- Email: [email protected]
Genetic and Environmental Influences on Milk and Milk Component Production1
Genetic and Environmental Influences on Milk and Milk Component Production 1 G. L. H A R G R O V E , D. A, M B A H , 2 and J. L. R O S E N B E R G E R 3 Department of Dairy and Animal Science The Pennsylvania State University University Park 16802
numerous estimates for milk yield and fat traits, corresponding information on protein traits is relatively inadequate. Bereskin and Lush (4) showed that common environmental correlation (C 2) among paternal sisters led to lower than expected correlations of consecutive sire proofs. The USDA (United States Department of Agriculture) has included C 2 in the model for sire evaluation (12). Influence of common environment on milk yield (.08 to .14) is documented (1, 12, 14, 16). However, corresponding information for fat yield (1) is inadequate. No estimates of C 2 are available for protein yield and percentages of protein and fat. Objectives of our study were to provide more information on heritability and genetic interrelationships of protein, fat, and milk traits and to estimate environmental correlations among paternal sisters for protein and fat traits in particular.
ABSTRACT
Estimates of genetic parameters and environmental correlations were from deviations from modified contemporary average of standardized records of first lactations of Holsteins in Pennsylvania. Records were extended to 305 days and adjusted for month of calving prior to calculation of deviations. Heritabilities from intrasire correlations among paternal sisters were .23, .26, . 22,. 71, and .64 for yields of milk, fat, and protein, and percentages of fat and protein. Genetic correlations were positive and generally high between yields and between percentages. Correlations of milk yield with percentages were strongly negative. Environmental correlations among paternal sisters in the same herd were .08,. 11,. 10, - . 0 4 , and - . 0 3 for yields of milk, fat, protein, and percentages of fat and protein. These statistics will be useful to plan optimum selection programs and to improve sire evaluation procedures, particularly for protein content and yield.
MATERIALS AND METHODS
INTRODUCTION
Recent consumer demand for dietary articles low in fat but high in protein has stimulated considerable interest in selecting dairy cattle for protein production. Selecting for protein content constitutes a change in breeding goals from current practice of selecting primarily for milk yield and secondarily for fat content. Prediction of selection results from new breeding goals are from estimates of genetic parameters important in genetic change. Although there are
Received February 25, 1980. 1journal Paper No. 5915, The Pennsylvania Agricultural Experiment Station. 21nstitute of Zootechnical Research of WAKWA, B.P. 65, Ngaoundere, Cameroon, West Africa. SDepartment of Statistics, Pennsylvania State University. 4Foss Electric Company, Hillerod, Denmark. 1981
J Dairy Sci 6 4 : 1 5 9 3 - 1 5 9 7
The data were deviations from modified contemporary average of standardized records of first lactations of Holsteins from 119 herds in Pennsylvania. Less than 1% of the records were owner-sampler, but several herds were on the alternate a.m.-p.m, plan of testing. All cows were milked twice a day. Analyses for fat and protein contents were at the Pennsylvania State University Central Milk Testing Laboratory where the testing is automated and has used Combi-unit 4 and Milko-Scan 3004 machines. Excluded were records of less than 90 days in milk, with unidentified sire or cow, or initiated by abortion. The lowest accepted age at calving was 19 too. Incomplete records of 90 or more days were extended to 305 days and corrected for age and month to a mature equivalent (ME) by factors developed by Cooper (6). Records of cows that went dry with less than 305 days in milk were not extended. A contemporary was defined as another individual <36 mo of age at calving that calved during the season from 2 mo before through 2
1593
1594
ItARGROVE ET AL.
mo after the month of calving of the cow. Paternal sisters were excluded from contemporary and noncontemporary groups. Noncontemporaries were older cows calving in the described season. The average of noncontemporaries calving in the same season was added to the contemporary total as one record and averaged. This average was adjusted for season average. A daughter having neither contemporaries nor noncontemporaries was excluded from the data. The adjusted average then was subtracted from the daughter record (each of the five traits individually) and credited with one-tenth the inferiority or superiority of the adjusted average from the calving-year average. This adjustment was an approximation of the adjustment used by USDA (7). From the resulting data set, only cows in first lactation calving from October, 1972, through September, 1978, were in analyses. To assure a full season for calculation of deviations, we included herds that had contemporaries calving 2 mo before and 2 mo after accepted records. Each sire had at least two daughters. This produced data set (II) in Table 1. From this set additional editing was applied to provide data sets (I) and (1II).
