Six sedimentary cycles, each hundreds of metres thick, have been recognised in the Surat Basin. The Jurassic cycles (nos. 1-4) typically started with high-energy deposition of coarse sediments, and ended with labile sand, silt, and mud. The environments are thought to have been braided streams, followed by meandering streams, swamps, lakes, and deltas. After a period of non-deposition or erosion, followed by a phase of high-energy deposition, the first Cretaceous cycle (no. 5) ended with marine mud; the second (cycle 6) started with paralic silt and sand and ended with shallow marine silt. The cycles are thought to be the result of global sea-level changes, characterised by rapid falls of sea level followed by slow rises, which, respectively, lowered and raised the base of erosion in the Surat Basin. During the Jurassic, the open sea lay several hundred kilometres to the east, and the sea only occasionally entered the basin, via the Brisbane and Toowoomba Straits. In the Cretaceous, the sea level was relatively higher and eustatic falls and rises of sea level led to alternating marine regressions and transgressions. The six Surat Basin cycles correspond in time to nine global sea-level oscillations. We think that some of the latter may have been too slight to be identified in the basin. There is also evidence that local isostatic movements may have exaggerated the impact of some global cycles and obscured that of others.

Dr Sangsters contribution on the ore deposits of the Cobar district (Sangster, 1979) is of interest in that it views from a different perspective a group of distinctive and economically important deposits whose origin is enigmatic. Despite the advances in ore genesis theory which have seen many of the former hydrothermal replacement deposits described in terms of volcanic processes operating at or near the ocean-ocean floor interface, workers familiar with Cobar geology have been circumspect in assigning to the Cobar deposits a similar mode of origin. The lead-zinc/copper zonation at the CSA and several of the idle mines which so impressed Dr Sangster had not gone un-noticed by local geologists (e.g. Robertson 1974, p. 180). That they have not taken the next, apparently simple step of considering this zonation in terms of the sedimentary-exhalative orebody model is for the good reason that the comparison does not survive detailed examination. In offering comment on Dr Sangsters paper it is appropriate to note firstly several areas where the bases for his interpretation in the CSA Mine area are wrong in fact.

OConnors comments regarding a proposed sedimentary-exhalative model for the CSA deposit are welcomed as additional contributions to discussions of the genesis of deposits in the Cobar district. His points with regard to errors of fact are answered below.

A series of transient electromagnetic (TEM) scale-model studies of the Elura zinc-lead-silver deposit near Cobar, NSW, has been carried out. The results are generalised inasmuch as they are not restricted to the Elura deposit, but may be applied to other pipe-like bodies. A model of Elura was cast out of typemetal into a graphite mould, which simulated a conductive host rock. For both one-loop and two-loop geometries the response of the body at early times is masked by the host rock, and the response of the body at late times is two to three times background. By studying the fall-off of response with depth of burial, it was found that there is an optimum time range in which a conductive body can be detected. This time range depends on the conductivity and size of the body and the conductivity of the host rock. The optimum loop size for detecting a body in a resistive environment is slightly larger than the size of the body in plan. If the body is contained in a conductive host rock, then a smaller loop will minimise coupling with the host rock and will enable detection of the body over a wider time range. However the smaller loop also emphasises lateral inhomogeneities in the overburden. The response obtained with a two-loop system is more complex than for a coincident-loop geometry, and can be positive or negative in sign, depending on the sample time, loop separation, depth of burial of body, and host-rock conductivity. For loop separations less than the depth of the body, the two-loop response closely resembles the one-loop response.

A new focal mechanism for the 1961 Robertson earthquake provides further evidence of contemporary thrust faulting in southeast Australia. When this is combined with the earlier work of Cleary and Doyle (1962) on the locations of the earthquake and its aftershocks, it seems that the earthquake was associated with a high angle (~80°) thrust fault, about 10 km wide and in the depth range 7 to 20 km, caused by northeast-southwest compression. This fault model is similar to that found by Mills and Fitch (1977) for the 1973 Picton earthquake, which also took place beneath the southern Sydney Basin. However, in situ stress measurements and earthquake data from immediately to the west in the Lachlan Fold Belt, suggest a different stress regime dominated by north-south compression. Thus, although the crust in southeast Australia is being compressed, the directions of maximum stress appear to change from region to region.

