60 
fornia walnut blight, and numerous ex- 
periments in cross breeding and hybrid- 
ization should be made to develop the 
good parent trees which we now have 
reason to believe can be produced by the 
breeding from many of the promising 
trees already known to us. 
THE PROBLEM OF TREE CROPS 
If chance hybridization and selection 
in the past has brought the worthless 
wild tree described by Sargent to the 
present perfection of the commercial 
Persian walnut, we certainly have great 
reason to anticipate large results from 
systematic hybridization if we can focus 
the resources of constructive science 
in that direction. 
The Persian walnut is but a type in 
the whole tree crop question. Its 
typical aspects may be summarized as 
follows: 
1. As acrop producer it is almost worth- 
less in its wild state. 
2. Some trees usually widely scattered 
are much superior to the average and 
worthy of propagation. 
3. Cross-bred strains from selected rare 
The Journal of Heredity 
(such have been produced in the gardens 
of Mediterranean lands). 
4. The possibilities of better varieties 
by hybridization have merely been 
glimpsed, not realized, and demand im- 
mediate work. ; 
The plant kingdom has never been 
systematically searched for useful plants. 
This is particularly true of trees, from 
which we have held back because of the 
time element; yet there is little doubt 
that forty or fifty species of wild trees 
are quite as promising of a good agri- 
cultural crop as was the wild walnut 
described by Sargent. By selection, 
propagation, cross-breeding and hybrid- 
ization, each of the fifty or more can 
probably, like the walnut, be made into 
a valuable crop producing food for men, 
and what is more important, agri- 
culturally, food for the beasts, if the 
problem is systematically handled. This 
field of endeavor also promises one of 
the most potent cures for the erosion 
problem. The hillside whose soil is 
pinned to the bed rock by the interlacing 
roots of crop yielding trees will not 
trees are much superior to any wild tree erode. It will stay and pay. 
Breeding Native Grapes 
Valuable work on self-sterility has been done at the North Carolina Experiment 
Station, principally with native grapes, but to a less extent with persimmons, 
blackberries and dewberries. It is found that the latter berries are in some cases 
self-sterile and in some cases self-fertile; grapes of the muscadine type (Vitis 
rotundifolia), however, have proved to be self-sterile in almost every instance. 
It is therefore necessary that the grower of such grapes should plant enough male 
vines to furnish pollen; and it has been found that there are two distinct wild types 
of male vine, the commoner one producing grapes that, when further propagated 
by seed, yield dark-colored fruit, while a rarer type, when perpetuated, yields 
light-colored fruit. All scuppernong grapes of the immediate generation are light 
colored. By the growth of several thousand hybrid seedlings, an effort is being 
made to find whether such characters as color of berry, persistance of holding 
fruit and size of fruit and size of fruit clusters are transmitted, and if so, by what 
laws. 
In this connection it will be of interest to mention that the United States De- 
partment of Agriculture is engaged in breeding a strain of muscadine grapes with 
perfect flowers, starting with a single individual of this character which was found 
among a lot of seedlings. The project promises to be wholly successful, and will 
make the production of these grapes much more profitable. Hybridization is also 
being used to produce varieties of grapes that will combine the vigor and disease- 
resistance of the Vitis rotundifolia and its large berries, with the hardiness and 
large clusters of the northeastern United States species (Vitis labrusca, etc.). 
