The 35 day evolution of the Hercules X-1 pulse profile:: Ginga observations and their implications

We observed Her X-1 using the Ginga observatory in the spring of 1989 with the primary intention of studying the evolution of the pulse profile through the course of the 35 day X-ray HIGH-LOW cycle in that source. These observations cover 16 separate days in two MAIN HIGH states and in the intervening SHORT HIGH state. We have augmented these data with four additional Ginga observations of Her X-1 taken for other purposes but useful for our study. We present light curves in the 1-37 keV energy band for the seven high states covered by these data together with a representative sample of pulse profiles. The signal-to-noise ratio for these profiles is generally excellent, and collectively they provide a sound base for studying the evolution of the pulse profile. Of particular utility is the 1989 May observation, which for the first time provides extensive coverage of a short high state at high photon counting rate. By combining pulse phase and frequency information from all three high state observations in 1989, we are able to determine the phase alignment of main high and short high pulse profiles with high confidence. We identify components in the Her X-1 pulse profile by their distinctive spectral signatures, and we establish the existence of a definite, repetitive pattern of pulse shape variations tied to the 35 day high-low cycle. Comparing pre-Ginga observations of pulse profiles superposed according to the 35 day phase indicates that this pattern has persisted over the past two decades. Moreover, the pulse phase alignment of the 1989 data allows the identification of components common to the main high and short high pulse profiles. One of the key elements of this pattern is the rapid change in pulse profile that occurs roughly 7 days into each main high state. In examining possible mechanisms for this interval of accelerated evolution, we are led to consider a class of models that involve dynamical changes in accretion flow geometry arising from neutron star obliquity. These models, however, suggest asymmetries in the X-ray illumination of the companion star that may conflict with extensive optical observations of the Her X-1 system. By contrast, kinematic changes in the geometrical aspect of obscuring matter flows near or within the magnetosphere that are tied to accretion disk precession provide a mechanism that may be consistent with observations.