S ' ' S HARVARD UNIVERSITY Library of the Museum of Comparative Zoology JOURNAL AND PROCEEDINGS OF THE ROYAL SOCIETY OF NEW SOUTH WALES 1977 PUBLISHED BY THE SOCIETY, SCIENCE CENTRE, 36 CLARENCE STREET, SYDNEY Royal Society of New South Wales OFFICERS FOR 1977-1978 Patrons His Excellency the Governor-General of Australia, The Honourable SIR JOHN KERR, k.c.m.g., k.c.v.o., c.b.e. His Excellency the Governor of New South Wales, SIR RODEN CUTLER, v.c., k.c.m.g., k.c.v.o., c.b.e., K.st.j. President W. H. ROBERTSON, b.sc. Vice-Presidents F. C. BEAVIS, B.sc., PH.D., F.G.s. W. E. SMITH, m.sc., ph.d., m.inst.p. E. K. CHAFFER D. J. SWAINE, m.sc., ph.d., f.r.a.c.i. P. D. TILLEY, B.A., PH.D. Honorary Secretaries M. J. PUTTOCK, B.sc. (eng.), m.inst.p. M. KRYSKO v. TRYST, b.sc., GRAD. DIP., A.M.AUS.I.M.M. Honorary Treasurer A. A. DAY, B.sc., PH.D., F.R.A.S., M.AUS.I.M.M. Honorary Librarian W. H. G. POGGENDORFF, b.sc. (agr.) Members of Council HELENA BASDEN, b.sc., dip.ed. J. W. G. NEUHAUS, m.sc., a.r.i.c., f.r.a.c.i. G. S. GIBBONS, M.sc., ph.d. S. J. RILEY, b.sc. (hons.), ph.d. G. C. LOWENTHAL, b.a., m.sc., ph.d., LAWRENCE SHERWIN, M.sc. f.a.inst.p. F. S. STEPHENS, b.sc., ph.d. D. H. NAPPER, m.sc. (syd.), ph.d. (cantab.) A.R. A.C.l. New England Representative : R. L. STANTON, m.sc., ph.d. South Coast Representative : G. DOHERTY, B.sc., ph.d. CONTENTS Parts 1-2 Geography : Catastrophic Channel Changes in the Macdonald Valley, New South Wales, 1949-1955. H. M. Henry . . . . . . . . . . . . . . 1 Geology : Does the Hunter River Supply Sand to the New South Wales Coast Today ? P.S.Roy 17 Geophysics : A Late Devonian Palaeomagnetic Pole for the Mulga Downs Group, Western New South Wales. B. J. J. Embleton . . . . . . . . . . 25 Mathematics : Series of Roots of a Transcendental Equation. Pietro Cerone and Austin Keane . . . . . . . . . . . . . . . . . . . . 29 Palaeontology : Floral Evidence for a Middle Triassic Age of the Gunnee Beds and Gragin Conglomerate, near Delungra, New South Wales. D. J. Bourke, R. E. Gould, R. Helby, R. Morgan and G. J. Retallack . . . . . . . . 33 The Tertiary Stratigraphic Palynology of the Murray Basin in New South Wales. 1. The Hay-Balranald-Wakool Districts. Helene A. Martin .. 41 Presidential Address 1976 : Leather — Why is it so ? Edric Chaffer . . . . . . . . . . . . 49 Report of Council, 31st March, 1977 . . 61 Parts 3-4 Report of Council, 31st March, 1977 (Continued from Volume 110, Parts 1 and 2) . . . . . . . . 73 Astronomy : Occupations observed at Sydney Observatory during 1974-1976. D. 5. King and K. P. Sims . . . . . . . . . . . . . . - . . . 81 Precise Observations of Minor Planets at Sydney Observatory during 1976. T. L. Morgan . . . . . . . . . . . . . . . . . . 87 Chemistry : The Essential Oil of the Fly- Repellent Shrub, Pterigeron bubakii. I. A. Southwell and J. R. Maconochie . . . . . . . . . . . . 93 IV CONTENTS Contents — Continued Geology : A Bottom Profile across Lake Eyre North, South Australia. J. A. Dulhunty 95 Influence of Hydrothermal Treatment on Physical and Chemical Properties of Chrysotile Asbestos. P. Hahn-Weinheimer and A. Hirner .. .. 99 Textures of the Carboniferous Ignimbrites in the Hunter Valley, N.S.W. B. Nashar and A. T. Brakel .. .. .. .. .. .. .. Ill Clarke Memorial Lecture : Petrogenetic Aspects of Some Alkali Volcanic Rocks. J. F. G. Wilkinson . . 117 Palaeontology : Fossil Marsupials from the Douglas Cave, near Stuart Town, New South Wales. John D . Gorier . . .. .. .. .. .. .. .. 139 Journal and Proceedings of the New South’'JfeIes VOLUME no 1977 PARTS land 2 Published by the Society Science Centre, .<5 Clarence Street, Sydney Issued 14th July, l‘)77 Royal Society of New South Wales OFFICERS FOR 1977-1978 Patrons His Excellency the Governor-General of Australia, The Honourable SIR JOHN KERR, k.c.m.g., k.c.v.o., c.b.e. His Excellency the Governor of New South Wales, SIR RODEN CUTLER, v.c., k.c.m.g., k.c.v.o., c.b.e., K.st.j. President W. H. ROBERTSON, b.sc. Vice-Presidents F. C. BEAVTS, B.sc., ph.d., f.g.s. W'. F. SMITH, m.sc., ph.d., m.inst.p. E. K. CHAFFER D. J. SWAINE, m.sc., ph.d., f.r.a.c.i. P D. TILLEY, B.A., PH.D. Honorary Secretaries M. J. PUTTOCK, B.sc. (eng.), m.inst.p. M. KRYSKO v. TRYST, b.sc., GRAD. DIP., A.M.AUS.I.M.M. Honorary Treasurer A. A. DAY, B.sc., PH.D., F.R.A.S., M.AUS.I.M.M. Honorary Librarian W. H. G. POGGENDORFF, b.sc. (agr.) Members HELENA BASDEN, b.sc., dip.ed. G. S. GIBBONS, M.sc., ph.d. G. C. LOWENTH.\L, b.a., m.sc., ph.d., F.A.INST.P. D. H. NAPPER, M.sc. (syd.), ph.d. (cantab.) A.R. A.C.I, of Council J. W. G. NEUHAUS, m.sc., a.r.i.c., f.r.a.c.i. S. J. RILEY, B.sc. (hons.), ph.d. LAWRENCE SHERWIN, m.sc. F. S. STEPHENS, b.sc., ph.d. New England Representative : R. L. STANTON, m.sc., ph.d. South Coast Representative : G. DOHERTY, b.sc., ph.d. Mus. COMP, zool;;, . . J_IBRAR-Y . SEP 2 8 1P77 Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 1-16, 1977 HARVARD UNIVERSIXY Catastrophic Channel Changes in the Macdonald Valley, New South Wales, 1949-1955 H. M. Henry ABSTRACT. Floods in the Macdonald valley, near Wiseman's Ferry, New South Wales between 1949- 1955 transformed the initially narrow V-shaped channel into a very wide channel with over- steepened banks. The difficulty of quantifying as distinct from recognising change in a river channel more than twenty years after it occurred is greatly increased in the absence of large- scale maps, aerial photography and streamflow records. Buried bridges, the downstream displacement of wharves, tidal changes, the ponding of fributary creeks, old photographs and survey plans, and information obtained from farmers have been used to reconstruct the character and extent of channel change. There are signs that since 1969 the river may have begun to resume its pre-1949 form by narrowing its channel and lowering its bed. INTRODUCTION Between June 1949 and February 1955 the Macdonald River, a tributary of the Hawkesbury River in eastern New South Wales, trebled the width of its bed for more than 30 kilometres above its tidal zone and aggraded its bed by almost three metres, and in doing so it completely changed the shape of its channel. During the preceding 80 years there had been a marked fall in the number of people living in the Macdonald valley; no extension of settlement had taken place, there had been no relevant change of land use, nor any interference by man with the river's natural flow. The Macdonald River is 180 km long, and has a drainage basin a little more than 2000 km^ in area. Its banks, from six to nine metres high in the study tract between Mario Creek and Wright's Creek (Fig. 1) have only been completely submerged by floodwaters on four occasions since 1867. Although average rainfall in the valley is about 750 mm, prolonged periods of low flow are common and periods of no flow occur. For fifteen years after the flood damage there was no apparent change in the channel of the Macdonald River. Since 1969, however, there are signs that the river may be returning to its pre- 1949 form by regrassing its channel , stabilising, extending and building up its low benches , and ; by degrading its bed. According to Stevens et al (1975),"... the effect of a particular flood event on river form is indicated by the ratio of that flood's peak discharge to the average annual peak-flood discharge". The history of the Macdonald since 1867 suggests this is an over-simplification. Lesser floods on the Macdonald between 1949-1955 were more destructive than the major floods which preceded them: clearly the whole meteorological and hydrological context must be taken into consideration. Nevertheless, the Macdonald River exemplifies what Stevens et al (1975) call "non- equilibrium river form". Hickin and Page (1971) have pointed out that the narrow sandstone valleys of central eastern New South Wales are particular- ly subject to devastation by major floods. The present study results from frequent visits made to the Macdonald valley since 1952. Fieldwork on channel changes was started in 1969 and levelling completed in 1974. THE SETTING The Macdonald River rises on the divide between the Hawkesbury and Hunter River systems at an altitude of about 650 m (Fig. lA) . In its down- stream tract the river is deeply incised into a dissected sandstone plateau 200-300 m high, while for nearly 50 km above its junction with the Hawkesbury it is flanked by a narrow and discontinuous floodplain. In few places is the steep-sided valley more than 400 m wide. The catchment area of the Macdonald lies entirely within sedimentary Triassic rocks of the Sydney Basin. The river is incised throughout its length through the overlying Hawkesbury Sandstone into the upper formations of the Narrabeen Group of sandstones, claystones and shales, with sandstone strongly predominating in the Macdonald valley (Galloway, 1967). At least two-thirds of the surface rock is Hawkesbury Sandstone, a fact reflected in the surface channel deposits which consist of well-sorted sub-angular to sub-rounded, medium to coarse sand. According to Probert (1971) recent sand deposits in the channel of the Macdonald River are almost gravel-free, with the silt fraction rarely exceeding five per cent. The Macdonald River is tidal for 13 km above its junction with the Hawkesbury, and for 40 km above the tidal limit its gradient does not exceed 40 cm per km. The valley fill is up to 50m deep for 25 km from Wiseman's Ferry. Results from seismic and resistivity probe surveys have been inconsistent (Probert, 1971) but data obtained by a geophysical group from Macquarie University suggest that the fill is 45 m deep 3 km above St. Albans. Owing to the rugged topography and barren sandy soils more than 95 per cent of the Macdonald drainage basin is uncleared. Although rainfall in the valley is fairly well distributed throughout 2 H. M. HENRY Figure 1. A. Location map of the study area B. Map of the main settled tract of the Macdonald River above tidal j influence as discussed in this I report. :| CHANNEL CHANGES IN THE MACDONALD VALLEY 3 the year, only floodplains and two small upland areas of less than 15 km^ are not still covered by the original dry schlerophyll forest. Like many coastal streams in New South Wales the Macdonald River appears to be incised within its floodplain. The banktops are inundated only by rare high floods. In discussing the streams of the Curnberland Basin west of Sydney, Pickup (1974) suggests as an alternative to incision that their channel capacities are related to events less frequent than one-to-two year floods, the most common he believes being floods with return periods of from four to seven years. However, if the Macdonald River is not incised it seems clear that, in its present form, itS' channel must be related to flows of more than twenty years recurrence interval, or have been formed during a period of considerably higher discharge than the present. BEFORE THE 1949-1955 FLOODS From the early 1820 's when the first white settlers moved into the Macdonald valley, it had experienced a succession of major floods at intervals of more than 20 years - specifically, in 1867, 1889, 1913 and 1949. Such floods do not appear to have produced noteworthy geomorphic changes - not even the 1949 flood which, with the possible exception of the 1889 flood, was the highest on record. However, the 1949 flood was but the first of a series of floods which occurred during the next six years, and although not as high as the floods of 1867, 1889, 1913 and 1949, those lesser floods effected a spectacular change in the river channel. The enlargement of the Macdonald River's channel, and its change from V-shaped to rectangular cross-sectional form between 1949- 1955, are illustrated in the paired photographs (Plates 1 to 6) . The river flats in the Macdonald valley had been surveyed before 1835 and the best land was soon granted by the Crown. Since arable land was so limited and there were five or more children on every farm in those days, farmers cultivated their floodplains to the banktops and down the riverward sides. How this could be done successfully for years at a time is suggested by the cross-sections of Figure 2 and the data of Table 1. Sloping banks above the river also provided the best pasture in dry times. The banks were important: farmers who could not possibly have foreseen the devastation to be caused by the 1949-1955 floods, used to take shovels when part of a bank was washed away and reshape it so that it would offer as little resistance as possible to further floods, and would then encourage the grass to grow over it. The practice before 1949 of using tree-trunks as footbridges suggests the narrowness of the river above St. Albans. In 1940 C.J. Bailey whose farm is 2 km above St. Albans placed a 9 m pole across the river, and until 1949 he was able to cross dry- footed even when the river rose by a metre. When it rises a metre at this point today the river is more than 50 metres wide (see item 8, Table 2). W. Bailey recalls when a pole would span the Macdonald almost anywhere above Mogo Creek. G. Sternbeck remembers the river as a creek running through reeds when he was a boy. It was then less than 8 m wide at points where it is more than 40 m from bank to bank today (Plates 7 and 8). Older farmers tell how until the early years of the century reeds 3 m high grew thickly along the banks from Higher Macdonald to Wright's Creek. These were usually Phragmites australis and the introduced Arundo donax and they were equally abundant in tidal and non-tidal zones. No doubt they reflected the sluggish flow of the river, and growing so densely their deeply-rooting rhizomes must have helped to protect the banks from erosion. The disappearance of the reeds was not due to the floods: when fencing wire became readily available many farmers took up dairying and the cattle very nearly wiped them out. There had been little fencing in the Macdonald before 1900; most farmers had only one or two head of cattle, and farm animals were tethered overnight. However in view of the lack of damage to river banks in the 1949 flood it seems unlikely that destruction of reeds was an important factor in subsequent erosion. MACDONALD RIVER FLOODS 1949-1955 The history of the floods on the Macdonald River between 1949-1955 is as follows: 1. The June 1949 flood, nearly 12 m high at St. Albans, was the first major flood after 1913. With the 1889 flood it was the highest on record. It did little damage to the banks but it deposited a metre of sand in the river bed from Mario Creek to Wright's Creek. 2. This unprecedented sanding-up continued during 1950, the wettest year in east- central New South Wales since continuous rainfall records began in 1858. The annual discharges of both the Hawkesbury and Hunter Rivers for 1950 are the highest on record. No bank-submerging flood occurred on the Macdonald in 1950 but, according to multiple regression analysis carried out by the hydrology section of the N.S.W. Public Works Department (N.S.W. Department of Public Works, 1975), the estimated annual discharge at St. Albans in 1950 was more than double that of any year since 1913, which it exceeded by more than 90 per cent. The estimated monthly discharge at St. Albans in June 1950 was more than double that of any other month between 1903 and 1973. 3. The August 1952 flood was 2 m lower than the 1949 flood at Higher Macdonald but was responsible for the first general destruction of banks along the Macdonald. Twice in one week floodwaters covered flats which had been completely submerged only three times since 1867. The flood raided the height of the bed by another metre. 4. There were more floods during 1953-54. They were lesser floods, but on the 4 H. M. HENRY Hunter (and, according to unpublished comparative flood analysis by the writer. Hunter River floods correlate more closely with Macdonald River floods in respect both of incidence and stage than do those of the Hawkesbury) the 1953- 19S4 floods were higher than any had been for nearly twenty years before 1949. 5. The last of the major floods occurred in February 1955. This was the highest flood ever recorded on the Hunter. On the Macdonald it only reached 6.65 m at St. Albans, and for the most part it did not break the banks, but in many places above Mogo Creek it was the most destructive of the floods, and it completed the demolition of benches between Mario Creek and Wright's Creek. BANK EROSION AFTER 1949 The widening of the channel by the 1949-1955 floods and its changed cross-sectional form can be seen by comparing the paired Plates 1 to 6. Without comparative aerial photography it would not be possible to show that the sites illustrated are typical of channel change from Mario Creek to Wright's Creek during the flood period. Systematic aerial photography of the Macdonald valley dates only from 1954, and most earlier R.A.A.F. photography has been lost or destroyed. However, the National Library in Canberra holds R.A.A.F. photographs of the Mellong area taken in 1941 which cover the Macdonald valley north from Upper Macdonald. Although they begin 10 km above St. Albans they do show that the Macdonald channel in 1941 was very narrow relative to its present form (Plates 7 and 8). Survey regulations published by the N.S.W. Department of Lands have always required surveyors to set out both sides of a watercourse on their plans, but old surveys are of disappointingly little assistance in determining the width of the Macdonald River before 1949. On comparing old surveys with old photographs it is clear that surveyors adopted the tops of the river's high shelving banks as its boundaries although on the average they were over-topped less than once every twenty years. As a result the river is shown on old surveys as being far wider than it was except at time of high flood; and difficulties are compounded because channel widening between 1949- 1955 so often consisted in the river stripping back banks which had sloped at less than 15 leaving over-steepened banks in approximately former banktop position. There are some places, however, where some measure of channel widening can be obtained from surveys made before 1949. In 1892 the surveyor Charles Robert Scrivener, who later became the first Commonwealth Surveyor and laid out the city of Canberra, spent ten weeks on the Macdonald River as a Department of Lands surveyor pegging out a new parish road between St. Albans and Melon Creek. At Higher Macdonald Scrivener surveyed an access road to Preston's Gully a kilometre downstream, endorsing his plan as follows: "About half the land taken for the road is good cultivable soil; the other is sandy and of little value being practically part of old bed of river raised some feet above present level of river. The road is practicable". Scrivener's access road was used for 50 years, but it disappeared in the 1949-1955 floods, for it ran along the middle of the widened river bed shown on Plate 6. Last century a racecourse was laid down inside the bend of the Macdonald River below Mogo Creek junction; it is shown in Scrivener's field notebooks of 1892. The course was registered with the Australian Jockey Club and races held there until the 1920's. A survey plan prepared in 1921 plots banktop positions in unusual detail. Comparison with a 1965 aerial photograph enlarged to the same scale shows that not only the banks but many acres of floodplain were demolished by the floods. A subdivision plan of land at Upper Macdonald prepared in 1947 when E. Jurd acquired two acres to build a new dairy shows how the 1955 flood cut swathes 30 m wide from banks which had survived the higher floods of 1949 and 1952. Jurd's land extended 43 m from the river and was ploughable to its boundaries. After the 1955 flood Jurd's dairy stood at the edge of a nine-metre drop to the channel; his cultivation paddock with its 165 metre frontage to the road had disappeared, at one point taking the road with it. For 400 metres further downstream the river had also stripped away the bank. When farmers began dairying they erected fences across the river bed consisting of posts driven into the sand two to four metres apart connected by fencing wire. These provide another measure of the widened channel. W. Bailey of Higher Macdonald had five fence posts in the river before 1949 but by 1960 their number had increased to twenty. Many other farmers, particularly above St. Albans, found it necessary to increase three- and four- fold the number of their fence posts between 1949-1955. At only one place between Mario Creek and Wright's Creek did significant widening of the old channel not occur. Five kilometres below St. Albans two consecutive meanders were cut-off by the 1952 flood, and 23 years later the upper cut-off meander still preserves some of the dimensions of the former stream (Plate 3, and Fig. 2, Section 3). CHANNEL AGGRADATION 1949-1955 The evidence that the bed of the Macdonald River aggraded by about 3 m between 1949-1955 is conclusive. 1. Buried Bridges Four bridges crossed the Macdonald between St. Albans and Higher Macdonald before 1949 (see Fig. IB). Although they stood up to 3 m above the bed all were buried in sand between 1949 and 1955. Fernance's bridge, three kilometres above St. Albans, stood 3 m high and many farmers rode their horses under it. The 1949 flood raised the bed by a metre here, and the 1952 flood raised it by another metre. After the 1952 flood subsided C.J. Bailey found the river bed a metre higher than the week before. His account of catastrophic sanding-up CHANNEL CHANGES IN THE MACDONALD VALLEY 5 by the 1952 flood is supported by other farmers. Aggradation went on and after the 1955 flood Femance's bridge was buried to its decking in sand. The Upper Macdonald road continued to use it until 1958, but it was now no more than a ford, and even at low stage the river flowed over it. Bailey's bridge, two kilometres upstream from Femance's bridge, stood 3 m above the bed before 1949. According to a local farmer the 1949 flood aggraded the bed by a metre at this point, and sanding continued until by 1955 only the tops of the bridge piles were above the sand. Jurd's bridge, seven kilometres above Bailey's bridge at Upper Macdonald, is presently spliced to the top of a nearly buried bridge which was once almost 3 m above the river. E. Jurd estimates that the bed aggraded by more than 2.5 m between 1949 and 1955. Sternbeck's bridge at Higher Macdonald (Plate 5) was also buried by sand between 1949 and 1955. The late Keith Sternbeck, the writer's field assistant for a year, used to ride a draught horse under the bridge in the 1930' s when return- ing to his uncle's farmhouse after ploughing. The 1949 flood raised the bed by nearly a metre and in 1953 a new bridge was built 100 m upstream. By 1956 the old bridge was buried and its decking was not re-exposed until June 1973 when the removal of debris from above the new bridge led to its re-excavation. Several months later it was again covered by a few centimetres of sand although the river was running over it. 2. High - Level Bridge at St. Albans The high-level bridge at St. Albans demonstrates the aggradation of the river less spectacularly. Plate 1, taken about 1910, shows the steel piles braced by three sets of crossed girders; in the 1973 photograph (Plate 2) only two-and-a-half sets are still above the sand. The distance between the horizontal girders is more than five metres. According to local residents nearly all the sand was deposited during 1949- 1955. 3. Aggradation of the River Bed above Higher Macdonald No bridge was ever built above Higher Macdonald and the old road forded the river eight times between Sternbeck's bridge and Mario Creek. Fifty years ago there were seven farms above the first sandy crossing, but by 1949 they had been abandoned. W. Bailey used to run cattle from Melon Creek to above Yengo Creek, and in dry times he would dig a soak in the river bed near the foot of a 3 m-high rock outcrop to water his cattle. According to Bailey those rocks were buried during 1949-1955. He considers that aggradation of the Macdonald River near Yengo Creek junction, as well as of tributary creeks in the area, was not less than at St. Albans and Higher Macdonald. Two hundred metres above sea-level, Kindarun is a cattle property on the upper reaches of the Macdonald River for which there are continuous rainfall records from 1914 onward. On the night of June 18 1949, 425 mm of rain fell at Kindarun - the gauge overflowed at 325 mm but 425 mm of water were measured in an oil drum on a neighbouring property next morning. River flats were destroyed in the resulting flood, the channel width was doubled in many places and large trees were torn out of the ground. According to Mr. A. Halton of Kindarun, little sand was deposited because the river was running too fast. However, by 1955 the bed had aggraded by two metres at Kindarun; it does not seem to have aggraded or degraded since. 4. Ponding of Tributaries Some of the tributaries of the lower Macdonald apparently did not aggrade at the same rate as the river and consequently formed lagoons near their mouths. Mogo Creek joins the Macdonald River 4 km above St. Albans and for about 10 km before reach- ing the river it flows along a flat-floored valley about 150 m wide. This is the St. Albans Common established in 1825. Farmers who hauled logs across the Common by horse team before 1949 say that during droughts there was not enough water at the lower end of the Common to water a horse. Since 1949, there has been a lagoon at the lower end of Mogo Creek more than 1 km long and 100 m wide; even in dry times when the river has stopped running the lagoon has not greatly diminished. When the river rises without a corresponding rise in Mogo Creek it backs up into the lagoon and floods the Wollombi road, which once formed part of the Great North Road . Other permanent lagoons formed near St. Albans between 1949 and 1955 are: 1. Dawes Run, a minor tributary of the Macdonald 2 km below St. Albans, flooded 40 ha of grazing land near its junction with the river during the 1949 flood; they are still under water. 2. A large lagoon east of the river above Femance's bridge used to dry out in very dry periods - it was dry in 1946. It filled in 1949 and has not been dry since; a rail-fence dividing two properties is still under water. 3. A large depression behind the west bank of the river, 2 km above the cut-off meanders at Central Macdonald, once often dry, has held a lagoon since 1949. Tributaries of the Macdonald River above Mogo Creek did not form ponded lagoons as a result of the floods but the lower reaches of some became heavily silted. Thompson Creek traverses a flat near C. Sternbeck's farm-house at Higher Macdonald and Scrivener called it a "marshy flat" in his field notebook. For 40 years Sternbeck made only minor repairs to the fences subdividing the flat, but after the 1949 flood he found it necessary to raise them by half a metre because they were being buried. During the next five years Sternbeck raised them again twice, two metres in all. The fences have not needed attention since 1956 because, according to Sternbeck, the flat has not continued to aggrade. 6 H. M. HENRY 5. Wharves and Tides Although the Great North Road ran through St. Albans from 1885 to 1929, farm produce was transported out of the valley by small river- boats. Corn was carted from Higher Macdonald to a landing near Fernance's bridge in the mid-19th century, but the top wharf on the Macdonald stood at St. Albans for nearly 50 years, and Wharf Street appears on all old plans. The wharf was moved 3 km downstream about 1870 and remained there until 1917 when continued sanding forced its further removal to Stoney Gully on the now cut-off meander. It is significant that each time the wharf was displaced downstream it was 3-4 years after a major flood. In 1892 an 18-foot sailing boat with sweeps arrived at St. Albans and its crew were told that a vessel of that size had not reached the village for more than twenty years ("Hawkesbury Herald", November, 1905). The crew was also told that the recent 1889 flood had been 1.5 m higher at St. Albans than the 1867 flood and considerably more destructive. Until 1949 the Macdonald River was tidal to St. Albans. The parish map showed the limit of tidal influence to be 300 m downstream but local residents say that very high tides flowed nearly 400 m beyond the high-level bridge. After the 1949 flood the Macdonald was tidal only to Wright's Creek, and this has been so until recently, when some spring tides have reached the nick-point at the lower cut-off meander (see later) . EXPLANATIONS OF CHANNEL CHANGE That sanding-up had been continuing for nearly a century before 1949 can be easily understood, but the changes in the Macdonald River's channel between 1949 and 1955 were not foreseeable since no significant change in land use or extension of settlement had occurred. According to local farmers, there was as much cleared land on the Macdonald in 1889 as in 1949, and the population was greater. In 1865 Canon Greaves wrote that 800 people lived on the Macdonald (Elkin, 1955). By 1933 the population of the St. Albans police patrol district, which included the whole of the Macdonald valley, had fallen to 575; and between 1932 and 1949 the number of electors enrolled in the Macdonald valley fell from 190 to 97 (Wiseman's Ferry subdivision, Hawkesbury State electorate) . In 1966 at the request of an inter-departmental committee on flood mitigation in the Nepean- Hawkesbury valley, the N.S.W. Soil Conservation Service investigated the origin of recent sand deposits on the Macdonald and Colo Rivers. It reported (Dyson, 1966) that the deposits had resulted from erosion of the steep sandy slopes bordering the valley and that bushfires had been the main factor contributing to such erosion, other important factors including burning-off to provide winter pasture and felling and transport of timber. Dyson (1966) thought that erosion due to farming was small compared with the extensive erosion and sanding which had occurred throughout the timbered area, but apparently did not realise that the sanding-up had taken place within a period of six years, for he did not mention the 1949-1955 floods. He also overlooked the channel widening and thereby missed one of the most important sources of the sediment. As Leopold (1956) pointed out: "Bank-cutting is usually a process of sediment-trading, erosion in one place and deposition in another". Figure 2 compares reconstructed cross-sections of the former channel with cross-sections taken at the same places today and suggests that much of the sand deposited could well have come from demolished banks upstream. There is no doubt that bushfires, burning-off and timber-getting contributed to aggradation of the Macdonald River, but catastrophic sanding-up only occurred between 1949-1955. Farming and related activities were not the primary causes of changes in the channel; these resulted from the cumulative effect of repeated high floods, probably accentuated by a raised floodplain water-table resulting from channel aggradation and the unpredent ed rainfall of 1950. One farmer at Upper Macdonald has said that he found bank erosion worst in places where he had previously noticed "springs" in the banks. Morris (1956), a local resident, described the floods of 1867, 1889 and 1913. He wrote of the floods of 1949-1952 (sic) as follows: "In 1949... the river banks were fringed in places with water gums which were there when Captain Cook landed, and they had withstood the three major floods mentioned as well as intermediate smaller ones by the dozen. But with the series of floods which started in 1949 the banks began to give way and a moderate-sized flood in 1952 took the lot - roots and branches as well as the banks on which they grew, and where the river skirted the mountainside it cut right back to the mountain rocks or hard formation. With the displacement of so much soil throughout the entire length of the cultivated Macdonald valley it followed naturally that the river bed would collect it. It did, and the bed is yards higher today than it was after the flood of 1889". Morris also referred to the permanent lagoons caused by the floods: "The result of this silting of the river bed is that acres of rich low-lying farmlands are now permanent lakes". Hack and Goodlett (1960) described a rainstorm on the Little River, Virginia, so violent that in some places it tore up the entire floodplain and washed away its cover of trees. By a remarkable coincidence it occurred on the same day as the 1949 Macdonald River flood, 18 June 1949. Hack and Goodlett considered that although floods like the Little River flood may have a recurrence interval of 600 years "they occur frequently enough to exceed in importance as erosive agents all intervening floods that do not damage the forest". The floods of 1949-1955 on the Macdonald River illustrate that possibility. The river's uncleared reaches and densely forested tributary gullies by no means went unscathed by the floods. A. Bailey describes large trees lying in all directions along the creeks after the 1952 and later floods; and at the creek heads the natural gibber swamps to which he used to take cattle in dry times had disappeared leaving only boulders and white sand. Other farmers describe tall trees falling as their root systems were undermined by successive floods, and CHANNEL CHANGES IN THE MACDONALD VALLEY 7 say that in many narrow valleys (e.g. Dawes' Run and Gorrick's Run) creek beds had doubled in width after the 1952 flood. FLOODS ON NEIGHBOURING STREAMS 1949-1955 Major flooding occurred on all rivers in east- central New South Wales between 1949 and 1955. On neighbouring Webb's Creek and Mangrove Creek (Fig. lA) the effects of the floods were comparable to those on the Macdonald River. Twenty kilometres up Webb's Creek from Wiseman's Ferry the creek bed aggraded by more than 2 m between 1949-1955 and banks were eroded although less uniformly than on the Macdonald. On Mangrove Creek the 1949 flood was more destructive than later floods and destroyed creek-flats back to the watershed. On June 17 and 18, 1949 Mangrove Mountain at the head of Mangrove Creek had 159 and 197 mm of rain respectively. During each flood on Mangrove Creek large quantities of sand were deposited until by 1955 the creek bed at Upper Mangrove had aggraded by 3 m. It has not aggraded since. Wollombi Brook has a drainage basin nearly equal in area to the Macdonald, and a mean annual discharge 50 percent greater. There was the same destruction of banks along the lower Wollombi as on the Macdonald, but above Wollombi the banks of the South Arm were not seriously eroded, and on the North Arm they were not eroded above Cedar Creek (Page, 1973). Deposition of sand was extremely heavy but appears to have been less heavy than on the Macdonald: at no point did the bed aggrade by more than 2 m during the flood period. Overbank deposition was particularly heavy on the South Arm which has a shallow channel . The Colo valley was less affected than the Macdonald by the 1949-1955 floods. Although the 1949 flood reached the record height of 20 m at Upper Colo, 30 km from the Hawkesbury, it did little damage to the banks, and the channel was not ravaged by ensuing floods to the same extent as the Macdonald. There was heavy deposition of sand in the tidal zone, but the bed seems to have aggraded only by about a metre above Central Colo, and according to local farmers the bridge at Upper Colo stands nearly as high above the bed as when built 40 years ago. Doyle's, Martindale and Baerami Creeks are tributaries of the Hunter or Goulburn Rivers but have a common watershed with the Macdonald or rise near its upper reaches. Their valleys widen where they reach the Coal Measures of the Hunter valley, and by comparison with the other streams mentioned their banks have a high silt and clay content. All three suffered severe bank erosion between 1949 and 1955, but deposition of sand was patchy and nowhere seems to have exceeded 60 cm. Ellis (1971), describing the effects of the floods on Baerami Creek, does not mention deposition of sand. EVIDENCE OF SUBSEQUENT READJUSTMENT OF CHANNEL FORM The channel of the Macdonald River changed little between 1955 and 1970. In 1970 the benches at the foot of the over-steepened banks were still narrow, hummocky and largely ungrassed. Aerial photographs show that there were only minor changes in their size and disposition between 1955 and 1970. The river bed remained wide and flat. Similarly, Page (1973) considered that the most distinctive form-element of the comparable Wollombi Brook was its almost flat bed. Since 1970 there have been the following signs of change: 1. Narrowing Bed In many reaches since 1970 the river has widen- ed the low benches, now levelled and grassed, and in doing so has established a stable course within its overwidened bed. Schumm and Lichty (1963) attributed the reconstruction of the Cimarron floodplain in U.S.A. to, interalia, unvegetated sand-bars becoming attached to banks in places where channels had been abandoned. On the Macdonald, a much narrower river, such a development only seems to be lasting when accompanied by a strong growth of grass. It is by no means clear why the Macdonald River began to narrow its channel again only after 1970. Since it was not due to changed land use or inter- ference with the natural flow of the river, the explanation must presumably be sought in the frequency and distribution of high and low flows. Local farmers account for the widening benches in terms of streamflow: a big flood, they say, demolishes sand-bars and either carries the sand away or spreads it over the bed, whereas a small flood merely lifts it and deposits it on the benches. The data in Tables 1 and 3, however, do not suggest any obvious meteorological or hydrological reason for the changes which have only become apparent since 1970. The explanation may be partly biological. The writer was told by a farmer on Martindale Creek, a tributary of the Hunter River having a common divide with the Macdonald (Fig. lA) that after a minor flood some years ago the channel floor was covered for miles with eucalyptus seedlings which grew several inches before they shrivelled and died. The rapid colonisation of parts of the Macdonald River channel by grass after 1969 may be analogous. In February 1969 a flood, about 3 m high at St. Albans, deposited large quantities of decomposing leaf-mould and other organic detritus in the bed of the lower Macdonald. In mid- 1969 a group from Macquarie University found gas pits nearly one metre in diameter in the Fernance's bridge area; the writer found other smaller pits near Bailey's buried bridge. Methane gas was being emitted from the pits which were found to overlie deposits of organic matter up to a metre deep covered by channel sand. It may be possible that the deposition of exceptional quantities of organic material in the Macdonald's bed in 1969 facilitated the growth of grass during the next four or five years. 2. Degrading Bed In September 1954 the N.S.W. Water Conservation and Irrigation Commission installed a staff gauge at St. Albans and, based on rating tables which have been recalculated at least yearly, cease-to-flow height in 1967 was about the same as in 1954, 8 H. M. HENRY although fluctuations over a range of about 25 cm had occurred. Cease-to-flow height has tended to fall slowly since 1967 but at the St. Albans gauge it appears to be still within 30 cm of its 1954 height. It is presumably not coincidence, however, that pools a metre deep have recently developed in reaches of the river between St. Albans and Central Macdonald which were ankle-deep before 1970, and that floods, including the 6 m flood of June 1974, have not filled them but have deepened and extend- ed them. Other evidence that the river may be lowering its bed is provided by high tides which today flow 1 km further upstream than they did in 1970, and the fact that a reach of the river 1 km below Wright's Creek, across which it was possible to wade at low tide two years ago, is now permanently more than a metre deep. There is one factor which will continue to delay the degrading of the Macdonald's bed, at least near present tidal limits. In adopting a shortened course across the neck of the lower cut- off meander at Central MacdoTiald in 1952, the river became superimposed on a bed of indurated ferruginous sand, the existence of which had not been known to local farmers. After 24 years the river is still incised less than 0.5 m into this bed the upper surface of which is 2 m above the former level of the river. The indurated bed is about 2 m thick, and overlies a deposit of clay with a high organic content, below which there is unconsolidated sand (M.F. Clarke, pers. comm.). Such a nickpoint has impeded the discharge of sand by the river, and probably helped to delay the draining of lagoons below St. Albans; however, pools a metre deep have now developed in the river bed less than 2 km upstream. 3. Future Reduction of Bank Slopes? In discussing the stability of the banks of the George's River near Sydney, Foster and Nelson (1971) considered them to be unstable when their slope exceeded 26 degrees. The Macdonald's banks are generally sandier than those of the George's River (Probert, 1971; Warner and Pickup, 1973), and at eight of the nineteen cross-section sites in Table 3 the slope of the tallest bank segment exceeds 30 degrees. Data in Table 3 suggest that the slope of the tallest segment of unaltered banks usually does not exceed fifteen degrees. It seems likely therefore that many of the flood- steepened banks of the Macdonald will eventually suffer a marked reduction in slope, and this too will contribute to restoring the shape of the former channel. CONCLUSION Amongst well-known papers describing rapid changes in river channels are reports by Smith (1940), Dobbie and Wolf (1953), Hack and Goodlett (1960), Schumm and Lichty (1963) , Stewart and Lamarche (1968) and Burkham (1972). The papers fall into two groups. Dobbie and Wolf (Lynmouth, Devonshire flood of August 1952) Hack and Goodlett (Little River, Virginia, flood of June 1949), and Stewart and Lamarche (Coffee Creek, California, flood of December 1964) describe single events of catastrophic intensity, while Smith (rivers of western Kansas), Schumm and Lichty (Cimarron River, Kansas) and Burkham (Gila River, Arizona) describe river metamorphosis caused by climatic fluctuations in subhumid and semiarid regions. The authors of the second group acknowledge the important role of exceptionalfy high floods in introducing a period of change. In his study of Wollombi Brook, Page (1973) considered that the 1949 flood, which he discusses as a catastrophic event, represented an "initial exceedance of a resistance threshold, after which erosion was continued by less spectacular flows". This would be an oversimplification if applied to the Macdonald River, where the period June 1949 to August 1952 inclusive (two major floods) was the principal sanding-up period, while the overlapping period August 1952 to February 1955 inclusive (also two major floods) witnessed the demolition of river-banks. If there were a catastrophic event on the Macdonald River, it was the August 1952 flood, as Morris (1956) indicates. On the morning after the August 1952 flood a Sydney newspaper reported: "The Macdonald River is now nearly 300 feet (90 metres) wide and a swathe has been cut through most of the farms between Higher Macdonald and St. Albans". (Sydney Daily Telegraph, 19th August, 1952). Burkham (1972) believed that the widening of the channel of the Gila River, Arizona, involving extensive destruction of the floodplain between 1905-1917, and its reconstruction between 1918- 1964, was a repetitive process occurring naturally; he seemed to envisage a cycle recurring at intervals of one or more centuries triggered off by catastrophic floods. There is no way of know- ing whether the 1949-1955 Macdonald River floods represent the destructive phase of such a cycle, but there is some evidence in the aerial photographs of 1941 that the former V-shaped channel was a recovery phase since the outlines of the present banks can be identified at many points. Furthermore the 1955 flood brought evidence to light at Higher Macdonald of a time when the channel may have been as wide as it is today: after the banks collapsed four tree-trunks were revealed in situ side by side, deep within and five metres below the surface of the high bank. Today they occupy a low bench a metre above the river bed. POSTSCRIPT As one of a number of projects to provide water for the Gosford-Wyong region before the end of the century, the N.S.W. Department of Public Works is considering building a 10-raetre-high weir near Fernance's bridge (Department of Public Works, 1975). What proposed engineering ingenuities can justify a plan to build a 10-metre- high weir on a sandbed stream of intermittent flow which aggraded its bed by three metres within six years at the weir site and for 30 km upstream although no artificial structure impeded the river's flow? Perhaps after further consideration the plan will not be proceeded with. However, the proposal may result in serious hydrological and geomorphological work being carried out on the Macdonald River for the first time. Whether undertaken by a government department or private research, it will be regrettable if further changes in the channel of this interesting sandbed stream, so accessible from Sydney, are not adequately monitored. CHANNEL CHANGES IN THE MACDONALD VALLEY TABLE 1 NUMBER OF DAYS BETWEEN OCTOBER 1954 AND DECEMBER 1973 ON WHICH MACDONALD RIVER EXCEEDED 1 METRE, 1.5 METRE AND 3 METRE LEVELS AT ST. ALBANS # Exceeded 1 m 1.5 m 3 m Oct. 1954 to Feb. 1955 18 10 4 1958 19 9 1 1959 11 3 0 Jan. to June 1960 2 0 0 Oct. 1961 to Apr. 1962 SO 22 1 Aug. 1962 to Dec. 1962 6 4 2 1963 43 25 5 1964 13 6 3 1965 3 0 0 1966 1 0 0 1967 26 8 2 1968 8 4 0 1969 24 11 0 June 1970 to Nov. 1970 1 0 0 Feb. 1971 12 9 2 July 1971 to Dec. 1971 1 0 0 May 1972 to Dec. 1972 2 1 0 1973 14 7 1 NOTE 1. The table includes all periods of more than two months duration between October 1954 and December 1973 during which continuous recording was maintained. 2. On only three days between February 1955 and June 1974 did the river exceed the five metre level at St. Albans, viz. 13 May 1962, 29 April 1963 and 10 June 1964. # Based on data supplied by the N.S.W. Water Conservation and Irrigation Commission. TABLE 3 ESTIMATED ANNUAL STREAMFLOW OF MACDONALD RIVER AT ST. ALBANS (million m^)* 1904 166.5 1928 92. 3 1952 404.5 1905 31.0 1929 164.7 1953 203.0 1906 9.6 1930 355.7 1954 96.0 1907 32.8 1931 287.1 1955 618.8 1908 244.8 1932 8.0 1956 442.6 1909 107.7 1933 24.8 1957 4.7 1910 102.6 1934 203. 1 1958 108.8 1911 83.0 1935 7.6 1959 70.5 1912 63.0 1936 17.5 1960 29.5 1913 583. 1 1937 33.2 1961 94.5 1914 62.0 1938 14.7 1962 269.5 1915 47.7 1939 26.9 1963 456. 1 1916 36.4 1940 9.7 1964 189.6 1917 35.0 1941 11.5 1965 13.6 1918 32.3 1942 164.4 1966 4.2 1919 2.6 1943 91.0 1967 132.8 1920 78.3 1944 7.9 1968 38.1 1921 362.1 1945 174.8 1969 153.3 1922 142.7 1946 204.3 1970 36.3 1923 5.5 1947 51.9 1971 171.6 1924 16.8 1948 80.0 1972 117.7 1925 19. 1 1949 504.4 1973 88.1 1926 202.9 1950 1154.4 1927 288.5 1951 314. 3 Average Annual Flow - 150.5 * From New South Wales Department of Public Works, 1975, Table 12. Linear measurements in metres. Slope angles in degrees. CHANNEL CHANGES IN THE MACDONALD VALLEY 11 gure 2. Cross-sections of channel at seven sites included in Table 3. The dashed lines showing e-1949 profiles are based on information from local farmers. 12 H. M. HENRY PLATE 1 Macdonald River at St. Albans bridge about 1910 showing pre-1949 channel (Photograph by Mrs. E. Jurd) PLATE 2 Macdonald River at St. Albans in 1973 showing widened channel and aggraded bed (see bridge pylons). CHANNEL CHANGES IN THE MACDONALD VALLEY 13 PLATE 3 River meander 5 kilometres below St. Albans in 1925 (By courtesy Government Printing Office). PLATE 4 Neck of cut-off meander in 1974 (this is the same meander illustrated in Plate 3) 14 H. M. HENRY PLATE 5 Sternbeck's since buried bridge at Higher Macdonald about 1940 (C. Sternbeck at approach to bridge). (Photograph by Mrs. 0. Sternbeck). PLATE 6 Site of Sternbeck's buried bridge in 1974 (the late K. Sternbeck is standing where C. Sternbeck stands in Plate 5. CHANNEL CHANGES IN THE MACDONALD VALLEY 15 PLATE 7 Aerial photograph of the Upper Macdonald River, 12 November 1971. Mellong area, run 4, No. 59537, reproduced by permission of the National Library of Australia and the Department of Lands, N.S.W. PLATE 8 Macdonald River, Higher Macdonald area, in 1970 (By courtesy Dept, of Lands) Scale 1:14550 16 H. M. HENRY ACKNOWLEDGEMENTS The writer's thanks are due to the Macdonald River farmers for information, photographs and hospitality; to K. Cadogan and H. McCubbin of Prospect County Council for access to survey data from 1833 to the present; to officers of N.S.W. Water and Conservation Commission and the N.S.W. Department of Public Works for infc^rmation; and to Professor J.L. Davies, Drs. R.J. Blong, P.J. Conaghan, W.N. Holland, G. Pickup S.J. Riley and M.A.J. Williams and to M.F. Clarke for invaluable advice and assistance. In conclusion the writer expresses his sorrow that Keith Sternbeck, field assistant and friend, and fourth generation farmer of Higher Macdonald, did not live to see this paper completed. REFERENCES Burkham, D.E., 1972. Channel changes of the Gila River in Safford Valley, Arizona, 1846-1970. U.S. Geol. Surv. Prof. Paper, 655-G. Dobbie, C.H. and Wolf, P.O. , 1953. The Lynmouth flood of August 1952. Proc. Inst. Civ. Eng., 2 (3), 522-588. Dyson, J.R., 1966. Sand deposits in the Macdonald and Colo Rivers. J. Soil Conserv. N.S.W., 22, 158-173. Elkin, A. P., 1955. THE DIOCESE OF NEWCASTLE. Published privately. 827 pp. Ellis, I., 1971. A history of Baerami Creek Valley, in "NJuswellbrook Chronicle" Print. Foster, D.N. and Nelson, R.C. , 1971. Report on bank stability, George's River. Univ. of N.S.W. Water Res. Lab. Tech. Report, 71/12. Galloway, M.C., 1967. Stratigraphy of the Fhatty-Upper Colo area, Sydney Basin, N.S.W. Journ. Proc. Roy. Soc. N.S.W., 101, 23-36. Hack, J.T. and Goodlett, J.C. 1960. Geomorphol- ogy and forest ecology of a mountain region in the Central Appalachians. U.S. Geol. Surv. Prof. Paper, 347. Hickin, E.J. and Page, K.J. , 1971. The age of valley-fills in the Sydney Basin. Search, 2, 383-384. Leopold, L.B. 1956. Land use and sediment yield in MAN'S ROLE IN CHANGING THE FACE OF THE EARTH, pp. 639-647. W. L. Thomas (Ed.). University of Chicago Press, Chicago. Morris, W. , 1956. One hundred years ago. Central Macdonald P. and C. Assoc. (Mimeo) . New South Wales Department of Public Works, 1975. REPORT ON INVESTIGATIONS FOR WATER SUPPLY TO GOSFORD-WYONG REGION. REPORT NO. 1. 3 vols. Page, K.J. , 1973. A field study of the bankfull discharge concept in the Wollombi Brook drainage basin, N.S.W. M.A. (Hons.) Thesis, Univ. Sydney. (Unpub) ) Pickup, G. , 1974. Channel adjustment to changed hydrologic regime in the Cumberland Basin, N.S.W. Ph.D. Thesis, Univ. Sydney. (Unpubl.). Probert, D.H. , 1971. Sand deposits in the Macdonald and Colo River areas. Geol. Surv. N.S.W. , 1971/703. Schumm, S.A. and Lichty, R.W., 1963. Channel widening and floodplain construction along the Cimarron River in southwestern Kansas. U.S. Geol. Surv. Prof. Paper, 352-D. Smith, H.T.U., 1940. Notes on historic changes in stream courses of western Kansas, with a plea for additional data. Trans. Kans. Acad. Sci. , 43/299. Stevens, M.A. , Simons, D.B. and Richardson, E.V. , 1975. Nonequilibrium river form. Journ. Hydro. Div. Proc. Araer. Soc. Civ. Eng. , 101/ HY5, 557-566. Stewart, J.H. and Lamarche, V.C., 1967. Erosion and deposition produced by the flood of December 1964 on Coffee Creek, Trinity County, California. U.S. Geol. Surv. Prof. Paper, 422-K. Warner, R.F. and Pickup, G. , 1973. Channel deterioration in the George' s River between Liverpool Weir and Little Salt Pan Creek. Pub. Report, Ban)cstown Municipal Council. 11 Fifth Avenue, Cremorne, N.S.W., 2090. (Manuscript received 3.9.76). Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 17-24, 1977 Does the Hunter River Supply Sand to the New South Wales Coast Today? P, S. Roy* ABSTRACT. The possibility that terrestrial sand is being added to the open coast by the Hunter River was investigated by sediment sampling. Medium-fine quartz sand infills the lower 9.5 km of the Hunter estuary. This deposit is the product of a now largely inactive, landward transport of marine sand from the open coast. Lithic river sand of terrestrial origin occurs upstream of the marine sand and becomes finer in a seaward direction. Minor amounts of very fine river sand are admixed with coarser marine sand in the lower estuary. Superficial deposits of terrestrial mud also occur in those parts of the lower estuary that have been dredged. Very fine river sediment is periodically flushed out to sea by floods. It does not accumulate in the high energy nearshore environment but settles in deeper water further offshore. Under present day conditions the Hunter River does not contribute significant amounts of sand to the adjacent beaches or nearshore zone. INTRODUCTION One factor in the present day sediment budget of sandy shoreline deposits (or "sweep zone" of Davies 1974) is the supply of river sand to the coast and its addition to the beaches. Bowen and Inman (1966) have shown the importance of this factor in California, where the construction of dams on coastal rivers has reduced the contribut- ion of terrestrial sand to the coast and caused a deficit in the sediment budget. In contrast. Ford (1963) suggested that rivers are not supplying sand to the New South Wales coast and "that the prevailing movement (of sand) is from the ocean into the entrances". Work by the Geological Survey of New South Wales in a number of estuaries in New South Wales (Roy and Peat, 1973 and 1975 a, 'b and in press) supports Ford's conclusion. On the New South Wales coast, the composition and texture of the estuarine sands have been found to be a useful indicator of their source, whether marine or terrestrial. Most of the New South Wales studies, however, were of estuaries with relatively small river inflows. The possibility therefore exists that, during floods, larger coastal rivers may carry terrestrial sand to the sea. The present day distribution of sediments in the coastal zone reflects processes operating during the Late Quaternary (Roy and Thom 1975). Throughout this time, processes were strongly influenced by glacio-eustatic events (Chappell 1974), especially the last two sea level fluctuations. During the most recent rise in sea level (the Post-glacial Transgression which culminated in the sea reaching its present level about 6,000 years ago - Thom and Chappell 1975) the coastal river valleys were drowned to form estuaries. There followed a phase of estuary infilling by both marine and fluvial processes. Marine sediments formed coastal barriers and tidal deltas in the estuary mouths, while river sediments were deposited in the central and upper reaches of the estuaries. During this phase of estuary infilling, the supply of terrestrial sand to the coastline is negligible. The stage of infilling reached today is mainly a function of the sizes of the original embayments and the supply of river detritus, the latter being related to river discharge. Many of the coastal rivers, especially the larger ones, have almost completely infilled their embayments with river sedments (producing flood plains) which have prograded over older estuarine and marine deposits. Davies (1974) argued that once the estuary is filled with sediment, river sand may be transported through the flood plain and delivered to the coast. However, river sand, to be deposited in the shallow near- shore zone and incorporated in the beaches, must be sufficiently coarse grained to attain dynamic equilibrium with the high-energy wave conditions of the open coast. The Hunter River, with its well developed flood plain, may have reached this stage. This report summarizes the results and conclusions of an investigation of the sediments in the lower Hunter River designed to explore the possibility that it currently supplies sand to the coast. The Hunter River is located 120 km north of Sydney. It is the fifth largest river (in terms of discharge) on the New South Wales coast with an average discharge of approximately 58 m^/sec (2000 cusecs) (Water Conservation and Irrigation Commission of N.S.W., 1971). The river is tidal for a distance of approximately 55 km from its mouth (Department of Public Works, N.S.W., tidal data, 1956 - plan 201/141). The port for Newcastle, the second largest city in N.S.W. , lies at the mouth of the Hunter River. Breakwaters and port facilities have been constructed, and the lower estuary has been dredged since 1905 (King 1911) to reclaim land and improve navigation. Consequently, much of the river bed in the port area reflects man's interference. SAMPLING Surface sediments were dredged at forty-three sites in the river channel between the river mouth and Raymond Terrace, 30 km upstream. Two samples were collected from Newcastle Bight outside the harbour mouth and two samples were taken from * This paper was submitted with the permission of the Under-Secretary, New South Wales Department of Mines. 18 P. S. ROY Figure 1 Bottom sediment sampling locations in the lower Hunter River DOES HUNTER RIVER SUPPLY SAND TO N.S.W. COAST TODAY? 