Subdwarf Studies. III. The Halo Metallicity Distribution

We have constructed the metallicity distribution of a sample of 372 kinematically selected halo stars, most having spectroscopically determined abundances, drawn from our southern subdwarf study and the northern program of Laird et al. AJ, 95, 1843 (l988a); 96, 1908 (l988b). We find that the metallicity distribution of this sample is in very good agreement with the predictions of Hartwick's ApJ, 209, 418 (1976) simple model over 2.5 decades of abundance. In particular, we find no significant deficit of metal weak stars, in contradistinction to the earlier investigation of Bond ApJ, 248, 606 (1981). We assess selection effects in the data, and disregard the most metal-rich 10% of the sample, which is probably contaminated by disk stars. We also compare several of Searle's The Evolution of Galaxies and Stellar Populations (Yale University Press, New Haven, 1977) stochastic models with the data. The field star data are not well fitted by that family of models, but we find very good agreement between the globular cluster metallicity distribution and a stochastic model with a mean of 10 enrichments per fragment. Previous studies have shown that, due to the small number of clusters available, the agreement between their metallicity distribution and that of the simple model is marginal. It is possible, but not established, that the clusters and the field stars have different parent distributions, a stochastic model producing the former, and a simple model the latter.