Mapping, Petrological and Geochemical Explorations of the Massabesic Gneiss Complex in New Hampshire Charles M

University of New Hampshire University of New Hampshire Scholars' Repository

Doctoral Dissertations Student Scholarship

Fall 2007 Mapping, petrological and geochemical explorations of the Massabesic Gneiss Complex in New Hampshire Charles M. Kerwin University of New Hampshire, Durham

Follow this and additional works at: https://scholars.unh.edu/dissertation

Recommended Citation Kerwin, Charles M., "Mapping, petrological and geochemical explorations of the Massabesic Gneiss Complex in New Hampshire" (2007). Doctoral Dissertations. 396. https://scholars.unh.edu/dissertation/396

This Dissertation is brought to you for free and open access by the Student Scholarship at University of New Hampshire Scholars' Repository. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of University of New Hampshire Scholars' Repository. For more information, please contact [email protected]. MAPPING, PETROLOGICAL AND GEOCHEMICAL EXPLORATIONS OF THE

MASSABESIC GNEISS COMPLEX IN NEW HAMPSHIRE

CHARLES M. KERWIN

BS Geology, Keene State College, 1996

MS Earth Sciences: Geology, U.N.H., 2000

DISSERTATION

Submitted to the University of New Hampshire

in Partial Fulfillment of

the Requirements for the degree of

Doctor of Philosophy

Earth Sciences: Geology

September, 2007

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 3277140

INFORMATION TO USERS

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 bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send 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.

UMI

UMI Microform 3277140 Copyright 2007 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346

c 2007

Charles M. Kerwin

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. This dissertation has been examined and approved.

-S jr: $6 Dissertation Co-Director, Jo Laird, Associate Professor of Geology

Dissertation Co-Director, Wallace A. Bothner, Professor of Geology

Peter J. Thompson, Affiliate Professor of Geology

Timothy T. Allen, Professor of Geology, Keene State College

. / > . U , ______G. Nelson Eby, Professor of Geosciences, University of Massachusetts at Lowell

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS

I wish to thank my committee members: co-primary advisors Jo Laird and Wally Bothner

for guidance throughout the entire project, Peter Thompson for countless hours of stratigraphic

discussions, Nelson Eby for the use of the INAA providing this project with Rare Earth Element

analyses and advice on how to interpret and present geochemical information and to Tim Allen

for allowing me to, help calibrate, optimize, work on method development and obtain analyses

with the XRF.

I also would like to thank: the many geologists from all over the United States that

contributed their thoughts and advice on complex geologic terranes and attended my geologic

talks and poster sessions and to the United States Geological Survey for providing funding for the

mapping portion of this study, Steve Allard for years of discussion and collaboration on south

eastern New Hampshire geology problems, Sean Kerwin and John Trautwein for many hours of

editing and last but not least, my family for many years of patience.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS

ACKNOWLEDGEMENTS iv

LIST OF TABLES ix

LIST OF FIGURES x

ABSTRACT xiii

CHAPTER PAGE

1. PREVIOUS WORK AND UNANSWERED QUESTIONS 1

Mapping 1

Stratigraphy and rock descriptions 8

Metasedimentary rocks 10

Central Maine Terrane 10

Devonian Littleton Formation (DL) 10

Madrid (Sm) and Smalls Falls Formation (Ssf) 10

Perry Mt. Formation (Sp) 11

Rangely Formation (Sr) 11

Merrimack Group 12

Kittery Formation (SOk) 12

Eliot Formation (SOe) 12

Berwick Formation (SOb) 12

Igneous rocks 13

Geochronology 13

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Petrology and geochemistry 16

Pressure and Temperature Estimates 16

Geochemistry 16

Purpose 18

2. METHODS 23

Field work 23

Petrography 23

Sample preparation for analytical geochemistry 24

Pressed pellets 24

Fused beads (major elements) 24

Loss on ignition for fusion techniques 24

Sample preparation for fusion 25

Sample preparation and analytical procedures for INAA

(Rare Earth Elements) 25

Sampling 26

Modeling and figures 28

3. GEOLOGY OF THE NORTHERN MASSABESIC GNEISS COMPLEX 30

General geology 30

Structure 35

Foliation 36

Metasedimentary rocks 36

Migmatites 36

Foliated granites and pegmatites 39

Faults 39

Campbell Hill-Hall Mountain Fault 39

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Flint Hill Fault 41

Gulf Shear Zone 43

Other 43

Mineral and crenulation lineations 43

Slickenlines 45

Mafic dikes and quartz veins 45

Summary and interpretation 48

4. PETROGRAPHY AND PETROLOGY 52

Introduction 52

General descriptions 52

Metasedimentary rocks 52

Metawacke 52

Pelitic schists 53

Migmatites 53

Granites 58

Mineral textures 66

Petrology 66

Interpretation 71

5. CHEMICAL ANALYSES 73

Rare Earth Elements 73

Metasedimentary rocks 75

Migmatites 75

Granites 80

Anomalous rocks 80

Major element analyses 84

vii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Granite fractionation 94

Comparison of chemistry with texture 94

Modeling 97

Comparisons with surrounding metasedimentary rocks 100

REE comparison 100

Trace element comparison 104

Summary and conclusions 106

CONCLUSIONS, INTERPRETATIONS AND FURTHER WORK

Conclusions 109

Interpretation 111

T ectonic history 111

Timing 111

Continued work 116

REFERENCES 117

APPENDICES 126

APPENDIX A XRF CALIBRATION CURVES 126

APPENDIX B LOSS ON IGNITION CALCULATION 147

viii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES

PAGE

1 Definitions 20

2 Measuring conditions for the XRF 27

3 a Instrument and sampling precision tables on pressed pellets 27

3b Instrument and sampling precision tables on fused beads 28

4 Mineral abbreviations 53

5 Mineral modes 64

6 Some partition coefficient values for the REEs 75

7 Chondrite normalized REE values determined by INAA 76

8 Slope calculations for the REE patterns La/Lu 78

9 Calculated Europium anomalies 79

10 Chemical data 85

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES

PAGE

1 Portion of the NH State Geology map of New Hampshire 2

2 Previously mapped areas around the MGC 4

3 Pictures of the three different migmatite types 7

4 Simplified map of the composite terranes that comprises New England 9

5 MGC geochronology map 15

6 MGC locations, pressure and temperature estimates map 17

7a, b Locations of basement terranes in northern New England 19

8 Model of evolution of the migmatite types and granites 22

9 Elevation diagram of the Gossville, North wood, Barrington, Candia,

Mt. Pawtuckawy and Epping 7.5 minute quadragles 31

10 Lithology map of the study area 32

11 Picture of a glacially polished basement outcrop of sheared pegmatite 34

12 Poles to foliation of the metasedimentary rocks 37

12a, b Fold axes and axial planes for the Candia quadrangle 37

12c, Fold axes for the remainder of the study area 37

12d. Synoptic diagram for the foliation fot the Northwood and Barrington quadrangle 37

13 a, b Poles to foliation of the migmatites and fold axes 38

14 Stereonet plots of the poles to foliation of the granites 40

15 Hillshade model (DEM) of the study area and adjacent quadrangles 42

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16 Plots of mineral and crenulation lineations 44

17 Map and rose diagram showing orientation of slickenlines within the study area 46

18a, b Rose diagrams showing the orientation of the Mafic dikes in the Candia and Mt.

Pawtuckaway quadrangles 47

18c, d Rose diagrams showing the orientation of the quartz veins in the Candia and

Northwood quadrangles 47

19 Generalized foliation pattern on bedrock geologic base map 49

20a, b Synoptic diagram of pole to foliation planes of the metasedimentary rock

and migmatites 51

2 la, b Outcrop photos of metawacke, tan and purplish weathering 54

21c, d Photomicrographs of metawacke 55

22a, b Outcrop pictures of typical Pelitic schist 56

22c, d Photomicrographs of typical pelitic schist 57

23 MGC Type 2 migmatite in plane and polarized light 59

24 Migmatites on QAP ternary diagram 60

25 Photomicrographs of Cg 90 (granite) plane and polarized fight 61

26 Picture of hand sample (Pg 304) of a granite in the Mt. Pawtuckaway quadrangle 62

27 Granites plotted on QAP ternary diagrams 63

28 Photomicrographs of various lithologies and textures showing evidence of

melting 67

29 Photomicrographs of three metasedimentary rocks that display textures cited as

being caused by decompression 68

30 Graphical representation of the reactions discussed in the text 69

31 Petrogenetic grid showing showing path of the rocks of the MGC 72

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32 Lithology map of the candia 7.5 minute quadrangle 74

33a, b Rare Earth Element plots for the metasedimentary rocks and migmatites 77

34a, b Rare Earth Element plots for the granites and identifier groups 81

35 Possible sub groups 82

36 Anomalous rocks 83

37a, b Major element plots for the granites 90

38a, b Granite discrimination diagrams 91

39a, b Major and trace element diagrams 92

40a, b Minor element graphs and fractionation diagrams 93

41 Granite fractionation diagrams of the incompatible elements Ba and

Sr vs. Eu/Eu* 95

42 Log Ba vs. log Europium anomaly with the migmatites Types 1-3 plotted on the

fractionation trendline from figure 41 96

43 Rock texture and chemistry comparisons Types 1,2,3 migmatites and granites 98

44 Rock texture and chemistry comparisons: Type 2 migmatite and granite (Cg 90) 99

45 Partial melt model for MGC Type 1 101

46 Partial melt model for MGC Type 2 102

47 Partial melt model for MGC Type 3 103

48a REE analyses from this study of the metasedimentary rocks 105

48b, c Comparison of REE data between metasedimentary rocks from the study area

with analyses of Silurian Rangeley and Perry Mt. formations 106

49 Trace element comparison with Silurian Rangeley, Perry Mt. and

Smalls Falls formations 107

50 Model from Figure 8 reconstructed to reflect findings in this study 112

Xll

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51 Folding during orogenisis 113

52 Migmatization of the metasedimentary rocks begins as micas dehydrate

and melts accumulate to produce the various migmatite types 114

53 Late faulting of the two major faults (CHHMF, FHF) allows the MGC to uplift

as a horst-like structure 115

xiii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT

MAPPING, PETROLOGICAL AND GEOCHEMICAL EXPLORATIONS

OF THE

MASSABESIC GNEISS COMPLEX

IN NEW HAMPSHIRE

Charles M. Kerwin

University of New Hampshire, September, 2007

The Massabesic Gneiss Complex (MGC) is a northeast trending belt of rocks that

exposes metasedimentary rocks, migmatites and granites. Early geochronological work indicates

that the MGC is Late-Proterozoic, but recent mapping (~112 mi2 Kerwin, 2000 and ~130 mi2,

this study) of the MGC and surrounding areas has found rock textures indicating that the

migmatites are possibly younger.

The MGC lies within an area of low relief that is mostly covered by glacial overburden

and is structurally defined as a regional anticlinorium in which early fold axes indicate refolding

during progression of the structural arching. Foliation of the metasedimentary and migmatite

rocks indicates they are related either lithologically or by tectonics or both. Faulting along the

two major faults in the area occurred during ductile and late brittle tectonic phases culminating in

horst-like uplifting of the MGC, most likely in Permian times.

Petrographically the rocks show evidence of compression and decompression melting

episodes in which muscovite and biotite dehydration reactions contribute to magmas that the

xiv

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mobilize and accumulate to produce the migmatite and rock textures observed during the

mapping phase of this project. Geochemically the metasedimentary rocks, the migmatites and

granites have two distinct Rare Earth Element patterns that are interpreted to show that some of

the rocks are related to the Barrington Granite and the rest to the Milford Granite. Discrimination

diagrams indicate that the granites (with the exception of the Barrington Granite) are all most

likely fractionated from a Milford-like source and are produced during continent-continent

collision along an active continental margin. Partial melt modeling of the metasedimentary rock

into migmatites indicates that the migmatites Types 1-3 (as outlined in Kerwin, 2000) are melted

Silurian rocks with one of the textures (Type 1) being restitic.

The results of these explorations are interpreted to show that metasedimentary rocks were

buried and melted producing migmatites of the MGC and evidence to support at least three and

possibly four melt episodes before uplift in Permian time.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1

PREVIOUS WORK AND UNANSWERED QUESTIONS

The geologic history of the Massabesic Gneiss Complex (MGC) is complicated in terms

of lithology, deformation and chronology. Methods that have been used to study the MGC are as

multifaceted as are the interpretations of its history. The primary focus of this work was to map

as much of the MGC as possible and construct a logical geologic history based largely on cross

cutting, lithologic and contact relationships and petrographic, chemical and tectonic analyses. In

order to cover all of the topics comprehensively, previous work will be discussed in three parts:

mapping, geochronology and geochemistry.

Mapping

The Massabesic Gneiss Complex (MGC) is a northeast-trending Late-Proterozoic belt of

enigmatic rocks in southeastern New Hampshire (Fig 1). In New Hampshire it is approximately

75 km long and 20km at its widest point. Billings (1956) included it as part of the Fitchburg

pluton following Emerson's (1890) description of rocks on strike with the MGC to the south.

Emerson described strongly foliated and massive granites that intruded and crosscut

Carboniferous and older schists that he interpreted to lie within a large synclinorium. Maczuga

(1981) described the Fitchburg plutonic rocks as containing potassium feldspar megacrysts on the

east of the pluton and interpreted these to be the result of metamorphism and the rocks

containing the megacrysts to be of higher metamorphic grade. Maczuga (1981, table 9) also

described the mineralogy and mode, including specific accessory minerals, all of which would

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHHMF SLF FHF 5 0 5 10 20 C ■ - miles 15 Explanation kilometers

Metasedimentary rocks Igneous rocks Central MaineTerrane ______

Devonian Littleton Formation K9AB Augite-Hornblende diorite and gabbro Silurian Madrid Formation (not shown) Permian Biotite Granite

Silurian Smalls Falls Formation Devonian (late) Concord Granite

Silurian Perry Mt. Formation Devonian two mica granite of southern N.H.

Silurian Rangeley Formation,upper Pink equigranular biotite granite (late Devonian)

Silurian Rangeley Formation, lower Devonian Spaulding Tonalite Exeter Diorite (Early Devonian) Merrimack Group S., | Silurian Newburyport Complex Siluro-Ordovician Kittery Formation Silurian Ayer Granodiorite Siluro-Ordovician Eliot Formation Siluro-Ordovician Eliot Formation Calef M ember Siluro-Ordovician Berwick Formation

Siluro-Ordovician Berwick Formation, Gove member Siluro-Ordovician Berwick Formation ■ Late-Proterozoic Massabesic Gneiss Complex unnamed member I

OZrz Rye Complex Faults B reakfast Hill CHHMF = C am p b ell Hill Hall Mt. Fault SLF = Silver Lake Fault FHF = FLint Hill Fault Figure 1. Portion of the Bedrock Geologic Map of New Hampshire (Lyons et al.,1997) showing rock formations as presently published. Lithology and stratigraphy of rock formations in the study area can be found in Chapter 1.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. correlate well with descriptions of the MGC rocks by Allard (1998) and Kerwin (2000,2001). It

was thought that the peraluminous nature and high potassium content of the rocks indicated an

anatectic origin for the Fitchburg plutonic complex (Maczuga, 1981).

Sriramadus (1966) mapped the Manchester, New Hampshire, 15-minute quadrangle

(Fig 2) and reported that the bedrock consists of granites, pegmatites and gneisses. The

abundance of exposed gneiss along the shores of Lake Massabesic in Manchester N.H. inspired

the name Massabesic Gneiss. In his study Sriramadus distinguished between a pink-microcline

gneiss and a white-oligoclase gneiss. He interpreted the pink gneiss as an intrusion of

microcline granite into schist, which permitted the designation of migmatite for these rocks

using the definition by Sederholm (1913). The white oligoclase gneiss was interpreted as being

modally equivalent to the schists of the upper Berwick Formation and could have been created

by heating the schists to produce a granitic rock, “leaving behind” biotite-rich rock (Sriramadus,

1966, p. 67). He also interpreted the contact between the Berwick Formation and the

underlying Eliot Formation to be gradational. Anderson (1978) mapped the northern region of

the MGC and noted that the MGC/Berwick contact appeared gradational and that the MGC was

conformable with the bounding metasediments to the south and fault bounded to the north in his

study area. Anderson did not observe contacts between the MGC and the local binary granites.

However, he inferred the contacts to be sharp, and he described the granites exposed in the

Deerfield area as highly sheared and associated with deformed pegmatites.

In 1979 Aleinikoff completed mapping of the Milford, N.H., 15-minute quadrangle

through which the MGC passes (Fig. 2). Two types of gneiss were described in this study, a

biotite-rich gneiss and a biotite-poor gneiss. The biotite-rich gneiss was considered a

paragneiss, and the biotite poor gneiss was interpreted to be an orthogneiss. The distribution of

these rocks within the MGC was reported to be predominately paragneiss crosscut by

orthogneiss. This terminology and interpretation lead to the conclusion that the paragneiss must

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHHMF SLF FHF miles

. - 5 kilometers

Explanation

CD Aleinikoff (1979) Milford 15' □ Kerwin (2001) Candia 7.5’ □ Kerwin (2000) CD Kerwin (this study) I '' I Sriramadus (1966) Manchester 15' B 1 1 H 1 Lars011 (1999) Fagan (1986) I— I Allard (1998)

□ Schulz (2004) Epping IS

Figure 2. Portion of the Bedrock Geologic Map of New Hampshire (Lyons et al.,1997) Rock key and explanation can be found in figure 1. Boxed regions indicate previous mapping projects relative to the MGC and surrounding areas.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. be older than the orthogneiss as described further in sections on geochronology and

geochemistry (Chapters 1 and 5, respectively). The Massabesic Gneiss was subsequently

renamed the Massabesic Gneiss Complex (Kelly, 1980). Kelly’s work focused on the absolute

age of the orthogneiss using Rb/Sr systematics (see section on geochronology, this chapter).

Kelly noted that the contacts between the MGC and surrounding metasedimentary rocks were

elusive and appeared gradational. Lyons (1982) interpreted the MGC to be an anticlinorium,

which was later confirmed by the mapping efforts of Fagan (1986) and Kerwin (2000). In 1985

the thought that the MGC might be a large-scale gravity emplaced klippe was introduced

(Hussey, 1985).

Fagan (1986) also noted the presence of migmatites within the Berwick Formation

between the Flint Hill Fault zone and the MGC, analogous to the transition zone of Allard

(1998) and the MGC Type 1 migmatite of Kerwin (1999). Fagan (1986) points out that the field

relationships indicate that if the MGC is late-Proterozoic, then the metasedimentary rocks must

also be late-Proterozoic, or alternatively, if the metasediments are Siluro-Ordovician, then so is

the MGC.

Pouliot (1994) investigated the nature of the “contact” between the Berwick Formation

and the MGC using ground magnetics, VLF resistivity and self-potential geophysical

techniques. The results of this work suggest that the study area is heavily faulted, most likely in

response to movement of the Flint-Hill Fault. Pouliot also concluded that the structures in the

area indicated pre-Mesozoic juxtaposition of the MGC with the Berwick Formation.

Originally mapped as Littleton by Freedman (1950) the pelitic schists southeast of the

MGC were subdivided from the Gove Formation into the Gove and Watson Hill Formations

byAllard (1998) based on differences in metamorphic grade. Allard was also unable to locate

an exact “contact” between the MGC and the Berwick Formation.

Mapping of 112 km2 completed by Kerwin (2000) in the Milford, New Boston,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pinardville and South Merrimack 7.5-minute quadrangles (Fig. 2) showed that the MGC is a

composite of migmatite gneisses, metasedimentary rocks of questionable affinity and granites.

With the exception of some obvious cross cutting intrusions, the majority of the

relationships between rock types within the MGC are gradational, making the distinction

between paragneiss and orthogneiss as described in the literature unwieldy. The impossibility

of differentiating the paragneiss and orthogneiss as described in the literature and the difficulty

in applying the twelve migmatite textures of Menhert (1968) led to the subdivision of the MGC

rocks using a semi-quantitative method based on a foliation spacing index (FSI). FSI is defined

as the number of foliations per unit length multiplied by the SIN of the dip angle plus the

outcrop face angle (Kerwin, 1999, p. 17-21). Type 1 (FSI = 2-3) resembles a migmatized

metasediment (Fig. 3) consisting of continuous biotite foliation with discontinuous felsic blebs

or lenses. Type 2 (FSI = 1-2) is described as roughly 50% leucosome and the foliation and

leucocratic layers are continuous. Type 3 (FSI = <1) migmatites are mostly leucocratic with

relic and flow foliation that is discontinuous to practically non-existent. Type 3 migmatites

often have large (l->2cm diameter) pink or white potassium feldspar crystals, which may be

xenocrysts or porphyoblasts, and magnetite crystals of various sizes. There is also a sub-type of

Type 3 that is the same in every way to Type 3 with the exception of a dark pink color, giving it

the appearance of a pegmatitic migmatite with a swirly foliation. The gradational “contacts” of

the migmatites and metasedimentary rocks led to the hypothesis that the gneisses were partial-

melt migmatites. The migmatites also appear to have been produced in part by the partial

melting of the bounding and overlying metasedimentary rocks (Kerwin, 2000) and re-melting of

the migmatites, mobilization and accumulation of the melts producing granites possibly as

discreet events or as a continuum. The Type 1 migmatites are considered correlative with the

paragneiss reported by Aleinikoff, and Types 2 and 3 are similar in description to the

orthogneiss. However, the relative abundance of the different rock types contrasts significantly

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Type 1 migmatite

Continuous biotite foliation with discontinuous felsic layers

Type 2 migmatite Continuous biotite foliation with continuous felsic layers

Type 3 migmatite Discontinuous biotite foliation with continuous felsic lenses approaching granite

Figure 3. Pictures of the three different migmatite types as described by Kerwin (1999). The majority of the exposed rock is Type 3 with lesser quantities of Type 2 and Type 1 migmatite being the least exposed. Scale is 3x9 inches (7.5x22.7 cm).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with what was reported in the literature (Aleinikoff, 1979).

The Pinardville quadrangle has been mapped at a scale of 1:24000 by Armstrong and

Burton (1999). They mapped three types of MGC migmatites and their Type 1 was considered

to be the original late Proterozoic rock. Criteria for determining the presence of Type 1 rocks

were based on the presence of amphibolite layers, most likely metamorphosed mafic dikes from

the rifting of Rodinia (Burton, personal communication, 2004). The amphibolite layers appear

to occur only in and to the southwest of the Manchester 7.5-minute quadrangle.

The regional structure in and around the MGC defines a broad asymmetric anticlinorium

that plunges slightly to the northeast (Kerwin, 2000). Robinson (1991) notes that it plunges to

the southwest in Massachusetts, possibly indicating a doubly plunging anticlinorium. This large

structure appears to expose deeper rocks in the Pinardville and Milford areas than to the

northeast which is probably why the older amphibolites of Armstrong and Burton are not

present to the northeast. The outcrops in the northeast end expose rocks of the MGC in

gradational contact with rocks that range from the Silurian Central Maine Terrane and Silurian

Berwick Formation to Devonian granites indicating that melting and interaction with younger

rocks took place until Permian time (Kerwin et al., 2004).

Stratigraphy and rock descriptions

The rocks within the study area as shown on the Bedrock Geologic Map of New

Hampshire (Lyons et al., 1997) include a variety of metasedimentary rocks and granites.

According to the state map (Fig. 2) the metasedimentary rocks in the vicinity of the MGC are

largely Silurian and possibly Siluro-Ordovician from the Central Maine Terrane and Merrimack

Group, and the granites range in age from Devonian to Permian. The regional terranes are

shown in figure 4.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Explanation Narragansett Basin Faults Connecticut Valley Trough Permian Intrusives NFZ - Norumega Fault Zone CH-HMF - Campbell Hill- Hall Mt. Fault Central Maine Terrane SL-FHF - Silver Lake-Flint Hill Fault Zone Bronson Hill Anticlinorium PFZ - Portsmouth Fault Zone Merrrimack Group WF - Wekepeke Fault Coastal Litho-Tectonic Block CNF - Clinton Newbury Fault BBF - Bloody Bluff Fault Putnam-Nashoba Terrane HVSZ - Hope Valley Shear Zone □ Avalon Affinities LC-HHFZ - Lake Char-Honey Hill Fault Zone figure 4. Simplified map of the composite terrranes that comprises New England geology; modified from state geologic maps of MA, CT, RI, NH and ME (Zen et al., 1983; Rodgers et al., 1985; Hermes et al., 1994; Lyons et al., 1997 and Osberg et al., 1984).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Metasedimentarv rocks

Central Maine Terrane

Littleton Formation (DP. The Littleton is a Devonian aluminous schist and meta-

greywacke that is unfossiliferous with the exception of a few locations in western New

Hampshire (Billings and Cleaves, 1934). It is a silvery grey medium-grained rock in which

layers commonly indicate topping directions. It presumably was derived from an easterly

source (Bradley and Hanson, 2002) and covered at one time the majority of New Hampshire

and eastern Vermont. Billings (1956) showed the majority of the metasedimentary rocks in

New Hampshire as Littleton Formation with many subdivisions (Billings, 1956). Subsequent

mapping and correlation with rocks from western New Hampshire and Maine led to the

realization that the Littleton Formation was most likely the youngest member of a larger

conformable stratigraphic sequence (Moench, 1971) and was likely correlative with the

Devonian Carrabasett and Seboomook Formations in Maine (Hatch et al., 1983). The Littleton

Formation rests conformably on the Silurian Madrid Formation.

Madrid (Sml and Smalls Falls (Ssfl Formations. The Madrid is a tan-weathering, fine

grained calcareous metasandstone with minor amounts of metashale and local coticule. It is not

found in great quantity within the study area. The Silurian Smalls Falls Formation (Ssf)

underlies the Madrid Formation. The Smalls Falls is a medium-grained, brownish-black to

yellow-weathered sulfidic schist. It has a distinctive yellowish-green and silverish-grey

appearance in outcrop and can contain 1 to 3 inch thick beds of metasandstone (Moench, 1971).

The bottom of the Smalls Falls Formation grades rapidly into the Silurian Perry Mountain

Formation.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Perry Mountain Formation fSp). The Perry Mountain Formation is a rusty-weathering

composite of light gray aluminous schists intercalated with micaceous quartzites. The layers are

typically a few centimeters thick. In the upper Perry Mountain the quartzites dominate the

outcrops while the lower part is dominated by pelitic schists. The quartzite layers are graded

and commonly cross-laminated (Moench, 1971). Coticule is locally common in the Perry

Mountain Formation. Underlying the Perry Mountain Formation is the Silurian Rangeley

Formation.

The Rangeley Formation (Sri. The Rangeley is a rusty-weathering pelitic schist that

contains layers of micaceous-quartzites and quartzo-feldspathic metawackes. Locally it

contains calc-silicate layers and boudins (Lyons et al., 1997). It has been mapped in the type

section as having three distinct parts. The upper is gray-metapelite with minor amounts of

metasandstone, the middle part is a polymictic conglomeratic schist and quartzite with

calcareous clasts and the lower part is a conglomeratic, sulfidic, metasandstone and schist

(Moench, 1971). Conformably underlying the Silurian Rangeley Formation is the laminated to

thin bedded, calc-silicate-bearing, fine grained metasandstone to metasiltstone of the Ordovician

Greenvale Cove Formation (Moench, 1971).

The Smalls Falls, Perry Mountain and Rangeley Formations are thought to be the distal

facies deposited in a southeasterly direction from the Bronson Hill Anticlinorium (Hanson and

Bradley, 1989) into a pre-Acadian basin that was flanked by the Bronson Hill Anticlinorium in

the west and an unknown (Avalon? Berry and Osberg, 1988) micro-continent to the east. The

Madrid Formation is made from sediments thought to have traveled along the axis of the basin in

a southwesterly direction (Hanson and Bradley, 1989), and the Devonian Littleton Formation is

the product of sediments shed from the east (from Avalon?) and transported northwesterly

(Hanson and Bradley, 1993).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Merrimack Group

Kitterv Formation (SOki. The Kittery Formation is a tan-weathering fine-grained

metasandstone to quartzofeldspathic rock. It contains primary sedimentary structures and is

thought to be a sequence of turbidites. Paleocurrent indicators within the Kittery Formation

suggest that sediment transport directions were westerly from a source to the east (Rickerich,

1983). The contact between the Kittery and the Eliot is gradational (Hussey, 1985; Escamilla-

Casas, 2003)

Eliot Formation fSOet. The Eliot Formation is a thin-bedded grey quartz mica phyllite

interlayered with fine grained quartz-biotite-plagioclase-actinolite granofels (Hussey, 1985).

