CE Cm STLAGE mum TEmmCUEs Am $TORAGE FACTORS RM TERMENTATTGM PARAMETERS """" ATTC CATTLE PERFORMANCE """" Thesis for The Degree 0T Ph D H MICHIGAN STATE umvsasm » MITCHELL RAY GEASLER T . 1970 LIBRARY {Hesu Michigan State University This is to certifg that the thesis entitled The Effect of Corn Silage Maturity, Harvesting Techniques and Storage Factors on Fermentation Parameters and Cattle Performance presented In] Mitchell Ray Geasler has been accepted towards fulfillment of the requirements for Ph.D. Animal Husbandry degree in ’ F»: 95;? E- ajor professor Date May 14, 1970 0-169 STORAG F01 Lid metabl Lhair rel smrage f In isekly ir increaset ensiled : airtight Mssure lechanis Simples Me maj the gem Silage l M M ( ABSTRACT THE EFFECT OF CORN SILAGE MATURITY, HARVESTING TECHNIQUES AND STORAGE FACTORS ON FERMENTATION PARAMETERS AND CATTLE PERFORMANCE BY Mitchell Ray Geasler Four experiments involving silo fermentation, animal performance and metabolic parameters in the ruminant were conducted to investigate their relationship to corn silage maturity, harvesting techniques and storage factors. In the Fermentation Study, ten corn silage harvests were made at weekly intervals from September 3 to November 5 at dry matters which increased significantly (P < .01) from 22.1% to 48,3%. Each harvest was ensiled in four 12" x 18” stainless steel experimental silos within an airtight Plexiglas chamber. The chamber and/or silos were equipped with pressure application and measuring devices, temperature measuring‘ mechanisms, total seepage collection and apparatus for taking daily samples from each unit° Four different pressures (0, 2.5, S and 10 psi) were maintained for a fermentation period of 12 days° As dry matter of the corn plant and/or maturity increased, lactic acid of the resulting silage was significantly reduced from a high of 5.8% to a low of\2.2% of DM (P < .01). Total nitrogen was significantly reduced from a high 1.3% of DM to a low of 1,1% of DM (P‘< .01). Water soluble nitrogen, expressed as a per cent of total nitrogen, dropped from a high of 32.2% A-_.E -a . w"... T61“ of Tsilage f< Thread gothS of plied. T ifectedby affementz In] silage on; Tlativet ET on Se [39.6% DT Cctober 15 'Tithigan finest dT M5 were MZSJT MM]? fiductior Mllctim from the TM harvest and the hMest ledium ‘ Mitchell Ray Geasler to a low of 15.3%. Seeplage decreased from a high of 16.90 ml./100 gm. of silage for the 22.1% DM silage to zero, irrespective of pressure, after the DM reached 34.7%. Pressure was apparently without effect on the end products of silage fermentation, as long as a minimum of~2.5 psiwas applied. The pattern of fermentation during the 12 day period was not affected by maturity and/or dry matter content; however, extensiveness of fermentation was markedly affected. In Feeding Trials 1 and 2, the effect of stage of maturity ofvcorn silage on yield per acre was investigated. Fineness of chop was also studied relative to dry matter stored per cubic foot of silo capacity. Harvests made on September 13 (28.2% DM), October 17 (48.2% DM) and November 14 (59.6% DM), 1966 and September 18 (30.7% DM), October 5 (34.7% DM) and October 19 (43.3% DM), 1967 were compared. In both years a 40-acre field of ”Michigan400" corn was initially divided into eight-row plots. At each , harvest date, two rows were harvested from each plot. The remaining two rows were picked to accurately determine grain yield per acre. In 1966, the 28.2% DM harvested silage yielded 5.11 tons of DM per acre, whereas the 48.2% DM silage yielded 4.57 tons per acre. This represented a reduction of 10.6%. The 59.676 DM silage yielded 4.06 tons per acre, a reduction of 20.5% from the 28.2% DM harvest, and a reduction of 11.0% from the 48.2% DM harvest. Dry matter yield peracre on the three dates in 1967 was-5.64, 5.86, and 5.56 tons, respectively, betweenthe 30.7% DM harvest andthe 34.7% DM harvest, a decrease of 1.41% between the first andthe last harvest. Two identical silos were filled each year for each harvest date, onewith fine chop silage (1/4 in.) and the other with medium chop silage (1/2 in. to 3/4 in.). Dry matter stored per cubic matter Ts. 12.13 I harvests, ‘ nest da 115 11.93 MTV. 1 Thad eat :aits th Mitchell Ray Geasler foot of silo capacity in 1966 was-13.40 1b. vs. 11.14 1b.; 11.99 lb. vs. $96 1b.; and 11.93 lb. vs. 10.97 lb. for fine and medium chop in the 2&2% DM, 48.2% DM and 59.6% DM harvested silages, respectively. In 1967, drynmtter stored in each silo was 12.32 lb. vs. 11.55 1b. and 13.15 1b. v5.12.l3 lb. per cubic foot of silo space for the 30.7% DM and 43.3% DM harvests, respectively. Combining fine and medium chop for each of the harvest dates in 1967, dry matter stored per cubic foot of silo capacity was 11.93 1b. vs. 12.64 lb. for the 30.7% DM and 43.3% DM harvests, respec— tively. In the fall of 1966, a 3 x 2 x 2 factorial experiment (12 lots of 9 head each) was initiated to study steer calf performance and carcass traits when fed the corn.silage harvested in the fall of 1966. Cattle fm128.2% harvested silage significantly (P 41.05) outgained_the 48.2% DM fed group (2.87 vs. 2.70 lb./day) and the 59.6% DM group (2.87 vs. 2.74 lb./day) but the 48.2% DM fed group was not significantly different from um 59.6% DM fed group. Carcasses from the 28.2% DM fed group were signifi- cantly superior to the 48.2% DM and 59.6% DM groups for all factors deter- mhung cutability. Pooled differences comparing fine and medium chop silage were small and insignificant; however, average daily gain values of 2.81 lb. for fine chop and 2.72 lb. for the medium chop silage approached significance (P <;.10). Within harvest dates, average daily gain was 2.89 lb. vs. 2.85 1b.; 2.78 lb. vs. 2.63 lb. and 2.78 lb. vs. 2.69 lb for fine and medium chop silage harvested at 28.2% DM, 48.2% DM and 59.6% BM in September, October and November, respectively. For 48.2% DM and 59.6% DM harvested silages, both fine and medium chop were fed as ensiled vs. regrinding just prior to feeding. No difference in average daily gain (2.72 lb. vs. 2.73 lb.) resulted; however, pounds of feed consumed per 100 pounds of gain favored the reground fed group (7.14 lb. vs. 6.84 lb.). “x 2 11 T196749 The com S mested Tharves1 slightly . intake), ‘ gain (6.6 lei group shied nc tattle fe grading ( Mitchell Ray Geasler A 2 x-2 x 2 x-2 factorial experiment (16 lots of 8 head each) was employed fl11967-1968 to study steer calf performance and carcass.traits when fed the corn silage harvested in the fall of 1967. Cattle fed the 30.7% DM harvested silage significantly (P <..05) outgained the group fed the 43.3% DM harvested silage (2.58 lb. vs. 2.46 1b.). This, coupled with a. slightly lower daily dry matter consumption (17.27 lb. vs. 17.62 1b. DM_ intake), resulted in a substantially lower feed requirement per pound of ‘gain (6.69 lb. vs. 7.16 lb. of 85% DM) in favor of the 30.7% DM silage fed group. Pooled results comparing the fine and medium chop silage showed no significant differences in animal performance. However, the cattle fed the fine chop silage produced significantly (P-< .05) higher .grading carcasses (high Good vs. middle Good). The Metabolic Study was conducted in the fall of 1967 and involved eight fistulated wether lambs to test various metabolic parameters in a 2 x.2 (two stages of maturity x two degrees of chop) factorially designed study. The 30.7% DM harvested silage was consistently lower in dry matter and nitrogen.digestibility, which resulted in a lower nitrogen retention. lbwever, all values were nonsignificant. Stage of maturity had no effect m1rumen volatile fatty acid.production and voluntary feed intake. The lambs fed fine chopped silage had nonsignificantly greater rumen VFA levels compared with the medium chop silage fed group. Lambs fed the fine chop silage had a significantly (F’<: .05) higher voluntary feed intake than those fed the fine chop (819.73 vs. 583.15 gm./day).. Dry matter digestibility, although nonsignificant, was lower for the fine chop silage (65.95% vs. 69.15%). THE EFFECT OF CORN SILAGE MATURITY, HARVESTING TECHNIQUES AND STORAGE FACTORS ON FERMENTATION PARAMETERS AND CATTLE PERFORMANCE by Mitchell Ray Geasler A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Animal Husbandry 1970 'IM‘JOT lissertat wtline c liograph iTperiel Timber: W—i /0-9»'70 Dissertation: Mitchell Ray Geasler candidate_for the degree of Doctor of Philosophy The Effect of Corn Silage Maturity, Harvesting Techniques and Storage Factors on Fermentation Parameters and Cattle Performance Outline of Studies: Major area: Animal Husbandry (Ruminant Nutrition) Minor subjects: Biochemistry and Extension Personnel Development Biographical Items: Experience: Member: Born: December 5, 1939: Lake, Michigan Undergraduate Studies: Michigan State University, 1958 - 1962 Graduate Studies: Michigan State University, 1966 - 1970 4-H Youth Agent, Cooperative Extension Service, Michigan State University, 1962 - 1966 Graduate Teaching Assistant, Michigan State University, 1966 - I970 American Society of Animal Science Society of Sigma Xi American Dairy Science Association FarmHouse Fraternity ii Thl lls advic His encou 111 mittee 111 Dr. l The w: it. llem My tec T loefer 1 lirligar 1 assistal his wif M the ACKNOWLEDGMENT The author extends his appreciation to Dr. Hugh E. Henderson for his advice, guidance and patient counsel throughout his graduate program. 1H5 encouragement and enthusiasm have been greatly appreciated. The author is further indebted to the other members of his graduate committee, Dr. J. T. Huber, Dr. Richard W. Luecke, Dr. Mason E. Miller, and Dr. D. Barrie Pursen for their sound advice and willing participation hithe writer's graduate program. Appreciation is also extended to Dr. Werner G. Bergen for his assistance in the development of the labora— tory techniques used in this research program. The author also wishes to thank Dr. Ronald H. Nelson and Dr. J. A. Hoefer for making the facilities of Michigan State University and the Michigan Agricultural Experiment Station available for this research. Appreciation is also extended to Mrs. Rodney Preuss for her assistance.and typing of this manuscript. The writer extends his sincere gratitude to his parents and to_ his wife, Margie, for their continued interest and encouragement through- out the author's career. list of 1 fist of I List of I I, In [1. Li TABLE OF CONTENTS List of Tables List of Figures List of Appendix Tables I. Introduction II. Literature Review Silage Fermentation Carbohydrate Fermentation Protein Breakdown. Corn Silage Maturity Summary‘ III. Materials and Methods Experiment 1 — Silage Fermentation Study Design Silage Sampling and Data Collection Silage Analysis Experiment 2 - Feeding Trial 1 Harvesting of Silage Feeding'Trial Experiment 3 - Feeding Trial 2 Harvesting of Silage Feeding‘Trial Experiment 4 - Metabolic Study Design Feeding Regime Sample collection Laboratory Analysis Statistical Analysis IV. Results and Discussion Experiment 1 - Silage Fermentation Study ’ Corn Silage Maturity Dry Matter Seepage Volume Silage pH Soluble Carbohydrate iv Page vi ix xi 13 17 32 33 33 38 38 39 43 43 44 47 48 50 50 51 52 53 53 53 55 61 Iable of C Wendi Appe Awe Appe be Apps Table ofContents (Cont.) Page Acetic Acid 61 Lactic Acid 64 Total Nitrogen. 66 Water Soluble.Nitrogen. 66 Water Soluble Nonprotein Nitrogen 69 Ammonia NitrOgen 72 Correlation_Coefficients of Fermentation Parameters. 72 Fermentation by Days 72 Interactions Between Stage of.Maturity and Rate of Fermentation 78 Silo Pressure 80 ExPeriment 2:- Feeding_Tria1 1 85 Chemical Analysis of Silagel 85 1 Dry Matter Yield and Silo Storage_Requirements- 90 Feeding Value of Mid-September vs. Mid-October vs. Mid—November Harvested Corn Silage 93 Fine vs. Medium Chop Silage ' 97 Reground vs. As Ensiled Feeding 97 Experiment 3.- Feeding Trial 2 ‘ 104 Dry Matter Yield per Acre and Silo Storage Capacity ‘ 109 Mid-September vs. Mid-October Harvested Corn Silage 112 Fine vs. Medium Chop Silage. 115 Experiment 4 - Metabolic Study» 115 Rumen pH and VFA Concentrations 115 Dry Matter Intake and Dry Matter Digestibility 119 Nitrogen Balance‘ ' 122 Correlation Coefficients 124 V. Summary 125 Bibliography 131 Appendices 140 Appendix I - Sample Calculation 140 Appendix II' - Design of Experiments 141 Appendix III - Raw Data 144 Correlation Coefficients 155 158 Appendix IV - Appendix V - Verification of Dry Matter Determinations LIST OF TABLES Table Page 1 Annett and Russel - Silage Characterization 6 2 King — Unavoidable Losses in the Silo 12 3 MSU 64% Supplement Formula 46 4 MSU 64-67 Supplement Formula 49 S Seepage Parameters 57 6 Simple Correlations — Fermentation Study 60 7 Mean Silage Parameters Relative to the Progress of Fermentation‘ 74 8 Mean pH and Deviations From the Mean Involved in the Interaction of Stage of Maturity and Process of. Fermentation. 79 9 Mean Lactic Acid Value and Deviations from the Mean. Involved in the Interaction of-Stage of Maturity and. Process of Fermentation 81 10 Mean Silage Parameters Relative to Silo Pressures 82 ll Mean.Temperatures Expressed as Deviations from Ambient. Temperatures Relative to Silo Pressure 84 12 Weather Data During the 1966 Harvest 36 13 Mean Silage Parameters Relating Fresh and Ensiled Materials used in Experiment 21- Feeding Trial 1, 87 14 Silage Parameters Relative to Stage of Maturity and Fineness of Ch0p used in Experiment 2 - Feeding Trial 1 88 15 Silage Parameters Relative to Stage.of Maturity and Fineness of Chop used in Experiment 2.- Feeding Trial 1 39 6 Effect of Stage of Maturity and Fineness of Chop on Dry 91 Matter and Dry Matter Yield.per Acre vi f Tables (Cont.) Page Effect of Stage of Maturity and Fineness of Chop on Silo Storage Requirements 92 Effect of Harvest Date on Rate of Gain and Feed. Efficiency 94 Effect of Harvest Date on Carcass Quality 95 Effect of Fine vs. Medium Chopped Corn Silage on Rate of Gain and Feed Efficiency 93 Effect of Fine vs. Medium Chopped Corn Silage on Carcass Quality 99 Effect of Fine vs. Medium Chopped Corn Silage Within Harvest Dates on Rate of Gain and Feed Efficiency 100 Effect of Fine vs. Medium Chopped Corn Silage Within Harvest Dates on Carcass Quality 101 Effect of As Ensiled vs. Regrinding of Corn Silage on Rate of Gain anthe d Efficiency 102 Effect of As Ensiled vs. Regrinding of-Corn Silage on Carcass Quality 103 Weather Data During 1967 Harvest 105 Mean Silage Parameters Relative to Stage of Maturity and Fineness of Chop used in Experiment 3 - Feeding Trial 2 106 Mean Silage Parameters Relative to Stage of Maturity and Fineness of Chop used_in EXperiment.3 - Feeding Trial 2 107 Mean Silage Parameters Relating Fresh and Ensiled Material used in Experiment 3*- Feeding Trial 2. 103 Effect of Stage of Maturity on Dry Matter Yield per.Acre. 110 Effect of Stage of Maturity and Fineness of Chop on Silo Storage Requirements 111 Effect of September vs. October Harvested Corn Silage on Rate of Gain and Feed Efficiency 113 vii f Tables (Cont.) Page. Effect of September vs. October Harvested Corn Silage on Carcass Quality 114 Effect of Fine vs. Medium Chopped Corn Silage on Rate of Gain and Feed Efficiency 116 Effect of Fine vs. Medium Chopped Corn Silage on Carcass Quality 117 Means of Rumen pH Values 118 Mean Rumen Volatile Fatty Acid Concentrations and Molar Per Cent 120 Means for Sheep.Parameters 121 Means for.Nitrogen Balance Study- 123 viii. LIST OF FIGURES Experimental Silo Unit Pressure Measuring Cell. The Silo Chambers. Sampling Procedure Sample Removal from the Chamber Schematic Diagram of Laboratory Analysis Conducted on Silage.Samp1es Mean.Dry Matter Content Relative to Stage of Maturity Mean Seepage Volume (ml/100 gm. Fresh Sample) Relative to Stage of Maturity Mean Silage pH Relative to Stage of Maturity Mean Soluble Carbohydrate Levels Relative to Stage of. Maturity Mean Acetic Acid Levels (Per Cent on Dry Matter Basis) Relative to Stage of.Maturity Mean Lactic Acid Levels (Per Cent on Dry Matter Basis) Relative to Stage of Maturity can Total Nitrogen (Per Cent on Dry Matter Basis) Relative to Stage of Maturity ater Soluble Nitrogen (Per Cent on Dry Matter Basis) elative to Stage of Maturity ater Soluble Nitrogen.Expressed as a Per Cent of Total itrogen Relative to Stage of Maturity oluble Nonprotein Nitrogen (Per Cent Dry Matter) elative to Stage of Maturity ix 37 4O 41 42 S4 56 59 62 63 65 67 68 7O 71 f Figures (Cont.) Page Ammonia Nitrogen (Nitrogen per 100 gm Dry Matter) Relative to Stage of Maturity 73 Mean pH and Carbohydrate Fractionization Relative to the Process of Fermentation 75 Mean Nitrogen Fractionization Relative to the Process of Fermentation 76 Effect of Stage of Maturity of Corn Silage on Total Dry Matter Accumulation in the Corn Plant 129 LIST OF APPENDIX TABLES 1x I Experiment 1 - Silage Fermentation Study — Silage Dry Matter Analysis of Variance Lx II Feeding Trial 1; Design of Experiment Feeding Trial 11; Design of Experiment Metabolic Study; Design of Experiment .x III Effect of Stage of Maturity and Fineness of Chop on Beef Cattle Performance (September Harvest) Effect of Stage of Maturity and Fineness of Chop on. Beef Cattle Performance (October Harvest) Effect of Stage of Maturity and Fineness of Chop on Beef Cattle Performance (November Harvest) September vs. October Harvest, Fine vs. Medium Chop, Zero vs. One Per Cent Concentrate Level September vs. October Harvest, Fine vs. Medium Ch0p, Zero vs. One Per Cent Concentrate Level Sheep Parameters Daily Nitrogen Metabolism Data_ Rumen Volatile Fatty Acids (um/ml) at T0 Rumen Volatile Fatty Acids (um/ml) at T2 T Rumen Volatile Fatty Acids (um/m1) at 4 xi Page 140 141 142 143 144 145 146 147 148 149 150 151 152 153 15th APPe @ 5 R1 pendix IV S M of Appendix Tables (Cont.) _ Page Rumen Volatile Fatty Acids (um/ml) at T6 154 ldiX IV Simple Correlation Coefficients - Experiment 4 - ’ 155 Metabolic Study It has oil food prodo' that can be ct oinoroole in Beef cattle r of conversion ooeil as th :oopiete reli finishing cat Segre; in general, portion as i hears that will be aVaj ontinoe to fool these r. Rese Stations ha “Wing and Itsomh h: in the ene; I. INTRODUCTION It has been theorized by most authorities on world population nd production that feeds having a high human caloric value and n be consumed directly by the human population will play a ole in ration formulation for meat animals by the year 2000. ttle rank a poor third to broilers and swine in the efficiency ersion_of concentrate feeds. Thus, competition by these species as the human population will force cattle feeders into nearly e reliance on high yielding roughages and plant residues for ng cattle. Segregation of the total gross energy value of.wor1d food plants ‘ ral, reveals almost as much gross energy in the stalk and leaf as is contained in the grain or tuber portion. Therefore, it that an ample supply of roughages in the form of crop residues available for ruminant feeding in the future and that beef can 3 to be a source of high quality nutrients in the human diet when se.residues. {esearch conducted at Michigan State University and other research ; has conclusively shown that silage is the best method for pre- and storing the nutrients of the growing plant. In Michigan, . has further demonstrated that no other crop will equal corn silage nergy production per acre of crop fed. Mud on silal‘ fineness 0 one. The 02 hring fez establish and prose (2 top on ) hanestee ( levels 0 oils an ( to metal high si] Much research has been reported on the_production and feeding of ilage; however, little is known relative to the effect of maturity, 55 of chop and pressure on silage fermentation and cattle perform- Therefore, the objectives of this study were: (1) To more closely define the changes taking place in the silo fermentation and identify factors controlling these changes, thereby ishing criteria for constructing an efficient container for storing eserving silage. (2) To evaluate the effect of stage of maturity and fineness of n yield per acre and silo storage capacity of corn silage when ted at various dry matter levels and/or stages of maturity. (3) To evaluate the effect of fineness of chop and dry matter on in—silo fermentation, particularly in the production of organic and nitrogen fractionization in the silo. (4) To test the effect of stage of maturity and fineness of chop lbOliC parameters and feedlot performance of beef cattle when fed .lage rations. est ill in th to o l II. LITERATURE REVIEW The utilization of corn.silage as an animal feed has been a well— lished practice for some time.