The piece of tripoli which is to be used in forming the globules, should be flat on one side, and so large that it may be handled conveniently, and protect the fingers from the flame. A piece four or five inches long, and three or four inches thick, will answer very well. The best tripoli for this purpose is of a white colour, with a fine grain, heavy and compact, and which, after it has been calcined, is of a red colour. This kind resists the fire best, is not apt to break when calcined, and the melted glass does not adhere to it. To calcine this tripoli, cover it well all round with charcoal nearly red hot, leaving it thus till the charcoal is quite cold; it may then be taken out. Let several hemispherical cavities be made on the flat side of the tripoli; they should be of different sizes, nicely polished, and neatly rounded at the edges, in order to facilitate the entrance of the flame. The large globules are to be placed in the large cavities, and the minuter ones, in the small cavities. The holes in the tripoli must never be touched with the finger. If it be necessary to clean them, it should be done with white paper; the larger globules may be cleaned with wash leather. The glass rods should be of various sizes, as of ¹⁄₁₀th, ¹⁄₂₀th, ¹⁄₃₀th of an inch in diameter, as clean and free from specks and bubbles as possible. TO MAKE SMALL GLASS MICROSCOPIC GLOBULES. Take two rods of glass, one in each hand, place their extremities close to each other, and in the purest part of the flame; when you perceive the ends to be fused, separate them from each other; the heated glass following each rod, will be finer, in proportion to the length it is drawn to, and the smallness of the rod; in this manner you may procure threads of glass of any degree of fineness. Direct the flame to the middle of the thread, and it will be instantly divided into two parts. When one of the threads is perfectly cool, place it at the extremity of the flame, by which it will be rendered round; and, if the thread of glass be very fine, an exceeding small globule will be formed. This thread may now be broke off from the rod, and a new one may be again drawn out as before, by the assistance of the other glass rod.