The Eliot Formation is in map contact with the Berwick Formation. Although the contact

relationship is unknown, the Eliot Formation is considered at the time of this writing to be the

bottom unit of the Merrimack Group (Lyons et al., 1997)

Berwick Formation fSObl. The Siluro-Ordovician Berwick Formation (Katz, 1917) is a

purplish-gray weathering fine grained quartzofeldspathic metawacke. It is often massive and

contains few primary structures (Walsh and Clark, 1999). It has two sub-units largely based on

divergence from the metawacke character of the Berwick. The subunits are the Gove member

and the Watson Hill member. Both of the members are pelitic schists, and the main differences

between them appear to be due to metamorphic grade. The Gove member is a medium grained

silvery muscovite-biotite-gamet-staurolite ± sillimanite schist (Freedman, 1950), and the

Watson Hill member is a medium to coarse grained, muscovite-biotite-gamet sillimanite schist

(Allard, 1998). The pelitic sub-units are not widespread and in New Hampshire only exist in

proximity to the Massabesic Gneiss Complex. Stratigraphically the Berwick Formation's place

has been the topic of discussions over the years. It was named by Katz (1917), and due to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the gneissic fabric it was considered it to be older than the Eliot and therefore at the bottom of

the Merrimack Group. Later Freedman (1950) thought it to be the youngest and at the top of

the Merrimack Group based on inferred conformities. It has been placed at the bottom of the

Merrimack Group on the Geologic State map of New Hampshire (Lyons et al., 1997).

Igneous Rocks

Igneous rocks play a role in constraining the ages of unfossiliferous metasedimentary

rocks when the age of the igneous rock is determined. The igneous rocks in the study area are

almost entirely plutonic with the exception of small outcrops of metavolcanics. The youngest

known Paleozoic pluton in the study area is the Milford Granite (southwest end of the MGC in

NH. See Fig. 2) in Milford NH. It is a medium to fine grained gray biotite, quartz, plagioclase,

K-feldspar ± muscovite granite (Aleinikoff, 1979). The Barrington Granite is a medium to

coarse-grained biotite, quartz, plagioclase, K-feldspar ± garnet granite.

Geochronology

Besancon, Naylor and Gaudette (1977) dated the MGC using U-Pb systematics of zircon

from a massive, foliated, magnetite-bearing, granitic orthogneiss. This work indicated an age of

600-630 Ma for the orthogneiss and was interpreted as an intrusion age. Besancon et al.

also described a sillimanite-bearing biotite gneiss with minor amphibolite (paragneiss) which

was considered older than the orthogneiss, and the MGC was cautiously considered an

allocthon and tentatively correlated with the “Eastern basement province” to the southeast. The

distribution of more recent geochronology data is summarized on figure 5. Aleinikoff (1979)

also did U-Pb-Th analyses on zircons. The focus of his work was to discriminate between the

paragneiss and orthogneiss of the MGC. Aleinikoff s work showed that the paragneiss had a

discordant age of 646 Ma. and the age of the orthogneiss was 475 Ma and also discordant.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. He explained the discrepancies as due to the heat from the orthogneiss intrusion contributed to

Pb loss in the paragneiss, and the heat from the intrusion of the 275 Ma Milford (NH) granite

likewise resulted in Pb loss from the orthogneiss. Subsequently, zircons were reanalyzed

(Aleinikoff et al., 1995) on a higher precision instrument (SHRIMP) and the age of the

orthogneiss was determined to be 623±8 Ma (Fig 5).

Kelly (1980) obtained rubidium/strontium whole rock analyses of the orthogneiss which

generated two isochrons; one with a slope indicating an age of 641± 26 Ma and another

indicating an age of 671 ±21 Ma. One of the rocks giving the older age was described as a

“massive, medium grained, biotite bearing, pink granitic gneiss” (Kelly, 1980, p.46). The two

isochrons were considered distinct because they had different initial 87Sr/86Sr ratios. However,

it was thought that the higher of the initial Sr ratios could be a function of contamination from

crustal rocks indicated by oxygen isotope data as proposed by Mose and Wenner (1980).

Mineral ages (Eusden and Barreiro, 1988) were also determined using potassium feldspar

and biotite, which gave ages of 308 ±10 Ma and 220 ±7 Ma. The mineral ages were

interpreted to be the result of a Permian tectonic overprint. They also reported metamorphic

ages for MGC paragneiss (titanite, 261±9 Ma) and the Berwick Formation (monazite, 277±5

Ma). Recent unpublished data (Wintsch and Dorais, personal communication, 2004) from

Ar^/Ar39 systematics from mineral separates taken from MGC rocks indicate that the MGC was

deep in the crust into Permian time and subsequently unroofed leaving no trace of Acadian

metamorphism.

The MGC has been intruded by the Milford pluton that is 275 Ma (Aleinikoff et al.,

1979) in the southwestern end, by granitic rocks of the Barrington Pluton (363 M.a., Eusden and

Barreiro, 1988) and by igneous rocks of Cretaceous age (122±2, 128±2 m.y.; Foland et al.,

1971; Eby, 1985, respectively) of the Pawtuckaway Mountain Complex at the northeastern end.

No one has attempted to quantify the timing of migmatization(s), to constrain the upper age of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. • Aleinikoff (1979) Milford Granite, 275 Ma. • Aleinikoff (1979) MGC paragneiss, 646Ma. • Aleinikoff (1995) MGC orthogneiss, 623±8 Ma. • Eusden and Barriero (1988) Barrington Granite, 364Ma. • Eusden and Baniero (1988) MGC paragneiss (sphene)metamorphic age, 261±9Ma • Eusden and Barriero (1988) Berwick, monazite, metamorphic age, 277±5 Ma.

• Kelly (1980) Rb/Sr mineral separates, 220±7Ma. • Kelly (1980) Rb/Sr mineral separates, 328±10Ma.

Figure 5 Map of the study and surrounding areas emphasizing sample locations where rock and metamorphic ages were obtained. Light grey outlines are geologic formation lines adapted from Lyons et al., 1997, Fig. 1, and the dark polygons are from Kerwin 2000 (southwest end of MGC) and this study (northeast end of MGC).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. migmatization and/or deformation or to determine what role the metasedimentary rocks might

have played in the formation of the migmatites.

Petrology and geochemistry

Pressure and Temperature Estimates

Pressure and temperature estimates (Allard, 1998) from mineral assemblages show

changes in depth from the Berwick Formation into the MGC (Fig. 6) and a similarity between

the pressure estimates from either side of the Flint-Hill Fault (FLF) indicating only ~lkm throw

on the fault (compare Allard's locations 1 and 4 with 2 and 3, Fig. 6). Thermobarometry

presented by Larson (1999) demonstrated an increase in temperature and pressure recorded by

the calc-silicate rocks from the Berwick Formation southeast of the MGC to the axis of the

MGC. Temperatures and pressures of 450-575°C and 4.5-6.0 kbars in the Berwick Formation to

690°C and 9.0 kbars in the MGC were reported. These temperatures and pressures agree well

with the anticlinorial interpretation of Lyons et al. (1982), Fagan (1986) and Kerwin (2000).

Schulz (2004) analyzed several samples of the Berwick Formation from the Epping 7.5-

minute quadrangle. He obtained a temperature range of 472-628°C and pressure estimates of

4.8-7.9 kbars. He interprets his data as showing two metamorphic events in which the early

event (Ml) is a higher metamorphic grade than the second event (M2). He also infers 4.8 km of

total vertical offset for the Flint-Hill Fault just to the northeast of Allard's area (Fig. 2).

Geochemistry

Dorias and Paige (2000) discriminated between Grenvillian and Avalonian plutons in

northern New England using geochemical and isotopic methods to identify different sources for

the granites. Their study used oxygen isotopes and trace element geochemistry from 36 plutons

to redraw the Grenville / Avalon basement contact to a more southerly position than was

inferred previously by Osberg (1978), Zen et al., (1983) and Zartman (1988) from radiogenic Pb

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pressure and temperature estimates

Allard (1998) Larson (1999) Schulz (2004) m 580-600°C 3.75 kbar 1 690°C 9 kbar - 628°C 7.9 kbar 600-615°C 4.75 kbar 2 575°C6kbar * 596°C 6.7 kbar # 540-570°C 4.45 kbar I 560°C5kbar «<»r 63khar . 585-595-C 5 4 450 C 45 k>” * “ BS 605-615°C 5-5.5 kbar # 472PC 4.8 kbar # 590-601°C 3.75 kbar » 506-590°C 3 kbar

Figure 6 . Locations, pressure and temperature estimates for rocks analyzed in the Berwick Formation and the MGC. Allard reported the numbers for the high and low end estimates that were calculated. Schulz also report two sets of calculated pressure and temperatures and interprets them as belonging to two metamorphic events in which M l is the higher values and M2 the lower values.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. systematics (Fig. 7). The crustal block boundaries as determined by chemical findings do not

agree with reported positions inferred from geophysical evidence (Unger et al., 1987; Stewart,

1989). The proposed placement of the basement contact (Dorais and Paige, 2000) puts the

MGC and the Merrimack Belt overlying Avalon basement in closer agreement with the

interpretation of Gromet (1989) that the MGC belongs to a group of Avalonian terranes exposed

by Mesozoic extension.

Another study by Dorais and Wintsch (2001) used epsilon Nd values from rocks

collected from the Merrimack Belt to invoke a Laurentian source for the sediments and magmas

in this belt. This seems at odds with the previous study that insists that the MGC and rocks to

the northwest are Avalon and rest on, or are melted from, Avalon basement, but would reflect

the idea that the basement underlying the study zone could be a composite of Laurentian

sediments and accreted basement terranes. Dorais, Wintsch and Kunk, (2003) calculated

pressure-temperature-time paths to show that the metamorphic conditions were high enough in

the Alleghenian to overprint most of the effects of the Acadian orogeny in southern New

England. They also concluded from_differences in the metamorphic grade between the MGC

and the metasedimentary rocks to the northwest and southeast that a “metamorphic and

structural discontinuity” exists above the MGC.

Purpose

The work done, past and present, shows that the MGC is largely composed of

migmatized rocks surrounded by metasedimentary rocks and containing inclusions of

metasedimentary rocks. The literature about migmatites includes many micro and macro

migmatite examples in which leucosomes (granitic material) are surrounded by melanosome

selvages (restite) (see Table 1 for migmatite nomenclature). If the melting took place in a

closed system the sum of the leucosome and melanosome would equal the original parent host

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 7. 7a. Locations of basement terranes in northern New England as proposed by Osberg (1978), Zen et al., (1983) and Zartman, (1988). Grenville (1.1 Ga) basement could be found along the NH/VT border and Avalon along extreme eastern Maine separated by a basement terrane (Medial NE) that was a combination of the two. 7b. Isotopic signatures of various plutons (Dorais and Paige, 2000) in New England suggest that the Avalon basement boundary is located much farther west and lies against Grenville with no intermediate basement terrane.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rock or paleosome. It is my opinion that in certain cases where the migmatization has occurred

as a local event a closed system hypothesis could be tested. However, in regional cases such as

the MGC a closed system, local event hypothesis is probably not realistic.

The main purpose of this work was to test the hypothesis that the MGC underwent

melting episodes (or one episode of long duration), in which the surrounding metasediments

underwent migmatization(s) creating rocks that are now collectively considered to belong to the

Late-Proterozoic Massabesic Gneiss Complex, and ultimately, the granitic sills and plutons

within the study area.

Table 1

Definitions Taken after Ashworth (1985, Table 1.1)

Migmatite terms Leucosome -The pale colored, quartzofeldspathic component within a migmatite. Melanosome -the dark colored, mafic mineral-rich layer often associated with -leucosomes as a rind. Neosome -portion of rock that includes leucosome + melanosome. Mesosome -Rock as it presumably appeared before alteration, usually intermediate in color between leucosome and melanosome. Interpretation of terms Paleosome -Hypothetical parent rock Protolith -Hypothetical parent rock Restite -Residual rock after melt extraction. Melt terms Anatexis- Partial melting Metatexis -Partial melt in which the melt/freeze process is not abundant enough to disturb pre-migmatization textures. Diatexis - Extensive melting in which pre-migmatite textures are disrupted and mafic minerals are included within the leucosome. Leucosome volume equals or is greater than non-leucosome material. Textural terms often found in the literature Stromatic -layered Agmatic looks like it has been brecciated Schollen -blocks of non-melt material rafting in leucosome Schlieren -streaks of non-leucosome within leucosome

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The hypothesis inherently includes the sub-hypothesis that if the migmatites were derived

from the metasediments then at least some of the MGC has to be younger than Late Proterozoic.

These hypotheses were tested by bedrock mapping, which examined crosscutting and gradational

relationships between migmatites and metasedimentary rocks, as well as between migmatites.

Mapping also permitted analysis of the structures, which enabled a reconstruction of the tectonic

history, in which events are categorized temporally. A second method involved whole rock

chemistry of a suite of metasedimentary rocks, the three types of migmatites and granites. Once

the chemistry of the rocks was determined, rare earth and trace elements were used to test the

hypothesis that the metasedimentary rocks were melted (at least partially) during one or more of

those melting events. Also it was hoped to determine if the melt from the metasedimentary rocks

was injected into the country rock producing migmatites or if the metasedimentary rocks are

partially melted in situ migmatites. Based on field relationships, the latter would suggest the

melts from the migmatites and metasedimentary rocks are the source rocks that produced the

granites. The results indicate that some rocks of the MGC may have been melted more than once

similar to an evolution model of Weber et al. (1985) (Fig 8) in which parent rocks are melted, and

melt is mobilized and pooled to produce rocks observed in the MGC and proximal granites.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Removal of silicate liquids

melt 2 ? segregation and collection

Partial melting melt 1 ?

Patent gneisses

Figure 8. Model of evolution of the migmatite types and granites modified from a model of Weber et al.,(1985). The parent gneisses could be Late-Proterozoic and Silurian metasediments. MGC Type 1 is a derivative of the parent gneisses and the Type 2 and 3 migmatites are derivatives of Type 1. MGC Type 2 may also contribute to production of MGC Type 3. Types 2 and 3 may contribute to Barrington granite and related sills. Removal of magmas or remelting of MGC types 2 and 3 result in the accumulation of granitic melts forming the Milford Pluton. This hypothesis was constructed based on relationships observed in the field. The events described above could be discreet events or have occurred as a continuum.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2

METHODS

Field work

Mapping was done at a scale of 1:24000 using U.S.G.S. 7.5-minute quadrangles as

topographic base maps. Standard pace and compass methods were used and locations

(uncorrected) were checked using a Garmin VHF/GPS 170. Location data were taken within the

NAD 1983, zone 19 UTM grid. Rock descriptions, location and structural information were

recorded and digitized using Arc View 33©. Structural data included were foliation, crenulation

cleavages, mineral lineations, minor and major fold axes and axial planes. The data were plotted

using Stereonet v. 6 (Almendinger, 2001) on to equal area, lower hemisphere stereographic nets.

Foliation planes were plotted as poles to the planes for which a best cylindrical fit great circle and

the pole to that plane (jt axis) were calculated. Sample size varied depending on the purpose for

which it was collected. Thin section samples were hand sample size. Samples for geochemical

analyses were approximately ten times the largest grain size, usually more. Oriented samples

were taken in places where structural information was warranted.

Petrography

Thin sections were cut perpendicular to foliation, and where sense of shear was obtained,

multiple thin sections were produced perpendicular to lineation and foliation and parallel to

lineation. Multiple thin sections were also made at locations where suspected compositional

differences between layers existed. Visually estimated mineral modes were recorded for all

samples except the granites and granitoid migmatites, which were point counted. Point counted

slides were done for 500-1000 counts. Mineral modes are summarized in chapter 4.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Photomicrographs were taken using a Nikon Cool Pix 4500 digital camera.

Sample preparation for analytical geochemistry

Several techniques were employed in order to analyze a range of elements by either X-

ray fluorescence (XRF) or Instrumental Neutron Activation Analyses (INAA). The heavy trace

elements were analyzed by XRF using pressed powder pellets. The lighter elements were

analyzed by means of homogenous fused beads (XRF) and the REE by INAA. XRF analyses

were accomplished on a Rigaku ZSX lOOe wavelength dispersive sequential XRF spectrometer at

the Keene State College Analytical Geochemistry Laboratory. Measuring conditions and

elements analyzed can be found in Table 2. The XRF was calibrated using pressed pellets and

fused beads made from a range of U.S.G.S. Rock Reference Materials. Calibration curves can be

found in Appendix A. Instrument precision was measured by running one bead and pellet

multiple times. The calculated level of uncertainty is from less thanl% RSD to around 10% RSD

(Table 3a and 3b). The higher levels of uncertainty are due to measured concentrations near the

lower detection limit. To test sample preparation precision, ten beads and pellets of the same

sample (MGC 2NR, table 3) were prepared and analyzed. Levels of uncertainty are much higher

(1-2X) indicating that sample preparation and sampling are the largest sources of error.

Pressed Pellets (trace elements)

All samples were crushed in a jaw crusher and split to a small fraction (~30 g) after

which they were pulverized in a shatterbox using a ceramic puck and container and stored in acid-

washed bottles. Six ± 0.001 grams of sample were weighed and 0.75 ±0.001 grams of binder

(3644 Ultrabind) was added. The sample and binder were mixed in a mixing mill for 1 minute

after which it was poured into an aluminum sample cup that had been placed in a hydraulic press.

The sample was compressed for 10 minutes at a pressure of ten tons.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fused beads (maior elements)

Loss on ignition for fusion techniques

Ceramic crucibles were weighed and the mass recorded. Approximately 1.5 gram of

sample was added to the crucible and the total mass was recorded. The crucibles were placed in a

muffle furnace at 1000°C for one hour, after which they were removed from the furnace and

placed in aluminum cooler/dessicators to cool (~0.5 -lhr). After the samples cooled they were

weighed again and material lost on ignition was calculated and expressed as a percent using the

formula LOI= g/(mws-om) where g = grams lost, mws = mass of the crucible plus the sample and

om = mass of the crucible (Appendix B).

Sample preparation for fusion

The flux was added to achieve a dilution of 10:1. This dilution was an appropriate

viscosity to pour the melt from the crucible into a mold. 0.9 ±0.001 grams of sample were pre

weighed in a mixing vial into which (pre-weighed) 4.5 ±0.001 grams of lithium tetraborate

(Li2B40 7) fluxing agent and 4.5±0.001 grams of lithium metaborate (LiB02) were added. The

sample and flux were mixed in a mixing mill for ten minutes, after which, the samples were

placed in a platinum/gold crucible with 0.5 piL of Lithium Bromide (LiBr) as a non-wetting agent

and placed in a muffle furnace at 1100°C for five minutes until melted (fused). The samples were

agitated and re-fused for an additional 5 minutes and poured into a platinum/gold mold. The

bottom surface of the mold is machined to a flat mirror finish so that no other sample surface

preparation was required.

Sample preparation and analytical procedures for INAA (Rare Earth Elements)

Approximately 200 mg of powdered sample was weighed into an acid cleaned, high

purity polyethylene vial. Approximately 0.2 ml of Au solution was added to the vial. The

solution was evaporated, and the vial was then heat sealed. Samples were irradiated for

approximately 6 hours at a flux of 4.0 x 1012 n cm'2 s'1. Ten to fourteen days after irradiation the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. samples were counted for 10,000 second intervals on a fully-automated dual detector [Ge(Li)

detector for high energy gamma rays and a LeGe detector for low energy gamma rays] INAA

system. Four to six weeks after irradiation the samples were recounted for 40,000 second

intervals on the same system. Each count set consists of the irradiation standard, 10 unknowns,

and a check standard. The irradiation standard was the U.S.G.S. rock standard AVG-1 using the

accepted values reported by Govindaraju, (1989). The Canberra spectroscopy program, Spectran-

AT, was used to locate gamma ray peaks and to determine gamma ray peak areas. Subsequent

data reduction was done using an in-house computer program. Peak areas were corrected for

gamma ray overlaps and the decay corrected to an initial start time. Where appropriate,

corrections were applied for uranium fission products. Geometry and flux corrections were made

using the relative intensities of the Au introduced by the Au spike. Repeated analysis of check

standards gives a precision and accuracy of better than 5% (relative) for Fe, Na, Sc, Co, Cr, La,

Ce, Nd, Sm, Eu, Gd, Tb, Tm, Yb, Lu, Hf, Ta, and Th and better than 10% for Rb, Ba, Sr, Cs, U

and Sb when element concentrations are 5x or greater than the detection limit.

Sampling

Samples (Fig. 30) are almost entirely collected from within the Candia quadrangle and

were chosen to reflect a wide distribution of lithologies to test the hypothesis that they belonged

to, or were derived from, the same rock formation. Ten metasedimentary rocks samples were

taken from locations throughout the quadrangle including a few from within MGC boundaries.

Migmatites were sampled to reflect the three types; two samples of MGC Typel and four of each

of the MGC Type 2 and 3. Migmatite leucosomes and melanosomes were not separated. Ten

granites were sampled to include foliated and non-foliated varieties, the Milford granite of

Milford, NH, and Barrington granite from Strafford, NH. Three additional granite samples were

taken from layer-parallel and cross-cutting sills that were originally intended for

geochronological work to constrain the upper age of migmatization in the MGC.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 2. Measuring conditions for the XRF

Measuring conditions for die XRF X-Ray tube: Rh, 50 Kv, 70Ma Filter out Attenuator 1/1 CoUimnatar fine detectors: SC-scintillation counter, PC-Proportional counter

Na KA TAP 55.120 200 PC 53.120 40 100-300 Mg KA TAP 45.910 200 PC 43.190 40 100-300 Al KA PET 144.610 40 PC 136.500 20 110-310 Si Ka PET 109.045 40 PC 104.140 10 114.630 10 100-300 P KA Ge 141.190 200 PC 143.190 40 130-290 K KA LiFl 136.675 45 PC 139.175 15 110-290 Fused Beads Ca KA LiFl 113.125 40 PC 106.000 10 119.200 10 110-290 Tt KA LiFl 86.110 200 SC 85.000 40 11-290 Fe KA LiFl 57.495 40 SC 54.760 40 61.000 10 80-400 Mn KA LEF1 62.950 40 SC 62350 40 63.550 40 100-300 Rb KA LLF2 37.975 200 SC 37.170 40 38.720 40 100-300 Sr KA LiF2 35.845 200 SC 35.260 40 36.880 40 100-300 Zx KA LiF2 32.085 200 SC 30.950 40 32.970 40 100-300 Ba KA LiFl 87.135 200 SC 88.500 40 120-270 Br KA LIF2 42.250 100 SC 42.250 20 43.720 20 100-300

Element hne CYvstal Peak (2 theta) time (s) detector BG1(2 theta) Time (s’) BG2 a theta) TImefs) PHA Rb KA LIF2 37.985 80 SC 37.170 40 38.720 40 100-300 Sr KA LiF2 35.845 80 SC 35.260 40 36.880 40 100-300 Y KA UF2 33.885 160 sc 33.030 80 35.260 80 100-300 Zr KA LiF2 32.085 80 sc 30.950 40 32.970 40 100-300 Pressed Pellets Nb KA LiF2 30.469 160 sc 29.870 80 30.820 80 100-300 La LA LiFl 83.100 160 PC 82300 80 84.300 80 150-230 Ce LB1 LiFl 71.740 160 PC 70.790 80 72.700 80 130-270 Nd LA LiFl 72.270 160 PC 71.200 80 72.900 80 150-260 Pb LB1 LiFl 28.285 160 SC 27.000 80 29.050 80 100-300 Th IA LiF2 39.275 160 sc 39.000 80 40.000 80 100-300 Table 3a. Instrument and sampling precision tables on pressed pellets

THIRTEEN Analyses of same specimen Sample Date & Time Rb S r Y Zr Mi La Ce Nd Pb Th mgc2NR new 4/15/03 23:42 94 215 43 350 2 0 42 82 43 6 7 mgc2NR new 4/16/03 1:16 93 215 44 352 21 45 8 7 43 5 7 mgc2NR new 4/16/03 2:50 94 216 44 351 20 42 81 46 6 6 mgc2NR new 4/16/03 4:24 94 217 44 353 20 41 86 44 6 7 mgc2NR new 4/16/03 5:58 95 217 44 353 20 42 84 43 5 7 mgc2 NR new 4/16/03 7:32 95 217 44 3 54 20 44 85 42 7 8 mgc2 NR new 4/16/03 9:07 95 218 44 355 21 40 91 43 6 7 mgc2 NR new 4/16/03 10:41 95 217 44 355 21 43 8 5 42 6 fi mgc2 NR new 4/16/03 12:14 95 218 4 5 355 20 42 87 41 6 8 mgc2 NR new 4/16/03 13:48 95 218 44 355 21 42 85 44 6 7 mgc2 NR new 4/16/03 15:23 95 219 44 356 2 0 41 93 43 6 7 mgc2 NR new 4/16/03 16:56 95 219 4 5 357 21 40 9 2 40 6 7 mgc2 NR new 4/16/03 18:31 96 218 44 357 21 43 92 41 6 6

AVERAGE of 13 95 217 44 354 2 0 42 87 43 6 7 1 RSD (%) 0.79 0.6 1.12 0.62 2.54 3.43 4.52 3.63 8.33 9.05 2 Std Dev (95% confidence) (ppm) 2 3 1 4 1 3 8 3 1 1

THIRTEEN specimens of same sample Sample Date & Time Rb S r Y Zr Nb La Ce Nd Pb Th mgc2NR wet 4/15/03 22:55 102 212 42 351 20 47 96 47 6 7 mgc2NR dry 4/16/03 0:29 93 213 43 338 20 36 73 35 7 6 mgc2NR 1 4/16/03 2:03 97 220 44 355 21 48 97 45 5 7 mgc2NR 2 4/16/03 3:37 96 215 43 332 21 39 78 40 5 7 mgc2NR 3 4 /1 6 /0 3 5:11 95 215 45 365 21 46 87 43 6 8 mgc2 NR 4 4/16/03 8:19 97 214 43 359 21 44 83 40 6 6 mgc2 NR 5 4/16/03 9:54 95 214 43 322 20 39 77 36 6 7 mgc2 NR 6 4/16/03 11:27 96 214 43 343 21 37 73 36 6 6 mgc2 NR 7 4 /1 6 /0 3 13:01 95 215 43 345 20 36 77 33 5 6 mgc2 NR 8 4/16/03 14:36 93 215 44 324 21 40 80 38 7 6 mgc2 NR 9 4/16/03 16:09 94 215 43 355 20 39 82 41 5 6 mgc2 NR 10 4/16/03 17:43 94 215 44 346 21 40 83 42 6 6 MGC 2 NR new AVERAGE of 13 frorr 95 217 44 354 20 42 87 43 6 7

AVERAGE 95 215 44 345 21 41 81 39 6 7 1 RSD (%) 2.44 0.9 1.77 3.84 2.53 9.96 9.26 10.4 11 .78 10.1 2 Std Dev (95% confidence) (ppm) 5 4 2 26 1 8 15 8 1 1

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 3b. Instrument and sampling precision tables on fused beads.

