Myron G. Best. Igneous and metamorphic 2003 Blackwell Science

Подождите немного. Документ загружается.

metamorphic paths causing the metamorphism. (The

homonym “terrain” refers to a geographic feature, such

as a mountainous terrain.) Figure 14.23 is an attempt to

relate the types of metamorphism from the previous

paragraph to geologic settings.

Ocean-ridge Metamorphism. The products of ocean-

ridge metamorphism are concealed on the seaﬂoor

except for scraps that have been locally accreted onto

island arcs and continental margins in subduction

zones as ophiolite (Section 13.6) in accretionary prisms

(Figure 14.24).

Hot, highly fractured magmatic rock exposed to

cold seawater at oceanic spreading ridges provides an

exceptional opportunity for movement of heat and

mass to create equilibrating metamorphic changes. The

global ridge system has been referred to as the “radia-

tor atop the oceanic magma engine” (Special Interest

Metamorphic Rocks and Metamorphism: An Overview

427

Trench

Fore-arc

basin

High P/T metamorphic rocks

in accretionary prism

(a)

Magmatic

arc

Oceanic crust

Lithosphere

Asthenosphere

1 and 4 closely

telescoped

(b)

Lithosphere

Asthenosphere

1. Zeolite

2. Prehnite–pumpellite

3. Blueschist

4. Greenschist

5. Amphibolite

migmatite and other partial

melts in crust

M-discontinuity

14.24 Schematic cross-sections through a continental margin subduction zone showing typical settings for regional and contact metamorphism.

Redrawn from Ernst (1976). (a) Underthrust imbrication of slices of arc-trench sediments and scraped-off oceanic lithosphere builds an

accretionary prism, or accretionary wedge (compare Figures 11.16 and 13.17). (b) Distribution of metamorphic facies (Figure 14.33) in

an orogenic belt, or orogen. Note that boundaries between facies in the magmatic arc in the core of the orogen are deﬂected nearer the

surface where the geothermal gradient is elevated. This is the setting of regional Barrovian-style metamorphism of the greenschist–

amphibolite facies series along an intermediate P metamorphic ﬁeld gradient that culminates with migmatites in the deeper crust. Con-

tact thermal metamorphism develops around magmatic intrusions, especially in the shallow crust, perturbing the facies boundaries as

drawn. The high P/T facies series (zeolite, prehnite–pumpellyite, and blueschist) develops in the accretionary prism adjacent to the trench

where the cool subducting oceanic slab acts as a heat sink, depressing geotherms.

Box 11.3). Tensile cracking during cooling and con-

traction of submarine lavas and feeder dikes, coupled

with active extensional rifting, promotes deep propa-

gation of fractures in the oceanic crust (Figure 13.1a).

Cracks allow seawater to circulate through the hot

rock, promoting hydrating mineral reactions that par-

tially to completely replace primary anhydrous mag-

matic phases. Heated seawater expands and buoys

upward out of the cracks, carrying ions leached from

the rock and dissolved in the hot brine. Sulﬁdes and

other solids precipitate in the cold seawater as this

solution exits from seaﬂoor hot springs, creating

particle-laden “black smokers” at vents along the ridge

(Figure 11.17). Basaltic rocks are metasomatized (or

hydrothermally altered, depending on one’s personal

viewpoint) to varying degrees by addition of Na and

water. Less mobile Ti, Al, and Mg may increase in

concentration in the protolith as a consequence of loss

of substantial Ca. Relict basaltic fabrics are preserved

in these spilites, whose metasomatic mineral com-

position is dominantly sodic plagioclase  chlorite 

quartz.

Relentless spreading of the seaﬂoor distributes the

products of ocean-ridge metamorphism across two-

thirds of the globe. As the hydrated crust is subducted

and heated at converging plate junctures, mineral

reactions liberate water that ﬂuxes the overlying man-

tle wedge and leads to the generation of arc magmas

(Section 11.4).

Regional Metamorphism. Within the continental crust

the most widespread of any metamorphism is regional

metamorphism. It is developed in the vast roots of oro-

genic belts (orogens) and, hence, is often called oro-

genic metamorphism. It involves broadly concurrent

deformation resulting from contractional stresses dur-

ing convergence of lithospheric plates in the subduc-

tion zone and recrystallization resulting from increases

in P and T in the thickened crustal welt (Figure 14.24).

