Partial melting and melt segregation in footwall units within the contact aureole of the Sudbury Igneous Complex (North and East Ranges, Sudbury structure), with implications for their

PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [Péntek, Attila]On: 10 July 2009Access details: Access Details: [subscription number 913079218]Publisher Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

International Geology ReviewPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t902953900

Partial melting and melt segregation in footwall units within the contact aureoleof the Sudbury Igneous Complex (North and East Ranges, Sudbury structure),with implications for their relationship to footwall Cu-Ni-PGE mineralizationAttila Péntek a; Ferenc Molnár a; David H. Watkinson b; Peter C. Jones b; Aberra Mogessie c

a Department of Mineralogy, Eötvös Loránd University, Budapest, Hungary b Department of Earth Sciences,Carleton University, Ottawa, Canada c Institute for Earth Sciences, Karl-Franzens University, Graz, Austria

First Published on: 10 July 2009

To cite this Article Péntek, Attila, Molnár, Ferenc, Watkinson, David H., Jones, Peter C. and Mogessie, Aberra(2009)'Partial meltingand melt segregation in footwall units within the contact aureole of the Sudbury Igneous Complex (North and East Ranges, Sudburystructure), with implications for their relationship to footwall Cu-Ni-PGE mineralization',International Geology Review,99999:1,

To link to this Article: DOI: 10.1080/00206810903101313

URL: http://dx.doi.org/10.1080/00206810903101313

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

http://www.informaworld.com/smpp/title~content=t902953900

http://dx.doi.org/10.1080/00206810903101313

http://www.informaworld.com/terms-and-conditions-of-access.pdf

Partial melting and melt segregation in footwall units within thecontact aureole of the Sudbury Igneous Complex (North and East

Ranges, Sudbury structure), with implications for their relationshipto footwall Cu–Ni–PGE mineralization

Attila Penteka*, Ferenc Molnara, David H. Watkinsonb, Peter C. Jonesb and

Aberra Mogessiec

aDepartment of Mineralogy, Eotvos Lorand University, Pazmany Peter setany, 1/C, Budapest 1117,Hungary; bDepartment of Earth Sciences, Carleton University, 1125 Colonel By Drive, Ottawa,Canada, K1S 5B6; cInstitute for Earth Sciences, Karl-Franzens University, 2 Universitatsplatz, Graz

8010, Austria

(Accepted 8 June 2009)

We performed detailed field and drill core mapping of partial melting features and felsicrocks (footwall granophyres, FWGRs) representing segregated and crystallized partialmelts within the contact aureole of the Sudbury Igneous Complex (SIC) in the 1.85 GaSudbury impact structure. Our results, derived from mapping within the North (WindyLake, Foy, Wisner areas) and East Ranges (Skynner, Frost areas) of the structure, revealthat partial melting was widespread in both felsic and mafic footwall units up to distancesof 500 m from the basal contact of the SIC. Texturally and mineralogically, significantdifferences exist between rocks formed by partial melting within and between localities.In general, however, melt bodies are dominated by different quartz-feldspar intergrowths(e.g. granophyric, graphic) and miarolitic cavities up to 5 cm in diameter. Major and traceelement compositions of Wisner and Frost FWGRs imply that they crystallized from meltsdominantly derived from partial melting of felsic Levack Gneiss and Cartier granitoidrocks, as well as from gabbroic rocks only at Frost. These results accord with ourobservations on in situ partial melting features and crystallized melt of microscopic scalein both felsic and mafic rocks. We conclude that partial melting occurred at a pressure of1.5 ^ 0.5 kbar and at temperatures up to 7508C in the Wisner area and up to 9008C in theFrost and Windy Lake areas. Segregations of partial melt into veins and dikes are presentin all localities, and were promoted by deformation of the Sudbury structure in thePenokean orogeny as indicated by dominant strike directions. Whereas veins and dikesreflect brittle conditions during melt migration, sheared melt pods in the Sudbury brecciamatrix indicate ductile conditions during their crystallization. Our results suggest a closegenetic association of partial melting, melt segregation, and hydrothermal processesresponsible for remobilization of Cu–Ni–PGE sulphides into and within the SIC footwall.

Keywords: partial melting; contact aureole; footwall granophyre; miarolite;geochemistry; Sudbury

Introduction

The Sudbury structure is one of the largest known impact structures on Earth, with an

estimated original diameter of 200–250 km (Dietz 1964; Grieve et al. 1991; Deutsch et al.

1995; Pope et al. 2004). The 1.85 Ga structure (Krogh et al. 1984) is host to the Sudbury

ISSN 0020-6814 print/ISSN 1938-2839 online

q 2009 Taylor & Francis

DOI: 10.1080/00206810903101313

http://www.informaworld.com

*Corresponding author. Email: [email protected]

International Geology Review

iFirst article, 2009, 1–35

Downloaded By: [Péntek, Attila] At: 22:43 10 July 2009

Igneous Complex (SIC), which crystallized from the impact melt sheet and is one of the

world’s most productive Ni–Cu–platinum group element (PGE) mining camps.

Crystallization of the superheated impact melt sheet resulted in thermal erosion and

assimilation of up to 800 m of footwall rocks (Prevec and Cawthorn 2002) and

development of a 1.2–2 -km-wide contact aureole in which partial melting also occurred

(e.g. Dressler 1984a, 1984b; Rosenberg and Riller 2000; Boast and Spray 2006). Molnar

et al. (2001) documented the mineralogy and fluid inclusion characteristics of granitic

veins, which they interpreted to be results of SIC-related partial melting, and introduced

the term ‘footwall granophyre’ (FWGR) for them. These authors discovered primary fluid

inclusions in quartz of miarolitic cavities of such veins which are very similar to fluid

inclusions from several Cu–Ni–PGE deposits in the footwall of the SIC; therefore they

suggested that fluids exsolving from crystallizing FWGRs may have played a role in the

formation of ore deposits. Despite possible economic significance, no detailed study of

FWGRs and the partial melting process itself has been performed up to now.

The aim of this paper is to document SIC-related partial melting of footwall units in the

North and East Ranges of the Sudbury structure. Detailed mapping and sampling were

performed in three areas (Frost, Wisner, and Windy Lake) in the vicinity of the SIC/footwall

contact, supplemented by large-scale mapping in two other areas (Foy and Skynner; Figure 1).

We describe petrography, mineral chemistry, and bulk-rock geochemistry of partial melt

bodies, and use these data to evaluate how and where SIC-related partial melting occurred

and how it may have been related to magmatic-hydrothermal processes and sulphide

mineralization.

General geology, the footwall lithologies, and contact aureole of the SIC

The Sudbury impact structure can be divided into three main units: impact-brecciated and

shock-metamorphosed footwall units, the SIC, and the overlying Whitewater Group units

(fallback breccia and sedimentary rocks) within the Sudbury Basin (Figure 1). The 30 by

60 km-large, ,2.5 km-thick SIC now forms an elliptical body crystallized from the melt

sheet, derived from the fusion of impacted target rocks. It is believed that the initial

superheated melt sheet with a temperature of at least 20008C has undergone magmatic

differentiation to form the layered sequence of the SIC main mass consisting of norite,

quartz gabbro, and granophyre (Naldrett and Hewins 1984; Ivanov and Deutsch 1999;

Prevec and Cawthorn 2002; Therriault et al. 2002; Zieg and Marsh 2005). The contact

sublayer at the base of the SIC is a noritic unit containing centimetre- to metre-scale

inclusions of adjacent footwall rocks and exotic mafic to ultramafic rocks (Naldrett et al.

1984; Lightfoot et al. 1997). It has major importance in hosting Ni–Cu–PGE orebodies,

especially in depressions of the footwall/SIC contact referred to as embayments, where

this unit may reach a thickness of up to 150 m. Another host of contact sulphides in

embayments is the footwall breccia which contains mostly locally derived clasts of

footwall rocks in a matrix of tonalitic to granitic composition (Dressler 1984b; Lakomy

1990; McCormick et al. 2002a). In the past 15 years, the focus of mineral exploration was

on the ‘footwall’-type deposits, which are situated up to 1 km distance below the contact

and are strongly enriched in Cu, PGE, and Au. These deposits appear to have formed

through a combination of magmatic and hydrothermal processes by fractionation and

remobilization of contact sulphides (see Ames and Farrow 2007; Pentek et al. 2008 and

references therein).

The Sudbury structure has a polyphase tectonic history including syn- and post-impact

deformation. It is believed that gravitationally-driven crater modifications must have

A. Pentek et al.2


caused large-scale movement in the Sudbury structure, but up to now only a few studies

have been published regarding this topic (e.g. collapse along concentric superfaults as

proposed by Spray 1997). The impact coincided with the Penokean orogeny (1.89–1.83 Ga:

Sims et al. 1989), which was a NW–SE directed compressional event. This orogeny is

believed to be dominantly responsible for the deformation of the Sudbury structure to its

present ellipsoidal shape, as well as for the generation and reactivation of northwest-

trending strike-slip faults, which offset the SIC on the North Range (Rousell 1984; Cowan

and Schwerdtner 1994; Riller 2005). In the North and East Ranges, the SIC is dominantly

underlain by Levack Gneiss Complex (LGC) and Cartier Batholith (CB) of the Archaean

Superior Province, whereas the footwall in the South Range is composed of Huronian

metavolcanic and metasedimentary rocks, as well as Creighton and Murray granites of the

Figure 1. (a) Geological map of the Sudbury structure showing the location of the studiedlocalities. (b) and (c) Location of studied FWGR occurrences in the Frost (b) and Wisner (c) areas(modified after unpub. maps by Wallbridge Mining Company Ltd. 2006).

International Geology Review 3


Palaeoproterozoic Southern Province. The LGC (2.71 Ga protolith ages: Krogh et al. 1984)

consists mainly of migmatitic tonalite gneiss, with common compositional layering (Card

et al. 1984; Dressler 1984a; Card 1994). Granulite-facies metamorphism and accom-

panying anatexis occurred at around 2.65 Ga (Krogh et al. 1984; James et al. 1991). The CB

originated by partial melting of the LGC, which is consistent with the 2.64 Ga age of its

emplacement (Meldrum et al. 1997). The Cartier granitoids range from granodiorite to

monzogranite, the latter being the most common. The gneisses and granitoid units were

later intruded by mafic bodies of the East Bull Lake suite (2.49–2.47 Ga: Krogh et al. 1984),

the Matachewan-type dike swarm (2.47–2.45 Ga: Heaman 1997), and the 2.2 Ga Nipissing

suite (Corfu and Andrews 1986).

These footwall rocks underwent impact brecciation at the 1.85 Ga Sudbury event. The

resulting pseudotachylite forms millimetre-wide veins to hundreds of metre-wide zones up

to a distance of at least 80 km from the SIC/footwall contact (Dressler 1984b; Rousell et al.

2003). These impactites, known as Sudbury breccia (SDBX), contain mostly clasts of

adjacent units in a fine-grained, recrystallized matrix. Wide zones of SDBX were very

important in the later history of the Sudbury structure, including the formation of ore

deposits in the footwall (Rousell et al. 2003), as they were conduits for SIC-related melts

and fluids.

Cooling of the superheated impact melt resulted in the development of an

approximately 1–2 km-wide contact aureole below the SIC (Coats and Snajdr 1984;

Dressler 1984b; Hanley and Mungall 2003). A recent study by Boast and Spray (2006)

emphasized the importance of later tectonic processes, which in some areas displaced and

obscured the contact aureole. According to these studies, the complete metamorphic

aureole comprises (with distances from the SIC) a zone of anatexis and assimilation

(¼ footwall breccia; up to ,25 m), pyroxene hornfels (up to 200–350 m), hornblende

hornfels (up to 600–1100 m), and albite-epidote hornfels (up to 1200–2000 m).

Several authors described observations, which suggest that partial melting of footwall

rocks also occurred during cooling of the SIC. Dressler (1984b) pointed out that

granophyric quartz-feldspar intergrowth in the matrix of the footwall breccia may be an

indicator of partial melting. Granitic dikes, intruding the main mass norite at the contact

with the Murray and Creighton plutons in the South Range, have also been suggested to be

the result of partial melting in the footwall granite and back-injection of these melts

into the SIC (Collins 1936; Davis 1984; Dressler 1984a). One of these granitic dikes gave a

U–Pb age of 1.855 Ga, which is consistent with its formation due to SIC-related partial

melting (Krogh et al. 1984). Rosenberg and Riller (2000) published microtextural

observations of in situ crystallized partial melts in Murray granite from the contact aureole

of the SIC. Granophyre-rich footwall breccia veins and dikes intruding underlying rocks

were observed by Dressler (1984b), who also pointed out the existence of medium-grained

to pegmatitic granophyre dikes cutting footwall breccia. Such felsic veins cutting Cu–Ni–

PGE mineralized footwall breccia, SIC norite, and SDBX were first studied by Marshall

et al. (1999) and Molnar et al. (2001) who referred to them as ‘FWGRs’ and described an

assemblage dominated by graphic-granophyric quartz-feldspar intergrowth and mega-

crysts of hornblende, clinopyroxene, and titanite.

