so DYNAMICS OF LIVING MATTER 



the rate of diffusion of salts through the membrane which is retarded in 



this case. A pure NaCl of the concentration — or f m only prevents 



the formation of an embryo when the egg is put into the solution imme- 

 diately after fertilization. If, however, the egg is put for the first twenty- 

 four hours after f ertiUzation into normal sea water and then into the pure 



NaCl solution of the above-mentioned concentration, the — or f w 



NaCl solution is not so toxic. In all probability the membrane of the 

 egg or the cells becomes more hardened or less permeable during the 

 first twenty-four hours. 



The antagonistic effects between two salts with a bivalent cation 

 are not so general, yet I found that \ c.c. of a ^g w SrClj solution 

 diminished somewhat the toxicity of a -^ w solution of MgCl^. 



While in the case of the membrane of the newly laid Fundulus egg, 

 the addition of a trace of a salt with a bivalent cation sufficed to anni- 

 hilate or diminish the toxic effect of a pure NaCl solution, we never 

 find such simple relations for the Fundulus after it is hatched or for 

 any living cell that is exposed directly to the solution without the inter- 

 ference of a dead membrane, like the one which surrounds the fish egg. 

 For such directly exposed living tissues or animals, it is a rule that a 

 pure NaCl solution of sufficient concentration requires, besides the 

 CaClj, a trace of KCl, in order to become harmless, as was shown in the 

 above-mentioned case of marine Gammarus. Besides, it is not possible 

 to substitute in that case for the Ca any bivalent cation; only Sr can 

 serve as a substitute for Ca in these cases. These limitations become 

 inteUigible on the assumption that the surrounding salts diffuse slowly 

 into the cells. As long as this diffusion is so slow that the secretory 

 activity of the cells or glands of an animal may remove them as fast as 

 they enter, the cell or the animal may Uve in such a solution. I consider 



this the reason why a Fundulus may live in a — NaCl solution, while it 



cannot live in a — NaCl solution. A second condition for the main- 



2 



tenance of Hfe is, according to this hypothesis, the continuation of the 

 action of the secretory mechanism. If the latter depends on the con- 

 tractile power of the protoplasm, as I believe it does, we can understand 

 that, in order to make a NaCl solution harmless, not only Ca but also 

 K are required. We shall see in a later lecture that apparently Na, 

 Ca, and K are required for the contractile phenomena of protoplasm. 

 Hober and Gordon* have pointed out the existence of an antago- 



* Hober und Gordon, Hofmeistet's Beitr'dge sur chemischen Physiologie und Pathologie, 

 Vol. 5, p. 432, 1904. 



