Lithium is an excellent tracer of mixing in stars as it is destroyed (by nuclear reactions) at a temperature around ~ 2.5 × 106 K. The lithium destruction zone is typically located in the radiative region of a star. If the radiative regions are stable, the observed surface value of lithium should remain constant with time. However, comparison of the meteoritic and photospheric Li abundances in the Sun indicate that the surface abundance of Li in the Sun has been depleted by more than two orders of magnitude. This is not predicted by solar models and is a long standing problem. Observations of Li in open clusters indicate that Li depletion is occurring on the main sequence. Furthermore, there is now compelling observational evidence that a spread of lithium abundances is present in nearly identical stars. This suggests that some transport process is occurring in stellar radiative regions. Helioseismic inversions support this conclusion, for they suggest that standard solar models need to be modified below the base of the convection zone. There are a number of possible theoretical explanations for this transport process. The relation between Li abundances, rotation rates and the presence of a tidally locked companion along with the observed internal rotation in the Sun indicate that the mixing is most likely induced by rotation. The current status of non-standard (particularly rotational) stellar models which attempt to account for the lithium observations are reviewed.