¹ It is possible that aisid like (ad)·ágathar is compounded with the prep. ad, which then is dropped whenever the stress falls on the first syllable. In any event, all the forms thus far noted are prototonic and consequently may belong under deuterotonic ad·ais.

² See HIL 1.10 s.v. (ad-)ágathar; Zur keltischen Wortkunde §185.

³ Note, however, aisim ‘I hate (recte “I fear”)‘, Fianaigecht 101.

⁴ For this and several of the ensuing references, see ZCP 18.306, §653; Zur keltischen Wortkunde §185; Fianaigecht 20, fn. 43.

⁵ Since the stressed vowel of no-da·ais rhymes with that of ais ‘back’, it is short—not long as marked in HIL 1.10 s.v. (ad-)ágathar; cf. Zur keltischen Wortkunde §185, fn. 2.

⁶ Compare nachit·ais, TBC (YBL) 50, 1325.

⁷ A Grammar of Old Irish 412, §666.

⁸ Ed. S. H. O'Grady, Silva Gadelica 1.72–82.

⁹ The Yellow Book of Lecan (facs. ed.) 173^b51.

¹⁰ The Book of Uí Maine (facs. ed.) 134^a31.

¹¹ MS Egerton 1782; cf. O'Grady 1.79. In the same prophecy (80) the future absolute 2d singular of cingid seems to recur. There it is spelled cichsi (cichsi, UM; ciesi, YBL).

¹² A study of the Irish ‘rhetorics’ will yield many other examples of the strong future which have not been recorded. Note in this connection on p. 80 in Bec mac Dé's prophecy íba 'thou shalt drink' for older íbae, and at·cí ‘thou shalt see’, which is corrupt for at·chichi, the readings of YBL and UM; cf. Thurneysen 402, 406.

¹³ To is the archaic form of the preverb do.

¹⁴ ZCP 13.103.

¹⁵ Altbu is obscure.

¹⁶ Entered by Pedersen, VGDKS 2.652, §849.2, under di-to-n-g-.

¹⁷ Among many instances illustrating the confusion between the preverbs di and do in Early Irish, observe, for example, di·toing, Triads, ed. K. Meyer 20, §159, for which, however, the variant reading do·toing is given.

¹⁸ Thurneysen 345, §534.

¹⁹ Ibid. 344, §533.

²⁰ Such at any rate is Kuno Meyer's opinion. In Cáin Adamnáin 43, he takes to·cuitchetar (20, §29) to be the perfect 3d plural of do·toing—not of tongid.

²¹ 2.53 s.v. (for-)laí.

²² Ériu 2.190, §58.

²³ Literally ‘weaves about’.

²⁴ For these perfective forms, see Ériu 9.41–2. That imb·fen, imm·fen also should have a deponential conjugation seems questionable; hence perhaps for ·imcoir read ·imcoe or ·imcoi, an emendation which finds some support in the fact that in all the examples so far recorded the initial consonant of the next word is an r.

²⁵ AL 4.126, 4.

²⁶ The subjunctive in Irish often expresses a command that need not be complied with at once.

²⁷ As resíu is followed normally by the ro-subjunctive, one might expect here ad·rogra in place of the ordinary subjunctive ad·gara, but exceptions occur; cf. Contr. Dict. Ir. Lang. s.v. resíu.

²⁸ Spelled farfia in Ml. 55^c16.

²⁹ Inp for imb likewise is found in inp·auch gl. ‘ago’, Thes. Pal. 2.360.

³⁰ H. Lewis and H. Pedersen, A Concise Comp. Celt. Gram. 401, §642. In the source cited by Lewis and Pedersen the c of ·rodcad, however, is lenited just as it also is in Ériu 12.168, 12.

³¹ Apparently forms of this verb in the active voice have not yet been noted.

³² Ériu 2.114, §39.

³³ Thes. Pal. 2.293, 6. Compare likewise sothcedach ‘fortunate’ and dothcedach ‘unfortunate’, AL 1.40, 11.

³⁴ Note, for instance, maid instead of maith ‘good’; cf. G. I. Ascoli, Il Codice Irlandese dell' Ambrosiana 2.374 f.

³⁵ Thurneysen 22, §30.

³⁶ See fn. 30.

³⁷ I am indebted to Dr. Osborn Bergin for this suggestion.

³⁸ Compare dotced (dothceth, UM) .i. di-toiced (di-thoiceth, UM), Anecd. 4.34, §409.

³⁹ R. Thurneysen, Zu irischen Handschriften und Litteraturdenkmälern 1.13, §1, 19.