Animal Science 2003, 76: /03/ $ British Society of Animal Science

Animal Science 2003, 76: 367-373 1357-7298/03/22330367$20 00 2003 British Society of Animal Science Genetic analysis of live weight and ultrasonic fat and muscle traits in a hill sheep flock undergoing breed improvement utilizing an embryo transfer programme J. A. Roden 1, B. G. Merrell 2, W. A. Murray 2 and W. Haresign 1 1 Institute of Rural Studies, University of Wales Aberystwyth, Llanbadarn Campus, Aberystwyth SY23 3AL, UK 2 ADAS Redesdale, Rochester, Otterburn, Newcastle on Tyne NE19 1SB, UK E-mail : jjr997@aber.ac.uk Abstract Genetic parameters for pre-weaning live weights and ultrasonic scanning measurements were estimated from a flock of Scottish Blackface sheep undergoing an embryo transfer programme. Maternal environmental effects could be evaluated without confounding with maternal genetic effects because embryos were transferred to unrelated recipient ewes. The data for the study were collected over a 7-year period (1993-1999) and related to a conventional hill farming system. The data were from 1465 lambs, the progeny of 60 sires, 263 donors and 784 recipient ewes. The only exception to the conventional farming system was at mating time when embryos were collected from selected donor ewes, following superovulation, and transferred to unrelated recipient ewes. Maternal environmental effects were important for birth weight (BW), 4-week weight (W4) and weaning weight (WW) but of less importance for ultrasonic fat depth (UFD), muscle depth (UMD) and muscle width (UMW). The heritabilities of the pre-weaning weights were moderate (0 17 to 0 23). The heritabilities of UFD and UMD were 0 44 and 0 27 respectively and were higher than found in previous similar studies. The heritability of UMW was low, 0 06. The genetic and phenotypic correlations among the pre-weaning weight traits were positive and moderate. There was a positive genetic ( + 0 25) and phenotypic ( + 0 24) correlation between UFD and UMD. The phenotypic and genetic correlations between BW and W4 and the scanning traits (UFD, UMD and UMW) were close to zero, and the correlations of WW with UFD, UMD and UMW were positive. The results of this study clearly demonstrate the importance of maternal environmental effects for lamb weights up to weaning and also suggest that genetic improvement for carcass composition in some populations of extensively reared hill sheep could be achieved more rapidly than previously thought possible. Keywords: embryo transfer, genetic parameters, live weight, sheep, ultrasonic measurements. Introduction Genetic improvement of hill sheep is an important objective within the UK sheep industry because of the importance of this sector in terms of maintaining the rural environment and its contribution of 34% of the genes to the slaughter generation (Meat and Livestock Commission, 1998). However, the extensive nature of the production system has meant that there have been few genetic studies of carcass and growth traits in hill sheep populations. 367 The studies of Atkins (1986) and more recently of Conington et al. (1995) have suggested that the heritability of pre-weaning traits in hill sheep populations is lower than those estimated for dual purpose type sheep (Fogarty, 1995). In addition, Conington et al. (1995 and 1998) and Bishop et al. (1996) have shown that, although in vivo measures of carcass composition (ultrasonic fat depth and muscle depth) have a moderate heritability in populations of Scottish Blackface sheep, the estimated values are again somewhat lower than those estimated in other

368 Roden, Merrell, Murray and Haresign breed types (Mercer et al., 1994; Fogarty, 1995). For genetic improvement programmes to be used effectively in hill sheep populations it is essential to have accurate estimates of the genetic parameters underpinning carcass quality traits and their correlations with other important maternal traits. The extensive nature of the production system and the relatively long period of dependence of the lamb on the ewe, suggest that maternal environmental effects are likely to have an important influence on lamb performance. It is, therefore, important to quantify this maternal effect when estimating genetic parameters. In order to separate the effect of the maternal genotype and the maternal environment an analysis of a large data