Problems in the astrophysics of accretion onto compact celestial bodies

Although during the last decade new observations and new theoretical results have brought better understanding of the physics of accretion onto compact objects, many old and several new questions and problems await answers and solutions. I show how the disc thermal-viscous instability model applied to both cataclysmic variable stars and X-ray binary transients compels us to conclude that assuming the existence in these systems of a flat accretion disc extending down to the accretor's surface or to the last stable orbit and fed with matter at its outer edge is too simple and inadequate a description of these objects. It is also clear that, in most cases, these discs cannot driven by (anomalous) viscosity only. The origin of the superhumps observed in cataclysmic variables and X-ray binaries is, contrary to the common opinion, still unknown. In accreting magnetic white dwarf systems outbursts not of the dwarf-nova type can be due to the magnetic gating instability and/or thermonuclear micronova explosions. Although the "typical" lightcvurves of X-ray transients can be described by analytical formulae (but their decay phase is not exponential), observations show that in many cases the light variations in these systems are much more complex. An elementary argument shows the impossibility of magnetars in pulsing ultraluminous X-ray systems, but we still do not have a complete, self-consistent description of supercritical accretion onto magnetized neutron stars and the resulting (necessarily beamed) emission. Although it is (almost) universally believed that active galactic nuclei contain accretion discs of the same type as those observed in binary systems, the evidence supporting this alleged truth is slim and the structure of accretion flows onto supermassive black holes is still to be determined.
Comment: Invited book chapter in "Highlights of the Compact Objects Sciences in the Last Decade", edited by Solen Balman, IU Press (Istanbul University Press, Turkey) Errors corrected, text added, Fig. 1.6 replaced/corrected