106 Erratic Phenomena of Lake Superior, 



places, sometimes strictly as brilliant as a polished metallic 

 surface, and everywhere these surfaces are more or less 

 scratched and furrowed, and both scratches and furrows are 

 rectilinear, crossing each other under various angles ; how- 

 ever, never varying many points of the compass on the same 

 spot, but in general shewing that where there are deviations 

 from the most prominent direction, they are influenced by 

 the undulations of the soil. It has been said, that the main 

 direction of these striae was from north-west to south-east, 

 but I have found it as often strictly from north to south, or 

 even from north-east to south-west; and if we are to express 

 a general result, we should say that the direction, assigned 

 by all our observations to the various scratches, tends to 

 shew that they have been formed under the influence of a 

 movement from north to south, varying more or less to the 

 east and west, according to local influences in the undula- 

 tions of the soil. It is, indeed, a very important fact, that 

 scratches which seem to have been produced at no great in- 

 tervals from each other, are not absolutely parallel, but may 

 diverge for ten, fifteen, or more degrees. There is one fea- 

 ture in these phenomena, however, in which we never observe 

 any variation. The continuity of these lines is absolutely 

 the same everywhere. They are rectilinear and continuous, 

 and cannot be better compared than with the efi^ects of stones 

 or other hard materials dragged in the same direction upon 

 flat or rolling surfaces ; they form simple scratches extend- 

 ing for yards in straight lines, or breaking off" for a short 

 space to continue again in a straight line in the same direc- 

 tion, just as if interrupted by a jerk. There are also deeper 

 scratches of the same kind, presenting the same phenomena, 

 only, perhaps, traceable for a greater distaiice than the finer 

 ones. These scratches, instead of appearing like the tracing 

 of diamonds upon glass, as the former do, would rather 

 assume the appearance of a deeper groove, made by the point 

 of a graver, or perhaps still more closely resemble the 

 scratches which a cart-wheel would produce upon polished 

 marble, if the wheel were chained, and coarse sand spread 

 over the floor. The appearance of the ^ock, crushed by the 



