found in two different formations, The crowns are elongate, narrow, relatively thin, and slightly inclined. The apex is blunt. The margins are entire, reaching right to the base, but are blunt; although worn, they show little sign of having been sharp. The outer face is slightly convex, medially depressed at the base, and extends below the normal base-line in a wide “tongue”, The inner face is more convex, smooth, The slenderness and the character of the outer face make its allocation to Tsurus rather uncertain. Occurrence, Basal grit member of the Tortachilla Limestone, Maslin Bay; and Strathalbyn cemetery road eutting—Manmnum Formation equivalent. Age. Upper Eocene, and Lower Miocene. Material Examined. Fifteen specimens. Genus GArncnarovon A, Smith (in Miller and Henle, 1838) Diagnosis. Teeth large to very large, triangular, serrate or dentate margin; outer face flat, inner face convex; sometimes with serrate lateral cusps. fd N, 5, PLEDGE Carcharodon megalodon (Charlesworth) (Plate 4, Fig. 12) Carcharias megalodon Charlesworth, 1837, p. 225, fig, 24. Carcharodon megalodon Agassiz, 1843, p. 247, pl. 29, McCoy, 1875. Dec, 2. pl, 11. fig. 4, p. 9 Woodward, 1889, p. 415, Ishiwara, 1921, p. 65, pl, 10, fe. 33, pl. 11, figs. 1-6. pl. 12, figs. 1, 2. Chapman and Cudmore, 19294, p. 133. Lericha, 1957. p. 32, pl. 3, figs. J-1S. Diagnosis. Very large, broad triangular, robust; outer face fat; apex slightly everted; no distinct lateral cusps. Margin with rounded dentes. Observations, This is one of the best known of fossil sharks’ teeth, mainly because of its large size. The sharks growing these teeth have been estimated to haye reached 30 metres in length, with a jaw gape of two metres. Although found in large numbers in some parts of the world, only relatively few teeth have been found in South Australia, An upper right lateral tooth, in the Depart- ment of Geology collection, has a crown about 45 mm high, and shows marked inclination towards the posterior, A fragment of enamel from the Strathalbyn locality may also belong to C. megalodon. The teeth are marked by a spatulate outline, the margins being convex towards the apex, and concave towards the base. Occurrence. The specimens at hand come from Barber's Quarry, Mount Gambier, and also trom Strathalbyn. Others have heen found at Pt, MacDonnell, Lake Bonney (S.E.). Morgan, and. Blanchetown, and elsewhere in the Murray Basin, Age. Lower to Middle Miocene, Material Examined. Seven specimens, iichiding five in the South Australian Museum, Carcharodon auriculatus (Blainyille) (Plate 4, Fig. 11) Squalus anviculatis de Blainvile, 18L8. p, 384, Carcharodon auriculatus Agassiz, 1843, p, 254. pl. 28. figs, 17-19. C_ angustidens Agassiz, 1843, p, 255, pl. 28, figs. 20-25. pl. 30, fies. 2, 3, McCoy. 1875, Dee. 2. pl LL, figs. 2, 3. p. 5. GC. auriculatus Woodward, 1889. p. All, Chapman and Cudmoere, 1924. p. 134. C. angustidens Leriche. 1957, p. 32, pl. 3, fig. 14, Diagnosis. Teeth robust, smaller and narrower than C. megalodon; single pair of serrated lateral cusps, especially large in lateral teeth, which have very narrow, rather oblique crowns. Observations. The specimen at hand is almost complete, lacking onlv the apex of the crown. The crown is crect, and narrow in its upper part, but flares suddenly about 1 cm above the base. The inner face is convex but medially depressed. he Jateral cusps are conrpound, ic, their outer edges lave two or more major points, all of which are serrated. The serrac are quite pointed, whereas the specimens of C, mevalodon at hand have rounded dentes. The root is strong, with widely divergent, flattened lobes. Occurrence, Port Willunga Beds, Aldinga Bay. Others haye been found at Mount Gambier and Tailem Bend, Age. Olivocene to Lower Miacene. Material Examined. Seven specimens, including six in the South Aus- tralian Museum. FOSSIL, ELASMOBRANCH TEETH OF 5,A, 155 Carcharodon carcharias (Linnaeus) (Plate 4, Fig, 13) Squalus carcharias Lirmaens, 1758, p, 235. Carcharodon rondeletii Miller and Henle, 1841, p. 70. Woodward, 1889, p. 420, Carcharodon carcharias Ishiwara, 1921. p. 68, pl. 17, fig, 3. Chapman and Cudmure, 1924, p. 135. Diagnosis, Carcharodon teeth, but relatiyely small, reaching 40 mm height; broad triangular, thin, straight margins, no lateral cusps. Observations. The specimen at hand consists of the main part of the crown. The margins Hare only slightly at the base, and are irregularly denticulated with well-separated rounded dentes. The outer face is slightly convex, but medially depressed. This species is still found in South Australian waters, and has world- wide distribution, Occurrence. Dry Creek Sands, Abattoirs Bore, Hd. Yatala. Age, Upper Miocene, Recent. Material Examined. One specimen, plus jaw of living shark. Carcharodon sp. indet. (Plaie 4, Fig. 14) Observations. A small tooth, obviously belonging to Carcharodon sp., has been found in the Blanche Point Marl. The apex and much of the root are missing, and the total height of the remainder is about 15 mm. The crown is uite thick (6 mm), and almost complete, except for apex and extremities of the base, and is inclined to the posterior. Only one of the cutting edges is preserved intact, and shows moderate-sized round-ended dentes. The tooth evidently comes from the posterior part of the jaw. Of the root only the central portion remains, preserving a median boss with a shallow groove. Occurrence. Blanche Point Marl, Blanche Point. Age. Upper Eocene—Carter’s Unit 2. Material Examined, One tooth. Jenkins’ collection, STRATIGRAPHIC DISTRIBUTION It is seen that, although sediments covering most of the Tertiary are present in South Australia, the known distribution of fossil sharks” teeth within them is discontinuous. This is mainly a collecting bias, due to favourable lithology for preservation and subsequent discovery, although assemblages from bores come from discrete levels. A number of assemblages from units in which fossil sharks’ teeth are rela- tively plentiful are listed below. They are treated in stratigraphic order. a. Knight Formaiion (lower part): Pecl’s Bure, Co. Robe, Hd. Ross, Sec. 19, 890 ft., E. & W.S, Bore No. 5, Naracoorte, 426 ft. Age. Middle Eocene, Species. “Notidanus” serratissimus, Myliobatis spp., Galeecerdo sp. indct., Carchorias macrotus, C. contortidens, C, maslinensis sp. nov., C. cuspidalus, C. rutoti, C. dubits, C. ensiculatus, Lamna obliqua, Isurus cf. desorii. b. Tortachilla. Limestone (basal grit), Maslin Bay. Age. Aldingan (Upper Eocene), Carter's Unit 1. L663 N, & PLEDGE Species. Carcharias macrotus, C, euspidatus, C. maslinensis sp. nov. Lamna cf. apiculata, Isurus (?) hastilis, Isurus (2) sp. These are all preserved as only the enamel shells of the crowns. e. Blanche Point Marl. Blanche Point and other places. Age. Aldingan, Carter's Unit 2. Species. “Notidanus” serratissimus, Pristis cf. cudmorei, Myliobatis sp., Carcharias macrotus, C, contortidens, C. ct. cuspidatus, C. cf. rutoti, C. masli- nensis, Lamne apiculata, L. vincenti, Isurus cf. hastalis, Carcharodon sp. indet. d. Strathalbyn, cemetery road cutting (IId. Bremer, Sec. 2548). Formation. Sublittoral equivalents of Mannum Formation. Age. Lower Miocene. Species. “Notidanus” sp. indet. (fragmental), Curcharoides ef. totuserratus, Carcharias macrotus, C, ct, cuspidatus, C. (2) maslinensis, Lamna apiculata, Isurus hastalis, 1. desorti, Isurus (?) sp. (same as in Tortachilla Limestone), Carcharadon megalodon. e. Marion No. 2 Bore, Sturt River Valley, Hd. Adelaide (depth 255 ft.), Formation, In or near Munno Para Clay member. Port Willunga Beds. Age. Batestordian-Baleombian (Lower Miocene), Carter’s Units 9-10, Species. “Noatidanus” sp, indet. (fragment), Carcharhinus cf. collatus, Carchariolamna cf. hereni, Carcharias contortidens, C_ cuspidatus. _ £. Cadell Marl Lens, Morgan Limestone (type