This article is from
Creation 13(3):46–50, June 1991

Browse our latest digital issue Subscribe

‘Fossil’ magnetism reveals rapid reversals of the earth’s magnetic field


Almost everyone is familiar with a compass needle. The magnetic properties of the metallic iron in the needle ‘force’ the needle to swing on its pivot until it lines up with the north-south direction of the earth’s magnetic field. Indeed, every molecule of the metallic iron in the needle has these magnetic properties.

‘Fossil’ magnetism

Iron also occurs in many types of rocks, not usually in its metallic form, but as an oxide mineral called magnetite, which as the name suggests is magnetic. Just as all the molecules in the compass needle align themselves along the earth’s magnetic field, so do the molecules in the magnetite grains in rocks.

This will happen at the time a magnetite particle in a sediment or volcanic ash comes to rest, or in a lava (hot volcanic rock) as it cools to 500°C. Once the sediment layer is deposited and buried, or the lava flow has cooled below 500°C, the direction of the earth’s magnetic field as recorded by magnetite grains in these rocks cannot usually be changed by subsequent geological events (except for metamorphism—the process of changes to rock under the influence of elevated pressures and temperatures), even if the direction of the earth’s magnetic field has subsequently changed. This magnetism in the rocks is thus in essence ‘fossilised’, and so is usually called palaeomagnetism.

Magnetic Reversals

The existence of this palaeomagnetism in the rocks has claimed a lot of attention since the 1960s. At that time it was discovered that there were what appeared to be magnetic ‘stripes’ in the rocks on the ocean floor. The stripes represented sections in the rock of normal (the same as today) and reversed directions of the earth’s magnetic field, and this has been used as evidence for so-called sea-floor spreading and continental drift.1

Since their discovery, a lot of questions have been raised regarding the validity of these magnetic polarity (direction) reversals. Doell and Cox2 state that, ‘The reversed magnetisation of some rocks is now known to be due to a self-reversal mechanism’. Thus Jacobs3 claimed that, ‘Such results show that one must be cautious about interpreting all reversals as due to field reversal and the problem of deciding which reversed rocks indicate a reversal of the field may in some cases be extremely difficult.’

However, since this initial skepticism, a number of careful field, laboratory, and theoretical studies, as reported by McElhinney4 and Jacobs5, have shown that self-reversal cannot explain more than a small percentage of the reversely magnetized samples. Indeed, Humphreys6 has recently reviewed the evidence for the validity of these ‘fossil’ magnetism studies and has found that fully half of all the 200,000-plus geological samples tested have a measurable magnetization whose direction (‘polarity’) is reversed with respect to the earth’s present magnetic field. He concluded that the variety, extent, continuity and consistency of the reversal data all strongly suggest that most of the data are valid, so that there is no option but to accept that reversals of the earth’s magnetic field must have occurred.

Geophysicists have now recognized a sequence of 26 such magnetic field reversals in rock extending from the ’Upper Miocene’ to the present, presumed to represent the past 5.5 million years of the evolutionist’s timescale. In the fossiliferous rock layers of the evolutionist’s past 600 million years, from the lowest metazoan (multi-celled) fossils of the so-called ‘Cambrian explosion’ to the present, there appear to have been recorded in a continuous sequence roughly 50 of these magnetic polarity reversals.

The Mechanism?

The problem with the interpretation of these magnetic data is the presumed mechanism for operation of the earth’s magnetic field, and thus the presumed multi-million year timescale for these reversals.

The earth’s magnetic field is generally considered by most geophysicists to be associated with electric currents in the earth’s innermost region, the core, which is believed to consist of a metallic iron-nickel mixture, and is presumed to operate like a dynamo. These electric currents are believed to be produced by the slow circulation of molten material that carries unequal amounts of positive and negative electric charge.7,8 Consequently, reversal of the earth’s magnetic field could be expected to be a slow process, and thus the evolutionary view has been that a transition from one magnetic polarity (direction) to the other generally took millions of years. Certainly, at least several thousand years have now been estimated as necessary for the completion of one such reversal.9

In any case, this operational mechanism preferred by most geophysicists, the so-called dynamo hypothesis, has many problems associated with it which have been well documented.10-13

Since the field reversals are recorded in the fossil strata, the reversals must have happened when the strata were being laid down. Many creationists argue that Noah’s Flood produced most of the fossil layers in a single year. Thus, these reversals of the earth’s magnetic field have to be envisaged as occurring on average every week or two during the Flood year.

