Answers are grounded in this paper and cite their sources.
In an earlier paper the variation of melt fraction with temperature near the solidus of a rock with the composition of a garnet peridotite was parameterized as a continuous function of temperature. A number of people have suggested that eutectic melting is a better description of the behaviour near the solidus. Therefore a 'eutectic' parameterization is used here, and requires a finite melt fraction to be produced at the solidus before the temperature can increase further. The resulting changes in melt volumes and compositions are small, and are of similar magnitude to the likely errors in the earlier calculations.
No takes yet. Share an insight, caveat, or question.
McKenzie et al. (1990) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: