U.S. Department of Agriculture, Clay Center, NE K. E. Gregory3, L. V. Curie and R. M. Koch4

Transcription

1 BREED EFFECTS AND HETEROSIS IN ADVANCED GENERATIONS OF COMPOSITE POPULATIONS FOR BIRTH WEIGHT, BIRTH DATE, DYSTOCIA, AND SURVIVAL AS TRAITS OF DAM IN BEEF CATTLE,* K. E. Gregory3, L. V. Curie and R. M. Koch4 U.S. Department of Agriculture, Clay Center, NE ABSTRACT Heterosis effects were evaluated as traits of the dam in F2 progeny of F1 dams and F3 and 4 progeny of F2 3 dams in three composite populations of beef cattle. Traits included birth weight, birth date, calving difficulty percentage, and survival percentage at birth, 72 h, and weaning for calves with dams of different age classes. Breed effects were evaluated for the nine parental breeds (Red Poll m], Hereford N, Angus [A], Limousin &I, Braunvieh PI, Pinzgauer [PI, Gelbvieh [GI, Simmental [SI, and Charolais [C]) that contributed to the three composite populations (MARC I = 1/4 C, 1/4 B, 1/4 L, 1/8 H, 1/8 A; MARC II = 1/4 G, 1/4 S, 1/4 H, 1/4 A; and MARC EI = 1/4 R, 1/4 P, 1/4 H, 1/4 A). Among calves with 2-yr-old dams, breed effects were significant for birth weight, birth date, calving difficulty percentage, and survival percentage at birth but not at 72 h and weaning. Calf survival at weaning was lowest for smallest (< p ) and largest (> p ) birth weight classes and did not differ among intermediate birth weight classes. Calves with difficult births with 2-yr-old dams were significantly heavier at birth (39.6 vs 35.4 kg) and had significantly lower survival at 72 h (87.1 vs 92.2%) and at weaning (77.4 vs 85.1%) than calves with 2-yr-old dams that did not experience difficult births. Among calves with dams 2 3 yr old and from dams of all ages, breed group effects generally were significant for the traits analyzed. Important breed group effects on dystocia and survival traits were observed independent of breed group effects on birth weight. Effects of heterosis were significant for birth weight for each generation of each composite population and for the mean of the three composite populations. Generally, heterosis effects for calving difficulty percentage were not significant. Effects of heterosis generally were significant for date of birth (earlier) for each composite population and for the mean of the three composite populations. Heterosis effects on survival to weaning percentage generally were positive but generally were not significant. Heterosis retained for birth weight, birth date, and survival percentage in combined F3 and 4 generation progeny of combined F2 and 3 generation dams did not differ (P >.OS) from expectation based on retained heterozygosity. These results support the hypothesis that heterosis in cattle for these traits is the result of dominance effects of genes. Key Words: Cattle, Heterosis, Breed Differences, Birth Weight, Dystocia, Calf Survival T. Anim. Sci : Published as Paper No. 9410, Journal Ser., Nebraska secretarial support. Agric. Res. Div., Univ. of Nebras)ca, Lincoln 3Roman L. Hruska U.S. Meat Anim. Res. Center, ARS, USDA, Clay Center, NE Appreciation is expressed to Gordon Hays, Wade 4Dept. of Anim. Sci., Univ. of Nebraska, Lincoln Smith, Steve Kappes, and Robert Bennett and their staff for the Operations support provided to this project, to Received October 29, Darrell Light for data analysis, and to Debbie Brown for Accepted April 5,

2 BREED AND HETEROSIS EFFECTS IN BEEF C A m 3575 lntroductlon Heterosis achieved through continuous crossbreeding can increase weight of calf weaned per cow exposed to breeding by 20% (Gregory and Cundiff, 1980). Breed characterization studies have revealed large differences among breeds for most bioeconomic traits (Gregory et al., 1982; Cundiff et al., 1986). Approximately 55% of the U.S. beef breeding population involving 93% of the farmers and ranchers who produce beef cattle are in production units of < 100 cows (USDA, 1987). Optimum crossbreeding systems are diffkult to adapt in herds that use fewer than four bulls (Gregory and Cundiff, 1980). Further, fluctuation in breed composition between generations in rotational crossbreeding systems can result in considerable variation among both cows and calves in level of performance for major bioeconomic traits unless breeds used in the rotation are similar in performance characteristics. Use of breeds with similar performance characteristics restricts opportunity to use breed differences in average genetic merit to meet requirements for specific production and marketing situations (Gregory and Cundiff, 1980). The potential of composite breeds as an alternative to crossbreeding for using heterosis and for using genetic differences among breeds to achieve and maintain a more optimum additive genetic (breed) composition was suggested fist by Dickerson (1969, 1973). Retention of initial (PI) heterozygosity after crossing and subsequent random (inter se) mating within crosses is proportional to (n - l)/n when n breeds contribute equally to the foundation (Wright, 1922; Dickerson, 1969, 1973). When breeds used in the foundation of a composite breed do not contribute equally, the percentage of mean F1 heterozygosity retained is proportional to 1 - e, where Pi is the fraction of each of n contributing breeds to the foundation of a composite breed (Dickerson, 1973). This loss of heterozygosity occurs between the F1 and F2 generations, and if inbreeding is avoided, further loss of heterozygosity in inter se mated populations does not occur (Wright, 1922; Dickerson, 1969, 1973). The primary objectives of this study were to 1) evaluate differences among parental breeds and 2) estimate the combined individual and maternal heterosis for birth weight, birth date, n i calving difficulty percentage, and calf survival percentage as traits of the dam in inrer se mated composite populations of cattle. Materials and Methods Mating Procedure. Matings were made to establish three composite populations (Table 1). In this experiment the F1 is defined as the first generation that reflects the final breed composition of a composite population. The origin of the sample of each parental breed and the three composite populations and more detail on mating procedure and management protocols of this experiment were reported by Gregory et al. (1991a). Composite populations were established by using the same sires and dams used in the parental breeds. All yeiuling females were exposed to yearling males for a natural-service mating season of 42 d. Since 1987 in Limousin and 1988 in Herefords, males 2 yr old or older have been mated with yearling females because of late puberty in both sexes of these breeds. Females 2 yr old and older were mated by AI for 28 d followed by natural service exposure for 28 d. for a mating season of 56 d. More than 80% of sires have been used in 2 2 yr. From 1978 until 1984, the mating season for yearling heifers was from mid-may until late June and for females 2 2 yr old was from the fist of June until late July. Since 1985, the mating season for yearling females has been from late May until near mid-july and for females 2 2 yr old has been from mid-june until near mid-august. This adjustment of about 2 wk in mating and calving season was made to allow greater synchrony of breeding and calving with nutritive and climatic environment. Mating season for yearling females has ended immediately before the start of the natural service part of the mating season for females 2 2 yr old in order to use the same singlesire mating pastures for both age classes. Females that were not pregnant were retained in all breed groups, unless they were nonpregnant in two successive years, until Since 1985, all nonpregnant females have been removed each year from all breed groups. Nonperformance criteria, such as age, color, and extremes in skeletal size, have been used to remove excess females to control population size for each breed group. The same basic criteria were applied to all breed groups. Where possible, an attempt has been

