were included. As indicated earlier, aside from 

 one running ripe fish, which was caught in January 

 1957, no ripe skipjack were available during the 

 periods of sampling. Those in the advanced stage 

 were found, however, in November, and January 

 to April. This suggests that their major spawning 

 season hi this area is from about November 

 through April. 



The spawning season may be further defined by 

 determining the temporal distribution of non- 

 spawning; i.e., early developing skipjack. It was 

 noted earlier that some classified as early de- 

 veloping possessed ovaries that either (1) contained 

 only early developing ova, aside from a few 

 residual ova from a previous spawning, or (2) 

 contained ova that had reached the developing 

 stage but were apparently degenerating. These 

 observations suggest, that during certain periods 

 of the year the ovaries may revert to a dormant 

 or early developing stage. A few skipjack classi- 

 fied as early developing were found in November, 

 January, and February; however, the greatest 

 numbers occurred in April and May, the months 

 in which fish in the advanced stage were declining 

 in number or were totally absent. I assumed 

 that such a distribution of early developing and 

 advanced skipjack indicated a decrease in spawn- 

 ing activity during these 2 months. However, 

 the situation is probably more complex, since 

 skipjack in all stages of development were found 

 simultaneously, and fish with developing ovaries 

 were found in all months sampled. In all proba- 

 bility, scattered spawning occurs throughout the 

 year, but the peak activity is from November 

 through April (or, roughly, the Southern Hemi- 

 sphere summer). The results of a study of the 

 distribution of skipjack larvae in this area, de- 

 scribed by Nakamura and Matsumoto, 1 sub- 

 stantiate this conclusion. The larvae were caught 

 in greater abundance contemporaneously with the 

 greatest numbers of fish with advanced ovaries. 



Skipjack spawning seasons in other areas, as in 

 Hawaii, appear to be typically long. The season 

 in Hawaii extends from late February, March, or 

 April to the first part of September (Brock, 1954). 

 Schaefer and Orange (1956) indicated that skip- 

 jack spawn in the vicinity of the Re villa Gigedo 

 Islands from April to December, and that spawn- 



ing is more intense in summer and fall. Con- 

 cerning the fish in the Philippines, Wade (1951, p. 

 469) states, "... there are indications that the 

 period from September to April, inclusive, may be 

 the principal spawning period." The spawning 

 season in the southern waters of Japan, however, 

 is relatively short. Yao (1955) postulated that 

 skipjack spawn during June-August in the waters 

 south of Kyushu and the Ogasawara area. 



FREQUENCY OF SPAWNING 



The presence of two or more modes in ova-size 

 frequency distributions and that of residual ova 

 in ovaries containing mature ova have been used 

 to hypothesize multiple spawning. Among in- 

 vestigators studying tuna, June (1953) concluded 

 on the basis of such evidence that yellowfin in 

 Hawaiian waters spawn more than once during a 

 season. So did Brock (1954) for Hawaiian skip- 

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

 and Otsu and Uchida (1959) for the Pacific 

 albacore. 



The ova-diameter frequency distributions for 

 skipjack in northeastern French Oceania were 

 typically bimodal, or sometimes multimodal in 

 the more advanced ovaries (fig. 4). Furthermore, 

 residual ova were found in ovaries of 23 of the 72 

 skipjack that were judged to be in the advanced 

 stage. 



MacGregor (1957) discussed this problem of 

 multiple spawning in some detail in an analysis of 

 Pacific sardine fecundity. He disagreed with 

 several criteria that have been used in the past to 



MICROMETER EXITS 



1 Nakamura, E. L. and W. M. Matsumoto. MS. Distribution of larval 

 tuna in Marquesan waters. Bureau of Commercial Fisheries Biological 

 Laboratory, Honolulu, Hawaii. (Manuscript.) 



Figure 4. — Ova-diameter distribution for a developing 

 skipjack ovary. 



SKIPJACK SPAWNING IN MARQUESAS AND TUAMOTU AREAS 



483 