Classic terranes include the Alps of southern Europe

and the Appalachian Mountains of eastern North

America, continuing, in a predrift connection, into the

Caledonides of the British Isles and Scandinavia. Rem-

nants of deeply eroded roots of Precambrian orogens

are represented in regional terranes in the Canadian

Shield and cratons in Australia, Africa, and other con-

tinents. Tectonites produced by regional metamorphism

are the familiar slates, schists, and gneisses.

Increased temperatures in the orogen are created as

geotherms adjust to the crust that is thickened by con-

tractional overthrusts and folds, magmatic underplates

(Figure 11.2), and stacks of volcanic deposits. Temper-

atures are sufﬁciently high in the lower crust to cause

partial melting and generation of calc–alkaline magmas

that ascend into the shallower crust as granitoid plu-

tons. Isostatic uplift and erosion during and following

orogeny exposes the crustal welt of metamorphic and

plutonic rock.

Orogens typically evolve over hundreds of millions

of years and experience more or less discrete pulses

of deformation or tectonic events related to changes

in character of converging oceanic plates and their

rates of convergence. Heating of the crustal welt may

accompany these pulses or occur sometime else, also

commonly in distinct episodes. Consequently, regional

terranes in orogens typically evolve through multiple

episodes of deformation and recrystallization, each

several million years in duration. Superposed meta-

morphic imprints may be of insufﬁcient intensity or

duration to completely erase earlier metamorphic

fabrics and mineralogical compositions, making it

possible to unravel at least a part of the history of

polymetamorphism.

Burial metamorphism. In very thick depositional

basins the more deeply buried sedimentary and vol-

canic material can be subjected to temperatures of

200–300°C sufﬁcient to cause recrystallization, or bur-

ial metamorphism. The classic terrane described by

Coombs (1961) is a thick stack (10 km) of Mesozoic

volcaniclastic rocks in southernmost New Zealand

(see Figure 18.30). Although of regional extent, burial

metamorphism has little or no associated penetrative

ductile deformation so that relict depositional fabrics

are usually well preserved (Figure 14.2) in rocks now

composed of low-T minerals such as zeolites. Asso-

ciated granitoids are also absent. Heated advecting

aqueous ﬂuids in the deep basins are likely involved,

so it is a type of hydrothermal metamorphism. A

currently active hydrothermal metamorphic system is

in the thick pile of sediment in the Colorado River

delta in the Salton Sea geothermal ﬁeld of southern

California and northern Baja Mexico (see Special

Interest Box 18.1).

Thick piles of sedimentary and volcanic rock accu-

mulated along passive continental margins, such as

along the eastern and Gulf of Mexico coasts of the

United States, that experience burial metamorphism

may subsequently suffer orogenesis as changes in plate

motion occur. In this polymetamorphism, the earlier

simple burial effects may be completely obscured.

Contact Metamorphism. Country rocks surrounding a

magmatic intrusion are subject to contact metamor-

phism. Intrusions liberate heat that conducts (Section

8.4.1) and advects via percolating ﬂuids into the coun-

try rock (Figure 4.12), producing a thermal gradient in

which T diminishes away from the magmatic heat

source. Oxygen isotope studies reveal that hydrother-

mal ﬂuids can advect many kilometers from an intru-

sion (Section 4.3.3). Many factors inﬂuence the extent

and conﬁguration of the thermal gradient in the

428 Igneous and Metamorphic Petrology

contact metamorphic aureole and how elevated the

temperatures become. Important factors include, for

the country rock, its composition, permeability to ad-

vecting ﬂuids (Section 8.5), pre-existing T, and depth

beneath the surface. Thermal metamorphism is most

conspicuous in shallow crustal country rock because

of the large thermal contrast with intrusive magmas.

Increasingly deeper, and therefore hotter, country

Metamorphic Rocks and Metamorphism: An Overview

429

45°25′N

mi 3

km 4

Slate

Granodiorite

Hornfels

Spotted slate

and semihornfels

Slate

Onawa

69°15′W69°25′W

(a)

(b)

Quartz

0 mm 0.2

(c)

Biotite

0 mm 0.2

Quartz

Muscovite

Andalusite

Cordierite

(d)

