464 Report oF rHe HORTICULTURAL DEPARTMENT OF THE 
number from tilled trees. These were weighed and gave 8.7 
grams as the average for the leaves of the sodded trees and 
11.5 grams for those of the tilled trees, making the leaves of 
the tilled trees one and one-third times as large as those of the 
sodded trees. This means, roughly speaking, that the assimi- 
lating powers of the tilled trees, considering size and taking 
no account of number of leaves, is one and one-third times as 
great for the tilled as for the sodded trees. . 
Leafing-time.—In the time of leafing there was a difference 
of three or four days, the leaves of the tilled trees opening first, 
probably because of the warmer soil. The leaves on the sod- 
mulch trees took on their autumnal tints a week or ten days 
earlier than did those of the tilled plat. As a natural conse- 
quence the foliage of the sodded trees dropped that much 
earlier, thus cutting short the growing season of the grassed 
trees and without doubt impairing not a little their future 
vitality and hastening the maturity of the fruit. But the shed- 
ding of foliage lets in the sun to give better color to the fruit, 
an advantage which offsets in some degree the disadvantage of 
early dropping of foliage. 
There was apparently little difference in the blooming time 
of the two plats, though observations did not cover all of the 
years and were not carefully enough made to warrant an exact 
statement in this regard. 
Annual growth of branches.—The annual growth of new 
wood is a most important criterion of the well-being of an 
apple tree, because it is not only a measure of vigor but of the 
bearing capacity, since buds may be formed on any of the 
wood older than one year. This is, however, a most difficult 
measurement to make since there is a wide variation in the 
annual growths of trees and even in different parts of trees. 
The plan first adopted in this work to show difference in 
annual growth was to take new wood from all trees but it was 
found that in this respect “outside” trees were too variable 
and averages therefore were taken only from “inside” trees. 
For the sod plat growths were taken only from the follow- 
ing: Trees 2 to 25 inclusive, rows 2 and 4; for the tilled plat, 
