Metamorphic petrology: Barrovian facies, index minerals, and the interpretation of pressure-temperature paths
Anchor (Master): Barrow 1893 Q. J. Geol. Soc. London 49:330; Eskola 1915 Bull. Comm. geol. Finlande 38; Miyashiro 1961 Am. J. Sci. 259:209; England-Richardson 1977; Spear 1993 Metamorphic Phase Equilibria and Pressure-Temperature-Time Paths; Holdaway 1971 Am. J. Sci. 269:97; Thompson 1976
Intuition Beginner
Deep in the Earth, heat and pressure cook rocks without melting them. This is metamorphism. Limestone becomes marble; shale becomes slate, then schist, then gneiss. The minerals inside rearrange themselves into new, more stable configurations as temperature and pressure rise.
In 1893, a geologist named George Barrow walked across the Scottish Highlands mapping the rocks. He noticed that the minerals in them changed in a consistent sequence as he moved from north to south. First he found chlorite, then biotite, then garnet, then staurolite, then kyanite, then sillimanite. Each mineral forms only within a specific range of temperature and pressure, like a thermometer and a depth gauge frozen into the rock.
Barrow had discovered index minerals. By reading which index mineral is present, you can tell how hot and how deep the rock was when it metamorphosed. This tells geologists about tectonic processes: mountains rising, continents colliding, ocean crust diving back into the mantle. Rocks at the surface today that carry sillimanite were once buried 25 kilometers deep.
Visual Beginner
| Zone | Index mineral | Approximate temperature | What the rock tells us |
|---|---|---|---|
| Chlorite | Chlorite | 200 to 350 C | Low grade, shallow burial |
| Biotite | Biotite | 350 to 450 C | Heating begins |
| Garnet | Garnet | 450 to 500 C | Medium grade |
| Staurolite | Staurolite | 500 to 600 C | Moderate pressure |
| Kyanite | Kyanite | 550 to 650 C | High pressure, deep burial |
| Sillimanite | Sillimanite | 650 to 750 C | High grade, hottest zone |
Walking south across the Highlands is like walking down into a buried mountain root, then back up: the grade rises from chlorite to sillimanite as the rocks get hotter, recording the heat of collision.
Worked example Beginner
Barrow mapped his original sequence in Glen Clova, Angus, in the eastern Scottish Highlands. Starting in the chlorite zone to the north and walking south, the rocks pass through each successive index mineral in order.
Step 1. The chlorite zone rocks record temperatures of about 300 C, corresponding to burial under a few kilometers of sediment. At this grade, the original clay minerals of the shale have just begun to recrystallize.
Step 2. The garnet zone marks about 470 C, the temperature at which garnet becomes stable in an aluminum-rich rock. The pressure at this point was about 5 kilobars, corresponding to roughly 15 kilometers of burial.
Step 3. The kyanite to sillimanite transition marks the highest grade. The temperature was about 600 C and the pressure about 6 to 8 kilobars, which corresponds to 20 to 25 kilometers of depth. These rocks were buried in the root of the Caledonian mountains during the Grampian Orogeny about 470 million years ago, when Baltica collided with Laurentia.
What this tells us: the Scottish Highlands expose a frozen cross-section through the root of an ancient mountain belt. Rocks now visible at the surface were once deep in the crust. Uplift and erosion over hundreds of millions of years stripped away the overlying 25 kilometers of rock.
Check your understanding Beginner
Formal definition Intermediate+
Metamorphism is the solid-state recrystallization of a rock under conditions of elevated temperature and pressure distinct from those of its formation, without wholesale melting. The thermodynamic state of a metamorphic rock at equilibrium is specified by its pressure, temperature, and bulk chemical composition; the mineral assemblage that crystallizes is the one that minimizes Gibbs free energy at those conditions.
Definition (metamorphic facies, Eskola 1915 [Eskola 1915]). Two rocks of identical bulk chemical composition metamorphosed at the same pressure and temperature develop the same mineral assemblage, regardless of their protolith or geographic location. A metamorphic facies is the set of pressure-temperature conditions over which a characteristic mineral assemblage is stable, independent of bulk composition.