The interpreted velocity-depth structure in the crust of the Lachlan Fold Belt in southeastern Australia indicates that velocity gradients, rather than discontinuous velocity changes, characterise the region. The velocity-depth morphology varies across the region. In the upper crust at depths less than 12 km, there is evidence in three areas for velocity decreases within a general velocity increase from about 5.6 km/s to 6.3 km/s. In the middle crust (depth range 16 km to about 35 km) a low velocity zone is interpreted within a general increase in velocity from 6.3 km/s to greater than 7 km/s; the prominence of the low velocity zone varies throughout the region. The upper mantle velocity is in the range 8.02 to 8.05 km/s; this velocity is reached at a depth slightly greater than 50 km under the highest topography in Australia, and at depths between 40 and 50 km elsewhere. There are similarities between the velocity-depth structure in southeastern Australia and that in the Appalachian Orogen of North America. Geochemical mixing and/or compositional inhomogeneity is a likely reason for the velocity-depth structures, and such inhomogeneity is probably still influencing current moderate earthquake activity and continuing highland uplift. In pre-Ordovician times the region of the Lachlan Fold Belt probably consisted of continental crust which was submerged and thickened by episodic crustal development similar to the processes which resulted in the Appalachian Orogen.

Fault-plane solutions determined for earthquakes in northwestern Australia (6 May 1978), central Australia (25 November 1978) and southeastern Australia (4 July 1977) each indicate nearly horizontal axes of maximum compressive stress. However, the azimuths of these axes are different from the azimuths of maximum stress axes determined previously for earthquakes in each area. This may be the result of a combination of warped stress fields at the junction of geologically different crustal blocks, and faulting in weakened zones of these blocks where the strike is oblique to the regional direction of maximum stress. Results in northwestern Australia can be explained by such effects.

The first Australian earthquake accelerograms were obtained from an accelerograph situated in the Dalton-Gunning region of New South Wales. Preliminary results were obtained for earthquakes on 23 November 1976 (maximum resultant acceleration 0.66 m/s^2), 30 June 1977 (0.21 m/s^2), 4 July 1977 (0.95 m/s^2) and 3 February 1979 (1.3 m/s^2). Maximum ground velocities were calculated for these earthquakes, and isoseismal maps drawn for the earthquakes at Bowning on 30 June 1977 and 4 July 1977. These data were used to test the validity of the relation Y = ae^bM R^-C for assessing ground motion, and hence determining seismic risk. Because of the uncertainties in the derivation of analytical expressions for ground motions, the fit to observed values of acceleration, velocity and intensity is considered to be good. The relation recommended for use by the Seismic Sub-Committee of the Australian National Committee on Earthquake Engineering included a depth-adjusting factor Co, where (R^2 = D^2 + h^2 + Co^2). When Co was omitted a better fit to the observed accelerations was obtained, but a poorer fit to observed velocities. The isoseismal pattern for the 4 July 1977 earthquake supports the radiation pattern expected from the faulting suggested by the focal mechanism and distribution of aftershocks. The isoseismals for the 30 June 1977 earthquake show a different radiation pattern; this suggests a different focal mechanism.

Regional airborne magnetic and gravity data and field observations have been used to define four geophysical domains in the Lachlan Fold Belt of New South Wales. Each has a different pattern of anomaly trend and amplitude. Three domains correspond with provinces containing rocks of similar lithology and age; the other corresponds with the Darling Basin. The magnetic data highlight minor variation in magnetite abundance, chiefly in igneous rocks, and, across the region, reflect slight differences in magma composition, structural trends, and possibly also style of volcanism and sedimentation. The gravity data also delineate the same structural trends and some major lithological variations. The boundary of the Darling Basin is gradational. The other domain boundaries are sharp, but with no evidence of major faulting. Within the geophysical domains a classification of magnetic anomalies by length, width, and amplitude appears capable of distinguishing between the various sources, thereby providing a useful mapping tool in regions with surficial cover. Circular anomalies of largest extent occur over granitoids, the smallest over small pipes and veins, and the intermediate size over basic and ultrabasic stocks. Elliptical anomalies are associated with magnetic granitoids and some basic and ultrabasic bodies. Sources of narrow anomalies include basic dykes, steeply tilted Palaeozoic lavas, ignimbrites, and serpentinites. Sources of complex zones include Tertiary basalt flows, piles of basic and intermediate lavas surrounding small intrusions, and some inhomogeneous granitoids. The regional gravity data reflect features with very large dimensions or large density contrasts. A major north-northeast trending Bouguer anomaly low corresponds with the Eastern Highlands, a region of high relief, thick crust, and extensive granitoids. A narrower low corresponds to another belt of granitoids trending north-northwest from Holbrook to Cobar. Small-scale gravity features include the Coolac Serpentinite and its associated basic rocks, the Mid-Silurian to Mid-Devonian Hill End Trough, and numerous regions of Late Devonian quartzose sedimentary rocks.

A deep crustal seismic reflection survey, conducted at Gundary Plains near Canberra, to test a digital seismic recording system, produced additional data for interpretation of seismic refraction profiles in the Lachlan Fold Belt. Good reflections were recorded down to the probable Moho, at an estimated depth of 41 km. The intracrustal reflections are characterised by bands of seismic energy, which probably represent velocity transition zones within the crust. Now that the technique has proved successful, longer seismic reflection traverses are needed to explore the major deep geological features of the Lachlan Fold Belt.