set with a good depth of pedigree recording is required; this is not generally available for hill sheep systems. However, in a population undergoing embryo transfer it is possible to estimate the maternal environmental effect without any confounding with the maternal genetic effect. The objectives of this study were firstly to estimate genetic parameters for pre-weaning live weights and ultrasonic scanning measurements in a flock of Scottish Blackface sheep that was undergoing a routine embryo transfer programme as part of a selection experiment, and secondly to assess the importance of maternal environmental effects on these traits. Material and methods Flock establishment and maintenance An open nucleus flock of 60 Scottish Blackface donor ewes was established in 1992 from a base population of approximately 1600 ewes maintained in eight hefted (discrete, self-replacing) flocks, under harsh hill conditions, at ADAS Redesdale in Northumberland. The selection history of these base flocks was one of subjective visual inspection to eliminate faults, and to improve body size, conformation and breed characteristics. Selection of ewes into the nucleus flock initially was based on an analysis of progeny weaning weight and carcass classification records (including carcass weight, and subjective fat and conformation scores). From 1992 to 1994, the nucleus flock was replenished in November each year with mature (mainly third and fourth crop) ewes from the base flocks only, which were selected on the basis of progeny records. Thereafter, it was replenished with a combination of additional mature ewes identified from the base population, along with mature ewes and gimmers (18-months old first mated animals) bred within the nucleus flock itself. Over this period approximately half the replacements were nucleus bred. From 1995, nucleus-bred replacements were selected on an index designed to achieve equal genetic improvement in live weight at scanning, and in ultrasonic measures of muscle depth and muscle width without changing mean ultrasonic fat depth within the population. The index was (0 59 EBV LWT ) + (0 40 EBV UFD ) + (0 41 EBV UMD ) + (0 36 EBV UMW ), where EBV LWT, EBV UFD, EBV UMD and EBV UMW represent the estimated breeding values for live weight at scanning, ultrasonic fat depth, muscle depth and muscle width respectively. Secondary selection criteria were used in order to ensure that all selected ewe and ram lambs were physically sound and had adequate fleece type and breed characteristics. This was based on subjective assessments of the traits. Initially Scottish Blackface rams borrowed from the industry were predominantly used to mate with nucleus flock ewes. These were subjectively judged to have good breed characteristics and superior carcass quality. From 1995 onwards, high index rams, either borrowed from the industry and progeny tested, or bred within the nucleus, were used to mate with nucleus flock ewes. Sheep management The recipient ewes used in the study were two regular-aged flocks of unselected Scottish Blackface ewes managed in a similar way to the base population flocks and the nucleus flock itself. In mid November each year all nucleus donor and recipient ewes were initially transferred from open hill grazing to inbye fields. They were then housed for a short period immediately before and during the mating period (a total of approx. 2 weeks). This was done to facilitate the induction of superovulation, semen collection, AI, embryo collection and embryo transfer. Details of the treatment protocols for donor and recipient ewes and the methods for collecting and transferring embryos have been provided elsewhere (Bari et al., 1999 and 2000). Intravaginal progestagen pesseries (Chronogest, Intervet Laboratories, Cambridge, UK) were used to synchronize ewes in batches of 20 donors and 100 recipients on alternate days over a 6-day period, and were inserted on day 0. Superovulation was induced in donor animals by giving eight twice-daily injections of 1 1 mg of a highly purified preparation of ovine FSH (Ovagen, Immunological Products Ltd, Auckland, New Zealand), covering the period from 60 h before until 24 h after removal of the progestagen pessaries on day 12. Recipient ewes were given an i. m. injection of 350 i. u. PMSG (Intervet Laboratories, Cambridge, UK) at the time of pessary removal on day 12.

