Search
Author
Title
Vol.
Issue
Year
1st Page

# Journal of Animal Science - Animal Products Comparison of variability in pork carcass composition and quality between barrows and gilts123

1. Vol. 94 No. 10, p. 4415-4426

Accepted: Aug 04, 2016
Published: October 7, 2016

4 Corresponding author(s):
View
Share

doi:10.2527/jas.2016-0702
1. M. F. Overholt*,
2. E. K. Arkfeld*,
3. D. A. Mohrhauser,
4. D. A. King,
5. T. L. Wheeler,
6. A. C. Dilger*,
7. S. D. Shackelford and
8. D. D. Boler 4*
1. * Department of Animal Sciences, University of Illinois, Urbana – Champaign 61801
Smithfield Foods, Denison, IA 51442
USDA, ARS, U.S. Meat Animal Research Center, Clay Center, NE 68933

## Abstract

Pigs (n = 8,042) raised in 8 different barns representing 2 seasons (cold and hot) and 2 production focuses (lean growth and meat quality) were used to characterize variability of carcass composition and quality traits between barrows and gilts. Data were collected on 7,684 pigs at the abattoir. Carcass characteristics, subjective loin quality, and fresh ham face color (muscles) were measured on a targeted 100% of carcasses. Fresh belly characteristics, boneless loin weight, instrumental loin color, and ultimate loin pH measurements were collected from 50% of the carcasses each slaughter day. Adipose tissue iodine value (IV), 30-min loin pH, LM slice shear force, and fresh ham muscle characteristic measurements were recorded on 10% of carcasses each slaughter day. Data were analyzed using the MIXED procedure of SAS as a 1-way ANOVA in a randomized complete block design with 2 levels (barrows and gilts). Barn (block), marketing group, production focus, and season were random variables. A 2-variance model was fit using the REPEATED statement of the MIXED procedure, grouped by sex for analysis of least squares means. Homogeneity of variance was tested on raw data using Levene’s test of the GLM procedure. Hot carcass weight of pigs (94.6 kg) in this study was similar to U.S. industry average HCW (93.1 kg). Therefore, these data are representative of typical U.S. pork carcasses. There was no difference (P ≥ 0.09) in variability of HCW or loin depth between barrow and gilt carcasses. Back fat depth and estimated carcass lean were more variable (P ≤ 0.0001) and IV was less variable (P = 0.05) in carcasses from barrows than in carcasses from gilts. Fresh belly weight and thickness were more variable (P ≤ 0.01) for bellies of barrows than bellies of gilts, but there was no difference in variability for belly length, width, or flop distance (P ≥ 0.06). Fresh loin subjective color was less variable (P < 0.01) and subjective marbling was more variable (P < 0.0001) in loins from barrows than in those from gilts, but there were no differences (P ≥ 0.08) in variability for any other loin traits or fresh ham traits. Overall, traits associated with carcass fatness, including back fat depth, belly thickness, and marbling, but not IV, were more variable in carcasses from barrows than in carcasses from gilts, whereas minimal differences in variability existed between carcasses of barrows and carcasses of gilts for traits associated with carcass muscling and lean quality.

## INTRODUCTION

Sex has a pronounced effect on the composition and quality of pork carcasses. Because barrows grew faster and reached physiological maturity earlier than gilts (Lee et al., 2013), barrows had greater HCW than gilts whether slaughtered at equal days on feed (Lee et al., 2013) or at equal BW (Davis et al., 2015). Barrow carcasses were also fatter than gilts when slaughtered at equal BW (Bereskin and Davey, 1978; Smit et al., 2014). Because barrows are typically fatter than gilts, barrow carcasses yielded a lesser percentage of lean cuts (Martel et al., 1988; Boler et al., 2014), and fresh bellies (Kyle et al., 2014), loins (Unruh et al., 1996), and hams (Uttaro et al., 1993) were fatter in barrows than in gilts.

