838 JOURNAL OF FORESTRY 



feet or more. Though no figure was given it is estimated that the 

 loss in the last instance was over 10 per cent. 



Thus the loss in the storm of this year was nearly four times as 

 great as the combined losses of the two earlier storms. It might be 

 added, for the sake of comparison, that it was about five times as 

 great as the combined losses of the storms of 1913 and 1914 on the 

 timber sales on the Whitman National Forest. 



The magnitude of the loss, however, is not the problem which is 

 created by these storms. The problem is, first and last, silvicultural, 

 and it is one ultimately of the method of cutting. Is the solution to 

 be found in lighter or heavier cutting? Is a radical change in method 

 necessary or will but a slight modification of the present method 

 obviate the loss? To answer these questions properly it is first neces- 

 sary to know what can be learned about the meteorological aspects of 

 severe windstorms : about their character, frequency, severity, and 

 extent. 



To have the benefit of professional knowledge and opinion regarding 

 windstorms and specifically those of the Northwest, a conference was 

 held with Mr. Wells, head of the U. S. Weather Bureau in Portland. 

 Mr. Wells said that there were no records at all in the weather office 

 on the character and the occurrence in detail of severe windstorms in 

 this region. No special study or observation had ever been made of 

 them. At the few points where complete weather stations are main- 

 tained, such as Portland, Baker, and Roseburg, consistent records of 

 wind velocity and direction have been kept for a number of years ; 

 but no special observation of the unusually strong winds have been 

 made, and at the small co-operative stations throughout the State 

 not even anemometer records are kept. Because of this, Mr. Wells 

 said that nothing at all was known about the periodicity of violent 

 winds. As to tornadoes in Oregon, the Weather Bureau had no 

 records, despite efforts to verify rumors of them which had arisen 

 from time to time. Also it was the opinion of Mr. Wells that tor- 

 nadoes do not occur in the Northwest, because the topography is too 

 unfavorable for their formation. Moreover, tornadoes can be identified 

 by their path through timber which is never more than a few hundred 

 yards wide, and in which the windfalls characteristically lie in one 

 direction on one side and in the opposite direction on the other side 

 of the path. It might be mentioned that this placing of windfalls has 

 never been observed in our experience. Mr. W'ells believes that all 



