SCOTT, STUDY OF CHANGES IN MVSTELUS CAN IS 27 



bility of the body wall to water. Garrey suspended Liinulus so that the 

 gills alone were immersed in a solution of one-half sea-water plus one-half 

 fresh water. A decrease in the osmotic pressure of the blood took place 

 which demonstrated the permeability of the gills. 



Sumner ('06) inferred that the structure of the skin of most teleosts 

 was an effective barrier to osmotic exchanges between the tissues of the 

 fish and the external medium. He devised an apparatus by which the 

 body was immersed in a solution of one concentration, while the gills were 

 bathed by water of another concentration. In an experiment with the 

 carp, Cyprinus carpio, the body of the fish was immersed in fresh water 

 and sea-water bathed the gills. There was a loss of weight at the end of 

 the experiment. In the second place, the body of this fresh water fish 

 was immersed in sea-water and fresh water was supplied to the gills. The 

 fishes not only continued to live longer than in the first instance, but 

 there was no loss in weight. The result showed that no osmotic changes 

 took place through the body membranes of the carp. When the body of 

 the tautog, Tautoga onitis, a marine form, was immersed in sea-water 

 and the gills were bathed with fresh water, the fishes died in from two to 

 three hours. On the other hand, when the gills were supplied with sea- 

 water and the body was immersed in fresh water, the fishes were appar- 

 ently not affected. These ingenious experiments of Sumner, in which it 

 will be noted that the fishes were not injured, contribute strong evidence 

 for the conclusion that the gills alone are concerned in osmotic changes. 

 Dakin ('08) called attention to the fact, as did Greene, in the case of the 

 salmon, that while the contents of the stomach of the lump sucker are 

 osmotically the same as sea-water, the osmotic pressure of the coelomic 

 fluid, though separated from the cavity of the intestine by a very thin 

 wall, is the same as the osmotic pressure of the blood, which is much less 

 than that of sea-water. He thus proved that the wall of the gut is nor- 

 mally impermeable to salts except in the processes of nutrition and was 

 inclined to the belief that the membranes are semi-permeable. 



From different points of view, the evidence indicates that the gills con- 

 stitute the pathway by which the osmotic changes take place. Sumner 

 alone has attacked the problem directly. Dakin criticised Sumner for not 

 excluding the gut as a possible factor. I concluded to investigate this 

 problem in the case of the dog-fish. The following facts justify Sumner's 

 conclusion : 



The average A of the blood of two dog-fishes immersed in fresh water 

 for sixty minutes was found to be 1.597°. A male Mustelus canis, sev- 

 enty-eight centimeters long, was pithed, the body cavity was opened, the 

 oesophagus was ligated, and the fish was placed on a support out of the 



