JOURNAL TRANSCRIPT
GENETIC AND ENVIRONMENTAL GROWTH TRAIT PARAMETER ESTIMATES FOR BRAHMAN AND BRAHMAN-DERIVATIVE CATTLE112 L. A. Kriese?, J. K. Bertrand3 and L. L. Benyshek3 The University of Georgia, Athens 30602 ABSTRACT
Beefmaster, Brahman, Brangus, and Santa Gertrudis field data records were used to determine genetic and environmental parameter estimates using a multiple-trait, pseudoexpectation approach. Adjusted birth weight, 205d weight, and postweaning gain records were analyzed for each b r d Also, Brangus weaning sheath and navel scores were both analyzed using a single-trait, pseudo-expectation method to determine genetic parameter estimates. Additive birth weight heritability (h2& estimates ranged from .22 to .37 and maternal birth weight heritability (h2d estimates ranged from .12 to .55. Estimates for 205d weight h2A for the four breeds varied from .21 to .25, and 205-d weight h2M estimates ranged from .15 to 21. Postweaning gain h2A estimates ranged from .16 to .56. The genetic correlation between direct and maternal portions of birth weight was negative for all breeds. This was also true for the genetic correlation between direct and maternal portions of 205-d weight, except in Brahman cattle, for which it was .15. The genetic correlation between additive portions of birth weight and 205d weight was large and positive in all breeds. A moderately positive correlation between 205d weight and postweaning gain was found for all breeds except Santa Gertrudis, whereas the environmental correlation between these two traits was a small to moderately negative estimate in all breeds. Brangus weaning sheath and navel score heritabilities indicated that genetic change for the size and shape of the sheath and navel area is possible. Key Words: Growth Traits, Sheath and Navel Score, Brahman Cattle, Heritability, Genetic Effects, Environmental Effects J. Aaim. Sci. 1991. 69:2362-2370
Introduction
Brahman and Brahman-derivative cattle, such as Brangus, Beefmaster, and Santa Gertrudis, are important beef cattle breeds in the southern and tropical regions of the United
‘A portion of this research was funded by the American Brahman Breeders Association, Houston, “X. the Santa Gertrudis Breeders International,Kingsville, TX, the Beefmaster Breeders Universal, San Antonio, ‘Ix,and the International B r a u p Breeders Association, San Antonio, Tx. %%is study was partially funded by the Georgia Agric. Exp. Sta. and was part of Hatch Project 873. 3Anim. and Dairy Sci. Dept. Received March 2, 1990. Accepted December 7, 1990.
States. These cattle are heat- and diseaseresistant and have become fixtures in the U.S. beef cattle industry. Genetic and environmental parameter estimates must be obtained for the development of sound breeding programs and National Cattle Evaluation (NCE) programs for these breeds. Limited information is currently available on genetic and environmental parameter estimates for breeds of Brahman and Brahmanderivative cattle. Aaron et al. (1987) reported heritability and genetic correlation estimates for birth weight and weaning weight using a large herd of Santa Gertrudis cattle and found parameter estimates comparable to those of Bos tuurus breeds of cattle. Also, Bertrand and Benyshek (1987) analyzed Brangus birth and weaning weight records and found parameter
2362
PARAMETER ESTIMATES FOR ZEBU-DERIVATIVE CATTLE
estimates near the upper end of field data estimates. An important breed characteristic of Brahman and Brahmanderivative cattle is the shape and size of the navel or sheath area. It is undesirable to have too little or too much skin in these areas; excess skin poses possible reproductive and health problems, and insufficient skin represents a lack of breed character. There are few studies in the literature that address this problem or attempt to determine the genetic makeup of these traits. Therefore, this study was conducted to determine 1) heritability and genetic and environmental correlation estimates for birth weight, weaning weight, and postweaning gain traits for Beefmaster, Brahman, Brangus, and Santa Gertrudis cattle using breed field data information and 2) heritability estimates for Brangus weaning sheath and navel scores. Materlals and Methods
Data Description. All information used in subsequent multiple-trait variancecovariance analyses came from NCE data pools. Breeds analyzed were Beefmaster, Brahman, Brangus, and Santa Gertrudis. Adjusted birth weight (BTHW), 205-d weight (Ww), and adjusted 365-d weight (YWT) traits were provided in each data set along with birthdate of the calf, weaning and yearling weigh dates, herd, sex, calf management information, and pedigree information. Each breed used Beef Improvement Federation (BIF) Guideline (1986) equations and age of dam correctiop factors to adjust birth weight and weaning weight records. Yearling weight records were also
2363
