Googlebot does spider site, but...

only a few documents show up in the index

[Martin Dunst]

Hello,

I've got a site that's been online since early 2003-11.
It's got an dmoz link, a Google Directory link and a few more.
Recently, its PR rose from 2 to 4.

The problem is that the site has 50+ documents, still a site:www.example.com search suggests that only 14-16 documents are indexed.
The serverlogs say that GoogleBot spiders the page. Moreover, I really think I've followed Google's quality guidelines:
- I don't use frames
- 95%+ of all documents are valid HTML 4.01
- I don't use ANY URL parameters
- no sessions
- no Javascript-links
- etc.

Content is generated dynamically and a set of RewriteRules assure that the client gets nice, readable ".html"-URLs.
The site has not been SEOed at all, it just uses valid markup and CSS and each document has a meaningful title / h1.

The day before yesterday, I thought there was a ray of light: The site:www.example.com search said that all (50+) documents were indexed, while most of them had a "Supplemental Result" tag.
"Finally", I thought.
Yesterday, it switched back to the old status: 16 documents in the index, only one of them has a description.

How can I possibly find out what Google is so picky about?

regards
Martin