This is not the case when glass really silvered by successful chemical processes is used, and when pure metallic silver is precipitated on to the surface of the best and flattest plate glass. When Mr. Walker and myself commenced our experiments in March, 1879, the so-called “Patent Silvered Glass” was expensive and confined to moderate-sized pieces of plate glass. Our first care, therefore, was to construct a table that could be brought by screws to a perfect level, and one that would carry a plate of glass at least twelve feet six long by six feet eight wide. Such a plate being most carefully cleaned, and quite free from grease, was placed upon the table, and levelled by means of spirit levels, just as a plate of glass used for the old collodion process would be levelled, in order that the fluid should not run off at one edge, leaving the other comparatively dry; and now came the knotty point—Which was the best silvering process to use? On consulting the best records of this art, we found valuable information in the _English Mechanic_, Vol. xxi., No. 542. The reader will find the following process very successful if minutely carried out in all its technical details— TO SILVER GLASS. Prepare two solutions. 1. Argentic nitrate is dissolved in distilled water, and ammonia added to the solution till the precipitate first thrown down is almost entirely re-dissolved. The solution is filtered and diluted, so that 100 cc. contain one gramme of argentic nitrate. N.B.—100 cc. are equal to rather more than 3½ fluid ounces. 2. Two grammes of argentic nitrate are dissolved in a little distilled water, and poured into a litre of boiling distilled water. 1·66 gramme of Rochelle salt is added, and the mixture boiled for a short time, till the precipitate contained in it becomes grey; it is then filtered hot. The glass, having been thoroughly cleaned with (1) nitric acid, (2) water, (3) caustic potash, (4) water, (5) alcohol, and lastly distilled water, is to be placed in a clean glass or porcelain vessel, the side to be silvered being placed uppermost. Equal quantities of the two solutions are then to be mixed and poured in, so as to cover the glass. This should be done while the glass is still wet with distilled water. In about an hour the silvering will be completed. Then pour off the exhausted liquid, carefully remove glass, wash in clean water, rub off silver where deposited where not required, allow to dry, and varnish silvered side with any thin varnish which does not contract much in drying. The time required for the operation depends on temperature. If the solutions be warmed to about 30°C., the silver is deposited in a few minutes; but it is safer to use them cold. The inside of test tubes, bulbs, &c., are silvered by putting the solutions into them, no second vessel being then required. Throughout the whole operation the most scrupulous cleanliness is the grand essential. 100 cc. are equal to rather more than 3½ fluid ounces.