36 
THE FOOD OF TREES IS IMBIBED [Part II. 
Rain and con¬ 
densation often 
shed inwards 
or towards the 
stem. 
Where branches form elbows, pointing down¬ 
ward, the dropping will sometimes almost amount 
to a stream, and will form a conspicuous mark 
on the road. Where boughs are horizontal, each 
drop falls perpendicularly, as soon as gravity 
overcomes cohesion; that is, as soon as the 
force of its weight is greater than its power of 
sticking to the bough. But not a drop of rain 
falls under trees in its natural form and size. 
In many cases accumulation goes on till the 
weight of water can bend the leaf to discharge 
its contents. 
Leaves are not fitted on to each other like 
tiles or slates on a roof; and it is impossible that 
the watershed of trees should be outward or 
from the stem, because the continuity of the 
outward channel is interrupted at the outward 
end of every individual leaf. But it is by no 
means impossible that the water-shed should be 
inward or towards the stem. And this, the 
very reverse of Koget’s fact, sometimes is the 
fact. Where leaves incline upward from the 
twigs, the twigs upward from the branches, and 
the branches upward from the stem, in rainy 
or condensing weather almost every drop of 
water is shed towards and down the stem; and 
the stem of a tree stands the model of a river, 
