JOURNAL OF THE ROYAL SOCIETY OF WESTERN AUSTRALIA VOLUME 56 PART 3 DECEMBER, 1973 PRICE: TWO DOLLARS REGISTERED FOR POSTING AS A PERIODICAL-CATEGORY A THE ROYAL SOCIETY OF WESTERN AUSTRALIA PATRON Her Majesty the Queen COUNCIL 1973-1974 President Vice-Presidents Past President Joint Hon. Secretaries Hon. Treasurer Hon. Librarian Hon. Editor A. F. Trendall, B.Sc., Ph.D., A.R.C.S., F.G.S. G. A. Bottomley, B.Sc., Ph.D. B. E. Balme, D.Sc. R. M. Berndt, M.A., Dip.Anth., Ph.D., F.R.A.I., F.F.A.A.A. . , M. Perry, B.Sc. (Agric.) (Hons.) G. Perry, B.Sc. (Hons.) . S. J. Curry, M.A. .... A. Neumann, B.A. A. J. McComb, M.Sc., Ph.D. S. J. Hallam, M.A. L. J. Peet, B.Sc., F.G.S. P. G. Quilty, B.Sc. (Hons.), Ph.D, J. A. Springett, B.Sc., Ph.D. G. M. Storr, B.Sc., Ph.D. J. C. Taylor, B.Sc., Ph.D., A.R.C.S. P. G. Wilson. M.Sc. P. R. Wycherley, O.B.E., B.Sc., Ph.D., F.L.S. 7. — Analyses of Western Australian iron meteorites by J. R. de Laeter ^ Manuscript received 18 July 1972; accepted 22 August 1972. Abstract All available analytical data on iron meteorites found in Western Australia are presented. The contents of nickel, cobalt, gallium and germanium in 33 of these meteorites was determined using X-ray fluorescence spectrometry, thus enabling these meteorites to be classified structurally and chemically. The new analytical data enabled a number of paired falls to be distinguished. Introduction The first recorded discovery of meteorites in Western Australia occurred in 1884 in the sub- district of Youndegin, some 110 km east of the town of York. Four pieces were recovered, and a description and chemical analysis of the largest of these specimens was given by Fletcher (1887). The analysis revealed a new type of graphitic carbon which was named cliftonite by Fletcher, who realised that its presence could be a sig- nificant clue to the origin and formation of meteorites. In fact meteorites contain far more informa- tion about the early solar system than was once believed (Anders 1971), and analytical data which have accumulated in the past two decades have played no small part in our present under- standing of the formation, evolution and chro- nology of the solar system. The monumental work of Suess and Urey (1956) on the abun- dances of the elements, depended largely on met- eorite abundance data. This work enabled theo- ries of element formation to be formulated, and the work of Burbidge et al. (1957) for example, depended to a large extent on the abundance table of Suess and Urey. Meteorite abundance data has continually been refined over the past decade and the recent abundance table of Cameron (1968) draws heavily on this informa- tion. It is probably true to say that our present understanding of nucleosynthetic processes in stars would not have been attained if accurate analyses of meteoritic material had not been available. It is significant that all the early meteorite discoveries in Australia were siderites (irons). The first Catalogue of Western Australian meteorites (McCall and Delaeter 1965) lists 29 ii'ons, 4 stony irons and 15 stones. There were almost twice as many irons as stones, despite the fact that on a world-wide basis the situation is almost the reverse (Mason 1962). This is un- doubtedly due to the fact that irons are much more easily recognised than are stony meteor- ites, and in Australia with its deserts and large 1 Department of Physics, Western Australian Institute of Technology, South Bentley, Western Australia 6102, and Honorary Research Associate of the Western Australian Museum. areas of arid land, irons are preserved for a much greater length of time than stony meteor- ites. Again it is probably significant that the Australian aborigine was not cognisant of the use of metals, and therefore had little interest in iron meteorites which may have been discovered. Since the publication of the Catalogue in 1965 a number of new meteorites have been reported, and many of these have now been incorporated in the Collection. This has necessitated the pub- lication of a Supplement to the Catalogue (McCall 1968) and a second Supplement is now in preparation. Many meteorites found in Western Australia have been extensively studied both in Australia and overseas. This interest is partly due to the pioneering work carried out by the late Govern- ment Mineralogist, Dr. E. S. Simpson who pub- lished a number of papers describing local meteorites over a period of nearly 40 years. Simpson always took great care to analyse the iron meteorites for the major elements — iron, cobalt and nickel, and for the minor elements copper, phosphorus, sulphur and carbon. Many of his results are extremely accurate and have been summarised in this work. Analytical method In this paper an attempt has been made to tabulate all the analyses that have been carried out on Western Australian iron meteorites. It is true that some of the earlier data can no longer be regarded as reliable, but it was felt that a complete record should be made at this time. On the other hand, where a number of analyses have been made by various authors, a recom- mended value has been given. The evaluation of analytical data for iron meteorites is by no means a simple task, and, as Moore et al. (1969) has pointed out, is sometimes more difficult than for stony meteorites, despite the simpler composition of siderites. In gener- al analysts avoid using samples which in- clude large troilite or schreibersite inclusions. Thus most analyses are not representative of the meteorites as a whole, but only of the metallic phase. For example the sulphur values usually refer to the metallic phase, but this is by no means obvious from many of the original publications. Certainly the sulphur content of the meteorite itself will be significantly different if troilite occurs to any extent. Again iron meteorites are inhomogeneous, and unless ade- quate sized samples are analysed, there can be no guarantee that the results are typical for the meteorite as a whole. Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 65 A more serious problem is that of evaluating accuracy and precision, particularly when differ- ent analysts and analytical techniques are in- volved. Work on the silicate rock standards G-1 and W-1 has shown that there is an alarmingly wide spread in the results obtained by different analysts, or even by the same analyst at dif- ferent times. (Pairbairn et al. 1951). Lest it be imagined that with the advent of modern methods of analysis such disparities no longer exist, it is instructive to examine the paper of Fleischer (1969) where new analytical data on these standard rocks is summarised. There is no reason to believe that the situation is any better for iron meteorites than for silicate rocks. In fact the present study reveals large scale discrepancies between analyses carried out on the same meteorite. Cobalt, for example, has invariably been overestimated in the old analyses. The elements in iron meteorites may be di- vided into four groups. Firstly, the major con- stituents of the metallic phases — iron, nickel and cobalt. These three elements usually account for over 99% of the total composition of iron meteorites. Secondly the important trace ele- ments gallium, germanium and iridium on which a chemical classification has recently been de- vised. Thirdly, the elements found dissolved in the metallic phases and in non-metallic mineral inclusions — carbon, phosphorus, chromium, sul- phur and nitrogen. Finally the trace elements such as copper, zinc and the noble metals. In point of fact almost every non-gaseous element occurs in iron meteorites, but unfortunately very few analyses have been made on these trace elements in the Western Australian iron meteorites. In addition to summarising the analytical data which could be located in the literature, each meteorite was analysed for nickel, cobalt, gal- lium and germanium as part of the present pro- ject. Nickel and cobalt, are major constituents of any iron meteorite, and traditionally have al- ways been regarded as essential analytical data. In point of fact cobalt is not particularly useful in classifying siderites, since its abundance range is very limted. The nickel content of an iron meteorite has long been recognised as one of the essential criteria used in classifying siderites, and it still serves an important role in any structural or chemical classification. The importance of determining the content of gallium and germanium has been recognised only in recent years. The significance of these elements stems from two main factors. Firstly, gallium concentrations in iron meteorites can vary by a factor of 400, and germanium concen- trations by a factor of 14,000. These ranges may be contrasted with cobalt, where the concen- trations do not vary by more than a factor of 2, and for nickel where it is unusual for the range in concentration to be greater than 4. Secondly, gallium and germanium concentra- tions in iron meteorites are highly correlated to each other and to nickel, and in addition are quantised into a number of distinct groups. After early work by Goldberg et al. (195D and Lover- ing et al. (1957), J. T. Wasson and his associates determined the abundances of gallium, german- ium, nickel and iridium in several hundred meteorites in order to elucidate the classifica- tion of iron meteorites, (Wasson 1967b, Wasson and Kimberlin 1967, Wasson 1969, Wasson 1970, Wasson and Schaudy 1971). On the basis of their analytical data, which were determined mainly by neutron activation analysis, eleven resolved chemical groups have been defined. It is believed that the groups may represent different meteorite parent bodies or perhaps regions within a parent body character- ised by different chemical or thermal environ- ments (Anders 1964). This chemical classifica- tion has been used in the present work. A Siemen’s S.R.S.-l X-ray fluorescence spec- trometer equipped with a molybdenum tube, lithium fluoride crystals and a scintillation de- tector was used for the analyses. A flat surface on each of the meteorites was polished with successive grades of carborundum paper until a smooth, highly polished surface at least 1.25 cm in diameter was prepared. This surface was exposed to the primary X-ray beam, and peak and background readings were taken for each of the four elements. The spectrometer was calibrated for each element by standard alloys and from a number of siderites with well- established composition. Details of the technique will be presented elsewhere (Thomas and DeLaeter 1972). The nickel, cobalt, gallium and germanium values measured in the project, must be quali- fied by error limits of ± 0.02%, rt 0.05%, ± 10 ppm and ± 10 ppm respectively. These errors represent the 95% confidence limits based on counting statistics of the experiment and the calibration of the instrument. They make no allowance for the heterogeneity of the sample. X-ray fluorescence spectrometry possesses the advantage of being non-destructive, but it suf- fers from the serious disadvantage of sampling only a very small volume of the specimen. The infinite thickness of an iron meteorite to the X-radiation used in this experiment was only of the order of a few thousandths of a cm, whilst the area of the incident X-ray beam was approximately 1.2 sq cm. In an effort to overcome this deficiency, each sample was analysed in at least two positions, but this did not alter the fact that the statistical errors stated above are probably less than the possible inhomogeneity errors imposed by the technique itself. The latter errors are certainly worse for the coarse octahedrites than the fine- grained siderites. Results Table 1 lists the classification, specific gravity, cooling rate and contents of the elements nickel, cobalt, gallium, germanium and iridium for the 37 iron meteorites which have been found in Western Australia. Multiple entries have been made for Mount Edith I and II, Premier Downs I and II and the Youndegin meteorites. There Journal of the Royal Society of Western Australia, Vol. 56, Part 3. December. 1973. 66 TABLE 1 ( 'hisaificofion and )niijor element (innl.ifffeft of iron meleoritex *('lasMtieation S]»ec'ifir (.'ooUna .Ni Co (ie Ir Namt* gra^■itv rate o- 0 0 Pl’io l)l>m ppm + Kclerences Stnic- (hem- 1K' ]{)8vr. ture istry Avoca (Western Australia) ()m lilA ( • 8 8-8(1 0 ■ ;52 48 Tills work 21-9 1 8 • (!.') 0 • ,52 •> 8 • 7(1 0-52 Hecommeiuled value lialfbiir Down^ . .. ()m-();) 0-49 14 (rosnells . . Anom. 1-An3 7 • 6 (1-04 0 ■ 45 49 252 This work Uundarin^ .. ... Oin lllA 7 • 7 8-18 0-50 10 50 4’hi.s work (1 8-18 0-5 13 8-32 0-(ll 19 38 4 8-22 0 - 53 1(1 43 Recommended value Ilais Om-OsT lllA 7-8 (1 • 93 0-49 32 Tills work 7-3(1 11 1(1-2 1 7-0 Recommended value Kunierina Opl IK' 7-8 9 -(11 0-54 37 95 This work ~1(MI 3 9 - r>o 13 9 - (19 3(1 - 8 93-4 8-1 8 39 - 3 1 9 • (14 37 94 Reconitnenecific Cooling gravitv rate °C/10«yr. Xi 0/ 0 Co o/ ,0 (4a ppm Ge Ir ])I)in P]>ni t References Struc- ture Chem- istry Wonyulgunna Om lllli 7-6 8-86 0-51 43 This work 2 (5 9-05 0-51 13 34 7 8-26 13 21-6 1 8-80 U-51 20 36 Recommended value Yarri Om lllA 7-8 7 • 77 0-49 13 40 This work 8-06 0 • 45 24 7-91 0-47 Recommended value Youanmi Om 111 A 7 ■ 6 7-78 0 • 50 30 'I'his work 8-08 0-87 13 20-5 1 7-80 0 • 50 Recommended value Youmlegin I .. . Og-Ogg 1 7-85 6-46 0 • 55 31 88 312 12 Youndegin 111 Og-Ogg 1 7-8 6-7 0-45 87 345 This work 7-01 0-93 15 90 • 6 1 6-8 0 • 45 90 Recommended value Youndegin Vll 1 Og-Ogg 1 6-92 0 • 49 88 322 7 Youndegin Og-Ogg 1 •—2 6 6-38 90-8 383 2-0 4 6-94 0 • 46 22 1 lie Structural classification used is given by liuchwald (Wasson, 1070). The Chemical classification adoi)ted is given liy Wasson (10(>7b). Wasson and Kiinberlin (1907). Wasson (1969). Wasson (1970) and Wasson and Schaudy (1971). + Heferences — 1 . l)e Laeter (1972) 2. McCall (1968) ■i. Coldstein (1969) 4. Wasson (1970) 5. Wiik it Mason (1965) 6. (ioldstein it Short (1967) 7. Lovering et al (1957) 8. Wasson (1969) 9. Hev(1966) 10. Ward (1898) 11. Mct'all it l)e Laeter (1965) 12. Similes et al (1967) 13. Siinjison (1938) 14. Frost n065) 15. Simpson (1916) 16. Wa.s.son it Kimberlin (1969) 17. McCall (1965) 18. Wasson it Schaudy (1971) 19. Simpson (1927) 20. Wasson (1967b) 21. Wasson (1971) 22. Moore et al (1969) 23. Sini])son (1907) 24. Cleverly it Thomas (1969) 25. l)e I.,aeter et al (1972) 26. Hodge-Smith it White (192(5) 27. McCall and Wiik (1966) 28. Schaudy (1972) 29. Taylor (1965) 30. Wasson (1967a) 31. Fletcher (1887). are actually eight separate fragments named Youndegin as detailed by McCall and DeLaeter (1965). Analyses have only been made on Youndegin I, III and VIII. It has not been possible to identify the particular fragment on which analysis by Goldstein and Short (1967), Wasson (1970) and Moore et al. (1969) were carried out, and their analyses have therefore been listed under the title “Youndegin”. Two of the 37 meteorites listed are not sider- ites, but are the metallic pieces of stony-iron meteorites. The first of these, Dalgaranga, rep- resents material recovered from a meteorite crater and described by McCall (1965) as a mesosiderite with octahedrite nodules. It was one of these nodules which was analysed in the present work. The second, Mount Egerton, has been described by McCall (1965) as a stony iron or possibly an enstatite achondrite usually rich in iron. Cleverly (1968) has examined further recoveries of this meteorite and classified it as an unbrecciated, metal bearing (21%), enstatite achondrite. A piece of the metal phase of Mount Egerton was examined in the present work. One entry in the Catalogue (McCall and DeLaeter, 1965) is Dowerin, which Simpson (1938) had thought was of meteoritic origin. Only one small 0.35 gm fragment of this “meteo- rite” was available and when this was subjected to X ray analysis by Reed (1972) the nickel to iron ratio was found to be <0.2%. It is therefore not a meteorite and has not been listed in the Tables. Another meteorite described by Simpson (1938) as a fine octahedrite is Lan- dor, but as only small fragments were available in the Collection it was impossible to analyse this meteorite except to confirm that it is a siderite. The structural classification used in this work has been devised by Buchwald (Wasson 1970). This classification has equal logarithmic band width intervals, each octahedrite class corres- ponding to a range of a factor of 2.5 in kamacite band width. The boundaries have been chosen so that most meteorites belonging to a particular chemical group fall within a single structural class and this procedure minimises the differ- ences between these classes and Tschermak- Prior classes. The chemical classification adopted in Table 1 is based on the nickel, gallium and germanium abundances of the meteorites, as described in the series of papers by Wasson and his associates. The specific gravities of 32 of the siderites listed in Table 1 were measured as part of the present study. Wherever possible inclusion — free pieces of the meteorite were used for the determinations, although it was not always pos- Journal of the Royal Society of Western Australia, Vol. 56, Part 3. December. 1973. 69 sible to be absolutely certain of the absence of troilite or schreibersite inclusions. The wea- thered outer surface of the meteorites was avoided in determining the specific gravity ex- cept for Loongana Station (iron), Murchison Downs, Nuleri and Premier Downs I and II where only the complete meteorites were avail- able. The specific gravity of Mount Egerton has not been recorded since silicate inclusions were contained in the fragment studied, though Cleverly (1968) reports a value of 7.66. The cooling rates of 13 of the meteorites have also been listed in Table 1. These data are based on electron microprobe measurements of dif- fusion gradients between gamma and alpha phases of meteoritic nickel iron (Goldstein and Short 1967; Goldstein 1969). The remainder of Table 1 deals with the abun- dance data of nickel, cobalt, gallium, germanium and iridium. Only 9 iridium values are quoted, all of which were determined in J- T. Wasson’s laboratory. Table 2 lists all the available data on those minor elements found dissolved in the metallic phases and in non-metallic mineral inclusions. Also listed are a number of trace elements. Zinc and tin have been determined by the stable isotope dilution method using solid source mass spectrometry by Rosman (1972) and DeLaeter and Jeffery (1967) respectively. These data are probably accurate to ±2% of the quoted values and since at least 1 g samples were dissolved for the experiment, heterogeneity errors should also be minimal. Lovering et al. (1957) deter- mined chromium and copper on a number of the samples using emission spectroscopy. Their data should be accurate to ± 5% of the quoted values. The data of Smales et al. (1967) on Ballinoo, Mount Edith I, Mount Magnet and Youndegin cover the widest range of trace elements ana- lysed by any of the authors. These data were obtained by neutron activation analysis and the accuracy of the results has been discussed in their paper. Table 2 Truce element (mulnrte>i of iron meteorites {in pi>m) .Naino (' J‘ S Cr C'u Zn As Mo IM Ag In Su Sh fRefprt'nces .\voca (Westani Austfalia) 1-2(5 1 1 Ballinoo 5000 •) 4800 300 (500 3 930 4 (5(5 233 < 1 7-0 8-5 3-8 0-04 |<0-01 0-17 (5 Diiketou 2410 300 7 Gundaring 48 15f> 5 Haig 1 2-3 0-1 1 8 ICnmeiina 920 4 0-03(5 1 0-7 ' 8 -Miliv Mitlv 2m) 2000 100 9 1-9(5 1 1 i 1 : 0-1.5 8 Mooranopiiiii .... 