A mapping method for the description of Warner–Bratzler shear force gradients in beef Longissimus thoracis et lumborum and Semitendinosus

Meat Science 72 (2006) 79–90 www.elsevier.com/locate/meatsci

A mapping method for the description of Warner–Bratzler shear force gradients in beef Longissimus thoracis et lumborum and Semitendinosus J.A.M. Janz

a,¤

, J.L. Aalhus b, M.E.R. Dugan b, M.A. Price

a

a

b

Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB, Canada T6G 2P5 Meat Research, Agriculture and Agri-Food Canada, Lacombe Research Centre, 6000 C&E Trail, Lacombe, AB, Canada T4L 1W1 Received 24 January 2005; received in revised form 6 June 2005; accepted 6 June 2005

Abstract A novel approach to mapping Warner–Bratzler shear of whole muscles was explored. The procedure was used on the Longissimus thoracis et lumborum (LTL) and Semitendinosus (ST) from six beef cattle by Wrst marking skeletally deWned anatomical landmarks on the muscle in situ. After removal from the carcass, further divisions were made while preserving sample orientation during cooking and preparation for shearing. Shear gradients were observed in all planes of the LTL, particularly the medial–lateral. The midsection of the ST had the lowest shears while superWcial locations of the cross-section had greater values. Muscle comparison indicated the ST was more uniform than the LTL. The mapping technique was subsequently used to identify localized eVects of altered carcass suspension on shear values and sarcomere length in the lumbar Longissimus from four beef cattle. This mapping method will provide guidance for further intensive investigation across the carcass musculature and under varying carcass conditions.  2005 Elsevier Ltd. All rights reserved. Keywords: Shear force mapping; Longissimus thoracis et lumborum; Semitendinosus; Beef; Altered suspension

1. Introduction National beef quality surveys conducted in Canada and the United States have indicated a deWciency in meeting consumer expectations, particularly for tenderness. Across a selection of chuck, loin, and round steaks evaluated during the Beef Consumer Satisfaction Benchmark Study (Beef Information Centre, 2002), only 68% of consumers were satisWed (score of 77 out of 10) with tenderness. For the inside round (Semimembranosus) alone, this value fell to 55%, while striploin (Longissimus * Corresponding author. Present address: Institute of Food, Nutrition and Human Health, Massey University, Private Bag 11 222, Palmerston North T6G 2P5, New Zealand. Tel.: +64 6 350 4336; fax: +64 6 350 5657. E-mail address: [email protected] (J.A.M. Janz).

0309-1740/$ - see front matter  2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.meatsci.2005.06.009

lumborum) steaks provided 82% of consumers with tenderness satisfaction. In the U.S. National Beef Tenderness Surveys conducted in 1990 and 1998 (Morgan et al., 1991; Brooks et al., 2000), 87.2% and 55.9%, respectively, of eye of round (Semitendinosus) steaks had a 68% chance of receiving a “slightly tough” sensory rating. National surveys provide strong indications from consumers about what makes a desirable product. For example, when consumers were asked to provide a reason for their less than ideal eating experience, tenderness complaints were most prevalent, and 88% of consumers cited “product deWciencies” as the cause (Beef Information Centre, 2002). Meat quality research can address consumer concerns by deWning the inherent muscle characteristics that might guide speciWc treatment of individual muscles to maximize their most positive qualities and to enhance those muscles that require improvement.

80

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90

Of work conducted since the mid-1960s describing beef tenderness trends, Segars, Nordstrom, and Kapsalis (1974) presented the most comprehensive analysis, using the entire length of each muscle of interest, and evaluating both longitudinal and cross-sectional trends. Evaluations of longitudinal trends have generally been based on three to six steak locations along the muscle (Henrickson & Mjoseth, 1964; Jeremiah & Murray, 1984; Gariépy, Jones, & Robertson, 1990; Wheeler, Shackelford, & Koohmaraie, 1996), in some cases restricted to only the lumbar portion of the Longissimus (Martin, Fredeen, & Weiss, 1970; Williams, Field, & Riley, 1983). As a part of their review of 40 diVerent bovine muscles, however, Belew, Brooks, McKenna, and Savell (2003) provided more extensive longitudinal sampling, focusing on several muscles for which this type of evaluation had not previously been conducted (for example, Psoas major, Gluteobiceps, Infraspinatus, Pectoralis profundus). In the cross-section, reporting of diVerences in the superWcial to deep plane has been rare (Smith, Carpenter, & King, 1969). Research tended to focus on medial to lateral diVerences, with most based on an evaluation of two to four core samples across each steak (Cover, Hostetler, & Ritchey, 1962; Hostetler & Ritchey, 1964; Alsmeyer, Thornton, & Hiner, 1965; McBee & Wiles, 1967; Hedrick, Stringer, Epley, Alexander, & Krause, 1968), although some studies made use of as many as Wve or six cores (Smith et al., 1969; Crouse, Theer, & Seideman, 1989; Berry, 1993). Investigations speciWc to the Semitendinosus have been minimal. Henrickson and Mjoseth (1964), Shackelford, Wheeler, and Koohmaraie (1997), and Denoyelle and Lebihan (2003) examined three to four locations along the length of the muscle, and Belew et al. (2003) and Torrescano, Sánchez-Escalante, Giménez, Roncalés, and Beltrán (2003) provided a mean value for whole ST for comparison to a variety of other muscles. Only Reuter, Wulf, and Maddock (2002) have made an in depth investigation. What characterizes the majority of these early reports is the limited number of muscle locations used for testing, along with the underlying assumption that these results could be extrapolated across the entire length and width of the muscle. A new trend in the comprehensive assessment of beef tenderness, however, has appeared, and is characterized by three recent reports describing detailed mapping methods. Kerth, Montgomery, Lansdell, Ramsey, and Miller (2002) examined striploins from 320 animals, and reasonably extensive sampling was performed in the muscle cross-section, however, only two longitudinal locations, restricted to the lumbar portion of the Longissimus, were studied. Zuckerman, Berry, Eastridge, and Solomon (2002) also studied the lumbar Longissimus, but conducted more extensive sampling with a greater number of steaks per muscle and more shear samples per steak

across seven muscle pairs. Unfortunately, shear values along the length of the muscle section were averaged at each shear measurement location and the opportunity to study longitudinal trends was lost. Reuter et al. (2002) mapped along the length of each of four muscles in the round from 10 animals, but averaged within the crosssection by zone or by steak. So, no single study examined all three dimensions of the given muscles, and none preserved anatomical relationships. Furthermore, all studies used aged product that was also frozen at some point prior to shear force evaluation, thus, introducing the potential to alter or mask inherent shear characteristics by permitting a period of postmortem proteolysis and a damaging freeze/thaw cycle (Geesink, Mareko, Morton, & BickerstaVe, 2001). The objective of this work was to describe a mapping method (Experiment 1) that could be used as a tool for the thorough examination of individual muscles by maintaining three-dimensional orientation and a relationship to anatomical landmarks. The practical use of the technique was demonstrated in the lumbar portion of the Longissimus to examine the eVects of altered carcass suspension (Experiment 2), an established treatment known to improve instrumental tenderness, on both shear values and sarcomere lengths.

