University of Southern California Dissertations and Theses

Magmatic foliations and layering: Implications for process in magma chambers

(USC Thesis Other)

Magmatic foliations and layering: Implications for process in magma chambers

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content IN FO R M A TIO N T O USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afreet reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MAGMATIC FOLIATIONS AND LAYERING: IMPLICATIONS FOR PROCESSES IN MAGMA CHAMBERS by Elizabeth Semele Yuan A Thesis Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE (Earth Sciences) August 1996 Copyright 1996, Elizabeth Semele Yuan. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 1381613 UMI Microform 138I6I3 Copyright 1996, by UMI Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. UMI 300 North Zeeb Road Ann Arbor, MI 48103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UNIVERSITY OF SOUTHERN CALIFORNIA T H E G RAD U A TE S C H O O L U N IV ER SITY P A R K LOS ANGELAS. C A L IF O R N IA 80007 This thesis, written by Elizabeth Semele Yuan under the direction of hsx. Thesis Committee, and approved by a ll its members, has been pre sented to and accepted by the Dean of The Graduate School, in partial fulfillm ent of the requirements fo r the degree of Master of Science r w , A u g u s t 13, 1996 THESIS COMMITTEE C hai rm an G y L 3 $ T f — 2 . Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. AC KNOW LE P.G EMEMTS Over the many years of my thesis, there have been countless people whose help was invaluable. There are many whom, no doubt, I have forgotten to mention, but whom I’ll remember long after these few pages have gone to press. To those people I apologise; your help WAS greatly appreciated. Firstly thanks must go to my committee; my adviser, Scott Paterson, for being my mentor, giving me the opportunity and guiding me through this research; Greg Davis, for his helpful comments, advice and help when it was needed most; finally Charlie Sammis, for his encouragement and agreeability. Thanks to Wallace Pitcher, who knows more than anyone else about the geology of Donegal, and who was willing to lend his expertise, time and invaluable information to our research group. Thanks also to the many people who contributed to this research through discussions and fieldwork; Bob Miller, Ron Vemon, Ken Fowler, Andy Crossland, Dave Mayo, Jeff Amato and everyone else in the structure group (all of whom, at one point, have talked with me about The Albatross). Thanks to the people who made the most important and exciting part of the research enjoyable - the fieldwork. In Belfast; thanks to the Hamsons for their unending hospitality, particularly Julie-Anne; a true friend throughout all these years, and Fionnula.; to those in Donegal - the McGills (Brid, Joe and family), who made both summers bearable by ii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. welcoming us into their home and treating us like family. Thanks also to Natalie Stone, who helped start the whole tradition of fieldwork in the rain, through the bogs and among the sheep. Finally, thanks to the one person who kept me sane (and insane, but healthily so) throughout that first season, Lesley Dinnett - without whose companionship, humor and creative cooking I probably would not have lasted the summer. Thanks to those people who helped me immeasurably through the "logistics” - to Rahul Bahadur (the delivery man), and the Main Office Goddesses; Rene Kirby, Cindy Waite and Desser Moton (Matt too, if he doesn't mind being referred to as a goddess); to Nick Brozovic, Vicky Hamilton, Kevin Weng, Jetty Lee, Avijit Chakraborty and Joyjeet Bowmik, all of whom, at one time or another, helped me with maps and graphics; and to Christa Dietzel and Tisa Major for their help with the finer aspects of research (fieldwork, drafting and samples). Finally, thanks to those people who were always there emotionally to support me; to my family, who encouraged me throughout graduate school and have done countless things to help me finish; to Robert Osborne for being a friend and a mentor; to Jennifer Trochez and Adam Woods for always being there to lend sympathetic ears; and to Dave Bowman, who in his characteristic way, was there to help in all aspects - moral support, scientific, logistical (and heck, even transportational). Last, but certainly not least, thanks to Malcolm Webster for being there for me regardless, throughout the years at USC. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This work was supported by grants from the National Science Foundation, Sigma Xi Society of Scientific Research, and the Graduate Student Reseach Fund of the Department if Earth Sciences, USC. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE QF.CflNTEM TS page ACKNOWLEDGEMENTS....................................................................... ii USTOFFIGURES................................................................................... vfi USTOFTABLES..................................................................................... x USTOFPLATES..................................................................................... xi ABSTRACT...............................................................................................xii INTRODUCTION..................................................................................... 1 Chapter 1: THE MAIN DONEGAL GRANITE Generalstatement............................................................................... 7 A) TYPES OF LAYERS i) Regular layering................................................................... 15 Lighter regular layers................................................. 15 Darker regular layers................................................. 21 Contacts....................................................................... 22 ii) Pegmatites.......................................................................... 31 ii) Assimilated layers.................................................................. 32 B) THEMINERALFABRICS............................................................... 34 Chapter 2: .DISCUSSION.......................................................................... 45 A) LAYER FORMING PROCESSES AND IMPLICATIONS 45 Layer Forming Processes Pertinent to the Main Donegal Granite................................................................ 52 0 SUBSOLIDUS PROCESSES.................................. 52 Metasomatism.....................................................52 ii) NEAR SOLIDUS PROCESSES.................................53 Filter Piessina..................................................... 54 iii) MAGMATIC LAYER-FORMING PROCESSES.............................................................. 55 Assimilation..................................................... 55 Liquid immiscibility...........................................55 Magma mixing/tningling....................................... 56 Magma injection...................................................57 B) FOLIATIONS AND UNEATIONS..................................................... 59 i) Classification of fabrics.....................................................59 C) FLOWGEOMETRIES.................................................................... 64 i) Accelerationflow.................................................................. 66 H ) Decelerationftow....................................................................68 in ) Vetodtygradientflow............................................................ 68 D) SUMMARY OF MAGMATIC LAYERING AND FOLIATIONS..................................................................................... 70 E) INTERPRETATION OF LAYERS AND FOLIATIONS INTHEMAINDONEGALGRANITE.................................................... 75 v Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i) THE FORMATION OF THE REGULAR LAYERS................. 75 ii) RELATIONSHIP OF THE LAYERS AND FOLIATIONS................................................................................ 83 F) IMPLICATIONS FOR OTHER ASPECTS OF THE BATHOLITH............................................................................................... 88 Chapter3: CONCLUSIONS........................................................................... 100 REFERENCESCITED....................................................................................105 APPENDIX I A) LAYER FORMING PROCESSES.......................................................113 i) MAGMATIC PROCESSES.......................................................113 Liquid immisa'bility..............................................................113 Magma mixing/mingling................................................... 114 Magma injection................................................................. 114 Assmilation.........................................................................114 ii) NEAR SOLIDUS PROCESSES.............................................. 114 Side wall in s/fucrystallization.........................................115 Crystal settling....................................................................115 Nucleation diffusion...........................................................116 Compositional convection/ Boundary layer fractionation............................................. 117 Double diffusive convection.............................................. 118 Filter pressing.....................................................................118 Flowsorting.........................................................................119 Sintering.............................................................................121 ii) SUBSOLIDUS PROCESSES................................................ 121 Metasomatism....................................................................122 Fluid-enhanced diffusion...................................................122 Stress-induced diffusion....................................................123 Transposition..................................................................... 123 B) LAYER TYPES AND INFERRED FORMATION PROCESSES........................................................................................... 124 Cumulus layering........................................................................... 124 Rhythmiclayering........................................................................... 126 Crypticlayering...............................................................................127 Schlieren........................................................................................127 Pegmatite.......................................................................................127 Sheets............................................................................................128 Mylonites........................................................................................ 128 APPENDIX II: GEOCHEMISTRY: X-RAY FRACTIONATION RESULTS................129-130 vi Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES Figure 1: Location map showing Main Donegal Granite and associated plutons........................................................................... 8 Figure 2: The marginal zone at Losset............................................................10 Figure 3: Map of the Lossett area showing the sheeted nature ofthecontact.............................................................................. 11-12 Figure 4: Simplified map showing sheet margins defined by country rock inclusions.................................................................. 13 Figure 5: Equal area stereonet projection showing poles to planes of regular layering in the Main Donegal........................ 16 Figure 6: Regular layers that depart from their usual NE-SW trend are cross-cut by later regular layers............................................... 17 Rgure 7: Millimeter-scale regular layers.........................................................18 Figure 8: Meter-scale regular layers with parallel magmatic foliations........................................................................................... 18 Rgure 9: Light, microcline-rich layers..............................................................19 Figure 10: Regular layers showing diffuse contacts between light anddarkbands.................................................................................23 Figure 11a: Dark layers cross-cut light layers (photograph)....................25 Figure 11 b: Dark layers cross-cut light layers (line drawing)................... 25 Figure 12a: Regular layering that has been folded and sheared (photograph)..............................................................................26-27 Figure 12b: Regular layering that has been folded and sheared (line drawing)..............................................................................26-27 Figure 13a: Boudined trondjhemitic dark layer being drawn out to form mineral-scale layers........................................................... 28 Figure 13b: Boudined trondjhemitic dark layer............................................. 28 vii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 14: Regular layers and magmatic foliations cross-cut by a late granitic, light layer that shows the beginning stagesof mixing................................................................................29 Figure 15: Pegmatitic dike showing millimeter-scale layers that are a result of fractionation processes...........................................33 Figure 16a: Schlieren that resemble some of the thinner dark bands (photograph).........................................................................35 Figure 16b: Schlieren that resemble some of the thinner dark bands (linedrawing)........................................................................ 35 Figure 17: Equal area stereonet projection showing lineations............ 36 Figure 18: Equal area stereonet projection showing poles to planes offoliations....................................................................................... 37 Figure 19a: Regular layers cross-cut by magmatic foliations (photograph)....................................................................................38 Figure 19b: Regular layers cross-cut by magmatic foliations (linedrawing)................................................................................... 38 Figure 20: S-C fabrics showing sinistral shear in solid-state deformedpegmatite........................................................................ 40 Figure 21: Stereonet plot showing S-C fabric orientations all indicating sinistral shear..................................................................41 Figure 22: Showing shear indicators and strain fields suggested byfolds............................................................................................. 44 Figure 23: Showing that the long axis of the strain ellipse might not be parallel to the flow directions--f!ow types and resulting strain ellipses.................................................................... 67 Figure 24: Showing foliations formed at angles to flow directions and planes...................................................................................... 74 Figure 25: Berger's filter pressing model........................................................77 Figure 26: Layers seperated by steeply-dipping quartzite rafts are “mini-sheets" that represent the emplacement of the magma.............................................................................................82 viii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 27: Diagram showing magma temperature vs. time and sequence of events in the Main Donegal Granite......................87 Figure 28: Schematic diagram showing the sequence of events in the Main Donegal Granite with respect to magma temperature, viscosity and crystal content vs. time and deformation......................................................................................89 Rgure 29: Schematic representation of opening shear zone, from Hutton, 1982.............................................................................91 Figure 30: Geologic map of the Main Donegal Granite............................ 93 Figure 31: Pegmatite dikes intrude the wall rocks at the SE tail ofthepluton...................................................................................... 95 Figure 32: Principle stress axis suggested by diking versus stress axes suggested by synplutonic folds in the wall rock and flattening fabrics in the pluton..................................................96 Figure 33: Mohr circle construction and block diagram showing how sheets may be emplaced perpendicular to o i................ 98 ix Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES Table 1: Layer-forming processes............................................................. 47-50 Table2: Layer-types....................................................................................... 51 Table3: Magmaticflowtypes........................................................................... 65 x Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF PLATES Platel: Map showing layering orientations in the Main Donegal Granite. Plate 2: Map showing foliation orientations in the Main Donegal Granite. Plate 3: Compilation map showing foliation orientations throughout the Main Donegal Granite. Plate 4: Map showing lineation orientations in the Main Donegal Granite. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT The Main Donegal Granite is a 400 Ma pluton located in the NW Caledonides of Ireland. The 40 x 10 km pluton is elongate NE-SW and is spectacularly layered on several scales. On the largest scale it is an amalgamation of sheets, tens of kilometers long and kilometers wide, that are intrusive in nature. Within these sheets are the "regular" layers-alternating, millimeter to meter wide, potassium feldspar-rich granites and potassium feldspar-absent trondjhemites. Field relationships suggest that these too, are intrusive in nature, with perhaps some mingling taking place. Both the sheets and the regular layers strike NE-SW and are steeply to vertically dipping. These features are overprinted by a penetrative foliation that is magmatic at the center of the pluton, but becomes progressively more solid state towards the margins of the pluton. This magmatic foliation is generally parallel to the sheets and layers, but occasionally cross-cuts their boundaries. Magmatic layers and foliations have been used to infer processes during pluton ascent, emplacement and post-magma chamber formation~in particular, magmatic flow taking place during these events. The assumption is that magmatic flow was parallel to foliations and to layer boundaries. Upon review of the processes that form magmatic layers, it is apparent that only the boundaries of layers formed by magmatic injection (intrusive in nature) and certain sedimentary cumulates can be used to infer magmatic flow planes at any time. Thus the margins of the sheets and regular layers in the Main Donegal Granite can be used to infer magmatic flow taking place during ascent and emplacement. On the other hand, magmatic foliations cannot be used to infer flow planes, as they are a result of magmatic strain, not flow--a point illustrated by the occasional cross-cutting nature of the magmatic foliations in the Main Donegal Granite. In the case of the Main Donegal Granite, this strain appears to be due to regional stresses. The parallelism of the magmatic and solid-state fabrics indicate that they may have formed due to the same stresses-stresses that may be related to a sinistral shear zone that borders much of the pluton. This shear zone may have also influenced the emplacement of the sheets and regular layers, although its exact role is unclear. Regardless of what causes deformation, when studying plutons, magmatic foliations, like most layer-types, should not be used to directly infer magmatic planes. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. INTRODUCTION The construction and evolution of a magma chamber is something that cannot be seen in action, but rather must be inferred from studies of structures, petrology and geochemistry in and around exposed piutons. Structures within piutons, such as compositional layering, foliations and lineations may give clues regarding the construction of, and dynamics within, the magma chamber. Therefore, understanding when and how such structures formed and, therefore, the processes they reflect, is of extreme importance in interpretations of magma chamber dynamics. A colleague of mine once said that before he was exposed to our research group he was under the impression that the interior of piutons were structurally uninteresting. In reality, piutons are often internally, highly varied and complex bodies. Almost all piutons exhibit structures in the form of mineral alignments (foliations and lineations) and/or layering. Most piutons contain some form of layering, whether it be ubiquitous throughout the pluton (for example the famous layered Skaergaard, Bushveld and Muskox intrusions) or only rarely found, perhaps at the margins of the body, for example the Chita and the Agua Negre pluton of Argentina (Aaron Yoshinobu, pers.comm., 1995). Igneous layers are sections of rock or magma, generally with parallel sides, that have distinct compositional and textural characteristics that distinguish them from adjacent portions of the pluton. They can be millimeters to 100's of meters thick, repetitive or random, and have sharp or diffuse boundaries. Traits exhibited by layers give clues to the processes by which they form. However, the interpretation of the significance of layers has long been the subject of debate. Initially, 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. layers were believed to have formed from the metasomatic replacement of pre-existing sedimentary rocks, and were thus used to argue for granitization (e.g. Scott, 1862 and 1864 and Blake, 1862, in Pitcher and Berger, 1972). Later, they were interpreted to have formed by crystal settling in a cooling and convecting magma chamber (Wager et af. 1960). It is now realized that magmatic layers can form by a variety of different processes in a magma chamber's history. These processes may operate during ascent, magma chamber construction, and post-cooling alterations, therefore complicating interpretations based on layer characteristics. One of the processes that is often inferred from magmatic layers, as well as magmatic foliations and lineations, is magmatic flow. A common assumption is that the layers, foliations and lineations preserve the orientation of paleo-flow planes and directions in the magma chamber; i.e. flow planes and directions were parallel to these structures (e.g. Balk, 1937; Pitcher and Read, 1959; Shelley, 1985; Guinberteau et ai., 1987; Ernst and Baragar, 1992; Nicolas, 1992; Smith et al., 1993; Philpotts and Asher, 1994;). Assuming that these structures do indeed represent magmatic flow, some investigators have further made inferences regarding the timing of such flow. For example, layers and foliations have been used to interpret early magma chamber processes, such as the ascent of magma (Vigneresse and Bouchez, 1990) and the construction or emplacement of the magma chambers (Mahood, 1986; Abbott, 1989; Vigneresse and Bouchez, 1990). They have been used to infer flow within the chamber after emplacement and during cooling. They have also been used to interpret the shape of intrusions, with the 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. assumption that the fiow was parallel to the walls of the magma chamber (Abbott, 1989; Ernst and Baragar, 1992; Vigneresse and Bouchez, 1990). In addition to flow, layers and foliations have been used to infer orientations of the X axis and XY plane of the strain ellipsoid (Balk, 1937; Blumenfeld and Bouchez, 1988; Benn and Allard, 1989; Nicolas, 1992; Yuan and Paterson, 1993a & b). Implicit in the assumption that magmatic layers, foliations and lineations all represent magmatic flow planes and directions is the idea that the magmatic layers, foliations and lineations are parallel to one another. The parallelism between layers and foliations is often just assumed. On closer examination, however, it is increasingly recognized that this is not the case. Magmatic foliations have been found cross cutting layer boundaries in several pluton studies (e.g. Berger and Pitcher, 1972; Benn and Allard, 1989; Blumenfeld and Bouchez, 1988; Nicholas, 1992) indicating that they cannot both represent the magmatic flow plane. If only one of these structures represents the flow plane (if the flow plane is preserved at all), what then, do the other structures represent? Thus, it is important to re-assess the relationship between magmatic fabrics and layering and to determine what processes they may, or may not, provide information about. Questions that need re examination include "which structures really do represent flow planes?”, and "what can we say about the timing of such flow?". One of the most spectacular layered piutons in the world is the Main Donegal Granite; a 400 Ma pluton that is located in the NW Caledonides of Ireland. The pluton is elongate NE-SW and is pervasively layered on several scales, from kilometer-wide sheets to millimeter-wide 3 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. petrographic layers. Almost all of the sheets and layers are NE-SW striking and steeply dipping. In addition, along its margin the granite is locally interlayered with the country layers, and internally contains abundant "raft trains" that mimic country rock stratigraphy ("raft trains" are apparently unrotated xenoliths, found in regular arrays, that are continuous from the country rocks, into the granite). Overprinting the layers is a penetrative magmatic foliation and fineation that is predominantly parallel to the layers, but occasionally cross-cuts them. The formation of the layers and of the foliation have been controversial for over a hundred years. In the mid-19th century the main hypothesis argued for a metamorphic, as opposed to magmatic, origin for the granites. Scott (1862) was one of the first to argue that the Granite was of metamorphic origin based on the fact that it "was bedded in accordance with the country rocks and was often interstratified with the adjacent sediments”. However, at the same time, Haughten (1862) argued that the Granite, in its central portions was "perhaps of igneous origin,...deriving its....gneissic character from the pressure exercised on it" (Pitcher, pers. comm., 1993). Callaway (1885), observing granitic veins that cut the foliation planes of the country rocks, also argued for an igneous origin, and attributed the foliation to strain imposed on an already consolidated magma, an opinion shared by Hull (1891), who attributed the parallelism of the granite foliations, layering and raft trains to a shearing event (Pitcher and Berger, 1972). in the early 1900's Grenville Cole suggested that the foliations in the granite were due to flow of the magma during emplacement, and "to some extent [to] subsequent pressure" (from Pitcher and Berger, 1972). 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pitcher and Read (1959) also originally believed that the foliation and layering in the Main Donegal Granite were due to magmatic flow. They concluded that the pluton had sheeted its way into the Dalradian metasediments and wedged them apart. One of the most detailed studies, conducted by Berger, suggested that the pluton had a far more complicated history involving intrusion of magma in individual sheets and filter-pressing in order to generate the layers (Berger, 1971; Pitcher and Berger, 1972). Finally in 1982, Hutton suggested that the Main Donegal Pluton was emplaced into an active shear zone. Space was created by distortion of one flank of this transpressional structure: a distortion best measured by comparing magnitudes of strain. This distortion was attributed to differential displacement along this NE-SW trending shear zone. Hutton hypothesized that the NE portion of the shear zone moved further than the SW portion, causing a doorstop-like effect that buckled open the shear zone. According to this model the resulting bowing out of the one wall literally caused magma to be sucked up into the crust (Hutton, 1982). Hutton's study focused mainly on the emplacement of the pluton and did not specifically address the formation of the layers. However, the controversy surrounding the formation of the layers and foliations still remains. Do both represent flow, as originally suggested by Cole (1906), and Pitcher and Read (1959), or other processes? If the foliations and/or the layers do not represent flow, do any other features in the pluton, and if so, what is the timing of that flow (e.g. during ascent, emplacement or subsequent to these events)? 