- Email: [email protected]
Genetics of Growth and Reproduction in the Turkey. 17. Changes in Genetic Parameters Over Forty Generations of Selection for Increased Sixteen-Week Body Weight1
Genetics Genetics of Growth and Reproduction in the Turkey. 17. Changes in Genetic Parameters Over Forty Generations of Selection for Increased Sixteen-Week Body Weight1 K. E. Nestor, J. W. Anderson, R. A. Patterson, and S. G. Velleman2 Department of Animal Sciences, Ohio Agricultural Research and Development Center, The Ohio State University, Wooster 44691 BW in the F line over 40 generations of selection were positively associated with BW at other ages (8 and 20 wk of age and at 50% production), shank length and width at 16 wk of age, days from stimulatory lighting to production of the first egg, and egg weight but were negatively associated with egg production, intensity of lay (maximum and average clutch length and rate of lay), and walking ability. Over the 40 generations of selection, genetic increases in BW in the F line were not associated with changes in broodiness or mortality to 8 wk of age. During generations 31 to 40, BW at 8 and 20 wk of age continued to increase in the F line, but there was no significant change in adult BW, and the only significant change in reproduction traits was for average clutch length (−0.030). Because the genetic changes in some correlated traits were not consistent in all generation intervals studied, the genetic correlation between the selected trait (16-wk BW) and the correlated trait apparently changed with selection.
Key words: turkey, body weight, egg production, reproduction, genetic parameter 2008 Poultry Science 87:1971–1979 doi:10.3382/ps.2008-00137
INTRODUCTION Early reports in the literature indicated that the heritabilities (h2) for BW of turkeys during the growing period are generally large. Nestor et al. (1967) summarized that the earlier published h2 estimates of BW in selected populations were 0.40, 0.42, 0.43, and 0.36, respectively, for birds in the age groups 0 to 8, 9 to 16, 17 to 24, and greater than 24 wk of age. Based on the data from 8 commercial flocks, Arthur and Abplanalp (1975) found that the h2 estimate of 18-wk BW was 0.42. Nestor et al. (1967) found that the h2 estimates ©2008 Poultry Science Association Inc. Received April 1, 2008. Accepted June 1, 2008. 1 Salaries and research support provided by state and federal funds appropriated to the Ohio Agricultural Research and Development Center, The Ohio State University, Wooster. 2 Corresponding author: [email protected]
in selected populations were lower than estimates in a randombred control (RBC) line when estimates were based on full-sib analysis but not when estimated from regression of offspring on mid-parent. In 2 RBC populations, h2 estimates of BW based on variance component analysis and parent-offspring regression were high and ranged from 0.40 to 0.68 (McCartney, 1961; Nestor et al., 1967; Havenstein et al., 1988). When based on the sire component of variance, Havenstein et al. (1988) found that the h2 estimate of 16-wk BW differed between sexes. Toelle et al. (1990) reported that scaling effects were not responsible for the sex difference in the h2 of 16-wk BW and that the genetic correlation between BW of the 2 sexes was close to unity. Short-term selection studies suggested that the realized h2 of BW was also high. Abplanalp et al. (1963) found that the realized h2 of 8- and 24-wk BW was 0.43 and 0.62, respectively. McCartney et al. (1968) observed
1971
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
ABSTRACT A line (F) of turkeys was selected over 40 generations for increased 16-wk BW. The base population for the F line was a randombred control population that was maintained without conscious selection and used to remove yearly environmental variation in the F line. Selection was effective in increasing 16-wk BW in the F line. Selection differentials based on the mean of the selected parents minus the mean of the entire population (intended) and intended selection differentials weighted for number of offspring produced (actual) did not differ consistently, indicating that natural selection was not opposing artificial selection during the reproduction of the F line. The realized heritability of 16-wk BW in the F line, based on the linear regression of the selection response on accumulated actual selection differential, declined with selection. For both sexes combined, the realized heritability was 0.309 ± 0.022 (SE), 0.268 ± 0.033, 0.268 ± 0.026, 0.166 ± 0.016, and 0.242 ± 0.004, respectively, for generations 1 to 10, 11 to 20, 21 to 30, 31 to 40, and 1 to 40. Genetic increases in 16-wk
1972
Nestor et al. 2
MATERIALS AND METHODS The base population of the selection study was a RBC population (RBC2) established in 1966 from pooled reciprocal crosses of 2 commercial large-bodied strains (Nestor et al., 1969). The RBC2 line was maintained using a random mating system with 36 parental pairs, but full-sib matings were avoided (Nestor, 1977a). A subline (F) of the RBC2 line was initiated by mass selection only for increased 16-wk BW. In the first generation, individuals used to reproduce the RBC2 line were selected by the method of Nestor (1977a), and the remaining individuals were used to start the F line with 2 weekly hatches assigned to each line (Nestor, 1977c). Individuals originally assigned to the RBC2 line that were not used in the reproduction of the line were available for use in the F line. The method of initiating the F line assured that a large number of families were included in the base generation of the F line, thus reducing founder effects. The F line was maintained with 36 parental pairs from generations 1 through 21. In generation 22 and later, the F line was reproduced with 36 males and 72 females, with each male being mated to 2 females.
Management changes have occurred during the long-term selection study. The RBC2 and F lines were maintained in flocks hatched in April and May during generations 1 through 11 and 15 through 27. Offspring were grown in confinement with sexes combined until 8 wk of age and then range-reared until 20 (generations 9 through 27) or 24 wk of age (generations 1 through 8). In generations 12 through 14, offspring were hatched in May and June and grown entirely in confinement. To estimate the effect of the difference in environment, during generation 28, one-half of the offspring from the first hatch was reared entirely in confinement, whereas the other half was reared in confinement until 8 wk of age and then range-reared until 20 wk of age. Based on data from the first hatch in generation 28, Noble et al. (1996) reported that there was no interaction between the RBC2 and F lines and environment on growth traits, suggesting that the environmental change did not affect the selection study. Offspring from other hatches in generation 28 and in all hatches in generations 29 and later were reared in confinement, sexes separate, from hatching through 20 wk of age. Offspring from the RBC2 and F lines were fed a declining protein, 6-ration system (Naber and Touchburn, 1970) during the growing period based on the schedule for males. Some minor improvements were made in the rations during the course of the selection study. However, in all generations, both lines were fed the same rations. In all generations, the rations met or exceeded National Research Council standards. Selected males of the RBC2 and F lines were housed in a pole shelter at 20 or 24 wk of age in generations 1 through 27 and in a windowless house at 20 wk of age in generations 28 through 40. Beginning at stimulatory lighting (14 h/d), the breeder males were fed a ration containing 15.3% CP, 0.93% calcium, 0.62% phosphorus, and 2,963 kcal/kg of ME. Selected females of the RBC2 and F lines were housed in a windowless breeder house and exposed to simulated declining daylight conditions until 8 wk before stimulatory lighting. At this time, light was restricted to 6 h per day. The hens were given stimulatory lighting of 14 h light per day at an intensity of 51 lx when they were approximately 39 wk of age. The hens were fed a ration containing 15.3% CP, 2.25% calcium, 0.64% available phosphorus, and 2,751 kcal/kg of ME beginning 1 wk before stimulatory lighting. Egg production records were obtained by trapnesting the hens 4 or more times daily. Broody hens were treated to decrease broodiness by first identifying the hens and then changing the environment (Nestor and Renner, 1966).
Traits Measured Body weights were recorded at 8, 16, and 20 (generations 9 through 40) or 24 wk of age (generations 1 through 8) and when the hens first achieved approxi-
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
that the realized h of 8- and 24-wk BW was 0.44 and 0.39, respectively. Based on 4 generations of selection, Mukherjee and Friars (1970) reported that the realized h2 of 12-wk BW ranged from 0.37 to 0.57 in different base populations. Johnson and Gowe (1962) observed that the values of the parameters of the growth curve of the turkey could be changed by selecting for increased BW at different ages. In the only long-term selection study in turkeys, Nestor et al. (2000) found that the realized h2 of 16-wk BW declined with selection, being 0.31, 0.27, and 0.24, respectively, in generations 1 to 10, 11 to 20, and 21 to 30. The decline in realized h2 appeared to be slightly different for males than females. Little or no association between egg production and BW was observed in earlier studies during the first few generations of selection for either increased egg production (Kosin and Becker, 1959; Shoffner and Leighton, 1962) or increased BW (Ogasawara et al., 1963; Mukherjee and Friars, 1970). Cook et al. (1962), Clayton (1971), and Arthur and Abplanalp (1975) estimated that the genetic correlation between BW and egg production was −0.1. In a long-term selection study for increased 16-wk BW, Nestor et al. (2000) found that, in general, egg production declined with genetic increases in BW, but the changes in egg production were not consistent across generations. The purpose of the present study was to analyze direct and correlated responses to long-term selection for increased 16-wk BW of turkeys over 40 generations of selection and to estimate changes in genetic parameters by analyzing the changes occurring at 10-generation intervals. Results in early generations of selection have been summarized previously (Nestor, 1977c, 1984; Nestor et al., 1996c, 2000).
1973
GENETICS OF GROWTH AND REPRODUCTION
Data Analysis The average increase in inbreeding coefficient per generation was estimated from one-half the reciprocal of the effective population size (Falconer, 1964). The effective population size was based on variation in family size. Changes in the selected and correlated traits over generations were estimated by the linear and quadratic regression coefficients of line means on generations with the values of the F line expressed as deviation from the RBC2 line. The significance of the regression coefficients was evaluated by t-tests. Whenever possible, the selection period was divided into intervals of 10 generations to evaluate possible changes in genetic parameters with selection. A separate analysis was also completed for the entire 40 generations. In addition, to evaluate the effect of selection over the entire period, a 1-way ANOVA of traits was used to estimate the effect of line in the 10th, 20th, 30th, and 40th generations of selection in the F line. Intended and actual selection differentials of 16-wk BW in the F line were obtained. The intended selection differential was defined as the average of the selected individuals minus the population mean, and the actual selection differential was the intended selection differential weighted for the number of offspring produced that contributed to the next generation. Realized h2 were estimated by the following: 1) the linear regression of selection responses in the F line, corrected for environmental fluctuations by expressing the values as deviations from the RBC2 line, on accumulated ac-
tual selection differentials; and 2) dividing the total corrected selection response by the total accumulated actual selection differential. For the estimates based on regression, the standard errors of the regression coefficient served as an approximation of the standard errors of the estimate.
