Gaelic Language of the South of Scotland. 167 



recognised that rocli: is not an old word either in English or 

 Scotch, even if rocks were not conspicuously absent in the 

 little green hill. Marchidun is not exactly either Gaelic or 

 Welsh; marc, for horse, must have dropped out of Gaelic 

 rather recently, for marcaire is still the word for horseman, 

 and mare is still used in Welsh ; while dun, as it still stands, 

 is more like Gaelic than Welsh. I am by no means sure 

 that the Celtic names may not be translations of Horsa's 

 Burg, the castle of a Saxon settler, and not really very old ; 

 which makes them especially interesting. 



Dunglass, of which one case occurs on the borders of 

 Berwickshire and East Lothian (the others are on the lower 

 part of the Clyde, and in Caithness) can only be a Gaelic 

 green or grey fort ; in this case, the low hill behind the 

 house; there is a sort of tradition of old fortifications there, 

 but if they ever were visible, they have been obliterated by 

 a bowling green of the last century. The site commands 

 a very wide view. 



Dunbar is perhaps more likely to be the town of St. Barr 

 than anything else. 



The Rhind Lectures for 1893, on the Place-Names of 

 Scotland, by Sir Herbert Maxwell, were a great advance, 

 as a whole, on any treatment of the subject there has been. 

 In fact the points open to correction were chiefly where a 

 previous observer had been misled by a theory, or some 

 accidental circumstance. In certain cases, like that of the 

 interesting island of Fidra, formerly Fetheray, nothing but 

 a knowledge of the place can explain the name. The lecturer 

 was inclined to make it a reminiscence of the old name of 

 Fothreve, which has now become Forth, and is not used as 

 the name of the country between the Forth and Tay, to the 

 upper part of which it belonged. But as Fitheray would 

 mean the island of the isthmus in Danish, one is inclined to 

 think that the tradition (which is really a bit of rudimentary 

 geology) that the rocks called the Brigs of Fidra had once 

 connected the island with the mainland had already arisen ; 

 though it is not likely they have ever done so since the 

 glacial period, they certainly do look very much like the 

 remains of a causeway. There is no improbability in the 

 Fothreve theory ; on the contrary, I am inclined to think 

 that the name of Inch Mickery, at Queensferry, is the last 



