Atlas of
Migmatites
About the cover: Front cover Metatex1te m1gmatite developed from a foliated granodionte show1ng ...
435 downloads
2218 Views
381MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Atlas of
Migmatites
About the cover: Front cover Metatex1te m1gmatite developed from a foliated granodionte show1ng leucosome 1n dilat1on structures. Laag Mountain area, Bntish Columb1a. Photograph by Paul McNeill. Rear cover D1atex1te m1gmat1te w1th schlieren, developed from a pelitic protolith. Quetico Subprov1nce. Ontano.
The Mineralogical Association of Canada gratefully acknowledges the financial contribution of rhe following organizations: Ca nadian Geological Foundation Natural Resources Canada
@
Atlas of
Migmatites Edward W. Sawy er
The Canad ian Miner alogis t Special Publica tion Mln er a logi c.aJ
Ass o ciatio n o f Can a d a
Association
min~ralogique
du canada
ritC·Cf itC NRC Research Press
9
© 2008 National Research Cou ncil of Canada All rights reser ved. o part of this publication m ay be reproduced in a retrieval system, or t ra nsmi tted by any m ea ns, electronic, mechanical, photocopyi ng, recording or otherwise, without the prior w rillen permission of the ational Resea rch Council of Ca nada, Ottawa, Ontario KIA OR6, Ca nad a. Printed in Ca nada on acid -free paper.@ ISBN 978-0-660- 19787-6 ISS 17 17-6387 N RC o. 46331
Library and Archives Canada Cataloguing in Publication Sawyer, Edward W illiam, 1951Atlas of migm atites/ E.W. Sawyer " RC Resea rch Press" Issued by: Nationa l Research Cou nci l Ca nada Co-published by Mineralogical Association of Ca nada Includes bibliographical references ISB 978-0-660-19787-6 l. M igmatite. 2. Migmatite - Pictorial works. 3. Pet rology. I. ationa l Research Council Canada II. Mineralogical Association of Canad a
Ill. Tit le. Q£475.M5S38 2008
552'.4
C2008-980022-2
NRC Monograph Publishing Program
Editor: P.B. Cavers (University of Western Onta rio) Editoria l Board: W.G.E. Caldwell, OC, FRSC (University of Western Ontario); M.E. Cannon, FCAE, FRSC (University of Ca lga ry); K.G. Davey, OC, FRSC (York Universi ty); M.M. Ferguson (University of Guelph); S. Gubins (Annual Reviews); B.K. Ha ll, FRSC (Dalhousie Un iversity); W.H. Lewis (Washington University); A.W. May, OC (Memor ial University of ewfound land); B.P. Dancik, Editor-in-Chief, RC Research Press (University of Alberta) Inquiries: Monograph Publishing Program, RC Research Press, National Research Council of Can ada, Ottawa, O ntario K IA OR6, Canada. Web site: http://pubs.nrc-cnrc.gc.ca Correct citation for this publication: Sawyer, E.W. 2008. Atlas of Migmatites. The Canadian Minera logist, Special Publication 9. N RC Research Press, Ottawa, On tario, Ca nada. 371 p.
Also in this series Encyclopedi a of Mineral Names W H . Blackburn & W H . De nnen Special Publication I ( 1997)
Glossary of Mineral Synonym s J. de Foures tier Special Publication 2 ( 1999)
Atlas of Micromorphology of Mineral Alteration and W eathering J. Delvigne Special Publication 3 ( 1998)
N ew Minerals 1995-1999 (2001) J.A. Mandarino Special Publication 4 (200 I)
The H ealth Effects of C hrysotile Asbestos: Contributio n of Science to Risk-Managem ent Decisions R.P. N o lan, A.M. Langer, M. Ross, F.J. Wick & R.F. Martin, eJ s. Special Publication 5 (200 I)
Mineral Species Discovered in Canada and Species Named after Canadians Las: l6 Ho rvath Special Publication 6 (2003)
Mineral Species First D escribed from Greenland O le V. Petersen and O le Johnsen Special Publication 8 (2005)
Available from: Mineralogical Association of Canada 490, rue de Ia Couronne Q uebec, Q C G l K 9A9 Canada www.mincralogicalassociatio n.ca
Editor, T he Canadian Mineralogist Robert E Martin
Table of Contents Preamble Preface
.................................................................................................................................................................................................................... xiii
........................................................................................................................................................................................................................... xiv
Acknowledgements Introduction
................................................................................................................................................................................... xv
............................................................................................................................................................................................................... 1
1. The scope of this atlas .................................................................................................................................................................................. 1 2. General terminology and definitions needed for work on migmatites ........................................ 2 2.1 The heritage of migmatite terminology ........................................................................................................................................... 2 2.2 A definition of migmatite .......................................................................................................................................................................... 3 2.3 Descriptive terms and definitions for the principal parts of a migmatite .................................................................... 4 Terms specific to the neosome ...................................................................................................................................................... 5 Terms for the other parts of a migmatite ................................................................................................................................. 7
3. Migmatites: the processes and morphologies ............................................................................................................... 8 3.1 The first-order morphological division of migmatites .............................................................................................................. 8 3.2 Temperatures, degree of partial melting, and fraction of melt .......................................................................................... 9 3.3 The partial-melting process ...................................................................................................................................................................... 9 3.4 A special case: melting under lithostatic stress conditions (so-called “static melting’) .................................... 10 3.5 The general case: melting under differential stress (so-called “dynamic melting”) ........................................... 10 3.6 Definitions of metatexite and diatexite .......................................................................................................................................... 12 3.7 The second-order morphological divisions in migmatites ................................................................................................... 13 Morphologies characteristic of metatexite migmatites ................................................................................................. 14 Patch migmatites ........................................................................................................................................................... 14 Dilation-structured migmatites ............................................................................................................................ 15 Net-structured migmatites ..................................................................................................................................... 15 Stromatic or layer-structured migmatites ..................................................................................................... 16 Transposition and the morphology of metatexite migmatites .................................................................... 17 Morphologies characteristic of diatexite migmatites ...................................................................................................... 17 Schollen or raft-structured migmatites ........................................................................................................... 17 Schlieric migmatites .................................................................................................................................................... 18 Diatexite migmatites ................................................................................................................................................. 18 High strain and the morphology of diatexite migmatites ............................................................................... 19
3.8 Migmatite morphologies outside the metatexite–diatexite division ............................................................................ 19 Fold-structured migmatites .................................................................................................................................... 19 Vein-structured migmatites ................................................................................................................................... 20 3.9 Descriptive terms that should be abandoned ............................................................................................................................ 20 Bedded migmatites ..................................................................................................................................................... 20 Agmatite ............................................................................................................................................................................ 20 Ptygmatic migmatites ................................................................................................................................................ 20 Ophthalmite migmatites ......................................................................................................................................... 20
4. Metasomatism and migmatites ..................................................................................................................................................... 21 4.1 Influx of aqueous fluid into hot rocks causing partial melting ........................................................................................... 21 Large-scale influx of fluid ................................................................................................................................................................. 21 Small-scale influx of fluid ................................................................................................................................................................. 22 4.2 Metasomatism and changes in the fertility of rocks ............................................................................................................... 22 4.3 Morphology of migmatites affected by infiltration metasomatism ............................................................................... 23
5. Microstructures in migmatites ...................................................................................................................................................... 23 5.1 Mineral paragenesis ..................................................................................................................................................................................... 23 5.2 Quantitative analysis .................................................................................................................................................................................. 24 The grain-contact method ............................................................................................................................................................. 24 Crystal-size distributions ................................................................................................................................................................. 25 Studies of grain size, aspect ratio, and orientation ......................................................................................................... 25 5.3 Diagnostic microstructures in migmatites .................................................................................................................................... 26 Microstructures produced in partial-melting experiments .........................................................................................26 Microstructures in the residual rocks, and evidence for partial melting ............................................................ 27 Microstructures in the melt-rich parts of migmatites; evidence for crystallization of the melt .......... 28 Magmatic and submagmatic foliations ..................................................................................................................................... 29 Melt inclusions ........................................................................................................................................................................................ 29 Cordierite–, garnet–, and orthopyroxene–quartz intergrowth microstructures ...................................... 30 Symplectitic intergrowths of quartz and plagioclase with mica ............................................................................... 31 Composition and zoning of plagioclase ................................................................................................................................... 31 Biotite composition and microstructures .............................................................................................................................. 32 Contact between leucosome and melanosome in metatexite migmatites .................................................... 33 Microstructure of schlieren in diatexite migmatites ....................................................................................................... 33 Microstructure of biotite-rich selvedges in migmatites ................................................................................................ 33