section, 4 miles downstream from Morgan). Age. Lower Middle Miocene, Carter's Unit 10. Species. Carcharhinus aculeatus, Galeocerda aduncus, Hemipristis serra, Carcharias macrotus, C. ef. cuspidatus, Lumna apiculata, L. crassidens, Isurus hastalis, I. desrii, [. retroflexns, Cavcharodon megalodon. g. Dry Greek Sands, as in Abattoirs Bore, Tennant’s Bore. Age. Upper Pliocene (Yatalan). Species. Myliobatis mourabbinensis,* Carcharhinus (Prionodon) aculeatus,* Carcharias ct. macrotus, C, contortidens,? Lamna sp., Carcharodon carcharias. As the accompanying Table 2 shows, some species are restricted to rela- tively short-lime ranges. This is partly an artificial zonation, caused by differen- tial preservation and paucity of specimens. Some species have short time ranges on a world seale, and the specimens at hand fall into them; others have long time ranges and may be represented by only a few specimens. Im several cases,. teeth apparently fall outside the accepted time ranges for the species with which they are compared. The following species are apparently restricted to the Middle Eocene in South Australia: Carcharias dubius, C. ensiculutus, Lamna ubliqua. Apparently restricted to Upper Eocene: Pristis of. cudmorei, Lamina vincenti. Found in both Middle and Upper Eocene: “Notidanus” serratissimus, Car- charias maslinensis, C. cf. rutoti. * These species were reported hy Créspin and Cotton (in Miles, 1952) from the Abattoirs Bore, but the matcrial has not been seen by the author. 157 FOSSIL ELASMOBRANCH TEETH OF S.A. Pieistocene dune limesands |/Blanchetown Clay [Hindmarsh Clay . _ fie & pases a F af 5a Ps “ oe, ae $10 Py | Yatalan | Norwest Bend Fm flee SOry Ck 3 se ee Math we oe we BM Oe gus arte fe tte b Wocene Katimoan [Toxton Sands | Sk Sk Jeter ge GP AR ob rb New aeerden et bal eat acted 4 - Boukpurnong Bed eft 6 ca me the ne hw Bg pe teed ry ee 4 + pe pete ne Fe ewe Mut eae fl eet fe 4 U.Miocene Mitchellian ar a er ae] Pa ee i oe oe oe Pe ene ey Rg ie ae ee ead ty ee pies gp Pte Bairnsdalian Pata Lst. « ‘ . t+ * ‘ . i a ny * rene + “2 A te oe ee M.Miocene a Gy WE Ww an e® “dateetlate 4p PR od ain Aether 4 Balcombion wt Os le Se eee ee ate 9 le CR) Fe nt ot Morgan ist, p Pare] * x x ae 18-19 Na “- 4 og ae WS gi tcrg* «8 *eg wget ene +} 'e ™ ro. e S we » = - 4 wey . 2 a ee sew Mw he * | L Miocene & | wh Pretec fae gw arly Ste te Fi the ee et bah 3 Mannum Fm, Port at Fes Ee oe ee ns cat ge e 5 Ettrick 7 3 ‘ es 98 @ sf «4 Cr ey . Dew % tre ta FH U Oligocene 3 Witlunga Ae on GS a Ba a hk whee RP es cals be She be oe § Fm. wt gg! ete ge fe OIE ee ee, . oes . ~ ' , 4 . al-32 Janjukion an gk gt ee cmt ett dip diate te pupae Cartons LiOligocene s Fa + wh bee ue EO ee ede te b> Wee Dee f *5 lh - ¢ = a g Beds 8 os one = 8 b&b Bem ew ee ta idee ee ad Soe ay-a6 Ranmork = sport pe lap AS Ae a ae ea ee eee nee eee EB BS — 5 7 tg Woks Gg Bitte 2 hie Bana ete Bees FS] Krisns Buccs et td db Bie gen gee tae BERS U Eocene 2 Blanche Point SP te Pa ee 2 EE eB tees op pet ot SS ra Beds Marl ar o- a3 -- - . oe . . . ee ste . . soe ome as oe , wr shal 2 x i Be = ee : TY] Sth Maslin Ss. EE at ete wat pres sues ZB Q% Johonnian Formation 4: a Se ye ee Dele Fe te hee TP ERSB M Eocene Pi EZ 3 8 Te SISESAI RL Be oy ee « Add ws’? ----- 4 R a eR yrs. et ae Austration ||| GAMBIER MURRAY — |] St. VINCENT § eg £ 7 ¥ 3 g gz Teazs Stages | ll] EMBAYMENT BASIN BASIN e *¥ = 38 38 € * 4 P 5 3 ~~ as uu > a & Re § 2 & & iS 2 Eee e ge $$ oe See a Suv © e s es ¢ * S&S 2 “Z| corretetion ofter Ludbrook(1967) Correlation % ea» z = 2 o£ = ef ges $ : $32 b 835 Bek s = of the Tertiary Rocks of the Australasian Keyion”, 4 & ae & & 5 eb ae. & a8 fr “yy Sind oz 5 5 3 & in “Tertiary Correlations ond Climate Changes in BS % 3 & 3 = g $8 B35 2 S$ eust 28a 3 ry seo - E = i. = a if & rm a 3 a3 ; the Pucific, Symposim 25. Sendai, Jepan, 22 ¢ FS 8u & Bd & Sucduuu § Sana 8-2 Sus 158 N.S. PLEDGE Found only in Gligncene; Carcharodon anrienulatus. Apparently restricted to Lower aud Middle Miocene: Carcharhinus collatus, Carcharoides totuserratus, Hemipristis serra, Galeocerdo aduncus, Carehario- Jamu cf, heroni, Lanina ef. crassidens, Carcharodon megaladon., Apparently restricted to Pliocene (and later); Carcharodon carcharias. Miocene to Pliocene: Carcharhinus aculeatus. The other species oceur during both Eocene and Miocene, and some range into the Pliocene, They raise the presently insoluble problem of whether such forms indeed represent single species. In comparison with the list of species given by Chapman and Cudmore (1924), some notable absentees are seen for South Australia. Heterodontus, represented in Victoria hy four species, has not been recorded, and there are fewer myliobatids. There are also fewer Carcharhinid sharks represented, and such species as Galeocerdo latidens and Sphyrna prisca have not been revo. nized. These omissions are apparently the result of unfavourable sedimentary environments combined with madcquate collecting in the requisite stratigraphic evels, The time is not yet ripe to establish vones based on sharks’ teeth asser- blages as Glikman has done, although a large collection from any one locality would indicate the appropriate Cainozoic epoch. Although many sharks are pelagic, it is doubtful whether their teeth could be accurately used for wider than regional correlation. ACKNOWLEDGMENTS Grateful thanks are given to Dr, N. H, Tadbrook, of the Palaeontology Section, Geological Survey of South Australia, who kindly allowed the author to stndy the Survey's extensive collection of teeth, Various other people made collections ayailable for study or helped in other ways; Dr, D. W. P. Corbett, Mr. C. J. Glover and Mr, H, Mincham of the South Australian Museum of Natural History, Mr. T. Darragh of The National Museum of Victoria, Mr. J, M, Lindsay of the Geological Survey of South Australia, Miss S. Summer, Messrs, R, J. F, Jenkins, W. J. Stuart, J. Brooke, M, R. Walter and R, Wright. Thanks are also given to Prof. M. F. Glaessner and Dr. M. Wade, who kindly read and criticized the manuscript. REFERENCES Acasstz, L. J. R., 1843, Récherches sir les Poissons Fosyiles, Vol. WI. Neuchatel, Amecuino, £., 1906. Les Formations Sedimentaires du Crétacé Supéricur et du Tertiare de Putegonie, An. Mits, nae, Buenos Airey. ser. 3, vol. 8. BratwyauLe, H. D. ve, 1818. Poissons Fossiles, Nouv, Diet. W'Hist. Nat. 27, Cuavean, F.. 1913. Note onthe Oecurrence of the Cainozaic Shark, Careharvitles, in Vietoria- Viet. Naturalist, 30, pp, 142-143. CitAuman, F., 1917. Some 'fYertiury Fish Teeth. Proc, Koy, Soc. Vict. 29 (ms.), pp. 134-141. Cuararan, F., 1918. Cretaceous and Tertiary Fish Remains frm New Zealand. NZ. Geol. Surv., Pal. Bull, 7. Carman. F.. and Cunmonr, F. A. 1924, Sone Cainozoic Fish Remains, with a Revision of the Growp, Proc, Roy. Soc. Vict.. 36 (1.5.), pp, 104-163, Cuarman, F,, and Prircranp, G. B.. 1904, Fossil Fish Remains from the Tertiaries af Aus- tralia, Part I, Proc, Roy. Soc. Vict., 17 (1s8.), pp. 267-297. Crarceswonrn, E., 1837. Notice of the ‘Teeth of Carcharias megalodon Ovcurring in the Kied Crag of Suttolk. Mag. Nat, Hist. (ns), 1, pp. 225-228. ; Tavis, J. W., 1888, On Fossil Fish Remains frow the Tertiary Formations of New Zealand. Sci. Trans, Roy, Dublin Sow... ser, 2, 4, pp, 1-62.. FOSSIL. FLASMOBRANCH TEETH OF S.A. 159 Eastman, C. RK. 1901, Systematic Paleontology, Pisces. Maryland Geol. Surv, Ment. Tiocene, pp. 98-110. Eastman, ©, BR. 1904. Systeriativ Paleontolagy of Minvene Deposits of Muryland, Pisces. Maryland Geol. Surv. Mem., pp. 71-93. Crimsrax, L. S., 1964. Akily Paleogena i ikh strativraficheskoe wunchenie, Akad. Navk SSSH, Hooyun, D. A, 1954. Pleistocene Vertebrates from Celebes. IX Elasmobranchii. K. Ned. Akad. Wetens. Prow. Ser. B, 57, pp. 4735-485, Horna, S. L.., 1939. On Two Small Collections of Fossil Fish Remains from Balasore, Orissa. Rec. Geol. Sury. India. 