Variation in pork carcass weight, composition, and quality characteristics exists (Cannon et al., 1996; Stetzer and McKeith, 2003; Klinkner, 2013). Carcass quality nonconformities cost the U.S. pork industry over $8.00 per pig (Stetzer and McKeith, 2003). Currently, producers attempt to limit variability in weight by marketing pigs in groups of similar live weights (Gerlemann et al., 2014). But even when 2 populations (e.g., barrows vs. gilts) have equal means for a trait, variability of that trait may differ between populations. Differences in variability would result in a greater proportion of carcasses from a population deviating from desired specifications. Understanding differences in variability between barrow and gilt carcasses will allow implementation of strategies to improve uniformity of pork products and maximize revenue from each carcass. The hypothesis was that even when multiple marketing groups were used to minimize variation in HCW, variation in carcass composition and meat quality traits would exist and the amount of variability would be different in barrows and gilts. Therefore, the objective of the study was to characterize variability differences in carcass composition and quality between barrows and gilts. ## MATERIAL AND METHODS Meat samples were obtained from a federally inspected abattoir; therefore, no Institutional Animal Care and Use Committee approval was necessary. Pigs (8,042 total) raised in 8 different barns representing 2 seasons (cold and hot) and 2 production focuses (lean growth and meat quality) were used in this study. Producers of barns A, C, E, and G had production programs focused on lean growth of pigs, and barns B, D, F, and H had production programs focused on meat quality. Pigs in barns A, B, C, and D, were slaughtered over a 7-wk period between February 11 and March 26 (cold season), and pigs in barns E, F, G, and H were slaughtered over a 7-wk period between July 29 and September 10 (hot season). All pigs were slaughtered at a single federally inspected commercial processing facility. From each barn, 3 marketing groups were selected based on weight as determined by site-specific protocols. Pigs were slaughtered over a 7-wk time period where the first marketing group from barns A, B, E, and F were marketed during the first week. The first marketing group from barns C, D, G, and H and the second marketing group from barns A, B, E, and F were marketed during the third week. The second marketing group from barns C, D, G, and H and the third marketing group from barns A, B, E, and F were marketed during the fifth week. The third marketing group from barns C, D, G, and H were marketed on the seventh week. ### Abattoir Data Collection Lairage procedures followed normal abattoir operating procedures. Pigs of the quality production focus were held in lairage overnight at the abattoir (approximately 13 h), and pigs from the lean growth program arrived at the abattoir approximately 7 h prior to slaughter. These differences were routine for those types of pigs at that plant because of the need to harvest pigs whose meat qualifies for a particular program at the same time. Pigs were rendered insensible by carbon dioxide stunning and terminated via exsanguination. Immediately after evisceration, carcasses were assigned a sequence identification number on the shoulder and ham and each pig’s respective lot tattoo was recorded. Carcass characteristics were evaluated on 7,684 pigs. Approximately 31 min postmortem, loin pH was collected at approximately the 10th rib on every 10th carcass (targeted 10% of the population) on each slaughter day. These carcasses were noted as the select pigs and used for in-depth quality analyses. Thirty-one minute pH data were collected with a REED SD-230 m (Reed Instruments, Wilmington, NC) fitted with a PHE-2385 glass combo electrode (OMEGA Engineering, Stamford, CT) during the first and second weeks of the cold season and a FC 200 B series electrode (Hanna Instruments, Woonsocket, RI) was used for all remaining weeks of the study. Carcasses of both cold and hot seasons were evaluated for HCW, back fat (BF) depth, and loin depth using a Fat-O-Meater probe (SFK Technology A/S, Herlev, Denmark), and estimated carcass lean percentage was calculated using an abattoir proprietary equation. Carcasses were blast chilled for approximately 100 min. As carcasses exited the chiller, adipose tissue cores, approximately 3.81 cm in diameter and consisting of all 3 adipose layers, were collected from the clear plate (adipose tissue located over the scapula and cervical vertebra) near the dorsal midline of the left side of every carcass. Adipose tissue cores from the selected 10% of carcasses from each slaughter day were used to calculate iodine value (IV) using gas chromatography. Fatty acid methyl esters (FAME) were extracted from adipose samples using the American Oil Chemists’ Society official method Ce 2-66 (AOCS, 1998). The resulting FAME were analyzed using the procedures of Arkfeld et al. (2015). Fatty acid methyl esters were normalized such that the area of each peak was represented as the percentage of the total area. Iodine values were calculated using fatty acid profile data (% FAME of total FAME) with the following American Oil Chemists’ Society (1998) equation: IV = (C16:1) × 0.95 + (C18:1) × 0.86 + (C18:2) × 1.732 + (C18:3) × 2.616 + (C20:1) × 0.785 + (C22:1) × 0.723, in which values in parentheses represent grams FAME/100 g total FAME. While carcasses were in