560 AMPHOTERIC COLLOIDS. V 



It seemed very important to test this possibility, since a decision 

 might serve as a crucial test for or against our purely chemical con- 

 ception of the behavior of colloids, according to which the effect 

 01 an electrolyte upon the physical properties of the colloid is due 

 to the formation of real chemical compounds between the colloid 

 and one of the ions of the electrolyte. If it can be shown that SO4 

 has a similar depressing effect as a bivalent cation we must also be 

 able to prove that SO4 combines with two molecules of gelatin; and 

 if it can be shown that POa, citrate, oxalate, succinate, etc., have no 

 such depressing effect upon the physical properties of gelatin but 

 behave like Br, CI, NO;, and acetate, we must also be able to prove 

 that one molecule of each of these acids combines with only one 

 molecule of gelatin. 



We are able to decide this question in favor of the purely chemical 

 theory of colloids. 



The decision was rendered possible through the proof that we can 

 ascertain the amount of acid in combination with a given mass of 

 gelatin by titration with NaOH.^ If we treat gelatin with acid and 

 titrate afterwards with NaOH, we obtain the mass of acid in com- 

 bination with the gelatin with the aid of two corrections; namely, 

 first by deducting from the titration value a constant value which is 

 1.8 cc. 0.01 N NaOH for 10 cc. of a 1 per cent gelatin solution, titrat- 

 ing to neutrality (pH = 7.0). This constant value 1.8 cc. 0.01 n 

 NaOH is the quantity of alkali required to bring 0.1 gm. isoelectric 

 gelatin to pH 7.0. The correctness of this method was demon- 

 strated in the case of HBr and HCl in which it was possible to com- 

 pare the amount of NaOH required for the titration of a 1 per cent 

 gelatin solution with the results of volumetric analysis for Br accord- 

 ing to Volhard's methods. It was found that the Br (or CI) num- 

 bers agreed with the NaOH numbers when the constant value men- 

 tioned above was deducted. In that paper^ we had titrated for 

 pH = 9.0, while in the experiments reported in the present paper 

 we titrate for pH = 7.0; as a consequence the value of the constant 

 given in the first paper was higher than 1.8 cc. 0.01 n NaOH. 



3 Loeb, J., /. Gen. Physiol., 1918-19, i, 363. 