adjusted using the suggested 365d weight equation. Adjusted postweaning gain (GAIN) was computed by subtracting WW from YWT. Weaning contemporary groups (WCG) were defied as herd, sex, weaning management code (creep vs no creep), and weaning weigh date for each breed. Adjusted postweaning gain contemporary groups were subsets of WCG and also included yearling management code and yearling weigh date. These definitions were the same as those used in each breed’s NCE analyses. Birth weight records were assigned to the animals’ WCG defiition. This was done to simplify the variance component analyses because it was assumed that both traits analyzed were either defined by the same contemporary group or the second trait was a subset of the first trait. For each breed analyzed in this study, there were many more weaning weight records available for analysis than birth weight records. Thus, by assigning the birth weight record to the WCG definition, all weaning weight records were used in the multiple-trait model variance component procedures. Subsets of the data were formed consisting of BTHW and WW, BTHW and GAIN,or WW and GAIN information for each breed. These data sets were then used in multiple-trait variancecovariance analyses to estimate vatiance and covariance parameters for each breed separately. In addition to the above information, the International Brangus Breeders Association (IBBA) provided weaning sheath and navel scores. Sheath and navel scores were obtained by certified IBBA inspectors at weaning age. These scores ranged from 1 to 5, with a score
TABLE 1. SHEATH AND NAVEL SCORE DESCRIPTIONS IN BRANGUS CATIZE
Description Score
Sheath
Navel
1 Excessively clean
Tight-hided overall. Displays a lack of breed character. The sheath is clean and free from excess leather. The prepucial opening is held at a 45’ angle to the body wall. Leather in navel area is slightly in excess of desired amounts.The prepucial opening is at a 90’ angle to the body wall. Excessive loose leather in navel area. Sheath is slightly pendulous,Prepucial opening is at a 90’angle to the body wall. Sheath is pendulous. Leather in navel area is greatly in excess of desired amounts. Prepucial openingmay be large.
Tight-hidedoverall.Displays a lack of breed character. The navel is clean and free ftom excess leather.
2 optimum 3 Acceptable
4Marginal 5 Unacceplable
Leather in tfie navel is slightly in excess of desired amounts. Excessive loose leather in navel area. Leather in navel is greatly in excess of desired amounts.
2364
KRIESE ET AL.
of 2 being optimum. Table 1 describes the type of sheath or navel that fits each score as defined by the IBBA. Although the scores are considered subjective, they are partially objective because they measure how far the sheath or navel is from the underline of the animal. Bulls and heifers were analyzed separately. Weaning contemporary groups were formed in the manner previously defined. The data sets were analyzed using a single-trait variance covariance analysis to estimate variance and covariance parameters for each trait. Heritabilities and Correlations. Variance and covariances were estimated using multiple-trait models and equating quadratics
composed of mixed-model solutions and righthand sides to their pseudo-expectations. This procedure is a multiple-trait extension of singletrait methods proposed by Schaeffer (1986) and the multiple-trait sire model given by Bertrand and Kriese (1990). Models employed were either a sire-maternal grandsire (MGS) for both traits or a sire-MGS model for the first trait and a sire model for the second trait. An important note is that the environmental covariance between the two traits was considered nonzero. The sire-MGS models were as follows:
where y1 and y2 were vectors of progeny the progeny vector. All sires and maternal records for the first trait and second trait, grandsires were represented in both solution respectively; c1 and c2 were fied contem- vectors across both traits regardless of whether porary group effects for the two traits; and SI they had a progeny or grandprogeny for that and s2, msl and ms2, and el and e2 were sire, trait. This method required that the contemmaternal grandsire, and random residual ef- porary groups be absorbed into the random fects, respectively, for the two traits. The X effects. In matrix notation, the multiple-trait and Z matrices were the incidence matrices of equations after absorption were as follows: 1’s and 0’s relating their specific effect back to
2365
PARAMETER ESTIMATES FOR ZEBU-DERIVATIVE CATTLE
