FLUID-INDUCED ALTERATION OF METASEDIMENTARY ROCKS IN THE SCOTTISHHIGHLANDS

Alexander Lewerentz

Fluid-induced alteration of metasedimentary rocks inthe Scottish Highlands

Alexander Lewerentz

©Alexander Lewerentz, Stockholm University 2017 ISBN print 978-91-7649-885-9ISBN PDF 978-91-7649-886-6 Printed in Sweden by Universitetsservice US-AB, Stockholm 2017Distributor: Department of Geological Sciences

Without music or anintriguing idea,colour becomes pallor,man becomes carcass,home becomes catacomb,and the dead are but for amoment motionless. Edgar Allan Poe

TABLE OF CONTENTS ABSTRACT ............................................................................................................................................ 1

SAMMANFATTNING ........................................................................................................................... 2

LIST OF PAPERS ................................................................................................................................... 3

1. METAMORPHISM AND METAMORPHIC FLUIDS ..................................................................... 4

1.1 Barrovian metamorphism .............................................................................................................. 4

1.2 Metamorphic fluids ....................................................................................................................... 8

1.3 Mechanisms for fluid flow ............................................................................................................ 8

1.4 Veins ............................................................................................................................................ 10

1.5 Fluid-rock interaction .................................................................................................................. 11

1.6 Quantification of metamorphic fluid fluxes ................................................................................ 12

2 GEOLOGICAL SETTING ................................................................................................................. 15

2.1 The Dalradian of Scotland ........................................................................................................... 15

2.2 Study area 1: Glen Esk ................................................................................................................ 16

2.3 Study area 2: Islay ....................................................................................................................... 16

3. METHODS ........................................................................................................................................ 19

3.1 Paper I ......................................................................................................................................... 19

3.2 Paper II ........................................................................................................................................ 19

3.3 Paper III ....................................................................................................................................... 21

3.4 Paper IV ....................................................................................................................................... 21

4. SUMMARY OF PAPERS ................................................................................................................. 22

4.1 Paper I ......................................................................................................................................... 22

4.2 Paper II ........................................................................................................................................ 25

4.3 Paper III ....................................................................................................................................... 26

4.4 Paper IV ....................................................................................................................................... 27

5. ACKNOWLEDGEMENTS .............................................................................................................. 30

6. REFERENCES .................................................................................................................................. 31

1

ABSTRACT Fluids, mainly H2O and CO2, are released from H- and C-bearing phases during prograde metamorphism. Because of the buoyancy of these fluids, they rise within the crust towards the surface of the Earth. Metamorphic fluids take advantage of permeable horizons, shear zones, fold hinges, fractures, and are channelled into high-flux zones. Fluid fluxes for channelized fluid flow may exceed background pervasive fluxes by several orders of magnitude. Meta-morphic fluids react with the surrounding rock during fluid flow, and altered zones are com-monly observed adjacent to high-flux conduits. Fluid-altered rock is texturally, mineralogical-ly, chemically, and isotopically different from rock unaffected by fluid flow. In this thesis, fluid-rock interaction is studied at two localities in the Scottish Highlands: Glen Esk and the Isle of Islay.

Glen Esk is one of the type localities used by George Barrow (1853-1932) to propose the concept of metamorphic zones and metamorphic index minerals as an approximate deter-mination of metamorphic grade. In several of the metamorphic zones in Glen Esk, index min-eral distribution is highly dependent on proximity to veins. The occurrence of index minerals is therefore not only controlled by pressure and temperature, but also by the availability of metamorphic fluids. Evidence of a retrograde fluid flow event from the North Esk Fault is observed in Glen Esk, for which a time-averaged fluid flux of 0.0003 – 0.0126 m3∙m-2∙yr-1 is calculated. The duration of the fluid event is estimated to between 16 and 334 kyr.

On the Isle of Islay, kyanite is observed in rocks of chlorite or lower-biotite metamor-phic grade, i.e. much lower temperatures than usually associated with kyanite formation. The favoured explanation for this is retrograde infiltration of extremely high-CO2 fluids, at least locally XCO2 > 0.7, at ~340°C, which altered these rocks and stabilised kyanite in a carbonate-bearing assemblage. Oxygen and carbon stable isotope profiles across the Islay Anticline re-veals highly channelized fluid flow along the axial region of this fold, with fluid:rock ratios at least four times higher than in rock farther away from the fold. Although carbon and oxygen isotope ratios of metacarbonate rocks were altered along the Islay Anticline, negative anoma-lies observed below and above the Port Askaig Tillite Formation cannot solely be attributed to metamorphic fluid flow, which implies that these rocks to varying degree retain their primary paleoclimatological isotopic signatures.

Keywords: metamorphic fluids, metamorphic fluid flow, fluid-rock interaction, metamorphic fluid fluxes.

2

SAMMANFATTNING Stora volymer H2O och CO2 frigörs som fluider under prograd metamorfos. Metamorfa fluider har lägre densitet än det omgivande berget, varför de stiger genom jordskorpan mot jordytan. Metamorfa fluider kanaliseras i permabla lager, skjuvzoner, veckaxlar, sprickor och andra högflödeszoner. Kanaliserade fluidflöden kan vara flera storleksordningar högre än bakgrundsvärdet för fluidflöde inom en bergart. Metamorfa fluider reagerar under transport med det omgivande berget och bildar fluidomvandlade zoner i anslutning till högflödeskanaler. Fluidomvandlat berg uppvisar texturella, mineralogiska, kemiska och isotopsammansättningsmässiga skillnader i jämförelse med berg som inte utsatts för fluidomvandling. I denna avhandling behandlas reaktioner mellan fluid och berg som studerats i två lokaler i de skotska högländerna: Glen Esk och Islay.

Glen Esk är en av de typlokaler som George Barrow (1853-1932) använde för att lägga fram konceptet om metamorfa zoner och metamorfa indexmineral som används för att ungefärligt uppskatta metamorf grad. I flera av de metamorfa zonerna är förekomsten av indexmineral i hög grad beroende av närhet till kvartsådror, vilket visar att bildandet av indexmineral inte bara styrs av tryck och temperatur, utan också av åtkomst till metamorfa fluider. I Glen Esk finns också spår av ett fluidflöde från North Esk-förkastningen, under retrograda metamorfa förhållanden, för vilket mededfluidflödet över tid uppgår till 0.0003 – 0.0126 m3∙m-2∙år-1. Denna fluidflödeshändelse beräknas ha pågått mellan 16 000 och 334 000 år.

På ön Islay i de sydvästra högländerna återfinns bergarter, som trots sin låga metamorfa grad i klorit- eller biotitzonen innehåller mineralet kyanit, dvs. temperaturer långt under vad som vanligen associeras med kyanitbildning. Detta förklaras med infiltration av fluider med extremt hög CO2-halt, åtminstone lokalt så högt som XCO2 > 0.7, vid ca. 340°C. Fluidomvandling av dessa bergarter stabiliserade kyanit tillsammans med karbonatmineral. Syre- och kolisotopprofiler över Islayantiklinen påvisar hög kanalisering av fluider längs dess veckaxeln. Förhållandet mellan fluid och berg var mer än fyra gånger så högt i närheten av veckaxeln jämfört lokaler längre ifrån densamma. Påverkan av metakarbonatbergarters isotopförhållanden har skett längs Islayantiklinen, men fluidpåverkan kan inte ensamt förklara de isotopanomalier som observerats under och ovan Port Askaig-tilliten, varför dessa bergarter kan ha bibehållit sin primära paleoklimatologiska isotopsignatur. Nyckelord: metamorfa fluider, metamorft fluidflöde, reaktioner mellan fluid och berg, kvantifiering av metamorft fluidflöde.

3

LIST OF PAPERS This thesis is based on the following papers that are referred to in the text by Roman numeral: Paper I Lewerentz, A., Skelton, A.D.L., Linde, J.K., Nilsson, J., Möller, C., Crill, P. &

Spicuzza, M.J. (2017). On the association between veining and index mineral distributions in Barrow’s metamorphic zones, Glen Esk, Scotland. Journal of Petrology, 1-27. DOI: 10.1093/PETROLOGY/EGX039.

Paper II Lewerentz, A. & Skelton, A.D.L. (submitted to American Journal of Science).

Fluid and carbon flux estimation of regional metamorphic fluid flow in Glen Esk, SE Scottish Highlands: the role of hydrodynamic dispersion for broaden-ing of an isotopic front.

Paper III Lewerentz, A., Skelton, A.D.L., Graham, C.M., Broman, C. & Däcker, E.

(manuscript). Post-peak metamorphic kyanite stabilisation in greenschist facies metasedimentary rocks on the Isle of Islay, SW Scottish Highlands.

Paper IV Skelton, A.D.L., Lewerentz, A., Kleine, B., Webster, D. & Pitcairn, I. (2015).

Structural channelling of metamorphic fluids on Islay, Scotland: implications for paleoclimatic reconstruction. Journal of Petrology 56, 2145-2172.

Papers I-III: Alexander Lewerentz was the main contributor with assistance from the listed co-authors in conducting field work, sampling, analyses and writing for these manuscripts. Paper IV: Alexander Lewerentz mainly contributed with sampling and field observa-tions/interpretations for a portion of the sample suite used in this study, as well as participated in the flux analysis of the isotope data.

4

1. METAMORPHISM AND METAMORPHIC FLUIDS

1.1 Barrovian metamorphism1 George Barrow (1853 – 1932) was one of the pioneers in the study of metamorphic rocks, and is famous for introducing the concept of metamorphic zones, presented in Barrow (1893; 1912a). His work is based on observations from large areas of the SE Scottish Highlands, where Glen Esk, and nearby valleys Glen Lethnot and Glen Clova, served as type localities. Barrow’s key observation was a systematic change in mineralogy and texture of the metased-imentary rocks found in the area, which represents a gradual increase in metamorphic grade. Barrow (1912a) suggested that these (and similar) metasedimentary sequences may be divid-ed into six metamorphic zones, and that the boundary of each of these zones is marked by the first occurrence of its index mineral. The six Barrovian: chlorite, biotite, garnet, staurolite, kyanite and sillimanite; tend to occur as porphyroblasts and are excellent tools for quick in-field determination of the metamorphic grade.

