Greek seals were normally attached to rings[782]. The definition of a coin as a piece of sealed metal was started by Burgon, who used it as an argument for the religious character of coin types and described the seal as “the impress of the symbol of the tutelar divinity of the city[783].” Macdonald very rightly rejects the theory, so widely current before the appearance of Ridgeway’s _Metallic Currency_, that all coin types had a religious origin. But he himself proceeds to hang on to the seal theory views of his own that are equally untenable, at least in the sweeping form in which he states them. He assumes that the seal must be always that of a state or king or magistrate, and that the device is usually heraldic. The latter point does not concern us here[784]: the former does, since it is incompatible with the views of Babelon and Lenormant that the earliest coins were not state issues. But the seal theory does not depend on the nature either of the seals or of the sealers. There is no reason why we should not accept Babelon and Lenormant on the private character of the earliest coins simultaneously with Burgon and Macdonald on the character of the first coin types as being simply seals. Indeed some of the evidence collected by Macdonald positively suggests that we should. In the fifth and fourth centuries B.C. at Athens public property was stamped with the public seal, τῷ δημοσίῳ σημάντρῳ[785]. The same practice was followed in the fifth century at Samos[786], and perhaps at Syracuse[787]. This use of public seals to stamp public property suggests corresponding private seals similarly used. At this point the literary evidence fails us, but archaeological documents come to our aid. Metal was not the only material that the ancients stamped in this way. Wine jars, bricks, and tiles were similarly stamped. Thousands of these stamps have come down to us, and many of them bear city symbols and a magistrate’s name. But they often bear also the name of the maker, and the Danish scholar Nilsson[788], who has made a thorough study of them, is inclined to regard them as essentially the seals of private manufacturers. The magistrate’s name may merely indicate the date: the state symbol may mean that the state was the consignee, or that the manufacturer enjoyed some sort of state protection, or was subjected to state taxation and control. If then such was the history of the stamp as applied to bricks and wine jars, we have a further reason for thinking that stamps as applied to the precious metals may have originally been put on them by private owners. May we not here have the true explanation of the marvellous seal of Gyges’ ring? May not the owner of this ring have been the first person to use his signet for stamping coins of metal, and may not this fact be the origin of the stories about its marvellous powers?