The Digestion of Cellulose by Sheep: The Extent of Cellulose Digestion At Successive Levels of the Alimentary Tract

Popoff (1875) observed the formation of methane in the rumen as a product of the fermentation of crude fibre; Zuntz described similar fermentative changes in the caecum, and Hofmeister (1880) showed that they also occurred in the colon. Tappeiner (1882) attempted to assess the extent of digestion of crude fibre of the food at different levels as it passed along the alimentary tract of the ruminant. He took samples of the contents of the rumen, small intestine, and caecum from a freshly killed sheep, and estimated the crude fibre which disappeared after ‘complete ‘fermentation at body temperature. From this he concluded that the greater part of the crude fibre was digested in the rumen, a smaller amount in the caecum, and none at all in the small intestine.

A number of workers (Williams & Olmstead, 1935; Norman, 1935; Crampton & Maynard, 1938) have stressed the serious limitations of conventional analyses which divide the carbohydrates of plant foodstuffs into crude fibre and N-free extractives, and have suggested that the analysis of such foods would be more logical and informative if the carbohydrates were divided into better defined groups, such as cellulose, lignin, and other carbohydrates. This proposal has been adopted and extended by Hale, Duncan & Huffman (1940) in their study of digestion in the bovine rumen. These authors developed a method for estimating the extent of digestion in the rumen by the use of ‘lignin ratios’ and concluded that 85% of the digestible cellulose in alfalfa hay was broken down there. ‘

The following experiments were carried out in this laboratory during 1936. They were mentioned in the Annual Report of this laboratory (Commonwealth of Australia, Council for Scientific and Industrial Research, 1937−38).

The lignin ratio procedure was used to determine the extent of cellulose digestion at successive levels of the alimentary tract of the sheep. The method is based essentially on a procedure suggested by Bergeim (1926) for the determination of digestion coefficients. After adding Fe₃O₄ to the food Bergeim determined the ratio of the amount of a given nutrient to the amount of iron in the food and in the faeces, and from this the percentage utilization was calculated. Later Heller, Breedlove & Likely (1928) used the natural iron content of the food as a reference substance, to ensure a more uniform distribution; Gallup (1929) and Gallup & Kuhlman (1931) made use of silica for the same purpose.

The validity of such procedures depends fundamentally on the complete indigestibility of the reference substance, and the exact concurrence of its passage through the alimentary tract with that of the nutrient in question. Knott, Murer & Hodgsen (1936) came to the conclusion that added materials, such as iron oxide, were unsuitable for the calculation of digestion coefficients in ruminants because they were unlikely either to remain evenly distributed in the fodder, or pass along the tract at the same rate as the rest of the food. For a similar reason silica was considered also to be unsuitable ; although it is a natural and intimate constituent of plant structures, sand is nearly always present in fodders and may gather in certain parts of the tract rather than move along it at an identical rate with the rest of the food. The absorption of part of the natural iron of the fodder debars its use as a critical reference marker.

Hale et al. (1940) have employed lignin as a reference substance for the calculation of digestion coefficients in the rumen. They claimed that it was digestible to a variable extent (up to 24%) in the lower levels of the tract. There have been other claims that lignin is not undigested in various animals including the ruminant (cf. Konig, 1907; Csonka, Phillips & Jones, 1929; McAnally, 1942), and there are others to the contrary (Paloheimo, 1925; Rogozinski & Starewska, 1927; Crampton & Maynard, 1938). As lignin is not a chemical entity the term, of necessity, has often been used loosely to signify a mixture of similar complexes, some of which may be prone to change in the alimentary tract, and thus may not be expected to react identically towards the various reagents and procedures which different authors have used for its estimation. Since the different methods employed are all empirical, the origin of the controversy is clear enough. Norman (1935) directed attention to this and pointed out that experiments which have depended on the fate of isolated lignin added to the diet are not conclusive since the biological availability of lignin is affected by the processes of isolation.

Lignin was used as a reference marker in the following experiments after it was proved that, as determined in a mixture of wheat straw and lucerne hay by the method of Norman & Jenkins (1934), it was not digested to any appreciable extent by sheep.

The evidence that lignin is not digested was drawn from the study of the intake and output of lignin by three sheep fed on a constant diet of wheat straw and lucerne hay for several months while confined to metabolism cages described by Marston (1935). During the fourteen day collection periods the fodder eaten and the faeces voided were measured with meticulous care. Lignin was determined by the method of Norman & Jenkins (1934) in samples from each and the balance drawn up. The data are summarized in Table 1.

It may be noted that these results are in agreement with those of Crampton & Maynard (1938) for the digestion of lignin by steers. A different method was used for the determination of lignin, and in which different fodders were fed.

Four sheep which had been fed for some weeks previously on a constant ration composed of wheat straw and lucerne hay were slaughtered, and samples of ingesta were taken from the rumen, omasum, abomasum, small intestine, caecum and colon. The dried, ground samples were subjected to analyses for cellulose and lignin. The method for the determination of lignin has been referred to previously ; the method used for the determination of cellulose was that recommended by Norman & Jenkins (1933).

Similar analyses were made of samples of the fodder which the animals had been fed.

Results of these analyses are listed in Table 2.

The data indicate that, of the cellulose present in the fodder, 40−45 % was digested before the food passed into the abomasum, and an additional 15−20 % was digested in the large intestine. During this second fermentation 7−11 % disappeared in the caecum, and 4−9% in the colon. No appreciable loss of cellulose occurred in the abomasum or in the small intestine.

Since the total digestion of cellulose amounted to about 60%, it is apparent that approximately 70% of the cellulose that was digested was broken down in the rumen, 17% in the caecum and 13% in the colon.

The claims of Hofmeister (1880) and Tappeiner (1882, 1884) that some digestion of cellulose occurs in the caecum and colon of ruminants were supported by the work of the Dresden school (Ellenberger, 1915-16), although it was stated that the quantity that disappeared in the colon was small. Trautmann & Asher (1939), however, found no fermentation of chemically pure cellulose which had been introduced directly into the caecum and colon of a goat, even after a prolonged sojourn there. They observed no loss of cellulose when rumen contents were introduced into the caecum, and after 24 hr. there, they considered it to be free of rumen flora. These workers claimed that the time ingesta stay in the caecum and colon, together with the fact that there is a rapid water resorption from these organs, must be inhibitive to fermentation processes.

The present experiments, however, support the contention of the earlier investigators that this portion of the tract plays a significant part in the digestion of cellulose by the sheep. Indirect evidence which further supports this view is to be seen in the work of Barcroft, McAnally & Phillipson (1944) who found that volatile fatty acids, the acknowledged products of cellulose fermentation, are present in the blood draining the caecum in considerably greater quantity than in the blood of the peripheral circulation.

1. Lignin, as determined in these experiments, has been shown to be undigested by sheep.

2. Examination of cellulose-lignin ratios in the fodder, and in material taken from successive levels of the alimentary tracts of sheep indicated that 70% of the digestible cellulose of the fodder was broken down in the rumen, 17% in the caecum, and 13 % in the colon. No digestion of cellulose occurred in the abomasum, or in the small intestine.

The author is indebted to Mr H. R. Marston, Chief of the Division, for suggesting the problem, for the specimens from which the lignin balances were determined, and for his help and criticism throughout.