- Email: [email protected]
Water distribution in porcine M. longissimus dorsi in relation to sensory properties
Meat Science 17 (1986) 213-231
Water Distribution in Porcine M. longissimus dorsi in Relation to Sensory Properties Stina Fjelkner-Modig* & Eva Tornberg? Swedish Meat Research Institute, POB 504, S-244 00 K/ivlinge, Sweden (Received: 29 August, 1985)
S UMMA R Y The water distribution in M. longissimus dorsi (LD) from purebred Hampshire and Swedish Yorkshire pigs was recorded by proton-pulseNMR. Three domains of water were seen with this type of method, • designated as free, extracellular and intracellular water, respectively. The relative proton population for the free water increased from 1"8% in raw, to 3.3% in fried, samples from Hampshire and to 4.2% for the Yorkshire samples. The relaxation time of the extracellular water increased for Hampshire samples from lOOms, when raw, to 108 ms and l l 4 m s for samples fried to centre temperatures of 68"C and 80"C, respectively. For Yorkshire samples it decreased from 122 to 108 and 109 ms, respectively. The relative proton population of the extracellular water decreased after frying (raw: 24"5%, 68°C: 18"6%, 80°C: 13"9%) for Hampshire samples, whereas the corresponding populations of protons for those of Yorkshire were 16"5%, 19.2% and 16"6%. The intracellular water had relaxation times of about 40ms (raw), 30 ms (68"C) and 28ms (80°C) for both breeds. The relative proton populations were: for Hampshire, 74"1% (raw), 77.7% (68"C), 83"4% (80"C) and, for Yorkshire, 81.6% (raw), 75"5% (68°C) and 77.2%
(8o°c). On average, the samples of Hampshire were more juicy and tender than those of Yorkshire. The sensory properties were related to the water distribution, but obvious influences of breed and end-point temperature * Present address: Nordreco, POB 500, S-267 00 Bjuv, Sweden. t To whom requests for reprints should be addressed. 213
Meat Science 0309-1740/86/$03-50 © Elsevier Applied Science Publishers Ltd, England, 1986. Printed in Great Britain
214
Stina Fjelkner-Modig, Eva Tornberg at frying were noted for the relationships. In general, the juiciness and tenderness of fried LD samples could fairly well be predicted by the water distribution in raw meat. However, due to the influence of breed and temperature, different variables are best for the prediction. Also, when the sensory properties were related to the water distribution in fried samples, the influences of breed and end-point temperatures were noted.
INTRODUCTION In a series of investigations the sensory properties of porcine M. longissimus dorsi have been found to be related to breed, cooking temperature and, for samples of Swedish Yorkshire, to composition and distribution of the intramuscular lipids (Fjelkner-Modig, 1986; FjelknerModig & Tornberg, 1986). The present paper deals with intramuscular water and its influence on eating quality. The greater part--about 8 5 % - - o f the muscle tissue water is located intracellularly in the myofibrils and the remaining 15 % is located in the extracellular space (Hamm, 1972, 1975). The water is primarily held by the structure built up by the myofibrillar proteins. The main part of the water is, as supposed by Offer & Trinick (1983), held by capillarity. A relatively small amount of the tissue water (4-5% of the total water) is restricted in motion due to the proximity of the protein molecules (Wismer-Pedersen, 1971; Hamm, 1975). This water is often called hydration water. The capacity of muscle tissue to retain water changes during the conversion of muscle to meat. The pH drop and the structural alterations (e.g. shrinkage of the myofibrils and increase in both the extracellular and intermyofibrillar spaces) that occur during rigor decrease the waterretaining capacity by commonly 1-3% and there is a loss of liquid (Taylor & Dant, 1971; Hamm, 1972; Heffron & Hegarty, 1974; Penny 1977; Offer, 1984; Offer et al., 1984). The amount of this loss, mainly occurring as drip, is influenced by a number of factors. Pork has, for instance, a higher water-retaining capacity than beef and horse meat (Wismer-Pedersen, 1971), but the variation within species is larger for pork than for beef (Sch6n & Scheper, 1960): PSE meat--mainly a quality problem for pork--has a lower water-retaining capacity than normal meat (Bendall & WismerPedersen, 1962). The amount and area of cut surfaces-especially those
Water distribution in porcine M. longissimusdorsi
215
perpendicular to the myofibrils--is also important for drip formation as the drip occurs mainly from the cut ends (Hamm, 1972). Fresh pork is always cooked before consumption. The cooking induces structural changes, which decrease the water-retaining capacity of the meat. The amount of liquid released from the meat during cooking is affected by several factors, such as the size of the piece of m e a t - especially the length-volume ratio cooking method, cooking time, final cooking temperature, the amount of connective tissue and the suspension of the muscle tissue (Bouton et al., 1976a,b; Bendall & Restall, 1983; Bogn/tr & Piicher, 1983; Offer et al., 1984). The water-retaining capacity is considered important for the sensory properties (Hamm, 1972, Lawrie, 1979). According to Hamm (1972, 1975) and Bouton et al. (1975) a high water-holding capacity is related to a high juiciness and also to a high tenderness. Contrary to these studies, Ritchey & Hostetler (1964), Ritchey (1965) and Fulton & Davis (1975) found no relationship between water-holding capacity and palatability. One important factor that may explain the contradictory results with regard to the relationship between water-retaining capacity and sensory properties of meat might be the methodology. As pointed out by Hermansson & Lucisano ( 1 9 8 2 ) a n d Tornberg & Nerbrink (1984), common methods for measuring the water-holding capacity of meat require that an external force is applied on the meat sample. This external force might cause damage to the meat structure. Therefore the results, given in the literature, concerning water holding in meat are difficult to interpret and compare as they depend on the method of measurement. A method not exposed to this disadvantage is the pulseN M R method. This method could therefore be considered to have a more absolute character. The pulse-NMR method also gives information about the water distribution within a meat sample and about changes occurring in the water distribution, when the meat sample is processed (Tornberg & Nerbrink, 1984). Since the distribution of water within the meat most probably affects the sensory properties, we believe that pulseN M R measurements will contribute to the understanding of the r61e of water for the eating quality of meat. Hence, the aim of this study was to characterize the water distribution within M . longissimus dorsi (LD) from Hampshire and Swedish Yorkshire pigs when raw and fried to 68°C and 80°C and to relate the water distribution within the samples to the sensory properties of LD.
216
StinaFjetkner-Modig, Eva Tornberg
MATERIALS AND METHODS Materials M. longissimus dorsi (LD, 7th vertebra thoracica-5th vertebra lumbaris) was taken from the right side of six Hampshire a n d six Swedish Yorkshire pigs. The animals were purebred and raised under identical conditions at Hermanstorp progeny testing station. The pigs for each breed came from three litters, each litter supplying one castrate and one gilt. The pigs were slaughtered during the week they reached a liveweight of 100 kg. The normal electrical stunning (110 V) slaughter method was used. The carcasses were conditioned at + 2°C for 3 days before the samples of LD were removed. Each sample of LD was cut into eighteen slices, 1.5 cm thick, and into two pieces of about 300 g. The analyses carried out on the various slices are shown in Fig, 1.
Meat colour value Meat colour value of LD was recorded by an EEL reflectance photometer (Evans Electroselenium Limited, Halstead, Great Britain) by using the Y-filter at a maximum wavelength of 550 nm. The reflectance value was recorded at three places on the surface of the transverse cut of LD at the 13th vertebra thoracica. The average value was registered. Meat with an EEL-colour value of 27 or higher is graded as PSE meat in the pig progeny testing.
pH value
r
=
Chemical a n a l y s i s
pH =
0
=
Drip l o s s
S
=
Sensory
EEL =
Co~oue vatue
SL
=
Sarcamere
F
Frying loss
W
=
Water dlstributian
=
analysis {eng,~h
Fig. 1. The analysis scheme for M. longissimus dorsi from pork. The sample slices were analysed raw and after frying to final centre temperatures of 68°C and 80°C.
Water distribution in porcine M. lonNssimus dorsi
217
Ultimate pH The p H value o f L D was recorded 3 days post m o r t e n by a p H meter (Portamess 651) with a glass electrode (Ingold 404) at the 13th vertebra thoracica. The measurements were carried out in the centre o f the transverse surface at a depth o f about 2 cm.
Drip loss The a m o u n t o f drip was determined by the m e t h o d of L u n d s t r 6 m e t al. (1984). Meat pieces (Fig. 1) o f about 300g were kept in trays at +2°C for 24 h and the drip was determined as percentage weight loss.
Frying and frying loss The 1-5-cm-thick slices were fried at 180°C on a double-sided griddle immediately prior to sensory evaluation. The frying was discontinued at a centre temperature of 68°C or 80°C for the slices shown in Fig. 1. The end-point temperature was recorded by thermocouples. Frying loss was calculated as the percentage loss in weight after frying. The frying loss was determined for two slices (Fig. 1) for each of the two end-point temperatures of 68°C and 80°C.
Sensory properties A trained expert panel consisting of ten men and women assessed four samples at each session. Sample pieces (4 x 3cm) were served one at a time immediately after frying. Between samples the assessors were asked to rinse their mouths with distilled water at r o o m temperature and eat an unsweetened biscuit. The sensory properties of fried LD were evaluated by a profile including the attributes: visible juiciness (1 = none, 9 = very large), initial juiciness (1 = none, 9 = very large), dryness in m o u t h (1 = none, 9 = very large), elasticity (1 = not elastic, 9 = very elastic), hardness (1 = v e r y soft, 9 = very hard), stringiness (1 = none, 9 = very large), chewing time (1 = v e r y short, 9 = v e r y long), chewing residual (1 = none, 9 = v e r y large) and total flavour intensity (1 = very weak, 9 = very strong) in accordance with the description given by Fjelkner-Modig (1986). The three attributes, visible and initial juiciness and dryness in m o u t h , were
218
Stina Fjetkner-Modig, Eva Tornberg
highly related, as were the attributes hardness, stringiness, chewing time and chewing residual (Fjelkner-Modig, 1986). Hence the sum of the attribute scores for visible and initial juiciness and dryness in m o u t h was considered representative of the total impression of juiciness and that of the attribute scores for hardness, stringiness, chewing time and chewing residual the total impression of toughness. These values were used for the statistical evaluation of the relationships between sensory properties and the water distribution in the meat.
