We review the evidence for a general shift-distance law for galaxies. This is shown directly by a redshift-distance diagram recently prepared by Sandage and Tammann for Sc I galaxies. The distances are measured from H II region diameters and from the diameters of the Sc 1 galaxies themselves for the more distant systems. We also show the redshift-apparent magnitude diagrams for several objects which are identifiable over a large range in distance and which prove to have a small range in absolute magnitude: these are supernovae of type I (Kowal and Sargent, 1973, unpublished), brightest cluster galaxies (Sandage, 1972a), and radio galaxies (excluding N-type galaxies) (Sandage, 1972b). Plots of redshift versus angular size for brightest cluster galaxies (Sandage, 1972c) also have the correct slope for this to be a distance effect. In summary, the tightness of the correlations for the various diagrams listed implies that there is a very close correlation between distance and redshift for galaxies out to at least z = Δλ/λ~0.2 and that there is no evidence for redshift anomalies among the objects in these diagrams.

Evidence for non-velocity redshifts in quasars and galaxies is reviewed. It is shown that all current statistical tests favour the association of at least some quasars with relatively nearby galaxies. It is shown that a statistically significant number of quasars falling close to faint galaxies have intermediate redshift (between z = 0.4 and 1.8). This confirms the previous result that those are the intrinsically most luminous quasars and keeps open the possibility that less luminous quasars can appear projected at large distances around very nearby galaxies.

It is shown that in four individual cases where quasars fall, projected, closest to bright galaxies, the galaxies show evidence of physical interaction. New evidence for perturbations of the inner isophotes of NGC 4319 by Markarian 205 and by a radio source on the opposite side is presented.

Evidence for systematically higher redshifts for compact and peculiar companion galaxies is reviewed. The intrinsic redshift-morphology relation for clusters of galaxies is commented upon and the relation to Tifft's work is noted. As a summarizing diagram, the individual associations of high redshift quasars and companions are used to show an empirical continuity of observed characteristics between compactness (taken to be a measure of youth) and excess redshift.

Finally some possible theoretical explanations for intrinsic redshifts are mentioned.

Arguments are given to support the hypothesis that a characteristic point on a cluster luminosity function, m*, corresponds to the same absolute magnitude in all clusters, and thus is an appropriate standard candle for determination of relative cluster distances. It is shown, however, that if m*is a good standard candle, then the first and tenth brightest cluster galaxies, m(1) and m(10), respectively, are not, for m* – m(1) and m* – m(10) both differ significantly from cluster to cluster. Moreover, m(1) and especially m(10) depend on cluster richness. Distances derived from them for remote clusters chosen for analysis are thus subject to selection effects that depend on cluster richness and on Bautz-Morgan type. Finally, a procedure suggested by Bautz and Abell is described whereby m* can be estimated from observed properties of distant clusters.

Distance determinations of the high-redshift galaxy NGC 7319 in Stephan's Quintet, based on 21-cm line observations, yield a value much smaller than the cosmological distance. At the 98.7 % level, a part of the redshift of this galaxy which could amount to 5000 km s−1 is anomalous.

Previous work on the redshift-magnitude banding correlation is briefly reviewed. New tests of the concept are applied successfully to a second cluster (A2199) and the outer portions of the Coma cluster. Using more than 200 redshifts in Coma, Perseus, and A2199 the presence of a distinct band-related-periodicity in redshift is indicated. Finally, a new sample of accurate redshifts of bright Coma galaxies on a single band is presented, which shows a strong redshift periodicity of 220 km s−1. An upper limit of 20 km s−1 is placed on the internal Doppler component of motion in the Coma cluster.

Redshift-magnitude bands are, therefore, now recognized to consist of discrete ‘spin states’ organized into ‘spin groups’ which show strong morphological associations. Bands are probably in turn organized into band systems. The individual spin states are suggested to represent distinct configurations of matter at the nuclear or fundamental particle structure level. Transitions between states in a time systematic sense of decreasing redshift (increasing internal binding) are suggested to occur in galaxy nuclei. Energy released in such transitions may be the driving energy in radio sources and related objects.

