Note that if the two circumstances I have hypothesized are not the case, i.e.
your pdf's lie under the starting url, e.g. 

<http://domain/.../some-directory/page-of-hundreds-links.html/pdf-folder/>

the settings I used in the preceding example should work under the default
"Options>Experts only" values (indeed now the links point resources lying
under the starting url).

So, if you have something to filter you'll put in "Options>Scan rules"
something as -* +*.pdf, and now you'll obtain the automatic download of all
your pdf's, all linked to your starting html page.