Bertrand and Kriese (1990) described a procedure for absorbing contemporary groups in cases involving two traits, in which there are no missing records for the first trait, there may be missing records for the second trait, and the contemporary groups for the second trait are subsets of the contemporary groups for the first trait. The procedure involves absorbing the contemporary groups separately for each trait, multiplying the absorbed matrices and vectors by the proper values h m the inverse residual variance-covariance matrix, and then combining the appropriate matrices and vectors. The hierarchical contemporary group structure between the two traits allows easier absorption of these effects in multiple-trait models in which the residual covariance between the traits is not zero, as was the case in the analyses presented in this study. For the analyses that contained only a sire effect for the second trait, the model was similar except that the maternal grandsire effect was eliminated for the second trait. Pseudoexpectations of quadratic forms and expectations of variances and covariances for each model were described by Kriese et al. (1991). The dataset composed of WW and BTHW progeny m r d s was analyzed using a sire-MGS model for both traits. The two datasets of WW and GAIN and BTHW and GAIN were analyzed using a sireMGS model for the first trait and a sire model for the second trait. Different models were employed to aid in convergence of variance component solutions. It has been our expenence with these multiple-trait models that if a trait is not maternally influenced, it becomes very difficult to estimate the various covariances associated with the model and the solutions will not converge and most probably will diverge. The expectations for the variance and covariance components using a sire-MGS model for both traits were:
No2,,)
= 1 / 1 6 0 ~ ~+, 1
/
k
~
~
+ 1/M2Mz, E(02el) = 11/1602~, + 3/kA1M1 + 3/4O2M1+ 02E1, E(Oe1e-J = 7 1 1 6 0 + ~ ~~ /~~~o A ~ M ~ + 5 1 8 0 +~~ /~ &~M ~~ M ~ + OE1E29 E(02,) = 11/1602A2 + 3/kA2M2 + 3/M2M2 02%. For the models that contained only a sire effect for the second trait, the expectations were similar except that terms containing ms2 and M2 were eliminated and resulted in the following changes in the residual expectations:
Brangus sheath and navel scores were analyzed using a singlstrait pseudo-expectation method (Schaeffer, 1986) and a sire-MGS model. The model was as follows: y = Xc
+
Zls
+
Zzms
+
e,
where y was a vector of observed weaning sheath or navel scores and c, s, ms,and e were vectors of fixed contemporary group effects and random sire, maternal grandsire, and residual error effects, respectively. X, Z1, and Z, were incidence matrices of 1's and 0's corresponding to their specific effect. The expectations for the variance and covariance components under this model were as follows:
Results and Discussion
Growth Traits. The number of progeny records, Contemporary groups, sires, and MGS from each breed's analysis are presented in Table 2. There was a large range of progeny records available for analyses of the different breeds. Tables 3 and 4 contain variance and
~
~
2366
ICRIESE ET AL.
TABLE 2. NUMBER OF RECORDS, CO"0RARY GROWS, SIRES, AND MATERNAL GRANDSIRES (MGS) FOR BRAHMAN AND BRAHMAN-DERIVATIVE GROWTH W ANALYSES Traita
contemponuy groups
sires
MGS
2,388 7,211 1.576
392 1,278 313
361 1.040 364
554 1,265
5,459 12,559 3,565
368 950 376
363 836 238
935 1,513
32,215 58,932 16,456
3,816 6.479 1.791
2,141 3,387 1,029
3,335 4,424
10,768 23,180 2,868
948 2.649 415
797 1,857 520
1,749 3,442
ReCOrdS
Beefmaster
BTHW
ww
GAIN Brahman BTHW
ww
GAIN Bmgus BTHW
ww
GAIN Smta GemUdis
BTHW
ww
GAIN
%THW = adjusted birth weight;
WW = adjusted 2054 weaning weight; GAIN = adjusted ~ ~ s t w -
covariance components and heritability and genetic correlation estimates for each breed. The variance and covariance components found in these tables were estimated using the outlined expectation equations. For each breed, three two-trait analyses were conducted. There were many more 205-d weight records reported than birth weight or postweaning gain records in each breed. In all four analyses involving birth weight and postweaning gain, there were relatively few post weaning gain records compared with the number of postweaning gain records involved in the analysis with 205d weight. There was a possibility that the sample of data for the birth weight postweaning gain analyses did not adequately represent the populations. However, except for the Beefmaster BTHW-GAIN dataset, heritabilities from the analyses of BTHW-GAIN, BTHW-WW, and WW-GAIN datasets were within .07 across a trait for each dataset. In most cases, the heritability estimates were within .01. Therefore, except for the Beefmaster BTHW-GAIN analysis, the variance component estimates for BTHW, WW, and GAIN were averaged across the three two-trait analyses for each breed, because the heritabilities were similar across the datasets. In the Beefmaster BTHW-GAIN analysis, there were only 327 GAIN records available for analysis, and the additive gain variance was not taken into consideration when reporting results for this paper. The variance and covariance components are provided for completeness, but the
e.
