As during evaporation the fluid carried off by caloric is entirely lost, being sacrificed for the sake of the fixed substances with which it was combined, this process is only employed where the fluid is of small value, as water, for instance. But, when the fluid is of more consequence, we have recourse to distillation, in which process we preserve both the fixed substance and the volatile fluid. The vessels employed for evaporation are basons or pans of copper, silver, or lead, Pl. II. Fig. 13. and 15. or capsules of glass, porcellain, or stone ware, Pl. II. A, Fig. 1. and 2. Pl. III. Fig. 3 and 4. The best utensils for this purpose are made of the bottoms of glass retorts and matrasses, as their equal thinness renders them more fit than any other kind of glass vessel for bearing a brisk fire and sudden alterations of heat and cold without breaking. As the method of cutting these glass vessels is no where described in books, I shall here give a description of it, that they may be made by chemists for themselves out of spoiled retorts, matrasses, and recipients, at a much cheaper rate than any which can be procured from glass manufacturers. The instrument, Pl. III. Fig. 5. consisting of an iron ring AC, fixed to the rod AB, having a wooden handle D, is employed as follows: Make the ring red hot in the fire, and put it upon the matrass G, Fig. 6. which is to be cut; when the glass is sufficiently heated, throw on a little cold water, and it will generally break exactly at the circular line heated by the ring. Small flasks or phials of thin glass are exceeding good vessels for evaporating small quantities of fluid; they are very cheap, and stand the fire remarkably. One or more of these may be placed upon a second grate above the furnace, Pl. III. Fig. 2. where they will only experience a gentle heat. By this means a great number of experiments may be carried on at one time. A glass retort, placed in a sand bath, and covered with a dome of baked earth, Pl. III. Fig. 1. answers pretty well for evaporations; but in this way it is always considerably slower, and is even liable to accidents; as the sand heats unequally, and the glass cannot dilate in the same unequal manner, the retort is very liable to break. Sometimes the sand serves exactly the office of the iron ring formerly mentioned; for, if a single drop of vapour, condensed into liquid, happens to fall upon the heated part of the vessel, it breaks circularly at that place. When a very intense fire is necessary, earthen crucibles may be used; but we generally use the word _evaporation_ to express what is produced by the temperature of boiling water, or not much higher. SECT. IV. _Of Cristallization._