262 BOTANICAL GAZETTE [OCTOBER 
While the root formation on the stem does not inhibit the shoot 
formation on the leaf if the root is exposed to moist air, the result 
is different if the roots are in water. If one leaf with a piece of 
stem (fig. 18) is put into a Petri dish, the bottom of which contains 
a thin layer of water, the stem will form enormous roots (Kr) at 
its basal end from the callus, and R2 from the basal end of the 
shoot (S) which grows out from the bud of the stem where the 
leaf was removed. On the other hand, the leaf has formed only 
a few small roots (R3) atone notch. (As a rule the notches of the 
leaf formed no roots in such an experiment.) In this case the roots 
of the stem which were functioning, and probably established the 
Fics. 18 AND 19 
usual root pressure, inhibited for a long time and in most cases 
permanently the regeneration in the leaf. 
If, however, we do not put the leaf immediately after it is cut 
from the plant into the Petri dish but suspend it first in moist air, 
the stem will form a shoot (S in fig. 19). Roots (R) may or may 
not be formed on the stem, but they will always be formed con- 
siderably later than the shoot, at least in winter. If after a month 
we put the leaf into a Petri dish, while the stem remains in moist 
air, the leaf (fig. 19) will rapidly form roots and shoots. The con- 
trast between the behavior of the leaves in this case and the one 
mentioned before is very striking. In the experiment represented 
in fig. 19, the roots (R) at the base of the stem could not establish 
