BUD VARIATION" IN THE VALENCIA ORANGE. 101 
relatively large number of variable fruits recorded on a few of the 
Productive trees of the Valencia strain, such as those listed in ranks 
1, 2, 6, 7, 10, etc., are almost entirely composed of those described 
briefly under the heading " Minor variations of fruits." Occasionally 
on one of these trees a small branch has been found bearing a few 
fruits of one of the major strains, but usually the variable fruits 
found on Productive trees of the Valencia strain are not of a character 
to lessen the commercial value of the crop. 
The working out of methods of interpreting these individual- 
tree performance records and of applying the knowledge gained 
from their study is essential. It is a simple matter to choose be- 
tween healthy trees of the same age grown under identical cultural 
conditions when they have such widely different records as those 
presented for the trees listed in ranks 2 and 88 in Table I. But to 
make a choice between similarly grown trees with such performance 
records as those listed in ranks 2 and 3 is quite a different problem. 
The average annual production of these two trees for the four years 
was practically identical, 178.2 pounds and 178 pounds, but the 
actual yields each year varied considerably. The tree listed in rank 
2 bore a gradually increasing amount of fruit during the 4-year period, 
while the one recorded in rank 3 bore heavily the first year, produced 
a lighter crop the next season, and an increasingly heavy one each 
succeeding year. The large number of variable fruits occurring on 
the tree in rank 2 is partly explained by the presence in the tree of 
a limb which produces fruits of the Ribbed strain. Both these trees 
appear to be of relatively high commercial value as crop producers 
and as sources of desirable bud wood for use in nursery propagation 
or for top-working trees of undesirable strains, and before any intel- 
ligent choice could be made between the two it would be necessary 
to know their performance records for one or two years more. 
The annual-production records by fruit sizes for these two trees for 
the year 1912 when compared with the records for the succeeding 
years show a great variation in the proportion of fruit of the various 
commercial sizes in the crops of the individual trees. This variation 
was not due to the quantity of fruit on the individual trees, for it is 
equally marked on both these trees, although there is a considerable 
difference in the weight and number of fruit borne by them that season. 
The general production of oranges of large sizes during some years is 
well recognized by citrus growers and is thought to be the result of 
seasonal climatic conditions. An inspection of the annual performance 
records of the other trees presented herewith lends weight to this 
theory, for it will be found that, without exception, all the trees 
produced a relatively small number of small-sized fruit during the 
season of 1912 in comparison with the season of 1913, irrespective of 
whether or not the total yield during that season was larger or smaller 