6. Whole-rock geochemistry in migmatite studies .................................................................................................... 34 6.1 A possible sequence of processes and some relevant questions .................................................................................. 34 6.2 Reference-point compositions ............................................................................................................................................................ 36 Determining protolith compositions (the starting material) ..................................................................................... 37 Determining the “melt” composition ...................................................................................................................................... 37 Residual rocks ........................................................................................................................................................................................ 40 Mineral compositions ......................................................................................................................................................................... 41 6.3 Diagrammatic representation ............................................................................................................................................................... 42 Matched triplet sets of samples .................................................................................................................................................. 43 General sets of samples ................................................................................................................................................................... 43
7. Migmatite-like rocks ..................................................................................................................................................................................... 47 7.1 Rocks formed by subsolidus segregation ....................................................................................................................................... 48 7.2 Models for the process of subsolidus segregation ................................................................................................................... 48 7.3 P–T conditions at which subsolidus segregation occurs ....................................................................................................... 48 7.4 The relationship between subsolidus segregation and migmatites ............................................................................... 49 7.5 Small-scale features of subsolidus segregations ......................................................................................................................... 49 The constituent parts ........................................................................................................................................................................ 49 Mineralogy of subsolidus segregation ...................................................................................................................................... 50 Microstructure ....................................................................................................................................................................................... 50 7.6 Outcrop-scale morphology ................................................................................................................................................................... 50 Stromatic, or layered, subsolidus segregations .......................................................................................... 50 Dilatant structures ....................................................................................................................................................... 51 Fleck structures ............................................................................................................................................................. 51 7.7 Rocks formed in syntectonic plutons and plutonic complexes ........................................................................................ 51 Syntectonic injection of magma ................................................................................................................................................... 51 Syntectonic crystallization of felsic plutonic rocks ........................................................................................................... 52 7.8 Vein complexes ............................................................................................................................................................................................... 52 7.9 Rocks formed in syntectonic plutonic and vein complexes compared with migmatites ................................. 53 Similarities ................................................................................................................................................................................................. 53 Differences ............................................................................................................................................................................................... 53
8. Working with migmatites .................................................................................................................................................................... 53 8.1 First-level map units .................................................................................................................................................................................... 54 8.2 Second-level map units ............................................................................................................................................................................ 54 8.3 Other considerations for mapping migmatites ........................................................................................................................... 55
9. Appendices .............................................................................................................................................................................................................. 56 9.1 Checklist of observations for each outcrop of migmatites ................................................................................................. 56 Observations on the neosome and paleosome ............................................................................................................... 56 Petrological observations in the study of migmatites ..................................................................................................... 57 Structural observations in the study of migmatites ......................................................................................................... 57 Way-up criteria in migmatites ....................................................................................................................................................... 57 Sampling of migmatites ...................................................................................................................................................................... 57 9.2 Glossary .............................................................................................................................................................................................................. 58
10. References .............................................................................................................................................................................................................. 62
The photographs
............................................................................................................................................................................................. 79
A. Some examples of migmatites ..................................................................................................................................................... 79 B. The parts of a migmatite ..................................................................................................................................................................... 83 Neosome and paleosome .................................................................................................................................................. 85 Neosome with leucosome and melanosome ........................................................................................................ 89 Neosome without distinct leucosome or melanosome ................................................................................. 99 Neosome in open-system migmatites ....................................................................................................................... 111 Variations within neosome .............................................................................................................................................. 120 From leucosome to leucocratic dikes in migmatites ....................................................................................... 128 Selvedges in migmatites ..................................................................................................................................................... 136
C. Metatexite and diatexite, the first-order division of migmatites ................................................... 142 Migmatites from the contact aureole of the Ballachulish Igneous Complex .................................... 144 Migmatites from the contact aureole of the Duluth Igneous Complex ............................................. 148 Upper amphibolite facies, regional migmatites from Saint-Malo, France .......................................... 150 Upper amphibolite facies, regional migmatites from the Opatica Subprovince, Quebec ....... 152 Granulite-facies, regional migmatites from the Ashuanipi Subprovince, Quebec ........................ 154
D. Second-order morphologies in migmatites ............................................................................................................... 156 The start of partial melting ............................................................................................................................................. 159 Metatexite migmatites with a patch structure .................................................................................................... 165 Metatexite migmatites with a nebulitic structure ............................................................................................. 169 Metatexite migmatites with leucosome in dilatant structures ................................................................... 171 Metatexite migmatites with a net structure .......................................................................................................... 181 Metatexite migmatites with a layered or stromatic structure associated with low strain ..... 190 Metatexite migmatites with layered or stromatic structure due to transposition ....................... 194 The transition from metatexite to diatexite migmatites .............................................................................. 201 Diatexite migmatites with schollen and with schlieren structures ......................................................... 207 Diatexite migmatites with schlieren structures .................................................................................................. 213 Diatexite migmatites ............................................................................................................................................................ 217 Diatexite migmatites at high strains .......................................................................................................................... 223
