WATER CONTENT OF PLANT PARTS. 45 
the stomata aided in the rise but are not necessary to it. So, then, a defi- 
ciency of water in the evaporating tissues would be the first cause of the drop 
in transpiration rate, while the evaporation rate is still increasing. If now 
the supply of water from below is unchanged, the smaller evaporation rate 
would allow a balance to be obtained when we should expect the relative 
rate to become a straight line for a time at least. It is hard to see how the 
gain could be sufficient to send the relative rate up again unless some other 
factor entered, either to check the rate of outgo or to increase the rate of 
intake from the lower parts. If Dixon’s hypothesis* of the ascent of sap be 
accepted, it might be supposed that the deficiency of water in the terminal 
tissues causes a “pull” on the water columns, which enter the tensile state, 
and thus the intake is increased. As soon as the supply begins to catch up 
with the demand the tensile pull is lessened and the relative rate begins to 
go down again. 
If the explanation is the latter one, based on Dixon’s hypothesis, then 
there might be a relation between the moisture content of different parts of 
the tree which would throw light on the matter. Consequently a series of 
moisture content determinations was carried out as, follows. 
DAILY COURSE OF WATER CONTENT OF LEAVES, TWIGS, AND STEMS. 
EXPERIMENTATION. 
ExpERIMENT XVIII. 
Three adult trees, designated I, II, III, were selected in open sunny 
places, and at two-hour intervals small limbs were sawed off and samples for 
the determination of water-content were taken as follows: Ij, IJ,, and III, 
were end twigs about 3 to 5 cm. long, with diameters from 0.1 to 0.2 cm.; 
I,, Iz, and III; were taken about a meter from the end twigs and had diam- 
eters about 1.2 to 1.6 cm.; Is, Iz, IIs, were taken 2 meters from the endtwig 
and were from 3.5 to 3.9 cm. in diameter. The work of sawing and cutting 
was done very quickly, with the aid of an assistant, and the pieces were 
weighed immediately after the cuts were made. The samples were allowed 
to dry in the sun for about a week and then dried to constant weight in an 
oven at 95° to 100° C. Accidents which happened to the end twigs of II 
left too few good readings for the results to be plotted. For the sake of a 
comparison with the water-content of a tree which had lost no branches, 
samples were taken from a fourth tree at 2" 30™ p.m. The two readings 
marked III, and IV, were taken about 4 meters from the tips, where the bark 
is brown and not green as in the other samples. In the sawing great care 
was taken to get branches widely separated on the tree, in order that the 
wounds might cause as little disturbance as possible. The evaporation rate 
was measured by exposing atmometer No. 11-1 in an open sunny place. 
*Dixon and Joly, On the ascent of sap, Proc. Roy. Soc. Lond., vol. 57 B. 
