The observational data on the radio properties of compact radio sources are presented, including their radio spectra, their occurence, and their structures, as well as the variations of the structures in the “superluminal” sources.

The radio structures mapped by aperture-synthesis techniques with the VLA, and discussed earlier at this conference by Fomalont, are energised by activity in galactic nuclei. The compact radio sources resolvable by VLBI techniques (Pauliny-Toth, these proceedings) are on scales ~105 times smaller; but even this still far exceeds the scale where - according to most theoretical ideas - the primary power production is concentrated. This central “core” - involving a large concentrated mass, probably a black hole of ~ 108 M⊙ accreting material from its surroundings - is the primary origin of the power for all categories of activity associated with active galaxies and quasars. In this paper, I shall principally discuss how this “core” could give rise to the fast-moving plasma which emerges in directed beams and energises the radio sources; I shall also comment on some interesting features of radio source structure.

Plerions, i.e. supernova remnants resembling the Crab Nebula, are characterized at radio wavelengths by having: 1) a centrally peaked brightness distribution, 2) flat radio spectrum (α > −0.3, Sν να), 3) high linear polarization, and 4) a highly ordered magnetic field. At higher frequencies (1011–1013 Hz depending upon the age) the spectrum turns over and attains a slope of α ~ −1. In particular, for the Crab Nebula the turnover frequency occurs at νc ~ 3×1012 Hz (cf. Fig. 1). Moreover, the optical spectrum displays a slope of −0.9 and the radiation is linearly polarized. Nonthermal emission is also detected in the X-ray and γ-ray domains (Fig.1). Similar spectral characteristics are found for other bona fide plerions. As discussed by Weiler and Panagia (1980), the plerion phenomenon is determined by the presence of a highly energetic and active central object (a fast spinning neutron star) which is able to both accelerate and inject continuously relativistic electrons into the remnant with a typically flat energy distribution roughly proportional to E−1.

From radio continuum observations it is known that the central regions of galaxies can be an important source of cosmic rays, in particular relativistic electrons. In order to get some insight in the properties of the central sources and their dependence on optical properties, especially morphological type, a large sample of galaxies (~ 400) was observed with the Westerbork Synthesis Radio Telescope at 1.4 GHz. The sample definitions, observations, reduction, survey characteristics and the analysis of the data are given in Hummel (1980, 1980a). Here we will concentrate on the results obtained for the central sources (> 20″).

We refer to a model of active galactic nuclei where a massive black hole (M≃108M⊙) is accreting spherically (1–3). Turbulent dissipation of the magnetic field produces an efficient heating. For low Thomson optical depth the main cooling mechanism is selfabsorbed cyclo-synchrotron (c-s) radiation from thermal electrons. The energy transfer of the thermal electrons to the c-s photons via Compton scattering (comptonization) becomes increasingly important for increasing optical depth. In ref. 2 the temperature profile of the inflowing gas was calculated balancing the heating and cooling mechanism but neglecting comptonization due to the difficulty of an analytic estimate of this process.

The physical conditions in the centre of spiral galaxies as our own galaxy still are not very well known. They offer themselves to more or less exotic speculations, some of which involve the possible existence of black holes and mechanisms for the production of high energy cosmic rays. In view of such argumentations I feel it worthwhile to look in more detail into recent progress obtained in the understanding of the dynamics of the central region of spiral galaxies which turns out to be intimately related to the galaxy as a whole, especially its spiral structure.

In the context of the proposed DUMAND-type detector for high-energy neutrinos, it is important to estimate the neutrino fluxes from powerful radio galaxies. This would indicate whether a detectable event rate can be expected, and would help choose among possible mechanisms for generating the radio lobes associated with many such galaxies. Among the models that have been proposed are: (a) a beam from the galactic core that interacts with the extragalactic medium; (b) ejection of clouds of plasma, particles, and magnetic fields into the lobes, either in an explosive burst, or by diffusive escape; (c) black holes or spinars ejected from the galactic nucleus by a gravitational slingshot mechanism.

The evidence that pulsars accelerate relativistic particles is reviewed, with emphasis on the γ-ray observations. The current state of knowledge of acceleration in strong waves is summarized, with emphasis on the inability of consistent theories to accelerate very high energy particles without converting too much energy into high energy photons. The state of viable models for pair creation by pulsars is summarized, with the conclusion that pulsars very likely lose rotational energy in winds instead of in superluminous strong waves. The relation of the pair creation models to γ-ray observations and to soft X-ray observations of pulsars is outlined, with the conclusion that energetically viable models may exist, but none have yet yielded useful agreement with the extant data. Some paths for overcoming present problems are discussed.

