Evaluation of growth, meat quality, and sensory characteristics of wool, hair, and composite lambs

Abstract

The objectives of this study were to evaluate the growth rates, carcass quality, shelf-life, tenderness, sensory characteristics, volatile compounds, and fatty acid composition of wool, hair, and composite (wool × hair) lambs. Twenty-one wether lambs [wool (Suffolk × Polypay/Targhee; n = 7), hair (Dorper × Dorper; n = 7), and composite (Dorper × Polypay/Targhee; n = 7)] were fed from weaning to finishing at the University of Idaho Sheep Center and subsequently harvested under United States Department of Agriculture inspection at the University of Idaho Meat Lab. At 48 h postmortem, carcass measurements were taken to determine the percent boneless closely trimmed retail cuts, yield grade, and quality grade. Loins were fabricated from each carcass and wet-aged at 0°C until 10-d postmortem. Following aging, 2.54-cm bone-in loin chops were cut and randomly assigned to 4 d of retail display, Warner–Bratzler Shear Force (WBSF), or sensory analyses. Thiobarbituric acid reactive substances were analyzed on days 0 and 4 of retail display while subjective and objective color measurements were observed once daily. Samples (24 g) were also collected for volatile compound and fatty acid analysis. A mixed model analysis of variance was used to assess breed differences. Discernable effects were considered at P < 0.05. Wool lambs had heavier hot carcass weights (P < 0.001), larger rib-eye area (P = 0.015), and higher dressing percent (P < 0.001) than the other breeds. There was an interaction observed between breed and days of retail display for browning (P = 0.006). On day 1 chops from the composite breed had more browning than chops from the wool breed. No differences were observed between groups for lean muscle L* values (P = 0.432), a* values (P = 0.757), and b* values (P = 0.615). Differences were not observed in lipid oxidation (P = 0.159), WBSF (P = 0.540), or consumer acceptability (P = 0.295). There were differences found for 7 of the 45 fatty acids detected and in 3 of the 67 volatile compounds detected. In conclusion, wool lambs were heavier and had a greater carcass yield than the hair lamb carcasses. Regardless of breed, consumers did not detect sensory traits that would impact their eating experience.

The current study found the wool group maintained higher growth from weaning until harvest with improved carcass yield and rib-eye area maintaining a superior product for the producers. There was minimal variation between groups for shelf-life and sensory analysis establishing that U.S. consumers are unable to detect breed differences.

Introduction

The United States Department of Agriculture (USDA) Economic Research Service (ERS) tracked lamb consumption finding it has been on a consistent decline since the 1940s, eventually plateauing at 1 pound per person, per year, per capita (USDA-ERS, 2020). Production operations have been notably declining alongside consumption, making up less than 1% of United States (U.S.) livestock industry receipts. According to National Agricultural Statistical Service, there was a drop from over 105,000 operations in 1990, to less than 80,000 in the last 2012 report (USDA-ERS, 2020). Conversely, the USDA completed an agriculture census in 2017 that showed an operation inventory of 101,387 with, 24% of operations owning less than 100 lambs, and 69% owning 24 lambs or less (USDA-ERS, 2020).

Sheep fall into two common categories in the United States, wool breeds and dual-purpose meat breeds. Synthetic fabrics have created a cheaper alternative, causing the value of wool production to drop from $49 billion in 2014 to $33 billion in 2018 (Persistence Market Research, 2019). Total U.S. shorn wool production during 2021 reached 22.5 million pounds, 3% down from 2020. The average price of $1.70 per pound was also down 1% from 2020 for an annual total of $38.2 million (USDA-NASS, 2022). With these drops in production, the market calls predominantly for fine wool lambs, such as the Merino or Rambouillet, their thin fiber diameter (17–24 microns) and fine crimp creates high-quality wool (LeValley, 2011). In order to get this fleece, sheep must be shorn once a year. Throughout the United States, there is a wide variety of shearing costs dependent on location and flock size. The average cost to shear is between $5 and 15 per head (Shroeder, 2022). This is affecting producers’ who raise market lambs and still shear their lambs as market values for medium-quality fleece are lower, causing producers to lose money annually (Pines, 2022).

The other common U.S. lamb category is dual-purpose that is raised mainly for meat. These sheep have fast growth rates and high carcass quality; examples include Suffolk and Hampshire. Crossbreeding to create a hybrid vigor lamb is a common way to gain the best attributes from the ram and ewe (Schoenian, 2006). One cross that shows breed complementarity is a Suffolk ram crossed with a Polypay ewe. The ram has the advantage of fast growth and heavy muscling while the ewe contributes prolificacy and lower feed requirements (Aaron, 2014).

Another type of sheep on the rise in the United States is the hair sheep. Rising from 1% in 1996 to 21.7% in 2011 (USDA, 2011). Local producers state that hair breeds, Dorper particularly, have milder flavor profiles than common wool meat breeds (The Dorper Difference, 2022). Dorper lambs are a superior option for U.S. producers when compared to other hair breeds like the St. Croix as Dorpers have larger frame sizes and are heavier (Burke and Miller, 2002, 2004). Breeds like the Dorper and Katahdin have varied shedding ability, causing shearing to be prolonged or not required, decreasing maintenance for producers (Kintzel, 2011).

One way to potentially increase lamb consumption and, therefore, production is to create a composite breed from hair and wool breeds. Based on previous research, utilizing a hair sire breed crossed with wool breeds opens the possibility of creating heterosis, and producing a sheep that would not have to be shorn (Duckett and Kuber 2001; Snowder and Duckett 2003; Burke and Apple, 2007). If this composite results in hybrid vigor, increasing growth rates and high carcass quality alongside a mild aroma and flavor, this could provide an opportunity for increased consumer demand and producer profits.

The objectives of this study were to evaluate the growth traits, carcass quality, shelf-life, sensory characteristics, volatile compounds, and fatty acids of wool, hair, and composite (wool × hair) lamb breeds.