Perthite

Andalusite

0 mm 0.2

Biotite

Quartz

Cordierite

Sillimanite

14.25 Contact metamorphic aureole surrounding the Onawa, Maine pluton in the Appalachian Mountains. (a) Generalized geologic map of

the Onawa granodiorite pluton and surrounding zoned aureole. Redrawn from Moore (1960). (b) to (d) Thin sections showing progres-

sive thermal metamorphism of pelitic country rocks in the aureole. (b) Slate. Country rock into which the magma was emplaced had

experienced previous regional metamorphism with development of a low-grade mineral assemblage of quartz  chlorite  phengitic

white mica  Fe–Ti oxides  graphite. Preferred orientation of platy grains deﬁnes a slaty cleavage. Angular quartz grains are relics of the

original clastic fabric of the parent shale. (c) Spotted semihornfels in outer part of contact aureole. Recrystallization under hydrostatic

conditions has completely obliterated the anisotropic fabric of the slate protolith, producing a somewhat coarser granoblastic

fabric. Most of the rock is composed of quartz  biotite  muscovite  graphite (black opaque grains). Large poikiloblasts of andalusite

(high relief ) and cordierite (low) are crowded with numerous tiny inclusions of matrix phases. On weathered rock surfaces, andalusites

stand out in relief, whereas cordierites etch out because of replacement by ﬁne-grained sericite and other alteration products; both form

spots. (d) Hornfels. Coarser granoblastic rock from the inner part of the zoned contact aureole. Anhedral andalusite was apparently

unstable and decomposing when the metamorphic reactions stopped. Samples provided courtesy of James M. Moore, Jr.

rock can exhibit wider aureoles but the thermal effects

may be progressively less evident because of ongoing

metamorphic processes and equilibration. Signiﬁcant

factors for the intrusion include its volume, shape,

T, composition, volatile content and composition,

whether the magma cools conductively (Section 8.4) or

convectively (Section 8.6), and whether magma replen-

ishment occurs. Contact aureoles range from barely

perceptible ones on the order of a centimeter wide next

to thin dikes that cool in less than a year to aureoles

measured in kilometers around large plutons that may

take tens of thousands of years to cool.

Aureole fabric around passively emplaced intru-

sions usually in the shallower crust (Section 9.4.3) is

commonly isotropic and granoblastic (Figure 14.25b,

c) because the country rocks are simply “cooked”

under hydrostatic conditions. Hence, metamorphism is

essentially thermal and deformation is so limited that

little or no preferred orientation of mineral grains re-

sults. The rock is typically a hornfels. However, around

forcefully intruded magmas usually in the deeper crust,

ductile deformation can create anisotropic fabrics (Fig-

ure 9.19b) that are indistinguishable in outcrop from

tectonites created by regional metamorphism.

In orogens, many granitoid plutons are post-tectonic

(Section 9.4.3) and postdate regional metamorphism

because magma ascent from deep crustal sources

may require a long interval of time after the thermal

“maximum” associated with peak tectonic activity and

magma generation. In such cases contact metamor-

phism is overprinted on earlier regionally metamor-

phosed country rocks, such as shown in Figure 14.25;

this is yet another expression of polymorphism.

In essentially closed-system aureoles little change in

chemical composition of the rocks can be discerned,

other than loss of water during progressive mineral

reactions. The metamorphism is virtually isochemical

as well as thermal.

However, in more open-system, metasomatic aure-

oles, major changes in country-rock composition take

place. Depending on the nature of the advecting ﬂuids

and country rock, a compositional gradient will be

established in addition to the thermal gradient. Com-

positional gradients are most evident where there are

sharp contrasts between compositions of the advecting

ﬂuids and country rock. The most striking common

examples occur where calcareous or impure carbon-

ate rocks (limestones and dolostones with variable

amounts of admixed quartz and clays) are increasingly

silicated nearer granitoid intrusions from which silica-

rich ﬂuids are expelled. The resulting metasomatic

calc-silicate skarns are made of actinolite, diopside,

andradite–grossular garnet, vesuvianite (idocrase), and

other Ca–Al–Mg–Fe silicates.

The source of the metasomatizing hydrothermal

solutions can be:

1. Exsolved water liberated from the crystallizing

magma intrusion.

2. Heated ﬂuids (“meteoric water”) in openings in the

country rocks.

3. Water liberated from nearby decomposing hydrous

minerals undergoing prograde dehydration reac-

tions (see below).

In addition to the normal rock-forming elements

(e.g. Si, Al, Fe) introduced into country rock by the

metasomatizing solutions, other elements (e.g. S, Zn,

Pb, Cu, Ag, Au) can also locally be introduced to form

ore deposits.

Dynamic Metamorphism in Ductile Shear Zones.