discussion will revolve around the heritability and correlation estimates. For each breed, heritability estimates were lower than estimates summarized by Woldehawariat et al. (1977) for BTHW, WW, and GAIN. This was not unexpected, because field data estimates are generally lower than heritability estimates from designed studies. Birth weight additive heritability (h2d, and maternal heritability (h2h;, estimates of .28 and .12 for the Brangus breed were similar to those reported by Bertrand and Benyshek (1987). Additionally, the h2A estimate (34) for Santa Gertrudis BTHW was similar to the heritability reported by Aaron et al. (1987) using King Ranch records. Brahman BTHW estimates of .37 and .18 for h2A and h2M, reSpeCtiVely, were comparable in magnitude to Brangus and Santa ~ertrudisestimates. However, the h2A estimate (.22) for Beefmaster BTHW was lower in magnitude, and the h2M estimate (.55) was much larger in magnitude than those of other Brahman and Brahman-derivative breeds in this study. This could be a function of the relatively small number of Beefmaster BTHW progeny records. All estimates of both additive and maternal heritabilities for BTHW in Brahman and Brahman-derivative cattle suggest that birth weight can be changed by selection in these breeds if desired. The genetic correlation between the additive and maternal portions of BW (rBM) was negative in all breeds. The magnitude of rgM was large in all
2367
PARAMETER ESTIMATES FOR ZEBU-DERIVATIVE CAlTLE
TABLE 3. GENETIC VARIANCE, COVARIANCE &), -ABILITY, AND CORRELATION ~ l " I b f A ~ FOR BEEFMASTER AND SANTA GWTRUDIS CAlTLE Beefmaster
ww
BTHW Traita
A'
Mbc
BTI-W Abv My Ah
3.80 -3.15 .22 -53
-3.15 9.44) -.53 .55
h'fh
WW
A, MY Ah
Mh
GAIN
A, Ah
symmetric
A
16.10
M
santa Gertrudis GAIN A
-1.51 10.04
-1.62 4.83
-.Cn .72 .29 -.11 131.05 -8.35 -8.35 130.85 .21 -.06 -.06 .21
-.06
4.01
.11 84.03 20.46 .50 .I2 216.09 .56
BTHW
WW
A
M
A
6.49 -3.28 .34 -58
-3.28 4.97 -.58 .26
17.73 -15.98 -7.87 16.23 .49 -.53 -25 .61 200.50 -72.04 -72.04 141.16 .25 -.43 -.43 .18
symmetric
M
GAIN A
3.66 3.62 .I4 .16 -16.52 24.74 -.11 .20 109.62 .26
QTHW = adjusted birth weighc WW = adjusted 205-d weaaing weigh^ GAIN = adjusted postweaning gain. bCo(variances) are on the first two lines of each Wail, heritabilities and correlations are on the next two lines for BTHW and WW. Variances and heritabilities are on the diagonals of each 2 x 2 cell. v = rows of (co)variances h = rows of heritabilities and correlations. GAIN Lines contain additive (co)variance, heritability and correlation estimates. 'A = additive; M = maternal
breeds except Brahman (-. 15). Negative rBM maternal values should be monitored when estimates were also reported in Brangus cattle selecting Brahman and Brahmanderivative (Bertrand and Benyshek, 1987), Simmental cattle to ensure a balance between additive and cattle (Burfening et al., 1981; Garrick et al., maternal values. Additionally, the correlation 1989). and Hereford cattle (Kriese et al., between direct and maternal effects, coupled with moderate heritability estimates, for both 1991). h eXamhhlg WW, h2A (21) and h2M (.15) BTHW and WW suggests that direct and estimates in Brangus cattle were very similar maternal components are important in Brahto estimates reported by Bertrand