UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effect of maturity on rumen degradation of tropical and temperate forage cell wall polysaccharides from… Mbugua, David M. 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_spring_mbugua_david.pdf [ 5.75MB ]
JSON: 831-1.0086137.json
JSON-LD: 831-1.0086137-ld.json
RDF/XML (Pretty): 831-1.0086137-rdf.xml
RDF/JSON: 831-1.0086137-rdf.json
Turtle: 831-1.0086137-turtle.txt
N-Triples: 831-1.0086137-rdf-ntriples.txt
Original Record: 831-1.0086137-source.json
Full Text

Full Text

EFFECT OF MATURITY ON RUMEN DEGRADATION OF TROPICAL AND TEMPERATE FORAGE CELL WALL POLYSACCHARIDES FROM LEAVES AND STEMS  by David Mwaura Mbugua B.Sc.(Agric.), University of Nairobi, 1988  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE  in THE FACULTY OF GRADUATE STUDIES Department of Animal Science We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA April, 1993 © David Mwaura Mbugua, 1993  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of A--Ai  ,  ^c  The University of British Columbia Vancouver, Canada  Date  DE-6 DE-6 (2/88)  ABSTRACT  Two main experiments were conducted in this study with the aim of better understanding the factors that influence the degradability of forages by ruminants. In the first part of the study, the degradability of leaf and stem fractions of two mature tropical forages, bana grass (Pennisetum purpureum) and silverleaf desmodium (Desmodium uncinatum) were determined by means of the In situ nylon bag technique. The effective degradabilities of both bana grass stems and leaves DM (50h)^were low (48.2% and 45.3% respectively) and did not differ (P > 0.05) for the two fractions. Desmodium stems were less degradable (P < 0.05) than the leaves (40.8% and 59.7% respectively). Neither grass leaves nor stems differed (P > 0.05) in effective degradability of NDF (49.0% vs 51.6% respectively) or its constituent polysaccharides. Bana leaves were higher (P < 0.05) in the potentially degradable DM, than the stems but did not differ (P > 0.05) in the rate at which this fraction was degraded. On the other hand, desmodium leaves were higher in the potentially degradable fraction and the rate at which it was degraded than the respective stems. NDF, cellulose and hemicellulose showed a similar trend. The second part of this study was aimed at determining whether the degree of plant maturity has any effect on the distribution of cellulose and linear and branched fractions of hemicelluloses. Two temperate forages, orchard grass ii  (Dactylis qlomerata) and tall fescue (Festuca arundinacea) harvested at different stages of growth and separated into leaves and stems were used. The cell wall content (NDF) and hemicellulose A (linear xylan) showed an increasing trend with maturity in orchard grass but not in tall fescue. Hemicellulose B (branched xylan) in the sample used in this study did not seem to change with increasing maturity. Stems (DM) of medium and late cut orchard grass were less effectively degradable (P < 0.05) than leaves (58.1% vs 62.0% and 52.6% vs 59.2% respectively). The potentially degradable fraction did not differ (P > 0.05) with maturation for the two fractions. The rate of degradation of this fraction was higher for leaves (P < 0.1). Tall fescue fractions did not differ (P > 0.05) in effective degradability and rate of degradation of the potentially degradable fraction. Cellulose and hemicellulose A and B showed a declining trend in their effective degradabilities (50h). Among these polysaccharides hemicellulose B showed a relatively higher degradability than the other polymers. Hemicellulose A seemed the least degradable. It appeared that the polysaccharides of leaves were relatively more degradable than those of stems. Tall fescue polysaccharides seemed to differ in effective degradability in both leaves and stems at the two cutting dates other than for hemicellulose A. Hemicellulose B in orchard grass was associated with relatively high rates of degradation compared to the other polymers.  Table of Contents Abstract ^ Table of Contents ^  iv  List of Tables ^  vii  List of Figures ^  ix  List of Appendix Tables ^ Acknowledgement ^  xi  1 GENERAL INTRODUCTION ^  1  2 LITERATURE REVIEW ^  5  2.1 PLANT FACTORS THAT INFLUENCE FORAGE QUALITY ^ 5 2.1.1 Introduction ^  5  2.1.2 Plant maturity ^  5  2.1.3 Plant anatomy ^  8  2.1.4 Plant biochemistry ^  10  2.1.5 Chemical composition of forages ^  11  2.1.6 Summary statement ^  22  2.2 CELL WALL CHEMISTRY, ITS RELATION TO DIGESTIBILITY ^ 23 2.2.1 Introduction ^  23  2.2.2 Cellulose ^  24  2.2.3 Hemicellulose ^  26  2.2.4 Phenolic-Hemicellulose complexes ^ 28 2.3 DIGESTION OF PLANT CELL WALLS BY RUMEN MICROBES ^ 31 2.3.1 Introduction ^  31  2.3.2 Mechanism of cellulose degradation ^ 33 2.3.3 Mechanism of hemicellulose degradation ^ 36 2.3.4 Summary statement ^  39  2.4 METHODS OF EVALUATING THE NUTRITIVE VALUE OF FORAGES ^ 41 iv  2.4.1 Introduction ^  41  2.4.2 Chemical procedures ^  41  2.4.3 Biological procedures ^  43 Enzymatic methods ^  43 In vitro methods ^  45 In situ techniques ^  47  2.4.5. Summary statement ^ 2.5 OVERALL SUMMARY ^  50 52  3 EXPERIMENTAL: DEGRADATION OF TROPICAL FORAGES ^ 54  3.1 Introduction ^  54  3.2 MATERIALS AND METHODS ^  55  3.2.1 Forages ^  55  3.2.2 In situ incubations ^  56  3.2.3 Experimental design ^  57  3.2.4 Chemical analyses ^  58  3.3 RESULTS AND DISCUSSION ^  59  3.3.1 Chemical composition ^  59  3.3.2 Degradation of DM, the cell wall and constituents ^  61  3.4 SUMMARY AND CONCLUSIONS ^  73  4 EXPERIMENTAL: DEGRADATION OF TEMPERATE FORAGES ^ 76  4.1 Introduction ^  76  4.2 Objectives ^  77  4.3 MATERIALS AND METHODS ^  78  4.3.1 Forages ^  78  4.3.2 In situ incubations ^  79  4.3.3 Experimental design ^  80  4.3.4 Chemical procedures ^  82  4.4 RESULTS AND DISCUSSION ^  87  4.4.1 Chemical composition ^  87  4.4.2 Rates and extents of DM and cell wall degradation ^  92 DM degradation ^  92 Degradation of cell wall polymers ^ 99 4.5 SUMMARY AND CONCLUSIONS ^  120  BIBLIOGRAPHY ^  122  APPENDIX ^  135  vi  List of Tables Table 3.1 Chemical composition of bana grass and desmodium  leaves and stems ^  Table 3.2 Extent of degradation of DM in bana grass and  silverleaf desmodium ^  Table 3.3 Extent of degradation of NDF in bana grass and  silverleaf desmodium ^  60 62 64  Table 3.4 Extent of degradation of hemicellulose in bana  grass and silverleaf desmodium ^ 65  Table 3.5 Extent of degradation of cellulose in bana  grass and silverleaf desmodium ^ 66  Table 3.6 Bana and desmodium DM degradation parameters ^ 68 Table 3.7 Bana and desmodium NDF degradation parameters ^ 68 Table 3.8 Bana and desmodium Cellulose degradation  parameters ^  Table 3.9 Bana and desmodium Hemicellulose degradation  parameters ^  Table 3.10 Effective degradabilities of DM at different  rumen passage rates ^  Table 3.11 Effective degradabilities of NDF at different  rumen passage rates ^  69 69 71 71  Table 3.12 Effective degradabilities of cellulose at  different rumen passage rates ^ 72  Table 3.13 Effective degradabilities of hemicellulose at  different rumen passage rates ^ 72  Table 4.1 Chemical composition of orchard grass.leaf and  stem fractions ^  Table 4.2 Chemical composition of tall fescue.leaf and  stem fractions ^  Table 4.3 Extent of DM degradation for orchard grass  leaf and stem fractions ^  Table 4.4 Extent of DM degradation for tall fescue.leaf  and stem fractions ^  91 92 95 95  Table 4.5 Orchard grass DM degradation constants ^ 97 vii  Table 4.6 Effective DM degradation values for orchard  grass ^  98  Table 4.7 Tall fescue DM degradation constants ^ 98 Table 4.8 Effective DM degradation values for tall fescue.. ^ 99 Table 4.9 Extent of degradation of cellulose in orchard  grass ^  101  Table 4.10 Orchard grass cellulose degradation constants ^ 102 Table 4.11 Extent of degradation of cellulose in tall  fescue grass ^  102  Table 4.12 Tall fescue cellulose degradation constants ^ 102 Table 4.13 Extent of degradation of hemicellulose A in  orchard grass ^  Table 4.14 Orchard grass hemicellulose A degradation  constants ^  Table 4.15 Extent of degradation of hemicellulose A in  tall fescue grass ^  Table 4.16 Tall fescue hemicellulose A degradation  constants ^  Table 4.17 Extent of degradation of hemicellulose B in  orchard grass ^  Table 4.18 Orchard grass hemicellulose B degradation  constants ^  Table 4.19 Extent of degradation of hemicellulose B in  tall fescue grass ^  Table 4.20 Tall fescue hemicellulose B degradation  constants ^  vii i  106 106 107 107 111 111 112 112  List of Figures Figure 3.1 DM degradability for desmodium leaves and stems ^ 63 Figure 4.1 Schematic presentation of the carbohydrate fractionation procedure ^  84  Figure 4.2 Orchard grass DM degradability curve ^ 94 Figure 4.3 Orchard grass hemicellulose B degradability curve ^  110  Figure 4.4 Cellulose degradability in orchard grass ^ 116 Figure 4.5 Linear xylan degradability in orchard grass ^ 116 Figure 4.6 Branched xylan degradability in orchard grass... ^ 117 Figure 4.7 Cellulose degradability in tall fescue grass ^ 117 Figure 4.8 Linear xylan degradability in tall fescue grass. ^ 118 Figure 4.9 Branched xylan degradability in tall fescue grass ^  ix  118  List of Appendix Tables Appendix Table 1 ANOVA for tropical forages DM (48h) ^ 137 Appendix Table 2 ANOVA for orchard grass DM (48h) ^ 135 Appendix Table 3 ANOVA for tall fescue grass DM (48h) ^ 135 Appendix Table 4 ANOVA for tropical forages DM (96h) ^ 135 Appendix Table 5 ANOVA for orchard grass DM (72h) ^ 136 Appendix Table 6 ANOVA for tall fescue grass DM (72h) ^ 136 Appendix Table 7 Degradation values for orchard grass cellulose ^  136  Appendix Table 8 Degradation values for orchard grass hemicellulose A ^  137  Appendix Table 9 Degradation values for orchard grass hemicellulose B ^  138  Appendix Table 10 Degradation values for tall fescue grass cellulose ^  139  Appendix Table 11 Degradation values for tall fescue grass hemicellulose.A ^  140  Appendix Table 12 Degradation values for tall fescue grass hemicellulose B ^  141  Appendix Table 13 Effective degradability for orchard grass cellulose ^  141  Appendix Table 14 Effective degradability for orchard grass hemicellulose A ^  142  Appendix Table 15 Effective degradability for orchard grass hemicellulose B ^  142  Appendix Table 16 Effective degradability for tall fescue grass cellulose ^  142  Appendix Table 17 Effective degradability for tall fescue grass hemicellulose.A ^  143  Appendix Table 18 Effective degradability for tall fescue grass hemicellulose B ^  143  x  ACKNOWLEDGEMENT  I wish to express my most sincere gratitude to my research supervisor Dr. R.M. Tait, Associate Professor, Department of Animal Science, for providing me with his support and invaluable guidance throughout the entire duration of this study. I am also most grateful to the other members of my graduate committee Dr. J.A. Shelford, Dr. L.J. Fisher and Dr. M. Pitt for their helpful comments and advice during the course of this research. Sincere thanks also go to the Kenya Government and the Canadian International Development Agency (C.I.D.A.) for awarding me the scholarship that made this study possible. My gratitude also goes to J.N. Mwangi of Simon Fraser University for his help in data analyses. Finally, I would like to dedicate this thesis to my mum, a really wonderful 'mami'. Thanks to God who made it happen.  xi  CHAPTER ONE GENERAL INTRODUCTION  Forty per cent or more of the earth's landmass is composed of rangeland which is more suited for grazing than for cultivation (Church, 1988). Products useful to humans from such areas would be greatly reduced if grazing animals were not available to utilize the vegetation to some degree. Forages provide the majority of feed for ruminant animals (Minson, 1990). Ruminants have played a major role in farming production, providing mankind with meat, milk, clothing and draft power. The tremendous diversity in these animals makes it possible for them to adapt to a wide range of climates; they are endowed with a capacity to feed on a wide range of temperate and tropical vegetation (Hobson, 1988). The unique presence of the rumen, a complex anaerobic microbial ecosystem in the foregut of these animals enables them to ferment forages of diverse nature and origin. The grazing animal exists in a highly dymamic situation in which its performance , in terms of growth , milk or wool production is determined by the quantity and quality of forage available. The principal nutritional constraint on animal productivity on a world-wide basis is the intake of digestible nutrients, particularly available energy (Reid and Jung, 1982). The primary sources of energy found in forages are the structural polysaccharides, cellulose, hemicellulose and pectins ( Dehority, 1991). Within a sward , forage quality varies not only with 1  species of plant, but also with the stage of growth, plant parts, and with climatic conditions ( Norton, 1982). Forage quality can be evaluated in terms of digestibility or fermentation of plant constituents and in terms of quantity of feed that ruminants consume (Akin, 1989). Variation in the digestibility of organic matter in forages is related to variation in the content of indigestible entities and the variable digestibilities of their potentially digestible structural carbohydrates (Ellis et al., 1988). The extended incubation of forage and cereal straws in the rumen leaves a residual fraction of cell wall polysaccharides highly resistant to further microbial degradation (Gordon et al., 1983). Variation in the digestibility of structural carbohydrates is due to both their intrinsic properties and to structural and chemical properties of the plant tissues formed by these structural carbohydrates and other chemical entities of cell wall contents. Factors limiting microbial degradation of plant cell walls have previously been attributed to features such as physical encrustation with lignin and cellulose structure ( Raymond, 1969). However Morrison (1973) suggested that lignin is more closely associated with hemicelluloses than with cellulose. Harkin (1973) suggested a possible existence of covalent linkages between lignin and the cell wall polysaccharides. Brice and Morrison (1982) reported that hemicellulose fine structure, particularly its xylose to arabinose ratio, was significant in influencing how this polysaccharide was 2  ^  degraded in the rumen. This ratio is known to increase with plant maturation. Hemicelluloses of pasture plants consist of a mixture of the following three major polysaccharide types: (a)linear A (hemicellulose A), ^a^water-insoluble heteroxylan containing uronic acid but only small amounts of arabinose; (b)  a more soluble hetero-xylan containing much more  arabinose and less uronic acid than linear A; (c)  a water soluble, highly branched polymer which in  addition to pentoses, is rich in galactose and uronic acid (b+c) = hemicellulose B (Gaillard et al., 1965). In their pure forms these polysaccharides are highly degraded by mixed cultures of rumen bacteria (Bailey and Gaillard, 1965). Factors that limit the utilization by ruminants of cellulose and hemicellulose need to be better understood. Factors such as cellulose crystallinity and linearity and branching of hemicelluloses have been suggested as potential limitations to cell wall digestibility by ruminants (Morrison, 1979). These factors might reflect on voluntary intake of the ruminant animal and hence productivity. In the work presented here, two mature tropical forages, bana grass (Pennisetum purpureum) and the legume silverleaf desmodium (Desmodium uncinatum) were separated into leaf and stem fractions and their cell wall degradation 3  characteristics studied. Their cellulose and hemicellulose components were also studied for these characteristics. The results were interpreted in relation to concentrations of these polysaccharides and lignin and suggestions were made to try to overcome some of the problems related to their indigestibility. To further understand the factors that may lead to low degradability of these polysaccharides in the rumen, a fractionation scheme for cell wall components was adopted which facilitated the separation of linear and more branched fractions of hemicellulose. This work was done using two temperate grasses; orchard grass (Dactylis qlomerata) and tall fescue (Festuca arundinacea) at different stages of maturity. They were also separated into leaf and stem fractions. Digestion kinetic models were developed for the various cell wall polymers. An effort has also been made to relate the observed results to possible effects on parameters that affect production, such as voluntary intake. The rapid decline in the digestibility of orchard grass within short periods of growth (Shelford and Fisher, 1988) warranted a closer study of the possible factors leading to this.  4  CHAPTER TWO LITERATURE REVIEW  2.1^PLANT FACTORS THAT INFLUENCE FORAGE QUALITY 2.1.1 Introduction  Forages are the major source of feed for ruminants in many parts of the world. The quality of a forage has its ultimate expression in terms of animal performance when it is fed as the sole source of digestible nutrients. Forages, however, are diverse in their characteristics, and this diversity results in variations in quality as an animal feed. Pasture plants differ in species, stage of maturity, morphology, anatomy, biochemistry, and in chemical components in cell contents and cell walls (Norton, 1982). These factors influence voluntary intake, digestibility and efficiency of utilization of absorbed nutrients. In the following sections a brief discussion is given covering differences in stage of maturity, anatomy, and biochemistry in relation to their effects on voluntary intake and digestibility. The role of chemical composition in this respect will be given a more thorough discussion.  2.1.2 Plant maturity  The factors which influence forage growth are dynamic, and change significantly with time. Consequently the chemical and physical characteristics of forages are influenced by the stage and the rate of growth of the crop, as well as by the 5  previous management of the sward. Mowat et al. (1965) studied the effect of maturation on crude protein (CP) content and in vitro dry matter digestibility (IVDMD) of leaves and stems of various temperate forages. They reported that the CP content of both leaf and stem portions decreased with increasing maturity. Norton (1982) made a similar observation for both temperate and tropical forages. In the early stages of growth, the cell contents (cell nucleus and the cytoplasm) may account for at least two-thirds of forage dry matter, with protein being the major contributor (Gill et al., 1989). As the plant matures, the proportion of the cell walls and its constituent fractions increases and the cell content decreases (Van Soest, 1982; Jung and Vogel, 1992). This situation leads to a decline in digestibility of forages. With maturation the protein content of legumes declines only slowly compared to grasses (Norton, 1982). Older leaves show the greatest decrease. The higher protein content in legumes than grasses and its maintenance with maturity may be associated with the continuous supply of nitrogen available from rhizobial fixation. Variations between legume species in protein content probably reflects on the effectiveness of rhizobial nitrogen fixation under different environmental conditions (Norton, 1982). Maturation has been reported to lead to decreased IVDMD of both leaf and stem fractions in grasses (Mowat et al., 1965). These workers observed that at early growth stages the digestibility of stems of grasses was higher than that of 6  leaves. In orchard grass, even at head emergence, stems were slightly more degradable than leaves. Shelford and Fisher, (1988) have also reported on the effect of advance in maturity in grasses on DM digestibility. These researchers reported a decline of 11 percentage points in the digestibility of orchard grass harvested within a space of 10 days. Gill et al. (1989) reported that cellulose, hemicellulose and lignin all increase with advancing maturity but indicated that it is the increased lignin, in particular its spatial distribution, which has the most significant negative effect on the rate and extent of digestion of a forage. Leaves of tropical grass species tend to be less digestible than those of temperate grasses (Hacker and Minson, 1981). These workers reported that in maize and tropical grasses, high temperatures during growth result in decreased leaf and stem digestibility. Increasing day temperature was reported to decrease digestibility in leaves of Cenchrus  ciliaris and Pennisetum clandestinum; however high night temperatures increased the digestibility of leaves of P.  clandestinum (Hacker and Minson, 1981). Both tropical and temperate grasses decline in dry matter digestibility at a mean rate of about 0.6% units per degree centigrade (Minson and Wilson, 1980).  7  2.1.3 Plant anatomy Leaf blades The amount and arrangement of tissues in forages can influence their quality. Tissues in the leaf blades of all grasses include vascular tissue (divided into phloem and xylem cells), parenchyma bundle sheath(s) surrounding the vascular tissue, sclerenchyma patches connecting the vascular bundles to the epidermises, single-layered abaxial and adaxial epidermal cells covered by a protective cuticle, and mesophyll cells between the vascular bundles and epidermal layers (Metcalfe, 1960). The first stable products of photosynthesis in tropical grasses are four-carbon compounds, while those of temperate grasses and dicotyledons are three-carbon compounds. Hence these plants are referred to as C4 and C3 respectively. Generally, grasses possessing the C3 pathway for photosynthesis have a higher ratio of mesophyll to vascular tissue than grasses that possess the C4 photosynthetic pathyway (characterized by a specialized leaf anatomy - the Kranz anatomy)(Akin, 1986). The mesophyll, devoid of lignin, is the tissue most easily degraded by microorganisms in the rumen. In tropical grasses intercellular air spaces represent only 3-12% of leaf volume compared with 10-35% in temperate species (Norton 1982). The lower surface area to weight ratio for both mesophyll and bundle sheath tissues restricts accessability of plant cells to microbial digestion in the 8  rumen (Hanna, et al 1973), thereby decreasing the rate of digestion of the bundle sheath and enclosed vascular tissue (Akin and Burdick, 1975). In C3 plants, the parenchyma bundle sheath is a distinct structure but its cell wall appears to degrade as rapidly as the wall of the mesophyll (Akin, 1986). This structure in C4 plants appears to be a major factor that often occupies a prominent part in the undegradable residue of C4 plants. Stems  Forage stems contain a large proportion of lignified tissues which resist microbial degradation. Stem anatomy of C4 and C3 plants seems to be similar, with the epidermis, sclerenchyma ring and vascular xylem occupying 28-34% of the cross-sectional area (Akin, 1989). These tissues are highly lignified and are totally indigestible. Parenchyma tissues vary in lignin histology and digestibility depending on age and species. In general, grasses tend to show the following trend in tissue degradability: mesophyl, phloem > epidermis, parenchyma bundle sheath > sclerenchyma > lignified tissues (Akin, 1982). Legume stems can be highly digestible in young plants, but as they mature they become lower in digestibility and contribute primarily to the indigestibility of the plant. In legumes it appears that parenchyma tissues do not lignify with plant maturity as in grasses (Akin, 1989). As such, this tissue is totally degraded in legumes.^Differential 9  lignification of legume tissues relative to the grasses is a factor responsible for differential breakdown, which may account for greater intake of legumes over grasses for a given stage of maturity (Demarquilly and Jarrige, 1974).  