0700 4200 i ' j 1 10 (590 1 4 <1 175 ' 1 5 Mount l)son (1919) 1 1 . Sini]ison (1927) 12. Moore et al (1999) 1:1. Sinii)son (1907) 14. Hodtfe-Sniith it White (1929) 15. {’leveiiy A’ 'rhonias (1999) 19. Fletcher (IH87). It is uncertain with what confidence one can quote the pre-1940 analyses of Ward (1898), Simpson (1907, 1916, 1927, 1938), Hodge-Smith and White (1926) and Fletcher (1887). Many of the data given by McCall and DeLaeter (1965) are also pre-1940 analyses performed by a num- ber of analysts, many of whom were employed by the Western Australian Government Chemi- cal Laboratories. Reed (1969) has analysed a number of the siderites listed in Table 2 for phosphorus, but it should be noted that the analyses refer to the kamacite phase only, and not to the meteorite as a whole. Because of the paucity of analytical information on these minor and trace elements, no recommended values have been given. Iron was not determined in this study. It is very difficult to determine a constituent making up approximately 90% of a sample with suffici- ent precision to be comparable with the other measurements. In order to determine an un- measured component the iron value would have to be precise to four significant figures. Many of the analyses in the literature list iron values, but these were usually determined by difference, and are therefore of doubtful value and have not been included in the tables. Nichiporuk and Chodos (1959) have analysed the troilite phase of 12 meteorites for 8 elements, but the only Western Australian siderite in the group is Ballinoo. They list the following data for this meteorite: iron 78.5%, nickel 4.82%, cobalt 0.30%, vanadium <13 ppm, chromium 0.12 ppm, copper 771 ppm, zinc <50 ppm and arsenic <50 ppm. The analytical method used was X-ray fluorescence, and the authors have estimated their accuracy limits for each of the elements analysed. Conclusions Since the publication of the Catalogue of Western Australian Meteorite Collections in 1965, a large and significant body of analytical data has been published by a variety of authors both in Western Australia and overseas. It is now recognised that the structural classification of iron meteorites depends on the kamacite bandwidth and can not be made simply in terms of the nickel content of the meteorite involved. Many of the older designations have there- fore been altered in terms of Buchwald’s classifications. The importance of the gallium-germanium groups has now been recognised as a means of Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 71 classifying iron meteorites on a chemical basis, and the groups have been used to infer genetic and environmental relationships between various meteorites. The theory of cooling rate deter- minations from nickel diffusion profiles has been developed in the last decade and this has en- abled the thermal history of meteorites to be investigated with important conclusions as to the size of the meteorite parent bodies. Another development in recent years has been the use of physical methods of analysis in de- termining the abundance of elements in a wide variety of materials. Data from neutron activa- tion analysis, X-ray fluorescence spectrometry, stable isotope dilution (using solid source mass spectrometry) and electron microprobe analysis have all been reported in this paper. Prior to 1965, traditional wet chemical analyses and emission spectroscopy were the only methods used in the abundance data included in the Catalogue. Many of the Western Australian meteorites had never been analysed for more than one or two elements, and some of the older data were suspect. The present project has in- cluded the analysis of every iron meteorite found in Western Australia for nickel, cobalt, gallium and germanium, including those meteorites dis- covered since 1965. One important outcome of the study has been to provide additional information on paired falls. A number of the Western Australian meteorites are believed to be fragments of the same fall masquerading under different names. The most famous case is that of the Youndegin shower. A diagram reproduced by McCall and DeLaeter (1965, page 57) shows the approximate locations of Youndegin I-VIII, Mooranoppin, Mount Stirling and Quairading. This group of 11 meteorites has a restricted geographical distri- bution, and on the evidence presented in this paper, are members of the same fall. It is hoped to procure additional samples of these meteorites so that a full investigation of all 11 meteorites may be carried out. Another group of meteorites which are thought to be part of the same fall are Loongana Station (iron) and the Premier Downs meteor- ites, (McCall and DeLaeter 1965). To these can now be added Mundrabilla, a large iron meteor- ite tentatively described as being connected to the other meteorites by McCall and Cleverly (1970). All these meteorites come from a re- stricted locality along the Trans-Australian Railway on the Nullarbor Plain. The chemical evidence presented in this paper supports the concept that these are members of the same meteorite shower. It is equally important to know which meteor- ites thought to be paired are, in fact, from separate falls. Mount Edith I and II were found in 1913 and 1914 respectively at locations some 3.2 km apart. They were therefore identified as belonging to the one fall, and the chemical evidence presented in this paper supports this contention. The data presented in this paper have re- inforced the necessity for some type of inter- laboratory comparisons to be made by those involved in meteorite analyses. It is now cus- tomary for laboratories making silicate rock analyses to use the set of standard rocks pro- vided by the United States Geological Survey. The National Bureau of Standards has also been active in producing a variety of standard metal alloys. However no existing metal standards serve the requirements for siderite analyses. It is to be hoped that pieces of a large iron meteor- ite, which have been tested for homogeneity and absence of significant mineral inclusions, might be distributed among laboratories involved in meteorite research, so that future analyses of iron meteorites might be made with more strin- gent confidence limits than has been possible in the past. Acknowledgments. — The author is indebted to Mr. W. W. Thomas for assisting with the X-ray analyses. Most of the meteorite samples were generously supplied by the Western Australian Museum Board to whom appreciation is expressed. The author would also like to thank Dr. G. J. H. McCall, Mr. W. H. Cleverly and Dr. C. Pearson for assisting in many ways. References Anders, E. (1964). — Origin, age and composition of meteorites. Space Sci. Rev. 3: 583-714. Anders, E. (1971). — Meteorites and the early solar system. Ann. Rev. Astron. Astrophys. 9: 1-34. Burbidge, E. M., Burbidge, G. R., Fowler, W. A. and Hoyle, F. (1957). — Synthesis of the elements in stars. Rev. Mod. Phys. 29: 547-650. Cameron, A. G. W. (1968). — A new table of abundances of elements in the solar system, In “Origin and Distribution of the Elements”. (L. H. Ahrens, ed.), 125-143, Permagon Press, New York. Cleverly, W. H. (1968). — Further recoveries of two impact-fragmented Western Australian meteorites, Haig and Mount Egerton. J. Roy. Soc. West Aust. 51: 76-88. Cleverly, W. H. and Thomas, R. P. (1969).— The Yarri Octahedrite iron meteorite. J. Roy. Soc. West. Aust. 52: 89-94. DeLaeter, J. R. (1972). — The isotopic composition and elemental abundance of gallium in meteor- ites and in terrestrial samples. Geochim. Cosmochim. Acta. 36, 735-743. DeLaeter, J. R. and Jeffrey, P. M. (1967). — Tin: its isotopic and elemental abundance. Geochim. Cosmochim. Acta. 31: 969-985. DeLaeter, J. R., McCall, G. J. H. and Reed, S. J. B. (1972).— The Redfields meteorite— a unique iron from Western Australia. Min. Mag. 39. 30-35. Fairbairn, H. W. et al. (1951).— A co-operative investi- gation of precision and accuracy in chemi- cal, spectrochemical and model analysis of silicate rocks. U.S. Geol. Surv. Bull. 980: 1-71. Fleischer, M. (1969).— U.S. Geological survey standards— I. Additional data on rocks G-1 and W-1. 1965-1967. Geochim. Cosmochim. Acta. 33: 65-79. Fletcher, L. (1887). — On a meteoric iron found in 1884 in the Sub-district of Youndegin, Western Australia, and containing Cliftonite, a cubic form of graphitic carbon. Min. Mag. 7 ; 121-130. Frost, M. J. (1965).— Notes on the composition and structure of the Duketon meteorite. J. Roy. Soc. West. Aust. 48: 128. Goldberg, E., Uchiyama, A. and Brown, H. (1951).— The distribution of nickel, cobalt, gallium, palladium and gold in iron meteorites. Geochim. Cosmochim. Acta. 2: 1-25. Goldstein, J. I. (1969).— The classification of iron meteorites. In “Meteorite Research”. (P. M. Millman, ed.), 721-737. D. Reidel, Dordrecht. Holland. Goldstein, J. I. and Short, J. M. (1967).— The iron meteorites, their thermal history and parent bodies. Geochim. Cosmochim. Acta. 31: 1733-1770. Hey, M. H. “Catalogue of meteorites” 3rd Edition. British Museum. London. Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December. 1973. 72 Hodge-Smith. T. and White, H. P. (1926).— The Ticraco Creek Siderite. Rec. Aust. Mus. 15: 66-68. Lovering, J. F., Nichiporuk, W., Chodos, A. and Brown, H. (1957). — The distribution of gallium, germanium, cobalt, chromium and copper in iron and stony-iron meteorites in relation to nickel content and structure. Geochivi. Cosmochim. Acta. 11: 263-278. Mason, B. (1962). — "‘Meteorites” J. Wiley and Sons, New York. McCall. G. J. H. (1965). — A meteorite of unique type from Western Australia. The Mount Eger- ton Stony-iron. Min. Mag. 35: 241-249. McCall. G. J. H. (1965). — New material from, and a reconsideration of, the Dalgaranga meteorite and crater, Western Australia. Min. Mag. 35: 476-487. McCall. G. J. H. and DeLaeter, J. R. (1965). — “Catalogue of Western Australian Meteorites” . Western Australian Museum, Perth. McCall. G. J. H. and Wiik, H. B. (1966).— The Warbur- ton Range nickel-rich ataxite. J. Roy. Soc. West. Aust. 49: 13-16. McCall, G. J. H. (1968). — “First Supplement to Catalogue of Western Australian Meteorite Collec- tions”. Western Australian Museum, Perth. McCall, G. J. H. (1968). — The Avoca meteorite. Mui. Mag. 36: 859-861. McCall, G. J. H. and Cleverly, W. H. (1970). — A review of meteorite finds on the Nullabor Plain, Western Australia, including a description of thirteen new finds of stony meteorites. J. Roy. Soc. West. Aust. 53: 69-80. Moore. C. B.. Lewis, C. F. and Nava, D. (1969). — Superior analyses of iron meteorites. In "Meteorite Research” (P. M. Millman, ed.), 738-748. D. Reidel, Dordrecht, Holland. Nichiporuk, W. and Chodos, A. A. (1959). — The concen- tration of vanadium, chromium, iron, cobalt, nickel, copper, zinc and arsenic in the meteoritic iron sulphide nodules. J. Geophys. Res. 64: 2451-2463. Reed, S. J. B. (1969). — Phosphorus in meteoritic nickel- iron. In “Meteorite Research” (P. M. Mill- man. ed.), 749-762. D. Reidel. Dordrecht, Holland. Reed, S. J. B. (1972). — Personal communication. Rosman, K. J. R. (1972). — A survey of the isotopic and elemental abundance of zinc. Geochim. Cosmochim. Acta. 36. 801-819. Schaudy, R., Wasson. J. T. and Buchwald. V. F. (1972). — The chemical classification of iron meteorites — VI : A reinvestigation of irons with Ge concentration lower than 1 ppm. Icarus. 17. 174-192. Simpson, E. S. (1907). — Notes on a meteorite from the Nuleri District of Western Australia. Bull. Geol. Surv. West. Aust. 26: 24-26. Simpson, E. S. (1916). — Analyses of meteorites found in Western Australia. Bull. Geol. Surv. West. Aust. 67: 135-140. Simpson, E. S. (1927). — Meteoric iron. East Mt. Magnet, Murchison Division. J. Roy. Soc. West. Aust. 13: 47-48. Simpson. E. S. (1938).— Some new and little known meteorites found in Western Australia. Min. Mag. 25: 157-170. Smales. A. A., Mapper, D. and Fouche, K. F. (1967).— The distribution of some trace elements in iron meteorites, as determined by neutron activation. Geochim. Cosmochim. Acta 31- 673-720. Suess, H. E. and Urey, H. C. (1956).— Abundances of elements. Rev. Mod. Phys. 28: 53-74. Taylor. S. R. (1965). — The Wolf Creek iron meteorite. Nature 208: 944-945. Thomas, W. W. and DeLaeter, J. R. (1972). — The analysis of nickel, gallium and germanium in iron meteorites by X-ray fluorescence spectrometry. X-Ray Spectrometry (In press) . Ward, H. A. (1898). — Four new Australian meteorites. Amer. J. Sci. 5: 135-140. Wasson, J. T. (1967a). — Differences of composition among Australian iron meteorites. Nature 216: 880-881. Wasson. J. T. (1967b). — The chemical classification of iron meteorites — I: A study of iron meteor- ites with low concentrations of gallium and germanium. Geochim. Cosmochim Acta. 31: Wasson, Wasson, Wasson, Wasson, Wasson, Wiik, H. 161-180. J. T. and Kimberlin, J. (1967). — The chemical classification of iron meteorites II : Irons and pallasites with germanium concentra- tions between 8 and 100 ppm. Geochim. Cosmochim. Acta. 31: 2065-2093. J. T. (1969). — The chemical classification of iron meteorites — III: Hexahedrites and other irons with germanium concentrations be- tween 80 and 200 ppm. Geochim. Cosmochim. Acta. 33: 859-876. J. T. (1970). — The chemical classification of iron meteorites — IV : Irons with Ge concen- trations greater than 190 ppm and other meteorites associated with Group I. Icarus 12: 407-423. J. T. and Schaudy, R. (1971). — The chemical classification of iron meteorites — V : Groups me and Hid and other irons with germa- nium concentrations between 1 and 25 ppm. Icarus 14: 59-70. J. T. (1971). — Personal communication. B. and Mason, B. (1965). — Analyses of eight iron meteorites. Geochim. Cosmochim. Acta. 29: 1003-1005. Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 73 8. — Eucalyptus forrestiana subsp. dolichorhyncha, a new taxon from Western Australia by M. I. H. Brooker^ Manuscript received 22 February 1972; accepted 22 August 1972 Abstract Eucalyptus forrestiana subsp. dolichorhyncha is described. Its distribution, historical status and relationship to E. forrestiana subsp. for- restiana are discussed. Eucalyptus forrestiana Diels subsp. dolicho- rhyncha M. I. H. Brooker subsp. nov. SLOBEB (Pryor and Johnson 1971). A subspecie typica operculo in rostrum elong- atum 1-3 cm longum contracto et alabastris fructibusque plerumque minoribus in pedicellis brevioribus differt. A subspecies differing from the typical form in the operculum which is abruptly contracted into an elongated beak, 1-3 cm long, and in the buds and fruit which are generally smaller and for shorter pedicels. (Figures 1, 2.) 1 II 1 1 1 1 1 M 1 q 1 1 1 CENTIMETRES A 1 5 6^ 1 1 1 1 1 INCHES 1 1 1 1 1 1 1 Figure 1. — A bud of Eucalyptus forrestiana subsp. forrestiana. 1 1 1 1 1 1 1 ITTj 0 1 ^ ^ 1 CENTIMETRES 4 1 5 0 1 1 1 1 1 1 INCHES I 1 1 1 1 2 1 Figure 2. — A bud of Eucalyptus forrestiana dolichorhyncha. subso. ‘Forestry and Timber Bureau, Canberra, A.C.T. 2600 Herbarium material Holotype: Grasspatch, Western Australia, J. W. Green 1252, 16.iii.l957. Other specimens: Grasspatch, C. A. Gardner 2225 (NSW, PERTH) 22.V.1924, C. A. Gardner and W. E. Blackall 1070 (PERTH) 15.x. 1931. C. A. Gardner (PERTH, PRI.) 31.1.1935, G. H. Burvill (PERTH) 6.viii.l959; Salmon Gums, R. Stamford (PERTH, NSW) 17.vii.1924; south of Grasspatch, J. H. Willis (MEL. 1009877) l.ix.l947; 2 miles south of Grasspatch, L. A. S. Johnson (NSW) 18.xii.l960; 2 miles south of Red Lake, R. D. Royce 4060 (PERTH) 18.iv.1953; 14 miles north of Gibson, M. E. Phillips (CBG 021991, FRI) 5.xi. 1962; 18.7 miles south of Salmon Gums, G. M. Chippendale 180 (FRI, MEL) 13. hi. 1967; 532 mile peg between Norse- man and Esperance, S. G. M. Carr 639 (PERTH) I. iv.l968; Gibson, A. Kessell 880 (PERTH) II. vi. 1969; 13.7 miles south of Salmon Gums, M. I. H. Brooker 2501, 2502 (PERTH) 15.ii.l970; 10 miles south of Peak Eleanora, J. S. Beard 5867 (King’s Park) 30.iii.l970. General distribu- tion is from Salmon Gums to Gibson on the Norseman-Esperance road. Discussion Subspecies dolichorhyncha has been widely referred to in literature and horticulture as E. forrestiana. Its limited area of distribution is crossed by the Norseman-Esperance road in a region where to both the west and east, subsp. forrestiana occurs over a wide tract of country, which has until recently been largely inaccessible (Beard 1973). Consequently subsp. forrestiana has probably been sampled far less frequently than subsp. dolichorhyncha and the latter rather than the former, has been regarded as typical. Nevertheless, both forms exist in Aus- tralian herbaria and both have been planted for ornamental purposes without taxonomic signi- ficance being attached to their differences. In the original description of E. forrestiana (1905), Diels stated that the operculum was pyramidal and illustrated it as such. Maiden (1917) made no reference to any unusual forms of the species and the illustration in the “Critical Revision” (Plate 95, nos. 1, 2), is very close to Diels’ original. Maiden (1929) made a further reference to E. forrestiana in which he referred to the “long rostrate operculum” of a specimen from Grasspatch (C. A. Gardner 2225), but he did not comment on the significance of the variation. This specimen, illustrated in Plate 283 (no. 6B), is subsp. dolichorhyncha. Gardner (1933) considered that Diels had de- scribed an aberrant form in respect of the oper- culum or alternatively that the beak of the Journal of the Royal Society of Western Australia, Vol. 56, Part 3. December. 1973. 74 operculum had become detached from Diels’ specimen. Evidence from recent field collections and from progeny which have been raised to the bud stage (Beard 1973) show that subsp. for- restiana does not produce a long beaked oper- culum. The bud apex is frequently scarred, how- ever. and it is probably on this feature that Gardner based his suggestion that the beak had become detached. Blakely’s redescription of E. forrestiana (1934) refers in part to subsp. doli- chorhyncha. A specimen of historical interest collected by Diels and Pritzel is Pritzel 479 (Perth) whose buds have opercula intermediate in morphology between those typical for the subspecies. From correspondence with the Director of the Botan- isches Museum, Berlin-Dahlem, where it is pre- sumed the type specimen of E. forrestiana was lodged and subsequently destroyed, it is not known what the relationship between Pritzel 479 and Diels 5331 (the type) is. If Diels 5331 is not extant we must take Diels’ description and drawing as typical of E, forrestiana and regard Pritzel 479 as either an intermediate form — a possibility not unlikely between two closely related forms whose distributions we know to be overlapping in some localities, or an immature bud variant. Acknowledgvients. — I wish to thank the OfRcers-in- Charge of the State herbaria in Perth, Melbourne and Sydney for the loan of specimens, and Dr. J. S. Beard who drew my attention to the two forms of E. for- restiana. Thanks are due to Mr. N. Hall for the photo- graphs and to Dr. L. A. S. Johnson for the latin translation. References Beard, J. S. (1973). — The distribution of Eucalyptus forrestiana Diels. J. Roy. Soc. W. Aust. 56: 12-13. Blakely, W. P. (1934). — “A Key to the Eucalypts”. The Worker Trustees, Sydney. Diels, L. and Pritzel, E. (1905). — “Fragmenta Phyto- graphiae Axistraliae Occidentalis” . Leipzig. Gardner, C. A. (1933). — Contributiones Florae Australiae Occidentalis. No. 8. J. Roy. Soc. W. Aust. 19: 79-93. Maiden, J. H. (1917). — “A Critical Revision of the Genus Eucalyptus”. Vol. 3: Govt. Printer, Sydney. Maiden, J. H. (1929).~”A Critical Revision of the Genus Eucalyptus”. Vol. 7: Govt. Printer. Sydney. Pryor, L. D. and Johnson, L. A. S. (1971). — "A Classi- fication of the Eucalypts” . Aust. Nat. Univ. Canberra. Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 75 9. — The ecology and distribution of Eucalyptus forrestiana Diels by J. S. Beard^ Communicated by M. I. H. Brooker Manuscript received 22 Feltruary 1972; accepted 21 November 1972 Abstract The distribution of the two subspecies of Eucalyptus forrestiana has been determined and the range mapped. Altitude, rainfall and asso- ciated soil types are described. Introduction Until recently Eucalyptus forrestiana Diels had been collected only along the main road from Norseman to Esperance, at the type local- ity. This was due to the impenetrability of the country on either side which had not then been settled and was covered with dense mallee in which there were few tracks and no roads. The land boom in Western Australian during the decade up to 1970 entailed the penetration of areas of potential farm land by tracks bulldozed for surveyors engaged in soil survey and land assessment, and it was possible to use these for botanical survey also. Ecology and distribution In November 1967 and in March and Septem- ber 1970 the writer visited this area for purposes of vegetation mapping and traversed many of iRoyal Botanic Gardens, Sydney, N.S.W. 2000. Present address: 6 Fraser Road. Applecross, W.A. 6153 the survey tracks. As E. forrestiana is a con- spicuous species it was readily possible to define its range. The ecology is slightly different on the east and west sides of the Norseman-Esper- ance road. On the west, from the sea between Hopetoun and Esperance a coastal plain slopes gently upwards for about 40 km. inland to an altitude of about 200 m., where the country levels off to a flat plain stretching for a further 30-40 km. This plain has a mallee soil with a differentiated profile of sand over clay, which is winter-wet due to the flatness of the country, with considerable areas having a gilgai surface. Such country with its special soil conditions car- ries a distinctive plant association in which Eucalyptus eremophila P. Muell. and E. forrest- iana are the dominants. The area occupied by this association is therefore the range of E. for- restiana which it is possible to map with some exactitude from aerial photography on the ac- companying diagram (Fig. 1). The range extends for 110 km. to the west of Truslove, and within the hatched area the species is very abun- dant. It has not been observed outside the area shown, presumably due to close association with the particular soil type. On the east side of the Norseman-Esperance road the country is in general similar to that to FORRESTIANA Sutap- (Mkherrhyncha Sutwc. forrastiana tLZJ Conlirajoua distributi> E. •ramcphila Patchy distribution !^d lake RCTCH trr ftaxtT Figure 1. — The distribution of Eucalyptus forrestiana. Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 76 the west and is likewise covered with dense mal- lee associations but there are no longer large extensive areas of the soil-type on which E. for- restiana occurs. The species appears to be in general less common and to be locally frequent in natches of badly drained or gilgai soil. It was recorded as far east as Mt. Ney (Beard 6388), 65 km. to the east of Truslove, beyond which on equivalent soils it appears to be replaced by E. dielsii. In this sector it occupies a belt about the same distance from the coast as its western distribution, indicating some relation to rainfall which averages between 335 and 385 mm. per annum. The gross distribution mapped is that of E. forrestiana subsp. forrestiana. Subspecies doli- chorhyncha M. I. H. Brooker (1973) has been observed only along the Norseman-Esperance main road where it mingles with subsp. forrest- iana, except for a single instance known; Beard 5867 was collected in a heath association 50 km. west of the main road, on a track 16 km. south of Peak Eleanora. It is possible therefore that subsp. dolichorhyncha does extend across the intermediate country between this point and its type locality. The vegetation of this belt of country is gener- ally described as “mallee” and contains many true mallee species of Eucalyptus in which a massive underground rootstock is developed. Following destruction of the top growth by a bush fire the plant sprouts again from the stock with numerous spindly stems. Neither E. for- restiana nor E. eremophila with which it assoc- iates adopts this growth form. Each tree_ is single-stemmed and originated as a seedling after the last bush fire. These species are killed by fire and do not regenerate by coppice. This behaviour has been described by Beard (1967) in E. platypus, which forms “thickets” some 150 km. further west in the Ravensthoroe area where it was noted that E. annulata and E. spathulata also share this habit. Such small trees, reach- ing heights of 5 to 10 m. are strictly-speaking not mallees and are known in Western Australia as “marlocks”. Technically the formation should be known as low forest. Reference Beard, J. S. (1967). — A study of patterns in some West Australian heath and mallee communities. Aust. J. Bot. 15: 131-9. Brooker. M. I. H. (1973 . — Eucalyptus forrestiana suhsp. dolichorhyncha, a new taxon from Western Australia. J. Roy. Soc. W. Aust. 56: 10-11. Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 77 10. — A progeny trial to obtain evidence of hybridity in two taxa of Eucalyptus by J. S. Beard ^ Communicated by M. J. H. Brooker Manuscript received 22 February 1972; accepted 21 November, 1972 Abstract A trial was conducted with progeny of Eucalyptus erythrandra Blakely and Steedman, and of E. forrestiana. Examination of surviving plants after seven years is considered to show that the former taxon is a hybrid form whereas the latter is not. Introduction In November 1962 sesd was collected from two individuals of Eucalyptus erythrandra Blakely and Steedman on the Ravensthorpe- Esperance road about 30 km east of Ravens- thorpe. Another seed collection was made from a population of small trees of E. forrestiana Diels near the 540 mile peg on the Esperance-Norse- man road between Scaddan and Truslove. E. erythrandra was described by Blakely in 1938 from a specimen collected by H. Steedman at Kundip, west of Ravensthorpe. Gardner, however, in 1933 described the same taxon as a new variety, var. robusta, of E. angulosa Schau., and later (Gardner, 1940) announced that “I have since received, through the Con- servator of Forests, specimens of this plant col- lected by Mrs. Daniells of Hopetoun, which exhibit a perfect series embracing on the one hand E. tetraptera Turcz., and E. angulosa Schau. on the other. Amongst the intermediate forms is typical E. erythrandra, which I con- sider to be a hybrid. The evidence in favour of this theory is quite clear.” Progency were raised from the two seed col- lections of E. erythrandra in the King’s Park Arboretum to seek evidence for hybridity. The same was done with the seed collection of E. forrestiana. The taxonomy of E. forrestiana has recently been clarified (Brooker 1973). The specimen and seed collected by the writer in 1962 are now correctly referred to as E. forrestiana Diels subsp. forrestiana and there is no reason to sus- pect interspecific hybridity. In 1962 however, the name E. forrestiana was being incorrectly re- stricted to a form of the species, occurring in the same locality, which differs in having a long beak to the operculum. This taxon has since been described as E. forrestiana subsp. doli~ chorhyncha M. I. H. Brooker. In 1962 the late Mr. C. A. Gardner considered the writer’s collection to be possibly a hybrid between E. forrestiana (as then conceived) and E. stoatei C. A. Gardn., and the seed was there- fore included in the progeny trial. iRoyal Botanic Gardens, Sydney, N.S.W. 2000. Present address: 6 Fraser Road, Applecross, W.A. 6153 Results Seed from the collections made was sown in the King’s Park nursery in summer 1962-1963, as follows: — E. erythrandra. Accession Nos. 2167/62, 2170/62 voucher specimen JSB. 2283. E. forrestiana. Accession No. 2155/62, voucher specimen JSB. 2343a. Plots were planted in the arboretum in May 1963 at 12' X 12' (3.45 x 3.45 m.). 2167/62, 4 lines of 9 “ 36 plants 2170/62, 4 lines of 9 36 plants 2155/62, 9 lines of 10 " 90 plants Each year thereafter the plants were weeded and watered weekly in summer. An assessment was made in May 1967, when in the 2155/62 progeny (E. forrestiana) there were 50 survivors ranging in height from 0.5 to 2 m., the larger ones coming into flower. Of 39 flowering speci- mens all reproduced the parental characteristic of short operculum and 4-ribbed fruit. In the two plots of E. erythrandra both survival and growth were poor. Only 22 plants out of 72 re- mained and these were mostly small and weak, with few flowering. It was observed however that there was a marked segregation into types resembling both reputed parents of the hybrid, E. tetraptera and E. angulosa (Gardner 1940). Final assessment was deferred until April 1970, seven years after planting, with this result — E. forrestiana. 46 survivors, 1.5-3 m tall, all flowering and fruiting copiously but fruits shed prematurely without liberating seed. No appar- ent differences between individuals in habit of tree or leaf size or shape. Short operculum, 46 (100%). 4 major ribs on fruit, 46 (100%). In- dications of minor ribbing on fruit between major ribs, 13 (28%). Slight variations in fruit length and thickness, and in the peduncle, appear but not seeming significant. E. erythrandra. 20 plants of which 2 dead; these still had leaves and were included in the ex- amination. Plants very variable, 12 assessed as resembling E. angulosa (upright branching habit, smaller, thinner leaves, compound in- florescences, small fruit of E. angulosa type),. 5 resembling E. tetraptera (straggly and decum- bent form, thick large rigid leaves, large soli- tary 4-angled fruit) and 3 intermediate. There was a difference between the two collections,- progeny of 2167/62 having 4 angulosa-iype, 5 tetraptera~type, 0 intermediates, 2170/62 having 8 angulosa-iype, 0 tetraptera-type, 3 inter- mediates. Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 78 Conclusions It is concluded that there is no evidence of hy- bridity in the progeny of E. forrestiana subsp. forrestiana, nor of variation to ssp. dolicho- rhyncha. There was no occurrence of the long- beaked form in the population, nor any inter- mediates. On the other hand Gardner’s supposition of hybridity in E. erythrandra is very clearly sup- ported. The only consideration here lies in the identity of the reputed parents. E. tetraptera occurs in the mallee-heath communities, east of Ravensthorpe where E. erythrandra forms are found but E. angulosa does not, being a coastal species. E. incrassata, on the other hand, an- other member of the “dumosa” group and of which E. angulosa has been considered a variety by some authorities, does so, and this is probably the actual parent. It has smaller fruits than E. angulosa but is sufficiently close to it not to affect materially the comparisons in the progeny trial. This paper is presented in order to draw attention to the utility of progeny trials con- ducted in Botanic Gardens to assist the elucida- tion of taxonomic problems. References Brooker, M. I. H. {1913) .—Eucalyptus forrestiana subsp. dolichorrliyncha, a new taxon from Western Australia. J. Roy. Soc. West. Aust. 56: 10-11. Blakely, W. F.. McKie, E. N. and Steedman, H. (1938).— Descriptions of four new species and two varieties of eucalyptus. Proc. Linn. Soc. N.S.W. 63: 65-9. Gardner. C. A. (1933). — Contributiones Florae Austra- liae Occidentalis 8. J. Proc. Roy. Soc. W. Aust. 19: 79-93. Gardner, C. A. (1940). — Contributiones Florae Austra- liae Occidentalis 11. J. Proc. Roy. Soc. W. Aust. 27: 165-210. Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 79 11. — Harris Sandstone-Yindagindy Formation relationships and possible position of permo-carboniferous boundary, Carnarvon Basin, Western Australia by J. F. Read^ P. J. Alcock^ and P. Hoseman^ Manuscript received 18 July 1972; accepted 19 September 1972 Abstract This paper describes the stratigraphic rela- tionships between plant-bearing sediments of possible Carboniferous or Permian age and underlying Yindagindy Formation sediments (Lower Carboniferous) exposed in the eastern Carnarvon Basin. The sequence is, in descending order: 3. Lyons Group (Permian), 2. Harris Sandstone, and 1. Yindagindy Formation (Lower Carboni- ferous). In the Moogooree area, plant-bearing sand- stones previously included in the Harris Sand- stone have been assigned to the Lyons Group by White and Condon (1959). The main reasons for the change were that the plant-bearing beds differed lithologically from the Harris Sandstone, and rested with erosional uncon- formity on a surface of marked relief. Detailed mapping in the Moogooree area indicates that these plant-bearing beds are similar lithologi- cally to the Harris Sandstone and that they are mildly disconformable upon the Yindagindy Formation. An angular unconformity which separates the Permian, Lyons Group from the underlying Harris Sandstone probably repre- sents the Carboniferous-Permian boundary in the area. Introduction This paper describes the stratigraphic rela- tionships between plant-bearing beds of Car- boniferous or Permian age and the underlying Lower Carboniferous Yindagindy Formation in the Moogooree area, eastern Carnarvon Basin, Western Australia (Fig. 1). The “Yindagindi limestone” of Teichert (1950) was renamed Yindagindy Formation by Condon (1954), who considered it to be of Carboniferous age because of its conformable relationship with underlying Carboniferous beds. Overlying plant-bearing sandstones in the Moogooree area were informally named the “Red Hill standstone”, by Teichert (1950). Condon (1954, p.31) renamed the plant-bearing beds the Harris Sandstone, making the type section approximately 31 km. (19 mi.) north-north west of Moogooree Homestead; at the type section, the Harris Sandstone was considered to con- formably underlie the Lyons Group and to dis- conformably overlie the Yindagindy Formation. Condon (1954, p.34) excluded the Harris Sand- stone from the Lyons Group because of lithologic and genetic differences. ^Department of Geology, University of Western Aus- tralia. Nedlands. Western Australia, 6009. Present address: Dept, of Geological Sciences, Virginia Polytechnic and State University, Blacksburg, Virginia. U.S.A. 2piacer Prospecting (Aust.) Pty. Ltd., Box 4315, G.P.O., Sydney, 2001. :^West Australian Petroleum Pty. Ltd., Perth, Western Australia. 6000. Later, the plant-bearing standstones around Moogooree were included in the Lyons Group as they were considered to be lithologically similar to non-tillitic parts of the Lyons Group and differed lithologically from the Harris Sand- stone at the type section (White and Condon 1959, p.58; Condon 1967, p.7). This implied that the Harris Sandstone was absent from the sec- tion exposed at Moogooree. Furthermore, Con- don (in White and Condon 1959, p.58) consider- ed that the plamt-bearing beds in the Moogooree area rested unconformably on an erosional sur- face of marked relief developed on Yindagindy Formation sediments. In contrast, Dickins and Thomas (1959) con- sidered that the plant-bearing sandstones around Moogooree were equivalent to the Harris Sandstone at its type section as the rocks formed a distinct stratigraphic unit on the basis of lithology, field occurrence and stratigraphic position and that they are distinguishable from the Lyons Group in that they are non-tillitic. In this paper, these plant-bearing sandstones will be referred to as Harris Sandstone. The Yindagindy Formation is considered to be of Lower Carboniferous age by Thomas (1962). Dickins and Thomas (1959) assign a Lower Permian age to the Lyons Group. How- ever, the age of the Harris Sandstone is at present inconclusive; White (in White and Con- don 1959) considers that the lepidodendroid plant material could be of Carboniferous or Permian age but Krausel (1961) implies that it is Carboniferous. Stratigraphy Determination of contact relationships be- tween the Lower Carboniferous Yindagindy Formation and the Carboniferous or Permian Harris Sandstone around Moogooree has been hindered by faulting, by poor exposure of Yindagindy Formation-Harris Sandstone con- tact and Harris Sandstone-Lyons Group contact and by the absence of adequate fossil material for determining age of the Harris Sandstone and basal Lyons Group beds. The Yindagindy Formation-Harris Sandstone contact at localities 1 to 4 (Fig. 1) was mapped by plane table at a scale of 1:1200; a traverse was run at locality 5. To facilitate mapping, the Yindagindy Formation was divided into 4 units. The sequence in the Moogooree area is, in descending order: Harris Sandstone: Thickness at type section, 85 metres (Condon 1967, p. 11). In the Moogooree area the Harris Sandstone occurs in fault-blocks which form low red hills; local total Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 80 Figure 1 Journal of the Royal Society of Western Australia. Vol. 56. Part 3, December. 1973. 81 thickness of the unit is not known. Contacts between the Harris Sandstone and the overlying Lyons Group are concealed beneath a residual cover of pebbles and boulders from glacial beds of the Lyons Group, to- gether with siliceous laterite fragments. The Harris Sandstone consists of thinly bedded or cross-bedded, well-sorted, medium- to coarse-grained quartz sand- stones; the sandstones are clean and free of detrital matrixL Cortical impressions and casts of lepidoden- droid stems are common near the base of the for- mation. The Harris Sandstone disconformably over- lies. Yindagindy Formation: Maximum thickness in the Moogooree area, 68 metres. Unit D . — (20 metres). This unit consists of fine- grained lime mudstones and interbedded, poorly exposed terrigenous sediments. The lime mudstones are lami- nated and have birdseye structures: they are poorly fossiliferous. The uppermost beds of the Yindagindy Formation are poorly exposed, thin ferruginous quartz sandstones (2 metres thick) and an underlying blue vuggy limestone which forms a distinctive marker bed. The blue vuggy limestone is characterized by abund- ant subvertical tubes up to 1 cm in diameter which are filled with sparry calcite. Locally where the quartz sandstone has been eroded, the blue, vuggy limestone marks the top of the formation. At locality 1. a limestone pebble conglomerate appears to be laterally equivalent to or overlies the blue, vuggy limestone bed. U7iit C . — (38 metres). This unit crops out as a low strike ridge. The unit consists of poor-sorted, medium- to coarse-grained calcareous feldspathic sandstones which are locally cross-bedded, together with poorly exposed siltstones and claystones, and intercalated oolitic limestones. Rare fossils include brachiopods, gastropods, serpulids and algal structures. UNIT B . — (4.5 metres). This unit crops out as a bluff. It consists of brown, thick-bedded, quartzose lAll specimens of Harris Sandstone from the Moogooree area examined in hand specimen and thin section were clean quartz sandstones with little or no detrital mat- rix: Condon {in White and Condon 1959, p. 58) de- scribed this sandstone as a silty quartz greywacke skeletal-fragment limestones and minor oolitic and coquinoid limestones. Fosiils include brachiopods. crinoids and bryozoans. UNIT A. — (5.5 metres). This unit is generally ob- scured by a rubble slope flanking the bluff formed by unit B limestones. It consists of fine- to coarse- grained, thin-bedded or cro-s-bedded, poor-sorted cal- careous feldspathic sandstones with two limestone horizons, the lower one marking the base of the yindagindy Formation: it conformably overlies, Williamhury Formation: Thickness at the type section, 235 metres (Condon 1967, p.69). Friable, poorly sorted, medium- to coarse-grained feldspathic sandstone. Harris Sandstone — Yindagindy Formation Contact Contacts between the Harris Sandstone and Yindagindy Formation are typically faulted, and the abundance of faults suggest that the area lies within a fault zone. The faults are evi- denced by stratigraphic discrepancy where upper beds of the Yindagindy Formation or lower parts of the Harris Sandstone are missing. Other features associated with the faults are slickensides in sandstones of the Harris Sand- stone and Yindagindy Formation, calcite veins in limestones and folding of limestone beds. Relative movement on faults is generally west — block down with variable lateral displacement. Stratigraphic relationships between the forma- tions were determined where sedimentary con- tacts have not been obscured by faulting. Such contacts occur in the southern portion of locality 1, in a fault block at locality 2, at the southern end of locality 4 and at locality 5. The stratigraphic succession and probable thickness Figure 2. — Columnar sections of the Yindagindy Formation, Moogooree area, localities 1 to 5. Blank pattern denotes soil covered areas. Journal of the Royal Society of Western Australia. Vol. 56, Part 3, December, 1973. 82 of the Yindagindy Formation at each locality are shown in Figure 2. Detailed maps of locali- ties 1 to 4 are shown in figs. 3 and 4. Locality f.— The Harris Sandstone at the southern end of locality 1 rests on blue, vuggy limestone and contains abundant cortical im- pressions of lepidodendroid stems. Locally, a limeston-pebble conglomerate (up to 1 metre thick) underlies the Harris Sandstone; the con- glomerate is probably stratigraphically equivalent to the blue, vuggy limestone. Dips of the Harris Sandstone, and Yindagindy Formation at the contact are flat-lying; dips are unreliable in determining stratigraphic relationships as some beds are displaced by faults. Locality 2.— The Harris Sandstone occurs as low residual mounds up to 3 metres high in a fault block in the central portion of locality 2. It contains plant fossils and rests upon the blue, vuggy limestone. The contact is well ex- posed and the sediments are almost horizontal. West of this fault block unit D is exposed as west dipping limestone bands, the topmost band being blue, vuggy limestone. West of this bed the Harris Sandstone crops out as rubble in soil. Locality 3 . — The sequence at locality 3 is heavily faulted and formation contacts are con- cealed beneath soil. A thick section of Yinda- gindy Formation is exposed at this locality (Fig. 2 ). Locality 4 . — Contacts are well exposed in and and south of a north-east flowing creek in the central portion of locality 4. Here, plant- bearing sandstones rest upon 2 metres of fer- ruginous, non-fossiliferous quartz sandstone which overlies blue, vuggy limestone. Lime- stones of unit D are well exposed at this locality. Dips of 20 to 25 degrees above and below the contact suggest conformable or dis- conformable contacts. Locality 5. — At locality 5, plant-bearing beds overlie thin ferruginous, non-fossiliferous sand- stones which rest on blue, vuggy limestone. Lithological correlations of Yindagindy Forma- tion limestones of unit D indicate that the blue, vuggy limestone is the same stratigraphic hori- zon in all areas mapped. Conclusions In the Moogooree area, the Harris Sandstone disconformably overlies a thin (2 metres) fer- ruginous sandstone or ( where this has been eroded), a distinctive, blue, vuggy limestone horizon. The ferruginous sandstone and under- lying blue, vuggy limestone are the uppermost beds of the Yindagindy Formation in the area. Locally, a limestone-pebble conglomerate (up to 1 metre thick) which is closely associated with blue, vuggy limestone, lies at the top of the Yindagindy Formation. The ferruginised top of the Yindagindy Formation and the sharp change in lithology from the Yindagindy Formation to the Harris Sandstone indicate a disconformity (Condon 1954, p.30). However, in the Moogooree area, Figure 3. — Geologisal map, locality 1. Journal of the Royal Society of Western Australia, Vol. 56, Part 3. December, 1973. 83 Figure 4.— Geological maps, localities 2. 3 and 4. Journal of the Royal Society of Western Australia. Vol. 56. Part 3, December, 1973. 84 the disconformity surface has little relief, prob- ably 2 metres maximum, and is essentially parallel to the underlying Yindagindy Forma- tion beds. There is no evidence for an ero- sional unconformity of marked relief at the base of the plant-bearing sandstones as re- ported by Condon (in White and Condon 1959*. Removal of parts of the Yindagindy Formation and Harris Sandstone by later faulting has brought Harris Sandstone or Lyons Group sedi- ments into contact with a truncated sequence of the Yindagindy Formation, simulating an ero- sional contact of high relief. Lithologically, the sandstones are similar to the Harris Sandstone at the type section, being clean quartz sand- stones rather than silty quartz greywackes as reported by White and (I^ondon (1959, p.58L Thus the plant-bearing beds are referred to the Harris Sandstone and not the Lyons Group as White and Condon (1959) and Condon (1962) proposed. South of Moogooree homestead, the Lyons Group progressively truncates older sediments till it finally rests upon the Precambrian base- ment rocks (Condon 1962). As the Harris Sandstone is concordant with the underlying Yindagindy Formation, an angular unconformity probably exists between Harris Sandstone and overlying Lyons Group, instead of the conform- able contact of Condon (1954). The contact between the Harris Sandstone (Carboniferous?, Krausel 1961) and the Lower Permian, Lyons Group (Dickins and Thomas 1959) may mark the Carboniferous-Permian boundary in the region. Acknowledgements. — This paper is part of a joint honours thesis submitted to the Department of Geology. University of Western Australia in 1966. Field work was funded by The University of Western Australia and a grant from West Australian Petroleum Pty. Ltd. The project was supervised by B. W. Logan; thanks are also due to J. J. E. Glover. P. J. Coleman and B. E. Balme of the Geology Department, The University of Western Australia. M. H. Johnstone, of West Aus- tralian Petroleum Pty. Ltd. introduced the authors to the area. Figures were prepared by C. Hughes and C. Bell. References Condon, M. A. (1954).