Compressional stresses in orogens create shallow

crustal brittle faults and deeper crustal ductile shear

zones. Major continental-scale faults and shear zones

include the San Andreas in California (see Figure 18.5),

the Brevard in the Appalachian Mountains (see Figure

18.7), the Moine Thrust in Scotland (see Figure 14.30),

the Alpine fault in New Zealand (see Figure 18.30),

and many others. Along any zone of displacement, high

rates of strain create cataclastic rocks by brittle defor-

mation, mylonite by ductile deformation, and very

locally pseudotachylite by frictional melting (Special

Interest Box 11.1).

Impact Metamorphism. Unrelated to tectonic settings

are bodies of rock impacted by meteorites that cause

impact metamorphism, also called shock metamor-

phism. Extremely high rates of strain resulting from the

transient high-speed shock wave create cataclastic fab-

ric, high-P phases, such as the silica polymorphs coesite

and stishovite, intracrystalline plastic slip (see Section

17.2.2), and local melting, producing pseudotachylite

(Grieve et al., 1995).

14.2.4 Metamorphic Grade

In the Onawa contact aureole (Figure 14.25), pelitic

hornfelses closer to the intrusion that contain higher T

minerals are said to be of higher metamorphic grade

than the pelites farther away made of lower T minerals.

Grade corresponds to equilibration T without restric-

tions as to P when comparing rocks of similar bulk

chemical composition. Differences in fabric are pos-

sible and, in fact, common in different grades; in the

Onawa aureole, the lower grade rocks are spotted

slates in contrast to the higher grade coarser hornfelses

near the contact. However, reliance on fabric, espe-

cially grain size, to infer metamorphic grade can be

very misleading because of the complex factors on

which fabric depends. Mineralogical differences are

more critical in establishing grade, especially as they

result from differences in the T of equilibration during

metamorphism.

430 Igneous and Metamorphic Petrology

How does a petrologist distinguish between high-T

minerals characterizing a high grade of metamorphism

from low-T and low-grade?

In Figure 14.25c and d, the major difference in the

mineral assemblage in the Onawa contact hornfelses is

the presence of muscovite and quartz in the outer,

lower T part of the aureole and the absence of this

mineral pair in the higher grade hornfels closer to the

intrusion that instead contain alkali feldspar and an

SiO

polymorph. This change in mineral assem-

blage can be modeled by the mineralogical reaction

14.1 thermal energy  (K,Na)Al

AlSi

(OH)

muscovite

 SiO

 (K,Na)AlSi

 Al

SiO

 H

quartz alkali feldspar

SiO

can be any of the polymorphs andalusite,

sillimanite, or kyanite, depending on P–T conditions.

Breakdown of muscovite  quartz is an endothermic

dehydration reaction, consuming heat and liberating

water. (Because of excess quartz in the Onawa aureole

rock relative to muscovite, it was not entirely con-

sumed in the reaction and higher grade hornfelses still

contain quartz.)

Thus, a general rule is that lower-grade metamorphic

mineral assemblages contain more volatile-rich phases

whereas higher-grade assemblages equilibrated at higher

T are made of less volatile-rich or volatile-free minerals.

Remember that in considerations of metamorphic

grade we compare rocks of similar bulk chemical com-

position: pelites with pelites, metabasites with meta-

basites, and so on; we don’t attempt to compare toads

with horses!

Devolatilizing prograde reactions liberate volatiles

with increasing T. Dehydrating prograde reactions,

such as the breakdown of muscovite, give off water as

T increases; the hydrous mineral assemblage, in effect,

“dries out” at higher T as a result of the endothermic

reaction. Prograde decarbonation reactions liberate

from calcite, dolomite, and other CO

-bearing

minerals. A metamorphic path that progresses toward

a peak or maximum T is referred to as prograde meta-

morphism. Thus the aureole rocks toward the intru-

sion at Onawa experienced prograde metamorphism,

as did the sequence of maﬁc rocks in Figure 14.6a to d.

After a peak T is reached during prograde meta-

morphism, there must logically be a reversal going

down T. This is clearly the case for rocks now exposed

at the surface. Reversal may happen soon (in a geolo-

gical sense), or many, possibly tens or hundreds of, mil-

lions of years after attaining the peak T. But high-grade

metamorphic mineral assemblages (and also magmatic

rocks made of high-T minerals) exposed at the surface

all over the world indicate that retrograde reactions

during the decreasing T of retrograde metamorphism

do not generally occur. Prograde mineral reactions are

not automatically reversed during cooling and uplift to

the surface for two apparent reasons. The ﬁrst follows

from the nature of reaction kinetics (Section 3.6.2)

whereby phases formed at high T typically persist

metastably at falling T because of slowing kinetic rates

of equilibrating reactions. The second is that prograde

reactions progressively liberate volatile ﬂuids, which

migrate out of the metamorphosing body. As T falls

after the peak metamorphic event, no volatiles are avail-

able within the system and little can enter into typically

impermeable bodies. Without available water, hydrous

phases more stable at lower T cannot replace high-T

anhydrous minerals. Primary magmatic minerals in

granitic rocks are typically only partially replaced (e.g.

