Back to about 200 Ma ago, charting the motions of plates is relatively simple using the striped patterns of magnetic field strength above the ocean floor, which reflect periodic reversals of polarity of the geomagnetic field. Post-Triassic plate motions can also be assessed in an absolute reference frame with the use of hot spot tracks. Since no ocean floor is older than 200 Ma, the method cannot be used before then. Instead, the inclination and direction of remanent magnetism in continental rocks, suitably corrected for any tilting by deformation, take on the role of tracking motions. The direction is taken as being towards the magnetic poles at the time a rock formed, whereas the inclination supposedly varies in a simple fashion with latitude as it does today; vertical at the poles and horizontal at the ancient Equator. The post-Triassic break-up of Pangaea allows the palaeomagnetic method to be tested, and for that period it holds up extremely well. The models that chart how continental masses separated from a late-Precambrian supercontinent, drifted and then clanged together in the Devonian to early Permian to form Pangaea use the assumption of a consistently dipolar magnetic field that was lined up with the Earth’s axis of rotation: about as uniformitarian as one can get. They are models that delight tectonicians and students alike. There is however, a period in Earth’s history, from about 750 to 600 Ma, when palaeomagnetic positioning gives worrying results. Evidence of glaciation occurs at nearly equatorial palaeolatitudes at least three times.

Taken at face value, these results form the basis for the ‘Snowball Earth’ hypothesis, and the 750 to 600 Ma period has been dubbed the Cryogenian. But there are two other ways of explaining what is about as far from uniformitarian as can be. Maybe there were long periods when the geomagnetic field was neither dipolar nor lined-up with the rotational axis, in which case palaeolatitudes for those periods would be totally meaningless. The other possibility, which is alarmingly odd, is that before about 600 Ma the angle between the Earth’s axis of rotation and the plane in which it orbits the Sun was not about 23.5°, but more than 58°. At a high obliquity, Earth’s rotation would then ensure that high latitudes were warmer than low ones, which would neatly explain away much of the evidence for ‘Snowball Earth’ conditions. It is a worrying idea, simply because some considerable force, i.e. a stupendous impact, would be needed to change the axial tilt from >58° to what it is now and probably has been throughout the Phanerozoic. Settling the matter once and for all seems now to have been achieved by David Evans of Yale University, using a simple yet ingenious approach (Evans, D.A.D 2006. Proterozoic low orbital obliquity and axial-dipolar geomagnetic field from evaporite palaeolatitudes. Nature, v. 444, p. 51-55).

Evans based his study on the uniformitarian assumption that conditions are just right for strong evaporation of shallow, enclosed seas between 15 to 35° of latitude either side of the Equator, which is where evaporite deposits are forming now. If true, and if the geomagnetic field has been much the same as it is now, except during reversals, then all evaporites should give palaeolatitude results with this narrow range. There are lots of them, going back to 2.3 Ga ago, and being quite soft it is easy to drill cores from them. Furthermore they contain wind-blown dust, the magnetic component of which would line up nicely with the geomagnetic field while salts crystallised. The results from 54 world-wide sample are quite a triumph, for no evaporite palaeolatitudes are further than 40° from the Equator, and their means fall within the modern latitude range of an excess of evaporation over precipitation. There are differences between different time periods – before Pangaea existed evaporites formed slightly closer to the Equator than in later times. The fact that they cluster also shows that the dominant component of the geomagnetic field has been consistently been a dipole. However, even though the fundamental assumptions on which palaeomagnetic measurements are based seem sound, there are still problems for the Snowball hypothesis. Are the magnetic measurements up to scratch and do the stratigraphic and radiometric ages of samples refer to the evidence for glaciation?

Bad news for lunar base

Whether or not the Moon becomes once again a target for exploration by astronauts, and for use as a launch pad for Mars, depend on whether there is any water there. There has been considerable optimism that perpetual shadows in some of the deep craters close to the south lunar pole might contain ice that has not been exposed to solar heating. There is a way of telling using radar imaging, and reconnaissance results from orbiting probes had suggested that ice was indeed there, hence the excited men in suits of various kinds. A check using far more revealing radar data produced using the Areceibo radio telescope – it has also produced images of Venus at far greater distances – show that both sunlit and shadowed areas on the Moon can give a signal that is theoretically that from ice (Campbell, D.B. et al. 2006. No evidence for thick deposits of ice at the lunar south pole. Nature, v. 443, p. 835-837). Since ice could never survive in full sunlight, the similar results cast great doubt on ice being anywhere else on the Moon. There also seems to be a correlation in degree of belief with degree of involvement with future lunar exploration preparation.