Aust. J. Agric. Res. 1975, 26, 393-406
Effectiveness of Selection for Growth Rate
in Zebu X British Crossbred Cattle. I
Pre-weaning Growth. -
G . W . Seifert
*
Division of Animal Genetics, CSIRO, Tropical Cattle Research Centre,
P.O. Box 542, Rockhampton, Qld. 4700.
Abstract
From random-bred populations of Africander cross (AX) and Brahman cross (BX), bulls and cows
were selected within breeds for high and low weight per day of age ( W / A )at c. 2 years of age and
mated assortatively. There was a similar and significant response to selection in both breeds in birth
weight and pre-weaning W / A as well as in pre-weaning gains. Although the BX animals were heavier
and gained more than the AX animals, the differences were not as great or as consistent as the effect
of selection.
Calf gains were inversely related to cow gains from birth to weaning, but positively correlated to
cow weight. During the pre-weaning period high W / A line cows gained significantly more than the
low W / A line cows, while AX animals gained more than the BX group. The higher gain of the high
W / A line cows and the higher weaning weight of their calves showed that these lines were more
efficient than the low lines in terms of production of weaning weight.
Very early calf gains from birth in November to January, and W / A in January, appeared to be
poor indicators of the genetic potential of the calf and of the permanent environmental effects affecting its growth. In contrast, the period from January to March, during which the most rapid gains
were experienced, reflected both the genetic potential and permanent environmental effects.
The previous lactational status of the dam only affected birth weight and day of birth significantly
and affected the AX calves more than the BX calves. This resulted in a significant interaction
between breed and previous lactational status for these two traits.
Introduction
Estimates of heritabilities of the growth rate of beef cattle reported by overseas
workers (Shelby et al. 1963; Pahnish et al. 1964) have generally been moderately high.
Therefore according to quantitative genetic theory, selection for improved growth
rate should be effective. A number of selection experiments with cattle (Rollins et al.
1962; Flower et al. 1964; Carter 1971 ; Newman et al. 1973) have in fact confirmed
that selection for growth rate would be effective, but all concluded that selection should
be for weight, rather than for gains.
Few estimates of the genetic parameters for growth in beef cattle exist in Australia,
and only Pattie et al. (1970) have obtained estimates for weaning weight. No estimates
exist for crossbred cattle in tropical Australia. However, Seifert and Kennedy (1972)
concluded that growth in Brahman and Africander x British cattle appeared to be
largely under the influence of additive genetic effects.
With the introduction of a National Performance Recording Scheme in Australia,
it is necessary to obtain estimates of the genetic parameters for growth, as well as to
demonstrate the effectiveness of selection for high growth rate. Many of the herds in
G. W. Seifert
northern Australia are crossbred herds based on the Brahman, and new breeds are
being synthesized from these, so that information in crossbred herds is also needed.
This paper reports the results of an experiment designed to measure the effectiveness of up and down selection for growth rate in Fg Brahman x British and
Africander x British populations.
Animals and Methods
The data presented were collected at the National Cattle Breeding Station, 'Belmont', Rockhampton, Qld. The environment has been described previously (Kennedy
and Turner 1959).
Animals
From random-bred Fz populations of Africander x Hereford (AH), Africander x
Shorthorn (AS), Brahman x Hereford (BH) and Brahman x Shorthorn (BS), all
male calves born during the normal 1962 calving period (October to December) were
left entire. The males were weaned at approximately weekly intervals when they
reached 205 days of age. All the female calves were weaned on 19 June 1963, when
their average age was 241 days.
Table 1. Number of bulls, cows and calves by breed and line
Bulls
Breed
No.
High W / A line
Cows
Breed
No.
AH
AS
Total
2
1
3
AS
AH
BH
BS
Total
1
2
3
BS
BH
All
6
63
35
98
I
Calves
No.
I
Bulls
Breed
No.
AH
AS
70
BH
BS
30
56
86
36
184
106
1
Low W / A line
Cows
Breed
No.
1
1
2
AS
AH
1
1
2
BS
BH
4
Calves
No.
