BSTRACT
GENETIC PARAMETERS IN

STRAIGHTBRED AND CROSSbRED
BEEF CATTLE POPULATIONS

by Richard John Dunn

The data included records on 375 steers and 362 heifers.
born over a four year period. 1960 to 1963 inclusive, at the
Fort Robinson Beef Cattle Research Station in northwestern
Nebraska. The calves were from 80 cows each from the Angus.
Hereford, and Shorthorn breeds. and were the progeny of 17
Angus, 16 Hereford. and 16 Shorthorn sires. The cows were
randomized to breeding pastures each year so that each sire
was bred to twice as many cows of his own breed as to each of
the other two breeds. Traits studied were birth weight.
weaning score. and adjusted ZOO-day weight (steers and
heifers); adjusted final weight. marbling score, final car-
cass grade. fat thickness, rib-eye area. and actual cutability
(steers); and adjusted 550-day weight (heifers).

The experimental design included run: sires each year
which caused a hierarchal design with sires nested within
breed of sire and year. but cross classified with breed of
dam. Steers and heifers were analyzed separately for all
traits.

Variance components for all traits were obtained from
analysis of variance tables. A least-squares and maximum
likelihood general purpose program was used to compute the

following estimates of genetic parameters and standard errors:

Components of genetic variance and covariance were
calculated for each of the mating types. There
were no differences in the sire components between
the straightbreds and crossbreds, indicating similar
additive genetic variance in the two groups.
Heritability estimates were obtained for the various
traits by the paternal half-sib correlation method.
Separate analyses were published for the pooled
within straightbred, and pooled within crossbred groups.
The analyses yielded heritability estimates from 0.15
to 0.85 with most standard errors between 0.3 and 0.4.
No difference was noted between the crossbreds and
straightbreds. The heritability estimates were
large enough to indicate that mass selection should
produce improvement in the traits.
Genetic and phenotypic correlations for the pooled
within straightbred analyses are presented. Standard
errors. in general, varied from 0.3 to 0.4. The
high genotypic and phenotypic correlations among
traits associated with weight (birth and ZOO-day
weight. final weight. 550-day weight. rib-eye.area.
and actual cutability) were the most important from
an economic standpoint.

Results indicate that mass selection for traits
associated with weight in the purebred population
should give simultaneous improvement in these traits.

Adjusted 200-day weight (steers and heifers),.ad-

Justed final weight (steers), and 550-day weight
(heifers) were recommended as traits to emphasize
in a mass selection program.

Estimates of the correlations between the genetic
ability of a sire to produce the same trait in
straightbred and crossbred prageny. and between his
genetic ability to produce one trait in the straight-
bred pOpulation. and another trait in the crossbred
pOpulation, were high. It was concluded that mass
selection in the purebred populations that make up
a cross would be as effective in improving commer-
cial production in the crossbreds as it would be in
improving commercial production in a breeding pro-
gram where the purebred bulls were bred to commer-

cial cattle of their own breed.

GENETIC PARAMETERS IN
STRAIGHTBRED AND CHOSSBRED

BEEF CATTLE POPULATIONS

5y
Richard John Dunn

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Animal Husbandry

1968

fif/ /

r///W

L": i“ I

DEDICATION

To Mom and Dad for their
unending inspiration, en-

couragement. love, and

sacrifice.

11

ACKNONLEDGEMERT

Sincere thanks and appreciation are expressed to Dr.
William T. Magee. the author's academic advisor. for his
guidance. encouragement. and assistance during the present
graduate program. His generous. patient. and understanding
-teaching and counseling. in and out of the classroom. will
always be rememberai.

Appreciation is also extended to the other members of
the author's Graduate Guidance Committee: Dr. John L. Gill.
and Dr. Lon D. Fc;illlqr‘. Dairy Department; Dr. Mason
E. Miller. Institute for Extension Personnel Development;
and Dr. Esther M. Smith. Anatomy Department.

The author is indebted to the Animal Husbandry Depart-
ment of Michigan State University. Dr. J. A. Hoefer. Acting
Chairman. for facilities and financial assistance provided
in the form of a Graduate Assistantship. Special thanks are
due Professor Graydon Blank for providing direction on this
assistantship. His cooperation and friendship will long be
remembered. The author is also grateful to all the staff.
employees. and graduate students in the department for their
support. assistance and companionship.

Appreciation is expressed to the United States Depart-
ment of Agriculture and University of Nebraska. Department
of Animal Science. for providing data and guidance on this
study. Special thanks are due Dr. Larry V. Cundiff. Investi-
gations Leader. NC-i Project. U.S.D.A.. Lincoln. Nebraska;

Dr. Robert M. Koch. Professor of Animal Science. University

111

of Nebraska; and Dr. Keith E. Gregory. Acting Director.
United States Department of Agriculture Meat Animal Research
Station. Clay Center. Nebraska.

The author is grateful to Dr. Walter R. Harvey. Depart-
ment of Dairy Science. Ohio State University. for running
part of the statistical analysis on the data. Also to Miss
Marsha Spieler. graduate student in Animal Husbandry; Mr.

w. L. Ruble. Research Associate. Statistics Department; and
Dr. William T. Magee for their help in the programming of
this study.

Grateful appreciation is expressed to Mrs. Kathryn Ide
for her skillful editing and typing of this manuscript. Also.
for her assistance and c00peration throughout the author's
graduate program.

Above all. the author is deeply grateful and indebted
to his parents. brothers and sister. other relatives. and
friends for their continual encouragement. support. under-

standing. and inspiration.

iv

TABLE OF CONTENTS

Introduction

Objectives

Review of Literature

Materials

Statistical Methods

Results and Discussion

Application

Summary

Literature Cited

21

27

46

67

74

79

Table

10

11a

11b

12

13

LIST OF TABLES

Heritability estimates for beef cattle
characters (Warwick. 1958)

Heritability estimates for beef cattle
characters (Review of Literature)

Genetic correlations for beef cattle characters.
(Review of Literature)

Summary of heritability averages obtained by
paternal half-sib method (Petty and Cart-
wright. 1966)

Summary of genetic correlation averages (Petty
and Cartwright. 1966)

Experimental design showing the total number of
sires. dams and calves by subgroup.

Description of the traits used in this study.

Example of the analysis of the data. steer
calves. Hereford sires.

Expected composition of variances and regression
coefficients for autosomal inheritance in non-
inbred strains, with no assortive mating.
(Dickerson. A.S.A.P.. 1960)

Analysis of variance and covariance for com-
puting heritabilities and genetic correlations

Covariance between sire averages for straight-
bred and crossbred calves including all traits
where the covariance of another trait with
yearling weight was calculated.

Components of covariance between sire averages for
straightbred and crossbred calves including the
traits where the covariance among different
traits was not calculated.

Components of variance for between sires (V(S))
and within sires (V(W)) for different mating
types and pooled averages for straightbred and
crossbred calves.

Heritabilities for the straightbred and cross-
bred steer and heifer calves. and genotypic
and phenotypic correlations for the straight-
bred steer and heifer calves.

vi

15
17

18

18

21

29

34

35

47

#8

49

50

52

Table Page

14 Correlations between the genetic ability of
straightbred beef cattle sires to produce
straightbred steer progeny (S) and their
genetic ability to produce crossbred steer
prOgeny (C) for the same trait. and for one
trait in the straightbreds and another in the 56 &
crossbreds (and vice versa). 57

15 Correlations between the genetic ability of
straightbred beef cattle sires to produce
straightbred heifer progeny (S) and their
genetic ability to produce crossbred heifer
progeny (C) for the same trait. and for one
trait in the straightbreds and another in the
crossbreds (and vice versa). 58

vii

LIST OF FIGURES

Figure

1 A path coefficient diagram showing the
correlations involved when a sire is mated
to several cows of his own breed to produce
straightbred offspring. and cows of another
breed to produce crossbred offspring.

2 Sire progeny means. by years. for the trait
adjusted weaning weight (in pounds) in
steers. Shorthorn sires on Shorthorn and
Angus dams.

3 Sire progeny means. by years. for the trait
adjusted 550-day weight (in pounds) in
heifers. Angus sires on Angus and Hereford
dams.

viii

45

61

 

INTHOJUCTION

The beef cattle breeder of today is getting caught in
a "cost-price squeeze". He needs to identify those animals
in his herd. or other herds.which W111 add the MOSt im~
provgg,ut 14 the traits that will contribute to productive
efficiency and better beef carcasses. He must then utilize
these animals in effective breeding plans that will help
»1m "1" - improvement in his herd. This. in turn.
will provide increased profits and a better means of living
for these beef cattle breeders. the packers and retailers of
beef. and will supply the consumer with an appealing. high-
quality. nutritious product at a reasonable cost. As
Gregory (1965) states: "One segment of the beef cattle
industry cannot be divorced from the other segments. From
a long term standpoint. there is an interdependence among
them. The commercial producer is interested in cows with a
long productive life that wean a high percentage of heavy.
high grading calves; the feeder desires rapid and efficient
feedlot gains; and the packer and retailer are interested in
the maximum amount of edible portion per unit of live or
carcass weight. The consumer expects this edible portion
to be tender. flavorful and juicy".

Beef cattle scientists need more basic knowledge on
genetic parameters in order to recommend better breeding and
selection methods and plans to these breeders. Measurement

procedures for economically important traits and reliable

_ 1 -

estimates of their genetic parameters (heritabilities. genetic
correlations. heterosis. etc.) are fundamental in the search
for this basic knowledge. As these measurement procedures

are developed. one must consider the relationship of the
measures to the trait being studied. the heritability of the
measurements. their genetic correlations with other traits
being studied. and the relative economic value of the trait
being studied. This information is essential to the future
potential of beef cattle breeding in this age of efficient.
scientific agriculture.

Heterosis has been observed in some of our domestic
animals for many years. Several studies are presently in
progress to study the influence of heterosis in beef cattle.
Several recent studies have used the British breeds because
they are the ones most frequently used for production of beef
in the northern two-thirds of the United States. The in-
fluence of heterosis is estimated by comparing the crossbreds
with the average of the straightbreds sired by the same bulls
and out of comparable cows. The difference between the
crossbreds and straightbreds is due to non-additive gene
effects. In general. crossbreeding has resulted in about
3% Increase in growth rate up to about 18 months of age. in
favor of the crossbred cattle. Heterosis in carcass traits
apparently is important only in those traits that are associ-
ated with growth of the animal (Gregory g£_§l.. 1965,
1966a. 1966b. 1966c; Gaines £2 21.. 1966. 1967; Brinks gt

- 3 -

al.. 1967; and Vogt 2E.§l~v 1967.)

Beef cattle scientists are generally in agreement that
the amount of genetic.change of economically important
traits in domestic animals. such as beef cattle. depends
largely on the size of the genetic variances and the herita-
bilities of the traits studied.

The heritability estimate is a key to improving the
population by breeding methods. It is considered one of the
most important parameters in animal breeding. Lush (1948)
states it as follows: "A characteristic is not inherited
as such. The thing inherited is the ability to respond in a
given manner to a given environmental circumstance. The ob-
served phenotype is the net result of these inherited
potentialities and the environmental circumstances. such as
nutrient supply. temperature. diseases. accidents. etc.
which they encounter. Between the genes which are trans—
mitted and the observed phenotype of the plant or animal is
a considerable gulf of time and of chemical and physiological
processes in which the genes interact with environmental
substances. forces and conditions. and also with the primary
and secondary products of each other. The complete story
of all that happens in this period includes the whole subject
matter of embryology and the physiology of growth and
development." Reviews reveal only one estimate of herita-
bility between straightbred and crossbred beef cattle (Miquel

and Cartwright. 1963). but in the last 20 years numerous

-4-

studies have been made.providing heritability estimates for
economic traits.in straightbred and purebred beef cattle.
These studies reveal'twytheritability of birth and weaning
weight. and conformation scores are moderately high. and
post-weaning growth rate and carcass traits are highly
heritable. Warwick (1958) concluded that beef cattle differ
in their inherent productivity and that those differences
are rather high in heritability. Gregory (1961) reported
that heritabilities for most of the economically important
traits. except fertility. seem high enough for selection to
be reasonably effective.

If selection pressure is applied to more than one trait
at the same time. progress is affected by the genetic inter-
relationship among the important traits. the amount of
selection practiced. and the generation interval. In order
to prepare effective breeding plans. estimates of genetic.
environmental. and phenotypic parameters for traits of
economic importance are needed. Accurate estimates of genetic
correlations allow us to predict direct and correlated re-
sponses in one trait resulting from selection for another
single trait. There are only a few estimates of genetic
correlations between traits. and studies on correlated re-
sponses in the literature. This is mainly because of the
great amounts of data that are necessary to provide reliable
estimates of these genetic correlations among the important

performance traits. Gregory (1961) feels that to date no

-5-

important genetic antagonisms have been shown among important
performance traits. but we have not shown experimentally

that mass selection is an effective breeding program for
these traits.

If non-additive gene action is a major source of genetic
variance. selection on the basis of purebred or straightbred
performance may not be effective for improving the perform-
ance of the crossbreds. The research by Gregory. using
these data. show that most of the traits exhibit hybrid
vigor. therefore some non-additive gene action must be involved.

Comstock (1960). referring to swine. stated the need
for estimates of genetic variance among crossbred sire
families. and the regression of crossbred sire progenies on
purebred sire performance. If this is low. selection for
purebred performance will be preportionately ineffective for
crossbred improvement. If negative genetic correlations are
important. he felt that the effective genetic variance among
crossbred sire families may be considerably greater than
that within the breeds.

To evaluate the expected progress in a crossbred popu-
lation when selection is practiced in a purebred Popula-
tion. estimates of genetic variances and covariances from
contemporary purebred and crossbred data in beef cattle are
needed. These estimates are not in the literature. These
are crucial questions that must be answered if crossbreeding

becomes a more important force in our future beef breeding

programs.
The

1.

5.

objectives of this study were:

Obtain estimates of the heritability of economi-
cally important live- animalfland carcass traits in
aastraightbred beef cattle pepulation.

Obtain estimates of the heritability of economi-
cally important live- animal.and carcass traits in

a crossbred beef cattle population.

Obtain estimates of genetic variances and covariances
from contemporary straightbred and crossbred data in
beef cattle.

Study the effect of genetic variances and covariances
from contemporary straightbred and crossbred beef
cattle data on the effectiveness of selection for
genetic improvement in beef cattle.

Estimation of interrelationships of straightbred

and crossbred progeny performance in the same trait.
and between one trait in the straightbred popula-
tion. and another trait in the crossbred population.
To obtain estimates of specific genetic parameters
in straightbred and crossbred populations of beef
cattle for various economic traits. and to use these
estimates to predict trait response to indirect
selection within populations of similar genetic

makeup.

REVIEW OF LITERATURE

Information is very limited in beef cattle breeding re-
search regarding differences in heritability estimates be-
tween straightbred and crossbred populations. genetic
correlations between productivity of traits in straightbred
and crossbred populations. and covariances between pr0geny
performance of traits in crossbreds. straightbreds. and be-
tween the two populations.

Some studies have been made in swine. This is probably
because crossbreeding research in swine has been conducted
.longer than in beef cattle.

Enfield and Rempel (1962) made a study to estimate the
covariance of sire effects in purebred and crossbred p0pula-
tions of swine. These covariance estimates were: weaning
weight. -2.42 i 3.67: average daily gain. 0.0040 : 0.0018;
and backfat probe. 0.0005 : 0.0007. The authors stated that the
genetic progress that can be made in improving crossbred per-
formance by selection within pure lines can be estimated if
the following is known: (a) selection intensity in the pure-
breds. (b) phenotypic variance of the selection criterion
in the purebreds. and (c) the covariance of sire effects in
the populations under selection and the population cross.
They concluded that practicing mass selection in both sexes
of both purebred p0pulations that make up a cross. would make
the expected improvement of the crossbreds the product of

the average selection differential in the purebreds. and the

-7...

- 3 -

ratio of four times the covariance of sire effects. divided
by the phenotypic variance in the purebreds. They made esti-
mates of this ratio which they felt were somewhat analagous
to heritability in the purebred populations. These were:
weaning weight. -.17; average daily gain. 0.92; and backfat
probe. 0.07.

Taylor gt El: (1965) studied genetic correlations be-
tween straightbred and crossbred swine. The straightbred
and crossbred daughters of 35 boars werecomposedcfl‘straight-
bred daughters having crossbred litters. daughters retained
to propagate their strain. and crossbred daughters. Traits
studied were: litter size and litter weight at birth. 21.
and 56 days. Three variance components and two covariance
components involving comparisons of crossbred daughters with
both of the other types of daughters were estimated for each
trait. One-half of the variance components were negative.
but,in general,they were smaller than the corresponding co-
variance components of the same sign. Only the covariance
components for litter weight at 21 days and one for litter
weight at 56 days were positive. Only three genetic correla-
tions were obtained since some negative variance components
were obtained. A positive genetic correlation of 0.18 was
obtained for litter weight at 21 days when comparing cross-
bred daughters to straightbred daughters. For litter weight
at 56 days the correlation was 0.61 between crossbred and

straightbred daughters and -.38 for crossbred daughters and

- 9 -

straightbred daughters having crossbred litters. They con-
cluded that since most of the covariance components were
negative. non-additive gene effects may be important sources
of variation in the traits studied.

