An appraisal of selection objectives and criteria for New Zealand Romney sheep with particular reference to wool traits : a thesis presented in partial fulfilment of the requirements for the degree of Master of Agricultural Science in Animal Science at Massey University

Abstract

Selection objectives and criteria were defined and appraised for a simple production system involving a New Zealand Romney breeding flock under North Island hill country conditions, in which all surplus offspring are sold as lambs. Through the availability of New Zealand Wool Board auction data for the seasons 1976/77 to 1980/81, the influence of wool quality traits on price was analysed by regression techniques. Traits examined included mean fibre diameter (MFD), style(S), mean length (ML) and yield (Y). For the 1980/81 season only, further data from the Coded Sales Assistance Report (C.S.A.R.) was available for scouring indicator (SI), colour indicator (CI), felted (F), pen stain (P), cotted (Co), tender (T), mixed length (LV) and mixed quality (QV). In addition, the effects on price of three non-fleece variables, lot weight (LW), mode of offering (MO) and New Zealand Wool Board market intervention policies (Int), were considered. Y was shown to have a major influence over greasy price. The relationship between price and ML was confirmed as being non-linear, with ML having a greater effect on the price of shorter wools. S and MFD were less influential. The control these four traits jointly exerted over greasy price ranged up to 74.0%, which was further enhanced by the introduction of quadratic terms. ML2 was the most important quadratic term. The inclusion of the C.S.A.R. and non-fleece related traits, failed to provide any further control over price. CI proved to be an effective substitute for S. Selection objectives were defined for greasy and clean wool, combined with short, long and mixed length categories. Economic weights for wool quality traits were directly calculated from the regression of auction price on the level of the traits. Economic weights for number of lambs weaned (NLW), weaning weight (WW), ewe body weight (EBW), greasy fleece weight (GFW) and clean fleece weight (CFW) were calculated using the marginal profit method. The relativities between the calculated economic weights were generally in good agreement with those of previously published estimates. For the selection objectives defined, various selection criteria were appraised. These included the traits in the selection objective, or their respective criteria, as well as hogget body weight (HBW), quality number (QN) and fleece character grade (CHG). NLW (dam), HBW and HGFW were of major importance in the selection index. The remaining traits were of only minimal value. On the basis of cost of measurement and value within the index, the full index was converted to a reduced index of NLW (dam), HBW and HGFW. In terms of accuracy of prediction and economy, this index was considered suitable for most commercial conditions. Further reduced indices were computed which generated less overall genetic gain, but which individual breeders may consider more appropriate to their particular requirements. Sensitivity analyses for HBW, NLW, GFW (CFW) and SC generally produced few changes of any consequence to the selection indices. Restriction of all genetic change in EBW significantly reduced the expected overall genetic gain.