Chemical analysis The contents of water (Nilsson, 1969), intramuscular lipids (SBRmethod, N M K L , No. 88, 1974), crude protein (Kjetdaht as modified by Nilsson, 1968) and hydroxyproline (Stegemann, 1958 as modified by Weber, 1973) were analysed for both raw and fried samples (Fig. I). Prior to the chemical analyses the epimysium sheath on :the slices was removed. The connective tissue content was determined by the amount of hydroxyproline in accordance with the method of Wyler (1972).
Sarcomere length A piece 10 x 5 m m was taken from the centre of those slices where the analysis of water distribution had been carried out. The pieces were cut along the fibre bundles into smaller pieces and fixed according to the procedure used by Cross et al. (1980-81). The length of the sarcomere diffraction bands was recorded by a helium-neon laser (Voyle, 1971). The sarcomere length was calculated in accordance with the formula used by Cross et al. (1980-81).
Water distribution The water distribution was recorded by a proton-pulse-NMR instrument (Bruker, Minispec, PC/20), mainly following the procedure of Tornberg & Nerbrink (1984). The measurements were carried out for raw and fried samples of LD, as shown in Fig. 1. Perpendicular to the slice surface a square-shaped (5.3 x 5-3 mm) rod was cut from the centre of the slice. The rod was cut carefully to avoid drip formation. The ends of the rod were cut off. The rod (about 8 m m long and with a weight of about 0.3 g) was put into a N M R tube (q57.5 mm) by means of a
Water distribution in porcine M. |on~ssimus dorsi
219
spatula and a glass rod. Muscle tissue and liquid, which had adhered to the tube wall, were removed with a paper tissue. Prior to the measurement the N M R tube with the sample rod was thermostatically held at 25°C for 30rain. The transverse relaxation time (T:) of the water protons within the meat rod was recorded at a frequency of 20 MHz. The T: measurements were made at 25°C by using the CarrPurcell-Meibom-Gill method (Meibom & Gill, 1958). The T: recordings were made for two different z spacings--500 and 4000 g s - - a n d at each measurement 49 scans were accumulated. The magnetisation value, M0 (corresponding to 100% water protons), was mostly determined at the "c spacing 4000 ~s. In two cases, however, it was easier to determine M0 for the measurements at z spacing 500 #s. M o was determined from the semilog plot of log M versus time and derived from the intercept of the straight line obtained from the first data arising from water proton relaxation. The relaxation data were analysed in multiexponential decay by curve decomposition using a microcomputer (Luxor ABC 80) as described b y Tornberg & Nerbrink (1984). The errors introduced by this calculation were for every sample estimated as being _< + 2 ms in T, for the fastest relaxing process and _< + 5 ms for the next process. The accuracy of the relative population of protons was estimated as being _< -!-_5% for the ' fastest process, _ +_6% for the next process and _< + 1% for the slowest process.
Statistical analyses Data were statistically treated and programmes of ANOVA one-way classification, Student's t test and linear regression were used.
RESULTS A N D DISCUSSION
Meat quality traits and composition All the samples of LD were taken from normal, graded pig carcasses. The Hampshire samples and the Swedish Yorkshire samples both had, on average, about the same EEL-colour value and ultimatepH (Table 1). This agrees well with an earlier investigation (Fjelkner-Modig & Persson,
23.2 5-4, 3'2 a 76'0;' 1.4 20.8" 3.2 1.84
x
Raw
(0'8) (o.i) (0'4) (0'4) (0-4) (0-6) (0.4) (0-05)
(SD)
(I "5) (1-2) (1. I) (0-4) (0.07) (2.6) (0-7) (0-7) (0.6) (0-8) (0.9) (0-8) (0.8) (0.6) (0.3)
7'0 7-I 1-8 6.4 3-6 3-0 3.8 2.6 3.8
(SD)
67"8" 2.5 28.0 2-9 1.74 21-2
x
68°C
lhtmpshire
5'9 5'6 2.7 5"7 4-2 3.4 4.4 3.2 4.0
66"3 2-5 29.4 2.9 1.60 28-2
x
(0.5) (0.9) (0.5) (0.5) (0.7) (0.6) (0.6) (0.5) (0.4)
(1 "0) (0.8) (0.5) (0.3) (0.06) (2-9)
(St))
80°C
24'0 5.5' 2"5 a 75" 1~' 1-3 22.9" 2.9 1.78
x
Raw
(4'0) (0.2) (0'9) (0"5) (0-3) (0.7) (0.6) (0.05)
(SD)
5.9 6'7 2.4 6-0 4.4 3.7 4.4 2.9 3.7
69'4 ¢ 2-3 27.9 2.7 1.68 21-0
x
(1'2) (0-9) (0.7) (0.9) (1"4) (0-9) (0.8) (0.8) (0.3)
(0"9) (0.5) (I .5) (0-3) (0.05) (2.1)
(St))
68°C
Swedish Yorkshire
5-2 5.1 3-4 5.8 4.7 3.8 4.6 3.4 3.7
67"6 2.7 29.2 2.6 1.55 27.2
x
(1'6) (1.6) (0.9) (0.8) (1"4) (0-9) (1.3) (1-0) (0-5)
(1.9) (0-9) (I .4) (0.4) (0.05) (2.5)
(SD)
80°C
* Valne in an intensity score with 9 units. Values within a row, descending from the same preparation and having the same letter, are significantly different. a, p <0.001; b, p < 0.01; c, d, e, p _<0.05.
EEL-colour wdue pit Drip (%) Water (%) lntra,nuscular lipids (%) Protein (%) Connective tissue (%) Sarcomere length (llm) Frying loss (%) Sensory analysis Visible juiciness* Initial juiciness* Dryness in mouth* Elasticity* Hardness* Stringiness* Chewing time* Chewing residual* Total Ilavour*
Trait
TABLE 1 Average Value (x) and Standard Deviation (SD) of the Different Traits of M. hmgissimus dorsi from Hampshire (n = 6) and Swedish Yorkshire (n = 6), when Analysed Raw and After Frying to End-point Temperatures of 68°C and 80°C
Water distribution in porcine M. lon~ssimus dorsi
221
1986). However, the average EEL-colour value of both the Hampshire and Yorkshire samples is about one unit higher than that found by Lundstr6m et at. (1984) for Yorkshire samples and those given by Fjelkner-Modig & Persson (1986) for samples of both the Hampshire and the Yorkshire breeds. The large standard deviation for the samples of Yorkshire (Table 1) is due to one sample having a high EEL-colour value, thus indicating PSE meat. The paleness and wetness appeared, however, only locally and a normal drip loss and ultimate pH was noted for the sample. A drip loss of 2-5% was recorded for the Yorkshire samples and this agrees well with information given by Lundstr6m et al. (1984). The Hampshire samples had a drip loss of 3-2%, which was significantly higher than those of the Yorkshire (Table 1). The results of the sarcomere length determination are given in Table 1. The Hampshire samples had throughout somewhat longer sarcomeres than the Yorkshire samples. This was also found by Fjelkner-Modig & Persson (1986). The shrinkage of sarcomere in length due to frying was the same for both breeds. Frying to 68°C resulted in a 5% reduction of sarcomere length and to 80°C in a reduction of about 13%. The contents of water, intramuscular lipids, protein and connective tissue were analysed for raw samples and also for samples fried to an internal temperature of 68°C and 80°C. The results of these analyses are given in Table 1. A breed difference was noted in the amount of water between the raw samples and those fried to 68°C. For raw samples the highest water content was noted for the Hampshire breed but, for the fried samples-both 68°C and 80°C--the highest content was noted for the Yorkshire breed. The Hampshire samples had a significantly lower amount of protein than those of Yorkshire, when raw, but for the fried samples the protein content was almost the same for the samples from both breeds, probably due to the different frying losses. The higher water content and the lower protein content of raw meat of Hampshire compared with that of Yorkshire has also been found by P. Barton-Gade, (pers. comm.). A rather low intramuscular lipid content was noted for the samples of both Hampshire and Yorkshire. Lundstr6m et al. (1984) and Fjelkner-Modig & Persson (1986) have reported an intramuscular lipid content of 1.7% for Yorkshire samples. For Hampshire samples the corresponding value was 2-0% (Fjelkner-Modig & Persson, 1986).
222
Stina Fjelkner-Modig, Eva Tornberg
Sensory analysis The samples of Hampshire were more juicy--both visibly and initially-more elastic, less dry in the mouth and tess hard than the Yorkshire samples. Furthermore, less stringiness, shorter chewing time and smaller chewing residual were noted for Hampshire than for Yorkshire. The total flavour intensity was about the same for both breeds (Table I). The breed differences were more pronounced at the final frying temperature of 68°C than at 80°C. The same breed differences have been found by Fjelkner-Modig (1986) but, in comparison, a somewhat higher juiciness and also a higher tenderness (i.e. shorter chewing time, smaller chewing residual and less stringiness and hardness) were noted in the present study for both the Hampshire and the Yorkshire breeds. This could be due to the treatment of the samples prior to sensory evaluation. Fjelkner-Modig (1986) used samples which had been frozen, stored at - 2 0 ° C for some time and thawed prior to frying. In the present investigation the sampIes were prepared fresh and in our own experience (S. Fjelkner-Modig, unpublished results) samples prepared flesh are assessed as being more juicy and tender than frozen samples. The sensory differences between the breeds were not statistically significant. However, the sensory differences between the breeds and also those between the two frying temperatures agreed well with those found by Fjelkner-Modig (1986), evaluating a much larger number of samples than in this study. Therefore, the breed differences noted in the present study are considered to be representative.
