4 BULLETIN 210, TJ. S. DEPARTMENT OF AGRICULTURE. 



(II), intermediate (III), oppressed (IV), and suppressed (V) may- 

 be followed. Since, however, the crown development of trees in the 

 codominant and intermediate classes may not be uniform, these two 

 classes may be subdivided and designated by the letters a, b, c, and d. 

 In this way there may be from seven to ten classes or groups of trees. 



At the time the trees are measured and divided into groups a note 

 is made for each tree as to whether or not it is bearing cones. These 

 data, which are very interesting in themselves, become absolutely 

 essential in case of partial seed production, when not all of the trees 

 bear cones. 



In the work of dividing the trees into groups, the investigator must 

 carefully examine each crown from all sides, observing its habitus. 

 Experience has taught that in order to avoid errors of crown classifica- 

 tion the total number of trees on a sample plot must not exceed 100. 

 The size of the sample plot will depend, therefore, upon the age, 

 density, and composition of the stand. 



The enumeration of the trees may be made by marking each tree 

 with white paper tags, and the record kept on a form similar to the 

 one given below: 



Number. 



Diameter. 



Class. 



Cones pres- 

 ent (+) or 

 absent (— ). 



1 



Inches. 

 25 

 13 



27 



IIIo 

 V 



+ 



1 











When in a mixed stand a tree of secondary species occurs, its name 

 should be given in the first column, but columns 3 and 4 left blank. 

 After the number of trees in each class is computed, a certain number 

 of sample trees are selected from the various classes. For ordinary 

 investigations, 10 per cent of the total number of trees on the sample 

 plot may be sufficient. For more intensive investigations, however; 

 a larger percentage should be taken. 



The more carefully the division into classes is done and the more 

 uniform the crowns of each class, the easier it is to select the sample 

 trees. It is advisable to select sample trees separately for each class — 

 that is, first select the trees of Class I, then take up the next class, 

 and so on. 



When not all trees in the stand are bearing cones, the sample trees 

 should be selected from the cone-bearing ones, and, in determining 

 the amount of seed production, the percentage of seed-bearing trees 

 for each group must be taken into account. 



The sample trees are felled, care being taken in falling that they 

 do not touch the crowns of other trees and thus knock off their own 



