- Email: [email protected]
Factors affecting the level of boar taint in entire male pigs as assessed by consumer sensory panel
Meat Science 61 (2002) 33–40 www.elsevier.com/locate/meatsci
Factors affecting the level of boar taint in entire male pigs as assessed by consumer sensory panel J. Babola,*, E.J. Squiresb, E.A. Gullettc a
Department of Food Science, Swedish University of Agricultural Sciences, PO Box 7051, 750 07 Uppsala, Sweden b Department of Animal and Poultry Science, University of Guelph, Guelph, Ontario, Canada N1G 2W1 c Department of Food Science, University of Guelph, Guelph, Ontario, Canada N1G 2W1 Received 12 March 2001; received in revised form 6 July 2001; accepted 8 July 2001
Abstract A sensory study was conducted to determine consumer responses to meat from entire male pigs in order to evaluate different factors involved in causing boar taint, including 16-androstene steroids, skatole, size of accessory sexual glands and live weight. Pork chops from 33 entire Yorkshire male pigs were divided into three groups according to the levels of 16-androstene steroids in salivary glands, high (H; 56–114 mg/g), medium (M; 35–55 mg/g) and low (L; 26–6 mg/g), and were distributed among 59 families. Pork chops from gilts were used as control (C). The families were asked to rate liking of cooking aroma, flavour and texture, and overall liking. Consumer ratings were best related to the size of bulbourethral gland and salivary gland, followed by the levels of 16androstene steroids in salivary gland. Meat from the L group received similar ratings to the C group, except the ratings for texture and overall liking were better than the C group. M samples were liked similarly to C, with the exception of some samples that had high fat skatole levels and which were liked less. H samples were liked less than C. The samples evaluated at the beginning of the experiment were liked more than those evaluated afterwards indicating sensitization of consumers to boar taint. The results indicate that levels of 16-androstenes in salivary glands can be used to estimate boar taint due to 16-androstenes; however, skatole measurements also need to be included. In addition, consumers are sensitive to some other maturity-related factors, which need to be considered when estimating the levels of boar taint. The effect of sensitization of consumers to boar taint requires more study. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Boar taint; 16-Androstene steroids; Skatole; Consumers
1. Introduction The incidence of boar taint in some entire male pigs is of concern when considering the use of entire males for pork production. Entire males have better production characteristics than castrates (reviewed by Babol & Squires, 1995; Walstra, 1974) which could be translated into significant economic gains for the pork industry. Raising entire males for pork production would also improve sustainability and animal welfare aspects of animal production systems. The major causes of the taint have been recognized as androstenone (Patterson, 1968) and other 16-androstene steroids (Garcı´a-Regueiro & Diaz, 1989) produced by testes, and skatole * Corresponding author. Tel.: +46-18-67-20-05; fax: +46-18-6730-80. E-mail address: [email protected] (J. Babol).
(Vold, 1970) which is produced by bacteria in the gut. These compounds accumulate in fat and generate offensive off-odours and off-flavours of meat, which are particularly well perceived by humans when the meat is heated. The levels of 16-androstene steroids in carcasses are related mostly to sexual maturity of pigs (Bonneau, 1982) whereas skatole levels are also dependent on environmental and dietary conditions (Jensen & Jensen 1997; Walstra et al., 1999). The relative contribution of skatole and androstenone to boar taint has been recently investigated in a large European study (Dijksterhuis et al., 2000; Matthews et al., 2000). This confirmed other studies that identified both compounds as being important in boar taint perception (Annor-Frempong, Nute, Whittington, & Wood, 1997), although with a stronger negative reaction towards skatole (Cameron, Penman, Fisken, Nute, Perry, & Whittington, 2000). The difficulties in clearly assessing the causes
0309-1740/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0309-1740(01)00159-0
34
J. Babol et al. / Meat Science 61 (2002) 33–40
of boar taint are understandable, taking into account the differences in odour perception among humans. Not all people are equally sensitive to androstenone (Koelega, 1980; Weiler et al., 2000) and these variations in sensitivity can be genetically determined (Wysocki & Beauchamp, 1984). Also, the variations in eating habits are important since processing, seasoning, as well as serving (hot vs. cold) have important effects on perception of boar taint (De Kock, Heinze, Potgieter, Dijksterhuis, & Minnaar, 2001; Pearson, Ngoddy, Price, & Larzelere, 1971). Consumer responses vary substantially from country to country (Matthews et al., 2000; Weiler et al., 2000). The consumer studies conducted in Canada showed a variable but generally relatively low sensitivity of consumers to boar taint (Cliplef, Grinwich, & Castell, 1984; Gullett, Partlow, Fisher, Halina, & Squires, 1993; Sather, 1995). There are indications that some other factors are involved in causing boar taint, since a significant portion of boar taint could not be explained by the levels of skatole and androstenone (Annor-Frempong, Nute, Wood, Whittington, & West, 1998; Babol, Squires, & Gullett, 1996; Bonneau, Le Denmat, Vaudelet, Veloso Nunes, Mortensen, & Mortensen, 1992). These factors might be of importance in light of the simulation studies by Bonneau et al. (2000) which showed that imposing low thresholds limits for skatole and androstenone levels in entire male carcasses would reduce, but not eliminate, consumer dissatisfaction with entire male meat. Several analytical methods have been developed to measure the degree of boar taint by measuring the amounts of boar taint causing compounds in carcasses (Claus, Herbert, & Dehnhard, 1997; Hansen-Møller & Andersen, 1994; Tuomola, Harpirio, Knuttila, Mikola, & Lo¨vgren, 1997). By using such methods in the slaughter houses, the tainted carcasses could be identified reducing the risk of meat with a high level of taint being sold to consumers. The Danish colorimetric procedure for measuring skatole equivalents in fat (Mortensen & Sørensen, 1984) has been used commercially with the level of skatole which cause a rejection of meat by consumers established for this method at 0.20–0.25 mg/g of fat. However, the methods for quantification of androstenone in fat, although they can be rapid (Claus et al., 1997), are not suitable for on-line use in the slaughter houses, and the threshold levels vary considerably between studies (Diestre, Oliver, Gispert, Arpa, & Arnau, 1990; Malmfors & Lundstro¨m, 1983). Boar taint due to 16-androstene steroids can be measured also by quantifying the total 16-androstenes in fat or in salivary glands by a colorimetric assays (Squires, Deng, & Wu, 1993; Squires, Gullett, Fisher, & Partlow, 1991). The objectives of this study were to evaluate the effects of different factors such as levels of skatole and 16-androstene steroids, live weight and sexual maturity on the perception of boar taint by consumers.
2. Materials and methods 2.1. Animals and sampling Pork loins were obtained from Yorkshire entire males slaughtered at the University of Guelph abattoir. At slaughter, weight of submaxillary salivary glands and length of bulbourethral gland were determined and samples of submaxillary salivary glands and backfat were taken for subsequent chemical analyses. The levels of 16-androstene steroids in salivary glands and in fat were measured by colorimetric procedures (Squires et al., 1993). The levels of skatole in fat were measured by colorimetric assay (Mortensen & Sørensen, 1984). Thirty-three entire males were selected and divided into three groups of 11 animals on the basis of the concentration of 16-androstene steroids in salivary glands. Groups were designated as high (H), medium (M) and low (L; Table 1). In addition to entire males, loins from six gilts were included as control (C). Gilts were slaughtered at 110 kg of live weight and analyzed for the amount of skatole in fat. Loins were obtained the next day after slaughter and vacuum packed and kept frozen at 35 C for a period of up to 3 months, until they were analyzed. 2.2. Consumer panel Fifty-nine families, who were pork consumers, were recruited from the university community to participate in the study which lasted 4 weeks. Each family received enough pork chops each week to feed the family. The pork chops from the four groups were randomized and delivered to the families so that each family evaluated one set of pork chops from each group, H, M, L, and C. The families did not know that they were evaluating meat from entire males. Choice of cooking method was left to the family, except that they were asked not to barbecue or cook in a sauce. The cooks were asked to rate the liking of the aroma during cooking and the entire family, except for small children, rated the pork for liking of flavour, texture, and overall liking. A sixpoint sensory scale was used: (1) like strongly, (2) like moderately, (3) like slightly, (4) dislike slightly, (5) dislike moderately, (6) dislike strongly. 2.3. Statistical analyses The effect of treatment on consumer responses was analyzed by ANOVA using the GLM procedure of SAS (SAS, 1991). The model included treatment, family, week number and the interaction term, weektreatment. The effect of week number on the level of boar taint parameters was analyzed using the same procedure. Tukey’s test was used to compare the means. The signal detection index (R-index; O’Mahony, 1988)
35
J. Babol et al. / Meat Science 61 (2002) 33–40
and M, and did not differ among M, L and C. For texture, the samples from L group were liked best, whereas no differences were observed between H and M groups. C group had intermediate values, which did not differ from the other treatments. For overall liking, the L group was evaluated as being liked the best followed by C and M, with H meat samples liked the least. Consumer liking of pork chops from entire males was also analyzed by the R-index procedure (Table 3). In this method, the sensory scores for entire male samples from a particular treatment group were compared to those for gilts, which were used as a control. R-indices represent the probability of detecting a difference between entire male and gilt samples. R-index of 50% indicates a similar liking between entire male and gilt meat; R-index above 50% indicates decreased liking, while R-index below 50% indicates increased liking of entire male meat compared to that of gilts. The H samples were liked less than control with respect to cooking aroma, flavour and overall liking. The liking of M samples was decreased only for overall liking (R-index, 55.9%). Meat from the L group differed from control only in texture; the R-index of 43.9% indicated that the texture of L chops was liked better than that of control samples. The frequency of dislike sensory scores (scores greater than 3) differed between the treatment groups for texture and overall liking but not for cooking aroma and flavour (Table 4). For texture, the L samples tended to
was used to evaluate the probability of detecting the difference between samples by consumers through comparing sensory scores for H, M, and L entire males to those for gilts (C). The frequency of dislike scores (scores higher than 3) was analyzed by Chi-squared (w2) test. Pearson correlation coefficients were calculated to relate sensory scores to various boar taint parameters. In addition, multiple regression analysis was conducted to evaluate various combinations of boar taint parameters and their squares and interaction terms in explaining the sensory responses.
