For a long time I thought the problem was the ink. The prints were coming out with a band of lighter coverage running roughly parallel to one edge — not white, just thinner, less saturated, a ghost of itself. I changed the ink. I changed the paper. I mixed the ink differently, rolled it longer on the glass, cut the amount and rolled again. The band stayed.
Then I thought it was the plate. I re-cut. I checked for debris. I lit it from the side and looked for warping. Nothing I could see explained what the print was showing me.
It took an embarrassingly long time to look at the brayer.
The brayer was fine — no visible flat spot, no cracking, rubber still supple. I'd cleaned it properly after every session. By every reasonable measure, it was not the culprit. And yet: it was the culprit.
What's actually happening
The issue was pressure — not across the roller, but along the handle. Specifically, I was applying more pressure at the near end of each pass than the far end, tipping the brayer very slightly forward as I rolled away from my body. The rubber contacted the plate unevenly, inking it unevenly, and the print faithfully recorded every bit of that.
The reason it took so long to diagnose is that uneven pressure of this kind doesn't show on the plate in any useful way. You look at an inked plate and you're looking at a surface that reflects light; the difference between a slightly thin layer and a correct one is nearly invisible to the eye. The print is a more honest record of what happened than the plate is.
What made it worse: the tipping wasn't consistent. Some passes were level, some weren't. So the effect varied print to print, which sent me looking for variables I was changing between prints — ink temperature, humidity, paper preparation — rather than something I was doing wrong consistently but not identically.
Obvious in retrospect. Maddening at the time.
What didn't fix it, and what did
Slowing down didn't fix it. Paying more deliberate attention to the motion, in the sense of watching myself do it, didn't fix it either — watching introduces its own changes and tells you what you want to see.
What helped first was changing my stance. I'd been reaching across the plate, which meant my arm was never at the same height relative to my body on the near side versus the far side of the pass. Moving the work so the plate was directly in front of me, not at an angle, levelled the motion out considerably. Not completely, but enough that the band narrowed.
What actually fixed it was changing grip. I'd been holding the brayer with the thumb on one side and fingers wrapping the top — a natural enough grip, but one that lets you lever the handle without meaning to. Holding it more like a rolling pin, with both ends of the axle supported rather than just the handle, eliminated most of the tip. The pressure distributes through the geometry of the tool rather than through the particular tension of one hand.
There's nothing clever about this. It's in any decent printmaking manual. I just had to fail for long enough to be ready to read it carefully.
One thing worth noting: a harder rubber brayer is less forgiving of this error, because it doesn't deform to compensate at all. A softer roller will spread slightly under uneven pressure and partially self-correct. But softer rollers have their own problems — they can compress into plate recesses and pick up ink you didn't want distributed to the whole surface. The fix to one thing is often the setup for the next thing.
The print is still the most honest account of what you did. The plate looks clean, the ink looks right, the studio looks organised. Then you pull the print and it says: near edge, again. At a certain point you have to believe it.