218 MICHIGAN SURVEY, 1908. 



that could be desired ; and one important advantage here is that most 

 of the bog areas are comparatively small, and all in a reasonably limited 

 aren, thus offering an unusual opportunity for their comparative study. 



There is, on the other hand, one difficulty that should be mentioned 

 in connection with any kind of field work on Isle Eoyale which takes 

 one far back from the shore, and that is the difficulty of penetrating 

 the dense tangles of the forest. The absence of roads, the limited num- 

 ber even of old blazed trails, the unusually dense underbrush, includ- 

 ing a very rank growth of Tax us minor (Ground Hemlock) and numer- 

 ous windfalls, together with the necessity of carrying by pack one's 

 supplies — all render the penetration of many parts of the island a mat- 

 ter of such difficulty that it has been remarked by all who have at- 

 tempted it. 



In the limited time at our disposal during the summer it was im- 

 possible to visit all of the 100 or more bog areas on the island: our 

 attention was therefore confined to a limited number of those which 

 are typical of a certain stage of development, or to those having in- 

 dividual points of special interest. 



Three general stages of the lake-bog series will be briefly touched upon, 

 (1) the open lake without marginal vegetation, (2) the partly open 

 lake with marginal vegetation of varying width, (.3) the wholly carpeted 

 bog area; the vegetative carpet in some cases being recent enough to give 

 beneath the feet, in other cases old and solid enough to be more or less 

 forested. 



The first, or open stage, includes only a few of the largest lakes of 

 Isle Eoyale, such as Lake Siskowit and Sargent Lake. Of these Lake 

 Siskowit is by far the largest, being at least a mile and one-half broad 

 in places. The principal reasons for the absence of vegetation in the 

 lakes of the first class seems to be that their size and openings renders 

 the sweep of the wind and the resulting wave action so vigorous that 

 even annuals cannot get a foothold along their shores. 



Wave action in a few places is clearly marked by a narrow but well- 

 defined beach, as along the north shore of Lake Siskowit. 



Another factor that has to be reckoned with in the larger lakes of 

 the first class is the work of ice. Ice destroys shore vegetation in two 

 ways, first — by pushing, due to expansion by freezing (and this total 

 expansion in a lake as large as Lake Siskowit is considerable) ; second 

 — the open expanse allows large ice fioes to blow ashore during the 

 spring break-up. The most interesting example -ot ice pushing noted 

 was along the north shore of Lake Siskowit, where there is an irre- 

 gular ridge, varying in height up to 15 feet and composed of bowlders 

 and various fragmental materials. Along this ridge there were, in 

 places, even over-turned trees of considerable size, pointing away from 

 the lake, back 20 to 25 feet from the present shore. This ridge seems 

 certainly to be the work of ice as in the case of the so-called "bowlder 

 rim" lakes of the western United States, or the ice floe ridges at Put- 

 in-Bay in Lake Erie. 



In drawing a line between the lakes that will long continue to remain 

 free from the encroachment of vegetation and those which are being 

 gradually captured by vegetation, the size and openness seem to be the 

 most impoi'tant factors, affecting the vigor of Avave action as well as 

 the work of ice in one or both of the ways suggested. 