— Progress report on the strati- graphy and structure of the Carnarvon Basin. Western Australia. Bur. Min. Resour. Aust Rep 15 Condon, M. A. (1962). — Kennedy Range, W.A. — 1:250,000 Geological Series. Bur. Min. Resour. Aust. Explan. Notes. 28 pp. Condon, M. A. (1967). — The geology of the Carnarvon Basin. Western Australia. Pt. 2: Permian Stratigraphy. Bur. Min. Resour. Aust. Bull. 77. Dickins, J. M. and Thomas, G. A. (1959). — The marine fauna of the Lyons Group and the Car- randibby Formation of the Carnarvon Basin, Western Australia. Bur. Min. Resour. Aust. Rep. 38: 65-96. Krausel. R. (1961). — Lycopodiopsis derbyi Renault und einige andere Lycopodiales aus den Gond- wana-Schichten. Palaeontographica, Abt. B, 109 (1-4): 62-92. Teichert, C. (1950). — Some recent additions to the stratigraphy of Western Australia. Bull. Amer. Ass. Petrol. Geol. 34 (9): 1787-1794. Thomas, G. A. (1962). — The Carboniferous stratigraphy of Western Australia. G.R. 4'^eme congr. Strat. Geol. Carb. (Heerlen, 1958) T III: 733- 740. White. Mary E. and Condon, M. A. (1959). — A species of Lepidodendron from the basal Lyons Group, Carnarvon Basin. Western Australia. Bur. Min. Resour. Aust. Rep. 38; 55-64. Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 85 12. — The genus Ctenotus (Laeertilia, Scincidae) in the South-West and Eucla Divisions of Western Australia by G. M. Storr‘ Manuscript received 20 February 1973; accepted 20 March 1973. Abstract The following 17 taxa are defined and keyed out: pantherinus pantherinus (Peters), pan- therinus ocellifer (Boulenger), lesueurii (Dum- eril & Bibron), fallens sp. nov., severus Storr, alleni sp. nov., mimetes Storr, uber uber Storr. atlas Storr, impar Storr, lancelini Ford, labil- lardleri (Dumeril & Bibron), gemmula sp. nov., delli sp. nov., catenifer sp. nov., schomburgkii (Peters), and brooksi euclae Storr. Lectotypes are designated for Lygosoma lesueurii Dumeril & Bibron and Tiliqua australis Gray. Intrcduction This is the fourth in a series of regional sur- veys of the genus Ctenotus. Previous papers covered the Eastern Division of Western Aus- tralia (Storr 1969), the Northern Territory (Storr 1970), and South Australia (Storr 1971). This paper is based on material in the Western Australian Museum (registered numbers without prefix) and a few specimens kindly loaned by the National Museum of Victoria (NMV), the South Australian Museum (SAM), Museum of Comparative Zoology (MCZ), Museum National d’Histoire Naturelle (Paris), and Zoological Museum of Humboldt University (Berlin). The number of specimens examined in each taxon were p. pantherinus (20), p. ocellifer (1), lesueurii (49), fallens (162), severus (10), alleni (2), mimetes (7>, u. uber (6), atlas (2), impar (45), lancelini (5), labillardieri (368), gemmula (25), delli (11), catenifer (14), schomburgkii (25), and brooksi euclae (30). In the descriptions of species quantitative characters are usually expressed as ranges with means in brackets. The term “palpebrals” here applies to the scales along the free edge of the upper eyelid. The term “calli” refers to thick- enings of the subdigital lamellae too broad to be called keels. “Presuboculars” are the scales aligned with and immediately posterior to the loreals. Key to species and subspecies 1. Dorsal and lateral pattern lacking ocelli 2 Dorsal and lateral pattern consist- ing wholly or mainly of black-and- white ocelli — pantherinus group 6 2. Dorsal and lateral pattern usually including longitudinal series of spots, blotches, dots and dashes, as well as stripes and lines 3 Dorsal and lateral pattern consist- ing solely of alternating longitudi- nal black and white stripes — taeniolatus group 11 1 Western Australian Museum, Perth, Western Australia 6000. 3. Subdigital lamellae smooth, callose or obtusely keeled; snout-vent length up to 105 mm Subdigital lamellae sharply keeled; snout-vent length up to 52 mm — schomburgkii group 4. Three supraoculars normally in contact with frontal; presubocu- lars 2 Usually two supraoculars in con- tact with frontal; presuboculars 3 labillardieri group 5. Subdigital lamellae smooth or broadly callose; SVL up to 105 mm — lesueurii group Subdigital lamellae obtusely keeled or narrowly callose; SVL up to 77 mm — leonhardii group 6. A black vertebral stripe (some- times incomplete or discontinuous) No vertebral stripe 7. A well-defined black vertebral stripe, narrowly edged with white Vertebral stripe absent or reduced to a black line 8. Nuchals normally 4 or 5; usually a white line on nape between para- vertebral and dorsolateral white lines, extending forward along edge of frontal and backward for vary- ing extent; a series of oblique white bars behind arm; midbody scale rows seldom more than 26; lamellae under fourth toe seldom less than 24 Nuchals normally 2 or 3; no white line between paravertebral and dorsolateral lines; no oblique white bars behind arm; midbody scale rows seldom less than 28; lamellae under fourth toe seldom more than 24 9. White dorsolateral line margined above by broad blackish laterodor- sal stripe and separated below from dark upper lateral zone by narrow hiatus of pale ground colour; lamellae under fourth toe less than 25; nuchals 2 or 3 White dorsolateral line margined above by narrow black laterodorsal stripe and in contact below with black of upper lateral zone; lamel- lae under fourth toe more than 25; nuchals 4 .... 10. Midlateral white stripe well de- fined; dark laterodorsal stripe not enclosing a series of pale spots ... Midlateral stripe not or barely dis- cernible; laterodorsal stripe enclos- ing a series of pale spots 11. White stripes totalling 11 (includ- ing a vertebral stripe) or 12 (with- out a vertebral stripe): nasals and prefrontals separated; labials usu- ally 7 White stripes totalling 8 or 10 (no vertebral stripe); nasals and pre- frontals seldom separated; labials usually 8 12. White dorsolateral line continuous; abdomen yellow; subdigital calli wide White dorsolateral line broken into a series of short dashes; abdomen white; subdigital calli narrow 4 16 5 12 7 10 p. pantherinus p. ocellifer 8 9 lesueurii fallens severus alleni mimetes u. uber impar atlas 13 14 Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December. 1973. 86 13. Hindleg reddish Hindleg yellowish 14. Hindleg yellowish brown, boldly marked with black and white: labials 8 Hindleg olive-grey, dotted with blackish brown; labials seldom more than 7 15. Back dark brown, unpatterned be- tween narrow black laterodorsal stripes Back olive grey, strongly patterned (broad blackish laterodorsal stripe often bearing a series of white dots; usually some trace of dark vertebral line) 16. Dorsal ground colour dark (olive grey or olive brown); presuboculars 2; plantars opposite fourth toe not differentiated Dorsal ground colour very pale (pinkish in life, white in alcohol); one presubocular; plantars oppo- site fourth toe enlarged and keeled lahillardieri lancelini gemmula 15 delli cateiiifer schomburgkii brooksi euclae Ctenotus pantherinus pantherinus Lygosoma (Hinulia) pantherinum Peters. 1866, Mber. Akad. Wiss. Berlin 1866; 89. Swan River, Australia. DiaQUosis. A large Ctenotus whose dorsal and lateral pattern consists mainly of black-and- white ocelli; subdigital lamellae sharply keeled; nasal grooved. Distribution. Northern interior of South-West Division, from the lower Murchison, south to Mt Lesueur. New Norcia and Quairading. Extra- limital in far southwest of Eastern Division (16 mi. S of Karalee*. Description. Snout-vent length (mm); 36-90 (67.9). Length of appendages (% SVL) : tail 152-186 (173), foreleg 22-31 (26.5), hindleg 39-45 (41.1). Nasals in contact. Prefrontals usually in con- tact, occasionally separated very narrowly. Supraoculars 4, first three in contact with frontal. Supraciliaries usually 7, occasionally 8, mean 7.2. Palpebrals 9-13 (11.4). Second loreal 0.8-1. 4 (1.02) times as wide as high. Upper labials usually 8, occasionally 9. mean 8.1. Ear lobules 4-8 (5.7), obtuse in juveniles, subacute in adults. Nuchals usually 2 or 3, occasionally 1 or 4. mean 2.7. Midbody scale rows usually 32 or 34. seldom 30 or 36, mean 33.4. Lamellae under fourth toe 22-25 (23.6). Dorsal and lateral ground colour coppery brown, occasionally washed with olive-green. Black vertebral stripe usually extending from nape to base of tail, but sometimes disappearing at midback or becoming broken posteriorly. Usually 5, sometimes 4, longitudinal series of ocelli on each side of body, viz. a paravertebral, a dorsolateral and two or three lateral, each ocellus consisting of a short white bar margined on each side by a short black bar. White bars of dorsolateral and midlateral ocelli in some specimens almost continuous enough to form stripes. Material. South-West Division: Janja Thicket, 16 mi. ENE of Kalbarri (37616); Galena (29627); Binnu (25599); 14 mi. NE of Morawa (17299); Mt Lesueur (11162-3); 7 mi. NE of Miling (17300-5); 7 mi. N of New Norcia (17306); Dangin (17307); near Quairading (2482); Wamenusking (8555); Bruce Rock (21371). Also the holotype (Berlin 5379). Ctenotus pantherinus ocellifer Lygosovia ocellifcrinii [Boulengerl, 1896. Ann. Mag. Nat. Hist. (6) 18; 342. Broome. W.A. (K. Dahl). Diagnosis. Differing from C. p. pantherinus mainly in lacking black vertebral stripe. Distribution. Arid mallee-spinifex zone of Eucla Division. Extralimital in Kimberley, North-West and Eastern Divisions of Western Australia and in Northern Territory and South Australia. Description. See Storr (1969; 99; 1970; 98). Material. 8 mi. E of Fraser Range (30756). Ctenotus lesueurii Lygosoma lesueurii Dumeril Sc Bibron, 1839, Erpetologie generale 5; 733. Nouvelle Hollande (Peron & Lesueur). Tiliqua australis Gray, 1839, Ann. Nat. Hist. 2: 291. Australia. [ = Lygosoma australe (Gray) of Peters, not Lygosoma australis Gray.] Diagnosis. A large member of the lesueurii group with a pale-edged vertebral stripe and a dark-edged dorsolateral line. Distinguishable from C. fallens by its brighter and more complex pattern ( including a pale dorsal line on nape between paravertebral and dorsolateral lines, extending forward along edge of frontal and for varying distances backward; and pale dark- edged oblique ventrolateral bars behind arm), more numerous nuchals (rarely less than 4), fewer midbody scale rows (rarely more than 26), and more numerous lamellae under fourth toe (seldom less than 24). Distribution. West coast and coastal plains south to Augusta, thence east to the Albany district. Description. Snout-vent length (mm); 43-101 (79.8). Length of appendages (% SVL); tail 200-253 (230), foreleg 24-30 (26.0), hindleg 41-53 (46.7). Nasals separated. Prefrontals in contact (occasionally separated very narrowly or by an azygous scale). Supraoculars 4, first 3 in contact with frontal, second very large. Supraciliaries 6-9 (rarely other than 7; mean 7.0). Palpebrals 10-14 (11.9). Second loreal 1.0-2. 2 (1.47) times as wide as high. Upper labials 8 (occasionally 9; mean 8.1). Ear lobules 3-7 (mean 4.7); obtuse in juveniles; acute, subacute or truncate in adults; third mostly largest. Nuchals usually 4, occasionally 5, rarely 3, mean 4.2. Midbody scale rows 24-28 (25.1). Lamellae under fourth toe 23-28 (25.9), smooth or broadly callose. Dorsal ground colour greyish brown in adults, coppery brown in juveniles. Blackish-brown vertebral stripe from nape to base of tail, much narrower than a paravertebral scale, with a narrow white margin, which in turn is narrowly margined with black. White dorsolateral line from rear of orbit to tail, on which it is suffused with brown; margined above with black (laterodorsal stripe). A dorsal line on nape between black paravertebral line and black laterodorsal stripe, extending forward along outer edges of frontoparietal and frontal. Upper lateral zone blackish brown with a series of white spots or dashes. White midlateral stripe from ear nearly to end of tail, anteriorly breaking up and tending to join with white ventrolateral spots which behind arm are Journal of the Royal Society of Western Australia. Vol. 56, Part 3, December, 1973. 87 modified into short oblique bars. A white line curving under orbit. Upper labials edged with brown. Reviarks. Of the three extant syntypes of Lygosoma lesueurii in the Paris Museum, I choose as lectotype no. 2982, collected by Peron & Lesueur [presumably on the west coast of Western Australia in 18011; this is the specimen whose measurements are given in the original description. Mr. A. F. Stimson of the British Museum tells me (in litt., 7 January 1972) that none of their specimens can be certainly identified as type of Tiliqua australis Gray. In order to stabilise that name, I designate the lectotype of Lygosoma lesueurii Dumeril & Bibron as neotype of Tiliqua australis Gray. Material. South-West Division: Meanarra Hill, 4 ml. E of Kalbarri (33529); Wittecarra Gully, 5 mi. SE of Kalbarri (33882); 11 mi. SSE of Kalbarri (33784); 1 mi. SSW of Kalbarri <33668-9, 33712); between Cockleshell Gully and Jurien Bay (12695); Favourite Island (17202); near mouth of Hill River (13440, 17205); Green Islets (17208-9); Muchea (462); Scarborough (28925); Mt Yokine (19112, 21714, 21885); Bedford (9121); North Perth (4840); Leederville (8885); Kings Park (17239); Nedlands (6672); South Perth (29652-3, 29772-5, 30252-4, 33384-5, 39085, 39983, 40842-5); Riverton (28330); Gos- nells (10978); Mandurah (40847); Augusta (30235); Chorkerup (4538). Ctenotus fallens sp. nov. Holotype. R 33780 in Western Australian Museum, collected by Lawrence A. Smith on 6 February 1969 at 11 mi. SSE of Kalbarri, Western Australia, in 27° 52' S, 114° 12' E. Diagnosis. Generally similar to C. lesueurii, differing in duller and simpler pattern (e.g., no dorsal pale line between paravertebral and dor- solateral pale lines, and no oblique ventrolateral barring behind arm), fewer nuchals (seldom more than 3), more numerous midbody scale rows (usually more than 26), and fewer lamellae under fourth toe (seldom more than 24). Distribution. Northern half of South-West Division, south to Pinjarra and inland to the Loop (lower Murchison), Balia Tank, Koolan- ooka Hills and Corrigin; also on Houtman Abrolhos (West Wallabi, Middle, Rat, Helsinki, and Hut Islands), Jurien Bay Islands (Green, Boullanger and Favourite), Wedge Island, Lancelin Island and Rottnest Island. Extra- limital in North-West Division. Description. Snout-vent length (mm) : 35-95 (68.7). Length of appendages (% SVL) : tail 175-261 (218); foreleg 20-32 (25.9); hindleg 35-54 (43.8). Nasals usually separated. Prefrontals in con- tact (very narrowly separated in one specimen). Supraoculars 4, first 3 (2 in 2 specimens) in contact with frontal. Supraciliaries 6-8 (mostly 7, rarely 6, mean 7.2). Palpebrals 9-14 (11.3). Second loreal 1.1-2. 3 (1.55) times as wide as high. Upper labials 7-9 (mostly 8, rarely 7, mean 8.1). Ear lobules 2-6 (3.8), subacute, acute or truncate in adults, obtuse in juveniles, second or third usually largest. Nuchals 1-5 (mostly 2, rarely more than 3, mean 2.3). Midbody scale rows 25-33 (28.1). Lamellae under fourth toe 17-26 (22.0), smooth or broadly callose. Dorsal ground colour dark or pale greyish brown in adults, blackish in juveniles. Blackish brown vertebral stripe from nape to base of tail, narrower than a paravertebral scale, with a narrow white margin, which in turn may be edged with black. White dorsolateral line from above temples to tail, on which it is suffused with brown, broadly margined above with black. Upper lateral zone dark brown with a series of white blotches, spots or dashes. White mid- lateral stripe from behind eye to tail. Lower lateral zone pale grey or pale brown, irregularly spotted with white. Indistinct white line curving under eye. Material. South-West Division; Gee Gie Out- camp, 21 mi. NNW of Murchison House (34039); Kalbarri (29924); The Loop, 22 mi. NE of Kal- barri (33868); Meanarra Hill, 4 mi. E of Kal- barri (33538); 19 mi. E of Kalbarri (33585); Red Bluff (33875, 37643); Lockwood Spring and Hawks Head Lookout, 20 mi. SE of Kalbarri (33847, 33872, 37565-7); 11 mi. SSE of Kalbarri (33776-83); 11 mi. SSW of Kalbarri (33666-7, 33711); Balia Tank, 23 mi. E of Ajana (40850); Koolanooka Hills (29580); Lake Arrowsmith (40849): 40 mi. S of Dongara (19762); Stock- yard Gully (26743-5); Green Head (30289); 5 mi. NE of Mt Peron (25284); Mt Lesueur (11161); 5 mi. N of Jurien Bay (17200); 5 mi. NE of Jurien Bay (30478); 5 mi. E of Jurien Bay (30503); Frenchman Bay (17206-7); main- land opposite Green Islets (17210); 10 mi. NE of LanceJin (22282); Lancelin (17214); Ledge Point (33428-31); 7 mi. N of New Norcia (17216-20); Beermullah (4807); between Mo- gumber and Gingin (30233); Gingin (40036); Culham (17221, 22450-4); Julimar Forest, 15 mi. NW of Toodyay (36332); 3 mi. SE of Bulls- brook (36329); Sorrento (41641-2); Balcatta (14860); North Perth (4838-9); City Beach (17238, 37742; NMV D9798); Cottesloe (7869); Wembley (13111); Gooseberry Hill (24689); 6 mi. E of Kalamunda (34082); Mundaring Weir (14856-7, 16532, 19656, 20594, 21228, 26448, 26475, 40019); 10 mi. SE of Sawyers Valley (22672); Darlington (5987-9, 21263-5); Bickley (13543); Roleystone (17240-5); Karragullen (17246-7); Gosnells (29328); Kelmscott (41177); York (7320, 12663-5); Corrigin (12434); Darling Scarp, between Pinjarra and Dwellingup (25095) ; West Wallabi Island, Houtman Abrolhos (17253-62. 19602-3); Middle Island, Houtman Abrolhos (27185) ; Hut Island, Houtman Abrolhos (37518, 41550); Rat Island, Houtman Abrolhos (41535, 41556); Helsinki Island, Houtman Abrolhos (41546); Green Island, 12 mi. N of Jurien Bay (17199, 17201); Favourite Island, Jurien Bay (17203); Boullanger Island, Jurien Bay (17204); Wedge Island (17211-3); Lancelin Island (17215); Rottnest Island (3266-8, 17129, 17222-36; NMV R974-81). Ctenotus severus ctenotus severus Storr, 1969: 101. Galena, W.A. (G. M. Storr) . Diagnosis. A medium-sized member of the lesueurii group, generally similar to C. fallens and differing mainly in colour pattern — black Journal of the Royal Society of Western Australia, Vol. 56, Part 3. December, 1973. 88 vertebral stripe absent or reduced to a line on foreback or neck and not white-edged; black laterodorsal stripe broad and sharp-edged; dark upper lateral zone separated from white dor- solateral line by a hiatus of pale ground colour. Distribution. Far northern interior of South- West Division, south to Galena and Gullewa. Extralimital in North-West Division (southern interior) and Eastern Division (southwest). Description. Snout-vent length (mm) : 54-91 (68.4). Length of appendages (% SVL) : tail 213-224 (218); foreleg 20-27 (23.8); hindleg 36- 46 (42.7). Nasals separated. Prefrontals in contact 'narrowly separated in one specimen). Supra- oculars 4, first 3 in contact with frontal. Supra- ciliaries 7 or 8 (7.3). Palpebrals 9-11 (10.4). Second loreal 1.2-1. 6 (1.34) times as wide as high. Ear lobules 4-6 (4.8). Nuchals 2 or 3 (2.7). Midbody scale rows 27-32 (29.7). Lamellae under fourth toe 19-23 (21.7). For further details of coloration, see original description. Material. South-West Division; Galena '17195-6. 19994-6, 25680-3); Gullewa (40848). Ctenotus alleni sp. nov. Holotype. R 33602 in Western Australian Museum, collected by Nicholas T. Allen on 17 Januai’y 1969 at 11 miles north of Galena, Western Australia, in 27° 41' S, 114° 39' E. Diagnosis. A member of the lesueurii group with reduced dorsal pattern, distinguishable from severus by its more numerous subdigital lamellae and nuchals, narrower black latero- dorsal stripe, and contact between white dorso- lateral line and black of upper lateral zone. Superficially similar to mimetes but readily dis- tinguishable by wide subdigital calli and by black upper lateral zone enclosing small white spots rather than large rufous rectangular blotches. Distribution. Far northern interior of South- West Division. Description (based on holotype and paratype). Snout-vent length (mm); 87, 78. Length of appendages (% SVL); tail 258, 264; foreleg 25, 24; hindleg 48. 48. Nasals separated. Prefrontals forming median suture. Supraoculars 4, first 3 in contact with frontal. Supraciliaries 7. Palpebrals 12-13. Second loreal 1.3-1. 9 times as wide as high. Upper labials 8. Ear lobules 4-6. Nuchals 4. Midbody scale rows 26-28. Lamellae under fourth toe 28-33, each with a moderately wide callus. Dorsal ground colour olive, darkest on head, palest on tail. Vertebral stripe reduced to a black line on nape. Narrow, clearcut, black laterodorsal stripe from brow to base of tail, about half a scale wide. White dorsolateral line from orbit to base of tail, on which it gradually becomes suffused with ground colour. Black upper lateral zone with one or two series of small white spots or short dashes. White mid- lateral stripe extending back nearly to end of tail, but barely extending forward to arm. Lower lateral zone blackish brown with one or two irregular series of short dashes. Paratype. South-West Division; 20 mi. NE of Yuna (26499). Ctenotus mimetes ctenotus mimetes Storr, 1969: 103. 12 miles east of Paynes Find, W.A. (D. A. Richards). Diagnosis. A member of the leonhardii group with long tail and legs, no vertebral stripe, well- developed white midlateral stripe, black latero- dorsal stripe narrow and not enclosing spots, and upper lateral zone consisting of alternating rectangular blotches of black and rufous. Distribution. Northern interior of South-West Division, south and west to Ajana, Carnamah, and Merredin. Extralimital in North-West and Eastern Divisions of Western Australia. Description. Snout-vent length (mm): 33-77 (63.3). Length of appendages (% SVL); tail 212-246 (228); foreleg 22-30 (25.2); hindleg 45-57 (51.3). Nasals separated. Prefrontals forming median suture. Supraoculars 4, first 3 in contact with frontal. Supraciliaries 7. Palpebrals 10-14 (12.2) . Second loreal 1.0-1. 