46 CHARLES F. W. MCCLUEE 



K was present began to gain weight in excess of the glove con- 

 taining Ringer in which K was absent. This form of behavior, 

 as previously shown (fig. 16), is characteristic of certain 

 skins through which, at the time of their removal from the 

 body, water does not immediately begin to diffuse when both 

 sides of the skin are bathed by solutions having the same 

 osmotic pressure. In one of these two experiments the initial 

 gain followed by the loss in weight (E, fig. 20) observed dur- 

 ing the first six hours of its suspension, is probably due to the 

 absorption and the loss of water by the skin itself, and not to 

 a transport of fluid through the skin into the glove during 

 the first three hours, and a transport from the glove through 

 the skin outward in the fifth and sixth hours. 



Eight more experiments of still another type were made on 

 skin gloves in which the inside of the skin was turned outward. 



Experiment 555 (fig. 21) 



The right and the left skin gloves of the same frog (R. 

 pipiens), with the inside of the skin turned outward, were 

 filled with Einger minus K and were suspended, respectively, 

 in Ringer and Ringer minus K. (For type of experiment see 

 diagrams G and H in fig. 21.) In five of these experiments 

 the behavior of the gloves was essentially similar to that 

 graphically shown in figure 21, in which, during the course 

 of twenty-four hours, the glove suspended in Ringer (G) lost 

 more in weight than did the glove containing Ringer minus 

 K when suspended in Ringer in which K was absent (H, fig. 

 21). In each instance a loss in weight of the glove was pre- 

 ceded by a gain, which occurred during the first hour's sus- 

 pension. This gain in weight, we have seen, is probably due 

 to an addition of water to the skin itself, through imbibition. 

 In three of these experiments a transfer of water through 

 the skin did not take place until after the gloves had been 

 suspended for several hours in the solutions. After the 

 transfer had been established, however, the rate of diffusion 

 was found to be influenced by the presence of K in the solu- 

 tion in which the gloves were suspended. When we compare 



