fiulleltn 790 fbecemfiea I954 is‘ Beef Cattle Perfornm nee at Bluebonnet Farm 1. Evaluation T est: for Gaining Ability A1 ,_,,..,-.;u- i E w < ,..-,L~;1.., Three Hereford bulls which were high gainers in the evaluation tests. Their gain ratios“ were 120, 127 and 140. respectively. They are now in breeding service at Bluebonnet Farm. .-..‘...V-', t in cooperation with the UNITED STATES DEPARTMENT OF AGRICULTURE TEXAS AGRICULTURAL EXPERIMENT STATION R. D. LEWIS. DIRECTOR, COLLEGE STATION, TEXAS DIGEST Efficient beef cattle must have the inherent ability to gain rapidly. Tests for this ability made at Bluebonnet Farm by self-feeding young beef cattle together at the same place with the c? feeds to determine differences among animals. Experiments were conducted to obtain more info A tion on the part played by heredity in causing individual differences in this character and to 1-‘ records on individual animals for use in the selection of breeding animals on Bluebonnet Farm for cattle belonging to private breeders. EPFPC NMFQ HQ Results of these tests indicate that the rate of gain is highly heritable and that it is possib, select for a large portion of the differences in the rate. Heritability was calculated from recor 853 animals, of 2 breeds, 1 cross and 3 sex classifications. These ranged from 21 to 57 percent. g heritability value considered most reliable is that obtained by the method of regression of pare - average offspring using gain ratio. Calculated from 73 parents and 291 offspring, it was 54 perch These high heritability figures add overwhelming support to the conclusion that selection for of gain will be effective when the individual’s own record is used. Progeny testing is useful, but limited in practical application. To make progress in gaining ability, high-gaining animals must r lected. ' mw-QI-I-mmr-imnol-rltnfi Records made by tested bulls and heifers are listed individually in a supplement to this b A available on request. These records are useful to identify high-gaining individuals. ' ACKNOWLEDGMENTS bi 4 I-Lbm l-d%n-O-lilfl This is a cooperative project with the U. S. Department of Agriculture, and is part of the ' Southern Regional Beef Cattle Breeding Project. R. T. Clark and E. J. Warwick, western and c». ern regional coordinators, have contributed to the progress of the project. A Appreciation also is expressed to the following for their assistance in these experiments: - Hill, superintendent, Bluebonnet Farm; R. E. Patterson, vice director, Texas Agricultural Expe 'i Station; J. C. Miller, J. K. Riggs and O. D. Butler, head, associate professor and associate prof *9 respectively, Department of Animal Husbandry, Texas A. & M. College System; and R. D. Turk, y Department of Veterinary Parasitology, Texas A. & M. College System. “l 4_Ln_an\4Aq1-l& CATTLE PRODUCTION is one of the most im- nt branches of agriculture in the Texas I areas, and is becoming increasingly im- n1: in many farming areas. Improved con- tifin practices result in more pasture, for- nd feed grain crops. These resources can be ‘- to better advantage through the produc- ‘of efficient meat-producing animals. Beef yerica’s preferred meat, and our economy de- ~ in large measure on more efficient conver- . of grass and feed crops into edible beef. _ growth usually is the most economical ‘ h. There is tremendous variability in the _ of different animals to grow on the same even among animals of the same breed, I d sex. Even a small increase in the aver- bility to gain rapidly, if increased and stab- 4 by breeding, would increase potential meat tion and the profit to commercial produc- E beef cattle. here have been only a few breeding experi- 4; directed toward this goal because of the tions of facilities at most research institu- f The acquisition of the former Bluebonnet ance Plant, McGregor, Texas, now named nnet Farm, made available to the Texas ultural Experiment Station enough land and ies for a full-scale breeding experiment beef cattle aimed at the improvement of i production characteristics. mall herds of registered animals were ac- J Herefords by purchase and Brahmans »-ns and gifts from interested breeders. Also, f: number of high grade Hereford cows purchased for use in test matings and in 's. Small groups of Santa Gertrudis and oll cattle were later added to the herd, but ‘mall amount of data obtained from them not included in the statistical analyses. his bulletin is the first of a series report- 5| results of investigations