Plate IV), if at all, during the excursion of the rock

system from the solidus to lower temperatures. In

metamorphic rocks, high-T anhydrous silicates, such

as garnet, are commonly incompletely replaced by a

hydrous Mg–Fe alumino-silicate, such as chlorite, if at

all. Retrogressive metamorphism (or alteration, as it

might be called by some geologists) is commonly local-

ized along fractures or shear zones in aggregates of

high-T minerals. Shearing not only produces a higher

energy state in the strained grains, making them less

stable and more amenable to replacement, but also

provides an avenue for advective inﬂux of volatile ﬂuid

into the otherwise impermeable rock body. Planar

veins of hydrous Ca–Al silicates, especially epidote, are

common in granitic rocks (Figure 14.26) and form by

reaction between plagioclase and water that invaded a

fracture, possibly during cooling below the solidus. In

Figure 14.27, an anhydrous eclogite assemblage has

been retrograded to a lower T hydrous assemblage in a

vein.

14.2.5 Metamorphic Zones

Terranes of metamorphosed rock commonly display

geographically limited, or mappable, metamorphic

zones. Each zone is deﬁned by some distinctive fabric

attribute or mineral assemblage in rocks derived from

compositionally similar protoliths. The sequence of

zones reﬂects the existence of gradients, or spatial vari-

ations, in conditions of metamorphism. Zones range in

dimension from that of an outcrop, or small portions of

it, to terranes measured in hundreds of square kilome-

ters. Zones are of considerable signiﬁcance in the elu-

cidation of transfers of heat that caused changes in the

T of metamorphism, of changes in P that were caused

by variations in depth of burial, and of changes in the

nature of advecting ﬂuids from different potential

sources. Where successive zones in a regional terrane

or contact aureole correspond essentially to a thermal

gradient they will be of different grade. Zones based

on some fabric attribute can be used to infer something

of states of stress and deformation. Zones and the

Metamorphic Rocks and Metamorphism: An Overview

431

gradients they reﬂect in regional terranes are of partic-

ular utility in understanding the evolution of orogens

and continents. The following three examples of meta-

morphic zones illustrate their connection to gradients

in equilibrating conditions.

Thermal Metamorphism in the Onawa Contact Aureole.

This aureole (Figure 14.25) was developed in previ-

ously regionally metamorphosed pelitic country rocks

and exhibits conspicuous textural and mineral zones

432 Igneous and Metamorphic Petrology

14.26 Vein of epidote (black) in granodiorite. Light-colored borders on the epidote consist of chloritized biotites and feldspars whose chalky

appearance is the result of replacement by very ﬁne grained sericite and other alteration products. Pocket knife 8 cm long rests on fresh,

unaltered granodiorite.

that resulted from a longer duration of conductive and

possibly advective heating and more elevated temperat-

ures closer to the intrusion. The ﬁrst visible indication

in hand sample of the thermally driven recrystallization

in the country rock slates in the outermost part of the

aureole is ill-deﬁned spots made of variably altered (to

ﬁne-grained phyllosilicates) poikiloblasts of cordierite

and/or andalusite. These poikiloblasts are slightly

lighter colored than the surrounding darker graphitic

matrix of the slate but tend to weather out on expos-

Metamorphic Rocks and Metamorphism: An Overview

433

ures, creating pits. With progressive recrystallization

toward the intrusion, slaty cleavage is obliterated and

the rock takes on a more massive appearance, becom-

ing ﬁrst a “semi” hornfels and ﬁnally nearer the intru-

sion a relatively coarse granoblastic hornfels (Figure

14.25d). Mineralogically, the transition from semihorn-

fels to hornfels was marked by the decomposition of

muscovite reacting with quartz to form alkali feldspar

plus sillimanite. Thus, one can recognize at least two

mineralogical zones, an outer lower grade zone of mus-

covite  quartz ( other phases) succeeded inward by

a higher grade zone of alkali feldspar  sillimanite (

other phases). The lower grade outer zone contains

more hydrous minerals—muscovite  biotite—than

does the inner zone—biotite only.

Metasomatic Skarn Zones at Crestmore, California.

This is an unusually well developed and now classic

example of mineral zones developed in calcareous

country rocks around a small granitic porphyry intru-

sion (Figure 14.28). The thermal gradient attained

exceptionally high temperatures next to the intrusion

but the successive zones are chieﬂy the result of a

strong chemical gradient imposed by hydrothermal

solutions expelled from the freezing magma.

The sequence of zones and ten subzones and their

constituent mineral assemblages as determined by

Burnham (1959) are, beginning with the lowest grade

outermost zone

Forsterite 1. calcite  brucite  clinohumite

zone  spinel

2. calcite  clinohumite  forsterite

 spinel

3. calcite  forsterite  spinel

 clintonite

Monticellite 4. calcite  forsterite  monticellite

zone  clintonite

5. calcite  monticellite  melilite

 clintonite

6. calcite  monticellite 

spurrite (or tilleyite)  melilite

7. monticellite  spurrite  merwinite

 melilite

Vesuvianite 8. vesuvianite  monticellite

zone  spurrite  merwinite  melilite

9. vesuvianite  monticellite

 diopside  wollastonite

Garnet zone 10. grossular  diopside  wollastonite

Note that once a particular “index” mineral, such as

monticellite, appeared in subzone 4 it persisted, pre-

sumably stably, not only through its corresponding

Retrograded eclogite

Fresh eclogite

(b)(a)

White mica

Sphene

0.5 mm

Jadeitic

pyroxene

Garnet

Rutile

Glaucophane

Epidote

Chlorite

14.27 Retrograde metamorphism of eclogite. (a) Hypothetical, schematic ﬁeld relations showing retrograde effects (shaded) controlled by two

sets of fractures along which hydrous ﬂuids gained access into the rock. Circle shows where thin section in (b) would be located.

(b) Thin section showing retrograde vein cutting diagonally from lower left to upper right through anhydrous omphacite  garnet 

rutile eclogite. Vein is made mostly of chlorite  epidote  glaucophane  white mica, which are all hydrous minerals. Note two garnet

grains on right partially pseudomorphed by chlorite, leaving embayed remnants; this is an example of a reaction texture in which the

reactant phase (garnet) and the product phase (chlorite) are both preserved. Sphene (titanite) has partially to completely replaced

rutile.

434 Igneous and Metamorphic Petrology

monticellite zone but also through the next higher

grade vesuvianite zone. It should also be noted that

calcite is stable in the six lower grade subzones but

is absent at higher grades, where only calc-silicate

minerals occur. Noting that the country rocks of

brucite marble (formed during an earlier metamorphic

event) far from the intrusion contain no quartz or

silicate minerals, it appears that the succession of zones

corresponds to increasing concentrations of silica and

elimination of CO

. Signiﬁcant amounts of Al were

introduced into the monticellite, vesuvianite, and gar-

net zones, in addition to Si, thereby stabilizing melilite,

vesuvianite, and grossular.

Many reactions could be written to account for the

progress of mineralogical changes through the zones

that consumed a lower grade mineral or mineral assem-

blage as a higher grade one was stabilized. A possible

silicating, decarbonation reaction producing some

vesuvianite in subzone 8 from the mineral assemblage

in 7 is

CaMgSiO

 2(2Ca

SiO

CaCO

)  3Ca

MgSi

monticellite spurrite merwinite

 4Ca

 15SiO

 12H

melilite in solution

→ 6Ca

(OH)

 2CO

vesuvianite

Another way to consider the progressive meta-

somatism is to graphically display (Figure 14.29) the

aureole chemical system in terms of the compositions

of the major minerals; these are made up essentially of

CaO, MgO, and SiO

and so can be represented in a

triangular diagram. (These composition diagrams are

described in Section 15.3.) This graphical analysis

ignores substantial Al in some minerals, as well as

minor Fe and the different temperatures that prevailed

in the successive zones. Despite the approximations,

the diagram clearly demonstrates that successively

higher grade zones have increasing proportions of

silica but decreasing CaO; MgO remains fairly constant.

These compositional variations are consistent with out-

ward migration into the country rock of silica-bearing

solutions derived from the intrusion and with outward

diminishing silication, or ﬁxation of SiO

, by reaction

with CaCO

and Mg(OH)

in the brucite marble

protolith.