5 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This thesis will summarize and assess different types of layering and foliations, as well as the modes and timing of their formation. It will focus particularly on the relationships between petrologic layers and magmatic foliations; specifically, what processes each may, or may not, reflect in terms of magma ascent, emplacement, and later magma chamber dynamics. These conclusions will be applied to a study of layering and foliations in the Main Donegal Granite, NW Ireland. In order to complete this study, fourteen weeks of field research in NW Donegal were carried out over two summers. During this time, detailed mapping of the southwestern portion of the granite was completed. Orientations and detailed observations of the layers and foliations were taken, with specific emphasis on the composition, texture, kinematics, relative orientations and cross-cutting relationships displayed by these features. Samples were also collected and taken for lab analysis of fabrics. One month of mapping of the Entiat pluton (a sheeted pluton, layered at its margins, located in the Northeastern Cascade mountains of Washington) was also completed in order to supplement the author’s understanding of layering found in piutons. This study brings together the author's observations of the Main Donegal Granite, as well as published and unpublished information from other studies of the Main Donegal Granite. In addition, present thinking regarding layer and foliation formation processes has been compiled with the intention to reassess what the layers and foliations represent. Initial results have been published in three abstracts (Yuan and Paterson, 1993a; Yuan and Paterson, 1993b; Paterson et al., 1994). 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C.HAPTfflLlj.T H E .MA I N D O N E G A L . g. BANI TE General statem ent The Main Donegal Granite of NW Ireland is an ideal subject for the study of layering and foliations in felsic systems, ft is layered on several different scales from kilometers to millimeters, as well as being strongly foliated. The formation of these layers has been controversial for many years. Whether or not they should be used to interpret emplacement, and/or post-magma emplacement dynamics, depends on our interpretation of how they formed. The Main Donegal Granite is an elongate NE-SW, 44 km by 10 km, 407 +23 Ma pluton (O'Connor 1982) located in the NW Caledonides of Ireland (Figure 1). The Granite was emplaced into Dalradian metasediments and slightly older Caledonian intrusions, which, together with the Main Donegal Granite, make up the Donegal Granite Complex. The Complex includes, in order of age, from oldest to youngest: the Thorr, Toories, Ardara, Rosses, Main Donegal Granite, and Trawenagh Bay piutons. The Main Donegal Granite cross-cuts all of the older piutons except the Toories monzodiorite (which is only exposed in isolated island outcrops to the west). The Trawenagh Bay pluton, located at the southwestern end of the Main Donegal Granite, has traditionally been classified as a separate, younger pluton; however, many workers have noticed its close ties to the former body. The main reason for drawing a boundary between the two bodies is the dearth of layers and foliations in the Trawenagh Bay pluton, structures that are so prevalent in 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ly & Trawenagh Bay pluton a Rosses Granite Main Donegal Granite If.'.':;| Ardara Granodiorite - Quartz Monzonite ^*-1 Thorr Granodiorite Trend lines of dominant foliation Figure 1. Location map showing Main Donegal Granite and associated piutons (modified from Pitcher and Read, 1959) 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the Main Donegal Granite. A shear zone exhibiting sinistral displacement is also located between the two bodies and overprints structures in both bodies. However, outside of the shear zone, the contact appears to be gradational. Compositionally, the Trawenagh Bay differs from the Main Donegal Granite only in that it has slightly more muscovite and garnet. Foliations and layering, that are very distinctive in the Main Donegal Granite, tend to fade out along strike into the Trawenagh Bay pluton. Layering within the Trawenagh Bay is rare, but isolated examples have been found (W.S. Pitcher, pers.comm., 1993). The Trawenagh Bay is thus treated as part of the Main Donegal Granite in this study. The Main Donegal Granite is composed of a series of separate, distinct sills, hereafter referred to as "sheets," that are meters to kilometers in thickness and kilometers long. They are elongate NE-SW, parallel to the margins of the pluton and are steeply dipping (Pitcher and Berger 1972). Hutton (1982) hypothesized that these represent the emplacement of the pluton by "lit-par-lit" diking. The sheeted nature of the pluton is best exposed at its margins where NE-SW-strikinging sheets of granite and pegmatite/aplite are separated by septum of country rock (Figure 2). These sheets coalesce inward to form the main body of the piuton as less and less of the country rock is preserved (Figure 3). Internally, the sheet boundaries are best defined by strings of country rock xenoliths (the famous "raft trains" of Pitcher and Berger, 1972; Figure 4). The xenoliths were assimilated to varying degrees, and are lenticular, with their long axes also elongated NE-SW. These raft trains can be traced from the wall rock into and throughout the pluton, 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 2. The marginal zone at Losset shows the sheeted nature of the pluton. More resistant outcrops are granitic and pegmatitic dikes, separated by politic country rock (forming the low-lying areas between outcrops). View is to the NE. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 3. Map of the Lossett area, showing the sheeted nature of the contact, and the sheets that coalesce inwards to form the main body of the pluton (map modified from Pande, 1954). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Y -S rw \x < « - , ^ W ' Lossatt Figure 4. Simplified map showing sheet margins defined by country rock inclusions ("raft trains") in the Main Donegal Granite. Key area locations (referenced later) are also shown. 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. maintaining the same orientations as their wall rock counterparts. Thus, while they are believed to be "free-swimming” (Pitcher and Read, 1959), they appear to still be in their original positions, and to represent the geographic positions of country rock and country rock structure before pluton emplacement. The batholith shows petrographic zoning on the largest scale, with even-grained grey granite being more abundant in the SW portion of the pluton and coarser granite being more abundant in the NW portion, inclusions of the even-grained granite have been found in the coarser granite and so the latter appears to be slightly older (Pitcher and Read, 1959). Published isotopic studies yield ®7Sr/8®Sr ratios of 0.7063. This low value has led investigators to suggest that the parent magma for the Granite was derived from the melting of upper mantle and/or juvenile lower crust, contaminated to varying degrees by mixing with upper crust (Halliday, et al., 1980; O'Connor et al., 1982). On a smaller scale, but present throughout the entire pluton, are several different types of petrographic layering that occur within the sheets. The formation and relative relationships of these layers are the subject of debate. Pitcher and Berger (1972) differentiated several types, many of which are probably subsets of each other. I have attempted to reclassify these based on my own work, incorporating Berger’s layer types into the following compendium. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A) TYRES _QF_U Y E BS 0 Regular, dua/mg By far the most common, but also the most enigmatic of the petrographic layer types, are the “ regular layers' of Pitcher and Berger (1972). These are alternating light and dark bands that, like the sheets, are vertical to steeply dipping and trend NE-SW (Figure 5), parallel to sheet margins and thus pluton sides (Plate 1). In a few areas these layers are folded, usually around inclusions (i.e. rafts) within the pluton, thus departing from their usual trend. In these areas, they are usually cut by later layers that do follow the prevalent NE-SW trend (Figure 6). These regular layers vary from meters to millimeters in thickness (Figures 7 and 8), and are defined as light and dark layers of varying mineralogy and texture. There is no particular pattern of repetition of light and dark layers or layer sets. Both types of bands are generally the same mineralogy (quartz, feldspar and biotite, plus accessory minerals), with the main difference being the occurrence of microcline and the relative abundance of plagioclase and biotite. Mflftfer .rs a u fa U fflrg . The lighter layers vary in composition and texture; however, in general they are microcline-rich (-40%; Figure 9), with lesser plagioclase (-25%), quartz (-25%) biotite (2-10%) and muscovite (<1%). Accessory minerals include apatite, zircon and garnet. Although percentages of each mineral may vary (Berger, 1971), their composition is still constrained enough to classify the light layers as syenogranites. 15 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Equal Area C.l. = 2.0 sigma Figure 5: Equal area stereonet projection showing poles to planes of regular layering in the Main Donegal Granite. Layers strike NE-SW and dip steeply. N = number of data points C.l. = contour interval Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 6. Regular layers that depart from their usual NE-SW trend are cross-cut by a later layer oriented in that direction. Note hammer handle in right hand comer for scale. Located at Doochary. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 7. Millimeter-scale regular layers. Located near Clashy. Figure 8. Meter-scale regular layers with parallel magmatic foliations. The left dark layer is interfingering with a light layer. Magmatic foliations maintain consistent orientations and thus cross-cut the interfingering contacts at high angles at their tips (see dotted line). Note also, the felsic vein cross-cutting regular layers. Magmatic minerals within that vein are aligned perpendicular to the dike walls and parallel to the prevailing foliation (parallel to the pencil). Located near Clashy. 18 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 9. Light, microcline-rich layers. Note the magmatic foliations parallel to the 15cm ruler. 19 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Texturally, the layers range from fine-grained, equigranular granites to extremely coarse-grained, porphyritic granites, with microcline phenocrysts. The coarse-grained samples can be so phenocryst-rich that they become "phenocryst-supported" (i.e. large, euhedral microclines form the dominant matrix of the rock, with the fine-grained minerals occurring in the interstices between these crystals). The entire spectrum between these two end-members (which basically involves an increase in the ratio of phenocrysts vs. equigranular groundmass) is present. Those that are phenocryst-poor and fine-grained tend to have more equal amounts of microcline vs. plagioclase. As the phenocryst concentration increases, the microcline content (logically) also increases. In thin section, microcline crystals range from small (3 mm) and anhedral, to large, euhedral phenocrysts that can occasionally reach more than 3 cm in length. Plagioclase occurs in lesser amounts, both as euhedral inclusions in the microcline, and as larger, more anhedral crystals in the groundmass. The small inclusions of plagioclase are often arranged zonally within the microcline megacrysts. Compositions range from andesine to oligoclase (Ani o - Anso)- Both feldspars are highly sericitized and myrmekite is abundant. Muscovite forms subhedral laths and can be found partially included in plagioclase. Subhedral biotite is often associated with the muscovite. It tends to be highly chloritized and in some cases is replaced by secondary epidote. Biotite can also be found with quartz interstitially between larger crystals. The quartz is sub- to anhedral, but magmatic (showing large crystals and fairly even extinction). 20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Accessory minerals include zircon, apatite, garnet and opaques (the rocks are ilmenite series). The opaques are anhedral and are included in zircon and, less commonly, in orthoclase. The zircons are euhedral, with pieochroic halos and like the euhedral garnets, are included in biotite. Apatite occurs as rounded crystals included in or associated with biotite, and occasionally quartz and feldspar. These observations suggest that the accessory minerals (garnet, apatite, zircon and opaques) were the earliest to crystallize. These were succeeded by muscovite, plagioclase, biotite and microcline; all of which overlapped in time, Finally, quartz and the last of the biotite crystallized. Sericitization of the feldspars, the growth of myrmekite and the replacement of the biotite are late stage processes that probably took place in the subsolidus state. Parker, nag/ar. fm /s Plagioclase is the most abundant mineral in the dark layers (40%) and is notably the dominant and often sole feldspar; the dark layers are almost completely devoid of microcline. Plagioclase compositions remain the same as in the lighter layers, ranging from andesine to oligoclase (Anio - Anso; a fact noticed by both Pitcher and Berger, 1972, and the present author). They are thus classified as trondjhemites (i.e. leucocratic quartz diorites or tonalites that are almost completely devoid of alkali feldspar; Barker and Arth, 1976). Plagioclase is sub- to anhedral, and although this forms the largest crystals in the darker layers, is rarely more than 2 mm in length. Thus, the difference in dominant feldspar mineralogy accounts for the 21 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. difference in texture between the light and dark layers; even the fine grained light layers tend to be coarser than the dark layers. Biotite forms the second most abundant mineral in the dark layers (20-30%). It is more abundant and less chloritized than in light layers, and is subhedral. It can be found included in plagioclase and quartz, as well as growing interstitiaily. Muscovite is rarely found. Quartz (20-30%), as in the light layers, is also anhedral and interstitial. The darker layers contain the same accessory minerals, minus the garnet, but in greater proportions than in the light layers (Berger, 1971). The textural relationships are the same, however. Zircon morphology does not change from light to dark layers (Pitcher and Berger, 1972, p.217), and most accessory minerals are included in biotite. Thus the accessory minerals appear to have been the first to begin crystallizing. Biotite and plagioclase followed, crystallizing at approximately the same time. Finally, quartz and the last of the biotite ended the sequence. Contacts The contacts between the light and dark layers are variable. In the field, they may be gradational, showing a gradual increase or decrease in microcline vs. biotite content (Figure 10). Conversely, they are often abrupt, showing sharp changes in microcline content, and are frequently cross-cutting. Material from the dark layers can be found included in the light layers, and dark layers "frequently wedge out into a host of light granite" (Pitcher and Berger, 1972, p.215). Material from the light layers is also found in the dark layers, although less frequently. Where the regular layers (both light and dark) depart from their usual NE-SW 22 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 10. Regular layers showing diffuse contacts between light and dark bands. 23 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. striking, steeply dipping orientation, they are cross-cut by bands of light (Figure 6) or dark material (Figures 11a and 11b). In these cases, the "cross-bands” (as they were called by Pitcher and Berger, 1972) are usually oriented NE-SW, and may or may not offset their hosts. In some cases offset may be great enough that cross-cut layers cannot be matched on either side of the “ cross-bands” (Pitcher and Berger, 1972, and personal observation; Figure 6). The regular layers that depart from the NE-SW trend often are folded and sheared, while the cross-cutting layers do not record any of this deformation (Figures 12a and 12b). Figures 13a and 13b show a 50 centimeter-wide dark band that has been boudined. Concurrently, this dark band is being streaked out with light band material to form millimeter- to centimeter-thick regular layers that fill the boudined necks. This interfingering and drawing out of thicker light and dark layers to form millimeter-scale regular layers is a frequent occurrence that results in layers with compositions intermediate between the two extremes, with occasional mineral bands of more mafic or more felsic composition. Figure 14 shows a light layer in the process of being disseminated to form magma of mixed composition. In the field, some of the cross-cutting contacts between light and dark layers may appear sharp; the change from light to dark layer is often abrupt both in texture and mineralogy, but on closer inspection the relationship is commonly not as clear. In thin section few, if any, examples of truncated crystals from either light or dark layers could be found (a relationship also noted by Pitcher and Berger, 1972, p. 218). This makes it difficult to determine which composition intruded which. Many of the crystals at the contacts are parallel to contacts, which may 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dginal continuation of margin oflighf layer Figure 11 a. and 11 b. Dark layers cross-cut light layers, suggesting that some dark layers are younger. Hammer is parallel to the magmatic foliation, which cross-cuts all features. 25 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 12a and b. Photo and line drawing of regular layering that has been folded and sheared. Note that this deformed layering has been cut by both a light layer (upper right corner, #4), and a dark layer (center, #5) that have not been affected by the deformation. This suggests that the pluton was deformed during emplacement. Note also that the magmatic foliation cross-cuts all features, but is deflected across the late, dark, cross cutting dike. Located at Glenleheen. Sequence of structure formation shown in line drawing (oldest to youngest) 1 - light, medium-grained, thin granite layers in darker, medium-grained trondjhemitic host 2 - shear 3 - light, coarse-grained dike with quartz-rich margins 4 - light, coarse-grained dike with euhedral microcline crystals 5 - dark, medium/coarse-grained dike with a few entrained microcline crystals 6 - magmatic foliation 26 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o .V/ / 27 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 13a. Boudined trondhjemitic dark layer being drawn out to form mineral-scale layers. Flow of mineral scale layers suggests dextral shear. Compare to figure 13b. Note cross-cutting dark band at the base of photo. Located at Glenleheen, near the old school. Figure 13b. Boudined trondhjemitic dark layer from the same area. Flow of mineral scale layers around boudins suggests sinistral shear. 28 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 14. Regular layers and magmatic foliations (both parallel to the ruler) cross-cut by a late, granitic, light layer that shows the beginning stages of mixing. Felsic minerals are drawn out parallel to the magmatic foliation (which is parallel to the older regular layers, but cross-cuts the late light layer contacts), at the right hand contact (middle of the photo). The left contact is mixed to a greater extent-the regular layers cannot be easily traced at this contact. Located at Doochary. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. account for the lack of observed truncation. However, the majority of contacts examined showed interlocking crystals and consertal textures across the transition. It might be assumed that a layer with a more lobate (convex-out) margin (both at the crystal scale and at the hand sample scale) was intruded by a layer with a more cuspate (concave-in) margin. Thus, theoretically one could infer which layers were older by looking at the shape of the contact. However, in some cases the light layers appear to have the lobate margins and in other cases the dark layers have the lobate margins. On the microscopic scale, the relationship may vary from crystal to crystal, again illustrating the difficulty in determining which layer intruded which. In samples where it appears that the darker layers have the more cuspate margin, contacts often show a concentration of biotite, apatite and zircon. In rare cases, small accumulations of biotite can be found “ dammed” behind microclines that protrude across the contact implying that the magma containing the biotite was flowing past the layer containing the microcline. These relationships are rare, however, and in general the biotite is euhedral and parallel to the margin. Feldspars show interlocking and consertal textures across the contact. Myrmekite is abundant. On the other hand, in samples where the lighter layers have the more cuspate margin there may be a slight increase in plagioclase vs. microcline content of the lighter band, implying some contamination from the darker layer. Again, there is little evidence of truncation, and crystals are consertal across the contact. Occasional biotites from the darker layers can be found included in the outer layers of zoned microclines 30 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. from the lighter layer, suggesting that they were incorporated into the microcline as it grew. These relationships imply that the lighter layer was intruded into the earlier darker layer. Thus, it is possible to find examples where light layers are younger and examples where dark layers are younger. iH Pegmatites The next most prevalent compositional type of layering in the Main Granite is pegmatite. Pegmatite occurs as blotchy, irregular patches that Pitcher and Berger term the “ earliest pegmatite bodies.” It is also found as coherent undeformed dikes, interpreted to be later intrusions (Pitcher and Berger, 1972). Both forms are found throughout the pluton, but are especially abundant in the area between the Main Donegal Granite and the Trawenagh Bay pluton. In outcrop, both types of pegmatite are pink to white and weather in an irregular, rounded fashion. Feldspar (microcline and plagioclase), quartz, muscovite and biotite are the dominant mineralogy. Accessory minerals include garnet, opaques, allanite, zircon and apatite. These accessories can be found included and associated with the muscovite and biotite. The most abundant feldspar is microcline, with lesser plagioclase. The microcline, which often shows graphic texture, forms the largest crystals and has inclusions of euhedral quartz and biotite. Quartz also appears as large euhedral crystals (up to 2 cm across) and as interstitial anhedral crystals that interleave with euhedral books of muscovite and biotite. In the patches of pegmatite there is little overall zonation. Crystals are euhedral, interfinger and can be 10 cm or more across, commonly occurring as 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. radial growths (seen especially well in large, euhedral growths of muscovite). Pegmatite dikes, on the other hand, often show comb textures, with a regular zonation of feldspar in the outermost portions, and muscovite, biotite and, finally, quartz in the very center. Other variations in zoning can be found. Figure 15 shows a pegmatite dike cross-cutting regular layering. The dike is aplitic near one wall and contains millimeter-wide, biotite layers. At the other wall the textures are pegmatitic. This zoning may be due to boiling triggered by decompression, fractional crystallization, or in situ crystallization. It is notable that coarse, light bands from the regular layers can be found grading into pegmatite dikes that have sharp contacts. These in turn grade directly into irregular zones of pegmatite that have gradational contacts with the surrounding granite. These dikes cannot be followed through the irregular zones. A direct correlation between pegmatites and the light, regular layers is thus suggested. iih Assimilated layers Whispy, thin layers that often bear a resemblance to the dark regular layers are formed from the assimilation of country rock. The included rock types (quartzite, metadolerite, Thorr granodiorite, Ardara granodiorite, limestone and pelites) have reacted with the magma to varying degrees, the pelitic lithologies being the most assimilated (Pitcher and Berger, 1972). These pelitic rafts (country rock inclusions) have been drawn out to form schlieren. The schlieren, being richer in biotite than the surrounding magma, appear similar to the dark regular layers. However, their xenolithic origin is clearly shown by the fact that 32 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 15. Pegmatitic dike showing millimeter-scale layers that are a result of fractionation processes. Mafic and felsic minerals have seperated out to form mineral bands and crystals in the felsic band show comb textures. Located at the Quarry at Doochary. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. some of the schlieren still show traces of pelite in their cores (Figures'! 6a and 16b), and can be traced along strike to raft trains of unassimilated pelite. B) MINERAL FABRICS Overprinting almost all of the layers and the sheets that make up the pluton, is a penetrative mineral fabric that is NE-SW trending, steeply dipping and generally parallel to all of the aforementioned structures (the pluton margins, sheets and regular layers; Plates 2, 3 and 4). This fabric is defined by magmatic minerals in the center of the pluton, but becomes increasingly defined by metamorphic minerals towards the outer margins of the body. Throughout most of the pluton the planar component (foliation) of this fabric is stronger than the linear component (S>L); however at some of the localities along the margins of the pluton the linear component becomes more prevalent and an L>S fabric is locally developed. In these areas the lineations trend NE-SW, but have shallow plunges in comparison to inner portions of the pluton, where their plunges tend to be steeper (Figure 17). The foliation remains NE-SW trending and steeply dipping throughout the entire pluton (Figure 18), although at the very NE of the pluton the fabric does become almost horizontal. In the rare cases where the regular layers depart from their normal NE-SW trend (for example, at the Doochary synform, shown schematically in Plate 1), the magmatic foliation can be found cross cutting layers at angles, without any deflection, still maintaining a consistent NE-SW orientation (Figures 8, 11a, 11b, 12,14,19a and 19b). 34 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. remains of pelite stiihrecognizable Figure 16a and b. Schlieren that resemble some of the thinner dark bands. Their xenolithic origin is betrayed by the remnants of pelite raft that have not been assimilated. Coherent, unassimilated pelite rafts can also be found along strike from this outcrop. Located along road from Glenleheen to Doochary. 