> No, honest. I don't think it is. Sorry. You're right of course. My misunderstanding was that fold recursively optimized its children, but as you correctly point out it only does the first ply. I'll check to see that I haven't missed any further optimization opportunities as a result. Roger --