1914] GATES—XEROPHILY 450 
of volume and leaf surface was irregular. With leaves of about 
the same size and thickness, the amount of water loss varied pro- 
portionally. In plants of Chamaedaphne from different plant 
associations, where both size and thickness of the leaves varied 
(fig. 2), it did not usually appear that volume was any constant 
function of the leaf area. 
While it is recognized that the measured leaf area is not the area 
of the water-losing mesophyll cells, it is believed that it furnishes 
Fic. 2.—Twigs of Chamaedaphne calyculata, showing the character of the leaves 
developed in the Larix association (bottles 1 and 2) and in the Chamaedaphne asso- 
ciation (bottle 3), at First Sister Lake; April 29, 1911. 
a satisfactory basis of comparison attained without the excessive 
difficulty that would attend the determination of the actual area 
of the surface abutting upon the intercellular spaces. The leaf 
surface, moreover, is the area through which the diffusion into the 
outer air takes place. 
To obtain a knowledge of the evaporating power of the air an 
open dish of water was run with several of the experiments. 
At the close of many of the experiments a section of the small- 
est part of the stem below any transpiring organs was cut out, 