RESULTS The total increase in inbreeding over the 40 generations was 17.1 and 33.6, respectively, in the RBC2 and F lines. The respective average increase per generation was 0.43 and 0.84%.
Changes in the RBC2 Line The RBC2 line exhibited positive linear trends in 8-wk BW of males and females (P ≤ 0.01) and in average clutch length (P ≤ 0.05) during generations 31 through 40 (Table 1). Changes in BW at 16 and 20 wk of age, mortality to 8 wk of age, egg production, days from beginning of stimulatory lighting to production of the first egg, maximum and average clutch length, rate of lay, and egg weight did not change significantly in the RBC2 line from generations 31 through 40. Over the 40-generation period, male and female BW at 8 and 20 wk of age had a significant negative linear trend (P ≤ 0.01), but there was a significant (P ≤ 0.05) positive quadratic regression coefficient in all cases. Rate of lay increased by 0.1% per generation (P ≤ 0.001), and egg weight decreased by 0.12 g per generation in the RBC2 line over the entire period of the study. The quadratic regression coefficient was significantly positive for maximum clutch length over the 40 generations. No significant linear or quadratic changes in the RBC2 line were observed for 16-wk BW of males and females, mortality to 8 wk of age, egg production, days from stimulatory lighting to first egg, and average clutch length based on data from 40 generations.
Direct and Correlated Changes in the F Line The total changes in growth and reproduction traits observed in the F line are given in Tables 2 and 3, respectively. During the first 30 generations of selection in the F line, there were large gains in BW at 8, 16, and 20 wk of age in males and females, adult BW in females, and shank length and depth at 16 wk of age in both sexes (Table 2). Associated with the large increases in BW and associated leg traits was a decrease in walking ability in both males and females as indicated by increases in walking ability score. Mortality to 8 wk of age did not change significantly due to selecting for increased 16-wk BW in the F line during the first 30 generations of selection. The F line exhibited substantial total gains in BW at all ages and shank length and shank width at 16 wk of age for males and females from 31 to 40 generations of selection. Changes in walking ability scores (positive in males and negative
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
mately 50% production (females only). Mortality was recorded to 8 wk of age. At 16 wk of age in some generations, shank width and length (Nestor et al., 1985) were measured and the birds were subjectively rated for walking ability. Each bird was given a rating of 1 to 5, where 1 = the bird had legs without any lateral deviation and had no difficulty walking and 5 = birds whose legs exhibited extreme lateral deviations or had extreme difficulty walking, or both. Ratings of 2, 3, and 4 represented intermediate values between these extremes. Egg production, except in 1 generation, was recorded for a 180-d production period beginning when the first egg was laid. In generation 15, the production period was shortened to 120 d as part of a successful program of eradication of a Mycoplasma meleagridis infection from both lines (Saif and Nestor, 1983). Egg records were analyzed to obtain days required from stimulatory lighting to production of the first egg and measurements of intensity of lay (rate of lay and maximum and average clutch length) and broodiness (total days lost from periods of nonproduction of 5 or more consecutive days including those at the end of the production period) according to the methods presented by Nestor (1972). Rate of lay was obtained by the following formula: number of eggs laid/(180 d − total days broody).
1974
Nestor et al.
Table 1. Linear regression coefficients of growth and reproduction traits on generations in the randombred control line Generations Variable
11 to 20
−0.054* 0.034
−0.012 −0.385* −0.040
−0.041** 0.009
−0.016 −0.007 −0.029* 0.297 −0.876 −0.179
0.039* 0.063* 0.0143 −0.522* 2.009* −0.021
0.034** 0.025 0.039 −0.293 −0.359 0.019
−0.031***1 0.004 −0.091**1,2 0.308 −0.184 −0.076
−0.070 0.007 −0.001 −0.194
0.133 0.014 0.003 0.230
0.165 0.037* 0.003 −0.034
0.0041 0.002 0.001*** −0.120***
0.171 1.638* 0.002 0.104 0.021 0.003 −0.012
21 to 30 0.057** 0.062 0.030
31 to 40 0.061** 0.031 0.036
1 to 40 −0.040***1 0.002 −0.198***1,2
1
Significant (P ≤ 0.05) positive quadratic regression coefficient. Based on generations 9 through 40. 3 Significant (P ≤ 0.05) negative quadratic regression coefficient. 4 Based on a 180-d production period. 5 Days from beginning of stimulatory lighting (14 h/d) to production of first egg. 6 Rate of lay = number of eggs laid/(180 d − total days broody). 7 Based on a 12-wk period. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. 2
in females) and mortality to 8 wk of age did not change greatly from generations 31 to 40. Based on gains per generation, the gains in BW at 8 and 16 wk of age and shank width at 16 wk of age were similar in generations 1 to 30 and 31 to 40. Gains in BW at older ages (20 wk of age in males and females and adult weight of females) and shank length at 16 wk of age were decreased in the later generations. A loss of 23.9 eggs per hen during a 180-d production period was a correlated response to the genetic gains in BW in the F line during the first 30 generations (Table 3). During generations 31 to 40, egg production increased by a total of 5.1 eggs per hen in the F line. This increase was probably not significant, because a difference of 10 eggs or more per hen is usually required for significance with the number of birds used and the variation in egg numbers. The loss in egg production of the F line during generations 1 through 30 was associated with an increase in days required from stimulatory lighting to production of first egg, and intensity of lay as measured by average and maximum clutch length and rate of lay. Further reductions of maximum and average clutch lengths apparently occurred in the F line during generations 31 through 40, but the changes in rate of lay during this period were positive. Egg weight increased in the F line during generations 1 through 30, but the increase was minor during generations 31 through 40. Based on the deviation from the RBC2 line, the linear regression coefficients in the F line of BW at 8 and 16 wk of age in males and females on generations of selection were positive and significant for generations 1 to 10, 11 to 20, 21 to 30, and 31 to 40 (Table 4). Like-
wise, the regression coefficients in the F line of BW at 20 wk of age in males and females on generations were positive and significant for generations 11 to 20, 21 to 30, and 31 to 40. For adult BW in females in the F line, the regression coefficients were positive and significant for generations 1 to 10, 11 to 20, and 21 to 30, but there was no significant linear change for generations 31 to 40 even though a relatively large change (0.67 kg) in adult BW (Table 2) was observed. Over the entire 40-generation period in the F line, there were significant positive linear regression coefficients for BW at 8, 16, and 20 wk of age in males and females and adult BW in females. A significant positive quadratic regression coefficient was observed for BW in the F line at 16 and 20 wk of age, suggesting that the rate of gain was not as great in later generations of selection. Significant linear regression coefficients were observed in the F line for shank length and width and walking ability scores of males and females from generations 14 through 40 (Table 4). Male shank width of the F line had a significant positive quadratic regression coefficient during this period. A negative quadratic regression coefficient was observed for walking ability scores of females of the F line. Mortality to 8 wk of age did not exhibit any significant change in the F line during generations 1 to 10, 11 to 20, 21 to 30, 31 to 40, or 1 to 40. Significant negative linear regression coefficients were observed for egg production on generations of selection in the F line when expressed as a deviation from the RBC2 line for generations 1 to 10, 21 to 30, and generations 1 to 40 (Table 5). For generations 1 to 40, there was a significant positive quadratic regres-
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
Male body weights, kg 8 wk 16 wk 20 wk Female body weights, kg 8 wk 16 wk 20 wk Mortality, 0 to 8 wk, % Egg production,4 no./hen Days to first egg5 Clutch length,4 eggs Maximum Average Rate of lay4,6 Egg weight,7 g
1 to 10
1975
GENETICS OF GROWTH AND REPRODUCTION
Table 2. Effect of selecting turkeys over 40 generations for increased 16-wk body weight on body weight at various ages, shank measurements, walking ability, and mortality Line1 Variable
F − RBC2
2.95 7.73 9.84
5.82 15.63 19.57
2.87*** 7.79*** 9.73***
1.99*** 5.82*** 8.06***
2.34 5.38 6.64 8.91
4.86 11.82 14.10 17.47
2.52*** 6.44*** 7.46*** 8.56***
19.9 16.0
22.4 17.9
13.0 11.0
RBC2
Male body weight, kg 8 wk 16 wk 20 wk Female body weight, kg 8 wk 16 wk 20 wk Adult3 Shank length,4 cm Males Females Shank width,4 mm Males Females Walking ability score4,5 Males Females Mortality, 0 to 8 wk, %
1.21 1.18 5.74
Change in F from 30th generation
Change per generation 1 to 30
31 to end
0.85 1.97 1.67
0.066 0.194 0.269
0.085 0.197 0.167
1.94*** 4.84*** 5.85*** 7.89***
0.58 1.60 1.61 0.67
0.064 0.161 0.195 0.263
0.058 0.160 0.161 0.074
2.55*** 1.91***
2.27*** 1.56***
0.28 0.35
0.076 0.052
0.035 0.044
17.2 15.8
4.23*** 4.76***
3.27*** 3.85***
0.96 0.91
0.109 0.128
0.120 0.114
2.84 2.47 10.01
1.63*** 1.29*** 4.27
1.50*** 1.51*** 4.1
0.13 −0.22 0.17
0.050 0.050 0.137
0.016 −0.028 0.017
1
RBC2 = randombred control line; F = subline of the RBC2 line selected for increased 16-wk body weight. From Nestor et al. (2000). 3 Body weight when hens first achieved 50% egg production, measured in 39th generation. 4 Measured at 16 wk of age in the 38th generation. 5 Birds were subjectively rated from 1 to 5 where 1 = birds whose legs did not have any defects and had no difficulty walking and 5 = birds whose legs exhibited extreme lateral deviation or had great difficulty walking. Ratings of 2, 3, and 4 represented intermediate values. ***P ≤ 0.001. 2
sion coefficient in the change in egg production of the F line over generations. For days from stimulatory lighting to production of the first egg in the F line, a significant positive linear regression coefficient was observed only for generations 1 to 10 and 1 to 40, and a significant quadratic regression coefficient was observed in generations 1 to 40. The linear regression coefficients of measures of intensity of lay (maximum and average clutch length and rate of lay) of the F line on generations of selection were consistently negative, but not always significant, for the various generation intervals.
