The thermal expansion of alkali halides at low temperatures - II. Sodium, rubidium and caesium halides

An investigation of the number-intensity distribution of the observed radio stars shows that they cannot be interpreted in terms of a homogeneous distribution of sources; there is an apparent increase in the spatial density or absolute luminosity of distant sources. It is also found that this increase is, within rather small statistical errors, independent of direction. It therefore appears that the solar system is situated at the centre of a spherical region in which the spatial density or luminosity of the sources increases with distance, uniformly in all directions. It may also be concluded that this increase does not persist for distances much greater than those to which sources have been observed individually. It appears impossible to explain the results in terms of sources situated within the galaxy, or by irregular clustering of extra-galactic sources. Furthermore, it may be shown from arguments based on the integrated radiation that the minimum radius of the spherical region is comparable with that of the optically observable universe. It is therefore suggested that the observed features are due to effects on a cosmical scale; such an explanation provides an immediate reason for the remarkably isotropic distribution of radio stars. Attempts to explain the observations according to steady-state theories offer little hope of success, but there seems every reason to suppose that they might be explained in terms of an evolutionary theory. No detailed comparison has yet been made between the observations and the predictions of any particular evolutionary model, but a general explanation of the observations seems possible if it is supposed that the majority of radio stars belong to an excessively rare class of object, having a local spatial density of about 2 $\times $ 10$^{-26}$ parsec$^{-3}$, and an absolute luminosity comparable with that of the intense radio source in Cygnus. On this interpretation, the difficulty of identifying radio stars with optical objects finds a natural explanation; only some tens of the main class of radio star would be within reach of the 200 in. telescope.