One of the main challenges facing standard hierarchical structure formation models is that the predicted abundance of galactic subhalos with circular velocities of 10-30 km/s is an order of magnitude higher than the number of satellites actually observed within the Local Group. Using a simple model for the formation and evolution of dark halos, based on the extended Press-Schechter formalism and tested against N-body results, we show that the theoretical predictions can be reconciled with observations if gas accretion in low-mass halos is suppressed after the epoch of reionization. In this picture, the observed dwarf satellites correspond to the small fraction of halos that accreted substantial amounts of gas before reionization. The photoionization mechanism naturally explains why the discrepancy between predicted halos and observed satellites sets in at about 30 km/s, and for reasonable choices of the reionization redshift (z_re = 5-12) the model can reproduce both the amplitude and shape of the observed velocity function of galactic satellites. If this explanation is correct, then typical bright galaxy halos contain many low-mass dark matter subhalos. These might be detectable through their gravitational lensing effects, through their influence on stellar disks, or as dwarf satellites with very high mass-to-light ratios. This model also predicts a diffuse stellar component produced by large numbers of tidally disrupted dwarfs, perhaps sufficient to account for most of the Milky Way's stellar halo.

5 pages, 2 figures, Submitted to ApJ