29
27
56
36
29
29
58
33
114
69
Within each breed group, approximately three times as many bulls as were required
for mating were selected on high and low weight per day of age, during February 1964,
when their average age was 434 days. The final selection from those previously selected
took place during September 1964, at a mean age of 657 days. All females, ranging in
date of birth from November 1957 to November 1962, were similarly divided into
high and low groups within breed and year of birth, on their age-corrected weights at
c. 2 years. The mating design and the numbers of bulls and cows mated are given in
Table 1.
Bulls were joined singly to c. 30 females during January and February 1965 for a
period of 7 weeks. Within the Africander and Brahman breed components, bulls were
mated to cows of the opposite British breed component. Similarly herds of approximately 100 Hereford x Shorthorn (HS) and 100 Shorthorn x Hereford (SH) cows
were mated to three randomly selected SH and HS bulls respectively.
Selection for Growth Rate in Cattle. I
The progeny from all matings were born from 25 October 1965 (day 0) to 11
January 1966, the mean day of birth being 5 December 1965 (day 41k15). The
maximum range in age of the calves was 78 days. All the calves were grazed together
except during mating from 19 January to 10 March 1966, during which period families
were re-formed and mated to single sires. Calves were weighed at birth, on 19
January, 10 March, and 27 April, and at weaning on 12 June 1966, when the average
age was 230 days. The cows were weighed on the 19 January and again on 12 June at
the weaning of their calves.
Statistical Analyses
The age-corrected weights of the progeny were analysed by the least squares method
(Harvey 1960) by using the following model.
Xijklm=u
+ Bi L j $- Ak f Sf + BLij + BAik $ BSil + GAjk
+ GSjl t Ask1 + PIDijkl + P2Gijklm +
m
PSWijklm
t eijklm
where X
ij,c,m
=weight at birth, or weight per day of age at January, March, April
or weaning (June) or gains between the above successive months;
a = overall
mean when all covariates equal zero;
Bi = effect of the ith breed (i = 1,2) ;
L j = effect of the jth line of selection ( j= 1, 2) ;
Ak=effect ofthe kth dam age ( k = l , 2 . .
. 4);
S , = effect of the lth sex (1 = 1, 2);
BL, BA, BS, GA, GS, AS = interactions between the main effects;
,Bn = partial regression of the XijkLm
on the day of birth of the calf (the
first calf was born on day 0 so that older calves have the smallest
day of birth) (n = l), gain of the dam from January to the time of
weaning of her calf (n = 2), birth weight of the calf (n = 3);
,,,= day of birth of the mth calf;
Dij
Gij,, ,= body weight gain of the dam from 19 January to 12 June (weaning) ;
Wij,, ,
= birth weight of the calf;
eijk,, = random error N.I.D. (0, 0:).
The same set of normal equations was used to analyse age-corrected weights (WIA)
and average daily gains between successive times. The W/A includes birth weight and
the cumulative gain up to the point of weighing, while the average daily gain represents only the gain between any two given times. Non-significant interactions were
eliminated in the final reduced models used to obtain the least squares constants. The
model was also used to analyse day of birth and dam weights and gains.
A model including the previous lactational status of the dam was used to analyse
a subset of the data which excluded maidens. This was necessary to avoid the confounding of the previous lactational status of maidens which were all non-lactating.
G. W. Seifert
Table 2. Analyses of variance of birth weight and pre-weaning
weights per day of age (WIA) of calves
Sourcea
DF
I
Birth
Jan.
Mean squares
Mar.
Apr.
Weaning
Breed (B)
Line (L)
Dam age (A)
Sex (S)
B x L
B x A
B x S
Lx A
L x S
A x S
Regr. B. day
Regr. dam gain
Error
aRegr., regression; B. day, day of birth; dam gain, dam gain from 19 January to 12 June (weaning),
Table 3. Least squares constants for birth weight and pre-weaning weights per day of age (WIA), and
their partial regressions on day of birth (PI), dam gain from January to June (weaning) (Pz),
and birth weight (03) of calves
Mean day of birth, 41.09 days; mean dam gain, 3.68 kg; mean birth weight, 27.98 kg
Birth
24.943
28.223
or a
pb
0.139
-0439
High W / A
LOWW / A
1.028
- 1.028
3 years
4 years
5 years
6 years
-1.404
-0.580
0.946
1.038
Males
Females
1.034
-1.034
PI
0~0810**
-0.0130
/%
Pa
-
Mar.
