>>>< snip >>
>>>
>>>>This is not the way it should work. Can you post some repro code?
>>>
>>>There's mskb#Q165410 that addresses this problem.
>>
>>That always kills me ... "Microsoft is researching the problem". About 10 years in this case. Ascending/Descending was a problem even in FPW (using 2+ Alias and the same table).
>
>At least they don't give up <g>

Ha har ... I can see it now ... some MS software engineer ... chained to his desk ... long grey beard ... searching for the elusive ascending/descending bug ... sort of like the Count of Monte Cristo ... LOL