Drip-irrigation scheduling techniques for fresh-market tomato (Lycopersicon esculentum Mill.) production were compared in three growing seasons (1989-91). Three regimes were evaluated: EPK [reference evapotranspiration (ET(o), corrected Penman) x programmed crop coefficients], ECC (ET(o) x a crop coefficient based on estimated percent canopy coverage), and SMD (irrigation at 20% available soil moisture depletion). EPK coefficients ranged from 0.2 (crop establishment) to 1.1 (full canopy development). Percent canopy coverage was estimated from average canopy width divided-by row width. Irrigation in the SMD treatment was initiated at -24 kPa soil matric tension, with recharge limited to 80% of daily ET(o). The EPK and ECC regimes gave similar fresh fruit yields and size distributions in all years. With the EPK scheduling technique, there was no difference in crop response between daily irrigation and irrigation three times per week. In all seasons, ECC scheduling resulted in less total water applied than EPK scheduling and averaged 76% of seasonal ET(o) vs. 86% for EPK. Irrigating at 20% SMD required an average of only 64% of seasonal ET(o); marketable yield was equal to the other scheduling techniques in 1989 and 1991, but showed a modest yield reduction in 1990. Using an SMD regime to schedule early season irrigation and an ECC system to guide application from mid-season to harvest may be the most appropriate approach for maximizing water-use efficiency and crop productivity.