74, pp. 199-215. Isurwana, ¥., 1921, Fossil Shark Teeth from the Neogene of Japan, Sci. Rep, Tohoku Inip, Univ. Ser, 2 (Geol,), 5, pp, 61-74. Jorpan, D. S., and Buar, GC, W., 1918. Supplementary Notes on Fossil Sharks. Univ, Calif, Publ, Bull. Dept. Geol., 7, pp. 243-256. aaa > M... 1902. Les poissons paléocénes de Ja Belgique. Mém. Mus. roy. Hist, ‘nat. Be Be, . Lenicne. M., 1905, Les. poissons éocénes de Ja Belgique. Thid,. 3. Lenwenge, M., 1910. Les poissons aligocénes de Ja Belgiqne. ITbid,. 5, Lenicns, M,,. 1926. Les poissons neogenes de la Belgique. Ibid. ser, 1. 32. Lericue, M., 1936. Les Poissons du Crétacé et du Nummolitiqae de Ande. Bull. See. Géol, France, ser. 5, 6 pp. 375-402. Lewicne, M,, 1957. Tes Poissons néosénes de Ia Bretagne de l’Anjou ot de la 'Touraine. Mém. Sou. Ceol. France, ser. 5. 81. Lericur, M., and Casirn, K., 1954, Les Fausies ichthyologiqie marine du Néogéene dos Indes orientales. Schweiz, paldont, Abh.. 70. Linnsay, J, M., and Snerreny, R. CG. 1966, Muuno Para Clay Member. Geol. Surv, & Aust, Qoart. Geol, Notes, 19, p. 7. TLinwanis, C., 1758. Systema Naturae. ed, LO. p, 235, McCoy, F,, 1875, Prodromus of the Palaeontology of Victoria, Dec, 2: 8-10. Geol. Surv. Viel. Mrs, K.. 1952. Geology and Underground Water Resources of the Adelaide Plans Are. Geol, Surv, &. Aust. Bull. 27, p. 234. Miitun, J.. and Henue, F, G, J., 1841. Systematische Beschreibung Plagiostemen, Berlin, Noernixe, T.. 1901. Vhe Fauna of the Miocene Bods of Burma. Geol. Surv, India, Mem, 1, pp, 372-378. Scere, T. D,. 1968, The Marine and Freshwater Fishes of South Austraha. Handhook of flora und Fauna of South Australia, Govb. Printing Office, Adelaide, Winrr, EJ. 1926. Eouene Fishes trom Nigeria. Geol. Sury. Nigeria, Bull, 10. Wie, E, L, 1930. The Vertebrate Faunas of the English Eocene. Vol Lo Brit, Mos, (Nat, Thist.), Woonwann, A. S.. 1886, On the Palaeontology of the Selachian Genns Notidanny Cuvier. Geol. Maz,, Dee. 3, 3. pp. 205-217, Woonwann, A. 5. 1889. Catalogue of Fossil Fishes in the British Musewn (Nat. Hist.), Pt, 0, Wouvwanp, A. 5. 1899. Tueth of Sharks and Skates fran English Eocene J’ornations: Proc, Geol, Assoc., 16, pp. 1-14. Zain, Ke von, 192% ‘Texthook of Paleontology, vol. IL revised by A. Sinith Woodward (1932), pp. 50-08. EXPLANATION OF PILATES (All figures of inner fuer except where stated } Puave | Tutra Glass Orthocdonta (all 2 x natural size). Fig, 1. “Notidenus’ serratissinius Ag. Blanche Point. Macl ( F17262), Fig, 2, “N2 ef. serratissimus Ag., Nurecoorte No. 3 Bore, 426 ft. (VG). big. 3. “Notidanuy”? serratissimus Ag, Blanche Point Marl (P17284 and Jenkins collestion). Pic, 4. Myliohatis spp., toothplites, articulating surfaces, Naracoorte No. 3 Bore: (V41). Fig, 5. Myliobatis sp. caudal spinc. Blanche Pot Marl (VAG). Fig, 6, Pristiy c& cudmorei Chap., rostral tecth, Blanche Point Marl (F172835. FITSLS). Fie, 7, Carcharhinus cf, collatus (Eastman), Marion No, 2 Bore, Sturt River Valley (17279, ¥17312-P17314). , Fig, 8. C, aeulwatus (Davis), River Morray Clits, probably Morgan (PT7270).. Fig. O. Hemipristiy serra Ag. Cadell Marl Liews, 4 tiles 8. of Morgan (ELT288 }. Fig, 10. a. Calvoeerdo adunens Ag., Morgim (E17275). hb. Guleocerdo sp., locality mnknown, possibly near Morgan (FI727L), Fiz. 11. Carcharoides of. tutuserrales Ameghing, Strathbyn (117287). 160 Fig, Tig. Fig, Fig. Fig, Fig, Fig, Fig, Fig, Fig. Vig. fig. ] ig. Fiz. N. 8. PLEDGE Puare 2 Carcharias maslinensis sp, noy. (all 2 x natural size) 1-4, E. & W.S. bore No. 5, Naracoorte (426 ft.), Knight formation. I. 2. tw , 4, 5 6, 8, as Holotype, immer face (17260). Nolutype, posterior profile. Paratypes--anteriors (V35). Paratypes (V1S): a. upper laterals: b, Jower Laterals. . Maslin Bay, basal. Tortachilla Liniestone—crown enamels only: 4, upper laterals (F17295-F 17300); b. lower laterals (F17301-F 17307); c. anteriors (F17289-F 17294). 7. Anterior tecth, Blanche Point Mar] (Carter's Unit 2), Maslin Bay (Brooke and Jenkins collections). Anterior tooth, lower Port Wilhmga Beds (Carter's Unit 3), Aldinga Bay (£17267). Puatr 3 (Natural size except where stated ) . 2, 3. Careharias macrotys (Ag,), large specimens, upper and lower anteriors. Blanche Point Marl, Maslin Bay (F17308-P17311, and Jenkins collection), * 6. a pyran (Ax.), Peel's Bore (390 ft.) near Robe; Knight Formation VIG, 4 . C. cl. macrotus (Ag.), Naracoorte Bore No, 5 (426 ft.) (V49). . C. cf, macrotus (Ag.}, Peel's Bore (390 ft.) (V53), . Caroharias sp, indet.. Naracoorte Bore No, 5 (V32). . CG, dubius (Ag.), Peel's Bore, near Robe (V27), . C, cantortidens (Ag.), Peel's Bore (V14). 2, C. cf. euspidatus (Ag,). (?) Blanche Point Marl (£17264). . GC. cf. cuspidutus (Ag.), Blanche Puint Marl, Maslin Bay (Jenkins collection). . C. cf. rutoti. (Winkler), Blanche Point Marl. Maslin Bay (F17263 and F17265). C_ensiculutus (Dayis). Naracoorte Bore No_ 5, x 2 (V36): a. inner face; b. outer face. . Carchariolamna cf. heroni Hora, Marion No, 2 Bore. Sturt River Valley. x 2 (F17280). PuatTe 4 Family Isuridae (all natural size). . Lamna abliqua (Ag.), “Deep yell, Murray Scrub”, probably from Eocene Knight Formation. equivalents (£17269). . L. apiculata (Ag.), Blanche Point Marl, Maslin Bay and Witton Bluff, Pt. Noarlunga F17261. F17266, and Jenkins collection), . L, tineenti (Winkler), Blanche Point Marl, Maslin Bay (F1728G6 aud Jenkins col- lection ). . L. ef. erassidens Ag., Murray Cliffs, probably near Morgan (P172725, . Tsures hastalis (Ag.), Murray Cliffs, probably Cadell Marl Lens near Morgan (F17273), . 1. ef, Aastalis (Ag.), basal Port Willunga Beds. Aldings Bay (F17268), . 1. desorii (Ag.). Strathalbyn, crown enamels only (1° 17282, F17315). IT, retroflexus (Ag.), upper right lateral tooth, Port Willunga Beds, Reddin’s Bore, Gawler River (270 ft.). (Photo by courlesy of J. M, Reddin,) . 1, retroflesus (Ag), anterior tecth, Murray Clills, near Morgan (F17274): a. inner face; b. profile. . Isurus (?) sp., basal Tortachilla Limestone, Maslin Bay (1°17283. 17316, F'17317). . Carcharodim auriculatuy Ag... Port Willunga Beds, Aldinga Buy (117276}. . G. megalodlen Charlesworth, Barbers Quarry, Mt, Gambier (F17277). . €. carcharias (Linn,), Dry Creek Sands, Abattoirs Bore, Pooraka (F17278). . Carcharodon sp. indel,, outer face, Blanche Point Marl, Witton Bluff, Port Noarlunga (Jenkins collection). PLATE | PLEDGE S. N, 9 N, S. PLEDGE PLATE So N. S. PLepce PLATE 4 STRATIGRAPHY OF LATE CAINOZOIC DEPOSITS IN SOUTH AUSTRALIA! BY J. B. FIRMAN” Summary The stratigraphy of late Cainozoic deposits in southern South Australia is set out for sedimentary basins and adjoining uplands in terms of environment and time. Middle Tertiary tectonic movements initiated late Cainozoic sedimentation. Tectonic movements at the end of the Pliocene produced a pattern of lineaments, which set limits to the inland penetration of rising seas, and controlled trends of stream courses and lake and continental margins. The Pliocene-Pleistocene boundary is well defined locally by faunal evidence, and regionally by an erosional break, which separates strongly ferruginized estuarine and fluvio-lacustrine sands from fluvio-lacustrine clays. The sequence of Pleistocene shallow marine and littoral deposits on the continental margin is subdivided into lower, middle and upper beds. Correlation with inland continental deposits is facilitated by through going soil stratigraphic units containing silcrete, ferricrete and calcrete. The shift from a cooler, wetter climate of the Pleistocene to a warmer and drier climate in the Recent is recorded inland by extensive dune sheets. STRATIGRAPHY OF LATE CAINOZOIC DEPOSITS IN SOUTH AUSTRALIA‘ by J. B. Firman? [Read 10 August 1967] SUMMARY The stratigraphy of late Cainuzoic deposits in_ southern South Australia is eet out tor sedimentary basins