34—7.? 1 mm V1 ~. A a. ‘1 ' “11“.1‘1111 ‘1 1:11.11 .: ‘13. 211.1: ;; 31,111 .1, ul ‘ V n 4 . '3' I,“ 1. 1:331 . . ' f ”(.111 Ian 1". ‘15 _, ,1 .:_ ‘. ,‘ A, "l' ' S?! W V“. i \ '1 ~" 1 - . 1 1%.“! ' 1-, - _ {A' V??? A i ., ‘2 . 91:", 1 114} J‘.‘ :‘I 1 -v' ff," ' 19 . '3“ 1 HM , ' I“ A!» ’8 )“1 1 ; 5:, . O ‘ M; 1” :1 f' 1“ . .1 , A" . ll 1 ‘ ‘ ." , " ‘1 .l V,.. 1 ‘ 111 .1 1 1 1 1.. 1 n 1. , . . E r ‘ .1 . :' ,1, ' " .1 '1 ‘1 .' ii, 1 ‘. A11 . v... .-— “.32,“ .«_::‘3’_ -nngé ”Maw .. .L" .51‘ 127.... W. V,‘ erI.« A A V". . —.A v. ‘ u “ ‘ .‘ A h A ._ — _ NEW-5:4 ..' - -—. -m....._._ .. x: :c:_, 3. 35:“: . .. 3..- - :9: LT'.” -1 1n. - A :; __ - 02;}: ~‘- V v V .5 . ., . .43 . - m‘ A _ WA : > J.;&.:. ~«,--—~...... ... 4.33:1: «-- V ,. “MM—”w“ . .3.- ‘-:=: ' r- . .7. - 7 -‘v ‘v ' _' A. .._..::::-; J-M=P.§2§4- - 1 Aura; $2.}. 2.3.53..- - - ' _ . . .... ._ . A A . ‘. — V_A~L>"_r..=‘=‘ ‘ ‘ . “ A L- A E - .. ' — :‘fi-gmé“ a... ”W... uv— .312“. ,7 1.; ‘3'! 111‘" n “‘1‘ ‘ir’. -EJJL; 1‘ unfit; ,5...“ $5?" .2... :3 1 r11! , ‘3‘} 1 1 a “‘11 113“" ~- 1%,")le “5:1? MW “—2-.-3. “28‘ 12,142.. ~F‘a‘ ”—1...- .151-‘3‘ 11 “1’11 3 1h 1 _ v-.?:.: “-53-; 352:: 3;».._ ‘ ‘ ”-2": 3;: 31 - «- _ U . :""==§«£ ‘ ‘ ~1’EE'5’7‘ , ~Z“ .m: '. A 54—2...- h . .. _A_ '. __ _ :H..::.9-‘ “=3"? —. -A 5. . -3 “3‘" " 5 5:." .32.... . fly- " I ' " ' ‘ A- . A’ . u.- '7. _ - . (.__ < ~ A - A . - ' " A . - .-. . . A .A- — . . _ - . ' . . : .._- ‘ . ~ _ A V . A A . v-Iv- . . . - . 1‘ 1 A . A . z. < . , A , . - . - . . . 1%. . _ . . . AA . ." A ....; A , V. _ ,1 .A ,, . , u - _ ... . I... . . A ‘ I ‘ . ~. . .'— - , 1 ~ - V 1 ‘ . , T‘ 5...- .413 ‘- -.'-— ~ I:=u~.:_=_;« Amt—nausea .1... . gems—Hm -—:.~“==-—- a L -‘-. ‘ ‘1 q? . . 11" 1 £11.19; {Pct} £151?}%§}1“1M‘i I? 1161.1“!!! Q‘Nlt {131111 11.111311. “film“ 1”. {g “m? 1‘15 “11151 1% .-—‘- EL “2" $52.:- fi" :51" .3“: 4““- 1—.— _...w ”:1?- "v‘ 111 .4" ting 1.63:; 333—: --_ a: - 'TFJEE' . “-‘E'fi :43; ‘an 5%.: JI.’ .729... :F‘flrm -‘ 193:... g “53:75 .3. .9: mag“; iii? .«5 cram-«M _:._ $375“ «r ,Fa’k“ " .21} - "$1.33 1 Iffifikfi‘af‘fl 1%: E ”133% Ltgzfix 5:. “ fig ,f ‘1 kiwi“! "114,253" TE 7‘ gar—7" .- ~ ‘. . A 5.. .1. .. .f“. . is? . (I... .2... r 1. *— as; g. 1 .1... r .‘HL ‘1: EM ".13: Rb}? 3.1.1 W11? 4‘ ‘tog’t‘flfie' Fart“ 214315 33m. ‘ L 1!.1h II‘AAIH‘ {fk’w y‘f‘ik . “111 . "\ t‘33.; * 31ng an 12111311 > I «I: V: .. $3: ' “r‘ 5.1, 1 PT; :1; '7 :é £35; 11'“ _ {F.: 43—75:“ . __ “A; 32.26 -an €92. «hf? .1“! ‘5.’ .‘L :fi-Arz“-::§..3‘;. r" "2.5.: 5 fi 3..“ ‘ ‘3: .- . .. €F 5%..‘Q- «2?; a??? 1 4&2; _ +3; .. E 'P—‘VV— . 3... w... ;_.3. lataiuw — 2“ - - t v was: a . tad ". '::m_r:v--“‘ :. a W 7' I I .1 111‘ ‘héf‘u 5}“ WM: 1 ‘ $111111..pr s 'l '1 «I: ‘ ’3‘:;« 3f“ :12: -. - - aw ’ a?" «'63:? .2966 ' ....‘....§. "“ “M-.. I" .a' ...4'~J‘ ......_ .A‘ - N .15“ . . - Dat 0-7639 ‘- . . [i y - . ..--\ fiflva 2 E53 20m Drum This is to certify that the thesis entitled THE POTENTIAL UTILIZATION OF SHORT ROTATION BIOMASS PRODUCED TREES AS A FEED SOURCE FOR RUMINANTS presented by Stephen Robert Baertsche has been accepted towards fulfillment of the requirements for Ph. D . degree in Animal Husbandry [/WW/ 7 Major prcgdéor ‘7/ '2, 753/754, / / %t A? r , x02 me, UVtKUU: rll‘L); 25¢ per day per item RETURNING LIBRARY MATERIALS: Place in book return to remove charge from circulation records III ‘ralfé \\\ l V“ ‘. - . “If/”II 1‘: '1 THE POTENTIAL UTILIZATION OF SHORT ROTATION BIOMASS PRODUCED TREES AS A FEED SOURCE FOR RUMINANTS BY STEPHEN ROBERT BAERTSCHE A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Animal Husbandry 1980 ABSTRACT THE POTENTIAL UTILIZATION OF SHORT ROTATION BIOMASS PRODUCED TREES AS A FEED SOURCE FOR RUMINANTS BY STEPHEN ROBERT BAERTSCHE Recently, there has been a resurgence in research interest in the potential use of alternative, noncon- ventional feed sources for animal production. Cellulose is the most abundant energy feed in the world but its full potential is yet to be exploited. In this study, as examination was made on very short rotation hardwood biomass trees as a potential feed source for ruminants. The specific objectives of this study were to (l) characterize and quantitate the gross chemical com— position of rapidly produced tree species (2) quantitate fiber components (celluose, hemicellulose and lignin) (3) to assess the fermentation of these rapidly produced tree species in the rumen by nylon bag technique (u) evaluate dry matter consumption, digestibility, rumen pH, rumen ammonia, VFA profiles and nitrogen balance of poplar biomass vs. alfalfa (5) to evaluate above objectives in terms of differing stages of growth and regrowth. EXPERIMENT I - CHEMICAL COMPOSITION. The selected chemical analyses indicated that biomass produced trees may be a nutritious forage source for ruminant animals. Five of the eleven biomass trees averaged over 20% crude protein with the average for all samples being 18.81% crude protein. The estimates of hemicellulose, cellulose, lignin and hemi- cellulosic sugars indicated that biomass tree samples were generally higher in cellulose, lignin and zylosezarabinose ratios when compared with alfalfa. The mineral contents of the biomass substrates over both harvest periods were com- parable with alfalfa in both macro and micro minerals mea- sured. Gross energy values were higher for alfalfa within both the early and late harvest periods. The biomass samples consisting of poplar, honey locust, black locust and aspen were all similar in their gross energy values over both harvests. EXPERIMENT II - ENSILEMENT STUDY. The fermentation para- menters measured within the ensilement study suggested ade- quate fermentation and preservation for all samples ensiled except elm, birch and willow which all exhibited significantly higher butyric acid values. EXPERIMENT III — NYLON BAG DEGRADABILITY. The degradability of samples incubated in the rumen of steers over a 6, 12, and 2H hour period exhibited a higher percentage of sample de— graded for the early harvested substrates plus those biomass species which were lowest in lignin within each harvest group. This trend was evident for dry matter, nitrogen, and acid detergent fiber disappearance. EXPERIMENT IV - INTAKE AND DIGESTIBILITY TRIAL. Within the feeding trial, as the percent of poplar increased from 66.6% to 100%, the intake of dry matter, crude protein, and energy decreased significantly vs. the 100% alfalfa and 33.3% poplar. However, the intake for the 66.6% poplar ration could be con- sidered adequate for dry matter, protein, and energy as this diet was significantly higher~(P*<.05) for intake values than the 100% poplar. The apparent digestibility coefficients suggested that the diets containing poplar were significantly less digestible in dry matter, crude protein, ADP and energy. However, both the 33.3% and 66.6% rations displayed dry matter digestibilities over 60%. Nitrogen retention and retention as percent of total intake were lower in value as the percentage of poplar increased in the diet. Differences in rumen fluid and blood parameters for lambs fed varying levels of poplar tended to be greater for the 100% alfalfa and 33.3% poplar, although no significant differences were reported. ACKNOWLEDGEMENTS I would like to express my appreciation to the following peOple whose efforts, knowledge, and understanding have aided me in my graduate program and the preparation of this thesis. Dr. M.T. Yokoyama for his guidance in my research work, critical reading of this manuscript, overall counseling in my academic work, and his patience while I was engaged with extension responsibilities. Dr. R.H. Nelson and the Animal Husbandry Department for the use of the facilities-and animals. Drs. W.G. Bergen, J.W. Hanover, DiR. Hawkins, and J.C. Waller for the added supervision of this experiment and numerous other consultations. Special thanks to E.L. Fink for her laboratory assis- tance and Marilee Kingsley for her careful typing of this manuscript. My parents, Mr. and Mrs. Wendell Baertsche, for their love, understanding, and valuable experiences that were given me while growing up on our family farm. Most of all, my wife, Vicky, for her support, love, and patience given me over my graduate program. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES. I. II. III. INTRODUCTION LITERATURE REVIEW. Background of Wood Use in Livestock Rations Wood Chemistry. . . . . . Wood Residues in Ruminant Rations Ground Whole Trees as a Ruminant Roughage Source. Irradiation of Wood Residues. Alkaline Treatment and Delignification. . Acid Hydrolysis of Wood and Wood Residues Summary . . . . . . . . . . . . . MATERIALS AND METHODS. Experiment I Harvest of Plants Collection of Samples . Dry Matter Calculations Crude Protein Determination Gross Energy Determination. Ash Determination . Ether Extract Determination Fiber Analysis. Derivatization of Hemicellulosic Sugars Mineral Analysis. Experiment II. Silage Analysis Harvest of Silage Experiment III General Design. . . Experimental Procedure. Experiment IV. General Design. Equipment Used. Feeding Program Preliminary Period. Preparatory Treatment Collection Period . . Rumen Fluid pH, Ammonia, VFA' 8. Blood Urea Nitrogen . . Statistical Analysis. iii 32 32 35 35 36 36 37 37 no LIL} 45 H5 45 us us 48 51 51 51 53 5H 5H 5H 55 55 56 IV. RESULTS Experiment I. Harvest. . . . Yields . . . . . Proximate Components Fiber Fractions. Mineral Content. Hemicellulosic Sugars. Experiment II Silage Proximate Components. Silage Fiber Analysis. Silage Mineral Analysis. Silage Nitrogen Distribution Silage pH and Organic Acids. Experiment III. Nylon Bag Degradability. Early vs. Late Harvest Dry Matter Disappearance Nitrogen Disappearance ADF Disappearance. Experiment IV Diet Composition Fiber Composition. Intake Comparisons Digestibility Coefficients Nitrogen Retention Rumen pH Rumen NH3. . Blood Ureal Nitrogen. Rumen Organic Acids. V. DISCUSSION. VI. GENERAL CONCLUSIONS VII. BIBLIOGRAPHY. iv 57 57 57 57 59 63 67 71 71+ 74 76 78 73 31 83 83 83 83 87 9O 93 93 94 96 98 98 99 99 99 103 11% 117 Table 10 ll l2 13 in 15 16 l7 LIST OF TABLES Chemical Composition of Wood Composition of Softwood and Hardwood Muka. Experimental Design for Rations and Metabolic Trend Genus Species for Alfalfa and Biomass Trees. Approximate Projected Yields of Alfalfa and Biomass Trees . . . . . . . . . (Early Harvest) Selected Proximate Analyses of Biomass Trees and Alfalfa. (Late Harvest) Selected Proximate Analyses of Biomass Trees and Alfalfa. Fiber Analyses of Alfalfa and Biomass Trees (Early Harvest) Fiber Analyses of Alfalfa and Biomass Trees (Late HarveSt). Mineral Contents of Biomass Trees and Alfalfa (Early Harvest) Mineral Contents of Biomass Trees and Alfalfa (Late Harvest). Hemicellulose Components Determined by Alditol Acetates. Selected Proximate Analyses of Ensiled Biomass Trees and Alfalfa Fiber Analysis of Ensiled Alfalfa and Biomass Trees Mineral Contents of Ensiled Alfalfa and Biomass Trees Nitrogen Distribution of Ensiled Alfalfa and Biomass Trees pH and Organic Acid Content of Ensiled Alfalfa and Biomass Trees . . . . . . . . . . . 14 52 58 59 60 61 64 65 68 69 72 75 77 79 80 82 l8 19 20 21 22 23 2M 25 Dry Matter Disappearance From Nylon Bags Nitrogen Disappearance From Nylon Bags ADF Disappearance From Nylon Bags. Chemical Composition of Treatment Diets. Fiber Analysis of Treatment Diets. Effect of Four Levels of Poplar on Nutrient Intake and Apparent Digestibilities Rumen pH, Rumen NH3, and Blood Urea Nitrogen Values. Rumen Fluid Organic Acid Composition vi an 88 91 9L+ 95 97 LIST OF FIGURES Figure l Biomass Species Black Locust and Honey Locust Utilized in Study . 2 Black Alder Photo: Specie Which Fixes Nitrogen 3 Flowchart for Alditol Acetate Analysis u Polysaccharide-Alditol— —Gas Chromatography Analysis. . . . . . . . 5 Diagram of Laboratory Analysis Conducted on Silage Samples. . . . 6 Ensilement Study: Harvested and Packed in Triplicate. 7 Nylon Bag and Paddle Utilized in Rumen Degradability Study . . . . . 