— — ______ — . _. W— ,7 _. _ 7—. — — ‘1 7 7 A , — — "—— ,. __ — — ___—__ — — 129 792‘ .THS VARIATION IN FEED EFFICIENCY NND CARCASS CHARACTERISTICS OF ENDIVIDUMLY FED SWINE 'Z‘HESIS FOR THE. DEGREE OF M 5‘ MICHIGAN STATE SOLLEGE MIKE V'ORK‘APICH 1955 THESJS This is to certify that the thesis entitled Variation in Feed Efficiency and Carcass Characteristics of Individually Fed Swine presented by Michael Vorkapich has been accepted towards fulfillment of the requirements for 1-1.8. degree in Animal Husbandry a/QW/ Major prfisor Date March 15, 1955 0—169 VARIATION IN FEED EFFICIENCY AND CARCASS CHARACTERISFICS OF INDIVIDUALLY FED SWINE By Mike Vorkapich A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Animal Husbandry 1955 THFQH‘ ACKNOWLEDGEMENTS The author wishes to express his sincere appreciation to Professor L.J. Bratzler, Department of Animal dusbandry, for his interest, contin- uous guidance, and constructive criticisms, in making this research pro- ject and manuscript a success. The writer is indebted to Dr. William b. Baten, Research Professor of Statistics, for his kindly help with the statistical analysis. For their co-operative willingness in helping with the slaughter and cutting of the experimental hogs, the author wishes to express his sin- cere thanks to co-workers, R.J. Deans and H.E. Townsend. To his mother and father, the author is very grateful and indebted for their assistance in making this manuscript a reality. .35i5N53553 VARIATION IN FEED EFFICIENCY AND CARCASS CHARACTEHISTICS 0F INDIVIDUALLY FED SWINE By Mike Vorkapich AN ABSTRACT Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements *for the degree of MASTER OF SCIENCE Department of Animal Husbandry 1955 K9. sagas Mike Vorkapich In the history of hog production in America, changes in type have been numerous and sometimes extreme. The present type change is dif- ferent from the situation which prevailed in other periods in that a ”meat- type" hog is in the process of being developed because of consumer pre- ference for leaner cuts. Swine carcass evaluation has received considerable attention in re- cent years due to the lower price of lard compared with the price of lean cuts of pork and the prices paid for live hogs. The most recent method for evaluating the carcass value of live hogs has been the probing technique. The feSvdrCn program herein reported deals with the feed effi- ciency of littermatos and evaluation of the carcass from live hog probes. Nine crossbred wigs, six of which were littermates of a Uuroc 9 Berkshire cross and three Yorkshire - Chester White (littermatcs), five barrows and four gilts, were divided into nine lots and placed on self- feeders containing a 16 percent protein ration. Throughout the experie ment, the pigs and the amount of feed consumed were weighed weekly. At 125 pounds the protein was reduced to 12 percent. The pigs were probed at three points; posterior to the shoulder blade, central lumbar region, and side of ham, at the live weights of 100, 150, and 215 pounds which was the slaughter weight. Carcass data included primal cut weights, body and leg length, backfat, and eye muscle measurements. The primal cut out ranged from 46.7 to 53.0 percent. Average backfat measurements ranged from 1.44 to 2.11 inches. Consumption of feed per 100 pounds of gain ranged from 337 to 422 pounds. Mike Vorkapich The second trial consisted of nine purebred Chester r"hite pigs (littermates), three barrows, and six gilts. The feeding, weighing and prohing locations were comparable to the first trial. Similar carcass data were obtained as in Trial 1. The primal cut out ranged from 44.2 to 48.9 percent. Average backfat measurements ranged from 1.70 to 2.22 inches. Consumption of feed per 100 pounds of gain ranged from 369 to 450 pounds. Analysis of the data disclosed the following results: 1. Feed efficiency among littermates varied and when analyzed it was found to be significant at the 5 percent level between breeds. 2. Variation in daily gain betweenlitternates ranged from .03 to .51 pounds between the purebred Chester Whites of Trial 2; from 0 to .04 pounds between the Yorkshire - Chester hhite cross; and from 0 to .22 pounds between the huroc - Uerkshire cross of Trial 1. Statistically, no significant difference was found when average daily gains were analyzed. 3. No significant differences were found when the data were analyzed for age at slaughter, carcass length, and dressing percent. 4. Significant differences were found between average backfat thick- ness between breeds. Highly significant differences were found between breeds of the percent of live and carcass primal cut out, and of the percent of lean cuts on live and carcass weight basis. 5. As the age at slaughter increases, the percent of lean cuts on live weight basis does not necessarily increase. 6. Highly significant correlations between percent of lean area in u...“— I Til-qr Mike Vorkapich cross section of rough loin and percent of live and carcass primal cut out were found. A highly significant relationship was found to exist between the eye muscle of the third rib and the eye muscle of the last rib. No correlation was found between the average live probe, shoulder probe, and last lumbar probe with the average carcass backfat thickness. II III IV VI VII VIII TABLE OF CONTENTS IN'I‘AVIJ‘DUCTIA‘NQ O O O O O O O OBJECT OF THE STUDY . . . . REVIEW OF LITERATURE. . . . EXPERIMENTAL PRUCEDURE. . . .A. B. Feeding Period. . . . Cutting and Slaughter RESULTS AND DISCUSSION. . . E. Feed Consumption. . . Daily Gain. 0 o o o o Carcass Measurements. Slaughter and Cutting Live Probe Measurements CUNCLUSIONS o o o o o o o 0 APPENDIX. ... . . . . . . . Procedure Data. BIBLIOGRAPHY... . . . . . . . Page 12 12 13 l7 17 18 20 22 31 33 35 44 ll. 12. 13. FORMULAE ANALYSIS ANALYSIS ANALYSIS ANALYSIS ANALYSIS ANALYSIS ANALYSIS ANALYSIS ANALYSIS ANALYSIS USED IN STATISTICAL ANALYSIS. . . . . . . 