At 12:04 PM 25/09/2001 +0100, Dave Waters wrote:
>... The assemblage Grt+Bt+Ms+Mt+Qtz will buffer fO2
>effectively, i.e. retard changes and hold it in a narrow range of
>values, because of the preponderant mass of condensed phases compared to
>the amount of O2 (or H2) that can normally be added or removed in a
>volatile phase. Except at extremely high fluid:rock ratio, only trivial
>changes in the annite and almandine contents of biotite and garnet would
>be needed to restore equilibrium, at an imperceptibly changed fO2.

But those trivial changes would be difficult to achieve for kinetic
reasons.  Equilibration that requires intracrystalline diffusion is
kinetically more difficult than equilibration that requires only
dissolution and growth of minerals through H2O-assisted grain-boundary
diffusion. Medium-grade Grt commonly has concentric compositional zoning -
proof of disequilibrium in respect to intracrystalline diffusion. Although
medium-grade Bt generally lacks concentric zoning, there is independent
evidence that it fails to equilibrate in divariant KFMASH assemblages.
According to the phase diagrams, XFe in Bt ought to vary smoothly across
each prograde zone and inflect at each of the isograds, but in fact it
tends to be constant across each prograde zone and change abruptly at each
isograd (eg see Chinner 1965 Min Mag 34, 132-143; Mohr & Newton 1983 Am J
Sci 283, 97-134). This suggests that Bt tends to approach equilibrium
during the univariant KFMASH reaction at each isograd (by means of the
general catalytic dissolution and regeneration of matrix minerals that
facilitates the growth of porphyroblastic minerals), while also suggesting
that Bt becomes progressively more metastable in respect to XFe with
increasing distance upgrade from each major isograd (due to the
sluggishness of intracrystalline diffusion).

   Accordingly, I am skeptical that the assemblage Grt+Bt+Mu+Mt+Qtz can
equilibrate fast enough to slide-buffer fO2 in a transient fluid. The
larger the number of additional minerals (Chl, St, Ky, Sil, Crd, Kfs,...)
the better would be its chances of doing so. In my experience low-variance
assemblages are not uncommon, and unfailingly they record maximal
petrological information.

Cheers, Dugald

Dugald M Carmichael                    Phone/V-mail: 613-533-6182
Dept of Geological Sciences and Geological Engineering
Queen's University                              FAX: 613-533-6592
Kingston  ON  K7L3N6             E-mail: [log in to unmask]