io8 
FLORIDA STATE HORTICULTURAL SOCIETY. 
RAINFALL AND COMPOSITION OF DRAIN¬ 
AGE WATER. 
Water began to come through the 
tanks on July 30th, and the first collec¬ 
tion was made on August 1st. Between 
this date and October 19, six other collec¬ 
tions were made, the seven collections av¬ 
eraging about 22 to 24 gallons each per 
tank. From July 7, when the tanks were 
opened, to October 19, the total rainfall 
was 24.11 inches. Computing, from these 
figures for the two thousandth part of 
an acre, it is found that the amount of 
rainwater falling on one of the tanks dur¬ 
ing this period was 330J4 gallons. The 
average amount collected from each tank 
during the same period was 160J4 gal¬ 
lons. It will thus be seen that nearly one- 
half of the water that fell on the tanks 
was collected as drainage water. No 
doubt the amount would have been some¬ 
what less had a crop of weeds or grass 
been allowed to grow over the entire sur¬ 
face, and had the trees been making a 
normal growth. 
We have made fairly complete anal¬ 
yses of these samples, and from these re¬ 
sults have calculated the total amount of 
fertilizing constituents washed out of the 
soil, and then converted these results into 
pounds of commercial materials per acre. 
For example: the total nitrogen found 
in tank No. 1 was calculated to its equiv¬ 
alent in nitrate of soda, and this amount 
multiplied by two thousand to bring it to 
pounds per acre. The results are set forth 
in Table I: 
LOSS OF NITROGEN. 
From this it is seen that nitrogen com¬ 
pounds are lost in much larger amounts 
than any of the other materials. This 
was to be expected, inasmuch as these 
compounds are not held in the soil in the 
same way that phosphoric acid and pot¬ 
ash are, but we had not expected to find 
such a wide difference. In this case, how¬ 
ever, it is probable that the loss is greater 
than it would be under ordinary field con¬ 
ditions, on account of the soil having been 
thoroughly stirred and aerated before be¬ 
ing placed in the tanks, and the subse¬ 
quent abundant rains having made the 
conditions for nitrification especially fa- 
favorable. However, the loss is prob¬ 
ably but a fair indication of what we may 
expect, and some light is thrown on the 
question of the time when these losses 
are heaviest. Since October 19th, 1910, 
the rainfall has been comparatively light, 
and the amount of water passing through 
the tanks correspondingly small. As a 
result, the loss of fertilizing constituents 
is very much less than during the rainy 
period. This accords with results ob¬ 
tained at Rothamsted, England. There 
the heavy rains come during the winter 
months, and it is found that the loss of 
nitrates is great when nitrogenous mate¬ 
rials are applied in the fall, but is reduced 
to a very small amount when these mate¬ 
rials are applied in the spring. There was 
found also a far greater loss of nitrates 
than of phosphoric acid and potash. 
LOSS OF OTHER CONSTITUENTS. 
Next after nitrogen, lime and magnesia 
are lost in greatest amounts, while potash 
and phosphoric acid are lost in compara¬ 
tively small quantities. 
The accompanying diagram shows the 
amounts, in parts per million parts of 
