>Rob,
>
>I no of no situation where _TALLY is inaccurate, if you have a reproducable example you should report it to MS so they can fix it.
>
>BTW, your work around will also fail if the array existed before the SELECT command that returns no rows. In this case the array will not be changed from what it was and the first element may contain a valid value.

Thanx for the feedback, Jim. Its been so long since we found this problem, that I don't have a specific example handy, but as I get some spare (?!!) time, I'll see if I can try it... I also seem to remember a reference to this in one of the FP magazines, some time past, but maybe not -- or maybe I misunderstood the reference...

As for the failure possibility you mentioned, I thought I had proposed to release the array first, then dimension it (to initialize it), then try to populate it, then test it. Why wouldn't that work reliably? What step is missing?

Thanx.

Rob