A note on relative motion in the general three-body problem