METHANOGENESIS IN MONOGASTRIC ANIMALS

BENT BORG JENSEN Danish Institute of Animal Science, Department of Nutrition, Research Centre Foulum, P O. Box 39,

DK-8830 Tjele, Denmark

Abstract. Studies of methanogenic bacteria present in monogastric animals are still scarce. Metha- nogens have been isolated from faeces of rat, horse, pig, monkey, baboon, rhinoceros, hippopotamus, giant panda, goose, turkey and chicken. The predominant methanogen in all except the chicken and turkey is species of Methanobrevibacterium. The chicken and turkey harbour species of Methanoge- nium. In pig the population of methanogenic bacteria is more than 30 times as dense in the distal colon as in the caecum. This finding is in agreement with the finding that the rate of methane production is much higher in the colon than in the ceacum. The amount of methane excreted clearly seems to depend on the amount of non-starch polysaccharide intake.

The directly measured methane production rate in pigs is from 3.3 to 3.8 times lower than the amount expected from stoichiometric estimates. These data, together with data showing that only small net amounts of hydrogen and small amounts of methane are produced in the ceacum and proximal colon where the microbial activity is high, clearly indicate that hydrogen sinks other than methane production are involved in hydrogen removal in the hindgut of pigs and probably also in other monogastric animals.

Methane production by monogastric animals is lower than methane production by ruminants. However, methane production by large herbivorous monogastric animals such as horses, mules and asses is substantial (up to 80 1 per animal per day). Methane production by rodents and avians is low. In general, methane production by wild animals is lower than methane production by domestic animals. It is concluded that the contribution of monogastric animals to the global methane emission is negligible, as it only represent about 5% of the total methane emission by domestic and wild animals of 80 Tg per year.

1. Introduction

Two different types of anaerobic microbial ecosystems produce significant amounts of methane. One type is the intestinal tract ecosystem such as the complex forestom- ach of ruminants and the large intestine of monogastric animals. This type of ecosystem does not completely convert substrates to CH4 and CO2. They accu- mulate significant quantities of acetate, propionate, and butyrate (Table I). The predominant substrates for methanogens in these ecosystems are H2 and CO2. The other type of ecosystem, e.g., swamps, rice paddies, terrestrial and marine aquatic sediments, and anaerobic digestion systems, are complete bioconversion systems. Acetate, propionate, butyrate, and H2 and CO2 are formed in the initial stages of the process. However, propionate and butyrate are further converted to acetate and H2; acetate, together with Hz and CO2 are substrates for methanogens resulting in complete bioconversion of organic matter to CH4 and CO2 (Table I).

The principal natural substrates in both types of ecosystem are polysaccharides, proteins and lipids. Production of acetate, propionate, butyrate, H2 and CO2 requires

Environmental Monitoring and Assessment 42:99-112, 1996. (~) 1996 Kluwer Academic Publishers. Printed in the Netherlands.

100 BENT BORG JENSEN

Table I Incomplete and complete anaerobic bioconversion of carbohydrate to methane

Incomplete (gastrointestinal tract)

57.5 C6H1206 --+ 65 CH3COOH + 20 CH3CH2COOH + 15 CH3CH2CH2COOH + 35 CH4 +

60 CO2 + 25 H20

Complete (swamps, ricepaddies, anaerobics sewage digestion systems)

57.5 C6H1206 ~ 172.5 CH4 + 172.5 CO2

Adapted from Miller (1991).

complex integrated activities between several different microbial species. Further, the removal of H2 by the methanogens has a pivotal influence on the activities of the non-methanogenic microorganisms in these ecosystems.

A major difference between the two types of ecosystems is turnover time. Intestinal bioconversion systems have turnover times of approximately 1 to 2 days. In contrast, the turnover times of complete bioconversion systems are weeks to months, because microbial conversion of acetate, propionate, and butyrate to CH4 and CO2 involves bacteria that have relative long generation times. These types of bacteria cannot be sustained in the testinal ecosystem due to their slow rates of multiplication.