and adjoining uplands in terms of environment and time. Middle Tertiary tectonic movements initiated Jate Cainozoic sedimentation. Tectonic movements at the end vf the Pliovene produced a pattern of lineaments which set limits to the inland penetration of rising seas, and controlled trends of stréam courses and Jake and contmental margins. The Pliocene-Pleistocene boundary is well-defined locally by faunal evi- dence, and regionally by an erosional break which separatcs strongly ferru- ginized estuarine and fluvio-lacustrine sands from: fluyio-lacustriné clays, The sequence of Pleistocene shallow marine and littoral deposits on the continental margin is subdivided into lower, middle and upper beds. Correla- tion with inland continental deposits is facilitated by throughgoing soil strati- graphic units containing. silerctc, ferrierete and calercte. The shift from a cooler, wetter climate of the Pleistocene to a warmer aml drier climate in the Recent is recorded inland by extensive dune sheets. INTRODUCTION This. paper sets out the stratigraphy of Jate Cainozoic deposits in the southern part of South Australia, The study includes the Murray, St. Vincent and Spencer Gulf, West Coast and Fucla Basins and the adjoining uplands, Wark has been concentrated on the southern margin of the continent where marine, transitional and continental deposits and related land surfaces oceur together, The paper is essentially a study of physical sequences, and is. a necessary preliminary to systematic radiometric dating and the study of changes in climate, flora and. fauna during the late Cainozoic. On the stratigraphic table, rock units are organised according to environ- ment and timc. The reader secking detailed information oun the identification and description of stratigraphic units and their relationships, is referred to the writers’ papers quoted herein. Tabulations in these papers suinmarize much of the evidence used as a basis for correlation in this paper. Environments shown on the table depend upon different models; geomorphic, sedimentary, geological and palaeontological concepts have been employed at various places. A uniform treatment of environments must await a unified system of classifica- tion. Those deposits which show the influence of climatic zonality are empha- sized on the table and in the text. 1 Published hy permission of the Director of Mines, South Australia. 2 Geological Survey of South Australia, Trans. Roy. Soo, §. Aust, (1967), Vol. 91, 166 J. B. FIRMAN LOCALITY MAP— MINOR LOCALITIES Place. names referred to in text and shown on Locality Map by numbers. 1, Avondale Homestead 20. Lake Torrens Craben 39. Molineyirx, Hundred of 2, Barossa Valley 22. Loxton 40. Point Labatt 3. Blanche Point 23. Lipson Reach 41. Port Augusta 4. Blanchetown 24, Lyndhurst 43. Port Wakefield 5. Bookpurmong 25, Maree 44, Renniark 6, Boston Bay 26, Marmon Jabuk 45. Ripon, Hundred af 7. Bungunnia Iontestead 27, Millewa 4G, Sir Joseph Banks Group 8. Chowilla 28. Parilla 47, Spencer Gulf Graben 10, Coomandook 29, Monoman Creek 45, Swan Reach 11, Coorong Lagoon 30, Moorlands 49. Tartanga Island 12. Fulham 31, Naracoorte 50. Telford Railway Station 13. Etadunna 32, Nildottic Sl, ¥Yallunda 14. Dry Creek 33. Nilpena 52. Callana 15, Hall Bay 34, Maslin Bay 53, Coffin Bay 16, Hallett Cove 35, Noarlunga-Willonga Basin 54. Lockleys 17. Hindmarsh 36, Padthaway 55. Marina 18. Karoonila 37. Pinnarso 56, Smart Bluff 19. Kimba 38. Peehinga 57,, Coomonga The study is based on cxamination of rock sequences revealed by drilling, regional stratigraphy and terrain analysis, and by investigations in support of mineral exploration, hydrology and soft-ground engineering, Although the paper has been prepared with the views of other workers in adjoining areas in mind, no attempt has been made to review their work in detail. STRATIGRAPHY AND GEOLOGICAL EVENTS TERITARY Tectonics and Sedimentation The nature and extent of Jate Cainozoic scdiments in South Australia reHects earlier tectonic movements which caused the retreat of the Miocene seas and initiated dominantly: clastic deposition throughout the remainder of the Cainozoic. For. asummary of early Tertiary tectonic events the reader is referred to Webb (1958). Sedimentary basins in Sonth Australia contain about 1,000 feet of Tertiary sediments, overlying Permian and Cretaceous rocks resting on a Precambrian basement. Tertiary biostratigrapby bas been descrihed for the basins generally by Ludbrook (1958, 1959, 1961 and 1963). The basins, as they occur now, were shaped during the Caimozoic. Within each basin, a very strong trend trom Miocene marine through transitional to continental sediments is apparent. Quaternary stratigraphic units are remarkably extensive, and the most important occur in hasins on the seaward margin of the State where transitional sequences link the stratigraphic record of the ocean basins and the continents. Upland Ferricrete and Sedimentary Equivalents Although this paper deals mainly with late Cainozoic stratigraphy, the processes responsible for some of the silcrete and ferricrete so prominently displayed in the late Cainiozoic landscape were in operation much earlier. One of the more important events during the Lower Tertiary was the formation of 1 Massive vesicular. cellular o¢ concreti¢mary ironstone overlying mottled and pallid zanes is termed “laterite”, Detailed stratigraphic stndies reveal that this classic profile contains layers that have been formed in dillerent ways at diferent times, the profile being as oli] as the first weathering of the parent rock, ‘hese considerations Joad to the use of ferricrete as a non-genetic term for the ironstone in this and other profiles. STRATIGRAPHY OF LATE CAINOZOIC DEPOSITS 167 NORTHERN ss ten, QUEENSLAND WESTERN AUSTRALIA oO uy a = 3 a = wy 2 VICTORIA Miles ¢ 300 Miles ; + oe: Kilometres O 400 Kilometres £ - = | 67-387 J.B. Firman S.A. Department of Mines 165 ]. B. FIRMAN ironstone in “laterite” profiles on old land surfaces in the npland areas moarginal to the basins. Lower Tertiary ferricrete is recorded by Harris and Oliver (1965) above and below Lower Tertiary sediments in the Barossa Valley, Ferricretes associated with the remnants of a high-level “peneplain” surface in the Mt, Lofty area have been described by Rix and Hutton (1953), Thomson anc Horwitz (1961), and by others quoted in Webb (loc, cit.) and Wells (1966) The Yallunda Ferricrete of the Lincoln Uplands (Johns, 1961; Firman, 1967) occurs in a similar geomorphic situation to the “peneplain” ferricate of the Mt, Lofty-Olary avd Flinders Ranges, The sedimentary sticcession in the Murray Basin has been examined for equivalents of the ferrierete in upland areas: oolitie siderite and “laterite” have been recorded in the Lower Pliocene Bookpurnong Beds; ferruginous beds aril cappings are found in the Upper Pliocene Loxton Sands; “ironstone” is recorded within the Upper Pliocene Parilla Sand and ortstein has been mapped at its surface; upper beds of Parilla Sand in the South Australia-Victoria border region are ferruginous and contrast with the non-ferruginous lower beds cropping out in River Ruiray cliffs. Ferruginisation in Lower Pliocene basin sediments ma correlate with the massive ironstone described by Gill (1958) in the Black Rock Member of the Lower Pliocene Sandringham Sands in Victoria. | Ferricrete (including ortstein) in Upper Pliocene basin sediments is correlated with ortstein on the eastern slopes of the Mt, Lofty Ranges, Geomorphic evidence shows that these deposits rest upon slopes incised through the “peneplain” ferri- cretes, and are therefore younger, and sedimentary evidence—notably similar lithology and close proximity to the Parilla Sand of the basin margin—suggests a correlation between ortstein-hearing beds on the eastern slopes of the ranges and the Parilla Sand of the type area in the Murray Basin. Silerete has been described mainly from inland areas of South Australia by Wopfner (1960, 1964), Wopfner and Twidale (1967), Stephens (1964), Stewart (1958), and others, but it bas also been recorded on Yorke Peninsula NOTES ACCOMPANYING TLE STRATICRAPHIC TABLE Sedimentary Basing A, Murray Basin, B, Mt. Lofty-Olary und Flinders Ranges. C. St, Vineent and Spencer Basins. D. West Coast Basins, E, Eucla Basin, Notes 1. For Pleistocene-Recent deposits, “Neritie’ is applied to shallow-water marine enviren- ments identified from sedimentary and veomorphic evidence, und ranviny from high- energy nearbeach situations fronting: the open ocean to low-energy near-heach situa- tions within the Gulfs. For Plincene deposits, “Neritie’” includes epinuritic shallow water Dry Creck Sands (Glaessner, M. E. and Wade, Mary. 1958) and marine shallow water Bookpuroong Beds (Undbrook, N. IL, 1958 and 1961), 8, Soils are superimposed on the units with which they are associated stratigraphically, Symbols in vertical bars show the time ranye, horizontal burs show the unit relationship. 