90 
PROFESSOR KOPP ON THE SPECIFIC HEAT OF SOLID BODIES. 
liquid and that of the glass are concentrated on the remainder, on the thermal action of 
the solid substance from which its specific heat is to be deduced. The results obtained 
by my method are less accurate when the residue is only a small fraction of the total 
result from which it is deduced. In individual cases, where this was unavoidable, I 
shall have to remark upon it. 
It may be said in favour of my method that, for a number of solid substances, no 
other method yet attempted is applicable either at all or with more prospect of a suc- 
cessful result. But this is less important than the proof furnished by my examination 
of very many substances, whose specific heat has been already determined by Neumann 
and by Regnault, that the specific heat of bodies may be determined by my method 
with an accuracy quite sufficient for many comparisons. But there are cases in which 
it is even advantageous not to heat the solid alone, but in conjunction with a liquid, and 
to bring them together into the water of the calorimeter. The chemical nature of the 
solid may necessitate this ; as, for example, when it readily alters on being heated in the 
air (compare § 34 in reference to amorphous boron) ; its physical structure may also 
render it desirable, as for instance if the substance has a large surface as compared with 
its mass, or is so porous that the thermal action due to humectation, and first observed 
by Pouillet *, takes place. Regnault has shown that this may be considerable f ; he 
states that for this reason the specific heat of some substances is found about too 
great. He appears to have estimated this thermal action by ascertaining the increase 
of temperature produced in the water of the calorimeter when the porous substance, 
whose temperature is that of the water and of the surrounding air, is dipped in it. But 
this action is probably far more considerable if, while heated, it is immersed in the 
water, because it then contains less air confined on its surface and in its pores J, and 
surface action can then act more intensely upon the liquid. The influence of this 
source of error cannot be measured exactly. It is unequal in different substances. In 
platinum it is small (Regnault found by his method that the specific heat of spongy 
platinum did not materially differ from that of massive pieces), while it may be con- 
* Ann. de Chim. et de Phys. [2] vol. xx. p. 141. 
t Ann. de Chim. et de Phys. [3] vol. i. p. 133. Regnault preferred to immerse the heated porous sub- 
stances, when they could be obtained in coherent pieces, directly in the water of the calorimeter. If they were 
enclosed in thin tubes and immersed, the equalization of temperature proceeded too slowly. Regnault abstained 
from enclosing at the same time a sufficient quantity of water in the tube to promote the circulation, because 
in that case the thermal action of the solid was only a fraction of that of the water added, on which the entire 
source of error falls. Regnault found also (ibid. p. 142) that in immersing anhydrous baryta, strontia, and lime 
in most carefully dehydrated oil of turpentine, there is such a thermal action that no useful result is to be obtained 
by his method for these oxides. 
i. To the examples already known, which show what influence temperature exerts on the quantity of air 
absorbed in a porous body, Regnault has added a very instructive one (Ann. de Chim. et de Phys. [3] vol. 
lxiii. p. 32). If amorphous boron, formed into disks by pressure in a steel mortar, was strongly cooled and then 
immersed in the water of the calorimeter (at the mean temperature), so considerable a disengagement of 
absorbed air was produced, that Regnault was compelled to give up the determination of the specific heat 
by this method. 
