It is important to understand the age at which sexual relations start in designing HIV prevention strategies. Most studies on age of sexual activity of young people provide estimated percentages of those that are sexually active in specific age groups, and tend either to not provide data for age at sexual debut, or to overlook the complexities of analysing data concerning sexual debut. This study considers the rate of entry into sexual relations in South Africa by providing the median age of women at time of first sexual relations as well as other percentiles, and analyses the hypothesis that age at sexual debut has been falling. The analysis uses data from the 1998 South African Demographic and Health Survey, including a nationally representative sample of 11 735 women, aged 15 to 49, interviewed in 1998. The analysis uses life-table techniques and multivariate analysis. About 8% of the respondents had had sex by age 15. The median age at time of first sex was approximately 18 years, and virtually all the women had had sex by age 23. There is evidence that the peak of the rate of entry into sexual relations occurs at age 18 and that younger cohorts of women are entering sexual relations at a younger age. The rate of entry into sexual relations is 14% to 20% faster for the younger cohorts, based on information given by the older respondents concerning their own behaviour at the same age. Age of entry into sexual relations of the women who participated in this study is compared to findings of demographic and health surveys in Tanzania and Zimbabwe, and rates of entry into marriage are also presented. The lag between entry into sexual relations and rate of entry into marriage is compared across countries. It appears that, whereas South Africans tend to enter into sexual relations later than Tanzanian counterparts and more or less at the same stage as Zimbabweans, their rates of entry into marriage are hugely delayed. The consequences for HIV infection control are discussed.