Genetic and Phenotypic Correlation of Some Body Measurements with Milk Yield in Nili Ravi Buffaloes of Pakistan

Each breed has some special physical attributes which are important for appropriate performance and functional survival. Some of the body measurements are considered very important from milk production point of view and they have been reported to be correlated with milk yield. Rehman (1996) and Khalid (2011) have reported positive phenotypic correlation of body length with milk yield. Alphonsus et al. (2010) has reported a very high genetic correlation of height at withers with milk yield in crossbred cows. Jaayid et al. (2011) has reported significant positive phenotypic correlation of heart girth with milk yield.

Genetic correlations between any two traits indicate that some of the same genes are controlling these traits and in case of positive genetic correlation between them selection for improvement in any of these traits will result in the improvement in the other trait as a correlated response. In case of absence of any milk recording system under field conditions, farmers generally select animals on the basis of phenotype. Information on genetic correlation of body measurements with milk yield can be used for indirect selection for improved milk yield in this breed. Keeping in view the importance of body measurements and their correlation with milk yield as reported in the literature, the present study was planned with the objectives to record different body measurements in Nili Ravi buffaloes and estimation of phenotypic and genetic correlations with milk yield.

Nili Ravi buffalo herds maintained at 5 Livestock Experiment Stations in Punjab and few private breeders were utilized in the present study. The data included 227, 297, 172, 165, 269 and 50 records at Pattoki, Chack Katora, Haroonabad, Khushab, Rakh Ghulaman and private breeder’s herds, respectively.

General management and feeding practices at these stations were almost similar and have been more or less the same. Adult animals were being maintained in open enclosures with sufficient covered area for shade and shelter during extreme weather conditions. Routine practice was to allow animals to graze on available fodders for 4-6 hours daily early in the morning after the end of milking depending on season. The lactating buffaloes were fed concentrate at the rate of one kg for every three kg of milk produced. Buffaloes were milked twice daily with an approximate interval of 12 hours at all the farms. All the body measurements were recorded for thrice with an approximate interval of three months with the the first measurement taken after about one month of calving.

Data recording on body measurements was started during July, 2010 and continued till June, 2012. Information including tag number of the buffalo, sire and dam number and pedigree records of sire and dam and other such recorded data were collected in addition to recording body conformation traits. Body measurements including heart girth, height at withers and diagonal body length were studied.

Environmental factors such as age of the buffalo at scoring, stage of lactation, parity, herd and season of scoring were included in the model for initial analysis. Data were analyzed by the mixed model procedure of the Statistical Analysis Systems (SAS, 2011).

The following general mathematical model was used (Model 1):

Yijklm = µ + Si + Hj + Pk + Tl +b1 (aijklm) +b2 (aijklm)2 + eijklm

Where

Yijklm is the record of mth buffalo at lth stage of lactation during kth parity of jth herd in ith season

µ is the overall population mean

Si is the effect due to ith season

Hj is the effect due to jth herd

Pk is the effect due to kth parity

Tl is the effect due to lth stage of lactation

aijklm is the age of buffalo at classification

b1 and b2 is the linear and quadratic regression coefficient of age at classification

eijklm is the random error associated with the observation on mth buffalo at lth stage of lactation during kth parity of jth herd in ith season.

Phenotypic and genetic correlations were estimated using Best Linear Unbiased Prediction (BLUP) evaluation techniques. Fixed effects observed to be significant in the initial analysis using the above model were included in the model for estimation of genetic and phenotypic correlations of body measurements with milk yield. Individual Animal Model was fitted under Restricted Maximum Likelihood (REML) Procedure outlined by Patterson and Thompson (1971). Bivariate analysis fitting animal model in ASREML computer program (Gilmour, 2009) was used for estimation of correlations which were calculated by applying the following formulas:

Phenotypic correlation (rp)= Cov pi. pj / σ² pi. σ² pj

Genetic correlation (rG)= Cov Ai. Aj / σ² Ai. σ² Aj

Environmental correlation (rE)= Cov Ei. Ej / σ² Ei. σ² Ej

Where,

σ² Ai = additive genetic variance for the ith trait

σ² Aj = additive genetic variance for the jth trait

σ²pi = phenotypic variance for ith trait

σ² p = phenotypic variance for jth trait

σ² Ei = residual variance for the ith trait

σ² Ej = residual variance for the jth trait

Cov pi. pj = phenotypic covariance for the traits i and j

Cov Ai. Aj = additive genetic covariance for the traits i and j

Cov Ei. Ej = residual covariance for the traits i and j

Separate data and pedigree files were prepared in excel sheets and analysis was performed in ASREML computer program (Gilmour, 2009). Pedigree records of buffaloes were traced back up to five available generations and these buffaloes were the progeny of 88 sires and 303 dams. Number of base animals were 119 with no pedigree records. Most of the phenotypic correlations of 305 days milk yield with body measurements were in low range. Body measurements have shown very low positive phenotypic correlation values with 305 days milk yield. Negative phenotypic correlations with 305 days milk yield were not observed. Positive genetic correlation values were observed for diagonal body length, height at withers and for heart girth. The results are presented in table 1 and 2.

