52 PRESIDENTIAL ADDRESS SECTION B. 



the distribution in the concentric shells as 2, 8, 8, 18, 18, 32, 32. 

 Langmuir gives this a physical basis by noting that areas of con- 

 centric spheres of radius 1, 2, 3, 4, 5 vary in the same proportion 

 as the numbers 2, 8, 18, 32, 50. He therefore assumes that these 

 higher atoms have spheres of radius 1, 2, 3, and 4, and that there 

 are two electrons in each unit of area in every sphere except the 

 innermost, thus giving the numbers 2, 8, 8, 18, 18, 32, 32, it being 

 assumed that each area-unit is filled twice over with a single 

 electron giving thus 2, 2 + 8, 2 + 8 + 8, 2 + 8 + 8+18, and so on, 

 these being the numbers 2, 10, 18, 36, 54, etc., which are the total 

 number of electrons present. This theory is so beautiful and com- 

 prehensive that it must be essentially true : at any rate it is likely 

 to supplant the older theory that the electrons are in revolution 

 round the nucleus like planets round a sun. Bohr has, however, 

 recently suggested that 32 is succeeded by 18 and 8, not by 50. 



The same theory, it may be noted, applies to the halogens 

 fluorine, chlorine, bromine, and iodine. Fluoride ion F' is 

 5#(10H+10# 9H+)50 which is an "inert" structure like that of 

 sodium ion and that of neon ( = 50(1OH+1O0 lOH+)50. So the 

 chloride ion is 9#(18H + 18i9 17H+)90 when its atomic weight is 

 35. When the atomic weight is 37* the constitution is 

 9(9(20+200 17H+)9#. Similarly argon of the atomic weight 40 

 (i.e., out of order in the Periodic table) is 90(2211+220 18H+)90. 

 Bromine ionf is 18#(45H+450 35H+)180, krypton being 

 180(47H+470 36H+)18fl. Iodine ion is 270(74H+740 53H+)27#, 

 xenon being 27fl(76H+760 54H+)270. 



The theory is, however, rather unsatisfactory for carbon and 

 all the elements below it, Carbon would be #(6H+60 6H+)0 with 

 a residual valency of plus 4, but, as I say later on, it is likely that 

 the nucleus itself is arranged like a tetrahedron (as the valencies 

 are) whereas the above is unsymmetrical. Similarly helium gas 

 would be 0(2H+2fl 2H+)#, which is very unsatisfactory from its 

 lack of symmetry. Helium gas has in reality a round molecule, 

 as shown by its Cp/Cv ratio. (See Fig. 5.) 



Again if Rydberg's formula be examined, surprise will be felt 

 that it does not read N = 2 (l 2 + l 2 +2 2 +2 2 +3 2 +3 2 +4 2 , etc.) in 

 which case it would be perfectly symmetrical. Assuming then this 

 form of the equation, the successive sums are 2, 4, 12, 20, 38, 56, 88, 

 these being therefore the number of electrons. Helium would be 

 second in this series and have a shell of 4 electrons instead of 2. 

 As it is neutral, one would have to assume a nucleus of 4H+, with 

 the shell of 4 electrons far out and in tetrahedral order. The 

 nuclear charge, the number of external electrons, and the atomic 

 number would all be the same, viz., 4, if two elements exist 

 between hydrogen and helium. This very neat and satisfactory 

 proposition is, however, knocked on the head by the existence of 

 the «-particle of radioactivity which has mass 4 and two positive 

 charges. It changes into helium on striking any kind of matter 



* See Aston, Phil. Mag. 1920, 611 and my prediction, '-Modern Alchemy and Transmutation' 



J.S.A. Association Anal. Chemists, 18/9/17. 

 + Since this was written, bromine has been fouud to be 79 plus 81, not 80. 