Genetic and Phenotypic Parameters
The following criteria were stipulated for creation (from data set II) of a data set for estimation of genetic and phenotypic parameters: a) A sire must have daughters in at least two herds, b) Each herd must have at least two sires, c) Only one daughter per sire per herd was accepted. When more than one daughter per herd was available, the accepted daughter was the one with the most contemporary sires or the most contemporaries if two had the same number of contemporary sires. In cases of a tie of both these criteria, low birth day (1 to 31) was the randomization criterion for daughter selection. These criteria favor selection of data from seasons with large numbers of sires. This produced data set (I) in Table 1. The mathematical model for each deviation record (D) included sire(s) and cow(c), both random with means zero and corresponding variances, o] and Oc 2. Thus, Dij= // + si + cij. Method 1 of Henderson (10) was used to estimate variance and covariance components. Heritability (h 2) estimates were four times the intrasire correlation (t) of paternal sisters (8). Journal of Dairy Science Vol. 64, No. 7, 1981
Standard errors were estimated as in (13). Genetic correlations were from sire components of variances and covariances (2), and their standard errors were approximated as in Falconer (8). Phenotypic correlations were estimated as in Becker (2) where phenotypic variances and covariances were sums of components. Common Environmental Correlation
The correlation among paternal sisters from environment common to them (C 2) was estimated as the difference between two intraclass correlations (1, 14). One additional data set was created from data set II that consisted of daughters of a sire in one herd only. Selection criteria were: a) A sire must have at least two daughters in one herd. b) Herds with higher numbers of daughters of a sire were selected, c) Each herd must have at least two sires. (Steps b and c were applied in an iterative fashion with increasing standards until only herds with maximum information were retained.) d) If daughters were distributed evenly in more than one herd, the lower numbered county-herd was accepted. This data set ( l i d is in Table 1. Intraclass correlations from I, II, and III were t l , t2, and t 3. Correlation among paternal sisters is expected from additive genetic causes (I) and common environmental (and nonadditive genetic) causes (II and III) according to whether paternal sisters were located in different (I) or the same (Ill) herds. If C 2 exists for a trait, the relation among the estimates of intraclass correlations is expected to be tl
zero or positive with an upper limit of o t i from
GENETIC AND ENVIRONMENTAL INFLUENCES TABLE 1. Distribution of observations by data sets used to estimate genetic parameters. Dataset
Sires
ka
Records
I II III
272 460 361
6.428 12.434 6.089
1758 5782 2207
ak, Coefficient of the variance component for sires in the expected mean squares formula.
the genetic component common to tl and ta. Therefore, the range for the standard error of (~2 W a S :
I
(~t3- ~t, ) .s if ~tz tl
= ~tl = (lower limit)
0~2
(~t3 + i~tl ) .s if 0ta tl = 0 (upper limit) The t2 is expected to be greater than tl by a fraction of C 2. This fraction should depend on how similar is the distribution of daughters to the distribution of daughters in I or III. Errors associated with estimates of intraclass correlations influence ~2. Thus a positive error in tl can deflate ~2. A negative error in tl leads • A to an inflated C 2. RESULTS AND DISCUSSION
1595
are intermediate to those from paternal sister correlations from deviation data by Thompson and Loganathan (15), Gaunt et al. (9), Wilcox et al. (17), and estimates from first lactations by Norman et al. (11). The estimate for protein yield is higher than in (9, 17). The estimate for fat percent is higher than estimates from other deviation data (9, 11, 15, 17, 18). However, standard error indicates it may be within the general range (.45 to .62) in these reports• The estimate for protein percent is higher than in (9, 17). Genetic Correlations
Genetic correlations between milk yield and components are in Table 2. Although correlations between yields agreed with (5, 9, 17), those involving fat yield were in the low range. Compared to correlations from Butcher et al. (5), antagonisms between milk yield and percentages of fat and protein are similar but slightly more negative than (17). The correlation between percentages of fat and protein is at the upper range of estimates (.55 to .74) by (5, 9, 17). Correlation between protein yield and protein percent (.09) was slightly lower than in (.14 to .37). There has been a consistent tendency for correlations between yield and percentage to be lower for protein than for fat. The remaining correlations agreed well with (5, 9, 17).
