unauthorized inspection, the old method of tying and sealing diptychs was reproduced as nearly as the change of material allowed. A strip cut from the lower edge of the parchment, but not detached, took the place of the cord formerly tied round the diptych; it was passed round the folded document, and the seal was applied over it as it had been over the cord. The seal thus affixed to a substance of little weight did not require such precautions to protect it from accidental damage as we infer in the case of Roman diptychs, and actually see in that of the very interesting wooden tablets of the Kushana period (roughly, first to third century a.d.) discovered in Turkestan by Sir M. A. Stein, with intact seals bearing heads from late Greek intaglio gems. To protect the seals upon their diptychs the Romans appear to have used small flat bronze capsules with hinged lids and holes in the sides through which the cord passed; the impression was made in the capsule upon wax enclosing the cord, and the lid was then closed. The writers of the Kharoshti letters in Turkestan adopted a simpler, but not less practical method. One leaf of their tablets was made thicker than the other across the middle of its outer side, and in this thickened part was sunk a cavity for the clay in which the string was secured by impression of the intaglio; by this means the device, which was below the level of the surrounding wood, was effectually guarded against injury. In view of the non-survival of intact Greek and Roman diptychs, the discovery of these inviolate documents in the sands of Central Asia possesses exceptional interest.
A second instance of continuity is furnished by the history of the hanging wax seal with a different impression on each face, like a coin. Mr. Poole produces evidence to show that this type, which served for authentication rather than for the security of documents, was suggested by the leaden papal bull, which also had an obverse and reverse, and was in like manner traversed by its cord. The use of lead for seals was general in the nearer East, and the Byzantine series covers many centuries; lead was probably chosen as the most effective substitute for wax, which melts too easily in a hot climate, or for clay, which is friable and easily damaged. The papal lead bull was in use in the seventh century, if not before, and the earliest examples of similar lead seals employed by secular rulers in the West were no doubt inspired by it; the author notes that the earliest of these at present known is that of Kenulf, or Coenwulf, king of Mercia (a.d. 800-10), which is to be seen in the British Museum. But some two hundred years passed before it occurred to any one that wax might be employed in place of lead for the pendent 'coin seal' on a royal charter, and it is interesting to learn that the change appears to have been made in England by Edward the Confessor; at any rate the oldest royal charter bearing a pendent 'coin seal' is one of his: on private charters the change was made earlier, as is proved by a document of a.d. 860-6 at Canterbury, which is reproduced by Mr. Poole in a full-page illustration. The reference to Kenulf's lead bull suggests the reflexion that wax for seals in England must have come into general use soon after his time, since the earliest of the Anglo-Saxon matrices in the British Museum, that of Æthelwald of Dunwich, is only some fifty years later in date. The reader will find in these pages a great deal of useful information as to the papal bull and its types, as to the
