294 
tion in size of these trees as they showed 
the difference in the nursery and con- 
tinue to showit three years after trans- 
planting into the orchard. In the 
orchard they are planted close together 
on uniform soil and are treated alike, 
so the difference cannot be attributed 
to local soil condition or nutrition. 
Is the difference due to the character 
of the bud union? The buds seem to 
have healed nicely in all trees used and 
exhibit no characters indicating that 
the formation of a poor union could be 
considered as causing the difference. 
Is it due to the roots having been 
injured thus resulting in dwarfing the 
tree? The roots were all examined 
when the trees were transplanted and 
all were found to be healthy and 
uninjured. Any injury or disease con- 
tracted since the trees were trans- 
planted could not be limited to the 
small tree rows only. 
Is it due to the kinds of buds used? 
All that can be said regarding this is 
that the buds were carefully selected 
from trees of known record and stand- 
ard type. It does not seem that the 
difference is to be explained in this way 
although this possibility cannot be en- 
tirely eliminated. 
IMPORTANCE OF SEEDLING STOCKS 
The only other factor that is likely 
to be the cause of the variation is the 
influence of the stocks used. The 
sweet orange stock used was merely 
ordinary sweet orange seedlings grown 
from unselected seed, the only extra 
precaution taken being merely to dis- 
card the smallest seedlings when trans- 
planting from the seed bed. About 
15 percent of the total number of seed- 
lings were discarded at that time. The 
universal custom pursued at present is 
to use either sweet, sour, grapefruit, 
lemon or trifoliate orange stock without 
reference to any particular kind within 
these great groups. Are the variations 
within the ordinary lots of sweet and 
sour orange seedlings sufficiently great 
to be assumed to account for these 
variations in size of nursery trees? 
Fortunately some evidence has been 
secured bearing on this point. 
The Journal of Heredity 
In 1915 the writer, with the help 
of Mr. W.M. Mertz and Mr. E. E. 
Thomas, made an examination of one 
sour orange nursery and selected sixteen 
seedlings that appeared to show differ- 
ent characters. At the same time in 
the same nursery four different types 
were selected in a bunch of sweet seed- 
lings. A more detailed examination 
would doubtless have revealed many 
more types but the only object in view 
at that time was merely to add ‘‘freaks”’ 
to our variety orchard. Buds were cut 
from each of these seedlings and two 
sour orange stocks were budded with 
each type. The trees from these buds 
are now five years old from the bud and 
have been set in the variety orchard 
for three years. All of the types se- 
lected present marked differences in size, 
foliage, character of branching and the 
like. The good vigorous types in the 
case of the sour orange selections are 
five times, or more, larger than the 
slow growing dwarf types. An indi- 
cation of this great difference can be 
obtained by comparing the photo- 
graphs of three typical trees shown in 
Figure 3, A, B and C. A represents a 
fine, vigorous, growing sour orange type 
while B and C represent slow growing, 
probably dwarf types. Similar dif- 
ferences in leaf size andshape are also 
exhibited. Compare for instance the 
size of leaves in A with those of C, 
which are both healthy trees. Two 
trees out of 16 of the sour orange types 
selected have lost the typical aroma of 
the sour orange, so far as the leaves are 
concerned. The four types of the 
sweet orange also differ in similar way 
in size and foliage characters. 
The great extent of this range of 
variation within the different species 
is shown equally as well by the large 
number and range of the named va- 
rieties that are grown. 
The sweet orange and sour orange 
seedlings grown for stock purposes are 
usually or at least frequently grown 
from seed, of unknown origin, and 
taken from different trees. We are not 
dealing with a homogeneous lot but 
with lots in which every individual 
differs from every other individual and 
