Abstract: This study was aimed to access information on genetic parameters i.e., heritability and genetic correlations of some economic traits of dairy cattle available in Bangladesh. For that, data materials were collected from Central Cattle Breeding Station and Dairy Farm, Savar, Dhaka recorded during 1992 to 2005 on Indigenous called Local, Friesian x Local (50 F-50% L) and Jersey x Local (50 J-50% L) cattle covered birth weight, first lactation milk yield, first lactation length and first calving interval. REML estimated heritabilities (h2) of BW were 0.365, 0.495 and 0.489 in Local, Friesian x Local and Jersey x Local cattle, respectively. The same for FLL, FLMY and FCI were 0.486, 0.428 and 0.128 in Local; 0.495, 0.495 and 0.497 in Friesian x Local cattle and 0.496, 0.499 and 0.496 in Jersey x Local cattle. The overall estimates were 0.500, 0.497, 0.498 and 0.499, respectively for BW, FLL, FLMY and FCI. The mediums to high estimates for all the traits indicate that the traits studied were largely influenced by additive gene action. Genetic correlations (rG) of BW with FLL, FLMY and FCI were very low except medium rG between BW and FCI in Local cattle. The direction of genetic association was positive except negative rG were between BW and FLMY and BW and FCI in Local cattle. First lactation length was positively genetically correlated with FLMY and FCI in a medium to high magnitudes in all the studied genetic groups. Genetic correlations between FLMY and FCI were medium in Jersey x Local cattle but low in Local and Friesian x Local cattle. The overall estimate was also low. The magnitude of REML estimated (co) variance and genetic parameters obtained in the present study were within the published range of estimates. Therefore, these estimates may be used in selection and breeding programmes for dairy cattle development in Bangladesh.
Introduction
Bangladesh possesses a large cattle population (24.5 million heads cattle, FAO, 2004) comprised of Indigenous, exotic and their crosses with Indigenous (ILRI, 2004). There is wide variation in terms of production as well as reproduction among the existing cattle. To enhance their genetic potential, several indiscriminate attempts were undertaken in past few decades but no remarkable responses are observed (Bhuiyan, 1997, 2001). In most of these programmes, scientific procedures for genetic evaluation of breeding animals-an estimate of the merit of an individual’s full complement of genes were not followed. But the pre-requisite for estimation of genetic merit of animals is the good estimate of genetic parameters i.e., heritability and genetic correlation. These genetic parameters are essential tool in animal breeding research and in the design and application of practical breeding programme (Koots et al., 1994).
In Bangladesh although many studies have been carried out on the comparison of phenotypic performances in cattle (Hossain et al., 2002; Uddin, 2001; Hirooka and Bhuiyan, 1995; Udo et al., 1990; Ahmed and Islam, 1987; Rahman et al., 1987), there are very few studies on genetic parameter estimates for economic traits (Bhuiyan, 1999; Hossain et al., 2002). For that matter, accumulated records maintained at the Central Cattle Breeding Station and Dairy Farm, Savar, Dhaka were utilized in the present study to estimate heritability and genetic correlation of four economic traits in dairy cattle.
Materials and Methods
The experimental data were collected from the record sheets maintained at the Central Cattle Breeding Station and Dairy Farm, Savar, Dhaka. Data recorded during the period of 1972 to 1982 were included in this study. Data were taken on Local, Friesian x Local (50 F-50% L) and Jersey x Local (50 J-50% L) genetic groups of cattle born in summer (April to June), rainy (July to September), winter (October to December) and spring (January to March) seasons. Mean and Standard Errors (SE) for the studied traits e.g., Birth Weight (BW), First Lactation Milk Yield (FLMY), First Lactation Length (FLL) and First Calving Interval (FCI) were estimated using SPSS computer program. The differences in means were tested using Least Significant Difference (LSD) method (Ahmed et al., 2003).
