M. J. Barhorst
K. M. Irvin1
S. J. Moeller
S. M. Neal
The Ohio State University
Department of Animal Sciences
1 For more information, contact at: The Ohio State University, 110F Animal Science Building, 2029 Fyffe Rd., Columbus, OH 43210; 614-292-6407; Fax: 614-292-2929; e-mail: irvin.3@osu.edu.
The primary objective of this experiment was to estimate both maternal and individual heterosis expressed by comparing the Yorkshire and Large White breeds and their crosses. A second objective of the experiment was, based on the estimates of heterosis, to conclude if the two breeds are the same or different. The study was conducted at the Western Branch of The Ohio State University's Ohio Agricultural Research and Development Center (OARDC), in South Charleston, Ohio.
Maternal heterosis was estimated by the performance of Yorkshire and Large White females and their crosses mated to Hampshire sires. Preweaning traits used in this analysis included birth weight, number of pigs born, number of pigs born alive, number of stillborn and mummified pigs, 21-day weight, weaning weight, and number of pigs weaned. Maternal heterosis was found (P < 0.05) for weaning weight (3.07%), average adjusted 21-day weight (3.08%), and average birth weight (3.26%).
Individual heterosis was estimated by the performance of individual pigs and average pig performance in pens. Postweaning traits used in this analysis were gain and feed efficiency. Individual heterosis estimates demonstrated crossbred pigs to have gained faster during the testing period than purebred pigs. Crossbred pigs had significant (P < 0.05) levels of heterosis for gain during the second half of the testing period (4.47%) and for gain during the entire testing period (3.87%).
Yorkshire and Large White swine were developed almost 250 years ago in England. Soon after this development, some of the Yorkshires were brought to the United States, while the Large Whites developed in England, according to Plager (1975). Since then, each breed has been subjected to different selection criteria and breeding schemes. Until recently, the two breeds had remained isolated from each other.
With the recent importation of the Large White breed to the United States, it is important to establish whether the two breeds are the same, or if through forces such as mutation, migration, and selection, the breeds have evolved into two distinct breeds. By mating the two breeds and subsequently crossing the breeds, a measure of heterosis can be obtained.
In 1992, a foundation herd was established at the Western Branch of The Ohio State University's Ohio Agricultural Research and Development Center (OARDC), South Charleston, Ohio, consisting of Large White (LW) and Yorkshire (Y) swine. The Large White pigs consisted of 40 gilts and eight boars from two different breeding farms. The Yorkshire pigs consisted of 40 gilts and eight boars from four different breeding farms. Both the Large White and Yorkshire pigs obtained for the experiment were bred to represent the four possible combinations (Y x Y, Y x LW, LW x Y, LW x LW) of females in the study.
Litter and sow data from 219 sows were utilized in analyses to estimate sow productivity and maternal heterosis for various traits. Litter data, consisting of information on 2,141 individual pigs, were used to estimate both the performance of the litter and the performance of the respective sow.
Large White and Yorkshire females (four different combinations) were mated to Hampshire boars. The Hampshire swine consisted of 12 boars from four different breeding farms. This was done to reduce breed of sire by breed of dam effects in the analysis. Data were collected over six farrowing seasons from November 1994 to November 1996.
Data were obtained from 566 pigs from the matings among Yorkshire and Large White swine. All pigs used in the analysis were purebred Yorkshires, purebred Large Whites, or resulted from reciprocal crosses of these two breeds. The foundation herds of Yorkshire and Large White pigs were mated to each other and subsequently crossed to produce the four possible combinations of offspring (Y x Y, Y x LW, LW x Y, LW x LW). These pigs were farrowed from March of 1993 through February of 1995. Feed efficiency and growth rates were calculated from the performance of the pigs and the pen during their maturation.
The results of this experiment are in agreement with Kacirek (1997) and Dimsoski (1996) who concluded, through the use of microsatellite loci variation, that Yorkshire and Large White swine represent different populations.
Maternal heterosis was expressed for several traits. In the analysis of differences among the two breeds and the reciprocal crosses, there were differences among breed of sire and breed of dam for 21-day weight and adjusted 21-day weight (P < 0.05). For both of these traits, the Y x LW pigs were the largest and the LW x Y pigs were the smallest. Significant differences existed between purebred Yorkshires and Y x LW pigs (P < 0.05) for these two traits. There were no significant differences between purebred Yorkshires and any other breed combination, while there were significant differences among Y x LW, LW x Y, and purebred Large White pigs (P < 0.05) for 21-day weight and adjusted 21-day weight. Yorkshire x Large White maternal genotype yielded significantly higher average birth weights than purebred Yorkshires and purebred Large Whites (P < 0.05). These results suggest the Y x LW maternal genotype produced the heaviest pigs from birth through 21 days of age and are in agreement with Schneider et al. (1982) and Gaugler et al. (1984). Maternal heterosis estimates associated with 21-day weight and adjusted 21-day weight were not significant (P > 0.05). Crossbred sows had 1.86% (P = 0.09) and 2.03% (P = 0.08) heterosis respectively, for these two traits. These results suggest that crossbred dams with Y x LW maternal genotype may provide a better environment for nursing their young.