Eskola (1915; 1920; 1929) later refined Barrow’s work by introducing the ACF and AKF chemographic diagrams, and thereby linked mineral assemblages to protolith composi-tion. Both diagrams use systems that are reduced to three components, because this is the maximum number easily represented in a two-dimensional diagram. The first step in this re-duction is grouping components together, meaning that they are treated together as one com-ponent under the assumption of equal partitioning within the group, as shown in Table 1. The second step is to project the diagram from components or phases that are in excess, in this case both the ACF and AKF diagrams are projected from quartz and H2O. Additionally, the ACF diagram is also projected from muscovite and albite, and the AKF diagram from musco-vite and plagioclase. A problem with both the ACF and AKF systems is that they ignore the effect of variability in the relative proportions of FeO and MgO, which is not appropriate for several common metamorphic minerals (e.g. chlorite, biotite, and garnet). As a solution, Thompson (1957) introduced a chemographic diagram specially constructed for metapelites: the AFM diagram, which treats FeO and MgO as separate components. This is enabled by projection from quartz, muscovite, and H2O, in combination with that Na2O, CaO, and MnO are ignored.

Table 1. Component reduction in the ACF, AKF, and AFM systems ACF AKF AFM

A = Al2O3 + Fe2O3 – (Na2O + K2O) A = Al2O3 – (CaO + Na2O + K2O) A = Al2O3 - 3K2O C = CaO K = K2O F = FeO F = FeO + MgO + MnO F = FeO + MgO + MnO M = MgO

1 This section is modifed from the author’s licentiate thesis (Lewerentz, 2015).

5

Fig. 1: Thompson (1957) A(K)FM diagrams for each of the six Barrovian zones, showing how assemblages for rocks with the composition of common pelites (shaded area) change with increased temperature, and how the six respective index minerals are introduced to these assemblages (modified after Winter, 2010). Mineral abbrevia-tions: Chl = chlorite, Cld = chloritoid, Grt = garnet, Ky = kyanite, Prl = pyrophyllite Sill = sillimanite, St = stau-rolite.

Chemographic diagrams only consider mineral assemblage stability as a function of bulk rock composition, and give limited information about changes in mineral assemblages and the reactions driving such changes. Equilibrium thermodynamics are widely used to in-

6

vestigate metamorphic reactions, a practice made possible by the compilation of thermody-namic databases containing mineral stability data (cf. Helgeson et al., 1978; Berman et al., 1985; Holland & Powell, 1985; Powell & Holland, 1985). Using such databases, it is possible to construct AFM diagrams for a series of temperatures and thus track the reactions that cause changes in mineral assemblages. For a rock of the average pelite composition, AFM diagrams may be constructed for temperatures representative of each of the six Barrovian zones, which represent the prograde part of the Barrovian metamorphic P-T path (Fig. 1). Such series of AFM diagrams also give limited information of possible reactions that occur during prograde metamorphism and introduce the Barrovian index minerals to the assemblages of an average pelite, and if these reactions are continuous or discontinuous (tie line flips).

Another application is construction of petrogenetic grids, i.e. pressure-temperature (P-T) diagrams of metamorphic reactions (cf. Albee, 1965; Hess, 1969; Thompson, 1976; Koons & Thompson, 1985; Spear & Cheney, 1989). In contrast to AFM diagrams, petrogenetic grids do not differentiate between rocks of different composition, but rather give all thermodynami-cally possible reactions for a given P-T window. Petrogenetic grids are thus powerful tools for investigation of possible P-T paths, e.g. by determination of the reactions responsible for the stabilisation of each of the six Barrovian index minerals along the prograde P-T path (Fig. 2). Petrogenetic grids may also be used to link retrograde metamorphic textures to possible reac-tions to investigate the retrograde part of the P-T path.

Fig. 2: Simplified petrogenetic grid in the K2O-FeO-MgO-Al2O3-SiO2-H2O (KFMASH) system with the biotite-in (a), garnet-in (b), staurolite-in (c), kyanite-in (d) and sillimanite-in (e) reactions marked on the typical Barro-vian P/T field gradient (modified from Winter, 2010).

Petrogenetic grids offer only limited information of which reactions will take place in a specific rock with a specific composition. To gain such information, thermodynamic data-bases may be used to construct pseudosections: these are P-T or temperature-composition (T-X) diagrams that take rock composition into account and show the stable mineral assemblages

7

for a specific rock composition over a range of P-T or T-X conditions (cf. Hensen, 1971; Powell et al, 1998). P-T pseudosections only show reactions that are thermodynamically sta-ble for a specified rock composition, and the position of these in P-T space. P-T pseudosec-tions are especially useful for investigation of metamorphic P-T paths. In the case of a rock with the composition of a typical pelite, the P-T pseudosection will therefore show: (1) stabil-ity fields of mineral assemblages, which in this special case equals P-T intervals for the Bar-rovian zones, and (2) P-T positions of index mineral reactions, which equals positions of the Barrovian isograds (Fig. 3).

T-X pseudosections are also specific for a given rock composition, and are utilised to visualise the impact rock or fluid composition has on mineral assemblage stability. These dia-grams are useful in the study of metamorphic fluids, and may for example be used to investi-gate how CO2-bearing fluids affect the mineralogy of the rocks they infiltrate.

Fig. 3: Pseudosection in the KFMASH system for a rock with the composition of a typical pelite (redrawn from Powell et al., 1998). The dashed red curve represents a hypothetical P-T path that would take a rock through each of the six Barrovian metamorphic zones. Mineral abbreviations: Chl = chlorite, Bt = biotite, Grt = garnet, St = staurolite, Ky = kyanite, Sil = sillimanite, And = andalusite, Ms = muscovite and Qtz = quartz.

8

1.2 Metamorphic fluids Metamorphic fluids are released from the crust by devolatilisation reactions when temperature increases as a result of orogenesis. They originate from H- and C-bearing phases that gradual-ly become unstable due to increased metamorphic grade. The main constituents are H2O and CO2, but small amounts of methane and dissolved salts (e.g. Na+, K+, and Ca2+) may also be present (Philpotts & Ague, 2009 and references therein). Metamorphic fluids are widely rec-ognised as a main driving force for metamorphic reactions. During metamorphic reactions, the reactants and products of these reactions are transported by diffusion, and diffusion rate is the chief kinetic limit on these reactions. The rate of diffusion when fully assisted by fluid is over 14 orders of magnitude faster than diffusion through silicate phases only. Hence, fluids make metamorphic reactions considerably faster (Ridley & Thompson, 1986).

Metamorphic fluids are a relatively recent discovery, and were proposed in the late 1950s and early 1960s (e.g. Greenwood, 1961; 1962). Metasomatism, the alteration of one phase into another but retaining the original volume, was one of the first studied fluid driven metamorphic processes (Korzhinskii, 1957), for which Thompson (1959) concluded that mass transfer is required and proposed metamorphic fluids as transport medium.

1.3 Mechanisms for fluid flow Metamorphic fluids are buoyant and therefore rise through the crust (Wood & Walther, 1986). Metamorphic fluid flow is either pervasive or channelled. Pervasive fluid flow affects the entire rock volume and is controlled by microstructures created during metamorphism (Ferry, 1994), whereas channelled fluid flow only affects rocks alongside preferred fluid pathways (Ague, 2003).

Pervasive fluid flow takes place on the scale of mineral grains. Microstrucutral con-trols on pervasive fluid flow include dihedral angles (the ‘wetting’ angle), mineral anisotropy, and microfractures. The geometry of fluid-filled pores is controlled by the fluid-solid dihedral angle (von Bargen & Waff, 1986), which is a function of the grain boundary energy and the fluid-solid interface energy (Smith, 1948). At dihedral angles >60°, the fluid in an isotropic mineral aggregate forms isolated pores and the porosity network is no longer connected, whereas an interconnected porosity network is stable at angles <60° (Bulau et al., 1979). The dihedral angle is generally smaller at lower temperatures, and reaches 60° at temperatures of ~450°C in a pure H2O fluid system (Holness, 1993). Yet, fluid flow is observed in metamor-phic rocks of considerably higher grade than this, for which additional controls have been proposed. Mineral anisotropy, such as flat surfaces and sharp corners on crystals, may in-crease the permeability of a mineral aggregate (Waff & Faul, 1992), to create an interconnect-ed porosity networks at higher temperatures. Coupled dissolution-precipitation is a process that creates porosity in mineral grains as fluid gradually penetrates the grains in a coupled dissolution-precipitation reaction, which may allow for fluid flow where no interconnected porosity network is available (Putnis & Putnis, 2007; Putnis & Ruiz-Agudo, 2013). Defor-mation may produce a network of evenly distributed and tightly spaced microfractures, which is utilised for pervasive fluid flow where porosity networks are non-existent (Ferry, 1994 and references therein).

9

Fig. 4: A summary of the different processes that transport metamorphic fluids in a channelled fluid flow setting (from Ague, 2003). Diffusive and mechanical dispersion dominate a minor flux component of cross layering flow, and is induced either by devolatilisation reactions or selvedge formation. Fluids that originate from these processes are channelled into high flux zones, either controlled by permeability contrasts as in the case of layer parallel flow at lithology boundaries or other permeable horizons, or in fractures (veining) which may be either parallel or crosscutting to layering.

10

Channelled fluid flow takes place in preferred fluid pathways, such as permeable li-thologies, shear zones, fold hinges and fractures (Fig. 4). Bickle & Baker (1990) used oxygen isotope profiles from alternating layers of schists and marbles to conclude that fluids were channelled within the relatively more permeable metapelites, and that fluid flow was thus controlled by permeability contrasts. Channelization of fluids in shear zones is widely report-ed, for example from granulite facies rocks in Norway (Austrheim, 1987) and blueschist to greenschist facies rocks on Syros in the Greek Cyclades (Pogge von Strandmann et al., 2015). Channelization of fluids in fold hinges is reported from greenschist facies rocks in the SW Scottish Highlands (Skelton et al., 1995; Pitcairn et al., 2010). Fracture-channelled fluid flow is for example reported from Barrovian sequences in New England, USA (Ague, 1994b; Whitney et al., 1996; Ague, 2011), at Stonehaven, SE Scottish Highlands (Ague, 1997; Mas-ters et al., 2000; Masters & Ague, 2005) and on the Shetland Islands, Scotland (Bucholz & Ague, 2010), as well as from accretionary prism-related metasedimentary rocks in a subduc-tion setting on New Zeeland (Breeding & Ague, 2002), eclogite facies rocks in the Swiss Alps (Widmer & Thompson, 2001), and blueschist facies rocks in the Greek Cyclades (Kleine et al., 2016b).