N a20 MgO A1203 Si02 P 2 0 5 K 20 CaO TKD2 F e2 0 3 wt% wt% wt% wt% wt% wt% wt% wt% wt% MGC 2 NR 3.14 1.308 13.505 69.783 0.273 1.654 3.508 0.894 5.709 MGC 2 NR 3.128 1.313 13.501 69.806 0.273 1.654 3.498 0.887 5.716 MGC 2 NR 3.132 1.308 13.518 69.768 0.276 1.659 3.496 0.89 5.736 MGC 2 NR 3.121 1.312 13.515 69.743 0.276 1.664 3.51 0.891 5.741 MGC 2 NR 3.126 1.312 13.51 69.72 0.276 1.665 3.512 0.896 5.757 MGC 2 NR 3.105 1.312 13.504 69.78 0.277 1.663 3.504 0.902 5.727 MGC 2 NR 3.098 1.303 13.524 69.799 0.276 1.654 3.507 0.889 5.721 MGC-2 NR 3.088 1.306 13.515 69.87 0.276 1.649 3 483 0.888 5.707 MGC 2 NR 3.098 1.299 13.499 69.894 0.277 1.647 3.48 0.885 5.702 MGC 2 NR 3.126 1.304 13.521 69.859 0.277 1.645 3.482 0.882 5.688 mgc 2 NR 3.096 1.298 13.497 69.887 0.279 1.654 3 482 0.885 5.696 mgc 2 NR 3.107 1.303 13.502 69.815 0.278 1.656 3.496 0.888 5.732 mgc 2 NR 3.12 1.309 13.511 69.776 0.281 1.664 3.492 0.892 5.735 mgc 2 NR 3.132 1.311 13.484 69.778 0.28 1.659 3.511 0.882 5.731 mgc 2 NR 3.119 1.312 13.496 69.776 0.282 1.664 3.513 0.887 5.732 mgc 2 NR 3.116 1.309 13.478 69.8 0.283 1.67 3.509 0.88 5.736 mgc 2 NR 3.103 1.302 13.492 69.806 0.283 1.659 3.509 0.889 5.728 mgc 2 NR 3.101 1.307 13.536 69.806 0.283 1.654 3.495 0.889 5.706

average 3.11 1.31 13.51 69.8 0.28 1.66 3.5 0.89 5.72 rsd% 0.48 0.36 0.11 0.07 1.16 0.41 0.33 0.6 0.31 2 st dev 0.03 0.01 0.03 0.09 0.01 0.01 0.02 0.01 0.04

Mn Rb Sr Zr Ba ppm ppm ppm ppm ppm MGC 2 NR 942 99 214 412 570 MGC 2 NR 941 100 210 411 573 MGC 2 NR 944 100 211 413 512 MGC 2 NR 955 99 211 415 592 MGC 2 NR 945 98 211 414 577 MGC 2 NR 940 103 212 416 591 MGC 2 NR 943 99 210 413 612 MGC-2 NR 940 99 212 413 514 MGC 2 NR 936 101 210 412 528 MGC 2 NR 937 97 207 409 496 mgc 2 NR 942 102 212 413 574 mgc 2 NR 940 97 209 412 574 mgc 2 NR 943 97 211 416 546 mgc 2 NR 949 100 210 413 646 mgc 2 NR 941 98 211 415 531 mgc 2 NR 946 99 213 415 517 mgc 2 NR 940 100 210 415 593 mgc 2 NR 945 99 212 413 564

average 942.72 99.28 210.89 413.33 561.67 rsd% 0.47 1.65 0.74 0.44 7.03 2 st dev 8.78 3.28 3.14 3.63 78.99

Modeling and figures

Partial melt models were run on an Excel spreadsheet, and magma mixing models were

run using the computer program PetMin3 (Eby, 1989). Partial melts were calculated from

average metasediment compositions and non-modal batch melting equations

C,=C0/ D0 +F(1-P)

CS=C0D/D0+F( 1 -P)

where Q = concentration of the element in the liquid, Cs = concentration of the element is the

residual solid, Cc = the concentration of the element in the original solid, F = the melt fraction,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. D= bulk distribution coefficient for the residual solid, D0 = bulk distribution coefficient for the

original solid and P = bulk distribution coefficient for the melt proportions.

D and D0 were calculated using the equation

Dj = x1Kd1+x2Kd2+x3Kd3...

where D; is the bulk partition coefficient for the element i, x, = proportion of mineral expressed

as a percent, Kd! = the mineral /melt partition coefficient for element i in mineral 1. P was

calculated using the equation, P = p1Kd]+p2Kd2+p3Kd 3 ...where P is the bulk distribution

coefficient for the melt proportions, p! = the weight fraction of mineral 1 in the melt, Kdj = the

mineral/melt distribution coefficient of the trace element in the melt (Rollinson, 1993). Kd values

are from Nash and Crecraft (1985) except for quartz values for Sr, Zr and values for any element

for the metamorphic mineral sillimanite for which values of 0.001 are assumed on the basis that

the probability of substitution of these elements into these minerals is low.

All figures were done in Adobe Illustrator 10 and printed from Adobe Acrobat 5 on an

Epson Stylus 7401 printer. All PetMin3 chemical models were adapted from screen capture

images and redrawn in Illustrator 10. Partial melt models from the spreadsheet and spider

diagrams were plotted using Kaleidagraph 3.5© and redrawn in Ilustrator 10©. Photographs

were scanned using an AGFA Snapscan touch scanner, imported into Adobe Photoshop 6© and

then imported to Illustrator 10 for annotation.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3

GEOLOGY OF THE NORTHERN MASSABESSIC GNEISS COMPLEX

General Geology

Bedrock mapping was done from 2000-2003 at a scale of 1:24000 using U.S.G.S. 7.5-

minute quadrangles as base maps. Areas covered included the Candia 7.5-minute quadrangle,

and half of each of the Northwood, Barrington and Mt. Pawtuckaway 7.5-minute quadrangles

(Fig-2).

Elevation is highest in the Northwood quadrangle (total relief of 275 m), and generally

decreases to the southeast in the Barrington quadrangle (total relief of about 125 m, Mt.

Pawtuckaway quadrangle total relief of around 190 m) (Fig. 9). The combined total relief for the

four quadrangles is about 320 m. The region has been multiply glaciated and is almost entirely

covered by less than a meter to greater than 30 m veneer of glacial ground moraine. The majority

of the hills are southeast-trending (bedrock-cored) drumlins. Outcrop is sporadic and is limited to

the lee side of roche moutonnees and higher terranes subjected to wind and water erosion,

streambeds and road cuts.

The Candia quadrangle is underlain by predominately migmatites and granites (Fig. 10,

Plate 1). Most lithologies occur in northeast striking belts. The preponderance of the migmatites

lessens to the northeast in the Northwood and Mt. Pawtuckaway quadrangles, and they disappear

altogether within the Barrington quadrangle. The granites adjacent to the north of the migmatites

are weakly foliated and contain small areas of metasedimentary rocks. In general the

granites north of the Campbell Hill-Hall Mt. Fault are unfoliated and contain within them

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Gossville Northwood Barrinqton

Candia Pawtuckaway Epping

Elevations in feet (m) ■ I 0-100 (0-30.48) H 601-700(183.1-213.4) ■ i 101 -200 (30.8-70) 701 -800 (213.6-244) ■ I 201 -300 (61.3-91.4) 801 -900 (244.1 -274.3) H 301-400 (91.7-122) 901-1000(274.6-304.8) index map ^ 9 401-500 (122.2-152.4) 1001-2000 (305-609.6) C=3 501-600(153-183)

Figure 9. Elevation diagram of the Gossville, Northwood, Barrington, Candia, Mt. Pawtuckaway and Epping, 7.5-minute quadrangles. Study area is outlined by white dashed line. Digital elevation model data used to build this map were provided by NH GRANIT. 31

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. GSZ

■ / / v ._ / /

CHHMF rIT I •

7.5-minute Quadrangle Key

1. Candia 2. Southern half of Northwood 3. Northern half of Mt. Pawtuckaway 4. Southern half of Barrington

Explanation

r-----1 Undifferentiated rocks of the Faults t-----i Pawtuckaway ring Dike Complex CHHMF-Campbell Hill-Hall Mt. Fault IIPegmatite sills GSZ- Gulf Shear Zone Granite, lined where foliated FHF-Flint Hill Fault ?-FHF, Calef Fault, Barrington Fault i "1 Metasedimentary rocks L__ 5 and granitic sills of the transition zone Metasedimentary rocks of the Central Maine Terrane (Ssf, Sp) Metasedimentary rocks of the Merrimack Group (SOb) Migmatites of the MGC MGC type 3 MGC type 2 MGC type 1

Figure 10. Lithology map of the study area showing spatial distribution of rock types and faults. See plate 1 for detailed geologic map and scale.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. metasedimentary rocks of the Central Maine Trough. The metasedimentary rocks in the study

area are rusty-weathering schists with lesser light brown-weathering calcareous fine-grained

metasandstones and quartzofeldspathic metawackes. The only primary structures preserved in

any of the metasediments appear along the southern border of the Candia quadrangle on a hill east

of Massabesic Lake (location C234, Plate 2). They (arguably) show topping direction toward the

southeast. Reconnaissance mapping in the Derry quadrangle south of this location indicates that

the southern exposure of the rusty weathering schist in the Candia quadrangle is a northern

extension of similar lithologies lying between the MGC and the Flint Hill Fault. The amount of

exposed metasedimentary rocks increases from the Candia into the Northwood and Mt.

Pawtuckaway quadrangles. The southern half of the Barrington quadrangle is largely granite with

metasedimentary rocks exposed around fault zones. The metasedimentary rocks in the

Barrington quadrangle are similar to those found in the other quadrangles, except in the southeast

comer where the metawacke lacks pelite layers.

All of the migmatite types exhibit diffuse boundaries with each other. In places Type 2

migmatites appear as enclaves within Type 3, and the leucocratic zones of the enclave appear to

flow into the surrounding migmatite. The Type 3 migmatites are often in diffuse contact with

granites and pegmatites, but in a few locations the granite and/or pegmatite bodies are clearly

crosscutting structures. The northernmost exposure of the MGC migmatites (location B233, Plate

2) is pavement outcrop. A group of three glacially polished outcrops show a Type 3 migmatite

(Fig. 11) with flow foliation and grain reduction of several cm scale K-spar grains.

The granites are fine to coarse-grained, gray to white binary granites, with the only

distinguishing factors being rare crosscutting relationships, a weak foliation and the presence of

magnetite, pink K-feldspars and large isolated K-feldspars. The foliation and the isolated

K-feldspars make the granites similar in appearance to Type 3 migmatites. Saddleback Mountain

(Northwood quadrangle) is composed of schist and metawacke intruded by pegmatite sills. The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 11. Picture of a glacially polished pavement outcrop of sheared pegmatite, the white board scale is ~7.6 x 23cm. The potassium feldspars are 10-20+cms in diameter. They are sheared and boudinaged and resemble sigma grains (macro scale). It is this location where the MGC surface expression terminates. 34

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. metasedimentary rocks are migmatized rusty-weathering staurolite, garnet schists that are

intercalated with a rusty-weathering thinly layered metawacke with minor calc-silicate. The

juxtaposition of these lithologies is apparent along the southern flank of the mountain and

throughout the quadrangles in this study. The pelite-metawacke layered rocks occur as erosional

remnants with pegmatite outcrops (north of the MGC) and granites south of the MGC. These MS

remnants are in contact with the dip slopes and overhangs of the sills with areas of no rock

exposure between the sills. It was the presence of these metasedimentary rock remnants that led

to the hypothesis that these areas were covered by the metasediments, intruded by the sills and

then weathered such that the more weather -resistant sills protected vestiges of the MS layers.

The areas containing this rock assemblage are mapped as transition zones (Plate 1).

Structure

Structural elements that were collected and recorded include biotite foliation, strike and

dip of planar surfaces, axial planes and fold axes of minor folds, trend and plunge of mineral

lineations (penetrative and surficial), strike and dip of joint surfaces and the orientation of mafic

dikes, pegmatite dikes and sills. Primary sedimentary structures (cross beds, facing directions)

were not recognized anywhere in the four quadrangles except for location 234 in the Candia

quadrangle.

In the mapping completed for this and previous projects (Kerwin 2000,2001), one

foliation surface was found to dominate the area with a secondary surface only locally developed.

In this work the dominant foliation is designated as S2, following Eusden's (1984) designation of

similar rocks mapped to the north and east.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Foliation

Metasedimentary rocks

Biotite foliation and the orientation of a local secondary foliation were measured in the

metasedimentary rocks. The stereonet plots for foliation in the Candia quadrangle indicate the

MGC and surrounding areas were broadly folded into a shallow northeast plunging anticlinorium

(Kerwin, 2001). The Northwood, Barrington and Mt. Pawtuckaway quadrangles (Fig. 12 and

12d.) also show the shallow northeasterly-plunging fold in a regional sense. The foliation

recorded from the Northwood quadrangle area, with the exception of the Saddleback Mt. area,

indicates a gentle regional fold in which the calculated fold axis (6) plunges gently to the

northwest. The foliation for the metasedimentary rocks in the Barrington quadrangle is similar to

the Candia data except the Barrington foliation is steeper (Fig 12). The average foliation plane

for each stereonet is represented by the bold black great circles. The average foliation plane for

the Candia and Northwood quadrangles are very similar (250 30NW, 255 38NW respectively).

The Barrington quadrangle average foliation is dissimilar in that it is represented by a plane that

dips steeply to the southeast (32 67SE) and not to the northwest. When the Northwood and

Barrington foliation data are plotted together (Fig. 12d) they very closely resemble that of the

Candia quadrangle.

The fold axes and axial plane data (Fig. 12a and b) for the Candia quadrangle show an

average axial plane dipping moderately to the northeast with the fold axes girdling the stereonet.

Figure 12a is not similar to figure 12c (fold axes for the remainder of the study area), which is

probably because most of the Candia metasedimentary rocks are from minor southwest verging

folds in proximity of the CHHMF.

Migmatites

The migmatite foliations are plotted in figure 13. In all quadrangles the migmatite

foliation reflects the orientations of the metasedimentary rock foliation with little variation. They

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. n = 9 Saddleback Northwood MS ioliation T he G ulf V MS biotite foliation n = 5 3

Barrington MS foliation

Quadrangle index

1. Candia Candia MS foliation 2. N orthwood o n = 6 9 3. Mt. Pawtuckaway 4. Barrington

n=25 n=51 n=26 n=99 Figure 12. Poles to metasedimentary rock foliation for the Northwood, Barrington and Candia quadrangles. The average foliation planes for each plot are in black. The rocks in the nothem Mt. Pawtuckaway quadrangle are primarily migmatites so there is no Mt. Pawtuckaway stereonet for the metasedimentary rocks. Figure 12a are the fold axes and 12b are the poles to the axial planes for the Candia quadrangle, average axial plane shown as black dashed line. 12c are the fold axes recorded for the remainder of the study area. 12d is a synoptic diagram for the foliation of the Northwood and Barrington quadrangles. Average fold axes are shown with black stars. See figure 10 for explanation of geologic units. ^

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. » ► Saddleback migmatite “ Northwood migmatite n=19 foliation n=10 foliation

Bamngton migmatite n=5 foliation

Quadrangle index

1. C andia 2. Northwood 3. Mt. Pawtuxckaway 4. Barrington

Candia migmatite foliation Mt. Pawtuckaway n=20 migmatite foliation

n= 59 n=17

Figure 13. Poles to biotite foliation for the migmatites in the study area. The stereonets show the data from each quadrangle. The average foliation planes for each plot are in black. Figure 13a fold axes from the Candia quadrangle and 13b fold axes from the other three quadrangles. Black stars represent the average fold axis. See figure 10 for explanation of geologic units.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. define the limbs of a broad anticlinorium with a regional axis plunging to the northeast with the

exception of the Saddleback migmatites. A few foliations that were considered to be from an Sn+1

generation plot in the same areas on the stereonets making the distinction tedious. It is not clear

if there is a reliable way to distinguish flow foliation from solid state metamorphic fabric, and in

the case of the MGC migmatites it could be either or both. Fold axes recorded for the migmatites

are scattered (Fig. 13a and b). The average fold axes for each plot are both moderately plunging,

trending northwest (Fig. 13a, Candia) and northeast (Fig. 13b, remainder of the study area). The

fold axes for the remainder of the study area (Fig. 13b) are consistent with the fold axes from the

metasedimentary rocks outside of the Candia quadrangle (Fig. 12c).

Foliated granites and pegmatites

In all of the quadrangles there are numerous granite and pegmatite intrusions. Typically,

just northwest of and within the MGC migmatites, they are foliated (Fig. 14). Whether the

foliation is a flow foliation or the result of tectonism is unknown. However, it parallels the

regional northeast trend of foliation.

Faults

Campbell Hi 11-Hall Mountain Fault

Campbell Hill, for which part of the fault is named, is located in the Manchester North

quadrangle west of the Candia quadrangle. The summit is 600 feet above sea level with 293 feet

of local relief and is composed entirely of glacially polished white bull quartz. The CHHMF is a

northeast striking fault with evidence of a lengthy history. The fault trace was previously placed

on the mapped contact between Berwick and Littleton Formations and is thought to have gone

through an extensive period of dextral movement culminating in upper crustal fracturing and

juxtaposition of two distinct terranes (Lyons at al., 1997). Whitaker (1983) mapped it to pass by

Saddleback Mountain in Northwood as shown by Freedman (1950). The CHHMF is a 0.5-1 km

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Q uadrangle index

3 1. Candia Foliated granite from quadrangles 2, 3 and 4 2. Northwood 3. Mt. Pawtuckaway 4. Barrington

Candia foliated granite

Total foliated granites

Figure 14. Stereonet plots of the poles to foliation of the foliated granites in the study area. The average foliation plane is bold. The Candia granites are foliated in a similar orientation to the regional fabric in that area. The other granites are similar with minor variation. See figure 10 for explanation of geologic units.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. wide zone of highly fractured and brecciated, silica impregnated and vein-filled country rock. It

crops out as a series of bull quartz hills that are easily traced from Milford, NH through Candia

on Hall Mountain and into the Northwood quadrangle (Fig 10). Fresh outcrops at the base of

Saddleback Mountain within a utility trench, exposed a sheared granitoid rock with a well-

developed S-C fabric that record a ductile phase of the CHHMF prior to brittle deformation.

Unfortunately the exposure was in such condition to prohibit any orientation of the fabric and

although samples were retrieved, no sense of motion was derived. Early folded quartz veins on

Hall Mountain (Location C 33) indicate a sinistral sense of motion and are cross cut by folded

quartz veins indicating dextral motion. In Northwood it crops out along Route 43 as white bull

quartz in pegmatite (Belanger, 2003) and Kerwin (this project). Northeastwardly beyond Lucas

Pond, exposure is lost in the pegmatites and granites of the Barrington Pluton. Reconnaissance

mapping in search of the fault in the northern half of the Barrington quadrangle by Misery Hill

and Stonehouse Pond, was not productive. Conversations with Peter Thompson (personal

communication, 2004) and mapping by Freedman (1950) indicate that a silicified area in the

northern part of the Barrington quadrangle may be worth further consideration. Interpretation of

digital elevation models (Fig. 15) and strike of the fault at the last point of exposure suggests

extending the fault trace from the Lucas Pond location to the northeast connecting the CHHMF

with the Hint Hill Fault in the Barrington quadrangle.

Hint Hill Fault

The Hint Hill Fault (FHF) crosses the southeast comer of the Candia quadrangle, passing

northeasterly through the Pawtuckaway and Barrington quadrangles (Plate 1, Fig. 10). Robinson

(1989) interpreted it to have had three phases of brittle deformation and one episode of

silicification. Post-silicification slickenlines indicate east side down as the last sense of

movement of approximately one kilometer offset (Allard, 1998). The surface expression of the

FHF is a series of hills of white bull quartz pods similar to those of the CHHMF, but without the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Index to 7.5 minute quadrangles

Figure 15. Hillshade model of the study area and adjacent quadrangles. The model shows the sun as if it were shining from an azimuth of 0 degrees and an inclination of 45 degrees. The trace of the Campbell Hill-Hall Mt. Fault with the north-striking ductile shear zone are shown in yellow and the trace of the Flint Hill Fault shown in red. Hillshade models are useful for locating fracture zones. Notice the area southeast of the Flint Hill Fault. It appears fractured and indicates the extent of the fault zone. Digital elevation data used to build this map were provided by NH GRANIT.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.5-lkm wide zone of silicified country rock. Oriented thin sections of sheared granites along

theFlint Hill Fault (location B208, see Plates 1 and 2) indicate that it also had a period of ductile

movement. Sheared quartz veins and cm-size sigma-like grains at the outcrop scale indicate a

sinistral sense of motion for the FHF during the earlier ductile phase.

Gulf Shear Zone

The Gulf Shear Zone (GSZ) is mapped as a splay of the CHHMF (Fig. 10) in the

northwestern comer of the Northwood quadrangle. It deforms Perry Mountain and Smalls Falls

Formations into sub-vertical layers that are locally silicified. Oriented thin sections have sigma

grains indicating an oblique sinistral movement (location N 128) along this zone.

Other

In the southeast comer of the Barrington quadrangle two faults were mapped (Plate 1).

They are silicified and contain box structures. The metasedimentary rocks crosscut by these

faults is often a pale greenish or gray color. These faults are inferred to be splays of a fault that

trends northeast along NH Route 125.

Mineral and crenulation lineations

Mineral lineations (stretched garnets, quartz and sillimanite pods) were measured. The

data are represented in figure 16. In the Candia quadrangle the mineral lineations are variable.

The Barrington quadrangle mineral lineations (n=4) trend to the northeast, but in the Northwood

quadrangle they trend to the northwest.

Crenulation lineations in the Candia quadrangle trend to the northwest, and to the

northeast in the Barrington quadrangle. In the Northwood quadrangle they are variable. Very

few lineations of both kinds were recorded from each quadrangle so they are combined in two

synoptic stereonets (Fig. 16a and b) for comparison. These synoptic diagrams are very similar to

the Candia fold axes in figure 13a.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. | n = 4 ■ __ ) ^ n = 1 0 Northwooil crenulation Northwood mineral lineation ~ lineation

n = 6 ■■ ' , . n = 4 Barrington crenulation Barrington mineral Candia mineral lineation lineation Q uadrangle / index

1. Candia 2. Northwood 3. Mt Pawtuckaway ■ 4. Barrington

* MU Candia crenulation lineation n = 6 a. n = io

Figure 16. Plots of mineral and crenulation lineations. Figures 16 a and b are total crenulations and mineral lineations respectively. See figure 10 for explanation of geologic units.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Slickenlines

Slickenlines are non-penetrative lineation caused by the friction created by faulting.

Figure 17 shows the slickenline data in terms of a rose diagram and map symbols. The majority

of the slickenlines trend northwest southeast and a minority northeast southwest. They are

interpreted to reflect a period or periods of brittle faulting. Not all of the slickenlines occur along

mapped faults raising the possibility that more faults exist than are currently recognized.

Slickenlines also are indicators of the latest movement of a fault and often contain steps showing

the sense of motion. Thirteen locations were found where the sense of motion was preserved

(Fig. 17). The sense of motion determined by most of the slickenlines is that the latest movement

of the MGC was up relative to surrounding rocks as a structural horst.

Mafic dikes and quartz veins

The study area is crosscut by numerous mafic dikes which presumably were emplaced as

the result of Mesozoic extension of a supercontinent (McHone, 1996). The orientations of the

dikes are shown in figures 18a and b. The rose diagrams show a northeasterly orientation of

mafic dikes, which is consistent with an extensional regime in which the maximum extension

direction was northwest - southeast. There are two locations within the study area that are

intruded by multiple mafic dikes. One is northeast of Massabesic Lake (Plate 1) in which a

several meter-wide exposure is crosscut either by multiple dikes over time or a very wide dike in

which mafic material was intruded along the axis of a widening fissure. The other area is on the

southwest side of Walnut Hill (Plate 1). This area is intruded by several different dikes with

widths varying from five centimeters to two meters. They also vary in grain size and

composition. Some are fine-grained, and others have 2 to 3 centimeter sized amphibole

phenocrysts.

In the Barrington and Mt. Pawtuckaway quadrangles a mafic dike is exposed that is

several meters in width and is magnetic. It is likely an extension of the Onway Dike, which

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Eqittlraa

1 ffMM t?ti .. ! circle=12%

/ / Quadrangle / t,* / / ./ index

A 1. Candia 2. Northwood 3. Mt. Pawtuckaway 4. Barrington

Location Sense of motion C 2 southeast side down C78 northw est side dow n C98 north side down C119 southeast side dow n C 125 w est side dow n B176 south west side down B238 south west side down B247 north west side down B181 south side dow n N 62 north side dow n N65 north west side down N94 north west side down N104 southwest side up Figure 17. Map and rose diagram showing orientation of slickenlines within the study area. See figure 10 for explanation of geologic units. The rose diagram indicates the primary direction of motion recorded by slickenlines was predominately northwest southeast. The table above lists the locations and sense of motion (shown in red on the map) where recorded.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Candia mafic dikes Candia bull quartz veins

E qual A rea

N= 13 north bqv N = 4 Mt. Pawtuckaway Northwood bull mafic dikes quartz veins

Figure 18. 18a. Rose diagram showing the orientation of the measured mafic dikes within the Candia quadrangle. 18b are the mafic dike orientations for the Mt. Pawtuckaway quadrangle. 18c and d show the bull quartz vein orientation data for the Candia and Northwood quadrangles, respectively.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. strikes northeast from Massachusetts, described by Freedman (1950) and Sundeen (1992) in

detail. The discontinuous exposures of the Onway dike were mapped from the southern end of

the Barrington quadrangle along Route 4 south of Mendums Pond to just north of Swains Lake.

Bull quartz veins (Fig. 18c and d) are exposed in several places, especially along inferred

fault zones. A small hill in the Candia quadrangle (location C218, Plate 2), tentatively named

Chester Hill, has ~0.5 -1 km2 exposed pavement outcrop of predominately Type 3 migmatites.

The migmatites are cross cut by several 5cm to 15cm mafic dikes that are following several en-

echelon fracture sets. The Chester Hill location also exposes several parallel bull quartz-filled

veins. The veins are 1mm up to several cms thick and show east-side-down offset of several

centimeters. Individually the offset doesn't seem significant, but over the 0.5 km of third

dimension exposure along Route 101, the cumulative offset was estimated to be more than a

meter. It is probable that these locally small offsets may collectively represent kilometer scale

offset over the distance of several quadrangles.

Summary and interpretation

Foliation in the MGC defines an asymmetrical anticlinorium (Fig. 12, and 19) with the

southeast limb of the structure dipping more steeply than the northwest limb. The anticlinorium

plunges slightly to the northeast. The foliation data from the Candia quadrangle and the synoptic

diagram (Fig. 12d) indicate the structure is consistent from the southwest into the northeast of the

study area. Axial plane data from the Candia quadrangle (Fig 12b) are different than those from

the rest of the study area. They were predominately collected from near the CHHMF and a few

from the metasedimentary rocks between the MGC migmatites and foliated granites (Plate 1).

The Candia folds record minor southwest verging folds which may be related to: emplacement

mechanisms of the local granites or possibly or sinistral ductile motion of the CHHMF. The GSZ

is interpreted to be a splay of the CHHMF, and if so, also shows sinistral motion for the ductile

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. GSZ

CHHMF

7.5-minute Quadrangle Key

2 1. Candia 2. Southern half of Northwood 3. Northern half of Mt. Pawtuckaway 4. Southern half of Barrington

Explanation

1------1 Undifferentiated rocks of the Faults L----- 1 Pawtuckaway ring Dike Complex CHHMF-Campbell Hill-Hall Mt. Fault m Pe§matite sills GSZ- Gulf Shear Zone l^rfffl Granite, lined where foliated FHF-Flint Hill Fault f----- 1 Metasedimentary rocks 7-FHF, Calef Fault, Barrington Fault i------1 and granitic sills of the transition zone m Metasedimentary rocks of the Central Maine Terrane (Ssf, Sp) m H I Metasedimentary rocks of the Merrimack Group (SOb) Migmatites of the MGC

■ I MGC type 3 Structural form symbols ■ 1 MCjC type 2 ------A------Strike and dip of foliation [ J MGC type 1 Horizontal foliation Trend and plunge of antiform

Figure 19. Generalized foliation pattern on bedrock geologic base map. See Plate 1 for detailed geologic map and scale.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. phase of the CHHMF. The fold axes collected in Candia (Fig. 12a) define a girdle that implies

that the minor folds recorded in figure 12b may have been refolded during the structural arching

of the anticlinorium. The primary migmatite foliation is sub-parallel to the metasedimentary

foliation with average foliation planes varying a little (Fig. 20). The calculated B axes for these

two stereonets are in close agreement implying that the migmatite foliation is relic foliation from

the earlier metasedimentary protolith or that these rocks were folded simultaneously. The biotite

foliation plotted in the Saddleback Mountain area is north of the CHHMF and appears to have

been unaffected by structural events taking place south of the fault.

The foliated granites are also parallel to sub-parallel to the regional biotite foliation of the

metasedimentary rock and migmatites. This suggests that the granites intruded along the

structural arch into areas of relatively less pressure, possibly at the onset of regional dilation,

resulting in a flow foliation mimicking the regional trend.

The rocks in and around the FHF also show evidence of ductile movement based on thin

section interpretation. The ductile movement of both faults is interpreted to be earlier than the

brittle motion based on the preservation of slickenlines on the rocks. The slickenlines show MGC

upward movement as a structural horst and brittle extension of the crust created fracturing and

emplacement of bull quartz as fracture filling. All were crosscut by the intrusion of mafic

magma.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Equal Area

C.l.= 2.0sigma

C J .= 2.0 sigm a

Figure 20. Synoptic diagrams of pole to foliation planes for all of the metasedimentary rocks (Fig. 20a) and migmatites (Fig. 20b). Average foliation planes are shown in bold black lines

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4

PETROGRAPHY AND PETROLOGY

Introduction

The rocks described in this chapter, with a few exceptions, are from the Candia

quadrangle. The sample and sample locations are listed using the first letter of the quadrangle

from which they came followed by the sample number which corresponds to a locality within the

quadrangle (Plate 2). It was apparent in the field that distinguishing the rocks could not be done

effectively using hand samples; therefore, micro analyses were needed to assist in this process.

