Figure 1 shows the spectrum of 3C279 (1253-05) in mid-1966 derived from observations at frequencies of 467, 630, 960, 1410, 2650 and 5000 MHz made by Batchelor, Brooks, Cooper, Milne, Roberts and the author with the CSIRO 210 ft. telescope. The figure also shows the spectrum obtained in 1964 by Kellermann with the same telescope. All the flux densities have been measured relative to a number of calibration sources on a scale previously determined by Kellermann. Although our flux densities are preliminary, further analysis should not change them significantly.

Using a network of ground-based stations stretching from Lae, New Guinea, to the Antarctic, and high altitude balloons and rockets, the Hobart cosmic ray group is studying several aspects of naturally-occurring high energy radiation.

Observations at a number of frequencies indicate that for at least two pulsars the average pulse shape has a slow but quite definite frequency dependence. Figure 1 shows average pulse shapes for CP 1919, CP 0950 and CP 1133. With the exception of those at 408 MHz these results were obtained at Parkes. The 408 MHz pulse shapes were obtained at Jodrell Bank by Lyne and Rickett. Circumstances of the observations are listed in Table I. Linearly polarized feeds were used at all frequencies.

Although the synchrotron emission from a single electron moving with an ultrarelativistic velocity βc in a magnetic field B0 is elliptically polarized, the characteristics of the polarization ellipse being determined by the position of the direction of observation n relative to the cone swept out by the direction u of the velocity vector about the direction of B0, the resultant emission from a distribution of such gyrating electrons is, to a first approximation, linearly polarized in the direction perpendicular to the projection of B0 on to the plane transverse to n. The reasons for this are that for a single electron the emission is effectively confined to within a small angular distance O(ξ) of u, where ξ = √ (1 −β2) ≪ 1 and that (i) the fourth Stokes parameter is to this approximation an odd function of the angle ip between n and u at its closest approach, and (ii) the number of electrons passing within equal angular distances ψ = ±O(ξ) are equal. It follows that if ξ is large enough it is possible for the next approximation to the fourth Stokes parameter for a distribution of electrons to be significant, so that the resultant emission is elliptically polarized.

Balloon borne alt-azimuth mounted X-ray telescopes flown from Hyderabad, India (17°N), Palestine, Texas (32°N), and Mildura, Australia (34°S), have detected hard X-rays (hv>20KeV) from discrete sources in the Cygnus region.

The cosmic ray flux in the energy range 100 MeV/nucleon ≤ E ≤ 1 GeV/nucleon is remarkable for its high degree of isotropy. Observed deviations from isotropy seldom exceed a few per cent and are commonly much smaller. The mechanism responsible for this isotropy is presumed to be multiple, large-angle scattering of the charged cosmic ray particles by irregularities of the interplanetary magnetic field. While generally precluding any hope of discovering a source-related anisotropy of the flux in this energy range, it is just this strong interaction of the cosmic rays with the interplanetary medium that allows the study of the small observed anisotropies, both persistent and transient, to yield considerable information about the structure of the interplanetary medium (the solar wind and its entrapped magnetic field).

Pulsars display a complicated pattern of amplitude variations. The scale of variations range from months through hours to sub-milliseconds. The purpose of this note is to present polarization data for CP 1919 and intensity distribution data for PSR 0833—45 which are relevant to pulse-to-pulse variations and the variation over periods of hours respectively.

Gold has suggested that pulsars are rotating neutron stars with strong magnetic fields. We have studied the simplest such model, one having an external dipole field aligned with the rotation axis.

In the field of cosmic ray galactic studies, two kinds of phenomenon are currently being investigated by members of the Hobart cosmic ray group, led by Dr. A. G. Fenton. In a combined project with the University of South Australia, the University of Tasmania is participating in an experiment whose aim is to locate sources of galactic X-rays in a part of the southern sky inaccessible from the northern hemisphere. The object of the other investigation, which forms part of a joint ANARE-University programme of cosmic ray research, is to elucidate the evidence for intensity maxima from fixed galactic directions in the otherwise isotropic charged primary radiation.

A modern coudé spectrograph for observing the spectra of stars and planets with maximum efficiency is a complex and versatile instrument, which represents an investment about equal to that for a 40-inch telescope. The trend toward special telescopes for special kinds of work began in the 1930s with Schmidt telescopes for wide angle photography, and has been followed by the use of specialized telescopes for photoelectric photometry, and more recently by telescopes for polarization measurements. Two telescopes designed for exclusive use with coudé spectrographs already exist—the 48-inch at Victoria and the 60-inch at Haute Provence. More are needed, particularly in the southern hemisphere.

Instruments were flown on the Pioneer 6 and 7 spacecraft during 1965-66 to study the degree of anisotropy of cosmic radiation in the energy range 7.5-90 Mev/nucleón. The instruments record the cosmic ray fluxes from each of four contiguous ‘quadrants’ of azimuthal rotation of the spacecraft, for each of three energy windows 7.5-45 Mev, 45-90 Mev, and 150-350 Mev for alpha particles and heavier nuclei. In addition, the counting rate of all particles of energy >7.5 Mev is recorded, thereby providing cosmic ray data of high statistical precision useful in the study of fast changes in the cosmic ray flux.

In a study of radio sources in the southern sky, observations have been made of 190 sources noted as ‘extended’, ‘possibly extended’ or ‘confused’ in the Parkes and other catalogues.

Radio contours are presented for Fornax A (NGC 1316), Virgo A (NGC 4486) and the central component of Centaurus A (NGC 5128). These three objects are intrinsically strong radio emitters identified with bright galaxies. The work forms part of a survey of all bright galaxies with the Molonglo cross down to a corrected total magnitude of 11m.0 with an extension down to photographic magnitude 12m.5 in an limited region of the sky.

The Parkes Interferometer was designed and constructed by the CSIRO Radiophysics Laboratory. It is unique in having continuously variable North-South and East-West baselines making possible plotting the visibility function of a source in a single observation. From an engineering point of view it is noteworthy in that the receivers, control circuits and computer employ semi-conductors exclusively.

The usual method of calculating corrections for precession nutation and aberration involves the use of the precession constants M, N, m, n and the Day Numbers A, B, C, D, E, J, J’. For the computer analysis of radio source data obtained with the Molonglo Telescope it seemed desirable to compute the corrections directly from the basic formulae, thus avoiding the trouble and possibility of accidental error in copying out the Day Numbers for each day’s observation. In this paper we quote in full the equations that are used in the analysis of the Molonglo data. These equations are abstracted from The Explanatory Supplement to the Astronomical Ephemeris and Smart’s Spherical Astronomy, and expressed in a form which parallels the organization of the computer routine. In addition to the observed position, the only other data required to compute the precession nutation and aberration corrections is the Julian date (J.D.). This basic parameter in the equations is very simply calculated to the required accuracy of a few minutes from the time and date of the observation. The computer routine that computes the corrections occupies 1350 words of programme in the KDF9 computer and takes 200 msec to compute one position. If a large number of computations are being carried out on continuous data, then the average time per computation is only 5 msec as a full computation of the coefficients does not have to be made for each point separately.