[ganong] PLACE-NOMENCLATURP: of new I5RUNSWICK 191 



other Indian names. On the othei' hand it is extremely difficult to find 

 examples of great change, that is, authentic examples derived from the 

 same documents as the cases above. By taking mis-spelled non-authori- 

 tative forms it is easy enough to find differences, which would not be real. 

 The best, almost the only marked example of change 1 have found is in 

 Grimrose, which is given by deMeuUes as Grimerasse, and is now called 

 in Maliseet Et-leem-lotch. The present Indian names, then, go back as a 

 rule with little change for two hundred years, and how much further no 

 one can say. From this we must infer that Indian place-names are very 

 constant in form ; and if it be true, as is often said, that Indian languages 

 are constantly and rapidly changing, their place-names must form their 

 most conservative and stable element, in which respect aboriginal and 

 civilized tongues are in agreement. Their constancy in a changing lan- 

 guage helps to explain also why so many of them are now not understood 

 by the Indians, and also shows how completely they have become proper 

 names and have lost the character of descriptive phrases. 



The tendency to stability in place-names (as well, indeed, as in other 

 words) unaided by maps or records is illustrated also by the fact that the 

 pronunciation used by country-people for Indian names is often nearer 

 the Indian form than is that generally current among people who use 

 books, and this also where there is no intercourse at present with the 

 Indians. It is thus with Madawaska, Jemseg, Piskahegan (see the 

 dictionary). 1 am iiiclined to think that the popular notion that 

 languages are most stable which have a literature, and are very unstable 

 without it, is not altogether true ; at all events it does not hold for 

 place-names. 



Three tribes of Indians live in New Brunswick, with about the same 

 distribution as at its discovery, the Micmacs occupying the entire G-ulf 

 of St. Lawrence slope from the Eestigouche to Nova Scotia, and the head 

 of the Bay of Fundy ; the Maliseets in the St. John valley, and the Pas- 

 samaquoddies upon the St. Croix and Passamaquoddy Bay. Their 

 relationships as shown in their language have not been fully worked out, 

 though Eand has supj)lied data from the Micmac tongue. Both are of 

 course of Algonquin stock. Micmac and Maliseet with much in common 

 are yet so distinct that members of one tribe cannot readily understand 

 those of the other; Passamaquoddy and Maliseet are, however, identical 

 or nearly so, and are very closely allied with the dialects of the New 

 England tribes. The Micmacs on the other hand seem to be related to 

 the Algonquin tribes of Canada. 



Though so difïerent in most features of their language, there is an 

 exact resemblance in many of their i^lace-namcs. This is shown both in 

 similar terminations, etc., as will be discussed be'.ow, and also in the fol- 

 lowing resemblances : 



Sec. II., 1896. 13. 



