Now let us examine some of the shapes which drops actually assume. I take a glass plate covered with a thin layer of grease, which prevents adhesion of water to the glass, and form upon it drops of water of various sizes by the aid of a pipette. You see them projected on the screen (Fig. 4). The larger drops are flattened above and below, but possess rounded sides and resemble a teacake in shape. Those of intermediate size are more globular, but still show signs of flattening; whilst the very small ones, so far as the eye can judge, are spherical. Evidently, the shape depends upon the size; and this calls for some explanation. If we take a balloon of indiarubber filled with water, and rest it on a table, the weight of the enclosed water will naturally tend to stretch the balloon sideways, and so to flatten it. A smaller balloon, made of rubber of the same strength, will not be stretched so much, as the weight of the enclosed water would be less; and if the balloon were very small, but still had walls of the same strength, the weight of the enclosed water would be incompetent to produce any visible distortion. It is evident, however, that so long as it is under the influence of gravitation, even the smallest drop cannot be truly spherical, but will be slightly flattened. The tendency of drops to become spherical, however, is always present. [Illustration: __Fig._ 5.—Formation of a sphere of orthotoluidine._] *Production of True Spheres of Liquids.*—Now it is quite possible to produce true spheres of liquid, even of large size, if we cancel the effect of gravity; and we may obtain a hint as to how this may be accomplished by considering the case of a soap-bubble, which, when floating in air, is spherical in shape. Such a bubble is merely a skin of liquid enclosing air; but being surrounded by air of the same density, there is no tendency for the bubble to distort, nor would it fall to the ground were it not for the weight of the extremely thin skin. The downward pull of gravity on the air inside the bubble is balanced by the buoyancy of the outside air; and hence the skin, unhampered by any extraneous force, assumes and retains the spherical form. And similarly, if we can arrange to surround a drop of liquid by a medium of the same density, it will in turn become a sphere. Evidently the medium used must not mix with the liquid composing the drop, as it would then be impossible to establish a boundary surface between the two. Plateau, many years ago, produced liquid spheres in this manner. He prepared a mixture of alcohol and water exactly equal in density to olive oil, and discharged the oil into the mixture, the buoyancy of which exactly counteracted the effect of gravity on the oil, and hence spheres were formed. The preparation of an alcohol-water mixture of exactly correct density is a tedious process, and we are now able to dispense with it and form true spheres in a more convenient way. There