1004.1 KICHARDS— THE ATOMIC WEIGHT OF XITROGEN. 121 



Among the other investigations those of Hardin and Hibbs were 

 very carefully carried out, but the quantities of material used 

 were so small that these experiments could hardly be expected to 

 determine accurately the second decimal place of the atomic 

 weight. They each weighed portions of substance containing on 

 the average only about 0.05 gram of nitrogen, and hence to be cer- 

 tain of a unit in the second decimal place of the atomic weights 

 the weighing must be certain to within 0.00003 gram. Such pre- 

 cision is almost impossible when one is using a vessel weighing 71 

 grams, as Hardin did. Moreover, Hardin seems to have made no 

 correction for the trace of electrolyte included in the film of silver 

 which formed one of his standards of reference. 



The work of Dean depended upon the volumetric analysis of 

 argentic cyanide. The method of work was too indirect to carry 

 great weight, even if volumetric analysis were at best a process 

 accurate enough for the degree of precision needed. 



The work of Sir William Ramsay and Miss Aston is interesting 

 because it involved the analysis of unstable compounds of azoimide. 

 The extraordinarily low result, a whole per cent, less than the 

 usually accepted value, is not easily explained. The authors tenta- 

 tively suggest once 'more the revolutionary assumption that the 

 chemical combining proportions are not constant. This idea is 

 by no means a new one, having been seriously advanced by J. P. 

 Cooke in 1855^ and again by Butlerov in 1882.^ In both these 

 older cases it is now fairly certain that there is no need of such an 

 iconoclastic assumption ; in Cooke's zinc antimonide crystals the 

 solid solution of excess of antimony or zinc in the crystals was 

 probably;the cause of the observed irregularities, and in Butlerov's 

 case the analytical data upon which the conclusions rested were 

 probably faulty. Much more recently the experiments of Heyd- 

 weiler have suggested that possibly a slight change in weight may 

 take place during chemical reaction ; but the changes which he 

 observed are so small as to be of entirely another order from this 

 deficiency of a whole per cent, in nitrogen. On studying Ramsay 

 and Aston's work it seems not impossible that the hydrolysis and 

 slightly reducing action of the weak and unstable nitrohydric acid,* 

 may have caused a deficiency of nitrogen in the salts which they 



1 Mem. Ajh. Acad., V, 23 (1855). 



2 Chem. Centralblatt, 1SS2, 740. 



aCurdus and Radenhausen,/. Pr. Chem. (2), J^3, 207 (1891). 



