hypothesize multiple spawning; he objected to the 

 criterion of the presence of two or more modes in 

 ova-size frequency distributions. He pointed out 

 that the presence of two or more such modes in 

 developing ovaries does not necessarily mean that 

 all the groups mature and that multiple spawning 

 will occur. 



Support for MacGregor's contention may be 

 found in Yabe's (1954) work on skipjack in Jap- 

 anese waters. He noted that skipjack smaller 

 than 47 cm. had relatively large ovaries from June 

 to August, but that the relative size of the ovaries 

 decreased after September. He suggested that 

 the ovaries shrank without spawning having 

 occurred. 



As noted earlier, my data also show some 

 ovaries that were classified as developing but 

 contained what appeared to be degenerating ova; 

 this finding, one may argue, is in agreement with 

 MacGregor's thesis. It also may be argued, 

 however, that these data merely indicate that there 

 may be a cessation of spawning during part of the 

 year and do not necessarily indicate that none of 

 the intermediate-sized ova will be spawned. 

 Furthermore, skipjack as small as 43 cm. possessed 

 advanced ovaries which also had residual ova from 

 a previous spawning. If their growth rate in this 

 area is similar to that of the species in Hawaii 

 (Brock, 1954), which is not entirely unreasonable 

 since skipjack in both areas seem to reach adult- 

 hood at about the same length, then these small 

 fish must have spawned no more than 3 or 4 

 months previously. Earlier than that they would 

 not have been large enough to be sexually mature. 

 These observations, although not conclusive, in- 

 dicate the possibility of multiple spawning within 

 a season for individual skipjack. 



FECUNDITY 



Four skipjack with fork lengths of 43 to 74 cm. 

 and with advanced ovaries were selected for 

 fecundity determinations. Small sections were 

 obtained from each of the ovaries and weighed to 

 the nearest thousandth of a gram. The number of 

 ova in the most advanced group in the sample was 

 determined and multiplied by the ratio of ovary 

 weight to sample weight to get an estimate of the 

 total number of mature ova in the ovary. The 

 results are presented in table 1, which also includes 

 results obtained by Yabe (1954) for skipjack in 

 Japanese waters. Estimates of the number of 



ova extruded at one spawning ranged from 0.1 

 million to 2 million, with an indication that the 

 number spawned is related to the size of the skip- 

 jack, the larger fish spawning more ova. Joseph 

 (1963) made fecundity determinations for 42 

 skipjack taken from the eastern Pacific Ocean. 

 His estimates ranged from about 0.2 million to 

 1.5 million ova per spawning for skipjack with 

 total lengths from 61.4 to 71.5 cm. Within the 

 limits of his data, the size of the skipjack and the 

 number of ova per spawning seemed to be related. 

 Among some of the other tuna, a relation between 

 the number of ova spawned and size of fish was 

 found for the bigeye in the Pacific (Yuen, 1955) 

 and Hawaiian yellowfin (June, 1953). 



Table 1. — Results of skipjack fecundity determinations 



i Data from Yabe (1954). 



GONAD INDEX 



Several investigators have discussed the possi- 

 bility of using the relative ovary weight (ovary 

 weight X10 3 /fish weight) as a measure of maturity 

 of tuna; e.g., June (1953), Yuen (1955), and Otsu 

 and Uchida (1959). Schaefer and Orange (1956) 

 also used the relative ovary weight to measure the 

 maturity of skipjack, the only difference being 

 that they used the cube of fish length instead of 

 fish weight and a factor of 10 s instead of 10 3 . 

 They called this relation the "gonad index." 



which is defined as t7./. = -^X10 8 , where "G.I." 



is the gonad index, "w" the weight of the ovaries 

 in grams, and "L" the fish length in millimeters. 

 Schaefer and Orange plotted the gonad index 

 against the 95th centile of the total ova-size 

 frequency distribution and found a linear relation, 

 the gonad index increasing with ova size, at least 

 in their Area II of the eastern Pacific. 



I calculated the gonad index for all the skipjack 

 in my collection. Because only a sample of the 

 mosl developed group of ova was measured, their 



484 



U.S. FISH AND WILDLIFE SERVICE 