3. Results The values of the different boar taint parameters in entire males are given in Table 1. With the exception of skatole, all other measurements were significantly different (P < 0.05) between H and L treatment groups. Except for levels of 16-androstenes in salivary glands and fat, the values of the parameters did not differ between H and M or M and L groups. Mean fat skatole levels were quite high in the H and M groups compared to the L group but the difference was not significant (P=0.231). The mean fat skatole concentration in gilts was 0.12 0.024 (standard error). A comparison of sensory scores revealed a treatment effect on all sensory attributes (P < 0.05; Table 2). The liking of cooking aroma and flavour was lowest for H
Table 1 Mean values and standard errors (S.E.) of boar taint parameters of entire male pigs with different levels of 16-androstene steroids in salivary glandsa Item
High (n=11)
Skatole (mg/g) 16–Androstenes in fat (mg/g) 16–Androstenes in salivary glands (mg/g) Bulbourethral gland length (cm) Salivary gland weight (g) Live weight (kg) a
Medium (n=11)
Low (n=11)
P-value
Mean
S.E.
Range
Mean
S.E.
Range
Mean
S.E.
Range
0.24 0.76a 78.5a 13.4a 81.5a 128.3a
0.057 0.12 6.36 0.51 6.60 4.76
0.06–0.68 0.38–1.75 56.3–114.1 9.5–15.3 47.7–118.1 106–157
0.22 0.37b 43.3b 11.9ab 65.8ab 121.8ab
0.062 0.52 2.05 0.72 6.85 3.70
0.04–0.64 0.05–0.67 35.1–54.7 9.1–17.1 46.1–115.4 106–143
0.11 0.33b 18.5c 10.5b 47.7b 110.2b
0.040 0.07 1.84 0.49 6.97 3.25
0.04–0.51 0.0–0.69 6.8–25.8 7.7–13.2 30.1–113.5 101–139
0.231 0.002 0.001 0.004 0.006 0.010
Means with different letters within a row differ (P <0.05).
Table 2 Mean consumer sensory scoresa and standard errors (S.E.) of meat from entire male pigs with different levels of 16-androstene steroids in salivary glandsb Sensory attribute
Cooking aroma (n=240)c Flavour (n=664) Texture (n=664) Overall (n=663) a b c
High (n=11)
Medium (n=11)
Low (n=11)
Control (n=6)
P-value
Mean
SE
Mean
SE
Mean
SE
Mean
SE
3.26a 2.64a 2.89a 2.80a
0.20 0.10 0.11 0.10
2.76ab 2.51ab 2.94a 2.74ab
0.20 0.10 0.10 0.11
2.59b 2.20b 2.34b 2.28c
0.17 0.09 0.09 0.09
2.54b 2.23b 2.61ab 2.43b
0.17 0.08 0.10 0.10
Sensory scale: (1) like strongly, (2) like moderately, (3) like slightly, (4) dislike slightly, (5) dislike moderately, (6) dislike strongly. Means with different letters within a row differ (P<0.05). Total number of evaluations.
0.008 0.001 0.001 0.001
36
J. Babol et al. / Meat Science 61 (2002) 33–40
receive lower percentage of dislike scores than other groups, whereas for overall liking the percentage of dislike scores was higher for both the H and M than for L and C samples. There was an effect of family on all sensory attributes (P=0.001). The rating of flavour, texture and overall liking were affected also by the week in which the samples were evaluated (Table 5). The interactions between week and treatment were not significant (P > 0.05). The samples evaluated in the first week of the study were liked more than those evaluated in the following weeks; however, the differences among weeks 2, 3 and 4 were not significant (P > 0.05). The distribution of samples to the families over the weeks was performed randomly according to treatment groups, which were established based on the levels of 16-androstene steroids in salivary glands. This resulted in differences in live weight in different weeks (P=0.001). Entire animals used in week 3 had a mean live weight of 135.0 4.15 (standard error) kg compared to 111.6 3.17, 112.1 3.48 and 119.8 4.05 kg for weeks 1, 2 and 4, respectively. Other measurements did not differ among the weeks (P > 0.05). The results of correlation analysis are summarized in Table 6. Cooking aroma, flavour, texture and overall
Table 3 Probabilities of detecting the differences by consumers (R-index values,%) between meat from entire male pigs with different levels of 16-androstene steroids in salivary glands and meat from gilts Sensory attribute
High
Medium
Low
Cooking aroma Flavour Texture Overall
62.3 57.2 54.9 56.9
53.3 55.5 54.5 55.9
NSa NS NS *
51.1 47.4 43.9 46.6
* ** NS *
NS NS * NS
*=P <0.05. **=P <0.01. a NS, P>0.05.
Table 4 Frequency of dislike scoresa (%) for meat from entire male pigs with different levels of 16-androstene steroids in salivary glands Sensory attribute
High
Medium
Low
Control
P-value
Cooking aroma Flavour Texture Overall
37.1 19.6 29.2 25.6
22.9 14.2 25.9 21.2
19.7 9.9 16.4 14.4
19.7 12.7 27.9 15.3
0.077 0.074 0.030 0.025
a Scores (4) dislike slightly, (5) dislike moderately and (6) dislike strongly are pooled together.
Table 5 Mean consumer sensory scoresa and standard errors (SE) for meat samples evaluated in different weeksb Sensory attribute
Cooking aroma (n=240)c Flavour (n=664) Texture (n=664) Overall (n=663) a b c
Week 1
Week 2
Week 3
Week 4
P-value
Mean
S.E.
Mean
S.E.
Mean
S.E.
Mean
S.E.
2.61 2.16b 2.36b 2.25b
0.16 0.07 0.08 0.07
2.98 2.48a 2.70ab 2.58a
0.20 0.10 0.10 0.10
2.93 2.53a 2.89a 2.73a
0.21 0.11 0.12 0.12
2.67 2.41ab 2.82a 2.68a
0.16 0.09 0.13 0.11
0.301 0.014 0.001 0.001
Sensory scale: (1) like strongly, (2) like moderately, (3) like slightly, (4) dislike slightly, (5) dislike moderately, (6) dislike strongly. Means with different letters within a row differ (P<0.05). Total number of evaluations.
Table 6 Correlations between sensory scores and various taint parameters in entire male pigs (n=33) Live weight Cooking aroma score Flavour score Texture score Overall liking score 16-Androstenes in salivary gland 16-Androstenes in fat Skatole in fat Bulbourethral gland length Salivary gland weight
0.38 0.46 0.48 0.52 0.60 0.43 0.46 0.66 0.80
*a ** ** ** *** * ** *** ***
*=P <0.05. **=P <0.01. ***=P< 0.001.23 a NS=P>0.05.
Cooking aroma score 0.74 0.58 0.67 0.38 0.31 0.01 0.64 0.59
*** *** *** * NSa NS *** ***
Flavour Texture score score
Overall score
0.81 0.91 0.44 0.36 0.28 0.59 0.58
0.41 0.26 0.49 0.57 0.61
*** *** *** * NS *** ***
0.93 0.34 0.18 0.42 0.47 0.48
*** * NS * ** **
* NS ** *** ***
16-Androstenes 16-Androstenes Skatole in salivary in fat in fat gland
0.73 0.31 0.57 0.60
*** NS *** ***
0.21 NS 0.32 NS 0.33 NS
Bulbourethral gland length
0.21 NS 0.36 * 0.83 ***
37
J. Babol et al. / Meat Science 61 (2002) 33–40 Table 7 Coefficients of determination (R2,%) for sensory attributes in different regression modelsa Independent variable
16-Androstenes in fat 16-Androstenes in salivary gland Skatole Bulbourethral gland Salivary gland Live weight Skatole+bulbourethral gland Skatole+salivary gland Skatole+bulbourethral gland+skatolebulbourethral gland
Dependent variable Aroma
Flavour
Texture
Overall
NSb 14.2 NS 40.8 34.4 14.6 NS NS 51.0
13.1 19.1 NS 35.0 34.6 21.5 NS NS 48.3
NS 12.0 18.3 22.3 22.7 23.4 33.6 NS NS
NS 17.2 24.3 32.3 37.8 27.4 46.9 46.2 NS
a
Only significant multiple regression models (P<0.05) in which the effect of both independent variables or interaction term on consumer responses of least one sensory attribute are presented. b Not significant (P >0.05).
liking were well correlated with bulbourethral gland length and salivary gland weight. Somewhat lower correlation coefficients were obtained between sensory scores and live weight and the levels of 16-androstene steroids in salivary glands. 16-Androstene steroids in fat, despite being closely correlated with 16-androstene steroids in salivary glands (r=0.73), were correlated only with flavour sensory scores. Fat skatole levels were correlated only with texture and overall liking scores. Multiple regression analysis indicated an effect (P < 0.05; Table 7) of two independent variables, skatole and bulbourethral gland length, on texture and overall liking scores, and skatole and salivary gland weight, on overall liking. For aroma and flavour, the interaction term between skatole and bulbourethral gland length was significant. All regressions were found to be linear.