Water distribution Three different relaxation times were, in general, found for both the raw and the fried samples. The water with the shortest relaxation time has been designated as intracellular water and that with medium relaxation time as extracellular water (Hazlewood et al., 1974, Foster et al., 1976). These designations might be questionable. However, the aim of this investigation was primarily to study the effect of cooking on the water distribution within meat and the relationships between the water distribution within meat and the sensory properties and therefore the actual meanings of the designations are not considered here. The water with the longest relaxation time, i.e. 1 s or more, is 'free water', as this
Water distribution in porcine M. longissimus dorsi
223
TABLE 2 Average Value (x) a n d Standard Deviation (SD) of the Relaxation Times and the Relative Population of Protons Recorded by Proton-Pulse-NMR for Samples of M. longissimus dorsi from Hampshire (n = 6) and Yorkshire (n = 6) Pigs. The Recordings were Carried Out for Raw Samples and for Samples Fried to 680C and 80"C
Sample
Raw
68*C
80"C
Hampshire
Yorkshire
Relaxation time
Relative proton population
Relaxation time
Relative proton population
(ms)
(%)
(ms)
(%)
x
(SD)
x
(SD)
x
(SD)
x
(SD)
> 1 500 I00 a 39.7
(8) (1.0)
1-8 24.5 b 74.1 c
(0. I) (3.2) (2-5)
> 1 500 122a 40-1
(13) (3.5)
1.8 16.5 b 81-6c
(0.2) (4.4) (3.6)
> 1 500 108 30.6
(21) (3.3)
3.3d 18.6 77-7
(0-4) (5-0) (6.9)
> 1 500 108 29-8
(10) (1-2)
4.2 d 19.2 75-5
(0-7) (4.2) (4.0)
(0.4)
> 1 500
(26)
13.9
(5-2)
109
(19)
16.6
(2-4)
(3,8)
83-4
(8.0)
(3-0)
77.2
(2-3)
> 1 500
114 28.6
3.2e
27-3
4-2 ~
(0.8)
Values within a row and with the same letter are significantly different. p _< 0-01: a, b, e and d. p <_ 0"05: e.
water corresponds to protons that do not exchange with the small amount of water, which is restricted in motion (i.e. the hydration water). The mean relaxation times and the mean relative populations of protons are given per breed in Table 2. The intracellular water has the shortest relaxation time (T3---about 40 ms for raw samples and about 30 and 23ms for the samples fried to 68°C and 80°C, respectively. Renou et al. (1984) have reported two relaxation times, where the fastest were 41 and 43ms for raw LD from Pietrain and Large White pigs, respectively. Our results for 7",.agree fairly well with their observations. A somewhat longer T,- value is given by Pearson et al. (1974). They investigated porcine muscles--mainly L D - - a n d found a 7",-value of 53 ms, when the samples were analysed 24 h after death. Moreover, for muscle tissue of other species, relaxation times of intracellular water varying from 35 to 45 ms have been reported. Hazlewood et al. (1974) noted, for instance, a Tt value of 44ms for rat, Foster et al. (1976) a Ti
224
Stina Fjelkner-Modig, Eva Tornberg
value of 45ms for rat and 35ms for barnacle, Lillford et al. (1980) a Ti value of 47ms for beef and, finally, Tornberg & Nerbrink (1984) have reported a Ti value of about 36ms for beef. The relaxation time of the extracellular water in raw samples was significantly longer for Yorkshire (122 ms) than for Hampshire (100 ms) samples. Pearson et al. (1974) have reported a relaxation time for extracellular water (Te) of 140ms. For other species Te values of about 120 ms (beef, Tornberg & Nerbrink, 1984), about 150 ms (beef, Lillford et al., 1980), 200ms (rat, Hazlewood et al., 1974 and Foster et al., 1976) and 400ms (barnacle, Foster et al., 1976) have been reported. The raw samples of Hampshire had a lower relative population of protons with the shortest relaxation time (Pi) than those of Yorkshire (Table2). Pearson et al. (1974) reported an intracellular water content of 86% for pork. Both Hazlewood et al. (1974) and Foster et al. (1976) found Pi values of 80--82% for rat muscle tissue whereas Lillford et al. (1980) reported Pi values of 94% for beef. The relative population of protons at the extracellular space (Pe) was 16.5% for the Yorkshire samples (Table 2). Pearson et al. (1974) noted approximately the same amount of water (14%). For muscle tissue of other species mo_stly P~ values of 10-14% have been reported (Hazlewood et al.. 1974; Foster et al.. 1976; Tornberg & Nerbrink, 1984): Compared with Yorkshire the Hampshire samples had a high content of extracellular water. All the Hampshire samples recorded had a rather high Pe value and the variation within breed was smaller for Hampshire than for Yorkshire. About 1-8% of the total water in raw muscle tissue was free for both breeds. 7',- decreased after heating for both breeds. This suggests an increase in protein concentration within the fibres, which could originate from a shrinkage of the myofibrils. The decrease in T,. from 68°C to 80°C was much smaller than that from raw to 68°C. Furthermore, for the extracellular water the changes in relaxation time induced by frying differed between the two breeds. T~ increased on frying for the samples of Hampshire. For the samples of Yorkshire longer T e values were noted in the raw meat than for those of Hampshire, but on frying the T~ decreased and was almost the same for the samples fried to 68°C as for those fried to 80°C (Table2). The relative proportion of intracetlutar water (Pc) increased by frying for the samples of Hampshire (3.6% from raw to 68°C and 5-7% from
Water distribution in porcine M. lon~ssimus dorsi
225
68°C to 80°C) while, for those of Yorkshire, a decrease of 6-1% in the relative amount of intracellular water was noted from raw to 68°C followed by an increase of 1.7% from 68°C to 80°C. This implies that there is an effect of breed on the changes in water distribution, which occur during frying. The change, due to frying, in the relative proportion of extracellular water also differed from breed to breed. For samples fried to 68°C almost the same Pe values (about 19%) were noted for the two breeds. But compared with the raw sample the Hampshire samples fried to 68°C had a lower proportion of extracellular water whereas, for the Yorkshire samples, a tendency to the opposite ratio was noted. The Pe value decreased to some extent from 680C to 80°C for both the breeds. The relative amount of free water was higher in the fried samples than in the raw ones.
The relationships between meat quality traits, frying loss and water distribution The results of the statistical analysis of the relationships between drip loss, EEL-colour value, final pH and frying loss, on the one hand, and water distribution for raw meat, on the other, are given in Table 3. An analysis of the relationships between frying loss for the samples fried to 68°C and 80°C and the corresponding water distribution of these fried samples was also performed and the results are shown in Table 4. As is evident in Table 3, an influence of breed was noted for the relationships between the meat quality traits and the water distribution. In most cases lower correlation coefficients were obtained for the Hampshire than for the Yorkshire samples. This might partly be a result of the smaller variations within breed in pH, EEL-colour value and drip loss that were recorded for the Hampshire samples (Table 1). For the Hampshire samples a tendency was noted that an increase in the amount of extracellular water was related to an increase in drip loss and decreases in both pH and EEL-colour values. These relationships could be expected as meat with a high EEL-colour value, i.e. PSE meat," mostly has a low pH value and a high drip loss and the extracellular water is easier released than the intracellular water. For the Yorkshire samples the meat quality traits were most related to the relaxation time of the intracellular water. A short relaxation time of the intracellular
Stina Fjelkner-Modig, Eva Tornberg
226
TABLE 3 The Results--Given as Correlation Coefficients---of Linear Regression Analysis of Meat Quality Traits, Frying Loss, Sensory Properties and Water Distribution for Raw Samples of M. longissimus dorsi from Hampshire (H) and Swedish Yorkshire (Y) Pigs
Trait
Water
Breed
T~
distribution in
Ti
Pf
raw meat
Pc
Pi
Sensory properties Juiciness
(68"C)
Juiciness
(80"C)
Toughness
(68"C)
Toughness
(80"C)
H Y H Y H Y H Y
--0.28 -0-09 0.57 -0-43 0-02 0"54 -0-23 0.41
0-68 0.75 0-19 0-45 - 0-97"* -0.04 -0-72 -0.33
0.47 -0.31 -0-73 -0-41 - 0-17 0-04 0-15 0.16
0-24 0.06 -0-80(*) 0.51 - 0-13 -0'66 0-10 -0"55
-0-35 -0-05 0"81" -0-47 0-07 0-71 -0-22 0.55
H Y H Y
-0-44 0-31 0-31 0-16
0.10 0"14 0.51 0.17
0-63 -0"35 0-02 -0-52
0-85* --0"51 0.29 -0-41
-0.70 0"60 -0-39 0-52
H Y H Y H Y
0-00 0"ll 0" 11 0-25 0.07 0-05
0'05 0-06 -0"40 0-43 0.03 0.12
0"37 -0-19 -0"52 -0-05 - 0-22 -0.24
-0-19 0"22 0"49 -0-03 - 0.09 0-24
Frying loss Frying loss (68*C) Frying toss (80"C) Meat
quality traits
Drip loss pH EEL-colour value
-0"26 -0"60 -0-69 0-59 0-19 -0-63
T = Relaxation time. P = Relative population of protons. Indexf = free water, e = Extracellular water, i = Intracellular water. Significance level: p _<0-10: (*), p <_0-05: * and p < 0-01: **
water, indicating a high protein concentration and, most probably, more denatured proteins, was associated with a high drip loss, a low pH and a high EEL-colour value. These relationships would also be expected as meat with a high EEL-colour value represents PSE meat and this meat has more denatured proteins than that o f normal meat quatity. The frying loss for samples fried to an end-point temperature o f 68°C could fairly well be predicted by measuring the water distribution in the raw meat for both breeds (Table 3). A high amount of extracellular water was associated with a high frying loss (r = 0"85, p <_ 0"05) for
H Y H Y
H Y H Y H Y H Y
Breed
-0.26 0.83* 0.23 0.11
0.30 0.15 0.57 0.19 - 0.40 0.70 - 0-34 -0.10
T,
-0.41 0.27 - 0-31 -0.11
0.26 -0.96** 0.78(*) 0.40 -0.45 0.45 - 0.07 -0.36
T,
-0.18 0.60 0.64 0.44
0.76(*) 0.24 0.30 0.50 - 0.67 0.48 - 0.62 -0.31
P~
-0.09 -0.55 - 0'21 0.31
-0.43 -0.19 -0.78(*) - 0.42 0.59 -0-22 0.47 0.38
P,
Water distribution in fried meat
0.23 0.55 0.13 -0.12
0-17 0.26 0.81 * - 0.11 - 0.33 0.22 - 0.42 0-15
P,
68 68 80 80
68 68 80 80 68 68 80 80
Frying temperature ('C)
T = Relaxation time. P = Relative population of protons. Index f = Free water, e = Extracellular water, i = Intercellular water. Signilicance level: p < 0.10, (*); p < 0-05, * and p < 0.01, **
Frying loss
Frying loss
Frying loss
Toughness
Toughness
Juiciness
Juiciness
Sensory properties
Trait
TABLE 4 The Results--Given as Correlation Coefficients--of Linear Regression Analysis of Frying Loss, Sensory Properties and Water Distribution for Fried Samples of M. longissimus dorsi from Hampshire (H) and Swedish Yorkshire (Y) Pigs
to to "-4
,o.
