﻿r 



> 



1903] UPLAND PLANT SOCIETIES 49 



In the northern half of the county, west of the Rouge River, 

 we find the heavier soils still retaining societies I, II, and III. East 

 of the valley of this river we find the country is much cut up. 

 The clay areas are small and rather well drained. They may be 

 occupied by any society from I to V. That they can support 

 society I is well shown by its occurrence in several places. Its 

 general absence from this region is perhaps due to another cause, 

 to be mentioned later. 



East of Sparta and northeast of Cedar Springs are perfectly 

 typical examples of society I growing upon light soil, the former 

 without trace of pine. In the western part of the Sheffield area 

 I was told by a resident that the clay was at least twenty feet 

 below the surface. But in these areas the soil is deeply covered 

 with humus. What the conditions are which cause the accumu- 

 lation of humus in one place and not in another apparently 

 similar place, I was unable to make out. This accumulation 

 most often occurs in rather low regions, where the sand would 

 normally remain moist longer than elsewhere. The maple-elm- 

 agrimony society grows to perfection on heavy soil with little 

 or no true humus. It is also found on lighter soil which has a 

 humus covering. 



In the southern half of the county it seems fairly clear, then, 

 that societies I and II will grow on rather deep sand if that be 

 covered wiih humus, and that when society V is found on clay it 

 is well drained and usually with little or no humus. Through- 

 out the county there is an obvious difference in humus content 

 between the areas occupied by societies IV and V, the sand of 

 the former being mixed with vegetable debris. The intermedi- 

 ate society III is found on the loamy soils and on the dryer and 

 better drained clay areas. 



b. The historic factor. — Besides the factor of relative water 

 content in the soils there is another which may be active in this 

 region, namely, what may be termed the historic factor. As the 

 ice sheet retreated slowly northward at the end of the last glacial 

 period, the portions of Kent county first uncovered were of 

 course in the southern part. And the first parts of the sand 

 plains to be uncovered lay also at the south, although these areas 



Mo. Bot. Garden 



ISu4. 



