19099] MERWIN AND LYON—SAP PRESSURE IN THE BIRCH 453 
thermometer increased 7° C. The corresponding increase in pressure 
was 30°". It is impossible that any part of the tree except the 
smallest twigs could have been heated during so short a time more 
than 3 or 4°. There must have been, therefore, little or no gas in 
the vessels of the tree. 
On April 21, 1906, from 2:00 to 2:45 P. M. (fig. 3), a rise of tempera- 
ture of 4°5 C. was accompanied by an increase of pressure of 20°". 
At sunrise on each of the mornings included in jig. 3 the pressure 
curves steepen greatly. 
During the afternoon of April 23, 1906 (curve B, fig. 3), the pressure 
was rising slowly till nearly sunset in response to a change of the 
weather with rising temperature. 
From the foregoing discussions it may be reasonably inferred that 
there are two chief pressure-producing agencies concerned in the 
phenomenon of sap pressure in the birch stem, namely root pressure 
and thermal expansion. The effects of both are modified consider- 
ably by evaporation and by infiltration of sap into the wood cells. . 
Root pressure and evaporation produce a daily oscillation of pressure, 
with the maximum shortly after sunrise and the minimum at sunset. 
Thermal volume changes in the tree cause a rise of pressure from 
sunrise till shortly after midday, and a fall from then till sunrise. 
Irregular minor oscillations of short period are caused by correspond- 
ing changes in air temperature or brightness of sunshine. The com- 
bined effect of the two agencies is to make the observed maximum 
come about midday and the minimum at sunset. The maximum 
is somewhat higher than would be produced by root pressure alone— 
in extreme cases twice as high. 
If a tree is tapped when the pressure is high, the flow of sap is at 
first copious, but the rate of flow lessens rapidly. The pressure, as 
measured anywhere in the trunk, also declines (see exp. 2 and fig. 5). 
The relation of pressure to flow during this period of falling is different 
for different relative positions of the gauge and the flowing orifice. 
Taking a theoretical case, the pressure as distributed over a radial 
section of a tree before tapping is represented in A, fig. 4. Lines of 
equal pressure are horizontal, and pressure increases downward. 
Shortly after tapping at a, the lines of equal pressure are as shown in 
B. A little later they are as in C. (The diagrams are constructed 