4. Discussion The development of bulbourethral glands as well as submaxillary salivary glands in male pigs is dependent on sexual maturity (Bonneau & Russeil, 1985; Booth, 1980), and thus the size of these glands can be used as a measure of sexual development status. Consumer responses to meat from entire male pigs were more closely correlated with the bulbourethral gland length and salivary gland weight than with the levels of 16-androstene steroids or skatole in carcasses. This is in close agreement with the results of evaluation of boar taint by a trained sensory panel (Babol, Squires, & Gullett, 1996). Thus, it appears that off-odours and off-flavours present in entire male meat are caused not only by increased levels of 16-androstenes and skatole but also by some other factors related to sexual maturity. These unidentified factors are probably quite volatile, as indicated by a high correlation between bulbourethral gland length and cooking aroma score (r=0.64). They are also likely to possess a similar odour to androstenone or
skatole, since they were detected by a sensory panel trained with oil solutions of both compounds (Babol, Squires, & Gullett, 1996). The significant effect of the interaction between the bulbourethral gland length and the levels of skatole in fat on aroma and flavour scores suggests a synergistic effect of sexual maturity-related factors and skatole on the perception of boar taint by humans. Contrary to the trained panel results (Babol, Squires, & Gullett, 1996), no combined effect of skatole and 16androstene steroids on sensory scores was observed in this consumer study. Sensory scores were more closely related to 16-androstene steroids, as measured in salivary glands, than to fat skatole levels. Skatole was found to have no effect on the perception of cooking aroma and flavour of the meat (P > 0.1, r=0.01 and r=0.28, respectively). This does not agree with most consumer studies which describe skatole as one of the major factors affecting the consumer acceptance of entire male meat (Matthews et al., 2000; Mortensen, Bejerholm, & Pedersen, 1986). Also, skatole, as opposed to androstenone, is perceived by most people as an offensive off-odour and it is unlikely that Canadian consumers are not sensitive to it. One explanation for the lack of correlation between skatole fat levels and aroma and flavour scores could be the fact that 72.7% (24 out of 33) of the meat samples used in this study had fat skatole levels below the threshold level of 0.25 mg/g fat. Thus, the sensory scores of a large portion of samples were probably not affected by skatole because skatole concentrations in these samples were below the levels detectable by consumers. This effect could have also been a result of specific cooking procedures, which were not controlled during this experiment, such as the use of seasonings masking the odour of skatole, and the cooking time and temperature as well as serving procedures. Skatole, compared to androstenone, has theoretically lower phase-portioning coefficients, and therefore, it is to a greater extent responsible for smell than flavour
38
J. Babol et al. / Meat Science 61 (2002) 33–40
of entire male meat (De Kock et al., 2001; Dijksterhuis et al., 2000). However, due to the same reason, it is possible that a considerable amount of skatole evaporated quickly from meat during cooking before the cooks and later families evaluated the meat. Skatole levels were negatively correlated with texture and overall liking scores. This negative effect of skatole on the consumer perception of texture and consequently the rating of overall liking could be a result of a bitter taste of skatole (Lundstro¨m et al., 1988). The scores for all sensory attributes were correlated with the levels of 16-androstene steroids in salivary glands, but only the flavour scores were correlated with the levels of 16-androstenes in fat. This finding is somewhat surprising, since fatty tissue present in meat as interor intramuscular fat is a major determinant of sensory characteristics of pork, and it is the high concentrations of odorous compounds in fat that cause boar taint. In studies with trained sensory panels, similar results were obtained by Babol, Squires, and Gullett (1996), but Squires et al. (1991) reported similar correlations between sensory scores for boar taint and the levels of 16androstenes in both salivary glands and fat. 16-Androstene steroids are synthesized in testes and their levels in pigs sharply increase at puberty (Bonneau, 1982). The increase in the synthesis of 16-androstene steroids is not linear and in more mature male pigs, fat levels of these steroids fluctuate considerably. The levels of 16-androstenes in salivary glands do not change at the same rate as those in fat due to the presence of a specific binding protein, pheromaxein, in salivary glands (Babol, Squires, & Bonneau, 1996). The levels of pheromaxein are related to the function of testes and consequently to the sexual development of male pigs. Hence, the concentrations of 16-androstenes in salivary glands of entire male pigs at any given time are probably more closely associated with the stage of sexual development than are the levels of these steroids in fat. This could be the reason for the correlations of 16-androstenes in salivary glands, but not 16-androstenes in fat, with the bulbourethral gland length and salivary gland weight (Table 6). Considering this, it is likely that the levels of 16-androstenes in salivary glands were more closely related to the sensory scores than were the levels of 16-androstenes in fat because they estimated the taint due not only to 16androstenes but also that due to sexual maturity. The negative evaluation of meat from H entire males clearly indicates that the level of 16-androstene steroids in salivary glands above 56.3 mg/g substantially increases the risk of rejection of pork by consumers. The positive evaluation of meat from L entire males (16androstenes in salivary gland below 25.8 mg/g) suggests that in these samples no taint was detected. The M samples (16-androstenes in salivary gland, from 35.1 to 54.7 mg/g) were evaluated more harshly than C samples only for overall liking. Thus, the M chops were liked
only slightly less than C chops, and virtually all M samples that were negatively evaluated had high skatole levels in fat (from 0.24 to 0.64 mg/g). Therefore, levels of 16-androstene steroids in salivary glands above 55 mg/g can result in boar taint due to 16-androstenes in fresh meat. However, in order to select tainted carcasses, skatole in fat should also be quantified so that no meat tainted due to this compound could enter the marketplace. Estimating boar taint on the slaughter line has to be as simple as possible to make the production of entire males economically justifiable. The use of the bulbourethral gland length as a prescreening test could reduce the number of animals requiring quantification of 16androstenes and skatole. The design of this study did not allow setting threshold limits for the bulbourethral gland length. However, in our study there were meat samples from seven entire males that did not receive negative evaluation (mean sensory scores above 3) from consumers, and all of these animals had bulbourethral glands of no more than 10.7 cm long. This value is similar to that proposed by Bonneau and Russeil (1985), who suggested that the same cut-off level can be used for all entire male pigs regardless of their breed and carcass weight. However, contrary to those findings, it seems that there are breeds, such as Duroc, that can have short bulbourethral glands and, at the same time, high level of boar taint compounds (Squires & Lou, 1995). The effect of sexual maturity on boar taint level requires further studies. It is likely that including a measure of sexual maturity status when sorting out tainted carcasses, and thus accounting for sexual maturity related factors other than 16-androstenes and skatole, would increase consumer satisfaction with entire male meat, which otherwise is difficult to achieve (Annor-Frempong et al., 1998; Bonneau et al., 2000). Another aspect that requires further investigation is the possible effect of sensitization of people to boar taint. Wysocki, Dorries, and Beauchamp (1989) reported that ability to smell androstenone was induced in initially insensitive people who were exposed to androstenone for six weeks. In this study, the samples evaluated in the first week were liked more than those evaluated in subsequent weeks. The decrease in liking of flavour and overall liking and the tendency of decreased liking of texture in week 2 compared to week 1 could not be explained by differences in the levels of skatole, 16-androstenes or other boar taint parameters. This indicates that people were more sensitive to boar taint after becoming familiar with the odour and flavour of tainted meat. It is thus possible that consumers who may initially accept meat from entire males may become less tolerable to it over time. However, it cannot be excluded that in this experiment the increased liking observed only for the first sample was due to a psychological effect and not to a sensitization effect. More research is required to evaluate the possible sensitization
J. Babol et al. / Meat Science 61 (2002) 33–40
to boar taint. If this effect is confirmed, the threshold limits for taint-causing compounds should be established at such levels as not to induce sensitivity in consumers. It needs to be emphasized that sensitivity of consumers to androstenone is a major factor accounting for the differences in the perception of boar taint by different populations (Weiler et al., 2000), and therefore, particular attention needs to be directed towards this compound. In contrast, the sensitivity of humans towards skatole has been shown to decrease over time (De Kock et al., 2001). Consumers used to meat from entire male pigs, such as those in UK, may also object less to tainted meat than those in the countries where male pigs are castrated because they perceive the taint as a normal pork odour/flavour (Matthews et al., 2000). Consequently, it may be possible that initially established low threshold values for boar taint compounds could be raised later after a prolonged exposure to entire male meat. According to the analysis of frequency of dislike scores and R-index analysis, meat from L group received better ratings than C for texture (Tables 3 and 4). ANOVA analysis showed a higher overall liking of L meat compared to that of gilts, while texture was evaluated similarly for both groups (Table 2). Therefore, it can be concluded that meat from entire males with low levels of 16-androstene steroids and skatole was liked more than that from gilts. This is in agreement with another Canadian consumer study (Sather, 1995). The reasons for this effect are not clear. It could be due to increased tenderness or due to a higher leanness of meat from entire males compared to that from gilts (reviewed by Babol, & Squires, 1995) making such meat more appealing and, as a result, more liked by consumers. Neither tenderness nor fatness level of the meat were determined in this investigation.
5. Conclusions Our results indicate that Canadian consumers like meat from entire male pigs with low levels of 16androstenes and skatole at least as well as meat from gilts. Therefore, providing that the level of boar taint is controlled, the use of entire male pigs for pork production would not result in consumer dissatisfaction. Measurement of 16-androstene steroids in salivary gland can be used as a measure for boar taint due to 16androstenes. In order to avoid negative reactions of consumers skatole analysis should be included as well. In addition, consumers are sensitive to some other maturity-related factors involved in causing boar taint. More research is needed to identify those factors; however, for practical purposes of boar taint estimation the measures of sexual development, such as the length of bulbourethral gland, may be effective. Finally, con-
39
sumers may become more sensitive to boar taint after initial exposure. This effect needs to be further investigated and should be considered when determining threshold levels for taint causing factors.
Acknowledgements This study was funded by the Natural Sciences and Engineering Research Council, the Ontario Ministry of Agriculture, Food and Rural Affairs and Ontario Pork.
