Both intravenous single injections of 6-tritium labeled glucose and primed continuous infusions of uniformly labeled carbon-14 glucose were applied to four nonpregnant Saanen goats. Four sequential treatments were high milk production, low milk production, "dry" with no milk production, and fasting during the dry period. The diet was 25% chopped alfalfa hay and 75% grain, and by design feed intake decreased sequentially (1.85, 1.49, .80) and 0 kg/day) with the progression from high production to fasting. Glucose irreversible loss averages from single injections of tritium labeled glucose were 199, 169, 96, and 58 mg/min for respective treatments; similar averages from continuous infusions of carbon labeled glucose for the first three treatments were 164, 132, and 76 mg/min, or about 20% less. Other blood glucose kinetic parameters were estimated after single injections, and averages declined from 5.9 to 4.5 g for pool size, 315 to 134 mg/min for total entry rate, and 115 to 76 mg/min for recycling as goats went from high production to fasting. Transfer of blood glucose to milk constituents was determined by continuous infusion. Transfer quotients for lactose were 73% for high production and 67% for low production; for casein they were 1.8 and 1.9%. The percentage of glucose irreversible loss accounted for in lactose was 31% for high production and 17% for low production whereas less than .2% was in casein.