Such changes are obviously very rapid compared to the evolutionist’s multi-thousand-year or million-year-plus time-scales predicted by their dynamo hypothesis. However, creationists Dr. Thomas G. Barnes (Professor Emeritus of Physics, The University of Texas at El Paso) and Dr. D. Russell Humphreys (physicist at the Sandia National Laboratories, Albuquerque, and Adjunct Professor of Physics at the Institute for Creation Research, San Diego) have argued convincingly for a viable alternative hypothesis to the dynamo. They propose freely decaying electric currents in the earth’s core.14-18 This mechanism accounts for the real-time decay of the earth’s field over approximately the past 150 years, the current generated from such field decay correlating well with calculations of the amount of current actually present within the core. In addition, it can account for the magnetic reversals recorded in the rocks having taken place in a matter of only days to weeks!19,20

Found! A field test

A convincing test of Humphreys’ proposal, that reversals of the earth’s magnetic field must have taken only a matter of days or weeks within the time-frame of the year-long biblical Flood only 4,500 or so years ago, would be to look for evidence of such rapid reversals within rock layers that would have formed, or are known to have formed, that rapidly. Indeed, Humphreys has suggested this himself.21 He suggested that the best candidates for strata which clearly formed within a few weeks and yet contained a full reversal, would be distinct lava flows thin enough that they would have cooled below 500°C within a few weeks.

There are several sites where reversal transitions are recorded in the rock layers in some detail, continuously tracking both direction and strength in small steps.22,23 However, a large portion of these magnetic reversal transitions are not recorded in a single, rapidly formed, rock layer, and so don’t meet the criteria for a test case. Now two geoscientists have reported their examination of such a lava flow and found just such a polarity transition recorded in it.24,25

Coe and Prevot are respected palaeomagnetists, who for some years have been involved with a large group of geoscientists undertaking detailed investigations of magnetic polarity changes in the immense pile of lava flows at Steens Mountain in Oregon, near the California-Oregon border. These lava flows have been studied along two traverses up the mountain and are regarded as Miocene, with the reversal record ‘dating’ from 15.5 +/- 10.3 million years ago.

Coe and Prevot carefully sampled a relatively thin (1.9 metres) lava flow, designated as flow number B51, at a point where their team’s previous investigations had suggested a rapid transition (magnetic polarity change) was likely to have been recorded. A group of nine lava flows with similar magnetic polarity directions (essentially ‘reversed’ polarity), the last one being B52, precedes a 90° jump to the ‘normal’ direction of flow B50 above, which is followed up the sequence by a group of six flows with directions that are indistinguishable from one another but very similar to the ‘normal’ direction of flow B50. Palaeomagnetic measurements by Coe and Prevot on the numerous samples they collected through the entire thickness of flow B51 show a bumpy but continuous transition from the ‘reversed’ polarity in the lava flows below to the ‘normal’ polarity in the lava flows above, just as expected.

Next comes the question as to how long it took for this magnetic reversal recorded in the lava flow to occur? This can be reasonably ascertained if it is possible to calculate how long it took for the lava flow to cool to 500°C, since at that temperature the magnetization of the magnetic grains in the basalt would be ‘frozen’.

For this purpose, Coe and Prevot used a very simple model of heat conduction that has been long established,26 and which suffices to calculate a reasonable value of the cooling time of a basalt flow, as verified by the use of actual temperature measurements from similar lava flows on the island of Hawaii.27

From their calculations, they concluded that the entire lava flow would cool to 500°C or below in about 15 days. This means that the magnetic polarity transition recorded in the lava flow had to be made in less than two weeks!

Lava Flows
‘ ABC’ indicate successive lava flows. ‘D’ earlier sedimentary rocks. Arrows show direction of the magnetism ‘fossilized’ in the cooled lavas.

An ‘astonishingly’ rapid reversal rate!