1.4 Veins Metamorphic fluids are somewhat problematic to study, as metamorphic fluid flow cannot be directly observed. The most direct approach to study metamorphic fluid flow is observation of fluid inclusions, i.e. μm-sized pockets of fluids and vapour trapped in solid mineral phases, which to some extent are representative of the original fluid composition. Other approaches almost entirely use indirect inference from fluid-driven reactions (fluid-rock interaction) or fluid precipitates.

Veins are relict fluid pathways, which form from fracture-channelled metamorphic fluid flow commonly. Veins consist of material (mainly quartz and/or calcite) that precipitat-ed from the fluid in the fracture. The morphology and petrography of veins depends on the conditions at and timing of vein formation, and can thus be used to assess the mode of fluid flow as classified by Bons et al. (2012). According to this classification scheme, veins may be subdivided into three groups based on growth plane localisation and the number of crack-seal events: antitaxial veins, syntaxial veins, and stretching veins. Antitaxial veins are initiated by fracturing (no crack-seal is needed) and grow in two directions from the centre of the vein towards the wall rock, commonly hosting crystals of fibrous habitus. Syntaxial veins grow inwards from one or both sides of the wall, host crystal of blocky habitus which also may be elongated, and have endured one or multiple crack-seal events. Stretching veins share most properties with syntaxial veins. In contrast they form cracks that crosscut previously precipi-tated material or wall rock and thus require multiple crack-seal events, and host stretched crystals and wall rock array inclusions. Syntaxial veins and stretching veins are both indica-tive of advective fluid flow (Bons et al., 2012 and references therein).

Quartz veins form from fluid-dissolved silica that may be derived from both local and external sources (Rubenach, 2013). Yardley & Bottrell (1992) reported that there was not or very little difference between vein and host rock oxygen isotope ratios, and interpreted this as evidence for locally sourced silica, or at least from within the same formation. Skelton et al (2015) reported similar vein-host rock relationships from the Isle of Islay, Scotland. In con-

11

trast, mass balance calculations of vein-host rock relationships in New England (USA) sug-gest that from 20-60% up to 90% of vein silica is externally derived (Penniston-Dorland & Ferry, 2008; Ague, 2011).

1.5 Fluid-rock interaction Metamorphic fluid flow cannot be studied directly; however, past fluid events may be inferred from evidence of fluid-rock interaction, i.e. alteration of rock infiltrated by fluids. Fluid-rock interaction takes place where fluid flow is pervasive, mostly along preferred pathways for channelized fluid flow. Fluid-rock interaction affects the texture, mineralogy, chemical com-position and isotopic composition of the rock. This is mostly seen as altered zones adjacent to preferred fluid pathways, but fluid-rock interaction may also serve as a buffer to preserve mineral assemblages outside their thermodynamically favourable P-T conditions. Carbon, and especially oxygen, stable isotope ratios are useful tools to fingerprint metamorphic fluid flow, as a binary H2O-CO2 fluid will alter both of these during fluid-rock interaction.

Bickle & Baker (1990) reported fluid-rock interaction along marble-schist contacts on Naxos, Greece, which altered the isotopic composition of the marble to create isotope fronts away from such lithological boundaries. Austrheim (1987) showed that fluid-rock interaction along shear zones in southwestern Norway buffered eclogite facies recrystallisation of granu-lites. In the SW Scottish Higlands, Pitcairn et al. (2010) showed that fluid-rock interaction along a region scale anticline caused complete resetting of the isotope signature within the axial area of this fold. Kleine et al. (2016b) demonstrated that fluid-rock interaction along fractures served as a buffer to preserve blueschist facies mineral assemblages at greenschist facies conditions.

Fluid-rock interaction along fractures (veins) are more commonly reported to result in altered zones (selvedges) whose width depends on the pervasiveness of fluid flow. Selvedge formation yields textural, mineralogical, chemical and isotopic alteration of vein-adjacent rock. Vein-adjacent selvedges are reported from Barrovian sequences at several localities worldwide. In New England, USA, garnet, staurolite and kyanite porphyroblasts are reported to be larger in size and more abundant in selvedges, whereas quartz, plagioclase and mica are less abundant than in surrounding wallrock (Fig. 5; Ague, 1994a; 1994b; 2003, 2011). These mineralogical changes are coupled with alteration of their chemical compositions, seen as gain of Li, Al, Mn, Fe, Zn, Y, and HREE, as well as loss of Na, Si, K, Sr, Ba, Sn, Eu, Pb and volatiles. Much like in New England, Bucholz & Ague (2010) reported increased kyanite crystal size and abundance in selvedges on the Shetland Islands in Scotland, coupled with mass gain of Al and REE, but loss of Si, Na, K, Ca, Rb, Sr, Cs, Ba and volaties. At Stonehaven in the SE Scottish Highlands, Ague (1997) and Masters & Ague (2005) observed increased garnet and staurolite modes and decreased muscovite modes in selvedges, which was coupled with mass gain of Na, Ca and Sr, but loss of K, Rb and Ba.

12

Fig. 5: (a) Example of a selvedge from the Wepawaug schist, New England, USA, where garnet, staurolite and kyanite porphyroblast are considerably larger in size adjacent to a vein. (b) Two plots of maximum kyanite length against distance from a vein shows that kyanite crystals gradually decrease in size away from the vein. Compiled from Ague (2011).

Selvedge formation also alters the oxygen isotope ratios of veins-adjacent rock and its constituent minerals, as the rock equilibrates with the fluid upon infiltration. If fluid-rock equilibration is complete, no gradients are observed in the vein-adjacent rock. If fluid-rock equilibration, on the other hand, is uncomplete, an isotope gradient is observed with the most fluid-like signatures seen adjacent to the vein, which gradually becomes more rock-like far-ther from the vein. Young & Rumble (1993) reported vein-adjacent garnet crystals overgrown by rims 0.8±0.5‰ lower in �18O than their cores, and attributed this to the passage of meta-morphic fluids. van Haren et al. (1996) found that selvedge-hosted garnet rims were ~1‰ higher in �18O compared to garnet rims outside selvedges.

1.6 Quantification of metamorphic fluid fluxes Over the years, different methods have been used to quantify the amount of fluid that takes part in fluid-rock interactions. The zero-dimensional fluid:rock (F:R) ratio is one of the sim-pler approaches. Taylor (1977) used oxygen isotope ratios to construct a volumetric model, the water-rock ratio (W:R):

13

where δ is the initial and final concentration of the tracer species. The W:R ratio may be used for evaluation and quantification of fluid driven reactions. The approach yields the relative proportions of fluid and rock that were needed to change the tracer species concentration of a rock into that observed in its altered state. This is made possible by that if an external fluid reacts with the surrounding rock, the isotopic signature of the rock will change and equilibrate towards that of the fluid. The degree of equilibration is a direct function of the degree of fluid-rock interaction. Several alternatives for calculation of F:R ratios have been proposed over the years, e.g. Ferry (1987) who proposed a model based on mass transfer of CO2 by which the W:R ratio is calculated through the relative proportion of fluid that was needed to carbonate an originally carbonate-free rock.

One-dimensional fluid fluxes calculate the amount of fluid that flowed along a line perpendicular away from a structural, lithological or other chosen boundary, either as a value integrated over the entire duration of fluid flow (time-integrated fluid flux) or averaged over the duration of fluid flow (time-averaged fluid flux). Bickle & Baker (1990) observed oxygen isotope displacement fronts away from marble-schist contacts and interpreted this as fluid flow into the marble. They applied advective-diffusive transport modelling to these displace-ment fronts to calculate time-integrated fluid fluxes (��t):

where � is the fluid velocity, � is the porosity, t is time, z1 is the distance of the geochemical front, �s is the density of the solid phase, �f is the density of the fluid phase, and KD is the sol-id-fluid partition coefficient. As an alternative approach, Ague (1994; 2003; 2011) used vein-selvedge-wall rock mass balance calculations to estimate one-dimensional fluid fluxes during Barrovian-type metamorphism.

Three-dimensional fluid fluxes provide more realistic flux estimates in cases of com-plex structural controls on fluid flow, such as folds. Such fluxes utilise geometrical solutions to combine several one-dimensional fluid flux estimates into a single regionally averaged flu-id flux. As an example, Skelton et al. (1995) used mass-balance calculations to estimate one-dimensional fluid fluxes for carbonation of a number of differently orientated metabasalt sills, geographically spread out over the area of a regional fold structure, and treated these as vec-tors in a geometrical 3D solution in order to calculate the regional upward fluid flux within the fold structure.

Metamorphic fluid flow is not homogenous, but as described above channelized into high-flux zones. This is also reflected in the high variable flux rate estimates available in the existing literature. Time-integrated background values for non-channelized pervasive fluid flow within permeable lithologies are ~102-103 m3m-2, whereas channelized fluid flow flux rates may reach above 106 m3m-2 (Fig. 6; Ague, 2014 and references therein). Because of this, it is important to establish the local and regional contexts for a fluid flux measurement, oth-erwise it remains unknown on what scale the calculated number is a representative value.

14

Fig. 6: Time-integrated fluid flux rates for a number of structural, geological and geographical contexts (from Ague, 2014). For the purpose of this thesis, the comparison between (p) the regional average for metamorphism and (l) individual veins in Scotland, is especially interesting.