Genetic parameters for ultrasonic traits in hill sheep 369 Donor ewes were hand-mated with their respective rams at the onset of oestrus. Where semen collection from individual rams proved possible, the donor ewes were also subjected to intrauterine insemination with fresh diluted semen 44 h after pessary removal, using a dose of approximately 30 10 6 motile sperm per horn. Embryos were collected using a semi-laparoscopic approach on either day 19 or day 20 (5 or 6 days after mating respectively) and recipient ewes each received a single embryo. Within 2 days of the MOET programme being completed for each batch of donor and recipient ewes, the recipient ewes were returned to hill grazing. Thereafter, recipient ewes and their resulting lambs were subjected to management and husbandry typical of an extensive Northumberland hill farming system. While confined to the hill, ewes received supplementary feeding in the form of ad libitum baled silage and self-help feedblocks as necessary in order to maintain body condition. All ewes remained on open hill grazing until they were housed immediately before the start of lambing. The average lambing date was 12 April, and ewes and lambs were returned to the open hill grazing within 10 days of lambing. Male lambs were reared entire, and all lambs were weaned at a mean age of 20 weeks, at which time they were transferred from open hill grazing to improved pasture. Following ultrasonic scanning ewe and ram lambs were selected as nucleus flock replacements. Each year all lambs were scanned on the same day; the average age varied between years from 20 to 29 weeks. Performance records Data were available from seven years of the selection programme (1993-1999). A summary of these data is presented in Table 1. Birth weight (BW), 4-week weight (W4) and weaning weight (WW) at approximately 20 weeks of age were recorded for all lambs. Since scanning and weaning were carried out at the same time in a number of the years a separate scanning weight was not included in this study. Scanning was performed at the level of the 3rd lumbar vertebra using a Dynamic Imaging real-time ultrasound scanner, with a 7 5 Mhz 56-mm probe. Three individual subcutaneous fat depth measurements were taken over the longissimus dorsi muscle, moving laterally between the vertical and transverse processes of the vertebra, from which the average fat depth was calculated (UFD). A single muscle depth (UMD) and a single muscle width Table 1 Summary of traits included in analyses Number of donors 80 60 40 20 Age Trait Mean s. d. No. Mean s. d. Pre-weaning traits BW (kg) - 1465 4 1 0 59 W4 (kg) 33 3 3 1465 14 8 2 04 WW (kg) 145 10 5 1025 37 2 4 75 Scanning traits UFD (mm) 171 26 6 977 2 5 1 36 UMD (mm) 171 26 6 977 22 3 2 96 UMW (mm) 171 26 6 977 42 4 4 62 BW: birth weight; W4: 4-week weight; WW: weaning weight; UFD: ultrasonic fat depth; UMD: ultrasonic muscle depth; UMW: ultrasonic muscle width. (UMW) measurement was taken at the deepest and widest point on the muscle respectively. The 1465 lambs recorded at birth were the progeny of 60 sires, 263 donors and 784 recipients. A number of recipients were used in more than 1 year. Of the donors, 69 had known sire and 185 had a known dam, approximately half of which, themselves had a known dam; of the sires seven had a known sire and dam. The average number of progeny per donor was 5 5 with a minimum number of 1 and a maximum of 38. Approximately 80% of donors were represented by three or more progeny. The distribution of progeny per donor is shown in Figure 1. Ultrasonic scanning data were available on 977 of the lambs, representing 60 sires, 241 donors and 636 recipients. Of the 784 recipients used in the study, 54% were represented by two or more lambs. Statistical analysis Data were initially analysed using Genstat (1993) in order to identify significant fixed effects, covariates and two-way interactions. The fixed effects included in this analysis were : year (seven classes), lamb sex 0 1-2 3-4 5-6 7-8 9-10 11-12 13-14 15-16 17+ Number of progeny Figure 1 Distribution of number of progeny per donor for pre-weaning traits.

