One thing we must consider when it comes to filtering stained negatives is that the filter's color is uniform, while the negatives color saturation (if that is the right term) varies with the image silver. If we had a blue filter in a dichroic head, as some do, We might see that as the filter strength is increased, the stain image image below some level is more blue and above it is more green. This is not a simple thing to analyze, because the paper local contrast should be higher for the part of the image for the image that lies below the blue level of the filter and lowr for the part above. Am I making myself perfectly unclear?
The magenta filter is minus green and the yellow is minus blue, of course. That doesn't make it any easier for me, but maybe it does for you guys.
Jay, all you have to do to check me out is the experiment I described. Bleach out the silver leaving the stain image. Print the thing on graded paper. I did that once for the public when I wrote the article "More Pyrotechnics" a few years ago.