Abstract
Many difficulties face the conventional interpretation of the red shift of quasars as a Hubble shift, with associated immense distances. These objects are not of galactic size or nature, and are not associated with galaxies or clusters of galaxies. The continuing energy source for such enormous powers for a period of 106 to 107 years has not been clearly revealed. The absence of the expected absorption for the Lyman-α spectral line of hydrogen is a new difficulty. Because of the relativistic limit on the diameter which can produce rapid fluctuations of light output, there may not be enough surface to radiate the required light.A similar and perhaps more serious difficulty exists for the fluctuating radio output. Calculations given here for synchrotron radiation self-absorption lead to a reasonably accurate formula for the angular diameter of a radio source. For the quasar 3C 273B these relations indicate a conflict with the usually assumed distance. However, the discrepancy may be explained in terms of strong variation of radio diameter with frequency. For CTA 102 the conflict is more serious, and could be explained —for cosmological distance—only by rejecting the data of Sholomitskii. These difficulties are removed by the hypothesis that the observed quasars were ejected from a gravitational collapse at the center of our own galaxy, which may have occurred roughly 5 million years ago. The resultant distances, of the order of a million lightyears, reduce the energy problem by a factor of 106 or 107. On this basis the optical diameter would be less than a light-hour, about the size of the earth's orbit. A rotating mass of a few thousand solar masses with this diameter would account for the unusual line width, could easily produce the required radiated energy, and could readily account for observed short fluctuation periods and variations in spectrum. It is suggested that the radio output may be produced by high-speed passage of the quasar through intergalactic gas. This would probably correspond to a radio size of a few light-years or less, in agreement with the fluctuations. Since the radio power would be considerably less than that of radio galaxies, it is suggested that radio galaxies may have ejected groups of quasars. This would explain the peculiarly distant locations of the radio sources for many such galaxies. The objections to this model that have been raised are apparently not fatal. In particular, the receding hydrogen cloud discovered by Koehler to be in the line of sight to 3C 273 is more plausibly interpreted as having been ejected from our own galaxy, in the manner observed for other galaxies, than as being associated with the Virgo cluster of galaxies. The latter interpretation, which would place 3C 273 further away, is in conflict with Lyman-α absorption data for 3C 9 and other quasars. Thus the local model seems to give a reasonable explanation not only of quasars but also of radio galaxies, bothv of which seem largely to defy explanation on other grounds. Whether or not this model is valid, it is clear that an understanding of quasars will radically change our understanding of the universe.