214 CHEMISTRY AND CRYSTALLOGRAPHY. 



despite the marvellous results obtained by Pope among the 

 organic compounds, 1 am compelled to believe that the funda- 

 mentals of the theory must be modified to a certain extent, so 

 that our present ideas of valency shall not be too much violated, 

 nor our knowledge of atomic volumes set at naught. 



The suggestion which the writer wishes to put forward 

 is one which, like I'ope's, must shock the chemist of the older 

 generation ; but since the latter is gradually becoming accustomed 

 to the idea of the composite nature of the " elements," he will 

 not be altogether unprepared for this further attack on the last 

 generation's creeds. Briefly, my new suggestion is that the 

 chemical ' equivalents ' have in most cases a real and separate 

 existence, and that their atomic spheres (equivalent-spheres) are 

 approximately all of the same size. Thus the element carbon 

 is (in my opinion) a tetrahedral arrangement of four equal 

 spheres, each of atomic weight 3 and of vplume very slightlv 

 greater* than that of the hydrogen atom : oxygen is to be repre- 

 sented as a juxtaposition of two equal spheres of atomic weight 

 8 : silicon as a tetrahedral arrangement of four spheres each of 

 atomic weight 7.08: glucinum is to be taken as two spheres of 

 atomic weight 4.55 each, and so on. (On the other hand, certain 

 elements, such as nitrogen and boron, both very similar to carbon, 

 are probably heterogeneous assemblages of spheres of different 

 atomic weight, though of approximately the same sizef.) I 

 may point out that the fundamental directive power of valency 

 follows at once from the position of the spheres. 



This theory at once disposes of the silicon difficulty, since 

 the fundamental spheres now become .nearly equal : in the silica 

 molecule, we would have a central group of four spheres each 

 of weight 7, in the hollows of which are four spheres of weight 8 

 to represent the two atoms of oxygen. The aggregate would have 

 the required two crystal forms. 



Similarly, water becomes a tetrahedral arrangement of 2 H 

 and 20 (0 :r=8), wihich brings out an analogy to H^Fg. Oxygen 

 gas becomies a regular tetrahedron of o's. Caustic soda would 

 be a similar case, which would also have the interest of being a 

 return to the formulation of fifty years ago ; thus Nao.oH 

 recalls XaO -f- HO, and also the intermediate stage of NagO H^O, 

 still used by agricultural chemists and other backwoodsmen. It 

 should be noted, however, that my scheme doubles all the old 

 formulae, e.g., sodium oxide is Na^o^ in place of NaO. Similarly, 

 alumina would become (Alo)*', where Al ^ 9 and o ^ 8 if. 



* As a first approximation to the relative volumes of the equivalent- 

 spheres, I suggest the sixth or the eighth root of the atomic weight. 

 This explains the considerable difference between the H and CI spheres 

 and the smaller difference between those of CI and I. 



t An analogy between Li and NH4 can, for example, be made by 

 assuming that Li is ZH4, arranged as a square pyramid where Z is the 

 sphere of atomic weight 3. 



JThus valency becomes simply the number of equivalent-spheres of 

 the same sort joined together in a crystal-aggregate. 



