CRAWFORD and KRETSCH: FISHY FLAVOR IN TURKEY 



convened and the air and nitrogen packs w^ere 

 randomly offered from day to day. Duncan's mul- 

 tiple range test (a = 0.05) was used to compare 

 the adjusted mean of the taste panel scores. The 

 scoring was: 1 = no fishy flavor, 5 = very fishy 

 flavor 



RESULTS AND DISCUSSION 



The results reported in Table 2 are to be inter- 

 preted with some caution because of the low level 

 of fishiness in the meat from turkeys fed 2% fish 

 oil for only 2 wk. Therefore, only trends are 

 indicated for the results in Table 2 where statisti- 

 cal significance could not be achieved. 



Tables 1 and 2 report Duncan's multiple range 

 test of the mean taste panel scores of breast meat 

 cooked in air or nitrogen from turkeys fed various 

 diets containing tuna oil and/or beef fat with and 

 without dietary a-tocopherol acetate or a-tocoph- 

 erol injection. All meats that contained 

 «-tocopherol gave taste panel scores that were 

 comparable to the scores for the control for all 

 methods of cooking. When breast meat is cooked 

 under nitrogen with a slight vacuum no appreci- 

 able difference in flavor is caused by any of the 

 diets. However, the breast meat from turkeys fed 

 diets containing 2% tuna oil (treatments 5C and 

 2B) did have slightly higher scores, although not 

 statistically different from the control (treat- 

 ments IC or IB, 47c beef fat). The breast meat 

 cooked in air from turkeys fed diets containing 

 2% tuna oil (treatments 5C and 2B) showed more 

 off flavor than those cooked in nitrogen when 

 each is compared to its control (treatments IC 

 or IB, 4% beef fat). Furthermore, the order and 

 rank of the scores for the air-cooked meat were 

 very similar to those of breast meat from whole 

 roasted turkeys previously reported by Crawford 

 et al. (1975). These turkeys were randomly 

 selected from the same groups of turkeys used in 

 this experiment and were roasted at 177°C to 

 center breast temperature of about 70°C. 



From the results of this experiment, it may be 

 concluded that cooking breast meat of potentially 

 fishy flavored turkeys under nitrogen is nearly as 

 effective in preventing fishy flavor development 

 as feeding a-tocopherol acetate (in the diets with 

 the tuna oil) and roasting in the normal manner 

 This implies that fishy flavor develops postmor- 

 tem and requires air for its development. Alter- 

 nately, it could be concluded that cooking under 

 nitrogen per se had practically no effect in pre- 



TABLE 1. — Duncan's multiple range test of mean' taste panel 

 scores^ for breast meat cooked in air or nitrogen from turkeys 

 fed various diets containing tuna oil and/or beef with and 

 without a-tocopherol acetate. 



'Mean taste panel scores connected by a common line are not signifi- 

 cantly different at tfie 0.05 probability level. 



^Taste panel scoring: 1 = no fishy flavor, 5 = very fishy flavor. Abbrevia- 

 tions: TO = tuna oil; E = mg d/ a-tocopherol acetate per kilogram of diet; 

 BF = beef fat; BF, = beef fat substituted for 2% TO +2% BF. 



^AII groups (except group IC, the control which was maintained on diet 

 with 4°o BF for all 1 6 wk) were fed a basal diet with 2% TO plus 2% BF from 

 8 to 14 wk of age and from 14 to 16 wk of age, they were fed a basal diet 

 with: group IC =4°b BF, group 2C = change to 4*^0 BF, group 3C = change 

 to 4°o BF -r 100 mg/kg a-tocopherol acetate, group 4C = change to 4''o BF 

 * 200 mg a-tocopherol acetate, group 5C = kept on 2% TO + 2% BF. 



■■These results for the breast meat of normally roasted whole turkeys 

 were previously reported by Crawford et al. (1975). 



Table 2. — Duncan's multiple range test of mean^ taste panel 

 scores^ for breast meat cooked in air or nitrogen from turkeys 

 fed various diets containing tuna oil and/or beef fat with and 

 without a-tocopherol acetate supplement or injection. 



'Mean taste panel scores connected by a common line are not signifi- 

 cantly different at the 0.05 probability level. 



^Taste panel scoring: 1 ^ no fishy flavor, 5 = very fishy flavor. Abbrevia- 

 tions: TO = tuna oil: E = milligrams d/ a-tocopherol acetate per kilogram of 

 diet; In = inject <i-tocopherol; BF = beef fat. 



^AII groups were fed a basal diet *4% BF to 14 wk of age and from 14 to 

 16 wk of age, they were fed a basal diet with: group IB = 4% BF, group 2B 

 = 2°o BF + 2°o TO, group 3B =2% BF + 2% TO (+ inject 170 mg of 

 a-tocopherol into thigh 72, 48. and 24 h before sacrifice), group 4B = 2% 

 BF + 2% TO + 100 mg a-tocopherol acetate per kilogram, group 5B = 2% 

 BF + 2% TO + 500 mg a-tocopherol acetate per kilogram. 



"These results tor the breast meat of normally roasted whole turkeys 

 were previously reported by Crawford et al. (1975). 



venting this development but that fishy flavor 

 had already developed in vivo and the heat of 

 cooking at 116°C for 80 min destroyed the compo- 

 nents which cause this flavor. Some observations 

 and recent work (Crawford unpubl. data) tend to 

 support the first conclusion. 



We have observed that the odor of fresh raw 

 turkey was insipid regardless of the type of diet- 

 ary oil. However, after comminuting and storing 

 in the refrigerator overnight, the flesh from tur- 

 keys fed tuna oil smelled fishy while the odor of 

 beef fat-fed turkeys remained rather insipid. 



91 



