A MAGNETOSPHERIC HEATING MODEL FOR THE EVAPORATION OF THE COMPANION TO PSR 1957+20

An orbital period derivative for the PSR 1957 + 20 system has recently been reported by Ryba & Taylor which indicates orbital contraction over a time-scale of 10(15) s. The angular momentum loss of the light companion is probably indicative of heavy mass loss, too high to be explained by previous evaporation scenarios. In addition, Fruchter & Goss have reported an eclipse in the time-integrated component of the pulsar emission at 1400 MHz at a phase of 0.23 to 0.24, slightly before the middle of the eclipse at lower frequencies, implying that the wind is shot out asymmetrically from the companion star's surface. In this letter we suggest that the companion has a magnetosphere which allows for extremely efficient and possibly uneven heating of the companion's surface, and partial collimation of the outflow.