19 TABLE 1 HUNTER RIVER AND NEARSHORE SEDIMENTS - SAMPLE COMPOSITION AND SOURCE (OF SAND COMPONENT) CLEAN SAND MUDDY SAND SANDY MUD MUD 10% MUD 10-50% MUD 50-90% MUD 10% SAND 2 (T) 26 (M*) 13 (T) 4 (T) 27 (M*) 17 (T) 5 (T) 30 (M) 20 (T) 6 (T) 31 (M*) 21 (T) 7 (T) 33 (M*) 22 (T) 8 (T) 37 (M*) 29 (M*) 9 (T) 38 (M*) 10 (T) 41 (M*) 50 (M*) 11 (T) 44 (M) 52 (T*) 12 (T) 45 (M) 14 (T) 46 (M) 15 (T) 47 (M) 16 (T) 19 (T) 23 (T*) 25 (T*) 28 (T*) 39 (T*) 40 (T) 1 (T) 3 (T) 18 (T) 35 (T) 24 (T*) 36 (T) 32 (T) 42 (T*) 34 (T) 43 (T) 49 (T) 54 (T) 48 (T) 55 (T) 51 (T*) 56 (T) 53 (T) 28 = sample location (Figure 1) (T) = terrestrial sand (M) = marine sand (T*) = mainly terrestrial with minor marine sand (M*) = mainly marine with minor terrestrial sand (N.B.: Mud component in sediment is of terrestrial origin) dredge spoil dumps on Kooragang Island. At a later date nine additional samples were collected near the estuary mouth 2 days after the peak of a 1 in 3 year flood. The flood reached Maitland, 45 km upstream of Newcastle, on the 26th January, 1976; it resulted in a 10 m rise in river level and a maximum discharge of 173,000 mega 1/day (J. McGlyn, W.C. and I.C., pers. comm.). Sample locations are shown in Figure 1. Also examined were drill hole samples from beneath Newcastle Harbour supplied by Coffee and Hollingsworth Pty. Ltd. and Dames and Moore Pty. Ltd. , consulting engineers. Sediment data from Stockton Beach were provided by Cheng Ly (Department of Geology, University of Newcastle, pers. comm.) and A. Gordon (P.W.D. , pers. comm.). Sand:mud (silt plus clay) ratios were determined and the sands were examined under a binocular microscope to estimate composition and texture. Detailed granulometric analyses were carried out on samples 1 to 47 by Cheng Ly (Ly, in prep.). Standard grainsize analysis of samples 48 to 56 was undertaken by the P.W.D. Hydraulics and Soils Laboratory. The samples are described in the appendix. SEDIMENT TYPES Grain size is largely a function of the depositional energy level. Very muddy sediments accumulate under quiet conditions; clean sand indicates higher energy levels. Clean sand (less than 10 per cent mud) comprises 56 per cent of the samples, muddy sand (10 to 50 per cent mud) comprises 14 per cent, sandy mud (50 to 90 per cent mud) comprises 16 per cent, and mud (less than 10 per cent sand) comprises 14 per cent. The composition and texture of the sand grains indicate the source of the sediment, whether terrestrial or marine. Marine sand, typified by that occurring on the ocean beach, is composed of medium sized, subrounded to well rounded quartz grains with less than 20 per cent lithic fragments; shell fragments are always present. River sand is composed of terrestrial detritus rich in lithic (rock plus feldspar) fragments; lithic content ranges from 50 to 70 per cent. Grains are angular to subangular and vary in grainsize from coarse to very fine. Shell fragments occur rarely and, in the Hunter River, were only found in samples seawards of Hexham. The heavy mineral suite in the marine sediment is dominated by rutile, zircon, ilmenite and leucoxene. In contrast, the river sediment is characterised by pyroxene, hornblende, epidote and magnetite (Ly in prep.) The samples are categorized in Table 1. All muds, sandy muds and, with two exceptions (samples 29 and 50) muddy sands, are composed dominantly of terrestrial material. Of these twenty-five samples, four contain traces of marine sand (samples 24, 42, 51 and 52), while the sand in 20 P. S. ROY samples 29 and 50 is predominantly marine. The remaining thirty-one samples are clean sands of which fifteen are of terrestrial origin, and five are of marine origin. The remaining eleven samples contain mixed marine and terrestrial material; of these seven are mainly marine and four are mainly terrestrial. SEDIMENT DISTRIBUTION The sediments sampled in the lower Hunter River in the pre-flood period fall into two groups that occupy different parts of the estuary. (The sediment distribution altered somewhat during the flood and is described later). Downstream from Raymond Terrace, river sediments occupy the channel to within 9.5 km of the sea. Seawards of this point sediments of marine origin dominate in the Hunter estuary and occur on the ocean beach and off the river mouth (Figure 2). Clean or muddy river sand is the dominant sediment type downstream of Raymond Terrace as far as the mouth of Fullerton Cove in North Arm Channel and almost half-way down the length of Kooragang Island in South Arm Channel. Mud or sandy mud occurs only rarely in this sector. A superficial layer of mud occurs in a 19 m deep scour hole at the junction of the Hunter and Williams Rivers. Sandy mud was found in the bed of the Williams River (sample 1) and at two sites near the bank of North Arm Channel (samples 18 and 24). The river sands show a progressive decrease in grainsize from medium - fine to very fine in a downstream direction. This grainsize trend is most apparent in North Arm Channel. In South Arm Channel the sands are slightly coarser, indicating a somewhat higher energy regime. Downstream oriented sand waves occur commonly above Kooragang Island and indicate a downstream direction of sediment transport during the ebb tidal flow as well as during river floods. The sediment in the lower 9.5 km of the estuary is mainly clean or slightly muddy marine sand which incorporates, as a minor component, very fine river sand. Clean, medium-grained marine sand (with no trace of a river component) was found offshore (samples 45 and 46) and also in dredge spoil on Kooragang Island (samples 30 and 47). The marine sands in the estuary are virtually identical in composition and texture to those in the nearshore zone and on Stockton Beach. As well as marine sand, mud and sandy mud were also encountered in the harbour area. The sand component of these muds is very fine grained and is of terrestrial origin. Drilling in tne harbour area penetrated up to 8 m of soft mud and fine sandy mud in parts of the channel that have been deepened by dredging. Shelly marine sands occur beneath the mud in these areas and overlie a stiff, weathered clay deposit of Pleistocene age. Figure 2. Generalised sediment distribution in the lower Hunter River and Newcastle Bight. DOES HUNTER RIVER SUPPLY SAND TO N.S.W. COAST TODAY? 21 DISCUSSION The predominance of sand sized sediment in the Hunter River estuary is characteristic of a depositional environment with moderately high energy levels. The sand in the upstream part of the estuary is of terrestrial origin. Its progressive fining in a downstream direction suggests a sorting mechanism associated with its gradual seaward transport. Its texture is thought to reflect normal estuarine conditions (tidal and river flow) rather than catastrophic events such as floods. A marked difference exists between the composition of the sediments in the Williams River (sandy mud) and the Hunter River (sand) immediate- ly above their junction. Because of its greater catchment area and discharge, the Hunter River is the major contributor of terrestrial sand to the estuarv, while the much smaller Williams River mainly supplies mud. The composition of the marine sand in the lower reaches of the estuary, especially the presence of estuarine shell species, indicates an estuarine channel or tidal deltaic environment of deposition. The shelly marine sands encountered by drilling beneath the dredged bed of the harbour belong to the same depositional unit. Tidal currents and waves are mainly responsible for the landward movement of marine sand from the open coast into the estuary mouth, although aeolian processes may have contributed some sand from the adjacent dunes at Stockton in the past. Fluorescent tracer tests conducted in Newcastle Harbour mouth by the P.W.D. (Boleyn and Campbell 1966) showed no evidence that marine sand was moving into the river mouth during the period of testing. The marine sand in the lower estuary is presumably a relict deposit, although still being reworked by currents at present. A compositional difference appears to exist between the surface and subsurface marine sands in the lower estuary; superficial samples contain a minor component of fine terrestrial sand, while sand dredged from beneath the channel bed and used to reclaim Kooragang Island is completely free of terrestrial sand (compare samples 26, 28, 30, and 37 from the river bed with samples 30 and 47 of spoil dredged from subsurface). This upward change from marine to more terrestrial sediment is in accord with the geological model of estuary infilling proposed above and reflects the gradual seaward migration of terrestrial sediments over older marine deposits. Siltation studies in Newcastle Harbour by the P.W.D. (1969) indicate that shallow areas, such as Fullerton Cove, act as settling basins for suspended sediment during floods. The mud is remobilized by wave turbulence during non-flood periods, and currents redistribute it throughout the port area. Here it accumulates under low- energy conditions in navigation channels and docks that have been deepened by dredging. These dredged areas act as traps for sediment moving seaward down the channels. Samples of mud collected in the port area contain traces of fine terrestrial sand indicating that some fine river sand is transported into the lower estuary. However, the almost complete absence of medium- TABLE 2 COMPARISON BETWEEN PRE-FLOOD AND FLOOD SAMPLES FROM THE LOWER HUNTER RIVER (SAMPLE LOCATIONS SHOWN IN FIGURE 1) Pre-Flood Flood 35 Mud 54 Mud 32 Sandy Mud 55 Mud 33 Mainly marine Sand 56 Mud 34 Very sandy Mud 53 Sandy Mud 41 Mainly marine Sand 52 Mainly muddy lithic Sand 43 Sandy Mud 51 Slightly sandy Mud 44 Marine Sand 49 Mud 50 Mainly muddy marine Sand 48 Sandy Mud 45 Marine Sand grained river sand in the mud, even at depth, suggests that very little sand of this size is reaching the estuary mouth. (By the same reasoning it would also appear that little marine sand from the open coast is moving landwards up the river channels) . Undoubtedly fine sediment in the lower Hunter River is flushed out to sea during floods. Samples collected near the river mouth during the flood in January 1976, when compared with pre-flood sediments from similar sites, show an increase in mud content and contain more fine to very fine river sand (see Table 2). The medium to coarse sand in some samples is predominantly marine in composition and is presumably derived from reworking of marine deposits in the lower estuary. The presence of this mainly fine sediment on the channel bed, despite the undoubtedly higher than normal current velocities, reflects a disequilibrium between the sediments and the hydrodynamic processes and confirms a seaward transport of sediment under these conditions. However, sediment transport under extreme flow conditions (ie. 1 in 100 year flood) cannot be extrapolated with confidence from this data. On Stockton Beach and in the nearshore zone, energy levels are too high for fine sand and mud to accumulate. Here the beach sands range from medium to coarse grained. Fine river sediment presumably moves farther offshore and is deposited in deeper water where energy conditions are sufficiently low. Sampling on the inner continent- al shelf off Nejvcastle by Shirley (1964), the Bureau of Mineral Resources (Davies and Marshall 1972) and Boyd (1974) has delineated a zone of carbonate-rich mud and sandy mud between water depths of 60 m and approximately 120 m, lying seawards of a nearshore zone of clean quartz (Figure 2). Boyd reasoned from grainsize data that the Hunter is the source of the shelf mud. An association between the sandy mud in the Hunter River and that on the continental shelf is supported by the mineralogical similarity of their clastic fraction (Byrnes and Holmes 1975). However, the question of when the shelf mud was deposited is 22 P. S. ROY unresolved. The possibility that the shelf mud was derived from the Hunter River in the past and accumulated as estuarine deposits during period(s) of lowered sea level cannot be discounted on sedimentological evidence alone (Davies and Boyd 1975). The composition of the sand forming Stockton Beach provides further data on the possible input of terrestrial sand to the coast. Studies by the P.W.D. (A. Gordon pers. comm.) show no evidence of a net littoral drift along the beach. Consequent- ly, a significant input of terrestrial sand would be expected to produce a recognizable change in beach sediment near the river mouth. Detailed granulometric and mineralogical analyses of beach sands by Cheng Ly show a number of trends : (i) The proportion of angular and very angular grains increases at the northeastern end of the beach. (ii) Lithic content increases slightly at both extremities. (iii) Heavy minerals are typically marine in character except at the north- eastern end of the beach where 'terrestrial' minerals such as pyroxene, hornblende, epidote and magnetite are dominant. The changes at the northeastern end of the beach are attributed to marine erosion of nearby Carboniferous volcanic rock outcrops. At the southwestern end of the beach however, the absence of parameters characterizing river sediment (ie. , significant quantities of terrestrial heavy minerals, angular grains, and rock fragments) is regarded as evidence against a significant supply of river sand to the coast. CONCLUSION Relict deposits of shelly, quartz- rich marine sands occur in the lower 9.5 km of the Hunter River estuary. These deposits were derived from the open coast and transported into the river mouth by waves and tidal currents during, and since, the Post-glacial Transgression. It is probable that this process is no longer active. Lithic river sands, derived mainly from the Hunter River rather than from the Williams River, occur in the upper estuary landwards of the marine sands. Sorting processes associated with the slow seaward transport of the river sand are thought to be responsible for its progressive fining in a downstream direction. Minor amounts of very fine river sand occur in deposits of mud and in the surface layer of the marine sand in the lower estuary. However in the subsurface, marine sands contain virtually no terrestrial material. This upward change in composition is probably the result of the gradual seaward progradation of younger fluvial deposits over marine sediments as the Hunter River estuary infills. To date the infilling process is incomplete. Under present day conditions, mud and very fine river sand accumulate in the lower estuary and are periodically flushed out to sea by river floods. The high-energy conditions in the nearshore zone preclude the deposition of this river sediment which is presumably carried further offshore and deposited in deeper water. The absence of a significant build up of river sediment in Stockton Beach near the river mouth is further evidence against a significant supply of medium-grained river sand to the coast in the recent past. Historical studies of shoreline changes along Newcastle Bight indicate a general erosional trend (B.G. Thom, University of Sydney, pers. comm.). The loss of beach sand is mainly attributed to aeolian processes and the inland migration of mobile transgressive dunes. In terms of sediment budget, erosion of Stockton Beach reflects an imbalance between gains and losses of sand to the coastal compartment. This study argues that the budget imbalance is at least partly due to the inability of the Hunter River to supply significant quantities of terrestrial sand to the coast. REFERENCES Boleyn, D.A. and Campbell, B.L. , 1966(7). Littoral drift in the vicinity of Newcastle Harbour Harbour entrance. Rep. Dept. Public Works N.S.W. Hydraulics and Soils Lab., 11. Bowen, A.J. and Inman, D. L. , 1966. Budget of littoral sands in the vicinity of Point Arguello, California. Tech. Mem., Coastal Engineering Research Centre, 19, 1129-1173. Boyd, R. , 1974. A marine geological investigation of the N.S.W. coast between Port Stephens and Norah Head. B.Sc. Hons. Thesis, Vniv. Sydney (unpub 1 ) . Byrnes, J. and Holmes, G. , 1975. Sediment samples offshore from and along the Hunter River. Rep. Geol. Surv. N.S.W., Miner alog. , 1975/41 (GS1975/246) (unpubl.) Chappell, J., 1974. Geology of coral terraces, Huson Peninsula, New Guinea: A study of Quaternary Tectonic Movements and sea level changes. Bull. Geol. Soc. Am., 85, 553-570. Davies, J.L., 1974. The coastal sediment compartment. Austr. Geogr. Studies, 12, 139- 151. Davies, P.J. and Boyd, R. , 1975. Variations in sediment texture on the continental shelf off Newcastle. Search (in press). Davies, P.J. and Marshall, J.F., 1972. B.M.R. Marine Geology cruise in the Tasman Sea and Bass Strait. Rec. Bur. Miner. Resour. Geol. Geophys. Aust. , 1972/93. Ford, A.R., 1963. River entrances of N.S.W. J. Inst. Eng. Aust. , 35, 313-320. King, C.W. , 1911. Sand-movements at Newcastle Entrance, New South Wales. Proc. Inst. Civil Eng. , 184, (2), 149-156. Ly, Cheng K. , . Depositional and mineralogical studies of Quaternary sedimentation in the DOES HUNTER RIVER SUPPLY SAND TO N.S.W. COAST TODAY? 23 APPENDIX SAMPLE DESCRIPTIONS Composition (%) Texture of Sand Fraction odinjp 1 c Mud Terrestrial Sand Marine Shell § Wood Grainsize Sorting Angularity 1 80 19 0 1 vf-m P A 2 10 89 0 1 f-m M A 3 100 0 0 0 - - - 4 0 100 0 0 m MW SA 5 3 97 0 0 f-m MP A- SR 6 0 100 0 0 f MW S-SR 7 2 96 0 2 m W A-SR 8 0 100 0 0 f MW SA 9 0 100 0 0 f-vf VW A 10 3 97 0 0 f W A 11 4 96 0 0 f MW A-SA 12 10 90 0 0 f MW A-SA 13 40 60 0 0 f-vf P A 14 3 92 0 5 f-m P A-SR 15 0 90 0 10 f-c P SA-SR 16 5 95 0 0 f W A-SA 17 18 82 0 0 f VW A-SA 18 80 17 0 3 vf w A 19 3 96 0 1 f w A-SA 20 30 65 0 5 f-vf MW A-SA 21 20 75 0 5 f-vf W A 22 40 57 0 3 vf w A 23 10 18 2 70 f-vf MW SA 24 80 13 4 3 f-vf MP A-SR 25 10 88 1 1 f-vf W A-SA 26 0 18 70 12 f-m MP SA-WR 27 0 5 80 15 m M R-WR 28 10 70 12 8 vf-m P A-SA 29 15 5 70 10 f-m MP SR-R 30 0 0 99 1 f-m M SR-WR 31 2 7 85 6 f-m M SR 32 70 25 0 5 vf VW A 33 0 10 74 16 m-c MP R-WR 34 50 40 0 10 vf VW A 35 98 2 0 0 vf MW A 36 98 2 0 0 vf W A 37 0 5 93 2 m MW SR-R 38 2 48 48 2 m MW SA-R 39 5 46 46 3 f-m M A-R 40 3 95 0 2 f M A-SA 41 5 25 65 5 f-m MP SA-R 42 95 3 1 1 vf W A 43 50 35 0 15 vf M A 44 3 2 70 25 f-m P SR-WR 45 0 0 97 3 f-m W SR-WR 46 0 0 90 10 f-m MW SR-R 47 0 0 100 0 m MW SR-R Sand Textural Symbols Grainsize Sorting Angularity vf very fine P poorly A angular f fine MP moderately poorly SA subangular m medium M moderately SR subrounded c coarse MW moderately well R rounded W well WR well rounded VW very well 24 P. S. ROY APPENDIX (CONTINUEDI FLOOD SAMPLES Composition Texture of Sand Fraction Mud Terrestrial Marine Shell 6 Grainsize Sorting Angularity Sand Wood 48 87 8 0 5 f-Tn P A-SA 49 94 3 0 3 f MW A-SA 50 25 10 58 7 f-c P SA-R 51 90 2 6 2 f-m P SA-SR 52 19 44 23 14 f-c P A-R 53 77 20 0 3 f-m P A-SA 54 98 2 0 0 vf W A 55 96 3 0 1 vf MW A 56 98 2 0 0 vf W A Sand Textural Symbols Grains! ze Sorting Angularity vf - very fine P poorly A - angular r - tine MP moderately poorly SA subangular m - medium M moderately SR subrounded c - coarse MW moderately well R rounded W well WR well rounded VW very well Newcastle - Port Stephens area of N.S.W. ph.D. Thesis, Univ. Newcastle, Newcastle (in prep.) P.W.D., 1969. Newcastle Harbour siltation investigation. Rep. Dept. Public Works N.S.W. , 114. Roy, P.S. and Peat, C. , 1973. The bathymetry and bottom sediments of Tuggerah, Budgewoi and Munmorah Lakes and the subsurface stratigraphy of Tuggerah Lake. Rep. Geol. Surv. N.S.W. , GS 1973/285 (unpublished). Roy, P.S. and Peat, C. , 1975a. Bathymetry and bottom sediments of Lake Macquarie. Rec. Geol. Surv. N.S.W., 17 (1), 53-64, 5 figs. Roy, P.S. and Peat, C. , 1975b. Bathymetry and bottom sediments of Lake Illawarra. Rec. Geol. Surv. N.S.W., 17 (1), 65-79, 6 figs. Roy, P.S. and Peat, C. , . Bathymetry and bottom sediments of Tuross estuary and Coila Lake. Rec. Geol. Surv. N.S.W. (in press) . Roy, P.S. and Thom, B.G., 1975. The N.S.W. shelf and coast: a model for development in the late Quaternary. Geology Abstracts ANZAAS, 46th Congress. Shirley, J. 1964. An investigation of the sediment on the continental shelf of N.S.W., Australia. J. Geol. Soc. Aust. , 7, 331-341. Thom, B.G. and Chappell, J. , 1975. Holocene sea levels relative to Australia. Search, 6 (3), 90-93. Water Conservation and Irrigation Commission, N.S.W., 1971. Water Resources of N.S.W. Government Printer, Sydney. Marine Geology Sub-Section, Geological Survey of N.S.W., Department of Mines, CAGA Centre, Bent Street, Sydney, 2000. (Manuscript received 11.10.76) Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 25-27, 1977 A Late Devonian Palaeomagnetic Pole for the Mulga Downs Group, Western New South Wales B. J. J. Embleton ABSTRACT. The thermal stability of the magnetism in specimens from fourteen oriented core samples of red-beds belonging to the Mulga Downs Group was satisfactorily tested using the progressive partial thermal demagnetization technique up to 620°C. Magnetic remanence considered primary and acquired at the time of rock formation was isolated in twenty-one specimens from twelve samples: they yield a palaeomagnetic pole situated at 54°S, 96°E (A95 = 11°). This supports an earlier result obtained from the Middle Devonian Housetop Granite of Tasmania but differs substantially from results obtained from the Lochiel Formation of southeastern New South Wales. The result from the Mulga Downs Group is interpreted as emphasizing the discrepancy between the Palaeozoic palaeomagnetic results from the main Australian platform, and those from the Lachlan Fold Belt. INTRODUCTION Palaeomagnetic investigations of Early Palaeo- ozoic rock formations have provided us with considerable data on which to base apparent polar wander models for the main platform of the Australian continent. Those data and some inter- pretations have been reported by McElhinny and Luck (1970), Luck (1972), Embleton (1972a, b), Embleton and Giddings (1974) and McElhinny and Embleton (1974). In addition, results obtained from Siluro-Devonian rock formations sampled in the southeastern region of Australia, reported by Briden (1966) and Luck (1973), have been inter- I preted within the framework of the plate tectonic j hypothesis. McElhinny and Embleton (1974) and I Embleton et at, (1974), following the earlier I interpretations of the tectonic evolution of the Tasman Orogenic Zone by Oversby (1971) , Solomon and Griffiths (1972) and Scheibner (1972), postu- lated that a region in southeastern Australia (the southern part of the Canberra-Molong High) though still rather poorly defined has undergone rigid rotation with respect to the main platform. Serpentinites which outcrop in a 'belt' from Kiandra in the south (the Coolac Serpentinite, Ashley et at., 1971) and northward to Nyngan were regarded as a possible remnant of the Late Palaeozoic (Devonian?) suture. On the plate tectonic model, basic and ultrabasic rocks associated with serpentinites may represent fossil oceanic crust emplaced originally as ophiolite (Coleman, 1971) during a period of either subduc- tion or obduction (Dewey and Bird, 1971) . Much of the interpretation was carried out by matching the Siluro-Devonian palaeomagnetic pole positions for the main platform and southeastern regions. A recent alternative model by Burns and Embleton (1976) explains the discrepancy in the palaeomagnetic data between the main Australian platform and the southeastern region in terms of internal deformation within the Lachlan Fold Belt, without attaching specific significance to the presence of the ultramafic and serpentinite bodies. The significance of the pole position obtained by Briden (1967) from the Middle Devonian Housetop Granite of Tasmania was uncertain within the plate tectonic scheme. It was originally claimed that the granite had suffered a Tertiary remagnetiza- tion, although McElhinny and Embleton (1974) have given an equally plausible explanation of the magnetization being of Devonian age. Support for that pole position now comes from some Upper Devonian sediments from western New South Wales. How these data fit the palaeomagnetic drift scheme remains uncertain, they in fact may emphasize the over-simplicity of the McElhinny-Embleton model. The results do serve to reinforce the principal thesis that the configuration of the Devonian geomagnetic field varies considerably from south- eastern regions to regions further west. PALAEOMAGNETIC RESULTS Oriented rock samples were collected using techniques described by Collinson et at. (1967, Chap. 1) from three localities in western New South Wales: (i) seven block samples from a section through conglomerates and coarse-grained white/grey sandstones at Mount Manara, about 60 km north of Ivanhoe, (ii) five block samples from a section of white and buff-coloured medium-coarse sandstones at the southern end of the Manara Hills, about 15 km southwest of Mount Manara, and (iii) fourteen core samples collected with a portable hand drill from a 20-30 m escarpment outcrop about 8 km northeast of Milton Grove Station (latitude 32 .8°S, longitude 143. 5°E). Only samples from locality (iii) retained measureable stable components of magnetization. None of the samples from localities (i) and (ii) produced repeatable results upon partial thermal demagnetization. The beds sampled near Milton Grove Station trend approximately east-northeasterly at 65° and dip to the northwest at 30° . This sequence of red sediments belongs to the Mulga Downs Group, generally regarded as Late Devonian in age. Roberts et at. (1972) cite evidence from Ritchie (1969) which supports a Middle Devonian age for the lower part of the succession. The highest beds may be as young as Early Carboniferous. Two specimens were cut from each core and il 26 B. J. J. EMBLETON TABLE 1 SUMMARY OF DIRECTIONS OF REMANENT MAGNETISM FOR SEDIMENTS OF THE MULGA DOWNS GROUP Treat- ment Dec° Inc° N R k “95° NRM 041 -65 28 22.60 5.0 13 300°C 041 -61 28 22.48 4.8 14 400°C 042 -62 28 22.52 4.9 14 500°C 044 -61 28 22.26 4.7 14 550°C 051 -60 28 22.89 5.2 13 600°C 056 -75 28 19.14 3.0 19 620°C 047 -67 28 19.83 3.3 18 Pole position based on the mean of 21 specimen virtual geomagnetic poles calculated after treatment at 550°C (see text): N = 21, R = 19.87, Dec = 042, Inc = -67, 095 = 8°, Lat. = 54°S, Long. = 96°E, Ag5 = 11°. NRM - natural remanent magnetization, Dec - declination, Inc - Inclination, N - number of unit vectors of which R is the resultant , k - the Fisher (1953) precision parameter, and “95(^95) - the semi-angle of the cone of confidence calculated at the 95% probability level (P - O.OS). their directions and intensities of remanent magnetism measured before treatment (NRM in Table 1) and after partial thermal demagnetization at six steps between 300°C and 620°C. The step-wise thermal demagnetization technique is described in Collison et Chap. 3). Specimen direc- tions of magnetization were shown to change little upon treatment and the intensity of magnetization decreased smoothly (see Fig. 1) as would be ex- pected for a fairly stable magnetic component. Table 1 lists the statistics (following Fisher, 1953) for each population obtained after successive stages of thermal demagnetization. The scatter Figure 1. The average demagnetization-intensity curve for the Mulga Downs Group red sediments. The bars indicate the standard error of the mean intensity obtained after treatment at the temperatures indicated. marginally decreased at 550°C, although no statisti- cal significance is attached to that. However, several specimens remain oblique to the principal population and those at an angular distance > 45° from the group mean (see Fig. 2) are considered intermediate and have probably retained substantial components of secondary magnetization: they were omitted from further calculations. Specimens from two samples were oppositely magnetized; they were inverted to obtain mean directions for the group. Before the bedding correction was applied, the mean direction of magnetization, dec. = 6°, inc. = -46°, “95 ” ®°> significantly different (18.1°) from the local present field direction, dec. = 9°, inc. = -64°, though the difference between the mean and the dipole field direction, dec. = 0°, inc. = -52° was only 7.2° (calculated for the 95% confidence level of the mean). However, the presence of a polarity reversal in the rock sequence is strong evidence for the stability of magnetization and that it is primary. Specimen virtual geomagnetic poles were averaged to yield the formation palaeomagnetic pole position: it lies at 54°S, 96°E (A95 = 11°). N Figure 2. The distribution of specimen directions after partial thermal demagnetization at 550°C. Those directions distinguished with a cross lie > 45° from the popula- tion mean (indicated by an open star) . A stereographic (equal angle) projection is used, open (closed) symbols denote upward (downward) pointing directions. The circle of confidence for the 95% probability level is also shown. The primitive represents the bedding plane. TECTONIC IMPLICATIONS The result is consistent with an earlier result for the Devonian obtained by Briden (1967) from a study of the Housetop Granite of Tasmania, viz. 67° S, 94°E. Embleton and Giddings (1974) have previously suggested that Tasmanian rock formations provide palaeomagnetic data which relate to the main plat- form of Australia. For example, the pole yielded PALAEOMAGNETISM OF DEVONIAN SEDIMENTS 27 by the Upper Cambrian Dundas Group (Briden, 1967) falls in a group of Cambro-Ordovician poles obtained from localities widely spread across the continent. However, Siluro-Devonian rock forma- tions in the southeastern mainland region of the Tasman Orogenic Zone yield results which have been regarded as contrary to a pattern of tectonic evolution based on the premise of a 'unified continent' during the Early and mid-Palaeozoic era. The model proposed by McElhinny and Erableton (1974) and detailed by Embleton et at. (1974) implies suturing along a line east of the Wagga Metamorphic Belt sometime during the Devonian. The presence of relatively undisturbed Upper Devonian sediments in the region of the Bogan Gate Platform would further suggest that tectbnic events affect- ing the area were more or less completed before sedimentation. Thus, on the plate tectonic model, the apparent inconsistency between the Late Devonian poles from (a) the Mulga Downs Group (presented here) and (b) the Lochiel Formation sampled in the vicinity of Nethercote (pole posi- tion is 320°E, S8°S) suggests the development of a protracted period of relative 'block' motion which ultimately involved the most easterly regions. The geological evidence on which the Burns and Embleton (1976) model is based, explains the post- Late Devonian relative movement between approxi- mately the Wagga Metamorphic Belt and the eastern coastal regions in terms of a Kanimblan (Early Carboniferous) deformation across the whole of eastern New South Wales. ACKNOWLEDGMENTS The samples were collected while the author was a faculty member of the Research School of Earth Sciences, Australian National University, with the assistance of Dr M.W. McElhinny, Mr C. Barton and a visiting fellow Dr R. Merrill, from the State University of Washington, Seattle, USA. Measurements were carried out in the Canberra laboratories with the kind permission of Professor A.L. Hales, Director of the School. REFERENCES Ashley, P.M., Chenhall, B.E., Cremar, P.L. and Irving, A.J., 1971. The geology of the Coolac Serpentinite and adjacent rocks east of Tumut, New South Wales. J. Proa. R. Soo. N.S.W., 204, 1-22. Briden, J.C., 1966. Estimates of directions and intensity of the palaeomagnetic field from the Mugga Mugga Porphyry, Australia. Geophys. J., 11, 267-278. Briden, J.C., 1967. Secondary magnetization of some Palaeozoic rocks from Tasmania. Pap. Proa. R. Soa. Tasmania, 101, 43-48. Burns, K.L., and Embleton, B.J.J., 1976. Palaeomagnetism and the structural history of the Lachlan region. Bull. Aust. Soa. Explor. Geophys., 7(1), 12-14. Coleman, R.G., 1971. Plate tectonic emplacement of upper-mantle peridotites along continental margins. J. Geophys. Res., 76, 1212-1222. CSIRO Division of Mineral Physics, NORTH RYDE NSW Collinson, D.W., Creer, K.M. and Runcorn, S.K.,1967. METHODS IN PALAEOMAGNETISM. Elsevier Publish- ing Company, Amsterdam, 609 pp. Dewey, J.F. and Bird, J.M. , 1971. Origin and emplacement of the ophiolite suite: Appalachian ophiolites in Newfoundland. J. Geophys. Res., 76, 3179-3206. Embleton, B.J.J., 1972a. The palaeomagnetism of some Proterozoic-Cambrian sediments from the Amadeus Basin, central Australia. Earth Planet. Sai. Lett., 17, 217-226. Embleton, B.J.J., 1972b. The palaeomagnetism of some Palaeozoic sediments from central Australia. <7. Proa. R. Soa. N.S.W., 105, 86-93. Embleton, B.J.J. and Giddings, J.W., 1974. Late Precambrian and Lower Palaeozoic palaeomagnetic results from South Australia and Western Australia. Earth Planet. Sai. Lett., 22, 355-365. Embleton, B.J.J. , McElhinny, M.W. , Crawford, A.R. and Luck, G.R., 1974. Palaeomagnetism and the tectonic evolution of the Tasman orogenic zone J. Geol. Soa. Aust., 21, 187-194. Fisher, R.A. , 1953. Dispersion on a sphere. Proa. R. Soa. London, A217 , 295-305. Luck, G.R., 1972. Palaeomagnetic results from Palaeozoic sediments of northern Australia. Geophys. J., 28, 475-487. Luck, G.R., 1973. Palaeomagnetic results from Palaeozoic rocks of southeast Australia. Geophys. J., 32, 35-52. McElhinny, M.W. and Embleton, B.J.J., 1974. Australian palaeomagnetism and the Phanerozoic plate tectonics of eastern Gondwanaland . Teatonophysias, 22, 1-29. McElhinny, M.W., and Luck, G.R., 1970. The palaeomagnetism of the Antrim Plateau Volcanics of northern Australia. Geophys. J., 20, 191-205. Oversby, B., 1971. Palaeozoic plate tectonics in the southern Tasman Geosyncline. Nature Phys. Sai., 234, 45-47. Ritchie, A., 1969. Ancient fish of Australia. Aust. Mus. Mag., 16, 218-223. Roberts, J., Jones, P.J., Jell, J.S., Jenkins, T. B .H., Marsden, M.A.H., McKellar, R.G., McKelvey,B.C. and Seddon, G. , 1972. Correlation of the Upper Devonian rocks of Australia. J. Geol. Soa. Aust., 18, 467-490. Scheibner, E. , 1972. Tectonic concepts and tecto- nic mapping. Rea. Geol. Surv. N.S.W. ,14,37-83. Solomon, M. and Griffiths, J.R., 1972. Tectonic evolution of the Tasman orogenic zone, eastern Australia. Nature Phys. Sai., 237, 3-6. (Manuscript received 24.3.1976) (Manuscript received in final form 20.9.1976) Journal and Proceedings. Royal Society of New South Wales. Vol. 110, pp. 29-31, 1977 Series of Roots of a Transcendental Equation Pietro Cerone and Austin Keane ABSTRACT. Series arising from the solution of a differential-difference equation are sununed by the use of Laplace transform methods and residue theory. The terms of the series depend on the roots of the transcendental equation p = e'nP , where n > 0 . INTRODUCTION SOLUTION OF THE DIFFERENTIAL-DIFFERENCE EQUATION Silberstein (1940), sums series arising from the solution of the differential-difference equation u' (t) = u(t - n) , t > n , u(t) = g(t) , 0 s t < n . Taking the particular value g(t) = 1 + t , Silberstein obtains the solution to (1), by the use of analytic continuation and an "orthogonality" condition, in the form; u(t) = I Pj ePj" Pj (1 + Pjh) (2) Many authors including Bellman and Cooke (1963) and El'sgol'ts and Norkin (1973), have solved (1) by use of Laplace transform techniques, the solution being given by ry+iR u(t) = Him J_ eP f(p)dp , t > 0 (7) R->" 2ti 'y-iR and u(0+) Aim = r-h. f(p)dp (8) ■*y-iR where u(0+) = Jlim u(0 + e) , e > 0 , e-K) where the pj's are the roots of the characteristic equation h(p) = 0 , h(p) = p - e-'iP (3) f(P) = q(p)/E(p) , rn q(p) = g(n)e*'^P + p g(x)e"P’'dx Jo and the sum is over all pj . From here on we will not specifically denote summation over pj . Then with the condition and y is such that Re(pj) < y where Pj are the roots of h(p) = 0 , which for the particular example are countably infinite in number and are distinct . u' (n) = u(0) (4) SILBERSTEIN 'S EXAMPLE holding, Silberstein concludes that (with n = 1) I 1 Pj(l + Pj) u(0) 1 (5) and Returning to equation (1) we have the transform U(p) of u(t) given by h(p)U(p) = g(0) + f e'P’‘g'(x)dx . (9) ^0 I = u(l) = 2 . pja + Pj) Assuming that g(x) can be differentiated n times in the interval (0,n) we can develop the integral to obtain It will be shown in this paper that condition (4) need not hold to obtain series identities by use of Laplace transform methods. It should be kept in mind that differing values of n result in different values of the h(p)U(p) = g(0) ^ g'(0) - e~''^Pg'(n) ^ g"(0) - e~^Pg"(n) ^ 2 P giving %g(0) = g(0) I + g'(0) Pj (1+npj) g'(n) y ^ - L ^ l+npj ^ (10) 30 PIETRO CERONE AND AUSTIN KEANE and Him 1 fY+iR dp R-x» 2TTi Y-iR p(p-e'^P) Him 1 ■Y+iR e"^P dp R-x« 27ii Y-iR p(p-e"^P) Him 1 ■Y+iR 1 1' R-+" 2Tii Y-iR p-e-'iP P dp = ^ • Since gW is an arbitrary function, equation (10) can only hold if y ^ ■ = Js i 1+np. ^ Pj (1+nPj) = 1 . (11) (12) It can be seen that the assumption g' (n) = g(0) removes the first series (11) and the two diffi- culties arising at t = 0 viz. the fact that the inversion integral equals h u(0+) and the integral Him j_ r ^ ^ 2wi I p A SIMPLE APPROACH Let us simplify the problem by taking g(t) = 1 . Then = KW • On inversion we obtain, using (7) and (8) u(t) = I f. np. t > 0 = I 1 1+nPi (13) whereas from the expansion -rnp U(p) = I r=0 p r+1 we obtain u(t) = I H(t-m) , t > 0 (14) r=0 where H(t) is the Heaviside unit function. Further from the identity 1 1 ^ 1 7^ ■ p"/^l*np) + n . )^"''"^-'(l-np) it follows using (13) and (14) that I-T t/n u(t) + nu(t-n) , t > n (15) = 1 + t I ^ H(t-m) . r=l Equation (15) gives the value of the sum of any inverse powers of pj and in particular with n = 1 , we can show yJ-=l yJ_=3 yJ^=li ^p. 2'i 2 3 2 ’ *^1 Pi Pi From (13) and (14), if we take t = 1,2,3 we have immediately equations (5) , (6) and ^ = j respectively. Pj (1+Pj) GENEPJiLISATIONS If we take bt g(t) = e , (b p^ , for any j ) in equation (1), it is a straight forward matter to obtain the result y T . = i Pj-b n ^ C2^nb)e nb 2(l-be^'’) (16) It is now possible to obtain any series \ (Pj-b)‘ by repeatedly differentiating both sides of (16) with respect to b . If we take _d_ db y J— ^ Pj-b we obtain 4b=-l n (1+nPj) 1+ne (17) Further generalisations may be obtained by using the derivative dpj dn l+nPj • (18) Then if we write F,(n) = I - L k (1+nPj)" (19) we obtain the recurrence relation ‘'k+2(''^ = ‘"k+l^^^ * TT^kU) (20) 1 with the starting values F^(n) = h , ^2^^^ ~ T+ne’ as in equations (11) and (17). It is then immediate from (20) that SERIES OF ROOTS OF A TRANSCENDENTAL EQUATION 31 1 l+ne (21) P4(n) 2+ne 2(l+ne)^ (22) ACKNOWLEDGMENT The authors would like to thank J.L. Griffith for his helpful comments on an early draft of this paper. REFERENCES Bellman, R. and Cooke, K.L., 1963. Differential- Difference Equations. Academia Press, N,Y. El'sgol'ts, L.E. and Norkin, S.B., 1973. Introduction to the Theory and Application of Differential Equations with Deviating Arguments. Academic Press, N.Y. Silberstein, L., 1940. On a Hystero-Differential Equation Arising in a Probability Problem. Phil. Mag. Ser 7, 29, 75-84. Department of Mathematics, The University of Wollongong, WOLLONGONG, N.S.W. 2500 (Manuscript received 22.11.1976) Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 33-40, 1977 Floral Evidence for a Middle Triassic Age of the Gunnee Beds and Gragin Conglomerate, near Delungra, New South Wales D. J. Bourke, R. E. Gould, R. Helby, R. Morgan and G. J. Retallack ABSTRACT. The Gunnee Beds, near Delungra, northern New South Wales, are a sequence of arkose, conglomerate, prominent sublabile sandstones, carbonaceous shale, and coal, unconformably overlying Late Palaeozoic rocks of the New England Fold Belt, and overlain by the Gragin Conglomerate with cobbles of quartz-feldspar porphyry and acid volcanics. Megafossil and microfossil floras obtained from the Gunnee Beds and Gragin Conglomerate indicate a Middle Triassic age. These are the only Triassic strata which crop out in the Warialda Trough on the west of the Woolomin-Texas Block of the New England Fold Belt. Deposition and preserv- ation of Middle Triassic terrestrial sediments both east and west of the New England region, indicate that the major unroofing of the New England Batholith had probably been completed by this time. INTRODUCTION In 1954, Rade (1954a) reported a Triassic megafossil flora from carbonaceous shales exposed in the bank of Warialda Creek, north-west of Delungra, N.S.W. Five species of the flora were subsequently identified by Dr. A.B. Walkom (in Rade, 1954b) and the containing strata, designated the Gunnee Beds, were correlated with the Ipswich Coal Measures of Queensland on the basis of the identified plant megafossils (Rade, 1954b). This was of interest as the Gunnee Beds were the only known Triassic strata outcropping at the base of the Surat Basin on the north-western margin of the Palaeozoic New England Tablelands region. Rade (1954b) correlated the overlying Gragin Conglomerate with the Bundamba Group of Queens- land and considered it to be of Jurassic age. Since then, however, these units have been shown on maps in part as Permian granite and Tertiary sediments by Chesnut and Cameron (1971) and Pogson and Hitchins (1974) . During recent mapping in the area (Bourke, 1974), D.J.B. relocated Rade's site in the Gunnee Beds (Figure 1); this led R.E.G. and G.J.R. to collect further specimens, including palaeopaly- nological samples, and to re-examine Rade's material housed in the Australian Museum. Palaeo- palynological samples were also collected from a shale and siltstone lense in the Gragin Conglom- erate, and a collection of plant megafossils from this unit examined. The palaeopalynological samples were investigated by R.M. and R.H. No attempt has been made here at major taxonomic description, the fossils merely being identified by comparison with known forms. Plant megafossils from the Gunnee Beds are housed in the Australian Museum (prefixed AMF) and the Geology Department, University of New England (prefixed UNEF) ; those from the Gragin Conglomerate and all figured microfossils in the Geological and Mining Museum of the Geological Survey of New South Wales (prefixed MMF and MMMC) . Stage co-ordinates refer to the rotary stage of the Geological and Mining Museum's Zeiss Photomicroscope No. 67500. Un- figured palaeopalynological material is lodged in the palynology collection of the Geological and Mining Museum, catalogue number 2319 (Gunnee Beds), and 2545, 2737, 2738 (Gragin Conglomerate). STRATIGRAPHY In the Warialda-Delungra district, Mesozoic sediments nonconformably overlie the New England Fold Belt (Scheibner, 1973; Leitch 1974; Rod, 1975) which in this area consists of highly deformed Palaeozoic cherts, slates, basic volcanics and tuffs (possibly the Woolomin Beds) intruded by granitic rocks of probable Late Carboniferous age. The basal Mesozoic unit of the sequence is the Triassic Gunnee Beds (Rade, 1954b); these may be a marginal equivalent to the Wandoan Formation of the Permo-Triassic Sydney-Bowen Basin. The Gunnee Beds are in turn overlain by the Triassic Gragin Conglomerate. Both the Gragin Conglomerate and the Gunnee Beds are restricted, in outcrop, to a small structural depression east of Warialda, named the Warialda Trough (Bourke, 1974) . Sediments of the Surat Basin overlie the Gragin Conglomerate. In the area shown on Figure 1, the Gunnee Beds consist of coarse-grained to granular sub- labile sandstone and arkose, interbedded with medium- to fine-grained labile sandstone and minor siltstone, carbonaceous shale and mudstone, and coal. The fossil plant horizons are interbedded with and overlain by prominent coarse-grained and granular sandstones. The sequence in this area is considered to represent the finer-grained, upper part of the Gunnee Beds. South and west of Figure 1, where the unit is topographically and probably stratigraphically lower than this sequence, the Gunnee Beds are composed of very coarse-grained to granular arkose, poorly-sorted paraconglomerate and coarse- to very coarse-grained sublabile sandstone. The unit ranges in thickness from 2 m to over 30 m. The Gunnee Beds appear to have been deposited under terrestrial conditions. It is evident from the immaturity of much of the Gunnee Beds that the sediment was transported only a short distance, and in many outcrops the arkose lies directly on the parent granitic rock. Restriction of the Gunnee Beds to the Warialda Trough and the Fig. 1. Geological map of the Delungra area lithology suggest that the unit may have been deposited in a river valley system with some dev- elopment of swamps and lakes. The Gragin Conglomerate is composed domin- antly of cobbles (with some pebbles) of quartz- feldspar porphyry and acid volcanics; clasts of granite, chert, and siliceous metamorphosed sediments such as mudstone, sandstone, and pebble conglomerate, are less abundant. The cobbles are about 120 mm in diameter, well-sorted and well- rounded. They are set in a matrix of medium- to coarse-grained sublabile sandstone. There are some lenses of laminar-bedded sublabile sandstone which are infrequently capped by lamellae of siltstone and carbonaceous shale. The conglom- erate is exposed as spectacular cliff faces up to 50 m high along Warialda Creek. The Gragin Con- glomerate is essentially conformable on the Gunnee Beds but shows an erosive basal contact. Sediments of the Surat Basin onlap across the Gragin Conglomerate, with the Hutton Sandstone overlying the unit on the western side of the Warialda Trough. Further east towards the margin of the Surat Basin, the Walloon Coal Measures and the Pilliga Sandstone (= Warialda Sandstone of Rade, 1954b) directly overlie the conglomerate. Some erosion of the Surat Basin sediments occurred prior to the area being covered by flood basalts in the Tertiary (Figure 2). Today the Gunnee Beds and Gragin Conglomerate are exposed in creek banks and other places where they have been exhumed from under the basalt or Surat Basin sediments. The sandstone and the conglomerate of the Gunnee Beds are extensively silicified (a silcrete, cf. Taylor and Smith, 1975) especially where they are overlain by basalt. t-EGAFOSSIL FLORA: GUNNEE BEDS The flora was collected from three localities (UNEL1586-1588) at GR 883221 on the Bingara MIDDLE TRIASSIC FLORAS NEAR DELUNGRA 35 A B j 0 I 2 KM Fig. 2. Geological cross section of the Delungra area. 1:100,000 sheet 9038, approximately 4.75 km NNW of Delungra (Figure 1) . The flora is dominated by species of the corystosperm Diaroidiim and closely related genera typical of the Triassic of Gond- wanaland. Locality UNEL1586 is in a black carbonaceous shale exposed on the southern bank of Warialda Creek. The shale grades upwards into the very fine sandstone and siltstone of UNEL1587. The fossil leaves are commonly oxidised with an oxidation corona developed into the dark matrix, but sometimes carbonised compression material does remain. Johnstonia aoriaoea is the most common of the fossils, which include: Diaroidium sp. cf. D. australe Jacob and Jacob 1950. [AMF45718-9] D. dubium (Feistmantel ) Gothan 1912. [AMF51140] D. lancifolium (Morris) Gothan 1912. [AMF45735] D. odontopteroides (Morris) Gothan 1912. [Figure 3.7] D. zuberi (Szajnocha) Archangelsky 1968; including various small and large fronds. [Figure 3.8] Diaroidium sp. ; like D. zuberi, but with pinnules coalescing in the apical part of the pinna and semicircular, wider than long in the basal portion. [AMF51141] ' Diaroidiopsis' sp. sensu Gould 1967. [Figure 3. 12] Johnstonia ooriacea (Johnston) Walkom 1925 s.s. ; with narrow lamina and entire margins. [Figure 3.5,6] J. stelzneriana (Geinitz) Frenguelli 1943; includes D. dentatum (Walkom) Anderson and Anderson 1970. [Figure 3.10] Johnstonia sp.; weakly lobed forms intermediate between J. stelzneriana and J. aoriaaea, including Diaroidium (intermediate sp. A) Anderson and Anderson 1970. [Figure 3.9] Xylopteris elongata (Carruthers) Frenguelli 1943. [Figure 3.11] Pteruchus sp. cf. P. sirmondsii (Shirley) Thomas 1933, sensu Townrow 1962; macroscopically intermediate between P. Johnstonii (Feistmantel) Townrow 1962 and P. sirmondsii. [UNEF15081-2] Pilophorosperma sp. cf. P. bumerense Thomas 1933. [UNEF15079; UNEF15080 is a similar specimen but about half the size] Isolated corystosperm seed; similar to that figured by Thomas (1933, fig. 33e) but rather more elongate. [UNEF15083] Rissikia media (Tenison-Woods) Townrow 1967. [Figure 3.14] Locality UNEL1587 is in a leached siltstone and very fine sandstone immediately underlying the bluff-forming sandstone and granule conglom- erate on the southern bank of Warialda Creek; the locality is about 1 m above UNEL1586. The plant remains, which are all fragmentary and naturally macerated to varying degrees, include: Cladophlebis lobifolia (Phillips) Seward 1900, sensu Walkom 1924. [Figure 3.1] Cladophlebis sp. cf. oblonga Halle 1914; similar specimen illustrated by Frenguelli (1947, pi. 5, fig. 7). [Figure 3.2] Asterotheaa menendezii de la Sota and Archangelsky 1962; probably includes fertile Cladophlebis australis of Walkom (1917). [Figure 3.3] Diaroidium lanaifolium (Morris) Gothan 1912; most like 'Thinnfeldia' acuta Walkom 1917. [Figure 3.4] Pteruchus johnstonii (Feistmantel) Townrow 1962. [AMF51132] Ginkgoites moltenensis (Seward) Du Toit 1927; includes G. digitata sensu Walkom 1917. [AMF51142] Rissikia sp. cf. R. apiaulata Townrow 1967. [Figure 3.13] Locality UNEL1588 is a grey shale underlying a thin coal seam in a small north-flowing tributary gully of Warialda Creek, about 70 m southeast of UNEL1586 and UNEL1587. This locality is probably a few metres stratigraphically lower than UNEL1586. Fossils collected: Diaroidium odontopteroides (Morris) Gothan 1912. [UNEF15084] 36 D. J. BOURKE AND OTHERS MIDDLE TRIASSIC FLORAS NEAR DELUNGRA 37 Pilophorosperma sp. cf. P. Bumepense Thomas 1933. [UNEF15085-6] Many of the forms present, e.g. D. odontop- teroides, D. dubium, J. ooriacea, J. stelzneriana, X. elongata, and P. johnstonii, are commonly given Middle to Late Triassic ranges (Stipanicic and Bonetti, 1969; Banks, Cosgriff, and Townrow, 1967; Anderson and Anderson, 1970). However local comparisons suggest a Middle Triassic (Anisian to Ladinian) age for the Gunnee Beds. Cladophlebis lobifolia sensu Walkom (1924) is only known from the Middle Triassic Esk and Neara Beds of Queensland and the Middle Triassic Nymboida Coal Measures of New South Wales (Flint and Gould, 1975); a basalt flow in the Nymboida Coal Measures has yielded a Middle Triassic radiometric date of 211 ± 5 x 10* years (Retallack, Gould, and Runnegar, in press). Johnstonia aoriaaea (s.s. with entire margins) is much more frequent in the Esk Beds than in the Late Triassic Ipswich Coal Measures (Jones and de Jersey, 1947, p. 19; de Jersey, 1972) , and also occurs at one horizon in the Basin Creek Formation of the Nymboida Coal Measures (UNEF14718 from UNEL1489) . Xyloptevis elongata is common in the Ipswich Coal Meajures, but is also found in the early Middle Triassic Hawkesbury Sandstone at Brookvale, N.S.W. (AMF18581); only one specimen has been found in the Gunnee Beds. The flora from the overlying Gragin Conglomerate is also of Middle Triassic age and so restricts the youngest age assignable to the Gunnee Beds. This Middle Triassic age determination was independently substantiated by palaeopalynological evidence . MICROFOSSIL FLORA: GUNNEE BEDS A palaeopalynological sample from the black carbonaceous shale at the plant megafossil locality UNEL1586 on Warialda Creek, was processed and examined by R.M. and R.H. The microfossil flora is dominated by Faloisporites australis (de Jersey) Balme 1970 [Figure 3. 18] and includes the following additional important forms: Aratrisporites parvispinosus sensu Helby 1973. [Figure 3.16, 17] Aratrisporites tenuispinosus sensu Helby 1973. Baoulatisporites aomaumensis (Cookson) Potonie 1956. Cycadopites follicularis Wilson and Webster 1946. [Figure 3.15] 'Guthoerlisporites' oanaellosus Playford and Dettmann 1965. Indospora alara Bharadwaj 1962. ' Nevesisporites' limatulus Playford 1965. Polypodiisporites rmtabilis Balme 1970. Protohaploxypinus samoiloviohii (Jansonius) Hart 1964. Protohaploxypinus sp. cf. P. jacobii (Jansonius) Hart 1964. ' Retusotriletes ' radiatus (Kara-Murza) . Uvaesporites sp. cf. U. verrucosus (de Jersey) Helby ex de Jersey 1971. Verrucosisporites sp. cf. V. oamarvonensis de Jersey and Hamilton 1967. The microflora recovered from the Gunnee Beds exhibits restricted diversity, but the mutual occurrence of A. parvispinosus, P. samoiloviohii, U. sp. cf. U. verrucosus, and V. sp. cf. V. oamarvonensis suggest a close comparison with the A. parvispinosus Assemblage Zone of Helby (1973) which is of Middle Triassic age. The apparent absence of Annulispora spp. and Craterisporites rotundus de Jersey 1970 from the Gunnee Beds association may differentiate it from the C. rotundus Zone which characterises the Late Triassic Ipswich Coal MeaSures (de Jersey, 1975) . This is substantiated by the occurrence in the Gunnee Beds of P. sp. cf. P. jacobii and V. sp. cf. K. oamarvonensis which do not appear to range into strata younger than mid-Ladinian (R.H.). An Anisian/Ladinian age is thus suggested. Helby (1973) indicated that the A. parvispin- osus Assemblage Zone also occurred in the Wandoan Formation, the Clematis Sandstone and the Moolay- ember Formation in the Bowen Basin of Queensland. Although the Gunnee Beds microfossil flora is clearly similar to the assemblages reported from the Wandoan Formation (de Jersey and Hamilton, 1969), the Clematis Sandstone (de Jersey, 1968), and the Moolayember Formation (de Jersey and Hamilton, 1967), none of the species regarded as diagnostic of the Duplexisporites problematicus Microflora by de Jersey (1975) were encountered. The apparent absence of these species and the restricted diversity of the Gunnee Beds assemblage may possibly be explained by the extreme proximal position of the Gunnee Beds in the Wandoan/ Moolayember drainage system. Fig. 3. 1, Cladophlebis lobifolia (Phillips) Seward sensu Walkom; portion of pinna showing characteristic pinnules. AMF51128, X 1.5. 2, Cladophlebis sp. cf. C. oblonga Halle. AMF51129, X 1.8. 3, Portion of fertile pinna of Asterotheoa menendezii de la Sota and Archangelsky . AMF51130, X 1.5. 4, Dicroidium lancifolium CMorris) Gothan. AMF 51131, X 1.5. 5, 6, Johnstonia ooriacea (Johnston) Walkom s.s. 5, AMF51133, X 0.9. 6, AMF51134, XI. 7, Dicroidium odontopteroides (Morris) Gothan. AMF51135, X 1.1. 8, Dicroidium zuberi (Szajnocha) Archangelsky. AMF51136, X 0.75. 9, Johnstonia sp. AMF51137, X 0.6. 10, Johnstonia stelzneriana (Geinitz) Frenguelli. AMF51138, X 1. 11, Xylopteris elongata (Carruthers) Frenguelli. AMF45745, X 1.35. 12, ' Dicroidiopsis ' sp. sensu Gould. AMF45710, X 0.8. 13, Rissikia sp. cf. R. apiculata Townrow. AMF51143, X 1.3. 14, Rissikia media (Tenison-Woods) Townrow. AMF45693, X 1.6. 15, Cycadopites follicularis Wilson and Webster. MMMC2010, 028/0886, X 800. 16, 17, Aratrisporites parvispinosus Leschik sensu Helby. 16, distal focus. 17, proximal focus. FMMC2011, 042/1102, X 800. 18, Faloisporites australis (de Jersey) Balme. MMMC2011, 050/1070, X 800. 