15

2 GEOLOGICAL SETTING

2.1 The Dalradian of Scotland The Dalradian Supergroup comprise mid-Neoproterozoic to Early Palaeozoic rocks of sedi-mentary origin, and is situated between the Highland Boundary and Great Glen Faults (Fig. 7; Stephenson et al., 2012). The Dalradian Supergroup is divided into four subgroups, in chrono-logical order: the Grampian, Appin, Argyll and Souther Highland Groups. The areas studied in this thesis expose rocks from the Appin and Argyll Groups (Islay), as well as the Southern Highland Group (Glen Esk), see below for further descriptions of these areas. Sedimentation occurred between ca. 730 Ma and 515 Ma (Tanner & Pringle, 1999; Tanner & Sutherland, 2007). The Appin Group rocks originate from shelf sediments and main lithologies include metapelites, -semipelites, and -psammites, as well as quartzites, calcsilicate rocks, metalime-stones and metadolostones (Anderton, 1985; Wright, 1988). The base of the Argyll Group is marked by diamictites that are most prominently exposed in the SW Highlands on Islay and the Garvellachs (Thomson, 1871; Spencer, 1971), but also found throughout the Grampian Highlands (Stephenson et al., 2012). These diamictites are overlain by metadolostones, quartzites, as well as basin- or basin margin-derived clastic metasedimentary rocks. The Southern Highland Group mainly comprises turbidite-related metasedimentary rocks that vary from pelitic to psammitic in composition (Stephenson et al., 2012).

Fig. 7: Geographical extent of the Dalradian Supergroup and its subunits (from Carty et al., 2012; 2014).

16

2.2 Study area 1: Glen Esk2 Glen Esk is one of Barrow’s type localities, and comprises all six metamorphic zones (Bar-row, 1893; 1912a; 1912b; Harker, 1912). Glen Esk hosts rocks of four distinct origins (Fig. 8), from south to north: (1) unmetamorphosed Siluro-Devonian sedimentary (mainly sand-stone) and volcanic rocks of the Old Red Sandstone Supergroup, (2) Ordovician metasedi-mentary rocks of the Margie and North Esk Formations within the Highland Border Complex, (3) Neoproterozoic metasedimentary rocks of the Glen Lethnot Grit and Glen Effock Schist Formations within the Dalradian Supergroup, as well as (4) the post-metamorphic Kincar-dine/Mount Battock granite (406 ±18 Ma, Oliver et al., 2008). Metamorphism in Glen Esk took place during the Grampian phase of the Caledonian orogeny, ca. 475-465 Ma (Chew & Strachan, 2014).

The Barrovian sequence of Glen Esk is studied in Papers I and II. This sequence com-prises rocks of the Dalradian Supergroups, and begins with the chlorite zone at the North Esk Fault, the boundary towards the Highland Border Complex, and ends where the sillimanite zone is cut by post-metamorphic granites some 8 km to the north. Typical mineral assemblag-es, all of which include muscovite and quartz, are chlorite, chlorite+biotite, chlorite+ +biotite+garnet, biotite+garnet+staurolite±chlorite, biotite+staurolite+kyanite±garnet, and biotite+sillimanite±K-feldspar±garnet±staurolite±kyanite (Barrow, 1893; Barrow, 1912a; Harte & Hudson, 1979; Harte, 1987; 2016). More detailed descriptions of the geology of Glen Esk are given in Papers I and II.

2.3 Study area 2: Islay In the eastern half of the Isle of Islay in the SW Scottish Highlands (Fig. 9), rocks of the Dalradian Supergroup of Neoproterozoic age (Halliday et al., 1989) are exposed. These are metamorphosed sandstones, mudstones and carbonate rocks of the Appin and Argyll Groups (Wilkinson, 1907; Bailey, 1917; Allison, 1933). Appin Group rocks are mainly phyllitic, and in places graphitic, metamudstones and carbonate rocks of the Kintra Dolostone, Ballygrant and Lossit Limestone Formations. These are overlain by the Argyll Group, initially by dia-mictites of the Port Askaig Tillite Formation (Thomson, 1871), and further up in the strata by metacarbonates of the Bonahaven Dolomite formation, metasandstones of the Jura Quartzite Formation, and phyllites of the Port Ellen Phyllite Formation within the Easdale Subgroup.

These rocks underwent prograde greenschist to epidote-amphibolite facies metamor-phism during the Grampian orogeny, at ca. 475-465 Ma (Chew & Strachan, 2014). Primary (D1/D2) deformation included regional-scale recumbent folding, which lead to the formation of the Islay Anticline (Roberts & Treagus, 1979). Secondary crenulations of D1/D2 fabrics are associated with later (D4), which occurred during uplift of the metamorphic complex. Peak metamorphism occurred between the D1/D2 and D4 phases (Graham et al., 1983). Fluid infiltration has occurred during two main events: (1) pre-D4 in a post-peak metamorphic set-ting at ~340°C (Kleine et al., 2015), and (2) post-D4 in a retrograde metamorphic setting at considerably lower temperatures (Fein et al., 1994). Both of these events lead to carbonation of the infiltrated rocks.

2 This section is modified from the author’s licentiate thesis (Lewerentz, 2015).

17

Fig. 8: Map of the field area in Glen Esk, SE Scottish Highlands. The main rock groups and formations, struc-tural boundaries, isograds are indicated (modified from Paper I; Harte, 1987; Tanner et al. 2013). Syn-metamorphic intrusions include the Morven-Cabrach (MCG), Boganclogh (BCG), Insch (IG), and Belhelvie (BG) gabbros, as well as the Aberdeen granite (AG).

The rocks around Kilnaughton Bay are studied in Paper III (Fig. 9), which include quartzites of the Jura Quartzite Formation and thin slivers of highly aluminous whiteschists and graphitic schists within this Formation, as well as calcareous schists of the Port Ellen Phyllite Formation. In Paper IV, metalimestones and metadolostones of the Kintra Dolostone, Ballygrant, Lossit Limestone and Bonahaven Dolomite Formatinos were studied along the full stretch of the Islay Anticline (Fig. 9). More detailed descriptions of the studied lithologies are given in the respective Papers III and IV.

18

Fig. 9: (a) General geology of the Isle of Islay, with the study area around Kilnaughton Bay for Paper III marked out in the southern-most part of the island. Samples for Paper IV were collected along the full stretch of the Islay Anticline. (b) Generalised stratigraphic column for Islay, in which the rocks studied in this thesis belong to the Appin and Argyll groups (modified after Skelton et al., 2015).

19

3. METHODS The different employed methods are briefly summarised below. For further details, see the respective Papers I-IV.

3.1 Paper I Paper I is based on studies of the metasedimentary rocks in Barrow’s type locality; Glen Esk. Outcrops in each of the six metamorphic zones were studied in field and sampled. Petrograph-ic descriptions and point counting was conducted using thin sections for polarisation micros-copy.

Whole rock compositions were determined by analysis of rock powder lithium-metaborate-tetraborate fused glass discs, using X-ray fluorescence (XRF) spectrometry for major elements, and by laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) for trace elements. Volatile component content was estimated as loss on ignition (LOI). Chemical compositions of individual mineral phases were determined by spot analysis (1-5 μm), using an electron microprobe analyser (EMPA). Stable oxygen isotope ratios were determined for hand-picked quartz and kyanite grains, using laser fluorination isotope ratio mass spectrometry (IRMS).

Equilibrium thermodynamic calculations were employed to construct pressure-temperature pseudosections, which were used to illustrate the how mineral stabilities differ between vein-adjacent selvedges and vein-distal rock.

3.2 Paper II Samples of syn-metamorphic veins and their host rocks were collected along a profile away from the Highland Boundary and North Esk Faults.

Whole rock total carbon content was determined for a suite of samples, using catalyti-cally aided combustion oxidation at 900°C and a Total Organic Carbon (TOC) analyser. Chemical compositions of individual mineral phases were determined by EMPA analysis. Stable isotope oxygen analyses were conducted using the same method as for Paper I.

One-dimensional fluid fluxes were calculated using (1) advection-dispersion model-ling of vein oxygen isotope data, and (2) advection-diffusion of the wallrock oxygen isotope and total carbon content data. Advection-dispersion modelling (1) is achieved analytically by the assumption that advection displaces and hydrodynamic dispersion broadens isotopic transport along fractures. Hydrodynamic dispersion is a combination of chemical diffusion (Dc) and mechanical dispersion (Dm), and results in variability of how fast isotopic transport is in different fractures (Fig. 10). Advection-diffusion modelling (2) is achieved analytically by the assumption of pure advective isotopic transport along parallel and evenly spaced fractures, which yields transverse diffusion in a pore fluid away from these fractures (Fig. 11).

20

Fig. 10: Schematic illustration of the advection-dispersion model employed in this study. A tracer (in this case oxygen isotopes) is transported by fluids in randomly spaced fractures of varying lengths and widths. Mechani-cal dispersion (Dm; i.e. variation in path-width, tortuosity and other factors) results in isotopic transport that is faster in some vein-pathways than others, which is illustrated by the difference between the shortest and the furthest travelled pathway. Chemical diffusion (Dc) and Dm combined equals the hydrodynamic dispersive com-ponent of isotopic front movement (figure and caption from Paper II).

Fig. 11: Schematic illustration of the model used for flux estimations based on the wall rock oxygen isotope and carbon content data (modified from Skelton et al., 2000). (a) Fluids infiltrate from the North Esk Fault and are transported purely by advection in equally wide and evenly spaced fractures, from which transverse diffusion changes the tracer ratio/concentration (δ18O/wt.% C) of the wall rock. (b) At the end of fluid flow, these altera-tions are locked into the wall rock, and the degree of tracer alteration generally decreases with distance from the fluid source (figure and caption from Paper II).

21

3.3 Paper III For Paper III, field relations were studied and samples collected of whiteschists and co-occurring lithologies, which are situated around Kilnaughton Bay, southern Islay. Additional samples of graphitic phyllites were collected along the Islay Anticline. Mineral assemblages were determined in thin section using petrographic microscopy.

Whole rock compositions and LOI were determined from whole rock powders, using the same methods as for Paper I. Chemical compositions of individual mineral phases were determined by EMPA analysis. Carbon and oxygen stable isotope ratios of carbonate phases were determined using IRMS. Raman spectrometry was used to determine the crystallinity of graphite, which is used as a geothermometer.

Temperature-XCO2 pseudosections were constructed to illustrate how increased CO2-content in the fluid component may change the stability fields of mineral phases and assem-blages.

3.4 Paper IV Metacarbonate rocks were sampled along the Islay Anticline and adjacent areas for petro-graphic analysis of veins and for stable isotopic analysis of veins and wallrock. Petrography was studied in the field, in polished hand specimen and in thin section, and veins classified according to Bons et al. (2012). Samples for C and O stable isotope analysis were collected in cm, m and km scale profiles to assess the scale of isotopic homogenisation.