Stanislaw. 33 g}, (1967) studied progeny records of 99
purebred boars that sired both purebred and crossbred litters.
The heritability estimates for 56-day weight. postweaning
daily gain. and probed backfat thickness in the purebreds
were 0.03 i .06. 0.28 i .06. and 0.55 1 .12. respectively.

The correSponding estimates within the crossbreds were 0.19

i .09. 0.39 i .10. and 0.47 t .13. The genetic correlations
within the purebreds between 56-day weight and postweaning
daily gain. 56-day weight and backfat thickness. and post-
weaning daily gain and backfat thickness. were 0.29 i .50.
-.05 i .53. and -.07 i .18. respectively. Within the cross-
breds the corresponding correlations were 0.20 I .21. 0.61 i
.16. and -.39 i .18. The authorscxnuihthithat improvement

in postweaning growth rate and probed backfat must come al-
most entirely from selection pressure applied to these traits.
but in the crossbred it appeared that postweaning daily gain
could be increased by selection for either 56-day weight or
less backfat. The sire components of covariance were 1.61.
0.0013 and 0.0023 for 56-day weight. postweaning daily gain
and probed backfat thickness. respectively.

Robison. gt El! (1964) studied boars with both purebred

and crossbred prOgeny to obtain preliminary estimates of the

-10..

effectiveness of selection in purebred populations for
achieving improvement in crossbred populations. Genetic
correlations between purebred and crossbred progeny means
for the two breeds of boars were 0.22 and 0.72 for 140-day
weight and 0.21 and.:>1400 for backfat at 140 days. They
concluded that selection for purebred performance would not
be effective for improving the crossbreds.

Louca and Robison (1967) estimated components of
variance and covariance from records on purebred and cross-
bred pigs sired by 76 boars. Heritability values from
paternal half-sib correlations were estimated for birth
weight. 154-day weight. litter size at birth. weaning. and
154 days. and backfat probe. The authors quoted these
estimates as 0.17 t 0.42. 0.09 i 0.29, and 0.05 i 0.20 for
birth weight. 0.35 t 0.34. 0.14 t 0.15. and 0.33 i 0.18 for
backfat probe. and 0.70 t 0.41. 0.81 t 0.35. and 0.65 t 0.24
for 154-day weight in the purebred boars. barrows. and gilts.
respectively. Corresponding values for the crossbred barrows
and gilts were 0.01 t 0.05 and 0.03 i 0.05 for birth weight.
0.03 t 0.04 and 0.00 t 0.04 for 154-day weight and 0.22 t
0.06 and 0.09 i 0.05 for backfat probe. The comparison of
the components of variance in the two breeding groups led
the authors to suggest that non-additive gene action was in-
volved. Both studies indicated to the authors that selection
on the basis of purebred performance would not be effective

for improving crossbred performance and that selection for

-11..

specific combining ability should provide a better means for
utilizing heterosis experienced in crosses. Genetically.
birth weight was negatively correlated with backfat probe
and positively correlated with weight at 154 days. Weight at
154 days was negatively correlated with backfat probe. The
authors felt that. in general. the magnitude and direction
of associations among traits were high enough to suggest that
simultaneous improvement in all traits would be possible.
Only one reference comparing heritabilities in cross-
bred and purebred (or straightbred) beef cattle was found in
the literature. Miquel and Cartwright (1963) compared
heritabilities for birth weight. weaning weight. and feedlot
gain from 10 years of data on Herefords (H). Brahmans (B).
and various crosses of the two breeds: B sire x H dam. Fi's
(BB). H X B Fl's (RB). backcrosses to H sires (HF1) and back-
crosses to B sires (BFl). They adjusted their data fer
effects of sex. year. and age of dam with least squares
analysis. Heritability estimates by the paternal half-sib
method for birth weight were: H. 0.15; B. 0.16; BH. 0.55;
HB. 0.50; HFi' 0.26; and BFl. 0.20. Heritability estimates
for weaning weight were: H. 0.24; B. 0.44; RR. 0.25; HR.
0.22; HFl. 0.07; and BFi’ 0.19. Estimates for feedlot gain
were: H. 0.74; B. 0.23; BH. 0.90; HB. 0.00; HFI. 0.42; and
BF1. 0.70. The authors noted that the estimated genic
variance and heritability ranked in the same order for all

characters with one minor exception. They concluded that

-12..

selection would be roughly as effective in crossbreds as in
purebreds.

One study was also located in the literature that dealt
with heritability estimates and genetic correlations between
straightbred and crossbred lambs. Bassett and Shelton (1966)
studied birth weight. slaughter score. and 120-day weight of
lambs out of straightbred ewes. and sired by rams of the
same breed. and rams of two other mutton breeds. Heritability
estimates for birth weight were 0.60 for the adjusted data
in the straightbred lambs. and 0.24 and 0.03 for the cross-
bred lambs. Estimates of heritability of adjusted 120-day
weight for the straightbred lambs were 0.47 and 0.39. as com-
pared with 0.25 and 0.45 for the crossbred lambs. Estimates
of heritability for adjusted slaughter score were 0.29 and
0.33 for the straightbred lambs. and 0.21 and 0.19 for the
crossbred lambs. The two estimates include data which were
both adjusted and not adjusted for environmental effects as
determined by least-squares analysis. In only one instance
did the crossbred data give a higher estimate of heritability
than the straightbred data. The genetic correlation between
birth weight and 120-day weight in the crossbred data was
0.87. This was the highest value found in the study. The
authors did not quote other genetic correlations.

The literature contains numerous studies of heritability
estimates for economic traits in straightbred and purebred
beef cattle. There are. also.21 few: estimates of genetic

correlations between economic traits in straightbred and

- 13 -

purebred beef cattle. Warwick (1958) attempted to summarize
studies which were known to have been reported up until that
time. Those heritability estimates. related to this study.
are given in table 1.

Table 1. Heritability estimates for beef cattle characters
(WarwickJ 1958)

f
L

 

 

No. of Av. of Range of
Character estimates _§§timates estimates
% %

Birth weight 15 41 11-100
Weaning weight 26 30 -13-100
Post-weaning
feedlot gain 13 45 19-70
Post-weaning
pasture gain 6 30 9-43
Carcass traits:

Carcass grade 5 34 —30-84

Rib eye area 3 69 69-72
Conformation grades:

Weaning 16 26 0-53

Slaughter 5 39 -13-63

 

Warwick also summarized the genetic interrelationship
of traits in beef cattle that had been studied prior to his
summary. Results prior to 1958 showed one study indicating
a negative relationship between maternal abilities and post-
weaning gaining ability; another giving positive genetic

correlations for gains of steers through three seasons,in-

-11)..

cluding a winter feeding period. grazing for a summer. and a
second winter feeding period; and one with positive genetic
correlations between gains in consecutive 84-day periods of a
252-day feeding period. Observed positive genetic. but nega-
tive environmental correlations between weaning score and
subsequent feedlot gains were found by one group of workers.
Several others studied genetic relationships among grade and
gains at various periods. Only a few reports on genetic
correlations had appeared at that time due to the large bodies
of data required for precise estimates.

Warwick¥38ummary has been used as a base for this review.
No attempt has been made to duplicate the literature cited
in his summary.

Table 2 is an attempt to summarize heritability studies
since Warwick completed his summary in 1958. including esti-
mates related to this study. Only heritability estimates
obtained by the paternal half-sib method are included. The
estimate is a simple arithmetic average of the values reported
by the research workers. No attempt has been made to adjust
values to number of head of cattle included. variance of the
estimate. etc. Where a worker reported more than one value
(different sexes. different breeds. different planes of nu-
trition. etc.) one combined average value was obtained.

.Again. this was a simple arithmetic average with no adjustment
made. If the author gave a pooled estimate. this was the one
used. Thus. the summary contains one estimate per trait. per

report. with each receiving equal weight.

_ 15 -

Table 2. Heritability estimates for beef cattle characters

 

No. of Av. of Range of
Character gggtimates estimates estimates References
%’ ZT
Birth weight 17 44 11-100 61.95. 46. 84.

90.82. 67! 6 D
49. 4. 80. 43.

77.50. 24. 33.
70.48.

Weaning score 27 37 5-70 10. 47.99. 3.
100. 85. 82.
72. 63. 57. 6.
78. 49. 18, 60,
36. 57. 62. 80.
55. 12. 69. 23.

91. 34. 50. 70.
48.

Weaning weight 40 35 -12-100 61. 97. 10. 68.
Q6. [‘70 8L}. 99’
90' 3' 100' 85’
82. 67. 6. 78.
64. 86. 36. 9.
88. 35. 89. 4,
57. 80. 79. 21.
76. 55. 43. 20.
12. 69' 23’ 42'
77' 50! 21+. 700

Final feedlot

weight 17 57 2-100 81. 84. 99. 5.
3. 100. 85. 82.
86. 9. 8. 88.
21. 19. 58. 13.

70.

Marbling score 2 47 31-63 9. 23.

Final carcass grade 9 57 0-135 10. 13. 3. 82.
56. 17. 21. 20.
23.

Eat thickness 8 48 24-74 13, 82, 86, 9,
17. 21.20. 23.

Rib eye area 9 58 12-156 13. 82. 56. 9.
17.44.21.20.23.

Actual cutability 1 23 23.

Long yearling 97. 3. 85. 66.

pasture weight 10 40 10-71 6.36.59.76.58,70

 

- 115 -

Several summaries of heritability estimates have been
made since Warwick's. Gregory (1961) presented a summary
almost identical to that of Warwick. He did not identify
the references included in each trait. Clark gt 31.,(1963)
included two tables of heritability estimates. but he did
not obtain an average value for each trait. Most of his
references were included in the summary by Warwick. but he
included a few later studies.

Table 3 is a summary of genetic correlations reported for
cattle since 1958. Warwick (1958) presented a brief dis-
cussion of the work up to that time. but did not present
actual values.

Petty and Cartwright (1966) prepared a summary of
genetic and environmental statistics for growth and confor-
mation traits of young beef cattle. A number of the traits
were the same ones studied in this work. They presented
weighted averages of paternal half-sib estimates. as well as
unweighted ones. The estimates were averaged by weighting
them with either the number of sires included in each esti-
mate or the estimated number of sires based on the average
number of offspring per sire in the other estimates of that
trait. Genetic correlations were given. with appropriate
sire-weighted averages determined by the "Z" transformation
method using the number of sires or the estimated number of
sires involved in the estimate when the actual number was
not given. These summaries are presented in tables 4 and 5.

The authors tried to select the most independent and pertinent

_ 17

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o c 053
mm mm o o .m m mazpmmn
.mm .m wHu» mscq
. ruHHHpmpso
mm mm Heap0¢
mm mm poem
.mH .Nm .mH .mm mm mm mm .Nw mm .mm mm whouon
mm mowmwo. mm m mm mmononnp
o o . . N m .Nm . .
H m NH mm m mm mm mm mm Han
math“? moéafim. . .
HO. WK. MN Nm .m N” MN Nm MN N Wdfihaw mmwo
h m .m .HH .m.HH w name Hmzflm
HH.I mm. 30. choom
H H H mm MN wGHHQLma
marmH. .. . . m o . . . ou.mm
no. on. mmman mwmooH m mm on .n.g .ps poH
H N N m.mm.&m .05 n mm JwOmQPgHm
Hm.-mH. SIS. Hd.umw.u $.30... maimm. 6m. .mnflfiméusimx
mm. om. mo. mm. mo. we. .oimmfimdod 334m .93
\M \N mi & H m . . . . . . mnHmmmx
HHm.-v-wo.u aw.-mo. deuom3. mo.-mH.u mNflHm.u .me.os
am.- mm.- 0H. no.. em. nH.- am. we. um.ms mpoom
N H N H a H m mH n mm . m mchmi...
mm.uad. mm.u©H. mm.:©%..
om. Ho.n 0H.I m:.n 00. do. moo. qu
H H H H m 5 epme
7.9: pHHHo awed mmcc much oaoom .93 .9: bacon .us
mgaemmn ambuo whonnHm uonsp mmmcpmo quHbama pOHnomw wchmcz wsHsmmz :paHm
.;Hem mica Hdzpoq baa HMSHd HmsHb

 

 

 

m3» mo

H.Hmnomed can
do pmmH esp pm chm mmcsoacmcm .owsma emu Hones: soupon on» cam .mopmaHpmm

.>m ms» mcHon Hones: chdHa can .mmumaHpmc no 90985: on» wsHmp amp

:85: do» exp :sz HanomeU on» mo uanH on» on and mnoHpmHohhoo oHumCoov

.mampomnmso mepmo moon you mQOprHmHHoo oHpmSoo

.m

mHnwa

_ 13 -

Table 4. Summary of heritability averages obtained by
paternal half-sib method (Petty and Cartwright.
1966)

W

 

 

 

Off. per
No. of Unweighted Weighted sire.weighted
Trait estimates average average average
Birth weight 21 .44? .44 16
Weaning score 24 .34 .36 16
Weaning weight 30 .33 .32 14
Final feedlot
weight 15 .56 .62 10
Yearling pas-
ture weight 9 .39 .41 10

 

Table 5. Summary of genetic correlation averages.
(Petty and Cartwright. 1966)

 

Off. per sire
Correlation weighted
between Overall rG Weighted rG average

 

Birth weight -
Weaning weight .58 .58 21

Birth weight - ’
Weaning score —.01 .36 13

Birth weight -
Final feedlot weight .61 .64 --

Weaning weight -
Weaning score .24 .39 13

Weaning weight -
Final feedlot weight .75 .79 11

Weaning weight -
Yearling pasture
weight ,ué .67 12

Weaning score - Final
feedlot weight .47 .43 _-

Weaning score - Year-
ling pasture weight -.08 -.O3 10

 

- 19 -

information available. The summary varies from the ones in-
cluded in this study since it contains selected references
dating back to the first studies (1946). Thus. it contains
references cited in Warwick's summary. as well as those
presented in this study.

In general. the review of literature summary of herita-
bilities in this study is higher than those of Warwick (1958)
and Gregory (1961). averaging about 9% higher in the traits
included in all summaries. The study is in rather close
agreement. however. with the summary of genetic statistics given
by Petty and Cartwright (1966). including both heritabilities
and genetic correlations. There are some differences in the
data included. as was discussed earlier. but in a number of
the heritability estimates and genetic correlations. the
majority of the references used were identical in both cases.
In some cases they organized the material in a slightly
different manner than that done in this study. The present
summary also contains some additional references not included
by Petty and Cartwright. or published after their study.

At best. heritability estimate averages and genetic
correlation averages presented can only be used as a guide.
but they probably give as good an overall picture as anything
presently available. All summaries combined represent over
100 studies. and include a high percentage of the work done
since 1946. when the first estimates were made. We must

realize. however. that the values may be biased by many yet

- 20 -

undetermined factors. Some of the estimates are based on

too few estimates to give the type of accuracy and confidence
useful estimates should have. It is hoped that these summaries
may be of value to research workers to compare their future
estimates with. and for them to use in formulating breeding

plans in beef cattle.

MATERIALS

Data for this study were obtained from an experiment
designed by Gregory gt El- (1965, 1966a. 1966b and 1966c)
to evaluate the effects of heterosis on traits of economic
value in beef cattle.

The plan of the experiment was described by Gregory gt
El- (1965. 1966a. 1966b. 19660). The experiment began in
1957. High grade heifer calves were purchased from Angus.
Hereford and Shorthorn breeders in Nebraska. Montana and
Colorado. All females were outbred and randomly selected
from large populations of their respective breed. The three
breeds were chosen because they are the most important ones
used for beef production in the United States. The plan was
to obtain calves of each breed from a number of sources.

Table 6 contains the design of the experiment. showing
the total number of sires. dams and calves by subgroup. The
numbers vary slightly from those shown by Gregory since all
calves not having complete data recorded on them were dropped
from this study. This was necessary for the computing of
genetic correlations and covariances.

Table 6. Experimental design showing the total number of
sires, dams and calves by subgroup.

Breed of sire and number of offspring
Dams Hereford Aggus Shorthorn Total

 

 

 

 

 

-_§reed No. M3 Fb M F m F ‘gffspring
Hereford 80 63 53 35 25 34 35 245
Angus 80 29 37 55 56 26 39 242
Shorthorn 80 33 33 34 27 66 57 250
Total 240 125 123 124 108 126* 131 737

an: male. bF: female

- 22 -

The original experimental design included eighty fe-
males of each of the three breeds. These females produced
four calf crops (1960. 1961. 1962 and 1963). Because all
females were the same age each year. calves produced during
the four years were out of 3-. 4-. 5-. and 6-year-old dams.
respectively. Females not calving in any given year were
removed from the experiment.