2.1.4 Plant biochemistry  The main factors regulating the intake of pasture by animals are the density of the sward, the level of fiber and its physical composition, provided there is adequate protein minerals and vitamins. Feed intake is severely depressed when crude protein content of pasture falls below 6-8% (Minson, 1982). Tissue nitrogen content of C4 plants is usually lower than that of C3, mainly because of the higher efficiency of nitrogen use associated with growth of these tropical plants (Norton, 1982). Brown (1978) indicated that the higher growth rate, nitrogen use efficiency and lower tissue nitrogen content of C4 plants than C3 plants is related to differences in the pathways of carbon fixation. The major difference in the biochemical pathways for C3 and C4 plants that might influence quality is that in C4 plants, the first photosynthetic products are oxaloacetic acid, malic acid and aspartic acid whereas in C3 plants phosphoglyceric acid is produced (Norton, 1982). In C3 plants, fraction 1 protein (ribulose-1,5biphosphate) represents about 50% of the soluble protein found in the mesophyll cells. The low activity of this enzyme is reported to be the main factor limiting carbon fixation during 10  photosynthesis. On the other hand the concentration of this enzyme is low (about 20%) in C4 plants and restricted to the bundle sheath cells (Bjorkmann et al., 1976). Phosphoenolpyruvic acid carboxylase is found in relatively low concentrations in C4 mesophyll cells but its higher activity results in high rates of CO2 fixation per unit of cellular protein when compared to C3 plants. Norton (1982) concluded that the low protein content often found in tropical grasses, even when fertilized with nitrogen, is an inherent characteristic of C4 plants, which is closely related to their survival under conditions of low soil fertility. Colman and Lazenby (1970) have shown that at higher temperatures (2335°C), the efficiency of nitrogen use by C4 plants is higher than at low temperatures (13-24°C), but tissue nitrogen content is also decreased by high temperatures.  2.1.5 Chemical composition of forages  The availability of nutrients in a feed  is essentially  determined by the chemical composition of the feed: first with respect to the concentrations of available  and unavailable  components and secondly, through organic  structures and  inhibitors that may limit the availability of the components with which they are associated. As pasture plants mature there is usually an increase in the proportion of fiber and a reduction in the protein and non-structural carbohydrates of the cell contents. In a study of 82 samples of grass and legumes, Van Soest (1965) reported 11  significant correlations between intake and neutral detergent fiber (r = - 0.65), protein (r = 0.54), and acid detergent fiber (r = -0.53) but not with lignin (r = -0.13). The voluntary intake of forage has been positively related to the digestibility of dry matter and energy (Minson, 1982). This relationship is associated with the main factors that control voluntary intake, namely the proportion of indigestible residue in the feed, the transit time of this residue through the rumen and the size of the rumen. The chemical composition of any given forage will vary depending on its species, stage of maturity, soil fertility, plant part and climatic conditions. The following sections focus on protein and carbohydrate sources in forages and how they relate to voluntary intake and digestibility. Herbage proteins The crude protein fraction, i.e. total nitrogen (N), of forages consists of proteins (true), amino acids, amides, ureides and nitrates (NPN). True protein constitutes about 6080% of total nitrogen (Minson, 1990). The protein in fresh forages may be classified into three main categories: fraction 1 - a single protein, i.e., ribulose-1,5-biphosphate (RUDP) carboxylase; fraction 2 - a mixture of cytoplasmic and other chloroplastic proteins; and fraction 3 - a mixture including chloroplastic, nuclear, and mitochondrial membrane proteins (Mangan, 1982). About 50% of the cell protein is RUDP carboxylase or fraction 1 protein. Fractions 1 and 2 are 12  soluble and are likely to be fermented in the rumen, although rates may vary for individual proteins (Nugent and Mangan, 1981). The extent of dietary protein degradation in the rumen has been correlated with protein solubility (Buttery and Lewis, 1982). Forage protein solubility may be influenced by the presence of tannins (Norton, 1982) which also reduce the susceptibility of the protein to microbial fermentation. Temperate grasses generally contain more crude protein (Cl') than tropical grasses, with mean concentration of 12.9 and 10.0% DM, respectively (Minson, 1990). Temperate and tropical legumes have the same leaf anatomy, C3 pathway of photosynthesis,^and similar levels of crude protein; concentration of Cl' were 17.5 and 16.6% DM,^respectively (Minson, 1990). The low protein content of tropical grasses poses a major limit to intensive forms of animal production, and the inclusion of tropical legumes in these pastures improves animal production by increasing protein availability for grazing animals (Norton, 1982). Immature forages contain high levels of crude protein of which more than 80% is true protein, found mainly in the cytoplasm. As immature forages contain low levels of cell wall constituents and are readily fermentable, they form a rich source of both protein and energy for rumen microbes (Hogan, 1982). As the plant matures there is a rapid decline in protein content, especially in grasses as the proportion of leaf decreases, which poses a major criterion in the management of forage quality for grazing animals. Protein content decreases in both stems and 13  leaves as the plant ages, but the decline is slower with leaves than with stems. Minson (1990)^reported that there is a rapid decline in voluntary intake when crude protein in a feed material falls below about 7.0% DM. When low quality roughages are not limited in quantity, protein is the most beneficial supplement (DelCurto, 1991). However, the exact mechanisms involved with the stimulation of feed intake are not clearly defined, and in all likelihood are a combination of several factors. The depressing effect on intake of a deficiency of protein appears to be caused by factors other than rumen distension, since grinding and pelleting protein deficient herbage has little effect on intake (Minson, 1967). The depression of intake appears to be caused by a deficiency of circulating amino acids,^since intake of a protein deficient diet can be increased if casein (but not urea) ^is infused into the duodenum (Egan, 1965). Herbage carbohydrates  Carbohydrates are the main repository of photosynthetic energy in plants and comprise roughly 50-80% of the dry matter of forages and cereals. The nutritive characteristics of these carbohydrates for animal feeding are variable, depending upon sugar components and linkages (Van Soest, 1982). Nutrient availability depends on the capability of microbial enzymes to cleave glycosidic bonds in plant carbohydrates and between the carbohydrates and other substances. Most polysaccharides 14  entering the rumen can be considered as belonging to one of two general types: plant non-structural carbohydrates such as sugars, starch and fructosans, and the structural polysaccharides which compose the greater part of all plant cell walls and which are considered to form the fibrous component of animal feedstuffs. Non-structural carbohydrates are readily available sources of energy to ruminant animals (Smith, 1973). The structural polysaccharides have a skeletal function in the living plant and are, by their very nature and organisation within the cell wall, far more resistant to microbial attack. However , it is the ability to utilise such materials as an energy source that provides ruminants with their particular ecological niche, and the ability of rumen micro-organisms to degrade plant polysaccharides efficiently is of paramount importance to the survival of the mature animal. Non-structural carbohydrates  In this section a brief discussion on the major nonstructural polysaccharides in relation to their distribution and degradability will be given. Suffice to mention that the monosaccharides, glucose and fructose, and the disaccharide sucrose, are the predominant sugars found in plant tissues and are mainly found in the cell contents. They are rapidly fermented in the rumen to yield volatile fatty acids (VFA), the major energy sources for the ruminant animal.  15 Herbage storage polysaccharides Starches and fructosans are the most abundant nonstructural polysaccharides. Starches are glucose polymers composed of D-glucopyranose units joined principally through al-4 glycosidic links (Smith, 1973). Starch is a composite of structurally distinct polymers: amylose and amylopectin (Chesson and Forsberg, 1988). Amylose consists of an essentially linear chain of al-4 linked sugar residues. Amyloses are degraded by randomly acting a-amylases which bring about a rapid reduction in viscosity and the release of low molecular weight oligosaccharides (maltodextrins) and by 8-amylases which remove successive maltose units from the nonreducing end of the chain (Chesson and Forsberg, 1988; Van Soest, 1982). On the other hand amylopectin is a highly branched molecule with al-6 links at the branch points and al4 links within the chains of each branch. About 50% of amylopectin can be degraded to maltose by the action of Bamylases, leaving a residue known as 'B-limit dextrin' which is protected from further attack by the 6-linked branch points. Complete hydrolysis of amylopectin requires the action of enzymes capable of hydrolysing the al-6 linkages, notably the glucoamylases and 'debranching enzymes'. Polymers of fructose occur in two forms, the inulins which serve as reserve carbohydrates in tubers in which the fructofuranosyl units are linked B2-1 and the levans in which the units are linked B2-6 (Smith, 1973). The latter are commonly found in temperate grasses, particularly in stems, 16  where they can reach levels of 1-5% of total dry matter (Smith,1973). Fructosans are rapidly and completely degraded in the rumen by organisms which include the holotrich (Williams, 1986) and entodiniomorph protozoa (Coleman, 1980). Many bacteria are able to utilise inulins (Hungate, 1966). The relationship between cellular components and feed intake depends on the contribution made by a specific constituent to the structure or volume of the plant. The rate of organic matter removal from the rumen plays a major role in limiting feed consumption. Organic matter can be removed from the rumen by digestion and by passage to the lower gut (Black et al., 1982). Soluble carbohydrates can be degraded by rumen microbes 30 times faster than storage carbohydrates which, in turn, are degraded 5 times faster than structural carbohydrates (Maeng and Baldwin, 1976). Thus, large differences in the proportions of these broad groups of plant constituents can markedly affect the rate of organic matter removal from the rumen and hence voluntary intake. Ulyatt (1981) suggested that the lower ratio of structural to readily fermentable carbohydrates in clover than in ryegrass was the major reason for the reduced apparent mean retention time of organic matter in the rumen and hence the greater voluntary feed consumption of sheep eating clover. Since structural carbohydrates are degraded at the slowest rate, they contribute most to dietary organic matter accumulation in the rumen (Black et al., 1982). This aspect is the subject of the next topic. 17 Herbage structural polysaccharides Cell walls form 30-80% of plant dry matter and vary as a source of energy. The cell wall of plants is a complex matrix composed of polysaccharides, proteins, phenolics (including lignin and tannins), water and minerals (Hatfield, 1989). Bailey (1973) has given a comprehensive review of the chemistry and biosynthesis of these polysaccharides. Structural carbohydrates in higher plants are represented by the polysaccharides cellulose, hemicelluloses and the pectic substances. The pectic substances include those substances that are more or less readily extracted from plant tissues with hot water, cold dilute acid, hot ammonium oxalate or hot EDTA solution (Bailey, 1973). The hemicellulosic fraction is defined as the group of polysaccharides extracted under alkali conditions after pectic material and lignin are removed from the matrix. The remaining material after the sequential extraction of pectins, lignin and hemicelluloses is defined as cellulose, or more precisely as a-cellulose, a pure cellulose B-glucan (Bailey, 1973). The nutritional value of a forage given to an animal is a function of the amount of forage eaten and the nutrients made available during its digestion. Provided the energy demand of the animal is not satisfied, differences in intake of plant material may result from either differences in the rate of removal of organic matter from the rumen and (or) differences in the amount of digesta which can accumulate in the rumen (Black et al., 1982). The importance of plant cell wall as the 18  primary restrictive determinant of intake has been demonstrated by Mertens (1973) as cited by Van Soest (1984). Using 187 forages, he showed that neutral detergent fiber (NDF), the fraction that is most representative of the total plant cell wall, was negatively correlated (r= -0.75) with voluntary intake. Functional fiber is needed to maximize chewing, saliva production and rumen function and to minimize the incidence of digestive disorders. On the other hand, productivity of cattle fed high forage diets is limited by the amount of forage that can be consumed (Beauchemin, 1991). In this case, forage intake and utilization are maximized by enhancing breakdown, digestion and throughput of fiber alleviating gut fill. Several concepts have been advanced regarding the action of the fill mechanism. Mertens (1973) point to the limiting effect of rumen fill. Dermaquilly et al. (1965) have suggested that the volume of indigestible matter is limiting through flow in the lower gut. Campling and Balch (1961) showed that removal of swallowed hay from the rumen had an immediate effect on the time the cow stopped eating. Cows could be encouraged to eat for very much longer than normal by removing the swallowed hay. Further Campling et al. (1961) as cited by Campling (1969) observed that cows fed hay or dried grass stopped eating when the reticulorumen contained similar amounts of dry matter and that the quantity of each roughage eaten was directly related to its rate of disapperance from the reticulorumen. Van Soest (1982) observed that the capacity 19  of the intestines to transport digesta was not a limitation to intake because daily fecal output increases with increasing voluntary roughage intake. Since the amount of cell wall is more highly associated with intake than any of its components or rate of fermentation parameters (Van Soest , 1975), strong support is given to a structural volume theory relating voluntary intake and forage quality. This involves two parts: the structural volume theory (relates to cell wall content) and rate of passage. The intake of a forage is usually increased when it is ground and pelleted (Minson, 1982), a result of decreased effective volume (Van Soest, 1975). However, the process of digestion does not decrease volume unless a reduction in particle size occurs. Van Soest (1975) considered the aspect of volume as relating to cell wall content which therefore influences the rumen fill effect. To alleviate the fill effect two mechanisms come into play, namely, rumination with reduction in particle size and passage rate. When feed enters the rumen, undigested feed residues must be reduced in size before they can leave the reticulorumen (Campling, 1969). Particle size reduction of feeds is mainly the result of chewing during eating and rumination. Microbial degradation and abrasion during movement within the rumen contribute to a lesser extent. Chewing of feeds during eating and rumination is a response to the physical property of fiber. Large feed particles are chewed more and reduced to a greater extent than smaller ones during eating (Beauchemin, 20  1991). For forages, between 15 and 55% of the long particles (by weight) are broken down during eating to a size that can be passed from the rumen. Rumination is usually the principal means by which particle size of feed is reduced, but not always. This is because the net contribution of rumination to breakdown of ingested feed particles depends on the efficiency of breakdown during feeding (Beauchemin, 1991). Once the large particles are swallowed after chewing, rumination very effectively reduces them. Fine particles pass from the rumen more rapidly than large ones (Campling, 1969). Thus, the higher intake of ground and pelleted forages could be interpreted as having saved the ruminant the trouble of ruminating forage particles to a size that will pass. If this is the case, the cell wall properties that determine the rate of breakdown and the amount of work required will affect intake and digestibility (Van Soest, 1975).  21  2.1.6 SUMMARY STATEMENT The foregoing discussion has stressed the role of indigestible components of forages on voluntary intake. The fraction of the fiber that is indigestible has been estimated to constitute up to 1/3 of the total fiber in grasses and about 1/2 in legumes (Fahey and Garleb, 1991). Variable intakes have been reported for diets with comparable amounts of cell walls (NDF) and this has been explained to be as a result of possible differences in rate and extent of cell wall digestion (Varga and Hoover, 1983). Since the overall rate and extent of cell wall digestion will depend on the rates and extents of digestion of individual polymers constituting the cell wall, a study of the latter parameters would cast some light on the understanding of overall digestion of the cell walls. This could also possibly explain why diets containing comparable levels of cell walls may lead to differences in voluntary intake. The present study was proposed as an effort to understand how different cell wall polymers are degraded within the rumen and to try and relate the results to variations in intake.  22  2.2 CELL WALL CHEMISTRY, AND ITS RELATION TO DIGESTIBILITY 2.2.1 Introduction  The degradability of plant cell walls is likely to be limited not only by the complex interrelationships between polysaccharide structures of the wall, but also by noncarbohydrate materials such as lignin (Ford and Elliot, 1987). Low molecular weight phenolic compounds associated with the cell wall have also been implicated as inhibitors of forage digestion in the rumen (Akin and Chesson, 1989). In this section the interrelationships amongst the wall compounds and how they relate to forage digestion in the ruminant animal are discussed. The pectins, whose degradability is known to be high, will not be discussed. The plant cell wall is the principal structural component of the plant cell and surrounds the protoplast and exterior of the cell membrane, the plasmalemma. Depending on cell type and the stage of maturity, the plant cell wall can be composed of up to three layers, the middle lamella, the primary cell wail and the secondary cell wall (Selvendran and O'Neill, 1987). The first cell wall structure to be synthesized by the cell is the middle lamella which is composed of pectic substances, primarily the Ca++ salts of galacturonate and rhamnogalacturonate. The next cell wall layer to be formed is the primary cell wall, which is usually found in young undifferentiated cells that are still growing (Selvendran and O'Neill, 1987). This layer consists mainly of polysaccharides, 23  primarily cellulose, hemicellulose and pectins. Besides these polysaccharides, this layer also contains approximately 10% protein, primarily extensin, with 30% of the amino acid residues being hydroxyproline (Fahey and Garleb, 1991). The secondary cell wall layer is formed once the cell has reached maturity. It is a rigid and thickened layer that has a distinct shape and property. The thickening is as a result of the laying down of the polyphenolic compound lignin, which begins in the primary cell wall region and extends onward to the middle lamella (Fahey and Garleb, 1991).  2.2.2 Cellulose  Cellulose is both a component of the primary and secondary wall material. It is composed of B-linked glucose residues which form chains of up to 14,000 residues in length (McNeil et al., 1984). The unique properties of cellulose are conferred by its secondary, rather than its primary, structure. The linear chains of B-1,4-linked glucose units aggregate through intra- and extramolecular hydrogen bonding to form microscopically visible fibrils (Delmer, 1987). The degree of order found within and between fibrils varies from regions in which the glucan chains are held firmly in parallel, and where X-ray diffraction studies indicate a high degree of crystallinity, to regions in which this order is somewhat reduced (amorphous regions)(Chesson and Forsberg, 1988). Other polysaccharides and lignin are closely associated with cellulose, but there is no evidence to suggest that 24  cellulose has covalent bonds with them (Hatfield, 1989). Cellulose is present in all plants and is probably the structural polysaccharide which contributes most to the rigidity and strength of plant structures (Bailey, 1973). The degradation of cellulose has been extensively studied in a number of microorganisms. One of the restraints to degradation either by enzymatic or chemical systems is the extensive hydrogen bonding that occurs in microfibrils to give cellulose a crystalline structure. Other factors postulated as potentially affecting cellulose digestion include encrustation by lignin and hemicellulose and a limited enzyme accessible space for cellulose hydrolysis (Fahey and Garleb, 1991). Kerley et al. (1988) defined enzyme accessible space by the size, shape and surface properties of microscopic and submicroscopic capillaries within fiber in relation to the size, shape diffusibility of microbial cellulase enzymes. Cellulose crystallinity is perceived to be a factor that regulates the rate as opposed to the extent of cellulose digestion. Van Soest (1982) reported that alfalfa cellulose prepared by chlorite delignification and alkaline extraction of hemicellulose had a slower digestion and greater overall digestibility than cellulose in intact alfalfa cell wall. He argued that the removal of lignin and hemicellulose may have allowed cellulose chains to become more closely aligned and thus resulted in greater crystallinity. The same worker (1975) indicated that lignin is more highly correlated to the degree of cellulose digestion than it is to rate of digestion. 25  2.2.2 Hemicellulose The constituents of the cell wall fall into three groups: the fibrillar polysacharides (mainly celluloses discussed in the previous section), the matrix polysaccharides (Bailey, 1973) and the encrusting substances (Selvendran and DuPont, 1984). The matrix polysaccharides are made up of linearly oriented polymers which are present at all stages of the development of the wall and also of highly branched polysaccharides that are deposited at particular stages of growth (Selvendran and DuPont, 1984). There are two major fractions in the matrix polysaccharides: (1) the pectic substances; (2) the hemicelluloses, which are represented by the polysaccharide soluble in alkali. Hemicellulose is the most complex of the plant polysaccharides. Hatfield (1989) has categorised hemicelluloses into four distinct components namely: the xylans, 8-glucans, xyloglucans and the mannans. The xyl'an fraction, assembled from 8-1,4-linked xylose residues, is the main hemicellulosic component found in grasses. The xylan is substituted with acetyl, arabinosyl and glucuronyl residues (Cheng et al., 1991; Hatfield, 1989). In ryegrass, wheat and barley straw, and white clover, 50% or more of the xylose residues may be substituted in the 0-2 and 0-3 positions, with 50 to 70% of these sustituents being acetyl groups (Cheng et al., 1991). The extent of substitution on the xylan backbone by arabinose varies from one in every two to eight xylose 26  residues (Hatfield, 1989). A proportion of the arabinose residues are further esterified with feruloyl and coumaroyl residues. In barley straw for example, 1 in every 31 arabinose residues was esterified with p-coumaric acid, and 1 in every 15 with ferulic acid (Mueller-Harvey et al., 1986). The other component of hemicelluloses, the B-glucans represent a unique molecule that seems to be restricted to immature tissues of grasses (Morrison, 1979) and endosperm cell walls of cereals (Hatfield, 1989). About 70% of the molecule is made up of glucose residues linked B-(1,4) and the remaining 30% linked 8-(1,3). Xyloglucans on the other hand are composed of a backbone of B-(1,4) linked glucopyranose units. This backbone is substituted with xylopyranose units by an a-(1,6) linkage. Xyloglucans are most abundant in primary cell walls of dicotyledons and may possess additional sugars mainly galactose and fucose (Hatfield, 1989). The mannans are frequently found as components of many seeds (galactomannans), but are also found as cell-wall components of many plants. The glucomannans and galactoglucomannans are basically linear molecules composed of B-(1,4) linked glucopyranose and mannopyranose residues. A series of hemicellulases exist in the rumen which can be isolated from cell free rumen fluid (Brice and Morrison, 1982). These workers studied the degradability of lignified and delignified hemicelluloses isolated from perennial ryegrass harvested at different maturity stages. They found that as the ratio of xylose:arabinose increased with plant 27  maturity (Morrison, 1980), there was an increase in hemicellulose degradability. The reason for this increase was that isolated hemicellulases cleave between two unsubstituted xylose residues. As a plant matures, however, lignin content increases and overrides the favorable effects of the increased xylose:arabinose ratio. However, it is also suggested that linear xylan may have a high affinity for the cellulose fibrils (McNeil et al. 1975). An increase in hydrogen bonding between xylan and cellulose microfibrils could lower the fermentability of both fiber fractions (Fahey and Garleb, 1991). It has been concluded that with young grass, the proportion of side chains is the major controlling factor of hemicellulose digestion while in older tissue the lignin content is the dominant factor (Morrison 1979).  