SUMMARY Four experiments were conducted to define reasons for failure of nonsurgical transfer of bovine embryos. Single embryos were transferred to the uterine horn ipsilateral to the corpus luteum-bearing ovary 6 to 9 days after estrus. All recipient heifers were within ± 24 hours of heat synchrony with the donors. All nonsurgical transfers were accomplished, using the Cassou apparatus with 0.5 ml French straws. Pregnancy was determined by rectal palpation on day 40 (31 to 34 days after transfer). In experiment 1, when embryos were transferred surgically, pregnancy rates were not different between the midventral approach (11/14, 77%) and the flank approach (10/14, 71%) but were lower (P < 0.05) when transfers were done nonsurgically (3/14, 36%). In experiment 2, transfer to the midportion of the uterine horn was compared with transfer to the cranial end of the uterine horn. Embryo survival in the midportion of the uterine horn tended to be higher (P < 0,06) when embryos were placed nonsurgically (15/20, 75%) than when transfer was accomplished surgically (9/20, 45%). Surgical transfer to the midportion of the uterine horn was not affected by sham nonsurgical manipulation. Pregnancy rate was not different between nonsurgical transfer to the middle portion of the uterine horn (15/20, 75%) and surgical transfer to the cranial portion (12/20, 60%). In experiment 3, pregnancy rate was not affected by digital compression to the uterine horn during withdrawal of the Cassou instrument after nonsurgical transfer. Pregnancy rates in experiment 4 were not different between the midportion of the uterine horn and the caudal portion of the uterine horn, using nonsurgical embryo transfer. An operator experienced in nonsurgical embryo transfer obtained higher pregnancy rates (23/40, 58%) than did an operator with no previous nonsurgical embryo transfer experience (14/40,35%, Ρ < 0.05). Amount of time to penetrate the cervix did not affect transfer results; however, increasing the time that the Cassou catheter remained in the uterus tended to decrease pregnancy rates (P < 0.07). It is concluded that experience in nonsurgical transfer is important in achieving acceptable success rates and that specific regional placement of the embryo in the uterine horn ipsilateral to the corpus luteum-bearing ovary is not imperative for nonsurgical transfer success.