Replacement and Lifetime Production Traits : Effect of Non-genetic Factors and Sire Evaluation

The present investigation was undertaken to study the effects of non-genetic factors and association among replacement and lifetime production traits. The data on 542 Holstein Friesian cows maintained during 1975-98 at State Cattle Breeding Project, Sector III, Hisar, were utilized. The average sex-ratio, abnormal births, mortality, culling and replacement rates on total calf born and total female calf born basis were 51.62, 8.50, 17.52, 31.05, 22.78 and 51.41 per cent, respectively. The study revealed that a minimum of 4 to 5 progenies are required per cow over its lifetime to replace itself. It indicated that each cow should produce a minimum of 2 female calves during its life so as to replace herself before being lost. The least-squares means for productive herd life, longevity and lifetime production were 1439.32±87.64 and 2419.18±8.25 days and 11317.95±913.15 kg, respectively. The heritability estimates for all replacement traits were very low indicating that sire selection may bring no desirable change in these traits. Heritability estimates were 0.178±0.157, 0.288±0.184 and 0.096±0.195 for corresponding lifetime production traits. Breeding values and ranking of sires were generated for replacement and lifetime production traits to estimate the rank correlations between these traits. Moderate desirable rank correlations were obtained between replacement rate and lifetime production traits indicating that sires proven on the basis of milk production are also expected to have better replacement rate. (Asian-Aust. J. Anim. Sci. 2002. Vol 15, No. 1 : 11-15)


INTRODUCTION
The effectiveness of selection depends on herd size, selection differential and selection intensity, which determine the rate of genetic improvement.Therefore, efforts should be made to increase the intensity of selection, which becomes possible by increasing the herd size and the number of offspring reaching the milking herd in the next generation.Genetic improvement is a long and expensive process, which becomes futile if the genetically superior germplasm fails to replace itself.This in turn depends upon reproductive efficiency and survival rate.
An individual has better fitness if it produces more surviving progeny in the same environment.The continuous use of highly selected sires to produce replacements virtually ensures an increase in the average genetic merit for milk producing ability as well as birth rate.An economic analysis indicated that ranking sires on stayability of daughters from birth to 48 months may increase profit for a broad range of replacement costs and returns from milk production.Thus, information on replacement rate is required for planning, operating and evaluating the breeding programmes for genetic improvement of the herd.Keeping this in view the present investigation was carried out to study the effects of non-genetic factors and association among herd replacement and lifetime production traits.

MATERIALS AND METHODS
The data for the present investigation were collected on 542 Holstein Friesian cattle maintained at the State Cattle Breeding Project, Sector III, Hisar, over a period of 24 years from 1975-98.Replacement traits were recorded as all-ornone traits (1 versus 0) on individual basis, while on population basis these were defined as follows : Sex-ratio: Percentage of male calves born.Lifetime production traits included in the study were: Longevity: The number of days from birth till disposal of the cow either due to culling or death.Herd productive life: The number of days from first calving till disposal of the cow from the herd.Life-time milk production: The total milk produced by the cow during its productive herd life.
The data for the present study were classified into seven Replacement traits were analysed for studying the effect of various non-genetic factors according to Robertson and Learner (1949) procedure for all-or-none traits.
Analysis of lifetime production traits was carried out using the following mixed model least-squares technique (model-2) of Harvey (1980): Y ijklm =µ+G i +P j +S k +A l +e ijklm Where, Y ijklm is the observation on the traits with various effects; µ is the overall population mean; G i is the random effect of i th sire NID (0, σ s 2 ); P j is the fixed effect of j th period of first calving (j=1,2,…,7); S k is the fixed effect of k th season of first calving (k=1,2,…,4); A l is the fixed effect of l th age of first calving group (l=1,2,…,6); and, e ijklm is the random error NID (0, σ e 2 ).The heritability estimates were worked out by the paternal half-sib correlation method.Sire evaluation was done by estimating the breeding value of the traits studied as deviation from the population mean using the formula as below: where, X oi is the average of the progenies of the i th sire, X p is the herd average, h 2 is the heritability of the trait, and n i is the number of daughters of i th sire.The association between breeding value of sires for replacement and lifetime production traits was estimated as rank correlation.

