This great divergence in the physical character of clays is due to their method of formation. Clay originated in the weathering of felspar, which chiefly consists of a double salt, a compound of the silicates of alumina and potash. Under the influence of air and water, this compound is decomposed, the potassium silicate passing into solution, whilst the aluminium silicate, being insoluble in water, is carried away by that medium. When the water can no longer carry the particles of aluminium silicate in suspension--for example when it reaches a sea or lake--the silicate settles down to the bottom, and a deposit of clay is formed. If the original felspar was very pure, and in particular very low in iron, the resulting clay will be of a handsome white colour. An example of this is afforded by kaolin, or porcelain earth, which is preferably used for making china. If, however, the felspar contained a considerable proportion of ferric oxide, the resulting clay is yellow; and if stones or mussel shells became incorporated with the clay prior to deposition, these bodies will be found as inclusions in the deposit, and such clay will require much troublesome preparation--grinding and levigation--before it is fit for use. For the purposes of the colour-maker, the most suitable clay is one that is pure white, free from inclusions, and does not change colour when exposed, in a finely divided state, to the action of the air. Many clays that were originally white gradually assume a yellow tinge on prolonged exposure to air and moisture, because the clay contained ferrous oxide, which changes, in the air, to the stronger pigment, ferric oxide. Many kinds of clay merely require a simple levigation to fit them for use as pigment. The lumps of freshly dug clay are placed in large tanks, etc., filled with water and stirred up continuously in order that, instead of forming a plastic mass which is very difficult to distribute in water, the particles detached from the lumps may pass at once into suspension. This turbid water is then transferred to another tank, etc., where the minute particles of clay are allowed to settle down, and the water becomes quite clear. Where this work is carried on on a large scale, it is advisable to put the freshly won clay into large pits close to the clay deposit, and to leave it there, covered with water, during the winter season. The freezing of the water breaks down the larger lumps of clay, by the resulting expansion, and this facilitates the subsequent levigation, the cohesion between the particles being destroyed.