370 Roden, Merrell, Murray and Haresign (male or female), recipient age (five classes : 2, 3, 4, 5 and 6 years old) and recipient ewe s heft (two classes). Date of birth was fitted as a covariate for pre-weaning traits (BW, W4 and WW) and age at scanning was fitted as a covariate for scanning traits (UFD, UMD and UMW). As all animals were born and reared as singles, birth type and rearing type were not included in the analyses. Restricted maximum likelihood methods were then used to estimate variance components using an animal model. The ASREML program (Gilmour et al., 1999) was used for this analysis. All traits were initially analysed in two univariate models, one fitting animal as the only random effect (model 1), and a second (model 2) in which the effect of maternal environment was estimated by fitting recipient as a second non-correlated random effect. The significance of the second random effect was tested using the likelihood ratio test, comparing the difference in 2log likelihoods ( 2logL) between the two models to a critical value from a chi-square distribution. Each combination of traits was then analysed with a bivariate model using the most appropriate model for each trait. Results Environmental effects Analysis of pre-weaning traits showed that year, lamb sex, heft and recipient ewe age all had significant effects (P < 0 05) on these traits. As expected, male lambs were significantly heavier (P < 0 001) than female lambs at birth, 4 weeks and weaning. Lambs carried and reared by younger ewes (aged 2 and 3 years) were significantly lighter than those carried by older ewes (aged 4, 5 and 6 years). In addition, there were significant interactions (P < 0 001) between year and sex, and year and flock for 4-week weight and weaning weight; and recipient ewe age and both flock and year for 4-week weight. The effect of date of birth was significant (P < 0 001) for all live-weight traits before weaning. The effect on birth weight was positive, indicating that lambs born later in the lambing period were heavier, but negative for 4-week weight and weaning weight as would be expected given that within a year all lambs were weighed on one date, irrespective of age. With respect to scanning traits, males had a significantly higher ultrasonic muscle depth than females (P < 0 001), but there was no significant difference in ultrasonic fat depth or muscle width. Recipient age had a significant effect on all three scanning traits (P < 0 05). Two-year-old ewes (gimmers) reared lambs that had significantly lower Table 2 Estimates of components of variance and heritability for pre-weaning traits using univariate models 1 and 2 Trait BW W4 WW Model 1 2 1 2 1 2 σ 2 a 0 06 0 06 0 51 0 58 2 09 2 13 σ 2 c 0 09 0 69 2 57 σ 2 e 0 21 0 12 2 24 1 51 10 44 7 87 σ 2 p 0 27 0 27 2 75 2 78 12 53 12 57 h 2 0 22 0 23 0 18 0 21 0 17 0 17 s.e. 0 049 0 046 0 048 0 049 0 045 0 044 c 2 0 31 0 25 0 20 s.e. 0 033 0 035 0 035 2logL 101 2 56 1 37 8 BW=birth weight; W4=4-week weight; WW=weaning weight. σ 2 additive genetic variance, a σ2 variance due to maternal c environment, σ 2 error variance, e σ2 phenotypic variance. p h 2 heritability, c 2 maternal environmental variance expressed as a proportion of phenotypic variance. 2logL = 2log likelihoods expressed as a deviation from model 1. UFD, UMD and UMW measurements than lambs reared by older ewes. There was a significant positive relationship (P < 0 01) between age at scanning and muscle depth, but not fat depth and muscle width. Genetic parameters The estimates of components of variance and heritability for pre-weaning traits are shown in Table 2. For all pre-weaning live-weight traits (BW, W4, WW), including the effect of the maternal environment improved the fit of the model and this is reflected by the improved log likelihood of model 2 compared with model 1. The heritability values for all three traits were moderate (0 17 to 0 23) and the c 2 values were of a similar magnitude (0 20 to 0 31). The trend of c 2 values suggests that the influence of the maternal environment became less important as the lamb progressed from birth to weaning. Table 3 Estimate of phenotypic (above diagonal) and genetic (below diagonal) correlations and mean heritability (on diagonal) for preweaning traits using a bivariate model (model 2) Trait BW W4 WW BW 0 23 +0 58±0 019 +0 32 ±0 026 W4 +0 71±0 091 0 21 +0 63 ±0 017 WW +0 34±0 153 +0 62±0 115 0 17 BW=birth weight; W4=4-week weight; WW=weaning weight.