2. Materials and methods 2.1. Experiment 1: Mapping method development All animals were British-type beef cattle Wnished on a grain/silage diet to 16–17 months of age and a mean liveweight of 582.2 kg § 11.1 SEM. On the morning of each of six kill days, one heifer was transported from the Lacombe Research Centre beef unit to the research abattoir. After carcass dressing, all adipose, lean, and connective tissues overlying the Longissimus thoracis et lumborum (LTL) and Semitendinosus (ST) muscles were removed from both left and right sides. This served to expose the length of both muscles (from the caudal side of the sixth lumbar vertebra to the cranial side of the fourth thoracic vertebra in the LTL; from distal to proximal attachment points in the ST) such that temperature decline under refrigeration conditions would be as equivalent as possible across all carcasses and sides. In order to moderate the carcass cooling conditions and reduce the potential for cold shortening in the exposed muscles, chilling commenced in a cooler maintained near 8.5 °C (range from 7.9 to 9.2 °C). At 7.5 h postmortem, sides were moved to a cooler held at conventional chilling temperature (mean 0.8 °C; range from ¡0.4 to 2.0 °C) for the remainder of the cooling period. At intervals throughout carcass chilling (0, 4, 6, 8, and 24 h), pH of the LTL was measured using an Accumet 1002 pH meter (Fisher ScientiWc, Edmonton, AB) and

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90

Orion Ingold spear-type electrode (Udorf, Switzerland). Measurements of pH were made along the mid-line of the LTL at the approximate mid-point of muscle depth (2–3 cm varying with longitudinal location), at four locations corresponding to the sixth lumbar (L6), second lumbar (L2), eleventh thoracic (T11), and fourth thoracic (T4) vertebrae. At these same locations, as well as at distal, mid-muscle, and proximal locations along the length of the ST, data logger probes (SAPAC Temprecord III ScientiWc Loggers, Sumaq Distributors, Toronto, ON) were inserted to record internal muscle temperature during carcass chilling. At 24 h postmortem, the LTL and ST were marked with food grade ink to identify the mid-line, the medial and lateral edges of the muscle, as well as the caudal side of each vertebrae from L6 to T4, inclusive, along the length of the LTL. Both muscles then were removed from both left and right carcass sides. Once removed from the carcass, the LTL was cut into steaks (4 cm thickness), beginning at the Wfth lumbar marking and proceeding to the fourth thoracic marking, discarding sections used for temperature and pH measurements. From the ST, 6 cm of lean tissue was removed and discarded from the extreme distal end, and 7 serial steaks were cut at 4 cm intervals, again, avoiding sections damaged by pH and temperature collection. The remaining portion of the ST, about 9 cm at the proximal end of the muscle, was also discarded. In order to mark a common, cross-sectional centre point across all steaks, the junction of the mid-line and the medial and lateral edges was marked on the cranial (LTL)/proximal (ST) face of the cross-sectional surface (Fig. 1). Steaks were cooked in a forced air convection oven (Bakers Pride, New Rochelle, NY) with thermocouples inserted through the medial side of each steak to continuously monitor internal temperature during cooking (Hewlett Packard 34970A Data Acquisition Switch Unit

Superficial Y axis

Medial

Central point of origin

Lateral

X axis

Deep

Fig. 1. Orientation of steaks for shear sample preparation around the central origin relative to X and Y axes. Oval represents the whole steak. Short vertical and dotted horizontal lines represent medial–lateral and superWcial–deep zones, respectively, used in locational analyses within the Longissimus thoracis et lumborum. Regions between axis names (medial, superWcial, lateral, deep) represent quadrant locations used in analysis of Semitendinosus.

81

with Benchlink software, Loveland, CO). Steaks were placed cranial (LTL)/proximal (ST) face upwards, on wire cooking racks to permit air circulation, in an oven preheated to 177 °C and removed when internal steak temperature reached 72 °C (LTL) or 70 °C (ST). Once removed from the oven, all steaks were placed in individual plastic bags and submerged in an ice bath to arrest cooking. When the steaks were cooled to 35–40 °C, purge was poured from the bags and samples were refrigeration at 4 °C until the following day. With the cranial/proximal surface facing upwards, each steak was divided into serial, rectangular shear samples. Using a twin scalpel with blades set at a 15 mm width, the surface of each steak was scored in a grid pattern that originated at the pre-marked centre point: the intersection of the X (muscle width: medial to lateral) and Y (muscle depth: superWcial to deep) axes (Fig. 1). Following the surface markings, each steak was cut into strips parallel to the Y-axis. Further division followed the grain of the muscle Wbres. All tissue within the epimysium and which yielded a complete 15 £ 15 mm sample was used for shear analysis and this provided 10–30 samples from each LTL steak and 15–25 from ST steaks. Samples were arranged on a tray lined with a coded grid in order to maintain locational orientation. Shearing was completed using an Instron 4301 Materials Testing System equipped with a Warner–Bratzler shear force cell and Series 9 software. Cross-head speed was set to 200 mm min¡1 and peak load was recorded in kgf then converted to N. 2.2. Experiment 2: Altered carcass suspension On the morning of each of four kill dates, one beef steer (average liveweight 652.1 kg § 10.5 SEM) was processed as described in Experiment 1 with the exceptions that the tissues overlying the Longissimus lumborum (LL) remained intact during carcass cooling and the right side of each carcass was suspended by the obturator foramen. All sides, regardless of suspension method, were chilled for 24 h in a cooler maintained at an average temperature of 1.8 °C. The following morning, the LL on each carcass side was stripped of fat, marked, removed, and divided in to 10 steaks, as described in Experiment 1. Beginning at the caudal end of the LL, steaks from both suspension treatments were assigned for measurement of sarcomere length (raw steaks) or Warner–Bratzler shear force (cooked steaks), respectively. With the cranial face upwards, each raw steak was scored with a double blade scalpel with the blades set 15 mm apart. Score marks began at the central origin and moved both medially and laterally to mark four intervals on either side of the origin for a total of 8 medial–lateral zones. A 2 g sub-sample from the centre of each zone was prepared and homogenized with 20 ml

82

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90

of an isotonic solution (0.25 M sucrose and 20 mM EGTA) for 10 s following the method of Aalhus, Best, Costello, and Jeremiah (1999). A wet mount slide of each homogenate was prepared and viewed under 1000£ magniWcation using phase contrast microscopy (Zeiss Axioskop, Germany). Six images from each homogenate were captured electronically (Sony 3 CCD camera with computer connection). Image analysis (Image Pro Plus, Media Cybernetics, Silver Spring, MD) was used analyse the repeating striated pattern (Aalhus et al., 1999; Devine, Walhgren, & Tornberg, 1999; Tornberg, Walhgren, Brøndum, & Engelsen, 2000) measured in pixels then converted to micrometres for statistical analysis. Cooking and shear force measurement followed the same procedures described for Experiment 1. 2.3. Data preparation and statistical analysis Prior to statistical analysis in Experiment 1 (mapping method), data from right side muscles were mirrored such that medial and lateral locations between sides were aligned. Exploratory analysis of location factors (general linear model procedure, GLM; SAS/STAT, 1990) indicated no signiWcant diVerence between sides, hence left and right data were merged prior to further data analysis. The univariate and means procedures (SAS, 1985) were applied to LTL and ST data sets to generate basic statistics for comparison of the two muscles. To permit closer examination of longitudinal sections of the LTL, shear data from the entire length of the muscle were divided into three “blocks” following analysis with the GLM procedure of SAS to determine appropriate break points (Fig. 2). Due to the nature of LTL muscle shape, with variation in cross-sectional area along its length, the muscle was further subdivided into “zones” to permit analysis of medial–lateral and superWcial–deep

shear trends. Three superWcial–deep zones were established within each block, while the number of medial– lateral zones varied from Wve to ten according to the width of steaks in each block. The GLM procedure was used for analysis of both shear and standard deviation (Aalhus, Dugan, Robertson, Best, & Larsen, 2004) data within blocks and zones of the LTL. The relatively circular cross-section of the ST permitted grouping of shear data by quadrants arranged around the pre-marked centre point. Analysis of shear and standard deviation data with the GLM procedure included steak, quadrant, shear sample location, and appropriate interactions in the model. Locational eVects in both muscles were considered signiWcant at the P 6 0.05 level and means separation for signiWcant eVects was completed with the probability of diVerence function. LTL pH data were analysed using a GLM that included location within the muscle and time postmortem. Warner–Bratzler shear and sarcomere length data from Experiment 2 (altered carcass suspension) were analysed according to a factorial design, using the GLM procedure of SAS, and included suspension treatment, steak location, medial–lateral zone, and appropriate interactions. EVects were considered signiWcant at the P 6 0.05 level with subsequent means separations completed with the probability of diVerence function. Correlation analysis (SAS, 1985) was also conducted to investigate the relationship between sarcomere length and Warner–Bratzler shear force at various crosssectional locations in the muscle.

3. Results and discussion 3.1. Experiment 1: Mapping method 3.1.1. Longitudinal trends in the Longissimus thoracis et lumborum Along the longitudinal axis of the LTL, shear was greatest in the central block (L1/T13 to T10/9, including the Canadian beef carcass grading site, T12/13), lowest in the thoracic block (T9 to T5), and intermediate in the lumbar block (L5 to L2/1) (Table 1). The range of shear values Table 1 Warner–Bratzler shear values (N) and standard deviations (SD; N) for longitudinal locations of the Longissimus thoracis et lumborum in Experiment 1

Fig. 2. Orientation of steaks and longitudinal block groupings with vertebral locations.

Longitudinal location

Shear

SEMy

SD

SEMz

Lumbar Central Thoracic

112.38a 115.42b 105.81c

0.78 0.59 0.78

23.93a 20.79b 16.87c

7.55 7.45 0.98

a,b,c: Values in the same column followed by diVerent letters are signiWcantly diVerent (P < 0.05). y SEM of shear value. z SEM of standard deviation.

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90

Shear SD

Shear (N)

120 b

b

b a

110 a

100

25 20

115

105

30

b

b

15

a

10

a a

Lumbar

SD (N)

125

83

5 Central Longitudinal locations

Thoracic

0

Fig. 3. Warner–Bratzler shear values and standard deviations (SD) of steaks within longitudinal block locations in the Longissimus thoracis et lumborum in Experiment 1.

amongst steaks within the longitudinal blocks is demonstrated in Fig. 3. Henrickson and Mjoseth (1964) reported very similar results demonstrating that the Longissimus was toughest near T13, became more tender towards T7, and was intermediate in tenderness between L6 and L2. In an examination of the posterior half of the Longissimus, Martin et al. (1970) reported contrasting data indicating lower shears near T11/12 with toughness increasing towards the lumbar region. Gariépy et al. (1990) also reported the lowest shears in the central (T12 to T8) versus thoracic (T7–T4) and lumbar (L5–T13) regions. Sampling in that study was limited, however, to the examination of one steak at each longitudinal location from each of which only three cores were sheared. Yet other reports have indicated no eVect of longitudinal location (Williams et al., 1983), however, limited sampling (Wheeler et al., 1996) and the use of taste panel tenderness evaluation (Jeremiah & Murray, 1984) hindered comparison with the present results. Variability within longitudinal blocks (Table 1) was greatest in the lumbar region and decreased towards the thoracic end of the muscle. Wheeler et al. (1996) also reported a lower shear standard deviation in the thoracic region. Henrickson and Mjoseth (1964), however, indicated the greatest uniformity in shear was found in the lumbar location and recommended the use of that portion of the muscle for meat quality investigation. Longitudinal variation in shear values may be related to variability in the degree of contraction experienced during rigor mortis development at diVerent locations along the length of the LTL. Unpublished time-lapse video data from our lab indicate a diVerential rigor contraction pattern along the length of the LTL during conventional chilling. Gariépy et al. (1990) also suggested diVerences in muscle restraint due to shackling during carcass dressing, carcass cooling, muscle Wbre angle, metabolic Wbre type, and connective tissue content as possible causes of intramuscular variation. 3.1.2. Cross-sectional trends in the Longissimus thoracis et lumborum Prior to carcass chilling, tissues overlying the LTL were removed to ensure that muscle chilling rate, and

