Associations between occupational exposures and spontaneous abortion (SA) in a cohort of female veterinarians were studied with pregnancy and job-exposure history data collected as part of a mixed-mode survey of all women graduating from U.S. veterinary colleges during the 11-year period 1970-1980 (N = 2,997; response rate 85.0%). Data analysis focused on SA risks among postgraduation pregnancies in relation to 1) type of clinical practice at the time of conception and 2) self-reported occupational exposures to 13 exposure entities. Multiple logistic regression was used to control for the potentially confounding effects of maternal age, gravidity, previous SA, and smoking and drinking behaviors. Pregnancies reported by veterinarians employed in all-equine practices were at highest relative risk of spontaneous abortion when compared with pregnancies reported by unemployed veterinarians [confounder-adjusted odds ratio (aOR) = 2.1; 95% confidence interval (CI), 0.6-7.4]. Agent-specific relative risk estimates ranged from 0.7 to 1.1, suggesting little or no excess risk. However, when analyses were restricted to small-animal practitioners, there was a weak association between SA risk and job-related exposure to ionizing radiation (aOR equals; 1.3; 95% CI, 0.8-2.0), a finding not inconsistent with the results of two other studies of female veterinarians. Although this study showed no strong associations, the results suggest a relationship between SA among female veterinarians and certain exposure types, and thus focus attention on other workers who encounter similar on-the-job hazards.