References Annor-Frempong, I. E., Nute, G. R., Whittington, F. W., & Wood, J. D. (1997). The problem of taint in pork—III. Odour profile of pork fat and the interrelationships between androstenone, skatole and indole concentrations. Meat Science, 47, 63–76. Annor-Frempong, I. E., Nute, G. R., Wood, J. D., Whittington, F. W., & West, A. (1998). The measurement of the responses to different odour intensities of ‘boar taint’ using a sensory panel and an electronic nose. Meat Science, 50, 139–151. Babol, J., & Squires, E. J. (1995). Quality of meat from entire male pigs. Food Research International, 28, 201–212. Babol, J., Squires, E. J., & Bonneau, M. (1996). Factors regulating the concentrations of 16-androstene steroids in submaxillary salivary glands of pigs. Journal of Animal Science, 74, 413–419. Babol, J., Squires, E. J., & Gullett, E. A. (1996). Investigation of factors responsible for boar taint. Food Research International, 28, 573–581. Bonneau, M. (1982). Compounds responsible for boar taint, with special emphasis on androstenone: a review. Livestock Production Science, 9, 687–707. Bonneau, M., Le Denmat, M., Vaudelet, J. C., Veloso Nunes, I. R., Mortensen, A. B., & Mortensen, H. P. (1992). Contribution of fat androstenone and skatole to boar taint: I. Sensory attributes of fat and pork meat. Livestock Production Science, 32, 63–80. Bonneau, M., & Russeil, P. (1985). The size of Cowper’s (bulbo-urethral) gland as an estimate of boar taint on the slaughter line. Livestock Production Science, 13, 329–334. Bonneau, M., Walstra, P., Claudi-Magnussen, C., Kempster, A. J., Tornberg, E., Fischer, K., Diestre, A., Siret, F., Chevillon, P., Claus, R., Dijksterhuis, G., Punter, P., Matthews, K. R., Agerhem, H., Be´ague, M. P., Oliver, M. A., Gispert, M., Weiler, U., von Seth, G., Leask, H., Font i Furnols, M., Homer, D. B., & Cook, G. L. (2000). An international study on the importance of androstenone and skatole for boar taint: IV. Simulation studies on consumer dissatisfaction with entire male pork and the effect of sorting carcasses on the slaughter line, main conclusions and recommendations. Meat Science, 54, 285–295. Booth, W. D. (1980). A study of some major testicular steroids in the pig in relation to their effect on the development of male characteristics in the prepubertally castrated boar. Journal of Reproduction and Fertility, 59, 155–162. Cameron, N. D., Penman, J. C., Fisken, A. C., Nute, G. R., Perry, A. M., & Whittington, F. W. (2000). Boar taint in pigs selected for components of efficient lean growth rate. Meat Science, 54, 147–153. Claus, R., Herbert, E., & Dehnhard, M. (1997). Comparative determination of the boar taint steroid androstenone in pig adipose tissue by a rapid enzyme immunoassay and HPLC method. Archiv fu¨r Lebensmittelhygiene, 48, 25–48. Cliplef, R. L., Grinwich, D. L., & Castell, A. G. (1984). Consumer acceptance of fresh pork and pork products from littermate boars and barrows. Canadian Journal of Animal Science, 64, 21–27.
40
J. Babol et al. / Meat Science 61 (2002) 33–40
De Kock, H. L., Heinze, P. H., Potgieter, C. M., Dijksterhuis, G. B., & Minnaar, A. (2001). Temporal aspects related to the perception of skatole and androstenone, the major boar odour compounds. Meat Science, 57, 61–70. Diestre, A., Oliver, M. A., Gispert, M., Arpa, I., & Arnau, J. (1990). Consumer responses to fresh meat and meat products from barrows and boars with different level of boar taint. Animal Production, 50, 519–530. Dijksterhuis, G. B., Engel, B., Walstra, P., Font i Furnols, M., Agerhem, H., Fischer, K., Oliver, M. A., Claudi-Magnussen, C., Siret, F., Be´ague, M. P., Homer, D. B., & Bonneau, M. (2000). An international study on the importance of androstenone and skatole for boar taint: II. Sensory evaluation by trained panels in seven European countries. Meat Science, 54, 261–269. Garcı´a-Regueiro, J. A., & Diaz, I. (1989). Evaluation of the contribution of skatole, indole, androstenone and androstenols to boar taint in back fat of pigs by HPLC and capillary gas chromatography (CGC). Meat Science, 25, 307–316. Gullett, E. A., Partlow, G. D., Fisher, K. R. S., Halina, W. G., & Squires, E. J. (1993). Effect of pig sex on consumer ratings for pork chops and bacon. Food Quality and Preference, 4, 201–205. Hansen-Møller, J., & Andersen, J. R. (1994). Boar taint—analytical alternatives. Fleischwirtschaft, 74(9), 963–966. Jensen, M. T., & Jensen, B. B. (1997). Tasty meat through diet design. Feed Mix, 5(4), 8–12. Koelega, H. S. (1980). Preference for and sensitivity to the odours of androstenone and musk. In H. van der Starre (Ed.), Olfaction and taste, VII (pp. 438). Oxford, UK: Information Retrieval Ltd. Lundstro¨m, K., Malmfors, B., Malmfors, G., Stern, S., Petersson, H., Mortensen, A. B., & Sørensen, S. E. (1988). Skatole, androstenone and taint in boars fed two different diets. Livestock Production Science, 18, 55–67. Malmfors, B., & Lundstro¨m, K. (1983). Consumer reactions to boar taint—a review. Livestock Production Science, 10, 187–196. Matthews, K. R., Homer, D. B., Punter, P., Be´ague, M.-P., Gispert, M., Kempster, A. J., Agerhem, H., Claudi-Magnussen, C., Fischer, K., Siret, F., Leask, H., Font i Furnols, M., & Bonneau, M. (2000). An international study on the importance of androstenone and skatole for boar taint: III. Consumer survey in seven European countries. Meat Science, 54, 271–283. Mortensen, A. B., Bejerholm, C., & Pedersen, J. K. (1986). Consumer test of meat form entire males in relation to skatole in backfat. In Proceedings of the 32th European Meeting of Meat Research Workers, Ghent (pp.23–66). Mortensen, A. B., & Sørensen, S. E. (1984). Relationship between boar taint and skatole determined with a new analysis method. In Proceedings of the 30th European Meeting of Meat Research Workers, Bristol (pp. 394–396). O’Mahony, M. (1988). Sensory difference and preference testing: the use of signal detection measures. In H. Moskowitz (Ed.), Applied sensory analysis of foods. (Vol. 1) (pp. 145–176). Boca Raton, FL: CRC Press.
Patterson, R. L. S. (1968). 5a-androst-16-ene-3-one, compound responsible for taint in boar fat. Journal of the Science of Food and Agriculture, 19, 31–38. Pearson, A. M., Ngoddy, S., Price, J. F., & Larzelere, H. E. (1971). Panel acceptability of products containing boar meat. Journal of Animal Science, 33, 26–29. Sather, A. P. (1995). Consumers and pork: Canadian consumer acceptance of fresh pork from entire male pigs. Meat Focus International, 4(9), 365–371. SAS. (1991). SAS/STAT User’s guide, Version 6.07. Cary, NC: SAS Institute Inc. Squires, E. J., Deng, W., & Wu, L. (1993). Colorimetric method for the determination of 16-androstene steroids in carcasses. In M. Bonneau (Ed.), Measurement and prevention of boar taint in entire male pigs (pp. 41–48). Paris: INRA Editions. Squires, E. J., Gullett, E. A., Fisher, K. R. S., & Partlow, G. D. (1991). Comparison of androst-16-ene determined by a colorimetric assay with boar taint estimated by a trained sensory panel. Journal of Animal Science, 69, 1092–1100. Squires, E. J., Lou, Y. (1995). Levels of boar taint in purebred entire male pigs in Canada. In Proceedings of a Meeting of the EAAP Working Group, Production and Utilization of Meat from Entire Male Pigs, Milton Keynes. Tuomola, M., Harpirio, R., Knuttila, P., Mikola, H., & Lo¨vgren, T. (1997). Time-resolved fluoroimmunoassay for the measurement of androstenone in porcine serum and fat samples. Journal of Agricultural and Food Chemistry, 45, 3529–3534. Vold, E. (1970). Fleishproduktionseigenschaften bei Ebern und Kastraten. IV. Organoleptische und gaschromatografische Untersuchungen wasserdampfflu¨chtiger Stoffe des Ru¨ckenspeckes von Ebern. Meldinger Norlandbruckhoegskole, 49, 1–25. Walstra, P. (1974). Fattening of young boars: quantification of positive and negative aspects. Livestock Production Science, 1, 187–196. Walstra, P., Claudi-Magnussen, C., Chevillon, P., von Seth, G., Diestre, A., Matthews, K. R., Homer, D. B., & Bonneau, M. (1999). An international study on the importance of androstenone and skatole for boar taint: levels of androstenone and skatole by country and season. Livestock Production Science, 62, 15–28. Weiler, U., Font i Furnols, M., Fischer, K., Kemmer, H., Oliver, M. A., Gispert, M., Dobrowolski, A., & Claus, R. (2000). Influence of differences in sensitivity of Spanish and German consumers to perceive androstenone on the acceptance of boar meat differing in skatole and androstenone concentrations. Meat Science, 54, 297– 304. Wysocki, C. J., & Beauchamp, G. K. (1984). Ability to smell androstenone is genetically determined. Proceedings of the National Academy of Science of the USA, 81, 4899–4902. Wysocki, C. J., Dorries, K. M., & Beauchamp, G. K. (1989). Ability to perceive androstenone can be acquired by ostensibly anosmic people. Proceedings of the National Academy of Science of the USA, 86, 7976–7978.