Fig. 4. Internal temperature decline during carcass cooling at various lumbar (L) and thoracic (T) locations along the Longissimus thoracis et lumborum in Experiment 1.

any subsequent eVect on tenderness/toughness, would be as similar as possible amongst all muscles and not aVected by diVerences in the depth, hence insulative value, of extraneous tissues. This procedure was undertaken knowing that it could increase the risk of cold shortening. While an attempt was made to moderate the carcass chilling environment, the temperature curves (Fig. 4) demonstrate that internal muscle temperature narrowly entered the region (below 10 °C before 10 h postmortem) commonly recognized to induce cold shortening. Since all the tissue overlying each muscle was removed, however, cold shortening was a risk along the entire muscle length, rather than in localized regions. As such, the observed gradients in shear values may have been exaggerated, as compared to a situation where cold shortening was not a risk, but not transformed. Muscle depth along the length of the LTL varied from 3–4 cm at the extremities to 5–6 cm at the central locations. A greater initial rate of temperature decline was observed in the thoracic region while temperature in the central area declined more gradually (Fig. 4). If cold temperature exposure was a factor in toughness/tenderness development, one would expect higher shear values at the superWcial location while the deep muscle location would have lower shear values since cold shortening (Locker & Hagyard, 1963; Marsh & Leet, 1966) would be limited with an increased insulative eVect of greater muscle thickness. While this statistically signiWcant diVerence was observed in the superWcial to deep crosssection of the lumbar region (Fig. 5(a)), the diVerence in mean shear value between the superWcial and deep locations was less than 1 kg (»9.8 N), the threshold for consumer detection (Aalhus et al., 1999). In the central and thoracic blocks, the trend was reversed with a signiWcant gradient of increasing shear value from the superWcial to deep locations indicating that the expected eVect of temperature on shear value did not occur. The trend in variability of shear values (Fig. 5(b)) in the superWcial–deep

84

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90 140 aa

b b

c

a

a

100

150

c b

140 Superficial

80

Central

60

Deep

40 20 0

Lumbar

a

SD (N)

25

a

c

110

a

c

b

c

a

de

5

6

de

de a

3

a

b

bd

Central b

c

e

de

de

Thoracic

d

4

7

8

9 10 Lateral

Cross sectional location 40

Central Deep

10

35 a

5

30 Lumbar

Central Thoracic Longitudinal location

SD (N)

b

b

1 2 Medial

Superficial

15

Lumbar

ad c

90

b

a

20

0

120

b b

b ad

100

Central Thoracic Longitudinal location

30

a

130 Shear (N)

Shear (N)

120

cd

25

Lumbar bcd

Central bc

bc

bc

20

Fig. 5. Warner–Bratzler shear gradient (a) and standard deviation (SD; b) for each longitudinal location across the superWcial–deep cross-section in the Longissimus thoracis et lumborum in Experiment 1.

dimension followed mean values with larger shear values having greater variability as indicated by standard deviation. This eVect was similar to that reported by Dugan and Aalhus (1998) who indicated a strong positive correlation between the two factors. Since some of the current results do not seem to be consistent with cold shortening theory, tension development during rigor onset may have had a greater impact on cross-sectional shear gradient than temperature during carcass cooling. This is particularly evident upon examination of the medial to lateral dimension. In the medial to lateral cross-section, shear diVerences were signiWcant and unique to each longitudinal block (Fig. 6(a)). In the lumbar region there were substantially larger shear values on the medial side. With a move towards the lateral edge, however, the shear values became lower and the magnitude of diVerence between adjacent sampling locations was reduced. In the central portion of the LTL, the shear gradient was somewhat sigmoidal with multiple peaks and valleys traversing the cross-section. Although, Zuckerman et al. (2002) provided no deWnition of the longitudinal location within the Longissimus, the surface plot they presented demonstrated a trend very similar in its serpentine pattern to the present results. Also in the current results, shear values in the thoracic area displayed a steady increase from the medial to the lateral positions. No signiWcant diVerence in standard deviation was observed in the medial– lateral zones of the central and thoracic blocks of the LTL; however, the lumbar region displayed signiWcant variability with a gradient declining from medial to

ad acd

Thoracic b

15

b

10 1 2 Medial

b

3

4

5

6

7

8

9 10 Lateral

Cross sectional location

Fig. 6. Warner–Bratzler shear gradient (a) and standard deviation (SD; b) for each longitudinal location across the medial–lateral crosssection in the Longissimus thoracis et lumborum in Experiment 1.

lateral (Fig. 6(b)). In general, the medial to lateral gradient was more pronounced than that observed in the superWcial to deep cross-section, an observation also reported by Kerth et al. (2002). Many published reports indicate the presence of a tenderness gradient in the medial to lateral dimension of the Longissimus, but there is considerable discrepancy about the direction of this trend. Supporting a gradient from lateral to medial are reports by McBee and Wiles (1967), Hedrick et al. (1968), Smith et al. (1969), Martin et al. (1970), Williams et al. (1983), Berry (1993), and Kerth et al. (2002). Conversely, Cover et al. (1962), Hostetler and Ritchey (1964), Alsmeyer et al. (1965), and Crouse et al. (1989) all reported a trend with larger shear values near the medial edge. Shear variability is aVected by sampling location, as well as sampling technique including carcass chilling, product ageing, cooking method, and shear sampling removal (Janz & Aalhus, 2002; Jeremiah, 2002), and each of these factors plays a role in the disparity of conclusions reported in the literature. 3.1.3. Postmortem pH in the Longissimus thoracis et lumborum While pH did vary (Fig. 7), likely due to variable chilling rates along the length of the LTL, cross-sectional

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90 6.8

pH

Cranial

Median plane

6.6 6.4

L6

6.2

L2

6.0

T11

5.8

T4

5.6

85

Myofibres

Medial

Lateral Direction of tension applied during carcass suspension Caudal

5.4 0

6

12 18 Carcass cooling time (h)

24

Fig. 7. Internal pH decline during carcass cooling at various lumbar (L) and thoracic (T) locations along the Longissimus thoracis et lumborum in Experiment 1.