Many rocks were examined, but this chapter focuses on the samples that were also subjected to

chemical analysis. Mineral abbreviations are listed in Table 4 and modes for all rock types are

summarized in Table 5.

General descriptions

Metasedimentary rocks

Metawacke

The metawackes are primarily fine-grained and tan or purplish weathered with quartz,

plagioclase, biotite ± actinolite with lesser calc-silicate layers (Fig. 21). The biotites are often

retrograded to chlorite, the plagioclase grains are rarely twinned, and the actinolite grains appear

relic. The quartz grains generally show undulatory extinction, and many of the plagioclase

twinning planes and micas are bent. Samples C15, C19, C23, C61 and C154 have grain boundary

textures that are cataclastic.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4 Mineral abbreviations (after Kretz, 1973)

Bt Biotite Me Microcline

Chi Chlorite Ms Muscovite

Grt Garnet PI Plagioclase

Kfs K feldspar Qtz Quartz

Ky Kyanite Sil Sillimanite

Liq Silicate melt St Staurolite

Pelitic schists

The pelitic schists are predominately rusty weathering and medium grained with quartz,

plagioclase, muscovite ± potassium feldspar, garnet, sillimanite (mostly fibrolite), staurolite and

tourmaline (Fig. 22). The plagioclase is sericitized, in some cases completely altered and often

untwinned; the quartz is undulatory and reactions can be seen along quartz and feldspar grain

boundaries. The potassium feldspar is perthitic and occasionally included within larger unaltered

plagioclase crystals. The sillimanite occurs as both prismatic and fibrolitic grains. Muscovite is

mostly primary with some secondary occurrences. Most of the primary muscovite has reacted

and in one instance forms a radial pattern and in another it is undulatory. The quartz grains often

look cataclastic with highly irregular grain boundaries. Sample C171 and C128 look

porphyroclastic with some quartz grain recrystallization; C234 has sillimanite

inclusions/intergrowths within plagioclase and muscovite crystals. Samples C23 and N4 have

cross micas.

Migmatites

The migmatites are medium to coarse-grained with quartz, plagioclase, biotite ±

potassium feldspar, muscovite and opaques (magnetite and/or pyrite). The main differences

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 21, outcrop pictures of tan weathering (21a, location N51) and purplish weathering (21b, location N10) metawackes. Both pictures were taken in the Northwood quadrangle but are representative of the metawacke units throughout the study area. Picture 21b is from the top of Saddleback Mt. and is in contact with a rusty weathering pelitic schist. The white board used for scale is 3x9 inches (7.5x22.7cm).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 21c. and d. Photomicrographs of metawacke. Scale in both pictures equals 2mm. 21c is plane polarized light and 21d is crossed polarized light. Qtz=quartz Bt=biotite and Pl=plagioclase. These pictures show the typical texture and grain size of the metawacke units. 55

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 22. Outcrop pictures of typical pelitic schists. Both pictures were taken in the Barrington quadrangle. 22a (location 206) shows the rusty weathering nature of the schists and the layering style. 22b (location 202) shows a rusty weathering schist (left-hand side of picture) in contact with a purplish weathering metawacke (right-hand side of picture). The gray area to the far right is a small pegmatite intrusion. White board used for scale is 3x9 inches (7.5x22.7cms).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 22c. and d. Photomicrographs of typical schist. Scale equals 2mm. 22c is sample N4 in plane polarized light and 22d is crossed polars. Bt=biotite, Qtz=quartz, Grt=garnet, Ms=muscovite, St=staurolite.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. between the three types of migmatites (Fig. 3) are the grain size and the occurrence of large

potassium feldspar (crysts, blasts?) and the presence of magnetite. Also, the amount of biotite

decreases from Type 1 to Type 3, and Type 3 foliation is disrupted to non-existent. The grain

edges are rimmed by a mineral film and grains are embayed and included by relic minerals (Fig.

23). The quartz is undulatory and the potassium feldspars are typically microcline. Plagioclase

often is sericitized, sometimes along twin boundaries. The migmatites were point counted and

plotted on QAP diagrams (Fig. 24). Four out of six samples plot in the tonalite field of

Strekeissen (1976) and two plot in the granite field. The two migmatites that plot in the granite

field are from locations where granites and migmatites were in contact.

Granites

The granites are medium to coarse-grained, gray to yellowish weathering with quartz,

plagioclase, potassium feldspar, biotite ± muscovite and some opaques (Fig. 25). Some of them

have a tectonic fabric, but the majority do not. A few of the granites have textures that in thin

section indicate melting along grain boundaries (see Mineral Textures, p. 62), and in one instance

melting is evident at the macro scale (Fig. 26). The quartz is undulatory and the potassium

feldspar is perthitic. Grain boundaries are highly irregular. Plagioclase in sample C 119 is partly

twinned and partly untwinned in the same crystal. Most of the plagioclase has some degree of

sericitization and in some cases has microcline inclusions. Many of the microcline and

plagioclase grains also have myrmekite inclusions

The granites were point counted between 500 and 1000 counts and then plotted on

quartz-alkali feldspar-plagioclase (QAP) diagrams (Fig. 27) to show variability with respect to

the felsic minerals and to biotite and muscovite (Fig. 25b and 25c, respectively). Figure 25a

shows that the "granites" are a combination of granodiorites and granites. Two samples, granite

sill Pg-2 and sample Cg 26-2 fall outside of the two major categories. Sample Cg 26-2 is also

distinct from the other granite samples in chemistry as is discussed in Chapter 5.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 23. MGC type 2 migmatite in plane light (top) and crossed polarized light (bottom). Notice the coarse-grained plagioclase include finer-grained plagioclase in other orientations, and quartz. Scale is 2mm.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Q

Quartz-rich granitoids

G ra n o - \A diorite \

Granite

|— | C3 218 ■ C 2 T H ^ Granites from figure 27 I I C 2 213 ■ C 2 119 I | C3 116 I I C3 119

Figure 24. Modal quartz-alkali feldspar-plagioclase of migmatite generated from point counting thin sections (500-1000 counts). Four of the six samples plot in the tonalite field of Strekeissen (1976) and two plot in the granite field. The two samples within the granite field were in contact with granites that were also sampled and point counted. The quartz, alkali feldspar, plagioclase compositions of the intrusions in the study area are shown for reference.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 25. Photomicrographs of Cg 90 plane light (top) and crossed polarized light (bottom). The granite textures are coarse grained, feldspars with smaller quartz inclusions. Plagioclase sericitization is not uncommon, myrmekite occurs at triple point boundaries, and biotite is often retrograded. Bt=biotite, Qtz=quartz, Pl=plagioclase. Scale equals 2mm. 61

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 26. Picture of a hand sample (Pg 304) of a granite in the Mt. Pawtuckaway quadrangle (refer to Plates 1 and 2). The granite has a foliation fabric and areas of biotite mica surrounded by quartz and feldspar leucocratic zones The scale is in cm. The biotite and leucocratic lenses appear flat along a cut made perpendicular to foliation and look circular in cuts made parallel to foliation (not shown).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Quartz

Granodiorites

□ C g T H Granites □ Cg21 Quartz -nch Cg 119 ■ Cg 90 granitoids B Pg-1 I I P 3 0 4 ■ I Cg WH H Pg-3 gB □ Quartz-rich □ Mg granitoid Grano- diorite Tonalite Granite □ C g 2 6 -2

a. Alkali Plagioclase feldspar

Figure 27. Granite thin sections were point counted between 600 and 1000 points. Figure 27a. are granite data plotted on a QAP diagram. Out of thirteen samples, five plot in the granite field, of which two are sills (Pg-1, Pg-3) from the Pawtuckaway quadrangle. Six samples plot in the granodiorite field and one in each of the quartz-rich granitoid and tonalite fields. Figure 27b. shows the same samples with biotite added to the calculation. The size of the triangle represents the relative abundance of biotite in each sample. Figure 27c. shows the samples with the relative abundance of muscovite as above. For samples with no muscovite the symbols are dots instead of triangles.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. P X N3 0.6 6.6 15.4 34.8 21.6 20.4 X X C C 15 XX XXX XXX X X X X X X C C 85 X XXXX XX XXX X X X 35 1.6 6.4 21.2 33.6 X X X 23 p+f f 0.6 B216 095 C23 x =mineral observed in thin section, sample was not point counted p=prismatic f=fibrolitic XXX XXXXXX X 21 11 56.6 19.6 20.8 C C 170 50b mast rd 171-1 N73 Cl C 169 f f XX XXX X XXX X 4.6 0.64.9 1.6 25.8 17.6 45.2 37.2 33.5 SOb rte 101 PI PI PI 42 Bt Bt St Bt St Ms Ky Sil Ms Sil 15 f+p Qtz Kfs 0 Qtz Kfs Qtz Grt Kfs 5.4 other Act Schists sample # sample # N4 B224 sample # C70-b C 19 C234-1 C71-1 xirmaline metawacke M etasedim entary rock Table 5. Mineral modes. See Plates and for 1 2 locations andTable 4 for mineral abbreviations.

Table 5 continued. s -w *C £> s 8 > 1“ O CM S ^ * ®®MS TO ^ E s o I- _ ^ CM - u I- ^ CO CO A U K - u t- m a "t ® iO ® ^ » C U - r U CO » a vD [fl ,J CM J , l f [ D v . O N \ m f- f m c\l N \ > C L LO LO to CO CO co *N CO 1 > ^ . Q !Z o> W # C f\J r- o - r u J \ f t— CO 0 0 N CM CO r r CO > > . - D r ^ 1/5 >; o m m S b ; ^ co ^ ^ r; oo ■3. J» C s: )0 3( £ , D J - OD N O f ) O ? 7 w m * cm j \ ( m l c a> O O C> ' - M CT> LO 2 u (riCO N ^ *— CD ' M LO “ M . . . CM O L O CM CM CM CM O L O M- Q_ E ■ M CM CM . ■ • ) D rvi o U) > 3 *-• m “ £ ?3 °> C Lft t— t t f L CM ) O j x x x x x x x jj, 1 ' O M M CO CM CM d ; M 0 0 0 0 n i £ O L CM *""" 65 CD MLO CM CO E ? >* § >s E 8 cu cr - LO 8-

£ O) m • . . oo . . • m O) <£ M • ^ CM O £ < ) O 0 ? N <\J00 “?

other myrm .9 myrm x seer 16.2 seer.! Mineral Textures

Mineral grains with highly irregular grain boundaries and relic mineral

inclusions/embayments within larger recrystallized minerals, quartz and plagioclase within twin

planes (Kerwin, 2000, p. 29) and "wormy" myrmekite occur in the migmatites (Fig. 28). These

textures can be produced by reactions of minerals and melts (Dell’Angelo and Tullis 1988;

Ashworth and McLellan, 1985; McLellan, 1983; Nyman et al., 1995; Fitzsimons, 1996). The thin

sections of the migmatites also contain cross micas and kyanite relics (Fig. 29) embedded in

sillimanite that have been interpreted to be evidence of decompression melting (Thompson and

Connolly, 1995).

Textures in the granite Pg 304 (Fig. 26) suggest at least two melt periods, one in which

the granite was made and a second in which the granite started to remelt, resulting melt zones of

quartz and feldspars encasing minerals of higher melting temperatures (biotite).

Petrology

Temperature and pressure estimates range from 506-590° C and 3 kbar just outside of the

MGC to 690° C and 9 kbar within (Allard, 1998; Larson, 1999; Schulz, 2004, see figure 6). The

pelitic schists are equivalent to that of Allard's Watson Hill (1998) sub-member of the Berwick

Formation, and that assemblage will be used for discussion of the petrology as a protolith for the

migmatites. The textures seen in the thin sections revealed severed possible reactions, which can

be used to track the petrology up through migmatization (Fig. 30). East of Watson Hill, staurolite

is being consumed by the prograde reaction [1] Mu + St = Grt + Bt + Sil + H20 ± Qtz (Allard,

1998). Reaction [1] also creates sillimanite, which is observed in many samples. Muscovite is

then broken down through the reaction [2] Mu + Qtz = Sil +Kfs +H20.

The alumino silicate textures may also give some insight with regards to temperatures

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a. Sample C3 218 b. Sample C 20*1

Sam ple C 2 13 Sam ple Cg TH

Figure 28. Crossed polarized light photomicrographs of various lithologies and textures showing evidence of melting. Figure 28a. is a MGC Type 3 migmatite from Candia, 28b. is a metasedimentary rock from Candia, which has a microcline relic within a plagioclase crystal. 28c. is a MGC Type 2 migmatite from Candia and 28d. is a granite from Tower Hill in Candia. The slides show inclusions within recrystallized grains, irregular grain boundaries and myrmekitic textures within feldspar. Scale = 2mm

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 29. Crossed polarized light photomicrographs of three metsedimentary rocks that display textures cited as being caused by decompression melting (outlined in white). Those textures include kyanite relics embedded in sillimanite (Fig. 29a.), and cross micas in figures 29b. and c. Scale is 2mm. 68

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A1A Sil

Ms +P1, Qtz

, Kfs

G xt Ms a. +P1, Qtz KO FMO

Grt

KO FMO N3

Ms [1] Mu + St = Grt + Bt + Sil + H20 ± Qtz +P1, Qtz

Grt

Ms ■------K O FMO +P1, Qtz N 4 2 , Kfs [2] Mu + Qtz = Sil +Kfs +H20 [3] Mu + H +Qtz = Kfs +Sil + melt See melt textures p. 67 KO A3& FMO Sil C 234

+P1, Qtz

, Kfs

Grt Bt [4] Bt + Sil + PI + Qtz = Grt +Kfs + melt FMO k2° See decomposition melt textures p. 68 C 2 T H

Type 3 migmatites and granites Figure 30. Graphical representation of the reactions discussed in the text. 30a is an index K O, A1 O , FeO +MgO (FMO) ternary diagram showing the mineral phases in the rocfes in moP%. Plagioclase and quartz are present in all samples. Sample C2 TH has Mu listed as a mineral phase but it is a minor occurrence and is interpreted to be secondary. See Table 5 for mineral assemblages for samples listed below each triangle. 69

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and pressures that these rocks underwent, by utilizing two factors determined experimentally.

The first is that sillimanite will not form below 535° C from mineral assemblages other than

kyanite and andalusite (Holdaway, 1971). Also the breakdown of staurolite before the reaction

kyanite to sillimanite requires pressures greater than 7 kbar (Richardson, 1968). The only

sample containing staurolite with the kyanite and sillimanite assemblage is N4 (Table 5, Plate 2)

from Saddleback Mt. in Northwood. As this is considered to be exterior to tectonic processes of

the MGC (see chapter 3), it provides a mineral assemblage from which the metamorphic

discussion can begin. None of the metapelites sampled within the MGC have this assemblage,

indicating that either the pressure was in excess of 7 kbar or, they are more related to

decompression reactions than prograde reactions.

At pressures lower than 4 kbar muscovite dehydration does not permit melting, but at

pressures equal to or greater than 4 kbars muscovite dehydration will produce a volume of melt

which is joined by biotite dehydration at higher temperatures and pressures (Spear, 1993). In

order for the melt to mobilize and segregate, a minimum of 30 volume % (critical melt fraction)

must be generated. The low porosity and lack of free H20 in the middle crust is not considered

sufficient to produce more than a few percent melt (Thompson and Connelly, 1995), so it would

seem unlikely that large volumes of melt could have been produced in the MGC. Reactions [1]

and [2] are dehydration reactions, and no free H20 is needed to produce the critical amount of

melt, even though the presence of magnetite suggests there was free H20 available. The fluids at

high temperatures and pressures would most likely have dissolved silicates and volatiles (C 02),

which could prevent partial melting by overpowering the / 0 2 buffering ability, and graphite

would precipitate and not magnetite (JoHannes, 1985, Cartwright, 1988). No graphite was

observed in or around the MGC, but in the Type 3 migmatites and in granites that appear to be

unfoliated Type 3 migmatites (Kerwin, 2004, stop #5) and a cross cutting granite sill (Sample Pg-

1) magnetite is abundant indicating water-rich fluids as opposed to C 02 -rich fluids

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. for the MGC rocks.

Interpretation

Based on petrography, the rocks underwent subsolidus prograde metamorphism (Fig. 31).

Starting with the assemblage of sample N3 and following the grey arrow counterclockwise until

they pass through the staurolite out line by the reaction (1] Ms +St = Grt +Bt +Sil+ H20 ± Qtz, to

become the Watson Hill assemblage of Allard (1998). The MGC then progresses through [2] Ms

+ Qtz = Sil + Kfs + H20, at which point the presence of Kfs indicates a boundary between MGC

and Berwick Formation rocks (Allard, 1998).

The MGC metamorphism continues through the granite melt curve and the dehydration

of muscovite initiates melting of the pelites through the reaction [3] Ms + PI +Qtz = Kfs +Sil +

melt creating Type 1 and 2 migmatites. The volume of melt produced by the muscovite

breakdown vapor absent reaction [3] will vary with depth because the solubility of water into

silicate melts increases with depth (Thompson, 1996), and should be proportional to the modal

abundance of hydrous silicates including amphibole and epidote, if present (Thompson, 1982).

The temperatures and pressures at which melting begins may also be affected by the addition of

phases that reduce melting temperature of minerals in the presence of, and without water. Some

of the schists (Table 5) have tourmaline as an accessory, many of the metasedimentary rocks have

apatite and many of the Type 3 migmatites and pegmatites have black schorl tourmaline and beryl

indicating the presence of fluorine and boron in the rocks. Both fluorine and boron have lowered

the melt temperature in experimental work (Manning and Pichavant, 1983) of granitic rocks, with

or without the addition of water, but significantly more than with water alone.

As pressure decreases (unroofing ?) the rocks undergo decompression melting and biotite

will begin to melt through reaction, [4] Bt + Sil + PI +Qtz = Grt + Kfs +melt (Le Breton and

Thompson, 1988) to produce Type 3 migmatites and granites until the rocks pass through the

granite melt curve again and all melts freeze inside the sillimanite stability field.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Temperature ( °C) 400 500 600 700 800 900

3 migmatites ” d granites

+P1, Qtz

MGC types 1 and 2

+ P I, Q tz

PI, Q tz

C 234

C2 TH_

Figure 31. Petrogenetic grid showing path of the rocks of the MGC. Starting with sample N3 and following the grey arrow in a counterclockwise direction the rocks are metamorphosed through the St breakdown reaction [1]. Increases in temperature and pressure dehydrates Mu and produces Kfs [2] and ultimately through the granite melting curve [3] creating migmatites Type 1 and 2. As unroofing allows the pressure to decrease the rocks pass through the biotite melting reaction [4] and Type 3 migmatites and granites are formed . Black stars represent pressure and temperature estimates of Allard (1998) and Larson (2000) of 506-590° C , 3 kbar and 690°C , 9 kbar respectively to constrain the curve. Ternary diagrams are from figure 30. Alimino-silicate stability fields are from Holdaway (1971), granite melt curve is from Luth, Johns and Tuttle (1964), phase diagram adapted from Tracy and Robinson (1983).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 5

CHEMICAL ANALYSES

The rocks for this portion of the study were mainly sampled from the Candia quadrangle

(Fig. 32), with a few from the Northwood, Barrington and Pawtuckaway quadrangles. The

Candia quadrangle has a representative suite of rocks, while one or more of the rock types

(metasediments, Types 1, 2 and 3 migmatites and granites) are absent from the other three

quadrangles. Thirty-four rocks were analyzed in which bulk rock major, trace and Rare Earth

Elements (REE) were determined. Ten metasedimentary rocks, two MGC Type 1, four MGC

Type 2, four MGC Type 3 and fourteen granites were analyzed.

Rare Earth Elements

The Rare Earth Elements (REE) are useful in determining whether a particular granite or

a migmatite is a derivation of a particular rock. The REE's have small variances in size and the

same valence (+3), with two exceptions; Europium, which can have a valence of +2 (Eu+2), and

Cerium, which can also be +4 (Ce*4). The small ionic radius and relatively high valence places

them in a high field strength group, and thus the compatibility or incompatibility with the host

rock may help to interpret petrological processes in a partial melt scenario. Mineral melt partition

coefficients (Kd) which are calculated both from experimental data and by measuring the

distribution between fine-grained groundmass and phenocrysts determine whether a particular

element in a mineral prefers to remain in the host (compatible) or the melt (incompatible). Kd

values >1 indicate the element is compatible, Kd

indicate equal distribution between mineral and melt (see table 6). The Chondrite-normalized

REE values for the rocks analyzed are tabulated in Table 7, and the analytical methods and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LEGEND Migmatites □ Granites (Cg) H U Type 1 (Cl) I I Metasedimentaiy rocks (C) N I H Type 2 (C2) ■ I Type 3 (C3) Figure 32. Lithology map of the Candia 7.5 minute quadrangle adapted from Plate 1. Locations of analyzed samples are shown in white dots with C, Cl, C2, C3, and Cg designations for metasedimentary rocks, Typel, Type 2, Type 3 migmatites and granites respectively. 74

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 6. Some partition coefficient values for the REE's.

K d b io tite ksD ar La 5.713 0.38 0.08 0.015 Ce 4.357 0.267 0.037 0.014 N d 2 .5 6 0.203 0.0 3 5 0 .0 1 6 Sm 2.117 0.165 0 .0 2 5 0 .0 1 4 E u 2 .0 2 5 .4 1 7 4 .4 5 0 .0 5 6 G d 0 .1 2 5 T b 1.957 0.0 2 5 0 .0 1 7 Yb 1.473 0.09 0.03 0 .0 1 7 L u 1.617 0 .0 9 2 0.033 0 .0 1 4 From Nash. W. P.. and Crecaft 119851

analytical errors are discussed in Chapter 2.

Metasedimentarv rocks

The REE's of the metasedimentary rocks (Fig. 33a) have similar concentrations and

similarly shaped chondrite-normalized REE patterns. In general these patterns have negative Eu

anomalies with similar overall steepness (La/Lu, Table 8) of 72.9-145.1, and the HREE's patterns

are concave up. The exceptions are samples C171, C170 and C15, which have near positive to

positive Eu anomalies (0.8-1.02, Table 9) and shallower slopes.

Migm atites

The migmatite's REE patterns show some similarities and deviations from those of the

metasedimentary rocks (compare Fig. 33a with b). The two Type 1 migmatites (samples C229

and one from the intersection of 101 and 101a in Milford, NH) are similar in concentration and

slope (Fig. 33b). These samples were selected by similarities in foliation spacing index numbers

(Kerwin, 1999) and texture, but are from opposite ends (southwest and northeast) of the MGC.

Two of the four Type 2 migmatites (C2 119 and C2 213) have similar Eu anomalies of 0.85 and

0.89 respectively and similar slopes of 101 and 110. The other two samples, C2 116 and C2 90

have pronounced negative Eu anomalies (0.62 and 0.35 respectively), but contrast in slopes (74.1

and 265.7 respectively). The four Type 3 migmatites are similar in slope, Eu anomalies and

concentration except sample C3 229. C3 229 was taken as a suite from the same locality, which

Element (ppm) ) u - r o r ^ o c O O l f i n i o t 0 - r O L O O l O ) f f O l ^ ) l f ) f f qin«-ff)inpiO(T)^p r o t b M - C N d M - f — c M t b r o mOfOWONNWCO^ rocduSTf^LO^^Nicn(O(i>qqu>ou)0(nu) Tl;.-p - . ; l T — . ^ * f T p ^ n p p ^ r m O - > h O - N O O - - * - N - * N m c — — M < - , - , — , - , — cmcm.— m c m c o c — . m c m c — * m c o c .— ! s r j \ c b t n L — * d C ( b O r c M ( o N ) t O n ^ i O o ( l O C O r>uifloa)tfU)a)U)N(o C q O C N O f N ^ S N O f t r oiw^d(J),

r— 00 o 00 o CO o o . N — r «— o CL

2 | 2 D) E

D N. CD M t“ CM tO Tf oq O LO CO 0 0 CMCO o N- M- M M*CM " M 0 0 00 O . CO N. UO Is. LO o CM CD CO 0> m o r~ “ CM *“ U 2 d < o o . _ k CM s I o t o t LO 00 oo cm Is.

o c\i oo O O O COCO O COCO sl \i c CO d o - r - O r COCM LO l rs ^ T}- O i— CO CM 0 CO 00 O O 00 O^N M - f CM O s r co M- u i N CO CO N ^t CM CM N i u LO N o o n in co co CM 00 - t N CM CO CM - r t— ^282 ^282 O O N LO CO _ M- CM CM CO CM CD rs rs. CM O CD CM JC JCJ CJ CJ CJ °9 0 CM CO 00 o o t CM . CD © ^ ^ LO lOP 9 o t 1— MC CM CM CM r—

O o p CM o n i CD w— «—

n i •tt E O(D CO MCM CM r—

D C CM CO - h CD O N p N; CO CM CMCO | CO M" LO CO M" oto > \ c in co c\i a> O CO O CO CO CO N. CO o L L O LO LO to O 0> o CO - r 0 D — CO to CM «— CD O 00 O CO 00 «— CM i—«— d M* o c m o t 0 M CM* o o t CM o 00 to CO co o o t t N p CM “>S m ^ 76 O — — i w •— O O in S COCO ooo oo ^ CM OCO CO Joo o CJ £ Q. O O - r COCO CO C\i •— ff> rs rs ff> O O) CO ^ LO s I C «- « CM t " CM CM d CD r— r— M- CO o t CM 0 0 — r CO CM . n CO CO CO hs. w o CM CO r— u

c\i M- O r~ o t M CM ^ CM LO o t O o O O O COCO COCO . - o t M- N. M- o t in d ^ o ^ cd o to t co O 00 CO < - h CM ^}- CM - r m s f © O CO tO CO t s ' I . N s r — CO f— - * r— a. d 6>

00 10 CO d cm CM CD o CM CL. D> ) 6

o t o t o 00 CM - r LO LO o CO CL

O00K to 0 0 CO LO M" 1-0 CO N CO ^ LO *— f T CO 9 »— CO o O ■ o t co s I ^ D C O Q - O M LO M- O «— C - O CO M - . o r O k T i-^pqinpgg !£ i-^pqinpgg T M M K CM ) f f tO N O LO CO N f N CM T CM LO «— CM CM «— I— CM ^ - r ^ O C M C M C — C C >- 't ' - ^ > LO CO CM tO - i O CM CM «— M- CM CO o t M > O — «— »— CO O 0> CM coropr-'-Tt®J2cM c o L O c o < D t o c o mmi n N N W CO N (J) (J) N CO W N N CM o t CM in CO ^ o o t o t ^ LO CM u )O) O) O ! ^ * C> * ^ A r ^ ^ iri o t © ^ CD Tf o t d m * t i o ' s o t CD o t 00 o o t o t LO M - CT> CM r LO f ’ CM < *— JC o CJ CJ D to p CO ,lfr CO o t m CO o t 0 0 LO CM k O M* CM CM o © o> ! O O N N N O o CO to CD CO J g CM CM g J CO CO co O O to CM is. CO CO to i 5 a) CMooo 2 h S Ol O) m ) 0 l O ) D p o s I CO p LO OCD CO CM § s ooco 2

. . V o t s - . CD N. M- s I D O ’t CO ID s I o is! to ^ q Op CO o t o t o - r CD CM O s r M*LO CM O O LO s r CM CO r~ oo o CO (D o o s f o t >

Table 7. Chondrite normalized (Nakamura, 1974) REE values determined by INAA for for the MGC rocks. C, M and P = Candia, Milford and Pawtuckaway quadrangles respectively. WH=walnut Hill, Th=Tower Hill and gB=Barrington granite. 1000

100

1 La Ce N d Sm Eu Gd Tb Tin Vb Lu

1000 ~i— r 1— r i — i---- 1—r "i— r T T

M1 101 A C1 2 2 9 C 2 9D C 2 116 C 2 119 C 2 2 13 C 3 2 29 C 3 119 100 C 3 21B C 3 120

1 _L_L i i _i_ i La Ce Nd Sm Eu Gd Tb Tm Vb Lu Figure 33. Rare Earth Element plots (Chondrite normalized) for the metasedimentary rocks (33a) and migmatites (33 b). Black =MS (C) Red=Type 1 migmatites (Cl), Purple =Type 2 (C2) and Green = Type 3 migmatites. See tables 8 and 9 and discussion in chapter 5. 77

S am p le La Lu La/ILu La/ILu sh a llo w Metasedimentary C79 100.61 12.65 7.95 ■ ■ ■ ■ ■ ■ ■ 2.65 roeks C234 175.76 15.59 11.28 C170 4.86 G 1 106.36 13.53 7.86 C171 5.94 v C ^ ^ CO Cvi o t d c o t o t 0 0 to o si c o ~ r m CNJ LO O LO . s f CNJ «— Tj* LO o u u 3 to o t ^3- p t CO o o LO 0 0 CO l s f o CNJ Oto t CO O co CNJ CO o t CD n D|sl s | CD o t 00 i— ! CO CO u fs. © •— LO n . s r LO LO - N o t ! I ■ —■■■r— * i o t . fs o t i r c s! fs . s f M- o t tr> OC o t CO LO 00 LO o CNJ O o 0 0 LO LO . s [ M »— w CM - « *— ' - r

CO • ■* o “ r t s ci O “ CD •“ ^ CO csi fs» tb r— «“ o t *■** co •” ■ ^r-r-r-r-^tnfOfOfOfO^^mSN © im ^ M o t o t 5 ? « s 5? O O 00 9 0 0 tO CO r r < ;r :r b c CO O cm •3- *

a OCO CO CO OLO CO LO o t o t S. fS fs.

