PROBLEMS IN ASTROMETRY: Parallax problems old and new

A bstract. The systematic corrections to the various parallax series used in the new Parallax Catalogue should probably not be applied. The average of these corrections over the entire sky is negative and if they are applied the result is that the average absolute parallax, at least for the southern stars, comes out smaller than the observed relative one. The printed absolute parallaxes would thus require corrections up to + "007. It is found from an as yet limited material that the relative parallaxes depend on the mean magnitude of the comparison stars. If this is confirmed the mean parallax of the I I mag. (pg) stars used as comparison stars must in many cases be very much larger than the statistical value derived from proper motions.