982 



MR. C. TATE REGAK OX 



Mem. Mas. Comp. Zool. xix. 1895, p. 82 ; Regan, Biol. Centr.- 

 Amer., Pisces, p. 93 (1907). 



Paragambasia Meek, Publ. Columbian Mus., Zool. v. 1904, p. 133. 



This genus, as now restricted, includes species in which the 

 male intromittent organ is moderately long, nearly -J- of the length 

 of the fish (without caudal), and is formed on a definite plan 

 which is subject to considerable modification in the different 

 species. The distal segments of the first prolonged ray are pro- 

 duced into processes directed more or less towards its apex ; the 

 anterior branch of the second prolonged ray is slender distally 

 and at some distance from the *end has an antrorse projection 

 which may be termed the " elbow " ; the posterior branch of this 

 ray ends in a I'etroi'se pointed hook or barb, and the segments 

 immediately proximal to the elbow of the anterior branch are 

 produced backwards into senate ; the third prolonged ray ends in 

 a hook more or less similar to that of the second. 



Of the 17 species here recognized I have seen males of 9, the 

 distal part of the anal fin of each of them is figured. Of Gani- 

 busia holbrookii (text-Q^g. 169 A) I have examined two adult males, 

 of 6^^. oligosticta (text-fig. 169 B)four, of G. nicaraguensis [text-fig. 

 168 A) two, of G. wrayi (text-fig. 168 B) four, of G. gracilior 

 (text-fig. 168 0) three, of G. dominicensis (text-fig. 169 C) two, 

 of G. nigropunctata (text-tig. 168 F) one, of G. punctata (text- 

 fig. 168 D) three, and of G. senilis (text-fig. 168 E) one. These 

 suffice to prove that the differences shown in the figures are really 

 specific; the number, form, and size of the modified segments 

 of the first prolonged ray, the number of segments distal to 

 the elbow, the form of the hook, and the number and length 

 of the serrse of the second prolonged ray, and the form of the 

 hook of the third, vary but slightly for each species. The most 

 noticeable variation is in G. oligosticta, the other examples 

 differing from the one figured in the addition of a segment 

 to each hook, so that the anterior hook has a short stem and 

 the posterior quite a long one. 



The following table gives the number of dorsal rays, of scales 

 in a longitudinal series, and the distribution of the species : — 