~S' E 5"
5' ~. '~
228
Stina Fjelkner-Modig, Eva Tornberg
Hampshire samples, whereas, for the Yorkshire samples, there was a tendency for high flying loss to be related to a high amount of intracellular water (r = 0-60, not significant). This implies that quite different relationships were obtained for the two breeds. A breed influence was also noted for the samples fried to 80°C. For the Yorkshire samples similar relationships between water distribution and frying loss were found for the samples fried to 80°C as for those flied to 68°C, whereas for the Hampshire samples, fried to 80°C, the relaxation time of the intracellular water was a more prominent factor than the extracellular water content. Frying loss, when related to the fried product, was, in the case of Yorkshire, to some extent related to the amount of free water and for the samples fried to 68°C also to the relaxation time of the extraceUular water (T~). Also, for the Hampshire samples, when fried to 80°C, the highest correlation coefficient was noted between frying loss and the amount of free water. For the Hampshire samples fried to 68°C the shrinkage of myofibrils (i.e. a decrease in T,-) was to some extent (r = -0-41) related to the frying toss.
The relationships between sensory properties and water distribution The relationships between the sensory properties, juiciness and toughness and the water distribution in raw meat are, per breed, shown in Table 3. Moreover, the sensory properties of the samples fried to the temperatures 68°C and 80°C, as related to the water distribution of the corresponding samples, are shown in Table 4. In general; it may be noted that about the same correlation coefficient, but with opposite sign, is found between the amounts of intra- and extracellular water and the sensory properties, the frying loss and the meat quality traits. This is because the amounts ofintra- and extracellular water are highly related, i.e. a decrease in P,. is followed by an increase in Pe, and vice versa. As is evident from Table 3, a breed difference with regard to the relationships between sensory properties and the water distribution in raw meat was noted. For both the Hampshire and the Yorkshire samples, juiciness assessed at 68°C was mostly related to the relaxation time of the intracellular water, whereas the amounts of intracellular and extracellular water were more determining for the juiciness achieved at 80°C. However, the juiciness of Hampshire samples, when fried to
Water distribution in porcine M. longissimus dorsi
229
80°C, was positively related to the amount of intraceUular water (r = 0"81, p _< 0-05) but for the Yorkshire samples a lower and negative correlation coefficient (r = -0-47, not significant) was obtained. As is evident from Table 3, the toughness of Hampshire samples was related to the relaxation time of the intracellular water for both temperatures. For meat from this breed it seems favourable, with regard to tenderness, to have a long Ti value, implying that less shrinked fibres gives a more tender meat. For the Yorkshire samples rather high correlation coefficients were found between toughness and the amounts of extracellular and intracellular water. This implies that a low amount of intracellular water, i.e. shrinked myofibrils, and a high amount of extracellular water is favourable with regard to tenderness. The results shown in Table 3 also imply that the juiciness and tenderness of fried meat could fairly well be predicted by the water distribution in raw meat. However, due to the obvious breed influence on the noted relationships, different parameters have to be used for each breed. An influence of breed was also noted for the relationships between sensory properties and the water distribution within fried meat. For the Hampshire samples we noted a tendency for more free water to be associated with more tender and juicy meat (samples fried to 68°C). The juiciness of samples fried to 80°C was, however, related to the amounts of intra- and extracellular water. For the Yorkshire samples a significant relationship was noted between juiciness and the relaxation time of intracellular water. This implies that a shrinkage of the myofibrils is related to an increase in juiciness for the samples fried to 68°C ( r = 0-98, p_< 0.01). For the samples fried to 80°C the juiciness showed a tendency to be related to the amount of free water. The toughness of the Yorkshire samples was associated with the relaxation time of extracellular water. For samples fried to 80°C, in general low correlation coefficients were noted. Overall, more significant relationships were noted between the sensory properties and the water distribution, recorded in raw and fried samples, for the Hampshire than for the Yorkshire samples. This could be a result of the influence of the amount and composition of intramuscular lipids on the sensory properties reported for Yorkshire samples but not for Hampshire samples (Fjelkner-Modig & Tornberg, 1986). With regard to flavour and how it is related to the water distribution, no clear cut pattern can be deduced from the statistical analysis. This
230
Stina Fjelkner-Modig, Eva Tornberg
could be due to the small variation in flavour noted from sample to sample. It should be observed from Tables 3 and 4 that a fairly high percentage of the variation in sensory properties was explained by water distribution. Hence, these results suggest that changes in water holding of meat is an important factor for determining the sensory quality of pork. ACKNOWLEDGEMENTS We wish to thank Dr H~kan Rud~rus for valuable discussions and criticism during the preparation of the manuscript and Gertrud Larsson (MSc) for her excellent technical assistance with the proton-pulse-NMR measurements. REFERENCES Bendall, J. R. & Restall, D. J. (1983). Meat Sci., 8, 93. Bendall, J. R. & Wismer-Pedersen, J. (1962). J. Food Sci., 27, 144. Bognfir, A. & Pfichner, H.-J. (1983). Fleischw&ts., 63, 943. Bouton, P. E., Ford, A. L., Harris, O. V. & Ratcliff, D. (1975). 3". Food Sci., 40, 884. Bouton, P. E., Harris, P. V. & Shorthose, W. R. (1976a). J. Text. Stud., 7, 179. Bouton, P. E., Harris, P. V. & Shorthose, W. R, (1976b). J. Food Sci., 41, 1092. Cross, H. R., West, R. L. & Dutson, T. R. (1980-81). Meat Sci., 5, 261. Fjelkner-Modig, S. (1986). J. Food Quality. (In press.) Fjelkner-Modig, S. & Persson, J. (1986). J. Anita. Sci. (In press.) Fjelkner-Modig, S. & Tornberg, E. (1986). J. Food Quality. (In press.) Foster, K. R., Resing, H. A. & Garroway, A. N. (1976). Science, 194, 324. Fulton, L. & Davis, C. (1975). J. Amer. Diet. Ass., 67, 227. Harem, R. (!972). Kolloidchemie des Fleisches, Parey, Hamburg, Berlin. I-Iamm, R. (1975). In: Meat (Cole, D. J. A. & Lawrie, R. A. (Eds)), Butterworths, London, 321. Hazlewood, C. F., Chang, D. C., Nichols, B. L. & Woessner, D. E. (1974). J. Biophysical, 14, 583. Heffron, J. J. A. & Hegarty, P. V. J. (1974). Comp. Biochem. Physiol., 49A, 43. Hermansson, A.-M. & Lucisano, M. (1982). J. Food Sci., 47, 1955. Lawrie, R. A. (1979). Meat science (3rd edn), Pergamon Press, Oxford. Lillford, P. J., Clark, A. H. & Jones, D. V. (1980). In: Water in polymers (Rowland, S. P. (Ed.)), ACS Symposium Series No. 27, 177. Lundstr6m, K., Bj~irstorp, G. & Malmfors, G. (1984). Proc. Sci. Meet. Biophysical PSE-Muscle Analysis, Vienna, 311.
Water distribution in porcine M. longissimus dorsi
23t
Meibom, S. & Gill, D. (1958). Rev. Sci. Instrum., 29, 688. Nilsson, R. (1968). Swedish Meat Research Institute, Anal. Instr., No. 10. Nilsson, R. (1969). Swedish Meat Research Institute, Anal. Instr., No. 6. NMKL, No. 88 (1974). Fedt. Bestemmelse i kod og kodvarer efter SBR. Offer, G. (1984). 30th Fur. Meet. Meat Res. Workers, Bristol, 87. Offer, G. & Trinick, J. (1983). Meat Sci., 8, 245. Offer, G., Restall, D. & Trinick, J. (1984). In: Recent advances in the chemistry of meat (Bailey, A. K. (Ed.)), The Royal Soc. Chem., London. Spec. Publ. No. 47, 71. Pearson, R. T., Duff, I. D., Derbyshire, W. & Blanshard, J. M. V. (1974). Biochim. Biophys. Acta, 362, 188. Penny, I. F. (1977). J. Sci. Fd. Agric., 28, 329. Renou, J. P., Monin, G. & Sellier, P. (1984). Proc. Sci. Meet. Biophysical PSE-Muscle Analysis, Vienna, 65. Ritchey, S. J. (1965). J. Food Sci., 30, 375. Ritchey, S. J. & Hostetler, R. L. (1964). J. Food Sci., 29, 413. Sch6n, L. & Scheper, J. (1960). Ziicktungskunde, 32, 488. Stegemann, H. (1958). Hoppe-Seyler's Zeitschrift fiir Physiologische Chemie, 311, 41. Taylor, A. A. & Dant, S. J. (1971). J. Fd. Techn., 6, 131. Tornberg, E. & Nerbrink, O. (1984). 30th Fur. Meet. Meat Res. Workers, Bristol, 112. Voyle, C. A. (1971). 17th Fur. Meet. Meat Res. Workers, Bristol, 95. Weber, R. (1973). Technical Report No. 7, Genve, Technicon International Division SA. Wismer-Pedersen, J. (1971). In: The science of meat and meat products (Price, J. F. & Schweigert, B. S. (Eds)) (2nd edn), W. H. Freeman & Co., San Francisco, 177. Wyler, O. D. (1972). Fleischwirts., 52, 42.
