Quaternary sediments are apparently more widespread and better represented in the northern Scott Plateau and Java Trench than in parts of the central Scott Plateau, where older rocks are exposed on the sea floor. Nevertheless, wherever Quaternary calcareous nannofossils are recovered from these areas, they are associated with reworked Upper Cretaceous and Tertiary forms. This may have been caused by bottom currents actively eroding parts of the central Scott Plateau and redistributing the fine fraction elsewhere on the plateau, Java Trench and nearby areas. These currents may have been active from the middle Miocene. Late Pleistocene to Holocene calcareous nannofossils occur throughout the near-surface (ca 1 m thick) sediments in the northern Scott Plateau, but are absent from such sediments in the Java Trench. However, Pleistocene nannofossil assemblages, older than those from the Scott Plateau, occur intermittently at lower levels in the near-surface (ca 1 m thick) sediments in the Java Trench. This difference is explained by suggesting that the present Nanno Solution Depth lies between 3290 m and 4950 m water depth in the Scott Plateau and Java Trench areas, but in earlier Quaternary times it fluctuated between 5090 m and 5424 m in the same areas.

During May-June 1973, a deep crustal seismic refraction survey was undertaken in the Bowen Basin, Queensland. Blasts from open-cut coal mines in the Basin were used as sources of seismic energy, and recordings were made at twenty-two sites on a 375 km-long line along the axis of the Basin between Goonyella in the north and Moura in the south. A four-layer crust was interpreted from the seismic data, with P-wave velocities of 4.00 ± .22, 5.33 ± .08, 6.39 ± .07 and 7.07 ± .02 km/ s respectively. The total thickness of the 4.00 and 5.53 km/s layers is about 6 km under Goonyella, and slightly more under Dingo 130 km north of Moura. Earlier magnetic and gravity work indicates that these two layers thin southwards towards Moura. They comprise folded Permian-Triassic sediments and possibly also Early to Middle Palaeozoic rocks. The 6.39 km/s layer probably represents the igneous or granitised basement. A lower crustal layer with a velocity of 7.07 km/s and thickness ranging between 5 and 6 km has been interpreted. The total crustal thickness at the centre of the traverse is 36 km, and the upper mantle P-wave velocity is 8.10 ± .11 km/s. A southward-dipping Moho may be interpreted, with the crustal thickness increasing from 35 km, 30 km south of Peak Downs, to 37 km near Dingo. The gravity field calculated from the model conflicts with the observed gravity, suggesting a more complex model than that defined by the seismic refraction data alone.

In 1976, four seismic reflection traverses were shot across, and close to, the eastern margin of the Galilee Basin to investigate the structure of the basins northeast margin with relevance to the extent of the Permian coal measures, and to investigate the relationships and extent of the underlying Devonian Adavale Basin and Devonian-Carboniferous Drummond Basin. The results show that the basins northeast margin is steep and faulted, and that there is only a narrow strip in which coal is likely to be found at easily mineable depths. About 30 km southwest of this margin, the sediments are undisturbed, with a southerly dip of about half a degree; here 2000 m of Galilee Basin sedimentary rocks overlie 700 m of Drummond Basin sediments - which extend further to the northwest than was previously thought. The eastern part of the Galilee Basin is underlain by the fluviatile sediments of the Drummond Basin. The more prospective Adavale Basin does not extend as far north as Jericho No. 1 exploration well. The Koburra Trough, along the northeast margin of the Galilee Basin, contains a thick sequence of Permo-Carboniferous sediments; results from the 1976 survey indicate that it could be bounded by large anticlinal or monoclinal structures which might provide petroleum traps; it is on this area that exploration should concentrate.

Chemically analysed lavas from Bagana, an active andesitic volcano on Bougainville Island, can be assigned to one of three age groups - pre-1943, 1943-53, or 1959-75. Lavas of the oldest group are chemically the most fractionated, whereas those of the 1943-53 group are the most mafic. The rocks of the youngest group, although intermediate in degree of fractionation have K2O, Rb, and Ba abundances similar to, or lower than, the rocks of the 1943-53 group. The three groups appear to represent distinct batches of magma that were successively erupted from Bagana, possibly from a high-level reservoir that was periodically emptied and refilled. The andesites are regarded as fractionates of mantle-derived mafic magmas. Most of the crystal fractionation probably took place during ascent from the mantle source region, and before entry into the reservoir beneath the volcano. An average chemical composition of the analysed Bagana andesites has major-element values close to those of the mean for more than 800 analysed late Cainozoic volcanic rocks from Papua New Guinea, and is proposed as a reference andesite composition for comparative studies.

Geochemical data are presented for a sequence of spilitic pillow basalts (Tumu River basalts) associated with peridotites and gabbros of the Marum ophiolite complex in northern mainland Papua New Guinea. The basalts are strongly differentiated from relatively magnesian types (Mg-value = 70) to ferrobasalts (Mg-value = 30) characterised by high levels of Fe, Ti, Zr, Nb, Y. The Tumu River basalts are enriched in large ion lithophile elements such as REE, Zr, Hf, Nb, P2O5, and compare with tholeiites from oceanic islands. Major and trace elements suggest extensive fractionation involving olivine, pyroxene, and plagioclase, followed by pyroxene, plagioclase, titanomagnetite, and ilmenite. Trace-element plots are used to examine fractionation processes and to estimate abundances in the parent magma. The calculated initial concentrations are compared with abundances and abundance ratios in least fractionated enriched and depleted tholeiites. The abundances in the parent magma are used to calculate source abundances for large (20-30%) degrees of partial melting. The levels range from 2-3 times chrondites for HREE, Ti, Y, Zr, Sc, and P2O5, to 3.5-5.5 times for LREE, and are similar to those inferred for other LREE-enriched tholeiites from both oceanic and continental areas. The chemistry of the basalts therefore reflects the mantle-source composition rather than a particular tectonic setting within an ocean basin.

