548 .Mr. Knight's Experiments on 
descent from the leaves, had been carried down by the wood, 
through my incisions, in the preceding experiments on the crab- 
tree; because I observed a very small increase in size, in the 
lower part of the stocks ; which, I think, could not have taken 
place without some matter derived from the leaves. But subse- 
quent observation induces me to believe, that the small quantity 
of additional matter found in the lower part of the stock came 
from a different source. In those experiments, I paid little at- 
tention to any small shoots which sprang from the trunk at 
some distance below the incisions ; and the buds which usually 
began to vegetate about Midsummer, were not always rubbed 
off, till some minute leaves appeared. Through these, I now be- 
lieve that a small quantity of sap was thrown into the bark, and 
carried up through its tubes, by capillary attraction, when the 
current from above was intercepted. For the increase of size 
in the stock always diminished, as it ascended towards the in- 
cision ; which, I think, would not have been the case, had it 
been produced by nourishment descending from the upper parts 
of the tree. 
Nothing has occurred, in the preceding experiments, to throw 
much light on the office of the medulla, to which Linnzeus and 
subsequent writers have annexed so much importance; but I 
will now endeavour to point out one of its offices. In the young 
and succulent shoot, this substance is extremely full of moisture; 
and, as there is an immediate communication between it and 
the leaf, through the central tubes, I conclude it forms a reser- 
voir, to supply the leaf with moisture, whenever an excess of 
perspiration puts that in a state to require it. Some reservoir 
of this kind appears to me to be necessary to plants ; for their 
