d of kernels is
measured by the total weight of kernels in the sample. The first crack
includes only those kernels that either fall out or can be removed
easily with the fingers. The remaining kernels are rescued with a pick
or by recracking. In my judgment, the score accurately measures the
merit of the samples. In the Mandeville, the large size is measured by
the weight of kernels which in part offsets poor cracking quality. Poor
cracking is usually caused by the edges of the halves being curved so
as to be bound in the shell. Much more testing should be done to
determine the value of the schedule.

Opinions regarding the ease of propagation of the butternut differ, but
mostly it is considered difficult to propagate, with often complete
failure. This merely means that the matter is not well understood. In my
own experience I have had just about as many failures as successes, and
must confess that I do not have much idea of what has been responsible
for either success or failure. Best results have been secured by using
inlay or bark slot grafts on stubs about 2 inches in diameter. This
agrees with the experience of Mr. Burgart, of Michigan, and Mr. Weshcke,
of Minnesota, who report that grafts must be made several feet from the
ground and not at the crown.

Shield budding has apparently not been satisfactory. Mr. D. C. Snyder
writes that chip budding is more successful. It is recommended by others
and I agree that grafting should be done early, just as growth starts
rather than later when trees are in leaf. Special care must be used in
tying the new shoots of the graft to braces to prevent breakage by wind
or birds. The butternut wood is very brittle and the grafts are often
lost by breakage. The whole matter of butternut propagation merits
further careful study.

Butternut varieties may be grafted on black walnut, butternut, or _J.
Sieboldiana_ stocks. Mr. Burgart, Mr. Weschcke, and Mr. D. C. Snyder
consider black walnut to be better than the others, giving a more
vigorous long lived tree. Varieties on butternut stocks are apparently
relatively short lived and _J. Sieboldiana_ stocks have a different
growth rate and are not hardy. Mr. Burgart uses bark slot grafts on
black walnut seedling stocks, 2-3 years old.

Butternut trees on their own roots transplant relatively easily because
there is no taproot as with the black walnut and the hickory, and there
are many fibrous surface roots that can be lifted when the tr