Origin of Biomes and Succession 279 



contrast (1) the particular genera mentioned {Rihes, Rubus, and 

 so on) and the understory herbs are also typical forest glade spe- 

 cies, and (2) in succession following complete denudation by fire 

 the coniferous forest often follows a preclimax community having 

 the aspen Popidus tremidoides as its dominant. Thus the budworm 

 attack has apparently reversed the succession but has produced a 

 subclimax community different from the usual one which follows 

 complete denudation of the area. 



Other insects occurring as outbreaks in coniferous forests pro- 

 duce much the same results. From an evolutionary standpoint this 

 series of events is similar to wind action because areas of forest 

 leveled by wind are often colonized in exactly the same fashion 

 as when the trees are killed by insect outbreaks. Both insects and 

 wind act most effectively on large, over-age trees and in a measure 

 can be considered together as a mechanism bringing about a sort 

 of cycle between the climax and some kind of pre-climax com- 

 munity. 



Plant diseases which have virtually exterminated particular spe- 

 cies of trees in North America have had only a limited ecological 

 effect. A fungus pathogen has almost eliminated the chestnut 

 Castanea dentata from the eastern deciduous forest. Certain or- 

 ganisms restricted to this host have also been removed. Where the 

 chestnut grew in nearly pure stands, the succession was reversed 

 to a certain extent as described for the balsam fir forests, but other 

 hardwoods filled the gap in the climax community. Where chestnut 

 trees were well mixed with other species only small glades were 

 produced, as would happen with the fall of a specimen of any 

 dominant tree species in the community. Losses of elms because 

 of the diseases phloem necrosis and Dutch elm disease produced 

 the same results. Thus except for associated host-specific species, 

 the loss or great reduction of these species resulted in neither per- 

 manent change in the general ecological conditions of their respec- 

 tive communities nor the introduction of new repetitive series of 

 subclimax stages that might serve for the perpetuation of species 

 confined to them. 



A review of the foregoing material on terrestrial community suc- 

 cession brings out the conclusions that ( 1 ) subclimax communities 

 and their constituent species have existed through a long part of 

 geologic time, undoubtedly in some form ever since complex biotic 

 associations came into existence; (2) mechanisms of various kinds 

 continually reverse succession and so produce new areas suitable 

 for the persistence of subclimax species; (3) all the stages of sue- 