Variance and covariance components of the traits were estimated using Restricted Maximum Likelihood (REML) approach by VCE4 computer program (Groeneveld, 1998). For REML analyses an animal model was used keeping year of birth and season of birth as fixed effect. However, for all traits analyses were carried out both within genetic group as well as over the genetic groups i.e. pooled data. The general single trait animal model used to describe the analysed traits was
Y = Xb + Za + Wc + e |
Where, Y is the vector of observation; X, Z and W are the known incidence matrices that associate with levels of b, a and c with Y; b is the unknown vector of year of birth; a is the unknown vector of animal’s breeding value; c is the unknown vector of season of birth and e is the vector of residual effects.
Covariance components and genetic correlations among traits were estimated using multi-trait animal model (Groeneveld, 1998) having animals additive genetic merit as only genetic term.
Results and Discussion
Data were collected from a total of 624 animals of which 489 had complete records for phenotypic and genetic studies of the concerned traits. Phenotypic mean, standard errors and mean comparison for the studied traits are presented in Table 1. Table 2 shows the effect of various factors on studied traits. Only data on animals having complete records (pedigree and performance) were used to estimate genetic parameters. The estimated variance components and heritabilities are shown in Table 3. Similarly, estimated covariance components and genetic correlation (s) among traits are presented in Table 4.
Phenotypic Performance
Local cattle had smallest birth weight (BWT) among the studied genetic groups
(Table 1). Udo et al. (1990) reported similar result
(15.6±0.40 kg) for Pabna cattle - an improved variety of Local cattle.
Birth weight of Friesian x Local cattle found in the present study was within
the range (17.28±0.436 to 23.05±0.32 kg) reported by Hirooka and
Bhuiyan (1995) and Nahar et al. (1989), Bhuiyan (1999) and Uddin (2001).
| Table 1: | Mean±SE for economic traits studied in three different genetic groups of dairy cattle |
| Figures in the parentheses indicate number of observation, Means with different superscripts in the same row differ significantly (p<0.05) | |
| Table 2: | Factors influencing the studied traits |
| * p<0.05 and ** p<0.01 | |
| Table 3: | Estimates of additive genetic variance and heritability for economic traits |
| Table 4: | Genetic correlations among economic traits |
Whereas Rege et al . (1994) found higher BWT for Jersey x Local crossbred than the present study. Mean First Lactation Milk Yield (FLMY) for Local, Friesian x Local and Jersey x Local cattle observed in the present study (Table 1) were almost similar with the results reported by Ahmed and Islam (1987) in Local, Rahman et al. (1987) in Friesian x Local and Hossain and Routledge (1982) in Jersey x Local cattle. The Local-Friesian crossbred produced more milk than Local-Jersey crossbred. The average FLL for Local, Friesian x Local and Jersey x Local and for pooled data (Table 1) were more or less similar. Ahmed and Islam (1987) in Local and Friesian x Local cows and Tekade et al. (1994) in Jersey crossbred observed nearly similar values for FLL. The average lengths between first and second calving for Local, Friesian x Local and Jersey x Local cattle and for pooled data were more or less similar (Table 1). Bhuiyan (1999) and Nahar et al. (1989) in Local and Friesian x Local and Rege et al. (1994) in Jersey crossbred also reported very similar range of results obtained in the present study.
Birth weight of cattle was found to be significantly affected (p<0.01) by genetic group (Table 2). Hirooka and Bhuiyan (1995) and Bhuiyan et al. (1992) also reported similar significant effect (p>0.05) of genetic group on BWT of calf. Likewise genetic group was also important with respect to FLMY and FLL (Hossain et al., 2002; Uddin, 2001; Bhuiyan, 1999; Sultana and Bhuiyan, 1997; Nahar et al., 1989). But FCI not found to be dependent on genetic group. Birth weight as well as subsequent FLMY, FLL and FCI of cows were influenced highly significantly (p<0.01) by year of birth. Similar results were reported by Hirooka and Bhuiyan (1995) and Bhuiyan et al. (1992) on BWT and Hossain et al. (2002), Uddin (2001), Bhuiyan (1999) and Sultana and Bhuiyan (1997) on FLMY, FLL and FCI. Except FLL season of birth effect was not important (p>0.05) for BWT, FLMY and FCI. Non significant effect of seasonal variation on studied traits may be attributed to similar feeding and management practices followed through the years.
