566 SCIENCE PROGRESS 



tribution of particular species is frequently limited in a seemingly- 

 very arbitrary manner, whereas the salinity of the particular 

 area in which they live is in reality the factor limiting their 

 distribution. The phenomenon may have indirect results also, 

 for the movements of fishes are in some cases considerably in- 

 fluenced by the distribution of the species of Crustacea or other 

 organisms upon which they feed principally. 



(b) Fishes. — In this group the division occurs between those 

 organisms which are at the mercy of the external medium and 

 those which maintain their own osmotic pressure more or less 

 irrespective of outside conditions. The cartilaginous fishes, the 

 elasmobranchs, communicate freely through their gill mem- 

 branes with the water, whereas the bony fishes, the teleosts, 

 possess gill membranes which are quite, or very largely, im- 

 permeable. 



Hence it results that in the former class there are sea-fishes 

 and fresh-water fishes. The same fish never passes from one 

 to the other, though it may pass from sea-water to less saline 

 estuarine water ; in which case its blood and body salts undergo 

 a change, or its total weight changes, till it is again in osmotic 

 equilibrium with its surroundings. Such changes have been 

 studied among others by Sumner, Bottazzi, and by Dakin. 



The teleosts, however, have a mean value for the osmotic 

 pressure of each species, even though there are in some of them 

 quite considerable changes with changing salinity. In the 

 anadromous fishes such as the salmon the impermeability of the 

 gill membranes is of a very high order. For between the 

 osmotic pressure of the fish at sea and in the river there is little 

 or no difference . The eggs , however, are deposited in fresh water, 

 and it is extremely probable that these possess a low osmotic 

 pressure, which gradually increases as maturity is reached. 

 Thus the method whereby the salmon has isolated itself from 

 its surroundings enables it to feed on the abundant life of the 

 sea and to pass again without injury into the safety of the upper 

 reaches of the rivers for the purposes of spawning. When one 

 considers that the freezing-point of the blood of most teleosts 

 is about — 0'6° to —o'q , corresponding to about 7*2 to io*8 

 atmospheres pressure at o° C, whereas that of the Atlantic is 

 about — 2*0 ° or 24 atmospheres, it is evident that the barrier 

 of an impermeable membrane is very necessary. 



(c) Reptiles. — This class does not appear to have been in- 



