THE UTILIZATION OF NON-PEOTEIN NITROGEN Urea, if fed in a suitable mixture and in the presence of readily available carbohydrate, such as starch, can replace part of the protein in the rations of ruminants, for both growth and milk production (33). This is true even when the protein replaced is of high nutritive value. Nonruminants cannot use urea in this way(34,35,36,37,38). It appears from recent work (39,40) that non-protein nitrogen is utilized by ruminants'because the rumen abounds with iodophilic bacteria. In the presence of added urea and added available carbohydrate the multiplication of these iodophilic bacteria leads to the synthesis in their bodies of considerable amounts of 'bacterial starch' and, of course, of protein. Ammonia derived preferably from intraruminal hydrolysis of urea can serve as a source of nitrogen for these bacteria. The source of carbohydrate seems to be of a more specific nature. Thus in vitro starch and various sugars, such as maltose and galactose, may provide for the energy requirements of the iodophile bacteria of the rumen(39). When, however, instead of receiving sugar or starch the bacteria are supplied with glycerol, multiplication does not occur and protein synthesis is replaced by proteolysis. It may well be that, while hexoses or more complex carbohydrates are available either for synthesis of bacterial protoplasm or for breakdown into fatty acids (41, 42, 43,44,45,46), glycerol is only available for the latter process. Pearson & Smith (39) are of the opinion that analyses of materials abstracted through rumen fistulas may often be"subject to large and relatively uncontrollable errors of sampling, as, indeed, McElroy & 6oss(46) admit. Pearson & Smith (39) have therefore preferred to incubate rumen contents in vitro using the microscopic appearance of the bacterial flora and the pH. of such incubated material as criteria of normality. They found that under certain easily reproducible conditions the chemical and bacteriological nature of such material is still comparable with normal rumen contents after 3-4 hi. of incubation in vitro. Using this method of incubation they have shown that urea added to rumen contents is rapidly hydrolysed to ammonia. They have also shown (39) that hydrolysis and synthesis of protein can both occur in the rumen, synthesis of protein being favoured by the addition of urea and a readily available carbohydrate. Smith & Baker (40) investigated the correlation between bacterial multiplication and protein and polysaccharide synthesis in the rumen. Counts of Protozoa, and of each of several morphologically distinct types of bacteria, were recorded both before and after incubation. The volume of zoogloea was also determined. All these, except the Protozoa, increased on incubation, the contribution made by Amylococcus chains being greatest. When material consisting mainly of Amylococcus chains was isolated by centrifugation of material freshly removed from the rumen, it was found to be capable of a twelvefold increase by reproduction when incubated in...