The mortality rate for male psychoanalysts was compared to that for the general white male population; for male physicians; and for male psychiatrists and neurologists. For psychoanalysts the rate was found to be significantly lower than for any of the otherthree groups. Several possible explanations for this low mortality rate are considered. Two major factors may be the careful screening of candidates for psychoanalytic training and their personal analysis. Possible methods of controlling for these factors are suggested.