Heritability
Phenotypic Correlations
Heritabilities for milk yield and constituents are in Table 2. Estimates for milk and fat yields
Phenotypic correlations in Table 2 agree with (4, 15).
TABLE 2. Heritabilities, phenotypic, and genetic correlations of milk yield and constituents,a Yield Trait Yields Milk Fat Protein Percentages Fat Protein
Milk
Fat
Percent Protein
(SE) .23 (.07) .40 (. 18) .83 (.07)
(SE) .78 .26 (.08) .69 (.11)
(SE) .94 .84 .22 (.07)
--.56 (.10) -.48 (.11)
.53 (.10) .34 (.13)
-.13 (.14) .09 (.15)
Fat
Protein (SE)
-.34 .31 -.15 .71 (.09) .77 (.04)
(SE) -.38 -.02 -.04 .58 .64 (.09)
aphenotypic correlations ~ .06 or ~< -.06 differ from zero, P~<.05. Genetic correlations below diagonal. Phenotypic correlations above diagonal. Heritabilities on diagonal (underlined). Journal of Dairy Science Vol. 64, No. 7, 1981
1596
HARGROVE ET AL.
TABLE 3. Intrasire correlations (t i) and common environmental correlations (~2) among paternal sisters, a Intraclass correlation Trait
Yields Milk Fat Protein Percentages Fat Protein
C~
tI
t 2
t3
.06 .06 .06
.08 .11 .09
.14 .18 .15
.08 .11 .10
.18 .16
.13 .13
.13 .14
--.04 -.03
t 3 -- t 1
(SE) (.01-.03) (.01--.03) (.01--.03) (b-.03) (b-.03)
aLower limit of standard errors of ~2 assumes at3 tl = a 2t~ and upper limit assumes at3tl = 0. Lower limit (b) not calculated when negative variance of ~2 was calculated.
Common Environmental Correlations
Environmental correlations among paternal sisters in t h e same h e r d are in T a b l e 3. A l t h o u g h ~2 f o r m i l k yield was in t h e l o w e r r a n g e o f estimates, .08 to .14 b y (1, 12, 14, 16), t h e e s t i m a t e f o r f a t yield is similar t o t h a t o f A r o r a a n d F r e e m a n (1). E s t i m a t e o f C 2 for p r o t e i n yield suggests similar c o m m o n e n v i r o n m e n t a l i n f l u e n c e s on this t r a i t as o n milk a n d f a t yields. T h e negative e s t i m a t e s f o r t h e p e r c e n t a g e traits are n o t e x p l a i n e d . D a u g h t e r s o f a sire in t h e c h o s e n h e r d c o u l d c o m e f r o m d i f f e r e n t seasons or years c o v e r i n g t h e p e r i o d w h e n t h e p r o j e c t was in progress. If t h e seasons a n d / o r years differed, this m i g h t r e d u c e similarity a m o n g p a t e r n a l sisters f r o m similar t r e a t m e n t . This w o u l d prevail if f i n d i n g s by Bereskin a n d F r e e m a n (3), t h a t d e v i a t i o n records still c o n t a i n residual h e r d - y e a r - s e a s o n effects, a p p l y to o u r data. If, h o w e v e r , t h e s e residual e f f e c t s are n o t i m p o r t a n t (14), o u r C 2 may be the maximum expected. E s t i m a t e s o f h 2 a n d C 2 are associated w i t h t l . Since e s t i m a t e s were: la 2 = 4 t l , ~2 = t3 t 1, e r r o r s in a given e s t i m a t e o f t i bias associated e s t i m a t e s o f h 2 a n d C 2. T h u s , positive e r r o r s in t l w o u l d lead to i n f l a t e d ~a2 a n d d e f l a t e d ~2. O u r e s t i m a t e s of h 2 for p e r c e n t a g e p r o t e i n a n d f a t are a m o n g the h i g h e s t in t h e l i t e r a t u r e . If t h e positive error in t l was large e n o u g h , a negative ~2 m i g h t o c c u r as for f a t a n d p r o t e i n ( T a b l e 3). Hence, t h e p a t t e r n o f t's a n d asso^2 ciated C for fat a n d p r o t e i n p e r c e n t s ( T a b l e 3) c o u l d occur. F u r t h e r m o r e , t2 n e v e r e x c e e d s t3, supporting the expected impact of C 2 on Journal of Dairy Science Vol. 64, No. 7, 1981
e s t i m a t e s o f t. O f t e n , d a t a sets used t o e s t i m a t e h e r i t a b i l i t i e s are similar t o d a t a set II. T h e t2 in T a b l e 3 indicate that estimates of heritabilities by intraclass c o r r e l a t i o n s of daughters from such d a t a sets m a y be i n f l a t e d b y a f r a c t i o n o f C 2. Hence, m a n y p u b l i s h e d h e r i t a b i l i t i e s (9, 11, 17) o b t a i n e d b y intraclass c o r r e l a t i o n m a y b e inflated.
