88 
PROFESSOR KOPP OjST THE SPECIFIC HEAT OF SOLID BODIES. 
The apparatus which I have just described is very simple. It is readily constructed; 
the chief point is to have two thermometers which have been compared with each other, 
one of them (f) graduated in tenths of a degree, while on the other ( d ) the tenth of a 
degree can be observed with certainty. The apparatus does not require much space ; 
yet, while the experiment is being made, rapid changes in the temperature of the sur- 
rounding air must be avoided. One observer only is required (all the experiments 
described in the sequel have been made without assistance). The experiments which I 
shall communicate prove that, by means of this apparatus, the specific heat of solid 
substances, even when only small quantities are taken (in most cases I worked with 
only a few grammes), may be determined with an accuracy not much less than that 
attained with larger quantities in more complicated processes. 
19. Yet, it is true, the accuracy of the results obtained by this process appears to be 
inferior to that attainable by the use of Neumann’s or of Regnault’s methods. I have 
investigated many substances, determinations of which have also been made by these 
physicists. I do not find that the numbers I have obtained deviate in one special direc- 
tion from those which these physicists have found, which moreover sometimes differ 
considerably among themselves * ; but that the certainty of the results I have obtained is 
less, is shown by the fact that the results of different experiments with the same substance 
agree less closely with one another than do those 'of Regnault and of Neumann. 
That my determinations are less accurate is probably least due to the circumstance 
that I did not use certain corrections, for instance, that I did not allow for the loss of 
heat in the calorimeter between the time when the heated body was immersed and the 
maximum temperature was attained f. I have endeavoured to diminish the uncer- 
tainty of the results from this source by having the temperature of the water in the calori- 
meter, before immersing the heated body, somewhat lower than that of the surrounding 
air. I have endeavoured to ensure comparability in my results for different substances 
by always operating as much as possible under the same circumstances ; that is, I 
endeavoured always to produce in the water of the calorimeter the same excess of 
temperature over that of the surrounding air. Without depreciating the interest and 
value of such corrections, I think that their application may be omitted if them practical 
importance is inconsiderable, and the increased difficulties which they necessitate pro- 
portionally large. It must be considered, in reference to such corrections, how far 
the accuracy, which the results obtained by their means claim, is not more apparent than 
real $. And further, that these corrections, where the conditions for their application 
really exist, are not considerable ; while, where they exert a considerable influence on 
the result, they may be uncertain, because the suppositions made in their development 
* Pape, in Poggendokee’s ‘Annalen,’ vol. cxx. p. 579, discusses the probable causes of these differences. 
f Another correction, which appears to me to he more important for the experiments in question, is, that the 
contents of the glass at the time at which the temperature of the water is at its maximum may he at a some- 
what higher temperature. This I have approximately taken into account. Compare §§ 23 & 24. 
X It is unnecessary to adduce examples where such corrections, proceeding from as comprehensive a basis as 