Models for the cooling history of the SIC also support the idea of partial melting in

footwall rocks. According to Prevec and Cawthorn (2002), the superheated melt sheet must

have consumed several hundred metres of its footwall, the assimilation process of which

was aided by the brecciated nature of the basement (Boast and Spray 2006). Recently, the

contact sublayer and the footwall breccia are believed to have formed as a result of this

process (Prevec et al. 2000; Prevec and Cawthorn 2002; Boast and Spray 2006).

A. Pentek et al.4


Methods of study

Mapping in the scale of 1:75 was carried out on trenched and high-pressure water-washed

outcrops of the Wallbridge Mining Company’s Wisner and Frost Lake properties

(Figure 1(a)). We also investigated selected diamond drill cores from Wallbridge’s Frost

Lake, Wisner, as well as Windy Lake properties, while large-scale (1:2000) mapping of

FWGRs was carried out in the Frost Lake, Skynner and Foy properties (Figure 1(a)).

Bulk rock geochemical analyses of grab samples and half-core samples were carried

out by ALS Chemex Ltd. The samples were analysed for major elements, as well as 50

base metals and trace elements (including Pt, Pd, Au) using inductively coupled plasma

mass spectrometry and inductively coupled plasma atomic emission spectrometry.

Mineral chemical data were acquired using a Camebax MBX electron microprobe by

wavelength dispersive analysis at Carleton University, Ottawa (Canada) as well as a Jeol

SEM 6310 scanning electron microscope with a wavelength and energy dispersive system

at the Karl-Franzens University, Graz (Austria). Analyses were carried out at 15 kV and

15 nA; counting times of 15–20 s or 40,000 counts were employed for each element

except for F (40 s) and Ni (40 s).

Occurrence and appearance of FWGRs

South and Southwest zones in the Wisner area (North Range)

The South Zone is situated in a SE–NW trending (,120–300 m wide), intensively

Sudbury brecciated zone which is located roughly 500 m NW from the Rapid River

contact Ni–Cu–PGE occurrence, the nearest known contact environment in the area

(Figure 1(c)). The Southwest Zone is also located in a brecciated zone of the footwall,

about 200 m North of the nearest exposure of contact sublayer. This sublayer forms the

base of an SIC unit, which has been offset about 400 m outward into the footwall along a

SE–NW striking fault. After the discovery of footwall Cu–Ni–PGE mineralization within

both areas in 2003–2004, bedrock was cleaned and washed in eight trenches providing

approximately 4400 m2 exposure.

At South Zone, trenches expose dominantly pink coloured medium- to coarse-grained

Cartier granitoid (quartz monzonite (QM) to granodiorite) rich in gneiss inclusions. The

Southwest Zone trenches are mainly composed of the same granitoid and medium-grained

felsic to mafic gneisses of the LGC. These rock units are intruded in both areas by late

Cartier granitic pegmatite dikes, mafic dikes of the Matachewan-type suites, and at South

Zone also by minor gabbro of the East Bull Lake suite (Wisner Gabbro). At South Zone,

these footwall rocks are strongly brecciated along a SE–NW striking zone, while thin (up

to several centimetres) SDBX veins also penetrate into the massive footwall units outside

this zone. These veinlets are either parallel or roughly perpendicular to the main SDBX

zone. At the Southwest Zone, breccia is very common at all trenches as massive zones, or

as dikes and veins having a preferred SE–NW strike. The typical ‘low sulphide’, PGE-rich

mineralization of the South Zone consisting of sulphide veinlets and disseminations is

interpreted to be hydrothermal in origin (Pentek et al. 2008); the very similar Cu–Ni–

PGE mineralization of the Southwest Zone was described briefly by Pentek (2006).

FWGRs cut all footwall rocks including SDBX (Figure 2) in both localities,

evidencing their post-impact origin. They may either be millimetre- to 40 centimetre-wide

and several metres-long veins and dikes (Figures 3(a) and 4(e)), or pods and irregular

bodies. Small (10–20 cm2) irregular intrusions of FWGR in SDBX show clear evidence of

syngenetic ductile deformation (Figure 3(c) and (d)). They appear to have penetrated

through the matrix of the breccia or smeared along the clasts. At some places, the matrix of



the breccia within a zone of several decimetres in diameter is entirely FWGR containing

clasts of the original matrix.

In all cases, FWGRs are characterized by a fine-grained aplitic texture, pink or grey

colour, and abundant miarolitic cavities up to 1 cm in diameter. The veins and dikes

intruded in a manner which reflects a transition between brittle and ductile conditions of

Figure 2. (a) Simplified geologic map of the Wisner Southwest area with locations of studiedtrenches and drill holes. Rose diagrams illustrate strike directions of FWGR veins and dikes.Geologic maps of representative parts of (b) the Blast trench (Wisner Southwest) and the (c) and (d)East trench (Wisner South; modified after Peterson and Fell, unpubl. map, Wallbridge MiningCompany Ltd. 2004) highlighting the occurrence of FWGRs, Cu–Ni–PGE sulphides, and samplelocations. Rose diagram in (c) illustrates strike directions of melt bodies at Wisner South.

A. Pentek et al.6


the host rock during emplacement: in gneisses and granitoids they tend to be relatively

straight, while they are usually more curvilinear in SDBX matrix. The contacts of the veins

and dikes are sharp-walled towards the granitoids and gneisses and more diffuse towards

the breccia matrix, whereas the irregular intrusions in SDBX always exhibit very diffuse,

gradational contacts. Generally, the veins and dikelets show a very dominant strike of

NW–SE (,140–3208) with the exception of one trench in the Southwest zone where their

strike is ENE–WSW (,80–2608; Figure 2(a)). In less intensely brecciated zones, the

FWGRs either bend into and intrude along the breccia veinlets or cut them without

changing orientation.

A close relationship of FWGR veins and Cu–Ni–PGE sulphide mineralization is

revealed by several observations, especially in the South Zone where terminations

of FWGRs are often gradually transformed into sulphide-rich hydrothermal veins

(Figure 2(d)). It is also common that patches of sulphides are enclosed by the granophyric

veins, occurring preferably within the miarolites. Sulphides are dominantly chalcopyrite

and millerite with a wide range of associated trace metal minerals, including platinum-

group minerals (Pentek et al. 2008).

Diamond drill cores from the Southwest Zone were also studied in order to get a better

understanding of the spatial distribution of FWGRs. These occur continuously up to a

vertical depth of 270–290 m in drill holes WIS-92 and WIS-93 and to a depth of 145 m

in WIS-94 (Figure 2(a)). However, the abundance of melt bodies does not decrease

Figure 3. Photographs of typical FWGRs from Frost ALZ and the Wisner occurrences. (a) Apliticdike from Wisner SW cutting QM. Note the abundant miarolitic cavities in the melt body, as well asthe partial melting of host rock next to the dike (indicated by arrows). (b) Amphibole-rich melt bodyin SDBX matrix next to gabbroic clast (Frost ALZ). (c) FWGRs in SDBX matrix at Wisner SW.Note the strong ductile deformation and the diffuse contacts of the melt bodies. (d) ‘Flame-like’veinlets of FWGR cutting through QM and Sudbury breccia (Wisner S). FWGR, footwallgranophyre; IGN, intermediate gneiss; QM, quartz monzonite; and SDBX, Sudbury breccia.



gradationally downward (away from the SIC) as may be expected in a contact aureole.

SDBX zones have experienced more intense partial melting and host melt bodies at depth

where they are absent from unbrecciated rocks. Other occurrences in the Wisner area

further constrain spatial distribution of FWGRs in relation to the SIC: in hole WIS-038,

Figure 4. Photographs of typical FWGRs from Frost ALZ, Windy Lake and Wisner S. (a) Cut slabof melt body seen in Figure 3(b) with miarolitic cavities filled by quartz and euhedral amphibole.Note the recrystallized zone of host gabbro at the contact to the melt body (indicated by arrow).(b) FWGR cutting mafic gneiss in Windy Lake drill core. Note the large miarolitic cavity in thecentral part of the vein, the coarse-grained granophyric-graphic texture, as well as long opaqueneedles, which are made up of rutile, ilmenite, and titanite. (c) Partially melted IGN in Windy Lakedrill core. Crystallized partial melt can easily be distinguished by its pink colour, the coarse-grainedgranophyric-graphic texture, miarolitic cavities (see arrow) and the black, opaque needles. (d) Meltbodies dominated by euhedral plagioclase laths in partially melted gabbroic rock (Windy Lake).Apart from the segregated pods, note also mm-sized felsic patches in the gabbro, which alsorepresent crystallized partial melt. (e) Aplitic FWGR veins cutting through SDBX matrix next to QM(Wisner S). (f) FWGR vein cutting partially melted gabbroic rock in Windy Lake drill core. The veinis dominantly made up of euhedral plagioclase laths; quartz occurs interstitially and in a miarolite(see arrow). FWGR, footwall granophyre; IGN, intermediate gneiss; QM, quartz monzonite andSDBX, Sudbury breccia.

A. Pentek et al.8


approximately 800 m from the contact, they occur only in the first 20 m; in WIS-40, WIS-

84, and the Broken Hammer Cu–Ni–PGE occurrence (Figure 1(c)), which are all located

1.2–1.3 km N of the contact, FWGRs do not occur. An important conclusion from these

more regional observations is that the texture and mineralogy of the melt bodies appears to

be independent of host rock and distance from the SIC.

Amy Lake Zone (ALZ) in the Frost Lake area (East Range)

The footwall Cu–Ni–PGE mineralization of the ALZ is located in the so-called Amy

Lake Breccia Belt, an up to 100 m wide SDBX body parallel to the basal contact of the SIC

(Figure 1(b)). This belt is also host to the Capre 3000 Cu–Ni–PGE deposit and connects

into the contact, where the Capre contact deposit is located. The ALZ is 300–600 m

horizontal distance north of the nearest exposed contact, which occurs at the base of an

SIC slab that was offset along the roughly N–S Amy Lake Fault (Figure 1(b)). Bedrock is

exposed on several trenches and consists of a wide variety of brecciated footwall rocks.

Migmatitic, mafic Levack gneiss and Cartier granite are dominant, but later mafic dikes

and bodies are also abundant; it is unclear to which intrusive suite they belong. SDBX

shows intense signs of ductile deformation: small clasts (up to several cm in diameter)

within the matrix are sheared and extremely elongated having a NNE–SSW orientation on

all trenches (Figure 5(b)); these features indicate ductile movement of the breccia.

FWGRs are abundant in all trenches within the breccia belt but occur only in

insignificant amounts in the East trench (Figure 5), which exposes similar footwall rocks,

however located outside of the brecciated zone. In the South trench, where fragmentation

was intense and the breccia is matrix-supported, irregular intrusions and pods of FWGR in

the matrix are most common (Figure 5). They have diffuse contacts with the matrix and

show signs of ductile deformation. As described also at Wisner, melt pods (up to 1 m in

diameter; Figure 3(b)) may replace the matrix of the breccia and contain strongly corroded

fragments of it. In the mafic clasts, the melt bodies occur as curvilinear ‘flame-like’

intrusions, which have sharp contacts and are aligned ENE–WSW. The North trench

exposes much less fragmented and thus clast-supported SDBX, with the matrix occurring

as dikes and veins. There, FWGRs occur mainly as veins cutting footwall rocks in a similar

manner to that described at Wisner. They have sharp contacts with the intruded rock and a

preferred orientation of NW–SE (120–3008 and 160–3408).

Melt bodies of the ALZ are texturally and mineralogically much more diverse, than

those observed at Wisner. Most commonly, they are pink, felsic rocks made up of

macroscopically visible granophyric to graphic quartz-feldspar intergrowth. Some of the

melt bodies also contain abundant euhedral amphibole up to 1.5 cm in length (Figure 4(a)).

Miarolitic cavities up to 5 cm in diameter are filled with quartz, epidote, and amphibole;

they locally contain also up to centimetre-sized patches of chalcopyrite. Amphibole-rich

melt bodies were only observed in the South trench, where SDBX contains abundant

gabbroic rocks. In the North trench, all FWGRs are felsic. It is important to emphasize that

miarolitic cavities tend to be much larger in irregular, ductile melt pods, than they are in

the brittle dikes and veins.

The PGE-rich sulphides occur in veinlets and disseminations accompanied by intense

hydrothermal alteration. A close genetic relationship between these mineralizing

hydrothermal processes and the FWGRs is revealed again by several observations including

the already mentioned sulphide-rich miarolitic cavities of some amphibole-rich melt bodies.

In some cases, a physical continuum between these melt bodies and hydrothermal veins can

also be observed, similar to that described from the Wisner South Zone.