3. Coorong Jayoon area, 4, Cottin Bay urea. 5. Deposits in the River Murray. Port Wakufield, ; 7. Upper member contains several layers of moderately hard caleretes 4%. Sediments of the St. Vineent Graben, 9 Limestone at the top of Hallett Cove Sandstone at Marino, Hillett Gove, Blanche Toint, 10. Sand below limestone (0) at Hallett Cove und Muslin Bay. 1], Sedimentary equivalents of the upland ferricvetes on basin margios are shown Lelow this line. 12, Deposits in the Barossa Vidley (Exainple from Rowlands Flat jn Tarris, Wook, andl Olliver, J, G,, 1965}. ; 3, Callana und Smart Blull examples in Morbes, B,C, (1966), 1. Lincoln Uplands. STRATIGRAPHY OF LATE CAINOZOIC DEPOSITS 169 LATE CAINOZOIC DEPOSITS IN SOUTH AUSTRALIA —-— STRATIGRAPHIC TABLE ENV | RONMENT #1 | LITTORAL, LAGOONAL, COASTAL LACUSTRINE, a NERITIC ESTUARINE AEOLIAN PALUDAL FLUVIAL - COLLUVIAL INLAND AEOLIAN SOILS 7 = 1 EM Leachee candy i ’ c, e zens Shelly, Felsarvcite Carbonate sediments SEMAPHORE SAND Fea B50 Recent | ceca Regent & horitgns tits Shell) sand D4 Salt crusts alluyium inal yands —_—_|— MOLINEAUX SAND * Soft ropy and ~ FULHAM SAND® columnar umisatet BUNYIP. SAND * oP pyle RECENT 4 , Voleante ash at Mount Gambier Gays sod alles in initermontane basins Talus and slope Geposits on surtaces pest Loveday. Soil ST. KILDA FORMATION °° LIPSON FORMATION YAMBA FORMATION * GOONAMBIDBAL FM." - YAMBA FORMATION *° o=6-0-6-0-0-0 4 ~c—o—a—o-o—o-| ff LOVEDAY SOIL* Sheets of POORAKA CLAY < WOORINEN FORMATION’ B Horizons of coh aealisn shell sand . OS™ nodular and platy ae TELFORD GRAVEL x th Pe Gypsiferous Soil " 4 - ‘ TERRA-KOSSA AND . RENDZINA Sand and MONOMAN Red-beown sand sheet sandy clay: AD LE In-solution pipes” FORMATION: over calerete AP BLACK EARTHS: RED. BROWN EARTHS | 1 Calerece in Py wy BAKARA SOIL (os = = = bs i ON” ; i me ED aaa BRIDGEWATER TELFURD GRAVEL" | Galereted shell sand Upper Member Lower" Member” RIPON GALGRETE. EISTOCENE = Terrestyial MIDDLE > | BUNGUNNIA.L3,* ats fp] Mancaniferous. seaming : FORMATION *> fa N ard ) Lower Mumiber y i Gypsum and Aw) COOMUNGA FM. IMM gypaerovy aly Caquina and calearenite ar base of Bridgewaser Fornarien? ZBLANCHETOWN CL ® A |/_——___|,,, Soil horizons of IF) naduiar carbonare 6, accumulation & ea oLAY Bodonone Soop soe xe COOMANDOOK FORMATION” Sande at Lachieys © CHOWILLA SAND” Limestone? HALLETT “sand atte eine ESAS *) Talus and slope deposirs ‘ flanking ranges as ae ficet® NORWEST BEND FM, * PARILLA SAND” |g aaa eeaaeacaa Limestone at Fehery Bay? ORY CREEK SANUS and bedrock highs eee) Ferncrete : : i: (oreste isi LOXTON SANDS (Upper bed) * LIOCENE | ; wa tt al LOXTON SANDS (ower bed) “ - ? ? LOWER PLIOCENE-MIOCENE ry Fider silere! LIMESTONE OF VARIOUS ENVIRONMENTS: Hyrine GAMBIER LIMESTONE: SSS SRS PATA spd MORGAN-MANNUM \imescone’? (imesvene ar Myponga, and MELTOW LIMESTONE!” PORT WILLUNEA REDS,* and WULLARAON LIMESTONE. © TERTIARY 7 pmrote :° oe) Older ferricreres 2 (YALLUNDA. le! FERRICRETE Shale with leaf of Lenco Uplate) HFFERENTIATED impresnons. in Silty and. MURNPEOWIE fancy ays Formation ®? Ce ewes 2 wy 170 |, B. PIRMAN (Crawford, 1965) and near Kapunda (Dickinson and Coules, 1957) and Ade- laide (Stewart, loc. eit.). Recent investigations on Islands of the Sir Joseph Banks Group by the Geological Survey of South Australia and on Eyre Penin sula from Kimba to Boston Bay (Firman, loc. eit.) have demonstrated its Oecur- ened over a wide area in the southern part of (he State. Fluvial processes have heen more active ou the southern margin of the continent, aud this may account for the erosion of all but a few remnants of silerete in upland areas and foe burial of silerete-bearing sediments on the basin margins. The occurrence of silerete south of the arid zone demonstrates that silerete lormation is not re- stricted to that zone. Other considerations, notably geologic age, the oced for a high water table favouring silica accumulation, and the presence of extensive ferruginous mottling in associated beds, favour a wetter chmate than the present. The approximate stratigraphic position of a 2Lower Tertiary silerete has heen shown at the base of the stratigraphic table following Forbes (1966), The stratigraphic position of the Boston Bay Silerete {s uncertain, bul the unit ds placed in the late Tertiary because of its position in the landseape and its rela- tlonship to younger deposits (Firman, loc, cit.). The silicified cap on the eroded surface of Parilla Sand in the Marray Basin is placed in the Pliocene on the table, but it could equally well be a Pleistocene feature, The land surface at the time of silicification is termed the Karoonda Surface (Firman, 1966a). In South Australia, clastic continental equivalents of late Tertiary marine sediments have a distinctive lithology and configuration. ‘Their most charac- teristic features are ferrugirfisation and the presence of extensive developments of ferricrete, These features are not found in younger deposits of known Quaternary age, QUATERNARY Late Pliocene-Pleistocene Tectonics Faulting ocemred throughout the Tertiary and, by_ the lose of the late Pliocene, a well-defined pattern of lineaments, including faults and major joints, had heen developed, Some lineaments parallel older faults and have throws up ta 200 fect, others are probably major joints. On photo-mozaics of the basin areas the fractures usually occur as lineaments mendiseg roughly northwest and northeast. The fractures occur in several sets and their intersection forms shomboid blocks, At this time, the inland side of the basins was upwarped and the seaward margin downwarped; the Mt. Lofty, Olary and Flinders Ranges were further aplifted and the extent of the sedimentary basins was reduced; limits were set to the inland penetration of rising, seas and the trend of river courses and of lake and continental margins was established. The pattern thus controls the tectonic and geomorphic framework of Quaternary sedimentation, Pleistocene" PLioceNE-PLEISTOCENE BOUNDARY A long period of erosion and sub-aerial weathering at the end of the Ter- Hary and the beginning of the Quaternary intervenes between a time of extensive quartz sand deposition in the Pliocene and a time of extensive clay deposition 1 The South Australian seqnence has been subdivided so that the Middle Pleistocene containy the Bridvewater Formution and associated ealeretes, Sedimentary evidence suggests a short time-span for the Upper Pleistocene, and a long time-span for the Lower Pleistucene when compared to the Middle Pleistocene. STRATIGRAPHY OF LATE CAINOZOIG DEPOSITS I7L in the Pleistocene. The period was interrupted locally by tectonic movements (sec Twidale et al, 1967). The sands at Lockleys in the St. Vineents