A quality-control project covering 56 pieces of mammographic equipment was carried out in Finland. The evaluation was based on the determination of contrast, film optical density, resolution, focus size, tube voltage, exposure time, half-value thickness, reproducibility and linearity of exposure, surface and exit dose, and defects at each centre when employing routinely used techniques. Dose and image quality were assessed using three phantoms. The results are compared with those obtained 5 years earlier and by other authors. Although the situation has, in many ways, improved considerably over the last 5 years (for instance, the average whole-breast dose has fallen on average by 84%, 7.37----1.10 mGy), there is still room for further improvement, especially in the optimization of the developing processes.