Such accuracy of combination as is obtained experimentally by precipitation can hardly be looked for in commercial glass manufacture. In the latter case, the formation of silicates from raw materials, and the formation of glass from the combination or mixture of silicates, have both to come into operation in the crucible. To ensure the production of definite silicates in the crucible, the raw materials must be mixed together in proper combining proportions, for if more metallic oxide be introduced than can chemically combine with the silica, the excess will form an impurity in the glass. To this cause is to be attributed, not only the blemishes visible in common glass, but also the decomposition of the substance of the glass, after long exposure to the action of air and moisture. The surface of old glass is commonly pitted and worn: this appearance is due to the solution, by moisture and carbonic acid, of an excess of material with which the silica has been unable chemically to combine ; or, in the first instance, to the materials having been mixed in improper proportions.
The quality which especially facilitates the manufacture of glass on a commercial scale is a curious viscosity or plasticity produced by the agency of heat. When solid glass is heated, it becomes gradually softer and softer, and passes by imperceptible stages into a liquid condition. It may be said to melt at the lowest temperature at which perceptible softening occurs, and to be fully melted when a further elevation of temperature does not make it more fluid ; but no precise tem perature can be given as its melting-point. At an intermediate stage between liquidity and solidity, which may conveniently be termed the stage of viscosity, glass is in an exceedingly favourable condition for manipulation. In this condition, it can be ladled, or poured, or gathered on a heated iron rod ; a solid mass, at the end of a hollow blowing-iron, will expand and become hollow by the pressure of the breath ; it can be moulded, pressed, or rolled ; so great is its ductility, that hollow or solid masses can be drawn out to immense lengths of tube or cane ; so weak is the cohesion of its particles, that a mass will lengthen by the force of gravity, and a contracted cup will expand into a flattened disc under the centrifugal force produced by rapid rotatory motion.
To maintain the glass in 'this state, it is necessary that glass-furnaces shall be capable of being worked with great regularity : sudden fluotuations of temperature affect not only the physical condition of the glass, but, by disturbing the homogeneity of the molten glass, produce strife and irregularities in its substance. The action of heat upon bodies is to develop a repulsive force between their molecules, which is continually struggling with molecular attraction. Under
its influence, bodies tend to expand, and finally to change their state of aggregation. This theory necessarily assumes the existence of pores or interstices between the molecules of matter, which increase in size with the growth of the repulsive force generated by heat. By this theory, many phenomena in the manufacture of glass may be explained. In glass in a viscous condition, the physical pores are large; and, as it passes from the viscous to the solid state, the pores must be gradually reduced in size throughout the entire mass. If the process of cooling be in any way hurried, the exterior of the mass will solidify sooner than the interior, and the latter will remain in a state of porosity and unstable equilibrium. This will espeoially be the case with glass, owing to the fact of its being an exceptionally bad conductor of heat. If a small quantity of molten glass be allowed to fall into water, a drop of glass is formed, similar to that represented at a, Fig. 727. If the thin end of this drop be cut into, whether by the action of hydrofluoric acid, the cntter's wheel, or a blow, the entire mass will be broken up, as at b c d. Until the surface is injured, the drop pos sesses extraordinary strength. These drape, commonly known as Prince Ruport's drops, afford an excellent illustration of the behaviour of glass too suddenly cooled. The exterior rapidly contracts, whilst the interior molecules are still in a state of mutual repulsion. The repellent molecules are only resisted by the intense solidity of an outer crust ; so soon, therefore, as this cruet is weakened, the entire mass is disin tegrated. Attempts have been made of late years to utilize the solidity and strength undoubtedly possessed by the crust of suddenly cooled glass. This is the principle of M. de la Bastie'e invention, to which further reference will be made.
In the manufacture of glass for all ordinary purposes, special provision has to be made to facilitate gradual cooling. This may be effeoted in either of two ways :—(1) By placing the manufactured glass, whilst still het, in a closed oven or kiln, and allowing the fire of the kiln gradually to die out ; (2) by exposing the manufactured goods to a permanent fire, and then gradually withdrawing them. The process of gradual cooling is technically known as " annealing." The kiln system is probably the better, especially where heavy goods have to be annealed. Its disadvantages are (1) the length of time required to effeot complete cooling, in some cases amounting to 1-2 weeks ; (2) the space ocoupied by the kilns, as there must always be at least two ; (3) the discontinuity of the process. The withdrawal system is especially suited to flint-glass works, where goods are produced rapidly, and it is essential that the cooling process shall be continuous.