Factors affecting the level of boar taint in entire male pigs as assessed by consumer sensory panel J. Babola,*, E.J. Squiresb, E.A. Gullettc a
Department of Food Science, Swedish University of Agricultural Sciences, PO Box 7051, 750 07 Uppsala, Sweden b Department of Animal and Poultry Science, University of Guelph, Guelph, Ontario, Canada N1G 2W1 c Department of Food Science, University of Guelph, Guelph, Ontario, Canada N1G 2W1 Received 12 March 2001; received in revised form 6 July 2001; accepted 8 July 2001
Abstract A sensory study was conducted to determine consumer responses to meat from entire male pigs in order to evaluate different factors involved in causing boar taint, including 16-androstene steroids, skatole, size of accessory sexual glands and live weight. Pork chops from 33 entire Yorkshire male pigs were divided into three groups according to the levels of 16-androstene steroids in salivary glands, high (H; 56–114 mg/g), medium (M; 35–55 mg/g) and low (L; 26–6 mg/g), and were distributed among 59 families. Pork chops from gilts were used as control (C). The families were asked to rate liking of cooking aroma, flavour and texture, and overall liking. Consumer ratings were best related to the size of bulbourethral gland and salivary gland, followed by the levels of 16androstene steroids in salivary gland. Meat from the L group received similar ratings to the C group, except the ratings for texture and overall liking were better than the C group. M samples were liked similarly to C, with the exception of some samples that had high fat skatole levels and which were liked less. H samples were liked less than C. The samples evaluated at the beginning of the experiment were liked more than those evaluated afterwards indicating sensitization of consumers to boar taint. The results indicate that levels of 16-androstenes in salivary glands can be used to estimate boar taint due to 16-androstenes; however, skatole measurements also need to be included. In addition, consumers are sensitive to some other maturity-related factors, which need to be considered when estimating the levels of boar taint. The effect of sensitization of consumers to boar taint requires more study. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Boar taint; 16-Androstene steroids; Skatole; Consumers
1. Introduction The incidence of boar taint in some entire male pigs is of concern when considering the use of entire males for pork production. Entire males have better production characteristics than castrates (reviewed by Babol & Squires, 1995; Walstra, 1974) which could be translated into significant economic gains for the pork industry. Raising entire males for pork production would also improve sustainability and animal welfare aspects of animal production systems. The major causes of the taint have been recognized as androstenone (Patterson, 1968) and other 16-androstene steroids (Garcı´a-Regueiro & Diaz, 1989) produced by testes, and skatole * Corresponding author. Tel.: +46-18-67-20-05; fax: +46-18-6730-80. E-mail address: [email protected] (J. Babol).
(Vold, 1970) which is produced by bacteria in the gut. These compounds accumulate in fat and generate offensive off-odours and off-flavours of meat, which are particularly well perceived by humans when the meat is heated. The levels of 16-androstene steroids in carcasses are related mostly to sexual maturity of pigs (Bonneau, 1982) whereas skatole levels are also dependent on environmental and dietary conditions (Jensen & Jensen 1997; Walstra et al., 1999). The relative contribution of skatole and androstenone to boar taint has been recently investigated in a large European study (Dijksterhuis et al., 2000; Matthews et al., 2000). This confirmed other studies that identified both compounds as being important in boar taint perception (Annor-Frempong, Nute, Whittington, & Wood, 1997), although with a stronger negative reaction towards skatole (Cameron, Penman, Fisken, Nute, Perry, & Whittington, 2000). The difficulties in clearly assessing the causes
0309-1740/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0309-1740(01)00159-0
34
J. Babol et al. / Meat Science 61 (2002) 33–40
of boar taint are understandable, taking into account the differences in odour perception among humans. Not all people are equally sensitive to androstenone (Koelega, 1980; Weiler et al., 2000) and these variations in sensitivity can be genetically determined (Wysocki & Beauchamp, 1984). Also, the variations in eating habits are important since processing, seasoning, as well as serving (hot vs. cold) have important effects on perception of boar taint (De Kock, Heinze, Potgieter, Dijksterhuis, & Minnaar, 2001; Pearson, Ngoddy, Price, & Larzelere, 1971). Consumer responses vary substantially from country to country (Matthews et al., 2000; Weiler et al., 2000). The consumer studies conducted in Canada showed a variable but generally relatively low sensitivity of consumers to boar taint (Cliplef, Grinwich, & Castell, 1984; Gullett, Partlow, Fisher, Halina, & Squires, 1993; Sather, 1995). There are indications that some other factors are involved in causing boar taint, since a significant portion of boar taint could not be explained by the levels of skatole and androstenone (Annor-Frempong, Nute, Wood, Whittington, & West, 1998; Babol, Squires, & Gullett, 1996; Bonneau, Le Denmat, Vaudelet, Veloso Nunes, Mortensen, & Mortensen, 1992). These factors might be of importance in light of the simulation studies by Bonneau et al. (2000) which showed that imposing low thresholds limits for skatole and androstenone levels in entire male carcasses would reduce, but not eliminate, consumer dissatisfaction with entire male meat. Several analytical methods have been developed to measure the degree of boar taint by measuring the amounts of boar taint causing compounds in carcasses (Claus, Herbert, & Dehnhard, 1997; Hansen-Møller & Andersen, 1994; Tuomola, Harpirio, Knuttila, Mikola, & Lo¨vgren, 1997). By using such methods in the slaughter houses, the tainted carcasses could be identified reducing the risk of meat with a high level of taint being sold to consumers. The Danish colorimetric procedure for measuring skatole equivalents in fat (Mortensen & Sørensen, 1984) has been used commercially with the level of skatole which cause a rejection of meat by consumers established for this method at 0.20–0.25 mg/g of fat. However, the methods for quantification of androstenone in fat, although they can be rapid (Claus et al., 1997), are not suitable for on-line use in the slaughter houses, and the threshold levels vary considerably between studies (Diestre, Oliver, Gispert, Arpa, & Arnau, 1990; Malmfors & Lundstro¨m, 1983). Boar taint due to 16-androstene steroids can be measured also by quantifying the total 16-androstenes in fat or in salivary glands by a colorimetric assays (Squires, Deng, & Wu, 1993; Squires, Gullett, Fisher, & Partlow, 1991). The objectives of this study were to evaluate the effects of different factors such as levels of skatole and 16-androstene steroids, live weight and sexual maturity on the perception of boar taint by consumers.
2. Materials and methods 2.1. Animals and sampling Pork loins were obtained from Yorkshire entire males slaughtered at the University of Guelph abattoir. At slaughter, weight of submaxillary salivary glands and length of bulbourethral gland were determined and samples of submaxillary salivary glands and backfat were taken for subsequent chemical analyses. The levels of 16-androstene steroids in salivary glands and in fat were measured by colorimetric procedures (Squires et al., 1993). The levels of skatole in fat were measured by colorimetric assay (Mortensen & Sørensen, 1984). Thirty-three entire males were selected and divided into three groups of 11 animals on the basis of the concentration of 16-androstene steroids in salivary glands. Groups were designated as high (H), medium (M) and low (L; Table 1). In addition to entire males, loins from six gilts were included as control (C). Gilts were slaughtered at 110 kg of live weight and analyzed for the amount of skatole in fat. Loins were obtained the next day after slaughter and vacuum packed and kept frozen at 35 C for a period of up to 3 months, until they were analyzed. 2.2. Consumer panel Fifty-nine families, who were pork consumers, were recruited from the university community to participate in the study which lasted 4 weeks. Each family received enough pork chops each week to feed the family. The pork chops from the four groups were randomized and delivered to the families so that each family evaluated one set of pork chops from each group, H, M, L, and C. The families did not know that they were evaluating meat from entire males. Choice of cooking method was left to the family, except that they were asked not to barbecue or cook in a sauce. The cooks were asked to rate the liking of the aroma during cooking and the entire family, except for small children, rated the pork for liking of flavour, texture, and overall liking. A sixpoint sensory scale was used: (1) like strongly, (2) like moderately, (3) like slightly, (4) dislike slightly, (5) dislike moderately, (6) dislike strongly. 2.3. Statistical analyses The effect of treatment on consumer responses was analyzed by ANOVA using the GLM procedure of SAS (SAS, 1991). The model included treatment, family, week number and the interaction term, weektreatment. The effect of week number on the level of boar taint parameters was analyzed using the same procedure. Tukey’s test was used to compare the means. The signal detection index (R-index; O’Mahony, 1988)
35
J. Babol et al. / Meat Science 61 (2002) 33–40
and M, and did not differ among M, L and C. For texture, the samples from L group were liked best, whereas no differences were observed between H and M groups. C group had intermediate values, which did not differ from the other treatments. For overall liking, the L group was evaluated as being liked the best followed by C and M, with H meat samples liked the least. Consumer liking of pork chops from entire males was also analyzed by the R-index procedure (Table 3). In this method, the sensory scores for entire male samples from a particular treatment group were compared to those for gilts, which were used as a control. R-indices represent the probability of detecting a difference between entire male and gilt samples. R-index of 50% indicates a similar liking between entire male and gilt meat; R-index above 50% indicates decreased liking, while R-index below 50% indicates increased liking of entire male meat compared to that of gilts. The H samples were liked less than control with respect to cooking aroma, flavour and overall liking. The liking of M samples was decreased only for overall liking (R-index, 55.9%). Meat from the L group differed from control only in texture; the R-index of 43.9% indicated that the texture of L chops was liked better than that of control samples. The frequency of dislike sensory scores (scores greater than 3) differed between the treatment groups for texture and overall liking but not for cooking aroma and flavour (Table 4). For texture, the L samples tended to
was used to evaluate the probability of detecting the difference between samples by consumers through comparing sensory scores for H, M, and L entire males to those for gilts (C). The frequency of dislike scores (scores higher than 3) was analyzed by Chi-squared (w2) test. Pearson correlation coefficients were calculated to relate sensory scores to various boar taint parameters. In addition, multiple regression analysis was conducted to evaluate various combinations of boar taint parameters and their squares and interaction terms in explaining the sensory responses.