Genetic parameters for ultrasonic traits in hill sheep 371 Table 4 Estimates of components of variance and heritability for scanning traits using univariate models 1 and 2 Trait UFD UMD UMW Table 6 Estimates of genetic and phenotypic correlations for preweaning and scanning traits using a bivariate model Trait UFD UMD UMW Model 1 2 1 2 1 2 σ 2 a 0 48 0 48 1 45 1 47 1 04 0 98 σ 2 c - 0 08-0 60-1 17 σ 2 e 0 59 0 51 4 05 3 44 14 24 13 13 σ 2 p 1 07 1 07 5 50 5 51 15 28 15 28 h 2 0 44 0 44 0 26 0 27 0 07 0 06 s.e. 0 074 0 074 0 064 0 064 0 043 0 042 c 2-0 08-0 11-0 08 s.e. - 0 041-0 055-0 052-2logL - 4 2-4 7-2 0 The estimated genetic and phenotypic correlations among the pre-weaning traits are shown in Table 3, with mean heritability values calculated from the bivariate models. The heritability values presented in Table 2 (using univariate models) and Table 3 (using a bivariate model) are very similar, as would be expected. The genetic correlations among live weights were all positive. The genetic correlation of birth weight and 4-week weight, and of 4-week weight and weaning weight were both relatively high. Estimates of components of variance and heritability for scanning traits are shown in Table 4. Unlike preweaning weight traits, the variance attributable to maternal environment for scanning traits was small. Inclusion of this effect in the analytical model only slightly improved the fit for fat depth and muscle depth, but had no effect for muscle width. The heritability for fat depth was relatively high. The heritability of muscle depth was moderate (0 27), but in contrast the heritability for muscle width was much lower (0 07). The c 2 values for fat depth and muscle depth, although significant, were very low. Table 5 Estimate of phenotypic (above diagonal) and genetic (below diagonal) correlations and mean heritability (on diagonal) for scanning traits using a bivariate model Trait UFD UMD UMW UFD 0 44 +0 24±0 035 +0 18±0 033 UMD +0 25±0 154 0 28 +0 53±0 025 UMW 0 06±0 268 +0 09±0 289 0 07 Genetic BW 0 28±0 143 0 18±0 172 0 04±0 272 W4 0 19±0 163 0 05±0 186 +0 05±0 285 WW +0 15±0 165 +0 31±0 172 +0 19±0 274 Phenotypic BW 0 09±0 035 0 01±0 035 +0 06±0 032 W4 +0 10±0 036 +0 15±0 035 +0 14±0 033 WW +0 30±0 033 +0 34±0 031 +0 31±0 031 See Table 1 footnote for abbreviations. The estimates of phenotypic correlations among UFD=ultrasonic fat depth; UMD=ultrasonic musclescanning traits (Table 5) were positive, but generally depth; UMW=ultrasonic muscle width; See able T 2 footnote low. The genetic correlations were also all low with for other abbreviations. large standard errors, indicating that they were not significantly different from zero. The correlations between pre-weaning and scanning traits are shown in Table 6. The genetic correlations of all three scanning traits with birth weight were negative, although not significantly different from zero. The phenotypic correlations were also close to zero. The genetic correlation of 4-week weight with fat depth was negative, but close to zero for muscle depth and muscle width. All of these genetic correlations were not significantly different from zero. The phenotypic correlations involving 4-week weight were all low and positive. Both the genetic and phenotypic correlations of weaning weight with all three scanning traits were positive. Discussion The use of data from a population undergoing a multiple ovulation and embryo transfer programme presents a unique opportunity for the estimation of genetic parameters. The key features that distinguish the data in this study are the pedigree structure which is characterized by relatively large maternal full- and half-sib families and the fact that the lambs included in the study were implanted as day 5 or day 6 embryos into unrelated recipient ewes, and subsequently reared by those ewes. For these lambs, there is therefore no confounding of the genetic and maternal additive or environmental effects, as would be the case for more conventional flocks used in previous studies. A typical sheep population consists of a relatively small number of sires mated to a large number of dams, each represented by a small number of progeny. The information contributed by the dams is UFD=ultrasonic fat depth; UMD=ultrasonic muscle depth; UMW=ultrasonic muscle width. of limited value for parameter estimation because

