1. Sheep fed at a constant rate were infused intraruminally with [1-14C]-acetate, -propionate or -butyrate during 5hr. periods. 2. Volatile fatty acids were estimated in the rumen contents and steady-state conditions were obtained. 3. Of the butyric acid carbon 60% was in equilibrium with 20% of the acetic acid carbon, and 2-3g.atoms of carbon were interconverted/day. 4. Little interconversion took place between propionic acid, acetic acid or butyric acid. 5. The net production rates for acetic acid, propionic acid and butyric acid were 3 7, 1-0 and 0-7moles/day respectively. 6. The production of volatile fatty acids accounted for 80% of the animal's energy expenditure.

1962are described which were designed to test the effects of this method of circulating the rumen contents on the metabolism of food in the rumen and the efficiency with which the system mixed PEG with the rumen contents. E X PER I M E N T A LAnimals and their treatment. Three mature Greyface wethers of about IZO lb live weight were fitted, by the method of Ash (1957), with rumen cannulas of z in. internal diameter.T h e animals were given 500 g of pelleted diet at 7.30 a.m. and again at 7.30 p.m. The rations were completely consumed within 15 min of feeding. Two diets were used: grass cubes consisting entirely of dried-grass meal, and maize cubes consisting of 50 yo dried-grass meal, 44 yo ground flaked maize, 5 yo linseed meal, 0.5 Yo steamed bone flour and 0-5 yo salt, with I g cobalt chloride hexahydrate added to each 1000 lb.

The compositions of a number of known mixtures of volatile fatty acids in aqueous solution and of volatile fatty acids in samples of rumen liquor were estimated by two gas-liquid chromatographic procedures, one of which operated in conjunction with an automatic titrating device and the other with a flame-ionisation detector. It was found that the method using flame-ionisation as the system of detection was more accurate and yielded results more rapidly than did that in which automatic titration was employed.