23 



square centimeter) that their absence does not materially alter 

 the usual appearance of the tree. Experiments on the tran- 

 spiration of palo verde, not yet published, have shown that as 

 compared with other desert forms, notably ocotillo, it has a low 

 rate of transpiration, and that the possible range of transpiration 

 is likewise small. That is, the "maximum" rate in summer 

 when the leaves are on is not so much greater than the " mini- 

 mum " when the leaves have fallen, as is the case with such other 

 desert plants as have been studied. The foliar history of the 

 palo verde is in complete accord with this observation. The tree 

 forms leaves slowly and retains them a relatively long time. 

 However, in cases of severe drought the leaves are shed and the 

 necessary adjustment of the rate of transpiration is accomplished. 



By the time of the summer rains in 1904 the leaves had fallen 

 from the palo verdes near the Laboratory ; after the rains had 

 come leaves were again organized. This course of events, which 

 may perhaps be the usual one, was not followed this season. 

 Possibly owing to the extraordinarily heavy spring rains, and in 

 part to small rains in June, the leaves were carried through the 

 dry portion of the earlier part of July until the time for the sum- 

 mer rains had come, and then, despite the fact that the rainfall of 

 summer was a meager one, they still persisted. With little doubt 

 the reduced amount of rain of the summer would not under ordi- 

 nary conditions of the atmospheric moisture be sufficient to per- 

 mit the retention of the leaves. That they were retained is in 

 the main due, I believe, to the high humidity which prevailed at 

 the time. 



A more striking example of the influence of high relative 

 humidity in extending the life of leaves is found in the ocotillo. 

 As was stated in an earlier paragraph the ocotillo when leafless 

 responds to an improvement in its water conditions by quickly 

 putting on a leaf-covering, and when dry weather returns this is 

 nearly as quickly exfoliated. It is of interest to note also that 

 this plant has a very high " maximum " rate and a very low 

 "minimum" rate, and, therefore, that the possible seasonal vari- 

 ation is considerably greater than that of palo verde, for instance. 

 The ocotillo then usually drops its leaves during a moderately 



