Changes in Length in Iron Wires by Magnetization. 345 



The question then arises, whether this last relation is direct, 

 or arises from some material property changing with the 

 temperature ? Is it possible that the velocity of light may 

 vary with temperature, apart from any relation to material 

 substances, or is the foregoing relationship dependent upon 

 some molecular function like that suggested, that is " mole- 

 cular domain ? " 



It seems impossible to answer the first part of this question. 

 The second part can be judged by analogy, which would 

 appear to suggest that a molecular function is more probable. 

 Thus, one might expect that the kinetic energy of the particles 

 is the property directly increased by rise of temperature, and 

 that many or all other properties of the molecule are conse- 

 quential upon that. The apparent volume of course increases. 

 That fraction of the volume which is measured by 



fJU 2 — 1 fl—1 



j?**"' or ~T ( "-^ 



also increases, whatever it may be called. This constant fact 

 appears to be a decided argument in favour of these ex- 

 pressions, giving " domain " and not " volume of matter " 

 simply. But again, if the plea for " domain " be granted, 

 there ought to be increase, not only when the liquid is heated, 

 but also when it is changed into a vapour. This is actually 



the case for — (v—{3), the value of which increases in the 



ratio 1 to 1*4 or 1*5. This variability may be compared with 

 that found by van der Waals for his constant b " (" Conti- 

 nuity &c.'% Phys. Soc. translation, p. 409). 



XXIX. On the Changes in Length produced in Iron Wires 

 by Magnetization. By Louis Trenchard More, Ph.D., 

 Johns Hopkins University*. 



THIS investigation was undertaken at the suggestion of 

 Professor Rowland, and has for its object the finding 

 of a relation between the change of length produced in iron 

 wires by magnetization and the intensity of magnetization 

 existing in the wire. It was hoped thus to obtain results 

 that would be comparable and to avoid certain errors common 

 to all previous work. 



* Communicated by the Author. 



