CONCERNING THE AGE OF THINGS 135
should exhibit irregular differences from year to year.
Furthermore, it is assumed that all the trees over large areas
of the earth's surface will be affected in like manner by the
climatic variations from year to year, and this has been found
to be substantially true for an area as large as England. Of
course, an oak growing hardily in Scotland may make smaller
absolute increase of trunk diameter than another living in
the lush Weald of Kent, but it is maintained that if, for
example, the Scot should grow '2 millimetres in 1960 and
3.5 millimetres in 1961, then its Kentish fellow, having
grown, say, 4 millimetres in 1960, will put on about
7 millimetres in 1961, and so on, in the same proportion.
Local climatic variations, and particularly the recent
large-scale interferences by Man such as fen-drainage
schemes, will undoubtedly invalidate this correspondence in
special cases, and timber grown in wet land is so irregular for
other than climatic reasons that no reliance for dating can be
placed upon it. Nevertheless, the pragmatic test of use has
shown that this method of dating can be relied upon, if used
with discretion and an understanding of its limitations.
The technique is as follows : Imagine a succession of two
good years, followed by three lean years, then another two
good ones, then one of the exceptionally good growing-years
near the peak of an eleven-year period : in this year a tree
is felled. Clearly, if we subsequently measure the distances
between successive growth-rings of a piece of timber from
this tree, we shall find some such series as 5.0, 4.8, 3.1, 3.0,
3.3, 5.1, 5.3, 7.0 millimetres (ending at the outermost ring
under the bark). Let us plot this succession of numbers as
a graph, at equal intervals along the x-axis, and we shall
have a curve which might be said to be the tree-growth or
climate characteristic of the period of years concerned—as
individual a stamp of that period as the unique finger-prints
of a person. Now, as Mr. A. W. G. Lowther did in
1943-1950, let us in imagination obtain pieces of well-grown
timber of known date, preferably pieces showing at least
one "waney" edge (which ensures that the outermost layer of
wood will be in evidence). With great labour, it should be pos-
sible to combine the ring-interval records of these individual
timbers, and so produce an unbroken record of climate, or
at least of tree-growth, as far back as reliably-dated timber