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.

I. Three heifers were changed from a diet of equal parts of hay and barley cubes (50:~o diet) to one entirely of barley cubes given in three equal feeds throughout the day. Feed intake was restricted to 80 % of calculated appetite at the time of change and this percentage progressively decreased as the live weights of the animals increased.2. The change of diet had no significant effect on the volume of rumen fluid but the rate of outflow from the rumen was significantly lower on the barley diet than on the 50: 50 diet.3. Animals on the restricted barley diet developed an exceptionally high rumen ciliate population and the bacterial population was shown by Gram films to include a number of organisms typical of roughage-fed animals. In culture, organisms of the genus Bacteroides were predominant but these appeared largely as cocco-bacilli in the Gram films. This microbial population was associated with a higher proportion of butyric acid than of propionic acid in the rumen fluid.4. Occasional fluctuations in ciliate populations occurred in all three heifers. Decreases in ciliate number were paralleled by increases in propionic acid and decreases in butyric acid but not necessarily by a fall in pH. Under these conditions Gram films showed increases in bacteriodes-type rods and in certain curved Gram-negative rods.5. Rumen ammonia concentrations were on average lower and showed a different diurnal pattern when ciliate numbers were reduced. Lactic acid concentrations were low and were not affected by the size of the ciliate population.6. When the three heifers were given the barley diet ad lib. there was a decrease in rumen pH and a complete loss of rumen ciliates. The rumen bacterial population and the volatile fatty acid proportions were similar to those seen during decreases in ciliate number at the restricted level of intake. These changes also occurred in a fourth heifer which was changed fairly rapidly from the 50: 50 diet to a restricted amount of the barley diet. 7.Two steers which had never had access to roughage were changed from ad lib. to restricted intake of the barley diet and were later given an inoculum of rumen ciliates. The rumen microbial population and the pattern of fermentation so produced were similar to those found in the heifers on the restricted barley diet.8. Anomalous values were noted for total counts of rumen bacteria when free starch grains were present in the rumen fluid.9. It is concluded that large ciliate populations and high proportions of butyric acid can be produced in animals fed exclusively on a barley diet by suitable adjustment of the intake and the method of feeding. I t is postulated that the ciliate population may be largely responsible for the high butyric acid concentrations.The rumen microbial populations of mature steers given ad lib. access to a barley diet have been described by Eadie, Hobson & Mann (1967) and have been shown to be similar in many respects to those found in young ruminating calves given large amounts of starchy concentrates (Eadie, 1962). A complete...

I . Two pairs of Friesian steers were changed from ad lib. to restricted intake of a pelleted barley diet and were maintained on this latter intake for periods of 18 or 25 weeks. The restricted level of intake was 70 g/kg0-73 and was adjusted weekly according to individual live weights. The daily allowance was given in three equal feeds during day-time.2. After an initial period of 9 weeks on the restricted diet, during which all four animals were kept free of rumen ciliate protozoa, one member of each pair was given an inoculum of rumen ciliates. Eight weeks later, the ciliate-free member of the younger pair of steers was similarly inoculated. Observations were made on the rumen bacterial and protozoal populations and on changes in rumen p H and volatile fatty acids (VFA) throughout each treatment period. The concentrations of urea and haemoglobin in blood and of glucose and amino acids in plasma were examined on one occasion in each animal.3. In the absence of ciliates, restriction of intake resulted in rumen pII values and molar proportions of VFA similar to those normally encountered on an ad lib. intake of a barley diet. A decrease in bacterial numbers and certain minor changes in bacterial types were observed on changing from ad Zib. to restricted intake but the resultant population under ciliate-free conditions was basically the same as that found later in the faunated animals. In culture, organisms of the genus Bacteroides werc prcdominant.4. Large populations of rumen ciliates were established in each animal inoculated. Relative to the ciliate-free periods, the presence of ciliates resulted in an increase in rumen pH, a reduction in total VFA concentration and a decrease in the ratio of propionic to butyric acid in rumen fluid. It is concluded that these changes are a direct effect of ciliate activity.5. Conditions within the rumen remained more stable from day to day when large ciliate populations were present than when ciliates were absent. In one animal, spontaneous fluctuations in ciliate number were accompanied by corresponding changes in rumen pH and VFA proportions. 6 . Significant differences were observed between faunated and ciliate-free animals in the concentration of urea in blood and of glucose in plasma; only minor differences were noted in blood haemoglobin and plasma amino acid concentrations.

The level of volatile fatty acid in the peripheral blood of sheep is considerably higher than in non-ruminants, arterial levels reaching 10 mg. per cent. (as acetic acid) or higher. That the level is related to fermentation in the rumen is indicated by its steady decrease during fasting. In sheep fed on rations of widely differing composition, 86-95 per cent. (molar basis) of the total volatile fatty acid in arterial blood was found to be acetic acid. Small amounts of propionic, butyric, and at least one other acid were also present. The removal of acetic acid from the arterial blood during its passage through the tissues of the head is considerable, and the extent of its removal is closely dependent upon its arterial level. There was no evidence that acids other than acetic were removed in significant quantities. The low normal blood-glucose level in fed sheep is associated with a smaller uptake of glucose by the tissues of the sheep than occurs in non-ruminants and this low uptake is compensated by a higher uptake of acetic acid. Arterio-venous differences of glucose and of volatile fatty acid have been measured in two dogs, after feeding and during fasting, and are compared with the results obtained from sheep. These findings are discussed in the light of present knowledge of the influence on ruminant metabolism of the rumen and the digestive processes therein.

I . Short-term in vitro incubations of whole rumen contents have-been used to measure the rate of production of volatile fatty acids (VFA) in two heifers given a diet containing 85 yo barley and 15 yo protein-mineral-vitamin supplement. Daily intake of the diet was restricted to 5.1 kg which was given in three equal feeds. In one heifer (no. 794), VFA production was measured over a 24 h feeding cycle; in the other (no. 832), a single 6 h feeding period was examined.2. The feeding schedule outlined above was found in previous work to result in a major alteration in the rumen environment and in the end-products of rumen fermentation relative to findings in animals given the same diet ad lib. Some of these earlier observations have been confirmed and extended in the present work.3. In heifer 794, the rate of VFA production reached a maximum (52-62 m-moles/l. h) I 5 minafter each feed and then declined steadily throughout the remainder of the feeding period. Consistent changes with time were observed in rumen pH and in the concentration of VFA and dry matter in rumen contents during each feeding period.4. Simultaneous measurements of rumen fluid volume and outflow allowed estimates to be made of VFA production and absorption in the entire rumen during each feeding period.In heifer 794, total VFA production amounted to 18.2 moles/day of which approximately 85 yo was absorbed directly from the reticulo-rumen. On a caloric basis, total daily production accounted for 36. j % of the digestible energy intake. j . In heifer 832 there were significant differences in the rate of VFA production and in rumen pH and VFA concentration between two sets of observations made some 8 weeks apart. These differences appeared to be related to changes in the rumen microbial population and in particular to a marked alteration in the numbers of ciliate protozoa. The possible role of rumen ciliates in modifying both the rate of VFA production and the environment within the rumen is discussed.