Apr.
Weaning
-0.355
2,813
0.977
1.613
0.896
1.278
0.762
1 .029
-0.041
0,041
-0.034
0.034
0.042
-0.042
0.041
-0.041
-0.041
-0.024
0.033
0.032
-0.037
-0.022
0.039
0.020
0.035
-0.035
0.024
-0.024
0.0002"
-0.0020**
0.0136**
-0.0001
--0.0016**
0 . 0099"*
Breed
-0.053
-0.073
0.053
0.073
Selection
0,088
0.049
-0.088
-0.049
Dam Age
-0.073
-0.049
-0.027
-0.156
0.038
0.247
0,038
-0.018
Sex
-0.014
0.039
-0.039
0.014
Partial Regression
AX
BX
+
Jan.
0.0466**
-0.0038
0.0453
0.0025**
-0*0019**
0.0193**
"a,Overall mean when day of birth, dam gain from January to weaning and birth weight are zero.
bp
=
C( + P1 (mean day of birth)
weight of calves).
+ /?z (mean dam gain from January to weaning) + 83 (mean birth
Selection for Growth Rate in Cattle. I
397
Table 4. Analyses of variance of successive average daily gains from birth to weaning of calves
Sourcea
DF
Breed (B)
Line (L)
Dam age (A)
Sex (S)
B x L
B x A
BXS
L x A
LXS
A x S
Regr. B. day
Regr. dam gain
Regr. birth wt.
Error
*P < 0.05.
1
1
3
1
1
3
1
3
1
3
1
1
1
146
Birth to
Jan.
Jan. to
March
0.018**
0.039**
0.040**
0~001
0 .023
0.001
0.000
0.002
0.001
0~004
2.105**
0.052**
0,015
0.006
0.002
0.059**
0.015
0.098**
0.009
0.016
0.007
0.006
0.001
0.007
0.209**
0.064**
0.097**
0.008
Mean squares
March to
April to
April
weaning
0.002
0.032**
0.005
0.020**
0.003
0.005
0.002
0.004
0.000
0.000
0.002
0.062**
0,005
0.003
0.004
0.063**
0.006
0.000
0.000
0.004
0.000
0.001
0.006
0.001
0.021**
0.022**
0.008
0,003
Birth to
weaning
0.034*
0.156**
0.020*
0.036*
0.000
0.009
0.000
0.003
0 000
0.002
0.016
0.156**
0.008**
0.006
3
"As for Table 2.
**P <: 0.01.
Table 5. Least squares constants for successive pre-weaning average daily gains, and their partial
regressions on day of birth (PI), dam gain from January to June (weaning) (Pe), and birth
weight (P3) of calves
Mean day of birth, 41.09 days; mean dam gain, 3.68 kg; mean birth weight, 27.98 kg
--
-
-
Birth to
Jan.
ua
pa
0.808
0.524
AX
BX
-0.041
0.041
High W/A
LOWW/A
0.019
-0.019
3 years
4 years
5 years
years
6i
-0.028
-0.024
0.049
0.003
Males
Females
0.001
-0.001
b'1
-0.0087**
-0.0009**
0.0027
PZ
P3
"See Table 3.
