>Bruce,
>
>Firstly, I would not necessarily expect the Resource Meter to report any change in RAM, given what has been reported about VFP's allocation and usage of memory. Have you seen fluctuations before (of substance) with it?
>
>My SWAG would be all the openings by SQL of the table(s) from which the Cursors are built, including possibly the cursors themselves.
>
>Might it be possible that your STUFF/SUBSTR are creating memvars which are hanging around?

Hi Jim - Well, I do see the R.M. move lower when I have object-related memory problems at times - but there is virtually none of that in this case...

I suspect the cursors, too, but I don't know why there should be a problem, and if there is why it doesn't show up sooner. It's the same cursor name every time, I've tried with/without NOFILTER, and I do USE IN at the end of each loop on the cursor.

The string appear innocent, I use only one 500 char long form property and a couple short locals...I've also tried it on three different machines...I'm stumped, it looks like the machines just get tired after a while :)