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C.l. = 2.0 sigma Figure 17: Equal area stereonet projection showing lineations trending NE-SW and mainly plunging shallowly to the SW. N = number of data points C.l = contour interval 36 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C.l. = 2.0 sigma Figure 18: Equal area stereonet projection showing poles to planes of foliations in the Main Donegal Granite. Foliations strike NE-SW and dip steeply. N = number of data points C.l = contour interval 3 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 19a and b. Photo and line drawing of regular layers cross-cut by magmatic foliations (shown by pencil and dotted line). Located at Doochary. 38 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. At the margins of the pluton the fabric continues across the contact with country rocks, with little deflection. In samples where the foliation is poorly defined, the biotite tends to be subhedral and the quartz tends to be blobby and interstitial between the larger feldspar crystals. In the more obviously foliated samples, biotite and quartz are still predominantly magmatic (although quartz does exhibit some subgrains) and form lines of crystals that mantle and undulate around the feldspars. Thus the biotite and quartz most often define a magmatic foliation. Occasionally in the lighter layers, aligned microcline crystals also help define the foliation (Pitcher and Berger, 1972, p.220). As biotite and quartz are not the largest crystals in the rock, it is often easier to identify the foliation in weathered outcrop (where the quartz stands out as the most resistant mineralogy) than under the microscope. It is notable that the orientations of magmatic microstructures indicate pure shear and flattening; there are no consistent tiling directions, pressure shadows or magmatic S-C structures. Thus a preferred direction of shear (sinistral or dextral) is not associated with the magmatic foliation and lineation. While the foliation is magmatic and appears to represent flattening throughout most of the pluton, at the margins it becomes progressively more subsolidus, and more intense, while remaining parallel to the magmatic foliation. Mylonitic fabrics can be found and S-C structures indicate a sinistral sense of shear (Figures 20 and 21). Fabrics become L>S and lineations are gently plunging (Pitcher and Berger, 1972; Hutton, 1982). The subsolidus fabric is mainly defined by secondary muscovite, recrystallizing quartz, biotite and to a lesser extent 39 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 20. S-C fabrics showing sinistral shear in solid-state deformed pegmatite. Pencil is parallel to C. Located at the Poisoned Glen. 40 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. • C-surface + S-surface mean C-surface mean S-surface 212, 82 W 217, 70 W Brockagh mean C-surface mean S-surface 14, 86 215, 89 Cloghleconnell s-surface c-surface mean C-surface = 44, 83 mean S-surface = 62, 89 Straboy s-surface c-surface mean C-surface » 48, 70 mean S-surface » 72, 66 Carbat Gap Figure 21: Stereonet plot showing S-C fabric orientations all indicating sinistral shear. Locations are shown in Figures 4 and 22. 41 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. plagioclase and microcline. The latter two minerals, although recrystallized around their edges, are the most resistant. The other minerals are deformed and aligned in the prevailing foliation direction around these relict phenocrysts. This subsolidus sinistral shear can be traced around almost the entire margin of the Main Donegal Granite proper (Figure 22). Hutton (1982) reports that at the NE tip of the pluton the shear zone splits into two strands around the sides of the pluton. However, while small, vertical shear zones could be found in that area, in general foliations were shallow dipping, showing no indication of the large shear directly at the pluton margin. Further away from the NE tip, sinistral shear could be found at both the northern and southern the pluton margins. Both strands strike northeast-southwest and dip steeply. The southernmost strand can be traced past the southwest tip of the pluton almost to Killybegs. The northernmost strand is less easily traced. In the southwest itdoes not border the Trawenagh Bay portion of the pluton, but cuts across the boundary between that portion and the Main Donegal Granite (Figure 22). From there the strand cannot be traced southwards past the Gweebarra river estuary; there is no evidence of it at the pluton contact in the Loch Doo area. I postulate that the orientation of the shear may swing so that it strikes east-west in this area, and underlies the estuary. A later fault that strikes northeast-southwest and cuts through the middle of the pluton can also be traced along strike to the estuary. This fault may have become located along, and reactivated this portion of the shear zone. However, it is important to note that this zone of shear does not 42 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. completely enclose the body of the Main Donegal Granite (specifically at the western end of the pluton at Loch Doo and around the Trawenagh Bay portion of the pluton; Figure 22). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. V / e u To K9yfc«g* / / Straboy Figure 22. Showing shear indicators, and strain fields suggested by folds Sinistral shear is suggested by S-C fabrics around the margins of the pluton (shown with half arrows). Southwestern S-C data shown in detail on stereonets. Other shear indicators are based on spot observations of S-C fabrics, as well as documentation in Hutton (1982). The shear zones are shown by the lined pattern. Flattening is suggested by synplutonic folds with NE-SW striking axial planes (locations shown with full arrows) and flattening fabrics in the pluton. Fold in the Loch Doo area is that pictured in figure 29. Other fold orientations are documented in Hutton (1977). 44 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2: DISCUSSION The most striking aspect of the Main Donegal granite is its regular layering. These layers are so ubiquitous that they must represent a significant process operating within the pluton. Is that process magmatic flow, and/or is it the same process that formed the prevalent and generally parallel, but sometimes cross-cutting foliation? If not, are the processes related? Can the foliations be used to make inferences regarding the processes that formed the layers or vice versa? Before we can address these issues directly, it is necessary to review processes that form layers, and the key features that would be particularly diagnostic of these processes. Foliation classification will also be discussed and finally, the formation of such structures in relation to magmatic flow. The conclusions from this discussion will have direct implications for the conclusions that can be drawn for the regular layers and foliation of the Main Donegal Granite. A) LAYER FORMING PROCESSES AND IMPLICATIONS Table 1 is a summary of processes that can lead to layer formation in igneous rocks. They are categorized and described from those interpreted to operate when the magma is predominantly well above the solidus (magmatic), to those that operate after much of the magma has crystallized (near solidus). Table 1 also lists recognized layer types, and catalogues the processes that have been suggested for their formation (and, therefore, the processes that they have been inferred to represent). 45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Obviously, different processes may be invoked for the formation of a single type of layer, so some layer-types will be listed under several different processes. The final column in Table 1 lists information that is provided by these layers, regarding the state of the magma chamber, f would ask the reader to pay particular attention to this information, as it wiil be pertinent to the discussion later in the text. For reference, Table 2 lists layer types and their diagnostic features. The processes that are especially applicable to the interpretations of the Main Donegal Granite regular layers, will be discussed in detail in the following section. For a more complete discussion of all other layer-forming processes, and layer-types, please refer to Appendix I. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Table 1. MAGMATIC PROCESSES Layer-forming processes PROCESS...... OPERATION TRAITS EXAMPLES SPECIFIC LAVER TYPES FORMED INFORMATION ------- PROVIDED BY LAYERS Liquid Immlsdbtllty Gradational to MIxIng/Mlngllng May be chemical or mechanical (due !o temperature, viscosity or density diHerences) Mechanical mixing (w iH only form layers i Incomplete) Chemical - sharp contacts, bul layers should be in chemical equilibrium and thus contain the same crystalline phases. Mechanical - sharp, tobate oontads, perhaps chilled margins DVtuse or sharp contacts dependxtg on the extent d mixlngfmingling. May resul In zones ol oompos’ lion intermediate between the two mixing components. Mixing magmas w i not be In chemical equllbrium. Chemical - Irregular pods dgrancphyreinthe Upper Zones B and C, Sltaergaard Intrusion (Roedder, 1969,1979) Mechanical - malic enclaves in Lamarck Granodlorile (Frost and Mahood, 1967) Regular layers, Main Donegal Granite (this thesis) Lamarck granodiorile (Fro6t and Mahood, 1967) Sheets Schlieren Slate at magma (chemical and physicaHemperalure, viscosity and density) in an already constructed magma chamber Magma ln]actlon Intrusion ol multiple magma pulses Sharp, possbtv cross culling contacts II earlier pulses cool before later pulses are Intruded. AKemathrety, actual pulses may not be ctstinguishabte, except as new layer sets in rhythmic/cyclic layering Calamity Peak pluton (Duke ei al., 1967) Olivine zone in Calamity Peak pluton (Husch, 1990) Main Donegal Granie (Pitcher and Read, 1969; Hutton, 1962) S u ile (^ ^ k » el al., 1967) Sheets Cyclic layers Boundaries ol sheets are parallel to magmatic H o w planes. Represent H o w during emplacement; may represent ascent. Cyclic layers represent chemical slate ol magma In an already constructed magma chamber Assimilation Melting and stretching ol wall rock xenolilhs Chemical contamination ol magma, schlieren Schlieren, Main Donegal Granite (Pitcher and Berger, 1972) Schlieren State at magma in relation to wall rock (temperature and vtsoosity) in an already constructed magma chamber; also may represent emplacement processes. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Table 1. Layer-forming processes (continued) Slde-wall In m Iiu crystallization In situ crystallization involving fractional crystallization Crystals grow perpendicular to layer boundaries, may lorm branching crescu mutates Skaergaard intrusion (McBirney and Noyes, 1979) Pegmatites Comb layers Crescumulales Lelsegang layers State of magma (oom positional, temperature) in an already constructed magma chamber Crystal sattllng precipitation flotation deposition Fractional crystallization and gravity processes Precipitation - heavier crystals sink Flotation - lighter crystals float Deposition - crystals deposited by density currents PrectpHalion • sequence corresponds lo 1 factional crystallization sequence of parent magma. May form Christmas tree structures' N a tio n ^ptagiociase accumulations at the lop of magma chamber Deposition - sedimentary structures, wash-outs, cross beds Deposition - erodonal channels In Nunarssuit complex (Parsons and Butterfield, 1981); layers In Hall Cove complex, Duke Island intrusion (Irvine, 1987) Precipitation cumulates Flotation cumulates Deposition cumulates (erosional channels, cross layers) Stale of magma in an already constructed magma chamber; Preckritalkm and Flotation cumulates provide information about the affect of gravity vs, viscosity vs. crystallization front migration speed. Deposition cumulates provide information about magmatic flow. Flow planes and directions are Interpreted as for sedimentary structures. Nucleatlon diffusion Nudeatlon rates that outstrip diffusion rales lead to fluids depleted ri fast-nudeallng minerals Repealed sets of layers Bushveld Complex (Wager, 1959) Skaergaard intrusion (McBirney and Noyes, 1979) Rhythmic layers (e.g. Lelsegang layers, Cyclic layers) Slate o( magma in an already constructed magma chamber- speolically nucioalion rales vs. diffusion rales. Compositional convection/ Boundary layer fractionation Fractionation or assimilation produces more or less dense liquids that rise or sink away from the crystal pile/wal Adcumulate pies (high in refractory components) Thin layers of malic, dense layers at base, less dense, leisic layers at lop of magma chamber Unit 10 of the Rhum Layered Intrusion (Kerr and Tat, 1984 & 1967) Palisade Intrusive Suite (Sawka, Chappell and Kistler, 1989) Pegmatites CoHolorm or corrugated bandng Slate ol magma in an already constructed magma chamber-- specriically, density differences, driven by crystallization. 00 Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Table 1. Layer-forming processes (continued) NEAR SOLIDUS PROCESSES (cont.) Doubla diffusive convection Opposing elf sets of thermal and compositional gradients lead lo a stratified chamber. Convection occurs In layers, diffusion across layer boundaries Layers of varying thickness (but thick enough lor convection fo occur). Perhaps depositions! cumulate layering if convection is strong enough. Skaergaard intrusion (McBirney and Noyes, 1979) Honningsvag Intrusive Suite (Robins et al., 1987) State of magma in an already constructed magma chamber-- specificaty, thermal vs. compositional differences. Filter pressing Uaulds migrate to zones of lower pressure during deformation (shearing and compaction) of crystal mush. Can only operate in bodes > 3 0 10112, with crystal contents <55%. Pairs of layers-one representing late liquids, one representing crystal much. Gradual composition change at contacts, zones of migration rich In late liquids (i.e. fluid relocation structures). Synplulonlc deformation (e.g. folding, shear, evidence of compaction) Regular layers of Ihe Mam Donegal Granite (Berger, 1971) Kinsman and Bethlehem Suites (Clark and Lyons, 1966) Skagit Gneiss (Hibbard, Magma chamber already constructed. Implies deformation of the magma. Provides limitations on crystal contents during layer formation. Flow sorting Crystals segregated by size and concentration due to the Wal, Bagnoid and Magnus efiects (caused by velocity gradients across Ihe H o w ) In dikes should show symmetrical distributions of crystals in cross section. Phenocrysts concentrated in center, liner crystals at margins Ewarara pluton (Goode, 1977); PtoumanCh Subalkaline Granite (Barriere, 1981) Schlieren Magma chamber already constructed. Imptes that magma was flowing, but directions and direct flow planes cannot necessarily be determined. Sintering (Ostwald Ripening) Agin^ooarsening process after accumulation of crystals. Larger crystals grow at the expense of smaller crystals in order to reduce total free surface energy of system tocrystafaizia. Jborelaa^ gradient of grain-size across the system. Textures are coarse and exhbil secondary structures (e.g. 'honeycomb' patterns) Banded Series, Stillwater Convex (Boudreau, Rhum complex (Hunter, 1987) Experiments (Nabatek, 1978; Jurewicz and Bruse, 1985; Park and Hansen, 1990; Means, 1994) Adcumulale layers Chemical stale of magma, controlled by fractionation, in an already constructed magma chamber. C O Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. Table 1. SUBSOLIDUS PROCESSES Layer-forming processes (continued) Metasomatism Alteration and/or replacement of pre existing minerals by the passage d late, melasomallclluids Met amorphic textures visfcle, perhaps deposition ol hydrothermal minerals Regular layers, Main Donegal Granite (Pitcher and Berger, 1972) Shap granite (La Bas, 1962) Rhum (Butcher et al. 1965) Pegmatites Movement ol lluids. May provide form ation aboU emplacement. Fluid-enhanced diffusion Fluids along grain boundaries enhance jpaln-scale processes sl i^ng^and^^Tor transport of components in lluids (by diffusion and llow) Quartz-rich layers marking channelways ol lltids, quartz- poor layers marking areas thatthidsdeploled. Textural coarsening In the Rhum complex (Hunter, 1967) Hermitage Massit (Gapats and Babarin, 1986) Metamorphlc banding Myloniles Movement ol lluids. Stress-Induced diffusion Due to applied stress, chemical components diltuse from low pressure to high pressure Mylonile banding removal ol quartz and concentration ol phytlosilicales in high pressure zones Phyllosilicate-rich C- planes in the Corcoesto granite (Burg and Ponce De Leon, 1985) Metamorphlc bonding Mytonitas Strain orientations. Transposition Physical rotation of pre existing layers and minerals New layer long axes are parallel to X- axis of strain ellipse Melamorphic banding Strain orientations cn o MAGMATIC LAYER TYPES D IA G N O STIC FEATURES Cum ulus Crystal Settling precipitation cumulates Piles of crystals perhaps inversely reflecting crystallization sequence of magma (i.e. earliest crystallizing minerals at bottom of pile flotation cumulates Usually consists of plagiodase accumulations at the top of the magma chamber deposition cumulates Sedimentary flow structures Crescum ulates (H arlstle cum ulates) Upward branching bladed mineral growths Adeumulate Cumulates concentrated in refractory components (pore material is absent) Orthocumulate Cumulates with refractory components and pore material present Mesocumulate Cumulates with refractory components and minor amounts of pore material present Com b Outward bladed mineral growths, oriented perpendicular to layer planes. Colloform or corrugated banding Regular bulges and troughs in layers R hythm ic Recurrence of distinctive layers, or sequences Cyclic Recurrence of distinctive layers or sequences that are indicative of the fractional crystaization path of the parent magma Leisegang layers Thin, monomineraiic, repeated sets of layers C ryptic (Phase) Gradational variation in soiid-solution mineral composition S c h lie ren Thin, wispy, discontinuous, malic layers P eg m atitic Coherent dikes or diffuse layers or blotches, with very coarse grained, interlocking, often euhedtal crystals S h e e t Laterally continuous, tabular layers, sharp contacts. Cross- cutting contacts implies a diking mechanism for formation M yio n itic Fine grained, streaked, aphanitic fabrics Table 2. Layer-types. 51 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Laver-Forminq Processes Pertinent to the Main Donegal Granite Processes that have been, or will be proposed for the formation of the regular layers in the Main Donegal Granite are reviewed below. They will be discussed from the earliest suggestions (metasomatism; Scott, 1862) to the most recent suggestions (magmatic injection; present author). i) SUBSOLIDUS PROCESSES The following layer-forming mechanisms are subsolidus processes; i.e. they operate in a crystal dominated, solid state system, where the magma is entirely solidified. Metasomatism Metasomatic layers form by the alteration or replacement of pre existing minerals during the passage of metasomatic fluids that are rich in dissolved constituents. They are recognized by the presence of metamorphic textures and compositions. This process is more common when fluids can flow with relative ease through the rock. Therefore, metasomatic replacement occurs most commonly at mid-ocean ridges, where the extruding magma heats the surrounding sea water which convects rapidly through the ridge crust. Metasomatism is also prevalent around shallow level plutons. Fluids in the country rock are heated by the magma, which may also contribute primordial fluids to the country rock liquids. Rapid convection is often aided by fractures formed in the aureole by the emplacement of the magma. 52 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. As fluid flow and diffusion in fluids are the processes by which constituents are transported during metasomatism, it is possible to move material in greater quantities, and much more rapidly, than by processes that involve solid state diffusion, which is a much slower process (Williams, 1972). Metasomatic replacement of pre-existing sedimentary layers (granitization) was used to explain the existence of igneous rocks (Scott, 1862 &1864; Blake, 1862, in Pitcher and Berger, 1972) before the concept of molten magma was accepted. It has also often been used to account for microcline phenocrysts occurring in layers or enclaves within granites (e.g. in the Main Donegal Granite (Berger, 1971; Pitcher and Berger, 1972; in the Shap Granite Complex (Le Bas, 1982); and in the Rhum complex (Butcher, 1985)). The regular layers of the Main Donegal Granite were considered prime examples of metasomatic layering because the granite is interbedded with country rock, contains abundant country rock xenoliths and is rich in microcline phenocrysts. However, the presence of clearly intrusive layers led later investigators to discount this theory (e.g. Callaway, 1885). ii) NEAR SOLIDUS PROCESSES The following layer-forming processes involve fractional crystallization, which is one of the most frequently called upon mechanisms for generating heterogeneities within magmas. It involves the crystallization of minerals from an originally homogenous melt and removal or isolation of those crystals from the melt. This results in two components: crystals which may accumulate by various processes to Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. form layers, and a liquid which is significantly different in composition from its parent magma. Filter pressing Heterogeneities that may result in layer formation can be generated by separating fluids from their fractionated crystal mushes. If the crystal mush is compacted, by shearing of the magma or by the compressive weight of the overlying crystals, liquids will migrate to zones of lower pressure (Best, 1982). Marsh (1981) noted that filter pressing is unlikely to take place in a magma if crystal contents are greater than 55%. As such bodies already exhibit crystal closest-packing, they dilate upon deformation (crystals move out of closest-packing in order to accommodate shear), and therefore have a tendency to pull liquid in, rather than push liquid out. Melt will only be extracted if the body is sheared enough that the crystals themselves start to deform. In addition, the rate of deformation must be greater than the rate of melt solidification; otherwise, the melt will freeze before it can be removed from the crystal mush. This will only occur in magma bodies >30km2. This places a lower limit on the size of bodies in which significant filter pressing can take place (Marsh, 1981). Layers formed by filter pressing should consist of “pairs” of layers that are geochemically traceable as a fractionate couple. One layer represents the incompatible elements and late crystallizing minerals, while the other represents the counterpart-the compatible elements and early crystallizing minerals. Layers should show a gradual compositional transition at their contacts, or fluid relocation structures. Berger (1971) 54 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. used filter pressing to argue for the formation of his “regular layers" in the Main Donegal Granite. Filter pressing has also been called upon to account for layers in the Kinsman and Bethlehem Suites (Clark and Lyons, 1986) and the Skagit Gneiss (Hibbard, 1987). iii) MAGMATIC LAYER-FORMING PROCESSES Magmatic layer-forming processes operate in a melt dominated system, where the magma is below the liquidus, but well above the solidus. Assimilation The melting of wall rock or xenoliths will produce heterogeneities that may form layers. If convection is taking place then these heterogeneities may be drawn out by flowing magma to form schlieren-type layers (e.g. in the Main Donegal Granite, Pitcher and Berger, 1972, p.215; Figures 16a and b, this text). Alternatively, assimilation at the walls and roof of a pluton may lead to the formation of layers with a distinctly contaminated composition immediately next to the boundary. Liquid immiscibility (mechanical) When two liquids are incapable of mixing they are immiscible. This immiscibility may be due to chemical or mechanical differences. Chemical immiscibility only occurs three compositional magma systems (mafic silicate magmas, C02-rich magmas and iron-rich tholeiitic magmas), none of which applies to the Main Donegal Granite. 55 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Therefore, only mechanical immiscibilty will be discussed here. For a more detailed review of chemical immiscibility please see Appendix I. Mechanical immiscibility may occur due to temperature, viscosity or density contrasts between two different magmas. At temperatures well above their liquidi, mafic and silicic magmas are mechanically miscible (Kouchi and Sunagawa, 1985). However, magmas that are below their liquidi are much more likely to be mechanically immiscible due to differences in viscosity. For example, sharp boundaries between mafic enclaves and felsic host rocks in the Lamarck Granodiorite are interpreted by Frost and Mahood (1987) to represent mechanical immiscibility between the two magmas. As mafic magmas have higher solidi than felsic magmas, at any given temperature, a mafic magma will be nearer its solidus than a felsic magma at the same temperature. Thus, a mafic enclave will contain more crystals and thus have a higher viscosity than the surrounding felsic magma. Because of this it will resist break-up and mixing due to convection, and will essentially be immiscible. If a magma is injected as a dike into a magma of different composition, it may maintain its coherence due to mechanical immiscibility and thus form layers. Magma mixing/mingling When two magmas in a chamber do not completely mix before crystallizing, layers that are formed may appear swirled and streaky. Complete mixing (or hybridization) will only occur if the magmas in question are similar in physical properties (e.g. density, viscosity) at the temperature at which mixing is taking place. Otherwise, only magma 56 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mingling will take place. This criterion will only be fulfilled if the magmas are compositionally similar (within 10% Si02 of each other; Frost and Mahood, 1987), for reasons explained above. Thus complete mixing and complete mechanical immiscibility can be considered end members of a spectrum of magma mixing. Silicic magmas are more likely than mafic magmas to approach the same viscosity of their felsic hosts, and are therefore unstable and likely to become mixed (Williams and Tobisch, 1994). Enclaves will only be deformed by magma flow if they have a similar, or lower viscosity than the magma. Therefore if an enclave has a viscosity close to, but still different from the surrounding, flowing magma, then it may be drawn out to form layers. For most microgranitic enclaves in felsic hosts, this will occur at temperatures between 950°C and 1050°C (Williams and Tobisch, 1994). Boundaries between incompletely mixed (i.e. "mingled") magmas may be diffuse (much like incompletely mixed paint), or may be coherent, solid bodies with sharp, distinctive contacts, depending on the extent of mixing. Magma injection The injection of magma pulses is a method of introducing different magmas into an area, and thus can operate in conjunction with magma mixing/immiscibilty. If the old and new magmas do not mix completely then layers may form. In extreme cases where almost no mixing takes place, layers are essentially self-contained dikes. They exhibit sharp (and often cross-cutting) contacts and perhaps an age gradation between dikes. Textures in the center of the dike may vary, depending 57 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. on whether flow sorting, differential injection, or in situ crystallization occurs. Magmatic injection has been used to account for the construction of whole plutons (e.g. the Main Donegal Granite; Pitcher and Read, 1959; the Calamity Peak pluton; Duke et al., 1988) as well as zones within plutons. Husch (1990) argued for injection to form the olivine zone in the Palisades Sill, a zone traditionally interpreted to have formed by gravity settling. The formation of layers in the Honningsvag Intrusive Suite has also been attributed to this process (Robins et al., 1987). If magma pulses are intruded into magma of similar mechanical and chemical properties, then mixing of the two is more likely to occur and contacts between the layers will be more diffuse. Addition of magma pulses has been used to account for the formation of new layer sets in rhythmic layering (see Table 2 for definitions). Differentiated injection is a form of magma pulsing that involves the sequential intrusion of magma pulses into a dike. Each of these pulses is tapped from different levels of a deeper magma chamber and therefore may have different crystal contents. The resulting layering is similar to that produced by the mechanical effects of flow in a dike, and so has been suggested as an alternative for the formation of some apparently "flow-sorted" dikes (McBirney, 1993). The two processes can be distinguished from each other by geochemistry; the groundmass of both the central, phenocryst- rich zone and the outer, phenocryst-poor zone should be the same for flow sorting, and should be different for differential injection (McBirney, 1993). 