194 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



Table 18. — Summary of covariance analysis of length-weight data on yellowtail 



I. COMPARISON OF SEXES BY QUARTER 



1 Expected less than once in 20 times by chance. 

 1 Expected less than once in 100 times by chance. 



since the length-weight data were from samples 

 taken at random and include representative num- 

 bers of males and females we regard the total 

 values for each quarter as representative and have 

 not segregated the data by sex. 4 



The estimated weight at each length occurring 

 in the landings has been obtained from the com- 

 bined data for males and females in the prepara- 

 tion of table 19. This will be used in the next sec- 

 tion to determine the number of fish landed. 

 Figure 12 indicates the average length-weight 

 relation. Meanwhile, it is interesting to note the 

 range in weight of the yellowtail in the landings. 

 When the central 98 percent was selected from the 

 data on average length composition (table 10) to 

 avoid the few very small or very large specimens, 

 the "lower limit" of size was 27.3 cm. (0.38 lb.) 

 and the "upper limit" was 46.7 cm. (2.17 lb.). 

 The average length was 35.87 cm. (0.93 lb.). The 

 smaller value reflects selection by the fishermen 

 as influenced by buyers interested in filleting the 

 fish. The buyers estimate that an average of 40 

 percent of the weight is recoverable as fillets. If 



"n — i — i — I — i — i — i — i — i — i — i — r 



* We have estimated the discrepancy arising from varying pro- 

 portions of the sexes by calculating the average weight of the 

 yellowtail In the second quarter (when the greatest difference 

 between the sexes occurs) for each sex by using the aggregate 

 formula, and we found that the maximum difference between the 

 sexes in average weight was 3.7 percent. Because of the small 

 difference and the poor representation of males In the length- 

 weight data for the second and third quarters, it appears unneces- 

 sary to compute the averages separately by sex. 



20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 



LENGTH- CENTIMETERS 



Figure 12. — Average length-weight relation of yellowtail 

 landed from the southern New England stock during 

 1943. The dotted lines are plus and minus twice the 

 standard error of estimate and enclose about 95 per- 

 cent of the observations. 



this was true of the small fish, 5 the lower limit of 

 desirable fillet weight would be about 0.076 pound, 

 or just over 1 ounce. The average-size fillet 

 weighed 0.186 pound, or about 3 ounces, and the 

 maximum 0.434 pound, or about 7 ounces. 



6 Small fish were disliked by the filleters because of higher cost 

 and lower yield, but data on fillet recovery were not obtained. 