The aim of the present review is to describe the characteristics of the ani- mal intestinal tract ecosystem of monogastric animals with special reference to methanogenesis.

2. The Digestive Tract of Various Monogastric Animals

There is a widespread tendency to overlook or underestimate the significance of microbial fermentation in monogastric animals. However, the digestive tract of some monogastric animals is quite complex and provide many environments for fermentative bacteria. In general the gastrointestinal tract is a tube extending from the mouth to the anus. It is divided into various well defined anatomic regions whose structure and function reflect the diet and the life style of the particular animal species; as a consequence animal species can be classified into several groups based on the characteristics of their gastrointestinal tract (Figure 1).

Ruminants represent the most developed and specialized group, from the point of view of microbial fermentation and the ability to use non-starch polysaccarides unavailable to animal digestion. However, pregastric fermentation is not restricted to ruminants, and has been reported in primates, rodents, ungulates and marsupials (Van Soest, 1984).

METHANOGENESIS 1N MONOGASTRIC ANIMALS 101

A,) Pony B) Pig

C) Mink [3) Goose Figure 1. The gastrointestinal tract of various monogastric animals. (A) Herbivore (horse); (B) omnivore (pig); (C) carnivore (mink); (D) avian (goose).

102 BENT BORG JENSEN

Monogastric herbivorous show a variety of adaptations. One group includes large non-ruminant herbivores such as the perissodactyls (horse, rhinoceros, etc.) and other large herbivores (elephant, etc.). In all of these, microbial fermentation is more important in the sacculated colon than in the caecum. Herbivorous rodents and other small herbivorous mammals characteristically exhibit large caeca and unsacculated colons. Adaptations employing the caecum as the main fermentative site are often associated with coprophagy as in the rabbit, hare and lemming.

In omnivores such as man and pig the caecum is much reduced, but the colon is sacculated (Figure 1). The sacculation of the colon and the relative small caecum may represent a special adaptation of the lower bowl to fermentation.

Carnivores (mink, cat, dog, etc.) comprise another class of mammals. Their gastrointestinal tract is simple. The intestine is relatively short with little or no caecal capacity and an unsacculated colon (Figure 1).

The anatomy of the avian gastrointestinal tract is most notably different from that of mammals in the mouth area, in the presence of a crop in the oesophagus, in the presence of a muscular stomach or gizzard and in most cases in the presence of two caeca. The large intestines of birds are short and are not sharply demarcated from the rectum and small intestine (Figure 1).

3. Microbial Acitvities in Various Regions of the Digestive Tract in Different Monogastric Animals

The distribution of the microbiota within the gastrointestinal tract differs between animal species. Investigations of the intestinal bacteria have concentrated on studies of the human intestine and the rumen. However, our knowledge of the components of the gut microbiota of the pig and the chicken is slowly but continuously increas- ing. Information for other species of monogastric animals is much less detailed.

The population level of the microbiota in various parts of the gastrointestinal tract of monogastric animals depends on the doubling time of the microorganism under the physico-chemical conditions in the part of the gastrointestinal tract under investigation, and the emptying rhythm of the particulate part (reservoir). The stomach and the crop is the first reservoir where the ingesta spend some time, and where the microflora may multiply. In contrast to humans, pig, rodents and fowl harbour a permanent microflora in the proximal regions of the digestive tract, consisting of lactobacilli and streptococci. These permanent populations can be achieved in the digestive tract of pigs, rodents and fowl because the lactobacilli and streptococci associate with the stratified squamous epithelial surface lining that part of the proximal digestive tract of these animal species (Tannock, 1992). In contrast to the stomach, most of the small intestine is unsuitable for bacterial proliferation in healthy animals, simply because the intestinal transit is too rapid to allow time for microbial division. Proliferation occurs in the upper part of the small intestine only when there is adhesion to the gut wall, or mechanical obstruction

