The amount of cytogenetic information on prostate cancer is relatively sparse when compared with that on other common adenocarcinomas. This is primarily due to frequent overgrowth in culture of the cancer cells by normal (diploid) cells. Although newly introduced techniques of in situ hybridization will undoubtedly reveal chromosome changes in a high percentage of primary and noncultured prostate cancers, such information is at present essentially limited to numeric changes, with structural changes (eg, translocations and deletions) not being readily or reliably ascertained with presently available in situ approaches. Future developments are likely to overcome these shortcomings and, thus, make possible detailed analysis of karyotypes in prostate cancer in a much higher proportion of these tumors than available up to now. The cytogenetic data on prostate cancer in the literature are reviewed in this article and, based on the frequent involvement of some chromosomes (nos. 2, 7, 8, 10, and 16) in and molecular data on prostate cancer, a suggested multistep process for the development of prostate cancer is presented. The steps in this process may not apply to all prostate cancers or in fact be the definitive ones, but the suggested scheme does indicate that a number of ordered genetic events are involved in the process of prostate cancer genesis. Epidemiologic and familial aspects of prostate cancer also have been discussed in this review as they might relate to the genetics of the disease.