233 
1918-19.] An Electron-Transference Hypothesis, etc. 
From expressions 2 - 87, 2’88, and 2*89 we see that the largest terms 
occurring in 2’82 are — and — jr?/3, the other terms being relative to 
i V 2“ 
this one of order 10“ 2 approximately. 
Hence we have 
2-90 
and therefore 
2-91 
P, 2 
*12 _ e n , e M 2 Q 
P«‘ lr^-.+ 
A4 
f ^ f ■ 
1 V 2J 2 
Pj + P 2 e 2 /Xi 6 2 />i 2 
Now 
hf 1 2 hf 
+ r9i+ E i 
_ 1 , / xp _ e /*2 y 2 
AM \ 2 f'Jh' 2 
and 
E. -£^1^1 = - of—. lcr 28 ') 
1 \2» J 
^i = 0(10- 24 ), 
hJi 
and so we may neglect (Ej — ^jr)fr relatively to 
' J v ' 1 1 h/i 
2-92 
and hence 
2*93 
••• p -- j p-‘ = ^ approx., 
Percentage contribution to the molecular refractive) _ 100 
index due to interaction of atoms J ^ \Vi 
Now h = number of electrons in the hydrogen atom when neutral, 
h> 2, 
since, if /i = l when one electron was transferred from the hydrogen 
atom, the on’y electrification left would be the positive electrification, 
and this would contribute a negative quantity to the atomic refractive 
index, and there is no evidence of the atomic refractivity of hydrogen 
being negative. 
Hence the percentage contribution to the molecular refractivity due to 
the interaction of the atoms is for h = 2 approximately 3 per cent. 
This investigation does not presume to give the accurate percentage 
contribution ; it merely gives an indication of the order of magnitude of the 
percentage, and with this proviso we may say that our result is in agree- 
ment with experimental evidence. 
(Issued separately January 8, 1920.) 