the equilibration cooler, vertebrae of all loins and bellies from odd numbered carcasses were labeled with sequence numbers that matched the ham and shoulder (targeted 50% of population each slaughter day). Approximately 22 h postmortem, carcasses were fabricated into primal pieces. Bellies (North American Meat Processors [NAMP] number 408; NAMP, 2007) and hams (modified NAMP number 401; NAMP, 2007) were collected and placed into combos for further analyses that same day. Loins were fabricated into boneless Canadian back loins (NAMP number 414; NAMP, 2007). Fresh muscle color (1 to 6, subjective scale), marbling (1 to 10, subjective scale), and firmness (1 to 5, subjective scale) were evaluated using National Pork Producers Council standards on the loin boning and trimming line at the time of cutting by an industry professional with over 10 yr of pork quality evaluation experience (NPPC, 1991, 1999). ### Bellies Skin-on bellies (NAMP number 408) were weighed and measurements of length and width (distance from the line of rib removal to the dorsal edge of the belly; measured at approximately the center of the belly from the anterior to posterior end) were recorded on approximately 50% of the bellies from each slaughter day. Belly depth (thickness) was recorded at 25, 50, and 75% the length of the belly along the longitudinal axis, beginning at the anterior end. The mean of the 3 belly depth measurements was reported. A subjective flop score of 0.5 to 5.0 in 0.5-unit increments was assigned to each belly. Subjective flop scores were anchored such that a score of 0.5 was characterized as an approximate flop distance of less than 5 cm, a score of 1 was 5.1 to 10 cm, a score of 2 was 10.1 to 15 cm, a score of 3 was 15.1 to 20 cm, a score of 4 was 20.1 to 25 cm, and a score of 5 was greater than 25 cm. Flop scores were evaluated by plant personnel with at least 5 yr of experience evaluating bellies. ### Loins Approximately 50% of the entire population of loins was selected from each slaughter day for further quality analyses and boneless loin weight. Instrumental L*, a*, and b* color evaluations were conducted on the ventral side at approximately 25 and 75% the length of the loin using a Hunter Miniscan XE Plus colorimeter (HunterLab, Reston, VA) with a D65 light source, 10° observer, and 25-mm port. Ultimate pH was recorded using a pH meter at the approximate midpoint of the ventral side of the loin. For data collected on cold-season wk-1 loins, a REED SD-230 m (Reed Instruments, Wilmington, NC) fitted with a PHE-2385 glass combo electrode (OMEGA Engineering) was used. Ultimate pH data collected on loins from cold-season pigs of wk 2 through 4, and all hot season pigs were evaluated using a HI 98160 Microprocessor Logging pH/ORP Meter (Hanna Instruments). ### Select Loins Loins from the selected 10% of the population of each slaughter day were vacuum-packaged and transported (1°C) to the U.S. Meat Animal Research Center (Clay Center, NE). Within 58 h of carcass cutting, loins arrived at the U.S. Meat Animal Research Center. Loins were immediately placed on carts in a single layer and ventral side up and aged (1°C). Loins were weighed (tared for vacuum-packaging bag) to record initial loin weight. At 20 d postmortem, loins were removed from their packaging and weighed to determine aged weight, and purge loss was calculated: [(initial weight, kg − aged weight, kg)/initial weight, kg) × 100]. Loins were then prepared for slicing with a Grasselli NSL 400 portion meat slicer (Grasselli SPA, Albinea, Italy). The posterior end of the loin (approximately 4 cm long) was removed by a straight cut perpendicular to the length of the loin at a point 5 cm posterior to the anterior tip of gluteus accessories. The anterior end of the loin was removed by a second cut made 396 mm anterior to the first cut, leaving a 396-mm-long center-cut loin section that fits the width of the Grasselli NSL 400 portion meat slicer. This approach maximized yield of chops, with the highest proportion of their mass/cross-sectional area comprising LM and excluded chops with a high proportion of their mass/cross-sectional area comprising other muscles (spinalis dorsi, multifidus dorsi, gluteus medius, and gluteus accessorius). Additionally, this approach standardized anatomical location of chop assignment across loins. Chops 5 and 6, which correspond approximately to the 11th rib region of the loin, were used for determination of slice shear force (SSF). Immediately after cutting, fresh (never frozen) chops were weighed to record initial weight. The following day (21 d postmortem), chops were cooked using a belt grill (Magigrill, model TBG-60; MagiKitch’n Inc., Quakertown, PA) to a desired internal temperature of 71°C. Cooked chops were weighed and cooking loss was calculated: [(initial weight, g − cooked weight, g)/initial weight, g) × 100]. Slice shear force was measured using the procedures of Shackelford et al. (2004) on 2 chops. The 2 SSF values were then averaged. ### Hams After hams were removed from carcasses, instrumental L*, a*, and b* were recorded on the gluteus medius and gluteus profundis using a Minolta colorimeter (Konica Minolta CR-400 colorimeter; Minolta Camera Company, Osaka, Japan; D65 light source, 0° observer, and 8-mm aperture) on the cut face of each ham. ### Select Ham Fabrication and Quality Characteristics Hams were transported in combos via refrigerated truck to the University of Illinois Meat Science Laboratory (Urbana, IL) where