It's actually 3.4 mm and I don't point this out because it's my nature to be a nit picker but rather because I'm guessing that if they could have made it 3 they would have. In other words I think they are probably up against the stops in getting to 3.4 and we are all aware of some of the shortcomings in that design. I think it would be possible to have shorter effective focal lengths but that such a lens would probably be appreciably more expensive.

Now that I'm at work with the Edmund catalogue handy I can answer my own question. Most of the lenses sold by Edmund are C or CS mount with back focus distances of, respectively, 17.5 and 12.5 mm. The XL2 has a back focus distance of 36 mm and thus wouldn't be able to use these lenses without an optical adapter i.e. not a simple spacer like the ones we use with F (back focal distance 46.5 mm) mount lenses.

I don't know anything about lens design. Could a fiber optic, like what they use for polaroind backs, work to transfer the image the extra distance? What type of optical solution would be used, any idea? Most of my books are still packed in boxes, so I can't get at them.

I'm no Carl Zeiss myself but I'd guess they would need to add an element to form a real image and then a relay lens to focus that image onto the CCD's. I suppose a fiber bundle would work. I keep coming back to the same thought: everyone is clamoring for a wider wide angle lens for the XL2 so that if it were possible to make one within reasonable cost constraints Canon would do it. They are primarily a lens company. As it is the best they can come up with is the 3:1 "Zoom" which isn't really a zoom lens but rather a "varifocal" in the sense that is has to be refocused when ever the focal length is changed. This is a tough problem.