310 THE STUDY OF PLANT COMMUNITIES * Chapter XI 



ficult elsewhere. That cross-dating and correlation with climatic 

 variation is possible in moist-temperate climates was demonstrated 

 by Douglass' studies in Europe and several parts of the United 

 States. In the Mississippi drainage area, the deviation from the 

 normal annual precipitation has been shown to affect ring growth 

 more than total precipitation, but the relationship is modified by 

 temperature and wind as they influence evaporation. 119 Ring 

 growth in New England has been shown, in a chronology from 

 hemlock, to have close correlation with climate as indicated by 

 exceptional and poor crop years. 164 



Since tree-ring analysis was originally begun with the hope that 

 it would show solar-terrestrial relationships, it was natural that, 

 with the establishment of long, dated chronologies, the data should 

 be studied for cyclic characteristics. Permanent periods, or those 

 of fixed length, showed no correlation; therefore, this idea was dis- 

 carded for one of cycle complexes in which any obvious or sig- 

 nificant recurrence of variation in data was considered to be cyclic. 

 On this basis, definite relationships were demonstrable between 

 sunspot activity in the past and terrestrial climate as recorded in 

 certain long-time chronologies of tree rings. An eleven- (ten to 

 twelve) year cycle is especially pronounced throughout the old 

 records and continues to be borne out, in a general way, for mod- 

 ern conditions. During periods of sunspot maximum, drought is 

 characteristic, and sunspot minimum is associated with excessive 

 precipitation. Application of the method to climatic prediction 

 may be possible as more long-time meteorological records are an- 

 alyzed and as more tree records of great length are worked out to 

 show the nature of prehistoric climates on many parts of the earth. 



THE RELICT METHOD 



As for several other phases of dynamic ecology, we are indebted 

 to Dr. E E. Clements for recognizing the potentialities of the relict 

 method, for demonstrating its usefulness, and for a clear and com- 

 plete exposition of the entire subject. The brief discussion that 

 follows can hardly avoid being a condensation of his ideas. 58 



"In the ecological sense, a relict is a community or fragment of 

 one that has survived some important change, often to become in 

 appearance an integral part of the existing vegetation!' The con- 



