Death of white holes

We critically review Eardley's argument which shows that, due to accretion, white holes are eventually converted into black holes---the death of white holes. While Eardley's main conclusions remain unscathed, we question some of the details of his analysis. We argue, together with Lake and Roeder, and Blau, that the death of white holes is caused by the gravitational effects of the accreting material, not by the blueshift instability of the Schwarzschild past horizon. We argue that it is not useful to discuss the death of white holes in terms of a time scale (which should be meaningful to external observers), and that it is preferable to discuss it in terms of the fraction of the white-hole mass which can be radiated to infinity. A rough measure of this fraction is M/mf, where M is the mass of the white hole, and mf the final mass of the system after accretion. We construct two different white-hole models, which illustrate the critical dependence of mf on the choice of model. It is shown that, in principle, mf can take any value between M (if there is no mass transfer between the cosmological and white-hole regions) and infinity (if there is an infinite mass transfer, which is possible in a flat or open universe).