MALAOOPTERYGII ABDOMINALES, 213 



FAMILY III. 



SILURIDtE. 



This family is distinguished from all others of the order by 

 the want of true scales, having merely a naked skin, or large 

 osseous plates. The intermaxillaries, suspended under the 

 ethmoid, form the edge of the upper jaw, and the maxillaries 

 are reduced to simple vestiges, or are extended into cirri. The 

 intestinal canal is ample, flexed, and without caeca; the blad- 

 der large and adhering to a peculiar bony apparatus ; the first 

 ray of the dorsal and pecto^^al is, almost always, a strong arti- 

 culated spine, and there is frequently an adipose one behind, 

 as in the Salmon. 



SiLURUS, Lin.(l) 



A numerous genus, easily recognized by its nudity, the mouth cleft 

 in the extremity of the snout, and in the greater number of the sub- 

 genera, by the strong spine which forms the first ray of the pectoral. 

 It is so articulated with the bone of the shoulder that the fish can 

 either depress it, or raise it perpendicularly, when it is immovable, 

 constituting a dangerous weapon, wounds from which are considered 

 as poisoned; an idea arising from the fact that tetanus frequently 

 ensues. 



The head is depressed, the intermaxillaries suspended under the 

 ethmoid and non-protractile, the maxillaries very short, but each 

 of them almost always continued into a fleshy cirrus, to which are 

 added others attached to ihe lower jaw or even to the nostrils. 

 There is no suboperculum to the gill-cover; the two superior lobes 

 of the stout and cordiform natatory bladder adhere to a peculiar bony 

 apparatus, which is connected with the first vertebrse. The stomach 

 is a fleshy cul-de-sac, the intestine long, ample, and without c3eca.(2) 



(1) Silurus and Glanis, two ancient names, at one time employed as synonymes, 

 and at another as the reverse, given to certain fishes of the Nile, Danube, and 

 Orontes, and of some rivers of Asia Minor. It is almost certain that they belong' to 

 this genus. 



(2) Hasselquist attributes caeca to the Schilh^i I have ascertained,#how-ever, that 

 the contrary is the fact. 



