GENERAL METHOD OF SAMPLE TREES. 47 



a tree which represents the average of the class ; in other 

 words, the sample tree should have the mean height, as near 

 as possible a circular section, a fairly straight and not a 

 forked stem, and an average extent of crown. Even with the 

 greatest care it is not always possible to avoid errors in the 

 selection ; hence it is generally advisable to take several sample 

 trees for each class, and to ascertain the average for the calcu- 

 lation of the volume of the class. The actual number of sample 

 trees depends on the desired degree of accuracy, and the total 

 number of trees in the class. At the same time the felling of 

 many sample trees is undesirable, hence their number should 

 be kept within reasonable limits. 



A further requirement is, that the sample tree should show 

 a basal area which corresponds exactly to the mean section of 

 the class. Such a tree is only in exceptional cases found, hence 

 it is necessary to take a tree as near as possible to the true 

 section and to modify the volume in proportion of the basal 

 areas of the true and false sample trees. Let v be = volume 

 of true sample tree, v f that of the false sample tree, s and s' 

 the corresponding basal areas, then v is found by the formula : 



and 



e. Determination of the Volume of Sample Trees. 



The volume of the sample trees is determined, either by 

 felling and measuring them on the ground, or by means of 

 form factors or volume tables. 



If the trees are felled, the stem and all straight pieces of 

 branches, in fact all regularly shaped parts, are divided into 

 sections of moderate length, from 3 to 10 feet, according to 

 the desired degree of accuracy, and the volume of each section 

 is ascertained separately according to the formula : 



Volume = section in the middle x by the length. 



