Modification of the Rumen Bacterial Flora by Feeding Cattle Ground and Pelleted Roughage as Determined with Culture Media with and Without Rumen Fluid

Abstract

The numbers and kinds of predominating bacteria and most probable numbers (MPN) of eellulolytie bacteria were determined in an experiment designed as a 2 X 2 Latin square, with two nonlaetating cows fed diets ad libitum consisting of artificially dried grass and the same grass ground and pelleted. Two nonselective anaerobic roll tube media--a rumen fluid n'mdium (Medium 98-5) and one in which tureen fluid was replaced by volatile fatty acids, heine, and small amounts of yeast extract and trypticase (Medium 10) were used to obtain colony counts and to isolate strains for presumptive identification. The mean colony count for Animal D, 15.2 X 109 per gram, was significantly higher than that for Animal C, 11.1 X 109 per gram. Counts were highly significantly higher when animals were fed ground grass (15.7 X 10 s per gram) as compared to long grass (10.5 X 109 per gram), and there was a highly significant difference between mean counts on Medium 10 (15 X 109 per gram) and Medium 98-5 (10.7 X 109 per gram). Much of the difference in counts for both culture media and diets was due to the much higher mean counts on Medium 10 (20.7 X 109 per gram) as compared to Medium 98-5 (10.9 X 199 per gram) when ground grass was fed. The MPN of cellulolytie bacteria were similar for the two animals when long grass was fed and for Animal C when either diet was fed. These counts were depressed when Animal D was fed ground grass. The presumptive identification of 74 to 109 strains of predominating bacteria, isolated from each medium inoculated from each animal on each diet, showed a remarkable similarity in distribution of species in the two animals