I think JMS has a point. The actual ground-out area isn't all that much darker than the rest of the rock -- certainly it isn't red-brown like the ring of dust around the ground surface.

Looking at the colorized MI image posted above, it looks like there are both white and dark portions of the matrix, in addition to the spherules. If you grind up that marbled white-dark material and throw it out as a well mixed dust, it should probably darken the areas where it lands.

What's odd, though, is that only the dust from the grinding looks very red. I guess the dark splotches may be a little red-grey (vs the blue-grey spheres) but it still doesn't seem to match the dust on the rock face and on the RAT assembly. Is something changing color when it's ground down to dust?

In fact, isn't that something hematite does -- color can depend on texture or particle size or something?