ON ELECTROLYSIS. 745 
and falling asymptotically to zero as the distance between them still 
further diminishes.' 
It is true that this fall of potential has not the result of postponing 
mutual discharge ad infinitum, however near atoms may approach; but I 
feel constrained to believe that no such discharge ever in practice occurs, 
from the fact (if it be a fact) that the whole of a liquid can be electro- 
lysed away, and from the assumption that any such discharged molecules 
would be wholly intractable to electrical influence. To explain non- 
discharge we can fall back on the rather dark fact of the enormous 
apparent dielectric strength of vacuum, combined with the very low 
potentials to which atoms are charged. 
I have tried an experiment of electrifying two falling clouds of lyco- 
podium oppositely, and allowing them to mix, or, so to speak, combine. 
The result has not yet been satisfactory, but lyeopodium granules are 
coarse and weighty bodies, and are unwieldy representatives of atoms. 
To get anything like such a quantity of each kind of electricity into the 
grains of a cubic millimetre of lycopodium as exists in the atoms of a 
cubic millimetre of water, the charge, and therefore the potential, of each 
granule would have to be enormous, something like a billion volts: for I 
find the diameter of a lycopodium granule about ‘004 centimetre. It is 
easy to see that a given quantity of both kinds of electricity, shared 
among a given bulk of equal and definitely arranged spheres, will raise 
the potential of each by an amount proportional to its superficies. 
It may be noted, as interestingly showing the enormous charges which, 
distributed among the atoms of a substance, produce such an insignifi- 
cant potential, i.e., as illustrating the prodigious electrical capacity of 
molecular arrangement, that if the opposite electricities were extracted 
from a milligramme of water, and given to two spheres one mile apart, 
those two spheres would attract each other with a force of ten tons!” (Cf. 
Helmholtz, Faraday Lecture, 1881.) 
A more hopeful substance than lycopodium, for constructing an 
artificial chemical compound with, is fine smoke, say of magnesia, which 
hovers in air for a long time. 
But, as Clark and I have observed, it is sufficient to electrify such 
1 Or one might work the argument conversely, and perhaps more plausibly, so as 
to obtain an estimate of atomic, as distinct from molecular, dimensions. Thus :— 
The total charge shared among the atoms of a gramme of water is perfectly well 
known, viz., 1073 electro-magnetic units of each kind. The heat produced during 
the formation of a gramme of liquid water from its gaseous constituents is, according 
to J. Thomsen, 4333 thermal units. If two bodies, or sets of bodies, each charged 
with that quantity of electricity (one set positive, the other set negative), approach 
each other so as to do this amount of work, their difference of potential must 
diminish, during the approach, by 1:7 volt. 
The fall may be from 1-7 to nearly 0, as the dissociation idea of electrolysis would 
suggést. Or it may be from 3 to 1-3, as the guess at atomic dimensions in the above 
text implies. Or atomic charge may be a thing acquired during the act of com- 
bination, as the hypothesis of zero charge in a molecular aggregate like HH or 00 
would imply ; in which case the necessary fall of potential must be 3-4 volts. 
The first of these alternatives gives for the diameter of an oxygen atom 
56 x 10% 23 ; 
1/z3 being the number of molecules in a gramme of water. 
2 Tf any cause could make the positive atoms in a drop of water group together 
and face the negative across a vacuum, a furious Leyden-jar, or ‘Globe-Lightning,’ 
would be produced. 
