120 RICHARDS — THE ATOMIC WEIGHT OF NITROGEN". [April 7, 



silver for weighing. Scott seems to have taken no especial precau- 

 tions as regards either of these complications, hence the possible 

 error of his work, even when corrected in the manner described 

 above, is as great as the difference between his work and that of 

 Stas. Thus it cannot be said to militate against that work, except 

 in pointing out that it is possible to obtain colorless ammonic 

 bromide. 



Scott also performed two analyses of ammonic chloride, which 

 gave a result (14.031) more nearly like that of Stas than that from 

 the bromide, but likewise somewhat lower. Since the analyses are 

 hardly numerous enough to be conclusive, and since no especial 

 pains seems to have been taken' to prevent the injurious effect of 

 the solubility of argentic chloride or the inclusion of argentic 

 nitrate in the precipitate, no further attention need be given to 

 this result, although it must be considered as more satisfactory 

 than the preceding work on the bromide. 



No light is shed upon the doubtful situation by the conclusions 

 of D. Berthelot^ and of Leduc^ concerning the density of nitrogen, 

 since they depend upon the precise fulfillment of the rule of 

 Avogadro, a generalization which is undoubtedly only an approxi- 

 mation.^ 



Other recently published results are recorded in the table below. 



Thomsen {Zeiischr. Phys. Chem., 13, 398, 1S94) 14.021 



Hardin (/. A7n. Ch. Soc, IS, 990, 1896) 14.01 



Hibbs (Thesis, U. of Pa., 1896) 14.01 



Dean (/. C/iet?i. Soc. Trans., 77, 1 17, 1900) 14-031 



Ramsay and Aston (^Gesell. d. Naturforscher 2ind Aerzte, Allg. Th. 



(1903) 8) 13-903 



Richards and Archibald ( Pi oc. Am. Acad., SS, 469, 1903) I4-039 



Thomsen's result was obtained by weighing the hydrochloric acid 

 and ammonia required to neutralize one another. In my recent 

 experiments on the atomic weight of magnesium, in cooperation 

 with Prof. H. G. Parker, it was found that hydrochloric acid gas 

 is by no means easy to dry thoroughly. It is therefore not impos- 

 sible that Thomsen weighed some water with his acid, thus causing 

 the weight of ammonia, and hence of nitrogen, to appear too low. 



1 ComJ>. Rend., 126, 954, 1030, 141 5 and 1 50 1 (1898). 



2 Leduc, Comp. Rend., 125, 299 (1S97). 



3 Ramsay and Steele, Phil. Mag. ^ Oct., 1903, p. 492. 



