A REVISION OF THE ATOMIC WEIGHTS OF IODINE AND SILVER. 



131 



The agreement in the air displaced as determined experimentally and as 

 calculated from the observed density is as close as could be expected, showing 

 that the vacuimi correction determined on page 129 is correct. It is to be noted 

 that if air is adsorbed by the powder the observed weight of air displaced would 

 be less than that calculated from the density. This seems to be actually the 

 case to a very slight degree. 



The Ratio of Silver to Iodine Pentoxide. 



Series I. By G. S. Tilley. 



Sample I of Iodic Acid and Sample B of Silver were Used. 



The atomic weights of iodine and silver calculated from the ratio of silver to 

 pentoxide depend, of course, upon the ratio of the atomic weights of silver and 

 iodine. The latter ratio has recently been subjected to careful revision by one 

 of us,2 and has been found to have the value 0.849943. Using this ratio the 

 atomic weight of silver calculated from the above data is 107.847. 



This value is highly sensitive to changes in the ratio of silver to iodine, a posi- 

 tive error of o.oi per cent producing a negative error of 0.03 per cent in the 

 atomic weight of silver. Hence, if the above value of the atomic weight of silver 

 is too low, it might be expected that the ratio of silver to iodine is too high. 

 From Baxter's results it seems certain that the ratio of silver to iodine is at any 

 rate no higher than the value used in our calculations. 



The result of the experiments which have been described is unexpected. For 

 although several recent investigations have shown that the atomic weight of 

 silver may be possibly as low as 107.87, referred to oxygen 16.000, no evidence 

 has yet been published which indicated a value as low as the foregoing. Ac- 

 cordingly a new series of experiments was undertaken, similar to the early one, 

 with even greater pains to avoid possible sources of error. 



^ The first two analyses were inadvertently mixed, and hence are combined in the table. 

 - Baxter: Proc. Amer. Acad., 40, 419 (1904); 41, 73 (1905); Jour. Amer. Chetn. Soc, 26, 

 1577; 27,876; ZeiL anorg. Chem., 43, 14; 46,36. See previous paper. 



