I. Acetate, propionate, butyrate and glucose w-ere each administered intravenously (2.09 M J (0.5 Mcal)/d) to lambs on either high-concentrate (HC) or high-roughage (HR) diets in two experiments with 4 x 4 Latin square design. During a control period all animals received a saline infusion.
2.In lambs on the HC diet glucose was the most effective metabolite tested for improving nitrogen retention. Acetate was 5 I yo, butyrate, 75 yo, propionate, 90 yo as effective as glucose in promoting N retention. Propionate infusion was most effective in increasing N retention in lambs on the HR diet. Acetate was 60 %, butyrate, 75 ?A, and glucose, 80 "/o as effective as propionate. Glucose infusion did not lower plasma urea N (PUN) concentration in lambs receiving the HC diet, but the three volatile fatty acids (VFA) did lower it ( I ' < 0.01). All energy sources significantly reduced the PUN concentrations with the HR diet, but differences between sources were not significant.3. Plasma concentrations of valine, isoleucine, leucine and phenylalanine were depressed to 57,71,89 and go % of their control values during glucose, propionate, butyrate and acetate infusion. Plasma concentrations of lysme and methionme were significantly decreased from control values ( P < 0.05) but there was no significant difference between energy sources.Lysine, valine, isoleucine and leucme were significantly (P < 0.01) higher in lambs given the HR diet than in lambs given the HC diet. The concentration of alanine was greater ( P < 0.05) during glucose infusion than during acetate and butyrate infusions, indicating net synthesis of alanine and possibly other non-essential amino acids during glucose infusion, since the plasma free amino acid N concentration increased despite decreased concentrations of plasma essential amino acids during glucose infusion. 4. The results indicated that glucose and propionate were superior to acetate and butyrate as energy sources for protein formation.5 . The poor relationship between PUN and N balance and plasma free cssential amino acids suggested that P U N concentration may have been inappropriate to assess the protein status under some dietary conditions. The three major volatile fatty acids (VFA) produced by rumen fermentation have not been widely investigated as sources of energy in protein synthesis in Iambs. Rook, Balch, Campling & Fisher (1963) and Orskov & Allen (1966) found no significant differences between the effects of the three major VFA on nitrogen retention when they were infused into the rumen or given as their salts in the food. Eskeland, Preston & Pfander (1971) and Eskeland, Pfander & Preston (1973) found that propionate promoted a higher N retention than butyrate, and acetate was the least effective when the respective energy sources were administered intravenously on an isoenergetic basis. All VFA were utilized less efficiently than glucose.The free amino acid concentration in blood of simple-stomached animals has been