>>Also, VFP does not take whole index to memory, if it would be so, it just could not work with large tables on network.
>
>When i took a second look I did come to a more detailed conclusion. Rushmore must have all the index records regarding the resultset in it's memory to work. If it cannot load all the index records rushmore will not work.
>
>Documentation Developers Guids FPW 2.6 Page D17-2
>
Walter,

So then the conclusion is; since the optimization may not work under certain conditions we should prevent it from working under all conditions?

That doesn't make sense to me. Having an index on deleted will optimize many queries and there are some that it can't do because of table size, however I am not willing to give up the optimization on those that it does work for.