Litter size of commercial European sow lines can be increased by the introduction of genes from the highly prolific, but obese Meishan. The introduction of these genes into sow lines may necessitate different nutritional strategies for lactation. A 2 × 2 × 2 factorial experiment was used to investigate the influence of breed type (Meishan synthetic 50% (M) or purebred European Landrace and Large White (W)) and parity (1 or 3) on response to dietery protein level (H = 178 g crude protein (CP) per kg 8 g lysine per kg, L = 136 g CP per kg 5 g lysine per kg over a 5-week lactation. M were lighter and fatter at farrowing than W (P < 0·001), consumed more food (M = 7·33 v.W = 6·98 (s.e.d. 0·17) kg/day, P < 0·05) but had higher maternal weight loss (M = 16·2 v.W = 11·9 (s.e.d. 1·88) kg, P < 0·05) and higher fat loss (M = 6·5 v. W = 26 (s.e.d. 077) mm P2, P < 0·001) during lactation. Litter growth rate was higher for M than for W (P < 0·005) and for H than for L (P < 0·005). Differences were found between breeds in the partitioning of food protein. Overall, higher correlations between lysine intake and litter growth rates were found in M (r = 0·55, P < 0·001) than in W (r = 0·13, P > 0·05). Lysine intake and sow weight loss were more correlated in W(r = 0·66, P < 0·001) than in M (r = 0·15), P > 0·05). Fat loss was correlated with initial fat depth in M (r = 0·61, P < 0·001) but not in W. Weaning to conception intervals showed a breed × diet interaction (loge days: ML = 1·38, MH = 1·55, WL = 1·73, WH = 1·43, s.e.d. 0·14, P < 0·05). It is concluded that M, with greater initial fat reserves, catabolize more body fat to support greater milk output. This is facilitated by provision of a high protein diet. The W sows do not utilize the extra protein for milk production but instead use it to minimize weight loss (maintain protein status of their own body) which may help prevent an extended weaning to conception interval.

Growth and survival from birth to weaning were monitored during three generations of crossbreeding between British Large White (LW) and Chinese Meishan (MS) pigs. The design allowed comparisons between sow genotypes ranging from zero to all MS genes, which were mated toLWor MS boars, to produce progeny with proportions of 0·0 to 0·5 or 0·5 to 1·0 MS genes, respectively. Crossbreeding parameters of both maternal and direct piglet performance were estimated for the first two parities using restricted maximum likelihood (REML) methods for litter traits (litter weight at birth, litter mean and within litter standard deviation of piglet weight at birth, proportion surviving to weaning, litter size and weight at weaning and litter mean piglet weight at weaning) and for traits of the piglet (birth weight, probability of survival and weaning weight). For litter traits, the estimated contribution of the additive maternal effect to the breed differences (MS-LW) was significant for litter mean piglet birth weight (–0·46 (s.e. 0·04) kg), survival to weaning (0·15 (s.e. 0·02)), litter size at weaning (1·6 (s.e. 0·16) piglets), litter weaning weight (–11·2 (s.e. 3·8) kg) and litter mean piglet weaning weight (2·54 (s.e. 0·24) kg). Adding litter size and litter mean piglet birth weight to the model removed the additive maternal contribution to the breed differences in survival, and litter size and reduced that for litter mean piglet weaning weight. The contribution of the direct additive effect to the breed difference (MS-LW) was significant for the within litter standard deviation in birth weight (0·018 (s.e. 0·006)), survival to weaning (0·12 (s.e. 0·02)) and litter size (1·12 (s.e. 0·64)) and weight (11·6 (s.e. 4·0) kg) at weaning, but not for piglet weight at birth or weaning. Fitting litter size and litter mean birth weight had comparatively little impact on the direct additive effects. There were significant maternal heterosis effects for litter weight at birth and litter size and weight at weaning, the estimated deviation of the F1 from the midpoint of the two purebreds 3·22 (s.e. 0·55) kg, 2·20 (s.e. 0·47) piglets, and 20·1 (s.e. 3·3) kg respectively, but none for survival or piglet weights. There were direct heterosis effects for litter weight and litter mean piglet weights, the estimated deviation of the Fjfrom the mid point of the two purebreds being 1·16 (s.e. 0·41) kg and 0·14 (s.e. 0·02) kg, for survival to weaning (0·04 (s.e. 0·02)) and for litter weight (11·2 (s.e. 2·5) kg) and litter mean piglet weight (0·96 (s.e. 0·17) kg) at weaning. Fitting litter size and litter mean piglet birth weight removed or reduced both maternal and direct heterosis effects. Individual piglet analyses gave similar results to analyses of the equivalent sow trait. It was concluded that in litters born to MS cows, the lower piglet survival and lower weaning weights were related to the larger litter sizes and lower piglet birth weights. For their birth weight, however, MS piglets have a greater ability to survive and thrive. The large direct and maternal heterosis effects observed for litter and mean piglet weight at weaning werepartly associated with the heavier birth weight of the crossbred piglet.