7a Nylon Bag Containing Known Amount of Sample Suspended into the Rumen of Cannulated Steer. vii 33 3M Ml H2 46 H7 49 50 INTRODUCTION Demographic projections and world production statis- tics pessimistically suggest that future food demands cannot be met, unless major advances are made in agricultural tech- nology. Chancellor and 6055 (1967), and Hodgson (1974) pro- jected agricultural statistics in the United States for the period 1980-2000 and indicated there must be a two to three fold increase in grain and forage production for the purpose of feeding livestock. With animal agriculture,certainuareas of the.world have limited production of nutritious roughage or fodder not only in terms of economics but also in pro- curement (McDaniels and Liebermann, 1979). As a result, there has been a resurgence in research interest in the po- tential use of alternative, nonconventional feed sources for animal production. Agricultural crop and processing resi- dues, aquatic plants, single cell protein, and recycled animal wastes are among a few of the many resources under investigation. The use of many of these resources, however, has not been without problems, either inherent (i.e. low availability of nutrients) or aquired (i.e. toxicity) (Hintz and Heitman, 1967). In many instances, these by—products must be physically or chemically treated to free the nutri- ents, or scrutinized for detrimental effects. While cellulose is the most abundant energy feed in l the world and the ruminant animal is an efficient util- izer of this feedstuff, its full potential has not been extensively exploited (Stone, 1976). Processed whole trees are being fed to cattle in many countries (i.e. United States, U.S.S.R.). However, this procedure has been only marginally successful because mature trees are low in avail- able nutrients due to their high lignin content (Young, 1976; Ievins g_._l., 1973). In contrast, there has been very little research on the feeding potential of short-rotation rapidly produced trees. The control of the stage of growth and uniform cultivation could improve nutrient content and availability considerably, and with careful selection of species (i.e. leguminous, rapidly growing species), the yields could be comparable to more conventional forages. Most chrrent research being done on the utilization of trees as an alternative feedstuff for animal production, have utilized either the by-products of the lumbering or wood-pulping industries, and as a result have been marginally successful in demonstrating their potential nutritional value. Probably the~largestobstacle in obtaining more de- sirable production from wood and lumber residues has been that wood itself is not a homogeneous material. Not only are there differences in the composition of the two major wood classifications, softwoods and hardwoods, but also the chemical composition varies in wood taken from different parts of the same tree (Pirie, 1978). Thus, the response of various woods to the same chemical treatment can be quite different. Consideration should be given to both 3 the source of raw material as well as the treatment applied when comparing the results of studies using mature trees or wood residues. The purpose of this research in contrast to previous research approaches, was to examine the use of very short rotation hardwood tree biomass as a potential feed source for ruminant production. The specific objectives of this study were to (l) characterize and quantitate the gross chemical composition of rapidly produced tree species; (2) characterize and quantitate fiber components (cellulose, hemicellulose, lignin) of these rapidly produced tree species; (3) to assess the fermentation of these rapidly produced tree species in the rumen by nylon bag technique; (4) to evaluate dry matter'consumption, digestibility, rumen pH, rumen ammonia,vvolatile fatty acid profiles, and nitrogen balance of poplar biomass vs alfalfa in a feeding trial with lambs; (5) to evaluate the above objectives in terms of differing stage of growth and regrowth (early vs late harvest). The ultimate objective and contribution is to identify those tree species which have the best potential as a ruminant feed source when rapidly produced. LITERATURE REVIEW Background of Wood Use in Livestock Rations Wood has evoked little interest as a dietary ingredient for ruminants although it contains 70 to 80% carbohydrate (Butterbaugh and Johnson, 1974). Only in times of national emergencies or dire feed shortages have any great amount of wood or wood residues been fed to ruminants. In the Scandinavian countries, more than 1.5 million tons of sul- fite and sulfate pulps from spruce, pine and fir were fed to cattle and horses during World War II when conventional feedstuffs were in low supply (Hvidsten and Homb, 1951; Nordfelt, 1947). Researchers (Anthony 23 El-, (1969); Kitts 33 gl-, (1969); El-Sabban 33 al., (1971); Dinius et 31., (1970) have studied the incorporation of whole trees and residues such as sawdust into ruminant rations and found inclusion of more than 25-30% on a dry matter basis usually depressed intake and gain. With regard to wood residues, most research emphasis has been on the utilization of wood-pulping and lumbering by-products as potential feed sources, and residues such as ammonium sulfite liquor and lignosulfonates, sawdust, bark, slashings and whole mature trees (Butterbaugh and Johnson, 1974; Croyle at al., 1975; Meitner, 1975). These residues however, have only been marginally successful in their potential feed usage, because their nutritive content is either very low or unavailable. u Wood Chemistry Chemical analyses reveal that wood is a complex carbo- hydrate substance which consists primarily of cellulose, hemicellulose and lignin. The association of these compo- nents gives rigidity to the cell walls making them capable of withstanding the stress, weight and structure of the tree (Stone, 1976). From a nutritional standpoint, only the cellulose and hemicellulose fractions can be utilized by the ruminant and the extent of their utilization is greatly dependent upon the percent of lignin associated with the tree product. The quantities of each of these components will differ from specie to specie, and also the chemical composition of the hemicellulose and lignin frac- tions. Therefore the feeding value of wood, whether it be residues from lumber or very young tree seedlings, is de- pendent upon its chemical composition. Killman and Cote (1968) showed the following table for the gross chemical composition of the two primary classes of wood. The values are expressed on an extractive—free basis. The extractives were removed by treatment with a neutral organic solvent such as alcohol, benzene, acetone or ether, although this treatment rarely renders the wood 100% ex- tractive free. Following this table, the chemical descrip- tion of the components is described by the above authors. TABLE 1. CHEMICAL COMPOSITION OF WOOD (All values in percent of extractive-free wood Hardwood Softwood Cellulose 40-44 40-44 Hemicellulose 23-40 25-40 O-Acetyl-4-0-methyglucurono-Xylan 20-35 --- Glucomannans 3-5 --- Arabino-4-O-methy1glucurono—Xy1an --- 10-15 Galactoglucomannans --- 15—25 Lignin 18-25 25-35 Pectin 1 1 Starch 5 5 Ash .2-.3 .2-.3 The primary constituent of all wood, soft or hard, is cellulose. Cellulose is a high molecular weight polymer con- sisting of B-D-glucopyranose residues linked together in straight chains by B-1, 4 glycosidic bonds (Stone, 1976). Cellulose is insoluble in water and aqueous alkali but is soluble in, and degraded by strong acids such as sulfuric or hydrochloric (Van Soest, 1963). Hemicelluloses are low molecular weight polysaccharides soluble in water at elevated temperatures. O-Acetyl-4-O - methylglucurono-Xylan, the primary hemicellulose of hard- woods, consists of a framework of (1—4) - linked B-D- Xylopyranose residues with 4—0-methyl-a-D-glucuronic acid residues as side chains. Glucomannans are made up of randomly distributed B-D-glucopyranose and B-D-mannopyranose residues linked together by (1-4) - glycosidic bonds. The hemicelluloses of softwood are more complex, both in respect to the number of hemicelluloses present and with respect to their structure. Arabino-4-0—methylglucurono-Xylan consists 6 7 'of a backbone of (1-4) - linked B-D-Xylopyranose units with 4-O-methyl-a-Ek glucuronic acid and a-L-arabinofuronose side chains. Galactoglucomannons, the predominant type of soft- wood hemicelluloses, have a backbone of randomly distributed (1-4) - linked B-D-glucopyranose and B-D-mannopyranose residues with a varying number of a-D-galactopyranose residues as side chains. Lignin is a three dimensional polymer of phenylpropane units linked together by'C-O-C :uniC-C"fionds. In softwoods each unit carries one phenotic oxygen and one methoxyl group,whi1e in hardwoods, only about half of the units con- tain an additional methoxyl group. Lignin cannot be hydro- lyzed by acids but is soluble in strong bases and pulping agents. The remaining components (pectin, starch, and ash) make up less than 10% of the nonextractable wood. Pectin is more abundant in bark than it is in wood. Its structure is largely unknown but contains galacturonic acids and minor quantities of arabinose and galactose. The most common constituents of the ash components are calcium, potassium and magnesium; existing as carbonates, phosphates, silicates, and sulfates. Wood Residues in Ruminant Rations According to Stone (1976) the annual harvest of timber, on a weight basis, totaled some 300 million tons of wood and 47 million tons of bark. About 190 million tons of wood and 27 million tons of bark were delivered as logs to U.S. mills or exported. The remaining 110 million tons of wood 8 and 20 million tons of bark was left in the forest because it was uneconomical as raw material for the existing indus- try. Because of the large amounts of available wood residues and the anticipated shortages of conventional roughages, several researchers have investigated the use of raw sawdust and other by-products of the lumber industry. Some of the early studies by Anthony and Cunningham (1968) utilized raw oak sawdust as the sole source of rough- age in finishing rations for growing lambs and steers. They reported that in both lamb and cattle finishing trials, slightly improved gains were obtained by the inclusion of 2.5% oak sawdust fed to steers, but increasing the sawdust level to 10%, reduced gains to those obtained on the all concentrate control. In a'subsequent study, Anthony 33 31. (1969) compared 15% sawdust to 15% coastal bermudagrass hay in a finishing ration to steers. They reported steers fed the hay rations consumed significantly (P<.01) more dry matter, had a faster rate of gain, and were more efficient than those fed the oak sawdust. El-Sabban et 31. (1971), also utilizing oak sawdust, investigated the effect of particle size as well as level of oak sawdust in cattle finishing rations. In the first trial, gains of steers fed 5 or 15% fine sawdust and 5 or 15% coarse sawdust were less than those of steers fed 5% ground timothy hay; however, only the 5% fine sawdust pro- duced significantly lower gains (P<.05). In the second trial, the level of hay in the control ration was increased to 15%, and compared to this particular treatment, the 9 inclusion of 5 and 15% fine sawdust resulted in significantly lower gains (P<.05). The coarse sawdust tended to produce slightly greater gains than the fine at both levels. In both trials, steers fed 15% sawdust tended to consume more feed and be less efficient than those fed 5% sawdust, but none of the wood containing nations were as efficient as the hay controls. The investigators also noted that the incidence of rumen parakeratosis and liver abcess appeared to be increased by the inclusion of sawdust. In an attempt to differentiate between hardwoods, Millet gt gt. (1970) used the modified tg ttttg technique of Mellenberger gt gt. (1970) and investigated 24 species of sawdust residues. He reported digestibilities of hard- woods to range from g low of 2% for red alder and sweetgum to a high of 35% for aspen. The softwoods studied were es- sentially indigestible with 0% for hemlock, pine, and spruce to 5% for Douglas fir. Following this same procedure, Feist gt gt. (1970) found similar variations between hard- woods and softwood species. Bender gt gt. (1970) also utilized the tg ttttg pro- cedure described by Mellenberger E£.il~ (1970) in deter- mining the effect of ball milling tree residues. They found that milling for 24 hours increased the 48 hour digestibility of aspen from 23.4% to 42.8% and from 15.1 to 50% for white birch, but had little effect on the soft- woods fir, hemlock, and spruce. Millet gt gt. (1970) reported that as ball milling time was increased from zero to 240 minutes, the 48 hr. tg vitro digestibility of red lO oak increased from 5 to 45% while the digestibility of aspen increased from 20% to 55%. It appeared that the first 20 to 30 minutes of milling appeared to have the major influence on digestibility. In regard to the digestibility comparisons of hardwoods vs softwoods, their work was in agreement with Bender gt gt. (1970) in which softwoods were reported to be much less responsive to ball milling than hardwood species. Mellenberger gt gt. (1970) determined a correlation of 0.994 between the tg_ttttg and $2.11K2 digestibilities of complete rations containing aspen sawdust. This work signi- fied the value of thejttttttg assay in predicting the digest- ibility of ground wood residues in the rumen itself. Subsequent work by Mellenberger gt gt. (1971) reported on a digestion trial using goats. The goats were fed either 0, 20, or 40% untreated aspen sawdust in both high roughage and high concentrate diets. Their results indicated that apparent digestible dry matter, digestible energy, and digestible carbohydrate decreased linearly with both types of rations as the percentage of sawdust increased. Apparent digestibility of the aspen sawdust was calculated by a least-squared plot of the digestion coefficients for the total ration, and then by extrapolating the resulting line to a point at which aspen would constitute 100% of the ration. By this method, it was calculated that the dry matter digestibility of aspen sawdust to be 41% in a high roughage ration and 28% when incorporated into a high con- centrate ration- ll Kitts gt gt. (1968) conducted digestion trials with mature wethers using increasing levels of black alder saw- dust. They reported that as the level of sawdust increased from 0 to 35%, the dry matter digestibility of a barley based ration decreased from 80.4 to 56.5% while cellulose digestibility decreased from 75.0 to 21.7%. Dinius and Baumgardt (1970) reported that lambs fed rations containing from 0 to 50% oak sawdust maintained a rather constant digestible energy intake because they ate more of the oak rations containing up to 35% sawdust. Beyond this level, total feed intake was not increased enough to maintain a constant digestible energy intake. In a subsequent study with lambs, Welton and Baumgardt (1970) determined that as the level of oak sawdust was increased from 30 to 50%, a significant decrease in ration dry matter digestibility was seen (P<.01) but had no effect upon digestible energy intake. Dinius gt gt. (1970) reported no significant differ- ences in the dry matter digestibility of high concentrate rations containing 0 or 10% of either aspen sawdust or oak sawdust that had been finely or coarsely ground. Because the inclusion of the 10% sawdust did not significantly lower digestibility when compared to the all concentrate control, they suggested that the sawdust itself was not digested but that its presence resulted in an improved digestibility of the concentrate portion of the ration. Kinsman gt gt. (1969) conducted a performance trial in which 32 finishing lambs were fed a 20% dried hardwood 12 (aspen) sawdust diet. They reported that in general, the ration produced very acceptable growth and carcass per- formance; however, no control ration was fed with which comparisons could be made. In another performance trial by Marion gt gt. (1959), they reported that growing steers fed a ration consisting of 34.1% ground mesquite wood plus 9.0% cottonseed hulls had similar gains to a control group fed 43.1% cottonseed hulls. They also reported that in a field study utilizing pregnant beef cows, those cows wintered on mesquite stems calved normally and were in similar body condition to cows fed cottonseed hulls. Vara gt gt. (1968) utilized cotton wood sawdust in rations fed to growing bulls. They fed either 14 or 28% cottonwood sawdust and compared these diets to a corn cob control ration, gains were 0.1 and 0.2 kg less, respectively, for the sawdust diets. No significant differences were seen in carcass yield, quality or total feed costs. Kitts gt gt. (1968) utilized black alder sawdust in a performance trial with yearling steers. They compared the alder sawdust to mixed hay at 10% of a barley based ration supplemented with either soybean meal or urea. They reported that on either nitrogen source, steers fed the hay diets made slightly greater daily gains and were more efficent, although no significant differences were found. Using cottonwood sawdust, McCartor gt gt. (1972) reported that steers fed 10% of this material as a rough- age source gained significantly less (P<.05) and were slightly less efficient than steers fed an all concentrate 13 control. Gilbert gt gt. (1973) conducted a feeding trial with finishing lambs utilizing hardwood sawdust as 15% of the total ration dry matter. Compared to a control diet of hay, the sawdust fed lambs exhibited a significant (P<.05) reduction in both daily gain and feed efficiency. Sawdust and wood residues have been used in several experiments to limit energy intake by cattle. Satter gt gt. (1970) included a pelleted concentrate containing 32% aspen sawdust in a diet fed to lactating dairy cows. They found that milk production as well as fat test to be equal to that produced on a conventional hay-concentrate dairy ration. Other work in this area was conducted by Cody gt gt. (1972) with young bull calves. They reported that intake could be limited by the inclusion of from 25‘ to 45% pine sawdust in the total ration. Subsequent studies by the same group showed that dietary sawdust levels of 15% or greater resulted in an increase of rumen parakeratosis and above 25%, sawdust produced rumen distention and ruminal and omasal impaction. Ground Whole Trees as a Ruminant Roughage Source The utilization of whole, mature trees has also been investigated as a potential feed source for cattle and sheep. In the U.S.S.R., about 500,000 tons of softwood is processed annually for oils, extracts, and for a ground meal called "muka", which is fed as a roughage at 3 to 8% of the total ration dry matter (Ievins gt gt., 1973; Young, 1976). Ievins et a1. (1973) claimed reduced susceptibility to disease, increased egg production, milk and meat production l4 and increased vitality in those animals being fed the ”muka” as a roughage source. However, these data have not been substantiated because animal performance data was unavail- able. A compositional analysis of softwood and hardwood muka has been done by (Keays and Barton, 1975). This following table suggests that the % protein content of the muka is not adequate for optimum animal performance (weight gain), but may be adequate for maintenance. TABLE 2. COMPOSITION OF SOFTWOOD AND HARDWOOD MUKA (Keays and Barton, 1975), (Air-Dried Basis) Component Spruce Birch Alfalfa Muka Muka Meal Protein, % 8.79 8 0 18.3 Fats, % 6.54 8 2 3.2 Cellulose, % . 35.6 18.0 26.2 Nitrogen-free extractives, % 34.0 56.8 41.8 Ash, % 4.4 4.2 9.6 Carotene, mg/kg 139.0 380.0 172.0 Riboflavin, mg/kg 6.0 4.0 13.2 Calcium, % 0.72 0.78 1.13 Phosphorus, % 0.17 0.26 .31 Potassium, % 0.44 0.73 1.34 Magnesium, % 0.59 0.30 0.20 Iron, mg/kg 158.5 101.0 212.0 Manganese, mg/kg 292.0 30.0 29.0 Copper, mg/kg 5.6 8.0 9.9 Zinc, mg/kg 31.5 121.0 16.0 Cobalt, mg/kg 158.0 90.0 360.0 A similar product has been prepared from cottonwood leaves obtained from 3 year old nursery grown Populus species (Dickson and Larson, 1977). This Wisconsin 15 research showed that the Populus leaves contained 19.5% crude protein, 3.2% soluble protein, 0.16% soluble amino acids, 16.4% total soluble sugars and 21.8% total non— structural carbohydrates, which compared favorably with alfalfa. No animal digestion studies, however, were done to ascertain the actual feeding value of the product. Research by Kamstra (1977) has shown that cattle will con- sume rations containing up to 48% ground and pelleted whole aspen.trees, and the product has been suggested as an emergency feed for cow-calf operations. The above prod- uct has also been successfully ensiled by this researcher. Early work done at the University of Hawaii by Kinch and Ripperton (1962) evaluated the biomass production of Leucaena glauca, or Koa haole, a small leguminous tree, and also examined animal performance. With 4.6 cuttings per year, or about 80 growing days between harvests, annual green forage yield averaged 33 tons per acre (82 tons/ hectare), with a crude protein content on a dry matter basis of 27.9%. Animal performance studies, however, showed that detrimental side effects (hair and wool loss, poor reproduction) were apparent with sheep and pigs fed this forage, but not with cattle. This effect was believed to be caused by the trees' high mimosine content, and may in part, have been responsible for the discontinued biomass propagation and cultivation of this tree in Hawaii. In another study by Enjmann gt gt. (1969) utilizing ensiled poplar trees fed gg libitum plus 0.23 Kg of soybean meal, they reported that sheep lost weight and that the 16 addition of 0.34 Kg of oats was necessary to meet the main- tenance requirement. Good results have been reported with wintering steers in Canada on diets containing either 20 or 40% silage prepared by grinding freshly cut aspen and poplar trees (Anonymous, 1966). The utilization by ruminants of either ground whole trees or raw sawdust residues have indicated that this material is unsatisfactory as a significant dietary compo- ent. According to Guggolz gt gt. (1971), the inability of ruminants to avail themselves of the energy from such car- bohydrate sources may be explained by one or more of the following: (1) Lignin acts as an inert barrier between the carbohydrate and the digesting enzyme, (2) the cellulose is too highly crystalline to be quickly available to enzyme action, or (3) silica inhibits carbohydrate digestibility. A number of pretreatments to improve the availability of wood carbohydrates have been examined. Summarizing, these treatments can be classified as: (1) irradiation, (2) alkaline delignification, and (3) acid hydrolysis. The above treatments and their effects will now be dis- cussed. Irradiation of Wood Residues The irradiation of basswood was investigated by (Lawton gt gt. 1951). They measured VFA production and dry matter disappearance tg ttttg and reported that at a treatment of 108 roentgens, the digestibility of the wood was comparable to that of hay. At levels above 108 roentgens, the tg vitro digestibility decreased which 17 indicated that the wood components were being converted to forms that were no longer readily fermentable. Millett gt gt. (1970) reported that irradiation with 108 roentgens resulted in maximum digestibility, but that hardwoods re- sponded better to irradiation than softwood species. A previous study by Kitts 23.2l- (1968) investigated the lE.Kl££2 digestibility of hemlock (a softwood) which was subjected to 5 different levels of gamma irradiation. There workers reported an increase in dry matter and cellu— lose digestibility up to 0.4 X 108 roentgens. The electron irradiation of either hard or softwoods appears to increase dry matter digestibility; however, this method of wood treatment has not been fully evaluated in feeding trials because of the high costs involved. Millett gt gt. (1970) reported an estimated cost to be about $150 per ton at the dosage of 108 roentgens. Alkaline Treatment and Delignification To increase the digestibility of wood and wood resi- dues by ruminant animals, research emphasis has been stressed on the process of delignification to make the car- bohydrate fractions more available. Some of the earliest work in this area was by Beckman (1918) in which he pat- ented a NaOH process for treatment of roughages and wood residues. In the Scandinavian countries more than 1.5 million tons of sulfate and sulfite pulps from spruce, pine and fir were fed to cattle and horses during World War II when feed supplies were short (Hvidsten and Homb, 1951; Nordfelt, . _ 18 1947). Hvidsten (1949) reported the digestibility of pulp produced from spruce wood by the sulfite method to be about 85.0%. The production of these highly digestible wood pulps demonstrated the ability of wood products and residues to serve as a energy supply for ruminants. However, because of the high costs and producers' unwillingness to accept this feedstuff, the practice of feeding such materials was discontinued after the war. More recent work has investigated the use of pulping procedures that exert a marked effect on digestibility of the material but that do not remove an appreciable amount of lignin. A study by Stanks (1961) reported that treat— ing aspen sawdust with a mixture of 15% NaClO. 5H20 plus 10 o\° NaOH resulted in an increase in the digestion of cellu- lose from 11.3 to 73.5 while decreasing the lignin con- tent only 2.4%. Another subsequent result of the above treatment was that 36.4% of the sawdust dry matter could be fermented to short chain acids, primarily succinic and acetic. Using a similar treatment as Hvidsten (1949), Clarke gt gt. (1971) compared sulfite processed Douglas fir wood pulp to barley in cattle finishing rations. They increased the level of pulp to 70% of the ration and found that at this extreme, a.marked decrease in gain resulted with little change in feed efficiency. This suggested a problem of acceptibility more than digestibility. In an tg vitro study with both softwood and hardwood delignified chemical pulps, Baker gt gt. (1973) found dry matter digestibilities 19 to range from 72 to 96%. They reported that eighty percent of the samples tested had digestibilities of 90%. To further verify the increase in dry matter digest- ibility due to alkaline treatments, Bender gt gt. (1970) conducted a digestibility trial utilizing N02 or NaC10. SHZO and found increases of 2 to 3 fold for aspen and other hardwood sawdust. However, they concluded that these treatments were too costly to be of practical im- portance. Theearliest work done utilizing sodium hydroxide as a means of increasing the dry matter digestibility of wood was demonstrated by (Wilson and Pigden, 1964). It had been known that the treatment of low quality forages with sodium hydroxide would increase their dry matter digest- ibility with the early development of the Beckman process. Wilson and Pigden (1964) utilized ground poplar wood and reported that alkali treatments up to about 9% (9 g NaOH/ 100 g wood) resulted in a linear increase in tg ttttg dry matter digestibility. Above this 9% level, however, no further increases in dry matter disappearance could be achieved. As the level of NaOH increased from 0 to 9%, the lE.Ki££2 digestibility of the wood increased from 3 to 40% while the digestibility of wheat straw increased from 30 to 70%. In a subsequent study, Feist gt gt. (1970) determined the amount of NaOH required for maximum digestibility of both quaking aspen and northern red oak. They reported values to be between 5 and 6 grams of NaOH per 100 grams of wood for the quaking aspen and red oak. 20 However, in a following study, the above researchers noted that the relative increase in digestibility was highly species dependent. They reported that the digestibility of basswood increased from 5 to 56% and paper birch from 9 to 38% whereas the digestibility of red oak increased from 3 to only 15% and American elm from 9 to only 14%. Work by Millett gt gt. (1970) reported similar findings as those by Feist gt gt. in regard to the amount of NaOH required for maximum digestibility as well as to the high degree of species dependency. They also demonstrated that treat- ment with NaOH on two softwoods (spruce and fir) had no significant effect in improving dry matter digestibility. In an attempt to verify the accuracy of an tg ttttg rumen technique developed earlier by Mellenberger gt gt. (1970), this,same group studied the digestibility of aspen sawdust, aspen bark, and alkali-treated aspen sawdust by goats. In regard to the alkali treatment, aspen sawdust was mixed with NaOH at (5 gm NaOH/100 gm wood) in a rotating digester for 2 hours. By incorporating levels of 0, 5, 10, 30, 45, and 60% of this material into both high roughage and high concentrate diets, they found that as the percentage of treated sawdust increased, the overall digestibility of the ration decreased. This decrease was linear in the high roughage rations only. In making comparisons with similar rations containing untreated sawdust, these researchers con- cluded that the NaOH treatment had increased the digestibility of aspen sawdust approximately 25%. The use of anhydrous liquid ammonia to treat woods and 2l wood residues has been studied by (Millett gt gt. 1970). This study involved steeping several species of sawdust in anhydrous ammonia under pressure for varying periods of time. It was determined that one hour of steeping and penetration by the anhydrous ammonia was long enough to obtain maximum digestibilities. Millett gt gt. (1970) also reported that a five day incubation period increased the digestibility of aspen sawdust from 33% for the control to 51% for the ammonia steeped product. In a subsequent study, Mellenberger gt gt. (1970) found that steeping aspen sawdust in liquid.ammonia for 1 hour increased the tg ttttg digestibility from 20 to 37%. Tarkow and Feist (1968) attempted to define the mecha- nism by which alkali treatments increase dry matter digest- ibility. By the utilization of a calcium ion exchange column, they measured the free carboxyl content of NaOH treated hard- wood and demonstrated that the effect of such bases is to saponify the esters of 4-0-methyl-D-g1ucuronic acid thus breaking the cross links between Xylan chains of the hemi- cellulose fraction and other polymeric units. As a result of this chemical action, there is a marked increase in the fiber saturation point or swelling capacity of the wood which provides for improved diffusion conditions of water soluble materials as well as improved enzyme substrate inter- actions. Feist gt gt. (1970) and Millett gt gt. (1970) have re- ported that the response of wood to alkali treatment is highly species dependent. Results from their studies have indicated 22 that hardwoods are more responsive to alkali treatments than softwoods. However, the digestibility of all species increases as their lignin content decreases. Work by Baker (1973) showed that at lignin contents of less than 7%, hardwood and softwood residues have essentially equal digestibilities of about 65%. However, at higher lignin contents, hardwood residues are more digestible than soft- wood residues. Baker (1973) concluded that the differences between these two classes could be explained by the follow- ing factors: (1) softwoods contain 25 to 50% more lignin than hardwoods, (2) differences exist in the lignin-car- bohydrate association between the two woods and (3) the lignin structure of softwoods differs from that of hardwoods. In determining differences in chemical structure between softwoods and hardwoods, Bender gt gt. (1970) reported that softwood lignins are almost exclusively built from guajacyl units while hardwood lignins contain both guajacyl and syringyl units. The significance of this is that the syringyl units might prevent cross-linking and the formation of a three dimensional lignin polymer which, in the softwood species, could effectively block theaccess of enzyme systems required for digestion. Acid Hydrolysis of Wood and Wood Residues 'As with alkaline treatment, final products of wood hy- drolysis with acid are greatly dependent upon the type of hydrolyzing process and the nature of the wood (hardwood vs softwood) that is being hydrolyzed. The following is a summary of acid hydrolysis and its effect upon wood 23 carbohydrates described by (Wise and John 1952). A mild form of acid hydrolysis is the treatment of wood with water. When treating wood with water at elevated tem- peratures, 10 to 30% of the weight of the wood will be solu- bilized. The similiarity of water hydrolysis and acid hy- drolysis is found in the action of hydrolytic degradation of the carbohydrate fraction. Conversely, alkali treatment acts in the manner of lignin solubilization which has already been described. Water hydrolysis will solubilize most of the hemicellulose fraction of wood residues but virtually does not affect the cellulose. The treatment of wood with dilute acids at elevated temperatures will result in at least some of the cellulose being hydrolyzed. However, when hydrolyzing with a concentrated acid followed by dilutions and boiling, 98 - 100% of the cellulose will be'converted to glucose. The treatments of using elevated temperatures and acid con- centrations that hydrolyze cellulose in a matter of hours, will readily convert hemicellulose and intermediate products into simple sugars in seconds or minutes. Under some hydro- lysis conditions, sugar decomposition is also promoted as is seen when cellulose is hydrolyzed rapidly. In the decompo- sition of sugars, the pentose sugars always break down more rapidly than the hexoses. Wise and John (1952) reported that the monosaccharide fraction are converted to organic acids which may represent up to 50% of the weight of the sugar originally formed with the remainder of the sugar being converted to humus material and gases. About 50% of zylose is converted to furfural, 24 and yields on a dry matter basis of the original wood range from 2.0% for softwoods to about 4.5% for hardwoods. The hexose sugars yield levulinic acid, among other substances, upon decomposition. An intermediate formed in this break- down is S-hydroxymethylfurfural, but this compound is highly unstable and is readily converted to levulinic acid. The products produced from sugar decomposition appear in the residue as a resinous tar due mainly to the polymerization of furfural, hydroxymethylfurfural, and levulinic acid. The solid portion of the residue is composed of approximately 15-30% lignin and about 10% extraneous substances such as tannin decomposition products, oils, and waxes. Considerable amounts of cellulose may remain in the residue depending upon the conditions and efficiency of the hydrolyzing process. The practical aspect of acid hydrolysis on good was described in the production of wood molasses. This process involved the percolation of a dilute solution of hot sulfuric acid, under pressure, through a column of sawmill wastes. The acid solution drawn from the bottom of the percolators is then neutralized and the resulting material filtered to remove the solid fractions. The remaining sugar or molasses is 50-60% dry matter. Work by Jones (1949) reported that wood sugar molasses fed to dairy cattle was equal in energy value to cane molasses when incorporated into rations for both growing heifers and lactating cows. Barrentine and Leveck (1949) compared oak wood molasses to ground white corn in beef cattle finishing rations and reported the feeding value of wood molasses to be equal to corn when 25 up to 25% of the corn dry matter was replaced by molasses dry matter. Wood molasses studies have exhibited favorable results; however, the high costs of production and limited use of the remaining residue after hydrolysis has limited its use by producers. As was mentioned previously, some of the early work at feeding acid and pressure treated wood products to livestock was done during and after World War I in Germany and Scandanavia. A review by Schneider (1943) stated that the chemical treatment of sawdust and bark can increase its digestibility, but these feeds when included in rations almost invariably depress the digestibility of one or more nutrients. Sherrard and Blanco (1921) conducted early studies with dilute acid hydrolysis of wood. They digested white pine sawdust with 1.8% sulfuric acid for 15 minutes under a steam pressure of 120 PSI and neutralized the resulting material with lime. When compared to the original sawdust material, the hydrolyzed product contained 20% less cellulose, the same amount of lignin, about 12% less crude fiber, and 16-18% more reducing sugars. They used the above material in a ration for lactating dairy cows and found adequate dry matter intake. However, no milk production or breeding data were presented. In a digestion trial with sheep, Archibald (1926) de- termined the feeding value of hydrolyzed wood produced by Sherrard and Blanco's (1921) procedure. Archibald reported the average dry matter digestibility to be about 46% for 26 white pine hydrolyzed sawdust and 33% for a similar product made from Douglas fir. Using these digestibilities, he calculated the net energy of the hydrolyzed wood products using the formula derived by Armsby (1917). The eastern white pine hydrolyzed sawdust had an apparent net energy value of 18.6 therms/100 lbs. but the net energy value of the Douglas fir hydrolyzed sawdust was a negative quantity, indicating that more energy was used up in the process of digestion than the material contained. In a second trial, Archibald compared 20% of the two wood products to 20% starch in rations for lactating dairy cows. They reported that fat content was not affected by treatment and there were little differences in body weight change, however, milk production was 4% less for cows fed Douglas fir hydro- lyzed sawdust and 1% less for cows fed pine hydrolyzed saw- dust when compared to starch controls. Bender gt gt. (1970) studied the effect of steam heat- ing of seven hardwood samples at 1750C for 2 hours on tg gtttg digestibility values. They reported this treat- ment increased 48 hour tg ttttg digestibilities by an average of 21.3% but had little or no effect on three soft- wood samples tested. In a subsequent study, Kitts gt gt. (1968) reported that the above increases in digestibility were not reflected in animal performance data. They compared 15 and 20% alder sawdust that had been treated under 2,000 psi of pressure and heated at 1170C to an untreated sawdust, and 15 or 20% hay in beef cattle growing rations. The hay fed cattle gained significantly faster (P<.01) and made more 27 efficient gains (P<.05) than those receiving either of the sawdust rations, while no differences were found be- tween treated and untreated sawdust. Studies by Butterbaugh gt gt. (1972) compared in- creasing levels of a 4:1 mixture of hardwood to pine saw- dust that had been hydrolyzed with 0.8% H2504 to alfalfa meal and reported that as the level of wood residue increased in the ration, dry matter and organic matter digestibility decreased (P<.05), but there were no effects upon nitrogen or cellulose digestibility. Work by Hudson (1971) utilized pine sawdust that had been hydrolyzed with 0.8% H2804 also. He conducted a digestibility study with lambs and found that the inclusion of 15% of this material resulted in a significant decrease (P<.01) in both dry matter and nitrogen digestibility. Following this study, Hudson conducted a nitrogen balance trial and comfirmed the depression in nitrogen digestibility of his first study, and demonstrated 0 a significant decrease (P<.05) in the 6 intake nitrogen retained and % absorbed nitrogen retained as a result of replacing alfalfa pellets with 15% hydrolyzed pine sawdust residue. Butterbaugh (1972) conducted a growth trial with lambs utilizing increasing levels of the 0.8% H2504 sawdust residue. He reported that as the level of this low acid product in- creased, weight gains decreased, but the amount of non—wood dry matter required per Kg of gain decreased indicating that some of the wood residue was being utilized as an energy source. In a similar study, Hudson (1971) had reported a 28 decrease in weight gains but had found that in both steer and lamb trials, treating the pine residue with either 2.5% NaOH or 2.5% NaOH plus heat would overcome the depression in gains but had no effect in overcoming the reduced effi- ciency observed when this material was fed without further treatment. Utilizing rats in growth trials and metabolism studies, Hudson (1971) utilized the same hydrolyzed pine sawdust as was used in the studies with ruminants. He reported that the feeding of hydrolyzed sawdust resulted in depressed gains and performance in monogastrics as was seen in ruminants. Additional treatments of the wood residue with NaOH, NH4OH, or combinations of either of these bases with heat was only partially effective in overcoming the depress; ing effects caused by this product. From his results, it was also included that the optimum levels of treatment were 2.5% for NaOH and 2.0% for NH4OH on a dry matter basis. However, as Hudson increased the dietary crude protein level to 24%, it was demonstrated that the detrimental effects of the sawdust upon animal performance were eliminated. He found that the supplementation of methionine or lysine, or both, was an effective means of increasing the performance of rats fed hydrolyzed sawdust with greater gains made when a combination of the two were fed. In subsequent work by Butterbaugh (1972), he conducted a digestibility trial with a 4:1 mixture of hardwood to pine sawdust that had been hydrolyzed with 2.3% H2504. His study found that in high alfalfa meal diets, dry matter, 29 organic matter, and nitrogen digestibilities were signifi- cantly depressed by the inclusion of 20 or 35% of this hydrolyzed product, but that the decrease in nitrogen digestibility was partially overcome by the supplementation of soybean meal. Johnson gt gt. (1973) utilized this same material and treatment in a digestibility study with sheep and reported that when hydrolyzed sawdust replaced cotton- seed hulls at levels of 25 and 50%, dry matter and organic matter digestibilities decreased but that the depression was not as severe as when hydrolyzed sawdust replaced alfalfa meal. In contrast with the alfalfa diets, the nitrogen digestibility increased when hydrolyzed sawdust replaced cottonseed hulls. In a performance study of growing steers, Johnson gt gt. (1973) wintered steers on high alfalfa meal diets containing 0, 20, 30, and 40% of a 2.3% acid hydrolyzed sawdust. They noted that as the level of wood residue increased, gains and the efficiency of gains decreased. Subsequently, Butterbaugh (1972) conducted a growth trial with lambs and reported that lambs consuming a basal alfalfa meal diet gained faster and were significantly more efficient (P<.05) than lambs receiving diets contain- ing 20 and 35% hydrolyzed sawdust. Johnson gt gt. (1973) found that adult ewes maintained on rations containing 25 and 50% hydrolyzed sawdust lost weight while the ewes on a control ration of cottonseed hulls gained approximately 3 pounds each over the 2 month period. Both Johnson gt gt. (1973) and Butterbaugh (1972) have reported that adding 30 acid hydrolyzed sawdust above 25%-30% dry matter will reduce both dry matter digestibility and animal performance. How— ever, Butterbaugh and Johnson (1974) fed increasing levels of rations treated with .8% H2804 and noted that rations containing up to 75% of the low acid treated residue were consumed well by growing lambs. There were no significant differences in weight gains of lambs fed the 25 or 50% residue rations when compared to the basal ration of alfalfa meal. However, feed efficiency tended to increase as the levels of wood residue increased over 25% of the diets. When supplemented with soybean meal, the 75% ration signifi— cantly decreased the dry matter/Kg gain requirements and increased weight gains. Summary ' . Most current research on the utilization of trees as an alternative feed source for animal production, have utilized only the by-products of the lumbering industry and as a result have been only marginally successful in demonstrating their potential nutritional value. It is a well known fact that for effective utilization of wood residues, the carbohydrate fraction be accessible to the action of rumen microorganisms. However, to establish this criteria, these wood by-products must usually be physically and chemically treated to destroy the highly lignified polysaccharide complexes, which resist fermentive and digestive processes. This added treatment is often economically unfeasible. 31 It is the purpose of this research to contrast previous research approaches. We will concern ourselves with the utilization of very short rotation tree production (hard- woods) as a potential feed source for animal production, specifically the ruminant. MATERIALS AND METHODS Four different experiments were performed: (1) a selected chemical analyses of ten biomass trees and an al- falfa control over two harvest periods (early vs. late .harvest), (2) a measurement of the ensilement characteristics, chemical composition, and rumen degradability of ensiled bio- mass and alfalfa samples (early harvest), (3) a determination of rumen degradability of biomass trees and alfalfa (early and late harvest) utilizing the dacron bag technique (Orskov and Mehrez, 1977), and (4) a digestion trial with wether lambs to determine ration digestibility, dry matter intake, nitrogen retention, rumen fluid pH, rumen ammonia, and blood urea nitrogen levels feeding varying levels of poplar vs. alfalfa meal pellets. EXPERIMENT I - Chemical Composition of Biomass and Alfalfa Samples A. Harvest of Biomass Trees and Alfalfa In the spring and summer of 1979, ten hardwood tree species were harvested twice from the Michigan State Univer- sity Tree Research Center. These materials were weighed on a freshly cut basis from premeasured plots and random samples of each specie were taken to analyze for dry matter and chem- ical composition. Potential dry matter yields were also cal- culated from this data. Both early and late harvest of alfalfa 32 Figure 1. Two biomass species utilized in the study. Above — Black locust, Below — Honey locust. Notice high leaf to stem ratio on the plants. 33 34 Figure 2. Black Alder — this specie and black locust on preceding page fixate nitrogen. 35 meal samples were harvested on the Harold Lietzke farm in St. Johns, Michigan. The alfalfa samples were collected and handled in the same manner as the biomass produced trees. B. Collection of Feed Samples for Analyses All samples of the biomass trees and alfalfa were ramdomly collected at several areas from each storage con- tainer and a composite was-made for each specie. Samples were ground through a 1 mm screen using a Wiley Milll prior to all chemical analyses with the exception of obtaining dry matter values. C. Dry Matter Percent All samples were analyzed for percent dry matter by recording initial wet weight and then drying the samples in an oven for 24 hours or longer. After complete drying, weights were recorded as percent of wet sample. D. Crude Protein and N Levels All samples were analyzed for N content using a semi- micro Kjeldahl digestion method with a Technicon Autoanalyzer II Sampler and Colorimeter. A 10% copper sulfate solution was used as a catalyst to assist in breaking down the organic matter. Potassium sulfate was added to raise the boiling point of the digestion process. The carbon and hydrogen lThomas - Wiley Mill, Arthur Thomas Co., Philadelphia, Pa. 36 of the organic matter were oxidized to carbon dioxide and water while the nitrogen was converted to ammonium sulfate. The procedure used was Official Methods of Analysis of the Association of Official Agricultural Chemists (1970). E. Gross Energy of Samples Gross energy values for each ration were obtained by utilizing the Parrl Adiabatic Oxygen Bomb Calorimeter. A previously weighed sample of each specie was placed in a combustion capsule. The capsule was placed in an oxygen bomb containing 25 atmOSpheres of oxygen. The oxygen bomb was covered with 2000 g of water in an adiabatic calorimeter. After the bomb and calorimeter had been adjusted to the same temperature, the sample was ignited with a fuse wire. The temperature rise was measured under adiabatic conditions. By multiplying the hydrothermal equivalent of the calorimeter times the temperature rise minus some small corrections for the fuse wire oxidation and acid production, the caloric content of the sample was calculated. F. Ash Values of Feed Samples Ash percentage was determined by igniting pre-weighed plant samples at 6000 C in a muffle furnace to burn off all of the organic material. The inorganic material which does not volatize at this temperature is regarded as ash. 1Parr Instrument Co., Moline, Illinois 37 Calculations were made on a dry matter basis with the weight of the residue ash expressed as a % of the original dried sample. C. Ether Extract Determination of Feed Samples Ether extract values were evaluated based on the principle that ether is continuously volatized, then condensed