Short communication The ATP concentration in the soil microbial biomass M. Contin a , A. Todd b , P.C. Brookes b, * a Dipartimento di Produzione Vegetale e Tecnologie Agrarie, University of Udine, Via delle Scienze, 208, I-33100 Udine, Italy b Soil Science Department and Statistics Department, IACR-Rothamsted, Harpenden, Herts AL5 2JQ, UK Received 2 March 2000; received in revised form 17 July 2000; accepted 9 August 2000 Adenosine 5 0 -triphosphate (ATP) occurs in all living cells but exocellular ATP has a half-life of less than 1 h (Conklin and MacGregor, 1972). Adenosine 5 0 -triphosphate therefore provides a useful indicator of life in soil. Soil ATP and biomass C concentrations are closely correlated in most studies so soil ATP can serve as a measure of soil microbial biomass (Jenkinson and Ladd, 1981). Our aim was to extend the original study of Jenkinson (1988) who correlated all data then available between ATP and biomass C. We there- fore examined appropriate published data available up to 1996 and some of our own, more recent, results (Contin, unpublished), in addition to Jenkinson's (1988) data. This gave a much larger sample, and measurements done under a wider range of conditions e.g. different methodologies, substrate-amended soils, soils below pH 4.8 and and soils from paddy-arable rotations. Despite this, soils from the Northern and Southern hemispheres, grassland, arable, woodland soils and soils from paddy arable rotations all had remarkably similar biomass ATP concentrations (range ca. 9±12 mmol ATP g 21 biomass C) to those of Jenkinson (1988). Jenkinson (1988) laid down strict criteria for the inclusion of biomass and ATP analyses in his original correlation. These were: 1. The soils had received a conditioning incubation of at least several days at 258C before analysis. 2. The soil moisture contents were below 60% water hold- ing capacity. 3. The soil pHs were not below 4.8. 4. All biomass C measurements were done by Fumigation± Incubation (Jenkinson and Powlson, 1976). 5. All ATP measurements were done following extraction with strongly acidic reagents. Under these conditions, and averaged over published values for a wide range of Australian, Danish, English and New Zealand soils from the surface depth of arable, grassland and woodland sites, the soil microbial biomass contained 11.7 ^ 2.74 (standard deviation; n  89) mmol ATP g 21 biomass C. Thus Jenkinson (1988) con®rmed the original ®ndings (Jenkinson et al., 1979; Oades and Jenkinson, 1979) that the ATP concentration of the soil microbial biomass is suf®ciently constant for it to serve as a `useful if rough measure of soil biomass'. This is only true provided the above conditions are ful®lled. For example, if neutral or alkaline extracting reagents are used (e.g. Paul and John- ston, 1977) the measured biomass ATP concentration is too small because phosphatases remain active long enough to denature a (variable) proportion of the biomass ATP (Brookes et al., 1987). Since Jenkinson (1988) published the original correlation there have been some signi®cant changes in biomass and ATP applications and biomass methodology. For example, biomass C is now almost invariably measured by fumigation extraction (FE) (Vance et al., 1987b) rather than fumigation incubation (FI) (Jenkinson and Powlson, 1976). The FE method permits measurement of biomass in situations where FI is invalid, for example soils ,pH 4.8 (Vance et al., 1989), soils containing actively decomposing substrates (Ocio and Brookes, 1990) and waterlogged (e.g. paddy soilsÐexcluded from this study) (Inubushi et al., 1989). Similarly, since the original ATP methodology was published (e.g. Jenkinson and Oades, 1979), ATP measure- ments have also been shown to be valid in acid soils (e.g. Vance et al., 1987a), soils where substrates are actively decomposing (Ocio and Brookes, 1990) and waterlogged soils (Inubushi et al., 1989). There have also been a large number of papers published throughout the world on ATP measurements in soil since Jenkinson's (1988) work. It therefore appeared appropriate to re-examine the world literature base of ATP and biomass relationships to see if Jenkinson's (1988) correlation still holds for the larger and more diverse database which now exists. Accordingly, his original data set (D. S. Jenkinson, perso- nal communication) was combined with all the data in the Soil Biology & Biochemistry 33 (2001) 701±704 0038-0717/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved. PII: S0038-0717(00)00204-2 www.elsevier.com/locate/soilbio * Corresponding author. Tel.: 144-1582-763133; fax: 144-1582- 760981. E-mail address: philip.brookes@bbsrc.ac.uk (P.C. Brookes).