1819.] Petit «nd Dulorig on the Tlieory of Heat. 189 



the selvages of the bunting being always placed in the first posi- 

 tion, which causes them to blow out without bagging. 



It would be a two-fold improvement, if, in running large, the 

 sheet of square sails was eased off; namely, first, the better division 

 of the stagnant air in front ; and secondly, the sails not being so 

 violently stretched, they would stand better on a wind. 



Article IV. 



Researches on some important Points of the Theory of Heat* 

 By MM. Petit and Dulong. (Presented to the Academy of 

 Sciences on April 12, 1819.) 



The considerations founded on the laws relative to the propor- 

 tions of chemical compounds enable us to form respecting the 

 constitution of bodies, ideas which, though arbitrarily established 

 in several points, cannot, however, be regarded as absolutely 

 vague and sterile. Convinced likewise that certain properties of 

 matter would present themselves under more simple forms, and 

 could be expressed by more regular and less complicated laws, 

 if we could refer them to the elements upon which they imme- 

 diately depend, we have endeavoured to introduce into the study 

 of some of the properties which appear more intimately connected 

 with the individual action of the material molecules, the most 

 certain results of the atomic theory. The success which we 

 have already met with makes us hope not only that this kind of 

 consideration may contribute materially to the further progress of 

 physics ; but that the atomic theory itself will receive from it a 

 new degree of probability, and will from it derive sure methods 

 of determining the truth among different hypotheses all equally 

 probable. 



Among the properties of matter to which the considerations 

 just mentioned are applicable, we shall choose, in the first 

 place, as having more particularly fixed our attention, those 

 which depend upon the action of heat. By directing our obser- 

 vations in a suitable manner, we have been led to discover 

 simple relations between phenomena, the connexion of which 

 had not been previously attended to ; but the numerous points 

 of view under which these phenomena may be examined, giving 

 to our researches an extent which does not permit us to embrace 

 the whole at one time, we have thought that it might be useful 

 at present to make known the results to which we have come. 



These first results relate to specific heats. The determination 

 of this important element has been, as is well known, the object 



* Translated from the Annates de Ctiimie et Physique, x. 3W. 



3 



