REPRODUCTION OF WESTERN YELLOW PINE. 79 



proper time, all of the seedlings in their near proximity will be 

 killed or retarded in growth by shade and root competition. Grant- 

 ing that a removal felling may be possible under special conditions, 

 the fact remains that nothing has been gained by the first cutting, 

 because in such open stands as our western yellow pine the removal 

 of a few overmature trees would have little effect in promoting seed- 

 ling establishment. In fact, if the shelterwood principle is applied 

 at all, germination should be secured under the virgin stand, and 

 within three years the mature and overmature trees should be re- 

 moved in one operation. If, as would inevitably be the case if this 

 method of cutting were adopted in general practice, the removal 

 fellings are omitted or deferred, we would have returned to the 

 group-selection method, differing from the standard practice only in 

 leaving a larger number of mature trees. Such cuttings would un- 

 doubtedly restock more promptly than the standard group-selection 

 cutting as far as the larger openings are concerned, but the actual 

 proportion of area restocked would be less, since few seedlings would 

 survive near the old trees. Against any benefits in the way of repro- 

 duction, which are at best doubtful, must be weighed the cost in loss 

 of increment, additional logging costs, and the deterioration of from 

 2,000 to 3,000 feet b. m. of merchantable timber above the amount 

 left in the standard group-selection method. The shelterwood sys- 

 tem was included in this experiment less with the idea that it would 

 ever be adopted in silvicultural practice than with the idea of provid- 

 ing a contrast with other systems. Most of the difficulties which 

 have Been pointed out were anticipated. It has been demonstrated 

 that the initial establishment of seedlings is favored by increasing 

 the number of large seed trees, which is one of the principal points 

 on which information was sought. An important point on which 

 stress was laid at the beginning of the experiment was the value 

 of a shelter wood in protecting young seedlings and thus favoring 

 survival. The evidence here obtained shows that the beneficial effects 

 attributed to the shelter of old trees were exaggerated, while associ- 

 ated detrimental effects were underrated. Present indications are 

 that, on the whole, the advantages of the shelterwood method are 

 outweighed by its disadvantages. 



In the scattered seed-tree method the outstanding advantages 

 are economy in logging and the creation of favorable conditions 

 for the growth of seedlings. A disadvantage which was anticipated 

 at the beginning of the experiment, but which has not materialized, 

 is injury to seedlings through lack of sufficient shelter. On the 

 contrary, present indications are that the removal of trees in excess 

 of requirements for seed reproduction is ordinarily a distinct aid to 

 reproduction. Although in this experiment the method was handi- 

 capped by too small a number of seed trees, the results obtained 