0F OF OF DE OF 0F OF OF GE GE VARIANCE VARIANCE VARIANCE VARIANCE VARIANCE VARIANCE VARIANCE VARIANCE VARIANCE VARIANCE PERCENT OF LEAN CUTS AT SLAUGHTER. o o o O O TABLES LISTED 0F OF DE OF OF OF DE OF OF OF 0N. Page FEED CONSUMED PER 100 POUNDS 0F GAIN AVERAGE DAILY GAIN. . . . . . . . . . AGE AT SLAUGHTER. . . . . . . . . . . CARCASS LENGTH. . . . . . . . . . . . AVERAGE BACKFAT THICKNESS . . . . . . DRESSING PERCENT. . . . . . . . . . . PERCENT OF LIVE WEIGHT PRIMAL CUT PERCENT OF CARCASS PRIMAL CUT OUT . . PERCENT OF LEAN CUTS 0N LIVE BASIS. . PERCENT OF LEAN CUTS UN CARCASS BASIS LIVE BASIS RANKED ACCORDING TO AGE CORRELATION AND REGRESSION COEFFICIENTS OF VARIOUS CARCASS MFASIJWEAN'PS. O O I O O O O O O O O O O O O O O O C O O O O O 16 17 19 20 21 22 23 25 26 29 30 APPENDIX LISTED Page TABLE I FEED DATA. . . . . . . . . . . . . . . . . . . . . . . 35 TABLE II FEED DATA. . . . . . . . . . . . . . . . . . . . . . . 36 TABLE III CARCASS MEASUREMENT DATA . . . . . . . . . . . . . . . 37 TABLE IV CARCASS MEASUREMENT DATA . . . . . . . . . . . . . . . 38 TABLE V SLAUGHTER DATA . . . . . . . . . . . . . . . . . . . . 39 TABLE VI SLAUGHTER DATA . . . . . . . . . . . . . . . . . . . . 40 TABLE VII CUTTING DATA . . . . . . . . . . . . . . . . . . . . . 41 TABLE VIII CUTTING DATA . . . . . . . . . . . . . . . . . . . . . 42 TABLE IX PERCENT LEAN AND FAT AREA OF EYE MUSCLE OF LAST RIB CIiOP O O O O O O O C C O O O O O O O O O O C O O O O O 43 INTRODUCTION In the history of hog production in America, changes in type have been numerous and sometimes extreme. The changes from a large, late ma- turing hog to a small, short-bodied animal and back again to a larger type, has not been the result of studied efforts to comply with market demands. Only during the last century have swine been prominent in the meat supply of the nation and the type has been raised to be adapted to the needs of the pork trade. The changes in type which have been made were the results of fads and fancies to a larger extent than through the preferences of the consumers. Greater refinement and quality were emphasized as producers became interested in fixing definite types for the breeds resulting in some de- crease in the size of the hogs, but, a radical change did not occur un- til the.time of the "hot-blood" boom in the Poland China breed. The preferred type was a very short, chuffy, lardy, early maturing pig which was one of the least profitable types of swine ever produced in America. other breeds weressomewhat affected by the trend toward small size, but the greatest extreme was reached by the Poland China breed. The require- ments of the porR trade had no part in influencing the change to the small type nor later to the big type. The present type change is different from the situation which pre— vailed in other periods in that a "meat-type" hog is in the process of being developed. There have been various concepts as to what character- istics make up a meat-type hog. Birmingham, Brady, Hunter, Grady, and Kiehl (1954) defined”a meat-type hog as a heavily muscled hog, free from excess fat, well balanced and uniform in length and depth of body, and produces a larger proportion of lean to fat, which will yield approxi- mately 50 percent primal cuts, be 29 to 30 inches long, have a backfat thickness of 1.3 to 1.6 inches and have a carcass weight of approximately 150 pounds." Zobrisky, Lasley, Brady, and Weaver (1954) defined a meat- type hog as well balanced, heavily muscled, well developed in the ham and loin, firm in flesh, trim of underline and jowl, and carries enough finish to produce a firm, high quality, high yielding carcass that has a desirable carcass length of 29 to 31 inches for a 200 to 230 pound hog. The solution to the production of a meat-type hog does not lie sole- ly in a specific plan of management or in any given breed. One method is selection for desirable carcass qualities. A rotational cross-breed- ing program would be another method for developing the meat-type hog. A third possibility would be limited feeding (Crampton, Ashton, and Lloyd, 1954) which would reduce the yield of fat and in turn increase the yield of lean meat with acceptable quality. Swine carcass evaluation has received considerable attention in re- cent years due to the lower price of lard as compared.with the price of lean cuts of pork and the prices paid for live hogs. According to Peter- son and Baird (1952) it is more difficult to evaluate the live hog than the carcass because the live hog is one step further removed from the wholesale cuts, which in the final analysis determine the value of the animal. The most recent metho( for evaluating the carcass value of live h0g5 has been the probing technique. The research PrOgram herein reported deals with the feed efficiency of littermates and evaluation of the carcass from live hog probes. OBJECT OF THE STUDY The lack of consumer demand for lard and its resultant low price has increased the price of the leaner and more desirable pork cuts. Both the producer and consumer would benefit materially from hogs that yield a higher percentage of the lean or muscled cuts. The swine pop- ulation must be altered to meet the market demand for a "meat-type" hog. The research problem herein reported was designed to study the feed effi- ciency of littermates individually fed ad libitum, carcass characteris- tics, and mecuanical measurement of backfat of live hogs as compared with carcass backfat measurements. REVIEW OF LITERATURE Very little investigation of variation between littermates concern- ing feed efficiency and carcass characteristics has been done. Blunn and Baker (1947) studied the data from 416 hog carcasses, 58 of which were from littermate pigs that received the same ration but were fed out in larger groups. They divided the feeding period roughly into two portions so that the gain made during the growing period (56 to 112 days) could be separated from the gain made during the fattening period (112 days to slaughter). According to McMeekan, (1940) most of the pigs' skeletal and muscular growth is made during the first 116 to 120 days and after that most of the increase in weight is due to fat