Coe and Prevot commented:

‘This period [of 15 days] is undoubtedly an overestimate…Nonetheless, even this conservative figure of 15 days corresponds to an astonishingly rapid rate of variation of the geomagnetic field direction of 3° per day.’28

They also estimated that the minimum change in magnetic field intensity was at an average rate of at least 300 gammas per day. This compares with typically measured rates of geomagnetic variation globally today of only a few degrees per century and about 150 gammas per year.29,30 No wonder Coe and Prevot found the calculated rate in lava flow B51 at Steens Mountain hard to believe:

‘The rapidity and large amplitude of geomagnetic variation that we infer from the remanence directions in flow B51, even when regarded as an impulse during a polarity transition, truly strains the imagination…’31

With due caution, Coe and Prevot thus felt prompted to search for alternative explanations. However, since other hypotheses required ‘special pleading’, they decided that the most straightforward interpretation explains the data best, that is, ‘the balance of evidence now in hand weighs in favor of rapid geomagnetic field variation.’32 They concluded:

‘We think that the most probable explanation of the anomalous remanence directions of flow B51 is the occurrence of a large and extremely rapid change in the geomagnetic field during cooling of the flow, and that this change most likely originated in the [earth’s] core. This interpretation must remain tentative until our investigation is completely finished, but, if true, it has important implications for the reversal process and the state of the earth’s interior.’33

How significant?

With just this one study completed on this flow, being the one and only example to date of a rapid magnetic polarity reversal found recorded in a rapidly formed single rock layer, we cannot be dogmatic that Coe and Prevot are correct, although as far as we can tell their work is very meticulous and quite thorough. They have been suitably tentative, although they have explored every other possibility as far as they can ascertain. Of course, part of their reticence to accept the implications of their own results is due to their strict adherence to the evolutionary time framework in which the postulated dynamo generating the earth’s magnetic field has supposedly operated.

However, so significant is this discovery, Coe and Prevot’s paper was highlighted and commented upon in the international weekly journal Nature. 34 In that report, the author was cautiously favourable to Coe and Prevot’s interpretation of the palaeomagnetic data, being unable to challenge the data presented by Coe and Prevot, or the analysis by which they arrived at their conclusions.

However, quite predictably, the same author seemed reluctant to abandon his ingrained evolutionary view that thousands of years are necessary for a geomagnetic reversal. Instead, he appeared to be hoping that some alternative explanation will eventually emerge which would relieve him from the implications of the field data for reversals. But, he also admitted:

‘Palaeomagnetism has a history of giving shocks to the geological and geophysical community. Usually these are initially unpalatable, although they are later accepted.’35


So, as Humphreys says, if Coe and Prevot are correct, we can infer important facts about the earth at the time when this ‘Miocene’ lava was flowing at Steens Mountain, a time which many creationists would place during the latter part of the Flood year, or soon thereafter.36 Some physical process must have then been at work in the earth’s core which could produce very rapid reversals of the earth’s magnetic field.

The magnetic field change found recorded in flow B51 at Steens Mountain was about 50,000 times faster than the 2,000 plus years previously thought to be the theoretical minimum time for geomagnetic reversals, and millions of times faster than the shortest reversals previously found recorded in geological strata, that is, according to the evolutionary time-scale. But these actual field data were found exactly as the creation scientist Humphreys, working within a young earth/Creation/Flood framework, predicted they would be. So if the magnetic reversals have occurred in days and weeks rather than thousands and millions of years, then the earth’s rock layers, in which there is a continuous sequence of these magnetic reversals, are by implication probably only thousands of years old. Thus these data are important new evidence for a young earth.