Effect of non-genetic factors
The period, season and parity of calving wise analysis of variance and average percentages of various replacement traits are presented in tables 1 and 2, respectively.The overall average incidence of sex ratio, abnormal births, culling, mortality and replacement rate was 51.62, 8.50, 31.05, 17.52 and 22.78 per cent, respectively.The nonsignificant effects of period, season and parity on sex ratio were in conformity with the results of Sethi and Rao (1981) and Khanna et al. (1983).The proportion of abnormal births was minimum in the third period (6.0%) and maximum in the last period (12.24%) but, the differences were nonsignificant (table 2).Similar results have been reported by Sharma and Jain (1984).A significantly higher proportion of abnormal births was observed in cows calving in the rainy season when compared to those calving in other seasons.This was in conformity with the results of Lathwal et al. (1993), while Khanna et al. (1983) observed nonsignificant differences among seasons.The incidence of abnormal births was not significantly different over parities, which was in line with the results of Lathwal et al. (1993) but contrary to those of Khanna et al. (1983).
Contrary to the report of Lathwal et al. (1993) culling rate was found to differ with the period of calving (p<0.01).The non-significant effect of season and parity on culling rate was in consonance with the results of Tomar and Verma (1988) and Lathwal et al. (1993).The mortality rate varied significantly among periods of calving, being highest during 1990-92 and lowest during 1993-95.These results are in conformity with those of Tomar and Verma (1988) and contrary to those of Lathwal et al. (1993).The results further revealed that season as well as parity were not significant sources of variations in female calf mortality.
The overall replacement rate was 22.78 and 51.41 per cent on the basis of the total calf born and the female calf born, respectively, which means that a very high proportion of calves born could not contribute to the future herd because of male births, abnormal births, culling and mortality of female calves up to the age at first calving.The study revealed that minimum 4 to 5 pregnancies were required per cow over its lifetime to replace itself.This was supported by the findings of Lathwal et al. (1993) in Red Sindhi cattle.It indicated that each cow should produce a minimum of two female calves during its herd life so as to replace itself or make genetic contribution to the future generation before being lost.Significant period differences in replacement rate indicated that the highest heifer replacements occured during 1984-86, and the lowest during 1981-83.The findings of Tomar and Verma (1988) were in conformity to the present results.The season of calving and the parity did not significantly influence both the replacement rates.
Analysis of variance and least-squares means for lifetime production traits depicting the effect of sire, AFC group, period and season of calving are given in tables 1b and 3, respectively.Longevity, productive herd life and lifetime production did not differ significantly among AFC groups, periods and seasons of first calving.These results were contrary to those of Durocq et al. (1991) and Rogers et al. (1988) for longevity and Hegde and Bhatnager (1985) and Reddy and Basu (1985) for productive herd life.The overall least-square means for longevity, productive herd  life and milk production were 2419.18±89.25 and 1439.00±87.64days and 11317.95±913.15kg, respectively.The lifetime milk production was higher in cows with lower age at first calving (up to 850 days) and those calving for the first time in the rainy season but the differences were non-significant (p<0.05).

Heritability estimates
Heritability estimates were calculated on sires having 9 or more progenies.The heritabilities for all the replacement traits were very low (between 0.041±0.010and 0.069±0.001)indicating that selection is not likely to yield favourable response in improving any of the replacement traits.These results are in consonance with those of Tomar (1984) and Rawal andTomar (1995, 1996).Heritability estimates for longevity, productive herd life and lifetime milk production were low to medium (0.288±0.184, 0.178±0.157and 0.096±0.195,respectively), indicating that sire selection through progeny testing is expected to respond favourably.

Sire evaluation
Sires distribution in various classes of replacement traits is presented in table 4. The sex ratio varied from 33.37 to 69.56 per cent among progenies of different sires.About 85% of the sires used did not deviate from the expected sex ratio of 50:50.It was observed that only one sire had lower male births (33.37%).The incidence of female calf mortality varied from 5.26 to 40.00 per cent.Very low (5-10%) and very high (above 30%) mortalities were observed in only about 14% of the sires each, while the remaining 72% of the sires had medium mortality rates.The culling rate was almost evenly distributed with 6-8 sires falling in each of the below 20, 20-40 and the above 40% classes.The study revealed that about half of the sires had replacement rates above 50% on the basis of the total calves.This indicates that even though the heritability estimate is low, selecting sires with higher replacement rates may result in improvement of this trait.
The ranges of breeding values for sex ratio, abnormal births, culling, mortality and replacement rate were -0.184 to 0.321, -0.068 to 0.147, -0.103 to 0.165, -0.092 to 0.149 and -0.147 to 0.237, respectively.It was observed that 51.61, 62.90, 61.29, 54.83 and 50.00 per cent of the sires had their breeding values below the corresponding herd averages for The breeding value for lifetime production traits was estimated by the weighted least-squares method as a deviation from the herd average.The ranges of breeding values for different sires were from -382.92 to 878.00 days, -297.40 to 555.48 days and -3,946.73 to 4,954.53 kg for longevity, productive herd life and lifetime milk production, respectively.The rank correlations among sire rankings on replacement and lifetime production traits are presented in table 5.It was observed that the coefficients of rank correlations were positive and low to medium for sex ratio with all other replacement traits.It was negative and close to zero for abnormal birth with culling and mortality rates and positive and medium for abnormal birth with replacement rate (0.211).The rank correlation of culling rate with mortality and replacement rate was also low negative, while that between mortality and replacement rate was close to zero.
Moderately high and positive correlations between ranking of sires on sex ratio with all the lifetime production traits were obtained.These estimates for abnormal births with all the lifetime production traits were low to medium positive and were negative and close to zero for culling rate with various lifetime production traits.
Moderate desirable rank correlations were obtained only between replacement rate and lifetime production traits indicating that sires proven on the basis of milk production are also expected to have better replacement rate.Other replacement traits have no bearing on sire evaluation on lifetime production traits and vice versa.

Table 2 .
Period, season and parity of calving wise least-squares means of incidence (%) of various replacement traits Figures in parentheses arethe number of observations in corresponding class.Percentages bearing different superscripts differ significantly (p<0.01)from each other.

Table 3 .
Least-squares means and standard errors of various lifetime production traits in different periods, seasons and age at first calving groups

Table 4 .
Sire distribution in classes of replacement traits

Table 5 .
Rank correlations among sire rankings on replacement and lifetime production traits