As a comparison to the ALZ, we performed mapping and sampling of FWGRs in other

parts of the Amy Lake breccia belt (e.g. drill hole WC-19) and other areas within the Frost

property (Frost West; Figure 1(b)), which do not host known occurrences of Cu–Ni–PGE

sulphides. The abundance of FWGRs was always found to be highest in SDBX zones, with

only minor signs of partial melting in unbrecciated rocks. Independent of host rock and

Figure 5. (a) Simplified geological map of the Frost ALZ with locations of the studied trenches.See Figure 2(a) for legend. (b) and (c) Geologic maps of the South trench (Frost ALZ) showing theoccurrence of FWGRs, Cu–Ni–PGE sulphides, and sample locations. (a) Rose diagram shows thedirections of clast elongation due to plastic deformation; each measurement represents a group ofcentimetre-sized clasts within an approximately 0.25–1 m2 area. (b) Rose diagram shows strikedirections of FWGRs.

A. Pentek et al.10


distance from the SIC/footwall contact, the texture and mineralogy of the melt bodies was

similar to those observed at the ALZ.

Windy Lake, Skynner and Foy areas

In the Windy Lake area (Figure 1(a)), exploration drilling projects outlined an embayment

structure hosting sub-economic disseminated Ni–Cu–PGE mineralization in sublayer and

footwall breccia, which together have a thickness of 150–200 m. Although the exact shape

and dip of the SIC/footwall contact in this area is not known, it has been estimated to be

,308 to SE by Rousell (1984), and the drill holes were deepened approximately normal to

this inferred dip of the contact. For the purpose of the current study, those drill holes were

chosen which exposed several hundred metres of the footwall below the contact and which

did not intersect major structures that may have disturbed the aureole. According to our

recent knowledge, the intersected footwall does not host Cu–Ni–PGE mineralization and

is dominantly made up of variable Sudbury brecciated Levack Gneiss and later mafic

intrusive bodies of different types.

Generally, FWGRs could be traced up to 500 m below the sublayer (Figure 6(b)), but

as seen also at Wisner, the abundance of FWGRs does not decrease gradationally from the

contact. Instead, zones of increased partial melting were observed, which often correspond

to the occurrence of SDBX zones.

The appearance of melt bodies is variable, but they mostly resemble the types described at

Frost. They are either pink or white, consisting mainly of quartz and feldspar: very common is

the coarse-grained granophyric to graphic type (Figure 4(b) and (c)), similar to felsic melt

bodies at Frost. Fine-grained, almost aplitic melt bodies, as well as textures dominated by

euhedral feldspar laths (up to 1 cm in length) are also common. This euhedral texture always

occurs in mafic gneisses and gabbros (Figure 4(d) and (f)), while the other types occur without

any dependence on host rock or distance from the SIC. Miarolitic cavities with quartz and

epidote up to 5 cm in diameter are abundant, especially in melt bodies hosted by felsic rocks

or SDBX (Figure 4(b)). A unique type of in situ to segregated FWGRs was observed in

gabbros of drill hole WWL-009 and -009B; they are dominantly made up of white, mm-sized

euhedral feldspar laths and euhedral biotite flakes up to 1 cm in length.

Two other areas without known footwall-style sulphide occurrences were included to

gain a more regional overview, especially on unmineralized parts of the footwall.

At Skynner (Figure 1), melt bodies occur in brecciated gneisses and mafic intrusives up to

1 km east of the SIC/footwall contact. Their appearance is very similar to those described

from the Frost area to the south of Skynner. At Foy, however we observed similar pink,

aplitic FWGRs containing small (mm-sized) miarolitic cavities as seen in the Wisner

localities, which are situated approximately 12 km to the NE. The aplitic melt bodies occur

in a SDBX belt that parallels the SIC contact, 900 m to the north.

Petrography and mineral chemistry of FWGRs

Felsic FWGRs of all studied areas consist of over 95 vol % quartz, plagioclase, and

K-feldspar, which occur in a variety of undercooling textures, which we describe following

the nomenclature by MacLellan and Trembath (1991). At Wisner, a crystallization sequence

of quartz-feldspar intergrowths may be observed (Figure 7(a)): granophyric (vermicular)–

micrographic (cuneiform)–micropoikilitic, which may finally develop into separate

subhedral to euhedral crystals of quartz and feldspar growing into miarolitic cavities. In

most of the samples from Frost and Windy Lake, euhedral plagioclase grains are suspended in



Figure 6. (a) Map of the Windy Lake area with the location of investigated drill holes, showing alsocontact metamorphic hornfels facies according to Boast and Spray (2006). (b) Stratigraphic columnsof two selected drill holes showing the occurrence of FWGRs (column a) indicating also their

A. Pentek et al.12


a groundmass made up of different quartz-feldspar intergrowths (Figure 7(b)); overall, these

textures are much more coarse-grained than those at Wisner. The chemical composition of

K-feldspar and plagioclase appears to be independent of texture and stage of crystallization:

Kfs is Or95–100; plagioclase is An0–6. Feldspars in melt bodies containing centimetre-sized

miarolites often exhibit intense sericitic alteration and replacement by clinozoisite and calcite;

in other cases, however, the grains are fresh.

The minerals accompanying this quartzo-feldspathic groundmass show only slight

variation within and between the localities. Amphibole (hornblende to actinolite) and

biotite occur as rare aggregates up to a few tens of micrometres, and rare epidote in

symplectitic intergrowth with albite. More abundant are opaque needles up to a few

millimetres in length, which have a core of rutile and ilmenite surrounded by titanite. Trace

apatite, zircon, and allanite also occur, but are more common in the miarolitic cavities.

FWGRs in mafic rocks (at Frost and Windy Lake) are dominated by euhedral

plagioclase laths (An26 – 48) with quartz-feldspar intergrowths (containing almost pure

albite and K-feldspar) occurring only in the interstitial spaces (Figure 7(c)). Zoned

amphibole ranging from hornblende to actinolite compositions is always present either as

millimetre-sized euhedral crystals or in the interstitial space between plagioclase laths.

The mafic melt bodies at Windy Lake are characterized also by abundant euhedral biotite

up to 1 cm in length (Figure 7(d)), which were absent at Frost. Apatite, allanite, and zircon

occur usually in trace amounts.

Miarolitic cavities (,1 mm to 5 cm) occur in all samples and represent the final stage

of crystallization. They are dominantly filled by subhedral quartz, albite, and K-feldspar,

which are usually accompanied by a variety of minerals. Aggregates of columnar,

chemically zoned epidote (Figure 7(e)) up to a few millimetres in length are very common

and often have an allanitic core. Euhedral or interstitial titanite (up to 1 mm in diameter) is

another abundant phase, often containing corroded inclusions of ilmenite. Amphibole may

occur as euhedral crystals up to a few millimetres in length in the miarolites or as ‘clouds’

of acicular inclusions within miarolitic quartz. They are usually actinolitic and ferro-

actinolitic in composition with a very important characteristic of such amphiboles at Frost

being the constantly high Ni contents (up to 0.5 wt % NiO). Irregular aggregates of

‘spongy’ (poikilitic) to acicular zircon often occur as inclusions within or along grain

boundaries of miarolitic quartz and albite and may be accompanied by euhedral apatite.

Independent from their occurrence in the groundmass or the miarolitic cavities,

hydrous silicates and apatite tend to be F-rich and contain less Cl: apatite has F contents of

2.3–3.5 wt %, with F/(F þ Cl) values over 0.90, while amphibole and biotite contain up to

0.7 wt % F and 0.2 wt % Cl.

In situ evidence of partial melting

While FWGRs represent segregated partial melts, evidence for in situ partial melting was

also observed at all localities. In granitic rocks or felsic parts of gneisses and leucosomes,

the originally white feldspar grains are often pink, and tiny (1–2 mm in diameter)

miarolitic cavities, as well as graphic textured pods may appear in them (Figure 4(c)).

textural types: coarse-grained granophyric-graphic (column b); dominated by euhedral feldspar laths(column c) and fine-grained, aplitic (column d). Yellow layers in stratigraphic column representSDBX zones. ab, Albite; ep, epidote; hbl, hornblende; px, pyroxene; FB, footwall breccia; GB,gabbro; LGC, Levack Gneiss Complex; and SL, sublayer.

R



Figure 7. Photomicrographs of FWGRs and partial melting textures. (a) A typical sequence ofquartz-feldspar intergrowth in Wisner melt bodies. The first stage is vermicular, granophyric, whichdevelops into cuneiform micrographic that is followed by micropoikilitic intergrowth in whichanhedral to subhedral quartz grains are enclosed in continuous feldspar. (b) Typical groundmass ofFrost melt bodies. Subhedral laths of plagioclase are included in quartz or irregular quartz-feldspar

A. Pentek et al.14


While partial melting of felsic rocks was observed in all localities, gabbro and mafic gneiss

showed signs of melting only at Frost and Windy Lake.

Partial melting and segregation of melt into FWGR veins can be traced

microscopically. Primary feldspar and quartz in the felsic rocks that have undergone

partial melting show typical contact metamorphic recrystallization features described

from other parts of the aureole (e.g. Dressler 1984b; Boast and Spray 2006): feldspar is

replaced by a fine-grained mass of decussate, stubby plagioclase (An22 – 28), while quartz is

recrystallized to an aggregate of clear, isometric grains with dihedral angles close to 1208.

These quartz aggregates often show cuspate and embayed grain boundaries towards

feldspar, which are signs of incipient partial melting. A more developed stage of melting is

represented by films and pods of granophyric to graphic quartz-feldspar intergrowths

along the grain boundaries resorbing the quartz aggregate (Figure 7(g)).

In gabbroic rocks from Frost ALZ and drill hole WC-19, we observed assemblages

dominated by fresh hornblende and stubby plagioclase (, An50 – 55), which are

accompanied by biotite, Fe–Ti oxides, and minor clinopyroxene (Wo44En45Fs11). Former

greenschist to amphibolite facies metamorphic assemblages of these rocks are preserved

as corroded, residual grains of actinolite, Mg–Fe amphibole, and epidote enclosed in fresh

hornblende. Crystallized partial melt is represented by narrow films of albite along

grain boundaries of hornblende and plagioclase, as well as pods and segregations on the

scale of tens to hundreds of micrometres in the matrix of the rock containing quartz,

albite ^K-feldspar.

In contrast to the mafic units at Frost, the assemblages at Windy Lake are dominated by

clinopyroxene (Wo40 – 42En42 – 44Fs15 – 16), rather than hornblende (Figure 7(f)). These are

accompanied by fresh, stubby plagioclase (An35 – 48), biotite, Fe–Ti oxides, apatite and in

one sample by strongly serpentinized orthopyroxene (Wo3En64Fs33). Remnants of

strongly corroded actinolite are enclosed within clinopyroxene. Pods of albite (An3 – 5) or

alkali feldspar (Or59 – 65Ab30 – 35An5) coexisting with plagioclase and biotite represent

crystallized partial melt. In a few samples, plagioclase and biotite are aligned, indicating

melt flow during recrystallization of the rock.

In all areas and rock types it can commonly be observed how in situ melt pods connect

and segregate to form microveins, and finally FWGR veins and dikes (Figures 7(h) and 8).

This is a clear evidence, that many of the FWGRs mapped in this study did not intrude far

distances, but crystallized actually from partial melts formed in the same area.

intergrowth. (c) Melt body dominated by euhedral plagioclase laths within a gabbroic rock fromWindy Lake. Interstitial space is made up of graphic quartz-feldspar intergrowth containing pure K-feldspar and albite. Plagioclase shows intense clinozoisite and sericite alteration. Note also allanitegrain. (d) Miarolitic cavity in FWGR hosted by gabbroic rock at Windy Lake. The cavity is filled byan assemblage of actinolitic to ferro-actinolitic amphibole, biotite, titanite, albite, K-feldspar, andquartz. (e) Group of euhedral epidote crystals in quartz filling a miarolitic cavity (Frost). Grainsexhibit chemical sector zonation demonstrated here by a difference in interference colours.(f) Restite assemblage of a partially melted gabbroic rock from Windy Lake. The rock is dominantlymade up of fresh clinopyroxene, plagioclase, magnetite, and biotite, but contains also corrodedremnants of the reactant actinolite. (g) Partially melted felsic gneiss from Wisner SW. Granophyricintergrowth represent in situ crystallized partial melt pools along the grain boundaries ofrecrystallized rock-forming quartz and feldspar grains. (h) Partially melted gabbroic rock from drillhole WC-19 in Frost. The groundmass is dominated by fresh hornblende, recrystallized feldspar, andremnants of actinolite. Former partial melt is represented by pods of quartz, feldspar, and biotite.Note the segregation of partial melt into a FWGR vein in the lower part of picture. act, actinolite; all,allanite; bt, biotite; cpx, clinopyroxene; hbl, hornblende; mag, magnetite; pl, plagioclase; qtz, quartz;ttn, titanite.

R



Bulk rock chemical composition

Host rocks

In addition to FWGRs, we also sampled their brecciated host rocks for geochemical analysis.

However, they can neither be regarded as protoliths, nor as restites, because of their

heterogeneous character, as well as the partial melting and later hydrothermal processes.