Definition (index mineral and isograd). An index mineral is a mineral whose first appearance in a regional metamorphic terrain marks a specific pressure-temperature condition. The map line joining locations of the first appearance of an index mineral is an isograd. In a Barrovian sequence (medium pressure, medium to high temperature) the index minerals appear in the order chlorite, biotite, garnet, staurolite, kyanite, sillimanite [Barrow 1893].
Definition (metamorphic grade). The grade of a metamorphic rock is a qualitative measure of the temperature it attained: low grade corresponds to chlorite-zone conditions, high grade to sillimanite-zone conditions. Grade increases monotonically with temperature along a Barrovian geothermal gradient.
Pressure-temperature regimes
Three regional metamorphic regimes are distinguished by pressure-temperature ratio [Miyashiro 1961]:
- Barrovian (medium-, medium-): typical geothermal gradient 20 to 30 C per kilometer, characteristic of crustal thickening at continental collision zones. Index sequence: chlorite to sillimanite, with kyanite as the high-pressure aluminum silicate polymorph.
- Buchan (low-, high-): gradient above 40 C per kilometer, characteristic of contact metamorphism around shallow intrusions. Index sequence: chlorite, andalusite, sillimanite, with cordierite; andalusite replaces kyanite as the diagnostic polymorph.
- High-pressure (high-, low-): gradient below 15 C per kilimeter, characteristic of subduction zones. Diagnostic assemblages: blueschist (glaucophane, lawsonite) and eclogite (omphacite plus pyrope-rich garnet).
Counterexamples to common slips Intermediate+
- "Higher grade means deeper burial." Not in general. Grade is set by temperature, not pressure alone. A Buchan sillimanite-bearing contact-metamorphic aureole formed at only 3 to 4 kilobars (10 to 12 kilometers) but at 650 C, while a subduction-zone blueschist formed at 15 to 20 kilobars (50 to 65 kilometers) but at only 350 to 450 C. The pressure-temperature ratio, not pressure alone, distinguishes regimes.
- "The aluminum silicate polymorph identifies the tectonic setting." The three polymorphs of (kyanite, sillimanite, andalusite) share the same chemical composition but differ in crystal structure, and each has its own stability field. Andalusite indicates low pressure, kyanite indicates medium to high pressure, and sillimanite indicates high temperature. A rock can contain two polymorphs if it crossed an isograd during its P-T history.
- "Metamorphism requires melting." It does not. Metamorphism is solid-state recrystallization. The boundary with igneous processes (anatexis) is marked by the wet solidus of the rock, typically 650 to 750 C for continental crust at depth; sillimanite-grade Barrovian rocks approach but generally remain below this boundary.
- "Facies and grade are synonyms." A facies is a pressure-temperature window with a diagnostic assemblage (greenschist facies, amphibolite facies, granulite facies, blueschist facies, eclogite facies). Grade is a one-dimensional temperature ranking along a single geothermal gradient. The greenschist facies is low grade; the granulite facies is high grade; the blueschist facies is high pressure but low grade.
Key result: Barrovian facies and the interpretation of metamorphic grade Intermediate+
Theorem (the index-mineral isograd sequence encodes the regional geothermal gradient). Let a regionally metamorphosed terrain of uniform bulk composition be exposed at the surface. Then the map ordering of isograds, from the chlorite isograd to the sillimanite isograd, is the projection onto the present surface of a monotone increase in paleo-temperature along the Barrovian geothermal gradient, and the spacing of the isograds in the field is determined by the slope of the pressure-temperature path. In particular, if the index-mineral sequence is chlorite, biotite, garnet, staurolite, kyanite, sillimanite, then the regional geothermal gradient during metamorphism was medium-pressure, in the range 20 to 30 C per kilometer, characteristic of crustal thickening [Barrow 1893; Eskola 1915].