3 (1.23) times as wide as high. Upper labials 8. Ear lobules 3-5 (4.0). Nuchals 3 or 4 (3.2). Midbody scale rows 26-28 (27.3). Lamellae under fourth toe 23- 28 (26.2), each compressed and bearing a dark obtuse keel or narrow callus. For coloration, see original description. Remarks. A specimen (26499) from the Yuna district was wrongly listed by Storr (1969: 104) under C. mimetes: it is actually a specimen of C. alleni. Material. South-West Division; 2 mi. W of Ajana (30321); Yuna (8303, 9027); Carnamah (407); Merredin (1265-6). Ctenotus uber uber Ctenotus uber Storr. 1969: 102. 22 miles southeast of Yalgoo, W.A. (P. J. Fuller). Diagnosis. A member of the leonhardii group, distinguishable from mimetes by presence of pale spots in dark laterodorsal stripe and absence of pale midlateral stripe. Distribution. Arid northeast of Eucla Division (Nullarbor Plain). Extralimital in North-West and Eastern Divisions of Western Australia. Description. Snout-vent length (mm); 44-61 (54.3) . Length of appendages (% SVL»; tail 157-172 (163); foreleg 22-27 (24.8); hindleg 45- 53 (47.0). Nasals normally separated (forming a median suture in one specimen). Prefrontals separated or in contact. Supraoculars 4, first 3 in contact with frontal. Supraciliaries 6 or 7. Palpebrals 10-13 (10.7). Second loreal 1.0-1. 4 (1.34) times as wide as high. Upper labials 8. Ear lobules 3-5 (3.6). Nuchals 3-5 (4.0). Midbody scale rows 30-32 (30.5). Lamellae under fourth toe 21-24 (21.8), each with an obtuse keel. Dorsal ground colour brown, darker and more olive on head, more coppery on tail and hind- legs. Vertebral stripe variably developed (at best a dark line from nape to base of tail; sometimes absent). Dark brown laterodorsal stripe enclosing an indistinct series of pale brown spots. Brownish-white dorsolateral line Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December. 1973. 89 sometimes broken into a series of spots. Upper lateral zone dark brown with 3 or 4 longitudinal series of pale brown dots. Lower lateral zone pale brown, spotted or variegated with brownish white. Material. Eucla Division: Seemore Downs (17284-5); 57 mi. NNE of Rawlinna (41592); Forrest (17286); 15 mi. S of Forrest (41594); 24 mi. S of Forrest (41593). Ctenotus atlas Ctenotus atlas Storr. 1969: 105. 17 miles south of Alley, W.A. (E. & H. Pianka). Diagnosis. A member of the taeniolatus group with 8 or 10 pale stripes. Further distinguish- able from impar by lack of pale vertebral stripe, nasals and prefrontals usually in contact, and upper labials usually 8. Distribution. Arid mallee-spinifex zone of Eucla Division. Extralimital in North-West and Eastern Divisions of Western Australia and in the interior of South Australia and of New South Wales. Description. For further details of colora- tion and scutellation. see original description. Material. Eucla Division: 11 mi. E of Fraser Range (30765-6). Ctenotus impar ctenotus impar Storr, 1969: 105. Tambellup, W.A. (F. R. Bradshaw). Diagnosis. A member of the taeniolatus group with 11 (regionally 12) pale lines and stripes, and nasals and prefrontals separated. Distribution. Southern half of South-West Division, north to the Gingin district, but absent from far southwest U.e. south of Busselton and west of the Fitzgerald). Extralimital in far southwest of Eastern Division. Description. Snout- vent length (mm): 30- 66 (51.8). Length of appendages (% SVL): tail 153-200 (176); foreleg 21-31 (25.7); hindleg 37- 50 (43.6). Nasals and prefrontals separated. Supra- oculars 4, first 3 (rarely 2) in contact with frontal. Supraciliaries 5-8 (mostly 7, mean 6.9). Palpebrals 8-13 (10.4). Second loreal 0.9-1. 5 (1.19) times as wide as high. Upper labials 7 (occasionally 6 or 8). Ear lobules 2-5 (3.7). Nuchals 2-4 (3.1). Midbodv scale rows 25-30 (27.7). Lamellae under fourth toe 18-24 (21.7), each with a dark obtuse keel. For coloration, see original description. Geographic variation. Over most of its range impar has eleven pale stripes including a broad whitish vertebral stripe. In the northwestern part of its range, i.e. from the Gingin district south to Pinjarra, the vertebral stripe is divided by a fine dark line, giving a total of 12 pale stripes. Material (additional to that listed in Storr 1969). South-West Division: Wanneroo (31450); Armadale (36676); Yunderup (37748-9); 17 and 18 mi. E of Pingrup (39853-7, 39874-6, 39931-2); Lake Magenta Reserve (39939-42); middle and lower Fitzgerald River (36960, 39004); 10 mi. N of Hopetoun (36248, 36287-9). Ctenotus lanceliiii ctenotus labillardieri lancelini Ford, 1969: 69. Lancelin Island. W.A. (J. Ford). Diagnosis. A large pale member of the labillardieri group with yellow legs streaked with blackish brown. Distribution. Only known from Lancelin Island, off west coast. Description. Snout-vent length (mm): 68-80 (75.7). Length of appendages (% SVL): tail 189-194 (192), foreleg 20-23 (20.8), hindleg 32-38 (33.4). Nasals separated (usually narrowly). Pre- frontals narrowly separated. Supraoculars 4, first 2 (3 in one specimen) in contact with frontal. Supraciliaries 7 or 8 (7.4). Palpebrals 11-13 (11.4). Second loreal 1.4- 1.8 (1.62) times as wide as high. Upper labials 8. Ear lobules 3 or 4 (3.4), subacute or obtuse, second the largest. Nuchals 3 or 4 (3.4). Midbody scale rows 24. Lamellae under fourth toe 22-24 (23.0), each bearing a wide callus. Dorsally pale brown, irregularly marked with dark brown (markings tending to orientate longi- tudinally). Poorly defined blackish brown laterodorsal stripe from temples to base of tail, enclosing an irregular series of pale spots. White dorsolateral line from orbit to base of tail. Blackish upper lateral zone enclosing an irregu- lar series of whitish spots and short dashes. White midlateral stripe from ear aperature to base of tail. Dark brown lower lateral zone variably marked with white, including in one specimen a ventrolateral stripe. Remarks. Lancelin Island is so small and close to the mainland that I find it hard to believe with Ford (1969: 74) that lancelini is only an insular representative of labillardieri. I think it more likely that lancelini is a northern representative of labillardieri, possibly surviving only on Lancelin Island. Material. South-West Division: Lancelin Island (18871-5). Ctenotus labillardieri Lyoosoma labillardieri Dumeril & Bibron 1839: 73. New Holland. Hinulia greyii Gray 1845: 76. Swan River. Diagnosis. A member of the labillardieri group with reddish legs heavily marked with black. Further distinguishable from gemmula, delli and catenifer by the white dorsolateral line con- tinuous ii.e. not broken into a series of short dashes) . Distribution. Humid coasts and near-coastal ranges of South-West and Eucla Divisions, north to the Swan River, east to the Thomas River, and inland to Mt Helena, Boddington, Rocky Gully and the Stirling Range; also on Eclipse and Bald Islands and the Archipelago of the Recherche. Description. Snout-vent length (mm): 25-76 (55.6). Length of appendages (% SVL) : tail 142-213 (184); foreleg 20-31 (25.0); hindleg 31-51 (39.4). Nasals separated (very rarely in short con- tact). Prefrontals separated (rarely in short contact). Supraoculars 4, first 2 (occasionally Journal of the Royal Society of Western Australia, Vol. 56. Part 3. December, 1973. 90 3) in contact with frontal. Supraciliaries 6-9 (7.0). Palpebrals 7-12 (9.6). Second loreal 1.0-2. 3 (1.54) times as wide as high. Upper labials 7 or 8 (7.4). Ear lobules 2-6 (3.8), obtuse or subacute. Nuchals 3 or 4 (occasionally 2 or 5, very rarely 6, mean 3.5). Midbody scale rows 24-31 (27.4). Lamellae under fourth toe 20-30 (24.5), each with a dark wide callus. Coloration in northern Darling Range.— Dor- sum brown or olive, without pattern except for narrow black laterodorsal stripe from brow to base of tail (on which it becomes increasingly broken). White dorsolateral line from brow to tail. Black upper lateral zone usually immacu- late, extending as a stripe forward through orbit nearly to tip of snout and backward nearly to end of tail. White midlateral stripe from upper lip to distal quarter of tail. Lower lateral zone blackish, enclosing a white ventrolateral stripe. Legs reddish brown, heavily blotched or streaked with black. Abdomen yellow. Coloration on Bald Island.— Dorsal ground colour olive. Black ragged-edged vertebral stripe usually present. Black laterodorsal stripe very wide, ragged edged, and enclosing a series of small white spots. White dorsolateral line not so straight as in northern specimens. Black upper lateral zone with 1-3 series of white dots. White midlateral stripe wavy, sometimes broken. Lower lateral zone blackish, irregularly spotted or variegated with white. Geographic variation. In most populations the coloration lies between the extremes de- scribed above. In the northern Darling Range the pattern is clear-cut and spotting is rare. Going south, the pattern gradually becomes ragged (especially in adults), black pigment in- creasingly invades the dorsum, the black stripes become dotted with white, and the white stripes become wavy or disjointed. For full discussion of variation, see Ford (1969). Material. South-West Division : Herne Hill (4908); Mt Helena (1978-9, 25583); Stoneville (27857-8); Greenmount (NMV D7615); Darling- ton (3340-1, 5985-6, 21262); Glen Forrest (627); Mundaring (8850, 14858. 21229, 26447); Mun- daring Weir (26476-7, 30678); 4-7 mi. E of Kalamunda (19492-4, 26816, 34337-8, 37475); Gooseberry Hill (4676); Kewdale (31946); Churchmans Brook (17981-2); Gosnells (4965); Bartons Mill (10262-4); Karragullen and 5 mi. SE (17987-8, 19118); Roleystone (17990); Ara- luen (31548); Canning Dam (12912); Wungong Brook (17985-6); Byford and 3 and 5 mi. E (17984, 19247, 19803); 4 mi. N of Jarrahdale (17983); Gleneagle (26291, 32470); Serpentine (17977-80); 6 mi. E of Keysbrook (17992-5); Banksiadale (6768, 34252-5); Dwellingup (39958-61, 39979, 40123) ; Boddington (10708- 10); Lake Clifton (17966-8); Collie and vicinity (17969, 19244-6, 22831); Margaret River (17953- 5); Mammoth Cave (66); Boranup (13417, 19833-4, 27850, 32465); Karridale (13446); Deepdene (12426. 12776, 36343); 5 mi. N of Augusta (37801-2, 37807) ; Augusta (30232) ; Cape Leeuwin (259, 263, 12783, 17956-65); Cal- gardup (7732); 7 mi. S of Nannup (27847); 10 mi. E of Nannup (21894); Carey Brook. Donnelly River (27848-9, 41039-49); 6 mi. NW of Manjimup (39725-9); Manjimup (5578-9. 5582, 8184, 19039-40, 37819); 7 mi. S of Man- jimup (17950-1); Pemberton (5580-1, 37968-9), Northcliffe (19489-91); Mt Chudalup (17970-3); Broke Inlet (26433); Nornalup (11039); Kent River (260-1'; Rocky Gully and 6 mi. W (17962, 41050-5); Pardelup (18004); Chorkerup (4514); Denmark, including Rudgyard, Valley of Giants, and Monkey Rock (19853-4, 22460-72, 24692-9, 24942-7, 31188-90, 31984, 37961, 37963-4); Albany (10946); King George Sound (NMV D2735-8); Eclipse Island (6803, 11278); Gidley Brook (27846); Lower Kalgan (19117); Two Peoples Bay (36354-7, 36383-4, 37834-5); Mt Many Peaks (17872-6); Waychinicup River (27845, 29699, 41148-55); Cheyne Beach (17926-47, 19976, 29700, 36010-3, 36015, 36018-9); Bald Island (17903-25, 19972-4, 40815-6); Porongorup Range (21805-8); Mt Toolbrunup (21809-15); Bluff Knoll (17974-5); Mt Bland (36882-5); Boonda- dup River (36918-9, 37187-94, 37199); east of Mt Barren (17976); Kundip (11005). Eucla Divi- sion; Oldfield River (30144); Dalyup River (17949); Dempster Head, Esperance (14948); 6 mi. NE of Esperance (SAM 5952); 5 mi N of Cape Le Grand (30802-3) ; Mt Le Grand (22529) ; mouth of Thomas River (36251-68); Figure of Eight Island (10119; NMV D8247); Boxer Island (NMV D9799, D9805); Thomas Island (10234); Mondrain Island (17901-2; NMV D8240); Middle Island (8684). Ctenotus gemmula sp. nov. Holotype. R 29640 in Western Australian Museum, collected by Magnus Peterson and Bryan J. Harty on 8 October 1967 at South Perth, Western Australia, in 32° 00' S, 115° 49' E. Diagnosis. A small member of the lahillardieri group, distinguishable from lahillardieri by its broken white dorsolateral stripe and narrower subdigital calli, and from delli by its 8 (rather than 7) upper labials and legs boldly blotched (not obscurely dotted) with black. Distribution. Southern half of South-West and Eucla Divisions, north to the Swan River and east to Israelite Bay. Description. Snout-vent length (mm); 31-58 (50.0). Length of appendages (% SVL); tail 163-203 (188); foreleg 21-29 (23.8); hindleg 34-47 (39.8). Nasals separated (rarely touching). Prefron- tals usually narrowly separated, occasionally in short contact. Supraoculars 4, first 2 in contact with frontal. Supraciliaries 6-8 (6.9). Palpe- brals 8-12 (10.4). Second loreal 1.1-1. 7 (1.34' times as wide as high. Upper labials 8. Ear lobules 2-5 (3.4), acute or subacute in adults, second usually largest. Nuchals 2-4 (3.3). Mid- body scale rows 24-28 (24.7). Lamellae under fourth toe 23-27 (24.9), each with a dark obtuse keel. Dorsally olive grey, unmarked except for narrow black laterodorsal line from brow to base of tail. A dorsolateral series of short white dashes from brow to base of tail. Black upper lateral Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 91 zone with or without a series of white spots, extending forward as a broken stripe through orbit nearly to tip of snout and backward on to proximal quarter of tail. White midlateral stripe wavy or broken into series of short dashes. Narrow dark grey lower lateral zone variably marked with white. Legs yellowish brown boldly marked with black and white. Paratypes. South-West Division: South Perth (29595-6, 29639-42, 29651, 29741, 29776-7, 30260, 34396, 37734, 37744-6, 40698, 40748, 40846, 41145, 41567); Rocky Gully (41056); 4 mi. W of Lake Cairlocup (41170); 11 mi. E of Greenshields Soak, Lake Magenta Reserve (39941). Eucla Division: 5 mi. W of Israelite Bay (31102). Ctenotus delli sp. nov. Holotype. R 37478 in Western Australian Museum, collected by John Dell on 29 April 1970 at 6 miles east of Kalamunda, Western Austra- lia, in 31° 57' S, 116° 08' E. Diagnosis. A small member of the labillardieri group with legs dark olive, finely peppered with black and white. Further distinguishable from labillardieri by broken white dorsolateral line, and from gemmula by upper labials usually 7 (rather than 8). Distribution. The northern Darling Range, from the Helena River to Mt Cooke. Description. Snout-vent length (mm): 34-63 <45.0). Length of appendages (% SVL) : tail 156- 179 (168); foreleg 24-32 (27.7); hindleg 35-45 (41.1). Nasals usually separated, occasionally in short contact. Prefrontals separated, usually widely. Supraoculars 4, first 2 in contact with frontal. Supraciliaries usually 7, occasionally 8. Palpe- brals 8-12 (9.7). Second loreal 1.3-1. 8 (1.52) times as wide as high. Upper labials usually 7, occasionally 8. Ear lobules 3-5 (4.0), obtuse, very small (especially compared to width of aperture). Nuchals 3 or 4 (3.3). Midbody scale rows 28-36 (29.9). Lamellae under fourth toe 23-29 (25.7), each with a dark, narrow callus. Dorsally dark olive brown, unmarked except for narrow indistinct black laterodorsal stripe from brow nearly to tail. White dorsolateral line from brow to tail (on which it becomes suffused with brown), broken, except anteriorly, into a series of short dashes. Black upper lateral zone from orbit nearly to end of tail, bearing one or more series of white dots. White mid- lateral stripe broken into series of short dashes, extending forward to below eye and indistinctly backward to about base of tail. Lower lateral zone dark grey flecked with white. Paratypes. South-West Division: 2 mi. SE of Darlington (22582) ; 4 mi. E of Kalamunda (39982); 6 mi. E of Kalamunda (19120, 34349, 36453, 37476-7, 37985); Karragullen (17989); Mt Cooke (40755). Ctenotus catenifer sp. nov. Holotype. R 21819 in Western Australia Mus- eum, collected by Julian R. Ford on 5 November 1963 on hill slopes above Cheyne Beach, Western Australia, in 34° 52' S, 118° 34' E. Diagnosis. A small member of the labillardieri group with legs olive brown longitudinally marked with series of black dots. Further dis- tinguishable from labillardieri by its broken white dorsolateral line and from delli and gem- mula by its heavily patterned back and lack of white midlateral stripe. Distribution. South coast of South-West Divi- sion, from West Cape Howe to Cheyne Beach, inland to Chorkerup; with a slightly different population 125 miles to northeast (near Ravens- thorpe) , Description. Snout-vent length (mm) : 33-58 (48.7). Length of appendages (% SVL): tail 168-191 (176); foreleg 21-29 (25.3); hindleg 33- 44 (38.1). Nasals separated. Prefrontals separated, usu- ally widely. Supraoculars 4, first 2 in contact with frontal. Supraciliaries 6-8 (7.2). Palpe- brals 8-10 (8.8). Second loreal 1. 1-1.4 (1.21) times as wide as high. Upper labials usually 7, occasionally 8. Ear lobules 3-6 (4.2), acute or subacute in adults, obtuse in juveniles, first or second usually largest. Nuchals 3 or 4 (3.5). Midbody scale rows 24-30 (26.7). Lamellae under fourth toe 21-25 (22.7), each with a dark obtuse keel or narrow callus. Dorsal ground colour olive grey flecked with black. An irregular black vertebral stripe oc- casionally present. Broad black laterodorsal stripe from brow to tail, ragged edged and bearing a series of pale dots. White dorsolateral line from brow to tail, more or less broken into a series of short dashes. Black upper lateral zone bearing a series of white dots. Dark grey lower lateral zone irregularly flecked with whitish. Remarks. For photograph and description of peculiar specimen from near Ravensthorpe, see Ford (1969). Paratypes. South-West Division: West Cape Howe (21823); Chorkerup (4251); Two Peoples Bay (36375, 36386, 40989-90) ; Waychinicup River (17877); Cheyne Beach (17935, 17942, 36013-5, 36318); Phillips River, 11 mi. W of Ravensthorpe (18005). Ctenotus schomburgkii Lygosoma schomburgkii Peters, 1863, Mber. Akad. Wiss. Berlin 1863: 231. Buchsfleld, S.A. (R. Schomburgk). Diagnosis. A member of the schomburkgii group, distinguishable from C. brooksi euclae by its dark dorsal colouration and two presuboculars. Distribution. Interior of South-West and Eucla Divisions. Extralimital in North-West and Eastern Divisions of Western Australia and in south of Northern Territory, northern South Australia, and western New South Wales. Description. Snout-vent length (mm): 30- 49 (40.7). Length of appendages (% SVL): tail 163-214 (188); foreleg 24-32 (27.0); hind- leg 43-61 (51.2). Nasals and prefrontals separated. Supraocu- lars 4 (rarely 5), with 3 (rarely 2) in contact with frontal. Supraciliaries mostly 7, occa- sionally 6, rarely 8. Palpebrals 7-10 (9.0). Sec- ond loreal 1. 2-2.1 (1.68) times as wide as high. Upper labials 7 (occasionally 6 or 8). Ear Journal of the Royal Society of Western Australia, Vol. 56, Part 3. December. 1973. 92 lobules 2-5 (3.2), short and obtuse, the first usually much the largest. Nuchals 2-5 (3.6). Midbody scale rows 24-29 (26.1). Lamellae under fourth toe 20-26 (23.0). Dorsally olive grey or olive brown, unmarked in south, variably marked with black in north (usually with a vertebral, dorsal, and laterodor- sal line). White dorsolateral line conspicuous only when margined above with black. Black upper lateral zone enclosing a series of pale vertical bars which may be very narrow or re- placed by one or two series of dots; extending as a black stripe forward through orbit to tip of snout and back to about middle of tail. Nar- row black lower lateral zone indented or spotted with white. Limbs boldly streaked with black. Geographic variation. In the far southwest of its range, i.e. north to the Darling Range and Merredin, schomhurgkii is completely unmarked dorsally. In the west there is a 75-mile gap be- tween our southernmost striped-back specimen (New Norcia) and our northernmost plain-back specimen (Bartons Mill). In the east there seems to be a gradual change from plain-back to striped-back; for example, two or four speci- mens from Holt Rock show a trace of the verte- bral stripe, and in our single specimen from Salmon Gums this stripe is still more strongly developed. I treat schomhurgkii binomially because I now regard pallescens of the Northern Territory as a full species. Material. South-West Division: 19 mi. E of Kalbarri (33539, 33553); 12 mi. N of Galena (33621-2); Caron (MCZ 33273); 7 mi. N of New Norcia (25673); Merredin (1267); Bartons Mill (10267); Boyagin Reserve (22516-7); East Pin- gelly (28315); Lake Magenta Reserve (39933, 39936-7, 40753); Holt Rock (34407, 34507, 36757, 37491); Lake Varley (29049). Eucla Division; Salmon Gums (30792); 12 mi. N of Seemore Downs (25862-3). Ctenotus brooksi euclae Ctenotus brooksi euclae Storr, 1971: 14. Eucla. W.A. (W. B. Alexander). Diagnosis. A member of the schomhurgkii group with whitish dorsal ground colour and a single presubocular . Distribution. Coastal sand dunes of Eucla Division, west to Eyre. Extralimital in far west- ern South Australia. Description and material. See Storr 1971: 14. References Dumeril, A. M. C. & G. Bibron (1839).— Erpetologie generate 5. Paris. Ford J. (1969).— Distribution and variation of the skink Ctenotus labillardieri (Gray) of southwestern Australia. J. Roy. Soc. W. Aust. 51: 68-75. Gray, J. E. (1845). — “Catalogue of the specimens of lizards in the collection of the British Museum”. London. Storr, G. M. (1969). — The genus Ctenotus (Lacertilia, Scincidae) in the Eastern Division of Western Australia. J. Roy. Soc. W. Aust. 51: 97-109. Storr, G. M. (1970). — The genus Ctenotus (Lacertilia, Scincidae) in the Northern Territory. R. Roy. Soc. W. Aust. 52: 97-108. Storr, G. M. (1971). — The genus Ctenotus (Lacertilia, Scincidae) in South Australia. Rec. S. Aust. Mus. 16(6): 1-15. Journal of the Royal Society of Western Australia. Vol. 56, Part 3, December, 1973. 