2 The Philippine Journal of Science igu 
water through its roots and that in two experiments the roots 
absorbed lithium nitrate, which was then transported to the stem 
and leaves. The passage of water through a submerged plant, 
however, does not show that the roots are of any advantage in 
obtaining nutrient salts, for, in most cases, it is doubtful if the 
water, in which the stems and leaves are growing, contains a 
much smaller percentage of nutrient salts than does the water in 
the soil. Moreover, even though the solution in the soil were more 
concentrated, it seems hardly likely that the cells of the growing 
region and thin leaves would contain a much higher percentage 
of nutrient salts than would diffuse into them from the surround- 
ing water, provided, of course, that the outer walls of the cells are 
permeable, and there is nothing to show that this is not the case 
in most submerged plants. If a concentrated solution did flow 
through the vessels of the stem and leaves, the salts in solution 
would tend to diffuse into the outside water before reaching the 
cells of the growing point and leaves where they would most 
likely be used. 
The writer, in some experiments which will be described in a 
future paper, has observed a rise of a solution of eosin at the 
rate of 4.2 cm a minute in the vessels of a cut stem of Elodea 
exposed to sunlight. The direction in which the solution passed 
through the vessels was found to depend on the position of the 
stem. When the basal end was placed in the solution the latter 
passed from the basal towards the apical end. If the tip was 
removed and the end of the stem nearest the apex placed in the 
solution while the basal end was supported above it, the eosin 
then passed through the vessels in the reverse direction. Bubbles 
of gas also pass through the stem along with the solution. If the 
end which is uppermost is cut the gas escapes as a stream of 
bubbles from this end. When the tip is intact and above the 
basal end the gas escapes in the same manner from the axils 
of one or more leaves. The gas would seem to afford an ex- 
planation of the movement of the water, for, under the condi- 
tions described above, a vessel is a capillary tube containing a 
column of water and gas, and such a column will tend to rise in 
a capillary tube immersed in water. As the column rises more 
bubbles of gas would be formed below and so the flow would 
tend to be continuous. The formation of the bubbles of the gas 
which IS mostly oxygen, would probably produce pressure in the 
vessel and this would again tend to carry the column upwards 
as any downward movement would have to overcome the pres 
