e. The evidence from the names of geographical localities in Gaul, both ancient and modern, goes the same way: Nantuates, Nantouin, Nanteuil, are derived from the Welsh nant=a valley, a word unknown in Gaelic.

f. The evidence of certain French provincial words, which are Welsh and Armorican rather than Erse or Gaelic.

g. An inscription on an ancient Celtic tablet found at Paris, A.D. 1711, and representing a bull and three birds (cranes), is TARWOS TRI GARANOS. Now, for the first two names, the Gaelic affords as good an explanation as the Welsh; the third, however, is best explained by the Welsh.

Bull = tarw, Welsh; tarbh, Gaelic.
Three = tri, Welsh; tre, Gaelic.
Crane = garan, Welsh; corr, Gaelic.

[§ 139]. The Pictish most probably Cambrian.—The evidence in favour of the Pictish being Cambrian rather than Gaelic lies in the following facts:—

a. When St. Columba preached, whose mother-tongue was Irish Gaelic, he used an interpreter—Adamnanus apud

Colgarum, 1, 11, c.32. This is a point of external evidence, and shows the difference between the Pict and Gaelic. What follows are points of internal evidence, and show the affinity between the Pict and Welsh.

b. A manuscript in the Colbertine library contains a list of Pictish kings from the fifth century downwards. These names are not only more Celtic than Gothic, but more Welsh than Gaelic. Taran=thunder in Welsh. Uven is the Welsh Owen. The first syllable in Talorg (=forehead) is the tal in Talhaiarn=iron forehead, Taliessin=splendid forehead, Welsh names. Wrgust is nearer to the Welsh Gwrgust than to the Irish Fergus. Finally, Drust, Drostan, Wrad, Necton, closely resemble the Welsh Trwst, Trwstan, Gwriad, Nwython. Cineod and Domhnall (Kenneth and Donnell), are the only true Erse forms in the list.

c. The only Pictish common name extant is the well-known compound pen val, which is in the oldest MS. of Bede peann fahel. This means caput valli, and is the name for the eastern termination of the Vallum of Antoninus. Herein pen is unequivocally Welsh, meaning head. It is an impossible form in Gaelic. Fal, on the other hand, is apparently Gaelic, the Welsh for a rampart being gwall. Fal, however, occurs in Welsh also, and means inclosure.

The evidence just indicated is rendered nearly conclusive by an interpolation, apparently of the twelfth century, of the Durham MS. of Nennius, whereby it is stated that the spot in question was called in Gaelic Cenail. Now Cenail is the modern name Kinneil, and it is also a Gaelic translation of the Pict pen val, since cean is the Gaelic for head, and fhail for rampart or wall. If the older form were Gaelic, the substitution, or translation, would have been superfluous.