Water Distribution in Porcine M. longissimus dorsi in Relation to Sensory Properties Stina Fjelkner-Modig* & Eva Tornberg? Swedish Meat Research Institute, POB 504, S-244 00 K/ivlinge, Sweden (Received: 29 August, 1985)
S UMMA R Y The water distribution in M. longissimus dorsi (LD) from purebred Hampshire and Swedish Yorkshire pigs was recorded by proton-pulseNMR. Three domains of water were seen with this type of method, • designated as free, extracellular and intracellular water, respectively. The relative proton population for the free water increased from 1"8% in raw, to 3.3% in fried, samples from Hampshire and to 4.2% for the Yorkshire samples. The relaxation time of the extracellular water increased for Hampshire samples from lOOms, when raw, to 108 ms and l l 4 m s for samples fried to centre temperatures of 68"C and 80"C, respectively. For Yorkshire samples it decreased from 122 to 108 and 109 ms, respectively. The relative proton population of the extracellular water decreased after frying (raw: 24"5%, 68°C: 18"6%, 80°C: 13"9%) for Hampshire samples, whereas the corresponding populations of protons for those of Yorkshire were 16"5%, 19.2% and 16"6%. The intracellular water had relaxation times of about 40ms (raw), 30 ms (68"C) and 28ms (80°C) for both breeds. The relative proton populations were: for Hampshire, 74"1% (raw), 77.7% (68"C), 83"4% (80"C) and, for Yorkshire, 81.6% (raw), 75"5% (68°C) and 77.2%
(8o°c). On average, the samples of Hampshire were more juicy and tender than those of Yorkshire. The sensory properties were related to the water distribution, but obvious influences of breed and end-point temperature * Present address: Nordreco, POB 500, S-267 00 Bjuv, Sweden. t To whom requests for reprints should be addressed. 213
Meat Science 0309-1740/86/$03-50 © Elsevier Applied Science Publishers Ltd, England, 1986. Printed in Great Britain
214
Stina Fjelkner-Modig, Eva Tornberg at frying were noted for the relationships. In general, the juiciness and tenderness of fried LD samples could fairly well be predicted by the water distribution in raw meat. However, due to the influence of breed and temperature, different variables are best for the prediction. Also, when the sensory properties were related to the water distribution in fried samples, the influences of breed and end-point temperatures were noted.
INTRODUCTION In a series of investigations the sensory properties of porcine M. longissimus dorsi have been found to be related to breed, cooking temperature and, for samples of Swedish Yorkshire, to composition and distribution of the intramuscular lipids (Fjelkner-Modig, 1986; FjelknerModig & Tornberg, 1986). The present paper deals with intramuscular water and its influence on eating quality. The greater part--about 8 5 % - - o f the muscle tissue water is located intracellularly in the myofibrils and the remaining 15 % is located in the extracellular space (Hamm, 1972, 1975). The water is primarily held by the structure built up by the myofibrillar proteins. The main part of the water is, as supposed by Offer & Trinick (1983), held by capillarity. A relatively small amount of the tissue water (4-5% of the total water) is restricted in motion due to the proximity of the protein molecules (Wismer-Pedersen, 1971; Hamm, 1975). This water is often called hydration water. The capacity of muscle tissue to retain water changes during the conversion of muscle to meat. The pH drop and the structural alterations (e.g. shrinkage of the myofibrils and increase in both the extracellular and intermyofibrillar spaces) that occur during rigor decrease the waterretaining capacity by commonly 1-3% and there is a loss of liquid (Taylor & Dant, 1971; Hamm, 1972; Heffron & Hegarty, 1974; Penny 1977; Offer, 1984; Offer et al., 1984). The amount of this loss, mainly occurring as drip, is influenced by a number of factors. Pork has, for instance, a higher water-retaining capacity than beef and horse meat (Wismer-Pedersen, 1971), but the variation within species is larger for pork than for beef (Sch6n & Scheper, 1960): PSE meat--mainly a quality problem for pork--has a lower water-retaining capacity than normal meat (Bendall & WismerPedersen, 1962). The amount and area of cut surfaces-especially those
Water distribution in porcine M. longissimusdorsi
215
perpendicular to the myofibrils--is also important for drip formation as the drip occurs mainly from the cut ends (Hamm, 1972). Fresh pork is always cooked before consumption. The cooking induces structural changes, which decrease the water-retaining capacity of the meat. The amount of liquid released from the meat during cooking is affected by several factors, such as the size of the piece of m e a t - especially the length-volume ratio cooking method, cooking time, final cooking temperature, the amount of connective tissue and the suspension of the muscle tissue (Bouton et al., 1976a,b; Bendall & Restall, 1983; Bogn/tr & Piicher, 1983; Offer et al., 1984). The water-retaining capacity is considered important for the sensory properties (Hamm, 1972, Lawrie, 1979). According to Hamm (1972, 1975) and Bouton et al. (1975) a high water-holding capacity is related to a high juiciness and also to a high tenderness. Contrary to these studies, Ritchey & Hostetler (1964), Ritchey (1965) and Fulton & Davis (1975) found no relationship between water-holding capacity and palatability. One important factor that may explain the contradictory results with regard to the relationship between water-retaining capacity and sensory properties of meat might be the methodology. As pointed out by Hermansson & Lucisano ( 1 9 8 2 ) a n d Tornberg & Nerbrink (1984), common methods for measuring the water-holding capacity of meat require that an external force is applied on the meat sample. This external force might cause damage to the meat structure. Therefore the results, given in the literature, concerning water holding in meat are difficult to interpret and compare as they depend on the method of measurement. A method not exposed to this disadvantage is the pulseN M R method. This method could therefore be considered to have a more absolute character. The pulse-NMR method also gives information about the water distribution within a meat sample and about changes occurring in the water distribution, when the meat sample is processed (Tornberg & Nerbrink, 1984). Since the distribution of water within the meat most probably affects the sensory properties, we believe that pulseN M R measurements will contribute to the understanding of the r61e of water for the eating quality of meat. Hence, the aim of this study was to characterize the water distribution within M . longissimus dorsi (LD) from Hampshire and Swedish Yorkshire pigs when raw and fried to 68°C and 80°C and to relate the water distribution within the samples to the sensory properties of LD.
216
StinaFjetkner-Modig, Eva Tornberg
MATERIALS AND METHODS Materials M. longissimus dorsi (LD, 7th vertebra thoracica-5th vertebra lumbaris) was taken from the right side of six Hampshire a n d six Swedish Yorkshire pigs. The animals were purebred and raised under identical conditions at Hermanstorp progeny testing station. The pigs for each breed came from three litters, each litter supplying one castrate and one gilt. The pigs were slaughtered during the week they reached a liveweight of 100 kg. The normal electrical stunning (110 V) slaughter method was used. The carcasses were conditioned at + 2°C for 3 days before the samples of LD were removed. Each sample of LD was cut into eighteen slices, 1.5 cm thick, and into two pieces of about 300 g. The analyses carried out on the various slices are shown in Fig, 1.
Meat colour value Meat colour value of LD was recorded by an EEL reflectance photometer (Evans Electroselenium Limited, Halstead, Great Britain) by using the Y-filter at a maximum wavelength of 550 nm. The reflectance value was recorded at three places on the surface of the transverse cut of LD at the 13th vertebra thoracica. The average value was registered. Meat with an EEL-colour value of 27 or higher is graded as PSE meat in the pig progeny testing.
pH value
r
=
Chemical a n a l y s i s
pH =
0
=
Drip l o s s
S
=
Sensory
EEL =
Co~oue vatue
SL
=
Sarcamere
F
Frying loss
W
=
Water dlstributian
=
analysis {eng,~h
Fig. 1. The analysis scheme for M. longissimus dorsi from pork. The sample slices were analysed raw and after frying to final centre temperatures of 68°C and 80°C.
Water distribution in porcine M. lonNssimus dorsi
217
Ultimate pH The p H value o f L D was recorded 3 days post m o r t e n by a p H meter (Portamess 651) with a glass electrode (Ingold 404) at the 13th vertebra thoracica. The measurements were carried out in the centre o f the transverse surface at a depth o f about 2 cm.
Drip loss The a m o u n t o f drip was determined by the m e t h o d of L u n d s t r 6 m e t al. (1984). Meat pieces (Fig. 1) o f about 300g were kept in trays at +2°C for 24 h and the drip was determined as percentage weight loss.
Frying and frying loss The 1-5-cm-thick slices were fried at 180°C on a double-sided griddle immediately prior to sensory evaluation. The frying was discontinued at a centre temperature of 68°C or 80°C for the slices shown in Fig. 1. The end-point temperature was recorded by thermocouples. Frying loss was calculated as the percentage loss in weight after frying. The frying loss was determined for two slices (Fig. 1) for each of the two end-point temperatures of 68°C and 80°C.
Sensory properties A trained expert panel consisting of ten men and women assessed four samples at each session. Sample pieces (4 x 3cm) were served one at a time immediately after frying. Between samples the assessors were asked to rinse their mouths with distilled water at r o o m temperature and eat an unsweetened biscuit. The sensory properties of fried LD were evaluated by a profile including the attributes: visible juiciness (1 = none, 9 = very large), initial juiciness (1 = none, 9 = very large), dryness in m o u t h (1 = none, 9 = very large), elasticity (1 = not elastic, 9 = very elastic), hardness (1 = v e r y soft, 9 = very hard), stringiness (1 = none, 9 = very large), chewing time (1 = v e r y short, 9 = v e r y long), chewing residual (1 = none, 9 = v e r y large) and total flavour intensity (1 = very weak, 9 = very strong) in accordance with the description given by Fjelkner-Modig (1986). The three attributes, visible and initial juiciness and dryness in m o u t h , were
218
Stina Fjetkner-Modig, Eva Tornberg
highly related, as were the attributes hardness, stringiness, chewing time and chewing residual (Fjelkner-Modig, 1986). Hence the sum of the attribute scores for visible and initial juiciness and dryness in m o u t h was considered representative of the total impression of juiciness and that of the attribute scores for hardness, stringiness, chewing time and chewing residual the total impression of toughness. These values were used for the statistical evaluation of the relationships between sensory properties and the water distribution in the meat.