and man and Brahmanderivative cattle. Benyshek (1987). However, the h2A estimate Additive heritability estimates for GAIN of .25 for Santa Gertrudis WW was lower than were variable between breeds. Estimates that reported by Aaron et al. (1987). Brahman ranged from .15 to .56. However, these and Beefmaster h2A and h2M estimates were estimates are in the range of estimates found in similar in magnitude to Brangus and Santa the literature (Woldehawariat et al., 1977; Gertrudis estimates. These additive and mater- Garrick et al., 1989). All GAIN estimates nal heritabilities suggest that genetic progress suggest that cattle can be selected for increased can be made for increased weaning weights in postweaning gains and yearling weights and Brahman and Brahmanderivative cattle. The that genetic progress can be made. genetic correlation between weaning weight A large, positive correlation between direct direct and weaning weight maternal ( r m ) BTHW and direct WW (rBW) was observable was negative and moderate in magnitude in across all breeds. This was in agreement with Brangus and Santa Gertrudis cattle. Moderate, reported rBw correlation estimates using negative r m correlations were also found in Hereford and Brangus cattle (Kriese et al., the literature for Brangus and Simmental cattle 1991) and Simmental cattle (Garrick et al., (Bertrand and Benyshek, 1987; Garrick et al., 1989) and with other reports (Woldehawariat 1989; Kriese et al., 1991). However, the rw et al., 1977). However, Quass et al. (1985) and estimate for Beefmaster cattle was small and Bourdon and Brinks (1986) found only a negative (-.M)whereas , the r m estimate for small, positive genetic correlation between Brahman cattle was small and positive (.15). birth weight and weaning weight in Hereford Quass et al. (1985) reported a r m of -.@ in and I Simmental cattle, respectively. Thus, when Simmental cattle. Although the relationship selecting for increased weaning weights in between weaning weight direct and weaning Brahman and Brahmanderivative cattle, birth weight maternal is not fully resolved, both weights should be monitored or dystocia weaning weight direct and weaning weight problems could arise.
2368
KRIESE ET AL.
TABLE 4. GENE'IIC VARIANCE, COVARJANCE @&,
HERITABILITY, AND CORRELATION ESTIMATES
FOR BRAHMAN AND BRANGUS CATTLE ~~
~
Brahman
ww
BTHW Traita
A '
BTHW Abv My
7.40 -.75 .37 -.15
Ah
&
A -.75 3.50
-.I5 .18
Av M V
Ah Mh
GAIN
A, Ah
symmetric
M
A
17.87 6.25 3.51 1.97 .63 .27 .18 .12 109.70 13.75 13.75 74.45 .23 .15 .15 .16
Brangus
GAIN
BTHW A M
9.31 12.29 .33 .63 64.32 13.73
5.60 -1.85 .28 -.52
ww A
-1.85 2.30 -52
.12
.59 .15
108.04 .31
symmetric
M
12.42 -6.34 -4.07 7.85 .48 -.29 -24 .55 121.75 -24.10 -24.10 87.50 .21 -.23 -.23 .15
GAIN
A 2.54 -1.96 .14 -.17 28.33 -.72 .34
-.01 56.65 .15
%THW = adjusted birth weight; WW = adjusted 205-d weaning weighs GAIN = adjusted postweaning gain. bCo(variauce) are on the fmt two lines of each trait, heritabilities and correlations are on the next two lines for BTHW and WW. Variances and heritabilities are on the diagonals of each 2 x 2 celL v = rows of (co)variances,h = rows of heritabilities and correlations. GAJN lines contain additive (co)variance, haitability and correlation estimates. ' A = additive; M = maternal.
