﻿38 BOTANICAL GAZETTE [January 



area. The southernmost, and hence earliest formed of these, 

 follows roughly a line drawn through Ross, Carlisle, Middleville, 

 and Hastings. The second passes through the middle of the 

 county, the southern edge being now the northern boundary of 

 the Grand River valley. The third moraine has its highest point 

 northeast of Cedar Springs, and extends in an irregular and 

 broken manner southeastward to the vicinity of Harvard, and 

 northward and westward to Kent City and Casnovia. The 

 second moraine is broken through by the valley of the Rouge 

 River, and the third by the same valley and also by the sand 

 plain north of Cedar Springs, These moraines are usually bor- 

 dered by sand plains on the outwash side and by till plains on 

 the side which was toward the ice sheet. 



Most of the surface soil of the county is predominantly 

 sandy. In classifying soils and designating them on the map, 

 no attempt has been made to distinguish the different grada- 

 tions between clay and sand. All soils which could not be termed 

 either clay or sand have been bunched together as loam, in 

 the broadest sense of that word, and denoted on the map by 

 dots. More accurate records were made, but it was found that 

 these minor differences of soil bore no apparent relation to the 

 nature of the societies recorded, and it was thought best not to 

 encumber the map with unnecessary details. Clay is denoted 

 on the map by horizontal lines, sand by an absence of any mark- 



* 



ing. Sand and gravel plains of limited extent lie along almost 

 every creek and about many of the lakes. No attempt has been 

 made to map these smaller deposits. 



3. Methods. — Owing to the large proportion of cultivated land 

 in the county, and the correspondingly small proportion which is 

 in an approximately natural state, a study of the natural plant 

 societies is necessarily a difficult one. From the more or less 

 natural areas which still remain, an attempt has been made to 

 reconstruct, as accurately as possible, the plant societies which 

 occupied the region at the time of settlement. The effects of 

 pasturing in wood lots have been allowed for so far as possible. 

 Information has been gathered from local residents as to the 

 nature of the forest which was removed in making certain fields 



r 