3 3576 GREGORY ET AL. made to maintain a similar age distribution of females in each breed group. Males and females from the F4 generation of each composite popdation were removed from the experiment at an age of 1 yr because no further loss of heterosis is expected in F3 generation (Table 1). Thus, the experiment was completed with production and growing the F4 generation progeny to 1 yr of age. Females in all breed groups were assigned to sires on a stratified random basis within ages in all populations. Half-sib or closer matings were avoided. Except as noted, the same basic criteria have been used to identify males for breeding use in all populations. In aii populations the intent has been to avoid extremes in regard to weight and condition and muscular and skeletal anatomy. Dystocia has been given some consideration in identifying breeding males for use in all breed groups. Larger scrotal circumference also has been favored, particularly in breeds that are late to reach puberty (Le., Hereford and Limousin). Polledness and color patterns of red or red with white markings have been preferred for males used in all generations of each composite population. An effort was made to maintain a broad pedigree base in all breed groups. The occurrence of genetic defects in some breed groups (i.e., double muscling in Gelbvieh, MARC I and MARC II; parrot mouth in Gelbvieh and Braunvieh, malocclusion in Hereford, Angus, and Simmental; hydmcephalus in Red Poll and MARC m; and ataxia in Simmental) resulted in some compromise of pedigree breadth by avoidance of carriers or close relatives of carriers. Management of Females. Female populations have been fed and managed consistently with their requirements to maintain all breed groups in similar condition. The general plan was to group females in three fully integrated management units under the day-today supervision of an operations coordinator who had operational responsibility for this project. When a composite population and its contributing parental breeds had similar feed and management requirements, they were managed together (i.e., all generations of MARC I, Braunvieh, Charolais, and Limousin have been TABLE 1. MATINGS TO ESTABLISH COMPOSITES, RETENTION OF HETEROZYGOSITY, AND EXPKTED RETENTION OF HETEROSIS Composite popalatiom Item MARC I MARC II MARCHI MCan Parents of F1 generation. (CxLH)x(BxLA) (GH)x(SA) (PA)xO - or or or (C X W XLH) (GA) X (SH) (PA)x WV Reciprocals Reciprocals Breed composition of PI and.25b,.25c,.25l 25G,.25S.ZP, 25R - subsequent generations.125h,.125a 25H,.25A.ZH, S A PI Heterozygosityb.wd F2 Heterozygosity F3 Heterozygosity Dam Progeny Heterosis F1 F2.78 $+.WP.75 $ 1 Hm.75 Ht+ 1 Hm.76 H!.98 Hm Heterosis F-2 F3.78 H! H!.75 Hm.75 H!.75 Hm.76 H:.76 Hm Heterosis 9 F4.78 H +.78 P.75 I H +.75 Hm.76 H +.76.Composite popdatiom were established from the same animats used in the purebred io-on. C = chamlais, L = Limousin, H = Hereford, B = Braunvieh, A = Angus, G = Gelbvieh, S = Shental, P = Pinzgauer, and R = Red Poll. bretention of initial (PI) heterozygosity following Crossing and subsequent random mating within the crosses (inter n se) is proportional to 1 - e, where Pi is the fraction of each of n brceds conhiiuijng to the foundation of a composite i population. Loss of heterozygosity occurs between the F1 and F2 generations. If inbreeding is avoided, further loss of heterozygosity does not occur. % denotes individual heterosis expressed by progeny of a given generation and P denotes mated heterosis expressed by their dams assumiog that retention of heterosis is proportional to retention of heterozygosity. Data were analyzed 8s traits of dam. d.94 instead of 1 because both sir- and dams of ~1 generation were onefoarth ~imouin.