Also, large, positive genetic correlations were present between direct WW and direct GAIN (rwG) in the Beefmaster, Brahman, and Brangus populations. This is in agreement with most previous estimates. However, a negative rwG (-.ll) was present in the Santa Gertrudis population. This seems contradictory to current theory, which indicates that genes controlling weaning weight growth also affect postweaning growth patterns. Neville et al. (1984a,b) conducted a study that included grade Angus, Polled Hereford, and Santa Gertrudis dams mated to purebred sires of the same breeds. The calves from the grade Santa Gertrudis dams mated to the purebred Santa Gertrudis sires weighed significantly more at birth and weaning and had the highest preweaning ADG in the study. However, this advantage did not carry over to the postweaning phase. Additionally, the Santa Gertrudis data were previously analyzed with 2,090 GAIN records and 15,290 WW records. The previous analysis represented the NCE data bank available at the end of the previous year (1988). As in this current analysis, the rwG was negative but larger in magnitude. With this move toward zero for the Santa Gertrudis rwG, it is probable that the rwG is not negative in the Santa Gertrudis breed. The current negative correlation between weaning weight and gain may be caused by a small sample size, coupled with a possible selection bias in the records that were reported. Often not all animals contained in a
weaning contemporary group will have a reported postweaning gain record. This may result in an inaccurate representation of the breed population. The environmental correlation between WW and GAIN was negative for all breeds (Table 5 ) , which is in agreement with the finding of Garrick et al. (1989). This negative correlation may be due to compensatory gain effects. The environmentalcorrelation between BTHW and GAIN varied greatly between breeds. Estimates ranged from -.06 to .18. Using Hereford and Brangus bull records, moderate, positive correlations have been reported between BTHW and GAIN (Kriese et al., 1991). However, a small, positive correlation between BTHW and GAIN was reported in Simmental cattle (Garrick et al., 1989). Interpretations of these environmental correlations should be made with caution because the method used to estimate environmental correlations is not free of selection bias (Bertrand and Kriese, 1990). However, no environmental correlation was substantially different from literature reports. Thus, Brahman and Brahman-derivative cattle should respond to environmental factors similarly to Bos taurus cattle. Sheath and Navel Score. Bull and heifer records were analyzed separately because reproduction is intrinsically linked to the sheath, but a relationship between reproduction and navel shape is currently unclear. Also, previous work showed that reranking of sires
2369
PARAMETER ESTIMATES FOR ZEBU-DERIVATIVE CATTLE TABLE 5. E"MENTAL VARIANCES, COVARIANCES (kg2), AND CORRELATIONS FOR BRAHMAN AND BRAHMAN-DERIVATIVJZ CATTLE Beefmaster Trait"b
BTHWWW
BTHW
6.90 .23
ww
GAIN
.U2
GAIN
11.73 .81 369.50 -61.11 -.25 167.62
Brangus
Brahman
BTHWWW 9.10 .29 -.06
B " W W
GAIN
14.21 -2.74 271.05 4 5 . 0 6 -.17 245.30
GAIN
13.65 25.62 12.13 .34 406.80 -28.03 -.08 313.45 .19
santa Gertrudis
BTHWWW GAIN 10.77 22.40 2.18 .30 518.06 -17.91 .06 -.04 308.65
'Variances are on the diagonal, covariances above the diagonal, and environmental correlations below the diagonal for each breed. bsTHW = adjusted birth weight; WW = adjusted 205-d weaning weight; GAIN = adjusted postweaning gain.
(Spearman correlation = .52) occurred among sheath and navel score expected progeny differences when sires had both male and female progeny (Kriese et al., 1988). In an additional study, the genetic correlation between weaning sheath and navel score was estimated from bulls having both male and female progeny in the same contemporary group (unpublished data). The genetic correlation was .51, which indicates that many of the same genes control both weaning sheath and navel score but also suggests that weanhg sheath and navel scores are not identical traits in males and females. Table 6 provides the number of weaning sheath and navel observations in each score classification. The data seem to follow a normal distribution. There were a total of 19,820 weaning sheath scores analyzed. These records represented 2,187 WCG, 2,933 MGS, and 2,044 sires. The h2A estimate for weaning The genetic sheath score was 21; h2M was correlation between the additive and maternal effects was -.39. For the weaning navel score analysis, there were 11,560 records from 1,389 WCG, representing 2,012 MGS and 1,222 sires. The h2A and h2M estimates for weaning navel score were .21 and .07, respectively. Again, the genetic correlation between additive and maternal effects was large and negative (-S3). Franke and Burns (1985) measured Brahman bull and heifer weaning sheath area and found a heritability estimate of .45. Our study indicates that both weaning sheath and navel score are under a moderate amount of additive genetic control and that improvement can be obtained for the sue and shape of the sheath or navel. However, there does not seem to be much maternal control of these traits. Exactly what the large, negative correlation between the direct and maternal effects indicates for both traits is unclear. It could indicate
.w.
