the terminus a quo, it is more hazardous to express any opinion as to the terminus ad quem. On the whole it seems probable that the deposit of these numerous objects took place at the close of the second period, in the withdrawal from Caledonia, prior to the establishment of the frontier line of Hadrian (i.e. circa 120 A.D.) from the Solway to the Tyne.
In the first place, such evidence as we possess suggests that no long period elapsed between the first and the second periods. Had the retirement been of a temporary nature, the articles that were hidden away would have been recovered. On the other hand, if the fort was abandoned in the early years of the second century and not reoccupied for more than a generation later—that is, till the advance of Lollius Urbicus—it is natural enough to suppose that all recollection of objects concealed may have passed away. Again, it will be noted that in none of the pits did fragments of Antonine pottery occur at the bottom in definite association with first century deposits. In the few cases when later pottery was met with, it was lying, as in Pit LIX, in the clay filling of the upper levels. Once more, the well at the Baths (Pit LVII) which is just one of those whose contents were such as to suggest that they had been thrown in on the occasion of some disaster, showed signs of having been left open at the abandonment to be filled up at the beginning of the succeeding period. A portion of the rim of a decorated bowl (Type Dragendorff 37), thick and with a heavy moulding suggestive of second century ware, was found at twelve feet, while on the surface the cobble foundation had been laid of double thickness and yet had subsided, as if; when it was placed there, the ground below it had been still soft from recent filling.
The relics from the well in the Principia (Pit I) must date from a subsequent abandonment. The very position of the well, standing as it did in its normal place in the outer courtyard of the last form that the Principia assumed, is of itself an indication that it belongs to a later period, and this is confirmed by the presence of a coin of Hadrian among the débris thrown into it. The strange medley it contained—the skulls, the broken armour, the ruins of demolished buildings, the buried altar—has its closest parallel in the spoil from the wells at Birrens and at Bar Hill.
115