Derivation. Consider a shale of fixed bulk composition undergoing progressive metamorphism along a pressure-temperature path parameterized by time or, equivalently, by depth during burial. The Gibbs phase rule for a system of chemical components and coexisting phases gives the variance
where the two additional degrees of freedom are pressure and temperature. For the six major components of a pelitic (aluminum-rich, potassium-bearing) shale, namely , , , , , , the variance with three coexisting phases is , and quartz, muscovite, and water are buffered throughout, fixing three degrees of freedom and leaving the assemblage to vary in the remaining two, which are pressure and temperature.
Each Barrovian index mineral appears through a specific mineral reaction that becomes thermodynamically favorable above a reaction boundary in space. The chlorite to biotite reaction releases iron and magnesium from chlorite into a new biotite structure at approximately 350 C. The biotite to garnet reaction produces the iron-magnesium-aluminum garnet almandine at approximately 450 C and 5 kilobars. The garnet to staurolite reaction forms the iron-aluminum silicate staurolite at approximately 500 C and 5.5 kilobars. The staurolite to kyanite reaction breaks down staurolite at approximately 600 C and 6.5 kilobars, releasing kyanite. The kyanite to sillimanite reaction crosses the polymorph phase boundary at approximately 600 to 650 C and 6 to 8 kilobars [Holdaway 1971].
Plotting these reaction points in space and fitting a single linear geothermal gradient through them, where is depth, is crustal density, and , yields in the range 20 to 30 C per kilometer. This matches the Barrovian medium-pressure regime. The same calculation for the Buchan index sequence chlorite, andalusite, sillimanite, cordierite yields above 40 C per kilometer, characteristic of low-pressure contact metamorphism; and the high-pressure sequence lawsonite, glaucophane, jadeite, aragonite yields below 15 C per kilometer, characteristic of subduction.
Corollary (depth of the Scottish Highlands metamorphism). The sillimanite isograd in Barrow's original mapping area records paleo-temperatures of about 650 C and pressures of 6 to 8 kilobars. Converting pressure to depth via with gives , in agreement with the 20 to 25 kilometer depth inferred from stratigraphic reconstruction of the Caledonian orogen. The Grampian Orogeny at about 470 million years ago buried the Dalradian Supergroup rocks to this depth during the collision of Baltica with Laurentia.
Bridge. The isograd-encoding argument builds toward 27.09.01 structural geology, where the same crustal-thickening structures (folds, nappes, thrust sheets) that buried the Scottish Highlands rocks are the load-bearing deformation features of every collisional orogen. The result appears again in 27.01.04 mantle plumes and large igneous provinces, where the high-heat-flow regime of a plume produces Buchan-style low-pressure metamorphism in the overlying crust, contrasting with the medium-pressure Barrovian pattern derived here. The foundational reason a single geothermal gradient fits all six Barrovian index minerals is that mineral reaction boundaries are thermodynamically determined functions of , so a regional terrain metamorphosed along one gradient samples a unique sequence of these boundaries. The central insight is that the bridge is between static mineral assemblages observed at the surface today and the dynamic pressure-temperature path the rock travelled through time.
Exercises Intermediate+
Advanced results Master
Barrow 1893: the index mineral sequence
George Barrow, mapping the Dalradian Supergroup of the Scottish Highlands for the Geological Survey, published in 1893 [Barrow 1893] the recognition that zones of distinct mineral assemblages succeed one another across the terrain in a fixed order. Each zone he named for its lowest-grade diagnostic mineral: chlorite, biotite, garnet, staurolite, kyanite, sillimanite. Barrow did not know the thermodynamic reactions underlying the sequence; he established the empirical mapping of mineral zones to spatial position, from which the concept of progressive regional metamorphism emerged. The medium-pressure regime he documented is now called Barrovian metamorphism, and his mapping area near Glen Clova remains a type locality.
Eskola 1915: the metamorphic facies concept
Pentti Eelis Eskola, working on the Orijärvi granitic region of southern Finland, published in 1915 [Eskola 1915] the recognition that rocks of identical bulk composition metamorphosed under the same pressure and temperature develop identical mineral assemblages. Eskola defined the metamorphic facies as a set of pressure-temperature conditions characterized by a specific diagnostic assemblage, and named the principal facies: greenschist, amphibolite, granulite, eclogite. The facies concept decoupled metamorphic grade from bulk composition for the first time, allowing petrologists to compare rocks of widely different chemistry by reading off the pressure and temperature of their formation from their assemblages alone.
