>I consider this a serious error since it involves a record pointer movement not specifically instructed by the program or initiated by the user, but for some reason the VFP team did not fix it. I don't know if it fell through the cracks or they just decided it wasn't serious (I would have a hard time fathoming that conclusion!).

From the sound of your description I concur. Even if it can be fixed with a few lines of code seems to show some redundant and now behaviour-split code in vfp. OTOH grids are complex beasts ;-)

regards

thomas