4 BREED AND HETEROSIS EFFECTS IN BEEP CATTLE 3577 managed together [Management Group 11; all generations of MARC II, Simmental, Gelb vieh, and Pinzgauer have been managed together Wagement Group 21; and all generations of MARC III, Hereford, Angus, and Red Poll have been managed together Management Group 31). The only deviation from this practice was during the 28-d natural service mating season when all females were in single-sire mating pastures and the 42-d mating season for yearling females. The Pinzgauer females were managed with composite MARC II for two reasons (i.e., to balance numbers among three integrated management units and because the feed and management requirements of Pinzgauer females are similar to those of Simmental and Gelbvieh). Even though the populations were grouped in the three management units, every effort was made to apply uniform management protocols among the three units. All groups received the same feed but the amounts were varied to be consistent with requirements. Improved pastures (cool- or warm-season grasses), winter feeding programs, and all basic management practices were the same. The sites were contiguous and were without boundaries (i.e., different management groups used the same pastures at different times). Two-year-old females were fed a mixture of corn silage and alfalfa haylage along with alfalfa and grass hay, beginning 2 to 3 mo before calving and continuing until pastures were adequate to meet their requirements (usually mid- to late April). All older females were fed mixtures of alfalfa and grass hay to meet nutritive requirements, usually from November until mid- to late April. After 1986, economic considerations favored feeding these females limited quantities of corn silage and alfalfa haylage during the period of winter feeding. Feeding Young Males and Females. Calves were weaned at an average age of about 180 d. Mean birth date was April 7 and calves were weaned the 1st wk of October in most years. Following an adjustment feeding period (28 d), females were fed diets composed of corn silage, alfalfa haylage, and protein-mineralvitamin supplement in varying proportions and lengths of time, depending on weather cmditions and weight gains of heifers: 1) Period 1, 2.34 Mcal of W g of DM, 11.62% CP, 2) Period Mcal of MJ3tg of DM, 12.34% CP, and 3) Period 3, 2.18 Mcal of MJ3tg of DM, 11.70% CP until they were placed on improved cool-season grass pasture from midto late April, &pending on adequacy to meet nutritive requirements. The three time periods were similar in length. Following an adjustment period after weaning (28 d), intact males were fed a diet composed of corn silage, rolled corn, and protein-mineral-vitamin supplement (2.69 Mcal of M4kg of DM, 12.88% CP) for 140 d. Data CoNection. Calves were weighed at birth, mid-mating season (end of AI mating period), at weaning and at 28,84,140, and 168 d postweaning. Yearling heifers were weighed at the start and end of the mating season and when palpated for pregnancy about 60 d after the end of the mating season. Thereafter, females were weighed, measured for height, and scored for condition three times each year (i.e., 13 before calving, 21 at the start of mating season, and 31 when palpated for pregnancy). Observations of estrus were ma& in yearling heifers starting approximately March 1 and continuing until the start of the mating season by visual observation using infertile males. Calving difficulty was subjectively evaluated using descriptive 1 = no difficulty, 2 = little difficulty by hand, 3 = little difliculty with calf jack, 4 = slight difficulty with a calf jack, 5 = moderate difficulty with calf jack, 6 = major difficulty with calf jack, 7 = Caesarean birth, and 8 = abnormal presentation). Percentage of calving difficulty was analyzed (scores 1 and 2 = 0; scores, 3, 4, 5, 6, 7, and 8 = 1). Analysis of Data. The data were analyzed by least squares mixed-model procedures (Harvey, 1985). The analytical model for birth and survival traits of all calves with 2-yr-old dams, calves with 2-yr-old dams with difficult births, and calves with 2-yr-old dams without difficult births included breed group, sire of cow within breed group (random), sex of calf, and year of birth of calf (1980 through 1989) (Tables 2 and 3). No main effect interactions were significant. For calves with dams 2 3 yr old and for calves from dams of all ages, the model included breed group, sire of cow within breed group (random), dam within sire of cow within breed group (random), sex of calf, year of birth of calf (1980 through 1989), age (3,4, 2 5 for 2 3 yr old dams and 2, 3, 4, 2 5 for all ages), and the interactions of breed group x age of cow and sex of calf x age of cow. Other main effect interactions were not

5 3578 GREGORY ET AL. significant (Table 4). Models used in specific analyses to clarify birth weight, gestation length, dystocia, and calf survival relationships are described when results from these analyses are presented. Breed groups included the nine parental breeds and the F2 generation progeny of F1 generation dams and combined F3 and F4 generation progeny of combined F2 and F3 generation (F2 and 3) dams. The F3 and F4 generation (F3 and 4) progeny of F2 and F3 generation (F2 and 3) dams were combined because a preliminary analysis showed that they did not differ (P =r.05), which is consistent with genetic expectation for retained heterozygosity (Table 1). Studentized range (D = Q.05 Em). as described by Snedecor and Cochran (1980), was computed to obtain approximations of differences required for significance among breed group means. Linear functions of means of parental breeds and of F2 progeny of F1 dams and of combined F3 and 4 progeny of combined F2 3 dams of each composite population were computed to estimate heterosis (Le., combined F3 and F4 generation progeny = F3 and 4 progeny and combined F2 and F3 generation dams = F2 and 3 dams). The F4 generation was removed from the experiment at 368 d (see Table 1 for genetic expectation for individual m] and maternal p] heterosis in each generation). TABLE 2. SUMMARY OF F STATISTICS, ERROR MEAN SQUARES, AND LEAST SQUARES MEANS FOR BIRTH AND SURVIVAL TRAITS OF PROGENY OF 2-YEAR-OLD DAMS (TEUITS OF DAM) df Birth Birth Calving survival or wt, date, daficulty, Birth, 724 Weaning, Item n k3 Julian 46 9% Analysis of variance Bread GOUP (B) Sirea (s/b)' Sex of calf (S) Year of birth Cy) Residuala Least squares means CI Breed group means Red Poll (RP) Hereford (H) Ansus ('4) Limousin (L) Braunvieh (B) pinzgauer CP) Gelbvieh (G) Simmental (S) Charolais Q Parental breed mean D.05b MARC I F1 F2 and F3 MARC II F1 F2 F3 MARC III F1 F2 and F3 D.05' ,629 4, ** ** 10.3** I ** 11.41** 2.W** , ** **.I ** 2.51** , ** %ire of cow within breed group and residual mean squares. %.OS is the approximate difference between means of parental breeds required for significance is the approximate difference between means of all bred groups required for significance. **P <.01.

