THE LUMINOSITY FUNCTION FOR STARS WITH -0.6-LESS-THAN-OR-EQUAL-TO-MV-LESS-THAN-OR-EQUAL-TO-6.5 IN THE GLOBULAR-CLUSTER M30 - IS THERE A PROBLEM WITH STELLAR EVOLUTION MODELS

The V band luminosity function (LF) for stars ranging from the tip of the giant branch to approximately 2.5 mag fainter than the main-sequence turnoff (MSTO) in M30 (= NGC 7099, GC 2137 - 234) is derived from CCD observations. Comparison with the theoretical LFs of Bergbusch & VandenBerg (1992) for [M/H] = 2, 16 Gyr models shows discrepancies which can be described either of two ways: (1) If the observations are normalized to the models over a few magnitudes fainter than the MSTO, then there is an excess (N(obs) > N(predicted) of subgiant stars within 0.5 mag of the MSTO at the level of 4.8 sigma and an excess of stars (approximately 20%) compared to the models all along the giant branch. (2) If the normalization is made to match predicted and observed numbers of stars on the giant branch, then there is a discrepancy between models and observations in the sense that there is a broad depression in the observed LF fainter than the MSTO that is not present in the models. Consideration of LFs generated from 12 Gyr models or models adjusted to a shorter (m - M)V = 14.5) distance modulus diminishes the magnitude of some of these discrepancies, but introduces other mismatches between the observations and models. LFs constructed from stellar models that have 10% of the stellar mass in an isothermal core qualitatively match the observations better than the ''standard'' models.