254 PROCEEDINGS OP THE AMERICAN ACADEMY 



By the method just described, the following results were obtained. 

 The letters a and b indicate different preparations. 



Wt. of Sb Br 3 Total Wt. of Ag 



taken. used. 



a 1. 2.5032 2.2528 



a 2. 2.0567 1.8509 



a 3. 2.G512 2.3860 



ft 4. 3.3053 2.9749 



b 5. 2.7495 2.4745 

 Mean value, 



Mean value of fifteen gravimetric de- 

 terminations previously published, 

 Theory Sb. 120 requires 

 „ Sb. 122 „ 



In order still further to control the work, we collected the bromide 

 of silver formed in the last two determinations, washing the precipitate 

 with the precautions which experience had shown to be necessary, and 

 determining its weight, first, after drying at 150° C, and, secondly, 

 after heating to incipient fusion. In b 6 there was a loss of -fe of a 

 milligramme ; in b 7 a loss of -fa of a milligramme only at the sec- 

 ond weighing. This is an absolute proof that there could be no 

 sensible occlusion of any tartaric acid or any tartrate by these pre- 

 cipitates, and, as stated in our original paper, the same test was fre- 

 quently applied, although not always, in our previous determinations. 

 It is also evident that these last experiments give us two essentially 

 distinct determinations of the atomic weight, although the materials 

 employed were identical with those of b 4 and b 5. 



Lastly, it is obvious that these gravimetric determinations, taken in 

 connection with the corresponding volumetric results, give us the most 

 conclusive evidence of the purity, both of the metallic silver used, and 



