It's become common practice to have website URLs point to a directory rather than an HTML file. So rather than "example.com/foo.html", one would go to "example.com/foo", which is a directory that contains an "index.html" file, and the server is configured to return that file when the directory is requested. (Often the directory will also contain any images or other related files that are needed by index.html.) wget seems to only realize this when the link ends in a trailing slash. That is, if the homepage at example.com includes a hyperlink to "example.com/foo/", wget will correctly download example.com/foo/index.html. But if the homepage links to "example.com/foo", wget will not recognize that this is a reference to an index.html file, and it will not be downloaded properly.
Isaac King
