New YorK AGRICULTURAL EXPERIMENT STATION. 357. 
plant may use more than it actually needs; that is, the same 
crop may thrive just as well in every respect, when it is supplied 
with less plant food than is indicated by a particular analysis. 
However, multiplication of data of this character is necessary in 
order that some kind of a general guide may be furnished for 
studying the plant-food requirements of crops. We need only 
be on our guard to interpret such data with some degree of free- 
dom and not regard each analysis as representing all conditions. 
The kinds and varieties of fruit trees used in our work were 
the following: Apple, Rhode Island Greening and Baldwin; 
peach, Champion, Elberta and Hills Chili; pear, Kieffer and 
Angouleme, Duchess de; plum, Grand Duke and Italian Prune; 
‘quince, Champion. The trees were selected with reference to 
securing typical representatives in the full vigor of bearing. The 
fruit, leaves and new growth of wood as represented by the tips 
of branches were carefully obtained, weighed, dried and ana- 
lyzed. Precautions were taken when necessary, to avoid the loss 
of foliage, by enclosing the entire tree with mosquito netting. 
A record was kept of the fruit that fell before the final picking. 
The fruit was picked at the stage of ripeness usual in common 
practice. The foliage was left until it showed a tendency to 
drop. ‘The twigs of new wood were removed soon afterward. 
THE AMOUNTS OF PLANT-FOOD PRESENT IN ONE SEA- 
SON’S GROWTH. 
We will consider by itself each kind of fruit used in our work. 
In the stone fruits, the pulp and stone were separated. The 
water given in the tables indicates simply the amount present in 
the different parts of the tree at the time the samples were taken. 
APPLE TREES. 
The following table gives the results secured with the two 
apple trees used in the work. The apple trees were about thirty 
years old. 
