An ingenious addition is put to the instrument in order to ascertain this directly. It is constructed so as to ascertain the number of particles that fall from a known height. Under a low-power microscope, and concentric with it, is mounted a tube 2 inches long and 1-1/2 inch in diameter, with a bottom and a cover, which are fixed to an axis parallel with the axis of the tube, so that, by turning a handle, these can be slid sideways, closing or opening the tube at both ends when required. In the top is a small opening, corresponding to the lens of the microscope, and in the centre of the bottom is placed the observing-stage illumined by a spot-mirror. The handle is turned, and the ends are open to admit the foggy air. The handle is quickly reversed, and the ends are closed, enabling the observer to count on the stage all the fog-particles in the two inches of air over it. The number of dust-particles in the air which become centres of condensation depends on the rate at which the condensation is taking place. The most recent observations show that quick condensation causes a large number of particles to become active, whereas slow condensation causes a small number. After the condensation has ceased, a process of differentiation takes place, the larger particles robbing the smaller ones of their moisture, owing to the vapour-pressure at the surface of the drops of large curvature being less than at the surface of drops of smaller curvature. By this process the particles in a cloud are reduced in number; the remaining ones, becoming larger, fall quicker. The cloud thus becomes thinner for a time. A strong wind, suddenly arising, will cause the cloud-particles to be rapidly formed: these will be very numerous, but very small--so small that they are just visible with great care under a strong magnifying lens used in the instrument. But in slowly formed clouds the particles are larger, and therefore more easily visible to the naked eye. Though the particles in a fog are slightly finer, the number is about the same as in a cloud--that is, generally. As clouds vary in density, the number of particles varies. Sometimes in a cloud one cannot see farther than 30 yards; whereas in a few minutes it clears up a little, so that we can see 100 yards. Of course, the denser the cloud the greater the number of water-particles falling on the calculating-stage of the instrument. Very heavy falls of cloud-particles seldom last more than a few seconds, the average being about 325 on the square inch per second, the maximum reaching to 1290. This is about four times the number counted in a fog. Yet the particles are so very small that they evaporate instantly when they reach a slight increase of temperature. CHAPTER IX FORMATION OF CLOUDS