372 Roden, Merrell, Murray and Haresign direct additive effects are confounded with both maternal additive effects and common environmental effects. To overcome this problem a large data set relating to a population with performance information covering a number of generations is generally required. However, the population in this study consists of relatively large maternal half-sib and full-sib families, in addition to the usual large paternal half-sib families. This means that dams contribute more useful information on direct additive effects, than in a conventional population, and, with the possible exception of cytoplasmic effects, these are not confounded with maternal additive effects. Therefore, although the total number of animals contributing performance data in this study is relatively low, the pedigree structure of the data is far more suited to accurate estimation of genetic parameters than a conventionally bred sheep population. This study also presents a unique opportunity to quantify the influence of the maternal environment on the early growth and measures of post-weaning body composition in hill sheep, managed in an extensive environment. The results indicate that, in this population, the maternal environment is indeed an important influence on pre-weaning growth of lambs, but not on ultrasonic measures of body composition measured post-weaning. The estimates of c 2 for pre-weaning traits are similar to those reported by Conington et al. (1995) for Scottish Blackface sheep, although the heritability values are considerably higher in this study. Saatci et al. (1999) reported the results of analysis of 12-week weight records of Welsh Mountain sheep and demonstrated the importance of the maternal environment on lamb performance. Their analysis, which separated the effect into temporary and permanent environmental effects, indicated that it was the temporary effects that were most important. In the present study it was not possible to separate these effects, but the variation accounted for was of a similar magnitude to the total variation accounted for by both permanent and temporary environmental effects in the study of Saatci et al. (1999). The heritability estimates for the pre-weaning traits are somewhat higher than those reported by Conington et al. (1995) (birth weight : h 2 = 0 07, 6 week weight : h 2 = 0 02), Bishop et al. (1996) (weaning weight : h 2 = 0 12) and Atkins (1986) (birth weight : h 2 = 0 13, pre-weaning weights : h 2 = 0 06) for Scottish Blackface sheep managed in a similar fashion to the flock at ADAS Redesdale. They are, however, more consistent with those summarized by Fogarty (1995) for dual-purpose breeds and Saatci et al. (1999) for Welsh Mountain sheep. Fogarty (1995) reported a mean heritability of 0 19 for birth weight and 0 20 for weaning weight, derived from 19 estimates, most of which were based on half-sib analyses. The low heritability values found in the studies of Conington et al. (1995), Bishop et al. (1996) and Atkins (1986) were attributed to the high level of environmental variation due to the extensive nature of the management system, typical in hill sheep systems. The estimates of heritability for ultrasonic measures of post-weaning body composition were considerably higher, but the c 2 values were similar to those reported in previous studies of Scottish Blackface sheep kept in an extensive environment (Bishop et al., 1996; Conington et al., 1998). Indeed, the estimates of heritability for fat depth are similar to those reported by Bishop et al. (1996) for Scottish Blackface sheep kept in an intensive environment (h 2 = 0 39), by Saatci et al. (1998) for Welsh Mountain sheep (h 2 = 0 40) and by Thorsteinsson and Eythorsdottir (1998) for Icelandic sheep (h 2 = 0 42). They are also considerably higher than those summarized by Fogarty (1995) for dual-purpose sheep (a mean of 0 28 (no. = 30) for fat depth). The heritability of muscle depth is also slightly higher than that found in previous studies (Bishop et al., 1996 (h 2 = 0 25) and 1998 (h 2 = 0 20); Saatci et al., 1998 (h 2 = 0 19)). These results suggest that