spectrum extends but a little way beyond the red into the dark space.— This explains Seebeck's discovery (page 165), viz. that the maximum of heat is found in different parts of the spectrum according to the nature of the prism. The smaller the number of calorific rays which any substance transmits (p. 215), whilst it retains the less refrangible rays, the more nearly does the maximum of heat in the spectrum formed by a prism made of that substance approach to the violet; whence, according to Seebeck, a prism of water places it in the yellow, of oil of vitriol in the orange, of crown-glass in the red, of flint-glass beyond the red; and with a prism of rock salt, which completely transmits even the least refrangible rays, the maximum of heat is situated, according to Melloni, in the dark space, and as far removed from the red as the red itself is from the violet. With some glass prisms, the maximum of heat approaches more nearly to the violet, in proportion as the ray of solar light passes through a thicker part of the prism,-because, in this longer passage through the glass, a greater number of the less refrangible heat-rays are retained. (Melloni.)

If light and heat are regarded as identical in substance, it may be supposed that the rays of heat approach more nearly to those of light in proportion as their motion is more rapid (which also implies increased refrangibility). The hotter the source of heat, the greater is the number of the more rapidly moving calorific rays emitted from it; and these pass without sensible absorption through glass and other diathermanous bodies, and then also through several, whilst the more slowly moving heat-rays are absorbed. (Ritchie.)

The subject of Radiant Heat has lately been further investigated by Knoblauch. (Pogg. 70, 205 and 337; 71, 1: abstr. Ann. Chem. Pharm. 74, 193.) The principal results obtained by this philosopher are as follows.

1. The quantity of radiant heat transmitted through diathermanous bodies is not (as former experiments seemed to show) directly proportional to the temperature of the source, but depends only on the constitution of the diathermanous body,-each body being permeated by certain calorific rays more readily than by others, whether those rays are emitted at a higher or at a lower temperature. Thus when rays of heat from incandescent platinum, the flame of alcohol, of an argand lamp, and of hydrogen gas, were made to pass through colourless glass and alum, and then to fall on a thermo-multiplier, the distance of which was so adjusted as always to give a deflection of 20° by direct radiation when the diathermanous bodies were removed, the following results were obtained:

Through rhombohedral tale-mica (Kali-und magnesia-glimmer), the heat from the flame of hydrogen passes less easily than that from either of the other three sources.-By experiments made with a cubical vessel containing hot water, also with a cylinder of untinned iron heated over au

Argand lamp, and a spiral of platinum wire heated to various degrees of incandescence, it was likewise found that when the heat emitted from one and the same body at various temperatures is made to pass through different diathermanous bodies, the quantity of heat transmitted is not proportional to the temperature of the source, but depends upon the nature of the diathermanous media.

2. With regard to the heating of bodies by radiation, Knoblauch shows that: The heating effect produced, when the radiated heat which reaches the bodies is of given intensity, is totally independent of the temperature of the source, and determined only by the nature of the absorbing bodies, which receive certain rays more readily than others.—Thus when the heat from an Argand lamp and a metal cylinder heated to 80° R. was made to fall on a plate of metal covered with lamp-black on the side turned towards the thermo-pile, but having its other side covered, in one case with carmine, in the other with black paper, the following results were obtained:

To determine the manner in which the heating of bodies by radiation is affected by their thickness, experiments were made with a metallic plate covered with layers of varnish, &c., of various thicknesses. The following table exhibits the results.

60 10.50 14.50

8.25 15 12 9.25 9.00

8.25 8.25 7.12 8.25 8.62 9.5 15.62 15.75 14.50 16.25 17.37 18.12 9.50 9.50 9.12 9.87 11.62 12.00

7.25

8.00

16 12

18.50

8.75

9.62

19.50

ylinder heated 35 6.50
to 80° R.... 60 10.50 17.50 18.12 20-12 20-75 18.62 20.25 21-37 22.12 17.00

Hence it follows, that within the limits of these experiments, the substances employed are more strongly heated, in proportion as their thickness is greater. Precisely the contrary result was obtained by Leslie and Melloni. The discrepancy is explained as follows. The amount of heat imparted to a body by radiation increases in proportion to the number of absorbing layers to which the heat can penetrate. But the heating effect attains its maximum at a certain thickness, beyond which the heat imparted by radiation cannot attain. Now in Knoblauch's experiments, the thickness was never too great to allow each successive layer to be

heated, and thus to act upon the metallic surface. In the observations of Leslie and Melloni, on the contrary, the interposed diathermanous plates were so thick, that only a small portion of the heat absorbed penetrated to the side which was turned towards the thermoscope.

