the tourmaline; but that the effect with non-luminous heat is very minute and difficult to perceive, on account of the secondary radiation. Though light is almost entirely excluded in one position of the pieces of tourmaline, and transmitted in the other, a vast quantity of radiant heat passes through them in all positions. Eighty-four per cent. of the heat from an argand lamp passed through them in the case where light was altogether stopped. It is only the difference in the quantity of transmitted heat that gives evidence of its polarization. The second slice of tourmaline, when perpendicular to the first, stops all the light, but transmits a great proportion of heat; alum, on the contrary, stops almost all the heat, and transmits the light; whence it may be concluded that heat, though intimately partaking the nature of light, and accompanying it under certain circumstances, as in reflection and refraction, is capable of almost complete separation from it under others. The separation has since been perfectly effected by M. Melloni, by passing a beam of light through a combination of water and green glass, coloured by the oxide of copper. Even when the transmitted light was concentrated by lenses, so as to render it almost as brilliant as the direct light of the sun, it showed no sensible heat. Professor Forbes next employed two bundles of laminæ of mica, placed at the polarizing angle, and so cut that the plane of incidence of the heat corresponded with one of the optic axes of this mineral. The heat transmitted through this apparatus was polarized from a source whose temperature was even as low as 200°; heat was also polarized by reflection; but the experiments, though perfectly successful, are more difficult to conduct. It appears, from the various experiments of M. Melloni and Professor Forbes, that all the calorific rays emanating from the sun and terrestrial sources are equally capable of being polarized by reflection and by refraction, whether double or single, and that they are also capable of circular polarization by all the methods employed in the circular polarization of light. Plates of quartz cut at right angles to the axis of the prism possess the property of turning the calorific rays in one direction, while other plates of the same substance from a differently modified prism cause the rays to rotate in the contrary direction; and two plates combined, when of different affection, and of equal thickness, counteract each other’s effects as in the case of light. Tourmaline separates the heat into two parts, one of which it absorbs, while it transmits the other; in short, the transmission of radiant heat is precisely similar to that of light.