Summary The epidemiological features of Kaposi's Sarcoma (KS) incidence in England and Wales in the period [1971][1972][1973][1974][1975][1976][1977][1978][1979][1980] (Bluefarb, 1957), but no population-based statistics exist.In Central and East Africa, KS has long been a common tumour, accounting for more than 10% of all malignancies in males in some countries (Cook et al., 1971). Before AIDS, the sex ratio in Africa was near to 1: 1 in children, but in adults over 10 males to each female were affected (Olweny, 1984).The AIDS-associated epidemic of KS in industrialised countries has epidemiological features which point to a transmissible agent, spread by sexual contact, plus HIV-mediated immunodeficiency, as the likely cause of this neoplasm (Beral et al., 1990 expected rates. Country of birth was unknown for four males and nine females with KS and these were excluded from calculation of standardised registration ratios. Confidence intervals for standardised registration ratios were calculated from tables of multipliers (exact limits) for estimating SMRs (Breslow & Day, 1982 (underlying and contributory) were obtained from extracts of death certificates of these individuals. The proportion of these people who were single in the two time periods was calculated and the difference in these proportions tested for significance using Fisher's exact test. In males, the proportions were adjusted for age differences by applying the England and Wales population 5 year age specific proportions of men who were single at the 1981 census to calculate expected numbers of single men in both time periods. The observed distribution of males who were single between the two time periods was then compared with the expected distribution under a null hypothesis of no difference in age-adjusted marital status between the two time periods, and tested for significance using the binomial distribution.