E. Other morphologies of migmatite ........................................................................................................................................ 225 Syn-anatectic folding: fold structures in migmatites ....................................................................................... 226 Migmatites with a vein structure ............................................................................................................................... 236
F. Microstructures characteristic of migmatites ......................................................................................................... 242 Results from quenched deformation-melting experiments: a starting point .................................. 246 Subsurface contact-aureoles: the Glenmore plug, Scotland ..................................................................... 250 Erupted, partially melted xenoliths: El Joyazo, Spain ....................................................................................... 252 Subsurface contact-aureoles: the Rum Igneous Complex, Scotland ................................................... 256 Shallow contact-aureoles: the Traigh Bhàn na Sgùrra sill on Mull, Scotland .................................... 258 Shallow- to medium-depth contact-aureoles: the Duluth Igneous Complex ................................. 262 Deeper contact-aureoles: the Ballachulish Igneous Complex, Scotland ............................................. 274 Regional migmatite terranes: the Ashuanipi Subprovince .......................................................................... 278 Regional migmatite terranes: the Opatica Subprovince ............................................................................... 282 Microstructures in residual rocks ............................................................................................................................. 284 Crystallization-induced microstructures in the melt-derived parts of migmatites: leucosome and leucocratic veins ........................................................................................................................... 292 Crystallization-induced microstructures in the melt-rich parts of migmatites: diatexite migmatites ...................................................................................................................................................... 305 Microstructures formed by flow in diatexite migmatites ............................................................................. 312 Inclusions of melt quenched to glass in minerals ............................................................................................... 316 Cordierite–, garnet–, and orthopyroxene–quartz intergrowth microstructures ....................... 320 Biotite–quartz and biotite–plagioclase symplectitic intergrowth microstructures ..................... 324 Biotite–sillimanite and biotite aggregates replacing garnet or cordierite .......................................... 328 Plagioclase .................................................................................................................................................................................. 329 Contact between the leucosome and melanosome in metatexite migmatites ............................ 330 Microstructure of schlieren in diatexite migmatites ........................................................................................ 343 Microstructure of biotite-rich selvedges in migmatites ................................................................................. 347
G. Migmatite-like rocks ................................................................................................................................................................................ 351 Layer-parallel or stromatic subsolidus segregations ........................................................................................ 352 Fleck segregations ................................................................................................................................................................ 354 Syntectonic plutons: rocks that resemble metatexite migmatites .......................................................... 356 Syntectonic plutons: rocks that resemble diatexite migmatites ............................................................... 362 Arrays of felsic veins that look like migmatites .................................................................................................... 366
Ada~
of MlgiTl.HHcs
- - - - - - - - - - - - - - - - - - - - - xiii
Pre am ble
It is with pleasure that I mtroduce the latest addition to the list of Spec1al Publications produced by the M1neralog1cal Association of Canada. This Atlas of Migmatites contributes d1rectly to the educational mandate of our Assoc1at1on. The book deals w1th migmatites, a very widespread group of rocks, especially common in the Archean crust that makes up a major proportio n of our country. Everyone has learnt a b1t about m1gmat1tes, but by and large, the subject matter "falls between the cracks" in our university curricula. Typ1cally, an upper-level undergraduate course in metamorphic petrology does set the stage for an understanding of anatect1c react1ons in terranes that have undergone metamorphism at conditions of upper amphibolite or granulite faoes. Students 1n such a course do learn about the concept of anatex1s and dehydration-Induced melt 1ng, but really not much about the extraordinarily complex array of products of such anatex1s, the 1mportance of fract1onal crystallization of the anatectic liquid, and the fate of assemblages of residual m1nerals. Furthermore, contact-related anatexis usually is not discussed. On the other hand, students in an upper-level course 1n igneous petrology do deal w1th products of partial melting in the crust, and can speak at length about the physical propert1es of a silicate magma and the results of its fractional crystallization. But they deal w1th ready-made plutons. Such students do not develop a good understanding of the steps that precede the formation of a pluton, where small batches of leucosome coalesce and rise through a deforming mass of neosome and resister litholog1es. As IS made clear below, Professor Sawyer has remed1ed the situat1on by writing the first authoritative treatise about the petrology of m1gmatites s1nce the work of Karl Richard Mehnert ( 1913 1996), of Berlin, published 40 years ago. He has abandoned the purely descript1ve approach of Mehnert 1n favor of an openly genet1c approach: a m1gmat1te 1s a rock that is the product of partial melting. If melting can be shown not to have taken place, the rock IS simply not a mlgmatite. Furthermore, he has recogn1zed the importance of
anatectic reactions not only 1n amphibolite-facies rocks, but also in the realm of the granul1te facies. He has exam1ned key occurrences of m1gmatites throughou t the world, and presents here 1n a systematic way the results of his scrutiny on all scales. Thus, the reader IS shown photographs of key exposures of m1gmat1tes 1n the field and petrographic details in 273 photographs, each with a deta1led capt1on, supplied by the author m 68% of the cases, the rema1nder being contributed by respected colleagues also active in the charactenzat1on of m1gmat1t1c assemblages. F1eld examples are presented from 12 countries, with the author personally involved 1n field studies 1n s1x of those. In th1s atlas, the author emphasizes the latest contributions in the area of migmat1te-related research, e1ther 1n a reg1onal settmg or 1n a contact aureole. Society-run, not-for-pr ofit organizations like the Mineralogical Assoc1ation of Canada find 1t very challenging to undertake such a publicat1on, ow1ng to the h1gh costs of product1on. In this 1nstance, the Assooat1on is most fortunate to have coproduced the volume w1th NRC Research Press. On behalf of the MAC. I thank Suzanne Kettley, Mark Bo1leau, and D1ane Candler for the1r valuable 1nput and their key roles in creating this book. In addit1on, I acknowledge the Involvement of P1errette Tremblay, Dwector of Publications of the Association, in ensuring close commumcatlons w1th N RC Research Press throughou t the preparation of this book. A lso, P1errette applied for and obta1ned funds from the Canad1an Geolog1cal Foundation and Natural Resources Canada, part of which paid for the pnnt1ng of a poster to publicize th1s book. I thank Vicki Loschiavo, who entered the ed1torial corrections on the master files of text and capt1ons as I progressed through the book. I am grateful to have had this opportun ity to help bnng th1s major contribut1on in the area of m1gmat1te research to fruition.
Robert F. Martin Editor, The Canadian Mineralogist
xiv - - - - - - - - - - - - - - - - - - - - -
Preface
Migmatites are some of the most confus1ng-look1ng, yet aesthet1cally pleas1ng rocks. The most stnk1ng cons1st of light-colored quartzofeldspath1c segregat1ons 1n a darker host, w1th the light segregat1ons show1ng a d1verse and often spectacular range of appearances, 1n some cases in swirls, 1n some others cross-cutt1ng, and in yet others discrete, or diffuse, or folded, or conta1n1ng high-grade metamorphic minerals like garnet and pyroxene. Migmatites justify their age-old name; they are "mixed rocks." Migmatites commonly look "squishy." This is no coincidence, for they are interpreted to be rocks frozen in the act of part1ally melting. A mixture of melt, disaggregated minerals, and unmelted rock is mechanically highly heterogeneous. It 1S no wonder that m1gmat1tes show such a w1de range of textures and structures. M1gmat1tes are of crucial 1mportance 1n understanding the genes1s of the huge volumes of gramt1c magma that are found 1n batholiths and that make up much of the continental crust. Migmatites are widely cons1dered to be examples of assemblages where such magma was caught in the act of being generated and escaping. M1gmatites thus furnish a un1que perspective on a fundamental process 1n the evolution of the Earth.
The arnval of this Atlas of Migmatites IS very timely. It 1s the first book devoted to m1gmatltes s1nce Mehnert's 1conic 1968 text. In the 1nterven1ng 40 years. dramatic strides have been made in the understand1ng of these complex rocks. Ed Sawyer is without quest1on the world's leading expert on migmatit1c textures and structures. In th1s atlas, he provides a wide range of superb field and thin -section photographs of migmatites from around the world, both his own and those of others, each carefully described and interpreted. The photographic part of the atlas is preceded by a substantial Introductory section in which he clarifies the dizzying range of descnptive terms for migmatites, both new and old. Although migmat1tes w1ll rema1n among the most complex of rocks to understand, th1s atlas goes a long way to mak1ng them more understandable.
David R.M. Pattison Univers1ty of Calgary
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - XV
Ackn owled geme nts
I would like to start by expresstng my gratttude to Mtke Brown and Richard Whtte for thetr encouragement and thetr reviews of the text. They pointed out my errors, omissions, and tracts of muddled wnting; the changes that they suggested have resulted tn a far better text. The rematntng faults are enttrely mine. Many people took the time to send me thetr photographs of migmatites for this atlas. The choice of which ones to use was far from easy, but each photograph, whether tncluded or not, has had an influence. I thank everyone who sent me a photograph or electronic image for his/her contribution. My perceptton of m tgmatttes and what happened wtthtn them has been influenced by seeing and dtscussing mtgmatites with many col leagues from around the world. To have seen such a wide range of mtgmatttes has been tnvaluable, and I am grateful for the hospitality and generosity of all t hose who have shown me their field areas. In particular, and more or less in chronologtcal order: Ned Chown (Grenvtlle Front), Ron Vernon and Bill Collins (Wuluma and Mount Hay), Dante! Lamothe and Alatn Leclair (Ashuanipi Subprovince), Mark Severson and Steve Hauck (aureole of the Duluth Igneous Complex), Mike Brown (Satnt-Malo), Gary Solar (Matne), Abdelali Moukhstl (Eastmain area), Geoff Clarke and Richard White (Mount Stafford, Broken Hill and Wuluma), Fernando Bea (Pena
Negra), Oltvier Vanderhaeghe (Masstf Central), and Dtrk van Reenen (Ltmpopo Belt). There has always been an easy exchange of t houghts and views wit hin the migmatite community, and next I would like to thank Marian Holness, Tracy Rushmer, Bernardo Cesare, Roger Powell, Alfons Berger, and Claudio Rosenberg for illumtnating some of the connections bet ween what one can see in migmatites and the various processes that have happened to them. In most endeavors, progress comes through sustained and focused effort, together with some serendipit y, of course. My work on migmatites, and ultimately this atlas, have been no different and would not have been possible without conttnutty of fund ing. Therefore, I am grateful for the philosophy behind the discovery grant program of the Natural SCiences and Engineenng Research Council of Canada, which funds long-term, cunostty-driven research programs. Finally, I thank Pierrette Tremblay and Robert F. Marttn for the tnvttation to write a book on migmat ttes and for thetr encouragement throughout, and to Robert for all his hard work tn reading and editing.
