We evaluated whether lactase activity is affected by EARLY feeding compared with STANDARD onset of feeding in a randomized trial in premature infants. The EARLY group (birthweight 1001 ± 173, gestational age 28 ± 1; mean ± SD) received milk (20 mL·kg-1·d-1) plus parenteral nutrition from day 4 to 15 days of age. The STANDARD group(birthweight 1099 ± 153, gestational age 28 ± 1) received only parenteral nutrition until day 15 at which time milk feedings were begun at 20 mL·kg-1·d-1. Milk intake then was advanced similarly in both groups. Lactase activity was measured at 10, 20, and 30 d of age using the urinary lactose/lactulose (L/LL) excretion ratio. Lactose and lactulose were administered in milk or water for 30 h. During the last 6 h, urine was collected for determination of the L/LL ratio. The ratio of urine to milk L/LL was calculated. Lactase activity is inversely related to the L/LL ratio. RESULTS: The L/LL ratio declined in both EARLY and STANDARD groups between 10 and 20 d (p < 0.01). The ratio was significantly lower in the EARLY group compared with that in the STANDARD group at 10 and 20 d of age:Table Conclusions: 1) Feedings begun at 4 d compared with those at 15 d resulted in greater lactase activity as measured by the urine to milk L/LL ratio; 2) Since early feeding appears to stimulate the development of lactase activity, feedings should not be withheld because of concerns about low lactase activity.