deposition. Blunn and Baker found a significant difference of +.023 pounds between litters within sires for average daily gain from weaning (56 days) to time of slaughter and a highly significant difference of +.090 inches between litters within sires for depth of backfat. How- ever, where they determined the analysis of variance for differences between littermate hogs (58), they found no significance in the depth of backfat (.044 inches) and average daily gain from weaning to time of slaughter (.017 pounds). A positive correlation of .292 was found be- tween average daily gain from.weaning to time of slaughter and depth of backfat. Blunn and Baker concluded that there may be a breed difference in the relation between fatness and rate of gain. Tucker, Dickerson, and Lasley (1952) studied the effects of full feeding from weaning to market weight on the crosses of Landrace with Duroc, Poland China and Hampshire, and the four purebred stock of each breed. They found that under full feeding, the crossbred hogs gained 7 percent faster and reached final weight 10 days earlier, but consumed 6 percent more feed daily and were no more efficient than the parental pure- breds. Dickerson and Crimes (1947) selected for high and low individual feed requirements per pound of gain in two strains of Duroc swine. All the pigs from each litter were full fed a mixed ration in individual pens and selection was based largely on differences between littermates. They found the differences to be 24 percent heritable. According to Headley (1947) variations in gain are due in part to differences in the quality and quantity of protein, composition of feeds, inherited characteristics of the individual animal and breed. Crampton, Ashton, and Lloyd (1954) studied the effect of restricting feed intake during the finishing period. They reported results obtained from 120 purebred Yorkshires which were fed over a period of two farrow- ing seasons (winter and summer), 60 pigs in each season. The hogs were full fed up to 110 pounds and then the feed was restricted until slaugh- ter weight. They found that restricted feeding increased the size of the loin muscle over the full fed hogs and also the lean area in the ba- con rasher. They also reported that the lean-fat proportions of the hog carcass were significantly correlated with size of the loin muscle and also the lean area of the bacon rasher. Merkel, Bray, Grummer, Phillips, and Bohstedt (1953) reported that by limited feeding, the length of the feeding period was increased and the average daily gain was depressed. In addition, they found that the dressing percentage and backfat thickness were decreased and signifi- cantly leaner and firmer carcasses were produced. The body length was also significantly increased. Winters, Sierk, and Cummings (1949) studied four different levels of feed intake and resultant carcass quality in swine. The first lot was self-fed the entire feeding period; the second lot was self-fed until the pigs weighed 125 pounds, then the feed was restricted to 3 percent of the body weight; the third lot was restricted feed to 3 percent of the body weight until the pigs weighed 125 pounds, and then self-fed to the slaughter weight of 215 pounds; the fourth lot was restricted to 3 percent of the body weight for the entire feeding period. They found that the range of feed efficiency of the four lots was from 365 to 391 pounds of feed per hundred pounds of gain with the restricted group be- ing the more efficient and the restricted, full-fed (third lot) the least efficient. They found that the third lot required a longer period (266 days) to reach the slaughter weight as compared to 206 days for the full—fed group. There was little difference between the remaining groups and the full-fed lot. They reported also that the group fed the restricted ration throughout, yielded the highest percent of primal cuts (72.11) whereas the full-fed group yielded the least.(69.47). The other two groups yielded approximately identical percent cut outs.. Crampton (1937) reported data from 7 Yorkshire litters, totalling .40 pigs which were fed individually, 20 hand-fed and 20 self-fed. He found that the self-fed hogs gained less than the hand-fed hogs. The self-fed hogs averaged 1.37 pounds of daily gain as compared to 1.50 pounds for the hand-fed hogs and,therefore,required slightly longer (7 days) to reach market weight. The feed required per 100 pounds of gain was 409 and 383 pounds respectively; the hand-fed hogs required about 50 pounds less feed to reach 200 pounds market weight. The carcasses of self-fed hogs were on the average shorter by about 7/8 inch than those hand-fed as reported by Crampton. Brampton (1940) set up an experiment feeding pigs individually which were full-fed 3 times daily to 100 pounds and then 2 times daily thereafter. He found no significant relationship between rate of gain and leanness or length of carcass. Hammond and Murray (1937) showed that the weight of the live hog and the weight of the carcass is directly related to the weight of the bacon sides. They found that the actual live weight of the hog affects the percentage of bacon it will yield much more than does breed or type. The actual rate of increase of the thickness of the backfat slows down as the sides increase in weight. With the increase in weight of the side, the changes in thickness of fat over the loin, shoulder, and flank are closely related. According to these workers, fat is deposited in the following order, shoulder, rump, and lastly in the region of the 10in. , . They also found that as the pig grows his length in- creases with age and live weight. In this study, they found that the backfat thickness varies in the different breeds for any given length of side. They found that crossbred pigs when compared with pure parent breeds at similar body weights are longer in body length but in backfat and belly thickness they are practically intermediate. Hetzer, Hankins, King, and Zeller (1948) reported from data of 141 crossbred pigs, that the live body measurement of length showed practi- cally no relationship with yield of trimmed and untrimmed cuts. Aunan and Winters (1952) useda coring device for carcass sampling to estimate the fat and lean content of the swine carcass. Locations at which the 70 carcasses of ll different breeds and crosses were probed with the coring device were: between the fifth and sixth ribs of the belly, between the eighth and ninth ribs of the loin, between the eleventh and twelfth ribs of the belly at the subcostal arches, the last rib of the loin, and the pocket of the belly. They reported that the fifth and sixth rib sample of the belly had the highest degree of association with with lean content of the carcass when percentage figures were used and gave the correlation coefficient of +.79 which was the highest of the 5 locations. A correlation coefficient of +.70 was found for the relation- ship between average backfat thickness and fat content of the carcass. Hazel and Kline (1952) reported backfat measurements from 96 live hogs. The sites chosen for probing were l'fi inches off the midline of the body above the longissimus dorsi; namely, behind the shoulder, mid- dle of the back, middle of the loin and middle of the loin over the ex- act midline of the body. They reported the following correlations be- tween the average backfat thickness on the carcass and the individual live—hog measurements: at the middle of the loin over the longissimus dorsi, +.67; and average of the four live-hog measurements, *.81. Hazel and Kline concluded that the location behind the shoulder was the most accurate single measure of carcass fatness and that the middle of the back was the poorest location for measuring backfat on live hogs. -10- Hazel and Kline (1953) studied backfat measurements of live hogs weighing approximately 210 pounds, which were probed prior to slaughter. Eight sites were probed. They made correlations between the percent lean cuts, percent fat cuts and the depth of fat measured by probing. The following results were obtained: behind the shoulder, over the longissi- mus dorsi -.69, +.76; middle of back over the longissimus dorsi -.55, +.54; middle of loin over the longissimus dorsi -.70, +.76; middle of loin over the lumbar vertebra -.48, +.53; top of the ham -.65, +.66; tailhead -57, +.43; side of shoulder, -.47, +.54; and side of ham -.29, +.40. Correlations between four backfat measurements taken on carcasses and the percentages of lean cuts and fat cuts were -.75 and t.79 re- spectively. From the figures obtained, they concluded that measurements at some of the sites reflect fatness and leanness as accurately as backfat measurements on the carcasses. They concluded from their data that the sites behind the shoulder, over the loin, and top of the ham have greatest accuracy in predicting leanness and carcass value. De Pape and Hhately (1954) studied probing measurements behind the shoulder, over the loin, and over the back, i inch from the mid-line, taken at monthly intervals from weaning to market weight. They found that the probes taken at ages younger than 140 days had very low pre- dictive values and that measurable differences occurred between breeds in backfat deposition at 140 days and older. They stated that the probes behind the shoulder and over the loin on both sides are good indicators of leanness and carcass value. These results seem to coincide with those found by Hazel and Kline.(1953). -11. Cole, Riley, Smith and Hobbs (1953) studied carcass data from 250 head of hogs that had been divided into six weight groups from 155 to 305 pounds at 30 pound intervals and classified as to types... chuffy (small), intermediate, and rangy (large)... which were handled uniform- ly as to feed, fill, and shrink. They reported the following results: 1. A highly significant correlation existed between type and dressing percentage. They found that the longest hogs with deepest chest measure- ments had the lowest dressing percent. 2. The rangy hogs had a higher percentage of their live weight in primal cuts, but this advantage dimin- ished as the live hog weight approached 300 pounds. 3. The rangy car- casses had 23.4 percent more spareribs of a leaner kind than the chuffy type. 4. They stated that a standard shrink should not be used in car- cass evaluation because of the significant differences in percentage of viscera between different types of hogs at any given weight. 5. They found that kill weight is probably a more constant basis for determining cutting yields as the total percentage of primal cuts based on packer or shipper style carcasses decreases as the live hog weight increases. Zobrisky, Brady, and Lasley (1953) studied carcass and live hog data from 207 hogs and reported significant negative correlations between the four lean cuts and the live hog backfat probes. They also reported sim- ilar results, but somewhat lower, between the same measurements and the five primal cuts. A significant positive correlation between the probes and the total fat was also reported. EXPERIMENTAL PROCEDURE A. Feeding Period Trial 1. Nine October crossbred pigs were started on experiment Janu- ary 5, 1954. Six were Duroc-Berkshire pigs at an average weight of 45 pounds (littermates) and three were Ybrkshire-Chester White pigs at an average weight of 76 pounds (littermates). At that time, the pigs, five barrows and four gilts, were weighed and divided into nine lots and placed on self-feeders containing a l6 percent protein ration consisting of: 652 lbs. Corn 100 lbs. Oats 130 lbs. Soy Bean Oil Meal 50 lbs. Tankage 10 lbs. Fish Solubles 25 lbs. Dehydrated Alfalfa Meal 14 lbs. Limestone 10 lbs. Dicalcium Phosphate 5 lbs. Iodized Salt 1 lb. Trace Minerals 0.5 lbs. Vitamins A and D 1 1b. Fortafeed 2-49-90 1.5 lbs. TM-S (3 #/ton) The pigs were continued on this feeding program until a weight of approximately 125 pounds was reached. Throughout the experiment, the pigs and the amount of feed consumed, were weighed weekly. Water was provided ad libitum. At 125 pounds, the ration was adjusted to give a 12 percent protein