S.M. Neal+ and K.M. Irvin*
+Agricultural Technical Institute and
*Department of Animal Sciences
Results from this experiment indicated some evidence of heterosis in the cross between Large White and Yorkshire breeds of swine. Average pig birth weights had higher heterosis than literature values. Other traits did not show statistically significant differences between pure and cross averages. The Large White x Yorkshire crossbred females tended to have superior litter sizes farrowed and weaned. Further study may provide more evidence of Yorkshire and Large White breed complimentarity and perhaps specific combining ability. More data are needed before conclusive statements can be made related to evidence of individual and maternal heterosis in Yorkshire and Large White breed crosses.
Many of the estimates of heterosis currently used resulted from experiments that were conducted in the early 1980s. There has been some interest in the possibility of heterosis within the Yorkshire breed since Large Whites have been imported and are being registered in the same breed registry as Yorkshires. Yorkshires in the United States originated from the Large White breed from England. For the most part, the lines have been separated until recently. As a result of importation of Large Whites, there has been speculation regarding the potential superiority of crosses of the two breeds relative to purebred Yorkshires and Large Whites. If, in fact, heterosis exists in the cross of the two breeds, estimates of breeding values and EPDs may be biased as presented in sire summaries. The primary objective of this work was (and is in the current experiment) to determine the presence and magnitude of individual and maternal heterosis in crosses of Yorkshire and Large White breeds of swine.
Experiment 1. This project was conducted at The Ohio State University Agricultural Technical Institute, Wooster. Foundation animals for the pure breeds included 10 unrelated boars (five of each breed) and 20 unrelated gilts (10 of each breed). These animals were mated randomly within breed to multiply the population size and resulted in 45 pure Large White females and 49 pure Yorkshire females. Both the original and subsequent females were mated randomly (avoiding inbreeding) to the original set of boars to produce progeny females of each of the following types: pure Yorkshire (Y), pure Large White (LW), Y sire X LW dam (YLW), and LW sire X Y dam (LWY). A portion of these four groups of females and contemporary gilts were evaluated to estimate individual heterosis. These four sets of gilts then were mated randomly to six Duroc sires to produce 205 litters for evaluation and estimation of maternal heterosis.
Breeding animals were fed a standard corn-soybean meal gestation and lactation diet. Pigs were weighed at birth and 21 days. Creep feed was not provided, and cross-fostering was not done. Traits for which data were collected for estimation of individual heterosis included average daily gain (ADG) from approximately 36 kg to 104 kg, carcass 10th rib fat depth (BF), carcass loin depth (LD), and estimated carcass percentage lean cuts (LC). Carcass information was obtained from data reported from a commercial slaughter facility using a Fat'o'meat'er probe.
Traits measured on litters for estimating maternal heterosis in sows included the number of piglets born alive (NBA); stillborns, mummies, and overlays, numbers weaned (NWEANED); average pig birth weights (BW); and average pig 21-day weaning weights (WW).
Analysis for the estimation of individual heterosis utilized a model including the effects of season, breed of sire, breed of dam, and the breed of sire by breed of dam interaction and appropriate covariates (ADG adjusted to a constant on-test weight; BF, LD, and LC adjusted to a constant end-test weight). Least squares means were estimated, and a linear contrast was used to compare the average of the pure breeds with the average of the reciprocal crosses. Litter data for the estimation of maternal heterosis were analyzed using a model which included the effects of season, parity, breed of dam and sire of the litter (random effect). and appropriate covariates (NWEANED adjusted for age at weaning; WW adjusted for BW, and age at weaning).
Individual Heterosis Data. Table 1 shows the number of observations and means for breed groups and crosses. Large White and YLW gilts gained faster than Yorkshire gilts (6% and 4.5%, respectively; P < 0.05). However, there was no difference between the pure and cross average. Large White and LWY were significantly leaner at the 10th rib than pure Yorkshire (13.6% and 11.9%, respectively; P < 0.05). Although not significant, crossbred pigs tended to be leaner (1.29%) than purebred pigs. Pure Yorkshire gilts had less loin depth than all other types (P < 0.05), and the reciprocal cross types averaged 0.545 mm more loin depth than the average of the purebreds (P > 0.05). Results of fat and loin depth translated into similar results for percentage lean cuts. Yorkshire gilts had 2.3% less percent lean cuts than LW (P < 0.05) and 1.8% less percent lean cuts than LWY (P < 0.05).
