728 PROCEEDINGS OP THE AMERICAN ACADEMY. 



Still, it is not quite identical with them. The experimental curves all 

 coincide more nearly with one another than they do with the theoretical, 

 and all show a certain asymmetry which that does not. In the case of the 

 experimental curves it is evident that the apex comes at a point where 

 more than one half the silver oxide is decomposed. Still the theoretical 

 and experimental curves fall much nearer together than was to have 

 been predicted, and it is possible that the difference that exists may be 

 due to faults in the experimental method; thus, for example, it is not 

 impossible that near the point of maximum velocity the reaction may 

 take place so rapidly that heat cannot be absorbed from the bath rapidly 

 enouo-h to maintain a constant temperature, — a condition which would 

 destroy the symmetry of the curve. Still, it is hardly probable that this 

 was the case, for if it had been there would have been a wide difference 

 between the curves for the highest and the lowest temperatures ; since at 

 the latter the maximum velocity is only one third or one fourth as great 

 as at the former. 



The Stddt op Different Varieties of Silver Oxide. 



In the experiments described hitherto the same silver oxide was used, 

 namely, a preparation by Merck labelled " highest purity." When this 

 was exhausted the experiments were continued with an oxide furnished 

 by Eimer & Amend. It was soon found, however, that with this sub- 

 stance no results could be obtained comparable with the preceding. The 

 reaction velocity was far more rapid than with the old oxide. Thus, at 

 340° the maximum velocity was passed in less than an hour, and in three 

 hours all action had ceased. Furthermore, the decomposition was irreg- 

 ular, so that under the same conditions inconstant results were obtained. 

 This sample of oxide was therefore abandoned, and another bottle of 

 Merck's procured. This was of the same grade and bore the same label 

 as that used in the previous experiments, but it, too, decomposed with 

 great rapidity. Thus, the velocity at 352" reached a maximum in thirty- 

 seven minutes, and the reaction was over in three hours, whereas with the 

 original sample at this temperature the maximum came at the end of four 

 hours. The great irregularities shown by the Eimer & Amend samples, 

 however, did not appear. Such irregularities most probably are due 

 to lack of uniformit y in the oxide. 



The great difference in the rapidity of decomposition of different 

 samples of oxide made it desirable to study the effect of the method of 

 prei^ration of the oxide upon its decomposition. It seemed probable 



