Both Holmberg (1961) and de Vaucouleurs (1963, 1973) found a significant difference between the mean velocities of the different morphological types in the Virgo cluster of galaxies. To check on any dependence of this result upon spatial distribution, Lewis (1975) plotted the position of every object in the compilation of de Vaucouleurs (1961) as a function of its type. Figure 1 is the resulting plot for the elliptical galaxies, which appear to lie within two well-defined and well-separated groups. The principal parameters of these groups are listed in Table I. Arp (1967) briefly noted the linear form of the northern group centred on NGC 4486. This paper examines the evidence for the physical existence of these groups.

The most prominent features in the spectra of Sgr A and Sgr B2 between 1636 and 1640 MHz are absorption lines corresponding to the 2π3/2, J = 3/2 ground state transitions of the isotropic species 18OH, which have rest frequencies of 1637.5 and 1639.4 MHz (Rogers and Barrett 1966). As a result of Doppler shifting, the lines of each transition are spread over a frequency range greater than 1.5 MHz. There are also present emission features which Zuckerman et al. (1971) have identified with the 110 -l11 transition of formic acid (HCOOH), which has a rest frequency of 1638.805 MHz. In view of some disagreement in the literature over the 16OH/18OH abundance ratio (Wilson and Barrett 1970; McGee et al. 1970), and of the baseline uncertainties in the spectra published in support of the HCOOH detection, we have obtained spectra of the two sources over the frequency range under discussion.

The majority of OH/IR stars have been identified as M-type Mira variables, but there are a few cases where the stars have been identified as M supergiants or late-type semi-regular variables. A large number of OH sources have also been discovered with the 1612 MHz OH characteristics of OH/IR stars but no apparent optical or infrared counterparts. In several cases the velocities of these unidentified sources (OH stars) have been outside the velocity range of the neutral hydrogen distribution, suggesting that some of these sources may represent a population with a higher velocity dispersion than Population I objects (Kerr and Bowers 1974a, b). The lack of infrared and optical counterparts may indicate that the sources are quite distant, since identified OH/IR stars are often bright infrared objects. A large-scale 1612 MHz sky survey has therefore been initiated to study the galactic distribution and kinematics of the OH stars and to investigate the possibility that they may represent an older population of stars (Bowers et al. 1974).

In the first decade or so of 21 cm studies the galactic spiral structure problem was treated in a very straightforward manner. A simple velocity-distance transformation was used to derive the locations of the spiral arms from the radial velocities of the main features on the 21 cm profiles. Some well-known diagrams were obtained in this way.

The ratio of the mass density in visible galaxies to the upper limit for the HI mass density determined from the lack of continuous Ly-α absorption in QSO spectra is 3 × 104. Thus, if galaxies form by condensation from an intergalactic medium, the process must be extremely efficient. Although no absorption due to a continuous distribution of intergalactic HI has been detected, many QSOs have been discovered that have absorption lines in their spectra, which have redshifts very different from the emission lines, and may be produced in clouds of integalactic HI. As would be expected if this were the case, the proportion of QSOs with absorption lines in their spectra is greatest for those which have large emission line redshifts. Another possibility, especially for those absorption lines in redshift systems with Zabs nearly equal to Zem , is that the absorption lines are produced in clouds surrounding the QSO.

The radio source, Cen A, is large and complex with many peaks in the brightness distribution over an area about 4 x 10 degrees. The peculiar elliptical galaxy NGC 5128 lies between the two strong inner radio brightness peaks and is centred on a weaker central radio source. This radio source is in the centre of the dust lane which divides the galaxy and may be related to the infrared, X-ray and γ-ray sources.

To date the Culgoora radioheliograph has recorded seven continuum bursts simultaneously at its three operating frequencies of 160, 80 and 43 MHz. In every case the radio emission is found to originate from stationary sources whose heights increase with decreasing frequency. The implications of these new observations are discussed below.

Multi-frequency observations with the Culgoora radioheliograph (Wild 1967); Sheridan et al. 1973) of five solar bursts of spectral type II (Wild et al. 1954) have now been analysed. The spectra of all five bursts show split-banded harmonic structure. Each type II burst passed through only two of the heliograph’s three observing frequencies (160, 80 and 43.25 MHz). Relationships between the positions, sizes and brightness temperatures of the various sources in each burst are presented and summarized; their implications for the propagation of shock waves and radio waves in the solar corona are briefly discussed.

The term ‘flare continuum’ (denoted by FC) was introduced by Wild (1970) to describe a strong, stationary continuum source at the metre wavelengths which occurs early in some flare events, particularly those of high energy. The flare continuum is distinct from the ‘storm continuum’ (SC), which occurs late in some flare events, becomes strongly circularly polarized and may continue as a ‘storm’. Both had previously been referred to as ‘stationary type IV’ bursts, to distinguish them from the original ‘moving type IV’ burst, IVM.

Determination of the large-scale structure of flare-produced disturbances requires either multiple spacecraft observations or an ensemble averaging of single spacecraft observations of a number of events. There is currently some confusion in the results of studies of shock normals derived from spacecraft data. Chao and Lepping (1974) suggest that the average shock shape near 1 AU is essentially spherical while Bavassano et al. (1973) suggest that the disturbance corresponds closely to the shapes predicted in the numerical simulations of de Young and Hundhausen (1971).