THE SELECTION SN'STEM 67 



be an actual outlay which often would not be justified. 

 The experiment has sometimes been tried of girdling 

 these trees. Doubtless this expedient would be practi- 

 cable m certain instances in public forests, but there are 

 usually more important cultural operations which demand 

 the use of whatever money there is to devote to such 

 purposes. Sometimes the danger of injury to young 

 growth from the windfall of girdled trees would fullv 

 offset the advantages secured. As a rule, therefore, the 

 cutting or girdling on a large scale of large, mature trees 

 which cannot be sold is not advisable. 



A poor market means that small individual trees can- 

 not be sold. It would be impossible, with a poor mar- 

 ket, except at considerable expense, to cut all the trees 

 below a diameter limit which are defective, unpromising, 

 or interfering with reproduction and with the growth of 

 other better trees. The amount of money which can be 

 spent on cutting small trees for these purposes depends 

 on the returns which would result. The question must 

 be worked out in the same way as any other proposed 

 investment. 



Cost of the System 



The cost of forestry with the selection system de- 

 pends on the following factors: 



1. Increased Cost of Logging. — In any selection for- 

 est, like most of our virgin forests, the lumberman has to 

 go over a good deal of ground for the timber. If there is 

 a further restriction of the amount cut by limiting the 

 size to only the largest trees, and by leaving seed-trees, 



