Walt,
Your correct about bluing being iron oxide, but not about the natural color. There are actually different forms of iron oxide molecule, ferric oxide, ferrous oxide, and ferrosoferric oxide. The differences are the number of iron and oxygen atoms in the molecule, and happen because iron atoms can be ionized by losing either two (++) or three (+++) electrons. The two forms do not combine into oxide molecules with the same numbers of oxygen atoms, because oxygen always ionizes to have two extra electrons (––), or at least that is so within the common scheme of things.
Red rust is ferric oxide, or hematite: Fe2O3. It is made up from the triple charged iron atoms combining with oxygen so the net numbers of negative and positive charges balance.
Hydrated ferric oxide, ochre, is a yellowish brown, and is Fe2O3·H2O
Greenish-Black iron oxide or wüstite is ferrous oxide, FeO.
Bluish-Black iron oxide is magnetite or ferrosoferric oxide, Fe3O4. It is an equal combination of the ferric and ferrous oxide molecules. It is what is on blued guns.
In the old days, plum brown was achieved by rusting and carding guns and saturating the red rust with oil to keep it from migrating into the surface. It was discovered you could convert red rust to magnetite with boiling water. The magnetite form does not migrate even when not oiled, and so it became a favored method of prepping the gun surface.
If you use molten oxidizing salts to blue a gun (hot bluing), temperature is critical, as George said. Overheating often gives red or purple results. It affects the surface iron ionization and with that, what the oxygen is combining with and the resulting balance of ferric and ferrous oxide molecules. The color has to do with that balance; just how complete the formation of magnetite is? The addition of nickel salts and other additives can make the finish blacker. As to changing color, though, I've got my great granddad's Smith $ Wesson, still as blue as it was in the late 1800's. I can't say I've ever noticed an actual color change in anything I own, though the color and intensity of incident light and whether or not the bluing is oiled and what with, can affect that.
Iron oxides are normally quite stable. Magnetite is mined and as a mineral has been stable in the ground for billions of years. Some specific influence would be required to alter it.