of Timler in different Directions. - 439 
which was kept open by a wedge, was extended.to the medulla, 
The wedge was then withdrawn, and the opposite sides of the di- 
vision instantly came in contact with great force. A second in- 
cision, similar to the preceding, was then made to commence at 
the bark, about an inch distant from the preceding, and to ter- 
minate, like that, at the medulla; by which means, a wedge of 
wood, an inch sauare at the bark, and ending in an edge at the. 
medulla, and ten inches in length, was wholly detached. This, 
nevertheless, did not quit its position, being retained in it by the 
expansion of the wooed from which it had been separated, 
The opposite sides of the same transverse sections .of wood 
were divided hy the saw in a direction diametrically opposite to 
that above mentioned ; under which circumstances, the expan- 
sion of the convergent cellular processes could not, as in the pre- 
ceding cases, occasion any pressure upon the sides of the saw, 
which consequently continued to move with perfect freedom. 
These circumstances led me to infer, that the medullary canal 
must be subject to considerable variations of diameter, with the 
increase or diminution of the quantity of moisture in the wood ; 
and I conceived that I should easily be able to ascertain the 
truth or falsehood of this conjecture by the following means. I 
selected, in winter, some parts of the stems of young trees as soon 
as they were felled, which J retained in such a situation as might 
occasion them to lose a considerable part of the water they ccn- 
tained, though not to such an extent as to destroy, or endanger, 
life. The medulla of these was then removed ; and the space it 
had occupied was filled with cylindrical pieces of metal, which. 
were so large that they could not be introduced without consi- 
derable force. ‘The pieces of wood were then deposited in a, 
damp soil, from which they absorbed much moisture ; and at the 
distance of ten d&#s J found the medullary canal so much en- 
larged, that the pieces of metal dropped through without any 
pressure being applied. ; 
I am prepared to prove, in a future communication, that the 
quantity of moisture in the alburnum is subject to great varia- 
tions in the living tree, and therefore I conclude that the'medul- 
lary canal frequently changes the extent of its diameter. 
It appears probable that, by means of this kind of expansion, 
the internal parts of timber trees so frequently: become rifted or 
cleft. Winds have Leen assumed by some, and frost by others, 
as the cause of these injuries. But winds cannot possibly be the 
cause, as pollared oak trees, upon which these can exert but 
very little power, are almost always rifted ; and the frost of this 
climate is rarely, or never, sufficiently intense to congeal the 
winter sap of trees. ‘This agent must also, J conceive, act sud- 
denly, if it act at all, and the trunks of large oaks cannot aA 
hed 
