A branch of stratigraphy that defines rock units as having different magnetic properties than those adjacent to them in the stratigraphic succession. The main property used is the polarity of the remnant magnetism in the rock, compared to the present magnetic field.

History and the Basics:
The discovery of the pattern of magnetic stripes on the sea floor marked the beginning of this branch of geology, and played a major role in plate tectonic theory. The stripes were confusing at first, but then the symmetrical pattern was noted, and when the pattern was compared to the ocean topography, it was discovered that the ridges served as the mirror plane. In other words, the mid-oceanic ridges were the source of new material on the ocean floor, and the magnetism at the time of cooling was preserved in this material.

If the magnetism preserved is in the same direction as today's magnetism then it is termed normal polarity, but if it is the opposite it is reverse polarity.

Magnetostratigraphy and Oceanic Crust:
Oceanic crust is composed of basalt, an igneous rock that contains magnetite and other minerals that can record the magnetic field present at the time of their crystallization. Above a specific temperature, the Curie point, the magnetic dipoles are free to move, so only heating above this temperature can change the magnetic record in the rocks. This makes the magnetism of oceanic crust fairly easy to study. However, this only provides information about the earth's magnetism to about 165 million years ago, because no oceanic crust older than this exists, due to subduction.

Magnetostratigraphy and Continental Crust:
Minerals in continental crust can also be used in magnetostratigraphy, but is more difficult since much of the accessible rock is not igneous. Metamorphic rocks can contain a magnetic record, but they can be misleading if, for example, one magnetic mineral has been heated above its Curie point and another contained in the same rock has not. They may also be misleading as they have often moved from their orientation during the cooling through the Curie point. So even if a good remnant magnetism exists, it may not tell the researcher anything since it has been moved. However, if the rock can be otherwise age-dated, the age can be compared to the magnetic polarity of the time based on oceanic crust to find whether magnetism was normal or reverse, and from this the relative movement of the rock can be determined.

Igneous rocks found on continental crust generally contain magnetic minerals like those on the ocean floor, but may have a different composition depending on their origin. Therefore it may be simple or more involved to measure the magnetism of these rocks depending on how strong their remnant magnetism is.

Sedimentary rocks often contain very few minerals that can record magnetism and require more sensitive methods. Sampling of sedimentary rocks is usually carried out by drilling a 2cm by 100cm core out of the rock, noting the three dimensional orientation of the core and the bedding. Six or more cores must be taken from each bed for statistical purposes. The remnant magnetism of the samples is determined in a laboratory by a magnetometer, which can often detect magnetism several orders weaker the earth's magnetic field.

This method is used to date and correlate sedimentary rocks only when other methods (such as those relying on lithologic properties, fossil content or absolute age dating) do not apply, since it is time consuming and expensive. There must be at least one point in the rock succession that can be dated absolutely, so that it can be compared to the magnetic reversals on published charts. Once the reversal history of the sedimentary succession has been established it can be compared to the chart to give ages and basic sedimentation rates. However, great care must be taken, as some reversal intervals may not be represented if sedimentation stopped for a period in time.

Reference: Nichols, G., 1999. Sedimentology and Stratigraphy. Blackwell Science Ltd., London.