Prof. Crookes has at length announced a theory in explanation of the movements exhibited by the remarkable "light mill" of his invention. He says: "The evidence afforded by the experiments is to my mind so strong as almost to amount to conviction, that the repulsion resulting from radiation is due to the action of thermometric heat between the surface of the moving body and the case of the instrument, through the intervention of the residual gas. This explanation of its action is in accordance with recent speculations as to the ultimate constitution of matter, and the dynamical theory of gases." The most refined means for exhausting the air from the glass bulb which contains the suspended vanes of the radiometer leave, and if they were to be carried to absolute mechanical perfection, would still leave a certain amount of gas in it. But Dr. Crookes has carried this attenuation so far that the number of gas molecules present can no longer be considered as practically infinite. Nor is the mean length of their paths between their collisions any longer very small compared to the size of the bulb. The latest use to which the radiometer has been put was to test the viscosity of gases at decreasing pressures. The glass bulb was furnished with a stopper lubricated with burnt rubber. This was fixed and carried a fine thread of glass which is almost perfectly elastic. To the end of this thread hung a thin oblong plate of pith to which a mirror was attached. The glass stopper being fixed, and the bulb capable of rotation through a small angle, it is evident that when the bulb is rotated the pith ball will remain at rest except as it yields to the friction of the air moved by the bulb. It does move, swinging a certain distance and then back, like a pendulum. The amount of this movement is carefully observed by a telescope, and recorded for five successive beats. As the pith and glass fibre form a torsion pendulum, it is evident that these beats will gradually die down in consequence of the resistance of the air. By exhausting the air to various degrees of rarity, it was proved that Prof. Clerk Maxwell's theory, that the viscosity of a gas is independent of its density, is correct. The logarithmic decrement of the first five oscillations (that is, the decrease, oscillation by oscillation, of the logarithm of the arc through which the pith vanes swing), was found to be nearly the same when the air was almost exhausted as when it was at its natural pressure, proving that its viscosity remained nearly equal for all pressures. Only in the exceptionally perfect vacuum referred to above did this logarithmic decrement sink to about one-twentieth of what it had commenced with. Repulsion of the vane by the action of light commences when this decrement is one-fourth of what it was before the exhaustion of air began. As the rarity of the air within the bulb increases the force of this repulsion begins to diminish, like the