『Abstract
　We review the nature and importance of soil factors implicated in the formation of secondary ferrimagnetic minerals in soils and palaeosols worldwide. The findings are examined with respect to temperate regions through a comprehensive analysis of over 5000 samples of surface soil from England and wales taken from a 5×5 km grid. Over 30 soil and environmental attributes are considered for each sample as proxies for soil forming factors. Measurements of low field magnetic susceptibility (mass specific) and frequency dependent susceptibility (mass specific and percentage) on each sample provide estimates of the concentration and grain size of ferrimagnetic minerals.
　Maps of soil magnetism across England and Wales show non-random distributions and clusters. One subset of data is clearly linked to contamination from atmospheric pollution, and excluded from subsequent analyses. The concentration of ferrimagnetic minerals in the non-polluted set is broadly proportional to the concentration of minerals falling into the viscous superparamagnetic domain size range (〜15-25 nm). This set shows clusters of high magnetic concentrations particularly over specific parent materials such as schists and slates, mudstones and limestones.
　Bivariate analyses and linear multiple regression models show that the main controlling factors are parent material and drainage, the latter represented by soil drainage classes and particle size. Together these two factors account for 〜30％ of the magnetic variability in the complete dataset. A second group of factors, including climate (mean annual rainfall), relief (slope and altitude), and organisms (land use, organic carbon and pH) have subordinate control. Climate, as represented by mean annual temperature, and also pedogenic time are deemed not relevant at these spatio-temporal scales.
　The findings are consistent with a largely abiotic system where the role of iron-reducing bacteria appears minor. At coarse spatial and temporal scales, secondary ferrimagnetic mineral formation is controlled by the weathering capacity to supply Fe to the surface soil. At finer soils, soluble Fe precipitates as ferrihydrite before transformation in response to periodically anaerobic conditions into other minerals including nanoscale magnetite.

Keywords: soil magnetism; magnetic susceptibility; frequency-dependence; soil forming factors; England and Wales』

Contents
1. Introduction and aims
2. Soil forming factors and magnetism
　2.1. Parent material
　2.2. Climate
　2.3. Relief and drainage
　2.4. Soil organisms and vegetation
　2.5. Time
3. Methods and techniques
　3.1. Field sampling and sample analysis
　3.2. Datasets of soil factors
　3.3. Data analysis
4. Magnetic patterns and soil factors
　4.1. Magnetic data
　4.2. Magnetic spatial patterns
　4.3. Association with soil factors
　　4.3.1. Parent material
　　4.3.2. Climate
　　4.3.3. Relief and drainage
　　4.3.4. Organisms and vegetation
　　4.3.5. Time
5. Interacting soil factors
　5.1. Correlation and multiple regression models
　　5.1.1. Considerations
　　5.1.2. Zero order correlations
　　5.1.3. Multivariate regression analysis
6. Discussion
　6.1. Magnetism and soil factors
　　6.1.1. Parent material
　　6.1.2. Climate
　　6.1.3. Relief and drainage
　　6.1.4. Organisms and vegetation
　　6.1.5. Time
　　6.1.6. Unexplained factorial variance
　6.2. SFM formation
　　6.2.1. SFMs and soil types
　　6.2.2. SFM concentrations and distributions
　　6.2.3. SFM conceptual model
7. Conclusions
Acknowledgements
Appendix A. Supplementary data
References