There was a significant positive quadratic regression coefficient for average clutch length and rate of lay of the F line in generations 1 to 40. Total days lost from broodiness in the F line did not exhibit any significant change in any generation interval. Most of the increase in egg weight in the F line occurred during generations 1 to 10 as indicated by a significant positive linear regression coefficient of egg weight on generations of selection during this period, with no significant change occurring in generations 11 to 20, 21 to 30, and 31 to 40. Over the entire period, the linear regression coef-
Table 3. Effect of selecting turkeys over 40 generations for increased 16-wk body weight on reproduction traits Line1 Variable 3
Egg production, no./hen Days to first egg4 Clutch length,3 d Maximum Average Total days broody3,5 Rate of lay,3,6 % Egg weight,7 g 1
RBC2
F
F − RBC2
Difference after 30 generations2
76.0 18.8
57.2 22.8
−18.8*** 4.0***
−23.9*** 4.3***
8.0 2.39 48.3 57.1 87
3.8 1.50 41.1 45.2 97
−4.2*** −0.89*** −7.2 −11.9*** 10***
−3.4*** −0.74*** 7.1 −13.8*** 9***
Change in F from 30th generation
1 to 30
31 to 40
5.1 −0.3
−0.797 0.143
0.510 −0.030
−0.8 −0.15 −14.3 1.9 1
−0.113 −0.025 0.227 −0.460 0.300
−0.080 −0.015 −1.43 0.190 0.100
RBC2 = randombred control line; F = subline of the RBC2 line selected for increased 16-wk body weight. From Nestor et al. (2000). 3 Based on a 180-d production period. 4 Days from beginning of stimulatory lighting (14 h/d) to production of first egg. 5 Total number of days lost during pauses in egg production of 5 or more consecutive days. 6 Rate of lay = number of eggs laid/(180 d − total days broody). 7 Based on the first 12 wk of egg production. ***P ≤ 0.001. 2
Change per generation
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
F
Difference after 30 generations2
1976
Nestor et al.
Table 4. Linear regression coefficients of growth-related traits on generations of selection in the F line after adjustment for changes in the environment1 Generations Variable
11 to 20
21 to 30
31 to 40
0.049*** 0.193***
0.054*** 0.115*** 0.193***
0.072** 0.258*** 0.347***
0.062** 0.178*** 0.284***
0.049*** 0.160***
0.050*** 0.115*** 0.158*** 0.260***
0.094*** 0.213*** 0.245*** 0.332***
0.054** 0.115*** 0.135*** −0.071
0.226***
1 to 40 0.071*** 0.196***2 0.245***2 0.065*** 0.158***2 0.180***2,3 0.222*** 0.058***2,5 0.046***5 0.131***5 0.126***5
−0.250
−0.100
0.439
0.669
0.024***5 0.033**5,6 0.012
1
The F line was selected for increased 16-wk body weight. To remove environmental variation, values for the randombred control were subtracted from those of the F line before the regression analysis. 2 Significant (P ≤ 0.05) positive quadratic regression coefficient. 3 Based on generations 9 through 40. 4 Body weight when hens first achieved 50% production. 5 Based on generations 14 through 40. 6 Significant (P ≤ 0.05) negative quadratic regression coefficient. **P ≤ 0.01; ***P ≤ 0.001.
ficient of egg weight on generations of selection was positive and significant, but there was a significant positive quadratic component in the change.
Selection Intensity, Selection Differentials, and Realized Heritabilities The number of birds available at 16 wk of age and percentage of offspring selected to reproduce the F line
are shown in Table 6. For both sexes combined, the percentage selected was 24.4 for generations 1 to 40. Selection pressure, as measured by the percentage selected, was greatest in generations 31 to 40, lowest in generations 11 to 20, and similar in generations 1 to 10 and 21 to 30. The number of females used to reproduce the F line was increased in generation 22 from 36 to 72. The intended and actual selection differentials did not consistently differ in the F line (Table 6). Realized
Table 5. Linear regression coefficients of reproduction traits on generations of selection in the F line after adjustment for changes in the environment1 Generations Variable Egg production,2 no./hen Days to first egg4 Clutch length,2 d Maximum Average Total days broody2,5 Rate of lay,2,6 % Egg weight,7 g
1 to 10
11 to 20
21 to 30
31 to 40
1 to 40
−2.961** 0.143**
−0.113 0.251
−1.061* 0.032
0.649 −0.071
−0.019***3 0.076***3
−3.64** −0.068** 2.405 −0.011*** 0.224*
−0.092 −0.050* −0.261 −0.010* −0.194
−0.157* −0.021 0.699 −0.003 0.181
−0.122 −0.030** 1.178 −0.001 −0.034
−0.084*** −0.019***3 −0.472 −0.003***3 0.243***3
1 The F line was selected for increased 16-wk body weight. To remove environmental variation, values for the randombred control were subtracted from those of the F line before the regression analysis. 2 Based on a 180-d production period. 3 Significant (P ≤ 0.05) positive quadratic regression coefficient. 4 Days from beginning of stimulatory lighting (14 h/d) to production of first egg. 5 Total number of days lost during pauses in egg production of 5 or more consecutive days. 6 Rate of lay = number of eggs laid/(180 d − total days broody). 7 Based on a 12-wk period at beginning of lay. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001.
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
Male body weights, kg 8 wk 16 wk 20 wk Female body weights, kg 8 wk 16 wk 20 wk Adult4 Shank length, cm Males Females Shank width, mm Males Females Walking ability score Males Females Mortality, 0 to 8 wk, %
1 to 10
1977
GENETICS OF GROWTH AND REPRODUCTION 1
Table 6. Selection intensity, selection differentials, and realized heritabilities in the F line
Generations Variable
1 to 10
11 to 20
21 to 30
31 to 40
364 275
324 384
418 592
1 to 40
2
376 305 24.8 23.2 24.0
32.4 29.8 31.1
0.6269 0.6178 −0.0091
0.6201 0.6128 −0.0073
0.5372 0.5398 0.0026
0.4409 0.4373 −0.0036
0.337 ± 0.030 0.280 ± 0.021 0.309 ± 0.022 0.392 0.321 0.356
0.320 ± 0.041 0.217 ± 0.033 0.268 ± 0.033 0.341 0.255 0.298
18.0 31.1 24.6
370 389
12.2 24.3 18.2
21.8 27.1 24.4
1.3054 1.3090 0.0036
1.2038 1.2006 −0.0032
0.9390 0.9350 −0.0040
0.6087 0.6141 0.0054
0.5280 0.5275 0.0005
0.5287 0.5297 0.0010
0.320 ± 0.026 0.218 ± 0.028 0.268 ± 0.026 0.306 0.241 0.274
0.202 ± 0.017 0.130 ± 0.025 0.166 ± 0.016 0.270 0.223 0.247
0.268 ± 0.004 0.215 ± 0.004 0.242 ± 0.004 0.270 0.223 0.247
1
The F line was developed from a randombred control population (RBC2) by selecting for increased 16-wk body weight. Based on birds alive at 16 wk of age. 3 I = intended selection differential (mean of selected birds minus mean of all birds); A = intended selection differential weighted by the number of offspring hatched. 4 Heritability ± standard error. 2
h2 of 16-wk BW in the F line were greater in males than in females and declined with selection, as estimated by the linear regression of response on the accumulated selection differential. For both sexes combined, the decline in realized h2 averaged about 4.8 percentage points for each 10 generations of selection, and the decline was largest (7.6 percentage points) in generations 31 to 40 from generations 21 to 30. The average decline in h2 estimates for each 10-generation interval was 4.5 and 5.0 percentage points in males and females, respectively. The realized h2 estimates based on estimates at certain points in time (last generation of each interval of selection) were, in general, greater than those based on regression for generations 1 to 10, 11 to 20, 21 to 30, and 31 to 40, but estimates by both methods were very similar for the entire selection period.