Whole rock carbon and oxygen stable isotope ratios of metacarbonates were deter-mined using IRMS.

Fluid:rock mixing lines were calculated to estimate the amount of metamorphic fluids that interacted with the rocks along the Islay Anticline. This was achieved for both the carbon and the oxygen data, following the approach of Taylor (1977):

where K is the rock:fluid partition coefficient.

22

4. SUMMARY OF PAPERS

4.1 Paper I3 As mentioned in the introduction of this thesis, the occurrence of Barrovian index minerals are largely controlled by pressure, temperature and bulk rock composition. However, in addi-tion to metamorphic grade, the occurrence and size of metamorphic minerals are controlled also by metamorphic fluids. In Paper I, the hypothesis is tested that channelled fluid flow in fractures has systematically changed the distribution of index minerals in the Barrovian se-quence of Glen Esk.

In the chlorite zone, chlorite is present in all collected samples in abundances varying from 23 to 36 vol%, and homogeneously distributed within the rocks (Fig. 12a). Vein density is <1% and the distance between veins is generally several meters or more, which results in only a small proportion of the rock mass being affected by fluid-rock interaction. In addition to this, oxygen isotope data show no signs of equilibration between fluid and country rock (Fig. 13a; see manuscript for details). It is interpreted that the distribution of chlorite as an index mineral is controlled by pressure, temperature, and protolith composition (P-T-x).

Most of the biotite zone is similar to the chlorite zone in terms of index mineral distri-bution, i.e. biotite size distribution is overall homogenous (Fig. 12a; Fig. 13b). However, one of the sampled localities is distinctively different than the rest of the biotite zone (Fig. 12b). This locality comprises a fold hinge which is intensely veined, with vein densities that reaches above 50% if only accounting for the area closest to the hinge. An estimation of biotite crystal size distribution that was conducted in the field shows the biotite crystals are largest at the fold hinge and gradual decrease away from the hinge on both limbs. This indicates a locally high degree of fluid-rock interaction. In all, the distribution of biotite as an index mineral is interpreted to be P-T-x controlled in most cases. For the fold hinge locality, crystal size is however fluid controlled by selvedge formation.

In the garnet zone the actual index mineral, garnet, is extremely scarce and most rocks in the garnet zone do not host any garnet. Where found, garnet is restricted to thin metapelite layers or rock adjacent to veins (Fig. 12c). Adjacent to veins, porphyroblast size is also larger. Mineral point counting shows that garnet abundance increases towards the vein, while quartz mode decreases. These trends are also coincident with a decrease in the oxygen isotope signa-ture of quartz, which is indicative of fluid-rock interaction (Fig. 13c). The concentration of Mn is an order of magnitude higher in the selvedge compared to the unaltered country rock, which is interesting as Mn shifts the garnet stability field towards lowers temperatures. In all, and including trace element data that are available in the attached manuscript, this is interpret-ed as selvedge formation, and thus distribution of garnet as an index mineral is mainly fluid controlled as opposed to the general view of a P-T-x control.

The staurolite zone is very similar to the garnet zone, and staurolite is restricted to a few intensely veined localities (Fig. 12d). At these localities, staurolite mainly occurs within cm wide zones adjacent to veins. Point counting shows that staurolite mode decreases away from

3 This section is modified from the author’s licentiate thesis (Lewerentz, 2015).

23

these veins, coincident with an increase in quartz mode. Selvedge formation, analogous to the garnet zone, is interpreted as the main control on distribution of staurolite as an index mineral.

Fig. 12: (a) Index mineral distribution in the chlorite and most of the biotite zone, (b) at Locality 05 in the biotite zone, (c) in the garnet zone, (e) in the staurolite zone, and (f) porphyroblast distribution in the sillimanite zone (figure from Paper I).

24

Fig. 13: Quartz oxygen isotope data from the chlorite (a) and biotite (b) zones, showing a random relationship between veins and vein-adjacent host rock. (c) Quartz oxygen isotope data from locality 09 in the garnet zone, showing a decrease in δ18O and quartz mode towards the vein. The p-value with which the null hypothesis (ran-dom variation about a constant value) can be rejected as well as the goodness-of-fit (expressed by the parameter R2) are shown. (d) Oxygen isotope profile for vein quartz and kyanite in the country rock from the kyanite zone, showing no variation after correction for Qtz-Ky fractionation (figure and caption from Paper I).

25

Kyanite distribution in the kyanite zone is in contrast to the garnet and staurolite zones rather homogenous (Fig. 12e). Vein density is around 13%, in comparison to 4% and 7% in the garnet and staurolite zones, respectively. This is indicative of even higher fluid-rock inter-action than in the two latter zones. Oxygen isotope data show that the country rock is com-pletely equilibrated with the fluid (Fig. 13d). Altogether, selvedge formation is again attribut-ed as the main control on index mineral distribution, albeit in the case of the kyanite zone, selvedges are wider and more narrowly spaced and thus overlapping.

In the sillimanite zone, distribution seems controlled more in the manner of the garnet and staurolite zones, and sillimanite is preferentially found in proximity to veins. Other por-phyroblasts, however, are more homogeneously distributed (Fig. 12f), which indicates that wide selvedges were formed. All vein-proximal samples from the sillimanite zone show mass loss of Si, which is consistent with the interpretation of selvedge formation. As an implica-tion, sillimanite distribution cannot be controlled by the same mechanism as all other porphy-roblasts. A possible explanation for this is that sillimanite formed during a later heating event that only affected the sillimanite zone, which previously has been argued by Chinner (1961; 1966) and Vorhies & Ague (2011).

Selvedge formation, i.e. removal of Si by fluids and relative enrichment of AFM com-ponents, is interpreted as the major control on index mineral distributions in the garnet, stau-rolite and kyanite zones in Glen Esk. The implication of this that the occurrence, distribution and size of Barrovian index mineral porphyroblasts, are not only controlled by metamorphic grade and rock composition, but also by access to metamorphic fluids.

4.2 Paper II In Paper II, advective-dispersive and advective-diffusive modelling is used to estimate fluid fluxes from regional oxygen isotope and carbonation fronts in Glen Esk. These also evaluate the importance of hydrodynamic dispersion along fractures and wall-rock diffusion for broad-ening of isotopic fronts.

Oxygen isotope and total carbon analyses of samples of Glen Esk yield three datasets: (1) �18O of veins, (2) �18O of wallrock, and (3) wt.% carbon of the wallrock. These are inde-pendently used to estimate three fluid fluxes: (1) ��t = 300 m3∙m-2, ��t = 2500 � 1500 m3∙m-

2, and (3) ��t = 135 � 22 – 172 � 29 m3∙m-2. Flux (1) and (3) are within the same order of magnitude, whereas flux (2) is one order of magnitude higher. The model used for flux (1) takes hydrodynamic dispersion into account, whereas the model used for (2) and (3) does not. The carbonation front used to calculate flux (3) has such short a length scale that hydrody-namic dispersion is assumed to be very marginal. Fluxes (1) and (2) show similar numbers which demonstrate that hydrodynamic dispersion played an important role in broadening of the isotopic front. It also shows that hydrodynamic dispersion may not be ignored when mod-elling isotopic fronts, at least not on the length scales of isotopic displacement in Glen Esk.

Furthermore, the duration of the fluid event was estimated to t = 16 – 334 kyr. Using this timescale, and ��t = 135 � 22 – 172 � 29 m, a time-averaged fluid flux of 0.0003 – 0.0126 m3∙m-2∙yr-1 is estimated.

26

4.3 Paper III Kyanite is generally an indicator of higher-grade Barrovian metamorphism, and typically oc-curs in metasedimentary rocks that have reached conditions of roughly ~6 kbar and ~600°C and above (Barrow, 1893; 1912; Baker, 1985; Dempster, 1985). On the Isle of Islay, kyanite is present in chlorite-free calcareous whiteschists of chlorite or lower biotite metamorphic grade. To investigate the mechanism behind kyanite stabilisation at such low temperatures, two possible hypotheses were explored: (1) bulk rock compositionally controlled stabilisation at peak metamorphic conditions, and (2) CO2-buffered stabilisation by fluids in a post-peak metamorphic setting.

In southern Islay, kyanite occurs within calc-phyllitic slivers of whiteschists within the psammites of the Jura Quartzite Formation. Some samples also contain chloritoid instead of kyanite. Most samples lack both kyanite and chloritoid, and contain the general assemblage of quartz + muscovite ± paragonite ± siderite ± dolomite-ankerite + rutile. Graphite is also ob-served within a few samples.

The whiteschists are very aluminous rocks. A temperature-XCO2 pseudosection, cal-culated for a compositions representative of the whiteschists, show that kyanite is a stable phase at CO2-free conditions above ~420°C (at 10 kbar; Skelton et al., 1995). Kyanite would in this case be accompanied by chlorite, chloritoid and epidote (Fig. 14). Increased XCO2 increases kyanite stability towards lower temperatures in assemblages with carbonate miner-als.

Two mechanisms for the stabilisation of kyanite may be hypothesised: (1) pressure, temperature and composition controlled stabilisation at peak-metamorphic temperatures, and (2) XCO2-buffered stabilisation during the main fluid event, which occurred at ~340°C in a post-peak metamorphic setting (Kleine et al., 2015).

(1) Kyanite and chloritoid are reported from similar low-temperature metamorphic rocks; however, always in assemblages with chlorite (Cruickshank & Ghent, 1978; Ashworth & Evirgen, 1984; Manby, 1983; Paradis et al., 1983; Ghent et al., 1989), which is also indi-cated by Fig. 14 for the whiteschists of Islay. Neither chlorite nor the other phases in this as-semblage is observed in these rocks, a pure composition controlled stabilisation of kyanite is therefore deemed unlikely.

(2) The observed kyanite-bearing assemblage is stable at higher XCO2 (Fig. 14), which points towards CO2-buffered stabilisation of kyanite. Kleine et al. (2015) showed that the main fluid infiltration event in the area occurred in a post-peak metamorphic setting at ca. 340°C. Imposing this temperature on the calculated XCO2-realtionships (Fig. 14), the fluid composition was highly carbonic with XCO2 above ~0.7, at least on a local scale. This is the favoured mechanism for kyanite stabilisation.

