Annu. Rev. Astron. Astrophys. 1980. 18: 165-218 Copyright © 1980 by Annual Reviews. All rights reserved

4.2. Host Spots

As we have seen in Section 2.2.1, hot spots occur at the outer edges of the most luminous sources and are sometimes seen within the diffuse lobes of the weaker ones.

4.2.1 PROPERTIES Hot spots have sizes that are typically a few kpc (see, for example, Readhead & Hewish 1976, Kapahi 1978). Hot spots in opposite lobes are usually collinear with the central radio core (Section 2.1.1). There are suggestions from the scintillation data that hot spots occur less frequently in sources whose overall sizes are larger than ~ 200 kpc (Readhead & Hewish 1976), or alternatively that the hot spots in the larger sources are bigger.

Hot spots have spectra similar to or slightly flatter than the surrounding diffuse emission (e.g. Hargrave & Ryle 1976, Jenkins & Scheuer 1976, Gopal-Krishna 1977, Gopal-Krishna & Swarup 1977, Burch 1977a, Spangler & Meyers 1978, Högbom 1979, Burch 1979b). Their integrated polarizations at frequencies above 1 GHz are typically 20-30% (see Hargrave & Ryle 1974, Strom & Willis 1979). Circumferential magnetic fields are indicated (see, for example, Dreher 1979) with the average direction of the magnetic field usually oriented roughly perpendicular to the overall extent of the radio component.

Although the presence of hot spots is clearly connected with the total radio-source luminosity, studies of the variation of hot-spot properties with luminosity are difficult to carry out. Such studies are largely confined to the most luminous and distant sources for which cosmological effects are important. The best-established correlation is undoubtedly the increase with luminosity of the separation of the hot spots from their parent nuclei (Section 2.2.3). Less credence should be given to the report that the average hot spot size increases with luminosity (Readhead & Hewish 1976). The initial interpretation of the (scintillation) data on which this report was based has been contradicted by high resolution interferometry (Kapahi & Schilizzi 1979).

During the last few years there have been several searches for optical emission from hot spots. Evidence has been found for such emission in 3C285 (Tyson et al. 1977), in 3C265 and 3C390.3 (Saslaw et al. 1978), in 3C33 (Simkin 1978), and in NGC 7385 (Simkin & Ekers 1979). The observed radio polarizations of the hot spots would argue against a thermal origin for the optical emission (Saslaw et al. 1978), although in the case of 3C33 Simkin has claimed that there may be weak emission lines present. The most plausible mechanisms for producing the optical emission would be direct synchrotron radiation or inverse Compton scattering of the microwave background by the electrons that produce the radio emission (Saslaw et al. 1978).

4.2.2 ORIGIN The collinearity of the hot spots with the nucleus in some double radio sources (Section 2.1.1) indicates that hot spots play a fundamental role in the radio-source phenomenon. It has been suggested that they are (a) regions where acceleration of electrons to relativistic energies occurs through energy pumped in from the nucleus (e.g. Rees 1971), (b) places where the bulk kinetic energy of the relativistic electrons created in the nucleus is thermalized to produce visible radiation (e.g. Blandford & Rees 1974), (c) the accumulation of plasmons, multiply ejected from the nucleus and decelerated by ram pressure (Christiansen et al. 1977), or (d) the location of compact supermassive objects ejected from the nucleus which produce and accelerate the relativistic electrons through interaction with circumgalactic gas (Saslaw et al. 1974).

In the first two cases the hot spots represent the end of a beam and they move outwards from the nucleus as the beam rams through the ambient medium. Although the hot spots need not be strictly "confined" in such a situation their motion will be governed primarily by ram pressure forces (e.g. Blandford & Rees 1974), and from the arguments of Section 3.1.2 an external medium with a density of _ext 10^-28 gm cm^-3 is needed. This lower limit was calculated using an upper limit of 0.1 c for the outward velocities of the hot spots, since for much greater outward speeds relativistic effects would result in a higher proportion of very unequal doubles than is observed (Mackay 1971, Katgert-Merkelijn et al. 1980, Longair & Riley 1979). Similar confinement arguments apply to the multiple-plasmon models.

To hold the hot spots together by gravitation (Burbidge 1967, Callahan 1976, Flasar & Morrison 1976) requires compact objects of mass 10⁹ M. This is unlikely. First, there is no evidence for structure in the hot spots on a scale much smaller than a kiloparsec. Second, in order to explain the observed alignment of the extended emission with the cores (Section 4.3.3), and with the optical galaxies and quasar polarizations (Sections 5.3), a polar ejection mechanism for the multiple ejection of compact objects from the nucleus is required. The slingshot process (Saslaw et al. 1974), the only mechanism to have been proposed for such ejection, is equatorial and therefore does not preserve memory of direction. Moreover it has difficulty in explaining the detailed brightness distributions of head-tail galaxies (see, for example, Downes 1980). Attempts to reconcile this theory with observations (e.g. Valtonen 1977) appear contrived.