568 Proceedings of the Royal Irish Academy. 



I referred just now to the change of the name of Down from Dun- 

 leth-glaise to Dun-da-leth-glas; and this leads me to notice the singular 

 mistake sometimes made by eminent Irish scholars as to this word da 

 when prefixed to place-names. It has been nearly always taken for 

 the numeral two, and in the case of Dun-da-leth-glas, Jocelyn trans- 

 lates : " the Portress of the two broken fetters." Colgan again : " Arx 

 duarum mediarum catenarum." But the name in the earliest autho- 

 rities, " Fiace's Hymn," Tigernach, and the "Book of Armagh," is 

 Dun leth glaise, and there is nothing about two in it. The fact has 

 been overlooked that da in such names is frequently not the numeral, 

 but a middle-Irish form of do, meaning " of," and pronounced short ; 

 and the legend invented by Jocelyn to account for the word two 

 here may be dismissed as worthless. Another instance of the 

 practice of inserting da before place-names is that of Glendalough. 

 The present pronunciation of the name correctly represents the 

 primitive form Gleann da, Loch, found in Broccan's Hymn in praise 

 of Brigit (7th century), which could never have originated from da, 

 two ; and I may say that the local pronunciation of such names will 

 generally be found to be quite accurate. But at a later period, when 

 da was supposed to mean two, it became necessary to alter the word 

 loch to the plural locha, for of course the dual numeral could not be 

 placed in agreement with a noun in the singular. Hence the Four 

 Masters, departing from the early form, write the name Gleann da 

 loeJia, " The valley of the two lakes, and Latin writers, " Yallis duo- 

 rum stagnorum." This erroneous form of the name derived some 

 colour from the fact that there is a second and smaller lake at the 

 place. But this has nothing to do with the early name Gleann da 

 Loch, "the Yalley of the Lake," which refers only to the principal 

 lake ; and !Moore was certainly right when he wrote of one only — 



" . . . that lake whose gloomy shore 

 Skylark never warbled o'er. " 



The name Gleann da Loch has been a source of some perplexity to 

 philologists when found in early texts, as it did not seem easy to 

 explain how the dual came to be connected with a noun in the 

 singular, on the assumption that da meant two. Windisch proposed 

 to solve the difficulty by reading li7id (lakes) for loch, but this is a 

 violent emendation for which there is no MS. authority (Irische 

 Texte, p. 29). 



One does not like to differ from so eminent a scholar as Dr. 

 O'Donovan, and no doubt it is his authority which has influenced 



