Keijo,
You definitely have very sharp eyes! Nice work with the Photoshop, too!
However, 3 pixels out of 225 is 1.3%, is not unreasonable for a hand-held mechanical affixer.
Also, perhaps this only applies to me, but if they were using scissors and in such a hurry or unconcerned as to miscut it by 40%, I'm very surprised they were able to get such a small variation in width. I know that I can't get that good with scissors even when I'm actually trying. That's why I use a paper cutter. I guess I'm trying to fathom how it could be so badly miscut, yet so nearly parallel! Off by 40%, but within 1.3% width tolerance! I would have thought if they took that much care, it would make more sense that they would have cut it at the perforations?
Anyway, we will simply have to agree to disagree. I've been involve with several non-profit groups doing volume mailings for nearly 2 decades, and I've yet to run across any that use scissors to separate the coils. But perhaps the groups I've worked with are in the minority? (Or maybe they just copy the way I do things?

)
Since neither of us were there when the stamp was applied (or at least I wasn't there

), we can only put up a best guess. The hypothesis of "happy scissors" cannot be ruled out. It does have the advantage of explaining why the width of the cuts are so much smaller than the actual stamp.
Interesting cover. Thanks for sharing the cover, and thanks to everyone who proposed possibilities!

k