1884.] Stress and Strain on" the Properties of Matter. 109 



calculated from the proportions of their components. Thermal 

 capacity is, therefore, a physical property which is not likely to be 

 altered to any appreciable extent by small impurities, so that the 

 results obtained by different experimenters agree very closely with each 

 other. 



It has been proved* that if e be taken to denote " Young's 

 Modulus," and a the mean distance between the centres of two 

 adjacent molecules, e X a 7 is in the case of most metals approximately 

 a constant. From this it would follow that the law of force proved 

 by Maxwell in his experiments on the viscosity of gasesf to exist 

 between the molecules of a gas is approximately true for solids, 

 accordingly the force between any two adjacent molecules of a solid 

 is approximately as the fifth power of the distance between their 

 centres. Now if we denote the atomic mass by A, the density by A, 

 the thermal capacity per unit mass by G m , and the thermal capacity 

 per unit volume by C P , we have the following relations : 



C m X A=a constant ; 



C v = A X C m ; 

 e x 7 =a constant ; 



OC I 



From these relations we obtain 



e 



=a constant ; 



* Lnc. elf., p. 32. 

 t " Phil. Trans.," 1866, vol. 126, Part I. 