93 Obituary Sir John Burton Cleland 1878-1971 Emeritus Professor Sir John Burton Cleland, Kt. (created 1964), C.B.E., M.D., Ch.M., P.R.A.C.P., the oldest surviving member of the Royal Society of Westei’n Australia, died in his sleep at Adelaide on August 11, 1971, at the age of 93 years. He was born at Adelaide on June 22, 1878. His father, Dr. W. L. Cleland, Colonial Surgeon of South Australia, put him up as a member of the local Field Naturalists’ Group as a youth, and he remained an active member for 80 years, actually contributing an article in its journal, the South Australian Naturalist, just prior to his death. He began his medical training in Adelaide but because of a strike by the honorary teaching staff in 1897 senior medical students were com- pelled to complete their courses elsewhere. He chose Sydney, because, as he related afterwards, he had heard that opportunities for “birding” were most favourable there! Subsequently he trained in England, at the London School of Tropical Medicine and the London Hospital (as a cancer research scholar), finally becoming Professor of Pathology at his old university, Adelaide (1920). Though he was distinguished in medicine he also became eminent in other fields — in ornithology (he became president of the Royal Australasian Ornithologists’ Union), in systematic botany and mycology, and in anthropology. He is credited as “the guiding spirit in the lavish anthropological expeditions to northern South Australia and the lower Northern Territory in the early and middle 1930s” which were organized by the University of Adelaide with the aid of Rockefeller funds. Though he was resident in this State only between 1906 and 1909, when he occupied the post of Government Pathologist and Bacterio- logist, he threw himself actively into the affairs of this Society’s predecessor — the West Aus- tralian Natural History Society. He was at once elected a member of the council, then presided over by the Anglican Archbishop of Perth, Dr. C. O. L. Riley, and delivered a lecture to the society on June 26, 1906, on “Demonstrations on some parasites of the blood.” This was followed by a paper, “An objection to the direct contin- uity of the germplasm, with a suggestion as to the part possibly played by hormones in heredity” (September 22, 1908), and a report on “A scientific trip to the north coast of Western Australia,” contributed to the same meeting. These were published in the journal of the Society. He was elected president of the Society for the year 1908-1909 but did not stay out the full term as he left the State and later took up the appointment of Principal Microbiologist in the New South Wales Department of Public Health. His presidential address, “The Australian fauna and flora and exotic invasions” (J. Nat. Hist. & Sc. Soc., 3 (1), 1910: 12-18) was read on his behalf by Bernard H. Woodward, director of the W.A‘. Museum. In this address Cleland was perhaps the first zoologist in this State to warn on the ecological and medical dangers of alien acclimatisations. One cannot but speculate as to how much richer the biological scene would have become in this State if Cleland had chosen to remain here. In Adelaide his fame grew and he was the recipient of many awards — the Verco Medal of the Royal Society of South Australia (1933). the Lord Medal of the Royal Society of Tas- mania (1939), the Australian Natural History Medallion (1952) and many fellowships. He was a member of many government-appointed boards and committees having to do with scientific and landscape-conservation matters. He was a conspicuous member of that band of “medical naturalists” who added so much to knowledge in Australia during the nineteenth and early twentieth centuries, and has been described as “the last of the old gentlemen- naturalists of the Charles Darwin tradition.’* Journal of the Royal Society of Western Australia. Vol. 56, Part 3. December. 1973. 94 OBITUARY Eileen Ruth Lathlain Johnson B.Sc. (Adelaide), M.A. (Toronto) 1896—1972 Mrs. E. R. L. Johnson, an Honorary Member of the Royal Society of Western Australia, died at her son’s home in Melbourne in August, 1972. Mrs. Johnson joined the Society as Miss E. R. L. Reed in 1922. An honours graduate in Botany from the University of Adelaide, she had come to the University of Western Australia to take up an appointment in the Biology Department. When the Botany Department became indepen- dent in 1929, Miss Reed was appointed Lecturer- in-Charge, a position she held until her marri- age towards the end of 1931. During this period Miss Reed was an active and enthusiastic mem- ber of the Society, serving on the Council and being Vice-President in 1932. In 1925-1926 an exchange lectureship was arranged between Miss Reed and Dr. Gertrude Wright of Toronto, who will also be remembered with affection by mem- bers of the Society of that period. In Toronto Miss Reed worked for and was granted the degree of Master of Arts of that University, After her marriage, Mrs. Johnson continued her membership of the Society, though with less time for Society activities. In 1954, after the death of her husband and with her children grown up, Mrs. Johnson rejoined the staff of the Botany Department of the University of Western Australia, and remained a valued and popular lecturer until her retirement and return to Ade- laide in 1965. In view of her scientific contribu- tions and her long and valued association with the Society she was elected an Honorary Member at that time. From 1928 Mrs. Johnson was associated with St. Catherine’s College, Univer- sity of Western Australia, and was Chairman of the College Council from 1956 to 1966; in recog- nition of this work she was awarded an M.B.E. in 1967. She also spent a three-year term on the University Senate, and was President of the Uni- versity Women’s Association. During her retirement, Mrs. Johnson was able to continue her botanical work at the Adelaide Herbarium through the courtesy of the Director, Dr. Eichler. The results of an early collecting trip to the Nullarbor Plain were published in the Society’s Journal, and a paper on the Western Australian species of Isoetes was almost ready for publication at the time of her death. Mrs. Johnson is survived by a daughter, Mrs. John Leslie, and a son. Dr. Andrew Johnson. She will be missed by a large circle of friends. Journal of the Royal Society of Western Australia, Vol. 56, Part 3, December, 1973. 95 INSTRUCTIONS TO AUTHORS Contributions to this Journal should be sent to The Honorary Editor Royal Society of Western Australia, Western Australian Museum, Perth. Papers are received only from, or by communication through, Members of the Society. The Council decides whether any contribution will be accepted for publication. All papers accepted must be read either in full or in abstract or be tabled at an ordinary meeting before publication. Papers should be accompanied by a table of contents, on a separate sheet, showing clearly the status of all headings; this will not necessarily be published. Authors should maintain a proper balance between length and substance, and papers longer than 10,000 words would need to be of exceptional importance to be considered for publication. The Abstract (which will probably be read more than any other part of the paper) should not be an expanded title, but should include the main substance of the paper in a con- densed form. Typescripts should be double-spaced on opaque white foolscap paper; the original and one carbon copy should be sent. All Tables, and captions for Figures, should be on separate sheets. Authors are advised to use recent issues of the Journal as a guide to the general format of their papers, includ- ing the preparation of references; journal titles in references may be given in full or may follow any widely used conventional system of abbreviation. Note that all illustrations are Figures, which are numbered in a single sequence. In composite Figures, made up of several photographs or diagrams, each of these should be designated by letter (e.g. Figure 13B). Illustrations should include all necessary lettering, and must be suitable for direct photo- graphic reproduction. To avoid unnecessary handling of the original illus- trations, which are usually best prepared between and 2 times the required size, authors are advised to supply extra prints already reduced. Additional printing costs, such as those for folding maps or colour blocks, will normally be charged to authors. It is the responsibility of authors to adhere to the International Rules of Botanical and Zoological Nomenclature. Palaeontological papers must follow the appropriate rules for zoology or botany, and all new stratigraphic names must have been previously approved by the Stratigraphic Nomenclature Com- mittee of the Geological Society of Australia. Thirty reprints are supplied to authors free of charge, up to a maximum of 60 for any one paper. Further reprints may be ordered at cost, provided that orders are submitted with the returned galley proofs. Authors are solely responsible for the accuracy of all information in their papers, and for any opinion they express. Journal of the Royal Society of Western Australia Volume 56 1973 Part 3 Contents 7. Analyses of Western Australian iron meteorites. By J. R. de Laeter. 8. Eucalyptus forrestiana subsp. dolichorhyncha, a new taxon from Western Australia. By M. I. H. Brooker. 9. The ecology and distribution of Eucalyptus forrestiana Diels. By J. S. Beard. 10. A progeny trial to obtain evidence of hybridity in two taxa of Eucalyptus. By J. S. Beard. 11. Harris Sandstone-Yindagindy Formation relationships and possible position of permo-carboniferous boundary, Carnarvon Basin, Western Australia, By J. F. Read, P. J, Alcock and P. Hoseman. 12. The genus Ctenotus (Lacertilia, Scincidae) in the South-West and Eucla Divisions of Western Australia. Obituary — Sir John Burton Cleland. Obituary — Eileen Ruth Lathlain Johnson. Editor: A. J. McComb The Royal Society of Western Australia, Western Australian Museum, Perth 321 10/10/73—625 WILLIAM C. BROWN, Governmenf Printer, Western Aus!ralia JOURNAL OF THE ROYAL SOCIETY OF WESTERN AUSTRALIA VOLUME 56 PART 4 DECEMBER, 1973 PRICE: TWO DOLLARS REGISTERED FOR POSTING AS A PERIODICAL-CATEGORY B THE ROYAL SOCIETY OF WESTERN AUSTRALIA PATRON Her Majesty the Queen COUNCIL 1973-1974 President Vice-Presidents Past President Joint Hon. Secretaries Hon. Treasurer Hon. Librarian Hon. Editor A. F. Trendall, B.Sc., Ph.D., A.R.C.S., F.G.S. , G. A. Bottomley, B.Sc., Ph.D. B. E. Balme, D.Sc. R. M. Berndt, M.A., Dip.Anth., Ph.D., F.R.A.I., F.F.A.A.A. , M. Perry, B.Sc. (Agric.) (Hons.) G. Perry, B.Sc. (Hons.) .... S. J. Curry, M.A. .... A. Neumann, B.A. . .. A. J. McComb, M.Sc., Ph.D. S. J. Hallam, M.A. L. J. Peet, B.Sc., F.G.S. P. G. Quilty, B.Sc. (Hons.), Ph.D. J. A. Springett, B.Sc., Ph.D. G. M. Storr, B.Sc., Ph.D. J. C. Taylor, B.Sc., Ph.D., A.R.C.S. P. G. Wilson. M.Sc. P. R. Wycherley, O.B.E., B.Sc., Ph.D., F.L.S. 13. — The origin of amphibolite and basic granulite bands in Precambrian gneisses of the south coast of Western Australia by N. C. N. Stephenson^ Manuscript received 22 August 1972; accepted 20 February 1973. Abstract. Evidence provided by chemical variation trends and by field associations suggests that amphi- bolite and basic granulite bands in the Pre- cambrian gneisses of the south coast of West- ern Australia have probably been derived from extrusive or intrusive basic igneous rocks, rather than calcareous or dolomitic shales. Introduction Precambrian gneisses with roughly east-west tectonic trends outcrop along the south coast of Western Australia between Point D’Entre- casteaux and Israelite Bay. These rocks are intruded by numerous syn- and late-kinematic granitic plutons also of Precambrian age (Turek and Stephenson 1966), and form part of the Albany-Esperance Block, an arcuate belt wrapped around the southern and southeastern margins of the Archaean Yilgarn Block. The gneisses are predominantly granitic in character, with intercalated basic, pelitic, and minor calc- silicate bands. Migmatitic types are common. The metamorphic grade varies from upper amphibolite to lower granulite facies. Granulite facies rocks from the Fraser Range, at the northeastern end of the Albany-Esperance Block, gave a Rb-Sr age of 1330± 15 m.y. (Compston and Arriens 1968, p. 585) and this could be the age of the main metamorphism throughout the Block. Bands and lenses of amphibolite (hornblende- plagioclase rocks) and basic granulite (horn- blende-plagioclase-pyroxene rocks) are common throughout the south coast gneisses. It is generally recognised that amphibolite and basic granulite may result from the meta- morphism of various rock types including extrusive and intrusive basic igneous rocks, basic tuffs, and calcareous or dolomitic shales. The problem of distinguishing between meta- igneous and metasedimentary amphibolites and basic granulites has concerned geologists for m.any years. The only published study of the amphibolites and basic granulites of the south coast is by Clarke et al. (1954) who concluded (p. 54), from a comparison of major element compositions, that they were probably derived from basic igneous rocks. Since publication of this reconnaissance study it has become widely accepted that similarity in major element com- position does not necessarily prove derivation from a basic igneous parent and various other approaches to the problem have been tried (e.g., Walker et al. I960; Leake 1964; Shaw and Kudo 1965). It is the purpose of this paper to re- ^Geology Department, University of New England, Armi- dale. New South Wales, 2351. Formerly Geology Depart- ment, University of Western Australia, Nedlands, Western Australia. examine the problem of the origin of the amphibolite and basic granulite bands in the south coast gneisses in the light of recent information. Methods Seven samples of amphibolite and basic granulate have been analysed for 13 major and 13 trace elements. SiO- 2 , ALOa, total Fe as Fe-iOa, MgO, CaO, K 2 O, Ti02, P 2 O-., and MnO were determined by X-ray fluorescence spectro- scopy using the techniques for sample prepara- tion and matrix corrections described by Norrish and Hutton (1964, 1969). Seven synthetic standards were used for calibration, and standard rocks G-2, GSP-1, AGV-1, BCR-1, PCC-1, and DTS-1 were analysed to check the accuracy of the method. FeO was determined by titration against standai’d ceric sulphate solution, and Na20 by flame photometry. Total water (HoO ' + H 2 O") was estimated as the ignition loss, with a correction (based on the residual FeO contents of the ignited samples) applied for oxidation of FeO. H 2 O- v/as deter- mined by drying the samples at 110°C. The precision of the analyses at the concentration levels encountered was, with 95% confidence, about ±1% of the amount present for Si 02 , AhO-.s, total Fe as Fe20.^, FeO, CaO, K 2 O, and T 1 O 2 , and ±5% of the amount present for MgO, Na20, P 2 O 5 , MnO, and H 2 O ^ Co, Rb, Sr, Y, Zr, Ba, La, Pb, and Th were determined by X-ray fluorescence spectroscopy using pressed powder samples (Norrish and Hutton 1964; Norrish and Chappell 1967, p. 205). Standard rocks W-1, G-2, GSP-1. AGV-1, BCR-1, PCC-1, and DTS-1 were used for cali- bration, and the procedure of Hower (1959) was used for matrix corrections. Interferences by FeK/^^l in Co analyses, RbK^ai,^ in Y analyses, and SrK; 6 i ,3 in Zr analyses were compensated by mathematical corrections based on measure- ments on samples spiked with known concen- trations of the interfering element (Leake et al. 1969, p. 63). Li, Ni, Cu, and Zn were deter- mined by atomic absorption spectroscopy. The precision of the analyses at the concentration levels encountered was, with 95% confidence, roughly d=5% of the reported concentration for Ni, Zn, Rb, Sr, Zr, and Ba; ±10% for Li, Co, and Y; ±15% for Cu, La, and Pb. Th was not detected. Modes were determined by counting 1000 points over a sam.ple area of about 500 mm^ Plagioclase compositions were estimated from measurements of the extinction angle X^ A 010 in sections perpendicular to x (Deer et al. 1963. Vol. 4, Fig. 55), orthopyroxene compositions Journal of the Royal Society of Western Australia. Vol. 56. Part 4, December. 1973. 97 s s <5 e « s a Si ot — '~H :£ O' 5 S-S a o ss « a o S '® a «c •NNd ^ II « « !2 5ft Si r. ^ «c a S « 1 1 \ CC I 1 ^ X CO 01X0100T-H01-+0 ! r-.U5XXl-OX>-H^ :01OO rH r 1 ! o] I ! " 1 w t>- OlOiOCO'MOS'+OX cooii-i -HiDoiXCOOOlOO :-HC>0 ID -H I'- X X I'- o T'. iD -H ]•» ]•» 01C005'>+r-i.--005 — MiDlDO] CilDr-lOlDX^'— l-HO— <00 o •1^ rH ^ o 1 1 -H 1 I -f 35 1-~ o X iD oi iD r- 1-1 w OCit-^XOXXOOli— icOi— 1-^ 35 01 -hD1XX'Mi-hOOOOO © i >o -t -- 1-1 © I o -HC035®cDI'-O35Xr--^f>-X © I-' cO O X iD iD X t 1 — 1 01 iD 0-1 X a iDXi— (XOlCC-'fi-HOOrHOO O a o 1 tc 1 -fOlXO-IXXCOiDX -+3:-t X -H 1- 35 oj lO 01 X CO 1-H 0 1 I Xti-HO-lcOiDOlOO :>-hCO o o , c0-^C0X’+01t'~c0-t‘'^XXX uD j X t'-3i'-it'-XOI35'*‘C0035 01'H CCr--XXc0O01f~iOCiOOO o 1— ' 1-H o * * 35uO»Dr^053505.-i01XI'«0-^‘ 35Xi-H05i-HXI^3535 01XX-' t CO [ OSXOIIDCOOXOCOOOO © 1 © ! 01 X ID CO X X -t D1 X 01 O 3i © ,—1 035t^-Hi-(wCi-tI-'C-^0101 •-t* C> 01 01 01 CO O Dl O 1 -H O 'H o o © io ^ ^ ^ © 1 ccXr-'COcool'^-tC’COi-HXX © ; X I'. X 35 1-H ID 01 CO •i' 1-1 ID >D 01 i-HC0-+- « M -t -H O cs ift t- rl M X 5-1 : O O "N X -f : cc lO rj O t iOCi'-C CO CO CO O'] -t lO CO CO X lO O: r>. X X X X 01 CO 05 O X CO UO 1— XXt-~01C005>-H--i\/ rH 01 ■-• V XcOcOt'--^X-t*X — -^OlOO i-t lO t-i r~ CO CO I'- CO o i-^ oi \ / » -t -- V ^-tOOi'^h'.OcOiOOCCOOiO 0JXXiO«Xt>.«XOl«Ol \ / T-H i-( IfS t-i 05 V XOI^XXCOX^t-i rMiOX-^OXOlXX rH O] X « liO V i-HcoxxC'^'^X'Ccoscvi'-in 01«X XOC0'^--t05Xr^\/ --I 01 rH uo V COOliCOOCOf'^iiOCOi-iXi-iiO I-HXX'-HOXXO] X05XX \/ T-H t-H Tt< Tf V »D 01 t— OI © iD X © X _©©X-t© ©©1--X : -H t- X X -C : iD 01 Ol 1-H l-H 01 1-H 1—1 r-i Ol 01 X © © X © © X © X © © X © Ol 1-H Ol : © -t X Ol 1-H : X Ol -t* 1-H 01 O © X I- t- X 1-H © © rH — © X -f X © X © X 1-H 1-H X © 1 -H X r-( : I- © : Ol rH rH r— © © ^ © X -H © •*+* X -H X © X X l-» I-» X : 1-H X © 1-H Ol '.re ^1 o ^ rH ©©©©© oioi-t*-^ ©o»xoi-t xx©x :Xl—©-+© :iDi-H-H'© 1-H X rH rH ©XrH©rH ©l--Xt-- ©1-HlD©© -^*©1—1© : lO C5 O ^ 1-t CO O ^ ^ ^ © Ol © © © © Ol © X © © X © © © © © Tt» Ol Ol r- -* rH 01 : CO ; Ol CO O pH 01 X X X © 1-H tJ< X © © © © rH hH Ol rt X Ol X © '•+' Ol © r- Ol Ol : ^ ^ : O'! ^ r- X Ol X © © CO CO CO O '*0 Ol X W CC pH ^ : X rf X Ol Ol : O pH rt O pH 1-^ © © r- t- 1-H © 1- I- 01 rH Ol ® t- © 1-H O © t- : rf l—l X rH 01 Ol : pH Ol 1*-* 01 Tj- O Ol © X rf rfi 1-H X © -H © X 01 1-H © © © © © © 1-H © © rH 01 Ol : X : 1-H 01 O ” 5 s i X > Journal cf tho Royal Society of Western Australia, Vol. 56. Part 4, December, 1973. 93 Niggli Values o CO O (M -H* 05 - t > (M O (N n OJ (M CO lO CO '-f X o 04 OJ X © X »ft >— I >ft © © © o 04 X 04 «-H « ©-H © ■-t -- t < X --H © © © »-H © 04 © ••t 1-1 1.0 © © © oi © © © 04 04 ^ 1-1 !>• X © © © -+<0I X © OJ'Tf Ol f- © -rf X I" © 04 © © © Ol 04 © © « X -+< © X 04 *-1 © 04 X Ol T-i © © X : Ol © X 04 : -iJii-tOl ii-b © © © 04 ; --i © O CO X 04 : 4^ -w l> © t- X X X : O CO i ; is ; 04 © X X © -Tt* © © © : X 04 X tr tr © X X -t X © 04 04 04 © © ,-H © 04 1-H ©® © to C .2 (U e if o a O © « - f « X © X © 01 © X © © X I> : © X ; © © ^ C rt c «S — CJ ■ ■ ■ c. OJ X n tC ,-‘-H o Uc CO ?r ^ ^ jr rt o o ^ ^ >c c o Eco rH p © P - :n ct o ; c oS •I « "S ! c > a > ' o © ' " O - f . Ci = I o I »-< 0/ ■ s ^ X p ^ 2i ' / o8 ^'5 « c^ii '=■ «3 - « >. "rt 1 "y L © *1 X Journal of the Royal Society of Western Australia, Vol. 56, Part 4, December, 1073. 99 from measurements of refractive index 7 (Deei et al. 1963, Vol. 2, Fig. 10), and clinopyroxene compositions from measurements of 2Vz and re- fractive index ^ (Deer et al. 1963, Vol. 2, Fig. 41). Description of the Amphibolites and Basic Granulites The amphibolite and basic granulite bands range in thickness from less than one metre up to several tens of metres, and are normally conformable with foliation and lithological banding in the surrounding gneisses. One notable exception occurs near Doubtful Island Bay where a discordant band of basic granulite clearly of intrusive igneous origin is well exposed. However, the great majority of basic bands could be interpreted as sedimentary or tuffaceous layers, or as igneous flows or sills. Most of the basic bands are broadly homo- geneous, but some are thinly layered suggesting either metamorphic differentiation or original (sedimentary?) layering. Recrystallisation dur- ing metamorphism has completely destroyed original textures, producing granoblastic or foliated fabrics. Hence the parent rock must be deduced from the pi’esent chemical composition. In this approach it is necessary to make the assumption, though difficult to substantiate, that metamorphism was isochemical (except for volatiles) . The 11 rock analyses on which this study is based comprise 4 previously presented by Clarke et al. (1954, Table IV) and 7 new analyses car- ried out by the author. The 11 analysed samples are from 5 widely separated localities (Torbay, Albany, Cape Riche, lower Pallinup River, and Point Irby; see Fig. 1) spanning a total distance of 130 km. The samples include representatives of the amphibolite facies and the amphibolite- granulite transitional facies, and cover all the more common variations in mineral assemblage. Samples showing evidence of metasomatism (usually granitisation; less commonly carbona- tion or scapolitisation) or of significant retro- gression have been carefully excluded from the study. Chemical analyses, C.I.P.W. norms, Niggli values, and modes for the samples studied are listed in Table 1. There appear to be no con- sistent differences in major element composi- tion, except perhaps water content, between the amphibolite and basic granulite samples. Hence the basic granulites are believed to be the higher-grade equivalent of the amphibolites. This belief is supported by the fact that these two rock types are generally not associated in the field. Discussion It is evident in Table 1 that the analysed samples are similar in major and trace element composition to average tholeiite and alkali basalt (Manson 1967; Prinz 1967). Following the system of Yoder and Tilley (1962, p. 352), four of the analysed samples may be classified as oversatur- ated tholeiite (normative quartz and hypers- thene), four as undersaturated tholeiite (norm- ative hypersthene and olivine), and three as alkali basalt (normative olivine and nepheline). However, bulk chemical similarity to basic igneous rocks does not prove an igneous origin. Most authors accept that mixtures of carbonate and shale in appropriate proportions may, after high-grade metamorphism with loss of volatiles, closely resemble basic igneous rocks in bulk composition. Attempts (e.g., by Evans and Leake 1960; Walker et al. 1960) to distinguish between meta-igneous and metasedimentary Figure 1. — Locality map. Journal of the Royal Society of Western Australia, Vol. 56, Part 4, December, 1973. 