4 Conclusion

The CU Vir magnetic field modelled by the method of "magnetic charges" verifies the assumption of Hatzes of a displaced dipole mode.

The distribution of chemical elements versus the magnetic field distribution is such that it is impossible to argue unambiguously, that the diffusion theory is justifiable. This uncertainty is connected first of all with the lack of observational data.

The absolute stellar magnitude of the star CU Vir is given by $M_{v} = 0\hbox{$.\!\!^{\rm m}$ }3$ (Gomez et al. 1988), the effective temperature is $T_{\rm e} = 12\,460$ K (Glagolevskij 1994). Hence the absolute bolometric stellar magnitude is $M_{\rm b} = -0\hbox{$.\!\!^{\rm m}$ }39$. This implies that the star is located in the Hertzsprung-Russell diagram between the Zero Age Main Sequence and the line of the luminosity class V, that is, the star has only recently been formed as a magnetic CP star after having arrived at the ZAMS (Glagolevskij & Chountonov 1998). It might be proven that the nonsymmetric structure of the surface magnetic field and the complex distribution of chemical elements are the result of recent formation of the magnetic field and chemical anomalies. The field is likely to rise not simultaneously to the surface in individual regions. If it is generated at all, then the generation conditions on the surface are dissimilar. These phenomena are undoubtedly related to the complex distribution of physical conditions in young stars, possibly as a result of fall out of big accretion masses during the previous evolutionary phases.

Additional measurements probably could throw light on the fine structure of the magnetic field in CU Vir.

Acknowledgements

This work has been performed with the support of the Academy of Sciences of Russia by the Grant "Astronomy".