Dear Friends,

    I hope I am not too late to weigh in on the Al-silicate triple point
discussion, especially now that you are onto vapor-absent melting.  Charlie
Guidotti sent me his latest correspondence on the triple point discussion.
Then he sent me all the other correspondence, so I decided to join the
Geo-Metamorphism Mailbase, which I should have done long ago.  Tripe point
is a good name for this rarely observed phenomenon.
    I agree with virtually all that has been said, but I think I can expand
a little on the information regarding the Picuris and Mt. Moosilauke
localities.  I would like to begin by sharing my notion of how and how well
rocks equilibrate to P-T conditions.  I believe that new phases BEGIN to
form in response to crossing P-T boundaries with relatively little
overstepping.  The extent of necessary overstepping should be related to:
(1) the available free energy, that is, reactions with high delta S, so that
delta G increases rapidly with T away from the P-T boundary, will proceed
the most rapidly;  (2) ease of nucleation of the new phase(s).  As stated by
some of you, sillimanite which formed from staurolite breakdown should
require less overstepping than sillimanite formed from andalusite.  In my
observations, the final disappearance of the unstable phase(s) often does
not occur.  This means that we must depend more on the nature of the new
phase and rely heavily on textural interpretation to get the right sequence
of events.
    I am familiar with the Picuris Mountains rocks.  In these rocks there
are no Mn-rich andalusites as there are in neighboring ranges.  I now
suspect, as you guessed, Charlie, that the Picuris rocks are polymetamorphic
as first suggested by Daniel, Thompson and Grambling (GSA National Meeting
abstracts, 1992, p. A264.  I can FAX it to you if you wish).  The earlier
event was kyanite to sillimanite in the Al-rich quartzites.  This took place
at about 4.2 kbar, 525 deg C. so that staurolite-garnet-biotite-muscovite-
graphite occurred without Al silicate in the nearby normal schists, but
chloritoid-Al silicate-hematite-ilmenite occurred without staurolite in the
neighboring quartzites.  Daniel et al. suggest that the widespread later
event was an isothermal decompression related to crustal extension.  It
produced mainly andalusite which was locally developed especially near
contacts between units at about 3.7 kbar, 525 deg C.  The P-T estimates are
mine for the Picuris.  Dugald, this would mean that even the Al-Mn schists
you mentioned were still not very near the triple point, but were at higher
T and perhaps lower P during the late stage metamorphism.  (These Al-Mn
schists would not be very good candidates for geothermometry because of
their unusual composition and their oxidized state).  In my earlier work I
believe I misinterpreted the Al silicate textures.  The most telling texture
is andalusite surrounding well-crystallized sillimanite with no sillimanite
outside the andalusite, and no fibrolite in these andalusite-bearing rocks.
    Regarding Mt. Moosilauke, I am pleased that Frank Spear has noted the
thrust fault between the Mt. Moosilauke schists and the Black Mountain area
aluminous quartzites.  I have suspected that to be the case.  The Mt.
Moosilauke schists have reached a high enough T that the staurolite in the
schists is breaking down to Al silicate-bearing assemblages.  In these rocks
5 have only sillimanite and 2 have both sillimanite and andalusite.  Hodges
and Spear (1982) don't try to guess which came first, but they do state that
the andalusite is highly sievy (unlike the Picuris andalusite) and that much
of the sillimanite is fibrolitic, presumably outside the andalusite.  I
would interpret that evidence to suggest that the rocks passed from the
andalusite field into the sillimanite field at fairly high T, and
sillimanite was the stable Al silicate at the culmination of metamorphism.
Sillimanite and andalusite both formed from staurolite breakdown and
sillimanite was also forming from andalusite.  Using my new G-B thermometer
and my still newer GASP geobarometer I get a T of 553-591 deg C for Hodges
and Spear's 7 rocks, and a P of 4.3-4.9 kbar for the 3 rocks which have
enough Ca in the garnet and plagioclase to be usable.  I suspect the Black
Mountain aluminous quartzitic rocks, which have kyanite, chloritoid and
staurolite, formed at about 530 deg C and 5 kbar.  The fluid may not have
been as pure water as in the Picuris quartzites, or the T may have been a
little higher.  In both cases the aluminous quartzites have a very high
Fe/Mg ratio.  As with the Picuris rocks, these localities must not have been
at triple point conditions.
    I'd appreciate hearing your comments, if you are not already bored silly
with this tripe.

Take care,     Mike



Mike Holdaway

Department of Geological Sciences
Southern Methodist University
Dallas, Texas 75275

Phone 214-768-2751
FAX 214-768-2701
Home 214-348-6579





%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%