A crossbreeding trial extending over three generations was used to investigate the genetic components contributing to the prolificacy of the Meishan breed in comparison with the Large White breed. Information on the number of teats and on body weight and litter size in the first two parities was recorded on purebred Meishan and Large White females and on reciprocal F1 and backcross females. Ovulation rate was also recorded for all litters, allowing the estimation of per litter prenatal survival. Crossbreeding parameters for direct, maternal and grandmaternal effects were estimated using restricted maximum likelihood analysis. There was a consistent advantage of three to four piglets born alive to the Meishan female compared with the Large White female. This was controlled by the maternal genotype, with no effect of the genotype of the litter itself. Both additive and heterosis effects were important, the contribution of additive maternal effects to the breed difference being similar across parities (4·0 (s.e. 1·1) and 4·2(s.e. 1·1), in the two parities respectively) and the maternal heterosis increasing slightly across parities (2·2 (s.e. 0·8) and 2·9 (s.e. 0·8), in the two parities respectively). The number born alive to F 1 females was similar to, or greater than, the number born alive to Meishan females. Ovulation rate was significantly higher in Meishan than in Large White females and this was controlled by additive gene effects which had a similar effect across parities, the weighted average of their contribution to the breed difference being 5·7 (s.e. 0·8) ova. Differences between the breeds in prenatal survival were small, although there was significant maternal heterosis, however, the maternal additive effect became significant after the inclusion of ovulation rate as a covariate. After adjustment for ovulation rate, the weighted average estimates across parity of the maternal additive contribution to the breed difference and the maternal heterosis for the proportional prenatal survival loere 0-14 (s.e. 0·05) and 0·13 (s.e. 0·03), respectively. This suggests that a combination of a high ovulation rate and especially a high level of prenatal survival for that ovulation rate led to the prolificacy observed in this sample of Meishan pigs. The inclusion of ovulation rate as a covariate in the analysis of number born alive confirms this view, as the maternal additive effect on litter size was only reduced by about one third and the heterosis effect was largely unchanged. Both numbers stillborn and mummified were increased in litters born to Meishan sows due to maternal additive effects, but the effects seemed largely a consequence of the increased ovulation rate as they became non-significant after its inclusion as a covariate in the model.

Two studies, one using a trained taste panel and the other a consumer panel, were carried out to evaluate the eating quality of the Meishan breed. Entire male and female pigs of four genotypes: purebred Meishan (MS) and Large Wliite (LW) and the reciprocal crosses (MS♂ × LW ♀ and LW♂ × MS♀) were reared in single sex groups and given a commercial diet ad libitum from 35 kg live weight to slaughter at around 70 kg. For the taste panel, a loin joint was roasted under standard conditions and samples of fat and lean were presented to the panellists. There were no statistically significant differences between the genotypes for tenderness, juiciness, flavour, odour, incidence of boar taint or overall acceptability. Purebred Meishan samples had a higher incidence of abnormal odours but lower cooking losses and shear force values than the other three genotypes. Meat from gilts was judged to be significantly more tender and juicy, with a weaker pork flavour but a lower incidence of abnormal flavours and higher overall acceptability than that from boars. There were statistically significant interactions between genotype and sex for tenderness, abnormal odour and shear force which mainly involved the purebred Meishans and were of little practical significance. In the consumer study, loin chops and leg joints from purebred LW and the two crossbred genotypes were evaluated. Households received two samples of the same type of joint from the same sex in two separate distributions. Each household received an LW sample and a sample from one of the crossbred genotypes. In general, consumers found the appearance of the joints from the three genotypes to be equally acceptable. For eating quality, the within-household deviations of the crossbred compared with the LW suggested that MS♂ × LW♀ samples were considered to be of better eating quality, particularly for juiciness (deviation -0·71, s.e. 0·24, P < 0·01). In contrast, LW♂ × MS♀ samples were generally considered inferior, particularly in terms of juiciness (+ 0·59, s.e. 0·26, P < 0·05) and flavour (+0·63, s.e. 0·27, P < 0·05). However, the overall acceptability of both crossbreds was considered little different from the Large White. Overall, the results of this work suggest little benefit in eating quality for the Meishan under United Kingdom production conditions.

