FISHERY BULLETIN: VOL. 79, NO. 3 



Skipjack 



Yellowfin 



FORK LENGTH(cm) 



Figure 5. — Fork length-frequency distributions for skipjack 

 (top) and yellowfin tunas (bottom) sampled during the study. 

 Individual samples have been pooled for the four study areas in 

 the Papua New Guinea region. 



correspondence between 6-mo intervals and calcu- 

 lated peaks is not precise, considering the underly- 

 ing assumption that all modal groups grew at 

 identical rates, the observed correspondence is 

 fairly good. The results indicate that the skipjack 

 tuna stocks exploited by the Papua New Guinea 

 fishery may exhibit two peaks in spawning activ- 

 ity, 6 mo apart. 



Naganuma (1979), using gonad indices of skip- 

 jack tuna caught in the western Pacific, demon- 

 strated the existence of two spawning groups: one 

 spawning in southern waters in the southern 

 summer (October-March) and the other in north- 

 ern waters in the northern summer. Data for lar- 

 val abundance in Papua New Guinea waters 

 (Nishikawa et al. 1978) indicate almost identical 

 peaks in spawning periodicity, although much 

 continuous spawning in equatorial waters is also 

 indicated. Spawning periodicity as determined 

 from gonad indices and larval distribution there- 



FIGURE 6. — Distribution of month of birth of skipjack tuna from 

 the Papua New Guinea fishery calculated by the von Bertalanffy 

 growth function, using an independent estimate of growth for all 

 modal fork length data. Distribution is shown in one-tenths of 1 

 yr. Dots indicate 6-mo intervals. 



fore corresponds to that determined from the re- 

 sults of the present study. 



However, as is clear from the skipjack tuna 

 modal data (Figure 3) and abundance (as inferred 

 from skipjack tuna CPUE, Table 1), this possible 

 6-mo spawning periodicity did not result in 

 semestral recruitment to the fishery in 1977-79. 

 Lewis (1980a), however, reported that there were 

 two groups of skipjack tuna present in the Bis- 

 marck Sea in 1972, apparently resulting from two 

 periods of recruitment 6 mo or more apart. As 1972 

 was a year in which skipjack tuna abundance was 

 exceedingly low, it might be expected that the con- 

 sequences of possible large-scale periodic recruit- 

 ment to the fishery would be more obvious then 

 than during periods of relatively high abundance, 

 as during the present study. 



Yellowfin tuna show restricted spawning 

 periods and semestral recruitment in many 

 fisheries (Hennemuth 1961; Davidoff 1963; Ma- 

 tsumoto 1966; Le Guen et al. 1969; Richards 1969; 

 Le Guen and Sakagawa 1973). That this is un- 

 likely to have occurred in equatorial western 

 Pacific waters during the period of the study is 

 clear from Figure 4; the fairly continuous produc- 

 tion offish implied by the large number of modal 

 groups passing through the fishery is unlikely to 

 be the result of one or two short spawning periods. 

 Although differential growth between elements 

 resulting from a protracted spawning period 

 might result in the observed spread in recruit- 

 ment, the youth of the fish, as inferred from rate of 

 growth estimates (Le Guen and Sakagawa 1973; 

 present paper), would necessitate either early 

 separation into groups offish exhibiting different 

 rates of growth or protracted spawning periods 

 (perhaps continuous spawning activity). 



Most identifiable groups of yellowfin tuna dis- 

 appeared from the pole-and-line catch at between 

 62 and 71 cm FL (Figure 4). Kikawa and 

 Warashina (1972) pointed out that the Japanese 



524 



