Even still, you can see from the uneven sanding that the surface is irregular. The lighter patches on the edges are contacting the sandpaper. The darker section in the center is not.
To fix this, I'm trying something new. I'm applying one coat of shellac to the low areas in an attempt to raise them up a little. I didn't document it, but I think it worked.
After the carver has done its work, the "perfect level" of the board doesn't matter. (The press can handle a little inconsistency, I think.) I noticed that it was slightly off level, so I'm using a long homemade wood block to sand down the high spots without messing up the level too much.
I pulled a few proofs and I'm not satisfied. Yes, it's better than yesterday, but if I'm committing hundreds of dollars of paper and renting studio time with an instructor to help pull this edition, I am going to spend the extra time to make sure the plate is as perfect as possible.
Here's the best proof from the run. Notice the relative type weight. Again, "May" is quite a bit heavier than "July". In the future, I will not use much type or I'll employ some other trick to mask the fact that it's really tough to get my plates within the 0.001" tolerance that's required for perfect type.