The object-glass of the great telescope was cast by M. Mantois, famous as the manufacturer of large lenses. The glass used was boiled and reboiled many times to get rid of all bubbles. Then it was run into a mould and allowed to cool very gradually. A whole month elapsed before the breaking of a mould, when the lens often proved to be cracked on the surface, owing to the exterior having cooled faster than the interior and parted company with it. At last, however, a perfect cast resulted. M. Despret undertook the even more formidable task of casting the mirror at his works at Jeumont, North France. A special furnace and oven, capable of containing over fifteen tons of molten glass, had to be constructed. The mirror, 6-1/2 feet in diameter and eleven inches thick, absorbed 3-3/4 tons of liquid glass; and so great was the difficulty of cooling it gradually, that out of the twenty casts eighteen were failures. The rough lenses and mirror having been ground to approximate correctness in the ordinary way, there arose the question of polishing, which is generally done by one of the most sensitive and perfect instruments existing-the human hand. In this case, owing to the enormous size of the objects to be treated, hand work would not do. The mere hot touch of a workman would raise on the glass a tiny protuberance, which would be worn level with the rest of the surface by the polisher, and on the cooling of the part would leave a depression, only 1-75,000 of an inch deep, perhaps, but sufficient to produce distortion, and require that the lens should be ground down again, and the whole surface polished afresh. M. Gautier therefore polished by machinery. It proved a very difficult process altogether, on account of frictional heating, the rise of temperature in the polishing room, and the presence of dust. To insure success it was found necessary to warm all the polishing machinery, and to keep it at a fixed temperature. At the end of almost a year the polishing was finished, after the lenses and mirror had been subjected to the most searching tests, able to detect irregularities not exceeding 1-250,000 of an inch. M. Gautier applied to the mirror M. Foucault’s test, which is worth mentioning. A point of light thrown by the mirror is focused through a telescope. The eyepiece is then moved inwards and outwards so as to throw the point out of focus. If the point becomes a luminous circle surrounded by concentric rings, the surface throwing the light point is perfectly plane or smooth. If, however, a pushing-in shows a vertical flattening of the point, and a pulling-out a horizontal flattening, that part is concave; if the reverse happens, convexity is the cause.