Originally Posted by Stephen Benskin
A one stop filter factor means flare doubles the shadow exposure 4 1/3 stops below the metered exposure. If the luminance range is shorter than the statistically average and the shadow exposure doesn't reach down 4 1/3 stops, then there's nothing there to add exposure to. Where the shorter luminance range shadow falls in this example will not experience a doubling of exposure.
I'm not following. Did you mean to say a one stop flare factor? Let's assume the two exposures I listed are both 4 1/3 stops below metered and that the film density under zero flare conditions would arbitrarily be 0.20 above FB+f in either case. In the first exposure, the intensity of light entering the camera is high, and so some of this light (from the portion of the image circle beyond the borders of the emulsion) might bounce around, eventually bouncing onto the film and adding some additional exposure. At least I think it might. Maybe the actual film density ends up being 0.23 above FB+f. In the second exposure, a much lower intensity of light enters the camera, but the shutter is open for longer. Would we still expect a density of 0.23? Or would we actually get the zero flare density of 0.20 because the light intensity is not high enough to cause significant reflections in the camera - even though the film is exposed for longer.

Perhaps there is no definite answer. Here's another way to think about the same example. You take a picture of a bright, uniform, featureless mid day sky and expose it to produce an expected net film D of 0.20. In the evening you take a picture of a middle grey, uniform, featureless evening sky and expose it to produce a net film D of 0.20 (so obviously this would be a significantly longer exposure than the picture of the mid day sky). Assume some camera flare in the first exposure leads to an actual net film D of 0.23. Would that occur in the second example too, even though the intensity of light entering the camera is lower? I'm thinking there would be less, or zero flare in the second exposure. Not sure.

Keep in mind this question concerns an in-camera film test with as little flare as possible (aside from some degree of lens flare which cannot be totally eliminated). The aim is to use a brightly lit white card for the entire series of exposures. In actuality, the lens is masked to restrict the projected image to within the borders of the emulsion. In other words, there should be no camera flare. But even though I would be using this mask anyway as an extra precaution, I'm wondering if it is necessary. So my example above would be the equivalent of not masking the lens.

Sorry again for repeating myself. I just want to make sure I really understand this.

Regarding the K-Factor discussion in Henry's book, as it happens I re-read that very section today, and I came away just as little understanding as when I first read the book. I wish he had started by explaining what the K-Factor actually is before getting into the ranges, calculations, standards etc. I still don't get what the K-Factor is. The only thing I think I understand is that it is not what Adams said it was. When you read Adams you come away thinking it is just a safety factor built into the meter.