REFERENCES
1 Arora, K. K., and A. E. Freeman. 1971. Environmental correlation between paternal half-sisters for milk and milk fat production. J. Dairy Sci. 54:880. 2 Becker, W. A. 1975. Manual of quantitative genetics. 3rd ed. Washington State University Press, Pullman. 3 Bereskin, B., and A. E. Freeman. 1965. Genetic and environmental factors in dairy sire evaluation. If. Uses and limitations of deviation records and the role of dams. J. Dairy Sci. 48:352. 4 Bereskin, B., and J. L. Lush. 1965. Genetic and environmental factors in dairy sire evaluation. III. Influence of environmental and other extraneous correlations among the daughters. J. Dairy Sci. 48:356. 5 Butcher, K. R., F. D. Sargent, and J. E. Legates. 1967. Estimates of genetic parameters for milk constituents and yields. J. Dairy Sci. 50:185. 6 Cooper, J. B. 1980. Age and month-of-calving adjustments and extension factors for milk, fat, and protein yields. M.S. thesis, The Pennsylvania State University, University Park. 7 Dickinson, F. N., R. L. Powell, and H. D. Norman. 1976. An introduction to the USDA-DHIA modified contemporary comparison. In The USDA-DHIA Modified Contemporary Comparison. USDA Prod. Res. Rep. No. 165. 8 Falconer, D. S. 1960. Introduction to quantitative genetics. The Ronald Press Co., New York, NV.
GENETIC A N D E N V I R O N M E N T A L INFLUENCES
9 Gaunt, S. N., C. J. Wilcox, B. R. Farthing, and N. R. T h o m p s o n . 1968. Genetic interrelationships o f Holstein milk composition. J. Dairy Sci. 51:1396. 10 Henderson, C. R. 1953. Estimation of variance and covariance c o m p o n e n t s . Biometrics 9:226. 11 Norman, H. D., B. T. McDaniel, and F. N. Dickinson. 1972. Conflicts between heritability estimates of mature equivalent and herd-mate deviation milk and fat. J. Dairy Sci. 55:507. 12 Plowman, R. D., and B. T. McDaniel. 1968. Changes in USDA sire s u m m a r y procedures. J. Dairy Sci. 51 : 306. 13 Swiger, L. A., W. R. Harvey, D. O. Everson, and K. E. Gregory. 1964. The variance of intraclass correlation involving groups with one observation. Biometrics 20:818.
1597
14 T h o m p s o n , G. M., and A. E. Freeman. 1970. Environmental correlations in pedigree estimates of breeding value. J. Dairy Sci. 53 : 1259. 15 T h o m p s o n , N. R., and S. L. Loganathan. 1968. Composition of cow's milk. I1. Genetic influences. J. Dairy Sci. 51:1933. 16 Van Vleck, L. D. 1966. Paternal half-sib correlations between pairs in the same and different herds. J. Dairy Sci. 49:195. 17 Wilcox, C. J., S. N. Gaunt, and B. R. Farthing. 1971. Genetic interrelationships of milk composition and yield. South. Coop. Ser. Bull. 155. 18 Wunder, W. W., and L. D. McGilliard. 1964. Heritabilities and genetic correlations for comp o n e n t s of milk in Holsteins and Guernseys. J. Dairy Sci. 47:700.
