develop and were
fertile. Some of them presented a peculiarity in growth of the
cotyledons and germ, both of which grew and protruded beyond the
involuere before the nuts were ripe, indicating that the germ had not
come to a state of rest during its usual period in the nut. This freak
appeared in only eight of the nuts, a larger number having normally
resting germs.

In all of these nuts it seemed to me we were probably dealing with
parthenogenesis. In order to make sure that no pollen had been carried
in by any sort of insect, I made check experiments last year, covering
pistillate flowers so carefully that there could be no question about
their having received no pollen. It was found that the chinkapin would
develop nuts freely in this way, and that the bitternut hickory,
shagbark hickory, and pignut (Hicoria glabra) would develop nuts
sparingly in this way.

I speak of the matter as parthenogenesis in advance of microscopic
examination of the ovules,--which will be made next year; but
parthenogenesis seems to be the most likely explanation. If this is the
case, the embryo has not been formed by the conjugation of two gametes,
as generally occurs in the algae and higher plants. It is possible that
the embryo in the unpollenized chinkapins does not originate from the
female gamete at all, but that it originates from a formative budding of
other cells in the ovule. We can speak of parthenogenesis only when the
embryo originates from a female gamete alone, _i. e._, without fusing of
protoplasmic mass of the female gamete with protoplasmic mass of the
male gamete.

Some of the nuts which I am calling parthenogens have developed plants
this year. The chief peculiarity to be observed is great disparity in
size between plants of the same age from the same parent tree. Some of
them grow very much more rapidly than the average plant of the species,
and others less rapidly when subjected to similar conditions of soil,
temperature and moisture.

We assume in biology that one of nature's objects in having two sexes is
to prevent early senescence of the allotment of protoplasm for a
species, and to avoid undue intensification of characteristics of one
parent. This is apparently nature's device for maintaining a mean type.
For man's purposes we may now make artificial selection of individual
plants which represent intensification of desirable characteristics of
one parent. The growing of trees from unfertilized ovules will
ap