The design of the experiment included using four bulls
of each breed per year. This was revised slightly. with
three Angus bulls being used in 1960. and six in 1963. This
made a total of 16 Hereford bulls. i7 Angus bulls. and 16
Shorthorn bulls. These bulls were obtained from breeders
or experiment stations in Montana. Wyoming. Colorado. North
Dakota. Nebraska. Kansas. Oklahoma. Iowa and Missouri. A
deliberate attempt was made to obtain bulls from different
sources. Some of the sires were inbred. but the inbreeding
was low. The sires represented a cross-section of their respec—
tive breed. The females were randomly allotted to breeding
pastures. with twice as many being exposed to bulls of their
own breed as to each of the other breeds.

Site of the experiment was the Fort Robinson Beef Cattle
Research Station. Crawford. Nebraska. The station is lo-
cated in the Mixed Sandy and Silty Tableland of northwestern
Nebraska. The average rainfall in the area varies from
zabout 15-18 inches per year. and the native range vegetation

is made up of short and intermediate grasses.

- 23 -

The management of the experimental herd included a
calving season extending from approximately February 10 to
May 1. with the majority of calves being dropped between
February 20 and March 31. Calves ran with their dams on
native pasture until weaning time. which was the first week
in October each year. Calves averaged from 200 to 210 days
of age at weaning time. All calves from each calf crop were
weaned on the same day. Birth weight was recorded on each
calf within 2“ hours or birth. All male calves were
castrated and all calves with horns were dehorned at the
time they were weighed. The weaning weight was recorded in Octo-
-bon. All weaning weights were adjusted to a constant age of
200 days. by multiplying the average daily gain by 200 and
adding the birth weight. A committee evaluated the beef
conformation of each calf at the time they were weaned. The
scoring system used is listed in table 7. Most of the calves
were given the choice grades.

All the heifers from each year were run together with
the exception of the breeding season of their dams. Those
dropped in 1960 and 1961 were managed for calving as 3-year-
olde; They were fed 1 pound of a #0} supplement per head
per day on native range during the first winter. The ration
was planned to produce gains of approximately 0.5 pound per
day during the 196-day wintering period from weaning in
October until April. The following_summer the heifers were
grazed on native summer range for a grazing season of 15“

days.

 

_ 2Q -

The heifers from the last two calf crops (1962 and 1963)
were managed for calving as 2—year-olds. They were grazed
on limited native range the first winter. but received about
5 pounds of concentrate feed per head per day in addition
to a liberal feeding of hay. This program was planned to
produce a daily gain of about a pound during the 196-day
period. These heifers were grazed during the following
summer on a short and intermediate grass range similar to
that used with the first two crOps of heifers.

Following weaning. all steer calves were fed for a post-
weaning period of 252 days. Weaning weight and date were
used as initial weight and the base date for the feeding
period. A growing-fattening ration with approximately 65%
TDN was fed to the steers. The 1960 steers were group fed.
while the 1961. 1962 and 1963 steers were on a prOgram of
individual self-feeding. The feeding program for the indi-
vidually fed steers consisted of a 2- to 3-week conditioning
period. followed by the use of individual self-feeders for the
rest of the 252-day period. The steers were randomly
assigned to their individual feeder. and were tied to their
feeder overnight except the 1963 steers which were fed part
of their feed during the day.

The feeding program consisted of a pelleted ration in
1960 and 1961. and one that was ground and mixed in 1962 and
1963. During most of the years about 1.5 pounds of long

grass hay per steer was fed free choice to the group before

- 25 -

they were fed their individual grain ration. Plenty of bunk
space was available for the hay and it was readily consumed.
Ample water was available to the steers in their group lots
when they were not in their individual feeding area. All
steers were slaughtered at the end of the 252-day feeding
period each year at an average age of 452 days.

Carcass measures and grades were obtained on all four
calf crops in a commercial beef cooler. Detailed carcass
cut-out data. using the right side of each carcass. were ob-
tained on calves from the 1961. 1962 and 1963 calf crops.
Processing of wholesale cuts was on the basis of boneless.
closely trimmed cuts (not over(L3 inch of fat on any surface).
Retail product. fat trim and bone were weighed and recorded.

All cuts were boneless with the exception of the loin
which contained the vertical and Spinous processes. and the
rib which contained approximately 6 inches of the rib bones.
The same cut-out procedure was used on all carcasses. Data
was adjusted to a complete carcass basis by multiplying the
cut-out values obtained from the right side of each carcass
by two.

Beef cooler carcass data were obtained by official
graders of the U. S. Department of Agriculture.

Actual cutability was not obtained for the steers pro-
duced in 1960. This was estimated from the percent retail
cuts estimated by the USDA equation,using data from the other
three calf crops as a guide. A list of the traits used in

this study is shown in table 7.

- 26 -

Table 7. Description of the traits used in this study.

 

 

 

 

 

 

Item Description
Steers and Heifers (Preweaning):
Birth weight Actual birth weight
Weaning score 7 low good. 8 = av. good.

9 high good. 10 a low
choice. 11 = av. choice.

12 = high choice. 13 = low
prime. 14 = av. prime. 15 =
high prime

200 day weight Average daily gain birth to
weaning x 200 + birth weight

Steers_LPostweaning):

Adjusted final weight Actual final weight minus
actual weaning weight + 200
day weight

Iarbling score 0 = tr.
1 = Sl-. 2 = 81. 3 = 81+
a = sm-. 5 = sm. 6 = sm+
7 = mt-. 8 2 Mt, 9 = Mt+
10 = md-. 11 = md. 12 = Md+
13 = sl.ab.-. 14 = s1.ab..
15 = sl.ab.+
16 = md.ab.-. 17 = md.ab..
18 = md.abe+
19 : ab- 20 = ab

Final carcass grade 7 = low good. 8 = av.good.
9 = high good. 10 = low

choice. 11 = av. choice.
12 = high choice. 13 = low
prime. 14 = av.prime. 15 =
high prime

Fat thickness A single measurement taken
3/4 the distance of the
longest axis of the rib-eye
recorded to the nearest .01

inch.
Rib-eye area Square inches
Actual cutability Lbs. of boneless closely

trimmed beef from the rouni.
loin. rib and chuck.
Heifers (Postweaning):

Adjusted 550 day weight 550 day weight minus actual
weaning weight + 200 day
weight

SfATISflCAL nJTAOJS

The statistical analyses of the data to measure the
effects of heterosis were explained by Gregory E£.El- (1965,
1966a. 1965b. 19660). Because the cattle used in that study
and this study are identical. with a few exceptions,
the analyses done by Gregory were not repeated.

Variance components for all traits studied were obtained
from analysis of variance tables.

The experimental design included a new crop of sires
each year which caused a hierarchal design with sires nested
within breed of sire and year. but cross-classified with
breed of dam. Year and sires were random effects. with the
other main effects being fixed. Steers and heifers were
analyzed separately for each trait. This W33 necessary
primarily for the analysis of post-weaning traits and corre-
lations between pre- and post-weaning traits because of
different management after weaning.

Since the same set of females were used throughout the
study. age of dam was confounded with years. Traits affected
by age of den cause the mean squares for years to be aug-

mented by the age of dam effects.

Method of Analysis:
The following general model was used to analyze the data
in the study. except where otherwise noted.

General Model:

Yijklmo g” + yij + sijklm + eijklmo

-28-

an effect common to all animals

‘:
n

trait

H
II

yij = an effect common to all animals calved in the
3th year.
Sijklm = an effect common to all calves sired by the
mth sire of the kth breed of sire when mated
to the 1th breed of dam in the 3th year.

eijklmo = the random experimental error.

General Design 3: the Experiment:
Table 8 is used to explain the general design of the

experiment.

Explanation of information 13 galls 2£.E§El£ §:

This table is an example of several that were used in
the analysis. This is an example of steer calves from Here-
ford sires. and using one combination of two traits as an
example: birth weight and yearling weight. A similar one
was used for heifer calves from Hereford sires for these two
traits. Similar tables were used for each pair of traits.
in each sex. and in each sire breed group (Hereford. Angus.
and Shorthorn). except where otherwise noted.

Each cell (A to 0) in the table was made up of mean

squares or covariances including the following sources:

Source:

Years
Sires/Years
Calves/Sires/Years
Each cell in the table included the following information:

- 29 -

 

 

Amumv Away
mommaam>oo

 

 

 

.34....2, __J..

A<umv Ammv
cosmaam>oo

 

 

 

H—fi-hq

snonpaosm
neaoacncm

 

 

 

 

heumv
msxs<
nanomehom

III'I‘I'I'I ill ll.

 

 

Ammv
ohomohcm

eownpewamnpm

 

 

 

 

A _
iiifiiii _

_ .9 _

i _

‘

 

 

,’!li)i. ‘4!!!

v6-3 17mm c- H sum H -...:..~ H elk H Tweed H: 2&4 sum a ed H :.N

 

mHHm

a
.

meme meme Home coma meme meme dean oemfi meme mead Heme oema_ nee»
_

 

 

 

 

:5 3m: oeseaneeoo sec pawn; medium» Ema passe: 5.3m fleas

.mcadm unomoacm .mobamc scope .ede on» we mdmhadzm 0:» mo oaassxm .m canes

spexqssoxo

A.

D.

Heritability of the traits birth weight (bw) and
yearling weight (yw) in the straightbred Hereford

population.

Genetic correlation between production of the
traits (bw) and (yw) in the straightbred Hereford

population.

Heritability of the traits (bw) and (yw) in the

Hereford-Angus crossbred population.

Genetic correlation between production of the
traits (bw) and (yw) in the Hereford-Angus cross-

bred population.

Heritability of the traits (bw) and (yw) in the

Hereford-Shorthorn crossbred population.

Genetic correlation between production of the
traits (bw) and (yw) in the Hereford-Shorthorn

crossbred population.

Correlation between the genetic ability of a sire
to produce straightbred Hereford progeny and his
genetic ability to produce Hereford—Angus cross-

bred progeny for the trait (bw).

Correlation between the genetic ability of a sire
to produce straightbred Hereford progeny and his
genetic ability to produce Hereford-Angus cross-

bred progeny for the trait (yw).

 

- 31 -

L* Correlation between the genetic ability of a sire
to produce the trait (bw) in straightbred Hereford
prOgeny and his genetic ability to produce the
trait (yw) in a Hereford-Angus crossbred pOpulation

(and vice versa).

M Correlation between the genetic ability of a sire
to produce straightbred Hereford progeny and his
genetic ability to produce Hereford-Shorthorn

crossbred progeny for the trait (bw).

N Correlation between the genetic ability of a sire
to produce straightbred Hereford progeny and his
genetic ability to produce Hereford-Shorthorn

crossbred progeny for the trait (yw).

0* Correlation between the genetic ability of a sire
to produce the trait (bw) in straightbred Hereford
progeny and his genetic ability to produce the
trait (yw) in a Hereford-Shorthorn crossbred popu-

lation (and vice versa).

* Due to the many combinations of traits involved. a limited
number were selected to study (birth weight. adjusted
weaning weight. adjusted final weight (steers) and 550-day
weight (heifers). and actual cutability). These traits were
selected because they seemed to be the most important

economically. Since adjusted final weight and 550-day

weight were thought to be the most important single traits

-32-

studied. the correlations involved combinations of the other

three traits with adjusted final weight and weight at 550

days of age.
Estimates of genetic parameters and standard errors of

the estimates were obtained by use of the Least-Squares and

I'laximun Likelihood General Purpose Program (LSMLGP) written

by Walter R. Harvey. Ohio State University. who assisted in I

the computations of the analyses on this portion of the
The pro- f

 

data. Details are given by Harvey (1960. 1964).
gram subrout inegwhich computes these genetic parameters.

calculates estimates that are identical to those obtained

by the Henderson (1953) Method 3.
The standard errors calculated by this program are

'approximate because no allowance is made for adjustments of

data for the effects of fixed factors. They thus afford

only minimum estimates of the standard errors (standard

errors are at least as large as these). Formulas used in

the program to obtain estimates of these standard errors

are modifications of those given by Swiger e3 a_l_.. (1961+)

The following model was used for computing estimates of

heritability and genetic correlations from paternal half-

SibS:

 

 

_ 33 -

Yljkl 3% + 3113+ Sljk + eljkl
)I = an effect common to all calves.e

1

trait

T13 ==an effect common to all calves calved in the 3th

year.

Sijk = an effect common to all calves by the kth sire

in the jth year.

eijkl = the random experimental error.

Dickerson (A. S. A. P.. 1960) and Lush (1948) have de-

 

scribed the expected composition of variances and mean
squares. Table 9,from Dickerson,summarizes these.

Methods of estimating variance and covariance components
with unequal sub-class numbers have been outlined by Hazel
g§_§;. (1943). Henderson (1953). and King and Henderson (1954).

Paternal half-sib heritability estimates and genetic
correlations for traits within a mating type were computed
using the analysis of variance form shown in table 10.

The form is similar to that used by Hazel g£_§l. (1943).
Sums of squares and cross products were computed in the
usual manner.

The within sire component V(w1) expresses the variation
between individual offspring by the same sire. The between
sire component V(Si) expresses the differences between the
true averages of sire groups. Therefore. offspring from
different sires have an expected variance V(W1) + V(Si). The

subscript i denotes the trait being studied. The between sire

.mcaasmm ngpHs mQOHpmHame HmHSmEQoHH>sm msz mcazpocow mo soHpompmpsH
asHezaocH HmsHsm mama so msoHpmbaomno wsosm mmosmhcmch ampsmssoaH>sc song cosmHHmbnuzmc
H.0pm .msoHpmhoSQw .msoHmeOH am
.mamoav wusmssoaH>sm mo moaasmm Hsonmch ans moaaposmw mo soHpomHousH scam conHHmb urmkv
.mamsasm newsman mHoQMHc .
pap .HmsHsm cso mo mQOHumeaomno Ham How oxHHm mH peg» pscssoaH>sm some mosmHHm> u m\e
.msmc psoammch aoau
moHHHamw newshon mamaaHe pan .mnHm Harm you oxHHm mH pmnp psossoaHesc soap cosmHHmb n w\v
.muomumm assumpms .
.mmmmww cap ems coppdsmsmnp 0:» ma muoHpmdboc oapofiow Hdep Seashon mosmHHmboo Homhv

 

 

.mpoommo Hmsaopms poohac CH noHpmHHmb OHHmsmw mosh cosmHHm> u Way

.opm .mpoomwc owmaobm mo SOHpomHopGH HooalcoHSH Scam moCmHamb uddwkw

.mcoapmeoc mosmszoc mo soHpomHopsH Hooalosp Bong cosmHHmb u 9mg“

.mcoHpmeoe cosmsHsoc £33 mpoommo .5» mo msoHuomaopsH Hooaaozp sow.“ oocmdagnawb
.mpoogmo .bm mo msoHpomHopsH Hooalozp 80.5 commahmbudwxc

.mpocamo mach owmaobm scam msoHpmece cosmsHsoc Sosa cosmHhmb n mks

 

 

W .mosow no mpoohmo ommaobm Bosh cosmHHmS n MB
2
_ H H H H H H oooH ooo.H ooo.H oooH oo.H 8H 3 + a + a
n a
+ m N‘s
H o o o o o o o o o o 3..esH :Hssz
o om.A H o o o Kev mammv mamvoomsv we. on. a .33
o mm.v o H H H STA mmeox amHsmsmHfl mm. mm. a .58
o 31v 0 o o 0 e8. 0 c 309A 0 mm. m .33
3 mu m o 2 oz <44 8 3. 5. a d
we we we me me a we we we we we mxe eommmemwmawm

 

 

 

 

 

283 4.5.393 53383 sob:
.. 20:5.
.wz as opraomme o: anz .msHmwpm cowasH so: sH couch .
HmaomOpsm How muscwwammooo GodmmoHMOH use moosmawmb mo :oHpHuoasoo ccpocaxm m canes

- 35 -

Table ML Analysis of variance and covariance for computing
heritabilities and genetic correlations.

m
Egpected Expected

 

 

Source d.f. Mean Squares Covariance
Years p-l

V(w1) + ky(sl) Cov(W1w1.)+
kOCov(Sisi,)

Sires/Years p(n-1)

 

Calves/Sires/ pn(k-i) V(w1) COV(W1W1.)
Years
Total pnk-i

component is multiplied by k0 which signifies the average

number of calves per sire group.

