It must represent the name of a district (or perhaps river), added to distinguish Bannaventa from other places of the same name. Muirchu’s statement that the _vicus_ was in or at Nentria (if this is the authentic reading) does not help us. But we can hardly doubt that he means by Nentria the same place which the Hymn of Fíacc means by _Nemthur_: l. 1, “Patrick was born in Nemthur.” This name might correspond to an old Celtic _Nemetoduron_ (a name preserved in _Nanterre_ near Paris).[363] This British Nemetodurum was, we may presume, in the same region as Bannaventa. The glossator in the oldest (eleventh century) MS. of the Hymn of Fíacc identified Nemthur with Ail Clúade, the Rock of Clyde, at Dumbarton. We are ignorant of his authority for this statement, which does not appear in any earlier source. The fact, however, that it is not inconsistent with the direct statements of earlier sources has procured credence for it. But it is inconsistent with the probabilities of the case. Patrick’s father was a decurion, and he must have lived in civilised Britain. We have no evidence that there were Roman towns with municipal constitutions in Strathclyde. The truth is that north Britain was little more than a large military frontier. It is generally supposed that Theodosius in A.D. 369 restored Roman rule, which had fallen back in the north as far as the Wall of Antoninus, and that the district which he recovered (recuperata provincia, Ammianus, 28, 3, 7), and which was renamed Valentia (by Valentinian, in compliment to his brother Valens), included the country between the Walls of Hadrian and Antonine. There is, strictly speaking, no direct authority for this conclusion; Ammianus does not indicate the position of Valentia. The supposition that it was in the north, and that Theodosius restored fortresses as far as the line of the northern wall, is, however, not improbable. But there is no probability that it was colonised[364] or became in the last half century of Roman rule anything more than a military district. The Rock of Clyde, at the extreme end of the Northern Wall, is the last place we should expect to find the _uillula_ of a Roman decurion; and the opinion that the home of Calpurnius was in that remote spot cannot be accepted without better evidence than an anonymous statement which we cannot trace to any trustworthy source.