-
Jan. to
March
-
March to
April
Overall Mean
0.303
0.255
0.377
0.308
Breed
-0.016
-0.008
0.016
0.008
Selection
0.023
0.014
-0.023
-0.014
Dam Age
-0.011
-0.012
0.008
-0.009
0.010
-0.021
0.024
0.011
Sex
0.029
0.013
-0.029
-0.013
Partial Regression
0 .0003
-0.0028**
-0.0010**
-0.0010**
0.0069**
0.0016
-
-
April to
weaning
-
-
Birth to
weaning
0.263
0.351
0.674
0.783
-0.012
0.012
-0.0334
0.034
0.023
-0,023
0.036
-0.036
-0.017
-0.01 1
0.012
0.016
-0.031
-0.020
0,025
0.026
0.002
-0,002
0.019
-0.019
0.0009**
-0~0006**
0.0019
-0.0008
-0.0016**
0.0053**
G. W. Seifert
Results
The analyses of variance for birth weight and pre-weaning weight per day of age ( WIA)
of the calves are given in Table 2, while the analyses of variance for pre-weaning
average daily gains for successive periods are given in Table 4. The least squares
constants for WIA and average daily gain and their partial regressions are given in
Tables 3 and 5 respectively. The least squares constants are for the reduced model
in which the partial regressions on birth weight and dam gains, as well as all the nonsignificant interactions, were eliminated.
The analyses of the day of birth and the dam weights and gain and their partial
regressions on birth weight and calf gain to weaning are given in Table 6 , while the
least squares constants are given in Table 7. Effects of previous lactation status are
included in these analyses, and maiden cows are omitted.
Table 6. Analyses of variance of day of birth of the calf and dam weights at January and June and dam
gain from January to June (weaning)
---
-
Sourcea
Breed (B)
Line (L)
Dam age (A)
Sex (S)
Lact. status (P)
B x L
B x A
B x S
B x P
L x A
L x S
L x P
A x S
A x P
S X P
B x A x P
Regr. B. wt.
Regr. calf g
Error
DF
1
1
2
1
1
1
2
I
1
2
1
1
2
2
1
2
1
1
99
Day of
birth
1963.12*"
106.88
121.70
175.22
409.90183.06
88.52
19.98
1184.28**
62.27
244.80
13.38
104.73
2.27
37.32
4.47
3519.34**
6988.46**
123.00
--
Mean squares
January
Weaning
dam wt.
dam wt.
991.13
7444.43**
10619~51**
3728.0612412.29
2751.79
1189.36
1335.36
12.02
2097.42
3890.94t
0.04
259.49
5463.01*
332.62
186.29
13534.91**
672.82
1036.43
1674.61
17952.58**
14266.46**
1279.03
9.66
1074.63
1750.84
828.72
187.99
2146.50
4904.60t
34.10
80.58
2411.59
1832.40
470.92
8054.96*
1130.92
1497.79
-
Dam
gain
5242.37**
2275.88*
400.80
639.80
2116.71"
387.14
81.31
60.14
295.09
41.21
58.61
36.35
54.83
638.72
603.61
73.18
707.04
3548.34**
347.96
**P< 0.01.
tP < 0.10
"Regr. B. wt., regression on birth weight. Regr. calf g, regression on calf gain from birth to weaning.
*P < 0.05.
Breed differences
Calf Weights, Gains and Day of Birth
The BX calves were heavier than the AX calves at all pre-weaning ages except
birth (Table 3). The advantage was only significant in March (P < 0.01) and at
weaning (P < 0.05) (Table 2). The BX calves gained more than the AX calves between dates of weighing. The differences were only significant from birth to January
(P < 0.01) and over the whole period from birth to weaning (P < 0.05) (Tables 4, 5).
399
Selection for Growth Rate in Cattle. I
The growth curves for the breeds and lines are shown in Fig. 1. The high and low
crossbred lines were heavier than the random-selected British, but the advantage was
not present at birth and appeared at the January weighing and increased toward
weaning.
AX calves were born earlier (P < 0 ~ 0 1 than
)
BX calves (Tables 6, 7).
Table 7. Least squares constants for day of birth of the calf and dam weights at January
and June and dam gain from January to June, and the partial regressions on calf
birth weight (PI) and calf pre-weaning gain (02)
Mean birth weight, 27.98 kg; mean pre-weaning gain, 142.41 kg
Day of
birth
January
dam wt.