Thompson 1957 and the AFM projection
John B. Thompson, Jr., in 1957 formalized the projection of pelitic assemblages onto the -- (AFM) plane, with quartz, muscovite, and water assumed present and projected out by the Thompson projection. The AFM diagram made the topology of coexisting phases in pelitic rocks graphically tractable: triangular diagrams showing the coexistence fields of chlorite, biotite, garnet, staurolite, kyanite, sillimanite. The topology changes with pressure and temperature, so AFM diagrams index the metamorphic grade directly, and the tie-line flips between, for example, chlorite-plus-garnet and biotite-plus-staurolite coincide with the Barrovian isograds.
Holdaway 1971 and the aluminum silicate triple point
Malcolm J. Holdaway in 1971 [Holdaway 1971] experimentally calibrated the triple point at and , resolving a discrepancy between earlier calibrations by Althaus and by Richardson. The triple point is the single at which andalusite, kyanite, and sillimanite coexist in equilibrium. Because the three polymorphs have the same chemical composition, their stability fields meet only at one point; a rock containing any two polymorphs equilibrated along the boundary between their fields, and a rock containing all three must have equilibrated at the triple point. The position of the triple point anchors the pressure calibration of every Barrovian and Buchan terrain.
Ferry and Spear 1978: garnet-biotite thermometry
John M. Ferry and Frank S. Spear in 1978 [Ferry-Spear 1978] experimentally calibrated the Fe-Mg exchange reaction between garnet and biotite, , whose equilibrium constant depends on temperature. The resulting thermometer, refined by subsequent calibrations, gives equilibration temperatures accurate to about 25 C for amphibolite-facies rocks. Applied to zoned garnet crystals, it recovers the prograde pressure-temperature path of the host rock point by point, since garnet preserves its growth composition due to slow intracrystalline diffusion.
Miyashiro 1961: paired metamorphic belts
Akiho Miyashiro in 1961 [Miyashiro 1961] described the paired metamorphic belts of Japan and the circum-Pacific: a high-pressure, low-temperature belt (blueschist to eclogite facies) on the oceanward side of a subduction zone paired with a low-pressure, high-temperature belt (Buchan-type, often with andalusite and sillimanite) on the continentward side. The pairing reflects the thermal structure of a subduction system: the subducting slab refrigerates the hanging wall close to the trench, producing high-/ metamorphism, while the volcanic arc above delivers heat to the crust, producing low-/ metamorphism. Paired belts have since been identified in the Alps, the Franciscan of California, and the Caledonides, providing a diagnostic tectonic signature of ancient subduction.
England and Richardson 1977: pressure-temperature-time paths
Philip C. England and Stephen W. Richardson in 1977 [England-Richardson 1977] modelled the thermal evolution of overthickened continental crust during collision. Crustal thickening buries rocks faster than they can heat, producing metamorphism at pressures higher than the thermal maximum; subsequent erosion uplifts and exposes the rocks, which heat up further during exhumation until thermal relaxation catches up. The resulting pressure-temperature-time paths are loops: pressure peaks before temperature, so the recorded mineral assemblage often does not correspond to the peak-pressure condition but to the peak-temperature condition, which is reached later at lower pressure. This explained why Barrovian terrains record mineral assemblages at lower pressure than the maximum burial depth, and placed metamorphic petrology on a quantitative tectonic basis.
Spear 1993: the P-T-t monograph
Frank S. Spear in 1993 [Spear 1993] synthesized the field into Metamorphic Phase Equilibria and Pressure-Temperature-Time Paths, the canonical modern reference. The monograph unifies Gibbs phase rule, thermodynamic databases, geothermobarometry, diffusion chronometry, and tectonic thermal modelling into a single framework for recovering the four-dimensional history of a rock from its mineral assemblage and zoning. The monograph formalized the inversion of mineral chemistry to pressure-temperature-time paths as the central quantitative program of metamorphic petrology.
