168 ESSAYS BIOGRAPHICAL AND CHEMICAL 



Many attempts have been made to devise some mathe- 

 matical relation between these atomic weights. So long 

 as there was reason to doubt the accuracy of the experi- 

 ments by means of which the atomic weights have been 

 determined, some such relation as the following had 

 considerable probability in its favour : Taking the 

 differences between the atomic weights of the elements 

 in the first column, lithium, sodium, potassium, rubidium, 

 and caesium, they are 



K-Na = 39-23 = 16; 



Rb-K = 85-39 = 46 = (3xl6) nearly; 



Cs-Rb = 133-85 = 48 = (3xl6). 



The differences are 16, 16, 3x16, and 3x16. Now 

 there are compounds of carbon and hydrogen which 

 possess the formulae, CH 4 , C 2 H 6 , C 3 H 8 , C 4 H 10 , C 5 H 12 , C 6 H 14 , 

 etc. ; and as the atomic weight of carbon is 12, and that of 

 hydrogen 1, the sum of the atomic weights, or, as they are 

 called, the molecular weights, are respectively 16, 30, 44, 

 58, 72, 86, etc., with a common difference of 14. We see, 

 therefore, that a set of compounds may so differ in mole- 

 cular weight as to present a regular series, with a common 

 difference. Nothing was more likely, then, than that 

 sodium should be regarded as a compound of one atom of 

 lithium with one atom of an unknown element of atomic 

 weight 16, or with two atoms of an unknown element of 

 atomic weight 8 ; while potassium might be looked upon 

 as a compound of an atom of lithium, with four atoms of 

 the element of atomic weight 8 ; and so on. But, unfor- 

 tunately for this simple theory, the differences between 

 the atomic weights of the elements are not exactly equal. 

 Instead of 16, the real difference between the atomic 

 weights of lithium and sodium is 16 '02; between potassium 