56.174
41.574
322,491
417.538
ua
fib
AX
BX
Breed
-5.031
5.031
High W / A
Low W / A
Selection
1.167
9.744
-1.167
-9.744
4 years
5 years
6 years
Dam Age
1.522
-21.444
1 .046
3 ,547
-2.568
17.497
Males
Females
Sex
1.368
-1.368
+
Lactating
Not lactating
Weaning
dam wt.
Dam
gain
-3.574
3.574
-6.309
6.309
Pvevious Lactational Status
2.262
-5.489
-2.262
5.489
Regressions
1.425**
2.794**
-0.381**
0.118
81
Dz
"a,Overall mean when calf birth weight and calf pre-weaning gain are zero.
bfi = u
+ 81 (mean birth weight) +
82
(mean pre-weaning gain).
Dam Weights and Gains
Dam weights did not differ significantly, but AX dams gained more (P < 0.01)
than the BX dams while suckling their calves (Tables 6, 7).
Dzrevences between selected groups
Calf Weights, Gains and Days of Birth
The progeny of lines selected for high W / A were significantly heavier than those
from the low lines at birth and maintained the significant advantage at all ages up to
weaning with the exception of the January weight (Tables 2, 3).
G . W. Seifert
Differences in average daily gain were highly significant (P < 0.01) (Tables 4, 5),
and were largest in the period from April to weaning.
Although calves from the high line were born later than those from the low line,
the differences were small and not significant.
Dam Weights and Gains
The up-selected dams were heavier (P < 0.01) than the low lines in January and
improved this advantage by better gains (P < 0.01) from January to weaning.
200
I50
Fig. 1. Calf weights
corrected for age of dam and sex
of the selected AX and BX groups,
and the random British group (HS).
5.xii.65, mean birth date; 12.vi.66,
date of weaning.
h
M
-r:
V
2on
.-
-u"-E
100
50
01
!
5.xii.65 19. i.66
t
I
I
1O.iii.66
27. iv.66
12.vi.66
Date of weighing
Eflects of dam age
Ca2f Weights, Gains and Days of Birth
Calf weight per day of age increased significantly with an increase in dam age,
except for the January weight (Tables 2, 3). Dam age only had a significant effect on
gain from birth to January, and on total gain from birth to weaning (P < 0.05)
(Tables 4, 5).
When dam weights at January and weaning, rather than dam gain, were included
in the model as covariates, the significance of dam age effects on W / A was eliminated,
with the exception of April weight. This indicates that a large part of the dam age
effect is due to the associated dam body weight differences.
The least squares constants for W / A (Table 5) also show that there was little
difference between dams 5 years and older, and that these classes could conveniently
have been grouped together.
Selection for Growth Rate in Cattle. I
Dam age had no significant effect on day of birth (Table 6), but older dams tended
to calve earlier than younger dams (Table 7) with the exception of maidens which
calved earliest.
Dam Weights and Gain
Older dams were heavier (P c 0.01) than younger dams in January and June
(Tables 6, 7), and where maidens were included they were the lightest. Although the
differences in gains were non-significant, the maidens gained the least and the older
cows the most.
Table 8. Analyses of variance including the effect of previous lactational
status of the dam on birth weight and pre-weaning ADG
I
Source
DF
Breed (B)
Line (L)
Dam age (A)
Sex (S)
Lact. status (P)
B x L
B x A
B x S
B x P
L x A
L x S
L x P
A XS
A XP
S X P
B1 (B. day)
P2 (dam gain)
Error
*P < 0.05.
1
1
2
I
1
1
2
1
1
2
1
1
2
2
1
1
1
97
**P < 0.01.
Mean squares
Birth
Jan-Mar.
Mar.-Apr.
31.275
111.082**
46.917*
72.009*
75.517*
0.066
2.457
2.156
87.632*
18.275
5.242
8.215
34.445
2.665
0.818
209.986**
17.834
15.001
0.0039
0.1004**
0.0135
0.1096**
0.0083
0.0065
0.0296t
0.0068
0.0033
0.0048
0.0023
0.0044
0.0115
0~0011
0.0002
0.1708**
0.0485*
0.0103
0.0004
0.0430**
0.0042
0.0182*
0.0119*
0.0019
0.0033
0.0021
0.0004
0.0006
0.0001
0.0001
0.0003
0.0029
0 0021
0.0004
0.0412**
0.0030
1-P < 0.10.
