BERYCOID FISH FAMILY POLYMIXIIDAE LACHNER 195 



Geographic distribution: Madeira Islands, Canary Islands, St. 

 Helena, Cuba. 



Discussion: This species is apparently considerably rarer in the 

 West Indian waters than lowei. Of 154 specimens of both species 

 captured and preserved from Cuban waters by the Atlantis Expedi- 

 tion, only four, or 2.6 percent, proved to be nobilis. These four 

 specimens were taken in water 295 fathoms in depth. Two of the 

 specimens were taken in a haul with lowei. 



Although Poey (1860) probably was not aware of the description of 

 lowei (Giinther, 1859) when he described venustus, he failed to recog- 

 nize specific differences, since part of his type material (MCZ 21812, 

 2 specimens) represents specimens of nobilis and lowei. 



Giinther (1859) distinguished lowei from nobilis on the basis of six 

 less rays in the dorsal fin, but when he had what he presumed to be 

 intermediate material from the Pacific and Indian Oceans he later 

 (1887, p. 34) lumped all into a single species, Goode and Bean (1895) 

 and others followed this procedure.^ 



Smith (1949) referred to his Natal specimens as nobilis. His figure 

 283 shows 29 soft dorsal rays, which is out of the range for this spe- 

 cies. Dorsal fin rays ranging from V, 27 to V, 38 were listed, but this 

 range was probably summarized from the literature. No specimens 

 from the Indian Ocean were available for my study. It appears that 

 specimens from this area may represent ja/)o/iica. 



The higher anal fin ray count in specimens from Madeira compared 

 to those from Cuba may show a real divergence in these populations 

 when larger collections become available. 



Polymixia lowei Giinther 



Plate 1 



Polymixia lowei Giinther, 1859, p. 17 (type locality, Cuba) ; 1887, p. 34 (in part. — 



Rivero, 1936, p. 57. 

 Dinemus venustus Poey, 1860, p. 161 (type locality, Cuba) (in part). 

 Polymixia nohilis Goode and Bean, 1895, p. 243 (in part). 

 Polymixia nobilis virginica Nichols and Firth, 1936, p. 2 (type locality. Cape 



Henry, Va.). 



A total of 235 specimens ranging in standard length from 60 to 

 198 mm. were examined from the following localities: Cuba (152 

 specimens), BM 1852.9.13.216 (holotype), MCZ 21812 (paratype of 

 Dinemus venustus Poey, 134 mm. standard length), and 150 MCZ 



' While this paper was ia press Maul published an account (Bol. Mus. Municipal do Funchal, Madeira, 

 No. 7, art. 17, pp. 9-11, 1954) in which he recognized two species on the basis of a higher dorsal count in 

 nobilis, 40 to 42 (spines plus rays), compared to 35 in lowei, and that the body of nobilis was "rather deep." 

 The range of variation of the dorsal rays is considerably greater than Maul indicates and the amount of 

 separation much closer. I find the depth of the body to vary considerably in respect to size and to some 

 extent sex, the females being somewhat deeper. I did not find any notable differentiation in body depth 

 when comparing specimens of equal size of these species. However, nobilis appears to attain a much larger 

 size and, consequently, greater depth. 



