Jochen, I did try something like this earlier.

The issue I ran into was that the downstream systems would let through some
Latin-2 characters but not all. The systems I am dealing with are not well
documented regarding anything dealing with encodings or unicode. I am going
to need to experiment letter by letter to determine whether the systems will
accept a letter 'e' with an accent, but not with a caron or umlaut.

I suspect, given my business sponsor's, we'll probably get into substituting
characters, so if the letter 'e' with an umlaut is not acceptable to the
downstream systems, we'll just pass the letter 'e' by iteself.

We also would like to know why a given record was not sent downstream. If I
can somehow determine that the first several characters of a string are in
Russian or Japanese, I can note it in our audit table, and skip to the next
record.

Thanks
Bill

Clive, thanks for the headsup on the regex. I just Goggled around and found
Michael Kaplan had an interesting blog article below. Maybe I can use this
info: http://blogs.msdn.com/michkap/archive/2005/09/13/464416.aspx

which lead me to this:
http://msdn2.microsoft.com/en-us/library/20bw873z.aspx

I might be able to make this work...

Thanks
Bill