38 D. J. BOURKE AND OTHERS MEGAFOSSIL FLORA: GRAGIN CONGLOMERATE A collection of plant megafossils was obtain- ed by D. Probert from small, khaki shale and silt- stone lenses interbedded with the conglomerate at UNEL1589. The locality is in a gravel pit near the junction of the Graman and Gragin Peak Roads at GR 831281 on the Bingara 1:100,000 sheet 9038. Determinations include: Dioroidium sp. cf. D. eskense (Walkora) Jacob and Jacob 1950; poorly preserved, with some transverse creasing of the pinnae. [MMF 17080, MMF17084] D. odontopteroides (Morris) Gothan 1912. [MMF 17085-6] D. zuberi (Szajnocha) Archangelsky 1968; small fronds like the one figured from the Gunnee Beds. [NWF17081-3] Lepidopteris madagasaar'iensis Carpentier 1935. [MMF17090] Presence of Dioroidium spp. indicates a Tri- assic age for the flora. Lepidopteris madagas- oariensis ranges from Early to Middle Triassic (Townrow, 1966) while D. odontopteroides ranges from Middle to Late Triassic (Townrow, 1967), so a Middle Triassic age for the Gragin Conglomerate is most likely. This is substantiated by the presence of a frond comparable to D. eskense which is restricted to Middle Triassic floras of eastern Australia (see Flint and Gould, 1975). MICROFOSSIL FLORA: GRAGIN CONGLOMERATE Several samples collected from siltstone and carbonaceous shale horizons at the top of a small sandstone lense of the Gragin Conglomerate in a cave on the eastern bank of Spring Gully (Bingara 1:100,000 sheet 9038, GR 759284) were subjected to palaeopalynological investigation. Spores and pollen recovered from the Gragin Conglomerate samples were even more restricted than those from the Gunnee Beds. Palynomorphs identified include: Converrucosisporites oameronii (de Jersey) Play- ford and Dettmann 1965. Cyaadopites folliaularis Wilson and Webster 1946. Diotyophyllidites mortonii (de Jersey) Playford and Dettmann 1965. Falaisporites spp. , including F. australis (de Jersey) Balme 1970. Guttatisporites visscheri de Jersey 1968. Neoraistriokia taylori Playford and Dettmann 1965. Osmundaoidites wellmanii Couper 1953. Polypodiisporites ipsviaiensis (de Jersey) Play- ford and Dettmann 1965. Punctatisporites spp. Punctatosporites walkomii de Jersey 1962. Rugulatisporites sp. Tubereulatosporites spp. Uvaesporites sp. cf. U. verruoatus (de Jersey) Helby ex de Jersey 1971. Vitreisporites pallidus (Reissinger) Nilsson 1958. Ciraulisporites parvus (de Jersey) Norris 1965 - acritarch. Falaisporites and Polypodiisporites were the most prominent genera in the samples. The micro- fossil association is clearly Triassic in age but the apparent absence of diagnostic species (discussed above in relation to the Gunnee Beds microfossil flora) precludes more definitive age assignment beyond suggesting that they are no older than Middle Triassic. DISCUSSION The Middle Triassic age assigned to the Gunnee Beds and Gragin Conglomerate is thus based on both plant megafossil and raicrofossil data. Fossil plant-bearing sediments of a similar age occur to the north in the Bowen Basin (de Jersey and Hamilton, 1967, 1969) and the Esk Trough of southern Queensland (de Jersey, 1972; Figure 4), to the east at Nymboida in the southern Clarence- Moreton Basin (Flint and Gould, 1975), far to the south in the Sydney Basin (Helby, 1973), and possibly to the south-west in the Gunnedah Basin as well (Hind and Helby, 1969; Bembrick, Herbert, Scheibner, and Stunz, 1973). The Delungra and Nymboida Middle Triassic sediments impose some constraints on interpret- ations of the New England region. The granitic rocks of the New England Batholith which intrude the Palaeozoic basement have been assigned Late Carboniferous to Early Triassic radiometric ages (Wilkinson, 1974), and the deposition and preser- vation of Middle Triassic coal measures on erosion surfaces of this complex probably means that most of the unroofing of the New England Batholith was completed by the Middle Triassic. The Gunnee Beds near Delungra have been deposited on granitic rocks of the batholith and contain much quartzose and arkosic sediment. The clasts of the overlying Gragin Conglomerate are mostly of quartz-feldspar porphyry with some granitic rocks. The Nymboida Coal Measures, unconformably overlying the Late? Palaeozoic sediments of the Coffs Harbour Beds (Korsch, 1971), contain the Bardool Conglomerate (McElroy, 1963). The majority of clasts in the Bardool Conglomerate are rhyolitic and/or ignim- britic; clasts of greywacke, siltstone, and chert from the underlying Palaeozoic sediments are also present. So it seems the source for much of the sediment in these Middle Triassic units was from erosion of the granitic rocks of New England and their volcanic and sub-volcanic equivalents. The granitic rocks underlying the Gunnee Beds are probably among the earliest intru- sives of the batholith, but even so they had already been exposed and eroded by the Middle Triassic. The later intrusives of the batholith show characters of high level emplacement (Wilkinson, 1974) which is not inconsistent with a relatively rapid erosion of the intruded country rocks . ACKNOWLEDGEMENTS D.J.B. and R.M. publish with the permission of the Under Secretary, New South Wales Department of Mines. Dr J. Pickett drew our attention to the collection of fossils from the Gragin Conglom- erate. Mr H. Butler and Mr K. Williams assisted in collection of plant fossils and provided valuable discussion. Dr A. Ritchie and Miss D. Jones kindly provided access to material in the Australian Museum. The maps were drafted by Mr M. Bone. REFERENCES Anderson, H.M. and Anderson, J.M., 1970. A preliminary review of the biostratigraphy of the uppermost Permian, Triassic and lower- most Jurassic of Gondwanaland. Palaeont. afr. , 13, suppl . , 1 . MIDDLE TRIASSIC FLORAS NEAR DELUNGRA 39 Fig. 4. Location map. Banks, M.R., Cosgriff, T.W., and Townrow, J.A., 1967. Triassic System in Australia. Proc. I. b.G.S. Gondwana Symposium, Buenos Aires. Abstr. , 19. Bembrick, C.S., Herbert, C. , Scheibner, E., and Stunz, J., 1973. Structural subdivision of the New South Wales portion of the Sydney- Bowen Basin. Quccpt. Notes geol. Surv. N.S.W., II, 1. Bourke, D.J., 1974. A structural subdivision of the Surat Basin in the Warialda area. Quart. Notes geol. Surv. N.S.W., 16, 1. Chesnut, W.S., and Cameron, R.G., 1971. Inverell 1:250,000 Geological Sheet. Geol. Surv. N.S.W. De Jersey, N.J., 1968. Triassic spores and pollen grains from the Clematis Sandstone. Pubis geol. Surv. Qd, 338. De Jersey, N.J., 1972. Triassic miospores from the Esk Beds. Pubis geol. Surv. Qd, 357. De Jersey, N.J., 1975. Miospore zones in the Lower Mesozoic of southeastern Queensland. In Campbell, K.S.K., (Ed.). Gondwana geology: papers presented at the Third Gondwana Symposium, Canberra, Australia, 2973, 159. A.N.U. Press, Canberra. De Jersey, N.J., and Hamilton, M., 1967. Triassic spores and pollen grains from the Moolayember Formation. Pubis geol. Surv. Qd, 336. 40 D. J. BOURKE AND OTHERS De Jersey, N.J. , and Hamilton, M. , 1969. Triassic microfloras from the Wandoan Formation. Rept geol. Supv. Qd, 31. Flint, J.C.E., and Gould, R.E., 1975. A note on the fossil raegafloras of the Nymboida and Red Cliff Coal Measures, southern Clarence- Moreton Basin, N.S.W. J. Rroo. R. Soa. N.S.y., 108, 70. Frenguelli, J. , 1947. El genero "Cladophlebis" y sus representantes en la Argentina. Rotas Mus. La Plata Paleont. , Secc. B. Paleobot., 2, 1. Gould, R.E. , 1967. The geology of the Slacks Creek area, south-east Queensland. Pap. Dep. Geol. Univ. .id, 6 (5) . Helby, R.J., 1973. Review of Late Permian and Triassic palynology of New South Wales. Spec. Pubis geol. Soc. Aust. , 4, 141. Hind, M.C. and Helby, R.J., 1969. The Great Artesian Basin in New South Wales. J. geol. Soc. Aust., 16, 481. Jones, O.A., and De Jersey, N.J., 1947. The flora of the Ipswich Coal Measures - morphology and floral succession. Pap. Dep. Geol. Univ. id, 3 (3). Korsch, R.J., 1971. Palaeozoic sedimentology and igneous geology of the Woolgoolga district, north coast. New South Wales. J . Proa. R. Soc. N.S.W., 104, 63. Leitch, E.C., 1974. The geological development of the southern part of the New England Fold Belt. J. geol. Soc. Aust., 21, 133. McElroy, C.T., 1963. The geology of the Clarence- Moreton Basin. Mem, geol. Surv. N.S.W. , Geo logy, 9 . Pogson, D.J., and Hitchins, B.L., 1974. New England 1 -.500,000 Geological Sheet. Geol. Surv. N.S.W. Rade, J., 1954a. Geology and subsurface waters of the Moree District, New South Wales. <7. Proa. R. Soc. N.S.W., 87, 152. D.J. Bourke, Department of Mines, P.O. Box 65, Armidale, N.S.W. 2350. R.E. Gould and G.J. Retallack, Department of Geology, University of New England, Armidale, N.S.W. 2351. Rade, J., 1954b. Uarialda artesian intake beds. J. Proc. R. Soc. N.S.W., 88, 40. Retallack, G.J., Gould, R.E., and Runnegar, B. , in press. Isotopic dating of a Middle Triassic megafossil flora from near Nymboida, north- eastern New South Wales. Proc. Linn. Soc. N.S.W. Rod, E. , 1975. Structural interpretation of the New England region. New South Wales. J. Proc. R. Soc. N.S.W., 107, 90. Scheibner, E., 1973. A plate tectonic model of the Palaeozoic tectonic history of New South Wales. J. geol. Soc. Aust., 20, 405. Stipanicic, P.N., and Bonetti, M.I.R., 1969. Consideraciones sobre la cronologia de los terrenos triasicos argentinas. In Gonduana Stratigraphy, I.U.G.S. Symposium, Buenos Aires, 1081. UNESCO, Paris. Taylor, G., and Smith, I.E., 1975. The genesis of sub-basaltic silcretes from the Monaro, New South Wales. J. geol. Soc. Aust., 23, 377. Thomas, H.H., 1933. On some pteridospermous plants from the Mesozoic rocks of South Africa. Phil. Trans R. Soc. Lond. , Ser. B, 222, 193. Townrow, J.A., 1966. On Lepidopteris madagas- cariensis Carpentier (Peltaspermaceae) . J. Proc. R. Soc. N.S.W., 98, 203. Townrow, J.A. , 1967. Fossil plants from Allan and Carapace Nunataks, and from the Upper Mill and Shackleton Glaciers, Antarctica. N.Z. J. Geol. Geophys. , 10, 456. Walkom, A.B., 1917. Mesozoic floras of Queensland. Part 1. - continued. The flora of the Ipswich and Walloon Series. (c) Filicales, etc. Pubis geol. Surv. Qd, 257. Walkom, A.B. , 1924. On fossil plants from Bellevue near Esk. Mem. Qd Mus., 8, 77. Wilkinson, J.F.G., 1974. The New England Bath- olith. Geol. Soc. Aust., Qd. Div., Field Conf. Guidebook, New England area, 24. R. Helby, 356 Burns Bay Road, Lane Cove, N.S.W. 2066 R. Morgan, Geological and Mining Museum, Geological Survey of New South Wales, 36 George Street, Sydney, N.S.W. 2000. (Manuscript received 3.4.1975) (Manuscript received in final form 10.8.1976) Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 41-47, 1977 The Tertiary Stratigraphic Palynology of the Murray Basin in New South Wales. 1. The Hay-Balranald-Wakool Districts Helene A. Martin ABSTRACT. The Hay, Balranald and Wakool districts are situated on entirely non-marine sediments, to the east of the limit of marine transgression of the Murray Basin. This palynological study of ten selected bores relates the sediments to the geological time scale by correlating the assemblages with the spore-pollen zones described for the Gippsland Basin. In this part of New South Wales, the deepest and oldest sediments, below 305m, are middle Eocene in age. Late Eocene assemblages are found below about 274m. Oligocene - early Miocene sediments form thick sections between approximate levels of 120m and 270m, and are encountered in every bore. Carbonaceous clays and lignites are common and the climate during deposition of these sediments must have been very humid, with lakes and swamps a feature of the landscape. Middle Miocene and younger assemblages have been recovered from only four of the bores examined. During deposition of the upper 100m, it appears that conditions were not favourable for pollen preservation. About the middle Miocene, the deposition of carbonaceous clays and lignites ceased and the rainfall must have decreased, although the continuance of the grey colour of the clays indicates that it was uniformly distributed through the year. with no marked dry season. INTRODUCTION The Murray Basin, over 259,000 square km. in area, extends into South Australia, Victoria and New South Wales. Tertiary sediments comprise the main bulk of the basin deposits, with a maximum thickness of some 500m in the central part, thinning out towards the basin margins. They have a complex stratigraphy which has been the subject of numerous reports (e.g. Ludbrook, 1961; Pels, 1960, 1969; Lawrence, 1966; Bembrick, 1974). These studies, however, have all been restricted to particular areas and there remain large parts of the basin that have been little investigated. Sediments in the eastern part of the basin are entirely non-marine and palynology is the only method readily available to relate them to the geological time scale. This paper presents the stratigraphic palynology from bores of the Hay, Balranald and Wakool districts in the east- central part of the basin, (see Fig. 1). There is one report of a palynological study in this area, that of Evans and Hodgson (1963) . Material from bores of this study have been re- examined in the light of the considerable advances made since these authors completed their report . GEOLOGY Of major significance to the Tertiary stratigraphy of the Basin is the eastern limits of the marine transgressions which extends from the west. Pels (1960) suggest that a probable basement ridge with a north easterly orientation, between the Lachlan and lower Darling Rivers, marks the limit of marine deposition. There is a dolomite bed at approximately 121 -152m in the Balranald No. 1 Bore, (see Fig. 1), but this is the only evidence of marine deposition in the area of this study. While not providing any additional information on the location of the limit of the marine transgression, present results are not inconsistent with Pel's proposal. A number of units have been described by previous authors, but only two have relevance to this non-marine area, the Renmark Beds (formerly the Knight Group, Harris, 1966b), and the Wunghnu Group (Lawrence, 1966) . The Renmark Beds are the most widespread of the units, and form the base of the Tertiary sequence. They consist of fine to medium-grained quartz sand; silt and siltstone which may be carbonaceous, dolomitic or calcareous; clay, usually carbonaceous, and lignites. The various sediments form alternating strate which inter- tongue. Evidence from elsewhere in the basin only indicates a Lower Tertiary age (Ludbrook, 1963; Lawrence, 1966) . Bembrick (1974) distinguishes the upper and lower Renmark Beds; the former are dominantly sandy with minor shale, siltstone and lignitic lenses whereas the latter are more argillaceous and carbonaceous. Because the Renmark Beds are overlain by several different younger units elsewhere in the basin, their upper boundary may not be synchronous . The Wunghnu Group (described from the Northern Plains of Victoria) overlies the Renmark Beds, beyond the limits of the marine transgress- ions. The sediments are fluviatile and lacustrine, highly variable in lighology, mottled throughout and contain numerous paleosols. The criterion used to discern the boundary between the Wunghnu Group and the Renmark Beds is the shallowest occurrence of the persistent carbonaceous beds which are characteristic of the latter (Lawrence, 1966) . The upper surface of the unit is defined as the present land surface. Selected bore logs are presented in Fig. 3. 42 HELENE A. MARTIN Five of the bores reached basement but the five most northerly ones were terminated before base- ment was reached. The logs show that the sediments are fine-grained and consist mainly of clay and silt with relatively minor amounts of sand. The latter are usually fine-grained with a particle size of about 2-2%nm or less. The clay colour is red-yellow-brown, as well as grey, down to about 90m, below which only various shades of grey are found. This boundary can be observed in every bore. Near Hay, it may be shallower, up to 67m, whereas further south, it is deeper, down to 109-112m. The grey colour found alternating with the various shades of red-yellow-brown in the upper 90m is invariably pale, and carbonaceous material has not been found here. Carbonaceous clays first occur below about 90m, where the red- yellow-brown coloured clays are entirely absent. Near Hay, the boundary appears to be shallower, but it is a layer of light grey above the carbonaceous clays which accounts for this apparent shallowness. If the previously described units are extended to this area, then the Wunghnu Group would extend down to about 90m (although mottling is not apparent in the bores described here) , and the Renmark Beds below. In the sediments of the Lachlan River Valley, Williamson (1964) has described two units; the upper one highly variable but mainly red-brown in colour (Cowra Formation) and the lower one grey in colour (Lachlan Formation). This division is based on the dominant rock type of the sand and gravels. In the upper unit it is a mixture of the various country rocks of the catchment area whereas in the lower, it is almost entirely quartz. In the sediments of this study, gravels are rarely encountered, and the sands, where they occur, are usually described as quartz so a division based on rock type is not possible. It seems probable, however, that the Cowra and Lachlan Formation together are the lateral equivalent of at least the upper part of the Wunghnu Group, and possibly the whole group. THE SPORE-POLLEN ZONATION Appendix 1 gives full title and Fig. 1 shows the location of bores selected for presentation in this paper. A number of other bores have been examined, but they repeat the evidence of those selected and presented here. Samples from these bores were treated as described in Martin, (1973a) . The assemblages presented in Appendix 2* have been correlated with the spore-pollen zones described by Stover and Patridge (1973) for the Gippsland Basin. These authors have assigned a time range to the zones from evidence of the foraminiferal assemblages in the interbedded marine strata. Fig. 2 presents the time ranges of the spore-pollen zones. By adopting this approach, it is assumed that similar events were synchronous in the Gippsland and Murray Basins, unless there is evidence to the contrary. For assemblages which are younger than those described by Stover and Partridge (1973) the tentative scheme of Martin (1973b) has been followed. Fig. 1. Locality Map. 1973 and Martin 1973b. *Appendix 2 is available from the author, on request. TERTIARY STRATIGRAPHIC PALYNOLOGY 43 In general, these assemblages fit the spore- pollen zones quite well. Occasionally some of the evidence appears contradictory, but when all of the evidence is examined, it is usually over- whelmingly in favour of one particular spore- pollen zone. The time ranges given by Stover and Partridge for some of the species have proven unreliable, but these have not been used in the determination of the spore-pollen zones. In many cases, there is a continuous gradation from one zone to another. 1 . The Lower Nothofagidites aspepus zone (for time ranges see Fig. 2) is the oldest of the Tertiary zones found anywhere in these studies. Two subdivisions have been recognised; an older or pre-Triorites magnifiaus and a younger, containing T, 'magnifiaus. The older subdivision contains a wealth of the distinctive species which terminate their range by the end of the Lower Nothofagidites asperus zone, and Pvoteaoidites spp. are abundant. Nothofagidites spp. far out- number Haloragacidites harrisii, and this shows that these assemblages cannot be any older than the Lower Nothofagidites asperus zone. The younger subdivision of this zone contains fewer of these distinctive species, but Triorites magnifiaus is present. Distribution: the older subdivision. Hay 1 at 317m and 329m and Bundy at 385-386m. The younger subdivision, Wakool 36078 at 280-333m and Wakool 36102 at 285-311m. Based on the number of distinctive species for the zone, these assemblages form a series with Bundy the oldest. Hay 1 inter- mediate and Wakool the youngest. A.D. Partridge (pers. comm.) considers that the Hay 1 assemblage may belong to the younger subdivision, but its relative position in the series is correct. 2. The Upper Nothofagidites asperus zone has a restricted diversity and Stover and Partridge (1973) consider that it is transitional between the Lower N, asperus and Proteaaidites tuberoulatus zones. The one occurrence of the Upper N. asperus zone in this study has Proteaaidites stipplatus as the dominant of the Proteaaidites spp. Curiously, Wakool 36078 has both the Lower N, asperus and P. tuberoulatus zones which simply intergrade without any indication of an assemblage of the Upper N. asperus zone . Distribution: Hay 1 at 283m. 3. The Proteaaidites tuberoulatus zone has been subdivided into three subdivisions, but the indicator species on which this is based are not often encountered in this study, so subdivision is not reliable. The best indicator for the base of the zone, Cyatheacidites annulatus is not common in this study, so the base has been delimited on the general characteristics of the assemblages. There is a gradation from the Lower N. asperus zone as its characteristic species drop out, one by one, in those bores with close sampling. In common with the older zones, the content of Nothofagidites spp. pollen is high and Myrtaaeidites spp. low. However, near the top of the zone, the relative frequencees change, with a decrease in Nothofagidites spp. and an increase in Myrtaoeidites spp . Distribution: (1) Bores which reached base- ment: Hay 1 at 122-210m: Balranald at 186m and 287-288m: Bundy at 207-213m and 277-278m: Wakool 36078 at 126-267m. (2) Bores which did not reach basement ended in the P. tuberoulatus zone : 30435 at 110-198m: 30383 at lll-178m: 30443 at 102-207m: 30464 at 97-140m: 36025/2 at ?113-139m. 4. The Triporopollenites bellus zone has a number of indicator species which are all uncommon to rare in this study. The relative content of Nothofagidites spp. is much lower and Myrtaoeidites spp. much higher than in the P. tuberoulatus zone, respectively. These two factors make it difficult to draw a line between the two zones. The assemblages of the P. tuberaalatus - T. bellus 'interzone' are quite distinctive in that they have a wealth of angiosperm pollen, especially the tricolpate - tricolporate types, many of which have not been described. Many of these pollen types are present in older assemblages but it is their association that is so distinctive in the P. tuberoulatus - T. bellus 'interzone'. All of the T. bellus assemblages belong to the older part of the zone because Haloragaaidites haloragoides , which first appears during this zone, has not been found. Distribution: Wakool 36078 at 94-108m and Hay 36025/2 at 105-?lllm. 5. Assemblages younger than the T. bellus zone are rarely encountered. Samples above this zone are usually barren or contain a restricted assemblage in such a poor state of preservation that it is unworkable. There are two exceptions, one of the Myrtaceae-Pasuar*ina phase (probably the older) and one of the Pleistocene-Recent. In both these assemblages, Myrtaceae type pollen is the dominant group with Nothofagidites either in very low percentages or absent, respectively. The Pleistocene-Recent assemblage has a higher content of Tubulifloridites spp. and Graminidites media, and together with Tubulifloridites spp. and Graminidites media, and together with Tubulifloridites pleistoaenious , these features are seen only in the Cowra Formation (Martin, 1969) . Distribution: 'Myrtsceae-Casuarina phase' in Wakool 36102 at 76-78m: Pleistocene-Recent in Wakool 36078 at 17-18m. DISCUSSION Fig. 3 presents the correlation of the stratigraphic columns. The oldest assemblages of this study belong to the Lower Nothofagidites asperus zone and are middle Eocene in age. They are found in the deepest sediments, below 305m, in two adjacent bores. Hay 1 and Bundy. Slightly younger assemblages, of about late Eocene age are found to the south in Wakool 36078 and 36102, below 274m. It is possible that these Eocene sediments extend further north, beneath the bores which did not reach basement. 44 HELENE A. MARTIN WAKOOL Fig. 3. Correlation of the palynological zones of selected bores. For location of bores, see Fig. 1. There is a single occurrence of the Upper Nothofagidites asperus zone in Hay 1. Where both Lower N. asperus and Proteacidites tuberaulatus zones occur in Wakool 36078 and 36102, there is a complete gradation between the two without any indication of the intermediate Upper W. asperus zone. This suggests that assemblages of the Upper N. asperus zone may have been produced by certain conditions of the environment and/or deposition. The Proteacidites tuberaulatus zone forms the thickest interval in every bore, between the approximate lievels of 120ra and 270m. Carbonaceous clays and lignites are common, and in some bores deposition of polleniferous sediments has been almost continuous. Both the sediments and pollen assemblages testify to widespread swamps and/or lakes. The top of this zone approximates the upper limit of the carbonaceous clays and is about middle Miocene in age. Using the criterion of Lawrence (1966), this is the top of the Renmark 'Beds, which here have an age much younger than reported elsewhere (Bembrick, 1974) . Many samples above the carboniferous clays have been examined but most have either yielded no pollen or have such a restricted assemblage with pollen so poorly preserved that it has been unworkable. The two occurrences of the T. bellus zone follow closely above the carbonaceous clays. The one Pliocene assemblage of the Myrtaceae- Casuarina phase and the one Pleistocene-Recent assemblage both occur in the Wakool bores and probably represent a local lake or billabong. TERTIARY STRATIGRAPHIC PALYNOLOGY 45 The age indicated is consistent with the postulated time equivalent to the Wunghnu Group and the Cowra/ Lachlan Formation. Today, the area around Wakool is low-lying and subject to inun- dation at times of higher river levels. Pels (1969) has described the Tertiary deposition in the Murray Basin 'in New South Wales and this study allows timing of some of the events. Deposition first started about the middle Eocene. The thick sections of carbonaceous clays and lignites, often containing fossil wood, were formed throughout the Oligocene and early Miocene, although some of the lignites may have started somewhat earlier. Pels describes the upper limits of the carbonaceous clays thus: " the upper limits of the lacustrine sediments form horizontal planes bounded by rock. The final lacustrine landscape was therefore one of a series of flat plains bounded by rock outcrops and stepping down to the west and southwest. This former landscape, including its rock outcrops is now buried to an average depth of 90m by fluviatile sediments " This ancient lacustrine landscape is middle Miocene in age and the overlying fluviatile sediments include late Miocene, Pliocene and Pleistocene deposition. The following climatic history can be reconstructed from the evidence of pollen assemblages and the nature of the sediments. There was a long period of a very humid climate throughout the Oligocene and early Miocene, when the rainfall was at least 180cm p.a., and it could have been very much more, judging from present day requirements of Nothofagus (Martin, 1969) . By middle Miocene time the climate had become less humid, carbonaceous clays were no longer being formed and the abundance of plant types requiring high humidity, e.g. Nothofagus, had decreased. The rainfall at this time was probably 150-180cm p.a., fairly uniformly distributed through the year, without a marked dry season (Martin, 1969, 1973b) . Conditions had become less suitable for pollen preservation which was restricted to a few localised lakes or billabongs, probably associated with the prevailing river systems. This decrease in humidity occurred in the Murray Basin at an earlier time than in the Gippsland Basin. Stover and Partridge (1973) state: "Specimens of Nothofagidites spp. are still abundant in the lower part of the (T. bellus) zone, and become gradually less common towards the top." (In contrast, the abundance of Nothofagidites spp. has been greatly reduced by the beginning of the T. bellus zone in the Murray Basin.) It is to be expected that decreasing humidity will become apparent in inland, continental districts, such as Western New South Wales, before its effects are seen in coastal locations like those of the Gippsland Basin. ACKNOWLEDGEMENTS I am indebted to the Water Conservation and Irrigation Commission of New South Wales for the financial assistance without which this work would not have been possible. Messrs. W.H. Williamson and D. Woolley of the Commission have provided invaluable assistance throughout. REFERENCES Bembrick, C.S., 1974. The Murray Basin In Markham, N.L. and Basden, H., eds.. The mineral deposits of New South Wales. Geol . Surv. of N.S.W. Cookson, I.C., 1947a. Plant fossils from the lignites of the Kerguelen Archipelago. B.A.N.Z. Antarat Res. Exp. 1929-31, Rep. Ser. ^ 129-142. Cookson, I.C., 1947b. On fossil leaves (Oleaceae) and a new type of fossil pollen grain from Australian brown coal deposits. Proa. Linn. Soo. N.S.W. 7^ 183-197. Cookson, I.C., 1950. Fossil pollen grains of proteaceous type from the Tertiary deposits of Australia. Aust. J. Scient. Res. 3(Ser. B) , 166-177. Cookson, I.C., 1953. Difference in microspore composition of some samples from a bore at Comaum, South Australia. Aust. J. Bot. 1, 462-473. Cookson, I.C., 1956. On some Australian Tertiary spores and pollen grains that extend the geological and geographical distribution of living genera. Proa. Roy. Soo. Viet. 69, 41-54. Cookson, I.C., 1959. Fossil pollen grains of Nothofagus from Australia. Proa. Roy. Soo. Viet. 71, 25-30. Cookson, I.C. and Pike, K.M., 1954a. A contribution to the Tertiary occurrence of Phyllaeladus and two other podocarpaceous types in Australia. Aust. J. Bot. 2, 60-68. Cookson, I.C. and Pike, K.M., , 1954b. Some dicotyledonous pollen types from Cainozoic deposits in the Australian region. Aust. J. Bot. 2, 197-219. Couper, R.A., 1953. Upper Mesozoic and Cainozoic spores and pollen grains from New Zealand. Palaeont. Bull. Geol. Surv. N. Z. 22. Couper, R.A. , 1954. Plant microfossils from New Zealand No. 1. Trans. Roy. Soo. N.Z. 81, 479-83. Couper, R.A., 1960. New Zealand Mesozoic and Cainozoic plant microfossils. Palaeont. Bull. Geol. Surv. N.Z. 32. Erdtman, G., 1960. On three new genera from the Lower Headon Beds, Berkshire. Bot. Notiser 113, 46-48. Evans, P.R. and Hodgson, E.A., 1963. A correlation of the Tertiary of A.O.G. Wentworth. No. 1, Woodside Oil Balranald No. 1 and Woodside Oil Bundy No. 1 Wells Murray Basin. Bur. Min. Res. Geol. & Geophys. Reeords, 1963/95. Germeraad, J.Q., Hopping, C.A. and Muller, J., 1968. Palynology of Tertiary sediments from tropical 46 HELENE A. MARTIN areas. i?ev. Palaeobot. Palynol. 189-348. Harris, W.F., 1955. A manual of the spores of New Zealand Pteridophyta. N.Z. Dept. Scient. and Res. Bull. 116. Harris, W.K., 1965. Basal Tertiary microfloras from the Princetown area, Victoria, Australia. Palaeontographiaa (B) . 115, 75-106. Harris, W.K., 1966a. Proteaoidites latrobensis. Taxon 15, 332-333. Harris, W.K., 1966b. New and redefined names in South Australian Lower Tertiary Stratigraphy. Geol. Sum). S.A., Quart. Notes, 20^ 1-11. Harris, W.K., 1972. New form species of pollen from southern Australian early Tertiary sediments. Trans. R. Soo. S. Aust. 96, 53-65. Krutzsch^ W., 1959. Micropalaontologische (Sporen- palaontologische) Untersuchungen in der Braunkohle des Geiseltales. Geologie 8 (Beih. 21/221, 1-425. Lawrence, C.R., 1966. Cainozoic stratigraphy and structure of the Mai lee Region, Victoria. Proa. Roy. Soc. Viet. 7^, 517-53. Ludbrook, N.H., 1961. Stratigraphy of the Murray Basin in South Australia. Geol. Sum). S.A. Bull. 36_. Martin, H.A., 1969. The palynology of some Tertiary and later deposits in New South Wales. Ph.D. Thesis. University of New South Wales. Martin, H.A., 1973a. The palynology of some Tertiary Pleistocene deposits, Lachlan River Valley, New South Wales. Aust. J. Bot. Supp. 6_, 1-57. Martin, H.A., 1973b. Urper Tertiary palynology in southern New South Wales. Geol. Soo. Aust. Special Publ. No. £, 35-54. McIntyre, D.J., 1965. Some new pollen species from New Zealand Tertiary deposits. N.Z. J. Bot. ^ 204-214. McIntyre, D.J., 1968. Further new pollen species from New Zealand Tertiary and uppermost Cretaceous deposits. N.Z. J. Bot. §_, 177-204. Mildenhall, D.C. and Harris, W.F., 1971. Status of Haloragaaidites (al Triorites) harrisii . (Couper) Harris comb. nov. and Haloragaaidites triorites Couper 1953. N.Z. J. Bot. 9, 297-306. Pels, S., 1960. The geology of the Murrumbidgee Irrigation area and surrounding districts. Bull. Hater Cons. & Irrig. Corm. N.S.W., 5. Pels, S., 1969. The Murray Basin. J. Geol. Soo. Aust. 16j 499-511. Potonie, R., 1956. Synopsis der Cattungen der Sporae dispersae. I. Sporites. Beih. Geol. Jb. 1-103. Potonie, R. , 1960. Synopsis der Cattungen der Sporae dispersae. III. Beih. Geol. Jb. 39, 1-161. Stover, L.E. and Evans, P.R., 1973. Upper Cretaceous-Eocene spore pollen zonation off- shore Cippsland Basin, Australia. Geol. Soc. Aust. Spea. Publ. No. £, 55-72. Stover, L.E. and Partridge, A.D., 1973. Tertiary and later Cretaceous spores and pollen from the Cippsland Basin, Southeastern Australia. Proa. Roy. Soo. Viat. S£, 237-286. Van der Hammen, T. and Wymstra, T.A., 1964. A palynological study of the Tertiary and Upper Cretaceous of British Cuinna. Leidse. Geol. Mededel. 183-241. Williamson, W.H., 1964. The development of ground-water resources of alluvial formation. Proc. Symp. on Water Resources use and Management, Aust. Acad. Sci., Canberra 1963, pp. 195-211. Melb. University Press. Wilson, L.R. and Webster, R.M. , 1946. Plant microfossils from a Fort Union coal of Montana. Am. J. Bot. 271-8. School of Botany, University of New South Wales, Box 1, P.O., Kensington, NS.W. 2033 (Manuscript received 20.10.1976) TERTIARY STRATIGRAPHIC PALYNOLOGY 47 APPENDIX 1 FULL TITLE AND LOCATION OF BORES The text uses abbreviated titles in referring to the bores. The full titles are given below: Balranald: Woodside (Lakes Entrance) Oil and Planet Oil Balranald No. 1 Well, 11 km W. of Balranald. Bundy: Woodside (Lakes Entrance) Oil and Planet Oil Bundy No. 1 Well, 6S km SSW. of Hay. Hay I: N.S.W. Oil and Gas N.L. Stratigraphic Test Bore No. 1, 48 km S. of Hay. Bores sunk by the Water Conservation and Irrigation Commission of New South Wales: 36025/2, 27 km ENE. of Hay 30464, 16 km N. of Carrathool 30383, 43 km SE. of Hay 30443, 1.6 km S. of Carrathool 30435, 26 km S. of Hay Wakool 36078, 19 km NW. of Wakool Wakool 36102, 10 km NW. of Wakool. Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 49 59, 1977. Leather — Why Is It So?* Edric Chaffer ABSTRACT. The fibre structure of hide, skin and leather is developed as one of the earliest fibre sciences, the fundamentals involved in its modification, and also that of its major protein constituent - Collagen - are highlighted, together with a brief outline of some processing involved in leather manufacture. Some of the important properties of leather affecting its everyday application and comfort are also elucidated. INTRODUCTION The founding fathers of the Royal Society of New South Wales, stipulated in the aims of our Society, that we are to "encourage studies in Science, Art, Literature and Philosophy, to promote and further the development of Science and allied disciplines and their applications." This presi- ■ dential address takes the form of a review and proposes to develop these objectives and to high- light their application within our everyday affairs. For although we are all fully aware that skin forms the outer covering layers of our own bodies, I would venture to suggest that few of us have given consideration to what its basic properties and functions are; or to how we use the skins of animals, converted as leather, for such everyday applications as clothing or footwear, or even for ' such a small but important usage as a metering cup washer within the carburettor of our motor vehicle. This address is on the fundamentals which are of importance in understanding about leather, and i to be able to do so, it is necessary to fully under- j stand the very important role which the original fibrous structure of animal skin continues to have on the final properties of leather. The major fibrous protein constituent of hide I and skin is the protein collagen and to produce I leather it is necessary to remove the other non- i collagenous components, and at the same time to physically and chemically modify the remaining network of collagen fibres. j Leather is considered to have been made when the pelt has been stabilized against breakdown by micro-organisms such as bacteria, fungi or enzymes. It is also resistant to the reversing action of I water and generally is also accompanied by an I increased resistance towards chemical and thermal degradation of protein. To obtain the above objectives it is very clear that it is necessary to control the physical characteristics of collagen <^ibres by chemical ‘Presidential address delivered to the Royal Society of New South Wales at Science House, Gloucester Street, Sydney, on April 7, 1976. means - and in doing this the tanner became the very first fibre scientist, for today the leather industry is recognized as the oldest of the many fibre industries. STRUCTURE OF THE SKIN Living skin is essentially composed of two distinct parts : - 1. The fibrous part. 2. The functional living organisms such as: the epidermis, the growing hairs, muscles for the hairs, the fat glands, sweat glands and connective tissue cells. In between the fibres and cells is the lymph containing soluble proteins. (Fig. 1 and Fig. 4j Epidermis Sebaceous Hair Shaft Hair Root Fig. 1. Diagram to show the general structure of ox hide. (After Sharphouse,1971j i 50 EDRIC CHAFFER unless the el-astin is digested or ruptured" it wiM restrain the even plumping and opening up of the skin, increasing the contrast between areas of the skin where it is present and those where it is relatively absent causing drawn and pebbled grain. It is one of the arts of the tanner to modify and minimize the effects produced by these elastin fibres. Epidermis • ' Epidermol -dermol jurictton ' Dermis Fig. 2. Diagram of block of skin showing epider- mis, epidermal -dermal junction and dermis. A portion of epidermis is peeled off to show papillae in the grain. (After Reed 1976) It is the fibrous part of the skin only which is of interest to the tanner, for the cellular struc tures such as the epidermis (Fig. 2) and soluble pro teins such as albumins, globulins and mucoids are removed before tanning proper begins. This is gene- rally done with solutions of hydrated lime and a re- ducing agent such as sodium sulphide, the residue remaining being removed in a bath by using tryptic enzymes. These enzymes attack the hair muscles (erector pili muscle) (Fig. 3) but not the collagen fibres as these are resistant to most enzymes except collagenase. The fibrous part of the skin is composed mainly ot collagen fibres occurring as fibre bundles being generally larger in the corium and smaller in the grain layer. In a sheath surrounding the collagen fibres is to be found the so-called "reticular tissue". This is currently understood to be a stage in the forma- tion of collagen, known as pro-collagen, but having differing chemical properties. The amino acid composition is similar to that of collagen, but it contains less proline and hydroxyproline and has a greater molecular weight. Also present in the skin structure are elastin fibres. These occur particularly in the grain layer and in the flesh layers. The amount and condition of elastin present is of considerable importance to the properties of leather as collagen and elastin differ in their response to swelling agents. In the course of pre- tanning processes the collagen swells more than the elastin so that the elastin is under stress. Thus, - Fig. 3. Vertical section grain showing epidermis erector pili muscle and sebaceous gland, (after Wilson, 1928) The involuntary hair muscle known as the erector pili muscle is also important (Fig. 3). In life it is affected by shock and cold which causes the hair to stand upright and also gives rise to "goose pimples". This ability of the erector pili muscle to contract still remains after death and can in cold conditions give rise to a coarseness of the grain surface of the hide which would be -permanently fixed in leather by tannage. To overcome this defect partial digestion of the erector pilli muscle by bating is carried out with tryptic enzymes. Bating also removes other unwanted proteins and helps to modify reactive sites on the long molecular polypep- tide chains of amino acids which go to make up the collagen fibrils, for it is on these reactive sites that the tannin molecules combine. Each individual hair or wool fibre is contained within its own individual hair follicle with its own fat gland (sebaceous gland) . This is important in sheepskins, particularly when they come from animals bred to give much greater yields of fine wool fibres, for under these conditions the skin has developed a layer of fat glands just under the grain layer leav- ing little room for the cross connecting fibres to connect the grain and the corium. This can result after the removal of the fat in a two layer effect in the skin and also a very open texture (Fig. 6). This defect will lead to greater softness and can make these particular types of skins suitable for clothing leather or oil tanned "chamois" leathers although unsuitable for many other uses. LEATHER-WHY IS IT SO? 51 Fig. 4. Vertical section of cow hide x 16 (After Wilson, 1928) Fig. 6. Vertical section of sheepskin x 50 (After Wilson, 1928) Fig. 5. Vertical section of goatskin x 50 (After Wilson, 1928) Fig. 7. Vertical section of pigskin x 20 (After Wilson, 1928) 52 EDRIC CHAFFER As a generalization for sheepskins we could say: the coarser the wool the better the quality of the skin. These fibres can be further split into individual fibres and these fibres can be further split into individual fibrils. The fibre structure of hides and skins also vary due to the following factors: - (a) Type of animal species. (b) From breed to breed within a species. (c) Age of the animal. (d) Sex of the animal . (e) Feeding conditions and environment. (f) Considerable variation over different parts of the same skin. Figures 4, 5, 6, 7, 8 and 9 show the effect of animal species upon the fibre structure of the following types of skin - cattle hide, goatskin, sheepskin, pigskin, kangaroo skin and snakeskin respectively. It will be noted particularly in the case of reptile such as snake, how we have a very horizontal weave pattern. This imparts very great strength but also means that it does not stretch in a shoe upper when in use like cattle hide. Therefore it is extremely important to have a good fitting for shoes at time of purchase. Fig. 10. Vertical section showing fibre bundles of calf skin. (After Wilson, 1928) This degree of splitting and opening up of the fibre bundles into fibrils is of fundamental import- ance to the properties required in the final leath- er. (Fig. 11) A Fig. 11. and fibrils. As an anology the fibre bundle of hide and skin can be likened to textile yarn. A textile yarn spun from coarse fibres and woven compactly will give a hard fabric with a firm harsh handle, which will nevertheless be strong. To produce a soft fabric, the textile manufac- turer chooses fine fibres and spins them into yarns with a minimum of twist to give the strength requ- ired and weaves the yard loosely into a fabric. When a tanner wishes to produce a solid firm leather he splits the fibre bundles as is indicated in Figure 11 at A, when he wishes to produce a soft pliable leather he splits them as indicated at C. If either the fibres or fibrils become separated as in B or D the leather looses quality. A sole leather with fibres split as in B will not be firm. An upper leather with fibres separated as in D will be weak and without fullness and handle. The unhairing and liming stage of leather processing is of great importance in the preparation for tannage as it decides the characteristics of the fibres which are to be tanned. In a sense, liming enables the tanner to vary the physical character- istics of the leather he produces by varying the diameter of the individual fibres that he tans. Liming will also decide the angle of weave of the fibres, that is whether the major proportion of the fibres and fibrils tend to lie parallel to the grain layer or whether they tend to run from flesh to grain. The corium is of major interest in leather pro- duction as it consists largely of a three-dimen- sional network of interwoven collagen fibres. It will be noted from Fig. 10, showing the cross section of cattle hide that the individual fibre bundles can be split up into individual fibres. In producing sole leather the fibre bundles should show subdivisions into fibres which have not separated, thus giving greater firmness. Also the fibre bundles of the middle section of the corium should be of a very much higher angle of weave. This will mean that the fibres have a very much greater abrasion resistance due to the fibres being LEATHER-WHY IS IT SO? 53 worn across their end sections rather than along their lengths. In the case of a leather for transmission belt- ing, which is expected to be flexible and of high tensile strength, it is desirable for the fibres to be subdivided into fibrils to obtain flexibility, also the flat or horizontal weave of the fibrils is desirable as it will enable the belt to withstand any pull to which it may be subjected in use. When extreme flexibility is required as in picking bands which are used in the textile industry, the fibres should be split completely into fibrils. It is of interest to consider the size of these fibres and fibrils, which may vary in thickness from 200.000 nanometers (i.e. 2,OOO,OO0 S) to 3.000 nanometers (i.e. 30,000 S) COLLAGEN Collagen is a complex protein which in living animals is synthesized by cells called fibroblasts and is secreted into the intercellular spaces where it aggregates to form fibrils. The newly emerging collagen is water soluble but is progressively con- verted into less soluble and more stable forms by cross linking between acid and basic side chains of adjacent molecules, and by hydrogen bonding between the side chains and between the polypeptide back- bones . In the years preceding 1950 various molecular structures were proposed for collagen, but in the year 1951 Pauling and Corey showed these to be untenable. The currently accepted model for collagen is a coiled coil or super helix composed of three helical polypeptide chains which are coiled into a larger helix, the whole rope-like structure being stabilized by interchain hydrogen bonding and also hydrophobic bonding and electrostatic interactions. (Fig. 12) Fig. 12 showing twisting of three polypeptide chains into a rope-like super helix or coiled coil. It has been shown that each of the three poly- peptide chains in the molecule contains 1,000 amino acid residues, and that collagen contains two unique features in that every third residue in the chain is glycine and in the non-polar regions of the molecule one of the other residues in each group of three is either proline or hydroxyproline. Collagen is characterized by the presence of the uncommon amino acids, namely hydroxyproline and hydroxlysine and the presence of very low cystine contents. (Fig. 13) AMINO ACID COMPOSITION OF COLLAGEN (Fig. 13) (All results expressed on moisture- and ash-free basis) Amino Acid g.per 100 g. g . residues per 100 g. Total Nitrogen 18.6 Amino-Nitrogen 0.46 - Amide-Nitrogen 0.66 - Neutral Amino Acids: Glycine 26.2 19.9 Alanine 9.5 7.6 Leucine ) Isoleucine) 5.6 4.8 Valine 3.4 2.9 Pheylalanine 4.2 3.7 Tyrosine 1.4 1.3 Tryptophane 0.0 0.0 Serine 3.4 2.7 Threonine 2.4 2.0 Cystine 0.0 0.0 Methionine 0.8 0.7 Proline 15.1 12.7 Hydroxyproline 14.0 12.1 Basic Amino Acids: Lysine 4.5 4.0 Hydroxylsine 1.3 1.1 Arginine 8.8 7.9 Histidine 0.8 0.7 Acidic Amino Acids: ■Aspartic acid 6.3 5.5 Glutamic acid 11.3 10.0 99.6 It may also contain a fair proportion of the amino acids in which the reactive group bears either a carboxyl group, an amide group or an amino group, which means that collagen has at the same time re- active amino acids sites which are acid, basic and non-polar . The number of molecular chains which make up a unit fibril can be calculated and has been estimated at 20,000 for one unit fibril of cow hide collagen of 110 nanometers in dimension. It is noteworthy that they have no fixed length, although from electron microscope photomicrographs we know that they are very long in relation to their width. It has been proved with collagen fibrils formed from solutions with a width of 100 nanometers and extending for lengths up to 10,000 nanometers that they contain up to 100,000 molecules. The rigid molecular structure and the highly ordered arrange- ment gives high mechanical strength to the fibrils and also explains the resistance of the collagen molecule to attack by enzymes and other reagents. There is almost no digestion of the molecule by pro- teolytic enzymes at temperatures below the denatura- tion temperature of the protein. 54 EDRIC CHAFFER Collagen can occur in either a solution of the solid phase, as collagen can be denatured by heat- ing, the hydrogen bond which stabilizes the helical structure is disrupted and the molecule is converted to a random coil formation which is known as gelatin. Very few people outside the leather industry real- ize that the study of the relationship between gela- tin and collagen and that they can be interconver- ted inspired much of the early work in the synthe- tic fibre industry. INFLUENCE OF STRUCTURE ON ACID AND ALKALINE SWELLING In briefly looking at the chemistry involved in liming and tanning we find that we must look at the response of collagen fibres to alkaline and acid solutions . When placed in water collagen takes the water up freely to about 300 per cent of its own weight. This water attaches itself to accessible active groups and peptide linkages which are not linked firmly to their opposite numbers in the neighbour- ing polypeptide chains. When acid is added to water it reacts with the collagen to form a collagen salt and the sites at which these salt linkages take place are the salt linkages within the collagen. For example, choosing hydrochloric acid; - After the salt formation, the chloride anion remaining associated with the substituted ammonium cation will cause the water in the collagen fibres to be more highly concentrated with respect to the chloride anion than the surrounding solution. In response to the excess osmotic pressure within the collagen to which this inequality in concentration gives rise, the water passes into the fibre and the collagen swells. A similar change occurs when sodium hydroxide is added to water in which collagen is immersed, in this case the hydroxyl anion from the alkali removes the hydrogen ions from the substituted ammmonium cations of the salt linkages and the carboxyl anion being left uncharged the polypeptide chains acquire a negative charge. The water within the collagen becomes highly concentrated with respect to the so- dium cations associated with the carboxy anions and water passes into the collagen and causes it to swell . The swelling of such fibres is restricted by the structure of the hide and skin, as is seen in the following Fig. 14. The lower of the three graphs gives the swollen weight of goatskin and pieces of belly and butt of ox hide after they have been immersed in acid and alkaline solutions of differing pH values. At the natural pH values the original salt linkages are intact, the swelling is at a minimum. As the pieces become strongly positively charged in acid solution or strongly negatively charged in alkaline solution the water passes into the fibres - on the acid side it reaches a maximum at pH2. It then falls because the concentration of anions in the outside solution is high and is not greatly exceeded by the concent- ration of anions within the fibres. The curves show that at all pH values the compact goatskin (Fig. 5^ swells less than the ox hide, further they show that the compact ox butt swells less than the belly part of the hide in which the fibres are less compactly woven. Fig. 14. The influence of hide and skin structure on water uptake at different pH values. (After Phillips, 1954) The top curve depicts gelatin which is very closely similar in chemical constitution to collagen, from which it is produced by forcing apart the long chain molecules. The absence of ordered and oriented molecular structure facilitates the entry of water between the polypeptide chains. As the curves show in the case of gelatin, water uptake at all pH values far exceeds that of either goatskin or ox hide. Nevertheless the shape of the curve shows that the chemical mechanism causing swelling is the same in non-fibrous gelatin as in the fibrous hide and skin. INFLUENCE OF LIMING ON WATER UPTAKE The swelling of hides and skins in acids and alkalis is of great importance to the tanner, for if a hide or skin is highly swollen the fibres will press against one another and the interfibril lary spaces will be closed. It then becomes very diffi- cult for tannins to diffuse into the hide and skin causing the tanning to be retarded and maybe arres- ted. In addition when the hides and skins swell the molecular structures of their fibres are placed under strain. During liming when the fibre bundles swell they also split into fibres, and the conditions of liming can be adjusted so that the fibres split into fibrils. But over and above these changes, liming opens up the molecular structure of the fibrils. Such opening up, which takes the form of forcing the molecular chains apart, would enable the unit fibrils and therefore the fibres of which they form part, to swell more freely and to take up more water (Fig. 15) It will be noticed that at all pH values the limed alkali-treated collagen takes up more water than the native untreated collagen, part of this increase becomes possible because of the loosening of the fibrous structure during liming, but the bulk is due to a loosening of the molecular structure of tne fibres and fibrils. This molecular loosening or LEATHER- WHY IS IT SO? 55 or Iqs* of cohesion is of great importance in tanning. Fig. 15. Water uptakes of native and alkali- treated collagens. (After Phillips, 1954) CHEMISTRY OF UNHAIRING Unhairing generally is concerned with the dis- persion of fibrous proteins: the tanner has to dis- perse the hair without weakening the collagen fibres of the hide. Hair and wool fibres are built up from cells which are much longer than they are thick. These cells are composed mainly of keratin, a protein which is similar in chemical composition to colla- gen. The polypeptide chains of hair contain, how- ever, much less proline and hydroxyproline, and they are not extended like collagen polypeptide chains, but are folded. These folded polypeptide chains of hair contain a double amino acid cystine, which is built into two polypeptide chains, linked together by a non-ionizable link. As the curves given in Fig. 16 show, this cross-linkage greatly restricts the swelling of the hair in both acid and alkaline solutions, for hair, like silk takes up much less water than collagen. Fig. 16 The degree of swelling of fibres of different materials at different pH values. (After Phillips, 1954) These cross-linkages can be broken chemically and then the fibre takes up water as freely as collagen. Sodium sulphide brings this change about because it is a reducing agent and reduces the cystine linkage of the cystine and so breaks the linkage between the two polypeptide chains. Fig. 17. A wool sodium fibre dispersed by sulphide a solution of 1 NH 1 1 NH 1 Reducing agent 1 NH 1 1 NH 1 CH.CH S 1 2 SCH..CH 2 1 CH.CH. SH 1 2 HSH.,.CH 2 1 1 CO 1 CO 1 CO 1 CO I As is shown above each polypeptide chain retains a cystine side chain. This is an acid side chain and will therefore contribute to the osmotic swelling of the fibre and facilitate its breakdown. (Fig. 17) Tanners found that if they put hides into a freshly made lime suspension (calcium hydroxide) it was difficult to loosen the hair later with sodium sulphide, but if they put the hides into a previous- ly used lime liquor before liming in a new lime liquor the unhairing proceeded satisfactorily. Chemists explain this as due to the old lime contain- ing reducing agents derived from the cystine libe- rated during the degradation of hair from previous packs of hides - these reducing agents would reduce the cystine cross linkages. But why should it resist the action of reducing agents? The answer is given below: - NH NH NH SCH .CH ^ 1 1 CH.CH 1 2 1 S CH CH+S 2 1 1 CO 1 CO 1 CO When the hair is treated with alkali the com- bined cystine looses an atom of sulphur and changes to a sulphide amino acid called lanthionine - the new sulphide cross-link formed in this manner is very stable and cannot be broken by reducing or oxidizing agents. 