Among the COS-B Galactic γ-ray sources two have been identified with the young pulsars PSR0531+21 and PSR0833–45, while for most of the others there are no convincing counterparts. An obvious explanation is that at least part of them are also pulsars. Because of the rarity of pulsar formation, an important issue for testing this possibility is to discuss whether relatively old pulsars (~105y) may be sources of γ-rays of the luminosity observed by COS-B (φ (100 MeV) 10−6cm−2s−1). On this line it is of great interest the claim of observation of γ-rays with SAS II from PSR1747–46 and with COS-B from PSR0740–28 and PSR1822–09, communicated at other conferences, but yet not confirmed. The three slow pulsars mentioned above were reported to have γ-ray luminosities of 1033 erg s−1 close to their intrinsic energy loss LT = Iωω. Their periods are about 0.1–1s and their distance ~1 Kpc.

Young pulsars apparently have a distribution of initial power outputs N (> Po−γ), with 1/2 < γ < 1 and Po ≳ 1038 ergs/sec. Assuming that ultra-high-energy (E ≳ 1015 eV) cosmic-ray nuclei are accelerated at the central pulsar, a young, dense supernova shell can be a powerful source of high-energy neutrinos. With an optical array placed in a volume of one km3 at great ocean depths, as proposed for the DUMAND detector, it is likely that ≳ 103 hadronic and electromagnetic cascades induced by neutrinos would be recorded for a stellar collapse within our Galaxy. Such supernovae occur about 8 times per century. Neutrinos from young supernova shells in the Virgo supercluster would be marginally detectable via neutrinos with N(> Po) ∝ Po−1/2, but unobservable if N(> Po) ∝ Po−1.

The properties of the radio continuum emission from spiral galaxies have been reviewed by Van der Kruit and Allen (1976) and by Van der Kruit (1978). In more recent years the major developments in the understanding of the radio continuum properties and the underlying physical conditions of galaxies have come from a number of surveys of large samples of objects. Some of these surveys (e.g. Hummel, 1980a) have good sensitivity and sufficiently high angular resolution to allow for the first time a clear separation of central sources and disk emission and a study of the properties of these components in a large number of galaxies. As a consequence some results already found, suggested or only suspected in previous detailed investigations of a limited number of objects are put on a firmer basis or entirely new aspects are revealed.

Size and shape of the “confinement volume Vo”, the region around our Galaxy where the c.r. particles live until they escape or loose completely their energy are essential parameters of models for the origin and propagation of the cosmic rays. Focussing the attention on the c.r. electrons, whose sources are definitely galactic and concentrated in the Disk, to confine Vo to the Disk or extend it to the Halo is full of consequences. For instance in the first case the sources can be everywhere inside Vo; by contrast in a Halo model the sources can reside only in some parts of Vo, the mechanisms of energy loss play different roles in different regions of the confinement volume and the particle lifetimes can be up to 100 times larger than in the Disk, (e.g. ref. 1).

The 408 MHz All-sky Survey has been made from four radio continuum surveys observed between 1965 and 1978, using the Jodrell Bank MKI telescope, the Effelsberg 100 metre telescope and the Parkes 64 metre telescope. A detailed description of the survey data reduction and calibration methods, with preliminary astronomical results will soon be published and a second paper will give an atlas of maps at the full survey resolution of 51′ arc between half power points.

The all-sky radio continuum map at 408 MHz presented at this symposium by Haslam et al. can be interpreted in terms of the large-scale 3-dimensional distribution of synchrotron emissivity in the Galaxy when due allowance is made for the thermal emission. Its derivation from a 2-dimensional map must involve a number of assumptions so it is instructive to compare the results of alternative approaches (described in detail in forthcoming papers by the present authors). In both cases the variation of emissivity in the galactic plane is obtained from the observed intensity profile at b=0o and then the z-variation is chosen to give the best fit to the complete map. The observed profile is shown in the figure with and without the contributions of catalogued supernova remnants and HII regions.