Basin that underlie the Hindmarsh Clay are placed in the Calabrian Stage at the base of the Pleistocene by Lid- brook (1963), The Plioeene-Pleistocene boundary has therefore been placed between the sands at Lockleys and the underlying Pliocene Dry Creek Sands. This same break in the sequence has been found in the Hallett Cove Sandstone, south of Adelaide, where Pleistucene fossiliferous limestone Gyerlies a lower bed uf sand which, according to Ludbrook (pers. comm.), contains a fauna equivalent to the Pliocene Yatalan fauna in the type section (see Twidale, Daily and Firman, 1967). In other sedimentary basins the Pliocene-Pleistocenc boundary has been supposed to lie directly below the Pleistocene mottled clay sequence (Ilindmarsh Clay and equivalents), but the evidence from Leckley's in St, Vincents Basin suggests that the boundary may lie deeper in the sequence in other basins also (Firman, 19652). On the south side of the Murray Basin on the Padthaway Archipelago, the Coomandook Formation overlies the Parilla Sand whieh is equivalent to the Dry Creek Sands, and is in turn overlain by the wedge of aeolian Bridgewater Formation, The seas in whieh this unit was deposited, penetrated inland as far as the Marmon-Jabuk scarp and the extension of a southeast trend to Naracoorte, No sedimentary evidence exists for greater inland penetration of any later Pleistocene high sea, Althongh Tate (1877) refers to debris at Third Creek near Adelaide ... attributed to the action of glaciers”, South Australia appears to lie outside the area of Pleistocene glaciation, Periglacial conditions may be indicated by late Pleistocene Gilgai soils, which resemble patterned ground and have shrink- and-swell structures that appear to be fossil, and by other structures in gravels near Peebinga and in cemented loess near Moorlands that may be due to trost-hieave. oo ConTINENTAL AQUEOUS SEDIMENTATION Fluvial and lacustrine deposition now occurred over a wide area in the southern Australian lowlands, In tectonic depressions flanking the Mt. Lofty- Olary and Flinders Ranges, such as the St. Vincents Graben, Spencers Gulf and Lake Torrens Graben, sequences up to 400 feet thick of irregularly bedded sandy clay with abundant sand and gravel lenses were laid down, The Hind- marsh Clay in the St, Vincents Graben is typical of the thick sequences and iy comparable to the Shepparton Formation described by Lawrence (1966) in the Riverine Plains of Victoria, In shallow basins, sequences about 50 feet thick of regularly bedded sandy clay with a few thin beds of sand were de- posited, The Blanchetown Clay in the Murray Basin, the Avondale Clay in hasins adjoining the northern Flinders Ranges and the Ochre Cove Beds in the Noarlunga-Wilhinga Basin (Ward, 1965) are zepresentative of the thin sequences, Diprotodon (Howchin, 1918) and Afaeropus have been recorded from these deposits. ‘Lhe widespread occurrence of the units suggests the transportation of large amounts of fine clastics during a pluvial climate, Formation of gypsiferous evapnrites and termination of clistic deposition marks a drier climate phase at the end of this time. Thin dolomitic micrites (Bungunnia Limestone of the Murray Basin, Nil- pena Limestone cast vf the Torrens Graben and un the west flank of the 172 J. B. FIRMAN Flinders Ranges) were Jaid down. The Bungtnnia Limestone was mentioned in Tate, Prof, Ralph (1855), who described “Travertine cover of thin-hedded sandy limestone (over) red and blue clay” (Blanchetown Clay) as long ago as 1885, A “Cypridiferous limestone” was collected from the Victorian Mallee im 1912 and matched with other limestones, notably a Diprotodan-hearing lime- stone from near Geelong, by Chapman (Chapman, Frederick, 1936). In the Murray Basin, the extent of the Bungunnia Limestone marker defines the lake system, called Lake Bungunnia in Firman (1965b), which contains the Blanche- town Clay. The position of the limestone high in the Murray River cliffs 50 to 100 feet above a river graded ty a modern higher sea-level, suggests a terrain with less local relief and a lower seaward gradient than at present, The position of the Bungunnia Limestone in the rock stratigraphic sequence outlined on the table is reliahly established. Other similar limestones have not vet been placed in sequence, Reconnaissance traverses suggest that a sequential arrangement of Pleistocene freshwater limestones exists; this would begin inland with the Bungunnia Limestone and extend south through younger Teshwater limestones of the south-east (Blackburn, Bond, and Clarke, 1967) to modern carbonate sediments described by Mawson (1929) and by Alderman (1964) in the Coorong lagoon adjoining the Southern Ocean. The Bimgunnia Lime- stone-Blanchetown Clay sequence is correlated with a limestone®-clay sequence (Forbes, loc. eit.) in the Maree-Lake Callabonna area which includes the Avoridale Clay of Firman (1967). The similarity of deposits suggests that the climate during deposition of Bungunnia Limestone hundreds of miles inland of the modern coast may haye been similar, in some respects, to that prevailing today in the lower south-cast adjoining the Southern Ocean- Aronian Drrosivion mv amr Coastan Zonr After the withdrawal of the sea in which the Coomandook Formation was laid down, the lower member of the Bridgewater Formation was deposited along the southern coastal margin, Bridgewater Formation was defined mm Victona and contains “Calcareous duncs and dune limestones” ({ Boutakoff, cigr On the sonthern edge of the Eucla Basin the lower aeolian member abuts directly against a fossil chff cut in Nullabor Limestone; on the west coast the lower member lies on a steeply rismg shoreline; in the south-east, the few known occurrences of the lower member are scattered across the Padthaway South Australian Museum. North Terrace, Adelaide, South Australia. Trans, Roy. Soc. $, Aust, (1967), Vol. 9b 188 MICHAEL J, TYLIER Field Notes: The males were located at each locality calling from vegeta- tion at heights of from three to eight feet fiom the level of the ground. ‘The call consisted of a single note, resembled a long squeak and had a duration of about four seconds, Sphenophryne mehelyi Parker When Parker (1934) described this species the only specimens availahle to him were two on Joan from the museum at Budapest which he designated type und paratype respectively, The specimens hal been reported previously (as Chaperina fusca) by Méhely (1901), ‘Che herpetological collection at Budapest was destroyed in 1956 and there are na known lopotypie specimens in existence, The specimens referred here to mehelyi agree in alinost all xespects with Parker's description and there is pertect agreement with the reported colouration of the type which, to judge from comparisons with species in the South Aus- tralian, Museum collection and descriptions in the literature, is quite unlike that of any other species of Sphenophryne. It is pertinent to note that the type locality (Sattelberg) is located on the Huon Peninsula, the portion of New Guinea in closest proximity Lo New Britain. Material Examined: 857 speeimens—Seuth Australian Muse 28.7785-7839, 7955-500U, 8186 (232 specimens), 6438 (15 specimens), Keravat Corrective Institulion, Keravat, Gazelle Peninsula, Variation and Comparison with Original Deseription; Of the above. speei- ments a series of 55 (.7785-7839) have heen examined in detail amd measured. Nineteen proved to be adult males, 27 adult females (of which 21 were gravid) and @ sexually immature specimens, The ranges and means of snout to vent lengths of the adnits were as follows: males 15°5-17-2 mm (mean 16:3 mm), femules 17-0-20-1 min (new 18-7 nn), Parker's figure of 20 mim as the size of the holotype was not the snout to vent length. Me states in an adeaunt of his descriptive methods Mist, p. 16), “The