to zero.

23 degrees. Invert the position of the plates, so that the blackened sides come into contact ; it will now sink down Remove either of the plates, and the liquor will again mount near four degrees. It is truly pleasing, says Mr. Leslie, to witness this varied spectacle, where the changes succeed each other as if performed by the fancied operation of magic. But those transitions, and even the measures of the diversified effects, are the necessary results of the principles already established. Compare the case where both the external surfaces of the screen are metallic with that in which they are covered with pigment. On the one side it receives five times less heat, and this heat is propagated with eight times less energy from the other. By the joint influence of those circumstances, therefore, its effect is 40 times less; which corresponds to about half a degree, a quantity scarcely distinguishable. When the screen consists only of a single plate blackened on the one side, the diminished effect is a mean between the receptive and the projecting powers, or 6 times smaller than where both surfaces are painted. This enfeebled impression is consequently equal to about four degrees.

The experimental inquiries of Mr. Leslie, which were directed particularly to the radiation of heat, its connection with the spaces through which it is propagated, the direction in which it moves, the projecting power of the heated body, and connection which subsists between it and the nature of the projecting surface, are highly important.

The canister being placed successively at different distances from the reflector, the effects of its removal, upon the differential thermometer in the focus, were noted; and an allowance was made for the changes in the focal length. It was found, in general, that the total corrected effects were inversely as the distances of the canister. This obstruction evidently cannot arise from the loss of heat in the atmosphere; for that cause would produce a diminution in a much more rapid series. To prove that the irregularity of the reflecting surface has no share in the phenomenon, Mr. Leslie shews that a concave glass speculum reflects the heat of a charcoal fire with an energy inversely as its distance. The diminution, then, can only

be owing to an imperfect reflection; and this was well demonstrated, by increasing the size of the canister, in such a proportion to its increased distance from the reflector, that it always subtended an equal angle there. Making the same correction as before, for the variation of focal length, the total effect of the removal upon the differential thermometer was found to be nothing more than might fairly be ascribed to the inaccuracy unavoidable in such delicate experiments. Hence the capital inference is drawn, that the impulses, by which heat and cold are propagated between distant objects, do not suf fer any sensible diminution of strength from the length of their progress through the atmosphere.

A very remarkable aberration is observed to take place in the reflection of heat. When the flame of a taper is withdrawn above two inches from the axis of the reflector, its image vanishes entirely from the focal ball of the differential thermometer. But the heated canister may be removed seven inches from the axis, before its impression on the instrument ceases to be distinguishable. Mr. Leslie gives us no computation of the extent of this penumbra of heat (if the expression may be used) he only explains the law by which the lateral motion of the heated body, from the reflector's axis, diminishes the total effect of the reflection upon the thermometer. The distance of the canister being proportioned to the series 1, 2, 3-7, the rise of the thermometer is as the powers of, whose exponents are the triangular numbers, 1, 3, &c. He also finds, by experiment, what he pretends is also a deduction from the aberration just now described, that the maximum of the effect produced by reflection is not in the focus, but nearer the speculum. His experiment is quite conclusive. The thermometer being at 58o in the optical focus, it rose to 80° half an inch nearer the reflector; and, half an inch beyond the focus, fell to 25°. The heat is therefore more diffused over equal spaces beyond the focus than between the focus and reflector.

If the side of the canister is turned gradually round, while its axis remains fixed in the axis of the reflector, the thermometer is less and less affected, as the radiating surface is more inclined from the perpendicular. Nothing can be simpler than the method adopted by Mr. Leslic

for ascertaining the law of this diminution. He placed a sliding screen between the canister and the reflector, and adjusted the slit so, that the radiating surface, at every part of its motion round its axis, subtended the same angle as the reflector. There was scarcely any effect produced upon the thermometer by the revolution of the canister. In like manner a cylindrical canister produces the same rise in the thermometer with a cubical one of an equal base and altitude; and from these, and other experiments founded upon the same principle, we may conclude, that the total action of a heated surface is equal to that of its orthographic projection, or proportional to the sine of its inclination.