In 1977 the R.V. Valdivia carried out a survey between Scott Plateau and the Java Trench, during which 1700 km of 24-channel seismic data, and 2550 km of bathymetric, gravity and magnetic data, were recorded; and 31 bottom samples were obtained, in water depths ranging from 2000 m to 5800 m. The Scott Plateau trends NNE and is bounded to the west by the Argo Abyssal Plain and to the north by the Roti Basin. The plateau is a foundered continental block, and lies at an average depth of 2000-3000 m. On the plateau the dominant fault direction is NW to WNW, an ancient strike direction on the Australian continent. The western margin probably formed as a series of NE-trending rifts and NW-trending transforms during Late Jurassic breakup. Canyons cut the western margin, and some of these appear to be fault-bounded. One such fault forms the northern margin of a major NW-trending feature, the Wilson Spur. This appears to be a transform fault and perhaps extends across the abyssal plain as far as the eastern end of the Java Trench. Valdivia seismic profiles suggest that, at the trench, it separates thrust-faulted continental crust to the east from oceanic crust to the west. This could explain the eastern termination of the deep part of the trench. The bathymetric depression of the Roti Basin, which lies southeast of the Java Trench, links the trench to the Timor Trough. The Argo Abyssal Plain slopes gently southward, with water depths ranging from 5000 m near the Java Trench to 5730 m in the south. Oceanic basement varies from smooth to hummocky and irregular, and is overlain by about 400 m of acoustically semi-transparent Late Jurassic and Cretaceous sediments, that is in turn unconformably overlain by 200 m of layered Tertiary sediment. Bottom samples taken by R.V. Valdivia from the outer Scott Plateau have provided new information about seismic sequences. They show that Callovian breakup was preceded by a period of basic volcanism and shallow marine sedimentation, that restricted shallow marine conditions followed in the Late Jurassic, and that bathyal carbonate sedimentation prevailed by the Late Cretaceous (Campanian). Quaternary marls cored on the northern Scott Plateau straddled the Pleistocene-Holocene boundary, and siliceous oozes cored on the southern slope of the Java Trench contain nannofossils which, below a few decimetres, are older than late Pleistocene. The Java Trench cores indicate that the calcite compensation depth was apparently between 5420 and 5700 m in the early or middle Pleistocene, and is above 4950 m now. The Scott Plateau cores indicate that the present calcite compensation depth in the region lies below 3290 m. On the Scott Plateau Holocene sedimentation rates are about 5 cm/I000 years, but in the Java Trench they are much lower. Manganese oxide crusts and nodules were recovered from the Scott Plateau, but their content of valuable metals was low.

A Workshop was held at the end of the 7th BMR Symposium in May 1978 on what the involvement of government organisations in Australia should be in geochemical surveys. This report summarises the proceedings and discussions, and presents the results of the Workshop.

Twenty short cores were recovered from the northern Scott Plateau (water depths ca 3200 m) and the southern slope of the Java Trench (water depths 4950-5790 m) by R.V. Valdivia in 1977; selected cores have been studied in detail. About 55 species of benthonic foraminifera were identified on the plateau, but only half as many in the trench. Only deep-water forms occur in the trench, suggesting an absence of transport from shallower depths. Fragile agglutinating forms are confined to the surface sediments, or are totally absent. The planktonic foraminifera from the plateau generally belong to tropical associations. The assemblages show that the Holocene-Pleistocene boundary lies at about 60 cm in the cores, and that carbonate solution is more marked above than below the boundary.

A system is described capable of simulating simultaneously a number of controlling parameters occurring in sedimentary environments. The system is composed of an experimental tank, reservoir tank, hydraulic, heating and illuminating attachments, and an electronic control programmed to regulate cyclic and non-cyclic operations of duration between 1 second and 10^5 seconds. The electronic component is capable of controlling several simulating systems simultaneously, their number being dependent on their complexity.

Rb-Sr isotopic data reveal a close temporal association between ore deposits and the emplacement of Upper Palaeozoic granitic intrusions in the Herberton tinfield of northeast Queensland. Three main granite types are present: Elizabeth Creek type, consisting mainly of leucocratic biotite adamellite and ranging in age from 292 or 299 m.y. to 320 m.y.; Herbert River type, mainly granodiorite and isotopically dated at 278 to 315 m.y.; and 280 m.y. old Mareeba type, which consists mainly of biotite adamellite. Six mineralising events have been detected, all but the youngest being related to Elizabeth Creek Granite. The oldest, about 320 m.y. ago, was associated with intrusion of apparently small granite bodies southwest and southeast of Irvinebank. The next, at 314 m.y., accompanied the intrusion of the main mass of Elizabeth Creek Granite in the Emuford-Watsonville-Coolgarra area. A slightly later event, at 309 m.y., was associated with the emplacement of petrographically similar granite near Herberton. Mineralisation in the Wolfram Camp and Bamford Hill areas appears to have taken place about 303 m.y. ago. All of these events were characterised by greisen formation. A subsequent mineralisation in the Herberton area, at about 297 m.y., is related to a fine-grained phase of Elizabeth Creek Granite, and is characterised by chloritic alteration. The youngest dated event, in which hydrothermal alteration led to the production of fine-grained green biotite, occurred at about 284 m.y., and may be related to an intrusion of Mareeba-type granite. None of the major economic products (Sn, W, Cu, Pb, Ag) were confined to a single mineralising event: tin deposits, for example, were formed during at least five of the six episodes, and tungsten during at least four.