Sex dz&vnces
Calf Weights, Gains and Days of Birth
Males were heavier (P < 0.01) than females at birth and at all other times, except
January (Tables 2, 3) when males were actually lighter than females, although this
difference was not significant. Differences in gains were significant only from January
to March and March to April (P < 0.01), and over the whole period from birth to
weaning (P < 0.05). During the other periods-birth to January and April to weaning-sex differences in gains were very small.
Males were born later than females, but the differences were not significant.
Previous lactational statzrs of dam
Calf Weights, Gains and Days of' Birth
In a separate analyses which excluded maidens, the previous
the dam significantly affected birth weight and gains from M
G. W. Seifert
analyses of variance for birth weight and gains from January to March and March to
April are given in Table 8.
Calves born from cows that were lactating the previous season were 1.996 kg
lighter than those that had not had a calf (P < 0.05). However, calves from previously
lactating cows gained faster than calves from the previously non-lactating cows, but
this advantage was significant only from March to April (0.026 kglday). When dam
gains were not considered as a covariate in the analyses, the differences were nonsignificant.
A non-significant weight advantage of calves from previously lactating cows
became apparent from March onwards, when their gain advantage cancelled their
birth weight deficit.
The other main effects were similar to those in the main analyses.
Dam Weights and Dam Gains
Effects of previous lactation on dam weights at January and weaning were nonsignificant, but previously lactating dams gained significantly more than those that
were not lactating,
Fig. 2. Breed x previous lactational
status of the dam interaction
on the birth weight of the calf
u
Lactating
Non-lactating
Previous lactational status of cow
Interactions
C a y Weights, Gains and Days of Birth
In the model which included the previous lactational status the interaction between
breed and previous lactational status was significant (P < 0.05) for birth weight and
is shown in Fig. 2. The birth weights of calves of lactating AX cows were much more
severely affected than those of the BX cows.
Although the AX cows calved earlier than the BX cows (P < 0.01), the differences
were small in the lactating cows, but large in non-lactating cows (Fig. 3). The BX
Selection for Growth Rate in Cattle. I
calving date was less affected by the previous lactational history, in contrast to the
AX non-lactating cows. A breed difference in calving date similar to that in nonlactating cows was found for the maidens.
Partial regressions and simple correlations
Calf Weights and Gains on Day of Birth
Calves born late in the season were heavier at birth than those born early, and this
weight advantage was maintained until April, after which it disappeared (Table 3).
Early-born calves grew faster (ADG) than late-born calves up until March (Table 5).
However, from April to weaning the late-born calves gained more rapidly than the
early-born calves, which probably reflected their phase of growth due to age.
From birth to weaning early-born calves grew faster than the late-born calves, but
the regression was not significant.
Fig. 3. Breed x previous lactational status
o
Lactating
Non. lactatmp
Previous lactational status of cow
Calf Weights and Gains on Dam Gain
Dam gains were negatively related to all gains and W/A,which indicated that cows
-with fast-gaining calves tended to lose most weight during lactation (Tables 3, 5).
,Calf Weights and Gains on Birth Weight
Heavy calves at birth remained significantly heavier than lighter calves at all ages
,(Table 3). This is to be expected, as birth weight is a component of the weights at
later ages. Birth weight was also significantly related to average daily gain from
January to March, and over the entire period from birth to weaning (Table 5).
Calf Birth Day on Birth Weight
Later-born calves were heavier (P < 0.05) than early calves (Table 7).
G . W. Seifert
Dam Weights and Gains on Calf Birth Weight and Pre-weaning Gain
Heavy calves at birth had heavier dams at January ( P < 0.01) and weaning
(P < 0.05) and fast-gaining calves depressed the gains of their dams (P c 0.01)
(Table 7).
Simple Correlations among Covariates and Dependent Variables
Correlations among the covariates and dependent variables are given in Table 9.