Edward W . Sawyer Universit e du Quebec
aChicoutimi
Int ro du cti on
I
I
TH E SCOPE O F T H IS ATLA S Migmatites are spectacular. complex-looking rocks that can insp1re, fasc1nate. or confuse geolog1sts. All migmatites v1ewed in an outcrop represent the sum of a senes of processes t hat acted 1n parallel. or sequentially. and the 1nfluence of vanous local factors (see the photogr aphs in section A) . In order to begin to understand the complexity and seem1ngly endless vanety in m1gmat it es, and to prov1de the reader context for the subJect of this book, some of the key factors and processes that make individual migmatites the way they are should be outlined at the outset. First. t here IS the pet rolog1cal process of par tial melt1ng. If the tempera ture becomes sufficiently high (above about 650°C), rocks may beg1n to part1ally melt. Some rocks have compos1t1ons that will produce more melt than others at a given temperature, a propert y of the rock called fertilIt y. Thus. one can Immediately see that comple xity beg1ns at this stage. because some rocks 1n a sequence will make more melt than others; some w ill not melt. The highest tempera ture atta1ned also has an effect. If the tempera ture only just surpasses the solidus. the m1gmat1te will conta1n a few small patches of melt scattered about in t he most fertile rock. If the max1mum temperature was. say. another 250° h1gher. then melting in the fertile layers might be pervasive, and well advanced 1n other less fertile rocks. Thus, the two m1gmatites would look completely different. even if generat ed from exactly the same sequence of rocks. Second. the nature of t he proto lith has an influence too. For example, the m1gmatite produced from a sequence of rocks that contains a single t h1n fertile layer (say a pelite) enclosed in a t h1ck sequence of 1nfertile rocks (say quartzite) will look completely different from one generated from a sequence that conta1ns a thin quartzit e in a thick sequence of pelites. Similarly. a m1gmatite generated from an 1sotrop1c rock (e.g.. a gran1te) wdl not look the same as one generated from a sequence of th1n ly bedded shales and sandstones.
Th1rd. deformation dunng part1al melt1ng has a maJor effect on the morpho logy of migmatites. Only rarely (e.g.. in some contact aureoles) do migmatites form under condit1ons where t here is essentially no deformation. Generally. migmatites are deformed w h1le they cont ain melt . Deform at ion of heterogeneous rocks produces vanations 1n the different ial st ress from place to place, and this sets up pressure gradients that dnve the movement of matenal. T here are likely to be vanations 1n viscos1ty or competence from rock to rock 1n the protolit h. but once partial melt ing of t he JUxtaposed rocks starts. much larger vanat1ons are cre ated. The part 1al melt has a viscos1ty that is much lower than that of nonmelted rocks; t herefore, dunng deformation, the melt 1n the rock moves more rapidly down pressure gradients t han the solid fract1on. and collect s 1n lowpressure s1tes. D eformat1on thus creates petrolog1cal diversit y. because t he melt is separated, or segregated. from the solids. which enhances the mechan1cal ext remes 1n the m1gmat1te. Deformat1on becomes concentrated, or partitioned. int o the weakest places; these are located where there 1s melt. Consequently. stra1n becomes more heterogeneous. and the m1gmatltes develop more and more complex and deforme d geometnes. Finally. t he penod of time t hat a migmatite has had to form also exerts an 1nfluence. The heat1ng and cooling cycle in contact aureoles is rapid (days to tens of thousands of years). such that melt may quench to glass or crystallize to granophyre, and pnmary microstructures, such as the shapes of the pores that t he melt occupied. are preserved because recrystallization did not reset microstructures. Strains are small. and deformation 1s restricted to small cracks and shear zones, so that the 1n1t1al geometncal relat1ons 1n the m1gmatite are generally preserved. In contrast. t he slow heat1ng and cooling (millions to tens of m1llions of years) of reg1onal metamorphic terranes mean that melts crystall1ze slowly. and subsequent recrystallization extens1vely modifies , or even eliminates, the pnmary microstructures. Moreover. large stra1ns can accumulate over long penods of t1me. so that the orig1nal geomet ry of the m1gmat1te 1s extensively folded. boudinaged, and transposed.
INTRODU CT IO N
2 ----------------------------------
Th1s book conta1ns two parts. One IS a summary of the advances 1n the understanding of m1gmat1tes and related rocks, including partially melted xenoliths of crustal rocks, that have been made s1nce Mehnert's ( 1968) influential book. The other consists of seven senes of photographs, each with a substantial explanatory caption, that Illustrate the aspects of migmat1tes covered 1n the first part. The two parts are complementary, and in the text, I refer to the 1llustrat1ons throughout, but the two parts could also be read separately, with the text part as a genenc discusSIOn of migmatites, and the Illustrated part as a senes of
2. GENERAL TERM INOLOGY AN D D EFINITIONS N EED ED FOR WO RK ON MIGMATITES
2.1 The heritage of migmatite terminology
The debate over what m1gmat1tes represent and the processes involved 1n the1r formation has been closely connected with the prevailing v1ews of how high-grade reg1onal Sect1on 2, on th1s page. beg1ns with the term1nology assometamorphism occurs. and how gramtes and gran1tic oated w1th m1gmat1tes; the various constituent parts of magmas are formed. Hutton (1795) believed that partial m1gmatltes are dealt with first, then the divers1ty 1n outcrop melt1ng changed pelit1c sed1ments and crystalline sch1sts appearance as a function of deformation, melt fract1on. and 1nto gne1sses. Lyell ( 1855) went a step further and considparent rock-type are descnbed. The terms necessary for ered that granites form where the part1al-meltmg process the study and descript1on of m1gmat1tes are defined; text was stronger and more complete. However, the present 1S shown in bold type to 1ntroduce a term where 1t is first terminology d1d not start to develop until later. One of defined, whereas text in italics is used to ind1cate a new the first terms specific to crustal melting is diatexis, introdefinition. Next, the processes that control the morphologduced by Gurich (1905) to describe cases where "partial" ICal d1vers1ty of migmatites are considered. so as to denve melting is complete, or where 1t has occurred throughout a a class1ficat1on scheme that 1s useful as a mapp1ng tool, and rock, to lead to the format1on of granite. Sederholm ( 1907) as a means for understand1ng why migmat1tes have a parmade the obv1ous comment that 1n most cases where a ticular morphology. A two-t1ered scheme of class1ficat1on term is requwed to explain the process of mak1ng gran1te. 1s proposed. All parts of migmatites and types of m1gmafus1on was not complete; he Introduced the term anat1tes ment1oned sect1on 2 are systematically Illustrated 1n texis, meamng melting, or remelting. of rocks, to cover the sect1ons B. C. D, and E. The weight g1ven to the Imporent1re range from 1nc1p1ent to complete melt1ng. tance of metasomat1sm and the 1ngress of H 0-nch flu1ds 1n the format1on of m1gmat1tes has changed considerably In the late mneteenth and early twent1eth centuries. maps1nce m1gmat1tes were first defined; current results and ping in the Precambrian sh1elds of Fennoscandia and North th1nk1ng are summanzed 1n sect1on 4. The interpretation of Amenca revealed many places where metasedimentary microstructures has always been central to metamorphic rocks appeared to pass 1nto gran1te. These transitional petrology and to the study of m1gmat1tes 1n part1cular. In zones attracted a great deal of mterest. Not surprisingly, recent years, the repertoire of microstructures that can be there appeared 1n the literature many opinions as to what used to identify a specific process has grown cons1derably, processes were involved; each hypothesis reflected an Indiand 1ncludes some microstructures derived from exper- vidual's own field experience. Iments as well as from natural rocks; both are discussed One school of thought. wh1ch saw Intrusions of gran1t1c 1n sect1on 5 and shown 1n sect1on F. Because the acqu1s1magma as the source of heat for the transformations. t1on of chem1cal compositions of whole rocks and m1nerals 1s now a rout1ne part of reg1onal mappmg. a bnef d1scus- developed from work done at the edges of granite plutons by Barro1s (1884). Lacroix (1898, 1900). M1chei-Levy s1on of what to sample, and the use of geochem1cal data (1893), Greenly (1903), and Sederholm (1907. 1923, 1n 1dentify1ng the processes contribut1ng to m1gmat1tes. are 1926). Sederholm (1907) described the manner 1n whiCh be not should and gu1de a th1s however, 6; g1ven 1n sect1on an older, foliated gran1te underwent "refus1on" and was regarded as an exhaust1ve survey of what can be done with into 1ntrus1ve masses by the 1njection of the Hango "reborn" compos1t1onal data on rocks and mmerals. Some rock types that are not formed by part1al melt1ng can resemble m1gma- (Finland) gran1te, which 1mparted what he called a "new eruptiv1ty" to it. Because he believed that the rocks 1n the t1tes; thus, 1n section 7, I d1scuss rocks that are commonly Hango area were produced by a comb1nation of processes m1staken for migmatites; examples of these are shown 1n section G . Finally, sections 8 and 9 deal with the prob- and not s1mply by partial melt1ng. Sederholm ( 1907) introduced the term migmatite to describe them. He defined lems of mapp1ng of m1gmat1tes; top1cs range from what to m1gmat1te as "the m1xture of two genet1cally d1fferent conof interpretation the of matter tncky show on maps to the .. . one 1s intrus1ve relat1ve to the other ... To stituents "generat1ons" of leucosome.
deta1led Interpretations applied to 1ndividual migmat1tes.
Ad,,, of 1\.t.gmanrc,
---------------------------------3
th1s group belongs the gneiss1c granit1c rocks wh1ch show 'net structure' charactenst1c of 1nc1p1ent meltmg. frequently 'blind-ending veins', breccia-like granites with innumerable fragments of more-or-less completely d1ssolved older rocks, and finally some stnped gran1tes 1n wh1ch only the still preserved parallel structure ind1cates a fa1nt remnant of the original properties of the rock." Sederholm ( 1907) called the m1gmat1te-forming process palingenesis. and although 1t spec1fically 1ncluded part1al melt1ng and d1ssolut 1on, he also regarded magma inJection and its assoCiated ve1ned and brecCiated rocks as fundamental to the process. The ve1ned rocks were later def1ned by Sederholm (1923) as arterite , and the brecCiated rocks as agmatite. In subsequent work, Sederholm ( 1926) placed more emphasis on the roles of ass1milat1on and the act1ons of fluids, for wh1ch he Introduced the term ichor, 1n the formation of m1gmat1tes. An alternative view of m1gmat1tes was developed by geologists work1ng 1n reg1onal metamorphic terranes. One of them, Holmquist ( 1916). found h1gh-grade gneisses that contain many small patches and ve1ns of granit1c material. As there were no gran1tes nearby, he Interpreted the patches and ve1ns to be the collect1on s1tes for part1al melt exuded from the m1ca-rich parts of the host gneiss. Holmquist gave these m1gmat1tes the name venite to emphas1ze their 1nternal ong1n (Holmqu1st 1916. 1920. 1921) and to distmgu1sh them from Sederholm's artentes. w1th thew veins of InJected material. The absence of nearby granites indicated to him that the heat source for the part1al melting lay deeper 1n the Earth. like that 1nferred for reg1onal metamorphism 1n general. Thus, Holmquist (1916) Introduced the term ultrametamorp hism 1n order to indicate the h1gher-than-normal degree of metamorphism requ1red for part1al melt1ng (i.e., anatexis). The term metatexis was introduced later by Scheumann ( 1936) to mean the general process of partial melt1ng 1n the continental crust, and he Intended that it replace Sederholm's terms anatex1s and m1gmat1te. However, the ex1st1ng term "anatexis" was perfectly adequate, and both metatex1s and d1atex1s disappeared from use. leaving anatex1s to mean part1al melt1ng of the continental crust. The dispute over the origin of gran1te (see the discuss1ons by Read 1957) also Influenced thought on m1gmatites. Bowen's ( 1928) demonstration that a gran1t1c magma IS the product of the extreme fractional crystallization of basaltiC magma, and the subsequent find1ng that there IS far more gran1te than could poss1bly be made by th1s mechan1sm, st1mulated new 1nterest in part1al melting of the continental crust. However. the lack of a structural model to expla1n how large volumes of gran1t1c magma move in the crust led to the so-called "space problem" and to the
idea that bodies of granite form in situ. Debate then turned to whether or not granite has a magmatic ong1n, and among the nonmagmatic protagonists, to the "wet" versus "dry" granitization d1spute. S1nce m1gmatites were commonly seen 1n the field 1n an Intermediate pos1tion between reg1onal metamorphic rocks and granite, op1nion as to thew origin also diverged. The legacy of that controversy was the development of a nongenetlc descnpt1ve termmology for m1gmatites. first by Dietnch and Mehnert (1960), and finally by Mehnert (1968). S1nce the publication of Mehnert's book 1n 1968. a very large maJority of work done po1nts to the fundamental role of partial melting in the formation of migmat1tes. There are few studies of subsolidus processes in m1gmat1tes, but they show that the leucocrat1c subsolidus segregat1ons present 1n the m1gmat1tes had formed prior to the onset of partial melting (e.g.. Blom 1988). Other authors have shown that metasomatism 1s not a significant factor in the formation of m1gmat1tes. although the 1nflux of aqueous flu1ds into rocks that are close to the1r solidus temperatures may provoke partial melting and the format1on of m1gmat1tes (e.g.. Patt1son 1991, Harns et al. 2003, Johnson et al. 2003). Thus. the modern view of migmatites corresponds closely to Holmquist's concept of ultrametamorph1sm, and to Sederholm's concept of anatex1s, but IS far from the concept of palingenes1s, or the vanous metasomatiC and subsolidus processes proposed dunng the granit1zation debate. In modern usage. m1gmat1te IS synonymous w1th anatex1te. Consequently, 1t IS now h1ghly debatable whether a general defin1t1on of m1gmat1te needs to be nongenet1c; 1n th1s book. I use a genetic. partial-melting-based definition and descnptlve term1nology for m1gmat1tes. Any defin1t1on of m1gmat1te should certa1nly 1nclude reference to basic characteristics 1n the field, the most important of wh1ch are that ( I) migmatites occur only in the h1gh-grade parts of reg1onal metamorphic terranes (upper amphibolite and granulite fac1es) and contact aureoles (pyroxene and san1din1te hornfels facies), and (2) because migmatites contain parts where melt1ng occurred, parts where melt was extracted. parts through wh1ch melt m1grated. parts where melt collected, and parts that d1d not melt at all. they are morphologically complex rocks at almost every scale of observat1on. from the m1croscop1c to the macroscop1c.
