Genetic parameters for subjective and objective wool and body traits in the Tygerhoek Merino flock

[摘要] English: Genetic evaluation systems require accurate estimates of genetic parameters. The genetic, phenotypic and environmental parameters for objectively measured wool, subjectively assessed wool and conformation traits as well as 16-month live weight were estimated for South African Merino sheep. Records of the Tygerhoek Merino resource flock were used to estimate these parameters. The database consisted of records of 4 495 animals, the progeny of 449 sires and 1 831 dams born in the period 1989 to 2004. The pedigree records used have been collected between 1969 and 2004. In this data, four lines were represented, namely a line selected on clean fleece weight, a line selected against rearing failure, a fine wool line, as well as an unselected control line. Single trait linear mixed models were used. Fixed effects that were assessed included birth status (single/multiple), sex (male/female), age of the dam in years (2-6+years), year of birth (1989-2004) selection line (1-4) and the sex*birth year interaction. Preliminary analysis showed that all the fixed effects had significant (P<0.05) effects on all the traits studied. The random effects included a combination of direct additive, maternal additive and maternal permanent environmental effects as well as the covariance between direct and maternal additive effects. The direct heritability estimates (h²a) for objectively measured traits were 0.38 for 16- month live weight (LW), 0.36 for greasy fleece weight (GFW), 0.40 for clean fleece weight (CFW), 0.65 for clean yield (CY), 0.37 for staple length (SL), 0.20 for staple strength (SS), 0.68 for fibre diameter (FD), 0.61 for coefficient of variation of fibre diameter (CVFD) and 0.61 for standard deviation of fibre diameter (SDFD). The maternal heritability estimates were 0.05 for LW, 0.09 for GFW, 0.10 for CFW and 0.05 for FD. The proportion of the total phenotypic variance due to the maternal permanent environment (c²pe) amounted to 5% for fleece weights. The correlation between direct and maternal genetic effects for LW, GFW and CFW were -0.28, -0.65 and -0.70 respectively. These results suggested that worthwhile responses to selection for objectively measured traits would result from directed selection. Estimates of h²a for subjectively assessed wool traits were 0.49 for wool quality (QUAL), 0.28 for regularity of crimp (ROC), 0.33 for wool colour (COL), 0.23 for wool oil (OIL), 0.21 staple formation (STAPL), 0.22 belly and points (BANDP), 0.50 for woolly face score (WFS), 0.15 for face cover score (FCS), and 0.45 for pigmentation (PIGM). Maternal genetic effects were significant for QUAL, ROC and COL amounting to 0.06, 0.03 and 0.07 respectively. The correlation between direct and maternal genetic effects for QUAL, ROC and COL were -0.62, -0.70 and -0.45 respectively. The h²a estimates for subjectively assessed conformation traits were 0.13 for topline (TOPL), 0.39 for total fold score (TOT), 0.15 for front quarters (FQ), 0.16 for pastern score (PS), 0.32 for hocks (HOCKS) and 0.37 for the general conformation score of the head (GEN). Among the subjectively assessed conformation traits the maternal genetic effects were significant only for PS at 0.05 while GEN was affected by c²pe at 5%. The correlation between direct and maternal genetic effects was -0.71 for PS. The improvement of subjectively assessed wool and conformation traits by selection seems possible from the results of present study. Genetic correlations (rg) among objectively assessed traits general agreed with literature values. Among the subjectively assessed traits favourable rg estimates were estimated between ROC and COL (0.31), for QUAL with ROC (0.49) and COL (0.26) and between STAPL and BANDP (0.58). The relationships between ROC and STAPL (-0.49) and for QUAL with STAPL (-0.45) and BANDP (-0.20) were unfavourable. Estimates of rg among subjective conformation traits were generally low, variable in sign and not significant. The only exceptions were rg's between HOCKS and FQ (0.71) and of GEN and TOPL with TOT (-0.31 and -0.47 respectively). Phenotypic and environmental correlations among objective, subjective wool and conformation traits were low to moderate. Maternal relationships of respectively 0.86 and 0.19 were found between GFW and CFW and between CFW and FD. The rg of LW with subjective wool traits were not significant, with the exception of moderate and favourable rg for LW with ROC (0.20) and for LW with WFS (0.20). These results suggested that heavier sheep are likely to have more even fleeces and higher scores for WFS. The rg were favourable with COL (-0.19), STAPL (0.35) and BANDP (0.42). CFW and CY were also favourably related to COL, STAPL, BANDP and QUAL where sheep with higher fleece weights and yield had wool with higher quality and better scores for COL, STAPL as well as BANDP. SL was favourably related to COL, BANDP and WFS and negatively related to OIL. These results indicated that sheep with longer staples had better scores for COL, BANDP and WFS as well as lower scores for OIL. Among the subjective wool traits only ROC was significantly related to SS at 0.33. Moderate and unfavourable genetic correlations were estimated for FD with QUAL, ROC and FCS, where sheep with a good quality wool that have more even fleeces and softer wool covering the face had reduced FD. The corresponding relationship with STAPL and BAND were unfavourable. Genetic correlations of CVFD and SDFD with QUAL and ROC were favourable and while these traits were unfavourably related to a number with STAPL. Phenotypic and environmental relationships between objectively measured traits and subjectively assessed wool traits were low to high and variable in sign. On the genetic level LW was favourably related to all the subjectively assessed conformation traits studied. Heavier sheep were generally plainer, had better scores for GEN, HOCKS, and FQ as well as for TOPL. Estimates of rg were favourable for CY with GEN (0.25), HOCKS (0.19), TOT (-0.26) and FQ (0.18), for FD with PS (-0.16) and TOPL (-0.18), for CVFD with HOCKS (-0.17), FQ (-0.33) and TOPL (-0.25), between CFW and GEN (0.23), and between SS and FQ (0.39). The relationships of SDFD with FQ and TOPL were negative. Unfavourable correlations occurred between SS and TOT (0.25), between FD and HOCKS and FD and TOT at 0.13, for fleece weights with TOT (GFW 0.48 and CFW 0.28) and TOPL (GFW 0.34 and CFW 0.29) and between CY and TOT at -0.26. Phenotypic and environmental relationships between objectively measured traits and subjectively assessed wool traits were low to high in magnitude and variable in sign. The results showed that selection for LW and objective wool traits will thus not seriously compromise subjective wool and conformation traits. The exception is FD, CVFD and SDFD, where unfavourable genetic relationships with STAPL and BANDP were found. Furthermore, from these results, it was clear that unrestrained discrimination against wrinkles in Merino sheep would adversely affect economically important objectively measured wool traits such as SS, FD and wool weight. Selection against wrinkles should be carefully monitored to minimise the risk of animals producing reduced fleece weights with broader fibres. It is therefore, proposed that animals with excessive wrinkles be culled from the breeding stock.

[发布机构] University of the Free State