1830] 
and the Communication of Heat. 
341 
infinite (their capacity being supposed constant) — becomes on the contrary more 
probable, now that we know that the specific heat diminishes with the tempera- 
ture: for the law of the diminution may he such, that the integral of the quan- 
tities of heat, reckoned down to a temperature infinitely low, may yet have a finite 
value. 
The law of cooling, such as we have just presented, and such as can be observed 
in vacuo , has reference to rates of cooling which are estimated from the depres- 
sions of the air thermometer. We may see from the comparisons of the other 
thermometric scales, which have been given in the preceding part of this enquiry, 
that in using any other sort of thermometer, the relations which we have discover- 
ed between the temperatures and the rates of cooling, would lose that character of 
simplicity and generality, which is ordinarily found to attach to the laws of nature. 
If the capacity of bodies for heat were constant, as measured on the scale of the 
air thermometer, the preceding law would give the expression of the quantities of 
heat lost in function of the corresponding temperatures. But as wc have proved 
that the specific heat of bodies is not constant on any thermometric scale, wc ace, 
that to obtain the real loss of heat, it is necessary to introduce one more element, 
that is to say, the change in capacity of the body submitted to experiment. In con- 
sidering the question under this point of view, it would be necessary to determine 
the law which regulates the variation of capacity of a certain body, nnd afterwards 
to determine, by direct experiment, the quantities of heat lost by the same body at 
fixed points of temperature, as measured by the air thermometer. Then multiply- 
ing the rates of cooling deduced by the aid of the preceding law into the corre- 
sponding capacities, we should obtain the actual loss of heat. It is not, of course, in 
the interval of the first 200 or 300 degrees on the centigrade scale, that we could 
hope to verify these conclusions. As the variation in capacity only begins to be 
sensible beyond this limit, it would be necessary to push our observations as far 
as temperatures of 500 or 000°. We may easily conceive the difficulties of such 
an enquiry. We have, however, succeeded in constructing an apparatus which an- 
swers all the requisite conditions ; and we have even made a great many experi- 
ments on the subject. But as our results do not as yet exhibit all the regularity 
which we hope to give to them, we shall defer the publication of them for the 
The means which Mr. Leslie has employed to measure the radiating power* of 
different surfaces, are well suited to determine the quantities of radiating heat lost at 
different temperatures. It is known that this method consists in estimating the effect 
of radiation in any body, by the heating of an air or mercurial thermometer, placed 
at a certain distance from the hot body, and (in order to render the effect more sen- 
sible! in the focus of a reflector. .... i. i 
It was by means of such an apparatus that Laroche obtained the result we have 
before referred to. Amongst the series of observations then made, there is one 
which in fact includes very high temperatures, but which cannot be turned to any 
Iccou'ne ta “mpeU*es determined by. ..rondure founded on t e 
hypothesis of the capacities being constant. The oatnher, repr.«n. mg he loss of 
heat are affected by another error winch proceeds from th. tin error of 
K" ""Ll'tolhowSronr ll wilfeven’ represent observation, of 
this kind, when freed llils between SS the' 
ported m the same memoir, and . wlut “ , ■ w . ere made with a vessel 
variation of capacity has but ItUklnBne ^ „ not havin([ eaceeded 200“, we 
of iron, full of mercury. .tin like manner we may neglect the correc- 
may suppose the specific beat constent l ^ ^ . mercu J ia , f licrui0 „ ie ter, in 
tion which should be applied to the mu hecau e in Laroche's eiperi- 
order to reduce them to those ot the air tie , (1 liquid the observed 
meats, the stem of the thermometer, no being “ er * e “ cqu», at least, to the 
temperatures must have been affected by an opposite error, equal, 
above correction. . . , -„ nnr tnrl bv this experimenter, wc 
Instead of taking each series se P ar » e > » beir) P deduce ' d f ro m the formula by 
have used what may be termed mean re. » Vbich will be found in the 4th 
which Mr. Biot has expressed these observations, which 
, . . , . „ • « Les nombres qui represen- 
1 This passage is obscure. The original is i as autre erreur qui provient de ce 
- les nertes%echal e ur.ontdV.lleur.^tfed «aa«me <1 J/ 
le rechauftmeut de son thermometre focal etait trop granu, p H 
tent 
4ue ie rccnauiiuieui tie suu iuv.«.w. ,, 
actitude de la loi de Newton ne lut tres sensible. 
