1885.] NATURAL SCIENCES OF PHILADELPHIA. 125 



matter to prove or disprove it hy experiment. As above said, 

 however, it is in consonance with physical laws in certain cases, 

 and in such cases it very probably gives a correct view of the 

 function of the organ. Yet there are many cases in which the 

 small size of the organ must render it nearly or quite useless for 

 anj^ such purpose, while its entire absence in very many instances 

 of active species of fish, shows that this function is of no special 

 value to the fish tribe as at present constituted, and suggests 

 that the original purpose of the air-bladder must have been very 

 different from that here surmised. A general examination of the 

 subject may aid us in gaining some definite conception of the 

 character of this original function. 



The air-bladder of fishes is an internal sac, occupying usually 

 the dorsal aspect of the body, and in some cases connected with 

 the intestinal canal by a pneumatic duct, though in the great 

 majorit} r of cases this duct is wanting, or its cavity is closed. 

 Thus, most generally, the bladder is a closed sac, containing gas 

 which could only have come from the blood-vessels, with which 

 it is abundantly provided in the form of retia mirabilia. This 

 gas, in fresh-water fishes, is nearly pure nitrogen. In ocean 

 fishes, particularly the deeper swimmers, oxygen is in excess, 

 and has been found in some instances to constitute as much as 

 87 per cent, of the contents. Some naturalists advance the 

 singular theory that the absolute weight of the fish may be 

 increased or diminished by compression or dilation of this gas, 

 as if the same quantity of gas could change its weight by a 

 variation in its density. But that the relative weight of the fish, 

 or its displacement of water, might be changed by a variation of 

 its body-volume, through a variation in the state of compression 

 of the air-bladder, is unquestionable, though in those numerous 

 cases where the bladder is very small its influence must be of 

 very little aid in the movements of the fish. 



In addition to its use in aiding the fish to ascend or descend 

 in the water, its dorsal position must also act to keep the back 

 of the fish uppermost. In certain cases it also doubtless 

 subserves another gravitative purpose — that of elevating or 

 depressing the anterior region of the body, at the will of the 

 fish. This is possible in those cases in which the bladder has a 

 considerable longitudinal extension. In some cases, it is pro- 

 longed into the tail of the fish. In others, it sends processes 



