April, 1923 ECOLOGICAL CLASSIFICATION OF PLANT COMMUNITIES 161 



The importance of geographic factors in their effect on vegetation is 

 admirably summed up by Livingston and Shreve ('21, p. 25) in the following 

 words : " We can, in brief, put it down as a law of plant geography that the 

 existence, limits, and movements of plant communities are controlled by 

 physical conditions. The conditions that control the movements of the com- 

 munity are those of the soil ; the conditions that control the broader geo- 

 graphical limits are almost solely those of the climate. The existence of the 

 community and the extent of the area occupied are, of course, controlled by 

 conditions of both soil and climate." Surely there can be no question as to 

 the fundamental importance, from an ecological standpoint, of a classification 

 of plant communities based wholly or in part on their geographical relations. 



The Association-complex. — Areas of essential geographic homogeneity 

 may be of considerable extent, as in the case of a sand plain or a salt meadow, 

 or they may be of comparatively limited extent, as in the case of a clay bank 

 or a rock face. In either case it is seldom that a tract of country of any size 

 is studied which does not exhibit sufficient geographic diversity to produce a 

 more or less pronounced diversity of habitats, with a consequent tendency to 

 diversity of plant associations. In other words, largely as a result of geo- 

 graphic differences of one sort or another the vegetation of any sizeable 

 tract of country comprises what for convenience may be characterized as an 

 association-complex, this term being given to any group of associations which 

 occupies a definitely circumscribed area. 38 



Geographically Defined Association-complexes. — When it comes to the 

 consideration of the geographic distribution of plant associations, and particu- 

 larly to the way in which associations are grouped into geographically defined 

 complexes, it becomes necessary to distinguish between the influence of cli- 

 matic and that of physiographic factors. Generally speaking, the subdivisions 

 of the earth's surface determined by climate are of relatively wide extent — 

 so wide that they are spoken of as climatic regions. In comparison with these, 

 the subdivisions determined by physiography are infinitely smaller. 39 A 

 region characterized by uniformity in climate may be anything but uniform 

 in the character of its physiography; and while the uniformity of climate in 

 such a region tends to favor uniformity in the ecological character of the 



38 Thus we can speak collectively of the associations of a naturally denned area, such 

 as a ravine or a lake or a climatic region, as an association-complex ; or we can refer by 

 the same term to the associations of an arbitrarily defined area, such as the State of 

 Connecticut. Interpreted in this sense the term association-complex corresponds in its 

 unrestricted usage to the term community, which by general agreement may be used in- 

 discriminately with reference to any specific assemblage of organisms, regardless of its 

 ecological rank. The term has been given a more restricted application, however, by 

 Du Rietz ('17). 



38 Exceptions to both these statements will of course be obvious. Thus marked dif- 

 ferences in climate are found within comparatively limited areas in mountainous regions, 

 while certain physiographic features, such as sand plains and salt marshes, may be essen- 

 tially widespread in their continuity. 



