[Warren Postm4]

I have an auto-generated index, and here is a sample of how it looks:

Banana ....
Orange ...
Banana ....
Pear ...
Banana ....
Steve....

I would like only one top level index heading:

Banana ....
Orange ...
Pear ...
Steve ...

I would have thought that Word would do this automatically. I have carefully
checked the XE entries, and the typing of "Banana:Orange" and "Banana:Steve"
both use the same first entry (Banana). But they appear three times in the
index, each with one item underneath, instead of with one combined top level
heading, and the three entries underneath it.

I opened it up in Office 2010 Word beta to see if it works better here, and
it does the same thing. I regenerated a new Index field into the document,
and it does it there too. I am stumped. I even did a search-replace with
Banana and replaced it with Apple, and it found and replaced all text and XE
entries, but my manual (which is not about fruit, but I figured I wouldn't
bore you all with technical terms that have nothing to do with my question)
index is totally full of duplicates like this. In other places, I have cases
where Word is working as expected. There is a space character in between
two words in all cases where it isn't working, and yet I know it's not some
other weird unicode character that looks like a space. I even took the spaces
out and changed them to dash (-) characters, and I get the same problem.
Then I changed them to a single letter (x) and it still happens. Word is
trying to make me lose all my remaining hair follicles, I think.