they were fabricated into subprimal pieces 4 to 7 d postmortem, following procedures of Boler et al. (2011). Briefly, a modified NAMP number 401 (rectus abdominus attached) was weighed, trimmed similar to a NAMP number 402 (NAMP, 2007), and weighed to obtain trimmed weight. Hams were then separated into 5 pieces: inside ham (NAMP number 402F; NAMP, 2007), outside ham (NAMP number 402E; NAMP, 2007), knuckle (NAMP number 402H; NAMP, 2007), inner shank portion (gastrocnemius muscle), and lite butt. Weights were recorded on all pieces. Identification of the inside ham, outside ham, and knuckle was maintained; however, inner shank and lite butt identification was not retained as they were not needed for further analysis. Instrumental L*, a*, and b* values (Konica Minolta CR-400 colorimeter; Minolta Camera Company; D65 light source, 0° observer, and 8-mm aperture) and ultimate pH (MPI pH meter; Meat Probes Inc., Topeka, KS; 2-point calibration at pH 4 and 7) were collected on the semimembranosus muscle (blonde spot; medial side) of each ham. ### Statistical Analyses Data were analyzed using the MIXED procedure of SAS 9.4 (SAS Inst. Inc., Cary, NC) as a 1-way ANOVA in a randomized complete block design with 2 levels (barrows and gilts). Barn (served as block), marketing group, production focus, and season were random variables. Homogeneity of variance of the residuals was tested using Levene’s test of the GLM procedure, which revealed that the variance of residuals of several variables were unequal. Therefore, a 2 variance model was fit using the REPEATED statement of the MIXED procedure, grouped by sex. The effect of sex was considered significant at P ≤ 0.05. Variances for each treatment (barrows and gilts) were calculated using the MEANS procedure. Homogeneity of variance was tested on raw data using the Levene’s test of the GLM procedure. Variances were considered different at P ≤ 0.05. Differences in variances of selected traits are illustrated by box and whisker plots. The bottom line of the box indicates quartile 1 (25th percentile); the middle line, the median (50th percentile); and the top line, quartile 3 (75th percentile). Interquartile range (IQR) was calculated as quartile 3 − quartile 1. An upper fence was calculated as quartile 3 + (1.5 × IQR); a lower fence was calculated as quartile 1 − (1.5 × IQR). Whiskers represent the highest and lowest points within the upper and lower fences that are not outliers, respectively. Any observation value greater than the upper fence or less than the lower fence was considered an outlier. Although outliers are factored into variance calculations, they are not graphically illustrated on box and whisker plots for clarity purposes. ## RESULTS AND DISCUSSION ### Carcass Characteristics Mean differences between carcasses from barrows and gilts were similar to previous reports (Bereskin and Davey, 1978; Lee et al., 2013; Davis et al., 2015). Specifically, carcasses from barrows had 0.94 kg greater HCW (Table 1; P < 0.0001) and 2.20 mm greater BF depth (P < 0.0001) than carcasses from gilts but had 1.25 mm less (P < 0.0001) loin depth, resulting in a 1.60 unit lesser (P < 0.0001) estimated carcass lean percentage. Barrows have lesser energy requirements for the deposition of lean tissue than gilts (Schinckel, 1994; Thompson et al., 1996). This results in excess energy being deposited as fat, leading to barrows having heavier carcasses, whether slaughtered at equal days on feed (Lee et al., 2013) or at equal BW (Davis et al., 2015). Barrows are generally fatter and have a calculated lower percentage of carcass lean than gilts (Bereskin and Davey, 1978; Xu et al., 2010; Smit et al., 2014). In contrast, gilts have a greater propensity to deposit lean tissue, reflected in greater LM area, reduced BF depth, and greater estimated carcass lean percentage than barrows (Smit et al., 2014). Moreover, the propensity of gilts to deposit less fat than barrows leads to gilts having greater concentrations of unsaturated fatty acids in adipose tissue and, thus, greater IV (Averette-Gatlin et al., 2002; Lee et al., 2013). This was reflected in an IV from gilt adipose tissue that was 1.95 units greater (P < 0.0001) than that from adipose tissue of barrows in the present study. View Full Table | Close Full ViewTable 1. Effect of sex on carcass characteristics least squares means and variability  No. Least squares means Variance Item Barrows Gilts Barrows Gilts SEM P-value Barrows Gilts Levene’s P-value HCW, kg 3,315 4,252 95.10 94.16 2.29 <0.0001 90.68 84.95 0.09 LM depth, mm 2,989 3,920 67.21 68.46 4.01 <0.0001 72.78 71.32 0.60 Back fat, mm 2,989 3,920 16.84 14.64 1.67 <0.0001 15.82 12.86 <0.0001 Estimated carcass lean, % 2,989 3,920 56.65 58.25 1.48 <0.0001 7.41 6.27 <0.0001 Iodine Value1 381 465 74.68 76.63 0.97 <0.0001 11.03 13.10 0.05 1Iodine value = [(C16:1) × 0.95 + (C18:1) × 0.86 + (C18:2) × 1.732 + (C18:3) × 2.616 + (C20:1) × 0.785 + (C22:1) × 0.723], in which values in pa¬rentheses represents fatty acid methyl ester (FAME) proportions (g FAME/100 g total FAME; AOCS, 1998). Approximately 95% of pigs slaughtered in the United States are marketed based on targeted carcass weight and lean estimates (Meyer, 2005). Therefore, emphasis has been placed by the pork industry to control variability in ending BW. A multiple marketing strategy is often used in commercial pig finishing operations, with pigs from a single barn being marketed over 2 (Gerlemann et al., 2013) to 6 wk (Gerlemann et al., 2014) to