Where unequal numbers of calves per sire group are

present. k0 is expressed by Hazel and Terrill (1945) as:
k0 = 4832 - 81mg
Ek(n-1)

where:

k = number of offspring per sire

n = number of sires

The between sire component (V(S))is calculated from the
1
“mule: 30(M5between sires' hswithin sires"
iHazel and Terrill (1945) outlined relationships between

components of variance and the genetic and environmental

variance for different systems of mating (non-inbred in this

study) :

-35-

Non-inbredgpgpulations

 

 

Source Component
variance of offspring V(S) (%)V(G)
V(W) (3/4)V(G)+V(E)
V(S)+V(W) V(G)+V(E)

 

This says. in effect. that the variance component for be-

tween sires. V(S). is considered to be an estimate of i of the

variance. and the variance component for within-sires.

genic
variance plus 3/4 of the

V(W). is considered to be non-genie

genic variance.
The genie variance. V(G). includes all the variance that

(nnibe attributed to the average effects of the individual genes

ffluenon-genic variance. V(E). includes the effects of environ-

ment.<dominance and epistasis.

In random mating populations. the contribution of the
various 'underlying correlations of different genic effects

to the phenotypic half-sib correlation are as follows:

rGG :-

‘. rDD = 0. and rII seems to be about 1/16 for two-gene
interactions. 1/64 for three-gene interactions. etc. (Lush.

1948).

In 'this study} the phenotypic variance of each calf is

made up of the genic variance plus the environmental

variance. This can be symbolized as follows:
V(P) = V(G) + V(E)

The components of variance and covariance in this study can

be attributed to (1) differences between calves by different

- 37 -

sires and (2) differences between calves by the same sire.
The estimates of the components of variance and co-
variance for our problem were derived as follows;
(methods of Hazel gt_§l”.1943):
Variance

Genie Variance - V(G):
V(Gi) = 4.0 V(Si)

V(Si) = variance between sires for the 1th trait

Variance among_ca1ves by the same sire - V(W):
V(Wl) = 3/4 V(Gl) +'V(E1)
V(wi) = within sire variance for the ith trait.

Non:genic Variance - VIED:

V(Ei) = V(Wi) - 3/4 V(Gi)

Phenotypic Variance - V(P):

Covariance

 

Genic Covariance:
COV(G1G13) = 4.0 COV(31311)

Covariance among calves by the same sire:
Cov(w1W1.) -.—. 3/4 Cov(GiGi.) + Cov(E1E1.)

Non-genie covariance:
COV(E1E1') = COV(W1W1.) ‘ 3/“ COV(G1G1.)

Phenotypic covariance:
COV(P1P1.) = COV‘GiGig) + COV(E1E1')

Discussions of heritability are given by Lush (1940. 1948.
1949). Of the methods described in these articles. the

paternal half-sib method is applicable to this study. The

general procedure for computing heritability by this method

is as follows: First, subtract the environmental component

tram their observed resemblance. Second. multiply the re-

sult by four. Third. correct for any deviations from random

mating. This estimate includes the genic variance. a small

amount of the epistatic variance. but none of the dominance

variance. It includes nothing from the maternal environmentbe-

cause paternal half-sibs are used. Tallis and Klosterman(1959)

discuss the factors that influence heritability estimates ob-
tained by the paternal half-sib method.

Heritabilities within each of the breeding groups were

computed as follows:

4V(Si)

 

V(Wi) + V(Si)

The estimates of heritability of the traits pooled for
the different breeding groups in the straightbred population
and pooled for different breeding groups in the crossbred
population were computed using pooled variance components.
These were obtained by adding together the V(Sl) and the
V(wi) for the different breeding groups within the straight-
bred population and within the crossbred population.

The same procedure as discussed earlier was then used.
The computer prOgram used in obtaining these estimates was

written by Marsha Spieler. graduate student in Animal Hus-
tendry at Michigan State University.

-39-

Phebhods fer estimating genetic. environmental. and pheno-

typic correlations have been developed and described by
Hazel. 33 El: (1943).

The following formula was used to obtain the genetic
correlation between two traits in a single population:

(COV $131.)

I‘ =
GiGit W(31) . V(Si')

 

 

where:

Cov 3181. = covariance between sire effects for the

two traits (i and i').

V(Si) = the between sire component of variance for

the 1th trait.

V(Si.) = the between sire component of variance for

the i'th trait.

The method of computing the correlation between the
genetic ability of a sire to produce straightbred progeny
and his genetic ability to produce crossbred progeny was
complicated because of the unequal numbers encountered in
the data (k being different for the two groups).

A method suggested by Robison (1967) was used to obtain
the covariances needed to estimate the correlation between

the genic value of a sire for producing straightbred and

crossbred calves. The covariance between.the straightbred

and crossbred calves by a sire was computed simply as the
covariance between the progeny means of the two groups. Robi-

son thought that the method seemed clear for equal numbers,

-LO-

but left some doubt when used with unequal numbers. His com-
parison of the method with one using an adjustment for un-
equal.tnnnbers (Robertson. 1962) resulted in essentially the
same results. Thus. he felt that this method would do a
reasonably good job of estimation. Comstock (1961) suggested
this simple covariance when he estimated the covariance of
sire effects in his population under selection and the popu-
lation.cross by using the simple covariance between straight-
bred and crossbred half-sib family means taking the two kinds
of families by pairs having common sires.

The covariance of sire effects of sires bred to their
own breed (straightbred progeny) and bred to one of the
other two breeds (crossbred progeny) was estimated using the

following general formula (Cov §bigéi):

a:

3m ijklm. Yijkl'm' £(Yijkl..)(Yijkl'..)

 

 

J - “3
where:
1 = trait
J = year

k a breed of sire
l a breed of dam

m = sire
n3: number of sires in 3th year

An example to symbolize how the covariance of sire

effects for sires bred to their own breed. and bred to other

-141...

breeds. for a given trait. is given below. We are using

trait #1 (birth weight). and Hereford sires with calves out

of Hereford and Shorthorn dams:

62-13%. Y1 insmO- E<Y13HHOO)(I1JHSOO)
3m n

j J_
(Enj‘ L")

When a covariance of sire effects was studied for one
trait in the straightbred progeny, and another in the cross-
bred progeny (and vice versa), the following procedure was
used. The example used is trait #1 (birth weight) in the
progeny sired by Hereford sires and out of Hereford dams, and
trait #4 (yearling weight) in the progeny sired by Hereford
sires and out of Shorthorn dams (and vice versa):

5 Y1 JHHm.Y4jHSm.' E(Y13HH. . ) (if-43113. .)
J “1 +

(Efnj ' 4)

5 Y], Y - ("f )(‘f )
JHHm. 1 JHSm. E 4.)HH. o 1 JHSO e
(5nJ-lfi

The formula used to estimate the correlation between the

/'

genetic ability of a sire to produce straightbred progeny and
his genetic ability to produce crossbred prOgeny for a single
trait. or one trait in one population and one in the other

population (and vice versa) was as follows:

-142-

Cov SP1 801

I‘ =
Gpicci

’VV(sp1)- v (8.1)

 

 

where:

Cmr§p1 §ci = covariance between the progeny means

of the straightbred calves (Sb) and
crossbred calves (SE) for the 1th

trait.

V(Spi) = the between sire component of variance for
straightbred calves by a sire for the 1th

trait.

V(sci) = the between sire component of variance for
crossbred calves by a sire for the 1th

trait.

Progeny means were computed using OnedWay Analysis of
Variance with.Unequal Number of Replications Permitted (UNEQI
Routine) which was programmed by Donald F. Kiel. and the de-
scription written by William Ruble. Donald F. Kiel. and Mary

E'- Rafter. Agricultural Experiment Station. Michigan State
university.

The covariances between the progeny means of straightbred
calves and crossbred calves by the same sire9within years, were
computed, as explained earlier. mixing a modification of the sub-
routine of the BAS’I‘AT Program prepared by William Ruble- Agri-

cultural Experiment Station. Michigan State University.

- 43 _

The between sire components of variance for straightbred
and crossbred calves by a sire were obtained from the Least
Squares and Maximum Likelihood General Purpose Program
(LSHLGP) written by Walter Harvey. Ohio State University.

The program for computing the correlations between the
genetic ability of a sire to produce straightbred progeny
and his genetic ability to produce crossbred progeny for a
single trait. or one trait in one population and another
trait in the other population was written by w. T. Magee,
Animal Husbandry Department. Michigan State University.

This correlation between the genetic ability of a sire
to produce straightbred progeny and his genetic ability to
produce crossbred progeny can be shown more clearly using a
path coefficient diagram.

Since the biometric relations between individuals can
be described in terms of correlations or variances,we can
illustrate them by using the method of path coefficients.
This method was described and used by Lush (1948).

Figure 1 is a path coefficient diagram illustrating the
correlations involved when a straightbred sire is mated to
several cows of his own breed. and to cows of one of the
other two breeds. The illustration shows the bull being mated
to three cows of his own breed. and to three cows of one of
the other breeds since the average number of offspring per
sire group in our study was between three and four (for steers

k0 = 3.64 and for heifers k0 = 3.33). The path between

-424-

the genie value of a parent and its offspring is i in a ran-

dom mating population in which neither the parent nor off-

spring is inbred.

 

 

 

A.mammpmaae use .mommcdaoc .uszQ0ha>co no mpochho one mcfimpnoo. mpomkkm camcmlmoc u m
spaadpmpdwmm> u n
.mnawtmamn no unaponosa on» nmospon coHpmHoHHoo u p
campocmna omwuobm u m
manposesa umbaomno u M
m decd>avsa so we ede> oasmw n c ”mom
R\
m0 5! (hmOSKI MAWQ
>,nH4l S (4 W... U

 

 

.p
K »\W «Amov
mo elll. mmo «I. (umo msaaammuo
.I N a s o c denounwamapm
a “a spas Ana mama
, s e m

mom As)

 

:

“lac endgameeo swappsmamnpm

 

 

 

 

 

 

 

 

.
mu
mo 00
\
\
\ noctum
\ -\ m“ s
5 u
Mo K «\m r\\\\\\\ \ mzahmwww
mm). moo ., bmooofi was NUQ cohnmmoao
.. w 1 a \ N osflie was
a. \
.p \
K a Q\\\\\
Hoomm. MHo :. Home

”ADV madnammmo 0099mm0hun m
.wsahammmo cunnmmono ooscoam on ocean weapons
mo msoo use .wsdnamuho coanuswamupm coupon on woman :so was me msoo amhobcm on

copes ma made a sens cobao>na macaumaoaaoo one wadsosm adawmac psoaoaummoo Spec < .H magmam

RESULTS AND DISCUSSION

Values for the variance components between sires and
within sires and the covariance between the sire averages
for straightbred and crossbred calves, by sexes. are shown
in tables 11a. 11b. and 12. Table 11a includes the covariance
between sire averages for straightbred and crossbred calves
including all traits where the covariance of another trait
with yearling weight was calculated. Table 11b includes the
covariance between sire averages for straightbred and cross-
bred calves. ineluding the traits where the covariance among
different traits was not calculated. Table 12 includes the
components of variance for between sires and within sires
for different mating types and pooled averages for straight-
bred and crossbred calves.

Petty and Cartwright (1966) have summarized weighted
averages of genetic variance estimates in purebred and
straightbred beef cattle for birth weight. weaning weight.
weaning score. final feedlot weight. and yearling pasture
weight. The genetic variances obtained in this study agree
reasonably well with those obtained by Petty and Cartwright.

but. in general. are slightly smaller.

-46-

-47-

mobamo cameo“

H
14

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

saozphosm u s .msw:¢ n o .UHOthcm u we mobamo cams ulnw
oH.oa o.ooH ofi so oo
mm.mo o.Hom oH os ss
Ho.oa o.ssH as oo oo
oo.HN o.s~a as no oo
mo.ou e.oH- ea os ss spaaanapso
oo.sm o.mo as so oo fleeces
m.smm
s.esm m.oos o.oHo o.mom o.soo No.omu mfi.oa: ea oH so oo
fi.osa o.oefi- o.modm o.s s.oHs as.su so.ma ofi ofi os ss
s.om o.Hoau m.ooo a.oofi- o.ooe mo.sfia oa.sa ma as on no as»
nos- a.sos- oésmfi Toma no? 349.. oo.mm.. i i: oo oo 53 3-03
s.ss o.omm o.s:mu o.o m.oH ss.eH: om.sau ma as os ss Ago
o.oom e.amH o.oau m.mon Hm.o so.os mg as so no .pz Hesse .ne<
m.ssm s.Hos s.s m.omo as oh so oo
o.omm- o.mafifi m.sH H.ms: ma ow os ss
m.Hem o.smH- m.maa o.smu ma as oo oo
o.soon e.osa N.Hmfl 0.2mm ea as no oo
s.om m.ou o.osu o.ssu oH as os ss .02
m.oms o.mmau s.aflfi o.os: ma as so on menus»; .fiee
os.o: ss.ooH mo.au so.sH a“ oH so oo
mo.o os.sofi ma.o oo.e oH oa os ss
oa.ms mo.soH so.m so.o o“ as oo oo
Ho.os ea.omu mo.e ma.sH as ea oo oo
oo.o: oa.m- ms.m ~o.m ma as os ss
sm.mo oo.sm om.s sm.oa o” as so on paws»: spasm
z s s m s
.pso .p3 .p3 .o: s a as m: smo-mufim emaummwm Amemnnpnmamnpmc
Haspoovsmeucoo .5.3<) scheme: 63 .9: spasm mmfim so .02 oocsu. $32 page
smcounmmoac%ludmne
.Uouaasoado

003 unwaox muddhmoh and: pump» nonponm ho oosmdhmboo 0:» 0H0SB madman Add
wzdnfiaond 00>H00 UohflmmOHo find U0hflp£wdmhpm How 00w090>0 0Hd0 3002903 oonmdhmfioo

.mfiH 0HQGB

- 48

mobamc cacao.“ umn

 

 

 

 

emanmmoso use conflunwddfiflu.uou numlflmhd ends cesspon oonmahwboo mo manoaomaoo

 

 

encepHosm u s .mswse u o .enocmnom a mo mm>Heo mama airs
oH~.o Hoo.1 msfi.: oma.fiu Hsm.o omo.u as oa so oo
ooH.o Hoo.o mma.o sHo.fl smd.o ooa.o oH ow os ss
smo.o ooo.o ooo.o mmo.o omo.o mom.o ma as oo oo
ofifi.o oHo.o ooo.H efifi.s mmm.o sem.o as as oo oo
oofi.o moo.s soo.u som.fi smo.o moH.o mg as os ss
oso.‘ ooo.o moa.o sso.H omm.o ooo.o oh 3” so no
a a a z m a
mmhm mso .xOAEu memew spoon 9h 0E Emmnmham Emanmhdm
unam pom .Hmo .sm meadows: oaoom madame: mcaam .JmmmamlwsHpmz
woosmaamboo : mo .02

.eopmasoamo pen
mes messes pscacmmae wsosm cosmaamboo 0:» opens madman on» msaesaosa mcbado

.DHH 0Hnt

- 49 -

mobamo oawaom

@

 

 

 

 

 

 

 

 

 

SHOSHHOSm nu N. .WSWQ¢ " 0 .UHOhmHmm N. W O mm>HQO QHQS nu EM
o.nnfifi fi.ofisfl mos.o msn.o oo.os on.on Azcs memwnmmowo
m.Hs s.mma msa.o mso.o HH.3H om.s Amos eoHoom mmwm9m><
s.nsfifi 3.3snfi moo.o mon.o ma.ma mo.ne A3c> memwncnwdswpm
m.mmm o.os “no.0 mso.o mo.o om.mfi Amv> eoaoom meanness
n.msOH H.nsofi oon.o sfio.o oo.on no.3: Azcs
~.Hsfiu m.oofi amo.n ono.o os.su Hm.oa Amv> os
N.onH n.Hssm nno.o moo.o nn.sm No.3: A3c>
s.oms s.nos: ooo.o msH.u Ho.om sm.ou Amos ns
s.mso H.sns ooe.o osm.o ns.nm Ho.n: A3c>
s.afio s.mns sso.o ssm.o mo... ms.mn amc> so
N.neeH o.asmfi sna.a oHs.o ns.an ns.en A3V>
s.mom- s.Hnfin moo.: sss.o om.mm so.o~ Amos no
o.Hns s.smsfi nmn.o moa.o os.oa os.oo A3c>
m.os o.mon sfia.o oso.o ma.oe oa.fiu Amc> sn
o.oooH s.somH noo.o Hoo.o oo.ms so.:o Ascs
n.0fiH- e.osH mm~.o Nno.o ss.o- oo.fimu Amos on
o.s0fifi s.smoH nmo.a smo.o ns.o~ so.nm gals
H.oHN o.os ooo.: amo.o os.o no.3 Amc> ss
n.msfia «.3smfi Hen.o sse.o os.on mm.nn Azcs
H.ssm m.ams mHH.o osfi.o sn.N sfi.nm Amos oo
s.HsNa m.snofi was.o meo.o on.eo oo.on Azcs
s.sofi o.oomn s:o.o ooo.o nn.o so.o Amc> nn

um, : m x mm m: cosmanm> sea I sham
.pz 9:003 .fidd choom \wcanmoz HQMA03 nundn mo 0onsom oomwu_wsdp0£
4. edema
.m0>aso cmnnmmOHo use connpnwamhum Hon newshobm deacon and momma Masada pnohomhdd
Hon AA3V>V mohdm manna: Una AAmV>v monam noosaon Hem monmdhmb Mo unaccomSoo .NH manna