56 EDRIC CHAFFER VEGETABLE AND MINERAL TANNING Collagen fibres are very susceptible to acid and alkaline waters, in addition bacteria can attack polypeptide chains breaking them down into their constituent amino acids. The function of tanning is to make the fibres of the hides and skins resis- tant to water and bacteria. We can, of course, make a kind of leather by dehydrating raw limed hide with acetone. The acetone replaces water in the fibres, and when the hide dries and the acetone evaporates the fibres do not stick together and the hide remains pliable. If this pliable piece of hide is re-wetted and then dried it becomes hard and homy; the wet fibres on drying have stuck together. Another reason for tanning is therefore to prevent the fibres from sticking together as well as to protect them from water and bacteria. We can, however, only make leather by protec- ting the molecular structure of the collagen fibres. The electron microscope and the small angle X-Ray diffraction patterns of collagen fibres have enabled us to see how tanning protects the molecular struc- ture of the collagen. In vegetable tanning we use extracts obtained from bark, wood, nuts, seed pods and leaves. One very important material is the tanning extract known as "Mimosa" derived from various Australian species of wattle which are today largely grown in South Africa. The vegetable tannins have hydroxyl and car- boxyl groups which can form hydrogen bonds and salt linkages with the molecular chains of collagen. The tanning molecules can also become linked to one another by hydrogen bonds - so that we can therefore picture the first stages of tanning as an associa- tion between tannins and exposed molecular chains. This is followed by association betweened fixed tan and other tanning molecules until all the available free space has been filled, when this takes place the fibre is tanned. Other synthetic tannins, known as syntans are also manufactured from the cheapest raw materials such as crude cresols condensed with formaldehyde, the resulting product then being sulphonated. Still more complex syntans are made by condensing simple syntans with sulphited cellulose, unfortunately these products have only about half the tanning ability of natural vegetable tannins at about twice the cost, although they can give some special effects which are not otherwise obtainable. CHROME TANNING Collagen however can be tanned with a far smaller quantity of mineral or inorganic tannins; for example, good shoe upper leather can be made using the most common inorganic tannin - basic chro- mium salts - when the pelt absorbs as little as two per cent of the tannin calculated as Cr^Oj. Currently single bath liquors are prepared by the reduction of solutions of sodium bichromate with sulphur dioxide or alternatively with acidified sodium bichromate solutions with glucose, the latter giving rise to the production of "masked" chrome 1 iquors . MlO^ ^OM M|0 \l/\ / \ / '^H.O OM 'M|0 The above formulae give some indication of how chrome tannins are formed and aggregate. Two hydrox- alated (CrOH) complexes are assumed to combine by displacement of two water molecules co-ordinated to the cationic chromium. This process can be repeated with other (CrOtl) units so that long chained aggre- gates of hydroxylated chrome tannins could arise. Additional complexity is introduced if the , primary units are dihydroxylated (Cr^COH)^. Basic complexes of aluminium and zirconium can also be produced to act as tannins. Mineral tannage stabilizes collagen by produc- ing non-ionizable cross linkages between polypeptide chains. These arise by the co-ordination of the carboxyl groups of the aspartic and glutamic acid side chains with the chromium atoms of the cationic chromium complexes. It is possible, although still uncertain, that the terminal amino groups of the lysine and argining side chains also co-ordinate with the chromium atoms. SCIENTIFIC ASPECTS OF USE OF LEATHER IN FOOTWEAR AND CLOTHING The use of leather for footwear and clothing is very much bound up with its hygroscopic nature. The tanner sometimes states that leather "breathes" and can demonstrate this by blowing air through the leather, but what is a more important constituent of leather "breathing" is its water va- j pour permeability, for we cannot achieve body or I foot comfort unless perspiration can be lost by evaporation. ' The amount of heat lose by the body or feet by evaporation of perspiration is dependent on the relative humidity of the air with which it is in contact . Let me remind you that relative humidity is the percentage ratio of the air to the water content at saturation. When the relative humidity is low the air is dry, and when it is high the air is damp. The relative humidity depends on temperature as well as on the moisture content of the air, because the amount of water needed to attain saturation increases rapidly with temperature. The air on an autumn morn- ing may be damp with dew forming on the grass, but by noon it may be very dry, not because the moisture content of the air has decreased, but because the temperature has increased. At dawn the temperature is low and the saturation water content is low, giv- ing a high relative humidity and dampness, whereas at noon the temperature is high and the saturation water content is high, giving a low relative humidity and apparent dryness. The water content in a given locality is in fact surprisingly constant, and the changes from dryness to dampness are largely reflec- tions of temperature changes. LEATHER— WHY IS IT SO? 57 Leather is hygroscopic and the amount of mois- ture it can hold depends on the relative humidity of the air with which it is in contact. Untanned skins including our own are also hygroscopic . From the standpoint of foot and body comfort and on account of the properties of leather fibres, it is of value to consider in detail the absorption of water vapour by collagen and leather. The water vapour uptake of collagen at diffe- rent temperatures and relative humidities is similar to those of other fibres such as nylon, silk, wool, vegetable tanned sole leather and chrome tanned upper leather (Fig. 18). Fig. 18. The absorption of water vapour at diffe- rent relative humidities by chrome leather (C) ; vegetable leather (L) ; wool (W) ; silk (S) ; nylon (N) (After Phillips, 1954) Leather is hygroscopic and the amount of mois- ture it can hold depends on the relative humidity of the. air with which it is in contact. The similarity of the shapes of all the curves is a reflection of the molecular structure of the fibres. The curves are all of this particular shape, because the water vapour which enters each fibre is held in two different ways. One being water directly linked to the mole- cular chains in each fibre. This water is some- times called "bound" water and accounts for the quick initial uptake, the other being water which at higher relative humidities, becomes more loosely linked to the bound water of the fibres. From Fig. 18 it should be noted how much more moisture at all relative humidities the chrome tanned and vegetable tanned leathers absorb than the textile fibres, wool, silk and hylon. In addition the chrome tanned fibre absorbs more water than the vegetable tanned leather because the fibres are not packed with tan molecules. We find that the real density of chrome leather steadily increases until it reaches a moisture con- tent of sixteen per cent on dry weight, as the mois- ture content increases still further, its real density decreases. An explanation of this initial increase and later decrease in real density with an increase in moisture content is that the regions where the chrome tannins have combined with the collagen have not been specially completely filled with tannin, and hence up to sixteen per cent moisture can pack into these regions without increasing their volume, thus the real density of the fibre increases. Above an absorption of sixteen per cent the water begins to increase the volume of the leather fibres and the density of the leather decreases. The additional water has to make room for itself by separating the molecular chains. This increases the volume and hence the density decreases. THERMOSTATIC PROPERTIES OF LEATHER Not only is leather enabled by its hygroscopic properties to absorb perspiration and to breath it out to the atmosphere,’ but in doing so it tends to oppose rapid changes in temperature (Fig. 19). Time , min Fig. 19. The rise in temperature of chrome leather during absorption of water vapour. (After Phillips, 1954) When damp leather looses moisture and comes to equilibrium with air of low relative humidity it becomes colder. The reason for this is that the leather supplies heat to the moisture, which is carried away as kinetic energy by the water vapour. Conversely when dry leather absorbs water vapour, it becomes warm, because as the water vapour is taken up as bound water it gives up its kinetic energy as heat. Let us consider what happens when leather clothing or shoes are taken on a winter's day from the warm dry air of a heated but well ventilated room to the cold damp air outside. Whilst inside the leather comes to equilibrium with the air of low relative humidity and its moisture content falls. Outdoors, the relative humidity is high and 58 EDRIC CHAFFER the leather begins to absorb water vapour, heat is liberated and this raises the temperature of the air in the leather. Temperature rises from twenty- five degrees Celsius to thirty-nine degrees Celsius in two minutes (plus fourteen degrees Celsius) . It is then cooled by air passing through and even after forty minutes its temperature exceeds twenty-five degrees Celsius. INFLUENCE OF MOISTURE CONTENT ON THE AREA AND STRENGTH OF LEATHER. Another important aspect of leather bearing on foot comfort is its dimensional stability to changes in moisture content caused by changes in relative humidity. In measuring changes in dimensions of chrome and vegetable tanned upper leathers, it has been found that chrome tanned leather is much more sensitive to changes in relative humidity. This means that leather shoes with chrome upper leather in damp weather will be too big and in dry weather they will be too tight to be comfortable. This has indeed solved the riddle of the weather prophet who foretells changes in weather by the pains in his corns, as the corns are merely respond- ing to the shrinkage of his shoes in dry weather. The moisture content of leather fibres has a pronounced influence on their strength. The strength of chrome leather increases with its moisture con- tent being in this regard similar to cotton, flax, hemp fibres - leather becomes weaker at lower rela- tive himiidities. In one example of a chrome leather where the moisture content varied from zero moisture to seventy per cent moisture - at one hundred per cent relative humidity the leather strength varied from a loss of forty per cent to a gain of thirty per cent - i.e. there was a seventy per cent variation in strength from zero to seventy per cent moisture content . In the case of vegetable tanned leather there occurred little difference in strength over the full range of moisture. THE CONTRIBUTION MADE BY THE FIBROUS STRUCTURE OF LEATHER TO FOOD AND BODY COMFORT Most people if asked would be puzzled why new synthetic fibres for clothing materials are not used in a continuous sheet instead of taking the trouble of converting them into fibres for spinning. The reason is that by using fibres which are spun into yarns and then woven into cloth, fabrics are pro- duced which are light in weight and of good tear resistance and flexibility, in addition such fabrics are good insulators against heat and cold. The fibres of fabrics entrap air which clings to the fibre surfaces and hence is more or less stat- ionary. Stationary air is a poor conductor of heat. The bulk density of both leather and wool felt is below 0.05 and their thermal conductivities do not greatly exceed that of air. This is not surprising, since it can be calcu- lated that the air |pace in wool felt of bulk density 0.38 per cm is seventy-seven per cent of the volume of the felt. Only in the more dense leathers and felts do the higher thermal conduc- tivities of the fibres themselves come into play, the thermal conductivities of wool being ten times that of air, while the thermal conductivity of leather fibre is somewhat higher. The bulk density of a leather largely determines the rate at which it transmits water vapour and therefore the rate at which it carries perspiration away from the foot. It is not always realized how much perspiration leaves a normal foot. ] The soles of feet perspire at five to ten times I the rate at which perspiration leaves the general ' body surface. ' The average rate of perspiration of the whole foot when at rest is six to seven milligrams per square2centimetre per hour. For a man's foot of 550 cm surface area this gives a total of 39.6 g. of perspiration moisture per twelve hour day. On a warm day with moderate exercise this rate is more than doubled. The ability of leather to absorb and transmit water vapour makes a very important contribution to foot health and comfort. CONCLUSION In bring this address to a close, I hope that I have made you realize a little of the science involved in such an everyday commodity as leather and the hides and skins from which it is derived. I trust that I have enabled you to recognize leather as one of the fibre industries and how knowledge of the chemistry and physics of matter in fibrous form . is rapidly growing, and also how the science of leather-making can explain the traditional methods of tanning. , ACKNOWLEDGEMENTS I wish to acknowledge my early guidance from the late Dr. Henry Phillips who in 1956 in his visit to > Australia highlighted this approach to leather as a fibre science. LEATHER- WHY IS IT SO? 59 REFERENCES Collagen and Leather Research in PROTEIN CHEMISTRY, REVIEW AND ABSTRACTS, 1973, pp. 3-53, CSIRO, Division of Protein Chemistry, Melbourne, S3 pp. Dempsey, Mary, et al, 1957. HIDES, SKINS AND LEATHER UNDER THE MICROSCOPE. The British Leather Manufacturers Research Association, Egham, 368 pp. Grassman, W. (Ed.), 1944, Die Haut in HANDBUCH DER GERBEREICHEMIE UND LEDERFABRI- KATION, 1(1), 1114 pp. Jordan-Lloyd, D. et al, 1948. PROGRESS IN LEATHER SCIENCE 1920-1945, British Leather Manufac- turers Research Association, London, 705 pp. Jordan-Lloyd, D. , 1943. LEATHER, Lecture on. The Royal Institute of Chemistry of Great Britain and Ireland, London, 31 pp,. O'Flaherty, F, Roddy, W.T. and Lollar, Robert M. , 1956. THE CHEMISTRY AND TECHNOLOGY OF LEATHER, Vols. 1-4. American Chemical Society Monograph Series, No. 134, Reinhold Pub. Co., New York, c/- W. Chaffer and Sons Pty Ltd 72 Gibbes Street CHATSWOOD. NSW 2067 Phillips, Henry, 1954. The chemistry of leather. Three Cantor Lectures on, J.R. Soa. Arts, 4934, (CII), 824-75. Reed, R. , 1972. ANCIENT SKINS PARCHMENTS AND LEATHERS. Seminar Press Ltd., London, 331 pp. Reich, G. , 1966. KOLLAGEN. Verlag Theodor Stein- kopff, Dresden 308 pp. Sharphouse, J.H., 1971. LEATHER TECHNICIAN'S HAND- BOOK. Leather Producers' Association, London, 349 pp. Stather, F., 1957. GERBEREICHEMIE UND GERBERI- TECHNOLOGIE. 3rd edn. Akademie-Verlag, Berlin, 948 pp. Wilson, John Arthur, 1928. THE CHEMISTRY OF LEATHER MANUFACTURES. 2nd edn, Vols 1 and 2, American Chemical Society Monograph Series No. 12, The Chemical Catalog Co. Inc., New York, 1181 pp. (Manuscript Received 25.4.77) Report of Council for the Year Ended 31st March, 1977 61 INTRODUCTION The past year has been a momentous one in the annals of the Society in that it has witnessed the transfer of our activities from Science House, Gloucester Street to the new Science Centre in Clarence Street. Although the Society's financial situation is still a matter of considerable concern to your Council, actions taken during the year have led to real economies and a reduction in the deficit. Hopefully this will continue and although a rapid improvement is unlikely in the immediate future the long-term prospects are good. MEETINGS AND LECTURES Council continued its policy of inviting speakers expert in their field to deliver lectures at our monthly meetings, the lectures being pitch- ed at the level of the well-informed layman. All meetings and lectures held during the year were well supported, the average attendance in excess of 45 is considered to be a vindication of this policy as well as being a reflection of the excellent standard maintained by our lecturers. Pollution Control Commission; Dr. L. L. Pownall, Chairman, N.S.W. Planning Environment Commission; Dr. A.J. Sutton, Director, Bureau of Crime Statistics and Research, N.S.W. Department of Attorney General and Justice. In addition, the following functions were held within the University of Sydney; June 16th: "An Evening at the Kacleay", Dr. P. Stanbury, Director, and Staff, the Macleay Museum. July 15th: Liversidge Research Lecture 1976: "Coordination, Topology and Structure in Metal Oxides", Professor R.L. Martin, Research School of Chemistry, Australian National University. ANNUAL DINNER The Annual Dinner was held in the Sydney Hilton Hotel on 15 March and was attended by 106 members and guests. The guest speaker was Dr. Harry Windsor, F.R.A.C.S. , the title of his address being "Surgery in Today's Society". The following meetings and lectures were held during the year, the venue for those from April through to August being the Large Hall, Science House, whereas for those held subsequently it was the Auditorium, Science Centre. April 7th: Annual General Meeting and President- ial Address, "Leather - Why is it so?", Mr. E.K. Chaffer, Director, W. Chaffer and Sons Pty. Ltd. May 5th: "Island Arcs and Ore Deposits", Professor R.L. Stanton, Department of Geology, University of New England. June 2nd: "Human Nutrition - Science or Trans- Science", Mr. M.V. Tracey, Chief, C.S.I.R.O., Division of Food Research. July 7th; "Science and Art: Some Issues", Mr. Elwyn Lynn, Curator, Power Gallery of Contemporary Art, the Power Institute of Fine Arts, University of Sydney. August 4th: "Some Spacecraft I have known". Dr. K.G. McCracken, Chief, C.S.I.R.O. , Division of Mineral Physics. September 1st: "Light and Colour", Dr. W.R. Blevin, Chief Research Scientist, C.S.I.R.O., National Measurement Laboratory. October 6th: "Some Possible Staging Places in Aboriginal Australian Migration from South- East \sia". Emeritus Professor N. W. G. Macintosh, Department of Surgery, University of Sydney. November 3rd: "Aboriginal Health and its Relationship to Vitamin C" , Dr. Archie Kalokerinos. December 1st; Symposium, "People, Planning and the Environment", Mr. R.P. Murphy, Assistant Director, Pollution Control, N.S.W. State AWARDS The following Awards for 1976 were made: Edgeworth David Medal: The Society's Medal: Clarke Medal; James Cook Medal : Liversidge Research Lectureship: Archibald D. Olle Prize: Summer School in Medic Mr. S. Harvey, Miss K. Ratanayagam. Professor R.H. Street Mr. E.K. Chaffer Dr. Lilian R. Fraser No Award Prof. R. L. Martin Dr. L.A. Drake ine Essay Prizes: Byron, Miss E. SUMMER SCHOOLS The two Summer Schools held during January for fifth-form students again proved to be most reward- ing for all concerned. The school having chemistry as its theme and entitled, "Chemistry and Colour", was held, as previously, at Macquarie University. It was attended by 59 students from 43 secondary schools. The second school entitled, "Medicine and Health Services", attracted 66 students from 63 schools. This school was based at Royal Prince Alfred Hospital and included visits to other medical institutions as well as State Government laboratories. MEMBERSHIP The membership at 31 March was: Honorary Members, 11; Life Members, 35; Members, 365; Associates, 60; Company Member, 1. The number of applications for membership received during the year was disappointingly small, none being received through the Summer Schools, and surely must be a reflection of the prevailing economic climate. Emeritus Professor P.A. Elkin and Sir Robert Price were elected to Honorary Membership. 62 REPORT OF COUNCIL PUBLICATION Volume 109 of the Journal and Proceedings was published during the year using the "type-it- yourself process. This innovation has proved most successful and has succeeded in holding publication costs for the time being. Dr. Day made on the Society's behalf a written submission to and personal appearance at the Industries Assistance Commission enquiry into the publishing industry. The Commission has yet to present its report to the Government. LIBRARY Demand on the library's resources was strongly sustained during the year. 2368 items were receiv- ed and processed. These comprised periodicals on exchange from some 356 societies and institutions, together with donations and periodicals purchased. 181 members, societies and other organizations used the library facilities during the year. Restricted finance allowed the library to be opened only two full days per week. The Librarian, Mrs. G. Proctor, maintained all the library's services under diffi- culty, especially during the transfer to Science Centre. The Library was moved to its new location in the 2nd week of December by the intensive effort of volunteer members and a team of students. FINANCE The accompanying financial statements show that a deficit of $11,771 was incurred on operations during the year. This decline in the Society's liquid resources was regretfully fore- seen by the Council, although unexpected factors made it larger than was budgeted for. No Government grant was received (the first such omission in about 110 years), and the Science Centre project was still in an establishment phase and therefore did not contribute any return to the Society. An additional burden, in excess of $5000, resulted from the need to move the office and library to the Science Centre late in the year, about two years earlier than previously anticipated. The resignation in April of the executive secretary, Mrs. V. Lyle, to proceed overseas, while much regretted, provided an opportunity to effect significant economy in salary payments by restricting office hours to twelve per week. A short trial quickly showed that no useful saving could be achieved by reduction of the librarian's hours, without very seriously disrupting the library service. The greatest possible economies were applied wherever practical to all the operations of the Scfciety and the voluntary work of councillors, ordinary members and friends of the Society assisted considerably in minimizing the deficit. Operations generating their own income were required to at least break even. The change to offset printing of the Journal and Proceedings saved about $5000. The net result of the year's operations is that the balance of the Resumption Reserve, established to support the Society following cessation of income from Science House, has now been exhausted and the remainder of the deficit was financed from the Society's general funds. (In comparing the 1976 figures with those for 1975 in the financial statements it should be noted that the 1975 financial year covered only ten months due to the change in closing date to 31st December. ) The response to the appeal for donations to assist in the continuation of long-standing subscriptions to journals for the library was most gratifying, yielding $1300. The bulk of this was invested to provide continuing income in future years. All donations of $2 or more to the "Royal Society of New South Wales Library Fund" are tax- deductible and will help to cover the cost of housing and operating the library. In 1977 this is anticipated to be close to $5000. SCIENCE CENTRE The new Science Centre was officially opened on 23 March, 1977, by His Excellency the Governor of New South Wales, Sir Roden Cutler. Lady Cutler was also present as were Mr. N. Wran, the Premier of N.S.W., Sir Eric Willis, Leader of the Opposition in State Parliament, and other distinguished guests. The Centre is now fully operational, the Society's Office and Library being housed on the 6th floor. Also on this floor is the office of Science House Pty. Ltd. which, in addition to managing the Science Centre, provides secretarial services for kindred societies. ROYAL SOCIETY ASSOCIATES Two years ago when this group was formed, largely from students who had attended the previous Summer Schools, Council held high hopes that it would prove to be a source of new members and life- blood for the Society. Unfortunately this hope has not been realized and this year the group has been forced into recess through lack of support. ACKNOWLEDGEMENTS Council wishes to acknowledge the excellent work carried out during the year by Mrs. Judith Day who attends to our secretarial affairs, to Mrs. Grace Proctor, the Assistant Librarian and Mrs. C. McKay. It also wishes to record its appreciation for the work carried out by all who were responsible for the success of the Summer Schools and monthly lectures. In addition. Council wishes to acknowledge the work of Dr. Alan Day, who managed the transfer of the Society from Science House, and also that of all his recruits who ably assisted in this task. CORRIGENDUM Report of Council for Year Ended 31st March, 1976. From the section headed Awards delete, "Liversidge Research Lectureship: Dr. R.L. Martin", and substitute, "Clarke Memorial Lecture, 1975: Dr. K.S.W. Campbell". REPORT OF COUNCIL 63 ANNUAL REPORT OF NEW ENGLAND BRANCH OF THE ROYAL SOCIETY OF NEW SOUTH WALES OFFICERS Chairman ; S.C. Haydon Secretary Treasurer : R.E. Gould Committee : R. L. Stanton, N.T.M. Yeates, R.D.H. Fayle, N.H. Fletcher Representative on Council : N.T.M. Yeates The following meetings were held : 18th June, 1976 : "Metals in Water", Dr. D.J. Swaine, C.S.I.R.O. Division of Mineralogy, Sydney. 21st October, 1976 : "Some Spacecraft I have known". Dr. K.G. McCracken, C.S.I.R.O. Division of Mineral Physics, Sydney. FINANCIAL STATEMENT Balance as at 31st March, 1976 $278.48 Credit - Interest to 29.6.76 4.91 - Interest to 31.12.76 4.11 $287.50 Debit - Air Fare for speaker - Miscellaneous $ 69.00 14.00 $ 83.00 Balance as at 31st December, 1976 $204.50 ANNUAL REPORT OF THE SOUTH COAST BRANCH OF THE ROYAL SOCIETY OF NEW SOUTH WALES OFFICERS Chairman : B. Clancy Secretary : G. Doherty Representative on Council : G. Doherty. No meetings of the Branch were held during the year. FINANCIAL STATEMENT Balance as at 31st March 1976 $46.20 Credit : Accumulated Interest $ 2.05 Balance as at 31st December 1976 $48.25 64 REPORT OF COUNCIL CITATIONS EDGEWORTH DAVID MEDAL The Edgeworth David Medal for 1976 is awarded to Associate-Professor Ross Howard Street for distinguished contributions to mathematical research. This award is restricted to scientists under the age of 35. Associate-Professor Street is a graduate of the University of Sydney and is now on the staff of Macquarie University where he rose from lecturer to associate-professor in 5 years. He has made major contributions to category theory, and has published an elegant analysis of the formal theory of monads. Already his published work marks him as a young scientist of exceptional understanding, in difficult fields where others have been indefinite, he has been decisive. Participation in overseas conferences has given prominence to his work, and led Professor S. Eilenberg of Columbia University, New York, to say of him: "This is the coming generation". Associate-Professor Street's work is of outstanding quality and depth, and well worthy of this award. THE CURKE MEDAL The Clarke Medal for 1976 for distinguished work in the natural sciences is awarded to Dr. Lilian Ross Fraser. As a Linnean Macleay Fellow, Dr. Fraser initiated her studies on the life history, morphology and physiology of "sooty moulds”, which culminated in the award of a D.Sc. from the University of Sydney. She joined the Department of Agriculture in the early 1940' s and continued research on fungi and fungal diseases of many native and crop plants. As a result of her work on the fungi causing root-rot in citrus and of her development of disease-resistant stocks, the citrus industry in the irrigation areas of the Murray and Murrimbidgee was saved from near-disaster about 1950. Later work dealt with the virology of citrus and diseases affecting grapes, vegetables and ornamentals She is held in high regard by her fellow scientists and by the man on the land, surely an important accomplishment. Internationally she is known for her research on the virology and pathology of citrus. At the invitation of the Government of India, she reported on problems of citrus-growing and disease control . Dr. Fraser is a past-President of the Linnean Society and has been active in the profession of botany. She retired recently from the Biological and Chemical Research Institute of the New South Wales Department of Agriculture, where she had been Chief Biologist. Dr. Fraser rightly joins the eminent group of Clarke medallists. THE SOCIETY'S MEDAL The Society's Medal for 1976 is awarded to Mr. Edric Keith Chaffer for service to science and to the Society. Mr. Chaffer's interest in the Royal Society of New South Wales started at school where he perused the Journal in the school library. As a schoolboy he attended the 1949 Clarke Memorial Lecture by Dr. F.W. Whitehouse, and was elected to membership in 1954. Since then he has shown a constant interest in the Society and continues to work for it. He is a past-President and has served as Honorary Secretary, Council member, chairman of the Section of Geology, director of Science House Pty. Ltd. and on library and publication committees. His extensive knowledge of the history of the Royal Society is constantly of value to members of the Council. As a director of W. Chaffer and Sons Pty. Ltd. , a tanning and leather-finishing company founded in 1887, Mr. Chaffer keeps a wide interest in leather and in the leather trade. For several years he was senior lecturer (part time) in the School of Tanning at the Sydney Technical College, and is active in the Leather Industry Research Association and other associations connected with leather. He is a member of the American Leather Chemists' Association, the Society of Leather Technologists and Chemists, and of the Verein fUr Gerberei-Chemie und-Technik. Clearly he is a man of standing in his profession. Following the study of geology at school, Mr. Chaffer did a part-time course in geology at the Sydney Technical College. He is a member of the Geological Society of Australia. Edric Chaffer is a dedicated member of the Royal Society of New South Wales. All aspects of our work REPORT OF COUNCIL 65 interest him and he contributes freely with his knowledge and participation. His service to our Society and to the profession of leather science makes him a worthy recipient of the Society's Medal. ARCHIBALD D. OLLE PRIZE The Archibald D. Olle Prize is awarded to Dr. Lawrence A. Drake for his paper entitled "Seismic Risk in Australia", which was judged the best paper submitted during 1976 and published in the Journal of the Society. Dr. Drake graduated B.A. with Honours in mathematics and B.Sc. in physics from the University of Melbourne. He then studied at the University of California, Berkley campus, where he graduated M.A. and Ph.D. At present Dr. Drake is Director of the Riverview Observatory and Senior Lecturer in the School of Earth Sciences at Macquarie University. His research interests are mainly seismology, the finite element method and surface wave responses of structures. SUMMER SCHOOL IN MEDICINE In January 1977, 66 selected students from Year 11 in Sydney schools attended a summer school in medicine arranged for the Royal Society of New South Wales by Dr. Lowenthal and Mrs. Krysko v. Tryst. We asked the students to submit essays based on what they had learnt during the week. These essays were sent to Dr. Lowenthal in Paris and he has judged them. It is my pleasure to announce the names of those who sent in the best 3 essays and to present each student with a copy of our Centenary Volume. First Equal Second Mr. Steven Harvey Miss Kathryn Byron and Miss Estelita Ratanayagam 66 Financial Statements for 1976 AUDITOR’S REPORT TO THE MEMBERS In our opinion: (a) the attached balance sheet and income and expenditure account are properly drawn up in accordance with the Rules of the Society and so as to give a true and fair view of the state of affairs of the Society at 31st December 1976 and of the results of the Society for the year ended on at date; and (b) the accounting records and other records, and the registers required by the Rules to be kept by the Society have been properly kept in accordance with the provision of those Rules. DOUGLAS R. WYLIE, PUTTOCK, KIELY AND CO. Chartered Accountants. By ALAN M. PUTTOCK Registered under the Public Accountants Registration Act, 1945 as amended. BALANCE SHEET to 31/12/76 RESERVES 9,040 Library Reserve(note 2(i)) 7,200 424,564 Resumption Reserve (note 2(ii)) 416,991 92,797 LIBRARY FUND (note 2(iii)) 93,822 806 LONG SERVICE LEAVE FUND 806 10,652 TRUST FUNDS (note 4) 11,374 37,822 ACCUMULATED FUNDS 35,742 $575,681 TOTAL RESERVES 8 FUNDS $565,935 CURRENT ASSETS 15 Petty Cash Imprest 153 Debtors for Subscriptions 1,075 153 Less Provision for Doubtful Debts 1,075 1,409 Other Debtors 8 Prepayments 12,534 Interest Bearing Deposit 10,108 Cash at Bank 187 1,865 7,426 10,420 $ 24,066 19,898 Less: CURRENT LIABILITIES 8,539 4 127 1,966 Sundry Creditors 8 Accruals Life Members Subscriptions - Current Portion Membership Subscriptions Paid in Advance Subscription to Journal Paid in Advance 13,684 4 40 1,362 $ 10,636 15,089 $ 13,430 2,122 2 13,600 14 NET CURRENT ASSETS Add: FIXED ASSETS Furniture and Office Equipment - at cost less Depreciation Lantern - at cost less Depreciation Library - 1936 Valuation Pictures - at cost less Depreciation 4,809 7,888 2 13,600 13 $ 15,738 $ 21,503 FINANCIAL STATEMENTS 67 $ 29,168 $ 26,312 Add: INVESTMENTS 16,680 Commonwealth Bonds 6 Inscribed Stock 27,180 20,000 Loans on Mortgage 400 Debenture Stock $37,080 27,180 1 509,490 Add: ASSOCIATED CORPORATIONS (note 3) Shares - at cost Advances and Loans - Unsecured 1 512,495 $509,491 $512,496 $575,739 $565,988 Less: NON-CURRENT LIABILITIES Life Members Subscriptions - 58 Non-Current Portion 53 $575,681 NET ASSETS $565,935 D.J. Swaine, President A. A. Day, Honorary Treasurer STATEMENT OF ACCUMULATED FUNDS For the Year Ended 31 December 1976 13.583 DEFICIT for year 11,771 87.583 Donations and Interest to Library Fund 1,303 3,767 Science House Partnership 1,438 Transfer from Library Reserve 1,840 Transfer from Library Fund 278 29,407 Transfer from Resumption Reserve 7,573 16,793 Accumulated Funds -Beginning of Year 37,822 $125,405 AVAILABLE FOR APPROPRIATION 37,045 87,583 Transfer to Library Fund 1,303 $ 37,822 ACCUMULATED FUNDS - Current Year $ 35,742 NOTES TO AND FORMING PART OF THE ACCOUNTS for the year ended 31st December, 1976. 1. SUMMARY OF SIGNIFICANT ACCOUNTING POLICIES Set out hereunder are the significant accounting policies adopted by the Society in the preparation of its accounts for the year ended 31st December, 1976. Unless otherwise stated, such accounting policies were also adopted in the preceding year. (a) Accounting Period In order to rationalise the time between the close of the Society's financial year and the holding 68 FINANCIAL STATEMENTS of the Annual General Meeting the Society's accounting period has been changed from 1st March - 28th February to 1st January - 31st December. As a result of the above change in policy the figures for the current period represent 12 months operations while the comparative figures shown represent 10 months operations. Cb) Depreciation Depreciation is calculated on a written down value basis so as to allow for anticipated repair costs in later years. The principal annual rates in use are: Furniture 7.5% Office equipment 15.0% (c) Deficit from Operations The balance of the Resumption Reserve, after lending $416,990 to Science House Pty. Ltd. , has been allocated to meet operating deficits. (see also notes 2 (ii) and 3). 2. MOVEMENTS IN PROVISIONS AND RESERVES (i) Library Reserve Balance at 1st January 1976 Less Transferred to accumulated funds re: Library recataloguing Typing and printing subject index Less donations towards printing sub j ect index 1975 1975 1976 1976 $ $ $ $ 10478 9040 1292 1840 681 - 1973 1840 535 1438 1840 Balance at 31st December 1976 $9040 $7200 (ii) Resumption Reserve Balance at 1st January 1976 Less Transferred to accumulated funds re: Operating deficit prior years Operating deficit current year 453971 424564 15994 . 13413 7573 29407 7573 Balance at 31st December 1976 $424564 $416991 Represented by: Shares in associated coporation Loans to associated corporation Short term deposits 1 416990 7573 1 416990 $424564 $416991 FINANCIAL STATEMENTS 69 ; 2. MOVEMENTS IN PROVISIONS AND RESERVES Continued 1 (iii) Library Fund 1975 1976 1 $ $ i Balance at 1st January 1976 5214 92797 Add donations and bank interest 87583 1303 92797 94100 Less Library purchases 278 i Balance at 31st December 1976 $92797 $93822 Represented by: Cash at bank 297 22 1 Loans to associated corporation 92500 92500 1 Commonwealth Bonds “ 1300 $92797 $93822 3. ASSOCIATED CORPORATIONS The Society has entered into a joint venture with the Linnean Society for the establishment of a Science Centre for New South Wales and to facilitate this. a company. Science House Pty. Limited has been formed in which each Society has 50% interest. Advances and loans to the company have been on an interest free basis repayable at call. No material repayments are anticipated prior to 31st December, 1977. 1975 1976 $ $ Total amount advanced 509490 512495 Representing: Resumption reserve 416990 416990 Library fund 92500 92500 Accumulated funds ■ 3005 $509490 $512495 4. TRUST FUNDS 1975 Clarke Walter Memorial Burfitt Prize $ $ $ Capital 7000 3600 2000 Revenue income for period “616 337 187 Less expenditure 360 “ “ 256 337 187 Balance from 1975 3396 954 912 Total Revenue $3652 $1291 $1^ Total Trust Funds $10652 $4891 $3099 70 FINANCIAL STATEMENTS 4. TRUST FUNDS Continued Liversidge Bequest Olle Bequest Total $ $ $ Capital 1400 - 7000 — — — Revenue Income for Period 131 117 772 Less Expenditure 50 - 50 81 117 722 Balance from 1975 548 1238 3652 Total Revenue $629 $1355 $4374 Total Trust Funds $2029 $1355 $11374 FUNDS STATEMENT FOR THE YEAR ENDED 31ST DECEMBER 1976 1975 1975 1976 1976 $ $ $ $ SOURCE OF FUNDS Donations and interest to library fund 87583 1303 Withdrawal of investments - 9900 Trust fund income 616 772 Reduction in working funds 9624 7607 Refund of capital Science House partnership 3767 “ $101590 $ 19582 APPLICATION OF FUNDS Operating deficit for the year Less : Items not involving the outlay of funds in the current period: Depreciation of fixed assets Provision for doubtful debts Funds applied to operations Loan to associated company Purchase of furniture and equipment Reclassification of life members subscriptions in advance Increase in investments Trust fund expenses 13583 11771 187 252 170 1014 13226 10505 87500 3005 100 6017 4 5 400 - 360 50 $101590 $ 19582 FINANCIAL STATEMENTS 71 INCOME AND EXPENDITURE ACCOUNT For the Year Ended 31 December 1976 INCOME 6,051 Membership Subscriptions - Ordinary 7,298 6 Membership Subscriptions - Life Members 5 158 Application Fees 27 6,215 7,330 1,950 Subscriptions to Journal 2,923 2,000 Government Subsidy 705 Donations - Printing Journal S Publications 117 10,870 Total Membership 5 Journal Income 10,370 4,136 Interest Received 4,152 1,924 Sale of Reprints 567 476 Sale of Back Numbers 60 123 Sale of Other Publications 343 4 Donations - General Annual Social Surplus 27 Summer School Surplus 739 17,533 16,258 Less : EXPENSES 800 Accountancy Fees 920 - Advertising 48 244 Annual Social - 110 Audit Fees 110 142 Cleaning 136 187 Depreciation 252 119 Electric Light 5 Power 131 96 Entertainment Expenses Journal 6 Publication Costs 73 9,073 Printing - Current Year Volume 4,645 681 Printing - Other Publications 372 175 Binding - 747 Wrapping 5 Postage 887 _ Legal Costs 5,904 250 409 Library Purchases 599 1,292 Library Recataloguing 1,840 - Library Relocation 1,967 463 Miscellaneous Expenses 63 715 Postage 973 822 Printing and Stationery - General 417 170 Provision for Doubtful Debts 1,014 4,006 Rent 4,767 328 Repairs and Maintenance 297 9,529 Salaries 6,946 - Secretarial Services 745 557 Superannuation Contributions - Employees 216 451 Telephone 361 31,116 13,583 DEFICIT for the year 28,029 11,771 AUSTRALASIAN MEDICAL PUBLISHING CO. LTD. 7t-79 ARUNDEL ST. . GLEBE. N.S.W. , 2037 1977 THE ROYAL SOCIETY OF NEW SOUTH WALES The Society originated in the year 1821 as the Philosophical Society of Australasia. Its main function is the promotion of Science through the following activities : Publication of results of scientific investigation through its Journal and Proceedings ; the Library ; awards of Prizes and Medals ; liaison with other Scientific Societies ; Monthly Meetings ; and Summer Schools for Senior Secondary School Students. Special Meetings are held for the Pollock Memorial Lecture in Physics and Mathematics, the Liversidge Research Lecture in Chemistry, and the Clarke Memorial Lecture in Geology. Membership is open to any interested person whose application is acceptable to the Society. The application must be supported by two members of the Society, to one of whom the applicant must be personally known. Membership categories are : Ordinary Members : $18.00 per annum plus $3 application fee. Absentee Members ; |15.00 per annum plus $3 application fee. Associate Members (spouses of members and persons under 25 years of age) : $5.00 per annum plus $2 00 application fee. Associate Members (with Journal) ; $12.00 per annum plus $2 application fee. Subscription to the Journal, which is published in four Parts per year, issued twice yearly in May and November, for non-members is $22 p.a. plus postage. For application forms for membership and enquiries re subscriptions, write to : The Royal Society of New South Wales, Science Centre, 35 Clarence Street, Sydney, 2000, N.S.W. The Society welcomes manuscripts of research (and occasional review articles) in all branches of science, art, literature and philosophy, for publication in the Journal and Proceedings. Manuscripts will be accepted from both members and non-members, though those from the latter should be communicated through a member. A copy of the Guide to Authors is obtainable on request and manuscripts may be addressed to the Honorary Secretary (Editorial) at the above address. Contents Geography : Catastrophic Channel Changes in the Macdonald Valley, New South Wales, 1949-1955. H. M. Henry . . . . . . . . . . . . . . 1 Geology : Does the Hunter River Supply Sand to the New South Wales Coast Today ? P. S. Roy . . . . . . . . . . . . . . . . . . 17 I Geophysics : A Late Devonian Palaeomagnetic Pole for the Mulga Downs Group, Western New South Wales. B. J. J. Embleton . . . . . . . . . . 25 Mathematics : Series of Roots of a Transcendental Equation. Pietro Cerone and Austin Keane . . . . . . . . . . . . . . . . . . . . 29 Palaeontology : Floral Evidence for a Middle Triassic Age of the Gunnee Beds and Gragin Conglomerate, near Delungra, New South Wales. D. J. Botirke, R. E. Gould, R. Helby, R. Morgan and G. J. Retallack . . . . . . . . 33 The Tertiary Stratigraphic Palynology of the Murray Basin in New South Wales. 1. The Hay-Balranald-Wakool Districts. Helene A. Martin .. 41 Presidential Address 1976 : Leather — Why is it so ? Edric Chaffer . . . . . . . . . . . . 49 Report of Council, 31st March, 1977 . . 61 AUSTRALASIAN MEDICAL PUBLISHING CO. LTD.. 71 -79 ARUNDEL ST. . GLEBE. N.S.W. , 2037 1977 Journal and Proceedinas of the New (SouthNi^les VOLUME 110 1977 PARTS 3 and 4 Published by the Society Science Centre, 35 Clarence Street, Sydney Issued 23rd Novenaber, 1977 NOTICE TO AUTHORS A "Style Guide to Authors" is available from the Honorary Secretary, Royal Society of New South Wales, 157 Gloucester Street, Sydney, N.S.W., 2000, and intending authors must read the guide before preparing their manuscript for review. The more important requirements are summarized below. GENERAL Manuscripts should be addressed to the Honorary Secretary (address given above). Manuscripts submitted by a non-member must be communicated by a member of the Society. Each manuscript will be scrutinised by the Publications Committee before being sent to an independent referee who will advise the Council of the Society on the acceptability of the paper. In the event of rejection, manuscripts may be sent to two other referees. Papers, other than those specially invited by Council, will only be considered if the content is substantially new material which has not been published previously, has not been submitted con- currently elsewhere, nor is likely to be published substantially in the same form elsewhere. Well- known work and experimental procedure should be referred to only briefly, and extensive reviews and historical surveys should, as a rule, be avoided. Letters to the Editor and short notes may also be submitted for publication. Original papers or illustrations published in the Journal and Proceedings of the Society may be reproduced only with the permission of the author and of the Council of the Society; the usual acknowledgments must be made. Offset printing with "Typeset-it-Yourself" preparation of a master manuscript suitable for photography is used in the production of the Journal. Authors will be supplied with a set of special format paper. An IBM Selectric (Golf Ball) typewriter with ADJUTANT 12 typeface must be used. Biological and reference material are shown in Light Italia. Symbol 12 has most type required for mathematical expressions and formulae. Detailed instructions for the typist are included in the Style Guide. PRESENTATION OF INITIAL MANUSCRIPT FOR REVIEW Typescripts should be submitted on heavy bond A4 paper. A second copy of both text and illustra- tions is required for office use. This may be a clear carbon or photographic copy.' Manuscripts,' including the abstract, captions for illustrations and tables, acknowledgments and references should be typed in double spacing on one side of the paper only. Manuscripts should be arranged in the following order: title; name(s) of author(s); abstract; in- troduction; main text; conclusions and/or summary; acknowledgments; references; appendices; name of Institution/Organisation where work carried out/or private address as applicable; date manuscript re- ceived by the Society. A table of contents should also accompany the paper for the guidance of the Editor. . • Spelling follows "The Concise Oxford Dictionary". The Systeme International d'Unites (SI) is to be used, with the abbreviations and symbols set out in Australian Standard ASIOOO. All stratigraphic names must conform with the Australian Code of Stratigraphic Nomenclature (re- vised fourth edition) and must first be cleared with the Central Register of Australian Stratigraphic Names, Bureau of Mineral Resources, Geology and Geophysics, Canberra. The letter of approval should be submitted with the manuscript. Abstract. A brief but fully informative abstract must be provided. .Tables should be adjusted for size fit the format paper of the final publication. Units of measurement should always be indicated in the headings of the columns or rows to which they apply. Tables should be numbered (serially) with Arabic numerals and must have a caption. Illustrations. When submitting a paper for re- view all illustrations should be in the form and size intended for insertion in the master manuscript If this is not readily possible then an indication of the required reduction (such as reduce to H size) must be deary stated. Note'. There is a reduction of 30% from the master manuscript to the printed page in the journal. Maps, diagrams and graphs should generally not be larger than a single page. However, large figures can be printed across two opposite pages. it Drawings should be made in black Indian ink on white drawing paper, tracing cloth or light-blue lined graph paper. All lines and hatching or stip- pling should be even and sufficiently thick to allow appropriate reduction without loss of detail. The scale of maps or diagrams must be given in bar form. Half-tone illustrations (photographs) should be included only when essential and should be present- ed on glossy paper (no negative is required). Diagrams, graphs, maps and photographs must be numbered consecutively with Arabic numerals in a single sequence and each must have a caption. References are to be cited in the text by giving the author's name and year of publication. Ref- erences in the reference list should fol low the pre- ferred method of quoting references to books, per- iodicals, reports and theses , etc ., and be listed alphabetically by author and then chronologically by date. Abbreviations of titles of periodicals shall be in accordance with the International Standard Organ- ization IS04 "International Code for the Abbreviation of Titles of Periodicals" and International Standard Organization IS0833 "International List of Period- ical Title Word Abbreviations" and as amended. Appendices should be placed at the end of the paper, be numbered in Arabic numerals, have a caption and be referred to in the text. Reprints. An author who is a member of the Society will receive a number of reprints of his paper free, An .author who is not a member of the Society may purcKase reprints. UMARY y^J;(Report of Council, 31st March, 1977— continued from Vol. 110, Parts 1 and 2) FEBl m ) { Abstract of Proceedings, Year ■■ i LOCATION .f Large Hall, Science House, 157 Gloucester I Street, Sydney. ij APRIL 7TH t t j 109th Annual General Meeting. The President i Mr. E.K. Chaffer was in the chair and 40 members ■. I and visitors were present. 4 new members were j elected and Council announced the admittance of !5 new associate members. Council also announced the election, by Council, to Honorary Membership I j of Professor Samuel W. Carey. Council advised of I j changes in subscriptions, the new rates being j Ordinary Membership $18.00, Absentee Membership ; ■< $15.00, Associate Membership $5.00, Associate 1 -. Membership (with Journal) $12.00, subscriptions to f- -.'i Journal (non-members and institutions) $22.00 plus postage. I \ The Clarke Medal was awarded to Dr. J.N. i [Jennings; the Edgeworth David Medal to Dr. F.J. ( »■ Ballard; the James Cook Medal to Dr. A. Walsh and j Messrs. D.R. Wylie, Puttock and Kiely were I elected Auditors. ! ^ The Presidential address "Leather - Why is it I so?" was given by Mr. E.K. Chaffer. {The incoming President, Dr. D.J. Swaine was installed and introduced to members. I MAY 3RD 1 891st General Monthly Meeting. The President •fDr. D. J. Swaine was in the chair and 68 members and visitors were present. 6 new members were \ ') elected and Council announced the admittance of 1 1 [new associate member. Ending 31st December,,ljn^^^|go WMVBMMTV Dr. D.J. Swaine was in the chair and 53 members and visitors were present. An address "Some Spacecraft I have known" was given by Dr. K.G. McCracken, Chief, Division of Mineral Physics, C.S.I.R.O. LOCATION Auditorium, Science Centre, 35 Clarence Street, Sydney. SEPTEMBER 1ST 895th General Monthly Meeting. The President Dr. D.J. Swaine was in the chair and 30 members and visitors were present. 1 new member was elected and Council announced the admittance of 1 new associate member. The President made a few remarks appertaining to the successful completion of the Science Centre and welcomed members and visitors to this, the first meeting in the new home of the Society. An address "Light and Colour" was given by Dr. W.R. Blevin, Chief Research Scientist, National Measurement Laboratory, C.S.I.R.O. OCTOBER 6TH 896th General Monthly Meeting. Mr. E.K. Chaffer, Vice-President, was in the chair and 43 members and visitors were present. 1 new member was elected. An address "Some Possible Staging Places in Aboriginal Migration from South East Asia" was given by Emeritus Professor N.W.G. Macintosh, Department of Surgery, University of Sydney. NOVEMBER 3RD iAn address "Island Arcs and Ore Deposits" was given by Professor R.L. Stanton, Department I of Geology, University of New England. ' -i I ^ JUNE 2ND i [ 892nd General Monthly Meeting. The President j ;Dr. D.J. Swaine was in the chair and 45 members i .’and visitors were present. Council announced the I f admittance of 1 new associate member. ' S An address "Human Nutrition - Science or Trans-Science" was given by Mr. M.V. Tracey, I .Chief, Division of Food Research, C.S.I.R.O. - [I JULY 7TH I - ; j 893rd General Monthly Meeting. The President ' ' ' Dr. D.J. Swaine was in the chair and 47 members and visitors were present. Council announced the * t admittance of 1 new associate member. i ■ c An address "Science and Art: Some Issues" I r|was given by Mr. Elwyn Lynn, Curator, Power ) il Gallery of Contemporary Art, University of Sydney. I! AUGUST 4TH 1 894th General Monthly Meeting. The President 897th General Monthly Meeting. The President Dr. D.J. Swaine was in the chair and 26 members and visitors were present. Mr. W.H. Robertson presented a few comments illustrated by photographs, on the total eclipse of the October 23. An address "Aboriginal Health and its Relation- ship to Vitamin C" was given by Dr. Archie Kalokerinos . DECEMBER 1ST 898th General Monthly Meeting. The President Dr. D.J. Swaine was in the chair and 41 members and visitors were present. Council announced that it had elected to Honorary Membership Sir Robert Price, Chairman of C.S.I.R.O., and Emeritus Professor P.A. Elkin. A symposium was held with the theme "People, Planning and the Environment". The panel of speakers comprised Dr. A.J. Sutton, Director, Bureau of Crime Statistics, N.S.W. Department of Attorney-General and Justice; Dr. L.L. Pownall, Chairman, N.S.W. Planning and Environment Commiss- ion; Mr. R.P. Murphy, Assistant Director, Pollution Control, N.S.W. State Pollution Control Commission. 74 ABSTRACT OF PROCEEDINGS LOCATION The Macleay Museum, University of Sydney. JUNE 16TH Members and gusests attended a private viewing of the exhibition "The Moving Frontier; Aspects of Aboriginal European Interaction" and of an exhibition on an aboriginal theme " A History of Australia". The Curator of the Museum, Dr. Peter Stanbury, gave a talk on the exhibitions. LOCATION Lecture Theatre No. 4, School of Chemistry, University of Sydney. JULY 15TH The Livers idge Lecture for 1976 was given by Professor R.L. Martin, Professor of Inorganic Chemistry, Research School of Chemistry, Institute of Advanced Studies, Australian National University, the title of the address being "Co-ordination, Topology and Structure in Metal Oxides". 