312 The Philippine Journal of Science 1922 
cautiously opened, allowing the air to be withdrawn from the 
suction system until the desired reduction in atmospheric pres- 
sure was indicated in the pressure gauge. The suction tube 
was then applied to the under surface of a citrus leaf, which was 
immediately submerged in an infusion of the citrus-canker bac- 
teria. The pinchcock was then opened and suction thus applied 
to the leaf surface. This caused withdrawal of air from the 
leaf through the stomata in the same manner as with a poro- 
meter, except that in this case, the leaf being submerged, air 
could not enter, and thus the pressure of the air inside the 
leaf was ultimately reduced to the same amount as in the 
suction tube. 
It was found that, by reducing the pressure in the apparatus 
and leaf sufficiently, the infusion would be drawn into the leaf 
tissues by atmospheric pressure and hydrostatic pressure com- 
bined. The hydrostatic pressure was of course low as compared 
to the differences in pressure between the atmosphere and the 
apparatus, since the dish in which the leaves were submerged 
was filled with liquid to a depth of about 3 centimeters only. The 
‘ presence of water in the intercellular spaces of the leaf could 
only be detected when these were flooded. For this reason it is 
quite probable that small amounts of water might, under pres- 
sure, enter the leaves through the stomata without being de- 
tected. Injection was only recorded when the intercellular 
spaces of the leaf were evidently flooded in spots. This visible — 
injection in most cases occurred at a considerable distance from 
the suction tube (the injector), thus clearly demonstrating that 
the reduced pressure caused by the suction tube extended 
throughout the internal air spaces of the leaf. 
The injection usually appeared at first as a single translucent 
spot on the leaf, which gradually spread and was generally 
followed by the appearance of several other spots at different 
points. If the pressure inside the leaf was not still further re- 
duced after injection was first observed, these spots did not spread 
rapidly nor involve a very large part of the leaf. Frequently 
the spread of the spots stopped after they had attained a diameter 
of 1 or 2 millimeters. From these facts it was deemed probable 
- that still smaller amounts of water might, in some cases, have 
been drawn into the leaf with less extreme ota without 
being detected. 
INJECTION TESTS 
In April and May, 1921, during the dry season, a number of 
tests were made of the pressure required to inject water into 
