a 1:1 sex ratio, while larger fish are 

 predominantly male, as they are in Hawaii. This 

 change in the sex ratio with increasing size may 

 also occur in Hawaii but it will have little effect 

 on the present analysis because the majority of 

 the Hawaiian specimens are considerably larger 

 than 80-90 pounds. 



Sex-linked differences in abundance and 

 growth have been investigated in other species 

 of tuna. Brock (1943) states that there was no 

 significant difference in the length of male and 

 female albacore, Germo alalunga (Gmelin), from 

 Oregon. He found the sex ratio to be 1:1 in fish 

 above 67 cm. Below 67 cm. males apparently 

 predominate, but he points out that "some of the 

 smaller females were wrongly called males." 

 Brock (1954) reports a 1:1 sex ratio for Hawaiian 

 skipjack, Katsuwonus pelamis (Linne), during 

 the months of March to August, but during Sep- 

 tember to December there were significantly 

 more males. 



Interyear Size Variation 



Evidence of interyear size variation was 

 sought in the largest samples for the years 1948- 

 1952, i.e., June-September. The grouped 

 weights for the four months of each year, plotted 

 as deviations from the 5-year average (fig. 3), 

 show that there are between-year differences in 

 the abundance and position of certain size groups. 

 However, the variations in abundance are snnall 

 and there seem to be no trends in the positions 

 of modcil groups that would bias a growth curve 

 based on the pooled data for all five years. 



Intrayear Size Variation 





 -2 



2 -" 



o 



> 



i 



UJ 



Z3 



UJ <^ 



cr 



u. 



I- -1 



z. 



UJ 



1 " 



UJ 



TTTTTT 



1948 

 N' 1815 



1 I I I! I I I I I I I 



1950 

 N = 2206 



1952 

 N=2450 



"TT 



\^^m. 



I I I I I I I I 



I I 



I I I 



I I I I I I I I 



I I I 



20 40 60 80 100 120 140 160 180 200 

 I I I I I I I I I I 



29 49 69 89 109 129 149 169 189 209 



WEIGHT IN POUNDS 



220 240 260 280 



I I I I 



229 249 269 289 



Kishinouye (1923), in speaking generally 

 of scombroid fishes fronn Japan, stated, "large 

 and old are caught at the beginning of the fishing 

 season, while at the end of the season only young 

 and small ones are found. " In the Hawaiian fish- 

 ery there is an indication that the smaller yellow- 

 fin (100-130 pounds) enter the fishery early in 

 the season (June and July) and larger fish (>130 

 pounds) appear a short while later (August and 

 September). This can be seen in figure 4, which 

 shows the percent deviations of individual months 

 from the 4 months' mean (June-September) 

 averaged over 5 years (1948-1952). This differ- 

 ential recruitment is not believed to eiffect the 

 interpretation of growth because the absolute 

 deviations are not great. 



Moore (1951) remarks that the catches 

 of the last 3 months of the year "were not large 

 in comparison with the catches of the sumnner 

 months, " causing erratic nnodes to appear. 



Figure 3. --Percent deviations from a 5-year 

 average of Hawaiian yellowfin weight- 

 frequency distributions for individual years 

 (pooled samples for June-September). 



The catches of the early months of the year are 

 also very snnall, and they too may introduce 

 this complication. This is illustrated by the 

 typical monthly weight-frequency distributions 

 for 1949 (fig. 5). 



Moore's use of percentage deviation to 

 follow the progression of modes through the 

 year may have biased the interpretation of the 

 modes in giving undue emphasis to the very 

 small sannples obtained early and late in the 

 year; this may be especially true for the 

 livebait-caught fish (fig. 5). This does not 

 necessarily mean that small, indefinite nnodal 

 groups which may appear for a few nrionths are 

 due to chance variation. They may or may 