2.2 A
definition of migmatite
Mehnert (1968) defined a m1gmat1te as"... a megascop1cally compos1te rock cons1sting of two or more petrographically different parts, one of which IS the country rock generally 1n a more or less metamorph1c stage. the other 1s of pegmatitiC, aplit1c, granitic or generally pluton1c appearance." For
I NTROD U CTION
4 -------------------------------
a nongenet1c defin1t1on, the use of the terms "pegmatitic", the terminology used to differentiate the old (nonpartially melted) from the newly created (t.e .. part1ally melted) parts "aplit1c", and "gran1t1c" was unfortunate, as the three carry migmatites 1s discussed. and then the terms that are used in unmistakable genetic implications. A second problem 1s describe the features and variat1ons withtn these broad to the 1mprecise refe rence to metamorphic grade. Ashworth at a smaller scale are outlined. This level of terivisions d ( 1985) rectified both of these problems in his defin1t1on of minology is, therefore. applicable to the fine-scale, detailed migmatite as "a rock fo und in medium-grade to high-grade of migmatites in a small outcrop (less than description metamorphic areas, that is pervasively inhomogeneous 2 ), or large hand-samples. m several on a macroscop1c scale , one part be1ng pale-coloured and consistently of quartzofeldspath1c or feldspath1c compoSition." Unfortunately, th1s nongenet1c definit1on does not Descriptive terms and exclude Sederholm's agmatites, rocks that Brown (1973) definitions for the principal had already argued are intrus1on brewas and not m1gmaparts of a migmatite t1tes. There is further advantage to be ga1ned by exclud1ng agmat1tes. The phrase "... two genetically different constitu- lmag1ne that one 1s able to compare a large outcrop ents ..." 1n Sederholm's ( 1907) defin1t1on of m1gmatite becomes (say 400 m') of m1gmatite w1th the rocks that ex1sted 1n Irrelevant 1f agmat1tes are excluded because for part1al the same outcrop before partial melt1ng occurred. Some melt1ng (and 1ndeed subsolidus segregation and metasoma- rocks. because they have su1table bulk compositions, will tism), the newly generated parts of a m1gmatlte ought to be have been affected by the part1al melt1ng. whereas othpetrogenetically related. Hence, the following revised definition ers, of unsuitable compositions. will not. Those rocks newly formed by part1al melting are called neosome, meaning is proposed. "new rock"; the term IS defined as follows. Migmatite: a rock found tn med1um and high-grade metamorphic areas that can be heterogeneous at the miCroNeosome: the parts of a migmat1te newly formed by, or scopic to macroscopic scale and that cons1sts of two. or more, reconstituted by, partial melting. The neosome may, or may not. petrographically different parts. One of these parts must have have undergone segregation tn wh1ch the melt and solid fracformed by partial melting and contam rocks that are petro- tions are separated. genetically related to each other (called the neosome) and to the1r protolrth through partial melt1ng or segregauon of A general charactenst1c of neosome IS that 1t has a coarser gra1n-s1ze (e.g., F1gs. Bl B4) than the rest of the m1gmathe melt from the solid fraction. The partially melted part Furthermore. the structure (e.g.. foltat1on, folds. t1te. typ1cally contams pale-colored rocks that are quartzolayenng) and microstructure (shape. s1ze, and onentat1on feldspathic, or feldspathlc, in compos1t1on. and dark-colored of gra1ns) that ex1sted pnor to part1al melt1ng are progresrocks that are ennched 1n ferromagneslan m1nerals. However. sively degraded as the degree and extent of part1al melting the partially melted part may simply have changed mmeralogy, 1ncrease. and eventually replaced by a new microstructure microstructure, and grain s1ze without developmg separate l1ght created by the neosome-form1ng processes. A neosome or dark parts. displays a wide range of morphology; as the examples in Note that the proportions are not spec1fied and that some, section B of th1s book show. the melt and sol1d fractions but not necessarily all, of the light-colored parts have to be have segregated in some (F1gs. BS B14). but not 1n others petrogenet1cally related to the other parts. There are four (F1gs. BI5-B25). The neosome 1n many m1gmat1tes 1s 1n Situ 1n found be may all not although m1gmat1te, a to parts bas1c (e.g.. the spat1ally focused neosome described by Wh1te et any part1cular m1gmatite, especially 1f the outcrop exam1ned al. 2004). but 1n migmat1tes where the melt fract1on was 1s small (less than several m ); scale IS an 1mportant factor h1gh. the neosome may have moved. If the port1ons (e.g.. 1n the study of m1gmat1tes. There are parts where part1al beds or compos1t1onal layers) of the rock from wh1ch the meltmg has occurred, parts from wh1ch the melt fract1on neosome formed can be 1dent1fied 1n the pre-m1gmat1te has been removed, parts where the melt fract1on has state. then these parts are called the protolith (Johannes collected. accumulated. or been InJected. and of course. 1985), or the pa rent rock (Ashworth 1985). Th1s means. there are parts that d1d not melt at all. Each of these parts of course, that the protolith cannot be present in a mlgmahas 1ts own w1de range of morpholog1es (or structure). tite; it will have been converted to neosome. m1neral assemblage, bulk compos1t1on. and microstructure Naming of the non-neosome parts of a migmatite has been (also called texture) ; consequently, there 1s a term1nology to by the Inconsistent use of terms, and by the use confused define and describe each. In the next section. I deal with the of inappropnate terms. Mehnert ( 1968) used the term definitions for the constituent parts of a migmat1te; these are illustrated in section B ofth1s book (F1gs. Bl B54). First. "paleosome," mean1ng "old rock" throughout the text of h1s
2.3
Ada .. of MlgmcHitC'
------- ------- ------- ------- ------- 5
book for the non-neosome part of a migmatite, but th1s usage is 1ncons1stent w1th the definition ("parent rock of a migmatite") given 1n the append1x of h1s book. To Ashworth
(1985), such a defin1tlon means that paleosome cannot be present 1n a migmatite: 1t must all have been converted to neosome by defin1t1on. Some authors (e.g., johannes and Gupta 1982) have used the term "mesosome" for the non-neosome part of a migmatite, but that IS problematical too, as the non-neosome part need not be mesocratic, and moreover, 1n many cases, the neosome has mesocratic parts. The root of the problem IS that there IS no clear opinion on which "o ld rock" IS the relevant one. If it IS the "old rock" of the pre-partial- melt1ng state. then paleosome refers to the litholog1es that will become neosome, but this usage is redundant. as adequate terms, i.e .. protolith or parent rock. already ex1st for that part. The alternative. that "old rock" refers to that part of the m1gmat1te not affected by part1al melt1ng and whiCh. therefore. contains only structures that pre-date the part1al melt1ng. IS far more useful. Hence, the follow1ng defin1t1on 1s proposed.
the development of leucocratic and melanocrat1c rocks 1n a neosome IS nearly always the result of well -defined petrological processes (e.g.. part1al melt1ng, segregation of the melt from the solid fraction, and fr act1onal crystallization) that have combined to create migmatites. this nomenclature has proven to be very useful and IS easily transferred to a genet1c scheme. The onset of partial melting changes a one-phase (solid) protolith to a two-phase (melt + solid) neosome. The melt fraction has a lower VISCOSity and density than the residual solid. and consequently, the two parts of a neosome can become separated. or segregated. The part of a m1gmatite from which the melt fraction has been removed can be defined as follows. Residuum : the part of the neosome that is predominantly the sol1d fracuon left after partial meltmg and the extraction of some, or all, of the melt fraction. Microstructures md1cotmg partial meltmg may be present.
Residuum is a general term; there is no particular reference to rock color or to m1neral assemblage. For some bulk compositions. the res1duum may be dom1nated by light-
of grams) IS e1ther unchanged, or only slightly coarsened, compared to that m s1milor rocks JUSt outs1de the reg1on affected by anatexiS.
colored m1nerals. such as feldspar or quartz; typically, because these m1nerals were so abundant 1n the protolith. they dominate the residuum as excess phases with respect to the melt-producing reaction. However, part1al melt1ng of the common crustal rock-types typically generates residua 1n which ferromagnes1an minerals are maJor constituents. Consequently. res1dua are most commonly melanocrat1c. and these are g1ven the special term melanoso me.
Paleosome ex1sts because its compos1t1on was such that 1t did not partially melt and did not become neosome (Olsen 1985); however, as some of the photographs 1n the atlas show, 1t 1s not always easy to decide what has melted and what has not. Further subd1v1S10n of the paleosome may be possible. and even useful. 1n mapp1ng some m1gmatites. If. in companng the pre- and post-partial-melt1ng states. certa1n
Melanoso me: the darker-colored port of the neosome m a m1gmat1te that 1s nch 1n dark mmerals such as b1ot1te, garnet. cord1ente. orthopyroxene. hornblende, clmopyroxene, and even oliVine. The melanosome 1s the solid, res1duol fractiOn (1.e.. residuum ) left after some. or all, of the melt fraction has been extracted. Microstructures md1cotmg port1ol meltmg may be present.
Paleosom e: the non-neosome part of a m1gmat1te that was
not affected by part1ol melting, and m wh1ch structures (such as foliations, folds. layenng) older than the partial meltmg ore preserved. The microstructure (s1ze. form, and onentatlon
litholog1es pers1st unchanged 1nto the h1ghest-grade parts of the migmat1te. then these paleosome lithologies can be called resisters (Read 1957) or r efractory. Layers of calc-silicate. quartzite, and metamaf1c rocks are common resister lithologies 1n migmatite terranes. The neosome and paleosome parts of m1gmatites are shown specifically 1n Figs. Bl B4 of section B, but as the other figures 1n sect1on B show, there IS no "typiCal" or "class1c" morphology to e1ther the neosome or the paleosome part of migmatites; their morphology is h1ghly varied.