Synthesis. The Barrovian-to-thermobarometry lineage builds toward a unified theory in which the mineral assemblage of a metamorphic rock is a thermodynamic recording of the pressure and temperature it experienced, and the chemical zoning of its constituent minerals is a recording of the path it took through space. The foundational reason a fixed index-mineral sequence recurs in collisional orogens worldwide is that pelitic bulk compositions share the same reaction topology in space, so any terrain following the same medium-pressure geothermal gradient crosses the same isograds in the same order; this is exactly the content of the Eskola facies concept. The central insight is that putting these together, Barrow's empirical zonation, Eskola's facies, Thompson's AFM projection, Holdaway's triple point, and the Ferry-Spear thermometer compose into a quantitative inversion engine from mineral assemblage to path. The pattern appears again in 27.09.01 structural geology, where the same crustal-thickening structures drive both the burial recorded by the metamorphic assemblage and the deformation recorded by the fabrics; the framework generalises from regional Barrovian metamorphism to every tectonic setting through the England-Richardson P-T-t loop; and the bridge is Miyashiro's paired belts, which connect the high-pressure subduction record to the low-pressure arc record in a single coherent thermal structure. Identifying the metamorphic facies with the tectonic regime closes the loop: blueschist means subduction, eclogite means deep burial, granulite means deep-crustal high-temperature stable-craton settings, and Barrovian amphibolite means continental collision.
Full proof set Master
Proposition (depth-pressure-temperature conversion along a Barrovian geotherm). Let the geothermal gradient during Barrovian metamorphism be degrees Celsius per kilometer of depth. Then the pressure-temperature coordinates of an index mineral isograd are related by , where is the surface temperature, is crustal density, and is gravitational acceleration. In particular, fixing two isograds in space determines and hence the metamorphic regime.
Proof. By definition of the lithostatic pressure gradient, , so . The geothermal gradient is the rate of temperature increase with depth, . Integrating from the surface (, ) gives , and substituting yields .
Given two index mineral isograds with measured coordinates and , solve the two linear equations
subtracting to eliminate :
For Barrow's chlorite isograd at and sillimanite isograd at , with and ,
consistent with the Barrovian medium-pressure regime (20 to 30 C per kilometer). Repeating for the Buchan sequence yields , and for the blueschist sequence .
Proposition (the England-Richardson P-T-t loop). In continental collision, crustal thickening by a factor buries a rock originally at depth to a new depth faster than thermal relaxation can warm it, so the peak pressure is reached before the peak temperature . Subsequent erosion at rate exhumates the rock along a path on which temperature initially continues to rise (because thermal relaxation lags the pressure drop), reaches a maximum at lower pressure than , then declines as the rock approaches the surface. Hence the recorded peak-temperature mineral assemblage corresponds to a pressure lower than , and the maximum burial depth is not directly readable from the peak assemblage alone.
Proof. Let the pre-thickening geotherm be and let crustal thickening by factor occur instantaneously at . Immediately after thickening, every rock at depth has been buried to depth , so its pressure jumps to , but its temperature is still the pre-thickening value . The new equilibrium geotherm under thickened crust would be with the new depth, giving , unchanged, so the long-time thermal state returns to the pre-thickening temperatures; but transiently, the thickened crust has a higher heat-producing volume and lower surface heat flux per unit depth, so the geotherm relaxes upward over a thermal timescale , where is the thickened-crust thickness and is thermal diffusivity. For , .
During this relaxation, a rock at depth follows a path in space. Pressure is always lithostatic, , and temperature evolves according to the one-dimensional heat equation
where is heat production per unit volume and is specific heat. Erosion at rate removes material from the top, advecting the rock upward: . Combining advection with conduction and integrating along the rock's trajectory, the temperature initially rises (because the thickened crust is heating up around it) before declining (because the rock approaches the cold surface). The peak temperature is reached at a depth shallower than , hence at a pressure lower than .
