Without adopting the explanation either of Prevost or Pictet, or advancing any hypothesis of our own on the subject, we shall deem it sufficient to show that the re -sults of the experiments with the reflectors afford no ar gument for the support of any theory, which might not be derived from direct radiation and absorption.
That we may begin with the simplest case, let it be supposed that the body A, Plate CXCIX, Fig. 5, is sur rounded on every side with bodies at an inferior tempe rature : it will immediately begin to throw off rays of car Ionic in all directions, and if the surrounding bodies be regularly and uniformly disposed, and have the same absorbing powers for caloric, it will cool equally on every side. If another body B, in every respect similar to A, at a temperature considerably lower than that of the other bodies, be now placed at no great distance from A, the body A will cool more rapidly than before, particularly on the side adjacent to B; and still more rapidly, if the surfaces a d, andfh, possess great radiating or absorbing powers. Since the cooling of A is promoted by the na ture of the surface/ h of B, and since those surfaces which radiate caloric most powerfully, also absorb it in the same degree in opposite circumstances, it is obvious, that any hypothesis, which assumes that emanations pro ceed mutually from one body to the other, during the pro cess of cooling, must have the same difficulty in explain ing why, in the present case, A is more rapidly reduced in temperature, when the radiation from B ought to be most abundant, as it would have in similar circumstances with the reflectors. Consequently, the objections which we urged against Prevost's theory, deduced from the experiment with the reflectors, that the greatest depres sion of the thermometer is exhibited when the surface of the cold body, placed in the focus of one of the reflec tors, is most favourable to radiation, apply with equal force in cases of cooling by direct radiation and absorp tion.
Let us now suppose that the radiations from the side b c, instead of being allowed to proceed from it in their ordinary course, are intercepted by some reflecting sur face, and thrown back in the same direction with the perpendicular emanations from the opposite side a d ; as this reflection can have no effect on the physical qualities of the emanations, but merely to change their direction, the only consequence of the arrangement we have sup posed will be to retard slightly the cooling of A, by the partial re-absorption of those rays which are thrown back upon b c. If we further suppose, that the reflected ema
nations now moving parallel to one another, are again intercepted by a reflecting surface placed behind B, and thrown by means of its figure upon the surface e g, the same effect must now be produced on this surface as if there had been a direct radiation toward it from b c. In deed, it might be inferred, a priori, that as the only func tion of the reflecting surfaces is to change the direction of the incident rays, we have no reason to expect, when they are employed, any effect different from that which is obtained by immediate and direct radiation.
We accordingly find, that the greatest effect is pro duced in diminishing the temperature of A, when the sur faces b c and r g are best fitted to radiate and absorb the calorific rays, just as we found the same thing to hold by direct radiation, when the surfaces a d and j h were in like circumstances. It is evident, therefore, that the rapid cooling of a hot body, when it is placed in the focus of a reflector, by means of a cold body placed in the fo cus of an opposite reflector, admits of being explained upon any hypothesis which affords a satisfactory expla nation of the simplest case of cooling, that by direct ra diation ; and that the experiments with the reflectors fur nish no argument in support of the doctrine of a frigo rifle principle of which we were not previously in pos session.
The degrees of natural cold which occur in the more inhospitable regions of the globe, are far sui passed by the reductions of temperature obtained by artificial means. The greatest cold that has hitherto been observed in the open air does not exceed — 50; and probably the extreme range of the inferior temperatures which occur in nature, is but little below that point.