181 
is weaker and the illumination lasts longer. Because the normal 
curvature has already shifted downwards a little way before the 
antiphototropic one becomes visible, the plants here assume the well- 
known S-shape. This case is illustrated in fig. 10 at an illumination 
of five minutes with 12.5 MC. 
With a light of still less intensity no more antiphototropic 
curvatures appear. We do find here after some time with the light- 
quantities that otherwise would have evoked an antiphototropic 
curvature, a considerable retardation in the velocity with which the 
curvature increases. When after the illumination these plants are not 
placed on the clinostat, they show the geotropic counter-curvature 
sooner than plants do that have been lighted for a longer or for a 
shorter period, and with which consequently this retardation does 
not oecur. In fig. 9, which shows the effects of an illumination with 
2'/, M.C. this curvature is seen at an exposition time of 10 minutes. 
This proves that, in order to effect an antitropic curvature it is 
in the first place necessary to destroy the sensibility in the anterior 
as well as in the posterior side entirely or nearly so, a process 
dependent on the supplied energy; but in the second place to enable 
the sensibility in the anterior side to gain an advantage over the 
sensibility in the posterior side, a process that can take place only 
if the sensibility in the anterior side has vanished sooner than in 
the posterior side. If, therefore, the antiphototropic curvatures appear 
only in the manner described, they cannot appear with illuminations 
of very short duration. We shall presently inquire whether this 
conception is correct (see below). 
That time is indeed an important factor in the process here 
discussed, is proved indubitably when we compare the effect of a 
unilateral illumination which is preceded by an omnilateral one 
with the effect of the same unilateral illumination, when it is followed 
by the omnilateral one. If it were only a question of the light- 
quantities in the antagonistic halves, the result would be the same 
in either case. This, however, is not so: whereas in the first case 
antitropic curvatures are seldom found and are never conspicuous, 
they appear regularly in the second case and are then generally 
well developed. 
The second case is easy of explanation with my theory: As the 
sensibility at the anterior side is completely, or almost completely 
destroyed by the unilateral illumination, there will be a gradual 
increase of it during the omnilateral after-illumination ; at the posterior 
side, on the contrary, the sensibility will still be more or less 
considerable at the end of the unilateral illumination, and it will, ° 
