^{page 74 note 1} It is perhaps better called a principle than a law, since it does not appear to operate with the strictness which we expect of phonetic laws. I have dealt with the question of the rhythmical basis of the form which was preferred toin ‘The Morphology of the Greek Comparative System: its Rhythmical and Repetitive Features“, A.J.P. vol. lxx (1949), pp. 159–70Google Scholar. For that reason I prefer not to discuss it more preferred fully here.

^{page 75 note 1} The exact nature of the termination of this word (a contraction of-ca?) and its accentuation are hard to explain. But Bechtel reminds us, Griech. Dial, ii, p. 295, that it is often difficult to discover the exact original form of the Laconian words in Hesychius, and we may dispense with that labour here.

^{page 75 note 2} Prof. Richardson has pointed out to me that a form with lengthened vowel could be ruled out by similarity to (from).

^{page 75 note 3} Emended by Porson to περαινερνιον. But the change is not necessary, and the second deftion of Photius seems to show that he did not really know what sort of ornament it was, and that he was hazarding a guess in the word is also found in Pollux, in a list of various see Kock, Com. Adesp. 1115.