The Evolution of Communities 211 



ganisms which feed on them or the understory species which depend 

 on them. Some relationships, however, are obhgatory in both direc- 

 tions, as for example that between plants and the animals which 

 pollinate them. 



This brief survey of biotic communities shows that they differ 

 from random mixtures of species because they are ordered in two 

 different ways. First, a community contains only those species which 

 are adapted to live in the ecological conditions prevailing in the 

 locale of the community. Second, an intricate web of various rela- 

 tionships exists between the different species comprising the com- 

 munity. These relationships may be obligatory with regard to one 

 or both species involved or they may appear to be relationships 

 between what are actually casual neighbors. The community is 

 therefore a mixture of species held together by a combination of 

 common ecological tolerances and by interspecies relationships 

 which are either partially or entirely obligatory. 



The origin and evolutionary history of these relationships would 

 therefore seem to be the crux of community evolution. Each in- 

 stance of a relationship existing between two species may be con- 

 sidered as one unit of community structure, and each unit may 

 have an evolutionary history independent of that of other units. 

 A community embraces thousands of such units, and the evolution 

 of these units will collectively represent the evolution of the com- 

 munity. The sequence of events in this evolution is (1) the evolu- 

 tion of many species, (2) the congregation or mixing of some of 

 these species, and (3) the subsequent evolution of the relationships 

 characteristic of modern biotic communities. 



THE CONGREGATION OF SPECIES 



The dynamics of the earth's crust, so important in the increase in 

 number of species, have been the chief causal agents in bringing 

 about a congregation of species. Topographic and climatic changes 

 not only divide the ranges of some species but also unite similar 

 ecological areas which were previously separated. Thus the suc- 

 cessive submergence and emergence of parts of the Central Amer- 

 ican isthmus would first unite, then divide the oceanic areas of 

 the region and conversely divide then re-unite the land areas on 

 either side of the break. Each union of a similar area would result 

 in a congregation of the species in the joined areas. Oscillatory 

 changes of lesser degree have produced the same effect many 

 times. It seems reasonable that this process of alternating change 



