We review the discovery of gamma-ray bursts, and summarise the new data from the Compton Gamma-Ray Observatory. We evaluate the evidence that gamma-ray bursts are at cosmological distances, and discuss applications of such gamma-ray bursts to cosmological problems.

The effect of varying magnetic field strength on the frequency of oscillatory motions for cellular multimode magnetoconvection has been investigated. In addition the influence of the thermal, viscous and magnetic diffusivities have been taken into account and the range of preferred horizontal scales established. The relationship between the period of oscillation and the magnetic field strength is determined.

The loss-cone-driven electron cyclotron maser instability is widely believed to be responsible for millisecond bursts of intense microwave emission often observed during solar flares. However, the maser radiation is strongly absorbed as it propagates outward from the corona and existing analytical models predict that this absorption should be sufficiently strong to prevent observable levels of the radiation from escaping, except under highly restrictive conditions. In order to address the problem of how microwave spike bursts can be observed at all, we present a numerical ray-tracing analysis which incorporates emission, propagation and absorption of fundamental cyclotron maser radiation in a realistic model of a coronal flux loop. It is found that the radiation can escape to a potential observer and that the physical conditions under which escape occurs are more restrictive for fundamental emission in the extraordinary (x)-mode than in the ordinary (0)-mode. Escaping radiation in the x-mode is found to be highly directional and chiefly observable toward the center of the solar disk, while escaping 0-mode radiation is found to emerge from the corona over a much wider range of directions, with some cases corresponding to observable radiation near the solar limb.

Clumpy, intense wave packets observed in situ in the Jovian and terrestrial electron foreshocks, and in the Earth’s auroral acceleration zone, point to the existence of non-linear plasma turbulence in these regions. In non-linear turbulence, wave packets collapse to short scales and high fields, stopping only when coherent wave-particle interactions efficiently dissipate the energy in the waves. The purpose of this paper is to examine the shortest scales and highest fields achieved during collapse in a strongly magnetised plasma, and identify parts of the solar system where the magnetised aspects of wave collapse are important.

The CANGAROO project incorporates two Čerenkov imaging telescopes at Woomera to obtain stereo images of very high-energy gamma-ray (and cosmic-ray) showers. The first stereo observations, with one imaging system, were made in March 1992, and preliminary stereo imaging observations began in July 1992. This paper describes the stereo imaging technique, the sources under investigation, and the indications from the first data sets.

Studies have been made of the extent to which extragalactic cosmic rays observed at the earth would be deviated by the galactic magnetic field. The distribution of these deviations is presented as a function of observed direction for two field models. Also using these two models, possible extragalactic directions are derived for the cosmic rays detected by the SUGAR giant air shower array as a basis of a possible technique for searching for extragalactic cosmic ray sources.

We describe a versatile infrared camera/spectrograph, IRIS, designed and constructed at the Anglo-Australian Observatory for use on the Anglo-Australian Telescope. A variety of optical configurations can be selected under remote control to provide several direct image scales and a few low-resolution spectroscopic formats. Two cross-dispersed transmission echelles are of novel design, as is the use of a modified Bowen-Burch system to provide a fast f/ratio in the widest-field option. The drive electronics includes a choice of readout schemes for versatility, and continuous display when the array is not taking data, to facilitate field acquisition and focusing.

The linearity of the detector has been studied in detail. Although outwardly good, slight nonlinearities prevent removal of fixed-pattern noise from the data without application of a cubic linearising function.

Specific control and data-reduction software has been written. We describe also a scanning mode developed for spectroscopic imaging.

We have measured the 2.3 GHz total and correlated flux densities on a baseline of 275 km of all sources in the Parkes catalogue which

1. (i) are south of declination +10°,

2. (ii) have a catalogued 2.7 GHz total flux density exceeding 0.5 Jy, and

3. (iii) have a 2.7/5.0 GHz spectral index flatter than −0.5.

More than 14% of the sample showed visibility amplitudes greater than 0.9, and more than 72% showed visibility amplitudes greater than 0.5. Of the sources with optical or other identifications 79% were quasars. In this paper we briefly summarise the results of this survey.

The Parkes-MIT-NRAO (PMN) 4.85 GHz continuum radio survey of the southern sky was undertaken in 1990 June and November. This survey was performed on the Parkes 64 m telescope using the NRAO seven-beam receiver. A point-source catalogue of 36,640 radio sources has been produced for the Southern Survey zone −87.5° ≤δ ≤ −37° and for the Tropical Survey zone −29° ≤δ ≤ −9.5°. The flux limit of this survey varies with declination and is typically about 30 mJy.

We have begun to cross-correlate the PMN data with sources contained in catalogues compiled at radio, and other, wavelengths. We have found associations for 96% of the PKSCAT90 2700 MHz database sources, and 95% of the Molonglo 408 MHz Catalogue sources within in the PMN Southern and Tropical Survey zones.

