I can do some testing today, but it will be on my Dell laptop Win2k with Personal Oracle 8 since I am working at home. I will not be able to do testing against a real Oracle server until middle of next week.

>Yep, that construct does work fine. I'm finding a fairly drastic (IMO) speed difference, however, between executing the stored procedure (.25s to .5s) and running the same statement through SPT (.01s to .03s). The SP/SPT in question is a simple SELECT that retrieves 301 records from a table of 170,000.
>
>That may seem like not a very good test, since we wouldn't normally get that many "child" records. I also did a test though that gets one record based on a PK and that too seemed to be about .25s for the SP vs. .01 or less for the (same) SPT.
>
>I'm going to do more testing of this and will post results back over on the FoxPro forum on that "CRUD SPs on Oracle or Not" thread, but have you done any performance testing of executing an SP through the Oracle ODBC driver vs. using SPT through the same driver?
>
>Kelly