58 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. B) FOLIATIONS & LINEATIONS 0 Classification of Fabrics Foliation and mineral lineation can be defined, respectively, as repeated or penetrative planar and linear structures in a rock mass due to aligned, planar or elongate minerals (Kearey, 1993). In further discussions regarding foliations, it will be assumed that the conclusions apply to lineations as well, even if lineations are not mentioned. Nomenclature used in previous papers regarding such fabrics (Paterson et al., 1989; Hutton, 1988) refer to the state of the rock during which the foliation formed, particularly the crystal versus liquid ratio. The percentage of crystals versus liquid controls viscosity, which in turn affects the manner in which the magma deforms. This will be reflected in the resulting microstructures. Categories include "magmatic," "sub- magmatic," and "solid state" (Paterson et al. 1989). Hutton (1988) referred to these states as "pre-full crystallization" (for magmatic and submagmatic) and "post-full crystallization" (for solid state). These categories are defined by the dominant mechanism by which deformation occurs. When studying igneous structures, one cannot ignore the importance of petrological processes (such as fractionation) in a magma, as they will invariably affect, if not determine, the rheology and thus the magma's behavior. Therefore, petrographic nomenclature will be adopted in place of the more structurally descriptive nomenclature previously used to specify the state of the magma. “Magmatic,” “sub magmatic," and “ solid state" are now referred to as “magmatic,” “near solidus,” and “subsolidus.” Criteria used to delineate in which of these 59 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. states the fabrics formed are summarized in Paterson et al. (1989). It should be noted that from the magmatic to subsolidus states there is a continuous transition, with more solid-state processes taking place as the magma cools to nearer its solidus, or magmatic processes predominating as a rock heats to nearer its melting point. Therefore, just as evidence of magmatic processes can be seen in bodies of rock where there is less than 10% melt present (Nicolas, 1993), evidence of what are traditionally viewed as metamorphic processes can be seen in magmas (Means 1994). Magmatic structures are those that form by processes which operate well above the solidus. Under these conditions, crystals are present, but in a small enough proportion that there is little crystal-crystal interference. Flow of melt and rotation of crystals is the main process by which deformation is accommodated. The percentage of melt versus crystals at which magmatic structures will form, varies depending on the movement of the magma. In static magmas with more than 50% melt, there are insufficient crystals to form a lattice network (Miller et al., 1988). Van der Molen and Patterson (1979) suggest that 35% melt is the minimum melt fraction necessary in order for crystals to flow without strong interactions. However, in melts that are flowing, magmatic structures may still form with melt contents as low as 10% (Nicolas, 1993). In this case, strong alignment of crystals in the shear direction would favor the concentration of melt into high strain zones, leaving the crystals little deformed. Further deformation would then be partitioned into the melt zones. Magmatic microstructures are typified by aligned, euhedral, internally undeformed, igneous crystals surrounded by equant, non-deformed quartz 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Paterson et al. 1989) (quartz deforms easily at all temperatures, and so is usually the first mineral to betray any signs of subsolidus deformation). Near solidus (or submagmatic) structures form by processes that operate close to, but above the solidus. Crystals are interlocking and are the most abundant phase, but with small amounts of interstitial melt still present. Magma behavior changes very rapidly as crystal percentages increase (from 60% crystals to 80% crystals, viscosity increases tenfold due to interacting grains; Wickham, 1987, Cruden, 1990); therefore structures formed in this state are often distinctive from magmatic textures and so are categorized separately. Near solidus structures tend to show the interaction of igneous crystals. The tiling of igneous minerals is common. Other telltale features of near solidus deformation are late magmatic minerals and mineral overgrowths in pressure shadows (Paterson et al. 1989). When there are enough crystals present to form an interlocking network, it is possible, even if there is melt present, for "metamorphic" processes to take place. For example, grain-boundary and phase- boundary migration, both processes traditionally thought to operate in the solid-state, have been recorded operating when there is still melt present (Means, 1994). Subsolidus (previously referred to as solid state, or metamorphic) textures form when the magma has cooled below the solidus and there is no liquid left. Subsolidus deformation involves intracrystalline deformation (e.g. subgraining) and recovery (e.g. creep, recrystallization), fracturing and boudinage of minerals, and the formation of bands of recrystallized grains. Not only does the behavior of a mineral 61 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. vary according to the temperature of deformation, but different minerals will deform differently when deformed at the same temperature. Thus the general temperature at which deformation takes place can be inferred by looking at mineral assemblages and microstructures. Finally, temperature of deformation can also be inferred by assessing whether the strain associated with the foliation formed is homogeneous or heterogeneous (Gapais, 1989). Aligned high temperature metamorphic minerals indicate subsolidus, high temperature deformation. Such deformation is also typified by extensive grain-size reduction and/or recrystallization, and elongation of minerals to form continuous bands. In particular, the recrystallization of high temperature minerals such as olivine (whose recrystallization indicates temperatures >1000°C; Raleigh, 1968), pyroxene, hornblende (whose recrystallization indicates temperatures >800°C; Biermann and Van Boermund, 1983), and feldspars (whose recrystallization indicates temperatures >450°C; Tullis, 1983) all indicate high temperature deformation. Fracturing and boudinage of these high temperature minerals is also common, with lower temperature minerals such as quartz and mica infilling the fractures and boudin necks (Simpson, 1985; Vernon et al. 1983, in Paterson et al. 1989). Strain associated with the foliation tends to be homogeneous. Ductile deformation is predominant and is often seen in the formation of mylonitic zones (Choukroune and Gapais 1983, Vernon et al. 1983 in Paterson et al. 1989). The transition from high to low temperature subsolidus deformation, although gradual, is arbitrarily placed at the amphibolite/greenschist boundary (450°C). Aligned low temperature metamorphic minerals 62 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. indicate subsolidus, low temperature deformation. Feldspars and other high temperature minerals behave brittlely, while low temperature minerals (especially quartz) undergo extensive recrystallization, often forming continuous "ribbons." The strain associated with the foliation tends to be heterogeneous, with a predominance of brittle deformation, often seen in the fracturing of individual minerals and the formation of cataclastic zones. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. C) FLOW GEOMETRIES Flow in a melt or solid rock can be described using a displacement field consisting of vectors that show particle motions. The flow directions are represented by the orientation of the vectors and the flow plane is the plane parallel to displacement vectors. Strain, the change in the shape of rock bodies often resulting in penetrative fabrics, will result from flow. It is important to note that flow directions (i.e. vector orientations) are not always parallel to the X-axis (X>Y>Z) of the strain ellipsoid. It is necessary, when attempting to determine paleo flow planes and lines in magma, to decide 1) whether flow planes and lines were parallel to the structures being used to infer flow; 2) whether flow planes and lines were parallel to the XY-plane and X-axis of the strain ellipsoid, respectively; 3) what processes are responsible for the magma flow (i.e. what processes that flow represents). The following discussion of magmatic flow and strain will show that the X axis and the XY plane of the strain ellipsoid resulting from flow may not be parallel to flow planes and directions. In addition it will illustrate that foliations can form at angles to magma flow planes and directions. Flow can be divided into two different categories: uniform (or homogeneous) or non-uniform (heterogeneous) flow. Markers (e.g. crystals) within these flows may behave passively or nonpassively (Table 3). In uniform flow velocity and displacement vectors have the same length and direction. If particles in such a flow are passive (i.e. the markers behave exactly as the melt does, so there is no viscosity or density contrast), then all the points in the particle will undergo the same 64 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. MAGMATIC FLOW UNIFORM FLOW vs. NON-UNIFORM FLOW PASSIVE MARKERS NON-PASSIVE MARKERS PASSIVE M ARKERS NON-PASSIVE MARKERS NO STRAIN A C C ELER A TIO N ALIGNMENT PARALLEL TO FLOW DIRECTION OR PLANE W NO STRUCTURE D EC ELER ATIO N LINEATION OR FOLIATION VELO C ITY G RADIENT SEE FIG URE 2 TU M B LE C O N TIN U O U SLY, S TA TIS TIC A LLY FORM A PREFERRED O R IE N TA TIO N PARALLEL TO T H E STRAIN ELLIPSO ID = FLOW PLANE AT H IG H STRAINS Table 3. Fabrics that form from different types of flow and marker behaviour cn cn amount of displacement as the melt. Therefore, strain is zero and the markers will maintain whatever original orientation they had before flow took place. If particles are non-passive (perhaps having a different viscosity or density than the surrounding melt), they tend to align with their long axes parallel to the flow direction in the flow plane (Balk, 1937). Thus, the resulting foliation and lineation reflect magmatic flow planes and lines. However, the mere presence of non-passive markers in a uniformly flowing material will lead to eddies and instabilities in the flow immediately surrounding the crystal. As a result, uniform flow at the mineral scale rarely occurs. Furthermore, widespread uniform flow is very rarely achieved, because magmas are rarely homogeneous, and vary in physical properties and flow dynamics. Non-uniform flow is much more likely to occur in melts. For simplicity, the following discussion will assume that markers in a heterogeneously flowing magma are passive. Different types of non- uniform flow have been investigated by Schmeling et al. (1988). Convection caused by the drag of wall rocks on the surface of a rising diapir of magma was modeled. The resulting types of flow can be divided up into three different categories (Figure 23) each of which resulted in a particular strain path (Mackin, 1947). i) Acceleration flow Acceleration flow applies to any flow where magma is channeled from a wide region into a narrow channel. For example, such flow would take place at the bottom of convection cells in a diapir as magma collects and flows upwards. In this case the X-axis and the XY-plane of the finite strain ellipsoid align parallel to the flow direction and flow plane 66 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CONVECTION a v e l o c it y g r a d ie n t f l o w ACCELERATION DECELERATION c d Figure 23. Showing that the long axis of the strain ellipse may not be oriented parallel to the flow direction. Therefore, if foliations and lineations represent strain, then they may not be oriented parallel to the direction of flow. Arrows in diagrams represent flow directions, while ellipses represent strain (foliations and lineations will be oriented parallel to the long axis of the strain ellipse). Representative crystals are shown rotating into parallelism with the long axis of the strain ellipse. a) Flow and resulting strain ellipses for fluid undergoing convective overturn in a risisng sphere (from Schmeling et al., 1988) b) The effect of wall rock drag on the velocity gradient of flow (modified from Mackin, 1947) c) Flow directions and resulting strain ellipses for fluid accelerating through a narrow aperture (modified from Mackin, 1947) d) Flow directions and resulting strain ellipses for fluid decelerating through a widening aperture (modified from Mackin, 1947) 6 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. respectively. Crystal long axes will tend to align parallel to the X-axis of the finite strain ellipsoid, and as this is parallel to the flow direction, the crystals can also be assumed to represent the flow planes and directions. ii) Deceleration flow Deceleration flow occurs where magma is flowing from a small conduit into a much larger area, and therefore is slowing down and spreading out. Forexample, such flow would take place at the top of a convecting diapir. In this case the X-axis and XY plane of the strain ellipsoid form approximately perpendicular to the flow direction and plane. Hi) Velocity gradient flow Velocity gradient flow will occur whenever there is drag along a surface, (i.e. if there is a viscosity contrast between two bodies, this type of flow will occur at their contact). This type of flow occurs along the sides of a convecting diapir and at any natural contact. It is especially applicable in relation to dikes or sheets, as well as layers, since in all such cases there is a compositional and viscosity difference, and therefore velocity gradients develop at their margins. Crystals will align parallel to the X-axis of the strain ellipsoid which, initially will be at an angle to flow directions, but with greater strain, will rotate into subparallelism with the flow direction. Therefore, in magmas that have undergone only a small amount of strain, the foliations and lineations may be at angles to the flow directions. In magmas that have undergone a great deal of strain, the flow directions, strain ellipsoid X-axes and XY- planes, and lineations and foliations will approach parallelism. 68 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In the case of non-uniform flow in which the markers are non passive, the position that a particle ends up in at any given time is a function of shear strain and the aspect ratio and initial orientation of the crystal (Hanmer and Passchier, 1991; Nicolas, 1992). At larger shear strains (y>5) the larger particles are more likely to be parallel with the flow direction (Fernandez and Laporte, 1991). The rotation of particles with aspects ratios >10 is so close to that of passive markers that they will rarely rotate past the shear direction. Preferred mineral orientations (lineations and foliations) statistically, most frequently form parallel to the X-axis and XY plane of the finite strain ellipsoid. This is initially at 45 degrees to the flow (or shear) direction, but rotates with continued simple shear towards the flow direction (Nicolas, 1992). Thus lineations and foliations will initially be at angles to flow directions, but will rotate over time into parallelism. Note that the lineations and foliations are marking the X axis and XY plane of the strain ellipsoid, which only at high strain, will be parallel to the flow directions. 6 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. P) SUMMARY OF MAGMATIC LAYERING AND FOLIATIONS What aspects of the pluton’s history do magmatic layers and magmatic foliations represent? And for that matter, do layers and foliations represent the same processes? Both have been used to interpret 1) magmatic flow, which may take place during ascent, emplacement, or after the magma chamber is constructed; 2) magmatic strain; and 3) the shape of the magma chamber. As seen earlier, layers can form by several different processes, which may operate at different times in the magma chamber history, and not all of which involve flow. Upon review of the layer-forming-processes table (Table 1), particularly the final column, it becomes apparent that magmatic injection and sedimentary deposition are the only processes capable of forming layers whose boundaries directly represent flow planes. (Note, for example, that smaller layers formed within injected dikes or pulses may form by means other than flow. Therefore it is only the boundaries of the layers themselves that represent flow planes). Thus, sheets and depositional cumulates are the only layer-types that may be used to directly infer flow planes. Other layers may form by processes involving flow, but flow planes and geometries cannot be directly inferred from their geometry (for example, magma mixing/mingling involves the flow of magma, but flow planes within that magma may not be parallel to resulting layer planes). This further limits the use of layers to interpret the shape of the magma chamber. Magmatic layers will not represent the shape of the magma chamber unless they represent flow that is parallel to the walls of the chamber, or 70 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. other processes that involve even, inward growth from the walls. Both of these conditions are almost impossible to determine unless the walls themselves are visible. However, if layers are determined to directly represent flow, it then becomes important to determine when that flow operated. This could be any time before the magma chamber cooled completely; however, the likelihood of flow-feature preservation increases the later that flow operates. Evidence for flow that operates during ascent is the most likely to be obliterated by later processes operating during and after emplacement of the magma (Clemens and Mawer, 1992). The difficulty in preserving such evidence is compounded by the difficulty in finding areas where such evidence would be located. Layers that represent magma ascent will often be located well below the base of the magma chamber. Therefore, one would have to find exposures of the pluton that showed the transition from ascent features to the magma chamber itself-i.e. the root, or floor of the pluton. It is equally hard to find layers that could represent magma chamber formation, as these, too, are not commonly preserved. In fact, very little of emplacement processes are preserved in magma chamber patterns due to the long-lived nature of such bodies. As with ascent, the dynamic nature of all magma chambers increases the likelihood that any emplacement-related flow structures will be obliterated by later activity in the magma. If emplacement processes are preserved, it would imply that the magma freezes as soon as it intrudes the country rock. Furthermore, the magma must experience very little penetrative deformation 71 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. afterwards, in the solid state, so as to preserve the evidence of formation. Layers formed by magma injection (i.e. individual sheets) are the only types that could represent flow during magma chamber formation. All other types of layering require the magma chamber to have formed well before the layers. Thus all other layer types only give clues to the nature and dynamics of the magma chamber once constructed. Most layer-forming processes (including flow) operate at this stage in the magma chamber’s history. However, as might be expected (and as stated before), due to the dynamic nature of the magma chamber, it is only the very last of these processes that is preserved. This observation is pertinent to the interpretation of magmatic foliations. As with layers, foliations can form at any time during or after magma ascent and emplacement. For magmatic foliations to form, there must be enough crystals present to form the foliation. Therefore, they can form as the minerals are crystallizing, or late in the magma cooling history, once the crystals have already grown. In addition, due to the transient and easily reset nature of magmatic crystal alignment, the magmatic foliations will reflect the last processes to act on the magma before it froze. For example, the observation that foliations often overprint layers, implies they must postdate layer-formation. In volcanic dikes, evidence of ascent and emplacement are much more likely to be preserved than in plutons, because these dikes cool quickly, and thus retain records of the operation of early processes. In contrast, deeper and slower cooling plutons are more likely to preserve only the latest processes in the magma chamber. Therefore, if a magma is intruded in a 72 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. very liquid state and cools very slowly, then very little to no information about ascent, emplacement, and early magma chamber dynamics will be preserved. Regardless of the timing of flow, as pointed out earlier in the discussion of magmatic flow, the X-axis of the magmatic strain ellipsoid is not always parallel to the flow direction (Figure 23). As minerals grow and align with their long axes parallel to the X-axis (Nicolas, 1992), they reflect magmatic strain. Thus, while certain types of magmatic layers may directly represent magmatic flow planes, foliations do not. They represent the XY-plane of the strain ellipsoid, which may or may not be parallel to the magmatic flow planes. This would explain the observation that magmatic foliations often cross-cut magmatic layer margins (Figure 24) (e.g. in Berger, 1971; Pitcher and Berger, 1972; Blumenfeld & Bouchez, 1988; Benn and Allard, 1989; Nicolas, 1992). This magmatic strain may be caused by local flow during magma ascent, emplacement, or later magma chamber dynamics (e.g. convection). It may also be a regional strain that is imparted during regional deformation. Finally, foliations cannot be used to interpret the shape of a magma chamber. The assumption in using them for this purpose, is that they represent the flow plane, which is parallel to the walls of the pluton. As foliations only directly represent magmatic strain, not flow, and may form at angles to flow planes, they should not be used to infer the shape of a magma chamber. 73 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Flew Direction Flew Plene Follitlon Figure 24. Showing foliations formed at angles to flow planes (which may be represented by petrographic banding) (modified from Lin and Williams, 1992). 'v l 4* E) THE INTERPRETATION OF LAYERS AND FOLIATIONS IN THE MAIN DONEGAL GRANITE i) THE FORMATION OF THE REGULAR LAYERS From the previous section it is clear that there are many processes that may form layers in plutons. In contrast, only one process forms foliations and lineations-strain. However, as with layer-forming processes, that strain can be a result of processes that operate at any point in the pluton’s history. Therefore, determining timing relationships between layers and foliations is of extreme importance. With the previous analysis in mind, the origins and significance of the regular banding, and the foliations in the Main Donegal Granite will be discussed. The origin of the regular layers is controversial. Originally their formation was attributed to pre-existing sedimentary structures that had undergone granitization (Scott, 1862, 1864, and Blake, 1862 in Pitcher and Berger, 1972). Subsequently, they were thought to be primary flow structures representing magma flowing from the northeast to the southwest end of the pluton (Pitcher and Read, 1959). The lack of sedimentary flow structures, as well as geometrical problems (where did the magma go once it reached the SW portion?), led to an intensive study by Berger (1971). He concluded that the layers formed by deformation of a crystal mush of trondjhemitic composition, with late, interstitial, potassium feldspar-rich fluids. This deformation led to the migration and localization of the late, potassium feldspar-rich fluids along 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. zones of dilation within the pluton, a filter pressing model (Figure 25). This hypothesis is based on three assumptions: 1) that deformation was synplutonic, 2) that all the light layers are younger than the dark layers, and 3) specifically, that the microcline is a late, almost metasomatic mineral. Synplutonic deformation is seen in the minor folding, boudinage and shearing of regular layers within the pluton (Figures 1 2 ,13a, and 13b). Evidence for the dark layers being older was based on observations that material from the dark layers can be found included in light layers; that dark layers “ frequently wedge out into a host of light granite” (Pitcher and Berger, 1972, p, 215); and that bands of light material can be found cross-cutting sets of regular layers (light and dark), especially when these layers are not in their usual NE-SW orientation (Figure 6). Finally, the euhedrality of the microcline led Pitcher and Berger (1972) to conclude that it was metasomatic, or at least formed extremely late in the magma’s cooling history. Indeed, the high concentration of microcline in the phenocryst-supported light layers, does suggest that they may represent late-stage ponding of fluids. Thus three of the criteria necessary for the filter pressing model are fulfilled. I agree with several aspects of Berger's model. However, there are some fundamental problems with his arguments. Both field and microstructural evidence indicate that the light and dark layers formed at the same time. Dark layers can be found cross cutting light layers (Figure 12; Berger classified these as separate “microgranite” dikes), indicating that not all light layers are younger. Further evidence of concurrence of the light and dark layers is given by the example of the dark boudinaged 76 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 25. Berger's filter pressing model for the formation of regular bands and cross-cutting bands. Shaded areas represent earlier trondhjemitic material (dark bands), light areas represent feldspathized zones of dilation (light bands). Long arrows represent shear directions, short arrows represent the direction of movement of alkalies (from Berger, 1971). 