6 BREED HETEROSIS EFFECTS IN BEEF C A m E 9 PP PK?m

7 3580 GREGORY ET AL. The mean square for sire of cow within breed group was used as the error term to test significance of differences among breed groups (F-test) and to compute (D = Q.05 [sq) to determine the approximate difference required for significance among parental breeds and as the error term for linear contrasts to estimate heterosis effects. Analysis of Variance Results and Discussion Breed Group. Breed group effects were significant for birth weight, birth date, calving difficulty percentage, and survival at birth percentage, but not survival percentage at 72 h and survival percentage at weaning in the analysis that included all calves with 2-yr-old dams (Table 2). Among calves with 2-yr-old dams that experienced dystocia, the effects of breed group were significant for birth weight and birth date but not survival traits, whereas among calves in which dystocia was not experienced, breed group effects were significant for birth weight, birth date, survival at birth, at 72 h, and at weaning (Table 3). Among calves with dams 2 3 yr old, breed group effects were significant for all birth and survival traits. Among calves from dams of all ages, the effects of breed group were significant for all birth and survival traits except survival at birth (Table 4). Sex. Among all calves with 2-yr-old dams, the effects of sex were significant for birth weight (39.4 kg for males vs 36.6 kg for females), birth date (78.2 for males vs 77.3 for females), and calving difficulty (65.8% for males vs 39.2% for females) (Table 2). Among calves with 2-yr-old dams that experienced dystocia, the effects of sex were significant for birth weight (41.2 kg for males vs 38.9 kg for females) and birth date (d 78.6 for males vs d 77.3 for females). Among calves from 2-yr-old dams that did not experience dystocia, the effects of sex were significant for birth weight (36.2 kg for males vs 35.3 kg for females) but not for birth date (d 77.2 for males vs d 77.6 for females) (Table 3). Among calves with dams 2 3 yr old, the effects of sex were significant for birth weight (43.8 kg for males vs 40.8 kg for females), birth date (d for males vs d for females), calving difficulty (11.1% for males vs 5.2% for females), and survival at 72 h (95.9% for males vs 96.8% for females) and at weaning (92.2% for males vs 94.3% for females) (Table 4). Among calves from dams of all ages the effects of sex were significant for birth weight (42.7 kg for males vs 40.0 kg for females), birth date (d 98.2 for males vs d 97.0 for females), calving difficulty (24.0% for males vs 14.7% for females), and survival at 72 h (94.1% for males vs 95.0% for females) and at weaning (89.4% for males vs 91.3% for females) (Table 4). Age. Among calves with dams 2 3 yr old, the effects of age of dams were significant for birth weight, birth date, calving difficulty, and survival at 72 h and at weaning. Calves with 3-yr-old dams averaged 1.4 kg lighter at birth than calves with dams 2 4 yr old (41.4 vs 42.8 kg), were born later, experienced a higher level of dystocia, and had lower survival at 72 h and at weaning (Table 4). Among calves with dams of all ages the effects of age of dam were significant for all traits analyzed with differences of greater magnitude than for calves with dams 2 3 yr old. Among calves with dams of all ages, calves with 2-yr-old dams were born earlier because of an earlier and shorter breeding season of yearling females than for females 2 2 yr old (Table 4). The significant interaction of breed group x age for birth weight, birth date, calving difficulty, and survival of calves with dams 2 3 yr old and with dams of all ages did not follow any pattern associated with direction of breed differences. The significant interaction of sex x age for calving difficulty, survival at birth, and survival at weaning of calves with dams 2 3 yr old resulted from a relatively higher level of dystocia and lower survival at birth and at weaning for male calves than for female calves with dams 3 yr old than with dams 4 and 2 5 yr old (Table 4). The significant interaction of sex x age for calving difficulty of calves from dams of all ages resulted from relatively higher levels of dystocia for male calves than for female calves with 2- and 3-yr-old dams. Differences Among Parental Breedr Differences among parental breeds include additive direct genetic effects (G') and additive maternal genetic effects (P). Calves With 2-Yr-Old Dams. Mean birth weight for parental breed calves was 37.7 kg.

