In our ordinary atmosphere there can be no clouds without dust. A dust-particle is the nucleus that at a certain humidity becomes the centre of condensation of the water-vapour so as to form a cloud-particle; and a collection of these forms a cloud. This condensation of vapour round a number of dust-particles in visible form gives rise to a vast variety of cloud-shapes. There are two distinct ways in which the formation of clouds generally takes place. Either a layer of air is cooled in a body below the dew-point; or a mass of warm and moist air rises into a mass which is cold and dry. The first forms a cloud, called, from being a layer, _stratus_; the second forms a cloud, called, from its heap appearance, _cumulus_. The first is widely extended and horizontal, averaging 1800 feet in height; the second is convex or conical, like the head of a sheaf, increasing upward from a level base, averaging from 4500 feet to 6000 feet in height. There are endless combinations of these two; but at the height of 27,000 feet, where the cloud-particles are frozen, the structure of the cloud is finer, like "mares' tails," receiving the name _cirrus_. When the cirrus and cumulus are combined, in well-defined roundish masses, what is familiarly described as a "mackerel sky" is beautifully presented. The dark mass of cloud, called _nimbus_, is the threatening rain-cloud, about 4500 feet in height. At the International Meteorological Conference at Munich, in 1892, twelve varieties of clouds were classified, but those named above are the principal. In a beautiful sunset one can sometimes notice two or three distances of clouds, the sun shedding its gold light on the full front of one set, and only fringing with vivid light the nearer range. Although no man has wrought so hard as Dr. Aitken to establish the principle that clouds are mainly due to the existence of dust-particles which attract moisture on certain conditions, yet even twenty years ago he said that it was probable that sunshine might cause the formation of nuclei and allow cloudy condensation to take place where there was no dust. Under certain conditions the sun gives rise to a great increase in the number of nuclei. Accordingly, he has carefully tested a few of the ordinary constituents and impurities in our atmosphere to see if sunshine acted on them in such a way as to make them probable formers of cloud-particles. He tested various gases, with more or less success. He found that ordinary air, after being deprived of its dust-particles and exposed to sunshine, does not show any cloudy condensation on expansion; but, when certain gases are in the dustless air, a very different result is obtained.