456 BOTANICAL GAZETTE [DECEMBER 
lower one 76.3°™.. The difference (14.4°™) is 2.8°™ more than the 
pressure of the sap column between the holes. This is the condition 
that would obtain if sap were being artificially pumped into the base 
of the stem and were evaporating slowly from the top. Friction of 
the sap in the ducts would cause the lower gauge to read higher than 
if no current were flowing. 
The gauge in the middle hole was then removed ‘and the sap 
allowed to flow from the hole. A few minutes later the upper hole 
was unstopped, but no sap flowed from it, though sap had flowed from 
it when the hole was made. The flowing of the hole below it had 
caused it to cease to flow. As soon as the middle hole began flowing, 
the pressure in the lower hole dropped rapidly to 23.4°™, and there 
remained nearly stationary for over an hour. The total drop in 
pressure in the lower hole was thus 52.9°™, but in the hole above the 
drop was 61.9°™. In other words, the pressure in the lower hole 
was 11.8°™ more than enough to raise sap to the level of the flowing 
hole. This, also, is a condition to be expected if a current of sap was 
flowing upward from the roots through the stem, overcoming friction. 
The flow from the middle hole was at first rapid—z7 drops in 
ro seconds—but it decreased in a few minutes to 8 drops per 10 
seconds, and at the end of 20 minutes to 4 drops. The flow then 
- continued at nearly this rate for more than an hour. Curves 4 and 
B, D and E, of jig. 5 are plotted from these observations. Curve C 
shows the drop of pressure from a similar experiment on another tree. 
Although the curves are nearly parallel, the ratio of flow to pressure 1s 
greater for the highest pressure than for the lowest. This relation 
may be explained by assuming that the copious flow of the first few 
minutes had a double source of supply; the larger part came irom 
the trunk, being forced toward the tap hole by the elastic expansion 
of the wood and the gas in the wood; and the smaller part came as a 
current from the roots. As soon as the excess of pressure in the stem 
had been relieved, the further and nearly uniform flow was kept up by 
the root pressure. Changes in the degree of cloudiness produced 
the waviness of the curves C, D, and E in jig. 5. 
During the 20 minutes that the flow was decreasing 790 drops 
(66°°) of sap escaped. Of this not more than 40° or less than 25 
could have come from the roots. (This will be seen by a study of 
