182 BOTANICAL GAZETTE [SEPTEMBER 
reached a length of nearly two millimeters. No distinct tendency 
to the appearance of the new shoots from the region of the mor- 
phological apex of the old shoot could be detected. Generally, 
however, a shoot was formed just back of the apex, but in the 
majority of cases they were produced at other points along the 
stem, and even from the very base. Occurring at the same time 
with the production of new shoots was an abundant growth of 
protonemata from the stem for its entire length. The regen- 
eration by new shoots was always in the way of axillary 
branches, in a manner similar to that which often occurs in 
nature. The protonemata were not, however, confined to the 
leaf axils, but grew as well from cells removed from the axillary 
regions. In the cultures in the light they originated generally 
from the side of the stem which was uppermost, while rhizoids 
were produced from the contact side and in greater abundance 
from the region of the stem which had formed a new shoot. This 
is shown in fig. z2. The cultures in the dark showed very 
rarely a protonema production, and in neither light nor dark was 
any bud formation noted from the stem protonemata. In sev- 
eral cases where the receptacles with the perichaetial leaves were 
placed in culture an abundant protonema production was noted 
from the end cells of the receptacle. A dissection showed these 
protonemata to originate from the cells lying between the base of 
the antheridia, archegonia, and paraphyses, and also from the 
basal cells of the paraphyses as shown in figs. zo, zz. All 
attempts to obtain protonemata from the paraphyses when sepa- 
rated from the stem were without effect. The material for 
growth was evidently drawn from the stem, and when this sup- 
ply was cut off the cells were not capable of independent 
growth. 
In order to determine whether the production of new shoots 
and protonemata was called forth by defoliation or not, whole 
plants were placed in exactly the same conditions as the defoli- 
ated stems. Regeneration by means of new shoots occurred, 
but not in the abundance that was noted in the defoliated stems, 
while no production of protonemata occurred and only occasion- 
