During the fall breeding season, 30 non-lactating mixed breed dairy goats were randomly allotted to one of three treatment groups. All does were implanted with 6 mg norgestomet for 11 d. Does in treatment A (n = 10) received a single injection of 750 IU pregnant mare serum gonadotropin (PMSG) at the time of implant removal, while does in treatment B (n = 10) were administered twice daily injections of 4, 3, 2 and 1 mg follicle stimulating hormone (FSH) given at 12 h intervals starting 48 h prior to implant removal. Control does (n = 10) received only the norgestomet implant. Estrus was detected in 90% of the does in treatments A and B and in 70% of control does within 48 h of progestagen removal. Mean time to onset of estrus was similar for treatments A and B but was shorter (P<0.05) than time to onset for control does. Four does in treatment A returned to estrus within 7 d after the initial post-implant estrus. Both gonadotropin treatments produced more (P<0.01) corpora lutea (CL) than controls. The total number of CL was similar among treatments, but treatment A does had more (P<0.05) abnormal CL than does in treatment B. The number of ova recovered from does in treatments A and B was not different; however, the number of embryos recovered and the number of transferable quality embryos was less (P<0.01) for does in treatment A than in treatment B. These findings suggest that FSH is superior to PMSG for superovulation of non-lactating dairy goats during the breeding season.