1878.]| On the Transpiration of Plants. 163 
done, the line of separation at the point where the edges of the 
cloth met, was remedied by allowing an overlapping of two inches 
or more and sealing by means of gum mucilage. It was now 
thought that evaporation from the vessel in which the plant was 
situated was next to impossible; but the question next arose, 
“ How is the plant to be supplied with the necessary moisture ?” 
This difficulty was overcome by taking a hollow cylinder of tin 
3-4 inch in diameter, and about 3 inches in length, and having 
made a hole of sufficient size in the cloth covering the pot, a few 
inches from the stem of the plant, introducing one end of this 
tube into the opening, the rubber cloth was tucked up and tied 
on it the same as in the case of the stem of the plant, the external 
opening of the tube being guarded by means of a cork. 
It would be useless, as well as illogical to assert that this ar- 
rangement would allow of no escape of moisture whatever, yet 
there is perfect safety in affirming that the quantity thus lost sinks 
into insignificance compared with the amount actually trans- 
pired by the plant itself. The loss by insecurity of this method 
could certainly not exceed a few grains per day. 
The plants were watered in the morning before weighing them 
_ for the day’s experiment, and just sufficient water was given them 
= to keep them in a natural state, the condition of the leaves being 
in all cases taken as a guide. After watering the plants in the 
_ Morning they were carefully weighed and then placed in the de- 
sired position and left undisturbed till evening, or any number 
_ of hours desirable, and then were again weighed (as a rule before 
any more water was given). ` The loss of weight, as before stated, 
was considered equivalent to the amount evaporated during the 
time of the experiments. 
Usually the observations were made for a day and a night, but 
the plants were also weighed in the evening so as to establish the 
relation between night and day evaporation during the same 
twenty-four hours. As before intimated the relationship between 
the dew point, temperature, etc., and the rapidity of transpiration 
was noted in most of the observations made. This was arrived 
pea i = 
rule, which it would be needless to detail here. 
_ With this brief yet, it is hoped, sufficiently comprehensible 
description of methods pursued, we shall pass to the notice of re- 
sults obtained me these arce 
