ABSTRACT The ability to respond to far-red-rich light is essential for seedlings germinating below dense canopies. Physiological and genetic studies have demonstrated that phytochrome A is the only photoreceptor mediating responses to far-red light. However, all phytochromes including phytochrome A are believed to be activated by red light and to be inactivated by far-red light. To address the fundamental question of why phytochrome A has its highest physiological activity at presumably inactivating wavelengths, we analysed light-induced degradation of phytochrome A in Arabidopsis. Rate constants were obtained for all reaction events in a two-step model of degradation. Based on biochemical data, the model includes a tagging mechanism preceding degradation. The parameterized model describes Pr accumulation, wavelength dependencies of degradation kinetics and steady-state levels as well as Pfr-induced Pr degradation. Subsequently, experimentally derived fluence rate response curves, action spectrum and response curves to dichromatic irradiation were compared to simulations based on the model of degradation. Two kinetically defined phytochrome subspecies, untagged Pfr and tagged Pr, have steady-state levels closely matching the physiological response curves. Therefore, sensing of far-red light by phytochrome A can be quantitatively explained based exclusively on regulated protein degradation.