Dugald et al.,
OK. I believe that, in the majority of cases, granitic melts are
produced at temperatures in excess of the solidi of their parent
rocks. So, yes, a phase is commonly exhausted during such partial
melting reactions. In the fluid-present case (with aH2O externally
controlled) T would then be governed by the power of the heat source,
thermal diffusivities, the heats of dissolution reactions
(progressive mineral dissolution in the melt) and any advective heat
loss. We could then have melt at any imposed fluid composition that
allowed us to be above the solidus. In contrast, the fluid-absent
case nails aH2O to a fixed value for any given T, at constant P. The
H2O budget is internally fixed. This also fixes the amount of melt in
the rocks. There's the difference.
Cheers,
JC
|