278 Origin of Biomes and Succession 



produced small grass supply would certainly produce extensive over- 

 grazed areas. It seems reasonable to believe that these biotic factors 

 coupled with ordinary weather cycles provided continuous areas 

 for many subclimax communities. 



Biotic or abiotic factors acting alone, however, may produce 

 different kinds of subclimax communities. In the reversal of succes- 

 sion described by Weaver ( 1954 ) , if the climax Andropogon prairie 

 is overgrazed, it is first replaced by subclimax blue grass Poa praten- 

 sis, which forms a uniform prairie cover; if this is overgrazed, sev- 

 eral species of herbs such as Verbena and Vernonia may become 

 abundant members of the blue grass prairie community. These 

 are not the same subclimax stages which follow complete destruc- 

 tion of the vegetation by fire or excessive wind or drought. The suc- 

 cession following these catastrophic events consists of communities 

 of annual grasses, then perennial grasses other than Poa pratensis, 

 and finally the Andropogon climax. Thus certain communities prob- 

 ably owe their existence solely to biotic factors acting with local 

 and unusual intensity on the climax community. Such communi- 

 ties are a product of the complexity of the community structure. 



Studies of phytophagous insects in Canadian forests indicate a 

 similar but more limited action of biotic factors in reversing suc- 

 cession ( Blais, 1952, 1954; Ghent, Fraser, and Thomas, 1957; Morris, 

 1958). The spruce budworm Choristoneiira fumiferana, the caterpil- 

 lar of a Nearctic moth which is always present throughout the bal- 

 sam fir and spruce forests of eastern Canada, is usually present in 

 low densities but occurs in outbreak numbers when two conditions 

 coincide. These two conditions are an almost pure dense stand of 

 balsam fir and a preponderance of mature trees in the stand. Under 

 these conditions the spruce budworm may completely defoliate the 

 trees for several successive years and kill all but the very young 

 ones. Winds usually fell the dead stems. In this area a new com- 

 munity springs up composed of very small balsam and spruce 

 trees (already present before defoliation) and a rank growth of 

 shrubs or shrubby herbs including Ribes, Riibus, Corylus, and 

 Aralia, plus many understory herbs including Lijcopodium and 

 Linnaea (Ghent, Fraser, and Thomas, 1957). This community has 

 subclimax angiosperm shrub species as its dominants but contains 

 well-advanced young growth of balsam and spruce up to several 

 feet high, which are the dominants of the climax. On this basis it 

 would seem as if the spruce budworm attack had moved the suc- 

 cession back from late in the climax to a stage late in the preclimax 

 community or to a stage very early in the climax community. In 



