1. Digestibility trials were conducted, and the rate of passage of hay was measured, with five Shorthorn cows during a period of normal diet and during two experimental periods in which diets low in hay and high in concentrates were given. In the first experimental period the concentrates were cubes of the wartime type sold as National Cattle Food No. 1, and containing a variety of constituents; in the second they were a mixture offlakedmaize (50%), weatings (35%) and decorticated ground-nut cake (15%). For convenience these are referred to as concentrate ‘cubes’ and ‘mixture’ respectively.
2. During the initial and final control periods the cows consumed daily 17–21 lb. hay, and about 4·5 lb. concentrates per 10 lb. of milk produced. In the initial control period the concentrates were the concentrate cubes, and in the final they were the concentrate mixture. The hay was reduced to 4 lb. daily during the two experimental periods and the remainder of the standard requirements of the animals were met by concentrates. The concentrates were the cubes in the first experimental period and the mixture in the second experimental period.
3. Seducing the hay to 4 lb. did not affect the fat content of the milk when the other food in the diet was the concentrate cubes, but there was a striking mean fall of 1·04% fat when the cubes were replaced by the concentrate mixture. This represented a loss of over 30% in the yield of fat.
4. Digestibility trials, conducted in the initial control and first and second experimental periods, indicated that the fall in milk fat content was not the result of changes in the amounts of dry matter, crude protein, ether extract, crude fibre, cellulose, cellulosans or pentosans not in cellulose digested. The essential difference between the diet of low hay with the concentrate mixture and the other diets given in this experiment was that it provided a high intake of starch yet had little of the physical property of roughage. It is concluded that depression of milk fat content results from a combination of these two factors and probably originates from changes in the physical and biochemical processes of the reticulo-rumen.
5. The intake of starch equivalent, as calculated from the intake of digestible nutrients, was close to standard requirements in all periods of the experiments, but there was a surplus of digestible crude protein.
6. The mean solids-not-fat content of the milk rose 0·48% at the time of the fall in milk fat, and this was entirely due to an increase in milk protein. After the return to normal diets the recovery of solids-not-fat was slower, but no less complete, than the recovery in milk fat.