23 



tion of keiykt. We know that a = A (page 21), and 





 =<? ^/ (formula iv, page 12). In other word* 



a 



the sample tree of average basal section is a true aver- 

 age tree. This case obtains only exceptionally in * 

 whole'crop, but is always approximately true for the 

 groups of trees composing its several girtli-chisses. 



(b) Height and form-factor a function of the girth and 



hence variable, or height constant, but form-factof 

 variable and a function of girth. The former con- 

 tingency may, as a rule, be assumed as always prevailing 

 in regular crops of one and the same age and origin ; 

 the former can very seldom happen. In either con- 

 tingency only girth-classes should be adopted. 



(c) Height irregularly variable, the form-factor being a func- 



tion of height. In this case height-classes must also 

 be formed. 



(d) Height and furm- factor both irregularly rariable.f\i\* 



is generally the case in irregular crops composed of 

 trees of various ages. Here also girth and height- 

 classes must be formed, and in addition several sample 

 trees should be measured for each class. 



B. GIRTH CONSTANT. This is always true of girth-classes, 

 and may be assumed to be true also of girth gradations. 

 "We have three distinct cases as follows : 



(a) Height, and form-factor constant or both of them functions 



of girth. In such a contingency every component 

 tree would be a correct average tree, but this is never 

 exactly the case even in the most regular crops. 



(b) Height constant or a function of girth, form-factor vari- 



able. This is the case generally met with. In order 

 to obtain a correct average form-factor several sample 

 stems must be measured. 



' (c) Height variable and form-factor a function of height. In 

 this case height-classes must be formed, thus reducing 

 the conditions (for each height-class) to those of case 

 (a). 



