HEIGHT GROWTH AS KEY TO SITi". 755 



would distinguish sites, once and for all. on the basis of the form of 

 forest which will ultimately, under natural conditions, prevail, Roth 

 would disregard this entirely, leaving the matter of site purely a func- 

 tion of the species which happened to be growing there. Zon's is a 

 matter of the succession of forests, presupposing that the forester is 

 interested mainly in the natural evolution of the stand ; Roth speaks of 

 site in terms of what it means for each and every species present, re- 

 gardless of whether it is there to stay or only by sufferance. 



It needs no argument to show that these are entirely independent 

 ideas, conflicting in no respect, except perhaps in the order of prece- 

 dence for immediate ai)plication. Even here there is really no dispute, 

 for each is important and each is bound to receive attention. The 

 object of this paper is not to argue one above the other as a necessary 

 program, but to point out ( i ) what seem to be some of the limitations 

 in regard to each: (2) why height growth as a site-indicator should be 

 adopted generally by the profession, and (3) how the two plans can 

 be made to fit in and lead to a fundamental and universal classification 

 of American forest sites. 



First, however, it is necessary to have in mind what qualities are de- 

 sirable in such a scheme. In the writer's opinion, these may be briefly 

 stated as follows : 



( 1 ) The plan should be fundamentally reliable for all future pur- 

 poses of management, but yet simple enough to permit quick and rea- 

 sonably accurate determinations in the field ; in other words, it should 

 be possible of immediate general use. subject to subsequent revision in 

 detail. 



(2) It should be universally applicable in jjrinciple. and yet adaptable 

 enough for intensive use in small areas. 



(3) While arbitrary lines must be drawn between site classes, these 

 need not be the same for the entire country ; standardization is more 

 important in principle than in detail. 



(4) The plan should provide a means of determining the site for all 

 forests and forest land everywhere, and not be limited to pure, even- 

 aged stands. 



If these requisites are correctly stated, each plan of classification can 

 be gauged by them. 



Any method of determining forest sites must em])loy an indicator, 

 whether this be the probable ultimate forest ("climax t\pe"), the 

 height growth of one or more species present, the ciu'rent annual vol- 

 ume increment of a full\' stocked pure stand, some herl) or shrub tvpical 

 of ;i locality, or merely tlie composition of the existing stand. Similar 



