THE BALANCING FUNCTION OF SALTS 121 



must either get into the cell or else change the surface of it in 

 such a way as to alter its permeability. Osterhout found that 

 salts are divided into two groups: one group behaves like calcium 

 chloride and causes the resistance of the cell to increase (or the 

 permeability to diminish) and later diminish; while the other 

 group, in which sodium chloride belongs, causes the resistance to 

 decrease (or the permeability to increase) from the very start 

 (Fig. 7). 



Those which belong in the first group are the chlorides, nitrates, 

 etc., of the bivalent and trivalent metals including CaCl 2 , CaN0 3 , 

 and MgCl2; while in the second group are to be found the cor- 

 responding salts of the monovalent metals such as KN0 3 , KC1, 

 NaN0 3 , and NaCl. Acids belong in the first group and alkalies 

 in the second. Any one of the substances in the first group as 

 well as any one from the second will cause the death of the plant 

 if given alone; but, when placed in the proper proportions with 

 an electrolyte from the other group, antagonism results and the 

 plant is not seriously injured. 



An Explanation of Antagonism. — As has been pointed out by 

 Osterhout, any theory of antagonism must explain the following 

 points : 



1. Why both sodium and calcium alone are toxic. 



2. Why antagonism results in mixtures of the two. 



3. Why they produce opposite effects upon the permeability 



of the protoplasm. 



4. Why the increase in resistance produced by calcium chloride 



is later followed by a decrease. 



5. Why all toxicity disappears in a balanced solution. 



Many theories have been advanced to explain the facts of cell 

 permeability and the phenomena of antagonism. The test of a 

 theory is the number of facts it can explain, the number of results 

 that can be predicted by its use, and its agreement with the facts 

 of the related fields (in this case colloidal chemistry and biophysics) . 

 At the present time there have been collected great masses of facts, 

 but no coherent scheme has been suggested to explain them as a 

 unit. Such work needs badly to be done. The general nature of 

 colloids, the behavior of salt molecules in solution (i. e., their 

 dissociation into ions bearing electrical charges), and the properties 

 of porous filters are sufficiently well known to provide us with a 

 base for future operations. Accordingly Raber has proposed a 



