4 Conclusions

Of the 86 stars classed for the first time in the near infrared, 27 had not MK classification to our knowledge. Of these 27, 4 are Am, 7 are F type classes V or IV, 11 are cool giants and 5 bright giants.

Let us also point out an example of confusion between Am and CS. Usually one finds Am stars called CS, but here we find an Am which is CS. The object is HD 70826 which figures as Am in Hauck (1986). We included it in our list of CS candidates because its correlation peak in the CORAVEL instrument for the measure of radial velocities did not correspond to an Am star.

We detect again, as Paper II, a large quantity of giants between types G7 and K2 with a maximum around G8 and an absence of early type G giants which predominated in the old classifications.

Our program of classification of cool components of CS stars is now complete (Papers II and III). From 266 stars observed as true CS or candidates, we found in the near infrared 184 spectral types of cool stars (G, K and M) and 82 corresponding to early type stars (B, A, Am, F).

We hope to continue this study in two directions. On one hand we wish to classify the hotter components in the blue region, since there exist a wide disparity of classifications for the same object. This would enable us to carry out a sound statistical study of the group. We notice that the majority of the radial velocities of these binaries are being studied by R. Griffin (Cambridge Observatory) and by ourselves with CORAVEL (OHP), so that for a certain number of objects the orbital elements are already known. The other direction is to confront the whole set of our classifications with the data from the Hipparcos-Tycho catalogues.

Acknowledgements

We appreciate very much the data and bibliographic references made available from the CDS at Strasbourg.