To the Editor.—
In the October 1985 issue of the Archives, Ellis et al1 stated that estrogen binding is seen in melanomas and melanoma precursor lesions.In a previous study of 45 cases of malignant melanomas, we concluded that estrogen receptors (ERs) were present in three (7%) of 45 cases of melanomas. If present, their concentration was found to be at the limit of sensitivity of our method. In the literature, the results reported are rather contradictory, since positivity percentages vary from 12%3 to 46%,4,5 concentrations being in general comparatively low. Some authors6 report that estradiolspecific receptors are not involved, the binding being artifactual instead.Recently, using a histochemical technique with a fluorescent-labeled estrogen compound, Ellis et al1 showed an important cellular fixation of this compound in several cases of melanoma. These techniques, widely used during the past few years in order to expose the