Table 9. Simple correlations among covariates and dependent variables
-
--
Day of
birth
-
Dam
gain
-
Birth
weight
-
Gain, birth
to Jan.
-
Gain, Jan.
to Mar.
-
Gain, Mar.
to Apr,
Gain, Apr.
to weaning
Gain, birth
to weaning
In contrast to the partial regressions, birth weight was negatively correlated with
average daily gain from birth to January, and positively correlated in all other gain
periods (P < 0.01).
Of the correlations among the covariates, only day of birth and birth weight were
significant ( P < 0.01).
The pre-weaning gains were positively correlated ( P < 0.01) with the exception of
the birth to January gains. Gain from birth to January was negatively correlated with
gain from January to March.
Discussion
Birth weight differences between the AX and BX calves were small and nonsignificant (Table 3), which is in contrast to the results of Seifert and Kennedy (1966)
and Kennedy and Chirchir (1971). However, in the previous analyses factors such as
dam age, previous lactational status and day of birth which significantly affected birth
weight, were not considered and could have accounted for some of the breed differences.
The birth weights of the AX calves were also much more affected by the previous
lactational history of their dams than those of the BX calves (Fig. 2).
The regressions of pre-weaning gains on birth weight showed that gain increased
with increasing birth weight, and this is in agreement with the results of Jeffery et al.
(1971). The regression was highest for the period from January to March, which
coincided with the most rapid growth phase (Fig. 1) and probably reflects the ability
of larger calves to utilize the flush of milk which coincided with the season.
The regression of calf gains on day of birth was negative from birth to March, but
positive from March to weaning; the overall regression (birth to weaning) was negative but non-significant. In contrast, the regressions of weights on day of birth were
positive until April but of declining magnitude, and became negative at weaning.
Selection for Growth Rate in Cattle. I
405
This declining trend reflects the erosion of the positive birth weight-day of birth
association by the negative relationship between gain and day of birth.
The strong negative regressions of the birth to March gains on day of birth support
the hypothesis of Lesmeister et al. (1972) and Seifert et al. (1974), that the older and
therefore heavier calves have the ability to utilize the flush of milk which coincides
with the break in the season and rapid growth (Fig. 1).
Increased birth weight is associated with a longer gestation, and this is reflected in
the increase in day of birth with increased birth weight as shown by the regression
coefficient (Table 6). The significantly earlier calving of the AX non-lactating cows
(Fig. 3) after removing the effect of birth weight, by covariance, therefore reflects the
earlier conception of the non-lactating AX females. The small difference in calving
date of all lactating females indicates that lactational anoestrus may be equally important in both breeds.
The effects of dam age on calf weights were similar to those reported in the literature (Blair et al. 1972). However, the differences were not significant when dam
weights at January and weaning were included as covariates in the model. This
indicates that the effect of dam age is largely through its effect on dam weight. Dam
gain, as a covariate, on the other hand, had no influence on the effects of dam age on
calf weights.
Dams that were lactating the previous year gained significantly more during the
pre-weaning period than those that were not lactating. This may have been due to
some compensatory effect, or to the fact that cows that calve regularly are better
adapted than those that do not.
Jeffery et al. (1971) and Jeffery and Berg (1972) found a positive relationship between calf weight and post-calvingweight of the dam in one year-season and a negative
relationship in another. In these data there was a negative relationship between dam
gains and calf gains and weights from March.
Males were significantly heavier than females at all ages, which is in agreement
with all other reports. Males gained significantly more than females only from
January to April, the period of most rapid gains. The males therefore appeared to
need a maximum nutritional environment in order to express their superior ability
t o grow.
Although the trends for the effects of the main factors and regressions on W/A
were similar in January to those in the other periods, there was a general lack of significance. This was mainly due to the large error variance, which probably resulted
from the relatively large variation in calf ages and weights at January.
The error coefficients of variation for W / A were lower than those for gains, and
for gains the smallest coefficient of variation coincided with the rapid-growing phase
from January to March. Gains from January to March were also repeatable with
later gains, which indicated that this may be a reliable weight to use in selecting for
high weaning weight.
