I used a search engine emulator which suggested that a local link from the index page index.html was interpreted as an absolute link ie. http // Welcome.html rather than inserting the domain. I've also received 404 errors which suggest this might be happening quite often.
(I wondered that this might be the simulator but using another page it did then correctly intepret the links.)
Does the server playing any part in the interpretation of a domain becoming domain+index.html and those links internal to that file being misinterpreted as absolute?
This from an effort to fathom Google's not indexing my site fully, having me wonder if the host of my site has changed something, expecting that Google will settle down I'm not concerned and I've added full links to my index files now to avoid the 'problem' (if there is one).
I'm guessing you are using the searchengineworld spider simulator? If it is, you are using relative urls on your site, and you've omitted the trailing slash at the end of the url (http //www.widget.com/) when entering it into sim spider, and this is why the links don't show up correctly.