ROMAN EXCAVATIONS AT YORK 55 



was discovered in the course of our excavations of the fourth-century 

 defences on these sides. 



A clue to the answering of this question was found during the excava- 

 tion, referred to above, in the Canon's garden. Underneath the rampart 

 belonging to the Constantian fortress wall were found remains of several 

 apartments of a barrack block : under the floors were discovered potsherds 

 that dated the structure to the early part of the second century. The dis- 

 covery of this barrack block here presumed an earlier defence line outside 

 the fourth-century defences. In 1928, our last season of excavation, in 

 the garden of the Territorial Association and also in one of the gardens 

 belonging to the School for the Blind, lying outside the fourth-century 

 wall, we attempted to find some trace of the earlier defences. Although 

 in this we were not successful, it must be remembered that during this 

 last season scarcity of funds forced us to limit our operations very strictly. 

 It was impossible to make the extensive examination that may be necessary. 

 One significant discovery, however, was made — namely, that of a very 

 heavy Roman road pointing from the fortress towards the old ford of 

 the Ouse at Clifton. This road was not coming from the fourth-century 

 gate but from a point in the fourth-century wall some sixty yards south 

 of the gateway. The presumption is that this road led to the gateway 

 of the earlier defences on the western side. It is possible that the earlier 

 defences on the southern side were discovered in 1883, when a massive 

 Roman wall was found under the General Post Office in Lendal running 

 parallel with the later defences. 



It would seem to follow, then, that Constantius, at the beginning 

 of the fourth century, reduced the size of the fortress. At the same time 

 reconstruction took place within the fortress itself. In the summer 

 of 1928, in preparing for the construction of an underground lavatory 

 near to the south-east corner, it was found that barrack buildings had 

 been levelled and a building of different character had been erected, 

 probably, so far as the evidence went, at the beginning of the fourth 

 century. This finding has been confirmed, and the evidence for a 

 Constantian reconstruction on a large scale has been greatly strengthened 

 by the discovery in 1930, in digging cellaring for the Mail Coach Inn 

 in St. Sampson's Square, of extensive remains of a large bath-house. 

 Owing to the public-spirited generosity of the owners, the Tadcaster 

 Brewery, these remains have been preserved. This bath-house stands 

 on a site that in the earlier fortress would normally be covered by barracks. 

 The structure with its bonding courses and similarity to the work at 

 the south-west corner dates itself to the beginning of the fourth century. 

 As Mr. S. N. Miller suggests {Ro?nan York : Excavations of ig26-'^, 

 p. 98), the probable inference will be that the Roman garrison no longer 

 lived in the fortress in permanent quarters, and that this will lend support 

 to the view ' that the military authority in Britain followed the practice, 

 originally authorized by Severus, of allowing the legionaries to live a 

 family life in the settlements and cities outside the walls of their strong- 

 holds.' 



York almost certainly suffered damage during the troubles of A. D. 367, 

 when the sixth Legion must for a time have abandoned the fortress. 