8 BREED AND HETEROSIS EppEcrS IN BEEF C4'ITL.E 3581 Birth weight ranged from 31.5 kg in Angus to 42.3 kg in Braunvieh (P e.01; Table 2). Mean birth date (Julian) for parental breed calves was 79. Birth date ranged from 74 in Red Poll and pinzgauer to 88 in Limousin (P <.01; Table 2). Breed rank for birth date is highly associated with age at puberty and(or) gestation length (Gregory et al., 1991a, b). Mean calving difficulty for parental breeds was 52.6%. Calving difficulty ranged from 31.8% in Angus to 73.8% in Braunvieh (P <.01; Table 2). Comparative results on calving difficulty of 2-yr-olds of these purebreeds have not been reported (i.e., breed differences in Ci + W). Mean calf survival at weaning for parental breeds was 80.4%. Survival at weaning ranged from 74.3% in Hereford to 85.4% in Red Poll (P <.05; Table 2). In a separate analysis of all calves with 2-yr-old dams, calving difficulty, and survival at birth, 72 h, and weaning were analyzed with birth weight (linear and quadratic) and birth weight within breed group (linear and quadratic) as covariates. For calving difficulty on birth weight, the linear regression was.042 (P <.01) and the quadratic regression was.oooo2 (P >.05). For survival at birth, the linear regression was (P <.01) and the quadratic regression was -.OOO4 (P e.01). For survival at 72 h, the linear regression was (P <.01) and the quadratic regression was (P <.01). For survival at weaning, the linear regression was (P e.01) and the quadratic regression was (P.01). The linear and quadratic regressions for calving difficulty and survival at birth, 72 h, and at weaning on birth weight within breed group were significant. The effects of breed group were important (P e.ol) for calving difficulty and survival at birth, 72 h, and weaning. The effects of sex were important (P e.01) for calving difficulty (e.g., males = 62% and females = 46%). These results show that in 2-yr-old dams each kilogram increase in birth weight resulted in a 4.2% increase in calves requiring assistance and that the response was linear. The negative linear and quadratic regressions (P.01) of survival at birth, 72 h, and weaning on birth weight reflect the importance of birth weight on calf survival and the curvilinearity of the effect. The significant effect of breed group on calving difficulty and survival at birth, 72 h, and weaning reflect important differences among breed groups for these traits, independent of breed group effects on birth weight. Thus, there seems to be some oppodty to reduce dystocia and to increase calf survival by consideration of factors other than birth weight, such as anatomical characteristics of dam and(or) calf. Similarly, the greater calving difficulty of males vs females documents important effects of sex on calving difficulty, independent of sex effects on biah weight. This is in agreement with the report of Mbnissier and Foulley (1977). To clarify further the relationship of birth weight with calving difficulty and survival at birth, 72 h, and at weaning for all calves with 2-yr-old dams, birth weights were grouped into five classes for each breed group and biah weight class was included in the model as a main effect in an analysis of calving difficulty and survival at birth, 72 h, and at weaning. Birth weight classes for each breed group were 1 < p ; 2 = > p to < p -.5 6; 3 > p-.5 6 to < p+.5 6; 4 = > p+.5 6 to e p ; and 5 = > p (Table 5). Mean 6 (residual standard deviation) for birth weight was 4.8 kg. Also, the model included breed group, sue of cow within breed group (random), sex, year, and the interactions of breed group x birth weight class and sex x birth weight class. Further, birth weight class was partitioned into orthogonal polynomials for a main effect (e.g., linear, quadratic, cubic, and quartic). The number of observations in each birth weight class was as follows: 1 = 284,2 = 995,3 = 1,628,4 = 894, and 5 = 335 (Table 5). The effects of breed group and birth weight class were important (P <.01) for all traits analyzed. Calf survival at birth, 72 h, and weaning for each birth weight class was, respectively, 1 = 95, 84, 70; 2 = 97, 93, 84, 3 = 96,90,83; 4 = 95,86,80; and 5 = 90,74,68 (Table 5 and Figure 1). The effects of sex were significant for calving difficulty (males = 59% and females = 48%) but not for survival traits. Mean calving difficulty by birth weight class W~S 1 = 15,2 = 33,3 = 54,4 = 73, and 5 = 92 (Figure 2). Only the linear regression was significant for calving difficulty. The linear, except for survival to weaning (P =.12), and quadratic regressions were significant for all survival traits. The cubic and quartic regressions were not important for any trait analyzed. The interaction of breed group x birth weight class was important (P <.01) for

9 3582 GREGORY ET AL. 11

10 BREED AND HETEROSIS EFPECTS IN BEEP CATIZE 3583 TABLE 5. LEAST SQUARES MEANS FOR CALF SURVIVAL PERCENTAGE TO WEANING BY BREED GROUP AND BIRTH WEIGHT CLASS FOR CALVES WITH 2-YEAR-OLD DAMS (TRAIT OF DAM) Birth weieht class >p-1.5ato >p-soto >p+.sato Item XI <p-15~ Cp-.5Q <p+so <p+1.5o >1+1.5O Allbreeds 4, a b Breed group Red Poll Hereford Angus Limousin Braunvieh F'inzgauer Gelbvieh Simmentd Charolais MARC I FI F2 and F MARC II F F2 and F MARC III F F2 and F anumber of observations for each subclass. %lean survival for each subclass. survival at 72 h and at weaning but not for other traits. The interaction of sex x birth weight class was significant for calving Miculty and survival at 72 h but not for survival at birth or at weaning. The significant interaction of sex x birth weight class for calving difficulty was the result of lower difficulty for female than for male calves in the intermediate birth weight classes (e.g., 2, 3, and 4). but there was no difference between sexes in the lowest and the highest birth weight classes (i.e., 1 and 5). The significant interaction of sex x birth weight class for survival at 72 h resulted from lower survival of female calves than of male calves in the two highest birth weight classes (i.e., 4 and 5) but equal or higher survival for female calves than for male calves in other birth weight classes. The linear effects of birth weight class on calving difficulty were similar to the results obtained in the analysis in which birth weight was included in the model as a covariate (Figure 2). The significant effect of birth weight class on calf survival at birth was primarily the result of reduced survival of calves with birth weights heavier than 1.5 CJ above breed group mean birth weight, whereas calf survival percentage at weaning was reduced for calves that were lighter than 1.5 IS below breed group mean birth weight as well as for calves that were heavier than 1.5 IS above breed group mean birth weight (Table 5 and Figure 1). Thus, the optimum birth weight for calf survival at weaning was an intermediate and did not differ (P >.05) between > p IS and > p IS from breed group mean (Table 5 and Figure 1). This range includes 85% of the calves in the data set analyzed (Table 5). This result is in agreement with the report of M6nissier and Foulley (1977). For calf survival at 72 h, birth weight classes 2 and 3 were highest, classes 1 and 4 were intermediate, and class 5 was lowest (Figure 1). These results reveal important differences among breed groups for calving difficulty and for survival at birth, 72 h, and at weaning other than through breed group effects on birth