75 Obituaries Ida Alison Browne Ida Alison Browne died in Sydney on October 21st, 1976 after a long illness. She was bom at 175 Sutherland St., Paddington on August 16, 1900 to the wife of William G. Brown. Drs. W.R. and Ida Browne After completing her secondary education at Fort Street Girls' High School she entered Sydney University in the faculty of Science in 1918. Her mind must have already been fully dir- ected towards a career in Geology, having topped the State in the 1917 Leaving Certificate exam in this subject and with supporting honours in Mathematics and Botany. Her University academic record bore out this intention with a series of High Distinction and Prizes, and in March, 1922 she graduated with First Class Honours, the Uni- versity Medal for Geology -Mineralogy and the Deas Thomson Scholarship, gaining the awards over the other First Class honours winner, H.G. Raggatt. In earlier years she had equalled or beaten such worthy candidates as G.D. Osborne and T.L. Willan. Her interests at this time were clearly in the petrological-mineralogical field with field mapping of only slightly less interest, while she appears not to have taken the Palaeontological sections available in Geology III. She was awar- ded a Science Research Scholarship on graduation but resigned in April, 1922 to become a demon- strator in the department replacing another woman graduate, Dorothy (Dip) Powell. She retained this position till early in 1927 when she accept- ed a Linnean MacLeay Fellowship in Geology. While a demonstrator she had published a paper on minerals at Broken Hill and three papers on the South Coast geology which was claiming her atten- tion. The years of her MacLeay Fellowship enab- led her to give full attention to this little known area of the state which was traversed only by rough bush tracks masquerading as highways . During this period she published seven papers, several quite lengthy, on various aspects of South Coast Geology. The regional stratigraphy and the igneous bodies such as Mt. Dromedary and Milton were particularly dealt with. Her thesis 'The Geology of the South Coast of New South Wales with special reference to the origin and relationships of the igneous rocks' gained her the award of D.Sc. in 1932. At that time she was only the second woman to achieve such a distinction at the University. In October, 1934 W.S. Dun, who had lectured part-time in Palaeontology at Sydney University for many years, died. Ida Brown had become a Demonstrator again in March of that year and al- though inexperienced in this branch of geology, she took over the task of mastering this new field and teaching in it. In March, 1935 she was appointed Assistant Lecturer in Palaeontology, to become successively Lecturer (1940) and Senior Lecturer (1945) resigning in August, 1950. It is characteristic of Ida Brown that she published only two items between 1934 and 1940, one a note (based on her earlier work) in an Agri- cultural paper, the other with Germaine Joplin on the fossiliferous Upper Devonian rocks at Mt. Lambie. Her time was devoted to studying palaeon- tology and not till she felt she had this subject firmly understood did she publish any research in this field. She had found an area of great inter- est, rich in mapping problems and fossils, in the Yass region, and this was to hold her attention for the next thirty years . In 1940 her first palaeontological paper appeared (a short note in fact) , and eleven more on various fauna appeared in due course. The palaeontological papers are interspersed between papers dealing with stratigraphy and structure, mainly of the Yass region. These stratigraphic papers complete with large, detailed maps epito- mize her love for field work and will, I feel, remain her most enduring geological memorial. In 1949 she married William Rowan Browne, long-time Reader in the Department of Geology at Sydney University. Following Browne's retirement in 1949 they enjoyed more than twenty years together in the field she acting as chauffeur and general guardian to the still active older geologist, specially on their long summer trips to the Kosciusko region. In this period W.R. Browne produced more than forty research papers and articles, Ida Browne ten including some of her major works. The roles of assistant and guardian were reversed dur- ing the last years when Ida Browne was afflicted by a slow, paralysing illness which kept her vir- tually bed ridden. Despite his age W.R. Browne maintained much of his vigour and this he devoted to caring for Ida and easing her pain. His sudden 76 OBITUARIES death late in 1975 meant a lonely last year for Dr. Ida. Almost all her papers were published by either the Royal Society or the Linnean Society of New South Wales, and to both of these she gave her active interest. She was President of the Royal Society in 1953-54 and served for many years on its Council. In 1945 she presided over the Linnean Society and was also a long-term Councillor of this body. The libraries of the Australian Museum and Wollongong University both benefited from her generous donation of books. One of her last acts was to approve the establishment of a W.R. Browne medal by the Geological Society of Australia. Ida Browne published few papers in association with other workers Cone with W.R. Browne, and three with other women geologists) . Only two papers were published overseas, one being a joint publi- cation. These two facts again are characteristic of her. She was happy to remain within the ambits of local science. Although aware of the litera- ture and the world beyond she was content to apply her knowledge to local problems. Furthermore, she was content to solve them essentially unaided. She belongs to an interesting period of the development of Australian Science during which a number of dedicated women quietly but firmly es- tablished themselves in fields which had been largely dominated by men. For this there is no doubt they owe some thanks to Edgeworth David and his wife but most of the effort was their own. Al- ways a dignified and stately lady. Dr. Ida will be remembered for her precise lectures and her unfail- ing courtesy. Her accurate work in the field will continue to be quoted long into the future. That is the memorial she would have appreciated. David Branagan Keith Edward Bullen On 23rd September, 1976 a most distinguished member of the Royal Society of N.S.W., Professor Keith Bullen, died in Auckland, New Zealand .while visiting relatives. Keith Edward Bullen was bom in Auckland on 29 June, 1906. After gaining a B.Sc. degree in physics and an M.A. (1st Class Honours) degree in mathematics, in 1927 he was appointed a Lecturer in mathematics at Auckland University College. From 1931 to 1933 he attended St. John's College, Cambridge, England, winning the Strathcona Exhibition. He worked with Sir Harold Jeffreys on the monumental task of constructing tables of travel times of seismic waves within the earth. The tables of the time were accurate to about one minute. The Jeffreys-Bullen tables are accurate to one or two seconds and are still, 30 years later, used for the location of earthquakes by the International Seismological Centre and the U.S. Geological Survey. The particular problem that Sir Harold Jeffreys set Keith Bullen was to work out the effect of the earth's flattening on the travel times of seismic waves within it. The re- sults of this work are clearly and logically set out in the Geophysical Supplement of the Monthly Notices of the Royal Astronomical Society of May, 1937. These so-called ellipticity corrections are not much used nowadays because, as they are only of about half a second, they are smaller than the variations of travel times of seismic waves caused by the differences of continental and oceanic crust near the earth's surface (two seconds). However, to estimate the ellipticity corrections, Keith Bullen had to study surfaces of equal den- sity within the earth. Knowledge of density within the earth at that time was extremely vague. Keith Bullen integrated the Adams -Williamson equation from the bottom of the earth's crust to the earth's centre, and, after some disappoint- s ments, obtained results consistent with the earth's moment of inertia. Keith Bullen was a lecturer at Hull Univer- sity College in England in 1933. After this he returned to Auckland University College. On 15 May, 1935 he married Florence M. Pressley. In the later 1930 's he produced a number of papers on New Zealand earthquakes, travel times of seismic waves, density within the earth, and on the crustal structure of New Zealand and the Pacific Ocean. I have reprints of 20 of these papers by me now as I write . From 1940 to 1945 Keith Bullen was Senior Lecturer in Mathematics at the University of Melbourne. From 1946 tor 25 years until his re- tirement he was Professor of Applied Mathematics at the University of Sydney. In 1949 he received the Lyall Medal. In 1952 he received the Hector Medal in New Zealand. In 1953 he received the Walter Burfitt Prize of the Royal Society of N.S.W. Keith Bullen has left us three clear, logical and care- fully written books on mechanics, seismology and the earth's density, as well as two Jess important books and a set of printed notes used at the International Institute of Seismology and Earth- quake Engineering in Tokyo. When I look back at a deceased person's life I find myself concerned, not only about his discoveries, his awards and his publications, but also about his honesty, his dedi- cation and the care with which he prepared lectures for the students who studied the courses he taught. Keith Bullen was well known at the University of Sydney for his honest, logical and clear lectures. Since he practised what he preached, a sentence from his Presidential Address to Section A of the Australian and New Zealand Association for the Advancement of Science in 1951 should go into his obituary: "In a world in which so little is done with meticulous care to detail, in which so rela- tively few are prepared to work with concentration OBITUARIES 77 over long hours, or to look at problems with an honest impartiality, the magnificent work of the natural scientist stands out." While at the University of Sydney, Keith Sullen was associated with three important and interesting research results. First, in 1946, he predicted from the sharp increase in the velocity of P waves at the earth's inner core boundary, that the inner core was solid. Second, the late Fr Burke-Gaffney, here at Riverview Observatory, colleceed seismic readings of four hydrogen bombs detonated in 1954. Keith Bullen estimated the times of the explosions to within 0.7, 0.4, 0.1 and 0.0 seconds respectively. The U.S. Atomic Energy Commission realized that that it was point- less to try to keep the times of large underground explosions secret, and has since released the times and locations of them all. These times and loca- tions have been of great assistance to studies of the interior of the earth. Third, in 1952 H. Benioff in southern California observed what he thought were oscillations of the whole earth, with a period of approximately 54 minutes. Benioff requested C.L. Pekeris, a mathematician from Israel, to calculate the periods of oscillation of the whole earth. To do this Pekeris had to use models of the earth which incorporated values of the density and incompressibility found by Keith Bullen. Keith Bullen was delighted when these models were confirmed by observations of spherical oscillations from the Chilean earthquakes in 1960. Keith Bullen received the Bicentennial Medal of Columbia University in 1954. He was President of the International Association of Seismology and the Physics of the Earth's Interior from 1955 to 1957. In 1957 he wss the Australian delegate at a most interesting conference in Washington on rockets and artificial satellites. In the course of the conference Sp-tnik 1 moved overhead. In 1959 he was Vice-President of the Scientific Committee for Antarctic Research of the International Council of Scientific Unions (SCAR) . He wrote : "Ostensibly all the problems of SCAR were purely scientific ones, but for a long time there was a steady political undercurrent from the exist- ence of several groups among the 12 nations repre- sented on SCAR. The first group included countries like the United Kingdom, Australia, France and Norway which had mdde substantial claims of Antarctic territory. The second group included Argentina and Chile which also made claims, but claims incompatible with those of Britain. The United States and the U.S.S.R. constituted two further groups which, while not wholly in accord with each other, had this in common that they declined to recognize the claims of any other country, and in fact asserted the rights of both to carry out scientific investigations wherever they chose in Antarctica. "In the early days of SCAR, one could never be quite sure whether an apparently innocent scientific proposal contained a significant politi- cal implication, and once again I learned how far scientists had moved from the old ivory-tower days. It appears that the labours of the SCAR scientists were not, however, in vain, and that they made an impact on governments, leading to the International Antarctic Treaty of 1959, which most people now seem to think is quite a good thing." Keith Bullen wrote these paragraphs in 1964. In 1961 he received the William Bowie Medal of the American Geophysical Union. In 1963 he received the Day Gold Medal of the Geological Society of America. In 1965 he received the Research Medal of the Royal Society of Victoria. From 1963 to 1967 he was Vice-President of the International Union of Geodesy and Geophysics. He was elected to the Pontifical Academy of Science in 1968. He received the Gold Medal of the Royal Astronomical Society and became an Honorary Fellow of the Royal Society of New South Wales in 1974. He was a Fellow of the Royal Society, a Fellow of the Australian Academy of Science, a Foreign Associate of the U.S. National Academy of Sciences, a Fellow of the American Geophysical Union, a Foreign Honorary Member of the American Academy of Arts and Science, a Foreign and Commonwealth Member of the Geological Society of London, an Honorary Fellow of the Royal Society of New Zealand, and he received the Sc.D. degree from the University of Cambridge. It only remains for me to say that Keith Bullen was respected and liked by all who worked with him, and to congratulate his widow, son and daughter on belonging to the family of so fine a man. Lawrence A. Drake 78 OBITUARIES Ernest Ritchie liri The sudden death of Ernest Ritchie, Professor of Organic Chemistry in the University of Sydney, at his home on 9th April, 1976, has deeply saddened colleagues and friends throughout Australia and beyond. He was a man of great hon- esty, sincerity and kindliness who was universally admired and respected. There is now everywhere a profound sense of shock that he is no longer with us, which we share with his wife, Maisie, and family, Susan, Robert and Ian. Em Ritchie was bom on 11th February, 1917, and grew up in the Eastern Suburbs district of Sydney. He attended Woollahra Superior Public School, Randwick Boys' Intermediate High School, and finally Sydney Boys' High School. With the help of a Public Exhibition Ritchie entered the Faculty of Science, Sydney University, in 1933 and as a result of a strong impression made by the then Senior Lecturer, Dr. Francis Lions, he embarked on his career in organic chemistry, graduating B.Sc. (HI) in 1937 and M.Sc. in 1939. Ritchie remained at the University, becoming Assistant Lecturer (1941), Senior Lecturer (1946), and Reader in Organic Chemistry (1961). In 1967 Ritchie was appointed to a Chair of Organic Chemistry and on the retirement of C.W.S. Shoppee he became head of the Department. During 1971-1973 he was Head of the School of Chemistry, following the retirement of R.J.W. Le Fevre. It is perhaps difficult to gauge the full measure of the man's achievement and contributions, such was his modesty and lack of desire to seek special recognition. He simply loved doing Organic Chemistry and seeing the subject flourish. His name is a by-word amongst former students of chemistry of the past 30 years, such was the quality of his lecturing. In research, some 60 students have gained higher degrees under his guidance: four now hold University Chairs, and many others have risen to senior industrial and academic positions, including 6 readers and associate professors. Ritchie's research interests were initially in the field of synthetic heterocyclic chemistry but when after World War II CSIRO initiated the Australian Phytochemical Survey the main thrust of his research effort developed, broadly on the chemistry of natural products of the Australian flora. Stemming from this, a total of over 150 research papers and several monographs have been published over a period of 30 years. The degree of D.Sc. was conferred on Ritchie in 1954 by the University of Sydney particularly for his pioneer- ing work in the biogenetic theory of the plant alkaloids. However, his research interests were to become much broader than this: sizeable and highly significant contributions were also made in most major areas of natural product chemistry. Collaborative efforts on problems of biological importance were especially satisfying, the most recent contribution being the isolation and identi- fication of the toxic substances responsible for St. George disease in cattle. MI ISIC MC iu: iiTia Recognition of Ritchie's work came early. He was the first recipient, in 1948, of the Edgeworth David Medal of the Royal Society of New South Wales, MDl and in 1963 he was awarded the H.G. Smith Medal by the Royal Australian Chemical Institute. Election to the Australian Academy of Science came in 1963. HI A member of the Royal Society of New South Wales since 1939, he contributed 19 papers to the Journal and Proceedings . Ritchie had a dry sense of humour and he could always take heart from the lighter side of a situation. He excelled in the art of understatement mn in the best Australian tradition. It was these qualities together with his fund of unfailing cheerfulness, warmth and enthusiasm which endeared him to his many friends. His untimely departure M is a source of great sadness to them. BICA By, SRIBl UlUl nisoi iiEcr Muni I lilis, I iim, MEMBERSHIP OF THE SOCIETY, APRIL 1977 79 During the year ended 31st March, 1977 the following changes in membership of the Society were effected. ELECTION TO HONORARY MEMBERSHIP JACKSON, Caroline Margaret (address unknown) CAREY, Professor Samuel Warren, D.Sc., Dept, of Geology, University of Tasmania, Hobart, Tasm., 7000. McMINN, Andrew, 80 Kissing Point Road, Turramurra, N.S.W., 2074. ELKIN, Emeritus Professor Adolphus Peter, C.M.G., Ph.D., 15 Norwood Ave., Lindfield, N.S.W. , 2070. MOORE, Peter Stanley, B. Sc. (Hons. ) , St. Ann's College, 187 Brougham Place, North Adelaide, S.A. , 5006. PRICE, Sir Robert, 314 Albert Street, East Melbourne, Viet. 3002. SKRINJARIC, Marica Catherine, 6 Hereford Street, Botany, N.S.W., 2019. ELECTION TO LIFE MEMBERSHIP SLADE, Rhonda Maree, 29 Irvine Street, Kingsford, N.S.W., 2032. MORT, Francis George Arnot ROUNTREE, Phyllis Margaret, D.Sc. THOMPSON, Denise Mary, 82 Awaba St., Mosman, N.S.W. 2088. ELECTION TO MEMBERSHIP RESIGNATION OF MEMBERS AITKEN, Janet Mary, B.Sc., M. Phil., 82 Mimosa Road, Greenacre, N.S.W., 2190. FRAZER, Geoffrey Leon, 4 Bradley Drive, Carlingford, N.S.W., 2118. HARDY, Clarence James, B.Sc., Ph.D., D.Sc., 12 Brassie Street, North Bondi, N.S.W., 2026. HENRY, Hugh Moore, B.A.(Syd.), B.A. (Macq.), Dip. Ed. (Syd.), 11 Fifth Ave., Cremorne, N.S.W., Richard Hugh McDonald Arnot David Somerset Bridges Gurij Gordijew Charles Mark Groden Robert John Gunthorpe Edith Lack John Francis Lovering Allan James George McGillivray Dirk Cornelius Van Dijk Neil Tolmie McRae Yeates 2090. RESIGNATION OF ASSOCIATE MEMBERS McANDREW, John, B.Sc., Ph.D., Division of Mineral Physics, C.S.I.R.O., P.O. Box 136, North Ryde, N.S.W., 2113. Allison Joan Baggs Roderic Gill OBITUARY MARTIN, Helene A., Ph.D., School of Botany, University of N.S.W., Kensington, N.S.W., 2033. Ida Alison BROWNE (1935; Pres. 1953; Life Member 1974) (deceased 21.10.76) MIKULSKI, John, B. Sc. (Hons . ) , 56 Kyooma Street, Tamworth, N.S.W., 2340. Betty Brunsdon BOWEN (1974) (deceased 3.7.76) MORGAN, Thomas Leslie, B.Sc., c/- Sydney Observatory, Sydney, N.S.W., 2000. Keith Edward BULLEN (1946; Honorary Member 1974) (deceased 23.9.76) ROY, Peter Stanton, B.Sc., Ph.D., 16 Hodgson Ave., Cyril Lloyd COOK (1968) (deceased 16.1.77) Cremorne, N.S.W., 2090. Adrian Noel OLD (1947) (deceased 17.11.76) STUBBS-RACE, Michael Anthony, 35 Dress Circle Road, Avalon, N.S.W., 2107. Ernest RITCHIE (1939; Life Member) (deceased 9.4.76) WALLACE, Harry Lachlan, B.E.(Mech.), 25 York St., Beecroft, N.S.W., 2119. WILSON, Mark Hume , B.A. , Dip. Ed., 10 Neil Street, Epping, N.S.W., 2121. ELECTION TO ASSOCIATE MEMBERSHIP BALAREZO, Oscar Walter Mendoza, 4/16 Oxley St., Glebe, N.S.W., 2037. BILLS, Ross Maynard, c/- St. Andrews College, University of Sydney, Newtown, N.S.W., 2042. Arthur Bache WALKOM (1919; President 1943; Life Member) (deceased 2.7.76) FINN, Anne Marie, lA Grove Ave., Penshurst, N.S.W. 2222. Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 81-86, 1977 Occultations Observed at Sydney Observatory During 1974-1976 D. S. King and K. P. Sims ABSTRACT. This paper presents the results of occultations observed with the 29 cm telescope during the years 1974-1976. The following observations of occultations were made at Sydney Observatory with the 29 cm telescope. A tapping key was used to record the times on a chronograph. The reduction elements were computed by the method given in the occul- tation supplement to the Nautical Almanac for 1938 and the reduction completed by the method given there. Since the observed times were in terms of co-ordinated time (U.T.C.), a correction of +0.01218 hours (=43.85 seconds) was applied to these observed times to convert them to ephemeris time with which The Astronomical Ephemeris for 1974 was entered to obtain the position and parallax of the Moon in terms of the FK4 co-ordinate system. Corrections of +0.01244 hour (=44.8 seconds) and +0.01269 hour (=45.7 seconds) were similarly applied to the observed times in 1975 and 1976 re- spectively before entering The Astronomical Ephemeris for 1975 and The Astronomical Ephemeris for 1976 to obtain the position and parallax of the Moon in terras of the FK4 co-ordinate system. The apparent places of the stars for the 1974 occultations were provided by H.M. Nautical Almanac Office. For the 1975 and 1976 occultations, the apparent places of the occulted stars were computed from their positions in the Catalogue of 3539 Zodiacal Stars for Equinox 1950.0 and the Smithsonian Astrophysical Observatory Star Catalogue using a Diehl Alphatronic computer. Table 1 gives the observational material. The serial numbers follow on from those of the previous report (Sims, 1973). The observers were D. S. King (K) , T. L. Morgan (M) , W. H. Robertson (R) , K. P. Sims (S) and H. W. Wood (W) . Except for occultations 788, 818, 826 which were reappearances at the dark limb, the phase observed was disappearance at the dark limb. Table 2 gives the results of the reductions which were carried out in duplicate. The Z.C. or S.A.O. numbers given in Table 1 are from the Catalogue of 3539 Zodiacal Stars for Equinox 1950.0 (Robertson 1940) and the Smithsonian Astrophysical Observatory Star Catalogue. REFERENCES Robertson, A.J. , 1940 Astronomical Papers of the American Ephemeris , 10 (2) Sims, K.P., 1973 J. Proc. Roy. Soc. N.s.W. 108, 1 Sydney Observatory Papers No. 73 TABLE 1 Serial No. S.A.O. or Z.C. No. Mag. Date U.T.C. Observer 770 0107 8.9 1974 Jan. 29 9 53 33.1 R 771 1033 6.8 1974 Feb. 4 10 31 26.3 S 772 092894 9.0 1974 Feb. 27 9 31 07.3 R 773 092896 8.6 1974 Feb. 27 9 36 02.2 R 774 076516 8.3 1974 Mar. 1 9 04 53.0 R 775 0960 6.6 1974 Mar. 3 9 16 32.5 W 776 0962 7.0 1974 Mar. 3 9 25 54.8 W 777 0964 7.0 1974 Mar. 3 9 37 06.8 w 778 1123 7.2 1974 Mar. 4 12 06 10.8 w 779 1127 5.9 1974 Mar. 4 13 02 06.7 w 780 097753 6.8 1974 Mar. 5 9 23 54.0 R 781 0594 6.9 1974 Mar. 28 9 20 42.1 w 782 076389 7.8 1974 Mar. 28 9 20 51.2 w 783 076869 8.6 1974 Mar. 29 8 18 24.8 R 784 076885 8.3 1974 Mar. 29 8 50 22.7 R 785 076913 8.6 1974 Mar. 29 10 00 52.4 R 786 076915 8.4 1974 Mar. 29 10 04 49.3 R 787 077851 7.9 1974 Mar. 30 8 41 42.5 W 788 145934 8.5 1974 Apr. 17 18 33 18.2 R 789 117890 7.9 1974 Apr. 30 12 46 39.3 S 790 138052 8.2 1974 May. 29 10 17 23.6 R 82 D. S. KING AND K. P. SIMS TABLE 1 (cont.) Serial No. S.A.O. or Z.C. No. Mag. Date U.T.C. Observer 791 138556 7.2 1974 May. 30 10 02 22.3 W 792 138592 7.8 1974 May. 30 12 41 17.4 S 793 1845 6.5 1974 May. 31 8 10 25.2 R 794 118054 8.8 1974 June 24 7 59 17.0 S 795 118059 8.7 1974 June 24 8 22 23.0 S 796 118073 8.6 1974 June 24 8 53 53.5 S 797 1705 7.5 1974 June 26 7 32 50.7 S 798 138911 8.0 1974 June 27 9 46 30.5 W 799 138935 7.6 1974 June 27 11 40 24.5 R 800 138945 8.6 1974 June 27 12 42 45.8 R 801 118373 8.5 1974 July 22 7 53 24.9 R 802 1778 7.1 1974 July 24 8 29 34.1 S 803 138552 8.7 1974 Aug. 20 8 53 23.7 R 804 1930 5.6 1974 Sep. 18 8 03 25.5 S 805 185128 8.6 1974 Sep. 22 9 06 15.9) 9 06 20.5) R 806 185127 8.8 1974 Sep. 22 9 09 47.6 R 807 185697 8.3 1974 Oct. 20 9 03 50.2 R 808 185721 8.7 1974 Oct. 20 9 39 22.8 R 809 2968 6.2 1974 Oct. 23 11 03 42.0) S 11 03 44. 2) 810 2969 3.2 1974 Oct. 23 11 12 36.4 S 811 163486 9.3 1974 Oct. 23 11 19 23.2 S 312 3277 7.8 1974 Nov. 22 10 04 37.6 S 813 0233 6.2 1975 Jan. 20 9 53 35.4 S 814 0739 7.4 1975 Feb. 20 10 59 24.5 R 815 1142 7.8 1975 Mar. 22 10 42 31.0 R 816 097018 8.0 1975 Mar. 22 11 20 01.4 R 817 097072 7.9 1975 Mar. 22 12 42 44.4 R 818 162239 7.4 1975 Apr. 3 17 26 21.0 R 819 0792 5.1 1975 Apr. 16 9 21 13.9 R 820 0905 6.7 1975 May. 14 8 04 13.0 M 821 097338 7.5 1975 May. 16 7 36 45.4 M 822 117890 7.9 1975 May. 18 7 43 29.4 R 823 1440 6.7 1975 May. 18 9 05 20.0 R 824 117923 8.3 1975 May. 18 10 07 08.2 R 825 1688 6.3 1975 May. 20 11 14 51.6 R 826 092548 7.1 1975 June 5 19 29 42.2 R 827 139025 7.8 1975 July 15 7 23 30.7 R 828 1853 4.9 1975 July 15 8 23 26.7 R 829 139037 8.3 1975 July 15 8 26 57.2 R 830 2111 7.0 1975 July 17 9 57 39.3 R 831 2241 5.0 1975 July 18 7 52 26.9 S 832 1815 4.8 1975 Aug. 11 9 03 45.6 S 833 158085 7.3 1975 Aug. 12 11 45 53.5 M 834 158631 8.8 1975 Aug. 13 9 21 13.2 S 835 158636 9.0 1975 Aug. 13 9 50 01.8 S 836 158682 8.8 1975 Aug. 13 11 54 42.0 M 837 159286 8.2 1975 Aug. 14 8 39 55.5 R 838 159309 8.0 1975 Aug. 14 10 07 07.0 R 839 2233 5.5 1975 Aug. 14 14 10 03.2 M 840 2498 4.5 1975 Aug. 16 7 47 18.7 R 841 157813 8.7 1975 Sep. 8 9 12 48.8 S 842 1900 7.2 1975 Sep. 8 9 18 21.0 S 843 2173 7.0 1975 Sep. 10 9 38 00.2 S 844 2182 6.3 1975 Sep. 10 10 53 07.7 M 845 159818 8.6 1975 Sep. 11 12 12 36.5 M 846 2338 6.6 1975 Sep. 11 12 48 24.6 M 847 185116 7.1 1975 Sep. 12 11 03 25.1 S 848 2465 7.4 1975 Sep. 12 11 16 32.5 S 849 162748 7.4 1975 Oct. 12 8 53 02.7 R 850 2975 7.0 1975 Oct. 13 8 36 58.3 R 851 163557 8.0 1975 Oct. 13 10 01 19.4 R 852 163571 8.0 1975 Oct. 13 10 15 04.3 R 853 163570 7.9 1975 Oct. 13 10 23 00.9 R 854 3125 6.9 1975 Oct. 14 15 29 59.8 M 855 185325 8.9 1975 Nov. 6 9 29 09.0 R 856 185337 8.3 1975 Nov. 6 9 47 12.3 R 857 145407 8.6 1975 Dec. 8 9 33 58.7 M 858 145406 7.9 1975 Dec. 8 9 35 58.3 M OCCULT ATIONS OBSERVED AT SYDNEY OBSERVATORY DURING 1974-1976 83 Serial No. S.A.O. or Z.C. No. Mag. Date U.T. C. Observer 859 145412 9.3 1975 Dec. 8 10 04 15.3 M 860 145434 8.7 1975 Dec. 8 11 06 54.7 M 861 3151 9.0 1975 Dec. 8 11 11 04.2 M 862 3464 7.1 1976 Jan. 7 11 39 23.1 R 863 0092 8.7 1976 Feb. 5 9 46 40.6 R 864 109398 8.8 1976 Feb. 5 10 03 59.7 R 865 1158 5.2 1976 Feb. 13 8 58 19.4 K 866 095544 7.9 1976 Mar. 10 10 42 55.6 R 867 0971 7.3 1976 Mar. 10 11 19 32.1 M 868 077821 8.8 1976 Apr. 6 9 01 23.5 R 869 077828 8.4 1976 Apr. 6 9 12 35.6 R 870 077840 8.8 1976 Apr. 6 9 20 25.0 R 871 077850 8.4 1976 Apr. 6 9 39 04.1 R 872 0915 4.7 1976 Apr. 6 11 03 26.1 R 873 077515 8.3 1976 May. 3 8 30 55.8 M 874 0873 7.9 1976 May. 3 9 08 13.5 M 875 077547 7.2 1976 May. 3 9 09 27.2 M 876 095939 8.5 1976 May. 4 8 26 59.6 M 877 095945 9.2 1976 May. 4 8 45 25.9 M 878 095965 8.1 1976 May. 4 8 58 48.1 M 879 095980 8.5 1976 May. 4 9 23 50.6 R 880 097009 8.8 1976 May. 5 8 59 56.4 S 881 097004 8.9 1976 May. 5 9 09 58.3 S 882 097030 8.1 1976 May. 5 9 58 59.0 S 883 097032 9.0 1976 May. 5 10 10 23.0 S 884 097056 9.1 1976 May. 5 10 24 03.2 S 885 097083 7.6 1976 May. 5 11 10 05.0 S 886 1271 5.9 1976 May. 6 10 08 51.5 s 887 098491 7.9 1976 May. 7 8 04 16.2 M 888 098487 8.3 1976 May. 7 8 08 22.8 M 889 098494 8.9 1976 May. 7 8 22 27.0 M 890 098344 8.5 1976 June 3 8 23 51.4 R 891 098341 8.7 1976 June 3 8 24 50.0 R 892 1457 6.7 1976 June 4 7 10 20.8 M 893 118059 8.7 1976 June 4 10 47 09.1 M 894 118073 8.6 1976 June 4 11 22 29.9 M 895 1843 6.9 1976 June 7 14 57 45.8 R 896 2117 5.3 1976 June 9 17 19 16.8 R 897 118408 8.6 1976 July 2 8 45 22.6 K 898 1564 6.6 1976 July 2 11 08 54.3 K 899 1792 7.1 1976 July 4 11 56 26.4 R 900 138821 8.7 1976 July 4 12 05 27.6 R 901 139285 8.1 1976 July 5 9 49 39.3 M 902 158546 7.4 1976 July 6 13 23 20.9 K 903 2331 6.4 1976 July 8 7 53 05.4 R 904 2345 6.9 1976 July 8 10 40 21.3 R 905 2658 5.4 1976 July 10 10 44 44.0 R 906 118767 8.8 1976 July 30 8 48 55.4 M 907 118783 7.6 1976 July 30 9 34 23.0 M 908 158173 8.8 1976 Aug. 29 9 05 51.1 S 909 158175 8.6 1976 Aug. 29 9 23 47.4 S 910 2361 4.8 1976 Sep. 28 8 38 50.9 s 911 2368 8.2 1976 Sep. 28 10 16 59.5 s 912 159953 8.9 .1976 Sep. 28 10 23 04.7 s 913 3477 6.6 1976 Oct. 6 10 30 39.7 K 914 162197 9.0 1976 Oct. 28 9 09 43.3 S 915 2789 7.3 1976 Oct . 28 9 35 30.1 S 916 162251 8.4 1976 Oct. 28 10 33 17.7 S 917 2722 7.1 1976 Nov. 24 10 21 23.5 K 918 128469 8.0 1976 Nov. 30 10 56 09.0 R 919 146372 8.3 1976 Dec. 26 11 08 56.7 R 920 3494 4.6 1976 Dec. 27 11 59 30.5 R 921 0413 6.8 1976 Dec. 31 10 48 29.0 S 84 D. S. KING AND K. P. SIMS Serial No. Luna- tion No. P q TABLE 2 pq q^ Ao pAo qAo Coefficient of Aa A 6 770 632 + 41 + 91 17 + 38 83 - 0.4 - 0.2 - 0.3 + 0.8 + 1.00 771 632 + 99 + 13 98 + 13 2 - 0.3 - 0.3 0. 0 + 13.9 - 0.03 772 633 + 91 + 41 83 + 37 17 - 0.3 - 0.3 - 0. 1 + 11.0 + 0.64 773 633 + 70 - 71 49 - 50 51 - 0.5 - 0.4 + 0.4 + 12.3 - 0. 50 774 633 + 57 - 82 32 - 47 68 + 1.2 + 0.7 - 1.0 + 8.9 - 0.77 775 633 + 80 + 59 65 + 48 35 - 1.0 - 0.8 - 0.6 + 12.0 + 0. 50 776 633 + 99 + 16 98 + 16 3 + 0.9 + 0.9 + 0.1 + 13.8 + 0. 05 777 633 + 85 - 52 73 - 45 27 + 1.4 + 1.2 - 0.7 + 10.9 - 0.62 778 633 + 95 - 30 90 - 29 9 - 2.1 - 2.0 + 0.6 + 12.1 - 0.51 779 633 + 79 - 61 62 - 49 38 + 0.4 + 0.3 - 0.2 + 8.9 - 0.78 780 633 + 83 + 56 69 + 47 32 + 0.8 + 0.6 + 0.4 + 13.7 0.30 781 634 + 97 + 24 95 + 23 6 + 0.5 + 0.5 + 0.1 + 13.0 + 0. 35 782 634 + 97 + 23 95 + 23 5 + 0.8 + 0.8 + 0.2 + 13.0 + 0.34 783 634 + 40 - 92 16 - 35 84 - 1.5 - 0.6 + 1.4 + 5.7 - 0.91 784 634 + 93 + 36 87 + 34 13 - 0.1 - 0. 1 0.0 + 12.8 ■f 0.38 785 634 + 87 - 50 76 - 43 25 + 0.4 + 0.4 - 0.2 + 12.1 - 0.49 786 634 + 89 + 46 79 + 41 21 - 0.6 - 0.5 - 0.3 + 12.2 + 0.47 787 634 + 81 + 58 66 + 47 34 - 0.7 - 0.6 - 0.4 + 11.9 + 0.51 788 634 - 54 + 84 29 - 45 71 - 1.3 + 0.7 - 1.1 -12.2 + 0.57 789 635 + 100 - 3 100 - 3 0 + 3.6 + 3.6 - 0. 1 + 13.7 - 0.40 790 636 + 70 - 72 48 - 50 52 + 2. 1 + 1.5 - 1.5 + 5.4 - 0.94 791 636 + 94 + 34 88 + 32 12 - 1.1 - 1.0 - 0.4 + 14.9 - 0.04 792 636 + 99 + 13 98 + 13 2 - 1.5 - 1.5 - 0.2 + 14.4 - 0. 25 793 636 + 51 - 86 26 - 43 74 + 0.8 + 0.4 - 0.7 + 2.6 - 0.99 794 637 + 97 - 24 94 - 23 6 + 0.3 + 0.2 - 0.1 + 12. 1 - 0.58 795 637 + 91 - 42 83 - 38 17 + 0.8 + 0.8 - 0.4 + 10.2 - 0.73 796 637 + 97 + 26 93 + 25 7 - 0.3 - 0.3 - 0.1 + 14.8 - 0. 12 797 637 + 100 + 6 100 + 6 0 - 0.2 - 0. 2 0.0 +14. 1 - 0.33 798 637 + 85 + 53 72 + 45 28 - 0.2 - 0.2 - 0.1 +14.5 •f 0. 19 799 637 + 99 - 13 98 - 13 2 + 0.2 + 0.2 0.0 + 13.0 - 0.48 800 637 + 21 - 98 5 - 21 96 + 1.9 + 0.4 - 1.9 - 2.2 - 0.99 801 638 + 70 - 71 49 - 50 51 + 1.4 + 1.0 - 1.0 + 5.6 - 0.93 802 638 + 54 + 84 29 + 45 71 + 0.7 + 0.4 + 0.6 + 12.1 + 0.58 803 639 + 88 + 47 78 + 42 22 - 0.3 - 0.3 - 0.2 + 14.8 + 0.11 804 640 + 32 + 95 10 + 30 90 - 0.6 - 0.2 - 0.6 + 8.7 + 0.80 805 640 + 99 - 10 99 - 10 1 - 1.0 - 1.0 + 0.1 + 13.8 - 0.09 806 640 + 86 + 50 74 + 43 25 - 3.0 - 2.6 - 1.5 + 11.9 + 0.50 807 641 + 100 + 4 99 + 4 0 - 2.0 - 2.0 - 0.1 + 13.8 + 0. 10 808 641 + 96 + 26 92 + 25 7 - 2.7 - 2.6 - 0.7 + 13. 1 + 0. 32 809 641 + 100 + 5 99 + 5 0 - 1.9 - 1.9 - 0.1 + 13.6 + 0.33 810 641 + 100 - 4 100 - 4 0 - 0.6 - 0.6 0.0 + 14.0 + 0.24 811 641 + 96 - 29 91 - 28 9 - 0.5 - 0.5 + 0.1, + 14.5 - 0.01 812 642 + 92 + 40 84 + 37 16 - 1.1 - 1.0 - 0.4 + 10.5 + 0.71 813 644 + 66 + 75 44 + 50 56 - 1.2 - 0.8 - 0.9 + 5.9 + 0.91 814 645 + 99 - 15 98 - 15 2 - 1.0 - 1.0 + 0.2 + 13.8 - 0.12 815 646 + 60 - 80 36 - 48 64 + 0.7 + 0.4 - 0.5 + 5.9 - 0.91 816 646 + 86 - 51 74 - 44 26 + 1.0 + 0.8 - 0.5 + 10.5 - 0.68 817 646 + 98 - 17 97 - 17 3 - 1.3 - 1.3 + 0.2 + 13.2 - 0.38 818 646 -100 + 8 100 - 8 1 + 0.9 - 0.9 + 0.1 -14.1 - 0.09 819 647 + 99 + 11 99 + 11 1 + 1.3 + 1.3 + 0.2 + 13.9 + 0. 10 820 648 + 95 + 31 90 + 30 10 - 0.1 - 0.1 0.0 + 13.6 + 0.24 821 648 + 42 - 91 18 - 38 83 + 1.6 + 0.7 - 1.4 + 2.8 - 0.98 822 648 + 49 + 87 24 + 43 76 - 0.2 - 0.1 - 0.2 + 11.3 + 0.65 823 648 + 61 + 79 37 + 48 63 - 0.3 - 0.2 - 0.2 + 12.5 0.54 824 648 + 75 + 67 56 + 50 45 + 0.8 + 0.6 + 0.5 + 13.8 + 0.37 825 648 + 98 + 19 97 + 18 4 + 0.3 + 0.3 + 0. 1 + 14.7 - 0.18 826 648 - 91 + 40 84 - 37 16 - 0.1 + 0.1 - 0.1 -14.5 + 0.12 827 650 + 90 + 44 81 + 39 19 - 1.1 - 1.0 - 0.5 -14.7 + 0. 12 828 650 + 67 - 75 44 - 50 56 + 1.9 + 1.3 - 1.4 + 5.7 - 0.93 829 650 + 70 + 71 50 + 50 50 - 0.1 - 0.1 - 0.1 + 13.3 + 0.44 830 650 + 100 + 8 99 + 8 1 - 1.1 - 1.1 - 0.1 + 14.2 - 0.14 831 650 + 83 - 56 68 - 47 32 + o oo + 0.7 - 0.5 + 10.4 - 0.68 832 651 + 91 - 40 84 - 37 16 + 0.4 + 0.4 - 0.2 + 10.8 - 0.69 833 651 + 97 + 26 93 + 25 7 + 0.3 + 0.3 + 0. 1 + 14.6 - 0.02 834 651 + 99 - 13 98 - 13 2 - 0.3 - 0.3 0.0 + 13.4 - 0.36 835 651 + 97 - 23 95 - 23 5 + 0.3 + 0.3 - 0.1 + 12.9 - 0.45 836 651 + 99 - 13 98 - 13 2 + 0.5 + 0.5 - 0.1 + 13.5 - 0.35 837 651 + 61 + 79 38 + 48 62 - 1.2 - 0.8 - 1.0 + 10.4 + 0.68 838 651 + 69 + 73 47 + 50 53 - 1.4 - 0.9 - 1.0 + 11.2 + 0.61 OCCULTATIONS OBSERVED AT SYDNEY OBSERVATORY DURING 1974-1976 85 TABLE 2 (cont.) Serial No. Luna- tion No. P q p2 pq q^ Aa pAa qAa Coefficient Aa A6 of 839 651 + 78 - 63 61 49 39 + 1.2 + 0.9 0.7 + 9.6 0.73 840 651 + 50 - 87 25 - 43 75 + 0.7 + 0.4 - 0.6 + 7.0 - 0.87 841 652 + 59 + 81 35 + 48 65 - 1.1 - 0.6 - 0.9 + 11.9 + 0.58 842 652 + 22 + 98 5 + 21 95 - 1.1 - 0.2 - 1.0 + 7.5 + 0.86 843 652 + 94 - 35 88 - 33 12 _ 0.2 - 0.2 + 0.1 + 12.2 - 0.51 844 652 + 29 + 95 9 + 28 91 - 1.3 - 0.4 - 1.2 + 6.5 + 0.89 845 652 + 75 - 66 56 - 50 44 + 1.0 + 0.8 - 0.7 + 9.7 - 0.73 846 652 + 55 - 84 30 - 46 70 + 1. 1 + 0.6 - 0.9 + 6.7 - 0.88 847 652 + 95 - 31 90 - 30 10 - 0.2 - 0.2 + 0.1 + 13.3 - 0.32 848 652 + 86 + 51 73 + 44 26 - 2.2 - 1.9 - 1.1 +12.0 + 0.51 849 653 + 38 + 92 14 + 35 85 - 0.5 - 0.2 - 0.4 + 2.7 + 0.98 850 653 + 68 - 74 46 - 50 54 - 0.4 - 0.2 + 0.3 + 12.3 - 0.53 851 653 + 75 + 66 56 + 49 43 - 1.7 - 1.3 - 1.1 + 8.0 + 0.83 852 653 + 74 - 67 55 - 50 46 + 0.4 + 0.3 - 0.3 + 12.9 - 0.46 853 653 + 88 + 48 77 + 42 23 - 2.2 - 1.9 - 1.0 + 10.5 + 0.69 854 653 + 91 - 40 83 - 37 16 - 1.7 - 1.5 + 0.7 + 14.7 - 0. 10 855 654 + 98 + 17 97 + 17 3 - 0.4 - 0.4 - 0.1 + 13.8 + 0. 19 856 654 + 87 - 49 76 - 43 24 - 0.6 - 0.6 + 0.3 + 12.3 - 0.48 857 655 + 77 - 64 59 - 49 41 + 0.4 + 0.3 - 0.3 + 13.8 - 0.37 858 655 + 96 + 28 92 + 27 8 - 0.6 - 0.6 - 0.2 + 12.2 + 0.57 859 655 + 90 + 44 80 + 39 19 - 1.7 - 1.5 - 0.7 + 10.6 + 0.70 860 655 + 90 + 42 82 + 38 18 - 1.9 - 1.7 - 0.8 + 10.7 + 0.69 861 655 + 98 - 20 97 - 19 4 + 2.5 + 2. 5 - 0.5 + 14.7 + 0. 12 862 656 + 96 + 27 93 + 26 7 - 0.8 - 0.8 - 0.2 + 12.1 + 0.59 863 657 + 96 + 28 92 + 27 8 - 0.3 - 0.3 - 0.1 + 12.1 + 0.58 864 657 + 43 + 90 18 + 39 81 - 0.6 - 0.3 - 0.5 + 1.7 + 0.99 865 657 + 96 + 28 92 + 27 8 + 0.2 + 0.2 + 0.1 +14.3 + 0.08 866 658 + 42 - 91 17 - 38 83 + 1.8 + 0.8 - 1.7 + 4.7 - 0.95 867 658 + 79 - 61 63 - 48 37 + 0.4 + 0.3 - 0.2 + 10.4 - 0.68 868 659 + 54 - 84 30 - 46 71 + 1.1 + 0.6 - 0.9 + 7.0 - 0.87 869 659 + 53 - 85 28 - 45 72 + 0.9 + 0.5 0.8 + 6.7 - 0.88 870 659 + 100 - 8 99 - 8 1 -»• 0.7 + 0.7 - 0.1 +14.0 - 0.14 871 659 + 69 + 72 47 + 50 53 + 0.1 + 0.1 + 0.1 +10.4 + 0.68 872 659 + 99 + 13 99 + 12 2 + 0.8 + 0.8 + 0.1 + 14.1 + 0.06 873 660 + 75 + 66 56 + 50 44 - 0.4 - 0.3 - 0.3 + 10.9 + 0.63 874 660 + 88 - 47 78 - 41 22 - 0.1 - 0.1 0.0 + 12.2 - 0.50 875 660 + 100 - 3 100 - 3 0 - 1.2 - 1.2 0.0 + 14.1 - 0.07 876 660 + 94 - 35 87 - 33 12 - 1.1 - 1.1 + 0.4 + 12.6 - 0.46 877 660 + 68 - 73 47 - 50 53 - 0.9 - 0.6 + 0.7 + 8.4 - 0.80 878 660 + 99 + 15 98 + 14 2 + 0.3 + 0.3 0.0 + 14.2 + 0.03 879 660 + 68 + 74 46 + 50 54 + 0.7 + 0.5 + 0.5 + 10.8 + 0.65 880 660 + 84 - 54 71 - 45 29 + 1.0 + 0.8 - 0.5 + 10.5 - 0.69 881 660 + 52 - 86 27 - 45 73 + 1.9 + 1.0 - 1.7 + 5.0 - 0.94 882 660 + 64 - 77 41 - 49 59 - 0.2 - 0.1 + 0.1 + 6.9 - 0.88 883 660 + 45 - 90 20 - 40 80 + 2.2 + 1.0 - 1.9 + 3.9 - 0.96 884 660 + 57 + 82 32 + 47 67 - 0.6 - 0.3 - 0.5 + 10.3 + 0.70 885 660 + 99 - 17 97 - 17 3 + 0.7 + 0.7 - 0.1 + 13.4 - 0.36 886 660 + 99 + 14 98 + 14 2 + 0.3 + 0.3 0.0 + 14.5 - 0.11 887 660 + 87 + 49 76 + 43 24 - 0.1 - 0.1 - 0.1 + 14.4 + 0.22 888 660 + 94 - 35 87 - 33 12 - 1.3 - 1.2 + 0.4 + 11.7 - 0.61 889 660 + 89 + 45 80 + 41 21 + 2.0 + 1.8 + 0.9 +14.6 + 0.17 890 661 + 64 - 77 41 - 49 59 - 0.2 - 0.1 + 0.1 + 5.9 - 0.92 891 661 + 51 - 86 26 - 44 74 + 1.6 + 0.8 - 1.4 + 3.6 - 0.97 892 661 + 64 - 77 41 - 49 59 + 1.2 + 0.8 - 0.9 + 5.5 - 0.93 893 661 + 97 - 23 95 - 22 5 + 0.8 + 0.8 - 0.2 +12.6 - 0.53 894 661 + 92 + 38 85 + 35 15 - 0.6 - 0.5 - 0.2 + 14.9 + 0.07 895 661 + 90 + 43 81 + 39 19 - 1.3 - 1.1 - 0.6 + 14.7 + 0.13 896 661 + 65 - 76 42 - 50 58 + 1.4 + 0.9 - 1.1 + 6.9 - 0.88 897 662 + 98 - 20 96 - 20 4 + 0.2 + 0.2 0.0 + 12.8 - 0.52 898 662 + 94 - 34 87 - 32 12 - 3.3 - 3.1 + 1.1 + 11.5 - 0.64 899 662 + 88 - 46 78 - 41 21 - 2.5 - 2.2 + 1.2 +10.3 - 0.72 900 662 + 93 - 37 86 - 35 14 - 1.4 - 1.3 + 0.5 + 11.3 - 0.65 901 662 + 8 -100 1 - 8 100 + 1.0 + 0.1 - 1.0 - 3.2 - 0.98 902 662 + 63 + 77 40 + 49 60 - 0.8 - 0.5 - 0.6 + 11.6 + 0.60 903 662 + 80 + 60 64 + 48 36 - 1.5 - 1.2 - 0.9 +12.2 + 0.51 904 662 + 90 + 43 81 + 39 18 - 1.8 - 1.6 - 0.8 + 13.3 + 0.34 905 662 + 90 + 43 81 + 39 19 - 1.7 - 1.5 - 0.7 +12.2 + 0.51 906 663 + 99 + 11 99 + 11 1 + 0.1 + 0.1 0.0 +14.6 - 0.23 907 663 + 91 + 42 82 + 38 17 - 1.0 - 0.9 - 0.4 +14.9 + 0.09 86 D. S. KING AND K. P. SIMS TABLE 2 (cont.) Serial No. Luna- tion No. P q p2 pq q^ Ao pAa qia Coefficient Aa A6 908 664 + 100 0 100 0 0 + 0.7 + 0.7 0.0 + 14.1 909 664 + 70 + 72 48 + 50 52 + 0.4 + 0.3 + 0.3 +12.7 + 910 665 + 100 + 7 99 + 7 0 - 1.6 - 1.5 - 0.1 + 14.2 - 911 665 + 57 - 82 33 - 47 68 + 0.6 + 0.4 - 0.5 + 7.1 - 912 665 + 89 + 45 79 + 40 20 - 3. 1 - 2.8 - 1.4 +13. 1 + 913 665 + 99 - 17 97 - 17 3 + 0.5 + 0.5 - 0.1 + 14.8 + 914 666 + 77 - 64 60 - 49 40 + 1.6 + 1.2 - 1.0 +12.2 - 915 666 + 59 - 80 35 - 48 65 - 0.6 - 0.3 + 0.5 + 10.1 - 916 666 + 100 + 0 100 0 0 0.0 0.0 0.0 + 14.2 + 917 667 + 99 + 11 98 + 11 1 - 2.2 - 2.2 - 0.2 + 13.9 + 918 667 + 70 + 71 49 + 50 50 - 1.6 - 1.1 - 1.1 + 6.5 + 919 668 + 93 - 35 87 - 33 12 - 2.9 - 2.7 + 1.0 + 14.9 - 920 668 + 37 - 93 13 - 34 86 - 0.5 - 0.2 + 0.5 + 9.8 - 921 668 + 50 - 86 25 - 44 75 + 1.0 + 0.5 - 0.9 + 9.8 - Sydney Observatory, Sydney, N.S.W., 2000. of 0.27 0.50 0.01 0.86 0.37 0.17 0.52 0.71 0.14 0.22 0.90 0.03 0.76 0.74 (Manuscript received 6.4.1977) Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 87-92, 1977 Precise Observations of Minor Planets at Sydney Observatory During 1976 T. L. Morgan ABSTRACT. Positions of 3 Juno, 6 Hebe, 7 Iris, 25 Phocaea, 39 Laetitia, 40 Harmonia and 148 Gallia obtained with the 23 cm camera are given. The programme of precise observations of selected minor planets begun by W.H. Robertson in 1955 is being continued and the results for 1976 are given here. The method of observation are des- cribed in the first paper (Robertson 1958). All the plates were taken with the 23cm camera (scale 116" to the millimeter). Four exposures were taken on each plate, except those for 148 Gallia which had two exposures. In Table 1 are given the means of the posi- tions for all the exposures for each of two sepa- rate groups of reference stars at the mean of the exposure times. The differences in the results for two groups of reference stars average 0?030 sec6 in right ascension and 0V40 in declination in the plates with four exposures. For those plates on which there were only two exposures the corres- ponding results were of 026 sec6 and 0V52. This leads to probable errors for the mean of the two results on the one plate of ofoi3sec6 in right ascension and 0V17 for the plates with four expo- sures, and0.011sec6 and 0V22 for the plates with two exposures. The result for the first pair of images was compared with the result for the last pair by adding the motion computed from the ephemeris for the plates with four exposures. The means of the differences were 0.01 3 seed in right ascension and 0V14 in declination. Comparison of the first and last exposures for the plates with two exposures gave mean differences of 0.057 sec6 and 0'.'74. The images of the plates with only two exposures were, in general, of poorer quality than the other since they were of fainter objects. Along with the fact that there were fewer measurements to be meaned, this explains the larger differences. It is ex- pected that the two results will be combined before they are used. However, they are published in the present form so that any correction to the positions of the reference stars may be conveni- ently applied by using the dependences from Table 2. No correction has been applied for aberration, light time or parallax, but the factors give the parallax correction when divided by the distances. The column headed "0-C" gives the differences be- tween the measured positions (corrected for paral- lax) and the position computed from the ephemeri- dies supplied by the Institute for Theoretical Astronomy in Leningrad. In accordance with the recommendation of Commission 20 of the International Astronomical Union, Table 2 gives for each observation the posi- tions of the reference stars and the dependences. The column headed "R.A." and "Dec." give the seconds of time and arc with the proper motion correction applied to bring the catalogue position to the epoch of the plate. The column headed "Stars" gives the Durchmusterung number taken from either the AGK3 or SAG catalogue. The first column gives a serial number which cross-references Table 1 and Table 2 and also the catalogue from which the reference stars were taken. All plates were reduced by both the methods of dependences and by first order plate constants using the same six reference stars. The r.m.s. residuals of the reference stars was 0'.'2 for AGK3 stars and 0'.'6 for SAG stars. The plates were measured by Mrs. A. Brown, Miss J. Fitt and Miss D. Teale who also assisted with the reductions. The observers at the tele- scope were D. S. King (K) , T. L. Morgan (M), W.H. Robertson (R) and K.P. Sims (S). References : Robertson, W.H. , 1958. Precise observations of minor planets at Sydney Gbservatory during 1955 and 1956. J. Roy. Soc. N.s.W. 92, 18-23. Sydney Observatory Papers No. 33 88 T. L. MORGAN TABLE 1 TABLE 1 POSITIONS OF MINOR PLANETS No. R.A. (1950.0) h m s Dec. (1950.0) o ' " Parallax 0 - C Factors s " s 3 Juno 1976 U.T. 1417 Mar. 09.57017 10 37 12.681 +05 1418 Mar. 09.57017 10 37 12.678 +05 1419 Mar. 29.49720 10 25 04. 260 +08 1420 Mar. 29.49720 10 25 04.251 +08 1421 Apr. 06.47842 10 22 31.290 +09 1422 Apr. 06.47842 10 22 31.280 +09 1423 Apr. 20.44093 10 21 42.695 +10 1424 Apr. 20.44093 10 21 42.687 + 10 1425 Apr. 28.44039 10 23 15.220 + 10 1426 Apr. 28.44039 10 23 15. 206 + 10 1427 May. 04.40008 10 25 16.873 + 10 28 May. 04.40008 10 25 16.884 + 10 29 May. 18.37473 10 32 36.877 + 10 i430 May. 18.37473 10 32 36.882 +10 1431 May. 24.35881 10 36 42.505 + 10 1432 May. 24.35881 10 36 42.553 + 10 1433 May. 31. 33968 10 42 06.386 + 10 1434 May. 31.33968 10 42 06.386 + 10 6 Hebe 19 76 U.T. 1435 July 08.77489 23 57 09.670 -04 1436 July 08.77489 23 57 09.645 -04 1437 July 19.76125 00 08 45.320 -05 1438 July 19. 761,25 00 08 45.342 -05 1439 July 29. 73650 00 17 09.759 -06 1440 July 29.73650 00 17 09.805 -06 1441 Aug. 03.72605 00 20 28.602 -07 1442 Aug. 03. 72605 00 20 28.606 -07 1443 Sep. 13.62625 00 21 01.078 -16 1444 Sep. 13.62625 00 21 01.072 -16 1445 Sep. 20.58474 00 16 58.971 -18 1446 Sep. 20.58474 00 16 58.908 -18 1447 Nov. 04.45911 23 59 57.763 -22 1448 Nov. 04.45911 23 59 57.773 -22 45 45 40.45 40.38 +0.038 ; -5.56 -0.01 -0.2 M 38 38 58.55 58.65 +0.005 -5.90 +0.03 +0.3 R 30 30 34.60 34.46 +0.020 -6.00 -0.01 -0. 1 R 32 32 12.70 12.20 +0.024 -6.11 +0.01 -0.5 R 51 51 38.06 38.43 +0.087 -6.09 -0.01 0.0 S 59 59 05.86 05.56 +0.009 -6.16 0.00 -0. 2 M 55 55 37.96 37.38 +0.033 -6. 15 -0.01 -0.2 R 46 46 11.11 10.62 +0.025 -6. 14 -0.04 +0.1 , K 29 29 49.15 49.38 +0.014 -6. 11 0.00 -0.3 M 35 35 31. 27 31.74 -0.022 -4. 29 -0.02 -0.5 K 18 18 13.28 12.80 +0.003 -4.20 +0.03 -0. 1 S 26 26 01.77 01.30 -0.007 -4.04 +0.03 -0.8 K 11 11 03.24 03. 16 -0.004 -3.90 +0.02 -0.7 M 59 59 06.72 06.84 +0.034 -2.55 +0.04 + 1.4 S 44 44 18.08 18.07 -0.029 -2.29 -0.01 +0.7 R 32 32 49.59 48.07 0.000 -1. 72 +0.01 0.0 S 7 Iris 1976 U.T. 1449 May. 24.76307 20 40 30. 361 -14 1450 May. 24.76307 20 40 30.414 -14 1451 May. 31.74917 20 42 25.916 -14 1452 May. 31.74917 20 42 25.877 -14 1453 June 07.72320 20 43 05.340 -13 1454 June 07.72320 20 43 05.334 -13 1455 June 08.72152 20 43 04.478 -13 1456 June 08.72152 20 43 04.555 -13 1457 July 05.65402 20 32 09.334 -13 1458 July 05.65402 20 32 09.356 -13 1459 July 19.61076 20 19 31.420 -13 1460 July 19.61076 20 19 31. 327 -13 1461 July 27.57269 20 11 12.374 -13 1462 July 27.57269 20 11 12.400 -13 1463 Aug. 03.54606 20 03 52.934 -13 1464 Aug. 03.54606 20 03 53.041 -13 1465 Aug. 18.50215 19 50 11.908 -13 1466 Aug. 18.50215 19 50 11.891 -13 52 52 17.28 18.00 -0.018 -2.85 -0.04 +0.3 M 23 23 39.39 39.06 -0.006 -2.91 -0.03 +0. 2 K 58 58 35.98 34.98 -0.029 -2.98 -0.03 -0.5 R 55 55 19.20 18.77 -0.026 -2.99 -0.05 -0.1 R 03 03 15.82 16.08 +0.018 -3. 11 0.00 +0. 1 K 04 04 57.75 56.64 +0.029 -3. 11 +0.04 -0.1 S 13 13 11.43 11.79 -0.004 -3.08 -0.01 +0.2 K 23 23 27.98 27. 16 -0.011 -3.06 -0.01 +0.4 M 50 50 08.28 08.47 +0.009 -2.99 0.00 +0.5 S PRECISE OBSERVATIONS OF MINOR PLANETS 89 TABLE 1 Ccont.) POSITIONS OF MINOR PLANETS R.A. No. (1950.0) h m s 7 Iris (cont.) 1976 U.T. 1467 Aug. 19.49930 19 49 27.152 1468 Aug. 19.49930 19 49 27.087 1469 Sep. 14.41120 19 40 51.056 1470 Sep. 14.41120 19 40 51.062 25 Phocaea 1976 U.T. 1471 May 03.76039 18 54 03.021 1472 May 03.76039 18 54 03.035 1473 May 24.71360 19 03 38.674 1474 May 24.71360 19 03 38.677 39 Laetitia 1976 U.T. 1475 Mar. 09.62833 12 20 30.435 1476 Mar. 09.62833 12 20 30.462 1477 Mar. 29.56258 12 05 47.977 1478 Mar. 29.56258 12 05 47.970 1479 Apr. 20.50121 11 51 52.716 1480 Apr. 20. 50121 11 51 52.708 1481 May 20.40632 11 46 25.490 1482 May 20.40632 11 46 25.485 1483 May 31.38934 11 48 54.100 1484 May 31.38934 11 48 54.072 40 Harmonia 1976 U.T. 1485 May 05.71730 18 16 56.386 1486 May 05.71730 18 16 56.400 1487 May 31.64588 18 04 58.818 1488 May 31.64588 18 04 58.914 1489 June 07.61087 17 58 30.131 1490 June 07.61087 17 58 30.204 1491 July 05.53388 17 29 06.748 1492 July 05.53388 17 29 06.702 1493 July 16.48002 17 20 38. 232 1494 July 16.48002 17 20 38.134 1495 July 29.45001 17 15 24.744 1496 July 29.45001 17 15 24.762 1497 Aug. 18.38939 17 18 38.559 1498 Aug. 18.38939 17 18 38.496 148 Gallia 1976 U.T. 1499 June 02.68894 19 14 56.976 1500 June 02.68894 19 14 56.968 1501 July 19.53105 18 39 33.054 1502 July 19.53105 18 39 33.030 1503 July 27.51487 18 33 31.892 1504 July 27.51487 18 33 31.930 1505 Aug. 18.43779 18 23 07.730 1506 Aug. 18.43779 18 23 07.696 Dec. Parallax 0 - C (1950.0) Factors O'" s " s -13 -13 51 51 58.63 58.14 +0.010 -2.99 -0.03 +0.7 -14 -14 30 30 41.28 41.85 -0.025 -2.89 +0.01 +0. 1 +00 +00 30 30 06.87 07.10 +0.014 -4.94 -0.01 -1.3 +08 +08 18 18 39.20 38.96 +0.037 -5.88 -0.02 -0.5 o + 55 59.79 -0. ,005 -5.34 +0, ,01 -0.1 +03 55 59.67 +06 29 00.59 -0. ,008 -5.65 -0. ,01 +0.2 +06 29 01.11 +08 32 42.16 +0. ,017 -5.89 -0, ,01 -0.4 +08 32 41.70 +09 13 16.01 -0. ,011 -5.96 +0. ,01 o 1 +09 13 15.66 +08 53 17.51 +0. ,012 -5.93 0. ,00 -0.3 +08 53 17.51 -21 14 17. 00 -0. ,013 -1. ,90 -0.04 +0. 1 -21 14 16. 26 -21 53 53. .60 +0. ,012 -1. ,79 +0.01 0.0 -21 53 52. 95 -22 07 41. 63 -0. ,025 -1, ,76 0.00 -0.2 -22 07 41. 15 -22 59 45. 23 +0, ,040 -1, .63 -0.01 +0.6 -22 59 44. 76 -23 17 12. ,56 -0, .020 -1. ,59 -0.04 +1.1 -23 17 12. ,83 -23 37 59. 39 +0, .011 -1, .53 -0.06 +1.0 -23 38 00. 13 -24 13 07. 30 -0, .017 -1. ,44 -0.01 +0.5 -24 13 06. 95 +04 25 42.44 +0.012 -5.43 -0.07 0.0 +04 25 42.98 +01 25 51.26 -0.003 -5.07 -0.07 +1.2 +01 25 50.79 +00 09 49.77 +0.025 -4.90 -0.03 -0.4 +00 09 49.12 -03 50 59.17 -0.003 -4.37 -0.08 +0.8 -03 50 58.76 S R R M M R M R M R K R K K S K K S K K 90 T. L. MORGAN TABLE 2 REFERENCE STAR POSITIONS AND DEPENDENCES No. Star Depend. R.A. Dec. No. Star Depend. R.A. Dec. 1417 + 7 2339 0.314544 41.210 41.10 1418 + 5 2359 0.361320 25.590 47.92 AGK3 + 5 2374 0.309125 44.993 04.47 AGK3 + 6 2329 0.265840 14.472 29.09 + 6 2335 0.376331 02.414 39.24 + 6 2339 0.372840 36.400 37.09 1419 + 9 2353 0.332811 15.494 40.38 1420 + 8 2355 0.331546 43. 360 27.11 AGK3 + 9 2357 0.329844 24.353 35.10 AGK3 + 9 2356 0.291070 22.328 54.80 + 8 2370 0.337346 30. 311 13.86 + 9 2364 0.377384 33.544 12.19 1421 + 9 2343 0.335182 24.957 56.07 1422 + 9 2341 0.321230 51.294 58.63 AGK3 + 10 2147 0.317978 23.434 11.71 AGK3 + 11 2223 0.351658 03. 179 53.43 +10 2154 0.346840 38.987 35.90 + 9 2359 0.327112 32.760 01.25 1423 + 10 2143 0. 298504 18.868 43.38 1424 +11 2210 0.324855 50.035 49.20 AGK3 + 11 2220 0.306297 19.344 28.79 AGK3 + 10 2146 0.332872 41.863 29.69 +11 2223 0. 395200 03. 179 53.43 + 11 2228 0.342273 27.026 03.99 1425 + 10 2145 0. 363336 27.572 16.61 1426 + 12 2203 0. 284180 57.977 16.23 AGK3 +11 2220 0.302736 19.344 28.79 AGK3 + 10 2147 0.355706 23.434 11.71 + 11 2236 0.333927 08.878 30.25 +11 2235 0.360114 41.574 57.77 1427 + 11 2222 0.332077 47.311 39.07 1428 +12 2208 0.360256 04.254 35.90 AGK3 + 10 2153 0.359047 04.078 02.47 AGK3 + 11 2228 0.345598 27.026 03.99 + 11 2238 0.308876 12.676 52.81 + 11 2243 0.294146 10.510 55.84 1429 + 11 2243 0.322149 10.509 55.84 1430 +12 2224 0.353376 53.637 33.59 AGK3 + 12 2235 0.352834 27.980 25.92 AGK3 + 10 2170 0.325860 52.040 20.53 +10 2174 0.325017 06.649 25.21 + 11 2255 0.320763 16.021 30.80 1431 + 12 2232 0.324764 19.681 16.94 1432 + 10 2174 0.302644 06.649 25.21 AGK3 + 10 2182 0.341428 43.106 34.76 AGK3 + 12 2239 0.350267 59.875 17.87 + 11 2267 0.333807 57.226 58.14 + 10 2190 0.347088 41.719 24.38 1433 + 11 2266 0. 363999 36.271 20.15 1434 + 11 2267 0.262875 57.226 58.14 AGK3 + 9 2399 0.317849 47.182 56.60 AGK3 + 9 2400 0.360746 55.656 12.38 + 11 2282 0.318151 25.429 20.01 + 11 2278 0.376380 28.987 25.37 1435 - 5 6088 0. 290730 04.661 01. 20 1436 - 5 6093 0.393154 17.894 00.32 SAO - 4 5989 0.377952 35.418 27.64 SAO - 6 6337 0.283916 26.204 21.69 - 4 6013 0.331318 46.789 26.79 - 4 6003 0.322930 58.050 06.14 1437 - 5 6117 0.304232 13. 227 40.64 1438 - 5 2 0.358632 02.202 30.71 SAO - 6 14 0.328708 11.294 57.38 SAO - 6 21 0.311564 46. 344 14.42 - 5 27 0.367060 01.509 02.21 - 5 23 0.329804 53.972 28.67 1439 - 7 32 0.304680 20.452 11.60 1440 - 8 29 0.346940 08.889 03.69 SAO - 5 40 0.339425 39.546 36.18 SAO - 6 46 0.326433 18.458 41.59 - 7 44 0.355894 03.343 06.33 - 6 60 0.326626 16.216 51.27 1441 - 9 54 0.332106 40.360 31.80 1442 - 8 39 0.265855 09.689 19.74 SAO - 7 44 0.404491 03.343 06. 33 SAO - 6 58 0.454990 28.736 00.43 - 6 67 0.263402 23.174 40.10 - 8 65 0.279155 35. 113 10.91 1443 -16 51 0.327876 12.531 37.78 1444 -16 52 0.402434 28.666 04.47 SAO -17 43 0.362740 59.710 31.64 SAO -17 55 0.208046 58.673 12.14 -18 62 0.309384 13.304 27.42 -18 52 0.389520 08.762 38.83 1445 -18 34 0.331176 01.307 23.76 1446 -20 32 0.350682 49.990 35.32 SAO -20 42 0. 339772 14.549 48.47 SAO -18 38 0.343082 20.153 38.96 -18 44 0.329052 41.785 27.41 -19 54 0.306236 25.076 30.63 1447 -21 6513 0.326364 29.261 32.34 1448 -2318114 0. 382938 28.205 18.46 SAO -2318127 0.370832 56.615 46.01 SAO -2216550 0.311524 40.000 56.41 -2318132 0. 302804 27.995 55.62 -2318135 0.305538 09.740 16.46 1449 -14 5821 0.338754 02.206 00.02 1450 -15 5750 0.344804 39.845 40.36 SAO -15 5767 0.388916 46.385 04.77 SAO -14 5829 0.310074 26.855 25.60 -15 5775 0.272330 57.701 31.76 -15 5785 0.345123 23.532 44.42 1451 -14 5829 0.308325 26.855 25.60 1452 -14 5821 0.273922 02.206 00.02 SAO -15 5767 0.343935 46.385 04.77 SAO -15 5772 0.378355 36.925 36.94 -14 5855 0.347740 49.039 40.21 -14 5856 0.347723 58.971 48.22 1453 -14 5829 0.315730 26.855 25.60 1454 -14 5826 0.359826 42.465 44.11 SAO -13 5755 0.334980 44.640 12.04 SAO -15 5790 0.299795 21.707 04.14 -15 5793 0.349290 51.103 07.47 -13 5761 0.340380 32.324 19.79 1455 -14 5821 0.325998 02. 206 00.02 1456 -13 5734 0.277973 09.493 59.77 SAO -13 5755 0.342733 44.640 12.04 SAO -15 5785 0.365208 23.532 44.42 -15 5802 0.331269 21.711 38.57 -14 5855 0. 356828 49.039 40.21 1457 -14 4780 0.344222 42.756 26.59 1458 -13 5695 0.370106 45.169 40.02 SAO -13 5700 0.357970 03.213 23.01 SAO -13 5702 0.307794 26.422 53.00 -12 5790 0.297809 08.244 36.41 -13 5720 0.322100 35.903 21.38 r PRECISE OBSERVATIONS OF MINOR PLANETS 91 TABLE 2 (cont.) REFERENCE STAR POSITIONS AND DEPENDENCES No. Star Depend. R.A. Dec . No. Star Depend. R.A. Dec . 1459 -13 5636 0.356374 53.730 21.16 1460 -13 5639 0.266491 37.702 34.44 SAO -14 5733 0.288982 48.077 55. 18 SAO -12 5704 0.367666 49.826 33.97 -12 5712 0.354644 07. 188 42.20 -13 5656 0.365842 35.381 47.13 1461 -14 5664 0. 343942 42. 151 47.53 1462 -12 5659 0. 324108 43.003 01.60 SAO -12 5674 0. 331036 23.946 48. 39 SAO -14 5677 0. 377146 21.849 42.64 -13 5624 0.325022 39. 124 19. 83 -13 5630 0. 298735 48. 206 30.37 1463 -12 5635 0.339223 36. Ill 42.54 1464 -13 5565 0. 373538 48. 141 23. 28 SAO -14 5636 0.368442 07.981 25.20 SAO -12 5644 0.329266 37.417 50.02 -13 5598 0. 292335 29.159 01.05 -14 5652 0. 297196 48. 157 00.54 1465 -14 5559 0.352513 10.975 44.68 1466 -14 5567 0.303872 44. 199 16.91 SAO -13 5506 0.335386 20.625 24.84 SAO -12 5563 0.338038 20. 231 23.25 -14 5593 0.312101 29. 378 17.46 -14 5590 0.358090 06. 552 30.49 1467 -13 5481 0.344948 26.994 33.01 1468 -13 5492 0. 344710 30.397 23.51 SAO -13 5501 0.313188 53. 195 35. 32 SAO -14 5571 0.313864 18.547 50.38 -14 5590 0.341864 06.552 30.49 -14 5593 0.341426 29. 378 17.46 1469 -14 5499 0.307483 21. 862 39.96 1470 -15 5431 0.288972 09.629 17.45 SAO -15 5448 0.342096 32.428 14.38 SAO -13 5459 0.367210 43.610 14.42 -14 5537 0. 350420 13.989 46. 14 -14 5531 0.343818 09. 831 12.06 1471 + 1 3816 0.369805 17. 828 58.82 1472 - 0 3589 0.286526 57.277 08.26 AGK3 + 0 4054 0.279632 13. 254 44.92 AGK3 + 1 3828 0.354045 47.155 30.81 - 0 3617 0.350563 33.664 49.68 + 0 4063 0.359428 59.846 03.71 1473 + 8 3957 0. 316932 19.395 17.41 1474 + 7 3947 0.345446 33. 781 44.95 AGK3 + 7 3961 0. 325425 57.686 09.92 AGK3 + 8 3975P 0.328407 28.710 24.99 + 8 3981 0. 357643 24.680 40.83 + 7 3971 0.326146 00.662 13.82 1475 + 5 2613 0.317862 25.931 04.79 1476 + 4 2602P 0.392019 34.906 29.35 AGK3 + 3 2638 0.387274 56.924 45.57 AGK3 + 4 2608 0.323830 28.661 27.55 + 4 2613 0.294864 14.353 28.47 + 4 2619 0.284150 43. 156 24.90 1477 + 6 2550 0.279700 20.991 54.51 1478 + 7 2511 0.331416 49.089 34.71 AGK3 + 6 2551 0.379351 22.549 59.87 AGK3 + 7 2519 0.338956 35.434 04.87 + 7 2520 0.340949 34.619 24.36 + 6 2560 0.329628 58.921 30.81 1479 + 8 2547 0. 315174 02. 724 03.98 1480 + 8 2549 0.359696 13.003 56.96 AGK3 + 9 2554 0. 340442 17. 582 56.00 AGK3 + 9 2558 0. 312078 36. 260 54.22 + 9 2567 0. 344384 02.580 52.99 + 9 2564 0.328226 57.820 53.39 1481 + 8 2540 0.320757 23.352 37.68 1482 + 9 2543 0.312660 29.579 45.68 AGK3 +10 2336 0.342701 54. 178 04.06 AGK3 + 9 2551 0.359496 04.531 52. 25 +10 2350 0.336542 52.715 50.11 +10 2345 0.327844 27.609 03. 12 1483 +10 2339 0.307470 50.941 30.35 1484 + 8 2541 0. 343334 34.923 27.28 AGK3 + 8 2547 0.365010 02.724 03.98 AGK3 +10 2350 0. 383667 52.715 50. 10 + 9 2558 0.327521 36. 259 54.22 + 9 2557 0. 273000 34.025 23. 88 1485 -20 5078 0. 287048 54.667 25.31 1486 -2212794 0. 343340 08.689 18.76 SAO -21 4954 0.373013 16.538 51.48 SAO -20 5097 0.278694 34.259 21.64 -21 4961 0.339940 17.567 47. 20 -21 4963 0.377965 59.560 29.85 1487 -2212509 0.335853 11.328 30.44 1488 -21 4842 0.309736 29.314 48. 80 SAO -2212617 0.302496 11. 135 32.78 SAO -21 4869 0.312524 55.732 40.87 -20 5015 0.361651 27.525 40.54 -2212618 0.377741 15.612 54.66 1489 -2212373 0. 254886 42.765 45. 22 1490 -22 4470 0. 268478 00.311 07.05 SAO -21 4810 0.352526 07.718 52.50 SAO -2212411 0. 388230 22.000 47.79 -2212530 0. 392588 12.309 51.86 -21 4842 0. 343292 29.314 48.79 1491 -2212058 0.355653 39.450 36.90 1492 -2212063 0.319204 04. 771 46.88 SAO -2313424 0.326986 25.591 13.82 SAO -2313403 0.362462 06.820 56. 35 -2212094 0.317362 31. 735 52.53 -21 4665 0.318335 22.498 52.52 1493 -2313331 0. 287558 17. 735 56.51 1494 -2313344 0.266304 24. 128 25.97 SAO -2212014 0.353110 03.711 54.42 SAO -2212020 0.372058 14.741 16.13 -2313379 0.359334 05.132 54.29 -2313373 0.361640 56.955 04.59 1495 -2211945 0. 340830 58.988 51. 35 1496 -2211935 0.306833 36.571 16.46 SAO -2413250 0. 322978 06. 200 36.67 SAO -2413262 0.328562 50.702 05.97 -2313328 0. 336192 07.482 02.24 -2313320 0.364605 23.793 02.07 1497 -2313313 0. 336407 11.630 53.70 1498 -2413278 0.312855 51.173 19.58 SAO -2413275 0.325618 35.147 19.00 SAO -2313328 0.326252 07.482 02.24 -2413327 0.337976 06.852 41. 54 -2313353 0.360893 47.309 04.36 1499 + 4 4031 0.318434 09.863 06. 26 1500 + 4 4039 0.354210 18. 763 45.12 AGK3 + 5 4101 0.336620 46.239 37.35 AGK3 + 3 3964 0.275798 58.833 54.72 + 3 3978S 0.344946 41.663 13.20 + 4 4065 0.369992 29.506 38.61 92 T. L. MORGAN TABLE 2 (contO REFERENCE STAR POSITIONS AND DEPENDENCES No. Star Depend. R.A. Dec. No. Star Depend. R.A. Dec. 1501 + 1 3738 0.322614 39.785 30.80 1502 + 1 3735 0.352116 18.202 08.28 AGK3 + 2 3661 0.359196 09.193 59.59 AGK3 + 1 3755 0.335816 42.424 23.38 + 0 2244* 0.318191 47.904 07.89 + 1 3770 0.312068 02.763 47.78 1503 + 0 3966 0.352847 33.204 10.56 1504 + 0 3965 0.334754 30.354 32.10 AGK3 - 0 3510 0. 308776 13.602 09.11 AGK3 + 0 3971 0.351710 28.293 08.05 + 0 3985 0.338376 47.093 32.06 - 0 3528 0.313537 53.104 38.39 1505 - 4 4452 0.292943 58.513 39. 19 1506 - 4 4451 0.416906 54.408 49.87 SAO - 3 4282 0.314580 45.382 44.67 SAO - 3 4283 0.282688 55.322 50.00 - 3 4288 0.392477 46.842 25.78 - 2 4643 0.300406 46.894 54.16 * AGK3 No. Sydney Observatory, Sydney, N.S.W., 2000. CManuscript received 6.4.1977^ Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 93-94, 1977 The Essential Oil of the Fly-Repellent Shrub, Pterigeron bubakii I. A. Southwell and J. R. Maconochie ABSTRACT. Steam distillation of Ptepigeron bubakii, a shrub used by the Barkly Tableland aborigines as a fly repellent, yielded an essential oil containing 3~caryophyllene and p-cymene as major components. Activity tests on the housefly showed no significant repellency. INTRODUCTION The genus Pterigeron (Compositae) comprises a number of aromatic species endemic to Australia currently undergoing taxonomic revision (C.R. Dunlop, pers. comm.). Pterigeron bubakii Comin is a small subshrub which is locally common throughout the black soil plains of the Barkly Tablelands in the Northern Territory and extending into both Queensland and Western Australia in similar habitats. The plant is unpalatable to stock and can form relatively thick communities in overgrazed areas. On Muckaty Station, N.T. , the aboriginal people used this species as a fly repellent. The women would place a coolamon containing a baby amongst the bushes to keep the bushflies away (L. Ulyatt, pers. comm.). Local inhabitants also placed a sprig in their hats as a repellent. ESSENTIAL OIL COMPOSITION Dried plant material (274g) was steam- distilled with cohobation in an all-glass apparatus (Hughes 1970) to yield 0.2% of volatile oil. Gas liquid chromatography on a Perkin-Elmer 900 gas chromatograph with a flame ionisation detector indi- cated two main constituents and several minor com- ponents. These were eluted with helium from an FFAP 15m X 0.5mm i.d. support coated open tubular column maintained at 60 for 3 minutes and then pro- grammed at 6 per minute to 170 . The following components were identified by their mass spectra determined on an AEI MS 30 spectrometer and con- firmed by co-chromatography with authentic samples: a-pinene (2%, relative retention time with respect to a-pinene (RR ) 1.00), B-pinene (3%, RR 1.51), a-phellandrene ^6%, RR 2.10), limonene (7%, RR 2.38), p-cymene (43%, RR 2.86), and B-caryophyl- lene (18%, RR^, 5.37). The identity of the major components p-cymene and caryophyllene was further confirmed by infra-red spectroscopy. Percentages were determined with a Hewlett-Packard 3370 A electronic integrator. REPELLENCY considered most likely to be active (Waterhouse 1947; McCulloch and Waterhouse 1947; Kerr, 1951) were tested as fly-repellents. Concentrations of 0.031-2.000 yg per yl of (i) 6-caryophyllene and (ii) the whole oil were shown to have no signifi- cant repellency against the housefly for periods of up to 30 minutes duration. CONCLUSION The components identified in the essential oil of Pterigeron bubakii are odiferous and have been used for the scenting of soaps and cosmetics (Guenther 1949) . As repellency tests showed no significant activity in the essential oil or its major sesquiterpene component B-caryophyllene, de- activation during distillation may have occurred. ACKNOWLEDGEMENTS The authors would like to thank the Mass Spectometry Unit, School of Chemistry, University of Sydney, for the g.l.c.-m.s. analysis, Mrs. B. Toyer, Museum of Applied Arts and Sciences, Sydney, for the g.l.c. and co-g.l.c. analysis and Dr. G. Holan, Division of Applied Organic Chemistry, C.S.I.R.O., Melbourne, for biological testing. REFERENCES Guenther, E., 1949. ESSENTIAL OILS. D. van Nostrand Co. London. Vol. 2. 852 pp. Hughes, A., 1970. Modified receiver for heavier than water essential oils. Chem. Ind. (Lond.) 1536. Kerr, R.W. , 1951. Adjuvants for Pyrethrins in Fly Sprays. C.S.I.R.O. Bulletin 261, 5. McCulloch, R.N. and Waterhouse, D.F., 1947. Labor- atory and Field Tests of Mosquito Repellents. C.S.I.R.O. Bulletin 213, 7. Waterhouse, D.F. , 1947. Insectary Tests of Repellents for the Australian Sheep Blowfly Luoilia auprina. C.S.I.R.O. Bulletin 218, 19. Both the steam volatile oil and the component 94 I. A. SOUTHWELL AND J. R. MACONOCHIE Museum of Applied Arts and Sciences. Harris St., Broadway, N.S.W. 2007 Animal Industry and Agriculture Branch Department of the Northern Territory P.O. Box 291, Alice Springs, N.T. 5750 (Manuscript received 18. 11. 