I had enough time--the bug was accommodating, up to the point when it realized it wasn't getting paid for this and flew off--to shoot a bracketed series by available light, and to also shoot with the built-in flash. A later test series of available-light shots showed that I should have anticipated some vignetting at 22mm (as well as at 10mm--both improved by stopping down), but the actual flash exposure of the flower showed that the built-in flash was a practical tool even here (at 22mm, but not at 10mm, where pronounced vignetting from the flash compounded the problem). And it helped me capture the bug with good definition (a little tweaking with Levels in Photoshop was needed, but that's not a reflection on the lens itself).

With the 10-22mm lens on the camera, I mounted the EOS 20D onto a Velbon ULTRA LUXi-F tripod for this moonrise over New York City's business district, as seen from Jersey City, New Jersey (f/4.5, 2.5 seconds, ISO 100). Notice the vignetting, which in this case helps frame the scene.

Anyway, as I started to say, flare could be a problem, and the lens shade may have helped. But what really did the trick was the internal baffling and flare-blocking diaphragm. I wanted to see this for myself, so I went out a couple of days sans shade and looked for damaging evidence. But could not find it. Light transmission in pictures without the lens shade was practically on a par with those taken with the shade (although I'd still recommend you spend the bucks on a lens hood, if only to protect the lens). I even pointed the lens at a setting sun. The most I got for my troubles was a tiny bit of flare, and an almost inconsequential flare ghost. Not a haunting experience, I might add.

Finally, I wanted to check a couple of other things, one being lens (curvilinear) distortion. At 22mm, pincushion distortion is barely noticeable. At 10mm, barrel distortion is apparent, but not necessarily to the extent you might expect in practice. Aside from that, I wanted to see how sharp the lens is corner to corner. Some falloff is expected at these wide focal lengths when the center of the image is in focus. And by veiling some detail, vignetting contributed to a perceived lack of definition around the perimeter of the image--but again, not readily discernible with most subjects. By the way, owing to the short throw, moving from one end of the zoom range to the other was quick--and smooth.

Now, before you run out and buy this lens, you might want to read Peter Burian's coming report on several non-OEM lenses that roughly fall within the same 10-22mm zoom range. I myself can't wait. This Canon EF-S 10-22mm USM lens has proven itself to me time and again. I've already reserved space for it in my camera bag.

You can see the AF/MF switch, distance scale, and zoom scale here. Equally important, the rear of the lens is designed to fit only Canon's newer APS-C cameras. The focusing ring, which operates whenever needed, is toward the back, with the wider zoom ring at the front.

Technical Specifications
Focal Length And Maximum Aperture: 10-22mm; f/3.5-4.5
Lens Construction: 13 elements in 10 groups
Diagonal Angle Of View: 107Þ 30 ft - 63Þ 30 ft
Focus Adjustment: Inner focusing system, with focusing cam
Closest Focusing Distance: 0.79 ft
Zoom System: Ring USM
Filter Size: 77mm
Max. Diameter x Length, Weight: 3.3x3.5", 13.6 oz
Street Price: $799

I photographed this neighborhood church with the lens at 10mm, for this dramatic view.

For more information, contact Canon U.S.A., Inc., One Canon Plaza, Lake Success, NY 11042; (800) 652-2666, (516) 328-5000; www.canonusa.com.