‘ Coppock and Stone (1968) refer to gs dating back to 1852 which report research using corn silage in and England. 1 In 1877, Goffart, of Burtin, France, described in a practical way hportant aspects of silage production. He discovered many important ;ions necessary in preserving the corn plant properly. Specifically, oposed reducing the length of cut from four centimeters to one centi- and the application of a cover weighted with stones or brick to othe exclusion of air. In a speech published in the official report of the State Board of lture of Pennsylvania for the year 1888, John Stewart of Morganya, 1vania said, "The use of ensilage is no longer an experiment; the s of its use in Ohio, New York, New Jersey and in our own state have its value and its practicability.” He went on to say, "The ensilage d to be better for milk than hay, and the cows will milk nearly as s when on grass. It costs us, by actual count, about fifty-nine cents n. When we fed hay, each cow ate two tons in a winter, and with_the feed of meal, did not do nearly as well as with the ensilage, while st of the daily food was eight times as much.”' In conclusion, he “I will state that the man who builds and uses a silo will save eighths of the cost of wintering his stock, and will keep them in (as good order, and get as much milk, beef or butter from them. The as come to stay. Formerly I thought it the rich man's luxury, but see it.is the poor man’s necessity, and if I only had a farm of twenty I would have a silo and keep twenty cows the year through.” F. H. King, in 1900, wrote ”Corn for silage--there is no crOp now lly grown which is so well suited to the production of silage as corn, wherever it will grow well to maturity. The unavoidable with it are very small; heavy yield per acre may be secured with. bertainty at moderate cost; and the silage made from it has less nonable features than that of most other crops." Silage production and preservation has progressed from this point any changes but with many questions left unanswered.. It is the pur- E this review to examine the state of knowledge of silage fermentation :hods of producing a quality product that will ultimately maximize performance. Other reviews covering these as well as other subjects concerning are to be found in Watson and Nash (1960), Barnett (1954), Coppock ene (1968) and Owens (1968). Fermentation Watson and Nash (1960) define silage as ”a succulent material pro— y a process of controlled vital changes from a green crop or other 1 of high moisture content. These changes which take place are mplex and depend on many factors.” Thus, to identify these changes erstand how they come about, one must first describe- the character corn plant and the resulting silage. ln 1 to com P isle 1. non tool fee extiz incoea: o olnost his fern phents ( T1 inett I___________________________________________ S In some very early work, Annett and Russel (1907) characterized 'n plant (green maize) and the resulting corn silage as shown in From this analysis, they concluded that the major changes took place during fermentation were a great reduction in nitrogen— tract (later found to be a breakdown of the soluble carbohydrate), ease in nonprotein nitrogen (due to the breakdown of protein) and st complete disappearance of the sugars. They also concluded that. rmentation does not affect the fiber content of the silage. Bennevet_al. (1964) presents a very complete breakdown of the com- iof the corn.plant and the content of each part including minerals. The actual pattern of fermentation has best been described by (1954) as a four—phase process. Phase 1. A relatively short phase during which the plant cells 11 respiring. This results in the production of carbon dioxide, lization of simple carbohydrates and a flow of water from the mass these biological happenings and the mechanical compression of the These events are accompanied by the evolution of heat. 3hase 2. A short time period in which small amounts of acetic 2 produced by coliform bacteria. e ’hase 3. The point of initiation of the lactic acid fermentation ; dependent upon the activity of lactic acid producing organisms, :illi and streptococci supported by adequate amounts of carbo- 'hase 4. The stage of quiescence in the mass during which the .cid production reaches its peak and remains at a high level. At nt the pH should be less than 4.2. Condon §£_al. (1969) have at this fermentation is complete at the end of eight days. in latte Sher Ext TABLE 1 Annett and Russel - Silage Characterization atter Extract .25 gen-free_Extract N otein N N (by difference) ent of total N present as NPN (as H2804) tatile Acids Green Maize (a) 16.81 0.48 1.78 9.33 4.21 1.00 0.285 0.214, 0.071 25.0 Maize Silage (‘8) 12. O 99 .39 .45 .38 .82 .98 .234 .137 .103 . .72 . .007 .006 TABLE 1 Annett and Russel - Silage Characterization Matter er Extract 6.25 rogen-free Extract er al N >rotein N JPN (by difference) cent of total N present as NPN s (as H2804) olatile Acids Green Maize (9o) 16. O. 25. 81 48 .78 .33 .21 .00 .285 .214 .071 0 Maize Silage (1») 12.99 0.39 1.45 0.234 0.137 0.103 43.72, 0.007 0.006 sported onion (1 obtained of the 5 ll silage-o plant on ntlsept lot the later We life 115 Peterson, Hastings and_Fred (1925), while working with corn silage orted that the oxygen in the mass that is used in the production of the bon dioxide during Phase 1 had disappeared almost entirely within five rs. They showed that maximum concentration of carbon dioxide was ained at 46 hours. At this_point, carbon dioxide comprises 60% to 70% the silo gases. Russel (1907) held that there were three agents involved in the age-making process during Phase 1: (l) The living maize cell; (2) the nt enzymes; and (3) the microorganisms. He showed by the addition of iseptics to the mass that the first two were "primary and essential,” the latter, the microorganisms, were only ”secondary and nonessential." er works (Peterson, Hastings and Fred, 1925) showed that the bacteria e useful, if not necessary in the production of acids which drop the This work was done by sterilizing the mass to stop respiration, then oculating it with microorganisms. i The microorganisms in the silage are the chief agents in the pro— ‘ion of the alcohol and organic acids according to Peterson, Hastings [Fred (1925). Kempton (1958) found that the initial number of bacteria he fresh crop bore no relationship to the final quality of the silage. @150 found that less than 0.1% of the bacteria on the crop at the time ensiling were capable of growing on lactobacillus selection medium. Gibson, Sterling, Keddie and Rosenberger (1958) showed that the nant bacteria of fresh herbage disappeared rapidly. All typical ge bacteria proceeded to multiply immediately if they were represented he herbage and if the temperature was appropriate for the organism. A complete review of the microbiology of silage is presented by .ton (1958). Salsbury, Maths heculture did not fo Ion closely followed The temperature silage quality has bee {1910) concluded that sever exceeded 24° to tively high tenperatu: ht beneficial (Coppo slotted that the great as the amount of air eluded that good sil to 38° [3. According silage fermentation llich controls the co state that lactic act on“ o) and only kil cluded that a temper tion. The work of S perature. In this n reached 80° F. At 1 again indicating an McCullough ( illicit silage was ha concluded that ambi on the final silage Salsbury, Mather and Bender (1949) concluded that viability of aculture did not follow pH of the silage during fermentation but much 3 closely followed a simple linear decline with length of fermentation. The temperature of the fermentation mass and its influence on Lage quality has been a question of great concern. Babcock and Russel ! 100) concluded that good silage could be made when the temperature Ier exceeded 240 to 260 C; this was in contrast to the concept that rela- Iely high temperature (550 C) for silage-making was not only inevitable :beneficial (Coppock and Stone, 1968). Eckles (1916), among others, rwed that the greatest factor causing variation in silage temperature zthe amount of air incorporated in the mass. Furthermore, he con- wed that good silage could be made at temperatures ranging from 100 380 C. According to Watson and Nash (1960), the temperature of the age fermentation is wrongly thought by some to be the sole factor ,ch controls the course of bacterial action. These authors also he that lactic acid producing organisms are most vigorous at 1220 F f C) and only killed at 1670 F (750 C). Benne and Wacasey (1961) con— Hed that a temperature of 800 to 1000 F is optimum for silage fermenta- h. The work of Shaw 25.21: (1951) related silage pH to silage tem— ture. In this work, silage pH did not go below 5.0 until the mass hed 800 F. At temperatures above 1000 F, the pH started to go up 'n indicating an alteration in the fermentation. McCullough (1969) reviewed work done at the Georgia station in» h silage was harvested on days of various ambient temperatures. He Eluded that ambient temperature during harvest had no apparent effect. he final silage temperature. Forage ensiled on a hot day may have atonerature greater in to a temperature The effect of h lyllechtel, Atkinson a they referred to as b peratures. Metabolic digestibility was red lnered from 5595 to 4 front 71% to 64% as a value of this silage llese changes result [1967) reported simi heated. Heating the effect. McCullough (l fermentation by char (1) Seriously color, has a strong sith a pH of 5 or al (2) Properly color, has a please acid taste indicati (3) Overheal and exhibit an odo: As stated p related to the oxy solely a function temperature greater than desired, but according to McCullough, it cools .to a temperature determined largely by cell respiration. The effect of high temperature (in excess of 60.50 C) was reported Bechtel, Atkinson and Hughes (1943). In this work_they described what y referred to as browning or darkening of the silage due to high tem— atures. Metabolic work with_this silage showed that dry matter estibility was reduced from 64% to 50%, protein digestibility was ered from 55% to 4% and nitrogen-free extract digestibility declined m 71% to 64% as a result of the high temperature. Also, the carotene ue of this silage was markedly decreased and the ash content increased. se changes resulted in a decrease of 50% in consumption. Gordon 67) reported similar results when experimental silos were intentionally ted. Heating these silages after fermentation had little, if any, Sect. McCullough (1969) summarized the effect of temperature on silage mentation by characterizing three distinct types of silage. (l) Seriously under-heated: This silage is usually a drab green 3r, has a strong odor, the tissues are slimy and have an insipid taste 1 a pH of 5 or above. (2) Properly heated: This silage is light green to yellow in r, has-a pleasant vinegary odor, tissues are firm and it has a sharp taste indicative of a pH below 4.5. (3) Overheated: These silages are from brown to black in color xhibit an odor from slightly burnt sugar to charred hay. As stated previously, the temperature of the mass is closely d to the oxygen.trapped in the mass. This, in turn, is almost a function of density of the silage mass, whether resulting from Insure externally ap 11 the silage. Kearne deluded immediately. mltherewas a greate hturu, butyric acid Kempton (1958) pillarily by density overheated and underw preserved silages wer roles of lactic acid silage was packed too lactate after two to disappeared, to be re The density 0 the silo, depend on n fineness of chop and of the silage mass, 1 Dexter, Huffm largely vertical in iovhich pressure at authors also report the pound per square thesillage. In a 6| In contrast, actually measured i utely 5 psi. Thes lleatly influenced 10 are externally applied or static pressure exerted from the weight e silage. Kearney and Kennedy (1962) concluded that air must be ded immediately. If not, a longer aerobic fermentation resulted, here was a greater loss in soluble carbohydrate and lactic acid. rn, butyric acid production was increased. ‘ Kempton (1958) also concluded that silage quality was determined ily by density or packing of the silage mass. Loosely packed silages ated and underwent primarily an acetic acid fermentation. Well- ved silages were firmly packed and contained as much as 150 micro— ,of lactic acid per gram of fresh weight. On the other hand, when was packed too tightly, it contained only about 100 micromoles of e after two to three days in the silo, all of which subsequently eared, to be replaced by butyric acid. The density of the silage mass and, in turn, pressures generated in lo, depend on weight of the silage mass which in turn is determined by 55 of chop and dry matter content of the ensiled material, and height silage mass, all of which will be discussed later in this review. Dexter, Huffman and Benne (1959) reported that pressures are 'vertical in a silo. (This is nacomparison with water or any fluid h pressure at any one point is equal in all directions.) These also report that the pressure in a silo is equal to approximately ad per square inch, per three foot depth, due to the weight of ge. In a 60-foot silo, this would equal 20 psi. 1 contrast, Yu, Boyd and Menear (1965) reported maximum pressure measured in a 30' x 60' upright silo was 700 psf, or approxi- psi. These authors also report that the pressure in a silo is afluenced by filling procedure. Boyd and Aldrich husing pure cellulos vlaeellulose exerted 51.9 lb. per cubic foo heu259a dry matter, 7 The losses enco density and pressure. seepage [Watson and N lydevvsity, in that t hereas, seepage loss of the ensiled materi laymatter for variou Murdock (1954) varied from 40‘ gallon loss when the silage Miller and Cli seven different fora: from12.296 to 38.6% v that moisture conten regulating seepage 1 equation: In this equa‘ Iaterial. This mod loss. ll Boyd and Aldrich (1959) showed the effect of dry matter on pressure ing pure cellulose at various dry matter levels. When 100% dry matter, ellulose exerted 93.5 lb. pressure per cubic foot, when 95% dry matter, lb. per cubic foot, when 50% dry matter, 78 lb. per cubic foot and 25% dry matter, 70 lb. per cubic foot. The losses encountered in silage production are also functions of ty and pressure. Losses are of two general types, fermentation and ge (Watson and Nash, 1968). The fermentation losses are influenced nsity, in that_the greater exclusion of air, the less will be the loss; as, seepage losses are a function of pressure and dry matter content e ensiled material. King (1900) reported losses as a per cent of atter for various layers in the silo (Table 2). Murdock (1954) reported seepage losses from concrete stave silos i from 40 gallons per ton of silage at 18% dry matter to no seepage when the silage reached 39% dry matter. Miller and Clifton (1954) examined 24 tower silos filled with different forages involved in six different experiments ranging l2.2% to 38.6% dry matter. They concluded, as have all other workers, voisture content of the ensiled material was the primary factor vting seepage loss° These authors proposed the following prediction .on: 2 % DM lost = 26.96 - 1.576 x +0.0230X In this equation, X is the per cent of dry matter of the ensiled al. This model accounted for 84% of the variation in dry matter 2nd lotton TABLE 2 Unavoidable Losses in the Silo (King, 1900) Pounds of Silage % of Dry yer in Layer- Matter Lost face 8,934 32.53 th, 8,722 23.38 th 14,661 10.25 th 48,801 2.10 th 13,347 7.01_ rd_ 7,723 2.75 nd 12,689 3.53 tom 12,619 9.47 McCullough (1969 silage dry matter to si ranging from 25% to 35% lives hypothesized the tieplant was below 255 vas above.359a dry matte condition in the silo. Huber, Thomas a losses with 44% dry ma silage (6.49s loss) and shady, Goodrich et al temptation loss in v Nicholson and iortheexpressed pur. to drying the materie acid production . McCullough (1‘ probably exerts no d its importance stems items as stage of me of packing the sil‘a; Carbohydrate Fermen The early wc striking feature 0: 311d an increase in vas due to bacteri 13 McCullough (1969) reported some work conducted by Axellson relating age_dry matter to silage pH, in which it was shown that dry matters ging from 25% to 35% produced the most suitable pH value (below 4.5). was hypothesized that pH was not reduced to the desired low level when splant was below 25% dry matter. The increase in pH when the silage above 35% dry matter was due to the inability to provide an anaerobic, dition in the silo. Huber, Thomas and Emery (1968) reported greater silo dry matter ses with 44% dry matter silage (15.1% loss) compared_to 36% dry matter age (6.4% loss) and 30% dry matter silage (7.0% loss). In another dy, Goodrich 32 al. (1967) reported the opposite result, with greater mentation loss in a 32% dry matter silage than at 45% dry matter silage.. Nicholson and Cunningham (1964) added shredded newspaper to silage the expressed purpose of increasing the dry matter and compared this drying the material. They found that drying had less effect on organic d