epmaniac - Forgive me for putting it this way, but are you asking theoretical questions about freely given editorial "reference" links, or are they questions about link buying choices? I ask because it sounds more like the latter, and my answer addresses some of those considerations.
There are a lot of signals that Google can pick up on, not the least of which is the quality of the other sites included among the other "references", and the relevance of the linking site. There's generally no good reason for a site to have 100 reference links on its home page, particularly when all the links are down at the bottom. At the least, it doesn't sound like intelligent site design.
ROS (run-of-site) links send their own signals, as do links buried in navigation and links with just the right repetitive anchor text. These usually indicating a special relationship between sites. 10,000 ROS links from a single domain doesn't give you much more benefit, if any, than one link from that same domain... but IMO the 10,000 links are likely to raise a flag for Google. Keyword anchor text will generally color the flag red.
a) one link from a PR 9 page which has 100+ external outbound links
b) one link from a PR 7 page which has ZERO or ONE external outbound link
In addition to the points Aristotle raises about the number if internal links on the page, I'd add a question about the pattern of outbounds from the site in general. Is the link from the PR7 page from the same domain as the PR9 site that has the 100+ external outbounds from home, some of which might be paid? Or is it from a clean site with no iffy links that is obviously very selective about what sites it links to?
And what kind of backlinks do these linking sites themselves have? Are they artificially boosting up their TBPR to make themselves attractive to link buyers? There's a test that martinibuster recommends which is very simple... Check to see if the linking site's backlinks come from SEO-type sites. If they do, I would avoid getting links from those sites.
