210 MB. F. DAT ON AMPHIBIOUS AND ■ 



surface of the pharynx, is seen in the amphibious Lepidosiren ; 

 but it seems to have been overlooked that the respiratory sac of 

 Saccobranchus is likewise connected with the commencement of 

 the alimentary canal and at its dorsal surface. 



The air-bladder, or swim-bladder, is found amongst the fresh- 

 water fishes of India in two distinct forms— (1) enclosed in a bony 

 case formed by a portion of the first or second vertebra or au- 

 ditory ossicles, (2) free in the abdominal cavity. Thei-e are inter- 

 mediate forms ; and (3) it may even be absent. In the Physo- 

 stomi, as in the Siluroids and Carps, whether enclosed in boue or 

 free in the abdominal cavity, it possesses a duct opening into 

 the alimentary canal. Much stress has also been laid upon the 

 interior of this swim-bladder being occasionally found subdivided, 

 as in the respiratory sacs of reptiles ; but the same has been ob- 

 served in the respiratory air-sac of the Cuchia eel. 



The pneumatic duct, as already observed, has been looked upon 

 as the homologue of the trachea, permitting in the embryo an ex- 

 tension of the mucous membrane lining the alimentary canal to 

 the swim-bladder. There are some objections to this view. The 

 pneumatic duct, it must be admitted, does open into the alimentary 

 canal at a spot situated somewhere between the end of the stomach 

 (as in the herring) and the pharynx. But in the majority of 

 Teleostean fishes which possess swim-bladders, more especially in 

 the marine forms (excluding Clupeidse), there is no pneumatic 

 duct and no connexion between the swim-bladder and alimentary 

 canal ; in many there is no swim-bladder at all, the same as in 

 most fishes there is no respiratory sac. 



I do not propose in this place to inquire into what is the homo- 

 logue of the swim-bladder, as I hope to do so at some future date ; 

 but I will merely point out that in the carp, although it possesses 

 a pneumatic duct leading from the swim-bladder into the alimen- 

 tary canal, it has also a communication with the acoustic apparatus ; 

 in fact, such in many fishes is effected either by a tubular pro- 

 longation or by chains of ossicles, termed by Weber the malleus, 

 incus, and stapes, in which Owen observed he had mistaken a 

 relation of analogy for one of homology. 



Lastly, I would observe that the artery to the swim-bladder 

 carries arterial blood from the abdominal aorta, coeliac artery, or 

 branchial vein, whereas that to the lungs or respiratory air-sacs 

 conveys venous but returns arterial blood. 



In the genera SaccobrancJius and Glarias we find a swim- 