77 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. layer (Figures 13a and 13b) that was drawn out with the light magma. While this does suggest that the dark layer was there first, it also implies that neither magma was completely crystallized. This is supported by the presence of magmatic textures in both light and dark layers. They can, therefore, be considered coeval. As no truncated crystals could be found at the contacts between layers, and crystals are interlocking across contacts, it is likely that both magmas were fairly low in crystal content when they intruded. Even magma with low crystal contents (<50%; Miller et al, 1988) has enough crystals present to be able to transmit compressive and shear stresses and, therefore, will be able to fracture and be intruded, especially if strain rates are high. Thus few actual crystals will be truncated; contacts between the "infilling dike and host pluton will be very irregular in detail and some crystals of the host pluton may become free-floating in the dike" (Hibbard and Watters, 1985). The presence of thin bands (centimeters, or less in diameter) and wispy layers also argues that the magmas were low in crystal content. High crystal contents would make the magma far too viscous for such thin layers to form. There is also petrographic evidence that the microcline phenocrysts of the lighter layers are not late, metasomatic minerals. Zonally arranged plagioclase inclusions within the microclines suggest that this mineral (of the supposedly earlier trondjhemitic, dark band parent) grew at the same time as the light layer microcline phenocrysts. Their zonal arrangements suggest that the included crystals nucleated and grew along the side of the growing microcline (if the microclines were replacive then the plagioclase orientation would be more random). In addition, Vernon 78 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (1986) argued that the included plagioclase has no counterpart of the same size in the groundmass and, therefore, is more likely to reflect early crystals that were incorporated in the microcline and isolated before they could grow to their full potential. In the filter pressing model, one would expect to see diffuse contacts, with gradual increases or decreases in microcline content, between light and dark layers. One might also expect to find microcline-rich paths along which the potassium-rich fluid migrated towards the zones of dilatancy. Instead, what is often seen is a complete lack of microcline in many of the dark layers (see Appendix II), and abrupt petrologic contacts between these and the light layers. Additionally, migration of potassium- rich fluids does not account for the greater amounts of biotite seen in the dark layers versus the light layers. In addition, according to Berger's model, filter-pressing takes place late in the cooling history of the magma, when the magma was predominantly a crystal mush with late, interstitial potassium feldspar-rich fluids. However, if crystal contents become >55% then a crystal mush tends to dilate upon deformation (Marsh, 1981), therefore it is unlikely that late-stage filter-pressing played a significant role in the formation of the regular layers. If one considers the orientation of the regular layers (NE-SW, and vertical) and assumes that they represent zones of dilatancy, as in the filter-pressing model, it suggests that the trondjhemitic parent magma was undergoing deformation with the principle axes of stress (c t - i) oriented NE-SW. This orientation is not supported by synplutonic deformation structures in the wall rock, which suggest that oi was 79 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. oriented NW-SE (see section F: “Implications for Other Aspects of the Batholith”, regarding the emplacement of the pluton, for a more detailed discussion). Alternatively, I propose a similar model which involves synplutonic deformation of a magma in order to orient, and to a certain extent, generate, the layers. To resummarise the evidence: the sharp and frequently cross-cutting relationships of some of the regular layers suggests a definite intrusive relationship. Some mechanical mixing is also suggested by the drawing out of light and dark layers. This, in addition to the occurrence of both light and dark cross-bands, suggests that both magmas were coeval. Low crystal contents are supported by the thin width of some of the layers, as well as the lack of evidence of truncation of crystals at contacts. Microcline does not appear to be a late, metasomatic mineral. Finally, synplutonic deformation is supported by the folding, shearing and boudinage of the regular layers, that are drawn out to form thinner layers, or are cross-cut by undeformed cross-bands. With this evidence in mind, I propose a model in which two separate, coeval magmas, one of trondjhemitic composition and one of granitic composition, inject into the same region either by injecting into country rock next to each other, or pulsing into a pre-existing sill (it is worthy of note that the orientation and nature of the stresses represented by this deformation still remain a problem). As these magmas are intruding, they may be modified by several different processes. Incomplete mixing (mingling) of the magmas (as seen in Figures 13a and 13b) would produce millimeter scale layers, or, if extensive enough, more homogeneous bodies of intermediate composition. The magmas may 80 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. also undergo shearing (as seen in Figure 12), which would help align the layers. Additional layer formation by strain alignment (suggested by the magmatic foliation) and perhaps even filter pressing (suggested by the phenocryst-supported light pods) would be aided by this shearing. Crystallization processes (such as boiling, in situ crystallization or fractional crystallization) may also play a minor role in forming some of the thinner layers (as seen in the pegmatite dike in Figure 15). In general, however, the layer-forming processes are mechanically driven. Finally, the magmas, once emplaced, could re-inject each other, aided by the synplutonic deformation. This model is similar to Berger’s in that it involves synplutonic deformation to help form the layers, however, it differs in that the process involves injection and mixing of coeval magmas early in the magma-cooling history, as opposed to the migration of late-stage metasomatic fluids. The model fits nicely with the "lit-par-lit" type of emplacement of the sheets within pluton (Hutton, 1982). The layers are simply a smaller manifestation of this mode of emplacement. Thus the difference between sheets and layers is mainly a scale dependent distinction; they both appear to have formed by the same process (magmatic injection). As sill walls are parallel to magmatic flow planes, the boundaries of sheets, and regular layers that are demonstrably intrusive, represent magmatic flow planes and can be used to infer flow planes during emplacement of that magma. The regular layers that are in contact with raft trains represent the earliest operation of this process, when magma was being emplaced into country rock (Figure 26). Contacts between other bands that are not in direct contact with country rock represent later flow during 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 26. Layers seperated by steeply-dipping quartzite rafts are "mini sheets" that represent emplacement of the magma. Located at Glenleheen, near old school (see figure 5 for location). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. emplacement of sills and pulses into the magma chamber. Internal features of the regular layers, and layers with gradational boundaries may also represent incomplete mixing and other processes, such as strain alignment and perhaps some minor filter pressing. ii) RELATIONSHIP OF THE LAYERS AND FOLIATIONS If many of the regular layers represent magmatic flow, what then do the magmatic foliations represent? As noted in the earlier discussion, such fabrics reflect strain of a deforming or flowing mass. That strain may have been due to regional strain imposed on the magma, or to local magmatic flow; in which case, the foliations may, or may not, be parallel to the flow planes, depending on the nature of the flow field. Due to their occasional cross-cutting nature (Figures 8 , 11a and b, 12a and b, 14, and 19), the magmatic foliations and lineations are interpreted to have formed after the regular layering. It is impossible then, that they represent magmatic flow planes, as any flow parallel to the foliation planes would have disrupted the layers themselves. I suggest instead that the magmatic foliations and lineations are not due to strain caused by local magmatic flow, but formed fairly late in the pluton’s cooling history, when the layers had already formed, and when there were enough crystals in the magmato record a fabric. The late timing of the foliation formation is superbly illustrated in Figures 12a and 12b, where it overprints all other features. As the foliations maintain a consistent NE-SW orientation, and usually only cross-cut layers that are not oriented in this direction, it is 83 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reasonable to believe that they are the result of a regional strain field that was imposed on the pluton as it cooled. Thus, they are interpreted to represent the resulting strain in the magma and to track the XY-plane and X-axis of the strain ellipsoid. The nature of the regional strain that produced the magmatic fabrics is uncertain (the orientation of the principle stress axes suggested by the foliations, layers and sheets also appears contradictory, and will be discussed in more detail in Section F). Regional flattening is indicated by the lack of shear sense indicators associated with magmatic foliations, and boudined dark layers that show opposite senses of shear (see Figures 13a and 13b), as well as by the orientation of synplutonic folds in the wall rock (with axial planes oriented NE-SW). However, if the magmatic fabrics are the result of the same strain field that produced the S-C fabrics in the outer portions of the pluton, then sinistral transpression, not pure shear caused their formation. There are three explanations that satisfy both of these apparently contrasting lines of evidence. The first involves the partitioning of strain in a transpressional environment. The sinistral component of shear may have been taken up in the outer portions of the pluton, while the inner portions recorded only the normal component. Alternatively, the second explanation requires that the strain field changed over the pluton’s history. Flattening may have predominated during pluton emplacement and magmatic fabric formation. The strain field may then have changed to one of sinistral shear, producing the S-C fabrics in the margins of the pluton. The outer portions of the pluton could have accommodated all the strain, shielding the inner portions, which would thus retain magmatic flattening fabrics. 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Finally, the third explanation calls on different rates of strain to account for the dichotomy. Flattening rates may have been relatively fast, and thus, easily recorded in the time it took for emplacement and cooling of the pluton. Shear rates, on the other hand, may have been too slow to be recorded during the relatively fast emplacement and cooling of the pluton. Once the pluton had cooled, the shear would have been recorded only in the outer portions of the pluton, for the same reason as stated before. If there is a small viscosity contrast between an intruding magma and its wall rock, then the magma will be emplaced as a diapir (i.e. a rounded body). Conversely, if there is a large viscosity contrast, then the magma will intrude as a dike (Rubin, 1992). Viscosity is controlled by crystal content. The sheets and the regular layers of the Main Donegal suggest that magma was emplaced as dikes. Thus, the magma, during emplacement, must have contained a high percentage of melt, in order to have achieved the necessary viscosity contrast with its wall rocks (experiments suggest viscosity contrasts up to 1016: Rubin, 1991). In addition, earlier pulses of magma must have cooled, crystallized and increased in viscosity sufficiently, so that later pulses (of more liquid magma) could intrude these earlier pulses, as dikes. However, if two magmas are of very different viscosities, they become mechanically immiscible, and therefore will not mix. As there is evidence that some mixing (albeit incomplete) has taken place (Figures 13a, 13b, and 14), the earlier and later pulses of magma must, at some point, have had similar crystal contents. These constraints suggest the following timing for the intrusion and cooling of the magma pulses that make up the 85 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Main Donegal granite: 1) Dikes intrude with low crystal contents, and begin to cool. Crystal contents and viscosity increase. 2) Later pulses intrude more crystalline (but still not solidified) early pulses. The heat exchange between the cooler, early pulse, and warmer, later pulse, leads to mixing of remaining melt phases, where both pulses are of similar crystal contents and viscosity. 3) The two pulses may also lose some of their coherency and mingle, aided by synplutonic deformation. 4) Finally the magmatic foliation forms late, when the magma has cooled to the point when there is little melt remaining (30-10% melt). This sequence is summarized diagramatically in Figure 27. 86 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission o f th e copyright owner. Further reproduction prohibited without permission. o o o diking and fractional crystallization (fast rates) £ O magma mixing/mingling foliation formation 3-10% melt«— »0% melt 12 900 magmatic ♦ t 15 600 + TIME DEFORMATION FIGURE 27: Schematic diagram showing the sequence of events in the Main Donegal Granite with respect to magma temperature, viscosity and crystal content vs. time and deformation (numbers are approximations, taken from Cruden, 1988) 00 -v l R IMPLICATIONS FOR OTHER ASPECTS OF THE BATHOLITH Ever since the concept of magmatic intrusion was accepted, the problem of accounting for how space is made for a pluton has troubled geologists. Many different mechanisms such as stoping, assimilation, volume loss etc. have been suggested, however, the only real means of making space, involves lowering the Moho or raising the earth's surface (Paterson and Fowler, 1993). All other mechanisms do not increase the volume of the crust, but simply transfer the material (and the "space problem") away from the site of emplacement. We refer to these processes as "Material Transfer Processes", or "MTP's." They can be separated into two categories: "near-field MTP's," i.e. processes that transfer material within the structural aureole of the magma, and "far-field MTP's", i.e. processes that transfer that material toward the Earth’s surface or the Moho (Paterson and Fowler, 1993). Most traditional emplacement processes are actually near-field MTP's. They deal with how material is moved away from the immediate area around the pluton and are the focus of most pluton-emplacement studies. They are summarized in Figure 28. The emplacement of a magma into a tectonic opening, such as a strike slip jog, or dilational shear zone, is an attractive model. It transfers country rock material away from the area where the magma is intruding, without the problems of other near-field MTP's, such as assimilation and/or stoping (which are limited thermodynamically), or ductile flow during diapirism and ballooning (for which the required strains in the 88 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 28. Pluton emplacement mechanisms. 1 - stoping; 2 - doming of roof, block elevation along faults; 3 - ductile wall rock deformation and aureole wall rock return flow; 4 - wall rock assimilation, anatexis, zone melting; 5 - lateral wall rock displacement by faulting or folding; 6 - emplacement into an extensional environment (from Paterson et al., 1991) 89 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. country rock are rarely found; Paterson and Fowler, 1993). The Main Donegal Granite was one of the first plutons for which this increasingly popular model was proposed. Based on the overall NE-SW trend of the pluton and its internal features, the penetrative subsolidus shear rimming the pluton, and the magmatic foliations in the center of the pluton, Hutton (1982) proposed that the pluton was emplaced, incrementally, as a series of sheets into a transtensional left-lateral strike slip shear zone. According to his model, the NE-SW trending fault opened due to greater displacement in the NE compared to the SW portion of the fault. This difference caused the fault zone to bend and open, creating space for the intruding sheets of magma (Figure 29). The overall tectonic environment in Hutton’s model is transtensional, with the shear zone active before, during and after emplacement of the Main Donegal. The shape of the older plutons in the region suggest that a shear zone was active before the Main Donegal Granite was emplaced. The Thorr granodiorite and the Ardara quartz monzodiorite/granodiorite, which are older than the Main Donegal granite, have roughly circular map patterns to the west, but have long, drawn-out "tails" extending to the NE. In the case of the Thorr, inclusions of the tails of these plutons can be traced, just like the other country rocks, as continuous raft trains from the SW all the way to the NW tip of the pluton. Tail inclusions of the Ardara can be traced halfway, as far NE as Carbat Gap (see location in Figure 4; Pitcher, pers.comm.). The Main Donegal granite (including the Trawenagh Bay pluton) mimics this pattern, with an elongate "tail" (the sheeted Main Donegal pluton) extending to the NE. The sheets of the Main Donegal pluton die out into the SW part of the intrusion (the 90 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 29. Schematic representation of opening shear zone, from Hutton 1982. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Trawenagh Bay pluton) (Pitcher and Berger, 1972) which forms a roughly circular map pattern. The shape of the pluton and the NE-SW trending sheets, layers, and magmatic foliations support the incremental opening of the shear zone during the emplacement of the granite. The overprinting, but parallel, subsolidus foliations, S-C fabrics and mylonitic deformation around the margins of the pluton imply that the shear zone continued to operate after the pluton had been emplaced. Although the shear zone was active during magma intrusion, its role in the actual transfer of country rock during emplacement of the pluton is debatable. If the shear zone is responsible for making space locally for the pluton by moving wall rock aside laterally, then the units on either side of the pluton should match up if the pluton were removed. No match can be made, although the geology in this area is complicated by folds and the Knockateen Slide (a ductile shear zone; Figure 30). The presence of the raft trains that are continuous from the wall rock, far into the pluton (Pitcher and Berger, 1972) suggest that these sections of wall rock, at least, have not moved to the lateral extent that would be expected with this model. The very slight bowing of country rock at the center of the pluton, (in map view; Figure 4), suggests there may be a minor amount of outward, lateral movement. However, it is certainly not enough to account for the entire area of the pluton. The shear zone should also border the entire pluton, yet this cannot be demonstrated at the southwest end near Loch Doo, nor at the NE end, where foliations are horizontal or gently dipping. The truncation of wall rock in this area, as well as at the NE end of the pluton (Figure 30) implies that in these 92 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. GEOLOGIC MAP OF THE MAIN DONEGAL PLUTON, NW DONEGAL, IRELAND (Aller Pilcher and Bargor, 1972) Intruslva Rocks Main Donegal Granite Rosses Granite Ardara Granodlortle ■ Quartz Monzonlte Thorr Granodlortle Dalradlan Kllmacrenan (Islay) Succession: NW lacles Crana Quartzlie Termon Pollies Slleve Tooey QuarUile Glencolumbkllle Pellle Falcarragh Limestone Falcarragh Pellle Loughros Formation Sessalgh-Clonmass Group Ards Quartzlie Aids Pellle I Creeslough Formation • Fault v Tectonic slide M i Miles Figure 30. Geologic map of the Main Donegal Granite, NW Donegal, Ireland to (after Pitcher and Berger, 1972) CO areas wall rock was removed by vertical displacement, rather than by horizontal displacement. In addition, upright, gently plunging folds with NE-SW trending axial planes (Figure 31) that are pluton-related (Hutton, 1977 & 1982) can be found on the sides, as well as at the ends of the pluton (e.g. at Loch Doo; Figure 31). This implies that the tectonic environment was transpressional rather than transtensional. It also suggests that the axis of principle strain (X axis) was oriented NW-SE and the axis of least principle strain (Z axis) was oriented NE-SW (Figure 32). This orientation is supported by the boudined layers, magmatic folds and magmatic foliations in the center of the pluton. Assuming strain and stress were coaxial, then the orientation of the principle stress axis (ctt) during emplacement was NW-SE. If this scenario is correct, then the sheets that make up the pluton, and the regular layers that were injected into these sheets, intruded parallel to the C 2 - < * 3 plane and perpendicular to c t-i - 02- In traditional Andersonian diking, dikes should intrude with their walls perpendicular to 0 2 - 03. This implies that either our interpretation of the orientation of the principle stress axes are erroneous, or the dikes intruded in a non-Andersonian manner. Hutton (1988) noted this contradiction and suggested that the orientation of the dikes was governed by pre-existing anisotropies (such as country rock orientations and the shear zone). Injection of magma at high angles to < n - 02 can be acheived in highly stratified rocks if magma pressures are high and deviatoric stress is low (Lucas and St-Onge, 1995). The resistance of a rock to rupture in a certain orientation can be thought of as a combination of the stress and the tensile strength of the rock perpendicular to the 94 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 31. Pegmatite dikes intrude the wall rocks at the SE tail of the pluton, at Loch Doo (see Figure 22). Dikes intrude both axial planes and bedding planes of the folded pelitic schists, and in turn, are folded themselves. Later offshots of magma cross-cut these folds, implying that the dikes are synplutonic. This axial plane strikes NE-SW and dips to the NW. This orientation, plus its location at the tail of the pluton, suggests that pure shear, or contraction was taking place during emplacement, not just sinistral shear (as in Hutton’s model). 95 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0*1 o r 03 X ' / W l according to boudins, foliations and wall rock folds. Suggests Non-Andersonian diking according to sheets Figure 32. Principle stress axis suggested by diking (see Figure 4 for map of sheets, Figure 5 for sheet and regular layer orientations), versus stress axes suggested by synplutonic folds in the wall rock and flattening fabrics in the pluton (with the assumption that stress and strain were coaxial). 96 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. plane in question (ctx + Tx). Once that stress and tensile strength are overcome (for example, due to melt pressure; i.e. Pm e it > < r x + TX ), then the rock will rupture along the plane with the least "strength". In order for a rock to rupture perpendicular to the principle axis of stress, oi + T-\ must be smaller than 0 -3 + T3 (Figure 33). In highly stratified rocks the difference between T 1 and T3 may be great enough for <ii + T, < 03 + 13. However, as tensile strength values are small in comparison to stress values, the deviatoric stress (01 - 03) must be small in order for differences in T| and T3 to affect the rupture direction of the rock. Another explanation would involve the rapid, 90° rotation of the stress field immediately surrounding the pluton. This rotation can occur during earthquake events in stick-slip type fault zones. However, as this switch takes place instantaneously, the timescales involved are probably too short to accomodate the emplacement of a kilometer long, 100m wide sheet. If the shear zone was operating during emplacement, then the magmatic features that suggest pure shear instead of simple shear (the layers, boudins and foliations) can be explained by partitioning of strain, or, alternatively, slow rates of shear in comparison to fast rates of intrusion and crystallization. Each pulse of magma intruded and cooled quickly enough, before much simple shear could take place, and thus, only record the pure shear strain during emplacement. The outer sheets which do show subsolidus S-C fabrics were exposed to syn- and post emplacement simple shear, and thus the pure shear fabrics were overprinted by subsolidus simple shear fabrics. This outer part of the pluton acted as a buffer zone for the core of the pluton, and prevented the 97 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31 • Pmett °3 * "melt High Melt Pressure P m elt> 3 i + T | Low Deviatoric Stress (03 + T3) > (cti +T1) Normal Stress (on) Intruded sheet T3 (Tensile strength parallel to foliation) T1 (Tensile strength perpendicular to foliation) Figure 33. Mohr circle construction and block diagram showing how sheets may be emplaced perpendicular to ai, due to a combination of pre-existing anisotropies and low deviatoric stress (modified from Lucas and St-Onge, 1995). 