of the Island of Trinidad, W. I. 389 



Even in the section so restricted, there appear to be em- 

 braced several genera, which we may at a future time name 

 and characterize. Miiller and Troschel have already separated, 

 under the name of Calophysus, a most distinct genus, the typical 

 species of which was formerly placed in this group by M. 

 Agassiz, in the " Selecta Genera et Species Piscium," of Spix, 

 and by M. Valenciennes in the " Histoire Naturelle des Pois- 

 sons." 



The " preceding " species to which Valenciennes alludes in 

 the paragraph which we have above cited, form a third section 

 in his arrangement ; they are distinguished by the round and 

 smooth head ; and the presence of only six barbies, the nasal 

 ones being absent. Two or more genera seem to be also em- 

 braced in the section. The type, P. raninus, Val., belongs to 

 a new genus, which we call Batkochoglanis. The species which 

 constitute this genus have an aspect nearly similar to the North 

 American Pimelodus, but their body is even shorter and stouter 

 than in that genus, and the anal, as in almost all of the South 

 American Pimelodinse, is much shorter, and only supported by 

 from eight or nine to fourteen rays. The caudal is either 

 rounded or emarginate, and the adipose fin small. Batrocho- 

 glanis will include the Pimelodus raninus of Val. as its type, 

 and as additional species P. bufonius, Val., P. mangurus, Val., 

 and P. charus, Val. 



The fish which Lacepede has indicated as the Tachisurus 

 Chinensis, from Chinese drawings, appears to be also the type 

 of a distinct genus, but as we have never been able to examine 

 it, we are not able to give its characters. 



Another genus which was placed by Valenciennes, in a fifth 

 section of Pimelodus, but which appears to have very little real 

 affinity to any other genera of Pimelodinoe, has recently re- 

 ceived from M. Dumeril the name of Couostoma. After the 

 separation from Pimelodus of these natural genera, and several 

 others that have been founded by naturalists, the genus to which 

 the name of Pimelodus should be restricted, deprived of the 