production. McCullough (1969) states that ”The dry matter content of a crop bably exerts no direct force on the events transpiring in the silo., importance stems from its usefulness as a relative measure of such ms as stage of maturity, protein content, and the relative difficulty packing the silage.” aohydrate Fermentation The early work of Annett and Russel (1908) showed that the most .king feature of silage fermentation was the breakdown of carbohydrates an increase in organic acids. Hunter (1921) proved that this breakdown due to bacterial action. Johnson gt al. ( pirate levels in the c hispaper reported, as soluble carbohydrate d: conclusively showed ch. fresh, not oven-dried, carbohydrate was destr When studying t degradation of carboh) them into volatile am volatile acid in sila, acid found in some si ajor nonvolatile aci present in larger am: The work of Co states that 80% of ti lrasemann (1928) als ration of these suga results in a loss oi represents a loss 0: The breakdow considerable amount and Fred, 1925), wt (Dexter, Huffman as Woodman and goes breakdown as lion ofvnitriogen-r 14 Johnson e£_al. (1966a) reported an extensive study of soluble carbo— nate levels in the corn plant and its breakdown during fermentation. s paper reported, as has earlier work, that a major portion of the .uble carbohydrate disappeared during ensiling. These workers also Elusively showed that carbohydrate determinations should be run on 5h, not oven—dried, samples. In their work, up to 60% of the soluble bohydrate was destroyed during drying. When studying the end products of silage fermentation (primarily pradation of carbohydrates to organic acids), it is necessary to divide am into volatile and nonvolatile acids. Acetic acid is the primary .atile acid in silage fermentation with traces of propionic and butyric .d found in some silages (Barnett, 1954). However, lactic acid (the or nonvolatile acid) is more important in silage fermentation and is sent in larger amounts (Watson and Nash, 1960). The work of Crasemann (1925), as reviewed by Watson and Nash (1960), tes that 80% of the losses in silage are due to carbohydrate degradation. semann (1928) also calculated the losses of energy due to the fermen— ion of these sugars. He found that the conversion of sugar to alcohol alts in a loss of 3.2%, whereas the conversion of sugar to acetic acid resents a loss of 39.3% of the potential energy. The breakdown of starch is not clear. Some work has shown that a riderable amount is broken down (Dox and Yoder, 1920; Peterson, Hastings Fred, 1925), while others show very little, if any, degradation ter, Huffman and Benne, 1959). Woodman and Amos (1924) reported that a portion of the fiber under- breakdown as a result of bacterial activity with the probable forma- of nitrogen—free extracts and organic acids. They also report that heresidual fiber has . ieoriginal fiber of 1 Many other compo hefermentation. Morp thrilling a silage sen fl-methyldentyric aci propanal, acetone, but ilso many esters were The effect of 1 mini performance has sylposium on voluntar in silage reduces fee acid to both green ch latter basis. These acetate per 100 kg. o reduce dry matter in The authors conclude energy from acetate In contrast 1 sorghum silage, cons from feeding silage contribution of sil dry matter intake v of the ration dry 1 The lactic ertensively. Klos 15 esidual fiber has been shown to possess greater digestibility than riginal fiber of the green crop. Many other compounds have been identified in silage as a result of ermentation.. Morgan and Pereira identified many compounds by steam lling a silage sample. Among them were C2 - C6’ isobutyric,<£ and_ Fethyldentyric acids, C2 - C6 aldehydes, 2 and 3 methybutanal, 2 methyl- inal, acetone, butanone, benyaldehyde, phenylacetaldehyde and furfural. imany esters were found, but were not separated by the distillation.. 1 The effect of the volatile fatty acids present in the silage on. 1 performance has been shown by many workers. Conrad (1966), in a sium on voluntary feed intake, concluded that the acetic acid level lage reduces feed intake. Dinius, Hill and Noller (1968) added acetic to both green chop and corn silage at levels from 0% to 6% on a dry I basis. These animals voluntarily consumed up to 112.8 grams of te per 100 kg. of body weight. Thus, acetic acid additions did e dry matter intake, but had no significant effect on caloric intake. uthors concluded that "The animals in these trials were substituting y from acetate for energy from forage.” In contrast to these reports, Senel and Owen (1966), working with am silage, concluded that any depression in dry matter consumption feeding silages is due to something other than the acetate and lactate ibution of silages. In fact, a significant (P‘< .01) increase in atter intake was found when acetate was added at levels up to 2.8% a ration dry matter. The lactic acid content of the silage has also been investigated sively. Klosterman g£_al.-(1960) found that one pound of lactic all in silage replace: mcluded that methods lactic acid content we Kempton [1958) level approaching 150 there was a relatively liter three weeks in t Schaadt and Jol heated corn silage. plate by the end of e: abundant than the D(- mt change with ferme Lactate acts a investigated by Hersh fermentation using or pyruvate, preferenti: propionic acids, wit? acid. In .the same 5 digestion. Prior to 194( of silage quality. that many other mea pH. Watson and h llolatile, nonvola1 the silage, an impl 16 in silage replaced 2.8 pounds of ration. From this, these workers. luded that methods of ensiling or treatment of silage to increase Lc acid content would be advantageous. Kempton (1958) reports that lactic acid increased rapidly to a . approaching 150 micromoles per gram of fresh material. Thereafter, ewas a relatively rapid decrease to an average of abouthO micromoles 'three weeks in the silo. Schaadt and Johnson (1968) studied lactic acid in some detail in ed corn silage. They found that the production of lactate was come by the end of eight days. The L(+) form of lactate was less ant than the D(-) isomer. This distribution of D and L forms did hange with fermentation time. Lactate acts as a metabolic intermediate in the rumen. This was tigated by Hershberger gt_al. (1956). They found, in an in_vi£rg ntation using ovine rumen microorganisms, that lactate, along with, ate, preferentially increased the rate of formation of acetic and nic acids, with lactate having the greatest effect on propionic In the same study, lactate decreased the rate of cellulose ion. Prior to 1940, pH was considered to be the most accurate indicator vage quality. McLean (1941) analyzed many silages and concluded any other measurements including dry matter were more accurate than Watson and Nash (1960) state that ”all the acids in silage ile, nonvolatile and amino) combine to give the total acidity of lage, an important value.” They go on to report that it is not wessary to determin plis a satisfactory ' Protein Breakdown The breakdown any authors. Watson this breakdown is pri ilso showed that pro silage. Russel (190 that the protein bre iatson and Nash. (196 red clover silage in prior to ensiling. protein breakdown di The breakdown pounds to amino acid more complex volatii mdBenne (1959) al: aresult of this do espentamethylene d The extent 0 reported that in th '35 degraded to nor protein hydrolysis the ensiled materi 17 ssary to determine all the different acids since, in their estimation, 5 a satisfactory index of the course of the fermentation process. :in Breakdown. The breakdown of proteins (proteolysis) is well established by authors. Watson anleash (1960) state that it is now conclusive that breakdown is primarily an action of plant enzymes. Hunter (1921) showed that proteolysis is a result of plant enzymes in normal e. Russel (1908) found, as a result of an experiment with maize, the protein breakdown was due to the tryptic enzymes of the cell. n and Nash (1960) reported the work of Kirsch (1930), who treated lover silage in many ways including autoclaving a fresh sample to ensiling. This treatment destroyed the enzymes and, as a result, in breakdown did not occur. The breakdown normally proceeds by way of relatively complex com- 5 to amino acids which can then be deaminated to form ammonia and ‘omplex volatile bases (Watson and Nash, 1960). Dexter, Huffman nne (1959) also report the formation of some longer chain acids as lt of this deamination. Russel (1907) reported many amines such tamethylene diamine, betaine, adenine, and others. The extent of proteolysis has varied greatly. Brody (1960) ed that in the silages he analyzed, up to 25% of the total nitrogen graded to nonprotein nitrogen. Later work by Brody (1965) showed n hydrolysis ranges from 18% to 29% depending on the dry matter of siled material. [on Silage Maturity The research in confined to work done pass or alfalfa sila utter was altered ra Wren reviewing lid of confusion in hogan (1954), workin lli dry matter, clas Ears b. Kernel Early Late m Early Wei l—o Kerne‘ leave; Kerrie leave The Ohio workers (J from 2096 to 71% dry Silage Maturity The research included in this review relative to maturity is being ned to work done with corn.and, in a few cases, sorghum crops. No or alfalfa silage research is included because percentage of dry r was altered rather than maturity. When reviewing the area of corn silage maturity, there is a great of confusion in terminology. Nevens, Harshbarger, Touchberry and a (1954), working with immature corn silages ranging from 15% to ry matter, classified the stage of maturity as follows: % Dry Matter Ears beginning to form 15 Kernels forming 17 Early milk 20 Late milk 23 Early Dent 25 Well-dented 28 Kernels hardening, most leaves green 30 Kernels hardening, fewer leaves green 32 i0 workers (Johnson, 1967 and 1968) worked with silages ranging 0% to 71% dry matter and used the following terminology: % Dry Matter Blister 20.2 to 21.3 Early milk 19.9 to 23.7 Milkiearly dough 21.9 to 27.0 Dough-dent 27.2 to 28.6 Glaze Flint Post-fro Mature Because of the loosed in the review tens. It has been sh torn plant has a defi oithe resulting sila of the plant, relati Stage of Maturity i Tassel Milk . Dough Glaze Ripe Other worker and Stone, 1968; Jo reported a similar However, the work Ind work reported lith the more matu % Dry Matter Glaze 33.5 to 34.6 Flint 38.4 to 47.2 Post—frost 46.7 to 50.8 Mature 71.0 to 71.7 Because of the variation in terminology, dry matter values will ased in the review and research presentation rather than maturity ns. It has been shown that the physical stage of development of the 1 plant has a definite effect on the chemical and nutrient composition the resulting silage. Hopper (1925) reported the proximate analysis the plant, relative to maturity, as follows: Stage of Maturity % DM NFE CP CF Ash E. Extract Tassel 13.5 50.5 11.6 27.7 8.5 1.7 Milk, 18.5 56.9 8.9 26.1 6.5 1.6 Dough 25.0 61.6 8.2 22.5 5.5 2.2 Glaze 32.7 62.2 8.3 21.4 5.4 2.6 Ripe 43.0 63.6 8.2 20.3 5.0 3.0 Other workers (Byers and Ormiston, 1966; Buck, Merrill, Coppock Stone, 1968; Johnson §t_al:, 1966; and Sprague and Leparulo, 1965) rted a similar analysis, especially the reduction in crude protein. 'er, the work of Owen and Webster (1958), working with sorghum silage, ork reported by Colorado State University in 1959 show the reverse, the more mature corn silage having a higher crude protein. Johnson at a_1_. (l elative to dry matter < iirdings : Per Cent Dry Matte Whole P1; 13.8 16.2 20.8 28,4 37.6 40.4 The reports re onerous. in generai hole the stage of p? mo; Nevens 3 fl., ilratzler, 1969; Bye id, 1966 and 19 The most com conducted by the Oh Thomas 311d Emery (] naxiom yield of a: the dough dent (28‘ WET. no actual da work by Johns0n et matter content of “ll 50°. J ’ respects. 20 Johnson et_al. (1963) analyzed the components of the corn plant .ve to dry matter content as the plant matured and reported these lgS: >er Cent Dr Matter of the Corn Plant and Com onents and % Ear Y . P Whole Plant Leaves Stalks Ears % Ears 13.8 20.5 13.4 ---- 0 16.2 17.4 17.1 12.2 15 20.8 20.8 17.4 23.7 41 28.4 24.5 21.7 38.1 52 37.6 27.8 18.9 53.1 66 40.4 37.7 24.3 62.4 66 The reports relating corn silage maturity to dry matter yield are >us. In general, they all conclude that dry matter yield is increased the_stage of physiological maturity of the plant (Bryant §£_al., Nevens e£_al., 1954; and Owen, 1958 and 1962) and then decreases .ler, 1969; Byers and Ormiston, 1964; Fowler g£_al., 1968; Gordon , 1966 and 1968; and Thomson and Rogers, 1968). The most complete studies relating maturity and yield have been ted by the Ohio workers (Johnson 33 31., 1966 and 1968) and-Huber, and Emery (1968). Johnson and McClure (1968) state that "the m yield of digestible energy per hectare would be achieved between Jgh dent (28% DM) and the glaze 64% DM) stage of maturity.”- How— 10 actual data on yieldwere presented in this paper. In earlier I Johnson 93 El” (1966), yield datawere maximized when the dry content of the stalks, leaves and ears were.approximate1y 20, 28 5 respectively. In the Michig: Has harvested at 30% levels, the dry matt hectare, respectivel and 1966) in that th about 35% dry matter In other stu< increased as much a: with dry matter lev of 27% in some stud The effect c mentioned by many a the later harvester had a significantl; MW matter. J. [71% dry matter) h “11 little if any This fluetu acid production wh lenblshire and Var in the earlier ha: lropionic’ butyri. respectiVely, Whe: a / lld 3.700, respect the plant mature d 21 In the Michigan State work of Huber, Thomas and Emery, silage harvested at 30%, 36% and 44% dry matter. At these dry matter els, the dry matter yields were 10.4, 12.2 and 10.2 metric tons per tare, respectively. This study supports the Iowa work (Hanway, 1963 1966) in that_the plant actually stops accumulating dry matter at it 35% dry matter, which is referred to as physiological maturity. In other studies, when dry matter levels were below 35%, the yield teased as much as 25% as it neared the 35% level. In studies working 1_dry matter levels above 35%, losses were great. These reached levels 27% in some studies. The effect of plant maturity on the fermentation in the silo is :ioned by many authors. Huber, Graf and Engel (1965) reported that later harvested silage (51% dry matter, hard dough stage of maturity) a significantly higher pH than did silages harvested at 34% and at dry matter. Johnson and McClure (1968) found that very mature silage dry matter) had a pH of 7.7 after fermentation, indicating that little if any fermentation had occurred. This fluctuation in pH is probably a reflection of the reduced production which occurs with dryer or more mature silages. Gordon, yshire and VanSoest (1968) reported higher amounts of organic acids he-earlier harvested_silages. They reported the levels of acetic, ionic, butyric and lactic acid to be 2.8%, 0.2%, 0.2% and 7.8%, ectively, when the silage was 32.4% dry matter, and 1.4%, 0.1%, 0.1% 3.7%, respectively, when the silage was 60.0% dry matter.. Johnson L. (1965) and Klosterman (1963) reported about the same changes, as )lant matured. In the paper by Klosterman (1963), lactic acid levels dropped from 11.33% 0 nsnse dry matter it One factor the duction in silage is ensiled; this being 1 lohnsonetal. (1966. they reached a peak and decreased linear soluble carbohydrate acid production in 1 Johnson fl 1: distribution in sil: reported tungstic a. tungstic acid solub reasured ammonia, “Wen produced d 44 increased in d1 mattrial does not 1 hoist silage, Silo losses and fermentation 1 1°55“ approached matter, In a Stud made at 24.306 and letter silage I‘esn later harvested si 22 upped from 11.33% of dry matter when the silage wast28.7% dry matter 13.75% dry matter when the silage was 66.6% dry matter. One factor thought to be directly related to the level of acid pro- ction in silage is the level of soluble carbohydrate in the plant when siled; this being the primary substrate for acid producing bacteria. hnson at al. (1966a) measured levels of carbohydrates and found that ey reached-a peak when the plant was approximately 25% dry matter, 1 decreased linearly until the plant was mature. This reduction in Luble carbohydrates followed virtually the identical pattern of organic .d production in the resulting silage. Johnson g£_al. (1967) reported an extensive study of the nitrogen :tribution in silage and how this relates to maturity. These authors lorted tungstic acid precipitable nitrogen as true protein and the gstic acid soluble nitrogen portion as nonprotein nitrogen. They also sured ammonia. They found the levels of ammonia and_nonprotein rogen produced during fermentation to be lower as the plant matured increased in dry matter. This is another indication that the dry erial does not undergo as extensive a fermentation as does the more t silage. Silo losses related to maturity are of two forms; seepage losses fermentation losses. Miller and Clifton (1965) reported that seepage es approached zero as the ensiled material approached 30% to 32% dry er. In a study reported by Sprague and Leparulo (1965), silages were at 24.3% and 32.5% dry matter. In this work, the early dent or er silage resulted in a 5.6% dry matter loss during storage. The r harvested silage, 32.5% dry matter, resulted in 3.8% dry matter loss. .silages above the 30% or 35% dry matter level, dry matter losses appear to be highest 1 nousiy, Huber, Thoma 11,396 and 44% dry levels were 7%, 6.4% greater dry matter lc oxidation due to a dc Coppock and 8‘ iron bacterial ferme runinants can use th acetic and lactic ac From this wor should be harvested utter losses durin ; I The effect 0 for corn silage hav Duncan (1960) repor period, and conclur corn silage was 67 52.8%. Probably t dry matter digesti (lets and 1968). Ia“ling from 20% < d'Igestibilities Wt latter digestibil the dough dent st thill leveled off, has a slight deer __ J r to be.highest for the higher dry matter-silage. As mentioned pre— 1y, Huber, Thomas and Emery (1968) reported three dry matter levels; 36% and 44% dry matter. Silo losses realized at these dry matter were 7%, 6.4% and 15.1%, reSpectively. One could assume that the r dry matter loss at the higher dry matter levels is due to increased ion due to a decrease in compaction. Coppock and Stone (1968) conclude that "dry matter losses resulting acterial fermentation may not reflect a net energetic loss because nts can use the primary end products of the bacterial fermentation; p and lactic acids.” From this work, it has been concluded by Hoglund (1964) that silages i be harvested between 32% and 35% dry matter to reduce total dry 3 losses during ensiling. The effect of stage of maturity on the various digestibility values trn silage have been investigated by many authors. Huffman and L (1960) reported the results of corn silage analysis over a 16-year 1, and concluded that the average per cent of digestible dry matter of ilage was 67.8%. In this same study, protein digestibility averaged Probably the most extensive work relating corn silage maturity and tter digestibility has been done by the Ohio group, Johnson e£_al. and 1968). In this work, eight different silages were harvested g from 20% dry matter to approximately 72% dry matter. Dry matter ibilities were established using lambs. It was concluded that dry digestibility increased from 66% to 72% as maturity increased up to gh dent stage of maturity, or approximately 28% dry matter. It veled off, with no significant difference from this point on. There light decrease after the 28% dry matter level. In other studies, asignificant increa drynatter levels ab aregression equatio percent of dry matt Per Cent In this equation, X estimate dry matter nlth it being the nu This would i that dry matter dig 19m t° a Point. naturity 0n dry mat 1969; Byers and Orr 1963, Kuhlman and ( Hill, 1963; and Po: many of these Stu d. lbilit)’ of the Sta and that this deer “grain in the to of the Plant is ap dry matter! 