6 Conclusion

The anisotropic mass loss by stellar winds is a complex 2-D problem, which influences stellar evolution. Here, we have devised a relatively simple method to treat carefully this problem within the context of the current 1-D stellar models.

We conclude that the anisotropic mass loss may play a significant role in the evolution of the fastest rotating OB stars. The polar enhanced mass loss rates make their rotation velocity faster during MS evolution and this may lead some stars close to the break-up limit.

For the stars with very high mass loss rates above $10^{-5}~M_{\odot}$ yr^-1, as is the case for OB stars with masses higher than 60 $M_{\odot }$, for WR stars and LBV stars, the anisotropic mass loss will shape and determine the evolution of the surrounding nebulae (Lamers et al. 2001), but will have little effect on the internal evolution, because the extra-torques due to the anisotropies have not the time to be transmitted inward by shear turbulence and circulation before the concerned layers are ejected.

For rotating stars with $T_{{\rm eff}}$ lower than about 21 000 K, an equatorial ejection may occur and have important consequences for the further evolution of the stellar angular momentum and for the instabilities oin the outer layers.

Acknowledgements

I express my gratitude to Dr. Georges Meynet for his many encouragements and for his great help during this work.