According to the documentation the -D option makes httrack 
crawl only to the child subdirectories of the URL 
specified. So:
httrack <http://www.shoesizes.com/bob/> -O /tmp/shoesizes -D
would download only www.shoesizes.com/bob/* stuff but not 
www.shoesizes.com/joe/* stuff. However, I tried to 
download www.geocities.com/psylockecentral with that very 
same options (tried both Winhttrack and httrack, version 
3.30) but it kept trying to mirror other sites in the 
geocities.com domain (referenced by psylockecentral links 
page), like www.geocities.com\lumamistic for example. Any 
idea what's wrong here?