2.2.4 Phenolic-hemicellulose complexes Lignin is generally considered to be the overiding factor in the limitation of cell wall degradation by microorganisms and their cell-free enzymes (Wallace et al., 1991). In the intact cell wall, lignin has been shown to be covalently bound to the hemicellulose (Morrison, 1979) and to exist, in part at least, as a lignin-carbohydrate complex (LCC), (Scalbert et al., 1985). In an attempt to understand the nature of LCCs, Morrison (1973) isolated an LCC from ryegrass (Lolium multiflorum). Further to this he investigated the structure of this molecule, isolated from (Lolium perenne), and reported that 28  there were at least three types of bonds between lignin and carbohydrate (Morrison, 1974). One type of bond could be cleaved by borohydride reduction, another cleaved by alkali and a third type was resistant to alkali. Saponification of plant fibers with alkali has shown that p-coumaric and ferulic acids and dimeric phenolic acids are bound to graminaceous cell walls (Hartley and Ford 1989). Treatment of bermudagrass (Cynodon dactylon) leaf blades for 24 hours with 0.1M sodium hydroxide resulted in a release of 86% of the ferulic acid, 65% of the dimers and 50% of the p-coumaric acid (Akin et al., 1992). Digestibility was increased from 6.5% in the untreated control to 56.6%. Lignin level was reduced after the alkali treatment. A higher level of alkali (14 NaOH) resulted in an increase in digestibility to 72%, and increased biodegradation of mesophyll and phloem tissues (Akin et al., 1992). Lignin-carbohydrate complexes have been isolated from the rumen (Gaillard and Richards, 1975; Neilson and Richards, 1982; Conchie et al., 1988). These workers postulated that these complexes are likely to represent phenolic rich fragments of the wall from which most of the sugar has been removed by the action of rumen microflora. Further they thought that any remaining sugar was likely to provide evidence of the nature of the LCCs in the intact wall. Gaillard and Richards (1975), have reported that the LCC from the rumen are soluble at pH 7 or higher but precipitate at pH 3. Neilson and Richards (1978) found that the soluble LCC produced in the rumen were not significantly affected by 29  further digestion in the rumen after passing into solution. Since this complex precipitated in the low pH of the abomasum and eventually appeared in the feces, they concluded that the amount of the complex digested was unlikely to be significant after it passed from the rumen. The complexes isolated from both the ovine and bovine rumen, tended to have either a high (> 100kDa) or low (< 4kDa) molecular weight (MW), the higher MW containing a higher carbohydrate concentration (Wallace et al., 1991). Reducing sugars (principally xylose and arabinose) were found in the low MW fraction, thus suggesting ether linkages to the phenolics (Wallace et al., 1991).  30  2.3 DIGESTION OF PLANT CELL WALLS BY RUMEN MICROBES 2.3.1 Introduction  The fundamentally important process of plant cell wall digestion by ruminants is dependent on colonization and digestion within the very complex microbial ecosystem of the rumen. Bacteria, fungi and protozoa colonize practically all plant materials that enter the rumen, with the exception of intact, outer plant surfaces which are reportedly not colonized by bacteria (Chesson and Forsberg, 1988). The rumen contains microbial species adapted to adhere to the specific incoming plant components in the physical condition they attain through ruminant feeding behavior. Adhesion is followed by successive microbial colonization within the adherent population, until effective digestive consortia are formed and the substrate is digested to release nutrients at maximal rate (Cheng et al., 1991). Adhesion is a prerequisite for the initiation of digestion of plant materials because this process strategically positions the microbial cell in contact with the substrate (Cheng et al., 1991). The major route of invasion of plant tissues by microbes appears to be via broken epidermal lesions (Chesson and Forsberg, 1988). Colonization by entry through stomata is comparatively unimportant with stem fragments of legumes and grasses, but it can be of greater importance for colonization of leaves (Cheng et al., 1983/84). The major species of bacteria that attach are the cellulolytic bacteria 31  Ruminococcus albus,  flavefaciens  R.  and  Fibrobacter  succinogens (formerly Bacteroides succinogens Montgomery et al., 1988) which are commonly found on cut edges of cell walls and damaged areas of cell surfaces (Akin, 1986). The Ruminococcus species appear to be loosely associated with cell walls, while F. succinogens exhibits a tight adhesion, frequently conforming to the surface of the material being digested (Chesson and Forsberg, 1988). The mechanism of attachment of different rumen microorganisms to cellulose and indeed other complex fibrous materials may involve specific binding by: cell-surface associated enzymes; adhesins (molecules on the microbial cell surface that bind to receptors on the plant material; or, possibly, non-specific ionic interaction (Forsberg, 1986). Rumen fungi and protozoa colonize plant fragments and degrade them to differing degrees (Akin, 1986). The cellulolytic fungus Neocallismatix frontalis has been shown to produce an active cellulolytic complex that carries out cellfree cellulose digestion (Wood et al., 1986). This is the only cell-free cellulolytic complex characterised from a rumen organism to date (Cheng, et al., 1991). Even though the mode by which these microorganisms attach themselves to plant materials is poorly understood, it has been found that anaerobic fungal zoospores have a preference for stomata and broken areas of plant particles from which soluble sugars diffuse and to which zoospores demonstrate a chemotaxic response (Chesson and Forsberg, 1988). Among the studied rumen 32  ciliate protozoa, EDidinium ecaudatum has been found to produce cellulase of high activity (Chesson and Forsberg, 1988).  2.3.2 Mechanism of cellulose degradation The rate of microbial digestion of feeds depends on three general factors: (1) plant structures that regulate bacterial access to nutrients; (2) microbial factors that control adhesion and the development of microbial consortia; and (3) complexes of oriented hydrolytic enzymes of the adherent microorganisms (Cheng et al., 1991). Isolated plant structural carbohydrates are readily digested by rumen microorganisms; however, their availability in the intact plant can be limited and varies with both plant species and maturity (Dehority, 1991). Rumen microbes responsible for cellulose digestion, produce cellulase enzymes in an active form. The cellulase system usually consists of three major types of enzymes which function synergistically in the hydrolysis of crystalline cellulose (Chesson and Forsberg, 1988). These are: endo-1,4-Bglucanase, cellobiohydrolase and B-glucosidase. The endoglucanase attacks cellulose in a random fashion that results in a rapid decrease in chain length and gives rise to cello-oligosaccharides. The cellobiohydrolase degrades cellulose by splitting off cellobiose units from the nonreducing end of the chain. The B-glucosidases hydrolyse cellobiose and soluble cello-oligosaccharides with a low degree of polymerisation, to glucose but cellulose is not 33  degraded. In a study of cellulose digestion from eleven forages using pure and mixed cultures of rumen bacteria as well as through in vivo digestion trials, Dehority (1991), reported a higher extent of cellulose digestion with F. succinogens A3c compared with R. flavefaciens B34b and in vivo cellulose digestion. Similar findings have been reported by Cheng et al (1983/84). succinogens is one of the rumen bacteria most active in growth related degradation of recalcitrant forms of cellulose such as cotton fibers and cellulose powder (Chesson and Forsberg, 1988). These gram negative, rod shaped bacteria are reported to produce high levels of endoglucanase and Bglucosidase (Groleau and Forsberg, 1981). Cells grown on cellulose produced 7-8 times more endoglucanase than either cellobiose-grown or glucose grown cells (Groleau and Forsberg, 1981) and it was suggested that endoglucanase is subject to a type of catabolite repression. The bacteria F. succinocrens is reported to form intimate adhesion to cellulose substrate (Cheng et al., 1983/84). They further reported that these bacterial cells were also associated with pit formation in the cellulosic cell walls of the grass leaves used in their study and that in many cases they were associated with complete digestion of these cell walls. Besides F. succinogens, other gram-positive coccus bacteria have been noted to be associated with pit formation and are reported (Cheng et al., 1983/84) to resemble those of pure cultures of Ruminococcus flavefaciens used by Latham et 34  al (1978a). Pettipher and Latham (1979) have studied the characteristics of the cellulase enzymes produced by R. flavefaciens. They reported that the pH optimum was between 6.4 and 6.6 and the temperature optimum between 39 and 450C. Cellulase^activity was^found^to^be mainly cell^associated during^exponential^growth,^but cell-free enzyme accumulated during  the stationary phase.^Another important  bacteria  is^Ruminococcus^albus.^Leatherwood  reported  that^this^bacterium produces^about  cellulolytic (1965)  has  20%^of  the  endoglucanase in rumen fluid, with some 58% of the cellulase being produced from the cells. Stack and Hungate (1984) discovered that 3-phenylpropionic acid in rumen fluid was the active compound which, upon inclusion in the growth medium in the place of rumen fluid, caused albus to grow faster and to synthesise a very active, cell-associated, high molecular weight cellulase. Cheng et al. (1983/84) concluded that this may indicate that selected endoglucanases are secreted during growth. Since rumen microorganisms and indeed those of the digestive tract have complex nutritional requirements and can only utilise one or two of the major polysaccharides, synergism among microorganisms can be important for the efficient use of forages by the ruminant animal. Synergism has been observed in the way of crossfeeding of hydrolysis products, utilization of end-products or production of an essential nutrient (Dehority, 1991). Russell (1985) reported a good example of crossfeeding of hydrolysis products when he 35  showed that non-cellulolytic bacteria could utilise cellodetrins released by cellulolytic bacteria. The combination of the cellulolytic species F. succinogens, R. flavefaciens or R. albus with non-cellulolytic Treponema or Butyrivibrio species sharply accelerates the rate of cellulose digestion, and while cells of F. succinogens alone are unable to clear cellulose agar, these microorganisms can clear large areas of the agar if they are combined with Treponema or Butyrivibrio species (Kudo et al., 1986). Ruminant cellulose digestion is very efficient because cellulolytic organisms are closely associated with symbiotic species that remove the products of their digestive enzymes, and thereby accelerate the digestive process (Cheng et al., 1991).  2.3.3 Mechanism of hemicellulose degradation  Digestion of hemicelluloses has been recognised for a long time. Dehority et al. (1962) traced this activity to the rumen microbial population and studied it in vitro with mixed cultures. They reported that, similar to cellulose digestion from forages, rate and extent of hemicellulose digestion decreased markedly with plant maturity. Xylanases are the main hemicellulose hydrolysing enzymes. They are more widely distributed than are cellulases among rumen bacteria (Chesson and Forsberg, 1988). Microorganisms often produce more than one xylanase, and F. succinogens, for example produces two (Chesson and Forsberg, 1988). B-  xylosidase, B - glycosidase and a - L - arabinofuranosidase are 36  essential for the complete degradation of oligomeric fragments arising from the hydrolysis effected by the polysaccharidedegrading enzymes widely distributed among rumen bacteria (Williams et al., 1984). The major hemicellulolytic bacteria in the rumen are:  Butyrivibrio fibrisolvens, Bacteroids ruminicola  and  Ruminicoccus species (Yokoyama and Johnson, 1988). Most of the predominant cellulolytic ruminicoccus species will degrade and utilise hemicellulose. Recent findings indicate that species of  B.  fibrisolvens, the most frequently isolated  hemicellulose-digesting ruminal microorganism, represent several strains but almost all strains are hemicellulolytic (Hespell, 1988). Unlike the cellulolytic species, strains of this bacteria use hemicellulose degradation products for growth. On the other hand not all strains of B. ruminicola are hemicellulolytic. Hespell (1988) attributes this to several factors, among them, the type of substrate used to grow the organism. He showed that the digestibility of alfalfa hemicellulose decreases with the age of the plant irrespective of the bacterial species that carries out the degradation. Coen and Dehority (1970) observed that even though B.  ruminicola strain D31d digested alfalfa hemicellulose, this strain was almost incapable of digesting bromegrass hemicellulose. This could be due to the chemical and structural differences between the two hemicelluloses since the bromegrass has no rhamnose or uronic acids and considerably more glucose than alfalfa (Hespell, 1988). 37  The biochemistry and enzymology of hemicellulose degradation by ruminal bacteria is not well understood. In  Ruminococcus albus, the cellulase and other enzymes appear to be organized into highly structured, high molecular weight, extracellular complexes of proteins and polysaccharides (Stack and Hungate, 1984). With F. succinogens, cellulase and xylanase activities are associated with large membraneassociated complexes (Groleau and Forsberg, 1981). 11,  fibrisolvens produces extracellular polysaccharides having a complex sugar composition. Hespell (1988) speculated that these extracellular complexes could aid the cells of various species to associate with the insoluble plant materials and aid in hindering diffusion of the enzymatically generated products away from the cell. The glycanases and glycosidases with roles in hemicellulose digestion are subject to carbon source-dependent regulation (Greve et al., 1984). The xylanase from R. albus is produced when cells are grown on filter paper but not when they are grown on cellobiose (Greve et al., 1984), and xylanase from R. flavefaciens is produced when cells grow on either filter paper or cellobiose (Pettipher and Lantham, 1979). Forsberg et al. (1981) reported that xylanase from F.  succinogens could only be produced by cells grown on filter paper and on Avicel but not on glucose. Chesson and Forsberg (1988) suggested that such findings could imply that xylanases from R. albus and F. succinogens are subject to repression. The distribution of glycanases and glycosidases among 38  rumen ciliate protozoa is widespread (Chesson and Forsberg, 1988). The highest activities have been found in Epidinium ecaudatum caudatum, Eremoplastron bovis Ostracodinium obtusum bilobum and Eudiplodinium maggi. Hemicellulase activity was first detected in the rumen ciliate protozoon Epidinium caudatum (Bailey et al., 1962). A wide range of glycoside hydrolases are synthesized by rumen protozoa (Williams et al., 1984). The glycoside hydrolases involved in the degradation of plant cell wall structural polysaccharides, including a-Larabinofuranosidase, B-D-galacturonidase, B-D-glucosidase and B-D-cellobiosidase, were higher in the cellulolytic entodiniomorphid ciliates. On the other hand anaerobic rumen fungi have also been reported to secrete a wide range of glycanases and glycosidases (Mountfort and Asher, 1985). Neocallismatix frontalis has been reported to produce a highly active, extracellular cellulase, several-fold higher in activity than that of the aerobic fungus Trichoderma reese (Wood et al., 1986).  2.3.4 SUMMARY STATEMENT  The foregoing discussion has dealt with the main microbes involved in cell wall degradation and on the necessity of a whole complement of them in cell wall degradation. The significance of synergism has been stressed. In most studies dealing with degradation of cell wall polymers, isolated pure forms have been used. However, in this situation any interactions between the wall polymers and their effect on 39  degradation cannot be detected. In the present study, the plant materials studied were incubated in their intact forms in the rumen with the hope that isolated polymers from the residues would give an idea of recalcitrant polysaccharides in the intact plant. This approach was preferable to incubating the isolated polymers and assessing their disappearance.  40  2.4^METHODS OF EVALUATING THE NUTRITIVE VALUE OF FORAGES 2.4.1 Introduction  Several techniques have been developed that attempt to estimate the contribution of feed protein or carbohydrate to the rumen. These methods are either animal or laboratory based. The following discussion will focus on chemical methods of assessing forage nutritive value as well as  in  vitro and in  situ methods.  2.4.2 Chemical procedures The conventional system of feed evaluation that has been in use for the longest time is the so-called proximate system of analysis. This system partitions dry matter into ether extract, crude fiber, nitrogen, nitrogen-free extract and ash (Van Soest, 1982). It is the basis on which TDN is calculated using the following assumptions: 1. ether extract recovers lipids and fats which contain 2.25 times the energy of carbohydrate. 2. All nitrogen is in protein which contains 16% nitrogen. 3. Crude fiber recovers the least digestible fibrous and structural matter of the feed. 4. The NFE represents highly digestible carbohydrates. Van Soest (1982) who has given an extensive review of this system argues that none of these assumptions are entirely true. The NFE contains the cumulative errors of all the other determinations. However, the greatest and most fundamental 41  error of the proximate system of analysis is the division of carbohydrates into crude fiber and NFE (Van Soest, 1982). Crude fiber has in some instances been found to be more digestible than NFE especially in grasses that contain more hemicellulose and soluble lignin. Other systems of feed evaluation based on comparative replacement values of feeds are used in Europe. They include the Scandinavian feed unit system and the starch equivaleht system (Van Soest, 1982). In the case of the Scandinavian feed unit system, the value of barley is taken as 100 and the relative quantity of other feeds required to replace the productive value of a unit of barley is taken as its feed unit equivalent. The starch equivalent is expressed as the net energy (NE) value of feed in units relative to the NE of lkg of starch. A significant improvement in forage quality evaluation methodology was the description of plant composition in terms of cell content and cell walls (Van Soest, 1967). According to this system, the cell contents comprise components that are digestible by enzymes secreted in the digestive tracts of all animals (Barnes, 1973), while the cell wall is comprised of those components that are only partially digestible and then only by rumen and cecal microorganisms. The cellular contents are soluble in neutral detergent and include minerals, lipids, sugars, organic acids, pectin, starch, soluble proteins and non-protein nitrogen. Van Soest and Moore (1966) estimated the cellular contents were 98% digestible and not affected by 42  lignification. The cell wall constituents (CWC) or neutral detergent fiber (NDF) are insoluble in neutral detergent and are only partially available. These constituents include mainly cellulose, hemicellulose, lignin, cutin and ash. Acid detergent fiber (ADF) on the other hand , quantitatively recovers cellulose, lignin, silica and unavailable nitrogen. The importance of ADF is that it represents a fractionation of the plant cell wall whereby hemicellulose is dissolved and partitioned from the cellulose and lignin which remain in the residue (Van Soest, 1984). ADF is highly negatively correlated (r = -0.75) to digestibility while NDF which reflects more on the volume of a forage consumed is highly negatively correlated (r = -0.76) to voluntary intake (Van Soest, 1984).  