the winters. Two other returns, which 

 may be more pertment, were from 

 releases ofT the northwestern Baha- 

 mas in May 1 963 and June 1 969 and 

 recaptures off Recife, Brazil, in 

 March 1965 and off Buenos Aires, 

 Argentina, in February 1973, respec- 

 tively (Figure 75). These two migra- 

 tions were definitely to wintering 

 grounds (in the northwest Atlantic 

 frame of reference) and one even ex- 

 tends the probable wintering area far 

 south of what was inferred from dis- 

 tributional data. The information 

 available leads to the assumption that 

 each of these fish had migrated south- 

 ward from the northern feeding 

 grounds m the fall preceding the win- 

 ter in which it was recaptured. Since 

 they had been at liberty for periods 

 of 2 1 and 45 months respectively, 

 however, they might possibly have 

 entered a different migratory pattern 

 of which we have no knowledge. In 

 any case, these returns do not pro- 

 vide information as to the routes 

 which they followed in their final 

 migrations. 



The recapture in December 1 973 

 at the "tail" of the Grand Banks of a 

 giant which had been released in 

 northeastern Newfoundland coastal 

 waters in August 1972 might indi- 

 cate a starting point for southward 

 migrations, but, since this fish had 

 been at large for about 16 months, 

 this was an "indirect", or presumed, 

 movement from coastal waters 



Japanese longlinc catch rates in 

 November 1965 (Le Gall 1974) were 

 relatively high (one to three fish per 

 1,000 hooks) in a narrow area be- 

 tween longitudes 59°W and 60°W 

 and extending from 44°N to 32°N 

 Although data on the sizes of the fish 

 have not been published, this area 

 might well represent the first part of 

 a southward migration of large blue- 

 fin from Canadian coastal waters to 

 their wintering areas. 



Consideration of the North At- 

 lantic circulatory system (Sverdrup 

 et al 1942) and the tendency of blue- 

 fin tuna to follow currents, however, 

 leads us to suggest that many of the 

 giants may follow a hypothetical mi- 

 gratory pattern following the North 

 Atlantic gyre around the Sargasso 

 Sea, first eastward, roughly along lati- 

 tude 40°N, then southward and fi- 



nally westward over a rather broad 

 front The proposed migrations Irom 

 the wintering areas through tlie spring 

 spawning areas to the summer feed- 

 ing grounds generally follow the re- 

 mamder of this gyre. Sella (1929a, 

 1929b) was most likely the first to 

 discuss the probable influence of the 

 Gulf Stream system on the distnbu- 

 tion of bluefin tuna in the Atlantic 

 Ocean. We consider the southward 

 migration of large bluefin tuna as 

 one of the most challenging prob- 

 lems which remain to be solved 



The wintering grounds of the 

 large bluefin which feed in north- 

 western Atlantic coastal waters in the 

 warm season ha\e not yet been 

 clearly defined The giant tuna are 

 less concentrated, and presumably 

 less active, during the cold season 

 than in their spawning and feeding 

 periods. They are dispersed over great 

 areas, and are almost always too far 

 below the surface to be seen, or to be 

 caught by surface gears Moreover, 

 the opportunities for bluefin from dif- 

 ferent spawning and feeding areas to 

 mix with each other may be at their 

 maximum during this period Except 

 for a very few tag returns, the major 

 sources of information on the distri- 

 bution and migrations of giant blue- 

 fin during the (northern) cold season 

 are the records of oceanic longline 

 fishenes. As already stated (Section 

 IVC3, and in this section), extensive 

 catch and effort data for the Japanese 

 longline fishery are available, but 

 there is little published information 

 on the sizes of the fish which it cap- 

 tures This deficiency increases the 

 uncertainties inherent in deducing mi- 

 gratory patterns from these data In 

 addition, the temporal and areal cov- 

 erage of the vast potential range of 

 the giant bluefin is incomplete, even 

 if the data from all years combined 

 arc considered collectively for each 

 month 



The contours of relative abun- 

 dance, as indicated by longline catch 

 rates (Figure 32), show two areas of 

 maximum abundance in the first quar- 

 ter of the year (January-March). One 

 of these is centered on the Greater 

 Antilles and includes the southeast- 

 ern Gulf of Mexico, the northwest- 

 ern Caribbean and Atlantic waters 

 north of central and eastern Cuba and 



Ilispaniola The other centers around 

 the "bulge" (the easternmost part) of 

 Brazil It extends northeastward 

 nearly halfway to the nearest parts of 

 Africa, northward to lafitude 15°N, 

 eastward to longitude 40°W and 

 southward, near the Brazilian coast, 

 to latitude 10°S Tag returns (Figure 

 75) show that giant bluefin do, at 

 least occasionally, migrate from the 

 western North AUantic into the South 

 Atlantic, 



The distributions of monthly 

 catch rales for individual years (Le 

 Gall 1974) and of average monthly 

 catch rates for groups of years 

 (Shiohama et al 1965, Shingu et al. 

 1975) permit more detailed estimates 

 of seasonal distributions and migra- 

 tions. 



Hxamination of the catch rates 

 in the Antilles area on this basis led 

 us to beliexe that the winter distribu- 

 tion centered on these islands actu- 

 ally covered all of the Gulf of Mexico 

 and the Caribbean Sea, and the adja- 

 cent Atlantic waters up to latitude 

 25°N and eastward to longitude 60°E. 

 Data for United States exploratory 

 and commercial fishing in this area 

 (Wathne 1959, H R Bullis, Jr., per- 

 sonal communications, personal ob- 

 servations of the senior author) indi- 

 cate that nearly all the bluefin taken 

 in these waters have been giants. 



The Japanese longline coverage 

 in this area during the cold season 

 has been light 1 he data indicate, 

 however, that bluefin were scarce 

 during most of the fall, but become 

 more abundant in December, Janu- 

 ary and February Coverage in March 

 was minimal, but bluefin were taken 

 in nearly half of the areas fished In 

 April, catch rates rose drastically in 

 the Atlantic waters of the area, evi- 

 dently because of the arrival of 

 spawning fish 



Catch rates of more than one fish 

 per 1 ,000 hooks were rare until April. 

 They occurred in only four instances, 

 each of which was in a different lo- 

 cality 



Wathne (1959) reviewed United 

 States exploratory and commercial 

 longline fishing in the Gulf of Mexico 

 and adjacent waters The commer- 

 cial fishery was directed toward yel- 

 kmfin tuna, T. alhacares. Wathne 

 noted that giant bluefin occurred in 



122 



