140 
to be produced by drawing out a liquid film which I shall 
have to notice particularly below ? Easy enough experi- 
ments on the contact electricity of metals will answer this 
question. If the contact-difference diminishes as the tem- 
perature is raised, it will follow from the Second Law of 
Thermodynamics, by reasoning precisely corresponding with 
that which I applied to the liquid film in my letters to you 
of February 2 and February 3 , 1858 ,* that plates of the two 
metals kept in metallic communication and allowed to ap- 
proach one another will experience an elevation of tempera- 
ture. But if the contact difference increases with tempera- 
ture, the effect of mutual approach will be a lowering of 
temperature. On the former supposition, the diminution of 
intrinsic energy in quantities of zinc and copper, consequent 
on mutual approach with temperature kept constant, will 
be greater, and on the latter supposition less, than I have 
estimated above. Till the requisite experiments are made, 
farther speculation on this subject is profitless : but what- 
ever be the result, it cannot invalidate the conclusion that a 
stratum of 2-5-07000^-0-0 of a centimetre thick cannot contain 
in its thickness many, if so much as one, molecular con- 
stituent of the mass. 
Besides the two reasons for limiting the smallness of 
atoms or molecules which I have now stated, two others 
are afforded by the theory of capillary attraction, and 
Clausius’s and Maxwell’s magnificent working out of the 
Kinetic theory of gases. 
In my letters to you already referred to, I showed that 
the dynamic value of the heat required to prevent a bubble 
from cooling when stretched is rather more than half the 
work spent in stretching it. Hence if we calculate the 
work required to stretch it to any stated extent, and multi- 
ply the result by f, we have an estimate, near enough for my 
present purpose, of the augmentation of energy experienced 
Proceedings of the Koyal Society for April, 1858. 
