According to some chroniclers, the Emperor Heraclius had already let loose the Shut-up Nations to aid him against the Persians, but it brought him no good, for he was beaten in spite of their aid, and died of grief. The theory that the Tartars were Gog and Magog led to the Rampart of Alexander being confounded with the Wall of China (see _infra_, Bk. I. ch. lix.), or being relegated to the extreme N.E. of Asia, as we find it in the Carta Catalana. These legends are referred to by Rabbi Benjamin, Hayton, Rubruquis, Ricold, Matthew Paris, and many more. Josephus indeed speaks of the Pass which Alexander fortified with gates of steel. But his saying that the King of Hyrcania was Lord of this Pass points to the Hyrcanian Gates of Northern Persia, or perhaps to the Wall of Gomushtapah, described by Vámbéry. Ricold of Montecroce allows two arguments to connect the Tartars with the Jews who were shut up by Alexander; one that the Tartars hated the very name of Alexander, and could not bear to hear it; the other, that their manner of writing was very like the Chaldean, meaning apparently the Syriac (_anté_, p. 29). But he points out that they had no resemblance to Jews, and no knowledge of the law. Edrisi relates how the Khalif Wathek sent one Salem the Dragoman to explore the Rampart of Gog and Magog. His route lay by Tiflis, the Alan country, and that of the Bashkirds, to the far north or north-east, and back by Samarkand. But the report of what he saw is pure fable. In 1857, Dr. Bellew seems to have found the ancient belief in the legend still held by Afghan gentlemen at Kandahar. At Gelath in Imeretia there still exists one valve of a large iron gate, traditionally said to be the relic of a pair brought as a trophy from Derbend by David, King of Georgia, called the Restorer (1089–1130). M. Brosset, however, has shown it to be the gate of Ganja, carried off in 1139. (_Bayer_ in _Comment. Petropol._ I. 401 _seqq._; _Pseudo-Callisth._ by _Müller_, p. 138; _Gott. Viterb._ in _Pistorii Nidani Script. Germ._ II. 228; _Alexandriade_, pp. 310–311; _Pereg._ IV. p. 118; _Acad. des Insc. Divers Savans_, II. 483; _Edrisi_, II. 416–420, etc.) NOTE 4.—The box-wood of the Abkhasian forests was so abundant, and formed so important an article of Genoese trade, as to give the name of _Chao de Bux_ (Cavo di Bussi) to the bay of Bambor, N.W. of Sukum Kala’, where the traffic was carried on. (See _Elie de Laprim._ 243.) Abulfeda also speaks of the Forest of Box (_Shará’ ul-buḳs_) on the shores of the Black Sea, from which box-wood was exported to all parts of the world; but his indication of the exact locality is confused. (_Reinaud’s Abulf._ I. 289.)