Soil Eiol. Biochem. Vol. 17, No. 4, pp. 463468, 1985 Printed in Great Britain. All rights reserved 0038-0717/85 $3.00 + 0.00 Copyright 0 1985 Pergamon Press Ltd BACTERIAL DYNAMICS DURING DECOMPOSITION OF ALDER LITTER A. KJ(~LLER,* S. SIRUWE* and K. VEX-BERG* Institut for Sporeplanter, University of Copenhagen, 0. Farimagsgade 2 D. DEL-1353 Copenhagen IL, Denmark (Accepted I5 December 1984) Summary-Different groups of aerobic and anaerobic bacteria were enumerated in decomposing alder litter (Alnus glutinosa) and the underlying soil. Unspecific media with soil extract were used for total counts and media with single carbon compounds or different nitrogen compounds for specific functional groups. The numbers of aerobic, amylolytic and proteolytic bacteria were high after litter fall and decreased towards the end of decomposition. The fluctuations in the anaerobic groups of organisms were mainly influenced by moisture. The numbers of denitrifiers in the soil were highest during decomposition of litter in the winter, while the numbers of ammonifiers in litter were always high. INTRODUCTION Decomposition of leaves and litter in deciduous forest soils has been studied for many years. These studies have included determination of weight loss and chemical anatysis of litter, identification of micro- organisms in successions (especially microfungi), bio- mass determinations, respiration measurements etc. The role of microfungi in decomposing processes has been reviewed by Wicklow and Carroll (1982), Pugh (1980), and Kjoller and Struwe (1982), but only few studies have considered the bacterial role during decomposition, e.g. Sundman (1970), Hisset and Gray (1973), Gray et al. (1974), Kauri (1982, 1983) treating aerobic bacteria and Skinner (1964, 1975), Shelley et al. (1983) treating anaerobic bacteria. The purpose of this study was to demonstrate possible changes in bacterial groups and numbers related to the progressive decomposition of alder litter. The study was carried out in an alder swamp which is a system, where nitrogen and moisture conditions are considered not to be limiting either for aerobic decomposition in litter and surface soil or for anaerobic activity in the deeper layers. The numbers of different groups of aerobic and anaerobic bacteria were determined by monthly enumerations on different media. All incubations were carried out at lO”C, which is close to the annual mean temperature in Denmark. MATERIAL AND METHODS The site investigated was an aider swamp situated along a small river, M&e&en. North of Copenhagen, previously used in a fungal study (Kjarller and Struwe, 1980). The site is waterlogged most of the year and the pH of the surface soil is between 6.5 and 7.0. Alder leaves and litter were collected each month *Present address: Department of General Microbiology, University of Copenhagen, Safvgade 83, DK-1307 Co- penhagen IL, Denmark. from litter fall in November until disappearance in June. Soil samples were taken each month at 3 depths; 2, 10 and 25cm, representing the fermenta- tion layer, the humus layer, and the grey humus layer respectively. All samples were processed immediately after collection. The moisture content was determined after drying the samples at 110°C for 24 h and loss-on-ignition was determined in four replicates after ignition at 550°C for 4 h. Total nitrogen was determined by the microKjeldah1 method (Barnard and Chayen, 1965) in duplicate. The isolation and enumeration methods used were the plate count technique (Holm and Jensen, 1972), the MPN test (De Man, 1975) and the roil tube method (roll tubes with N, atmosphere) (Hungate, 1972). The results obtained from both the plate count and the roll tube enumerations imply the same sources of errors, due to choice of media, inhibition, separation of bacterial aggregates etc. All MPN enumerations were carried out as S-tube decimal dilution-tests. The 99% confidence limits for each MPN determination are for most values less than +0.5 times the actual value. All plate counts and roll-tube counts were made in five replicates, and in most cases mean values of approx. 100 colonies were obtained. Assuming a Poisson distribution, the individual values are deter- mined with an accuracy of *20%, corresponding to an interval of twice the standard deviation. Table 1 shows an overview of the groups of bacteria, methods and media for their enumeration. Composition of substrates and media Soif extract agar (Holm and Jensen, 1972>- glucose, 1.0 g; peptone, 1.0 g; yeast extract, 1.0 g; K,HPO,, 1.0 g; agar, 20.0 g; soil extract, 400 ml; distilled water, 600 ml; and actidione (cyclohex- imide), 40 mg. Cellulose, pectin, starch, gelatine and chitin agar were prepared according to Kjoller and Struwe (1980) with addition of 40 mg actidione to 1000 ml medium. 463