High resolution observations of the radio continuum emission from nearby galaxies at several frequencies provide information about the cosmic ray electrons. Optimum results are expected by combining low-frequency synthesis observations with high-frequency single dish data, e.g. Westerbork 610 MHz / Effelsberg 10.7 GHz (1′ resolution) or Cambridge 150 MHz / Effelsberg 2.7 GHz (4′ resolution). Maps of 15 nearby spiral galaxies at 10.7 GHz have been made with the Effelsberg 100-m telescope. Maps of M31 and M33 are available at 2.7 and 4.8 GHz.

The distribution of interstellar nucleons is dominated by gas at temperatures ranging from a few K to a few hundred K. At the warmer end of this temperature range, the gas is predominantly in the form of ubiquitously distributed atomic hydrogen. The colder gas is almost entirely molecular; it resides in compressed clumps, confined principally to the inner-galaxy, and is most effectively traced by observations of carbon monoxide. In this paper, we focus on some of the problems which currently hinder derivation of the morphology and total number of nucleons in the galaxy. For the atomic gas, these problems involve optical depth effects in HI profiles, the amount of cold HI residing in molecular clouds, and the form of the outer-galaxy rotation curve. For the molecular gas, the problems involve the uncertainties in the conversion from CO intensities to H2 densities, including the possibility of composition gradients across the galaxy, the total number and typical size of molecular clouds, and the possibility that the molecular material in the region of the galactic nucleus is distributed differently from the material in the galaxy at large.

It is well known that, when cosmic rays stream along the field lines of the galactic magnetic field at a velocity which exceeds the Alfven velocity in the medium, they excite hydromagnetic waves of a wavelength comparable to their Larmor radius. These waves, in turn, scatter the cosmic rays, forcing them to reduce their bulk speed (Wentzel 1974, and references there-in). In a stationary situation, the bulk speed of the cosmic rays will depend on their scale height, and on the strength of the relevant damping mechanisms affecting the waves. Isotope observations imply that cosmic rays of energy ≲ 1 GeV are confined in the galaxy for a time ~ 2.107 yrs (Garcia-Munoz et al. 1977); a simple interpretation of the composition data in the energy range 1–100 GeV implies that the confinement time T decreases as the rigidity ε = cp/eZ increases: Tα ε−a, with a = 0.3 − 0.5. This energy dependence of the confinement time should hold at least up to ε ≃ 3.106 GV, to be consistent with the lack of structure in the cosmic ray spectrum in the range 10–3.106 GV (see Cesarsky 1980 and references therein).

Parker (1966, 1969) proposed that the interstellar medium could be subjected to instabilities, able to destroy its structure in a time shorter than the galactic evolution time. He emitted the idea that this phenomenon could lead to halo formation, escape of magnetic field and cosmic rays from the galaxy, via some sort of interstellar “bubbles”, and also to cloud formation. He showed then that if the interstellar medium was well represented by a simple horizontal stratified equilibrium model, perturbations would effectively grow and destroy the structure.

The existence of hydromagnetic waves (waves whose frequency ω is less than the ion gyrofrequency Ωi = eB/mic) in a collisionless magnetized plasma with β, the ratio of plasma pressure to magnetic pressure, much greater than unity is required in theories for Fermi acceleration of cosmic rays by converging scattering centres at a shock front, in theories for the adiabatic cooling of cosmic rays due to trapping by plasma instabilities in an expanding supernova remnant (Kulsrud and Zweibel 1975, Schwartz and Skilling 1978) and in theories for resonant scattering of cosmic rays by hydromagnetic waves in the hot phase of the interstellar medium (Holman et al. 1979). Hydromagnetic waves may be damped by thermal ion cyclotron damping for wavenumbers k≳Ωi/vi, where vi = (Ti/mi)1/2 is the average thermal ion speed, and by linear Landau damping for non-zero angles of propagation with respect to the ambient magnetic field (Foote and Kulsrud 1979). Damping by both these processes is strong in a high-β plasma where there are many particles travelling at the phase speed of the waves. Hydromagnetic waves propagating along may be damped by nonlinear wave-particle interactions, the most important of which is thermal ion Landau damping of the beat wave of two Alfvén waves. This nonlinear process has the effect of transferring energy from the waves to the particles and can therefore be considered as a damping process for the waves.

Secondary antiprotons are a potentially interesting probe of cosmic ray propagation because their production cross section is strongly energy-dependent, increasing by more than two orders of magnitude between 10 and 1000 GeV/c. This is quite unlike the case for fragmentation cross sections of complex nuclei, which are virtually constant with energy. Moreover, the flux depends primarily on the environment seen by protons which need not be identical to that probed by other nuclei.