24 gnificant increase in dry matter digestibility has been reported at matter levels above 28%. In the work of Thomson and Rogers (1968), gression equation is reported, relating dry matter digestibility and tent of dry matter of the corn crop. This regression equation is Per Cent dry matter digestibility = 71.21 — 0.14X l l ris equation, X = dry matter of the crop being ensiled. Noller, 3r, Rumsey and Hill (1963) also report a regression equation to rate dry matter digestibility. Their equation is Y = 70.88 + 0.06X X being the number of days after the blister stage of maturity. This would indicate that these workers agree with the_0hio work; dry matter digestibility does increase as maturity progresses, at to a point. Other workers have also investigated the effect of ity on dry matter digestibility (Bratzler, 1969; Buck e£_al., ‘Byers and Ormiston, 1964; Caldwell and Perry, 1967; Hill and Noller, iKuhlman and Owen, 1962; Nevens, 1933; Noller, Warner, Rumsey and 1963; and Perry EE.E£°’ 1968), and report varying results. From >f these studies, Coppock and Stone (1968) concluded that the digest— .y of the stalk and leaves of the corn plant decreases with maturity, at this decrease is compensated for by an increase in the proportion in in the total plant, so that the total dry matter digestibility plant is approximately constant throughout a range of 20% to 50% tter. The effect of 1 reported as varying 1 reported no differenc they were working wit $211969) both re lower with higher dry to this problem of p and French (1956), i protein was directly iced. It was earlie silages decreases wi it may explain the c In the report of (31. in the feed increas to 9% crude protein as crude protein 1e Per cent of crude p nines the digestib; The work of also concluded tha higher dry matter digestible energy The effect nany authors. Cor rations ranging f: m“ affecting "is fed —— J 25 The effect of maturity on crude protein digestibility has been ted as varying in results. Gordon, Derbyshire and Humphrey (1966) ted no difference in crude protein digestibility in the silages were working with. However, Hunt and VanderNoot.(1961) and Goering . (1969) both reported that protein digestibilities were consistently with higher dry matter silage, or later harvested silages. Related is problem of protein digestibility is the report of Glover, Duthie rench (1956), in which they concluded that the digestibility of crude in was directly related-to the per cent of crude protein of the It was earlier shown in this review that crude protein in corn as decreases with maturity and therefore, if this report is correct, I explain the decrease in protein digestibility previously reported. :,report of Glover e£_al. (1956), the digestibility of crude protein a feed increased very rapidly at low protein levels from about 2% crude protein. Thereafter, digestibility increases more slowly tde protein levels increase. This report concluded that the total nt of crude protein in the feed, irre5pective of its nature, deter- the digestibility of the protein. The work of Hunt and VanderNoot (1961), previously referred to, oncluded that digestible energy increased with later harvested, or dry matter, silages. This was one of the few papers reporting ible energy. The effect of maturity on dry matter intake has been reported by ithors. Conrad, Pratt and Hibbs (1964) analyzed many different 5 ranging from 52% to 80% digestibility. They concluded that the 5 affecting intake, when a ration low in digestibility (52% to 66%) 3d, were such things as body weight, reflecting roughage capacity, my undigested residi ofpassage. When ra‘ to 80%), the factors aetabolic body size, ration. Corn silage lherefore, we must 1 to the animal's cap: Most authors likewise increases. correlation coeffic hatter intake= Ag; [1963) and Klosterr Tl matter intake ' ltport 0f Klosterm is actually poorer 0963) reported vc more mature silagt Silages fed to la tion with the nor In attempt to maturity, Thou silages and Cone: to dry matter cor tionship’ Since ftEdlng time did eluded that the . rod . l p usts in the J 26 undigested residue per unit of body weight per day, reflecting rate assage. When rations were fed that were high in digestibility (67% 9%), the factors affecting intake were other parameters, such as )olic body size, production of the animal and digestibility of the >n. Corn silages would fall in the high digestibility type of ration. efore, we must look for factors affecting intake which are not related to animal's capacity to consume more food. 1 Most authors have found that as dry matter increases, silage intake dse increases. Johnston and Cook (1970) reported a significant *lation coefficient of 0.65 between dry matter of the silage and dry r intake. ‘Again referring to the Ohio work of Johnson and McClure ) and Klosterman e£_al. (1963), both reports indicate an increase in atter intake with increasing dry matter of the silage. However, the t of Klosterman 33 al. (1963) also states that the more mature silage tually poorer in feed efficiency. Noller, Warner, Rumsey and Hill ) reported voluntary intake by heifers was 20% to 30% higher for the mature silages. Owen (1962) reporting on the work with sorghum es fed to lactating dairy cows, also reported an increase in consump- with the more mature sorghum silage. In attempting to explain the increase in dry matter intake relative turity, Thomas, Moore, Okamoto and Sykes (1961) worked with alfalfa as and concluded that consumption was linearly and positively related v matter content of the silage. However, this is a secondary rela— rip, since changing the dry matter content of the silage or hay at fig time did not alter consumption. Therefore, these authors con— 1 that the variation in dry matter intake was due to fermentation :ts in the silage. There have been many other reports of dry matter intake increaSing re Huber, and 31356“ loss and Meyers, 19 by Goering g’g EL ( matter intake- In were used. There were I rumen parameters. alfalfa and sudex : percent of rumen rate were higher w The majorit performance has be lyers and Ormistor Derbyshire and Var Johnson (1929) all maturity of the c Significant incre cluded that this altually there we Tuber, Thomas and increasing maturi Reports 01 as the work with allthors have rep. "1°51 Of these re the I‘ation than ke increasing relative to maturity (Bryant_e£_al., 1966; Bryant, r, and Blaser, 1965; Huber, Graff and Engel, 1965: Marshall, Nordon, and Meyers, 1966; and Owen egual., 1967). In contrast, a report, oering §£_al. (1969) concludes that maturity had no effect on dry er intake. In this study, silages ranging from 23% to 47.8% dry matter used. There were no reports of the affect of corn silage maturity on n parameters. However, Mahopatro and Leffel (1964), working with lfa and sudex silages at various dry matter levels, reported that .ent of rumen acetate was lower and per cent of propionate and buty- ‘were higher when either hay or dry silages were fed. The majority of the work relating corn silage maturity to animal armance has been done with lactating dairy cows. The reports by s and Ormiston (1964), Gordon, Derbyshire_and Humphrey (1966), Gordon, rshire and VanSoest (1968), Marshall e£_al. (1966) and White and :on (1929) all showed no increase in milk production related to ~ity of the corn plant. Huber, Graff and Engel (1965) reported a ficant increase in milk production as maturity increased, but con- d that this was due to an increase in dry matter intake, and that Hly there was no effect on the efficiency of milk production. Later, , Thomas and Emery (1968) reported milk yields were decreased with sing maturity of silage. Reports of feeding trials with growing animals are not so numerous work with lactating dairy cows. However, a considerable number of s have reported results using various corn silage maturities. In f these reports, the silage has constituted a larger portion of tion than is the case with the lactating cow. Therefore, these rials are a more dir unfunded with a dif nller, Warner, Rumse vlgteen chop corn wz' significantly faster son with the corn si lanb (1965) reported and59.7%. In this 2.35 lb., and 2.17 l significantly faste: 7.41 and 7.78 for tj level [1969) report latter silage. The silages were fed tc the wet silage, am conversion was exP: land of gain in m tilt dry matter) , another study at I hitter feed convei higher feed intake lain, higher dres: net retum Per ac: significant diffe 10% or 45°, dry ma significpmt diffs I [1964) rePlanted p 28 v ials are a more direct evaluation of the silage being fed but yet are nfounded with a difference in dry matter intake. In one study, by 11er, Warner, Rumsey and Hill (1963), heifers were used in a comparison green chop corn with corn silage. In this report, the heifers_gained gnificantly faster.(l.66 lb./day) on the green chop material in compari- rwith the corn silage (1.10 lb./day). Zimmerman, Newmann, Hinds and m (1965) reported three moisture levels of corn silage; 72.7%, 66.4%, .59.7%. In this study, the average daily gains reported were 2.16 lb., 5 1b., and 2.17 lb. per day, respectively, with 2.35 lb. being a nificantly faster gain than the other two. Feed efficiency was 7.31, 1 and 7.78 for the three moisture levels, respectively. Burroughs and e1 (1969) reported work at Iowa.State University using 32% and 44% dry ter silage. They found no difference in average daily gain when the ages were fed to beef steers. However, feed efficiency was better for wet silage, and the wetter silage produced greater returns. Feed version_was expressed as net energy for maintenance plus gain per 1d of gain in megacalories. These values were 6.65 for the wet silage dry matter), and 6.48 for the dry silage (44% dry matter). In her study at Iowa State University, Fowler 23_al. (1968) reported er feed conversion for a 32% dry matter silage. However, because of er feed intakes, the 45% dry matter silage showed a higher rate of. , higher dressing percentage, higher carcass grade and therefore a higher return per acre. The Minnesota group (Goodrich g£_al., 1967) found no ificant difference in daily gain when silages were fed at 32% versus or 45% dry matter. Likewise, Klosterman e£_al. (1963) reported no ificant difference in average daily gain. In later work, Klosterman 4) reported-heifers gained slightly faster with slightly better feed efficiency when fed hatter silage. Thi Perry, Mohler, and trials comparing s:' were consistent di: this trial favorin analysis was repor There have lost due to passag reported that an a consumed in corn 5 was true when sil; reported by Buck .' tor, there was we at 35% dry matter centimeters was 0 respectively, Th [1963), that the corn Silage, is C Many authc on the varioUs p; been Surname. batters (62°. dry would Cause the ; heater field dr humane COntent (8.930. ,5! 5 479 A 29 iciency when fed a 41.4% dry matter silage compared with a 37.7% dry ter silage. This difference was attributed to an increased feed intake. ry, Mohler, and Beeson (1961) reported on two extremely short feeding als comparing silages of 29.7% dry matter and 37.0% dry matter. There a consistent differences in average daily gain and feed efficiency in 5 trial favoring the 29.7% dry matter silage; however, no statistical .ysis was reported. There have been many questions relative to the amount of energy ; due to passage of kernels in the feces. Huffman and Duncan (1959) trted that an average of only 2.7% of the whole kernel dry weight umed in corn silage was voided in the feces of lactating cows. This true when silage averaged from 26% to 28% dry matter. In a trial rted by Buck gt_al. (1969) relating kernel passage to silage dry mat- there was very little difference found. When working with silage. 5% dry matter, relative to 40% dry matter, the average sieve size in imeters was 0.676 compared with 0.634 for the two dry matter levels, actively. Therefore, it could be concluded, as did Coppock and Stone 3), that the loss in passage of whole kernels when cattle are fed ‘silage, is of relatively little importance. Many authors have summarized the effect of maturity of corn silage 1e various parameters already reviewed. Owens, Jorgensen and Voelker ) summarized their work.by stating that harvesting at high dry rs (62% dry matter) instead of at medium dry matters (39% dry matter) cause the following changes: (1) a lower dry matter yield; (2) er field dry matter loss; (3) higher per cent of ear.loss; (4) lower ene content; (5) lower total acid concentration during fermentation 0 6 vs. 5.47% of dry matter); (6) higher pH (3.88 vs. 4.18). In this nial also, voluntary gains were in favor o {1942] recommended th work indicates that 2 30% dry matter, the I the stalks only 20% . indication of maturi naturity, is the dry 'oyh‘evens and Duncar natter is below 35% iatter is not a goor eluded that "the op feeding value coinc The review t increasing maturit) 111(1) a decrease r lit“; (2) probablj decrease in storag the resulting 5113 intake of Silage c. matter will result signifiCant increr { losses in Convent: on digestibility ‘ themuntary int The Effect , r . .J 30 .al also, voluntary dry matter consumption,.milk production and weight .ns were in favor of the higher dry matter silage. Nevens and Duncan M2) recommended that silage be harvested at 30% dry matter. Their k indicates that at this stage of maturity, when the entire plant is dry matter, the ears will range between 40% and 50% dry matter and stalks only 20% dry matter. The Illinois work also showed the best ication of maturity, as far as a measurement for accurately determining urity, is the dry matter of the leaf portion of the plant, as reported Nevens and Duncan (1949). This is particularly true when the dry ter is below 35%. When above 35%, the leaves dry rapidly and leaf dry ter is not a good indication of plant dry matter. Gordon (1967) con—. led that "the optimum harvest stage for corn as-judged by yield and ling value coincides with lower moisture contents.” The review by Coppock and Stone (1968), reports that the effect of :easing maturity of the corn plant from 20% to 35% dry matter results :1) a decrease of green forage, but a significant increase in dry forage .d; (2) probably a small increase in harvest loss; (3) a_significant rease in storage losses; (4) no consistent effect on digestibility of resulting silage dry matter; and (5) an increase in the voluntary ke of silage dry matter. Allowing the plant to mature beyond 35% dry er will result in (1) little gain in dry matter production; (2) a ificant increase in field losses; (3) the possibility of greater storage es in conventional tower and horizontal silos; (4) no consistent effect igestibility of the resulting silage dry matter; and (5) an increase in voluntary intake of silage dry matter. The effect of fineness of chop on silage parameters and animal per— ince has been studied by very few authors. Many have speculated as to this effect. For in packing depends upon chopped, but they c:' lolic parameters an in this area has be that of Huber, Sand Sandy and Huber (19 {3%, three silar normal field chop : then mixing them w the entire Plant w The metabolic Stud free extracts Was digestibinty was was highest on trw 2‘1 kil(’grams per respectiVelye In M and 26:3 kil With a milk fat I tionperiod (100$ fat were attribuw work reported by TepOrted' This lain when growin agreed with Mill In a“Otkre w on kernel paSSag 31 ~effect. For instance, Benne-and Wacasey (1961) stated that proper, ing depends upon moisture content of the crop and how finely it is ped, but they cite no particular experiments. In reference to meta: c parameters and actual animal performance, the most extensive work his area has been done by the Virginia group. Two different reports, of Huber, Sandy, Miller and Poland (1966) and that of Miller, Poland, y and Huber (1968), review this work. In the 1968 report by Miller l., three silages of 44% dry matter were utilized. The first was a a1 field chop silage. The second treatment was grinding the ears and mixing them with regular field chop stalks. In the third treatment, entire plant was rechopped. All treatments were made prior to ensiling. metabolic study indicated digestibility of the dry matter and nitrogen- extracts was higher on treatments two and_three. Crude protein stibility was highest on treatment three, and crude fiber digestibility righest on treatment one. Dry matter intake values were 2.2, 1.9 and (ilograms per hundred kilograms of body weight for the three treatments, actively. In the milking trial, fat corrected milk (FCM) was 28.0, and 26.3 kilograms per day for the three treatments, respectively, a milk fat test of 86%, 75.8% and 81.1% compared to the standardiza— period (100%). The low values for treatments two and three in milk ere attributed to a low acetate to propionate ratio. In the earlier reported by Huber §£_al. (1966) a similar milk fat depression was ted.. This earlier work also reported no difference in average daily when growing dairy heifers were fed the silages. However, this study. 