6 Conclusions

We have solved a number of Hipparcos missing or poor solutions taking advantage of the results of the Tycho-2 project. Intensive use has been made of the Tycho-2 imaging and double star software.

One interesting conclusion from our investigation is that stars in the Hipparcos Catalogue with position errors above some 60 mas, should be used with great caution. We believe these solutions are mostly very wrong, although we did not select our sample for testing that idea.

Many of the systems we present, have separations in the unfortunate range of approximately $13\hbox{$^{\prime\prime}$ }- 20\hbox{$^{\prime\prime}$ }$, which is particularly difficult for the Hipparcos instrument and the derived separations should be used with some caution.

There are still probably a few hundred unsolved problems left among Hipparcos stars with poor or missing solutions. Some, we have been unable to study and others we have studied but rejected because we did not fully trust our solution. Additional information will be needed for their resolution.

Acknowledgements

This work was supported by the Velux Foundation of 1981 and by the Danish Space Board.

For three stars we have been using the Digitized Sky Survey, produced at the Space Telescope Science Institute from images made at the UK and Palomar Schmidt telescopes.