Chemical analysis The contents of water (Nilsson, 1969), intramuscular lipids (SBRmethod, N M K L , No. 88, 1974), crude protein (Kjetdaht as modified by Nilsson, 1968) and hydroxyproline (Stegemann, 1958 as modified by Weber, 1973) were analysed for both raw and fried samples (Fig. I). Prior to the chemical analyses the epimysium sheath on :the slices was removed. The connective tissue content was determined by the amount of hydroxyproline in accordance with the method of Wyler (1972).
Sarcomere length A piece 10 x 5 m m was taken from the centre of those slices where the analysis of water distribution had been carried out. The pieces were cut along the fibre bundles into smaller pieces and fixed according to the procedure used by Cross et al. (1980-81). The length of the sarcomere diffraction bands was recorded by a helium-neon laser (Voyle, 1971). The sarcomere length was calculated in accordance with the formula used by Cross et al. (1980-81).
Water distribution The water distribution was recorded by a proton-pulse-NMR instrument (Bruker, Minispec, PC/20), mainly following the procedure of Tornberg & Nerbrink (1984). The measurements were carried out for raw and fried samples of LD, as shown in Fig. 1. Perpendicular to the slice surface a square-shaped (5.3 x 5-3 mm) rod was cut from the centre of the slice. The rod was cut carefully to avoid drip formation. The ends of the rod were cut off. The rod (about 8 m m long and with a weight of about 0.3 g) was put into a N M R tube (q57.5 mm) by means of a
Water distribution in porcine M. |on~ssimus dorsi
219
spatula and a glass rod. Muscle tissue and liquid, which had adhered to the tube wall, were removed with a paper tissue. Prior to the measurement the N M R tube with the sample rod was thermostatically held at 25°C for 30rain. The transverse relaxation time (T:) of the water protons within the meat rod was recorded at a frequency of 20 MHz. The T: measurements were made at 25°C by using the CarrPurcell-Meibom-Gill method (Meibom & Gill, 1958). The T: recordings were made for two different z spacings--500 and 4000 g s - - a n d at each measurement 49 scans were accumulated. The magnetisation value, M0 (corresponding to 100% water protons), was mostly determined at the "c spacing 4000 ~s. In two cases, however, it was easier to determine M0 for the measurements at z spacing 500 #s. M o was determined from the semilog plot of log M versus time and derived from the intercept of the straight line obtained from the first data arising from water proton relaxation. The relaxation data were analysed in multiexponential decay by curve decomposition using a microcomputer (Luxor ABC 80) as described b y Tornberg & Nerbrink (1984). The errors introduced by this calculation were for every sample estimated as being _< + 2 ms in T, for the fastest relaxing process and _< + 5 ms for the next process. The accuracy of the relative population of protons was estimated as being _< -!-_5% for the ' fastest process, _ +_6% for the next process and _< + 1% for the slowest process.
Statistical analyses Data were statistically treated and programmes of ANOVA one-way classification, Student's t test and linear regression were used.
RESULTS A N D DISCUSSION
Meat quality traits and composition All the samples of LD were taken from normal, graded pig carcasses. The Hampshire samples and the Swedish Yorkshire samples both had, on average, about the same EEL-colour value and ultimatepH (Table 1). This agrees well with an earlier investigation (Fjelkner-Modig & Persson,
23.2 5-4, 3'2 a 76'0;' 1.4 20.8" 3.2 1.84
x
Raw
(0'8) (o.i) (0'4) (0'4) (0-4) (0-6) (0.4) (0-05)
(SD)
(I "5) (1-2) (1. I) (0-4) (0.07) (2.6) (0-7) (0-7) (0.6) (0-8) (0.9) (0-8) (0.8) (0.6) (0.3)
7'0 7-I 1-8 6.4 3-6 3-0 3.8 2.6 3.8
(SD)
67"8" 2.5 28.0 2-9 1.74 21-2
x
68°C
lhtmpshire
5'9 5'6 2.7 5"7 4-2 3.4 4.4 3.2 4.0
66"3 2-5 29.4 2.9 1.60 28-2
x
(0.5) (0.9) (0.5) (0.5) (0.7) (0.6) (0.6) (0.5) (0.4)
(1 "0) (0.8) (0.5) (0.3) (0.06) (2-9)
(St))
80°C
24'0 5.5' 2"5 a 75" 1~' 1-3 22.9" 2.9 1.78
x
Raw
(4'0) (0.2) (0'9) (0"5) (0-3) (0.7) (0.6) (0.05)
(SD)
5.9 6'7 2.4 6-0 4.4 3.7 4.4 2.9 3.7
69'4 ¢ 2-3 27.9 2.7 1.68 21-0
x
(1'2) (0-9) (0.7) (0.9) (1"4) (0-9) (0.8) (0.8) (0.3)
(0"9) (0.5) (I .5) (0-3) (0.05) (2.1)
(St))
68°C
Swedish Yorkshire
5-2 5.1 3-4 5.8 4.7 3.8 4.6 3.4 3.7
67"6 2.7 29.2 2.6 1.55 27.2
x
(1'6) (1.6) (0.9) (0.8) (1"4) (0-9) (1.3) (1-0) (0-5)
(1.9) (0-9) (I .4) (0.4) (0.05) (2.5)
(SD)
80°C
* Valne in an intensity score with 9 units. Values within a row, descending from the same preparation and having the same letter, are significantly different. a, p <0.001; b, p < 0.01; c, d, e, p _<0.05.
EEL-colour wdue pit Drip (%) Water (%) lntra,nuscular lipids (%) Protein (%) Connective tissue (%) Sarcomere length (llm) Frying loss (%) Sensory analysis Visible juiciness* Initial juiciness* Dryness in mouth* Elasticity* Hardness* Stringiness* Chewing time* Chewing residual* Total Ilavour*
Trait
TABLE 1 Average Value (x) and Standard Deviation (SD) of the Different Traits of M. hmgissimus dorsi from Hampshire (n = 6) and Swedish Yorkshire (n = 6), when Analysed Raw and After Frying to End-point Temperatures of 68°C and 80°C
Water distribution in porcine M. lon~ssimus dorsi
221
1986). However, the average EEL-colour value of both the Hampshire and Yorkshire samples is about one unit higher than that found by Lundstr6m et at. (1984) for Yorkshire samples and those given by Fjelkner-Modig & Persson (1986) for samples of both the Hampshire and the Yorkshire breeds. The large standard deviation for the samples of Yorkshire (Table 1) is due to one sample having a high EEL-colour value, thus indicating PSE meat. The paleness and wetness appeared, however, only locally and a normal drip loss and ultimate pH was noted for the sample. A drip loss of 2-5% was recorded for the Yorkshire samples and this agrees well with information given by Lundstr6m et al. (1984). The Hampshire samples had a drip loss of 3-2%, which was significantly higher than those of the Yorkshire (Table 1). The results of the sarcomere length determination are given in Table 1. The Hampshire samples had throughout somewhat longer sarcomeres than the Yorkshire samples. This was also found by Fjelkner-Modig & Persson (1986). The shrinkage of sarcomere in length due to frying was the same for both breeds. Frying to 68°C resulted in a 5% reduction of sarcomere length and to 80°C in a reduction of about 13%. The contents of water, intramuscular lipids, protein and connective tissue were analysed for raw samples and also for samples fried to an internal temperature of 68°C and 80°C. The results of these analyses are given in Table 1. A breed difference was noted in the amount of water between the raw samples and those fried to 68°C. For raw samples the highest water content was noted for the Hampshire breed but, for the fried samples-both 68°C and 80°C--the highest content was noted for the Yorkshire breed. The Hampshire samples had a significantly lower amount of protein than those of Yorkshire, when raw, but for the fried samples the protein content was almost the same for the samples from both breeds, probably due to the different frying losses. The higher water content and the lower protein content of raw meat of Hampshire compared with that of Yorkshire has also been found by P. Barton-Gade, (pers. comm.). A rather low intramuscular lipid content was noted for the samples of both Hampshire and Yorkshire. Lundstr6m et al. (1984) and Fjelkner-Modig & Persson (1986) have reported an intramuscular lipid content of 1.7% for Yorkshire samples. For Hampshire samples the corresponding value was 2-0% (Fjelkner-Modig & Persson, 1986).
222
Stina Fjelkner-Modig, Eva Tornberg
Sensory analysis The samples of Hampshire were more juicy--both visibly and initially-more elastic, less dry in the mouth and tess hard than the Yorkshire samples. Furthermore, less stringiness, shorter chewing time and smaller chewing residual were noted for Hampshire than for Yorkshire. The total flavour intensity was about the same for both breeds (Table I). The breed differences were more pronounced at the final frying temperature of 68°C than at 80°C. The same breed differences have been found by Fjelkner-Modig (1986) but, in comparison, a somewhat higher juiciness and also a higher tenderness (i.e. shorter chewing time, smaller chewing residual and less stringiness and hardness) were noted in the present study for both the Hampshire and the Yorkshire breeds. This could be due to the treatment of the samples prior to sensory evaluation. Fjelkner-Modig (1986) used samples which had been frozen, stored at - 2 0 ° C for some time and thawed prior to frying. In the present investigation the sampIes were prepared fresh and in our own experience (S. Fjelkner-Modig, unpublished results) samples prepared flesh are assessed as being more juicy and tender than frozen samples. The sensory differences between the breeds were not statistically significant. However, the sensory differences between the breeds and also those between the two frying temperatures agreed well with those found by Fjelkner-Modig (1986), evaluating a much larger number of samples than in this study. Therefore, the breed differences noted in the present study are considered to be representative.