1 with Miller_e£_al. (1968) on differences in digestibility coefficients. In another study, Buck 33 al. (1969) reported the effect of recutting 'nel passage and feeding value of corn silage. In this work, recutting thepiant before ens feces but had no sir digestible energy v: lower in dry matter this may account fc (1963) reported no with the flail tho; the flail Chopper ‘ it would break all Mary In summary, maturity to harves the range of 30% 1 workers, this woul There has ' of fineness of ch cluded that a fin 32 plant before.ensiling reduced the amount of kernel dry matter in the s but had no significant effect on the total digestiblenutrients or stible energy values of the silages. In this study, Buck used silages er in dry matter than those used in the work of Huber and Miller, and w may account for some of the differences he.reported. Kolari et_§i, 6) reported no differences in cattle performance when silage was made tthe flail chopper as compared with a regular chop. In this work, flail chOpper was hypothesized to affect cattle performance because ould break all kernels. This was found not to be true in that study. gry In summary, most authors agree that the most,effective stage of. rity to harvest corn silage to optimize all factors is somewhere in, range of 30% to 35% dry matter; or, using the terminology of the Ohio ers, this would be in.the dent to glaze stage of maturity. There has been very little work completed to establish the effect ineness of chop. However, many authors have eluded to this and con- ed that a fine chop is desirable, if not necessary. Aseries c two feeding trial sertation. Mater headings. All 5: in Experiment 1. M. silage harvestec liarvest began or pressures were l Experime designed by the Withstand an if diameter by 18 were drilled ir effluent to eSl welded in plat: Steel tubing t Effluent. Sil inserted betwe titre welded t( III. MATERIALS AND METHODS A series of four different experiments—-a fermentation study, feeding trials, and a metabolic study--is included in this dis- :ation. Materials and methods are presented under experimental Hngs. All silages used were characterized by the methods described Experiment 1. Experiment 1 - Silage Fermentation Study Design. A 10 x 4 factorial design was utilized to study corn ge harvested at ten harvest dates and ensiled under four pressures.. est began on August 27, 1969 and terminated November 5, 1969. Silo sures were 0, 2.5, 5.0 and 10.0 pounds per square inch (psi). Experimental silos used in this study and shown in Figure l were gned by the author and constructed with 3/16—inch stainless steel to stand an internal pressure of 25 psi. Silos measured 12 inches in ater by 18 inches high (1.39 cubic feet). A series of 3/32-inch holes drilled in the bottom and on opposite sides of the silo to allow rent to escape from the silage mass. A stainless steel funnel was. ed in place beneath the bottom of the silo and connected by stainless tubing to the series of holes on each side of the silo for collecting ent. Silo pressure was applied by using a three-ton hydrolic jack ted between a stainless steel floating plate and a rigid steel struc- welded to the sides of the silo and extending over the tap. 33 Th: i fe”reenter: “rile eqi to fOilow 34 FIGURE 1 Experimental Silo Unit This stainless steel cylinder was used as the silo unit in the mentation study. It was equipped with pressure application and meas; 98. equipment, a seepage collection system and a temperature thermis er follow silage mass temperature. Silo . of Plexiglas locally in a sity Profess inserted in as follows: Air w llhen pressur exerted by t chamber woul the pressure escape hose. cylinder, be applied. Seep; beneath the were filled Cowstructed “Ibex With zero pressur themister c were recorde Four apleliglas and fiVe fee It had a 12_ ”I 35 Silo pressures were measured with a pressure cell constructed of Plexiglas and shown in Figure 2. Pressure cells were fabricated locally in accordance with the design of J. Boyd, Michigan State Univer: sity Professor of Agricultural Engineering. A pressure cell was inserted in the bottom of each miniature silo. Pressures were measured as follows: Air was pumped into the external chambers of the pressure cell. when pressure within the chamber equaled or slightly exceeded the pressure exerted by the silage mass upon the chamber, the diaphragm covering the_ :hamber would be raised and air allowed to escape through the center of :he pressure cell forming a bubble in a water vessel at the end of the escape hose. The apparatus gave rough estimates of pressure in each :ylinder, but was not accurate enough to determine exact pressures being ipplied. Seepage was collected in a two—liter plastic cylinder placed eneath the collection funnel in each silo. At each harvest, four silos ere filled with the same corn plant material. Three of the silos were onstructed as described. The fourth was a heavy plastic five—gallon arboy with a 3/16-inch Plexiglas plate as a cover which was used as the arc pressure silo. Each silo was equipped with a temperature measuring termister connected to a temperature measuring gauge. Temperatures re recorded at the time of filling, and at 24 hour intervals thereafter. Four silos were filled at each harvest. All four were placed in Plexiglas chamber (see Figure 3). The chamber was four feet square 1 five feet high and equipped with six black plastic surgical gloves., had a 12-inch cube port for entry and removal of samples during the 36 FIGURE 2 Pressure Measuring Cel l This Plexiglas cell was constructed-to measure pressure in the silo mit. It laid flush to the bottom of the silo. Wildi | clam}, 37 FIGURE 3. The Silo Chambers Silo units representing the four pressure levels were placed in a 4' Plexiglas chamber which was infused with 002 to maintain anaerobic. itions. The temperature measurement unit can be seen between. the two bers. L_ ferment so ask of "Mil harves and br filled two to chop w choppe allowe periot ready such in WI was r. procet Silagt be fI‘l thOpp “era. also 38 fermentation period. The chamber was infused with carbon dioxide (C02) to as to maintain an anaerobic atmosphere. Silage. The silage used in this study was harvested from a plot pf ”Michigan 400” corn maintained specifically for this use. At each arvest date, two rows were chopped with a Fox self-propelled chopper nd brought directly to the Beef Cattle Research Center. The silos were illed and placed in the chamber as soon as physically possible (varied from wo.to six hours) after chopping the corn plant. Uniformity of fineness of 10p was maintained by the use of a recutter screen inserted in the topper throughout the ten harvests. The silage from each harvest was flowed 12 days of fermentation in the silo. At the end of the lZ-day riod, the apparatus was dismantled, and the silos were unloaded and_made_ ady for the next harvest which started two days later. By utilizing ch a time schedule, the 10 harvests which were made weekly were run two sets of the miniature silos and anaerobic chambers. Sampling and Data Collection. The weight of the ingoing silage ; recorded as the silos were being filled at each harvest. The same wcedure was followed when emptying each silo. Two samples of the ingoing age were taken; one for oven dry matter determination, and another to frozen for later analysis. However, two to six hours had lapsed between pping and sampling. During the lZ—day fermentation period, datawererecorded daily for page volume, and temperature of the silage mass. A silage sample was D taken daily from the top of each silo by removing the pressure jack floating plate. Pressure was reapplied after taking the sample. age samples were passed out of the chamber through the port so as to 39 maintain an anaerobic atmosphere (see Figures 4 and 5)._ At the end of the lZ—day fermentation period, each silo was unloaded and the contents sampled. All samples were frozen for later analysis. Silage Analysis. A schematic diagram of analysis conducted is :hown in Figure 6. Immediately after thawing the silage samples, total dtrogen was determined-by macro—Kjeldahl procedures and per cent dry atter determined by oven drying for 24 hours at 550 C. (See Appendix V or verification of this method.) Silage extracts were prepared by homogenizing a 25 gram aliquot of he sample in an Lourdes homogenizer with 100 ml of distilled and deionized ater for one minute and straining through two layers of cheesecloth. A 3 ml aliquot of the extract was used for determining pH and soluble nitro- an. pH-was determined on a Corning Model 12 pH meter and soluble nitrogen 5 determined by micro—Kjeldahl procedures. The remainder of the extract was deproteinized using one ml of 50% lfosalicylic acid (SSA) and nine ml of extract. The sample was then ntrifuged at 18,000 rpm for 10 minutes and stored in a refrigerator for ter analysis. Volatile fatty acid content of the silage was determined injecting samples of the deproteinized silage fluid described above :0 a Packard gas chromatograph. Colormetric procedures of Barber and merson (1941) were used to determine lactic acid content of the depro- nized sample. Soluble carbohydrate determinations were made using the deproteinized ract according to the procedure of Johnson et_al. (1966), but modified ghtly. In the modified procedure, the same volumes as called for by nson et al. were used, but instead of allowing the particles to settle, V were spun in a centrifuge at 10,000 rpm for five minutes before 40 FIGURE 4 Sampling Procedure Samples were removed from the top of each silo unit daily. This as done by removing the floating plate and jack to get to the silage mass. fter the sample was taken,.the pressure was reapplied.t 41 FIGURE 5 Sample Removal from the Chamber The daily samples were removed through this one cubic foot port. The port was flushed thoroughly with C02 prior to opening it into the chamber each day. ' 42 FIGURE 6 Schematic Diagram of Laboratory Analysis Conducted on Silage Samples Sample -—————€> Total nitrogen (macro-Kjeldahl) Dry Matter (See Appendix III) E.X T R A C T (25 gm. silage/100 ml distilled water) pH <%————— . 44~_;> Sulfosalacylic acid (SSA) l extract (deproteinization) Soluble Nitrogen Nonprotein nitrogen micro-Kjeldahl) (micro—Kjeldahl) NHS — Nitrogen (Conway) VFA's.(gas chromatograph) Lactic acid (Barker and Sommersen) Soluble carbohydrate 43 aking an aliquot of the supernatant for carbohydrate determination.. This emoved all starch particles which might have remained in the soluble raction. This resulted in an analysis of only the actual soluble carbo— ydrate portion_of the corn plant. Results, using this modified procedure, are very similar to the Johnson g£_al. data; however, values were much ower. Nitrogen.fractionization of the silage was done as follows: (1) otal nitrogen was determined on the freshly thawed silage sample by macro- jeldahl procedures; (2) Total soluble nitrogen was determined on the ater extract of the sample before deproteinization by micro-Kjeldahl rocedures; (3) Using the deproteinized extract, total water soluble onprotein nitrogen was determined by micro-Kjeldahl procedures; (4) The ifference between water soluble nitrogen and water soluble nonprotein itrOgen was called water soluble protein; and (5) Ammonia nitrogen in 1e water soluble nonprotein nitrogen fraction was determined by the l bthod of Conway (1950). Experiment 2 - Feeding Trial 1 The experimental design of this trial was a 3 x 2 x 2 factorial, ilizing 12 lots of steers with 9 head per lot, a total of 108 steers. eatments started were: Three harvest dates, two degrees of chop at, rvest, and two degrees of regrinding at feeding time. Harvesting of Silage. Corn silage ("Michigan 400") was harvested September 18, October 17, and November 14, 1966. Two silos were filled each harvest date; one with a fine chop silage (3/8-inch) and one with edium chop silage (1/2-inch to 3/4-inch). The corn field was divided 0 eight-row plots. Two rows of each plot were harvested in September, 44 v0 in October, two in November, and two were harvested as ear corn mwdiately following the November harvest to establish grain yield per :re. The September and October harvests were ensiled in four 16' x 50' ancrete stave silos, and the November harvest was stored in two 12' x 50' oncrete stave silos. No additives were applied to the silage. Each load of silage was sampled for per cent dry matter during arvest to compute yield per acre and silo storage capacity. October and November harvested silage, both fine and medium chop, ere fed ”reground" and "as ensiled.” Regrinding was done.by running the ilage through a hammer mill immediately after removing it from the silo nd just prior to each feeding. Thus, physical form of the reground fine nd medium chopped silages was the same, which provided an opportunity to etermine if cattle performance differences were due to physical form or o a difference in silo fermentation. The September silage was not eground because its high moisture level caused difficulty in grinding. ‘ Feeding Trial. Choice Hereford steer calves, averaging 475 pounds pen purchased in mid-October, 1966, were acclimated on a ration of corn ilage and protein before use on this trial. They were put on experiment gnuary 13, 1967 at an average weight of 538 pounds. ‘ The cattle were randomly assigned by weight to 12 lots of 9 head ich, and treatment combinations were assigned at random (see Appendix II)° All steers were weighed on two consecutive days and the average of 2? two weights was used as the initial and final weight on the experiment. ey were assigned blocks on the basis of their first-day weight and ran- mly assigned pens from each block following the second—day weight. 45 All lots of cattle were fed twice daily a ration comprised of a full feed of the appropriate corn silage, 1% of body.weight daily in rolled shelled corn, (adjusted every four weeks according to the average lot weight) and one pound of MSU—64 supplement per head daily (see Table 3). All ration ingredients were combined in,a horizontal mixer and thoroughly mixed prior to each feeding. No vitamin A was included in the supplement since all cattle were used in another study which_eva1uated methods and potency of injectable vitamin A on a within—lot basis. All cattle remained on feed for 180 days and were slaughtered on July 12, 1967 at an average weight of 1,036 pounds. Following slaughter, the carcasses were allowed to hang in the cooler for 48 hours, were graded by a Federal grader, and tracings made of the 13th rib. All estimates of carcass quality and desirability were made by a Federal grader. Shrinkage to market averaged 2.43% for all cattle, computed after La lOO-mile haul using slaughter weights over off—experiment weights. Dressing percentage values were computed by using cold carcass weight Tover off—experiment weight. During the course of the experiment, all silage refused by the cattle was removed from the bunk and reweighed. Because of the degree of chop treatment, this was necessary to get a true evaluation of feed efficiency. The amount refused by each lot of cattle can be determined by subtracting the amount of corn silage consumed from the amount fed as listed in the Inimal performance data. 46 TABLE 3 MSU-64% Supplement Formula Pounds Per One Pound Daily Ingredient 1, 000 Pound Mix Provides 45% Feed Grade Urea 130.0 0.64 lb. Protein 50% Soybean Oil Meal 538.2 Ground Shelled Corn 115.0 Dicalcium Phosphate (26.5% Ca. - 20.5% P) 100.0 12 gm. Ca. - 9 gm. P. Trace Mineral Salt 100.0 1.6 oz. .Sodium Sulfate (22.5% S) 8.1 800 mg. iAueromycin (50 gm./1b.) 1.5 75 mg. Vitamin D (9,000 IU/gm.) 2.2 9,000 IU Stilbestrol Premix (2 gm./1b.) 5«0 10 mg. l (Not more than 37% protein equivalent derived from Urea) 4'7 Experiment 3 - Feeding Trial 2 l A 2 x 2 x 2 x 2 factorial design was utilized (16 lots of 8 head ach) to study two harvest dates, two degrees of chop, two concentrate revels and two replications. ] This experiment differed from Experiment 2 - Feeding Trial 1 in hat two concentrate levels were fed. One—half of the cattle received a ation made up entirely of corn silage and supplement and the other group eceived a full feed of corn silage plus 1% of body weight daily in shelled orn and_supplement. Differences in concentrate level were included in his experiment to investigate the possibilities of an interaction between orn silage maturity, fineness of chop and concentrate level. Data from his portion of the experiment are not presented since no interactions, areround. Harvesting of silages. As in Experiment 2 - Feeding Trial 1, the .eld of ”Michigan 400" corn was divided into eight—row plots with two. ms harvested on September 18 (30.7% dry matter), two rows on October 3 4.7% dry matter), and two rows on October 19, 1967 (43.3% dry matter). e October 3 harvest was made by the Dairy Department and only yield ta are presented. To establish grain yields per acre, the remaining two vs of the eight—row plot were harvested in mid-November as ear corn. For the September 18 and October 19 harvest dates, two silos were led; one with a fine chop (3/8-inch) and one with a medium chop (1/2- h to 3/4-inch) silage. The silage was ensiled in four 16' x 50' con- te stave silos with metal roofs; four of the same silos used in Experi- t 1. No additives were used in any of the silages. As in Experiment 1, I load of silage was weighed and sampled for dry matter determinations. entation of silage took place for a minimum of 30 days before being fed. 48 ‘ Feeding Trial. Choice Hereford steer calves, averaging 460 pounds hhen purchased in mid-October, 1967 were used in this trial.- The steers 1ere fed a ration of corn silage and_one pound of MSU—64 supplement per ead daily for 30 days prior to being placed on trial November 17, 1967, eighing 478 pounds. All cattle were implanted with 24 milligrams per ead of stilbestrol on December 13, 1967 and reimplanted with 36 milli: rams per head on April 6, 1968. They were weighed on two consecutive days td the average of the two weights was used as the initial and final weight. e steers were assigned to blocks on the basis of their first—day weight Ld randomly assigned to the reSpective treatment combination following he second—day weight. A11 lots were fed a completely mixed ration twice daily of the )pr0priate corn silage, MSU—64—67 protein supplement (one pound per head. :r day--see Table 4) and the appropriate level of concentrate. Four cattle in each lot receiving 0% concentrates (representing 8 weight range in each lot) were terminated when the average weight of 1 cattle reached approximately 1,025 pounds. The remaining four cattle re left on feed to be terminated when they reached 1,150 pounds. The ne procedure was followed for terminating the lots receiving 1% concen: Ltes. Immediately following the final Weight, all cattle were trucked lmiles to a commercial slaughtering plant, were allowed to stand over- ht and were slaughtered during the morning of the next day. Following ughter, the carcasses were allowed to hang in a cooler for 48 hours are measurements were taken. Loineye and fat tracings were made of the 1 rib. A11 estimates of carcass quality and desirability were.made by »deral grader. 49 TABLE 4 MSU 64-67 Supplement Formula Pounds Per One Pound Ingredient One Ton Mix Daily Sapplies 45% Feed Grade Urea 230 0.32 lb. Protein. 