It may have to do with how frequently those documents are updated and how it is done. If you have to maintain an index of billions of documents one of the things you do is look at how frequently they change. The ones that change every day get much more attention than the ones that change only occasionally, but of course the behavior is somewhat random because it has to pay a visit or two in order to figure out how to deal with a particular document. It also of course is demand driven, based on user queries. Finally, I don’t know much specifically about Google but some search engines are smarter then others about following forwarding links, and it may depend on how the page was set up to forward. As a rule if you set the page to forward the end user’s browser automatically to the new location a search engine will likely follow along too, but if you don’t it won’t.

None of that is particularly helpful unless you have access to the old pages so you can set them up for browser forwarding, or can convince the current owners of the domain to put browser forwarding into place. I doubt there is any way for a few documents among billions to get Google-the-company’s manual attention directly, but again I don’t know much about Google specifically.

Thanks for the thoughts. The forwarding was effective enough that google caches the target page rather than the forwarding page. Google just isn’t giving an adequate entry for the target page even though the latter has been out there over a year, it’s the one almost all third-party links, directories etc. point to, and in any event until recently it was the page google presented most prominently (rather than the forwarding page).

So the short of it is that it’s a mystery. Maybe google will answer my email to them. They say they respond to such things.