6 3 8 
Journal of Agricultural Research voi. xxni. No. 8 
The larger number of weeds on the unburned area in the third period 
may have been due to the later start of the vegetation on these plots. 
It may be concluded from these data that the treatment of the plots 
had nothing to do with the increase over 1918. If the burning was 
responsible for these changes, there would have been a greater difference 
in the number of weeds on the burned and unbumed areas. 
In 1920, weeds were generally less abundant than in the preceding 
year and there were no consistent differences due to the treatment of 
the plots. The 1921 burning reduced the number of weeds somewhat, 
the differences in each case apparently being significant. 
The data seem to show beyond doubt that burning does not favor 
the growth of weeds but that on the contrary there is a tendency towards 
a decrease in their number on the burned plots as compared with the 
unburned plots. 
THE EFFECT OF BURNING ON SOU, TEMPERATURES 
The principal object of recording the soil temperature was to determine 
if burning resulted in a warmer soil early in the spring, and if so, whether 
this difference could be correlated with the growth of vegetation. The 
thermograph bulbs were placed at depths of one and three inches respec¬ 
tively, because the majority of roots are found at about the latter depth 
and because germinating seed is ordinarily found at a depth of one 
inch or less. 
The thermographs were usually started immediately after burning 
and they were all kept running until June 10. In 1921, they were 
started before burning in order to see if the heat generated by the burning 
vegetation penetrated the soil. The records show that no rise in tem¬ 
peratures took place in either the one-inch or three-inch bulbs, indicating 
that the heat of combustion is confined entirely to the immediate ground 
surface. It seems, therefore, that unless there is considerable accumula¬ 
tion of vegetation the fire does not heat the soil sufficiently to cause 
material damage. .Usually most plants are merely scorched and it is 
probable that weeds and grass seeds which are lightly covered escape 
injury entirely. Moreover if the burning is done soon after a rain, as 
was done in these experiments, the moisture in the ground will aid in 
protecting the seeds. 
The maximum and minimum temperatures were assumed to be the 
most significant temperatures in relation to vegetation and are the only 
ones considered in this paper. The mean maximum and mean minimum 
for each year are presented in Table V which follows. Figures 5, 6, 7 
and 8 have been prepared to show the change in temperature with the 
advance of the season for each year. 
TablK V .—Mean maximum and minimum temperatures at a depth of one inch on burned 
and unburned plots {1918 to 1921) 
"Year. 
Mean maximum temperatures. 
Mean minimum temperatures. 
Burned. 
Unbumed. 
Difference. 
Burned. 
Unbumed. 
Difference. 
1918. 
81.7 
67. 2 
+ 14-5 
50. 2 
50.0 
-0.4 
1919 . 
71. 6 
67. I 
+ 4-5 
54-7 
51-9 
~\~2. 8 
1920. 
88.9 
70-3 
+18.6 
60. 7 
53-3 
+ 7-4 
1921. 
77.0 
66.4 
+10.6 
49.6 
42. 8 
+6.8 
Average. 
79.8 
67.7 
-|~I2. I 
53-8 
49 - 7 
+4. 1 
