EXOSAT ME sample of dwarf novae

We present our analysis of the EXOSAT ME archival data on dwarf novae. Many dwarf novae were observed with EXOSAT; unfortunately, a significant fraction resulted in non- or marginal detection with the ME instrument. Our analysis of the the brightest dwarf nova, SS Cyg, shows little correlation between the count rate and the hardness ratio, contrary to an earlier report (King et al./ 1985, Nature 313, 290). This is likely to be the result of the significant improvement in the understanding of ME background since 1985. The apparent lack of correlation favors a compact boundary layer between the accretion disk and the white dwarf as the hard X-ray emitting region, rather than an extended ``corona.'' According to our spectral fit, all dwarf nova data are compatible with a single-temperature (2--->20 keV) Bremsstrahlung continuum with an iron emission line near 6.7 keV. However, several sources show significant epoch-to-epoch variability, the origin of which is poorly understood at the moment. Dwarf novae deserve a more sensitive scrutiny in the hard X-ray region, using forthcoming missions such as ASTRO-D. With this knowledge, we re-investigate the contribution of dwarf novae to the Galactic ridge X-ray emission. They are spectrally indistinguishable, an important reason for suspecting a strong connection. Using a reasonable estimate of the spatial distribution of dwarf novae, these systems appear to be a major, if not the dominant, component of the ridge emission.