Background. The goal of this paper is to provide a preliminary description of the marital status for a large number of childhood cancer survivors participating in the Childhood Cancer Survivor Study (CCSS). Procedure. This report includes children and adolescents (<21 years of age) diagnosed with cancer between 1970 and 1986 at 25 oncology centers in the United States and Canada who survived at least 5 years from diagnosis. Self-reported data from 10,425 survivors are used in this preliminary descriptive summary. The proportion of survivors ever married and divorced/separated is compared to the U.S. population according to age-specific groups. The median age of the survivor population at diagnosis was 7 years and 26 years at the time martial status was ascertained. Excluded from this assessment are children <15 years of age at the time of study, those whose martial status was unknown, and those married prior to diagnosis. Data for marital status of the U.S. population, as tabulated in the Bureau of Census 1995 Update, is used as a general comparison to the survivor population. Results. Overall, 32% of the survivors reported being married or living as married, 6% being di-vorced or separated, 0% being widowed, and 62% having never been married. In general, compared to the U.S. population, survivors were less likely to have ever married, particularly females and whites, but, once married, were less likely to divorce/separate, again particularly females and whites. Black survivors were generally found to be more likely to have married, with males and blacks more likely to divorce/separate once married. Comparison of childhood tumor types suggested that survivors of CNS tumors, particularly males, were less likely to have ever married and more likely to divorce/separate compared to those with other cancer diagnoses and the general U.S. population. Conclusions. This interim evaluation of the CCSS cohort provided preliminary data describing a suggested decreased likelihood of marriage, which may be influenced by gender and/or race. These patterns must be confirmed within the entire CCSS cohort and comparisons made with an appropriate sibling comparison group before making final conclusions. Med.