Journal of Dairy Science Vol. 64, No. 7, 1981
numerous estimates for milk yield and fat traits, corresponding information on protein traits is relatively inadequate. Bereskin and Lush (4) showed that common environmental correlation (C 2) among paternal sisters led to lower than expected correlations of consecutive sire proofs. The USDA (United States Department of Agriculture) has included C 2 in the model for sire evaluation (12). Influence of common environment on milk yield (.08 to .14) is documented (1, 12, 14, 16). However, corresponding information for fat yield (1) is inadequate. No estimates of C 2 are available for protein yield and percentages of protein and fat. Objectives of our study were to provide more information on heritability and genetic interrelationships of protein, fat, and milk traits and to estimate environmental correlations among paternal sisters for protein and fat traits in particular.
ABSTRACT
Estimates of genetic parameters and environmental correlations were from deviations from modified contemporary average of standardized records of first lactations of Holsteins in Pennsylvania. Records were extended to 305 days and adjusted for month of calving prior to calculation of deviations. Heritabilities from intrasire correlations among paternal sisters were .23, .26, . 22,. 71, and .64 for yields of milk, fat, and protein, and percentages of fat and protein. Genetic correlations were positive and generally high between yields and between percentages. Correlations of milk yield with percentages were strongly negative. Environmental correlations among paternal sisters in the same herd were .08,. 11,. 10, - . 0 4 , and - . 0 3 for yields of milk, fat, protein, and percentages of fat and protein. These statistics will be useful to plan optimum selection programs and to improve sire evaluation procedures, particularly for protein content and yield.
MATERIALS AND METHODS
INTRODUCTION
Recent consumer demand for dietary articles low in fat but high in protein has stimulated considerable interest in selecting dairy cattle for protein production. Selecting for protein content constitutes a change in breeding goals from current practice of selecting primarily for milk yield and secondarily for fat content. Prediction of selection results from new breeding goals are from estimates of genetic parameters important in genetic change. Although there are
Received February 25, 1980. 1journal Paper No. 5915, The Pennsylvania Agricultural Experiment Station. 21nstitute of Zootechnical Research of WAKWA, B.P. 65, Ngaoundere, Cameroon, West Africa. SDepartment of Statistics, Pennsylvania State University. 4Foss Electric Company, Hillerod, Denmark. 1981
J Dairy Sci 6 4 : 1 5 9 3 - 1 5 9 7
The data were deviations from modified contemporary average of standardized records of first lactations of Holsteins from 119 herds in Pennsylvania. Less than 1% of the records were owner-sampler, but several herds were on the alternate a.m.-p.m, plan of testing. All cows were milked twice a day. Analyses for fat and protein contents were at the Pennsylvania State University Central Milk Testing Laboratory where the testing is automated and has used Combi-unit 4 and Milko-Scan 3004 machines. Excluded were records of less than 90 days in milk, with unidentified sire or cow, or initiated by abortion. The lowest accepted age at calving was 19 too. Incomplete records of 90 or more days were extended to 305 days and corrected for age and month to a mature equivalent (ME) by factors developed by Cooper (6). Records of cows that went dry with less than 305 days in milk were not extended. A contemporary was defined as another individual <36 mo of age at calving that calved during the season from 2 mo before through 2
1593
1594
ItARGROVE ET AL.
mo after the month of calving of the cow. Paternal sisters were excluded from contemporary and noncontemporary groups. Noncontemporaries were older cows calving in the described season. The average of noncontemporaries calving in the same season was added to the contemporary total as one record and averaged. This average was adjusted for season average. A daughter having neither contemporaries nor noncontemporaries was excluded from the data. The adjusted average then was subtracted from the daughter record (each of the five traits individually) and credited with one-tenth the inferiority or superiority of the adjusted average from the calving-year average. This adjustment was an approximation of the adjustment used by USDA (7). From the resulting data set, only cows in first lactation calving from October, 1972, through September, 1978, were in analyses. To assure a full season for calculation of deviations, we included herds that had contemporaries calving 2 mo before and 2 mo after accepted records. Each sire had at least two daughters. This produced data set (II) in Table 1. From this set additional editing was applied to provide data sets (I) and (1II).
