vignetting is darkening, not bluring. I think the out-of-focus has already been answered, but yes a poorly seated adapter could do the same thing, though I think it is less likely (seating the adapter is orders of magnitude easier than aligning yourself perfectly parallel in my experience).

As for darkening, the top is @ ~86%, middle @ ~80% and bottom at ~65%, a light falloff consistent with a camera hot-shoe mounted flash, not vignetting, and the corners are not significantly darker than their respective middles. If he is vignetting with that setup (and that aperture) on DX I'd be shocked.

??p
If you do not see light fail of on this shot then my monitor is failing.
The sharp part of this image is lower left corner to lower middle left, as you go further from this area it is more and more blurred. Such an uneven blur could hardly be caused by diffraction (unless your diaphragm blades are seriously broken) but only if your lens is not aligned and parallel to the subject so some parts of the image are out of focus because DOF was not enough to cover it all. You say you used f/16 but not how close you were, anyway this could be caused by either camera not being parallel to the subject or reverse mount adapter is not perfectly aligned and is causing unsharpness and distortions especially visible in the lower right part.
Either that or my monitor is completely broken already.

??t
Vignetting as in darkening of corners? This image shows no sign of vignetting.

alphanikonrex: "But what about the bottom right compared to the bottom middle? How come they look so different?"
You're out of plane both ways. More top to bottom, but also some left to right.

It is called vignetting and it affects corners more then the middle …

Diffraction. Squint your eyes. Diffraction.

Yup. Soap's probably right, as usual :^)

But what about the bottom right compared to the bottom middle? How come they look so different?

Thanks for the help everyone :^)

I think Soap's analysis is right on.

I assumed you want a reproduction ratio of 1:1 or greater. Reversing the lens gives a little more working distance.

F16 is way past the point where diffraction has kicked in.

EDIT:
I'm not saying it is or it isn't softer than you should expect and I understand why you want to increase DoF at cost when doing macro, but at this f-stop point you're not testing the lens, rather physics outside its control.

Question 2 is - how sure are you that the subject is perfectly parallel to the sensor plane? At the focal length you're using minute shifts away from a perfectly coplanar situation will appear as softness, and be noticed first at the edges.

AND judging by the fact you are soft up top and sharper as you go down the image I suspect this is a problem with the test shot provided.

EDIT 2:
The closer I look the more sure I am that you are not seeing softness as a result of the lens, but rather a subject tilted in relation to the focal plane.

corners are definitely too soft

If you "go cheap" on a macro lens then you really can't complain about the "cheap" results can you?

if you really want good marco shots with out breaking the bank, then get a cheeps focusing bellows on ebay. I've gotten great shots with mine using a 50mm F/1.8.

Turning the lens backwards is just a trick.

Try a true macro such as a 55mm or 60mm reversed. It should give better results. If you don't have one, you can buy a 55mm F3.5 or F2.8 manual focus on ebay for a good price. I have a couple of old 55mm F3.5 lenses that I use for just this purpose. Check out Bjorn Rorslett site for his lens reviews.

Actually, quite a few times I have part of my subject in one of the corners, and it looks really strange. I don't like it at all.

You know Alpha in real Macro shots of insects and other tiny things I don't think that the corners would be a problem anyway

Pete

I use a 50mm ƒ/1.2 mounted in reverse on the camera for macro. I've got a suspicion that the corners of the images are no good, so I took a test shot:

Full Size: http://homepage.mac.com/adbilge/corner2.jpg

Ignore the ƒ/# in the metadata, it's actually ƒ/16.

So do the corners look bad? Any idea why that might be? After all, isn't this supposed to be a fabulous lens?