Terms specific to the neosome The nongenet1c terminology of Mehnert (1968) IS based on changes in the relative proportions of light- to dark-colored m1nerals that occur as a m1gmat1te forms. Because
The complement to res1duum is, of course, derived from the anatectic melt. and th1s part of a neosome IS called the leucosom e. Leucosom e: the l1ghter-colored port of the neosome m a m1gmat1te. cons1stmg dommontly of feldspar and quartz. The leucosome 1s that part of the m1gmot1te denved from segregated partial melt; 1t may cont01n microstructures that ind1cote crystallization from a melt. or a magma Leucosome may not necessarily hove the composition ofon anatectiC melt; fractional crystallization and separation of the fractionated melt may hove occurred.
The size. physical form. onentation. and grain s1ze are not factors 1n determ1n1ng what IS res1duum. leucosome, or
INTRODUCT ION
6 --- -------- -------- ------ ------ -- - -
melanosome, and the photographs in sect1on B show some of the range of d1versity present 1n m1gmatites. However, these parameters should all be recorded. as they may be important in fully descnb1ng all the types of leucosome and of melanosome (or res1dua) 1n a migmat1te. and could be of s1gn1ficance 1n determ1n1ng exactly how each rock 1n a migmatite formed. The sol1d, res1dual fract1on of a neosome can, 1n most orcumstances , be regarded as being in place (i.e., tn s1tu), but the melt phase is potentially mob1le, a fact recogn1zed 1n the older literature by use of the term "mobilizate" for the melt-derived parts of a m1gmatite. Therefore, a set of terms that convey how far the fract1on of anatectic melt has moved from where 1t formed to where 1t crystallized are useful 1n mak1ng a more complete field-based description of the 1ndiv1dual products of melt1ng in a m1gmat1te. and for the subsequent understand1ng of the petrogenetiC relationships between the melt products and the rocks around them. The following terms are proposed and adopted throughout the book.
product of crystallization of on anamelt that has segregated anatectic on of port tectic melt, or at the Site where the rem01ned has (rom 1ts res1duum, but melt formed. In situ leucosome: the
In-source leucosome: the product of crystallization of on
anatectic melt, or port of on anatectiC melt, that has m1grated away (rom the place where 1t formed, but IS still w1thm the confines of 1ts source foyer. Leucocratic vein or dike: the product of crystallization of on anatectic melt, or port of on anatecuc melt, that has m1grated out of 1ts source foyer and has been tnjected mto another rock, wh1ch maybe nearby, or farther away, but IS still m the reg1on affected by the anatectic event. Granite (or granodiorite , tonalite, etc.) dike or sill : the product of crystallization of a (efs1c melt that has m1grated
out of 1ts source reg1on completely, and IS InJected mto host rocks of lower metamorphic grade or mto nonmetomorphosed rocks. Each of the four can be recogn1zed by a comb1nat1on of morpholog1cal and geochemical charactenst1cs, and by the petrogenetiC relationsh1p between the melt-denved part and 1ts host. In s1tu leucosome 1s n contact w1th 1ts own res1duum, wh1ch typ1cally forms a melanosome around it (see, for example, Figs. BS and Btl). Contacts between the leucosome and melanosome are generally diffuse on a m1llimeter to cent1meter scale. The composition of an m s1tu leucosome corresponds to an in1tial anatectic liqu1d or, if some melt was lost after crystallization had started, to
a cumulate composit ion denved from an in1t1al melt. The petrogenetic relat1onship between the leucosome and host 1s qu1te specific; the leucosome is denved from the anatectic melt with respect to which the melanosome is the res1duum. However. if some loss of melt has occurred, an excess of res1duum relative to leucosome can be expected. An in-source leucosome IS derived from melt that has moved and, therefore, may be d1scordant and have sharp borders w1th 1ts host (e.g., Fig. B27); many occurrences, however, are stromatiC (i.e., layered; see below) , and have some d1ffuse and some sharp contacts. An 1n-source leucosome may have a composition corresponding to an 1nitial anatectic melt, a cumulate or a fractionated anatectic melt. Because 1t IS denved from a melt that has moved, an 1n-source leucosome 1s detached from its own residuum, but as 1t remains 1n 1ts source layer; 1t IS hosted by another. s1m1lar res1duum (melanosome) that was formed from the same source layer. Thus, there IS a petrogenetic connection between the leucosome and 1ts host (e.g., similar Mg-number, mineral assemblage, or ISOtopic compos1t1ons). but they are not exact complements. The res1dual host may, for example, have formed from a slightly lower. or h1gher. degree of part1al melting, or may have expenenced a h1gher, or lower, degree of melt loss. Leucocrat1c ve1ns or dikes typ1cally have sharp contacts. 1rrespect1ve of whether they are discordant or concordant to the structure 1n their host (e.g., Fig. B28). There IS no dwect petrogenetiC relat1onsh1p between the leucocratic vein or dike and 1ts host rock; the age of crystall1zat1on 1n the former should be cons1stent with the metamorphic age 1n the latter. however. Leucocrat1c vems or dikes generally have compositions that ind1cate derivat1on from a fractionated anatectiC melt, but some have compositions of 1nitial melts. and others, compositions of cumulates. Gran1te d1kes typically have sharp contacts w1th the1r host and. because they were InJected 1nto cool hosts, may have fine-grained border zones (chilled marg1ns). There is no petrogenetic relat1onsh1p between a gran1te dike and 1ts country rock. The dike may be a product of crustal anatexis, or 1t may be denved from the crystallization of a felsic. 1ntermed1ate. or even mafic magma. Gran1t1c dikes commonly have a compos1t1on consistent w1th crystallization from a fract1onated melt. Well-defined bands of leucosome are very common in migmat1tes, but as the photographs 1n the book show. a leucosome may also be poorly defined and may form somewhat nebulous patches. Furthermore. 1t 1s qu1te common for several different forms of leucosome to be present 1n the same outcrop of migmatlte (e.g.. Oliver and Barr 1997). Most outcrops are relatively small, and on these. 1t may not
Aria> of Migmatites
- -- -- - - - - - - - - - - -- -- - - - - - - - - - 7
be possible to trace leucocratic veins, or dikes, back to t heir
Magma: a s11icate liquid that contains crystals, wh1ch might
source layers and to examine t heir relationship with 1n situ leucosome. However, the root zones of leucocratic veins
hove crystallized from the melt (llqwdus phases), be the solid products of the melting reaction (also called peritectic products by some authors), or be minerals in excess.
and dikes can be fou nd in some larger outcrops (see Figs. B43-B48), and these show a characterist ic petrological continuity from in situ leucosome t hrough in-source leucosome to the leucocratic dikes and, commonly, with a strong structural cont rol (i.e., by shear zones or folds) on the initial locat ion of the dikes. The best-known form that the residual part of a migmatite can take is the melanosome that occurs along the margin of 1n situ leucosome (e.g., Figs. BS-BI4) . Far less att ention has been paid to t he other, and far more common, morphologies that the melanosome can adopt in migmatit es; it may occur as patches, as continuous layers, and in some cases, as irregularly shaped bodies (e.g., Kenah and Hollister 1983, Sawyer 2001, Wh ite et al. 2004). Some residual rocks formed in migmatites in granulit e-facies terranes are not particularly melanocratic, although they are very w1despread (e.g., Guernina and Sawyer 2003). The neosome in many migmatites has segregated into clearly defined leucosome and melanosome. However, if the melt and residual solid fractions do not separate, leucosome and melanosome do not form, or are far less evident. For these nonsegregated examples, t he term "neosome", used alone, suffices. Examples of a neosome that has not developed leucosome and melanosome parts through segregat ion are shown in Figs. B15- B26. The recogn ition of neosome and it s leucosome and melanosome part s is, generally, a simple matter in migmatites developed from mesocratic or melanocratic protoliths (e.g., metagreywackes, metapelites, metadiorites, and metamafic rocks). It may be a much more difficult matter if the protolith is leucocratic (e.g., some granites, tonalites, trondhjemit es, and metapsammites); the dilution of the mafic minerals by feldspar and quartz means that subsequent changes in modal proportions produce only subtle changes in color, which may be difficult to detect in the field. For such rocks, distinguishing the neosome from the paleosome in the field may best be accomplished by using changes in microstructure, such as fabric or grain size, rather than changes in mineral modes or color. Examples of this approach will be found in the atlas (e.g., Figs. BIS and Bl6). Two furthe r terms specific t o the neosome remain to be defi ned. Melt: a silicate liqwd without crystals.