Genetic and Phenotypic Parameters
The following criteria were stipulated for creation (from data set II) of a data set for estimation of genetic and phenotypic parameters: a) A sire must have daughters in at least two herds, b) Each herd must have at least two sires, c) Only one daughter per sire per herd was accepted. When more than one daughter per herd was available, the accepted daughter was the one with the most contemporary sires or the most contemporaries if two had the same number of contemporary sires. In cases of a tie of both these criteria, low birth day (1 to 31) was the randomization criterion for daughter selection. These criteria favor selection of data from seasons with large numbers of sires. This produced data set (I) in Table 1. The mathematical model for each deviation record (D) included sire(s) and cow(c), both random with means zero and corresponding variances, o] and Oc 2. Thus, Dij= // + si + cij. Method 1 of Henderson (10) was used to estimate variance and covariance components. Heritability (h 2) estimates were four times the intrasire correlation (t) of paternal sisters (8). Journal of Dairy Science Vol. 64, No. 7, 1981
Standard errors were estimated as in (13). Genetic correlations were from sire components of variances and covariances (2), and their standard errors were approximated as in Falconer (8). Phenotypic correlations were estimated as in Becker (2) where phenotypic variances and covariances were sums of components. Common Environmental Correlation
The correlation among paternal sisters from environment common to them (C 2) was estimated as the difference between two intraclass correlations (1, 14). One additional data set was created from data set II that consisted of daughters of a sire in one herd only. Selection criteria were: a) A sire must have at least two daughters in one herd. b) Herds with higher numbers of daughters of a sire were selected, c) Each herd must have at least two sires. (Steps b and c were applied in an iterative fashion with increasing standards until only herds with maximum information were retained.) d) If daughters were distributed evenly in more than one herd, the lower numbered county-herd was accepted. This data set ( l i d is in Table 1. Intraclass correlations from I, II, and III were t l , t2, and t 3. Correlation among paternal sisters is expected from additive genetic causes (I) and common environmental (and nonadditive genetic) causes (II and III) according to whether paternal sisters were located in different (I) or the same (Ill) herds. If C 2 exists for a trait, the relation among the estimates of intraclass correlations is expected to be tl
zero or positive with an upper limit of o t i from
GENETIC AND ENVIRONMENTAL INFLUENCES TABLE 1. Distribution of observations by data sets used to estimate genetic parameters. Dataset
Sires
ka
Records
I II III
272 460 361
6.428 12.434 6.089
1758 5782 2207
ak, Coefficient of the variance component for sires in the expected mean squares formula.
the genetic component common to tl and ta. Therefore, the range for the standard error of (~2 W a S :
I
(~t3- ~t, ) .s if ~tz tl
= ~tl = (lower limit)
0~2
(~t3 + i~tl ) .s if 0ta tl = 0 (upper limit) The t2 is expected to be greater than tl by a fraction of C 2. This fraction should depend on how similar is the distribution of daughters to the distribution of daughters in I or III. Errors associated with estimates of intraclass correlations influence ~2. Thus a positive error in tl can deflate ~2. A negative error in tl leads • A to an inflated C 2. RESULTS AND DISCUSSION
1595
are intermediate to those from paternal sister correlations from deviation data by Thompson and Loganathan (15), Gaunt et al. (9), Wilcox et al. (17), and estimates from first lactations by Norman et al. (11). The estimate for protein yield is higher than in (9, 17). The estimate for fat percent is higher than estimates from other deviation data (9, 11, 15, 17, 18). However, standard error indicates it may be within the general range (.45 to .62) in these reports• The estimate for protein percent is higher than in (9, 17). Genetic Correlations
Genetic correlations between milk yield and components are in Table 2. Although correlations between yields agreed with (5, 9, 17), those involving fat yield were in the low range. Compared to correlations from Butcher et al. (5), antagonisms between milk yield and percentages of fat and protein are similar but slightly more negative than (17). The correlation between percentages of fat and protein is at the upper range of estimates (.55 to .74) by (5, 9, 17). Correlation between protein yield and protein percent (.09) was slightly lower than in (.14 to .37). There has been a consistent tendency for correlations between yield and percentage to be lower for protein than for fat. The remaining correlations agreed well with (5, 9, 17).
