A respirometric experiment was used to assess the toxicity of a toxic element polluted-sewage sludge on two soil samples (SN and SO) differing, mainly, in their organic composition, texture and carbonate content. The essayed rates were 1.67, 3.34, 8.35 and 16.7% (dry weight basis), corresponding to total Zn, Cr and Ba concentrations ranging from 371, 105 and 514 to 3711, 1054 and 5 1.4 mg kg(-1), respectively. The experiment consisted of the incubation of the amended soil samples under controlled temperature and humidity conditions, followed by the observation of both, daily and cumulative basal respiration curves, and the comparison of the curves parameters from the fitted exponential mathematical models. Water-soluble potentially toxic elements concentration was determined at the end of the incubation period to confirm the reason of the toxicity differences found in the respirometric tests. All the sludge amended samples showed an increase of soil respiration compared to the unamended control, being positively related to the waste rate. A respiration inhibition appeared later (4 and 20 days in SN and SO soils, respectively, for the lower dose) and the time of delay depended on the waste rate used, being larger as waste dose increased. The toxicity level differed in the two soils essayed, mainly due to their basal microbial activity, and to their different properties, which determine the released elements bioavailability. Water-soluble elements determination at the end of the incubation revealed the presence of soluble Zn, Cr and Ba in different concentrations in the two soils. For instance, Zn concentration was, at the highest application dose, of 131 and 23 mg kg(-1), for SN and SO, respectively. So, the sludge toxicity was best observable at low rates and at incubation time periods long enough to allow the mineralisation of the waste oxidable organic matter. (c) 2006 Elsevier B.V. All rights reserved.