9i 
certain Tropical Trees. 
stem-members shorten, and consequently the leaves are closer 
together ; the youngest leaves shade themselves under the 
older ; the young leaves stand generally approximately 
upright in the direction of the axis bearing them, and thus 
make but small angles with the most powerful sun’s rays.’ 
As far as I am aware, similar divergences called forth by 
differences in insolation have not been described in trees ; 
yet Browne a grandiceps shows an agreement with those 
divergences too obvious to be accidental. 
In the shade-growing tree the leaves and leaflets soon 
become horizontally disposed ; the main branches make wide 
angles with the main trunk, which divides within a foot or 
so of the ground into several branches. These main branches, 
spreading horizontally, give rise to lateral members, which 
also make but small upward angles— -some running out away 
from the centre of the tree, others toward it. Hence the 
tree assumes a sort of ‘ umbrella ’ outline. The tree growing 
in the sun is, on the contrary, pyramidal (applying the word 
in the sense in which it is used in describing the shapes of 
such trees as the Poplar) ; its main branches rise up steeply, 
and its leaves have a much more marked vertically down- 
ward tendency. The branches of the shade-growing tree are 
longer than those of a tree growing exposed to the sun ; 
and consequently the former has a more spreading habit. 
When it is remembered that Brownea grandiceps is a tree, 
albeit a small one, the apparent paradox propounds itself — 
a tree, by its nature a seeker after light, loves the shade. 
Does it, then, seem unreasonable to suppose that such 
plants, whose constitution is so unadapted to withstand great 
light, should not only have the means, by virtue of the power 
of movement of their leaves, to protect, if need be, their old 
blades from an excessive sun, but also have special means 
of protection for young leaves. 
Since the course of the foregoing argument is at times 
interrupted by descriptions of experiments, I wish here to 
briefly sum up- — to point out how far previous views are 
supported by these experiments, how far they seem to me 
