THE UPWARD TRANSLOCATION OF FOODS IN WOODY PLANTS 287 
removal from the stumps of felled trees, however, may have been due to 
an excess of water following removal of the transpiring surface and resulting 
in a check on respiration or in death. When the stumps were healthy, as 
indicated by the development of shoots, the starch did disappear. Hartig 
failed to state the time at which trees were felled as well as the relative 
number of stumps which retained or lost their food stores. 
The fact that the diameter growth of a trunk was very much decreased 
below a ring is additional proof that there is no large excess of food stored in 
the roots. Hartig explained this weak growth below a ring as due to the 
inability of the food, which he considered as moving up through the xylem, 
to move radially to the cambium. He believed that only that food coming 
through the phloem could be used in cambial growth. 
Other data which have been considered as proof of the use of food from 
the roots for spring shoot growth have been presented by Leclerc du Sablon 
(1906) who determined the effects of ringing at different seasons on the 
amounts of carbohydrates found in roots and stems of a number of woody 
plants. He concluded that, as a general rule to which there may be 
exceptions, the roots of woody plants act as storage organs from which the 
carbohydrates move up in the spring. The data he offered, however, are 
far from conclusive. Some results he obtained in ringing experiments on 
the pear are presented in table i. 
The analyses for April 13 alone suggest that upward translocation from 
the roots might have taken place in the spring and that the ring has pre- 
vented this upward transfer, for in that tree ringed February 9 the roots 
Table i . Data from Leclerc du Sablon to show effect of ringing on distribution of food 
between roots and stems of pear trees. Total carbohydrates expressed as percentage of dry 
weight. 
Date at which Sample Taken 
Not Ringed 
Ringed Feb. 9 
Ringec 
May 8 
Roots 
Stems 
Roots 
Stems 
Roots 
Stems 
Feb. 18 
30-3 
23.0 
Apr. 13 
22.4 
21.3 
25.6 
18.3 
June 16 
27.9 
237 
27.9 
29.5 
17-5 
29.0 
Aug. 4 
29.2 
24.7 
26.5 
33.2 
18.3 
27.0 
Sept. 24 
33.8 
25.7 
19.3 
29.1 
21.4 
29.5 
Dec. I 
29.3 
254 
17.4 
25-9 
17.5 
25.8 
have a higher content than the check and the stems a lower content. But 
in a preliminary series (in 1904) he found similar differences between 
individuals taken at one time. For the quince, samples taken from four 
different plants on March 17 showed a maximum difference in carbohydrate 
content per 100 grams of dry material of 5.7 grams for roots and 4.9 grams 
for stems. With the pear, the corresponding differences were respectively 
2.7 and 1.4. It is true that trees differing in external characteristics were 
definitely chosen for these samples, but similar differences might easily 
