W. A. Norton—Theories of Heat. 189 
remaining after their removal augmented by slow degrees from 
sam tO 5455 of the length. The atoms of the individual 
molecules of the iron must then have taken up a corresponding 
number of slightly varying positions of equilibrium, giving rise 
to new intensities in the forces of action of one molecule on 
another at a given distance. Now such a series of slightly dif- 
ering positions of equilibrium of the molecules, 1s eit irrec- 
oncilable with any other hypothesis than that of the capability 
of variations, by indefinitely small degrees, in the intensities of 
the atomic forces exerted at given distances. ‘ 
bserved cannot have place unless the constituen sr 
Molecules are capable of taking an indefinite number of slightly 