* 3 f s CD t o CD CD CO 00 o c> ^ d> c> c> o o o o o o o d o d o d d d o LU

CD CD CD r— LO f s CM CM CO to co M" CD f s T— LO CM CO M N N s N ffl f s I s l o c o LO LO CO CM h - CO N- CO O' O o CO q M; CD q CO CM 0 0 '° ? ^ ro irt o co n CO |s! rs! r— M" CM CM O LO LO CD LO O CD M- LO CM o GO CO LO CO O to CM rs ! rs! ^ «— CO CM CM CM CM *— CO CM CM CO 00 «— CM CO CM CM »— <- i- N CO f- t

CD CO LO LO — LO *— r— CM LO r—

,_ CM CM CO fs. CD CO 00 t o CO *— CM r s CO CM o LO O CM CO CO 00 LO CO CD LO CO LO fs CM q q q CO fs. CM q LO *— q 00 LO CD CO q r - O q LO q CM o r- «— M; CD 00 CO r_! is! n ! r-1. cd o CM LO rs! cd t© CD CD CM od r-1 o r— CO fs! d M" CM M" CO CM CO LO CO CO CM M- f s CM CM CM CO CO CM r” CM M" C\J Cvj CO rf ^ Tf

c d j r o> o r s CM O *— :P- r~ •»“ CJ CJ u cj U u CJ u CL 0 2 >% £5 B

I CO Q. Cu Nakamura (1974) and color coded by lithology.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. had representations of more than one rock type in contact or proximity. This was also done with

samples C2 90 and Cg90 (MGC Type 2 and granite) and C2 119, C3119 and Cg 119 (MGC Type

2 and 3 and granite) to test the hypothesis that the textures were derivatives of the same rock, (see

Comparisons of chemistry with texture below).

Granites

The REE plots for the granites (Fig. 34a) also can be subdivided into two categories that

either resemble those of the metasedimentary rocks (negative Eu anomalies and shallow slopes,

see Tables 8 and 9 for calculated values) or those that resemble the Type 3 migmatites (flat to

slightly positive Eu anomalies and steeper slopes). To simplify the discussion two granites have

been selected as identifiers of the two categories (Fig. 34b). The first category (Barrington

group) is one in which the slopes are relatively shallow and the Eu anomalies are negative (Fig.

35a). The second group (Milford group) includes those in which the slopes are steeper (Fig. 35b)

and the Eu anomalies are slightly negative to near positive.

"Anomalous11 rocks

Figure 36 is a composite of all the rock types that did not neatly fit into the above

categories with the two identifier granites plotted in bold black. Two metasedimentary rocks

(C170, C171) are both within or very near the FHF zone, which may have introduced some

metasomatic fluids into the rocks. Both C170 and C171 have a REE pattern closer in slope and

Eu anomalies to the Milford granite than any other metasedimentary rock analyzed. The C2 213

REE plot is almost identical to C 171 (see locations on Fig. 32) and was correlated with rocks of

the Berwick Formation based on field observations (Larson, 1999). Sample Cg 26-2 lies within

the CHHMFZ and has most likely had some influences of migrating fluids over the span of fault

activity. Tower Hill granite (Cg TH) with its unique pattern of high HREE's doesn't lie near any

mapped tectonic feature. The topography in this area suggested the presence of a fault but no

surface expression could be found and CgTH remains anomalous.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1D00 Pg-1 esiiPg-2 C g 119 C gB 6 C g 9 0 C g WH C g TH C g 26-2 too

U Ce Nd Sm Eu Gd Tb Tm Vb Lu 1UUU

100

M g

Nd Sm Eu Gd Tb Tm Yb Lu

Figure 34. Rare Earth Element plots (chondrite normalized) for the granites (34a). Two REE patterns can be observed from all of the data and typified by the Barrington granite and Milford granite (34b). One in which the Eu anomaly is strongly negative and a relatively shallow slope (gB). The other has a weak negative or positive Eu anomaly and a steeper slope (Mg). All of the samples in this study can be grouped into one or the other of these characteristics.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. -«— M1 101A •e— C1 229

h — C3 229 ♦H&79 « — C 2 3 4 C1 ■*— C15

■a— C171 ■0— P 101 100

La Ce NdSm Eu Gd lb Tm Vb Lu

— ♦ - -C 2 119 —B--C 2 213 --- 0— C3 119 1000 i I I r ! I r i — r —*- -C 3 218 ----1—- C3 120 — &—-Pg-1 —•—-P g -2 — ■—-p g -3 100 --- ♦—- C g 119 — *— M g

La Ce Nd Sm Eu Gd lb Tm Yb Lu

Figure 35 Possible sub groups. The Barrington granite with rocks of similar REE patterns (35a). The Milford granite and rocks with similar REE patterns (35b). The rocks were grouped by pattern regardless of rock type.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1000

C2 213 Mg C170 c 171

100 -

Ls. Ce Nd Sm Eu Gd Tb Tm Yb Lu

Figure 36 shows the rocks that did not fit the two REE pattern groups. The samples C170 and Cg 26-2 are found in or very the Hint Hill Fault and the Campbell Hill-Hall Mountain Fault zones respectively. Proximity to faults zones and fluids utilizing the fractures may account for the anomalous REE patterns. Cg TH does not lie within or near a visible fault, however, the topography in that area suggests faulting although none was found.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Major element analyses

Several major and trace elements were analyzed by XRF (Table 10). The granites are all

peraluminous (Fig. 37a). Figure 37b is a Quartz, Alkali feldspar, Plagioclase ternary diagram

with eutectic tickmarks (Tuttle and Bowen, 1967) showing PH20 values of 5,3,1 and 0.5. None of

the granites plot on minimum melt compositions with the exception of the Barrington granite

which plots on the Ph2O0.5 eutectic. Granite data are also organized into discrimination diagrams

to help understand their development in a tectonic context. Figure 38a is a diagram taken from

Batchelor and Bowden (1985) that uses discriminate functions R1 vs. R2 where R l= 4Si-

1 l(Na+K) -2(Fe+Ti) and R2 = 6Ca +2Mg +A1. The values used in the discriminate functions are

calculated by the equation, ((Wt % oxide / molecular wt)* the number of cations))* 1000, giving

millications per lOOg sample. The tectonic discrimination diagrams generated by this method

indicate the granites were formed in syn-collisional and active continental margin environments.

In the lower diagram (Fig. 38b) plotting the compositions of Ta vs. Yb (Pearce, et al., 1984) put

the granites in this study in either the volcanic arc granite (VAG) field, which includes granites

from an active continental margin or the syn-collisional granite field which includes syn and post

tectonic granites.

The major and trace elements are also plotted against Si02 (Fig. 39) and fractionation

diagrams (Th vs. U, Zr vs. Hf and Nb vs. Ta, Fig. 40) to see if there are trends from one rock type

to another, of which, there are few. There are vague negative trend lines in some of the figure 39

diagrams: such as in A1203 and F e ^ vs. Si02 (Fig 39a) and Ti02 vs. Si02 (Fig. 40a). However,

the rock types that make up the trends are mixed in these diagrams, and it is not apparent that the

migmatites were derived from the metasedimentary rocks and the granites from the migmatites.

In the fractionation diagrams (Fig. 40b.) there are some positive tends (Zr vs. Hf and Nb

vs. Ta) that do not show one rock type fractionating into another, but do demonstrate some

fractionation processes taking place. The Nb vs. Ta diagram shows granites Cg TH, Cg 21, Cg

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. K 20 1.842 4.282 2.5 1 6 3 338 5.991 6.149 4.9 7 6 4 .4 2 5 0 15 2.001 0.15 P205 0 156 0.1 0 3 3.74 0 .1 2 9 0 .0 7 6 5.21 0 .0 9 2 0 .1 3 4 2.4 0 .0 6 9 4 .4 1 6 0.1 1 9 Si02 66.693 0.085 69.808 0.283 1.659 7 3 .4 1 3 65.416 0.235 6 02 68,36174.451 0.164 5.389 74.72871.888 0.321 0.095 4.6 4.843 7 1 .6 0 4 15 18 68.156 0.307 2.508 14.43 71.709 14.79 71.806 A I203 15.512 66.941 0.147 2.644 18.069 13.71611.598 74.038 78.369 0.084 0.051 6.05 3.091 12.69313.094 71,80511.74614.081 70.773 0 15813.132 75.04512.601 68.985 0.145 2.356 73.603 0 074 2.482 0.048 2.795 1.91 11.177 72.627 0.154 1.897 13.136 71.094 li.69915.698i5.595 78.46 70.601 70.182 0.047 0.137 15.157 2.609 72.258 0 088 4.645 11.617 72.292 0 15213.492 0 05 14.448 73.64 0.083 1.856 13.807 15.717 69.593 0.244 14.08915.981 14.521 75.13114 .4 9 8 73.081 0.086 4.744 13 .9 0 6 20.049 61.613 0.127 5.062 14.681 2 0 .6 5 3 2.91 2.6 6 1.673 1.743 2.6 0 4 0.201 0 .4 8 3 0 .5 2 8 0.3 8 2 0 .2 4 5 1.85 2 6 82 3.07 2.391 13.154 71.929 6.11 0 .1 7 8 N a20 MgO 1.846 2.571 1.899 1.481 2.613 0 .4 0 4 3.201 1.283 2.123 3.623 1.128 2.9 1 5 2.2 3 3 2.8 8 2 0.7 2 5 3.641 0.5 0 2 2.8 5 6 0.7 1 8 2.431 2.5 6 73.103 1.169 1.302 0 7 5 3 2.943 0 .4 1 7 3.0873.1 8 3 0 .7 0 6 0 .2 6 6 3.6 7 8 0.7 6 7 3.5 4 3 0 .4 6 8 4.1 6 73.4 0 4 0 .7 6 8 3 1 043.1 5 6 0.601 4 .0 1 4 3.387 0 .5 6 7 172 3.0 6 8 193 696 1.853 146 4 4 0 481 9 0 0 2.1 8 2 5 0 0 .6 7 1.4 7 41.931 134 2 85 5.0 4 2 6 95 3 .7 9 6 5 1 4 6 .4 8 4 1.1 1 9 1.163 4 2 8 1.733 2 05 3.3 7 3 1227 5.445 653 2.43 2.868 5 .8 4 5 674 2 .2 8 7 3 0 4 4 .7 9 9 4 .3 2 45 .7 2 8 3 19 9 4 0 3 .9 1 44 .6 2 3 4 7 4 5 8 0 4 .6 2 5 5 37 2 .4 0 6 156 8 .7 8 2 4 5 0 4 .2 3 9 2 49 4.261 3.0752 .5 3 3 411 196 3.429 0.849 F e203 MnO 0.5 2 1.283 0.725 0.373 2.327 266 0.7 4 2 0.6 7 6 5 .0 6 4 0.387 2.6980.1 1 3 308 0.327 2.242 195 0.0 5 2 0.287 1.961 198 CaO Ti02 1.426 1.817 1.353 1.561 0.265 1.699 231 1.783 0.321 2.253 0 .6 7 6 5.1 4 42.437 0.652 0.785 0.8 6 6 0.9 4 9 2.282 0.3 7 9 3.316 0.707 3.509 0.8 8 9 2.335 0.545 1.896 0.5 1 3 0.367 0.5 4 4 0.915 0.1 8 5 Ml 101a 1.403 0.765 C C 1 C 234 Cg TH P 101 C 170C 171Cl 229 2.969 8.5 2 7 0.701 0.6 6 2 3.305 C 169 2.507 C 15 C 79 2.722 Pg-2 Cg 86 1.491 Pg-i Cg 26-2 Cg 148 1.574 0.365 C3 229C3 119C3 218 1.244 1.555 2.606 0.198 0.176 1 71 • gB M g C3 120 1.609 0.259 Cl 67 C2 90C2 116 C2 213 1.897 0.585 C2 119 Cg 90 1.584 0.35 Pg-3 Cg WH 0.9 8 8 Cg 119 Cg 21 0.903 0.1 C, M and P= Candia, Milford and Pawtuckaway quadrangles respectively, = Mg granite, respectively, Milford granite quadrangles gB = Pawtuckaway Barrington P= and and Milford M C, Candia, Table 10. Major and trace element data for MGC rocks in wt% oxides. wt% in rocks for MGC Table 10. data calculations. in used element data Major trace cells Shaded and are WH = Walnut Hill and TH = Tower Hill. Tower = TH Hill and = Walnut WH rocks C 61 granites m igm atites Fusion beads metasedimentary

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 87 381 341 589 893 999 1543 Zr Ba 216266 721 273283 506 178 2 1 6 8 207 345 199 5 10 228 2 46415 2 26 2 6 0 593 173 9 68 2632 60 1829 232 4 304 Sr 176 192178 321 133 2 28 2 34 159 243 699 351 2 09 235 4 45 2 1 0 2 08 2 48 155 1051 775 2 16 6 08 4 4 0 378 219 4 03 341 2 093 3 181 281 Rb 86 8 0 97 143 211 267 373 183 109 164 329 516 170 181138 2 214 1 0 439 213 207 133 160 179 2 10 195 2 76 165 240 307 9 0 65 459 261 50 48 184 Fusion beads C C 79 C 234C 1 C 170 189 C 171 76P 101 94 114 C 15 C 61 Cl 67 M1 101a 234 97 353 1078 Cl 229 C2 90C2 119 C2 213C3 2 29 C3 164 119 100 135 C3 218C3 1 20Pg-i Pg-2 Pg-3 74 88 206 264 427 Cg 90 Cg 21 Cg WHCg TH Cg 26-2B g M g 235 593 94 59 53 59 434 312 Cg 119 Cg 148 154 385 270 1625 rocks C 169 granites Cg 86 204 213 182 823 migmatites C2 116 226 metasedimentary and Pawtuckaway quadrangles respectively, respectively, granite, quadrangles gB = Barrington Pawtuckaway and Hill. = Tower TH Hill and WH = Walnut granite, Mg = Milford Table 10 continued, minor elements (fusion beads). (fusionelements beads). minor continued, 10 Table P= M and Milford C, Candia,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 9 9 8 9 1 4 9 6 6 11 Th 11 12 11 10 13 12 11 14 11 38 27 31 26 23 ppm 9 Pb 10 14 15 18 12 20 36 19 39 ppm ■ ■■11- ■■11- " ■ 9 Nd 39 36 25 39 13 38 5 0 26 29 824 8 10 49 11 26 2621 4 0 33 19 42 24 18 21 2727 32 39 ppm Ce 62 29 27 68 26 62 28 25 92 118 58 181 PPm 2 94 97 24 7 7 -1 20 6 16 -2 32 La 13 29 13 51 1 18 42 35 46 86 62 118 2627 52 53 21 36 67 74 16836 59 65 11 62 934 032 53 7 0 65 14 30 12 6174 146 1 60 48 8 12 24 31 6 26 3234 67 5820 88 32 50 81 46 81 34 67 78 ppm 188 9 Nb 12 14 30 10 11 10 13 21 12 10 11 21 ppm V " .: £ , v Zr 61 6046 14 13 172176139 20 22 188 187 183 13 194 14 ppm 179 17 2 442 08245 13 17 212 2 16 207 2 88329 14 2 8 8 25 2 38 . 9 2 44 13 279 2 5 43 1 6 5 12 269 .1 8 3 Y 7 2 0 8 4 4 15 10 10 19 28 28 43 29 21 ppm Sr 54 51 10 91 18 85 94 199 28 182 34 2 50 13 175 132 170 ppm 222 30 205240 19 16 326 2 26 209751 8 585 11 385 ■ 5 "' ■ 270207 115 21 205 4 38 6 Rb 81 80 188 82 98 93 192 102 167 23 139 167 173 30 ppm 214 134 159 223 100 ■47 2 00 175571 391 13 314 222 Sample C 1 C 79C 234 132 C3 229 Ml 101a C 15 173 C2 11 6C2 119 C2 213 227C3 119C3 218 130 88 9 218 Cl 67 C2 90 161 C3 120Pg-1 Pg-2 85Cg 86 155 203 11 C 170 C 171 C1 229 5 Pg-3 Cg 9 0 g B Cg TH 195 Cg 21 Cg 119 Cg WH 2 26 Cg 148 161 Cg 26-2 239 granites Milford and Pawtuckaway quadrangles respectively, gB = Barrington granite, respectively, Mg granite, = Milford granite, quadrangles gB = Pawtuckaway Barrington and Milford WH = Walnut Hill and TH = Tower Hill. Tower TH = Hill and WH Walnut = Table 10 continued. continued. Table elements 10 = Minor Candia, P pellet). (pressed calculations. XRF and C, in M used cellsdata are Shaded m igm atites rocks C 169 Pressed Pellets m etasedim entary C 61

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. OO in © © t o © O O O © © O O © Q © to CO ^ 0 « N CT> © 1— N O N © to N. © *J CO < ~ ai <6 o> M* M* CM -CO © CO CM © O CM - © CMtOCMCMCM M- CM CO CM -CO CM CM CO CM CM © 5© © ft-©©©!— -M-©' ©

© O © © © O t o ' 2 ’© © w o to "f © o mi ro U) •-I CX5 -r- s s - * n . © rslvcvi ^ to -00 5? m1 . 10 “» ! = ' * I*. LO ’© LO UsU)tO ®-°kw

0 - 0 O O -•'■>/)-..^I. ■ yyi■,- jkl ; •. _ O © ©OO ’h- CO O CO 1 0 o o S © o © m o © o NOr-©^....r-T- >£0 on •- ro t o 00 'CM icj> fto in c*. *00 'CO -LO (O - o > 't\i O LO CM CO •00 lO 10 r~ CO "M* CO l/ids., cn m to ro m CO CM CO M- CO M- CO -CO CO LO CM CM©©©i—©©y

0 - 0 © o SS BS n o> - S n n 88 -8 8 : 8 " LO '© LO © t'8 8 g 8 g 8g-8 8 g 8 © «— i— CM CD CM tD U) in 't © © O 00 lO CO in 8*2 CM 'M- CM ^ “ ■ ts . CO CO CO t o ? m ’ m 1 0 )z LO XTi N . *LO a ss - - ® ,2 •£ »2 - 2 o o o O o © © © OOO © 8,8 88 o © O O 888888 8 C C t£> to to © M“ M* © ©LOLoOM-'ai © CM© CM«- CM© © © © •—M" ©M" ©CM 00© •—M- ©M* © © © m to £ ©LO t o fs. © C M tO tO CT> <— M" CM LO 1— <0 to *- tD to © © CT) «- in :5k K o>S °?co tfioo oo cm LO ,LO CO •-; © © in .h. to eot co to cm ": • © C0 A** to w ■ ^ CM CM* LO CO M* CO lO »— © M" M ft ri CT> CM CM CM* -CM M* © 2 *” © o o o O © OOO© m © © o © O © o o o o o o o © o © o o © O © 0 0 0 0 CM © N- © N; O © qOOOOqiOOq 8 M- •— «— trj r*>» CT) LOSNNOJN^r^oi "Sod CM »— CO M’ M- © © CONCOCOCMOOM'CM© *— f\j «— *— (M t~ * — •— CM»— «— (M»— g o ©_ o OO O g CO o © o OOOOO ©OOO 2 -a O © LO . © © ©9*n(0 O CM © ” 8 8 qooLnooiooo 2 c to 0 0 ■*- - « =to •CM © © cdcxDin©©©^^^ h - CO 00 ~ © © , CM • CM © '© © CM © © © ©

o o o o Jt' -fO h*» CO © o go £ ; ^ © o ooooooo© co CO CO N M* LO v t a o •...■:■■■■■'. iff ^ ^ M; to N- © © . M; a- CO OO ’U) N (D CO r- to CO csi © a » in CO ^ CO ^ ° ■— .ru ■•-■ :•*- w~m'-r~- ■ 2-od -in CM © CM O -LO CM -M" © © ft^ CM

LO '© 8 8g°^Poooo- N. CM _ o o © o o o o o 88 182 ■JCO CO ■O tO O l CNJ ^ CO -ft- tO © «— t o 1 CO — ,.fi ' CO LO 2

o © 0 0 0 o © ©©©^j © _ m r- LO CM Xf M* CTV to o © CO to CM 0> 1ft Al Aft 4N :1ft lifts S o © 000©000t0© V i CM © to © © © © q ‘to A- ?o> t o r-T LO CM 1— © CO *—• 2 ~ f\t 2 CM ^ 00 ^ «— CM t~ r-. W ™ •“ •" 2 ™ CM* © © © :© M* M" © CM

N- «— IT) O © •— © S-

O M* CO CO O 1C LC N © O CM ©M‘©M-M’tO©M"0 O N tO N 00 © N- 00 to 00 M; 1—1—-O^LOCT^LO.— CO® 1 0 is. m- in M- LO M“ M" t— M" >— r-^ M" © CM CM *-© « —-CM O *--(\j*— a-

© © CO © © © O CT) T ^ CP> {—» fw •“ r~~ © CM O — CM *- *- CM - - ID O 5 I (fi ro to *- © i r N- O *- CM m <— •— cm CM «- CM — W r- CO O) > h N CM CM CM CM O) O O) O) O O) O) 1 OCJCJUUOCJOOQ. OOOO o o u o o o a Pawtuckaway Pawtuckaway quadrangles respectively gB=Barrington granite,Mg= Milford granite, WH= Walnut Hill ro ri and Tower TH = Hill, nd = not determined it Table continued. 10 Total INAA Data. Shaded cells are data used in calculations. C, M and = P Candia, Milford and

S a m p le ______Stn Eu Gd Tb Tm Yb Lu Y Zr Hf Nb Ta Th U Sb P b Metasedimentary C 79 5.97 1.24 5.42 0.83 0.41 2.88 0.43 28.00 269.00 6.92 13.00 0.85 9.54 2.17 nd 10.00 rocks C 234 9.85 1.90 10.10 1.44 0.56 3.65 0.53 28.00 265.00 6.85 17.00 1.05 19.00 3.95 0.37 27.00 © ■o VO © CM i o CM CM © p p o O o o o o CD CD CD CM* . t f m MCO CM m 8 CO o o O m o o (O CO o o O o o o r“ t r CM CO o o o o o o t CD o o CD o t ro o r“ o t t f iri *— 0 L - oooo r - r ■ £ CM CO m o o t o t DCD CD CO m v t CO o t O O M; o o m o o ML LO LO CM CM 00 fv uS *” Tf CM CO o 0CD 00 p o *— O© CO LO iri ■ :0 CM :C0 - T 0> © N © m CM o « o c m O m i *n viv cm n i O CM - f CO CM^ - h CMCO O CO cvi m O CO O o c O © VO VO *— >r—'CO 0 0 © O »**• O . O CO CO «— M" >— *— CM (M N. O CO(O © O O O O O O O p o p p ■ ^ CD o r - e dodo o t v o o cm* o o cvl i r t - ^ r tO o v co tv m fs. CD fv tO . N CM 00 CM ^ CO co o v o c o t O o t CD h*. v p m m p >vo v r o MCM CM *~ © Q. § o § 1— M CMCM CVJ O 8 8 o t d O © r— -»"• tM

rsi O ft ’f CO 8 lo d d 00 o rv CM © ( O ' — —

^ m *f :* d d d d ro CD m m MCM CM CM © r— O

■•'«T D * m M*

d CO o r o ri»— ,r-i CM CD CM m o . d cm d © © © IV o CM CM s CM m 00 n i in © O d d d n40 © in © d in © fv © m t— CL ) 6 ' •.t—. p o o t O *“ ■o o o o o © *n © d CM LO CD M* o OO o © d o © CO o o - O |V cvi o DCD CD © rv oro ro f i o M©CM © CM CL C C c O) ) 6 ) 6 ,© ,ro d o t CM O C CM •CD O 73 CO o CM o D00 VD CM © in © © O OO o t rv © M* O d M tf •M; o n i M© © CM p M© CM © © o t CL .00 p O r r V r d d o W CM o t O 00 T3 CM 00 m o CM M- CM CO rv o © -M; © 00 o *“ © CM o © o CM © © © D© CD © M1 M* o u S*t p :.sr m p tspoicop!Jo) o ^ m o o m o o O t CM v f o t CO .to o o o o ■CM o o o o ■if j p - r - r CO fti tO N o o o O O OOO O O N i O ’ i^ « i u ) i ni — " r t o © *— o rv O fv *© M* © r © - © CM CM © in O d O M- o © LO o t »-* © to © 'o :0 C 'M D N CD 'CM CD 0 : .'to ■— o o IV © v f © © CM to OOO© * © ID %© © O dodo ^ M- CMCM fM - i _ CD O O O CO CO CM CD o o © o CD o o fs. 0 0 © o o o CM CM o o o © m CD © CM oo ) O d p © m rv M1 D© CD © © f n ■ft m m o CO CO CM © o © © v o cvi © © © CM m o CM p in o *” Q CM in O M CO M* !

c m p .co ■CM. . •M- r- .r CO ■**“ ' OO CD o © © oo o © o t o r~ © — r |v N- to CM o o © © © o o o o o o t © o o t © o t »— © © © T-^ © © © p m d © CD Dl Dl CD 5 Z 05 -r- © O (D © o o o © *” rv d »— © o d © in rv d r—

Table 10 continued. Total INAA Data. Shaded cells are data used in calculations. C, M and P = Candia, Milford and Pawtuckaway quadrangles respectively. gB=Barrington granite, Mg= Milford granite, WH= Walnut Hill and TH = Tower Hill, nd = not determined I" 1 1 1 ...... I 1 l ...... I Metaluminous Peraluminous

Peralkaline i i i i , ..._J, ...... 1____ 1 ...

AI2Q3/ (Ca0+Na20+K20)

Cg WH Cg 21

Cg 148

'Cg TH Mg

• Cg 26-2 P Figure 37. Major element plots of the granites. Figure 37a is a diagram, which shows the granites are peraluminous on a molecular aluminum saturation index. Figure 37b is a Quartz, Alkali feldspar, Plagioclase ternary diagram of Tuttle and Bowen (1967) with 5, 3, 1 and 0.5 PH20 tickmarks indicating the locations of experimental values of ternary eutectics in the presence of water. None of the granites plot on a eutectic point with the exception of the Barrington granite.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2000 2000 mantle fractionation

pre-plate collision

1000 xollisn

late orogenic 1000 syn-collision' anorogemc Mg

1000 2000 3000 4000 Cg 26-2

1000 2000 3000 4000 R1

Within-plate granites (WPG)

(SCOLG) \ Syn-collisional granites From figure 27, p. 63

• Granite • Granodiorite 0 Tonalite Cgll9 X V*g-2 (ORG) ^ Quartz-rich •C g90 Ocean-ridge granites granitoid ,P g -l (VAG) *Pg-3 Volcanic arc granites I 1—1 J-l .u.,1____ i_ 1 10 100 Yb Figure 38. Granite discrimination diagrams. 38a R1 discriminate function (4Si-11 (Na+K)-2(Fe+Ti) vs R2 discriminate function (6Ca+2Mg+Al) diagram of Batchelor and Bowden (1985) Figure 38b is Ta vs Yb granite discriminate diagram of Pearce et al. (1984). The classification of tectonic environments contains the following possibilities: ORG can be granites associated with normal or anomalous ocean ridges and back arc or fore arc basins, VAG are granites found in tholeiitic and calk-alkalic basalt ocean arcs and in active continental margins. WPG are granites found in ring dike complexes, attenuated contintental crust and oceanic islands, SCOLG are syn and post tectonic granites in continent-continent collisions and syn-tectonic granites found in continent-arc collisions.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced a. Fe203 total CaQ A1203 Figure 39 a shows A1203, CaO and Fe203 vs Si02, b) shows K20, shows b) Si02, vs Fe203 and CaO A1203, shows a 39 Figure diagrams. oxide) (wt% element trace and Major b, and 39a Figure MgO and Na20 vs Si02. Envelopes have been drawn around the migmatites migmatites the around drawn been granites. Pg have the and Envelopes Si02. vs Na20 and MgO 64 02 i0 S Tp imtts Milfordgranite • (Mg) Pggranite Typemigmatites 3 sills O ■ metasedimentary rocks « Type 1 migmatites ♦ granites ♦ Type1migmatites ye2mgaie oBarrington o granite(gB) Typemigmatites 2 060 80 64 68 02 i0 S ♦ ♦ 0.8 40

0.6 30-

0.4-

10 - 0.2 ♦ o

U 500 0.8 400

0.6 300

0.4 200

0.3 Zr/Hf~33 100

Hf 30 1.5

X! 0.5

a. ♦ ♦ Nb/Ta~16 72 76 Si02 b. i « metasedimentary rocks Pg granite sills Ta Type 1 migmatites granites Type 2 migmatites Barrington granite (gB) □ Type 3 migmatites Milford granite (Mg) Figure 40.40a are minor element(wt%) graphs of MnO, P205 and Ti02 vs Si02 with envelopes drawn around the migmatites and Pg granites. Figure 40b are three fractionation diagrams expressed in ppm with trend lines and slopes for reference.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 26-2 and Barrington granite (gB) plotting away from the trendline. These four granites also plot

in a different field in figure 38b.