Therefore, we compare the composition of FWGRs with reported data on similar units

outside of the SIC contact aureole, which may be regarded as the best approximation of the

possible protoliths: CB rocks (Meldrum et al. 1997); LGC rocks (Prevec 1993; Ostermann

1996; Meldrum et al. 1997); and mafic dikes (Lightfoot and Naldrett 1996; Phinney and

Halls 2001). As a comparison to LGC, we also use our felsic LGC samples from fresh parts

of Windy Lake drill holes, because reported averages usually included mafic parts of the

gneisses. Major and trace element compositions of our samples of gabbros from Frost ALZ

suggest that they belong to the Nipissing suite (Table 1). The gabbro at Windy Lake,

however, has element concentrations and ratios, which do not give any clear information

about their relationship to any of the known mafic suites.

FWGRs

A total of 25 FWGR samples from three areas (Wisner South, Wisner Southwest referred to

as Wisner S and SW below, and Frost ALZ) were analysed for major and trace elements. To

avoid contamination with the host rock, and because of possible inhomogeneity resulting

from their coarse-grained character, as well as the abundance of miarolitic cavities, we could

only analyse melt bodies, which allowed a sample size of at least 250 cm3.

The major element composition of FWGRs at the two Wisner localities are essentially

identical (Table 1), therefore these two areas are discussed together. The SiO2 contents of

melt bodies range from 74 to 78 wt %, with an average of 75.5 wt %. The concentration of

Al2O3, FeO, MgO, and CaO shows only little variation and mostly overlaps with felsic

Cartier granitoids (Figure 9). The large variation of Na2O (2–6 wt %) and K2O

(1–6 wt %), as well as the strong negative correlation of these elements (r ¼ 20.92)

corresponds well with the observed albite and K-feldspar contents of individual samples.

Figure 8. Thin section maps showing the distribution of crystallized partial melts in (a) IGN and (b)SDBX. These maps illustrate how melt pods segregate and collect into veins, which contain restitefragments. Note the absence of melt pods in the mafic part of the gneiss.

A. Pentek et al.16


Tab

le1

.C

hem

ical

com

po

siti

on

of

FW

GR

s,h

ost

rock

s,an

dp

oss

ible

pro

toli

ths.

FW

GR

sH

ost

rock

san

dp

roto

lith

s

Wis

Sav

g.

(n¼

3)

Wis

SW

avg

.(n

¼1

2)

Fr

AL

Zav

g.

fels

ic(n

¼7

)F

rA

LZ

80

42

73

FL

GC

avg

.(n

¼8

)C

Bav

g.

GB

Fr

avg

.(n

¼2

)G

BW

l7

04

06

1M

atav

g.

Nip

avg

.

SiO

27

6.4

77

5.4

07

7.3

37

0.1

06

4.2

66

8.4

95

1.0

55

0.1

04

9.7

15

1.8

1A

l 2O

31

2.0

01

2.3

11

1.3

07

.70

15

.77

15

.63

14

.15

13

.30

12

.43

14

.47

FeO

a1

.21

1.2

11

.12

5.3

14

.40

2.4

81

1.0

21

4.3

11

5.6

61

0.2

9C

aO1

.55

1.3

21

.66

5.5

74

.60

2.7

39

.84

8.7

49

.13

10

.02

Mg

O0

.38

0.3

30

.56

5.0

32

.16

1.0

56

.88

5.0

34

.61

8.1

0N

a 2O

3.7

73

.82

5.8

02

.65

4.1

14

.53

2.2

63

.25

2.6

32

.12

K2O

3.3

73

.54

0.3

60

.16

1.4

92

.74

0.7

00

.53

0.5

50

.87

TiO

20

.26

0.2

80

.29

0.3

10

.58

0.2

80

.68

1.5

22

.17

0.6

5M

nO

0.1

90

.04

0.0

20

.10

0.0

70

.03

0.1

90

.20

0.2

40

.19

LO

I0

.39

1.2

30

.59

0.9

71

.47

1.3

31

.80

1.2

42

.10

1.4

4T

ota

l9

9.5

89

9.4

89

9.0

19

7.9

09

8.9

39

9.2

99

8.7

09

8.5

69

9.2

49

9.9

6

Pb

12

6.6

72

23

.33

10

3.5

01

70

.00

99

6.2

55

16

.08

39

2.6

71

06

0.0

01

23

2.1

93

05

.48

Ba

72

0.0

01

32

5.0

01

05

.00

13

2.0

05

87

.50

10

84

.09

20

5.0

03

30

.00

16

0.5

92

90

.00

Rb

45

.00

51

.00

6.0

02

.00

22

.41

43

.37

22

.55

14

.10

22

.53

32

.00

Sr

25

4.0

01

38

.00

12

1.0

01

63

.50

56

6.2

56

19

.64

23

3.5

05

33

.00

17

2.1

21

69

.00

Cs

0.1

60

.24

0.0

90

.06

0.3

90

.23

0.4

50

.67

1.4

00

.68

Y6

.87

6.5

38

.74

15

.00

12

.34

3.2

71

5.5

01

8.7

04

2.8

81

6.0

0Z

r9

4.2

71

11

.46

16

1.2

79

8.8

04

5.2

66

6.3

76

3.5

03

7.5

01

64

.29

66

.00

Hf

2.7

03

.05

4.7

63

.10

1.6

61

.79

1.8

01

.60

4.3

81

.20

Nb

6.5

35

.03

7.4

14

.20

6.0

02

.12

2.4

54

.20

n.a

.5

.10

Ta

0.3

60

.27

0.5

10

.30

0.2

80

.08

0.2

00

.28

0.7

10

.18

V1

7.0

01

7.0

81

7.2

96

1.0

08

1.2

53

5.0

02

41

.00

35

3.0

03

95

.47

22

4.0

0T

h1

1.1

09

.13

7.1

24

.40

2.1

51

.90

1.5

01

.90

2.3

11

.69

U1

.23

0.8

51

.26

0.9

00

.28

0.2

80

.29

0.4

00

.61

0.5

0L

a1

9.7

32

0.3

01

0.9

41

9.8

03

2.0

32

2.9

48

.55

20

.90

16

.31

6.2

9C

e4

1.4

03

7.8

32

4.2

73

8.6

06

6.2

43

8.5

81

7.7

04

6.1

03

7.9

01

3.8

7P

r4

.67

4.4

52

.79

4.2

88

.07

4.2

92

.38

6.0

7n

.a.

1.7

9N

d1

6.0

71

5.3

11

0.2

91

6.5

03

0.8

51

4.4

01

0.2

02

5.7

02

4.0

67

.37

Sm

2.7

92

.59

1.8

43

.19

5.2

22

.06

2.4

85

.49

5.8

62

.08



Tab

le1

–continued

FW

GR

sH

ost

rock

san

dp

roto

lith

s

Wis

Sav

g.

(n¼

3)

Wis

SW

avg

.(n

¼1

2)

Fr

AL

Zav

g.

fels

ic(n

¼7

)F

rA

LZ

80

42

73

FL

GC

avg

.(n

¼8

)C

Bav

g.

GB

Fr

avg

.(n

¼2

)G

BW

l7

04

06

1M

atav

g.

Nip

avg

.

Eu

0.4

00

.61

0.4

41

.08

1.4

41

.09

0.7

81

.67

1.8

30

.65

Gd

2.5

12

.39

1.7

43

.27

4.3

11

.81

2.6

35

.07

n.a

.0

.39

Tb

0.3

10

.29

0.2

40

.48

0.5

20

.18

0.4

50

.73

1.1

22

.50

Dy

1.4

11

.36

1.4

42

.74

2.5

50

.69

2.9

23

.99

n.a

.2

.26

Ho

0.2

90

.27

0.3

10

.55

0.4

80

.13

0.6

30

.76

n.a

.0

.69

Er

0.7

80

.76

0.9

81

.65

1.3

20

.33

1.8

02

.07

n.a

.1

.66

Tm

0.1

00

.11

0.1

60

.22

0.1

80

.05

0.2

80

.28

n.a

.0

.44

Yb

0.7

10

.70

1.0

21

.55

1.1

00

.27

1.7

41

.67

4.0

31

.43

Lu

0.1

10

.11

0.1

50

.22

0.1

60

.05

0.2

60

.24

0.6

00

.22

Cu

46

.43

23

.31

31

.33

11

3.0

02

2.3

41

1.8

1n

.a.

20

5.0

0n

.a.

13

0.0

0N

i1

4.3

78

.45

51

.66

92

4.0

03

1.0

31

3.0

3n

.a.

59

.20

43

.72

13

8.0

0P

t,

0.0

05

,0

.00

50

.02

0.3

0,

0.0

05

,0

.00

5n

.a.

,0

.00

5n

.a.

0.1

0P

d,

0.0

01

,0

.00

10

.04

0.4

2,

0.0

01

,0

.00

1n

.a.

,0

.00

1n

.a.

n.a

.L

a/L

u1

86

.76

19

4.0

07

7.7

89

0.0

02

37

.68

33

5.0

03

3.1

38

7.0

82

7.0

72

8.5

9C

e/Y

bN

16

.12

14

.99

6.6

06

.92

16

.67

29

.40

2.8

67

.67

2.6

12

.69

Th

/U8

.40

10

.48

8.0

84

.89

6.9

06

.20

5.2

44

.75

3.7

93

.38

All

val

ues

are

aver

ages

,ex

cep

tfo

rse

lect

edsa

mple

s8

04

27

3(a

mp

hib

ole

-ric

hm

elt

bo

dy)

and

70

406

1(W

indy

Lak

eg

abb

ro).

Dat

aso

urc

es:

CB

(Mel

dru

metal.

19

97);

Mat

(Ph

inn

eyan

dH

alls

20

01);

Nip

(Lig

htf

oo

tan

dN

ald

rett

19

96).Location:

Wis

S,

Wis

ner

So

uth

;W

isS

W,

Wis

ner

So

uth

wes

t;F

rA

LZ

,F

rost

AL

Z.Rock

type:

FL

GC

,fe

lsic

Lev

ack

Gn

eiss

Com

ple

x;

CB

,C

arti

erB

ath

oli

th;

GB

Fr,

gab

bro

Fro

st;

GB

Wl,

gab

bro

Win

dy

Lak

e;M

at,

Mat

achew

anin

tru

siv

es;

Nip

,N

ipis

sin

gin

tru

siv

es;

avg

.,av

erag

e;n,

nu

mb

ero

fan

aly

ses;

n.a

.,n

ot

anal

yse

d.

aT

ota

lir

on

exp

ress

edas

FeO

.b

All

trac

eel

emen

tsin

ppm

.

A. Pentek et al.18


Figure 9. Harker diagrams showing concentrations of major elements in FWGR samples (alwaysas a function of SiO2), as well as their Levack gneiss, Cartier granitoid, and gabbroic host rocks.



In the Qtz–Ab–Or plot of normative compositions (Figure 10), FWGRs project between

the 0.5 and 2 kbar cotectic lines of water-saturated haplogranite (Qtz–Ab–Or), with the

average (Qtz41Ab36Or23) on the 0.5 kbar cotectic near the minimum melt composition.

In the An–Ab–Or ternary, melt bodies plot into the trondhjemite and granite fields,

with the average composition (An8Ab56Or36) in the Ab-rich part of the granite field.

The rocks are metaluminous to peraluminous, with average Al2O3/(Na2O þ K2O) and

Al2O3/(CaO þ Na2O þ K2O) values of 1.22 and 0.99, respectively.

The concentration of SiO2, Al2O3, FeO, MgO, and CaO in felsic FWGRs at Frost have

again little variation with averages relatively similar to the Wisner samples. However, in

contrast to melt bodies at Wisner, Na2O and K2O concentrations also show little variation.

All samples are Na2O-rich and there is no significant negative correlation (r ¼ 20.41) of

the two elements. The Na2O-rich character of these melt bodies is also evident in the Qtz–

Ab–Or ternary of normative compositions, where all samples plot near the intersection of

the 0.5 and 1 kbar cotectic lines with the Qtz–Ab join. These rocks are all metaluminous

(average A/NK ¼ 1.14 and A/CNK ¼ 0.88) and have a trondhjemitic composition. The

FWGRs containing abundant amphibole from Frost have major element concentrations

quite similar to the felsic ones, but are shifted towards the composition of Frost gabbro

(see sample 804273 in Figure 9 and Table 1).

The concentration of Sr is quite similar in FWGRs of Wisner SW and Frost, while it is

a bit higher at Wisner S (Figure 11). In contrast, concentrations of Ba, Rb, and Cs show

large variation within and between the studied occurrences, usually with much lower

Plotted are also fields of the CB, LGC, footwall breccia matrix (FWBX), ‘remobilized footwallbreccia and granophyric dikes’ (FWBX dikes), as well as averages of Matachewan and Nipissingintrusions (data from Coats and Snajdr 1984; Dressler 1984b; Lakomy 1990; Lightfoot and Naldrett1996; Meldrum et al. 1997; Phinney and Halls 2001; McCormick et al. 2002b). Fr, Frost; Ma,Matachewan; Ni, Nipissing; Wl, Windy Lake; 73, amphibole-rich sample 804273 and Avg., average.

