Another thing that all SEM pictures of highly polished edges always strikingly display, is something they don't show... A sawtooth pattern at the very edge. It's never there, and still, it should be there, at least if we presume that honing is only abrasion. If that was true, the scratches, left by the hone, would end in a jagged line. And yet, they do not. We always see a fluent line. According to Verhoeven, abrasion is not the only contibuting factor to honing. He quotes two other factors: debris deposit and plastic flow. The first does exactly what it says: honed away steel piles up, inside scratches and on top of the original steel. It happens mostly in the honing direction, but not exclusively. The second is a bit more difficult to understand: it's they same process that makes steel bendable and rollable. This is called "plastic deformation"
Elastic/Plastic Deformation If the same principles only occuring on the surface of a steel object, they are called "plastic flow".
Burs exist of steel that was shaped by one or both of those two non-abrasive factors. They have to, because abrasion cannot allow something to grow. Debris deposit and plastic flow can.
I don't think such steel looks different. If you look at the burred edge of a cabinet scraper under magnification, you can't see any difference between the bur and the original steel, other than the bur being a part that's bend over. For those that not know about "sharpening" scrapers: the scraper is first honed to a perfect 90° corner, a tool called a "burnisher" is used to push the 90° corner into a bur shape. It requires surprisingly little force.
My point is that the final part of the edge is much more formed by plastic flow than by abrasion.