ration by increasing the corn to 752 pounds and reducing the soybean oil meal to 60 pounds, and tankage to 20 pounds per 1,000 pounds of mixed feed. The amounts of all the other ingredients remained the same. Re- cords were kept of the feed consumed by each lot from which feed efficienp cy was calculated. The pigs were probed at three points; posterior to the shoulder blade, central lumbar region, and side of hmm,at the live weights of 100, 150, and 215 pounds which was the slaughter weight. An incision, é-inch in width, transversely to the longissimus dorsi, was made at the mentioned points by a lancet. A metal ruler was inserted into the fatty tissue until it reached the muscle tissue and a reading was taken which included the skin. Trial 2. Nine March purebred Chester White littermates, three barrows and six gilts, averaging 48 pounds were divided into nine lots and start- ed on experiment May 13, 1954. Although one barrow was excluded from feeding and carcass data due to illness, only the longissimus dorsi area data of this pig was used. The feeding and weighing procedure and probing positions were identical with Trial 1. B. Cutting and Slaughter Procedure Trials 1 and 2. The animals were taken off feed at approximately 215 pounds and given access to fresh water fer a period of 24 hours prior to -14- slaughter. At the time of slaughter, a live weight was recorded which was used as a basis for calculating live weight cut out, live weight percent lean cuts, dressing percent, and percent shrink. The hogs were slaughtered packer style and chilled for 24 hours at a temperature of 32° - 36° F., at which time a chilled carcass weight was recorded, ex- cluding the leaf fat and kidney. All carcass measurements were made and recorded in inches. The length of the body was measured from the anterior edge of the first rib near the first thoracic vertebra to the anterior edge of the aitch bone. The leg length was measured from the anterior edge of the aitch bone to the coronary band. The number of ribs of each carcass was recorded. Backfat measurements were made over the first rib opposite the first thoracic vertebra; over the seventh thoracic vertebra; over the last rib; and over the midpoint of the last lumbar vertebra. The backfat thickness for each carcass was calculated by averaging these measurements. The carcasses were cut into primal cuts and the weights of each recorded. The neck bones, jowl, clear plate and forefoot cut across the knee joint, were removed from the 2§ rib shoulder. The remaining cut, the New York style shoulder, was weighed as the first primal cut. The ham was removed between the second and third sacral vertebrae on a line perpendicular to the hind leg. The tail, flank, surplus fat, and foot (at the hock joint) were removed. A skinned ham was made, leaving about 3/8 inch of fat on the ham. This cut was weighed as the second primal cut. The rough loin and belly were separated along a line beginning one inch below the tenderloin muscle at the posterior end to about one inch - 15 - from the end of the backbone of the blade end. At this time, tracings were made of the cross-sectional area of the right rough loin at the blade end and also between the last two ribs. A planimeter was used to determine the area of lean and fat from these tracings and the percent of lean was calculated. A ch0p containing the next to the last rib was removed from the rough loin and saved for photographic records. The rough loin was weighed in order to determine the loin index by compari- son with the weight of the trimmed loin. The back fat was removed from the loin leaving about a fi-inch of fat on the loin. The trimmed loin was weighed as the third primal cut. The spare ribs were removed from the belly which was trimmed and weighed as the fourth and last primal cut. 0f the four primal cuts, the skinned ham, New York style shoulder and trimmed loin, were considered as the lean cuts in calculating the percent of lean cuts. Analysis of variance and t-tests were calculated for feed consumed per hundred pounds of gain, average daily gain, age at slaughter, dress- ing percent, average backfat, carcass length, percent of live weight cut out, percent of carcass cut out, percent of lean cuts on live basis, and percent of lean cuts on carcass basis, according to the methods of C.H. Goulden (1952). Correlation coefficients (Snedecor, 1946) were deter- mined between percent lean area in cross section of rough loin and both the percent carcass cut out and percent live weight cut out; between eye muscle area of the third rib and last rib; between average probe, last lumbar probe, and shoulder probe with the average backfat measurements. The statistical formulae used are shown in Table 1. - 16 - TABLE 1 FORMULAE USED IN STATISTICAL ANALYSIS Analysis of variance: (Goulden, 1952) . 2 SK2 - Séfil- - Total sum of squares. (5x1)2 + (51(2)2 + (sxn)?! ‘ C.T. ' Between sum of N1 N2 Nn squares. t-test 21-22 gggl- 22)'-\/VZ—12+ UZZSE'L t a . ill - ‘Zé) V N q2&_ . 72 ) (table for t) ' Significant level between means. Correlation analysis: (Snedecor, 1946) (5X)(SY) sxy-_T_ ,. . xy 2 (5)02 (an2 \/(SX " -fi--)(SY2‘- N ) sxr _ (stzgsrg b I 5x2 - (sxf‘ RESULTS AND DISCUSSION A. Feed Consumption All the animals in Trials 1 and 2 were individually fed. The range in feed efficiency of the littermates for Trial 1 was 337 to 422 pounds of feed per 100 pounds of gain. For the littermates of Trial 2, the range was 369 to 450 pounds of feed per 100 pounds of gain. 0n the av- erage, Trial 1 consumed 381.6 pounds of feed as compared to 404.1 pounds for Trial 2 which showed that there was a variation in feed efficiency between Trials. See Appendix Tables I and II. From these data it appears that the crossbred pigs (Duroc - Berkshire and Yorkshire — Chester White) of Trial 1 were more efficient than the purebred Chester White pigs of Trial 2. The difference in feed efficien- cy may have been due to the climatic conditions at the time they were be- ing fed. Trial 1 was made during the winter season whereas Trial 2 was conducted in the summer. A statistical analysis was made of the feed consumed per 100 pounds of gain. The analysis of variance given in Table 2 follows the method given by Goulden (1952). TABLE 2 ANALYSIS OF VARIANCE 0F FEED CONSUMED PER 100 POUNDS 0F GAIN Source Degrees Sum Mean of of of Square F-Value variation Freedom Squares ___ Total 16 16303 Breed 2 6938 3469 5.19% Error 14 9365 669 The analysis of variance of feed consumed per 100 pounds of gain was made as shown in Table 2. Here the F-test shows a significant dif- ference at the 5 percent level in the feed consumed per 100 pounds of gain between breeds. It appears that heredity factors played an impor- tant role in causing the differences in feed consumption for 100 pounds of gain. A t-test (Table l) was made to determine where the significance existed. The difference between the average consumption of the Yorkshire - Chester White and Duroc - Berkshire gave a t value of 2.68 which was significant at the 2 percent level. Similarly, the difference between the average consumption of the Chester White breed and Duroc - Berkshire gave a t value of 2.62 at the 2 percent level. There was no significant difference between the Yorkshire - Chester White and Chester Whites since the average consumption difference was less than the standard deviation of 18.3. 8. Daily Gain The variation in average daily gain between littermates of Trial 1 was 1.58 to 1.80 pounds and the range for Trial 2 was 1.40 to 1.91 pounds. The differences in daily gain between littermates ranged from .03 to .51 pounds for the purebred Chester White hogs of Trial 2; from 0 to .04 pounds within the Yorkshire - Chester White cross; and from 0 to .22 pounds within the Duroc - Berkshire cross of Trial 1. The littermates of the purebred Chester Whites showed the most variation. There was little difference in average daily gain for the three groups. The average daily gain of the Duroc - Berkshire cross was 1.67 pounds, of the Yorkshire - Chester White cross, 1.68 pounds, and for the Chester White, 1.62 pounds. King, Gobble, and Henning (1952) found that crossbred hogs made greater average daily gains of 1.26 pounds per day or .09 pounds more per day than the pedigreed hogs. TABLE 3 ANALYSIS OF VARIANCE OF AVERAGE DIALY GAIN Source Degrees Sum _Hean of of of Square F-Value Variation Freedom Squares Total 16 44.92 Breed 2 .01 .005 .002 Error 14 44.91 3.21 The analysis of variance of average daily gain was made as shown in Table 3. The F-test did not show any significant difference in the aver- age daily gain between breeds. The age at slaughter within littermates varied in Trial 1 from 166 to 187 days and in Trial 2 from 157 to 193 days. However, between breeds the averages of the age at slaughter show very little difference. For the Duroc - Berkshire cross, the average slaughter age was 175 days, for the Yorkshire - Chester White, 179 days and for the purebred Chester Whites, 176 days. King, Gobble, and Henning (1952) found that the cross bred pigs showed a definite economic advantage in the number of days required to - 20 - reach market weight. The crossbreds averaged 158 days to reach approxi- mately 225 pounds as compared to 176 days for the pedigreed hogs. TABLE 4 ANALYSIS OF VARIANCE OF AGE AT SLAUGHTER Source Degrees Sum Mean of of of Square F-Value Variation Freedom Squares Total 16 1850 Breed 2 36 18 0.139 Error 14 1814 129.6 The analysis of variance of the age at slaughter was made as shown in Table 4. The F-test did not show any significant difference in the average age at slaughter between the breeds. AC. Carcass Measurements Variation of carcass length between littermates of Trial 1 ranged from 29.25 to 31.1 inches as compared to 27.7 to 30.7 inches of the pure- bred Chester White hogs of Trial 2. The average difference of carcass length between breeds was very small. The average carcass length of the Duroc - Berkshire was 29.91 inches, Yorkshire - Chester White, 30.01 inches and for the purebred Chester White, 29.85 inches. See Appendix Tables I and II. - 21 - TABLE 5 ANALYSIS OF VARIANCE OF CARCASS LENGTH source Degrees *Sum ‘Hean of of of Square F-Value Variation Freedom ngares Total 16 5.93 Breed 2 .07 .035 .084 Error 14 5.86 .419 The analysis of variance of carcass length was made as shown in Tab- le 5. The F-test did not show any significant difference in the carcass length between breeds. The average backfat measurements for Trial 1 were fairly low as compared with those of Trial 2. In Trial 1, the range of the average backfat measurements was from 1.44 to 2.11 inches as compared to 1.70 to 2.22 inches for the Chester White hogs of Trial 2. There was a variation in the backfat thickness between littermates as shown in Appendix Tables III and IV. However, between breeds, the average backfat differences were large. The depth of fat was greater at the carcass locations than at the live hog locations most nearly corresponding to them. The average of the' four carcass measurements of both Trials was 1.84 inches while the aver- age of the two live hog locations was 1.56 inches as measured by probing. See Appendix Table IV. - 32 - TABLE 6 ANALYSIS OF VARIANCE OF AVERAGE BACKFAT THICKNESS Source Degrees "Sum ‘Hean of of of Square F-Value Variation Freedom Squares Total 16 .69 Breed 2 .32 .16 6.15% Error 14 .37 .026 The analysis of variance of the average backfat thickness was made as shown in Table 6. Here, the F-test shows a significant difference at the 5 percent level between the average backfat thicknesses of the breeds. A t-test (Table 1) was made to determine where the significance ex- isted. A difference between the average backfat thickness of the York- shire - Chester White hogs and Duroc - Berkshire produced a t-value of 3.25 which was highly significant at the 1 percent level. The t-value between the Chester White hogs and Duroc - Berkshire cross gave a signi- ficant difference at the 5 percent level. There was not any significant difference between the purebred Chester White hogs and Yorkshire - Chester White cross since the difference was