The 35 day cycle of the X-ray binary Hercules X-1

The assumption that the clock of the 35 day period is a precession in HZ Her is tested. A model for the accretion region is constructed, based on this assumption, which consists of four distinguishable components: a tilted, twisted, optically thick accretion disk; an Alfven shell whose optical depth varies gradually with polar angle; a hot, optically thin corona; and a time-varying accretion stream between HZ Her and the disk. The twisted shape of the disk is calculated in detail, including the effects due to radiative pressure from the X-rays. The assembled configuration of disk, shell, corona, and stream is shown to provide an adequate qualitative explanation for practically all the 35 day related phenomena of the system. New explanations are suggested for the occurrence of hard X-rays in the middle of the off period, for the gradual X-ray turnoff, and for the 35 day variability of the optical pulsations. Arguments based on tidal evolution theory imply that a precession in HZ Her can only be present if a (so far unknown) mechanism has opposed the expected gradual decrease of obliquity in the system.