Heritabilities
The estimated h2 value of BWT obtained analyzing the pooled data
was 0.50, which was higher than the estimates obtained within genetic groups
(0.365 in Local, 0.495 in Friesian x Local and 0.489 in Jersey x Local). Medium
to high h2 of BWT found in the present experiment was within the
published range (0.29±0.14 in Brown Swiss to 0.64±0.56 in Friesian
x Local) reported by Mandal and Sachdeva (1999), Plasse et al.
(2002), Akbulut et al . (2002) and Bhuiyan (1999). Estimated heritability
values of FLMY separately for three genetic groups and pooled data were high.
This high h2 value indicated that FLMY is largely controlled by additive
gene action. Medium to high h2 for FLMY was also reported by Hossain
et al. (2002), Gaur et al. (1999) and Ageeb and Hillers (1991).
The h2 of FLL in studied genetic groups and for pooled data were
almost similar. The high h2 for FLL found in the present study were
within the literature range (0.32±0.51 in Friesian x Local cows to 0.80±0.25
in Friesiwal-Friesian x Sahiwal crossbred) reported by Bhuiyan (1999), Nanavati
et al. (1998), Tekade et al. (1994) and Gaur et al. (1999).
The h2 value of FCI for Local was low whereas the same for Friesian
x Local, Jersey x Local and pooled data were high (Table 3).
But nearly zero to high h2 (0.0034±0.07 in Sahiwal to 0.65±0.21
in Frieswal) of FCI was reported by different authors (Sethi et al.,
1997; Gogoi et al., 1992; Bhoite et al., 1999; Gaur et al.,
1999). Low h2 for FCI in Local cattle is important in decision making
to the breeder that for improvement of calving efficiency emphasis need to be
given on management practices. One possible reason for low heritability of FCI
in Local cattle could be that the pedigrees of the Local cattle were not complete
as with other genotypes.
Genetic Correlations
Estimates of genetic correlations (rG) among the studied quantitative
traits of dairy cattle are depicted in Table 4. Estimates
of rG for BW and FLL, BW and FLMY and BW and FCI were very low in
all of the studied genetic groups of dairy cattle except medium rG between
BW and FCI in Local cattle. Negative rG of BW with FLMY and FCI were
only in Local cattle indicating that genes influencing BW subsequently retarded
FLMY and FCI. Low estimates of the present study indicate that BW of individuals
is not so important in selection programme to improve FLL, FLMY or FCI in dairy
cattle. Genetic correlations between FLL and FLMY in studied genetic groups
of dairy cattle obtained in the present study were within the published range
(Katoch and Yadav, 1990; Roy and Katpatal, 1988). The rG between
FLL and FLMY in Jersey cattle reported by them were 0.126±0.06 and 0.89±0.07,
respectively. Bhuiyan (1999) reported that the rG between FLL and
FCI in Friesian and Friesian x Local graded cows were 0.19 and 0.14, respectively.
Results of the present study were higher for all the three genetic groups. This
high genetic correlation between FLMY and FCI is also interesting to the breeders
as lifetime production is associated with calving interval. Genetic correlations
between FLMY and FCI obtained in the present study were medium to low for the
three genetic groups of dairy cattle. Such medium to low estimates of rG
between these two traits was within the published ranged for different
breeds (Lara et al., 1989; Herbert and Bhatnagar, 1989; Roy and Katpatal,
1988). They reported that rG between lactation milk yield and calving
interval were 0.33±0.30 in Holstein cows, 0.06±0.29 in Karan Swiss
cows and 0.10±0.33 in Jersey cows, respectively.
Above all, the differences observed among the estimated phenotypic and genetic parameters with that of others reported elsewhere in the world could be attributed to differences in sample size, genetic groups used, environmental conditions (feeding and management practices), models and procedures employed to estimate parameters and so on. However, since these estimates were found using data on animals maintained under the standard conditions of feeding, management and environmental conditions of Bangladesh, these values may well be used for genetic evaluation of animals at national dairy cattle herd. Moreover, the cattle breeding programme in the country need to be better organized to support scientific animal breeding practices such as individual animal recording, genetic parameter estimation using REML procedure and animal selection based of BLUP based breeding values in order to realize expected genetic improvement in the national herds.