744 REPORT—-1885. 
experience is for the atom of electricity to be smaller than any quantity 
hitherto measured. 
Whether the charge of a monad atom be indivisible or not, it is cer- 
tainly a natural unit of electricity, and it becomes of great interest to 
calculate its value. This is easily possible to the same degree of approxi- 
mation, as we know the size of an atom. The electrical charges in the 
atoms of a gramme of water are known accurately enough, from ordinary 
electro-chemical-equivalent determinations, viz. 1°5 x10!3 electrostatic 
units of each kind; the number of molecules in a gramme of water may 
be considered as something between 10*4 and 10”; and accordingly the 
charge of a monad atom is something like 10-'! or 10-! electrostatic 
units. 
Now this is very small, less than the hundred trillionth of a coulomb, 
and if it were really an ultimate atom of electricity, it is wholly unlikely 
that the fact would have been noticed. It is possible, however, to think 
of some phenomena which may afford an indication one way or the other, 
and I shall venture to suggest one or two later. 
The charge of an atom is so small that its potential cannot turn out 
high, on any customary hypothesis as to actual atomic magnitude, pro- 
vided ordinary considerations of electrostatics apply to atoms. It is 
difficult to know whether they apply or not until it can be shown 
that absolute vacuum has a specific inductive capacity. The transparency 
of interstellar space, and the velocity of radiation in it, would seem re- 
spectively to answer the question in the affirmative, and to suggest 
unity as its value. At any rate, when one has no other mode of tackling 
atomic charges it would seem reasonable to try the ordinary electrostatic 
laws on them, and see how they fit and what happens. 
Consider therefore the following problem. 
A number of equal spheres, each charged with a definite quantity of 
electricity, are commingled with the same number of similar spheres 
each charged with the same quantity of opposite electricity, the poten- 
tials of the spheres being so low that mutual discharge does not occur 
even during a collision, or so-called ‘contact,’ of the spheres, and some 
law of force being assumed between the spheres irrespective of their 
charges. 
I do not propose to attack the problem thus vaguely suggested, 
because many persons can do it far more easily and thoroughly than I; 
but certain facts are patent. The potential of each sphere must be lower 
in the ‘combined’ state than in the isolated, and, unless an atom be 
assumed to be extravagantly small compared with the spaces between 
them in the liquid state, the potential of each isolated atom is but a few 
volts. 
Facts are known which suggest sizes for the actual substance of the 
molecules, but, without pressing them, one may assume that whereas in 
the liquid state the distance of the atoms apart is about 10-8, the radius 
of each of them is about 10-9 or even 10—! centimetres; then, the 
charge of a monad atom being as aforesaid 10-1! or 10~!%, it follows that 
its potential, when isolated, is about 3 volts. 
If such an atom pair off with another of opposite sign, the potential of 
each will fall as they approach ; becoming, when the distance between 
their nearest points is one tenth of the radius of either, about 1-2 volt,} 
* See table by Sir William Thomson in Llectrostatics and Magnetism, § 142. 
