Genetic evaluation of type traits in hypertrophic Piemontese cows

The aim of this study was to estimate genetic parameters for type traits of hypertrophic Piemontese cows. Seven traditional type trait evaluations (70 to 100 grid scores), 2 body measurements (cm), and 13 linear description traits (1 to 9 grid scores) recorded on 21,757 Piemontese primiparous cows reared in 990 farms were used. Data were analyzed using a multiple-trait (22 traits) animal model with canonical transformation, accounting for a unique design matrix with the following effects: herd-year-classifier, days in milk, age at calving, and the genetic additive cow effect. Heritability estimates of traditional type evaluation traits were low for thorax, rump, feet and legs, and dairyness (<= 0.10), intermediate for fleshiness and overall score evaluations (0.13 to 0.15), and medium to high for body size (0.26). Genetic correlations of dairyness with all the other traditional type traits were low (from -0.14 to 0.16), those of feet and legs were moderate (0.19 to 0.44), and the remaining 5 traits were high (>= 0.55), with an exception regarding fleshiness and body size (0.28). Medium-high heritability estimates were obtained for withers height (0.31) and trunk length (0.21), with a very high genetic correlation between these traits (0.97). The genetic correlations of body measurements with body size were also very high (about 0.96), high with thorax, rump, and overall score (0.47 to 0.59), and moderate with the other traditional type traits (0.04 to 0.27). Heritability estimates of all linear traits were moderate (0.09 to 0.15), with the exceptions of top line (0.07) and condition score (0.05). Genetic correlations between linear traits were generally low to moderate (from -0.11 to 0.44) with the only exceptions of the 6 fleshiness traits and body condition, which showed very high correlations (0.60 to 0.96). Moreover, skeletal traits as top line, bone thinness, and head scores presented moderate genetic correlations (0.51 to 0.65). Genetic correlations between linear traits and traditional type traits were consistent with the trend observed between type traits. In conclusion, body measurements seem to describe body size better than traditional evaluation or linear descriptors. The genetic correlations among type evaluation and linear description traits suggest the need for a reduction in the number of traits scored, particularly of those relating to muscular development.