TJieory of Capillarity. 343 
that he had not considered the "repulsive force of heat 
within the liquid " — a phrase of which the nearest modern 
equivalent would be " the repulsive or elastic reaction between 
the molecules considered with reference to temperature;"" and 
indeed the passage which I have quoted is, I believe, the only 
one in which Laplace takes any elastic reaction into consi- 
deration. 
In a short paper published in the Bulletin de la Society Phi- 
lomathique de Paris, 1819-1822 (referred to by Poisson), after 
pointing out that the chemical attraction is very great while 
the capillary effect is very small, he says: — " La theorie que 
j'ai donnee de ces phenomenes embrasse Paction des deux 
forces dont je viens de parler en prenant pour fexpression 
integrale de Feffet capillaire, la difference des deux integrates 
relatives a l'attraction moleculaire et a la force repulsive de 
la chaleur ce qui repond a l'objection du savant physicien 
M. Young." Poisson, however, points out that this answer 
is quite insufficient; and in his own theory, by examining 
carefully into the nature of the internal equilibrium in which 
the "repulsive force of heat " is involved, concludes (1) that 
there must be a very rapid increase of density close to the 
surface as we descend into the liquid, and (2) that but for 
this great variation of density none of the phenomena of 
capillarity would occur. Clerk Maxwell, in his article in the 
Encyclopaedia Britannica on this subject, speaks of the latter 
conclusion as mathematically wrong, by which I think he 
means that in liquids, for example, much less compressible 
than any we know of, there would still be a surface-tension, 
though the variation of density near the surface was vanishingly 
small. 
Laplace does indeed point out, on p. 494 in the General Con- 
siderations appended to the Second Supplement, that the com- 
pression of the liquid is zero at the surface, and increases with 
extreme rapidity, reaching a constant value before a sensible 
depth below the surface is attained. But he is content with 
this cursory observation. Had he attended sufficiently to this 
point, he would probably not have spoken as he does in the last 
paragraph of his treatise of the idea of a surface-tension, where 
he endorses the view of Segner: — " Qu'elle n'etait qu J une 
fiction propre a representer les phenomenes, mais que Ton ne 
devait pas admettre qu'autant qu'elle se rattachait a la loi 
d'une attraction insensible a des distances sensibles." 
These words of Laplace have, I believe, contributed greatly 
to hinder the acceptance of the idea of a surface-tension. That 
he should have held such a view is, however, rot surprising; 
for it can be shown that it is in virtue of this very extension of 
