It is the thesis of this chapter that the plates are part of one recognisable mode of mantle convection, driven by the top thermal boundary layer of the mantle. I argued in Chapter 8 that there is no fundamental reason why the modes of flow driven by the different thermal boundary layers should be the same. You will see in Chapter 11 that mantle plumes seem to be a distinct mode of mantle convection driven by a bottom thermal boundary layer, and that in the mantle these two modes seem to behave in substantially different ways, and are not even very strongly coupled. It is therefore useful to identify explicitly two modes of mantle convection: the plate mode and the plume mode. I have previously also referred to the plate mode as the plate-scale flow [1, 2] because the flow seems to be quite strongly constrained to have the horizontal scale of the plates. This will be addressed in Section 10.2.

In the first half of this chapter I present a series of numerical models that illustrates the influence that various material properties have on the form of mantle convection. We look at the influence of the mechanical properties of the lithosphere, the effect of the increase of mantle viscosity with depth and the possible role of phase transformations. We also estimate the amount of heat transported by the plate–mantle system. In the second half of the chapter we look at how well the resulting conception of mantle convection matches observational constraints, especially surface topography and heat flow and the internal structure revealed by seismic tomography. The chapter concludes with a summary of the conception of the plate mode of mantle convection that is developed here.