Magnetostratigraphy is dependent upon variations in the magnetic properties of rocks as a basis for geological correlation. The most widely used property is the direction of the primary remanent magnetism, which records the geomagnetic field polarity at the time of formation of the rock. Sets of magnetic polarity reversals in sedimentary sequences can be correlated between sections. Furthermore, biostratigraphical information from the sequences often permits correlation of the observed polarity sequence with the appropriate part of the radiometrically-calibrated geomagnetic polarity time scale (GPTS). This allows the assignment of a numerical age to each of the beds containing a polarity reversal within the succession. Because polarity reversals effectively happen simultaneously over the whole surface of the Earth, they can be used for global correlations.

The pattern of geomagnetic polarity reversals that forms the basis of the GPTS for the Cenozoic and late Mesozoic is well established from ocean floor spreading magnetic anomalies, so that the GPTS for these periods is now well-defined (e.g. Berggren et al 1995). A problem in magnetic stratigraphy is the occurrence of magnetic overprints, acquired during later thermal or chemical events in the sediment/rock history. Often, these can be removed by incremental demagnetisation, to isolate the primary component. However, in some circumstances the overprint is more stable than the primary component, or it may completely replace the latter.

Throughout the Cenozoic, including the Holocene, the GPTS is age-calibrated by combined palaeomagnetic and radiometric age investigations of terrestrial lava sequences such as those on Iceland, and by Milankovitch cyclicity in sedimentary sequences.  A further source of information is biostratigraphical age determinations on sediments which immediately overlie specific ocean floor spreading magnetic anomalies in the ocean basins and which have been cored by deep-sea drilling.

The basic magnetostratigraphical unit is the magnetic zone or magnetozone, which usually is identified as a rock interval characterised by a specific (either normal or reverse) dominant magnetic polarity. The interval of time corresponding to a particular magnetic polarity zone is the magnetic chron, which has a typical duration of about 10⁵ to 10⁷ years. Subchrons are shorter intervals of opposing polarity within a chron and superchrons are longer intervals of dominantly normal, reverse or mixed polarity. Small scale perturbations in ocean floor magnetic polarity records which may represent very short geomagnetic polarity events are called cryptochrons.

The four most recent magnetic chrons, which extend from the Holocene to the late Miocene, were named after pioneering workers in geomagnetism. In younger to older order, these are the Brunhes normal, Matuyama reverse, Gauss normal and Gilbert reverse polarity chrons. Each of these contains distinctive subchrons, such as the Olduvai subchron which is located within the Matuyama chron, during the Early Pleistocene Subepoch. These and earlier chrons are now conventionally labelled from the corresponding ocean-floor spreading magnetic anomaly number. The number is usually suffixed by the letter n or r, according to whether the dominant magnetic polarity is normal or reverse. It is prefixed by the letter C (Cenozoic) to discriminate the time interval (e.g. chron C21n) from the corresponding magnetic anomaly number (e.g. anomaly 21).

A magnetic zone may sometimes be defined in terms of magnetic susceptibility or some other distinctive magnetic property of the rock, instead of the magnetic polarity. Magnetic susceptibility is essentially a mineralogically-controlled parameter, that reflects the composition, concentration and grain size of magnetic minerals within the rock. Fluctuations in susceptibility may reflect climatic, tectonic or other controls on sedimentation, that can be correlated locally, regionally or globally. For example, susceptibility fluctuations in loess sequences and deep marine successions have been correlated with oxygen and carbon isotope stages, reflecting global climatic changes.

The magnetostratigraphical record is being applied increasingly to help resolve correlation problems between different environments, particularly between terrestrial and deep-sea sequences.

*This guide is based on that produced by Rawson et al. (2002) for the Stratigraphy Commission of the Geological Society of London.