diVerences in Warner–Bratzler shear, particularly in the medial–lateral direction, were larger than those observed longitudinally. This was especially evident in the lumbar region which had a pH postmortem proWle nearly identical to the mid-muscle locations during carcass chilling indicating that a factor other than pH was in eVect. In the absence of a substantial pH or temperature eVect, tension inherent to the LTL during establishment of rigor mortis appears to be a crucial inXuence on the development of the observed gradients and variability of shear values. 3.1.4. Internal tension development in the Longissimus thoracis et lumborum The LTL has a complex pattern of attachment with multiple insertion points to bone and connective tissue, and with myoWbres varying in angle relative to the longitudinal axis of the muscle (Eisenhut, Cassens, Bray, & Briskey, 1965; Jones, Burson, & Calkins, 2001). Both skeletal attachment and intramuscular Wbre arrangement aVect muscle tension postmortem and have been implicated in controlling sarcomere length and associated characteristics (Locker, 1960; Eisenhut et al., 1965). According to Koohmaraie, Doumit, and Wheeler (1996), meat toughening is due to sarcomere shortening during rigor development and in the absence of sarcomere shortening, shear force does not increase. The biomechanical function of the LTL is extension and lateral Xexion of the spine (Jones et al., 2001). During conventional carcass suspension from the Achilles tendon, tension on the Longissimus, particularly in the posterior half of the muscle, is permitted to slacken. Extension of the hip joint, hence Xexion of the spine, results in decreased strain placed on the LTL. Reduced resistance to ongoing postmortem muscle contraction increases the potential for extensive sarcomere shortening to occur as a result of rigor development. Shear values were lowest and most consistent in the thoracic region of the LTL, indicating that tension development may be fairly uniform across this location. Unpublished observations from carcass dissection con-

Fig. 8. Approximation of muscle Wbre angles in the medial to lateral cross-section of the lumbar region of the Longissimus thoracis et lumborum.

ducted in our lab indicate the oblique Wbre direction of thoracic myoWbres is uniform across the muscle, as compared to the central and lumbar areas in which Wbre angle varies in the cross-section. The smaller thoracic cross-section may also promote more consistent development of rigor tension as compared to areas with a considerably wider cross-section. Focusing on the lumbar segment of the LTL, both shear values and standard deviation were signiWcantly lower at the lateral versus medial side of the muscle. Berry (1993) reported the presence of unspeciWed, inherent properties in the lateral portion of the loin steaks that made it unlike other sections in terms of tenderness. Examination of myoWbre direction in the medial to lateral cross-section of the lumbar region revealed a shift in Wbre angle across the LTL (Fig. 8). On the medial side Wbres angled from caudal/medial to cranial/lateral. On the lateral side Wbres where nearly parallel with the longitudinal axis of the muscle. Lower shear values and reduced variability could be explained by the diVerential tension placed on the myoWbres during carcass suspension. In the medial area of the lumbar region strain is placed on the myoWbres at an angle to their long axis. The lateral Wbres experience force along their length possibly creating greater resistance to rigor contraction, and resulting in the lower shear values observed in the lateral versus medial region of the lumbar cross-section. 3.2. Semitendinosus mapping Along the length of the Semitendinosus, Warner– Bratzler shear was greatest in the distal section, lowest in mid-region, and intermediate in the proximal segment (Fig. 9), although the greatest diVerence between longitudinal locations was only 21.67 N with a 12.94 N diVerence between the extremities. Standard deviation of shear (Fig. 9) displayed a similar trend with variability the lowest in the mid-region. Investigating the “aVranchi” method of muscle trimming used in France, Denoyelle and Lebihan (2003) also indicated that the extremities of the ST were less tender. When mapping tenderness in the four major muscles of the round, Reuter et al. (2002) reported that shear values in the ST

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90

140

Shear

SD

Shear (N)

120

c

100 80

ac ab b

Surface plots indicating average cross-sections of LTL and ST locations are presented in Fig. 10. While the ST had lower mean and median shear values, likely a result of a slightly lower end-point cooking temperature, it is the variation of those values that is of particular interest (Table 3). The standard deviation for shears in the LTL was approximately twice as large as that in the ST and this is demonstrated quite clearly by the minimum–maximum range. Literature reports seem to vary on the ranking of ST shear against the Longissimus. Henrickson and Mjoseth (1964), Belew et al. (2003), and Torrescano et al. (2003) all indicated that the ST was tougher than Longissimus samples. In a similar comparison, though, Cross, Carpenter, and Smith (1973) observed lower shear values in the ST, but no diVerence between the ST and the Semimembranosus, Biceps femoris, and Rectus femoris. There seems to be solid agreement, however, that the standard deviation of shear values in the ST is consistently lower than all other muscles in these comparisons. Although generally regarded as a muscle of lower value that the LTL, the greater uniformity within the ST could be exploited by meat researchers seeking to minimize the volume of within-muscle sampling where large numbers of carcasses are included in treatment comparisons. Shear value distributions were also unique to each muscle. While both distributions tended to be slightly skewed towards higher shear values, this was more pronounced in the LTL. Furthermore, as indicated by kurtosis values, the ST distribution did not have as pronounced a peak near the mean and had more bulk in the central portion of the curve.

5

20 0

20

10

b

40

3.3. Comparison of LTL and ST

15

ac

a

60

25

SD (N)

86

a

b

c

d

e

1

2 Distal

3

4

5

c

d

6 7 Proximal

0

Steak location Fig. 9. Warner–Bratzler shear values and standard deviation (SD) of steaks across the distal to proximal plane of the Semitendinosus in Experiment 1.

were lowest in the middle and higher at the ends. Shackelford et al. (1997) drew a slightly diVerent conclusion and reported a gradient of decreasing shear force from proximal to mid to distal, with signiWcant diVerences at each level. In their experiment, however, muscles were aged 16 days and stored frozen prior to analysis. Furthermore, a Brahman inXuence was present in some of the animals and may have had an impact on connective tissue content and tenderness (Pringle, Williams, Lamb, Johnson, & West, 1997). Temperature data collected along the length of the ST during carcass cooling (data not shown) indicated that the proximal and distal extremities initially chilled at a more rapid rate than did the mid-section that had a greater tissue depth, perhaps decreasing the impact of refrigeration on muscle contraction during early carcass cooling. In the cross-section (Table 2), the largest shear values were observed in the superWcial–medial quadrant, and the lowest in deep–lateral, which also had the least variability. The superWcial–lateral and deep–medial quadrants had intermediate shears. The proximity of the deep–lateral quadrant to the musculature of the hip may have resulted in a slower cooling rate than that in more superWcial locations. The greater instrumental tenderness values of the superWcial locations may have resulted from a cold shortening eVect on the contractile apparatus, although sarcomere length was not evaluated in this experiment. Table 2 Warner–Bratzler shear values (N) and standard deviations (SD; N) for quadrant locations of the Semitendinosus in Experiment 1 Quadrant

Shear

SEMy

SD

SEMz

SuperWcial–lateral Deep–lateral SuperWcial–medial Deep–medial

105.32a 94.83b 116.01c 105.03a

0.98 0.88 0.98 0.98

15.20a 11.67b 17.26a 15.49a

0.78 0.78 0.78 0.78

a,b,c: Values in the same column followed by diVerent letters are signiWcantly diVerent (P < 0.05). y SEM of shear value. z SEM of standard deviation.