The objective of the present experiment was to test whether the relationship between feed composition and the efficiency of protein utilization in pigs was different between two very different breeds. Fortyeight entire male pigs, half Large White × Landrace (LW ×) and half Chinese Meishan (CM), were assigned at 13 kg live weight either to an initial slaughter group or to one of six feeding treatments for 6 weeks. The two feeds used were starch, with 12.52 MJ metabolizable energy (ME) and 0 kg digestible crude protein (DCP)/kg, and a high-protein feed, with 11.95 MJ ME and 0.362 kg DCP/kg. In the first week the LW × pigs were given 400 g high-protein feed/d with 0, 100, 200 or 300 g starch/d, or 200 g high-protein feed/d with 400 or 500 g starch/d. The six diets had ME:DCP ratios of between 33 and 119 MJ/kg. The diet compositions were held constant within each of the six treatments but the allowances were increased weekly on time-based linear scales to be 2.25 as great in week six as in week one. The allowances for the CM pigs were 0.75 of those for the LW ×. The calculated efficiency of protein utilization, ep, was found to be directly proportional, up to a maximum value, to the ME:DCP ratio of the diet for both breeds, as found previously for LW × male and female pigs. The overall constant of proportionality was ep = 0.0108 (SE 0.00024) ME:DCP, the value of which did not differ significantly between the two breeds. The maximum observed ep values were 0.710 (SE 0.014) and 0.824 (SE 0.016) for the LW × and CM pigs respectively. It is likely that the lower maximum value for the LW x pigs was an experimental artefact due to increased feed spillage by these pigs on the two highest starch allowances used. In previous experiments the maximum value of ep was found to be 0.813 for LW × pigs, a value similar to that found here for the CM pigs. The results are consistent with the view that the two very different pig breeds have the same net efficiency in using protein and use the same rules for partitioning a scarce resource such as energy.

The prospect that genes from the Chinese Meishan pig will be used to improve reproductive performance of European pigs requires that the Meishan is evaluated for traits of economic importance and the genetics of any breed effects investigated. Entire male and female pigs of four genotypes; purebred Meishan (MS) and Large White (LW) pigs and both reciprocal Fl crossbred genotypes (MS ♂ × LW ♀and LW ♂ × MS ♀), were fed ad libitum and slaughtered at about 70 kg. After slaughter, carcass weight, dimensions and conformation were recorded and fat depths at the shoulder, mid back and loin were measured on the carcass, together with fat depths and eye muscle dimensions recorded on a section cut at the last rib position. Drip loss, muscle reflectance and pH were measured and subjective assessments of fat firmness, fat separation, muscle marbling and muscle colour were made. Genotypic means and genetic crossbreeding effects (direct additive and heterosis effects and maternal additive effects) were estimated using restricted maximum likelihood procedures. When compared with the LW, MS carcasses had a similar killing-out proportion but were shorter with a poorer conformation. MS carcasses had substantially greater fat depths and reduced eye muscle dimensions in comparison with the LW. Differences between genotypes for fat depths and muscle dimensions were largely controlled by direct additive gene action, there was little evidence of direct heterosis and hence Fl crosses were intermediate between the purebreds. Significant genotype × sex interaction meant that genotypic differences for these traits were much greater in females than in males. Significant direct additive genetic differences between female genotypes were found for fat firmness, fat separation and drip loss, but these effects may in part have been a consequence of subcutaneous fat depth differences between genotypes. The degree of marbling was greater in MS females than in females of the other three genotypes and these genotypic differences did not seem to be associated with subcutaneous fat depth. Muscle from MS males was darker in colour with a lower reflectance than that from LW males. Genotype × sex interaction was such that genotypic differences for meat and fat quality traits differed between sexes.

Genes from the Chinese Meishan pig have the potential to enhance reproductive performance of European pigs. In order to allow prediction of the impact of Meishan genes in a range of alternative improvement programmes all traits of economic importance must be evaluated and genetic crossbreeding effects estimated. Entire male and female pigs of four genotypes, purebred Meishan (MS) and Large White (LW) pigs and both reciprocal Fl crossbred genotypes (MS ♂ × LW ♀ and LW ♂ × MS ♀), were farrowed in Edinburgh and subsequently performance tested at either Edinburgh or Newcastle. In Edinburgh, animals were penned in groups of four and fed ad libitum between pen mean weights of approximately 30 and 80 kg. At the end of test fat depths at the shoulder, last rib and loin were measured ultrasonically. In Newcastle, animals were penned in groups of six and fed ad libitum between pen mean weights of approximately 30 and 70 kg. Genotypic means and genetic crossbreeding effects (additive and heterosis direct effects and additive maternal effects) were estimated using restricted maximum likelihood.

When compared with the LW, the direct additive effect of genes from the MS produced increased growth rate up to the time of weaning, no change in growth rate between weaning and start of test and greatly reduced growth rate during the performance test. The maternal additive effect of genes from the MS was to reduce growth rate up to the time of weaning, with little effect thereafter. There was substantial direct heterosis for growth rate in all periods measured, but heterosis was less in males than in females during the performance test. The combined effect was such that, within sex, the LW and the two crossbred genotypes were of similar ages when they reached 70 and 80 kg, but MS pigs were 38 to 60 days older. The direct additive effect of MS genes was to increase subcutaneous fat levels and there was little evidence for the effects of maternal genes or direct heterosis on these traits. There was a direct additive effect of MS genes reducing food intake and increasing food conversion ratio and there was direct heterosis for increased food intake. There were significant interactions between genotype and sex. Male and female LW pigs had a similar performance but male MS pigs had slower growth rates on the performance test with lower food intakes, food conversion ratios and subcutaneous fat levels than the females.