Materials and Methods

IACUC statement of approval

This project was reviewed and approved by the University of Idaho Institutional Animal Care and Use Committee (IACUC) on 10/02/2020 Protocol number IACUC-2020-57.

IRB approval

This project was reviewed by the University of Idaho Institutional Review Board (IRB) and considered Exempt under Category 6 at 45 CRF 46.104(d)(6).

Rearing

Two University of Idaho (UI) Sheep Center breeding groups included Targhee/Polypay ewes with a Suffolk ram (wool group) and Targhee/Polypay ewes with a Dorper ram (composite group). Wool wether lambs (n = 7) and composite wether lambs (n = 7) were born at the UI Sheep Center and purchased by the research project at an average age of 69 d. All lambs were castrated between 30 d and 45 d. The hair (Dorper) wether lambs (n = 7) were purchased from a local producer at an average age of 79 d. The purchased hair lambs entered a 30-d quarantine at the UI Sheep Center upon arrival. Once the trial began, all lambs were raised on the University of Idaho total mixed ration (TMR), formulated by a university nutritionist (Table 1). Feed was weighed each feeding (07:00 and 15:00), the amount fed varied depending on bunk score. The targeted bunk score was 0.5 (scattered remaining feed, majority of the bunk bottom exposed; Rusche, 2020). Following a 30-d quarantine, in the same environment as the other two groups, the hair lambs were integrated into the same pen as the wool and composite lamb groups. The study started on the date of hair sheep arrival. Growth measurements, weight (kg), heart girth circumference (cm), and shoulder height (cm) were taken on day 0 (08:00) and the two final days before each harvest was averaged. Average daily gain (ADG) was calculated using the following equation:

Table 1.

Composition of total mixed ration fed to lambs from day 0 of study to harvest

Ingredients	DM percent	As fed percent
Feeder Alfalfa	28.2	17.4
Performix sheep	5.86	4.59
Corn	27.3	17.2
Barley	31.3	19.9
Sodium bicarb	2.28	1.31
Magna fat	4.54	2.62
Salt	0.48	0.26
Water	0	36.8
Dry matter	54.8%

$${\displaystyle \begin{array}{l}{\displaystyle ADG=\frac{Finalweight-Initialweight}{DaysFed}}\end{array}}$$

Feed analysis

Feed samples were taken weekly, placed into gallon freezer Ziploc bags, and frozen in a –20°C freezer. Samples were then mixed into 0.23 kg. monthly composite feed samples and shipped to Dairy One forage lab for TMR, near infrared reflectance spectroscopy (NIR), and Wet Chemistry Mineral feed analysis.

Harvest

Lambs were harvested under USDA Inspection at the University of Idaho Meat Lab when they reached ~0.5 cm of 12th rib fat thickness as estimated by trained University of Idaho personnel. There were three harvest days, the first harvest date contained three lambs from each group ranging in age from 171 d to 205 d, harvesting a total of nine. Lambs were on feed for 90 d for the first harvest group. The second harvest date contained three lambs from each group ranging in age from 183 d to 223 d, harvesting a total of nine. Lambs were on feed for 111 d for the second harvest. The final harvest date contained two lambs each, from the wool and composite groups and three lambs from the hair group ranging in age from 189 d to 219 d. Lambs were on feed for 125 d for the final harvest group. Directly after harvest, hot carcass weights (HCW) were recorded before being placed in the cooler. At 48 h. postmortem, back fat (BF), body wall, rib-eye area (REA), conformation, maturity, and flank streaking were evaluated these values, percent boneless closely trimmed retail cuts (%BCTRC; Boggs et al., 2006), quality grade (QG), and yield grade (YG) were calculated following USDA guidelines (USDA, 1992). Carcasses were fabricated, and the lamb loin (IMPS 231) was removed following North American Meat Processors Association’s Institutional Meat Purchase Specifications (USDA, 2014). Loins were then vacuum sealed (FlairPak 300. 10 × 14; Appleton, WI) with bone guard (Waltons, #4001207; Wichita, KS) and aged at 0°C until 10 d postmortem.

Product preparation

Following the 10-d aging period, 2.54-cm bone-in loin chops were cut to use the longissimus thoracis for all further analyses, utilizing a bandsaw and randomly assigned to retail display (n = 1), thiobarbituric reactive substance (TBARS) (n = 1), Warner–Bratzler shear force (n = 2), consumer sensory panel (n = 3), volatile compounds analysis (n = 1), or fatty acid methyl esters (FAMES) (n = 1). The retail display and TBARS chops were placed in Styrofoam trays (CKF Inc. #88102, Langley, BC, Canada), and overwrapped with an oxygen-permeable polyvinyl chloride (PVC) film (oxygen transmission rate: 1,450 cc/645cm² per 24 h; water vapor transmission rate: 17.0 g/645 cm 2 per 24 h; Koch Industries, Inc. #7500-3815; Wichita, KS). The remaining chops were vacuum sealed (FlairPak 300, 6 × 10 and 8 × 10; Appleton, WI), and frozen at –20°C, until analyses were conducted.

Retail color

Retail display chops (n = 2) were displayed in a glass-fronted retail display case (Model GDM-69, True Manufacturing Co., O’Fallon, MO) with fluorescent lighting (408 lux) at 2–4°C for 4 d. To avoid effects due to display case location, chops were rotated daily after each measurement. The chops were allowed to bloom for at least 60 min before subjective and instrumental color analysis and before TBARS samples were acquired. Oxygenated lean color (1 = extremely bright cherry-red, 8 = extremely dark red), surface discoloration (1 = no discoloration 0%, 6 = extensive discoloration 81–100%), amount of browning (1 = no evidence of browning, 6 = dark brown), discoloration (1 = none, 5 = extreme), color uniformity (1 = uniform, no two-toning, 5 = extreme two-toning), and bone marrow color (1 = bright reddish-pink to red, 7 = black) were evaluated by three evaluators daily (08:00) following American Meat Science Association Guidelines (AMSA, 2012). The evaluators were within an 18–59 age range. Evaluators were trained with pictures and color tiles to assist in standardizing. To analyze instrumental color, color measurements were taken daily (08:00) using a Nix Pro 2 Color Sensor (Nix Sensor Ltd., Hamilton, ON, Canada) with a 2° standard observer with a 14 mm-diameter area of measurement. The instrument was set to illuminant D₆₅ and Commission International de’l Eclairage (CIE) L*(lightness), a* (redness), and b* (yellowness) values were recorded.

