216 ROYAL SOCIETY OF CANADA 



rocky lands w lie M thciv are iortile intervales near l>y. l'.ul this argu- 

 ment from prol.ability must also be used with caution, for there aro 

 many imaginable causes, personal, political or other, which may, at that 

 period, have outweighed the influences which would determine us at the 

 present time. If probability of this kind fails, then one has to lake 

 documents or old ma])s alone, and do the best they allow. But these 

 sources of evidence are by no means perfect, for aside from the fact 

 that the}' do not often mention precise sites, these being supposed to bo 

 either well known or not of sufficient importance to be especially men- 

 tione<l in documents whose objecbis different, there is the further fact that 

 they are sometimes misleading unintentionally through reliance on errone- 

 ous tradition or through any of the other[causcs which mislead us to-da5\ 

 There is a curious tendency in the minds of most students to place a reli- 

 ance upon a very old document that would not be given to one that is 

 recent, and any fragment, if only old enough, is usually accepted as 

 almost unquestionable. One may almost say that in general the older a 

 document is, the more trust do students put in it, and the more elabor- 

 atelj' will they build theories upon it, whereas the same if but few years 

 old would not receive notice. In such a study as the present, undoubtedly 

 the best documents are detailed maps made at the time on the spot by 

 surveyors or officers, especiallj' if prepared to accompany official reports, 

 where there is every inducement to truthfulness and no reason for false- 

 hood. With such maps one can go to the ground and by measurements 

 locate a spot where stood some structure of which neither trace nor tradi- 

 tion remains. General maps of small scale are much less trustworthy ; 

 for simjily practical reasons connected with their drawing or engraving 

 may make them locate places wrongly, as, for instance, a fort might be 

 placed on the wrong side of a river because there is more space for it and 

 its name there than on its proper side. In locating early settlements 

 after the period of surveys there is little difficulty, for records and maps 

 are carefully preserved in the official Land Offices, where they are easily 

 accessible, and may readily be compared with the modern topography. 

 Old jilanfl, indeed, often contain most imj)ortant hints upon our present 

 subject, for the early surveyors in unsettled districts naturally put all 

 available information and marked all possible localities upon their maps. 

 Historic sites are, of course, of all degrees of importance, from those 

 of events of world-wide interest and importance down to those so local as 

 to bo not regarded by any but the most curious ; and in this study it has 

 been difficult to draw a line between those to be included antl tho.sc to be 

 omitted. 1 have tried to err rather upon the side ol" including too much, 

 but I hope the reader will not find much that is too trivial. Lengthy as 

 this study appears, however, it is so far from exhausting its subject from 

 a l<K-al point of view, that it is to be regarded rather as a foundation for 