For spiral galaxies the nuclear radio emission is usually dominated by a complex distribution of emission with median diameter of 200 pc and median power of 1019 W Hz−1 sr−1 at 1400 MHz. There is a large range in both power and diameter. The power is independent of morphological type for the normal spirals but is correlated with the absolute optical magnitude and with the infrared emission. For Seyfert galaxies the emission is generally stronger, in some cases by several orders of magnitude (e.g. NGC 1275, 3C 120).

Elliptical galaxies have been found with very compact radio sources, some less than a parsec in diameter. These are as powerful as the strongest spirals (~ 1021 W Hz−1 sr−1). Even stronger compact nuclear sources are now being found in the nuclei of those elliptical galaxies which also have extended radio sources (the radio galaxies). The presence of nuclear sources of this strength is so highly correlated with the presence of extended sources that this suggests a continuing involvement of the nucleus.

Several problems concerning the nuclei of galaxies are particularly interesting at present, either because new data have become available from advances in observational techniques, or else because recent observations suggest that earlier conclusions should be revised. These problems are discussed in this article.

Spectrophotometry in the near infrared shows that the late-type stars in the nucleus of M31 are mostly late giants; the M/L ratio in the nucleus is approximately 15, a value in agreement with that calculated independently from the velocity dispersion of the stars.

Infrared emission at 10µ has now been detected in 50 galaxies; the Seyfert-type galaxies are the brightest at 10 µ and there is evidence of variations of the 10 µ flux in NGC 4151, which rules out re-radiation by dust as the mechanism of infrared emission. In each of the galaxies NGC 253 and NGC 3034 an infrared source of dimension ~ 200 pc is present in the centre which coincides in position and dimension with the central component of the radio source at 1415 MHz; since both galaxies are rich in dust and in early-type stars it is plausible that, in these cases, re-radiation by dust is the origin of the infrared emission. Most of the other galaxies detected at 10 µ are spirals with intense and narrow emission lines.

Comparison of the optical and radio properties of nuclei of galaxies show that galaxies with compact nuclear radio sources are more likely to have optical spectra with emission lines than are galaxies without central radio sources. This correlation holds for elliptical and spiral galaxies. Its cause is not understood at present.

The relatively accurate profiles now available for the permitted lines of several Seyfert galaxies show that mass motions are the main broadening agent for these lines. Variations of line intensities have been observed in several Seyfert galaxies. The most likely explanation of these variations is that they are due to the variations of the photoionizing flux; in some cases, however, weakening of the permitted lines could be caused by high-velocity mass motions present in the region emitting the broad component of these lines.

From the investigation of the relatively nearby galaxies it is estimated that at least 5 % of galaxies have expansion motions or non-thermal optical fluxes.

This paper reviews current ideas about the possible constituents of galactic nuclei, and the mechanisms for ejecting gas and massive objects.

The isophotal surfaces of central bulges of normal spirals, often described as ‘spheroidal’, are shown to depart from ellipsoids of revolution and to resemble MacLaurin or Jacobi spheroids with zonal distortion; two typical examples, NGC 4565, 5746, are illustrated. Weaker examples are NGC 2683, 4594 and 891. Stronger examples are NGC 7332 and 128.

The true nucleus of an ordinary spiral is at most a very small object (~ 10 pc) resolved only in the nearest galaxies. The apparently larger nuclei seen in more distant spirals and lenticulars, in particular the so-called N-types, are probably artifacts due to insufficient resolution. It is possible that even in M31 the small ‘nucleus’ is merely the seeing-convolved image of the central peak in the r1/4 luminosity distribution of the spheroidal component.

In the nearest barred lenticulars and early spirals each large ellipsoidal ‘nucleus’ (~ 1.0 × 1.5 kpc) in the centre of the bar or central lens is found to include a second bar and inner nucleus on a smaller scale, in the approximate ratio of the corresponding Jeans' lengths. This is illustrated by high resolution photographs of NGC 1291; another example is NGC 1326.

Recent spectroscopic observations of the nucleus of NGC 1275, together with results from earlier workers, confirm the variability of the emission lines and indicate corresponding variations in the electron temperatures and densities of different zones. These changes may be associated with the microwave outbursts from this source.

Recent theoretical studies of the consequences of fierce tidal interactions between galaxies are reviewed and compared.