Maternal Heterosis Data. Results of the litter data are shown in Table 2. Although not significant, Y and LWY had about 0.5 more pigs born alive than both LW and YLW dams. This trend continued for the number of pigs weaned for LWY dams only. Perhaps in a larger experiment, the superiority of the LWY cross for number weaned would be validated (LWY vs Y, P < 0.16). Reciprocal cross dams averaged 0.4 more pigs per litter weaned (P > 0.05). Average piglet birth weight was lowest for Y compared to all other types (P < 0.05). The linear contrast of purebred average compared to crossbred average pig birth weight was significant, indicating that maternal heterosis was present for average pig birth weight (6.18%, P < 0.05). No significant differences were observed in average pig weaning weights.
Experiment 2. A follow-up experiment with a similar design is currently underway at the Western Branch of the Ohio Agricultural Research and Development Center at South Charleston. This experiment was initiated with the purchase of 80 gilts and 16 boars, half of which were pure Yorkshire and the other half pure Large White. Four breed types were produced and include pure Yorkshire, pure Large White, Yorkshire X Large White, and Large White X Yorkshire. Individual heterosis data were obtained on rate of growth from 45 to 110 kg (ADG), average backfat thickness at 110 kg (AVEBF), and pen measurements of feed conversion (F/G).
| Table 1. Least squares means for growth and carcass data in gilts for breeds and breed crosses. | |||||||||||
| ADG | BF1 | LD1 | LC | ||||||||
| Breed | n | Mean | n | Mean | n | Mean | n | Mean | |||
| Y X Y | 88 | 0.67 ± 0.012a | 59 | 21.84 ± 0.71a | 59 | 42.17 ± 1.05a | 59 | 56.6 ± 0.28a | |||
| LW X LW | 37 | 0.71 ± 0.015b | 28 | 19.23 ± 0.76b | 28 | 46.21 ± 1.11b | 28 | 57.9 ± 0.30b | |||
| Y X LW | 60 | 0.70 ± 0.015b | 46 | 21.03 ± 0.81 | 46 | 44.97 ± 1.21b | 46 | 57.1 ± 0.33 | |||
| LW X Y | 40 | 0.68 ± 0.017 | 38 | 19.51 ± 0.81b | 38 | 44.50 ± 1.21b | 38 | 57.6 ± 0.33b | |||
| Cross minus pure | 0.0 | - 0.265 | 0.545 | 0.10 | |||||||
| % Individual heterosis | 0.0 | - 1.29 | 1.23 | 0.17 | |||||||
| 1 Measured in mm.
ab Means in the same column with different letters differ (P < 0.05). | |||||||||||
| Table 2. Least squares means for litter traits for breeds and breed crosses. | |||||||||||
| NBA | NWEANED | BW1 | WW1 | ||||||||
| Breed | n | Mean | n | Mean | n | Mean | n | Mean | |||
| Y X Y | 49 | 10.26 | 48 | 8.93 ± 0.60 | 49 | 1.32 ± 0.062a | 48 | 5.10 ± 0.22 | |||
| LW X LW | 48 | 9.74 | 48 | 9.09 ± 0.59 | 48 | 1.43 ± 0.061b | 48 | 5.08 ± 0.21 | |||
| Y X LW | 51 | 9.72 | 50 | 9.14 ± 0.58 | 51 | 1.50 ± 0.059b | 50 | 5.02 ± 0.21 | |||
| LW X Y | 57 | 10.25 | 55 | 9.68 ± 0.57 | 57 | 1.42 ± 0.058b | 55 | 4.89 ± 0.20 | |||
| Cross minus pure | - 0.015 | 0.40 | 0.085 | -0.135 | |||||||
| % Maternal heterosis | - 0.15 | 4.44 | 6.18c | -2.65 | |||||||
| 1 Measured in kg.
ab Means in the same column with different letters differ (P < 0.05). c Linear contrast of pure vs cross (P < 0.05). | |||||||||||
Maternal heterosis data were obtained on each of the breed types as dams. Fourteen Hampshire boars were used to produce offspring from the four sets of dam breed combinations. Traits measured included number of pigs born alive (NBA), number weaned at 21 days (NWEANED), average pig within a litter birth weight (BW), and average pig within litter 21-day weight (WW). These data sets were analyzed in a similar manner to that described above.