METHANOGENESIS IN MONOGASTRIC ANIMALS

Table II Microbial activity in intestines from various animals

103

Species Body weight (kg)

Intestinal contents ATP ATP ATP (mg) (#g (kg BW) -1) (#g (kg BW~ -1)

Mink Goose Duck Rat a Rat b Chinchilla Pig Pig Pig Cow

1.0 0.1 120 120 4.0 0.5 130 180 4.0 0.6 140 200 0.1 0.1 800 450 0.1 0.3 2800 1600 0.5 0.5 1000 910 10 4.1 420 750 40 27.2 1030 1800 90 75.9 840 2600

500 5000 8000 41000

a Low fiber diet, b high fiber diet.

leads to stasis of digesta, with pathological consequences. Alimentary stasis is the rule of the lower part of the small intestine (ileum) and the large intestine (caecum, colon and rectum) and all gut bacteria have sufficient time to multiply there, resulting in a large microbial population. The hydrolytic digestive function of the large intestine of pigs and other monogastric species is carried out by this rich and diverse population of anaerobic bacteria (Moore et al., 1987). As shown in Figure 2, the density of the microbial population in the caecum and colon of pigs amounts to 101~ bacteria per g of digesta, comprising more than 400 different species, of which only few have been described in full (Moore et al., 1987).

The microbial activity (measured as the concentration of ATP) in various seg- ments of the digestive tract of several monogastric animals is shown in Figure 3. In all animal species highest microbial activity was found in caecum and the proximal segments of the colon; however, it is obvious that microbial activity differs between the animal species.

Table II shows the total microbial activity found in the entire gut per kg of metabolic weight of the animal. These data clearly indicate that the microbial activity is much more important to ruminants than to monogastric animals. Further, the data illustrate that the microbial activity is higher in pigs than in fowl, and that the microbial activity depends on the age of the animal and on the diet. Probably the single most important factor influencing the microbial activity in the gastrointestinal tract of monogastric animals is the amount and type of substrate available to the microbiota. In particular, non-starch polysaccarides are the principal energy substrate for large intestinal microbial fermentation, and the amount as well as the chemical and structural composition of the carbohydrate are important factors for

104 BENT BORG JENSEN

11-

10

"T 8 A c~

o-1

7 i1 C)

C~ ..9.o 6

Z,

~x~X anaerobe X

bacteroides x ~ lac tobac i l l i

~ X ~cteria

:oliforme

enterococci

~ y e a s t

_

I J small intestine [,..J colon

;tomach caecum Figure 2. The density of selected populations of the microbiota in various regions of the digestive tract of pigs.

the microbial activity in the digestive tract (Bach Knudsen et al., 1991, 1993a, b; Jensen, 1988; Jensen and J~rgensen, 1994; Macfarlane and Cummings, 1991).

4. Methanogenic Bacteria in Monogastric Animals

Studies of methanogenic bacteria present in monogastric animals apart from man is still scarce and mainly qualitative. Methanogens have been isolated from fae-

METHANOGENESIS IN MONOGASTRIC ANIMALS 105

c- (9 C 0

C7~

n

::1,

Figure 3. animals.

120

100

80

80

40

20

0

D

ceecotrofe

J7, J smelt intestine stomach

gizzard

rabbit

rat (high fiber)

chicken

~ pig

goose

I ]~ colon

caecum Microbial activity in various regions of the gastrointestinal tract in different monogastric

ces of rat, horse, pig, monkey, baboon, rhinoceros, hippopotamus, giant panda, goose, turkey, and chicken (Miller and Wolin, 1986; Miller, 1991). The predomi- nant methanogen in all except the chicken and turkey is species of Methanobre- vibacterium (Miller and Wolin, 1986). The chicken and turkey harbour species of Methanogenium (Miller and Wolin, 1986). Methanobrevibacterium is also the most common methanogen in faeces of man and insects and in the forestomach of the bovine rumen (Miller and Wolin, 1986).

