634 Journal of Agricultural Research vol xxiii, no. 8 
inch under the surface and the other at 3 inches. Two others on the 
plot which was never burned also recorded the temperature at depths 
of 1 and 3 inches. The thermographs were kept in operation continu¬ 
ously from the time the field work began early in the spring until the 
latter part of June or the first of July. 
QUADRAT CHARTINGS 
In order to study the effect which the different treatments had on the 
principal plants, 7 small quadrats, each 10 inches square, were laid 
out in each plot. (PI. 2, A.) The quadrats were bounded by a wooden 
frame, across which, in both directions, were woven strands of ordinary 
fine twine, held in place by small nails driven into the frame. These were 
placed at intervals of one-half inch. The twine divided each quadrat into 
400 squares, each one-half inch on a side. These frames were perma¬ 
nently staked down so that exactly the same area would be charted each 
time. Sheets of paper similarly divided were used for recording the 
data, and upon these the observer placed the exact composition as he 
found it in each square of the quadrat. On account of the small size of 
the squares, it was not necessary to sketch the size of each grass tuft 
(PI. 2, B). Instead, the results were compiled by squares for grasses and 
sedges and by specimens for the weeds. For example, quadrat A, 1 
might contain 150 squares of big bluestem, 75 of little bluestem, and 15 
specimens of dwarf fleabane. The method of using the squares as the 
basis of measurement, rather than the exact area occupied by each plant 
is not generally recommended, but in this case it was used, not only 
because it simplified the work but also because of the small size of both 
the individual grass plants and the squares on the quadrat frame. It 
has proved successful and has shown the relative change from year to 
year in a satisfactory way. 
YIELDS 
In order to arrive at the difference in amount of forage actually pro¬ 
duced on the different plots, a number of cuttings were made on each 
area. This was generally done in September. An iron hoop, the rim 
of which included an area of one ten-thousandth of an acre, was used. 
The hoop was permitted to fall more or less at random and the material 
within the borders cut 1 inch above ground with a pair of sheep shears. 
The material was then dried and weighed and the yields per acre 
calculated. 
EXPERIMENTAL RESULTS 
THE EFFECT OF BURNING ON GRASSES 
Quadrats on both the early-burned area and the unburned area were 
charted three times each season. From this data the number of squares 
which contained grasses, sedges, and weeds was compiled. A summary 
of the data on grasses and sedges is given in Table II. The dates of 
burning were March 23, 1918, March 13, 1919, April 2, 1920, and April 7, 
1921. 