less than 0.11. Rust (1953) found no significant differences of average backfat thickness among related purebred Chester White hogs. D. Slaughter and Cutting Data. From the data, Appendix Tables V and VI, there was very little variation between littermates in dressing percent. The range in dressing percent for Trial 1 was from 72.3 to 75.9 percent with a mean average of 74.8 percent. As for Trial 2, the range in dressing percent was from 72.8 to 76.8 percent with a mean average of 75.1 percent. The two mean percentages compare favorably with the average of the two Trials which was 74.9 percent. The mean difference between the two Trials was 0.3 percent which is too slight to show any significance. TABLE 7 ANALYSIS OF VARIANCE OF DRESSING PERCENT Source Degrees Sum Mean of of of Squares F-Value Variation Freedom Squares Total 16 25.71 Breed 2 0.5 ' .25 .139 Error 14 25.21 1.80 The analysis of variance of the dressing percent of the breeds was made as shown in Table 7. Here, the F-test did not show any significant difference in the average dressing percent between the Duroc - Berkshire cross, Yorkshire — Chester White cross and Chester White hogs. Rust (1953) found no significant difference statistically in dress- ing percent of related purebred Chester White hogs. Between littermates of Trial 1, the variation in percent primal cut out varied from 46.7 to 53.0 percent while the percent primal cut out of Trial 2 ranged from 44.2 to 48.9 percent. A difference between the means of the percentage live weight cut out shows that the crossbred hogs of Trial 1 cut out higher in primal cuts, (51.5 percent) than the Chester White hogs (47.2 percent). - 24 - TABLE 8 ANALYSIS OF VARIANCE OF PERCENT OF LIVE WEIGHT PRIMAL CUT OUT Source Degrees Sum Mean of of of Square F-Value Variation Freedom Squares Total 16 100.05 Breed 2 70.58 35.29 16.73** Error 14 29.47 2.11 The analysis of variance of the percent of live weight primal cut out was made as shown in Table 8. Here, the F-test shows a highly signi- ficant difference in the percent of live weight primal cut out between breeds at the 1 percent level. A t-test was used to determine where the highly significant differ- ence in percent of live weight primal cut out existed. A highly signi- ficant difference in favor of the Duroc - Berkshire over the Yorkshire - Chester White showed up when the t value was found to be 3.98 at the 1 percent level. A t value of 5.48 was fOund to be highly significant at the 1 percent level between the Duroc - Berkshire and Chester Whites of Trial 2. No significant difference was found between the means of the Chester Whites of Trial 2 and Yorkshire - Chester White. Rust (1953) found no significant difference in percent of live weight primal cut out among related purebred Chester White hogs. The variation of percent of carcass cut out for Trial 1 ranged from 61.7 to 71.0 percent as compared to 60.7 to 65.2 percent in Trial 2. There was variation in percent of carcass primal cut out between individ- ual littermates as can be seen from the data compiled in Appendix Tables VII and VIII. From the averages of percent of carcass primal cut out, the hogs in Trial 1 were superior to those in Trial 2. Trial 1 had a mean average of 67.1 percent as compared to 62.9 percent for Trial 2. TABLE 9 ANALYSIS OF VARIANCE 0F PERCENT OF CARCASS PRIMAL CUT OUT Source Degrees Sum Mean of of of Square F-Value Variation Freedom Squares Total 16 185.16 Breed 2 140.16 70.8 22.06** Error 14 45.00 3.21 The analysis of variance of the percent of carcass primal cut out was made as shown in Table 9. The F-test shows a highly significant dif- ference between breeds in carcass primal cut out at the 1 percent level. A t-test was made to determine where the highly significant differ- ence existed. A t value of 4.49 between the means of the carcass primal cut out of the Duroc - Berkshire and Yorkshire - Chester White proved to be highly significant at the 1 percent level. Similarly, a t value of 6.30 between the means of the same carcass characteristic of the Duroc - Berkshire and purebred Chester Whites showed a highly significant differ- ence at the 1 percent level. The variance of the means between the Yorkshire - Chester White hogs and Chester White was less than 1.27 and therefore no significant difference was shown. -26.. Rust (1953) found no significant difference in the percent of carcass ‘primal cut out when treated statistically among related purebred Chester White hogs. Between littermates there was a difference in the percentage of lean cuts based on live weight. Between the Duroc - Berkshire littermates, the difference ranged from 0 to 3.8 percent, between the littermates of the Yorkshire - Chester White from .1 to 1.0 percent and from 0 to 5.4 percent between littermates of the Chester Whites. The mean averages of the percent of lean cuts on live basis for the Duroc - Berkshire was 39.3 percent, for the Yorkshire - Chester White 36.3 percent, and 35.8 percent for the purebred Chester White. It appears that there is a dif- ference in the percent of lean cuts on live basis among the breeds. TABLE 10 ANALYSIS OF VARIANCE OF PERCENT OF LEAN CUTS 0N LIVE BASIS Source Degrees Sum Mean of of of Square F-Value Variation Freedom Squares Total 16 79.46 Breed 2 44.12 22.06 8.75** Error 14 35.34 2.52 The analysis of variance of percent of lean<2uts on live weight basis between breeds was made as shown in Table 10. Here, the F-test shows a highly significant difference between breeds in the percent of lean cuts on live weight basis at the 1 percent level. A t-test was made to find out where the highly significant differ- ence existed between the breeds. The mean difference between the Duroc - Berkshire cross and Yorkshire - Chester White cross gave a t value of 2.68 which was significant at the 5 percent level. However, the mean difference between the Duroc - Berkshire and purebred Chester Whites was highly significant at the 1 percent level which gave a t value of 12.91. No significant difference existed between the means of the Yorkshire v