3. With respect to the radiating powers of different substances at the same temperature, Knoblauch confirms the law laid down by Melloni, viz. that the radiating power of a body is influenced by scratching its surface, only in so far as its density and hardness are thereby altered:-also the result previously obtained by both Rumford and Melloni, that: the radiating power increases with the thickness of the radiating film,—a law which furnishes another proof of the correspondence between radiation and absorption. Knoblauch likewise observes that the equality of the radiating and absorbing powers is absolutely true as regards one and the same body; but that with respect to different bodies, it cannot be maintained that a body which at a certain temperature exhibits a higher radiating power than another, necessarily also possesses a greater absorbing power;-for the proportion between the quantities of heat absorbed by two bodies varies with the nature of the calorific rays.-Lastly, it is shown that the radiating power of a body is the same, however different may be the calorific rays by which it is heated.

4. The heat radiated from the most various solid bodies, such as metal, wood, porcelain, leather, cloth, pasteboard, &c.-of different thicknesses and different conditions of surface appears, when tested by all the means at our command, to be of the same nature, in whatever manner it may have been excited.-In the experiments by which this result was obtained, the temperatures of the sources of heat varied from 25° to 90° R. This result is of some interest with reference to the determination of specific heat; for if the ice in the calorimeter were to absorb the heat radiated from different substances in different degrees, the quantity of ice melted would not be a direct measure of the quantity of heat.

5. Alteration of heat by Irregular Reflection. Melloni has remarked that a white surface reflects, with various degrees of intensity, the heat of a Locatelli's lamp, according as it is used with or without the glass chimney, also the heat of incandescent platinum, and that of a metal cylinder heated to 400° C.-Metallic plates with rough surfaces are the only bodies which reflect equally the heat from all sources, whilst lamp-black gives a scarcely perceptible dispersion with any.-The following table contains the results of a number of experiments in relation to this subject. The source of heat used was an Argand lamp, and the heat, after reflexion from the various substances mentioned at the bead of the table, was made to traverse the several diathermanous media mentioned in the first column,-the object being to determine whether the calorific rays, after diffuse reflection from those various surfaces, would pass through the different media in equal or unequal quantities.

The rays of heat reflected from certain homogeneous bodies passed in unaltered proportion through the diathermanous media. Such was the case with birchwood, cork, and mahogany; also with the simple metals and metallic alloys.

Heat is therefore altered by diffuse reflection in very different ways; in a high degree by some bodies, not at all by others. These alterations, in the case of unpolished bodies, are independent of their degree of roughness:-in the case of metallic surfaces, it is even indifferent whether they are used in a state of high specular polish or in any other condition of surface.

By making use of incandescent platinum, the flame of alcohol, and a heated metal cylinder as sources of heat, it was found that: The changes produced in heat by irregular reflection are affected by the nature of the source of heat as well as by the nature of the reflecting surface. And in particular, that the modifications, which are very considerable in the rays of the Argand lamp, are less in those of red-hot platinum, still less in those of the alcohol flame, and in the case of the metal cylinder heated to any temperature between 20° and 90° R, they become absolutely nothing.

It is easily seen how by these modifications the rays of heat reflected from different substances, may to a certain extent, alter their relations one to another. Thus, the heat of an Argand lamp when reflected from carmine, passes through gypsum with less facility than when reflected from white velvet. The rays of incandescent platinum pass equally well through gypsum after reflection from those surfaces; and the heat of an alcohol flame passes through that medium after reflection from carmine better than after diffuse reflection from white velvet.

On repeating the experiments with the four above-mentioned sources of heat with reference to a different object, it was found that surfaces which affect equally the rays from any one source of heat, e. g. of an Argand lamp, likewise modify in an equal degree the rays from any other source. The following lists contain those substances which scatter the rays of heat in such a manner that, as far as regards their passage through red glass, blue glass, alum, rock-salt, calcspar, and gypsum, they are not to be distinguished one from the other. The bodies in (1) have likewise this peculiarity,-that the heat irregularly reflected at their surfaces is undistinguishable from non-reflected heat.

(1.) Gold, silver, platinum, mercury, iron, tin, zinc, copper, lead, alloy of lead and tin, brass, German silver, untinned iron plate. (2.) Gypsum, chalk, white lead, white oil-colour, porcelain, linen, white paper, blue paper, white cotton, grey calico, Paris green, green cinnabar, chrome yellow, black lac. (3.) Birch-wood, cork, mahogany, yellow marble. (4.) White satin, black satin, white taffetas, black taffetas. (5) Blue velvet, black velvet. (6.) Yellow leather, brown morocco. (7.) Light cloth, black cloth. (8.) Blue flock-paper, green flock paper. (9.) White wool, red wool. (10.) Cinnabar, oxide of copper.

Of the following substances, those contained in the same division exhibit a similar but not exactly equal action.

(11.) Carmine, madder, red flock-paper. (12.) White velvet, white wool, green flock-paper. (13.) White lead, Diessbach blue. (14.) Black velvet, green oil-cloth. (15.) Black paper, black glass. (16.) Coal, coke, plumbago. (17.) Lamp-black, animal charcoal.

The following substances, with reference to the dispersion of heat, cannot be included in either of the preceding groups.

(18.) Ultramarine. (19.) Peroxide of tin. (20.) Tannate of per