Thus, the first step is to calibrate the metallicity and age color indices with objects where the mean metallicity and age are known, i.e. star clusters with well mapped CMD's. A sample of northern globulars is shown to the left, a plot of mean [Fe/H] versus vz-yz color.

The relationship between color and [Fe/H] is excellent, however there are some notable problems. One is that the range of cluster metallicities (-2.5 to -1), and ages, is limited compared to the expectations for galaxy metallicities (solar or greater). In fact, the addition of dwarf and giant ellipticals, with spectroscopic metallicity measurements, to the diagram indicates a change in slope from globulars. This is, of course, due to the fact that galaxies are not made of a single metallicity stellar population, but rather are a composite of stars with a range of metallicities.