38 Remarks on the Atomic Constitution of Elastic Fluids. 



with the ultimate chemical atoms. We have already had oc- 

 casion to refer to the postulates employed by Laplace as the 

 basis of his profound mathematical inquiries into the consti- 

 tution of elastic fluids. Now if heat be combined with the 

 particles of matter by a principle of attraction or affinity (as 

 supposed by Laplace), it is impossible to conceive such affinity 

 to be exercised by aggregates of atoms, and yet not to be the 

 attribute of the single atoms, of which such aggregates are 

 composed. And if the ultimate atoms be endowed with an 

 affinity for caloric, no reason can be assigned why union 

 should not take place between caloric and each ultimate atom 

 singly; nor, the molecules of heat being self-repulsive, why 

 the ultimate atoms after such union should not become mu- 

 tually repulsive. The contrary hypothesis of Dr. Prout in- 

 volves the anomaly of supposing heat to have a combining 

 affinity for two or more atoms, while it is destitute of such af- 

 finity for single atoms ; and also that of supposing two atoms 

 to have relations towards two atoms, or three towards three, 

 which do not obtain between single atoms. 



It is furthermore apparent, that the question respecting 

 the mode of union between heat and the molecules of bodies, 

 is not limited to the constitution of elastic fluids, but must 

 equally comprehend the conditions of liquid and of solid. 

 Now the relations of several simple bodies to heat, established 

 by the experiments of Dulong and Petit, point unequivocally 

 to the chemical atoms, as determining the measure of specific 

 heat. In the thirteen simple substances which were the sub- 

 jects of experiment, they found that the product of the spe- 

 cific heats into the atomic weights was invariably a constant 

 quantity, and consequently that the ultimate atoms contained 

 precisely the same quantity of caloric. These results have 

 been since confirmed by various German experimenters*, and 

 can only be reconciled with the doctrine of the combination 

 of heat with the ultimate chemical atom. 



In recapitulation, it has been shown, 



1st, That the law of Mariotte, and that of the equal expan- 

 sibility of the different gases, are mathematically derived from 

 elements altogether foreign to the numerical relations of their 

 ultimate molecules ; and that no corollary is contained in those 

 laws, determining equality of atoms in a given volume of the 

 different elastic fluids. 



* See Mr. Johnston's excellent and comprehensive Report on Chemistry, 

 in the Rejiort of the First and Second Meetings of the British Association for 

 the Advancement of Science, p. 418. 