Mr. Leslie having ascertained that bodies differ very widely in their power of projecting, absorbing, and reflecting heat, found that the chemical qualities of the heating surface have a considerable influence upon its projecting power; the effect of tin being 12, iron or steel operates as 15, mercury above 20. All oxydes acquire a greater action as they recede from the metallic state. Lead being as 19; when tarnished by exposure to the air, it becomes as 45, while minium is as 80. Sealingwax and rosin are nearly equal to paper, and ice is as 85. The polish of the radiating surface diminishes its action, where that is not naturally great. The roughening of glass does not heighten its projecting power; but that of tin is doubled by covering it with furrows. This singular effect cannot be owing to the greater surface which the roughened metal exposes; for the increase of surface is precisely counterbalanced by the increase of obliquity; and, moreover, it is found that the addition of cross furrows, by striating the surface in the other direction, nearly destroys the effect of the first operation. The thickness of the radiating surface greatly affects its powers of action. A thin film of isinglass produces a radiation as 26; a thick one as 42: but when the thickness exceeds the thousandth part of an inch, any subsequent increase does not augment its action. Mr. Leslie thinks that the difference in projective power, which is observable in several of the cases above noticed, may be resolved into the variations of the bodies with respect to hardness and softness. He reasons this matter with his wonted inge

nuity, and shows that the addition of moisture, and still more the addition of a mucilaginous substance, considerably augments the action of a surface painted black. The quality of colour is the last to which our attention is directed, and Mr. Leslie seems disposed to doubt whether it exerts any influence at all in modifying the projective and absorbent powers of bodies; a point which he conceives is incapable of strict solution, because a change of colour must always be attended with an alteration in the structure of the substance: and, rigorously speaking, this is no doubt true. But there is one mode of inducing a change of colour, by means of a change in the body's structure, which is known, and for which allowance may be made by our author's experiments; we allude to the scoriæ on the surface of metals, from slight oxydation. If the oxydation is found always to augment the metal's action, as different metals assume different hues in the beginning of the process, a comparison of several, in this respect, will enable us to estimate how far colour operates. Perhaps even the change induced upon the substance of vegetable tinctures, by weak, acidulous, and alkaline solutions, is so little proportionate to the alterations which their colours undergo in the mixture, that an approximation might be obtained from experiments with paper dyed in this manner.

Mr. Leslie repeated several branches of this inquiry, to ascertain the various reflecting powers of different surfaces. Those which absorbed and projected heat most copiously were, in general, found to reflect least, though by no means in a reciprocal proportion. Glass reflecting as 10, tin-foil reflected as 85, lead as 60, steel as 70; tin-foil shining with mercury as 50, and brass as 100. A tin reflector had its power reduced to one tenth, by being striated in one direction, although this operation did not at all change the limits of its focus. A coat of tallow reduced the powers of the reflector to one twelfth; a coat of olive oil to two fifths; a coat of isinglass to three tenths, but, as it dried and became thinner, only to seven tenths: a thin irridescent coating reduced the reflection only to four fifths. The general rule then holds good, that the reflecting powers of bodies bear some inverse ratio to their absorbent and projective powers; although

so many circumstances unite in modifying the proportion that we are unable, as yet, to express it by one universal law.

It is therefore evident that caloric is thrown off from bodies in rays, which are invisible, or incapable of exciting vision, but which are capable of exciting heat.

These invisible rays of caloric are propagated in right lines, with extreme velocity, and are capable of the laws of reflection and refraction.

The heating agency, however, is different in the different coloured rays of the prismatic spectrum. According to Dr. Herschel's experiments, it follows inversely the order of the refrangibility of the rays of light. The least refrangible possessing it in the greatest degree.

Sir Henry Englefield has lately made a series of experiments on the same subject, from which we learn that a thermometer having its ball blackened rose when placed in the blue ray of the prismatic spectrum in 3' from 55° to 56°; in the green, in 3′ from 54° to 58°; in the yellow in 3' from 56° to 62°; in the full red, in 2' from 56° to 72°; in the confines of the red, in 21 from 58° to 7310; and quite out of the visible light, in 21 from 61° to

79°.

Between each of the observations, the thermometer was placed in the shade so long as to sink it below the heat to which it had risen in the preceding observation ; of course its rise above that point could only be the effect of the ray to which it was exposed. It was continued in the focus long after it had ceased to rise; therefore the heats given are the greatest effects of the several rays on the thermometer in each observation. A thermometer placed constantly in the shade near the apparatus was found scarcely to vary during the experiments.

Sir Henry made other experiments with thermometers with naked balls, and with others whose balls were painted white; for which we refer the reader to the interesting paper of the baronet,* from which the above experiments are transcribed.

The coloured rays emitted from the sun and combustible bodies, since they excite heat and vision, must consist of a mixture of heat-making rays and rays of light.

* Journals of the Royal Institution, No. x. p. 202.