Heritability
Phenotypic Correlations
Heritabilities for milk yield and constituents are in Table 2. Estimates for milk and fat yields
Phenotypic correlations in Table 2 agree with (4, 15).
TABLE 2. Heritabilities, phenotypic, and genetic correlations of milk yield and constituents,a Yield Trait Yields Milk Fat Protein Percentages Fat Protein
Milk
Fat
Percent Protein
(SE) .23 (.07) .40 (. 18) .83 (.07)
(SE) .78 .26 (.08) .69 (.11)
(SE) .94 .84 .22 (.07)
--.56 (.10) -.48 (.11)
.53 (.10) .34 (.13)
-.13 (.14) .09 (.15)
Fat
Protein (SE)
-.34 .31 -.15 .71 (.09) .77 (.04)
(SE) -.38 -.02 -.04 .58 .64 (.09)
aphenotypic correlations ~ .06 or ~< -.06 differ from zero, P~<.05. Genetic correlations below diagonal. Phenotypic correlations above diagonal. Heritabilities on diagonal (underlined). Journal of Dairy Science Vol. 64, No. 7, 1981
1596
HARGROVE ET AL.
TABLE 3. Intrasire correlations (t i) and common environmental correlations (~2) among paternal sisters, a Intraclass correlation Trait
Yields Milk Fat Protein Percentages Fat Protein
C~
tI
t 2
t3
.06 .06 .06
.08 .11 .09
.14 .18 .15
.08 .11 .10
.18 .16
.13 .13
.13 .14
--.04 -.03
t 3 -- t 1
(SE) (.01-.03) (.01--.03) (.01--.03) (b-.03) (b-.03)
aLower limit of standard errors of ~2 assumes at3 tl = a 2t~ and upper limit assumes at3tl = 0. Lower limit (b) not calculated when negative variance of ~2 was calculated.
Common Environmental Correlations
Environmental correlations among paternal sisters in t h e same h e r d are in T a b l e 3. A l t h o u g h ~2 f o r m i l k yield was in t h e l o w e r r a n g e o f estimates, .08 to .14 b y (1, 12, 14, 16), t h e e s t i m a t e f o r f a t yield is similar t o t h a t o f A r o r a a n d F r e e m a n (1). E s t i m a t e o f C 2 for p r o t e i n yield suggests similar c o m m o n e n v i r o n m e n t a l i n f l u e n c e s on this t r a i t as o n milk a n d f a t yields. T h e negative e s t i m a t e s f o r t h e p e r c e n t a g e traits are n o t e x p l a i n e d . D a u g h t e r s o f a sire in t h e c h o s e n h e r d c o u l d c o m e f r o m d i f f e r e n t seasons or years c o v e r i n g t h e p e r i o d w h e n t h e p r o j e c t was in progress. If t h e seasons a n d / o r years differed, this m i g h t r e d u c e similarity a m o n g p a t e r n a l sisters f r o m similar t r e a t m e n t . This w o u l d prevail if f i n d i n g s by Bereskin a n d F r e e m a n (3), t h a t d e v i a t i o n records still c o n t a i n residual h e r d - y e a r - s e a s o n effects, a p p l y to o u r data. If, h o w e v e r , t h e s e residual e f f e c t s are n o t i m p o r t a n t (14), o u r C 2 may be the maximum expected. E s t i m a t e s o f h 2 a n d C 2 are associated w i t h t l . Since e s t i m a t e s were: la 2 = 4 t l , ~2 = t3 t 1, e r r o r s in a given e s t i m a t e o f t i bias associated e s t i m a t e s o f h 2 a n d C 2. T h u s , positive e r r o r s in t l w o u l d lead to i n f l a t e d ~a2 a n d d e f l a t e d ~2. O u r e s t i m a t e s of h 2 for p e r c e n t a g e p r o t e i n a n d f a t are a m o n g the h i g h e s t in t h e l i t e r a t u r e . If t h e positive error in t l was large e n o u g h , a negative ~2 m i g h t o c c u r as for f a t a n d p r o t e i n ( T a b l e 3). Hence, t h e p a t t e r n o f t's a n d asso^2 ciated C for fat a n d p r o t e i n p e r c e n t s ( T a b l e 3) c o u l d occur. F u r t h e r m o r e , t2 n e v e r e x c e e d s t3, supporting the expected impact of C 2 on Journal of Dairy Science Vol. 64, No. 7, 1981
e s t i m a t e s o f t. O f t e n , d a t a sets used t o e s t i m a t e h e r i t a b i l i t i e s are similar t o d a t a set II. T h e t2 in T a b l e 3 indicate that estimates of heritabilities by intraclass c o r r e l a t i o n s of daughters from such d a t a sets m a y be i n f l a t e d b y a f r a c t i o n o f C 2. Hence, m a n y p u b l i s h e d h e r i t a b i l i t i e s (9, 11, 17) o b t a i n e d b y intraclass c o r r e l a t i o n m a y b e inflated.
