Mar. 17,1923 
Further Studies in Photoperiodism 
911 
flowering in every case. The vegetative period of the Biloxi variety, for 
example, was reduced from 94 days to 21 days. In this experiment the 
reduction in the duration of the daily light period was no greater than 
actually occurs during the growing period of the Biloxi. The light in¬ 
tensity was reduced by shading to about 3,500 foot-candles, while the 
natural decrease up to the close of the growing season (October 1) only 
is to an average of about 6,200 foot-candles for the same hours of the 
day (9 a. m. to 3 p. m.), and to about 4,500 foot-candles for the entire 
io-hour day (7 a. m. to 5 p. m.). It is to be noted, also, that the reduced 
intensity shows no selective action on the different varieties with respect 
to time of flowering, whereas this is the characteristic feature of the 
varietal response to change in light duration. The varietal responses 
to the different light periods, in fact, are strictly in accord with the natural 
behavior of the varieties as to earliness of maturity. There are signifi¬ 
cant differences, also, in the effects of change in intensity and change in 
duration of the light on the heights attained by soybeans. In the former 
paper (7) it was shown that under the natural day length of summer, 
reduction of light intensity causes increased elongation of the stem in 
soybeans and that, in fact, with heavy shading a typical viny form o£ 
growth develops. In Table VI it is shown, however, that under a 10- 
hour day heavy shading failed to cause increased stem elongation in 
comparison with plants exposed to the full sunlight for 10 hours daily, 
except in the case of the Otootan variety. The effect of shortening the 
duration of the light period is the reverse of that produced by reducing 
the intensity; that is, there is decided reduction in stem growth, as is 
shown in Table VI. Here, again, the effect of a shortened light period 
on the stature attained by the plant is in accord with the progressive 
decrease in stature which results when the date of planting is made 
later and later in the season, an effect well shown in the former paper 
(7, PL 78, B). Decrease in light intensity fails to account for either the 
initiation of flowering or the slowing down in stem growth in soybeans 
which occur with advance of the season after the summer solstice. 
A similar experiment was made with the Silverskin onion, using the 
same shade that was employed in the test with soybeans. With the 
light intensity thus reduced to one-third the normal, plantings of the 
onion made June 16 grew and developed bulbs in substantially the same 
maimer as in the open sunlight. At most, the dying back of the tops 
was delayed a week and the bulbs were about the same size as those of 
the controls. Under a io-hour exposure to full sunlight, on the other 
hand, there was no bulbing and the tops remained green throughout the 
summer. These differences in behavior are well shown in Plate 18, A. 
There is no evidence, therefore, that bulb formation in the onion is due 
to excessive intensity of sunlight in the summer. 
In the light of these experiments it would seem that use of such terms 
as “quantity of solar radiation'* in relation to specific plant responses 
should be abandoned where such terms involve both intensity of the 
radiation and duration of the daily exposure. Not only is there fre¬ 
quently no approach toward equal effects for equal changes in these two 
factors but these effects may actually be in opposite directions. The 
futility of attempting to apply quantitative energy relations of this sort 
to the process of flowering in the soybean plant is further emphasized in 
the following data. Beginning June 30, one lot of Peking seedlings was 
placed in darkness from 10 a. m. to 4 p. m. daily; a second lot was simi- 