98 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. inner magmatic, flattening fabrics and structures from being overprinted by subsolidus simple shear. The Main Donegal Granite is constructed of a series of sheets that contain layers and cross-cutting magmatic and sub-solidus foliations, implying that the pluton was constructed incrementally. Thus, the formation of the pluton, layers and foliations was a time transgressive process. Each time a new sheet was intruded, layers would form. As the sheet cooled, a magmatic foliation would form. This would have to happen repeatedly throughout the entire construction of the pluton, and suggests that the regional strain field, and thus, probably the stress field, did not change throughout this period. Older sheets, layers, and foliations might be expected to show substantial rotation if they were not intially parallel to the shear zone. As few do, it is suggested that either most of these features were originally parallel to the shear plane, or, that not only the formation of the layers, boudins and foliations, but the entire emplacement of the pluton was fairly rapid in relation to the operation of the shear zone. Thus the orientations of all of these structures remain consistently NE-SW. 99 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3: CONCLUSIONS The interpretation of what magmatic layers and foliations represent is crucial to our understanding of the processes that operate in a pluton's history. This study has shown that some of our traditional assumptions, in particular, that magmatic layers and foliations represent magmatic flow planes, are often erroneous. The Main Donegal Granite is layered on several different scales, from kilometer-wide sheets, to millimeter-wide layers, all of which are parallel to or cross-cut by a penetrative magmatic foliation, which, in turn, is cross-cut by a subsolidus foliation at the pluton margins. The formation, and relationships of these features to one another, can be used to interpret several aspects of the magma chamber’s history. Sheets and regular layers provide rarely-preserved information regarding flow during emplacement. They also illustrate later magma chamber processes such as magma mixing and some fractionation processes. Finally, the foliations can be used to interpret strain and possibly make inferences regarding the orientations of regional stresses at the time of emplacement. I will begin with the earliest of the features to form, and work chronologically forwards, discussing what each structure represents: The first features formed were the sheets that make up the entire pluton. The continuous raft trains that mark many of their margins, as well as their obvious intrusive nature at the pluton contacts, indicate that they are dikes. The only layers that may represent magmatic flow during ascent and emplacement are those formed by magmatic injection (i.e. 100 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. intrusive in nature); all other types of magmatic layers form by processes that operate after the magma chamber is constructed- Therefore, the margins of the sheets that make up the Main Donegal Granite imply that magmatic flow planes during ascent and emplacement of the pluton were oriented striking NE-SW and steeply dipping. Do the regular layers that persist throughout most of the pluton represent flow then? The clearly intrusive nature of many of the regular layers in the Main Donegal Granite suggest that they, too, were formed by diking and pulsing, and are simply smaller manifestations of the processes that formed the sheets. Thus their margins do represent magmatic flow during ascent and emplacement. Other regular layers (for example, the millimeter-scale layers, or layers with gradational contacts) show evidence of processes that operated after emplacement, such as mixing and fractionation, and do not necessarily represent flow planes. The magmatic foliations and lineations in the Main Donegal Granite must have formed late in the pluton’s cooling history, as they are consistent throughout the pluton, overprinting the regular layers and sheets, and often cross-cutting layer margins. Their cross-cutting nature implies that they are not parallel to magmatic flow planes. As illustrated in Chapter 2, magmatic foliations (and lineations) represent magmatic strain, not magmatic flow. They may form at angles to flow planes and directions and therefore should not be used to infer magmatic flow planes. The orientation of some magmatic foliations at angles to layers in the Main Donegal Granite, illustrates this perfectly. The strain that these foliations represent, appears to be due to regional stresses, and not stresses caused by local magmatic flow, as the foliations maintain a 101 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. consistent NE-SW orientation throughout the pluton and cross-cut some internal contacts. The gradual transition to, and parallelism with solid state foliations that represent the shear zone (which is a strain feature caused by regional stresses), also suggest that the magmatic foliations represent those same regional stresses. The subsolidus fabrics (sub-solidus foliations, mylonites and S-C structures), located at the outer margins of the pluton, overprint all features and therefore were the last to form. The increasing intensity towards the margins of the pluton indicate the presence of a sinistral shear zone at, or near the pluton sides, but not at the pluton "ends" where the shear zone cuts across the pluton and then cannot be traced. The parallelism of the subsolidus fabric and the other features that formed earlier, suggest that they may have formed due to the same stresses. If this is the case then it is possible that the shear zone operated throughout the pluton's history, before, during and after pluton emplacement. As the magmatic foliation represents magmatic strain (and assuming that stress and strain were coaxial), the corresponding parallelism of the sheets, layers, and pluton itself, suggests that regional stress may have influenced the emplacement of the sheets and layers. In addition, the parallelism these features with the earlier layered "tails” of other plutons, and the later subsolidus foliations around the pluton, suggest that the regional stresses maintained their orientation before, during and after emplacement of the Main Donegal Granite. Structures within the Main Donegal Granite (e.g. boudins showing inconsistent directions of shear) indicate contraction during 102 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. emplacement, as do folds found at the ends and sides of the pluton. Conversely, S-C structures on the margins of the pluton suggest sinistral shear. This dichotomy may be a result of differences in the rates of emplacement and cooling versus rates of shear, changes in the strain field, or a partitioning of strain. Regardless, all of the above structures suggest a transpressional environment, not the transtensional environment that is necessary for Hutton’s model of an opening shear zone. Alternatively, the shear zone may not have played a role in the making space for the pluton at all. It should, theoretically, border the entire pluton, but instead, cuts across the pluton in the Trawenagh Bay area, and is untraceable at the southwest end of the pluton. In addition, the difficulty in matching up wall rock on either side of the shear zone, as well as the continuity of raft trains from the wall rock into, and throughout the pluton, suggest vertical movement of wall rock, instead of lateral movement, to produce space for magma. There remains a dichotomy between the orientation of the stresses suggested by the structures in and around the pluton, and the emplacement of the sheets (and regular layers). Boudins, foliations and folds suggest that < ti (axis of maximum principle stress) was oriented NW-SE. However, the sheets and regular layers, if representing traditional Andersonian diking, suggest that c t i was oriented NE-SW. Pre-existing anisotropies (plus low deviatoric stress), or rapid rotation of stress fields around the pluton may account for this discrepancy. What is clear, upon review of layer-forming processes, is that few layers are likely to directly represent magmatic flow. Only those formed by magmatic injection, or those formed by depositional sedimentary 103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. processes (a result of density currents or compositional convection in an existing magma chamber) can be used to infer magmatic flow planes. Thus the only layers that can represent magmatic flow planes during ascent and emplacement, are those formed by magmatic injection. Therefore, sheeted complexes, such as the Main Donegal Granite, are the only plutons where layers can be used to infer these early processes. All other layers within plutons represent processes that operate after the magma chamber has been constructed. Within this category, only depositional cumulates, and the margins layers that have formed by later pulsing can be used directly to infer paleo-flow planes within the magma chamber. Magmatic foliations, on the other hand, represent magmatic strain, which may be due to local flow (taking place at any point in the magma chamber history), or may be caused by regional tectonic stresses. Regardless of what causes the deformation, magmatic foliations, like most layer-types, should not be directly used to infer magmatic flow planes. For this reason, they also should not be used to interpret the shape of the magma chamber. In addition, if magmatic foliations are to be preserved, they must form late in the magma chamber cooling history. Consequently, they only record strain that was imposed on the magma just before it froze. Therefore, unless other evidence can be found that suggests that the strain fields remained constant from ascent to cooling, they also should not be used to make any inferences about ascent and emplacement. 104 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. REFERENCES CITED Abbott, R.N., 1989, Internal structures in part of the South Mountain batholith, Nova Scotia, Canada: Geological Society of America Bulletin, v. 101, p. 1493-1506. Balk, R., 1937, Structural Behaviour of Igneous Rocks: Geological Society of America Memoir, v. 5, p. 1-177. Barker, F. and Arth, J. G., 1976, Generation of trondjhemitic-tonalitic liquids and Archean bimodal trondjhemite-basalt suites: Geology, v. 4, p. 596-600. Barriere, M., 1976, Flowage Differentiation: Limitation of the "Bagnold Effect" to the Narrow Intrusions: Contributions to Mineralogy and Petrology, v. 55, p. 139-145. Barriere, M., 1981, On Curved Laminae, Graded Layers, Convection Currents and Dynamic Crystal Sorting in the Ploumanac’h (Brittany) Subalkaline Granite: Contributions to Mineralogy and Petrology, v. 27 p. 214-224. Benn, K. and Allard, B., 1989, Preferred Mineral Orientations Related to Magmatic Flow in Ophiolite Layered Gabbros: Journal of Petrology, v. 30, p. 925-946. Berger, A.R., 1971, The origin of banding in the Main Donegal Granite, N.W. Ireland: Geological Journal, v. 7, 2, p. 347-358. Best, M.G., 1982, Igneous and Metamorphic Petrology: New York, W.H. Freeman and Company, 630 p. Biermann, C. and Van Roermund, H.L.M., 1983, Defect structures in naturally deformed clinoamphiboles - A TEM study: Tectonophysics, v. 95, p. 267-278. Blake, E.H., 1862, On the primary rocks of Donegal: Journal of the Geological Society of Dublin, v. 9 , p. 294-300. Blumenfeld, P. and Bouchez, J.-L., 1988, Shear criteria in granite and migmatite deformed in the magmatic and solid states: Journal of Structural Geology, v. 10, 4, p. 361-373. Boudreau, A.E., 1987, Pattern formation during crystallization and the formation of fine-scale layering, in Parsons, I., eds., Origins of Igneous Layering: Dordrecht, D. Reidel Publishing Company, 196, p. 453-471. 105 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Burg, J. P. and Iglesias Ponce De Leon, M., 1985, Pressure-solution structures in a granite: Journal of Structural Geology, v. 7, p. 431-436. Butcher, A. R., Young, I. M. and Faithful, J. W., 1985, Finger structures in the Rhum Intrusion: Geological Magazine, v. 122, p. 491-502. Callaway, C., 1885, On the granitic and schistose rocks of northern Donegal: Quarterly Journal Geological Society of London, v. 41, p.221-239. Clark, R. G., Jr. and Lyons, J. B., 1986, The Kinsman intrusive suite: Journal of Petrology, v. 27, p. 1365-1393. Clemens, J. D. and Mawer, C. K., 1992, Granitic magma transport by fracture propagation: Tectonophysics, v. 204, p.339-360. Cruden, A.R., 1988, Deformation around a rising diapir modeled by creeping flow past a sphere: Tectonics, v. 7, p. 1091-1101. DeVore, G.W., 1969, Preferred mineral distributions of polymineralic rocks related to non-hydrostatic stresses as expressions of mechanical equilibrium: Journal of Geology, v. 77, p. 26-38. Duke, E.F., Redden, J.A. and Papike, J.J., 1987, Calamity Peak layered granite-pegmatite complex, Black Hills, South Dakota: Part 1. Structure and emplacement: Geological Society of America Bulletin, v. 100, p. 825-840. Ernst, E.R. and Baragar, W.R.A., 1992, Evidence from magnetic fabric for the flow pattern of magma in the Mackenzie giant radiating dike swarm: Nature, v. 356, p. 511-513. Fernandez and Laporte, 1991, Significance of low symmetry fabrics in magmatic rocks: Journal of Structural Geology, v. 13, p. 337-347. Frost, T.P. and Mahood, G.A., 1987, Field, chemical, and physical constraints on mafic-felsic magma interaction in the Lamarck Granodiorite, Sierra Nevada, California: Geological Society of America Bulletin, v. 99, p. 272-291. Gapais, D., 1989, Shear structures within deformed granites: Mechanical and thermal indicators: Geology, v. 17, p. 1144-1147. Gapais, D. and Babarin, B., 1986, Quartz fabric transition in a cooling syntectonic granite (Hermitage Massif, France): Tectonophysics, v. 125, p. 357-370. 106 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Goode, A.D.T., 1977, Vertical igneous layering in the Ewarara layered intrusion, central Australia: Geological Magazine, v. 114, p. 365- 374. Guinberteau, B., Bouchez, J.L. and J.L., V., 1987, The Mortagne granite pluton (France) emplaced by pull-apart along a shear zone: Structural and gravimetric arguments and regional implication: Geological Society of America Bulletin, v. 99, p. 763-770. Haeussler, P. J. and Paterson, S. R., 1993, Tilting, burial, and uplift of the Guadalupe Igneous Complex, Sierra Nevada, California: GSA Bulletin, v. 105, p. 1310-1311. Haughten, S., 1862, Experimental researches on the granites of Ireland: Part III. On the granites of Donegal: Quarterly Journal Geological Society of London, v.18, p.403-420. Hibbard, M.J. and Watters, R.J., 1985, Fracturing and diking in incompletely crystallized granitic plutons: Lithos, v. 18, p. 1-12. Hibbard, M. J., 1986, Deformation of Incompletely Crystallized Magma Systems: Granitic Gneisses and their Tectonic Implications: Journal of Geology, v. 95, p. 543-561. Hunter, R. H., 1987. "Textural equilibrium in layered igneous rocks", in I. Parsons, ed., Origins of Igneous Layering. D. Reidel, Dordrecht, p. 43-503. Husch, J.M., 1990, Palisades Sill: Origin of the olivine zone by separate magmatic injection rather than gravity settling: Geology, v. 18, p. 699- 702. Hutton, D.H.W., 1977, A structural cross-section from the Aureole of the Main Donegal Granite: Geological Journal, v. 12, p. 99-112. Hutton, D.H.W., 1982, A tectonic model for the emplacement of the Main Donegal Granite, Ireland: Journal of the Geological Society, London, v. 139, p. 615-631. Hutton, D.H.W., 1988, Granite emplacement mechanisms and tectonic controls: inferences from deformation studies: Transactions of the Royal Society of Edinburgh: Earth Sciences, v. 79, p. 245-255. Irvine, T.N., 1982, Terminology for Layered Intrusions: Journal of Petrology, v. 23, p. 127-162. 107 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Irvine, T. N., 1987, "Layering and related structures in the Duke Island and Skaergaard intrusions: Similarities, differences, and origins", in I. Parsons, ed.. Origins of Igneous Layering. D. Reidel Publishing Company, Dordrecht, p. 184-245. Jurewicz, S. R. and Bruce, E., 1985, "The distribution of partial melt in a granite. The application of liquid phase sintering: Geochimica et Cosmochimica Acta, v.49, p. 1109-1121. Kearey, P. and 1993, The Encyclopedia of the Solid Earth Sciences. Blackwell Scientific Publications, Oxford, 713p. Kerr, R.C. and Tait, S.T., 1986, Crystallization and compositional convection in a porous medium with application to layered igneous intrusions: Journal of Geophysical Research, v. 91, B3, p. 3591-3608. Komar, P.D., 1972a, Mechanical interactions of phenocrysts and flow differentiation of igneous dikes and sills: Geological Society of America Bulletin, v. 83, p. 3443-3448. Kouchi, A. and Sunagawa, I., 1985, A model for mixing basaltic and dacitic magmas as deduced from experimental data: Contributions to Mineralogy and Petrology, v. 89, p. 17-23. Le Bas, M. J., 1982, "The Caledonian granites and diorites of England and Wales", in D. Sutherland, ed., Igneous Rocks of the British Isles. Wiley, Chichester, p. 191-214. Lucas, S.B. and St-Onge, M.R., 1995, Syntectonic magmatism and the development of compositional layering, Ungava Orogen (northern Quebec, Canada): Journal of Structural Geology, v.17, p.475-491. Mahood, G.A., 1986, Waxing and waning magma chambers: a comparison of zoning in ignimbrites and plutons: Intl. Mineral. Assoc., Abstracts, 14th Gen. Mtg., p. 163. Martin, D., 1990, Crystal settling and in situ crystallization in aqueous solutions and magma chambers: Earth and Planetary Science Letters, v. 96: 336-348. McBirney, A.R. and Noyes, R.M., 1979, Crystallization and Layering of the Skaergaard intrusion: Journal of Petrology, v. 20, p. 487-554. McBirney, A.R., 1993, Magmatic differentiation: mechanisms and effects, in eds., Igneous Petrology: London, Bartlett Publishers International, p. 131-172. 108 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Means, W.D. and Park, Y., 1994, New experimental approach to understanding igneous texture: Geology, v. 22, 4, p. 323-326. Miller, C.F., Watson, E.B. and Harrison, T.M., 1988, Perspectives on the source, segregation and transport of granitoid magmas: Transactions of the Royal Society of Edinburgh: Earth Sciences, v. 79, p. 135-156. Nabelek, P. I. and Taylor, L. A., 1978, Nucleation and growth of plagioclase and development of textures in a high-alumina basaltic melt: Proceedings of the Lunar and Planetary Science Conference #9, p. 725-741. Nicolas, A., 1992, Kinematics in Magmatic Rocks with Special Reference to Gabbros: Journal of Petrology, v. 33, 4, p. 891-915. Nicolas, A., Freydier, C., Godard, M. and Vauchez, A., 1993, Magma chambers at oceanic ridges: How large?: Geology, v. 21, p. 53-56. O'Connor, P.J., Long, C.B., Kennan, P.S., Halliday, A.N., Max, M.D. and Roddick, J.C., 1982, Rb-Sr isochron study of the Thorr and Main Donegal Granites, Ireland: Geological Journal, v. 17, p. 279-295. Pande, I.C., 1954, The geology of the Kilmacrenan district, County Donegal [PhD]: Imperial College, London. Park, Y. and Hanson, B., 1993, An experimental investigation of Ostwald ripening of crystalline phases in silicate melt: Geological Society of America Abstracts with Programs, v. 25, p. 96. Parsons, I., 1979, The Klokken gabbro-syenite complex, South Greenland: cryptic variation and origin of the inversely graded layering: Journal of Petrology, v. 20, p. 653-694. Parsons, I. and Butterfield, A. N., 1981, Sedimentary features of the Nunarssuit and Klokken syenites, S. Greenland: Journal of the Geological Society of London, v. 138, p. 289-306. Paterson, S.R. and Fowler, T.K. Jr., 1993, Re-examining pluton emplacement processes: Journal of Structural Geology, v. 15, p. 191- 206. Paterson, S.R., Vernon, R.H. and Fowler, T.K. Jr., 1991, Aureole tectonics: Reviews in Mineralogy: Contact Metamorphism, v. 26, p.673-714. 109 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Paterson, S.R., Vernon, R.H. and Tobisch, O.T., 1989, A review of criteria for the identification of magmatic and tectonic foliations in granitoids: Journal of Structural Geology, v. 11, 3, p. 349-363. Paterson, S.R., Yuan, E.S., Miller, R.B., Pitcher, W.S., 1994, Emplacement of the Main Donegal Granite, Ireland: Geological Society of America Abstracts with Programs, v.26, p.80. Philpotts, A.R. and Asher, P.M., 1994, Magmatic flow-direction indicators in a giant diabase feeder dike, Connecticut: Geology, v. 22, 4, p. 363- 366. Pitcher, W.S. and Berger, A.R., 1972, The Geology of Donegal: A study of Granite Emplacement and Unroofino: New York, John-Wiley and Sons, 435 p. Pitcher, W.S. and Read, H.H., 1959, The Main Donegal Granite: Geological Society of London Quarterly Journal, v. 114, p. 259-305. Pitcher, W.S. and Spencer, M.O., 1972, Solid Geology of Northwest and Central Donegal, Map 1 for The Geology of Donegal: A Study of Granite Emplacement and Unroofing. Pitcher, W.S. and Berger, A.R., Wiley-lnterscience, New York. Raleigh, C.B., 1968, Mechanisms of Plastic Deformation of Olivine: Journal of Geophysical Research, v. 73, p. 5391-5406. Robin, P.-Y.F., 1979, Theory of metamorphic segregation and related processes: Geochim. Cosmochim. Acta, v. 43, p. 1587-1600. Robin, P. Y. F., 1979, Theory of metamorphic segregation and related processes". Geochim. Cosmochim. Acta, v.43, p. 1587-1600. Robins, B., Haukvik, L. and Jansen, S., 1987, "The organization and internal structures of cyclic units in the Honningsvag Intrusive Suite, North Norway: Implications for intrusive mechanisms, double diffusive convection and pore-magma infiltration", in I. Parsons, ed., Origins of Igneous Layering. D. Reidel Publishing Company, Dordrecht, p. 287-312. Roedder, E. W., 1959, Fluid inclusions as samples of the ore-forming fluids. Geological Society of America Bulletin,: v. 12, p. 1663. Roedder, E., 1979, Silicate liquid immiscibilty, in H. S. Yoder Jr., eds.,_ The evolution of the igneous rocks: 50th anniversary perspectives. Princeton University Press, Princeton, New Jersey, p. 15-57. 110 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rubin, A. M., 1992, Dikes and Diapirs in Cracking Viscoelastic Media: EOS, v.72, p. 279. Sawka, W.N., Chappell, B.W. and Kistler, R.W., 1990, Granitoid Compositional Zoning by Side-wall Boundary Layer Differentiation: Evidence from the Palisade Crest Intrusive Suite, Central Sierra Nevada, California: Journal of Petrology, v. 31, p. 519-553. Schmeling, H., Cruden, S.R. and Marquart, G., 1988, Finite deformation in and around a fluid sphere moving through a viscous medium: implications for diapiric ascent: Tectonophysics, v. 149, p. 17-34. Scott, R.H., 1862, On the granitic rocks of the southwest of Donegal, and the minerals therewith associated: Journal of the Geological Society of Dublin, v. 9, p. 285-294. Scott, R.H., 1864, On the granitic rocks of Donegal, and the minerals therewith associated: Journal of the Geological Society of Dublin, v. 10, p. 13-24. Shelley, D., 1985, Determining paleo-flow directions from groundmass fabrics in the Lyttelton radial dykes, New Zealand: Journal of Volcanology and Geothermal Research, v. 25, p. 69-79. Simpson, C., 1985, Deformation of granitic rocks across the brittle-ductile transition: Journal of Structural Geology, v. 7, p. 503-511. Sinha Roy, S., 1977, Mylonitic microstructures and their bearing on the development of mylonites - an example from deformed trondjhemite of the Bergen Arc region, SW Norway: Geological Magazine, v. 114, p. 445-458. Smith, J.V., Miyake, Y. and Yamauchi, S., 1993, Flow direction and groundmass shear zones in dykes, Shimane Peninsula, Japan: Geological Magazine, v. 130, 1, p. 117-120. Tullis, J., 1983, Deformation of feldspars, in Ribbe, P.H., eds., Feldspar Mineralogy: 2, p. 297-323. Vernon, R.H., 1986, K-feldspar Megacrysts in Granites - Phenocrysts, not Porphyroblasts: Earth-Science Reviews, v. 23, p. 1-63. Vernon, R.H., Williams, V.A. and D'Arcy, W.F., 1983, Grainsize reduction and foliation development in a deformed granitoid batholith: Tectonophysics, v. 92, p. 123-145. 111 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Vigneresse, J.L and Bouchez, J.L., 1990, Use and misuse of geophysical data to determine the shape at depth of granitic intrusions: Geological Journal, v.25, p.249-260. Wager, L.R. and Brown, G.M., 1968, Layered Igneous Rocks: Edinburgh, Oliver and Boyd, Ltd., 588p. Wager, L.R., 1959, Differing Powers of Crystal Nucleation as a Factor Producing Diversity in Layered Igneous Intrusions: Geological Magazine, v. 96, p. 75-80. Wager, L.R., Brown, G.M. and Wadsworth, W.J., 1960, Types of Igneous Cumulates: Journal of Petrology, v. 1, p. 73-85. Wickham, S.M., 1987, The segregation and emplacement of granitic magmas: Journal of the Geological Society of London, v. 144, p. 281- 297. Williams, P.F., 1972, Development of metamorphic layering and cleavage in low grade metamorphic rocks at Bermagui, Australia: American Journal of Science, v. 272, p. 1-47. Williams, P.F., 1990, Differentiated layering in metamorphic rocks: Earth- Science Reviews, v. 29, p. 267-281. Williams, Q. and Tobisch, O.T., 1994, Microgranitoid enclave shapes and magmatic strain histories: constraints from drop deformation theory: Journal of Geophysical Research, v. 99, B12, p. 24,359-24,368. Yuan, E.S. and Paterson, S.R., 1993a, Petrographic banding and foliations in the Main Donegal Granite: which represents flow?: Geological Society of America annual Cordilleran and Rocky Mountain Section, Reno, Nevada, p. 168. Yuan, E.S. and Paterson, S.R., 1993b, Evaluating flow from structures in plutons: Geological Society of America Abstracts with Programs, v.25, p. 305. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX I A) LAYER-FORMING PROCESSES i) MAGMATIC PROCESSES Magmatic processes operate in a melt dominated system, where the magma is below the liquidus, but well above the solidus. Liquid immiscibility (for a review of mechanical immiscibility see text, p.57) There are three magma systems that may yield immiscible liquids due to chemical differences. Iron-rich sulfide liquids can separate out from mafic silicate magmas that contain only a minor amount of sulfur (a few hundredths of a percent by weight; Best, 1982). C0 2 -rich alkaline magmas may yield alkali- and silica-rich liquids, as well as C0 3 -rich liquids that are immiscible. Evidence of this immiscibility is seen in small liquid inclusions in minerals in nephelinite (ijolite)-carbonitite complexes (Best, 1982). Finally, iron-rich tholeiitic magmas, in the late stages of crystallization, may separate out into a mafic, iron and phosphorous-rich liquid and a felsic, silica-rich liquid. This immiscibility field was first outlined in experiments by Roedder (1959 & 1979). Irregular pods of granophyre found in gabbroic rocks in the Upper Zones B and C of the Skaergaard intrusion are believed to form in this manner (Roedder, 1979). Often immiscible liquids may not separate out until much of the magma has already crystallized. Therefore, they will not form layers, but instead will be situated in the interstices between crystals (McBirney, 1993). It may be difficult to determine if layers are formed by the 113 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. separation of two chemically immiscible liquids, or by some other process such as incomplete mixing (i.e. mechanical immiscibility). Layers that truly formed from the separation of two immiscible liquids should have sharp contacts, should be in chemical equilibrium with each other and, therefore, should contain the same crystalline phases (McBirney, 1993). Layers may form as a combination of both chemical and mechanical differences. For example, if two chemically immiscible liquids have different densities, then one may rise and float above the other to form layers. Alternatively, layers may form entirely due to mechanically- induced immiscibility. Magma mixing (see text, p.58) Magma injection (see text, p.59) Assimilation (see text, p.57) ii) NEAR SOLIDUS PROCESSES The following layer-forming processes all involve fractional crystallization, which is one of the most frequently called upon mechanisms for generating heterogeneities within magmas. It involves the crystallization of minerals from an originally homogenous melt and removal or isolation of those crystals from the melt. This results in two components: crystals which may accumulate by various processes to 114 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. form layers, and a liquid which is significantly different in composition from its parent magma. Side wall in situ crystallization This is the process of in situ crystallization, where crystals grow in boundary layers along the walls and roof of a pluton. McBirney and Noyes (1979) called on a combination of in situ fractionation, nucleation diffusion and double diffusive convection (see below) to account for the formation of layers in the Skaergaard Intrusion. Comb layering, where crystals grow outward from the walls of a magma chamber, is an example of in situ crystallization. The texture and orientation of crystals will be a function of temperature gradients and the composition of the static boundary layer (McBirney and Noyes, 1979). Different compositional and physical patterns will form depending on the nature of the crystallization process (e.g. if diffusion rates keep up with crystallization rates), and the physical nature of the magma (static or flowing?). For example, if diffusion rates are slower than crystallization rates, then the chemical, or physical inhomogeneities within the magma may be preserved, as the magma did not have time to homogenize before it was frozen. Crystal settling Layer-forming processes may also involve mechanical sorting by gravitational segregation or density currents. With gravitational segregation more dense crystals sink to the bottom of the magma chamber and less dense crystals float to the roof zone. Crystals must be 115 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. heavy, or light enough that the rate of sinking/floating is great enough for them to escape the crystallization front (McBirney and Noyes, 1979). Precipitation or floatation can only take place if crystals are forming in suspension in the magma chamber. This will only happen if critical supersaturation for homogenous crystallization is exceeded throughout the chamber; i.e. in magma that is cooling quickly. Slowly cooling magmas do not reach supersaturation in the inner portions of the magma chamber, and therefore minerals will only crystallize on the walls, roof, and floor (Martin, 1990). Density currents in magma, like currents in water, deposit layers of crystals in a sedimentary fashion. These currents may be caused by processes such as compositional convection and boundary layer fractionation (see below). Depositional cumulates are recognized by sedimentary features; e.g. erosional channels in the Nunarssuit complex (Parsons and Butterfield, 1981), and layers in the Hall Cove complex, Duke Island intrusion (Irvine, 1987). Nucleation diffusion Differing rates of nucleation may also cause zoning in plutons. Those minerals with the highest rates of nucleation (hereafter referred to as mineral A) will tend to crystallize first, drawing certain components out of the liquid. As nucleation rates generally outstrip diffusion rates, a zone of liquid will form that is depleted in the components involved in the nucleation of A. This limits further growth of mineral A and allows mineral B, which has a slower nucleation rate, to crystallize out of the remaining liquid constituents (Best, 1993). Eventually crystallization may proceed 116 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. far enough away from the initial mineral A layer, that the liquid at the crystallization front is no longer depleted in mineral A constituents. The process may then be repeated. Continued recurrences of this sequence of events would produce rhythmic layers such as those found in the Bushveld Complex of South Africa (Wager, 1959) and the Skaergaard Intrusion (McBirney and Noyes, 1979). Compositional convection/Boundary layer fractionation Fractionation produces a residual liquid that is less dense than the fractionated crystals and the original fluid. If the porosity of the crystal pile is high enough (>0.05; Kerr and Tait, 1986), then the light liquid may escape and be replaced by denser, more saturated fluid. Alternatively, the processes of melting of wall rock, crystallization of magma and absorption of volatiles at a boundary can make the residual liquid more dense than the rest of the magma. This will cause the liquid layer to sink along the boundary and collect at the floor of the magma chamber (McBirney, 1993). This density-driven compositional convection, or boundary layer fractionation, will continue to occur as long as the velocity of the fluids convecting out of the pile is greater than the velocity of the advancing solidification front (otherwise the entire pile will freeze). This mechanism has been used to account for the unusually high percentage of refractory components found in adcumulus piles (Kerr and Tait, 1986). If this process was operating in an intrusion, one might expect to see thin layers of dense, mafic minerals at the base and less dense, felsic mineral layers at the top of the chamber. The formation of Unit 10 of the Rhum 117 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Layered Intrusion has been attributed to compositional convection (Kerr and Tait, 1984 & 1987). Sawka, Chappell and Kistler (1989) call upon boundary layer fractionation initiated by a double diffusive gradient (differing diffusive rates of heat and compositional diffusion) to form the zoning seen in the Palisade Intrusive Suite in the Sierra Nevada. Double diffusive convection (i.e. thermogravitational diffusion) In crystallizing magmas, the thermal and compositional gradients may have opposing effects on the overall density gradient. Temperature decreases, and thus density increases towards the roof of the chamber. Conversely, there should be a density decrease going upwards in the chamber due to crystallization and the formation of less dense residual liquids that rise to the top of the magma chamber. The combination of these two opposing factors leads to the formation of a stratifiea chamber, with convection occurring within each layer and diffusion of heat and chemical components across the boundaries (Best, 1982). Layers formed in this manner can be of varying thickness, but by definition should be thick enough for convection to occur. If convection is vigorous enough there may be flow sorting and evidence of depositional cumulate layering. This process has been used to account for the formation of certain layers in the Skaergaard intrusion (McBirney and Noyes, 1979) and the Honningsvag Intrusive Suite (Robins et al., 1987). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Filter pressing (see text, p.55) Flow sorting When melts containing suspended crystals flow through an area with boundaries (i.e. in a dike or along the walls of a magma chamber) sorting of the crystals by size and concentration, can occur. There are three different forces caused by flow which can lead to this segregation (Komar, 1972a; Barriere, 1976). 1) Crystals flowing close to the wall will be repelled towards the center of the flow due to mechanical interaction with the boundary. This is known as the wall effect, and is only effective in the vicinity of the wall itself (Barriere, 1976). 2) The Magnus effect occurs when a crystal suspended in Poiseuille flow (flow with a parabolic velocity profile) is rotated due to the differential flow velocity across its length. The rotation of the crystal gives rise to a “ transverse force” that causes the crystal to move toward the center of the flow (Barriere, 1976). The Magnus effect is dominant in magmas with <2% crystals (Barriere 1981). 3) The Bagnold effect is the most important of the three flow sorting forces. It affects crystal populations, rather than working on the scale individual crystals. The Bagnold effect occurs in magma flowing through dikes with crystal concentrations greater than 13% (Barriere, 1976). Drag occurs at the walls of the dike, creating a velocity gradient. This in turn causes shear, with the shear magnitude being greatest at the margins where the velocity gradient is steepest, and the least towards the center of the dike where flow is more uniform (McBirney, 1993). The velocity gradient tends to cause more crystal-crystal interaction, which in turn produces a "dispersive shear pressure" (McBirney, 1993), or "grain 119 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dispersive pressure" (Komar, 1972a), that drives crystals out of the zone of maximum shear towards the center of the dike. Crystals will tend to remain concentrated near the center of the dike as the gradient is more uniform, and therefore fewer crystal-crystal interactions occur. Plug-like flow (i.e. flow which has a velocity profile that is convex outward in the direction of flow) is necessary for this phenomenon and can occur in psuedoplastic (non-Newtonian) fluids. It can also occur as a consequence of the interplay of grain interactions and the resulting grain dispersive pressure. If there is a high concentration of crystals and frequent crystal-crystal interactions then size sorting may also occur (Komar, 1972a). Layers formed by flow sorting should show symmetrical distributions of crystals in cross section. Thus symmetrical layer sets can be interpreted to represent individual dikes. The center of the layers should contain the greatest concentrations of phenocrysts, while smaller crystals can be found at the margins. Flow sorting has been called upon to account for layers in the Ewarara pluton, Central Australia (Goode, 1977) and the Ploumanc’h Subalkaline Granite of Brittany (Barriere, 1981), to name but a few. Barriere (1976) noted that the Bagnold effect is only effective in dikes that are less than 100 meters wide. Therefore, other mechanisms are necessary to produce crystal concentration and size sorting in dikes of greater dimensions. Differential cooling, differentiated injection of magma, convection and crystal settling (Barriere, 1976) are alternative mechanisms to produce such sorting, although the latter two processes would not produce symmetrical layer sets. 120 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Sintering This is an “aging” or “ coarsening" process that takes place after crystals have nucleated and accumulated. It has been modeled and observed in numerous experiments (e.g. Nabalek, 1978; Jurewicz and Bruse, 1985; Park and Hansen 1993; Means, 1994) Variations in particle size within a crystal assemblage can cause the pile to become unstable. Larger crystals grow at the expense of smaller crystals in order to reduce the total free surface energy of the system. The resulting layers are thin, with a width proportional to crystal grain size, and show coarse textures and secondary structures. There is an overall grain size gradation across the system. Textures are coarse and often exhibit secondary structures (e.g. “honeycomb patterns"; Nabalek, 1978). This process has been used to explain fine scale layering in the Banded Series of the Stillwater Complex (Boudreau, 1987) and the Rhum complex (Hunter, 1987). iii) SUBSOLIDUS PROCESSES Deformation processes that may lead to layer formation, operating in the solid state, are categorized as ductile or brittle. However, this classification is somewhat misleading, as deformation that looks ductile on the macroscopic scale, may be formed by brittle processes on the microscopic scale. Thus the transition from brittle to ductile behavior should be considered gradual (Williams et al., in press). On the macroscopic scale, whether brittle or ductile deformation takes place is dependent not only upon temperature, but also on pressure, 121 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. composition, water content and strain rate. Therefore, both brittle and ductile processes may operate at high and low temperatures. In general, however, ductile processes such as diffusive mass transfer (e.g. Nabarro- Herring and Coble diffusion) and crystal-plastic flow (e.g. dislocation glide and twinning) predominate at high temperatures (and low strain rates), whereas brittle processes, such as fracture and cataclasis, predominate at low temperatures (and high strain rates). Metasomatism (see text, p.54) Fluid-enhanced diffusion The presence of fluids along grain boundaries can also enhance grain-scale processes, such as grain boundary sliding. Layers formed by such processes will show little evidence of crystal plastic processes, suggesting that abundant water must have been present in order to assist grain boundary sliding (Williams, 1990). Such fluids would move up pressure gradients, weakening the pathways along which they travel and causing more slip along these channels. In addition, the fluids would dissolve, carry and redeposit components, such as quartz, as they moved upwards. Thus quartz-rich layers would mark channelways were fluids were oversaturated and depositing silica, and areas where fluids were dissolving quartz would becoming passively enriched in undissolved components (e.g. phyllosilicates). Textural coarsening in the Rhum complex (Hunter, 1987) and the Hermitage Massif (Gapais and Babarin, 1986) has been attributed to fluid-enhanced diffusion. 122 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Stress-induced diffusion Stress can also have a strong effect on metamorphic processes that lead to layer formation. If stress is unequally distributed in a body, metamorphic segregation of minerals into bands may occur. Chemical components will diffuse from areas under high pressure to areas under low pressure, in order to maintain the equilibrium chemical potential of the system (Robin, 1979). In addition, if there is an original compositional heterogeneity in the rock, such as body with alternating quartz-rich and mica-rich layers, then the competency and thus chemical potential of those layers will vary. The mica-rich layers would tend to have a higher chemical potential and therefore, any silica (which is very mobile) in the mica-rich layers, would migrate towards the quartz-rich layers. Thus new layers could form, or original layering would be intensified by this inter layer diffusion (Robin, 1979). Processes like that described above have been used to account for the formation of mylonite banding, which forms at high pressures (Sinha Roy, 1977, in Robin, 1979). Burg and Ponce De Leon (1985) proposed that pressure solution (i.e. stress-induced diffusion) was responsible for the removal of quartz, and consequent concentrations of phyllosiiicates in thin layers along C-planes in the Corcoesto granite of northwestern Spain. Transposition Rotation of both pre-existing layers (macroscopic) and minerals (microscopic) may also lead to layer formation. However, it is likely that most transposed layering forms as a result of a combination of physical 123 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rotation and ductile folding/flow (Best, 1982), with the new layers forming with their long-axes parallel to the X axis of the strain ellipsoid. B) LAYER TYPES AND INFERRED FORMATION PROCESSES Cumulus layering The first type of igneous layering extensively studied was cumulus layering. This is the process proposed for forming layers in such famous localities as the Bushveld, Stillwater, Muskox and Skaergaard intrusions. It involves the buildup of crystals due to fractional crystallization in a pre existing magma chamber. Initially the formation of cumulus layering was attributed to gravity settling of crystals in a largely liquid (and less dense) magma (Wager et al., 1960). However, in later studies, it became clear that “ cumulus” layers can form in several different manners, not all of which include gravity settling. For example, studies of plagioclase crystal density versus magma densities indicate that the density of such crystals is too low for them to have sunk. Instead they would have floated, thus implying that plagioclase “ cumulus" layers cannot have formed by the traditional crystal settling process (McBirney and Noyes 1979). In addition, "cumulate" layers occur along the walls and roof of intrusions like the Skaergaard (Best, 1982), and the Guadelupe Igneous Complex (Hauessler and Paterson, 1993), which would preclude their accumulation by sinking. Therefore, a “ cumulate” is now defined as “ an igneous rock characterized by a cumulus framework of touching mineral crystals or grains that were evidently formed and concentrated primarily through fractional crystallization," (Irvine, 1982). Fractional crystallization Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. is the genetic process for the formation of the cumulates. However, the manner of their buildup is no longer indicated by this new definition. Sub-classifications (listed below) have been erected in order to impart implications for the actual buildup of the crystals: Crystal settling is invoked for the formation of the following layer types. All imply that the magma contained enough melt that crystals could travel freely. This type of cumulus layering is most likely to be found in mafic intrusions as these magmas are the least viscous and therefore crystals can move through the magma with ease. Precipitation cumulates are those that form from crystals precipitating and settling out of a magma - the traditional “cumulate” process. Such layers should reflect the fractional crystallization of a magma as expected from Bowen's Reaction Series, with mafic minerals at the base of the chamber grading into felsic at the top. Flotation cumulates form as a result of less dense crystals collecting at the top of a magma chamber; plagioclase (which should form crystals less dense than the melts from which it crystallizes) accumulations are often accounted for in this manner. Depositional cumulates are layers formed by sedimentary processes within the magma chamber (e.g. minerals deposited by convection currents). Plagioclase-rich layers that are found at the bottom of magma chambers (when they are expected to be found at the top) interlayered with minerals of greater density are believed to be carried to the lower regions by such convection currents. Depositional layers may show graded bedding, cross-bedding and scouring. 125 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Other types of cumulates that can form in mafic or felsic plutons and do not involve transport of the crystal fractionates are those that form due to accretion (i.e. in situ). For example, crescumulates are delicate, branching cumulate crystals that grow outwards from the walls of the cooling chamber (Irvine, 1982). They resemble tendrils of frost on a cold window. Comb layers also grow in situ, on the walls of magma chambers or dikes. Elongate, bladed crystals oriented perpendicular to layering planes grow outward from the walls as the magma cools. The difference between crescumulates and comb layers is in their form (crescumulates branch, comb layers often interfinger). The preservation of the crystals at ninety degrees to the walls implies that the magma was static. Rhythmic layering Rhythmic layering is a descriptive term used to indicate the recurrence of distinctive layers or sequences. The type locale is the Skaergaard Intrusion (Wager and Brown, 1968), which exhibits macrorhythmic (1-5m thick) and microrhythmic (1-3 cm thick) layering (Irvine, 1987). Wager (1959), suggested that rhythmic layers in the Bushveld complex of South Africa were formed due to the differing powers of crystal nucleation of chromite, bronzite and bytownite. Alternatively, if the genesis of these units is due to the periodic influx of new magma pulses, then the layering is called cyclic (Irvine, 1982). In cyclic layering each sequence exhibits mineral trends (whether in the order of appearance, their modal ratios, or their individual mineral compositions) that are indicative of the parent magma's fractional crystallization path (Irvine, 1987). 126 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Cryptic layering Cryptic layering shows a gradational variation in solid-solution mineral composition (e.g. increases in Fe vs. Mg content of pyroxenes) in an igneous body. This type of layering can be found in the Skaergaard Intrusion (Wager and Brown, 1968) and in the Klokken roof series (Parsons, 1979) and is believed to be a result of fractional crystallization. Certain solid-solution phases may appear or disappear at certain points in an intrusion. This is known as phase layering. Schiieren Schlieren layers are discontinuous, thin, wispy, layers that have higher concentrations of mafic minerals than the surrounding rock. These layers may form by xenolith melting and disaggregation during flow (i.e. partial assimilation) (Pitcher and Berger, 1982; Hatch et al., 1972), or enclave disaggregation during flow (Frost and Mahood, 1987). In order for this to have occurred the enclaves must have had similar viscosities as the surrounding host. Alternatively, the schlieren layers may be native to the magma, and simply form from the concentration of mafic minerals in certain zones due to flow sorting (Barriere, 1981). In either case, the magma chamber must have been dynamic, with convection occurring. Pegmatite Pegmatite layers can occur as dikes with sharp intrusive boundaries, or as more diffuse layers or blotches of coarse-grained crystals. 127 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pegmatites are generally late-stage in the magma crystallization history, and commonly occur at the top of the magma chamber. Their coarse textures imply crystallization from hydrous magmas. Pegmatites with sharp boundaries may be formed by either diking, or the melting and resolidification of pre-existing igneous layers (Irvine, 1987). Diffuse blotches may represent late, low-density, fractionated fluids involved in processes like boundary layer fractionation. Sheets These are tabular igneous intrusions that may form by numerous processes (e.g. diking, magma mixing, double diffusive convection). Studies of contacts and to a lesser extent of internal features will help differentiate which of these processes are responsible in each individual case. Mylonites Mylonites are fine grained, streaked, aphanitic fabrics (Best, 1982, p.389). They are produced by ductile, solid-state flow such as that which occurs during stress-induced diffusion (Robin, 1979). 128 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. SAMPLE* MOO250a-D OSF-O OSB-D OSI2C-D MDG230-D MDG320-D MOG342-D 0882*0 OSA-D MDG205-L OSD-M OSD-L SI02 66.43 67.36 67.83 67.06 68.6 69.66 70.62 71.01 71.12 71.27 71.31 71.92 TI02 0.48 0.51 0.41 0.48 0.44 0.36 0.29 0.24 0.2 0.17 0.26 0.13 AI203 17.16 17.27 16.85 16.51 16.67 15.84 15.32 15.45 15.71 15.14 15.24 15.04 reo * 3.S 3.14 3.07 3.35 3.07 2.81 1.86 1.61 1.48 1.27 1.04 0.03 MoO 1.24 1.04 1 1.16 1.14 0.03 0.71 0.66 0.54 0.51 0.59 0.35 MnO 0.078 0.06 0.062 0.066 0.0040 0.047 0.045 0.042 0.033 0.036 0.035 0.024 cao 3.1 3.05 2.06 2.69 3.04 2.55 1.03 1.06 2.01 1.42 1.65 0.56 N»20 4.80 5.86 5.10 4.06 4.86 4.58 4.06 3.91 4.74 3.78 4.26 2.07 K20 2 1.6 1.3 1.71 1.56 2 3.81 3.53 2.33 4.5 3.53 6.78 P20S 0.10 0.22 0.11 0.10 0.14 0.17 0.11 0.00 0.06 0.05 0.05 0.05 LOI 0.60 0.70 1.06 0.76 1.10 0.86 0.68 0.07 1.00 1 0.70 1.26 TOTAL 00.758 100.9 00.842 00.856 100.7140 00.807 00.535 00.492 00.3 09.146 00.655 100 Fa* 0.730 0.751 0.764 0.743 0.720 0.752 0.723 0.705 0.732 0.713 0.766 0.728 A/CNK 1.085 1.021 1.1 1.108 1.006 1.108 1.071 1.110 1.124 1.107 1.101 1.136 KN 6.87 7.45 6.40 6.67 6.42 6.58 7.87 7.44 7.07 8.28 7.8 0.75 Pari* oaf million Rb 130 122 111 127 112 107 145 134 100 165 133 199 Sr 270 280 350 271 330 273 354 275 310 28B 232 317 Da 102 100 102 00 125 181 801 403 283 766 392 1240 Ga 26.1 26 27.6 25.8 23.4 23.9 19.7 18.6 21.1 18.7 20.4 15.8 Y 21.8 23.4 16 22,8 15.5 16.0 15.2 15.2 10.6 13.1 16.6 12.3 La 16.4 27 28.1 28.6 20.4 28.4 13.7 10.7 13.6 22.1 40.4 22 Zr 224 206 171 217 142 107 135 110 140 104 140 78.3 Nb 28.2 22 37 26.5 16.6 18 16.1 10.8 15.5 12.1 10 11.0 Cu 5.8 0.4 0.6 0.6 1.2 6.4 1.2 4.0 0 0.1 0.0 0 Zn 05.0 107.1 02.7 03.5 84.7 76.1 47.4 45.9 40.7 35.7 52.1 25 Cr 20.4 23 25.0 27.3 17.2 18.5 26.2 18.8 14.6 17.2 16.9 13.3 HI 0 1.7 2.7 2.7 0 0 2.0 0.2 0 0 0 0 Th 0.1 8.5 6.7 10.0 7.0 8.1 6.2 10.3 8.2 7.8 22.3 8.4 U 2.7 2 3.6 2.6 2 3.1 3.4 2.8 2 3.3 2.2 3.2 Pb 17.1 17.7 16.1 14.4 16.3 18.4 25.7 22.3 21.2 25 25.3 35 S 83.7 72.3 62.7 62.7 55.3 55.8 40.5 32.6 39.7 41.1 38 35.4 a 07l 05.0 106 07.7 100 80.8 6B.5 64.8 82.6 70.4 76.4 89 M to Appendix II: Geochemistry X-Ray Fractionation results for "regular layers." Sample numbers ending in "D" (e.g. OSF-D are the darker layers; sample numbers ending in "L" are the lighter layers; sample OSD-M is a sample layered on the millimeter scale and is a combination of light and dark layers. Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission. Appendix II. Geochemistry X-Ray Fractionation results for "regular layers" (continued) SAMPLE# M0O239-L OSF-L M0G334-L MDQ347-L MDG2SOa-L OSE-L SI02 72.2 72.60 73.13 73.21 73.4 73.4 T I0 2 0.23 0.15 0.015 0.14 0.03 0.2 A I203 14.08 14.05 14.74 14.30 14.35 14.65 FEO* 1.40 0.00 1.11 0.04 0.10 1.21 MoO 0.55 0.35 0.36 0.42 0.11 0.36 MnO 0.020 0.024 0.034 0.027 0.012 0.010 CaO 1.04 1.33 1.60 0.03 0.44 0.64 N i2 0 3.58 3.20 3.6 3.31 2.15 3.61 K20 4.00 5.40 4.30 6.66 0.14 6.13 P205 0.00 0.07 0.06 0.06 0.03 0.00 LOt 0.60 0.38 1.35 1.07 0.34 1.32 TOTAL 00.046 00.674 100.330 00.747 00.102 100.300 Fa* 0.73 0.742 0.756 0.604 0.612 0.77 A/CNK 1.001 1.001 1.00 1.141 1.002 1.10 KN 7.65 0.74 7.06 6.67 10.29 0.64 Part* par million Rb 131 155 154 170 214 163 8r 338 205 313 294 264 215 Ba 014 005 716 780 1047 050 Ca 17.4 10.7 17.4 10.2 13.6 16.6 Y 13.1 14.4 11.0 20.2 7.0 10.0 La 22.0 24.0 20.7 10.6 1.3 7.0 Zr 121 02 100 03.2 0.0 120 Nb 0.4 3 0.5 11.3 0 11.4 Cu 0.2 0.0 3.6 0 3.0 0.1 Zn 41.7 30.7 33.5 25.0 10.8 27.4 Cr 16.4 16.0 10.7 15.0 17.6 14.4 Nl 0 0 0 0 0 0 Th 0.5 9.5 0.4 11.0 0 2.2 U 2.2 2.6 2.6 2.0 3.5 2.7 Pb 25.0 33.2 23.2 30.3 37.6 22.7 S 30.7 31.7 41.0 33.3 33.7 54 a 71.7 76.2 65 05.5 00.0 102 CO o PLEA SE N O TE : Oversize maps and charts are filmed in sections in the following m anner; L E F T T O R IG H T, TO P TO BO TTO M , W ITH SM ALL OVERLAPS The following map or chart has been refilmed in its entirety at the end of this dissertation (not available on microfiche). A xerographic reproduction has been provided for paper copies and is inserted into the inside of the back cover. Black and white photographic prints (17" x 23") are available for an additional charge. UMI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72 64 ' 6 4 | 34 ' 6 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 'O Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 59 64 33/ 8 2 83 ,1 5 48 67 64, .40 35, 34] 8 3 : .7 4 83 £ 7 n reproduction prohibited Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 39. S O ' 88 90 69 60, 82 6 2« X 82 n ^ g r - 'Z & ' 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 88 ^ 31 47~ ~39^ ^ ! 82 50 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36 3 5 , 841 .74 40, 22 68 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Miles 2 Kilometer Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 82 1 Miles 2 Kilometers ns in the Main Donegal Granite e ciuion. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PLEASE NO TE: Oversize maps and charts are filmed in sections in the following manner: L EFT T O R IG H T, TO P TO BOTTOM , W ITH SM A L L OVERLAPS The following map or chart has been refilmed in its entirety at the end of this dissertation (not available on microfiche). A xerographic reproduction has been provided for paper copies and is inserted into the inside of the back cover. Black and white photographic prints (17" x 23") are available for an additional charge. UMI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 86 36 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8 8 ^ . * y a s * 4 »rcy Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 / / Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. \ o o 77, _ 66 78 78 X My* P I t 7 8 K < *y i f 'm- i 88 X / 7 ^ filsrL X*> G*l l s ^ r r f W t ' t a f y ? % . M 4 1 2 -^ . X * » 84, '80 72 £ E Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Miles Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 88 77 60’ Miles 0 1 i i Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78, Miles Kilom Plate 2. Map showing foliation orientations in the Main Donegal Granite. Located at the southwestern portion of the pluton. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Miles 2 Kilometers itions in the Main Donegal Granite Fthe pluton. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PLEASE N O T E : Oversize maps and charts are filmed in sections in the following manner: L E FT T O R IG H T, TO P TO BOTTOM , W ITH SM ALL OVERLAPS The following map or chart has been refilmed in its entirety at the end of this dissertation (not available on microfiche). A xerographic reproduction has been provided for paper copies and is inserted into the inside of the back cover. Black and white photographic prints (17” x 23”) are available for an additional charge. UMI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 5 6 Kilometers Plate 3. Compilation map showing foliations in the Main Donegal Granite are predominantly NE-SW striking and steeply dipping. Combined data compiled from Pitcher and Spencer (1972) and author's research (data shown with black foliations symbols from author, data shown with open symbols from Pitcher and Spencer). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PLEASE NOTE: Oversize maps and charts are filmed in sections in the following manner: L E FT T O R IG H T, T O P TO BO TTO M , W ITH SM ALL OVERLAPS The following m ap or chart has been refilmed in its entirety at the end of this dissertation (not available on microfiche). A xerographic reproduction has been provided for paper copies and is inserted into the inside of the back cover. Black and white photographic prints (17” x 23") are available for an additional charge. UMI Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission Z4- Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. y? Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Miles Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 Miles 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 43 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Miles 2 Kilome Plate 4. Map showing lineation orientations in the Main Donegal Granite. Located at the southwestern portion of the pluton. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 Miles 1 1 2 Kilometers Donegal Granite. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Structure and origin of echinoid beds, unique biogenic deposits in the stratigraphic record