with beef cattle f ebonnet Farm. Others will include: selec- the Bluebonnet Farm herd based on per- nce; carcass evaluations of Hereford and , ross Hereford x Brahman steers; weight ctively, animal husbandman and geneticist, assist- imal husbandman and geneticist, Bluebonnet Farm, egor, Texas; and animal husbandman, U. S. Depart- _of Agriculture, formerly stationed at Bluebonnet , now superintendent, Chinsegut Hill Sanctuary, sville, Florida. Cattle Performance at Bluebonnet Fawn I 1. Evaluation Tart: for Gaining Ability BRUCE L. WARWICK, T. C. CARTWRIGHT and M. W. HAZEN * for age at weaning; regularity of breeding; use of heat tolerance tests; and hybrid vigor. Preliminary evaluation tests for gain were started in 1948-49 with a group of 42 unregis- tered Hereford calves. The first full-scale eval- uation test for gain was put into operation in 1949-50 with bulls and heifers owned by Blue- bonnet Farm. At that time the test also was made available to private breeders for testing bull calves. The 1950-51 test was opened to private breeders to test both heifer and bull calves. HERITABILITY In any group of animals of like age, breed and sex, maintained under similar conditions, some will grow or gain more rapidly than others. Part of this variation may be due to differences in heredity and part to differences in reaction to the environmental influences which cannot be identified easily. Through certain statistical pro- cedures these two portions of the differences may be separated and measured (statistically termed variance). The part due to selectable inherited differences has been labeled “Heritability.” This can be measured effectively only when the sires or dams of tested individuals are known, or if one or both parents and the individual have been tested. Heritability is expressed in percent and rep- resents that portion of the variation that is sub- i.e. below 15 percent, direct selection on the in- dividual’s own record is relatively ineffective in increasing the records of the offspring. Where ject to selection. When the heritability is low, CONTENTS Page Digest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 3 Heritability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 3 Heritability of Gaining Ability . . . . . . . . . . . . . .. 4 Heritability Studies at Bluebonnet Farm . . . . .. 5 The Gain Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 6 Records of Completed Gain Tests . . . . . . . . . . .. 7 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 7 Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . _ . . . .. 8 Literature Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 8 heritability is 30 percent or higher, selection bas- ed on the individual’s own record is highly effec- tive in increasing the average of the offspring. Animals should be selected whose records are above the average of the group. This is a theo- retically valid method for permanently raising the average production of the offspring over the average production of the group from which the parents are selected. However, the offspring of selected parents will not be as much above the average as their immediate parents were above the average of the group from which they were selected. This is where heritability figures ap- ply. As an example, if both selected parents have gain ratios of 110, each Would be 10 percent above the average of the group from which selected. If 5O percent heritability is assumed, the offspring would be expected to average 5 percent above the average of the group from which their parents were selected. Progeny from selected parents will be expected to vary about as widely as the group from which the parents were selected. Therefore, if enough progeny are produced, a few may exceed their parents in gaining ability. HERITABILITY OF GAINING ABILITY Until recently this problem received little at- tention but results now available are encourag- ing. Knapp and Nordskog (1946, 1946A) , published heritability figures for economic c acters in beef cattle. Based on 177 steers an bulls, heritability of rate of gain was foun be 75 percent by the paternal half-sib correla method. By the regression of offspring-on-pa method, heritability was 97 percent. Knapp Clark (1947), after gathering data on 422 st, and 52 bulls in the same experiment, compui heritability as 72 percent. Patterson, et al. (19 reported on the rate of gain of animals tested the Balmorhea substation of the Texas Agri tural Experiment Station for a