Reproductive Performance of Crossline and Pureline Dairy Heifers1

Reproductive Performance of Crossline and Pureline Dairy Heifers 1 C. Y. L I N , 2 A . J. M c A L L I S T E R , 2 T. R. B A T R A , 2 A. J. L E E , 2 G. L. R O Y , 3 J. A. V E S E L Y , 4 J. M. W A U T H Y , s and K. A . W I N T E R s Research Branch Agriculture Canada

ABSTRACT

INTRODUCTION

Data of 2779 purebred and crossbred heifers collected from five research stations of Agriculture Canada were used to study additive and nonadditive genetic effects on ages at first heat and at first breeding and conception rate at first service. Of these heifers, 2378 heifers had information on ages at first conception and at first freshening, days from first service to conception, and gestation length. The model included station, year of birth, sire, breed additive, maternal, and heterosis effects where sire effects were treated as random. Station differences were a significant source of variation for all reproductive traits. Year of birth had significant effects on four of seven reproductive traits. Breed additive effects for all genetic groups were not significant except for Finnish Ayrshire and American Holstein. No significant maternal effects were detected. Of 21 combinations of heterosis effects, six combinations showed significance. Partial regression coefficients ranged from negative to positive, suggesting that breed additive, maternal, and heterosis effects could increase or decrease for each percent increase of genetic contribution, depending upon the trait, breed group, and type of inheritance.

Received October 17, 1983. 1Animal Research Centre Contribution No. 1188. ~Anirnal Research Centre, Ottawa, Ontario, K1A 0C6. 3Lennoxville Research Station, Quebec. 4Lethbridge Research Station, Alberta. s Normandin Research Station, Quebec. Charlottetown Research Station, Prince Edward Island. 1984 J Dairy Sci 67:2420-2428

Heifer reproductive problems represent a major economic loss to the dairy industry before heifers have an opportunity to initiate a first lactation and generate income. Poor heifer reproduction can lower the herd calving rate and retard rate of genetic progress. Reproductive problems are primary reasons for disposal of lactating cows (1, 2, 4, 5, 19, 24). Few studies have examined genetic effects on heifer reproduction. Heifer reproductive performance has received attention partly because of the large proportion of heifer disposals for reproductive failure and partly because reproductive efficiency determines how soon a heifer's productive life begins. Lin and Allaire (14) estimated that the economic value of a 1-mo decrease of age at first calving was equivalent to an increase of 138 kg milk in first lactation in terms of total profit to 72 mo of age. Batra and Touchberry (3) found that the gestation period of Holstein x Guernsey crossbreds was 1.2 days shorter than that of purebreds. However, Little and May (15) found that crossbreds of AberdeenAngus and British Friesian breeds had a longer gestation period than British Friesian (278.4 vs. 273.4 days). McDowell (17) summarized several USDA (United States Department of Agriculture) crossbreeding experiments in the S-49 Southern Regional Cooperative Research Project and concluded that crossbreds tended to surpass purebreds in overall breeding efficiency. The extent of heterosis varied with breeds used as sires or dams, location, management practices, specific crosses, and among reproductive traits. A lower proportion of barren heifers that were bred but failed to calve and a higher first service conception rate have been reported for crossbred versus purebred heifers in the UK (United Kingdom) (7). The purrpose of this study was to compare purebred and crossbred heifer reproductive performance and to evaluate environmental

2420

GENETIC EFFECTS OF HEIFER REPRODUCTION effects of station (herd) and year of birth on these traits. M A T E R I A L S A N D METHODS Experimental Procedure

