Ovarian cancer (OC) occurs in 1 in ~2,500 women in the general population and presents in a late clinical stage (Stage III) in more than 80% of patients. This 80% has a 5-year survival rate of only ~35%. If OC could be detected near onset (Stage I), in contrast, it is generally believed that the 5-year survival rate could be increased to greater than 90% under existing treatment protocols. Petricoin et al. have reported a dramatic breakthrough in detection of Stage I OC, using mass spectroscopy of serum proteins and a proprietary genetic algorithm. Horner has demonstrated that even under a 10-bin compression of the intensity range of the NCI/FDA 2003 OC reference spectra, several widely available, non-proprietary phylogenetic algorithms can classify without error >90% of the spectra in that set at the 95% confidence level. Here I report that an apomorph analysis of the results of applying the Maximum Parsimony phylogenetic method to the NCI spectra yields several distinct OC diagnostic subsets of the spectra, none of which intersects, and each of which discriminates at least as well as, those identifiedby Petricoin et al.