Annu. Rev. Astron. Astrophys. 1999. 37: 603-648 Copyright © 1999 by Annual Reviews. All rights reserved

7.3. Resolved UV Star Populations

The HST has sufficient sensitivity to probe directly the UV star populations of nearby galaxies. Three UV imaging studies of hot stars in M31 and M32, all based on FOC observations of small fields, have appeared to date (King et al 1992, Bertola et al 1995, Brown et al 1998a). Although the two earlier studies suffered from serious calibration difficulties (see Brown et al 1998a), all three detected UV-bright stars and agree that luminous PAGB stars cannot account for more than a small fraction of the total FUV light. The photometry of Brown et al (1998a) has a detection limit of m(1750 Å) ~ 24.5, which is not deep enough to reach the HB itself but does encompass PAGB, PEAGB and AGB-manqué luminosities. Brown and colleagues identify a large number of stars consistent with expectations for the descendants of EHB stars with M_ENV ~ 0.002-0.05 M. However, most of the UV light is produced by unresolved stars, presumably on the EHB. The integrated HUT or IUE spectra are consistent with models, normalized by the resolved samples, in which only 2% of the total population passes through the EHB channel in M31 and only 0.5% does so in M32. The lifetime of the PAGB channel, which makes up the rest of the post-HB population, is so short that few resolved objects are expected in the observed fields. Somewhat unexpectedly, the shapes of the luminosity functions for the resolved stars in M31 and M32 are similar (despite significant differences in optical absorption line spectra). M32 differs only in having fewer total stars (per unit UV surface brightness) above the detection threshold. Brown and coworkers also find that about 10% of the brighter resolved population is not explainable by existing post-HB evolutionary tracks.

Any process that reduces HB envelope masses in a significant fraction of the population can have influence extending beyond the UV region. EHB stars do not become AGB stars, and if most of the evolving population passes through the small-envelope channel, the AGB contribution to the integrated optical/IR spectrum (mainly longward of 6000 Å) of a galaxy will decrease. Changes in the AGB should also be detectable with the surface brightness fluctuation imaging method (Tonry & Schneider 1988). Ferguson & Davidsen (1993) find that the incidence of planetary nebulae, which should also decrease if the AGB population decreases, is anticorrelated with bluer 1500-V colors. This is important circumstantial evidence that small-envelope HB stars are implicated in the UVX. The correlation should be pursued with a larger sample of galaxies, and radial dependences within galaxies should be studied as well.

The existing deep imaging studies therefore provide good support for the EHB interpretation of the UVX. Imaging to the level of the EHB itself within the Local Group can probably be secured with HST/STIS and HST/ACS. Worthey (1993) has described how the surface brightness fluctuation technique can be applied to faint hot stars to extend the effective depth of such UV imaging.