3. Results The values of the different boar taint parameters in entire males are given in Table 1. With the exception of skatole, all other measurements were significantly different (P < 0.05) between H and L treatment groups. Except for levels of 16-androstenes in salivary glands and fat, the values of the parameters did not differ between H and M or M and L groups. Mean fat skatole levels were quite high in the H and M groups compared to the L group but the difference was not significant (P=0.231). The mean fat skatole concentration in gilts was 0.12 0.024 (standard error). A comparison of sensory scores revealed a treatment effect on all sensory attributes (P < 0.05; Table 2). The liking of cooking aroma and flavour was lowest for H
Table 1 Mean values and standard errors (S.E.) of boar taint parameters of entire male pigs with different levels of 16-androstene steroids in salivary glandsa Item
High (n=11)
Skatole (mg/g) 16–Androstenes in fat (mg/g) 16–Androstenes in salivary glands (mg/g) Bulbourethral gland length (cm) Salivary gland weight (g) Live weight (kg) a
Medium (n=11)
Low (n=11)
P-value
Mean
S.E.
Range
Mean
S.E.
Range
Mean
S.E.
Range
0.24 0.76a 78.5a 13.4a 81.5a 128.3a
0.057 0.12 6.36 0.51 6.60 4.76
0.06–0.68 0.38–1.75 56.3–114.1 9.5–15.3 47.7–118.1 106–157
0.22 0.37b 43.3b 11.9ab 65.8ab 121.8ab
0.062 0.52 2.05 0.72 6.85 3.70
0.04–0.64 0.05–0.67 35.1–54.7 9.1–17.1 46.1–115.4 106–143
0.11 0.33b 18.5c 10.5b 47.7b 110.2b
0.040 0.07 1.84 0.49 6.97 3.25
0.04–0.51 0.0–0.69 6.8–25.8 7.7–13.2 30.1–113.5 101–139
0.231 0.002 0.001 0.004 0.006 0.010
Means with different letters within a row differ (P <0.05).
Table 2 Mean consumer sensory scoresa and standard errors (S.E.) of meat from entire male pigs with different levels of 16-androstene steroids in salivary glandsb Sensory attribute
Cooking aroma (n=240)c Flavour (n=664) Texture (n=664) Overall (n=663) a b c
High (n=11)
Medium (n=11)
Low (n=11)
Control (n=6)
P-value
Mean
SE
Mean
SE
Mean
SE
Mean
SE
3.26a 2.64a 2.89a 2.80a
0.20 0.10 0.11 0.10
2.76ab 2.51ab 2.94a 2.74ab
0.20 0.10 0.10 0.11
2.59b 2.20b 2.34b 2.28c
0.17 0.09 0.09 0.09
2.54b 2.23b 2.61ab 2.43b
0.17 0.08 0.10 0.10
Sensory scale: (1) like strongly, (2) like moderately, (3) like slightly, (4) dislike slightly, (5) dislike moderately, (6) dislike strongly. Means with different letters within a row differ (P<0.05). Total number of evaluations.
0.008 0.001 0.001 0.001
36
J. Babol et al. / Meat Science 61 (2002) 33–40
receive lower percentage of dislike scores than other groups, whereas for overall liking the percentage of dislike scores was higher for both the H and M than for L and C samples. There was an effect of family on all sensory attributes (P=0.001). The rating of flavour, texture and overall liking were affected also by the week in which the samples were evaluated (Table 5). The interactions between week and treatment were not significant (P > 0.05). The samples evaluated in the first week of the study were liked more than those evaluated in the following weeks; however, the differences among weeks 2, 3 and 4 were not significant (P > 0.05). The distribution of samples to the families over the weeks was performed randomly according to treatment groups, which were established based on the levels of 16-androstene steroids in salivary glands. This resulted in differences in live weight in different weeks (P=0.001). Entire animals used in week 3 had a mean live weight of 135.0 4.15 (standard error) kg compared to 111.6 3.17, 112.1 3.48 and 119.8 4.05 kg for weeks 1, 2 and 4, respectively. Other measurements did not differ among the weeks (P > 0.05). The results of correlation analysis are summarized in Table 6. Cooking aroma, flavour, texture and overall
Table 3 Probabilities of detecting the differences by consumers (R-index values,%) between meat from entire male pigs with different levels of 16-androstene steroids in salivary glands and meat from gilts Sensory attribute
High
Medium
Low
Cooking aroma Flavour Texture Overall
62.3 57.2 54.9 56.9
53.3 55.5 54.5 55.9
NSa NS NS *
51.1 47.4 43.9 46.6
* ** NS *
NS NS * NS
*=P <0.05. **=P <0.01. a NS, P>0.05.
Table 4 Frequency of dislike scoresa (%) for meat from entire male pigs with different levels of 16-androstene steroids in salivary glands Sensory attribute
High
Medium
Low
Control
P-value
Cooking aroma Flavour Texture Overall
37.1 19.6 29.2 25.6
22.9 14.2 25.9 21.2
19.7 9.9 16.4 14.4
19.7 12.7 27.9 15.3
0.077 0.074 0.030 0.025
a Scores (4) dislike slightly, (5) dislike moderately and (6) dislike strongly are pooled together.
Table 5 Mean consumer sensory scoresa and standard errors (SE) for meat samples evaluated in different weeksb Sensory attribute
Cooking aroma (n=240)c Flavour (n=664) Texture (n=664) Overall (n=663) a b c
Week 1
Week 2
Week 3
Week 4
P-value
Mean
S.E.
Mean
S.E.
Mean
S.E.
Mean
S.E.
2.61 2.16b 2.36b 2.25b
0.16 0.07 0.08 0.07
2.98 2.48a 2.70ab 2.58a
0.20 0.10 0.10 0.10
2.93 2.53a 2.89a 2.73a
0.21 0.11 0.12 0.12
2.67 2.41ab 2.82a 2.68a
0.16 0.09 0.13 0.11
0.301 0.014 0.001 0.001
Sensory scale: (1) like strongly, (2) like moderately, (3) like slightly, (4) dislike slightly, (5) dislike moderately, (6) dislike strongly. Means with different letters within a row differ (P<0.05). Total number of evaluations.
Table 6 Correlations between sensory scores and various taint parameters in entire male pigs (n=33) Live weight Cooking aroma score Flavour score Texture score Overall liking score 16-Androstenes in salivary gland 16-Androstenes in fat Skatole in fat Bulbourethral gland length Salivary gland weight
0.38 0.46 0.48 0.52 0.60 0.43 0.46 0.66 0.80
*a ** ** ** *** * ** *** ***
*=P <0.05. **=P <0.01. ***=P< 0.001.23 a NS=P>0.05.
Cooking aroma score 0.74 0.58 0.67 0.38 0.31 0.01 0.64 0.59
*** *** *** * NSa NS *** ***
Flavour Texture score score
Overall score
0.81 0.91 0.44 0.36 0.28 0.59 0.58
0.41 0.26 0.49 0.57 0.61
*** *** *** * NS *** ***
0.93 0.34 0.18 0.42 0.47 0.48
*** * NS * ** **
* NS ** *** ***
16-Androstenes 16-Androstenes Skatole in salivary in fat in fat gland
0.73 0.31 0.57 0.60
*** NS *** ***
0.21 NS 0.32 NS 0.33 NS
Bulbourethral gland length
0.21 NS 0.36 * 0.83 ***
37
J. Babol et al. / Meat Science 61 (2002) 33–40 Table 7 Coefficients of determination (R2,%) for sensory attributes in different regression modelsa Independent variable
16-Androstenes in fat 16-Androstenes in salivary gland Skatole Bulbourethral gland Salivary gland Live weight Skatole+bulbourethral gland Skatole+salivary gland Skatole+bulbourethral gland+skatolebulbourethral gland
Dependent variable Aroma
Flavour
Texture
Overall
NSb 14.2 NS 40.8 34.4 14.6 NS NS 51.0
13.1 19.1 NS 35.0 34.6 21.5 NS NS 48.3
NS 12.0 18.3 22.3 22.7 23.4 33.6 NS NS
NS 17.2 24.3 32.3 37.8 27.4 46.9 46.2 NS
a
Only significant multiple regression models (P<0.05) in which the effect of both independent variables or interaction term on consumer responses of least one sensory attribute are presented. b Not significant (P >0.05).
liking were well correlated with bulbourethral gland length and salivary gland weight. Somewhat lower correlation coefficients were obtained between sensory scores and live weight and the levels of 16-androstene steroids in salivary glands. 16-Androstene steroids in fat, despite being closely correlated with 16-androstene steroids in salivary glands (r=0.73), were correlated only with flavour sensory scores. Fat skatole levels were correlated only with texture and overall liking scores. Multiple regression analysis indicated an effect (P < 0.05; Table 7) of two independent variables, skatole and bulbourethral gland length, on texture and overall liking scores, and skatole and salivary gland weight, on overall liking. For aroma and flavour, the interaction term between skatole and bulbourethral gland length was significant. All regressions were found to be linear.