R

Figure 10. Qtz–Ab–Or normative plot of FWGR samples and the averages of CB and LGC (CBdata from Meldrum et al. 1997; LGC data from Prevec 1993; Ostermann 1996; Meldrum et al. 1997and own samples), as well as Wisner FWGR samples. Plotted are also cotectic lines for 0.5, 1, and2 kbar, as well as minimum melt compositions for 1, 2, 5, and 10 kbar at aH2O ¼ 1 (Tuttle and Bowen1958; Luth et al. 1964). For legends see Figure 9.

A. Pentek et al.20


values at Frost. In accordance with this, Rb/Sr values are also higher at Wisner S and SW

(0.17 and 0.37), than at Frost (0.05). The scatter of these elements mimics that of K2O: at

Frost, Rb, Ba, and Cs show strong positive correlation with K2O wt % (r ¼ 1.00, 0.95 and

0.77, respectively), whereas at Wisner, Rb and Ba correlate positively with it (r ¼ 0.94

and 0.61).

Compared with LGC and CB rocks, FWGRs are overall depleted in REE (Figure 12).

While at Wisner concentrations are quite uniform, at Frost they are within a wide range

(Figure 12(c)). If normalized to average CB, the REE pattern of melt bodies from Wisner S

and SW are almost identical with a smooth increase towards the HREE, while the Frost

samples have a more steeply increasing trend, which cuts the Wisner patterns (Figure

12(a)). Normalized to an average LGC composition (gained using own samples and

published data by Prevec 1993; Ostermann 1996; Meldrum et al. 1997;) we get very

irregular REE patterns, but if we normalize it to the average of our own felsic to

intermediate Levack gneiss samples, the patterns are again smooth (Figure 12(b)). In this

latter plot, Wisner samples show an almost horizontal, slightly increasing trend, while the

melt bodies from Frost are again more steeply increasing; a small negative Eu anomaly

may be observed in all localities. These patterns result in the more shallow decreasing

trend of FWGRs on the chondrite-normalized plot in comparison to LGC and CB (Figure

12(c); see also La/Lu values in Table 1). Chondrite-normalized Ce/Yb values (Ce/YbN)

Figure 11. Ba versus Sr diagram and Harker diagrams showing Zr, P, and U concentrations ofFWGR samples, as well as their Levack gneiss, Cartier granitoid and gabbroic host rocks. Forlegends, abbreviations, and data source see Figure 9.



are also very useful in determining the possible source rock of the FWGRs. At Wisner S

and SW this value averages 16.1 and 15.0, respectively, which is very similar to the

average value of our LGC samples (16.7), but much lower than that of average CB (29.4).

At Frost, melt bodies have an average Ce/YbN value of only 6.6, which is more

comparable to that of gabbroic rocks in the same area (Table 1).

In FWGRs, Zr and U (and similarly Hf and Th) concentrations show a wide variation,

with the fields of separate localities overlapping (Figure 11) and Th/U ratios varying

greatly between 2.3 and 40.8. There is an enrichment of these elements in comparison to

LGC and CB samples, but they are still lower than the published average CB values. The P

contents of melt bodies fit well into the trend of decreasing P concentrations with

increasing SiO2 wt % shown by the LGC and CB samples: data points of the two Wisner

localities overlap well and show a quite limited range, while the more felsic samples from

Frost have slightly lower contents.

Comparison with footwall breccia

As a comparison, we have also plotted data of footwall breccia matrix reported by Coats

and Snajdr (1984), Dressler (1984b), Lakomy (1990) and McCormick et al. (2002b), as

well as samples of ‘footwall breccia dikes’ (Lakomy 1990), ‘remobilized footwall breccia’

and ‘granophyric dike’ (Dressler 1984b) in Figures 9 and 11. The fields outlined by the

major element concentrations of footwall breccia matrix from various localities in the

Sudbury structure usually have a wide range and overlap with the fields of LGC, the CB, as

well as Nipissing and Matachewan intrusives. A compositional similarity of footwall

breccia matrix, with adjacent Levack gneisses was already observed by Lakomy (1990). In

comparison to FWGRs, the limited reported data (Lakomy 1990; McCormick et al. 2002b)

on Ba and Sr concentrations of footwall breccia matrix reveal a relatively similar pattern,

although shifted towards higher Sr values. The composition of the remobilized footwall

breccia dikes is similar to the FWGRs of this study with the slightly lower SiO2 contents

(73.4 wt % in average) being the only difference.

Figure 12. Average REE patterns of FWGRs normalized to (a) average CB; (b) felsic LGC (ownsamples) and (c) chondrite normalized plot showing the range of FWGR patterns (grey field, FrostALZ; solid line, Wisner S) and patterns of average CB and LGC (data source: see Figure 10).

A. Pentek et al.22


Thermobarometry

An attempt to estimate the pressure during the formation of FWGRs was made by Molnar

et al. (2001), who used the total Al contents of hornblende, which resulted in values from

1.1 to 1.6 kbar. Assuming lithostatic conditions, this corresponds to about 4–6 km depth of

crystallization, which is in good agreement with the estimated 5 km depth of erosion in the

Sudbury structure (Dressler 1984b). In our melt bodies, only few amphibole grains

enabled us to use Al-in-hornblende barometers (e.g. Hammarstrom and Zen 1986;

Hollister et al. 1987; Anderson and Smith 1995), with resulting values between 0.7 and

2.1 kbar (Table 2). The fact that our FWGR samples plot between the 0.5 and 2 kbar

cotectic lines on the Qtz–Ab–Or ternary is compatible with these data. Overall, we

consider 1.5 ^ 0.5 kbar as the most realistic pressure estimate for the partial melting

process and the crystallization of these melts.

In our attempt to obtain the crystallization temperatures of segregated FWGRs, the

almost pure composition of albite and K-feldspar and the low Na-contents of amphibole in

most of the samples did not enable us to use amphibole–plagioclase thermometry or

ternary feldspar thermometers. The very pure composition of feldspars can only be

attributed in some cases to later hydrothermal alteration, in most instances however they are

absolutely fresh and are regarded as primary compositions. The occurrence of magmatic

albite and albite-poor K-feldspar was described by Tuttle and Bowen (1958), who

suggested that unmixing of plagioclase and potassium feldspar will go on if crystallization

takes place in the presence of volatile phases. This is confirmed for example by experiments

of London et al. (1989) who observed almost pure albite at temperatures of 575–6008C. For

those FWGRs, which crystallized within mafic rocks, we also tried to use the semi-

quantitative, graphic thermometer based on the TiO2 contents of amphiboles, which was

calibrated for MORB compositions (Ernst and Liu 1998). The resulting amphibole

crystallization temperatures are from 7808C down to below 5508C, with the latter derived

from actinolites growing in miarolitic cavities. The quartz-feldspar textures, which make up

the groundmass of the melt bodies, also bear information on the crystallization

temperatures: in their experimental studies, MacLellan and Trembath (1991) observed

granophyric and micropoikilitic intergrowth within the same overlapping temperature

range, between 730 and 6508C. However, micropoikilitic textures were found to persist to

lower temperatures in accordance with the crystallization sequence observed in our

samples, in which micropoikilitic textures overgrow the granophyric ones. In the same

experiments, skeletal intergrowth (common at Frost and Windy Lake) crystallized from

800 down to 6008C.

In contrast, hornblende and plagioclase in the mafic restites from Frost and Windy

Lake show clear textural evidence for equilibrium crystallization (i.e. they are intergrown

and included in each other) and were also suitable from their composition for amphibole–

plagioclase (edenite-richterite) thermometry (using the HbPl 1.2 software according to

Holland and Blundy 1994). The resulting temperatures (for P ¼ 0 kbar) are 807–877 at

Frost and 826–934 at Windy Lake (Table 2). Using the semi-quantitative thermometer

by Ernst and Liu (1998), the temperatures are much lower (up to 8808C, but usually

between 700–8008C), however, relative temperature changes are similar to changes

shown by the amphibole–plagioclase thermometer. This suggests that this thermometer

may be used in our case to reveal relative temperature differences between samples, but

does not give reliable absolute temperatures. The intimate intergrowth of alkali feldspar

and plagioclase in partial melt pools within restites from Windy Lake enabled the use

of ternary feldspar solution thermometers by means of the SOLVCALC computer



program (Wen and Nekvasil 1994). The average temperatures between 810 and 8708C (for

P ¼ 1–2 kbar) are interpreted to represent the crystallization temperature of the

partial melt pods and are in good agreement with our results on amphibole and plagioclase

pairs.

An estimate of the partial melting temperature could also be given by the zircon

saturation thermometer of Watson and Harrison (1983) as our samples contained newly

formed, acicular and spongy (poikilitic) zircon grains and had Zr concentrations varying

within a narrow range. The resulting temperatures are 757 ^ 198C at Frost, 749 ^ 228C at

Wisner Southwest and 727 ^ 218C at Wisner South. As we will see, these temperatures

are in agreement with petrogenetic considerations of the partial melting temperature in the

Wisner area. In the case of Frost, however, the resulting temperature is approximately

1008C lower, than the temperatures gained from amphibole–plagioclase and that

suggested by melting experiments.

Table 2. Results of thermobarometric calculations of FWGR and restite assemblages obtainedusing the following methods: Al in hornblende barometry (Hammarstrom and Zen 1986; Hollisteret al. 1987; Anderson and Smith 1995); amphibole–plagioclase thermometry (Holland and Blundy1994); ternary feldspar solution thermometry (using various models in the SOLVCALC computerprogram; Wen and Nekvasil 1994); Zr saturation thermometry (Watson and Harrison 1983).

Al in hornblende barometry (kbar)Number of

analysed grains (n)Hammarstrom &

Zen (1986) Hollister et al. (1987)Anderson &Smith (1995)

FR387 3 1.3 1.1 1.5 (0.9)WL233 10 1.0 0.7 1.3 (0.7)WISSWR2 1 1.7 1.5 2.1 (1.4)

Amphibole–plagioclase (edenite-richterite) thermometry (T8C)

Number ofanalysed pairs (n) 0 kbar 5 kbar

FR137 7 836 (807–862) 844 (819–866)FRALZ378 5 864 (854–877) 862 (852–873)WL224 4 856 (826–884) 844 (815–877)WL233 3 890 (865–934) 888 (866–926)

Ternary feldspar solution thermometry (T8C)

Number ofanalysed pairs (n) 1 kbar 1.5 kbar 2 kbar

WL231 7 809 (775–840) 811 (773–844) 813 (780–842)WL232/1 3 822 (789–851) 836 (793–877) 836 (795–893)WL232/2 3 870 (829–905) 872 (830–912) 867 (832–918)

Zr saturation thermometry (T8C)

Number ofsamples (n) Zr (ppm) M T (8C)

Frost ALZ 6 155 (77–241) 1.71 (1.46–2.18) 757 (722–774)Wisner S 3 94 (63–130) 1.57 (1.48–1.73) 727 (704–745)Wisner SW 12 111 (73–150) 1.48 (1.12–1.84) 750 (710–784)

FR, Frost drillhole WC-19; FRALZ, Frost ALZ; WL, Windy Lake and M, (Na þ K þ (2 £ Ca))/(AlxSi) in bulkrock sample.

A. Pentek et al.24


Discussion

Partial melting of granitoids and felsic gneisses

In our aim to understand the conditions of partial melting below the SIC, a key role is given to

textures, which are generally interpreted as signs of partial melting (e.g. Mehnert et al. 1973;

Sawyer 1999, 2001; Clemens and Holness 2000; Rosenberg and Riller 2000) and which

we commonly observed in felsic rocks of all localities: (1) granophyric quartz-feldspar

films and pockets along the grain boundaries of rock-forming quartz and feldspar and (2)

irregular, cuspate or embayed grain boundaries of rock-forming quartz and feldspar. The

absence of such textures around the primary hydrous phases (biotite and amphibole) indicates

that the melting reactions in the case of these rocks involved dominantly quartz, plagioclase,

and K-feldspar and can thus be described in the Qtz–Or–Ab–An–H2O system.

In the water-saturated haplogranite system (Qtz–Or–Ab–H2O), melting will start at a

temperature of approximately 7008C at the estimated 1.5 kbar pressure (e.g. Tuttle and

Bowen 1958; Johannes and Holtz 1996 and references therein; Figure 13). Although our

protoliths cannot be considered as haplogranites, because the composition of plagioclase

in these rocks is usually between An20 and An30, experiments prove that the difference

in the solidus temperature of the granite system is less than 108C up to plagioclase

compositions of An40 (Johannes 1984). In the water-saturated subsystems Qtz–Ab and

Qtz–Or, melting begins between 750 and 8008C at 1.5 kbar pressure (see references

above). However, even in rocks containing quartz and only one type of feldspar, melting

may start at the water-saturated eutectic of the Qtz–Or–Ab–An system if the fluid phase

carries the missing Na or K. Melting experiments with natural granitoid rocks confirm the

results obtained in the above mentioned model systems (e.g. Piwinskii and Wyllie 1970;

Piwinskii 1973; Figure 13). Some studies (e.g. Wyllie and Tuttle 1964; Manning 1981;

Pichavant 1981) have shown that the addition of F, B2O3, and Li results in the decrease

of solidus temperatures by up to 1008C in granitic systems. Although we have no data

indicating elevated concentrations of these elements in the protoliths or the FWGRs

themselves, we cannot rule out the possibility that fluids present during the partial melting

process had significant contents of these elements (e.g. high F contents of SIC-related

hydrothermal fluids is suggested by Jago et al. (1994)).

