ECOLOGY OF NORTHERN MICHIGAN. 47 



Great Lakes, it was followed by plants and animals, so that when it finally 

 disappeared the different forms of life were left distributed in a north and 

 south direction approximately in the order of their northward migration. 

 The wide distribution of the boreal forms in northern North America is thus 

 seen to be very intimately related to the character and extent of the pene- 

 plained nature of the region. 



Toward the south, with the retreat of the last ice sheet, the boreal forms 

 became restricted to local areas, as in bogs for example, but further north 

 they tend to become of wider distribution. This is shown in the Porcupine 

 Mountains by the general distribution of forms that about Ann Arbor, Mich- 

 igan are confined to the tamarack swamps, by the presence of boreal forms 

 in the climax forest to the exclusion of more southern forms, and by the fact 

 that there are few forms restricted to the bog societies. That this spread- 

 ing out of the conditions with which the boreal forms are associated affects 

 the more nearly related habitats first is shown by the number of boreal 

 forms in the flood-plain biota while on the higher ground the southern 

 forms become more numerous. When the present biotic types reached 

 this region, the various societies took possession of the different habitats 

 to which they were adapted, but, owing to the fact that the processes were 

 not in equilibrium these societies could not remain fixed. This is shown, 

 at the present time, where the changes were taking place rapidly, as for 

 example on the mountain top, by the fact that the conditions are being 

 changed in the direction of those that prevail on the low land, and the biota 

 of the cliff habitats is being supplanted by the deciduous forest types which 

 occupy the lower levels where the processes approach an equilibrium. As 

 the processes approach an adjustment, the changes take place more slowly, 

 and the more resistant forms may persist for a considerable time in the 

 succeeding society, as is shown by the presence of the isolated White Pines 

 in the deciduous forest. The deciduous forest society thus represents the 

 climax biota in the Porcupine Mountains, as it is associated with those 

 conditions toward which the other habitats are tending. 



From the dynamic nature of the processes that give rise to the environ- 

 mental factors in the different habitats, it follows that a study of the dis- 

 tribution of forms in a particular region should be made from the standpoint 

 of the processes involved, and, since the conditions in particular habitats 

 may become of geographic extent, the same methods are applicable to general 

 distributional problems. This leads naturally to the conclusion so well 

 expressed by H. R. Mill ('05, p. 10) for geography in general: "Geography 

 was defined long ago as the science of distribution; but the old idea was 

 statical distribution, the laying down on maps of where things are; now 

 we see that we ought to go further and discuss also how the things got 

 there, why they remain there, whether they are in transit and if so how 

 their path is determined. We are learning to look on distribution from its 

 dynamical side, the earth with all its activities being viewed as a machine 

 at work." 



