Thanks for all the suggestions to solve my parsing problem as posted
Thanks to Lou especially for solving the problem.
It was a mixed content case. Whereas the content model of <note> only
allows either pcdata or a sebsequent series of para's, I had inserted a
whitespace somewhere, so the parser assumed that the note was terminated
and had troubles parsing on.
(apart from a clumsy absence of a <P> in the last line of the example.
The version which does parse:
<P ID="td1.1.001" N="1.001">
Electronic Streuvels Project