Lipid oxidation

Thiobarbituric acid reactive substances, an indicator of lipid oxidation were analyzed on days 0 and 4 of retail display following the protocol in Section XI, Appendix O of the Meat Color Measurement Guidelines (AMSA, 2012). Approximately 0.5 g of muscle samples were excised from the Longissimus dorsi avoiding large pieces of fat, connective tissue, and the exposed edges of the chop. Samples were then prepared and analyzed in duplicate.

Cooking

Chops were thawed for 24 h at 4°C before being cooked to a target internal temperature of 71°C using clamshell countertop grills (Cuisinart Griddler Deluxe Model GR-150), chop location on grills were rotated to avoid location influence. The internal temperatures were continuously monitored using thermometers (32311-K Econo-TempTM Thermocouple, Atkins, Middlefield, CT) inserted into the geometric center of each chop. The temperature was recorded at time off grill as well as peak temperature. Cook loss was determined using the equation below.

$${\displaystyle \begin{array}{l}{\displaystyle CookLoss=\frac{Rawweight-Cookedweight}{Rawweight}x100\mathrm{\%}}\end{array}}$$

Warner–Bratzler shear force

Two chops from each lamb were used for Warner–Bratzler shear force (WBSF) analysis. Chops were cooked as described above, then cooled to room temperature. Once cool, each chop had three 1.27-cm cores removed parallel to the muscle fiber direction following the guidelines of Section VIII, Appendix B. a. (AMSA, 2015). The cores were then sheared perpendicular to the muscle fiber direction using a WBSF machine (G-R Manufacturing, Manhattan, KS). Peak shear force values were recorded from each core, the average of the cores from each chop was then analyzed, determining the WBSF of each chop.

Consumer sensory analysis

A consumer taste panel was implemented in the Margaret Ritchie School of Family and Consumer Sciences Mary Hall Niccolls Building Test Kitchen on the University of Idaho campus following the AMSA guidelines (AMSA, 2015). Consumer panelists (n = 74) were asked to sign a consent form prior to sampling as well as complete a demographics survey. Panelists were given salt-free soda crackers and tap water between each sample to cleanse their palate. After cooking as described above, chops were sliced into three equal portions approximately (1.27 × 1.27 × 2.54 cm) and placed individually into 2 oz plastic disposable covered labeled cups, three chops were used from each lamb and three samples were cut from each chop, panelists tasted two to three samples.

Samples were placed in a warmer heated to a range of 37–51°C (H-Series Insulated Food Pan Carrier, Cambro Manufacturing, SKU: CAM-UPCH400) prior to being served to a panelist who received one sample from each group with randomly assigned three-digit, blind-coded sample numbers. The consumer panelists were also asked to fill out a questionnaire rating samples for overall acceptability, texture, juiciness, and flavor using a 10-point scale (10 = like extremely, and 1 = dislike extremely). Panelists were also asked if they could detect an off flavor and if they would purchase the product.

Volatile compound analysis

Volatile compound analysis was performed at Texas Tech University following the protocol by Gardner and Legako (2018). Samples (24 g) were prepared at the University of Idaho by cooking as described previously. Samples were allowed to cool overnight in a refrigerator before being cubed and frozen in liquid nitrogen. Samples were then homogenized, placed into labeled Whirlpak 4 and 18 oz standup bags, and stored at –80°C, until being shipped on dry ice to Texas Tech University. In a sealed glass vial, 5 g of the sample was weighed. The volatile compounds were collected from the headspace and then extracted through a solid phase microextraction. Extraction and injection of volatile compounds were then carried out using a Gerstel automated MPS sampler. Volatile compounds were separated by gas chromatography using VF-5ms capillary column. After separation, the volatile compounds were detected by a mass spectrometer, and the identity was validated utilizing external standard comparisons of ion fragmentation patterns. Quantitative determinations (ng/g) of compounds of interest were then conducted by an external standard method.

Fatty acid methyl esters

Fatty acid methyl ester analysis was performed at Texas Tech University following the protocol described by O’Fallon et al. (2007) which determined the fatty acid concentration of chops. Samples (24 g) were prepared at the University of as described in the volatile compound methods. Powdered samples (1 g) were weighed into 20 ml glass vials. An internal standard (C13:0) was added to the vials (1 ml). Next, 0.7 ml of 10 N potassium hydroxide and 5.3 ml of methanol was added to the tube for derivatization. This tube was then incubated for 1.5 h at 55°C. Following incubation, tubes were then placed in cold water for 20 min, 0.58 ml of 24 N sulfuric acid was then added, the tubes were then re-incubated under the previous conditions. Following final incubation, 3 ml of hexane was added, tubes were vortexed for 5 min, and then centrifuged for 5 min at 3500 rpm. Gas chromatography-flame ionization detection was used to detect fatty acid concentration. Following separation, fatty acid identity and quantity were confirmed using external standards and calculated on a weight basis.

Statistical analysis

Data were analyzed utilizing the mixed model analysis of variance to assess breed differences unless otherwise noted. Grading was analyzed with the day of harvest as a random effect. The retail display was analyzed using a heterogenous autoregressive covariance structure to account for repeated measures within groups and a Tukey’s adjustment for multiple comparisons with the day as a repeated measure. Breed-by-day interaction was evaluated for color measurements and lipid oxidation. Subjective color scores were averaged across the three trained evaluators prior to analysis. Shear force and volatile compounds were analyzed using final cook temperature as a covariate. Consumer taste panel willingness to purchase and off-flavor were analyzed using a generalized mixed model assuming a binary distribution and summarized as percentages. All statistical analyses were run using statistical analysis system (SAS) V 9.4 (SAS Inc., Cary, NC). Discernable effects were considered at P < 0.05.

Results and Discussion

Growth traits

Limitations of the current study include the number of lambs per group, this number of replications is a constraint on the following conclusions. Growth traits are included in Table 2. There were no differences found between groups for day 0 live weight (P = 0.058). Burke and Apple (2007) conducted research that differs from the current study, with Suffolk lambs at weaning being heavier than Dorper lambs. There were differences for day 0 shoulder height (P < 0.001), and heart girth circumference (P < 0.011). The wool group was taller than the composite group which was taller than the hair group. The wool and composite groups had larger heart girth circumferences than the hair group. Previous research has not been found that includes values for weaning shoulder height or heart girth circumference between breeds. There were differences found between groups for final live weight (P < 0.001), shoulder height (P < 0.001), and heart girth circumference (P < 0.001). The wool group had an overall higher final weight. The wool group was taller and had a larger heart girth circumference than the composite which was taller and had a larger heart girth circumference than the hair group.

Table 2.

Estimated mean effects of composite, hair, and wool lamb growth and carcass traits

	Groups			SEM	P Value
Traits	Composite	Hair	Wool
Live traits
D 0 Weight (kg)	26.7	21.6	29.9	2.3	0.058
D 0 shoulder height (cm)	21.5b	19.0c	24.5a	0.5	<0.001
D 0 Heart girth circumference (cm)	69.9a	61.7b	72.0a	2.2	<0.011
Weight (kg)	48.7b	43.9b	60.5a	2.1	<0.001
Heart girth circumference (cm)	86.2b	81.4c	93.7a	1.3	<0.001
Shoulder height (cm)	59.7b	54.3c	69.7a	0.4	<0.001
ADG (kg)1	0.22	0.22	0.25	0.02	0.129
Carcass traits
HCW (kg)	27.5b	25.8b	35.5a	1.2	<0.001
Back fat (cm)	0.40ab	0.36b	0.44a	0.02	0.022
Body wall (cm)	0.79b	0.87b	1.05a	0.05	0.006
REA (sq cm)	15.92b	17.10b	19.26a	0.73	0.015
Dressing percent	51.63b	52.45b	54.81a	0.51	0.001
Yield grade	2.02ab	1.71b	2.44a	0.17	0.022
Quality grade2	11.8a	10.8b	12.4a	0.3	0.004
%BCTRC3	47.78b	48.49a	46.48b	0.49	0.031
Flank streaking scores4	12.6a	10.8b	12.8a	0.5	0.012

¹Average daily gain.

²Quality Grade Scale: Low Choice: 10 – Average Prime: 14.

³Percent Boneless Closely Trimmed Retail Cuts (Boggs, et al., 2006).

³Flank Streaking Scale: Traces: 10 to Moderate: 14.

^a–cWithin a row and trait, means without a common superscript differ (P < 0.05).

Throughout the feeding period, the composite group performed similarly in ADG to the wool group, it did have a smaller final heart girth circumference and lighter final weight; however, it maintained its height and heart girth circumference over the hair group. In agreement with the current study, Burke and Apple (2007) demonstrated that purebred Suffolk was heavier at slaughter than the 7/8^ths Dorper crosses. Furthermore, it was noted, from weaning to slaughter, the Dorper and Suffolk cross wethers had heavier weights than the other hair group. Lopez-Carlos et al. (2010) found that Dorper ram lambs had a greater heart girth circumference compared to other hair groups, including Katahdin and Blackbelly, which were not different from each other. Even though the current study found differences in final weight there was no difference observed for ADG (P = 0.129). Results of the current study are in agreement with Burke and Apple (2007), where it was found that the 7/8^ths Dorper crosses and purebred Suffolk wethers had comparable ADG to each other, even with the Suffolk wethers finishing at a higher weight. Other studies including Cloete et al. (2000) and Notter et al. (2004) have also found that Dorper lambs commonly have similar ADG when compared to many wool breeds.

Carcass traits

Summary results for all carcass traits are recorded in Table 2. There were differences seen between groups for HCW (P < 0.001), dressing percent (P < 0.001), and REA (P = 0.015). The wool group had heavier HCW, greater dressing percentages, and larger REA than the other two groups which were not different from one another. Monaco et al. (2014) also found that Suffolk lambs had heavier carcasses than Dorper carcasses. Another study by Snowder and Duckett (2003) showed that Columbia ewes crossed with Dorper-sired lambs had higher slaughter weights than Columbia or Suffolk-sired rams. When looking at the dressing percent, the same study found that dressing percentage was similar between Suffolk and Dorper-sired wethers. Burke and Apple (2007) differ from the current study, as it found that 7/8^ths Dorper cross and purebred Suffolk wethers both had similar, greater dressing percent’s than Katahdin wethers.

There were differences detected between the groups for back fat (P = 0.022) and body wall (P = 0.006). The wool group had thicker back fat than the hair group and thicker body wall than the other two groups. Other studies like Duckett and Snowder (1999) found differing results showing that Dorper-sired lamb carcasses had similar external fat thickness as Suffolk-sired lambs. There were observed differences between groups for %BCTRC (P = 0.031). The hair group had a higher value for %BCTRC over the other groups. The previously stated study in agreement with the current study reported the Dorper-sired lambs had a mild advantage over the Suffolk-sired lambs for trimmed loin, rack, and leg cut weights. There were differences found for yield grade (P = 0.022) between groups. The wool group had a higher yield grade than the hair group. Burke and Apple (2007) however, showed comparable yield grades between Dorper and Suffolk-sired lambs. Differences were found between groups for the quality grade (P = 0.004). The wool group had greater quality grades than the hair group but did not differ from composite groups. All lambs were determined to be young lamb based on skeletal and lean maturity. From a study by Nell (1998) reported that the Dorper-sired lamb crosses had similar carcass quality grades to the Suffolk-sired lambs. Another study by Snowder and Duckett (2003) found similar results with the Dorper, Suffolk, and Columbia-sired lambs having comparable carcass quality results. There were also differences found for flank streaking scores (P = 0.012). The wool and composite carcasses had higher flank streaking scores than the hair group. The composite lambs displayed improved flank streaking and quality grading over the hair lambs; however, the hair lambs did manage to have improved %BCTRC.

