﻿1903] UPLAND PLANT SOCIETIES 55 



of the soil}^ Mere field observations can neither destroy this 

 hypothesis nor establish it. 



On the other hand, the present series of observations seems 

 to show that the historic factor may be a very important one in 

 the distribution of the plant societies of Kent county, and the 

 test of the hypothesis offered in this connection is to be 

 obtained through observation and comparison, and through 

 them alone. 



It is hoped that in the future studies may be made of other 

 areas lying north and south of this one, and the results carefully 

 mapped and published. Emphasis is here laid on the map, for 

 by it alone can a satisfactory comparison be instituted. Photo- 

 graphs are apt to be too superficial to be of any accurate use, 

 though they would undoubtedly be valuable in connection with 

 the map. 



Hull Botanical Laboratory, 



The University of Chicago. 



^4 Similar conclusions to the one here expressed have been published by Mr. 

 Bruncken. Bruncken, E., Studies in plant distribution, i. On the succession of 

 forest types in the vicinity of Milwaukee. Bull. Wis. Nat. Hist. Society 2: 17-28 

 1902. 



* 



