Fifty-eight treatment cycles in an in vitro fertilization/gamete intrafallopian transfer (IVF/GIFT) program were studied to compare the efficacy of two urinary methods, hemagglutination test (Higonavis) and enzyme immunoassay (Ovustick), in detection of spontaneous luteinizing hormone (LH) surge. If an isolated rise in urinary LH level was taken as indicative of LH surge, the false-positive rate was 36.7% for Higonavis and 10.2% for Ovustick. The difference was statistically significant (P less than 0.001). If only a sustained rise in urinary LH was taken to indicate LH surge, the false-positive rate was 6.1% for Higonavis and 0% for Ovustick. In the seven cycles with a spontaneous plasma LH surge, there was a positive correlation between the plasma LH levels and the two urinary assay methods in six cycles (85.7%). Compared to plasma LH, there was a mean delay of 17.4 hr by the Higonavis test and 15.6 hr by the Ovustick test. If a sustained rise in urinary LH levels was taken as indicative of LH surge, both methods are quite accurate but the Ovustick appeared to be more specific.