DISCUSSION Usually some changes in environment occur during the course of a long-term selection study. A major change (from range-rearing to complete confinement) in the rearing environment for the lines involved in the present study was made in generation 28. To evaluate the influence of this environmental change on line comparisons, Noble et al. (1996) used individuals from the first hatch in that generation to compare the performance of various lines, including those used in the current study, in the 2 environments. A sample of full-
sibs from each family was reared in each environment. There was no line × environment interaction for BW at 16 and 20 wk of age and leg measurements (shank length, width, and depth) at 16 wk of age. The interaction was significant for walking ability score of males at 16 wk of age. However, the cause of the significant interaction was due to variation in performance of a line not included in the present study. Therefore, the change in rearing environment probably did not affect the results of the current study. Significant trends over generations were observed for several traits of the RBC2 line for various periods of measurement. The direction of the changes was not always consistent over the course of the selection study. Studies by Nestor (1977b) and Noble et al. (1995) using the RBC2 line and other RBC populations suggested that the changes in the control lines were due to environmental changes. Long-term selection studies for increased BW have been conducted at Virginia Tech using chickens (Dunnington and Siegel, 1996), at the University of Georgia (Anthony et al., 1996; Marks, 1996), and at The Ohio State University (Nestor et al., 1996a,b) using Japanese quail and at The Ohio State University (Nestor et al., 1996c, 2000) using turkeys. Direct responses to selection have been measured in all of the selection studies. In general, realized h2 estimates declined with selection in chickens (Liu et al., 1994), in Japanese quail at the University of Georgia (Marks, 1996) and The
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
Number of birds RBC2 line F line Percentage selected Males Females Both sexes Selection differentials,3 kg Males I A I − A Females I A I − A Realized heritability Linear regression4 Males Females Both sexes Final generation Males Females Both sexes
1978
Nestor et al.
increased in the F line only during generations 1 to 10. The results of the present study and those of Nestor et al. (2000) suggested that the magnitude of genetic correlations, as well as the h2, may change with selection. Nestor (1971) postulated that the change in the genetic relationship between egg production and BW during the growing period in turkeys selected for increased egg production was due to the relationship between intensity of lay, egg production, and BW. In early generations of a turkey line selected for increased egg production, large increases in egg production were not associated with decreases in BW. During this time, the increases in egg production were primarily associated with decreases in broodiness. In later generations of the egg line, when increases in intensity of lay had a greater influence on the increases in egg production of the egg line than broodiness, BW declined. The decline in egg production observed in the F line in the present study and that of Nestor et al. (2000) was primarily due to reduction in intensity of lay, and there was no significant change in broodiness. The results of Nestor et al. (2000) and the present study support the hypothesis of Nestor (1971) that the negative association of BW and egg production is mediated through intensity of lay.
REFERENCES Abplanalp, H., F. X. Ogasawara, and V. S. Asmundson. 1963. Influence of selection for body weight at different ages on growth of turkeys. Br. Poult. Sci. 4:71–82. Anthony, N. B., K. E. Nestor, and H. L. Marks. 1996. Shortterm selection for four-week body weight in Japanese quail. Poult. Sci. 75:1192–1197. Arthur, J. A., and H. Abplanalp. 1975. Linear estimates of heritability and genetic correlation for egg production, bodyweight, conformation and egg weight of turkeys. Poult. Sci. 54:11–23. Clayton, G. A. 1971. Egg production in turkeys. Br. Poult. Sci. 12:463–474. Cook, R. E., W. L. Blow, C. C. Cockerham, and E. W. Glazener. 1962. Improvement of reproductive traits and body measurements of turkeys. Poult. Sci. 41:556–563. Dunnington, E. A., and P. B. Siegel. 1996. Long-term divergent selection for eight-week body weight in White Plymouth Rock chickens. Poult. Sci. 75:1168–1179. Falconer, D. S. 1964. Introduction to Quantitative Genetics. Oliver and Boyd, Edinburgh, UK. Havenstein, G. B., K. E. Nestor, V. D. Toelle, and W. L. Bacon. 1988. Estimates of genetic parameters in turkeys. 1. Body weight and skeletal characteristics. Poult. Sci. 67:1378–1387. Johnson, A. S., and R. S. Gowe. 1962. Modification of the growth pattern of the domestic turkey by selection at two ages. Pages 57–63 in Proceedings of the 12th World’s Poultry Congress, Sydney, Australia. Kosin, I. L., and W. A. Becker. 1959. Shifts in the phenotype of lines within a turkey strain, following several generations of selective breeding. Poult. Sci. 38:1220. (Abstr.) Liu, G., E. A. Dunnington, and P. B. Siegel. 1994. Responses to long-term divergent selection for eight-week body weight in chickens. Poult. Sci. 73:1642–1650. Marks, H. L. 1996. Long-term selection for body weight in Japanese quail under different environments. Poult. Sci. 75:1198–1203.
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
Ohio State University (Nestor et al., 1996a), and in turkeys (Nestor et al., 2000; present study). In the present study, the realized h2 of 16-wk BW based on the linear regression of response on accumulated actual selection differentials was 0.202, 0.130, and 0.166, respectively, for males, females, and sexes combined during generations 31 to 40, which indicated that a plateau in response is not likely in the near future of this line. Estimates of realized h2 of 16-wk BW in the F line over the entire 40 generations of selection were very similar when based on linear regression or response divided by total accumulated actual selection differentials at the 40th generation. When the selection period was divided into 10-generation intervals, h2 estimates based on a single point measurement of response divided by the accumulated actual selection differentials in the final generation were generally greater than those obtained by regression of response on accumulated actual selection differentials (Nestor et al., 2000; present study). During reproduction, natural selection was apparently not opposing artificial selection, because the intended selection differential (deviation of selected birds from the population mean) and actual selection differential (intended selection differential weighted by the number of offspring produced) did not differ (Nestor et al., 2000; present study). This result was unexpected, because genetic increases in BW in the F line were associated with a decrease in egg production (Nestor et al., 2000; present study) and with a decrease in hatch of fertile eggs (Nestor et al., 2000). The F line was reproduced early in the reproduction period when egg production, fertility, and hatch of fertile eggs were at a maximum. Likely, if the reproduction period was extended over a longer period of time, the intended and actual selection differentials would be different. Measurements of correlated responses were made routinely for some traits and only periodically for other traits in the chicken selection study at Virginia Tech and in the Japanese quail selection study at the University of Georgia. In the Japanese quail selection study at The Ohio State University (Nestor et al., 1996b) and in the turkey study at The Ohio State University (Nestor et al., 2000; present study), many correlated traits were measured most generations so that changes with selection in the correlated traits could be measured. The results of the study of Nestor et al. (2000) and the present study indicated that the genetic relationship, particularly in magnitude, between 16-wk BW and many correlated traits changed with selection. Based on the linear regression coefficients on generations of selection, the changes in number of eggs laid were negative and significant for only generations 1 to 10 and 21 to 30. For intensity of lay traits (maximum and average clutch lengths and rate of lay), the changes were larger in generations 1 to 10 than in other generation intervals. Of the intensity of lay measures, only average clutch length changed significantly (−0.030, P ≤ 0.01) during generations 31 to 40. Days from stimulatory lighting to production of the first egg and egg weight
GENETICS OF GROWTH AND REPRODUCTION
cursor in Coturnix coturnix japonica. 11. Correlated responses over thirty generations. Poult. Sci. 75:472–477. Nestor, K. E., W. L. Bacon, Y. M. Saif, and P. A. Renner. 1985. The influence of genetic increases in shank width on body weight, walking ability, and reproduction of turkeys. Poult. Sci. 64:2248–2255. Nestor, K. E., M. G. McCartney, and N. Bachev. 1969. Relative contribution of genetics and environment to turkey improvement. Poult. Sci. 48:1944–1949. Nestor, K. E., M. G. McCartney, and W. R. Harvey. 1967. Genetics of growth and reproduction in the turkey. 1. Genetic and non-genetic variations in body weight and body measurements. Poult. Sci. 46:1374–1384. Nestor, K. E., D. O. Noble, J. Zhu, and Y. Moritsu. 1996c. Direct and correlated responses to long-term selection for increased body weight and egg production in turkeys. Poult. Sci. 75:1180–1191. Nestor, K. E., and P. A. Renner. 1966. New management system for broody turkey hens. Ohio Rep. 51:83–84. Noble, D. O., D. A. Emmerson, and K. E. Nestor. 1995. The stability of three randombred control lines of turkeys. Poult. Sci. 74:1074–1078. Noble, D. O., K. E. Nestor, and C. R. Polley. 1996. Range and confinement rearing of four genetic lines of turkeys. 1. Effects on growth, mortality, and walking ability. Poult. Sci. 75:160–164. Ogasawara, F. X., H. Abplanalp, and V. S. Asmundson. 1963. Effect of selection for body weight on reproduction in turkey hens. Poult. Sci. 42:838–842. Saif, Y. M., and K. E. Nestor. 1983. Eradication of a Mycoplasma meleagridis in an experimental flock of turkeys. Ohio Poult. Pointers 22:2. Shoffner, R. N., and A. T. Leighton. 1962. Photoperiodicity and selection for egg production in the turkey (Meleagris gallapavo). Pages 37–41 in Proceedings of the 12th World’s Poultry Congress, Sydney, Australia. Toelle, V. D., G. B. Havenstein, K. E. Nestor, and W. L. Bacon. 1990. Estimates of genetic parameters in turkeys. 3. Sexual dimorphism and its implication in selection procedures. Poult. Sci. 69:1634–1643.