Data were collected from dairy cattle breeding project of Agriculture Canada (AC). This breeding project was initiated in 1972 and conducted cooperatively by five research stations of Agriculture Canada at Charlottetown, Prince Edward Island; Normandin, Quebec; Lennoxville, Quebec; Ottawa, Ontario; and Lethbridge, Alberta. The breeding project was undertaken primarily to determine the response to pure-line selection for first-lactation, 308-day protein yield in two lines of dairy cows and to study the effectiveness of crossbreeding between the two selected lines. For ease of presentation, the following notation for each of the original genetic groups in the crossbreeding will be: Agriculture Canada Holstein (ACH), Canadian Holstein (CH), United States Holstein (USH), Agriculture Canada Ayrshire (ACA), Canadian Ayrshire (CA), United States Ayrshire (USA), Finnish Ayrshire (FA), United States Brown Swiss (BS), and Norwegian Red Cattle (NR). Details of design and management procedures of the breeding project are described by McAllister et al. (16). Briefly, the ACH cows were inseminated with semen from ACH, CH, and USH bulls to create H-line foundation animals. Similarly, the ACA cows were inseminated with semen from the ACA, CA, USA, FA, BS, and NR bulls to create A-line foundation animals. The ACH and ACA cows were from closed populations that had been selected for yields of total milk solids (12). They were markedly different in milk yield. Bulls of the different genetic groups were introduced to broaden the genetic base of the A and H line populations and in particular to raise the additive genetic value of the A line for protein yield and body size. Subsequent to production of H and A pureline foundation animals, the two lines were closed and maintained as closed breeding populations. The H and A pureline foundation animals were not mated randomly inter se before they were used to produce the crossline foundation animals. Panmictic mating to break up linkage groups in

2421

the pureline foundation populations would have delayed conducting the crossbreeding phase of the project for about 15 to 20 yr. The H and A foundation females were mated with bulls of their respective lines to produce pureline progeny. Concurrently, about one-third of each of the H and A foundation females were mated with bulls of the other line over 4 yr to produce the crossline foudation animals. Heifers derived from pureline and crossline foundation groups were available for study of additive, maternal, and heterosis effects on heifer reproductive traits. Calves were separated from their dams within 24 h of birth and reared individually in calf stalls on a limited whole-milk feeding program for the first 8 wk. All heifer calves underwent an individual feeding trial from 26 to 34 wk of age. Calf starter-grower was fed for ad libitum consumption up to a maximum of 2.5 kg per day for the first 34 wk. From 34 to 50 wk of age the starter-grower was fed at 1.8 kg per day. Hay or silage was fed for ad libitum intake to heifers until 2 wk before first calving. Heifers were observed for heat twice a day. All heifers were bred at first heat after reaching 350 days of age. Frozen semen in pellet form was used for all inseminations. Any heifer not confirmed pregnant by rectal palpation by 574 days of age was removed from the herd. Heifer reproductive traits were age at first observed heat, age at first breeding after 350 days of age, age at first conception, age at first calving, conception rate at first service, number of days from first service to conception, and gestation length (days from last service to first calving). A total of 2779 purebred and crossbred heifers had information on ages at first heat and at first breeding and conception rate at first service. Of these, 2378 heifers had information on ages at first conception and at first freshening, days from first service to conception, and gestation length. Therefore, separate analyses were applied to both sets of data. Data Analysis

The statistical model for analysis of reproductive traits was: Yijklmn = M+ ~kiai+ Zkj aj + ]~kij hij + •k*jmj+Hk+Tl+Sm+eijklmn Journal of Dairy Science Vol. 67, No. 10, 1984

2422

LIN ET AL.

where Yijklmn is individual observation; M is the intercept; k i and k'l are, p pro ortion of genes c o n t r i b u t e d by the i m ~enetic group through the sire and the jtn genetic group through the dam; ai and aj are additive effects of the ith and jth genetic groups; kij is the p r o p o r t i o n of genes c o n t r i b u t e d by the ith and jth genetic groups; hij is the heterosis e f f e c t b e t w e e n the ith and jth genetic groups; k* i is the p r o p o r t i o n o f genes in the dam "from the jth genetic group; mj is the maternal effect of the jth genetic group; H k is the k th station effect; T1 is the lth year o f birth effect; S m is the m t h sire effect; and eijklmn is a r a n d o m residual. All effects e x c e p t sire and residual were fixed. Because sires are treated as r a n d o m rather than fixed, inclusion of sires in the m i x e d m o d e l analysis will n o t adjust out breed group differences. However, exclusion of sires f r o m the m o d e l w o u l d inflate residual variance. Mixed m o d e l analyses were applied to the reproductive traits as t h e y are theoretically

superior to least squares analyses (11). The inverse of relationship m a t r i x (A - 1 ) was n o t used. The well-known m i x e d m o d e l equations (10) are of the f o r m :

rxx x z kz'x

][:]

