SPAWNING OF ALBACORE IN THE SOUTH PACIFIC OCEAN 



155 



Stage in September 1958 is probably due to tlie 

 very small sample (7) collected during the month. 



It seems clear from figure 4 that the peak of 

 spawning activity in the central South Pacific 

 occurs at some time during the southern summer 

 months, between September and March. It 

 seems logical to assume that the peak spawning 

 period begins around December and extends into 

 January or February, when the percentage of the, 

 late developing stage gradually decreases. This 

 interpretation is based on the assumption that 

 this decrease reflects a gradual development of the 

 ovaries through the advanced, ripe, and spawning 

 stages. 



This southern summer spawning is in contrast 

 to the albacore of the North Pacific, which appear 

 to spawn during the northern summer months 

 (Uej'anagi, 1957; Otsu and Uchida, 1959b). 

 Areal variation in gonad development 



Smce ovaries in the advanced stage were so few, 

 and those in the ripe stage were entirelj* missing 

 from the samples, we have considered late devel- 

 oping ovaries to be indicative of spawning fish. It 

 has been postulated by Otsu and Uchida (1959b) 

 that when albacore are close to spawTiing, they be- 

 come unavailable to the hook and line fishery. 

 There is also a distinct possibility that the ovaries 

 develop very rapidly after passing the late devel- 

 oping stage (Ueyanagi, 1957) and that they are in 

 the advanced stage for but a ver}- brief interval 

 before spawiung takes place. This would then 

 make sampling of the later stages a matter of 

 sUght chance and may well account for the small 

 numbers of such stages found. Although the more 

 advanced stages are missing from our collection, we 

 believe that reasonable inferences as to spawning 

 can be drawn from the occui'rence of ovaries in the 

 late developing stage. 



In figures 5 and 6 and tables 3 and 4 are shown, 

 by area and by time, the percentage of occurrence 

 of late developing and advanced ovaries in the 

 samples. Since the data revealed no clear longi- 

 tudinal differences in the seasonal development of 

 the ovaries (fig. 5), only the latitudinal difTerences 

 are discussed here. The areas were arbitrarily 

 divided latitudinally as follows: 0°-10° S., 10° S.- 

 20° S., and 20° S.-30° S. 



Some differences are apparent. In the area 

 south of latitude 20° S. nearly all of the ovaries 

 sampled were in the early developing stage, indi- 



lecE-iec 



I to) I lui I im I tuJ I ^ n^' I 



(31 



I 



(01 I (0) I (0) I (0) I H I (0) I 10) 1 (0) I (01 I (01 , (01 I (01 I (01 I (01 J (01 



I tui ] tut I iu> j I 



JUL AUG SEP OCT NOV DEC I JAN FEB ^WR APR MAY JUN AIL AUG SEP I 



1957 



1958 



F"iGURE 5. — Seasonal an(i longitudinal variation in ovary 

 development, central South Pacific Ocean (0°-30° S.). 

 (The percentage of late developing and advanced ovaries 

 and the sample number are shown for each month.) 



eating that there is very little, if any, spawning in 

 that area. On the other hand, in the area between 

 10° S. and 20° S., 287 out of 387 pairs of ovaries 

 were found to be in the late developing stage, in- 

 dicating that most of the albacore spawning south 

 of the Equator takes place in these latitudes. Sam- 

 pling was inadequate between the Equator and 

 10° S., but this area may probably be considered 

 an extension of the middle area. 



620267 0-^2 