minimize variation in ending BW. The homogenization of ending BW is due to reduced competition as groups of pigs are removed, increasing ADG, ADFI, and G:F of the remaining pigs (DeDecker et al., 2007). Pigs were marketed in the present study to provide a desired HCW; therefore, it was expected that variability of HCW would not differ between carcasses of barrows and gilts (Table 1; Fig. 1a and 1b; P = 0.09). Furthermore, loin depth variability was not different between sexes (P = 0.60). Estimated carcass lean percentage of barrows was more variable (P < 0.0001) than that of gilts and was likely driven by the fact that BF depth was also more variable (P < 0.0001) in carcasses of barrows. This is possibly due to the fact that barrows have greater potential to deposit fat, and thus, the range in fat deposition in barrow populations is greater than in populations of gilts. In contrast, IV was less variable (P = 0.05) in barrows than in gilts. The greater variability of gilt IV may have been a reflection of the differences in adiposity between barrows and gilts; because the gilts are typically leaner than barrows (Smit et al., 2014), any change in fat deposition is likely to result in a more substantial change in fatty acid profile and IV in gilts as compared with the fatter barrows. According to the most recent estimates, excessive carcass fatness accounted for a$0.85/carcass loss in revenue, whereas excessive carcass leanness represented a $0.10/carcass loss of revenue to packers (Stetzer and McKeith, 2003). Although there was no difference in variability between sexes for HCW (possibly due to an effective multiple marketing strategy), the variability of BF depth and estimated carcass lean represent opportunities for producers and packers to use sex-specific management strategies to control variability, reduce the number of carcasses deviating from desired carcass specifications, and ultimately increase the utility and value of carcasses. Figure 1. Variability of selected carcass characteristics of barrows and gilts: (A) HCW (kg), (B) loin depth (mm), (C) back fat depth (mm), (D) estimated carcass lean (%), and (E) Iodine value (AOCS, 1998). ### Fresh Belly Characteristics Bellies of barrows were 4.0% heavier, 0.27 cm wider, and 0.19 cm thicker and had 0.39 unit greater (more firm) subjective flop score than bellies of gilts (Table 2; P ≤ 0.0001), but there was no difference (P = 0.22) in belly length. Barrow carcasses were both heavier and fatter than gilts, which explains barrows having heavier bellies that were both deeper and wider. Greater belly weight, thickness, and firmness in barrows compared with gilts have been previously reported (Uttaro et al., 1993; Kellner et al., 2015). View Full Table | Close Full ViewTable 2. Effect of sex on fresh belly characteristics least squares means and variability  No. Least squares means Variance Item Barrow Gilts Barrows Gilts SEM P-value Barrows Gilts Levene’s P-value Weight, kg 1,623 2,022 7.61 7.32 0.28 <0.0001 1.41 1.21 <0.01 Length, cm 1,623 2,022 69.38 69.23 1.37 0.20 17.67 19.19 0.10 Width, cm 1,623 2,021 36.00 35.73 0.74 <0.0001 6.20 5.78 0.13 Depth,1 cm 1,623 2,022 2.64 2.44 0.08 <0.0001 0.18 0.15 <0.01 Flop2 1,623 2,020 2.26 1.88 0.21 <0.0001 0.69 0.64 0.06 1Average belly depth was calculated as the average of measurements at 25, 50, and 75% the length of the belly (anterior to posterior). 2Subjective flop scores were anchored such that a score of 0.5 was characterized as a flop distance of less than 5 cm, a score of 1 was 5.1 to 10 cm, a score of 2 was 10.1 to 15 cm, a score of 3 was 15.1 to 20 cm, a score of 4 was 20.1 to 25 cm, and a score of 5 was greater than 25 cm. In the present study, barrow carcasses were fatter and had more variable BF depth than gilt carcasses, and this relationship was reflected in fresh belly characteristics. Fresh belly weight (P < 0.01) and belly depth were more variable (P < 0.01) in carcasses from barrows than in carcasses from gilts (Table 2; Fig. 2). Of all the pork primals, the belly has the greatest percentage of fat; thus, the belly’s weight and thickness are correlated with changes in adiposity (Kyle et al., 2014). As barrows have greater potential to deposit fat, the range of belly fatness within in a barrow population is likely to be greater than in a population of gilts, thereby increasing the potential for greater variability to exist, as was observed in the present study. There was no difference (P ≥ 0.06) in variability between barrows and gilts for belly length, width, or flop score. Nonconformity of belly thickness, particularly if bellies were too thin, accounted for$1.00/carcass loss of revenue (Stetzer and McKeith, 2003), a loss of value that is passed through the supply chain to bacon processors. Belly weight, depth, and flop are weakly correlated with commercial bacon slicing yield; even so, they have been reported to be among the best indicators of bacon processing yields (Kyle et al., 2014). Although these traits are not entirely indicative of finished product characteristics, reducing the variation of these fresh belly traits could potentially reduce variation in commercial bacon processing yields and slice characteristics. The most recent estimates by the U.S. Bureau of Labor Statistics (2016) reported that sliced bacon was the most valuable pork retail cut; therefore, any opportunity to improve the uniformity and predictability of fresh bellies has the potential to increase revenue for packers and processors.

