Code:
However, I'm not sure this will do the trick. The way you're suggesting will get only the HTML, but I'd also need to download all the images (JPGs & C) linked in the document.
Any further hint?
I don't a lot of time to think about this today, but here's one thought -- adding one level of complexity.
Navigate to the page on that website, and then go to "Save Page As...". In the "Save As" dialog box, be sure to choose "Web Page, Complete" as the format to save it as. That will save the file as an .HTML file (say, science.html) and will create an folder (science_files) where it will put all the images. The only problem is that the page you just saved has all the menus and other nonsense.
So NOW open the science.html file you just saved in Firefox, highlight the part you want and view its code and follow the procedure I outlined above. The image links will link the ones on your harddrive rather than the remove site. So long as you save the .html file in the same folder as you saved the original, I think calibre (or whatever) should be able to find them when you convert to .mobi.
I'll have to test that later, however.
P.S. Didn't see DDHarriman's post. I'm a big fan of soPDF, but I'm not sure that's the way to go here. Try both methods and see what you prefer.