11 3584 GREGORY ET AL. 6 / d,atz /"- P- <u-l 50 >p-1.5a to <p BIRTH WEIGHT CLASS Figure 1. Relationship of calf survival at birth (x), at 72 h (0). and at weaning (0) to birth weight class in 2-yr-old females for all breed groups. weight. The breed group effect was not different (P >.OS) across birth weight classes for calving difficulty and survival at birth, but the breed group effect was different (P <.01) across birth weight classes for survival at 72 h and at weaning (Table 5). Some breed groups did not differ greatly in survival at weaning across birth weight classes (e.g., Red Poll, Charolais, Pinzgauer, and Composite MARC I), whereas other breed groups did differ greatly in survival at weaning across birth weight classes (e.g., Braunvieh, Angus, Limousin, Gelbvieh, and Composite MARC IQ Table 5). Thus, this interaction does not follow any pattern associated with differences among breed groups in biological type for major bioeconomic traits. Calves With 2-Yr-Old Dams With and Without Dincult Births. Mean birth weight of calves with Micult births was 4.2 kg greater >p+.50 to >.q+l.sa 2 3 BIRTH WEIGHT CLASS 4 5 Figure 2. Relationship between calving difficulty and birth weight class in 2-yr-old females for all breed groups.

12 BREED AND HETEROSIS EPFECTS IN BEEF CATTLE 3585 (P c.01) than mean birth weight of calves that did not experience dystocia (39.6 vs 35.4 kg). Mean survival to weaning was 7.7% greater (P <.01) in calves that did not experience dystocia than mean survival to weaning in calves that did experience dystocia (85.1 vs 77.4%; Table 3). Among calves with difficult births, mean birth weight ranged from 33.8 kg in Angus to 44.7 kg in F inzgauer (P <.01). Mean survival to weaning ranged from 71.5% in Limousin to 84.1% in Angus (P >.OS). Among calves in which dystocia was not experienced, mean birth weight ranged from 30.5 kg in Angus to 39.5 kg in Braunvieh (P c.01). Mean survival to weaning ranged from 75.2% in Hereford to 90.4% in Pinzgauer (P <.01; Table 3). In a separate analysis of calves with 2-yr-old dams with difficult births, survival at birth, 72 h, and weaning were analyzed with birth weight (linear and quadratic) and birth weight within breed group (linear and quadratic) as covariates. For survival at birth, 72 h, and weaning the regressions (linear and quadratic) were important (P <.01) (e.g., suwival at birth and -.OOO8, survival at 72 h and -.001, and survival at weaning and ). The regressions (linear and quadratic) on birth weight within breed group were all important (P <.Ol), except the quadratic regression for survival at weaning, which only approached significance (P =.lo). The effects of breed group were significant on survival at birth, 72 h, and weaning in this analysis. In a separate analysis of calves with 2-yr-old dams without difficult births, survival at birth, 72 h, and at weaning were analyzed with birth weight (linear and quadratic) and birth weight within breed group (hear and quadratic) as covariates. For survival at birth, 72 h, and weaning the regressions (linear and quadratic) were not consistent (e.g., survival at birth [P <.05] and [P >.05], 72 h -.OOO6 [P >.05] and [P <.01], and at weaning.0058 [P <.05] and.0014 [P <.Ol]). Thus, among calves that did not experience dystocia, higher birth weights resulted in higher calf sutvival to weaning. The regression of survival at birth, 72 h, and weaning (linear and quadratic) on birth weight within breed group were significant, except the quadratic regression for survival at weaning. The effects of breed group were significant for survival at birth, 72 h, and weaning. Calves With Dams 2 3 Yr Old. Mean birth weight of parental breed calves with dams 2 3 yr old was 42.0 kg. Birth weight ranged from 34.6 kg in Angus to 47.4 kg in Pinzgauer (P <.01; Table 4). Mean birth date (Julian) of parental breed calves with dams 2 3 yr old was 105. Birth date ranged from 99 in Angus to 109 in Limousin and Braunvieh (P c.01; Table 4). Mean calving difficulty of parental breed calves with dams 2 3 yr old was 8.6%. Calving difficulty ranged from.9% in Angus to 15.9% in F inzgauer (P <.01; Table 4). Mean calf survival at weaning of parental breed calves with dams 2 3 yr old was 92.6%. Survival at weaning ranged from 88.1% in Simmental to 95.6% in Red Poll (P <.01; Table 4). In a separate analysis, gestation length, birth date, birth weight, calving difficulty, survival at birth, and survival at 72 h were analyzed for calves with dams 2 3 yr old. Data on gestation length were not available on calves with 2-yr-old dams. The model used in this analysis included breed pup, sire of cow within breed group (random), sex, year of birth, age of cow, and the interactions of breed group x sex, breed group x age, and sex x age. Breed group effects were significant for gestation length, birth date, birth weight, and calving difficulty but not for survival at birth and at 72 h. A second analysis of these traits was conducted with gestation length, gestation length within sex, and gestation length within brsed group as covariates. The linear regressions of all traits on gestation length (d) were significant (e.& birth date, 1 d; birth weight,.4 kg; survival at birth,.l%; and survival at 72 h,.2%). The regressions of all traits, except birth weight, on gestation length within sex were significant. The regressions of all traits, except birth date, on gestation length within breed group were significant. Gestation length accounted for 90% of the breed group variation in birth date, 14% of the breed group variation in birth weight, and 31% of the breed group variation in calving difficulty. Calves With Dams of All Ages. Mean birth weight of parental breed calves with dams of all ages was 40.9 kg. Birth weight ranged from 33.9 kg in Angus to 46.2 kg in Pinzgauer (P <.01; Table 4). Mean birth date (Julian) of parental breed calves with dams of all ages was 99. Birth date ranged from 93 in Angus to 104 in Limousin (P <.01; Table 4). Mean calving difficulty of parental breed calves with