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
McCartney, M. G. 1961. Heritabilities and correlations for body weight and conformation in a randombred population of turkeys. Poult. Sci. 40:1694–1700. McCartney, M. G., K. E. Nestor, and W. R. Harvey. 1968. Genetics of growth and reproduction in the turkey. 2. Selection for increased body weight and egg production. Poult. Sci. 47:981–990. Mukherjee, T. K., and G. W. Friars. 1970. Heritability estimates and selection responses of some growth and reproductive traits in control and early growth selected strains of turkeys. Poult. Sci. 49:1215–1222. Naber, E. C., and S. P. Touchburn. 1970. Ohio Poultry Rations. Ohio Cooperative Extension Service Bulletin 343. The Ohio State University, Columbus. Nestor, K. E. 1971. Genetics of growth and reproduction in the turkey. 3. Further selection for increased egg production. Poult. Sci. 50:1672–1682. Nestor, K. E. 1972. Broodiness, intensity of lay, and total egg production of turkeys. Poult. Sci. 51:86–92. Nestor, K. E. 1977a. The use of a paired mating system for the maintenance of experimental populations of turkeys. Poult. Sci. 56:60–65. Nestor, K. E. 1977b. The stability of two randombred control populations of turkeys. Poult. Sci. 56:54–57. Nestor, K. E. 1977c. Genetics of growth and reproduction in the turkey. 5. Selection for increased body weight alone and in combination with increased egg production. Poult. Sci. 56:337–347. Nestor, K. E. 1984. Genetics of growth and reproduction in the turkey. 9. Long-term selection for increased 16-wk body weight. Poult. Sci. 63:2114–2122. Nestor, K. E., J. W. Anderson, and R. A. Patterson. 2000. Genetics of growth and reproduction in the turkey. 14. Changes in genetic parameters over thirty generations of selection for increased body weight. Poult. Sci. 79:445– 452. Nestor, K. E., W. L. Bacon, N. B. Anthony, and D. O. Noble. 1996a. Divergent selection for body weight and yolk precursor in Coturnix coturnix japonica. 10. Response to selection over thirty generations. Poult. Sci. 75:303–310. Nestor, K. E., W. L. Bacon, N. B. Anthony, and D. O. Noble. 1996b. Divergent selection for body weight and yolk pre-
1979
Key words: turkey, body weight, egg production, reproduction, genetic parameter 2008 Poultry Science 87:1971–1979 doi:10.3382/ps.2008-00137
INTRODUCTION Early reports in the literature indicated that the heritabilities (h2) for BW of turkeys during the growing period are generally large. Nestor et al. (1967) summarized that the earlier published h2 estimates of BW in selected populations were 0.40, 0.42, 0.43, and 0.36, respectively, for birds in the age groups 0 to 8, 9 to 16, 17 to 24, and greater than 24 wk of age. Based on the data from 8 commercial flocks, Arthur and Abplanalp (1975) found that the h2 estimate of 18-wk BW was 0.42. Nestor et al. (1967) found that the h2 estimates ©2008 Poultry Science Association Inc. Received April 1, 2008. Accepted June 1, 2008. 1 Salaries and research support provided by state and federal funds appropriated to the Ohio Agricultural Research and Development Center, The Ohio State University, Wooster. 2 Corresponding author: [email protected]
in selected populations were lower than estimates in a randombred control (RBC) line when estimates were based on full-sib analysis but not when estimated from regression of offspring on mid-parent. In 2 RBC populations, h2 estimates of BW based on variance component analysis and parent-offspring regression were high and ranged from 0.40 to 0.68 (McCartney, 1961; Nestor et al., 1967; Havenstein et al., 1988). When based on the sire component of variance, Havenstein et al. (1988) found that the h2 estimate of 16-wk BW differed between sexes. Toelle et al. (1990) reported that scaling effects were not responsible for the sex difference in the h2 of 16-wk BW and that the genetic correlation between BW of the 2 sexes was close to unity. Short-term selection studies suggested that the realized h2 of BW was also high. Abplanalp et al. (1963) found that the realized h2 of 8- and 24-wk BW was 0.43 and 0.62, respectively. McCartney et al. (1968) observed
1971
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
ABSTRACT A line (F) of turkeys was selected over 40 generations for increased 16-wk BW. The base population for the F line was a randombred control population that was maintained without conscious selection and used to remove yearly environmental variation in the F line. Selection was effective in increasing 16-wk BW in the F line. Selection differentials based on the mean of the selected parents minus the mean of the entire population (intended) and intended selection differentials weighted for number of offspring produced (actual) did not differ consistently, indicating that natural selection was not opposing artificial selection during the reproduction of the F line. The realized heritability of 16-wk BW in the F line, based on the linear regression of the selection response on accumulated actual selection differential, declined with selection. For both sexes combined, the realized heritability was 0.309 ± 0.022 (SE), 0.268 ± 0.033, 0.268 ± 0.026, 0.166 ± 0.016, and 0.242 ± 0.004, respectively, for generations 1 to 10, 11 to 20, 21 to 30, 31 to 40, and 1 to 40. Genetic increases in 16-wk
1972
Nestor et al. 2
MATERIALS AND METHODS The base population of the selection study was a RBC population (RBC2) established in 1966 from pooled reciprocal crosses of 2 commercial large-bodied strains (Nestor et al., 1969). The RBC2 line was maintained using a random mating system with 36 parental pairs, but full-sib matings were avoided (Nestor, 1977a). A subline (F) of the RBC2 line was initiated by mass selection only for increased 16-wk BW. In the first generation, individuals used to reproduce the RBC2 line were selected by the method of Nestor (1977a), and the remaining individuals were used to start the F line with 2 weekly hatches assigned to each line (Nestor, 1977c). Individuals originally assigned to the RBC2 line that were not used in the reproduction of the line were available for use in the F line. The method of initiating the F line assured that a large number of families were included in the base generation of the F line, thus reducing founder effects. The F line was maintained with 36 parental pairs from generations 1 through 21. In generation 22 and later, the F line was reproduced with 36 males and 72 females, with each male being mated to 2 females.
Management changes have occurred during the long-term selection study. The RBC2 and F lines were maintained in flocks hatched in April and May during generations 1 through 11 and 15 through 27. Offspring were grown in confinement with sexes combined until 8 wk of age and then range-reared until 20 (generations 9 through 27) or 24 wk of age (generations 1 through 8). In generations 12 through 14, offspring were hatched in May and June and grown entirely in confinement. To estimate the effect of the difference in environment, during generation 28, one-half of the offspring from the first hatch was reared entirely in confinement, whereas the other half was reared in confinement until 8 wk of age and then range-reared until 20 wk of age. Based on data from the first hatch in generation 28, Noble et al. (1996) reported that there was no interaction between the RBC2 and F lines and environment on growth traits, suggesting that the environmental change did not affect the selection study. Offspring from other hatches in generation 28 and in all hatches in generations 29 and later were reared in confinement, sexes separate, from hatching through 20 wk of age. Offspring from the RBC2 and F lines were fed a declining protein, 6-ration system (Naber and Touchburn, 1970) during the growing period based on the schedule for males. Some minor improvements were made in the rations during the course of the selection study. However, in all generations, both lines were fed the same rations. In all generations, the rations met or exceeded National Research Council standards. Selected males of the RBC2 and F lines were housed in a pole shelter at 20 or 24 wk of age in generations 1 through 27 and in a windowless house at 20 wk of age in generations 28 through 40. Beginning at stimulatory lighting (14 h/d), the breeder males were fed a ration containing 15.3% CP, 0.93% calcium, 0.62% phosphorus, and 2,963 kcal/kg of ME. Selected females of the RBC2 and F lines were housed in a windowless breeder house and exposed to simulated declining daylight conditions until 8 wk before stimulatory lighting. At this time, light was restricted to 6 h per day. The hens were given stimulatory lighting of 14 h light per day at an intensity of 51 lx when they were approximately 39 wk of age. The hens were fed a ration containing 15.3% CP, 2.25% calcium, 0.64% available phosphorus, and 2,751 kcal/kg of ME beginning 1 wk before stimulatory lighting. Egg production records were obtained by trapnesting the hens 4 or more times daily. Broody hens were treated to decrease broodiness by first identifying the hens and then changing the environment (Nestor and Renner, 1966).
Traits Measured Body weights were recorded at 8, 16, and 20 (generations 9 through 40) or 24 wk of age (generations 1 through 8) and when the hens first achieved approxi-
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
that the realized h of 8- and 24-wk BW was 0.44 and 0.39, respectively. Based on 4 generations of selection, Mukherjee and Friars (1970) reported that the realized h2 of 12-wk BW ranged from 0.37 to 0.57 in different base populations. Johnson and Gowe (1962) observed that the values of the parameters of the growth curve of the turkey could be changed by selecting for increased BW at different ages. In the only long-term selection study in turkeys, Nestor et al. (2000) found that the realized h2 of 16-wk BW declined with selection, being 0.31, 0.27, and 0.24, respectively, in generations 1 to 10, 11 to 20, and 21 to 30. The decline in realized h2 appeared to be slightly different for males than females. Little or no association between egg production and BW was observed in earlier studies during the first few generations of selection for either increased egg production (Kosin and Becker, 1959; Shoffner and Leighton, 1962) or increased BW (Ogasawara et al., 1963; Mukherjee and Friars, 1970). Cook et al. (1962), Clayton (1971), and Arthur and Abplanalp (1975) estimated that the genetic correlation between BW and egg production was −0.1. In a long-term selection study for increased 16-wk BW, Nestor et al. (2000) found that, in general, egg production declined with genetic increases in BW, but the changes in egg production were not consistent across generations. The purpose of the present study was to analyze direct and correlated responses to long-term selection for increased 16-wk BW of turkeys over 40 generations of selection and to estimate changes in genetic parameters by analyzing the changes occurring at 10-generation intervals. Results in early generations of selection have been summarized previously (Nestor, 1977c, 1984; Nestor et al., 1996c, 2000).
1973
GENETICS OF GROWTH AND REPRODUCTION
Data Analysis The average increase in inbreeding coefficient per generation was estimated from one-half the reciprocal of the effective population size (Falconer, 1964). The effective population size was based on variation in family size. Changes in the selected and correlated traits over generations were estimated by the linear and quadratic regression coefficients of line means on generations with the values of the F line expressed as deviation from the RBC2 line. The significance of the regression coefficients was evaluated by t-tests. Whenever possible, the selection period was divided into intervals of 10 generations to evaluate possible changes in genetic parameters with selection. A separate analysis was also completed for the entire 40 generations. In addition, to evaluate the effect of selection over the entire period, a 1-way ANOVA of traits was used to estimate the effect of line in the 10th, 20th, 30th, and 40th generations of selection in the F line. Intended and actual selection differentials of 16-wk BW in the F line were obtained. The intended selection differential was defined as the average of the selected individuals minus the population mean, and the actual selection differential was the intended selection differential weighted for the number of offspring produced that contributed to the next generation. Realized h2 were estimated by the following: 1) the linear regression of selection responses in the F line, corrected for environmental fluctuations by expressing the values as deviations from the RBC2 line, on accumulated ac-
tual selection differentials; and 2) dividing the total corrected selection response by the total accumulated actual selection differential. For the estimates based on regression, the standard errors of the regression coefficient served as an approximation of the standard errors of the estimate.
RESULTS The total increase in inbreeding over the 40 generations was 17.1 and 33.6, respectively, in the RBC2 and F lines. The respective average increase per generation was 0.43 and 0.84%.