Figure 2.

Variability of selected fresh belly characteristics of barrows and gilts: (A) belly weight (kg), (B) belly depth (cm), and (C) flop score (no median line displayed for flop score because barrow median was equal to first quartile and gilt median was equal to third quartile)

### Fresh Loin Characteristics

As previously discussed, carcasses from barrows had a greater HCW than carcasses from gilts; however, fresh, boneless loins from barrow carcasses weighed 0.18 kg less (Table 3; P < 0.0001) than those from gilt carcasses. This is likely the result of gilts being more heavily muscled and leaner at the 10th rib, as indicated by greater loin depth and reduced fat thickness. The pH of loins measured at 31 min postmortem were lesser in LM of barrows than in gilts (6.53 vs. 6.56; P = 0.03); however, at 24 h postmortem, barrows had a greater loin pH than gilts (5.71 vs. 5.69; P < 0.0001). Subjective marbling and firmness scores of loins from barrows were greater (P ≤ 0.01) than those from gilts. Although there was no difference in subjective color (P = 1.00), L* values of loins from barrows were 0.97 units greater (P < 0.0001), a* values were 0.14 units greater (P < 0.0001), and b* values were 0.28 units greater (P < 0.0001) than those of loins from gilts. It is not surprising the difference in L* was not detected in the subjective color evaluation, as differences in L* must typically exceed 2 units to be detectable by an evaluator (Zhu and Brewer, 1999). There were no differences in purge loss or cook loss (P ≥ 0.14) between loins from barrows and loins from gilts. Loin chops from barrows had 0.97 kg lesser (P < 0.01) SSF value than those from gilts. The greater instrumental toughness of gilts than barrows is in contrast to several previous studies that reported no effect of sex (Skelley and Handlin, 1971; Martel et al., 1988; Cisneros et al., 1996). However, other studies have reported LM of barrows to be more tender than that of gilts (Nold et al., 1997; D’Souza and Mullan, 2002). The greater tenderness of barrows may be due to having more marbling than gilts. Intramuscular fat in the loin muscle has been reported to have a small but significant, positive correlation with both sensory and instrumental tenderness (Huff-Lonergan et al., 2002).

View Full Table | Close Full ViewTable 3.

