When we consider life as a biogeochemical agent in the Earth system then it is quite well understood how bioactivity has altered the composition of Earth's atmosphere from a CO 2 dominated to the present nitrogen-oxygen composition (e.g., [1]). In the present review article [2], A. Kleidon attempts to outline a more general theory in which life would factor in the evolution of the entire planet, including its interior. While Kleidon rightly points out that the biosphere transforms significant amounts of solar energy into chemical energy through metamorphic reactions in sedimentary basins, it is not clear whether or not the Earth's interior machinery can make any use of this energy and if so, just how this energy is being used.
The latter question as such is not an entirely new one. Rosing et al. [3] have speculated that the growth of continents may be powered by chemical energy derived from solar energy through photosynthesis, deposited in sedimentary basins, and fed into the interior through subduction of sedimentary rock. Dyke et al. [4] have recently attempted to calculate a power balance for the cycle that would be closed by the rise of the continents and erosion of continental rock that -in turn -feeds the sedimentary basins.
It is widely agreed in the astrobiology literature that plate tectonics is a prerequisite for -at least -evolved life (e.g., [5]). Plate tectonics is essential in forming continental shields and free land masses. In addition and perhaps more importantly, models of the thermal and magnetic evolution of the terrestrial planets [6] suggest that plate tectonics is essential in cooling the deep planetary interior and drive the geodynamo either by cooling the core at a sufficient rate to drive thermal convection or by removing heat at a sufficient rate to allow the growth of an inner core. Growth of that core may release buoyancy that would again drive the dynamo.
But just how will plate tectonics arise and why is the Earth the only planet in the solar system that has plate tectonics? Theorists of mantle convection have learned to differentiate between mostly two basic forms of convection in fluids whose viscosity depends strongly on temperature and on the presence and concentration of free water: Stagnantlid and mobile-lid convection. In stagnant-lid convection, an immobile layer through which most of the temperature and viscosity variation occurs overlies the convecting layer. On a planetary scale this would be a planet whose outer shell would be immobile like on Mars, Venus and Mercury. In mobile-lid convection, the outer shell participates in the convection cycling. Plate tectonics would be one particular form of mobile-lid convection on a planetary scale. This mode of convection requires the plate to be ductile enough to allow it to participate in the flow. The temperature dependence of rock viscosity as such would certainly not allow a mobile lid. But if the rheology of the Earth is con-