In our opinion, it is necessary to assume water-saturated, or at least water-present

conditions during partial melting of the felsic rocks, especially in the case of Wisner. Melting

in a dry Qtz–Or–Ab–An system would require roughly 10008C (Johannes and Holtz 1996

and references therein), which high temperature may have obtained right at the SIC contact,

but is quite unrealistic several hundred metres into the footwall. Our observations on the

spatial distribution of FWGRs are also indicating fluid-present conditions of partial melting.

As we have seen, the intensity of partial melting does not decrease gradually away from the

contact, as one would assume in a dry system. Instead, we observed zones of increased partial

melting, which often correspond with SDBX zones. In several cases, we observed

that unbrecciated granitoids did not show any signs of partial melting, while brecciated zones

at the same distance from the contact hosted abundant FWGRs. The better development

of partial melting in SDBX zones is, in our opinion, caused mainly by the presence of

fluids, but was probably also promoted by the fine-grained character of the breccia matrix.

Partial melting of mafic rocks

Experimental studies on natural basalts, amphibolites, and gabbros have shown that partial

melting of such amphibole-bearing rocks produces metaluminous to peraluminous



quartzo-feldspathic melts. Most of the experiments were performed under high pressures

and water-absent conditions to model the generation of granitic magmas by dehydration

melting processes in the lower crust (e.g. Rushmer 1991; Wyllie and Wolf 1993; Johannes

and Wolke 1994). However, a few experiments were also done at low pressures relevant to

our topic and/or under water-saturated conditions (Heltz 1976; Spulber and Rutherford

1983; Beard and Lofgren 1991; Koepke et al. 2004). From these studies we may conclude

that the beginning of dehydration melting of amphibole-bearing assemblages at 1–2 kbar

pressure requires temperatures in the range of 850–9008C (Figure 13) and that there is no

significant difference in the solidus temperature assuming water-saturated conditions.

The chemical and modal composition of the rock does not appear to have a great effect on

the location of the solidus, either.

While Beard and Lofgren (1991; Equations (2) and (3)) described melting reactions

resulting in the formation of clino- and orthopyroxenes, Heltz (1976; Equation (1)) also

observed reactions, which did not yield pyroxenes. Below we list a few reactions by these

authors:

Hbl þ An50 þ Fe–Ti oxides 5 HblðMg;Al-richÞ þ An75 þ Fe–Ti oxides þ Melt: ð1Þ

9 Qtz þ 33 Amph þ 5 Plag 5 2 Mgt þ 6 Opx þ 20 Cpx þ 19 Melt : ð2Þ

29 Plag þ 15 Amph þ 3 Ilm 5 3 Mgt þ 9 Cpx þ 35 Melt: ð3Þ

Note that restite plagioclase is more calcic and restite amphibole is richer in AlIV, Mg,

Ti, and A-site cations than the starting compositions.

Figure 13. P–T diagram showing solidus curves of the water-saturated systems (1) Qtz–Ab–Or,(2) Qtz–An40–Kfs, (3) Qtz–Ab, (4) Qtz–Kfs, the dry system (5) Qtz–Ab–Kfs, as well as thedehydration melting of amphibolites (6–9). Plotted are also solidus curves of natural granitoid rocksas determined by Piwinskii and Wyllie (1970) and Piwinskii (1973) (grey fields; QM, quartzmonzonite; GRD, granodiorite; TON, tonalite); the effect of the addition of 1 and 2 wt % F on thelocation of the solidus at 1 kbar (Manning 1981). Dashed lines highlight the estimated 1.5 ^ 0.5 kbarfor the pressure during partial melting in the SIC footwall. Data sources: 1–5, Johannes and Holtz(1996) and references therein; 6, Beard and Lofgren (1991); 7, Johannes and Wolke (1994); 8,Spulber and Rutherford (1983) and 9, Wyllie and Wolf (1993).

A. Pentek et al.26


Although the exact modal composition of the protoliths is not known, corroded

remnants of actinolite, Mg–Fe amphiboles ^ epidote in the mafic samples from Frost

indicate that they may have been greenschist facies metagabbros. For the gabbros at Frost

ALZ, our geochemical data implies that they belonged to the Nipissing suite, but in the

case of mafic rocks from hole WC-19, we have no reliable information. The residual and

newly formed assemblages in the restites suggest the following general melting reaction

which resembles Heltz’s reaction (1) listed above ( þ Fe–Ti oxides on both sides):

Act þ PlagðAn,50Þ^ Mg–Fe Amph ^ Ep ¼ Hbl þ PlagðAn50–55Þ þ Melt ^ Cpx:

According to hornblende–plagioclase thermometry, this melting reaction may have

occurred at temperatures of at least 8808C, which is in good agreement with the

experimentally determined solidi of mafic rocks described above.

For the mafic restites at Windy Lake we have no information on the peak temperature

of partial melting, however, using ternary feldspar thermometry we could reveal that

crystallization of partial melt to plagioclase and alkali feldspar may have occurred from

about 870 down to 8008C. Although the rocks are now dominantly made up of anhydrous

products of the melting reaction or phases crystallized from the melt, strongly corroded

silicic amphiboles also occur, which are reactants of the melting process. We have to

emphasize that the modal composition of the protolith is again unknown, but the few

reactant remnants and the produced assemblages suggest melting reactions very similar to

those described by Beard and Lofgren (1991; see reactions (2) and (3) above).

Melt segregation and its structural control

Generally, the ability of granitic melts to segregate from their host rocks depends on many

factors including the intensity of syn-melting deformation, the rate and amount of melt

produced, the melt viscosity, and host rock fabric (e.g. Clemens and Droop 1998; Sawyer

2001). Some authors also emphasize that fluid-absent melting involving the breakdown

of a hydrous silicate is accompanied by bulk volume increase (positive DVmelting) and

fluid-present melting has a negative DVmelting (e.g. Brown and Rushmer 1997;

Clemens and Droop 1998). The increase in volume during fluid-absent melting results

in an increase of the pore pressure, which promotes melt segregation. However, the effect

of tectonic deformation during partial melting appears to be of much greater significance

(Rushmer 2001). Especially in the case of granitic melts (in contrast to low-viscosity

basalts) is the density contrast to the host rock not enough for compaction-driven melt

segregation. In these cases, high strain rates effectively assist the extraction of melt (e.g.

Brown and Rushmer 1997; Rutter 1997; Sawyer 2001). Although it is usually believed that

melt segregation can only occur once a certain minimum melt fraction is reached,

experiments by Rosenberg and Handy (2001) show that under dynamic conditions, melt

migration may occur at melt fractions as low as 1–5 vol %.

As we have seen, there is a continuity of melt segregation from grain scale to outcrop

scale in all our studied localities. Our mapping revealed that pre-existing inhomogeneities

acted as channels for melt migration. Melt formed in felsic bands of Levack gneiss often

migrated along foliation (Figure 2) and was emplaced into pre-existing shear zones, which

are now filled by FWGR dikes. In a similar way, SDBX veins represent channels along which

melt could preferably migrate. However, we have shown that veins and dikes of FWGR cut

also through homogeneous granitic rocks with preferred structural orientations. In these cases

it is clear that melt segregation occurred under deformation and migration was mainly



controlled by strain directions. Rosenberg and Handy (2001) revealed that melt migration

may either occur subparallel to the principal shortening direction, or subperpendicular to it

and the extent to which these former melt migration channels are preserved depends on this

orientation. Especially, in the Wisner area, FWGRs show a preferred strike of SE–NW,

which is subparallel to the principal shortening direction of the Penokean orogeny. In one of

the Wisner trenches however, the FWGRs have a dominant strike of ENE–WSW, which is

subparallel to the SIC/footwall contact. In this case it may be possible that strain related to

crater modification during the crystallization of the SIC may have promoted the migration of

melt, or that melt used pre-existing brittle structures related to the cratering process.

The relatively straight veins and dikes having sharp contacts toward the host rock

indicate that they intruded under brittle conditions; these usually occur in unbrecciated

rocks or along veins or dikes of SDBX. However, matrix supported SDBX zones within

the same areas contain melt bodies which may be extremely sheared, smeared onto clasts

and have diffuse contacts toward the matrix. In the Frost area, we observed that even small

clasts are plastically deformed and elongated. These observations indicate that partial

melting and melt migration occurred under both ductile and brittle conditions. As pointed

out by Fournier (1999), the ductile vs. brittle behaviour of rocks depends on many factors.

For example, the presence of water allows plastic behaviour at low temperatures, while

high strain rates promote brittle behaviour at higher (even magmatic) temperatures. It is

therefore probable that increased strain rates due to the Penokean orogeny facilitated

movement of the partial melt along dilatant brittle structures mainly subparallel to the

principal shortening direction in unbrecciated rocks, while the water-rich SDBX zones

were still in a ductile condition.

Crystallization of FWGRs: the transition from magmatic to hydrothermal processes

An important question regarding the crystallization of the melt bodies is whether they

could have migrated long distances (in the range of several 100 m) or if they froze rapidly

within tens of metres after segregation. Near minimum melts just above their solidus are

incapable of large-scale migration as either decompression or a decrease in temperature

causes rapid crystallization (Droop and Clemens 1998). This is demonstrated in our case

by crystallization of partial melt on thin-section scale or where segregated partial melt

crystallized in veins within one metre of the source rock. However, the occurrence of

brittle veins and dikes, which may be followed up to 10 m distance and zones of several

parallel-running ‘rootless’ veins (Figure 2), indicate that there may have been also a more

large-scaled melt migration along focused channels. It has to be emphasized, that the

conditions in the contact aureole of the SIC did not follow the ‘ordinary’ trend of

downward increasing pressure and temperature. There, the pressure and temperature

gradients were either opposite (assuming a simple, flat-lying contact) or changed in a

complicated manner due to the irregular contact in embayment structures. This means, that

partial melts may have migrated longer distances without crystallizing if the conditions

allowed them to stay above their solidus. The fact that we observed partial melting in

footwall rocks even up to approximately 500 m from the SIC implies that melts formed

near the contact may have migrated several hundred metres into the footwall without

experiencing a decrease in temperature below the solidus.

It is evident that melt bodies containing abundant miarolitic cavities reached water

saturation during their crystallization. Molnar et al. (2001) described primary, high

salinity fluid inclusions in quartz from FWGRs and pointed out that these fluids probably

played a significant role in the hydrothermal system active during the crystallization of the

A. Pentek et al.28


SIC. Large, centimetre-sized miarolites usually occur in irregular melt bodies which

crystallized under ductile conditions, while veins and dikes emplaced under brittle

conditions contain only mm-sized cavities. It is very likely that fluids exsolving from the

partial melts under plastic conditions could not leave the system and were therefore

‘trapped’ in the crystallizing melt pod. In these cases, large miarolites formed, which are

filled by hydrothermally precipitated minerals (quartz, actinolite, epidote) and earlier

formed magmatic phases are strongly altered. During crystallization of veins and dikes,

the exsolving fluid could leave the crystallizing melt and migrate along the same brittle

structures. Not only these fluids, but also the Cu–Ni–PGE ore-forming hydrothermal

fluids migrated along the same structures, as seen at South Zone (Pentek et al. 2008). Apart

from the spatial association of Cu–Ni–PGE sulphide occurrences and FWGRs in many

cases, there are other indications that these ore-forming fluids not only used the same

channels, but were present syngenetically with the crystallizing partial melt: (1) PGE-

bearing sulphide–silicate veins in the continuation of FWGRs; (2) PGE-bearing sulphide

patches and inclusions within the miarolites of melt bodies and (3) high Ni contents of

amphiboles crystallized in melt bodies from the Frost area.

Summary and implications for the significance of partial melting in hydrothermal

Cu–Ni–PGE deposition in the footwall of the SIC

We observed FWGRs in all studied localities of the North Range (Windy Lake, Foy, and

Wisner) and the East Range (Skynner and Frost) of the Sudbury structure, while

Molnar et al. (2001) described melt bodies from the Onaping–Levack embayment

(Craig, Hardy, and Barnet areas). These results of regional scale confirm earlier

observations and models describing the cooling history of the Sudbury melt sheet, which

suggested that partial melting of footwall rocks may have been widespread in the contact

aureole of the SIC. Putting our results in the context of recent knowledge on the cooling

history of the melt sheet, we summarize the following points regarding partial melting in

the footwall of the SIC:

(1) The impact melt sheet, now preserved as the SIC was superheated to an initial

temperature of at least 20008C (Ivanov and Deutsch 1999; Prevec and Cawthorn

2002; Zieg and Marsh 2005). Temperatures of 1100–12008C at the basal contact

induced complete melting of the proximal rocks and led to footwall assimilation of

up to 800 m thickness (Prevec and Cawthorn 2002).