Changes in the RBC2 Line The RBC2 line exhibited positive linear trends in 8-wk BW of males and females (P ≤ 0.01) and in average clutch length (P ≤ 0.05) during generations 31 through 40 (Table 1). Changes in BW at 16 and 20 wk of age, mortality to 8 wk of age, egg production, days from beginning of stimulatory lighting to production of the first egg, maximum and average clutch length, rate of lay, and egg weight did not change significantly in the RBC2 line from generations 31 through 40. Over the 40-generation period, male and female BW at 8 and 20 wk of age had a significant negative linear trend (P ≤ 0.01), but there was a significant (P ≤ 0.05) positive quadratic regression coefficient in all cases. Rate of lay increased by 0.1% per generation (P ≤ 0.001), and egg weight decreased by 0.12 g per generation in the RBC2 line over the entire period of the study. The quadratic regression coefficient was significantly positive for maximum clutch length over the 40 generations. No significant linear or quadratic changes in the RBC2 line were observed for 16-wk BW of males and females, mortality to 8 wk of age, egg production, days from stimulatory lighting to first egg, and average clutch length based on data from 40 generations.
Direct and Correlated Changes in the F Line The total changes in growth and reproduction traits observed in the F line are given in Tables 2 and 3, respectively. During the first 30 generations of selection in the F line, there were large gains in BW at 8, 16, and 20 wk of age in males and females, adult BW in females, and shank length and depth at 16 wk of age in both sexes (Table 2). Associated with the large increases in BW and associated leg traits was a decrease in walking ability in both males and females as indicated by increases in walking ability score. Mortality to 8 wk of age did not change significantly due to selecting for increased 16-wk BW in the F line during the first 30 generations of selection. The F line exhibited substantial total gains in BW at all ages and shank length and shank width at 16 wk of age for males and females from 31 to 40 generations of selection. Changes in walking ability scores (positive in males and negative
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
mately 50% production (females only). Mortality was recorded to 8 wk of age. At 16 wk of age in some generations, shank width and length (Nestor et al., 1985) were measured and the birds were subjectively rated for walking ability. Each bird was given a rating of 1 to 5, where 1 = the bird had legs without any lateral deviation and had no difficulty walking and 5 = birds whose legs exhibited extreme lateral deviations or had extreme difficulty walking, or both. Ratings of 2, 3, and 4 represented intermediate values between these extremes. Egg production, except in 1 generation, was recorded for a 180-d production period beginning when the first egg was laid. In generation 15, the production period was shortened to 120 d as part of a successful program of eradication of a Mycoplasma meleagridis infection from both lines (Saif and Nestor, 1983). Egg records were analyzed to obtain days required from stimulatory lighting to production of the first egg and measurements of intensity of lay (rate of lay and maximum and average clutch length) and broodiness (total days lost from periods of nonproduction of 5 or more consecutive days including those at the end of the production period) according to the methods presented by Nestor (1972). Rate of lay was obtained by the following formula: number of eggs laid/(180 d − total days broody).
1974
Nestor et al.
Table 1. Linear regression coefficients of growth and reproduction traits on generations in the randombred control line Generations Variable
11 to 20
−0.054* 0.034
−0.012 −0.385* −0.040
−0.041** 0.009
−0.016 −0.007 −0.029* 0.297 −0.876 −0.179
0.039* 0.063* 0.0143 −0.522* 2.009* −0.021
0.034** 0.025 0.039 −0.293 −0.359 0.019
−0.031***1 0.004 −0.091**1,2 0.308 −0.184 −0.076
−0.070 0.007 −0.001 −0.194
0.133 0.014 0.003 0.230
0.165 0.037* 0.003 −0.034
0.0041 0.002 0.001*** −0.120***
0.171 1.638* 0.002 0.104 0.021 0.003 −0.012
21 to 30 0.057** 0.062 0.030
31 to 40 0.061** 0.031 0.036
1 to 40 −0.040***1 0.002 −0.198***1,2
1
Significant (P ≤ 0.05) positive quadratic regression coefficient. Based on generations 9 through 40. 3 Significant (P ≤ 0.05) negative quadratic regression coefficient. 4 Based on a 180-d production period. 5 Days from beginning of stimulatory lighting (14 h/d) to production of first egg. 6 Rate of lay = number of eggs laid/(180 d − total days broody). 7 Based on a 12-wk period. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. 2
in females) and mortality to 8 wk of age did not change greatly from generations 31 to 40. Based on gains per generation, the gains in BW at 8 and 16 wk of age and shank width at 16 wk of age were similar in generations 1 to 30 and 31 to 40. Gains in BW at older ages (20 wk of age in males and females and adult weight of females) and shank length at 16 wk of age were decreased in the later generations. A loss of 23.9 eggs per hen during a 180-d production period was a correlated response to the genetic gains in BW in the F line during the first 30 generations (Table 3). During generations 31 to 40, egg production increased by a total of 5.1 eggs per hen in the F line. This increase was probably not significant, because a difference of 10 eggs or more per hen is usually required for significance with the number of birds used and the variation in egg numbers. The loss in egg production of the F line during generations 1 through 30 was associated with an increase in days required from stimulatory lighting to production of first egg, and intensity of lay as measured by average and maximum clutch length and rate of lay. Further reductions of maximum and average clutch lengths apparently occurred in the F line during generations 31 through 40, but the changes in rate of lay during this period were positive. Egg weight increased in the F line during generations 1 through 30, but the increase was minor during generations 31 through 40. Based on the deviation from the RBC2 line, the linear regression coefficients in the F line of BW at 8 and 16 wk of age in males and females on generations of selection were positive and significant for generations 1 to 10, 11 to 20, 21 to 30, and 31 to 40 (Table 4). Like-
wise, the regression coefficients in the F line of BW at 20 wk of age in males and females on generations were positive and significant for generations 11 to 20, 21 to 30, and 31 to 40. For adult BW in females in the F line, the regression coefficients were positive and significant for generations 1 to 10, 11 to 20, and 21 to 30, but there was no significant linear change for generations 31 to 40 even though a relatively large change (0.67 kg) in adult BW (Table 2) was observed. Over the entire 40-generation period in the F line, there were significant positive linear regression coefficients for BW at 8, 16, and 20 wk of age in males and females and adult BW in females. A significant positive quadratic regression coefficient was observed for BW in the F line at 16 and 20 wk of age, suggesting that the rate of gain was not as great in later generations of selection. Significant linear regression coefficients were observed in the F line for shank length and width and walking ability scores of males and females from generations 14 through 40 (Table 4). Male shank width of the F line had a significant positive quadratic regression coefficient during this period. A negative quadratic regression coefficient was observed for walking ability scores of females of the F line. Mortality to 8 wk of age did not exhibit any significant change in the F line during generations 1 to 10, 11 to 20, 21 to 30, 31 to 40, or 1 to 40. Significant negative linear regression coefficients were observed for egg production on generations of selection in the F line when expressed as a deviation from the RBC2 line for generations 1 to 10, 21 to 30, and generations 1 to 40 (Table 5). For generations 1 to 40, there was a significant positive quadratic regres-
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
Male body weights, kg 8 wk 16 wk 20 wk Female body weights, kg 8 wk 16 wk 20 wk Mortality, 0 to 8 wk, % Egg production,4 no./hen Days to first egg5 Clutch length,4 eggs Maximum Average Rate of lay4,6 Egg weight,7 g
1 to 10
1975
GENETICS OF GROWTH AND REPRODUCTION
Table 2. Effect of selecting turkeys over 40 generations for increased 16-wk body weight on body weight at various ages, shank measurements, walking ability, and mortality Line1 Variable
F − RBC2
2.95 7.73 9.84
5.82 15.63 19.57
2.87*** 7.79*** 9.73***
1.99*** 5.82*** 8.06***
2.34 5.38 6.64 8.91
4.86 11.82 14.10 17.47
2.52*** 6.44*** 7.46*** 8.56***
19.9 16.0
22.4 17.9
13.0 11.0
RBC2
Male body weight, kg 8 wk 16 wk 20 wk Female body weight, kg 8 wk 16 wk 20 wk Adult3 Shank length,4 cm Males Females Shank width,4 mm Males Females Walking ability score4,5 Males Females Mortality, 0 to 8 wk, %
1.21 1.18 5.74
Change in F from 30th generation
Change per generation 1 to 30
31 to end
0.85 1.97 1.67
0.066 0.194 0.269
0.085 0.197 0.167
1.94*** 4.84*** 5.85*** 7.89***
0.58 1.60 1.61 0.67
0.064 0.161 0.195 0.263
0.058 0.160 0.161 0.074
2.55*** 1.91***
2.27*** 1.56***
0.28 0.35
0.076 0.052
0.035 0.044
17.2 15.8
4.23*** 4.76***
3.27*** 3.85***
0.96 0.91
0.109 0.128
0.120 0.114
2.84 2.47 10.01
1.63*** 1.29*** 4.27
1.50*** 1.51*** 4.1
0.13 −0.22 0.17
0.050 0.050 0.137
0.016 −0.028 0.017
1
RBC2 = randombred control line; F = subline of the RBC2 line selected for increased 16-wk body weight. From Nestor et al. (2000). 3 Body weight when hens first achieved 50% egg production, measured in 39th generation. 4 Measured at 16 wk of age in the 38th generation. 5 Birds were subjectively rated from 1 to 5 where 1 = birds whose legs did not have any defects and had no difficulty walking and 5 = birds whose legs exhibited extreme lateral deviation or had great difficulty walking. Ratings of 2, 3, and 4 represented intermediate values. ***P ≤ 0.001. 2
sion coefficient in the change in egg production of the F line over generations. For days from stimulatory lighting to production of the first egg in the F line, a significant positive linear regression coefficient was observed only for generations 1 to 10 and 1 to 40, and a significant quadratic regression coefficient was observed in generations 1 to 40. The linear regression coefficients of measures of intensity of lay (maximum and average clutch length and rate of lay) of the F line on generations of selection were consistently negative, but not always significant, for the various generation intervals.
