Precipitation of Colloids. 583 



for which the values of N" are nearly equal. This result, 

 however, requires confirmation. Freundlich gives one value 

 for a trivalent ion in the precipitation of As 2 S ;i , namely, 

 aluminium, which if we take N" constant, gives the result 



3 = -0001096 N- " 09 . 



This, combined with the two equations above, gives for the 

 minimal concentration of cations with approximately equal 

 values of JST " 09 , 



: 0-0008184 N" " 09 : 0-0001096 N -0 ' 09 . 



C 1 :C 2 :C 3 =0-55N 



= 5018 : 7-47:1 



= (70'8) 2 : 7-17:1, 



a bad result from the point of view of the above hypothesis, 

 but which, it may be pointed out, rests only on one result, 

 and that the measurement of a small quantity, the concen- 

 tration of the aluminium ion being *000093 gm. atoms per 

 litre. 



It is of interest now to see whether precipitation by anions 

 follows a similar rule. In investigating this matter, we are 

 met with the difficulty of not knowing what value to assign 

 to N in the case of a complex ion, which is the most usual 

 type of anion. We can, however, take three values from 

 some data by Freundlich on the precipitation of colloidal 

 Fe(OH) 3 . The left-hand column shows the minimal con- 

 centration of the anion of potassium salts. The Fe(OH) 3 

 was positively charged and at a concentration of -00163 mols. 

 per litre. 



Table III. 



Colloid :— Fe(OH) 3 -00163 mols. per litre. 



. . Minimal 



concentration. 



**Sio- Atomic h 0gl0 - 

 Minimal Atomic 

 . .. 1 number. , 

 concentration. [ number. 



CI' -00848 



Br' -0120 



I' 0145 



3928 17 1-230 

 2-079 35 1-544 

 2-161 53 1724 



The result, it will be seen, is satisfactory ;is far as it goes. 

 We obtain the equation 



C = -00215N ' 45 , 



in which it will be noticed that the value of n is positive. 



