Evaluation of the temperature sensitivity of soil organic matter (SOM) decomposition is critical for forecasting whether soils in a warming world will lose or gain carbon and, therefore, accelerate or mitigate climate warming. It is usually described, using Arrhenius kinetics, as increasing with the stability of the substrate in laboratory conditions, where substrate availability is non-limiting and where chemical recalcitrance, therefore, predominantly regulates stability. However, conditions of non-limiting subtrate availability are rare in the undisturbed soil, where physicochemical protection of substrates may control their stability. The aim of this study was to assess the temperature sensitivity of decomposition of SOM with contrasting stability in the field. Our conceptual approach was based on in situ measurements of soil CO2 efflux at a range of temperatures from root exclusion plots of increasing age (1 month and three decades) and, therefore, with SOM of increasing stability. From a set of short-term measurements in spring, using diurnal temperature variation, the relative temperature sensitivity of SOM decomposition decreased significantly (p < 0.0001) with increasing SOM stability, and was weak (Q(10) < 1.3) in long-term root exclusion plots. This result was confirmed in a similar set of short-term measurements repeated later in the year, in summer, as well as from an analysis perfomed at the seasonal timscale. We provide direct field evidence that the temperature sensitivity of SOM decomposition decreases with increasing stability, in direct contrast with Arrhenius kinetics prediction, and therefore show that stability of SOM in the field cannot be the sole result of chemical recalcitrance. We conclude that the physicochemical protection of SOM, which controls SOM stability in the field, constrains the temperature sensitivity of SOM decomposition under field conditions. (C) 2019 The Author(s).