460 
Joumal of Agricultural Research 
Vol. XXVIII, No. 5 
The pinnae midway of the leaves measured averaged in length for daily exposures 
of 10, 12^£, and 13J^ hours, 1.9, 2.7, and 3.5 inches, respectively, while the controls 
exposed to a normal short day, averaged in length from 1.8 to 2.2 inches. 
Similarly, but in a less pronounced degree, the terminal pinna was affected, the 
average lengths for the 10, 12J^, and 13hour exposures being 2.1, 2.5, and 2.8 
inches, respectively, and for their controls ranging from 2 to 2.3 inches. 
These measurements show very definitely that a day length of 12J^ or 133^ 
hours is much more favorable to leaf development than a day length of 10 or 11 
hours, the longer day causing the development of a much larger leaf, longer in 
both midrib and pinnae. These measurements on plants grown under an arti¬ 
ficially shortened and an artificially lengthened day explain the pronounced 
difference in appearance between the plants in the field in late fall and winter, 
with their slow growth, short internodes, and small leaves, and the plants in late 
spring, with their rapid growth, long internodes, and large leaves. This contrast 
is shown in Plate 2, B, on a branch obtained from row No. 16, planted July 19, 
1922, and photographed May 9, 1923. 
SUMMARY 
The blossoming of Tephrosia Candida was inhibited both by a day length 
shortened to 10 hours, and by a day length of 13.2 hours (length* of the longest 
summer days in Mayaguez, P. R.). 
Shortening the day from the long day of summer to a 12-hour length promptly 
induced blossoming, whereas a continuation of exposure to the long summer days 
inhibited it. 
Under a day length artificially protracted to 13J^ hours for 10 weeks, buds 
failed to appear, but on shortening the day length they promptly appeared. 
Under exposure to a maximum day length of 12J^ hours, later reduced to 12 
hours, the plants displayed much less tendency to come into blossom than was 
the case when reduction was made from a longer day. 
There was still less tendency to blossom when the day, instead of being 
shortened, was lengthened from 10 to 12 hours. 
While blossoming may occur under a 12-hour day, whether the preceding days 
have been too short or too long for blossoming, the impulse to blossom apparently 
is much more pronounced when the preceding days have been too long rather than 
too short, days of too great a length followed by shorter days quickly inducing 
the reproductive stage. 
Although autumn is the normal blossoming season for T. Candida , heavy 
blossoming was induced in April through artificial manipulation of the length of 
light exposure or day length. 
Day length not only determines the blossoming season for T. Candida , but it 
also affects its growth in a pronounced and decided manner, the longer days 
producing growth with longer internodes and larger leaves. 