Water distribution Three different relaxation times were, in general, found for both the raw and the fried samples. The water with the shortest relaxation time has been designated as intracellular water and that with medium relaxation time as extracellular water (Hazlewood et al., 1974, Foster et al., 1976). These designations might be questionable. However, the aim of this investigation was primarily to study the effect of cooking on the water distribution within meat and the relationships between the water distribution within meat and the sensory properties and therefore the actual meanings of the designations are not considered here. The water with the longest relaxation time, i.e. 1 s or more, is 'free water', as this
Water distribution in porcine M. longissimus dorsi
223
TABLE 2 Average Value (x) a n d Standard Deviation (SD) of the Relaxation Times and the Relative Population of Protons Recorded by Proton-Pulse-NMR for Samples of M. longissimus dorsi from Hampshire (n = 6) and Yorkshire (n = 6) Pigs. The Recordings were Carried Out for Raw Samples and for Samples Fried to 680C and 80"C
Sample
Raw
68*C
80"C
Hampshire
Yorkshire
Relaxation time
Relative proton population
Relaxation time
Relative proton population
(ms)
(%)
(ms)
(%)
x
(SD)
x
(SD)
x
(SD)
x
(SD)
> 1 500 I00 a 39.7
(8) (1.0)
1-8 24.5 b 74.1 c
(0. I) (3.2) (2-5)
> 1 500 122a 40-1
(13) (3.5)
1.8 16.5 b 81-6c
(0.2) (4.4) (3.6)
> 1 500 108 30.6
(21) (3.3)
3.3d 18.6 77-7
(0-4) (5-0) (6.9)
> 1 500 108 29-8
(10) (1-2)
4.2 d 19.2 75-5
(0-7) (4.2) (4.0)
(0.4)
> 1 500
(26)
13.9
(5-2)
109
(19)
16.6
(2-4)
(3,8)
83-4
(8.0)
(3-0)
77.2
(2-3)
> 1 500
114 28.6
3.2e
27-3
4-2 ~
(0.8)
Values within a row and with the same letter are significantly different. p _< 0-01: a, b, e and d. p <_ 0"05: e.
water corresponds to protons that do not exchange with the small amount of water, which is restricted in motion (i.e. the hydration water). The mean relaxation times and the mean relative populations of protons are given per breed in Table 2. The intracellular water has the shortest relaxation time (T3---about 40 ms for raw samples and about 30 and 23ms for the samples fried to 68°C and 80°C, respectively. Renou et al. (1984) have reported two relaxation times, where the fastest were 41 and 43ms for raw LD from Pietrain and Large White pigs, respectively. Our results for 7",.agree fairly well with their observations. A somewhat longer T,- value is given by Pearson et al. (1974). They investigated porcine muscles--mainly L D - - a n d found a 7",-value of 53 ms, when the samples were analysed 24 h after death. Moreover, for muscle tissue of other species, relaxation times of intracellular water varying from 35 to 45 ms have been reported. Hazlewood et al. (1974) noted, for instance, a Tt value of 44ms for rat, Foster et al. (1976) a Ti
224
Stina Fjelkner-Modig, Eva Tornberg
value of 45ms for rat and 35ms for barnacle, Lillford et al. (1980) a Ti value of 47ms for beef and, finally, Tornberg & Nerbrink (1984) have reported a Ti value of about 36ms for beef. The relaxation time of the extracellular water in raw samples was significantly longer for Yorkshire (122 ms) than for Hampshire (100 ms) samples. Pearson et al. (1974) have reported a relaxation time for extracellular water (Te) of 140ms. For other species Te values of about 120 ms (beef, Tornberg & Nerbrink, 1984), about 150 ms (beef, Lillford et al., 1980), 200ms (rat, Hazlewood et al., 1974 and Foster et al., 1976) and 400ms (barnacle, Foster et al., 1976) have been reported. The raw samples of Hampshire had a lower relative population of protons with the shortest relaxation time (Pi) than those of Yorkshire (Table2). Pearson et al. (1974) reported an intracellular water content of 86% for pork. Both Hazlewood et al. (1974) and Foster et al. (1976) found Pi values of 80--82% for rat muscle tissue whereas Lillford et al. (1980) reported Pi values of 94% for beef. The relative population of protons at the extracellular space (Pe) was 16.5% for the Yorkshire samples (Table 2). Pearson et al. (1974) noted approximately the same amount of water (14%). For muscle tissue of other species mo_stly P~ values of 10-14% have been reported (Hazlewood et al.. 1974; Foster et al.. 1976; Tornberg & Nerbrink, 1984): Compared with Yorkshire the Hampshire samples had a high content of extracellular water. All the Hampshire samples recorded had a rather high Pe value and the variation within breed was smaller for Hampshire than for Yorkshire. About 1-8% of the total water in raw muscle tissue was free for both breeds. 7',- decreased after heating for both breeds. This suggests an increase in protein concentration within the fibres, which could originate from a shrinkage of the myofibrils. The decrease in T,. from 68°C to 80°C was much smaller than that from raw to 68°C. Furthermore, for the extracellular water the changes in relaxation time induced by frying differed between the two breeds. T~ increased on frying for the samples of Hampshire. For the samples of Yorkshire longer T e values were noted in the raw meat than for those of Hampshire, but on frying the T~ decreased and was almost the same for the samples fried to 68°C as for those fried to 80°C (Table2). The relative proportion of intracetlutar water (Pc) increased by frying for the samples of Hampshire (3.6% from raw to 68°C and 5-7% from
Water distribution in porcine M. lon~ssimus dorsi
225
68°C to 80°C) while, for those of Yorkshire, a decrease of 6-1% in the relative amount of intracellular water was noted from raw to 68°C followed by an increase of 1.7% from 68°C to 80°C. This implies that there is an effect of breed on the changes in water distribution, which occur during frying. The change, due to frying, in the relative proportion of extracellular water also differed from breed to breed. For samples fried to 68°C almost the same Pe values (about 19%) were noted for the two breeds. But compared with the raw sample the Hampshire samples fried to 68°C had a lower proportion of extracellular water whereas, for the Yorkshire samples, a tendency to the opposite ratio was noted. The Pe value decreased to some extent from 680C to 80°C for both the breeds. The relative amount of free water was higher in the fried samples than in the raw ones.
The relationships between meat quality traits, frying loss and water distribution The results of the statistical analysis of the relationships between drip loss, EEL-colour value, final pH and frying loss, on the one hand, and water distribution for raw meat, on the other, are given in Table 3. An analysis of the relationships between frying loss for the samples fried to 68°C and 80°C and the corresponding water distribution of these fried samples was also performed and the results are shown in Table 4. As is evident in Table 3, an influence of breed was noted for the relationships between the meat quality traits and the water distribution. In most cases lower correlation coefficients were obtained for the Hampshire than for the Yorkshire samples. This might partly be a result of the smaller variations within breed in pH, EEL-colour value and drip loss that were recorded for the Hampshire samples (Table 1). For the Hampshire samples a tendency was noted that an increase in the amount of extracellular water was related to an increase in drip loss and decreases in both pH and EEL-colour values. These relationships could be expected as meat with a high EEL-colour value, i.e. PSE meat," mostly has a low pH value and a high drip loss and the extracellular water is easier released than the intracellular water. For the Yorkshire samples the meat quality traits were most related to the relaxation time of the intracellular water. A short relaxation time of the intracellular
Stina Fjelkner-Modig, Eva Tornberg
226
TABLE 3 The Results--Given as Correlation Coefficients---of Linear Regression Analysis of Meat Quality Traits, Frying Loss, Sensory Properties and Water Distribution for Raw Samples of M. longissimus dorsi from Hampshire (H) and Swedish Yorkshire (Y) Pigs
Trait
Water
Breed
T~
distribution in
Ti
Pf
raw meat
Pc
Pi
Sensory properties Juiciness
(68"C)
Juiciness
(80"C)
Toughness
(68"C)
Toughness
(80"C)
H Y H Y H Y H Y
--0.28 -0-09 0.57 -0-43 0-02 0"54 -0-23 0.41
0-68 0.75 0-19 0-45 - 0-97"* -0.04 -0-72 -0.33
0.47 -0.31 -0-73 -0-41 - 0-17 0-04 0-15 0.16
0-24 0.06 -0-80(*) 0.51 - 0-13 -0'66 0-10 -0"55
-0-35 -0-05 0"81" -0-47 0-07 0-71 -0-22 0.55
H Y H Y
-0-44 0-31 0-31 0-16
0.10 0"14 0.51 0.17
0-63 -0"35 0-02 -0-52
0-85* --0"51 0.29 -0-41
-0.70 0"60 -0-39 0-52
H Y H Y H Y
0-00 0"ll 0" 11 0-25 0.07 0-05
0'05 0-06 -0"40 0-43 0.03 0.12
0"37 -0-19 -0"52 -0-05 - 0-22 -0.24
-0-19 0"22 0"49 -0-03 - 0.09 0-24
Frying loss Frying loss (68*C) Frying toss (80"C) Meat
quality traits
Drip loss pH EEL-colour value
-0"26 -0"60 -0-69 0-59 0-19 -0-63
T = Relaxation time. P = Relative population of protons. Indexf = free water, e = Extracellular water, i = Intracellular water. Significance level: p _<0-10: (*), p <_0-05: * and p < 0-01: **
water, indicating a high protein concentration and, most probably, more denatured proteins, was associated with a high drip loss, a low pH and a high EEL-colour value. These relationships would also be expected as meat with a high EEL-colour value represents PSE meat and this meat has more denatured proteins than that o f normal meat quatity. The frying loss for samples fried to an end-point temperature o f 68°C could fairly well be predicted by measuring the water distribution in the raw meat for both breeds (Table 3). A high amount of extracellular water was associated with a high frying loss (r = 0"85, p <_ 0"05) for
H Y H Y
H Y H Y H Y H Y
Breed
-0.26 0.83* 0.23 0.11
0.30 0.15 0.57 0.19 - 0.40 0.70 - 0-34 -0.10
T,
-0.41 0.27 - 0-31 -0.11
0.26 -0.96** 0.78(*) 0.40 -0.45 0.45 - 0.07 -0.36
T,
-0.18 0.60 0.64 0.44
0.76(*) 0.24 0.30 0.50 - 0.67 0.48 - 0.62 -0.31
P~
-0.09 -0.55 - 0'21 0.31
-0.43 -0.19 -0.78(*) - 0.42 0.59 -0-22 0.47 0.38
P,
Water distribution in fried meat
0.23 0.55 0.13 -0.12
0-17 0.26 0.81 * - 0.11 - 0.33 0.22 - 0.42 0-15
P,
68 68 80 80
68 68 80 80 68 68 80 80
Frying temperature ('C)
T = Relaxation time. P = Relative population of protons. Index f = Free water, e = Extracellular water, i = Intercellular water. Signilicance level: p < 0.10, (*); p < 0-05, * and p < 0.01, **
Frying loss
Frying loss
Frying loss
Toughness
Toughness
Juiciness
Juiciness
Sensory properties
Trait
TABLE 4 The Results--Given as Correlation Coefficients--of Linear Regression Analysis of Frying Loss, Sensory Properties and Water Distribution for Fried Samples of M. longissimus dorsi from Hampshire (H) and Swedish Yorkshire (Y) Pigs
to to "-4
,o.
~S' E 5"
5' ~. '~
228
Stina Fjelkner-Modig, Eva Tornberg
Hampshire samples, whereas, for the Yorkshire samples, there was a tendency for high flying loss to be related to a high amount of intracellular water (r = 0-60, not significant). This implies that quite different relationships were obtained for the two breeds. A breed influence was also noted for the samples fried to 80°C. For the Yorkshire samples similar relationships between water distribution and frying loss were found for the samples fried to 80°C as for those flied to 68°C, whereas for the Hampshire samples, fried to 80°C, the relaxation time of the intracellular water was a more prominent factor than the extracellular water content. Frying loss, when related to the fried product, was, in the case of Yorkshire, to some extent related to the amount of free water and for the samples fried to 68°C also to the relaxation time of the extraceUular water (T~). Also, for the Hampshire samples, when fried to 80°C, the highest correlation coefficient was noted between frying loss and the amount of free water. For the Hampshire samples fried to 68°C the shrinkage of myofibrils (i.e. a decrease in T,-) was to some extent (r = -0-41) related to the frying toss.
The relationships between sensory properties and water distribution The relationships between the sensory properties, juiciness and toughness and the water distribution in raw meat are, per breed, shown in Table 3. Moreover, the sensory properties of the samples fried to the temperatures 68°C and 80°C, as related to the water distribution of the corresponding samples, are shown in Table 4. In general; it may be noted that about the same correlation coefficient, but with opposite sign, is found between the amounts of intra- and extracellular water and the sensory properties, the frying loss and the meat quality traits. This is because the amounts ofintra- and extracellular water are highly related, i.e. a decrease in P,. is followed by an increase in Pe, and vice versa. As is evident from Table 3, a breed difference with regard to the relationships between sensory properties and the water distribution in raw meat was noted. For both the Hampshire and the Yorkshire samples, juiciness assessed at 68°C was mostly related to the relaxation time of the intracellular water, whereas the amounts of intracellular and extracellular water were more determining for the juiciness achieved at 80°C. However, the juiciness of Hampshire samples, when fried to
Water distribution in porcine M. longissimus dorsi
229
80°C, was positively related to the amount of intraceUular water (r = 0"81, p _< 0-05) but for the Yorkshire samples a lower and negative correlation coefficient (r = -0-47, not significant) was obtained. As is evident from Table 3, the toughness of Hampshire samples was related to the relaxation time of the intracellular water for both temperatures. For meat from this breed it seems favourable, with regard to tenderness, to have a long Ti value, implying that less shrinked fibres gives a more tender meat. For the Yorkshire samples rather high correlation coefficients were found between toughness and the amounts of extracellular and intracellular water. This implies that a low amount of intracellular water, i.e. shrinked myofibrils, and a high amount of extracellular water is favourable with regard to tenderness. The results shown in Table 3 also imply that the juiciness and tenderness of fried meat could fairly well be predicted by the water distribution in raw meat. However, due to the obvious breed influence on the noted relationships, different parameters have to be used for each breed. An influence of breed was also noted for the relationships between sensory properties and the water distribution within fried meat. For the Hampshire samples we noted a tendency for more free water to be associated with more tender and juicy meat (samples fried to 68°C). The juiciness of samples fried to 80°C was, however, related to the amounts of intra- and extracellular water. For the Yorkshire samples a significant relationship was noted between juiciness and the relaxation time of intracellular water. This implies that a shrinkage of the myofibrils is related to an increase in juiciness for the samples fried to 68°C ( r = 0-98, p_< 0.01). For the samples fried to 80°C the juiciness showed a tendency to be related to the amount of free water. The toughness of the Yorkshire samples was associated with the relaxation time of extracellular water. For samples fried to 80°C, in general low correlation coefficients were noted. Overall, more significant relationships were noted between the sensory properties and the water distribution, recorded in raw and fried samples, for the Hampshire than for the Yorkshire samples. This could be a result of the influence of the amount and composition of intramuscular lipids on the sensory properties reported for Yorkshire samples but not for Hampshire samples (Fjelkner-Modig & Tornberg, 1986). With regard to flavour and how it is related to the water distribution, no clear cut pattern can be deduced from the statistical analysis. This
230
Stina Fjelkner-Modig, Eva Tornberg
could be due to the small variation in flavour noted from sample to sample. It should be observed from Tables 3 and 4 that a fairly high percentage of the variation in sensory properties was explained by water distribution. Hence, these results suggest that changes in water holding of meat is an important factor for determining the sensory quality of pork. ACKNOWLEDGEMENTS We wish to thank Dr H~kan Rud~rus for valuable discussions and criticism during the preparation of the manuscript and Gertrud Larsson (MSc) for her excellent technical assistance with the proton-pulse-NMR measurements. REFERENCES Bendall, J. R. & Restall, D. J. (1983). Meat Sci., 8, 93. Bendall, J. R. & Wismer-Pedersen, J. (1962). J. Food Sci., 27, 144. Bognfir, A. & Pfichner, H.-J. (1983). Fleischw&ts., 63, 943. Bouton, P. E., Ford, A. L., Harris, O. V. & Ratcliff, D. (1975). 3". Food Sci., 40, 884. Bouton, P. E., Harris, P. V. & Shorthose, W. R. (1976a). J. Text. Stud., 7, 179. Bouton, P. E., Harris, P. V. & Shorthose, W. R, (1976b). J. Food Sci., 41, 1092. Cross, H. R., West, R. L. & Dutson, T. R. (1980-81). Meat Sci., 5, 261. Fjelkner-Modig, S. (1986). J. Food Quality. (In press.) Fjelkner-Modig, S. & Persson, J. (1986). J. Anita. Sci. (In press.) Fjelkner-Modig, S. & Tornberg, E. (1986). J. Food Quality. (In press.) Foster, K. R., Resing, H. A. & Garroway, A. N. (1976). Science, 194, 324. Fulton, L. & Davis, C. (1975). J. Amer. Diet. Ass., 67, 227. Harem, R. (!972). Kolloidchemie des Fleisches, Parey, Hamburg, Berlin. I-Iamm, R. (1975). In: Meat (Cole, D. J. A. & Lawrie, R. A. (Eds)), Butterworths, London, 321. Hazlewood, C. F., Chang, D. C., Nichols, B. L. & Woessner, D. E. (1974). J. Biophysical, 14, 583. Heffron, J. J. A. & Hegarty, P. V. J. (1974). Comp. Biochem. Physiol., 49A, 43. Hermansson, A.-M. & Lucisano, M. (1982). J. Food Sci., 47, 1955. Lawrie, R. A. (1979). Meat science (3rd edn), Pergamon Press, Oxford. Lillford, P. J., Clark, A. H. & Jones, D. V. (1980). In: Water in polymers (Rowland, S. P. (Ed.)), ACS Symposium Series No. 27, 177. Lundstr6m, K., Bj~irstorp, G. & Malmfors, G. (1984). Proc. Sci. Meet. Biophysical PSE-Muscle Analysis, Vienna, 311.
Water distribution in porcine M. longissimus dorsi
23t
Meibom, S. & Gill, D. (1958). Rev. Sci. Instrum., 29, 688. Nilsson, R. (1968). Swedish Meat Research Institute, Anal. Instr., No. 10. Nilsson, R. (1969). Swedish Meat Research Institute, Anal. Instr., No. 6. NMKL, No. 88 (1974). Fedt. Bestemmelse i kod og kodvarer efter SBR. Offer, G. (1984). 30th Fur. Meet. Meat Res. Workers, Bristol, 87. Offer, G. & Trinick, J. (1983). Meat Sci., 8, 245. Offer, G., Restall, D. & Trinick, J. (1984). In: Recent advances in the chemistry of meat (Bailey, A. K. (Ed.)), The Royal Soc. Chem., London. Spec. Publ. No. 47, 71. Pearson, R. T., Duff, I. D., Derbyshire, W. & Blanshard, J. M. V. (1974). Biochim. Biophys. Acta, 362, 188. Penny, I. F. (1977). J. Sci. Fd. Agric., 28, 329. Renou, J. P., Monin, G. & Sellier, P. (1984). Proc. Sci. Meet. Biophysical PSE-Muscle Analysis, Vienna, 65. Ritchey, S. J. (1965). J. Food Sci., 30, 375. Ritchey, S. J. & Hostetler, R. L. (1964). J. Food Sci., 29, 413. Sch6n, L. & Scheper, J. (1960). Ziicktungskunde, 32, 488. Stegemann, H. (1958). Hoppe-Seyler's Zeitschrift fiir Physiologische Chemie, 311, 41. Taylor, A. A. & Dant, S. J. (1971). J. Fd. Techn., 6, 131. Tornberg, E. & Nerbrink, O. (1984). 30th Fur. Meet. Meat Res. Workers, Bristol, 112. Voyle, C. A. (1971). 17th Fur. Meet. Meat Res. Workers, Bristol, 95. Weber, R. (1973). Technical Report No. 7, Genve, Technicon International Division SA. Wismer-Pedersen, J. (1971). In: The science of meat and meat products (Price, J. F. & Schweigert, B. S. (Eds)) (2nd edn), W. H. Freeman & Co., San Francisco, 177. Wyler, O. D. (1972). Fleischwirts., 52, 42.