length of a frog, is the maximum distance in a straight line between the tip of the snout and the hinder side of the thighs behind the vent.” This “total length” would exceed the snout to vent length by about one millimetre, placing the size of the holotype very close to the mean snout to vent Jength of the females vx- amined, As noted by Zweifel (1962), Méhely (1901) cites a figure of 24-3 mm for the same specimen and, since it is highly probable that his measurement is the “total length” m vogue at that time, this would indicate a snout to vent lenwth of ahout 23 mm, ‘Vhe specimens tally with Parker's original deseription in the following respects; the presence of a denticulated prepharyngeal ridge with an adjacent median swelling; the distance between the eyé and the tip of the snout is approxt- nately equal to the horizontal diameter of the eye; the loreal region iy oblique; the nostrils are nearer the tip of the snout than the eye; the interorbital space is equivalent to twice the diameter of the upper cyelid; the fingers are short and equipped with small dises; the toe dises are larger than the finger discs: wher the hindlimb is adpressed along the side of the body the kuee reaches the eye. The colouration of the present specimens agrees in all respects with that of the holotype. The canthus rostralis of mehelyi is said to be rounded but most of the Keravat specimens exhibit only the slightest curvature and in some it is definitely straight. A further point of divergence concerns the length of the fourth finger. MICROHYLID FROGS OF NEW RRITAIN 18S Parker reports that in the holotype it is equivalent to iliree-quarters of the inter narial span, whereas its length exceeds the internarial span in the present material, ‘The tympanum is said to be distinct and, “rather less than half the diameter of the ‘eve’. This is truc of the majorily of specimens, but in those in which the skin on the side of the head is at all loose, the site of the tympanum cannot be detected externally, The skin appears stnouth when superficially examined, but minute tubercles ure visible when it is examined beneath the inicroscope. In Parker's key to Sphenophryne the tibia of mehelyi is said to equal ahout half the distance from snout to vent, and to reach the couplet containing mehelyi involves agreement with, “tibio-tarsal articulation reaching the cye’. As indi- cated in a previous paragraph, the knee of the adpressed hind limb of the picsestt material will reach the eye, but the tibia length is consistently less than 1alf of the snout to vent length, A frog whose tibia Jength is exactly half the snout to vent length has a TL/S-V ratio of 0:500, whereas the range of the jwesent material is +354--432, The significance of these points of divergence from the original description caunot be determined until adequate topetypic material is available. At the present time it is considered that the interests of nomenclatural stability are best served by regarding the New Guinea and New Britain populations. von- specific. Comparison with Other Species: Of the thirteen species currently recognised five have snout to vent lengths exceeding 34 mm (curnuta, macrorhyncha, palmipes, rhodadactyla and schlaginhaufeni), The maximum recorded size uf the present material (20-1 mm) therefore prevents any confusion with these SPECIES, Sphenophryne hrevicrus, brevipes, and crassa lack dilated discs on the fingers (present in mehelyi) whilst polysticta has a bluntcr snout, shorter hind- limbs and a different colouration, Of the three Australian species reported hy Zweifel (1962, 1965), 5. fryi has shorter limbs (TIL/S-V 0:31-0:38, mean 0-346, as opposed to 0-35-0-43, mean 0-390) and attains a larger size than mehelyi (of more than 25 specimens reported by Zweifel (1962), only one adult had a snout to vent length of Jess than 20 mm, whereas 20-1 mm is the maximum tor mehelyt). Sphenophryne plunlalis is evidently also a larger species than mehelyli and the specimens re- ported by Zweifel were uniform in hearing, “ill-defined, and irregitlar dark brown spots”, on the dorsal surface of the body. There is no trace of any dorsal mark- ings on any of the 337 mehelyi. The size and TL/S-V ranges of robusta are similar to those of mehely: bunt the reported colouration of the former (although variable) is evidently quite distinct from that of mefhelyi. : Ficlil Notes: Vhe specimens were found congregating bencath piles of decamposing vegetation, consisting of rotting grass and reeds laid upon ground covered with leaf mould, The piles covered arcas of up to five square yards, As many as 250-300 mehelyi were estimated to occur together but, when the yege- tation was removed the frogs rapidly dispersed in all directions and rarely were more than 50 or 60 captured ut any one site. Numerous females were found at the base of the piles sitting upon clumps of eggs at the fout of small depressions excavated to a depth of about two inches in the leaf mould. Each egg clump consisted of five or six eggs connected to one 190 MICHAEL J. TYLER another (in the form of a bead necklace) by a very fine cord. Subsequent measurements of the eggs revealed a mean diameter of 3-3 mm for those at a stage prior to gastrulation, and 4-8 mm for those containing well developed juvenile froglets apparently near to emergence. Six eggs is virtually the maximum number that a female of this diminutive species could cover with her body. Three months prior to the collection of the frogs the area in which they were found was virgin rain forest, but subsequent deforestation had completely eliminated the natural environment of the species. It therefore remains un- certain whether the aggregation of large numbers of frogs in breeding condition is a natural phenomenon, or simply necessitated at Keravat by the reduction in the amount of rotting vegetation occurring in the area. REFERENCES Menery, L. (1901), Beitrige zur kenntniss der Engystomatiden von Neu-Guinea, Termesz. Fiz. 24, pp. 169-271. Parker, 11, W. (1934). A monograph of the frogs of the family Microhylidae. London, British Museum (Natural History), viii + 208 pp. Tyten, M, J. (1965), Transfer of the New Britain frog Hyla brachypus (Werner) to the microhylid genus Oreophryne. Mitt. Zool. Mus. Berlin, 40 (1), pp. 3-8. Zweiret, R. G, (1962), A Systematic Review of the Microhylid Frogs of Australia. Amer. Mus. Novit., No. 2113, pp. 1-40, Zweiret, R. G. (1965), Revisionary Notes on Australian Microhylid Frogs of the Genus Sphenophryne. Ibid., No. 2214, pp. 1-9. A NEW SPECIES OF FROG OF THE HYLID GENUS NYCTIMYSTES FROM THE HIGHLANDS OF NEW GUINEA BY MICHAEL J. TYLER* Summary A new species of Nyctimystes is described. The maximum recorded snout to vent length of males is more than 80 mm, and the species shows distinct affinities with N. humeralis (Boulenger). However, it lacks the characteristic humeral spine exhibited by males of that species, and is also distinguished by its undivided nuptial pad and by differences in colouration. A NEW SPECIES OF FROG OF THE HYLID GENUS NYCTIMYSTES FROM THE HIGHLANDS OF NEW GUINEA by Micmac. J. Tycer* [Read 12 October 1967] SUMMARY A new species of Nyctimystes is described. The maximum recorded snout ta vent length of males is more than 80 mm, and the species shows distinct affinities with N. humeralis (Boulenger). However, it lacks the characteristic bumeral spine exhibited by males of that species, and is also distinguished by its undivided nuptial pad and by differences in colouration. INTRODUCTION In 1963 and 1964 Mr. Barry Craig donated to the South Australian Museum several large collections of frogs obtained in the vicinity of Telefomen in the Western Highlands of New Guinea. Amongst the specimens were eleven representatives of an undescribed species of the hylid genus Nyctimystes. In 1965. Mr. Craig was a member of