292 BOTANICAL GAZETTE [ocroBeR 
small test tubes. Each stem was carefully sealed into its tube, but 
a capillary glass tube alongside the stem permitted air to enter to 
take the place of absorbed water. To show how much of the total 
loss was due to the cortex, control experiments were made with twigs 
deprived of their leaves. As it was found that the losses through the 
cortex sometimes amounted to 15 per 
cent. of those through the leaves, the 
plan of covering the entire cortical sur- 
face with cacao wax (a mixture of half 
beeswax and half cacao butter) was 
finally adopted. Weighings were made 
on a balance sensitive to less than 5™ 
and the period in most cases allowed for 
transpiration (two to four hours) usually 
secured a loss of weight of more than 
200™ for the least active set of leaves 
employed. Only sunny days were 
chosen for the observations, which were 
all made out of doors. The thermome 
ter ranged, during the season of the 
experiments and the hours of the - 
occupied by them, from 18 to e 
(usually from 20° to 25°). Most ‘ 
the work was done between 12:00 @ 
5:00 P. M., and the per cent. of ie 
| tive humidity at 3:00 P. M. was 17 mi 
| under 55. The determination gird 
. per cent. of total illumination 
Fic. 8.—Sun leaves of Olea: which shade leaves have been develo ; 
the branch stands nearly vertical aeema ta We ae important part of “oe 
and the leaves in general point fi the form, struct 
upward. set of observations On: ge 
ure, or functional activity of suc / 
and many photometric observations were made on bs 2 ayer 
discussed. Unless otherwise indicated, the per cents: eee ne! 
the illumination at or near midday, at the season stated. pane 
the shade leaves studied had grown in about the following a 
of illumination: Quercus, 1.5-5 per cent.; Olea, 4-0 PE 
