Mar. i7,1923 
Further Studies in Photoperiodism 
917 
gain and loss of water. This implies, perhaps, control of permeability. 
Proper distinction must be made, therefore, between regulation of the 
external supply of water and control of the internal mechanism involved 
in dealing with phenomena of hydration. It seems probable that the 
light period, acting through control of the internal mechanism, is capable 
of exercising a far more refined regulation of the internal water supply 
of the plant's parts than is ordinarily accomplished by change in the 
external supply of water. The experiments with Sagittaria recorded on 
page 878 indicate that this is true. Growing this plant as an aquatic 
at most delays flowering by some 10 days as against growth under 
mesophytic conditions, while appropriate change in the light period 
resulted in sharp response. As already pointed out, growing soybeans 
under semixerophytic conditions in contrast with mesophytic conditions 
was entirely without effect in initiating flowering. It may occasionally 
happen, perhaps, that the available external supply of water will be 
just right to favor a definite response such as successful flowering. It 
seems likely also that plants which are not particularly sensitive to 
change in the light period, including typical everbloomers, such as the 
cotton plant (Gossypium spp.), may be more subject to disturbance of 
the normal flowering and fruiting processes (shedding of blossoms or 
young fruits, etc.) by such factors as change in the water supply or the 
ratios between the different elements of plant food in the soil. In 
addition to its apparent regulatory action on the internal mechanism 
controlling hydration, it is to be expected that the duration of the light 
period will tend to modify the external evaporative forces involved in 
transpiration. It seems possible that this may be a factor of importance 
in the action of the long days of summer in checking the growth of 
certain plants, although no definite evidence on the subject is available. 
Regardless of its actual causative significance, there seems to be no 
doubt that hydration is subject to definite regulation by length of day 
and, hence, change in the degree of hydration is definitely correlated 
with change in form of expression in the plant. Apparently the optimal 
light period for apogeotropic growth also is optimal for maximum hydra¬ 
tion. Progressive loss of power to fully utilize carbohydrate and other 
nutrient material in accomplishing increase in stature seems to be corre¬ 
lated with progressive loss of capacity to maintain the transpiration 
stream. Development of sexually produced as well as vegetative repro¬ 
ductive structures, tuberization, and the general phenomena of dormancy 
seem to follow as natural sequels. 
CONCLUSION 
In this paper considerable data are presented which tend to demonstrate 
the importance of the seasonal range in length of day as a factor in 
initiating and regulating sexual reproduction in plants, thus confirming 
the results of investigations reported in an earlier article. Experiments 
are described, also, which bear on certain quantitative aspects of this 
response. Results of fairly extensive investigations on the relation 
of length of day to the formation of tubers and bulbs are presented 
in some detail. Observations, largely of a preliminary nature, are 
reported on length of day as a factor in various other responses of the 
plant, including character and extent of branching, root growth, forma¬ 
tion of pigment, abscission and leaf fall, dormancy, and rejuvenescence. 
Apogeotropic growth or increase in stature as affected by the daily light 