z'z + i op,,

Lz'rJ

Mixed m o d e l analyses algorithms (10) were used for c o m p u t i n g m a x i m u m likelihood estimates o f sire and residual variance c o m p o n ents for reproductive traits. Initial variance ratios of 39(i.e., h 2 = .1) were used to start iteration for reproductive traits. Initial solutions to m i x e d m o d e l equations were o b t a i n e d by direct m a t r i x inversion and f o u r rounds of iteration. Sire and residual c o m p o n e n t s of variance were estimated after each iteration to be used for n e x t iteration. N u m e r a t o r sum squares (Q) and F statistics were c o m p u t e d at the end of the last iteration. Genetic effects (ai, h!j, and mj) and prop o r t i o n of additive genetic, heterozygotic, and maternal contributions (i.e., ki, kij, and k~'j) by each genetic group were defined in the m o d e l in the same way as outlined by R o b i s o n et al. (22), who e x t e n d e d Gardner and Eberhart's m e t h o d (9) to analyze crossbreeding data o f dairy cattle. In retrospect, the techniques of

TABLE 1. Number of animals by genetic group. Genetic group a

No. animals

Genetic group

ACH ACA ACH × ACA CH X ACH USH × ACH ACA X ACH CA X ACA USA X ACA FA × ACA BS X ACA NR X ACA ACH X (CH X ACH) ACH X (USH X ACH) (USH X ACH) X ACH (USH X ACH) X ACA (FA X ACA) X ACH

4O2 246 26 121 81 14 18 19 59 24 51 24 18 110 18 26

(FA X ACA) X ACA (BS X ACA) × ACH (BS X ACA) X ACA (NR × ACA) × ACH (NR X ACA) X ACA (CH X ACH) X (CH X ACH) (CH X ACH) X (USH X ACH) (USH X ACH) X (CH X ACH) (NR X ACA) × (FA × ACA) ((USH X ACH) X ACH) X ACH ACH X ((CH × ACH) X ACH) (CH × ACH) X ((CH X ACH) X ACH) (CH X ACH) X ((USH X ACH) X ACH) ((BS X ACA) X ACA) × ACH miscellaneous

No. animals 60 30 21 24 61 23 17 16 13 17 14 10 16 10 789

aACH = Agriculture Canada Holstein, CH = Canadian Holstein, USH = United States Holstein, ACA = Agriculture Canada Ayrshire, FA = Finnish Ayrshire, BS = United States Brown Swiss, NR = Norwegian Red Cattle. Journal of Dairy Science Vol. 67, No. 10, 1984

GENETIC EFFECTS OF HEIFER REPRODUCTION regressing breed group means on proportion of gene contributions in analysis of crossbreeding data have been used by Touchberry (23) and Koger et al. (13). The data consisted of 2779 heifers with a combination of 359 genetic groups. The genetic groups with more than 10 heifers in the data are in Table 1. Coefficients for 9 breed additive effects, 9 maternal effects, and 36 heterosis effects between the 9 original genetic groups were computed for all heifers according to their pedigrees. They are too numerous to be presented in the paper. However, a listing of these coefficients is available on request. Because coefficients for breed additive and maternal effects each sum to one, there are dependencies among equations corresponding to breed additive effects and among equations corresponding to maternal effects. The breed additive, maternal, and heterosis effects involving breed groups CA and USA were not fitted in the model because, as shown in Table 1, the number of heifers involving breed groups CA and USA are too few to make statistical testing meaningful for these two breed groups. Exclusion of breed groups CA and USA from the model also removed dependencies among equations for breed additive and maternal effects, respectively. Thus, breed additive and maternal effects due to CA and USA are effectively included in the intercept M, and all others are deviations from CA and USA. Partial regressions estimated in this model are adjusted for the intercept or other main effects fitted. The full residual variance-covariance matrix is utilized implicitly in estimation of standard errors of partial regressions.

RESULTS A N D DISCUSSION Environmental Factors

Mixed model analyses of heifer reproductive traits from the fourth round of iteration are in Table 2. Effect of station was highly significant (P<.01) for all reproductive traits. This reflects differences in climate, herdsmen, housing facilities, and types of roughage among the five cooperating research stations in spite of attempts to standardize grain feeding and management across stations.

2423

Year of birth was highly significant (P<.01) for days from first service to conception and ages at first heat, at first conception, and at first freshening but was not significant for age at first breeding, conception rate at first service, and gestation length. Because all heifers were bred at first heat after reaching 350 days of age, this restriction seemed to reduce the effect of year of birth on age at first breeding after 350 days of age. No time trend was observed for those traits that showed a significant effect of year of birth.

