CLASSIFICATION OF PLANT COMMUNITIES 311 



changes within a comparatively short period of time: along 

 streams such rapid changes are commonly associated with the 

 erosion of ravines in uncompacted rock and the building up and 

 destruction of flood plains; along the coast they are associated 

 with the erosion of bluffs in uncompacted rock, and the develop- 

 ment or destruction of coastal swamps (e.g., salt marshes), 

 beaches and sand dunes. In this connection, see especially 

 Cowles (7, 8). But, in the large, rapid changes such as these are 

 the exception rather than the rule. Over much the greater part 

 of the earth's surface the changes due to topographic agencies are 

 consummated so slowly that their influence on the vegetation, 

 like that of climate, becomes apparent only when geologic 

 periods of time are taken into account. It is indeed open to 

 question whether in the main the successions due to topographic 

 agencies do actually take place more rapidly than those due to 

 changes in chmate. In the glaciated regions of the eastern 

 United States, for example, the land surface for the most part 

 has undergone little alteration since the recession of the con- 

 tinental ice sheet; yet during this period it is generally agreed 

 that there have ensued .profound cUmatic changes, which have 

 been accompanied by correspondingly great transformations in 

 the character of the vegetation (in this connection, see Nichols 

 18, pp. 237, 245, 249). 



If account is taken of the three types of succession defined 

 by Cowles, then it is evident at once that vegetation can never 

 attain a condition of equilibrium. As Cowles (8, p. 81) aptly 

 phrases it, 'Sve have a variable approaching a variable rather 

 than a constant." While conceding, however, the far-reaching 

 importance of the cHmatic and topographic changes which have 

 ensued and which will continue to ensue in geologic time, it seems 

 to the writer that, in attempting to solve the relatively contem- 

 poraneous problems of dynamic plant geography, much more is 

 to be gained than lost by postulating the cUmatic conditions of 

 the present, and by ignoring topographic changes, except m 

 so far as these manifestly proceed with sufficient rapidity as to 

 become effective within the present climatic era. 



