THE TESTING OF THE LAW. 37 



alike in this: that they differ from all other elements in 

 having no combining power whatever. It has been found im- 

 possible to make them unite chemically with any other 

 substance in nature. They thus constituted a new and 

 separate family of elements for which there seemed to be 

 no place in the periodic law. Now the law undertakes to 

 find a place for any and every element in nature. The 

 validity of the law was threatened. The case was serious. 

 Let us see how neatly and completely the law was vin- 

 dicated. One of the most curious facts in connection with 

 the periodic law is the regular increase in the valence of 

 the Groups. By valence is meant the power of an atom 

 to unite chemically with other atoms, in other words " the 

 grabbing-power " of the atom. If we consider these val- 

 ences, for the present, as so many little arms to hold on 

 with, we find that the elements of the different groups, 

 from Group I to Group VIII, have 1, 2, 3, 4, 5, 6, 7 and 8 

 arms respectively. This variation in the " grabbing-power " 

 is well exemplified in the way in which the atoms in (let us 

 say) Series 3 unite with fluorine to form fluorides: 



1 atom of sodium unites with 1 atom of fluorine to form 1 molecule of fluoride 



magnesium 



aluminum 



silicon 



phosphorus 



sulphur 



while the element chlorine does not unite with fluorine at 

 all, owing apparently to the fact that they both belong to 

 the same family, are alike in their nature, and are, so to 

 speak, sister elements. 



Now these five new elements, helium, neon, argon, kryp- 

 ton and xenon , have no combining power whatever, no 

 arms at all, so to speak; hence they should constitute a 

 separate group coming before Group I, the members of which 