2.4.3 Biological procedures In this section the use of enzymatic,  in  vitro and  in  situ  techniques of evaluating the nutritive value of forages will be discussed. Enzymatic methods Enzymatic techniques have been employed to study the digestibility of the nutritional components of by forages various workers (Jones and Hayward, 1973; 1975; Hungate et al., 1983). Jones and Hayward (1973), observed that there are several advantages of using enzymatic methods over fermentation methods. They cited the variations observed 43  between animals and those due to different diets as examples which can be overcome by use of enzymes. The enzymes are either acquired from commercial sources or extracted from the rumen fluid. The most common sources of cellulase are the fungi Trichoderma viride and Trichoderma reesei. Proteases are used to study protein degradability. Even though there are advantages of using commercial enzyme preparations (low cost, time reduction and less contamination of feed residue) over live microbial cultures there are concerns, particularly with commercial proteases, about their specificity in relation to ruminal proteolytic activity (Nocek, 1988). Mahadevan et al. (1987) concluded that the use of non-rumen proteases in an in vitro system for prediction of feed protein degradation may be of limited value or misleading, since protease other than of ruminal origin may not have the same action on feed proteins. Enzymes derived from different fungal sources have been reported to differ in their ability to solubilize herbage and cellulose (Jones and Hayward, 1975). Among four different fungi, Trichoderma viride was found to be the most active, solubilising 70% of cellulose paper in 24 hours. Nocek and Hall (1984) have further reported that use of enzyme procedures generally results in less solubilization of dry matter than does use of ruminal microbes. They observed lower cell wall digestion with enzyme combinations than with in situ values. They concluded that this may be associated with limited enzyme, change in incubation environment with time, or 44  inadequate complements of enzyme types to simulate ruminal digestion. The possibility of lack of physical association of the commercial enzyme preparation with fiber particles could not be ruled out. Nocek (1988) has concluded that enzymatic digestion techniques may be more suitable for measuring relative differences between feedstuffs than providing absolute digestibility values. Accuracy of prediction appears to depend greatly on the forage or feedstuff in question, and the relative complement of enzymes used in the incubation. In vitro methods  The most commonly used in vitro system is the batch fermentation method. Under this we have the single-stage and the two-stage systems. Even though there have been many modifications to the original method used by Tilley and Terry (1963) the basic elements remain the same. The first stage of a batch in vitro procedure simulates rumen digestion of structural carbohydrates. The two-stage fermentation technique attempts to measure not only the digestible fibrous fraction but also the digestible soluble fraction. The second stage involves the extraction of the residue from the first stage with acid-pepsin (Tilley and Terry, 1963) or neutral detergent solution (Van Soest et al., 1966). The acid-pepsin digestion stage simulates the in vivo breakdown of feed and microbial protein by the digestive enzymes of the abomasum in the ruminant. Dry matter 45  disappearance is greater following neutral detergent extraction due to the solubilization of the cell wall. Thus, in vitro procedures with neutral detergent have been equated to true digestibility and procedures with acid pepsin equated to apparent digestibility (Van Soest et al., 1966). High correlations between one-stage in vitro and in vivo cellulose digestion or dry matter disappearance have been reported but interlaboratory comparisons reveal high variations. Barnes (1967) reported mean in vitro cellulose (24h) values ranged from 40.0 to 63.9% among 14 laboratories, and in vitro dry matter digestibility (24h) ranged from 38.7 to 53.5% among five laboratories. The addition of the secondstage acid-pepsin digestion gave higher correlation coefficients and lower errors compared to results from the one-stage technique. The greatest source of uncontrolled variation in any in vitro system is the inoculum (Barnes, 1973). Use of washed cell suspensions or rumen fluid strained through many layers of cheesecloth have been used successfully to reduce variability between analysis (Rode and Satter, 1984). However this method may not be recommendable since the removal of forage particles is likely to remove the most active cellulolytic microorganisms. Other sources of random error include donor diet, buffer medium, sample size, and sample (Barnes, 1973). Besides random error, in vitro techniques are associated with predictive errors. These are errors emanating from the use of mathematical models to predict in vivo 46  measurements from in vitro values (Barnes, 1973). Inability of the batch culture to adequately account for the dynamic situation within the rumen led to the development of semi-continuous and continuous systems that more closely approximate ruminal conditions (Abe and Kumeno, 1973; Czerkawski and Breckenridge, 1979). Slyter and Putnam (1967) indicated protozoal numbers were lower and viable bacteria numbers slightly higher when compared with ruminal environment of a steer fed the same diet. Czerkawski and Breckenridge (1977) on the other hand, observed that protozoa numbers were similar between  in  yivo data and in vitro simulations.  Mathematical descriptions of continuous culture kinetics have been expanded to account for the more complex behavior of substrates and microbes in activated sludge and other heterogenous ecosystems (Hashimoto et al., 1982) but have not been used to describe the rumen ecosystem to date. In situ techniques The in situ technique, also variously referred to as the dacron bag, nylon bag, in sacco technique has been used extensively to study rumen degradability of nitrogen (N), dry matter (DM), organic matter (OM) and fiber fractions (Susmel et al., 1990). This method involves suspension of feed material in the rumen and allows intimate contact of the test feed with the ruminal environment. The method allows the description of a degradability kinetic (Orskov et al., 1980), which generally follows a first-order model, and the quantity 47  of nutrient effectively degraded in the rumen can be calculated from kinetic coefficients of degradability and the outflow rate of feed from the rumen. The nylon bag technique is subject to considerable variability. Sources of variation include: size and type of bags; cloth mesh size; sample size and fineness of grind; number of samples per trial; diet of the host animal; method of suspension in the rumen; location and time in the rumen; and method of cleaning and rinsing bags after removal from the rumen (Barnes, 1973). An extensive review by Nocek (1988) of the in situ technique has addressed these problems and produced some recommendations. On the question of sample and bag size, he suggests the optimum sample size to be that which will provide adequate residue at the end of extended rumen incubation for chemical analysis without overfilling the bag so as to delay bacterial attachment, increase lag time and underestimate digestion rate. He recommends a range in sample size to surface area ratio of 10 to 20 mg/cm2 for both forage and concentrate diets. Grinding of forages increases surface area per unit weight of sample and the surface area accessible for microbial attachment. Fineness of grind is important as it determines how much material is passively lost through the pores of the bags and also the digestion rate. Generally, longer and coarser materials are associated with slower rates of digestion and greater variation (Nocek, 1988). Grinds of less than lmm have little effect on digestion rate. Van Karen and 48  Heinemann (1962) demonstrated no difference in DM disappearance for forages ground through screens of 0.28, 0.42 and 0.84 mm. However, decreasing particle size to < 0.6mm may also cause clumping of the sample, thus decreasing digestion rate (Figroid et al., 1972). Nocek (1988) suggests a grind of 2mm for protein supplements and by-product type ingredients and 5mm for forage materials. Utilization of forages involves complex interactions among plant components, microorganisms in the rumen and the animal. These factors interact to influence both digestibility and voluntary intake. Disappearance of feed from the digestive tract can be described by two major processes, digestion and rate of passage (Mertens and Ely, 1982). Waldo et al. (1972) proposed that fiber exists in two definable components -potentially digestible and indigestible. While the indigestible fraction disappears from the rumen by passage only, the potentially digestible fraction disappears by passage, digestion and absorption. The digestive process can be divided into rates of digestion, digestion lag and potentially digestible fractions (Mertens and Ely, 1982). Initial work on digestion kinetics did not consider a lag phase and only assumed the existence of three pools (Orskov et al., 1980): pool A that was considered to be rapidly degraded in the rumen; pool B that was considered to be potentially degradable with time and pool C representing a fraction that appeared undegradable in the rumen irrespective of time. For protein degradation, DM and 49  other components, Orskov and McDonald (1979) had shown that if the percentage protein disappearance (p) from samples incubated for time t is described by the equation p = a + b (1 _ e-ct) and if k is the fractional rate of passage from the rumen, then the effective degradability can be calculated as P = a + bc/(c + k) where p = the actual degradation after time 't'; a = the component of protein degraded rapidly and forms the intercept of the degradation curve at time zero, b = the potential degradability of the component of protein which will, in time, be degraded. c = the rate constant for the degradation of 'b' (%/hr). P = the effective degradability of protein. With fibrous feeds there is often a lag phase in the degradation of DM due to the time taken for adherence of cellulolytic organisms to the substrate (Orskov, 1991). As such McDonald (1981) has improved on the above equation by incorporating a lag time factor associated with the degradation of the b component.  2.4.4 SUMMARY STATEMENT  Chemical and biological (in vitro and in situ) methods of assessing the nutritive value of feed materials for ruminants were discussed. The weak and the strong points of each method were also mentioned. The in situ method was chosen for this study. This was partly because it eliminates variations (in in vitro systems) 50  emanating from use of inoculum which has been filtered through cheesecloth and may therefore be associated with low cellulolytic and hemicellulolytic activity. It was also felt that more substrate material could be used with this method than with an in vitro one.  51  2.5 OVERALL SUMMARY  The foregoing literature review has dealt with aspects that influence or have a potential influence on the degradability of forages by ruminants. Such plant factors as the species, maturity and anatomy of morphological parts were discussed. The role of chemical composition, with an emphasis on the cell walls as the primary physical factor regulating voluntary intake in ruminants was stressed. Having established the role played by the cell walls in forage utilization, a discussion of the main factors that influence the breakdown of its constituent polysaccharides was given. Such factors as cellulose crystallinity, linearity and branching of hemicelluloses, formation of covalent linkages with other simple or complex phenolics of the wall were discussed. Microbial degradation of cell walls was covered, with an emphasis on the significance of synergism in wall breakdown. Finally the various methods used in the laboratory to study digestibility were discussed. The present study was initiated with the aim of first studying the factors that limit utilization of two popular Kenyan forages harvested during the dry season. While it is well accepted that the concentration of the cell walls and lignin are the dominant factors regulating forage utilization, a study of the contribution of the constituent polysaccharides to this aspect have not. This is particularly so with these two forages. Past research has shown that the degree of linearization 52  or branching of hemicelluloses may contribute significantly to the overall degradation of plant cell walls. However, factors leading to differential degradations of these wall constituents have not been studied and are mainly speculations. In this study it was postulated that such polymers would be associated with different rates and extents of degradation in the rumen. Such differences were thought to lead to differences in overall cell wall degradation. In the second part of this work, the effect of maturity on the degradability of these polymers in leaves and stems were studied.^Their rates and extents of degradation were determined^with the hope of identifying recalcitrant polysaccharides and the factors leading to this.  53  CHAPTER THREE EXPERIMENTAL: DEGRADATION OF TROPICAL FORAGES  3.1 Introduction There is an enormous amount of literature on utilization of both tropical and temperate forages by ruminants (Minson, 1990; Mowat et al., 1965). One of the major limiting factors to animal production in Kenya is related to the seasonality of forage production which is the uneven distribution of rainfall. During the wet season there is abundant forage and milk production during this period is usually high. However, during the dry season there is a marked decline in milk yield due mainly to a scarcity of adequate forage of high quality to meet the energy demands of the lactating cow. In a country like Kenya where forage plays the major role of meeting both the energy and protein demands of the cow, an understanding of the factors that limit forage utilization is very important. It is generally believed that the concentration of the cell wall in any forage is the dominant factor regulating energy intake. During the dry season, the available forage for cattle feed in Kenya is usually of a mature nature and therefore can be a major limitation to energy and protein intake. It is well documented that protein concentration declines with forage maturation while the cell wall concentration increases (Minson, 1990). These effects are detrimental to animal production. 54  The broad objective of this study was to determine the factors that may limit forage utilization of two widely used tropical forages, bana grass (Pennisetum purpureum) and the legume silverleaf desmodium (Desmodium uncinatum) harvested during the dry season in Kenya. Proper selection of ration ingredients is important in avoiding possible depressions in digestibility and hence available supplies of energy and amino acids. Therefore the present study was undertaken with the following specific objectives in mind: (a) To determine the rate and extent of degradation of the cell wall and its constituents in leaf and stem frations of bana grass and silverleaf desmodium. (b) To compare the degradation characteristics of the grass versus that of the legume. (c) Provide some possible suggestions of alleviating the problem of having inadequate feed of a high quality during the dry season.  3.2 MATERIALS AND METHODS 3.2.1 Forages Two tropical forages were used in this study. Bana grass and silverleaf desmodium were harvested in Kenya during the dry period of September to November of 1991. Bana grass was harvested on Oct. 24th and was a four month regrowth. Silverleaf desmodium was harvested on Oct. 17, 18 and 19th. Like bana grass, desmodium was at the flowering stage during 55  harvesting. Only one sample of each forage was used because it was felt that replications could be generated by using several cows. Also the forages were initially intended to be used for a different experiment (involving a lot of chemical work) rather than the one reported here. After each harvest the legume and the grass were manually separated into leaves and stems as soon as could be possible and then put into polythene bags which were placed in a deep freezer at -8°C. After completion of the harvesting the forage fractions were dried in a forced-draught oven maintained at 600C until sufficiently dry for grinding. The weights of the dried materials were taken and recorded. The stem to leaf ratios determined on the dried materials were 1.75 and 1.35 for bana grass and desmodium respectively. The forage fractions were then ground in a laboratory hammer mill (Christy and Norris, UK) to pass a 2mm sieve screen.  3.2.2 In situ incubations To determine the degradability characteristics of these forage fractions, they were incubated in three Holstein heifers adapted (three weeks) and maintained on a purely grass hay diet. The heifers were fitted with permanent rumen cannulae. About 7g of the ground substrate was weighed into nylon bags measuring 10cm x 20cm with a mean pore size of 53gm (Bar Diamond, Inc. Idaho, U.S.A.). The mouth of each bag was tied and secured with a nylon string before the bags were put 56  in a weighted mesh bag and placed in the ventral sac of the rumen. The substrate to bag surface area ratio was maintained at 17.5mg/cm2. To model the degradability of DM, cell wall (NDF), cellulose and hemicellulose, duplicate substrate samples were incubated in each cow for 12, 18, 24, 48 and 72h. In this first run degradabilities did not reach an asymptote and therefore this experiment was repeated for the following incubation times 12, 18, 24, 48, 72, 96 and 120 hours. Incubations were conducted in reverse time order such that all the bags were removed at the same time at the end of 120h. Zero hour disappearance of dry matter was estimated by placing substrate samples in static water at room temperature for 30 minutes. All the bags were washed under gently flowing tap water (Orskov et al., 1980) before drying in a forced-draught oven at 60°C for 48h. The dried bag plus residue were weighed to determine DM disappearance.  3.2.3 Experimental design Degradability data for the two forages and their leaf and stem fractions were analysed in an 8 (incubation time) x 2 (forage) x 2 (plant part) factorial arrangement in a randomised complete block design. The three cows served as the blocks. The interaction between forage and plant part was also determined. 57  Degradability parameters, 'a' the soluble fraction, 'b' the insoluble but fermentable fraction and 'c' the rate Of degradation of the insoluble fraction 'b' (Orskov and McDonald, 1979) were estimated using the Eureka program (1987). To estimate the effective degradabilities of the DM, the cell wall and its constituents assumed values for the outflow rate (k) were chosen (0.02 and 0.04/h). All statistical analyses and comparisons were conducted using the General Linear Model of SAS 1985).  3.2.4 Chemical analyses DM content of the forage fractions was determined by drying samples to constant weight at 1050C in a forced-draught oven. Ash was determined by igniting the samples at 500°C in a muffle furnace. Crude protein was determined by the method of Parkinson and Allen (1975). NDF and ADF were determined by the method of Waldern (1971). Lignin (permanganate) and cellulose were determined by the method of Goering and Van Soest (1970). Hemicellulose was calculated as the difference between NDF and ADF.  3.3 RESULTS AND DISCUSSION 3.3.1 Chemical composition  Table 3.1 presents the chemical composition values for the two forages and their leaf and stem fractions. As stated earlier the forages were harvested at the flowering stage. The legume fractions contained a higher CP content than the grass fractions. In both cases, leaves showed a trend of higher CP content than stems. Bana stems had the highest cell wall content (78.29%) indicating very low cell contents. Desmodium leaves had the lowest cell wall content. Leaves in both cases were lower in cell wall than stems. Lignocellulose (ADF) content was highest in desmodium stems (62.61%) than in the other fractions. Hemicellulose concentration was lower in both legume fractions relative to the grass fractions. Grass leaves also contained a higher hemicellulose content than grass stems. The grass leaves and stems had a higher concentration of cellulose than the legume. In both cases stems were higher in cellulose than the leaves. Lignin concentration was observed to be greater in stems than in leaves in both the grass and the legume. Legume stems had a very high concentration of lignin (15.43%) relative to the other forage fractions. Ash content was higher in grass leaves than in grass stems but the situation was reversed in the case of the legume.  59  Table 3.1 Chemical composition of bana grass and desmodium leaves and stems (%DM) Treatments Item  Bana stems  CP  4.4  NDF  Desmodium  leaves  stems  leaves  6.8  9.0  25.5  78.29  73.26  71.34  49.08  ADF  51.62  45.77  62.61  41.35  Hemi.  26.67  27.49  8.73  Cellulose  36.08  33.40  25.67  19.31  Lignin  10.75  6.37  15.43  9.24  Ash  9.93  12.39  10.67  10.17  7.73  The results reported here are consistent with previous reports on differences between leaves and stems and between legumes and grasses for both temperate and tropical forages. Norton (1982) for example, indicated that the cell wall content of leaves is usually lower than that of stem, with differences between leaf and stem being greater in legumes than in grasses. In this study, stems were higher than leaves in cell wall content by 5.03 percentage points while in legumes the difference was 22.26 percentage points. The low concentration of hemicelluloses in legumes is well documented. Bailey (1973) reported a lower concentration of total hemicellulose and cellulose in legume leaves compared with grass leaves but no marked differences were observed between legume and grass stems. Lignin and cellulose concentrations were higher in stems than in leaves of the two forages which indicates on the former organ's rigid structure required to 60  hold the plant erect. Ash levels in both forages were high and it was speculated that this may have been the result of contamination of soil minerals effected by the dusty and windy conditions that prevailed during the period prior to harvesting of these forages.  3.3.2 Degradation of DX, the cell wall and its constituents Extent of degradation of DX and the cell wall. Table 3.2 shows the in situ degradation values of DM for the two forage fractions. Only the 48h incubation statistical comparisons are shown in this table, but data on all the other incubation times are also included. Appendix Table 1 shows the ANOVA for DM at 48h for the two forages. 48h comparisons were shown as it was felt that these values could reflect on the extent of degradation at normal rumen retention time (about 50h Van Soest, 1975). At 48h, bana stems DM was more degradable (P < 0.05) than that of bana leaves. This was unexpected but it should be noted that the extent of degradation at 48h depended on the rate at which degradation was changing for both leaf and the stem fractions at that particular time. Desmodium leaves DM was significantly more degraded (P < 0.05) than stems. This is consistent with other findings (Akin, 1989) which indicate that the lbw digestibility of mature legume plants is primarily caused by the low digestibility of the stems. This has been attributed to the presence of heavily lignified xylem, phloem cap and 61  interbundular cells which form a formidable barrier to degradation (Akin, 1989). A significant (P < 0.05) interaction between forage and plant components was observed. This was possibly because the plant parts were obtained from the same original material.  Table 3.2 Extent of degradation of DM in bana grass and silverleaf desmodium. Treatment  INCUBATION TIME (h) 0  12  18  24  18.5  30.8  38.6  44.3  59.8a 63.0  65.8  67.6  B. Leaves 16.0  24.2  34.2  41.3  56.2b 61.6  65.7  67.7  D. Stems  17.5  24.9  33.8  37.6  49.7c 52.9  55.3  56.8  D. Leaves 19.6  39.8  57.5  62.2  73.0d 74.7  76.2  78.2  S.E.M.  1.41  1.30  1.49  0.59  0.43  B. Stems  0.15  48  72  0.52  96  120  0.36  a,b,c,d Means followed by a different letter within the 48 and column differ (P < 0.05).  DM degradability  Desmodium stems and leaves  Figure 3.1  Table 3.3 Extent of degradation of NDF in bana grass and silverleaf desmodium. Treatment  INCUBATION TIME (h) 0  12  18  24  18.5  30.1  37.5  44.2  60.2a 63.8  66.6a  68.6  B. Leaves 19.9  28.2  40.5  48.0  63.2b 68.0  71.0b  73.2  D. Stems  19.8  23.2  32.6  37.7  49.4c 53.2  54.8c  57.2  D. Leaves 37.9  49.8  65.1  70.5  77.9d 79.5  80.4d  82.3  S.E.M.  1.17  1.22  1.37  0.62  0.42  0.31  B. Stems  0.21  48  72  0.48  96  120  a,b,c,d Means followed by a different letter within the 48 and 96h columns differ (P < 0.01). Degradation of bana grass cell walls (NDF, Table 3.3) was higher (P < 0.01) for leaves than for stems at 48h and 96h. A similar situation was observed with the legume. Grass stem cell walls were more degradable (P < 0.01) than those of legume stems, but legume leaf cell walls were more degradable (P < 0.01) than those of grass leaves. It is well established that legumes have higher lignin concentrations than grasses and that the concentration is greater in stems than in leaves (Jung, 1989). Smith et al. (1972) reported that legume cell walls had a lower extent of digestion than grass cell walls and were also associated with higher lignin concentration's. This could explain the observation in this study where grass stems were more degradable than those of legumes. In the case of leaves, Van Soest (1982) notes that grass leaves have both a metabolic and a structural role through the midrib which may 64  result in lower degradability of this tissue compared with legume leaves which mainly have a metabolic role. Grass leaf cellulose was degraded to the same extent at 48h. as stem cellulose (Table 3.5). Similarly no significant difference was observed between grass leaf and stem degraded at 96h. In the case of the legume significantly more (P < 0.05) leaf cellulose was degraded compared with that of stern. Degradation of the hemicellulose in grass fractions did not differ (P > 0.05) at 48h but that of leaves was more degradable (P < 0.05) than that of stems at 96h (Table 3.4). On the other hand, the legume fractions differed (P < 0.05) in their hemicellulose degradability at the two incubation times. Desmodium stems had the lowest extent of hemicellulose digestion at 48h and 96h.  