254 BOTANICAL GAZETTE [ocromse 
into a vessel containing a 3 per cent. solution of ether. This seemed to 
anzsthetize the stem without killing it, and roots appeared just above 
this portion. The effect of the anesthetic probably was to prevent 
any passage of stimuli through this part of the stem either up or 
down. Experiment 30 is instructive in this connection. It will be 
recalled that surrounding 
the stem at any place 
with water in glass cylin- 
ders will not start roots 
at that place if the con- 
nection with the roots 
below is unbroken; but, 
as shown in fig. 7, if a 
notch is cut in the stem 
some distance below, 
thus severing connection 
with the roots, roots will 
appear above the notch 
in the water, apparently 
coming from those bun- 
dles severed by the notch. 
In other words, the water 
supplies a favorable con- 
dition for root develop- 
ment, and the cells are 
able to act as soon as the 
connection with the roots | 
below is broken. If the 
air be moist, roots come | 
out also immediately | 
Fic. 8.—Phaseolus: Stem cut off at base; lower above the notch. cee 
end submerged. Portion of next internode above When a stem 18 | 
oo by water as in fig. 4, and stem severed; ranged as in fig. 8, roots 
che -- slightly separated. Roots appeared always arise on the part 
just above the cut, bi 
on the part just below it, even though this piece is inverted with its 
end in water, as in fig. 9. This difference in the behavior of the cells 