Breed of dam differences existed for weaning weight, average adjusted 21-day weight, and average birth weight (P < 0.05). For all three of these traits, the breed combination Y x LW produced the largest values, whereas the purebred Yorkshires produced the smallest values. Heterosis levels for these traits were 3.0% (P = 0.04), 3.08% (P = 0.04) and 3.26% (P = 0.002), respectively. These results suggest the maternal genotype Y x LW would be the preferred breed combination to utilize in a breeding program. This genotype appears to produce the largest pigs from birth through 21 days. Although there were no significant differences among breed of sire and breed of dam for birth weight, heterosis for this trait was 2.63% (P < 0.05). Differences for weaning weight and average adjusted 21-day weight may also be explained by the crossbred maternal genotype providing a more suitable environment for the young pigs. A higher average litter birth weight was a result of crossbred dams having heavier individual pigs at birth, rather than having more pigs per litter. Johnson and Omtvedt (1973) reported similar findings. Although not significant, it appears that the purebred dams had more success in keeping their offspring alive from birth to 21 days of age (P = 0.82). One would expect that sows producing heavier pigs at 21 days provided a more suitable environment for their young. These results are somewhat different than expected because of the crossbred pigs having higher 21-day weights. However, crossbred dams still weaned more pigs than purebred dams (P = 0.38). Crossbred dams exhibited 10.63% heterosis for number of stillborn pigs (P = 0.70).
Significant differences existed between sexes. Barrows and gilts differed for 21- day weight, adjusted 21-day weight, and birth weight (P < 0.05), and for weaning weight (P < 0.10), with barrows being heavier for each trait. Parity differences existed for number born (P < 0.10), average birth weight, and average adjusted 21-day weight (P < 0.01). Significant differences existed among all four parities for average birth weight (P < 0.05). Parity two outperformed parity one for average adjusted 21-day weight (P < 0.05). Parity did not affect number of stillborn or mummified pigs. Parity effects were also not significant for survival rate or number of pigs alive at 21 days (P > 0.10).
Several traits were affected by farrowing season. Farrowing season was highly significant for 21-day weight, adjusted 21-day weight, birth weight, weaning weight, number born, number at 21 days, average adjusted 21-day weight, survival from birth to 21 days, and average birth weight (P < 0.01). These results are in general agreement with Buchanan and Johnson (1984). Farrowing season was not significant for number born alive (P = 0.06) or survival from birth to weaning (P = 0.08). Fall-born litters had significantly lower birth weights (P < 0.05) than spring- born litters and the litter farrowed during the summer. On average, spring-born litters had more pigs at 21 days and higher average birth weights than did litters farrowed in the fall. Spring-born litters having higher average birth weights than fall-born litters may be a result of extreme environmental conditions during the summer season.
There were no significant differences among breed combinations for gain during the first half of the testing period. Differences between purebred Yorkshires and the LW x Y breed combination existed for gain in the second half of the testing period and for total gain (P < 0.05). The LW x Y breed combination gained 90 lb. (40.96 kg.) versus purebred Yorkshires which gained 83 lb. (37.70 kg.) (P = 0.005) for the second half of the testing period. The LW x Y breed combination also gained the most for the entire testing period, whereas purebred Yorkshire pigs gained the least among the four breed combinations (P < 0.01). Purebred Yorkshires had the highest average backfat measurement and purebred Large Whites had the lowest average backfat measurement (P < 0.05). Breed of dam differences were significant for average backfat measurements (P < 0.01). Breed of sire was an important effect for Y x LW and purebred Yorkshire pigs (P < 0.10) for gain during the second half of the testing period. Heterosis was significant for gain during the second half of the testing period (P = 0.05) and total gain (P = 0.03) with values of 4.47 and 3.87%, respectively. These results are in agreement with Toelle and Robison (1983), Johnson (1981), and Sellier (1976), indicating that crossbred pigs gain more rapidly than purebred pigs.
Significant differences existed among farrowing seasons for all traits analyzed. Spring-born litters out-gained fall-born litters (P < 0.01) which was in agreement with Johnson et al. (1973). Winter-born litters also out gained summer-born litters (P < 0.01). There were significant differences between summer-born and winter-born litters (P < 0.05), and among spring-born, summer-born and winter-born litters (P < 0.05) for total gain. Since there was only one summer farrowing season, the environmental conditions of that summer may have caused the pigs slower growth rates during that farrowing season.
