182 
therefore, still decrease during the omnilateral illumination. It follows, 
then, that it will reach in this side its lowest value only after some 
time has elapsed, and meanwhile it will have reached at the anterior 
side a higher value, and, as the advantage will remain with this 
side till the original sensibility in both sides will have been restored, 
it is to be expected that an antitropic curvature of some importance 
will arise. If a unilateral illumination is preceded by an omnilateral 
one (apart from those cases in which the omnilateral fore-illumination 
is very weak) the occurrence of an antiphototropic curvature would, 
on the contrary not be in keeping with my theory. I was, therefore, 
originally inclined to call in question the reports concerning the 
occurrence of antitropic curvatures under these circumstances, the 
more so as these reports are extremely doubtful: under almost the 
same conditions under which in one case curvatures were found, 
they could not be discovered in the other. In view of the results, 
to be recorded lower down, the possibility of their accuracy may, 
however, be admitted, at least in a certain number of cases. 
If, besides the cause discussed here, there should be still another 
to call forth antiphototropic curvatures, it must manifest itself, as 
stated above, if the time-factor is eliminated as much as possible, 
i.e. if the duration of the illumination is shortened as much as 
possible. The advantage in the anterior side will then gradually 
decrease, and consequently the antiphototropic curvature also. To settle 
this question | made some experiments, which need not be described 
in detail, as the way in which they were performed did not differ 
from that used in my earlier experiments. Suffice it to say here that 
in some experiments the quantity of light amounted to 9000 M.C.S. 
and the period of illumination to */,, 3, 12, 48, and 192 seconds, 
while in other experiments an amount of 12000 M.C.S. was used, 
and a period of illumination of 1, 4, 16, 64, and 256 seconds. After 
the illumination was concluded the plants were put on the clinostat 
on which they remained for 3 or 4 hours. Subsequently the curva- 
tures were compared. It now appeared that every where antiphototropic 
curvatures came forth and that their magnitude differed only very 
little. It would seem indeed, as if the curvatures occurring with 
longer periods of illumination were generally slightly stronger, but 
the differences were so insignificant that they did not afford any 
certainty in this respect. 
The experiments described go to show that there must be indeed 
still another factor by which antiphototropic curvatures can be 
brought about. Since the factor of time had been eliminated as much 
as possible, we must have to do here with a direct result of the 
