An interview to detect histories of sexual abuse was administered to consecutive attenders at a gynecology clinic on two occasions, on one occasion by a male interviewer and on the other by a female interviewer. Fifty-six subjects were assessed, and at least partial agreement was found in 70 percent between the two interviews. Approximately one-third of incidents were reported at only one of the interviews, with gender of interviewer making little apparent difference. Contrary to expectation, subjects appeared more forthcoming about their sexual abuse at the first interview. There is some evidence to suggest that it is the order of the interviews rather than the gender of the interviewer that is the important factor, since the number of subjects reporting incidents and the number of incidents reported to a male and female do not differ significantly. The lower reporting at the second interview might be due to context, given the shorter time and perhaps consequent lack of rapport. The first and second interviews were different in respects other than the gender of the interviewer. It is possible that differences in personality or style among the interviewers contributed to differences. It is also possible that the familiarity of the questions and their sequence contributed to a wish by some subjects to withhold information from another, albeit somewhat familiar, interviewer. 2 tables and 11 references