TRANSACTIONS OF THE SECTIONS. 



59 



1 atom water is greater than the mean by about ^^^th. Hence it is im- 

 possible that the heat evolved can be a consequence of a diminution, 

 -when no such diminution exists. In all the other compounds there is 

 a diminution of the specific heat, but it does not correspond with 

 the heat evolved. The greatest takes place when one atom of oil of 

 vitriol is mixed with three atoms water. It amounts, in that case, to 

 about ^th, and the heat evolved is 208°. But when one atom of oil 

 of vitriol is mixed with two atoms of water, the heat evolved is 219° ; 

 yet the diminution of specific heat is only about -^, and consequently 

 less than when the heat evolved is only 208°. The same want of coin- 

 cidence exists in every part of the table. Hence it follows, that when 

 oil of vitriol and water are mixed the heat evolved is not the conse- 

 quence of a diminution of the specific heat. 



Dulong and Petit observed that when the atomic weight of a simple 

 body is multiplied by its specific heat the product is a constant quantity. 

 Dr. Thompson has shown in a paper published in the third volume of the 

 Records of General Science, that this constant quantity is 0-375. It fol- 

 lows from this law, that the same absolute quantity of heat exists in com- 

 bination with every simple atom ; that the differences of the specific 

 heats of different simple bodies are owing to a difference in their atomic 

 weights. 



In the same paper it is shown that when the atomic weight of a 

 compound body is multiplied by its specific heat, the product is always 

 a multiple of 0*375 by a whole number, which number depends upon, 

 or at least is connected with the number of atoms of which the com- 

 pound body is composed. If the number multiplying 0-375 be equal 

 to the number of atoms in the compound body, then it follows that 

 every atom of the compound body retains all the heat with which it 

 was combined when in an isolated state. If the multiple be less 

 than the number of atoms, then the compound contains less heat than ex- 

 isted in its elements, and the difference between the multiple and the 

 number of atoms gives us the proportion of heat wanting. 



Let us apply this method to the combinations of oil of vitriol and 

 water. The following table exhibits the result. 



