1. The rates of gluconeogenesis from most substrates tested in the perfused livers of well-fed rats were about half of those obtained in the livers of starved rats. There was no difference for glycerol. 2. A diet low in carbohydrate increased the rates of gluconeogenesis from some substrates but not from all. In general the effects of a low-carbohydrate diet on rat liver are less marked than those on rat kidney cortex. 3. Glycogen was deposited in the livers of starved rats when the perfusion medium contained about 10mm-glucose. The shedding of glucose from the glycogen stores by the well-fed liver was greatly diminished by 10mm-glucose and stopped by 13.3mm-glucose. Livers of well-fed rats that were depleted of their glycogen stores by treatment with phlorrhizin and glucagon synthesized glycogen from glucose. 4. When two gluconeogenic substrates were added to the perfusion medium additive effects occurred only when glycerol was one of the substrates. Lactate and glycerol gave more than additive effects owing to an increased rate of glucose formation from glycerol. 5. Pyruvate also accelerated the conversion of glycerol into glucose, and the accelerating effect of lactate can be attributed to a rapid formation of pyruvate from lactate. 6. Butyrate and oleate at 2mm, which alone are not gluconeogenic, increased the rate of gluconeogenesis from lactate. 7. The acceleration of gluconeogenesis from lactate by glucagon was also found when gluconeogenesis from lactate was stimulated by butyrate and oleate. This finding is not compatible with the view that the primary action of glucagon in promoting gluconeogenesis is an acceleration of lipolysis. 8. The rate of gluconeogenesis from pyruvate at 10mm was only 70% of that at 5mm. This ;inhibition' was abolished by oleate or glucagon.