period of 7 y” Six to 10 bull progeny per sire were tested, ing a total of 814 progeny. Using the half correlation method, they concluded that he bility was extremely high for this character, actual numerical value being close to 100 perc Knapp and Clark (1950) after adding more Q reported revised figures that ranged from 65 i; cent by the half-sib correlation method to 77 cent by the regression of offspring-on-pa method. Kincaid et al. (1952) computed he, bility of growth rate from gain data avail for both sires and progeny. These varied f“ 0 to 42 percent and averaged 22 percent for, bulls fed individually and 12 percent for 55 h‘ ers on pasture. Cartwright (1954) working ~i Figure 1. Bulls and heifers on gain test at Bluebonnet Farm. xyoung bulls and heifers in heat tolerance es, found heritability of 19 percent for rate ‘y in on shadeless pastures during the summer. ‘rson et al. (1954) revised the heritabilities 4 the Balmorhea work after collecting more and improving the technique of analysis. '- found an average value of about 53 per- l ABILITY srunnzs AT BLUEBONNET FARM fata collected from studies conducted for 4 ._,(1_949-53) at Bluebonnet Farm involved : ttle,‘587 of which were raised on the farm. ” from 853 of these cattle could be used for lation of heritability. These data give heri- _'ty figures that agree fairly well with prev- reports. Since bulls, heifers and steers 0f breeds 0r crosses were tested, some diffi- es in analysis were encountered because of mely unequal numbers of animals of each f‘ To overcome the inequality and dispro- '0nality of numbers in the various subdivis- i and to eliminate differences due to test year Lbreed, the data were analyzed on the basis of ain ratio. The gain ratio was computed for I animal by dividing its total gain on test by average total gain on test of animals of the year, breed, sex and ration group and mul- ‘ng by 100. For example, Hereford bull “BF ' gained 375 pounds in the 1952-53 test. The ge for Hereford bulls that year was 309 ds so the gain ratio for “BF 831” is: 75/309), or 121. in average animal would have a gain ratio A00. An animal above 100 would be above ge, below 100 would be below average. Using this gain ratio as the measure of gain- ability. heritability was computed by two ‘ods. The first was obtained by the method 4 correlation between half-sibs. Because of rences in pretest environment and in gene- ariation among cow herds of the various es on which the test animals were raised, ,b1'lity values were computed separately for _ ttle raised at Bluebonnet Farm, as well as ll‘ cattle tested at Bluebonnet. The Here- A Brahman and the F, (Hereford x Brah- groups also were treated separately. A few ny groups of Aberdeen-Angus. Santa Ger- is. Charbrav and Shorthorn cattle and some i ls with unknown sires were tested at Blue- et Farm but these data could not be used in ollowing heritability figures. The herita- figures obtained are: 1' included No. of animals Heritability’ , percent 'mals 853 38 ls raised at BF 587 34 refords 329 51 ords raised at BF 189 21 ahmans 124 46 ans raised at BF 33 46 ereford x Brahman) 359 33 lraised at BF) The second method was by the regression of offspring-on-parent. Calculations were made from tested parents, either male or female, and their offspring. In most cases only one parent of the offspring had been tested. All of these cattle were raised at Bluebonnet Farm. Sixty- seven parents and the average of their offspring were included in this calculation. Heritability was calculated to be 57 percent. Also available were five additional sires tested at Balmorhea. Offspring from two of these sires were raised and tested at Bluebonnet Farm. Those from the other three sires were raised at the Texas Range Station near Barnhart and tested at Bluebonnet. When this additional information is added the figure is changed to 54 percent. Calculation of heritability by the regression of offspring-on-par- ent is considered the best method (Lush, 1940) since it is a more direct measure of what a par- ent is transmitting to its offspring and is more reliable. These calculations, again, are based on the gain ratio computed for each animal. This may not be exactly the same as the absolute gain within breed, sex, year and feed level, but the use of the gain ratio has two advantages. One is that of overcoming difficulties in analysis and the other is that selection must be made on the basis of this gain ratio or some similar method of ranking the