A program to identify the optical counterparts of PMN Southern Survey point sources, S4.85 GHz ≥ 70 mJy, using the COSMOS database, is under way. To facilitate this programme we are improving the positional accuracy of PMN sources with observations made at the Australia Telescope National Facility compact array. We have developed a new “snapshot” mode of observing to process the large number of sources (~ 8000) in our sample. It is possible to obtain accurate positions from three snapshots efficiently with a total integration of < 3 minutes.

This paper discusses the observation of molecular hydrogen line emission outside near-infrared wavelengths, and in particular the opportunities afforded to molecular astrophysics by studies in the mid- and far-infrared.

we use Monte Carlo simulations of the period distribution of magnetic cataclysmic variables to show that the synchronised AM Herculis-type systems and the asynchronous Intermediate Polars are essentially unrelated systems following different evolutionary scenarios.

Spectroscopic imaging of the SNR RCW 103 has revealed extensive emission in near-infrared lines of H2and [Fe II], where the blast wave is encountering a molecular cloud. The H2 appears to be located outside the [FeII], a morphology which challenges our understanding of shock wave physics. It is suggested that reverse shocks may be responsible for the phenomenom.

The radio emission associated with SN 1987A appears to be synchrotron emission resulting from the acceleration of electrons at the interface between the outward moving shock wave and clumps of circumstellar material. The Australia Telescope Compact Array is now able to resolve this region, which has dimensions of ~ arcsec, revealing a slight (10%) asphericity in the distribution of the low density gas within the [OIII] circumstellar ring. Assuming that the radio emission arises from a region just behind the shock front, we deduce a mean radial expansion velocity, from 1987 to 1992, of 29 200 kms. First observed contact of the shock with the [OIII] circumstellar ring could occur as early as mid-1993, depending on the deceleration in the intervening gas. This will probably be closely followed by shock-excited optical lines, a strong X-ray outburst and a further increase in the radio emission.

The ideas and work presented in this paper date almost entirely from 1977 when the original paper was submitted. The referee suggested that we check out several other strong northern candidates before publication, but upon resubmission several years later the editor noted that as it was now accepted that DA,F white dwarfs did not exist there was little point in publishing the paper. In view of the fact that our preprint was the main stimulus for this status change and reference to this work remains ‘1977 unpublished’, we would like the paper to be published and thus complete the historical record.

Observations of some of the stars considered by Weidemann as the most likely DA,F candidates are discussed together with theoretical computations of white dwarfs and subdwarfs. The reddest stars BPM 23251 and G 82–23 are shown to be K subdwarfs and L 532–81 (EG62) is a DA white dwarf. It is suggested that the DA,F stars, if they exist at all, are extremely rare and that the metal abundance of the DA sequence may be less than 10–74 of the solar abundance.

From extensive spectroscopic observations we derive the three-dimensional structure of the Homunculus Nebula surrounding η Carinae. The analysis shows that the two lobes are hollow and incomplete, and that the gas was ejected at similar epoch. Consequently the distribution of material we see reflects its initial velocity. Using the observed expansion velocity, and the known time of ejection, we derive a distance to η Car of 2.2 ± 0.2 kpc.

We discuss a generalisation of the synchrotron bubble model which has been applied to short-lived radio transients which can peak and decay over just a few days. The assumptions of the simple model imply that when the flux is increasing with time, it must also be an increasing function of frequency. Observations of two recent radio transients, Nova Muscae 1991 and the Galactic Centre Transient, include the first data showing such a rising phase, and in both cases the radio flux was a decreasing function of frequency during the observed rising phases. Thus the simple synchrotron bubble model is inadequate, at least for these events. A fundamental feature of the simple model is the assumption that the process accelerating the radiating electrons ceases before the radio emission can escape. We relax this assumption by including an injection of electrons, with a constant energy spectrum, into the synchrotron bubble.

Who discovered QSOs? What was the first QSO discovered? To Maarten Schmidt (1963) goes the credit for realising that the emission lines he saw in a 13 mag star were the Balmer series in hydrogen at a redshift of 0.158. The first QSO had been recognised. But discovery is a complex process. Schmidt observed the ‘star’ because it was associated with a radio source 3C 273 by means of an accurate radio position. This was the first of several instances where accurate radio positions have enabled significant progress in the QSO story.

Charles Todd was the first Government Astronomer and Superintendent of Telegraphs in South Australia. Most widely known for his instrumental role in the construction of the Overland Telegraph, linking Australia and England, Todd also established the Adelaide Observatory and made valuable contributions to both astronomy and meteorology.

In March 1963 a paper by Hazard, Mackey and Shimmins in Nature, announced an accurate position for the radio source 3C 273. This was subsequently identified as the first known quasar. Through an innocent editorial mistake, the affiliation of the first author was wrongly attributed. This caused long-term animosity between the Physics Department of the University of Sydney and the CSIRO Division of Radiophysics.