PDF

The Triassic organic-rich flat clam biotope: A synthesis of paleoecological and climate modeling analyses

PDF

Paleoseismology, geomorphology, and fault interactions of the Raymond fault, Los Angeles County, California

PDF

Uptake and transport of desmopressin and its lipidized derivatives in Caco-2 cells

PDF

Investigating the influence of atmospheric changes on the variability of the North Pacific using paleoproxy data and a fully coupled GCM

PDF

Dynamic fluvial systems and gravel progradation in the Himalayan foreland

PDF

Large epifaunal bivalves from Mesozoic buildups of western North America

PDF

Investigation of arc processes: Relationships among deformation, magmatism, mountain building, and the role of crustal anisotropy in the evolution of the Peninsular Ranges Batholith, Baja California

PDF

Late Pleistocene ventilation history of the North Pacific as seen through delta carbon-13 records of the Southern California borderland

PDF

Groundwater modeling and geochemical tracer (CFC-12 and tritium) distribution in the Abalone Cove landslide, Palos Verdes, California

PDF

Determination of paleoearthquake age and slip per event data, and Late Pleistocene-Holocene slip rates on a blind-thrust fault: Application of a new methodology to the Puente Hills thrust fault,...

PDF

Late Pleistocene-Holocene depositional history of the California Continental Borderland

PDF

Active deformation at Canyonlands National Park: Distribution of displacements across the grabens using spaceborne geodesy

PDF

Magma mingling/mixing in a heterogeneous, multi-pulse magmatic system: An example from the Jackass Lakes pluton, central Sierra Nevada batholith

PDF

Changing characteristics of deformation, sedimentation, and magmatism as a result of island arc -continent collision

PDF

Systematic analysis of crustal anisotropy and attenuation using seismic data associated with the 1999 Chi-Chi, Taiwan, earthquake

PDF

The feminized city: Reading Wang Anyi's "Ballad of Eternal Sorrow"

PDF

Barnacles as mudstickers? The paleobiology, paleoecology, and stratigraphic significance of Tamiosoma gregaria in the Pancho Rico Formation, Salinas Valley, California

PDF

Variable reverberation in performance spaces: Passive and active acoustic systems

PDF

Seafloor precipitates and carbon-isotope stratigraphy from the neoproterozoic Scout Mountain member of the Pocatello Formation, Southeast Idaho: Implications for neoproterozoic Earth history

Asset Metadata

Creator Yuan, Elizabeth Semele (author)

Core Title Magmatic foliations and layering: Implications for process in magma chambers

Degree Master of Science

Degree Program Earth Sciences

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Geology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-7104

Unique identifier UC11341068

Identifier 1381613.pdf (filename),usctheses-c16-7104 (legacy record id)

Legacy Identifier 1381613.pdf

Dmrecord 7104

Document Type Thesis

Rights Yuan, Elizabeth Semele

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA