“‘When one places one of these mirrors facing the sun, and causes it to reflect, upon a very near wall, the image of its disk, one sees distinctly appear therein the ornaments or characters which exist in relief upon the back. Now the cause of this phenomenon, which arises from the distinct employment of fine copper and of crude copper. If on the back of the mirror one has produced, by casting it in a mould, a dragon arranged within a circle, one engraves deeply on the face of the disk an exactly similar dragon. Next, one fills the deeply-chiselled cuts with a somewhat baser copper; then one incorporates this metal with the first, which ought to be of a finer quality, by submitting the mirror to the action of fire; after which one flattens and smoothens down the face of the mirror, and spreads over it a slight layer of lead (tin?). “‘When one turns toward the sun the polished disk of a mirror so prepared, and reflects its image upon a wall, it distinctly presents bright tints and dark tints, which come, these from the purest parts of the copper, those from the baser parts.’ “Ou-tseu-hing, to whom we owe the preceding explanation, tells us that he has seen a mirror of this sort broken into fragments, and that he recognized for himself the accuracy of his description.” Ayrton quotes from a Japanese work, the _Shim-pen-kamakura-shi_, or New Collection of Writings about Kamakura, a description of a temple-mirror, which when looked at obliquely shows the face of a Buddhist priest, not resembling in the least the raised ornament on the back (see p. 45, _supra_). The same authority refers to the _Kokon-i-to_, or Genealogy of the Old and New Physicians, for an alleged process of producing magic effects on mirrors by treating the surface with a peculiar paste. The recipe is as follows: “Take ten parts of _shio_ (gamboge), one of _funso_, and one of _hosha_ (borax). Powder them thoroughly, and mix them to the consistency of a paste with a little dilute glue. If any pattern be drawn on the surface of a mirror with this paste, and then allowed to dry, the pattern will be seen, even after polishing, if looked at obliquely.” It appears that this process fails in reality to give any result. The process of inlaying described by Ou-tseu-hing is also an error. The magic effect is certainly not produced in this way. Probably he was misled by the circumstance that flaws in the bronze castings are sometimes filled by the insertion of soft copper beads. [Illustration] [Illustration] APPENDIX IV. ANALYSES OF JAPANESE MIRRORS. Mm. Champion and Pellet (_Industries de l’empire Chinois_, p. 64) give the following composition for Chinese mirrors: Copper 50·8 Tin 16·5 Zinc 30·5 Lead 2·2 ------ 100·0 ------