STELLAR OXYGEN ABUNDANCES .2. ON THE LOCATION OF THE BREAK IN THE [O/FE] VERSUS [FE/H] PLANE

We examine the question of the location of the break in the [O/Fe] vs [Fe/H] relation. Fitting simple functions to a relatively homogeneous data set, we find that goodness of fit does not currently favor any unique break point in the range -1.7 less-than-or-equal-to [Fe/H] less-than-or-equal-to -1.0. Statistical analysis of fits to existing Be-9 data, which we use as a proxy for O, results in the same conclusion. However, we find other evidence which does suggest that the break occurs near [Fe/H] approximately -1.0. This is higher than recently claimed values near [Fe/H] approximately -1.7. However, breaks at intermediate [Fe/H] values cannot be excluded. Some of the qualitative implications of this result are discussed with particular attention to the possibility of significant delays between the end of halo formation and the beginning of star formation in the disk. A gap in the observed [O/H] distribution may betray the occurrence of such a hiatus. The gap suggested by other workers to be present seems to persist despite the addition of more O data. However, the observed gap seems to be related to the well known shortfall of stars with -1.2 less-than-or-equal-to [Fe/H] less-than-or-equal-to -0.8. More work is still required to discern whether this latter shortfall is only a selection effect or related to a genuine ''G dwarf problem.'' We argue that present data indicate that [O/Fe] is constant in the halo (at least for [Fe/H] greater than or similar -3). For [Fe/H] greater-than-or-equal-to -1.0, ordinary least-squares regressions indicate that the slope in the [O/Fe] vs [Fe/H] relation is -0.35 to -0.40, a fair amount less than the value (approximately -0.5) usually assumed. However, the ordinary least-squares bisector regression, which may be more appropriate for determining the relation between [O/Fe] and [Fe/H], yields a steeper slope (-0.7). Until a concensus can be reached on which regression is applicable for what purposes, the constraints provided by an ''observed'' [O/H]-age relation (which does not use true O abundances, but adopts O abundances based on Fe abundances) on several issues of Galactic chemical evolution should be regarded with caution.