CHEMISTRY — BAXTER. 2I3 



A second sample of material was prepared from meteoric iron. Nickel 

 was the chief impurity contained, the greater part of which was eliminated 

 as sulphide and the remainder by precipitating the ferric hydroxide many 

 times with a large excess of ammonia, as previously described. Metallic 

 iron was obtained by processes similar to those enumerated above. 



Eighteen titrations against silverwith the first sample of material yielded as 

 an average value for the atomic weight of iron 55.835 (Ag= 107.870), while 

 16 gravimetric analyses gave an average value 55.831. In the case of the 

 meteoric material 5 titrations gave as an average result 55.829 and 5 gravi- 

 metric determinations 55.827; 4 of the analyses with the meteoric material 

 were, however, subject to slight uncertainty. The 6 analyses which were 

 free from known error gave an average value 55.832. 



The difference between the results with the two samples is no greater than 

 the probable experimental error, and therefore there can be no doubt that 

 the two specimens of material were identical. The final result for the atomic 

 weight of iron is less than one one-hundredth of a unit lower than the value 

 previously obtained by Baxter with ferrous bromide which had been sub- 

 limed in porcelain tubes and which was found to contain a small quantity of 

 alkali bromides extracted from the tubes. This investigation will be pub- 

 lished very shortly. 



The investigation upon lead bromide begun by Mr. Worsham (see Year 

 Book No. 8) was continued by Mr. Thorvaldson. The previous difficulty in 

 obtaining fused lead bromide which would dissolve completely in water was 

 apparently due to hydrolysis of the salt during solution, with the formation 

 of an insoluble basic bromide. Hydrolysis was prevented by adding a small 

 amount of acetic acid to the water in which the salt was dissolved. Several 

 preliminary experiments with material fused in a current of nitrogen and 

 hydrobromic-acid gases and analyzed as described above in the case of iron, 

 indicate a value somewhat higher than that previously obtained by Baxter 

 and Wilson. This investigation will be continued. 



Mr. G. W. Harris undertook the determination of the atomic weight of 

 arsenic by the titration of arsenious oxide against iodine. The method con- 

 sisted in subliming recrystallized arsenic trioxide in a current of dry gas 

 into a weighed glass tube. After the solution of the trioxide in dilute 

 sodium hydroxide in a vacuum to prevent oxidation by the air, and after 

 neutralization of the excess of sodium hydroxide with phosphoric acid, a 

 weighed very nearly equivalent amount of pure iodine, which also had been 

 sublimed in a current of dry air into a weighed tube, was dissolved in the 

 arsenite solution. The end-point was found in the presence of starch by 

 adding very dilute iodine or arsenite solution, with due observance of the 

 precautions to preserve neutrality during the titration by means of soluble 

 phosphates, which have been shown by Washburn to be necessary. 



Arsenic trioxide sublimed in air yields a value for the atomic weight of 

 arsenic of 74.96, referred to silver (107.870) and iodine (126.920), while 