REFERENCES
1 Arora, K. K., and A. E. Freeman. 1971. Environmental correlation between paternal half-sisters for milk and milk fat production. J. Dairy Sci. 54:880. 2 Becker, W. A. 1975. Manual of quantitative genetics. 3rd ed. Washington State University Press, Pullman. 3 Bereskin, B., and A. E. Freeman. 1965. Genetic and environmental factors in dairy sire evaluation. If. Uses and limitations of deviation records and the role of dams. J. Dairy Sci. 48:352. 4 Bereskin, B., and J. L. Lush. 1965. Genetic and environmental factors in dairy sire evaluation. III. Influence of environmental and other extraneous correlations among the daughters. J. Dairy Sci. 48:356. 5 Butcher, K. R., F. D. Sargent, and J. E. Legates. 1967. Estimates of genetic parameters for milk constituents and yields. J. Dairy Sci. 50:185. 6 Cooper, J. B. 1980. Age and month-of-calving adjustments and extension factors for milk, fat, and protein yields. M.S. thesis, The Pennsylvania State University, University Park. 7 Dickinson, F. N., R. L. Powell, and H. D. Norman. 1976. An introduction to the USDA-DHIA modified contemporary comparison. In The USDA-DHIA Modified Contemporary Comparison. USDA Prod. Res. Rep. No. 165. 8 Falconer, D. S. 1960. Introduction to quantitative genetics. The Ronald Press Co., New York, NV.
GENETIC A N D E N V I R O N M E N T A L INFLUENCES
9 Gaunt, S. N., C. J. Wilcox, B. R. Farthing, and N. R. T h o m p s o n . 1968. Genetic interrelationships o f Holstein milk composition. J. Dairy Sci. 51:1396. 10 Henderson, C. R. 1953. Estimation of variance and covariance c o m p o n e n t s . Biometrics 9:226. 11 Norman, H. D., B. T. McDaniel, and F. N. Dickinson. 1972. Conflicts between heritability estimates of mature equivalent and herd-mate deviation milk and fat. J. Dairy Sci. 55:507. 12 Plowman, R. D., and B. T. McDaniel. 1968. Changes in USDA sire s u m m a r y procedures. J. Dairy Sci. 51 : 306. 13 Swiger, L. A., W. R. Harvey, D. O. Everson, and K. E. Gregory. 1964. The variance of intraclass correlation involving groups with one observation. Biometrics 20:818.
1597
14 T h o m p s o n , G. M., and A. E. Freeman. 1970. Environmental correlations in pedigree estimates of breeding value. J. Dairy Sci. 53 : 1259. 15 T h o m p s o n , N. R., and S. L. Loganathan. 1968. Composition of cow's milk. I1. Genetic influences. J. Dairy Sci. 51:1933. 16 Van Vleck, L. D. 1966. Paternal half-sib correlations between pairs in the same and different herds. J. Dairy Sci. 49:195. 17 Wilcox, C. J., S. N. Gaunt, and B. R. Farthing. 1971. Genetic interrelationships of milk composition and yield. South. Coop. Ser. Bull. 155. 18 Wunder, W. W., and L. D. McGilliard. 1964. Heritabilities and genetic correlations for comp o n e n t s of milk in Holsteins and Guernseys. J. Dairy Sci. 47:700.
Journal of Dairy Science Vol. 64, No. 7, 1981