The traditional approach to the extragalactic distance scale rests on a pyramid of primary, secondary and tertiary indicators of increasing range and decreasing accuracy. This multi-step procedure, fraught with the danger of cumulative errors, has led in recent years to two main, widely diverging scales: the ‘long’ scale (Sandage and Tammann 1974, 1975, 1976) implying a Hubble constant and the ‘short’ scale (de Vaucouleurs 1978, 1979; de Vaucouleurs and Bollinger 1979; de Vaucouleurs and Peters 1981; de Vaucouleurs et al. 1981) leading to H0 ≃ 100. Several authors have already shown that the long scale rests on very precarious foundations (Capaccioli and Fasano 1980, de Vaucouleurs 1981, Hanes 1980); counter arguments have been offered in its defence and to criticize the short scale (Tammann, Sandage and Yahil 1979, Sandage and Tammann 1982). For want of a conclusive test a consensus was still lacking.

The cataclysmic variables are close binary systems consisting of a late type star and a collapsed star, usually a white dwarf, undergoing mass exchange. According to the standard model, the late type star (the secondary) fills its Roche lobe and material escaping from the inner Lagrangian point is transferred to the primary by means of a mass transfer stream and an accretion disc. The spectroscopic and photometric properties of most cataclysmic variables can be understood in terms of radiation from the various components of such a system, with the accretion disc usually dominating in the optical region (see Warner (1976) for a review). However, recently a new class of cataclysmic variables has been discovered with distinctive optical properties that are inexplicable in terms of the standard model. These systems known as the AM Herculis type variables have provided the first direct evidence for the presence of strong magnetic fields in the white dwarfs of some cataclysmic variables. We present here a review of some of the important properties of these variables with emphasis on the unique system VV Puppis which has provided the first unequivocal evidence for high harmonic cyclotron radiation from white dwarfs.

This review surveys some recent observations of polarization in solar spectral lines with emphasis on their theoretical interpretation. Solar optical polarimetry is concerned primarily with measuring solar magnetic fields, a quest which began with Hale’s (1908) observation of Zeeman splitting in a sunspot. Today a wide range of optical polarimeters is used for solar research (see Bauer (1981) for a comprehensive review). For example, a standard magnetograph (Babcock 1953) detects circular polarization in selected wavelength bands in the wings of a Zeeman sensitive line, providing information on the line of sight or longitudinal field component. The most sophisticated instrument is a Stokes polarimeter (e.g. Bauer et al 1980, 1981) which measures with high spectral resolution the profiles of all four Stokes parameters (I,Q,U,V) across a spectral line. Such a device potentially provides all the data needed to infer the vector magnetic field.

Many excellent reviews of gravitational wave astronomy have appeared in the literature (see for example Thorne 1980 and references cited there). Much has also been written on the potential of the second generation of Weber-bar detectors which, operating at low temperatures and exploiting superconducting technology, have design sensitivities up to a million times greater than those of a decade ago. Indeed such has been the intensity of activity in this area, that even before this generation of cryogenic detectors has come of age, a third generation, quantum mechanical in outlook, is already being conceived.

In this contribution, which is an expanded version of an invited lecture at the 1982 A.G.M. at Noosa Heads, the author recalls some of the early work in radio astronomy from Dover Heights.

Since the work of Wu and Lee (1979) there has been renewed interest in the classical theory of electron cyclotron masers (Lee and Wu 1980, Lee et al. 1980, Wu et al. 1981, 1982, Hewitt et al. 1981, 1982, Melrose et al. 1982, Omidi and Gurnett 1982, Melrose and Dulk 1982). A useful idea in these recent developments of the classical theory concerns a geometric interpretation of the classical gyroresonance condition

where Ωe, is the nonrelativistic gyrofrequency, s = 0, ± 1, ± 2,… is the harmonic number, is the Lorentz factor and ║ and ┴ denote components parallel and perpendicular to the magnetic field. In v┴ − v║ space (1) represents an ellipse with centre v║ = vc, v┴ = 0, eccentricity e0 and semi-major axis V parallel to the v┴ axis, with

Observations of extremely high brightness temperatures in astrophysical objects imply that coherent emission processes must be occurring. An emission process may be coherent due to an inverted energy distribution leading to maser action, or due to bunching of particles in a region less than or of order of the wavelength of the radiation. Calculations of emission by bunches have generally used either a fluid model, e.g. Buschauer and Benford (1976), or a single-particle approach which requires all particles to have the same velocity, e.g. Saggion (1975) uses elements of both. In this paper we investigate gyromagnetic emission by bunches using a single-particle approach which includes the effects of differing particle velocities on the radiation.

