is declared by the polished surface the small ring-stones of the sort retain in spite of all the injuries of time and wear. The composition appears to be the same with that of the Egyptian blue enamel, the "artificial cyanos" of Theophrastus,[1] so largely applied to the decorative productions of the national art. That such enamelling had for its object the making terra cotta and steaschist pass for true lapis lazuli is made evident by Ælian's notice that the "High Priest of the Egyptians used, when administering justice, to wear round his neck an image of the goddess Truth, carved in sapphirus."[2] The nature of this badge of office is abundantly attested by the existence of the numerous tablets in artificial cyanos, bearing figures of deities, and similarly intended for pendant jewels. Sir H. Davy found by experiment that the cyanos used in Roman fresco-painting could be exactly reproduced by fusing together, for the space of two hours, 15 parts pure carbonate of soda, 20 pulverised flint, and 3 copper filings. A similar mixture, the proportion of flint somewhat increased for the sake of hardening it, would produce a paste with all the qualities and appearance of the antique specimens. As Alexandria, upon the decay of Sidon, became the chief seat of the glass manufacture (one of its fabricants, Firmus, being actually wealthy enough to dispute the empire with Aurelian), it is more than probable that paste gems "vitreæ gemmæ e vulgi annulis,"[3] formed a large part of her exports; and that, for the species requiring it, the artificial cyanos (the invention of the country) was especially put into requisition.
The actual process of making paste gems can be briefly described, if minute technical details be omitted. The impression of the work to be imitated is taken in a mixture of fine tripoli and pipe-clay, rammed down in a little iron case of the dimensions required. This forms the matrix, which, after drying, is placed within the furnace, with a bit of glass of the proper colour laid upon it. This is watched until observed to become plastic, and then carefully squeezed down with an iron spatula coated with French chalk to prevent adhesion. After annealing, the glass, on removal from the matrix, presents an exact counterpart of the