Chester White and Chester White hogs. Rust (1953) found no statistical significance in percent of lean cuts on live weight basis among related purebred Chester White pigs. Variation in the percent of lean cuts on carcass basis between lit- termates of the Duroc - Berkshire ranged from 48.6 to 54.4 percent with a mean average of 52.6 percent. .For the Yorkshire - Chester White hogs, the range varied from 47.6 to 49.5 percent with a mean average of 48.4 percent. The purebred Chester White hogs ranged from 44.6 to 50.4 per- cent with a mean average of 47.7 percent. The average percent of the lean cuts on carcass basis of the Duroc - Berkshire, which was 52.6 per- cent, was above the total average of both trials which was 49.6 percent. TABLE 11 ANALYSIS OF VARIANCE OF PERCENT OF LEAN CUTS 0N CARCASS BASIS ___A L Source Degrees Sum Mean of of of Square F-Value Variation Freedom Squares Total 16 136.38 Breed 2 87.63 43.81 12.59** Error 14 48.75 3.48 -28- The analysis of variance of percent of lean cuts on carcass basis between breeds was made as shown in Table 11. The F-test shows a high- 1y significant difference at the 1 percent level between the breeds in percent of lean cuts on carcass basis. A t-test was made to determine the exact significant difference be- tween breeds. The mean difference of percent of lean cuts on carcass basis between the Duroc - Berkshire cross and Yorkshire - Chester White cross produced a t value of 3.18 which was highly significant at the 1 percent level. Similarly, when the mean difference between the Duroc - Berkshire cross and purebred Chester White hogs was taken, a t value of 4.85 was found and it proved to be highly significant at the 1 percent level. However, there was no significant difference between the York- shire - Chester White and purebred Chester White. Rust (1953) found no statistical significance in percent of lean cuts on carcass weight basis among related purebred Chester White hogs. ~‘ "'1 -7 773mm: TABLE 12 PERCENT OF LEAN CUTS ON LIVE BASIS RANKED ACCORDING TO AGE AT SLAUGHTER Trial 1 Age % Trial 2 Age % FEE—N3: (Days) Lean Cuts FIE—N3: (Days) Lean Cuts 12-5 166 40.4 1-4 157 34.4 12-3 166 40.3 1-1 157 32.4 12-2 173 36.6 1-6 171 36.5 9-4 179 36.9 1-10 171 35.5 9-3 179 36.0 1-8 178 37.2 9-5 179 35.9 1-12 186 36.4 12-4 180 40.3 1-9 193 37.8 12-9 180 37.8 1-7 193 36.4 12-8 187 40.4 1-5 193 34.6 As the age of slaughter increases, the percent of lean cuts on live basis does not necessarily increase. ,In the Duroc - Berkshire cross of Trial 1 (litter number 12), their ages at slaughter varied from 166 to 187 days and the percent of lean cuts on live basis from 36.6 to 40.4 percent. Among the Yorkshire - Chester White littermates of Trial 1 (litter number 9) the age at slaughter was identical but the percent of lean cuts on live weight basis varied from 35.9 to 36.9 percent. In Trial 2, there was variation between the littermates as their ages varied from 157 to 193 days and the percent of lean cuts on live basis from 32.4 to 37.8 percent. Ammfia mo maoamv pcofiofimywoo commemumoe Hp .Ho>eH wcooaea H vs vasofimesmfim zanmfim as mmomxowcu vmmxomn mmdohdo mmeo. Nmmo.+ sm.ae se.me ma.vm we omeeo>e m> shone assess «wee mmocxowcv «mwxomn ammoamo oooo. 3303.+ hm.Hm mo.mm mH.om ma oumuo>m m> enema aerwwoxn mmedxoflsv «tuxedo mmeommo omeo. ammo.+ sm.Ho e~.mm sm.ee we ammum>e m> enema msea maeao>e nae whoa mo maemse ohm mmom. msmamo.+ mm.moe mm.mee ma.mm we as nae scene we assess can «So «so Haefima mmmoamo «doomed m> mace m was ac Hmmm. ##hvmb.+ mm.ammm> oo.>mmam No.wm>mm ma compose mmoao me some need «awoken .wso wso Hmsfiaa anwwoa o>fiH pdooeoa m> deed :wsoa mo mawm. ##wmam.+ mm.mvmmv vo.vaemm No.>m>mm ma dowuoem mmoao me some mama vdoouom new 53 em Hp a nauseoemrpcm we as m 9 55m “Mn: madoaeasmmo: unseemo mszmzmxbm m3 mazmHoHLmuoo zaHnndxoux 22¢ 22He< «04 w.vvm w.omv m.omh >.mow m.Hmv m.mvm m.oam N.mv~ v.mmm o.am~ Hugo; you m.mm «.mv m.vs m.mm v.sv- v.5m o.HN H.8m m.mm 4.e~ mwmca>< w.moa H.mvH m.mmm m.mma m.NvH «.mmm H.mc «.ms p.mm m.mm H3909 mMMMIE' mrmw:1;.a v.ms unmwl >.mv m.mm H.Hm, v.vm v.am m.m~ mum m #04 m.mm o.mv v.v> v.mm v.8N «.mm N.om N.>m m.mm o.rm mum m «o; m.mm m.mv m.m> b.vm N.mv m.ooa m.am m.mm m.mm v.wm vnm b #04 m.mm w.mm m.mm o.mm m.am m.moa m.vm m.wm m.mm w.mm owmuo>< w.mmm >.mam m.cmv m.mHv «.mom m.amm m.>¢a o.H>H .>.wma h.wha Haves v.0v v.mm m.vm H.mm m.mm m.moa m.vm m.mm m.vm v.Hm muma m “Q” m.hm m.mm m.om m.mm H.0m n.ooa H.wm m.hm N.mm m.mN mama a pod m.ov o.vm m.mm H.o> m.mm m.HHH m.mm m.mm >.mm v.am vuma v #04 v.0v v.vm v.mw c.H> w.mm m.ooH v.mm o.mm m.mm m.mm mama m you m.cv o.mm N.ww m.mo o.mm m.mHH ¢.vm m.Hm m.»m m.om mINH N you w.mm m.mv w.mb m.vm b.mv w.OOH x.mm b.mm o.mm m.bm muNH H uQ~ .«2 .9: mmao.mnq maso “so 930 «so «so muse mo afioq hovasosm mHHon as: o>wq «0 & undo mo R smog Hasfium R Hmaflum R.93 anwmm cosimue .>.z voaafiua coqnfixm .oz menu some musu Gama adwos mmduumo .pk o>«a Ndwpqv my:u mo wmwfioz mo: H Hufiua mqm< ~.Hmm m.mvm ¢.mHmH e.coHH ~.mmm m.HmeH m.msm m.em4 ~.mH¢ «.mwv Haves m.om~ e.me H.mmm o.mom p.8em m.em> m.moH. H.mom w.mnH ¢.v- Hdaoe a HaHuH :;n-m.vvm m.ocv m.sme ~.moo m.Hm4 m.m4m m.aHm‘ m.av~ «.mmm o.Hom Hauoa H Hche m was H «HmHua ho «nan maHsaso m.mm p.84 m.ve m.mo m.»v m.mm ».om o.om e.m~ H.m~ mm.c.>< «on 0.0mm >.Hmm H.mmm o.mom b.8em m.eme m.mmH H.mo~ m.amH v.v- Hosea was «.0» m.~4 o.ms >.~o o.mv m.OOH m.H~ ~.mm m.vm o.o~ NH-H m «o3 v.4m H.mv H.mm m.mm m.ae v.mm m.eH «.mm m.m~ H.8N qu s “on v.8m p.54 m.me N.Nm m.ev m.mm «.mm m.mm H.mm H.e~ euH o «as m.mm m.mv m.me m.vm m.ev o.HoH N.NN 8.5m H.mm p.wm m-H m Hos N.>m «.mv o.as H.mo H.84 m.mm o.mm v.mm m.HN m.mm m-H 4 «on v.mm 0.44 H.mo e.om m.sv H.4m o.mH ~.mm o.mm m.mm H-H m was m.>m v.0m H.om ~.mm m.mv e.moH m.H~ a.©N w.m~ v.Hm m-H a god m.mm «.84 m.me o.mo m.©v H.mm ~.mH H.mm o.- ~.w~ OH-H H 504 .53 .93 undo .mpq .530 «so “so «so «no .550 Mo aHos amanozm zHHom an: «>Hq mo a -cdo no a new; HaeHcm a HmeHsa a .53 Hapoa cmscha .».z uoasmcmimmaame .32 .558 saws menu ado; Hayes «assume .53 m>Ha A..nqv mass mo psmHox mo: N HAHua <9 mqm