>A while back I asked for help in finding duplicate records. I've worked this job for a week and I have better information to give. This is an auditing application. First, there are several million records in the table. The auditors want to look for records where several of the fields either match or maybe values were transposed (like vendor number and transaction number or invoice number). They want to search the table an only return a set of records where there are not just unique records (one of everything), but where two or more records meet the criteria. Suggestions?

You can still do this with SQL try the following:

(in a program file)
Select * from myTable where testit() = .T. into cursor abc

function testit
llRetVal = .F.
do case
case mytable.field1 = mytable.field2
llRetVal = .T.
case soundex(mytable.field3 = soundex(mytable.field4)
llRetVal = .T.
case ...
case ...
endcase
return llRetVal