76 ) (Manuscript received in final form 24.5.77) Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 95-98, 1977 A Bottom Profile Across Lake Eyre North, South Australia J. A. Dulhunty ABSTRACT. A bottom profile constructed from soundings obtained across the northern half of Lake Eyre North, shows cross-sectional features of the Warburton and Kalaweerina Grooves of tectonic origin, and areas termed the Cooper and Neales Depressions believed to have been formed by outwash of estuarine sediments and inshore scouring. Very recent or contemporary tectonic development of the grooves, supports previously recorded evidence of contemporary tectonism. INTRODUCTION Two long narrow parallel sided grooves extend from north to south across the northern half of the bed of Lake Eyre North, which slopes gently to the south at an average gradient of 3 cm per km (Fig. lA) . They are almost linnear, with very slight uniform curvature, and are approximately parallel to each other and to the general align- ment of the eastern and western shores of the lake. The Warburton Groove is the wider and longer, ex- tending some 85 km from near the northern shore towards Belt Bay. The Kalaweerina Groove is narrower and shorter extending about 40 km towards Madigan Gulf. Floodwaters entering through the estuaries of the Warburton River and its distributary Kalaweerina Creek, flow south along the Warburton and Kalaweerina Grooves towards Belt Bay and Madigan Gulf respectively. Water from minor floodings, and the first water to arrive before major floodings, flows along the grooves as chan- nels. When the whole lake bed is covered during major flooding, the grooves are completely sub- merged. As they frequently carry flowing water, the grooves could normally be expected to become meandering water courses owing to the extremely low gradient of the surface over which they pass, and the unconsolidated nature of the lake bed silt in which they have developed. However, absence of meandering and their near linnear nature, strongly suggest tectonic origin and control rather than development from normal processes of flowing water. Whilst the presence of the grooves was evid- ent from aerial observation and photography, no- thing was previously known about their depths or cross-sectional profiles. In 1974, while the lake bed was completely submerged by major flood- ing, the author obtained a number of soundings in the vicinity of the Cooper Estuary and south along the eastern shore. Also, in 1975 an east-west line of soundings was obtained across the lake from the Cooper Estuary, by R. Clark and A. and M. Atkinson in their boat Ibis, working in conjunc- tion with an investigation by the author. These soundings gave a precise bottom profile across the lake and the two grooves, and also two depressed areas close to the eastern and western shores. Numerous additional soundings obtained during 1974-75 by J. A. T. Bye of Flinders University, and P. J. Dillon, J. C. Vandenberg, G. D. Will and J. Clark of the South Australian Department of Engineering and Water Supply (Pers. Com. 1974-76) confirmed the occurrence of the depression near the eastern shore extending south from the Cooper Estuary, for which the name "Cooper Groove" was proposed. However, the name "Cooper Depression" is preferred by the present author and used in this paper, as discussed later. The purpose of the present paper is to record the data obtained by R. Clark and A. and M. Atkinson in conjunction with the author and the profile constructed across the lake, and to con- sider results, together with other observations, in relation to the significance and development of the geomorpho logical features concerned. ORIGINAL OBSERVATIONS The first recorded sighting of any part of the Warburton Groove was by J. Ross who in 1869 des- cribed water flowing from the northern shore across the bed of Lake Eyre towards a "deep channel of water" which was undoubtedly the Warburton Groove (Mason 1955). Fifty years later, G. H. Halligan (1923) made the first aerial examination of the lake and described and figured two tongues of water extending to the north, which were the two grooves, but he did not recognise them as linnear depres- sions. In 1929, Madigan (1930) described, as seen from the air, a dark area running north, in Lake Eyre North, which "appeared to be a depression in the lake bed". It was the Warburton Groove, and he was the first to see it as a groove-like featura During the 1949-50 major flooding of the lake, the role oF the Warburton Groove as a water channel was appreciated from aerial reconnaissance, and named the Warburton Groove by Bonython (1955, 1960), who also described water flowing along the Kalaweerina Groove but the name Kalaweerina was not used. When regional vertical aerial photography be- came available in about 1960, the true form and full extent of the two grooves commenced to emerge, and was confirmed by a Skylark Rocket photograph taken from an altitude of 240 km in 1972 (Dulhunty 1975) . The tectonic significance of the Warburton and Kalaweerina Grooves in association with the geological development of Lake Eyre, has been dis- cussed by Wopfner and Twidale (1967) and Williams (1973) . The first precise mapping of the two grooves by photointerpretations, and use of the 96 J. A. DULHUNTY ft A BOTTOM PROFILE ACROSS LAKE EYRE NORTH 97 name "Kalaweerina Groove" was by Williams (1973, 1975) . t THE BOTTOM PROFILE The positions of the Warburton and Kalaweer- ina Grooves, and the Neales and Cooper Depressions are illustrated in the locality map. Fig. lA, to- gether with the profile section line X-Y and river systems entering Lake Eyre. Soundings were made from the sailing craft Ibis, with a graduated pole at points fixed by log line and compass course along the true west bear- ing of the section line X-Y. Where it was nec- essary to deviate from the true west bearing, sounding values were transferred from actual sounding points to the line X-Y, along north-south perpendiculars as illustrated in Fig. IB. This was acceptable as deviation distances were rela- tively very small, and the significant depth changes under investigation were those in an east- west direction. The detail plan of sounding points along the profile section X-Y is shown in Fig. IB, together with the exact positions and widths of the Warburton and Kalaweerina Grooves, and the general areas of the Neales and Cooper Depressions, as determined by aerial photointerpretation. Imme- diately beneath this plan in Fig. IB, is the pro- file section placed with shoreline points and sounding points directly beneath corresponding points in the plan. The profile was constructed by plotting measured depths, at sounding points, in relation to mean lake level of -11.1 m A.H.D.in August 1975. Extrapolation of the profile to the shorelines beyond closely spaced soundings, is indicated by broken lines . RESULTS AND CONCLUSIONS The profile constructed from soundings shows definite groove-like features corresponding closely in position and width with the Warburton and Kala'- weerina Grooves plotted in plan from aerial photo- graphy. The Warburton Groove is the deeper and wider of the two. They stand out very clearly in profile, but their actual depths of 54 and 13 cm in relation to widths of 4.6 and 2.7 km respectively, indicate the extremely shallow nature of the grooves, and the "flatness" or remarkably low relief across the lake bed. There is a slope of 4 cm per km to the west, between the Cooper Depression and the Kalaweerina Groove, then a slight rise to a low ridge between the two grooves. Overall, the bed of the lake exhibits a slight average fall of about 8 mm per km from east to west, in the vicinity of the profile section. The general areas of the Neales and Cooper Depressions, near the western and eastern shores are shown in plan by broken lines in Figs. lA and IB and appear in the profile section in Fig. IB. They are visible in aerial photography of the dry lake bed, but differ considerably in appearance from the two grooves. They have ill-defined sides, are vague in shape and form, and lack de- finite orientation. From low altitude aerial observation of both areas in 1974-75, and examina- tion of the Cooper Depression by boat in 1974, there appeared to be offshore silt of sand banks and inshore channels in the areas of the depres- sions . As they may be due to outwash sand and silt from the Cooper and Neales Estuaries, and in- shore scouring by water flowing south along the shores, the term "depression" is used in this paper so as not to infer tectonic significance as in the case of the "grooves" in the profile. The profile shows the eastern sides of the Warburton and Kalaweerina Grooves steeper than their western wides. When considered in relation to their widths of 4.6 and 2.7 km, this feature would seem to be of some significance, although un- certain in meaning. If the grooves are related to faulting it may mean similar elements of movement beneath each of the grooves. The long narrow nature of the grooves, and their well-defined sides, would seem to indicate sharp fault displacement rather than warping or flexing. Floodwaters from the Warburton River and Kalaweerina Creek, flow south west across the lake bed, then into and south along their respective grooves (Fig. lA) . Aerial photography (South Australian Department of Lands 1961-64 photography) indicates continuation of the grooves to the north from the places where the floodwaters flow into them, as illustrated by broken lines in Fig. lA. This strongly supports previous opinions that they originated as tectonic features. It also indic- ates that floodwaters flow along the grooves as pre-determined channels, and have not been able to develop meanders owing to continued or contemporary tectonic influences or controls. There is no evidence of antiquity of the grooves, and absence of any significant modification by flowing water may mean that they have developed very recently, or are at present developing as a result of contem- porary tectonism evidence of which, at Lake Eyre, has been previously recorded by Wopfner and Twidale (1967), and Williams (1973, 1975). ACKNOWLEDGEMENTS It is wished to acknowledge (i) the co-opera- tion of Messrs. R. Clark and M. and A. Atkinson, and use of their boat Ibis, in obtaining soundings across Lake Eyre North as described in the text, (ii) assistance of Muloorina Station in field investigations, (iii) funds provided by the Australian Research Grants Committee, and (iv) research facilities of the Department of Geology and Geophysics, University of Sydney. REFERENCES Bonython, C. W. , 1955. LAKE EYRE, SOUTH AUSTRALIA. THE GREAT FLOODING OF 1949-50. The Report of the Lake Eyre Committee, Roy. Geogr. Soc. Aust., S. Aust. Br. Griffin Press, Adelaide, 27-36 pp. Bonython, C. W. , I960 . A Decade of Watching for Water in Lake Eyre. Proa. Roy. Geogr. Soc. Aust., S. Aust Br., 61, 1-8. Dulhunty, R., 1975. THE SPELL OF LAKE EYRE, Lowden Publishing Company, Kilmore, p. 81. Halligan, G. H., 1923. Expedition to Lake Eyre, South Australia. Rept. Aust. Assoa. Adv. Sat., 16, 120-128. 98 J. A. DULHUNTY low Madigan, C. T., 1930. Lake Eyre, South Australia. Geogr. Joujp., ?6, 215-240. Mason, B., 1955. LAKE EYRE, SOUTH AUSTRALIA. THE GREAT FLOODING OF 1949-50. The Report of the Lake Eyre Committee, Roy. Geogr. Soc. Aust., S. Aust. Br. Griffin Press, Adelaide, p.l9. Williams, A. F., 1973. Lake Eyre Hap Sheet and Explanatory Notes. GEOLOGICAL ATLAS OF SOUTH AUSTRALIA. 1:250,000 series. Geol. Surv. S. Aust., Adelaide. Willi ams, A. F., 1975. Noolyeana Map Sheet. GEOLOGICAL ATLAS OF SOUTH AUSTRALIA. 1:250,000 series. Geol. Surv. S. Aust., IdI Adelaide. Wopfner, H. and Twidale, C. R., 1967. Geomorpho- logical history of the Lake Eyre Basin, in LANDFORM STETHES FROM AUSTRALIA AND NEW GUINEA, J. H. Jennings and J. A. Mabutt (Eds.). A.N.U. Press, Canberra, 118-143 pp. Department of Geology and Geophysics, The University of Sydney, N.S.W., 2006. (Manuscript received 13.7.1977) DiTS fibr than prod tons anno ;indi sipi Coup l(5la has late play indy iat{ I Hn! subs Jons the expl yeai indi pro! It( I«tl /197' thi) ton: ■odi sp« ton: Chet ) liai Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 99-110, 1977 Influence of Hydrothermal Treatment on Physical and Chemical Properties of Chrysotile Asbestos* P. Hahn-Weinheimer and a. Hirner ABSTRACT. Aqueous suspensions of natural chrysotile fibres of equivalent specific surface area, originating from Advocate, Balangero and Zidani, and synthetic chrysotile were treated hydrothermal ly at pressures ranging from 1 to 300 kp cm”^ and temperatures from 20 to 250 C for 2 hours to 120 days. The resulting solid and liquid phases were investigated by physical and chemical methods. The influence of a second mechanical cleaning process and of hydrothermal treatment on struct- ural, chemical and technical properties of the fibres and on the dissolved phase of chrysotile in water are presented. It can be demonstrated that the mechnical qualities of the fibres depend mainly on the constitution of the fibre surface, and to a lesser degree on the amounts of accessory minerals. Thermodynamic data, as activation energies, activity product constants and Gibbs free energies are calculated and compared with those reported in literature. INTRODUCTION Of all asbestos minerals, chrysotile is the fibre most used in industry and accounts for more than 90% of the demand for asbestos fibres. The production of fibre has increased from the fifty tons at Thetford, Canada in 1877 to the world annual output of 4.3 million tons in 1975, an indication of the still growing economic significance of chrysotile (Lincoln, 1975). Compared with the quality of man-made fibres (glass filament, steel, carbon etc.) chrysotile has an important advantage in terms of basic material properties per unit cost. Chrysotile plays an essential role in the asbestos cement industry, and is used in a wide range of friction , materials for transmitting power or converting ; kinetic energy into heat energy. All known substitutes fail on the grounds of efficiency, consistency, or cost effectiveness. Unfortunately the average extrapolated life for economical ■ exploitation of chrysotile does not exceed forty If years. ■ This study on chrysotile was stimulated by 1| industry, on the one hand, and by unsolved X problems in fundamental research, on the other. 1 It differs from synthetic experiments (Bowen and 1 Tuttle 1949, Yang 1960, 1961, and Kurczyk et al ! 1975) in that aqueous suspensions of fibrous } chrysotile were used, instead of highly ; concentrated gels, with the aim to improve or 4 modify the qualities of the fibres. JWe exposed chrysotile fibres of equivalent specific surface areas to various hydrothermal i conditions, and investigated the physical and » chemical properties of the treated fibres. As the w quantities of treated fibres were in the range of H a few grams up to 50 g per run we were able to ■ perform all analytical work on the same material. ■ * An extract from this publication was presented I Sydney entitled "Investigation of the Behaviour of ( Investigations on the properties of ground chrysotile in water are described by Holt and Clark (1960), Hostetler and Christ (1968), and Choi and Smith (1972). Ball and Taylor (1963) carried out similar experiments, but on a more anolied basis by adding several oxides and hydroxides to the aqueous suspensions of chrysotile In our study, however, the hydrothermal treatment was performed with unground chrysotile fibres under such conditions that, in most instances, the conversion of chrysotile into forsterite, talc, enstatite and antigorite did not take place. Our results are in agreement with recently developed model of Evans et al (1976) which was derived from field observation. MATERIALS AND METHODS The starting materials for the experiments consisted of chrysotile fibres of grade 4D from three different localities, Newfoundland, Italy and Greece. Advocate asbestos (AM) originates from the Advocate deposit situated in the Burlington Peninsula, 5 km north of the small town of Bale Verte. The main body is a harzburgite with dunitic enclaves; it covers an area 3 to 5 km wide and 75 km long. The ultrabasics are mainly near the western contact, where the eruptives of Bale Verte are exposed. The fibre bearing rock (exploited area 1200m x 500m) is wrapped in dark chloritized schists of volcanic origin. The cross fibre is mined by open pit operation. Geological studies were published by Church (1969). The Balangero (BM) open pit mine is located in northern Italy, 30 km northwest of Turin in the hills of San Vittore. This slip fibre deposit occurs in northeastern digitation of the large pre-Triassic peridotitic massif of Lanzo. Although the Lanzo massif and the associated the 25th International Geological Congress 1976 in ’sotile under Hydrothernal Conditions". 100 P. HAHN-WEINHEIMER AND A. HIRNER sequence of green rocks have been studied for more than a century, there exists only one geological study of the Balangero deposit (Riccio, 1969). Riccio concludes that an original Iherzolite has been serpentinised as well as antigorised. These phenomena as well as the fibre formation are attributed to alpine tectonics. The chrysotile asbestos from Zidani CZM) belongs to the slip fibre type. The deposit is located in Thessalia, northern Greece and is mined by extensive underground operations. The ultrabasic body is an outlier of the large ophiolitic sequence of rocks forming Vourinos mountain. Macroscopic examinat- ion of the fibre bearing rocks shows consideration shearing. The country rocks are considered to be Jurassic. No geological study of Zidani is known. The fibre containing rocks were processed in the mining companys' mills as follows: Separation of the fibres by crushing in mills; successive milling for opening and classification of the fibres according to their length and quality. Fibres with a specific surface area of about 20,500 cm^g"^ (determined by N2 -absorption) which did not differ more than 7% from each other, were used. In a wet cleaning process in the Company's labs particles smaller than 100 urn, i.e. the dust- fraction were removed. The remaining cleaned fibres were opened again to the above mentioned specific surface area and named AMC, BMC, ZMC. Synthetic material, which we obtained through the kindness of Prof, de Vynck/Gent, was synthesis- ed from gels in 15 days according to a method similar to that of Bowen and Tuttle (1949). The synthesised material showed DTA and X-ray diffraction peaks characteristic of chrysotile, but no fibrous texture could be observed under the microscope. Electron microscopic examination, however, showed fibrils of 700 to 12,000 % in length with 40 to 70 X inner and 140 to 240 X outer diameters, together with relics of starting gels. A hundred runs of aqueous suspensions of unground cleaned and uncleaned chrysotile fibres from Advocate, Balangero and Zidani were carried out in an oilbath and in 2 litre and 10 litre autoclaves. Pressures were varied from 1 to 300 atms. and temperatures from 2 to 250 C, and the length of runs from 2 hours to 120 days. Pressures SPAGHETTI TUBING (TEFLON) PRESSURE LINE LID OF AUTOCLAVE HOLDER PIECE SEALING BOLTS FILLER (TEFLON) FILTER FILTER HOLDER SAMPLE CELL (TEFLON, 314 ml) SOLID BALLS (TEFLON) Figure 1. Bleeding Device in 2L-Autoclave after Dickson et al. (1963) CHRYSOTILE ASBESTOS 101 _2 pressure P(kp cm ) temperature T(°C) 1 1 50 1 50 100 200 100 200 300 20 80 80 100 100 100 100 200 200 200 fibres AM ZM cleaned fibres AMC BMC ZMC dust-fraction AM BM synthetic chrysotile XX XX X XX XX X X X X X X X X X X X X X X X X X X X X X X X X X X TABLE 1 CONDITIONS OF HYDROTHERMAL EXPERIMENTS IN OILBATH AND 2 LITRE AUTOCLAVE fibre P ( kp cm ^ ) / /T(°C) cone, of chrysotile (%) duration additives ( cone. 1%) AM 50/ 150 2.9 80 hrs B,S,A 100/150 2.9 80 hrs B,S,A 200/ 150 2.9 80 hrs B,S,A 100/ 100 9 50 hrs B,S BM 100/ 100 9 100 hrs B,S 200/ 100 9 100 hrs B,S 100/200 9 100 hrs B,S 200/200 9 100 hrs B,S ZM 100/ 100 9 50 hrs B,S AMC 50/150 2.9 80 hrs SZ 100/150 2.9 80 hrs SZ 200/150 2.9 80 hrs SZ 100/200 9 100 hrs B,S 200/200 9 120 d BMC 100/200 9 100 hrs B,S 200/ 200 9 120 d S ZMC 100/200 9 100 hrs B,S 200/ 200 9 120 d B ^^weight % of fibre B = brucite S = SiOa A = AI2O3 SZ = Si02 Cliquid and solid, resp.) + 20% and 40% ZMC, resp. TABLE 2 HYDROTHERMAL EXPERIMENTS IN 10 LITRE AUTOCLAVE 102 P. HAHN-WEINHEIMER AND A. HIRNER Figure 2. Aggregated chrysotile fibres from Balangero (mode: SEM) . Figure 3. Chrysotile fibres from Zidani (mode: SEM). Figure 4. Aggregated particles in chrysotile (mode: SEM). CHRYSOTILE ASBESTOS 103 and temperatures were monitored and recorded by a ten-channel printer during the runs. The precision of the pressure and temperature regulation devices was 1% and 0.3%, respectively. Immediately after the termination of the runs the suspensions were filtered through 0.45 urn pore size filters, and the solid and liquid phases were analysed separately. The reaction vessel and the 2 litre autoclaves were constructed in such a manner that enabled samples to be bled off during the runs (Fig. 1). The various hydrothermal conditions of the runs in the oilbath and the 2 litre autoclaves are compiled in Table 1. These experiments may be regarded as fundamental research, whereas those in closed 500 ml teflon vessels in the 10 litre autoclave are of a more applied character. The conditions of these experiments are listed in Table 2. In addition, analogous runs were tried in 0.01 normal ammonia. Of course, blanks were carried out for each run. The solid phase was analysed by physical and chemical methods: the transition from single fibrils to aggregated bundles of fibres could be observed by electron microscopy and single fibres were identified by electron diffraction; X-ray diffraction revealed structural differences between serpentine minerals and made possible the determination of the contents of contaminating minerals. Chemical compositions of the solid state were determined by wet-chemical methods. X-ray fluorescence spectroscopy, instrumental neutron activation analysis and electronprobe microanalysis. The surface properties of the different kinds of fibres were investigated by zetapotential measurements. The mechanical properties of chrysotile asbestos in cement products were tested with respect to filtrability and strength (Hirner, 1976). The concentrations of hydrated magnesium and silicon in the liquid phase were analysed by atomic absorption and spectrophotometry, respectively. For the various physical and chemical analyses the fibres were prepared in non destructive manner by suspension methods; for X- ray fluorescence analysis, the calcined samples were fused with lithium metaborate. RESULTS Characterisation of Starting Material The inner and outer diameters of the three types of natural fibres range from 50 to 90 A and from 200 to 500 X, respectively, which is in general agreement with diameters observed by Yada (1971). Uncleaned Advocate fibres are well shaped and soft; those from Balangero are straight and not very flexible, and show geometric shapes in aggregates (Fig. 2) . Uncleaned fibres from Zidani are not well shaped, scaly, with adhering particles (Fig. 3). Fig. 4 shows mineral particles which were identified as platy aggregates of lizardite and antigorite. These minerals were found in all kinds of fibres on a microscopic scale. Diffraction patterns proved our fibres to be mainly clinochrysotile; in addition antigorite was identified in BM, and orthocrysotile was found in ZM and the synthetic material. The contents of accessory minerals are presented in Table 3. The highest amounts of dolomite and magnetite were found in Zidani and Balangero fibres, respectively. The chemical compositions of the untreated fibres are illustrated on Figure 5. The groups of major, minor and trace elements including extreme traces are clearly separated. In spite of the small variations in the major elements a relatively high amount of magnesium is characteristic of the Advocate fibre and a relatively high amount of silica of the Zidani fibre. The surface charge differs between the three fibres; the following zetapotentials were determined: +54 mV for Advocate, +24 mV for Balangero and +18 mV for Zidani. The different surface potentials affect the different mechanical properties. Asbestos cement cakes with fibres from Advocate have a lower density, but higher strength (340 kp cm"^) than those from Zidani (245 kp cm"^). TABLE 3 ACCESSORY MINERALS ASSOCIATED WITH CHRYSOTILE FIBRES (%) DETERMINED BY X-RAY DIFFRACTION ANALYSIS Origin Brucite Dolomite Calcite Magnetite Advocate 0.2 <0.1 0.4 2.5 Balangero 0.2 0.4 0.6 6.3 Zidani <0.1 1.9 0.3 3.1 Influence of Mechanical Cleaning Residual amounts of orthochrysotile in AM and antigorite in BM were almost quantitatively remov- ed; in the Zidani fibre, however, orthochrysotile could not be removed, a fact which indicated that orthochrysotile is present as intergrown fibres. As the amounts of accessory minerals are reduced the integral intensities of the principal reflections (002) and (004) are increased by 25% in the Advocate fibre, by 17% in Balangero fibre, and by 9% in the Zidani fibre. In the Balangero fibre the appreciable amounts of contaminating minerals could be removed down to the limits of detection, i.e. 0.1% for brucite, 0.1% for dolomite, 0.2% for calcite and 0.2% for magnetite. This holds true for magnetite in the Advocate fibre. Brucite in this fibre and dolomite and magnetite in the Zidani fibre, however, are so intimately intergrown that they still remained after mechanical processing. This procedure proved to be efficient in reducing the chromium content in the Advocate and Zidani fibres by a factor of 2.5 to 3.5. In the Balangero fibre the nickel, manganese and titanium contents were lowered by a factor of about 2. Because of the removal of interfering and specifically heavier contaminants the density of asbestos cement decreased, as if the fibre material had passed a cleaning process. Filtrability and strength of the Balangero and Zidani fibres increased simultaneously which clearly indicates that the fibres are the only 104 P. HAHN-WEINHEIMER AND A. HIRNER I I I — •- a> to I— < O LL O oco»ocoE^ ^si o »- OZZONItOtO6 (0.3% still remained). By simple exposure to water brucite can be removed from the cleaned Advocate fibre. 25% to 35% of the contaminating calcium present was dissolved, from all kinds of fibres, when treated hydrothermally. TABLE 4 ZETA-POTENTIALS OF CHRYSOTILES (mV) no cleaning special cleaning origin blank blank treated in H^O normal hydro- thermal conditions Advocate +54 +48 +53 +52 Balangero " + 1% SiO^ + 24 +26 +28 +10 +23 -26 Zidani " + 1% brucite + 18 +17 -13 -31 + 0 -26 Table 4 illustrates the influence of hydro- thermal treatment on the electrical potentials of the fibre surfaces. It is surprising that the surface charge of the Advocate fibres could not be changed. A complex hydrated magnesium layer surrounding the fibres may be the explanation; this layer seems to be much more resistent to hydrothermal influence than brucite. The surface of Balangero fibres with medium zeta-potentials is more sensitive to hydrothermal conditions, whereas the surface charge of Zidani fibres with low zeta-potentials changes its sign even in water under normal conditions. A reversal of the sign of the surface charge can be produced with the Balangero fibres when silica is added (Table 4). As a result of our experiments concerning adsorption of brucite on Zidani fibres, we suggest that experiments concerning the chemical adsorption of organic polymers on unsaturated silanol groups of chrysotile, as described by Cossette and Lalancette (1975) would be particularly successful with hydrothermally treated Zidani fibres. The behaviour of the surface as already described is most important in regard to the contact of the various fibres with the basic medium of the cement slurry and, therefore, the mechanical qualities can be correlated with out observations of the zetapotentials . By hydro- thermal treatment filtrability and strength (determined in the Company's labs according to a modified F.V.T. test) were improved for the Advocate fibre, but not for Balangero and Zidani fibres. The results of the more detailed investigation of the Advocate fibre with respect to filtrability and strength are compiled in the diagram of Figure 6. Filtrability is strongly enhanced by additives (1%) of silica, brucite and alumina at higher temperatures (200 C) , whereas the strength was remarkably reduced if fibres were treated in diluted ammonia. Liquid Phase The behaviour of magnesium in solution under hydrothermal conditions is described in the following; under normal conditions (at 1 kp cm'^ and 20°C, abbreviated 1/20) 7.7 ■ 10"^ mole/1 MgO from AMC, 2.9 ■ 10'^ mole/1 from BMC, and 5.0 • 10'^ mole/1 from ZMC were dissolved after one day. Equilibrium was already reached in this time and maintained until the 20th day. At the higher temperature of 80 C the dissolution of magnesium was strongly enhanced: after 20 days the amounts had increased by factors of 2.9, 2.6, and 1.5 for AMC, BMC and ZMC, respectively. Contaminating minerals containing magnesium, such as brucite were increasingly dissolved at higher temperatures. The amount of dissolved magnesium decreased with increasing pressure. Magnesium ions are prevented from moving from the fibre surface into solution at pressures >50 kp cm"^ and temperatures >100 C. The dissolved silica in aqueous phase behaved differently for the three types of chrysotile fibres: 5.7 ■ 10'^ mole/1 (Si02)aq were dissolved in suspensions of ZMC reaching solubility equilibrium at 1/20 after one day. Increasing temperature caused higher solubility of (Si02)aq whereas pressure had almost no influence. As the silicon layers cover a great area of the surface of the Zidani fibre, continuous dissolution of silicon from the surface is guaranteed; it cannot be excluded, however, that amorphous silica originating from the inner space of the tubular fibres is being dissolved, too. 2 • 10'^ mole/1 dissolved silica were determined in the suspensions of BMC under normal conditions; equilibrium was reached after one day. At 100 C the solubility increased by a factor of 4. Further 106 P. HAHN-WEINHEIMER AND A. HIRNER 1) 100/100 in H2O 2) 100/200 in H2O 3) 50/100 in 0.01 nNH40H 4) 50/200 in 0.01 nNH40H M.R.A.2 • witliout additives (kg cm ) X + 1% Si02 Figure 6. Filtrability and strength of asbestos cement cakes Advocate fibre after treatment. CHRYSOTILE ASBESTOS 107 t.102 100 10- 1) ZMC, CSi02)aq dissolution 2) BMC, (Si02)aq dissolution 3) AMC, Mg^'*' dissolution 4) Chrysotile from Quebec, Mg^'^ dissolution Choi and Smith (1972) 2.5 I I I ■ ' I I I I — 3.0 3.5 Figure 7. Activation Energy Plot ■I -103 increase of the P/T-conditions did not noticeably change the amounts of dissolved Si02. In the case of AMC a distinct decrease of (Si02)aq in the solution was generally observed from the first to the twentieth day, i.e. by a factor of 2.3 at 1/20 up to >10 at 300/200. This indicates that the silicon layer in Advocate chrysotile exposes a relatively small area of contact to water where a local saturation of (Si02)aq is instantly reached. The pH showed almost constant values in the range of 6.5 to 8.0 during all the runs. In the filtrates the conductances were always correlated with the concentrations of the ions. THERMODYNAMIC DATA In principle there are two possible mechanisms responsible for dissolution of serpentine: solid state diffusion and/or dissociation of the mineral surface. Luce et al. (1972) showed for lizardite that dissolution is a function of time squared; strong acid causes a linear relationship. Dissol- ution of ground chrysotile is affected only by dissociation of the large "brucite-like" surface of the fibres. After the first minutes of contact with water hydroxyl groups of chrysotile are already dissolved and final pH is reached. The dissolution of Mg^'*' is proportional to the temperature and is terminated within the first hour at T>45°C. In the following the reactions on the surface of chrysotile equilibrate by readsorpt- ion of MgOH'*' and Mg(H20)6^''' (Choi and Smith 1972). Dissolution of ground chrysotile in water is incongruent for short periods (minutes to hours) and congruent for long periods (days to months). The simplified description given above cannot be fully accepted for the dissolution processes of unground fibres. The latter showed solubility equilibria of the major elements only at 1/20 for BMC and P = 1 kp cm”^ and T<100°C for ZMC for one to twenty days. Within the first day no equilibria were detected for any kind of fibre. A stationary dissolution process under hydrothermal conditions could not be observed in any run. In order to get data suitable for thermodynamic calculations the above mentioned fluctuations and trends were averaged or extrapolated. Activation energies of dissolution for the main constituents of unground chrysotile in water were calculated from the time-dependent solubility plots, assuming a first order reaction. By plotting log [Rate (=slopej,ax)/T] versus 1/T, as it is shown on Figure 7, the energy of activation for the magnesium ions and the silica complexes in water were determined by the slopes of the graphs. For magnesium dissolution from AMC 6.4 kcal/mole (Choi and Smith (1972) found 5.5 kcal/mole), and for silica dissolution from BMC and ZMC 11.25 and 8.5 tz.l kcal/mole were calculated, respectively. If these values are compared with that of 5.3 kcal/mole for self- diffusion of water molecules in water (Glasstone et al. 1941), it is clear that the dominant process of dissolution of unground chrysotile is the diffusion of magnesium and silicon from the surface of the fibre into the suspending medium. r 108 P. HAHN-WEINHEIMER AND A. HIRNER in Hi CO < CTJ u U 4) O u •H rH X w ZMC § equilibrium in HgO ON r^ rlA CM r NO O ON +1 BMC manufactured fibres >100 ^ o 0 o CO ON O ON #»CM ^ r NO o ON +I AMC CO lA r NO O ON 4*1 w r X'-' ^ 3 H 0) 0 0 3 ON 0 • t X -H ON X 0} 0 Pi Vh r tA h ^ -P >N 0 (h O CO tA r OJ w a (j U 0 >. 3 \x> o A >.2 N s •H On > s •H CM r ti -P ^ ai \o o CQ (Uw Z 01 0 bO os +1 3 3 ‘H H c TJ I tA (0 '9^4) 4) 3 0 +1 E H ^ H <1) J. 3 o -3* W *0 » C Sh •H ON NO 0 C «l >.A U ON £ 01 Z 01 o bO H CO 3 y— N •H -p r- c 1 lA 4) \D B •H •rl ON H *0 3 >< CJ •> bO ^ 3 0 3 0 C w s •H H -P lA NO •H 4) 3 4) C\] ON 2 bC 0 E • • £ 01 >N lA ^ b ON >.2 E '-N (Q O B 4) •p 0 3 O o lA 3 ON 3 Xi C\J 0 r •H •rt £ lA o U iH W + 1 T3 •H 3 lA H 3 -H •» O tA » 4) NO 4> ON 2 41 1 3 •0 0 •H H 3 0 0 « X 0 ■ 5H2O 3Mg^'^ -i- 60H~ -t- 2H4Si04^ Kchr = [H4Si04]2 AG° = - RT InKchr = ‘ C^^^f^a with a = reactants and b = products. The Gibbs free energy was derived from the experiments in the oilbath at 80 C using thermo- dynamic data from King et al. (1967) and Hostetler and Christ (1968). In Table 5 our results are compared with those of other authors; of course, the data were recalculated for standard conditions. In general, the values for the Gibbs free energies of fibres of different localities may differ by 2 to 3 kcal/mole. The Advocate fibre is more stable from the standpoint of energy than of Zidani; the one from Balangero holds a mid- position. King et al. (1967) and Hostetler and Christ (1968) found for New Idria chrysotile a value which is exactly the mean between our values for AMC and BMC. The value of Bricker et al (1973) seems to be too high for an average which can be explained by the fact that he used in his calculation a value for talc which is still under discussion (Chernosky 1973). CONCLUSIONS 1) By a mechanical cleaning process of chrysotile fibres all other serpentine minerals are removed from clino-chrysotile. 2) Clinochrysotiles from Advocate and Balangero are resistant to hydrothermal treatment; but in those from Zidani antigorite is formed. 3) The dust-fraction contains significantly more chrysotile if treated hydrothermal ly. 4) With the exception of magnetite accessory minerals are removed by mechanical cleaning and hydrothermal treatment from the fibre, but not from the dust-fraction. Contaminating minerals, which are epitactically intergrown with fibres, are hydrothermal ly dissolved. S') Some minor and trace elements - such as aluminium - are located in the chrysotile structure and their amounts are neither mechanically nor hydrothermal ly influenced. 6) The Gibbs free energy of the three kinds of fibres investigated differ by 2.5 kcal/mole. The calculated activation energies for the dissolution of magnesium and silicon demonstrate that these processes are typical of diffusion from the fibre surface into solution. 7) The surface of chrysotile fibres is of importance to their mechanical properties. Fibres with high positive surface charges show good mechanical qualities and are hydrothermally resistant. Fibres with low positive surface charges, however, change their potentials to negative values and are able to absorb positively charged complexes. Summarizing: chrysotile fibres with high surface charge, such as Advocate, are suitable for the manufacture of asbestos cement products; fibres with low surface potentials, such as Zidani, should be more advantageous for the development of composite materials. ACKNOWLEDGEMENTS We are grateful to Eternit S.A. , Belgivun and its director A. Gosseye for financial support and Professor P. Ney, University of Cologne for intrumental help. REFERENCES Ball, M.C. and Taylor, H.F.W., 1963. An X-ray study of reactions of chrysotile. J. Appl. Chem. London, 13, 145-150. Bowen, N.L. and Tuttle, O.F., 1949. The system Mg0-Si02-H20. Bull. Geol. Soc. Am. , 60, 439-460. Bricker, O.P., Nesbitt, H.W. and Gunter, W.D., 1973. The stability of talc. Amer. Mineral. , 58, 64-72. Chernosky, J.V. , 1973. The stability of chrysotile and the free energy of formation of talc. Geol. Soa. Amer. Abstr. , 5(7), 575. Choi, I. and Smith, R.W. , 1972. Kinetic study of dissolution of asbestos fibers in water. <7. Colloid and Interface Soi. , 40/2, 253-262. Church, W.R., 1969. Metamorphic rocks of Burlington Peninsula and adjoining areas of Newfoundland, and their bearing on continent- al drift in the North Atlantic. Amer. Assoc. Petrol. Geol. Mem. , 12, 212-233. Cossette, M. and Lalancette, J.M. , 1975. Polyhydroxysilicate-polymer reaction products. 3rd Int. Congr. Phys. dhem. Ash. Min., Paper 5.20., Quebec, Canada. Dickson, F.W., Blount, C.W., and Tunell, G. , 1963. Use of hydrothermal solution equipment to determine the solubility of anhydrite in water from 100°C to 275 C and from 1 bar to 1000 bars pressure. Amer. J. Sci. , 261, 61-73. Evans, B.W., Johannes, W. , Oterdoom, H. and Trommsdorff, V. , 1976. Stability of chrysotile and antigorite in the serpentinite multisystem. Schueiz. Min. Petrograph. Mitt. , 56, 79-94. 110 P. HAHN-WEINHEIMER AND A. HIRNER Glasstone, S., Laidler, K.J. and Eyring, H. , 1941. The theory of rate process. McGraw Hill, New York. Hirner, A. , 1976. Beeinflussung der physikalischen und chemischen Eigenschaften von technisch aufbereitetem Chrysotil-Asbest durch hydrothermale Behandlung. Diss. Technical University, Munich, W. Germany. Holt, P.F. and Clark, S.G., 1960. Dissolution of chrysotile asbestos in water, acid and alkali. Nature, 23, 237. Hostetler, P.B. and Christ, C.L., 1968. Studies in the system Mg0-Si02-C02-H20 (I); The activity-product constant of chrysotile. Geoahim. Cosmoohim. Acta, 32, 485-497. King, E.G., Barany, R. , Weller, W.W. and Pankratz, L.B., 1967. Thermodynamic properties of forsterite and serpentine. U.S. Bur. Mines, Rep. Invest. 6962. Kurczyk, H.G., MUnchberg, W.I. and Radermacher, G.I., 1975. Products of hydrothermal reaction in the system. Mg0-Si02-H20. 3. Symposium f. Silikatchemie, Brtlnn, CSSR. 17-19 June, 1975. Lincoln, B., 1975. Asbestos: a world resource, I- 3rd Int. Congr. Rhys. Chem. Ash. Min., Paper 1:1, Quebec, Canada. Luce, R.W., 1969. Dissolution of Magnesium silicates. Ph.D. Thesis, Stanford University, Stanford, California. Luce, R.W., Bartlett, R.W. and Parks, G.A. , 1972. Dissolution kinetics of magnesium silicates. Geoahim. Cosmoohim. Acta, 36, 35-50. Riccio, M. L. , 1969. Le Serpentine amiantifere du Balangero Tesi de Laurea in Geologia applicata Universita degli Studi di Torino, Italy. Wildman, W.E., Whittig, L.D. and Jackson, M.L., 1971. Serpentine stability in relation to formation of iron-rich montmorillonite in some California soils. Amer. Mineral., 56, 587-602. Yada, K. , 1971. Study of microstructure of chrysotile asbestos by high resolution electron microscopy. Acta Crystallogr. , A27 , 659-664. Yang, J.C., 1960. The system Mg0-Si02-H20 below 300°C. I: Low temperature phases from 100 to 300°C and their properties. J. Amer. Ceram. Soc. , 43, 542-549. Yang, J.C., 1961. The growth of synthetic chrysotile fiber. Amer. Mineral. , 46, 748- 752. Department of Geochemistry, Technical University, Munich, West Germany. (Manuscript received 21.12.1976) (Manuscript received in final form 10.5.1977) Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 111-116, 1977 Textures of the Carboniferous Ignimbrites in the Hunter Valley, N.S.W. B. Nashar and a. T. Brakel ABSTRACT. Rocks of the Carboniferous calc-alkaline volcanic series in the Hunter Valley, New South Wales, possess well-defined eutaxitic devitrification and possibly vapour-phase textures and show varying degrees of welding and compaction. The devitrification and vapour-phase textures comprise axiolitic, spherulitic, granular and granophyric forms. The eutaxitic textures suggest that the ignimbrites were emplaced by an ash-flow mechanism with welding commencing after the flow came to rest accompanied by compaction due to the overlying lithostatic load. INTRODUCTION Occurring within the Hunter Valley is a petro- graphic province of Carboniferous volcanic rocks of the andesite-dacite-rhyodacite - rhyolite associat- ion, first described in a series of papers by Osborne (1922a, 1922b, 1925, 1927, 1928a, 1928b), Browne (1927) and Scott (1948). As indicated in Figure 1 the Carboniferous rocks crop out in a belt which extends beyond the Hunter Valley in a north- westerly direction between Tamworth and Gunnedah to east of Moree. The volcanics were developed extensively and regularly throughout the Carbonifer- ous System from Upper Visean to Upper Stephanian time. Scott (1948) and Osborne (1950) first recognised some of these rocks as ignimbrites. ROCK TYPES AND MINERALOGICAL COMPOSITION - The rocks range in composition from pyroxene andesite (approx. 59% Si02) to rhyolitic ignimbrite (approx. 78% Si02) . The pyroxene andesites are the only rocks which can be identified as normal lavas. The others are either ignimbrites or devitrified rock whose original texture has been obscured. Because the magmatic history of the rocks does not favour equilibrium between phenocrysts and groundmass the mineralogical composition and prop- ortion of the phenocrysts do not reflect the chemical composition. Quite acid rocks contain plagioclase and ferromagnesian phenocrysts expected from a more basic magma. The acid components are 'hidden' in the groundmass. On the basis of chemical analyses Nguyen (1976) has termed the rocks pyroxene andesite, hornblende dacitic ignim- brite, rhyodacitic ignimbrite and rhyolitic ignimbrite. Their phenocryst assemblage and percentage and groundmass percentage are given in Table 1. The term ignimbrite is used by the authors in the same way as ash flow tuff is used by Ross and Smith (1961) . The term includes welded tuff because, as will be seen below, the ash flow tuff may be unwelded, partially welded or densely welded. (Smith, 1960). MODE OF OCCURRENCE The Carboniferous ignimbrites in the Hunter Valley occur usually in lithological units ranging TABLE 1 PHENOCRYST AND GROUNDMASS PERCENTAGE ROCK TYPES Qtz K-Feld. PHENOCRYSTS Plag. Biot. Hble Hyp. , Fe-Ti Oxides GROUND MASS Rhyolitic ignimbrite 2 5 2 2 89 Rhyodacitic ignimbrite 6 4 10 3 77 Hornblende dacitic ignimbrite rare 27 2 6 3 rare 2 60 Pyroxene andesite 31 7 5 3 54 B. NASHAR AND A. T. BRAKEL Fig. 1. Distribution in New South Wales of Carboniferous rocks in which ignimbrites occur IGNIMBRITE TEXTURES 113 in thickness from a few metres to about 100m but some are greater. For example, the Mt Bright Rhyolitic Ignimbrite Member near Pokolbin (Brakel, 1973) has been estimated to be about 460m thick and the Paterson Volcanics at Mt Tangorin 310m (Slee, 1968). The aerial extent also varies. Some, such as the Mirannie Volcanic Member and Paterson Volcanics, cover a considerable area and have been used as marker units for a distance of up to 65km. Some units comprise one ash flow while others are multiple and comprise two or three. Terminations of ignimbrite units have been observed and in all instances the rock grades into an agglomerate in which phenoclasts of rounded volcanic material in places exceed by volume the proportion of matrix. Tuffs and tuff breccias are associated with the agglomerate. The units become thinner until they lens out. TEXTURES In handspecimen the rocks are porphyritic. The phenocrysts average about 4mm in size and are set in either a lithic or vitroclastic groundmass which is generally quartzo-feldspathic in composit- ion wherever devitrification has occurred. The plagioclase phenocrysts in the lighter coloured rocks are often pink in colour. The rhyolitic ignimbrites often contain abundant pumiceous frag- ments (Plate lA) . The lithoidal varieties vary in colour depend- ing upon the degree of alteration. The more acid rocks are buff, pink, cream or white but when fresh are usually dark grey as are the andesites. Occasionally, the ignimbrite may be light to dark green in colour due to the presence of secondary celadonite in the matrix. The vitric varieties are black, or greenish black, pitchstones and are often cross-cut by dark red devitrified and haemat- itized layers. These rocks, when weathered, take on a brick red lithoidal appearance. Under the microscope, phenocrysts, particular- ly of quartz, are usually seen to be resorbed and embayed and it is not uncommon for those of biotite and hornblende to be resorbed and their former presence marked by granules of magnetite. The abundance of phenocrysts varies from 5 to 55%. The groundmass of all rock types except the pyroxene andesites is usually eutaxitic and examples of poorly welded, moderately welded and strongly welded ash flow tuffs abound (Plate IB, C, D, E, F). The shapes of the shards in thin section are mostly cuspate but Y, U or C and 0 shapes are common. The shape depends upon the degree of initial fragmentation, subsequent compaction of the ash flow tuff and plane of section. In the loosely compacted rocks, complete bubble shapes and bubbles in various stages of fragmentation can be clearly seen (Plate IB, C) . In many cases compaction has resulted in the shards being compressed against or even moulded around the phenocrysts (Plate IF) . Pumice and lithic fragments of variable size (up to 25mm) are common and these, too, are frequently elongated and aligned parallel with the shards. Often, the ends of the pumice fragments are frayed . Devitrification of shards and pumice is commonly developed. The most usual forms are axiolitic (Plate IIA), granular (Plate IIB) and spherulitic. The crystallization within the spherulitic form is of three types. The first comprises radiating fibres. In the ignimbrites it is sometimes difficult to discern the shape of the spherules when viewed in plane light but when the nicols are crossed the radiating fibres are seen to cut across the shards. The second, although resembling the radiating type in ordinary light (Plate HE) is seen to have a granular texture when observed between crossed nicols (Plate IIF) while the third, again indistinguishable from the first two types in plane light (Plate IIC) reveals grano- phyric texture between crossed nicols (Plate IID). The latter two types occur in the rhyolitic lavas and there is a possibility that they may be the result of vapour -phase crystallization rather than the products of devitrification. Vapour-phase crystallization has been described by Brakel (1967), Marchoni (1968) and Frater (1970). This consists of silica mineralization in cavities in the Y- and 0-shaped shards and small accumulations within what are believed to be pore spaces between gas bubbles. The quartz granules are coarser than those of the devitrification products. Whether these and/or the spherules in the rhyolites are vapour -phase phenomena is uncertain. If they are, the original form of silica would probably have been cristobalite and/or tridymite which has inverted to quartz. Granophyric crystallization has been observed in fine-grained Carboniferous acid rocks from the Gloucester District but no trace of eutaxitic texture is evident in them. Whether their texture represents a coarser than usual devitrification product or primary crystallization from an intrusive melt could not be determined. SIGNIFICANCE OF THE TEXTURES In spite of their age, these Carboniferous lithoidal and vitric volcanic rocks possess well- defined eutaxitic and devitrification textures typical of younger ignimbrites found and described elsewhere. According to Smith (1960) , ignimbrites have been emplaced as hot avalanche-type masses, which probably contained hot gases, and therefore may have been to a greater or less extent autoexplosive. Welding commenced after the flow came to rest and was accompanied by compaction due to the overlying lithostatic load, resulting in shards of glass (or devitrified glass) and fragments of pumice and rock being aligned in a parallel arrangement as indicated in Plate lA, D, E, F. Using Smith's (1960) model of zones of welding which requires zones of no welding at the base and top of the cooling unit and zones of partial welding separating these from a central zone of dense welding, the degree of welding of the shards indic- ates the position of the specimen within a cooling unit. Thus, not only the zonal pattern within the 114 B. NASHAR AND A. T. BRAKEL PLATE I A B C D E F ’umiceous fragments in rhyolitic ignimbrite (Scone 310975) Ion welded rhyodacitic ignimbrite. Plane light (Camberwell 352992) >oorly welded rhyodacitic ignimbrite showing various shapes of shards. ’lane light (Camberwell 352992) toderately welded rhyodacitic ignimbrite. Plane light (Camberwell 171982) Strongly welded rhyodacitic ignimbrite. Plane light (Camberwell 101971) /ery strongly welded rhyodacitic ignimbrite showing compaction of shards iround phenocrysts. Plane light (Camberwell 258985) - IGNIMBRITE TEXTURES 115 0 005 mm PLATE II A Rhyolitic ignimbrite showing axiolitic crystallization of the shards. Crossed nicols (Cessnock 292417) B Rhyolitic ignimbrite showing granulitic crystallization of the shards. Crossed nicols (Cessnock 292417) C Spherulitic crystallization within rhyolite. Plane light (Dungog 940108) D Granophyric crystallization within the spherules depicted in C. Crossed nicols E Spherulitic crystallization within rhyolite. Plane light (Dungog 940108) F Granulitic crystallization within the spherules depicted in E. Crossed nicols 116 B. NASHAR AND A. T. BRAKEL unit may be determined but also the cooling units themselves may be defined. To give two examples, Brakel (1973) has recorded the thick Mt Bright Rhyolitic Ignimbrite Member near Pokolbin whicli shows a gradation from loosely welded texture at its base to extreme welding at its top. At the top of the unit is an unconformity, indicating that the upper, less strongly welded zones have probably been removed by erosion. Prater (1970) has successfully defined welding zones in ash flow tuffs in the Rouchel Brook-Back Creek Area accord- ing to Smith's model. The eutaxitic texture of most of the rocks suggests that they were emplaced by an ash-flow mechanism. The ash-flows, similar to nuSes ardentes, are regarded as mixtures of gas-emitting shards and pumice shreds, with some lithic fragments. At the base and top of the flows, where the opportunity for the loss of heat and volatiles was greatest, the shards became rigid soon after being formed and consequently the material deposited was unwelded or poorly welded. In the centre of the ash-flows, because of the higher concentration of volatiles and heat, the shards and pumice fragments were still in a plastic state when the flows came to rest, and welding took place aided by lithostatic load. REFERENCES Brakel, A.T., 1967. A re-examination of the geology of the Mt View Range-Sweetman's Creek area. B.Sa. (Hons) Thesis , Univ. Newaastle. (Unpub 1 . ) . Brakel, A.T. , 1973. The geology of the Mt View Range District, Pokolbin, N.S.W. J. Proa. Boy. Soa. N.S.W. , 105, 61-70. Browne, W.R., 1927. The geology of the Gosforth district. J. Proa. Boy. Soa. N.S.W. , 60, 213-77. Prater, K.M., 1970. The geology of the Rouchel Brook-Back Creek district, N.S.W. B.Sa. (Hons) Thesis, Univ. Newaastle. (Unpubl.). Marchoni, D.L., 1968. The geology of the Westbrook District, N.S.W. B.Sa. (Hons) Thesis, Univ. Newaastle. (Unpubl.), Nguyen, Van Van, 1976. The geochemistry and petro genesis of the Lower Carboniferous calc- alkaline Volcanic Rock Association in the Hunter Valley, New South Wales. Ph.D. Thesis Univ. Newaastle. (Unpubl.). Osborne, G.D., 1922. The geology and petrography of the Clarencetown-Paterson district. Parts i and ii. Proa. Linn. Soa. N.S.W., 47, 161-98; 519-34: (1922a and b respectively). Osborne, G.D., 1925. The geology and petrography of the Clarencetown-Paterson district. Part iii. Proa. Lin. Soa. N.S.W., SO, 67-79. Osborne, G.D., 1927. The geology of the country between Lamb's Valley and the Paterson River. Proa. Linn. Soa. N.S.W., 52, 85-103. Osborne, G.D., 1928a. The Carboniferous rocks between Glennies Creek and Muscle Creek, Hunter River district. New South Wales. Proa. Linn. Soa. N.S.W., 55, 565-87. Osborne, G.D., 1928b. The Carboniferous rocks in the Muswellbrook-Scone district, with special reference to their structural relations. Proa. Linn. Soa. N.S.W., 55, 588-97. Osborne, G.D., 1950. The Kuttung vulcanicity of the Hunter-Karuah district, with special reference to the occurrence of ignimbrites. J. Proa. Boy. Soa. N.S.W., 85, 288-301. Ross, C.S. and Smith, R.L., 1961. Ash-flow tuffs: their origin, geologic relations and identifications. U.S. Geol. Surv. Prof. Paper 566. Scott, B. , 1948. The geology of the Stanhope district, N.S.W. J. Proa. Boy. Soa. N.S.W., 81, 221-47. Slee, K.J., 1968. The geology of the Cranky Comer basin, N.S.W. B.Sa. (Hons) Thesis, Univ. Newaastle. (Unpubl.). Smith, R.L., 1960. Zones and zonal variation in welded ash flows. U.S. Geol. Surv. Prof. Paper 545F. B. Nashar Department of Geology The University of Newcastle, N.S.W. 2308 A.T. Brakel Geological Survey of Western Australia Regional Geology Branch PERTH, W.A. 6000 (Manuscript received 10.5.1977) Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 117-138, 1977 Petrogenetic Aspects of Some Alkali Volcanic Rocks* J. F. G. Wilkinson ABSTRACT. The roles of low- and high-pressure fractional crystallization are examined and res- ervations expressed concerning their validity as controls in the genesis of certain alkaline volcanics. These reservations arise largely because of very limited data documenting the existence of the appropriate complementary crystal extracts required by fractional crystalliz- ation models. Low-pressure fractionation trends in alkali feldspar-bearing olivine nephelinites from the Inverell area, New South Wales, are defined by the compositions of schlieren and leucocratic veinlets. These indicate the capacity of olivine nephelinites to ultimately yield phonolitic derivatives via transitional malignitic compositions. The vitric residuum in a highly undersaturated vitrophyric phonolite from the Dunedin Volcano, East Otago, is mildly peralkaline. However, peralkalinity in potential derivatives of such highly undersaturated aa- free salic parents, crystallizing nepheline before Ca-bearing alkali feldspars, is largely suppressed until the parent melts experience extensive crystallization on or close to the respective nepheline-alkali feldspar field boundaries. The efficacy of high-pressure fractionation (P > 10 kb), largely controlled by kaersutitic amphibole and lesser olivine and clinopyroxene, in the genesis of certain 'evolved' (but never- theless displaying relatively high 100 Mg/Mg+Fe ratios) alkaline volcanics (hawaiites, nepheline mugearites, nepheline benmoreites, etc.) containing Cr-diopside Iherzolite xenoliths is examined with particular reference to specific eruptives from the Newer Volcanics of Victoria. This model finds little support in the (relative) abundances or frequencies of occurrence of the phases (megacrysts) comprising the alleged high-pressure crystal extracts. It is proposed that at least some 'evolved' high-pressure alkaline volcanics may be the products of direct partial melting of relatively iron-rich Iherzolites (100 Mg/Mg+Fe ~ 80-86). INTRODUCTION It is an honour indeed to be invited by the Royal Society of New South Wales to be their Clarke Memorial Lecturer. In the 1975 Clarke Lecture my old friend K.S.W. Campbell spelt out with strong conviction that W.B. Clarke "really loved" fossils. I am unable to state with the same confidence whether W.B. Clarke had the same feelings towards alkaline rocks but I hope, however, that he would have had at least a certain fondness for them. Fractional crystallization, the central theme of the Lecture, is the genetic control most commonly invoked to explain the diversity of lava types (particularly those more evolved than basalt) in anorogenic volcanic assemblages. It becomes a potentially viable control on the intervention of any mechanical process which removes and inhibits equilibration between early-formed crystals and their parent melts. Low-pressure fractional crystallization trends in alkaline melts can be identified and evaluated by studies of differenti- ated intrusives (commonly sill-like and relatively restricted in thickness and lateral extent), schlieren, and vitric residua in extrusives. Although these studies may delineate the character of potential low-pressure differentiates their viability in explaining the spectrum of composit- ions within a specific volcanic province is never- theless subject to a number of constraints. For * The Clarke Memorial Leoture, delivered before the Royal Society of New South Wales, 14th July, 1977. example, it is not merely sufficient to demonstrate chemical coherence and reasonable materials-balance relations between assumed derivatives, parent liquids and potential (zoned) crystal extracts (the liquidus or near-liquidus phenocrysts in the parents). Within the volcanic province as a whole, there should be at least some tangible evidence of the fundamental corequisite to a fractional cryst- allization model involving multicomponent melts, namely the complementary crystal fraction (s) to each liquid derivative (= evolved aphyric lavas) generated during an event which is often inter- preted as both protracted and complicated. The same constraint, which will be subsequently dis- cussed in more detail, is equally applicable to genetic models based on fractional crystallization controls at high pressures. Coombs and Wilkinson (1969) have summarised low-pressure fractionation data for a spectrum of alkaline liquids, differing in Na/K and Fe/Mg ratios and levels of undersaturation (alkali basaltic to basanitic) . The reader is referred to this paper for a summary of low-pressure fraction- ation trends displayed by differentiated alkaline melts, intrusive and extrusive. However, there were no data at that time documenting the low- pressure fractionation behaviour of olivine nephel- inite - it is, admittedly, either absent or else only a minor component in most alkaline volcanic provinces but olivine nephelinites nevertheless may occur on an essentially semi-regional scale, as in the East African rift system. Low-pressure fractionation trends in a representative from the felsic end of the highly undersaturated alkaline volcanic spectrum (a vitrophyric phonolite) will also be examined. 118 J. F. G. WILKINSON TABLE 1 ANALYSES OF OLIVINE NEPHELINITE, SCHLIEREN, AND SOME CONSTITUTENT MINERALS Analysis number IH 2H IN 2N 3 4 5 6 C.I.P.W, IH . norms 2H Si02 41.92 46.10 43.8 44.1 - - 52.1 51.7 Or 13.3 24.5 Ti02 2.29 1.84 - - . 22.4 18.3 1.04 2.01 Ab 3.1 13.1 AI2O3 13.29 18.10 33.5 33.2 0.46 0.95 1.81 0.80 An 9.7 10.8 Cr203 - - - - 0.46 0.13 - - Ne 18.5 19.6 FG2 0 3 4.14 4.21 0.80 0.79 25.7 32.6 - 24.2 Di 25.8 12.5 FeO 9.03 5.08 - - 47.9 44.1 7.33' 1.8 01 13.6 4.2 MnO 0.23 0.17 - - 0.77 0.93 - 0.32 Mt 6.0 6.0 MgO 8.60 3.39 - - 1.90 1.79 13.7 2.82 11 4.4 3.5 CaO 10.25 6.62 0.14 0.20 0.20 - 22.9 5.61 Ap 3.4 2.4 Na20 4.43 5.80 16.8 16.7 - - 0.57 10.3 H2O 2.3 3.6 K2O H2O+ 2.27 1.43 4.16 2.38 4.86 4.58 “ “ ■ 0.10 Total 100.1 100.2 H2O- 0.87 1.27 - - - - - - 62.9 54.3 P2O5 1.39 0.99 - - - - - - D.I. 34.9 57.2 Total 100.14 100.11 99.9 99.6 99.8 98.8 99.5 99.7 Ne Ks 57.7 21.7 46.7 24.3 'Total Fe reported as FeO. = 100 Mg/Mg+Fe^'" Qz 20.6 29.0 IH Average host olivine nephelinite (4 analyses). Inverell 1:100,000, Sheet 9138, Series R65I, grid ref. LN090040. Analyst G.I.Z. Kalocsai. 