Granite fractionation

Fractionation diagrams compare elements as they separate from a crystallizing magma as

a function of their compatibility or lack thereof, in a mineral. Some of the granites in this study

are foliated but are otherwise indistinguishable and fractionation diagrams provide a method in

which to relate them to each other. In figure 41 the granites show fractionating trends where the

Milford granite lies at or near the top suggesting it may be parental. In both of the fractionation

diagrams in figure 39, the Barrington granite (sample gB) appears distinct. These results argue

against the earlier use of REE patterns that suggested two groups of rocks matching either the

Milford group or Barrington granite group.

Originally it was hypothesized that migmatization of the metasedimentary rocks may

have generated the magmas for the granites. This was tested using the same fractionation

diagrams (Fig. 41), but with the various migmatite compositions (Fig. 42). None of the

migmatite compositions fall on the trendline defined by some of the granites, so by this method

the migmatites and the granites appear distinct. This could be further tested by analyzing the

leucosomes from the migmatite samples and plotting them on the fractionation curve.

Comparisons of chemistry and texture

Several samples were taken from localities that had more than one texture in contact or

proximity to one another. Three locations were chosen: C229 (MGC Type 1 and 3), C90 (MGC

Type 2 and granite) and Cl 19 (MGC Type 2 and 3 and granite). The hypothesis is that the

migmatites are derivatives of the metasedimentary rocks and that migmatization progressed in a

way that Type 1 migmatites would become Type 2 migmatites and Type 2 would become Type 3

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 Cg 11

/ Pg-2 - Cg 86^C g 90

/ * Cg TH - o> Cg 26-;

0.1 100 1000 Log (Ba)

Pg-1 1

Cg 90

LU Cg21 'Cg WH Cg 26-2

0.1 10 100 Log Sr « metasedimentary rocks ■ Pg granite sills Type 1 migmatites ♦ granites Type 2 migmatites o Barrington granite (gB) □ Type 3 migmatites • Milford granite (Mg) Figure 41. Granite fractionation diagrams of the incompatible elements Ba and Sr vs. Eu/Eu* (Europium anomaly) shows a fractionation trend suggesting that the Milford (type) granite may be the source to many of the granites in this study. The suggestion that Milford granite is parental would indicate that the granites may be Permian in age and would argue against the earlier interpretation form REE patterns that there was a Barrington and a Milford group of rocks. Tower Hill granite (Cg TH), the three granite sills, Pgl-3, and Barrington granite plot away from the trend suggesting that they have different magma sources. 95

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. t— i—r ii n i j t— i— i"ii i ii | r

C3 218 □ C3 119 C2 213v C2 119 □ C3 1 * C l 2 2 9 ^ 3 C2 l ld “ vUJ 3 M l 101a LU □ C3 229 OI o C2 90

0.1 I 1 I 1111 ■4— 4 —4 . ,| 10 100 1000 Log (Ba)

Type 1 migmatites Type 2 migmatites □ Type 3 migmatites

Figure 42. Log Ba vs. log Europium anomaly with the migmatites Types 1-3 plotted on the fractionation trendline from figure 41. The migmatites do not plot on the trendline suggesting that they do not share magma sources or are not the magma source for the granites analyzed in this study.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. migmatites and Type 3 migmatites would become granites. If this were true, diagrams comparing

these rock chemistries might show the progression. The results are shown in figure 43. Figure

43a compares Types 1 and 2 migmatites and granite from locality C l 19. Figure 43b is

from locality C229 and are Type 1 and Type 3 migmatites, and figure 44 shows a Type 2

migmatite and a granite from locality C90. The diagrams show some enrichment of LREEs and

some depletion in the HREEs as would be expected in partial melt progressions from lower grade

to higher-grade migmaties. The major element plots for these rocks show some chemical

variation between different rock types. These diagrams show that the chemistry of these different

rock types is the same within analytical error (as reported in Chapter 2) suggesting that the

foliation spacing index (FSI) numbers used to subdivide the rocks does not also indicate the

textures are products of progressive migmatization events.

Modeling

Non-modal batch melting models (details in Chapter 2) were used to calculate

hypothetical migmatite compositions from a representative average of metasedimentary rock

composition. The model was built to follow the melt equation ([3], chapter 4) Ms +P1 +Qtz =

Kfs +Sil +melt, which was observed in thin sections of some of the metapelite samples. Batch

melting implies that the melt resides with the host in equilibrium until expelled as a batch.

Mineral melt partition coefficients (Kd) are not available for Sillimanite, so Kd values of 0.001

were used assuming that the probability of substitution of this element into appropriate sites of

the minerals is low. For the same reason, the Kd for the mica phase is for biotite mica, even

though equation [3] contains muscovite. Also, crystal fractionation was not considered in the

modeling of the migmatites, but will be considered in the section of this chapter concerning

granites.

The metasedimentary rocks were hypothetically melted to model (an average of) Type 1,

2 and 3 migmatite compositions. The calculated melt compositions that closest fit the averaged

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1000 too

C2119 C2119 C3119 C3 119 10

100 ■g <*

0.1

1 $102 AI2O0 Fe203 K20 0*0 N*20 hfcO 1102 P205

1000 100

C1 229 C1 229 C3 229 C3 229

100 *o

1 0.01 Lb. Ce Nd Srt Eu Gd lb Tan yb Lu SI02 AJ203 F*203 K20 C*0 N«20 h^O 1102 P205

Figure 43. Plots of rocks samples where multiple samples were taken from a single locality with various textures represented in this study on REE and wt% oxide graphs, a) Type 1 and 2 migmatite and a granite from Cl 19. b) Locality C 229 are a Type 1 and a Type 3 migmatite.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1000

C290 Cg 90

100

fi oa s 1-S IS u

Nd Sen Eu <3d TbLa Ce Nd Sen Eu <3d TbLa

100

C 290 Cg 90

SJO? A1203 Ft? 03 K20 CtO Na20 MgO T103 F303

Figure 44 compares a Type 2 migmatite and a granite from locality C 90 on REE and wt% oxide graphs.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Type 1 migmatite composition produced a melt fraction of 035, by melting modal proportions

ofminerals as follows: 0.091 and 0.091 for mica and plagioclase, 0.0 for K-feldspar and 0.82 for

quartz (Fig. 45). The model also calculates the composition of the residual solid (Cs). The

measured Type 1 composition appears to fit the plot of the residual solid better than the melt

composition with respect to the Rb, Sr and Ba concentrations. The Type 2 composition was

modeled using a higher melt fraction (F) of 0.5 and different mineral proportions. The calculated

melt was plotted against the measured compositions and the calculated residual solid composition

(Fig. 46). The measured values plot somewhat between the two calculated compositions. The

Type 3 migmatite (Cl) composition (Fig. 47) was modeled to within 350 ppm using a melt

fraction of 0.8. Errors in these models are most likely being generated from trying to model

average compositions and without any consideration to pressure and temperature conditions, but

they are good enough to assist in answering some of the fundamental questions about the MGC.

Comparisons with surrounding metasedimentary rocks

The majority of the metasedimentary rocks mapped and analyzed for this study are

shown in other works (see chapter 1) to belong to Merrimack Group (Berwick Formation) rocks.

During the mapping phase of this study it became clear that some if not all of the

metasedimentary rocks would correlate better with metasedimentary rocks of the central Maine

Terrane (Rangeley Formation) or a broader definition of Berwick Rocks would be needed. Once

the chemical analyses were completed it was possible to see if the chemistry would validate the

field observations.

REE comparison

The data for the metasedimentary rocks in and around the MGC are compared to

published data of the Silurian Rangeley, Perry Mountain Formations (Solar and Brown, 2001),

and with high, medium and low metamorphic grade Devonian Littleton Formation (Moss et al.,

100

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MGC Type 1 melt model

800

•measured mgc 1 ■ calculated mgc 1 melt • calculated mgc 1 residual solid — ». -ave MS 600

■400

200

Rb Sr Ba Cs Pb Zr Th U element Melt Observed Residual F 0.35 Mineral mode (Wt%) Calculated (ppm) Obs MGC1 Dif (Abs) Calculated (ppm) Mica 0.3 Cl Rb 88.78 Rb 134.5 45.72 Cs Rb 182.77 P R b 1.15 Plagioclase .31 Cl Sr 8.81 Sr 170 161.19 Cs Sr 308.52 P S r 1.46 K-Feldspar 0.03 Cl Ba 68.71 Ba 529 460.29 Cs Ba 776.41 P B a 2.30 Quartz 18.3 Cl Cs 0.60 Cs 3.86 3.25 Cs Cs 17.47 PCs 0.30 Cl Pb21.31 Pb 10 11.31 Cs Pb 16.16 PPb 0.16 Wt fraction in melt Cl Zr 238.77 Zr 241 2.23 Cs Zr 218.65 P Z r 0.22 M ica 0.091 Cl T h 15.67 Th 20.06 2.28 CsTh 9.01 PTh 0.10 Plagioclase 0.091 C1U 2.67 U 1.72 1.04 Cs U 2.52 PU 0.10 K-Feldspar 0.0 2 687.32 2 dif 481.49 Quartz .818

Figure 45. Comparison of theoretical (models) with actual results for MGC Type 1 migmatites. Non-modal batch partial melt models were used to melt an average metasediment composition (ave MS) into a Type 1 migmatite composition. F= melt fraction, P= bulk distribution coefficient, Cl = composition of the melt and Cs is the composition of the residual rock. Calculated results of the melt (red) are plotted against the observed (measured, in black) results and the calculated residual solid (blue). Absolute values (Abs) of the difference between the calculated and observed (Dif (Abs)) are summed in the table.

101

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MGC Type 2 melt model 1000

♦ — calculated mgc 2 melt e — calculated redidual solid mgc2 — measured mgc 2 800 * - -a v e MB

600

200

Rb Sr Ba Cs Pb Zr Th u element Melt Observed Residual F 0.5 Mineral mode (Wt%) Calculated (ppm) Obs MGC2 Dif (Abs) Calculated (ppm) Mica 0.3 Cl Rb 131.7 Rb 175.67 43.97 Cs Rb 198.37 P Rb 2.25 Plagioclase 0.31 Cl Sr 8.88 Sr 155.67 149.79 Cs Sr 393.5 P Sr 1.68 K-Feldspar 0.03 Cl Ba 288.53 Ba 539.67 4251.13 Cs Ba 959.11 P Ba 13.83 Quartz 18.3 Cl Cs 0.61 Cs 4.71 3.56 Cs Cs 22.2 P C s 1.42 Cl Pb 31.97 Pb 13.0 18.97 Cs Pb 15.04 P Pb 1.02 Wt fraction in melt Cl Zr 337.27 Zr 188.0 149.27 Cs Zr 209.29 PZr 1.05 Mica 0.455 Cl Th 16.61 Th 11.99 4.62 Cs Th 8.06 PTh 0.46 Plagioclase 0.0 Cl U 2.77 U 1.29 1.48 Cs U 2.39 PU 0.37 K-Feldspar 0.273 2 619.78 2 dif 726.37 Quartz .27 Figure 46. Comparison of theoretical (models) with actual results for MGC Type 2 migmatites. Non-modal batch partial melt models were used to melt an average metasedimentary rock composition (ave MS) into a Type 2 migmatite composition. F= melt fraction, P= bulk distribution coefficient, Cl = composition of the melt and Cs is the composition of the residual rock. Calculated results of the melt (red) are plotted against the observed (measured, in black) results and the calculated residual solid (blue). Absolute values (Abs) of the difference between the calculated and observed (Dif (Abs)) are summed in the table. 102

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MGC Type 3 melt model 800

• - -ave MS ■e— calculated mgc3 — measured mgc3 600

£ Q. Q.

200

Rb Sr BaCs Pb Zr Th U element Melt Observed F 0.8 Mineral mode (Wt%) Calculated (ppm) Obs MGC3 Dif (Abs) Mica 0.3 Cl Rb 130.0 Rb 85.0 45.0 PRb 1.76 Plagioclase 0.31 Cl Sr 9.81 Sr 194.0 184.19 P Sr 4.15 K-Feldspar 0.03 Cl Ba 745.61 Ba 738.17 7.44 PBa 10.44 Quartz 18.3 Cl Cs 0.61 Cs 4.50 4.39 PCs 1.03 Cl Pb 29.99 Pb 12.70 17.32 PPb 0.% Wt fraction in melt Cl Zr 263.67 Zr 201.0 62.67 PZ r 0.78 Mica 0.33 Cl Th 12.35 Th 0.94 11.42 PTh 0.34 Plagioclase 0.181 C1U 2.22 U 13.89 11.66 PU 0.29 K-Feldspar 0.217 2 344.11 Quartz 0.28

Figure 47. Comparison of theoretical (models) with actual results for MGC Type 3 migmatites. Non-modal batch partial melt models were used to melt an average metasediment composition (ave MS) into a Type 3 migmatite composition. Calculated results (red) are plotted above against the measured (Obs) results (black) and the absolute values (Abs) of the difference between the calculated and observed (Dif (ABs)) are summed in the table. F= melt fraction, P= bulk distribution coefficient.

103

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1995). No published Berwick Formation data was found. Figure 48a shows the range of the Rare

Earth Element data for some of the metasedimentary rocks in this study and the two analyzed

MGC Type 1 migmatites shaded in blue for comparison with the other analyses. Figure 48b

compares the range of these data with REE analyses from the Rangeley and Perry Mountain

Formations (Solar and Brown, 2001). Figure 48b compares the range of data from this study with

REE analyses for the Devonian Littleton Formation (Moss et al., 1995).

Trace element comparison

Figure 49a shows several trace element analyses for the metasedimentary rocks from this

study with the range of the data shaded in blue for (49b) comparison with trace element data from

the Silurian Rangeley, Perry Mt. and Smalls Falls Formations (Solar and Brown, 2001).

Based solely on the diagrams 48 and 49, it would appear that there is little difference

between the chemistries of these rocks. In the field the Berwick Formation is predominately

metagraywacke in composition. The CMT rocks (Silurian Perry Mountain and Rangeley

formations) are predominately schists with metawacke layers and pelitic schists. The

metasedimentary rocks mapped in this study are typically a schist and metawacke layered rock.

Based on lithologic and chemical similarities the metasedimentary rocks in the study area would

correlate better with CMT rocks than with the massive metagraywacke Berwick Formation.

However, as this topic is not the focus of this work and critical Merrimack Group chemical data

for comparison are unavailable, I respectfully leave the question for future consideration.

Summary and conclusions

The REE patterns indicate that the different types of rocks in the study area are not

particularly dissimilar. The migmatites and granites plot over the field created by the

metasedimentary rocks with the only distinctions being Europium anomalies and slopes of the

plots. The granites are peraluminous, and based on tectonic discrimination plots were most likely

104

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1000

C169 C170

P101 C79

100 C 15 C61

g Z 10

a. La Ce Nd Sm Eu Sd 1b Tm Vb Lu

Figure 48a. Ree analyses from this study of the metasedimentary rocks. The range of the data is shaded in blue to aid with comparisons between the rocks in this study with Central Maine Terrane rocks from other studies. Metawackes are blue, schists are gold.

105

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Central Maine Terrane 1000

R angeley

■S— Perrry M t 100 T9 NV s ¥ II

La Ce Nd Sm Eu Gd Tb Y Yb Lu

Devonian Littleton Formation

1000

low grade

100 W0>

IIii

La Ce Nd Sm Eu Gd Tb Y Yb Lu Figures 48b and c. Comparison of Ree data between metasedimentary rocks (b) analyzed in the study (blue shaded area) with data from analyses of Silurian Rangely and Perry Mountain Formations (Solar and Brown, 2001) and (c) from low, medium and high metamorphic grade Devonian Littleton Formation (Moss et al., 1995). 106

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. MGC metasedimentary rocks tooo

c 1 7 0 C 169

* - - c 7 9

too c 15 c61

SrRb ¥ Zr Nb Pb Central Maine Terrane

tooo

— - Sp - B - .S p —■ ■ §p — *- - Sp too -Sp - B - -Sp — Ssf

Rb SrV ZrNb Pb

Figure 49. Comparison of metasedimentary rock data from the MGC (49a) with published analyses from the Silurian Central Maine Terrane rocks (49b) of the Rangeley (Sr), Perry MT. (Sp) and Smalls Falls (Ssf) Formations (Solar et al., 2001) superimposed onto the range of MGC data (blue)

107

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. created in continent-continent collision and active continental margin environments.

Preliminaryconclusions based on REE patterns of the granites and migmatites indicated two sub

groups of rocks; one was related to the Barrington Granite and the other the Milford Granite.

Later interpretation based on fractionation patterns indicated that the granites were probably all

related to a parental source of Milford Granite type magma with the exception of some of the

granite sills, Tower Hill granite, Cg 26-2, Cg 21 and the Barrington granite. Plotting the

migmatite compositions against the trendline generated in the granite fractionation diagram

(Fig. 42) suggests that the migmatite and the granites are unrelated.

Modeling of the migmatites using basic partial melt calculations indicates that a low-melt

-fraction partial melt created the Type 1 migmatite. However it was difficult to match the

measured composition to the calculated melt composition with respect to Rb, Sr and Ba. The

results of this model could be interpreted to show that the bulk rock composition for the MGC

Type 1 is a better fit as a restite than a migmatite based on the calculated residual rock

concentrations. The best Type 2 migmatite model shows the measured bulk rock compositions

are neither melt nor residual indicating that probably these rocks are a combination of both.

Perhaps analyzing Type 2 migmatites should involve separating leucosome and melanosome and

analyzing them along with bulk rock analysis. Type 3 migmatites modeled well as a high melt

fraction migmatite suggesting that these rocks were either deeper or froze in areas where melt

was migrating and collecting or both.

Finally, in an effort to identify the metasedimentary rocks that were melted to produce

the migmatites, the metasedimentary rocks in the study were compared to similar bounding

metasedimentary rocks. The comparison of lithology with the CMT indicates an affinity of the

MGC to the CMT and not Merrimack Group (Berwick Formation) rocks as they are

predominately pelitic schists and metawackes but more chemical data are needed.

108

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 6

CONCLUSIONS, INTERPRETATIONS AND FURTHER WORK

The hypothesis (Chapter 1) that the Massabesic Gneiss Complex (MGC) underwent

melting episodes in which surrounding metasedimentary rocks were migmatized possibly as

many times as three, has been carefully examined. The migmatizations produced melts that

generated migmatite textures and/or accumulated to become the granite bodies found in and

around the MGC. It was also hypothesized that if the metasedimentary rocks were of the Central

Maine Terrane or Merrimack Group, the migmatization of these rocks must post-date the late-

Proterozoic designation of the MGC and would make the majority of the rocks in the MGC

younger than previously thought.

The dominant biotite foliation (Ch. 3) in the migmatites largely shares the same

orientation as the foliation in the metasedimentary rocks and suggests that the migmatite foliation

is relic. That combined regional foliation in both migmatites and metasedimentary rocks defines

a structural arch that plunges slightly to the northeast. Weak foliation in the foliated granites

(refer to Fig. 10, p. 32 and Fig. 14, p. 40) may have formed during intrusion to a structurally

higher level. The fold axes orientations recorded in Candia support later folding (or refolding)

that may be related to emplacement of granites or subsequent tectonic movement along the

Campbell Hill- Hall Mt. Fault (CHHMF). The CHHMF and the Flint Hill Fault (FHF) show

evidence for ductile oblique sinistral shear and brittle dextral movement. Cross-cutting

relationships indicate that brittle motion is younger than the ductile motion. The CHHMF and the

FHF represent late normal faults in which the motion was northwest side down for the CHHMF

109

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and southeast side down for the FHF; the arched MGC is thus a horst. The petrography of the

MGC rocks indicates prograde metamorphism in which muscovite is dehydrated (producing

H20); ultimately the rocks pass through the granite melt curve (refer to CH. 4, Fig 30, p. 69, Fig.

31, p. 72) producing partially melted metasedimentary rocks (migmatites). Later at lower

pressures, MGC rocks pass through the biotite melting curve in decompression melting reactions

producing Type 3 migmatites and granites. Ultimately, the rocks cool down, pass back through

the granite melt curve and all melts solidify.

The Rare Earth Element (REE) chemical data (Ch. 5) from several samples of

metasedimentary rocks, migmatites and granites taken in and around the Candia quadrangle (Fig.

32) define two categories based on the slopes of the REE patterns and Europium anomaly

similarities. This same pattern is not supported, however, by analyses of the granite chemistry.

Fractionation plots (Fig. 41, p. 95) of the peraluminous granites indicate that most are

fractionated products of a Milford granite (like) source with the exceptions of two granite sills

and a dike (samples Pg 1, Pg 2 and Pg 3, respectively) and the Barrington granite. The bulk rock

migmatites do not plot on the fractionation trendline (Fig. 42, p. 96) of the granites suggesting

that they are not related to the granites. Tectonic discrimination diagrams of the granites indicate

they were produced in active continental margins and continent-continent collision (Fig. 38, p.

91).

Partial melt modeling (Figs. 45-47) of the metasedimentary rocks was interpreted such

that Type 1 migmatites are textures relating to removal of melt from the rock and are restitic. The

Type 2 and 3 migmatite textures reflect higher amounts of leucocratic material mobilized to

locations of relatively lower pressure. The metasedimentary rocks that were melted are

tentatively correlated with those of the Central Maine Terrane based on chemical similarities

(Figs. 48 and 49) and lithology (Ch. 3).

110

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Conclusions

The MGC underwent at least three melting episodes: one that produced the Barrington

granite, major migmatization event(s) producing three migmatite types and a third that produced

granite bodies within the MGC. Bulk rock analyses of the migmatites indicates that they did not

contribute magmas to those later granites but future analysis of the leucosome from these

migmatites will test this important question. The migmatites appear to be products of melting

Silurian metasedimentary rocks. The migmatites are younger than late-Proterozoic and the

majority of the granites are most likely Permian in age. The earlier model (Fig. 8, p. 22) is

reconstructed to relate to the findings of this study (Fig. 50).

Tectonic history

The metasedimentary rocks that become MGC migmatites are folded during orogenesis

(Fig. 51) thickening the crust and at metamorphic conditions that exceed temperatures and

pressures sufficient to partially melt them Once melting begins (Fig. 52) and the critical melt

fraction is surpassed, melts mobilize and leave the rocks that become Type 1 migmatites and

accumulate in rocks that become Type 2 and Type 3 migmatites. Progression of the melting

causes the MGC melt-bearing rocks to rise buoyantly producing a regional arch. Normal faulting

along the Campbell Hill-Hall Mt. Fault and the FlintHill Fault exposes the MGC as a horst-like

structure (Fig. 53).

Timing

The dominant S3 biotite foliation of the metasedimentary rocks is probably Devonian;

however, high grade metamorphism is known to have continued into the Permian (Eusden and

Barreiro, 1988), but deformation of early folds as suggested by the fold axes in the Candia

quadrangle is most likely late Carboniferous or early Permian. Nappe-style southeast verging

folds as depicted in Fig. 51 are most likely Acadian (Devonian) as similar folds north of the study

area are truncated by the Nonesuch River Fault (correlated with the CHHMF) in the late

111

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Muscovite and biotite partial melting, migration and pooling of magma

I ?

Partial melting of parent gneisses by dehydration of muscovite Devonian 1 ■ Silurian metasedimentary rocks NH Plutonic Series

late-Proterozoic Fig. 50. Model from figure 8 reconstructed to reflect findings in this study. MGC Type 1 migmatites are partially melted metasedimentary rocks producing in situ stromatic migmatites (restites ?). Further melting caused by dehydration of biotite produces larger melt fractions which mobilize and contribute to MGC Type 2 and 3 migmatites. It appears unlikely that magmas from these migmatites were the source for the Milford type granites of Permian age. 112

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Temperature (°C)

♦PI, Qo

Type 3 migmatites and granites et (kms) Depth

♦H.QB & MGC a 3 £ 6 -

♦H.Qe

NW SE -i 0

15 15

Figure 51. Folding and or faulting during orogenesis thickens the crust metamorphosing MGC rocks progressing to temperatures and pressures sufficient to partial melt the rocks. Star represents location of the rocks in pressure temperature space.

113

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Temperature (°C) 400 500 600 700 800

MGC types la n d 2

# 1 *■ * * + + _ ■ NW + * * + + ** <■ ' N

Figure 52. Shows temperatures and pressures of MGC rocks as they pass through the granite melt curve and reaction [3]. Migmatites Type 1 and 2 are being produced and the presence of melts creates gravitational instability and arches the MGC. Star represents location of the rocks in pressure temperature space.

114

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Temperature ( °C)

3 migmatites ad granites

Figure 53 shows late faulting of the two major faults (CHHMF, FHF) allows the MGC to uplift as a horst-like structure bringing the rocks up and allowing them to melt by decompression melting (reaction [4]) producing MGC Type 3 migmatites. After the rocks cool to temperatures below the granite melt curve they freeze. The MGC is exposed as a narrow region of melted metasedimentary rocks and granites. Star represents location of the rocks in pressure temperature space.

115

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Devonian or Carboniferous (Eusden et al., 1987). The foliation in the adjacent granites is most

likely related to the arching phase of the anticlinorium here interpreted as late Carboniferous to

Permian.

Continued work

Future work in the Massabesic Gneiss Complex should include further mapping at a scale

of 1:24,000 or larger, of the area southwest of the Candia quadrangle at least to the Pinardville

quadrangle. This area has not been mapped in detail since 1966. Attention should be paid to the

migmatite contacts and relations between them and the granites and metasedimentary rocks.

Larger scale work could also be done in areas with good exposure to examine the foliation as a

flow indicator in the Type 3 migmatites.

The present work provides a database into which new data from continuing chemical

investigations could be added to enhance the history o f die Massabesic Gneiss Complex. A

comprehensive study of the metasedimentary rocks within and those that bound the MGC

(Rangeley, Perry Mountain of the Central Maine Terrane, Berwick Formation and the Merrimack

Group) focusing on chemistry and structural details would help to redefine the tectonic boundary

between the Central Maine Terrane and Merrimack Group rocks. A chemical database for the

Berwick Formation is badly needed. Analyses of leucosomes and more analyses of granites that

are mapped as Barrington might lend new insight to sources for the granite magmas and

comparisons with the one Barrington sample in this study. The addition of closer spaced

thermobarometry studies to increase the P-T space resolution would be useful. Careful mapping

and observations, additional geochronological data of the granites and granite sills would

constrain some of the events listed in this work, and some isotopic (Nd/Sm...) work that would

shed more light on the Proterozoic nature of the MGC.

116

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. REFERENCES

117

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. REFERENCES

Aleinikoff, J. N., Zartman, R. E. and Lyons, J. B., 1979, U-Th-Pb Geochronology of the Massabesic Gneiss and the granite near Milford, south central New Hampshire: New evidence for Avalonian Basement and Taconic and Alleghenian disturbances in Eastern New England: Contributions to Mineralogy and Petrology, v. 71, p. 1-11.

Aleinikoff, J. N., Walter, M. and Fenning, M.C., 1995, U-Pb ages of zircon, monazite and sphene from rocks of the Massabesic Gneiss Complex and the Berwick Formation, New Hampshire and Massachusetts: Geological Society of America, abstracts with programs, Northeastern section, v. 27, no. 1, p. 26.

Allard, S. T., 1998, Metamorphic petrology, structure and bedrock geology of the Berwick Formation within the Mt. Pawtuckaway 7.5 minute quadrangle, Raymond and Nottingham, NH: unpublished M.S. thesis, University of New Hampshire, 125 p.

Almendinger, R. W., 2001, Stereonet 6, computer program.