4. Discussion The development of bulbourethral glands as well as submaxillary salivary glands in male pigs is dependent on sexual maturity (Bonneau & Russeil, 1985; Booth, 1980), and thus the size of these glands can be used as a measure of sexual development status. Consumer responses to meat from entire male pigs were more closely correlated with the bulbourethral gland length and salivary gland weight than with the levels of 16-androstene steroids or skatole in carcasses. This is in close agreement with the results of evaluation of boar taint by a trained sensory panel (Babol, Squires, & Gullett, 1996). Thus, it appears that off-odours and off-flavours present in entire male meat are caused not only by increased levels of 16-androstenes and skatole but also by some other factors related to sexual maturity. These unidentified factors are probably quite volatile, as indicated by a high correlation between bulbourethral gland length and cooking aroma score (r=0.64). They are also likely to possess a similar odour to androstenone or
skatole, since they were detected by a sensory panel trained with oil solutions of both compounds (Babol, Squires, & Gullett, 1996). The significant effect of the interaction between the bulbourethral gland length and the levels of skatole in fat on aroma and flavour scores suggests a synergistic effect of sexual maturity-related factors and skatole on the perception of boar taint by humans. Contrary to the trained panel results (Babol, Squires, & Gullett, 1996), no combined effect of skatole and 16androstene steroids on sensory scores was observed in this consumer study. Sensory scores were more closely related to 16-androstene steroids, as measured in salivary glands, than to fat skatole levels. Skatole was found to have no effect on the perception of cooking aroma and flavour of the meat (P > 0.1, r=0.01 and r=0.28, respectively). This does not agree with most consumer studies which describe skatole as one of the major factors affecting the consumer acceptance of entire male meat (Matthews et al., 2000; Mortensen, Bejerholm, & Pedersen, 1986). Also, skatole, as opposed to androstenone, is perceived by most people as an offensive off-odour and it is unlikely that Canadian consumers are not sensitive to it. One explanation for the lack of correlation between skatole fat levels and aroma and flavour scores could be the fact that 72.7% (24 out of 33) of the meat samples used in this study had fat skatole levels below the threshold level of 0.25 mg/g fat. Thus, the sensory scores of a large portion of samples were probably not affected by skatole because skatole concentrations in these samples were below the levels detectable by consumers. This effect could have also been a result of specific cooking procedures, which were not controlled during this experiment, such as the use of seasonings masking the odour of skatole, and the cooking time and temperature as well as serving procedures. Skatole, compared to androstenone, has theoretically lower phase-portioning coefficients, and therefore, it is to a greater extent responsible for smell than flavour
38
J. Babol et al. / Meat Science 61 (2002) 33–40
of entire male meat (De Kock et al., 2001; Dijksterhuis et al., 2000). However, due to the same reason, it is possible that a considerable amount of skatole evaporated quickly from meat during cooking before the cooks and later families evaluated the meat. Skatole levels were negatively correlated with texture and overall liking scores. This negative effect of skatole on the consumer perception of texture and consequently the rating of overall liking could be a result of a bitter taste of skatole (Lundstro¨m et al., 1988). The scores for all sensory attributes were correlated with the levels of 16-androstene steroids in salivary glands, but only the flavour scores were correlated with the levels of 16-androstenes in fat. This finding is somewhat surprising, since fatty tissue present in meat as interor intramuscular fat is a major determinant of sensory characteristics of pork, and it is the high concentrations of odorous compounds in fat that cause boar taint. In studies with trained sensory panels, similar results were obtained by Babol, Squires, and Gullett (1996), but Squires et al. (1991) reported similar correlations between sensory scores for boar taint and the levels of 16androstenes in both salivary glands and fat. 16-Androstene steroids are synthesized in testes and their levels in pigs sharply increase at puberty (Bonneau, 1982). The increase in the synthesis of 16-androstene steroids is not linear and in more mature male pigs, fat levels of these steroids fluctuate considerably. The levels of 16-androstenes in salivary glands do not change at the same rate as those in fat due to the presence of a specific binding protein, pheromaxein, in salivary glands (Babol, Squires, & Bonneau, 1996). The levels of pheromaxein are related to the function of testes and consequently to the sexual development of male pigs. Hence, the concentrations of 16-androstenes in salivary glands of entire male pigs at any given time are probably more closely associated with the stage of sexual development than are the levels of these steroids in fat. This could be the reason for the correlations of 16-androstenes in salivary glands, but not 16-androstenes in fat, with the bulbourethral gland length and salivary gland weight (Table 6). Considering this, it is likely that the levels of 16-androstenes in salivary glands were more closely related to the sensory scores than were the levels of 16-androstenes in fat because they estimated the taint due not only to 16androstenes but also that due to sexual maturity. The negative evaluation of meat from H entire males clearly indicates that the level of 16-androstene steroids in salivary glands above 56.3 mg/g substantially increases the risk of rejection of pork by consumers. The positive evaluation of meat from L entire males (16androstenes in salivary gland below 25.8 mg/g) suggests that in these samples no taint was detected. The M samples (16-androstenes in salivary gland, from 35.1 to 54.7 mg/g) were evaluated more harshly than C samples only for overall liking. Thus, the M chops were liked
only slightly less than C chops, and virtually all M samples that were negatively evaluated had high skatole levels in fat (from 0.24 to 0.64 mg/g). Therefore, levels of 16-androstene steroids in salivary glands above 55 mg/g can result in boar taint due to 16-androstenes in fresh meat. However, in order to select tainted carcasses, skatole in fat should also be quantified so that no meat tainted due to this compound could enter the marketplace. Estimating boar taint on the slaughter line has to be as simple as possible to make the production of entire males economically justifiable. The use of the bulbourethral gland length as a prescreening test could reduce the number of animals requiring quantification of 16androstenes and skatole. The design of this study did not allow setting threshold limits for the bulbourethral gland length. However, in our study there were meat samples from seven entire males that did not receive negative evaluation (mean sensory scores above 3) from consumers, and all of these animals had bulbourethral glands of no more than 10.7 cm long. This value is similar to that proposed by Bonneau and Russeil (1985), who suggested that the same cut-off level can be used for all entire male pigs regardless of their breed and carcass weight. However, contrary to those findings, it seems that there are breeds, such as Duroc, that can have short bulbourethral glands and, at the same time, high level of boar taint compounds (Squires & Lou, 1995). The effect of sexual maturity on boar taint level requires further studies. It is likely that including a measure of sexual maturity status when sorting out tainted carcasses, and thus accounting for sexual maturity related factors other than 16-androstenes and skatole, would increase consumer satisfaction with entire male meat, which otherwise is difficult to achieve (Annor-Frempong et al., 1998; Bonneau et al., 2000). Another aspect that requires further investigation is the possible effect of sensitization of people to boar taint. Wysocki, Dorries, and Beauchamp (1989) reported that ability to smell androstenone was induced in initially insensitive people who were exposed to androstenone for six weeks. In this study, the samples evaluated in the first week were liked more than those evaluated in subsequent weeks. The decrease in liking of flavour and overall liking and the tendency of decreased liking of texture in week 2 compared to week 1 could not be explained by differences in the levels of skatole, 16-androstenes or other boar taint parameters. This indicates that people were more sensitive to boar taint after becoming familiar with the odour and flavour of tainted meat. It is thus possible that consumers who may initially accept meat from entire males may become less tolerable to it over time. However, it cannot be excluded that in this experiment the increased liking observed only for the first sample was due to a psychological effect and not to a sensitization effect. More research is required to evaluate the possible sensitization
J. Babol et al. / Meat Science 61 (2002) 33–40
to boar taint. If this effect is confirmed, the threshold limits for taint-causing compounds should be established at such levels as not to induce sensitivity in consumers. It needs to be emphasized that sensitivity of consumers to androstenone is a major factor accounting for the differences in the perception of boar taint by different populations (Weiler et al., 2000), and therefore, particular attention needs to be directed towards this compound. In contrast, the sensitivity of humans towards skatole has been shown to decrease over time (De Kock et al., 2001). Consumers used to meat from entire male pigs, such as those in UK, may also object less to tainted meat than those in the countries where male pigs are castrated because they perceive the taint as a normal pork odour/flavour (Matthews et al., 2000). Consequently, it may be possible that initially established low threshold values for boar taint compounds could be raised later after a prolonged exposure to entire male meat. According to the analysis of frequency of dislike scores and R-index analysis, meat from L group received better ratings than C for texture (Tables 3 and 4). ANOVA analysis showed a higher overall liking of L meat compared to that of gilts, while texture was evaluated similarly for both groups (Table 2). Therefore, it can be concluded that meat from entire males with low levels of 16-androstene steroids and skatole was liked more than that from gilts. This is in agreement with another Canadian consumer study (Sather, 1995). The reasons for this effect are not clear. It could be due to increased tenderness or due to a higher leanness of meat from entire males compared to that from gilts (reviewed by Babol, & Squires, 1995) making such meat more appealing and, as a result, more liked by consumers. Neither tenderness nor fatness level of the meat were determined in this investigation.
5. Conclusions Our results indicate that Canadian consumers like meat from entire male pigs with low levels of 16androstenes and skatole at least as well as meat from gilts. Therefore, providing that the level of boar taint is controlled, the use of entire male pigs for pork production would not result in consumer dissatisfaction. Measurement of 16-androstene steroids in salivary gland can be used as a measure for boar taint due to 16androstenes. In order to avoid negative reactions of consumers skatole analysis should be included as well. In addition, consumers are sensitive to some other maturity-related factors involved in causing boar taint. More research is needed to identify those factors; however, for practical purposes of boar taint estimation the measures of sexual development, such as the length of bulbourethral gland, may be effective. Finally, con-
39
sumers may become more sensitive to boar taint after initial exposure. This effect needs to be further investigated and should be considered when determining threshold levels for taint causing factors.
Acknowledgements This study was funded by the Natural Sciences and Engineering Research Council, the Ontario Ministry of Agriculture, Food and Rural Affairs and Ontario Pork.
