180 Tlie Bishop of Bristol [May 19, 



used in its general sense for " of the race of," not in its limited sense 

 for " son of." I do not remember any Ogam inscription which uses 

 the better known formula for " of the race of," Ua or O', i.e. " grand- 

 son of." As far as I have been able to see, the difference in principle 

 between the formation of a familv name from Mac and from the 

 more distant 0' would provide a very interesting subject of investiga- 

 tion. It is probably not known to all English readers that O'Neill, 

 for instance, is properly only used as the name of a man ; an Irish 

 woman in an English hospital is addressed by an Irish physician as 

 Mary Nin Neill, in the feminine, not Mary O'Neill in the masculine. 

 Other examples of the patronymic Decedda are as follows : In 

 another Kerry inscription, maqi deccedda ; at Dunbel, Kilkenny, 

 maqi decquedda; at Ballycrovane, co. Cork, maqi deccedda; at Bal- 

 lintaggart, co. Kerry, maqi deccedda ; at Killeen Cormac, maqi 

 ddecceda ; these are all in Ogam script : at Penrhos Lygwy, in Latin 

 characters, hie iacit maccu deceti ; and at Buckland Monachorum, 

 near Tavistock, Sarini fili macco decheti. 



The cross on this stone is, no doubt, a great deal later than the 

 ogams. The ogams on the two arrises are both read the same way, 

 instead of up one side and down the other. This is evidence that 

 they are two separate inscriptions, not one continuous phrase ; and it 

 suggests that the stone was originally in a horizontal position, per- 

 haps as the lintel of a rude entrance to the pagan cilleen. 



Fortwilliam, co. Kerry. Fig. 14 is a stone in the vestibule of 

 the Library of Trinity College, Dublin. It is in excellent preserva- 

 tion, and the ogams are beautifully bold and regular and clear ; and 

 yet it presents considerable difficulties. The Irish antiquarians 

 appear to be agreed about the readings, and my rubbing and exam- 

 ination falls in with their view, though I have felt it necessary to 

 mark two of the scores with a query. The chief difficulties are (1) 

 that it is not easy to distinguish vowels from consonants, whether 

 by size or by relative position; and (2) that there are rows of vowel 

 scores without spaces to separate them into their proper numbers. 

 My first score is quite as large as a consonant, and yet, with the 

 Tinnahally examples before us, we seem compelled to take it as the 

 vowel a. This gives anm fedlliostoi macui eddoini. The patronymic 

 eddoini need not give us any trouble, it is probably the same as 

 Aedan. The fedll, again, need not trouble us ; there are other names 

 of this character, as Feidhlimidh, or Fedhlimidh, our Phelim. The 

 oistoi, or iostoi, appears to be too much for the science of interpre- 

 tation. The only point on which I feel it possible in my ignorance 

 to fasten is the group of four scores which is rendered st. There 

 is no doubt that the Book of Ballymote assigns to it this value of st, 

 though, so far as I know, it has not been found on any ogam stone. 

 Considering the recognised fact that care was not taken by the author 

 of this inscription to separate his lines of dots into their proj)er 

 groups, I venture to suggest that we should read these four scores as 

 two groups of two scores ; that is, as gg, not with the usual pronun- 



