GENERAL METHOD OF SAMPLE TREES. 49 



(2) The sample trees in the several classes differ in basal 

 area somewhat from the mean basal areas. 



If the volumes of the approximate sample trees are v/ ; t? a ' ; 

 v s ' . . . and the corresponding basal areas = / ; s/ ; s 3 ' ; 

 . . . . then 



v _v 1 'xs l v _ r a ' X s 2 . v _<Xg 3 



K ! - X %! , K 2 - ', - X 7? 2 ) ' 3 -- 7 - X W 3 . . . 

 Si S 2 % 



As 



s x x w x = Si = basal area of the first class, 



s 2 x ?i 2 = $2 = basal area of the second class, etc. 



the volume of the wood is : 



Si S 2 S 3 



(3) Several sample trees are measured in each class. In 

 that case : 



_ + < + V" + . ) X S l . _ + V" + V*" + . . .) X S t 



l . y _ * 



/ 7/ //x 



etc., and 

 F _+V+V // + - - .) x S, + V+V+ . . .) x 2 



' " ' " //X 



g. Clubbing together several Classes, leading to the Method of the 

 Arithmetical mean Sample Tree. 



In order to shorten the method described above, and to 

 reduce the number of sample trees to be felled^ several, or all, 

 classes may be clubbed together into a group. 



Let 



WL w 2 , ^ 3 ... be the numbers of trees in the several classes 

 u s.>, s 3 . . . ,, basal area ,, ,, ,, 



Jin h. 2 , h 3 . . . ,, heights ,, ,, 



/i,/ 2 >/3 ... form factors 



and 



s, h, f the basal area, height and form factor of the mean 

 tree of all classes thrown together. 



VOL. III. E 