Table 3.4 Extent of degradation of hemicellulose in bana grass and silverleaf desmodium. Treatment  INCUBATION TIME (h) 12  18  24  18.5  34.3  37.4  45.4  B. Leaves 15.9  24.4  39.2  D. Stems  17.4  21.3  D. Leaves 22.2  B. Stems  S.E.M.  0  48  72  96  120  61.6a 64.4  66.2a  68.1  45.6  64.0a 68.3  70.1b  73.0  21.9  27.7  37.2b 42.8  41.0c  44.6  23.4  56.3  65.2  77.7c 80.6  81.0d  83.0  1.21 2.89  1.63  2.56  2.15  1.17  1.13  1.01  a,b,c,d Means followed by a different letter within the 48 and 96h columns differ (P < 0.05).  65  Table 3.5 Extent of degradation of cellulose in bana grass and silverleaf desmodium. Treatment  INCUBATION TIME (h) 0  12  18  24  18.5  35.3  38.8  44.5  B. Leaves 16.0  27.0  38.3  D. Stems  17.4  26.0  D. Leaves 14.1 S.E.M.  B. Stems  1.02  48  72  96  120  63.9a 67.6  71.4a  72.7  45.7  63.5a 69.1  72.7a  74.2  37.6  39.3  56.9b 59.3  60.7c  63.2  31.4  48.2  48.3  63.5a 63.8  65.2d  68.6  2.61  1.47  1.95  0.81  0.51  0.50  1.11  a,b,c,d Means followed by a different letter within the 48 and 96h columns differ (P < 0.05). DEGRADATION CHARACTERISTICS OF DM, THE CELL WALL AND ITS CONSTITUENTS Tables 3.6 to 3.8 are a presentation of the degradation kinetic parameters, as defined by Orskov and MacDonald (1979), for DM, the cell wall and its constituents. Leaves of desmodium had a higher (P < 0.05) insoluble but potentially degradable (fraction ,b,) DM compared with desmodium stems. However, grass leaves did not differ (P > 0.05) from grass stems in this fraction. Grass leaf and stem fractions did not differ (P > 0.05) in their cell wall (NDF) fraction 'b' content. This could be compared with the results of Fritz et al. (1990) who observed a higher concentration of cell walls in leaves than in stems of sorghum x sudangrass genotypes. The rates of degradation associated with this fraction in the two morphological parts did not differ (P > 0.05). This was unexpected more so because leaves are known to contain highly degradable and unlignified cells (for example the mesophylls). It is however noted that lignin is not related to rate of 66  degradation of the cell wall but rather determines the extent of degradation. The potentially degradable hemicellulose was higher (P < 0.05) in grass leaves than in stems. However, this fraction did not differ (P > 0.05) for cellulose in both leaves and stems of the grass. A similar situation existed with the legume. No difference (P > 0.05) was observed in the rates of degradation of these polymers in both grass and leaf fractions. The rate of degradation of legume cellulose and hemicellulose were 1.3 and 1.5 higher respectively in leaves than in stems. The amounts of these polymers potentially degradable in the rumen were higher (P < 0.05) in leaves than in the stems. Generally the grass fractions were higher than the legume fractions in potentially degradable cellulose and hemicellulose. This could be related to the lower levels of these polymers in the original legume plant materials. Amongst the grass and the legume plant parts, legume leaves had significantly (P < 0.05) higher rates of degradation of the cell wall constituents. The grass and the legume did not differ in lag phase possibly because of the large error. Smith et al. (1972) observed that legume cell walls were digested at a faster rate than grass cell walls, even though they generally had a higher lignin content and lower extents of digestion than grasses. The higher rate of degradation of legume leaves in this study, could lead to higher rate of digestion of the whole plant.  Table 3.6 Bana and desmodium DM degradation parameters DEGRADATION PARAMETERS  Treatment  a  b  lag (h)  c  Bana^Stems  19.0  54.7 ac  3.8  0.03  Bana Leaves  16.0  58.2 c  7.0  0.04  Desmodium Stems  17.4  45.1 b  7.1  0.05  Desmodium Leaves  14.1  52.7 a  5.3  0.08  1.46  1.4  0.01  S.E.M.  1.42  a t b f c Means followed by different letter(s) within a column differ (P < 0.05). a= soluble material (%) b= potentially degradable material (%) c= rate of degradation of 'b' (%) Table 3.7 Bana and desmodium EDF degradation parameters DEGRADATION PARAMETERS  Treatment  a  b  lag (h)  c  Bana^Stems  18.5 a  50.1 a  5.6  0.04 a  Bana Leaves  19.9 b  52.3 a  8.2  0.05 ab,  Desmodium Stems  20.1 b  35.9 b  10.1  0.05 ab  Desmodium Leaves  37.9 c  42.5 c  8.6  0.1 b  0.17  0.64  1.1  0.01  S.E.M.  a ( b / c Means followed by different letter(s) within a column differ (P < 0.05).  Table 3.8 Bana and desmodium cellulose degradation parameters Treatment  DEGRADATION PARAMETERS b  a  lag (h)  c  Bana^Stems  18.9  54.7 a  3.8  0.03 a  Bana Leaves  16.0  58.2 a  7.0  0.04 a  Desmodium Stems  17.4  45.1 b  7.1  0.05 a  Desmodium Leaves  14.1  52.7 a  5.3  0.08 b  1.46  1.4  0.01  S.E.M.  1.42  a t b Means followed by different letter(s) within a column differ (P < 0.05).  Table 3.9 Bana and desmodium hemicellulose degradation parameters Treatment  DEGRADATION PARAMETERS a  b  lag (h)  c  Bana^Stems  18.5  50.0 a  3.5 a  0.04 a  Bana Leaves  15.9  56.0 b  8.7 b  0.06 a  Desmodium Stems  14.8  29.3 c  13.0 c  0.06 c  Desmodium Leaves  20.8  60.0 b  10.9 bc  0.1 b  S.E.M.  1.69  1.62  0.8  0.01  a ( b i c Means followed by different letter(s) within a column differ (P < 0.05). EFFECTIVE DEGRADABILITY OF DM, THE CELL WALL AND ITS CONSTITUENTS  Results for the effective degradabilities (based on both the rate of digestion and the assumed rates of passage through the rumen) are presented in Tables 3.9 to 3.12. The effective degradability of DM was higher (P < 0.05) for grass leaves than for the stems. Effective degradability of cellulose and hemicellulose did not differ (P > 0.05) between bana leaves and stems. However, cell walls were more degradadable (P < 0.05) in leaves than in stems. Reid et al. (1973) made the same observations with the same grass in Uganda where they reported no significant differences in DM digestibility between leaf and stem until the regrowth was 16 weeks. In this study the grass was harvested as a 4 months regrowth. In their study Reid et al. (1973) observed that the digestibility of the whole plant at 16 weeks was 48.9%. In this study at 50h the degradability of stems was 48.2% and that of leaves 45.3% (Table 3.9). In the case of the legume leaf and stem fractions, cell wall components were more effectively degraded in leaves than in stems (P < 0.05). DM was significantly (P < 0.05) more effectively degraded in bana stems compared with desmodium stems. But desmodium leaf DM was more degradable (P < 0.05) compared to that of bana leaves. In Australia, Graham (1967) described a mature sample of uncinatum as highly lignified material of low digestibility (about 40%), but found that 70  sheep in metabolism cages ate relatively large amounts of it, so that levels of energy retention were unusually high for such a coarse feed. The low degradability of desmodium stems could be related to the observed low and slow rate of degradation. The low digestibility of the whole plant may be as a result of the low degradability of stems as observed in the present study. Stems were also associated with the highest concentration of lignin (15.43%). Table 3.10 Effective degradabilities of DX at different rumen passage rates. Treatment  MEAN FLOW RATES k = 0.02  k = 0.04  Bana Stems  48.2a  38.8a  Bana Leaves  45.3b  34.9b  Desmodium Stems  40.7c  33.0b  Desmodium Leaves  59.6d  49.1c  0.31  0.67  S.E.M.  a ( b I c t d Means followed by different letter(s) within a column differ (P < 0.05). Table 3.11 Effective degradabilities of EDF at different rumen passage rates. Treatment  MEAN FLOW RATES k = 0.02  k = 0.04  Bana Stems  48.2 a  30.4a  Bana Leaves  51.0 b  40.4b  Desmodium Stems  40.9 c  33.3c  Desmodium Leaves  67.6 d  59.4d  0.42  0.46  S.E.M.  acb,c,d Means followed by different letter(s) within a column differ (P < 0.05). 71  Table 3.12 Effective degradabilities of cellulose at different rumen passage rates. Treatment  MEAN FLOW RATES k = 0.02  k = 0.04  Bana Stems  51.2 a  40.9 a  Bana Leaves  50.2 a  38.5 ab  Desmodium Stems  44.8 b  35.8 b  Desmodium Leaves  50.0 a  40.4 a  0.88  1.15  S.E.M.  a/b, Means followed by different letter(s) within a column differ (P < 0.05).  Table 3.13 Effective degradabilities of hemicellulose at different rumen passage rates. Treatment  MEAN FLOW RATES k = 0.02  k = 0.04  Bana Stems  49.2 c  39.9 c  Bana Leaves  49.7 c  38.2 c  Desmodium Stems  30.6 a  24.3 a  Desmodium Leaves  61.8 b  49.5 b  1.00  1.45  S.E.M.  a f b f c Means followed by different letter(s) within a column differ (P < 0.05).  3.4 SUMMARY AND CONCLUSIONS  Understanding the factors that limit forage utilization, particularly in those countries that depend on these materials as the sources of energy and protein for their livestock, is very important. In this study two locally popular tropical forages, bana grass and silverleaf desmodium (legume) were harvested during the dry season in Kenya and studied for their rates and extents of degradation in the rumen. Stems of both forages were found to be high in cell wall content, ADF and lignin. Desmodium leaves were low in both the cell wall and ADF. The legume fractions were low in hemicellulose relative to the grass. However, their lignin content were high particularly in the stems. Cellulose and hemicellulose were high in the grass fractions. Results from this study indicated that the two forages were poorly degraded in the rumen. Desmodium stems were particularly so, a finding that seems to agree with the report of Akin (1989) that the low digestibility of the mature whole plant may be because of the low digestibility of the stems. Desmodium leaves were however moderately degradable. This could be related to a high content of rapidly degradable cell types like the mesophyll cells (Minson and Wilson, 1980). There was no difference in the rate of degradation of DM in the grass fractions but this was higher in desmodium leaves compared to the stems. Grass leaves were higher in content of potentially degradable cellulose and hemicellulose than the 73  stems. The rates of disappearance of these polymers in both leaves and stems was not different. In terms of the effective degradabilities of DM and the cell wall components the results indicated that the grass fractions were not different. Legume fractions were however different in this respect, with the leaves having higher effective degradabilities than stems. The effect of higher passage rates especially with forages of poor quality is to reduce their degradability in the rumen and this was evident in this study by comparing two passage rates, 0.02 and 0.04/h. The low degradability of the two tropical forages in this study may be related to their high cell wall and lignin contents. It was clear with the desmodium stems that lignin could have played a major role in limiting the degradation of the cell wall and its constituents. Because of the high degradability of the legume leaves and its low cell wall content and high crude protein content it is suggested that this forage could be combined with the grass during the dry season to improve intake. It is however noted that desmodium is associated with high levels of tannins, but since its intake seems to be relatively high in sheep (Graham, 1967) it is possible that with cattle the intakes may even be better.' It is suggested that during the wet season when there is abundant forage of high quality (high protein content and low cell wall concentration) some form of conservation, particularly ensiling, can be adopted. Fisher and Shelford 74  (1988) have enumerated the possible ways through which forages lose their nutritive value during the ensiling process. Delaying cutting should be avoided as it is one of the ways nutritive loss occurs.  CHAPTER FOUR EXPERIMENTAL: DEGRADATION OF TEMPERATE GRASSES  4.1 Introduction The role of the cell walls in regulating voluntary intake and digestibility is well documented. Since the cell walls have a major role contributing to the fill effect in ruminants, factors which influence their rate and extent of degradation will reflect on voluntary intake. Such factors as cellulose crystallinity, linearity and branching Of hemicelluloses as previously discussed, have profound effects on cell wall degradation parameters (rate and extent of degradation). These parameters will also differ depending on the type of cells being degraded. Mesophyll cells, which comprise most of the leaf tissue, are practically unlignified and therefore highly degradable. Most experiments on cellulose and hemicellulose degradation have been conducted on the isolated polymers which reportedly are rapidly degraded. This scenario may be very different from that of the intact plant. In proposing this study, it was hypothesized that the overall rate and extent of cell wall degradation will be determined by the rates and extents of degradation of individual polymers. It was speculated that linear and branched hemicellulosic polymers will differ in these parameters and that maturity could have a significant effect in this respect. The main objective of this study was to 76  determine the degradability of cellulose, ^linear xylan (Hemicellulose A) and a mixture of linear and branched shortchain heteroxylans (Hemicellulose B) by incubating the intact plant parts in the rumen and then extracting those polymers that survive rumen incubation.  4.2 Objectives The specific objectives of these experiment were: 1. to determine the effect of maturity and species on the levels of cellulose, linear xylan and branched heteroxylan in both leaf and stem fractions of orchard grass and tall fescue; 2. to investigate whether there are any differences in the degradability of each of these polymers between leaves and stems; 3. to investigate whether there are differences between the degradability of linear xylan (hemicellulose A) from mixed linear and branched heteroxylans (hemicellulose B) within the same plant part; 4. to relate the degradabilities of the various polymers to lignin and its concentration on the respective plant parts. 5. to interprete the observed results in terms of solving practical animal production problems.  77  4.3 MATERIALS AND METHODS 4.3.1 Forages  Two forages, orchard grass (Dactylis alomerata) cv. Mobite and tall fescue (Festuca arundinacea) cv. Courtney, were used in this experiment. They were harvested in the spring of 1992 from plots at Agriculture Canada Research Station - Agassiz B.C. Orchard grass was harvested on three different dates; May 11th., May 20th., and May 29th. This particular grass shows rapid decline in digestibility within short growth periods (Shelford and Fisher, 1988). Tall fescue was harvested on two dates; May 20th. and May 29th. A third harvest, June 5th. was discarded after it was found that it was not a first cut as originally intended. These grasses were manually separated into leaves and stems. Leaves were defined as those parts that form the leaf blade while stems included the leaf sheaths. Dead leaves were discarded. The separate fractions were then dried in a forced draught oven maintained between 55 - 60°C until they were sufficiently dry for grinding. The weights of the dried leaf and stem fractions of the two grasses for each harvest date were taken and recorded. Wet weights of the separated leaf and stem fractions were not taken. After drying the materials were then ground to pass through a 2mm sieve in a laboratory hammbr mill (Orskov et al., 1980).  78  4.3.2 In situ incubations  Three Holstein heifers fitted with permanent rumen cannulae were used in this study. They were adapted (two weeks) and maintained on a purely grass hay diet fed twice daily (0800 and 1400hrs). About 7g of the ground substrate was weighed into nylon bags ( size 10cm x 20cm with mean pore size of 53Am) and the mouth of each bag tied and secured with a nylon string. The substrate weight to bag area ratio was 17.5mg/cm2 which is well within the recommended range of 10 to 20mg/cm2 (Nocek, 1988). Only one run was possible because of a shortage of cannulated animals at the time, inadequacy of original forage material and a tight schedule for the available animals. To model the degradability of DM, cellulose, linear xylan and branched heteroxylan components of the forage, duplicate substrate samples were incubated in each cow for 12, 18, 24, 48 and 72 hours. The bags were secured in a weighted mesh bag which was held about 50cm from the mouth of the cannula. All the bags were placed in the ventral sac of the rumen. Samples to be exposed for 72 hours were placed into the rumen fir'st about one hour before feeding time. Introduction of the remaining samples was conducted in reverse order over time, such that, the last samples (12h incubation) were introduced 60h after introduction of the 72h samples. Thus, all bags were removed at the same time, 72h after entry of the first set with the hope of minimizing variation in washing. Zero hour 79  disappearance of DM was determined by placing substrate sample bags in static water in a basin for 30 minutes. All the bags were washed under gently flowing tap water until rinsings were clear. This took approximately 2 minutes per bag. After gentle squeezing (Orskov, 1980) to remove excess water the bags were dried in a forced draught oven at 55°C for 48h. The bags were then weighed to determine DM disappearance.  4.3.3 Experimental design Data on orchard grass and tall fescue were analysed separately since the available information (physiological maturity) could not justify their combined analysis and comparison. Data on orchard grass DM degradability were analysed as a 6 (incubation time) x 3 (stage of maturity) x 2 (plant part) factorial in a completely randomised block design. Data on tall fescue DM degradability were analysed as a 6 x 2 x 2 factorial also in a randomised complete block design, where cows (3) served as the different blocks. A total of 216 observations for orchard and 144 for tall fescue were made. Due to a severe limitation in the amount of DM residue left after incubation, particularly for the earlier stages of growth, residues from all the cows were pooled which rendered the polymer results statistically not analysable. Incubation of about 7g resulted in less than lg of residue which after 80  treatment with neutral detergent solution would have been reduced by half. While there was adequate amount of leaves from the earlier stages of growth for triplicate incubations, stems were severely inadequate. Other options were to conduct the experiment in several runs but the demand for the few cannulated cows made this impossible. Polymer degradability results are therefore reported as trends rather than statistically defined differences. The results were fitted to the following equation using Eureka program which performs Least Square Fit to find the function (of the required form) that best matches the points x, f(x) where x is the % degradation value and f(x) is the incubation time intervals: p = a + b(1 _ et) where 'p' is the percentage disappearance at time t, 'a' the soluble degradable fraction or physically lost material, 'b' the insoluble but degradable fraction, and 'c' the degradation rate constant. When t = 0, p was fixed as the disappearance due to washing. All these constants were later used to calculate effective DM, cellulose, linear xylan or branched heteroxylan degradation (P) using the formula: bc P= a + ^ c + k where k is the assumed outflow rates ( 0.02, 0.04, ^0.06 and 0.08). All the statistical analyses and comparisons were conducted using the General Linear Model (least square means) 81  of the SAS package (1985).  4.3.4 Chemical procedures With the exception of the degradation measurements, DM concentration was determined by drying samples to constant weight in a forced-draught oven at 100°C. Ash was measured by igniting samples in a muffle furnace at 500 to 505°C. Nitrogen was determined by the sulfuric acid oxidation method (Parkinson and Allen, 1975) and then CP calculated as N x 6.25. Neutral detergent fiber (NDF), acid detergent fiber (ADF), cellulose and permanganate lignin analyses were carried out according to the method of Goering and Van Soest (1970). Hemicellulose was calculated as the difference between NDF and ADF. Determination of cellulose,^linear zylan and branched heterozylan The residues were ground to pass through a lmm sieve opening in a Wiley mill before further analysis. All analyses for each polysaccharide were conducted in triplicate from pooled samples (across cows). To facilitate the extraction of the various polymers any cell contents and lignin remaining in the residues were first removed by treating the materials with neutral detergent solution and then delignifying them using sodium chlorite (Gaillard and Bailey, 1968). Ten grams (10g) of the ground 82  residue was boiled with 300m1 of neutral detergent solution for one hour over a heating block. The neutral detergent residue (NDR) was then dried at 400C for 48h. A 6g sample of the NDR was heated with 170m1. of water containing 1.9g of sodium chlorite and 0.35m1. of glacial acetic acid, in a water bath under a hood for lh. at 70 - 80°C (Wise et al., 1946; Morrison, 1975). The reaction flask was sealed by inverting a 25ml. erlenmeyer flask onto its neck (Wise et al. 1946). At intervals of 15 minutes equal amounts of acetic acid and sodium chlorite were added and the mixture gently stirred. After lh the delignified residue was washed free of chlorite with large amounts of water and then air-dried after washing with acetone. At this stage the material (holocellulose) was ready for polymer extraction.  Figure 4.1. Schematic presentation of the carbohydrate fractionation procedure Rumen sample residue (ground lmm)  [Boil in neutral detergent solution lh.] Neutral detergent residue [Na chlorite, 80°C, lh.] Holocellulose [10% KOH, 16h, room temp.]  Hemicellulose^ (soluble)^  a-cellulose (residue)  [pH=4.5-5, 4°C, 16h] Centrifuge; 17,300g, 30min.  