698 
Journal o f Agricultural Research 
Vol. XXX, No. 8 
Table IV .—Important climatic characteristics of the several national forest sub- 
* Data based on Bulletin Q, Weather Bureau, > Data lacking. 
United States Department of Agriculture. * Figures are for April-May-June. 
6 S. in southern part, NW. at north end. 1 Range, 130 to 180 days. 
c NW. and S. in western part, S. and SW. in east. m Figures are for Oetober-November. 
* Range, 85 to 120 days. ” Minnesota, 90 days; Michigan, 120 days. 
* Range, 65 to 105 days. • Minnesota, 37°; Michigan, 41°. 
f SE. and S. in northwest, W. and SW. in south, r Minnesota, 49°; Michigan, 52°. 
» E. and SE. in May. « Minnesota, 105 days; Michigan, 130 days. 
* SE. in May, July, August, September. ' Minnesota, N. in spring, S. and SW. summer and 
■ Range, 95 to 130 days. fall; Michigan, SE. in spring and fall. 
acre separately. If, however, numbers 
of factors can be grouped together in 
such a way as to reduce the number of 
items that must be considered in 
rating, and so as to make it possible to 
use the data concerning the history of 
past fires, which are already at hand 
or obtainable, it may be possible to 
develop a method that can be applied. 
This involves the principle of classifica¬ 
tion of risks, somewhat analogous to 
that used in the insurance business, 
where rating is based on the probable 
losses for a class of risks, rather than 
for each individual risk separately. 
CLASSIFICATION OF RISKS 
The object of such a classification 
should be to throw together into one 
class all forest tracts whose factors of 
risk are so substantially similar that 
the probable fire loss and suppression 
cost per unit of area, over a period of 
years, will be fairly uniform. Because 
of the nature of the data available for 
study, and also in order that ratings 
can be applied in working out actual 
protection organizations, the classifi¬ 
cation must be along rather broad lines, 
with a minimum of detail. Accord¬ 
ingly, the following general scheme was 
followed: 
1. To allow for general difference in 
climatic and seasonal factors and the 
resultant differences in general forest 
conditions, the western national forest 
region, exclusive of Alaska, was divided 
into 21 subregions, mainly on the basis 
of climatic characteristics. (See Table 
IV and figs. 2 and 3.) These sub- 
regions are as follows: 
1. Northern Rocky Mountains. 
2. Western Montana. 
3. Northern Idaho. 
4. Eastern Washington. 
5. Western Washington and Oregon. 2 
6. Southern Oregon and Northern California Coast 
Ranges. 
7. Southeastern Oregon and northeastern Califor¬ 
nia. 
8. Eastern Oregon and southwestern Idaho. 
2 The study did not cover Alaska or the eastern national forests. 