animals. Heritability calculated from this ratio gives a better estimate of prog- ress to be expected. These analyses of the Bluebonnet data are presented in more technical detail by Warwick and Cartwright (1954). These heritability val- ues, and the others reported in the literature, in- dicate that heritability of rate of gain in beef cattle is high. Even if the lower figures are near- er correct, marked progress can be expected by testing and selecting high-gaining bulls. Herita- bility values discussed here show considerable variability. They were computed from tests at four stations with different genetic material. Heritability applies directly only to the group from which data are collected. Also, some of the Balmorhea and Bluebonnet data included pro- geny groups from different ranches. Thus, pre- test ranch differences are reflected in the analy- sis as genetic differences in the sires. This re- duces the reliability of the calculated values. The Bluebonnet values, especially for the Herefords, are lower for the animals raised at Bluebonnet than for the entire group tested, which includes cooperator cattle from different ranches. This does not mean that heritability of rate of gain is lower for Bluebonnet cattle, but that their en- vironmental influences are more similar. These heritability values for gaining ability, gathered independently from various sources and- computed by different methods, indicate that con-t siderable progress can be made by a relatively: simple selection program. Also, the data are ex- tensive enough to assume that a fair sample has been tested and that almost any group of beef 5 cattle can be expected to fall Within the range of reported values for heritability of rate of gain. THE GAIN TEST The gain evaluation test has been standard- ized as far as possible. The feed lot test begins in November and continues for 140 days after an adjustment period of 2 weeks. The earlier tests were for 154 days. Calves must have been drop- ped between November 1 of the previous year and April 30 of the year the test is started. The self-fed ration for all breeding cattle consists of a mixture of cottonseed meal, ground sorghum grain, ground hegari fodder, ground J ohnsongrass hay, and either ground alfalfa leaf meal or vitamin A concentrate, or both. Con- centrates are increased to about 30 to 35 percent during the adjustment feeding period. Steers are fed in two groups: one fed the same ration as described for the bulls and heifers. the other fed a high concentrate ration of 66 to 68percent con- centrate. Other than some of the cattle raised at Bluebonnet and the Texas Range Station. all tested animals Qrn guhiepfl". tn rgqiatratinn; nun. 1.110‘ ‘Hon ‘Fihqt A vnarq (fifivnrod hv ‘l-lwiq qnqlvcliq_ flap animals were in qirn ‘nrnqenv o-rnu-nq nf 9 ny- mare The reflliirement of a. minimum, Dumber per mire group for breeding animals has been abandoned to allow more extensive testing of individuals However. steers of anv lrnown breeding vnav he entered in progeny groups of 3 to 6. it‘ otherwise qualified. These steers contribute to the evalua- tion of sires, breeds and crosses, and to further study of heritability. Cattle go on test at ages differing as much as 6 months. There also is a difference in weight because of age and finish, especially in cattle of cooperators. The correlation coefficient between initial weight (beginning of test) and gain ratio (a measure of subsequent gain within year, breed, sex and ration group) based on the 920 animals tested at Bluebonnet Farm was 0.11. Between initial age and gain ratio it was 0.00. These fig- Figure 2. Self-feeding of gain test cattle at Bluebonnet Farm. " e _ 6 Figure 3. Weighing gain test cattle at Bluebonnet F 0 ures indicate that there is little relationship .< tween those variables and that neither age ~ initial weight had any appreciable effect in. termining subsequent gain. Stated differe within the limits set here, age is of no valuf predicting gain on the test and only about 1“ cent of the variability in gain can be predii from a knowledge of initial weight. Most c tested thus far entered the test in medium fl with only a few approaching what might be sidered fat. Definite extremes probably w. affect the test, but within the range from med thin to slightly fat, the test appears valid. S' lar conclusions were reached by Patterson, e1 (1949) and Patterson, et al. (1954) on the at Balmorhea. i 1a.. H. g- a e-a a a. as Selection for high gaining ability, as