We have some evidences that the old Romans practised, at least experimentally, the art of printing with ink. The British Museum has a stamp with letters engraved in relief, that was found near Rome, and which seems to have been made for the purpose of printing the signature of its owner. The stamp is a brass plate, about two inches long and not quite one inch wide. A brass ring is attached to the back of the plate which may have been used as a socket for the finger, or as a support when it was suspended from a chain or girdle. On the face of the stamp are engraved two lines of capital letters, huddled together in the usual style of all old Roman inscriptions, cut the reverse way, as it would now be done for printing, and enclosed by a border line. An impression taken from this stamp would produce the letters in the accompanying illustration, which may be translated, _the signature of Cecilius Hermias_. Of Cecilius Hermias we know nothing. He may have been a civic official who used this stamp to exempt himself from the trouble of writing, or a citizen who tried to hide his inability to write. [Illustration] If this stamp should be impressed in wax, the impression would produce letters sunk below the surface of the wax in a manner that is unlike the impressions of seals. The raised surface on the wax would be rough where it should be flat and smooth. This peculiarity is significant. As this rough field unfitted it for a neat impression on any plastic surface, the stamp should have been used for printing with ink. [Illustration: An Old Roman Stamp. [From Jackson.]] The accompanying illustration is that of a brass printing stamp in the British Museum, which is preserved as a specimen of old Roman workmanship.[6] The letters were cut in relief, in reverse order, and with a rough counter or field. This roughness proves that it could not have been used to impress wax. [p038] Brass stamps of similar construction and of undetermined age have been frequently found in France and Italy. All of them are of small size, and contain names of persons only. [Illustration: Roman Stamps. [From Jackson.]] The illustrations annexed, of two engraved brass stamps of eccentric shapes, were also copied from the originals in the British Museum. As the letters are roughly sunk in the metal, and are not fitted for stamping in wax, it is supposed that the stamps were made for impression with ink. They are regarded as Roman antiquities, of undoubted authenticity, but the meaning of the inscriptions, the special purposes for which they were made, and the period in which they were employed, are unknown. The difficulty connected with the proper fixing of ink upon these stamps of brass, of which a subsequent notice will be made, is one of many causes which prevented the development of this experimental form of printing.