Quantification of the number of methanogenic bacteria in the faeces of different animal species including man has been done by Sorlini etal. (1988). Highest values were found in pig faeces, followed by human, cattle and rabbit faeces (Table III). Butine and Leedle (1989) showed that in pigs the population of methanogenic

106 BENT BORG JENSEN

Table III

Numbers of anaerobic bacteria and methanogenic bacteria present in faeces from various mammals

Species Anaerobes a Methanogens a

Mink 4 x 10 I1 1 x 107

Cattle 3 x 1011 1 x 106 Pig 3 x 1011 1 x 108

Rabbit 7 x 109 4 104

a Bacteria per gram dry weight (MPN). Data adapted from Sorlini et al. (1988).

bacteria was more than 30 times as dense in the colon as in the caecum; a finding that is in agreement with the results of Jensen and JCrgensen (1994) who found that the rate of methane production and the concentration of methane in the caecum and the proximal colon were low, followed by a steady increase in the successive segments of the hindgut. This finding is further supported by the results of Robinson et al. (1989), who found the rates of methane production of colonic samples to be ninefold higher than those of caecal samples.

5. Methane Production in the Gastrointestinal Tract of Pigs

Apart from man and pig very little information exists about methane production in the digestive tract of monogastric animals. The main products of non-starch polysaccharide fermentation in the gastrointestinal tract of pigs are short chain fatty acids, lactate, and various gases (Macfarlane and Cummings, 1991). The short chain fatty acids produced are physiologically important especially in the large intestine, where butyrate in particular is required to maintain the health of the epithelial cells lining the gut. Three gases, H2, CH4 and CO2, are produced in appreciable volumes by the intestinal microbiota. The main part is excreted in flatus, while only a smaller part is absorbed into the bloodstream and excreted in expired air. It has been indirectly shown by using respirometry that methanogenesis occurs in the lower gastrointestinal tract of pigs. In contrast to ruminants, methanogenesis accounts for minimal loss of digestible energy in pigs (Christensen and Thorbek, 1987). A number of investigators have used breath and flatus H2 and CH4 measurements to quantitate fermentation in the large intestine. However, very little information exists about the relative rates of gas production in different regions of the gastrointestinal tract of monogastric animals.

In a recent study by Jensen and J~rgensen (1994) the microbial activity, the composition of the gas phase and the relative rate of gas production were investi- gated in various regions of the gastrointestinal tract of pigs (120 kg LW) fed either a low (5% NSP) or a high fibre diet (27% NSP). A dense population of culturable

METHANOGENESIS IN MONOGASTRIC ANIMALS 107

anaerobic bacteria, a high ATP concentration and high adenylate energy charge (AEC) values were found in the last third of the small intestine, indicating that a substantial microbial activity takes place in that portion of the gut.

The highest microbial activity (highest bacterial counts, highest ATP concen- tration, high adenylate energy charge, and low pH) was found in the caecum and proximal colon. Higher microbial activity was found in the stomach and all seg- ments of the hindgut in the pigs fed the high fibre diet than in the pigs fed the low fibre diet. Very good correlations were found between the concentration of gases in specific portions of the gut and the in vitro production rates at the same site. The highest concentrations and highest production rates for H2 were found in the last third of the small intestine, while only small concentrations and low production rates of H2 were detected in caecum and colon. No methane in the gas phase and no methane production could be detected in the stomach or small intestine. The production rate and concentration of methane was low in the caecum and the proximal colon, followed by a steady increase in the successive segments of the hindgut. A very good correlation was found between in vivo and in vitro measurements of methane production. The amount of CH4 produced by pigs fed the low fibre diet was estimated to be 1.4 litres per day per animal (120 kg LW) and the corresponding amount in vivo to be 1.0 litre per day. Substantially higher amounts of CH4 were produced by pigs fed the high fibre diet (12.5 litres per day in vivo and 12.2 litres per day in vitro).