Quantitatively, the rock's trajectory in is a clockwise loop when is plotted on the vertical axis: pressure rises steeply during thickening, temperature rises more slowly during relaxation, then both decline during exhumation, but pressure declines faster than temperature, producing a path that crosses isotherms at angles. The peak-temperature assemblage equilibrates at but at . Garnet zoning, recording the prograde path, can recover the loop's shape; inclusions in garnet (such as rutile, which stabilizes at high pressure) preserve evidence of the high-pressure early phase even when the rim matrix equilibrated at lower pressure.
Connections Master
Minerals, rocks, and the rock cycle
27.02.01. The chapter survey introduces the three rock classes and the transformations between them. This unit deepens the metamorphic branch of that cycle: the index-mineral sequence catalogued here is the quantitative refinement of the qualitative progression slate to schist to gneiss, anchoring the rock-cycle picture to specific pressure-temperature conditions and specific tectonic regimes. Without the survey's mineralogical vocabulary (silicate structures, mica group, feldspar group, aluminum silicates), the Barrovian index minerals would be opaque labels; with it, they become precise geothermobarometers.Structural geology and rock deformation
27.09.01. Regional metamorphism and regional deformation are coeval in orogenic belts: the same crustal-thickening structures (folds, thrust nappes, ductile shear zones) that bury rocks to Barrovian depths also impose the foliation and lineation fabrics recorded in schists and gneisses. The England-Richardson P-T-t loop derived in this unit requires the overthickening that only large-scale thrusting can deliver, and the metamorphic grade mapped by Barrovian isograds maps directly onto the structural domains identified by finite-strain analysis. The Scottish Highlands type locality of Barrow is also the type locality of Caledonian thrust tectonics, including the Moine Thrust.Mantle plumes, hot spots, and large igneous provinces
27.01.04. Where Barrovian metamorphism records the medium-pressure, medium-temperature regime of continental collision, the contact metamorphism driven by a mantle plume or large igneous province delivers the low-pressure, high-temperature Buchan regime. The andalusite-and-cordierite assemblages of contact aureoles around plume-driven intrusions are the thermal signature of sustained basaltic underplating at the base of the crust, and they contrast directly with the kyanite-bearing assemblages of collisional settings. The two regimes are the end members of the crustal metamorphic spectrum, distinguished by whether heat is delivered conductively from below (plume) or advectively by burial (collision).Permian-Triassic mass extinction and the Siberian Traps
27.08.04. The Siberian Traps large igneous province, the leading kill-mechanism candidate for the end-Permian extinction, intruded into and metamorphosed vast volumes of country rock, producing contact-metamorphic aureoles and releasing carbon and sulfur from the heated sediments. The Buchan-style thermal metamorphism around the traps intrusions is the direct application of the low-pressure, high-temperature regime catalogued in this unit, and the thermobarometric methods developed for Barrovian terrains extend quantitatively to the contact-metamorphic aureoles that baked the Tunguska Basin sediments and drove the end-Permian carbon-isotope excursion.
Historical & philosophical context Master
Charles Lyell, in the third volume of his Principles of Geology (1833), systematized the doctrine of uniformitarianism, the principle that geological features are explained by processes operating at the rates observable today. Metamorphism, as a slow recrystallization driven by heat and pressure over geological time, fits this principle; Lyell used the term "metamorphic" in something close to its modern sense. The intellectual debt of every later worker in the field, from Barrow onward, is to Lyell's insistence that the same physics applies across deep time.
George Barrow, a field geologist with the Geological Survey of Scotland, published his index-mineral zonation in 1893 [Barrow 1893] without thermodynamic apparatus. His paper identified the sequence chlorite, biotite, garnet, staurolite, kyanite, sillimanite across the Dalradian Supergroup of the southeastern Highlands and named each zone for its lowest-grade diagnostic mineral. Barrow did not know the mineral reactions that produced his zones, nor the pressures and temperatures they recorded; his contribution was the empirical mapping of a consistent regional pattern, which subsequent workers recognized as the signature of progressive metamorphism. The medium-pressure regime he documented is now called Barrovian metamorphism, and the Glen Clova area remains an international reference section.