Individual Heterosis Data - Preliminary Results. Table 3 shows preliminary results of individual heterosis from the second experiment. Heterosis values for ADG and BF were 6% and 6.9%, respectively. Neither of these numbers was statistically significant; however, both were greater than the estimates from the first experiment. It should be noted that in the first experiment, only gilts were included in the individual heterosis phase, which partially may explain the differences in results. Pure Large White pigs were significantly leaner (3.58 mm) than LWY pigs. The YLW cross was represented by relatively few numbers compared to the other groups. Individual heterosis for F/G was estimated to be near zero. Pure Large White pigs tended to be the most efficient in converting feed to gain. For all of these traits, more data will be needed to draw reliable conclusions. Comparison of the two experiments does not confirm any of the trends seen in the data.
Maternal Heterosis Data - Preliminary Results. Maternal heterosis data for the second experiment was somewhat in agreement with the data collected at Wooster (Table 4). Even though the number of litters represented in this data set was relatively small, similar trends were apparent.
LWY sows had the largest NBA and were significantly greater that pure LW. LWY sows also had the greatest number weaned in both data sets; however, the difference was greater in the first experiment. Heterosis for NBA was estimated at 8.62 (P > 0.05) but was near zero in Experiment 1. Maternal heterosis for NWEANED was near zero. However, as seen in the first experiment, pig birth weights were significantly greater in crossbred pigs (9.56%, P < 0.05). YLW sows had heavier pigs born on average than pure Y dams (P < 0.05), and a similar trend was seen in results of Experiment 1.
| Table 3. Least squares means for performance data.1 | ||||||||
| ADG2 | BF3 | F/G | ||||||
| Breed | n | Mean | n | Mean | n | Mean | ||
| Y X Y | 96 | 0.73 ± 0.04 | 96 | 18.78 ± 1.00ab | 8 | 3.41 ± 0.11 | ||
| LW X LW | 68 | 0.77 ± 0.06 | 68 | 16.62 ± 1.53a | 6 | 3.07 ± 0.14 | ||
| Y X LW | 36 | 0.84 ± 0.08 | 36 | 17.63 ± 1.97ab | 4 | 3.33 ± 0.15 | ||
| LW X Y | 131 | 0.75 ± 0.03 | 131 | 20.20 ± 0.65b | 10 | 3.21 ± 0.10 | ||
| Cross minus pure | 0.045 | 1.215 | 0.03 | |||||
| % Individual heterosis | 6.0 | 6.9 | 0.93 | |||||
| 1 Number of pigs is 331.
2 Measured in kg. 3 Measured in mm. ab Means in the same column with different letters differ (P < 0.05). | ||||||||
| Table 4. Least squares means for litter traits for breeds and breed crosses.1 | |||||||||||
| NBA | NWEANED | BW2 | WW2 | ||||||||
| Breed | n | Mean | n | Mean | n | Mean | n | Mean | |||
| Y X Y | 22 | 10.35 ± 0.72ab | 21 | 8.71 ± 0.54 | 22 | 1.32 ± 0.05b | 21 | 6.45 ± 0.23ab | |||
| LW X LW | 23 | 9.38 ± 0.80a | 21 | 8.23 ± 0.59 | 23 | 1.40 ± 0.06ab | 21 | 6.55 ± 0.26ab | |||
| Y X LW | 8 | 10.38 ± 1.08ab | 8 | 8.01 ± 0.79 | 8 | 1.56 ± 0.08a | 8 | 7.12 ± 0.35b | |||
| LW X Y | 21 | 11.05 ± 0.71b | 21 | 8.76 ± 0.52 | 21 | 1.42 ± 0.05ab | 21 | 6.28 ± 0.23a | |||
| Cross minus pure | 0.85 | -0.085 | 0.13 | 0.20 | |||||||
| % Maternal heterosis | 8.62 | -1.00 | 9.56c | 3.08 | |||||||
| 1 Number of litters is 74.
2 Measured in kg. abMeans in the same column with different letters differ (P < 0.05 ). c Linear contrast of pure vs cross (P < 0.05). | |||||||||||
The authors wish to express appreciation for partial sponsorship of this project provided by the Ohio Pork Producers Council and the American Yorkshire Club. The authors also gratefully acknowledge the work of Terry Meek, ATI swine unit manager, and David Owens and other OARDC Western Branch employees.