376 
AMERICAN JOURNAL OF BOTANY 
[Vol. io, 
were removed, leaving the twig bare as indicated in figure i. On one twig 
from each set (no. i) a ring was made immediately below the upper leaves. 
On a second twig (no. 2) in each set two rings were made, one just below 
the upper leaves and a second at the base of the defoliated part. The third 
twig (no. 3) was ringed at the base of the defoliated part and not at the point 
just below the upper leaves. Sodium nitrate was added to the soil around 
Fig. 1. Showing method of ringing so that the amount of water passing through the 
defoliated part of each stem would be approximately the same whether or not this part is 
isolated from the roots by a ring. 1. Check, medium carbyhodrate content in defoliated 
part. 2. Ringed, low carbohydrate content in defoliated part. 3. Ringed, high carbo¬ 
hydrate content in defoliated part. 
the bushes at the time of ringing, August 25, and the twigs were cut for 
analysis on October 3. An analysis was made of the defoliated part of each 
stem as indicated by brackets in the figure. It was assumed that the 
amount of water moving through the stem would be approximately the 
same in numbers 1 and 2, since both had the same number of leaves on the 
stem at the apex and both were ringed immediately below these leaves. 
Since the defoliated part of the stem in no. 1 W'ould probably have a 
higher sugar content than number 2, as it was not isolated by a ring from 
the leaves at the base, and since any such difference in carbohydrate con¬ 
tent might distinctly alter its tendency to retain nitrogen, the third twig 
