CHROMIUM. 117 



and Marchand, made three distillations of mercuric chlo- 

 ride with lime, and got the following results : 



12.048 gnn. HgClj gave 8.889 grm. Hg. 73. 7S0 per cent. 



12.529 " 9-2456 " 73-794 " 



12.6491 " 9-3363 " 73-810 " 



Mean, 73-795, dr .006 



Combining these series we have : 



Turner 73-79i> ± -005 



Millon 73-845. ± .010 



Svanberg 73-795. ± -006 



General mean 73-798, =n -0034 



In this mean Turner's figures undoubtedly receive undue 

 weight, for, on experimental grounds, they are probably in- 

 ferior to both of the other series. It is better, however, that 

 the general mean should remain as it is, than that I should 

 deal arbitrarily with any of the data. 



We now have three figures to calculate from : 



Per cent, of Hg in HgO 92.601, zb .0014 



" HgS 86.2127, ±z .0027 



HgCl^ 73-798, ± .0034 



These give us three values for the atomic weight of mer- 

 cury and a general mean as follows : 



From HgO Hg = 199.786, ± .059 



From HgS " = 200.016, zt .088 



From HgClj " = 199.239, ± .086 



General mean " =z 199.712, ±z .042 



If = 16, then this becomes 200.171. 



CHROMIUM. 



Concerning the atomic weight of chromium there has 

 been much discussion, and many experimenters have sought 

 to establish the true value. The earliest work upon it hav- 