2H Average melanocratic schlieren (4 analyses). Analyst G.I.Z. Kalocsai. IN Nepheline in olivine nephelinite IH (ISi+AI+Fe®'*' = 16.05, based on 32 0; Ne?8 . sKs 1 6 . gQZ4 . 9 wt.%). 2N Nepheline in schlieren 2H (ZSi+AI + Fe^'*' = 16.05, based on 32 0; Neys . gKs 1 5 . bQzs . 6 wt.%). 3 Titanomagnetite in host olivine nephelinite (FeO and FeaOs calculated on ulvospinel basis from microprobe analysis with total Fe expressed as FeO) . 4 Titanomagnetite in schlieren (FeO and Fe20a calculated on ulvospinel basis) . 5 Titaniferous salite in host olivine nephelinite (Zxy = 2.00, based on 6 0; Ca4 8 . oMg4 0 . 0 Fe 1 2 . 0 atom %) . 6 Groundraass acmitic pyroxene in melanocratic schlieren (Zxy = 1.99, based on 6 0) . SOME LOW-PRESSURE FRACTIONATION TRENDS Olivine Nephelinites and Associated Schlieren, Inverell Area, New South Wales It is now reasonably well established that low-pressure fractionation of basanitic melts may ultimately yield phonolitic derivatives {see Coombs and Wilkinson, 1969, Fig. 4). Although deduced in a generalised way for some provinces {e.g. Saggerson and Williams, 1964; King, 1965) the nature and extent of fractionation trends of even more undersaturated mafic liquids, namely olivine nephelinites, have yet to be documented quantitatively. Certain olivine nephelinites in the Inverell area, north-eastern New South Wales, carry relatively more leucocratic schlieren (host- residuum field relations clearly indicate 'mother- daughter' consanguinity), thereby providing an opportunity to study the low-pressure derivatives from potentially plagioclase-free ultra-alkaline nephelinitic parents. The 'most extensive development of ultra- alkaline extrusives in north-eastern New South Wales is located several kilometres west of the township of Inverell where the olivine nephelinites (as yet undated radiometrical ly) are the youngest eruptives in a volcanic province of diverse mag- matic affinities. Cainozoic volcanic activity in this area commenced with olivine tholeiites and tholeiites of early Miocene age, soon followed by a mildly undersaturated sodic alkaline series, in which alkali olivine basalt, nepheline basanite and hawaiite are represented (Wilkinson, 1966; McDougall and Wilkinson, 1967; Duggan, 1972; Wilkinson and Duggan, 1973; Wellman and McDougall, 1974) . PETROGENETIC ASPECTS OF SOME ALKALI VOLCANIC ROCKS 119 Na Fig. 1. Microprobe analyses of pyroxenes in Inverell olivine nephelinites and associated raelanocratic schlieren. Total iron is determined as FeO, and FejOs calculated by allotting equivalent molecular proportions of Fe203 to NaaO. Open circle, average clinopyroxene composition in olivine nephelinite. Solid circles, cl inopyroxenes in melanocratic schlieren. The schlieren-bearing olivine nephelinites contain prominent phenocrysts of olivine in a fine-grained groundmass largely composed of prismatic titaniferous salite (average composition Ca49Mg39Fei2 ; of. Table 1, No. 5) which is often included in anhedral plates of nepheline (NeaeKsia Qzs; Table 1, No. IN). Titanomagnetite (Table 1, No. 3; mol.% Fe2Ti04 = 62;4) is relatively abundant and minor soda sanidine (Ab3 90r6 oAni ) and zeolite are groundmass constituents. The larger olivine phenocrysts are Mg-rich {mg = 85; mg = 100 Mg/Mg+Fe where Fe = total Fe calculated as FeO) but olivine phenocrysts, commonly zoned, generally have core compositions in the range mg = 78-70. The compositions of groundmass olivines average mg - 60, and are similar to the compositions of olivines {mg = 59) in the coarser- grained melanocratic schlieren. Rare Cr-diopside Iherzolite xenoliths composed of olivine {mg = 90) , aluminous orthopyroxene (Ca iMg9 iFes) , aluminous chrome diopside and brown spinel are also present in the olivine nephelinite. Contacts between the nephelinitic hosts and the coarser- grained 'mini-pegmatitic ' schlieren (up to 5 cm wide) may be somewhat ill-defined in outcrop but thin-sections indicate a fairly abrupt change in relative average grainsize across contacts. Increases in felsic phases at the expense of olivine, clinopyroxene and titano- magnetite yield modal compositions for the wider (analysed) schlieren appropriate to (but somewhat more leucocratic than) malignite (Troger, 1935). A fairly typical schlieren mode (vol.%) is olivine 2, clinopyroxene 15, nepheline 32, alkali feldspar 31, opaques 9, zeolite and mesostasis 11. The clinopyroxene in these particular schlieren is mainly a titaniferous salite, similar in compos- ition to the clinopyroxene in the olivine nephel- inite host (Fig. iL However, some salite pheno- crysts are now rimmed by green acmitic pyroxene and discrete crystals of acmite-rich clinopyroxene are also present in the groundmass (Table 1, No. 6; Fig. 1), sometimes in a reaction relationship with titanomagnetite. Figure 1 illustrates the solid solution between early-crystallized clino- pyroxenes belonging to the Di-Hd series and later schlieren variants which display progressive Na Fe®'^ ->■ CaMg substitution. The clinopyroxene crystallization trend is thus similar to certain clinopyroxene trends in the Shonkin Sag laccolith, Montana (Nash and Wilkinson, 1970). Nepheline (Table 1, No. 2N) is euhedral towards alkali feldspar, an essentially An-free soda sanidine (Ab370r63). Titanomagnetite (Table 1, No. 4; 51.6 mol.% Fe2Ti04) is depleted in Ti02, Cr203 and MgO, and enriched in MnO, compared with titanomagnetites in the host olivine nephelinite. Areas of mesostasis are dominated by curved alkali feldspar microlites and zeolites. The most evolved 'schlieren' form leucocratic 120 J. F. G. WILKINSON DIOPSIDE Co Mq Si2 Og Fig. 2. Phase equilibrium diagram of the system nepheline-diopside-silica (Schairer and Yoder, 1960). The Inverell olivine nephelinites (Table 1, Analysis IH) and associated melanocratic schlieren (Table 1, Analysis 2H) are indicated by a solid circle and solid square, respectively. Arrows indicate generalised low pressure fractionation trends of olivine nephelinites resulting from the successive separation of Fo-rich olivine, Ca-rich clinopyroxene and nepheline. veinlets (2-5 ram wide) composed of alkali feldspar and nepheline (collectively these phases comprise more than 95% of individual veinlets, with alkali feldspar>>nephel ine) , plus minor Ca-rich pyroxene and zeolite (a potassic phill ipsite) . An average composition of the nepheline cores (Neso Ks i sQzi,) is similar to the compositions of nephelines in the host olivine nephelinite and in melanocratic schlieren (Table 1) but the nephelines are zoned to rim compositions with either higher or lower Na/K ratios than the cores. The former trend is accompanied by enrichment in Si (_af. Brown, 1970) . The sanidines in these highly leucocratic schlieren display differing degrees of enrichment in Ba. Average core compositions of optically clear sanidines essentially devoid of Ba (Ab4 20rs yAn i ) are slightly less Or-rich than -those in the host and in melanocratic schlieren but are zoned out to more Or-rich compositions (Abs sOre 4An i ) (Fig. 3). Slightly turbid alkali feldspars are hyalophanes (with up to 5.2% BaO) , the most Ba-rich variant having the composition Csi sAbsaOrs^Ani . The in- terpretation that these veins represent the most evolved residua in this association is consistent with the enrichment in Ba (x2) and Rb (x3) in the melanocratic schlieren, relative to the host nephelinite. Clinopyroxene remains a Ca-rich salite (Cai, 7Mg4 iFei 2) , and is thus similar in composition to the Ca-rich pyroxenes in the olivine nephelinite and the melanocratic schlieren. Discussion The host rocks conform modally with olivine nephelinite but despite low ah and moderate an (deriving in part from AI2O3 in clinopyroxene and titanomagnetite) , they are not sufficiently under- saturated to permit the appearance of normative lo (compare the compositions of "average" olivine nephelinites listed by Nockolds (1954) and Le Maitre (1976) which contain 6.5 and 13.6% la, respectively). However the Inverell olivine nephelinite (Table 1, No. IH) is similar in com- position to a slightly more Mg-rich example (M = 70.8; M = 100 Mg/Mg■^Fe^'*^ when FeO and FeaOj are known) from Flinders Island, Australia. The 1 bar melting relations of this particular olivine nephelinite have been investigated by Tilley and Thompson (1972). Olivine appears on the liquidus at 1304 °C and is followed in the crystallization interval by clinopyroxene (1150°C) and nepheline (1063°C) . The Inverell olivine nephelinites display a similar order of crystallization and their low-pressure liquidus temperatures, as indicated by experimental data on lavas with comparable compositions, were probably close to 1200°C (Thompson, 1973) . The chemistry of the cl inopyroxene-rich schlieren (Table 1, No. 2H) is similar to that of the "average" malignite (Nock- olds, 1954), the latter being relatively more sodic (Na/K = 3.3) and also mildly peralkaline (1.8 aa) . PETROGENETIC ASPECTS OF SOME ALKALI VOLCANIC ROCKS 121 The field relations, modal compositions, order of crystallization of the host olivine nephelinites (D.I. = 34.9; D.I. = differentiation index, Zqz + ab * or + ne la) , and the respective phase chemistries indicate that the melanocratic pegmatoidal schlieren (D.I. = 57.2) and the leucocratic veinlets have been produced by low- pressure fractionation of Mg-rich olivine (becoming increasingly Fe-rich in successive fractions), Ca- rich clinopyroxene, and a Fe2Ti0i,-rich titanomag- netite. Olivine, clinopyroxene and titanomagnet- ite display the most extensive cryptic variation. In contrast the principal felsic phases (nepheline and alkali feldspar) retained relatively constant compositions. Nevertheless, fractionation of a sodic nepheline (Na/K = 5.3) from the host neph- elinitic liquid (Na/K = 3.0) yielded malignitic residua which are relatively more K-rich (Na/K = 2.0) and which contain nephelines (of early crystallization) slightly enriched in Qz, a consequence of the increased silica activities of the residual liquids (Table 1). The principal mineral phases in the olivine nephelinite-schlieren association define two (somewhat simplified) sub-systems; (i) the mafic assemblage olivine-Ca-rich cl inopyroxene-titano- magnetite which collectively controlled schlieren decreases in TiOa, total Fe, MgO, CaO and M values, and an increase in AI2O3; and (ii) the felsic assemblage nepheline-alkali feldspar, major phases in the Ab-Or-Ne-Ks quadrilateral of the residua system NaAlSi0i,-KAlSi04-Si02 (H2O) , which were largely responsible for alkali controls. When projected (as a function of normative Ne, Di and Qz) in the Fe- and K-free system nepheline (NaAlSi04) -diopside (CaMgSi20e) -silica (Si02) (Schairer and Yoder, 1960), the host olivine nephelinite (Ne42 • 6Dis4 ■ 4QZ3 . 0) plots in the forsterite field and the composition of the average malignitic schlieren (Nesg . oDi27 . ?Qz 1 3 . 3) falls close to the intersection of the olivine, diopside and nepheline phase boundaries (Fig. 2). This particular system appears to be the only synthetic system containing the assemblage forst- erite + diopside + nepheline + liquid which approximates to olivine nephelinite. The order of crystallization of the natural liquids closely resembles that in the undersaturated portion of the simplified synthetic system where successive derivative liquids, following crystallization of forsterite, diopside, and finally nepheline, trend towards the lowest melting composition (1068°C) located on the NaAlSi04-Si02 join (the join I diopside-albite in the anhydrous system is an equilibrium thermal divide). Residual liquids are thus albite- and nepheline-rich (with albite »nepheline) and their analogues in the Inverell assemblages are represented by the highly leuco- cratic schlieren with somewhat comparable alkali feldspar-nepheline ratios. Probably as a result of relatively rapid cooling there are only limited petrographic data in the olivine nephelinites consistent with a reaction relation between olivine and liquid (a feature of the synthetic system) to yield diopside, indicated by dis- continuous, localised salite rims to some olivine phenocrysts. Olivine would be expected to react out only under conditions of equilibrium crystal- lization. In the residua system projection the salic components (less an) of the host olivine nephel- inite plot in the nepheline field (Fig. 3) and successive liquids were displaced (in projection), following the initial crystallization of abundant nepheline, towards a natural nepheline-alkali feldspar field boundary. Nepheline was then joined in the crystallization interval by alkali feldspar. Evidence of peralkalinity in the malignitic schlieren is provided largely by the acmitic rims to salite phenocrysts and by the acmitic pyroxenes in their ' groundmasses ' . Peralkalinity was thus restricted to relatively well-evolved stages in the crystallization histories of these schlieren (Na + K/Al = 0.78), largely following extensive crystallization of nepheline and an essentially An-free soda sanidine. The 'plagioclase effect' (Bowen, 1945), as a potential control in the development of peralkalinity, was therefore minimal. Although it is not possible to obtain com- positions of the most leucocratic veins (interpre- ted as the most evolved differentiates) by direct analysis, approximate compositions nevertheless may be derived indirectly from the compositions of their felsic phases, assuming that the compositions of the residual liquids which ultimately crystal- lized to alkali feldspar-nepheline assemblages fall on or close to the joins between these phases in the residua system. The salic components (less an) of the malignitic schlieren 2H approach (in projection) the low-temperature area in the under- saturated quadrilateral, determined experimentally under anhydrous and hydrous (PH2O = 1 kb) condit- ions (Fig. 3). More evolved residua generated largely by clinopyroxene fractionation must remain in this low-temperature area and hence retain a phonolitic character; they cannot evolve to relatively more saturated or more undersaturated compositions. The join between the Or-rich sanidine rims and the core compositions of the coexisting nephelines in leucocratic veinlets (complex nepheline zoning is disregarded) passes through compositions whose parameters in the low- temperature area in the residua system are approx- imately Ne4oKS3oQz3o i.e. adjacent to but falling at more potassic compositions than the points R and T^i in Figure 3. These parameters may be recast to yield Si02 58.4, AI2O3 24.0, Na20 8.7, K2O 8.9, which obviously define a potassic phono- litic composition. The Na/K ratio of this liquid (1.5) continues the trend of K-enrichment indicated by the melanocratic schlieren (Na/K = 2.1), com- pared with the olivine nephelinite (Na/K = 3.0), and now largely reflect controls resulting from the separation of major soda sanidine (Na/K = 0.78), less potassic than the liquid. Fractionat- ion obviously took place at very low pressures. The compositions of their most evolved different- iates (devoid of modal leucite) actually plot well within the leucite field in the undersaturated quadrilateral of the residua system at 1 bar (Schairer, 1957) . It thus appears that the field of leucite in the Ca-bearing residua system under- goes contraction even at very low pressures and that the minimum melting composition is displaced to more K-rich compositions, compared with the Ca-free system. Finally, it may also be noted that the subordinate phonolites associated with the nephelinite volcanic centres of eastern Uganda display relative enrichment in K (King, 1965). 122 J. F. G. WILKINSON Fig. 3. Host olivine nephelinites (solid circle IH) and melanocratic schlieren (open circle 2H^ from the Inverell district, N.S.W., and their respective nephelines and alkali feldspars plotted in the undersaturated quadrilateral in the residua system (see Table 1) . Analyses 3C and 3R (solid squares) denote core and rim compositions of soda sanidines in leucocratic veinlets (see text). The join 3-3R (dashes) between nepheline core composition 3 and the most Or-rich sanidine rims is also indicated. Portions of the nephel ine-alkal i feldspar boundary curves at 1 bar (solid lines; Schairer, 1957) and PH2O = 1 kb (dashed lines; Hamilton and Mackenzie, 1965) are shown. The cross Tm - minimum on nepheline-alkali feldspar phase boundary at PH2O = 1 R = ternary reaction point in the anhydrous system. The low-pressure fractionation trends in the Inverell olivine nephelinites are similar, at least to compositions with D.l. ~ 50, to those in the melanocratic nepheline basanite (atlantite)- pegmatoid association at Omirai, East Otago (Coombs and Wilkinson, 1969). However the Inverell fractionation trends can be traced to much more evolved (phonolitic) compositions whose generation was at all times independent of plagioclase fractionation. At Omimi, calcic plagioclase is a prominent phase in the host basanites and potash oligoclase (zoned to soda sanidine) an important constituent of the associated pegmatoids. Previous appraisals of low-pressure fract- ionation trends of olivine nephelinitic liquids have been based largely on broad field associat- ions and not tested by the compositions of un- doubted natural derivatives. The following fractionation sequences for natural assemblages have been proposed: - (i) melanephel inite or olivine nephelinite ■* (basanite or tephrite) ->■ phonolite (Wright, 1963; Spencer, 1969); (ii) ankaratrite ->■ (melanephelinite and nephelinite) -* phonolite; (SaggerSon and Williams, 1964); (iii) melanephel inite-nephelinite ->■ phonolite or trachyte (King, 1965); (iv) olivine nephelinite ->■ olivine nephel- inites with increasing K/Na ratios -»■ nepheline benmoreite -<■ mafic phonolite (Sutherland, 1974). Experimental data on the system Na20-Al203- Pe203-Si02 at 1 bar (Bailey and Schairer, 1966; Bailey, 1974) indicate that the quaternary reaction point ac + hem + ne + ab + liquid (the 'ijolite point') is linked via "malignite" on a univariant crystallization path ac + ne + ab + liquid to a quaternary eutectic equivalent to peralkaline phonolite. The present study thus generally confirms the broad trends inferred from natural associations and experimental studies and indicates that phono- litic differentiates may derive from olivine nephelinite via transitional malignitic composit- ions. It also highlights some differences from the previous proposals. In the production of phonolitic differentiates from olivine nephelinites it is not necessary to assign a r8le to "transit- ional" plagioclase-bearing basanitic or tephritic types. Furthermore, the production of trachytic derivatives by fractional crystallization is highly unlikely. It must also be emphasized that on this occasion the volume of derivative phonolite is trivial (probably much less than 1% of that of the parental liquid) and that phonolitic derivat- ives are produced only when fractionation is well advanced. In this context it may be significant that phonolitic field associates of the Inverell olivine nephelinites have not yet been found. It is not clear whether the present data can be applied directly to even more undersaturated Ze-bearing olivine nephelinites fractionating at low pressures (the Ic derives partly from the Ks component of nepheline). These olivine nephelin- ites may be highly sodic (Bailey, 1974, Table 2) and presumably the amount of alkali feldspar in such rocks may be minimal. However the relevant experimental data (Figs. 2, 3) indicate that alkali feldspar only appears as a significant phase at a late stage in the crystallization histories of successive residua and hence, via nepheline fractionation, should be stored initially in successive derivative liquids. PETROGENETIC ASPECTS OF SOME ALKALI VOLCANIC ROCKS 123 TABLE 2 ANALYSES OF VITROPHYRIC PHONOLITE, ITS RESIDUAL GUSS, AND PHENOCRYSTS C.I.P.W. norms Analysis number IR IG IN IF 2 IR IG S1O2 52.75 53.8 45.9 63.6 49.6 Or 25.0 25.6 TiOz 0.29 0.14 - - 0.48 Ab 26.7 28.3 AI2O3 21.62 20.4 33.6 22.7 2.87 An 0.0 - FeaOa 0.96 1.05 0.27“ - - Ne 32.9 28.4 FeO 3.10 3.43 - - 18.6^ Ac - 3.2 MnO 0. 13 0.10 - - 0.58 Ns - 0.2 MgO 0.15 0.10 - - 5.74 Di 8.3 6.9 CaO 2.04 1.58 1.62 3.28 20.5 01 0.5 2.2 NazO 10.38 10.1 16.0 7.94 1.17 Mt 1.4 - K2O 4.24 4.30 3.20 2.15 - 11 0.6 0.3 H2O + 3.61 - - - - Ap 0.2 - H2O- 0.40 - - - - H2O 4.0 - P2O5 0.07 - - - - Total 99.6 95.1 Total 99.74 95.0 100.6 99.7 99.6 D.l. 84.6 82.3 'Total Fe reported as FezOs, ^Total Fe reported as FeO. Ne 56.0 53. 1 IR Vitrophyric phonolite from margin of phonolite dyke. Ks 16.8 17.7 Rocky Point, 1.5 km north-north-east of Port Chalmers, East Otago. Grid reference: Dunedin 1:63,360. Qz 27.2 29.2 164/909802. Analyst G.I.Z. Kalocsai. IG Residual glass in IR. IN Nepheline phenocrysts in IR (ZSi+Al+Fe®'*' = 16.06 based on 32 0; Neeo ■ 2Ks i i . sQze . o wt.%). IF Anorthoclase phenocrysts in IR (Ori 3 . zAbe 9 . gAn i e . 9 wt.%). 2 Sodian ferrosalite phenocrysts in IR (Exy = 2.02, based on 6 0; Ca4 7. 7Mgi a • eFes 3 . 7 atom %) . A Highly Undersaturated Vitrophyric Phonolite from the Dunedin Volcano, New Zealand The phase petrology of moderately under- saturated phonolitic liquids plotting in the alkali feldspar field in the residua system and the compositional trends of liquid residua generated essentially by low-pressure alkali feldspar fractionation have been evaluated by Carmichael (1964) and Nash et at. (1969) . There are apparently no data defining fractionation trends and compositions of residua from more highly undersaturated salic melts whose crystal- lization commenced in the nepheline field in the residua system. Natural melts with this com- position are apparently quite rare (see Hamilton and Mackenzie, 1965, Fig. 5); the nosean phono- lite of Wolf Rock, Cornwall, is one example for which chemical data on the host and its felsic phases are available (Tilley, 1959) but this volcanic is essentially holocrystalline and its fractionation behaviour at low pressures is unknown. The highly undersaturated vitrophyric phono- lite (Table 2, No. IR) discussed herein was collected from the margin of a phonolite dyke about 0.5 metre wide, intruding alkali basalt (Careys Basalt) at Rocky Point, some 1.5 km north- east of Port Chalmers, near the eroded core of the Dunedin V'olcano (see Allen, 1974, Fig. 2). The vitrophyric phonolite contains conspicuous pheno- crysts of euhedral nepheline (Table 2, No. IN) and euhedral to subhedral anorthoclase (Table 2, No. IF), which sometimes displays marginal resorption. Green sodian ferrosalite (Table 2, No. 2), and titanomagnetite (with 51.3 mol. % Fe2Ti04) are minor phenocryst phases. The phenocrysts are set in abundant pale green glass containing a minor microlite component which consists largely of pale green, feathery quench clinopyroxene (sometimes localized around opaque oxide granules). Rare microlites of apatite and dark brown(?) aenigmat- ite are also present. The mode of the analysed specimen (Table 2, No. IR) is: nepheline 16, anorthoclase 9, clinopyroxene 2, glass (+ micro- lites) 73, titanomagnetite trace (vol.%). Textural relations between phenocrysts indicate that titanomagnetite and clinopyroxene crystallized initially and that they were followed in the cryst- allization sequence by relatively abundant nepheline and later by anorthoclase. The Rocky Point phonolite (Table 2, No. IR) is distinctly sodic (NazO/KzO = 2.4; Nese-oKsie.e QZ27.2 wt.%), and highly undersaturated (ne = 32.9). In major element chemistry it is very 124 J. F. G. WILKINSON Fig. 4. Vitrophyric phonolite (IRJ , near Point Chalmers, East Otago, its residual glass (IG), nepheline (IN) and anorthoclase phenocrysts (IF) plotted in the undersaturated quad- rilateral in the residua system. Analysis numbers refer to analyses listed in Table 2. Portions of the nepheline-alkali feldspar boundary curves at 1 bar (dashed lines: Schairer, 1957) and at PH2O = 1 kb (solid lines: Hamilton and MacKenzie, 1965) are shown. The cross T^ = minimum on nepheline-alkali feldspar phase boundary at PhjO = 1 kb; NesoKsigQzai and 750°C. The ternary reaction point R in the anhydrous system (at approximately Ne4 6KS2 1QZ3 3) falls close to 2G. Analyses 2F, 2G, 2R (solid circles) respectively denote anorthoclase phenocrysts and residual glass in a kenyte from South Victoria Land, Antarctica (Carmichael, 1964, Analyses IR, IG, IF). Compositions X and Y (open circles) are synthetic compositions (Runs 811 and 847, respectively) used to determine 3-phase boundaries at P^jO = 1 kb (Hamilton and MacKenzie, 1965). similar to the Dicks Hill phonolite, Otago Penin- sula (Price and Coombs, 1975) which is, however, somewhat less undersaturated (ne = 25.9) and which also displays a higher Fe203/Fe0 ratio, a feature of other phonolites in the Dunedin Volcano (Price and Coombs, 1975; Price and Chappell, 1975). Discussion Microprobe analysis of the glass (Table 2, No. IG; Nes 3 . ] Ks 1 7 . 7QZ2 9 . 2) indicates that it is mildly peralkaline, with minor ao and a trace of ns. Points for microprobe analysis of the glass (using a defocussed beam) were restricted to areas of glass devoid of quench phases and, within the domain of the section, well removed from phenocryst phases. To some extent the appearance of normative ac and ns has been determined by the somewhat arbitrary (but probably the most rational) procedure of recalculating the Fe203 and FeO contents of the glass in the same ratio as that of the host. The extent and character of possible changes in the glass chemistry as a result of hydration (particularly as this might affect Na20 and K2O) cannot be evaluated but because of com- positional consistencies with experimental data (Fig. 4), it is considered that significant changes in the chemistry of the glass probably have not occurred. In Figure 4, some relevant experimental data for the quadrilateral NaAlSi308-KAlSi30e-NaAlSi0i,- KAlSiO^ are indicated, for the anhydrous (1 bar) and hydrous systems (PH2O “ ^ kb) . At the latter pressure the temperature minimum Tm on the nepheline-alkali feldspar field boundary (T^ = NesoKsi 9QZ3 1 ; T = 750 ± 7°C) is located close to the ternary reaction noint R in the dry system (R is located at ap nately Ne46Ks2iQz33 at 1020°C) but it indicates the displacement, with increasing Ppi^Q) the field boundary in the An- free system towards more undersaturated composit- ions (see Hamilton and MacKenzie, 1965; Morse, 1969). A similar trend exists in analogous systems with increasing An-contents (Norris and MacKenzie, 1976). The host phonolite, its nepheline and anorth- oclase phenocrysts (recalculated An-free), and the residual glass have been plotted in the residua system (Fig. 4). In this projection the host fall in the nepheline field, so that initial precipit- ation of this phase moved the composition of residual liquids (in projection) towards a natural nepheline-alkali feldspar field boundary located at relatively less undersaturated compositions. Nepheline was then joined by alkali feldspar, in this instance an An-rich anorthoclase. This inter pretation of the felsic crystallization sequence is consistent with textural relations- between the felsic phenocrysts. It will be noted that the residual liquid approaches but does not reach the experimental field boundaries (Fig. 4). If leach- ing of alkalis (particularly Na20) has occurred, the original obsidian would have contained more ns but its position in the residua system would remain essentially unchanged unless it is assumed that Si and A1 were also removed from the glass. The similar configuration of the three-phase boundary delineated by the join IN-IR-IG and experimentally determined three-phase boundaries suggests that alkali leaching, if it occurred, was not selective. If Si and A1 have been in fact removed the residual liquid IG would plot closer to the experimental nepheline-feldspar field boundaries, a trend which is also achieved if conventional CIPIV normative procedures are dis- regarded and all excess Na20, reflecting PETROGENETIC ASPECTS OF SOME ALKALI VOLCANIC ROCKS 125 peralkalinity, is allotted to ns, FeaOs being assigned exclusively to mt. The parameters of the glass in the residua system then become Nes2.3 Ksi 7. 3QZ30 • 4 and the glass now plots very close, perhaps coincidentally, to the hydrous experimental field boundary determined at PH2O “ ^ Be this as it may, the crystallization history of the melt may have been polybaric, and hence by analogy with the experimental system the natural field boundary may have varied its position slightly in response to varying Ph20- This interpretation is consistent with the occurrence of partially resorbed anorthoclases of (?) early crystallization which subsequently were in dis- equilibrium with liquid as the field boundary shifted towards the Ab-Or join, probably in response to decreased Pp]20 following loss of water along dyke contacts. Nepheline continued to crystallize, now accompanied by a somewhat later generation of euhedral anorthoclase. Quenching finally occurred at a temperature of approximately 775°C, suggested by the composition of the nepheline phenocrysts (Hamilton, 1961) . Carmichael (1964) and Nash et al. (1969) have demonstrated the peralkaline character of residua (in which Na + K > Al) in vitrophyric phonolites whose felsic crystallization commenced in the alkali feldspar field. Peralkalinity is a con- sequence of the so-called 'plagioclase effect' (Bowen, 1945). In synthetic mixtures of NaAlSi30e with a Ca-bearing second component (e.g. CaSiOs) the feldspar which initially separates is a Ca- bearing plagioclase. Peralkalinity thus may result from the preferential incorporation of Ca (and Al) into sodic feldspar crystallizing from a melt devoid of a normative CaAl2Si20e component and rocks with di and/or wo in their norms have com- positions that would permit the 'plagioclase effect' to operate (Table 2, Nos. IR, IG, IF). The norm of the glass and the sparse develop- ment of groundmass aenigmatite indicate the peralkaline character of the residuum in the Rocky Point phonolite but the level of normative peralkalinity, following the precipitation of 25% of highly aluminous felsic phenocrysts from an apparently an-free liquid, is actually relatively unimpressive. In contrast the host kenyte 2R (Fig. 4) initially contained 5.8 an but its vitric residuum, following the extraction of 20% anortho- clase (Or 1 7Ab6 3An2 0) , was completely purged of an and now contains 4.2% aa (Carmichael, 1964). The plagioclase effect in the Rocky Point phonolite was partly suppressed by the early separation of abundant sodic nepheline which was relatively more peralkaline (Na + K/Al = 0.89) than the anortho- clase of later crystallization (Na + K/Al = 0.68). Consequently it is unlikely that crystallization of felsic phases from strongly undersaturated salic melts will induce significant peralkalinity in the residua until the latter became essentially Ca-free (and Al-deficient) following the pre- cipitation of relatively abundant Ca-bearing alkali feldspar. In highly undersaturated salic volcanics this phase might ultimately dominate the felsic mineralogy but in this particular phonolite it is still largely occult in the vitric residuum (in part in di) . Pronounced peralkalinity will thus occur only at relatively evolved stages in the fractionation of highly undersaturated ae-free phonolitic liquids in which nepheline is a near- liquidus phase and it must follow somewhat pro- longed crystallization on or close to the respect- ive alkali feldspar-nepheline field boundaries. As with olivine nephel inite-phonolite associations, already discussed, it is debatable whether the low- pressure fractionated derivative liquid and crystal extract can be 'cleanly' separated at such an advanced stage of crystallization. The composition of the residual liquid at this advanced stage is unknown but consideration of the configuration of a three-phase triangle (assuming equilibrium crystallization) whose base (assuming limited nepheline enrichment in Ks) is the join IN-IR (bulk rock) - (more Or-rich alkali feldspar, indicated by anorthoclase zoning to more Or-rich rims) and whose apex is defined by a more evolved residuum IG' (on or close to the field boundary) suggests that IG will move towards minimum melting, increasingly peralkaline compositions IG' which would be relatively more potassic (in projection) than T^,, the experimentally determined minimum composition in the Ca-free system at PH2O = 1 kb. This postulated trend agrees with experimental data on the calcium-bearing system NaAlSi04- KAlSi0i,-CaAl2Si20e-Si02 at Ph,0 = 1 kb where the eutectic composition in the plane containing 3% An is now close to Ne4oKs3oQzso (Norris and Mackenzie, 1976). A corollary of this observation is that some Al may have been lost during hydration of the vitric groundmass in the host which, as a pristine vitrophyric obsidian, may have contained some normative an {of. Table 1, No. IR) , developed largely at the expense of di. OPERATIONS AT HIGH PRESSURES Following Kuno's classic (1964) account of high-pressure megacrysts of aluminous ortho- and clinopyroxene coexisting with Cr-diopside Iherzo- lite xenoliths in an "alkali olivine-basalt or hawaiite" (% = 8.8; normative plagioclase 100 aniab+an = An4i*) at Taka-sima, north Kyushu, an increasing number of moderately evolved 'inter- mediate' alkaline volcanics bearing the stamp of high-pressure (upper mantle) ancestry has been described. Most occurrences are from the Caino- zoic volcanic provinces in eastern Australia but overseas localities of high-pressure alkaline volcanics with 'evolved' characteristics are also known (e.g. Hutchison et at., 1975; Chapman, 1976). The volcanics contain Cr-diopside Iher- zolite xenoliths, which are frequently accompanied by various megacryst species of demonstrable high- pressure origin. The hosts range in composition from hawaiite and nepheline hawaiite (Wilkinson and Binns, 1969; Green and Hibberson, 1970; Green et at., 1974; Irving and Green, 1976; Ellis, 1976) through mugearite and nepheline mugearite (Green et at., 1974; Irving and Green, 1976) to nepheline benmoreite (Price and Green, 1972; Green et at., 1974). It may be commented that the Pigroot (North Otago) "mafic phonolite" (Price and Green, 1972) is difficult to classify according to the nomenclature of Coombs and Wilkinson (1969). Its normative plagioclase is highly sodic (Ans) but it has a D.I. (56.3) which is more characteristic of mugearitic lavas. The most evolved Iherzol ite-bearing alkaline volcanic so far reported appears to be the phonolite from the Bokkos plug, Nigeria (Wright, 1969; Irving 126 J. F. G. WILKINSON and Price, 1974). Complete chemical data on this phonolite apparently have yet to be published but it has a relatively high mg value (50.9) (MgO = 1.1, = 1.9) and is rich in alkalis (Na20 + K2O = 13.3). In addition to the examples noted above, over fifty Tasmanian occurrences of "fractionated Iherzolite-bearing alkaline rocks (with m^-numbers < 68)" have been recognized by Sutherland (1974) . Some of these will be examined subsequently in slightly more detail (Fig. 6). The origin of these relatively 'evolved' alkaline volcanics must be sought ultimately at pressures greater than 9-10 kb and hence genetic controls operated in a pressure regime(s) markedly different from that usually envisaged for the generation of liquids with comparable compositions. The latter have been widely interpreted as relatively low-pressure fractional crystallization derivatives of more mafic parents in sub-volcanic 'magma chambers'. However the genesis of former upper mantle residents has also been interpreted in more-or-less similar terms. That is, the pre- ferred model relates the origin of 'high-pressure' hawaiites and more evolved alkaline volcanic liquids to crystal fractionation under hydrous conditions, or, more specifically, to fraction- ation of 'wet' basanitic (7-8% H2O) or alkali basaltic parents at upper mantle pressures (15- 20 kb) dominated by kaersutitic amphibole, accompanied by minor but variable olivine and clinopyroxene (± biotite) (Irving and Green, 1972; Green et al., 1974; Irving and Green, 1976; Ellis, 1976; see Borley et al., 1971; Kesson and Price, 1972; Flower, 1973; Price and Chappell, 1975). High-Pressure Fractionation The viability of this model, as it might relate to specific 'evolved' volcanics from various Australian volcanic provinces, will be examined with particular reference to one of the best documented Australian volcanic provinces, namely the Upper Pliocene to Holocene Newer Volcanics (Newer Basalts) of Victoria and South Australia (Irving and Green, 1976). In addition to tholeiites this province contains a spectrum of alkaline volcanics ranging from alkali basalt to hawaiite, and basanite to nepheline mugearite. Many lavas within this compositional spectrum display evidence of their upper mantle ancestry (Irving and Green, 1976). The widespread basan- ites (the average M value for 25 Iherzolite- bearing examples = 65.6; Table 3) are mildly potassic (Na20/K20 = 1.86) and are associated - generally at separate eruptive centres - with nepheline hawaiites (average M = 63.6; Na20/K20 = 2.11) and nepheline mugearites (average M = 63.0; Na20/K20 = 1.96). These M values are based to a large extent on the Fe203/Fe0 ratios adopted by Irving and Green (1976), namely 0.20 for bas- anites (0.25 for Mount Leura basanites) , 0.20 for nepheline hawaiites, and 0.50 for nepheline mug- earites. All examples included in these averages contain Cr-diopside Iherzolite xenoliths {of. Frey and Green, 1974) and high-pressure megacrysts also occur at many localities (Irving, 1974b; Mass and Irving, 1976; Ellis, 1976). It seems most unlikely that crustal contamination has played a role in the genesis of the evolved members of the Newer Basalts because the ®^Sr/®®Sr ratios for 14 examples (of all major chemical types) are within the normal range for uncontaminated modem basalts (Irving and Green, 1976). Compared with their analogues from the type areas or from classical well-documented volcanic series elsewhere, the eastern Australian Iherzolite- bearing hawaiitic and mugearitic volcanics collect- ively display important compositional differences. More specifically, they are generally significantly richer in MgO, for a given D.I., than similar volcanics elsewhere (Fig. 5), reflected in the relatively high M values of the Victorian Newer Volcanic averages, already noted. On the basis of normative plagioclase compositions and different- iation indices these lavas are appropriately termed hawaiite, nepheline mugearite, etc. (Coombs and Wilkinson, 1969) but their MgO contents, at least to D.I. ~ 55, approach or may exceed the MgO contents of 'average' alkali olivine basalts (Fig. 5). For example, the average mugearite composition computed by Nockolds (1954) has a low M value (W = 39.3) and the type mugearite from Skye (Muir and Tilley, 1961) has M = 35.6, reflected in the relatively Fe-rich olivines and cl inopyroxenes in mugearites from other classic localities (Muir and Tilley, 1961). In contrast, olivine phenocrysts in the nepheline mugearite from Mt Anakie (East) are relatively Mg-rich (F073; Ellis, 1976). The relatively high M values of the Iherzolite-bearing volcanics in Figure 5 are at the same time accomp- anied by relatively high Na20 + K2O, oxides which largely determine the differentiation indices. There is also a tendency for these particular Iherzolite-bearing volcanics to be relatively sodic. In order to emphasize the evolved character of the Iherzolite-bearing lavas plotted in Figure 5, selection was restricted to examples (with the sole exception of the Auckland Island hawaiite; Green and Hibberson, 1970) with normative plagioclase compositions less calcic than An4o- Although relatively rich in MgO, the volcanics under dis- cussion display no obvious divergence from the 'generalised' trend illustrating variation in total Fe (expressed as FeO + Fe203), the Iherzo- lite-bearing types of intermediate D.I. plotting in a more-or-less random fashion with respect to the trend defined by average compositions of well- documented and type rocks (Fig. 5). The compositions of the average nepheline basanite, nepheline hawaiite, and nepheline mug- earite from the Newer Volcanics listed in Table 3 indicate the constraints, discussed below, on the character of crystal extracts composed of kaer- sutite, olivine and clinopyroxene. The most sign- ificant compositional parameters defining this moderately undersaturated lineage are M, normative plagioclase composition, and D.I. With increasing degree of 'evolution', M decreases only slightly, normative plagioclase becomes decidedly more sodic, total alkalis increase (Na20 + K2O increases from 6.0 to 8.0%) and the respective D.I.'s increase from 35.9 to 52.0 (Table 3). If anything, norm- ative nepheline decreases slightly, in contrast to similar lineages in some provinces which have been interpreted as the products of low-pressure fractionation of more mafic parents. The genesis of the more under saturated 'evolved' members of the Newer Volcanics via high- pressure fractionation of basanitic parents will PETROGENETIC ASPECTS OF SOME ALKALI VOLCANIC ROCKS 127 3,4 5,6,7 8 9 10 IU2 13 Average nepheline basanite B , ’ Newer Volcanics , Victoria • Hawaiite , nepheline hawaiite A Mugearite , nepheline mugearite ■ Nepheline benmoreite o Averages after Nockolds (1954) ® Averages after LeMaitre (1976) ° Hawaiian averages * Hebridean examples Fig. 5. Variation of MgO and FeO + Fe203, with differentiation index D.I., of evolved alkaline volcanics containing Cr-diopside Iherzolite xenoliths. The generalised curves indicate variation in MgO and FeO + Fez O3 in ('average') alkali basalt (AB) , hawaiite (H) , mugearite (M) , and benmoreite (B) . Hawaiian averages (open squares) are based on the data of Macdonald and Katsura (1964) and the Hebridean examples (crosses) refer to the analyses cited by Muir and Tilley (1961) and Tilley and Muir (1964). 1, hawaiite, Auckland Island (Green and Hibberson, 1970). 2, hawaiite, Mt Kurweeton, Victoria (Irving and Green, 1976). 3, nepheline hawaiite (average of 2), Mt Elephant, Victoria (Irving and Green, 1976). 4 and 5, nepheline hawaiites, Runnymede and Flinty March, Tasmania (Green et al. , 1974). 6, hawaiite, near Kyogle, New South Wales (Wilkinson and Binns, 1969). 7, hawaiite, Kingston, Tasmania (Green et al. , 1974). 8 and 9, nepheline mugearites from Mt Franklin (average of 2) and The Anakies (East) (average of 3), Victoria (Irving and Green, 1976). 10, nepheline mugearite (2102), The Anakies (East), Victoria (Irving and Green, 1972, 1976). 11, hawaiite, Redcliffe plateau, central Queensland (Green et al. , 1974). 12, 'mafic phonolite', northeast Otago (Price and Green, 1972). 13 and 14, nepheline mugearite, Mt Leslie, central Queensland, and nepheline benmoreite, Mt Mitchell, southeast Queensland (Green et al. , 1974). With the exception of the Auckland Island hawaiite (No. 1), all xenolith-bearing volcanics have 100 aniab+an < 40. The nomenclature of the volcanics plotted in Figs. 5 and 6 follows that adopted by the various authors. be examined semi-quantitatively as a function of the extraction of kaersutite alone, or kaersutite plus lesser olivine and clinopyroxene. The composition of the kaersutite extract (with total Fe expressed as FeO) has been based on the average composition of two kaersutite megacrysts (2110A and B8; Irving, 1974b) from the nepheline mugearite of The Anakies (East) (reflecting the most abundant compositional generation at this locality). The clinopyroxene composition is based on the average of two megacrysts from the nephel- ine mugearite at Mt Franklin (Irving, 1974b), and that of the olivine megacrysts (apparently unrecorded as yet in the Newer Volcanics) has been taken as mg = 86 (i.e. olivine of this composition has been assumed to be in equilibrium with liquids with M ~ 64). Because there is only very limited variation in the M values of members of the spectrum nepheline basanite nepheline mugearite. it has been assumed that there should be relatively little change in the mg values of ferromagnesian megacrysts precipitated from melts within this compositional range (it must be noted, however, that amphibole megacryst mg values do in fact vary sympathetically with host compositions; Ellis, 1976) . It is also recognized that successive crops of high-pressure phases will precipitate from liquids of changing composition and hence removal of crystal fractions with constant compositions (although a widely adopted practise in the literature) may not provide accurate indications of residua compositions. It is also assumed, despite the reservations of some workers (isotopic data are by no means conclusive on a possible accidental origin for high-pressure megacrysts in alkaline volcanics; see Stuckless and Irving, 1976) , that ferro- 128 J. F. G. WILKINSON TABLE 3 COMPOSITIONS OF MODERATELY UNDERSATURATED REPRESENTATIVES OF THE VICTORIAN NEWER VOLCANICS AND HYPOTHETICAL HIGH-PRESSURE DERIVATIVES Analysis number 1 2 3 4 5 6 7 SiOz 45.7 47.6 49.4 46.2 46.9 49.0 50.3 TiOa 2.8 2.4 2.2 2.8 2.8 2.3 2.2 AI2O3 12.9 13.5 15.6 13.3 13.7 14.5 15.2 FeaOs 2.0 1.9 3.5 1.9 1.8 3.4 3.1 FeO 10.1 9.5 7.0 9.6 9.2 6.8 6.2 MnO 0.2 0.2 0.2 0.2 0.2 0.2 0.2 MgO 10.8 9.3 6.7 10.0 9.0 7.2 5.6 CaO 8.6 8.9 6.7 8.6 8.5 8.9 8.9 Na20 3.9 4.0 5.3 4.1 4.4 4.5 4.9 K2O 2.1 1.9 2.7 2.3 2.4 2.2 2.3 P2O5 0.9 0.8 0.7 1.0 1.1 1.0 1.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Fe203/Fe0 0.20 0.20 0.50 0.20 0.20 0.50 0.50 100 Mg/Mg+Fe^'^ (AO 65.6 63.6 63.0 65.0 63.6 65.4 61.7 Na20/K20 1.86 2.11 1.96 1.78 1.83 2.05 2. 13 C. I, . P.W. norms Or 12.2 11.1 16.1 13.3 14.5 12.8 13.3 Ab 12.6 19.4 25.7 14.7 16.1 25.1 29.3 An 11.4 13.1 10.8 11.1 10.3 12.8 12.8 Ne 11.1 8.0 10.2 10.8 11.4 7.1 6.5 Di 21.3 20.7 14.3 20.2 19.8 20.1 19.2 01 21.1 18.5 11.9 19.4 17.5 10.4 7.4 Mt 3.0 2.8 5.1 2.8 2.6 4.9 4.4 11 5.3 4.6 4.1 5.3 5.3 4.4 4.3 Ap 2.0 2.0 1.7 2.4 2.7 2.4 2.7 Total 100.0 100.2 99.9 100.0 100.2 100.0 99.9 100 an/ab+an 47.5 40.3 29.6 43.0 39.0 33.8 30.4 D.I. 35.9 38.5 52.0 38.8 42.0 45.0 49.1 1 Average nepheline basanite. Newer Volcanics (25 analyses) . 2 Average nepheline hawaiite (7 analyses). 3 Average nepheline mugearite (6 analyses). 4 Derivative after extraction of 5% kaersutite, 2.5% olivine and 2.5% clinopyroxene (see text) from the average nepheline basanite (No. 1). 5 Derivative after extraction of 10% kaersutite, 5% olivine and 5% clinopyroxene from the average nepheline basanite (No. 1). 6 Derivative after extraction of 10% kaersutite, 5% olivine and 5% clinopyroxene from the average nepheline hawaiite (No. 2). 7 Derivative after extraction of 15% kaersutite, 7.5% olivine and 7.5% clinopyroxene from the average nepheline hawaiite (No. 2). Analyses 1-3 are based on analyses listed by Irving and Green (1976). PETROGENETIC ASPECTS OF SOME ALKALI VOLCANIC ROCKS 129 magnesian megacrysts are generally cognate with their hosts. For many megacryst occurrences this interpretation is consistent with appropriate high-pressure experimental data. Rejection of a strictly cognate origin of megacrysts virtually eliminates any opportunity to evaluate potential high-pressure fractionation trends quantitatively. So far as can be ascertained, there is no evidence of hybridism in megacryst-bearing volcanics, i.e. petrological evidence which would support an accidental relationship between megacryst and a particular host. On this premise the compositions of high-pressure nepheline hawaiites and nepheline mugearites plus their respective megacryst pop- ulations ideally should approximate the compos- itions of their less evolved, essentially un- fractionated parents, depending, of course, on megacryst abundances at specific localities. In evaluating compositional controls on derivative liquids following the extraction of high-pressure megacrysts, the proportions of the latter have been applied in only a generalised way. It is possible to provide an 'ideal' solution for a particular melt-derivative liquid combination but an approach of this type permits only res- tricted application to the much wider problem of high-pressure fractionation controls. A crystal extract composed solely of kaer- sutite is inappropriate because its removal results in increases in the M values of the derivatives, a result of kaersutite MgO (7% MgO, on an anhydrous basis) being less than MgO of the parent basanite. Olivine extraction is a highly efficient mechanism to decrease M values and hence for this particular lineage, in which M values remain more or less constant, olivine cannot be a major phase in the crystal extract. Crystallization of excessive high-pressure jadeitic K-free clinopyroxene would yield deriv- atives with decreased Na20/K20 ratios - a trend which is opposite to that displayed by the natural assemblages (Table 3). An extract composed of kaersutite, olivine and clinopyroxene in the somewhat arbitrary ratio of 2:1:1 has been selected to test the fraction- ation model semi-quantitatively. Removal of 10% of this extract from the average nepheline basan- ite (Table 3, No. 1) yields a derivative similar to the average nepheline hawaiite (Nos. 2 and 4) as indeed does the removal of 20% of this extract (Table 3, No. 5), which also results in a some- what excessive increase in alkalis. Extraction of 20% and 30%, respectively, of the kaersutite- olivine-clinopyroxene fraction from the average nepheline hawaiite - presumably the next stage in the high-pressure operation - yields derivative M values, D.I.'s and normative plagioclase com- positions that are essentially compatible with the average nepheline mugearite (Table 3, Nos. 3, 6, 7) but it does not produce an adequate decrease in CaO, and Na20 + K2O have not increased sufficiently to match the alkali contents of the natural nepheline mugearites. Furthermore, the derivatives now show a decided decrease in their ne contents, a result of the removal of signif- icant amounts of a highly undersaturated phase, namely kaersutite (normatively akin to olivine nephelinite) . Of more relevance, however, is the magnitude of the crystal extract (some 30-40 wt.