Melloni had investigated the laws of the radiation and absorption of radiant heat in solid and liquid matter; but its radiation and absorption by gases and vapours was unknown previous to the experiments of Mr. Tyndall. The apparatus employed was a horizontal brass tube four feet long, between two and three inches in diameter, polished inside, and closed air-tight at each end by a plate of rock-salt, which transmits more heat than any other substance. The air could be pumped out of the tube by one pipe, and the gas or vapour for the experiment introduced by another. Close to one end of the brass tube there was a thermo-electric pile connected with its goniometer. On each side of this arrangement there was a vessel of water kept at the boiling point. These two vessels were so placed that when the rays of heat from one of them passed through the exhausted tube, and fell upon one face of the thermo-electric pile, their effect was so neutralized or balanced by the rays of heat falling on the opposite face of the pile from the other, thus the needle of the goniometer was steadily maintained at zero, and its deflection instantly showed the absorbent effect produced by any gas or vapour that was admitted into the exhausted tube. Since aqueous vapour has a very exalted absorbent power, a gas or vapour was rendered perfectly dry before its absorbent capacity was determined. For that purpose the pipe that introduced it into the brass experimental tube was so constructed that the gas had first to pass over fragments of pumice-stone wet with strong sulphuric acid, which absorbed its moisture and dried it. Common atmospheric air, however, was not only dried in this manner, but it was deprived of its carbonic acid by passing over caustic potash, and many other precautions were taken to prevent the possibility of error. Under the ordinary pressure of the atmosphere, when the experimental tube was exhausted, the needle of the goniometer stood at zero, but as soon as pure dry atmospheric air was introduced into the tube its absorption caused the needle to move from zero to 1°.