1. Motivation

My study of these data was motivated by the tentative identification by Nagase et al. (1992) of the high energy cut-off in Cen X-3's spectrum with cyclotron opacity. It seemed that, if a CSRF was detectable, pulse phase resolved measurements would help distinguish between a slab and a column emission geometry which cause radiation to be emitted as pencil or fan beams, respectively (see Figures 4.3 and 4.3). The energy of a cyclotron absorption feature is expected to be greatest when the angle between the line of sight and the magnetic field is zero (Mészáros and Nagel (1985a)). Thus for a pencil beam the CSRF energy should be greatest at the pulse maximum (when the line of sight is closest to the magnetic field direction). For a fan beam it should be at a minimum at the pulse maximum (when the line of sight is normal to the magnetic field). Comparing their calculations with observations of the CSRF in Her X-1 Mészáros and Nagel (1985b) concluded that the pulsar radiated as a pencil beam. For luminous pulsars such as Her X-1 and Cen X-3 it is expected that there is a radiation deceleration shock above the surface of the neutron star producing a column geometry and a fan beam (e.g. Wang and Frank (1981)). In the case of 4U 1538-52, Bulik et al. (1992) modeled the pulse phase dependence of the spectrum. They obtained quantitative agreement with pulse phase resolved observations of the cyclotron line made with Ginga and found evidence for differences in temperature and opening angle for the polar caps.