If alumina be fused with potash, aluminate of potash, soluble in water, is made; if, however the solution is too concentrated, a certain quantity of the alumina will be precipitated; but if it be dilute, the whole of the alumina will remain in solution. When aluminate of potash of specific gravity 1A.12 is mixed with a solution of silicate of potash of specific gravity 1A.2, no precipitate or gelatinization will take place for some hours; the more dilute the solution, the longer will it remain without gelatinization, and of course the thinner it will be, and the greater power of penetration it will have when applied to a porous surface. When solutions of aluminate of potash and of silicate of potash of greater density are mixed together, a jelly-like substance is almost immediately formed, and sometimes even the whole mass gelatinizes. If this jelly be allowed to dry slowly, it will contract, and at last a substance will be left behind sufficiently hard to mark glass, though the time for this hardening may be from one to two years; and on examination it is found that this substance has very nearly the same chemical composition as felspar, and is perfectly insoluble in ordinary mineral acids. Now, suppose a dilute solution of this mixture to be applied to the surface of stone, the silicate and aluminate of potash will gradually harden and fill up the interstices of the stone; and as both the substances entering into combination are contained in the same solution, they will both penetrate to the same depth. Inasmuch as the artificial felspar is not acted upon by destructive agents which would disintegrate the stone, it becomes a bonding material for its loosened particles, and at the same time gives a case-hardening to the stone, which no doubt will as effectually protect it against atmospheric influences as in the case of the hardening of the natural one. We have a tolerable guarantee that this will be so, if we consider the number of enduring minerals into the composition of which silica, alumina, and potash enter, and also of the almost imperishable character of granite, which is so largely composed of felspar. Many experiments have been performed on an exhaustive scale with these materials, and in every case it has been found that they have answered the expectation of those who have thus tested them. It is, however, necessary to state, that in making these experiments, great care must be used to employ the mixed substance in solution before gelatinization has set in, for if this has occurred, even to the slightest extent, a surface coating is formed on the stone, which, not having formed a bond with it, easily rubs off.