- 5o -

 

 

 

 

0:.Nsfi 0s0.fi 0H0.0 «m0.a 0sn.s s.HmsN n.0oss Azcs meesnmmowo
0:.sm o00.0 o00.0 mm0.0 0om.u m.s0o 0.Nos Amc> umaoos memeso><
ns.onH was.0 sso.0 :HN.H nos.o N.ms0m n.nfisn sacs memwnpewsenpm
on.mm mfifi.o Noo.o mmm.o mna.« m.smc 3.0Hma Amv> ueaoom memehe><
00.ns sso.0 m~0.0 0nn.fi sHH.:fi H.Hmnm 0.saom A3c>
00.00H nnm.0 000.‘ oms.n :00.on o.:nm m.mmom Amc> os
ms.0sfi 000.0 mH0.0 anH.H mso.n s.osam 0.0sns Azcs
ms.nm- osfi.: 300.0 0HH.0 sso.fi s.mnmfi s.0sssu Amos ns
00.nfifi 0:0.0 nH0.0 nns.0 Nno.s m.nsnm a.ooaa Azcs
00.0 000.0 o00.0 ms0.0 0:s.0 s.ssfin fi.nonfi “00> so
s0.sHH s00.H HH0.0 mmm.fi nms.s o.msfim 0.0No0 63c>
os.nm Hmfi.u 0N0.0 HON.0 0H0.H 0.so o.mme Amc> no
0H.mnfi msa.s nm0.0 ono.0 :00.n 0.smmm 0.0noo sacs
os.os nsfi.0 000.: oss.0 ss0.0 0.smm H.0onu Amos sn
ms.nnm «om.a mm0.0 sns.0 mos.o 0.mamm s.s0om A3c>
Hm.«nn 0NN.- s00.0 fism.0 as~.0 0.som 0.Hs0~ Amos on
as.0ofi noe.0 om0.0 oos.0 on0.0H n.Hmam H.0mon A3c>
o0.o sHH.0 000.0 m30.- onn.- H.oom m.nss Amos ss
om.mnH sno.0 0fi0.0 sso.0 onn.o 0.noam N.nmnn Azc>
HV.H0 nn0.0 300.0 omo.0 nos.m N.ssfifi s.nosm A00> oo
sm.mnH sn0.0 sso.0 osa.m sso.0 m.0ssn 0.sosn A30>
HH.0m m0H.0 ”00.0 mma.0 snn.Hl| o.s~m- H.mneu Amos nn
3 z z z s a a
.p30 eeue .xodzp euenw ehoom .93 zeu .uz easedhes. sea I 09am
HeSpo< esonnsm pea meeoueo .sm wsHHnHez uomn .nus .sm .nus no oohsom emNpiwsHpez
sauna

 

 

 

 

 

 

 

 

.mebaeo uennmmOHo use uehppswaehpm you mewehebe ueaooa use mesh» wsapes psohemmdu how
AA3V>V mewae saspds use Asmv>v eeham seeSpen Hos ooseaneb he musesoaaoo

.A.psooc NH canes

_ 51 -

Table 13 contains the paternal half-sib heritability
estimates for the straightbred and crossbred steer and
heifer calves. and the genetic and phenotypic correlations
for the straightbred steer and heifer calves. The original
statistical analyses of the data contained a separate
analysis for each mating type and sex (18 separate analyses).
as well as the pooled analysis separately for each sex. The
k0 was very small (3.64 for steers and 3.33 for heifers). and
the degrees of freedom for sires was 13 to 15. therefore, the
standard errors of both the heritability estimates and the
genetic correlation estimates were large. making them of
little use as point estimates. The reason farcdassifying the
data into such small groups was to get the variance components
needed to calculate the genetic correlations between the per-
formance of the straightbred and crossbred progeny by a sire.
The pooled within straightbred and crossbred analyses for
each sex were useful for showing trends. The pooled analyses
still have rather large standard errors because k0 is small,
most values being between0.3 and0.4. The sampling errors for
the separate analyses were considered too large to Justify
publishing them.

Values greater than one are found in the table. It is
impossible for a true heritability or genetic correlation to
be greater than one. These estimates above one result from
sampling errors. They are presented to make it possible to
obtain an unbiased average of values. It was not possible to

calculate certain genetic correlations because of some negative

sire components of variance.

 

 

 

 

 

cameos H.Pu .eHes u so .ueHDmmOHo u an .uehnuswaespe unme
sn.0 sn.0 0n¢0 0fi.0 -m0.0 30.0 No.0 0H.0 an.0 s sunfifineamo
no.0 u He5p0<
no.0 oo.o no.- mo.- oo.- 03.0 3m.o NH.o om.o m emse
00.0 0 eso-nsm
3m.- sm.- mm.o 3m.o mm.o fi3.o mm.o om.o Ho.- m mmesxcacp
:m.o 0 pea
03.- sfi.- so.0 on.0 no.0 H0.0 3s.0 «0.0 mfi.- m meeew
mo.o o .900 .9&
03.- 0m.- 00.H no.0 03.0 0H.0 HH.0 sH.0 0H.- a 00000
nfi.- 0 .nwe:
.
oz sec .03
a, so.0 00.0 mfi.0 0H.0 m0.- an.0 as.0 ss.0 3s.0 30.0 3H.0 0o.0 an.0 m sme-onm
_ 00.fi on.0 0 .Aav .0:
.5 .203
m0.s 00.a 00.0 00.0 nm.- 00.0 3H.H so.0 0H.0 om.0 00.0 3o.0 an.0 m .0:
00.0 3m.0 . .2003 .000
0n.- sH.0 mH.0 n0.0 so.0 s0.- 03.- 00.- 03.- nH.0 03.0 00.0 so.0 m mwoom
so.0 s3.0 0 .2003
ss.0 30.0 sn.- 03.- ms.- 3s.0 no.0 30.0 H0.0 00.0 3m.- s3.0 no.0 mm .03
:1 3s.0 o3.0 no opwsm
z
2 seem a e z m z s a s 2 em as wmmmwe pawns
.030 esc.x00sp cues esoom .p: .03. .393 .secz encom.seex .ps spasm
Hessee -an pas .seo.ss .nsez see-0nn .ee .mue .000 .

 

 

uHMHB

A.Hesoweau 0:» so puma 0:» on one escapeacnnoo oucspososa use .Hesoweau

on» so pawns can on one escapeaehnoo capesem .Hesowedu on» so sewage meapaaanepanemv

.mcbaeo scenes use scene uesnpswaenpm on» you escapeaeaaoo cacsposesa use each»
nosew use .me>aeo Amado: use scope uenpemono use uehnpsedehpm 039 new medpdaanepdhem .MH canee

 

-53...

The numbers available in this study are too small to give
meaning to point estimates. Tallis and Klosterman (1959)
found that the number of offspring per sire played a very
important part in determining sizes of errors of estimate.
Extensive data are needed since estimates are subject to large
sampling errors. They have published a table giving the
minimum number of animals needed to estimate heritability with
a certain standard error. In referring to this review of
literature heritability summary. for example. it is found
that traits in this study have had average values from 35%
to 60% when studied by other research workers on large numbers
of data during the past 20 years. Using weaning weight as an
example. and referring to the table. it is seen that to obtain
a standard error of0.05 would require about 306 sire groups,
13 offspring per sire. or about 3.978 offspring. A standard
‘ error ofcxij would require 35 sire groups, 13 offspring per
sire. or about 455 offspring.

Tallis (1959) has also prepared a table giving the opti-
mum sizes of progeny groups for the estimation of genetic
correlations with (n':>1000). An example using this table
will be the genetic correlation between weaning weight and
final weight (genetic correlation average of0.72 on 9 esti-
mates quoted in this review of literature). Tallis'
table discloses that the optimum size of progeny groups would
be about 7. with about 143 sires. making a total of 1.000
offspring.

When the data were pooled within sexes for the straight-

 

- 54 -

bred calves. all of the heritability estimates and most of the
genetic correlations are within the range of estimates re-
ported by previous research workers. In a number of cases
they are quite close to the average estimates reported in

the Review of Literature summary. Very few previous esti-
mates of genetic correlations are available on the carcass
traits. The phenotypic correlations are in reasonably

close agreement with the summary by Petty and Cartwright

(1966).

iunlrrrrs~
. I.

This study did not indicate any differences between the
heritability estimates of the straightbred and crossbred
calves by a sire. indicating that selection should be as
effective in the crossbreds as the straightbreds. This was
in agreement with the only other study in the literature
comparing heritabilities in crossbred and straightbred beef
cattle (Miquel and Cartwright. 1963). They concluded that the
estimated genie variance and heritability ranked in the same
order for all characters with one minor exception. Bassett
and Shelton (1966» working with heritability estimates and
genetic correlations between straightbred and crossbred lambs.
found only one instance where the crossbred lambs provided a
higher estimate of heritability than the straightbred lambs.
Louca and Robison (1967) quoted higher heritabilities
for purebred pigs than fru' crossbred pigs in the traits
they studied. but the heritabilities given by Enfield and
Rempel (1962) for swine data did not seem to indicate any
definite trend.

- 55 -

In order to obtain the correlation between the genetic
ability of a sire to produce a given trait in straightbred
progeny and his genetic ability to produce the same trait in
crossbred progeny. and the correlation between the genetic
ability of a sire to produce one trait in straightbred
progeny. and another trait in crossbred progeny (and vice
versa). the components of covariance between the progeny means
in tables 11a and 11b were used along with the sire components
of variance in table 12. The method used is discussed in

the section on Statistical Methods.

 

Tables 14 and 15 contain these correlations for the
steer and heifer calves by mating types and pooled by sexes.
As discussed earlier. k0 is small. and the degrees of
freedom for sires is too small to make useful point
estimates. but the values obtained should indicate
trends.

The pooled values for the correlations are high in both
sexes. with the correlations being higheriru-the steers than
furthe heifers. Out of the 17 pooled correlations. possible
to compute. in both sexes. about half have a value of 1.0 or
greater. with several others being higher than(18. This
indicates. in these data. that the correlation of a sire's
genie ability to sire straightbred and crossbred calves is
probably high.

Since so many of the estimated correlations among the
genetic values of the sires were above one. it seemed ad-

visable to make other calculations to determine if the pro-

HIHDI‘H! 3..“ I H. I

snosphosm u s .eswss u o .unouesem n no
.eoseaheb no usesossoo made ebapemes he emseoen uepeHSOHeo nos seapeaesscc umre

 

-35

 

00.H 00.0 s0.0 sn.0 m 00.0 0n.H n3.H H0.H 000000
afi.0 o0.0 m m m 00.0 00.0 a0.3 m oH os ss
0 0 as.- m m m m m m 30 ns ss
sn.0 00.0 00.- 00.- ns.- 03.0 00.0 0H.- 3n.0 oH so oo
3n.0 0 0H.H 0n.0 so.0 0a.0 0 0oyH no.0 30 no 00
3n.0 s0.- 0 00.fi 30.H 0 m 00.0 m 30 sn nn
0 0 00.0 30.0 o0.0 0n.“ 0 n3.0 00 3a on nn
gun sow-1‘ A00 000 000 A00 [000.00 000
.puo ecse .xQHSp cuesw enooe .03 .sees 09000 00900 sensenam seQIeHHm
Hessca mse-nam pee .weo.ss .0002 .ee.00< .000 .seoz A00.03 co .oz ommup masses
A00 ,00 A00 ’00 A00 A00 A00.ss A00 20000
.020 eene .xoacp cuesw whose .03 .sees ehoom Amv.us

H00000 000-000 see .000.20 .900: .ee.s00 .000 .seose spasm

 

 

.Aemnob ceab usev mueanmeono esp

s0 aespose use eueanpswaespm esp s0 uses» use you use .paenp esee 0:» Hon Acv ssemona

peeve uesneeono eosuonm on spaadne capesem Hams» use Amv aseMOHa Reese uenppswaenpe
eosuona cu menam causes weep uesnpswaeapm so hpdaape capesom esp seespen escapeaeasoo .3H eases

 

30.0 00.H o0.0 000000

 

 

. 00.0 00.H n os ss
mu m m 0 ns ss
. 00.0 3s.0 oH.0 so oo
03.0 s0.0 3H.- no oo
0 m 0 sn nn
30.0 m 0 on nn
A0 0 00 -0 0 00 so 0 00 sen-00am sea-00am
.ps .s-H .ps .0.“ .03 .s.“ ems.» gape:
.000 - .333 .000 - .03 .30 -

-0000 000000 .0003 .004 .03 spasm

 

 

.Aeeseb eoab usev euenpeeono esp s0 Hespose use euennpsmaeaue

exp s0 paehp eso new use .paeap esee es» Hem on asewOHa Heepe uennmmono

eosuons on spaaane capescw sacs» use any hseMOHs Heepe uennpswaeape eosuona
on cease causes ween uennpswaenpm no mundane capesew esp seezpen escapeaenhoo A.psoov 3H eases

 

 

 

 

sHOSpHosm u s .eswss u w .uHomesem u no
.eose0ae> so musesomsoo 000m e>0pemes he eeseoen uepeHSOHeo nos s00peHeHHoo u me
00.0 30.0 00.0 n0.0 3o.0 s3.0 000000
n0.0- 0 03.- m 0 0 n0 os ss
00.0 00.0 s3.0 n0.- 0 00.0 n0 ns ss
. 00.0 00.0 03.0 00.0 03.0 m 30 so oo
my
. 00.0- 00.- 0n.0- m m 0o.0 30 no oo
0 0 m 00.0 no.0 0n.0 n0 sn nn
0n.0 00.o m 0 00.0 mm n0 on nn
“0 0 00 00 0 00 A00 .0: 000
.03 .03 seulonn ADV .uz chose 00v meh0e
000-0nn 000-0nn .000 - 000 .000: .00< .0003 - .0: 00000 00 .02
.00.... .0: .000 - .0: 000 - $0 .3. 300.88 A00

.0003 .000 .03.:0000 -0nn .000 .0003 .000 .0003 .03.:ps0m

)||||

 

 

 

 

 

 

 

 

 

 

 

.Aeehe> 000> usev muehnmeOHo 0:»
s0 Hespose use mueanpsM0ere esp s0 00enp eso Hon use .paehp esee esp new 000 hsemona
00000: ueaneeoso eosuonm cu sp0H0Qe c0pesew 000:» use Amy asewona 00000: ueanusteapm
ecsuoas on 00000 cappeo seep uehnps00enpe so zpda0ne 00pesem esp seespep ms00peaehnoo .00 wanes

-59...

cedure followed had a bias in it. One value which could be
used is the phenotypic correlation between the averages per sire
for straightbred and crossbred offspring. The first step in
this procedure was to graph several of these to get a visual
appraisal of the relationships involved.

Two examples of these are sire progeny means, by years,
for adjusted weaning weight in the steers {Eu‘ Shorthorn sires El?

bred to Shorthorn and Angus cows and similar means.

 

by years, for adjusted 550-day weight in the heifers tku‘
Angus bulls bred to Angus and Hereford cows. These are
shown in figures 2 and 3. These two relationships were
selected as examples since they represent one of the largest
positive correlations. and one of the largest negative corre-
lations in traits.

By visual analysis. it can be seen that the correlation
in figure 2 is highly positive. and that the correlation in
figure 3 could give us a negative value.

Phenotypic correlations were not obtained between the
progeny of a sire bred to cows of his own breed to produce
straightbred progeny and to cows of one of the other two
breeds to produce crossbred progeny for all traits used in the
study. but the two examples above were done on a hand cal-
culator for illustrative purposes.

A phenotypic correlation of<162 was computed for adjusted
weaning weight between progeny means of straightbred steer
calves from Shorthorn bulls bred to Shorthorn cows and progeny

means of crossbred steer calves from the Shorthorn bulls bred

for the trait adjusted

by years.
weaning weight (in pounds) in steers. Shorthorn Sires

OX".

. -.-_.. __
T
?

Sire progeny means,
I?)

1

L

1

I

Figure 2.

 

moo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

4 . . . . . . . w 4 _ 4 m . u - M.
: ; . : :. ; I: .: -, .- ..
4 _ 4 . . ._ _ w . . _. . ._ mo
1...--.- 1.1.. .. .1. - _. .-
m m U . . . . s w c. . . — .OII ...I I...
I4... .3. . -. --.-.,.-+.-.I-.-_.:. .... .3 . . .* u . . s
4.. . T... T. . .; - . .-..- a.-. -... ... .-.--..-.-...--T ... 8
.II - -...r2-.. I I ...-.-_. _+_- . I -.1 .+ -w L14: 4.- .. 1%-L.-I Io IL-IIVII TF1... 1.... .- -. UO
...-. .4... .....__..4._. :1-.. ._ so
. -- .. z j... -3--. . . . . ..
. . _ - . - . . I
. 2.141.. 2..-. ._.....-..-.._. .. . I . . . . . . 3. s
. . . . . -o -.: I. I. I
I . . r. .. f...- ......44-.--. . . . . m .-
{M II II . “II I 11!.” . V IL .OI I. -1I - -4. - ”I «.I--.I o- P 1.101-. o h: -4.- .MV- , I HO
0 .. + + . __ ._ a a 4. ~ .. 4 . m M 4 . ._ T
MIFI IFI e I--MI-.§IIIFIIQ $.16- If )9 .o. +IK.4 Iw.I‘ - § I- Q-II.*II Q.....0.-1.§IIO.I- *II AMI-.430...- . b-WIIIIMPI I‘ moo
” . . . .-_ .- ~ .09.. H . . . . n m . H11 >
m IIOI .» - +II o I * .-.§II I. I. II....-4..* I4.-- -MI%IIIWII.WII4IIII lll’IIM-IIIMIII‘I. 00
. . . o . I . . *
IE1 I...-I+ I... -4- «II-.11 .IIII.I-_-I+-I.1..4ir.-.r+1-1..-I... -.1...II..I . mo
, _ . a l _ .. . .. .. u H . . .. . . II. II.
“.111. IT +- kul-IWY- o WIIW. ”If-O. e .o W-II-WII”..- V o .fiI :6. I”. e .s 1+. .+II4.:I .fiI. 9.1:“.1..MI| . 111+. 4&0
ST. I .h I 4 I _ III+IIIMIIIOHIIIIQ III-IA-JH~.PIII~II-b1l w. IPIIIO . .14ri 9| ILIIIA IF II IHIII1IIQIIPIII+IIII I O. IIrIIIhlII I4I .- mo
. - . .
m....-.-.-..fi . . W . . . ._ ....-....,.1.. . .-........_--.-.... .-.-._I.-:._.-..I.. 1.- 1:11. do
mrfit H * w o .w * e. M .. -. w. a o W m 0.... o Wa- 6. o :4LW 4 .+-...: IWIIvl..I_.1II..o. .ImIITI-L to
.1. . _ . . . . ._ N .
ditr II b #11? III}! I+QVII+I fII+IIW .IQIIJIII‘ILI III+I .IIO-I ”II 11 filIIIIIWIIIfllp. mo
. . _ n _ . _ .. . . n . _ .
wTIfLIILIfiII+ IIHIIWIIIAII QI- IIIOI.IM+ .e IH .I‘gI10hV‘vwd III 4 +1.1 1 Ir . 4 .TI‘IIWIIINIIIA NO
)_{1_ I-IHI-IWIIVI- .9 I? II” It“? ITIHIII. III LII.” q oIIIQII b I’IINI. .J.OI.1ImI4 +1 .._f.III+IIIWII «III-f- IN}- 1 1.1 fl .1 T A H. O
. . _ . . . . . . . . _
91... ..... .. .9-- . . . . .. 1.2-401... 14....-.7-4-..41.11:-.-. .50
n .11- IIIOIIIIIQIIOIII+III+IIPIIIVII§II+III I. FIIHI o 4 -wI..4 eO-W a; + w .wI 4. ..*.3.5. ...-.I4I.WI 4 mo
. _ _ . m . . ._ .I I .
MIV. 1 fir III IIIIII+I IQIIQI III ILII- 4.1....va '- IILIIIMI .IIIIII+_.II If .IItIII III-VIII: .YII-4I. #1 I+II..QIIIOII I .I Mr _ mo
1“ L: I I IfI-Io. e v .9 11:... . . o ...r A. III-I9 .1“... .I aI. ._+.III_Q:I If .IQI .f-+a- .4lIIIILIIQ-I4il. 11.11 ‘NO
t M a . _ . H . n _ . ’iIfII. mo
VOAWI* O I . . .— w . o 4 fl . WI.-.V 4 0| .4; 6 0.. I C 9 . {+I :0! .I-OIOIIJII'_ .
e1- . .L: .-. +1.. 4 . .. T- .2... :1.-. - - . s
.I__ -+...- II- “II-III- IrIII... 1111.1»..14 LI r.-. . . . -..I-. .. . . ..I I1 IV 4 to

 

 

 

-1..-

.. . . a HIII-1.1.1... _I+.Iw.+._+1I.-.-I.+:+1+--1 -.I...--+ . 114141-4141 mo
_ . _ . . . P r . +

H1.-. 1 1.. I. -1. +I-IILI-4-+I. ._ .- Wfi No

. .21.-4* 1.7.1-1-?111 . 1 4-4,-_--..----.:..; . .. .._ 4...... so

 

 

 

 

 

 

 

 

IFI+:.I:E%1.IHI+:ri+IIt+ %!v2-- .2.. a -s.-fl+-. moo

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- 4 .
. m . . .
A IIUIIAII III“. IAfiI 141 .TI IIIATII.I-ILT . M? v b . + -4 1.41:4 1.1-: +- . «1M 00
. . __ 1 411-....- .F mo
_ TECH -4 . . . . m3
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 0
OO/Ea/654AJ210987/Or3h4.3210987/05432109m7
/o -t . Q, .4 1J ac
rss )
60u8d64ann
7hmrwoer0 a
S rriree
(AOIDSDLY-Ivm

77 (Shorthorn straightbred
sire progeny mean)

for the trait adjusted 550-
Angus sires (6) on

- 61

-

~ progeny means, by years,

V

I

day weight (in pounds) in heifers.

Sir

Figure 3.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(6) and Hereford (5) dams.

 

 

 

 

 

 

 

 

 

 

Angus

 

 

 

 

 

 

II I-I- . . . . . . _ 4 . --. 4 - _ m .- 4 - .M.. I III- 2- . . I. w 4. a .wI
. . . . . n . .
+I+II.#:+-+-~.--.-. 4.. +.e-+ * w w . . . . 4 . g-.sg . 4 .;+II-4 to
. n . . _ . . . .. _
4.1.11...th 4. I”. . . . .r M . 4 - .wlf- h I .I --+ II.+-1II-I 11.1.4...- m 4-4+II4... 1 I .1 m0
. .. . + n . . . w . . . . . I -. .. PW: -__ I. . .I . H L A V .IL NO
WIII..+I€I+IwI+I4-4-It-:IIII_-4IWIMEL-. II .1154 ..4 1.. - - my“ . 1.4 H
4I4I+I+I.Iw:+-+I¢I4InI4III.III .43.-. r I-I-+I-I+II-IIII-11l+m4ééomwFm --Il. Ho
“ _ . n n _ 4. . ._ w _ . . t
.ILII+;. 4 .- . h..-. . .. I-. . + 4:...I-I-Izp-et-wII- 1I+.;.trI+P+52E7thyg--..4-Lono
m _ . . U. .. n "- .. . . + u 3. . . . t llhlt . m
«III-II H- h... r 9 w .44. v H-» .I .4. I A I -I* -IF IATIILII: .1 fIIIvI .aIw111 lJ.r11.I -_ Ifiry. 2 =1: fi-IIIIIV I114. mo
. _ . . . . . 4 . . . . . . . _
. “-.--I . 9 M u 9 ~ H ~ 4. a ..... W- + I» f. ”I I... 4 I .L-_.. I .IPI/ o.- 4 +UO» IhIIR T IlrI. Inf-I..fIIb-III¥II.I~ mo
_...._ ..... -. .. . .. . ... .9123 _
.1.-.- . . 7.1.-.. 7.- . . . . .. . 41.1.2.1. no
“III I* h m h 4 w M d H .. . . . p w h w 4 H I "- o v .- h .v I I .91. T III L-Irl-FIIIIII. 00
u _ . . . .. .. . _ . . . fl . .. _ . .. _ u
H I. I:I.PI..+ -. T . u I+ N ”-.O H 6 V w v .+ 9 v .o .r I. VIII Hf.- hflfIIrIDL Irl IIVI 1+ IMI 14.1 III A. mo
“ F -. .- . . .. 4 . _ . . . . . _ . . v _ 4 . .-
.--_.-.,. .....,3...-......--- I. ._- . . . . . . _. -. . -_. I .. . .1. 1.. .5
_ . . . 4 . n . . - . . . . . .
.I+i+4 ... . w .. . . .a..+.fi .LIM-I I- ..- 1 +. .14 41w11. wo
1.1....- . .g. . .1... . .. .- . .-.. .-_..-...-.-...-._.-I. .8
u _ w . .. w . n . _ . .. .. . u . ._ _. 4 g * t .. .
.++.+++...+-r,-++.%a4..-.-....T-rir:11.;
_ n _ _ W _. H .. . . . .IIF . _ p .. I r .1. w .. F mOO
. . .. . .:1.I4-- I.-II- . 4 4 . 41.. 1 - - .
_ . u . . .. _ . . .. Z . I I:- . L 00
. _- 1 14 H1 L 11411.1 1 I L1.- I.1 I_I + I1 I1 I 14 14 ..I 1.. -
.11. . w.:.I.-+LLI.II.1 .- .- r21 Irrww- - L . mo
.1 _. . .4 .1 -_1 _ 1H _ . _ 24: . .— 1~III "1| .- 4 _ .- 1+
.r 1 1” _1 4. 0 4 II4II0III+II1 I. -.I...II.L.II-I+.II«.I 431-.» .I+II. +:.. -».-..II.. III.- 1.4.IIO- IIFII JIIIIIILrI-Ie-I %.I-r.-IAYII+IIFIIMIII+IIIO V0
. . _ . . . . . . . _ . ~ . . . _ . .
. . . . 1 _ It- - I . 2.. .1 TI: I II I I I . I III I I L11 II
.. .............-._-...-..-.-._...:.._.-..I 4.1.1... . .. .8
. 4 4 1 0 1.11119 +I II I. r- ell-f-L II.O.I .I+ IITIILUIII .TII-.P.I IT- .I. I.I..b.II .OII. .4.I-II0I.I IlkIW-IILTIIWIILYI-AI .m- w .0 mo
. .. _. . w . . . . . _ 1+- . . _ _ _I . fl . . _
.. 4. -_ . ._ I. - .1. . 2-..... 41.-I.II.--.--.-. ..4 . . I. .. .. . ..:. so
_ . _ . . _ . O. .. . .. . _.
., ..-+I-. 4-.--..,.-. .. ..._ . . . 4 . . .0.-. -. 7. _ .. .11..-. mo
_ . u _ . . . F _ H. . .
I4: - 4. I y F - + . ”I 1* - H g I 9 9 «I W . :9 . .4. v1..- Ar . . _9. Im. iw‘llfI! NO
. _ . _ n4. . . _ .. . _ . .. _
W.- +IIWI11IT.IWJ|*.I.IIIVAI.VIIWI..h9.l).+..!.w. ibuIés-ILT .I..T$a .1. «I 4 HO
+IJI 11+:- +1.19 III-IL . 1101-011 I. 1.71.73 I I VOO
. . _ _ .
- . ...- VJ- . .L..- - H 4 gm.-. gel-.11-.-- mo
_ .
.. . . .-.,-.-..-..I4-...I+.- III-1.-.-.. - - .. -- Isl mo
. H . . . . :F .. _ . . . _ . L: r I
IIIWIIIL 1+ I _ 4 I +- 15 . . -41: 9 . . I LT i7 . _ II ..+I I“ I6 IUTIIILI IfiIIfltI I1 I “O
I.- .. .r I. .I.---HI-L_- -1.--- 4- I411...- I 8
_ u . .
IIHIIIILII I: FIII I .49: I I .f -F I4 +. ._I: It: I f I- I; .4 III+-.II. I II+ I 1% _ I I I441. -41I4rll LWII a mo
- H _ -

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. _
- I+II*II.I+ III->21 VIII+II+£II I... I: II I... II 3...!!- I 4TI$IILTI¢IIII I “II; to

 

 

 

I .
H-
50’ -

 

 

 

progeny 80
7o
60

nean)

66 (Angus straightbred sire
progeny mean)

-62-

to Angus cows. Likewise. a phenotypic correlation of -.29 was
computed for 550-day adjusted weight between progeny means of
straightbred heifers from Angus bulls bred to Angus cows and
progeny means of crossbred heifers from the Angus bulls bred
to Hereford cows. The phenotypic correlations were product
moment correlations among the sire averages.

In order to check the model and statistical procedures
for determining the correlations involved when a sire is
mated to several cows of his own breed to produce straightbred
offspring. and mated to several cows of another breed to pro-
duce crossbred offspring. with an average being obtained on
each of the progeny groups. a path coefficient diagram, as
shown in figure 1. was drawn.

The methods of Lush (19h8) are used in computing the
phenotypic correlation by the path coefficient diagram method.
As illustrated in figure 1. the correlation between the genie
value of a parent and its offspring is one-half. For the
heritability estimate for weaning weight the one chosen was the
average ofCL35 from table 2. This is the best point esti-
mate available at the present time for beef cattle. The
value of the path between the phenotype of one offspring by a
sire and the average phenotype of several progeny by a sire

is given by Lush as:
1

— WEI + (n-1)tj

where t _ the correlation between the phenotypes
of individual offspring by a sire

(t = k of heritability in half-sibs)
number of offspring by a sire.

X

 

 

:3
II

 

- 63 -
The paths in figure 1 show the relationship between the

observed phenotypic correlation and the biological cause of

the correlation. This relationship is as follows:
2 2 2 2
r--=n I‘ XI‘
POpPoc [} GsGo) h Gsstc

the correlation between the genic value

of a sire and the genic value of his

 

offspring.

SI

3"

(D

'1

(D
~51
wl
II

h = heritability
n and x = (defined earlier)

the correlation between the genetic

*3
C)
C)

II

ability of a sire to sire straightbred
offspring and his genetic ability to
sire crossbred offSpring.

If the correlation between the genetic ability of a sire
to sire straightbred offspring and his genetic ability to sire
crossbred calves for adjusted weaning weight is equal to 1.00.
as this study would indicate. working through the problem
gives a correlation of 0.21. This can be defined as an esti-
mate of the expected phenotypic correlation between progeny
means of straightbred and crossbred calves by a sire for wean-
ing weight. Since this value is only about one-third of the
product moment correlation computed from the actual progeny
means. it indicates that the sampling errors have made this ob-
served covariance among sire averages different than the true

covariance.

-64-

This adds confidence to the belief that the values in tables

17 and 18 are not above 1.00 because of mistakes in. developing

the procedure to estimate them. rather.than a procedural bias.
The statistical analyses of the data used in the experi-

ment, to measure the effects of heterosis. were explained

by Gregory e3 §_l_. (1965, 1966a. 1966b, 1966c). These calcula-

tions were not repeated in this experiment. but a summary

of the appropriate parts will be discussed since their re-

 

sults have a direct bearing on the results and discussion of

men~ *-'-:;....a

this experiment.

In Gregory's analysis of variance. years were highly
significant (E’<:.01) for birth weight (heifers), weight at
200 days (steers and heifers). weaning score (steers and
heifers). 550-day weight(heifers). rib-eye area, carcass
grade. actual cutability, fat trim. and adjusted final weight
(steers). All interactions involving years were relatively
unimportant. Gregory interpreted this to meanlthat. within
the range of environments of the 4 years at one location
and within the genotypes of the breeds of sires and dams.
genetic-by-environment interactions were not important for
the traits studied, however, year effects were highly signi-
ficant. In this study sires are nested within years.
This nesting prevents the sires from interacting with years
(since a nested factor cannot interact with a factor of a

larger classification).

-55..

Gregory's interactions between breed of sire and breed
of dam were significant (P <.01) for birth weight and 200-
day weight (both sexes). 550-day weight (heifers). final
weight. and rib-eye area (steers). Gregory stated that
these significant interactions reflected the importance of
heterosis on these traits. The heterosis effect was highly
significant (P<<:.Ol) for all pre-weaning traits (birth
weight, weaning weight at 200 days. and weaning conformation
score) for both sexes, adjusted final weight (steers). 550-
day weight (heifers), and most carcass traits associated
with or affected by weight. The effects of heterosis de-
creased with increasing age. Interactions of sires with
breed of dam were not significant for any of the traits
studied. indicating that the difference in heterosis effects
between sires of the same breed were not important. When
Gregory compared this interaction with that between breed of
sire and breed of dam it indicated to him that heterosis was
due to the effects of the breed of sire, rather than to the
effect of sires within breeds.

In the analysis of variance done by Gregory. difference
between sires within breed of sire and year were highly sig-
nificant (P’<:.01) for birth weight and weaning score (steers
and heifers), and significant (P <1.05) for ZOO-day weight in
the heifers. In the 1962-63 heifers sire differences for
550-day weight were highly significant. In the steers, final
‘weight. actual cutability. and fat trim showed highly signi-

ficant differences among sires. while rib-eye area and carcass

 

-66-

grade showed significant differences. These sire differences
within breeds indicated to Gregory that additive genetic
variance was important on the traits studied.

