104 Prof. F. W. Clarke on the Results obtained 



The magnificent accuracy of Stas's manipulations could 

 hardly receive more striking confirmation than is afforded by 

 these figures. 



The atomic weights of the remaining elements I give below 

 in the order of their calculation. I also give the authorities 

 whose experiments are combined in the values presented. 

 Elaborate details there is, of course, no room for. They must 

 be sought in the complete monograph when the latter is pub- 

 lished. 



Nitrogen. — Data by Regnault, Pelouze, Marignac, Turner, 

 Penny, and Stas. General mean of all, N= 14*0210 ; 

 ±•0035. This value involved the values previously found 

 for 0,K,Na,Ag, and CI. 



Carbon. — Data by Liebig and Eedtenbacher, Maumene, 

 Dumas and Stas, Erdmann and Marchand, and Stas. General 

 mean of all, C = 12*0021 ; ±*0019. Rejecting the series by 

 Liebig and Redtenbacher, and by Maumene, which involve 

 constant errors, we get from the remaining series a value of 

 0=11-9736; ±*0028. If 0=16, this becomes 0=12-0011. 

 The ratio between oxygen and carbon, therefore, is a ratio 

 between two whole numbers, 16 and 12. 



Barium. — Determined by Berzelius, Turner, Struve, Pelouze, 

 Marignac, Dumas, and Stas. General mean, Ba = 136*763 ; 

 ±•031. If 0=16, Ba = 137*007. 



Strontium. — Data by Pelouze, Dumas, and Marignac. 

 General mean, Sr=87*374; ±*032. 



Calcium. — Experiments by Berzelius, Erdmann and 

 Marchand, and Dumas. Ca=39*990; +*010. The best single 

 result by Erdmann and Marchand gives Ca=39*905 ; or, if 

 = 16, Ca=39*997. 



Lead. — Berzelius, Turner, Marignac, Dumas, Anderson, 

 and Stas. Some data were rejected. The final, most pro- 

 bable value is Pb= 206*471 : ±*021. 



Fluorine. — Berzelius, Louyet, Dumas, and De Luca. Gene- 

 ral mean, F=18'984 ; ±-007. 



Phosphorus. — The only good determinations for this ele- 

 ment are those by Schrotter and by Dnmas. General mean of 

 both series, P=30'958 ; ±*007. 



Boron. — Berzelius, Laurent, Dumas, and Deville. In mean, 

 B=10-941; ±-023. Not well determined. 



Silicon. — Pelouze, Dumas,and Schiel. Si = 28*195 ; ±*031. 

 This constant needs thorough experimental revision. 



Lithium. — Mellet, Diehl, Troost, and Stas. General mean, 

 Li=7*0073; ±*007. 



Rubidium. — Bunsen, Piccarcl, Godeffroy. Rb = 85*251 : 

 ±•0045. 



