Page:Archaeological Journal, Volume 3.djvu/278

250 "Stablesianorum equitum," (which are recorded to have been stationed at the mouth of the Garienis,) than the more insulated and aquatic situation of Burgh. According to Woodward's map of Roman Norfolk, the balance of dry land is very little in favour of either; but, from examinations of the country, I am inclined to believe that there was a free passage along the coast, from Caister to Happisburgh, and that, so far from the sea having receded in that line, it has nearly swallowed up two parishes, viz.. Little Waxham and Eccles, and greatly encroached upon others since the Roman period. The finding Roman coins at Eccles, which I have done, and some remains at Horsea, as I am credibly informed, prove that there was such a communication and access along the sea coast. But, however this might be, in one respect Camden decidedly has the advantage. The grandeur of the remains of the camp at Burgh favours his opinion; and, probably, this was the reason why, as Spelman says, "Camdeno Burgh arrisit;" whereas the existence of the walls of a camp at Caister near the sea, mentioned by Spelman, has been questioned, and it has been hinted that he confounded the comparatively modern dwelling-house of the Fastolfes, called Caister castle, with a Roman camp.

Now, in justice to Spelman, I will mention a few facts which I have observed. Fragments of sepulchral urns, of pottery, and of glass, are found very extensively and in great profusion in the parish; I traced them in a line from a quarter of a mile to the north-east, to three quarters of a mile to the south-west. They are found in the greatest abundance in a field on the west of the church, where tradition has fixed the Roman camp. In this spot one can scarcely use a spade without meeting with foundations of buildings, and broken pieces of Roman tiles lie scattered on the surface. The vault, or building of Roman tiles, described by the Rev. Thomas Clowes, was discovered here: Roman coins are found in different parts of Caister, but most abundantly in this field. As far as my observation goes, those found at Caister are more ancient than those found at Burgh, which are chiefly of the period of Constantine, whereas coins of M. Antoninus and of Commodus Antoninus are very common at Caister. Among them one of John Zimisces, who succeeded to the empire in A.D. 961, may deserve especial notice, as it appears remark-