There was a significant positive quadratic regression coefficient for average clutch length and rate of lay of the F line in generations 1 to 40. Total days lost from broodiness in the F line did not exhibit any significant change in any generation interval. Most of the increase in egg weight in the F line occurred during generations 1 to 10 as indicated by a significant positive linear regression coefficient of egg weight on generations of selection during this period, with no significant change occurring in generations 11 to 20, 21 to 30, and 31 to 40. Over the entire period, the linear regression coef-
Table 3. Effect of selecting turkeys over 40 generations for increased 16-wk body weight on reproduction traits Line1 Variable 3
Egg production, no./hen Days to first egg4 Clutch length,3 d Maximum Average Total days broody3,5 Rate of lay,3,6 % Egg weight,7 g 1
RBC2
F
F − RBC2
Difference after 30 generations2
76.0 18.8
57.2 22.8
−18.8*** 4.0***
−23.9*** 4.3***
8.0 2.39 48.3 57.1 87
3.8 1.50 41.1 45.2 97
−4.2*** −0.89*** −7.2 −11.9*** 10***
−3.4*** −0.74*** 7.1 −13.8*** 9***
Change in F from 30th generation
1 to 30
31 to 40
5.1 −0.3
−0.797 0.143
0.510 −0.030
−0.8 −0.15 −14.3 1.9 1
−0.113 −0.025 0.227 −0.460 0.300
−0.080 −0.015 −1.43 0.190 0.100
RBC2 = randombred control line; F = subline of the RBC2 line selected for increased 16-wk body weight. From Nestor et al. (2000). 3 Based on a 180-d production period. 4 Days from beginning of stimulatory lighting (14 h/d) to production of first egg. 5 Total number of days lost during pauses in egg production of 5 or more consecutive days. 6 Rate of lay = number of eggs laid/(180 d − total days broody). 7 Based on the first 12 wk of egg production. ***P ≤ 0.001. 2
Change per generation
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
F
Difference after 30 generations2
1976
Nestor et al.
Table 4. Linear regression coefficients of growth-related traits on generations of selection in the F line after adjustment for changes in the environment1 Generations Variable
11 to 20
21 to 30
31 to 40
0.049*** 0.193***
0.054*** 0.115*** 0.193***
0.072** 0.258*** 0.347***
0.062** 0.178*** 0.284***
0.049*** 0.160***
0.050*** 0.115*** 0.158*** 0.260***
0.094*** 0.213*** 0.245*** 0.332***
0.054** 0.115*** 0.135*** −0.071
0.226***
1 to 40 0.071*** 0.196***2 0.245***2 0.065*** 0.158***2 0.180***2,3 0.222*** 0.058***2,5 0.046***5 0.131***5 0.126***5
−0.250
−0.100
0.439
0.669
0.024***5 0.033**5,6 0.012
1
The F line was selected for increased 16-wk body weight. To remove environmental variation, values for the randombred control were subtracted from those of the F line before the regression analysis. 2 Significant (P ≤ 0.05) positive quadratic regression coefficient. 3 Based on generations 9 through 40. 4 Body weight when hens first achieved 50% production. 5 Based on generations 14 through 40. 6 Significant (P ≤ 0.05) negative quadratic regression coefficient. **P ≤ 0.01; ***P ≤ 0.001.
ficient of egg weight on generations of selection was positive and significant, but there was a significant positive quadratic component in the change.
Selection Intensity, Selection Differentials, and Realized Heritabilities The number of birds available at 16 wk of age and percentage of offspring selected to reproduce the F line
are shown in Table 6. For both sexes combined, the percentage selected was 24.4 for generations 1 to 40. Selection pressure, as measured by the percentage selected, was greatest in generations 31 to 40, lowest in generations 11 to 20, and similar in generations 1 to 10 and 21 to 30. The number of females used to reproduce the F line was increased in generation 22 from 36 to 72. The intended and actual selection differentials did not consistently differ in the F line (Table 6). Realized
Table 5. Linear regression coefficients of reproduction traits on generations of selection in the F line after adjustment for changes in the environment1 Generations Variable Egg production,2 no./hen Days to first egg4 Clutch length,2 d Maximum Average Total days broody2,5 Rate of lay,2,6 % Egg weight,7 g
1 to 10
11 to 20
21 to 30
31 to 40
1 to 40
−2.961** 0.143**
−0.113 0.251
−1.061* 0.032
0.649 −0.071
−0.019***3 0.076***3
−3.64** −0.068** 2.405 −0.011*** 0.224*
−0.092 −0.050* −0.261 −0.010* −0.194
−0.157* −0.021 0.699 −0.003 0.181
−0.122 −0.030** 1.178 −0.001 −0.034
−0.084*** −0.019***3 −0.472 −0.003***3 0.243***3
1 The F line was selected for increased 16-wk body weight. To remove environmental variation, values for the randombred control were subtracted from those of the F line before the regression analysis. 2 Based on a 180-d production period. 3 Significant (P ≤ 0.05) positive quadratic regression coefficient. 4 Days from beginning of stimulatory lighting (14 h/d) to production of first egg. 5 Total number of days lost during pauses in egg production of 5 or more consecutive days. 6 Rate of lay = number of eggs laid/(180 d − total days broody). 7 Based on a 12-wk period at beginning of lay. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001.
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
Male body weights, kg 8 wk 16 wk 20 wk Female body weights, kg 8 wk 16 wk 20 wk Adult4 Shank length, cm Males Females Shank width, mm Males Females Walking ability score Males Females Mortality, 0 to 8 wk, %
1 to 10
1977
GENETICS OF GROWTH AND REPRODUCTION 1
Table 6. Selection intensity, selection differentials, and realized heritabilities in the F line
Generations Variable
1 to 10
11 to 20
21 to 30
31 to 40
364 275
324 384
418 592
1 to 40
2
376 305 24.8 23.2 24.0
32.4 29.8 31.1
0.6269 0.6178 −0.0091
0.6201 0.6128 −0.0073
0.5372 0.5398 0.0026
0.4409 0.4373 −0.0036
0.337 ± 0.030 0.280 ± 0.021 0.309 ± 0.022 0.392 0.321 0.356
0.320 ± 0.041 0.217 ± 0.033 0.268 ± 0.033 0.341 0.255 0.298
18.0 31.1 24.6
370 389
12.2 24.3 18.2
21.8 27.1 24.4
1.3054 1.3090 0.0036
1.2038 1.2006 −0.0032
0.9390 0.9350 −0.0040
0.6087 0.6141 0.0054
0.5280 0.5275 0.0005
0.5287 0.5297 0.0010
0.320 ± 0.026 0.218 ± 0.028 0.268 ± 0.026 0.306 0.241 0.274
0.202 ± 0.017 0.130 ± 0.025 0.166 ± 0.016 0.270 0.223 0.247
0.268 ± 0.004 0.215 ± 0.004 0.242 ± 0.004 0.270 0.223 0.247
1
The F line was developed from a randombred control population (RBC2) by selecting for increased 16-wk body weight. Based on birds alive at 16 wk of age. 3 I = intended selection differential (mean of selected birds minus mean of all birds); A = intended selection differential weighted by the number of offspring hatched. 4 Heritability ± standard error. 2
h2 of 16-wk BW in the F line were greater in males than in females and declined with selection, as estimated by the linear regression of response on the accumulated selection differential. For both sexes combined, the decline in realized h2 averaged about 4.8 percentage points for each 10 generations of selection, and the decline was largest (7.6 percentage points) in generations 31 to 40 from generations 21 to 30. The average decline in h2 estimates for each 10-generation interval was 4.5 and 5.0 percentage points in males and females, respectively. The realized h2 estimates based on estimates at certain points in time (last generation of each interval of selection) were, in general, greater than those based on regression for generations 1 to 10, 11 to 20, 21 to 30, and 31 to 40, but estimates by both methods were very similar for the entire selection period.
DISCUSSION Usually some changes in environment occur during the course of a long-term selection study. A major change (from range-rearing to complete confinement) in the rearing environment for the lines involved in the present study was made in generation 28. To evaluate the influence of this environmental change on line comparisons, Noble et al. (1996) used individuals from the first hatch in that generation to compare the performance of various lines, including those used in the current study, in the 2 environments. A sample of full-
sibs from each family was reared in each environment. There was no line × environment interaction for BW at 16 and 20 wk of age and leg measurements (shank length, width, and depth) at 16 wk of age. The interaction was significant for walking ability score of males at 16 wk of age. However, the cause of the significant interaction was due to variation in performance of a line not included in the present study. Therefore, the change in rearing environment probably did not affect the results of the current study. Significant trends over generations were observed for several traits of the RBC2 line for various periods of measurement. The direction of the changes was not always consistent over the course of the selection study. Studies by Nestor (1977b) and Noble et al. (1995) using the RBC2 line and other RBC populations suggested that the changes in the control lines were due to environmental changes. Long-term selection studies for increased BW have been conducted at Virginia Tech using chickens (Dunnington and Siegel, 1996), at the University of Georgia (Anthony et al., 1996; Marks, 1996), and at The Ohio State University (Nestor et al., 1996a,b) using Japanese quail and at The Ohio State University (Nestor et al., 1996c, 2000) using turkeys. Direct responses to selection have been measured in all of the selection studies. In general, realized h2 estimates declined with selection in chickens (Liu et al., 1994), in Japanese quail at the University of Georgia (Marks, 1996) and The
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
Number of birds RBC2 line F line Percentage selected Males Females Both sexes Selection differentials,3 kg Males I A I − A Females I A I − A Realized heritability Linear regression4 Males Females Both sexes Final generation Males Females Both sexes
1978
Nestor et al.
increased in the F line only during generations 1 to 10. The results of the present study and those of Nestor et al. (2000) suggested that the magnitude of genetic correlations, as well as the h2, may change with selection. Nestor (1971) postulated that the change in the genetic relationship between egg production and BW during the growing period in turkeys selected for increased egg production was due to the relationship between intensity of lay, egg production, and BW. In early generations of a turkey line selected for increased egg production, large increases in egg production were not associated with decreases in BW. During this time, the increases in egg production were primarily associated with decreases in broodiness. In later generations of the egg line, when increases in intensity of lay had a greater influence on the increases in egg production of the egg line than broodiness, BW declined. The decline in egg production observed in the F line in the present study and that of Nestor et al. (2000) was primarily due to reduction in intensity of lay, and there was no significant change in broodiness. The results of Nestor et al. (2000) and the present study support the hypothesis of Nestor (1971) that the negative association of BW and egg production is mediated through intensity of lay.