m ured in these tests, can be effective in raising, average gaining ability of a herd. Calves rais Bluebonnet Farm were tested to study the t retical basis for inheritance of gain zle. heri =. ity, and to obtain the actual individual records‘. use in selection. The tests were opened to pri" breeders to supplement the studies of heritabi to provide comparison with calves raised on periment at Bluebonnet Farm and to make "a able to the breeders entering the calves and to sible purchasers an authentic record of each mal for their use for selection purposes. i mil-leg: =1 At Bluebonnet Farm, sires are being s__ ted as much as possible from the top 25ipe a in gain for the group tested the same year. addition to the sires raisedand tested at 3* bonnet Farm, two Hereford bulls were purc in 1949 with gain test records from the 19f, Balmorhea test. These were the two hig gaining bulls that year. Records of the |:_ of these two sires add direct support to the clusions reached based on the theoretical c0 erations. These results will be presented in second bulletin of this series. Each year all y ers belonging to Bluebonnet Farm which a '9 the low 25 percent of all heifers of the same b are sold for slaughter. It is too early to ate the results of this culling. ng__np_.nn_.nn_.an_l ii Ii ifin-Iuéuul H I erSa i Charolaise. daily gain 0f heifers. é 1 for illustration. RECORDS OF COMPLETED GAIN TESTS ilt isgimportant to make available in perman- orm the results of the tests completed to date luebonnet Farm for study and use by other A To this end we have listed individu- jtllbullsiypand heifers tested from 1949 through . ‘ Included are the following breeds and es: Hereford, Brahman, Aberdeen-Angus, ,1 ‘Gertrudis, Charbray, Shorthorn, Red Poll The only unregistered animals "imals not approved for registry included inregistered Hereford heifers raised at Blue- let Farm and the Texas Range Station, and “Brahman heifer the first year. been listed within breed and sex groups by of the gain ratio, i.e., 100 (total individual /avérage gain for group). These tables show h animals had superior gaining ability. Sires ~: progeny were tested are listed separately phabetical order by owner. Each sire is num- p» and this number is the sire code appear- the individual-record tables. ce, a sire’s rank, owner’s complete name and ess, and other information can be determin- fFOI‘ each breed and cross, information is pre- s» in the following order: individual records i lls, averages by year for daily gain of bulls, iidual records of heifers and averages by year The sire list follows f dividual records of all breeds and gives the name and address of the owner. The records for the five tests from 1949-50 through 1953- The records By cross ref- These lists have been prepared as a supple- F to this bulletin and are available upon re- A partial list of one breed is given in This is arranged like upplement listing, the animals with the high- ain ratios first. Only the highest seven, one average and the two lowest of this breed and sex are included. One of the bulls with a daily gain of 2.3 pounds was the highest gaining Hereford in the year tested and had a gain ratio of 125. Another bull tested a different year gained 2.4 pounds per day but the gain ratio was only 100. This shows that the actual pounds of gain must be considered in relation to the performance of the other animals of the same group tested the same year, i.e., the gain ration. It is recommend- ed that the gain ratio be used directly for selec- tion, using only sires with gain ratios above 100, preferably those with the highest ratios. The wide range of variation between animals of the same breed, sex and year group emphasizes the possibility of greatly increasing average growth or daily gain by selection. DISCUSSION It has been established that gaining ability, probably the most important character in beef cattle, can be improved by selecting high gaining individuals as herd sires and cow replacements. Records such as those given in the supplemental table are necessary to establish which are super- ior individuals. Since a lot of weight is given to the records in selection, some careful study of the records is required. Gains are different from year to year and among sexes and breeds. These should be taken into consideration, especially if animals are tested under different conditions. It is necessary to evaluate an animal relative to its sex group which was tested under the same conditions. An animal’s daily gain in pounds is interesting and has some value but of more im- portance is the relative value of the record of the animal. It has been found very useful at Blue- bonnet Farm to use the gain ratio for this pur- TABLE 1. EXAMPLES OF PERFORMANCE OF HEREFORD BULLS TESTED, 1949-50 — 1953-541 I . Registry . Daily Gain Rank in Year tested c‘ d animal number sue codez owners gain. lbs. ratio Year Sire Individ. -3# F. 909 . . . . . . . . . . 