Barrovian Zones in Pelites in Scotland. Historically,

the ﬁrst systematic study of mineralogical variations in

progressively metamorphosed rocks was made near the

turn of the last century by George Barrow (1912) in the

regional terrane just north of the Highland Boundary

fault in Scotland southwest of Aberdeen (Figure 14.30).

The regionally metamorphosed rocks of the Scottish

Highlands comprise a 13-km thick sequence of shale,

sandstone, limestone, and maﬁc lava that was folded

and thrust into nappes during the early Paleozoic Cale-

donide Orogeny. Following more or less concurrent

metamorphism during the deformation, numerous

post-orogenic granitoid plutons were intruded.

Barrow found signiﬁcant variations in the miner-

alogical composition and fabric of the pelites across the

Brucite marble

Monticellite zone

Idocrase zone

Garnet zone

Quartz monzonite

porphyry

0 m 400

14.28 Schematic cross-section through the granitic porphyry and

surrounding metasomatic zones at Crestmore. California. See

also Figure 14.29. Outermost forsterite zone lying between the

brucite marble country rocks and the monticellite zone is

omitted. Redrawn from Burnham (1959).

Calcite Periclase

CaO

MgO

Spurrite

and

tilleyite

Merwinite

Wollastonite Diopside

Monticellite

Forsterite

Clinohumite

7,8

SiO

CO

14.29 Compositional relations in the system CaO–MgO–SiO

with

excess CO

at more or less ﬁxed P but variable T. The path of

progressive silication of the carbonate country rocks at Crest-

more is shown by the sequence of numbers 1–10 representing

the skarn mineral assemblages that make up the metasomatic

zones. See also Figure 14.28. Redrawn from Burnham (1959).

during progressive metamorphism requires careful

study of mineral phase equilibria and reactions (Chap-

ter 16). In the thermal metamorphism in the Onawa

contact aureole, the ﬁrst appearance of the index

mineral pair alkali feldspar  sillimanite, produced by

breakdown of muscovite  quartz (reaction 14.1; Fig-

ure 14.31), likely depended on a critical change in T;

if so, the isograd represents the line of intersection of a

paleo-isothermal surface with the ground surface. In

the Crestmore metasomatic zones, successive isograds

were dictated chieﬂy by changes in silica activity.

In Figure 14.30, northwest of Aberdeen, there are

regional Buchan zones deﬁned on the basis of andalu-

site and cordierite in pelitic rocks. If equilibrium pre-

vailed, pressures cannot have exceeded about 4 kbar

(Figure 14.31), otherwise the other two Al

SiO

poly-

morphs would have been stabilized, as in the zonal suc-

cession from kyanite to sillimanite in typical Barrovian

zones in the remainder of the Scottish Highlands. The

lower pressures of Buchan metamorphism are also

consistent with the presence of Mg–Fe aluminosili-

cate cordierite rather than denser almandine garnet

(another Mg–Fe aluminosilicate) that is widespread in

Barrovian zones.

Geographic relations of Buchan and Barrovian

zones and variations in the grade of the latter provide

Metamorphic Rocks and Metamorphism: An Overview

435

terrane, whereas other protoliths showed less variation

in their constituent mineral assemblages because of

less sensitivity to changing metamorphic conditions.

Northward from the boundary fault, he mapped a

sequence of pelitic mineral zones beginning with the

lowest grade slates and slightly coarser phyllites of the

chlorite zone made basically of chlorite  white mica

 quartz. Succeeding zones were identiﬁed by the ﬁrst

appearance of a particular index mineral that com-

monly persisted into succeeeding higher grade zones

(see Figure 18.13). The onset of the next higher grade

biotite zone was marked by the ﬁrst appearance of

biotite in phyllites and coarser schists that contain addi-

tional chlorite  white mica  quartz. The beginning

of the garnet zone was delineated by the appearance of

an almandiferous garnet and of succeeding zones by

the ﬁrst appearance of staurolite, kyanite, and ﬁnally

sillimanite in the highest grade schists and gneisses.

Similar Barrovian metamorphic zones, or segments of

them, in pelitic rocks have since been documented in

many other orogenic belts around the world.

