ECOLOGY OF ISLE ROYALE. 51 



these with the falling of the lake have migrated outward, as is sug- 

 gested by the courses of the main streams occupying the rock valleys. 



The very immature condition of the drainage shows that during the 

 present post-Glacial cycle only comparatively slight changes have 

 modified the relief from the condition in pre-Glacial times; it is thus 

 largely an inherited topography, hence the consequent drainage. 

 It should perhaps be added, however, that the date of the faulting is 

 not definitely known; it may be very ancient, but the weight of the 

 ice sheet may have had considerable influence. It thus seems probable 

 that with the decline of the lake level there has been an increase and 

 downward elongation of the stream environments, and that their course 

 has been determined largely by the pre-Glacial topography, supple- 

 mented, of course, by the southward tilting of the land. The lake 

 basins have had an origin similar to that of the streams and have 

 tended toward extinction by tilting, inwash, organic debris and to a 

 limited extent by the downcutting of outlets. On account of the rela- 

 tively small amount of erosion by the ice sheet it is probable that the 

 shallow swamps and the smaller streams were influenced more by the 

 ice than those features related to the greater relief of the surface; 

 even moderate tilting would considerably influence such an environ- 

 ment, because within the major valleys the divides are generally low. 



The origin of certain land habitats only remains to be considered. 

 These have undergone a complex succession of changes. The resistent 

 lava of the Greenstone had been the least reduced by erosion so was the 

 first to emerge from the lake level. This was first a beach, and as the water 

 fell from its crest the upper beach migrated to lower levels and the 

 land habitat continued to increase in area. The beach line itself 

 expanded laterally, if not in width, as the area of the island increased. 

 When once the exposed rocks were beyond the reach of the waves, 

 weathering and erosive processes were initiated which tended to pro- 

 duce a residual soil. Plant remains from lichens were perhaps the 

 first humus formers, and it is probable that it was not until the 

 period of Lake Algonquin that the lake drift which was washed ashore 

 became a source of such material; but winds, birds, lake currents and 

 the waves may all have contributed pioneers of the higher plants. 

 The harbors at the northeastern end of the island would tend to strain 

 out the drift from the southwestward flowing current and the return- 

 ing one along the southeastern coast of the island would tend to lodge 

 drift in Washington Harbor and the Siskowit Bay region. 



As the water continued to fall to lower levels, the land biota 

 followed down the slopes behind the receding beach. By the 

 Nipissing stage, the vegetation and many animals were probably 

 well established and had begun to actively encroach upon the 

 swamps and lakes and thus tended to increase the land habitat. 

 With the tilting that followed the formation of the Nipissing 

 beach, a readjustment must have taken place between the land 

 and water habitats, but to what degree their relative areas were 

 influenced is not known. During the initial elevation ponding would 

 be expected at the northeastern end of the island, but with a greater 

 elevation this same area would be well drained, as the divides in the 

 valleys are low and the transverse drainage near the central part of 