3.4. Experiment 2: Altered carcass suspension 3.4.1. Warner–Bratzler shear When the interaction between suspension method and steak location was plotted, the Warner–Bratzler shear force trend within both treatments was similar to that reported for the lumbar segment of the Longissimus in Experiment 1. Shear values increased with an increasing distance from the caudal aspect of the muscle (Figs. 3 and 11); however, the altered suspension technique resulted in lower shear values at every point of comparison to the conventional treatment (Fig. 11). Furthermore, less variation between steak locations was observed following altered suspension, with a range and standard deviation of 11.10 and 21.97 N, respectively, as compared to 31.97 and 28.34 N using conventional suspension. A signiWcant interaction also existed between suspension treatment and location in the medial to lateral cross-section. Similarly to results in Experiment 1, the

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90

87

Fig. 10. Surface plots of average Warner–Bratzler shear values in the cross-section of the lumbar (a), central (b), and thoracic (c) regions of the Longissimus thoracis et lumborum and across the Semitendinosus (d). Rotation of plots and directional axes have been varied to optimize surface view.

LTL Na Mean Median Standard deviation Minimum–maximum Range Kurtosis Skewness

2317 112.68 109.44 24.91 45.01–232.61 187.60 0.87 0.73

ST 697 104.93 106.50 12.16 77.67–130.72 53.05 ¡0.21 0.22

a

Number of shear samples evaluated across all animals; averaged across sides.

conventional suspension method resulted in a higher shear value along the medial edge of the lumbar Longissimus (Fig. 12). Aitch bone suspension acted to reduce the

140

b bc

130 Shear (N)

Table 3 Comparison of descriptive statistics for Warner–Bratzler shear values (N) from the Longissimus thoracis et lumborum (LTL) and Semitendinosus (ST)

c d

120 110

CONV a

AS c

100

bc

90 80

bc

b a

Caudal

Cranial Steak Location

Fig. 11. Warner–Bratzler shear gradient in the caudal to cranial direction of the Longissimus lumborum following conventional (CONV) or altered (AS) suspension in Experiment 2.

diVerence in shear values between the medial and lateral extremes and to reduce shear value overall, as similarly reported by Sørheim et al. (2001).

88

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90 150

Shear (N)

a

130

c

bc de

110 ab

def

d a

a

90

c

a

bcf

be

ab

b

CONV AS

c

70 1 2 Medial

3

4

5

6

7

8 Lateral

ML Zone

Fig. 12. Warner–Bratzler shear gradient in the medial to lateral crosssection of the Longissimus lumborum following conventional (CONV) or altered (AS) suspension in Experiment 2.

3.4.2. Sarcomere length The interaction between suspension treatment and medial–lateral zone location (Fig. 13) indicated that suspension method inXuenced the cross-sectional pattern of sarcomere length in the medial to lateral direction. Altered suspension created a smooth curve across the cross-sectional locations, maximizing sarcomere length in the central area. At all points of comparison, sarcomere length following altered carcass suspension was signiWcantly longer than with conventional suspension. In Experiment 1, it was speculated that variation in Wbre angle across the Longissimus in the medial to lateral direction resulted in variable tenderness across the muscles and lower shear values on the lateral side of the lumbar region (Fig. 6(a)). Variable resistance to rigor contraction was the proposed mechanism (greater on the lateral side), although sarcomere lengths were not measured in that study. Results of Experiment 2, however (Fig. 13), demonstrated signiWcantly longer sarcomeres in the lateral aspect of the lumbar Longissimus and a correspondence of longer sarcomere length to signiWcantly lower shear value (Fig. 12) in this region as compared to the medial side of the muscle within the conventional suspension treatment.

2.2 cd

SL (um)

2.0

d

d

de bce a

ab

AS c

1.6 1.4

CONV

f

1.8

a

ab

1 2 Medial

abc

3

4

abc

ab

ab

5

6

a

7 8 Lateral

ML Zone

Fig. 13. Sarcomere length gradient in the medial to lateral cross-section of the Longissimus lumborum following conventional (CONV) or altered (AS) suspension in Experiment 2.

Table 4 Correlation co-eYcients (r) between sarcomere length and Warner– Bratzler shear force in the medial to lateral cross-section of the lumbar Longissimus following conventional and altered suspension in Experiment 2 Cross-sectional location

Conventional suspension

Altered suspension

r

P

r

P

1 (Medial) 2 3 4 5 6 7 8 (Lateral)

¡0.52 ¡0.26 ¡0.20 ¡0.08 ¡0.27 0.15 0.20 0.04

0.02 0.26 0.39 0.73 0.26 0.52 0.41 0.87

¡0.39 ¡0.61 ¡0.46 ¡0.49 ¡0.44 ¡0.32 ¡0.45 ¡0.001

0.09 <0.01 0.04 0.03 0.05 0.17 0.05 0.99

3.4.3. Relationship between sarcomere length and shear force The relationship between sarcomere length and shear force was examined at each location within the medial to lateral cross-section. Within the conventional suspension treatment, the only signiWcant relationship that existed was at the extreme medial location (Table 4). When the same analysis was conducted within the aitch bone suspension treatment, Wve of eight locations yielded signiWcant relationships (Table 4). The improved relationship between sarcomere length and shear with altered suspension indicated that a greater proportion of shear variability was attributable to sarcomere length at the longer sarcomere lengths generated by aitch bone suspension. Smulders, Marsh, Swartz, Russell, and Hoenecke (1990) reported longer sarcomeres were associated with greater tenderness with nearly 30% of variability in taste panel tenderness scores accounted for by variability in sarcomere length. This group also reported that the relationship between tenderness and sarcomere length deteriorated at shorter sarcomere length. Regardless of their signiWcance, the correlation coeYcients deWning the shear/sarcomere length relationship in Experiment 2 were relatively low (Table 4), despite the importance of contraction state to tenderness/toughness development. Tarrant (1998) indicated that the straightforward relationship between toughening and actomyosin bridge formation is not always reXected in the relationship between sarcomere length and toughness possibly due to (1) variability in the strength of the bond, or (2) early acceleration of tenderization that obscures the relationship. Taylor (2003), however, presented a diVerent view indicating that actomyosin bond strength was not important, but that Wbre diameter may play a crucial role in the determination of meat tenderness or toughness. Additional investigation into the causative factors of tenderness/toughness is required, and the mapping system could be used to further understand the variability in tenderness and its contributing factors.

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90

4. Conclusions The divergence of results across previous reports on muscle characterization suggests that a limited sampling region limits the ability to make deWnitive conclusions about the tenderness gradients in a given muscle. In the past, conclusions about transverse shear gradients were often based on single locations with results extrapolated to describe longitudinal tenderness trends in a muscle of considerable overall size or length (>1 m in the LTL). In some cases, a lack of reporting on anatomical landmarks to orient sample collection within the carcass makes even casual comparison of results diYcult. The mapping technique presented in Experiment 1 was useful for describing Warner–Bratzler shear gradients within the LTL, and provided a more thorough examination of the ST than any previously published report. The procedure included extensive sampling volume using a method that was easily repeated due to the provision of speciWc anatomical landmarks. Furthermore, all cross-sectional sampling was conducted relative to a common centre point and parallel to muscle Wbre direction to aid with the reproducibility of the technique. Although mapping techniques may not be logistically feasible for use in all treatment comparison studies, they are invaluable for the discovery of inherent trends within individual muscles and can provide guidance for sampling location in meat quality research (Reuter et al., 2002; Belew et al., 2003). As expected in Experiment 2, the incorporation of altered carcass suspension reduced shear force by 25% with a concomitant 28% increase in sarcomere length. Results also showed that altered suspension improved the relationship between shear and sarcomere length, although variability in shear was not entirely accounted for by this single physical characteristic. Although other physical and biochemical traits could have been examined, sarcomere length was chosen for intensive examination to demonstrate that multiple measurements can be successfully gathered from within single muscles using the mapping method. Typically, these measurements are mutually exclusive due to their destructive nature and their diVering requirements for raw versus cooked material; however, the extensive sampling technique of the mapping method allows their collection from within the same muscle rather than across muscles or between animals. Consumers are willing to pay a premium for beef of guaranteed tenderness (Boleman et al., 1997) and the mapping results indicate that certain regions of the Longissimus and Semitendinosus can be expected to have signiWcantly lower Warner–Bratzler shear values. Reuter et al. (2002), for example, suggested that centre cut ST steaks could be marketed as “premium”. In addition, use of the mapping method under a variety of conditions and in diVerent areas of the carcass muscu-

89

lature will permit a more insightful exploration of relationships amongst tenderness and inherent biochemical and physical characteristics. Data of this type could then be applied to solve current, practical problems through the exploitation of inherent characteristics in the development of muscle speciWc tenderness enhancing strategies.

Acknowledgements We extend thanks to numerous staV at the Lacombe Research Centre for their assistance in the completion of this work: B. Starr and K. Grimson from the Beef Unit for animal care and transportation; D. Brereton and C. Pimm from the Meats Centre for slaughter and processing; R. Thacker and D. Best for laboratory advice; and F. Costello, I. Larsen, and K. Landry for assistance with sample processing. Thanks to Dr. J. Basarab, Alberta Agriculture, Food and Rural Development, for statistical guidance.

References Aalhus, J. L., Best, D. R., Costello, F., & Jeremiah, L. E. (1999). A simple, on-line processing method for improving beef tenderness. Canadian Journal of Animal Science, 79, 27–34. Aalhus, J. L., Dugan, M. E. R., Robertson, W. M., Best, D. R., & Larsen, I. L. (2004). A within-animal examination of postmortem ageing for up to 21 d on tenderness in the bovine Longissimus thoracis and semimembranosus muscles. Canadian Journal of Animal Science, 84, 30–304. Alsmeyer, R. H., Thornton, J. W., & Hiner, R. L. (1965). Some dorsal– lateral location tenderness diVerences in the Longissimus dorsi muscle of beef and pork. Journal of Animal Science, 24, 526–530. Beef Information Centre. (2002). Beef consumer satisfaction benchmark study. Available from http://www.beeWnfo.org/pdf/csr.pdf [12 January 2004]. Belew, J. B., Brooks, J. C., McKenna, D. R., & Savell, J. W. (2003). Warner–Bratzler shear evaluations of 40 bovine muscles. Meat Science, 64, 507–512. Berry, B. W. (1993). Tenderness of beef loin steaks as inXuenced by marbling level, removal of subcutaneous fat, and cooking method. Journal of Animal Science, 71, 2412–2419. Boleman, S. J., Boleman, S. L., Miller, R. K., Taylor, J. F., Cross, J. R., Wheeler, T. L., et al. (1997). Consumer evaluation of beef of known categories of tenderness. Journal of Animal Science, 75, 1521–1524. Brooks, J. C., Belew, J. B., GriYn, D. B., Gwartney, B. L., Hale, D. S., Henning, W. R., et al. (2000). National beef tenderness survey – 1998. Journal of Animal Science, 78, 1852–1860. Cover, S., Hostetler, R. L., & Ritchey, S. J. (1962). Tenderness of beef. IV. Relations of shear force and Wber extensibility to juiciness and six components of tenderness. Journal of Food Science, 27, 527–536. Cross, H. R., Carpenter, Z. L., & Smith, G. C. (1973). EVects of intramuscular collagen and elastin on bovine muscle tenderness. Journal of Food Science, 38, 998–1003. Crouse, J. D., Theer, L. K., & Seideman, S. C. (1989). The measurement of shear force by core location in Longissimus dorsi beef steaks from four tenderness groups. Journal of Food Quality, 11, 341–347. Denoyelle, C., & Lebihan, E. (2003). Intramuscular variation in beef tenderness. Meat Science, 66, 241–247.