Anderson, T. R., 1978, The Northern termination of the Massabesic Gneiss, New Hampshire: unpublished M.A. thesis, Dartmouth College, Hanover NH, 111 p.

Armstrong, T. R., Aleinikoff, J. N. and Burton, W. C., 1999, Structural and geochronologic constraints on the tectonothermal evolution of the Massabesic Gneiss Complex, southeastern New Hampshire: Geological Society of America Abstracts with Programs, v. 31, p.44.

Arth J.G., 1976, Behavior of trace elements during magmatic processes-a summary of theoretical models and their applications: Journal of Res U.S. geological survey, v. 4, pp 41-47.

Ashworth, J. R., 1985, Introduction. In: Migmatites (ed. Ashworth, J. R.): Balckie Glasgow, p. 1-35.

Ashworth, J. R. and McLellan, E. L., 1985, Textures: Migmatites (ed. Ashworth, J. R.): Balckie Glasgow, p. 180-203.

Batchelor, R. A. and Bowden, P., 1985, Petrogenetic interpretation of granitoid rock series using multicationic parameters: Chemical Geology, v. 48, p. 43-55.

Belanger, T., 2003, Polydeformation and fault relocation at Saddleback Mountain, southeastern New Hamspshire: unpublished senior thesis, University of New Hampshire, 17 p.

118

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Besancon, J., Gaudette, H. E.,and Naylor, R. S., 1977, Age of the Massabesic Gneiss, southeastern New Hampshire: Geological Society of America, abstracts with programs, v. 9, No. 3, p. 242.

Berry, H. N. and Osberg, P. H. 1988, A stratigraphic synthesis and tectonic model for pre- Devonian rocks of Maine and western New Brunswick: Geological Society of America Abstracts and Programs, v. 20, p. 6.

Billings, M. P. and Cleaves, A. B., 1934, Paleontology of the Littleton area, New Hampshire: American Journal of Science, v. 28, pp. 412-4-38.

Billings, M. P., 1956, The geology of New Hampshire, Part II, Bedrock Geology: New Hampshire State Planning and Development Commission, 200 p.

Bradley, D. C. and Hanson, L. S., 2002, Paleocurrent analysis of a deformed Devonian foreland basin in the Northern Appalachians, Maine, U.S.A.: Sedimentary Geology, v. 148, p. 425-447.

Cartwright, I., 1988, Crystallization of melts, pegmatite intrusion and the Inverian retrogression of the Scourian complex, north-west Scotland: Journal of Metamorphic Geology, v. 6, p77-93.

Dell' Angelo, L. and Tullis, J., 1988, Experimental deformation of partially melted granitic aggregates: Journal of Metamorphic Geology, v. 6, p. 495-515.

Dorais, M. J., Wintsch, R. P and Kunk, M. J., 2003, P-T-t Conditions of the Massabesic Gneiss Complex, New Hampshire: Alleghanian loading and exhumation of a portion of the Avalon Terrane: Geological Society of America abstracts, Northeast section Annual Meeting, Halifax, Nova Scotia, v. 35, p. 24-25.

Dorais, M. J. and Wintsch, R. P., 2001, A Laurentian provenance of the Merrimack belt, New England Appalachians: Geological Society of America abstracts, Annual Meeting, Boston, p.261.

Dorias, M. J. and Paige, M. L., 2000, Regional geochemical and isotopic variations of northern New England plutons: implications for magma sources and for Grenville and Avalon basement-terrane boundaries: Geological Society of America Bulletin, v.112., p.900-914.

Dwight B. C., and Hanson, L. S., 2002, Paleocurrent analysis of a deformed Devonian foreland basin in the Northern Appalachians, Maine, USA: Sedimentary Geology, v. 148, p. 425- 447.

Eby, G.N., 1996, PetMin3, computer program.

Eby, G. N., 1985, Geology and geochemistry of the Mount Pawtuckaway ring-dike complex, White Mountain igneous province, N H .: Geological Society of America abstracts with programs, v. 17. p. 17.

119

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Emerson, B. K., 1890, Porphyritic and gneissoid granites in Massachusetts: Geological Society of America Bulletin, v. 1, p. 559-561.

Escamilla-Casas, J. C., 2003, Bedrock geology of the seacoast region of New Hampshire, U.S.A.: unpublished Ph.D. thesis, University of New Hampshire,

Eusden, J. D., 1984, The bedrock geology of part of the Alton, New Hampshire and Berwick, Maine 15’ quadrangles. Unpublished MS thesis, University of New Hampshire.

England, P. C. and Thompson, A. B., 1984, Pressure-temperature-time paths of regional metamorphism, I, Heat transfer during the evolution of regions of thickened continental crust: Journal of Petrology, v. 25, p. 894-928.

Eusden, J. D., 1988, The bedrock geology of the Gilmanton 15 minute quadrangle New Hampshire: unpublished Ph.D. thesis, Dartmouth College.

Eusden, J. D. and Barreiro, B., 1988, The timing of peak high-grade metamorphism in central- eastern New England: Maritime Sediments and Atlantic Geology, v. 24, pp. 241-255.

Eusden, J. D., Bothner, W.A. and Hussey, A. M. II, 1987, The Kearsarge-Central Maine synclinorium of southeastern New Hampshire and southwestern Maine: Stratigraphic and structural relations of an inverted section: American Journal of Science, v.287, p. 242- 264.

Eusden, J. D., 1984, The bedrock geology of part of the Alton, New Hampshire and Berwick, Maine 15’ quadrangles. Unpublished MS thesis, University of New Hampshire.

Fagan, T. A., 1986, Metamorphic petrology and structure of a transect from the Berwick Formation into the Massabesic Gneiss Complex, Southeastern New Hampshire: Unpublished M.S. thesis, University of New Hampshire, 108 p.

Fitzsimons, I. C. W., 1996, Metapelitic migmatites from Brattstrand Bluffs, East Antarctica- metamorphism, melting and exhumation of the mid crust: Journal of Petrology, v. 37, n. 2, p. 395^414.

Freedman, J., 1950, Stratigraphy and Structure of the Mt. Pawtuckaway Quadrangle, southeastern New Hampshire: Bulletin of the Geological Society of America . v. 61, p. 449-492.

Foland, K. A.,Quinn, A. N., and Giletti, B. J., 1971, K-Ar and Rb-Sr Jurassic and Cretaceous ages for intrusives of the White Moutnain Magma Series, northern New England: American Journal of Science, v. 270, p. 321-330.

Govindaraju, K., 1989, 1989 compilation of working values and sample description for 272 geostandards. In Govindaraju, K. (ed.) Geostandards newsletter, v. 19, Special Issue, 113.

Gromet, P. J., 1989, Avalonian terranes and late Paleozoic tectonism in southeastern New England; Constraints and problems: Geological Society of America Special Paper 230, p. 193-211. 120

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hanson, L. S. and Bradley, D. C., 1993, Late Silurian to early Devonian paleogeography of the Kearsarge-Central Maine Basin revealed by paleocurrents and sedimentary facies: Geological Society of America Abstracts with Program, v. 25, n.6, p. A- 360 Hanson, L. S. and Bradley, D. C., 1989, Sedimentary facies and Tectonic interpretation of the lower Carrabasett Formation, North-Central Maine, in Tucker, R. D. and Marvinney, R. G., eds., Studies in Maine Geology: Maine Geological Survey, v. 2, p. 101-126.

Harris, N. B. W., and Ingor, S., 1992, Trace element modeling of pelite-derived granites: Contributions to Mineralogy and Petrology, v. 110, pp 46-56.

Hatch, N. L., Moench, R. and Lyons, J.B., 1983, Silurian-Lower Devonian stratigraphy of eastern and south central New Hampshire. American Journal of Science, v. 283, pp. 739- 761.

Hatch, N. L. and Moench, R. H., 1984, Bedrock geologic map of the wilderness and roadless areas of the White Mountain National Forest, Coos, Carrol and Grafton Counties, New Hampshire: U.S. Geological Survey Miscellaneous Field Studies Map MF-1594-A, scale 1:125,000.

Holdaway, M. J., 1971, Stability of andalusite and the aluminum silicate phase diagram: American Journal of Science, v. 271, p. 97-131.

Hussey, A. M., II, 1985, The bedrock geology of the Bath and Portland 2 degree map sheets, Maine Geological Survey, Open file report, no. 85-87,80p.

Johannes, W., 1985, The significance of experimental studies for the formation of migmatites. In: Migmatites (ed. Ashworth, J. R.): Balckie Glasgow, p. 36-85.

Katz, W. J. Jr., 1917, Stratigraphy in southwestern Maine and southeastern New Hampshire: U.S. Geological Survey Professions! Paper, v. 108-1, p. 165-177.

Kelly, W. J. Jr., 1980, An isotopic study of the Massabesic Gneiss Complex: Southern NH: Unpublished M.S. thesis, University of New Hampshire, 121 p.

Kerwin, C.M., Allard, S. T., Laird, J. and Bothner, W. A., 2004, The rocks speak and the Proterozoic shrinks, results from mapping and REE chemistry of the Massabesic Gneiss Complex, N.E.I.G.C.

Kerwin, C. M., 2001, Geology of the Candia 7.5 minute quadrangle: New Findings: Geological Society of America abstracts with Programs, v.33, p. 30.

Kerwin, C. M., 2000, Structure and bedrock geology of the Massabesic Gneiss Complex in Southern New Hampshire: unpublished M.S. thesis, University of New Hampshire, 70 p.

Kerwin, C. M., 1999, Regional structure of the Massabesic Gneiss Complex and quantification of the degree of migmatization in the Milford quadrangle, southwestern New Hampshire: Geological Society of America Abstract with Programs, v. 31, P-A-27.

121

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Larson, T. E., 1999, The metamorphic history of the Massabesic Gneiss Complex and the Berwick Formation of southern New Hampshire: Unpublished M.S. thesis, University of New Hampshire, 117 p.

Le Breton, N., and Thompson, A. B., 1988, Fluid absent (dehydration) melting of biotite in metapelites in the early stages of crustal anatexis: Contributions to Mineralogy and Petrology v.99, p.226-237.

Luth, W., Johns, R.H. and Tuttle, O.F., 1964, The granite system at pressures of 4 to 10 kilobars: Journal of Geophysical Research, v. 69, p. 759-773.

Lyons, J. B., Bothner, W . a., Moench, R. H. and Thompson, J. B. Jr., 1997, Bedrock geologic map of New Hampshire, 2 sheets, U.S. Geological Survey, scale 1:250,000

Lyons, J. B., Boudette, E. L. and Aleinikoff, J. N., 1982, The Avalonian and Gander Zones in central and eastern New England: St. Julien, P. and Beland, J., (eds), Major structural zones and faults of the Northern Appalachians, Geological Society of Canada, Special Paper, v. 24, p. 43-66.

Maczuga, D., E., 1981, The petrology and geochemistry of the Fitchburg Plutonic Complex, central Massachusetts: Unpublished M.S. thesis, University of Massachusetts, Amherst, pp. 128.

Manning D. A. C. and Pichavant, M., 1983, The role of Fluorine and Boron in the generation of granitic melts: Migmatites, Melting and Metamorphism, Proceedings of the geochemical group of the Mineralogical Society, EDS: Atherton, M. P. and Gribble, C. D., Shiva Publishing.

McKenzie, D., 1984, The generation and compaction of partially molten rock: Journal of Petrology, v. 25, p. 713-765.

McLellan, E., L., 1983, Contrasting textures in metamorphic and anatectic migmatites: an example from the Scottish Caledonides: Journal of Metamorphic Petrology, v. 1, p.241- 262.

McHone, J., G., 1996, Broad terrane Jurassic flood basalts across northeastern North America: Geology, v. 24, n. 4, p. 319-322.

Menhert, K. R., 1968, Migmatites and the origin of granites. Elsevier, Amsterdam, pp. 393.

Moench, R., 1971, Geologic maps of the Rangeley and Phillips quadrangles, Franklin and Oxford Counties, Maine: U.S. Geological Survey Miscellaneous Geologic Investigations, map 1-106.

Mose, D., and Wenner, D., 1980, Rb-Sr Whole-rock isochrons and 180/160 data from selected Appalachian plutons; Geological Society of America, Southeastern Section meeting abstracts with program, v. 12, p.202.

122

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Moss, B., Haslin, L.A., Dymek, R.F. and Shaw, D.M., 1995, Redetermination and reevaluation of compositional variations in metamorphosed sediments of the Littleton Formation, New Hampshire; American Journal of Science, v. 295, no.8, p. 988-1019.

Nakamura, N., 1974, Determination of REE, Ba, Fe, Mg, Na and K in carbonaceous and ordinary chondrites: Geochimica Cosmochimica Acta, v. 38, p.757-775.

Nash, W. P., and Crecaft, H.R., 1985, Partition coefficients for trace elements in silicic magmas: Geochimica Cosmochimica Acta, v. 49, p. 2309-2332.

Nyman, M. W., Pattison, D. R. M. and Ghent, E. D., 1995, Melt extraction during formation of K-Feldspar and Sillimanite Migmatites, west of Revelstoke, British Columbia: Journal of Petrology, v. 36, n. 2, p 351-372.

Osberg, P. H., Hussey, A. M. II and Boone, G.M., (eds.), 1985, Bedrock Geologic Map of Maine, Maine Geologic Survey, Scale 1:500,000.

Pearce, J. A., Harris, N. B. W. and Tindle, A. G., 1984, Trace element discrimination diagrams for the tectonic interpretation of granitic rocks: Journal of Petrology, v. 25, p. 956-983.

Pouliot, J. B., 1994, A geophysical and geological investigation along the eastern margin of the Massabesic Gneiss Complex, Raymond, New Hampshire: Geological Society of America, abstracts with programs, v. 26, no. 3, p. 83.

Richardson, S. W., 1968, Stuarolite stability in a part of the system Fe-Al-Si-O-H: Journal of Petrology, v. 9, p. 467-489.

Rickerich, S. F., 1983, Sedimentology, stratigraphy and structure of the Kittery fromation in the Portsmouth, New Hampshire region: unpublished M.S. thesis, University of New Hampshire, pp. 115.

Robinson, P. and Goldsmith, R., 1991, Stratigraphy of the Merrimack belt, central Massachusetts: in Hatch, N. L. (ed.) The Bedrock Geology of Massachusetts, G1-G37, U.S. Geological Survey Professional Paper, p. 1366-J.

Robinson, J. C., 1989, A brittle fracture analysis of the Flint Hill Fault Zone, Southeastern New Hampshire, unpublished M.S. thesis, University of Massachusetts,pp. 91

Rodgers, J., compiler, 1985, Bedrock Geological Map of Connecticut: Connecticut Natural Resources Atlas Series, Bedrock Geologic Map, 2 sheets, scale 1:125,000

Rollinson, H., 1993, Rare earth elements: Using Geochemical Data. Longman Singapore Publishers, pp.351

Sederholm, J. J., 1913, Uber die Enstehung der migmatitisschen Gesteine: Geo. Rundschau., Band TV, pp. 174-185.

Schulz, J. E, 2004, Metamorphic petrology and structural implications of the calcareous (Unnamed) member of the Berwick Formation with the Epping 7.5 minute quadrangle, southeast New Hampshire, pp. 123

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Solar, G. S. and Brown, M., 2001, Petrogenesis of migmatites in Maine, USA: possible source of peraluminous leucogranite in plutons: Journal of Petrology, v. 42, no. 4, p. 789-823.

Spear, F. S., 1993, Metamorphic phase equilibria and pressure temperature-time paths: Mineralogical Society of America Monograph.

Spear, F. S., Kohn, M. J., and Cheney, J. T., 1999, P-T paths from anatectic pelites: Contributions to mineralogy and petrology. V. 134, p. 17-32.

Stewart, D. B., 1989, Crustal processes in Maine: American Mineralogist, v. 74, p. 698- 714.

Streckeisen, A. L., 1976, To each plutonic rock its proper name: Earth Science Reviews, v. 12, pp. 1-33.

Sriramadus, A., 1966, The geology of the Manchester Quadrangle, New Hampshire: N.H. Department of resources and Economic Development Bulliten, v. 2, pp.78.

Sundeen, D. A. and Huff, M. C-, 1992, Petrography, petrology and K-Ar geochronology of hyperbyssal mafic and silicic Mesozoic igneous rocks in southeastern New Hampshire: Eastern North American Mesozoic Magnetism, EDS: Puffer, J. H. and Ragland P. C., Geological Society of America Special Paper no. 268, p. 75-94.

Thompson, A. B., 19%, Fertility of crustal rocks during anatexis: Transactions of the Royal Society of Edinburgh: Earth Sciences, v, 87, p. 1-10.

Thompson, A. B. and Connolly, J. A. D., 1995. Melting of the continental crust: some thermal and petrological constraints on anatexis in continental collision zones and other tectonic settings: Journal of Geophysical Research, v. 100, p. 15565-15579.

Thompson, A. B., 1982, Dehydration melting of pelitic rocks and the generation of H20 undersaturated granitic liquids: American Journal of Science, v 282, p. 1567-1595.

Tracey, R. J. and Robinson, P., 1983, Acadian migmatite types in pelitic rocks of central Massachusetts: Atherton, M. P. and Gribble, C. D., (eds.) Migmatites, melting and metamorphism; Proceedings/ Meeting High Grade metamorphism, migmatites and melting of the Geochemical Group of the Mineralogical Society of the University of Glasgow, P. 163-173.

Tuttle, O. F. and Bowen, N. L., 1958, The origin of granite in the light of experimental studies in the system NaAlSi30 5 - KAlSi3Og - Si02 - H20: Mem. Geological Society of America, No. 74.

Unger, J. D., Stewart, D. B. and Phillips, J. D., 1987, Interpretation of migrated seismic reflection profiles across the northern Appalachians in Maine: Geophysical Journal of the Royal Astronomical Society, v. 89, p. 171-176.

Walsh, G. J. and Clark, F. C. Jr., 1999, Bedrock Geologic map of the Windham quadrangle, Rockingham and Hillsborough Counties, New Hampshire: U. S.G.S. open file report 99- 8, pp. 27

124

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Weber, E. B., Barbey, P., Cuney, M. and Martin, H., 1985, Trace element behavior during migmatization. Evidence for a complex melt-residuum-fluid interaction the St. Malo migmatite dome (France): Contributions to Mineralogy and Petrology, v. 90, p. 52-62. Wintsch, R.P., Sutter, M. J., Kunk, J. N., Aleinikoff, J. N. and Dorais, M. J.,1992, Contrasting P- T-t paths: Thermochronologic evidence for the final assembly of the Avalon composite terrane in the New England Appalacians: Tectonics, v. 11, no. 3, pp. 672-689.

Whitaker, L. A., 1983, The geology of the Saddleback Mt. area, Northwood quadrangle, southeastern New Hampshire: unpublished M.S. thesis, University of New Hampshire, pp. 131

Wood, D. A., Tamey, J., Varet, J., Saunders, A. D., Bougault, H., Joron, J. L.,Treuil, M. and Cann, J. R., 1979b, Geochemistry of basalts drilled in the North Atlantic by IPOD Leg 49: implications for mantle heterogeneity: Earth Planet Science Letters, v. 42, P. 77-97.

Zartman, R.E., 1988, Three decades of geochronological studies in the New England Appalachians: Geological Society of America Bulletin, v. 100, p. 1168-1180.

Zen, E., Goldsmith, R., Ratcliffe, N. M., Robinson, P. and Stanley, R. S., 1983, Bedrock Geologic map of Massachusetts: United States Geological Survey, scale 1:250,000.

125

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDICES

126

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX A XRF CALIBRATION CURVES

127

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration ______Application ______Beads 14MajorFP2

0.80

0 .20 -

0.20 0.40 0.60 0.80

Theor. int

No. Sample ______Meas. int. ______Theor. int. ______Calculated______Deviation Corrected 1 AGV-2 0.60037 0.61719 0.62565 0.00846 n BHVO-2 0.32613 0.31883 0.34449 0.02567 3 DNC-I 0.26047 0.27395 0.28641 0.01245 4 G-2 0.54020 0.60164 0.56396 -0.03768 5 * GSP-2 0.28886 0.40994 0.30628 -0.10365 6 RGM-I 0.60260 0.60100 0.62793 0.02693 7 SDC-1 0.29914 0.30598 0.31682 0.01084 8 W-2 0.30228 0.31702 0.32004 0.00302 9 AGV-2 b 0.60462 0.61719 0.63001 0.01282 10 DNC-I b 0.27097 0.27395 0.28794 0.01399 1 1 G-2 b 0.53548 0.60164 0.55912 -0.04252 12 GSP-2 b 0.33581 0.40994 0.35442 -0.05552 13 RGM-I b 0.59785 0.60100 0.62307 0.02207 14 W-2 b 0.29881 0.31702 0.31649 -0.00053

*:Deselected samples RigjC*§CU

128

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration Application Beads 14MajorFI;2

Clement line Me-KA

...... ll ~ Aim* - Him + C 4.0- I.inearf straight) A 7.54365e-00l 4.2*833e-0<)2 f Accuracy 3.77929e-002 Corr factor 9.99506e-00 I 9 Inter, sld No Fixed point No 2 . 0- ■ Weiehtinu Normal

u-o-l------1------1...... - — r - ■ 0.0 1.0 2.0 3.0 4.0

Theor. ini

o. Sample Meas in I Theor. int. Calculated Deviation Corrected I AGV-2 0.66724 0.55748 0.54622 -0.01126 BHVO-2 2.96545 2.20790 2.27991 0.07202 y DNC-1 4.06735 3.12082 3.11115 -0.00967 4 G-2 0.25903 0.23234 0.23829 0.00595 5 * GSP-2 0.25967 0.30010 0.23877 -0.06133 6 RGM-I 0.09692 0.08687 0.11600 0.02913 7 SDC-I 0.68144 0.53512 0.55694 0.02182 8 W-2 2.45391 1.95125 1.89402 -0.05722 9 AGV-2 b 0.64465 0.55748 0.52918 -0.02830 10 DNC-I b 4.11444 3.12082 3.14668 0.02586 II G-2 b 0.25396 0.23234 0.23446 0.00212 12 GSP-2 b 0.30659 0.30010 0.27417 -0.02593 13 RGM-I h 0.09527 0.08687 0.11475 0.027X9 14 W-2 b 2.46030 1.95125 1.89885 -0.05240

•: Deselected samples Rigaku

129

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration Application Heads 14Majorl lJ2

Hlement line AI-KA

It A im ' i Dim 1 C Linear

o-M>-H------1------t1—------0.0 5.0 10.0

Theor. int.

No. Sample Meas. int. Theor. int. Calculated Deviation Corrected 1 AGV-2 16.77969 9.78587 9.56672 -0.21914 BHVO-2 14.06891 7.59040 8.06623 0.47583 3 DNC-I 18.91486 10.35289 10.74860 0.39571 4 G-2 14.76308 8.91248 8.45047 -0.46201 5 * GSP-2 11.17621 8.66307 6.46505 -2.19803 6 RGM-1 14.69840 7.96870 8.41467 0.44598 7 SDC-1 16.88298 9.28064 9.62390 0.34326 8 W-2 14.95504 8.71782 8.55673 -0.16109 0 AGV-2 b 16.33770 9.78587 9.32207 -0.46380 10 DNC-I b 19.15002 10 35289 10.87876 0.52587 11 G-2 b 14.73953 8.91248 8.43744 -047504 12 GSP-2 b 13.12254 8.66307 7.54239 -1.12068 13 RGM-1 b 14.75088 7 96870 8.44372 0.47503 14 W-2 b 15.06281 8.71782 8.61639 -0.10143 15 wise mine a 0.33808 0.29108 0.46586 0.17477 10 wise mine b 0.32358 0.29108 0.45783 0.16674 *

*:Deselected samples Rignku

130

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration Application Beadsl4Majorl-f'2

Element line Si-KA

80.0- 1 p It = Aim2- Blm - C r Lineart straight) A- B 142353e+000 60.0- © C -4.15882e-00l Accuracy 3.59435e(000 % Corr. Cacter 9.82720c-001 #• Inter, std No 40.0- p • Fixed point No Weighting Normal

2 0 .0-

/ o.ir-j------1...... ""I”------0 50 100

Theor. ml.

No. Sample Meas. int. Theor. int. Calculated Deviation Corrected 1 AGV-2 44.00812 63.26708 62.23 114 -1.03594 2 BIIVO-2 39.16875 52.13068 55.34214 3.21146 3 DNC-I 36.16479 49.08720 51.06591 1.97871 4 G-2 50.52403 73.90788 71.50675 -2.40112 5 * GSP-2 38.03191 71.55174 53.72381 -17.82793 6 RGM-1 58.69321 78.94970 83.13586 4.18616 7 SDC-i 52.80505 71.27626 74.75387 3.47761 8 W-2 38.79555 55.02986 54.81088 -0.21898 9 AGV-2 b 43.02770 63.26708 60.83548 -2.43160 10 DNC-I h 3648569 49,08720 51.52271 2.43551 1 1 G-2 b 50.34643 73.90788 71.25393 -2.65394 12 GSP-2 b 44.37244 71.55174 62.74976 -8.80198 13 RGM-I b 58.74958 78.94970 83.2161 1 4.26641 14 W-2 b 39.13228 55.02986 55.29023 0.26037 15 wise mine a 75.78792 108.56763 107.47075 -1.09687 16 wise mine h 75.73248 108 56763 107.39183 -1.17580 *

*:Deselected samples Rigaku

131

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration Application Beadsl 4MajorFP2

Element line P-KA 1.5

It - A ltrr Blm * C Linear(straighl) A- B 6.5455 le-001

1.0 - c- I 20568e-002 A ccuracy 2 .4 7 6 5 1 e-002 Corr. facter 9.95794e-001 Inter, std No /• Fixed point No W eighting Nonnal

0.5-

U . U ' I ■ ■■ ...... 1 ...... 0.0 0.5 1.0

T heor int

No. Sample Meas. int. Theor. int. Calculated Deviation Corrected 1 A G V -2 1.27269 0.83345 0.84510 0.01165 2 BHVO-2 0.76383 0.46387 0.51202 0.04815 3 D N C-I 0.191 10 0.1 1992 0.13714 0.01723 4 G-2 0.33883 0.23654 0.23384 -0.00270 5 * GSP-2 0.59677 0.49134 0.40267 -0.08867 6 * RGM-I 0.13360 0.00000 0.09950 0.09950 7 SDC-I 0.40649 0.27031 0.27813 0.00782 8 W-2 0.32344 0.23980 0.22377 -0.01603 9 A G V -2 h 1.22944 0.83345 0.81679 -0.01666 10 DNC-1 b 0 18922 0.11992 0.13591 0.01600 1! G -2 b 0.33166 0.23654 0.22915 -0.00739 12 G SP-2 h 0.67380 0.49134 0 .45309 -0.03825 13 W-2 b 0.31768 0.23980 0.21999 -0.01980

^Deselected samples Rigcikll

132

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration ______Application ______B e a d s 1 4 M ajo tT ’P2

30.0-

20.0-

10.0-

0.0 10.0 20.00.0 30.0

Theor. int.

No. Sam ple ______M eas. int. ______Theor. int. ______C alculated______Deviation Corrected 1 AGV -2 15.67383 9.93284 10.10879 0.17595 2 B H V O -2 2.98474 1.79865 1.84525 0.04659 3 D N C-I 1.28998 0.79298 0.74156 -0.05142 4 G-2 22.87727 15.17702 14.79991 -0.37711 5 * G SP-2 22.49614 18.35360 14.55170 -3.80190 6 RGM-I 24.23889 14.56626 15.68663 1.12037 7 SDC-I 18.38834 11.34936 11.87657 0.52721 8 W -2 3.31057 2.13573 2.05744 -0.07829 9 A G V -2 b 15.17507 9.93284 9.78398 -0.14886 10 DNC-1 b 1.28516 0.79298 0.73843 -0.05456 11 G -2 b 22.67490 15.17702 14.66811 -0.50891 12 G SP-2 h 25.81451 18.35360 16.71273 -1 64087 13 RGM -1 b 24.15093 14.56626 15.62935 1.06309 14 W-2 b 3.31840 2.13573 2.06253 -0 07320

*: Deselected samples I* *-■

133

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity' Calibration Appl ication Beads 14MajorFP2

Element line Ca-KA

______6 0.0- It - A im 2 + Blm - C w I Linear! straight) * A- B 8.79674e-00l C-- 5.94669e-00l Accuracy 7.78606e-001 8- 4 0 .0 - Corr. facter 9.99306e-001 E / Inter, std No a u Fixed point No s Weighting Normal f t ■ 2 0 .0 -

f 1n u.u i 00 20.0 40.0 60.0

Theor int.

