Short-term and medium-term effects of liming (CaCO3), fertilization [SCa(NO)3)2.NH4NO3] , and acidification on soil bioactivity were measured in a spruce stand in Southern Germany. The experiment was set up in a randomized block design. Acid precipitation lowered the pH, liming increased the pH, while fertilization caused only small alterations in p H values. Significant differences in soil moisture occurred only in the mineral horizons. The soil ATP content of the humus layers decreased in all plots (control included) up to day 100. On all sampiing dates, a pronounced decrease in ATP content followed the acidification. Minor decreases in ATP were observed after fertilization, while liming produced no defined effects. Similar trends, but less pronounced, were observed in the mineral horizons. Only a few significant correlations were found between p H values and ATP or between moisture and ATP within a treatment and sampling date.
Key words Lime
Fertilizer
Acidification SoiI organisms 9 ATP content to evaluate the consequences of these amendments on soil microflora and soil animals (B&~tth et al. 1980; Zelles et al. 1987 a, b;Nohrstedt et al. 1989;Funke 1991;Badalucco et al. 1992), but conflicting results have often been obtained, possibly due to non-standardized methodology and the heterogenous and complex soil environment (Badalucco et al. 1992). Furthermore, only a few sampling dates or replicates have been examined in these studies, either immediately after treatment or years later.
It is also important to differentiate clearly between soil biomass and soil activity. To obtain a complete assessment of soil activity, it is necessary to use a number of different methods, reflecting different aspects of soil metabolism (Zelles et al. 1987a;Nannipieri et al. 1990).
The objective of the present study was to compare both short-term and medium-term effects of liming, fertilization, and acidification on one aspect o f soil bioactivity (ATP content) in relation to soil moisture and soil pH. Numerous replicates and sampling dates were examined without pseudoreplication using a completely randomized block design (Hurlbert 1984).