64 TRANSPIRATION IN A DESERT PERENNIAL. 
experiment XXI, in which it appears that a plant which shows the dip and 
rise under conditions of high evaporation does not show them under condi- 
tions of lower evaporation. The water content of twigs and leaves suggests 
also a theory for the cause of the rise succeeding une ae which isbased on 
Dixon’s theory for ascent of sap in trees. 
(7) Parkinsonia trees in sunlight show hourly ci hges in the relative 
transpiration rate, the amount of opening of stomata, in water content of 
leaves and twigs, and in leaf temperature, and these have evident inter- 
relations which are held to be governed by the ratio of the demand to the 
available supply of water. 
Since, as is well recognized, the transpiration-absorption water-balance is 
probably the most vital factor governing the occurrence and distribution of 
plants in desert and semi-desert regions, the facts brought out in this paper 
are obviously connected with the success of Parkinsonia as a desert peren- 
nial. The story of the responses of the adult tree to the coming on of 
drought conditions is as follows. First, the leaflets begin closing earlier each 
day until finally they remain open only a few minutes at dawn and at twi- 
light. Second, the transpiration amount islessened with the drying out of the 
soil. Third, the leaflets drop off and later the rachises. Fourth, the twigs and 
small branches begin to die until finally, when extreme drought conditions 
prevail, whole limbs are lost without injury to the vitality of the plant, and 
thus the tree passes through the drought period in spite of the large amount 
of evaporating surface still exposed at the time of the falling of the leaves. 
In addition to the seasonal responses, the tree has a daily response which 
consists in a closure of leaflets, followed several hours later by a lessening 
of actual transpiration rate, while the evaporating power of the air is still‘ 
increasing. This decrease is accompanied by a closure of stomata, a low- 
ered water content of leaves and twigs, and a slight rise in leaf temperature. 
The drop is followed by a rise, but in general the maximum transpira- 
tion is not again reached for that day. That seedlings do not respond so 
readily to seasonal changes is to be expected from the size and structure 
of their leaves, since these differ little from those of green-house-raised 
plants. Seedlings are frequently found in drought seasons with dead leaves 
attached, showing that the leaves were killed outright without falling. 
Perhaps the more mesophytic type of leaf can not endure the “incipient 
drying” to so great an extent as the more xerophytic type found on the 
adult trees. It is hoped that the studies now in progress will throw more 
light on the variations of transpiration and root absorption with | soil 
moisture. 
Pi 