REFERENCES Abplanalp, H., F. X. Ogasawara, and V. S. Asmundson. 1963. Influence of selection for body weight at different ages on growth of turkeys. Br. Poult. Sci. 4:71–82. Anthony, N. B., K. E. Nestor, and H. L. Marks. 1996. Shortterm selection for four-week body weight in Japanese quail. Poult. Sci. 75:1192–1197. Arthur, J. A., and H. Abplanalp. 1975. Linear estimates of heritability and genetic correlation for egg production, bodyweight, conformation and egg weight of turkeys. Poult. Sci. 54:11–23. Clayton, G. A. 1971. Egg production in turkeys. Br. Poult. Sci. 12:463–474. Cook, R. E., W. L. Blow, C. C. Cockerham, and E. W. Glazener. 1962. Improvement of reproductive traits and body measurements of turkeys. Poult. Sci. 41:556–563. Dunnington, E. A., and P. B. Siegel. 1996. Long-term divergent selection for eight-week body weight in White Plymouth Rock chickens. Poult. Sci. 75:1168–1179. Falconer, D. S. 1964. Introduction to Quantitative Genetics. Oliver and Boyd, Edinburgh, UK. Havenstein, G. B., K. E. Nestor, V. D. Toelle, and W. L. Bacon. 1988. Estimates of genetic parameters in turkeys. 1. Body weight and skeletal characteristics. Poult. Sci. 67:1378–1387. Johnson, A. S., and R. S. Gowe. 1962. Modification of the growth pattern of the domestic turkey by selection at two ages. Pages 57–63 in Proceedings of the 12th World’s Poultry Congress, Sydney, Australia. Kosin, I. L., and W. A. Becker. 1959. Shifts in the phenotype of lines within a turkey strain, following several generations of selective breeding. Poult. Sci. 38:1220. (Abstr.) Liu, G., E. A. Dunnington, and P. B. Siegel. 1994. Responses to long-term divergent selection for eight-week body weight in chickens. Poult. Sci. 73:1642–1650. Marks, H. L. 1996. Long-term selection for body weight in Japanese quail under different environments. Poult. Sci. 75:1198–1203.
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
Ohio State University (Nestor et al., 1996a), and in turkeys (Nestor et al., 2000; present study). In the present study, the realized h2 of 16-wk BW based on the linear regression of response on accumulated actual selection differentials was 0.202, 0.130, and 0.166, respectively, for males, females, and sexes combined during generations 31 to 40, which indicated that a plateau in response is not likely in the near future of this line. Estimates of realized h2 of 16-wk BW in the F line over the entire 40 generations of selection were very similar when based on linear regression or response divided by total accumulated actual selection differentials at the 40th generation. When the selection period was divided into 10-generation intervals, h2 estimates based on a single point measurement of response divided by the accumulated actual selection differentials in the final generation were generally greater than those obtained by regression of response on accumulated actual selection differentials (Nestor et al., 2000; present study). During reproduction, natural selection was apparently not opposing artificial selection, because the intended selection differential (deviation of selected birds from the population mean) and actual selection differential (intended selection differential weighted by the number of offspring produced) did not differ (Nestor et al., 2000; present study). This result was unexpected, because genetic increases in BW in the F line were associated with a decrease in egg production (Nestor et al., 2000; present study) and with a decrease in hatch of fertile eggs (Nestor et al., 2000). The F line was reproduced early in the reproduction period when egg production, fertility, and hatch of fertile eggs were at a maximum. Likely, if the reproduction period was extended over a longer period of time, the intended and actual selection differentials would be different. Measurements of correlated responses were made routinely for some traits and only periodically for other traits in the chicken selection study at Virginia Tech and in the Japanese quail selection study at the University of Georgia. In the Japanese quail selection study at The Ohio State University (Nestor et al., 1996b) and in the turkey study at The Ohio State University (Nestor et al., 2000; present study), many correlated traits were measured most generations so that changes with selection in the correlated traits could be measured. The results of the study of Nestor et al. (2000) and the present study indicated that the genetic relationship, particularly in magnitude, between 16-wk BW and many correlated traits changed with selection. Based on the linear regression coefficients on generations of selection, the changes in number of eggs laid were negative and significant for only generations 1 to 10 and 21 to 30. For intensity of lay traits (maximum and average clutch lengths and rate of lay), the changes were larger in generations 1 to 10 than in other generation intervals. Of the intensity of lay measures, only average clutch length changed significantly (−0.030, P ≤ 0.01) during generations 31 to 40. Days from stimulatory lighting to production of the first egg and egg weight
GENETICS OF GROWTH AND REPRODUCTION
cursor in Coturnix coturnix japonica. 11. Correlated responses over thirty generations. Poult. Sci. 75:472–477. Nestor, K. E., W. L. Bacon, Y. M. Saif, and P. A. Renner. 1985. The influence of genetic increases in shank width on body weight, walking ability, and reproduction of turkeys. Poult. Sci. 64:2248–2255. Nestor, K. E., M. G. McCartney, and N. Bachev. 1969. Relative contribution of genetics and environment to turkey improvement. Poult. Sci. 48:1944–1949. Nestor, K. E., M. G. McCartney, and W. R. Harvey. 1967. Genetics of growth and reproduction in the turkey. 1. Genetic and non-genetic variations in body weight and body measurements. Poult. Sci. 46:1374–1384. Nestor, K. E., D. O. Noble, J. Zhu, and Y. Moritsu. 1996c. Direct and correlated responses to long-term selection for increased body weight and egg production in turkeys. Poult. Sci. 75:1180–1191. Nestor, K. E., and P. A. Renner. 1966. New management system for broody turkey hens. Ohio Rep. 51:83–84. Noble, D. O., D. A. Emmerson, and K. E. Nestor. 1995. The stability of three randombred control lines of turkeys. Poult. Sci. 74:1074–1078. Noble, D. O., K. E. Nestor, and C. R. Polley. 1996. Range and confinement rearing of four genetic lines of turkeys. 1. Effects on growth, mortality, and walking ability. Poult. Sci. 75:160–164. Ogasawara, F. X., H. Abplanalp, and V. S. Asmundson. 1963. Effect of selection for body weight on reproduction in turkey hens. Poult. Sci. 42:838–842. Saif, Y. M., and K. E. Nestor. 1983. Eradication of a Mycoplasma meleagridis in an experimental flock of turkeys. Ohio Poult. Pointers 22:2. Shoffner, R. N., and A. T. Leighton. 1962. Photoperiodicity and selection for egg production in the turkey (Meleagris gallapavo). Pages 37–41 in Proceedings of the 12th World’s Poultry Congress, Sydney, Australia. Toelle, V. D., G. B. Havenstein, K. E. Nestor, and W. L. Bacon. 1990. Estimates of genetic parameters in turkeys. 3. Sexual dimorphism and its implication in selection procedures. Poult. Sci. 69:1634–1643.
Downloaded from http://ps.oxfordjournals.org/ at Serials Section, Dixson Library on June 2, 2015
McCartney, M. G. 1961. Heritabilities and correlations for body weight and conformation in a randombred population of turkeys. Poult. Sci. 40:1694–1700. McCartney, M. G., K. E. Nestor, and W. R. Harvey. 1968. Genetics of growth and reproduction in the turkey. 2. Selection for increased body weight and egg production. Poult. Sci. 47:981–990. Mukherjee, T. K., and G. W. Friars. 1970. Heritability estimates and selection responses of some growth and reproductive traits in control and early growth selected strains of turkeys. Poult. Sci. 49:1215–1222. Naber, E. C., and S. P. Touchburn. 1970. Ohio Poultry Rations. Ohio Cooperative Extension Service Bulletin 343. The Ohio State University, Columbus. Nestor, K. E. 1971. Genetics of growth and reproduction in the turkey. 3. Further selection for increased egg production. Poult. Sci. 50:1672–1682. Nestor, K. E. 1972. Broodiness, intensity of lay, and total egg production of turkeys. Poult. Sci. 51:86–92. Nestor, K. E. 1977a. The use of a paired mating system for the maintenance of experimental populations of turkeys. Poult. Sci. 56:60–65. Nestor, K. E. 1977b. The stability of two randombred control populations of turkeys. Poult. Sci. 56:54–57. Nestor, K. E. 1977c. Genetics of growth and reproduction in the turkey. 5. Selection for increased body weight alone and in combination with increased egg production. Poult. Sci. 56:337–347. Nestor, K. E. 1984. Genetics of growth and reproduction in the turkey. 9. Long-term selection for increased 16-wk body weight. Poult. Sci. 63:2114–2122. Nestor, K. E., J. W. Anderson, and R. A. Patterson. 2000. Genetics of growth and reproduction in the turkey. 14. Changes in genetic parameters over thirty generations of selection for increased body weight. Poult. Sci. 79:445– 452. Nestor, K. E., W. L. Bacon, N. B. Anthony, and D. O. Noble. 1996a. Divergent selection for body weight and yolk precursor in Coturnix coturnix japonica. 10. Response to selection over thirty generations. Poult. Sci. 75:303–310. Nestor, K. E., W. L. Bacon, N. B. Anthony, and D. O. Noble. 1996b. Divergent selection for body weight and yolk pre-
1979