P. 14.—Gaelic settlements in S.-W. Britain. Sources: Cormac’s _Glossary_, s.v. mogeime (ed. Stokes); the text (of which an extract was printed by Zimmer in _Nennius Vindicatus_, p. 85) from Bodleian MSS. Rawlinson, B. 502 f. 72 c., and Laud 610 f. 100 a. 1, edited with translation by Kuno Meyer (_Y Cymmrodor_, 14, 101, _sqq._). Cp. also _Historia Brittonum_, § 14, p. 156, ed. Mommsen. See Zimmer, _ib._ on these settlements. He determines the date of the composition of the Rawlinson-Laud document as about A.D. 750. The question as to the survival in West Britain of the descendants of an ancient pre-British Goidelic population (maintained by Professor Rhŷs, combated by Professor K. Meyer, see _Transactions of Hon. Soc. of Cymmrodorion_, 1895-6, p. 55 _sqq._) does not affect the reality of a later Goidelic settlement in historical times. On the Dessi cp. Rhŷs, _Studies in Early Irish History_, p. 56; _Origin of Welsh Englyn_, pp. 26, 73, 179 (in _Y Cymmrodor_ xviii. 1905); _Celtic Britain_ (ed. 3), p. 247 _sqq._ P. 10.—The statement that Man was never conquered by Rome might have to be modified if, as has occurred to me, the words of Tacitus at the beginning of _Agricola_ 24, _naue prima transgressus ignotas ad id tempus gentes crebris simul ac praeliis domuit_, record a descent of Agricola on that island. In the context an expedition to Caledonia seems excluded. [_Prima_ is unintelligible. It may be an instance of the common confusion of _un_ with _im_, _ima_ (for _una_) having been taken as an abbreviation of _prima_.] P. 14.—“Perhaps were conditions of military service.” A stone of Killorglin (now in the Dublin Museum) bears the ogam inscription GALEATOS, which Professor Rhŷs explains as genitive of Latin _galeatus_, a helmed soldier. He thinks that it was a name “first given to a Goidel who had worn the Roman _galea_, that is one who had served in the Roman army. For one cannot help comparing it with _Qoeddoran-i_ as connected with _Petrianae_, and the name _Sagittarius_, of which the genitive _Saggitari_ was found in ogam on a stone discovered at Burnfort, in the neighbourhood of Mallow, in Co. Cork. All three names are presumably to be explained on the supposition of Goidelic touch with Roman institutions, especially the military system; not to mention such Latin names as _Marianas_, _Latinus_, and _Columbanus_, or their significance, so to say, in this context” (reprint from _Proceedings of Royal Society of Antiquaries of Ireland_, Part i. vol. xxxii. 1902, p. 16). CHAPTER II P. 16.—Calpurnius: _Confession_, 357₄ (Armagh MS. has _Calpornum_; Cotton MS. _Calpornium_; Brussels MS. of Muirchu, _Cualfarni_, 495₇). Properly a _nomen_, but, like other _nomina_, adopted as a _cognomen_. As such, however, it does not seem to occur often. Cp. _C.I.L._ xiv. 570; iii. Suppl. Pars post. 14354²⁸; viii. 9157.