Retail subjective color

There was a group-by-day interaction for subjective browning color (P = 0.006; Table 3). The composite group chops browned at a faster rate initially than the wool group chops. There was no difference between groups subjective color-scoring discoloration (P = 0.065; Table 4), oxygenated lean color (P = 0.407), surface discoloration (P = 0.107), and color uniformity (P = 0.082). There was a difference in bone marrow color between groups (P = 0.019; Table 4) with the chops from the wool group having darker marrow than the composite group.

Table 3.

Estimated mean effects of day of retail display and breed group (composite, hair, and wool) on Longissimus dorsi amount of browning

	Day of	Groups			(P = 0.006)
Trait	Display	Composite	Hair	Wool	SE
Amount of browning	0	1.3c	1.2d	1.1d	0.1
	1	2.1bx	1.7cxy	1.4cy	0.1
	2	2.2ab	2.3b	2.0b	0.1
	3	2.7a	2.7b	2.3b	0.1
	4	3.1a	3.4a	3.2a	0.1

1 = no evidence of browning, 6 = dark brown.

^a–cWithin a column, means without a common superscript differ (P < 0.05).

^x–yWithin a row, means without a common superscript differ (P < 0.05).

Table 4.

Estimated mean effects of breed group (composite, hair, and wool) on Longissimus dorsi subjective and objective color

	Groups			SE	P value
Traits	Composite	Hair	Wool
Subjective color
Oxygenated lean color1	4.7	5.1	5.0	0.2	0.407
Discoloration2	2.3	2.4	1.9	0.2	0.065
Surface discoloration3	2.2	2.3	1.8	0.1	0.107
Color uniformity4	2.2	2.2	1.9	0.1	0.082
Bone marrow color5	2.3b	2.3b	3.0a	0.2	0.019
Objective color
Lean L*	36.28	35.09	35.65	0.64	0.435
Lean a*	13.79	13.46	13.96	0.49	0.757
Lean b*	11.07	10.62	11.21	0.45	0.615
Marrow L*	46.42	47.81	49.16	1.36	0.387
Marrow a*	22.56	20.55	19.98	1.25	0.328
Marrow b*	17.61	15.69	16.18	0.69	0.141

¹1 = extremely bright cherry-red, 8 = extremely dark red.

²1 = none, 5 = extreme.

³1 = none (0%), 6 = extensive (81–100%).

⁴1 = uniform, 5 = extreme two-toning.

⁵1 = bright reddish-pink to red, 7 = black.

^a,bWithin a row, means without a common superscript differ (P < 0.05).

Differences between days were seen for discoloration (P < 0.001; Table 5), surface discoloration (P < 0.001), oxygenated lean color (P < 0.001), color uniformity (P < 0.001), and bone marrow color (P < 0.001). In general, discoloration increased with visual desirability decreasing as days progressed. This is consistent with previously reported research that followed beef steak retail display discoloration (Colle et al., 2015, 2016).

Table 5.

Estimated mean effects of days of retail display on lamb Longissimus dorsi subjective and objective color

	Day of display					SE	P value
Traits	0	1	2	3	4
Subjective color
Oxygenated lean color1	3.7e	4.2d	5.0c	5.7b	6.0a	0.3	<0.001
Discoloration2	1.2e	1.7d	2.4c	2.8b	2.8a	0.3	<0.001
Surface discoloration3	1.4d	1.8c	2.3b	2.6a	2.5a	0.2	<0.001
Color uniformity4	1.3d	1.9c	2.9b	3.5a	3.7a	0.1	<0.001
Bone marrow color5	1.4e	1.9d	2.6c	3.2b	3.6a	0.1	<0.001
Objective color
Lean L*	35.55	35.84	36.21	35.18	35.59	0.64	0.272
Lean a*	15.47a	15.14a	13.71b	12.64c	11.73d	0.48	<0.001
Lean b*	11.85a	11.82a	10.64b	10.37ab	10.16b	0.45	0.010
Marrow L*	45.67	47.27	47.35	49.74	48.94	1.32	0.079
Marrow a*	22.71ab	23.25a	22.06abc	19.48bc	18.67c	1.42	0.006
Marrow b*	16.25	18.12	17.31	15.70	15.07	0.94	0.087

¹1 = extremely bright cherry-red, 8 = extremely dark red.

²1 = none, 5 = extreme.

³1 = none (0%), 6 = extensive (81–100%).

⁴1 = uniform, 5 = extreme two-toning.

⁵1 = bright reddish-pink to red, 7 = black.

^a–cWithin a row and trait, means without a common superscript differ (P < 0.05).

Retail objective color

Objective color measurements for lean showed no differences for L*, a*, or b* values between groups (P = 0.435, P = 0.757, and P = 0.615, respectively; Table 4). Burke and Apple (2007) had similar results after a 30-min bloom period, with the Dorper and Suffolk carcasses having no differences between breeds for L*. It was also determined that the Dorper carcasses had redder Longissimus muscle scores, with higher a* values than Suffolk and St. Croix carcasses making their chops appear more yellow, contrary to the current study. Different instruments were utilized for objective color measurements between trials as well as different bloom time lengths, potentially resulting in divergent values. Geesink et al. (2000) determined that higher L*, a*, and b* values are considered more desirable to consumers. This study did not find differences in color, potentially due to the fact the lambs were of similar age and maintained the same diet.

