232 Proceedings of the Royal Irish Academy. 



do. Assuredly it was not an alphabet for literary purposes which he was 

 inventing; as such it would have been abandoned within a week, as being 

 totally impracticable. You have only to write a short sentence or a line of 

 poetry in Ogham to be convinced of this. The grouping of the scores from 

 one to five shows that he was inventing a system of finger signs, whereby 

 secret communications could be made, after the manner of the deaf and dumb 

 alphabet, by those initiated. For this purpose he wanted an alphabet the 

 number of whose signs was a multiple of five. Now the form of the Bonian 

 alphabet at his disposal contained eighteen signs. This we infer from the 

 omissions in the Ogham. These eighteen signs were 



ABCDEGHILMNOQESTVZ 



— the V having of course both a vocalic and a consonantal value, as we 

 assume in the inscription before us, where the second W, taken as = UV, 

 inverts into TU. 



To make his alphabet an even multiple of five he first added two signs, 

 for consonant V and for NG. Thus he had 



A B C D E G H I L M N Q E S T U Z V NG. 



He then separated out the vowels ; and this, together with the subsequent 

 distinction which he observed of broad and slender vowels, shows that we 

 have to do with the work of a grammaiian — 



AOUE I: BCDGHLMNQBSTZVNG. 



He next, as was long ago noticed by the late Eev. E. Barry, took as one 

 of his groups the initials of the first five numerals, as they would have been 

 in the Irish of his time. These were H, D, T, C, Q. Of course, the numeral 

 " four" should begin with Q as well as " live," but in the latter it is followed 

 by a strongly emphasized ' u,' which perhaps enhances its quality. At any 

 rate, the alphabet now stood thus — 



I |i in i'i| inn 



AOUEI :HDT C Q:BGLMNESZVNG. 



Of the remaining letters he began at the beginning, and took every 

 second — 



AOUEI : HDTCQ : B L N S V : G M E Z NG. 



Next he rearranged the remaining group, beginning with M (probably 