Breed Additive Effects

Partial regressions for Finnish Ayrshire (FA) additive effects were significant (P<.05) for ages at first heat, at first conception, and at first freshening and approached significance (P = .05) for number of days from first service to conception, conception rate at first service, and gestation length (Table 2). Partial regression for USH additive effect was highly significant (P<.01) for gestation length. Partial regressions for breed additive effects for ACH, CH, ACA, BS, and NR were not significant for all reproductive traits. Only a small number (2 to 5) of bulls were in each of the breed groups other than ACH and ACA. Breed Maternal Effects

Partial regressions for breed maternal effects were not significant for any breed group. However, regressions for Brown Swiss maternal effects for age at first breeding approached significance at 5% probability. This seems to suggest that breed maternal effects might be ignored in estimating breed additive and heterosis effect for reproductive traits. The absolute values of partial regression coefficients for breed additive effects are generally larger than those for the maternal effects for all traits (Table 3). This suggests that variance of breed maternal effects should not be expected generally to be greater than variation of breed additive effects. It is probably for this reason that maternal variances usually are ignored in estimating genetic parameters. However, greater contribution of variation by breed additive effects than by breed maternal effects does not mean necessarily that the former effect is economically more important than the latter. Journal of Dairy Science Vol. 67, No. 10, 1984

2424

O

O

.-&

e~ eq

<

~

Q~

~

~

~

~

Journal of Dairy Science Vol. 67, No. 10, 1984

~

~

'M

LIN ET AL.

"Q

Q~m

"

e-

~

~

~

m

~

Q ~ M ~

~

. . . . .

~

~

M

"

'~

.~QM

"

~

•

.

~-.

•

•

GENETIC EFFECTS OF HEIFER REPRODUCTION

Partial regressions for breed maternal effects were statistically significant for milk yield (22) but not for b o d y weights and b o d y measurements (21). However, no statistical significance of breed maternal effects were reported for heifer reproductive traits in the literature, including this study. Therefore, the importance of breed maternal effects appears to depend upon traits studied•

e~

e q t'~

eq

.~

2425

'7m

- o Heterosis Effects

"o

" 4tt~ •d"

~

<

~

II

V u

< m

N

~

-~

~

z

m~ ,zz

< <

-~ u

Of the 21 partial regressions for combinations of effects of heterosis studied, 15 were not significant for any reproductive trait (Table 2). McDowell et al. (18) found that crossbreds were superior to purebreds by 3.2% in overall breeding efficiency. McDowell (17) reported that heterosis for breeding efficiency was not significant except for first breeding to conception in Holstein x Brown Swiss crosses. Parekh and Touchberry (20) found nonsignificant heterosis for reproductive traits (age at first service, age at calving, and days from first service to conception) in a Holstein and Guernsey crossbreeding experiment at University of Illinois. Ages at first heat and at first breeding showed significant regressions for heterosis effects (P<.01) for heifers derived from the mating between Holsteins and Ayrshires of Agriculture Canada, that had been selected for yields of total milk solids (12). The CH x NR and ACA × F A heifers showed significant regressions for heterosis effects (P<.05) for age at first conception. The ACH × BS heifers exhibited significant regressions for heterosis effects (P<.05) for days from first service to conception and approached significance (P = .05) for ages at first conception and at first freshening. Regressions for CH × NR heterosis effects were significant for ages at first conception and at first freshening, days from first service to conception, and conception rate at first service. Regressions for ACA × F A heterosis effects were significant for ages at first conception and at first freshening. In addition, they approached significance (P = .05) for days from first service to conception. Gestation length showed significant regressions for heterosis effects for ACH X USH, ACH x ACA, and USH × NR (Table 2). It also approached significance (P = .05) for CH × USH Journal of Dairy Science Vol. 67, No. 10, 1984

2426

LIN ET AL.

and CH × NR. Brown Swiss has the longest gestation period among dairy breeds (6, 8). However, BS did not show significant heterotic effects in crossing with other breed groups. The magnitude of a trait for a given breed is not necessarily associated with heterosis as heterosis is from gene interaction (dominance or epistasis). Partial regression coefficients of additive and heterotic effects are generally larger than those of maternal effects (Table 3). This suggests that additive and heterosis effects are a more important source of genetic variation of reproductive traits than maternal effects.

