Field studies were conducted to determine the dissipation and movement of metribuzin and metolachlor applied at conventional rates to a Verndale sandy loam (Udic Argiboroll) in north-central Minnesota under irrigated potato production in two years. The rapid dissipation of both metribuzin and metolachlor was found during the initial l0 to 15 days in both years, and more than 70% of the applied herbicide dissipated during this period. From 10 to 15 days after application up to the end of growing season in both years, the levels of both herbicides decreased slowly with time. Metolachlor dissipated at a slower rate than metribuzin in surface soil and could carry over to the next cropping season. Metribuzin and metolachlor were detected in only 6 and 1 of 154 soil samples in the first year and in 3 and 4 of 225 soil samples in the second year, taken from 15 to 75 cm, respectively. Fifty to 67% of water samples from suction samplers at 135-cm depth contained detectable levels (＞0.4μg L-1) of herbicides in both years. Under laboratory conditions degradation of both herbicides was much slower than their dissipation in field. Therefore, it appeared that leaching might be an important dissipation pathway for metribuzin and metolachlor under irrigated potato production.