I have a large DB - >200 records. Some of the records relating to one supplier are not actually dupes but records that should be grouped as one entry or anyway identified and perhaps copied to another DB whilst deleting them from the current DB. Example for a std record: Supplier A, Sale ref, Sale date, Item ref, description, cost, colour and size. Usually these are unique records but for one supplier in particular a record might look like this:

Supplier A | 0909 | 8/31/10 | 388 | T-shirt | $40 | Red | Medium

Supplier A | 0909 | 8/31/10 | 388 | T-shirt | $40 | white | Medium

Supplier A | 0909 | 8/31/10 | 388 | T-shirt | $40 | blue | Medium

If I look for dupes in certain fields by inserting ! into each field I don't seem to get just the records above but other ones too even from other suppliers. Not sure I'm clear. Can anyone make out what I'm doing wrong? Thanks.

In your example, you do not show any duplicate records. They represent three different shirts, one of each color.

Try defining a calculation field that would return:

"Supplier A | 0909 | 8/31/10 | 388 | T-shirt | $40"

then look for duplicates in this field.