IIllll

Partial Regression Coefficients

Coefficients for breed additive, maternal, and heterosis effects were expressed as decimal fractions in fitting the model. Partial regression coefficients derived from these fractions were converted to partial regression coefficients based on percentages of gene contributions as in Table 3. This was accomplished by dividing the original partial regression coefficients by 100. Standard errors for all partial regression coefficients are in Table 3. Because only USA sires and no USH cows were in this project, the proportion of gene contribution for breed additive and maternal effects of USH never reached a maximum of 100%. Statistically, it would appear less accurate to predict genetic effects of this breed group at 100% of gene contribution. This is also true for CH, BS, FA, and NR, which contributed only sires and no darns to the project. From Table 3 gestation length decreased significantly by .16 days for each percent increase of USH genes controlling breed additive effects. A decrease of gestation length might be of value in maintaining short calving intervals, shortening generation interval, reducing calf size, and possibly lessening calving difficulty. Each percent increase of Finnish Ayrshire (FA) genes responsible for breed additive effects would decrease ages at first heat, at first conception, and at first freshening by .93, .91, and 1.0 days. Gestation length would increase by .09 and .19 days for each percent increase of heterozygosity between ACH and USH and between USH and NR. The ACH × ACA heterotic effects tended to reduce gestation period (.02 day) and ages at first heat (.23 day) and at first Journal of Dairy Science Vol. 67, No. 10, 1984

I l l l l

I

I' I" I" I" ] ' ~

I"

d2

e~ q.,

o=

,. ,. ~o

"d M r," .d

<

.,.

e"

Breed maternal ACH CH USH ACA FA BS NR Heterosis between ACH and CH ACH and USH ACH and ACA ACH and FA ACH and BS ACH and NR CH and USH CH and ACA CH and FA CH and BS CH and NR USH and ACA USH and FA USH and BS USH and NR ACA and FA ACA and BS ACA and NR FA and BS FA and NR BS and NR

-.O8 -.09 -.07 -.11 --.04 --.46 --.12

.18 .20 .21 .18 .23 .26 .23

-.01 .05 -.05 -.01 -.18 --.29* --.11

.10 .11 .12 .10 .13 .15 .13

-.02 -,05 -.23 -.01 .03 --.37 .03

.17 .20 .21 .17 .21 .24 ,22

-.01 -.03 -.21 .01 ,05 -.37 .05

.18 .20 .21 .18 .22 .25 .22

-.03 -.05 -.22 -.00 .13 -.11 .04

.15 .16 .17 .15 .18 .21 .18

.05 -.14 -.03 -.04 -.16 .03 --.10

.17 .19 .21 .17 .22 .25 .22

.00 .01 .02 .02 .02 .00 .02

.01 .02 .02 .01 .02 .02 .02

-.07 --.05 -.23** ,20 -.20 --.09 .04 -.23 -.09 -.62 .92 .07 -.36 .39 -.34 .36 -.45 .04 .66 .40 -.26

.17 .24 .06 .24 .43 .27 .36 .20 .53 .65 .76 .27 .63 .70 .70 .22 .44 .26 .63 .40 .70

-.01 -.03 --.15"* .21 -.02 -.11 .07 -.09 -,19 .10 .69 .06 .53 .08 -.39 .08 -.12 -.07 .48 .20 -.19

.09 .13 .03 .14 .25 .15 .21 .11 .30 .37 .43 .15 .36 .40 .40 .13 .25 .15 .36 .23 .40

-.01 .10 -.05 .35 -.74 --.07 .32 -,09 .07 -.1I -1.65' .18 .43 --.28 .08 .44* .60 -.03 .37 .05 -.85

.16 .25 .06 .25 .40 .28 .37 .19 .51 .61 ,79 .28 .61 .70 .66 .22 .40 .27 .61 .41 .66

-.02 .18 -.07 .36 -.72 -.11 .39 -.10 .13 -.10 -1.81" .23 .43 -.15 .28 .47* -.56 -.06 .44 .03 -.91

.16 .26 .06 .25 .41 .28 .38 .19 .51 .62 .80 ,28 .62 .71 .67 .23 .41 .27 .62 .42 .67

