A long-standing problem in epididymal physiology is the fate of unejaculated spermatozoa in the cauda epididymidis under conditions such as congenital absence of the vas deferens, long-term vasectomy, or castration. There is no convincing evidence for significant absorption of spermatozoa, defective or otherwise, by spermiophagy or dissolution in the epididymis of normal animals. Spermiophagy by epithelial cells or intraluminal macrophages may take place if the duct ruptures and granulomas form (e.g., after experimental ligation), although there is no quantitative information on the rate of sperm removal by this means. In one animal model (the rabbit), the epididymis is unusually resistant to granuloma formation and has provided unique insights into a phenomenon that is suggested to be present in all species. Spermatozoa retained in the rabbit cauda epididymidis by placing ligatures on the vas deferens and corpus epididymidis degenerate after several weeks but do not decrease significantly in numbers. After castration, however, they die very rapidly and >90% disappear. It is hypothesized that, in the normal androgen-maintained epididymis, degradative pathways are present in the luminal fluid that are constitutively inhibited by survival signals emanating from the epithelium. In the absence of androgen, the intraluminal mileau changes and death signals predominate that activate degradative pathways via the ubiquitin-proteasome system, DNAses, etc., to mediate dissolution of sperm organelles and nucleoprotein. It is suggested that the latter condition is the default situation and is only prevented by the stimulatory action of androgens on the epididymal epithelium.