were combined in Z-centimeter groups. The 

 Honolulu market sizes, obtained in pounds, were 

 combined into 10-pound weight groups for analysis. 

 Tables 3 through 10 (Appendix) contsiin zili of the 

 grouped measurements. 



Prior to locating modes, the data were 

 smoothed by a three-itenn moving average and 

 the following criteria were imposed: 



1. No modes shall be recognized in 

 distributions containing less than 100 fish. 



2. Each mode shall be separated from 

 every other mode by troughs dipping at least five 

 fish below modal peaks cifter smoothing. 



3. Each mode nnust be present in the data 

 for at least two adjacent months (Hawaiian data 

 only). 



4. At least two of the frequency classes 

 making up a modal group must contain no less 

 than 15 individuals each before smoothing, or 

 at least 10 individuals if the mode is present in 

 two adjacent months. 



5. The mode shall be the peak of the 

 smoothed distribution or the center of two or 

 more minor peaks which differ by less than five 

 fish in height. 



INTERPRETATIONS OF SIZE FREQUENCIES 



Yellowfin from the Hawaiian Islands and 

 those from various areas along the Equator differ 

 widely in average size. Hawaiian fish are the 

 largest, on the whole, and there is an increase 

 in size from west to east along the Equator 

 (Murphy and Shomura 1953). In view of this, 

 three areas were examined independently: 

 Hawaii, central equatorial region (155 W. to 

 180 longitude), and the Trust Territory or 

 western equatorial region (134 E. to 179 E. 

 longitude). 



Hawaiian Yellowfin 



Sexual Differences in the Catch 



The mesin size of the males is greater 

 than the mean size of the females in the dominant 

 size groups sampled by the longline. For the 

 fish sampled at the Honolulu market from April 

 through Septennber, 1951, the dominant female 

 modal group is centered at about 120 pounds and 

 that of the male at about 130 pounds (fig. 2). 

 Associated with the difference in mean size is an 

 inequality in the sex ratio, for example, the fish 

 examined fronn Novennber 1950 through October 



"1 I I I I I I I I I r 



I I I I I I I I 



^1 FEMALES N:I6I 

 I I MOLES N = 256 



AREA OF OVERLAP 



WEIGHT IN POUNDS 



Figure 2. --Weight-frequency distributions of 

 Hawaiian yellowfin, April-September 1951. 



1951 had an average ratio of 1:06 in favor of 

 the males (table 1). Monthly ratios do not 

 deviate sufficiently to appreciably affect the 

 positions of the modes in unsexed samples, 

 although there is some indication that the pre- 

 dominance of males is greater from October 

 through February. 



Table 1. --Monthly sex ratios of Hawaiian 

 yellowfin 



A change in sex ratio with increasing 

 size has been discussed by workers concerned 

 with the equatorial region (Nakamura 1949, 

 Shimada 1951a, Murphy and Shomura 1953, 

 and Otsu 1954). They indicate that near the 

 Equator yellowfin under 80 or 90 pounds have 



