Art. X. — On Entropy Meters. 

 By E. F. J. Love, M.A., F.R.A.S., 



Demonstrator and Assistant Lecturer in Natural Philosophy in the 

 University of Melbourne. 



(With Diagram). 



[Read 1st July, 1897.] 



(1) Introductory. 



The physical quantity, termed "the Thermodynamic Function" 

 by Rankine, and the " Entropy " by Clausius — a name which has 

 been finally adopted for it by physicists — is first encountered by 

 the student when he is a little way advanced in the study of the 

 dynamical theory of heat, and he generally finds it a stumbling- 

 block. This is scarcely to be wondered at, seeing that no less a 

 genius than Clerk Maxwell succeeded in giving three different 

 — and totally inconsistent — definitions^ of this conception in 

 successive editions of his "Theory of Heat." Yet it is definite 

 enough ; entropy is as definite a proj^erty of any portion of 

 matter as is its volume, or temperature, or pressure. When a 

 body receives heat from external sources its entropy is increased, 

 when it gives up heat its entropy is diminished ; the change in 

 the entropy being measured by the ratio which the transferred 

 heat bears to the absolute temperature at which the transfer 

 takes place. 



Special interest attaches to entropy, inasmuch as it is the only 

 physical property which always increases when heat is supplied, 

 and diminishes when heat is abstracted. A body may receive 

 heat- — e.g. during fusion or vaporisation- — -without rising in 

 temperature ; cases are even known — e.g. saturated steam — such 

 that the entrance of heat is accompanied l)y a fall in temperature. 

 Similarly we do not always find tlie volume of a body increased 

 by the entrance of heat ; silver iodide contracts when its 

 temperature rises, so does ice during fusion, thougli botli are 

 absorbing heat. But no considerations of this kind come in 

 with regard to entropy ; if heat enter tlie body its entropy rises, 



1 Two of tliese, moreover, were hopelessly wroiijf ; the third was correct. 



