Abstract:
Proc. Aust. Soc, Anim. Prod. Vol. 18 SIRE SELECTION EFFICIENCY IN A QUEENSLAND MERINO STUD USING WOOLPLAN Mary ROSE* In a Peppin Merino stud using WOOLPLAN, the productivity of classer's grades was examined and the selection efficiency of sires estimated for the years 1986 There were significant differences between grades for clean fleece to 1988. Reserve rams had higher clean fleece weights, fibre diameters and weight. There was considerable selection indices than sale rams in each year. Efficiency of sire variation in the production characters in both grades. selection using a selection index ranged from 61 to 71% suggesting that some Sires also had emphasis was placed on the selection index in sire selection. finer than average wool in all years. INTRODUCTION WOOLPLAN, the national performance recording scheme for Merino and other nonpedigreed wool sheep, became available to studbreeders in Queensland in 1986. It provides several breeding objectives as options and a number of measured For the studbreeder it meant a change from selection criteria (Ponzoni 1987). selection of rams using independent culling levels for the measured characters to the use of a selection index which combined a number of criteria into a single score. During the introduction of WOOLPLAN studbreeders were supplied both the WOOLPLAN output and the old fleece testing report with which they were familiar. The old report showed the measured characters and ranked the animals on the basis of clean fleece weight percent. This enabled breeders to see that WOOLPLAN, while providing a more accurate assessment and ranking of the rams than the measured criteria, did not lead to the selection of a different type of ram from those selected by past methods. Because of breeders' requests, WOOLPLAN's standard output also lists the raw data for the measured characters. This study examined the classing and selection of rams in a stud using WOOLPLAN and estimated the sire selection differentials for measured characters and for selection index. In addition the relative performance of the reserves, flock ~ rams and visual culls and the variation in these grades were estimated. XATKRIAJL AND HE!IBODS The stud and its management The stud was a Peppin Merino stud located in south west Queensland, All rams were shorn as lambs and classed prior to shearing at 12 months with approximately 8 months' wool growth. In each year of the study (1986, 1987, 1988) rams were classed as reserves, flock rams or culls before hogget shearing. Potential sires were chosen with the aid of measurement at 15 months and final sire selection was undertaken prior to joining at 18 months of age, The numbers in each class are shown in Table 1. - * Wool Biology Laboratory, Department of Primary Industries, Yeerongpilly, Qld, 4105, 340 Proc. Aust, Soc. Anim, Prod. Vol. 18 Table 1 Numbers of rams classed visually as reserves, flock rams and culls at hogget shearing and the number of stud sires selected * Values in parenthesis are % of total drop; n-a. data on culls unavailable Breeding objective and selection criteria The breeding objective in each year was to increase clean wool production while The selection criteria were restricting fibre diameter to its existing level. clean fleece weight and fibre diameter. In the first year live weight was also measured as a selection criterion; the objective being to increase reproduction Because of the rate, hogget live weight and cast-for-age ewe live weight. problems of measuring large numbers of rams off-shears at similar rates of fill and the perceived small genetic changes likely, this measurement was omitted in following years. Calculations Sire selection differentials were calculated by expressing the selection differential achieved for the character or index as a percentage of the maximum possible selection differential (Riches and Turner 1955). The maximum value would have been obtained if the required number of rams were selected solely in order of the character or index. RESULTS Productivity of classer's grades The performance of the reserves, flock rams and visual culls for each year (1986-88) is shown in Table 2. There were significant differences between reserves and flock rams in each year; the reserves having higher clean fleece weights, fibre diameters and selection indices. Considerable variation, as measured by the standard deviation, was observed for each character in each grade at the three shearings. Table 3 shows the means for measured characters and selection indices of the three drops of rams and the selected sires and the efficiency of selection achieved, Sires selected in each year had significantly higher clean fleece weights and finer wool than the mean of the drop. In 1986 sires had a significantly higher live weight also. 341 Proc. Aust. Soc. Anim. Prod. Vol. 18 Table 2 Mean (s-d.) production and selection index for rams in the classer's grades at 12 months with 8 months' wool growth (1986-88) Table 3 Means for measured characters and selection indices for ram drop and selected sires for the years 1986-88 and the sire selection efficiency * Mean for reserves and flock rams only The efficiency of sire selection based on the selection index ranged from 61% {1986 drop) to 71% (1987 drop). Sires selected were not confined to the classer's reserves alone and the percentages chosen from rams not in the reserves were 16, 57 and 35 for the 1985, 1986 and 1987 drop respectively. The selection efficiency for clean fleece weight ranged from 55 (1985 drop) to 76% (1987 drop). The selection efficiency for live weight was 54% for the 1985 drop. 342 Proc. Aust. Soc. Anim, Prod. Vol. 18 These findings on the relative productivity of classer's grades and the variation within the grades are generally consistent with those observed previously in this stud, although for a different classer, by Rose (1980) and The variability of production within classer's grades McGuirk et al. (1982). is of great importance not only to studs but to their clients and underlies the demand for grading of sale rams based on measured productivity. It should be noted that the variation observed in these grades was generally much less than the phenotypic variances used in WOOLPLAN (Ponzoni 1987). The selection efficiency for clean fleece weight, when only clean fleece weight and fibre diameter were measured, was 76% and 70% for the 1986 and 1987 drops These values are also similar to those reported by McGuirk et respectively. When live weight was also included at the 1986 shearing the al. (1982). selection efficiency for clean fleece weight was much lower which suggested that emphasis was being given to both characters and that they are not very highly correlated phenotypically (rp = 0.10 for this drop). This study has confirmed that this stud gave effective emphasis to the selection index in the selection of sires and did in fact use rams with high selection indices which were not classed into the reserves by the classer. This was contrary to the previous policy in this stud when rams were selected There was also some primarily on clean fleece weight (McGuirk et al. 1982). emphasis placed on selection of sires for wool quality. Although the objective used would restrict fibre diameter or reduce it marginally, this stud continues the policy of discarding any ram more than two standard deviations above the mean for fibre diameter regardless of selection index. Selection using a selection index has been shown to be the most efficient method when selecting for a number of traits together. However there are a number of visually assessed characters which breeders think important and which are not presently included in the WOOLPLAN options. However by defining their objective clearly a stud such as this may attach more realistic importance to these traits and reassess rams with high scores for selection as sires. This is not possible where studs measure only reserve rams and this stud is one of only a small number of Queensland studs which presently measure all reserve and The use of a two-stage selection of sires would ensure that studs sale rams. did not overlook high producing rams. ACKNOWLEDGEMENTS I would like to thank the studbreeder and his staff for their assistance in undertaking this study. The work was supported by the Wool Research and Development Fund, REFERENCES MCGUIRK, B-J., ROSE, Mary and SCOTT, R. (1982). Aust. J. Exp. Agric. Anim. Hush. 22:274. PONZONI, R-W. (1987). In 'Merino breeding programs in Australia', p-25, editor B.J. McGuirk. (Australian Wool Corporation: Melbourne). RICHES, J.H. and TURNER, Helen Newton (1955). Aust. J. Agric. Res. 6:99, ROSE, Mary (1980). Proc, Aust. Soc. Anim. Prod. 13:428. - 343