Effect of sex on fresh loin weight and quality characteristics least squares means and variability

 No. Least squares means Variance Item Barrow Gilts Barrows Gilts SEM P-value Barrows Gilts Levene’s P-value Boneless loin wt, kg 1,757 2,213 3.65 3.83 0.20 <0.0001 0.23 0.23 0.57 31-min pH 344 428 6.53 6.56 0.07 0.03 0.04 0.04 0.78 24-h pH 1,763 2,224 5.71 5.69 0.03 <0.0001 0.02 0.02 0.16 Color1 3,233 4,140 3.07 3.07 0.20 1.00 0.30 0.33 <0.01 Marbling1 3,233 4,140 2.36 1.87 0.49 <0.0001 0.88 0.71 <0.0001 Firmness2 3,233 4,140 2.80 2.67 0.13 <0.0001 0.01 0.03 0.08 L*3 1,727 2,207 53.05 52.48 0.97 <0.0001 6.32 5.91 0.15 a*3 1,727 2,207 7.42 7.28 0.36 <0.0001 1.36 1.29 0.20 b*3 1,727 2,207 13.86 13.58 0.38 <0.0001 1.02 1.07 0.28 Purge loss, % 364 439 0.86 0.91 0.29 0.14 0.35 0.41 0.36 Cook loss, % 371 445 17.09 17.21 0.78 0.35 4.12 3.97 0.76 Slice shear force, kg 371 445 14.90 15.06 2.40 <0.01 26.97 32.12 0.18
1Subjective color and marbling scores based on National Pork Producer Council quality standards (NPPC, 1999).
2Subjective firmness scores based on National Pork Producers Council standards (NPPC, 1991).
3L* measures darkness to lightness (greater L* value indicates a lighter color), a* measures redness (greater a* value indicates a redder color), and b* measures yellowness (greater b* value indicates a more yellow color).

Subjective color was less variable (Figure 3; P < 0.01) in loins from barrows than in loins from gilts. However, subjective marbling scores of loins from barrows were more variable (P < 0.0001) than those of loins from gilts. The greater variability of marbling score in the barrow population was similar to the results observed for other traits associated with carcass adiposity and, likewise, was likely due to the greater potential of barrows to deposit fat compared with gilts. There was no difference (P ≥ 0.08) in variability between loins from barrows and loins from gilts for boneless weight, 30-min pH, 24-h pH, subjective firmness, instrumental color, purge loss, cook loss, or SSF. Nonconformity related to loin quality issues represents an estimated \$2.13/carcass direct loss to packers (Stetzer and McKeith, 2003). In the present experiment, there was little difference between sexes in variability for most loin quality traits. However, the existence of greater variability in color in gilts and greater variability in marbling in barrows represents an opportunity to improve the uniformity of loin products offered to consumers.

Figure 3.

Variability of selected fresh loin characteristics of barrows and gilts: (A) boneless loin weight (kg), (B) 24-h pH, (C) marbling (NPPC, 1999), (D) L* (lightness), (E) a* (redness), and (F) slice shear force (kg). [

### Fresh Ham Characteristics

There was no effect of sex on skin-on, bone-in ham weight (Table 4; P = 0.35). However, skinned and trimmed ham weights of barrows were 0.16 kg less (P < 0.01) than those of gilts. The heavier trim weight of hams from gilts compared with those from barrows is in agreement with the data of Boler et al. (2011), who reported that trimmed hams of barrows weighed less and made up a lesser percentage of HCW than trimmed hams from gilts. The greater skinned and trimmed ham weight of gilts is likely due to their carcasses being more heavily muscled than those of barrows, as indicated by gilt carcasses having greater loin depth and heavier boneless loins than barrow carcasses. It would be expected, then, that with heavier muscling, the skin-on, bone-in ham weight of gilts would be greater than barrows, but no difference was observed. Barrow carcasses were fatter (as indicated by greater BF depth) and the lack of difference in skin-on, bone-in ham weight may be due to the difference in muscling being compensated with greater amount of subcutaneous fat of the hams from barrows. The greater muscling of gilt hams is further substantiated by the fact that the inside ham, outside ham, and knuckle of barrows weighed less (P ≤ 0.01) than those of gilts, although there was no effect of sex on shank or lite butt weight (P ≥ 0.62). The heavier weight of the ham muscle components of gilts compared with barrows further indicates that the greater weight of trimmed hams from gilts is due to a greater degree of muscling in gilt carcasses.

View Full Table | Close Full ViewTable 4.

