Vines . — On Epinasty and Hyponasty. 435 
that an explanation which holds good in the one case is correct 
in the other. 
So far I have dealt exclusively with dorsiventral members 
which, when fully exposed to light of sufficient intensity, place 
their upper surface at right angles to the incident rays ; to 
members, that is, which are diaheliotropic. But it must not 
be overlooked that there are dorsiventral members which take 
up a different light-position, presenting their margin to the 
incident rays, and which have hence been described as para- 
heliotropic. This phenomenon is presented both by young 
growing leaves and by mature motile leaves. So far as I am 
aware, no observations have been made on growing leaves of 
this kind with the view of ascertaining whether they are epi- 
nastic or hyponastic, and I have not had time this season to 
make any experiments with them. But some inferences can 
be drawn as to the conditions of paraheliotropism in mature 
motile leaves, at least in those cases, which seem to be re- 
latively few, in which the paraheliotropic movement is not 
effected by means of torsion. For example, Darwin 1 has 
pointed out that the leaflets of Robinia fall at night, whereas 
when exposed to bright sunlight they rise above the hori- 
zontal. In accordance with what has been said above in 
explanation of nyctitropic movements, that of the leaflets of 
Robinia may be ascribed to epinasty. The assumption of the 
paraheliotropic position is, therefore, effected in opposition to 
the inherent epinasty of the leaflets. 
I11 Averrhoa bilimbi the leaflets fall downwards at night 
until they are vertical, and are therefore probably epinastic 
like those of Robinia. But they likewise sink on exposure to 
bright light into an obliquely descending position, so that the 
effect of the inherent epinasty of the leaflets is modified to only 
a slight extent. 
This kind of reaction to light cannot, I think, be satis- 
factorily explained otherwise than on the assumption of a 
special paraheliotropic irritability. 
There remains to be considered yet another class of move- 
1 Darwin: Movements of Plants, p. 445. 
Gg 
