EXPERIMENTAL KNOWLEDGE OF THE PROPEKTIES OF MATTER. 479 



elements among those to whicli the law of atomic beats applies, but witb- 

 out success. 



It is noticeable that Regnanlt, finding so large a number of elements 

 conform to the law of specific beats, is at last disinclined to admit any 

 exception for elements in the solid state, and is only deterred from 

 doubling the atomic weight of carbon with the object of including this 

 element by finding how widely the values of the specific heats ' differ 

 from one another for carbon from different sources. 



The constancy of specific heat in spite of allotropic variations of the 

 body examined, as mentioned above, is disputed by Bettendorf and 

 Wiillner ^ for selenion and for arsenicum ; but for the former one may 

 prefer Regnault's result, and the latter is perhaps open to criticism. A 

 marked exception is, no doubt, the case of diamond, the specific heat of 

 which is much less— between 100° and 0°— than that of other varieties 

 of carbon. The highest temperature in Regnault's experiments on the 

 specific heats of solid bodies was 100° or, in a few cases, a few degrees 

 higher ; generally it was lower than 100°, and much lower in cases where 

 the fusing-point of the body was low. 



Wmstyn's Law of Molecular Heats. 



In 1848 "Woestyn^ enunciated the following law as to the specific heats 

 of compounds: ' If C is the specific heat and A the formula-weight of a 

 compound of certain elements, the atomic weights of which are a],a.^, . 

 and the specific heats c,, c,, and if to,, «,, ... be the number of atoms of 

 each element in the formula of the compound, then AC = n^a^c^ + n^a^c.2 



It is taken as a fact by Woestyn that different elements have different 

 atomicheats— that a,c,,a2a2, • • . are not all equal ; and, therefore, as an 

 accident if the molecular beats of compounds of similar atomic com- 

 position are equal or approximately so. Thus it follows from his principle 

 that lead sulphate and calcium sulphate should have different molecular 

 beats, the difference being that between the atomic heats of lead and cal- 

 cium ; it also follows from his principle that there should be the same 

 difference between the molecular heats of the carbonates, of the prot- 

 oxides, of the nitrates, chlorides, bromides, iodides, chlorates ... of this 

 pair of metals ; and similarly for any other pair of metals which form 

 salts and oxides of similar atomic composition. 



_ By taking the differences between the known molecular beats of an 

 oxide of lead and corresponding sulphide, and so for mercury and for bis- 

 muth, and antimony for the same amount of combined oxygen, be finds 

 numbers which according to his law should be equal, being the difference 

 in each case between the atomic heats of sulphur and oxygen, and which 

 are, in fact, nearly equal. The law, tested by a considerable number of 

 compounds, is found to give results very fairly in accordance with it. 



Extension of Dulong and Petit's Law to Compounds. 



Although Neumann, and after him Regnault, showed most interest- 

 ing relations among molecular beats of compounds, from which Regnault 



' Aim. Chim. (3), 1, 1841, p. 204. 



^ Pogg. 133, 1868, p. 293 ; and Aim, Chim. (4), 14, 1868, p, 476, 



' Ann. Chim. (3), 23, 1848, p. 296. 