Jensen and JOrgensen (1994) concluded from their results that although the highest microbial activity was found in the caecum and proximal colon and although hydrogen production is an obligate part of anaerobic fermentation in the hindgut, only small net amounts of hydrogen were produced in these segments of the gut. In contrast to the rumen, where all microbially produced H2 is metabolized to methane, their results show that only small amounts of methane were produced in the caecum and proximal colon in pigs. This strongly indicates that hydrogen sinks other than methane production are involved in hydrogen removal in the caecum and proximal colon of pigs.

In contrast to man where about half the population are methane-producers, all pigs seems to produce methane in the hindgut (Christensen and Thorbek, 1987; Jensen and JCrgensen, 1994; Jcrgensen, unpublished; Zhu etal., 1993). As shown by Christensen and Thorbek (1987), the daily CH2 excretion by pigs increases linearly with increasing LW, probably as a result of increasing feed intake. The amount of CH4 excreted per kg dry matter feed intake varied from 2.5 to 5.2 litres depending on the composition of the diet. Zhu et al. (1993) and Jensen and J~rgensen (1994) found that the amount of methane excretion increase with increasing fibre content in the diet. Calculating the daily amount of non-starch polysaccharide intake in the experiments of Christensen and Thorbek (1987), Zhu et al. (1993), Jensen and JOrgensen (1994), and JCrgensen (unpublished results) and plotting this daily NSP intake versus the daily amount of methane excreted, clearly shows that the amount of methane excretion by pigs is dependent on the NSP intake (Figure 4).

108

14

BENT BORGJENSEN

12

A

10 "0

C o 8 e ~

0 X | 6 t -

e- - 4

2

0 0 9

O

A A

,o e.6 0 9

~

O

0

0 I I I I I 1 I I I

0 100 200 300 400 500 600 700 800 900 1000

NSP consumed (g/day) Figure 4. Relation between daily amount of non-starch polysaccharides (NSP) consumed and daily methane excretion in pigs. Symbols correspond to different studies; A, Christensen and Thorbek (1987); A, Jensen (unpublished); l , Jensen and Jcirgensen (1994); O, Jr (unpublished); O, Zhu et al. (1993).

6. Stoichiometric Estimates of Methane Production in Pigs

Methane and propionate are the two major recognized hydrogen sinks in microbial fermentation in the rumen (Hungate, 1966; Orskov et al., 1968). A stoichiometric carbon-hydrogen balance equation based on measurements of VFA molar propor- tions and the amount of carbohydrate fermented can be used to predict the amount of methane produced (Van Soest, 1982). This technique has been widely used in ruminant studies and appears to give sensible results (Orskov et al., 1968; Whitelaw et al., 1970). However, this is not the case for pigs. As shown by Zhu et al. (1993), the stoichiometric estimates of methane was from 3.3 times (cereal-based diet) to 3.6 times (cereal-based diet supplemented with 30% unmolassed sugar-beet pulp) higher than those directly measured. The same type of calculations with the data of Jensen and JCrgensen (1994) and J~rgensen and Jensen (1994) showed that the directly measured methane production rates were 3.8 times lower than the amount expected from the stoiochiometric estimate. These data together with the data of

METHANOGENESIS IN MONOGASTRIC ANIMALS 109

Jensen and J~rgensen (1994) showing that only small net amounts of hydrogen and small amounts of methane were produced in the caecum and proximal colon, clearly indicate that hydrogen sinks other than methane production are involved in hydrogen removal in the hindgut of pigs and probably also in other monogastric animals.

7. Hydrogen Sinks other than Methane Production

As pointed out by Zhu et al. (1993) and Jensen and JCrgensen (1994) a number of possible pathways for disposal of H2 other than methane production exist in the gut. These include saturation of unsaturated fatty acids, reduction of nitrate to ammonia, reduction of CO2 to acetate, reduction of sulphate to sulphide, reduction of oxygen that diffuses from the blood into the gut lumen and microbial synthesis of lipid and amino acids.

