The Mime headers are faulty. You are coding Content-Transfer-Encoding: 7bit  when you are including an 8-bit character xA3. To be correct you should also be coding charset="CP1252" as that charset has a representation for the pound sign. Or use the html sequence &#A3  See the table:

http://www.microsoft.com/globaldev/reference/sbcs/1252.mspx 

Martin 

My question is not specific to Pmail but I feel sure that there must be
readers who are experts in RFCs etc relating to email.

I'm working on a website where the user inputs data which is automatically
emailed to me with a copy to the sender. Characters other than ASCII hex20
(space) to hex7F (~) or the £ sign (Unicode 00A3) are rejected before the
email is sent. I've had problems with the £ sign not appearing in the
receiver's (me in this case) email unless I disabled AVG's scan of incoming
messages.

By including:

MIME-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

in the header of the message it works correctly with the AVG scan enabled.

Is that likely to overcome the problem with the vast majority of email clients?

Bob

[quote user="BobKellock"]My question is not specific to Pmail but I feel sure that there must be
readers who are experts in RFCs etc relating to email.

I'm working on a website where the user inputs data which is automatically
emailed to me with a copy to the sender. Characters other than ASCII hex20
(space) to hex7F (~) or the £ sign (Unicode 00A3) are rejected before the
email is sent. I've had problems with the £ sign not appearing in the
receiver's (me in this case) email unless I disabled AVG's scan of incoming
messages.

By including:

MIME-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

in the header of the message it works correctly with the AVG scan enabled.

Is that likely to overcome the problem with the vast majority of email clients?

Bob
[/quote]

The most general solution would be using UTF-8 encoding because (almost) all languages can be encoded using UTF-8. For details you may check out this Unicode page. If you need to evaluate character codes you may also find this converter page useful.