Evermanv (uul Kenrhdl — Fish from Central Ecuador. 97 



reason, the absence of an adipose fin in that genus is of no im- 

 portance and the only (hstinctivc cluiracter is tlie very improb- 

 able absence of ventral lins. 



Regarding this genus Regan remarks that it is allied to Arges, 

 (lifTt'ring only in the absence of ventral fins; that it is possible 

 tliat tins feature is ainiormal or accidental and that the genus 

 may be founded on a specimen of Arges hracJii/rephalvs or an 

 allied species. 



Astrohlepus grixalvii is known only from the very imperfect 

 description and monstrous figure of Humboldt ; and it probably 

 never will be found so long as one of this group without ventral 

 fins is looked for. It is not improbable that Cydojyimn guentheri, 

 occurring in the same river basin in which the Astrohlepus was 

 found common enough to l;)e used as food by the inhalntants is 

 identical with it. Huml:)oldt says regarding it : 



Tlie Fescado negro, which is largely eaten at Popayan, is not found in 

 that part of the Cauca River which is nearest the city. The physical 

 cause of this phenomenon is quite remarkable. A river impregnated 

 with sulphuric acid descends from the volcano of Purace to which the 

 inhabitants give the name of Vinegar River. It is known by the beautiful 

 cascade which it makes at the foot of the volcano. From the point where 

 the waters of the Vinegar River mingle with those of the Cauca River, as 

 far as four miles farther down, the latter is without fish, although in its 

 uppei- part the fish are quite abundant. Small quantities of the acitl, which 

 were taken for our chemical analyses, are often considerable enough to 

 injure the organization of fishes. 



Regan recognizes nineteen species of Arges, to eight of which 

 he gives new names. Of the eight supposed new species four are 

 founded on forms which had been regarded as known species. 

 Arges hoidengeri is based on Stygogeneshumboldtii of Boulenger, Ijut 

 not of Swainson . Arges eigenmanni is the Cycloplum cydojmm of 

 Putnam, and Eigenmann and Eigenmann (not P. ryclopum 

 Huml)oldt). Arges cyclopum is redescribed from four specimens 

 from some unknown locality. Arges vaUlanti is based on a 

 single specimen sent from the Paris Museum as Brontes prena- 

 dllla. Arges fissidens is founded on a part of Boulenger' s Arges 

 whymperi (two specimens). 



There is a key to the species the main divisions of which are 

 based on the character of the adipose fin. The other divisions 

 are based on extent of ventrals and pectorals, character of teeth, 

 position of ventral, etc. Among our specimens there are indi- 



