10 BULLETIN 1495, U. S. DEPARTMENT OF AGRICULTURE 
in the chaparral type, and by 13.79 acres in the brush-field type, in 
both of which most fires are crown fires. 
Surface fires generally develop where an irregular low canopy of 
flash fuels exists, such as characterizes the grass, woodland, western 
yellow pine, mixed conifer, and Douglas fir types. In the first two 
of these, grasses and weeds are the most important fuels, and the 
average rate of spread is greater than in the commercial timber 
types, where needles and twigs are the important fuels. This differ- 
ence in fuels is reflected in the average rates of spread of 17 and 5.3 
acres per hour, respectively, for the grass-woodland and timber 
roups. 
‘ Ground fires occur where a compact and poorly aerated layer of 
humus and duff permits only a small supply of oxygen to reach the 
fire and where there is a scarcity of flash fuels. Under normal con- 
ditions ground fires eat their way slowly. The observed rate of 
spread for sugar pine-fir is 4.33 acres per hour, and for fir, the lowest 
in the entire scale, 1.07 acres per hour. 
THE FIRE PROBLEM 
NEED OF AN INDEX OF FIRE DANGER 
Most of the information heretofore available on forest cover types 
has dealt with climatic or botanical differences and has been of little 
direct value in attacking the fire problem. That a study of types 
from the fire-control standpoint is exceedingly difficult is evident 
from the great differences in age, height, and density of the mature 
timber, amount of young growth present and other variables, even 
within what is classed as a single major type. ‘Thus, at first glance 
it might appear that any attempt to isolate the cover-type factor as 
it affects the start and spread of fires is foredoomed to failure unless a 
very large number of subdivisions can be recognized. This would be 
a theoretically correct approach to the problem, but the results of 
such a study could find little use until a detailed ‘fire reconnaissance” 
of forest areas should recognize and map similar types. No such de- 
tailed classification of hazards is now in prospect. 
The practical and immediate job of fire control has been the reduc- 
tion of area burned and in most cases it has been impossible to recog- 
nize values demanding special intensive protection. Moreover, it 
has been necessary to provide much protection in types of generally 
very low intrinsic value, such as the chaparral and grassland, since 
these are of great value in preventing erosion, and also are contiguous 
to or are intermingled with commercial forest or restocking brush 
fields, with no natural or artificial barrier to prevent fires originating 
on the low-value lands from spreading on to more valuable properties. 
It will only be as more funds become available for fire control in the 
future, and as the whole level of protection effort is greatly raised, 
that a more detailed classification of risks, hazards, and values will be 
possible in fire control. 
At present not even an approximate expression of the magnitude of 
differences between types has been worked out, although it is com- 
monly recognized that the fire season is longer and that fires spread 
more rapidly in cover types of the lower elevations than at higher 
altitudes. What is needed at once is an index, even though crude, of 
the difference in fire control between the major cover types. 
©} | 