The most obvious feature of the polarization of the radio emission from most pulsars is the rotation of the plane of linear polarization across pulses. The original interpretation of this in terms of the magnetic pole model (Radhakrishnan 1969, Radhakrishnan et al. 1969, Radhakrishnan and Cooke 1969) accounts for the variation of position angle extremely well for some pulsars (e.g. Manchester and Taylor 1977, Manchester 1978). Conversely, this provides strong support for the magnetic pole model for pulsar emission. It also suggests that the emission is basically linearly polarized as implied by virtually all proposed emission mechanisms, e.g. the reviews by Ginzburg and Zheleznyakov (1975) and Arons (1979). However, there are two features of the polarization which require a separate explanation. First, some pulsars have a moderately high degree of circular polarization, even in the integrated pulse profile (Manchester 1971, Lyne, Smith and Graham 1971). In some pulsars the average degree of circular polarization can exceed the average degree of linear polarization, e.g. in PSR 0835-41 and 0959-54 (McCulloch et al. 1978). Second, some pulsars exhibit the phenomenon of transitions between orthogonal elliptical polarizations (Manchester, Taylor and Huguenin 1975, Backer, Rankin and Campbell 1976, Cordes and Hankins 1977, Cordes, Rankin and Backer 1978). In many pulsars the orthogonal polarizations have substantial circular components, e.g. in PSR 1133 + 16 (Manchester et al. 1975) and PSR 2020 + 28 (Cordes et al. 1978).

In the usual model of discrete cosmic X-ray sources, it is assumed that a massive star loses matter to a companion neutron star, either through a stellar wind, or by means of Roche lobe overflow into an accretion disk around the neutron star (Pines 1980). The irifalling matter becomes ionized and is channelled along the magnetic field lines of the neutron star, so that it is accreted onto the star in a very small region at the magnetic polar caps. The physical conditions at these points are assumed to be: magnetic field B ~ 1012 G, plasma temperature ~107-108 K, and infall plasma velocity ~ 0.3 c-0.5 c.

In recent years, a number of numerical experiments have simulated various aspects of the early stages of star formation (see Tscharnuter 1980 for a discussion and review). In all but one of these experiments, however, the effects of the interstellar magnetic field have been neglected, although observations (for example, see Verschuur 1969) suggest that in some interstellar clouds, the magnetic energy is comparable to, or even greater than, the gravitational and thermal energies. It is also believed (Mouschovias 1981), that at least at the early, diffuse stages of collapse, where the ionizing radiation can penetrate deep into the cloud, the bulk of the neutral matter will feel the magnetic forces via collisional coupling with the ionized matter. Thus there exist no observational, nor theoretical reasons justifying the neglect of the interstellar magnetic field in these numerical simulations.

High-resolution observations indicate that very strong, small scale magnetic fields in the solar photosphere are concentrated into ropes which emerge through it. The scale of these ropes is only a few hundred kilometres across (Stenflo 1976) and their strength is estimated to vary between 1,000 and 2,000 G (Harvey 1977). These features are closely related to photospheric granular convection. The flux is observed as X-ray bright spots and sheet-like crinkles in the dark intergranular lanes, and is buffeted and shifted about by the granules (Dunn and Zirker 1973). Further, the crinkles can be resolved into separate features which outline small micropores as though a flux sheet at the end of a convection cell has separated into several isolated tubes (Galloway and Weiss 1981).

The polytropic stellar model with index n = 0 has a uniform density distribution throughout, and consequently its physical radius is essentially arbitrary because the surface density condition, ϱ = 0, is never satisfied. This surface anomaly, which is not associated with the other polytopic models for 0 < n ≤ 5, could be a constraining factor in certain astrophysical applications involving the n = 0 polytrope. For example, in some circumstances it may be appropriate to utilize the simple physical formulation of the model but on the other hand inappropriate to disregard any zero boundary requirements for the surface density. A sequence of new E-type (as defined below) composite analytical solutions to the Lane-Emden equation, based on the indices 0 and 1, has been developed which eliminates this physical indetermination. The associated polytropic models can be classified as essentially uniform density models. Specifically, they have a large central uniform density n = 0 zone matched, in a physically consistent way, to a small outer n = 1 zone which has a steep density gradient giving ϱ = 0, along with T = 0 and P = 0, at the radial distance corresponding to the first zero of the composite solution.

Recent numerical experiments (Norman et al. 1980), which simulate the axisymmetric collapse of a rotating, self gravitating cloud, show that spurious angular momentum transport can seriously affect the evolution of the cloud. In particular, it may determine if a ring-like density enhancement will occur. The spurious angular momentum transport can arise either from an explicit artificial viscosity, which might be required if shocks occur, or from an implicit viscosity due to truncation errors in the difference equation approximation to the exact equations. In donor cell schemes like those used by Tohline (1980) and Boss (1980) spurious angular momentum transport is due to truncation errors in the difference equations. For axisymmetric problems the errors are usually not serious since the typical length of a cell in the computational grid is very much less than the length scale of the cloud. We would expect the errors to be much greater when fragmentation occurs because the length scale of a fragment may only be comparable to that of three or four cells.