Objective color measurements for lean showed no differences in L* for days of retail display (P = 0.272; Table 5). There were differences for objective color a* measurements between days (P < 0.001; Table 5). Lean a* had similar values on days 0 and 1, then decreased from day 2 through 4. It is important to mention all groups had positive redness values, greater than the 14.5 known for being consumers threshold for acceptability (Holman et al. 2020). There were differences found between days (P = 0.010; Table 5) for b* lean color. The b* value on day 0 was greater than day 2 and day 4, and day 1 was higher than day 2 and day 4. Although there are not many studies on lamb, beef has been shown to decrease in redness and yellowness over retail display time (Colle et al. 2015, 2016, 2019).

There were no differences found between groups for bone marrow L*, a*, or b* values (P = 0.387, P = 0.328, and P = 0.141, respectively; Table 4). Limited research has been found on lamb chop bone marrow discoloration. No differences were found between days of retail display for bone marrow L* and b* values (P = 0.079 and P = 0.087; Table 5). However, there were differences for bone marrow a* values between days of retail display (P < 0.006; Table 5). On day 1, bone marrow a* values were greater, with a higher redness, than days 3 and 4 and day 2 had a higher value than day 4, There is limited research on lamb bone marrow scores however the bone marrow color decline was expected after oxygen exposure causes iron oxidation in the hemoglobin as shown in beef studies (Lanari et al., 1995; Gill, 1996).

Lipid oxidation

When evaluating lipid oxidation, there were no differences between groups (P = 0.159; Table 6). There was a difference between days of retail display (P < 0.001) with chops having greater lipid oxidation on day 4 than day 0. The day 0 to day 4 lipid oxidation levels increased from 0.21 to 1.02 mg of Malondialdehyde (MDA)/kg of muscle. Tarladgis et al. (1960) and Greene and Cumuze (1982) determined the threshold for consumers to detect off-flavors in meat is TBARS > 1.0 showing the potential for consumers to detect an off-flavor in this study’s day 4 samples. Further research is required to determine if consumer sensory analysis on the final day of display would have increased off-flavor detection.

Table 6.

Estimated mean effects of Longissimus dorsi Lipid Oxidation¹, WBSF (kg), and percent cook loss comparison between breed groups (composite, hair, and wool)

	Groups
Trait	Composite	Hair	Wool	SE	P value
Lipid oxidation	0.50	0.89	0.59	0.14	0.159
WBSF (kg)2	3.79	3.31	3.45	0.31	0.540
Cooking loss %	26.50	24.26	25.08	0.68	0.084

¹Warner–Bratzler shear force.

²Lipid oxidation (mg MDA/kg muscle).

Cooking loss

Percent cooking loss for the groups was similar (P = 0.084; Table 6). This differs from Burke and Apple (2007), who found cooking losses were greater in Suffolk chops when compared to Dorper or Katahdin chops. Another study by Estrada-Leon et al. (2022), found that Dorper-crossed Katahdin lambs had increased cook loss compared to purebred Dorper lambs. Although cooked to the same temperature, Burke and Apple (2007) used a convection oven instead of a grill and air movement could create differences between trials.

Warner–Bratzler shear force

The WBSF analysis had no differences between groups (P = 0.540; Table 6). This differs from results found by Duckett and Snowder (1999) where Dorper-influenced lambs had increased tenderness levels with decreased Warner–Bratzler Shear Force average scores that were 1.1 kg lower than Suffolk-influenced lambs when comparing chops aged 10 days. Another study by Snowder and Duckett (2003) with differing results from the current study shows that Dorper-sired lambs have 30% lower WBSF values than Suffolk-sired lambs. Burke et al. (2003) also found that Longissimus chops from Dorper, Katahdin, and St. Croix hair breeds had lower (more tender), shear force values than Suffolk lamb chops. In the current study, tenderness levels potentially improved for the wool group as they had higher levels of flank streaking and improved quality grades over the hair group. In addition, only one ram was used for each of the groups causing limited genetic variability. Finaly, the lambs were of similar age, fed the same diet and all had comparable ADG, these similar management practices and growth rates potentially are some reasons for similar tenderness levels in the chops.

Consumer sensory analysis

Consumer demographics (Table 7) show that over 65% of panelists ate lamb less than five times per year. However, there were a high number of consumers (76.8%) willing to purchase the samples from the taste panel with no differences between groups (P = 0.254; Table 8). Positively, there were also low numbers (12.8%) of consumers who felt the samples had an off-flavor, with no differences being detected between groups (P = 0.759; Table 8). Consumer demographics are subject to change based on the location of the consumer sensory analysis.

Table 7.

Consumer sensory analysis panelist demographics summary statistics (n = 74)

Age	n	%
18–19	7	9.5
20–29	43	58.1
30–39	10	13.5
40–49	2	2.7
50+	12	16.2
Gender
Male	30	40.5
Female	44	59.5
Ethnicity
Caucasian	63	85
Hispanic/Latino	6	8.1
Asian	2	2.7
American Indian	1	1.4
Black	1	1.4
Not indicated	1	1.4
Lamb consumed per year1
0	16	21.6
1–4	32	43.3
5–8	10	13.5
9+	16	21.6
Form most consumed2
Chop	31	41.9
Roast	14	18.9
Ground	5	6.8
Other	14	18.9
Not indicated	10	13.5

For ¹ and ² consumers were asked:

¹Please indicate the number of meals a year in which you consume lamb.

²Please indicate the form in which you most commonly consume lamb.

Table 8.

Consumer sensory analysis percentages for consumer satisfaction between lamb breed groups (composite, hair, and wool)

	n	%
Willingness to purchase1		(P =0.254)
Yes	149	76.8
No	45	23.2
Off flavor2		(P = 0.759)
Yes	25	12.8
No	170	87.2

¹Consumer satisfaction: Would you be willing to purchase this product? no/yes.

²Off-flavor: This is your based on your ability to detect an off-flavor of the sample: no/yes.

