Trichloroacetonitrile (TCAN) is a by-product of the chlorine disinfection of water containing natural organic material. When administered by gavage to pregnant Long-Evans rats in a medium-chain triglyceride vehicle, tricaprylin oil (Tricap), at a volume of 10 m l/kg, TCAN induced fetal cardiovascular anom alies at doses as low as 1 m g/kg/d (Smith et al., 1988). A slight but possibly biologically significant increase over the water control group in adverse pregnancy outcomes (resorptions, reduced fetal weight, and anomalies) was observed in the Tricap control group. This led us to reexamine the developmental effects of TCAN in a second vehicle, corn oil (CO). Five groups of approximately 20 pregnant female rats received TCAN in CO at 15, 35, 55, and 75 mg/kg/d, and in Tricap at 15 m g/kg/d (10 m l/kg dosing volume). Corn oil, Tricap, and water served as vehicle controls. Animals were treated by oral intubation on gestation d 6- 18 (vaginal plug=d0). Five out of 20 dams (75 m g/kg) died during treatment. Adjusted maternal weight gain was lower in females receiving 35 mg/kg TCAN or greater. The mean percent of nonlive implants per litter was elevated at 55 and 75 m g/kg TCAN (CO). The TCAN dose-response curve for fetal (but not maternal) effects was shifted to the right when CO was compared to Tricap. Fetal weight was reduced at 15 m g/kg TCAN (Tricap) and at 55 m g/kg TCAN (CO). When TCAN was adm inistered in CO, the mean frequency of soft-tissue malformations decreased with significantly fewer septal and great vessel cardiovascular defects observed. We hypothesize that the volatile haloacetonitrile, TCAN, may interact with the Tricap vehicle in such a way that effects on the developing cardiovascular system are potentiated. The lowest observed adverse effect level for TCAN (CO) was determined to be 35 kg/kg.