socTii ami-:rican fisiip:s 



33 



In June, 1913, I described the remarkable little Fterohrycon 

 IcDidojii from a single specimen about an inch long which I had 

 taken out of a littl(;^ rivulet at Boca de Raspadura, Colombia. 



Further study shows that these very diverse fishes, inhabiting 

 the extreme northwest and extreme southeast corners of tropical 

 South America, are related, all of them possessing the technical 

 characters enumerated below. 



As far as known individual species of this subfamily have 

 restricted distributions which would point to the ready modifica- 

 tion of the Glandulocaudin^ under different environments. Four 

 species of Gep/i yi-ocJxn-aj' occupy contiguous teri-itory in Colom- 

 bia and Panama; and three species of Gia nd n I'na mhi are found 

 in as many rivers of southeastern Brazil. Cal iirirJi.Ui ys iporangd' 

 and possibly Pseudocorynopoma dorice are exceptions, being found 

 over wider territory, the former in the rivers emptying into the 

 Atlantic between the Parahyba and some distance north of Rio 

 Grande do Sul, and the latter from Rio Grande do Sul to Buenos 

 Aires and Paraguay. 



Of greatest interest is the ahdity to develop sexual dimorphism. 

 In the majority of the Characins. the difference between the male 

 and the female are not great. In the Glandulocaudina?, however, 

 the males frequently are (juite different from the females, and of 

 particular interest is tlie fact that this difference sometimes ap- 

 pears in one organ and sometimes in another. The opercle in 

 the males of one species is excessively modified, some of the scales 

 in another, the fins in still others ; and generally there is a gland- 

 ular pouch with specially modified scales on the caudal fin of the 

 males in all of the species. In many of the species the low^r 

 caudal fulcra of the males are modified and may be separated as 

 a spur from the rest of the fin. 



A second point of general interest is the ability to produce a 

 similar result by very diverse means. 'Adaptations arise when- 

 ever needed if they are at all possible.' If we grant that it is 

 desirable to have a moA^able spot near the middle of the body, 

 then this desideratum has been reached in Stevardia through the 

 prologation of its opercle; in Ptero'brycon by the prolongation of 

 a scale; in one species of Pseudocorynopoma and one of Grpliyro- 

 charax by tipping the pectoral with black or brown. (Permanent 

 black spots near the middle have been accomplished by one species 

 of Rhoadsia and one of Yrsicat)'u.s. members of other subfamilies'). 



The mouth in all the species is oblique and the pectorals are 



2—748 



