122 BOTANICAL GAZETTE — [aucust 
the beginning of the experiment. The last operation is the insertion 
of the stem of the plant into the cylinder. This should be done as 
quickly as possible and the exact time noted. The stop-cocks in the 
U-tubes are then turned so as to allow free passage of air, and the 
water is started running from the aspirator. As the water flows from 
the aspirator, a steady current of air is thereby drawn into the transpi- 
ration cylinder, where it passes over the leaves, out through the outflow 
tube at the top, thence through the pentoxid tubes, where both the 
moisture of the normal air and that given off by the plant is absorbed. 
One tube is usually quite sufficient if it is fresh, but since I used the 
same tube a number of times without refilling, it was thought best to 
use a second tube as a guard, and as a means of indicating the exhaus- _ 
tion of the water-absorbing capacity of the first tube. The flow : 
of water from the aspirator was so regulated that each experiment 
lasted approximately an hour. The area of leaf surface, the cubic 
contents of the transpiration cylinder, and the capacity of the aspirator 
in the experiments made were approximately in the proportion of 
1:5:500. Of course the leaf surface varied considerably in the 
different experiments, since it was impossible to obtain the same 
amount each time. It will thus be seen that the air in the transpira 
tion cylinder was changed 100 times in 60 minutes, or once in every 
36 seconds. The rise in humidity in the cylinder due to transpiration 
of the plant can therefore easily be controlled by regulating the 
rapidity of the flow of air through the cylinder or by changing the 
amount of leaf surface inclosed. The exact time when the watel 
ceases to flow from the aspirator is noted, the stop-cocks in the U-tubes 
are cut off, and the plant severed from the parent stem at the point 
where it enters the cylinder. The plant may be kept in the transpira- 
tion cylinder until ready to be weighed. When the flow of air throug® 
the cylinder ceases, the contained air soon reaches the saturatio? 
point and the plant is thus kept fresh until ready for weighing and jor 
putting the leaves in the press for area determination. 
The increase in weight in the pentoxid tubes gives the sum of the 
amounts of water transpired by the plant in the given time, plus the 
water present in the quantity of air used; therefore it becomes 
necessary to know the exact water content of the air. This may be 
found by use of the sling psychrometer, the wet and dry bu 
