A measurement of the magnetic field direction at the site of major flares

Lundstedt et al. (1981) showed that the direction of the photospheric magnetic field at the site of a flare is a good predictor of the solar wind velocity observed at Earth four days later. We describe here how the field direction was obtained, and discuss possible errors involved in the determination of the angle. The discussion also includes a characterization of the solar active regions.Now at Institute for Astronomy, Lund University, Lund, Sweden.