A southern sky survey of H I in the velocity range — 340 km s−1 to +380 km s−1 has shown that a long filament of H I extends from the Small Magellanic Cloud (SMC) region down to the South Galactic Pole and connects with the long H I filament discovered recently by Wannier and Wrixon (1972) and van Kuilenburg (1972). There is also some evidence that the feature continues on the other side of the Magellanic Clouds and crosses the galactic plane at l = 306°. The whole filament, which follows very closely a great circle over its entire 180° length, is given the name ‘The Magellanic Stream’. It may have been produced by gravitational interaction between the SMC and the Galaxy during a close passage (20 kpc) of the SMC some 5 × 108 yr ago although it is impossible to account for the observed radial velocities along the Stream unless some force other than gravity is invoked to act on the Stream as well.

Some results are discussed of recent – still largely unpublished – high-resolution studies of bright spirals, both in the radio continuum and in the 21-cm line radiation. Special emphasis is given to observations of M51, M81 and NGC 4258.

The most important new data are (1) Estimates of the contrast in gas density between arms and interarm regions (M81 and M101); the contrast appears to be quite strong. It provides an important parameter for the density wave. (2) Evidence for a pronounced decrease of gas density along the arms when these are followed towards the centre. Considerable line radiation is observed from outer regions where the optical intensity of the arms is very low. (3) Extension of rotation curves to larger distances from the centre. (4) Data on the motions in the spiral waves. (5) Determination of the shift between synchrotron and optical arms, providing direct evidence for the formation of stars as a consequence of the passage of the interstellar gas through the spiral density wave. From the shift measured in M51 the formation time is found to be roughly ten million years. If the data for (2) are ascribed to the depletion of the gas by star formation, the average net fraction of the gas consumed in star formation is found to be between 2% and 3% per passage through the spiral wave (Table I). (6) Separation of nuclear and disk synchrotron radiation in spirals. (7) Evidence for a recent expulsion of about 108 solar masses from the nuclear region of NGC 4258.

A review is given of recent theoretical and observational work on the density wave theory of spiral structure. Emphasis is put on the kinematic picture, and the question whether modern observations reveal the existence of density waves is discussed.

The particle resonance in a galaxy is of special interest because, according to the recent theory of Lynden-Bell and Kalnajs (1972), it provides the excitation of the density waves. In particular it is of interest to know if the density wave reaches the particle resonance, because the dispersion relation derived by Lin and Shu (1966) has a forbidden region if the parameter where (with α0 the surface density and κ the epicyclic frequency), is larger than one. If Q= 1, Lin and Shu find two waves going through the resonance but the amplitude tends to infinity there.

For spiral galaxies, investigations have been made as to the behaviour of density waves at the critical regions of the Lindblad and co-rotation resonances and in particular as to their role in the maintenance and generation of these waves. We shall also discuss the conditions under which an inner resonance ring may be formed. Furthermore, we find that the co-rotation region acts as a wave-amplifier.

Maps with 0′.5 resolution are presented of the distribution of neutral hydrogen in two galaxies. The barred spiral NGC 5383 contains much hydrogen, in strong differential rotation, in its outer parts; some H I concentrations with possibly anomalous velocities are observed in the regions of bar and nucleus. In the giant Scd spiral M101, the H I distribution corresponds closely with the optical spiral pattern, except for a lack of hydrogen in the central region.

Spectroscopic observations of the central part of NGC 4736 show strong noncircular motions connected with the triple radio source. It is also found that a central ring of H II regions is probably expanding at about 30 km s−1.

The luminosity of a spiral arm is believed to originate primarily in the very young, newly forming stars; and the spiral arm itself to be a spiral wave which is capable of triggering the formation of the young stars selectively along the wave crest. A semi-empirical study of the density wave patterns predicted in the density wave models of twenty-five external galaxies has been made and one result of this study is presented here. It is found that those galaxies of the sample whose models predict the possibility of strong shock waves are also the galaxies which exhibit long, well-developed spiral arms; and those galaxies whose models predict weak shock waves are also the galaxies which exhibit less-developed spiral structure. This trend is seen through a correlation between w⊥0, the velocity component of basic rotation normal to a spiral arm, which is an important parameter in determining the shock strength on the one hand, and luminosity class, which is a measure of the degree of development of spiral structure on the other.