To the Editor.
—We read with interest the article in the July issue of the Archives, "Comparison of Prognosis After Enucleation vs Cobalt 60 Irradiation of Melanomas," by Dr Gass.1Several limitations are listed in the article, some of which cannot be remedied. For example, the sample size is small and the patients were not randomized to treatment groups. However, unbalanced distribution of base-line characteristics can be controlled by multivariate analysis when assessing treatment effects. In fact, results of such analyses change the conclusion of the article, ie, the evidence for difference in treatment effect on survival is inconclusive, and the estimated difference between enucleation and cobalt irradiation is not statistically significant.We coded and reanalyzed data presented in the first two tables of Dr. Gass' article. Similar to other studies on prognostic factors among melanoma patients,2 the following variables were evaluated, as shown in Table 1: (1)