24 WALTER E. CARREY. 



aquarium water and later obtained positive tests for these sub- 

 stances in the blood of Carcimis m<znus. We may conclude then 

 that the membranes of marine invertebrates as a class are per- 

 meable both to water and to salts and act exactly " like a dialyzer 

 membrane." 



The path taken by the exchanged material is not so certainly 

 known. Fredericq 6 assumes that the branchial membranes of 

 Carcimis imcmis are the permeable structures but publishes no 

 evidence supporting the view. Quinton 19 takes it for granted 

 that it is the external wall which is permeable, and this seems to 

 be true for Aplysia, the form with which he worked, as has been 

 shown by Ph. Bottazzi and P. Enriquez. 2 



That the outer wall is the permeable structure of worms may 

 be shown by a very simple experiment performed by the author 

 on Nereis and Chatopterus. Ligatures were passed about the 

 animals close to either end and drawn tight thus completely 

 closing the alimentary canal ; care was taken to avoid abrasion 

 of the integument. When placed either in fresh water or dilute 

 sea water swelling and increase in weight was obtained. 



Liuiuliis is an animal on which the permeability of the gills may 

 be easily demonstrated. These structures are borne on the ab- 

 dominal segment which may be bent ventrally to an angle of 

 about 90. When placed in this position and so propped up in 

 the aquarium that only the abdomen is under the surface of the 

 water, the mouth parts may be as much as fifteen centimeters 

 above the surface. No water can enter the alimentary canal, 

 nevertheless in equal parts of fresh water and sea water 

 ( J = 1.03), six hours sufficed to render the integument and 

 gills tense and swollen. A freezing-point determination showed 

 that J of blood had changed from 1.82 to - 1.32. In 

 another experiment with Linnilns the animal was placed astride a 

 narrow dish of fresh water and so supported that only the gills 

 dipped beneath the surface with each rhythmic oscillation. After 

 eight hours enough water had been absorbed to bring J down 

 to - - 1.41, the freezing point of the water had also changed from 

 0.02 to 0.20, and a copious precipitate of silver chloride was 

 obtained. The gills of Limulus are permeable both to water and 

 to salts. Metals in proteid combination, e. g., copper of haemo- 



