BrightKidz Hub

An Old Beam Can Be Dated to a Single Year

About 15 minutes

Somewhere in a museum there is a blackened piece of oak taken out of the roof of an old house. The little card beside it does not say roughly eight hundred years old. It says the tree was cut down in the winter of 1236.

That is an odd thing to know about a lump of wood. Nobody was standing there writing it down.

One ring is one year

In places with a real winter, a tree cannot grow all year round. In spring it grows fast and the new wood is pale and open. By late summer it slows, and the wood it lays down is darker and tighter. Then it stops altogether until the following spring.

Starting again after a full stop leaves a visible edge. Cut the trunk across and those edges are the rings. Each one is one growing season, so each one is one year.

So a beam with 180 rings came from a tree that grew for about 180 years. That much is easy.

Now try counting

Here is where the obvious idea falls over.

To count rings back to a date, you have to know where to start counting from. With a tree still standing in a field that is no trouble at all: the ring just under the bark is this year, the one inside it is last year, and you can walk backwards as far as the trunk goes.

The beam in the museum is not standing in a field. It was cut down a long time ago, and when is exactly the thing you are trying to find out. Counting inwards from its outer ring only tells you how many years before an unknown year something happened.

It is a ruler with the numbers rubbed off. You can measure the length of a thing with it. You cannot say where it sits.

What the rings are really recording

Rings are not all the same width, and the differences are not random.

A year with a kind spring and enough rain leaves a wide ring. A cold year, or a dry one, or a year the tree spent recovering from something, leaves a narrow one. The ring is a record of how good that year was for growing.

And now the useful part. Every oak in the same valley lived through the same weather. A hard summer was hard for all of them. So they all recorded the same run of good years and bad years, in the same order.

A count is common. A pattern is not

"180 rings" is a single number, and thousands of trees share it. It cannot pick anything out.

But narrow, narrow, very wide, narrow, wide, wide, very narrow, wide is not a number. It is a shape. And a long enough shape usually fits the record in one place only, the way a key fits one lock and rattles uselessly in all the others.

A cut log end, and two ring patterns waiting to be matched On the left, the sawn end of a log: three rings drawn one inside another, not quite sharing a centre, with a small solid dot at the heart and a single crack running out from it to the lower edge. On the right, two rows of upright bars. The lower row is long, ten bars of clearly different heights standing in a line. The upper row is short, only five bars, and its pattern of heights is the same run of tall and short that appears part-way along the lower row. The short row sits above the left-hand end of the long one, so the two patterns do not line up yet. Sliding the short row along until its tall bars sit over the long row's tall bars is the whole method.
The long row is a pattern of years already dated, built up from wood we know the age of. The short row is the mystery beam. Nothing is being counted here. The two patterns are being slid past one another until they agree.
A cut log end, and two ring patterns waiting to be matched On the left, the sawn end of a log: three rings drawn one inside another, not quite sharing a centre, with a small solid dot at the heart and a single crack running out from it to the lower edge. On the right, two rows of upright bars. The lower row is long, ten bars of clearly different heights standing in a line. The upper row is short, only five bars, and its pattern of heights is the same run of tall and short that appears part-way along the lower row. The short row sits above the left-hand end of the long one, so the two patterns do not line up yet. Sliding the short row along until its tall bars sit over the long row's tall bars is the whole method.

The short row of bars is the beam. The long row underneath is a stretch of years already worked out. Tall means a wide ring and a good year; short means a narrow ring and a hard one.

Slide the beam along, one ring at a time, until its run of tall and short bars sits over the same run in the row below.

There is only one position where every bar agrees, and at every other position at least one bar is plainly wrong. That single fit is the whole method. Once the beam sits in the right place, you are no longer reading the beam — you are reading the calendar underneath it.

Reaching back further than any living tree

The oldest oak in a wood might be three hundred years old. That gets you back three hundred years and no further, which is not much use for a Roman fort.

So the record is built in overlapping links.

Start with a living tree and you have three hundred years, dated exactly, because its outermost ring is this year. Now find an old barn built long ago from timber cut while that tree was young. The barn's outer rings will match the living tree's inner rings, and where they match, the barn's date is fixed. The barn's own inner rings now reach back further than the living tree ever did.

Then find something older still that overlaps the barn. Then something older than that.

Each link is checked by pattern-matching, not assumed. Chains built this way now stretch back thousands of years, and every year in them was proved by a fit.

Why can a long pattern of ring widths date a beam when the number of rings cannot?

What it cannot do

This is a method with edges, and they are worth knowing.

It needs a real winter. A tree in a rainforest, where every month is much like the last, often makes no clear annual rings at all, so there is nothing to count and no pattern to match.

It needs enough rings. A short piece with forty of them makes a short pattern, and a short pattern fits in too many places to prove anything.

It needs the right neighbourhood. Wood from a different region grew under different weather, so its pattern will not match your record even if the date is right.

And it needs the outside of the tree. Carpenters trim the soft outer wood off a beam before using it. If those rings are gone, you can no longer say the year it was cut — only that it cannot have been cut before the last ring you still have.

That last one is not a failure. It is a different, weaker kind of answer, and it turns up again and again in this work. A dropped coin gives exactly the same shape of answer, for exactly the same reason.