The opinion that all the solar heat radiated into infinite space is wasted, starts moreover from a hypothesis which is not proved, and which is highly improbable—namely, that only an extremely small portion of the sky is covered with celestial bodies. That might certainly be correct if we assumed, as has formerly been done, that the majority of the celestial bodies must be luminous. We do not possess, however, any reliable knowledge of the number and size of the dark celestial bodies. In order to account for the observed movements of different stars, it has been thought that there must be in the neighborhood of some of them dark stars of enormous size whose masses would surpass the mass of our sun, or, at least, be equal to it. But the largest number of the dark celestial bodies which hide the rays from the stars behind them probably consist of smaller particles, such as we observe in meteors and in comets, and to a large extent of so-called cosmical dust. The observations of later years, by the aid of most powerful instruments, have shown that so-called nebulæ and nebulous stars abound throughout the heavens. In their interior we should probably find accumulations of dark masses. The light intensity of most of the nebulæ is, moreover, far too weak to permit of their being perceived. We have, therefore, to imagine that there are bodies all through infinite space, and about as numerous as they are in the immediate neighborhood of our solar system. Thus every ray from the sun, of whatever direction, would finally hit upon some celestial body, and nothing would be lost of the solar radiation, nor of the stellar radiation. As regards the radiation-heat exchange, the earth might be likened to a steam-engine. In order that the steam-engine shall perform useful work, it is necessary not only that the engine be supplied with heat of high temperature from a furnace and a boiler, but also that the engine be able to give its heat up again to a heat reservoir of lower temperature—a condenser or cooler. It is only by transferring heat from a body of higher temperature to another body of lower temperature that the engine can do work. In a similar way no work can be done on the earth, and no life can exist, unless heat be conferred by the intermediation of the earth from a hot body, the sun, to the colder surroundings of universal space—_i.e._, to the cold celestial bodies in it. To a certain extent the temperature of the earth’s surface, as we shall presently see, is conditional by the properties of the atmosphere surrounding it, and particularly by the permeability of the latter for the rays of heat.