Sperm count distributions among exposed and control groups at a dibromochloropropane (DBCP) manufacturing plant were remarkably similar. Yet reproductive histories from 60 exposed men obtained in conjunction with the semen analyses indicated that fertility had been reduced during exposure. Ratios of observed to expected births or standardized fertility ratios (SFRs) were computed for reproductive experience at parities of 1 or greater. The SFR for the period at risk from DBCP exposure (SFR = 0.63) was significantly lower than those derived from the entire not-at-risk period (SFR = 1.21) or the portion related to nonexposed employment at the plant preceding exposure (SFR = 1.33). Significant reductions would have been evident at least 18 years prior to the year in which the histories were obtained. The effect on fertility seems to have been greatest during the initial period of DBCP production. Most fertility reduction occurred after 3.5 years of exposure. Fertility returned to normal following cessation of exposure (SFR = 1.18), although it appeared to remain subnormal for about two years. Wherever there is concern about the potential for adverse reproductive effects in the workplace, data suitable for fertility analyses should be collected during annual medical examinations.