that the uterus influences the shape and size of the calf's sheath and navel areas during gestation. Implications
This study was conducted to characterize four breeds of Brahman and Brahman-derivative cattle in terms of genetic and environmental parameters for growth traits. There do not seem to be any genetic parameters contradictory to traditional Bos taurus estimates, although the maternal portion of variation may play a more important role in Brahman and Brahman-derivative cattle. The large, negative genetic correlations for direct growth and maternal portions of variation between and within traits must be considered when selecting within these cattle. Selection for extreme growth or maternal ability could be detrimental for the other trait if it is not monitored because of the antagonistic correlation present between direct and maternal portions of variation. Environmental correlations were of the same direction and magnitude as other reported estimates, indicating that Brahman and Brahmanderivative cattle re-
TABLE 6. NUMBER OF BRANGUS WEANING SHEATH AND NAVEL OBSERVATIONS BY SCORE CLASSIFICATION
Scorea 1
2 3 4 5
Weaning sheath observations
navel observations
1,378 9,819 7.114 1,369 140
1,118 6.564 3229 591 58
weaning
aDescription of scores is given in Table 1.
2370
KRIESE ET AL.
spond to environmental factors similarly to Bus taurus cattle. Also, Brangus sheath and navel scores may be improved through selection. Literature Cited
Aaron, D. K.,F. A. Thrift and N. R. Parish. 1987. Genetic parameter estimates for preweaning growth traits in Santa Gertrudis cattle. J. Anim. Sci. 651495. Beef Improvement Federation. 1986. BIF Guidelines for Uniform Beef Improvement Programs. (5th Ed.). Raleigh, NC. Bertrand, J. K. and L. L. Benyshek. 1987. Variance and covariance estimates for maternally influenced beef growth traits. J. Anim. Sci. 6(:728. Bertrand, J. K.and L. A. Kriese. 1990. Two methods for parameter estimation using multipletrait models and beef cattle field data. J. Anim. Sci. 68:2310. Bourdon, J. M. and J. S. Brinks.1986.Scrotal circumference in yearling Hereford bulls: adjustment factors, heritabilities and genetic, environmental and phenotypic relationships with growth traits. J. Anim.Sci. 62:958. Burfening, P. J., D. D. Kress and R. L. Friedrich. 1981. Calving ease and growth rate of Simmental-sired calves. III. Direct and maternal effects. I. Anim. Sci. 53:1210. Franke, D. E. and W. C. Bums. 1985. Sheath area in Brahman and grade Brahman calves and its association
with preweaning growth traits. J. Anim. Sci. 61:398. Ganick, D. J., E. J. Pollak, RL. Quass and L. D. Van Vleck. 1989. Variance heterogeneity in direct and maternal weight traits by sex and percent purebred for Simmental-sired calves. J. Anim. Sci. 67:2515. Kxiese, L., K.Bertrand and L. Benyshek 1988.Heritability estimates for Brangus weaning sheath and navel scores. Anim. Sci. Div. 1987.Beef Cattle Rep. p 30. Special h b l . No. 52. Univ. of Georgia, Athens. Kriese, L.A., J. K.Bertrand and L. L. Benyshek 1991.Age adjustment factors, heritabiMes and genetic correlations for scrotal circumference and related growth traits in Hereford and Brangus bulls. J. Anim. Sci. 69: 478. Neville, W. E.,Jr., B. G. Mullinix, Jr. and W. C. McCormick. 1984a. Grading and rotational crossbreeding of beef cattle. II.Calf performance to weaning. I. Anim. Sci. 58:38. Neville, W. E.,Jr., B. G. Mullinix, Jr. and W. C. McCormick. 1984b. Grading and rotational crossbreeding of beef cattle. III. Postweaning and carcass traits of steers. J. Anim. Sci. 58~47. Quass, R.L.,M. A. Elzo and E.J. Pollak. 1985.Analysis of Simmental daw estimation of direct and maternal genetic (c0)variances.J. Anim. Sci. 61(Suppl. 1):221. (Abstr.) Schaeffer, L. R 1986. Pseudo expectation approach to variance component estimation. J. Dairy Sci. 69:2884. Woldehawariat, G.,M.A. Talamantes,R. R. Petty, Jr. and T. C. Cartwright. 1977. A summary of genetic and environmental statistics for growth and conformation characters of beef cattle. Texas Agric. Exp. Sta. Dept Anim. Sci. Tech. Rep. No. 103.