VEGETATION MANAGEMENT EGLER 



311 



and are not part of an invading relay succeeding the shrub stages, 

 then the dollar costs of brush (tree) control are on an entirely dif- 

 ferent plane. These trees once removed are not capable of reinvading 

 the remaining shrublands and will not invade unless bare soil (their 

 requisite for reinvasion) is again produced. In short, such species can 

 be eliminated, not controlled. 



Figure 3. — A diagrammatic presentation of initial floristic composition. According to this 

 theory, weeds, grasses, shrubs, and trees were all present in the soil at the time of last 

 cropping, fire, or other catastrophe, as roots, seedlings, or dormant seeds. The successive 

 physiognomic stages are then due to the progressive development of these plants, with 

 the later stages taking longer to grow up and crowd out the preceding stages. 



THE APPLICATION OF VEGETATION DEVELOPMENT 



It is not to be assumed that one or the other of these hypotheses, 

 "relay floristics" or "initial floristic composition," is exclusive in any 

 one region. Every case of vegetation development known to me is a 

 combination of both, with initial composition playing a larger role 

 than had been previously anticipated. Every tree species in every 

 type of vegetation development has a role worthy of impartial in- 

 vestigation if lowest-cost brush control is to be obtained for the 

 land manager. 



GRASSLAND vs. SHRUBLAND 



Consideration of the above principles leads inevitably and logically 

 to the next thought on vegetation management of these lands. Both 

 forests (4) and annual weeds (1) are not permissible cover types. 



284725—54- 



-21 



