290 PROPERTIES CONFERRED BY COLLOIDAL CONSTITUENTS 



respect. If, however, they are immersed in solutions which are 

 hypertonic to these, for example in eight per cent, cane-sugar or 0.8 

 per cent, sodium chloride, they lose a quantity of water and in twenty- 

 four hours they are found to have shrunk decidedly in volume. Evi- 

 dently, then, the sugar or salt cannot enter the limiting membranes of 

 the cells of the skin while water can pass through them freely in the 

 direction tissues -> external medium. One might imagine, therefore, 

 that the epithelium of a tadpole resembles an ordinary semipermeable 

 membrane, permitting the passage of water but not of dissolved 

 substances. If this were really the case, then on immersing the tad- 

 poles in solutions which are hypotonic to 0.6 sodium chloride we should 

 expect them to take up water and to increase in volume just as much as 

 they decrease in volume in hypertonic solutions. This does not occur, 

 however, and tadpoles immersed in hypotonic solutions do not take up 

 water to any greater extent than from isotonic solutions. We can only 

 infer, therefore, that the superficial epithelium of the tadpole permits 

 the passage of water from within outward, but not in the reverse 

 direction; that this membrane is permeable to water in the direction 

 tissues -> external medium, but not in the direction external medium 

 -> tissues. 



The property of one-sided permeability is displayed by many living 

 membranes, but not by all. A very striking contrast is shown in this 

 respect by the Pavement Epithelium which lines the peritoneal cavity, 

 on the one hand, and the Columnar Epithelium which lines the lumen of 

 the small intestine, on the other. Thus Heidenhain introduced 50 c.c. 

 of a three per cent, solution of Glucose into the peritoneal cavity of a 

 dog, and at the same time 44 c.c. of the same solution into an isolated 

 loop of the intestine. After ninety minutes the quantity and composi- 

 tion of the residual fluid in the peritoneal cavity were as follows : 



Quantity of fluid. Glucose. Sodium chloride. 



19.5 c.c. 1 . per cent. . 55 per cent. 



while in twenty-five minutes the composition of the residual fluid in 

 the loop of intestine was as follows : 



Quantity of fluid. Glucose. Sodium chloride. 



19 . c.c. 3 . 8 per cent. . 04 per cent. 



From the peritoneal cavity both water and glucose had issued into 

 the tissue-fluids, the glucose even more rapidly than the water, while 

 sodium chloride, which was absent from the fluid originally introduced, 

 had diffused from the tissues into the peritoneal cavity. The peri- 

 toneal epithelium, therefore, behaved like a membrane of parchment, 

 permitting the passage of dissolved substances in either direction in 

 proportion to their relative concentrations on the two sides of the 

 membrane. 



From the intestinal loop, both water and glucose had issued into the 

 tissue-fluids, water somewhat more rapidly than glucose. But prac- 

 tically no sodium chloride had diffused into the intestinal fluid from the 