Gregory gt §l° (1965, 1966a. 1966b, 1966c) have given
least-squares means and constants for breed of sire, breed
of dam and interactions for each sex. and means for sexes
combined. estimated by an analysis of least-squares. For
this reason the means for the groups shown in their paper are
not listed here.

In this study. means of progeny of individual sires and
standard deviations within sexes were computed for each of
the sires when bred to cows of their own breed and to cows of
each of the other two breeds. Allowing for certain missing
cells. this added up to 137 averages in the steers (on 9
individual traits) and 138 averages in the heifers (on 4
individual traits). Partly because of the size of this table
(1785 averages for individual traits). and since it does not
directly apply. in itself. to the final results. it was not
included. These means (If progeny of individual sires were
.computed for use in obtaining the covariances needed to esti-
mate the correlation between the genic values of a sire for
'producing straightbred and crossbred calves, discussed

earlier.

 

APPLICATION

Basic knowledge on genetic parameters is necessary in
the improvement of livestock through better breeding and
selection methods. These parameters are calculated from a
particular population at a particular time. They may or may
not be the same for another population or another period of m!
time. h"'

Even though a deliberate attempt was made to obtain '

 

heifer calves from several sources and an attempt was made K
to obtain bulls from different sources, the sample used in a
this study is not representative of the Angus, Hereford and Short-
horn breeds. Also, the breeds used in the experiment cannot
be considered a random sample of all beef cattle breeds.
Thus, all statements made in reference totflmme data. cannot
be expected to hold true for all beef cattle. Also. as
mentioned earlier. the average number of offspring per sire,
and degrees of freedom for sires are too small to give
'useful point estimates. All comments must be on the
basis of trends.
The paternal half-sib heritability estimates computed
in this study agree reasonably well with the average esti-
mates summarized in the Review of Literature. These esti-
mates represent many thousands of cattle. and over
100 studies during the past 20 years.
The heritabilities in this study ranged from 0.15 to

0.85 with only two of the 13 values below about 0.40. Values

- 67 -

-68-

for these traits from this Review of Literature summary
ranged from 0.23 to 0.58. Research workers have concluded
that the heritabilities for the pertinent traits are high
enough for selection to be reasonably effective (Gregory.
1961). Results from this study agree with this conclusion.

In this study. there did not appear to be any difference
in the heritability estimates between the straightbred and
crossbred calves. This was in agreement with the study of
Miquel and Cartwright (1963) discussed earlier. Also. there
were no apparent differences in the estimated sire components
between the straightbred and crossbred calves. This informa-
tion would lead to the same conclusion as Miquel and Cart-
wright made when they concluded that selection would be
roughly as effective in crossbreds as in the purebreds
(straightbreds in this study).

In general. the pooled within straightbred analyses. by
sexes. indicate high genetic correlations (0.64 to 1.23) be—
tween the traits pertaining to weight. rib-eye area. and
actual cutability. The phenotypic correlations follow the
same pattern except for the correlations between rib-eye area
and the other traits(0.26 too.59). Weaning score is nega-
tively correlated genetically with all traits pertaining to
weight and actual cutability (0.0 to -.50). and is slightly
correlated with final carcass grade. fat thickness. and rib-
eye area (0.13 to 0.25). while being more highly correlated
with marbling score (0.69). Phenotypic correlations of

weaning score with the other traits are also low, varying from

 

- 69 -

0.07 to 0.36. with the highest values being between weaning
weight and weaning score. The carcass traits. marbling

score. final carcass grade. and fat thickness are highly
correlated genetically with each other (0.61 to 1.00). nega-
tively correlated genetically with birth weight. rib-eye area.
and actual cutability (-.2U to -.73) and slightly correlated

with the other traits for weight(0.09 toO.23). Phenotypic T1_
correlations for these traits follow the same pattern with

negative correlations (-.01 to -.12) with birth weight and

 

rib-eye area, and moderate correlations with the other 5'
traits.

One of the main objectives of this study was the esti-
mation of interrelationships of straightbred and crossbred
progeny performances in the same trait. and between one
trait in the straightbred population and another trait in
the crossbred population. These correlations between the
genetic ability of a sire to sire straightbred and crossbred
progeny for one trait. and his ability to sire one trait in
the straightbred progeny. and another in the crossbred pro-
geny. were high. with many being greater than one. Between
one-third and one-half of the values could not be computed
because of negative sire variance components. Values were
higher in the steers than in the heifers. No explanation can
be given for this. Only 13 out of 108 correlations were
negative. In the steers only 10 out of 72 individual corre-
lations. and 2 of the 12 pooled correlations were belowCLSO.

In the heifers. one-third of the 36 individual correlations.

- 7o -

and 5 of the 6 pooled correlations were below 0.50. No trends
among correlations were noted on the basis of specific mating
types. The only pooled correlation low in both steers and
heifers is the one between birth weight and adjusted final
weight. It would seem that traits measured in both sexes
would need to be low in both sexes to be significant as it

is hard to visualize that there should be any biological
differences between a bull's ability to sire steers and
heifers. The correlation for fat thickness (measured only in
the steers) is also much lower than the other values.

This study indicates that the correlation between a
sire's genetic ability to sire straightbred and crossbred
prageny for one trait. and his ability to sire one trait in
the straightbred progeny. and another in the crossbred pro-
geny is high. Thus. a sire that sires good straightbred
calves should also sire good crossbred calves (and vice-
versa) when bred to comparable cows.

Mass selection has been the most common method for im-
proving livestock. This is the method in which animals are
selected on their own phenotypic values. Only the production
traits (birth weight. weaning weight and score. and final
weight) can have direct selection applied to them. Other
traits (carcass traits) require another system. such as sib-
selection.

By selecting calves with heavy weaning weights and final
weights. it should be possible to produce heavy cattle

which make rapid and efficient

 

‘Egigl_,imc

- 71 -

gains in the feedlot. and will produce a maximum amount of
edible portion per unit of carcass weight. Indications are
that adjusted weaning weight and adjusted final weight would
be the best criteria to use in mass selection. The associa-
tions among the traits concerned with weight are high enough
to recommend that simultaneous improvement in all of these

traits would be possible.

 

Gregory §£_gl. (1965, 1966a. i966b. 1966c) have

analyzed the influence of heterosis in the cattle used in

 

this experiment. In general they found that crossbreeding E!
has resulted in about a 3% increase in growth rate up to
about 18 months of age. in favor of the crossbred cattle.
Heterosis seems to be important only in carcass traits that
are associated with the growth of the animal. Heterosis has
been measured by comparing the crossbred offspring by a
sire with the straightbreds out of comparable cows. The
difference between the crossbreds and straightbreds is due 'ro
non-additive gene effects. Other studies (quoted earlier)
have found similar advantages. in favor of the crossbreds.

If future experiments continue to observe heterosis
in beef cattle. it is probable that the majority of the
commercial cattle production in the future will be made up
of crossbred cattle. If this is the case. one of the main
questions will be : Will practicing mass selection in both
sexes of both purebred populations that make up a cross be
effective in improving the performance of the crossbred off—

spring? This question is one of the most important ones

_ 72 -

trying to be answered in this study.

Results of this study indicate that:

1. There are no differences in the heritability esti-
mates between the straightbred and crossbred calves.
thus the effectiveness of selection would be
roughly the same in the crossbreds and the straight- E!
breds. ‘

2. Heritability estimates for the traits studied seem

 

high enough so that mass selection for them should «-*
be effective. 5
3. The associations among the traits concerned with
weight are high enough to recommend that simul-
taneous improvement in all of them would be possible.
4. Indications are that the correlations between the
genetic ability of a sire to sire straightbred and
crossbred offspring for one trait. and his ability
to sire one trait in the straightbred prOgeny. and
another trait in the crossbred progeny are high.
Thus. it can be concluded that simultaneous mass selection
for traits associated with weight in the purebred popu-
lation should be effective for improving the crossbred
population.
The rate at which the crossbred performance can be
improved by mass selection in both sexes of both purebred
populations that make up a cross can be expressed using the

following formula:

-73-

mean in the = (I‘G Gc‘v th O/GC(z/b))

crossbred p
population
where:
rG G = the correlation between the genie value
p c

of the purebred and the genie value of
the crossbred. n

h p: heritability in the purebred population.
7/b==selcction intensity.

 

0’s = the standard deviation of genie. values of g

the crossbred population.

If the correlation between the genie value of the pure-
bred and the genie value of the crossbred is near 1.0 (as
this study would indicate) it can be seen. from the above
formula. that mass selection in both sexes of the pure-
bred population that make up a cross will be as effective in
improving commercial crossbred performance as if the selec-
tion were practiced improving commercial straightbred

performance.

SUEMARY

The data included records on 375 steers and 362 heifers,
born over a four year period 1960 to 1963 inclusive, at the
Fort Robinson Beef Cattle Research Station in northwestern
Nebraska. The calves were from 80 cows each from the Angus.
Hereford. and Shorthorn breeds. and were the progeny of 17 Il_
Angus. 16 Hereford. and 16 Shorthorn sires. The cows were 4

randomized to breeding pastures each year so that each sire ;

 

was bred to twice as many cows of his own breed as to each of E‘
the other two breeds. Traits studied were birth weight. 5
Weaning score. and adjusted ZOO-day weight (steers and heifers);
adjusted final weight. marbling score. final carcass grade.
fat thickness. rib-eye area. and actual cutability (steers);
and adjusted 550-day weight (heifers).
The experimental design included a new crop of sires
each year which caused a hierarchal design with sires nested
within breed of sire and year. but cross classified with breed
of dam. Steers and heifers were analyzed separately for all
traits.
Variance components for all traits were obtained from
analysis of variance tables. A least-squares and maximum
likelihood general purpose program was used to compute esti-
mates of genetic parameters and standard errors.
The results of the study were as follows:
1. Components of genetic variance and covariance were
calculated for each of the mating types. There were
no differences in the sire components between the

- 7h -

- 75 -

straightbreds and crossbreds. indicating similar
additive genetic variance in the two groups.
Heritability estimates were obtained for the various
traits by the paternal half-sib correlation method.
Separate analyses were obtained for each mating type
x sex sub-group (18 separate analyses). and pooled
within straightbred. and pooled within crossbred Fa

analyses were obtained for each of the sexes. The i

 

separate analyses were subject to large sampling

errors because of limited offspring per sire group.

 

and small degrees of freedom for sires. They were
not published. The pooled analyses yielded herita-
bility estimates fromCL15 t00.85 with most standard
errors between0.3 and0.u. No difference was noted
between the crossbreds and straightbreds. The heri-
tability estimates were large enough to expect im—
provement in the traits with mass selection. The
results indicate that the same type of selection
programs used to improve straightbred populations
would also be effective in improving crossbred
populations.

Genetic and phenotypic correlations were obtained for
the same groups as discussed in 2, except for the
pooled within crossbred group. which was not com-
puted. The within straightbred analyses. by sex.
are published. Some of the genetic correlations

could not be computed because of negative sire com-

ponents of variance. Standard errors. in general.

- 76 -

varied fromCl3 toCLQ. The traits associated with
weight (birth and ZOO—day weight. final weight. 550-
day weight. rib-eye area, and actual cutability) were
highly correlated both genetically and phenotypically
with each other. Weaning score was negatively corre-
lated genetically. and slightly correlated pheno-
typically. with the above mentioned traits. and :1
slightly correlated both genetically and phenotypically

with the other carcass traits (marbling score. final

 

carcass grade. and fat thickness). These carcass “l
traits were highly correlated. genetically. with each
other. and negatively correlated with birth weight.
rib-eye area. and actual cutability. Their phenotypic
correlations with birth weight and rib-eye area were
negative. and they were moderately correlated with
the other traits studied.

Results indicate that a selection program based
on mass selection would be effective in improving
traits associated with weight in the straightbred
population. The size and direction of association
among traits associated with weight, would suggest
that simultaneous improvement in all of these traits
would be effective. Adjusted ZOO-dayweight (steers
and heifers). adjusted final weight (steers). and
550-day weight (heifers) were recommended as traits
to emphasize in a mass selection program.

Correlations were obtained between the genetic ability

- 77 -
of a sire to produce the same trait in straightbred
and crossbred prOgeny. and between his genetic abil-
ity to produce one trait in the straightbred popula-
tion. and another trait in the crossbred population.
In general. the correlations were high. Many were
greater than 1 due to sampling errors. Due to un-
equal numbers of offspring per sire. the covariance
was computed simply as the covariance between
straightbred and crossbred pregeny means. Sire com-
ponents of variance were obtained from the program
used to obtain the genetic correlations. A path co-
efficient diagram was used to illustrate the rela-
tionships involved

In general. the correlations were high indicating.
despite large sampling errors. that the expected values
of the correlations in this study are near 1.00. This
would indicate that selection in purebreds that are
used in a crossing program would be effective in im-
proving the performance of the crossbreds.
A formula was proposed to measure the expected re-
sponse in the crossbred population. from mass selec-
tion in the purebred populations that make up the
cross. If the correlation between the genie value of
the straightbred and the genie value of the crossbred
is near 1.00. as this study indicates. it was con-
cluded that mass selection in the purebred pepulations
that make up a cross would be as effective in im-

proving commercial production in the crossbreds as

 

—78-

it would be in improving commercial production in a
breeding prOgram where the purebred bulls were bred

to commercial cattle of their own breed.

 

9.

10.

11.

12.

LITERATURE CITED

A.S.A.P. 1960. Techniques and Procedures in Animal Production
Research. American Society of Animal Production.

Bassett. James W. and Maurice Shelton. 1966. Crossbreeding
influence on heritability estimates. J. Animal Sci.
25:877 (Abstr.).

Blackwell. R. L.. J. H. Knox. C. E. Shelby and R. T. Clark.
1962. Genetic analysis of economic characteristics of
young Hereford cattle. J. Animal Sci. 21:101.

Brinks. J. 3.. R. T. Clark and N. M. Kieffer. 1963. Sex differ- Fl
ences in response to inbreeding in a line of Hereford .‘i
cattle. Proc. West. Sect. Am. Soc. Animal Sci. 14:V-1.

Brinks. J. 5.. R. T. Clark. N. M. Kieffer and J. R. Quesen-
berry. 1962. Genetic and environmental factors affecting if
11

 

performance traits of Hereford bulls. J. Animal Sci. 21:

777.

Brinks. J. S.. R. T. Clark. N. M. Kieffer and J. J. Urick.
196a. Estimates of genetic. environmental and phenotypic
parameters in range Hereford females. J. Animal Sci. 23:
711.

Brinks. J. S.. J. J. Urick. O. F. Pahnish. B. W. Knapp and
T. J. Riley. 1967. Heterosis in preweaning and weaning
tgaits among lines of Hereford cattle. J. Animal Sci.
2 3278.

Brown. C. J. and Warren Gifford. 1962. Estimates of herita-
bility and genetic correlations among certain traits of
performance tested beef bulls. J. Animal Sci. 21:388
(AbStro)o

Busch. D. A. and C. A. Dinkel. 1967. Heritability estimates
for certain beef traits. J. Animal Sci. 26:1465 (Abstr.).

Carter. R. C. and C. M. Kincaid. 1959a. Estimates of genetic
and phenotypic parameters in beef cattle. II. Heritability
estimates from parent-offspring and half-sib resemblances.
J. Animal Sci. 18:323.

Carter. R. C. and C. M. Kincaid. 1959b. Estimates of genetic
and phenotypic parameters in beef cattle. III. Genetic
and phenotypic correlations among economic characters.
J. Animal Sci. 18:331.

Christian. Lauren LaRoy. 1963. Analysis of growth and pro-
duction traits in twin and non-twin beef cattle data.
Ph.D. Thesis. University of Wisconsin. Madison. Wisconsin.

- 79 -

13.

14.

150

16.

17.

18.

19.

20.

21.

22.

23.

24.

1hr} , C. J.. D. Chambers. L. E. Walters. J. V. White-
' ‘ :. scephens. 19CZ. Heritability estimates
éhararzcristics. J. Animal Sci. 21:38? (Abstr.)
Clark. R. 1.. J. c. Brinks,Ralph Bogart. L. A. Holland.

EnubiceJ. 0. F. Pahnish. J. A. Bennet and R.
..-1-tian. 1063. Beef cattle breeding research in the
western region. Ore. Agr. dxp. Sta. Res. Bul. 73. ‘

fir
v

Comstock. R. E. 1960. Problems and evidence in swine breed-
ing. J. Animal Sci. 19:75.

Comstock. R. E. 1961. Reciprocal recurrent selection with
reference to swine breeding. An. Rpt. and Troc. Reg. -
SWine Brd. Lab. a

 

Cundiff. L. 7 Doyle Chambers. D. F. Stephens and R. L.

Willhan. 1994. Genetic analysis of some growth and car-
cass traits in beef cattle. J. Animal Sci. 23:1133.

Cunningham. E. P. and C. R. Henderson. 1965. Estimation of
genetic and phenotypic parameters of weaning traits in
beef cattle. J. Animal Sci. 24:182.

