Agricultural by-products (wheat straw, WS; sugar beet, SB; corn cobs, CC; rape extracted meal, REM; soya bean hulls, SBH; bagasse, B; almond shells, AS; and olive stones; OS) were analyzed by dynamic and isothermal thermogravimetry (TG) in inert and oxidative environment. The goal was to evaluate the effect of oxygen presence on thermolysis and to compare the environmental impact of individual samples. In the inert environment at 250°C, the sequence according to decreasing residue amount under dynamic conditions is B (95%) CC WS SBH AS OS REM SB (80%). This represents the decreasing role of hemicellulose and pectin components on thermal stability of individual samples. At 500°C the order changed to: REM (35%) OS AS SB SBH B CC WS (24%). The TG/derivative thermogravimetry (DTG) values determined in air atmosphere could not be used for deter- mination of individual macromolecular components due to overlapping of DTG maxima. For most of the studied sam- ples, the measured residues under dynamic conditions were greater in the inert than in the oxidative environment. Ac- cording to the thermooxidation maxima, the maximal dif- ference of measured residues on dynamic TG curves run in nitrogen and oxygen environments in the order of decreas- ing thermooxidation resistance: B (32%; 325°C) AS (29%; 450°C) OS (23%; 450°C) SBH (20%; 475°C) WS (21%; 300°C) SB (13%; 325°C) REM (15%; 500°C). This is due to the largest protein content of REM (29%) as well as pectin, hemicellulose, and lignin composition of samples. Under isothermal conditions the rate constants were determined by linear regression method and are considered as initial rate constants of material degradation. For inert conditions, the smallest rate constants and biggest activation energy (E*) were determined on CC. The values of the rate constants increase in the following order CC WS SBH B OS AS SB REM. In general it can be concluded that the rate constant values measured in oxygen are greater than analogical values measured in nitrogen. The determined E* values in oxygen were greater for AS, WS, B, SBH, and OS than the analogical values determined in nitrogen. They were decreasing in the following order: AS WS B SBH OS. On CC, SB and REM, E* values were lower in comparison with the E* calculated for inert conditions. We assume that when the E* value decreased in the presence of oxygen and when simultaneously smaller residue was formed under dynamic conditions, then the material is more unsuitable as a fuel from environmental point of view. It is because more oxygen was consumed and more material was gasified than on samples with closer values in both environ-