  1. Ham, K.A., Snelling, A.A. and Wieland, C, 1990. What about continental drift? Have the continents really moved apart? The Answers Book, Creation Science Foundation, Brisbane, Australia, chapter 2, pp. 27–41. Return to text.
  2. Doell, R. and Cox, A., 1967. Magnetization of rocks., In: Mining Geophysics, vol. II, Society of Exploration Geophysicists, p. 452. Return to text.
  3. Jacobs, J.A., 1967. The Earth’s Core and Geomagnetism, Pergamon Press, Oxford, p. 106. Return to text.
  4. McElhinny, M.W., 1973. Palaeomagnetism and Plate Tectonics, Cambridge University Press, Cambridge, pp. 108–111. Return to text.
  5. Jacobs, J.A., 1984. Reversals of the Earth’s Magnetic Field, Adam Hilger Ltd, Bristol, pp. 29–38. Return to text.
  6. Humphreys, D.R., 1988. Has the earth’s magnetic field ever flipped? Creation Research Society Quarterly 25(3):130–137. Return to text.
  7. Jacobs, Ref. 5, pp. 13–18. Return to text.
  8. Merrill, R.T. and McElhinny, M.W., 1983. The Earth’s Magnetic Field, Academic Press, London, pp. 209-263. Return to text.
  9. Opdyke, M.D., Kent, D.V. and Lowrie, W., 1973. Details of magnetic polarity transitions recorded in a high deposition rate deepsea core. Earth and Planetary Science Letters,vol. 20, pp. 315–324. Return to text.
  10. Barnes, T.G., 1972. Young age vs. geologic age for the earth’s magnetic field. Creation Research Society Quarterly  9(1):47–50. Return to text.
  11. James, R.W., Roberts, P.H.and Winch, D.E., 1980. The Cowling anti-dynamo theorem. Geophysical and Astrophysical Fluid Dynamics  15:149–160. Return to text.
  12. Inglis, D.R., 1981. Dynamo theory of the earth’s varying magnetic field. Reviews of Modern Physics 53(3):481–496. Return to text.
  13. Humphreys, D.R., 1986. Reversals of the earth’s magnetic field during the Genesis Flood. Proceedings of the First International Conference on Creationism, Creation Science Fellowship, Pittsburgh 2:113–126. Return to text.
  14. Barnes, Ref. 10. Return to text.
  15. Barnes, T.G., 1973. Electromagnetics of the earth’s field and evaluation of electric conductivity, current, and joule heating in the earth’s core. Creation Research Society Quarterly 9(4):222–230. Return to text.
  16. Barnes, T.G., 1983. Origin and destiny of the earth’s magnetic field. Institute for Creation Research Technical Monograph No. 4, Institute for Creation Research, San Diego, California. Return to text.
  17. Humphreys, D.R., 1983. The creation of the earth’s magnetic field. Creation Research Society Quarterly 20(2):89–94. Return to text.
  18. Humphreys, D.R., 1984. The creation of planetary magnetic fields. Creation Research Society Quarterly 21(3):140–149. Return to text.
  19. Humphreys, Ref. 13. Return to text.
  20. Humphreys, D.R., 1990. Physical mechanism for reversals of the earth’s magnetic field during the Flood. Proceedings of the Second International Conference on Creationism,Creation Science Fellowship, Pittsburgh, Vol. 2 (in press). Return to text.
  21. Humphreys, Ref. 13, pp. 121. Return to text.
  22. Bogue, S.W. and Coe, R.S., 1984. Transitional paleointensities from Kauai, Hawaii, and geomagnetic reversal models. Journal of Geophysical Research 89(B12):10341–10354. Return to text.
  23. Prévot, M. et al, 1985. How the geomagnetic field vector reverses polarity. Nature 316:230–234. Return to text.
  24. Humphreys, D.R., 1990. New evidence for rapid reversals of the earth’s magnetic field. Creation Research Society Quarterly 26(4):132-133. Return to text.
  25. Coe, R.S. and Prévot, M., 1989. Evidence suggesting extremely rapid field variation during a geomagnetic reversal. Earth and Planetary Science Letter 92:292–298. Return to text.
  26. Jaeger, J.C., 1968. Cooling and solidification of igneous rocks. In: Basalts, H.H.Hcss and A. Poldervaart (eds), Interscience, New York,  2:503-536. Return to text.
  27. Shaw, H.R., Hamilton, M.S. and Peck, D.L., 1977. Numerical analysis of lava lake cooling models, I. Description of the method. American Journal of Science 277:384–414. Return to text.
  28. Coe and Prévot, Ref. 25, p. 296. Return to text.
  29. Langel, R.A. and Estes, R.H., 1985. The near-earth magnetic field at 1980 determined from Magsat data. Journal of Geophysical Research 90:2495–2509. Return to text.
  30. Langel, R.A., 1987. The main field. In: Geomagnetism, Vol. 1, J.A. Jacobs (ed), Academic Press, London, pp. 249–512. Return to text.
  31. Coe and Prévot, Ref. 25, p. 296. Return to text.
  32. Coe and Prévot, Ref. 25, p. 297. Return to text.
  33. Coe and Prévot, Ref. 25, p. 297. Return to text.
  34. Fuller, M., 1989. Fast changes in geomagnetism. Nature, vol. 339, pp. 582–583. Return to text.
  35. Fuller, Ref. 34, p. 583. Return to text.
  36. Humphreys, Ref. 24, p. 133. Return to text.