References Annor-Frempong, I. E., Nute, G. R., Whittington, F. W., & Wood, J. D. (1997). The problem of taint in pork—III. Odour profile of pork fat and the interrelationships between androstenone, skatole and indole concentrations. Meat Science, 47, 63–76. Annor-Frempong, I. E., Nute, G. R., Wood, J. D., Whittington, F. W., & West, A. (1998). The measurement of the responses to different odour intensities of ‘boar taint’ using a sensory panel and an electronic nose. Meat Science, 50, 139–151. Babol, J., & Squires, E. J. (1995). Quality of meat from entire male pigs. Food Research International, 28, 201–212. Babol, J., Squires, E. J., & Bonneau, M. (1996). Factors regulating the concentrations of 16-androstene steroids in submaxillary salivary glands of pigs. Journal of Animal Science, 74, 413–419. Babol, J., Squires, E. J., & Gullett, E. A. (1996). Investigation of factors responsible for boar taint. Food Research International, 28, 573–581. Bonneau, M. (1982). Compounds responsible for boar taint, with special emphasis on androstenone: a review. Livestock Production Science, 9, 687–707. Bonneau, M., Le Denmat, M., Vaudelet, J. C., Veloso Nunes, I. R., Mortensen, A. B., & Mortensen, H. P. (1992). Contribution of fat androstenone and skatole to boar taint: I. Sensory attributes of fat and pork meat. Livestock Production Science, 32, 63–80. Bonneau, M., & Russeil, P. (1985). The size of Cowper’s (bulbo-urethral) gland as an estimate of boar taint on the slaughter line. Livestock Production Science, 13, 329–334. Bonneau, M., Walstra, P., Claudi-Magnussen, C., Kempster, A. J., Tornberg, E., Fischer, K., Diestre, A., Siret, F., Chevillon, P., Claus, R., Dijksterhuis, G., Punter, P., Matthews, K. R., Agerhem, H., Be´ague, M. P., Oliver, M. A., Gispert, M., Weiler, U., von Seth, G., Leask, H., Font i Furnols, M., Homer, D. B., & Cook, G. L. (2000). An international study on the importance of androstenone and skatole for boar taint: IV. Simulation studies on consumer dissatisfaction with entire male pork and the effect of sorting carcasses on the slaughter line, main conclusions and recommendations. Meat Science, 54, 285–295. Booth, W. D. (1980). A study of some major testicular steroids in the pig in relation to their effect on the development of male characteristics in the prepubertally castrated boar. Journal of Reproduction and Fertility, 59, 155–162. Cameron, N. D., Penman, J. C., Fisken, A. C., Nute, G. R., Perry, A. M., & Whittington, F. W. (2000). Boar taint in pigs selected for components of efficient lean growth rate. Meat Science, 54, 147–153. Claus, R., Herbert, E., & Dehnhard, M. (1997). Comparative determination of the boar taint steroid androstenone in pig adipose tissue by a rapid enzyme immunoassay and HPLC method. Archiv fu¨r Lebensmittelhygiene, 48, 25–48. Cliplef, R. L., Grinwich, D. L., & Castell, A. G. (1984). Consumer acceptance of fresh pork and pork products from littermate boars and barrows. Canadian Journal of Animal Science, 64, 21–27.
40
J. Babol et al. / Meat Science 61 (2002) 33–40
De Kock, H. L., Heinze, P. H., Potgieter, C. M., Dijksterhuis, G. B., & Minnaar, A. (2001). Temporal aspects related to the perception of skatole and androstenone, the major boar odour compounds. Meat Science, 57, 61–70. Diestre, A., Oliver, M. A., Gispert, M., Arpa, I., & Arnau, J. (1990). Consumer responses to fresh meat and meat products from barrows and boars with different level of boar taint. Animal Production, 50, 519–530. Dijksterhuis, G. B., Engel, B., Walstra, P., Font i Furnols, M., Agerhem, H., Fischer, K., Oliver, M. A., Claudi-Magnussen, C., Siret, F., Be´ague, M. P., Homer, D. B., & Bonneau, M. (2000). An international study on the importance of androstenone and skatole for boar taint: II. Sensory evaluation by trained panels in seven European countries. Meat Science, 54, 261–269. Garcı´a-Regueiro, J. A., & Diaz, I. (1989). Evaluation of the contribution of skatole, indole, androstenone and androstenols to boar taint in back fat of pigs by HPLC and capillary gas chromatography (CGC). Meat Science, 25, 307–316. Gullett, E. A., Partlow, G. D., Fisher, K. R. S., Halina, W. G., & Squires, E. J. (1993). Effect of pig sex on consumer ratings for pork chops and bacon. Food Quality and Preference, 4, 201–205. Hansen-Møller, J., & Andersen, J. R. (1994). Boar taint—analytical alternatives. Fleischwirtschaft, 74(9), 963–966. Jensen, M. T., & Jensen, B. B. (1997). Tasty meat through diet design. Feed Mix, 5(4), 8–12. Koelega, H. S. (1980). Preference for and sensitivity to the odours of androstenone and musk. In H. van der Starre (Ed.), Olfaction and taste, VII (pp. 438). Oxford, UK: Information Retrieval Ltd. Lundstro¨m, K., Malmfors, B., Malmfors, G., Stern, S., Petersson, H., Mortensen, A. B., & Sørensen, S. E. (1988). Skatole, androstenone and taint in boars fed two different diets. Livestock Production Science, 18, 55–67. Malmfors, B., & Lundstro¨m, K. (1983). Consumer reactions to boar taint—a review. Livestock Production Science, 10, 187–196. Matthews, K. R., Homer, D. B., Punter, P., Be´ague, M.-P., Gispert, M., Kempster, A. J., Agerhem, H., Claudi-Magnussen, C., Fischer, K., Siret, F., Leask, H., Font i Furnols, M., & Bonneau, M. (2000). An international study on the importance of androstenone and skatole for boar taint: III. Consumer survey in seven European countries. Meat Science, 54, 271–283. Mortensen, A. B., Bejerholm, C., & Pedersen, J. K. (1986). Consumer test of meat form entire males in relation to skatole in backfat. In Proceedings of the 32th European Meeting of Meat Research Workers, Ghent (pp.23–66). Mortensen, A. B., & Sørensen, S. E. (1984). Relationship between boar taint and skatole determined with a new analysis method. In Proceedings of the 30th European Meeting of Meat Research Workers, Bristol (pp. 394–396). O’Mahony, M. (1988). Sensory difference and preference testing: the use of signal detection measures. In H. Moskowitz (Ed.), Applied sensory analysis of foods. (Vol. 1) (pp. 145–176). Boca Raton, FL: CRC Press.
Patterson, R. L. S. (1968). 5a-androst-16-ene-3-one, compound responsible for taint in boar fat. Journal of the Science of Food and Agriculture, 19, 31–38. Pearson, A. M., Ngoddy, S., Price, J. F., & Larzelere, H. E. (1971). Panel acceptability of products containing boar meat. Journal of Animal Science, 33, 26–29. Sather, A. P. (1995). Consumers and pork: Canadian consumer acceptance of fresh pork from entire male pigs. Meat Focus International, 4(9), 365–371. SAS. (1991). SAS/STAT User’s guide, Version 6.07. Cary, NC: SAS Institute Inc. Squires, E. J., Deng, W., & Wu, L. (1993). Colorimetric method for the determination of 16-androstene steroids in carcasses. In M. Bonneau (Ed.), Measurement and prevention of boar taint in entire male pigs (pp. 41–48). Paris: INRA Editions. Squires, E. J., Gullett, E. A., Fisher, K. R. S., & Partlow, G. D. (1991). Comparison of androst-16-ene determined by a colorimetric assay with boar taint estimated by a trained sensory panel. Journal of Animal Science, 69, 1092–1100. Squires, E. J., Lou, Y. (1995). Levels of boar taint in purebred entire male pigs in Canada. In Proceedings of a Meeting of the EAAP Working Group, Production and Utilization of Meat from Entire Male Pigs, Milton Keynes. Tuomola, M., Harpirio, R., Knuttila, P., Mikola, H., & Lo¨vgren, T. (1997). Time-resolved fluoroimmunoassay for the measurement of androstenone in porcine serum and fat samples. Journal of Agricultural and Food Chemistry, 45, 3529–3534. Vold, E. (1970). Fleishproduktionseigenschaften bei Ebern und Kastraten. IV. Organoleptische und gaschromatografische Untersuchungen wasserdampfflu¨chtiger Stoffe des Ru¨ckenspeckes von Ebern. Meldinger Norlandbruckhoegskole, 49, 1–25. Walstra, P. (1974). Fattening of young boars: quantification of positive and negative aspects. Livestock Production Science, 1, 187–196. Walstra, P., Claudi-Magnussen, C., Chevillon, P., von Seth, G., Diestre, A., Matthews, K. R., Homer, D. B., & Bonneau, M. (1999). An international study on the importance of androstenone and skatole for boar taint: levels of androstenone and skatole by country and season. Livestock Production Science, 62, 15–28. Weiler, U., Font i Furnols, M., Fischer, K., Kemmer, H., Oliver, M. A., Gispert, M., Dobrowolski, A., & Claus, R. (2000). Influence of differences in sensitivity of Spanish and German consumers to perceive androstenone on the acceptance of boar meat differing in skatole and androstenone concentrations. Meat Science, 54, 297– 304. Wysocki, C. J., & Beauchamp, G. K. (1984). Ability to smell androstenone is genetically determined. Proceedings of the National Academy of Science of the USA, 81, 4899–4902. Wysocki, C. J., Dorries, K. M., & Beauchamp, G. K. (1989). Ability to perceive androstenone can be acquired by ostensibly anosmic people. Proceedings of the National Academy of Science of the USA, 86, 7976–7978.