Consumer taste panel results are shown in Table 9. There were no differences between the groups for ratings on acceptability (P = 0.295; Table 9). Monaco et al. (2014), however, showed that Dorper/Santa Inez and Suffolk samples were given some of the highest accepted sensory evaluations. This shows that the Dorper influence improves Santa Inez sensory panel acceptability. Differences were also not observed between groups regarding tenderness, however, showed that Dorper/Santa Inez and Suffolk samples were given some of the highest accepted sensory evaluations. This shows that the Dorper influence improves Santa Inez sensory panel acceptability. Differences were also not observed between groups regarding tenderness(P = 0.254) which was expected due to the finding of no differences between groups for WBSF. The Snowder and Duckett (2003) study differed in this result, showing that there was an increased tenderness rating for consumer taste panel in Dorper-sired lambs than in Suffolk-sired. Notter et al. (2004), in agreement with Snowder and Duckett (2003) also found that trained sensory panelists rated Dorper-sired lamb chops to be more tender than Suffolk and Dorset-sired lambs. Including a trained sensory panel in future studies could illuminate differences between breeds.

Table 9.

Estimated mean effects of consumer sensory analysis between lamb breed groups (composite, hair, and wool)

	Groups			SEM	P value
Traits	Composite	Hair	Wool
Acceptability	7.4	7.0	7.4	0.2	0.295
Tenderness	6.6	6.6	7.2	0.3	0.254
Juiciness	6.7	6.5	7.1	0.3	0.174
Flavor	7.0	6.9	6.6	0.2	0.435

¹Scale, 10 = like extremely, extremely tender, extremely juicy, and like flavor extremely, respectively; 1 = dislike extremely, not at all tender, extremely dry, and dislike flavor extremely, respectively.

Differences between groups for juiciness were not detected (P = 0.174). The Dorper-sired lambs also showed higher panelist ratings for juiciness in Snowder and Duckett (2003) contrasting with the results from the current study. There were no discernable differences between groups for flavor (P = 0.435). It was previously noted lipid oxidation on day 4 surpassed the threshold for consumers to detect off-flavor. Panelists were provided day 0 samples. Further research to determine if increased off-flavor detection would occur, after day 4 of retail shelf-life is needed. Consumer scores averaged between 6 (liked slightly) and 7 (liked moderately) for all traits: acceptability, tenderness, juiciness, and flavor. The study by Snowder and Duckett (2003) is in agreement with these results, portraying similar flavor intensity scores among Dorper-sired, Columbia-sired, and Suffolk-sired lambs. Cramer et al. (1970) also showed no differences in flavor intensity due to breed or sire breed.

Volatile compound analysis

Summary results for volatile compounds are recorded in Table 10. The two major pathways for volatile compound creation are lipid oxidation reactions and Maillard reactions (Mottram, 1998). A meta-analysis by Watkins et al. (2013) was conducted looking at lamb aroma. Fifteen volatiles were found to be the top impact compounds for baseline lamb aromas, these included 2-acetyl-1-pyrroline (popcorn, roasted), (E,E)-2-4-decadienal (fatty, fried), chlorine), decanal/2,4-(E,E)-heptadienal (roast meat, potato), 2,3-diethyl-5-methypyrazine (nutty, roasted), dimethyl trisulfide (sulfur), -4-ethyloctanoic acid (mutton-like), furaneol (caramel), €-2-heptenal (fish, fried), methional (cooked vegetables, potato), 4-methylphenol (stable, animal), (Z)-2-nonenal (plastic, octanal (lemon, floral), 1-octen-3-one (mushroom, earth), and (E)-2-octenal (grass).

Table 10.

Estimated mean effects of aroma compounds found in the cooked Longissimus dorsi headspace (mg/g) and fatty acid composition of total lipids (mg/g) from groups

	Groups			SE	P value
Traits	Composite	Hair	Wool
Volatile compounds1
Octane	8.87a	7.96a	6.70b	0.57	0.022
Tetradecane	0.32b	0.52a	0.48a	0.05	0.019
Phenylacetaldehyde	3.08b	3.18a	3.18a	0.02	0.005
Fatty acids2
C18:1trans	0.33b	0.24b	0.53a	0.05	0.004
C18:2n–6	1.16b	1.21b	1.43a	0.07	0.029
C20:1n–5	0.06ab	0.04b	0.08a	0.01	0.045
C20:1n–8	0.08ab	0.06b	0.10a	0.01	0.027
C20:5	0.03ab	0.05a	0.02b	0.01	0.022
C22:4	0.01b	0.02ab	0.03a	0.003	0.026
C24:1n–9	0.01b	0.02b	0.03a	0.003	0.011
C18:0	3.14	2.92	3.04	0.21	0.767
C16:0	5.54	4.58	5.78	0.42	0.126
Total SFA	9.71	8.38	9.89	0.66	0.244
Total MUFA	9.01	7.84	9.94	0.68	0.125
Total PUFA	2.10	2.27	2.37	0.13	0.347

^a,bWithin a row and trait, means without a common superscript differ (P < 0.05).

¹There were 67 compounds were detected, 3 showed differences.

²There were 45 fatty acids detected, only 7 showed differences.

In the present study, there were 67 volatile compounds detected for the three groups. Of these, only 3 were deemed significant. None of the volatile compounds that were of significance in this study are among the top 15 compounds that make up the lamb aroma baseline. There were differences between groups for Octane (P = 0.022). The hair and composite groups had larger values than the wool group. A study by Grabež et al. (2019) found that Octane has a positive correlation to citric acid which was determined by Lugaz et al. (2005) to be correlated to increased sour sensations as well as increasing overall gamy flavor. There were also differences between groups for Tetradecane (P = 0.019). The hair and wool groups had higher values than the composite group. Previous research has not been found discussing the effect of Tetradecane on lamb volatiles.