Terms for the other parts of a migmatite In morphologically complex migmatites, t here are, inevitably, some part s that are neit her leucocratic nor melanocratic. Mesosome has been used as a descriptive term for these, but its usefulness in any genet ic sense is very limit ed . In part, t his is because of inconsistent usage in the past. For Henkes and Johannes ( 198 1), mesosome corresponded to the protolith, whereas for Ashworth (1985) , it was the paleosome. A far more serious problem is that mesocratic rocks can occur in either the neosome or the paleosome; hence, the petrogenetic significance of mesosome is inherently less precise than that attached to leucosome or melanosome, wh ich are, by definition, par ts of the neosome. For this reason, I suggest that mesosome should not be used. Because mesocratic rocks can occur in the neosome or paleosome (or both), it is necessary to record where the particular mesocratic rock is located in the migmatite (e.g., in the paleosome, or in the neosome) . If the melt and solid fractions do not separate in a neosome. it will not develop leucosome and melanosome, and overall, the neosome wil l be mesocrat ic. However, in such cases, a coarser grain-size and a more isotropic fabric will typically have replaced the microstructure and fabric t hat pre-dated anatexis, and w ill aid in ident ifying what is neosome. As more work is done on migmatites, it is becoming clear that some example of 1n situ and in-source leucosome and many leucocratic dikes, or veins, in migmatite terranes are separated from their host rocks by a narrow rim that is compositionally, mineralogical ly, and microst ruct urally different from the host (see Figs. B49-B54). These compositional rims are not the residuum left after extraction of anatectic melt; therefore, the term used for them is se lvedge. Field observations from many migmatite terranes indicate that selvedges are most commonly developed around leucocratic veins and dikes; the injected melt thus does not seem to have been in chemical equilibrium wit h its host. Consequently, three main mechanisms have been proposed for the formation of selvedges: (I) reaction between the host rock and an aqueous nuid exsolved as the melt that produced the leucosome or leucocratic dike crystallized, (2) reaction between minerals in the adjacent wallrock and the injected anatectic melt, and (3) diffusional exchange of components between the anatect ic melt (or the crystallized leucosome or leucocratic dike) and its host in response to activity gradients.
INTRODUCTION
8 ---------------------------------
Selvedges can be leucocrat1c, mesocrat1c, or melanocrat1c. However, a very common type cons1sts of a th1n (mmwlde) b1otite-nch (or hornblende-nch 1n metamafic rocks) melanocrat1c nm adjacent to leucocrat1c dikes and some cases of in-source leucosome; these are called m afic selvedges. Mafic selvedges are also common at the margin of granitic dikes and sills intruded into low-grade country rocks; mafic selvedges are. therefore. not found only 1n m1gmat1tes.
3. MI GMATITES: THE PROCESSES AND MORPHOLOGIES The next step IS to introduce the terms that describe the overall appearance of a migmat1te at a scale larger than a small outcrop. Mehnert ( 1968) introduced 13 morphological terms (agmatitic. diktyon1t1c. schollen, stromatic, surre1tic. folded, ptygmatic, veined, ophthalmitic, stictolit hic, schlieren, nebulitic, and homophaneous) for th1s purpose. His scheme was not explained 1n terms of petrological processes, and so, overall, it has proved unsatisfactory for three mam reasons. Fwst, this scheme does not prov1de a way of understand1ng how one morphology of a migmatite is related to another. Second, 1t does not prov1de any 1ns1ght 1nto the petrolog1cal processes that contributed to the format1on of m1gmat1tes. Consequently, the scheme has turned out to have little pract1cal use 1n making a map of a migmatite terrane. Th1rd, the terms were designed to be nongenet1c, but 1n pract1ce most of the morpholog1es are h1ghly ongin-specific, i.e., they occur exclus1vely as a result of partial melt1ng. The overall morphology of migmatites will be considered next, with the intention of determining wh1ch factors control their appearance at the scale of a large outcrop. Th1s approach leads to a much clearer p1cture of how different morpholog1es of m1gmat1te form and are related to one another, 1n addit1on to prov1ding a workable bas1s for mapping migmatites.
3.1
The first.. order morphological division of migmatites
Field stud1es from a great many anatect1c terranes (Breaks et al. 1978, Brown 1979, Jamieson 1984, Weber et al. 1985, Sawyer and Barnes 1988, W1ckham 1987a, Bea 1991, Obata et al. 1994, Sawyer 1998, Oliver et al. 1999, Solar and Brown 2001, Johnson et al. 2001a, White et al. 2005) reveal a similar change in migmat1te morphology with 1ncreas1ng metamorphic grade. Examples from several m1gmat1te terranes are
shown in sect1on C of th1s book. In the lower-grade parts of anatect1c terranes, paleosome IS dominant 1n the migmatlte (i.e., the proport1on of neosome to paleosome 1s low), and old, pre-partial-melt1ng structures such as bedding, compositional layering, foliat1on, and folds are widely preserved 1n it. The neosome part 1s charactenzed by narrow bodies of leucosome of various orientations, bordered by melanosome, with its residual mineralogy and bulk composition. From a rheolog1cal aspect. the bulk behavior of these m1gmatites d1ffers little from solid rocks that are not partially molten, although the rocks are weaker. Toward the h1gher-grade parts of anatect1c terranes, there 1s a change 1n m1gmat1te morphology; neosome becomes the dom1nant feature. It is s1gn1ficant that 1n many of these higher-grade m1gmat1tes. leucosome IS far more abundant than res1dual matenal. The latter typ1cally occurs as schlieren of mafic m1nerals 1n the leucosome, together w1th schollen or rafts of paleosome and melanosome. Overall, paleosome 1s not abundant 1n h1gher-grade m1gmat1tes, and may even be absent. Charactenst1cally, pre-part1al-melt1ng structures are absent (except where they are preserved as schollen or rafts); they were destroyed during neosome formation and replaced by syn-anatectic structures, typically a magmatiC or submagmatic-foliation, or a flow banding. In terms of rheology, the neosome 1n those m1gmatites was magma-like. The trans1t1on from one morphology to the other IS gradual 1n some m1gmat1te terranes. The progress1ve change 1n the morphology from the lower-grade to the h1gher-grade parts of m1gmat1te terranes is systematically covered in section D of the book. However, the passage from one to the other is abrupt 1n many terranes, and commonly tectonic, as 1t coinodes w1th doma1ns of h1gh shear strain (e.g., Brown and Solar 1998a, Solar and Brown 2001). The abrupt nature of the many contacts suggests that some type of threshold behav1or may be controlling the changeover. Exactly the same change in the morphology of migmatites occurs between the outer and 1nner parts of contact aureoles affected by part1al melt1ng (Flood and Vernon 1978, Pattison and Harte 1988. Grant and Frost 1990, Hobson et al. 1998, Barnes et al. 2002, Johnson et al. 2003) . Th1s two-fold morpholog1cal diVISion of migmat1tes readily falls 1nto the old scheme of metatex1s and diatexis. The paleosome-dominated types were regarded as having formed from low degrees of part1al melt1ng. and were called metatexites, whereas the neosomedominated ones were Interpreted to be the result of nearly complete fus1on, and called diatexites (e.g., Mehnert 1968, Brown 1973, Ashworth 1985). The reasons why only these two basic types of m1gmatites (usefully called metatexite and diatexite) form needs to be understood before the terms can be properly defined. To do th1s, 1t 1s necessary
Atla, of Mtgm
c>-
f
0.8
1.0
_...,--···-~---····r·····-·····..······..········..···:··-··············-·······-·········
t) ()
Fraction of Melt 0.4 0.6 nebulites
patch
~
c
·~
0.2
0
net
1 I
d;ate,;te
li ~i
!: