^{page 271 note 1} Cf. Sayce, , writing about the same time in this Journal (10, 1931, p. 784)Google Scholar: “the imitation of [the Phœnician or Sinaitic-Phœnician letters] in the cuneiform script [of Ras Shamra].”

^{page 271 note 2} Cf. the suggestion made0 by Virolleaud, M. as early as 1929: “L'alphabet de Ras Shamra est-il plus ancien que celui des Phéniciens ? N'en est-il, au contraire, qu'une imitation ou une adaptation?” (Syria, 1929, p. 310Google Scholar.)

^{page 273 note 1} As to the prototypes of the other nineteen signs I find myself in agreement, broadly speaking, with Sprengling and Olmstead as regards Nos. 15, 25, 28 (Gaster leaves these unexplained), and with Gaster as regards Nos. 3, 6, 12, 16, 18, 23. Concerning the prototype of the remaining ten we agree to a great extent.

^{page 274 note 1} In m the zigzag has been simplified into a single wedge. The formula applies to all the rest of the suggested eighteen equations if as is probable the y of six wedges was derived from a form a little younger than that of Gezer and the fuller forms of n and r from forms a little nearer to the Sinaitic than the n and r (?) of Ain Shems. Sinaitic forms in the Table are from Butin, , Harvard Theol. Rev., 1932, 129Google Scholar: Phœnician from Ahiram, except ṣ (wanting there) from Abiba'l (No. 22) and the arrow of Rouisseh (No. 23). In Nos. 9, 16, and 5, z, m, and the doubtful h are given, rather differently from Grimm, as I see them in the published copy and photograph of the Ain Shems ostrakon.

^{page 274 note 2} The second specimen of samek in the Table is Butin's unidentified sign xxiii from Sinai, No. 357, where it would make the proper name sm' (as in Sabæan).

^{page 275 note 1} The similar (or connected ?) adaptations of aleph and hē in Greek suggest that 'ain may have, as in Greek, been occasionally used for o; and, in fact, Bauer has recently noticed that there is a probable example of this use in ṣp'n = ṣpn, ṣapōn, Syria, 1934, p. 154 (OLZ., 1935, 129 f.).

^{page 276 note 1} For some of the rare examples see Syria, 1931, p. 194, n. 1; 1934, p. 79, 21, and 32; p. 245, 6, and 15.

^{page 276 note 2} Syria, 1934, p. 149.

^{page 277 note 1} In this case the duplication might well be derived from variants that already existed in the non-cuneiform writing; but all the other duplicates could be sufficiently explained by secondary developments within the cuneiform alphabet. If I am not mistaken the major mythological texts contain no duplicate forms except those for š/ś (and the occasional variant to l [15a]). The fact would be favourable to the hypothesis that No. 26b was not, originally at least, merely a graphic variant. On these signs see Syria, 1931, p. 197, n. 2; Bauer, , OLZ., 1935, 131Google Scholar; Syria, 1934, p. 306, 17; p. 330, 20. No. 26a corresponds both with š and ś, but No. 26b could always, so far as I have observed, be equated to ś. [The observation has already been made by Ginsberg, H. L., JRAS., 1935, p. 45Google Scholar.]

^{page 277 note 2} So p. 135. P. 140, “between S. III and Phœnician”: misprint?