The tube was again exhausted; the needle stood at zero, but was deflected from zero to 1° as soon as the tube was filled with oxygen. A similar experiment was made with nitrogen and hydrogen with the same result. Thus, dry air and the elementary gases, oxygen, nitrogen, and hydrogen, have the same absorptive power, and consequently they all deflected the needle of the goniometer one degree. The whole amount of radiant heat that passed through the exhausted tube produced a deflection of 71° 5ʹ; hence taking as unit of heat the amount that would deflect the needle one degree, the number of units expressed by 71° 5ʹ is 308, consequently the absorption of each of these four gases amounts to 100/308 or 0·3 per cent. The most delicate tests could not show any difference between the three first, but Professor Tyndall had reason to believe that hydrogen has the lowest absorptive power of all gases and vapours, though he was unable to express the amount. The absorptive power of all four is very much less than that of every other gas or vapour, and invariably deflects the needle to 1°, which thus becomes the unit of comparison. Olefiant gas, the most luminous of the constituents of coal gas, possesses the highest absorptive power of the permanent gases. When sent into the exhausted tube it deflected the needle of the goniometer from 0° to 70° 3ʹ, which is equivalent to 290 units. The whole heat that passed through the exhausted tube before the gas was admitted produced a deflection of 75° or 360 units, consequently more than 7/10ths or 81 per cent. of the whole heat was cut off by the olefiant gas. Such opacity to heat in so transparent a gas is quite marvellous. A current of it was sent into the open air between the thermo-electric pile and one of the sources of heat, and although it was perfectly invisible, it instantly deflected the needle of the goniometer from 0° to 41°. In order to ascertain the relation between the density of the gas and the quantity of heat extinguished or absorbed, an ordinary mercurial gauge was attached to the air-pump. The experimental tube was exhausted, and the needle of the goniometer stood at zero. Then, from a graduated glass vessel, measures of olefiant gas, each amounting to the 1/50th of a cubic inch, were successively sent through the drying pipe into the exhausted tube. The amount of the heat absorbed and the depression of the mercurial column corresponding to each measure of gas as it was introduced, was registered from one to fifteen measures. This experiment showed that for very small quantities of gas, the absorption is exactly proportional to the density or tension. One measure of the gas only produced a depression of the mercurial column amounting to the 1/367th part of an inch, or about the 1/15th of a millimetre.