TRANSACTIONS OF SECTION H. 889 



literature, for example, all that has been contributed to its vast stores from this 

 native source, you would find that you left a wide and unwelcome void. 



But the question must present itself sooner or later, with what race outside 

 these islands we are to compare or identify our mound-dwellers. I am not pre- 

 pared to answer, and I am disposed to ask our archaeologists what they think. In 

 the meantime, however, I may say that there are several considerations which 

 would impel me to think of the Lapps of the North of Europe. But even supposing 

 an identity of origin could be made out as between our ancient mound-inhabiting 

 race and the Lapps, which, I am told, is craniologically impossible, it would 

 remain still doubtful whether we could expect any linguistic help from Lapland. 

 The Lapps now speak a language belonging to the Ugro-Finuic family, but the 

 Lapps are not of the same race as the Finns ; so it is possible that the Lapps have 

 adopted a Finnish language and that they did so too late for their present language 

 to help us with regard to any of our linguistic difficulties. One of these lies in 

 our topography : take for instance only the names of our rivers and brooks — there 

 is probably no county in the kingdom that would be too small to supply a dozen 

 or two which would baffle the cleverest Aryan etymologist you could invite to 

 explain them ; and why ? Because they belong in all probability to a non-Celtic, 

 Don- Aryan language of some race that had early possession of our islands. Never- 

 theless it is very desirable that we should have full lists of such names, so as to 

 see which of them recur and where. It is a subject deserving the attention of 

 this Section of the British Association. 



We have now loitered long enough in the gloom of the Pecht's house : let us 

 leave the glamour of the fairies and see whether any other race has had a footing 

 in these islands before the coming of the Celts. In August 1891 Professor Sayce 

 and I spent some fine days together in Kerry and other parts of the south-west of 

 Ireland. He was then full of his visits to North Africa, and he repeatedly assured 

 me that, if a number of Berbers from the mountains had been transferred to a village 

 in Kerry and clad as Irishmen, he would not have been able to tell them by their 

 looks from native Irishmen such as we saw in the course of our excursions. This 

 seemed to me at the time all the more remarkable as his reference was to fairly 

 tall, blue-eyed persons whose hair was rather brown than black. Evidence to the 

 same effect might now be cited in detail from Pi'ofessor Haddon and his friends' 

 researches among the population of the Arran Islands in Galway Bay. Such is 

 one side of the question which I have in my mind : the other side consists of 

 the fact that the Celtic languages of to-day have been subjected to some dis- 

 turbing influence which has made their syntax unlike that of the other Aryan 

 languages. I have long been of opinion that the racial interpretation of that fact 

 must be, that the Celts of our islands have assimilated another race using a 

 language of its own in which the syntactical peculiarities of Neo-Celtic had their 

 origin : in fact that some such race clothed its idioms in the vocabulary which 

 it acquired from the Celts. The problem then was to correlate those two facts. 

 I am happy to say this has now been undertaken from the language point of view 

 by Professor J. Morris Jones, of the University College of North Wales. The 

 results have been made public in a book on The Welsh People recently published 

 by Mr. Fisher Unwin. The paper is entitled 'Pre-Celtic Syntax in Insular Celtic,' 

 and the languages which have therein been compared with Celtic are old Egyptian 

 and certain dialects of Berber. It is all so recent that we have as yet had no 

 criticism, but the reasoning is so sound and the arguments are of so cumulative a 

 nature, that I see no reason to anticipate that the professor's conclusions are in any 

 danger of being overthrown. 



At the close of his linguistic argument, Professor Morris Jones quotes a French 

 authority to the effect, that, when a Berber king dies or is deposed, which seems to 

 happen often enough, it is not his son that is called to succeed him, but the 

 son of his sister, as appears to have been usual among certain ancient peoples of 

 this country ; but of this more anon. In the next place my attention has been 

 called by Professor Sayce to the fact that ancient Egyptian monuments represent 

 the Libyans of North Africa with their bodies tattooed, and that even now some 

 of the Touaregs and Kabyles do the same. These indications help one to group 



