Mar. 17,1923 
Further Studies in Photoperiodism 
883 
the io-hour day as compared with the 13-hour day. By referring to 
Table I it will be seen that a 13-hour day delayed flowering in varying 
degrees in different varieties of sorghum as compared with a io-hour 
day. As previously noted, flowering in chrysanthemum and poinsettia, 
also, was delayed by a 13-hour exposure as compared with a io-hour 
exposure. As a rather extreme case reference is made to the behavior 
of wild sunflower as described on page 875. In this instance the unfolding 
of the flower buds was delayed for several months by the long daily light 
exposure. 
As the light period is shifted away from the flowering optimum and 
toward the optimum for vegetative development one of the effects is 
cleistogamy. Thus, the sparse flowering of Mikania in spring, previously 
referred to, when the increasing day length is moving toward the optimum 
for vegetative activity, is usually associated with a cleistogamous type 
of blossom. The cleistogamous flower of the violet during the long 
summer days is to be regarded as a step away from the open, blue, 
colored blossom of spring and toward the purely vegetative form of 
development. These results are in line with observations of Bouch6 
(4) on certain species, reported in 1875. 
CHANGE FROM THE REPRODUCTIVE TO THE PURELY VEGETATIVE FORM 
OF ACTIVITY 
It is of interest to inquire further into the effect of changing the plant 
from the optimal light exposure for sexual reproduction to one approxi¬ 
mating the optimal for vegetative activity. If made in the earliest 
stages of development of the plant the change in light exposure would 
have little or no qualitative effect, while if made only after senescence 
has reached an advanced stage the only action of importance would 
necessarily involve rejuvenescence. This latter condition will be con¬ 
sidered in later paragraphs. 
Plantings of Cosmos bipinnata Cav. in 12-quart galvanized iron buckets 
germinated July 5, each bucket containing about a dozen seedlings. 
At the outset all plants were exposed to a io-hour day, a light period 
which quickly forces flowering. Beginning July 6, one bucket of seedlings 
was transferred on each alternate day from the io-hour day to the full 
daylight exposure of summer at Washington, D. C. All control plants 
remaining under the io-hour day were showing flower buds on July 26. 
All plants transferred to the long-day conditions on or after July 18 
likewise were showing flower buds on July 26, while the plants transferred 
during the period July 6 to 16, were showing no flower buds as late as 
September 2. It is evident that an exposure to a io-hour day for not 
less than 10 days is necessary to bring the flower buds into evidence. 
The dates of first open blossoms ranged from August 9 for the plants 
remaining under the io-hour day till July 26, to August 21 for those 
transferred to the long-day conditions on July 18. A notable feature 
was that of the numerous flower buds formed on the plants which were 
removed from the io-hour light exposure on July 18 and July 20 only 
a few were able to open, the others being permanently suppressed, a 
result which has often been observed in cosmos and other plants. The 
foreshortening of the primary axis, with resultant crowding together 
of nodes, associated with the formation of the terminal flower bud, is 
followed by considerable swelling of the nodal tissues immediately be¬ 
neath the bud as a result of the stimulus from the longer day. New 
