io8 
OTIS F. CURTIS 
ing from a deficiency of osmotically active substances, is not clear. From 
the available data it would seem, however, that the former is of first im- 
portance. In a number of instances it was found that the part above a ring 
would grow slightly, and later would wither completely when the compe- 
tition for water became more severe. Table 6 indicates, that, as would be 
expected, the ringing not only decreases the concentration above the ring 
but also tends to increase that below. 
These few determinations cannot be fully relied upon for all details, but, 
assuming that they are approximately correct, it can be readily seen why the 
part above a ring cannot compete for water with other parts unless it retains 
its leaves. This therefore would explain the withering of ringed shoots 
that lack leaves, and shows that the leaves do not directly draw water to 
the phloem as suggested by Hanstein, but rather that they supply the 
phloem with osmotically active substances, thus indirectly enabling it to 
obtain its own water. 
It might be stated in this connection that the writer has obtained con- 
siderable evidence that one important factor among several possible factors 
that may be concerned in the matter of polarity and inhibition is associated 
with a local* distribution of foods and of osmotically active substances. 
The same substance, possibly sugar, may act in both roles. It has been 
found, for instance, that the upper part of a rapidly growing shoot may have 
a concentration that would give a pressure over two atmospheres greater 
than the concentration in the middle or lower part of the same shoot. 
The "inhibition" of shoot growth at nodes below the terminal one may be 
due to a lack of sufficient food and to inability to compete successfully for 
water rather than to a backward flow of some "inhibitor." In fact, as was 
shown in a previous paper (Curtis, 191 8), it is possible to reverse the 
polarity by placing the base of a shoot in a strong sucrose solution. The 
writer has found that other substances are even more efficient than sucrose 
in thus altering the polarity of a shoot. The check in growth following 
ringing cannot be due to retardation in removal of an "inhibitor" produced 
by the leaves, for, if leaves remain, the growth is greater than if they are 
removed. 
The Translocation of Food to Fruits 
It has been recognized that at least part of the food carried to growing 
fruits probably moves through the phloem. Hanstein (i860) found that 
ringing below fruits, if no leaves were left above the ring, resulted in checking 
their further development. In order to determine the effect of ringing on 
the transfer of food to fruits, twelve fairly well matched pairs of young 
Wealthy apples were selected. The leaves from the stems close to the fruit 
were removed, and the stem of one of each pair was ringed so that no leaves 
were above the ring. The greatest length and the greatest diameter of each 
fruit were taken. The same thing was done with three pairs of Rhode 
