Jim;

Yes, but if you did 1,000,001 iterations it would make a "really big" difference! :)

Tom


>John,
>
>I just ran the exact same test under VFP 6.0 and I found that in 6 the m. is faster by 0.1 seconds on 1,000,000 assignments. So it seems , in my testing, your premiss is backwards and it is VFP 6 that is faster (by 0.1 seconds) for using the m. and VFP 7 is slower by the same amount.
>
>However, I do point out that the difference is 0.1 seconds over 1,000,000 iterations. This translates into 0.0000001 per assignment. Hardly earth shattering.