^{page 148 note 1} The Greek prototype of the Latin alphabet is probably to be seen on the Numasioi-inscription (Journ. Phil. xvi. 196). It bears a striking resemblance to that of Veii (cf. the table in Müller's Handb. vol. i.) and this is true of the Umbrian and Etruscan alphabets. We may therefore, I think, fairly regard the Greek alphabet in its Veian form as the parent of Italic alphabets proper. Geographical reasons also render this probable.

^{page 148 note 2} The answer to this may be that ^*cānimen ^*gēnimen were the forms; but this would need great hardihood.

^{page 148 note 3} L. L. vii. 3. p. 70 (Steph.). The passage is corrupt but I think it cannot be reasonably doubted that this is the sense. Curiously enough he goes on to derive cano from this same root cas.

^{page 149 note 1} κερω may also belong to this root. —κερω which becomes κερω as κτανω etc. became κτενω etc. Cf. —Lit. skiriu. The root was therefore s-ka^xr, and gives also -corium, and -scortum. The guttural may have been velar, see Etyma Latina, s.vv.

^{page 149 note 2} Compare Brugmann's similar explanation of verna, Grds. ii. p. 137.

^{page 149 note 3} It is quite probable that the simple numeration from 1 to 24 preceded the more elaborate decimal system.

^{page 149 note 4} These were either one suffix by Ablaut, or three independent suffixes.

^{page 149 note 5} So regula implies linca, afterwards transferred to the instrument.