Dams from the high lines were significantly heavier than those from the low lines
at January and June, although they were selected on their 2-year-old W/A, which
indicated that they had maintained their weight advantage throughout their lives.
The high line dams also managed to gain significantly more during the pre-weaning
period in spite of the higher pre-weaning gains of their calves, which indicated that
they probably were better-adapted animals.
G. W. Seifert
Selection for high W/A at 2 years significantly increased all weights, except in
January, and all gains. Birth and weaning weights as well as gains are components of
weight at 2 years, and this response to selection reflects the indirect selection or the
correlated response for these traits.
McDonald and Turner (1972) found that genetic and permanent environmental
components of maternal ability and the covariance of individual and maternal effects
accounted for 15-20 % of the variation in birth weight and 35-45 % of the variation
in daily gain from birth to weaning. The direct maternal effects due to the larger body
size of the high W / A dams probably also contribute to the pre-weaning traits. However, when the partial regressions on dam weight were included, selection still contributed significantly to the variance. Line differences after adjusting for dam weight
and gains give a minimum estimate of the genetic effect.
Whatever the cause of the response to selection, the similar results in the AX and
BX animals indicate that a large amount of additive genetic variance exists in these
populations for these traits. Productivity can therefore be improved by selecting on
these traits in this environment.
Acknowledgments
I wish to thank Mr A. J. Short and Mr K. G. Bean for their valuable technical
assistance in compiling and analysing the data, Mr H. G. Turner for his advice and
guidance, and the staff of the National Cattle Breeding Station for their assistance in
handling and weighing the cattle.
References
Blair, L. G., Wilson, L. L., and Ziegler, J. H. (1972). J. Anim. Sci. 35, 1155.
Carter, A. H. (1971). Proc. N.Z. Soc. Anim. Prod. 31, 157.
Flower, A. E., Brinks, J. S., Urick, J. J., and Willson, F. S. (1964). J. Anim. Sci. 23, 189.
Harvey, W. R. (1960). U. S. Dep. Agric., A. R. S., No. ARS-20-8.
Kennedy, J. F., and Chirchir, G. I. K. (1971). Aust. J. Exp. Agric. Anim. Husb. 11,593.
Kennedy, J. F., and Turner, H. G . (1959). CSIRO Aust. Div. Anim. Hlth. Prod. Divl. Rep. No. 8
(Ser. S.W.-3).
Jeffery, H. B., Berg, R. T., and Hardin, R. T. (1971). Can. J. Anim. Sci. 51, 561.
Jeffery, H B., and Berg, R. T. (1972). Can. J. Anim. Sci. 52,23.
Lesmeister, J . L., Burfening, P. J., and Blackwell, R. L. (1972). J. Anim. Sci. 36, 1.
McDonald, R. P., and Turner, J. W. (1972). J. Anim. Sci. 35,1146.
Newman, J. A,, Rahnefeld, G. W., and Fredeen, H. T. (1973). Can. J. Anim. Sci. 53,l.
Pahnish, 0. F., Roberson, R. L., Taylor, R. L., Brinks, J. S., Clark, R. T., and Roubicek, C. B. (1964).
J. Anim. Sci. 23, 562.
Pattie, W. A., Williams, L. G., and Dettmann, E. B. (1970). Proc. Aust. Soc. Anim. Prod. 8, 115.
Rollins, W. C., Carroll, F. D., Pollock, J. W. T., and Kudoda, M. N. (1962). J. Anim. Sci. 21, 200.
Seifert, G. W., and Kennedy, J. F. (1966). Proc. Aust. Soc. Anim. Prod. 6, 257.
Seifert, G. W., and Kennedy, J. F. (1972). Proc. Aust. Soc. Anim. Prod. 9, 143.
Seifert, G. W., Rudder, T. H., and Lapworth, J. W. (1974). Aust. J. Exp. Agric. Anim. Husb. 14, 277.
Shelby, C. E., Harvey, W. R., Clark, R. T., Quesenberry, J. R., and Woodward, R. R. (1963). J. Anim.
Sci. 22, 346.
Manuscript received 22 August 1974