44 TRANSPIRATION IN A DESERT PERENNIAL. 
Experiment X, which was an attempt to investigate this relation in 
another way, brings out the fact that the relative transpiration of a plant of 
potted Parkinsonia microphylla in a soil of 3 per cent moisture is only one- 
sixtieth to one-hundredth the amount the same plant gives when its soil 
moisture is 14 per cent, and that this change takes place without loss of 
leaves. Now, experiment XVII shows that when the same plants, which 
experiment X shows behave similarly under similar conditions of soil- 
moisture, are run together under identical atmospheric conditions but under 
different conditions of soil-moisture, the stomatal diameters in the plant 
having a soil of 3 per cent moisture are about one-seventh of those from the 
plant having a soil of 14 per cent moisture. Evidently a lowering of the 
soil moisture means a lessening of the relative transpiration and also a 
shutting up of the stomata, but it will be noted that the-transpiration rates 
are not proportional to the linear dimensions of the stomatal openings. 
Lloyd’s calculations for Fouquieria splendens* show that the closure of 
stomata is not the cause of changes in relative transpiration rate. This is 
perhaps true of other plants, and with this in view the following hypothesis 
was framed as the starting-point for the work which follows. The morning 
drop is caused by a deficiency in water in the transpiring tissues, due in turn 
to the supply from the lower tissues not being equal to the demand of the 
evaporating power of the air. This deficiency of water in the leaves causes 
a closure of stomata simply because water is drawn from the guard cells in 
the same manner as from the other cells. The diminished size of the open- 
ings would cause a rise of humidity within the intercellular spaces and thus a 
checking of the evaporation rate within the leaf, even though the stomata 
did not close down enough to check the theoretical amount of vapor diffused 
through the openings. Thus the closure may act to further check transpi- 
ration, and meantime the ratio of outgo to intake has lessened and all the 
cells, including the guard cells, gain water and the latter open mechanically, 
after which transpiration goes on more rapidly, but does not reach the former 
high rate unless the ratio of outgo to intake reaches unity. However, in the 
case of young plants without leaves the early maximum and drop occur; 
and in the case of the leafless branch of an adult tree the drop is followed by 
a subsequent rise. It will be seen from table 17 that the stomata of twigs 
remain nearly closed all of the time, and this leads to the conclusion that the 
lessening of the relative transpiration rate before the maximum evaporation 
rate for the day has been reached may take place without a change in sto- 
matal opening. 
In experiment I the readings were not taken at frequent enough intervals 
to discover a rise after the drop, if it occurred. In experiment II a flattening 
of the curve occurs after the drop, but no distinct rise; while in experiment 
IV the rise after the drop is less in amount than in cases where plants in 
leaf were used; this may mean that the closure and subsequent reopening of 
*Lloyd, F. E., The physiology of stomata, Carn. Inst. Wash. Pub. 82, pp. 136-137, 1908. 
