muller's topknot. 171 



Another mark of distinction is dcscrihod as existing in the 

 roughness of the under side; but with respect to this we 

 hesitate with the expression of a doubt as regards its constancy. 

 There are variations of proportion in the parts of the bodv 

 which may be observed when these fishes are brought together, 

 but the sepai'ation of the ventral fins from the anal in Bloch's 

 Topknot, whereas they are closely joined together in the 

 present species, is a more decisive ground of distinction between 

 them; although a question may arise whether the not very 

 distant separation of these organs should be deem.ed sufficient 

 to consign them, as has been done, to different genera. The 

 underlapping of the termination of the dorsal and anal fins 

 beneath the body, as here represented, appears to afford a 



more characteristic distinction for constituting a separate genus, 

 as distinguishing these two species of fish from all others of 

 this family with which we are acquainted, as well in form as 

 probably in habit. 



The length of the example described which was of the usual 

 size, was six. inches and a half, and in width with the fins 

 extended four inches and a half; but without measuring the 

 fins the breadth was three inches; from the snout to the 

 hindmost extent of the gill-covers one inch and six eighths; 

 the gape wide, opening downward; the mystache wide, teeth 

 small; the lower eye advanced before the upper. The lateral 

 line rises in an arch over the pectoral fin, and from thence 

 backward straight. The upper surface, as well of the head as 



