Conclusions

Although the inverse-Compton model accounts for the luminosity of high energy photons from LSI+61 , and the low ratio of x-ray to -ray emission, it is difficult to reconcile with the spectral properties of the keV x-rays. Below the low-energy break in the inverse-Compton spectrum, the luminosity should decrease with decreasing energy, producing a hard keV spectrum, in contrast to the observed spectrum which is very soft. The keV spectrum is much more readily explained by the shock model. We are naturally lead to an explanation for the high energy photons from LSI+61 in which the keV x-rays are produced in circumstellar gas shocked by the expanding plasmon of relativistic electrons and the -ray emission arises from inverse-Compton scattering of stellar photons off those relativistic electrons with energies greater than . High spectral resolution x-ray observations to search for thermal line emission, and time monitoring of the -ray flux to search for periodic variation in phase with the average radio light curve would provide straight-forward tests of this model.

If the -ray luminosity is produced by Inverse-Compton scattering, a particle energy injection rate of erg s is required. This energy rate could be produced by accretion onto a neutron star or by a relativistic wind from a rapidly spinning, young pulsar. Our data do not allow us to discrimate uniquely between these possibilities. However, the lack of a high luminosity outburst at keV energies, as is observed in other accretion-driven x-ray binary systems, argues against the accretion scenario for LSI+61 .

This research was supported by a grant from the Natural Sciences and Engineering Research Council of Canada. M. Peracaula acknowledges financial support from the Ministerio Espaol the Educación y Ciencia and partial support by CICYT(ESP93-1020-E) and DGICYT(PB91-0857). The National Radio Astronomy Observatory is operated by Associated Universities Inc., under contract with the U.S. National Science Foundation.