GENETIC-IMPROVEMENT OF EXTERIOR AND LEG S OUNDNESS IN SWINE

Twelve exterior and leg traits used in two different conformation scor ing systems were analysed. Genetic correlations were estimated between conformation traits and to the performance traits daily gain and back fat thickness. The first non-linear scoring system described four exte rior traits and one overall leg trait on a scale of nine (1 to 9), in which nine is the desireable and one the undesireable score. The secon d scoring system described linearly one exterior trait and six single leg traits on a scale of five (-2 to +2), in which zero described the optimum and the negative and positive scores the degree and direction of deviation from the optimum. Data were collected at the central test station of a Northern German breeding company, in which boars were gr oup-penned and ad libitum fed during the individual performance test f rom 100. to 170. days of age. Data of 2122 boars from the breeds Large White (1109) and Landrace (1013) were present for the analysis. Herit ability estimates ranged from .37 to .55 for performance traits, .08 t o .37 for exterior traits and .11 to .39 for leg traits. High genetic correlations were found between daily gain and the exterior traits hei ght, lenght, muscularity and feeding condition (r(g) = .57 to .97) whe reas correlations between daily gain and leg traits were low (r(g) = - .32 to .21). Highest estimates were found between front view of front legs, liquid at hock joints and dairy gain in Large White boars and be tween side view of hind legs length of claws and daily gain in Landrac e pigs. Moderate genetic correlations were found between backfat thick ness and the exterior traits feeding condition, muscularity and type ( r(g) = .37 to .46). Linear scored leg traits hardly influenced perform ance traits. Liquid at hock joints increased by straight and O-shaped rear legs in Large White boars (r(g) = -.41 and -.12) as well as by si ckled and X-shaped rear legs in Landrace boars (r(g) = .16 and .37). X -shaped rear legs were often found together with sickled hind legs (r( g) = .23 and .58) and front view of front legs and rear view of hind l egs showed deviations from the optimum in the same direction (r(g) = . 39 and .53). The results also showed, that breed differences in severi ty of leg weakness were present ih linear scored traits and that breed s differed in heritabilities and genetic correlations. The use of a li near scoring system were recommended for improvement of leg soundness, because for each trait much more information about deviations from th e optimum were obtained, each leg trait represented separate genetic i nformation as indicated by low genetic correlations among each other, and each leg trait showed moderate heritabilities.