90

J.A.M. Janz et al. / Meat Science 72 (2006) 79–90

Devine, C. E., Walhgren, N. M., & Tornberg, E. (1999). EVect of rigor temperature on muscle shortening and tenderisation of restrained and unrestrained beef M. Longissimus thoracis et lumborum. Meat Science, 51, 61–72. Dugan, M. E. R., & Aalhus, J. L. (1998). Beef tenderness: within Longissimus thoracis et lumborum steak variation as aVected by cooking method. Canadian Journal of Animal Science, 78, 711–714. Eisenhut, R. C., Cassens, R. G., Bray, R. W., & Briskey, E. J. (1965). Fiber arrangement and micro-structure of bovine Longissimus dorsi muscle. Journal of Food Science, 30, 955–959. Gariépy, C., Jones, S. D. M., & Robertson, W. M. (1990). Variation in meat quality at three sites along the length of the beef Longissimus muscle. Canadian Journal of Animal Science, 70, 707–710. Geesink, G. H., Mareko, M. H. D., Morton, J. D., & BickerstaVe, R. (2001). Electrical stimulation – when less is more. Meat Science, 57, 145–151. Hedrick, H. B., Stringer, W. C., Epley, R. J., Alexander, M. A., & Krause, G. F. (1968). Comparison of factors aVecting Warner–Bratzler shear values of beef steaks. Journal of Animal Science, 27, 628–631. Henrickson, R. L., & Mjoseth, J. H. (1964). Tenderness variation in two bovine muscles. Journal of Animal Science, 23, 325–328. Hostetler, R. L., & Ritchey, S. J. (1964). EVect of coring methods on shear values determined by Warner–Bratzler shear. Journal of Food Science, 29, 681–685. Janz, J. A. M., & Aalhus, J. L. (2002). The complexity of measuring beef tenderness using Warner–Bratzler shear methodology for whole muscle samples. In T. Nakano & L. Ozimek (Eds.), Food science and product technology (pp. 37–64). India: Research Signpost. Jeremiah, L. E. (2002). Meat cookery: A review. In T. Nakano & L. Ozimek (Eds.), Food science and product technology (pp. 1–36). India: Research Signpost. Jeremiah, L. E., & Murray, A. C. (1984). The inXuence of anatomical location within the Longissimus dorsi muscle on overall tenderness and on the concentration and solubility of intramuscular collagen. Canadian Journal of Animal Science, 64, 1045–1047. Jones, S. J., Burson, D. E., & Calkins, C. R. (2001). Muscle proWling and bovine myology. CD-ROM. University of Nebraska-Lincoln, National Cattlemen’s Beef Association, Cattlemen’s Beef Board. Available from http://bovine.unl.edu/. Kerth, C. R., Montgomery, J. L., Lansdell, J. L., Ramsey, C. B., & Miller, M. F. (2002). Shear gradient in Longissimus steaks. Journal of Animal Science, 80, 2390–2395. Koohmaraie, M., Doumit, M. E., & Wheeler, T. L. (1996). Meat toughening does not occur when rigor shortening is prevented. Journal of Animal Science, 74, 2935–2942. Locker, R. H. (1960). Degree of muscular contraction as a factor in tenderness of beef. Food Research, 25, 304–307. Locker, R. H., & Hagyard, C. J. (1963). A cold shortening eVect in beef muscles. Journal of the Science of Food and Agriculture, 14, 787– 793. Marsh, B. B., & Leet, N. G. (1966). Studies in meat tenderness. III. The eVects of cold shortening on tenderness. Journal of Food Science, 31, 450–459. Martin, A. H., Fredeen, H. T., & Weiss, G. M. (1970). EVects of sampling location and carcass fatness on tenderness of steaks from the

Longissimus dorsi of yearling Shorthorn bulls. Canadian Journal of Animal Science, 50, 235–241. McBee, J. L., & Wiles, J. A. (1967). InXuence of marbling and carcass grade on the physical and chemical characteristics of beef. Journal of Animal Science, 26, 701–704. Morgan, J. B., Savell, J. W., Hale, D. S., Miller, R. K., GriYn, D. B., Cross, H. R., et al. (1991). National beef tenderness survey. Journal of Animal Science, 69, 3274–3283. Pringle, T. D., Williams, S. E., Lamb, B. S., Johnson, D. D., & West, R. L. (1997). Carcass characteristics, the calpain proteinase system, and aged tenderness of Angus and Brahman crossbred steers. Journal of Animal Science, 75, 2955–2961. Reuter, B. J., Wulf, D. M., & Maddock, R. J. (2002). Mapping intramuscular tenderness variation in four major muscles of the beef round. Journal of Animal Science, 80, 2594–2599. SAS (1985). SAS user’s guide: Basics, version 5. Cary, NC: SAS Institute, Inc. SAS/STAT. (1990). User’s guide, version 6 (vol. 2, 4th ed.). Cary, NC: SAS Institute Inc. Segars, R. A., Nordstrom, H. A., & Kapsalis, J. G. (1974). Textural characteristics of beef muscles. Journal of Texture Studies, 5, 283– 297. Shackelford, S. D., Wheeler, T. L., & Koohmaraie, M. (1997). Repeatability of tenderness measurements in beef round muscles. Journal of Animal Science, 75, 2411–2416. Smith, G. C., Carpenter, Z. L., & King, G. T. (1969). Considerations for beef tenderness evaluations. Journal of Food Science, 34, 612–618. Smulders, F. J. M., Marsh, B. B., Swartz, D. R., Russell, R. L., & Hoenecke, M. E. (1990). Beef tenderness and sarcomere length. Meat Science, 28, 349–363. Sørheim, O., Idland, J., Halvorsen, E. C., Frøystein, T., Lea, P., & Hildrum, K. I. (2001). InXuence of beef carcass stretching and chilling rate on tenderness of M. longissimus dorsi. Meat Science, 57, 79–85. Tarrant, P. V. (1998). Some recent advances and future priorities in research for the meat industry. In Proceedings 44th international congress of meat science and technology (pp. 2–13), 30 August–4 September 1998, Barcelona, Spain. Taylor, R. G. (2003). Meat tenderness: theory and practice. Brazilian Journal of Food Technology, 6, 56–66 (special issue 49th ICoMST). Torrescano, G., Sánchez-Escalante, A., Giménez, B., Roncalés, P., & Beltrán, J. A. (2003). Shear values of raw samples of 14 bovine muscles and their relation to muscle collagen characteristics. Meat Science, 64, 85–91. Tornberg, E., Walhgren, M., Brøndum, J., & Engelsen, S. B. (2000). Prerigor conditions in beef under varying temperature- and pH-falls studied with rigometer, NMR and NIR. Food Chemistry, 69, 407–418. Wheeler, T. L., Shackelford, S. D., & Koohmaraie, M. (1996). Sampling, cooking, and coring eVects on Warner–Bratzler shear force values in beef. Journal of Animal Science, 74, 1553–1562. Williams, J. C., Field, R. A., & Riley, M. L. (1983). InXuence of storage times after cooking on Warner–Bratzler shear values of beef roasts. Journal of Food Science, 48, 309–310 312. Zuckerman, H., Berry, B. W., Eastridge, J. S., & Solomon, M. B. (2002). Shear force mapping: A tool for tenderness measurement. Journal of Muscle Foods, 13, 1–12.