Archaeomagnetic dating relies on the measuring the orientation of iron particles in burnt deposits towards the
magnetic pole. The pole moves around, but magnetised deposits stay fixed on its position at the time of burning.
We can measure the difference between their orientation and the present position of the pole, which can give us the
date of the burning episode.
When material such as clay or earth is heated to above 650 degrees Celsius (the Curie Point), such as in a hearth
or kiln, the existing magnetism of iron particles in the soil is wiped clean and they are re-magnetised. Magnetic
particles are always oriented towards the magnetic north pole, and this is fixed at the time of burning. When
structures are repeatedly burnt, we can sometimes measure the date of separate burning episodes by sampling
different fired layers.
Small samples of soil or burnt building materials are prepared in situ by having small plastic discs glued to the
surface of the layer. The discs are marked with a line which points towards the present position of the magnetic
pole - which is measured with a highly accurate compass. Then the discs and the small blocks of soil attached
beneath them are carefully removed. In the lab afterwards, the difference in orientation between the line showing
the present magnetic pole and the orientation of the magnetised particles in the soil, reflecting the pole's
position at the time of burning, can be determined.
These differences in magnetic orientation, which can give us an accurate date, are compared to a known curve of the
movement of the earth's magnetic pole. This is not always a straightforward process - the curve sometimes doubles
back on itself in a wave pattern reflecting the back and forth movements of the pole over time - therefore a sample
can apparently give two or more separate dates, sometimes centuries apart. The archaeologist would have to decide
which one is the most likely based on other indicators such as finds or radiocarbon dates. The curve has been
well-documented for Britain, but in many other areas of the world there is a lot less certainty in using this
Return to Dating Methods.
- C. Renfrew and P. Bahn, Archaeology: theories methods and practice, Thames and Hudson, London 2000, p. 158-9.