As early as 1922 Waksman emphasized the use oj microbiological analysis of soil as an index of soil fertility. Since that time there have been few comparative investigations of soil microbiology and relatea soil properties. For a part of Exmoor, three main soil phases were identified within the Ashcombe soil series: unreclaimed, reclaimed and reverted. At sites within these phases four major groups of microorganismsfilamentous fungi, yeasts, actinomycetes and bacteria -were enumerated at several depths in the profile, using a soil-dilution plate technique. Significant differences were found between soil phases in terms of the numbers of various groups of microorganisms at depths of between 30-50 mm and 0-20 mm. Microbial estimates generally declined with depth down the profile. For all depths in the reclaimedphase, actinomycetes -and bacteria exceed filamentous fungi and yeasts by at least an order of magnitude. This distinction is not as clear-cut in the unreclaimed profiles. The reverted profiles occupy positions intermediate between the reclaimed and unreclaimed patterns. The relations between different groups of microorganisms within and between profiles are described in some detail. Microbial ratios F: Y, A : B and F: B were determined. In the reclaimed phase F: Y is less than or close to unity, whilst in the reverted and unreclaimed phases the ratio is considerably greater than 1-00. A : B is greater than 100 in the case of two unlimed reverted sites, but is less than 1-00 for all other samples. The numerical estimates of microorganisms and selected ratios are used to illustrate soil phase differences resulting from the activities of man and his grazing animals in an upland environment. Ecological interpretations of the results are attempted. Introduction Microorganisms are widely acknowledged as being of fundamental importance in soil formation and development, by virtue of their intimate and vital links with the decomposition of organic material. Considerable literature relates to the presence of various groups or species of microorganisms in particular soils, or in some defined zone of the soil such as the rhizosphere (e.g. Parkinson & Waid, 1960; Alexander, 1961; Waksman, 1967; Gray & Parkinson, 1968; Gray & Williams, 1971, all of which contain excellent reviews). Many quantitative studies have been made of single soil profiles, with particular reference to certain microbial groups, and of several profiles with reference to a single group of microorganisms. For instance, bacterial numbers in soils at different depths were estimated by Waksman as long ago as 1916. Carmo-Sousa (1969) reviewed both the quantitative and qualitative distribution of yeasts in soils from a wide range of environments. Burges (1963) examined fungal activity in the various horizons of a Cheshire podzol, and for the same soil Parkinson & Coups (1963) measured the physiological activity of the soil population. A comparative study of actinomycete populations in acid, podzolic and neutral mull forest soils was made by Corke & Chase (1964), whilst Strzelczyk et al. (1969) have investigated actinomycetes from several arctic soils. As early as 1922, Waksman suggested the use of microbiological analysis of soil as an index of fertility. Later, using counts of microorganisms from fifty soils in New South Wales, Jensen (1934) established consistent relationships between the numbers of bacteria, actinomycetes and fungi, and certain soil properties. Numbers of all three groups were correlated with organic matter. Soil reaction was significantly correlated with numbers of fungi. Of particular significance in the context of the present work are the relations which Jensen describes between soil properties and the ratios of groups