50% Soybean Oil Meal 1,259 0.32 lb. Protein + Cane Molasses 50 Binder_ Dicalcium_Phosphate (26.5% Ca. — 20.5% P) 200 12 gm. Ca. - 9 gm. P TraCe Mineral Salt (High Zn.) 200 50 gm. Salt Sodium Sulfate (22.5% S) 40.4 2.06 gm. 8* \ureomycin (50 gm./1b.) 3.0 75 mg.- fitamin A (10,000 IU/gm.) 13.2 30,000 IU itamin D (9,000 IU/gm.) 4.4 9,000 IU Ratio of.1 part S to 11.3 parts N in urea or 1 part Slto 22.6 parts N n supplement. 50 Shrinkage to market averaged 2.2% for a11.catt1e, calculated by using weights taken after the 100—mi1e haul over off-experiment weights. ressing percentage was computed as in Trial 1. Experiment 4 — Metabolic Study Design. A 2 x 2 replicated factorial design was utilized to study wo silage maturities and two degrees of chop (see Appendix II). Silages tudied were the same as those utilized and described for Experiment 3 — eeding Trial 2. The trial was conducted concurrently with Experiment 3 — ceding Trial 2 and was initiated on March 16, 1968 and completed on arch 23, 1968. Eight mature Cheviot wethers (one year of age) fitted with rumen. Lnulae (Jarrett, 1948) and averaging 31.3 kg. at the start of the trial are.utilized to study metabolic parameters of the respective silages. The leep were fed the respective silage for two weeks in 4' x 4' individual ens before being placed in a collection crate for one week. After the eep had been in the collection crate for one day; feed intake, water take, urine and fecal output were measured and sampled for analysis over six-day period. On the seventh day, while the animals were still in e collection crate, rumen samples were taken just prior to feeding and two, four and six hours postfeeding. Feeding Regime. The sheep were fed twice daily, at 8:00 a.m. and 5:00 p.m. in amounts which assured agilibitum intakes.. The ration wa5~ nposed of the respective corn silage which was removed from the silo it prior to each feeding plus a mineral supplement added at 3% of the .age dry matter. The silage was weighed and thoroughly mixed with the teral supplement prior to each feeding. After all ration ingredients were 51 mixed, samples were taken for laboratory analysis and dry matter deter— ination. Unconsumed feed was weighed, sampled and discarded before the :00 a.m. feeding each day. Water was provided ad.libitum throughout the rial. Sample Collection. Total fecal collections were made by fitting ach sheep with a heavy plastic bag cemented to the posterior of the sheep. eces was removed on a daily basis and weighed. Two per cent was retained or dry matter determinations and frozen for later analysis. Total urine was collected in a two—liter glass bottle which con— ined 25 ml of 20% sulfuric acid and one ml of 10% copper sulphate. The tal volume was measured and then diluted with water up to 1,800 ml. ie-sixth of the diluted urine from each of the six days' collections was atained for later analysis. The pH of the rumen samples was determined with a Corning Model 12 ,meter, and whole rumen contents were strained through two layers of eesecloth to which one ml of saturated mercuric chloride was added per- ml of the strained rumen fluid. This mixture was retained for volatile tty acid and rumen ammonia analysis. Laboratory Analysis. Dry matter of feed presented, the consumed 2d, and feces was determined by drying the sample at 1050 C for 24 hours e Appendix V)._ Silage samples were analyzed in accordance with the cedures outlined and described for Experiment 1 - Silage Fermentation dy. Total nitrogen of the dry feces ground through a 20 mesh screen determined by the macro—Kjeldahl procedure. The micro—Kjeldahl procedure was used to determine total nitrogen urine. 1 52 Rumen voltatile fatty acid concentrations were determined on a ackard gas chromatograph. Samples were prepared by mixing five m1 of trained rumen fluid with one m1 of 25% metaphosphoric acid, centrifuging t 12,000 x g for five minutes. The peak.areas were converted to micro- oles per ml and moler percentages by comparison to standard solutions nalyzed at.the same time. Statistical Analysis All data reported in this dissertation were analyzed on an IBM 300 computer at the Michigan State University Computer Laboratory. lalysis of variance and correlation coefficients-(AOAC, 1960) have been mputed on all trials in order to more precisely define the significant lationships among the variables studied. Because of unequal numbers in periment l - Silage Fermentation Study, a least squares procedure was ed (Harvey, 1960). In the model were included harvests made on ptember 3, 17, October 1, 15, 29 and November 5; pressure levels of 2.5 i 5 psi; and days 1, 2, 3, 5, and 12 of the fermentation. Regular llysis of variance (AOAC, 1960) was used to test the effect of stage of :urity and pressure when the process of fermentation was not included. example of the analysis of variance and the Duncan's new multiple range cedures are shown in Appendix I. IV. RESULTS AND DISCUSSION Experiment 1 — Silage Fermentation Study. Complete results of this experiment.are shown in tabular and, graphic form (Figures 7 through 19 and Tables 5 through 11). Results are summarized and presented on (1) the effects of stage of maturity of the corn plant on the resulting silage after twelve days of fermentation in the silo, (2) characterization of the silage fermentation from the_ time of filling the silo through day 12, (3) interactions of silage_ maturity and rate of fermentation (items 1 and 2),and (4) the effects of varying silo pressures on silage fermentation parameters. As stated in procedures, a two- to six-hour time lapse occurred )etween chopping the plant material and collecting the initial fresh, ;ample. Therefore, fermentation was well under way when the fresh sample as collected as verified by chemical analysis. Results of the fresh ample.ana1ysis are presented, but not included in the discussion due to he atypical nature of the sample. Corn Silage Maturity Dry Matter. Dry matter content of silages at the end of the 12-day armentation for the fresh material, as well as each of the four pressures :udied, are shown in Figure 7. Dry matter values for the fresh material Id the mean values for the four pressures studied are graphically shown .Figure 7. As would be expected, a highly significant difference 53 r______________________________________,$1_1 FIGURE 7 Mean Dry Matter Content Relative to Stage of Maturity 30 —- — — Fresh / // / 20 . Mean I I l I I i 1 t I Sept. Sept. Sept. Sept.. Oct. Oct. Oct. Oct. Oct. Nov. 3 10 17 24 1 8 15 22 29 5 Harvest Date 21.1 23.4 25.2 25.4 31.3 32.1 35.2 40.6 45.2 43.2 18.1 20.7 27.2 24.3 27.4 32.0 31.5 39.1 48.1 42.9 21.6 23.4 27.3 23.8 28.6 29.1 38.0 37.9 48.2 42.7 21.7 24.9 31.0 26.2 28.7 32.0 34.8 36.3 47.7 39.5 26.9 26.1 32.2 29.1 30.6 29.7 33.5 37.2 49.1 43.0 22.1 23.8 29.4 25.9 28.8 30.7 34.5 37.6 48.3 42.0 rsignificantly different (P < .01) error of the means = 0.990 54 55 P <'JN).occurred in dry matter content of silages harvested over the eriod September 3 through November 5. Per cent dry matter of the corn lant increased from a low of 22.1% on September 3 in,a linear relation—. hip_to a high of 48.3% on October 29, an increase of 3% per week through he harvest season. No explanations can be offered for the nonlinear esults obtained on September 17 and November 5. Research such as Nevens E_al. (1954), Johnson et_al. (1967, 1968) and_Huber et_al. (1968) all sport a similar relationship between dry matter content of silage and irvest date. Seepage Volume. Volume of seepage (effluent) escaping from the .10 during the 12-day fermentation, and expressed as ml per 100 grams of lage placed in the silo, is shown and the mean graphically illustrated Figure 8. Volume was-linearly related to both harvest date or dry tter content of the silage and pressure applied to the silo. All dif- rences in seepage volume were highly significant (P<( .01). The early harvest (22.1% dry matter) produced the greatest quantity seepage (16.90 ml per 100 gm. of silage) and, as silage dry matter :reased, seepage volume decreased, until no seepage was collected when 2 silage reached 34.5% dry matter. Murdock (195% reported no seepage en corn silage reached 39% dry matter. These data agree with those of 1er and Clifton (1965) who concluded that seepage loss was determined marily by the dry matter content of the corn crop. As volume decreased, dry matter concentration of the seepage reased, as shown in Table 5. This might be explained by relating the iage dry matter to maturation of-the corn plant. As the plant matures 1 22.1% dry matter in the September 3 harvest, to 25.9% dry matter in September 24 harvest, the starch content would increase with the, rtifldcflldo). ”WMAHUEW Minn-3. .( I. y i... EIIIIf________________________________w"77 FIGURE 8 Mean Seepage Volume (mVlOO gm. fresh sample) Relative to Stage of Maturity of Corn Silage, 15.0 - 0.0 . “" Mean. 5.0 _L l I l l n a l I Sept. Sept. Sept.> Sept. Oct. Oct. Oct. Oct. Oct. Nov. 3' 10 17 24 l 8 15 22 29 5 Harvest Date 12.01 5.18 0.76 0 0 O 0 0 O 0 10.37 14.15 5.39. 5.75 O 0 0 0 0 0 28.31 21.98 17.83 17.72 3.93 3.95 0 0 0 0 16.90 13.77 7.99 7.82 1.31 1.32 0 O 0 o significantly different G’<..01) error the mean = 608.08 56 TABLE 5 Seepage Parameters_ Harvest Date Sept. 3 Sept. 10 Sept. 17 Sept. 24 % Dry Matter 5.46 6.88 10.26 10.42 % Total Nitrogen .1 1.41, 2.28 2.17 1.70 % Ash 1 11.41 10.92 7.01 6.91 1 Per cent of total dry matter.. 58 velopment of the grain portion of the plant. Although analyses were t,conducted, visual appraisal of the seepage samples collected clearly dicated that a large quantity was expelled within the seepage in the ter-harvests. Per cent total nitrogen and ash, expressed on a seepage y matter basis, are also shown in Table 5 with no major differences und. Fermentation of the seepage was in progress when these samples e.collected; therefore, further characterization was not done. Silage pH. Results of pH determined on the silage after 12 days' entation are shown and the mean graphically illustrated in Figure 9. iighly significant increase in pH occurred from 3.52 to 4.65 as the_ ,age increased in dry matter from 22.1% to 48.3% which was significantly related with dry matter (r = 0.64-—see Table 6). Since these silages e not treated with a buffering and/or neutralizing agent such as lime— ne, urea, etc., pH would be expected to reflect total quantity of anic acids found in the silage. This was the case as verified by a 11y significant correlation coefficient of -0.52 and -O.77 (see Table 6) acetic and lactic acid content of the silage, respectively. It has 1 shown at the Michigan Station (Henderson, unpublished data) that e is no relationship between pH and organic acid content when silages treated with buffering and/or neutralizing agents. Often,treated ges with the highest organic acid content will have the highest pH‘ es. When working with untreated silages, pH has been extensively used 1 indicator of silage quality (Barnett, 1954) and these data support concept. However, it is of no value when estimating quality of. :ed silages. FIGURE 9, Mean Silage pH Relative to Stage of Maturity 7.0 “" Mean 6.0 5.0 4.0 I I I i L I l I l Sept. Sept.. Sept. Sept. Oct. Oct. Oct. Oct. Oct. Nov. 3 10 17 24 1 8 15 22 29 5 Harvest Date 3.40 3 69 3.60 3.81 4.00 4.10 4.45 5.45 5.90 5.63 3 50 3.66 3.50 3.80 4.10 4.02 4.25 3.15 4.30 4.20 3.55 3 55 3.50 3.82 4.05 4.30 4.25 4.20 4.20 4.40 3.61 3.60 3.50 3.80 4.00 4.00 4.10 4.20 4.10 4.35 3.52 3.63 3.53 3.81 4.04 4.08 4.28 4.23 4.63 4.65 re significantly different (P< .01) I error of the means = 0.204 59 . H O . V m x. a E . .. oraa> Hmoaufinu H 11 nI11111ta1In11IIIIII1IrIIIlununulnllurnllllllulu. 1 m d N O .m .u P r 1 H A N N . e o w. c M N .... m .n... w. v m .u ..... _ w .w m M 1.... u a a H... .m u n w m... M L A N S S N D e H e S 0/0 0/0 0/0 0/0 0/0 0/0 0/0 p s oo.H ..mm.o ..ae.o ..44.o ..ae.o ..w6.o ..44.o ..6m.o- ..w4.o- ..44.o 0:0 marsfiom oo.a ..oa.o ..ow.o .roa.o ..4a.o ..Hm.o ..Hw.o- ..aa.o- .wm.o eau< canoes a OO.H ¥¥vo.o ¥*OB.O *rom.o om.o skmw.OI stm.Ol ¥mm.O Ufio< UHH$U< w m OO.H aroo.o **©®.O 8800.0 krwn.01 8800.01 *fim.o Z I 22 w N oo.H ..ea.o ..om.o .rm6.o- .rmo.o- .rmm.o 2az ranrfiom o m s oo.H ..om.o ..H6.o- . ..N6.o- ..aa.o z ofinsaom ON: r oo.H ..mm.o- em.o- 6N.o ermonpfiz a oo.H ..vo.o ram.o- Hoops: sum w oo.H .om.o- 2a Aoemfiv amumerema Acapm coaumucoEHom I mcowumfloanou onEflm oo.H HE .oESHo> ommmoom 6O 61 Soluble Carbohydrate. Results of water soluble carbohydrate levels in corn silage conducted on the fresh silage sample and after 12 days in the silo are shown in Figure 10, with mean values graphically illustrated. The erratic values obtained on the fresh sample are pro— bably due to the partial fermentation which had occurred prior to taking the sample as explained at the beginning of this section. A small but :onsistent and highly significant decrease with maturity occurred in soluble carbohydrate content of the silage sample taken after 12 days in :he silo. This resulted in a highly significant correlation coefficient 1f —0.56 (see Table 6) between per cent dry matter and soluble carbohydrate evel. A highly significant (P <\ 4" Po\° / \\ >‘ (H ‘ p a) >\ / II \ 54 L3 0 . ‘z" 2 5‘ ’ \\\ Q . / \\‘H i: 3041.0 0 2_ U; Q o\° v13 "-1 m . U o E ,- 2 < E o U :0 <1: IE 0 O' S -1 H l I (6 I! g I O 10'*0 O- I“ < LA; 1 g 1 ' 0 Fr 1 2 3 5 12 1 Days of Fermentation 75 Soluble NPN Soluble Nitrogen FIGURE 19 Mean Nitrogen Fractionization Relative to the Process of Fermentation 0.04 0.05 0.02 ‘ €0.01 0) p 4.) c6 2 >\ H O KH 0 +5 0.2 0.) U 5.. (D Q. “’ 0.1- (I) N 'U (D W U) 2 g. 0.3 U-l U) (D :3 T5 0 2 > fl H H :5 0.1 L g 1 I L I . Fr 1 2 3 5 12 Days of fermentation 76 77 As shown in Figure 18, pH decreased very rapidly from 5.55 in the fresh material to 4.59 on the first day of fermentation and further reduced to 3.98 during the remaining 11 days. Total acetic and lactic acid levels were 2.02% of silage dry matter on day 1 and were 5.02 on day 12. Thus, reduction in pH is accompanied by an expected level of organic acid accumulation. The rate of production of lactic acid appeared to be linear from day 1 through day 12, Whereas the rate of production of acetic acid appeared to increase rather rapidly through day 3, level off from day 3 to day 5 and then decrease through day 12. Values for acetic acid are in agreement with Barnett (1954) who concluded that acetic acid increased very rapidly through the first two phases of fermentation (day 1 through day 3). He further concluded that acetic acid production con—. tinued at a slower rate thereafter, which is in conflict with these data which show a net reduction in total acetic acid levels following day 5. Barnett (1954) also concluded that lactic acid increases at a slow rate during phase one and two but at an accelerated rate later in the fermenta- tion (phases 3 and 4). These data do not support this conclusion, as previously pointed out. The net reduction in acetic acid production following day 5 may be explained on the basis that acetic acid was used as a bacterial substrate (energy source) by lactic acid producing bacteria which continues to increase in activity throughout the 12-day fermentation period. Soluble carbohydrate levels, as shown in Table 7 and Figure 18 increase very rapidly on day l, appeared to level off on day 2, decrease very rapidly on day 3 and decrease at a very slow rate through day 12. It seems clear from these data that soluble carbohydrate levels and total organic acid levels are negatively associated. 0rganic_acid 78 levels increased at a rate approximately two times greater than the reduction in soluble carbohydrate. Johnson gtpal. (1966a) showed a. similar relationship between.soluble carbohydrate and lactic acid content of corn silage. This relationship would be expected if an anaerobic atmosphere existed in which anaerobic glycolysis could occur by action of the active bacteria. Each mole of simple sugar; e.g. glucose, metabolized in this manner would produce two moles of lactic acid. Soluble nitrogen fractions shown in Table 7 and Figure 19 increased linearly from day 1 through day 12. Increasing values for nitrogen frac— tions during the fermentation would indicate that proteolysis continued to occur throughout the 12—day fermentation period.. Data on total nitro- ‘gen level results are variable and inconsistent, and differences obtained during the lZ-day fermentation are probably due to sampling error. Little or no change in total nitrogen occurred, as would be expected. Interactions Between State of Maturity and Rate of Fermentation. There were four significant interactions between stage of maturity and fermentation rate; (1) pH, (2) per cent acetic acid, (3) per cent lactic acid,and (4) soluble carbohydrate. The mean pH values and deviations from the mean for each harvest and the day of fermentation are shown in Table 8. Upon examination of these data, it is clear that_in earlier harvests, the pH decreased linearly throughout the lZ-day fermentation period, whereas, in later harvests, the pH increased through day 5, and then decreased to day 12. As a possible explanation of this interaction, bacterial activity in the high moisture silage, pointed out previously, would become very active early in the fermentation. This accelerated bacterial growth is stimulated by the availability of high levels of soluble carbohydrate and 79 TABLE 8 Mean pH and Deviations From the Mean Involved in the Interaction of Stage of Maturity and Process of Fermentation Days 1 2 3 5 12 Y 2 ( 0.00) c 0.00) ( 0.01) . {-0.13) c 0.12) [](-0.58) 4.01 4.01 3.91 3.95 3.50 3.88 ( 0.34) c 0.04) {—0.24) {-0.14) {-0.34) 4 .4.59 4.18 4.08 3.50 4.12 ( 0.04) (—0.05) {—0.02) (-0.22) ( 0.25) (—0.17) 6 4.46 4.37 4.29 4.27 4.06 4.29 {-0.02) (-0.03) ( 0.02) {-0.08) c 0.11) ( 0.16) Harves‘ 8 4.73 4.72 4.66 4.74 4.25 4.62 0 ( 0.08) ( 0.04) (-0.12) ( 0.39) 1 5.06 5.09 4.25 4.85 11 (-0.36) {—0.05) ( 0.63) {-0.22) ( 0.54) 4.77 4.97 5.83 4.30 5.00 _ ( 0.13) ( 0 13) -( 0.02) - ( 0.20) (-O.48) X) 4.59 4.59 4.48 4.66 3.98 4.46 80 the anaerobic condition in the-mass. In the later harvests (dryer silage) soluble carbohydrate would be lower and a less anaerobic condition would exist and, therefore, bacterial growth would be less. Concurrently, proteolytic enzyme activity of the plant would continue as normal and form volatile bases which could account for the rise in pH. At some, unknown point after five days of fermentation, the bacterial population becomes active and their production of organic acids overshadows the plant proteolysis, which lowers thede of the mass to the level found on day 12. The mean per cent lactic acid and deviations from the means by harvest dates and days of fermentation are shown in Table 9. These data_ support the proposed explanation for the interaction involving pH, in that the per cent lactic acid increases linearly in the early harvested, high moisture silages. However, in the later harvests, lower in moisture content, lactic acid production did not start until after day 5. The production of lactic acid is thought to be a direct indication of bacterial activity. Silo Pressure The effect of silo pressure on various silage_fermentation para- meters is shown in Table 10. It should be pointed out again that the silage stored in the zero pressure silo was not maintained in an anaerobic atmosphere. Silo pressure had a profound effect on volume of seepage which was nil at zero pressure and increased in a linear fashion to 15.62 ml per 100 grams of silage.stored at 10 psi. For all other fermentation para- meters studied and presented in Table 10, there appeared to be little or 81 TABLE 9 Mean Lactic Acid Value and Deviations from the Mean Involved in the, Interaction of Stage ovaaturity and Process of Fermentation Days 1 2 3 5 12 E 2 (-0.88) {—0.17) c 0.94) ( 0.61) {-0.50) c 2.29) 2.39 3.45 4.94 4.97 5.73 4.30 4 c 0.11) (—0 51) (-0.15) ( 0.55) ( 0.73) . 