Hans Jenny 127 



In terms of the oscillation volume theory (75), the constant K com- 

 prises four oscillation volumes, conditioned by the two different ions 

 each being held on two different adsorbents. For the system under con- 

 sideration: 



K = 



l_ v N 



V* 



JfNa 

 _ ^NH4_ 



For interacting mono-divalent ions we write: 

 NH 4 



Ion-X 



NH, 



-f- Ca Amberlite 



Ion-X Ca + 



NH 4 

 NH 4 



Amberlite 



Applying the base exchange equations of Krishnamoorthy, et al. (26), 

 the following equilibrium equation may be derived: 



Ca(ii Ca + NH 4 ) I Ion-X 



(nh 4 



Amberlite 



NH 4 Ion-X 



Ca(i| Ca + NH 4 ) Amberlite 



-= k 



The symbol k A represents the equilibrium constant for the system 

 Ca-Amberlite -f- 2 NH 4 C1 and k x the constant for Ca-Ion-X -j- 2 

 NH 4 C1. 



In numerous experiments Sengupta has shown that the exchange of 

 ions of equal valency (NH 4 -Na, NH 4 -K, Ba-Ca) between two ad- 

 sorbents is normal, that is, the observed equilibrium constant is identi- 

 cal, within experimental error, with the calculated constant (Table III). 

 The ammonium-cesium pair possibly constitutes an exception. 



This identity does not exist for ion pairs of unequal valency (Na-Ca, 

 NH 4 -Ca). If large particles of ammonium-Amberlite are mixed with 

 small particles of calcium-Amberlite (same type of adsorbent), the ex- 

 change constant should be 1, but according to Table III it is over twice 

 as large — namely, 2.1 1. In every case listed in Table III, the observed 

 exchange constant differs profoundly from the calculated one. 



Interestingly enough, the abnormality disappears when free salts are 

 added to the adsorbent mixtures. Now, the observed exchange constant 

 agrees with the calculated one. At present we do not know whether 

 this peculiar behavior of the mono-divalent ions is the result of experi- 

 mental technique or of colloid interaction. 