Unsatured fatty acids are saturated distal to the ileal-caecal junction (J~rgensen and Just, 1988) and Christensen and Thorbek (1987) have shown that inclusion of soy-bean oil in a basal diet reduced the amount of methane excreted by pigs. Further, the pig intestinal microbiota contains acetogenic bacteria (DeGraeve et al., 1990) and sulphate-reducing bacteria (Butine and Leedle, 1989). It has been shown that when sulphate is available, sulphate-reducing bacteria have a higher substrate affinity for hydrogen than methanogenic bacteria and that methane excre- tion only occurs when sulphate is absent or limiting (Lovley et al., 1982; Lupton and Zeikus, 1985). On the other hand, it is known that acetogenic bacteria are dis- placed by methanogenic bacteria in competition for available hydrogen (Prins and Lankhorst, 1977). Thus acetogenic bacteria will only become active when there is little hydrogen uptake by sulphate-reducing bacteria or methanogenic bacteria. However, acetogenesis may be important in the caecum and proximal colon of pigs, since it has been shown that pH may be an important factor in controlling the rate of hydrogen uptake (Gibson et al., 1990). In vitro studies have shown that sulphate-reducing and methanogenic bacteria in human faeces are relatively pH-sensitive, preferring an environment that is neutral or slightly alkaline, whereas highest rates of acetogenesis occurred at acidic pH. Since the pH in the caecum and proximal colon of pigs is low (pH 5.0-5.5) whereas the pH in the distal colon approaches neutrality (pH 6.5-7.0) acetogenesis may be the dominating hydrogen sink in the caecum and proximal colon while methanogenesis may dominate in the distal colon.

8. Methane Production by Various Monogastric Animals

The daily methane production by various domestic and wild monogastric animals together with the daily methane production by ruminants such as cow and sheep are

110 BENT BORG JENSEN

Table IV Daily intestinal methane production by various domestic and wild animals

Species Body weight CH4 produced CH4 produced (kg) (litre day- t) (litre day- l

(kg body weight)- 1)

Domestic animals

Cow 500 230 0.46 Sheep 40 30 0.75 Horse 500 80 0.16 Mules/assess 250 40 0.16 Pig 100 10 0.10 Man 50 2 0.05 Goose 0.4 0.05 0.01 Rat 0.3 0.001 0.03

Wild animals Warthog 45 4 0.09 Zebra 200 20 0.10 Rhinoceros 800 60 0.07 Hippopotamus 1000 70 0.07 Elephant 1700 110 0.06

Data from Champ et al. (1990); Christensen and Thorbek (1987); Crutzen et al. (1986); Jensen (unpublished) and Miller (1990).

shown in Table II. It is obvious that methane production by monogastric animals is lower than methane production by ruminants. However, methane production by large herbivorous monogastric animals such as horses, mules and asses is also substantial. Methane production by rodents such as the rat and by avians such as the goose is low. In general, methane production by wild animals is lower than methane production by domestic animals, probably due to a lower feed intake.

9. Overall Methane Production by Monogastric Animals

Crutzen et al. (1986) have estimated the global production of methane by several monogastric domestic animals and have also tried to estimate the contribution of wild monogastric animals to the global production of CH4. The yearly production by pigs was estimated as 1.5 kg per pig per year in the developed countries and to 1.0 kg per pig per year in developing countries. Multiplying those values by the pig population of 500 million in the developing countries and of 300 million in the developed countries, this yields about 1 Tg CH4 per year.

The mean yearly methane production by horses was estimated to 18 kg per animal, and the world population of 65 million horses therefore to produce a total

METHANOGENESIS IN MONOGASTRIC ANIMALS 1 1 1

of 1.2 Tg CH4 per year. A mean production rate of 10 kg per animal per year was estimated for the global mule and donkey population of 54 million, leading to a total emission of 0.5 Tg CH4 per year by these animals.

Methane production by wild monogastric animals, of which zebras, elephants and other large herbivore are the most important contributors, was estimated to be less than 0.6 Tg per year.

