[274]. The old form of the name Atholl is Athfhotla; and in the Prophecy of St. Berchan, one of the kings, who represents Kenneth M‘Alpin, is said to have died for bruinnibh Eirenn, on the banks of Erin. He died at Forteviot, on the river Earn.
[275]. Words and Places, by the Rev. Isaac Taylor, p. 258. This argument appears to have been first used by Mr. Kemble in his Saxons in England, vol. ii. p. 4, but his line of demarcation is quite different from Mr. Taylor’s. He says—‘The distinctive names of water in the two principal languages appear to be Aber and Inver.’ He then gives a list of seven Abers in Wales, and in Scotland eleven Abers on the south-east side of his line, and twelve Invers on the north-west; but the contrast is produced by simply omitting the Invers which are on the same side with the Abers, and the Abers which are to be found among the Invers. Mr. Taylor adds—‘The process of change is shown by an old charter, in which king David grants to the monks of May “Inverin qui fuit Aberin.” So Abernethy became Invernethy, although the old name is now restored.’[restored.’] This is quoted without acknowledgment from George Chalmers, with the usual result of second-hand quotation, that of perpetuating error. The true reading in the charter is ‘Petnaweem et Inverin que fuit Averin;’ and it means in the ordinary charter Latin that these places formerly belonged to a person called Averin. Abernethy never became Invernethy. The two places are distinct from each other: Invernethy at the junction of the Nethy with the Earn, and Abernethy a mile farther up the river.
[276]. Diefenbach, in his Celtica, vol. i. p. 23, is of this opinion. He says, ‘Aber gehört völlig beiden Sprachaesten an.’
[277]. Mr. Bannister, in his Glossary of Cornish Names, has no Abers, but an Appledor.
[278]. Taylor, Words and Places, p. 232. With what success he attempts to make this out his list of Pens will show. Leaving out those in Dumfriesshire, Ayrshire, and Haddington, where there was originally a Welsh-speaking people, ‘we find,’ he says, ‘the Cymric form of the word in the Grampians,’ which is utter nonsense, ‘the Pentland Hills,’ which is a corruption of Petland Hills, as the Pentland Firth is of the Petland Firth, ‘the Pennguaul Hills,’ which have no existence, and ‘Pendrich in Perth,’ which is a corruption of Pittindriech. The whole of this part of Mr. Taylor’s work is tainted with phonetic etymology; e.g., he says, ‘From llevn, smooth, or from linn, a deep still pool, we obtain the names of Loch Leven, and three rivers called Leven in Scotland.’ The old form of this name Leven is ‘Leamhan,’ which means in Irish an elm-tree. The Welsh equivalent is Llwyfan.
[279]. Perhaps Pennan, the modern name of a headland at the Moray Firth, may be an exception, but we have not its old form.
[280]. Pette is the form of this word in the Book of Deer, and it appears to mean a portion of land, as it is conjoined with proper names, as Pette MacGarnait, Pette Malduib. It also appears connected with Gaelic specific terms, as Pette an Muilenn, ‘of the mill.’ With the article it forms Petten, or Pitten, as in Petten-taggart, termed in a charter of the church of Migvie (St. Andrews Chartulary, preface, p. 21) ‘terra ecclesiæ.’ It is Pettan t-saguirt, the priest’s land. In the same Chartulary (114) the ‘villula quæ dicitur Pettemokane’ is afterwards apparently called ‘domus cujusdam viri nomine Mochan.’ It is synonymous with Both, a dwelling, as we find Bothgouanan, near Elgin, has become Pitgownie, and Badfodullis, near Aberdeen, Pitfoddles. Dr. Stuart points out, in his introduction to the Book of Deer, p. lxxxiv., that Pit and Bal are frequently used indiscriminately.
[281]. As in Fothuirtabhaicht now Forteviot, Fothurdun now Fordun, Fothenaven now Finhaven.
[282]. These laws are taken from The Four Ancient Books of Wales, where the subject of the race and language of the Picts is fully discussed in Chapters VII., VIII., and IX. This has, of course, led to some repetition, and in one respect the author has been led to modify the views there stated. An examination of the old forms of the Cornish names in the Manumissions in the Bodmin Gospels, printed in the Revue Celtique, vol. i. p. 332, has led him to see that there is a British element in the proper names in the list of Pictish kings, and that that element is not Welsh, but Cornish.