Table 1: Least squares means for body measurements in Nili Ravi buffaloes

Table 2: Phenotypic and genetic correlation of some body measurements with 305 days milk yield in Nili Ravi Buffaloes

Phenotypic correlation of diagonal body length with milk yield was very low in the current study (<0.05). Sieber et al. (1988) reported a significant phenotypic correlation of body length with milk yield as 0.21. Bayram et al. (2006) has also reported a significant positive phenotypic correlation of body length with milk yield as 0.28. Alphonsus et al. (2010) has reported this correlation as 0.25 in crossbred cows. Jaayid et al. (2011) has reported significant positive phenotypic correlation of body length with milk yield as 0.32 in Iraqi buffaloes. Musa et al. (2011) has reported this correlation highly significant as 0.54 in Kenana cattle. The very low phenotypic correlation in the current study might be due to relatively small data set or due to species differences.

Genetic correlation of diagonal body length with 305 days milk yield has been observed considerable as 0.16±0.0003. Khan (2009) has reported almost similar genetic correlation of diagonal body length with 305 days milk yield as 0.13±0.00 in Sahiwal cows. Alphonsus et al. (2010) has reported this correlation as 0.048 in crossbred cows. Diagonal body length in the current study has shown a low but positive genetic correlation with milk yield and this trait might be considered in the selection program for Nili Ravi buffaloes.

Phenotypic correlation of height at withers with 305 days milk yield was found as 0.04±0.02. Jaayid et al. (2011) has reported a similar phenotypic correlation of height at withers with milk yield as 0.03 in Iraqi buffaloes. Slightly higher phenotypic correlation of height at withers with milk yield was reported by Lin et al. (1987) as 0.12 in Holstein cows and Khan (2009) as 0.09±0.06 in Sahiwal cows. Sieber et al. (1988) reported a significant phenotypic correlation of height at withers with milk yield as 0.22. Bayram et al. (2006) has also reported a significant positive phenotypic correlation of 0.30 in Holstein Friesian cows. Alphonsus et al. (2010) has reported this correlation as 0.23 in crossbred cows. Khalid (2011) has reported this correlation as 0.20 in Nili Ravi buffaloes. Musa et al. (2011) has reported this correlation highly significant as 0.28 in Kenana cattle.

Genetic correlation of height at withers with 305 days milk yield was observed as 0.08±0.00. Lin et al. (1987) and Van Niekerk et al. (2000) have reported higher estimates as 0.35±0.13 and 0.33, respectively. Khan (2009) has reported this correlation as 0.17± 0.00. Alphonsus et al. (2010) has reported a very high genetic correlation as 0.59 in crossbred cows. These reports are not in agreement with the findings of current study. Lower genetic correlation of height at withers with milk yield suggested that this trait is not much important, however further research is needed to investigate this relationship in Nili Ravi buffaloes.

The findings of current study indicated a phenotypic correlation of heart girth with 305 days milk yield as 0.04±0.04. Sieber et al. (1988) reported a positive but non significant phenotypic correlation of heart girth with milk yield as 0.18. Rehman (1996) has reported phenotypic correlation of heart girth with milk yield as 0.14±0.02. Bayram et al. (2006) have reported a highly significant positive phenotypic correlation of 0.30 in Holstein Friesian cows. Alphonsus et al. (2010) has reported this correlation as 0.17 in crossbred cows. Khalid (2011) and Yakubu (2011) have reported a very high phenotypic correlation of heart girth with milk yield as 0.63 and 0.53 respectively. Jaayid et al. (2011) has reported significant positive phenotypic correlation of heart girth with milk yield as 0.36 in Iraqi buffaloes. Musa et al. (2011) has reported this correlation highly significant as 0.36 in Kenana cattle. The findings of the current study do not agree with most of the above reports. The reason might be relatively small data set or some genetic or environmental differences.

Genetic correlation of heart girth with 305 days milk yield was found as 0.14±0.00. Khan (2009) has reported corresponding value as 0.22±0.00 in Sahiwal cows. Lin et al. (1987) and De Haas et al. (2007) have also reported higher genetic correlation of heart girth with milk production as 0.30 and 0.39, respectively. Alphonsus et al. (2010) has reported a very high genetic correlation as 0.82 in crossbred cows. These reported values are high and do not agree with the findings of current study. Genetic correlation of heart girth with milk yield although not very high but seems to be important and can be considered for indirect selection for milk yield through heart girth measurement.

Diagonal body length in the current study has shown a low but positive genetic correlation with milk yield and this trait might be considered in the selection program for Nili Ravi buffaloes. Genetic correlation of heart girth with milk yield although not very high but seems to be important and can be considered for indirect selection for milk yield through heart girth measurement.

The cooperation of the officer incharges of public sector livestock farms and the members of buffalo breeders association of Punjab is highly acknowledged. The project was funded by the Higher Education Commission of Pakistan.

We want to make it clear that there is no conflict of interest with any of the financial organization regarding the material discussed in the current manuscript.