Therefore, as pointed out by Crutzen et al. (1986), it must be concluded that the contribution of monogastric animals to the global methane emission is negligible, as it only represents about 5% of the total methane emission by domestic and wild animals of 80 Tg per year.

References

Bach Knudsen, K. E., Jensen, B. B., Andersen, J. O. and Hansen, I.: 1991, 'Gastrointestinal Implica- tions in Pigs of Wheat and Oat Fractions. 2. Microbial Activity in the Gastrointestinal Tract, Br. J. Nutr. 65, 233-248.

Bach Knudsen, K. E., Jensen, B. B. and Hansen, I.: 1993a, 'Digestion of Polysaccarides and Other Major Components in the Small and Large Intestine of Pig Fed Diets Consisting of Oat Fractions Rich in fl-D-Glucan', Br. J. Nutr. 70, 537-556.

Bach Knudsen, K. E., Jensen, B. B. and Hansen, I.: 1993b, 'Oat Bran but not/3-Glucan-Enriched Oat Fraction Enhanced Butyrate Production in the Large Intestine of Pigs', J. Nutr. 123, 1235-1247.

Butine, T. J. and Leedle, J. A. Z.: 1989, 'Enumeration of Selected Anaerobic Bacterial Groups in Caecal and Colonic Contents of Growing-Finishing Pigs', Appl. Environ. Microbiol. 55, 1112- 11116.

Champ, M., Barry, J.-L., Bonnet, C., Berot, S. and Delort-Laval, J.: 1990, 'The Role of Cell Wall Polysaccharides and c~-Galactosides in the Flatus Induced by the Consumption of a Legume Seed (Lupin) in the Rat', Sci. Aliments. 10, 317-323.

Christensen, K. and Thorbek, G.: 1987, 'Methane Excretion in the Growing Pig', Br. J. Nutr. 57, 355-361.

Crutzen, P. J., Aselmann, I. and Seiler, W.: 1986, 'Methane Production by Domestic Animals, Wild Ruminants, other Herbivorous Fauna, and Humans', Tellus 33B, 271-284.

DeGraeve, K. G., Grivet, J. P., Durand, M., Baumatin, P. and Demeyer, D.: 1990, 'NMR Study of 13CO2 Incorporation into Short-Chain Fatty Acids by Pig Large-Intestinal Flora', J. Gen. Microbiol. 36, 579-582.

Gibson, G. R., Cummings, J. H., Macfarlane, G. T., Allison, A., Segal, I., Vorster, H. H. and Walker, A. R. P.: 1990, 'Alternative Pathways for Hydrogen Disposal During Fermentation in the Human Colon', Gut 31, 679-682.

Hungate, R. E.: 1966, The Rumen and its Microbes, Academic Press, London. Jensen, B. B.: 1988, 'Effect of Diet Composition and Virginiamycin on Microbial Activity in the

Digestive Tract of Pig', in: Buraczewska, L., Buraczewski, S., Pasturewska, B. and Zebrowska, T. (eds.), Digestive Physiology of the Pig, Polish Academy of Science, Jablonna, pp. 392--400.

Jensen, B. B. and Jc~rgensen, H.: 1994, 'Effect of Dietary Fibre on Microbial Activity and Microbial Gas Production in Various Regions of the Gastrointestinal Tract of Pigs', AppL Environ. Microbiol. 60, 1897-1904.

Jr H. and Jensen, B. B.: 1994, 'The Effect of Dietary Fibre on Digestibility, Microbial Activity, and Microbial Gas Production in Various Regions of the Gastrointestinal Tract of Pigs', in: Souffrant, W.-B. and Hagemeister, H. (eds.), Vlth International Symposium on Digestive Physiology in the Pig, Bad Doberan, Germany, 4-6 October, pp. 273-276.

