The scheme of proximate analysis ordinarily used in routine chemical examination of livestock feeds is demonstrably inadequate as an index of their general feeding values. Criticism may be levelled at all five proximate principles but, because of their relative proportions in the feed making up the largest part of livestock rations, the carbohydrate fractions,—crude fibre and nitrogen-free extract,—are of particular interest and importance. In 1938, Crampton and Maynard (1938) proposed that the carbohydrates be partitioned into three fractions, viz;., 1) cellulose, 2) lignin, and 3) other carbohydrates. This scheme was believed to be an improvement on the crude fibre—nitrogen-free extract partition as an index of the energy value of feedingstuffs of high lignin content such as pasture herbage. The plan however, was not entirely satisfactory, because of the problems of quantitative lignin determination in materials containing appreciable amounts of protein. In the belief that the lignin—cellulose—“soluble carbohydrate” partition of the total “carbohydrate” component of feeds was sound biologically as an index of this useful energy value.