(2) When cooling of the melt sheet reached a stage where temperatures at the

immediate footwall were not enough to consume all rock types, the downward

eroding assimilation front came to a halt and a chaotically mixed zone formed.

The contact sublayer and footwall breccia have been recently interpreted to

represent this mixed zone (Prevec et al. 2000; Prevec and Cawthorn 2002) with

igneous matrices of different composition and restitic clasts, which could not be

further assimilated. The dioritic matrix of the footwall breccia near the sublayer is

more and more granitic toward the footwall (Lakomy 1990; McCormick et al.

2002a) which is in accordance with assimilation focused increasingly on felsic

lithologies as temperatures decrease away from the SIC. The major element

composition of remobilized footwall breccia dikes described by Dressler (1984b)

and Lakomy (1990) fit well between the compositional data of footwall breccia

matrix and melt bodies in the footwall sampled in our study, which suggests that

these felsic lithologies represent partial melt bodies formed at the contact, similar

to those described by Molnar et al. (2001).



(3) Prevec and Cawthorn (2002) suggested that partial melting of footwall rocks

may have occurred to depths of 200–500 m below the melt sheet. Our field

observations and drill hole mapping confirm this result as we could trace partial

melting to a depth of approximately 500 m below the base of the sublayer.

However, the degree of partial melting does not decrease gradually away from the

contact, but appears to have been localized, especially to SDBX zones. Outside

this distance however, FWGRs do not exist even in the Sudbury brecciated zones.

(4) Melting of mafic rocks in the Windy Lake and Frost areas suggests that temperatures

in the contact aureole may have reached 850–9008C up to at least 200 m from the

contact. It is possible however, that such high temperatures were only attained

below embayment structures, where the thickened melt sheet transmitted more heat

into the footwall. In these zones, mafic rocks could partially melt either under water-

absent conditions (dehydration melting) or in the presence of fluids.

(5) In other areas (e.g. Wisner localities), which were either more distal to the original

contact or situated below SIC outside of embayment structures, temperatures

reached only about 7508C, mafic rocks did not undergo partial melting and melting

of felsic rocks required water-saturated conditions. Thus actual fluid saturation of

rocks defined localization of melting processes. Zones of SDBX probably

contained pore fluids derived from the near footwall at or after brecciation and

were therefore more water-rich than unbrecciated granites and gneisses. SDBX

zones also acted as conduits for SIC-related melts and fluids (Rousell et al. 2003),

thus fluid flow along breccia zones may also have been important in promoting

partial melting. Of course, fluid flow may have focused also on other conduits

(such as brittle structures, contacts etc.) and facilitate partial melting in felsic

rocks. Similar low-temperature partial melting driven by water infiltration into

a contact aureole was suggested by Holness and Clemens (1999) for the

Ballachulish Complex, as well as by Harris et al. (2003) for the Bushveld

Complex. As a result of this irregular occurrence of FWGR it is not possible to

establish concentric zones of decreasing partial melting, similar to those described

according to the contact metamorphic mineral assemblages (Dressler 1984b;

Hanley and Mungall 2003).

(6) The difference in partial melting temperature and the contribution of melts derived

from various protoliths is responsible for the textural, mineralogical and

compositional differences among FWGRs from Windy Lake, Frost, and Wisner

areas. While the bulk rock composition of Wisner samples is in accordance with

partial melting of only felsic LGC and probably CB units, data of FWGRs at Frost

indicates also significant contribution from the melting of gabbroic rocks.

(7) Melt segregation occurred in all localities and was promoted by the ongoing

deformation during the Penokean orogeny and pre-existing zones of weaknesses

in the host rock (e.g. foliation, SDBX veins). While veins and dikes reflect brittle

conditions during melt migration, sheared melt pods in SDBX matrix indicate

ductile conditions of the matrix during partial melt crystallization.

Finally, we want to emphasize that there is a close genetic association of partial

melting, segregation of FWGRs and the hydrothermal processes responsible for sulphide

remobilization into and/or within the footwall. Although the actual physical connection is

not preserved, we suggest that melt bodies occurring together with hydrothermal sulphides

at Frost ALZ and Wisner South may have originated from contact or footwall

environments hosting pre-existing SIC-related sulphide mineralization. Molnar et al.

A. Pentek et al.30


(2001) described FWGRs cutting sulphide-bearing footwall breccia and showed that high

temperature and high salinity fluids exsolving from them were capable of leaching out

mobile metals (including Cu, Pt and Pd) from the magmatic sulphides. Strong alteration

and halogen halos within sulphide-bearing sublayer and footwall breccia evidence the

interaction of hydrothermal fluids with the sulphides (McCormick and McDonald 1999;

McCormick et al. 2002a). The close association of FWGRs and hydrothermal sulphides at

Wisner and Frost ALZ suggests that partial melts and metal-bearing fluids may have

migrated into and/or within the footwall using the same structures.

Obviously, partial melting also occurred in contact environments that did not host

magmatic sulphides and as our study revealed, also several hundred metres into the footwall.

In these cases, we will observe similar FWGRs and associated hydrothermal features, but

because fluids exsolving from the crystallizing partial melts had no interaction with pre-

existing sulphides, there will be no spatial association with footwall Cu–Ni–PGE sulphides

(e.g. Frost West, Skynner, Foy). Overall, partial melting processes described in this paper

appear to have importance also from an economic standpoint and further studies need to focus

on the significance of these processes in hydrothermal Cu–Ni–PGE mineralization within

the contact aureole of the SIC as well as other large igneous complexes hosting Ni–Cu–PGE

sulphides (such as Duluth or Bushveld), where partial melting of footwall rocks also appears

to have occurred (e.g. Sawyer 2002; Harris et al. 2003; Johnson et al. 2003).

Acknowledgements

We would like to acknowledge Wallbridge Mining Company Ltd. for the financial and logisticalsupport of this study and especially Bruce Jago, Alar Soever, and Joshua Bailey for their help anddiscussions. The work was funded by the Canada-Hungary Science and Technology AgreementProject CAN-02/04 to F. M. and D. H. W., NSERC grant no. 7874 to D. H. W., the HAESF SeniorResearcher Fellowship to F. M., as well as the Society of Economic Geologists Foundation HughE. McKinstry Student Research Award and CEEPUS scholarships to A. P.

References

Ames, D.E., and Farrow, C.E.G., 2007, Metallogeny of the Sudbury mining camp, Ontario, inGoodfellow, W.D., ed., Mineral deposits of Canada: a synthesis of major deposit-types, districtmetallogeny, the evolution of geological provinces, and exploration methods: GeologicalAssociation of Canada, Mineral Deposits Division, Special Publication, p. 329–350.

Anderson, J.L., and Smith, D.R., 1995, The effects of temperature and fO2 on the Al-in-hornblendebarometer: American Mineralogist, v. 80, p. 549–559.

Beard, J.S., and Lofgren, G.E., 1991, Dehydration melting and water-saturated melting of basalticand andesitic greenstones and amphibolites at 1, 3, and 6.9 kb: Journal of Petrology, v. 32,p. 365–401.

Boast, M., and Spray, J.G., 2006, Superimposition of a thrust–transfer fault system on a large impactstructure: implications for Ni–Cu–PGE exploration at sudbury: Economic Geology, v. 101,p. 1583–1594.

Brown, M., and Rushmer, T., 1997, The role of deformation in the movement of granitic melt: viewsfrom the laboratory and the field, in Holness, M.B., ed., Deformation-enhanced fluid transport inthe Earth’s crust and mantle: Mineralogical Society Series 8, p. 82–110.

Card, K.D., 1994, Geology of the Levack Gneiss Complex, the northern footwall of the Sudburystructure: Ontario Geological Survey Paper 94-1C, p. 1–10.

Card, K.D., Gupta, V.K., McGrath, P.H., and Grant, F.S., 1984, The Sudbury structure: its regionalgeological and geophysical setting: Ontario Geological Survey Special Volume 1, p. 25–44.

Clemens, J.D., and Droop, G.T.R., 1998, Fluids, P–T paths and the fates of anatectic melts in theEarth’s crust: Lithos, v. 44, p. 21–36.

Clemens, J.D., and Holness, M.B., 2000, Textural evolution and partial melting of arkose in a contactaureole: a case study and implications: Visual Geosciences, v. 5, no. 4, p. 1–14.



Coats, C.J.A., and Snajdr, P., 1984, Ore deposits of the North Range, onaping–levack area, Sudbury:Ontario Geological Survey Special Volume 1, p. 327–346.

Collins, W.H., 1936, The life history of the Sudbury Nickel irruptive, Part III, environment:Transactions of the Royal Society of Canada, 3rd Series, v. 30, p. 29–53.

Corfu, F., and Andrews, A.J., 1986, A U-Pb age for mineralized Nipissing diabase, Gowganda,Ontario: Canadian Journal of Earth Sciences, v. 23, p. 107–109.

Cowan, E.J., and Schwerdtner, W.M., 1994, Fold origin in the Sudbury basin, in Lightfoot, P.C., andNaldrett, A.J., eds., Proceedings of the Sudbury–Noril’sk Symposium: Ontario GeologicalSurvey Special Volume 5, p. 45–55.

Davis, G.C., 1984, Little Stobie mine: a south range contact deposit: Ontario Geological SurveySpecial Volume 1, p. 361–370.

Deutsch, A., Grieve, R.A.F., Avermann, M., Bischoff, L., Brockmeyer, P., Buhl, D., Lakomy, R.,Muller-Mohr, V., Ostermann, M., and Stoffler, D., 1995, The Sudbury structure (Ontario,Canada): a tectonically deformed multi-ring impact basin: Geologische Rundschau, v. 84,p. 697–709.

Dietz, R.S., 1964, Sudbury structure as an Astrobleme: Journal of Geology, v. 72, p. 412–434.Dressler, B.O., 1984a, General geology of the Sudbury area: Ontario Geological Survey Special

Volume 1, p. 57–82.Dressler, B.O., 1984b, The effects of the Sudbury event and the intrusion of the Sudbury Igneous

Complex on the footwall rocks of the Sudbury structure: Ontario Geological Survey SpecialVolume 1, p. 97–138.

Ernst, W.G., and Liu, J., 1998, Experimental phase-equilibrium study of Al- and Ti-contents ofcalcic amphibole in MORB-a semiquantitative thermobarometer: American Mineralogist, v. 83,p. 952–969.

Fournier, R.O., 1999, Hydrothermal processes related to movement of fluid from plastic to brittlerock in the magmatic-epithermal environment: Economic Geology, v. 94, p. 1193–1211.

Grieve, R.A.F., Stoffler, D., and Deutsch, A., 1991, The Sudbury structure-controversial ormisunderstood: Journal of Geophysical Research, v. 96, p. 22753–22764.

Hammarstrom, J.M., and Zen, E-an, 1986, Aluminum in hornblende: an empirical geobarometer:American Mineralogist, v. 71, p. 1297–1313.

Hanley, J.J., and Mungall, J.E., 2003, Chlorine enrichment and hydrous alteration of the SudburyBreccia hosting footwall Cu–Ni–PGE mineralization at the Fraser Mine, Sudbury, Ontario,Canada: Canadian Mineralogist, v. 41, p. 857–881.

Harris, N., McMillan, A., Holness, M., Uken, R., Watkeys, M., Rogers, N., and Fallick, A., 2003,Melt generation and fluid flow in the thermal aureole of the Bushveld Complex: Journal ofPetrology, v. 44, no. 6, p. 1031–1054.

Heaman, L.M., 1997, Global mafic magmatism at 2.45 Ga: remnants of an ancient large igneousprovince?: Geology, v. 25, p. 299–302.

Heltz, R.T., 1976, Phase relations of basalts in their melting range at PH2O ¼ 5 kb. II meltcompositions: Journal of Petrology, v. 14, p. 249–302.

Holland, T., and Blundy, J., 1994, Non-ideal interactions in calcic amphiboles and their bearing onamphibole–plagioclase thermometry: Contributions to Mineralogy and Petrology, v. 116,p. 443–447.

Hollister, L.S., Grissom, G.C., Peters, E.K., Stowell, H.H., and Sisson, V.B., 1987, Confirmation ofthe empirical correlation of aluminium in hornblende with pressure of solidification of calc-alkaline plutons: American Mineralogist, v. 72, p. 231–239.

Holness, M.B., and Clemens, J.D., 1999, Partial melting of the Appin Quartzite driven by fracture-controlled H2O infiltration in the aureole of the Ballachulish Igneous Complex, ScottishHighlands: Contributions to Mineralogy and Petrology, v. 136, p. 154–168.

Ivanov, B.A., and Deutsch, A., 1999, Sudbury impact event: Cratering mechanics and thermalhistory, in Dressler, B.O., and Sharpton, V.L., eds., Large meteorite impacts and planetaryevolution II: GSA Special Paper #339. Boulder: Geological Society of America, p. 389–398.