13 ~~ 3586 GREGORY ET AI. TABLE 6. EFFECTS OF RETAINED HETEROSIS ON BIRTH AND SURVIVAL TRAlTS OF PROGISTY OF 2-YEAR-OLD DAMS (TRAITS OF DAM) Birth Calving Birth swival WeigW diffkulty, date, weaning, Item kg 46 Julian 46 Linear contrasts Heterosis MARC I F1 minuspurebrads 2.3** ** 2.6 F2 and F3 minus purebreds 3.w* ** H F1 minus H Fz and F3' MARC II F1 minus purebreds 2.P* ** -.5 F2 and F3 minus purebreds 2.9'. 8.6** 44** H F1 minus H Fz and F3' ** MARC m Fi minus purtbreds 12** ** 3.5 F2 and F3 minus purebreds 1.4** H F1 H F2 and F3' Mean heterosis AU composites F1 minus purebreds F2 and F3 minus purebreds.87 H Fq min~ H F? and Fqa 1.8** ** ** 4.5* -2.7** %inear contrasts of observed and expected heterosis to test hypothesis that re* heterosis is proportional to retained heterozygosity when H + Hm = 1 and H' =.S and I"' = 5. (See Table 1 for # and Hm expectations for each generation). *P <.os. **P <.01. dams of all ages was 19.7%. Calving difficulty ranged from 8.8% in Angus to 28.5% in Braunvieh (P c.01; Table 4). Mean survival at weaning of parental breed calves with dams of all ages was 89.7%. Survival at weaning ranged from 86.4% in Simmental to 93.4% in Red Poll (P c.01; Table 4). Heterosis All characters were analyzed as traits of the dam (e.g., F2 generation progeny of F1 generation dams and F3 and 4 generation progeny of F2 3 generation dams). The F3 and F4 generation progeny of F2 and F3 generation dams were combined because they did not differ (P >.05), which is consistent with genetic expectation assuming grand maternal heterosis = 0 (Table l). Expectation for maternal heterosis is (.94 P) for composite MARC I and (1 P) for composites MARC II and MARC III for the F1 generation dams. Expectation for individual heterosis for the F2 generation progeny of F1 generation dams is (.75 H!) in composites MARC 11 and UARC III and (.78 H') in composite MARC I. Thus, expectation for mean heterosis for the F2 generation progeny of F1 generation dams of the three composite populations is ( P). Expectation for mean heterosis for the F3 and subsequent generation progeny of F2 and subsequent generation dams of the three composite populations is (.76 H! +.76 Hm; Table 1). Calves With 2-Yr-Old Dams. Heterosis effects for birth weight were significant in F2 generation progeny and F3 and 4 generation progeny of F1 generation and F2 3 generation dams for each of the three composite populations and for the mean of all composite populations (Table 6). Heterosis retained for birth weight did not differ (P >.05) from expectation based on retained heterozygosity. The effects of heterosis on calving difficulty were not consistent. Heterosis effects for calving difficulty were significant for F3 4 generation progeny of F2 and 3 generation dams for composite MARC II population and for the mean of the three composite populations. Heterosis effects were not significant for the F2 generation progeny of F1 generation dams of composite MARC II or for any

14 BREED AND HETEROSIS EFFECTS IN BEEF CATTLE 3587 generation of the two other composite populations or for the mean of F2 generation progeny of F1 generation dams of the three composite populations (Table 6). Heterosis retained for calving difficulty Mered (P <.01) from expectation based on retained heterozygosity only in the MARC 11 composite population. The F2 generation progeny and F3 and 4 generation progeny of F1 generation and F2 and 3 generation dams were born earlier (P <.01 except for F3 and 4 generation progeny of F2 and 3 generation dams in composite MARC III) than their parental purebreds, reflecting earlier conception that likely resulted from a younger age at puberty in the composite populations (Gregory et al, 1991b). Heterosis retained for birth date did not differ (P >.05) h m expectation based on retained heterozygosity. Although the effects of heterosis on survival to weaning were not significant, the values were positive except for F2 generation progeny of F1 generation dams in the composite MARC II population (Table 6). Calves With Dams 2 3 Yr Old. The effects of heterosis on birth weight were significant for all generations of the three composite populations and for the mean of the three composite populations (Table 7). Heterosis retained for birth weight did not differ (P >.05) from expectation based on retained heterozygosity (Table 7). Heterosis resulted in an earlier birth date, and mean heterosis for birth date was sisnifcant for F2 generation progeny of F1 generation dams and for F3 and 4 generation progeny of F2 and 3 generation dams. Heterosis retained for birth date did not differ (P >.05) from expectation based on retained heterozygosity (Table 7). The effects of heterosis on calving difficulty generally were not significant for the three composite populations or for the mean of the three composite populations in either generation (Table 7). The effects of heterosis on survival at 72 h and at weaning were significant for F2 generation progeny of F1 generation dams in composite populations MARC II and MARC III and for the mean of the three composite populations (Table 7). Heterosis retained for calf survival did not differ (P >.05) from expectation based on retained heterozygosity (Table 7). Calves From Dams of All Ages. The effects of heterosis on birth weight were significant for all generations of each of the three composite populations and for the mean of the three composite populations. Heterosis retained for birth weight did not differ (P >.OS) from expectation based on retained heterozygosity (Table 7). The effects of heterosis on birth date favored (P <.01) the three composite populations and the mean of the three composite populations. Heterosis retained for birth date did not differ (P >.05) from expectation based on retained heterozygosity (Table 7). The effects of heterosis on calf survival at weaning were positive and significant for F3 and 4 generation progeny of F2 and 3 generation dams in composite population MARC II and for F2 generation progeny of F1 generation dams in composite population MARC III and ap proached significance (P =.06) in 73 4 generation progeny of F2 and 3 generation dams in composite population MARC I. Mean heterosis was significant for survival at weaning for F2 generation progeny of F1 generation dams and approached significance (P =.a) for F3 and 4 generation progeny of F2 a d 3 generation dams. Heterosis retained for survival at weaning did not differ (P >.OS) from expectation based on retained heterozygosity (Table 7). General These results show large differences among breeds in calving difficulty, particularly in calves with 2-yr-old dams. Calves with micult births with 2-yr-old dams were significantly heavier at birth and had significantly lower survival at 72 h and at weaning than calves with 2-yr-old dams that did not experience difficult births. Large differences were observed among breed groups in calving difficulty and calf survival independent of breed group effects on birth weight. This result suggests some opportunity to reduce dystocia and to increase calf survival by consideration of factors other than birth weight, such as anatomical characteristics of dam and(or) calf. Similarly, greater calving difficulty was observed in male calves than in female calves independent of sex effects on birth weight, indicating that anatomical differences between sexes likely contribute to dystocia. Calf survival at weaning was lowest (P <.05) in smallest and in largest birth weight classes and did not differ (P >.05) among intermediate birth weight classes, documenting that interme diate birth weights are optimum for increased survival.

