Thilo,
I think that I understand what you are saying. My purposes in measuring the Blue and Magenta/Green channel and comparing them to the visual reading is not to measure the effects of a color addition brought about by the stain. It is instead to measure the effects of that stain on actual negative density as that would impact upon the light spectrum that is exposing a paper.

In the case of Azo, which exposes primarily to blue and upper band UV, the effects of stain upon the density presented to the passage of that quality of light must be a valid comparator. At least I have found it to be true in my experience. The addition of a value of .30 pyro stain density onto a visual density of 1.10 would be a sizable effect in the case of Azo. In fact if I were relying on a visual channel reading of 1.60 to achieve a 1.30 net contrast range (subtracting .30 as the actual low density). I would be dismayed to find that the negatives were in fact printing with a density range of 1.60 and be of too extreme a contrast to print on Azo. When one is shooting 8X10 and 12X20 negatives the cost of these errors would be regretful.

Now on to the matter of green/magenta on VC materials. I think that we can agree that green or magenta are the colors that affect the higher contrast emulsion on VC materials. It would appear to me that any effect of staining density upon the transmission of those colors would be affective of that high contrast emulsion. While my tests indicate that Pyrocat and ABC pyro are not as influential on the magenta/green channel as upon the blue channel, there is indication that the pyro stain does in fact affect density as it affects the transmission of the magenta/green spectrum.

I would appreciate any further thoughts that you have on this matter.