33 



8. Number, ielection and meagnremeut of the sample trees. 



The number of sample trees to be measured will depend on the 

 method of survey adopted, and on the size and the degree of uni- 

 formity of the crop. The method of survey which requires the 

 smallest number is that of the average tree, 2 or 3 stems sufficing 

 if the crop is r^ular. In survey by girth-and-height classes from 

 6 to 15 trees must be measured. Draudt's method requires from 

 10 to 15. 



Since special circumstances, independent of" the method of sur- 

 vey, often limit the number of trees that may be felled, the choice 

 of the method of survey must, in such cases, be regulated by the 

 number of trees which it is permissible to fell. It must, however, 

 be borne in mind that the greatest accuracy in measuring the 

 sample trees can never compensate for the smallness of their 

 number. Averages derived from the measurement of a large num- 

 ber of trees will, even if the measurements have been made less care- 

 fully, be more trustworthy than those obtained with the help of a 

 smaller number ol trees. There is less error in estimating the 

 contents of a large number of standing trees with the aid of ^ olume 

 Tables, or by means of the ricA^-height, than iu measuring up an in- 

 sufficient number of felled trees »;* the most accurate manner possible. 



The care exercised in selecting. the sample stems must be in 

 proportion to the smallness of their number. They must be as 

 nearly as possible correct representatives of their class in respect 

 of height, form and branching, and all forked and otherwise 

 abnormal trees should be avoided. Moreover, the boles should be 

 perfectly regular at the places where the girth has to be measured. 

 "When a diameter calliper is used, at least two diameters should be 

 taken at each place to obtain a correct average. When the number 

 of sample trees to be felled is large, they should be situated at 

 different points of the entire crop. The sample trees should always 

 be chosen immediately after the enumeration survey has been com- 

 pleted, and by the same person who has conducted this survey, and 

 who has therefore a correct and vivid general impression of the 

 character of the crop and the component trees. 



The cubing of the sample trees should be effected as accurately 

 as possible by the rules already laid down for the measurement of 

 felled trees. 



