This myth is a frequent source of incorrect assumptions about Google. People will often see that a site with fewer incoming links than their own site has a higher PageRank, and assume that PageRank is not based on incoming links.
The fact is that PageRank is based on incoming links, but not just on the number of them. Instead PageRank is based on the value of your incoming links. To find the value of an incoming link look at the PR of the source page, and divide it by the number of links on that page. It's very possible to get a PR of 6 or 7 from only a handful of incoming links if your links are "weighty" enough.
Also remember that for PageRank calculations every page is an island. Google does not calculate PageRank on a site-wide basis -- so internal links between your pages do count. This is very important, as instituting a proper structure for your internal links can drastically improve your rankings.
Similar to Myth #3, people will sometimes look for backwards links to a site on Google and fine none, but if the site does have a PR listed and it is in Google's cache, they know that the toolbar isn't just guessing.
The reason for this is that Google does not list all the links that it knows about, only those that contribute above a certain amount of PageRank. This is especially evident in a brand new site. By default, all pages in Google have a minimum PR. So even a page without any incoming links has a PR value, albeit a small one. If you have a brand new site with 20 or 30 pages, all of which Google has spidered, but you have no incoming links from other sites, then your pages will still have a PageRank resulting from these internal links. As your home page is likely linked to from every page on your site, it might even get a PageRank of up to 1 or 2 from all these little boosts. However, in this situation searching for incoming links will likely yield 0 results.
You can also see this happening on pages that have been around for awhile. For instance, find a page with 0 incoming links listed in Google but that you know has links. Check to see if Google has spidered it by checking its cache, if it is in the cache then Google has spidered it. Then the cache of the page that links to it, to see if Google has spidered that one as well. If both pages are cached but no link is listed then we know Google knows about the link, they just choose not to list it.
Chances are if you look at the linking page with the Google Toolbar installed, you'll notice the page has a very low PR. Furthermore, if you count the number of links on the page, you'll notice it has alot, maybe over 20. So you're dividing a very low PR among over 20 links. Thus each link carries very little weight, so Google doesn't list these links when you search for them. However, Google does count the links, which is why the page you tested will still have a PR showing up even though it has no backwards links.
It's very important to remember how Google lists incoming links. Often, people see their number of incoming links drop, and they think they have lost those links. In reality, the linking page could have lost some weight and consequentially, the links might have dropped below the value threshold that's required in order for links to be listed. Or the linking page could have added more links, causing each link's share of the weight to be lower, and again causing the link to drop below the value threshold. In either case the link is still counted, it just isn't listed.
Why does Google do this? Perhaps the answer has to do with technical limitations. If the average number of links per page is 20 then Google would have to deal with over 60 billion links, which might create an index that was too large to be publicly searchable.