1. The paper describes the development and construction of an apparatus for maintaining a normal microbial population of the rumen under strictly controlled conditions over long periods of time, 2. The apparatus is simple to construct and operate. It is possible to do four replicate experiments at the same time.3. The results of three experiments are given. The experiments showed that when the steady-state was reached it could be maintained indefinitely, with the type and quantities of products of fermentation very similar to those in the rumen of donor animals, including the maintenance of normal protozoal populations for up to 49 d.4. It was found that within wide ranges, the digestibility of rations and the output of products were independent of dilution rate.5. Except for the lowest 'level of feeding', the digestibility was independent of the level of feeding. The output of products was proportional to the amount of food digested and was the same as would be expected in sheep on similar rations.6. An experiment in which a ration of hay was changed to a mainly concentrate ration showed that the fermentation characteristics were determined mainly by the food given.Many types of artificial rumen apparatus have been described (see review Czerkawski, 1976~). The apparatus constructed in this laboratory (Czerkawski & Breckenridge, 1969) was designed specifically for short-term work involving 6-8 h incubations and the possibility of measuring the gaseous exchanges was one of the main requirements. This apparatus and a small-scale version of it (Czerkawski & Breckenridge, 1970) have been used extensively in the studies of the mechanism of methane production and its inhibition. It was possible to 'trace out' several biochemical pathways and to develop new inhibitors of methane production.Several experiments in vivo (Czerkawski, Christie, Breckenridge & Hunter, 1975; Czerkawski, 19763) showed that the inhibition of methane production gives rise to gross changes in rumen metabolism. Some of these changes come about gradually over a period of weeks rather than days or hours and it is difficult to exert sufficient control or to interpret the results of experiments with animals. Therefore there was great need for a suitable longterm artificial rumen technique. A simple continuous culture procedure was discounted because it does not simulate the conditions in the rumen sufficiently (e.g., see Isaacson, Hinds, Bryant & Owens, 1975). Usually the protozoal concentrations decrease and often protozoa disappear altogether. Sometimes the bacterial metabolism bears little relation to that of the rumen of the donor animal and it is often necessary to stimulate the fermentation by infusing clarified rumen fluid. Another drawback in using a commercial fermentor is the provision of one or at most two reaction vessels; this limits severely the type of comparative work that can be undertaken.In designing the apparatus described here the following requirements had to be met: (a) the reaction vessels should contain solid phase (partly ...

1. A long-term experiment was made with the Rumen Simulation Technique (Rusitec), in which the fermentation of a mixed ration of hay (10 g/d) and bruised barley (5 g/d) was compared with the fermentation of the same diet in the presence of 2, 10 and 50 mg monensin/d. 2. Monensin depressed the production of acetic and butyric acids, markedly increased the production of propionic acid and virtually, eliminated the production of isovaleric acid. The production of methane was decreased in the presence of monensin, but this decrease could be accounted for entirely by the changes in the production of volatile fatty acids and redistribution of metabolic hydrogen. 3. The digestibility of dry matter (DM) in the rations declined in the presence of monensin. Determinations of the rates of digestion showed that the digestion of the readily-fermented food in the initial stages was not affected by monensin, but that at 24 h digestion had been inhibited by monensin. The inhibition was due entirely to its effect on the digestion of the fibrous components. Digestion of non-fibrous material was not affected. 4. The efficiency of microbial growth, expressed as g dry weight/mol ATP formed (YATP) and in terms of DM digested, tended to be increased by monensin. This however occurred only at high, non-practical doses. 5. Urease (EC 3. 5. 1. 5) was induced by the addition of urea of the fermentation, but monensin had no effect on urease activity. Although monensin increased the activity of protease in washed suspensions, more food protein apparently escaped degradation. This may have been due to decreased deaminative activity. 6. Monensin altered the microscopic appearance of the fermentation fluid, and changed the activity of some enzymes in sonicated extracts, including alkaline phosphatase (EC 3. 1. 3. 1), acetate kinase (EC 2. 7. 2. 1) and succinate dehydrogenase (EC 1. 3. 99. 1). These results are discussed in terms of known sensitivities of rumen microbes to monensin and their contribution to the fermentation as a whole.

I . Linseed oil was incorporated gradually into the diet of four sheep until the animals received 90 g additional fat/d. Attempts were made to measure changes in concentration of substances and rates of synthesis in the rumen directly, and by incubation of rumen contents in vitro (zero-time technique).2 . The high-fat diet increased the dilution rate and the volume of rumen contents and decreased the synthesis of diaminopimelic acid in the rumen. The number of protozoa decreased and the number of bacteria increased in the rumen of animals receiving the high-fat diet.3. The concentration of volatile fatty acids (VFA) in the rumen decreased for sheep given the high-fat diet, but the capacity of rumen contents to produce VFA in vitro increased.4. The incorporation of radioactivity from [14C]acetate into lipids during incubation of rumen contents in vitro increased with the amount of linseed oil in the diet. The greatest proportional increase was with the bacterial fraction of rumen contents.5. In the group of four animals used, one animal showed consistent differences in the magnitude of the measured variables. This animal appeared to have a smaller rumen, a lower dilution rate and larger concentrations of some substances in the rumen. A higher proportion of fatty acids appeared to be synthesized by the micro-organisms from this animal. (1972) found that the net lipid synthesis was greater with high-fat diets. In our earlier work (Czerkawski, 1966) when the intake of linseed-oil fatty acids was increased gradually to a level that constituted 18 % of the total energy content of the diet, there was inhibition of methane production, a small increase in faecal excretion of cellulose and crude lipid, but no change in faecal excretion of nitrogen. Further experiments in which linseed-oil fatty acids or linseed oil were added to the diet of sheep in increasing amounts (Czerkawski, 1973), showed that the changes in the fermentation patterns were similar for all sheep, but in some, lipid metabolism in the rumen appeared abnormal in as much as there was considerable accumulation of palmitic acid and smaller amounts of myristic acid. It was concluded on the basis of these experiments that the increases in concentration of palmitic acid were due to the increased synthesis of this compound in the rumen, and an indirect estimate of the magnitude of this synthesis was made. The value was similar to those obtained by direct measurements reported by other workers. However, there were a number of puzzling features. For instance, the changes in fermentation pattern, such as inhibition of methane production, decreased concentration of acetate and butyrate in the rumen, or increased excretion of lipid in the faeces, occurred in all sheep irrespective of whether there was a large synthesis of palmitic acid or not. It was suggested that the increase in lipid synthesis occurred in all sheep, but in those sheep that had unchanged concentrations of palmitic acid, other fatty acids and possibly more complex lipids were being synthesized. I...

I. Experiments with a small-scale artificial rumen have shown that of forty-two cornpounds tested the majority were fermented, as judged by the production of volatile fatty acids, but methane production was associated only with the fermentation of formate, certain hydroxyacids, pyruvic acid, primary alcohols (methanol, ethanol, propanol and butanol), glycerol and methyl compounds. With primary alcohols there was a stoichiometric relationship between methane production and the oxidation of the alcohols to the corresponding acids.2. The fermentation of rhamnose and 1,z-propanediol was studied in detail. With both compounds there was a temporary accumulation of lactic acid and a continuous net production of propionic acid. The initial rate of acetate production w-as rapid with rhamnose but decreased subsequently, whereas propionate continued to increase. With propanediol the net rate of production of acetate was slow at first and then increased. There was no increase in the production of butyric acid with either rhamnose or propanediol, and the endogenous methane production was inhibited by 20-40 %. There was evidence for the formation of an unidentified compound during fermentation of rhamnose and propanediol.

No abstract