Objective: This study was designed to evaluate the effects of endogenous LH surge, GnRH agonist (GnRH-a) or human chorionic gonadotropin (hCG) as ovulation trigger on pregnancy rate by intrauterine insemination (IUI). Method: Patients received daily 100 mg of clomiphene citrate (CC) for 5 days starting on the third day of the menstrual cycle followed by human menopausal gonadotropin (hMG) for ovulation induction. Follicles larger than >16 mm in diameter were present in the ovary, frequent LH tests in urine were introduced to detect an endogenous LH surge. Final follicular maturation and ovulation were induced by GnRH-a 0.1 mg (s.c.) or hCG $5,000{sim}10,000$ IU (i.m.) administration except natural ovulation. Pregnancy was classified as clinical if a gestational sac or fetal cardiac activity was seen on ultrasound. Results: There were no differences in age, duration of infertility and follicle size, but more ampules of hMG were used in GnRH-a group compared to hCG 10,000 IU treated group (p<0.05). Lower level of estradiol ($E_2$) on the day of hCG or GnRH-a injection was observed in hCG 10,000 IU group than other treatment groups (p<0.01). The overall clinical pregnancy rate was 19.8% per cycle (32/162) and 22.2% per patient (32/144). Pregnancy rate was higher in natural-endogenous LH surge group (37.5%, 9/24) than GnRH-a (18.8%) or hCG treated group (20.9% & 13.9%), but this difference was not statistically significant. No patient developed ovarian hyperstimulation. Abortion rate was 22.2% (2/9) in hCG 5,000 IU group. Delivery or ongoing pregnancy rate was 37.5% (9/24), 18.8% (3/16), 16.3% (7/43) and 13.9% (11/79) in endogenous LH surge, GnRH-a, hCG 5,000 IU and hCG 10,000 IU treatment groups, respectively. Conclusion: These results support the concept that use of natural-endogenous LH surge in stimulated cycles may be more effective to obtain pregnancies by IUI than GnRH-a or hCG administration.