Abstract Low temperature oxidation (LTO) of hydrocarbon liquids generally results in a more viscous end product; this has clearly been shown in the literature of the past 30 years. However, under the right conditions, LTO can be used to achieve viscosity reduction in heavy oils. The In Situ Combustion Group at the University of Calgary conceived of a two-stage LTO process whereby oil is contacted with air, first at low, then at elevated, temperatures. The first, low temperature, step incorporates oxygen into some of the hydrocarbons, yielding labile bonds that should break at lower-than-usual temperatures. Once these free radicals are formed, the second step promotes bond cleavage at higher temperatures, resulting in shorter chain hydrocarbons. In a field situation, this process would be analogous to first injecting air into a formation at low temperature, then starting a steam soak or steam flood. Experimental runs carried out on Athabasca bitumen examined the effects of oxygen partial pressure, temperature, reaction time, and the presence of rock and brine. On completion of each experiment, the gas composition was determined using gas chromatography, water acidity (pH) was measured, and the hydrocarbon products were analysed for coke and asphaltenes contents, viscosity, and density. Some instances of viscosity reduction have been observed; these are linked to lower oxygen partial pressures, higher second stage temperatures and longer run times. This paper discusses the experimental work, and estimates the optimum conditions for successful viscosity reduction of a given heavy oil. Introduction Heavy oil and oil sands are important hydrocarbon resources that total over 10 trillion barrels, nearly three times the conventional oil in place in the world. The oil sands of Alberta alone contain over two trillion barrels of oil. In Canada, approximately 20﹪ of oil production is from heavy oil and oil sand resources(1). The application of thermal energy to increase heavy oil recovery has become more popular as conventional reserves decline. Steam injection accounts for the majority of the thermal recovery projects currently in operation; however in situ combustion offers many theoretical advantages if the operational characteristics of the process are incorporated in the design and operation of the field project. A major difficulty encountered in operating in situ combustion processes is low temperature oxidation (LTO), which involves oxygen addition reactions that occur at temperatures lower than 300 °CDATA [C. Typically, the byproducts of these reactions are oxidized hydrocarbons that have an increased polarity. This makes them more viscous, and thus detrimental to the in situ combustion process. Because of the major impact that LTO can have on the performance of an in situ project, a significant number of investigations have been carried out on the nature and effect of LTO reactions(2-27). In some circumstances, however, it may be beneficial to subject oil to LTO. The experimental results of Cram and Redford showed that air/steam combinations can provide better recovery rates and better thermal efficiencies than steam alone, at comparable volumes of steam injected, when the process is carried out in the low temperature oxidation region((28). It is believed that energy generation by the exothermic oxidation reactions is a significant factor in the LTO process.