There were differences between groups for Phenylacetaldehyde, an aldehyde (P = 0.005). This study found that hair and wool groups had higher values showcasing favorable composition for phenylacetaldehyde. One study by Wang et al. (2021) found lambs of 2–6 mo had phenylacetaldhyde as one of the main flavoring compounds. The above study defines phenylacetaldhyde as having a honey flavor with a low odor threshold, Wang et al. (2021) also found as the animal aged, fatty acid composition changed causing different aromas and flavors. Aldehydes such as phenylacetaldehyde, are some of the main components for meat flavor, most have a strong flavor with a low flavor threshold to produce stronger off-flavors or rancidity over time (Elmore et al., 2005; Calkins and Hodgen, 2007). The animals were of similar age and maintained the same diet, minimizing outside effects on the compounds caused by anything other than breed differences.

Fatty acid methyl esters

Analyses detected 45 fatty acids for the three groups, summary results for the fatty acids are recorded in Table 10. There was limited evidence for breed type differences for the fatty acid analysis with only seven fatty acids being detected as significant. There were differences between groups for omega-6 fatty acid, Linoleic acid, C18:2n–6 (P = 0.029), Elaidic acid, C18:1trans (P = 0.004), Nervonic acid, C24:1n–9 (P = 0.011), and long-chain omega-6 fatty acid. The wool group had higher levels of all the above acids than the other two groups. There were differences found between groups for Adrenic acid, C22:4 (P = 0.026). The wool group had a higher value than the composite group. This is contrary to the study by Fisher et al. (2000) who found a high correlation between 18:3 (positive) and 18:2 (negative) PUFA with flavor intensity for taste panelists. When comparing fatty acid levels with the taste panel results, the current study found no flavor intensity differences. In contrast with the previous study, the wool group had an increased omega-6 linoleic acid, C18:2n–6 presence compared to the other groups with no negative effects seen for consumers. Numerically there was a difference in the amount of 18:2n–6 seen between the trials, with Fisher et al. (2000) Suffolk-fed concentrates reached 1.88 mg/g whereas for this study the Suffolk only reached 1.44 mg/g, this potentially played a role in the decreased impact on consumer flavor analysis. There were differences between groups for the omega-3 unsaturated fatty acid eicosapentaenoic acid, C20:5 (P = 0.022). The hair group had a higher value than the wool groups.

There were differences between groups for eicosenoic acid C20:1n–5 (P = 0.045) and C20:1n–8 (P = 0.027). The wool group had higher values than the hair group. There were no differences between the groups for stearic acid, 18:0 (P = 0.767) or palmitic acid, 16:0 (P = 0.126). These results are agreeable with Webb and Casey (1995) who found no breed differences between Dorper and South African Mutton Merino breeds in subcutaneous adipose tissue for stearic and palmitic acids. This is notable as Sañudo et al. (2000) demonstrated that stearic acid is heavily correlated to acceptability for flavor. These results are unlike a study by Snowder and Duckett (2003), which compared Dorper and Suffolk-sired lambs from Columbia ewes, who found that Dorper-sired lambs have lower levels of palmitic (16:0) and palmitoleic (16:1) acids in the Longissimus muscle. It also found conversely, that Dorper-sired lambs had a 6.8% greater concentration of stearic acid (18:0). The concentration of total muscle SFA (P = 0.244), MUFA (P = 0.125), and PUFA (P = 0.347), did not differ significantly across breed types to influence consumer perception. Other studies (Duckett and Greiner, 2005; Zhang et al., 2020) found Dorper-sired lambs had lower levels of MUFA’s than other breeds including Suffolk-sired, and higher levels of PUFA that may negatively impact the odor profile than Tan and Hu. There is notable variation in fatty acid profiles between studies. Cruz-Sanchez et al. (2022) determined that breed, sex, and lambing type can impact MUFA and PUFA levels in lamb. Type of feed can also have an effect on fatty acid levels in lamb (Duckett and Kuber, 2001). The current study fed all lambs on the same diet potentially limiting noticeable differences.

Conclusion

A strategy to improve the carcass characteristics and flavor profiling of lamb is needed to aid in enhancing production and thereby improving the consumption of lamb in the United States. Previous studies have shown increased benefits to carcass size and improved flavor traits for consumers utilizing crossbreeding between hair and wool lambs. The current research found minimal improvements of benefits in the wool and hair crossbred composite group when compared to the purebred hair lambs for growth and certain carcass characteristics and no improvements for flavor characteristics. It requires noting that there were likely some limitations to the study due to the small number of experimental units in each treatment group. However, the current research can serve as a base for future investigation of wool and hair composite crossbreeding. Additional research is required to develop a more effective composite breed to optimize carcass quality and flavor characteristics to improve profits for producers and palatability for consumers. This is not a beneficial cross (Dorper × Polypay/Targhee) to the industry due to being lighter weight and poorer muscled, as it is unable to compete with that of the larger framed meat breeds. Finally, there was no added benefit of improved acceptability for consumer and retail perception.

Glossary

Abbreviations

ADG

average daily gain

AMSA

American Meat Science Association

BCTRC

boneless closely trimmed retail cuts

BF

back fat

ERS

economic research service

FAMES

fatty acid methyl esters

HCW

hot carcass weight

IACUC

Institutional Animal Care and Use Committee

IMPS

Institutional Meat Purchase Specifications

IRB

Institutional Review Board

MDA

Malondialdehyde

NAMP

North American Meat Processors Association

NASS

National Agricultural Statistical Service

NIR

Near Infrared Reflectance Spectroscopy

REA

rib-eye area

QG

quality grade

SAS

statistical analysis system

TBARS

thiobarbituric acid reactive substance

TMR

total mixed rations

U.S.

United States

USDA

United States Department of Agriculture

WBSF

Warner–Bratzler shear force

YG

yield grade

Funding statement

This research was funded by the National Sheep Industry Improvement Center (Award Number: AG5072). We gratefully acknowledge the financial support that was provided by them. Support for this research project was also provided by the Idaho Agriculture Experiment Station.

Conflict of interest statement

There is no known conflict of interest.