82 M. Acbille Cazin on Internal Work in Gases. 



internal work in a given change of state is evidently of very 

 great importance in the establishment of a law of molecular forces. 

 The mechanical theory of heat gives us the means of discovering 

 this sign in a great number of cases. 



It is generally admitted that, in a material system, positive 

 work caused by the use of heat only is correlative to the ap- 

 pearance of a number of thermal units proportional to this work, 

 and that negative work is correlative to the disappearance of 

 a like number of thermal units. By saying that work is con- 

 verted into heat in the first case, and heat into work in the 

 second, we express this correlation in a convenient manner for 

 drawing conclusions. Hence when a body experiences a change 

 such that some internal work has been effected and a quantity 

 of heat has disappeared, without revealing any other pheno- 

 menon, mechanical or thermic, this work must be negative. 

 This is what happens in the melting of ice when it is occasioned 

 by the disappearance of a certain quantity of heat coming from 

 external bodies, and when no pressure is exerted on the surface 

 of the ice. In the most general case, where heat disappears ex- 

 ternally to a body whilst some of it appears internally, and ex- 

 ternal and internal work are effected at the same time, we shall 

 be able to deduce from the measure of the first three quantities 

 the value and sign of the last. 



In gases we are accustomed to take into account only the ex- 

 ternal work, and can hardly conceive attractive molecular forces. 

 For a long time we have imagined these bodies as assemblages 

 of molecules mutually repelling each other. This was a way of 

 explaining their expansibility ; and the repulsion was attributed 

 to caloric when this agent was invoked in order to interpret 

 the effects of heat. The mechanical theory of heat rejects this 

 explanation, and leads to other hypotheses on the constitution 

 of gases. By supposing that the molecules of gases are en- 

 dowed with certain motions, we easily explain the expansibility 

 and the principal properties of this kind of bodies. Although 

 these hypotheses cannot express the truth so long as they are 

 devised with the object of explaining approximate laws, such as 

 those of Mariotte and Gay-Lussac, they are nevertheless impor- 

 tant ; for they disengage us from the old ideas, and certainly pre- 

 pare other, more complete hypotheses, which will be suggested 

 by a more profound knowledge of the phenomena. 



Thus it is especially to experiment that we must have recourse, 

 if we wish to possess exact notions on the work in gases. As every 

 thing leads to the belief that the molecular forces in these bodies 

 are simply directed according to the lines of the centres, it may 

 be supposed that they are subject to a more simple law than 

 those of solids and liquids, and it may be hoped that this law will 

 be more easily discovered. 



