On the breeding areas of the Swordfish (Xiphiax) 445 



this maximum, while there is a decline in May through June. Thus, even if Xiphias 

 breeds throughout the year in this tropical area, analysis of the data in this way sug- 

 gests that the maximum is from February to April. Arata (1954, pp. 234 239), on 

 the other hand, from his own records and those of others which note the appearance 

 of the fry or of mature specimens, concludes that the peak of the spawning season 

 off Florida and Georgia is from April through September. In this connection, it 

 should be noted that our data too indicate the presence of a large number in August 

 (Table I). However, this figure is due to the fact that the Dana, on her passage through 

 the southern part of the Sargasso Sea in 1928, chanced on six small specimens in 

 postlarval stages and two a little larger (33-0 and 46-3 mm in length) at the three 

 stations in that area. Consequently, it seems possible that this particular year was 

 an unusually favourable one for the reproduction of this species. 



It should be stressed here that the area where the fry discussed in this paper were 

 found must be one of the species' chief spawning areas, because equally intensive 

 fishing in adjacent areas to the north and east was carried out from Danish vessels 



Table I 

 The monthly distribution of catches and number of specimens in postlarval stages less 

 than 20 mm in length per 100 hours of fishing 



Month I II III IV V VI VII VIII IX X XI XII Total 



No. hrs. 



fishing 45 156 274 365 164 214 44 20 4 113 61 216 1690 



No. post- 

 larvae 



<20mm 1 6 12 14 I 2 6 1 I 5 49 



No. per 

 100 hrs. 

 fishing 2-3 4-6 4-4 40 0-5 4-5 300 0-9 15 2? '0 



using the same type of gear without taking a single postlarval specimen. Although 

 the area southeast of the main area was fished less intensively, it was sampled rather 

 adequately without finding a single specimen. LUtken (1880) mentions some catches 

 (undated) of specimens in young stages, some as small as 2 cm in length, taken with 

 drag nets at the surface from sailing ships in the centre of the tropical-subtropical 

 Atlantic, Hence, we know that the species must also breed there as well. Undoubtedly, 

 spawning is much more scattered there than in the area delimited by the Danish 

 surveys. Thus, when the Dana crossed from Capetown to the Canary Islands via 

 St. Helena-Freetown in February-March 1930, no Xiphias larvae were taken there 

 despite intensive fishing. In any event, this does not seem to have been the n^^ht 

 season to find them in this part of the eastern tropical-subtropical Atlantic. 



Mediterranean. In the Mediterranean, where environmental conditions are suitable 

 for the development of a rather sizeable stock of Xiphias. only a single \tphias larva 

 has been found to date in the Danish collections. It was taken in the Balearic Sea 

 off Algiers in late September. At Messina (Sella, 1911 ; Sanzo, 1922; 1930), this 

 species breeds in the summer (June-August) at a time when the surface temperatures 



