Quaternary Science Reviews, Vol. 10, pp. 73-85,1991. 0277-3791/91 $0.00 + .50 Printed in Great Britain. All rights reserved. © 1991 Pergamon Press plc ENVIRONMENTAL MAGNETISM m A PERSONAL PERSPECTIVE Frank Oldfield Department of Geography, University of Liverpool, P.O. Box 147, Liverpool L69 3BX, England Environmental magnetism is a versatile methodology applicable to a wide range of environmental studies and contexts. The complexity and unfamiliarity of many natural magnetic assemblages, coupled with the very versatility and sensitivity of the measurements can often limit interpretations at present, and there is an urgent need to improve the theoretical and quantitative basis for interpreting the results derived from these newly accessible techniques. There is also a need to make the results both more objective and more readily communicable to specialists in related fields. Examples are shown which illustrate the range of applications of magnetic measurements, and suggestions given as to the ways in which, in the future, environmental magnetists may be able to overcome current problems. INTRODUCTION Over the past fifteen years there has been a rapid expansion in the use of magnetic measurements, in effect as a newly opened window on order in environ- mental systems. Opening the window depended on advances in electronics and computing which made the view through it widely accessible to many scientists other than the specialist geophysicists and 'palaeomagi- cians' who had taken the first glance. Looking through it has been the privilege of a rather diverse band of environmental scientists, many of them, like me, poorly qualified to make the most of the newly unfolding vistas, yet so beguiled by the prospect that they have been unable to turn away. Coming into environmental magnetism from an earlier research career devoted largely to the time consuming business of pollen analysis, what first impressed me was the relative speed with which apparently interpretable magnetic measurements could be carried out. I spent a lot of time repeating to fellow scientists a proselytizing liturgy which invariably included the words 'cheap, simple, rapid and non-destructive'. It took the gentle chiding of Ed Deevey, to whose memory I dedicate this review, in his encouraging appraisal of the chapter outlines of a proposed (and now published) book, to bring home to me the self-evident truth that the really significant virtues of the methodology lay in its capacity to address problems which were inaccessible to existing techniques. Nose to tree, I'd lost my view of the wood. This was for me just one of several examples of Ed Deevey's quite extraordinary ability to place the activity of research into the broader context of scientific concern and endeavour. This ability has been reflected in scientific prose of arresting insight and quite startling elegance which, in failing to emulate, we may neverthe- less aspire to honour. Some attempt to review the state of the environ- mental magnetists' art is timely for we have passed through that first trusting, romantic stage of rapid and sometimes rather hazardous speed-skating over ever larger patches of ice of unknown thickness, into the second stage of much more closely critical reappraisal, the stage when the cracks begin to show. Whitehead (1929) maps out these stages delightfully and antici- pates a third involving some emergence of a canon or synthesis forged from the reconciliation of creation and critique. This still lies a long way ahead, though it is just possible that we can identify ways forward which avoid the worst of the blind alleys. I begin by setting out a personal view of the strengths and weaknesses of the methodology as it has developed up to the present, then try to suggest some of the ways in which we may hope to achieve a fuller realization of its potential. THE PRESENT POSITION The rocks of the lithosphere are the only significant primary sources for almost all the magnetic minerals which pass into the atmosphere and hydrosphere and become the concern of the environmental magnetist. Weathering and soil formation transform iron com- pounds in ways which reflect both the mineralogy of the parent material and the biogeochemical processes acting within the regolith under prevailing and, in many cases, past conditions. In consequences, profiles of magnetic properties from a deep, mature regolith provide a well differentiated expression of weathering and of soil forming processes at each depth (Fig. la). This concept of magnetic 'horizonation' is a far cry from the early preoccupation with the simple 'enhance- ment' of susceptibility in topsoils, as a result of the neoformation of magnetite and/or maghemite by fire or 'fermentation' (Mullins, 1977), or even bacterial action (Fassbinder et al., 1990). Nevertheless, it is no more than a starting point for further study since it raises more questions than it answers. Clearly, the magnetic mineral and grain size assemblages at each depth are beautifully sensitive to changing biogeochemical condi- tions, but how quickly do these assemblages form? How stable and persistent within the regolith are they once formed? How close to quantitative specification of 73