27

Fig. 14: T-XCO2 pseudosection for sample AL15CF-XRF13, a kyanite-bearing whiteschist from Carraig Fhada (figure and caption from Paper III). At CO2-free conditions, kyanite is stable above ~420°C. Kyanite stability increases with increased T-XCO2 and the observed kyanite-bearing assemblage (in bold) is stable at XCO2 > 0.7 at 340°C, the main fluid infiltration event in the area (Kleine et al., 2015).

4.4 Paper IV Water is the main constituent of metamorphic fluids (Ferry & Baumgartner, 1987) and δ18O is therefore the most readily altered isotope ratio (of the two), whereas δ13C values only are sig-nificantly altered if fluid quantatites are large (Bickle & Baker, 1990). Because of this, δ13C signatures are often inferred to have withstood later metamorphic fluid alteration, i.e. that primary sedimentary values are preserved, and thereby usable for palaeoclimatologocial re-constructions. Paper IV aims to investigate evidence for structurally channelled metamorphic fluid flow along the Islay Anticline, and to evaluate how such fluid flow may have modified carbon and oxygen stable isotope ratios of metacarbonate rocks alongside this anticline. The study also aims to analytically lift off any metamorphic overprint from the δ13C ratios of these rocks, in order to see if these ratios hold any primary palaeoclimatologically significant signa-

28

tures. This is achieved by sampling and stable isotope analysis of metacarbonate rocks along cm, m, and km scale profiles across the Islay Anticline.

Stable isotope analyses show that δ18O values are typically 2-3‰ lower in the axial region of the Islay Anticline. In central Islay, δ13C values are 4-8‰ lower within the axial region. However, this km-scale trend is not seen in the northern parts of the island, where δ13C values vary extremely little over distance from the fold axis. This points towards struc-tural channelling of metamorphic fluids through the axial region, which gradually altered and lowered the isotopic signature of the surrounding lithologies, which is also supported by that higher vein densities (up to 50%) are observed within the axial region.

Metre-scaled isotope profiles across carbonate layers show little variation within the axial area, whereas profiles farther away show ~10‰ variation within the layer. Similarly, no difference is seen between veins and country rock within the axial region, whereas a differ-ence is observed farther away. This indicates that fluid (veins) and country rock fully equili-brated within the axial area, whereas fluid and country rock was out of equilibrium farther away from the Islay Anticline.

Calculation of fluid:rock ratios (see above; Taylor, 1977) yield > 30:1 within this axial region, which is at least four times greater than the regional mean ratio of 7.6 ± 1.5:1 for car-bonate rocks on Islay. This supports the interpretation that metamorphic fluids were chan-nelled through the axial region of the Islay Anticline. Using fluid-rock mixing lines, the met-amorphic fluid was calculated to have δ18O = 15.3‰ and δ13C = –6.1‰, as well as XCO2 = 0.2.

Removal of the effects of metamorphic fluid flow on δ13C values recorded by meta-carbonate rocks shows that metamorphism contributed to but cannot have created the large negative δ13C anomalies seen within the Lossit Limestone and Bonnahaven Dolomite For-mations. As an implication, palaeoclimatological interpretations drawn from these isotope signatures cannot be refuted as an effect of later metamorphic fluid alteration.

Based on the observations and analytical data, as well as interpretations of these, it is concluded that: � Channelling of isotopically light metamorphic fluids produced vein networks within en

echelon axial regions of the Islay Anticline fold system. � These metamorphic fluids caused extensive modification (lowering) of δ18O and more

locally δ13C values within the axial regions of this fold system. � Fluid:rock ratios within the axial regions of the fold system exceeded 30:1. This is at least

four times larger than the regional mean of 7.6 ± 1.5:1. Inferred axial-planar channelling of metamorphic fluids along the Islay Anticline corroborates the previous finding by Skelton et al. (1995) of axial channelling of metamorphic fluids along the nearby Ar-drishaig Anticline.

� ‘Removal’ of the effects of metamorphic fluid flow reveals that a negative δ13C excursion (the ‘Islay anomaly’) recorded in carbonate rocks from the Lossit Limestone Formation can at least in part be attributed to metamorphic fluid flow. On the other hand, we cannot explain negative δ13C values in the Bonahaven Dolomite Formation by metamorphic flu-id flow.

29

� Mixing lines on δ13C – δ18O plots can be used to extrapolate from δ13C values that have been modified by metamorphic fluid flow to protolith δ13C values that can be more relia-bly used in paleoclimatic studies.

30

5. ACKNOWLEDGEMENTS I thank my supervisor, Alasdair Skelton, for feedback and comments during the time of my PhD studies. Co-supervisors Charlotte Möller and Patrick Crill are also equally thanked for their feedback. Josefin Linde and Jonas Nilsson are acknowledged for their work on the stau-rolite zone in Glen Esk, a project they undertook as their respective MSc theses. Dan Zetter-berg is greatly thanked for assistance in field, as well as with sample preparations and other practical issues. I thank Krister Junghahn for assistance with various practical issues at the department.

Runa Jacobsson and Cora Wohlgemuth-Überwasser are recognised for whole rock composition analyses. John Valley and Mike Spicuzza are grateful thanked for making the oxygen isotope analyses possible, and for a warm welcoming in Madison, Wisconsin. For their assistance with the electron microprobe in Uppsala, Jaroslaw Majka and Abigail Barker are also thanked. Curt Broman is thanked for fluid inclusion and Raman on graphite analyses.

I thank Graham and Isabelle Alison for their hospitality and the numerous enjoyable bar meals and lock-ins at the Port Charlotte Hotel on Islay. Davy, Donny and Ciara are warm-ly thanked for brightening up dark and cold March evenings with live music in the Port Char-lotte Hotel bar. These evenings have left a permanent imprint on my memory. Likewise, Da-vid Blacklaws is thanked for the hospitality given at the Panmure Arms in Edzell. Last but not least, I also thank all other friends and colleges I have forgotten to mention, but nevertheless directly or indirectly contributed to my progress.

31

6. REFERENCES Ague, J.J. (1994a). Mass transfer during Barrovian metamorphism of pelites, south-central

Connecticut. I: evidence for changes in composition and volume. American Journal of Science 294, 989-1057.

Ague, J.J. (1994b). Mass transfer during Barrovian metamorphism of pelites, south-central Connecticut: II, Channelized fluid flow and the growth of staurolite and kyanite. Ameri-can Journal of Science 294, 1061-1134.

Ague, J.J. (1997). Crustal mass transfer and index mineral growth in Barrow’s garnet zone, northeast Scotland. Geology 25, 73-76.

Ague, J.J. (2003). Fluid infiltration and transport of major, minor, and trace elements during regional metamorphism of carbonate rocks, Wepawaug schist, Connecticut, USA. American Journal of Science 303, 753-816.

Ague, J.J. (2007). Models of permeability contrasts in subduction zone mélange: Implications for gradients in fluid fluxes, Syros and Tinos Islands, Greece. Chemical Geology 239, 217–227.

Ague, J.J. (2014). 4.6 – Fluid flow in the deep crust. In: Holland, H.D. & Turekian, K.K. (eds.). Treatise on Geochemistry (2 ed.). Elsevier, 203-247.

Ague, J.J. (2011). Extreme channelization of fluid and the problem of element mobility dur-ing Barrovian metamorphism. American Mineralogist 96, 333-352.

Albee, A.L. (1965). A petrogenetic grid for the Fe-Mg silicates of pelitic schists. American Journal of Science 263, 512-536.

Allison, A. (1933). The Dalradian Succession in Islay and Jura. Quarterly Journal of the Geo-logical Society of London 89, 125-144.

Anderton, R. (1985). Sedimentation and tectonics in the Scottish Dalradian. Scottish Journal of Geology 21, 407–436.

Ashworth, J.R. & Evirgen, M.M. (1984). Mineral chemistry of regional chloritoid assemblag-es in the Chlorite Zone, Lycian Nappes, South-west Turkey. Mineralogical Magazine 48, 159-165.

Austrheim, H. (1987). Eclogitization of lower crustal granulites by fluid migration through shear zones. Earth and Planetary Science Letters 81, 221-232.

Bailey, E B. (1917). The Islay Anticline (Inner Hebrides). Quarterly Journal of the Geologi-cal Society of London 72, 132–164.

Baker, A.J. (1985). Pressures and temperatures of metamorphism in the eastern Dalradian. Journal of the Geological Society, London 142, 137–148.

Barrow, G. (1893). On an intrusion of muscovite-biotite gneiss in the south-eastern Highlands of Scotland, and its accompanying metamorphism. Quarterly Journal of the Geological Society of London 49, 330-358.

Barrow, G. (1912a). On the geology of Lower Dee-side and the southern Highland border. Proceedings of the Geologists’ Association 23, 274-290.

Barrow, G. (1912b). Lower Dee-side and the Highland border. Proceedings of the Geologists’ Association 23, 268-273.

Baxter, E.F., Ague, J.J. & Depaolo, D.J. (2002). Prograde temperature-time evolution in the Barrovian type-locality constrained by Sm/Nd garnet ages from Glen Clova, Scotland. Journal of the Geological Society, London 159, 71-82.

32

Berman, R.G. & Brown, T.H. (1985). The heat capacity of minerals in the system K2O-Na2O-CaO-MgO-FeO-Fe2O3-Al2O3-SiO2-TiO2-H2O-CO2: representation, estimation, and high temperature extrapolation. Contributions to Mineralogy and Petrology 89, 168-183.

Bickle, M.J. & Baker, J. (1990). Advective-diffusive transport of isotopic fronts: an example from Naxos, Greece. Earth and Planetary Science Letters 97, 78-93.

Bons, P.D., Elburg, M.A. & Gomez-Rivas, E. (2012). A review of the formation of tectonic veins and their microstructures. Journal of Structural Geology 43, 33-62.

Breeding, C.M. & Ague, J.J. (2002). Slab-derived fluids and quartz-vein formation in an ac-cretionary prism, Otago Schist, New Zealand. Geology 30, 499–502.

Breeding, C.M., Ague, J.J., Bröcker, M. & Bolton, E.W. (2003). Blueschist preservation in a retrograded, high-pressure, low-temperature metamorphic terrane, Tinos, Greece: impli-cations for fluid flow paths in subduction zones. Geochemistry, Geophysics, Geosystems 4.