Linear Xylan^ (sediment)  Supernatant [4 vol. 95% ethanol]  ^ [Centrifuge; 17,300g, 30min] Mixed linear^Supernatant and branched heteroxylans ^(discarded) (sediment) a-cellulose Triplicate samples (1g) of the delignified residues were extracted by shaking with potassium hydroxide (10% w/v, 20m1.) under nitrogen at room temperature for 16h in 50m1 centrifuge bottles (Gordon and Gaillard, 1976). The extract was filtered and then centrifuged at 300 x g to remove any particulate matter. The sediment was combined with the extracted residue and then washed free of alkali with water followed by acetone and dried to constant weight. This residue formed the first polymer, a-cellulose. After drying its weight was taken and recorded. Linear zylan - A Linear xylan was determined on the supernatants from the previous step. The supernatant was acidified with 50% acetic acid to pH 4.5-5 to precipitate the linear xylan (Gordon and Gaillard, 1976). The solution was placed in a refrigerator overnight to facilitate this process. The precipitate, creamish in color, was centrifuged off at 17,300 x g for 30 minutes (Jung et al, 1991). After washing this precipitate, it was dried in a freeze drier overnight. The weight was then taken and recorded. Since this precipitate contained a lot of salt from the alkali extraction step it was found necessary to ash it to determine the absolute weight of the linear xylan. Ashing was done at 6000C since at 5000C it was found to leave a residue of uncombusted carbon. Absolute linear xylan was 85  calculated as the difference between the previously recorded value and the ash value. Mixed linear and branched heteroxylans Branched heteroxylan was determined on the supernatants from the previous step by decanting it into 4 volumes of 95% ethanol (Jung et al., 1991; Gordon and Gaillard, 1976). The precipitate, which formed immediately, was collected by centrifugation at 17,300 x g and then dried in a freeze-drier overnight. The weight was recorded. After ashing as in the case of linear xylan, the weight of the absolute branched heteroxylan was calculated as the difference in weight between the original unashed material and the weight of the ash.  4.4^  RESULTS AND DISCUSSION  4.4.1 Chemical composition  The chemical composition of both orchard grass and tall fescue leaf and stem fractions are presented in Tables 4.1 and 4.2 respectively. The leaf to stem ratios of early, medium and late cut orchard grass were 1.71, 1.20 and 1.16 respectively. Those of early and medium cut tall fescue were 2.42 and 3.10 respectively. The quality of orchard grass declined from the early cut to the late cut as indicated by the decrease in CP and the increase in NDF and ADF in both leaves and stems. Leaves showed a trend of higher CP than stems at the three stages of growth. Neutral detergent fiber (NDF), a measure of the cell walls (Van Soest, 1982), rose by 9.35 percentage points within two weeks for stems and by 7.41 percentage points within the same period of time for leaves. Acid detergent fiber (ADF), a measure of lignocellulose and indigestibility rose by 8.33 percentage points between early and late cut stems. With leaves ADF rose by 5.77 percentage points between the two cuts. The results with tall fescue were unexpected. Crude protein values were very low as shown in Table 4.2. While the low CP values would probably indicate that this grass was past heading visual observation and the leaf to stem ratio indicated otherwise. The crude protein content of forages is influenced by the stage of growth and the level of nitrogen 87  fertilizer applied during growth (Jones and Wilson, 1987). Young vegetative growth is high in protein but the protein content rapidly falls as the proportion of leaf decreases. Protein content decreases in both leaves and stems as the plant ages but the decline is more rapid with stems than with leaves (Norton, 1982). Our results for orchard grass are in agreement with these reports. However, those of tall fescue were inconsistent. Crude protein content was determined at U.B.C. and verified at Agriculture Canada Research Station Agassiz. Cellulose: This important structural component of plants rose from 28.32% for early maturity stems to 33.79% for late maturity stems in orchard grass. Leaves had a tendency to be lower cellulose than stems at all stages of maturity. The increase in cellulose content with maturity was smaller with leaves compared with stems. This reflects on the role of cellulose, particularly in stems, of providing strength and rigidity to the whole plant. Tall fescue showed a similar trend to orchard grass but the increase in cellulose was smaller. Both orchard grass and tall fescue leaves showed a trend of lower cellulose than stems. These results are in agreement with past findings regarding changes in cellulose with growth. For example, Bailey (1973) indicated that cellulose content in leaves may rise rapidly from the levels (8 - 10%) of the early stages of growth to much higher levels (21 - 22%) as was the case with ryegrass. 88  Hemicellulose A: Linear^xylan,^referred to^as hemicellulose-A in Tables 4.1 and 4.2, was first defined by Gaillard (1962) as a water insoluble heteroxylan containing uronic acid and a small proportion of arabinose. The levels of this polymer were observed to increase with maturity for both leaves and stems of orchard grass. For stems, there was an increase in this polymer by 3.75 percentage points from early to late maturity. Apart from the early maturity stage, leaves and stems had a similar trend in linear xylan content. On the other hand, tall fescue showed the same trend. Few studies have been conducted in which the quantitative distribution of this polymer in leaves and stems has been determined. However, past research has shown that as a plant matures the ratio of xylose to arabinose and glucose residues increases, indicating an increase in linearity of the hemicellulose (Reid and Wilkie, 1969; Morrison, 1980). Hemicellulose B: Branched heteroxylans or hemicellulose B have been defined as a mixture of small molecular weight xylans which are richer in side chains together with more complex molecules containing galactose, glucose and rhamno se .  (Bailey, 1973). The distribution of these polymers did not seem to differ in both leaves and stems of orchard grass with maturity. Tall fescue showed the same trend as orchard grass. Chemical fractionation of hemicellulose B has shown that it contains both a linear and a highly branched polymer (Gaillard, 1965). Morrison (1980), looked at the change in 89  proportion between the linear and the branched polymers with maturity in both leaves and stems of S26 cocksfoot (Dactylis  qlomerata) and S170 tall fescue (F. arundinacea). In both cases, and for both leaves and stems, ageing resulted in an increase in the linear molecule at the expense of the branched one. The overall effect of this change on hemicellulose B is to make this molecule more linear at later stages of maturity, even though no change may be observed in total extractable hemicellulose B. Data presented here show trends that suggest no changes in this cell wall component, but it is possible that the linear component associated with it may have increased at the expense of the branched polymer with maturity. Lignin content in both leaves and stems of orchard grass and tall fescue were low as shown in Tables 4.1 and 4.2. Other than in the early stage of maturity of orchard grass steins showed a higher content than leaves. Tall fescue showed the same trend.  Table 4.1. Chemical composition of orchard grass leaf and stem fractions (DM). Item (%) CP NDF ADF a-Cell. Hemi-A Hemi-B Total Hemi A+B Lignin Ash  Treatments Early Stems  Early Leaves  8.1  16.2  6.0  14.3  5.2  13.2  58.5  54.8  61.3  56.3  67.8  62.2  34.2  30.2  38.0  31.2  42.5  36.0  28.3  25.7  30.6  26.4  33.8  28.8  8.6  10.4  10.7  9.7  12.4  11.4  11.9  10.3  10.8  10.1  11.6  11.2  20.6  20.7  21.5  19.8  24.0  22.5  3.0  2.9  4.0  10.3  8.8  Medium Stems  8.28  Medium Leaves  3.4 9.7  Late Stems  Late Leaves  5.8  4.6^.  9.9  8.8  Table 4.2 Chemical composition of tall fescue leaf and stem fractions (% DM). Item  CP  Early Stem  Early Leaves  Treatments Medium Stems  Medium Leaves  4.9  9.9  4.5  9.6  NDF (%)  54.6  52.8  55.1  53.1  ADF (%)  33.2  31.7  34.3  31.8  a-Cell.  27.1  26.2  27.9  26.7  Hemi-A  9.8  9.8  9.8  10.4  Hemi-B Total A + B  9.6  10.9  11.5  10.7  19.4  20.7  21.3  21.1  Lignin  3.0  3.0  4.0  3.6  Ash  5.6  6.0  7.5  7.2  4.4.2 RUMEN RATES, EXTENTS AND EFFECTIVE DEGRADABILITIES OF DM AND CELL WALL POLYMERS Degradation of forage DM Results on the extent of DM degradation in the rumen for orchard grass and tall fescue are presented in Tables 4.3 and 4.4 respectively.^Appendix Tables 2, 3, 5, and 6 are a presentation of the ANOVAs of DM for the two forages. ^.The effect of maturity on DM disappearance is well documented (Norton, 1982; Van Soest, 1982). In this study, the disappearance of DM was significantly higher (P < 0.05) with 92  leaves than with stems. Comparisons were made for all the incubation times but statistical comparisons are shown only for 48 and 72h. Significant differences (P < 0.05) were observed between leaves at the early stage and the latter stages of growth in 72h DM disappearance. However medium and late harvested leaves did not differ (P > 0.05) in DM disappearance. A drop of 6.45 percentage points in DM disappearance between early and late cut leaves was observed. Stems were significantly different (P < 0.05) in DM loss at 72h at all the three stages of growth. A drop in DM disappearance at 72h of 11.3 percentage points occurred for stems. Figure 4.2 shows the degradation curve of orchard grass DM at early and late maturity for both leaves and stems. Disappearance of tall fescue leaf and stem fractions did not establish a clear or obvious trend. Leaves were significantly higher (P < 0.05) in DM disappearance (72h) compared to stems at the early maturity stage but not at the second stage. These results are in agreement with previous reports regarding digestibility of leaves and stems. Leaves are usually more digestible than stems (Akin, 1986) but can be equal or only slightly less digestible at earlier stages of growth (Buxton et al., 1985). This is related to differences in cell-wall concentration and chemical composition between leaves and stems (Hornstein et al., 1989). The fact that no significant differences were observed between either leaves or stems at the two harvest stages of tall fescue may indicate 93  that this particular grass does not change in quality within short growth periods.  DM degradability Orchard grass  Figure 4.2  95  Table 4.3 Extent of DM degradation for orchard grass leaf and stem fractions. Treatments  TIME OF INCUBATION (h) 0  12  18  24  48  72  Early Stems  26.62  40.70  51.87 62.84  75.17a  78.83a  Early Leaves  24.82  43.16  56.83 68.83  77.04a  82.67d  Medium Stems  23.14  38.83  44.41 53.55  66.88b  74.58b.  Medium Leaves 20.28  36.33  52.31 67.41  73.73c  78.04a  Late Stems  17.77  34.25  42.15 49.61  63.99d  67.53c  Late Leaves  18.53  40.06  48.97 59.93  68.70b  0.368  1.17  s.e.m.  1.91  1.00  0.80  76.22ab 0.86  a,btc,d Means within the same column (48 and 72h) followed by different letter(s) differ (P < 0.05). Table 4.4 Extent of DX degradation for tall fescue leaf and stem fractions. Treatments  TIME OF INCUBATION (h) 0  12  18  24  48  72  Early Stems  35.36  52.23  57.40  59.95  70.87b  77.21b  Early Leaves  31.27  52.22  61.24  62.19  77.18a  83.70a  Medium Stems  33.57  49.46  56.13  62.39  74.43a  80.69a -  Medium Leaves 31.46  50.63  59.58  66.92  75.15a  82.40a  s.e.m.^11^0.39  1.25  1.17  1.54  1.51  0.94  a,b Means within the same column (48 and 72h) followed by different letter differ (P < 0.05). A noticable difference between early and late harvests was in their fraction 'a', the immediately soluble material. Late cut forage fractions were associated with significantly lower (P < 0.05) 'a' values than early-cut fractions. This could possibly indicate the lower concentration of soluble matter as 96  reflected by an increase in NDF content. It could also probably reflect on the greater resistance to breakage associated with older more lignified tissues which may have led to lower passive loss of dry matter from the bags. The second fraction denoted by constant 'b'^is insoluble but potentially degradable (Orskov, 1991). No significant differences (P > 0.05) were observed between leaves and stems in this fraction at the three stages of growth of orchard grass. In tall fescue (Table 4.7) leaves had significantly higher (P < 0.05) DM 'b' fraction than stems and also the older material was significantly lower (P < 0.05) than younger material in this fraction. No significant differences (P > 0.05) were observed in the rates of disapperance, denoted by the letter 'c', the 'b' fraction between leaves and stems and amongst different maturities for both orchard grass and tall fescue (Tables 4.5 and 4.7 respectively). The extent of degradation occurring in the rumen is determined by both the rate of digesta passage within the rumen and its rate of degradation. Tables 4.6 and 4.8 show the effective DM degradation values of orchard grass and tall fescue respectively, calculated for various rates of passage. Other than for the early maturity stage, leaves and stems of orchard grass differed (P < 0.05) in effective DM degradation. Late cut stems were significantly lower (P < 0.05) in effective degradability than stems of early cut but not the medium one. Early and medium cut leaves did not show any 97  differences (P > 0.05) in effective degradability. However, late cut leaves were significantly lower (P < 0.05) than earlier cuts in effective degradation. Tall fescue leaves and stems did not have any significant differences (P > 0.05) in effective degradability either at the earlier or latter maturity stage. This can be explained by the fact that the characteristics of the two cuts were not sufficiently different in maturity.  Table 4.5 Orchard grass DM degradation constants Treatments  DM DEGRADATION CONSTANTS a^(%)  b^(%)  c  Early Stems  24.97c  60.13de  0.035c  Early Leaves  23.30f  63.30de  0.043h  Medium Stems  21.87f  62.33de  0.028c  Medium Leaves  18.40k  64.27d  0.044h  Late Stems  16.97k  56.73e  0.034c  Late Leaves  17.90k  62.33de  0.043h  s.e.m.  0.49  2.30  0.005  a= % solubility; b= potentially degradable fraction [%] and c= rate of degradation of the 'b' fraction (/h]. ccd,e,f,k Means within the same column followed by different letter(s) differ (P < 0.05). c,h Means within the last column followed by different letter differ (P < 0.1)  Table 4.6 Effective DM degradation values (%) for orchard grass Treatments  Mean outflow rate variables k=0.02  k=0.04  k=0.06  k=0.08  62.97db  52.83ab  46.90bc  43.10ab  Early Leaves  66.07b  55.73a  49.43b  45.20a  Medium Stems  58.13ad  47.47bc  41.63cd  37.97bd  Medium Leaves  62.03db  51.53ab  45.17bc  40.80ab  Late Stems  52.60a  42.93c  37.43d  33.83d  Late Leaves  59.16cd  49.00bc  42.93cd  38.83bd  Early^Stems  I^s.e.m.  1.84  1.98  1.94  1.86^I  k= rumen digesta flow rate Uh]. acb,c,d Means within the same column followed by different letter(s) differ (P < 0.05).  Table 4.7 Tall fescue DM degradation constants Treatments  DM Degradation Constants a^(%)  b^(%)  Early Stems  35.80a  53.73a  0.036a^*  Early Leaves  30.97b  56.50a  0.044a  Medium Stems  33.33c  44.03b  0.032a  Medium Leaves  31.00b  52.77ab  0.047a  s.e.m.  0.71  2.79  a ( b f c Means within the same column followed by different letter(s) differ (P < 0.05).  c^(%)  0.008  Table 4.8 Effective DM degradation values (%)^for tall fescue grass. Treatments  Mean outflow rate variables k= 0.02*  k= 0.04 *  Early Stems  63.97  56.53  52.23  49.40  Early Leaves  68.16  59.10  53.67  50.00  Medium Stems  65.87  56.67  51.67  48.37  Medium Leaves  67.03  58.56  53.33  49.80  1.57  2.02  2.08  2.04  I^  s.e.m.  k= 0.06^*  k= 0.08*  .  *means within these columns not different (P > 0.05) Degradation of forage cell wall polymers Cellulose Degradation of cellulose was determined by isolating the fraction that survived rumen incubation. Table 4.9 shows the extent of cellulose disappearance from the rumen at various times. Degradation of this polymer was high in orchard grass at all stages of growth at 72h ranging from 71.5% in late cut stems to 84.7% in early stems. Based on the actual degradation values and the standard deviations (72h), it would seem that both leaves and stems at the early stage of harvest of orchard grass were not different. Similarly, it would appear that more cellulose disappeared from leaves than from stems at the late maturity stage of growth. Akin (1986) observed that plant tissues differ in rates of disappearance, suggesting that tissue morphology also may affect accessibility and rate of digestion. The surface area that is accessible to enzymes was positively related to rate of digestion when surface area was 10 0  increased by swelling of cotton linters (Stone et al., 1969). There was a general trend of cellulose degradability declining in both leaves and stems as these morphological parts matured. Results of rate of degradation presented in Table 4.10 show that cellulose had a trend of being more rapidly degraded in leaves than in stems at all stages of growth other than in the early one in orchard grass. However, maturity did not seem to result in much change in rate of cellulose degradation in stems. Younger leaves seemed more rapidly degraded relative to older, late harvested leaves. The potentially degradable fraction, 'b', declined from 73.37% at early maturity to 65.10% at late maturity for stems. Leaves showed a similar trend declining from 70.63% to 62.70%. Based on the standard deviations it would appear that both leaves and stems of early and medium harvests possibly were not different in their potentially degradable fraction. Table 4.11 shows the results for the extent of degradation of tall fescue cellulose. No clear trend was observed in the degradation of this polymer in both leaves and stems. However, based on the 72h degradation values and their standard deviations it would appear that earlier cut stems were less degradable than the corresponding leaves (79.5± 0.04 vs 84.4± 0.04). Rate of cellulose degradation did not seem to differ in stems at the two growth stages (Table 4.12). However, leaf cellulose seemed to be degraded faster in the medium stage of growth than in the early one. 101  Digestion of cellulose is incomplete but this depends on the type of tissues and cells being digested (Chesson and Forsberg, 1988). Darcy and Belyea (1980) reported that cellulose of late cut orchard grass was less digested than that of early cut. Mowat et al. (1965), also observed with  in  vitro studies that leaves of earlier cut orchard grass were less digested than stems. The results reported here showed a trend consistent with their findings.  Table 4.9 Extent of degradation of cellulose in orchard grass. (Mean ±SD; n=3) Treatments  TIME OF INCUBATION (h) 0  12  18  24  48  72  Early Stems  26.7 37.1 50.1 62.9 77.6 + 0.11 ± 0.18 ± 0.16 ± 0.18 ± 0.18  84.7 ± 0.10  Early Leaves  24.9 52.8 54.5 67.8 76.9 ± 0.14 ± 0.56 ± 1.44 ± 0.42 ± 0.31  84.3 ±^0.38 .  Medium Stems  22.5 37.9 43.3 53.6 73.4 ± 0.47 ± 0.62 ± 2.23 ± 0.77 ± 0.45  77.8 ± 0.06  Medium Leaves 20.3 33.8 43.1 66.2 74.3 ± 0.02 ± 0.50 ± 0.32 ± 0.09 ± 0.25  78.9 ± 0.17  Late Stems  17.8 31.7 41.3 51.2 70.4 ± 0.41 ± 0.86 ± 0.26 ± 0.03 ± 3.76  71.5 ± 0.31  Late Leaves  18.5 40.9 48.8 60.4 67.7 ± 0.09 ± 0.46 ± 0.36 ± 0.40 ± 0.24  79.1 ± 0.28  Table 4.10 Orchard grass cellulose degradation constants (Mean ±SD; n=3) CELLULOSE DEGRADATION CONSTANTS  Treatments  b (%)  a (%)  c  Early Stems  24.10± 0.100 73.37± 1.041  0.025± 0.0006  Early Leaves  25.17± 0.058 70.63± 0.651  0.044± 0.0021  Medium Stems  21.37± 0.252 71.03± 1.950  0.024± 0.0010  Medium Leaves  17.43± 0.116 69.97± 0.611  0.035± 0.0032  Late Stems  16.23± 0.451 65.10± 0.854  0.029± 0.0006  Late Leaves  18.43± 0.058 62.70± 0.854  0.039± 0.0010  Table 4.11 Extent of degradation of cellulose in tall fescue grass (Mean ±SD; n=3). Treatments  TIME OF INCUBATION (h) 0  12  18  24  48  72  Early Stems  35.2 52.4 57.3 61.3 73.0 + 0.21 ± 0.17 ± 0.79 ± 0.28 ± 0.14  79.5 ± 0.04  Early Leaves  59.9 76.6 31.5 48.2 59.1 ± 0.44 ± 0.22 ± 0.68 ± 0.19 ± 0.75  84.4 ± 0.04  Medium Stems  33.6 52.0 58.4 66.2 78.1 ± 0.10 ± 0.14 ± 0.11 ± 0.13 ± 0.71  84.8 ± 0.08  75.3 Medium Leaves 31.5 48.8 57.0 64.9 ± 0.06 ± 0.40 ± 1.10 ± 0.42 ± 0.07  83.6 ± 0.12  Table 4.12 Tall fescue cellulose degradation constants (Mean ±SD; n=3) Treatments  CELLULOSE DEGRADATION CONSTANTS a (%)  b (%)  Early Stems  35.77± 0.058  56.00± 0.600  0.033± 0.0012  Early Leaves  31.37± 0.058  60.00± 0.458  0.030± 0.0006  Medium Stems  33.43± 0.153  47.60± 0.265  0.034± 0.0000  Medium Leaves  31.20± 0.100  56.77± 0.404  0.034± 0.0015  103  c^.  Cellulose degradability is thought to be influenced by several factors. On the one hand extensive hydrogen bonding in cellulose microfibrils is thought to influence the rate at which this polymer is degraded while on the other hand lignin shielding is thought to impact on the extent of its degradation (Hatfield,1989). Morrison (1979) indicated that since the structure of cellulose does not change significantly during the growing period, then cellulose of a young plant ought to be as digestible as that of an older plant. Lignin was found to increase with maturity in the present study and more so with stems than with leaves. The presence of a high proportion of unlignified mesophyll cells in leaves (Akin, 1986) would probably explain the higher degradability of leaf cellulose than that of stems. The fact that tall fescue did not seem to show any differences in cellulose digestibility at the two stages of growth would indicate that this grass was harvested at a fairly young stage and the cutting interval too short for any significant changes to be detected. Hemicellulose A Table 4.13 shows the results of the extent of linear xylan (hemicellulose A) degradation at various incubation times. These results suggest that there was a rapid decline in degradability of this polymer, dropping from 75.94 to 57.02% in two weeks for orchard grass stems. In leaves however, no such drastic changes were observed. Degradability of this 104  polymer remained high (k$ 70%). Based on the standard deviations these results suggest that hemicellulose A may not have been differently degraded in both leaves and stems of orchard grass at the earlier stage of growth. The rate of degradation of this polymer tended to increase for stems with maturation (Table 4.14). The same situation was observed with leaves. While this seems to contradict the observation on the overall extent of degradation, it has been reported that both rate and extent of degradation of a cell wall polymer are influenced by different factors (Hatfield, 1989). Lignin influences the extent of degradation while factors like accessibility of the cell wall by hydrolytic enzymes, intermolecular bonding will affect the rate but not necessarily the extent of polymer degradation. These results would tend to support the hypothesis that lignin does not influence the rate of hydrolysis of cell wall polymers but rather sets an upper limit to the extent to which these polymers can be degraded. There is good evidence that, in grasses, at least part and maybe all of the lignin is covalently bound to hemicellulose (Morrison, 1974). Gordon and Gaillard (1976) showed that in wheatstraw about 78% of the lignin in this material was associated with linear xylan (hemicellulose A) while only 8% was associated with the more branched fraction (hemicellulose B). Increase in linearity of hemicellulose is believed to have positive effects on hemicellulose degradability (Brice and Morrison, 1982). This 105  is because of an increase in the proportion of contiguous unsubstituted xylose residues. Hydrolysis of hemicelluloses by rumen hemicellulases is reported to take place at the bond between two unsubstituted xylose residues. An increase in lignin content however counteracts the positive effects of hemicellulose linearization that accompany maturation (Morrison, 1979). Tall fescue showed the same trend as orchard grass in the 72h extent of linear xylan degradation (Table 4.12). Like cellulose in earlier harvested stems, hemicellulose A tended to be less degradable than the corresponding polymer in leaves at the same stage of growth. The rate of degradation of the potentially degradable fraction, 'b', showed an increasing trend in stems with maturation but not in leaves (Table 4.7). The potentially degradable fraction, b, dropped from 60.30 to 55.07% in stems from early to medium maturity.  