450 
Journal of Agricultural Research 
Vol. XXVIII, No. 5 
tassels in August and September, and opened first blossoms from September 23 
to October 15. The August buds required on the average, 49 days to open, and 
the September buds, but 29 days. Group 5, however, with its artificially short¬ 
ened 12-hour daily light exposure, showed tassels from June 27 to July 18, and 
opened first blossoms from July 20 to August 4, the interval between the first 
observance of tasseling and the date of blossoming averaging but 18 days. 
If the autumn blossoming were merely the result in due time of the long summer 
days alone, group 4, which was subjected to an artificially lengthened day in 
spring merging into the long days of summer, would have been expected to 
blossom before the normal blossoming season. Four of the 12 plants in this 
group died before October, and the others opened their first blossoms between 
September 23 and October 15. 
Group 5 was subjected to the same artificially lengthened day in spring, but 
the daily light exposure was shortened to 12 hours after June 6. The plants in 
two cans in this group failed to make vigorous growth, one dying of a root dis¬ 
ease in August, and none blossoming. The remaining nine plants opened their 
first blossoms between July 20 and August 4, which was 7 to 9 weeks ahead of 
the earliest blossom in group 4. 
The behavior of these two groups showed that blossoming was inhibited so 
long as exposure to the long days of summer continued, but that it was promptly 
induced by shortening the length of the daily exposure to 12 hours. 
Plants in cans No. 1 to 18, inclusive, A series, were given a 10-hour day 
exposure at first, and all passed the autumn blossoming season, 1922, without 
flowering. The plants in 13 of the cans were of sufficient age to blossom, how¬ 
ever, and their controls did blossom. Blossoming had been completely inhib¬ 
ited by a 10-hour day length. 
In November, these 18 cans were divided into three groups, group 1 continu¬ 
ing with a 10-hour day, group 2 receiving a 12j-hour day, and group 3 a 13J- 
hour day for 10 weeks, at the end of which time budding was not evident. The 
artificially lengthened day was then shortened by one-half hour for each group. 
After four weeks, it was again shortened by half an hour for group 3. Two 
weeks later it was brought to a 12-hour length, where it was maintained for 
both groups. 
A week after the first reduction in day length, budding appeared to be start¬ 
ing. A week later, budding was noted on 7 of the 14 plants in group 3 (1 in can 
No. 12, and 3 in can No. 15 having died), which had been subjected to the 
longer day, and on 1 of the 16 plants in group 2. With the exception of 2 
weak, poor plants in can No. 3, which had been kept the longest of its group at 
a 10-hour day, and 1 plant which died budded in April (can No. 6), all the 
plants in group 3 opened their first blossoms between April 4 and April 27. 
(PI. 4, A.) 
Group 2, which had received only a 12£-hour light exposure as its longest day, 
showed a much less pronounced tendency to come into blossom tfyan did group 
3. The first plant opened its first flower March 31, and 3 plants came into 
blossom in May, and 1 in each of the months of June, July, and August. Of 
the remainder, 7 died between May and October, and none flowered. 
The light exposure of group 1 was changed on March 14 from 10 to 12 hours. 
Two plants came into blossom in July, and 2 in August, but the others failed 
to blossom. 
From the behavior of these plants under varying day lengths, it is concluded 
that a day length of 10 hours is too short, while one equaling that of a June 
day in this latitude, 13.2 hours, is too long, for flowering, and that an interme¬ 
diate day length is necessary to induce blossoming. While blossoming may occur 
under a suitable day length, whether the preceding days have been too long 
