PEOFESSOK KOPP ON THE SPECIFIC HEAT OF SOLID BODIES. 
83 
be held as a reason for explaining why the determinations of the specific heats of solid 
elements do not even approximately obey Dulong and Petit’s law, nor those of solid com- 
pounds of analogous chemical constitution Neumann’s law. Neither law is universally 
valid, although I have found that Neumann’s law applies in the case of many compounds 
of analogous atomic composition, to which, on account of their totally different chemical 
deportment, different formulas are assigned ; and even in cases in which these laws have 
hitherto been considered as essentially true, the divergences from them are material. 
Each element has the same specific heat in its solid free state and in its solid com- 
pounds. From the specific heats to be assigned to the elements, either directly from 
experimental determination, or indirectly by calculation on the basis of the law just 
stated, the specific heats of their compounds may be calculated. I show the applicability 
of this by a great number of examples. 
In reference to this calculation of the specific heats of solid bodies I may here make 
a remark. The agreement between the results of calculation and experiment is often 
only approximate ; it is then natural to urge that the two ought really to agree more 
closely. To that the question may be allowed : What means are there of even approxi- 
mately predicting and calculating beforehand the specific heat of any inorganic or 
organic solid compound when nothing but its empirical formula is given 1 to which 
among the numbers OT, 0'2, 0 - 3 may it come nearest! The cases in which 
differences exist between calculation and observation, enumerated in § 103 to 110, may 
be set against this uncertainty. 
My proof of the propositions given above is based on determinations made by earlier 
inquirers, and on a not inconsiderable number of my own. I first describe the method 
by which I worked, and then give the results which I have obtained by its means. 
PAET II. — DESCRIPTION OF A METHOD OF DETERMINING THE SPECIFIC HEAT OF SOLID 
BODIES. 
16. I have worked by the method of mixture. It is not necessary for me to discuss 
the advantages which this method has over that of the ice-calorimeter, at any rate in 
requiring smaller quantities ; nor, as compared with the method of cooling, need I dis- 
cuss the uncertainties and differences in the results for the same substance, which are 
incidental to the use of this method, and which Regnault has detailed*. 
The method of mixtures has been raised by Neumann and by Regnault to a high 
degree of perfection. Although by Neumann’s method it is possible to determine more 
accurately the temperature to which the body investigated is heated, Regnault’s method 
allows larger quantities to be used. Regnault’s process gives the specific heats of 
such substances as can be investigated by it as accurately as can at all be expected in 
the determination of this property. In the case of copper and steel, it is not merely 
possible to determine their specific heats by its means, but also to say whether and how 
* Ann. de Chim. et de Phys. [2] vol. Ixxiii. p. 14 ; [3] vol. ix. p. 327. 
