CHEMISTRY — BAXTER. 209 



Baxter, Gregory P., Harvard University, Cambridge, Massachusetts. Grant 

 No. 579. Determination of atomic zveights. (For previous reports see 

 Year Books Nos. 3, 4, 5, 6, and 7.) $1,000. 



With the assistance of grant 579 the following researches were carried on 

 under Professor Baxter's direction : 



Since the atomic weights of silver and iodine referred to oxygen calculated 

 from the ratio of silver to iodine pentoxide (recently published, see Year Book 

 No. 7, pp. 189-192) are dependent upon a knowledge of the ratio of the 

 atomic weights of silver and iodine, and since a given percentage error in 

 the latter ratio is multiplied by three in the atomic weights of silver and 

 iodine, it is highly desirable to reinvestigate the ratio of silver to iodine. Such 

 an investigation is now in progress. 



The analysis of tri-silver phosphate, begun some time ago by Dr. Grinnell 

 Jones (see Year Book No. 6, p. 184), has been completed. Four different 

 samples of tri-silver phosphate were precipitated under as widely different 

 conditions as possible. Sample O was prepared by adding a dilute solution 

 of silver nitrate to a dilute solution of di-sodium phosphate; sample N by 

 neutralizing the mother-liquor of sample O with ammonia. Sample P re- 

 sulted from the addition of a solution of sodium ammonium hydrogen phos- 

 phate to a solution of silver nitrate, and sample R from the addition of 

 sodium di-ammonium phosphate to silver nitrate. The different samples 

 were thoroughly washed with water and dried. Portions of the phosphate 

 were heated in a current of pure dry air for several hours at 400° C. ; then, 

 after being weighed, they were analyzed by solution in nitric acid and pre- 

 cipitation with hydrobromic acid. 



A slight amount of residue insoluble in dilute nitric acid was found to have 

 essentially the same proportion of silver as the phosphate itself. This resi- 

 due was always dissolved in more concentrated nitric acid before the pre- 

 cipitation of the silver bromide ; hence no correction for the residue is 

 necessary. 



Since silver phosphate can be fused only with the greatest difficulty, the 

 determination of the water in the dried salt was accomplished by dissolving 

 the phosphate, after being dried at 400°, in fused silver chloride and collect- 

 ing the water set free in a weighed phosphorus pentoxide tube. The water 

 actually found in the salt was less than 0.0005 per cent, a negligible quantity. 



The density of the salt was found by displacement of toluol in two experi- 

 ments to be 6.37. 



By determining the gain in weight when air was admitted to an exhausted 

 tube containing a weighed amount of silver phosphate, it was shown that the 

 salt does not adsorb appreciable amounts of air. 



