Soaps are insoluble in strong caustic alkaline solutions, and therefore saponification (as the decomposition of fats by alkalis is called), does not readily take place in them, and for this reason the soap-boiler generally dilutes his caustic soda solutions to a strength not exceeding 18° Tw. (sp. gr. 1·090) in gravity, and separates the soap at the end of the operation, by the addition of brine, in which it is insoluble. An easier method, and one which is often useful for the preparation of small quantities of special soaps for fat liquors and the like, is as follows.[160] 10 lb. of a _good_ caustic soda, free from common salt, is dissolved in 4 gallons of water, and 75 lb. of oil or fat is warmed to about 25° C. or just sufficiently to render it liquid, and the soda solution is added in a thin stream, with constant stirring, which must be continued until the mass becomes too pasty. It is now set aside in a warm place for at least twenty-four hours, during which saponification gradually takes place. For leather purposes, a neutral soap, with a slight excess of fat, is generally advantageous, so that the fat may be increased to 80 lb.; or, in place of this, the operation will be facilitated by the addition of 5 lb. of commercial oleic acid. If soft soap is desired, 14 lb. of caustic potash may be used in place of the 10 lb. of caustic soda. The hardness or softness of soaps varies to some extent with the fat used, but potash soaps are always much softer than the corresponding soda soaps. It is obvious that with soaps made in this way, all the glycerin remains mixed with the soap. If, on testing, the soap does not prove to be free from caustic, it may be re-melted, which will generally complete the reaction. Before attempting to work with large quantities, a laboratory experiment is desirable, using 10 grm. of soda in 40 c.c. of water, and 75 to 80 grm. of oil or fat. The neutrality or freedom of the soap from caustic alkali may be tested by touching a freshly cut surface with an alcoholic solution of phenolphthalein, which the least trace of caustic soda or potash will render pink. [160] Carpenter, ‘Soap, Candles and Lubricants,’ p. 144. If solutions of soaps are mixed with those of salts of the heavy metals or of the alkaline earths, a mutual decomposition takes place, the acid of the salt combining with the alkali of the soap; and the fatty acid with the metallic base, to form a metallic soap. Most of these soaps are sticky masses, insoluble in water, but not unfrequently soluble in turpentine or petroleum spirit, if previously thoroughly dried, so that some of them have been applied to the production of varnish. Alumina soaps are occasionally used to thicken mineral oils, or render them more viscous. The general reaction of the stearin soap with calcium sulphate is shown in the following equation, though in practice it is sometimes more complex: