rted
directly from Japan. His trees are growing in the outskirts of
Westfield, Chautauqua County, N. Y., and within a mile of Lake Erie.

In July of this year, Dr. E. A. Scott of Galena, Md., called the
attention of the writer to a number of fine trees in his small town, all
of which had been grown by him from _J. sieboldiana_ seed obtained from
a tree nearby and "every one" of which was bearing "butternuts," as he
and his neighbors call them. The American butternut does not occur in
that part of Maryland which is on the upper end of the Chesapeake
Peninsula, probably 10 miles from Chesapeake Bay. Both black and Persian
walnut trees are very common in that region. The tree which bore the
original seed is a typical Japanese walnut. It stands at the end of a
row of Persian walnut trees along the driveway of a private country
lane. There are several black walnut trees, perhaps 500 yards to the
southwest, but no butternuts for many miles. As the Persian and Japanese
walnuts blossom at about the same time and the black walnut considerably
later, it would seem altogether probable that if any cross had taken
place it would have been Japanese x Persian, rather than Japanese x
black. The chances of a Japanese x butternut cross would have been so
remote as to be altogether improbable.

Many years ago, Judge F. P. Andrus of Almont, Mich., planted one tree
each of Persian and Japanese walnuts in his dooryard. Both soon came
into bearing. Squirrels planted nuts in the ground and presently the
yard was filled with offspring, the majority of which were of the type
now called butterjaps. The trees were extremely vigorous but the nuts
were of so little value that all were finally cut down. Butternut trees
are common in Michigan and butternut pollen may have been responsible
for these crosses but circumstantially the evidence pointed much more
strongly to Japanese x Persian crosses than to Japanese x butternut
crosses.

Other cases of these sorts might be cited, but the evidence which the
writer has been able to bring together up to the present month,
September, 1933, strongly indicates that butterjaps may be due to either
an actual cross with a Persian or black walnut and possibly with
butternut or to reversion to a parent oriental type. So far, it has been
out of the question to hazard a reasonably safe assumption as to the
staminate parent of all particular crosses by merely studying the
botanical characteristics of the butterjap o