the Australian Star Mountains Expedition and obtained 650 frogs in an area approximately 40 miles west of Telefomen. Included in the collection were four additional specimens of the same specics. The methods of measurement employed in the description of the new species conform to those used by Tyler (1963). Nyctimystes zweifeli new species Holotype: South Anstralian Museum R.5426. An adult male collected at Telefomen, Western Highlands, New Guinea, on November 24th, 1963, by B. Craig. Definition: A particularly large specics of Nyctimystes in which males have a snout to vent length of up to 83 mm, The hindlegs are very long with a tibia Jength to snout to vent length ratio (TL/S-V) of -588--654. The veins of the palpebral venation are largely vertical in their vrientation with a tendency to form a reticulum. Description of Holotype: The vomerine teeth are in two roughly oval and promincnily raised series between the oval choanae. The tongue is less than half as wide as the mouth, oval in shape and with a very slight posterior inden- tation. The head is broader than long (head length 26-8 mm, head width 29-4 min), The snout is. blintly rounded when viewed from above and strongly rounded in profile. The canthus rostralis is slightly curved and the loreal region oblique. The nostrils are inconspicuous and separated from one another by a distance which is slightly less than the distance between the naris and the eye (eve to naris distance 7-2 mm, internarial span 7-0 mm). The eye is relatively small and not prominent, its diameter (7-8 mm) only slightly greater than * South Australian Moseum, North Terrace. Adelaide, South Australia. Trans. Roy, Soc. S. Aust, (1967), Vol. 91. 192 MICHAEL J. TYLER the eye to naris distance, The pupil is completely dilated. The jalpebral venation forms a dense pattern of predominantly vertical lines. The tympanum is small (diameter 3:7 mm) with only the inferior portion of its annulus visible. The distance from the snout to the vent is 75-9 inm. The fingers are long with large terminal discs, broad lateral flanges and llattened subarticular tubercles (Fig. la). The webbiny between the third and fourth fingers reaches the base of the subarticular tubercle heneath the peuulti- mate phalanx on the third, and midway up this tubercle on the fourth, Fingers in decreasing order of length, 3, 4, 2,1. The dises of the second, third and fourth fingers are larger than the tympanum. The toes are almost fully webbed, the webbing reaching the dises of all toes except the fourth where it extends as far as Ue subarticular tubercle at the base of the penultimate phalanx, and continues to the disc as a broad lateral flange (Fig. 1b). The toe discs are only slightly smaller than the finger discs, and ull are larger than the diameter of the tympanum, There is a small kidney- shaped inner but no outer metatarsal tubercle. The legs are extremely long; the tibia length is 47-4 mm and the tibia length to snout to veut length ratio (TL/S-V) is 0-625. When the hindley is adpressed the heel reaches well beyond the tip of the snout; when the hindlegs are laid alony the side of the body the knee and elbow overlap considerably; when the hindlegs are bent at right angles fo the axis of the body the heels overlap slivhtly. The skin covering the dorsal surfaces of the head, body and limbs is smoath. The lateral surfaces of the body are slightly granular, the throat and chest finely granular and the abdomen and undersurface of the thighs coarsely granular. There is an oxtremely prominent supratympanic fold, a sharp fold on the posterior surface of the tocearm, and a less conspicuous tarsal faldl, The dorsal surfaces of the head, body, forearms and tibia are a very dark Plum colour with. a hint of a crimson suffusion on the dorso-lateral portion of the body. On the thighs the plum occupies the median portian whilst the anterior and posterior surfaces are white, Jeavily suffused with large, regular spots of plum. The sides of the body, the tarsus and foot are similarly marked. The abdomen is ercam and the vemainder of the ventral surfaces grey witli obscure brownish mottling, There is a sub-gulay vocal sac with paired openings in the Aoor of the muuth near the angles of (he jaws, and an extremely large, black nuptial pad at the hase of the first finger. Variation: There are eight paratypes: South Australian Museum R.SSL2 881) collected at the type locality during the period November 14th-18th, 1983, by B, Craig. All specimens are males (six adults and two juveniles), Six addi tional male specimens from other localities are clearly representatives of N, sweifeli; South Australian Museum R.5275, $811, Okfekarnan, approximately 10 miles west-north-west of Telefomen; R.6476, 6499, Kawolabib (141°O5’ lonw,, 5°08" lat.), approximately 10 miles south-east of Capella, Star Mountains; 8.6477, Begibajen near Kawolabib and R.6478, Tabek Creek, Kawolabib. The complete range of allitude of these localilies is 4,700-5,500 ft. The range of snout to vent length of the arlult males is 60-8-83-4 mm, but the smallest specimen with a ouplial pad (8.8815) measures 73-6 mm. There is very little variation in the proportions of the coniplete series. In one specimen the internarial span is equal to the eye to nariy distance, and iss the remainder slightly less than it, the eye to naris to internarial span ratio HYLID GENUS NYCTIMYSTES FROM THE HIGHLANDS OF NEW CUINEA 193 14 MICHAEL J. ‘TYLER {E-N/IN) being 1:000-1-12] with a mean of 1-063. This is an exceptionally long-legged species, for the 'T1./S-V range is °588-'654, with a mean of -625. The supratympanic fold is conspicueus in all specimens, whilst the fold of skin on the posterior surface of the forearm of two shows a tendency to form tubereles. In the remainder it is a continuaus fold as found in the holotype. Finger webbing and the palpebral venation are features in which there is scant variation. (Four examples of the palpebral venation are depicted in Fig. Ic, ) The dorsal colouration varies from a very dark plum to pale violet. In the darkest specimens patehes of dark slate are visible. Ina few specimens there are a few tiny, brilliant white rings scattered on the dorsal surface, Markings on the thighs appear to increase in intensity with age. In the largest specimens the pale background of the anterior and posterior surfaces is almost completely obseured by irregular patches of plum and slate. There appears to be a similar correlation with markings on the ventral surface of the body, for in juveniles and small males it is an immaculate pale cream. In three specimens there are no islands of pigmentation on the lateral bedy surfaces vr on the anterior and posterior surfaces of the thighs. The size and number of the spines which make up the nuptial pad vary with the size of the individual, In the largest males the spines are larger and less numerous than in the smaller specimens. Comparison with Other Species; The large size of sweifeli clearly distin- quishes it from all species except granti, humeralis, kubori, montana, naritiose, papua, perlmeiri and pulehra. (The species referred to here as N. montana is montana (Peters and Doria) and not the species described by Parker, 1936, for which the substitute name cheesmant wis proposed by Tyler, 1965.) The only specimen of granti reported in the literature is the holotype female with a length of 100 mm. A female specimen from Antares (Rijksmuseum Van Natunrlijke Historic, Leiden, No, 12110) approximately 250 milus cast of the type locality has been examined by the writer, It is a female with a snont tu vent Jength of 70 mm; it agrees with the original description of Boulenger (1914) and perfectly reproduces the striking vermiculated pattern of the dorsul surface seen in the illustration accompanying it. From a comparison of this specimen with