I recall here how latitude-dependent rotation imposed by the solar convection zone on the top of the radiation zone would burrow deep into the interior, owing to thermal diffusion, in any laminar and purely hydrodynamic model. Since helioseismology has shown that this differential rotation remains confined in a thin boundary layer, the tachocline, it means that the radiative spread is inhibited by another physical process; this process may be purely hydrodynamic (non-MHD), which is the scope of this chapter, or it may involve magnetic fields: those are considered by Garaud in Chapter 7 of this book. I will show that the confinement of the tachocline can be achieved through an anisotropic turbulent viscosity, whose cause and plausibility are discussed. Other hydrodynamic mechanisms are examined, such as internal gravity waves, which may also play a role in the tachocline. An alternative possibility is that the tachocline is fully embedded in the layer of penetrative convection, in which case no differential rotation would be applied on to the radiation zone.

Introduction

In 1990, I was invited with Ed Spiegel to give the principal lectures at the Woods Hole summer school. The theme of that year, ‘Stellar Fluid Dynamics’, was covered extensively by Ed, and I chose to focus on problems related to the rotation of stars. My last lecture, as it happened, was devoted to ‘flow between the Sun's convection and radiation zones and transport of chemicals’.