Table 4.13 Extent of degradation of hemicellulose A in orchard grass (Mean ±SD; n=3). Treatments  TIME OF INCUBATION (h) 0  12  18  24  48  72  Early Stems  25.1 34.2 42.5 50.9 64.9 ± 0.11 ± 1.86 ± 1.10 ± 1.79 ± 0.47  75.9^. ± 0.57  Early Leaves  23.7 37.1 42.5 55.2 66.1 ± 0.24 ± 0.70 ± 4.03 ± 0.87 ± 0.91  75.7 ± 1.42  Medium Stems  21.5 31.7 32.7 37.1 56.9 ± 0.07 ± 1.41 ± 0.58 ± 0.01 ± 1.87  63.5 ± 0.61  Medium Leaves 15.9 23.8 30.4 53.1 65.6 ± 0.11 ± 1.34 ± 0.71 ± 0.50 ± 3.08  68.5 ± 0.73  Late Stems  32.4 38.4 53.1 16.4 23.4 ± 0.39 ± 5.80 ± 1.68 ± 1.05 ± 1.52  57.0 ± 0.51  Late Leaves  48.1 18.5 39.8 55.1 63.1 ± 0.16 ± 3.12 ± 1.72 ± 1.59 ±0.73  72.4 ± 0.63  Table 4.14 Orchard grass hemicellulose A degradation constants (Mean ±SD; n=3) Treatments  HEMICELLULOSE A DEGRADATION CONSTANTS a (%)  b (%)  c  Early Stems  24.10± 0.656 72.60± 1.212  0.014± 0.0006  Early Leaves  23.83± 0.322 66.00± 1.253  0.022± 0.0023  Medium Stems  21.47± 0.231 62.87± 1.563  0.018± 0.0012  Medium Leaves  15.80± 0.458 71.87± 2.702  0.021± 0.0015  Late Stems  16.53± 0.058 54.07± 4.055  0.021± 0.0025  Late Leaves  18.63± 0.666 54.67± 2.401  0.043± 0.0031  Table 4.15 Extent of degradation of hemicellulose A in tall fescue grass (Mean ±SD; n=3). Treatments  TIME OF INCUBATION (h) 0  12  18  24  48  72  Early Stems  31.1 43.9 47.4 48.4 62.6 ± 0.15 ± 0.91 ± 2.03 ± 1.90 ± 0.18  71.0 ± 026  Early Leaves  33.5 36.1 47.4 45.0 65.3 ± 0.07 ± 2.28 ± 0.43 ± 0.56 ± 1.25  78.2 ± 0.75  Medium Stems  31.5 48.7 51.4 56.6 69.6 ± 0.57 ± 1.54 ± 0.60 ± 1.08 ± 0.94  78.3 ± 0.94  Medium Leaves 31.6 42.0 47.3 57.1 66.1 ± 0.04 ± 0.67 ± 1.65 ± 1.43 ± 0.68  79.0 ± 0.54  Table 4.16 Tall fescue hemicellulose A degradation constants (Mean ±SD; n=3) Treatments  HEMICELLULOSE A DEGRADATION CONSTANTS a (%)  b (%)  c  Early Stems  35.10± 0.781  60.30± 1.229  0.012± 0.0006  Early Leaves  31.80± 0.872  65.73± 0.551  0.015± 0.0006  Medium Stems  34.03± 0.231  55.06± 0.987  0.022± 0.0000  Medium Leaves  31.17± 0.416  63.23± 1.050  0.016± 0.0023 Homicellulose B The degradability of the more branched hemicellulose (hemicellulose B) at various incubation times is presented in Tables 4.17 and 4.19 for orchard grass and tall fescue respectively. These values suggest that for orchard grass, hemicellulose B was likely to be more degradable (72h) in stems than in leaves at the earlier stage of growth (88.3± 0.05 vs 86.1± 0.67) but this scenario changed at late maturity 108  when more of this polymer tended to be more degraded in leaves than in stems (81.0± 0.67 vs 73.8± 1.71). Irrespective of the morphological part, degradability of this polymer declined with maturity. It has been shown that as grass matures, the linear component of hemicellulose B increases (Morrison, 1980). Since this component is more degradable than the branched one it would be expected that hemicellulose B would become more degradable as grass matures. However, it has been shown that the linear component is the one closely associated with lignin. An increase in this cell wall component with maturity would therefore lead to a decline in the degradability of hemicellulose B. Based on the standard deviations leaf hemicellulose B in tall fescue tended to be more degradable than that of stems at the earlier growth stage, but did not at the later stage of maturity. This could be related to an increase in the more degradable linear component of hemicellulose B and only a limited increase in lignin. Figure 4.3 shows the rumen degradation curve for hemicellulose B for both the leaf and stem fractions of orchard grass. The rate of degradation of the potentially degradable fraction of this polymer did not seem to differ with maturation for stems but showed a tendency to increase in leaves (Tables 4.18 and 4.20). Based on the standard deviations it would appear that fraction 'b' was higher in early than in late cut stems in orchard grass but not in 109  leaves. Similarly leaves appear not to have differed from stems in each of the growth stages. Other than for the early cut leaves, the degradation rate of this polymer seemed the same for all the plant parts of tall fescue.  Hemicellulose B degradability Orchard grass  Figure 4.3  111  Table 4.17 Extent of degradation of hemicellulose B in orchard grass (Mean ±8D; n=3). Treatments  TIME OF INCUBATION (h) 0  12  18  24  48  72  Early Stems  71.4 83.3 50.0 59.4 25.8 + 0.28 ± 1.71 ± 2.81 ± 1.92 ± 0.55  88.3 ± 0.05  Early Leaves  80.0 58.3 71.1 23.4 44.4 + 0.40 ± 1.20 ± 1.01 ± 1.22 ± 0.59  86.1 ± 0.67 .  Medium Stems  72.9 45.9 55.5 22.2 40.8 ± 0.14 ± 0.18 ± 0.95 ± 1.71 ± 0.86  78.0 ± 0.56  75.5 44.8 67.0 36.9 Medium Leaves 17.9 ± 0.06 ± 2.59 ± 2.18 ± 2.75 ± 0.19  78.6 ± 0.79  Late Stems  53.3 58.8 72.8 16.8 39.0 ± 0.07 ± 2.47 ± 1.34 ± 0.49 ± 2.39  73.8 ± 1.71  Late Leaves  72.5 54.9 63.9 17.8 44.4 ± 0.07 ± 0.95 ± 1.27 ± 0.54 ± 0.12  81.0 ± 0.67  Table 4.18 Orchard grass hemicellulose B degradation constants (Mean ±SD; n=3) Treatments  HEMICELLULOSE B DEGRADATION CONSTANTS a (%)  b (%)  c  Early Stems  25.87± 0.404 66.27± 1.002  0.042± 0.0020  Early Leaves  23.37± 0.404 66.50± 0.265  0.042± 0.0015  Medium Stems  22.10± 0.200 65.97± 1.002  0.028± 0.0017  Medium Leaves  17.90± 0.700 67.73± 1.716  0.037± 0.0012  Late Stems  16.80± 0.200 60.73± 1.550  0.047± 0.0031  Late Leaves^_17.75± 0.570 63.55± 1.345  0.048± 0.0021  Table 4.19 Extent of degradation of hamicellulose B in tall fescue grass (Mean ±SD; n=3). Treatments  TIME OF INCUBATION (h) 0  12  18  24  48  72  Early Stems  25.5 53.0 61.7 62.3 73.7 ± 0.58 ± 3.04 ± 3.72 ± 1.24 ± 0.37  80.1 ± 0.84  Early Leaves  23.1 56.0 59.5 67.8 76.8 ± 0.54 ± 0.89 ± 1.92 ± 1.39 ± 0.53  85.5 ± 0.14  Medium Stems  22.1 57.5 61.5 66.0 78.1 ± 0.12 ± 1.22 ± 0.33 ± 1.05 ± 0.45  84.7 ± 0.53  Medium Leaves 17.6 55.2 61.7 68.2 78.1 ± 0.53 ± 1.55 ± 1.39 ± 0.79 ± 0.79  86.1 ± 1.08  Table 4.20 Tall fescue hemicellulose B degradation constants (Mean±SD; n=3) Treatments  HEMICELLULOSE B DEGRADATION CONSTANTS a (%)  b (%)  c  Early Stems  35.77± 0.058  56.00± 0.600  0.033± 0.0012  Early Leaves  31.27± 0.153  59.03± 2.043  0.030± 0.0000  Medium Stems  33.43± 0.153  47.60± 0.265  0.034± 0.0006  Medium Leaves  31.20± 0.100  56.77± 0.404  0.034± 0.0015  As indicated earlier hemicellulose B is composed of a linear and a branched component both of which determine the total degradability of hemicellulose B. Bailey and Gaillard (1965) isolated both linear hemicellulose A xylan and hemicellulose B and studied the action of rumen microbial enzymes on these polymers. They reported that the branched B polymers ( isolated from hemicellulose B by complexing with iodine), from both grass and legume were more resistant to hydrolysis than the linear polymers. Of the linear polymers 113  (including linear A), the B fraction was hydrolysed most rapidly which may explain why hemicellulose B was more digestible than hemicellulose A in red clover (Bailey and Macrae, 1970). The resistance to hydrolysis of the branched B polymers by the rumen microorganisms was suggested to be due, most probably , to the high content of uronic acid in this polymer. In this study hemicellulose B had a trend of higher degradability after 72 hours of rumen incubation than cellulose or hemicellulose A. It appeared that hemicellulose A was the least degradable of the polymers studied. The observations of Gordon and Gaillard (1976) that only a small proportion of the total lignin is associated with hemicellulose B may explain the high extent of degradation of hemicellulose B observed in these experiments. Arabinofuranosidase activity has been found in cell free extracts of the rumen (Williams and Strachan, 1984) and is capable of removing the arabinose residues from the xylans. The activity of such glycosidases renders highly substituted arabinoxylans more susceptible to xylanase degradation. However, the three dimensional organization of the matrix may play a role in the extent and/or rate of such side chain removal (Hatfield, 1989). Greve et al. (1984) observed that an arabinofuranosidase from Ruminococcus albus extensively removed arabinose residues from isolated hemicellulosic materials (47% removal) but had a more limited activity against isolated cell walls (1.2%). The arabinofuranosidase 114  has a molecular weight of approximately 300,000. Such a large enzyme may have difficulty penetrating the matrix to remove arabinose side chains. This limitation may in turn limit the activity of xylanases that can penetrate to the xylan substrates. It is possible then that the high degradability of hemicellulose B could be related to a lower proportion of the highly branched fraction relative to the linear one and therefore a higher activity of the xylanases. Indeed, Morrison (1980) showed the ratio of the linear to the branched component to be about 65:35 in leaves and 83:17 for stems. This ratio was dependent on maturity, being higher with older material. The decline in degradability of hemicellulose B with age could be related to the formation of hydrogen bonds between the linear xylans and cellulose (Hatfield, 1989) or lignin (Buxton and Brasche, 1991). The rate of digestion of fiber components is a constraint on total fiber digestion and factors altering this rate affect fiber digestibility (Allen and Mertens, 1988). Similarly, digestibility of fiber decreases as the rate of rumen turnover increases. Figures 4.4 to 4.9 present the results for the effective degradabilities of the various cell wall polymers at a rumen passage rate of 0.02/hr. Data for these bar charts are presented in appendix Tables 10 to 15. In first-order systems, retention times are reciprocals of fractional rates of passage. In this study relative comparisons were made only for a 0.02/hr. passage rate which corresponds to a rumen mean 115  retention time of 50h. The results obtained indicate that the effective degradability of the polymers studied was far from complete ranging from 44.10% (hemicellulose A) to 70% (hemicellulose B). Further the results indicate that irrespective of the forage material, stage of growth or plant morphological part, hemicellulose B was the most effectively degradable cell wall polymer. This is likely to be associated with the high rate of digestion of this polymer relative to the others. Bailey et al. (1976) reported in their in vivo work using grass hemicelluloses, that all grass hemicellulose fractions were present in the feaces with a slightly lower proportion of hemicellulose B relative to other hemicellulose. Compared to feed fractions arabinose was lower in faecal hemicellulose B but not in the other fractions. Their results were comparable to those of Ford (1973) who observed preferential removal of hemicellulose arabinose, glucose and galactose relative to xylose with a proportional increase in hemicellulose A xylan in the faeces compared to the feed. Maturity tended to have a lowering effect on the effective degradability of all the polymers studied in orchard grass fractions. No clear pattern was observed with the tall fescue. As a plant matures and the hemicellulose component increases in linearity there is a tendency for the linear xylans to form hydrogen bonds with cellulose (Morrison, 1979; Hatfield, 1989). The extent of this interaction might prevent the binding of cellulolytic microorganisms or impede the 116  hydrolytic activity of a microorganism's cellulolytic complex (Hatfield, 1989), thus lowering the extent of degradation of these polymers.  Fig. 4.4^Cellulose degradability Orchard grass  Fig. 4.5^Linear  xylan degradability Orchard grass  117  Fig. 4.6 ^Branched  Orchard grass  1 1  20^29^11 cutting date (May/92)  Fig.^4.7  20  Cellulose degradability Tall fescue grass  ^24^20  ^  Cutting date (May/92) stems  29  Leaves  E".::1 stems  ^20  ^=]  xylan degradability  ^  118  MN  Leaves  29  Fig. 4.8^Linear  xylan degradability  Tall fescue grass  Fig. 4.9  Branched xylan degradability Tall fescue grass  119  4.4.3 SUMMARY AND CONCLUSIONS Structural carbohydrates of forages are not completely digestible by ruminants. The nature of these limitations are poorly understood. In this study, an attempt was made to understand the effect of maturity on the major cell wall carbohydrates of leaf and stem fractions of two temperate grasses, orchard grass and tall fescue. Further to this, an attempt was made to understand the factors that may lead to their reduced degradability. The polymers studied were cellulose, a linear molecule of hemicellulose commonly referred to as hemicellulose A and another fraction that is associated with high levels of arabinose and therefore highly branched, simply referred to as hemicellulose B. This study showed that orchard grass declines rapidly in quality with maturity within short growth intervals. Tall fescue however did not show this trend. Cellulose in both leaves and stems increased with maturation. Similar observations were made with Hemicellulose A, which was also observed to be equally distributed between leaves and stem's. Hemicellulose B was equally distributed betweem leaves and stems, but maturation did not seem to influence its level. Hemicellulose B was the most effectively degradable polymer among those studied. This was related to its high rate of degradation and possibly to the presence of a highly degradable linear xylan associated with it. Maturity had a significant lowering effect on the degradability of all the 120  polymers. Hemicellulose A was the least effectively degraded of the polymers studied. Leaf polysaccharides were more degradable than those of stems other than in the case of the earliest stage of growth when it was observed that some stem polysaccharides were more degradable. Tall fescue did not show a good trend in polymer breakdown. However, irrespective of the stage of growth, or plant part, hemicellulose B was the more degradable fraction compared with the others. From this study it is concluded that the rapid decline in quality of orchard grass could be related to the presence of recalcitrant cell wall polymers. Cellulose and hemicellulose A were observed to increase with maturity and were both incompletely degraded. Hemicellulose B was highly degradable at the earlier growth stages but declined with maturity. The possibility that linear xylans could have interacted through hydrogen bonding with cellulose and thus rendered the respective hydrolases ineffective on these polymers cannot be ruled out. In the present study lignin may not have been the major player in limiting utilization of these polymers since its level was fairly low. However, it is now well established that it is the chemical composition of lignin rather than its quantity that is important in inhibiting digestibility. This study tends to support the hypothesis that the amount of linear xylan, relative to the branched xylan, may be an important factor influencing both the rate and extent of cell wall digestion. Presumably, linear xylans have a greater 121  propensity to form physical and/or chemical associations with other cell wall polymers, especially lignin, that would decrease cell wall digestibility. This study showed that polymers associated with leaves were more degradable than those of stems. It is concluded from this study that a breeding program aimed at reducing some of the recalcitrant polymers, or alternatively reducing the lignin level may be beneficial to animal production. Similarly a breeding program aimed at increasing the leafiness of the forages could enhance their utilization by livestock. In the short term a management practice directed at harvesting the forages at an early stage to ensure high leafiness and adequate DM may also be utilized.  BIBLIOGRAPHY Abe, M. and Kumeno, F. 1973. In vitro simulations of the rumen fermentation: apparatus and effects of dilution rate and continuous dialysis on fermentation and protozoal population. J. Anim. Sci. 36: 941-947. Akin, D.E. and Burdick, D. 1975. Percentage of tissue types in tropical and temperate.grass leaf blades and degradation of tissues by rumen microorganisms. Crop Sci. 15: 661-668. Akin, D.E. 1979. Microscopic evaluation of forage digestion by rumen microorganisms- A Review. J. Anim. Sci. 48:701-710. Akin, D.E. 1982.Microbial breakdown of feed in the digestive tract. In: Nutritional Limits to Animal Production from Pastures, Ed. Hacker, J.B. Commonwealth Agricultural Bureau, pp. 210-223. Akin, D.E. 1986. Chemical and biological structure in plants as related to microbial degradation of forage cell walls. In: control of digestion and metabolism in ruminants, Ed. Milligan, L.P.; Grovum, W.L. and Dobson, A. Prentice-Hall, Engelwood Cliffs, New Jersey, pp 139-157. Akin, D.E. and Chesson, A. 1989. Lignification as the major factor limiting forage feeding value especially in warm conditions. Int. Grassi. Congr. 16: 1753-1760. Akin, D.E. 1989. Histological and physical factors affecting digestibility of forages. Agron. J. 81:17-25. Akin, D.E.; Hartley, R.D.; Rigsbay, L.L. and Morrison III, W.H. 1992. Phenolic acids released from bermudagrass (Cynodon dactylon) by sequential sodium hydroxide treatment in relation to biodegradation of cell types. J. Sci. Food Agric. 58: 207214. Allen, M.S. and Mertens, D.R. 1988. Evaluating constraints on fiber digestion by rumen microbes. J. Nutr. 118: 261-270. Bailey,^R.W.,^Clarke, R.T.J. and Wright,^D.E.^1962. Carbohydrases of the rumen ciliate Epidinium ecaudatum (Crawley): action on plant hemicellulose. Biochem. J. 83: 517523. Bailey, R.W. and Gaillard, B.D.E. 1965. Carbohydrases of the rumen ciliate Epidinium ecaudatum (Crawley):Hydrolysis of plant hemicellulose fractions and B-linked glucose polymers. Biochem. J. 95: 758-766. Bailey, R.W. and Macrae, J.C. 1970. The hydrolysis by rumen and caecal microbial enzymes of hemicellulose in plant and 123  digesta particles. J. Agric. Sci. Camb. 75: 321-326. Bailey, R.W. 1973. Structural carbohydrates. In: Chemistry and biochemistry of herbage, Ed. Butler, G.W. and Bailey, R.W. Academic Press, London and New York, Vol. 1, pp 157-211. Bailey, R.W., Monro, J.A., Pickmere, S.E. and Chesson, A. 1976. Herbage hemicellulose and its digestion by the ruminant. Landbouwhogeschool Wageningen. Miscellaneous papers #12, pp. 1-15. Barnes, R.F. 1967. Collaborative In vitro rumen fermentation studies on forage substrates. J. Anim. Sci. 26:1170-1178. Barnes, R.F. 1973. Laboratory methods of evaluating feeding value of herbage. In: Chemistry and biochemistry of herbage, Ed. Butler, G.W. and Bailey, R.W. Academic Press, London and New York, Vol. 3, pp. 179-214. Beauchemin, K.A. 1991. Physiological role of fiber in ruminant diets: A rationale for effective feeding of forage. In: Proceedings of the Pacific Northwest Nutrition Conference, Ed.Henschel, K.K. pp 67-78 Bjorkmann, 0.; Boynton, J.; Berry, J. 1976. Comparison of the heat stability of photosynthesis, chloroplast membrane reactions, photosynthetic enzymes, and soluble protein in leaves of heat adapted and cold adpted C4 plant species. Carnegie Institution of Washington Yearbook 75: 400-407. Black, J.L.; Faichney, G.J. and Sinclair, R.E. 1982. Role of computer simulation in overcoming limitations to animal production from pastures. In:Nutritional limits to animal production from patures, Ed. Hacker, J.B. Commonwealth Agricultural Bureaux, pp 473-493. Brice, R.E. and Morrison, I.M. 1982. The degradation of isolated hemicelluloses and lignin-hemicellulose complexes by cell-free, rumen hemicellulases. Carbohydrate Res. 101:93-100. Brown, R.H. 1978. A difference in N use efficiency in C3 and C4 plants and its implications in adaptation and evolution. Crop Sci. 18: 93-98. Buttery, P.J. and Lewis, D. 1982. Nitrogen metabolism in the rumen. In: Forage protein in ruminant animal production. Occassional publication of British Society of Animal Production no. 6 Edinburgh, U.K.:BASP pp.1-11. Buxton, D.R., Hornstein, J.S., Wedin, W.F. and Marten, G.C. 1985. Forage quality in stratified canopies of alfalfa, birdsfoot trefoil, and red clover. Crop Sci. 26: 180-184. Buxton, D.R. and Rusell, J.R. 1988. Lignin constituents and 124  cell wall digestibility of grass and legume stems. Crop Sci. 28: 553-558. Buxton,^D.R.^and Brasche, M.R. 1991.^Digestibility of structural carbohydrates in cool-season grass and legume forages. Crop Sci. 31: 1338-1345. Campling, R.C. and Balch, C.0 1961. Br. J. Nutr. 15: 531. Campling, R.C. 1969. Physical regulation of voluntary intake. In "Physiology of digestion and metabolism in the ruminant" Ed. Phillipson, A. T. Oriel Press Ltd. pp. 226-234. Cheng, K.J.; Stewart, C.S.; Dinsdale, D. and Costerton, J.W. 1983/84. Electron microscopy of bacteria involved in the digestion of plant cell walls. Anim. Feed Sci. and Technol. 10:93-120. Cheng, K.J.; Forsberg, C.W.; Minato, H. and Costerton, J.W. 1991. Microbial ecology of feed degradation within the rumen. In: Physiological Aspects of Digestion and Metabolism in Ruminants; Proceedings of the Seventh International Symposium on Ruminant Physiology. Academic Press, Inc. pp. 595-624. Chesson,^A.^and Forsberg,^G.W.^1989.^Polysaccharide degradation by rumen micro-organisms. In " The Rumen Microbial Ecosystem" Ed Hobson, P.N. Elsevier Science Publishers, New York, N.Y. pp. 251-284. Church,^D.C.^1988. The classification and importance of ruminant animals. In: The Ruminant Animal-^Digestive physiology and nutrition. Ed. Church, D.C. pp. 1-13. Coen,^J.A.^and Dehority, B.A.^1970.^Degradation and utilization of hemicellulose from intact forages by pure cultures of rumen bacteria. Appl. Microbiol. 20: 362-368. Coleman, R.L. and Lazenby, A. 1970. Factors affecting the response of tropical and temperate grasses to fertilizer nitrogen. Proceedings of the 11th Int. Grassl. Congr. Surfers Paradise, Australia, pp. 392-397. Coleman, G.S. 1980. Rumen Ciliate Protozoa. Advan. Parasitol. 18: 121-173. Conchie, J.; Hay, A.J. and Lomax, J.A. 1988. Soluble lignincarbohydrate complexes from sheep rumen fluid their composition and structural features. Carbohydr. Res. 177: 127151. Czerkawski, J.W. and Brekenridge, G., 1979. Experiments with the long-term rumen simulation technique (Rusitec); use of soluble food and an inert solid matrix. Br. J. Nutr. 42:229125  245. Darcy, B.K. and Belyea, R.L. 1980. Effect of delignification upon in vitro digestion of forage cellulose. J. Anim. Sci. 51:798-803. Dehority,^B.A.;^Johnson, R.R. and Conrad,^H.R.^1962. Digestibility of forage hemicellulose and pectin by rumen bacteria in vitro and the effect of lignification thereon. J. Dairy Sci. 45: 508-512. Dehority, B.A. 1969. Pectin-fermenting bacteria isolated from the bovine rumen. J. Bacteriol. 99:189-186. Dehority, B.A. 1991. Effects of microbial synergism on fiber digestion in the rumen. Proc. Nutr. Soc. 50:149-159. Dekker, R.F.H. 1989. Biodegradation of the hetero-1,4-linked xylans. In: Plant cell wall polymers, biogenesis and biodegradation, Ed. Lewis, N.G. and Paice, M.G. American Chemical Society, pp. 619-629. DelCurto, T. 1991. Influence of protein supplements on the utilization of low-quality roughages. In: Proceeedings of the Pacific Northwest Animal Nutrition Conference, Ed. Henschel, K.K. pp 225-239. Delmer, D.P. 1987. Cellulose biosynthesis. Ann. Rev. Plant. Physiol. 38:259. Dermaquilly, C., Boissau, J.M. and Cuylle, G. 1965. Proc. IX Internat. Grassi. Congr., Sao Paulo, Brazil, 1: 877. Demarquilly, C.and Jarrige, R. 1974. The comparative nutritive value of grasses and legumes. Vaextondling 28: 33-41. Egan, A.R. 1965. Nutritional status and intake regulation in sheep. 2. The influence of sustained duodenal infusions of casein or urea upon voluntary intake of low protein roughagas by sheep. Austr. J. of Agric. Res. 16:451-462. Ellis, W.C., Wylie, M.J. and Matis, J.H. 1988. Dietarydigestive interactions determining the feeding value of forages and roughages. In: Feed Science, Ed. E.R. Orskov. pp. 177-229. Eureka: The solver. 1987. Borland International, Inc. Scotts valley California. Fahey, G.C. Jr. and Garleb, K.A. 1991. Forage fiber chemistry and utilization. In: Proceedings of the Pacific Northwest Nutrition Conference, Ed. Henschel, K.K. pp. 51-65. 126  Figroid, M., Hale, W.H. and Theurer, B. 1972. An evaluation of the nylon bag technique for estimating rumen utilization of grains. J. Anim. Sci. 35: 113. Fisher, L.J. and Shelford, J.A. 1987. Taking a look at your silage making. In: Dairy Producers Short Course, Abbotsford, B.C. Ministry of Agric. and Fisheries, pp. 7-13. Ford, C.W. and Elliot, R. 1987. Biodegradability of mature grass cell walls in relation to chemical composition and rumen microbiol activity. J. Agric. Sci., Camb. 108: 201-209. Forsberg, C.W.; Beveridge, T.J. and Hellstrom, A.H. 1981. Cellulase and xylanase release from Bacteroids succinogens and its importance in the rumen environment. Appl. Environ. Microbiol. 42: 886-896. Forsberg, C.W. 1986. Mechanism of bacterial attachment to dietary fiber in the rumen. In: Proceedings of 13th Int. Congr. of Nutr. Ed. Taylor, T.J. and Jenkins, N.K. Libbey, London, pp. 193-195. Fritz, J.0., Moore, K.J. and Jaster, E.H. 1990. Digestion kinetics and cell wall composition of brown midrib sorghum x sudangrass morphological components. Crop Sci. 30: 213-219. Gaillard, B.D.E. 1965. The relationship between the cell-wall constituents of roughages and the digestibility of the organic matter. J. Agric. Sci., Camb. 59:369-373. Gaillard, B.D.E.; Bailey, R.W. and Clarke, R.T.J. 1965. The action of rumen bacterial hemicellulases on pasture plant hemicellulose fractions. J. Agric. Sci. Camb. 64:449. Gaillard, B.D.E. and Bailey, R.W. 1968. The distribution of galactose and mannose in the cell wall polysaccharides of red clover (Trifolium Dratense leaves and stems. Phytochemistry 7: 2037-2044. Gaillard, B.D.E. and Richards, G.N. 1975. Presence of soluble lignin-carbohydrate complexes in the bovine rumen. Carbohydr. Res. 42: 135-145. Gill, M., Beever, D.E. and Osbourn, D.F. 1989. The feeding value of grass and grass by-products. In: Grass- its production and utilization, Ed. Holmes, W., Blackwell Scientific Publications, pp. 89-129. Goering, H.K. and Van Soest, P.J. 1970. Forage fiber analysis. Agric. Handbook No. 379. Agric. Res. Serv., U.S. Dept. Agric., Wash., D.C. Gordon, A.H., Lomax, A.J. and Chesson, A. 1983. glycosid,ic 127  linkages of legume, grass and cereal straw cell walls before and after extensive degradation by rumen microorganisms. J. Sc!. Food Agric. 