We evaluated (1) whether the sample transport time could lead to a significant loss of carbon through microbial respiration and to a change of measured respiration rates, which can be a problem in areas difficult to access, with a long travel time from field to laboratory; (2) whether the method used to quantify heterotrophic respiration for agricultural soils is adequate for horizons that remain always water-saturated or close to saturation. Surface horizons and deep Bh of Amazonian podzols were sampled and kept under refrigeration to maintain moisture of sampling time. Incubations of aliquot of the same sample were initiated on the sampling day and 3, 6, 9 and 12 days after sampling. Other aliquots were conducted on a tension table to given water potential (60 cm H2O ) prior to incubation. Soil samples, whether disturbed or not, should not be dried but kept at sampling moisture in semi-open plastic bags under refrigeration at 4 °C, respiration monitoring must be conducted without prior water potential adjustment. In such conditions,12 days between sampling and beginning of measurement did not affected respiration results. The method used for agricultural soils gave different results and does not make sense for soils under perudic moisture regime.

No abstract