100 amphibolites on the basis of abundance levels of certain trace elements have not produced com- pletely diagnostic criteria because observed ranges of trace element concentrations in basalts and in carbonate-shale mixtures greatly over- lap. mg Figure 2— Niggli c-mg plot for south coast amphi- bolites (circles) and basic granulites (dots) (after Leake 1964). 100 mg Figure 3. — lOOmg-c-(al-alk) plot for south coast amphi- bolies (circles) and basic granulites (dots) (after Leake 1964). However, Leake (1964) has suggested that chemical variation trends in suites of samples, rather than absolute abundance levels, may be significant, and this approach is followed here. Perhaps 11 analyses are insufficient to reliably define variation trends, but nevertheless some interesting results emerge. Niggli c-mg and lOOmg-c- (al-alk) plots of the south coast am- phibolites and basic granulites (Figs. 2 and 3, after Leake 1964) closely follow igneous trends, and Ni and Co concentrations appear to in- crease with increasing Niggli mg values (Figs. 4 and 5) as expected in igneous rocks, but not in carbonate-shale mixtures (Leake 1964). Therefore derivation of the amphibolite and basic granulite bands in the south coast gneisses from extrusive or intrusive basic igneous rocks, or possibly basic tuffs, seems likely. e cL CL mg Figure 4. — Ni-mg plot for south coast amphibolites (circles) and basic granulites (dots). 40 “ • • 0 0 E o d. d 20 - o u A 1 1 1 0-3 0-4 0-5 0-6 0-7 mg Figure 5. — Co-mg plot for south coast amphibolites (circles) and basic granulites (dots). However, Orville (1969) has demonstrated that rocks composed largely of hornblende and plagioclase, with relatively small amounts of other minerals, must plot in a limited field along the hornblende-plagioclase tie-line on the ACF diagram, and hence approximate the field of basic igneous rocks (see Fig. 6, after Orville 1969). A consideration of the chemical composi- tions of hornblende and plagioclase leads to the conclusion that most amphibolites must also plot within the field of basic igneous rocks on c-mg and lOOmg-c- (al-alk) diagrams. Orville (1969) has also shown that hornblende-plagio- clase rocks can, because of their restricted bulk compositions, represent only a very limited composition range in carbonate-shale mixtures (see Fig. 6). Therefore it can be argued that hornblende-plagioclase rocks, regardless of their origin, might be expected to approximate basic igneous rocks in major element composi- tion, and might be expected to show igneous, rather than sedimentary, trends on variation diagrams based on major element composition. Because mixtures of carbonate and shale are unlikely to be confined to the restricted range of proportions which yields amphibolite on metamorphism (shale with 20-40% dolomite: see Orville 1969, Fig. 5), it follows that amphi- bolites derived from carbonate-shale mixtures are likely to be intimately associated in the field with calc-silicate rocks (e.g., plagioclase-diop- side grossular rocks) and pelitic gneisses (e.g., Journal of the Royal Society of Western Australia, Vol. 56, Part 4, December, 1973. 101 A C Diop. F Figure 6. — ACF plot for south coast amphibolites (circles) and basic granulites (dots) (after Orville 1969). quartz - plagioclase - garnet - biotite - cordier- ite gneisses), and transitional assemblages. Associations of this type, though not unknown, are uncommon in the south coast gneisses; the amphibolite and basic granulite layers are normally interbanded with granitic gneiss. Furthermore, lithologies representing bulk compositions transitional between amphibolite and pelitic gneiss and between amphibolite and calc-silicate rock are rare in the region in question, so there does not appear to be a con- tinuous variation from calc-silicate to pelitic rock through amphibolite and basic granulite. The amphibolite and basic granulite appear to represent a distinctly defined rock type rather than an intermediate member of a carbonate- shale-derived series. The lack of close associa- tion in the field between calc-silicate rocks, amphibolites, and pelitic gneisses also rules out Orville’s (1969) model for the genesis of metasedimentary amphibolites involving chem- ical reaction between adjacent incompatible carbonate and pelitic assemblages. Conclusions It is concluded that amphibolite and basic granulite bands in the gneissic complex of the south coast of Western Australia have been de- rived mainly from extrusive or intrusive basic igneous rocks rather than calcareous or lolo- mitic shales because: (i) they follow igneous trends on c-mg, lOOmg-c-(al-alk), Ni-mg, and Co-mg variation diagrams: (ii) they are not usually closely associated in the field with calc-silicate and pel- itic gneisses. Derivation from an igneous parent is further supported by the occurrence at Doubtful Island Bay of at least one baud of basic granulite showing discordant, cleai'ly intrusive relations with surrounding rocks. However, the possibility that the parent rocks were basic tuffs cannot be ruled out in the absence of relict textures, but this is perhaps unlikely in view of the gen- eral scarcity of basic tuffs in the geologic column. Acknowledgements. — This paper is an incidental pro- duct of a Ph.D. project carried out under the super- vision of Professor R. T. Prider and Mr J. G. Kay in the Geology Department, University of Western Aus- tralia. The major element X-ray fluorescence analyses were done in the Physics Department, Western Aus- tralian Institute of Technology in collaboration with Messrs G. Kerrigan and W. Thomas with the permis- sion of Dr J. de Laeter. The remaining chemical and X-ray fluorescence analytical work was done with the assist- ance of Mr P. E. Bannister. Computer programs writ- ten by Messrs Bannister, S. M. Bye. and D. A. C. Williams, and Drs J. Hallberg and W. R. O’Beirne were used. Mr B. Whan drafted the diagrams, and Mr R. Large critically reviewed the manuscript. References Clarke, E. de C., Phillipps, H. T., and Prider, R. T. (1954). — The Pre-Cambrian geology of part of the south coast of Western Australia. J. Roy Soc. W. Aust, 38: 1-64. Compston, W. and Arriens, P. A. (1968).— The Pre- cambrian geochronology of Australia. Canad. J. Earth Sci. 5: 561-583. Deer. W. A., Howie. R. A., and Zussman, J. (1963).— "Rock-Forming Minerals”, Vols. 1-5. Long- mans, London. Evans. B. W. and Leake, B. E. (I960).— The composi- tion and origin of the striped amphibolites of Connemara, Ireland. J. Petrol. 1: 337- 363. Hower, J. (1959). — Matrix corrections in the X-ray spec- trographic trace element analysis of rocks and minerals. Amer. Min. 44: 19-32. Leake, B. E. (1964). — The chemical distinction between ortho- and para-amphibolites. J. Petrol. 5: 238-254. Leake. B. E., Hendry, G. L., Kemp. A., Plant, A. G., Harvey, P. K., Wilson, J. R.. Coats, J. S. Aucott, J. W., Liinel, T., and Howarth, R. J. (1969). — The chemical analysis of rock powders by automatic X-ray fluorescence. Chem. Geol. 5: 7-86. Manson, V. (1967). — Geochemistry of basaltic rocks: major elements. In "Basalts”. (H. H. Hess and A. Poldervaat, eds.) 215-269. John Wiley, New York. Norrish, K. and Chappell. B. W. (1967). — X-ray fluores- cence spectrography. In "Physical Methods in Determinative Mineralogy”. (J. Zussman, ed.), 161-214. Academic Press, London. Norrish, K. and Hutton, J. T. (1964). — Preparation of samples for analysis by X-ray fluorescent spectrography. Divl. Rept. Div. Soils CSIRO 3/64. Norrish, K. and Hutton, J. T. (1969). — An accurate X-ray spectrographic method for the analy- sis of a wide range of geological samples. Geochim. Cosmochim. Acta 33: 431-453. Orville, P. M. (1969). — A model for metamorphic dif- ferentiation origin of thin-layered amphi- bolites. Amer. J. Sci. 267 : 64-86. Prinz, M. (1967). — Geochemistry of basaltic rocks: trace elements. In "Basalts”. (H. H. Hess and A. Poldervaart eds.) 271-323. John Wiley: New York. Shaw, D. M. and Kudo, A. M. (1965). — A test of the discriminant function in the amphibolite problem. Min. Mag. 34: 423-435. Turek, A. and Stephenson, N. C. N. (1966). — The radio- metric age of the Albany Granite and the Stirling Range Beds, south-west Australia. J. Geol. Soc. Aust. 13: 449-456. Walker, K. R., Joplin, G. A., Lovering, J. F., and Green, R. (1960). — Metamorphic and metasomatic convergence of basic igneous rocks and lime-magnesia sediments of the Precambrian of northwestern Queensland. J. Geol. Soc. Aust. 6; 149-178. Yoder, H. S. and Tilley, C. E. (1962). — Origin of basalt magmas : an experimental study of natural and synthetic rock systems. J. Petrol. 3: 342-352. Journal of the Royal Society of Western Austral.a, Vcl. 5G, Part 4, December, 1973. 102 14. — The petrology of the Mt Gardner Adamellite, near Albany, Western Australia by N. C. N. Stephenson^ Manuscript received 20 February 1973; accepted 17 July 1973. Abstract. The Mt Gardner Adamellite is emplaced in Precambrian amphibolite facies gneisses of the Albany-Esperance Block, about 30 km east of Albany, Western Australia. It is a composite pluton composed mainly of coarse-grained, por- phyritic adamellite intruded by dykes of micro- adamellite and by very minor pegmatite and quartz veins. Field evidence strongly suggests that the pluton was intrusively emplaced as a magma or crystal mush. Chemical data are consistent with derivation of the microadamel- lite dykes from the porphyritic adamellite magma by concentration of residual liquids, perhaps by filter pressing, during the later stages of crystal- lisation. The pegmatite and quartz veins are possibly the products of more advanced frac- tionation. The magma probably originated by anatexis of crustal rock below the present level of emplacement of the pluton during the orogeny responsible for regional metamorphism of the country rocks. Introduction Mt Gardner Adamellite is the name proposed for a granitic pluton situated at the southern end of Two People Bay on the south coast of Western Australia, about 30 km east of Albany 1 Geology Department, University of New England, Armidale, New South Wales 2351. Formerly Geology Department, University of Westrn Australia, Nedlands, Western Australia. (Figure 1). It is named after Mt Gardner, a prominent topographic expression of the pluton, located at 35° GO'S latitude and 118° lO'E longi- tude. The Mt Gardner Adamellite is one of a number of granitic plutons emplaced in the Precambrian high-grade metamorphic rocks of the Albany-Esperance Block, and has not been described previously. The purpose of this paper is to discuss the origin of this pluton in the light of new field, petrographic, and chemical data. A geological map is attached (Figure 2). The chemical and modal analytical methods used in this study have been summarised else- where (Stephenson 1973). Mesonorms were calculated using the method of Barth (1962), and plagioclase compositions were estimated from measurements of the extinction angle X' A 010 in sections perpendicular to x (Deer et al. 1963, Fig. 55). Sample numbers refer to the collection of the Geology Department, Uni- versity of Western Australia, Country Rocks Introduction The basement rocks of the south coast part of the Albany-Esperance Block are Precambrian Figure 1. — Map of the Albany district showing the location of the Mt Gardner Adamellite. Journal of the Royal Society of Western Australia, Vol. 56, Part 4, December, 1973. 103 gneisses with roughly east-west tectonic trends. These gneisses are predominantly granitic in composition, with intercalated metasedimentary and metabasite bands. Migmatites are common. The metamorphic grade varies from upper amphibolite to lower granulite facies. Granulite facies rocks from the Fraser Range, at the northeastern end of the Albany-Esperance Block, gave a Rb-Sr age of 1330±15 m.y. (Comp- ston and Arriens 1968) and this could be the age of the main metamorphism throughout the block. The gneissic country rocks surrounding the Mt Gardner Adamellite have been largely ob- scured by Recent unconsolidated dune sands and by the Southern Ocean. However, they appear to be largely granitic in character, with occa- sional thin dioritic bands and lenses. Petrography 1. The granitic gneiss is poorly foliated, equigranular fine- to medium-grained, and highly leucocratic. Quartz, oligoclase, and microcline in sub -equal amounts are the major constituents. Biotite is the main accessory and a few samples contain a little green hornblende. Minor accessories include magnetite, sphene, epidote, muscovite, apatite, allanite, and zircon. The texture is predominantly granoblastic, com- monly modified by the occurrence of biotite either in small streaky aggregates or in well oriented disseminated flakes. In some samples, especially those collected close to the margin of the Mt Gardner Adamellite, microcline tends to corrode and enclose other minerals, suggest- ing metasomatic growth. Chemical analyses and modes of four representative samples are pre- sented in Table 1. 2. The dioritic gneiss is a fine- to mediuni- grained, equigranular, dark grey, mesocratic rock composed mainly of andesine, green horn- blende, biotite, and quartz. Microcline is locally present. Minor accessories include magnetite, sphene, apatite, allanite, and zircon. The texture is usually weakly foliated due to the preferred orientation of hornblende and biotite. The chemical analysis and mode of a representative sample is presented in Table 1. Metamorphic facies The mineral assemblages most common in the gneisses around the Mt Gardner Adamellite may be summarised as follows: 1. Granitic gneiss — Quartz-plagioclase-microcline-biotite± hornblende. 2. Dioritic geneiss — Quartz-plagioclase-biotite-hornblende± microcline. These assemblages are characteristic of the amphibolite metamorphic facies (Turner 1968). Journal of the Royal Society of Western Australia, Vol. 56. Part 4, December, 1973. 104 Table 1 Chemical analyses and modes of granitic and dioritic. gneisses from Mt Gardner Granitic gneiss Dioritic gneiss 50578 56585 56592 56593 54513 SiO, 72-26 76-12 77-51 76-15 56-20 Al -,03 12-93 . . 12-21 11-83 12-82 14-98 FeoOa 2-59 1-22 0-47 1-00 2-35 Feb 1-53 1-00 0-70 1-01 5-47 MgO 0-52 0-23 0-13 0-42 5-19 CaO 2-03 1-21 0-54 1-30 6-68 Na .,0 3-57 3-19 3-25 3-38 3-00 K.b 3-24 4-30 5-22 4-17 3-29 H.,0- 0-36 0-21 0 - 33 0-22 0-99 H„0- Oil 0-13 0-10 0-16 0-14 Tib, 0-34 0-16 0-06 0-15 1-29 P,05 0-09 0-02 0-01 0-04 0-69 MnO 0-06 0-05 0-02 0-03 0-15 Total 99-63 100-05 100-17 100-85 100-42 Trace Elements (p.p.m.) Li 16 10 22 10 18 Co 4 <3 “3 3 32 Xi 26 20 19 24 87 Cu 6 5 4 4 49 Zn 78 54 44 51 101 Rb 137 172 267 143 177 Sr 240 101 22 108 593 Y 36 42 134 35 42 Zr 259 138 97 154 .303 Ba 1501 1141 277 1696 1667 La 67 59 37 48 67 Pb 37 37 51 28 32 Th 12 12 27 20 22 Modes (vol. %) Quartz 34-1 35-5 38-4 35-5 10-4 K-feldspar 26-1 30-8 35-8 27-0 7-5 Plagioclase 33-7 31-5 24-0 34-8 37-8 Biotite 4-2 1-4 1-7 tr 14-9 Hornblende 2-0 27-7 Rest i-9 0-8 0-1 0-5 1-7 Plag. An% 29 19 13 18 34 Structure Determination of the structure in the gneisses around the Mt Gardner Adamellite is difficult be- cause of inadequate outcrop. Lithological band- ing and foliation in the gneiss appear to be mutually parallel. In isolated outcrops near the northern margin of the pluton these fea- tures strike roughly north-south and dip gently (20°-30°) west. This represents a marked local departure from the regional east-west strike in the south coast part of the Albany-Esperance Block. A weak mineral lineation, defined by pre- ferred orientation of elongate biotite flakes, plunges southwest at about 15°-25°. This is assumed to be a b-lineation. Mt Gardner Adamellite Field occurrence and facies The Mt Gardner Adamellite outcrops prom- inently over an area measuring 6 - 2 ' x 3i- km, but the actual dimensions of the pluton may be greater because the margins are almost com- pletely obscured by the Southern Ocean or by Recent aeolian sand. The pluton is topographi- cally expressed as a group of dome-shaped hills which rise steeply from the sea to a maximrun elevation of 400 m at Mt Gardner. Two main facies have been recognised: (i) Porphyritic adamellite (ii) Microadamellite. Porphyritic adamellite constitutes the bulk of the pluton. It is a fairly homogeneous, massive rock with abundant megacrysts of K-feldspar up to 4 X 4 x 2 cm in size set in a medium- grained allotriomorphic to hypidiomorphic granular groundmass. Microadamellite occurs as minor dykes up to a few metres wide within the porphyritic adamel- lite. It typically shows fine-grained, allotrio- morphic granular texture, but the local presence of anhedral megacrysts of K-feldspar up to 2 x 2 x 1 cm in size produces a seriate texture in places. Mineralogy Both facies of the Mt Gardner Adamellite are composed mainly of K-feldspar, plagioclase, and quartz in order of decreasing abundance, with biotite the main accessory. Minor accessories in- clude magnetite, sphene, muscovite, metamict, allanite, epidote, and zircon. Chemical analyses, mesonorms and modes of representative samples are presented in Table 2. The K-feldspar is microcline, occurring as an- hedral megacrysts and in the groundmass. Larger grains show a strong tendency to corrode and enclose the other minerals, and some mega- crysts are strongly poikilitic. Crosshatching may be well developed, rudimentary, or absent, and Carlsbad twinning is common. Perthitic tex- ture is usually conspicuous, with film, string, and patch types most common. The plagioclase is anhedral to subhedral, albite-twinned oligoclase-andesine (An28-An.3i), commonly with albite rims on grain boundaries in contact with K-feldspar. Patchy alteration to saussurite or sericite is not unusual. Quartz is anhedral and commonly shows undulose extinction. Biotite is anhedral to subhedral with X light brown, Y - Z dark brown. In the porphyritic facies it tends to be concentrated with the minor accessories in wispy aggregates, whereas in the microadamellite it occurs as disseminated flakes commonly showing a preferred orientation. Alteration to chlorite is evident in some samples. Textures and crystallisation history The crystallisation history of the Mt Gardner Adamellite is not easily determined from grain relations. Apatite and zircon may be included in magnetite and snhene, and all these minerals tend to be euhedral against, and enclosed by, biotite. Biotite and quartz are occasionally en- closed by plagioclase, and these three minerals (especially plagioclase) are commonly corroded and enclosed by K-feldspar. Plagioclase is locally euhedral against quartz, and quartz is commonly interstitial to both feldspars. Hence the order in which the minerals commenced to crystallise appears to be: (i) apatite and zircon; Journal of the Royal Society of Western Australia. Vol. 56. Part 4. December, 1973. 105 Table 2 Chemical analf/nest, inesoTwrmft. and modes the porphyritic adamellite and m icroadamellite facies of the Mt (rurdner Adamellite Por])liyritic Adamellite Microadamellite 5(5582 65625 65629 6.5626 (55628 65630 SiO., .... 6(5 -07 67-74 67 -89 69-42 68-88 71-43 ALO 3 .... 1(5-25 16-18 15-20 15-07 14-87 14-33 FeoOg .... 2-25 1-42 1-73 1-51 1 - 52 1-88 FeO .... 1 -4:5 1-14 1 -31 1 -52 1-63 1-11 MfzO .... 0-90 0 • 53 0-81 0-71 1 - 00 0-62 CaO . . 2-68 2-25 1 • 99 2-14 2-11 1-22 X;uO .... 3-71 3 - 50 3-14 3-29 3-14 2-85 Rod .... 5-49 6-00 6 • 43 5-25 5-56 5-83 H 0 O+ . 0-54 0-38 0-47 0-80 0-87 0 - 62 HoO- .... 0-08 0-10 0-13 0 • 09 0-11 0-18 TiO.> .... 0-64 0-38 0-54 0-49 0-52 0-43 P,Os 0-24 0-20 0-19 0-13 0-18 0-11 MnO .... 0-06 0-05 0-06 0-05 0 • 05 0-03 Total 100-34 99-87 99-89 100-47 100-44 100-64 Mesonorms (mol. %) Q 17-63 19-10 20-35 24-70 24-17 28-75 Or 29-93 33-85 35-87 28-61 29-68 33-11 Ab 33-52 31-71 28 • 58 29-92 28-59 25-98 An 9 - 59 8-64 6-87 8-19 7-58 3 • 93 C 0 - 79 0-89 0 • 65 112 1-04 2-13 Bi 4-31 3-10 4-27 4-53 5-83 2-94 Mt 2-36 1-40 1-83 1 • 59 1-60 1 • 99 Sp .... 1-34 0-79 1-14 1 -02 1-09 0-90 Ap 0-50 0-42 0-40 0-27 0-38 0-23 Trace Elements (p.p.m.) Li 19 13 21 17 19 16 Co 7 4 5 3 4 6 Ni 33 33 19 33 18 19 Cii 6 8 5 6 9 5 Zn 79 68 70 70 68 65 Rb 187 211 222 208 196 221 Sr 754 641 606 533 469 218 Y 56 49 6(5 18 77 18 Zr 456 336 409 353 490 364 Ba 3789 3818 3962 2747 2949 1448 .... 179 115 106 182 191 136 Pi) 44 48 49 45 41 38 Th 23 18 25 39 39 52 Modes (vol. %) Quartz K-feld- 19-1 19-6 19-7 23-0 23-6 27-4 s])ar Pla^io- 39-5 43-6 44-8 36-9 39-1 41-2 clase 34-7 28-9 28-9 31-2 30-6 24-4 Biotite 4-2 5-1 4-1 0-8 5-4 5-5 Best .... 