ive them good care and gradually to remove all of the
butternut top.

Each fall, the first two years after I had grafted all these walnuts, I
cut and stored enough scionwood from each variety to maintain it if the
winter should be so severe as to destroy the grafts. Unfortunately, the
grafts had developed so well, even to the actual bearing of nuts by
three varieties, that in 1940 I did not think this precaution was
necessary. Then came our catastrophic Armistice Day blizzard, the most
severe test of hardiness and adaptability ever to occur in the north.
Many of our hardiest trees suffered great injury from it, such trees,
for instance, as Colorado blue spruce, limber pine, arborvitae; cultured
varieties of hickories, hiccans, heartnuts; fruit trees, including
apples, plums and apricots, which bore almost no fruit the next summer.

Although not one variety of English walnut was entirely killed, all,
except one, suffered to some degree, and it was not until late the
following summer that several varieties began to produce new wood. The
variety which showed the greatest degree of hardiness is "Firstling,"
originally known as Letter F. Although the primary buds on the Firstling
were nearly all killed, very few of the small branches were affected and
the union itself suffered no injury. Second in hardiness is Kremenetz,
much of its top being killed, but its union being only slightly
affected. No. 64 was affected in about the same amount as Kremenetz.
Increasing degrees of tenderness and, of course, decreasing degrees of
hardiness, were shown by the many other varieties, some of which may
never recover completely from the shock of that blizzard. The seedling
trees suffered only slight damage so that I expect that they are hardy
enough to produce fruit here.

I cannot conclude this chapter without mentioning certain observations
I have made regarding hardiness, which, although they require more
specific study, I wish to describe as a suggestion for further
experimentation by either amateur or professional horticulturists. My
theory is that a determination of the hardiness factor of an English
walnut tree can be made according to the color of its bark. I have seen
that a tree having thin bark which remains bright green late into the
fall is very likely to be of a tender variety. Conversely, among these
Carpathian walnuts, I have found that varieties whose bark becomes tan
or brown early in autumn show much more hardiness tha