Pentti Eelis Eskola, working on the Orijärvi region of Finland between 1915 and 1920 [Eskola 1915], generalized Barrow's zonation into the metamorphic facies concept: rocks of identical bulk composition equilibrated at the same pressure and temperature develop the same mineral assemblage. Eskola defined the greenschist, amphibolite, granulite, eclogite, and contact-metamorphic facies, separating metamorphic grade from bulk chemistry for the first time. Akiho Miyashiro in 1961 [Miyashiro 1961] extended the framework to plate-tectonic scale with the paired metamorphic belts of Japan, reading the high-pressure-low-temperature and low-pressure-high-temperature pairing as the thermal signature of subduction. Philip England and Stephen Richardson in 1977 [England-Richardson 1977] recast metamorphic petrology in dynamical terms, showing that the recorded assemblage reflects a path through space, not a single static condition. Frank Spear's 1993 monograph [Spear 1993] unified the field into a quantitative inversion program: from mineral assemblage and zoning to pressure-temperature-time path.
Bibliography Master
@article{Barrow1893,
author = {Barrow, George},
title = {On an intrusion of muscovite-biotite gneiss in the south-eastern {Highlands} of {Scotland}, and its accompanying metamorphism},
journal = {Quarterly Journal of the Geological Society of London},
volume = {49},
number = {1-4},
pages = {330--358},
year = {1893},
}
@article{Eskola1915,
author = {Eskola, Pentti E.},
title = {On the relations between the chemical and mineralogical composition in the metamorphic rocks of the {Orij\"arvi} region},
journal = {Bulletin de la Commission g\'eologique de Finlande},
volume = {38},
pages = {109--145},
year = {1915},
}
@article{Miyashiro1961,
author = {Miyashiro, Akiho},
title = {Evolution of metamorphic belts},
journal = {Journal of Petrology},
volume = {2},
number = {3},
pages = {277--311},
year = {1961},
}
@article{EnglandRichardson1977,
author = {England, Philip C. and Richardson, Stephen W.},
title = {The influence of erosion upon the mineral facies of rocks from different metamorphic environments},
journal = {Journal of the Geological Society of London},
volume = {133},
number = {3},
pages = {201--213},
year = {1977},
}
@book{Spear1993,
author = {Spear, Frank S.},
title = {Metamorphic Phase Equilibria and Pressure-Temperature-Time Paths},
publisher = {Mineralogical Society of America},
address = {Washington, D.C.},
year = {1993},
}
@article{Holdaway1971,
author = {Holdaway, Malcolm J.},
title = {Stability of andalusite and the aluminum silicate phase diagram},
journal = {American Journal of Science},
volume = {271},
number = {2},
pages = {97--131},
year = {1971},
}
@article{FerrySpear1978,
author = {Ferry, John M. and Spear, Frank S.},
title = {Experimental calibration of the partitioning of {Fe} and {Mg} between biotite and garnet},
journal = {Contributions to Mineralogy and Petrology},
volume = {66},
number = {2},
pages = {113--117},
year = {1978},
}
@article{Thompson1957,
author = {Thompson, John B.},
title = {The graphical analysis of mineral assemblages in pelitic schists},
journal = {American Journal of Science},
volume = {255},
number = {11},
pages = {747--792},
year = {1957},
}
@book{Lyell1833,
author = {Lyell, Charles},
title = {Principles of Geology, Being an Attempt to Explain the Former Changes of the Earth's Surface, by Reference to Causes Now in Operation},
volume = {3},
publisher = {John Murray},
address = {London},
year = {1833},
}
@book{Winter2010,
author = {Winter, John D.},
title = {Principles of Igneous and Metamorphic Petrology},
edition = {2},
publisher = {Pearson},
address = {Upper Saddle River, NJ},
year = {2010},
}
@book{PhilpottsAgue2022,
author = {Philpotts, Anthony R. and Ague, Jay J.},
title = {Principles of Igneous and Metamorphic Petrology},
edition = {3},
publisher = {Cambridge University Press},
year = {2022},
}
@book{Yardley1989,
author = {Yardley, Bruce W. D.},
title = {An Introduction to Metamorphic Petrology},
publisher = {Longman Scientific and Technical},
address = {Harlow},
year = {1989},
}