.01 .12 .07 .11 -.68* .03 .28 .00 .43 -.19 -1.81"* .11 .42 -.26 .55 .33* -.43 .08 --.17 -.02 -.50

.13 .21 .05 .21 .34 .23 .31 .16 .43 ,51 .67 .24 .52 .59 .56 .19 .34 .23 .51 .34 .55

-.09 .09 --.03 --.15 .47 -.13 ,30 .04 .08 .16 1.62" .13 .79 .82 .22 -.32 .27 --.25 .30 -.31 -.01

.16 .23 .06 .24 .42 .26 .35 .20 .51 .63 .73 .26 .61 ,68 .68 .22 .42 .25 .61 .39 .68

-.00 .09** -.02* .01 .02 --.03 .07 -.01 .07 ,01 --.16 .05 --.00 .13 .19"* .03 .04 --.03 .07 -.02 -.06

.01 .02 .01 .02 .04 .03 .04 .02 .05 .06 .08 .03 .06 .07 .07 .02 .04 .03 .06 .04 .07

aAll partial regression coefficients were based on the percentage of gene c o n t r i b u t i o n s e xc e pt that the partial regression coefficients for c o n c e p t i o n rate at first service were based on the fraction of gene contributions. W < o ox

Z

E~ © rv

0 t~ Z

bACH = Agriculture Canada Holstein, CH = Canadian Holstein, USH = United States Holstein, ACA = Agriculture Canada Ayrshire, FA = Finnish Ayrshire, BS = United States Brown Swiss, N R = Norwegian Red Cattle. *Significant at P<.05. **Significant at P<.O1.

Z o

00

t,# 4~

2428

LIN ET AL.

b r e e d i n g (.15 d a y ) for e a c h p e r c e n t increase o f h e t e r o z y g o s i t y . T h e A C H x BS h e t e r o t i c effects w o u l d r e d u c e n u m b e r o f days f r o m first service to c o n c e p t i o n b y .68 d a y for each p e r c e n t increase of h e t e r o z y g o s i t y . T h e CH x N R h e t e r o t i c effects w o u l d r e d u c e ages at first c o n c e p t i o n 1.65 days a n d at first f r e s h e n i n g 1.81 days, t h e n u m b e r of d a y s f r o m first service to c o n c e p t i o n , a n d i m p r o v e c o n c e p t i o n rate. O n t h e c o n t r a r y , A C A x F A h e t e r o t i c effects w o u l d increase ages at first c o n c e p t i o n .44 d a y a n d at first f r e s h e n i n g .47 day, require greater n u m b e r o f days f r o m first service to c o n c e p t i o n .33 day, a n d s h o w p o o r e r c o n c e p t i o n rate for each p e r c e n t increase o f h e t e r o z y g o s i t y . Alt h o u g h A C A × F A crossbreds did n o t s h o w desirable h e t e r o s i s in r e p r o d u c t i o n , b r e e d a d d i t i v e effects o f F A were desirable (Table 3). T h e sign a n d m a g n i t u d e of partial regression c o e f f i c i e n t s for b r e e d additive, m a t e r n a l , a n d h e t e r o s i s effects varied d e p e n d i n g u p o n traits a n d b r e e d groups. H o w e v e r , inclusion of o n l y t h e N R m a y have b e e n s u f f i c i e n t to i m p r o v e r e p r o d u c t i v e p e r f o r m a n c e o f crossline heifers as c r o s s b r e e d i n g b e t w e e n N R a n d CH resulted in desirable h e t e r o t i c effects o n r e p r o d u c t i o n ( T a b l e 3). ACKNOWLEDGMENTS

T h e a u t h o r s g r a t e f u l l y a c k n o w l e d g e operat i o n a l and t e c h n i c a l staff at all t h e c o o p e r a t i n g research s t a t i o n s for t h e i r c o n t r i b u t i o n s to this long b r e e d i n g p r o j e c t . REFERENCES

