any of these,
equally simple and inexpensive, and so much better in its action that
scions may be kept by it two and three years in about the same condition
as when severed from the parent tree; and to prove this statement I have
here with me for your examination scionwood of several kinds of nut and
fruit trees that have been kept in the Harrington graft box one year and
two years. At the present time I have no older wood in my graft box, for
the simple reason that in the summer of 1928 the cover of the box, which
had been in several years, rotted so that the top caved in, leaving it
open to too much air, thus in time spoiling what wood was in it; and
before putting in new wood in November I had to dig out the old box and
replace with a new one. For wood will rot in time in the ground. I have
had, at different times in the past, scionwood in my box three years
old, much of it seemingly still good. I have not used any of it for
grafting at three years, but I have with good success the second year
old from cutting. I started experimentally with this method and box
thirty years ago and there has not been a year since in which I have not
used it, so you may readily understand that it is not an untried theory
I am giving you. A much valued member of our society, J. F. Jones of
Lancaster, Pa., now deceased, wrote me at one time, "You undoubtedly
have the best method of keeping scionwood known at the present day," and
Prof. Close, head of the Pomology Department of Agriculture, Washington,
D. C., made the same statement to me.

My own box is located in an evergreen grove on dry land, but a shady
position to the north of a building might answer fairly well. Until the
last eight years my box was for a long period, under and between two
large butternut trees growing out in the open, except at the northward.
In my opinion it is highly desirable to cut and store all scionwood
before severe temperatures of the winter occur, preferably between
Thanksgiving and Christmas because very severe freezing is liable to
produce some little injury to the cambium layer, at least in some years,
and if that injury be even very slight it will usually spell failure
when used.

The graft box, as I am using it, is about thirty inches long by eighteen
inches deep and fifteen inches wide. It has a solid cover but has a six
inch square hand hole through on top in front, covered by a loose board
lying flat and about ten inches square and butting back