Jago, B.C., Morrison, G.G., and Little, T.L., 1994, Metal zonation patterns and microtextural andmicromineralogical evidence for alkali- and halogen-rich fluids in the genesis of the Victor Deepand McCreedy East footwall copper orebodies, Sudbury Igneous Complex: Ontario GeologicalSurvey Special Volume 5, p. 65–75.

A. Pentek et al.32


James, R.S., Sweeny, J.M., and Peredery, W., 1991, Thermobarometry of the Levack Gneisses-footwall rocks to the Sudbury Igneous Complex: Lothoprobe, Abitibi-GrenvilleTransect Report,v. 32, p. 179–182.

Johannes, ., 1984, Beginning of melting in the granite system Qz–Or–Ab–An–H2O: Contributionsto Mineralogy and Petrology, v. 86, p. 264–273.

Johannes, W., and Holtz, F., 1996, Petrogenesis and experimental petrology of granitic rocks: Berlin:Springer-Verlag, p. 335.

Johannes, W., and Wolke, C., 1994, Bildung von tonalitischen Schmelzen, ein Schritt in derEntwicklung der kontinentalen Kruste: European Journal of Mineralogy, v. 6, p. 129.

Johnson, T.E., Gibson, R.L., Brown, M., Buick, I.S., and Cartwright, I., 2003, Partial melting ofmetapelitic rocks beneath the Bushveld Complex, South Africa: Journal of Petrology, v. 44,p. 789–813.

Koepke, J., Feig, S.T., Snow, J., and Freise, M., 2004, Petrogenesis of oceanic plagiogranites bypartial melting of gabbros: an experimental study: Contributions to Mineralogy and Petrology,v. 146, p. 414–432.

Krogh, T.E., Davis, D.W., and Corfu, F., 1984, Precise U–Pb zircon and baddeleyite ages for theSudbury Area: Ontario Geological Survey Special Volume 1, p. 431–446.

Lakomy, R., 1990, Implications for cratering mechanics from a study of the Footwall Breccia of theSudbury impact structure, Canada: Meteoritics, v. 25, p. 195–207.

Lightfoot, P.C., Doherty, W., Farrell, K., Keays, R.R., Moore, M., and Pekeski, D., 1997,Geochemistry of the main mass, Sublayer, Offsets and Inclusions from the Sudbury IgneousComplex, Ontario: Ontario Geological Survey, p. 231 Open File Report 5959.

Lightfoot, P.C., and Naldrett, A.J., 1996, Petrology and geochemistry of the Nipissing gabbro:exploration strategies for nickel, copper and platinum group elements in a large igneousprovince: Study, v. 58, p. 81, Ontario Geological Survey.

London, D., Morgan, G.B., and Hervig, R.L., 1989, Vapor-undersaturated experiments withMacusani glass þH2O at 200 MPa, and the internal differentiation of granitic pegmatites:Contributions to Mineralogy and Petrology, v. 102, p. 1–17.

Luth, W.C., Jahns, R.H., and Tuttle, O.F., 1964, The granite system at pressure of 4 to 10 kilobars:Journal of Geophysical Research, v. 69, p. 759–773.

Manning, D.A.C., 1981, The effect of fluorine on liquidus phase relationships in the systemQz–Ab–Or with excess water at 1 kbar: Contributions to Mineralogy and Petrology, v. 76,p. 206–215.

Marshall, D.D., Watkinson, D.H., Farrow, C.E.G., Molnar, F., and Fouillac, A.M., 1999, Multiplefluid generations in the Sudbury Igneous Complex: fluid inclusion, Ar, O, Rb and Sr evidence:Chemical Geology, v. 157, p. 1–19.

McCormick, K.A., Fedorowich, J.S., McDonald, A.M., and James, R.S., 2002a, A textural,mineralogical, and statistical study of the footwall Breccia within the Strathcona Embayment ofthe Sudbury structure: Economic Geology, v. 97, p. 125–143.

McCormick, K.A., Lesher, C.M., McDonald, A.M., Fedorowich, J.S., and James, R.S., 2002b,Chlorine and Alkali Geochemical Halos in the Footwall Breccia and Sublayer Norite at theMargin of the Strathcona Embayment, Sudbury Structure, Ontario: Economic Geology, v. 97,p. 1509–1519.

Mehnert, K.R., Busch, W., and Schneider, G., 1973, Initial melting at grain boundaries of quartz andfeldspar in gneisses and granulites: Neues Jahrbuch fur Mineralogie Monatshefte, v. 4,p. 165–183.

Meldrum, A., Abdel-Rahman, A.-F.M., Martin, R.F., and Wodicka, N., 1997, The nature, age andpetrogenesis of the Cartier Batholith, northern flank of the Sudbury structure, Ontario, Canada:Precambrian Research, v. 82, p. 265–285.

Molnar, F., Watkinson, D.H., and Jones, P.C., 2001, Multiple hydrothermal processes in footwallunits of the North Range, Sudbury Igneous Complex, Canada, and implications for the genesis ofvein-type Cu–Ni–PGE deposits: Economic Geologist, v. 96, p. 1645–1670.

Morrison, G.G., 1984, Morphological features of the Sudbury structure in relation to an impactorigin: Ontario Geological Survey Special Volume 1, p. 513–522.

Naldrett, A.J., and Hewins, R.H., 1984, The main mass of the Sudbury Igneous Complex: OntarioGeological Survey Special Volume 1, p. 235–252.

Naldrett, A.J., Hewins, R.H., Dressler, B.O., and Rao, B.V., 1984, The contact sublayer of theSudbury Igneous Complex: Ontario Geological Survey Special Volume 1, p. 253–274.



Ostermann, M., 1996, Die Geochemie Impaktschmelzdecke (Sudbury Igneous Complex) imMultiring-Becken Sudbury: [Ph.D. thesis]: Germany, Westfalische Wilhelms-UniversitatMunster, 118 p. (in German)

Pentek, 2006, The Broken Hammer Zone and other hydrothermal Cu–Ni–PGE occurrences in thefootwall of the Sudbury Igneous Complex, Wisner Property, North Range, Sudbury Structure:[M.Sc. thesis]: Budapest, Hungary, Eotvos Lorand University, 146 p. (In Hungarian)

Pentek, A., Molnar, F., Watkinson, D.H., and Jones, P.C., 2008, Footwall-type Cu–Ni–PGEmineralization in the Broken Hammer area, Wisner Township, North Range, Sudbury Structure:Economic Geology, v. 103, p. 1005–1028.

Phinney, W.C., and Halls, H.C., 2001, Petrogenesis of the Early Proterozoic Matachewan dykeswarm, Canada, and implications for magma emplacement and subsequent deformation:Canadian Journal of Earth Sciences, v. 38, p. 1541–1563.

Pichavant, M., 1981, An experimantal study of the effect of boron on a water-saturated haplograniteat 1 kbar pressure: geological applications: Contributions to Mineralogy and Petrology, v. 76,p. 430–439.

Piwinskii, A.J., 1973, Experimental studies of igneous rock series, central Sierra Nevada batholith,California: Part II: Neues Jahrbuch fur Mineralogie Monatshefte, v. 5, p. 193–215.

Piwinskii, A.J., and Wyllie, P.J., 1970, Experimental studies of igneous rock series: felsic body suitefrom the Needle Point pluton, Wallowa batholith, Oregon: Journal of Geology, v. 78, p. 52–76.

Pope, K.O., Kieffer, S.W., and Ames, D.E., 2004, Empirical and theoretical comparison of theChixculub and Sudbury impact structures: Meteoritics and Planetary Science, v. 39, p. 97–116.

Prevec, S.A., 1993, An isotopic, geochemical, petrographical investigation of the genesis of EarlyProterozoic mafic intrusions and associated volcanism near Sudbury, Ontario: [Ph.D. thesis]:Edmonton, Alberta, Canada, University of Alberta, 165 p.

Prevec, S.A., Lightfoot, P.C., and Keays, R.R., 2000, Evolution of the sublayer of the SudburyIgneous Complex: geochemical, Sm–Nd isotopic and petrologic evidence: Lithos, v. 51,p. 271–292.

Prevec, S.A., and Cawthorn, R.G., 2002, Thermal evolution and interaction between impact meltsheet and footwall: a genetic model for the contact sublayer of the Sudbury Igneous Complex,Canada: Journal of Geophysical Research, v. 107, p. B8, doi: 10.1029/2001JB000525.

Riller, U., 2005, Structural characteristics of the Sudbury impact structure, Canada: impact-inducedversus orogenic deformation-a review: Meteoritics and Planetary Science, v. 40, p. 1723–1740.

Rosenberg, C.L., and Handy, M.R., 2001, Mechanisms and orientation of melt segregation pathsduring pure shearing of a partially molten rock analog (norcamphor–benzamide): Journal ofStructural Geology, v. 23, p. 1917–1932.

Rosenberg, C.L., and Riller, U., 2000, Partial-melt topology in statically and dynamicallyrecrystallized granite: Geology, v. 28, p. 7–10.

Rousell, D.H., 1984, Structural geology of the Sudbury Basin: Ontario Geological Survey SpecialVolume 1, p. 83–96.

Rousell, D.H., Fedorowich, J.S., and Dressler, B.O., 2003, Sudbury Breccia (Canada): a product ofthe 1850 Ma Sudbury Event and host to Cu–Ni–PGE deposits: Earth Science Reviews, v. 60,p. 147–174.

Rushmer, T., 1991, Partial melting of two amphibolite: contrasting experimental results under fluid-absent conditions: Contributions to Mineralogy and Petrology, v. 107, p. 41–59.

Rushmer, T., 2001, Volume change during partial melting reactions: implications for meltextraction, melt geochemistry and crustal rheology: Tectonophysics, v. 342, p. 389–405.

Rutter, E.H., 1997, The influence of deformation on the extraction of crustal melts, in Holness, M.B.,ed., Deformation-enhanced fluid transport in the Earth’s crust and mantle: Mineralogical SocietySeries 5, p. 82–110.

Sawyer, E.W., 1999, Criteria for the recognition of partial melting: Physics and Chemistry of theEarth, Part A, v. 24, no. 3, p. 269–279.

Sawyer, E.W., 2001, Melt segregation in the continental crust: distribution and movement of melt inanatectic rocks: Journal of Metamorphic Geology, v. 19, p. 291–309.

Sawyer, E.W., 2002, Report on thin sections from DDH WM-1, Spruce Road Cu–Ni Deposit, SouthKawishiwi Intrusion, Duluth Complex: Natural Resources Research Institute, University ofMinnesota, Duluth, Report of Investigations, NRRI/RI-2002/13, 13 p.

A. Pentek et al.34


Sims, P.K., Van Schmus, W.R., Schulz, K.J., and Petermann, Z.E., 1989, Tectono-stratigraphicevolution of the Early Proterozoic Wisconsin magmatic terranes of the Penokean orogen:Canadian Journal of Earth Sciences, v. 26, p. 2145–2158.

Spray, J.G., 1997, Superfaults: Geology, v. 25, p. 579–582.Spulber, S.D., and Rutherford, M.J., 1983, The origin of rhyolite and plagiogranite in oceanic crust:

an experimental study: Journal of Petrology, v. 24, p. 1–25.Therriault, A.M., Fowler, A.D., and Grieve, R.A.F., 2002, The Sudbury Igneous Complex: a

differentiated impact melt sheet: Economic Geology, v. 97, p. 1521–1540.Tuttle, O.F., and Bowen, N.L., 1958, Origin of granite in the light of experimental studies in the

system NaAlSi3O8–KAlSi3O8–SiO2–H2O: Geological Society of America Memoir, v. 74,p. 153.

Watson, E.B., and Harrison, T.M., 1983, Zircon saturation revisited: temperature and compositioneffects in a variety of crustal magma types: Earth and Planetary Science Letters, v. 64,p. 295–304.

Wen, S., and Nekvasil, H., 1994, SOLVCALC Computer Program for feldspar thermometry:Computers and Geosciences, v. 20, p. 1025–1040.

Wyllie, P.J., and Wolf, M.B., 1993, Amphibole dehydration-melting: sorting out the solidus, inPrichard, H.M., Alabaster, T., Harris, N.B.W., and Neary, C.R., eds., Magmatic processes andplate tectonics: Geological Society London Special Publications, v. 76, p. 405–416.

Wyllie, P.J., and Tuttle, O.F., 1964, Experimental investigation of silicate systems containing twovolatile components. Part III. The effects of SO3, P2O5, HCl, and Li2O in addition to H2O on themelting temperatures of albite and granite: American Journal of Sciences, v. 262, p. 930–939.

Zieg, M.J., and Marsh, B.D., 2005, The Sudbury Igneous Complex: viscous emulsion differentiationof a superheated impact melt sheet: Geological Society of America Bulletin, v. 117,p. 1427–1450.



Documents

Partial melting and melt segregation in footwall units within the contact aureole of the Sudbury Igneous Complex (North and East Ranges, Sudbury structure), with implications for their