1.82 1.93 2.65 5.22 2.75 6 (—0.36) (-0.15) (—0 22) ( 0.24) ( 0.49) ( 0.91) 1.53 2.09 2.40 3.22 5.34 2.92 ( 0.26) c 0.12) (40.23) (—0.22) ( 0.07) {—1.02) Harvest 8 0.22 0.43 0.46 0.83 2.99 0.99 10 ( 0.20) {—0.15) {—0.05) (—1.19) 0.34 0.73 2.70 0.82 11 ( 0.87) ( 0.02) (—0.33) (—0.56) (—1.73) 0.12 0.00 0.01 1.65 0.27 -— {—1.03) (-O.67) {-0.29) ( 0.06) ( 1.93) X 0.98 1.33 1.71 2.07 3.94 2.01 82 TABLE 10 Mean Silage Parameters Relative to Silo Pressures different. Pressure per square inch . 4 l 0 2.5 5.0 10.0 s'e' Seepage Volume, m1. 0.000 339.000 706.500 2168.000 384.58 pH 4.403a 3.848 3.982 3.926 0.1288 Per Cent Dry Matter 31.130a 32.060 32.280 33 740b\ 0.6264 Per Cent Nitrogen 1.242 1.196 1.251 1.249 0.0363 Per Cent Soluble Nitrogen 0.2489 0.3102 0.2722 0.3041 0.0173 Per Cent Soluble Non— Ah B protein Nitrogen 0.2118 0.2779 0.2602a 0.2825B 0.0130 Per Cent NH3 Nitrogen 0.0404 0.0471 0.0443 0.0426 0.0032 Per Cent Acetic Acid 1.1180 1.106 1.102 0.956 0.0984 Per Cent Lactic Acid 3.028A 4.466 3.882 4.055 0.2140 Soluble Carbohydrates 14.537 16.528 16.040 16.117 1.2574 1 . Ten observations per mean. aP < .05. P < .01. Values with no subscript or having the same subscript are not significantly 83 no difference between silages stored under 2.5, 5, and 10 psi. The 5 psi. silo would probably be the most,representative of the normal upright_farm silos, as reported by Yu, Boyd and Menear (1963). However, virtually all values differed for silages stored under zero psi. Therefore, all degrees. of pressure applied in this study resulted in an anaerobic atmosphere which produced a high quality silage and no benefit was derived from pres— sures above 2.5 psi. On the other hand, zero pressure was not sufficient to maintain an anaerobic atmosphere. As a consequence, silage produced at 0 pressure was inferior in all fermentation parameters studied. Other authors have reported pressure measurements in the silo, but none of those reviewed reported the effects of pressure on silage fermen— tation. Temperature of the silage mass as shown in Table 11 followed the same pattern as previously discussed. All degrees of pressure applied resulted in virtually no increase in temperature above ambient, whereas a rise of.2.83o C above ambient was observed on day 1 in the silage stored at zero psi. This relative increase in.temperature continued throughout the 12-day fermentation. The temperatures reported in this study are extremely low, ranging around 250 C. They may not be representative of normal silo conditions, because of the small volume of silage and the rapid dissipation of any tem- perature which might have been produced during the fermentation.. Babcock and Russell (1900) concluded that good silage could be made at.these temperatures, however. 84 TABLE 11 Mean Temperatures Expressed as Deviations from Ambient Temperaturestelative to Silo Pressure Pressure per square_inch. 1 s.e. 0 i 2.5 l 5.0 l 10.0 Temperature (Deviation from ambient temperature, degrees C.) Day 1 2.83°A 0.710 0.130 -0.19° 0.354 Day 2 1.05oA -0.10° -0.329 —0.890 0.279 a Day 3 1.010 0.09° 0.06o —0.30° 0.322 Day 4 1.03°a 0.45° 0.29° 0.03° 0.229 A Day 5 0.900 -0.27° -0.35° -0.65° 0.170 A Day 6 0.85° -0.30° —0.55° -0.78° 0.185 Day 7 0.680A —0.619 —0.80° —0.90° 0.304 Day 8 1.00°A —0.54° \40.57° -O.68° 0.314 Day 9 0.050A —1.22° —1.30O —1.33° 0.206 Day 10 0.780A -0.14° —0.24° —0.35° 0.138 A 1 1 Day 11 0.720 —0.45° —0.52° -0.88° 0.210 1 ‘ 1Ten observations per mean } ap < .05 p < .01. Values with no subscript or having the same subscript are_not significantly different. ‘ 85 Experiment 2 - Feeding Trial 1 Weather data, including freeze dates, snowfall and wind velocity, was recorded during the silage harvesting periods and is reported in Table 12. There was no frost prior to the September 13 harvest date. Between September 16 and October 20, the conclusion of the October harvest, freezing occurred on five different nights. There was no snowfall, and the wind reached 20.3 mph on one.day. Between October 20,and November 15, freezing occurred on 14 of the 26 nights. It also snowed a total of four days with a maximum accumulation of nine inches. All snow had melted prior to the November harvest. Maximum wind velocity of 19.7 mph, occurred on one day during this time. Table 13 shows the results of analysis of the silage samples taken during harvest and again during the course of the experiment. These data characterize changes which occur during the fermentation process and are in complete agreement with the results obtained in Experiment 1 — Fermenta— tion Study, involving a wide range of silages harvested at various dry matters which has been previously presented and discussed. Chemical Analysis of Silage. Results of the chemical analyses conducted on the six composite silage samples taken during the course of. the feeding trial are shown in Tables 14 and 15. Dry matter content of the corn silage averaged.28.2% for the September, 48.2% for the October, and 59.6% for the November harvested silages. In each case, the fine chopped silage had a higher dry matter value than medium chopped silage (September, 30.4% vs. 27.9%; October, 49.6% vs. 45.4%; November 60.7% vs. 58.2%). This was probably due to the greater moisture evaporation from the more finely chopped corn plant during the harvesting process. 86 TABLE 12 Weather Data During the 1966 Harvest1 Days Temp. Dropped Below Freezing Snowfall Wlnd 2 Mph high Date Temp. Amount Accumulation during period Prior to Sept. 16 September 16 - 0 First Harvest. Sept. 16 31 26 27: Oct. 1 30 4 20.3 6 25° 12 26° October 20 — Second Harvest Oct 20 280 25 26° 26 23° 29 29° 30 15° 0 Nov. 2 258 2.5" 3 24 5.5“ 8" 19-7 4 90 Trace 9” 5 24° 0.1" 7" 6 29° 5" 7 2H 12 21° 0 13 19° 14 24° November 15 — Third Harvest Nova 15 210 1Weather data reported as recorded at the U. S. Weather Bureau, Capitol City Airport, Lansing, Dates not listed are days during w fall below freezing. Michigan. hich the temperature did not 87 00.00 00.0 00.00 00.00 00.00 00.00 202 00 0 m0 00.0000+ 000.0 000.0 00.000 + 000.0 000.0 00.000 + 000.0 000.0 2 - :z 0 00.00 00.00 00.00 00.00 00.00 00.00 2 0000000 Hops: mo w mm 00.00 00.00 00.00 00.00 . 00.00 00.00 2 00000 00 0 00 00.000 + 000.0 000.0 00.0mm + 000.0 000.0 00.000 + 000.0 000.0 202 0000000 0 00.00 00.00 00.00 00.00 00.00 00.00 2 00000 00 0 00 000.0 000.0 000.0 000.0 000.0 000.0 00000000000 000 z oHQDHOmcH 0000: 0 00.00 00.00 0N.00 00.00 00.00 00.00 2 00000 00 0 00 00.000 + 000.0 000.0 00.000 + 000.0 000.0 00.000 + 000.0 000.0 2 0000000 00002 0 00.0 00.0 00.0 00.0 00.0 00.0 000.0 x 20 aflopOHm opSHu w 00.0 00.0 00.0 00.0 00.0 00.0 00000002 00000 0 ”c0000N0:000o00m coachufiz o mm.o o mm.m o 000< ofiuoo< 0 000.0 0 00.0 0 00.0 0 000< 600000 0 mflmmn 000005 zap 0 co wommohmxo.mo:00> 00< 00.0 00.0 00.0 00.0 00.0 00.0 2 00.00 00.00 00.00 000002 000 0 omcmgu poafimcm swohm omcmgo pofifimcm :mohm omcmco pofifimcm Amman Honfio>oz gmnouuo Honsopmom madfihoumz woaflmcm paw amohm mcwpmaom mHoHoEMHmm owmwfim :00: H HMHHH wcwpmom I N.u:ofifihomxm CH pom: MH mam/0H 88 NN.H wm.H om.o mm.o mm.m vm.o 00.0 mw.o Ho.H mv.m mm.N pHo< UHpoo< w om.N no.m mm.o Hm.H mw.m 0N.o mm.o mw.o mH.N Hv.o mN.m UHU< UHHUmA w 000005 200 we 0:00 00m 0 00 wommmtho m®500> HH< wm.v mm.# om.m mm.v vn.m mv.m mH.m oo.v wH.v on.m wh.m 2m Hw.mv Nm.wv mm.mm mH.wv HN.wN mH.wm 05.00 wm.mv No.mv nw.mN vv.om Houumz Aha w OH HH N H v m ampszz oHHm .002 00000 202 .000 .0000 .002 0000 .002 0:000 .002 0:000 0cmoz 0:0500009 .>oz .000 .0mom vH mqmHpmHom 0000050000 ow00Hw 89 N0.0N mm.mm m0.5m 0N.mN 00.00 0N.5m 00.0m 00.0w 5w.mm 0m.00 5N.50 202 00 0 00 500.0 500.0 m00.0 500.0 N00.0 000.0 N00.0 000.0 N00.0 500.0 000.0 2 I 0:2 w 0w.m0 m0.5m 0m.m0 05.N0 00.w0 00.00 0N.00 00.00 m0.0m 00.00 0N.mm 2 .000 _ 0N2 mo 0 00 0m.0m mm.0m 00.0w 00.0N 0N.0m mm.0m m0.mm 0m.0m 00.0N 00.00 0m.5m 2 00005 mo 0 00 mm0.0 000.0 0wm.0 0N0.0 m00.0 500.0 0wm.0 000.0 mm0.0 mm0.0 0mm.0 202 0003000 0 00.50 0m.w0 5w.M5 05.00 00.0w 0N.05 mm.M5 NM.05 00.00 00.50 N0.N0 2 00009 Mo 0 m0 M50.0 000.0 m00.0 mm0.0 mN0.0 000.0 000.0 000.0 N00.0 000.0 0mm.0 00000000000 200 2 000000000 ON: 0 Nm.Nm 00.0m N0.0N 0N.0m 00.00 05.mm N0.0N 00.0N mw.0m wm.m0 mm.5m 2 000o5 mo 0 00 50m.0 000.0 m00.0 5m0.0 500.0 000.0 mmm.0 000.0 000.0 050.0 00m.0 2 0005000 0m: 0 00.0 mm.m 00.0 00.0 m0.m 00.0 00.0 00.0 00.0 00.0 0m.m 0mm.0 x 20 0000000 00:00 0 00.0 00.0 mm.0 0m.0 00.0 0m.0 0m.0 5m.0 00.0 0m.0 00.0 2 00005 0 000000000000000 000005 200 mo 0:00 000 0 m0 000000mx0 m0500> 00< :0mo000z = 00 00 N 0 0 m 000532 0000 .002 0:00 .>oz .000 .0000 = .002 000m .002 0000 .002 000m 00002 000500005 : .>oz .000 .0000 @000 00 00050500 050 20003002 00 00000 00 0>00000m 0000050000 000000 0 00009.0:00000 . N 0005000mxm 50 0000 m0 mum/0.0. 90 All other factors analyzed with the exception of total nitrogen. showed the same trends which occurred in Experiment 1 and previously discussed. Total nitrogen remained constant (Table 15) across all harvest dates instead of declining as the corn plant matured as in Experiment 1. There were no significant differences in any fermentation parameters between the fine and medium chop. Difference between mean values for harvest dates, with the exception of total nitrogen, were significant (P <’.05); however, none of the differences between fine and medium ch0p proved to be significant. Dry Matter Yield and Silo Storage Requirements. As shown in Table 16, dry matter yield per acre was decreased 10.6% by delaying the harvest 34 days from mid—September to mid—October (5.11 tons vs. 4.57 tons). The trend continued through the November harvest with an additional 11.1% decrease in yield when harvest was delayed 28 days from mid-October to mid—November (4.57 tons vs. 4.06 tons). This gave a combined decrease of 20.5% when harvest was delayed 62 days from mid—September to mid—November. Most of the published data would indicate that corn silage dry matter yield per acre increases until dry matter content of the corn plant reaches approximately 35%. It then levels off for a few days and subse— quently decreases at a rapid rate depending upon weather conditions. (Johnson and McClure, 1968; Huber et_al., 1968; Hanway, 1963, 1966; and Gordon, 1966). Since no harvests were made between the dry matter levels of 28% and 48%, maximum dry matter yield per acre was probably missed. The effect of stage of maturity and fineness of chop on pounds of silage dry matter stored per cubic foot of silo capacity is shown in Table 17. It is interesting to note that in all cases, dry matter stored 91 .umo>aw:,gomso>oz 0:“ Mom mmam can umo>am:.aoaouoo map How .umo>aa: Ho paopmom map How owmfiflm Ea wow mo £0p Hog :Hoo boa .zn mm.v op pcofim>flswo mmz mHLH mm.e Hmam .enoo 2o rmw mo on0a\.sp m.ea on coucsoem owmflflm,wo pmo>gma Honso>oz esp mcwzoaaom waoflx 2900 one ”ouoz www.mm emeaaaou wmfi.mm esfleoz amo.HH- amm.om- amm.mfl aoo.e son.oo mafia ea .>oz amfi.we eoefinaou . www.me suave: aao.ofi- amm.mfl Ham.e sNo.me mean an .boo wfimwa eoganEou saw RN asfleoz Hmo.afi HHH.m aee.om «and MH .omom .pu .pgo 2o wow so woofi . NH o mH m flay dose no mean Eonm omcdgu w oao<\m:oe 2Q .>< ooamoo pmo>amm oao< Mom waofir houumz xan bum hound: Ann :o @030 we mmocosfim was zpfiazumz mo QMMpm mo uoommm ,oaamamoz wo.oH- rm.oHu om.m enamou mm.HH mcflm 5H .uoo wa.on eH.HH mmamou 3.2 on: 2 .Eom NH .uoo mH .pmom moco omywoo cam ommHHm moau _ och cmmzpon {um .59 non mo 4 . wwwmm Eonm omcwgu w ,omzwgu w 2o mo .mnH ooamoo p I mucoEoHstom omeOpm oHHm co mono mo mmocmaHm paw powsumz mo.ommum mo uuowmm NH mHm HH< mm.m mm.m oo.v vn.m mo.v mm.m mn.m mm.m mm mm.mm oo.om om.me oa.om om.oe om.ce ow.om oa.Hm Henna: sea a N H v m 902852 oHHm .ve: cane .oeo .uaem .eez .eeua .eez pane mama: psoEpmoue accepoo Honaopmom N awash mcfiwoom I m pcoafinomxm :a wow: moau mo mmozocflm can xufiHSsz mo omeum cu o>fipmfimm mamuoEmHmm ommafim new: 5N mam00 mo memo Mom 0 m0 commogmxo Wos0e> 00< c0000002000000m comonpfiz = N 0 0 0 .00: 0000 .000 ..0000 = .002 0000 .002 0:00 mama: pcoaquAH = 0000000 Hensoumom N 00000 maavoom : m u:oEaHOQXm :0 com: . moau mo mmodmafim use 00005002 mo ommwm ou o>00d~o¢.mhouo&dhmm omofiflm can: wN mqmflh .111IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIILHH-II 108 TABLE 29 Mean.Silage Parameters Relating Fresh and Ensiled Material, Used in Experiment 3 — Feeding Trial 2 September October Fresh Ensiled; Change Fresh Ensiled Change‘ % Dry Matter 30.70 43°30 pH 5.70 3.74 5.22 4.00 All values expressed on a dry matter basis % Lactic Acid 0 5.35 O 2.90 % Acetic Acid 0 1.98 0 0.82 Nitrogen fractionization_ % Total Nitrogen 1.44 1.22 1.61 1.24 % Crude Protein (N x 6.25) 9.00 7.50 10.06 7.75 % Water Soluble N 0.357 0.522 +146.22% 0.222 0.507 +228.38% as % of Total N 24.79 42.79 13.79 40.89 % Water Insoluble N (by difference) 1.083 0.698 1.388 0.733 as % of Total N 75.21 57.21 86.21 59.11 % Soluble NPN 0.176 0.393 +223.30% 0.201 0.379 +188.S6% as % of Total N 12.22 32.21. 12.48 30.56 as % of Water Soluble N 49.30 75.29 90.54 74.75 % NH - N 0.013 0.052 +400.00% 0.016 0.056 +350.00% as % of NPN 7.38 13.23 7.96 14.78 109 total nitrogen was not exPerienced in these harvests as was the case in Bxperiment.2, both differing from results obtained in Experiment 1. Dry Matter Yield per Acre and Silo Storage Capacity. Average per cent dry matter of the silage and dry matter yield per acre for each harvest date are shown in Table 30. Dry matter yield per acre increased 4.0% (5.64 tons vs. 5.86 tons) between the September 18 and October 3 harvests and decreased 6.1% (5.86 tons vs. 5.56 tons) between the October 3 and October 19 harvest. These results are consistent with results obtained during the 1966 harvest and reported previously in Experiment 2. As con- cluded by Huber e£_al. (1968), corn silage dry matter yield per acre appears to be maximized at about 35% dry matter and little is to be gained by purposely delaying harvest beyond this point. The effect of stage of maturity and‘fineness of ch0p on silo storage capacity is shown in Table 31. The length.of time needed to fill each silo varied slightly, due to weather conditions, available labor, etc. For the mid—September har- vest, the silo filled with fine chop silage required two days, whereas three days were required to fill the silo with medium chop silage. In mid-October, one day and two days were required for filling the silos with fine and medium chop silages, respectively. In no case were silos refilled after initial filling. These time periods are presented due to the pos— sible effect of filling time on the silo storage capacity. Dry matter stored per cubic foot of silo capacity was increased 5.56% (11.93 1b. vs. 12.16 lb.) by delaying harvest from mid¢September (30.7% dry matter) to mid-October @3.3% dry matter). These.results do not agree with results obtained in Experiment.2 where dry matter stored per 110 TABLE 30 Effect of-Stage of Maturity on Dry Matter Yield Per Acre Harvest Degree Per cent Tons/Acre % Change From Date of Chop DM Sept..l8 100% DM 30% DM Sept. 18—22 Fine Medium Combined 30.7% 5.64T 18.8 Oct. 2-4 1 Fine 34.7% 5.86T 19.5 +4.0% Oct. 19—21 Fine 46.3% Medium 40.3% Combined 43.3% 5.56T 18.5 —l.4% 5 1 Harvested for Dairy Department 1 1 1 111. TABLE 31 Effect of Stage of Maturity and Fineness of Chop on Silo Storage Requirements De ree o/ Lbs. Of DM, 04: Chan e Harvest ' 5f DM per cu. ft. between.gfine. % Change From Date Chop of silo & medium chop Sept. 18 Mid—September Silo 3 Fine 31.1% 12.32 Silo 4 Medium 30.2%, 11.55 Average 30.7% 11.93 —6.25% Mid—October Silo 1 Fine 46.3% 13.15 Silo 2 Medium 40.3% 12.13 Average 43.3% 12.64 -7.77% +5.56% 112 cubic foot of silo capacity was reduced 11% by delaying harvest from mid— September (28% dry matter) to mideOCtober (48% dry matter) and further reduced 7% by delaying harvest to mid—November (60% dry matter). This discrepancy may be partially explained by the 1966 harvest requiring one day to.fill each silo and the 1967 harvest requiring two to three days to fill each silo, and thus allowing more time for settling and compaction while filling. A The_September harvested fine chop was 6.25% higher than the coarse chop in pounds of dry matter stored per cubic foot (12.32_lb. vs. 11.55 1b.). The same trend continued in the October harvest (a 7.77% advantage with the fine chop) due to the greater compaction of the finer chopped material (13.15 lb. vs. 12.13 lb. per cubic foot). These results are in agreement with similar results obtained during the 1966 harvest. Mid—September vs. Mid—October Harvested Corn Silage. Pooled results of the effect of harvest date on rate of gain, feed efficiency and carcass quality are shown in Tables 32 and 33.. Complete performance of all lots 1 are shown in Appendix III. Cattle_fed mid-September harvested silage .gained significantly (P < .05) faster than cattle fed mid-October harvested silage (2.58 lb. vs. 2.46 lb.). Their higher rate of gain coupled with a slightly lower-daily dry matter consumption (17.27 lb. vs. 17.62 1b.), resulted in a substantially lower feed requirement per cwt. of gain (660 1b. vs. 716 1b.). These factors were the basis for the lower cost of .gain shown for cattle fed the midsSeptember harvested silage ($11.58 vs. $12.39 per.cwt. gain). All carcass traits favored the cattle fed the September harvested silage; however, these differences were sma11_and non— significant. These results are in complete agreement with the results previously reported in Experiment 2. III:::______________________________________"—"’' 113 TABLE 32 Effect of September vs. October Harvested Corn Silage on Rate of Gain and Feed Efficiency I (November 17, 1967 to July 1, 1968) l September October 214 Days on Experiment. Harvest Harvest 1, 2, 8, 9, 3, 4, 5, 6, Lot Numbers 10, 12, 7, ll, 14, 15 13, 16 No. of animals 64 64 Av. initial weight, lbs. 478 478 Av. final weight, lbs. 1031 1004 Av daily gain, lbs. 2.58a 2.46 Av. daily ration, lbs. Corn silage fed 38.64 28.23 85% DM shelled corn 3.11 3.11 Protein supplement 0.98 0.98 TOTAL 85% DM basis 17.27 17.62 Feed consumed per 100 lbs. ‘gain, lbs. 669 716 Daily feed consumed per 100 lbs. body weight, lbs. TOTAL 85% DM basis 2.29 2.38 Concentrates 2 1.24 1.28 Roughage 1.05 1.10 Concentrate:Roughage Ratio 3 24:76 23:77 Feed cost per 100 lbs. gain 4 $11.50 $12.39 Live selling price per cwt. $25.30 1 $24.32 1 Performance data includes all animals in the treatment, whereas car— cass data includes a random slaughter of one—half of the animals. N Does not contain grain content of corn silage. 