2-5 2-8 2-5 2-1 1-3 1-5 Pliifl. An% 28-30 28 30 28 28 31 , 29-30 (ii) magnetite and sphene (hi) biotite : (iv) plagioclase; (v) K-feldspar. The position of quartz in the crystallisation sequence is regarded as doubtful. There was probably a large overlap between the crystallisation ranges of the felsic minerals. Several textural features of the Mt Gardner Adamellite give rise to speculation regarding certain aspects of the petrogenesis of the pluton. The tendency for biotite and minor accessory minerals to occur in aggregates suggests that these minerals may be refractory remants of parent rock or xenoliths rather than products of magmatic crystallisation. The preferred orientation of disseminated biotite flakes in the microadamellite parallel to intrusion margins is probably a primary flow structure. The strong tendency of K-feldspar to corrode, enclose, and replace the other minerals suggests a post-mag- matic (autometasomatic) phase of K-feldspar growth. Perthite and albite rims on plagioclase are believed to have developed by subsolidus reorganisation of albite exsolved from K-feld- spar (see Phillips 1964). The common occurrence of undulose extinction in quartz and K-feldspar, and occasional fractured feldspar grains and bent biotite flakes suggest some post-consolida- tion deformation of the pluton. Chemical analyses A comparison of the analyses in Table 2 shows that the porphyritic adamellite and micro- adamellite facies of the Mt Gardner Adamellite are very similar in composition. The micro- adamellite tends to be slightly richer in Si 02 and Th, and slightly poorer in AI2O3, P2O5, Sr, and Ba than the porphyritic adamellite. Minor intrusions Both major facies of the Mt Gardner Adamel- lite are cut by occasional small veins of pegma- tite and quartz a few centimetres in width. The relative ages of the pegmatite and quartz veins are not known. The pegmatite is a coarse-grained, hypidio- morphic-textured rock composed mainly of quartz, oligoclase, and microcline, with minor magnetite and biotite. The quartz veins are coarse-grained, allotriomorphic-textured, and composed almost entirely of quartz. Xenoliths Xenoliths are fairly common in the porphyri- tic adamellite. They occur as clearly defined, angular blocks up to about 20 m across, showing little evidence of assimilation. The lithologies represented are restricted to those found in the nearby country rocks, with xenoliths of granitic gneiss outnumbering those of dioritic gneiss by at least ten to one. Chemical analyses and modes of representative samples are presented in Table 3 for comparison with the analytical data for the country rock gneisses listed in Table 1. The xenoliths generally show random orientation of their internal foliation, and therefore appear to have been rotated during theii* incorporation in the pluton. The microadamellite dykes contain few xenoliths, mostly of porphyritic adamellite. It is concluded that the xenoliths in the Mt Gardner Adamellite have been rafted from the adjacent wall and roof rocks. Their nature is consistent with intrusive magmatic emplacement of the pluton, rather than metasomatic em- placement, but there is no evidence to suggest that they have been transported from signifi- cantly greater depth. Contact relations Contacts between the Mt Gardner Adamellite and surrounding gneisses are mostly obscured, either by the Southern Ocean or by superficial deposits. The sole exception occurs at the north- Journal of the Royal Society of Western Australia, Vol. 56, Part 4, December. 1973. 106 Table 3 Chemical ajiahjses and modeff of firanitic and dioritic gneiss xenoliths in the Mf. Gardner Adamellite Granitic gneiss Dioritic gneiss SiO.. 56583 71-45 65631 78-91 65627 56-19 AlA 13-66 10-79 13-44 Fe.Os 1-57 1-16 1-87 FeO 1-24 0-50 5-05 HsrO 0-42 0-15 7-55 C.aO 1-24 0-49 5 - 93 Xa .,0 2-86 1 -98 2-37 K ,0 5-90 6-01 4-60 H„0* 0-57 0-28 0-95 H..O- 0-13 0-13 0-10 Tib,, 0-33 0-07 1 -05 P 2 O 5 0-08 0-02 0-72 MnO 0-04 0-14 Total 99-40 100-49 99-96 Trace Elen Li lents (p.]).in 14 .) 2 35 Co 5 <3 37 Ni 31 15 204 Cu 6 5 7 Zn 61 27 113 Rb 238 191 188 Sr 249 219 1136 Y 25 8 29 Zr 297 88 216 Ba 1454 1330 3913 La 316 <6 84 Pb 51 34 33 Th 78 <5 16 Modes (vol. %) i Quartz 1 27*3 43*7 13-4 K-feldspar 35-6 40-7 1 14-9 Plagioclase 31-2 14-2 : 23-5 Biotite 4-0 0-4 ; 21-7 Hornblende 25-0 Rest i-9 i-0 1 1-5 Plag. An % 28 20 31 ern margin where about 600 m of the contact is exposed at Point Valiant. Here the contacts are sharp in detail, but the pluton margin is some- what indefinite, being defined by a wide zone in which gneiss and adamellite are intermingled. Porphyritic adamellite is interbanded with the gneiss in lit-par-lit fashion on a large scale, and irregular discordant intrusions of porphyritic adamellite and microadamellite into gneiss are common. Small veins of quartz and pegmatite are also fairly numerous. The gneiss in this contact zone locally shows development of K- feldspar porphyroblasts, suggesting K-metaso- matism, but there is no evidence of major thermal effects in the contact rocks, nor is there any sign of marginal chilling in the adamellite. It is concluded that the contact relations are consistent with intrusive magmatic emplace- ment of the Mt Gardner Adamellite, rather than metasomatic emplacement. Discussion Petrogensis of the Mt Gardner Adamellite Contact relations and the nature of xenoliths strongly suggest that the Mt Gardner Adamel- lite was intrusively emplaced as a magma or crystal mush. The spatial association and simi- larity in composition between the porphyritic adamellite and microadamellite facies suggest a close genetic relationship between them. The nature of this relationship is investigated below. Despite the strong similarity in bulk chemical composition it is evident in Table 4 that K/Rb, Ba/K, Ba/Rb, and Sr/Ca ratios are slightly lower in the microadamellite than the porphyri- tic adamellite. Consideration of the substitution behaviour of Ba and Rb for K, and of Sr for Ca during progressive fractional crystallisation of granitic magma (Taylor 1965) suggests that these results are consistent with the micro- adamellite dykes being the residual product of fractional crystallisation of the porphyritic adamellite magma. The pegmatite and quartz veins are possibly the product of more ad- vanced fractionation. Q Figure 3.— Mesonormative Ab-Or-Q proportions for the Mt Gardner Adamellite samples compared with the cotectic lines for the system An-Ab-0r-Q-H20 where Ab/An = 3.8, for water vapour pressures of 2 kb (from von Platen 1965) and 7 kb (inferred from von Platen 1965, and von Platen and Holler 1966). In Figure 3 (after von Platen 1965, and von Platen and Holler 1966) the Mt Gardner Adamel- lite samples are compared with phase relations in the system An-Ab-Or-Q-HoO for the appropriate Ab/An ratio of 3.8. The microadamellite samples approximate the minimum melting composition for water vapour pressures around 7kb, whereas the porphyritic adamellite samples plot signifi- cantly further from the Q-corner. Consideration of the crystallisation behaviour of melts in the system An-Ab-Or-Q-H.O (see von Platen 1965) shows that these results support the contention that the microadamellite dykes are the residual product of fractional crystallisation of the por- phyritic adamellite magma, and suggests that fractionation occurred at a water vapour pressure of 7 kb or less. Filter pressing resulting from tectonic dis- turbance of the pluton during the later stages of crystallisation is seen as a likely factionation mechanism. Journal of the Royal Society of Western Australia, Vol. 56, Part 4, December, 1973. 107 Origin of the magma Stephenson (1973 in prep.) has argued that similar granitic plutons nearby are the product of anatexis of crustal rocks during the orogeny responsible for the high-grade meta- morphism of the country rocks. A similar origin for the Mt Gardner Adamellite magma seems likely, although there are no radiometric or Sr isoptope data available to confirm or refute this suggestion. The country rocks in the vicinity of the Mt Gardner Adamellite belong to the amphibolite facies, and hence m,ay have attained a tempera- ture high enough to cause substantial anatexis. Furthermore, the country rocks are composed mainly of granitic gneiss and therefore could yield large amounts of granitic magma on par- tial melting. Thus it is possible that the Mt Gardner Adamellite may be the product of ana- texis more or less in situ. However, this pos- sibility can be ruled out on chemical grounds. It is reasonable to assume that elements con- centrated in the final stages of fractional cry- stallisation of magma should also be concen- trated in early-formed anatectic melts. Hence a magma formed by partial melting should show lower K/Rb, Ba/K, Ba/Rb, and Sr/Ca ratios Table 4 Element ratios for the Mt Oardner Adamellite and the gneissic country rocks K/Kl) Ba/K xlO® Ba/Rb Sr^Ca xl0“ I’orphvritic Adamellite 56582 294 8:i 20 39 05625 284 ! 4 18 40 05029 290 74 18 4:3 MicroadamelHte 05620 252 0:3 18 35 05628 284 04 15 31 056:10 264 :10 0-6 25 Granitic (Jneiss 56578 196 50 11 17 56585 208 :12 o-o 12 56592 10:1 0-4 1-0 5-7 5659:3 242 49 12 12 Dioritic (Gieiss 5451:} 154 01 9-4 1:3 than the parent rock (see Taylor 1965). Com- parison of these ratios for the Mt Gardner Adamellite with those for the granitic and dio- ritic gneiss country rocks (Table 4) suggests that the pluton cannot have been formed in situ by partial or complete melting of the country rocks. Therefore an origin at greater depth is assumed. Acknowledgements —This paper is based on part of a Ph.D. project supervised by Professor R. T. Prider and Mr. J. G. Kay of the Geology Department, University of Western Australia. The major element X-ray flu- orescence analyses were carried out in the Physics De- partment, Western Australian Institute of Technology with the permission of Dr. J. de Laeter and under the instruction and supervision of Messrs G. Kerrigan and W. Thomas. The remaining chemical and X-ray flu- orescence analytical work was done with the assistance of Mr. P. E. Bannister. References Barth, T. F. W. (1962). — A final proposal for calculating the mesonorm of metamorphic rocks. J. Geol. 70: 497-498. Compston, W and Arriens, P. A. (1968).— The Precam- brian geochronology of Australia. Canad. J. Earth Sci. 5: 561-583. Deer, W. A., Howie, R. A., and Zussman, J. (1963). — “Rock-Forming Minerals, Vol. 4, Framework Silicates” . Longmans, London. Phillips, E. R. (1964). — Myrmekite and albite in some granites of the New England Batholith, New South Wales. J. Geol. Soc. Aust. 11: 49-59. Platen, H. von (1965). — Experimental anatexis and genesis of migmatites. In “Controls of Metamor- phism”. (W. S. Pitcher and G. W. Plinn, eds.) 203-218. Oliver and Boyd, Edinburgh. Platen, H. von and Holler, H. (1966). — Experimentelle Anatexis des Stainzer Plattengneises von der Koralpe. Stelermark, bel 2, 4, 7 und 10 kb H 2 O — Druck. Neues Jb Miner. Abh. 106: 106-130. Stephenson, N. C. N. (1973). — The petrology of the Mt Manypeaks Adamellite and associated high-grade metamorphic rocks near Albany, Western Australia. J. Geol. Soc. Aust. 19: 413-439. Stephenson, N. C. N. (in prep.).— The petrology of the Albany Adamellite and Torbay Adamellite plutons, near Albany. Western Australia. Taylor, S. R. (1965). — The application of trace element data to problems in petrology. In “Physics and Chemistry of Earth”. L. H. Ahrens et al., eds.) Vol. 6: 133-213. Pergamon Press, Oxford. Turner, F. J. (1968). — “Metamorphic Petrology: Minera- logical and Field Aspects”. McGraw-Hill, New York. Journal of the Royal Society of Western Australia, Vol. 56, Part 4, December, 1973. 108 15. — A preliminary investigation of the Yilmia Meteorite by J. Graham^ and A. C. Theron^ Manuscript received 20 March 1973; accepted 17 April 1973 Abstract. The meteorite found at Yilmia Hill in West- ern Australia has been shown by optical and electron microprobe analysis to be an enstatite chondrite of type II (or class E6). It is charac- terised by relatively abundant sinoite and tri- dymite. Several criteria indicate a low value for the temperature of final equilibration although the phase assemblage is beginning to be affected by wealthering. Introduction Only sixteen meteorites of the enstatite chon- drite type were reported by Mason (1966). They were subdivided by Keil (1968) into two types (I and II) with two intermediate stones, on the basis of the iron content and other mineralogical characteristics which will be further noted below. A new find was made on Australian Selec- tion (Pty) Limited ground at Yilmia Hill near Spargoville, W.A., late in 1969 by Mr. G. Coulson, but was unrecognised until June 1971. At this time a second occurrence was discovered only about 400m south of the first. The two occur- rences were designated Yilmia I and Yilmia II. It was not possible for the present authors to carry out the extensive statistical work required for a complete analysis of the meteorite. How- ever, sufficient mineralogical and electron microprobe data have been accumulated to place it unequivocally into Type II of the enstatite chondrites and to point out some of its peculiar- ities. A limited point count on a representative field of view showed that the percentage of metallic nickel-iron was above average, and of silicates below average, for this type of meteorite. The class symbol according to the Prior-Mason system as modified by Keil (1969) would be Ce 2 , while the Van Schmus and Wood (1967) group would be E6. The analytical data presented in the section on mineralogy are only given as approximate figures, since in view of the vari- ability of the phases, individual analyses are almost meaningless. Description of the find Both masses occurred on fiat, soft, lateritic soil which is underlain at a depth of about 30 cm with a hard laterite. The absence of outcrop in this area accounts for the ease with which the meteorites were found. The two masses were discovered three km north north-east of Yilmia Hill Trig Station at latitude 31° 12' and longi- tude 121° 3T (Figure 1). The meteorites were partially exposed with the lower portions buried about 8 cm deep. A scatter pattern of smaller fragments (Figure 2A), per- haps thrown off on impact, northwards of the 1 CSIRO, Division of Mineralogy, Wembley, Western Australia 6014. 2 Australian Selection (Pty) Limited, Kalgoorlie, West- ern Australia 6430. Yilmia I mass, together with the fact that the two impact sites lie on a north-south line, sug- gests that the meteorites came in on a south- north trajectory. Unfortunately the Yilmia II site was disturbed before the position of the loose fragments relative to the main mass could be recorded. The scatter patterns represent something of a mystery, for this type of meteorite should have sufficient metal to resist fragmentation on impact, especially in soft soil. The alternative suggestion that fragmentation is the result of weathering does not explain the scatter over a distance of one metre or the relatively un- weathered state of the exposed surfaces. Although the lateritic area surrounding the sites was searched carefully, no further discoveries were made. The main mass of Yilmia II (figure 2 B and C) is the best preserved and roughly arrow shaped, the blunt point being originally directed towards the north. The side that faced south is almost flat. The buried fragments of Yilmia I, when pieced together, Indicate a roughly similar shape. The buried portions were severely weathered to a rusty yellow brown colour. The weathered lower side of Yilmia II is strongly exfoliated or layered, unlike the exposed upper half which is apparently devoid of layering. The exposed surfaces are dark brown, fairly smooth and un- dulose. Irregular and pentagonal crack patterns were seen on some of the more weathered sur- faces. Freshly cut surfaces had a bluish colour which became oxidised upon exposure within a matter of days. Yilmia I weighed 16.2 kg of which the main mass weighed 8.6 kg. Yilmia II weighed 23.8 kg, the weight of the main mass being 11.3 kg. The greater part of these finds are preserved as samples 13192 and 13197 respectively at the Western Australian Museum. Almost 100 gm at the Kalgoorlie School of Mines have sample numbers 10951.1 and 10951.2. Small samples exist at CSIRO Laboratories, Perth, and in the Geology Department, University of Melbourne. All the mineralogical data in this paper refer to the find Yilmia I (see Figure 2A). Analyses were carried out on an MAC-400S microprobe analyser operating at 20 kV for all elements except carbon, nitrogen and oxygen, which were analysed at 8 kV. For the most part metallic standards were used, although comparison was made with other oxides, sulphides and silicates. Apatite was used as a phosphorus standard. Mineralogy From the mineral assemblage shown in Table 1, and the detailed composition of the phases, Journal of the Royal Society of Western Australia, Vol. 56, Part 4, December. 1973. 109 Journal of the Royal Society of Western Australia, Vol. 56, Part 4, December. 1973. 110 :nCh£S CM APPmM^MA^SLY A-fS^ PACE Figure 2.— (a) Scatter pattern of mass Yilmia I, north being to the right of the photograph. The white card marks the spot from which came the portion of the meteorite described in this paper. It is possible that this portion was moved to this spot after discovery of the site. (b) Main mass of Yilmia II, showing smooth surfaces exposed to the air, and the layered buried portion. The pointed (northern) side is towards the viewer. (c) Yilmia II viewed obliquely from above relative to position as found. The right hand side was facing north. Journal of the Royal Society of Western Australia, Vol. 56, Part 4, December, 1973. Ill Table 1 Phases Present in the Yilmia Hill Meteorite Opaque Phases : Phase Ideal Composition Minor Elements Present (A])proximate %) native iron* a Pe 2-4 Xi, 0-4 Co, 0-2 Cr, 0-1 Ti kamacite a Pe« 7-5% M) o-~l Xi, —1 Si. 0-3-0-6 Co, ---O-l Zn taenite y Fe (8-55% Xi) 12-16 Xi.-^l Si. 0-4 Co troilite FeS 1-l.i Cr, ,V-1 Ti. 0-1 Mil, 0-1 V, .^0-1 Zn, -05 Co alabandite MnS Up to 21 Fe, 2 Mg, 0-1 Cr, <0-5 Ca, 0-1 Zn daubreelite FeCr.,S4 3 Mil. —-15 Zn oldhamite CaS 2 Fe. 0-6 Mu. 0-3 Si schreibersite Fegl* 23-26 Xi, 0-15 Co, some S, — -05 Zn ])cntlandite* tFe, Xi) Co graphite a-C * May be an alteration product. Transparent Phases: Phase Ideal Composition enstatite MgSiOa oligoclase (Na. Ca) (Al, Si) 4()8 tridvmite SiO„ sinoite Si.,X.,0 clinopyro.xeue (Fe, Mg, Ca) SitJg the meteorite is a Type II enstatite chondrite. In thin section some relict chondrules can be distinguished. A feature of the sections is the segregation into metal-rich and sulphide-rich regions (Figure 3 A and B) and into oligoclase- rich and estatite-rich regions, on a scale of a few mm. Although the sections studied were fairly fresh, and most of the phases were un- altered, much of the oldhamite had been weath- ed away, and no good section of this mineral was exposed. In some areas, kamacite and ala- bandite were partly altered to an oxide phase (Figure 30. Even where the original phase was kamacite, sulphur was present in the product. Some of this has undoubtedly been introduced in the weathering solutions, but there also seems to be a thin rim (- Um) of sulphide surround- ing many kamacite grains, and in some cases this is left unaltered by the mild weathering, showing the outline of the original grain. One weathered grain was apparently after schreiber- site. Analysis of these regions is not possible on a quantitative basis, because they are hetero- geneous on a micron scale. However, many microprobe traverses show the presence of nickel-rich regions in which Ni may exceed iron by as much as five to one, and the total metal counts may exceed those from a monoxide. Sulphur is sometimes present, and the example shown in Figure 4 could be interpreted as a very small grain of pentlandite, containing some enrichment of cobalt. These phases are evidently not in equilibrium. Siliceous Minerals Enstatite — This phase is an enstatite of low manganese content. Its iron and calcium con- tents are as expected for Type II, but its alu- minium content is lower than usual. Full quan- titative data were not obtained. OUgoclase and tridymite are also present; probe data were obtained only for identification purposes, so no minor element trends can be quoted. The Si02 phase was not obvious in our thin section, and a few grains were removed from the polished section for Debye-Scherrer analysis. It was clearly identified as tridymite and was quite abundant. Sinoite, Si2N20 occurred in several large grains in the polished section. It forms hard, columnar crystals, which normally fluoresce strongly under electron bombardment. Some areas, however, fluoresced only weakly, and it is evidently dangerous to take the fluorescence as the major diagnostic criterion for this mineral. Opaque minerals Kamacite is, as usual, the most abundant opaque phase. It contains 5-7-2% nickel, 0.3-0. 6% cobalt, and up to 1% silicon. Apart from the even lower silicon content, these figures are typical of kamacite from Type II enstatite chondrites, although some zinc (up to 0.1%) also seems to be present in the Yilmia material. One grain of native iron was observed adjacent to troilite and the more typical kamacite: when carbon coated, its colour was distinctly different from that of the latter. This contained only 2.4% nickel and no silicon. The minor elements present included 0.4% Co, 0.2% Cr, and 0.1% Ti, figures which suggest that it may have been formed by weathering from the adjacent troilite, since the phase assemblage is still highly reduc- ing. Ni and Co would need to be contributed from kamacite and taenite. Graphite is generally associated with the kamacite. Taenite occurs in several grains, usually ad- jacent to kamacite, and was recognised by its high nickel content. This is then the third enstatite chondrite to contain taenite, and the second of Type II. The nickel content of 12-16% Joiirnal of the Royal Society of Western Australia, Vol. 56. Part 4, December, 1973. 112 Journal of the Royal Society of Western Australia, Vol. 56. Part 4, December, 1973. 113 S3S o U +:> 0) “s s W '55 S ^ O O ?-< r> it on is" t-S OJ bC^ cS "fe c 33 ^ iC “S cS - O >rt -^rG S CD E-< a ■5 -2 . 0. -§, : bC-C C ^ 0) 42 O Sh O) O Cd CQ a a ^ u « S ;=J a ■r-l dJ O K tn (1) CL) (U - c3