Abstract
Two contradictory results on the evolution of SO galaxies now exist in the recent literature: either S0's are old (zformation > 2) and are evolving passively or most of them form at z < 0.5, as implied by the deficit of S0's in intermediate-redshift (z ~ 0.5) clusters. The resolution of this controversy may be that the apparent deficit of S0's has been derived from a quantity—the E-to-S0 ratio—that is prone to morphological classification errors. Once all sources of error are taken into account, the E-to-S0 ratios of clusters at different redshifts are fully compatible, and no additional creation of S0's at z < 0.5 is required by the data. Furthermore, there is no deficit at all of S0's in the intermediate-redshift cluster for which we have morphological types of very high quality and thus derive an E-to-S0 ratio with a small error.