TIMBER GROWING AND LOGGING PRACTICE IN CALIFORNIA 19 
The problem of protecting cut-over lands involves study of three 
principal questions : 
(1) What .are the causes of the fires, and what can be done to 
eliminate these causes and prevent fires from starting? 
(2) Can fires on cut-over lands be controlled after they start, and 
what can be done to prevent large uncontrolled fires? 
(3) With the foregoing questions answered, what form and 
method of protection will deliver satisfactory results ? 
CAUSES OF FIEES 
Only a few of the fires that start on cut-over lands in California 
are ever reported, and detailed study of the causes of fires is there- 
fore difficult. Of an average of 1,050 fires each year handled by 
the Forest Service in the 12 heavily timbered national forests 
in California, only 56 are reported as due to logging, and of these 
only 13, or 23 per cent, are reported as fires of over 10 acres in size. 
This is certainly a very incomplete record of logging fires. Taken 
at face value it indicates that logging operations are a very minor 
factor of risk, whereas experience and study of operations prove 
beyond question that prevention and suppression of logging fires 
is the main problem in protecting cut-over lands. Until recently 
no record of logging fires on private land has been kept by any of 
the important lumber companies, so that the very meager data now 
available can at best serve only as an indication of the actual situa- 
tion. 
One of the larger companies, which in 1922 made a serious eifort 
to reduce and control fires on its own lands, has the following record 
of fires for one year by causes : 
Logging railroad engines 144 
Donkey engines 32 
Miscellaneous 10 
Total fires 186 
Both the railroad and the donkey engines are wood burners, and 
the great number of fires started by them in spite of the use of spark 
arresters shows clearly the high risk (or likelihood of fires starting) 
owing to the logging operations themselves. The miscellaneous fires 
include a few fires that escaped in cleaning along rights of way, a 
couple of lightning fires, and a camper fire. If these data are 
representative, as they are believed to be, they show beyond question 
that the principal risk on recently cut-over areas is from the logging 
operations. As a corollary of this, it is obvious that the risk areas 
are first, narrow strips along operating railroads; second, narrow 
belts around donkey settings ; third, and probably less important, 
highways through cut-over land which are used by the general pub- 
lic. Undoubtedly, too, some slash fires start from smoking in the 
woods by employees of lumber companies, and from fires started 
for warmth and left unextinguished. 
These considerations lead to the further conclusion that once an 
operation has gone from a particular cut-over area and logging 
railroads no longer touch it, the risk is reduced to the normal or 
blanket risk for that region, from occasional lightning, smoking, or 
camper fires. Experience indicates that such is the case. The great 
