Genetic parameters for milk production of ewes in four South African woolled sheep flocks

M.A. Snyman1#, S.W.P. Cloete2,3 & W.J.Olivier1

1Grootfontein Agricultural Development Institute, Private Bag X529, Middelburg (EC), 5900; 2Directorate Animal Sciences, Private Bag X1, Elsenburg, 7607 ; 3Department of Animal Sciences, University of Stellenbosch, Private Bag X1, Matieland, 7602, South Africa;   

# E-mail: Gretha Snyman


Background: Selection for early weight gain in sheep is often complicated by variable and sometimes unfavourable direct-maternal genetic correlations. Early weight gains rely heavily on maternal milk supply available to lambs. Genetic information on the milk production of ewes may help to clarify the discrepancies pertaining to direct and maternal (co)variance components of weaning weights in the literature.


Aim: To estimate (co)variance components for milk yield of ewes and the weaning weights of their lambs.


Methodology: Animals from the well-documented Carnarvon Afrino, Elsenburg Merino, Cradock fine wool Merino and Grootfontein Merino resource flocks were used to obtain data on the estimated milk production of ewes and the weaning weights of their lambs (Ethical committee approval: DECRA/R12/08-Elsenburg; GVE/AP2/7-Grootfontein). Milk production was determined at approximately 21 and 84 days after lambing using the oxytocin method and used to estimate total lactation yield. The number of individual lactation records ranged from 966 at Grootfontein to 1613 at Carnarvon. Genetic parameters were derived from two-trait models in ASREML. 


Results: Milk yield were affected by ewe age in all flocks but the Cradock flock, with mature ewes generally outperforming two-tooth maidens. Ewes suckling multiples generally had a higher milk production than those ewes suckling singles. The heritability (±SE) of milk yield amounted to 0.21 ± 0.03 at Carnarvon, 0.29 ± 0.07 at Elsenburg, 0.10 ± 0.04 at Cradock and 0.02 ± 0.03 at Grootfontein. Genetic correlations of maternal milk production with weaning weight amounted to 0.29 ± 0.25 at Carnarvon, 0.64 ± 0.19 at Elsenburg, 0.28 ± 0.16 at Cradock and 0.23 ± 1.02 at Grootfontein. The correlations of the additive effect for milk yield and the maternal genetic effect for weaning weight were 0.77 ± 0.11 at Carnarvon, 0.99 ± 0.17 at Elsenburg, 0.37 ± 0.12 at Cradock and 0.55 ± 1.38 at Grootfontein. Milk production was highly correlated with total weight of lamb weaned at Carnarvon (0.68 ± 0.10), Cradock (0.92 ± 0.11) and Grootfontein (0.66 ± 1.00). Body weight before mating was related to maternal milk production only in the Carnarvon Afrino flock, yielding a genetic correlation of 0.33 ± 0.10.


Discussion: High genetic correlations were obtained between total milk production and reproduction, here represented as total weight of lamb weaned in three flocks. Moderate to high genetic correlations were also derived between maternal milk production and early body weights in all four flocks, while the additive correlations of milk yield with the maternal effect of weaning weight ranged from high to almost unity.


Conclusions and recommendations: There was no adverse genetic correlations among ewe milk yield, early lamb weights and reproduction. Selection on breeding values for any one of these traits, and particularly on maternal breeding values for weaning weight in the case of ewe milk production, are likely to also benefit the other traits. Further research should focus on adding information on ewe size also in the Elsenburg, Cradock and Grootfontein flocks, in order to obtain an indication of the efficiency of production. The effect of selection based on maternal breeding values on the production potential in sheep should also be quantified.



Proceedings 49th SASAS congress, Stellenbosch