Since the cyclicality of RAND() is certainly greater than, say, the
number of rows, it's exceedingly unlikely to duplicate numbers unless
you're generating a huge simulation. In that case, XL's RAND function in
all versions is suspect enough that you should not use it for any
serious modelling work.
You may want to look at posts regarding RAND() in the archives
http://groups.google.com/advanced_group_search?q=group:*excel*
by Braden, Grove, Lewis and others.
Perhaps you're trying to do something different - post back if so.
