Tomatoes (Lycopersicon esculentum Mill.) were grown on an Arredondo fine sandy soil to evaluate the effects of water quantity applied by drip irrigation scheduled by pan evaporation in a 3-year study. Water was applied to polyethylene-mulched tomatoes at 0, 0.25, 0.50, 0.75, and 1.0 times pan evaporation in one application per day. Irrigation was also scheduled with tensiometers to apply water to maintain soil water tension above 10 cb. The response to irrigation varied with rainfall during the three seasons. In an extremely dry season, fruit yields were doubled by irrigation. Total fruit yields were highest with irrigation quantities of 0.75 and 1.0 times pan and significantly lower with 0.25 and 0.50 times pan. In an extremely wet season, fruit yields were not influenced by water quantities from O to 1.0 times pan. In a third season that was wet from the middle to the end of the season, irrigation more than doubled the marketable fruit yield. However, with an increase in water quantity from 0.25 to 0.75 times pan, yield increased only from 65.9 to 74.1 t·ha-1. Water uses during the three seasons with 0.75 pan were 31.8, 31.1, and 29.6 cm, respectively. Fruit yields were similar with the 0.75-pan and 10-cb tensiometer treatments, but water uses with the latter treatment were 15.8, 17.0, and 18.4 cm during the three seasons, respectively. Tomato leaf N concentrations were reduced slightly with each increase in water quantity applied, even though N was applied with drip irrigation. Leaf N concentrations with the 10-cb treatment were generally equal to or higher than the concentrations with 0.75 pan.