A REVISION OF THE ATOMIC WEIGHT OF ARSENIC. 77 



A solution of hydrochloric acid was purified by distillation after dilution. 



Hydrochloric-acid gas was generated by dropping C. P. concentrated sul- 

 phuric acid into C. P. concentrated hydrochloric acid. The acids were shown 

 to be essentially free from arsenic. 



All the water used in the research was purified by distilling the ordinary dis- 

 tilled water of the laboratory, once with alkaline permanganate and then once 

 alone, in both cases with the use of block-tin condensers which required no 

 cork or rubber connections to the distilling flasks. 



Quartz or platinvmi vessels were always employed in place of glass, whenever 

 glass was unsuitable. 



METHODS OF ANALYSIS. 



The first method of analysis employed was that of converting the silver arse- 

 nate into silver chloride by heating in a current of hydrochloric-acid gas. Since 

 this process does not involve transfer of material it should be capable of giving 

 results of great accuracy. Glass and porcelain are unsuitable for containing 

 the arsenate during this process on account of the certainty of their being at- 

 tacked. The first attempts at using quartz for the purpose resulted in slight 

 etching of the surface of the tube where it came in contact with the salt. Ex- 

 perience showed, however, that with careful manipulation the attacking of the 

 quartz could be wholly prevented. The vessel used to contain the arsenate was 

 a quartz tube nearly 2 cm. in diameter but joined to small tubes at each end. 

 These tubes were exactly like those employed by Richards and Jones in the 

 conversion of silver sulphate into silver chloride.^ After the tube had been 

 weighed by substitution for a coimterpoise similar in shape and size, a suitable 

 quantity of silver arsenate was introduced, and the tube and contents were 

 heated in a current of pure dry air for between 7 and 8 hours at 250° C. Al- 

 though this treatment is not sufficient to expel last traces of moisture, it was 

 hoped that by uniform treatment of the arsenate in all the analyses the propor- 

 tion of water retained by the salt could be reduced to a constant percentage, 

 which could be determined in separate experiments. 



The complete drying of the salt by fusion was not permissible because of de- 

 composition of the arsenate at temperatures in the neighborhood of its fusing 

 point. During the drying of the arsenate the quartz tube was surrounded with 

 a cylinder of thin platinum foil and was contained in a hard-glass tube connected 

 with an apparatus for furnishing a current of pure dry air. The hard-glass tube 

 was heated by means of two aluminum blocks 15 cm. by 13 cm. by 5 cm., one 

 placed above the other, the upper surface of the lower block and the lower sur- 

 face of the upper being suitably grooved to contain the tube. The blocks were 

 bored to contain a thermometer, the bulb of which was located near the middle 

 of the tube. This oven (fig. 4) could be readily maintained at constant temper- 



^ Pub. Car. Inst., No. 69, 69(1907); Jour. Am2r.Chcm.S0c., 2g,S;i$', Zeit. anorg. Chem., 

 55, 80. 