15 3588 GREGORY ET AL.

16 BREED AND HETEROSIS EFFECTS IN BEEP CAl'TLE 3589 These results show a significant heterotic effect on birth weight in progeny of cows of all ages. The effects of heterosis on calving difficulty were not consistent and generally were not significant. Thus, the effects of heterosis on birth weight are not reflected in increased calving difficulty as traits of the dam. These results are interpreted to suggest that heterotic effects on cow size likely are sufficient to accommodate increased calf birth weight resulting from heterosis without increasing dystocia. Heterosis retained for all traits analyzed did not differ (P >.05) from expectation based on retained heterozygosity. Gregory and Cundiff (1980) presented results showing that the proportion of retained heterosis was not less than the p'oportion of retained heterozygosity in rotational crossbreeding. This suggested that heterosis in cattle is the result of dominance effects of genes. In a comparison of inter se mated F3 vs F1 generation Hereford-Angus crosses, Koch et al. (1985) reported that heterosis retained was greater than expected based on retained heterozygosity for four growth-related traits (postweaning gain, final weight, carcass weight, and ribeye area), equal to expectation for six traits (day born, birth weight, calving ease, preweaning gain, weaning weight, and fat thickness), and less than expected for three traits (survival rate, pregnancy rate, and marbling). implications Because large differences exist among breed groups in dystocia and calf survival, independent of breed group effects on birth weight, there seems to be opportunity to reduce dystocia and increase calf survival by consideration of factors other than birth weight, such as anatomical characteristics of dam and(or) calf. Heterosis effects on cow size seem to be sufficient to accommodate increased calf birth weight, resulting from heterosis, without increasing dystocia Heterosis retained for birth weight, birth date, and calf swival in advanced generation of com- posite populations does not differ &om expectation based on retained heterozygosity. Thus, heterosis in these traits seems to be due to dominance effects of genes and probabiy reflects recovery of accumulated inbreeding depression. Literature Cited Cundiff, L. V., K. E. Gregory, R. U Koch and G. E. Dickerson Genetic diversity among cattle breeds and its use to inmase production &iciency in a temperate environment. In: Proc. 3rd World Cow. on Genet. AppL to Livest. Prod. Lincoln, NE Dickerson, G. E Experimental approaches in utilizing breed resources. Anim. Breed. Ah@. 37:191. Dickerson, G. E Inbnxding and heterosis in animals. In: Roc. of the Anirn. Breeding and Genet. Symp. in Honor of Dr. Jay L. Lush. pp Am. SOC. of Anim. Sci., Champaim E. Gregory, K. E. and L. V. Cundiff Crossbreeding in beef cattle: Evaluation of systems. J. Anim. Sci. 51: Gregory, K. E., L. V. cundiff and R. M. Koch Comparison of crossbreeding systems and breeding stocks used in suckling kds of continental and tempexate areas. Plenary Session. In: Proc. 2nd World Congr. on Genet. Appl. to Livest. Prod. Madrid, Spain. 5:482. Gregory,K.E.,L.V.CundiffandR.M.Koch. 1991a. Breed CEFtCts and hetemsis in advanced generations of composite populations for preweaning traits of beef cattle. I. Anim. Sci Gregory, K E., D. D. Lunstra, L. V. Cmdiff and R. M. Koch. 1991b. Breed effects and heterosis in advanced generatiom of composite populations for puberty and scrotal traits of beef cattle. I. Anim. Sci Harvey, W. R User's guide for LSMLMW, mixed model least-squares and maximum likelihood computer program. The Ohio State Univ., Columbus (Mimeo). Koch, R. M, G. E. Dickerson, L. V. Cundiff and K. E. Gregory Heterosis retained in advanced gmerations of crosses among Angus and Hereford cattle. J. him. Sci Mhissier, F. and J. L. Foulley Present situation of calving problems in the EEC: Incidence of calving difficulties and early calf mortality in beef herds. In: Calving Problans and Early Viability of the Calf, Current Topics in Veterinary Medicine and Animal Science. EEC 430. Snedecor, G. W. and W. G. Cochran Statistical Methods (7th Ed.). Iowa State Univ. Press, Ames. U.S.DA.. Agricultural Statistics Government printing office, Washington, DC. Wright, S Effects of inbreeding arid crossbreeding on guinea pigs. III.USDA Bull. 1121, Washington, DC.