%) required to pass from nepheline basanite to nepheline mug- earite - a natural extract that must have retained a more-or-less constant proportion of phases to account for the geochemical coherence, in terms of major and minor element chemistry, of nepheline mugearites at different eruptive centres. It now becomes appropriate to examine the nature of the megacryst assemblage at those centres in the Newer Volcanics where the lineage nepheline basanite nepheline mugearite is developed. High-pressure fractionation as the major genetic control, applied to the more 'evolved' undersaturated representatives of the Newer Volcan- ics, and indeed to similar volcanics elsewhere, becomes even more questionable when the general character of the megacryst assemblage at the Newer Volcanic localities and other eastern Australian alkaline provinces is examined. Of seventeen localities of Newer Volcanics where the represent- atives of the nepheline basanite, nepheline hawaiite, nepheline mugearite lineage occur, megacrysts have been recorded only at eleven localities. At these, black vitreous clinopyroxene is either the most abundant ferromagnesian mega- cryst (anorthoclase may be the most abundant mega- cryst species at some localities) or is the sole megacryst species at eight localities, kaersutite being essentially absent (Irving, 1974b; Mass and Irving, 1976) . Kaersutite is very abundant at only one locality, namely The Anakies (East) and it has been recorded as a 'very rare' megacryst at Mt Noorat. It may be noted that Ellis (1976, Table 2) records the abundance of amphibole megacrysts at The Anakies (East) as 'scarce'. Kaersutite is also a rare species in the nepheline hawaiite at Lake Keilumbete (Ellis, 1976). Olivine megacrysts apparently have yet to be recognised in the Newer Volcanics but olivine was clearly a high-pressure phase at some centres (e.g. Mt Leura, Mt Noorat and Mt Shadwell), indicated by the occurrence at these centres of cognate wehrlite inclusions belonging to the Ti-augite series. However at these particular centres Ti-augite inclusions are subordinate to Cr-diopside Iherzolite xenoliths by a factor of twenty or more (Irving, 1974a). The Tasmanian Iherzolite-bearing hawaiites, mugearites and benmoreites (and their more under- saturated equivalents) are also relatively rich in MgO relative to D.I. (Sutherland in Leaman, 1976; Sutherland, unpublished data), with a tendency for MgO to decrease somewhat more regularly with increasing D.I. than the Victorian and other examples (Figs. 5 and 6); FeO + Fe20s tends to remain more or less constant with increasing D.I. Unfortunately many analyses (Leaman, 1976, Table 2) display atypically high Fe203/Fe0 ratios but ten analyses (unpublished) have M values between 74 and 56 {M =68-49 when the analyses are corrected on a basis of Fe203/Fe0 = 0.15). The XENMEG catalogue (Wass and Irving, 1976) lists only one Tasmanian megacryst locality with kaersutite (abundance not stated), namely the nepheline hawaiite at Flinty Marsh (Green et al. , 1974). Indeed, megacrysts appear to be comparatively rare in Tasmanian Cainozoic alkaline volcanics. Three Queensland localities of Iherzolite-bearing 'evolved' alkaline volcanics have been listed by Green et al. (1974) . Kaersutite is abundant only in the Mt Mitchell nepheline benmoreite and 'sparse' kaersutite has been recorded in the Mt 130 J. F. G. WILKINSON • Hawaiite , nepheline hawaiite ^ Mugearite.nepheline mugearite ■ Benmoreite , nepheline benmoreite Fig. 6. Variation, with respect to differentiation index D.I., of MgO and FeO+FeaOs of evolved Tasmanian alkaline volcanics containing Cr-diopside Iherzolite xenoliths (Sutherland, unpublished data). Normative plagioclase compositions in all cases are less than 40. Generalised curves are those depicted in Fig. 5. Leslie nepheline mugearite where it is there subordinate to anorthoclase. The patchy and variable distribution and the aberrant proportions of megacryst species (in which zoning is by no means conspicuous) at these centres thus do not favour a model based largely on kaersutite fractionation. At more than fifty localities of Iherzolite-bearing 'evolved' alkaline volcanics in eastern Australia, kaer- sutite megacrysts are only sufficiently abundant at three (The Anakies (East), Mt Mitchell, and Spring Mount (Wilkinson, 1962; Binns et at. , 1969)) to suggest that amphibole fractionation might have exerted significant compositional controls in any derivatives. It could be argued that the aberrant character of megacryst popul- ations at the outcrop does not detract from the fractionation model because of pre-eruptive removal of (certain) megacrysts, either as a result of gravity or perhaps some form of flow differentiation. These would not explain the preferential removal of kaersutite (p ~ 3.2) relative to the more widespread and relatively more abundant megacrysts of cl inopyroxene (p ~ 3.4). Furthermore, kaersutitic amphibole is generally rare in Ti-augite wehrlite inclusions in the Newer Volcanics (Irving, 1974a) where it occurs largely as an intercumulus phase (Ellis, 1976). Finally, it must be noted that the relative frequency of occurrences of megacrysts in the Newer Volcanics (clinopyroxene > feldspars > kaersutitic amphibole > (olivine)) is the same as in the (dominantly) Cainozoic alkaline volcanic rocks of eastern Australia (Wass and Irving, 1976). Clinopyroxene is present at approximately 40% of megacryst localities in eastern Australia but kaersutitic amphibole occurs at only 11% of localities. Binns et al. (1969) provisionally concluded that high-pressure fractionation offers the possibility of only limited variation in derivative liquid compositions. So far as I am aware, very few data have appeared subsequently to materially change this conclusion and indeed it has been essentially confirmed by evaluation of potential high-pressure fractionation trends induced by megacryst precipitation at two north-eastern New South Wales localities (Lawlers Creek and Boomi Creek; Wilkinson, 1975a, 1975b). For example, removal of 20% olivine and 20% clinopyroxene from a hypothetical olivine nephelinite parent to yield the host olivine nephelinite at Lawlers Creek results in a moderate decrease in M values but the composition of the derivative is otherwise essent- ially unchanged (Wilkinson, 1975a). In view of the problems associated with the production of nepheline mugearite from basanite the possibility of deriving even more felsic (phonolitic and trachytic) liquids from parental mafic melts via high-pressure (amphibole) fractionation becomes even more remote. Whether such felsic liquids can be generated by partial melting in the upper mantle, as suggested by Wright (1971), is still uncertain and requires experimental examination. This proposal does, however, receive support from the isotopic data on certain northeastern Nigerian trachytes and phonolites (Grant et al. , 1972). PETROGENETIC ASPECTS OF SOME ALKALI VOLCANIC ROCKS 131 The formation of cognate ultramafic cumulates belonging to the Ti-augite inclusion group indicates gravitative movement and accumulation of ferromagnesian phases at elevated pressures (c/. Kushiro et al. , 1976) but once megacryst pre- cipitation during a period of magma stagnation has taken place at upper mantle pressures, pre- servation of megacrysts in their alkaline hosts (especially where megacrysts co-exist with Cr- diopside Iherzolite xenoliths) indicates very rapid ascent of liquid + megacrysts (+ xenoliths) to low-pressure eruptive regimes {of. Kushiro et al. , 1976; Biggar and Clarke, 1976) . It is thus a reasonable assumption that many megacryst populations (as now observed at the outcrop) are representative of the extent of high-pressure crystallization and the character of the high- pressure precipitate. One of the very few well -documented examples of fractionation at elevated pressures in which there is close correlation between field and experimental data has been provided by Knutson and Green (1975) who have demonstrated the potential production of mugearite from near- saturated hawaiite at deep crustal pressures (6.5-8 kb), following extraction of olivine, clinopyroxene, orthopyroxene, plagioclase and minor ilmenite and apatite (collectively equiv- alent to approximately 27% crystallization of the hawaiite parent). It should be noted, however, that plagioclase (generally absent from upper mantle megacryst populations) comprises 56% of the megacryst-cumulate assemblage, that Cr- diopside Iherzolites are absent from the mildly undersaturated alkaline volcanics in this partic- ular province, and that the parent hawaiite {M = 45.9) is itself interpreted as a fractionated derivative from transitional olivine basalt (with M = 65-75). A further example may be cited. The Al- spinel ultramafic-mafic inclusion suite at Boomi Creek, northwest of Barraba in northeastern New South Wales (Wilkinson, 1975b), has been inter- preted in the light of textural, mineralogical and chemical data as the remnants of a layered ultramafic-mafic 'pluton' which initially crystal- lized at pressures in the vicinity of 10 kb and subsequently re-equilibrated at subsolidus temp- eratures {oa 950°C) and comparable pressures. Fractionation at moderate pressures of parental K- poor subalkaline magma, controlled largely by extraction of olivine and subcalcic clinopyroxene, decreased the saturation levels of derivative liquids. This interpretation favouring moderate pressure fractionation is internally consistent with available experimental data but what is more important, it finds support in the major develop- ment of ultramafic cumulates, a prime component to the fractionation model. An Alternative High-Pressure Model Ferromagnesian fractionation has been invoked as the major control in the genesis of high- pressure mugearites, hawaiites, etc., mainly because their M values have been considered too low for these liquids to have been initially in equilibrium with parent peridotite with mg' ~ 90 - or, more specifically, pyrolite with M = 89.2. It has been proposed that Iherzol ite-bearing alkaline basalts with M ~ 70 represent unfractionated upper mantle liquids whose rapid movement to the surface inhibited ferromagnesian fractionation en route. In contradistinction, somewhat similar Iherzolite- bearing alkaline volcanics (with M - 60-68) have been interpreted as 'second order' derivatives - the end-products of later, superimposed high- pressure fractionation controlled (largely) by ferromagnesian phases. This proposal derives in part from the assumption that only basaltic liquids can be generated by partial melting in the upper mantle, an assumption that is, of course, implicit in the definition of pyrolite - "the composition of pyrolite is defined by the property that it is required to produce a basaltic magma upon partial melting, leaving behind a residual refractory peridotite" (Ringwood, 1975, p. 180). An additional implication in the proposal that liquids with M = 70±2 must have been in equilibrium with source peridotite containing olivines with mg = 88-90 (Green et al. , 1974) is the assumption that the upper mantle is essentially homogeneous in its M value. This is also open to question. Upper mantle peridotite heterogeneity is well illustrated by the data of Hutchison et al. (1975) which clearly indicate the variation between suites of Cr-diopside Iherzolites and associated ultramafic inclusions from 5 localities in the Massif Central, France. It is also indicated by the relatively Fe-rich character of several Cr-diopside Iherzolite xenoliths briefly reported below. If the major elements of Cr-diopside Iherzolite xenoliths are residual, following extraction of a basaltic component (up to 20% partial melting) (Frey and Green, 1974; Ringwood, 1975, pp. 177- 178; Irving, 1976) it is more-or-less axiomatic that the mg values of 'undepleted' source peridot- ites were less than those of their respective residua. Differences in the mg values of parents and residua will vary according to the degree of partial melting envisaged in the source peridotite. It is difficult to generalise on the composition of upper mantle peridotite xenoliths but the available data give the impression that the most commonly analysed representatives of the Cr-diopside Iherzolite xenolith group have mg = 90±2. If these xenoliths are residua, the mg values of many un- depleted peridotites must have been significantly less than 90, in accordance with experimental data indicating increases in mg values (at a given pressure) of residual phases in peridotite as a function of temperature increase (~ degree of partial melting). Thus olivine mg in residual garnet peridotite increases by about 2.5% with 15- 20% partial melting at 20 kb (Mysen and Kushiro, 1976) and by a similar amount over a temperature increase of 80-100°C (a comparable temperature increase induces 15-20% partial melting of garnet peridotite; Mysen and Kushiro, 1976) for water- saturated pyrolite (less 40% olivine) at 20 kb (Green, 1976) {see references in Wilkinson, 1976, p. 196). If K[) (the olivine-liquid Fe-Mg distribution coefficient) is indeed a function of fo2, T, P and the composition of the source material (Mysen, 1975) it becomes increasingly difficult to adopt Kp (and olivine-liquid Fe-Mg partition relation- ships) as a critical parameter in the recognition of unfractionated primary upper mantle liquids. If these complexities are disregarded for the time 132 J. F. G. WILKINSON being (_of. Green, 1976), an alternative but feasible interpretation can be applied to the histograms (analysis frequency vs. basalt M value) for 130 Iherzol ite-bearing volcanics from eastern Australia (Green et al. , 1974, Fig. 1). On adjusted (lower) FeaOa/FeO ratios (0.15-0.20) - a reasonable procedure because of the susceptibility of alkaline volcanics to alteration - approximately 70% of the specimens (W = 60-68) can be inter- preted (if ferromagnesian fractionation is dis- regarded) as partial melts initially in equilib- rium with olivine with mg = 83-88. Depending on the degree of partial melting, the latter would be more Mg-rich than olivines in the undepleted source peridotites. This interpretation assumes minimal high- pressure ferromagnesian fractionation of the partial melts in question. Since olivine is the most efficient phase in decreasing the M values of derivative liquids (olivine should occur on or near the liquid! of alkali basaltic melts at pressures up to 12-15 kb and at higher pressures for {?7-rich basanitic liquids) evidence in favour of high-pressure olivine fractionation should be provided by widespread and, at many localities, relatively abundant olivine megacrysts. In fact, olivine is a relatively rare megacryst species in eastern Australian alkaline volcanic provinces (Mass and Irving, 1976) and, so far as can be ascertained, in provinces elsewhere. Orthopyrox- ene is also an efficient megacryst phase in decreasing M values of derivative liquids but like olivine it is also a comparatively rare megacryst species (see Wilkinson and Binns, 1969). The high frequency of occurrence of cl inopyroxene megacrysts suggests that many alkaline liquids crystallized under conditions where cl inopyroxene was the major liquidus or near-liquidus phase at elevated pressures but relatively extensive pre- cipitation of clinopyroxene is necessary to induce any significant changes in the M values of any derivatives. Extraction of jadeitic clinopyroxene also inhibits enrichment in Na in derivative liquids. It is concluded that high- pressure ferromagnesian fractionation was gener- ally minimal for these particular xenolith- bearing lavas because magmas which transported Cr-diopside Iherzolite xenoliths rapidly to the surface were unlikely to have left olivine or pyroxene megacrysts behind. It is important to emphasize that olivine compositions in the range mg = 83-88 represent olivine compositions coexisting with melt. In view of the compositional variation of phases in residual peridotites as a function of the degree of partial melting (Mysen and Kushiro, 1976) these mg values must therefore be higher than those in the undepleted source peridotite - often by some 2-3% at least. Hence the mg values of many 'parental' olivines were thus most likely in the range mg = 81-86 before the onset of partial melting. The problem of generating liquids with M values less than 68 by partial melting of highly magnesian upper mantle source peridotites diminishes to a large extent if it is assumed that the primitive upper mantle in at least some regions of magma generation is relatively Fe-rich, compared with the compositions of the upper mantle source peridotites most commonly envisaged (.mg ~ 90) . This proposal has been discussed by Wilkinson and Binns (1978), with particular reference to the genesis of the extraordinarily voluminous contin- ental flood 'basalts' (often more appropriately termed tholeiitic andesites), but also tested with least squares calculations assuming a relatively Fe-rich source peridotite (mg = 83.7) and a derivative corresponding to the average hawaiite (mg = 53.1; Nockolds, 1954). The proposal is highly relevant to genetic considerations of the hawaiites and mugearites under discussion. Its validity obviously depends to a large extent on the documentation of xenoliths of the Cr-diopside Iherzolite group with M (or mg) values significantly less than 88. Some data on a number of these inclusions are now available. The most Fe-rich example is a Cr- diopside Iherzolite xenolith from an analcimite at Spring Mount, New South Wales (mg = 79; Wilkinson and Binns, 1978). The most commonly described type - admittedly apparently rare compared with relatively more Mg-rich Iherzolites - appear to have mg values between 85 and 87. These include two other Iherzolite xenoliths (mg ~ 85) from northeastern New South Wales localities (Wilkinson, 1975a, 1975b), five examples from Hawaiian alkaline hosts (mg = 85-87; White, 1966; Kuno and Aoki, 1970; Kushiro, 1973), one from Ogusoyama, south- western Japan (mg = 86; Ishibashi, 1970), and one from Tarreyres, Massif Central, France (mg = 86; Hutchison et al . , 1975). The mg values are based either on xenolith analyses or, more commonly, on analyses of their constituent olivines and pyroxenes - xenolith mg values are reasonably well defined by the mg values of their respective olivines and ortho- pyroxenes. For a given xenolith mg < M but when MgO is very high and pristine FeaOa of the xenolith is low (cf. Frey and Green, 1974) mg closely approximates to M. In the absence of other data specifically defining the affinities of a particular peridotite inclusion, phases from Cr- diopside xenoliths (and hence the affinities of the xenolith) generally may be discriminated from phases in Ti-augite ultramafic inclusions by the relatively higher Cr203 and lower TiOa contents of the clinopyroxenes in Cr-diopside Iherzolites (Wilkinson and Binns, 1978). Hutchison et al. (1975) interpreted the Fe- rich character of the Tarreyres xenolith (T33) as probably the result of "magmatic enrichment in Fe, Ti and other fusible oxides". This interpretation derives largely from the rather restrictive premise that undepleted upper mantle should yield only partial melts no more evolved than basalt. The chemistry of xenolith T33 can be equally well interpreted as a true reflection of its relatively undepleted character - AI2O3 and CaO are 4.00 and 4.38% respectively, and it may therefore be a potential parent to partial melts with M values of 60-63. There are apparently few, if any, published trace and minor element data of hawaiitic and mugearitic eruptives of high-pressure origin and of relatively Fe-rich Cr-diopside Iherzolites (mg = 80-87), thereby permitting more detailed chemical evaluations of 'mother-daughter' relationships. PETROGENETIC ASPECTS OF SOME ALKALI VOLCANIC ROCKS 133 If, however, it is assumed that the most Mg-rich olivine phenocrysts in relatively 'evolved' alkaline extrusives reflect the compositions of residual peridotitic olivines with which the partial melts were initially in equilibrium (c/. Green, 1976) it is possible to at least test the feasibility of their direct relation to comparat- ively Fe-rich Iherzolites, assuming minimal high- pressure ferromagnesian fractionation. For example, the nepheline benmoreite from Mt Mitchell in southeast Queensland {M = 53.1, D.I. = 69.7), selected by Green et at. (1974) as an example of a fractionated upper mantle liquid derived from basanite largely as the result of kaersutite extraction, contains olivine phenocrysts as Mg- rich as Fo63 and hence it could have been a partial melt initially in equilibrium with olivine with mg = 81-83. Olivine phenocrysts in alkaline volcanics from the eastern Azores (Boone and Fernandez, 1971) have compositions largely in the range Fosz-Foeo and there are other examples of 'basaltic' phenocrystal olivines in the Foao- Foee range {see Brown, 1967, Table III; Brown and Carmichael, 1969; Green and Hibberson, 1970). Olivine phenocrysts in Newer Volcanics nepheline basanites and nepheline hawaiites are as Mg-rich as Fo86-Fo63 (Ellis, 1976). High-pressure olivine megacrysts and olivines in cognate Ti- augite ultramafic inclusions are rarely more Mg- rich than mg = 88-86 and the most Mg-rich varieties in individual inclusion suites commonly extend to relatively more iron-rich compositions (Wilkinson, 1975a). If it is again assumed that ferromagnesian fractionation of the parent melts, prior to olivine megacryst precipitation, was minimal, the most Mg-rich liquidus olivine com- positions suggest that the host liquids were initially in equilibrium with residual peridotite with mg (maximum) = 88-86 {of. Carter, 1970). Detailed analytical data on olivine phenocrysts from a spectrum of evolved alkaline volcanics of suspected high-pressure origin are obviously required, due consideration being given to dis- criminating cognate phenocrysts from xenocrystal olivines. Accidental xenocrysts of olivine (and tschermakitic orthopyroxene) with mg = 80-87 would of course provide an indirect indication of the relatively Fe-rich character of upper mantle peridotites . The partial melting of relatively Fe-rich Iherzolites, as primitive components of the upper mantle, is suggested as an alternative to a model based on ferromagnesian fractionation in the generation of 'evolved' alkaline volcanics, at least to examples with M values between 55 and 65. It does not, however, clarify the problem of significant alkali enrichment in the partial melts, reflected in relatively sodic normative plagioclase compositions and intermediate dif- ferentiation index values. Enrichment in Na may reflect major compositional controls exerted by early entry of jadeitic clinopyroxene into the liquid (the clinopyroxene in the Spring Mount Iherzolite xenolith {mg = 79) contains 2.77 NaaO; Wilkinson and Binns, 1978). There is, of course, also the possibility that the source peridotites originally contained some sodic calciferous amphibole {of. Hariya and Terada, 1973) and/or phlogopite, phases which would preferentially enter derivative melts on relatively small degrees of partial melting of the source Iherzolites. At present there are some obvious constraints to a partial melting model based on relatively Fe- rich source peridotites. For example, there are at present only comparatively limited chemical and mineralogical data on 'undepleted', relatively Fe- rich Iherzolite xenoliths (M = 80-86) and hence there is an obvious need to investigate xenolith populations in alkaline hosts at the sampling level adopted by Hutchison et at. (1975). Furthermore, only one Fe-rich Iherzolite {mg = 85.3) has been subjected to high-pressure experimental studies (Kushiro, 1972). The partial melt (represented by clear, pale green glass essentially devoid of quench crystals) produced in this xenolith at 20 kb, 1460°C anhydrous has a basanitic aspect (ne = 10.0) but the liquid is Al-rich, and consequently its analysis yields a highly calcic normative plagio- clase. The M-value (based on Fe20a/Fe0 = 0.20) is 67; when Fe203/Fe0 = 0.15, M equals 66. The origin (s) of natural igneous assemblages may be constrained by experimental data but the most plausible genetic models are those in which the most fundamental characters of the natural assemblages (field, mineralogical and chemical) and relevant experimental data collectively provide a reasonable level of internal consistency. A comparatively simple single-stage partial melting model based on relatively Fe-rich source peridot- ites for the genesis of hawaiite, nepheline mug- earite, etc. - when these lavas are of demonstrable upper mantle origin - is offered as an alternative to a more complex multistage model requiring the initial generation of alkali basaltic or basanitic magmas, (or in some models, picritic), followed by what appear to be unduly excessive high-pressure crystal extracts which in general are not repres- ented in requisite amounts at the outcrop or in the proportions required by models based on ex- perimental data. CONCLUDING REMARKS Similar comments are also relevant in genetic interpretations of the alkaline volcanic spectrum in many continental and oceanic eruptive centres. We are here concerned with the genesis of alkaline eruptives with differentiation indices between 50 and 65 and not only with the trachyte-phonolite problem already highlighted by Chayes (1963). Although admittedly a difficult - in many instances, probably quite impracticable - procedure limited by poor outcrop and other constraints, very few studies detail or attempt to detail the relative volumes of the various eruptives and so relate these to a fractional crystallization model. This type of information has been provided for Saint Helena (Baker, 1968), although it may be commented that the "basalt" {M = 50.5, normative plagioclase An49, D.I. = 35.7) selected as parent to a con- tinuously variable volcanic series is closer in composition to hawaiite than to basalt. Of more concern, however, is the paucity of data on the complementary cumulates (either cognate inclusions or concentrations of single (zoned) crystals) required by the fractional crystalliz- ation model, particularly the relatively mafic assemblages necessary in the production of deriv- atives with intermediate differentiation indices. Borley (1974) has summarised the wide variety of inclusions in oceanic alkaline rocks. Although 134 J. F. G. WILKINSON inclusions may be abundant at some centres, such as Rodriguez Island (Upton et al. , 1967), they have apparently yet to be subjected to detailed and systematic phase mineralogical studies and evaluated as potential by-products of fraction- ation controls. Clearly, models based on fractional crystallization would assume much higher levels of internal consistency and con- viction if inclusion abundances and compositions were evaluated in this way. Are the cognate extracts actually present but ignored? If they are in fact absent this in itself demands some explanation if fractionation is adopted as the prime genetic control. In rebuttal, it could be argued that, because cumulates are rare or absent, they have remained unsampled at depth in the volcanic pile. This may well be so but it is also not unreasonable to assume that at some time(s) during the history of major volcanic centres, earlier cumulates should be sampled and brought to the surface by a rejuvenative phase(s) of volcanic activity. As an example, a detailed geochemical eval- uation of the mildly potassic Gough Island volcanic sequence led Zielinski and Frey (1970) to conclude that fractional crystallization of a basaltic parent involving the removal of olivine, pyroxene, feldspar and olivine was the prime petrogenetic control. Yet "all but one of the rocks considered were fine-grained with only a few phenocrysts" and "a criticism of the fractional crystallization model is that no cumulative rocks were obtained by Le Maitre except for the picrite basalt" (Zielenski and Frey, 1970, p. 246, p. 253). According to Le Maitre (1965), gabbroic xenoliths (ol ivine-cl inopyroxene-orthopyroxene- calcic plagioclase assemblages) occur abundantly in many of the basalt and trachybasalt flows, dykes and tuff horizons on Gough Island. He interpreted these xenoliths as fragments of upper mantle above the source area of the Gough magma. Finally, it must be noted that fractional crystal- lization models require what appear to be unduly large amounts of crystal extracts e.g. up to 40- 43% of various combinations of olivine, clino- pyroxene, kaersutite or spinel (e/. Brown and Carmichael, 1969; Gunn et dl. , 1970). In some studies it is assumed, more-or-less axiomatically, that a chemically and petrograph- ically related series of alkaline lavas comprises a 'fractionated series' and a fractionation model is then tested by subtracting various phases, in the 'right' proportions, from a parent which is ideally assumed to be basaltic. The parent com- position is most commonly adopted, partly because, I suspect, it has been more-or-less automatically assumed that only basaltic liquids are primary upper mantle derivatives. There is also a tendency to assume that ankaramites and oceanites, by virtue of prominent phenocrysts of clinopyroxene and olivine, represent the (low pressure) cumulates which are equated, at least in part, with the crystal extracts necessary to yield more evolved derivatives from some basaltic parent. This interpretation should be applied with caution. On the basis of 1 -atmosphere melting studies Thompson and Flower (1971) have indicated that certain phenocryst-rich ankaramites (MgO 10-12%, CaO 10.1-12.3%) from Anjouan, Comores Archipelago, may have been formed by prolonged in situ crystal- lization of initially all-liquid magmas, with no appreciable relative movement of early-formed phenocrysts and liquids. The genesis of alkali volcanic successions of extended composition, as commonly envisaged, thus evolves around a number of somewhat complicated, more-or-less 'ideal' processes. It commences with the partial fusion of (homogeneous) upper mantle peridotites and concomitant production of basaltic parents. If the most primitive partial melts are considered picritic (e/. Wilkinson, 1976) they are then required to shed olivine enroute to lower pressure regimes. The high-pressure extraction process whereby divorce occurs between partial melts and refractory residua must be extra- ordinarily efficient - at present there is appar- ently no evidence which demonstrates convincingly that a derivative partial melt has carried re- fractory residua of its parent to low-pressure regimes. The parent basalt melt than fractionates under relatively low-pressure conditions, crystal- lizing in the 'right' proportions phases which have differing crystal shapes, sizes and densities. The models often ignore phase relationships indicated by the experimentally determined crystal- lization behaviour of comparable melts, both at low and high pressures. The crystal extract is then largely or totally removed (with little or no evidence in the volcanic pile of its former presence) and so the production of the 'fraction- ated' series begins. On the evidence of a chemical discontinuity near the basalt-"andesite" transition in many anorogenic volcanic suites, Thompson (1972a) has argued that "most tholeiitic andesites, hawaiites and trachyandesites" are generated at upper mantle depths as a result of high-pressure partial crystallization of bodies of basaltic magma. Release of evolved liquid residua only occurs when the masses are approximately half solid and the process is essentially one based on advanced fractionation. Alternative petrogenetic models for evolved alkaline volcanics, particularly those in the hawaiite-benmoreite spectrum, might also include: (i) partial fusion of relatively Fe-rich upper mantle Iherzolites; and (ii) partial or (? complete) fusion of magmas already crystallized high in the upper mantle. Partial melting of a Snake River olivine tholeiite at 8 kb yields a tholeiitic andesite partial melt (Thompson, 1972b) and similar experimental data on alkali basaltic and related compositions are obviously desirable. At the same time, however, it must be commented that accidental high-pressure inclusions with compositions significantly more evolved than peridotitic (ultramafic) are apparently relatively rare in alkaline volcanics, if we are to judge from the data presently available on eastern Australian occurrences (Wass and Irving, 1976). It has not been a purpose of this Lecture to discredit models based essentially on fractional crystallization processes. Rather it represents an attempt to direct attention to specific areas where detailed interpretative data are meagre but obviously required. More specifically, a number of major problems will undoubtedly be clarified by detailed studies of inclusion populations from alkaline hosts. The current situation on alkaline volcanic genesis has, I believe, been succintly summarized by Carmichael et al. (1975, p. 426): "The petrologist of today is much better equipped PETROGENETIC ASPECTS OF SOME ALKALI VOLCANIC ROCKS 135 than his predecessors to assess quantitatively the chemical trends of fractional crystallization. Quantitative models have been set up for a number of oceanic lava series - St Helena, Gough, Thing- muli in Iceland, some of the Hawaiian volcanoes, Tenerife in the Canaries. Models, however, they remain - and open to competition with yet other models that fit the growing data equally well or better. Many of the chemical data on which they are based can equally well fit more complex models in which the final stage is partial or complete fusion of previously consolidated (possibly already differentiated) magma. In view of this and of uncertainties regarding the site and pressure-temperature conditions of differentiation, no model can establish with certainty the actual course of differentiation. It serves rather to illustrate the kinds of events and processes that might have participated in some unique line of evolution whose varied products are tangibly represented as rocks of a unique volcanic series". ACKNOWLEDGEMENTS I thank L. Sutherland for generously providing unpublished analyses of Tasmanian Iherzolite-bearing alkaline volcanics and H.D. Hensel for assistance with microprobe data. Research was supported by the University of New England. REFERENCES Allen, J.M. , 1974. Port Chalmers breccia and adjacent early flows of the Dunedin Volcanic Complex at Port Chalmers. N.Z. J. Geol. Geophys., 17, 209-223. Bailey, D.K., 1974. Nephelinites and ijolites, in THE ALKALINE ROCKS, pp. 53-66. H. Sorenson (Ed.). John Wiley and Sons, London. 622 pp. Bailey, D.K. and Schairer, J.F., 1966. The sys- tem Na20-Al203-Fe203-Si02 at 1 atmosphere, and the petrogenesis of alkaline rocks. J. Petrol., 7, 114-170. Baker, I., 1969. Petrology of the volcanic rocks of Saint Helena Island, South Atlantic. Bull. Geol. Soo. Am., 80, 1283-1310. Biggar, G.M. and Clarke, D.B., 1976. Reaction rate of xenocrysts in synthetic basalt melts. Nat. Envir. Res. Counail Pub. Ser. D, 6, 227-230. Binns, R.A., Duggan, M.B. and Wilkinson, J.F.G., 1969. High-pressure megacrysts in alkaline lavas from northeastern New South Wales. Am. J. Soi., 269, 132-168. Boone, G.M. and Fernandez, L.A., 1971. Pheno- crystic olivines from the eastern Azores. Mineral. Mag., SB, 165-178. Borley, G.D., 1974. Alkaline provinces - oceanic islands, in THE ALKALINE ROCKS, pp. 311-330. H. SjSrenson (Ed.). John Wiley and Sons, London. 622 pp. Borley, G.D., Suddaby, P. and Scott, P., 1971. Some xenoliths from the alkalic rocks of Teneriffe, Canary Islands. Contrib. Mineral. Petrol., 31, 102-114. Bowen, N.L., 1945. Phase equilibria bearing on the origin and differentiation of alkaline rocks. Am. J. Soi., 243A, 75-89. Brown, F.H., 1970. Zoning in some volcanic nephelines. Am. Mineral., 55, 1670-1680. Brown, F.H. and Carmichael, I.S.E., 1969. Quater- nary volcanoes of the Lake Rudolf region: 1. The basanite-tephrite series of the Korath Range. Lithos, 2, 239-260. Brown, G.M., 1967. Mineralogy of basaltic rocks, in BASALTS: THE POLDERVAART TREATISE ON ROCKS OF BASALTIC COMPOSITION, VOL. 1, pp. 103-162. H.H. Hess (Ed.). Interscience Publishers, New York. 482 pp. Carmichael, I.S.E., 1964. Natural liquids and the phonolitic minimum. Geol. J. (Liverpool), 4, 55-60. Carmichael, I.S.E., Turner, F.J. and Verhoogen, J., 1975. IGNEOUS PETROLOGY. McGraw-Hill Book Company, New York. 739 pp. Carter, J.L., 1970. Mineralogy and petrology of the earth's upper mantle based on the partial fusion - partial crystallization model. Bull. Geol. Soo. Am., 81, 2021-2034. Chapman, N. , 1976. Inclusions and megacrysts from undersaturated tuffs and basanites, east Fife, Scotland. J. Petrol., 17, 472-498. Chayes, F., 1963. Relative abundance of inter- mediate members of the oceanic basalt- trachyte association. J. Geophys. Res., 68, 1519-1534. Coombs, D.S. and Wilkinson, J.F.G., 1969. Lineages and fractionation trends in undersaturated volcanic rocks from the East Otago Volcanic Province (New Zealand) and related types. J. Petrol., 10, 440-501. Duggan, N.T., 1972. Tertiary volcanic rocks of the Inverell area, northeastern New South Wales. B.Sc. (Hons.) Thesis, Univ. New England. (Unpubl . ) Ellis, D.J., 1976. High-pressure cognate inclusions in the Newer Volcanics of Victoria. Contrib. Mineral. Petrol., 58, 149-180. Flower, M.F.J., 1973. Evolution of basaltic and differentiated lavas from Anjouan, Comores Archipelago. Contrib. Mineral. Petrol., 38, 237-260. Frey, F.A. and Green, D.H., 1974. The mineralogy, geochemistry and origin of Iherzolite inclusions in Victorian basanites. Geochim. Cosmoahim. Acta, 38, 1023-1059. 136 J. F. G. WILKINSON Grant, N.K., Rex, D.C. and Freeth, S.J., 1972. Potassium-argon ages and strontium isotope ratio measurements from volcanic rocks in northeastern Nigeria. Contrib. Mineral. Petrol., 35, 277-292. Green, D.H., 1976. Experimental testing of 'equilibrium' partial melting of peridotite under water-saturated, high-pressure con- ditions. Canad. Mineral., 14, 255-268. Green, D.H., Edgar, A.D., Beasley, P., Kiss, E. and '.'/are, N.G., 1974. Upper mantle source for some hawaiites, mugearites and benmore- ites. Contrib. Mineral. Petrol., 48, 33-43. Green, D.H. and Hibberson, W. , 1970. Experimental duplication of conditions of precipitation of high-pressure phenocrysts in a basaltic magma. Phys. Earth Planet. Interiors, 3, 247-254. Gunn, B.M., Coy-Yll, R., Watkins, N.D., Abranson, C.E. and Nougier, J. , 1970. Geochemistry of an oceanite-ankaramite-basalt suite from East Island, Crozet Archipelago. Contrib. Mineral. Petrol., 28, 319-339. Hamilton, D.L., 1961. Nephelines as crystal- lization temperature indicators. J, Geol. , 69, 321-329. Hamilton, D.L. and Mackenzie, W.S., 1965. Phase- equilibrium studies in the system NaAlSiOi, (nepheline) -KAlSi04 (kalsilite) -Si02-H20. Mineral. Mag., 34, 214-231. Hariya, Y. and Terada, S., 1973. Stability of richterites o-tremol ites 0 solid solution at high pressures and possible presence of sodium calcic amphibole under upper mantle conditions. Earth Planet. Sci. Letters, 18, 72-76. Hutchison, R. , Chambers, A.L., Paul, D.K. and Harris, P.G., 1975. Chemical variation among French ultramafic xenoliths - evidence of a heterogeneous upper mantle. Mineral. Mag., 40, 153-170. Irving, A.J. , 1974a. Pyroxene-rich ultramafic xenoliths in the Newer Basalts of Victoria, Australia. Neues Jahrb. Mineral. Abhand. , 120, 147-167. Irving, A.J. , 1974b. Megacrysts from the Newer Basalts and other basaltic rocks of south- eastern Australia. Bull. Geol. Soa. Am., 85, 1503-1514. Irving, A.J., 1976. On the validity of paleo- geotherms determined from xenolith suites in basalts and kimberlites. Am. Mineral., SI, 638-642. Irving, A.J. and Green, D.H., 1972. Experimental study of phase relationships in a high- pressure mugearitic basalt as a function of water content. Abstracts with Programs, Geol. Soo. Am. Mtgs. , 4, 550-551. Irving, A.J. and Green, D.H., 1976. Geochemistry and petrogenesis of the Newer Basalts of Victoria and South Australia. J. Geol. Soa. Australia, 23, 45-66. Irving, A.J. and Price, R.C. , 1974. Mantle-derived phonolitic rocks from Nigeria, Eastern Germany and New Zealand (abstr.). EOS, Trans. Am. Geophys. Un. , 55, 487. Ishibashi, K. , 1970. Petrochemical study of basic and ultrabasic inclusions in basaltic rocks from northern Kyushu, Japan. Mem. Fac. Sci. Kyushu Univ. , Ser. D., 20, 85-146. Kesson, S. and Price, R.C., 1972. The major and trace element chemistry of kaersutite and its bearing on the petrogenesis of alkaline rocks. Contrib. Mineral. Petrol., 35, 119-124. King, B.C., 1965. Petrogenesis of the alkaline igneous rock suites of the volcanic and intrusive centres of eastern Uganda. J. Petrol., 6, 67-100. Knutson, J. and Green, T.H., 1975. Experimental duplication of a high-pressure megacryst/ cumulate assemblage in a near-saturated hawaiite. Contrib. Mineral. Petrol., 52, 121-132. Kuno, H. , 1964. Aluminian augite and bronzite in alkali olivine basalt from Taka-sima, northern Kyushu, Japan, in ADVANCING FRONTIERS IN GEOLOGY AND GEOPHYSICS, pp. 205-220. Osmania University Press, Hyderabad. Kuno, H. and Aoki, K., 1970. Chemistry of ultra- mafic nodules and their bearing on the origin of basaltic magmas. Phys. Earth Planet. Interiors, 3, 273-301. Kushiro, I., 1972. Partial melting of synthetic and natural peridotites at high pressures. Ann. Rept. Geophys. Lab. (Carnegie Inst. Yrbk. 71), pp. 357-362. Kushiro, I., 1973. Origin of some magmas in oceanic and circum-oceanic regions. Tectono- physios, 17, 211-222. Kushiro, I., Yoder, H.S. and Mysen, B.O., 1976. Viscosity of basaltic and andesitic liquids at high pressures. Ann. Rept. Geophys. Lab. (Carnegie Inst. Yrbk. 75), 614-618. Leaman, D.E., 1976. Hobart. Tasmania Dept. Mines, Geol, Surv. Rept., 1:50,000 Series. Le Maitre, R.W., 1965. The significance of the gabbroic xenoliths from Gough Island, South Atlantic. Mineral. Mag., 34, 303-317. Le Maitre, R.W., 1976. The chemical variability of some common igneous rocks. J. Petrol., 17, 589-637. Macdonald, G.A. and Katsura, T. , 1964. Chemical composition of Hawaiian lavas. J. Petrol., 5, 82-133. PETROGENETIC ASPECTS OF SOME ALKALI VOLCANIC ROCKS 137 McDougall, I. and Wilkinson, J.F.G., 1967. Potassium-argon dates on some Cainozoic volcanic rocks from northeastern New South Wales. J. Geol. Soc. Australia, 14, 225-234. Morse, S.A. , 1969. Syenites. Ann. Rept, Geophys. Lab. (Carnegie Inst. Yrbk. 67), 112-120. Muir, I.D. and Tilley, C.E., 1961. Mugearites and their place in alkali igneous rock series. J. Geol., 69, 186-203. Mysen, B.O., 1975. Partitioning of iron and mag- nesium between crystals and partial melts in peridotite upper mantle. Contrib. Mineral. Petrol., 52, 69-76. Mysen, B.O. and Kushiro, I., 1976. Compositional variation of coexisting phases with degree of melting of peridotite under upper mantle conditions. Ann. Rept. Geophys. Lab. (Carnegie Inst. Yrbk. 75), 546-555. Nash, W.P., Carmichael, l.S.E. and Johnson, R.W., 1969. The mineralogy and petrology of Mount Suswa, Kenya. J. Petrol., 10, 409-439. Nash, W.P. and Wilkinson, J.F.G., 1970. Shonkin Sag laccolith, Montana: I. Mafic minerals and estimates of temperature, pressure, oxygen fugacity and silica activity. Contrib. Mineral. Petrol., 25, 241-269. Nockolds, S.R., 1954. Average chemical compos- itions of some igneous rocks. Bull. Geol. Soc. Am., 65, 1007-1032. Norris, G. and Mackenzie, W.A. , 1976. Phase relations in the system NaAlSi0i,-KAlSi0i,- CaAl2Si20e-Si02 at PhjO “ ^ 5)nvir. Res. Council Pub. Ser. D, 6, 79-81. Price, R.C. and Chappell, B.W. , 1975. Fractional crystallization and the petrology of the Dunedin volcano. Contrib. Mineral. Petrol., 53, 157-182. Price, R.C. and Coombs, D.S., 1975. Phonolitic lava domes and other features of the Dunedin volcano. East Otago. J. R. Soc. N.Z., 5, 133-152. Price, R.C. and Green, D.H. , 1972. Lherzolite nodules in a "mafic phonolite" from north- east Otago, New Zealand. Nature, 235, 133- 134. Ringwood, A.E. , 1975. COMPOSITION AND PETROLOGY OF THE EARTH'S MANTLE. McGraw-Hill Book Company, New York, 618 pp. Saggerson, E.P. and Williams, L.A. J. , 1964. Ngurumanite from southern Kenya and its bearing on the origin of rocks in the northern Tanganyika alkaline district. J. Petrol., 5, 40-81. Schairer, J.F., 1957. Melting relations of the common rock-forming oxides. J. Am. Ceram. Soc., 40, 215-235. Schairer, J.F. and Yoder, H.S., 1960. The nature of residual liquids from crystallization, with data on the system nepheline-diopside-silica. Am. J. Sci., 258-A, 273-283. Spencer, A.B., 1969. Alkalic igneous rocks of the Balcones province, Texas. J. Petrol., 10, 272-306. Stuckless, J.S. and Irving, A.J. , 1976. Strontium isotope geochemistry of megacrysts and host basalts from southeastern Australia. Geochim. Cosmochim. Acta, 40, 209-213. Sutherland, L.S., 1974. High-pressure inclusions in tholeiitic basalt and the range of Iherzol ite-bearing magmas in the Tasmanian volcanic province. Earth Planet. Sci. Letters, 24, 317-324. Tilley, C.E., 1959. A note on the nosean phonolite of the Wolf Rock, Cornwall. Geol. Mag., 96, 503-504. Tilley, C.E. and Muir, I.D., 1964. Intermediate members of the oceanic basalt-trachyte association. Geol. For. Stockh. Fork., 85, 434-443. Tilley, C.E. and Thompson, R.N., 1972. Melting relations of some ultra alkali volcanics. Geol. J. (Liverpool), 8, 65-70. Thompson, R.N., 1972a. Evidence for a chemical discontinuity near the basalt-andesite transition in many anorogenic volcanic suites. Nature, 236, 106-110. Thompson, R.N., 1972b. Melting behaviour of two Snake River lavas at pressures up to 35 kb. Ann. Rept. Geophys. Lab. (Carnegie Inst. Yrbk. 71), 406-410. Thompson, R.N., 1973. One-atmosphere melting behaviour and nomenclature of terrestrial lavas. Contrib. Mineral. Petrol., 41, 197- 204. Thompson, R.N. and Flower, M.F.J., 1971. One- atmosphere melting and crystallization relations of lavas from Anjouan, Comores Archipelago, western Indian Ocean. Earth Planet. Sci. Letters, 12, 97-107. Troger, W.E., 1935. SPEZIELLE PETROGRAPHIE DER ERUPTIVGESTEINE. Berlin. 360 pp. Upton, B.G.J., Wadsworth, W.J. and Newman, T.C., 1967. The petrology of Rodriguez Island, Indian Ocean. Bull. Geol. Soc. Am., 78, 1495-1506. Wass, S.Y. and Irving, A.J., 1976. XENMEG - A catalogue of occurrences of xenoliths and megacrysts in basic volcanic rocks of eastern Australia. Australian Museum, Sydney. 441 pp. Wellman, P. and McDougall, I., 1974. Potassium- argon ages on the Cainozoic volcanic rocks of New South Wales. J. Geol. Soc. Australia, 21. 247-272. 138 J. F. G. WILKINSON Wilkinson, J.F.G., 1962. Mineralogical , geochem- ical and petrogenetic aspects of an analcite- basalt from the New England district of New South Wales. J. Petrol., 3, 192-214. Wilkinson, J.F.G., 1966. Residual glasses from some alkali basaltic lavas from New South Wales. Mineral. Mag., 35, 847-860. Wilkinson, J.F.G., 1975a. Ultramafic inclusions and high-pressure megacrysts from a nephel- inite sill, Nandewar Mountains, north-eastern New South Wales, and their bearing on the origin of certain ultramafic inclusions in alkaline volcanic rocks. Contrib. Mineral. Petrol., 51, 235-262. Wilkinson, J.F.G., 1975b. An Al-spinel ultramafic- mafic inclusion suite and high pressure mega- crysts in an analciraite and their bearing on basaltic magma fractionation at elevated pressures. Contrib. Mineral. Petrol., S3, 71-104. Wilkinson, J.F.G., 1976. Some subcalcic clino- pyroxenites from Salt Lake Crater, Oahu, and their petrogenetic significance. Contrib. Mineral. Petrol., 58, 181-201. Wilkinson, J.F.G. and Binns, R.A., 1969. Hawaiite of high-pressure origin from north-eastern New South Wales. Nature, 222, 553-555. Wilkinson, J.F.G. and Binns, R.A., 1978. Relat- ively iron-rich Iherzolite xenoliths of the Cr-diopside suite: A guide to the primary nature of anorogenic tholeiitic andesite magmas. Contrib. Mineral. Petrol, (in press) Wilkinson, J.F.G. and Duggan, N.T. , 1973. Some tholeiites from the Inverell area. New South Wales, and their bearing on low-pressure tholeiite fractionation. J. Petrol., 14, 339-348. White, R.W., 1966. Ultramafic inclusions in basaltic rocks from Hawaii. Contrib. Mineral Petrol., 12, 24 5-314. Wright, J.B., 1963. A note on possible different- iation trends in Tertiary to Recent lavas of Kenya. Geol. Mag., 100, 164-180. Wright, J.B., 1969. Olivine nodules and related inclusions in trachyte from the Jos Plateau, Nigeria. Mineral. Mag., 3?, 370-374. Wright, J.B., 1971.- The phonolite-trachyte spectrum. Lithos, 4, 1-5. Zielenski, R.A. and Frey, F.A., 1970. Gough Island: evaluation of a fractional crystal- lization model. Contrib. Mineral. Petrol., 29, 242-254. Department of Geology, University of New England, ARMIDALE, N.S.W. 2351 (Manuscript received 18.8.1977) Journal and Proceedings, Royal Society of New South Wales, Vol. 110, pp. 139-145, 1977 Fossil Marsupials from the Douglas Cave, Near Stuart Town, New South Wales John D. Gorier ABSTRACT. Marsupial remains from the Douglas Cave include Thylacinus cynocephalus , Dasyurus viverrinus , Sminthopsis sp. , Antechinus sp. , Phascogale tapoatafa, Perameles nasuta, Isoodon obesulus , Cercartetus nanus, ?Acrobates sp. , Lasiorbinus sp. , Potorous tridactylus , Thylogale stigmatica, Wallabia bicolor. Macropus sp. , M. agilia, M. giganteus , M. titan and M. altus. A radiocarbon date of 29,200 years was obtained from near the base (Frank, 1969) of the older cave earth unit. INTRODUCTION Douglas Cave was discovered by 1896 by a Mr. R.J. Wilson (Leigh, 1897) in a limestone outcrop on Portion 93, Cooper Parish, in the County of Wellington about 8 km southwest of Stuart Town, central western N.S.W. (Figure 1). Leigh (1897) reported that a passage off the main chamber, the Bone Cave, was "literally packed with the fossil- ised bones of both herbivorous and carnivorous marsupials". A collection of bone was forwarded to the then Government Palaeontologist, W.S. Dun, who identified Thylacinus spelaeus Owen, Dasyurus sp. , and Macropus sp. (Dun, 1897). The next investigation was by Frank (1969), who mapped the cave, which he called the Douglas Cave, and described its clastic sediments and probable history. Animal remains were collected from pits dug by Frank into the two cave earth units in Douglas Cave, and were lodged at the Bureau of Mineral Resources, Canberra. The Initial description of this material was as part of the author's unpublished Honours thesis at the Australian National University. CAVE STRATIGRAPHY Frank (1969) recognised two cave earth units and sank a pit into each of these. Pit I in the older cave earth unit (Unit 1) reached a depth of 2.4 metres but was not excavated to bedrock. A radiocarbon date of 29 200 ±^888 " 1555 was obtained from charcoal near the base of the pit (Frank, 1969). Dates of 24 100 ±700 Gx - 1476 (Pit II) and 26 200 ±^88 Gx 1477 (Pit I) were obtained from bone at the top of cave earth Unit I (Frank, 1972). As dates obtained from bones are probably underestimates, deposition of Unit 1 may have been rapid. A further date from bone near the base of Unit 1 in Pit I gave a date of 26 100 ±1888 Gx - 1478 (Frank, 1972). Frank (1969) showed that Unit 1 was deposited through a solution pipe that opened some 30 000 years ago, and is probably about 2.5 metres thick. Bones* were collected down to 1.8 metres. After deposition of Unit 1 ceased, flowstone was deposit- ed and forms the base of Unit 2. The second period of deposition was initiated by the opening of the present entrance as a collapse doline. * in the Bureau of Mineral Resources possession FAUNA The distribution of the marsupial fauna from the two cave units is shown in Tables 1 and 2. Apart from marsupial remains, the cave earth units contain at least two species of lizard and several types of rodents which are at present undetermined. PALAEONTOLOGY Family Macropodidae Subfamily Macropodinae Genus Macropus Shaw, 1790 Macropus giganteus Shaw, 1790 Only one specimen from the Douglas Cave, DC4 from the top 15 cm of Pit I, is assigned to Macropus giganteus. The specimen, a right mandibular ramus, is of a juvenile with a partly shed DP_, and erupted M. , and a semi-erupted M (Plate 2, figures 4, 5). Comparison with measurements from Tedford (1967) and Bartholomai (1971) are shown in Table 4. Genus Macropus Shaw, 1790 Macropus titan Owen, 1838 Two specimens (DCl and DC8) are referred to the extinct giant kangaroo Macropus titan. Both occur in the 30-60 cm level of Pit I and represent two individuals. The specimens are both mandibular fragments containing molar teeth (Plate 1, figures 1,2,3), and compare closely with characteristics of M. titan as described by Bartholomai (1975). Measurements of the molar teeth are given in Table 3 together with measurements of M. titan given by Marshall (1974) and Tedford (1967). Length is plotted against posterior width in Figure 2 for the third lower molar of the Douglas Cave specimens, following the format of Marshall (1974). Both specimens lie towards the centre of the Macropus titan field and so are probably typical members of that species. DISCUSSION Lack of Larger Marsupial Remains Comparison of the Douglas Cave fauna with other similar cave faunas indicates a lack of the larger marsupials: the diprotodontids and sthenurine kangaroos in particular. This omission 140 JOHN D. GORTER Figure 1. Locality map of the Douglas Cave. from the fauna is surprising in view of the dates (24 - 29 x 10^ years B.P.), which place the older deposits well within the known stratigraphic range of the larger marsupials. The terrain around the cave is limstone karst, and such an environment may have proved unsuitable for animals as large as diprotodontids . As the cave opening is located high on a low rise with no higher ground in the vicinity, large remains would have been unlikely to have been washed in through the solution pipe described by Frank (1969), through which Unit 1 was deposited. From the fossil distribution (Table 1) it can be seen that no large marsupials are present in the Pit I deposit until the 60-120 cm level and that they become common only in the higher levels. This may reflect widening of the solution pipe, allowing larger macropods to fall in. Large marsupials are similarly absent from the lowest level of Pit 11 but make their appearance at the 112 cm level. This may reflect the development of the collapse doline through which Unit 2 was deposited. In conclusion, the absence of the larger marsupials from the deposit is thought to be caused by such factors as mode of deposition of the cave earth units and the geographical location of the cave, rather than extinction of larger marsupials before the case was opened. On the Relationships between Macropus titan and M. giganteus The only published occurrence of Macropus titan and M. giganteus in the same deposit is from Mammoth Cave, Western Australia (Tedford, 1967). However, Merrilees (1968) revised the Western Australian faunal list, and identified the Mammoth Cave Macropus as M. fuliginosus , not M. giganteus . Bartholomai (1975) further suggests that the occurrence of M. titan in Western Australia should also be checked. Thus the association of the two forms in Douglas Cave has relevance to the problem of the phylogenetic relationship of M. titan to M. giganteus . Tedford (1967) said that M. gigant- eus is "not readily derivable" from M. titan: whilst their molar characteristics are basically similar, those of titan are said to be slightly more complicated (Anderson, 1929). Marshall (1973) considers M. titan to be an example of Pleistocene gigantism, with M. giganteus as the surviving smaller form and recently Bartholomai (1975) concluded that separation of the two forms at the specific level is justified on size differential and various morphological grounds, and that "close relationship with the recent M. giganteus appears highly likely" (p. 205). Marshall (1974) placed the date of transition between the two forms at about 20 000 years B.P. Discoveries of M. titan at Lancefield, Victoria, in swamp deposits dated as young as 12 000 years B.P. (D. Horton, personal communication 1976) put Marshall's estimate in question. However, Wright (1975) reports preliminary dates of about 12 500 B.P. on collagen and about 16 000 B.P. on apatite from bones recovered at Lancefield. Wright suggests these dates should be treated as minima because of the chemical environment of the swamp, and postulates that the bones are probably considerably older than 16 000 years. Flood (1974) referred macropodid material from Cloggs Cave, Victoria from deposits dated at 22 980 +2 000 B.P. (ANU-1220), to M. titan. J. Hope (written communication, 1976) has re-examined this material and considers that the macropodid is referable to M. giganteus. Thus the oldest previously known Macropus giganteus is about 23 000 B.P., while the yougest M. titan is probably considered older than the nominal 16 000 FOSSIL MARSUPIALS 141 TABLE 1 MINIMUM NUMBERS OF INDIVIDUALS FROM SAMPLED LEVELS IN CAVE EARTH UNIT 1, PIT I Species Bone Room Floor Top 15 cm 30-60 cm 60-120 cm 120-180 cm Thylacinus cynocephalus 1 _ _ Lasiorhinus sp. - 1 - - - Dasyurus viverrinus - - - - 1 Sminthopsis sp. - - - 3 4 Antechimis sp. - - - 4 7 Phascogale tapoatafa - - - 1 - Cercartetus nanus - - - 3 1 ^Acrobates sp. - - - 1 1 Perameles nasuta - - - 4 5 Isoodon obesulus - - - 2 5 Potorous tridactylus - - - 1 - Thylogale stigmatica - 1 - - - Wallabia bicolor - 1 - - - Macropus sp. Present Present Present Present - M. agilis - 1 - - - M. giganteus - 1 - - - M. titan - - 2 - - M. altus - 4 1 1 TABLE 2 MINIMUM NUMBERS OF INDIVIDUALS FROM SAMPLED LEVELS IN CAVE EARTH UNIT 2, PIT II Species Top 70 cm 112 cm 120 cm Bottom 30 cm 30 cm Sminthopsis sp. 2 Antechinus sp. 1 Dasyurus viverrinus 1 Perameles nasuta 1 Isoodon obesulus Cercartetus nanus 1 MacTopus altus 1 2 I 1 I 2 r 142 JOHN D. GORTER TABLE 3 DIMENSIONS OF LOWER TEETH OF MACROPUS TITAN (DC 1, DC 8) IN MILLIMETRES. L = Length, AW = Anterior width, PW = Posterior width. , measured across the lophids. Specimen P3 Ml M2 "3 M4 DC 1 L . . ca 16 17.1 19 AW - - 9.6 10.6 10.6 PW 10.4 10.5 10.4 DC 8 L 7.9 . 14 16.5 AW - - 10.2 10.5 - PW ■ 9.25 10 10 - Marshall L 13-14 13.8-16 15.5-17.4 17-19 (1974) AW - 7.7-8 8-9.7 9.5-11.6 9.9-11.6 PW - 8. 2-8. 7 9.5-10.4 9.2-11 9.3-11.2 Tedford L . 15.5-19 16. 1-20.2 (1967) AW - - - 9.2-11.2 9.4-11.6 titan PW “ “ 8.5-10.5 “ 8.9-11.2 TABLE 4 DIMENSIONS OF LOWER CHEEK TEETH OF SPECIMENS ASSIGNED TO MACROPUS GIGANTEUS IN MILLIMETRES Specimen “P3 Ml M2 »3 M4 DC 4 L 8.2 10.7 12-13 . AW 5.6 6.6 7.0 - - PW 6.0 6.8 - - - Range for L lower molars, AW from PW Bartholomai (1971) - 8.6- 11.7 4. 7- 6. 8 8.9-13.2 5. 7-7.9 11.4-14.3 6.6- 8.5 11. 8-14.9 6.7- 8.7 Range for L _ 11.4-14.4 12.5-15 lower molars, AW - - 6.7- 8.8 8 - 8.9 from PW Tedford (1967) 6. 3-8. 5 7.1- 8.5 FOSSIL MARSUPIALS 143 Plate 1. All illustrations are two-thirds natural size. Figure 1. Macropus titan from 30-60 cm level of Pit I, Douglas Cave. Occlusal view of broken left mandible DC 1 (CPC 17108). Figure 2. Lingual view of same. Figure 3. Labial view of same. Figure 4. Macropus altus from 30-60 cm level of Pit I, Douglas Cave. Labial view of broken left mandible DC 9 (CPC 17109). r 144 JOHN D. GORTER Plate 2. All illustrations are two-thirds natural size. Figure 1. Macropus altus from 30-60 cm level of Pit I, Douglas Cave. Occlusal view of broken left mandible DC 9 (CPC 17109). Figure 2. Macropus altus from 30-60 cm level of Pit I, Douglas Cave. Occlusal view of broken left mandible DCll (CPC 17110). Figure 3. Lingual view of same. Figure 4. Macropus giganteus from top 15 cm of Pit I, Douglas Cave. Occlusal view of broken right mandible DC 4 (CPC 17111). Figure 5. Labial view of same. FOSSIL MARSUPIALS 145 B.P. date from Lancefield. In view of the close stratigraphic relation of the two species in the Douglas Cave, it is probable that there was a period of co-existence of the older M. titan with the more recent M. giganteus. E e 15 n o x: '* 10- a> (/> o 0. 54- 10 Macropus titan qci ! ‘^Dce Macropus giganteus 1 — I — I — I — I — I — I — I — I — I — I — I — T— 1 15 20 25 Length of Msfmm) Figure 2. Scatter diagram of LM3 against PWM3 of Macropus titan from the Douglas Cave, compared with the fields for M. titan and M. giganteus as presented by Marshall (1974). ACKNOWLEDGEMENTS Dr. Jeanette Hope of the Australian National University read the first draft of this paper and made many helpful suggestions. The figures were drawn by M. Moffatt. Comparison with material in the possession of J. Hope, C.R. Tidemann, J.H. Calaby, and D. Horton was made in the course of this study. I would like to express my thanks to M.D. Plane, my supervisor during this study while the writer was at the Australian National University. This article is published with permission of the Director, Bureau of Mineral Resources. REFERENCES Anderson, C., 1929. Palaeontological Notes No. 1. Macropus titan Owen and Thglacoleo carnifex Owen. Rec. Aust. Mus. , 17(1), 35-49. Bartholomai, A. , 1971. Morphology and variation of the cheek teeth in Macropus giganteus Shaw and Macropus agilis (Gould). Mem. Old. Mus., 16(1), 1-18. Bartholomai, A. , 1975. The genus Macropus Shaw (Marsupialia: Macropodidae) in the Upper Cainozoic deposits of Queensland. Mem. Qld. Mus., 17(2), 195-235. Dun, W.S., 1897. Annual report of the Assistant Palaeontologist and Librarian for the year 1896. Ann. Rep. Dept. Mines and Agriculture, N.S.W. for 1896, 152-154. Flood, J. , 1974. Pleistocene man at Cloggs Cave: his tool kit and environment. Manlcind, 9, 175-188. Frank, R. , 1969. The clastic sediments of Douglas Cave, Stuart Town, New South Wales. Helictite, 7(1), 3-13. Frank, R.M., 1972. Sedimentological and morphological study of selected cave systems in eastern New South Wales, Australia. Ph.D. Thesis, Australian National University , Canberra . (Unpub 1 . ) . Leigh, W.S., 1897. Stuart Town, Caves. Ann. Rep. Dept. Mines and Agriculture, N.S.W. for 1896, 152. Marshall, L.G., 1973. Fossil vertebrate faunas from Lake Victoria region, S.W. New South Wales, Australia, Mem. Nat. Mus. vict. , 34, 151-171. Marshall, L.G., 1974. Late Pleistocene mammals from the "Keilor Cranium Site", southern Victoria, Australia. Mem. Nat. Mus. Vict. , 35, 63-86. Merrilees, D. , 1968. Man the destroyer: late Quaternary changes in the Australian marsupial fauna. J. Roy. Soc. West. Aust., 51, 1-24. Tedford, R.H., 1967. The fossil Macropodidae from Lake Menindee, New South Wales. Univ. Calif. Pubis. Geol. Sci., 64, 1-156. Wright, R.V.S., 1975. Preliminary excavation report - Lancefield Swamp, 1975. Rept. Australian Institute of Aboriginal Studies. (Unpub 1 . ) Bureau of Mineral Resources, Geology and Geophysics, Canberra, A.C.T., 2601. (Manuscript received 1.1.77). THE ROYAL SOCIETY OF NEW SOUTH WALES The Society originated in the year 1821 as the Philosophical Society of Australasia. Its main function is the promotion of Science through the following activities ; Publication of results of scientific investigation through its Journal and Proceedings ; the Library ; awards of Prizes and Medals ; liaison with other Scientific Societies ; Monthly Meetings ; and Summer Schools for Senior Secondary School Students. Special Meetings are held for the Pollock Memorial Lecture in Physics and Mathematics, the Liversidge Research Lecture in Chemistry, ami the Clarke Memorial Lecture in Geology. Membership is open to any interested person whose application is acceptable to the Society. The application must be supported by two members of the Society, to one of whom the applicant must be personally known. Membership categories are : Ordinary Members : SI 8.00 per annum plus $3 application fee. Absentee Members : $15.00 per annum plus $3 application fee. .Associate Members (spouses of members and persons under 25 years of age) : $5.00 per annum plus $2.00 application fee. .\ssociate Members (with Journal) : $12.00 per annum plus $2 application fee. Subscription to the Journal, which is published in four Parts per year, issued twice yearly in May and November, for non-members is $22 p.a. plus postage. For application forms for membership and enquiries re subscriptions, write to : The Royal Society of New South Wales, Science Centre, 35 Clarence Street, Sydney, 2000, N.S.W. The Society welcomes manuscripts of research (and occasional review articles) in all branches of science, art, literature and philosophy, for publication in the Journal and Proceedings. Manuscripts will be accepted from both members and non-members, though those from the latter should be communicated through a member. A copy of the Guide to Authors is obtainable on request and manuscripts may be addressed to the Honorary Secretary (Editorial) at the above address. r Report of Council, 31st March, 1977 (Continued from Volume 110, Parts 1 and 2) . . . . . . . . 73 Astronomy : Occultations observed at Sydney Observatory during 1974-1976. D. S. King and K. P. Sims . . . . . . . . . . . . . . . . . . 81 Precise Observations of Minor Planets at Sydney Observatory during 1976. T. L. Morgan . . . . . . . . . . . . . . . . . . 87 Chemistry : 'I'he Essential Oil of the Fly-Repellent Shrub, Pteiigeron bubakii. I. A. Southwell and J . R. Maconochie . . . . . . . . . . . . 93 Geology : A Bottom Profile across Lake Eyre North, South Australia. J. A. Dulhunty 95 Influence of Hydrothermal Treatment on Physical and Chemical Properties of Chrysotile Asbestos. P. H ahn-W einheimer and A. Hirner . . . . 99 Textures of the Carboniferous Ignimbrites in the Hunter Valley, N S.W. B. Nashar and A. T. Brakel . . . . . . . . . . . . . . Ill Clarke Memorial Lecture : Petrogenetic Aspects of Some Alkali Volcanic Rocks. J . F. G. Wilkinson . . 117 Palaeontology : f ossil Marsupials from the Douglas Cave, near Stuart Town, New South Wales. John D. Gorier . . . . . . . . . . . . . . . . 139 AUSTRALASIAN MEDICAL PUBLISHING CO. LTD.. 71-79 ARUNDEL ST. . GLEBE. N.S.W. , 2037 1977 c/lcme Bookbinding Co., Inc. 300 Summer Street Boston. Mess. 02210