the potential for genetic progress for body composition traits in hill sheep maintained under extensive conditions may be higher than previously thought, but may also vary greatly between populations. The heritability of ultrasonic muscle width is considerably lower than that of muscle depth. This most probably reflects the fact that this parameter was more difficult to measure accurately than was muscle depth. The genetic and phenotypic correlations among the pre-weaning weight traits were positive and moderate in magnitude, as would be expected. The correlations of these traits with the scanning traits, were generally close to zero or slightly positive. The results of this study have demonstrated the importance of the maternal environmental effect on pre-weaning growth of hill lambs. Given the magnitude of this effect it should clearly be accounted for in models used for genetic evaluation of pre-weaning traits. The estimates of heritability for ultrasonic measures of body composition are considerably higher than those reported in previous studies of apparently similar populations. This suggests that genetic progress for carcass

Genetic parameters for ultrasonic traits in hill sheep 373 composition in some populations of hill sheep could be achieved more rapidly than previously thought possible. Acknowledgements Financial support for this work from the Ministry of Agriculture, Fisheries and Food is gratefully acknowledged as is the guidance and loan of rams provided by the Blackface Sheep Breeders Association. References Atkins, K. D. 1986. A genetic analysis of the components of lifetime productivity in Scottish Blackface sheep. Animal Production 43: 405-419. Bari, F. Y., Khalid, M., Haresign, W., Merrell, B., Murray, A. and Richards, R. I. W. 1999. An evaluation of the success of MOET in breeds of hill sheep maintained under normal systems of hill flock management. Animal Science 69: 367-376. Bari, F. Y., Khalid, M., Haresign, W., Murray, A. and Murray, B. 2000. Effect of mating system, flushing procedure, progesterone dose and donor ewe age on the yield and quality of embryos within a MOET programme in sheep. Theriogenology 53: 727-742. Bishop, S. C., Conington, J., Waterhouse, A. and Simm, G. 1996. Genotype environment interactions for early growth and ultrasonic measurements in hill sheep. Animal Science 62: 271-277. Conington, J., Bishop, S. C., Waterhouse, A. and Simm, G. 1995. A genetic analysis of early growth and ultrasonic measurements in hill sheep. Animal Science 61: 85-93. Conington, J., Bishop, S. C., Waterhouse, A. and Simm, G. 1998. A comparison of growth and carcass traits in Scottish Blackface lambs sired by genetically lean or fat rams. Animal Science 67: 299-309. Fogarty, N. M. 1995. Genetic parameters for live weight, fat and muscle measurements, wool production and reproduction in sheep: a review. Animal Breeding Abstracts 63: 101-143. Genstat 5 Committee. 1993. Reference manual. Oxford University Press, Oxford. Gilmour, A. R., Cullis, B. W., Wellham, S. J. and Thompson, R. 1999. ASREML: program user manual. NSW Agriculture, Orange, Australia. Meat and Livestock Commission. 1998. Sheep yearbook 1998. Meat and Livestock Commission, Milton Keynes, UK. Mercer, J. T., Brotherstone, S., Bradfield, M. J. and Guy, D. R. 1994. Estimation of genetic parameters for use in sheep sire referencing schemes. Proceedings of the fifth world congress on genetics applied to livestock production, Guelph, vol. 18, pp. 39-42. Saatci, M., Ap Dewi, I. and Ulutas, Z. 1998. Correlations between weaning weight, carcass traits and ultrasonic body measurements in Welsh Mountain sheep. Proceedings of the British Society of Animal Science, 1998, p. 84. Saatci, M., Ap Dewi, I. and Ulutas, Z. 1999. Variance components due to direct and maternal effects and estimation of breeding values for 12-week weight of Welsh Mountain lambs. Animal Science 69: 345-352. Thorsteinsson, S. and Eythorsdottir, E. 1998. Genetic parameters of ultrasonic and carcass cross-sectional measurements and muscle and fat weight of Icelandic lambs. Proceedings of the sixth world congress on genetics applied to livestock production, Armidale, vol. 24, pp. 149-152. (Received 3 April 2002 Accepted 12 December 2002)