Insulin (0.4 IU/kg/d) or saline was injected subcutaneously into six goats at 08:00 for 5 days in a double switch-back design in Expt 1. A Latin square design was used in Expt 2 with four treatments, insulin (0.3 IU/kg/d), glucose (50 g/d in 20% solution), insulin plus glucose, or saline, infused intravenously into eight goats at 2-h intervals from 06:00 to 14:00 for 1 day. Milk yield per day/feed consumed was significantly depressed (P < 0.05) by 0.28 units by insulin in Experiment 1. Average mammary vein glucose concentration was significantly lower (P < 0.05) after the treatment. Goats receiving five treatments insulin single day in Expt 2 had no significantly changed (P > 0.05) milk production and average mammary vein glucose concentration. Infusion of glucose with or without insulin depressed average milk fat content (P < 0.05), while the latter treatment significantly elevated average mammary glucose concentration (P < 0.05). When the lowest milk production was recorded (10:00 to 14:00), the elevations of mammary vein glucose after glucose infusion, and milk fat after insulin were significant (P < 0.05). Fluctuations in precursors for the mammary gland induced by exogenous insulin and glucose can be rapidly accommodated by the organs through a similar mechanism as occurs during the feeding and fasting cycle.