Significant differences existed among sexes for all traits. Boars gained the most, followed by barrows, and then by gilts over the entire testing period (P < 0.01). The average backfat measurements demonstrated boars to be leanest, followed by gilts, and then barrows (P < 0.01). The results for gain and backfat measurements were expected.
For the first half of the testing period, purebred Yorkshire pigs consumed more feed than purebred Large White pigs (P < 0.01). For the rest of the testing period, the LW x Y crossbred pigs consumed more feed than the purebred Large White pigs (P < 0.01). Feed efficiency only differed for the first half of the testing period between purebred Large White and Y x LW pigs (P < 0.10) favoring the Y x LW genotype. An unexpected result was that dam breed effects were highly significant for all feed and gain traits (P < 0.01). These results are in disagreement with McLaren et al. (1987a). Breed of sire effects are normally associated with gain and feed efficiency. However, none of the heterosis estimates was significant for these traits. Although not significant, crossbred pigs exhibited negative heterosis (P = 0.08) for feed efficiency for the first half of the testing period. While not significant, crossbred pigs had negative heterosis levels of 6.33 and 1.94% for feed efficiency during the second half of the testing period and overall, respectively. Feed efficiency and gain are moderately heritable traits. Fahmy and Bernard (1972) suggested that relatively low heterosis levels for feed efficiency is greatly affected by additive gene action. Crossbred pigs exhibiting negative heterosis for feed efficiency does not necessarily suggest that these two breeds of swine are the same.
Sex effects were significant for several traits. Gilts consumed more feed than boars during the second half of the testing period (P < 0.10). Boars were significantly better feed converters than barrows during the first half of the testing period and over the entire testing period (P < 0.01). Gilts were also better feed converters than barrows (P < 0.01) over the entire testing period. These results are in agreement with McLaren et al. (1987a).
Farrowing season affected some of the traits. Spring-born litters on average consumed more feed and gained more weight than did summer-born litters during the second half of the testing period (P < 0.01). Overall, spring-born litters converted feed more efficiently than did litters born in the summer (P < 0.01).
The results of this experiment suggest that while similar, Yorkshire and Large White swine are in fact different breeds. Breed differences for many of the traits analyzed were not significant; however, several differences between the two breeds existed. It is evident that these two breeds, through forces such as mutation, migration, random drift, and selection, have diverged into two distinct populations.
Buchanan, D. S. and R. K. Johnson. 1984. Reproductive performance for four breeds of swine: crossbred females and purebred and crossbred boars. J. Anim. Sci. 59:948.
Dimsoski, P. I. 1996. Variation at microsatellite loci in Yorkshire and Large White swine. Ph.D. Dissertation. The Ohio State University, Columbus, OH.
Fahmy, M. H. and C. S. Bernard. 1972. Heterosis in crosses between three lines of Yorkshire swine selected for feed efficiency and carcass quality. Can. J. Anim. Sci 52:444.
Gaugler, H. R., D. S. Buchanan, R. L. Hintz, and R. K. Johnson. 1984. Sow productivity comparisons for four breeds of swine: purebred and crossbred litters. J. Anim. Sci. 59:941.
Johnson, R. K. 1981. Crossbreeding in swine: experimental results. J. Anim. Sci. 52:906.
Johnson, R. K. and I. T. Omtvedt. 1973. Evaluation of purebreds and two-breed crosses in swine: reproductive performance. J. Anim. Sci. 37:1279.
Johnson, R. K., I. T. Omtvedt, and L. E. Walters. 1973. Evaluation of purebreds and two-breed crosses in swine: feedlot performance and carcass merit. J. Anim. Sci. 37:18
Kacirek, S. L. 1997. Variation at microsatellite loci in the Large White, Yorkshire, and Hampshire breeds of pigs. M.S. Thesis. The Ohio State University, Columbus, OH.
McLaren, D. G., D. S. Buchanan, and R. K. Johnson. 1987. Growth performance for four breeds of swine: crossbred females and purebred and crossbred boars. J. Anim. Sci. 64:99.
Plager, W. L. 1975. History of Yorkshires and American Yorkshire Club. Ch. 1. American Yorkshire Club, Lafayette, Indiana.
SAS. 1995. ASAS User's Guide: statistics® and statistical analysis system. 6.11 Edition. SAS Institute, Inc., Raleigh, NC.
Schneider, J. F., L. L. Christian, and D. L. Kuhlers. 1982a. Crossbreeding in swine: genetic effects on litter performance. J Anim Sci. 54:739.
Sellier, P. 1976. The basis of crossbreeding in pigs: a review. Livestock Prod. Sci. 3:203.
Toelle, V. D. and O. W. Robison. 1983. Breed prenatal, breed postnatal, and heterosis effects for postweaning traits in swine. J. Anim. Sci. 57:313.