J~rgensen, H. and Just, A.: 1988, 'Effect of Different Dietary Components on Site of Absorption Site of Disappearance of Nutrients', in: Buraczewska, L., Buraczewski, S., Pasturewska, B. and

1 12 BENT BORG JENSEN

Zebrowska, T. (eds.), Digestive Physiology of the Pig, Polish Academy of Science, Jablonna, pp. 230-239.

Lovley, D. R., Dwyter, D. E and Klug, M. J.: 1982, 'Kinetic Analysis of Competition Between Sulphate Reducers and Methanogens for Hydrogen in Sediments', Appl. Environ. Microbiol. 43, 1373-1379.

Lupton, E S. and Zeikus, J. G.: 1985, 'Physiological Basis for Sulphate Dependent H2 Competition Between Sulfidogens and Methanogens', Curr. MicrobioL 11, 7-12.

Macfarlane, G. T. and Cummings, J. H.: 1991, 'The Colonic Flora, Fermentation, and Large Bowel Digestive Function', in: Phillips, S. E, Pemberton, J. H. and Shorter, R. G. (eds.), The Large Intestine: Physiology, Pathophysiology, and Disease, Raven Press, New York, pp. 51-92.

Miller, T. L.: 1991, 'Biogenic Sources of Methane', in: Rogers, J. E. and Whitman, W. B. (eds.), Microbial Production and Consumption of Greenhouse Gases, Methane, Nitrogen Oxide and Halomethanes, American Society for Microbiology, Washington D.C., pp. 175-188.

Miller, T. L. and Wolin, M. J.: 1986, 'Methanogens in Human and Animal Intestinal Tracts', System. Appl. Microbiol. 7, 223-229.

Miller, T. L., Wolin, M. J. and Kusel, E. A.: 1986, 'Isolation and Characterization of Methanogens from Animal Faeces', System. Appl. Microbiol. 8, 234--238.

Moore, W. E. C., Moore, L. V. H., Cato, E. P., Wilkins, T. D. and Kornegay, E. T.: 1987, 'Effect of High-Fibre and High-Oil Diets on the Faecal Flora of Swine', Appl. Environ. Microbiol. 53, 1638-1644.

~rskow, E. R., Flatt, W. P. and Moe, P. W.: 1968, 'Fermentation Balance Approach to Estimate Extent of Fermentation and Efficiency of Volatile Fatty Acid Formation in Ruminants', J. Dairy Sci. 51, 1429-1435.

Prins, R. A. and Lankhorst, A.: 1977, 'Synthesis of Acetate from CO2 in the Caecum of Some Rodents', FEMS Microbiol. Lett. 1, 255-258.

Robinson, J. A., Smolenski, W. J., Ogilvie, M. L. and Peters, J. P.: 1989, 'In vitro Total-Gas, CH2, H2, Volatile Fatty Acid and Lactate Kinetics Studies on Luminal Contents from the Small Intestine, Caecum, and Colon of the Pig', Appl. Environ. Microbiol. 55, 2460-2467.

Sorlini, C., Brusa, T., Ranalli, G. and Ferrai, A.: 1988, 'Quantitative Determination of Methanogenic Bacteria in Faeces of Different Mammals', Curr. MicrobioL 17, 33-36.

Tannock, G. W.: 1992, 'Genetic Manipulation of Gut Microorganisms', in: Fuller, R. (ed.), Probiotics. The Scientific Basis, Chapman and Hall, London, pp. 181-207.

Van Soest, P. J.: 1982, Nutritional Ecology of the Ruminant, O & B Books, Inc., Corvallis, Oregon. Whitelaw, E G., Hyldgaard-Jensen, J., Reid, R. S. and Kay, M. G.: 1970, 'Volatile Fatty Acid

Production in the Rumen of Cattle given an All-Concentrate Diet', Br. J. Nutr. 24, 179-195. Zhu, J. Q., Fowler, V. R. and Fuller, M. E: 1993, 'Assessment of Fermentation in Growing Pigs Given

Unmolassed Sugar-beet Pulp: a Stoichiometic Approach', Br. J. Nutr. 69, 511-525.