Dearborn. Delwyn and C. A. Dinkel. 1959. Evaluation of final
weight in the selection of performance tested bulls. J.
Animal Sci. 18:1464 (Abstr.).

Doane. Ted H. 1960. Inheritance of production traits in beef
cattle. Ph.D. Thesis. hansas State tniversity, Manhattan.
Kansas.

DuBose. L. E and T. C. Cartwright. 1967. Relationships among
production and carcass traits in cattle. J. Animal Sci.
26:203 (Abstr.).

Enfield. F. D. and W. E. Rempel. 1962. Covariance of sire
effects in purebred and crossbred populations of swine.
J. Animal Sci. 21:971 (Abstr.).

Fitderald. Marion. 1961. Estimation of genetic parameters
of certain production and carcass traits in beef cattle.
N.S. Thesis. South Dakota State University, Brookings.
South Dakota.

Fitzhugh. H. A. Jr. 1965. A biometrical evaluation of
weights of beef cows and performance of their progeny.
Ph.D. Dissertation. Texas A & M University. College
Station. Texas. '

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

-81..

Gaines. J. A.. W. H. McClure. D. W. VOgt. R. C. Carter and
C. M. Kincaid. 1966. Heterosis from crosses among British
breeds of beef cattle: fertility and calf performance to
weaning. J. Animal Sci. 25:5.

Gaines. J. A.. G. V. Richardson. W. H. McClure, D. W. Vogt
and R. C. Carter. 1967. Heterosis from crosses among
British breeds of beef cattle: carcass characteristics.
J. Animal Sci. 26:1217.

Gregory. K. E. 1961. Improvement of beef cattle through
breeding methods. North Central Regional Publ. 120. Web. J
Agr. Exp. Sta. Res. Bul. 196. :

Gregory. K. E. 1965. Symposium on performance testing in .
beef cattle: Evaluating postweaning performance in beef
cattle. J. Animal Sci. 24:248.

 

Gregory. K. E.. L. A. Swiger. R. M. Koch. L. J. Sumption.
W. W. Rowden. and J. E. Ingalls. 1965. Heterosis in pre-
weaning traits of beef cattle. J. Animal Sci. 24:21.

tfi; “w

Gregory. K. E.. L. A. Swiger. R. M. Koch. L. J. Sumption.
J. E. Ingalls. W. W. Rowden and J. A. Rothlisberger.
1966a. Heterosis effects on growth rate of beef heifers.
J. Animal Sol. 25:290.

Gregory. K. E.. L. A. Swiger. L. J. Sumption. R. M. Koch,
J. E. Ingalls. W. W. Rowden. and J. A. Rothlisberger.
1966b. Heterosis effects on growth rate and feed
efficiency of beef steers. J. Animal Sci. 25:299.

Gregory. K. E.. L. A. Swiger. L. J. Sumption. R. M. Koch.
J. E. Ingalls. W. W. Rowden and J. A. Rothlisberger.
1966c. Heterosis effects on carcass traits of beef
steers. J. Animal Sci. 25:311.

Hagan. Fay. C. E. Lindley. C. J. Christians and J. C. Taylor.
1964. A study of economically important preweaning traits
in beef cattle and herd bull selection. Ann. Livestock
Field Day Report. Miss. Agr. Exp. Sta.

Hallman. Lester Clark. Jr. 1965. Performance of calves en-
rolled in the Pennsylvania beef cattle improvement pro-
gram. Ph.D. Thesis. Pennsylvania State University. Uni-
versity Park. Pennsylvania.

Hamann. H. K.. S. Wearden and W. H. Smith. 1963. Estimation
of genetic and environmental factors affecting weaning
weights of creep-fed cattle. J. Animal Sci. 22:316.

Harricharan. H.. R. W. Bratton. C. R. Henderson. 1967. Esti-
mates of some genetic parameters of economic importance
in Angus cattle. J. Animal Sci. 26:875 (Abstr.).

37.

38.

39-

40.

41.

42.

43.

an.

1+5.

46.

47.

48.

49.

- 82 -

Harvey. Walter R. 1960. Least-squares analysis of data with
unequal subclass frequencies. U.S.D.A.. Agricultural
Research Service. ARS 20-8.

Harvey. Walter R. 1964. Computing procedures for a generalized
least-squares analysis program. Mimeographed paper pre-
sented at the Analysis of Variance Workshop. Ft. Collins.
Colorado. 50 pp.

Hazel. L. N.. Marvel L. Baker and C. F. Reinmiller. 1943.
Genetic and environmental correlations between growth
rates of pigs at different ages. J. Animal Sci. 2:118. .1—

Hazel. L. N. and C. E. Terrill. 1945. Heritability of wean-
ing weight and staple length in range Rambouillet lambs.
J. Animal Sci. 4:347.

Henderson. C. R. 1953. Estimation of variance and covariance
components. Biometrics 9:226.

 

E , I; 'I‘" ‘VIiirgt‘..sj .

Howarth. Birkett. Jr. 1960. The evaluation of weaning weight.
12-month weight and 12-month type score in the selection
of replacement heifers. M.S. Thesis. South Dakota State
University. Brookings. South Dakota.

Kieffer. Nat N.. Doyle Chambers and D. F. Stephens. 1959. In-
heritance of certain maternal traits in beef cattle. J.
Animal Sci. 18:1464 (Abstr.).

Kieffer. Nat M.. R. L. Hendrickson. Doyle Chambers and D. F.

Stephens. 1958. The influence of sire upon some carcass
characteristics of Angus steers and heifers. J. Animal
Sci. 17:1137 (Abstr.).

King. 5. C. and C. R. Henderson. 1954. Variance components
analysis in heritability studies. Poul. Sci. 33:147.

Lesley. J. P.. B. N. Day and J. E. Comfort. 1961. Some genetic
aspects of gestation length. and birth and weaning weights
in Hereford cattle. J. Animal Sci. 20:737.

Lehmann. R. P.. J. A. Gaines. R. C. Carter. K. P. Bovard and
C. M. Kincaid. 1961. Selection indexes for weanling
traits in beef calves. J. Animal Sci. 20:53.

Loganathan. S. 1962. The genetic and environmental factors
affecting the performance of Hereford calves from birth
to weaning. M.S. Thesis. university of Tennessee. Knox-
ville. Tennessee.

Loganathan. S.. R. J. Cooper and C. S. Hobbs. 1965. Factors
affecting preweaning performance of Hereford calves. J.
Animal Sci. 24:849 (Abstr.).

50.

51.

52.

53-
54.

55.

56.

57.

58.

59.

60.

61.

62.

-83-

Lombard. J. H. 1963. Heritabilities of pre-weaning and wean-
ing characteristics of beef cattle. Proc. S. Afr. Soc.
Anim. Prod. 2(2):121.

Louca. A. and 0. W. Robison. 1967. Components of variance and

covariance in purebred and crossbred swine. J. Animal
8C1. 26:267.

Lush. J. L. 1940. Intra-sire correlations or regressions of
offspring on dam as a method of estimating heritability
of characteristics. Proc. Amer. Soc. Anim. Prod. 23:293.

Lush. J. L. 1948. The genetics of populations. Mimeo. Publ.

Lush. J. L. 1949. Heritability of quantitative characters in
farm animals. Proc. 8th Intern. Congr. Genetics 1948.

Magee. W. T.. J. Brinks. R. H. Nelson and G. A. Branaman.
1961. Some factors affecting weaning weight and score of
bezf calves. Quart. Bul. Mich. Agric. Exp. Sta.. 43:

55 . -

Magee. W. T.. R. H. Nelson. G. A. Branaman. L. J. Bratzler
and A. M. Pearson. 1958. Some factors affecting carcass
grade in steers. J. Animal Sci. 17:649.

Mahmud. Anuwar and Estel H. Cobb. 1963. Factors affecting
weaning weights. preweaning gains. and conformation scores
of beef calves in Hawaii. J. Animal Sci. 22:820 (Abstr.).

Makarechian. Mahmoud. 1960. Factors affecting the yearling
productive traits in beef cattle. Ph.D. Thesis. Michigan
State University. East Lansing. Michigan.

Marchello. J. A.. D. W. Blackmore and J. J. Urick. 1960.
Heritability of 18-month weight of heifers and the rela-
tionship of this weight to the birth and weaning weight
of the heifers first calf. Proc. West. Sect. Am. Soc.
Animal Sci. 113X-1.

Marlowe. Thomas J. and D. W. Vogt. 1965. Heritabilities.
phenotypic correlations. and genetic correlations in-
volving preweaning gain and weaning grade of beef calves.
J. Animal Sci. 24:502.

Meade. J. H.. Jr.. J. C. Dollahon. J. C. Taylor and C. E.
Lindley. 1959. Factors influencing weaning weights of
Hereford and Angus cattle in Mississippi. J. Animal Sci.
18:1149 (Abstr.).

Meyerhoeffer. D. C.. R. C. Carter and B. M. Priode. 1960.
Early selection of beef calves. J. Animal Sci. 19:1222
(Abstr.).

 

63.

64.

68.

69.

70.

71.

72.

73.

74.

-84-

Meyerhoeffer. D. C.. R. C. Carter and B. M. Priode. 1963.
Sex differences in heritability of traits in beef calves.
J. Animal Sci. 22:240 (Abstr.).

Minyard. J. A. and C. A. Dinkel. 1965. Heritability and re-
peatability of weaning weight in beef cattle. J. Animal
Sci. 24:1072.

Miquel. Cesar and T. C. Cartwright. 1963. Comparison of
heritabilities in crossbred and purebred cattle. J.
Animal Sci. 22:821 (Abstr.).

Pahnish. O. F.. R. L. Roberson. R. L. Taylor. J. S. Brinks. F!
R. T. Clark and C. B. Roubicek. 1963. Genetics of growth
traits in range cattle. J. Animal Sci. 22:821 (Abstr.). .

 

Pahnish. 0. E.. R. L. Roberson. R. L. Taylor. J. S. Brinks.
B. T. Clark and C. B. Roubicek. 1964. Genetic analyses is:
of economic traits measured in range-raised Herefords at a
preweaning and weaning ages. J. Animal Sci. 23:562.

Pahnish. O. F.. E. B. Stanley. Ralph Bogart and C. B. Roubi-
cek. 1961. Influence of sex and sire on weaning weights
of Southwestern range calves. J. Animal Sci. 20:454.

Parker. Charles Frederick. 1963. A biometrical evaluation of
certain genetic and environmental parameters in a large
herd of Santa Gertrudis cattle. Ph.D. Thesis. Texas A &
M University. College Station. Texas.

Petty. R. B. Jr. 1966. A biometrical evaluation of growth
and conformation traits and comparison of selection pro-
cedures of young British breed beef cattle. M.S. Thesis.
Texas A & M University. College Station. Texas.

Petty. R. E.. Jr. and T. C. Cartwright. 1966. A summary of
genetic and environmental statistics for growth and con-
formation traits of young beef cattle. Dep. Tech. Rep.
Tex. Agric. Exp. Sta.. No. 5: 55 pp. (Mimeo.).

Roberson. Robert L.. 0. F. Pahnish. R. L. Taylor. J. S.
Brinks. R. T. Clark and A. M. Lane. 1963. Genetics of
grade. condition and weight in range cattle. J. Animal
Sci. 22:822 (Abstr.).

Robertson. Alan. 1962. Weighting in the estimation of vari-

ance components in the unbalanced single classification.
Biometrics 18:413.

Robison. 0. W. 1967. Information on the covariance between
the straightbred and crossbred progeny performance of,a
sire. with unequal numbers in the progeny groups. Pri-
vate communication.

75-

76.

77-

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

- 85 -

Robison. 0. W.. A. Louca and J.E. Legates. 1964. Purebred and
crossbred performance of paternal half-sibs of swine. J.
Animal Sci. 23:853 (Abstr.).

Rollins. W. C. and K. A. Wagnon. 1958. Genotype environment
interaction in a beef cattle herd. Proc. Xth Int. Congr.
Genet. (Montreal). 1958. Vol. II (Abstr.): 240.

Sanfiorenzo. John Hayward. 1964. Comparison of selection in-
dexes constructed for widely different beef cattle types.
Ph.D. Thesis. Texas A & M University. College Station.
Texas 0

Scarth. R. D.. P. J. Phillips. G. W. Sherritt. J. H. Ziegler
and T. B. King. 1965. Estimating genetic and phenotypic
parameters for performance traits. carcass traits and
live measurements from selected groups of beef sires. J.
Animal Sci. 24:1219 (Abstr.).

Sewell. H. B.. J. E. Comfort. B. N. Day and J. F. Lesley.
1963. Genetic and environmental factors influencing the
weaning weight of beef calves. Res. Bul. Mo. Agric..Exp.
Sta. No. 823.

Shelby. C. E.. R. T. Clark. J. R. Quesenberry and R. R. Wood-
ward. 1957. Heritability of some economic traits in record
of performance bulls. J. Animal Sci. 16:1019 (Abstr.).

Shelby. C. E.. R. T. Clark. J. R. Quesenberry and R. R. Wood-
ward. 1960. Heritability of some economic characteristics
in record of performance bulls. J. Animal Sci. 19:450.

Shelby. C. E.. W. R. Harvey. R. T. Clark. J. R. Quesenberry
and R. R. Woodward. 1963. Estimates of phenotypic and
genetic parameters in ten years of Miles City R.O.P.
steer data. J. Animal Sci. 22:346.

Stanislaw. C. M.. I. T. Omtvedt. R. L. Willham and J. A.
Whatley. Jr. 1967. A study of some genetic parameters in
purebrgd 2nd crossbred populations of swine. J. Animal
801. 2 31 o

Swiger. L. A. 1961. Genetic and environmental influences on
gain of beef cattle during various periods of life. J.
Animal Sci. 20:183.

Swiger. L. A.. K. E. Gregory. R. M. Koch. W. W. Rowden. V.
H. Arthaud and J. E. Ingalls. 1963. Evaluating post-
weaning gain of beef calves. J. Animal Sci. 22:514.

Swiger. L. A.. K. E. Gregory. L. J. Sumption. B. C. Breiden-
stein and V. H. Arthaud. 1965. Selection indexes for
efficiency of beef production. J. Animal Sci. 24:418.

SW1ger. L. A.. W. R. Harvey. D. 0. Everson and K. E. Gregory.
1964. The variance of intraclass correlation involving
groups with one observation. Biometrics 20:818.

 

88.

89.

90.

91.

92.

93.

91+.

95.

96.

97.

- 86 -

Swiger. L. A.. R. M. Koch. K. E. Gregory and V. H. Arthaud.
1961. Effect of length of the feeding period on accuracy
of selection for gain and feed consumption in beef cattle.
J. Animal Sci. 20:802.

Swiger. L. A.. R. M. Koch. K. E. Gregory and V. H. Arthaud.
1962. Selecting beef cattle for economical gain. J.
Animal Sci. 21:588.

Swiger. L. A.. R. M. Koch. K. E. Gregory. V. H. Arthaud. W.
W. Rowden and J. E. Ingalls. 1962. Evaluating pre-
weaning growth of beef calves. J. Animal Sci. 21:781.

Synar. Harry R. 1958. The relatinnships of weaning production
factors in range cattle. M.S. Thesis. University of
Arizona. Tucson. Arizona.

Tallis. G. M. 1959. Sampling errors of genetic correlation
coefficients calculated from analyses of variance and
covariance. The Aust. J. Stat. 1:35.

Tallis. G. M. and Earle W. Klosterman. 1959. Efficient esti-
mates of heritability from paternal half-sib correlations.
J. Animal Sci. 18:622.

Taylor. J. C.. R. H. Miller and H. 0. Hetzer. 1965. Genetic
correlations between straightbred and crossbred swine.
J. Animal Sci. 24:852 (Abstr.).

Thornton. J. W.. J. A. Gaines and C. M. Kincaid. 1960. Esti-
mates of parameters of growth in beef heifers. J. Animal
Sci. 19:1228 (Abstr.).

Vogt. D. W.. J. A. Gaines. R. C. Carter. W. H. McClure and
C. M. Kincaid. 1967. Heterosis from crosses among

British breeds of beef cattle: post-weaning performance
to slaughter. J. Animal Sci. 26:443.

Wagnon. K. A. and W. C. Rollins. 1959. Heritability estimates
of post-weaning growth to long yearling age of range beef
heifers raised on grass. J. Animal Sci. 18:918.

Warwick. E. J. 1958. Fifty years of progress in breeding
beef cattle. J. Animal Sci. 17:922.

Wilson. L. L.. C. A. Dinkel and D. E. Bay. 1962. Genetic
parameters and selection indexes for beef cattle. J.
Animal Sci. 21:97? (Abstr.).

Wilson. L. L.. C. A. Dinkel. D. E. Bay and J. A. Minyard.
1963. Beef cattle selection indexes involving conforma-
tion and weight. J. Animal Sci. 22:1086.

 

a»; hrilfiiuuqfiiwiln. .-
. ..: p..- .. 11..

0’5er