3 Does contain grain content of corn silage. 4 Feed prices used: Corn silage - $7.50 per ton on 30% DM basis; Shelled corn — $1.20 per bushel; MSU—64 supplement - $5.50 per cwt. Values with no subscript or having the same subscript are not signifi— cantly different. A = (P < -011, a (P4 .05). 114 TABLE 33 Effect of September vs. October Harvested Corn Silage on Carcass Qualit (November 17, 1967 to July 1, 1968) 214 Days on Experiment September October Harvest. Harvest Lot Numbers 1, 2, 8, 9, 3, 4, 5, 6, 10, 12 7, 11, 14, 15 13, 16 Carcass evaluation No. of animals 32 32 Carcass grade 5 11.34 10.88 Marbling score 6 . 14.19 14.19 Fat thickness, 13th rib, inches 0.60 0.72 Ribeye area, sq. inches 11.18 11.02 % Kidney, heart and pelvic fat. 1.92 1.61 Cutability 7 49.65 49.15 . Cold carcass weight, lbs. 605 587 Dressing per cent 58.50 56.64 Carcass price per.cwt. $43.25 $42.94 Beef produced per acre of corn fed, lbs. 1667 1558 Gross returns per acre of corn fed $422.00 $377.00 5 Carcass grade values: 7 = Standard; 10 = Good; 13 = Choice; 16 = Prime 6 Marbling values: 11 = Slight; 14 = Small; 17 = Modest; 20 = Moderate; 23 = Slightly Abundant 7 Per cent boneless, trimmed, retail cuts Values with no subscript or having the same subscript are not significantly different. A =_(P 4 .01), a = (P < .05)._ \ 115 Fine vs. Medium Chop Silage. Pooled results of all fine and medium chop comparisons (September and October combined) are shown in Tables 34 and 35. Cattle fed fine chop silage gained at a slightly faster rate than cattle fed medium chopped silage (2.55 lb. vs. 2.50 lb. daily). Likewise, dry matter consumption was slightly greater for the fine chop silage fed ‘group than the medium chop silage fed group (2.39% of body weight daily vs. 2.28%). The cattle fed the fine chop silage produced significantly higher grading carcasses (11.41 vs. 10.81) which resulted in a significantly (P < .05) higher carcass price ($43.30 Vs. $42.89 per cwt.). For all other comparisons, differences were small and nonsignificant.. Again, results are in agreement with results reported in Experiment 2. Experiment 4 — Metabolic Study Rumen pH and VFA Concentrations. Results of this comparison are shown in Table 36. Neither stage of maturity nor fineness of chop significantly influ— enced mean rumenva (Table 36). The mean rumen pH for the September harvested silage was 6.12 and for the October harvest, 6.17. The fine Chopped silage produced a mean rumen pH of 6.16 while the medium chopped silage produced a mean of 6.13. Rumen pH is primarily due to the concentration of volatile fatty acids in the rumen, which can occur either by ingestion of feedstuffs con- taining volatile fatty acids or from rumen microbial fermentation. Rumen pH values for the sheep fed these silages relative to time after feeding (Table 36) exhibit a normal.pattern (Fenner et al., 1967) of decreased pH during active fermentation up to two hours postfeeding and then increased PH as fermentation declines by six hours. 116 TABLE 34 Effect of Fine vs. Medium Chopped Corn Silage on Rate of Gain and Feed.Efficiency . (November 17, 1967 to July 1, 1968) 1 Fine Medium 214 Days on Experiment Chop Chop l 2, 3, 7, 10, 1, 4, 5, 1 Lot Numbers 11, 12 6, 8, 9, 14, 16 13, 15 No. of animals 64 64 Av. initial weight, lbs. 477 479 Av. final weight, lbs. 1022 1014 Av. daily gain, lbs. 2.55 2.50 Av. daily ration, lbs. Corn silage fed 32.82 34.04 85% DM shelled corn 3.13 3.09 Protein supplement (MSU 64-670) 0.98 0.98 TOTAL 85% DM basis 17.90 16.99 Feed consumed per 100 lbsl gain, lbs. TOTAL 85% DM basis 702 . 680 Daily feed consumed per 100 lbs. body weight, lbs. TOTAL 85% DM basis 2.39 2.28 Concentrates 2 1.26 1.24 Roughage 1.13 1.04 Concentrate:Roughage Ratio 3 23:77 24:76 Feed cost per 100 lbs. gain 4 $12.17 $11«80 Live selling price per cwt. $25.45 $24.17 1 Performance data includes all animals in the treatment, whereas carcass data includes a random slaughter of one—half of the animals. The remainder were fed to heavier slaughter weights. 2 Does not contain grain content of corn silage. 3 Does contain grain content of corn silage. o _ 4 Feed prices used: Corn silage — $7.50 per ton on 30% DM ba51s; Shelled corn - $1.20 per bushel; MSU—64 supplement - $5.50 per cwt. values with no subscript or having the same subscript are not significantly different. A =‘ (p < .01). a = (P< .05). 117. TABLE 35 Effect of Fine vs. Medium Chopped Corn Silage on Carcasstuality (November 17, 1967 to July 1, 1968) : Fine Medium 214 Days on Experiment ! Chop Chop 2, 3, 7, 10, l, 4, 5, Lot Numbers ‘ 11, 12, 6, 8, 9, ‘ 14, 16 13, 15 Carcass evaluation: No. of animals 32 32 Carcass grade 5 11.41a 10.81 Marbling score 6 14.13 14.25 Fat thickness, 13th rib, inches- _ 0.60 0.73 Rib eye area, sq. inches 11.16 11.03 % Kidney, heart and pelvic fat 1.83 1.70 Cutability 7 49.72 49.15 Cold carcass weight, lbs. 599 593 Dressing per cent 58.78 56.36 CarcaSs price per cwt. $43.30a $42.89 5 Carcass grade values: 7 = Standard; 10 = Good; 13 = Choice; 16 = Prime. 6 Marbling values: 11 = Slight; 14 = Small; 17 = Modest; 20 = Moderate; 23 = Slightly Abundant ' 7 Per cent boneless, trimmed, retail cuts. Values with no subscript or having the same subscript are not significantly different. A = (P.< .01), a = (P.< .05). QNOE 80AM wfiOflubfirHQmo'O HDOHH .cmoE Mom mcoflum>aomno 039w mmo.o MH.o 0H.o hH.o Na.o ovuo.o ma.o 0H.® mo.o 0H.w .M . 1. . o mNO.o wH.w MH.© mN.© no.0 mmo.o mm.© ON 0 or 0 mo 0 H . . v ovo.o 00.0 mo.® 00.0 mo.o omo.o 00.0 00.0 00 0 mo 0 H . . . . N mno.o mm.m mo.o 00.0 mm.m HOOH.O mm.m mo 0 om m 00 o H . . . o wOfi.o ww.o om.o ov.o wv.o ONmH.o o¢.o ow 0 mm o 00 o H v.6.m suave: mafia .poo .pmom a.m.m suave: oqflm Beebe: ocflm QEHH mcafimemm memo: uncapmoae nonouoo Hopaoumom mozfin> mm cmESm mo mamoz om m4m cossm new: mm mamnomno 950m :moE.Hom muoauw>uomno ozkm H . s annunpannnwnn monaa Hm.mc wh.mo vH.mo vo.m© omv.m mm.mo om.wo mo.wo mm.mo kuumz kHQ . mamu.noupmz mw.wha on.wnm MM.ONN ON.BMN mv.mhd om.wom ON.me om.mwN NAHQ Hmoom nawu.oxmucfi uo.mv smH.mwm amn.maw mv.on mv.Nmo vn.m© mm.mom om.Hmw mm.omm mm.hwh Hound: zho .o.n eases: mesa .uoo .uanm .n.n aspen: mesa sedan: mafia N H memo: “coaunone nmnouoo Honsoumom wkopoEmnmm moogm new mcmoz mm,mgm< .05) for the lambs fed the fine vs. medium chopped silage (4.82 gm vs. 3.51 gm). A difference of this magnitude would be expected due to the significant increase in dry matter intake between the lambs fed the fine and medium ch0pped silage. There was no significant difference in fecal nitrogen. for the lambs fed the September vs. October harvested silages. .cmoe Hog m:0flua>nomno Aden .cmoE.nom macapm>nomno ozhm .muo.o u anemonMaemnmw m4.am mw.w4 am.mm Nu.om em.He mw.m4 4e.mm 4s.n4 4 spannnnnnnmhe z cannnaa< Na.o4 em.w4 4n.a4 wN.O4 mm.e4 mm.m4 mm.mm mo.e4 ennnomnn 2 no a mm 2 enennpnm mm.mN mm.mm mm.ow mH.om on.mm Nw.vm Hm.nH 44.NN OMMpnfi 2 mo w mm 2 vocfimuom mH.4m AN.mN mm.w~ mm.am 6N.mm mm.mm 4H.mm om.m~ manned 2 do a mn.z Aeneas: mm.N4 AH.Hm m4.44 mm.m4 4o.wm mH.om 6m.a4 6N.Nm orange 2 mo s we 2 Hanna Num.o mm.u NN.N M4.N NA.H 4Na.o e4.m H4.N mm.H 4o.N flame\5mv.eomonpne enefinpne mm.e om.a Na.e ow.e HH.6 mm.a mm.e mo.a flsne\5mv.cnmonune ennnnoxm mmm.o Nw.N mm.m me.m mm.N mum.o mw.m 64.N an.m om.N flane\2nv.:omonune semann: 4A.4. 06.4 mo.m nm.4 Hm.m 6w.4 wfi.4 4m.4 mame\5mv.eowonpne eonnonn< omm.o .Hm.m .Nm.4 no.4 mm.4 mum.o 6N.m mm.4 ea.m ma.4 flsne\smv_:nwonpae Hanna mem.o mm.w N4.m 6H.m mm.w 4om.o am.w mn.m 46.5 mo.m fisne\emu.6snoea enmonuuz «.6.m sense: mafia .uuo .panm H.6.n Essen: mafia sedan: mean memo: psoEpmonH Honouoo nopfiopmom Annum oocwflam :omoapflz now memo: mm mqm< Ho.N mm.N om.N mm.N mn.N mm.N mw.N Nv.N .wnH .chm.xHHmw .>< mam wooH Hmm NHoH oNOH VNoH mvoH omoH .mnH .uzmHoz Hmch .>< wmv Luv mmv wmv Luv mwv mnv wnv .mnH .HHMHos HmHuHcH .>< oH oH 0H 0H 0H 0H 0H 0H maooum,wo .oz OON wNN OON wNN OON wNN OON wNN unoEHHomxm co mxmn mH w o m w v 0H w m HH w n m w H mH w w NH w oH vH w N .02 p04 NH we NH we NH we NH No woumaucoocoo mo Ho>oH eusoz mch susoz och mono mo beamed umo>amm Honouoo ,pmo>gm: Honsoumom umo>amz mo open Ho>oq oumaunoucou pcou Hod one H HwomH .H AHBN ow NH Honso>ozv .m> onoN .moau EDHUQZ .m> onHm MN mqmgm: Honoooo .m> Honfioumom 148 .ucmwcznm xHuamHHm u mN .0 Hopmpowoz n.0N EHHQ H OH we 3m: MH2333 Mom 0N.Hm n :Hou voHHoaw mo HousmzmHm Eowcma m mo ..muamHoz Hougmdem HoH>mo: 0p vow wsHocH mpmw mmmoamo mamaogz .pcosu HmHmas 2m New “mono: HQUHOHU n mH M NH :0 20p Hon om.nm I .mHSo HHmuoH .voEEHHu .mmoHocoa pcooHom MHHasmnyH ”@000 H OH HHEMHHm u HH Hwhmvcmum u n ommHHm :Hou .owwHHm choc mo ucounoo :HNHM :Hmucoo mooo m l\ ”mosz> mcHHnsz o "mous> opmam ammoamu m .uzo Mom om.mw n unoEonmsm ”wow: mooHHm poem N .ommHHm :Hoo mo pcopcoo :Hmum :Hmpnoo no: moon N wcha cam HowchEoH 6:5 mono esp EH mHmEHzm HHm mopsHocH dump oozmaaomaom H .mHmchm ogu mo mHmsuozo .m wcm v moHan waaomg< How mouocgoom om.Nvm oo.mvw mN.mvw ww.NvN mN.mvw mo.va mo.mvw vv.mvm .pzo Hog ooHHm mmmoamu Nn.mm mw.vm wn.wm NN.mm em.wm mm.nm v0.00 wo.nm N .ucoo Hem mchmoHo wHo vmm mum mam HHo Hmm wNo Hmm .mnH .oanos mmwouwo pHou an.mv mm.om mm.om on.mv vm.wv nm.om mo.mv ow.wv n quHHnwpsu om.H om.H mN.H mo.H oo.N mm.H mH.N Hw.H Ham oH>HoAH , paw pHmo: .xmszM N wm.oH mm.oH em.oH mm.HH NN.HH Hm.HH ,mm.oH ON.HH .:H .Um .moam.oxonHm Hm. mm. me. we. Na. av. No. mm. .cH .HHH :pmH .mmoeHUHze pea oo.nH oo.vH mN.mH om.NH mo.mH mm.NH wm.oH wm.vH o mgoow mcHHnHmz mN.oH mN.HH mH.HH ww.oH mN.HH om.oH mH.NH om.HH m owmnm mmmohmu w w w w w w w w mHmEHc< mo .02 ":0Hpm5Hm>m mmmuamu om.NNw No.MNw Nv.me mm.me mv.me NN.VNN 0N.on ow.VNw .pzo Hem ooHHm MEHHHow o>HH NH No NH No NH No NH No mopaupeoocou Ho Ho>mH ESHwoz och Eszoz och mogu Ho oouwoo umo>Hm2 Honouoo umo>Hmz Honsoumom pmo>Hmz Ho mama HwoaH .H Hst on HemH .HH Hanam>ozu Ho>oH opmaucoocou pcou Hem oco .m> 0H6N .mozu aszoz .m> och .umo>me HopOpoo .m> Homaoumom 3N mHm cossm mm.mN wH.o 05.0 wn.m nw.NH N mm MH.N4 mm.m mN.H oo.wH mm.NH H NM fl mm.om om.n mw.o on.VN mm.NN v MM 1 NM.Nm mm.m mo.H oo.HN vm.vN v OM Nw.om Hn.m MH.H no.mH Hm.ON N MN NH.Nm Hm.HH on.H ow.mH HH.wN H MN NB.N@ Nm.HH ow.H VM.mN mo.MN M NN ON.vv No.0 ov.o ww.mH Nm.NH M 0N HmHOH QHHLHDM oHHAHSHOMH oHcOHmon oHuoo< .oz .02 oHHm momgm N o mHm cossm 153 MH.HM 05.5 N0.H Nw.w H0.MH N mm 5M.0m M0.0 ww.H M5.0N HH.HN H VM ov.Hv Nm.m Ho.o Hv.oH ov.mH v MM Ne.mv Ne.o No.0 NH.5H N0.HN N om MH.wv NH.0H NM.H NN.mH 5M.HN N MN NM.oo vo.NH MH.H Hm.oH v0.0M H MN No.0v MN.5 Hw.o M5.mH MN.HN M NN NN.om 5N.w vH.H 0N.0N HM.0N M 0N HmpOH UHHAHSN ngxwznomH quOHmogm oHuoo< .oz .02 oHHm ammam v 5 pm HHE\E:U mvHo< xuumm oHHumHo> coszm 5 mqm55 mvHo< Auumm oHHumHo> cmesm w mHm 5vzpw UHHOHmpoz I v pqoaHHomxm I IIIIIIIIIIIIIIIIIIIIIIIIIIIIII------—-—---———— Hopuma Houuma HHQHPNE Houuma Houuws Houuma Houume Houudfi .Hmupaa oxmuaH comonqu oxmucH :oMOHsz 5H0 5H0 590 5H0 5H0 590 5a 5H0 5H6 'U oHanmmee N oHHHpmomHe N oHHHHmomHe N oHnHHmmme N oHnHumomHv N mHnHHmomHe N oHnHomomHv N oHHHHmoNHw N oHnHomomHv N oxwucH genome 5H0 oHMHcH Houuwe 5H0 .oxwucH Hopuma 5H0 oxmucH Hopume 5 oxmch.HouumE 5 oxmunH Houuma 5 oxmucH Hounds 590 oxmucH Hopuaa 5 oxach Hopgma 5 oxmqu Houpms 5 Hoppma 5H0 ommHHm Houpme 5H0 omeHm Hmugma 5H0 omeHm Houpma 590.6maHHm Hopuae 5Hw ommHHm HoupwE 5HV.0MMHHm Hmpume 550 cudafim Hounds 5H0 omeHm Hoppma 590 ommHHm HoprE 5H0 ommHHm HopumE 5H0 omaHHm IIIIIIIIIIIIIIIIIIIIII 1mpcoHonmoou.:0HpmHonaou 6H H oHnaHHm> lIIIIIIIIIIIIIIIIIIIIIIIIII daHm H mam<fi I >H XHQmem< n IUF~0U\ n .1 slay... 1 x.» >LI..I.I.ILILI NNN5.0 UHom UHHUmH omeHm zmz ommHHm www.m Mmm smaHHm 2az,omaHHm mmm.o z 32 omaHHm zNz.6maHHm MHN.oI 0Hom onomH ommHHm noNOHuHc oanHow Houd3.omeHm *wmo.0 Mam ommHHm cowoanc oHnsHom Honda ommHHw *mo5.o z :z ommHHm nomopr: oHnsHom Hops: ommHHm 05v.0I zaz omeHm comoapHn oHnsHom Hops: omnHHm 0N4.o eHoa 6HpuNH owaHHm 66H6:606H.=6N6H6Hz 50H.0 Mam omeHm quucouoH QQMOHHHZ H5N.0I z :z QMNHHm coHpcoHoh :omoysz owN.0I , . an omeHm :OHucouoH cowoanz Hoo.0 cowoauH: oHnsHOm Heads omeHm :oHuaopoH somoHsz 500.0 0Hom QHHONH oMMHHm ,comoypH: 5H¢CHHD “w MNm.o Mum ommHHm cowomuHc 5hmcHH0 11 z :z.oMMHHm :mmoHHH: 5HmcHH0 _ 2&2 ommHHm :omowuHc 5hmcHHO comoHpH: oHnsHOm Nona: omaHHm comoupHc 5amcHH: :OHunouoH :omoHsz dowoanc 5chHHD wHom oHpomH ommHHm .comoanc Hmoom mm ommHHm cowoprc Hauom z Mmz.ommHHm comoan: Hmoom . zmz mmmHHm ammoan: Hmoom cowoapH: mHnsHom Noam: ommHHm somoan: Hwomm EOHHEonH somoHHHz :monoH: Hmumm comoan: 5HmcHH0 :oMOHuH: Hnoom wHom UHpomH ommHHm ,oxmucH comowqu mm omeHm oxmch :omoyqu 2 M22 omeHm onch cowonqu oxmucH somoyqu oxmsz :QNOMsz oxmch :OMOHHHZ IIIIIIIIIIIIIIIIIIIIIIIII H.6HanHN> IIIIlllllllllllllllllllllll . zaz omaHHm cowoan: oHnsHom Hopmz ommHHm COHpcopoH comoysz HH oHHNHHm> 505um oHHonmuoz I v pcoaHHomxm I mucmHoHHmmoo :OHHNHeNHou onEHm \o)«‘I‘IJ .0 ‘1‘)..61 .( I. 157. oemH .Hoooeoem H NNN. HHo. v,av I. . owe. Hmo..v.N0 I IIIIIIIIIIIIIIIIIIIIIIIIIII. HmosHm> HmoHanu 13.0.3.? MONMO eHoa 6HpuaH omaHHm mm omaHHm mm o eHua oHpumH ommHHm .. z Mzz emaHHm ma omaHHm z mmz oNaHHm I1II||IIII1|11|IIII .HH oHHNHam> H oHanHm> i111|1|ln||11 5vsum UHHonmpoE I v pamEHHomxm I muaoHUHmmoou GOHumHoHHou onEHm H.06660H mHmH xHQZmNN< APPENDIX V VERIFICATION OF DRY MATTER DETERMINATIONS VERIFICATION OF DRY MATTER DETERMINATIONS The accuracy of dry matter determination of feedstuffs containing volatile acids and bases has been questioned by many authors, but no accurate method has been devised which does not involve use of exhaustive extraction and distillation techniques. The two methods most commonly used in work with silages and similar fermentation products are oven—drying and distillation. Distillation pro- cedures involve the use of organic solvents which are nonmiscible in water and have a boiling point higher than water (commonly toluene is used for this purpose, Bidwell and Sterling, 1923). Oven—drying at 650 C, as described by Barnett (1954), not only removes most of the water but also some organic matter, distinctly noticeable because of the pleasant aroma associated with drying silage (Fenner and Barnes, 1965). Forbes (1943) used drying in a vacuum oven for 22 hours at 500 C and employed a closed system, drawing heated, dry, and COz-free air through the sample into a red—hot furnace, where a platinum catalyst oxidized the organic matter into CO2 and water. The water and CO2 were trapped quantitatively in concentrated sulfuric acid and flaked sodium hydroxide, respectively. The increase in weight of sodium hydroxide represented the CO2 from the oxidized organic matter removed from the sample by the drying air. The amount of removed organic matter was calcu— lated, assuming that it represents acetic acid only. 158 159 McDonald and Dewar-(1960) used a similar approach with a regular oven at 1000 C. Hot, dry and COZ-free air was pumped through the sample. and through a Liebig condenser. The precipitate was collected in a salt and ice—cooled vessel. Before entering the atmosphere, the air was forced to pass through traps of silica gel to remove the water, then through soda-lime for the absorption of volatile acids and, finally, through a standard acid solution for removal of volatile bases. This assured a com-. plete recovery of organic matter. Ammonia, ethanol, acetic, propionic, butyric, and lactic acids were determined quantitatively in the condensate and added to the oven-dried dry matter. Fenner and Barnes (1965) reported that the use of organic solvents for the determinations of dry matter was first reported in 1904. Perkins (1943) reported that 95% of the acetic acid of the sample was found in the water, after having been removed from the dry matter by the toluene method. Fenner and Barnes (1965) concluded that, in general, with good corn silages, the toluene-extracted water required only the titration values of the steam distillate_for volatile bases and acids to make the dry matter correct. If this is not done, they concluded, the error could reach a 10% underestimation of dry matter. To verify the procedure used in this study, corn silage dry matter determinations were conducted in the following ways: 1. Toluene (AOAC method with 2—1/2 hours of distillation). 2. Oven dry matter — 1050 C. a. 24-hour drying. b. 48-hour drying. 160 3. Oven dry matter -.550 C a. 24—hour drying b. 48-hour drying The following results were obtained: Per Cent Dry Matter as Determined By: Sample 105° c Oven 50° c Oven °’° Of No. Error 1 Toluene % 24 hrs. 48 hrs. 24 hrs. 48 hrs. % % % % 1 25.5 25.0 25.0 24.1 23.7 7.1 2 24.0 24.3 23.8 23.1 23.1 7.6 3 25.5 22.5 21.6 21.7 21.7 14.9 4 51.1 42.0 41.2 40.0 40.0 21.7 5 50.3 40.8 39.6 40.9 40.9 18.7 6 55.8 43.3 43.1 42.3 42.3 24.2 I Mean error 15.7 L I 1 l % error is the error between the mean of the oven determination and the toluene determination. Shown in this table are the results comparing the toluene distil- lation with various oven drying methods. In this work the error comparing the two methods ranged from 7.1% to 24.2% with a mean error of 15.7%. The oven dry matter values certainly appeared more valid and it was concluded \ that not all of the water had been removed during the distillation with the toluene. To further verify methods and procedures, a second trial was conducted utilizing 161 1. Toluene (2—1/2 hour distillation) using a I ground sample. 2. Oven drying (550 C) using a ground sample. The following results were obtained: Dry Matter Determination as Sample Analyzed on a Ground Silage Sample by: % of No., 0 Error Toluene 105 C Oven % % l 33.73 30.25 10.37 2 33.34 33.13 0.63 3 33.60 34.31 2.13 Average error 4.37 These results compare the two methods after the silage samples were chopped in a Waring Blender for one minute. This reduced the partical size and the results compared much better than before with only 4.37% error. The pH values of the toluene distillate certainly provided evidence that the volatile acids were not remaining in the sample as dry matter. After reviewing the literature and the results of the above trials, it was concluded that oven drying at 550 C for 24 hoursIgave satisfactory results. It was further concluded that the accuracy would not be improved without going to extreme distillation and recovery methods for dry matter determinations, as described by Forbes (1943) and McDonald and Dewar (1960), which was not within the financial resources of this project. VH1 cHIGnN sTnTE UNIV. NH 619 LIIHLWHIHIIWJIES :Hl ll!