A COMPARISON OF FIELD ESTIMATES AND LABORATORY ANALYSES OF THE SILT AND CLAY CONTENTS OF SOME WEST MIDLAND SOILS

Abstract

SummaryField estimates of silt and clay contents were compared statistically with the silt (2‐60 μm) and clay (<2 μm) contents determined by the pipette method of 184 soil horizons from a large range of west Midland soils (clay contents 1–81 per cent and silt contents 3–71 per cent). Regression equations were calculated (a) for each of four surveyors using combined topsoil and subsoil data and (b) for surface and subsoils separately using combined data for the four surveyors. There is slight evidence of ‘operator bias’ and a tendency for all four surveyors to underestimate clay in surface horizons though these conclusions have limited significance because of the small number and the distribution of some of the sample populations. Single equations for the combined data from all operators and all horizons explained 75.5 per cent of the variation in field estimates of silt content and 85.4 per cent of the variation of clay estimates. For silt, the regression line almost passes through the origin with a slope not significantly different from unity, whilst for clay, the line intercepts the y‐axis close to zero and has a slope of 0.904. The results show that, with experience and adequate reference samples, surveyors can confidently estimate the particle‐size distribution of a wide range of soils. These studies were done during soil mapping in Staffordshire, Hereford and Worcester, and Salop using the newly introduced soil classification (Avery, 1973) and revised Handbook (Hodgson, 1974) of the Soil Survey of England and Wales.