THE 

 LONDON, EDINBURQH, and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE. 



[FIFTH SERIES.] 



NOVEMBER 1884. 



XLIII. On the Numerics of the Elements. 

 By Edmund J. Mills, D.Sc, F.R.S.* 



I EMPLOY Professor James Thomson's expression "nu- 

 meric " as an abbreviation of " numerical value." The 

 numeric of an element is the numerical value ascribed to its 

 symbol. 



Ever since Dobereiner's timef the relations between the 

 numerics of the elements have been felt to be a fascinating 

 study; but it was not until the discovery by NewlandsJ of 

 the periodic function, that a closer and really comprehensive 

 grasp of the subject was rendered possible. The theory of 

 the periodicity of compounds was subsequently developed by 

 myself §, under the name of " Cumulative Resolution. M 



It has of late years been admitted that it would be desirable 

 to place the periodicity of the elements on a more definite 

 numerical basis, and to include in it, if possible, a much 

 closer approximation to actual numerics than it has hitherto 

 contained. Accordingly, in 1879, I drew up a list of mean 

 values of the numerics of the elements ; there was but little 

 difficulty in finding that these constituted a succession of 

 geometric series. At this point the work was laid aside for 

 further consideration. The appearance, in 1882, of Clarke's 

 able and exhaustive summary of all the evidence relating to 



* Communicated by the Author. 



t Pogg. Ann. xci. p. 301 (1829). For this reference I am indebted 

 to Professor Ferguson. 



t For an epitome of Newlands's work, see ' The Periodic Law ' (London : 

 Spon, 1884). 



§ Phil. Mag. [5] iii. p. 498. For the periodicity of the infinite olefines 

 of boiling-point and melting-point, see Phil. Mag. [5] xvii. p. 173. 



Phil Mag. S. 5. Vol. 18. No. 114. Nov. 1884. 2 D 