British Geological Survey (1998). South Islay. Scotland sheet 19. Solid and drift geology. 1:50 000 provisional series.

Bucholz, C.E. & Ague, J.J. (2010). Fluid flow and Al transport during quartz-kyanite vein formation, Unst, Shetland Islands, Scotland. Journal of Metamorphic Geology 28, 19-39.

Bulau, J.R., Waff, H.S. & Tyburezy, J.A. (1979). Mechanical and thermodynamical con-straints on fluid distribution in partial melts. Journal of Geophysical Research 84, 6102-6108.

Carty, J.P., Connelly, J.N., Hudson, N.F.C. & Gale J.F.W. (2012). Constraints on the timing of deformation, magmatism and metamorphism in the Dalradian of NE Scotland. Scot-tish Journal of Geology 48, 103–117.

Carty, J.P., Connelly, J.N., Hudson, N.F.C. & Gale J.F.W. (2014). Erratum: Constraints on the timing of deformation, magmatism and metamorphism in the Dalradian of NE Scot-land. Scottish Journal of Geology 50, 95.

Chew, D.M. & Strachan, R.A (2014). The Laurentian Caledonides of Scotland and Ireland. In: Corfu, F., Gasser, D. & Chew, D.M. (eds.). New Perspectives on the Caledonides of Scandinavia and Related Areas. Geological Society of London Special Publications 390, 45-91.

Chinner, G.A. (1961). The origin of sillimanite in Glen Clova, Angus. Journal of Petrology 2, 312-323.

Chinner, G.A. (1966). The distribution of pressure and temperature during Dalradian meta-morphism. Quarterly Journal of the Geological Society 122, 159-181.

Connolly, J.A.D. (2005). Computation of phase equilibria by linear programming: a tool for geodynamic modelling and its application to subduction zone decarbonation. Earth and Planetary Science Letters 236, 524-541.

Connolly, J.A.D. (2009). The geodynamic equation of state: what and how. Geochemistry, Geophysics, Geosystems 10.

Cruickshank, D.R. & Ghent, E.D. (1978). Chloritoid-bearing pelitic rocks of the Horsethief Creek Group, southeastern British Columbia. Contibutions to Mineralogy and Petrology 65, 333-339.

33

Dempster, T.J. (1985). Garnet zoning and metamorphism of the Barrovian type. Contributions to Mineralogy and Petrology 89, 30–38.

Eskola, P. (1915). On the relation between the chemical and mineralogical composition in metamorphic rocks of the Orijärvi region. Bulletin of the Geological Society of Finland 44.

Eskola, P. (1920). The mineral facies of rocks. Norsk Geologisk Tidskrift 6, 143-194. Eskola, P. (1929). On the mineral facies. Proceedings of the Geological Society, Stockholm

51, 157-172. Fein, J.B., Graham, C.M., Holness, M.B., Fallick, A.E. & Skelton, A.D.L. (1994). Controls on

the mechanisms of fluid infiltration and front advection during regional metamorphism: A stable isotope and textural study of retrograde Dalradian rocks of the SW Scottish Highlands. Journal of Metamorphic Geology 12, 249–260.

Ferry, J.M. (1987). Metamorphic hydrology at 13-km depth and 400-550°C. American Min-eralogist 72, 39-58.

Ferry, J.M. (1994). A historical review of metamorphic fluid flow. Journal of Geophysical Research 99, 487-498.

Ferry, J. M. & Baumgartner, L. (1987). Thermodynamic models of molecular fluids at the elevated pressures and temperatures of crustal metamorphism. In: Carmichael, I. S. E. & Eugster, H. P. (eds) Thermodynamic Modeling of Geological Materials: Minerals, Flu-ids and Melts. Mineralogical Society of America, Reviews in Mineralogy 17, 323–365.

Geological Survey of Scotland (1898). Bowmore. Sheet 19. Solid geology. 1:63 360 series. Ghent, E.D., Stout, M.Z. & Filippo, F. (1989). Chloritoid-paragonite-pyrophyllite and

stilpnomelane-bearing rocks near Blackwater Mountain, Western Rocky Mountains, British Columbia. Canadian Mineralogist 27, 59-66.

Graham, C.M., Greig, K.M., Sheppard, S.M.F. & Turi, B. (1983). Genesis and mobility of the H2O-CO2 fluid phase during regional greenschist and epidote amphibolite facies met-amorphism: a petrological and chemical isotope study in the Scottish Dalradian. Journal of the Geological Society, London 140, 577-599.

Graham, C.M., Skelton, A.D.H., Bickle, M. & Cole, C. (1997). Lithological, structural, and deformation controls on fluid flow during regional metamorphism. In: Holness, M.B. (ed.). Deformation-enhanced Fluid Transport in the Earth’s Crust and Mantle. Chap-man & Hall, 196–226.

Greenwood, H.J. (1961). The system NaAlSi2O6-H2O-argon. Total pressure and water pres-sure in metamorphism. Journal of Geophysical Research 66, 3923-3946.

Greenwood, H.J. (1962). Metamorphic reactions involving two volatile components. Year Book Carnegie Institution of Washington 61, 82-85.

Halliday, A.N., Graham, C.M., Aftalion, M. & Dymoke, P. (1989). The depositional age of the Dalradian Supergroup: U-Pb and Sm-Nd isotopic studies of the Tayvallich Volcan-ics, Scotland. Journal of the Geological Society, London 146, 3-6.

Harker, A. (1912). The North Esk. Proceedings of the Geologists’ Association 23, 273. Harris, A.L. & Pitcher, W.S. (1975). The Dalradian Supergroup. In: A Correlation of Precam-

brian Rocks in the British Isles (eds Harris AL, Shackleton RM, Watson J, Downie C, Harland WB, Moorbath S). Special Report of the Geological Society of London, 6, 52-75.

34

Harte, B. (1987). Glen Esk Dalradian, Barrovian metamorphic zones. In: N.H., Kneller, B.C., Gillen, C. (Eds.). Excursion Guide to the Geology of the Aberdeen Area. Scottish Aca-demic Press (for Geological Society of Aberdeen), Edinburgh, pp. 193–210.

Harte, B. (2016). Barrow’s zones in Glen Esk, Angus, Scotland. Aberdeen Geological Society. Harte, B., & Hudson, N.F.C. (1979). Pelite facies series and the temperatures and pressures of

Dalradian metamorphism in E Scotland. In Harris, A.L., Holland, C.H. & Leake, B.E. The Caledonides of the British Isles; reviewed. Geological Society of London – Special Publication 8, 323-337.

Helgeson, H.C., Delany, J.M., Nesbitt, H.W. & Bird, D.K. (1978). Summary and critique of the thermodynamic properties of rock-forming minerals. American Journal of Science 278, 1-299.

Hensen, B.J. (1971). Theoretical phase relations involving cordierite and garnet in system MgO-FeO-Al2O3-SiO2. Contributions to Mineralogy and Petrology 33, 191-214.

Hess, P.C. (1969). The metamorphic paragenesis of cordierite in pelitic rocks. Contributions to Mineralogy and Petrology 24, 191-207.

Holland, T.J.B. & Powell, R. (1985). An internally consistent thermodynamic dataset with uncertainties and correlations: 2. Data and results. Journal of Metamorphic Geology 3, 343–370.

Holland, T.J.B. & Powell, R. (2011). An improved and extended internally consistent ther-modynamic dataset for phases of petrological interest, involving a new equation of state for solids. Journal of Metamorphic Geology 29, 333-383.

Holness, M.B. (1993). Temperature and pressure dependence of quartz-aqueous fluid dihedral angles: the control of adsorbed H2O on the permeability of quartzites. Earth and Plane-tary Science Letters 117, 363-377.

Kleine, B.I., Pitcairn, I.P., Skelton, A.D.L. (2015). The mechanism of infiltration of meta-morphic fluids recorded by hydration and carbonation of epidote-amphibolite facies metabasaltic sills in the SW Scottish Highlands. American Mineralogist 100, 2702-2717.

Kleine, B. I., Skelton, A. D. L., Huet, B., & Pitcairn, I. K. (2014). Preservation of blueschist-facies minerals along a shear zone by coupled metasomatism and fast-flowing CO2-bearing fluids. Journal of Petrology 55, 1905–1939.

Kleine, B.I., Zhao, Z. & Skelton, A.D.L. (2016b). Rapid fluid flow along fractures at greenschist facies conditions on Syros, Greece. American Journal of Science 316, 169-201.

Koons, P.O. & Thompson, A.B. (1985). Non-mafic rocks in the greenschist, blueschist and eclogite facies. Chemical Geology 50, 3-30.

Korzhinskii, D. S. (1957). Physicochemical Basis of the Analysis of the Paragenesis of Min-erals. Izvestiya Akademii Nauk SSSR, Moscow (transl. Consultants Bureau, New York, 1959).

Lewerentz, A. (2015). Vein controlled crystal size distributions of Barrovian index minerals. Licentiate Thesis in Geology, Department of Geological Sciences, Stockholm Universi-ty, 64 p.

Manby, G.M. (1983). A reappraisal of chloritoid-bearing phyllites in the Forland Complex rocks of Prins Karls Forland, Spitsbergen. Mineralogical Magazine 47, 311-318.

35

Masters, R.L., Ague, J.J. & Rye, D.M. (2000). An oxygen and carbon isotopic study of multi-ple episodes of fluid flow in the Dalradian and Highland Border Complex, Stonehaven, Scotland. Journal of the Geological Society, London 157, 367–379.

Masters, R.L. & Ague, J.J. (2005). Regional-scale fluid flow and element mobility in Bar-row’s metamorphic zones, Stonehaven, Scotland. Contributions to Mineralogy and Pe-trology 150, 1-18.

Oliver, G.J.H., Chen, F., Buchwaldt, R. & Hegner, E. (2000). Fast tectonometamorphism and exhumation in the type area of the Barrovian and Buchan zones. Geology 28, 459–462.

Oliver, G.J.H., Wilde, S.A. & Wan, Y. (2008). Geochronology and geodynamics of Scottish granitoids from the late Neoproterozoic break-up of Rodinia to Palaeozoic collision. Journal of the Geological Society, London 165, 661–674.