A coronal hole is a region of the solar corona characterised by diverging magnetic fields of single polarity and lower-than-average densities (and probably temperatures). It is now generally accepted that coronal holes are the source of high-speed streams in the solar wind (Munro and Withbroe 1972, Kopp and Holzer 1976, Steinolfson and Tandberg-Hanssen 1977, Munro and Jackson 1977; see Pneuman 1980 for references to most of the theoretical and observational papers on coronal holes since their recognition in 1968). We consider an infinitesimal field-aligned flow tube with cross-sectional area A(r) where r is the heliocentric radius (Figure 1), and we adopt the functional form

A(r)/A(r ) = (r/r0)8

where r is the coronal base radius, and 5 is a parameter which measures the divergence of the flow. If s = 2 then the flow is purely spherically symmetric, while if s > 2 the flow is more strongly divergent as is expected to be the case for a coronal hole. The cross-section of a typical coronal hole is shown schematically in Figure 2.

Ground-based observations of Type III bursts made with spectrographs and spectro-polarimeters, at frequencies above the ionospheric cut-off, reveal that most bursts (excluding storm Type IIIs) have fundamental (F) and harmonic (H) structure (Wild et al. 1959; Dulk and Suzuki 1980). An example of F-H bursts is given by Sheridan (1978). Such bursts are produced by streams of electrons travelling along open magnetic field lines and exciting plasma oscillations which are converted to electro-magnetic waves at both the F and H frequencies of the local plasma frequency in the corona.

We describe a method for reducing displacements in radio source positions caused by ionospheric refraction. Our method is an improvement on that used by Wild et al. (1959), who assumed all frequencies in the 40-70 MHz band came from the same position above the Sun and applied an f-2 correction for ionospheric refraction. Our method retains the dispersion of source position with frequency, which is inherent in the radio source, but allows for the f-2 ionospheric effect. We also discuss ways of estimating the absolute source positions from a knowledge of ionospheric density gradients. A typical example of ionospheric variations on solar records is shown in Figure 1(a). Here we have plotted the observed source positions on an f-2 scale. The measurements refer to a solar storm continuum burst which occurred on 1981 May 9 following an importance 2B flare at 08°N., 38°E., heliographic coordinates. From previous observations of such events we think it highly likely that the true source positions are stationary and displaced with decreasing frequency outwards along a line close to the radial direction above the flare. Figure 1(a) shows that: (a) the 160, 80 and 43 MHz source displacements vary as f-2; (b) there is a systematic slow drift of the source positions towards the south and east which increases with increasing hour angle; (c) superimposed on this steady drift is a quasi-periodic variation in source position with a period ~20 min.

A type of metre-wavelength continuum is observed in close association with solar flares and Type II radio bursts. This continuum is stationary in position, often intense, of a long duration (occasionally lasting well over an hour) and highly polarized (Robinson and Smerd 1975; Robinson 1978). It is thought to be produced by the conversion of Langmuir waves to electromagnetic waves near the local plasma frequency. Because of its close association with Type II bursts, the class has been designated as Type II related flare continuum, or FCII. In this paper I examine a particularly clear example of such an event, with emphasis on the relation between the Type II and the start of the continuum.

A Type II solar radio burst is a relatively narrow-bandwidth metre-wavelength emission which drifts outward in the solar corona at a velocity between 500 and 1500 km s-1. It was first described by Wild and McCready (1950) and since then it has been the subject of numerous investigations (see e.g. McLean 1974; Nelson and Robinson 1975).

Duley (1974) has shown that, at the temperatures usually associated with interstellar gas clouds, we would expect CNO grains to be present. During gravitational collapse these grains migrate to the centre of the gas cloud, leading to an enhancement of the heavy-element abundance in the core (Prentice 1976, 1978). It was Krautschneider (1977) who verified such a scenario, by considering the dynamical collapse of gas and grain clouds. If such an initial radial abundance inhomogeneity existed, Prentice (1976a) showed that this configuration may well survive the later convective mixing phase and thus approach the zero-age main-sequence (ZAMS) with a small (-v 3% by mass) metal enhanced core.

Evidence exists for mode energy changes in beat Cepheid variable stars over the 20-30 year interval during which photoelectric observations of the objects have been obtained. For TU Cas, Hodson, Stellingwerf and Cox (1979) have found that the first overtone amplitude has been decreasing with time from 0.4 to 0.25 mag over a 67 year timebase; these authors also included some early visual estimates in their analysis. For U TrA, Faulkner & Shobbrook (1979) have found that the first overtone pulsation has been increasing relative to the fundamental over a twenty year timebase.