34:1341-1350. Gordon, A.J. and Gaillard, B.D.E. 1976. The relationship between and carbohydrate in the hemicellulose A, B and C fractions extracted from lucerne and wheatstraw with alkali. Landbouwhogeschool Wageningen. Miscellaneous Papers 112, pp 55-65. Graham, N. McM.^1967. The net energy value of three subtropical forages. Aust. J. Agric. Res. 18: 137-147. Greve, L.C.; Labavitch, J.M. and Hungate, R.E. 1984. a-Larabinofuranosidase from Ruminococcus albus 8: purification and possible role in hydrolysis of alfalfa cell wall. Appl. Environ. Microbiol. 47: 1135-1140. Groleau, D. and Forsberg, C.W. 1981. The cellulolytic activity of the rumen bacterium Bacteroids succinogens. Can. J. Microbiol. 27: 517-530. Hacker, J.B. and Minson, D.J. 1981. The digestibility of plant parts. Herbage Abstr. Vol. 51, #9 pp. 459-482 Hanna, W.W.; Monson, W.G.; Burton, G.W. 1973. Hitological examination of fresh forage leaves after in vitro digestion. Crop Sc!. 13: 98-102. Hartley, R.D. and Ford, C.W. 1989. Phenolic constituents of plant cell walls and wall degradability. In: Plant cell wall polymers biogenesis and biodegradation, Ed. Lewis, N.G. and Paice, M.G. American Chemical Society, pp. 137-145. Hashimoto, S. Fujita, M. and Baccay, R.A. 1982. Substrate microbial behavior and kinetics in the anaerobic digestion of night soil. J. Technol. 60:55-65. Hatfield, R.D. 1989. Structural polysaccharides in forages and their degradability. Agron. J. 81: 39-46. Hespell, R.B. 1988. Microbial digestion of hemicelluloses in the rumen. Microbiological Sciences vol.5 no.12, pp. 362-365. Hobson, P.N. 1988. Preface. In: The Rumen Microbial Ecosystem. Ed. Hobson, P.N. Elsvier Science Publishers, New York, N.Y. Hogan, J.P. 1982. Digestion and utilization of proteins. In: Nutritional limits to animal production from pastures Ed. Hacker, J.B., Commonwealth Agric. Bureaux, Farnham Royal, U.K. pp. 245-257. Hornstein, J.S., Buxton, R.D. and Wedin, W.F. 1989. Cell128  carbohydrates in leaves, stems, and herbage of alfalfa and red clover. Crop Sci. 29: 1319-1324. Hungate, R.E. 1966. The rumen and its microbes. Academic Press, Inc. New York. Hungate, R.E., Stack, R.J., Greve, R.C. and Labavitch, J.L. 1983. An enzymatic approach to cell wall structure. S. Afr. Anim. Sci. 13:51. Jones, D.I.H. and Hayward, M.V. 1973. A cellulase digestion technique for predicting the dry matter digestibility of grasses. J. Sci. Fd. Agric. 24: 1419-1426. Jones, D.I.H. and Hayward, M.V. 1975. The effect of pepsin pretreatment of herbage on the prediction of dry matter digestibility from solubility in fungal cellulase systems. J. Sci. Fd. Agric. 26: 711-718. Jones,^D.I.H. and Wilson, A. 1987. Nutritive quality of forages. in: the nutrition of herbivores. Proc. 2nd Int. Symp. of the nutrition of herbivores, Ed. Hacker, J.B. and Ternouth, J.H. Academic Press. Sydney Jung,^H-J.^G.;^Ralph, J. and Hartfield, ^R.D.^1991. Degradability of phenolic acid - hemicellulose esters: A model system. J. Sci. Food Agric. 56: 469-478. Jung,^H-J.^G. and Vogel, K.P. 1992. ^Lignification of switchgrass (Panicum virgatum) and big bluestem (Andropogon gerardii) plant parts during maruration and its effect on fibre degradability. J. Sci. Food Agric. 59: 169-176. Kerley, M.S.; Fahey, G.C. Jr. Gould, J.M. and Ianotti, E.L. 1988. Effects of lignification, cellulose crystallinity and enzyme accessible space on the digestibility of plant cell wall carbohydrates by the ruminant. Food Microstructure 7:59 Kudo, H.; Cheng, K.J. and Costerton, J.W. 1986. Interaction between Treponema bryantii and cellulolytic bacteria in the in vitro degradation of straw cellulose. Can. J. Microbiol. 32: 244-248. Latham, M.J.; Brooker, B. E.; Pettipher, G.L. and Harris, P.J. 1978(a). Ruminococcus flavefaciens cell coat and adhesion to cotton cellulose and to cell wall in leaves of perennial ryegrass (Lorium perenne). Appl. Environ. Microbiol., 35:156165. Latham, M.J.; Brooker, B.E.; Pettipher, G.L. and Harris, P.J. 1978(b). Adhesion of Bacteroids succinogenes in pure culture and in the presence of Ruminococcus flavefaciens to cell walls of leaves of perennial ryegrass (Lorium Derenne). Appl. 129  Environ. Microbiol., 35:1166-1173. Leatherwood, J.M. 1965. Cellulase from Ruminococcus albus and mixed rumen microorganisma. Appl. Microbiol. 13: 771-775. Maeng, W.J. and Baldwin, R.L. 1976. Dynamics of fermentation of a purified diet and microbial growth in the rumen. J. Dairy Sci. 59: 636-642. Mahadevan, S., Sauer, F.D. and Erfle, J.D. 1987. Preparation of protease from mixed rumen microorganisms and its use for the in vitro determination of the degradability of true protein in feedstuffs. Can. J. Anim. Sci. 67:55. Mangan, J.L. 1982. The nitrogenous constituents of fresh forage. In: Forage protein in ruminant animal production. Occassional publication of British Society of Animal Production no. 6 Edinburgh, U.K.:BSAP pp. 25-40. McDonald,^I. 1981. A revised model for the estimation of protein degradability in the rumen. J. Agric. Sci. Camb. 96: 251-252. McNeil, M.; Albersheim, P.; Taiz, L. and Jones, R.L. 1975. The structure of plant cell walls. Plant Physiol. 55:64. McNeil, N.; Darvill, A.G.; Fry, S.C. and Albersheim, P. 1984. Structure and function of primary cell walls of plants. Ann. Rev. Biochem. 53: 625-663. Mertens, D.R. and Ely, L.O. 1982. Relationship of rate and extent of digestion to forage utilization- A dynamic model evaluation. J. Anim. Sci. 54: 895-905. Metcalfe, C.R. 1960. Anatomy of monocotyledons. I. Gramineae. London, Oxford Univ. Press pp 731. Minson, D.J. 1967. The voluntary intake and digestibility in sheep, of chopped and pelleted Digitaria decumbens (pangora grass) following a late application of fertizer nitrogen. Brit. J. of Nutr. 21: 587-597. Minson, D.J. and Wilson, J.R. 1980. Comparative digestibility of tropical and temperate forage - a contrast between grasses and legumes. J. Austr. Inst. Agric. Sci. pp. 247-249. Minson,^D.J. 1990. Forage in ruminant nutrition. Academic Press, Inc. PP 9-58. Moore, K.J. and Cherney, J.H. 1986. Digestion kinetics of sequentially extracted cell wall components of forages. Crop Sci. 26:1230-1235. 130  Montgomery, L.; Flesher, B. and Stahl, D. 1988. Transfer of Bacteroids succinogenes (Hungate) to Fibrobacter gen. nov. as Fibrobacter comb. nov. and description of Fibrobacter intestinalis sp. nov. Int. J. Syst. Bacteriol. 38:430-435. Morrison, I.M. 1973. Isolation and characterization of lignincarbohydrate complexes from Lolium multiflorum. Phytochemistry. 12: 2979-2984. Morrison, I.M. 1974. Structural investigations on the lignincarbohydrate Complexes from Lorium Derenne. Biochem. J., 139: 197-204. Morrison,^I.M.^1975.^Delignification and hemicellulose extraction of cell walls of Lolium Derenne and Triforium Dratense. Phytochemistry 14: 505-508. Morrison,^I.M.^1979. Carbohydrate chemistry and rumen digestion. Proc. Nutr. Soc. 38:269-274. Morrison, I.M. 1980. Changes in the lignin and hemicellulose concentrations of ten varieties of temperate grasses with increasing maturity. Grass and Forage Sci. 35: 287-293. Mountfort,^D.O.^and Asher, R.A.^1985.^Production and regulation of cellulase by two strains of the rumen anaerobic fungus Neocallismatix frontalis. Appl. Environ. Microbiol. 49: 1314-1322. Mowat, D.N., Fulkerson, R.S., Tossell, W.E. and Winch, J.E. 1965. The in vitro digestibility and protein content of leaf and stem portions of forages. Can. J. Plnt Sci. 45: 321-331. Mueller-Harvey,^I. and Hartley, R.D. 1986. Linkage of pcoumaroyl and feruloyl groups with cell-wall polysaccharides of barley straw. Carbohydr. Res. 148: 71-85. Neilson, M.J. and Richards, G.N. 1982. Chemical structures in a lignin-carbohydrate complex isolated from the bovine rumen. Carbohydr. Res. 104:121-138. Nocek, J.E. and Hall, M.B. 1984. Characterization of soyhull fiber digestion by in situ and in vitro enzymatic procedures. J. Dairy Sci. 67: 2599. Nocek, J.E.^1988. In situ and other methods to estimate ruminal protein and energy digestibility: a review. J. Dairy Sci. 71: 2051-2069. Norton,^B.W.^1982. Differences between species in forage quality. In: Nutritional limits to animal production from pastures, Ed. J.B. Hacker, Commonwealth Agricultural Bureau, pp 89-110. 131  Nugent, J.H.A. and Mangan, J.L. 1981. Characteristics of the rumen proteolysis of fraction 1 (18S) leaf protein from lucern (Medicacto sativa L). Br. J. Nutr. 46:39-58 Orskov, E.R. and Mcdonald, I. 1979. The estimation of protein degradability in the rumen from incubation measurements weighted according to rate of passage.J. Agric. Sci. Camb. 92: 499-503. Orskov, E.R., Deb Hovell, F.D. and Mould, F. 1980. The use of the nylon bag technique for evaluation of feedstuffs. Trop. Anim. Prod. 51: 195. Orskov, E.R. 1991. Manipulation of fibre digestion in the rumen. Proc. Nutr. Soc. 50: 187-196. Parkinson, J.A. and Allen, S.E. 1975. A wet oxidation procedure suitable for the determination of nitrogen and mineral nutrients in biological materials. Commun. Soil Sci. Plant Anal. 6:1. Pettipher, G.L. and Latham, M.J. 1979. Characteristics of enzymes produced by Ruminococcus flavefaciens which degrade plant cell walls. J. Gen. Microbiol. 110: 21-27. Raymond, W.F. 1969 . The nutritive value of forage crops. Adv. Agron. 21:1-108. Reid, C.S.W.; Lyttleton, J.W. and Mangan, J.L. 1962. Bloat in cattle. XXIV. A method of measuring the effectiveness of chewing in the release of plant cell content from ingested feed. N. Z. J. Agric. Res. 5: 237-246. Reid, J.S.G. and Wilkie, K.C.B. 1969. Total hemicelluloses from oat plants at different stages of growth. Phytochemistry 8:2059-2065. Reid, R.L., Post, A.J., Olsen, F.J. and Mugerwa, J.S. 1973. Studies on the nutritional quality of grasses and legumes in Uganda. I-Application of in vitro techniques to species and stage of growth effects. Trop. Agric. (Trinidad) 50: 1-15.  Reid, R.L. and Jung, G.A. 1982. Problems of animal production from temperate pastures. In : Nutritional limits to animal production from pastures, Ed. J.B. Hacker, Commonwealth Agricultural Bureau, pp 21-43. Rode, L.M. and Satter, L.D. 1984. Biological procedures for estimating nutritive value. In: Laboratory evaluation of fibrous feeds - relationships to nutrients requirements and utilization by ruminants. Proceedings of a symposium on ruminant physiology, Banff, Alberta Canada, Ed. Knipfel, J.E., section 4. 132  Russell,^J.B.^1985.^Fermentation of cellodextrins by cellulolytic and non-cellulolytic rumen bacteria. ^Appl. Environ. Microbiol. 49: 572-576. SAS/STAT Software. 1985. General linear model procedure. SAS Institute Inc. N. Carolina. Scalbert, A.; Monties, B.; Guittet, E. and Rolando, C. 1985. Ether linkage between phenolic acids and lignin fractions from wheat straw. Phytochemistry 24: 1359-1362. Selvendran,^R.R.^and O'Neill, M.A. 1987. Isolation and analysis of cell walls from plant material. Meth. Biochem. Anal. 32: 25. Selvendran, R.R. and Dupont, M.S. 1984. Problems associated with the analysis of dietary fibre and some recent developments. In: Development in food analysis techniques-3, Ed King, R.D., pp. 1-68. Shelford, J.A. and Fisher, L.J. 1988. Silage additives. In: Dairy Producers' Short Course, Clearbrook, B.C., Ministry of Agric. and Fisheries, pp.25-31. Silley,^P.^1985. A note on the pectinolytic enzymes of Lachnospira multiparus. J. Appl. Bacteriol. 58:145-149. Slyter, L.L. and Putman, P.A. 1967. In vivo vs. iv vitro continuous culture of ruminal microbial populations. J. Anim. Sci. 26: 1421-1428. Smith, D. 1973. The nonstructural carbohydrates. In:Chemistry and biochemistry of herbage Vol.1, Ed. Butler, G.W. and Bailey, R.W.). Academic Press. pp. 105-155. Smith, L.W. Goering, H.K. and Gordon, C.H. 1972. Relationships of forage compositions with rates of cell wall digestion and indigestibility of cell walls. J. Dairy Sci. 55:1140-1147. Stack,^R.J.^and Hungate,^R.E.^1984.^Effect of 3phenylpropionic acid on capsule and cellulases of Ruminococcus albus 8. Appl. Environ. Microbiol. 48: 218-223 Stone, I.E., Scallan, A.M., Donefer, E. and Ahlgren, E. 1969. Digestibilities as a simple function of a molecule of similar size to a cellulase enzyme. Adv. Chem. Ser. 95:219. Susmel, P., Stefanon, B., Mills, C.R. and Sphangero, M. 1990. Rumen degradability of organic matter, nitrogen and fibre fractions in forages. Anim. Prod. 51: 515-526. Tilley, J.M.A. and Terry, R.A. 1963. A two stage technique for the in vitro digestion of forage Crops. J. Brit. Grassl. Soc. 133  18: 104-111. Ulyatt, M.J. 1981. The feeding value of temperate pastures. In: Grazing animals, Ed. Morley, F.H.W., Amsterdam, Elsevier, pp. 125-141. Van Keuren, R.W. and Heinmann, W.W. 1962. Study of a nylon bag technique for in vivo estimation of forage digestibility. J. Anim. Sci. 21: 340. Van Soest, P.J. 1965.Voluntary intake in relation to chemical composition and digestibility. Symposium on factors influencing voluntary intake by ruminants. J. Anim. Sci. 24: 834-843. Van Soest, P.J and Moore, L.A. 1965. Proc. 9th Int. Grassld. Congr. (Sao Paulo, Brazil) 1, pp. 783-789. Van Soest, P.J., Wise, R.H., and Moore, L.A. 1966. Estimation of true digestibility of forages by the in vitro digestion of cell walls. Proc. 10th Int. Grassi. Congr. Helsinki, Finland, pp.438-441. Van Soest, P.J. 1967. Developing a comprehensive system of feed analysis and its application to forages. J. Anim. Sci. 26:119-128. Van Soest, P.J. 1967. J. Anim. Sci. 26: 119-128. Van Soest, P.J. 1975. Physico-chemical aspects of fiber digestion. In: Digestion and metaboplism in the ruminant. Ed. McDonald, I.W. and Warner, A.C.I., University of New England Publishing Unit, Armidale, N.S.W., Australia, pp 351-365. Van Soest, P.J. 1982. Nutritional Ecology of the Ruminant. Cornell University Press. Van Soest, P.J. 1982. Nutrtional ecology of the ruminant. Comstock, Cornell University Press. Van Soest, P.J. 1984. Chemical procedures for estimating nutritive value. In: Laboratory evaluation of fibrous feeds relationships to nutrients requirements and utilization by ruminants. Proceedings of a symposium on ruminant physiology, Banff, Alberta Canada, Ed. Knipfel, J.E.), section 3. Varga, G.A. and Hoover, W.H. 1983. Rate and extent of neutral detergent fiber digestion in feedstuffs in situ. J. Dairy Sci. 66: 2109 Waldern, D.E. 1971. A rapid micro-digestion procedure for neutral and acid detergent fiber. Can. J. Anim. Sci. 51:67. 134  Waldo, D.R., Smith, L.W. and Cox, E.L. 1972. Model of cellulose disappearance from the rumen. J. Dairy Sci. 55: 215129. Wallace, G.; Chesson, A.; Lomax, J.A. and Jarvis, M.C. 1991. Lignin-carbohydrate complexes in graminaceous cell-walls in relation to digestibility. Anim. Feed Sci. Tech. 32: 193-199. Williams, A.G. and Strachan, N.S. 1984. The distribution of polysaccharide-degrading enzymes in the bovine rumen digesta ecosystem. Curr. Microbiol. 10: 215-220. Williams, A.G., Withers, S.E. and Coleman, G.S. 1984. Glycoside hydrolases of rumen bacteria and protozoa. Curr. Microbiol., 10: 287-294. Williams, A.G. 1986. Rumen holotrich ciliate protozoa. Microbial Rev. 51: 25-49. Wise, L.E., Murphy, M. and D'Addieco, A.A. 1946. Chlorite holocellulose, its fractionation and bearing on summative wood analysis and on studies on the hemicelluloses. Paper Trade J. 122:35. Wood, T.M., Wilson, C.A. Macrae, S.I. and Joblin, K.N. 1986. A highly active extracellular cellulase from the anaerobic rumen fungus Neocallismatix frontallis. FEMS Microbiol. Lett. 34: 37-40. Yokoyama, M.T. and Johnson, K.A. 1988. Microbiology of the rumen and the intestine. In: The ruminant animal, digestive physiology and nutrition, Church, D.C.), Prentice Hall, Engelwood Cliffs, New Jersey, pp. 125-144.  135  APPENDIX Appendix table 1 ANOVA for tropical forages (48h) ^ Source df^SS^MS^F^P > F Treatment^3 forages(F)^1^66.6^66.6^31.6^0.0001 components(C) 1^577.7^577.7 274.3^0.0001 F x C^1 1089.6^1089.6 517.3^0.0001 Block (cow)^2^39.9^20.0^9.5^0.0015 Error^18^37.9^2.1 Corrected total 23 ^1811.8  Appendix table 2 ANOVA for orchard grass DX (48h) Source  df^SS  Treatment 5 component(C) 1^98.5 maturity(M) 2 576.4 CXM 2^12.1 Block (cow) 2^343.5 Error 28^107.0 Corrrected total  MS 98.5 288.2 6.0 171.7 3.8  F 25.8 75.4 1.6 44.9  P > F 0.0001 0.0001 0.2235 0.0001  35^1137.4  Appendix table 3  ANOVA for tall fescue grass DX (48h)  Source  df^SS  Treatment 3 component(C) 1 maturity(M) 1 C x M 1 Block (cow) 2 Error 18  74.0 3.5 47.0 48.7 244.9  Corrrected total  418.1  23  136  MS  F  74.0 3.5 47.0 24.4 13.6  5.44 0.26 3.45 1.79  P > F 0.0315 0.6181 0.0796 0.1955  Appendix table 4^ANOVA for tropical forages (96h) Source  df^SS^MS^F  Treatment^3 forages(F) 1^0.03 components(C) 1^646.0 F x C 1^661.6 Block (cow) 2 1.7 Error 18 20.0 Corrected total  23  Appendix table 5 Source  0.02 582.0 596.1 0.8  0.8797 0.0001 0.4697 0.0001  1329.3  ANOVA for orchard grass DM (72h) df^SS^MS^F  Treatment 5 component(C) 1 maturity(M) 2 C x M 2 Block (cow) 2 Error 27 Corrrected total  0.03 646.0 661.6 0.9 1.1  P > F  34  248.0 446.2 49.9 229.6 120.2  248.0 223.1 25.0 114.8 4.5  55.7 50.1 5.6 25.8  Pr>F 0.0001 0.0001 0.0092 0.0001  1140.9  Appendix table 6  ANOVA for tall fescue grass DM (72h)  Source  df^SS^MS  Treatment 3 component(C) 1 maturity(M) 1 C x M 1 Block (cow) 2 Error 18  74.0 3.5 47.0 48.7 244.9  Corrrected total  418.1  23  137  74.0 3.5 47.0 24.4 13.6  P > F 5.44 0.26 3.45 1.79  0.0315 0.6181 0.0796 0.1955  Appendix table 7 Degradation values for orchard grass cellulose Treatment  Incubation time (h) 0  EARLY STEMS  LEAVES  MEDIUM STEMS  MEDIUM LEAVES  LATE STEMS  LATE LEAVES  12  18  24  48  72  26.62  37.32  50.27  62.91  77.46  26.83  36.96  50.06  62.70  77.79  26.67  37.10  49.96  63.05  77.50  25.03  52.16  53.02  67.55  76.54  24.83  52.85  55.90  67.59  77.10  24.76  53.27  54.64  68.30  77.06  22.77  38.52  42.71  53.92  73.92  77.73  22.73  37.28  41.43  54.22  73.36  77.84  22.93  37.81  45.86  52.76  73.02  77.78  20.25  33.28  43.23  66.25  74.60  79.05  20.27  34.27  43.33  66.15  74.10  78.71  20.29  33.89  42.73  66.08  74.29  78.91  18.27  32.40  41.03  51.17  74.68  71.81  17.72  30.72  41.49  51.17  68.46  71.21  17.47  31.87  41.48  51.12  67.92  71.39  18.53  41.05  49.17  60.90  67.52  78.83  18.44  41.21  48.61  60.20  67.75  79.09  18.61  40.34  48.49  60.23  67.99  79.39  138  .  Appendix table 8 Degradation values for orchard grass hemicellulose A Treatment  Incubation time (h) 0  EARLY STEMS  EARLY LEAVES  MEDIUM STEMS  MEDIUM LEAVES  LATE STEMS  LATE LEAVES  12  18  24  48  72  24.99  32.49  41.18  51.14  65.08  75.63'  25.10  36.19  43.16  49.00  65.32  75.59  25.21  34.01  43.01  52.54  64.42  76.59  23.40  36.30  40.05  54.27  65.00  74.18  23.83  37.19  40.26  56.00  66.65  75.91  23.83  37.67  40.14  55.33  66.50  75.00  21.40  30.29  32.91  37.12  57.42  62.84  21.47  31.79  33.11  58.43  63.83  21.54  31.11  32.02  37.11  57.80  63.95  15.80  23.31  29.98  53.67  65.48  68.48  16.00  22.90  31.24  52.69  65.56  67.77  16.01  23.10  30.05  53.00  66.64  69.21  15.93  25.99  32.62  37.25  53.82  57.36  16.53  25.18  31.97  39.32  52.07  56.43  16.67  24.89  31.58  38.60  52.33  57.27  18.32  32.52  49.89  53.55  62.61  72.17  18.46  34.00  48.05  55.08  62.71  71.85  18.63  33.52  48.42  55.23  63.93  73.06  139  Appendix table 9 Degradation values for orchard grass hemicellulose Treatment  Incubation time (h) 0  EARLY STEMS  EARLY LEAVES  MEDIUM STEMS  MEDIUM LEAVES  LATE STEMS  LATE LEAVES  12  18  24  48  72  25.87  49.92  61.70  72.02  83.59  88.36  25.47  51.74  60.64  69.23  83.56  88.37  26.00  48.33  61.37  72.92  82.63  88.28  22.97  43.10  58.74  72.48  80.61  85.30  23.37  44.65  57.14  70.18  79.45  86.42  23.76  44.48  59.02  70.58  80.24  86.49  22.10  40.59  46.19  54.79  73.29  78.39  22.34  40.94  46.61  55.70  73.49  78.31'  22.10  40.85  45.80  54.18  71.91  77.39  17.90  35.88  45.44  65.49  75.40  78.76  17.96  35.72  46.71  65.23  75.37  79.34  17.84  35.14  45.30  65.24  75.72  77.77  16.73  40.50  52.33  58.96  71.49  74.80  16.80  39.05  52.80  58.24  73.95  74.81  16.87  39.57  54.85  59.17  70.10  71.84  17.77  43.57  54.15  64.39  72.37  81.72  17.81  45.45  55.34  63.39  73.43  80.94  17.67  44.23  54.12  64.09  72.60  80.38  140  Appendix table 10 Degradation values for tall fescue grass cellulose Treatment  Incubation time (h) 0  EARLY STEMS EARLY LEAVES MEDIUM STEMS MEDIUM LEAVES  12  18  24  48  72  35.35  52.40  56.43  60.96  73.11  79.44.  35.33  52.50  57.79  61.50  72.84  79.41  34.97  52.16  57.82  61.33  72.96  79.58  31.23  48.21  59.20  59.74  76.44  84.37  32.02  48.48  58.31  60.11  76.72  84.39  31.27  48.05  59.65  59.82  76.66  84.44  33.66  52.17  58.53  66.07  77.27  84.81  33.56  51.95  58.47  66.32  78.47  84.88  33.47  51.91  58.32  66.27  78.48  84.72  31.45  48.37  57.64  65.38  75.23  83.73  31.55  49.19  57.54  64.73  75.20  83.66  31.46  48.92  55.68  64.60  75.34  83.50  141  Appendix table 11 Degradation values for tall fescue grass hemicellulose A Treatment'  Incubation time (h)  I^0 EARLY STEMS EARLY LEAVES MEDIUM STEMS MEDIUM LEAVES  12  18  24  48  72  31.11  44.98  49.63  48.29  62.43  70.76  30.97  43.47  46.94  49.56  62.66  70.90  31.27  43.34  45.66  50.36  62.78  71.26  47.61  45.53  66.56  79.08  33.56 33.59  34.45  47.65  45.06  64.10  77.70  33.46_  34.98  46.88  44.42  64.92  77.90  31.47  49.72  51.58  56.14  70.00  78.30  32.10  49.45  50.71  57.88  70.30  79.31  30.97  46.93  51.85  55.91  68.54  77.43.  31.58  42.62  45.44  57.71  65.63  79.34  31.51  41.97  48.66  55.50  65.85  79.29  31.58  41.28  47.69  58.17  66.90  78.38  142  Appendix table 12 Degradation values for tall fescue grass hemicellulose B Treatment'  Incubation time (h) 0  EARLY STEMS EARLY LEAVES MEDIUM STEMS MEDIUM LEAVES  12  18  24  48  '^  25.75  50.21  61.18  61.26  73.98  72^I 79.38  24.81  52.48  60.31  61.97  73.46  79.88  25.87  52.53  61.69  63.68  23.37  57.03  58.06  58.84  77.44  85.30  23.37  55.36  61.69  58.32  76.59  85.57  23.43  55.68  58.81  56.22  76.47  85.51  22.22  61.92  38.88  66.73  78.19  81.02  21.98  61.37  36.77  66.57  78.41  84.24  22.10  61.33  36.78  64.83  77.55  84.24'  17.90  56.35  63.18  67.42  78.86  87.01  16.99  55.42  60.43  68.99  78.05  86.49  17.90  55.75  61.48  68.08  77.28  84.93  81.02  Appendix Table 13 Orchard grass cellulose effective Degradability Treatments  Mean outflow rate variables k=0.02  k=0.04  k=0.06  k=0.08  Early Stems  65.34± 0.207  52.78  46.09  41.93  Early Leaves  67.03± 0.321  57.14  51.03  46.88  Medium Stems  60.00± 0.200  47.90  41.57  37.68  Medium Leaves  61.91± 1.448  50.05  43.18  38.71  Late Stems  54.57± 0.304  43.40  37.27  33.39  Late Leaves  59.84± 0.053  49.33  43.08  38.93  143  Appendix Table 14 Orchard grass hemicellulose A effective Degradability Treatments  Mean outflow rate variables k=0.02  k=0.04  k=0.06  k=0.0.8  Early Stems  55.40± 0.608  44.23  39.10  36.13  Early Leaves  58.50± 1.510  47.37  41.63  38.17  Medium Stems  51.53± 0.757  41.20  36.17  33.17  Medium Leaves  52.07± 0.611  40.10  34.07  30.40  Late Stems  44.10± 0.400  35.03  30.43  27.67  Late Leaves  55.70± 0.361  46.73  41.20  37.50  Appendix Table 15 Orchard grass hemicellulose B effective Degradability Treatments  Mean outflow rate variables k=0.02  k=0.04  k=0.06  k=0.08  Early Stems  70.70± 0.100  59.73  53.10  48.63  Early Leaves  68.40± 0.300  57.40  50.73  46.27.  Medium Stems  60.57± 0.586  49.27  43.10  39.20  Medium Leaves  61.77± 0.493  50.33  43.60  39.20  Late Stems  59.37± 0.404  49.57  43.47  39.27  Late Leaves  62.60± 0.608  52.30  43.83  41.37  Appendix Table 16 Tall fescue cellulose effective degradability Treatments  Mean outflow rate variables k= 0.02  k= 0.04  k= 0.06  k= 0.08  Early Stems  65.44± 0.071  57.32  52.69  49.70  Early Leaves  67.11± 0.068  56.82  51.13  47.52  Medium Stems  68.76± 0.189  59.23  53.75  50.19  Medium Leaves  66.82± 0.232  57.17  51.63  48.04  144  Appendix Table 17 Tall fescue hemicellulose A effective degradability Treatments  Mean outflow rate variables k= 0.02  k= 0.04  k= 0.06  k= 0.08  Early Stems  57.87± 0.208  49.17  45.27  43.07  Early Leaves  59.60± 1.054  49.43  44.70  41.97  Medium Stems  62.93± 0.569  53.63  48.87  45.97  Medium Leaves  59.63± 1.823  49.53  44.77  41.97  Appendix Table 18 Tall fescue hemicellulose 13 effective degradability Treatments  Mean outflow rate variables k= 0.02  k= 0.04  k= 0.06  k= 0.08  Early Stems  65.43± 0.058  57.33  52.73  49.73  Early Leaves  66.43± 1.155  56.30  50.70  47.17  Medium Stems  68.73± 0.208  59.27  53.73  50.20  Medium Leaves  66.83± 0.208  57.20  51.63  48.03  145  _.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items