One can envisage a three-dimensional surface of

constant grade during the metamorphism whose inter-

section with the present ground surface is a mappable

line, called an isograd, that is marked by the appear-

ance of a particular index mineral. Determining what

changing intensive variable or variables caused the ﬁrst

appearance of a particular index mineral at an isograd

Barrovian zones

Sillimanite

Kyanite

Garnet

Biotite

Chlorite

Ocean

miles

Buchan

andalusite

zone

Fault

Aberdeen

Edinburgh

Glasgow

Highland Boundary Fault

Atlantic

Great Glen ]

Moine

Thrust

14.30 Simpliﬁed map of Barrovian mineralogical zones in regional

metamorphic rocks of the Caledonide orogen in Scottish

Highlands. Thin staurolite zone between garnet and kyanite

zones is omitted. Areally restricted Buchan zones lie north of

Aberdeen. Redrawn after Gillen (1982).

P (kbar)

Depth (km)

KYANITE

SILLIMANITE

ANDALUSITE

granite

solidus

Pa  Qtz

Ab  Al

SiO

 H

Ms  Qtz

Kfs  Al

SiO

 H



400 500 600

T (°C)

700 800

Water-Saturated

14.31 Breakdown conditions for paragonite and muscovite at P

 P

ﬂuid

 P compared to stability ﬁelds of Al

SiO

poly-

morphs (long dashed boundary lines). Decomposition of nat-

ural paragonite–muscovite solid solutions in the presence of

quartz occurs between the stippled bands. In systems where

P P, muscovite  quartz breaks down at lower T (line

of open circles). See Table 14.1 for key to mineral abbrevia-

tions. Redrawn from Spear (1995).

436 Igneous and Metamorphic Petrology

as it equilibrated under peak conditions. For example,

it can be inferred that the kyanite-bearing schist in

Figure 14.9 equilibrated at a higher P than did the

andalusite-bearing slate in Figure 14.10 because kyanite

is denser than andalusite. (Kyanite is actually the densest

SiO

polymorph.) However, we must hasten to add

that this may only be true if the kyanite schist equili-

brated at an equal or higher T than the andalusite slate,

as may be appreciated from the stability diagram for

the Al

SiO

polymorphs in Figure 14.31. Because of

the positively sloping boundary line between the sta-

bility ﬁelds of kyanite and andalusite, it may be pos-

sible for kyanite to have equilibrated at a lower P and

lower T than the andalusite in the slate. Other minerals

in the two rocks may provide additional information

indicating their relative temperatures of equilibration,

resolving ambiguities such as this.

Hence, two or more minerals comprising the meta-

morphic mineral assemblage in a rock provide the cru-

cial information regarding the equilibrating intensive

variables. The minerals occurring in a metamorphic

rock—its mineralogical composition—may differ from

its mineral assemblage equilibrated under some particu-

lar set of intensive variables at the peak of prograde

conditions. Because of sluggish kinetics, relict phases

surviving metastably from the protolith or formed

along the prograde path, as well as possible phases

crystallized along the retrograde path, may occur in the

Chlorite

zone

Biotite

Garnet

B Biotite isograd G Garnet isograd

Pelitic

rocks

Quartzo-feldspathic

rocks

Maﬁc

rocks

Calcareous

rocks

14.32 Hypothetical, schematic Barrovian mineralogical zones and bounding isograds (thicker lines). Mineralogical variations in interlayered

maﬁc, quartzo-feldspathic, and calcareous rocks are independent of those in pelitic rocks on which the zones and isograds are based;

consequently, position of deﬁning isograds can only be interpolated through these compositions. Isograds are independent of contacts

between chemical rock groups and rock structure, premetamorphic folds in this example.

insights into gradients in P and T that prevailed in the

Scottish segment of the Caledonide Orogen.

Metamorphic zones and their bounding isograds do

not necessarily correspond to stratigraphic and struc-

tural contacts on geologic maps (Figure 14.32). Condi-

tions of metamorphism are commonly independent of

patterns of pre-metamorphic protolith deposition and

deformation. On the other hand, isograd patterns can

be useful in unraveling the effects of tectonism and

magmatism. For example, isograd “surfaces” in a con-

tact terrane might dome over an underlying intrusion

and sharply offset isograds might be the result of post-

metamorphic faulting, such as across the Great Glen in

Figure 14.30.

14.2.6 Intensive Variables and Stable Mineral

Assemblages

Intensive thermodynamic variables prevailing in a

metamorphic system govern phase equilibria. The ﬁnal

equilibrium state of the system is generally assumed

to be that attained at the peak T in the course of a

progressive metamorphic path where reaction kinetics

are the most favorable. Unless proven otherwise, equi-

librating mineral reactions during the subsequent retro-

grade path are less prominent or even nonexistent.

A single mineral can provide limited insight into

what intensive variables—P, T, and concentrations of

chemical components—prevailed in the rock system