and allow- ed to reflux through the feed sample, extracting ether soluble materials. The extract was then collected in a beaker. When the process was completed, the ether was evaporated under a hood and collected in another container and the remaining ether extracted residue was dried and weighed. The final calculations were made on a dry matter basis with the weight of the ether extract expressed as a % of the dried original sample. - H. Fiber Analysis Values of Samples Neutral Detergent Fiber- this procedure attempts to divide the dry matter of feeds very near the point which separates the nutritively available and soluble constituents from those which are incompletely available or dependent on microbial fermentation. The specific procedure used was described by Van Soest and Wine (1967). A previously weighed sample was placed in a Berzelius beaker for refluxing. The following reagents were added in order: neutral detergent solution, decalin, and sodium sulfite. The mixture was heated to boiling for five to ten minutes and then reduced and refluxed for 38 60 minutes. Previously tared crucibles were placed on a filtering apparatus. Beakers were swirled and contents were poured into each crucible and a vacuum was applied. The remaining mat was washed twice with acetone, and dried at 1050 C over— night and weighed. Calculations were made on the dry matter basis with the weight of the dried NDF fraction expressed as a % of the original dry sample weight. Acid Detergent Fiber - this fraction supposedly repre- sents ligno-cellulose in feedstuffs. The residue also includes silica, however. The difference between the cell walls and acid detergent fiber is an estimate of hemicellulose, al- though this difference does include some protein attached to cell walls. The acid detergent fiber is used as aypreparatory step for lignin determination. The procedure used was that of Van Soest (1963). A previously weighed sample was placed into a Berzelius beaker for refluxing. Reagents of acid-detergent solution and decalin were added and the mixture was heated to boiling for 5 minutes. The heat was then turned down and the material was refluxed for exactly 60 minutes. The volume was then filtered on a previously tared crucible to which a vacuum had been applied. The remaining mat was washed twice with acetone and then dried at 1050 C overnight and weighed. The cal- culations were made on a dry matter basis with the weight of the dried ADF fraction expressed as a % of the original dried sample. 39 Permanganate Lignin — this procedure of fiber deter- mination utilized the acid detergent fiber procedure as a preparatory step. The detergent removed the protein and other acid-soluble material which would interfere with the lignin determination. The principle of the procedure is that the acid detergent fiber residue is primarily lignocellulose of which the cellulose is dissolved by the permanganate solutions. The remaining residue consists of lignin and acid-insoluble ash; however, with samples containing large amounts of cutin this also is measured as part of the lignin. This is an indirect method for lignin, utilizing per- manganate, and allows the determination of cellulose and insoluble ash in the same sample. The insoluble ash is an estimate of silica content, which in many forages is a factor in reducing digestibility. The crucibles from the acid detergent fiber procedure were placed in a glaSS'tray with one end of the tray 2 to 3 cm. higher so the acid could drain away. To each crucible 30 to 40 ml. of the permanganate solution was added. The mats of the material were broken up with a stirring rod to allow better sample contact with the solution. The samples were left in contact with the solution for 90 minutes. New solution was continually added at all times during the digestion process. At the end of digestion time, the permanganate solution was promptly suctioned off. Approximately 20 ml. of demineralizing solution was then added and allowed to stand until the solution color changed. At the end of this time, this solution was filtered off and 40 the digestion was considered complete by the completely white color indicated. The calculations were made on a dry matter basis with the weight of the dried lignin fraction 9 expressed as a o of the dried ADF fraction. I. Derivatization of Hemicellulosic Sugars to Alditol Acetates The 10 biomass samples and alfalfa sample from the first harvest were delignified with sodium chlorite and hydrolyzed with trifluroacetic acid. The hydrolyzed sugars were derivatized to their corresponding alditol acetates and analyzed on the gas chromatograph. Summary of Procedure of Alditol Acetates A 50-60 mg delignified sample (F) was transferred to a 18 X 150 mm heavy duty Wheaton tube containing 10 ml of 1 N trifluroacetic acid. Myoinositol (5-7 mg) (G) was added to each tube as an interval standard. Each delignified sample was run in triplicate. The tubes were sealed with slotted stoppers and aluminum seals and hydrolyzed at 1200 C for 60 min in an autoclave. After the hydrolysis, the sam- ples were filtered through a preweighed scintered glass crucible (No. 3). The filtrate was then collected and poured into a 50 ml round bottom flask. The residue was washed with 50 ml of deionized distilled water and dried for 24 hours. The filtrate was evaporated to dryness under a stream of filtered air in a 600 C water bath. The hydrolyzed hemicellulose sugars in the filtrate were reduced to their respective alditols with sodium borohydride (l g) in 1 N ammonia (50 ml) for 1 hour with occasional swirling. The Ml .m osswfim >Ia_Eo a . a mac/40%;: $32.0. 0.833385% A 99. 0.505535» $333830: 2.29.2 zoquxo N0.9.12 .89.. 4 0.20m: mEOm oz< §m<4mOF>O AiZOrEmOE wh<4mo m_w>._>O.E 42 POLYSACCHARI DE - ALOITOL- GAS CHROMATOGRAPHY ANALYSIS POLYSACCHARIDE l HYDROLYSIS H‘ (3,0 I H-c-OH l w REDUCTION HO‘C‘H ____.> HO- p-H H-C-OH I H-C-OH I H MONOSACCHARIDE (e.g. GALACTOSE) ALDITOL ACETATE: Figure 4. '1 H-c-OH ' ACETYLATION . H‘f‘OH -—-’ ALDITOL ffi3-(?.+1 lH3517¥rE HO-c-H I H-c-OH 'a H-C-OH I H ALDITOL H .. V H- C— O- C— CfH I H S.) H Hc - o-c - CH l ” -O-j-H " 0" I‘ H 9 H Hc-O-c-cH l 9 H H - C-O—C—CH H ' H H3 The reduction was stopped with glacial acetic acid until the reaction stopped. Ten milliliters of methanol was then added and evaporated to dryness under a stream of filtered air in a 600 C water bath. Five more 10 ml methanoladditionswere made and evaporated to dryness as before. Acetic anhydride (4 ml) was then added and the round-bottom flask was sealed and wired down. The mixture was then heated at 1200 C for 1 hr. in an autoclave. GLC Quantification of Alditol Acetates A 3-5 ul sample of the autoclaved mixture was injected into a gas chromatograph Hewlett Packard 5840A Gas Chromato- graph equipped with a hydrogen flame ionization detector. A stainless steel column (120 X 0.3 cm) packed with 0.2% polyethelene glycol adipate, 0.2% polyethelene glycol succinate, and 0.4% silicone X F - 1150 on Gas-Chrom P (100-120 mesh) was used. Other GLC parameters were as follows: Column temperature, programmed between 135-2000 C with a lO-min holding at 1350 C after injection of the sample followed by l0 C/min increase in temperature; helium flow rate of 30 ml/min, injection temperature of 2109 C, detector temperature of 2500 C, attenuation of 32 X and range of 1 mV. Calculations of Results The following formula was used to calculate the percent hemicellulosic sugars: O 6 sugar = C X wt. Of sugar BC/D Sample wt X A 100 wt. of cellulose in E = JE/F wt. of sugar in E = GIE/FH 44 Where A, dry matter of ground sample; B, oxalate fiber residue scrappings; C, weight of oxalate fiber at 0 min; D, weight of oxalate fiber at 30 min; E, weight of holo- cellulose sample (50 mg); G, weight of myoinositol; H, myoinositol GLC peak area; 1, sugar GLC peak area; J, weight of cellulose. J. Mineral Analysis of Biomass Trees and Alfalfa Determination of Ca, P, K, Na, Mg, Fe, Zn, A1, Cu, Mn - these elements were all evaluated using a SMI 111 Direct Current Spectrophotometerl in the MSU Natural Resources Analytical Laboratory. The method of sample preparation was described by Walsh (1971). The plant tissues were finely ground after drying at approximately 12 hours. A sample Of 1.0-1.5 gm. of dried plant tissue was weighed and placed into a porcelain crucible. This was ashed in a muffle furnance at 5000 C for 2 to 4 hours. The ash was then dissolved in 5 ml of 20% (ZN) HCL to place the residue sample in complete solution. The solution was filtered through an acid-washed filter paper into a 50 ml flask and collected for analysis on the SMI III. lSpectramics Inc., Lexington, Mass. H5 EXPERIMENT II - Silage Fermentation Study A. Harvest of Biomass Trees and Alfalfa At the time of first harvest, three samples per tree - species and an alfalfa control were harvested, wilted to 30-40% DM, chopped uniformly as possible, and packed in laboratory glass silos for ensiling. Each sample was sealed tightly and had C02 gas bubbled through the sample for 2 minutes each. The samples were stored at the MSU Agricultural Fermentation Laboratory during the 24 day ensilement period. B. Silage Analysis A schematic diagram of analysis conducted is shown in the following figure. Immediately after removal from the experimental silos, total nitrogen was determined by macro-Kjeldahl procedures and percent dry matter determined by oven drying for 24 hours at 550 C. Silage extracts were prepared by homogenizing a 25 gm aliquot of the sample in a Sorvill homogenizer with 100 ml of distilled and deionized water for one minute and straining through two layers of cheesecloth. A 20 ml aliquot of the extract was used for determining pH and soluble nitrogen. The pH was determined on-a Beckman Model 4500 pH meter. The remainder of the extract was deproteinized using one ml of 50% sulfosalicylic acid (SSA) and nine m1 of ex- tract. The sample was then centrifuged at 18,000 rpm for 10 minutes and stored in a refrigerator for later analysis. Volatile fatty acid content of the silage was determined by H6 Schematic Diagram of Laboratory Analysis Conducted on Silage Samples Sample > Total nitrogen (macro-Kjeldahl) Dry Matter { Selected proximate components E X 1‘}? A CI'T (25 gm. silage/100 ml distilled H20) pH (______., Sulfosalacylic acid (SSA) extract i 7 deproteinization) Soluble Nitrogen NH3 - Nitrogen (Conway) ____+ VFA'S (gas chromatograph) Lactic acid (Barker and Sommersen) FIGURE 5. Bergen gt al. (1968) H7 Figure 6. ENSILEMENT STUDY - All biomass species were harvested and packed in triplicate. All samples were gassed with C02. H8 injecting samples of the deproteinized silage fluid described above into a Hewlett Packard 5840A Gas Chromotograph. Colori- metric procedures of Barber and Sommerson (1941) were used to determine lactic acid content of the deproteinized sam— ple. EXPERIMENT III - Rumen Degradation Using Dacron Bag Technique A. Design of Study Four rumen fistulated crossbred steers weighing approx- imately 363 Kg were utilized in the study. The steers were housed in the metabolism room of the Beef Cattle Research Center for the duration of the period. Each steer was fed an equal amount of grass-legume hay morning and night with trace mineral salt blocks and water presented free choice. The degradability of each potential feedstuff was measured over a time period of 6-12-24 hours using the arti- ficial bag technique described by (Orskov and Mehrez, 1977). Degradability of dry matter, nitrogen and ADF was measured on each specie. Each feedstuff was incubated in the rumen of each of the four steers in duplicate giving a total of eight values per treatment. B. Experimental Procedure Dacron bags containing 2000 holes/cm2 were employed in the study. Each bag was made to a size of 12 X 5 cm and was stitched with rounded corners in order to avoid ac- cumulation of the test feed and to facilitate easy removal of the residues. The bags were washed and dried to a 49 Figure 7. ABOVE: Nylon bags (2000 holes/cmz) were tied to teflon paddle. Elastrator bands were placed over top of bag to insure proper seal of bag. FOLLOWING PAGE: Nylon bag containing known amount of sample was suspended into the rumen of a cammulated steer. 50 Figure 7a. 51 constant weight at 1000 C. The required number of bags, each containing a known amount of samples, (3-4 gms) were each tied to a steel leader on a hard plastic paddle. The bags were soaked in water for approximately 5 minutes and then suspended in the rumen. At the end of each incubation interval, the bags were removed from the rumen and washed thoroughly under running tap water until the rinsing water was colorless. They were then dried to a constant weight at 1000 C. The proportion of dry matter which had disappeared was calculated from the amount incubated and that left in ' the bag after incubation. Acid detergent fiber and nitrOgen values were then calculated from the remaining residue. EXPERIMENT IV - Digestion Trial A. Design of Study A 4 X 4 Latin Square design was employed to compare the dry matter intake, digestible energy intake, and di- gestion coefficients for dry matter, crude protein, di- gestible energy, and acid detergent fiber. Other parameters measured included rumen fluid pH, blood urea nitrogen, and rumen ammonia values. The four treatment diets in- cluded 100% alfalfa, 33.3% poplar - 66.6% alfalfa, 66.6% poplar - 33.3% alfalfa, and 100% poplar. The experimental design and rations are show in Table I. B. Equipment Used Metabolism Cages - sheep digestion cages were used which permitted the feeding of a known amount of feed and 52 TABLE 3. EXPERIMENTAL DESIGN FOR RATIONS AND METABOLIC TRIAL Feeding Period l 2 3 4 Lamb Number 1 A B C D 2 D A B C 3 C D A B 4 B C D A lRation Code: U0003> 100% alfalfa 33.3% poplar - 66.6% alfalfa 33.3% alfalfa - 66.6% poplar 100% poplar 53 water and the quantitative collection of urine. Urine Containers - plastic containers were used to collect the daily urine volumes under the metabolism cages. Five liter plastic bottles were used to store the urine during the collection period. Feces Collections - feces were collected in collection bag harnesses and emptied both morning and night and wet weights were taken. Covered plastic buckets were used to store the feces during the collection period until subsequent analysis for dry matter, nitrogen, gross energy and acid detergent fiber content was performed. Scales - a portable Toledo scale was used to weigh the feed, feces, urine, and lambs at the beginning and end of each feeding trial. Preparation of Feed — previously weighed mixtures of the various rations were delivered to the Harold Lietzke Farm Pellet Mill at St. Johns, Michigan for processing. After pelleting the rations, the pellets were stored in dry plastic containers and sealed to avoid moisture and other contamination. C. Feeding Program Four Suffolk wethers weighing approximately 40-50 Kg were housed in the metabolism room at the MSU Beef Cattle Research Center during the entire experiment. The sheep were fed ad libitum once a day at 8:00 a.m. and were adjusted to new diets for 14 days prior to each 7 day collection period. Daily DM intake was reduced to 90% of the ad libitum feeding after the first 7 days of the adjustment period to insure adequate consumption. Fresh 54 water was given to the lambs both in the morning and night during the collection period. Trace mineralized salt was provided free choice to all the lambs during the entire study. The 7—day fecal collections were weighed and mixed thoroughly and subsampled for later analyses. Urine was collected in plastic bottles containing 20 ml concentrated sulfuric acid and an aliquot sample of daily urine was saved for nitrogen (N) analysis. D. Preliminary Period The purpose of the preliminary period was to acclimate the lambs with the metabolism cages, make the necessary equipment adjustments to insure that the feces and urine were collected properly, and adjust the animal to its in- take of feed in relation to the gxcretion of feces and urine. A preliminary period of 14 days for each lamb was used to assure maximum consumption of each ration until the conditions of the experiment were met. E. Preparatory Treatment All lambs were shorn, vaccinated with Type D toxoid for enterotoxemia, drenched with Loxon for internal parasites, and all feet were trimmed. Rumen cannulas were inserted in each lamb 2 months in advance of the initial collection period. F. Collection Period The collection period for feces and urine ran for 7 consecutive days with the feed intake carefully measured. 55 Each afternoon before the collection was initiated, the cages and collection area was cleaned thoroughly. Each collection period began on the morning after the animal had been eating a constant amount of feed for the pre- liminary period. During the collection period, a random sample of the feed that was weighed out for feeding was saved for analysis. Feces and urine were removed from their containers, weighed, and stored in a freezer for .subsequent analysis. F. Rumen Fluidng, Rumen Ammonia, and Volatile Fatty Acids Rumen fluid samples were taken from each lamb at the end of each collection period and were analyzed for pH level by a Beckman Model 4500 pH meter. This rumen fluid was then strained through cheesecloth and samples were analyzed for iumen ammonia values in mg % on the Orion Ammonia Ion Electrode Model 95-10. VFA analysis was conducted using the Hewlett Packard 5840A Gas Chromatograph. G. Blood Urea Nitrogen Blood samples were collected in 10 ml heparinized vacutainers from the jugular vein of each of the lambs. The samples were then centrifuged at 3,000 rpm to separate plasma and cell contents and the plasma obtained was frozen. Urea nitrogen was determined using the Conway procedure (Conway, 1960). Conway dishes were prepared by adding 1 ml boric acid solution to their inner well and 1 ml of glycerol to the depression around the outside of the plate. Exactly .5 ml of the plasma was pipetted into the 56 one side of the outer well and then diluted with distilled water. A urease solution was added to the plate to convert the urea in the sample to NHS. After the enzyme reaction, K2 C03 was added to all the urease plates to release the ammonia. The plates were allowed to diffuse one hour on the rotator. They were then titrated, recorded, and calcu- lated in mg/100 ml. H. Statistical Analysesr" All of the data from the digestion study, N balance, and the measured rumen and blood parameters were analyzed for treatment differences by the Latin Square analysis of variance method on the Hewlett Packard 9825A. Separation of mean values was conducted using the Studentized range test found in Statistical Tables by (Rahlf and Sokal, 1974). RESULTS Harvests of Biomass Trees and Alfalfa. Table 4 lists the eleven species which were utilized in this study. Ten hardwood tree species were selected for their rapid growth rate and ability to regenerate after cutting. Alfalfa (Medicago sativa) was selected as a control for comparison with the tree species. Alfalfa was chosen for a control because it represented a commonly used feedstuff that is high in nutrient value and could serve as a standard for rating the biomass tree species. The approximate projected yields of alfalfa and biomass species are presented in Table 5. Values are given in both' tons/acre and metric tons/hectare with alfalfa yielding the highest values with 4.5 and 10.1 respectively. Based on dry matter weights after two harvests, values for the biomass trees ranged from a low of 4.0 for willow to 8.5 for poplar in metric tons/hectare. Four of the biomass species, aspen, black alder, black locust, and honey locust all yielded over 7.0 metric tons/hectare. Selected proximate analysis for biomass and alfalfa species are shown in Tables 6 and 7. Table 6 represents the early harvested samples while Table 7 exhibits the values for the same species harvested 8-9 weeks later. Field har- vested dry matters for the (early harvest) ranged from 27.49 57 58 Haws: xflflmm mopfiouaop mSHDQOQ monpcmomflsp mfimpflpoaw moficm>axmccom mseflxmsm mamoflsoam mafia: msowwpxmmd masuom «HOHOHOpSOmQ macflnom «snap maca< mopfloHZEosp mSHDQom mEHmmHuHm mszucmfifl< m>flpmm omwofipoz oEmz oawfipcoflom onHflz swamom pmsooq >oco: cm< :ooso Eam cupflm pmzooq xomam poca< Momam comm< mszucmafl< mm~mw~< oEmz :oEEou mmmmb mm H I\\OO\<1'LOO \OLDQ'BOOQ‘ N.n o.m 0.5 H.ofi osmpoo:\m:OH oasuoz moan: agape: m.< oHo<\maos peofie>flsem .m.: BOHHHZ HHHQOQ pmsooq xoco: cm< :ooso Sam cosflm pmsooq sowam Hopa< xomam comm< memmH< coupauucnpeoes Seeds Hmmmmk wm Gmkumhomm me<2~xommm< .m m4m

segues xgm n .mcoflpmcflasopop oopnu wo mowmso>m Ono Ohm mosam>m mm.m 4H.N Na.k Hm.ma am.fim zoflsflz Hm.4 SN.m NN.S NA.HN S4.am pafiaoa 04.4 wm.m Sm.m me.NN mm.wN Bazooa keno: Ho.4 4H.m 4N.w 4w.AH 4w.AN am< compo HH.4 aw.N 4N.e SN.SH ma.a~ eflm HN.4 Hfi.m ma.w am.m~ Hm.om pmsuoa guess Ho.4 Ho.s km.e mw.ON a4.AN goea< xomam ma.4 Nm.m MN.m Hm.mH NN.mN copam OH.4 4e.m Ha.e 44.SH Nw.aN soam< Hk.m Ho.m N.a 4N.AH ma.kN msaoeafia< 4S.4 44.m 40.H . mH.NN HN.SN HLHHCH< AEw\Hwozv w.uocswxm pocpm Mdmm¢ w.¢m®popm opssu pouumz xsa coflpmoHWHpcopH ucmfiz xwpocm mmoso n o pupae: Spa wo :oflpflmomEou w4mAm coupes >43 n .mCOflwdCHEHOHOU mmhfiu WO mmeOK/N 03”.. ohm. WQSHw>w e4.m N4.N 54.x Sw.a SN.Nm 204443 44.4 4m.N Nm.e 44.aH mN.om aaaaoa NH.4 Hfi.m H4.e MN.oN 4N.4N omsooa some: Ha.m 4m.N M4.w 44.mfl wa.am am< goose ma.m mm.N HN.A a4.mfi N4.om 54m 4N.4 Ho.m 4w.m Hm.NN ka.am SHSOOS xuaam mm.m m4.m Nw.o 4k.wfi NN.Nm coea< guaam HN.4 HH.4 Nm.m HH.mH N4.Hm gugnm 40.4 wo.m N4.A M4.AH HH.aN coam< m4.m Hm.N Hw.k mm.efi 4N.4N m2:5:44a< mm.4 ak.N NN.w w4.om Nw.aN HLHHGH< Aew\awoxv w.pomsuxm Hocum w.£m< 4.:flouosm condo coupon: >4: OHOHHHOHWHucopH “swam xmsocm mmosw . . hopumz age we :oflpflmomaou mqmn .mcoHHHCHEpouow oossu mo owmpo>m ozu can mosfim>m mw.w4 m4.mH mm.mfi m.om wm.mN Ho.©H mm.ma NH.NH No.54 mm.HN an.wfi 4.3moflsfifioo mo.om Hw.MH 44.04 Hm.mH Hm.ma N4.NH m4.ma mm.HN mm.m4 4N.NH HN.4H w.om0H:HHoowEo: m4.oH 4m.m ©4.m H4.w NH.oH 44.5 mH.w Hm.w NN.m NH.NH HH.m 4:2ch3 4m.mm mn.wm mm.mm HH.@N m.mm No.4m w4.mm mN.om 4H.BN om.mm Nw.mN m4mm¢ H4.m4 4m.N4 mw.mm No.44 Hw.4m 44.44 mc.m4 ©4.m4 mm.N4 4H.o4 mm.wm wammm wmsooq :m< umsooq poea< mg: .m. 304442 HHHQOQ xoco: compo Efim Momam xomam cosflm comm< uceflfl< mammH< neoflpauacaueoes Omega 48mm>mAmqD .ma04um:HE4Opop oo4cp wo om44o>m 0:4 O44 mo344>m 65 40.44 44.44 44.44 44.44 04.04 304443 44.44 44.44 44.44 44.44 40.44 444000 44.44 44.44 44.0 44.44 40.44 443604 40:0: 04.44 04.44 44.44 44.44 44.44 :44 eooco 40.44 40.44 04.44 44.04 44.44 E44 44.44 04.44 44.4 40.44 44.44 46444 40.04 44.44 4440 40.44 40.44 40044 40444 44.44 40.44 40.0 04.44 04.44 40444 44.44 40.44 44.04 44.44 44.44 c004< 40.44 44.44 44.44 04.04 44.44 4244:4444 44.44 44.44 44.4 40.04 44.44 044444< 4040424400 4cm04344oo45o: 4.:4c444 4.00< 4.442 :0444O4444e004 4:440 44mm>4000: :m< :0040 54m 445000 xo04m 4004< Momam :O440 comm< mzsucmH4< «Wamma< 00444044400004 40440 4440>0<= 440400 00 :mz4zzmhmn mEZmzomzou mm04=44m042mz .44 mqmm ozu ope mosam>m Hm.m am.m 4H.a HV.NH 5H.Nq aoHHHZ 9N.4 mH.m om.o wo.HN om.mm eaaaoa NH.4 He.m HN.© ao.NN S¢.Hm omsoos keno: mw.m 4a.N 4m.a Hm.afi om.Nm zm< compo Ha.m m4.m afl.a mq.mH am.aq Sam MN.4 Hm.m mw.a GH.MN mN.Hm omsooq Mamam aa.m QH.¢ HN.a 44.0N HH.¢m goefi< goasm mo.q Nm.¢ HH.m 55.4H qw.om copflm aw.m Hm.m Na.o afl.oa HN.¢m cogm< mo.m 4m.N qm.o mm.wfi NN.Hm msaocafifl< oq.¢ mm.m H¢.a ew.NN Ha.mm meaWH< me\HmUv: wawumhpxm hmfium a waCHou‘Ohm GVShU hopumz \Akfl DSOMHQUMprCGUH waHQ zmgocm mmogo Houumz kg: mo :ofluflmomfioo mq n .mcoflumcfiehouov mossy wo ommgo>m esp own mosam>1 om.w~ 5.5H w.mH mm.w~ HB.¢N mm.ma HN.@H Nu.mfi om.aH mm.oH N.mH $.0mofisfifiou om.oH NH.NH mo.oH m.mH mm.AH m.oa ofi.mfi w.o~ am.HH 0H.OH a.mfi w.omoH:HHoofiEo: HH.NH ow.m v0.0 no.9 ma.ma v0.5 NN.© wo.w HN.@ wa.ma mm.c a.:a:mfls oa.am ©m.mm vw.mm Nv.mN ww.om om.om mq.wm mm.vw mm.w~ om.om m¢.HN a oo.m¢ mo.mm ow.om No.0v ov.vm om.mw mm.mv ma.mv «v.0v mo.mv nv.om a zofifluz. anamom pmsooq zoco: cm< :oonu Eam pmsooq Momam nowa< xomam mmmmm& mm<20~m Qz< qa oz» ohm mo:_a>m m.cm a.om a.am a.mm N.am a.om em.m~ o.a~ a.~v n.4m w.oQ .z Magoo co a .AZmzzv-Azmv ac.a mm.“ eo.m __.m mc.e c.¢ m~.w a.o o¢.m am.v m.c .m swoon do a .z-m:z e.No a.mm _.¢m a.oc m.mm m.mm o_.mo c.4o a.mv ~.~o m.~m .mgdmooo do a .2 o~£s~0m:_ ”v.5m 2H.ve a.mv a~.am 5.2? cm.H¢ om.sm om.mm vfl.sm Na.mm N_.AV .m fimuou do a .z assaiom cc. NN.1 _¢._ am. No. am.~ AA. we. ~m.u No.2 a¢.~ « .Az-m:zi-azmv 4.. 4N. .N. o_. _1. 