canister, to mingle its effects upon the thermo meter with those of any of the adjoining sides; until some part of the reflector is brought to occupy some part of each of the two quadrants of the circle. But it is easy to place the reflector in such situations as to mingle the effects of any two sides of the canister, upon the thermometer, in any proportion we please. For example,

Suppose the reflector, and also the thermometer to be fixed fast, in their proper situation, upon the piece of wood M, extending a little beyond the canister at one end, and a little beyond the circumference of the circle at the other, and moveable upon a centre under the middle of the pedestal, upon which the canister G is placed. We shall be enabled, by this means, to remove the reflector, and the thermometer, to any part of the circle we please, without deranging the apparatus. Suppose the reflector to be, by this means, turned round to the right hand, until the line g, drawn from the centre of the canister, to the extreme edge of the right side of the reflector, be brought to correspond precisely with the line e. In this situation, the whole of the reflector will still remain within that quarter occupied by the black side of the

D

canister; and it will, in this situation, produce precisely the same effect upon the thermometer, as it did when it was placed directly in front of

it.

Let the board be now moved round still farther to the right hand, until the axis of the reflector L, correspond precisely with the line e. In this situation, it is obvious, that the black surface of the canister will act upon one half of the reflector, and the clear surface upon the other half of it: and by this means, an effect will be produced, compounded of one half of the power of the black surface which is 50 degrees, and one half of the power of the clear surface, which is 6 degrees; and in consequence, the coloured liquor in the thermometer will, in this situation of the reflector, immediately descend to 56 degrees.

We shall afterwards have occasion to show the particular effects which may be produced upon the thermometer, by the combined effects of any two of the four different surfaces of the canister, corresponding to any degree of the circle occupied by the breadth of the reflector.

Let us now return again to the consideration of the 6th Experiment, and show the cause of that declension in the effect upon the thermo

meter, which is said in the 7th Experiment, to have been produced by removing the pane of glass to a greater distance from the canister.

Let the frame, containing a pane of glass of six inches long, and of a sufficient breadth to cover the whole of the black face of the canister, be placed at I, (see plate 1) two inches from the front of the canister. The glass, in this situation, will extend from the line e to the line f, and consequently it will there intercept the whole of the rays of heat which issue from the black surface of the canister, and which are diffused through that quarter, or 90 degrees of the circle. In this situation Mr. Leslie says, in in the 6th Experiment, that only one-fifth part of the rays, or one-fifth part of the whole unobstructed power of the black surface of the canister, was transmitted through the glass to the reflector: and he supposes, that the other fourfifths of the rays were absorbed by the glass in their passage through it. It is, however, as probable that these rays were reflected from the surface of the glass.

Now remove the glass screen to m, one foot distant from the front of the canister. It will, in this situation, cover only 18 degrees of the circle, between the lines i and k; and conse

quently can receive only one-fifth part of the rays of heat proceeding from the canister which fellupon it in the former situation, when it was placed near the canister. For in that situation, the rays of heat discharged by the black surface of the canister, passed through the glass in a condensed state; the glass being there, near the centre of the heated body from whence they proceed. But when the screen is removed to m, it is obvious, that those rays of heat are greatly diffused; and the force of these rays, which fall upon it in that situation, must, in consequence, be diminished in proportion to that diffusion: and, as has been already observed, they can possess no more than one-fifth of the power which they had in the first situation of the screen. And if four-fifths of these diffused rays which fall upon the glass in this distant situation, and which are only one-fifth part

of the whole rays that proceed from the black

surface, are also absorbed by the glass in their passage through it, it is not possible that the remaining fifth-part of these diffused rays could produce an effect upon the thermometer of more than four degrees. For as the black surface of the canister, when the glass-screen was placed immediately before it, produced an effect upon

the thermometer of only 20 degrees, it could not, when the screen was removed to m, produce more than one-fifth part of that effect; and the fifth part of 20 is just 4 degrees. This is precisely one-twenty-fifth part of the full and unobstructed effect. But as it is probable, that more than four-fifths of these diffused rays would be absorbed or reflected, it might, perhaps, in that situation, produce an effect of only one-thirtieth part, which he says it actually did. We will now proceed to

EXPERIMENT VIII.

Remove the pane of glass, and in its stead attach to the frame a sheet of writing paper; "dispose the apparatus properly, and, having "placed the screen two inches before the black"ened side of the canister, fill this with boiling 66 water. The liquor of the differential ther"mometer will now rise to 23 degrees; an ef"fect equal nearly to the fourth part of what it "produced without the intervention of the "screen."

This, in opposition to Mr. Leslie's opinion, is surely a proof that heat as well as light is transmitted through the paper.

There is, however, a circumstance in this ex