木の長い記憶

The long memory of wood

A tree has no diary, but it keeps something very like a ledger. In temperate regions, where growth stops each winter and resumes each spring, a tree adds one layer of wood a year, and the layers show in a cut trunk as concentric rings: a paler band of fast spring growth followed by a darker, denser band laid down later in the season. Counting the pairs gives an age. That much is familiar to most schoolchildren. What is less familiar is how much more the rings can say.

The thickness of a ring depends on the conditions in which it grew. A year of generous rain and moderate warmth tends to produce a broad ring; a drought, a late frost or an attack by insects tends to produce a narrow one. A trunk is therefore a long sequence of wide and narrow bands, and the sequence is not random. Trees growing in the same region respond to the same weather, so their ring patterns resemble one another, much as the handwriting of one family might.

It was this resemblance that gave rise to the science of dendrochronology. Early in the twentieth century an astronomer working in the dry south-west of the United States was searching for a link between sunspots and the earth's climate, and he turned to trees as a convenient record. The link he sought proved elusive, but the method he developed did not. By sliding the pattern from a young tree against the pattern from an older one, he found stretches where the two matched closely, and so could be joined together. The technique is called cross-dating, and it is the heart of the discipline.

Cross-dating allows a chronology to be extended far beyond the life of any single tree. A living oak provides the recent centuries; a beam from a medieval barn overlaps with it and carries the sequence back further; a trunk preserved in a peat bog or a river bed overlaps with the beam, and so on. Painstakingly assembled in this way, continuous sequences from European oak and pine now stretch back more than ten thousand years, with each ring assigned to a specific calendar year. In the mountains of California, meanwhile, certain bristlecone pines are themselves more than four thousand years old, which makes them among the oldest individual living organisms known.

The uses are surprisingly varied. An archaeologist who finds a timber in an ancient building can sometimes say not merely when the tree was felled but, if the outermost ring survives, in which season. A museum can test whether the panel of an old painting came from a tree that was growing when the picture is claimed to have been made. And because rings can be dated with such precision, they are used to check and correct the radiocarbon method, which would otherwise drift out of step with the calendar.

None of this means that a tree ring is a thermometer, and researchers are careful to say so. Width is influenced by temperature, moisture, soil, competition from neighbouring trees and the age of the tree itself, and separating these influences takes statistical care. Some species, particularly in the tropics where there is no cold season, form no clear annual rings at all. Even in favourable regions, a stressed tree may skip a ring in a very poor year or add a false one after a mid-summer pause, and an unwary reader may be misled by either.

Yet the remarkable fact remains that a piece of wood, left to itself, preserves a patient and largely honest record of the years through which it lived. It cannot say what any year meant to the people who endured it. It can only say whether the rain came. But to a historian trying to understand why a settlement was abandoned, or why harvests failed across a region in the same decade, that is no small thing.