Here's a program that will do it for real.
"To use WebReaper, simply enter a starting URL, and hit the Go button. The program will then download the page at that URL, parsing the HTML as it goes, looking for links to other pages and objects. It will then extract this list of sub-links and download them. This process continues recursively until either no more links fulfil WebReaper's filter criteria or your hard disk becomes full - which ever happens first!"
http://www.simtel.net/pub/pd/54400.html