THE BAADE-WESSELINK METHOD AND THE DISTANCES TO RR LYRAE STARS .8. COMPARISONS WITH OTHER TECHNIQUES AND IMPLICATIONS FOR GLOBULAR-CLUSTER DISTANCES AND AGES

We summarize all techniques that derive M(v) or M(bol) as a function of [Fe/H] for RR Lyrae variables. Baade-Wesselink analyses indicate <M(v)(RR)> is-proportional-to 0.16[Fe/H] and <M(bol)(RR)> is-proportional-to 0.21[Fe/H]. We revise recent results from main-sequence fitting to account for metallicity-dependent errors in B - V from model isochrones, finding <M(v)(RR)> is-proportional-to 0.12[Fe/H]. We redefine the equilibrium temperature scale for RR Lyrae stars, a crucial quanity for comparison of theory with observations and the period-shift analyses of Sandage. We readdress the period-shift analyses for RR Lyrae stars in globular clusters and in the field, finding in both cases <M(bol)(RR)> is-proportional-to 0.19[Fe/H]. All methods now agree that, within the errors, <M(v)(RR)> is-proportional-to 0.15[Fe/H] and <M(bol)(RR)> is-proportional-to 0.19[Fe/H]. There is need to invoke an anticorrelation between helium and heavy-clement abundances. If [O/Fe] = + 0.3, the Galaxy's age may be as high as 20 +/- 3 Gyr, and the most metal-poor clusters are on average older than the more metal-rich ones. If [O/Fe] is-proportional-to -0.4]Fe/H], the Galaxy's age is about 14 +/- 2 Gyr, and the cluster's ages do not correlate with metallicity. Regardless of the oxygen abundance, there is a significant range in ages among the intermediate-metallicity clusters. The spread correlates with horizontal branch morphology. The Oosterhoff classes are very distinct in a plot of RR Lyrac richness versus metallicity, suggesting that the division results from a sudden change in the way the horizontal branch is populated at [Fe/H] almost-equal-to - 1.7.