1 Ace, D. L. 1973. One calf out of four never reaches a year. Hoard's Dairyman:981. 2 Allaire, F. R., H. E. Sterwerf, and T. M. Ludwick. 1977. Variations in removal reasons and culling rates with age for dairy females. J. Dairy Sci. 60:254. 3 Batra, T. R., and R. W. Touchberry. 1974. Birth weight and gestation period in purebred and crossbred dairy cattle. J. Dairy Sci. 57:323. 4 Burnside, E. B., and J. W. Wilton. 1970. Anatomical traits as they relate to productive utility. J. Dairy Sci. 53:837. 5 Burnside, E. B., S. K. Kowalchuk, D. B. Lambroughton, and N. M. MacLeod. 1971. Canadian dairy cow disposal. I. Difference between breeds, lactation numbers and seasons. Can. J. Anita. Sci. 51:75. 6 DeFries, J. C., R. W. Touchberry, and R. L. Hays. 1959. Heritability of the length of the gestation period in dairy cattle. J. Dairy Sci. 42: 598. 7 Donald, H. P., and W. S. Russel. 1969. Some aspects of fertility in purebred and crossbred dairy Journal of Dairy Science Vol. 67, No. 10, 1984

cattle. Anita. Prod. 10:465. 8 Foote, R. H. 1981. Factors affecting gestation length in dairy cattle. Theriogenology 15:553. 9 Gardner, C~ O., and S. A. Eberhart. 1966. Analysis and interpretation of the variety cross diallele and related populations. Biometrics 22:439. 10 Henderson, C. R. 1973. Sire evaluation and genetic trends. Proc. Anim. Breeding Genet. Syrup. in Honor of Dr. J. L. Lush, Am. Soc. Anim. Sci., Am. Dairy Sci. Assoc., Champaign, ]L. 11 Henderson, C. R., Jr., and C. R. Henderson. 1979. Analysis of covariance in mixed models with unequal subclass numbers. Commu. Statist. Theor. Methods A8: 751. 12 Hickman, C. G. 1971. Responses to selection of breeding stock for milk solids production. J. Dairy Sci. 54:191. 13 Koger, M., M. Peacock, W. G. Kirk, and J. R. Crockett. 1975. Heterosis effects on weaning performance of Brahman-Shorthorn calves. J. Anita. Sci. 40:826. 14 Lin, C. Y., and F. R. Allaire. 1977. Relative efficiency of selection methods for profit in dairy cows. J. Dairy Sci. 60:1970. 15 Little, W., and R. M. May. 1979. The effects of rapid rearing and early calving on the subsequent performance of dairy heifers. Anita. Prod. 29:131. 16 McAllister, A. J., T. R. Batra, J. P. Chesnais, J.P.F. Darisse, J.A.B. Emsley, A. J. Lee, J. Nagai, G. L. Roy, J. A. Vesely, and K. A. Winter. 1978. The National Cooperative Dairy Cattle Breeding Project. ARC Tech. Bull. No. 1, Agric. Can., Ottawa. 17 McDowell, R. E. 1982. Crossbreeding as a system of mating for dairy production. Southern Coop. Serv. Bull. No. 259. Louisiana State Univ., Baton Rouge. 18 McDowell, R. E., J. A. Velasco, L. D. Van Vleck, J. C. Johnson, G. W. Brandt, B. F. Hollon, and B. T. McDaniel. 1974. Reproductive efficiency of purebred and crossbred dairy cattle. J. Dairy Sci. 57:220. 19 O'Bleness, G. V., and L. D. Van Vleck. 1962. Reasons for disposal of dairy cows from New York herds. J. Dairy Sci. 45:1087. 20 Parekh, H.K.B., and R. W. Touchberry. 1981. Comparative reproductive performance of crossbreds and contemporary pure-bred dairy cattle: heifers and their first calving dry periods records. Ind. J. Anim. Sci. 51:756. 21 Robison, O. W., M. G. Kelly, B. T. McDaniel, and R. E. McDowell. 1980. Genetic parameters of body size in purebred and crossbred dairy cattle. J. Dairy Sci. 63:1887. 22 Robison, O. W., B. T. McDaniel, and E. J. Rincon. 1981. Estimation of direct and maternal additive and heterotic effects from crossbreeding experiments in animals. J. Anim. Sci. 52:44. 23 Touchberry, R. W. 1970. A comparison of the general merits of purebred and crossbred dairy cattle resulting from twenty years (four generations) of crossbreeding. 19th Annu. Natl. Breeders Roundtable, Kansas City, MO. 24 White, J. M., and J. R. Nichols. 1965. Reasons for disposal of Pennsylvania Holstein cattle. J. Dairy Sci. 48:512.