Fishes of the Western North Atlantic 17 



of the upper lobe. The number of myotomes and dorsal fin-ray chambers is often larger 

 also in B. flatae, although there is no clear distinction between the two in these respects. 

 B. flatae differs from B. bermudae (which it resembles in the shape of the caudal fin) in 

 its more numerous myotomes (at least 59) and dorsal fin-ray chambers (at least 275). 



Additional Description. Anterior end of notochord in rostrum extends forward in a 

 straight linej rostrum marked ofF from dorsal fin by a shallow notch j caudal fin lanceolate 

 but asymmetrical, the origin of its lower lobe anterior to origin of its upper lobe by a dis- 

 tance V3 to % as great as length of upper lobe, about midway between its tip and atrioporej 

 anus considerably posterior to midpoint of lower lobe of caudal; distance from tip of 

 caudal to anus V3 as long as from anus to atriopore; dorsal fin Y^ to ^,4 as high as dis- 

 tance from its crest to the margins of the metapleura; highest dorsal fin-ray chambers 

 3 to 6 times as high as long; dorsal fin-ray chambers 280 to 330; precaudal fin-ray cham- 

 bers 19 to 33; myotomes 37 to 40 anterior to atriopore, 13 to 17 between atriopore and 

 anus, and 6 to 9 posterior to the anus, the recorded totals from 58 to 65; gonads 26 to 31. 



Color. Presumably as in B. caribaeum (p. 14), but no specific inform.ation is avail- 

 able. 



Size. Recorded specimens have ranged from 28 to ^d mm. in length. 



D evelof mental Stages. Presumably as in other members of the genus. 



Habits. Nothing is known of the habits of B. flatae to differentiate it from its rela- 

 tives. 



Range. Specimens positively identified as B. flatae are known up to the present time 

 only from oflF the mouth of the Rio de La Plata and from southern Brazil (San Sebastiao I., 

 the vicinity of Rio de Janeiro). But it seems probable that the Lancelets that have been 

 recorded as B. caribaeum from Santos, from Santa Catharina at the mouth of the Amazon, 

 from the Rio de La Plata and from Buenos Aires, were B. flatae. 



Synonyms and References: 



Branchiostoma flatae Hubbs, Occ. Pap. Mus. Zool. Univ. Mich., 102, 1922: 10 (descr., off Rio de La Plata); 

 Jordan, Evermann and Clark, Rep. U.S. Coram. Fish. (1928), 2, 1930: 7, footnote. 



Probable References: 



Amfhioxus miiUeri Moreau, Bull. Acad. Roy. Belg., (2) 59, 1 875 : 3 1 2, i pL, 1 2 figs. (micr. anat. of notochord, 

 Rio de Janeiro, Brazil). ^- 



Branchiostoma caribaeum^^ Gunther, Rep. Zool. Coll. "Alert," Brit. Mus., 1884: 32 (in part, specimen from 

 Botafogo, near Rio de Janeiro, Brazil) ; Jordan and Evermann, Bull. U.S. nnt. Mu?., 4.- (l), 1S96: 4 (in 

 part, Brazil incl. in range) ; Bull. U.S. nat. Mus., 47 (4), 1900: pi. I, fig. I (probably B. flatae because of 

 shape of caudal) ; Evermann and Marsh, Bull. U.S. Bur. Fish., 20 (l), 1902: 59 (ill., after Jordan and 

 Evermann 1900, as above, but account is of B. caribaeum, N. Carolina) ; Barbour, Bull. Mus. comp. Zool. 

 Harv., 46, 1905: no (in part, because Rio de La Plata incl. in distrib.) ; Ribeiro, Arch. Mus. nac. Rio 

 de J., 14, 1907: 146, 199; Fauna brazil., Peixes, 2(1), Fasc. I, 1923: 4 (in part, because Brazil incl. in 



32. The name miilleri would have priority over flatae if the specimens in question actually were identical with 

 the latter, as the locality suggests. But Moreau gave no account of their external characters, nor is it likely that 

 the sections on which his studies of microscopic anatomy were based are still in existence. 



33. Sometimes spelled "carribaeum." 



