258 BOTANICAL GAZETTE [OCTOBER 
was about equal. These experiments have been repeated so often 
that they can be asserted to form reliable demonstration experi- 
ments, during the winter months at least. 
We have already stated that in a completely isolated leaf (as in 
fig. 1) the roots in the notches in the leaf do not begin to grow until 
a few days after the bud in a stem (5S in fig. 2) has begun to grow. 
It would seem, therefore, that we might weaken the inhibiting 
influence of a piece of stem, as shown in the experiment in fig. 2, if 
FIGS. 14 AND 15 
we inhibit or retard the shoot formation of the bud on the stem. 
This can be done, as we have already stated, by not removing the 
other leaf on the stem, as in fig. 3. It is not necessary, however, to 
leave the whole leaf attached to the stem; it suffices if we leave 4 
piece of the stalk of a leaf attached as in fig. 14. In this case 4 
leaf (with a piece of stem and a piece of the stalk / of the other leaf) 
were suspended November 12 in moist air. The bud in the axilla 
of the stalk was by the presence of the latter prevented from 
growing out, and after some time roots (R) were formed at the basal 