Oliver, N.H.S. (1996). Review and classification of structural controls on fluid flow during regional metamorphism. Journal of Metamorphic Geology, 14, 477–492.

Paradis, S., Velde, B. & Nicot, E. (1983). Chloritoid-pyrophyllite-rectorite facies rocks from Brittany, France. Contibutions to Mineralogy and Petrology 83, 342-347.

Penniston-Dorland, S.C. & Ferry, J.M. (2008). Element mobility and scale of mass transport in the formation of quartz veins during regional metamorphism of the waits river for-mation, east-central Vermont. American Mineralogist 93, 7–21.

Philpotts, A.R. & Ague, J.J. (2009). Principles of igneous and metamorphic petrology, 2nd

edition. Cambridge University Press, Cambridge, 667 p. Pitcairn, I.K., Skelton, A.D.L., Broman, C., Arghe, F. & Boyce, A. (2010). Structurally fo-

cused fluid flow during orogenesis: the Islay Anticline, SW Highlands, Scotland. Jour-nal of the Geological Society, London 167, 659-674.

Pogge von Strandmann, P.A.E., Dohmen, R., Horst R. Marschall, H.R., Schumacher, J.C. & Elliott, T. (2015). Extreme Magnesium Isotope Fractionation at Outcrop Scale Records the Mechanism and Rate at which Reaction Fronts Advance. Journal of Petrology 56, 33–58.

Powell, R. & Holland, T.J.B. (1985). An internally consistent thermodynamic dataset with uncertainties and correlation: 1. Methods and a worked example. Journal of Metamor-phic Geology 3, 372-342.

Powell, R., Holland, T.J.B. & Worley, B. (1998). Calculating phase diagrams involving solid solutions via non-linear equations, with examples using THERMOCALC. Journal of Metamorphic Geology 16, 577-588.

Putnis, A. & Putnis, C.V. (2007). The mechanism of reequilibration of solids in the presence of a fluid phase. Journal of Solid State Chemistry 180, 1783–1786.

Putnis, C.V. & Ruiz-Agudo, E. (2013). The mineral–water interface: where minerals react with the environment. Elements 9, 177-182.

Ridley, J. & Thompson, A.B. (1986). The role of mineral kinetics in the development of met-amorphic microtextures. In: Walther, J.V. & Wood, B.J. (Eds.). Fluid-rock interactions during metamorphism. Advances in Physical Geochemistry 5, 195-212. New York, Springer-Verlag, 218 p.

Roberts, J.L. & Treagus, J.E. (1979). Stratigraphic and structural correlation between the Dalradian rocks of the SW and Central Highlands of Scotland. In: Harris, A.L., Holland,

36

C.H. & Leake, B.E. (eds.). The Caledonides of the British Isles – Reviewed. Special Publication of the Geological Society of London, 8, 199-204.

Rubenach, M (2013). Structural controls of metasomatism on a regional scale. In: Harlov, D.E. & Austrheim, H. (eds.). Metasomatism and the Chemical Transformation of Rock: the role of fluids in terrestrial and extraterrestrial processes. Lecture Notes in Earth System Sciences. Springer, 93-140.

Scheele, M. & Hoefs, J. (1992). Carbon isotope fractionation between calcite, graphite and CO2: an experimental study. Contributions to Mineralogy and Petrology 112, 34-45.

Skelton, A.D.L. (1996). The timing and direction of metamorphic fluid flow in Vermont. Contributions to Mineralogy and Petrology 125, 75–84.

Skelton, A.D.L. (1997). The effect of metamorphic fluid flow on the nucleation and growth of garnets from Troms, North Norway. Journal of Metamorphic Geology 15, 85-92.

Skelton, A.D.L., Graham, C.M. & Bickle, M.J. (1995). Lithological and Structural Controls on Regional 3-D Fluid Flow Patterns during Greenschist Facies Metamorphism of the Dalradian of the SW Scottish Highlands. Journal of Petrology 36, 563-586.

Skelton, A.D.L., Lewerentz, A., Kleine, B., Webster, D. & Pitcairn, I. (2015). Structural Channelling of Metamorphic Fluids on Islay, Scotland: Implications for Paleoclimatic Reconstruction. Journal of Petrology 56, 2145-2172.

Skelton, A.D.L., Valley, J.W., Graham, C.M., Bickle, M.J. & Fallick, A.E. (2000). The corre-lation of reaction and isotope fronts and the mechanism of metamorphic fluid flow. Contributions to Mineralogy and Petrology 138, 364-375.

Smith, C.S. (1948). Grains, phases, and interfaces: an interpretation of microstructure. Trans-actions of the Metallurgical Society of AIME 175, 15-51.

Spear, F.S. & Cheney, J.T. (1989). A petrogenetic grid for pelitic schists in the system SiO2-Al2O3-FeO-MgO-K2O-H2O. Contributions to Mineralogy and Petrology 101, 149-164.

Spencer, A.M. (1971). Late Pre-Cambrian glaciation in Scotland. Geological Society of Lon-don Memoir 6, 99 pp.

Stephenson, D., Mendum, J.R., Fettes, D.J. & Leslie, A.G. (2013). The Dalradian rocks of Scotland: an introduction. Proceedings of the Geologists’ Association 124, 3–82.

Tanner, P.W.G. & Pringle, M. (1999). Testing for a terrane boundary within Neoproterozoic (Dalradian) to Cambrian siliceous turbidites at Callander, Perthshire, Scotland. Journal of the Geological Society, London 156, 1205–1216.

Tanner, P.W.G. & Sutherland, S. (2007). The Highland Border Complex, Scotland: a paradox resolved. Journal of the Geological Society, London 164, 111–116.

Tanner, P.W.G., Thomas, C.W., Harris, A.L., Gould, D., Harte, B., Treagus, J.E. & Stephen-son, D. (2013). The Dalradian rocks of the Highland Border region of Scotland. Pro-ceedings of the Geologists’ Association 124, 215-262.

Taylor, H.P. (1977). Water/rock interactions and the origin of H2O in granitic batholiths: Thirtieth William Smith lecture. Journal of the Geological Society 133, 509-558.

Thompson, J.B. (1957). The graphical analysis of mineral assemblages in pelitic schists. American Mineralogist 42, 842-858.

Thompson, J.B. (1959). Local equilibrium in metasomatic processes. In: Abelson, P.H. (Ed.). Researches in Geochemistry, Vol. 1, John Wiley (1959), 427–457.

37

Thompson, A.B. (1976). Mineral reactions in pelitic rocks: II. Calculation of some P-T-X(Fe-Mg) phase relations. American Journal of Science 276, 425-454.

Thomson, J. (1871). On the stratified rocks of Islay. In: Report of the 41st Meeting of the Brit-ish Association for the Advancement of Science, Edinburgh. John Murray, pp. 110–111.

van Haren J. L. M., Ague J. J. & Rye D. M. (1996). Oxygen isotope record of fluid infiltration and mass transfer during regional metamorphism of pelitic schist, Connecticut, USA. Geochimica Cosmochimica Acta 60, 3487–3504.

Viete, D.R., Oliver, G.J.H., Fraser, G.L., Forster, M.A. & Lister, G.S. (2013). Timing and heat sources for the Barrovian metamorphism, Scotland. Lithos 177, 148-163.

von Bargen, N. & Waff, H.S. (1986). Permeabilities, interfacial areas and curvatures of par-tially molten systems: Results of numerical computations of equilibrium microstruc-tures. Journal of Geophysical Research 91, 9261-9276.

Vorhies, S.H. & Ague, J.J. (2011). Pressure-temperature evaluation and thermal regimes in the Barrovian zones, Scotland. Journal of the Geological Society, London 168, 1147–1166.

Vorhies, S.H., Ague, J.J. & Schmitt, A.K. (2013). Zircon growth and recrystallization during progressive metamorphism, Barrovian zones, Scotland. American Mineralogist 98, 219-230.

Waff, H.S. & Faul, U.H. (1992). Effects of crystalline anisotropy on fluid distribution in ul-tramafic partial melts. Journal of Geophysical Research 97, 9003-9014.

White. R.W., Powell, R. & Johnson, T.E. (2014a). New mineral activity-composition rela-tions for thermodynamic calculations in metapelitic systems. Journal of Metamorphic Geology 32, 261-286.

White. R.W., Powell, R. & Johnson, T.E. (2014b). The effect of Mn on mineral stability in metapelites revisited: new a–x relations for manganese-bearing minerals. Journal of Metamorphic Geology 32, 809-828.

Whitney, D.L., Mechum, T.A., Kuehner, S.M. & Dilek, Y.R. (1996). Progressive metamor-phism of pelitic rocks from protolith to granulite facies, Dutchess County, New York, USA: constraints on the timing of fluid infiltration during regional metamorphism. Journal of Metamorphic Geology 14, 163-181.

Widmer, T. & Thompson, A.B. (2001). Local origin of high pressure vein material in eclogite facies rocks of the Zermatt-Saas zone, Switzerland. American Journal of Science 301, 627–656.

Wilkinson, S. B. (1907). The Geology of Islay. Memoir Geological Survey of Scotland. Winter, J.D. (2010). An introduction to igneous and metamorphic petrology, 2nd edition. Pren-

tice Hall, Upper Saddle River, 702 p. Wood, B. J. & Walther, J. V. (1986). Fluid flow during metamorphism and its implications

for fluid–rock ratios. In: Walther, J. V. & Wood, B. J. (eds). Fluid–rock interactions during metamorphism. Springer, pp. 89–108.

Wright, A.E., 1988. 15. The Appin Group. In: Winchester, J.A. (Ed.). Later Proterozoic Stra-tigraphy of the Northern Atlantic Regions. Blackie, 177–199.

Yardley, B.W.D. & Bottrell, S.H. (1992). Silica mobility and fluid movement during meta-morphism of the Connemara schists, Ireland. Journal of Metamorphic Geology 10, 453–464.

38

Young, E.D. & Rumble, D. (1993). The origin of correlated variations in in-situ 18O/16O and elemental concentrations in metamorphic garnet from southeastern Vermont, USA. Ge-ochimica et Cosmochimica Acta 57, 2585-2597.

Zack, T. & John, T. (2007). An evaluation of reactive fluid flow and trace element mobility in subducting slabs. Chemical Geology 239, 199-216.