564 
Journal of Agricultural Research 
Vol. XXXI, No. 6 
In the summer of 1924 further tests were conducted with yellow 
Cosmos. The seed germinated January 26. The preliminary 
treatment of the seedlings was the same as in the tests of the preceding 
year. The control plants remaining outdoors through the summer 
showed flower buds October 1. Tests involving three different light 
treatments were made: (1) upper portion of plant subjected to 
continuous darkness, and lower portion exposed to 10 hours of 
daylight daily; (2) upper portion exposed to continuous darkness, 
and lower portion allowed to receive all the daylight of summer days; 
(3) entire plant maintained in continuous darkness, and subsequently 
returned to all the natural illumination of summer days. 
The first series, which was carried out to confirm the results of 
the experiments made the preceding year, was begun June 3. The 
S er portion of the plant was restored to the full length of dav- 
t on July 9. Possibly because of the somewhat cooler weather 
wtich prevailed during this time, it was possible to prolong the 
exposure to darkness beyond the lengths of tne exposures in the pre¬ 
vious experiments, without irreparably damaging the darkened part 
of the plant. The lower part of the plant, exposed to 10 hours of 
daylight, showed flower buds June 28, and the first blossom opened 
July 15. In this instance, the upper, darkened portion likewise 
showed flower buds on June 28, and tne first open blossom on July 15. 
The test was repeated, beginning July 9, and in this test it was 
possible to continue the exposure to total darkness for only 19 days. 
The lower part of the plant, in response to the 10 hours of daylight, 
showed flower buds July 22, and first open blossom August 8; while 
the upper, darkened part showed flower buds August 26, and first 
open blossom September 27. 
The first test of the second series was begun May 31. The upper 
portion of the plant was restored to the full length of daylight on 
July 3, after having been darkened for 34 days. The lower portion 
of the plant, exposed to full daylight, showed no flower buds through 
the summer, and the same was true of the upper part which had been 
darkened. Flower buds finally appeared in the top of the plant at 
the same date as in the controls, namely, October 1. In a second 
test, begun July 3, the top of the plant was excluded from light for a 
period of 25 days while the lower portion, as previously, received 
throughout the test the light of the full length of day. No flower 
buds appeared on the plant till October 1. 
In the third series, three plants were kept in darkness from July 12 
to July 22. When returned to outdoor conditions, the plants were 
in a critical condition, due to advanced etiolation, and two of them 
perished. The third individual survived, turned green, and resumed 
vegetative development. No flower buds appeared till October 1. 
DISCUSSION OF RESULTS 
These tests furnish no evidence that darkness in itself is capable 
of initiating flowering. Considering the whole plant, under the 
conditions of the test, exposure of Cosmos to continuous darkness 
failed to result in reproductive activity, while it has been seen that 
exposure to a short period of daylight seemed to promptly promote 
flowering and fruiting, and that exposure to a long period of daylight 
apparently inhibits these processes. Likewise, in the tests in localiz- 