7303034 6 Bluebonnet 3.1 140 53 1 1 B. F. 828 . . . . . . . . . 7303031 7 Bluebonnet 2.9 132 53 2 2 emere 27th . . . . . . 6155170 26 Storts 3.0 130 51 1 1 #1 F. 3129 . . . . . . . . 7877819 89 Bluebonnet 3.1 129 54 4 1 Domino Chief 23d 6529152 23 Pazdral 2.9 128 52 2 1 nnet Farm 685 . . 6792515 6 Bluebonnet 2.8 127 52 1 2 - . . . . . . . . . . . . . . . . 5756748 3 Bluebonnet 2.3 125 50 1 1 F. 1047 . . . . . . . . . 7877811 89 Bluebonnet 2.4 100 54 4 8 . . . . . . . . . . . . . . . . 6346452 6 Bluebonnet 1.2 51 51 8 21 F 3274 . . . . . . .. 7877822 89 Bluebonnet 1.0 43 54 4 12 Daily gain. lbs. Gain ratio Number tested Average Lowest Highest Lowest Highest 11 1.8 1.2 2.3 64 125 31 2.3 1.2 3.0 51 130 38 2.2 1.7 2.9 76 128 32 2.2 1.5 3.1 69 140 12 2.4 1.0 3.1 43 129 lcomplete list of bulls and heifers tested is available as a supplement to this bulletin. “es. registry numbers and owner included in the supplement. r's names and addresses included in the supplement. pose. The gain ratio simply denotes the animal's standing in relation to the average of its group. To the breeder who is interested in improv- ing beef cattle for economic purposes that are useful to the commercial producer, it is recom- mended that he give rate of gain first consider- ation, being careful in obtaining and evaluating records. GLOSSARY Correlation: A measure of the closeness of as- sociation of two things. Perfect correlation is represented by 1.0, and no correlation by 0. F1: First generation produced by crossing two breeds. Gain Ratio: As used in this bulletin, it is 100 times the total gain on test of an individual divided by the average total gain on test of all animals of the same year, breed, sex and ration group. See text for an example. Heritability: Statistically, the fraction of the variation between animals due to selectable hereditary differences. See text for a more complete discussion. Regression: Average change in one of two char- acters per unit change in the other. For ex- ample, the regression of weight on age meas- ures the average number of pounds (or other measure of weight) of weight increase for an increase in age of one unit (days, months, etc.). Sib: A brother or sister. brothers or half-sisters. Half-sibs are half- LITERATURE CITED Cartwright, T. C. 1954. Responses of beef cattle to ambient temperatures. Journal of Animal Scienc press). q Kincaid, C. M., R. C. Carter and J. S. Copenhaver. ~ Heritability of rate of gain from progeny tests beef cattle. Journal of Animal Science 112741 stract). ' Knapp, Bradford Jr. and A. W. Nordskog. 1946. tability of growth and efficiency in beef cattle. J“ nal of Animal Science 5:62. “ Knapp, Bradford Jr. and A. W. Nordskog. 1946a. l tability of live animal scores, grades and certain. cass characteristics in beef cattle. Journal of ' mal Science 5:194. ' Knapp, Bradford Jr. and R. T. Clark. 1947. Genetice environmental correlations between growth ra -' beef cattle at different ages. Journal of Animal‘ ence 6:174. _- Knapp, Bradford Jr. and R. T. Clark. 1950. Revised .9 mates of heritability of economic characteristi beef cattle. Journal of Animal Science 9:582. ’ Lush, J. L. 1940. Intra-sire correlations or regre of offspring on dam as a method of estimating‘ p tability of characteristics. Proc. American S of Animal Production pp. 293. Patterson, R. E., J. H. Jones, J. J. Bayles and R. V. bough. 1949. Performance-testing and pr p testing of beef breeding stock as an aid to sele Journal of Animal Science 8:608. , 1' I Patterson, R. E., T. C. Cartwright, J. H. Jones an _ Bayles. 1954. Performance testing in the sel of beef breeding stock. Journal of AnimalSc (in press). P Warwick, Bruce L. and T. C. Cartwright. 1954. » tability of rate of gain in young growing beef ~ Journal of Animal Science (in press). The individual records of all bulls and heifers tested at Bluebonnet Farm 1949 through 1954 are available as supplementary tables to this Bulletin 790. Write to Agricultural Information Office Texas A. & M. College System College Station, Texas