Effect of sex on fresh ham weights and quality characteristics least squares means and variability

 No. Least squares means Variance Item Barrow Gilts Barrow Gilts SEM P-value Barrows Gilts Levene’s P-value Skin-on wt, kg 379 460 11.6 11.66 0.24 0.35 1.17 1.24 0.58 Trimmed wt, kg 379 461 9.70 9.86 0.26 <0.01 0.85 0.94 0.39 Inside wt, kg 376 460 1.60 1.67 0.07 <0.0001 0.05 0.06 0.36 Outside wt, kg 381 462 2.19 2.26 0.10 <0.01 0.09 0.10 0.80 Knuckle wt, kg 381 462 1.30 1.35 0.04 <0.0001 0.03 0.03 0.18 Shank wt, kg 381 462 0.69 0.69 0.02 0.62 0.01 0.01 0.46 Lite butt wt, kg 380 461 0.35 0.35 0.02 0.96 0.01 0.01 0.40 Gluteus profundis L*1 3,253 4,157 40.85 40.30 0.53 <0.0001 12.83 13.12 0.12 a*1 3,253 4,157 15.81 15.76 0.13 0.36 4.93 4.59 0.90 b*1 3,252 4,156 3.82 3.62 0.26 <0.0001 2.85 2.85 0.32 Gluteus medius L* 3,255 4,160 45.84 45.51 0.56 <0.0001 11.78 11.04 0.16 a* 3,253 4,160 9.11 9.03 0.44 0.07 3.74 3.07 0.30 b* 3,253 4,160 2.46 2.23 0.36 <0.0001 2.52 2.49 0.56 Semimembranosus L* 381 457 46.79 46.35 0.42 0.04 10.31 9.40 0.46 a* 381 458 9.38 9.67 0.37 0.02 3.82 3.17 0.14 b* 380 457 1.49 1.49 0.23 0.99 2.57 2.32 0.44 pH 381 459 5.68 5.67 0.07 0.63 0.08 0.08 0.94
1L* measures darkness to lightness (greater L* value indicates a lighter color), a* measures redness (greater a* value indicates a redder color), and b* measures yellowness (greater b* value indicates a more yellow color).

The gluteus profundis of barrows had 0.53 greater L* (P < 0.0001) and 0.20 unit greater b* value than gilts, but there was no difference (P = 0.36) in a* between the gluteus profundis of barrows compared with gilts. The gluteus medius of barrows had 0.33 unit greater (P < 0.0001) L* value and 0.23 unit greater (P < 0.0001) b* value than the gluteus medius from gilts but did not differ (P = 0.07) in a*. Barrow semimembranosus muscles had 0.44 unit greater (P = 0.04) L* value but 0.29 unit lesser (P = 0.02) a* value than the semimembranosus from gilts. There was no effect of sex (P = 0.99) on b* value or ultimate pH (P = 0.63) of the semimembranosus. Greater L* and b* values in the gluteus medius of barrows compared with that of gilts, with no difference in a*, has been previously reported (Uttaro et al., 1993). Although there were significant differences in color between ham muscles of barrows and ham muscles of gilts, the magnitude of differences were not likely great enough to be discernible by consumers.

Furthermore, there were no differences (Table 4; Figure 4; P ≥ 0.12) in the variability of barrows and gilts for any fresh ham characteristics. The homogeneity of variance between barrows and gilts for ham characteristics, especially for skin-on, bone-in ham weight, was unexpected. With the disparity in variability observed in carcass fatness and estimated carcass lean, it was expected that a similar effect on variability of skin-on, bone-in hams would have been observed. It is likely that because the ham is among the leanest primals, any differences between barrows and gilts in the pattern of fat deposition would not have been substantial enough to cause differences in the variability of the overall ham weight. Additionally, the individual ham muscles were trimmed to uniform subcutaneous fat depth, and therefore, the differences in variability of traits associated with adiposity of the other primals would not be observed in the ham. The lack of difference in variability between sexes indicates that hams can be managed and processed in any manner, irrespective of whether they were from barrows or gilts.

Figure 4.

Variability of selected fresh ham characteristics of barrows and gilts: (A) skin-on, bone-in ham weight (kg), (B) trimmed ham weight (kg), (C) L* (lightness), and (D) semimembranosus pH.

### Conclusions

Even though a multiple marketing group strategy was used to minimize variability of HCW, barrows were more variable than gilts for traits associated with fat deposition, with the exception of IV, and there were minimal differences in variability between sexes for traits associated with carcass lean or quality traits not directly related to fat. Understanding the differences in variability offers the opportunity for the pork industry to minimize variation, improve product uniformity and consumer eating experience, and, ultimately, capture additional revenue from each carcass by managing variability of the pig due to sex.