2. Mass Transfer

The spectral classification of the companion star (O6-8f), the presence of residual emission at mid-eclipse, and the broad eclipse transitions indicate that there is a strong stellar wind. This wind reprocesses the emission from the compact source and would be expected to have a significant effect on the X-ray spectrum. The observed high state luminosity $\sim$ 10³⁸ erg s^-1 (van der Klis et al. (1980)) is too large to be accounted for by spherical wind accretion alone (Petterson (1978)). Thus, at least in the high state, most of the mass transfer must take place by another mechanism such as tidal enhancement of the wind or Roche lobe overflow.

Because of the high X-ray luminosity of the source it is expected that there is a region of extreme ionization surrounding the neutron star and extending to a distance greater than the radius of the companion (Blondin (1994b); Hatchett and McCray (1977)). This would rule out the presence of a normal line-driven wind from the region of the companion facing the neutron star. This is because highly stripped ions are transparent to the ultraviolet radiation that drives such a wind. Day and Stevens (1993) have suggested that when the source is in the high luminosity state the accretion is instead driven by a wind which is thermally excited by the intense X-ray flux. However a line-driven wind may still be driven from the part of the companion not illuminated by the X-ray source. This shadow wind will quickly become ionized when it is exposed to the X-ray source and it will cease to be radially accelerated by radiation from the companion (Blondin (1994a)). This stalled wind may be responsible for episodes of absorption detected around orbital phase 0.6 (Pounds et al. (1975))