51. AN. as. SN. nu. mN. a .z-m:z TN._ mx._ _a._ 3a.~ 4¢._ ai.N mN.N Nm.i cm.s Nm.~ mm._ a .m o~s3~0m=_ :5. cv._ Nc._ c_._ mc._ cm.~ so._ em. am.“ m_.~ Na.~ a .m umns_om _¢.N_ ac._~ mo.NN ~m.al M¢.m_ o..m~ ce.¢~ an.q. a~.as mm.ms va.- * .cuoooua canto ma._ o<.m mm.m ca.N a¢.~ ia.m a~.m om.N ao.m aa.~ mo.m a .m sauce mygsam “daumm mmmwmm .anw- mam. mmmmmm mmmw¢ mmmmm mmmma mmmmmMMMQ mmmmmmw ammm xo:o: :oopc xocH: xum_m amzzzé mmm osu one mozaw>£ .Houpme >96 we w a we woupomoa ohm momxfimcm HH ago.v;y u:oa0uuflc xfiuccuMgflcmwm ohm mumwpumpoazm ucoaouaws :wflz mosfim> =QEAH«V‘AHAWQ~_AM WnQAvAUufi—n—w em.oN Hm.vm mm.mv mm.om em.- Nc.~v mfi.~m a.mm ”a.mm mw.o~ mo.~m «N ¥.mv .H.M.U w.m m.U.£ Mg¥-u Sun— H.U .H.M.U .H.M.U.a— H.~u M.w no.m_ Hnme mm.mm o.~m m.v~ hm.mm hm.m~ -.c~ mo.n~ -.mfi ~m.~e NH fied ; a m; .331 .36 .2 . a... :5... a... .3. a 5v.v~ N~.QN m.o~ n~.- mq.m on.m~ oo.w~ m~.<~ e~.-. em.- mo.mN o 3.6.8 1;... .1»... in... a... flaié Md... a... fig... 5.6.6 w... mmflmu~< ace moose mmmeofim mo umo>umz open a.mN N.m maamm ~o.wm .m.n~ mm.ov N . m ~m.mm fl . m we.o~ mv.wm em x.e _mfl.omn W.“ ..u.n x.e ;.a Magma n.e.u Magma x.e M.“ ~v.- me.m~ o . m am.om ~m.n~ N.e . ~.H ~v.c~ .m mw.- no.me - E.M.O ¥.H.©.U Swn$fl R.H.—v.0 C.....— .mwfi.0mn~ HWfisUmfi Hsv—«Qg—u v—Ww.mku E.~.W.O M.m mm.h~ souv~ m«.o~ owns“ “NMVH m«.a~ mw.wN @m.m~ o~.m~ ~o.m~ va.~m c ~.o x n o.n : a m a 2 a a E o g a a x c a a «.m.m a.m.o.a H.o.m.o w.m zofiafiz “name; umsooa xozo: :w< :oouu E~ umsooa xowfim pov~< xomam soumm :omm< mssucm~m< amuawa< sewuwnso:_ . mo muse: . ma~mu~< vca moohh mmmEOMm mo umo>umz zgpmm awoumasocw moo~\uv mammhw no zmzax mze z_ :mkz 20:; muz¢a .mH mqm ~.x.a.~.;.m AHc.v;v acououuww xfiucmufim«:mfim mum muQMHUmuoasm asepowwwv nu“: mozfim> a.m.c.o.a.1 .v .mm me. e N .mm mo.~N N . v N .o« No.om «N.mm em.cN em.mm «N wax.w N¢.a ;.nmm m.n ~.o :mmam “Mn “.9 m.w.n .x.o m.m . om.m N . m o .oN am.o~ a .vm m.mm HN.oN ~m.¢m wN.NN ca.me NH . . . . a . H VN.mH . N o.mN mm.vN mN.mN ne.m~ e.wN o MM.W— mwumM— p—NW.MV~NQ Mhuwfl .m.U “MM¢Q mmm Sgnaw £.£.m M.U w.w mwfimwwé. 6:5 mommy? mm~w=_o..nm uo um®>.~m__ QHGJ NNMwN mmuve mmuwm mmumv wm.mN NN.Nm we.~e «$.am w~.mv m._m w~.mc «N V. e a n ._ a 2.. 3.... z... “a... m6 a a... V_ 6.6 .m m.m o.m om.me m.~ me.ON . ~.vm m.m m.Nm No.9N mo.we NH meauwo HNQNVNU Saw flwflchU Haw “MAM“ MQUQD “Mes—UWU .HQMAU XoW-Qq—V Mam ce.mfi «N._N mo.~m No.mN mm.N~ cv.mm cm.mN m.NN we.~N caumm mm.em o 5.9.m a.m.o ~.c.o.m u.u.m>. E.m a.m.m m.a.m.m x.muo.n.m N.w.o E.H u o m onHMz and on umzooq Noco: :m< :oouo sfim umsooa xoe~m uow~< xommm :ohflm :omm< mscucm~fl< mwfimwfi< :ofiumasocH mo muse: mmfimma< cam moo“? mmmEONm mo umo>hm: xakmm Avouwnsucw woom\mv mammkm no zmzzz mzk ZH amh2 20¢; muz 66o.v;. 6:6606666 >6ucaumufizu6m ohm muz66om6oazm 6:066666c £663 moz6a> H.Xqfl.d..-.nw “Avg—UQUAQAG 6 .6 6.6 6 . 6 6 . 6 .6 6.666..66 66.66 66.66 66.6 66.66 66 “.0 flwc.U.—6A~ “mums. Aw .fi.M.—.v «U; H.0M—VCU n—“mqnfi Haw fl; .5.— M.:.U.D.‘w M.U.U.U ¥.Q.—d ”.3 .6.6 66.6 wm.6 66. 6 66.6 6w} H66 .66 66.66 66.66 66.6 66.66 66 6.6 6.6 6 .6 6.6 6. 6 6 6 6.6 6 6 6. 6 6 . 6. .6. 6 6.6.6.6.6 6.6.6.6 6.6.6 6.6 66.6 .6.6 66.66 66. 6 66.6 .6. 6. 666.6 66.6 66.6 66.6 66.66 6 6 6.6.6 6 6.6.6 6.6.6.6 .6 6 6 6...6.6.6 . .6. 6. 6 6.6.6.6.6.6 6.6.6 6.6.6 6.6 6.6 . ahdmm~< cca much? mm6506m ho umo>6az 0666 6*... 66.66 6: 66.66 66.66 m6..6 66.66 66.66 66.66 66.66 66.66 66 6 ..6 6 6 6.6 6.6.6 .6 6.6.6.6 6.6 6.6.6.6 .6 6 6.6.6.6.6 6.6.6.6.666.6.6.6.6.6 6.6.6 6.6 66.6 66.6. 6... 6666 66.6. 6.66 6m.66 66.66 6.66 66.6. 66. 6 6 “66 6. ..6 6 6.6 6.6 6. 6.6.6.6 6.6.6.6 6. .6.6 6.6.6.6 6.6.6.6 6.6.6.6 6.6.6.6 . 6 .6. 66.6 .6. 66 66.6 66.6 6m.66 . 66.66 66.66 66.6 66. 6 6mm66 6 — 1 U ~1U.Q.—w n— M . £.1.U.D.H~ H..—.V.U :. .£.¢w p—.U.£..—w ——.U.fi~.._w «.£.—J.U.D ..Hv A... 20666: 6:mo; umsoog cho: :m< cooho E6: umsooa xuwflm hoca< xumHQ :666m :omm< mszu:666< MMme6< :o6uagsoz6 - auamw~< wzc moo.6 mmcso«= mo umo>6c: >6amm .o muse: 6666666666 6666\66 666666 66 26266 666 z. 666<66626 66<6 26662 666. 662<6<666<666 .66 .66 66666 92 illow to 17.31 for black locust. In the 24 hour sampling, the biomass trees ranged from 10.25 for ailanthus to 21.64 for black locust with an overall average of 15.07 for the biomass spcies. Alfalfa (23.01) was highest overall but not significant differenct was shown between it and black locust (21.64) and honey locust (20.6). Alfalfa and black locust were significantly higher (P<.01) than aspen (13.85), willow (11.17) and ailanthus (10.25). Alfalfa, black locust and honey locust were significantly higher (P<.05) than green ash (16.46), birch (15.44), black alder (14.72), elm (14.29) and poplar (14.29). Utilizing the late harvest samples, the six hour bio- mass samples averaged 7.21 compared to the early harvest at 9.03. At six hours, alfalfa was the highest value over- all at 13.43. This value was significantly higher (P<.01) than birch (8.21), aspen (8.15), green ash (7.94), poplar (7.81), black alder (7.04), elm (6.04), willow (3.43) and ailanthus (2.48). The range for the biomass trees ranged from a low of 2.48 for ailanthus up to 11.61 for black locust. Black locust and honey locust were significantly greater (P<.05) than the other biomass species tested. Within the twelve hour period, alfalfa, black locust and honey locust displayed no significant differences. The values for the biomass species ranged from 3.81 for willow to 15.47 for black locust with an average of 9.50 compared to 12.12 for the early harvest. Alfalfa was significantly higher than aspen (10.33), black alder (10.22), poplar (9.14), green ash (8.39), elm (7.21), ailanthus (4.11) and 93 willow (3.81). Within the trees, black locust and honey locust were significantly greater (P<.05) than aspen and black alder. At 24 hours, the extent of ADP disappearance ranged from 6.28 for willow up to 16.58 for honey locust. Alfalfa was again highest overall at 18.97. This was significantly greater (P<.01) than black alder (12.10), green ash (11.81), poplar (11.64), aspen (10.56), elm (10.38), ailanthus (7.04) and willow (6.28). No signifi- cant difference was seen between alfalfa and black locust (16.58) and honey locust (16.07). The overall average for the biomass late harvest species was 11.71 compared to 15.07 for the early harvest. Feeding Trial Results - The chemical composition of the four diets fed during the lamb feeding trial are found in Table 21. All of the diets were pelleted and their dry matters ranged from 93.71% for (100% poplar) to 94.12% for the (66.6% poplar - 33.3% alfalfa) ration. Ash values ranged from 6.87% for (100% poplar) to 7.42 for (100% alfalfa). Crude protein values were similar in value and ranged from 18.21% for (100% poplar) to 19.71% for (100% alfalfa). Ether extract was highest for (100% alfalfa) at 3.57% and lowest for (100% poplar) at 3.14%. Gross energy values in Kcal/gm ranged from 4.16 for 100% poplar to 4.53 for 100% alfalfa. The fiber analysis of the four treatment diets is found in Table 22. % NDF ranged from 39.12% for (100% alfalfa) to 43.92% for (100% poplar). The range for ADP 94 .mflmmg hoppme xgw a :o momzfimcm ohm xmhoco mmoym can pomppxo gocpo .cflouo~m owspo .cmm :o mucoEopsmmoE pcoscomQSm - wcflpoaaom popwm :oxmu mm: poppms xamm 0H.4 Am.¢ m¢.q mm.¢ EM\Hauz .smgocm mmOpo efi.m mm.m oe.m Am.m w .oomuoxm pocom HN.wH VA.wH mm.wfi HA.mH w .cflopogd oesgu aw.o qw.o mN.a N4.A a .:m< Ha.mm NH.4@ Nm.mm 40.4a w .quomz ska flwoofiv wm.mmv wm.mmv. HwOOHV Emou guano; mwfimofi< amazon «wamwfi< flwo.oov flwo.oov Hmfimom mwfimwfi< mmHmHQ sz29 a ,mcoflpmcflehowow oopzu wo owm~o>m map opw mozam>m am.o~ Ho.ofi AM.NH Hm.mm Na.mq Awooav seamed Hm.mH oA.HH Hw.OH NH.om mw.fle Awm.mmv mmfimwfl< Awo.oov pmamom OA.QN m.HH ao.m qu.@m 40.H4 Awm.mmv anamom flwo.oov moammfi< Hw.ON mq.fifi Aw.o no.5N Nfi.am AwQOHV awfimwfi< w.omofizfiaou omofizfifiooHEo: w.:flcmag w.am< w.aaz ncoflomuHWHucoeH boas m44 97 £aMnWa® mfio.vmv pcogowmfiw xfipcwoflmficmflm ohm mpmflhomgomzm pcopowmfiw :uflz moSHw> w.o.n.m me.mN CAH.mN om4.mm mam.Nm camped z Hapoo do w .coflpcopom m MHM.A m.Wmo.a o.oNA.HH ,oNH.NH sme\m .coflpcouoz z :ofiuCopom :omoggfiz ;.DHN om m.n4mflam o.mqo so o.a ouqo m smumcm m.nAN mm w34.3 mm o.n.mmm a4 o.aam m4 gonna use wagon efiu< w.nHH.Nm w5.1.93 o.n.mmfi.am o.amN.No :Hoooga mango new.mm m.mNo.No m.a¢m.mo a.mow.ao Hopoaz spa Hwy sewaflnaomomae ocogaaa< wma.oofi ov¢.mma nom.mmm www.mmm Hmox axmwocm manfiumomflm omm.mam nmfi.mom w.wvm.mwm o.mmm.omm Hmox .xwgocm mmOho mm.¢N oa.mN Hmnmm BN.VN ponflm ucowpopom wflo< woa.mfi w.omo.mH mav ca . www.ma :fiopoym evapu nHw.NA w.mam.mw a.mmm.ow a.mqa.aw 2866a: spa flmA.Oa:1mx\mo oxaocfl oeoaposz .4-flwocfiv m -Awm.mmv N -nwm.mmv H -flwooflm soau anamoa mofimmfi< Hmfidoa mofimofl< Awo.oov flwo.oov teamed mwfimoa< mmHHHAHmHHmmuHQ Hzmmmq mDom quQmmm mo Hommmm .mm mqm<9 98 coefficient times the gross energy intake was highest for diet 1 at 255.78 kilocalories. This value was signifi- cantly higher (P<.01) than the other three rations in the trial. Diets 2, 3, and 4 were also all significantly different from each other at (P<.01). Apparent digestibility coefficients are also shown in Table 23. Dry matter digestibility ranged from 53.86% for diet 4 up to 69.86 for diet 1. Diet 1 was significantly greater (P<.01) than diet 4 and also greater (P<.05) than diets 2 and 3 at 65.54% and 62.02% respectively. Crude protein digestibility was also highest for diet 1 at 62.23% and was significantly greater (P<.01) than diet 4 at 53.86%. Ration 1 was also significantly higher (P<.05) than ration 3 at 44.04%. Acid detergent fiber had the lowest overall values for the digestibility coefficients‘measured.- Values ranged from 35.27% for diet 4 up to 45.89% for diet 1. Diet 1 was significantly greater (P<.01) than diet 4 and at (P<.05) than diet 3 at 38.28%. No significant difference was seen between diets 1 and 2 or 3 and 4. The coefficient of digestible energy was highest for diet 1 at 64.6% com- pared with the lowest value for diet 4 at 50.21%. Overall diet 1 was significantly higher (P<.01) than diet 3 (54.34%) and diet 4. There was also significant differences (P<.05) between each of the treatment means. Nitrogen retention and retention as a percentage of the total intake are also shown in Table 23. Nitrogen retention in grams/day was highest for diet 1 at 12.12 and ranged down to 7.31 for diet 1. Diet 1 was significantly 99 higher (P<.05) than ration 4 at 7.31 and ration 3 at 9.65. There was no significant difference between diets 1 and 2 or 3 and 4. The nitrogen retention as a percentage of total nitrogen intake ranged from 25.31% for diet 4 up to 32.37% for diet 1. Diets 1 and 2 were significantly greater (P<.05) than both rations 3 and 4 at 25.17% and 25.31% respectively. The values for rumen pH, rumen NH3, and blood urea nitrogen values are shown in Table 24. Rumen pH values ranged from 6.23 for diet 1 up to 6.77 for diet 4. No significant differences were found between any of the treat- ment means. Blooc urea nitrogen values were highest for ration 1 at 24.21 and lowest for diet 4 at 22.87. No significant differences were reported. The values for rumen NHS ranged from 11.43% for diet 4 up to 12.13% for alfalfa. Again, no significant differences were found between treatment means. Rumen fluid organic acid composition was shown in Table 25. Values are given for acetic, propionic, butyric and valeric acid. Values for acetic acid ranged from 10.32 for ration 4 up to 11.35 for ration 1. No significant differences were found between treatment means. Propionic acid values ranged from 3.31 for diet 4 up to 5.20 for diet 1. These values were different significantly (P<.05). Values for diets 2 and 3 were not significantly different than both diets 1 and 4. Butyric acid values ranged from 1.92 for diet 4 up to 2.12 for diet 1. No significant differences were found between treatment means. Valeric acid values were the lowest organic acid values measured. 100 .mcwoe psoEumogp :oo3uon moocogommfiw ucmoflwflcmflm oz N .wcflwoow popwm .M: N onmEmm who: mosflm> HH zmoomHHz n.a flHE ooa\oHoEEv mm wommopmxo ohm monam>m HH. ma. NH. ea. oflpofiw> Nm.H Hw.H No.N NH.N oflpxpsm nam.m nmmv.w nano.v woN.m oflcoflmo~d Nm.OH Hw.oH ow.oH mm.HH oflpoo< -flwOOHV m -flwm.mmv m -Awm.mmv - “wooflv .awbH pmamom wmewH< pmamom meme< ”$0.001 flwo.oov tandem mmamwfi< wonHHmomzou QHQ< qu