^{page 296 note 1} The farmer, Mr. C. V. Everall, notified Dr. A. W. J. Houghton, who took it into safe-keeping at his house, Oak Wood, Pulverbatch, near Shrewsbury, kindly provided the first-named writer with details, photographs, and a squeeze, and made it accessible for inspection.

^{page 296 note 2} The site of discovery is about 575 yards north-east of the north-east angle of the Forum at Wroxeter, grid ref. SJ 568091; see Kenyon, Kathleen M., Archaeologia, lxxxviii (1940), pi. lxviiiGoogle Scholar. For the line of the defences in this sector see V.C.H. Shropshire, i, 223, fig. 10, Atkinson Wroxeter, 330, pl. 72.

^{page 296 note 3} Dr. G. Webster (to R. P. W.) in providing details about the latest date for Roman artefacts on the site.

^{page 296 note 4} Mr. P. A. Barker has kindly provided the following note on a late timber building which he is reporting in Medieval Archaeology, vol. xii.

‘In 1966 a small area in the north-west corner of the baths insula at Wroxeter close to the north-south street was stripped, revealing, immediately under the topsoil, the plan of a timber and wattle building adjoining and lying parallel to the street. The northern half of the building had a clay floor with a sandstone hearth, and was separated by a narrow passage from a hard-standing made of broken roof-tiles set on edge. The southern end of the building had been destroyed by a large pit, but sufficient of the plan remained to show that it had been bow-sided, 12 feet wide at its maximum and at least 36 feet long. In form the building bears a remarkable resemblance to the longhouses familiar from later deserted village sites, but pottery included in its clay floor was of fourth-century date, as were the pottery and coins of the filling of the later pit. As no pottery from the period from about 400 to about 1100 has yet been recognised in Shropshire the absence of post-Roman pottery does not pre-elude a sub-Roman or post-Roman date for this building, though whatever its date its immediate antecedents appear to be Germanic rather than Celtic. It can be demonstrated that this is the last structure at this point in the town, and as it lies in the highest part of the field there is every reason to hope that the latest levels may be preserved else where.’

^{page 296 note 5} At the time of discovery the plough damaged two areas of the text, and a thin flake, J in. wide by ¾ in. long, broke away from the top right margin and removed the upper right tip of the terminal letter X.

^{page 297 note 1} For dating the inscription the form of the letters gives little help for they have been crudely fashioned with a pick. Such a style can be illustrated on several quarry-inscriptions, of which one (R.I.B. 1009, on the north side of the river Gelt) is datable to A.D. 207. So the style would not be out of place in a context of the late fourth or early fifth century. It continued to be used throughout the fifth and first half of the sixth century, and occurs in A.D. 540 on Nash Williams, E.C.M.W., pi. viii, fig. 84, no. 104.

^{page 297 note 2} In line 3 letter 1 cannot be A and x hardly fits, Letter 6 is probably L with sloping foot, for it is unlikely that one 1 was immediately followed by a second 1. Letter 8 has sustained damage. Two strokes linked at the foot make v; if this had been preceded by a short vertical stroke, now lost in damage, the complete letter could have been N.

^{page 297 note 3} Cf. M. A. O'Brien, Corpus Genealogiarum Hiberniae, i, 564.

^{page 298 note 1} Printing q for what was really a labialized velar stop, roughly qu in English quod; to print maququos would be rather misleading, as the sound was a single one, q^w, not kw.

^{page 298 note 2} On these names, see E. MacNéill in Proceedings of the Royal Irish Academy, xxvii c, 15 (1909), p. 365 f.

^{page 298 note 3} Cf. Mac-Cuilind, a spelling of this; O'Brien, op. cit., p. 681.

^{page 298 note 4} On these, see the present writer's ‘Notes on the Ogam Inscriptions of Southern Britain’ in Fox, C. and Dickins, B., The Early Cultures of North-West Europe (Cambridge, 1950), pp. 199Google Scholar ff.

^{page 298 note 5} On all these inscriptions, see Macalister, R. A. S., Corpus Inscriptionum Insularum Celticarum, i (Dublin, 1945), 306 ff.Google Scholar; Nash-Williams, V. E., The Early Christian Monuments of Wales (Cardiff, 1950)Google Scholar, passim; and the present writer, Language and History in Early Britain (Edinburgh, 1953Google Scholar; abbr. L.H.E.B.), chapter v and elsewhere.

^{page 298 note 6} See‘Notes on the Ogam Inscriptions’, pp. 203 ff.

^{page 298 note 7} See L.H.E.B., p. 154.

^{page 298 note 8} See Macalister, op. cit., no. 496.

^{page 299 note 1} See R.I.B. i, no. 721.

^{page 299 note 2} Op.cit., p. 12. Cf. L.H.E.B., pp. 158 ff.

^{page 299 note 3} Nash-Williams's identification of the Paulinus is by no means certain.

^{page 299 note 4} See L.H.E.B., pp. 128 f., 141 f.

^{page 299 note 5} L.H.E.B., pp. 139–41.

^{page 299 note 6} L.H.E.B., p. 84 f.

^{page 299 note 7} L.H.E.B., pp. 137, 139, 143.

^{page 300 note 1} l′ and n ′ are palatalized l and n, and l means a weak, indistinct variety of the short vowel in English hit.

^{page 300 note 2} L.H.E.B., p. 166 ff. One would have expected Cvnorigos (or -as; or perhaps Latinized -is) maqvi, etc.

^{page 300 note 3} L.H.E.B., pp. 117 ff.