Recent Work on Transpiration. 133 
the size of the pores by deformation of the leaf resulting from 
suction. The air-stream should be drawn through the leaf only 
when readings are actually being taken, since some stomata have 
been shown to be sensitive to the stimulus of the continued 
passage of air. There is also probably a desiccating action to be 
feared. Some stomata are sensitive to the shock produced by 
handling the leaf, and it is therefore advisable to move the plant 
as little as possible, or to allow it sufficient time to recover from 
any necessary handling before taking readings. It was also made 
clear that the resistance offered by the inter-cellular spaces of the 
leaf to the passage of the air-stream could not be neglected, and a 
modification of the form of the cups attached to the leaf was found 
desirable. The experiments also showed that the stomata on different 
parts of the same leaf behave similarly in their reaction to the chang¬ 
ing environment, but the behaviour of stomata on different leaves 
is not so uniform. It is therefore advisable, when dealing with 
a shoot, to fix leaf-chambers to several leaves and to take a com¬ 
posite reading from them all, as has already been suggested by 
Balls. The results of this work showed that with certain modifica¬ 
tions and some special precautions, the porometer method gave 
consistent readings, and it was therefore considered satisfactory. 
Experiments were then conducted on the relation of stomatal 
change, as indicated by porometer readings, to transpiration rate. 
It was recognised as essential that the influence of the external 
conditions included in Livingston’s “ evaporating power of the air,” 
should be either controlled or eliminated. Recourse was had to 
Livingston’s “ relative transpiration ” method (19), but it was soon 
realised that this method was not adequate under all conditions, 
failing as it did to take into consideration the different effects of 
air movements on the rate of evaporation from different surfaces. 
Renner’s work on various evaporating surfaces (26, cf. p. 128), showed 
that the effect of wind on the rate of evaporation from surfaces of 
different areas was not the same. Livingston (18) also showed 
that it was impossible to standardise different atmometers by 
comparison with a unit area of water surface or with any particular 
selected atmometer. It appears, therefore, that “ relative tran¬ 
spiration ” cannot under all conditions be considered to give a 
measure of the transpiring power of the plant. 
The desirability of controlling the movements of air in the 
vicinity of the plant had early been recognised from a priori con¬ 
siderations, and efforts were made to obtain this control. After a 
series of trials with ordinary electric fans, a specially constructed 