No. Sam ple M eas. int. Theor. int. Calculated Deviation Corrected 1 A G V -2 24.23402 22.28339 21.91271 -0.37068 2 B H V O -2 57.39062 49.41834 51.07971 1.66137 2 DNC-1 55.83528 49.91262 49.71152 -0.20110 4 G _2 8.39344 8.20037 7.97816 -0.22220 5 * GSP-2 7.46574 8.79460 7.16209 -1.63251 6 RGM-1 5.65720 4.82558 5.57116 0.74558 7 SDC-1 6.83647 6.02554 6.60853 0.58300 8 W-2 51.69846 46.83151 46.07247 -0.75904 4 AGV-2 b 23.71327 22.28339 2 1.45462 -0.82878 10 D N C -I b 56.34293 49.91262 50.15809 0.24547 1 1 G -2 b 8.36153 8.20037 7.95009 -0.25028 12 G SP-2 b 8.52675 8.79460 8.09543 -0.69917 13 R G M -I b 5.62923 4.82558 5.54655 0.72098 14 W -2 b 51.85067 46.8315! 46.20636 -0.62514

*:Deselected samples Rigciku

134

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration Application Beads 14MajorFP2

Element line Ti-KA

It A ln F 4 Bini C • Linear( straight) A B l.05723e+001 C 2.98804e-001 Accuracy 2,305l8e-00l Corr. Factcr 9.98394e-00l Inter, std No Fixed point No W eighting Norm al • r ...... ' •

U.O-J------1...... 1------— r 0.0 3.0 10.0 15.0 20 0

T h eo r int.

No. Sam ple Meas. int. Theor. int. C alculated Deviation C orrected 1 AGV -2 0.56859 6.52389 6.31016 -0.21373 2 B IIV O -2 1.55354 16.49324 16.72329 0.23005 3 DNC-I 0.26617 2.91283 3.11282 0.19999 4 G-2 0.25438 2.96439 2.98816 0.02377 5 * G SP-2 0.28056 4.12142 3.26500 -0.85642 6 R G M -I 0.14925 1.67545 1.87670 0.20125 7 SD C -I 0.57823 6.46199 6.41206 -0.04993 8 W -2 0.57478 6.43671 6.37555 -0.06116 9 A G V -2 h 0.55297 6.52389 6.14495 -0.37894 10 D N C-I b 0.26987 2.91283 3.15191 0.23909 11 G -2 b 0.25372 2.96439 2.981 18 0 .01679 12 G SP-2 b 0.32573 4 .12142 3.74252 -0.37890 13 R G M -I b 0.15159 1.67545 1.90145 0.22600 14 W -2 b 0.57543 6.43671 6.38243 -0.05428

:Deselected samples HiflClkll

135

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration Application beads 14MajorFP2

Element line Mn-KA

It - Alnt’- Blm < c Linear(straight) A- B* 3.81592e+000 0.60- c- 8. |9334e-003 Accuracy 6.57745e-002 Corr. Pacter 0 97872e-00I Inter, std N o 0 .40- Fixed point N o W eighting N orm al

0 .20-

0.00 0.0 1.0 2.0 3.0 4.0

T heor in!

N o. Sam ple______M eas. int. ______Theor. int. Calculated ______Deviation ______C orrected 1 AGV-2 0.44652 1.74999 1.71209 -0.03790 -i BHVO-2 0.75451 2 77800 2.88734 0.10934 3 DNC-I 0.66700 2.56292 2.55342 -0.00949 4 G-2 0.14144 0 53182 0.54790 0.01608 5 * G SP-2 0.14110 0.73144 0.54664 -0.18481 6 RGM-I 0.17218 0.65038 0.66522 0.01484 7 SDC-1 0.55187 2.06724 2.11408 0.04684 8 W-2 0.73076 2.82559 2.79671 -0.02888 0 AGV-2 b 0 43516 1.74999 1.66874 -0.08125 10 D N C-I b 0.67020 2.56292 2.56562 0.00270 I 1 G-2 b 0.16836 0.53182 0.65063 0.11880 12 G SP-2 b 0.16800 0.73144 0.64928 -0.08217 13 RGM-I b 0.16434 0.65038 0.63530 -0.01507 14 W -2 b 0 .72422 2.82559 2,77175 -0.05384

*:Dcsclected samples Rigaku

136

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IT Sensitivity Calibration Application Beadsl4MajorFP2

100- Flement line Fe-KA

It - A lttr B lm < C * I,ineart straight) A- 2.90698e+000 • 5.83332e+000 .. A ccuracy 5.772056+000 C orr. 1'actcr 9.96920e-001 50- •/ Inter, std No * Fixed point N o W eighting Norm al

I •

u ...... Ns. ! 1 0 100 200 300

Theor. ini

No. Sam ple Meas. int. Theor. int. Calculated Deviation Corrected 1 A G V -2 43.19312 135.74380 131.39500 -4.34880 2 B H V O -2 79.82857 235.19398 237.89363 2.69965 3 D N C -l 65.03639 195.86906 194.89302 -0.97605 4 G-2 16.93388 54.92894 55.05982 0.13088 5 * G SP-2 25.36614 100.27563 79.57225 -20.70339 6 RGM-I 12.88240 38.99250 43.28223 4.28972 7 SDC-1 47.67165 132.79227 144.41399 11.62172 8 W-2 68.34760 204.80894 204.51863 -0.29031 9 A G V -2 b 42.35430 135.74380 128.95654 -6.78726 10 DNC-1 b 65.51297 195.86906 196.27842 0.40936 1 1 G-2 b 17.14945 54.92894 55.68647 0.75753 12 G SP-2 b 28.701 11 100.27563 89.26695 -11.00869 13 RGM-I b 12.84268 38.99250 43.16677 4.17426 14 W -2 b 68.21630 204.80894 204.13694 -0.67200

*:Deselected samples Rigaku

137

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration Application Beads l4MajorFP2

Element line Rb-KA

1.5- It - A im 2 - B lm + C l.inearf straight) 9 A- B- 2.l8249e*000 C - -1 38845e-002 1.0- Accuracv l.03332e-00l Corr. facter 9.96004e-00l © Inter, std No Fixed point No Weighting Normal

0.5 ' Qt / 0.04 0.0 1.0 2.0 .0

Theor. int.

Overlap correction El. line M ethod Coefficient Br-K A Meas. int. t-l.64588e-O03)

No. Sample Meas. int. Theor. int. Calculated Deviation C orrected 1 AGV-2 0.41816 0.87777 0.87061 -0.00716 0.40527 2 BHVO-2 0.05505 0.1 1334 0.08092 -0.03243 0.04344 3 DNC-i 0.03203 0.05970 0.02904 -0.03066 0.01967 4 G-2 1.11260 2.36057 2.35570 -0.00487 1.08573 5 « G SP-2 1.29639 3.26679 2.75086 -0.51593 1.26678 6 RGM-1 1.10226 2.13838 2.32348 0.18510 1.07096 7 SDC-I 0.79767 1.7021 1 1.71586 0.01376 0.79256 8 W -2 0.11627 0.24778 0.20920 -0.03858 0.10222 9 AGV-2 b 0.41481 0.87777 0.86488 -0.01289 0.40264 10 DNC-I b 0.06032 0.05970 0.07239 0.01269 0.03953 1 1 G -2 b 1.08528 2.36057 2.30301 -0.05756 1.06158 12 GSP-2 b 1.422X3 3.26679 3.04766 -0.21913 1.40277 13 RGM-1 b 1.09295 2.13838 2.30283 0.16445 1.06150 *: Deselected samples R i o a k u

No. Sam ple Meas. int. Theor. int. C alculated Deviation Corrected 14 W -2 b 0.16046 0.24778 0.27505 0.02727 0.13239

138

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration Application Beads 14Ma jorFP2

6.0 Element line Sr-KA

It A im - Blm - C Linear! straight) A- B 2.12618e+000

C'- 1.64343e-001 A ccuracy 1,79792e-OOI Corr. lacier 9.98341 e-0() I Inter, std No Fixed point No W eighting Normal

2.0-

OO 0.0 5.0 10.0

Theor. int.

Overlap correction El. line Method Coefficient Rb-K; M eas. int. (l.944l2e-002)

No. Sam ple M eas int. Theor. int. Calculated Deviation C orrected 1 AGV-2 4.45209 9.48026 9.64703 0.16677 4.45997 B H V O -2 2.39866 4.92238 5.26613 0.34375 2.39951 3 DNC-I 0.83356 1.92506 1.93745 0.01239 0.83394 4 G -2 3.30349 7.42961 7.23303 -0.19658 3.32460 5 * G SP-2 1.25284 3.58051 2.88046 -0.70005 1.27746 6 RGM-I 0.71870 1.75955 1.73670 -0.02285 0 73952 7 SDC-I 1.20043 2.69585 2.74944 0.05359 1.21584 8 W-2 1.13581 2.49653 2.58350 0.08698 1.13780 9 A G V -2 b 4 38607 9 48026 9.50656 0.02630 4.39390 10 D N C -I b 0.85583 1.92506 1.98563 0.06057 0.85660 11 G -2 b 3.27132 7.42961 7.16364 -0.26597 3.29196 12 G SP-2 b 1.45179 3.58051 3.30910 -0.27141 1.47907 13 R G M -I b 0.69617 1.75955 1.68841 -0.07114 0.71681 *:Deselectcd samples Rigaku

No. Sam ple ______M eas._int. ______T heor. int. ______C alculated______D eviation ______C orrected 14 W-2 b 1.13082 2.49653 2.57414 0.07762 1.13339

139

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Empirical Calibration Application ______Rock' I enTraceEmp

Component Y

. \ Element line Y -KA ...... X = At3- Bl2- Cl - D ...... 0 O a Linear( straight)

O ♦ V A-

V B- O ...... C l.08814e+003

o o D- 2.97081 e+000

C D O . Accuracy 6.64637c-00l V ...... r Corr. factor 9.98047e-00l * Inter, std Rh-KAC / Fixed point No o > Weighting Normal o *

¥ ¥ / /

0 50 100

Std. valuefppm)

Matrix correction ______W»X( I + K rsum(AFj)»sum(QFjFk)rsutn(RFj/( I - IOQ/Wi)))+sum(BFj)-suni(DFjFk)+C Type 1st Comp. 2nd Comp. Coefficient Type st Comp. 2nd Comp. Coefficient B Rb (-2.39625e-00l) B Th (-1.22499e-002)

N o. Sample Int. ratio Std. value Calculated Deviation A pparent 1 * NPk Rind 0.03651 (8.4) 8.10468 -0.29532 42.99576 2 * BF Intrusive 0.06676 (30.5) 30.83950 0.33950 75.27601 3 * K insm an 0.05338 (22.4) 20.39030 -2.00970 63.06686 4 * BF Schist 0.07406 (37.3) 30.18795 -7.11205 90.67555 5 * Bg N Road 0.07957 (55.3) 4 4 .2 8 8 4 1 -11.01159 100.56016 6 * BF Cneiss 0.05832 (25.7) 23.18923 -2.51077 68.94627 7 * SQ D Ft 1 0.05749 (33.4) 33.10314 -0.29686 65.81964 8 * SQD Sargent 0.05320 (33.2) 32.62392 -0.57608 61.43700 9 AG V -2 0.03135 20 20.57174 0.57174 36.51299 10 BHVO-2 0.02392 26 26.63897 0.63897 28.36302 1 1 DNC-I 0.01470 18 17.88352 -0 11648 19.07831 12 G-2 0.04496 1 1 10.84893 -0.15107 52.04246 13 GSP-2 0.07816 28 28.02241 0.02241 87.99431 *: Deselected samples Rigaku

No. Sam ple Int. ratio Std. value Calculated Deviation Apparent 14 * RG M -I 0.05369 (25) 25.26994 0.26994 61.12747 15 W -2 0.02197 23 21.82062 -1.17938 28.06152 16 Wise Mine Quartz -0.00253 (0) 0.21381 0.21381 0.00000 140

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration

Application Beadsl4Maiort-P2 6 . 0- Element line Zr-KA

It =- A im 2 B lm ' C I.inearf straight) A-

4.CH B-- 1.88673e~000 C- 9.l6223e-OOI Accuracv 2.80097e-001 o / Corr. facter 9.94919e-001 Inter, std N o ¥ Fixed point N o W eighting N orm al 2.1H ...... P v N \ o

0.0 0.0 5.0 10.0

Theor. int.

Overlap correction El. line M ethod Coefficient Sr-KA M eas. int. (-I.47l23e-O0l)

No. Sample M eas. int. Theor. int. C alculated D eviation I A G V -2 C orrected! 2.15555 4.16181 3.74508 -0.41673 B H V O -2 1.49935 1.36228 2.74132 2.82035 0.07903 D N C -I 1.00922 0.05916 0.63802 0.79628 0.15826 -0.06357 4 G -2 3.24177 6.0217 5 6.10957 0.08783 2.75257 5 G SP-2 4.33766 10.26965 8.74544 -1.52422 4.14963 6 R G M -I 2.0 0 9 7 0 4.41136 4.50253 0.09117 7 SD C -I 1.90081 2.89737 5.45721 6.04524 0.58804 2.71848 8 W -2 0.511 17 1.659 01 1.56476 -0.09425 0.34374 0 A G V -2 b 2.13371 4.16181 3.72221 -0 4 3 9 6 0 10 D N C -I b 1.48723 0.02057 0.63802 0.71714 0.07912 -0.10552 I I G -2 b 3.24979 6.02175 6.13377 0 1 1203 12 G SP-2 b 2.76540 5.04294 10.26965 10.02020 -0.24945 4.82528 R G M -I b 2.00207 4 .4 1 136 4.49444 0.08307 *■. Deselected samples 1.89652 t i g a k u No. Sam ple M eas. int. Theor. int. C alculated D eviation 14 W -2 b Corrected! 0.51890 1.65901 1.58049 -0.07852 0.35207

141

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission K«frv '' *-

Empirical Calibration Application ______RockTenT raceEmp

Com ponent Nb Element line N b-K A

X - A lC B I-- Cl - D Lineart straight) A* B C= 9 . 15797c-0 0 2 D 2.93679e+(>00 o K A ccuracy 7.2750 le-001 e Corr. factor 9.96907e-001 Inter std Rh-KAC 0 .020- Fixed point N o W eighting N orm al

0 000- 0.0 20.0

Std. valuetppm)

Matrix correction W=X( l-K-sum ( AFi)-suni(Oiil:k)-sum(KFi4 I - 100'W'i)ll-sum(BFi)-stini(PF[Fk)-C Type I st Comp. 2nd Comp. Coefficient [ Type 1st Comp. 2nd Com p. Coefficient B <-2.21986e-002)

No. Sam ple Int. ratio Std. value Calculated Deviation ----I J B I t i p t f c ------A pparent 1 * NPk Rind 0.00399 (5.7) 6.40214 0.70214 5.88647 * BF Intrusive 0.03034 (26.9) 30.04497 3.14497 27.57706

"x * K insm an 0.01671 (14.6) 17.74656 3.14656 15.09725 4 * BF Schist 0.02039 (19.2) 20.78377 1.58377 20 02801 5 * Bg N Road 0.01792 (17.6) 18.11624 0.51624 18.82758 6 * BF Gneiss 0.01732 (12.9) 18.23236 5.33236 13.47050 7 * SQ D FH 0.03524 (30.6) 34.46451 3.86451 31.34143 8 * SQD Sargent 0.03898 (33.6) 37.89424 4.29424 34.33699 9 AQ.V-2 0.01388 15 15.20757 0.20757 15.44397 10 B 11 VO -2 0.01764 (18) 18.51788 0.51788 18.57716 11 DNC-I -0.00076 (3) 1.83797 -1.16203 3.39957 12 G-2 0 0 1 0 5 2 (12) 12.33038 0.33038 12.24418 13 G SP-2 0.02682 27 26.87594 -0.12406 27.62156 ♦D eselected s.-unnC-s

142

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FP Sensitivity Calibration Application B eads H M ajo rF P 2

0.060- Klement line Ba-LA

A im - * Blm I. ineart straight) X A - B I.I3360e+001 0 .040- C 6.69207e-003 A ccuracy 2.40540e-002 Corr. facter 9.9251 le-001 Inter sld No Fixed poinl No 9/ W eighting N orm al 0 . 020-...... c/ 4

0.000 0.00 0.20 0.40 0.60

T heor. int.

Overlap correction El. line Method Coefficient Ti-KA Meas. int. (-4.54343e-003)

N o Sam ple Meas. int. Theor. int. Calculated D eviation C orrected 1 A G V -2 0.03238 0.33121 0.34452 0.01331 0.02980 2 BH VO -2 0.00636 0.03679 -0.00127 -0.03806 -0.00070 3 DNC-I 0.00628 0.03340 0.06413 0.03073 0.00507 4 G-2 0.04886 0.54466 0.54745 0.00279 0.04770 5 * G SP-2 0.02867 0.38891 0.31724 -0.07167 0.02740 6 RGM-I 0.02384 0.23630 0.26925 0.03296 0.02316 7 SDC-I 0.01882 0.18869 0.19023 0.00154 0.01619 8 W -2 0.00673 0.04787 0.05339 0.00552 0.00412 9 A G V -2 b 0.02932 0.33121 0.31058 -0.02063 0.02681 10 D N C -I b 0.00324 0.03340 0.02953 -0.00387 0.00201 1 1 G -2 b 0.04726 0.54466 0.52941 -0.01525 0.04611 12 GSP-2 b 0.03250 0.38891 0.35835 -0.03056 0.03102 13 R G M -I b 0.02369 0.23630 0.26742 0.03112 0.02300 *:Deselected samples Rigakll

No. Sam ple M eas. int. Theor. int. Calculated D eviation C orrected 14 W-2 b 0.00540 0.04787 0.03828 -0.00959 0.00279

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Empirical Calibration Application ______Rock I'enT racchnip

/ Component La / Element line La-LA / / ,/ X = A!3-* Bl2- Cl < D /• / Linear! straight) / / A- / B- / C I.06l87e*003 / ? D- -7.71100e-00l / Accuracy 7.65887e' 000 C o i t . factor 9,92727e-00l « Inter, std No > Fixed point No Weighting Normal •/ _ ......

• *

\------0 100 200 300

Std. value(ppm)

No. Sam ple Intensity Std. value Calculated D eviation A pparent 1 * N Pk Rind 0.01493 (17) 15.08262 -1.91738 2 * BF Intrusive 0.07637 (90.5) 80.32393 -10.17607 3 * Kinsman 0.04070 (43.9) 42.44702 -1.45298 4 * BF Schist 0.02874 (52.2) 29.74705 -22.45295 5 * Bg N Road 0.05990 62.83493 9.53493 . <53-3 > 6 * BF Gneiss 0.03350 (30.4) 34.80155 4.40155 7 * SQ D FH 0.12161 (126) 128.36294 2.36294 8 * SQD Sargent 0.09091 (89.4) 95.76353 6.36353 9 AGV-2 0.03797 38 39.54811 1.54811 10 B H VO -2 0.00772 15 7.42654 -7.57346 II DNC-I 0.00152 3.6 0.84294 -2.75706 12 G-2 0.09563 89 100.77555 1 1.77555 13 GSP-2 0.16166 180 170.89085 -9.10915 14 RGM-I 0.02284 24 23.48202 -0.51798 15 SDC-I 0.04765 42 49.82702 7.82702 16 Wi2 0.00902 10 8.80697 -1.19303

*:Deselected samples Rigakll

144

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Empirical Calibration Application RockTenT raceEmp

0 40- C om ponent Ce

/ Element line Ce-LB I

/ / / X = Al-’* BI 2 « Cl + D Linear(straight) ...... J it'.,...... 0.3 0 - / A~ B" / C- 1 32S46C-003 / D- -I 75260e*000 / / A ccnracv 9,92723e+000 5, 0 .20- ...... r \ ...... V. Corr. iaclor 9,96157e-001 Inter, std No .V-N

. . . . Fixed point No

...... Weighting Normal / 0 . 10- ...... o x ......

u.wvr-]------I r 0 200 400

Std value(ppm)

No. Sample Intensitx Std. value Calculated Deviation Apparent 1 N Pk Rind 0.01934 (34.3) 23.93976 -10.36024 -> BF Intrusive 0.13992 (193) 184.12512 -8.87488 3 Kinsman 0.06346 (92.7) 82.55129 -10.14871 4 * BF Schist 0.04753 (107) 61.38897 -45.61103 5 Bg N Road 0.10340 (115) 135.60987 20.60987 6 BF Gneiss 0.05561 (64.4) 72.12290 7.72290 7 SQD FH 0.20153 (268) 265.97136 -2.02864 8 SQD Sargent 0.15219 (199) 200.42529 1.42529 9 A G V -2 0.05197 68 67.28732 -0.71268 10 B H V O -2 0.02338 38 29.30673 -8.69327 1 1 G-2 0 13208 160 173.71002 13.71002 12 G SP-2 0.30401 410 402.11165 -7.88835 13 RGM-I 0 03467 47 44.30501 -2.69499 14 SD C -I 0.07759 93 101.32239 8.32239 15 W -2 0.01834 23 22.61131 -0.38869

' Deselected samples Rigpcil«.«j

145

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Empirical Calibration Application Rock I enTraccEmp

/ Component Nd Element line Nd-LA

/ / X - A ll- Bl: - Cl + D / / l.inear( straight) / ...... *ft A- / FC / C= 6.84340e^<)02 / / l>~ -2.66204eH)00 / / ...... / ...... Accuracy 5.19359e ‘ 000 Corr. factor 9,97l02e-001 / / Inter std No ...... Fixed point No m Weighting Normal V

/ 0 100 200 300

Std. value!ppm}

No. Sam ple Intensity Std value Calculated Deviation A pparent 1 * N Pk R ind 0.01463 (13.4) 7.34986 -6.05014 2 * BF Intrusive 0.1 1952 (81.4) 79.13029 -2.26971 3 * Kinsman 0.06100 (39.9) 39.08271 -0.81729 4 * BF Schist 0.03964 (45.8) 24.46520 -21.33480 5 * Bg N Road 0.09395 (52) 61.63172 9.63172 6 * B F G neiss 0.04788 (26.9) 30.10417 3.20417 7 * SQ D Fl l 0.15979 (1 12) 106.68867 -5.31 133 8 * SQD Sargent 0.14347 (95.7) 95.52024 -0 .17976 q A G V -2 0.04780 30 30.04942 0.04942 10 BHVO-2 0.03282 25.0 19.79800 -5.20200 11 DNC-1 0.00494 5.2 0 71860 -4.48140 12 G-2 0.08799 55 57.55305 2.55305 13 GSP-2 0.29219 200 197.29531 -2.70469 14 RGM-I 0.03693 19 22.61064 3.61064 15 SDC-1 0.07542 40 48.95089 8.95089 16 W -2 0.01883 13 10.22409 -2.77591 *

*:Deselected samples ffXifsuku

146

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Empirical Calibration ______Application ______Rock 1 cn'l raccEmp

0.020 Component Pb Element line Ph-LBl

/• X - Al’- Bl: - Cl + D Linear( straight) 0.015- A- B- C- l.98873e-003 D - 1.3 1887c- 001 Accuracj 6.88608e-001 0 .010- Corr. factor 9.98.392e-00l Inter, std Rh-KAC Fixed point No Weighting Normal

0.005-

0.000- 0.0 20.0 4 0.0 60.0

Std. value(ppm)

No. Sam ple ______Int. ratio ______Std. value ______C alculated______D eviation ______Apparent 1 * N Pk Rind 0.01685 (49) 46.69603 -2.30397 -) * BF Intrusive 0.00368 (21) 20.50827 -0 49173 2 * Kinsm an 0.00867 (36) 30.43674 -5.56326 4 * BF Schist -0.00027 (25) 12.65617 -12.34383 5 * B g N Road 0.00309 (23) 19.33555 -3.66445 6 * BF G neiss 0.00351 (23) 20.17418 -2 82582 7 * SQD FH 0.01074 (28) 34.55110 6 55110 8 * SQD Sargent 0.00882 (32) 30.72775 -1.27225 9 AGV-2 -0.00017 13 12.85248 -0.14752 10 * DNC-I -0.00370 (6.3) 5.82659 -0 4 7 3 4 1 1 1 G-2 0.00897 30 31.02153 1,02153 12 GSP-2 0.01424 42 41.50908 -0.49092 13 RGM-1 0 .0 0 5 4 1 24 23.95490 -0.04510 14 SDC-1 0.00577 25 24.66201 -0.33799 15 * W-2 -0.00298 (9.3) 7.25300 -2.04700 16 * Wise Mine Quartz -0.00653 (0) 0.19769 0.19769

*:Deselected samples Rigakll

147

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX B LOSS ON IGNITION CALCULATIONS

147

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.5006 4.5005 4.5004 4.5005 4.5005 4.5004 mLiB02 4.5 4.5003 4.5006 4.50054.5001 4.5004 4.5006 mLi2B407 mLiB02 mmv mLi2B407 mmv mmv mLi2B407 mLiB02 0.90070.9001 0.9002 4.50050.9004 4.5004 4.5009 4.5001 4.5006 0.90090.9001 0.9008 4.5009 4.5004 4.5 4.5006 0.90030.90070.9001 4.5004 4.50030.9007 4.5001 4.5005 4.5007 4.5007 0.9005 4.50070.9006 0.9007 0.9007 4.5005 4.5006 0.9005 4.5009 0.2 1.02 0.72 0.98 0.74 0.53

0.011 0.0139 0.0169 1.12 0.0084 0.55 0.0061 0.41 mix vial flux added mad loss loi (%) mad lossmad loi (%) loss loi (%) w/flux = mmv + mass of mmv = mass ofsample into 29.0025 36.5014 0.0104 34.6147 34.519932.561432.7193 0.054234.4672 0.0558 0.0512 3.5 3.36 3.58 34.929334.484632.4249 0.0167 0.0121 0.0077 1.15 0.84 0.52 29.1743 0.0031 34.502132.7325 0.0146 36.509434.6285 0.0078 34.6648 0.0065 0.39 29.0164 34.6257 34.5167 29.1774 32.7362 34.6713

27.5968 33.1488 31.2254 31.2235 32.7409 Cl 30.9552 32.4326 C61 33.0261 34.5741 C79 30.9567 32.6172 C15 33.4911 34.946 P101 35.0604 36.5118 C169 C171 C170 C167 33.0901 34.5184 C234 33.0549 34.4967 C2 90 35.0587 36.5172 C2116 33.1475 34.6346 C2119 33.0249 Cl 229C2 213 33.0859 27.5929 C llO l-a mws =mas ofcup with sample sample (metaseds) om mws mad =mad mass drying after om = original mass of cup sample (migmatites typel) om mws sample (migmatites type2) om mws Appendix B Loss on ignition calculations metasedimentary rocks (fusion beads)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. t" m ‘J^tTt^tTtTf'^TtTtTtTt^TtTt s e

in in

T—

in (Nvo inin »n o 8 $ o o o o d 0 0 0 0 0 0 0 0 0 0 0 0 0

on in 3 ■3 § § $ ! § £ £ E E cm i n (N X ro co ON^-iJi/icNrtgioji/iiSJSyDfN

\ n 3 t in X O CNr- ro co in Ov CN| ai CN ro § m C4 8 ?] B s 8 d 8 r-: vo O e: CNin vo CMOn VOin in vo c4 in

vo VO i-H ro i-Hro NlO^fOOO^^ON

(N UIC (N 8 n CN * (N K O' e

"O co S tO | U U U U -—ID- — 1 g, " 1 E E & QO < E

149

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.5003 4.5001 4.5007 4.5007 4.5007 4.5001 4.5009 4.5004 4.5004 mLiB02 mLiB02 4.5006 4.5007 4.5007 4.5006 4.5005 4.50064.5008 4.5007 4.5008 4.5003 4.5009 4.5005 4.5009 4.5004 4.50064.5007 4.50044.5008 4.5008 4.5002 10.509 10.5081 10.5026 10.5087 mmvw/flux mmv mLi2B407 mmv mLi2B407 0.9001 0.9004 0.9003 0.9005 0.9009 0.90090.9004 4.5001 0.9009 mmv 3.5012 10.5032 3.5032 3.5024 10.5012 1.15 3.5012 1.42 3.5083 0.57 3.5029 10.5074 ! 0 .4 5 (%) 13.68 3.5039 0.33 0.9008 0.06 0.9013 4.5001 4.5002 0.84 0.9005 0.94 0.9009 0.720.48 0.901 0.9005 #DIV/0! 0.9007 #DIV/0! 0.9005 loss loi (%) 0 0 0 0 0 0 0.0149 0.38 0.0227 0.1412 3.56 0.6217 0.0453 0.0564 loss loi (%) loss loi 0.018 0.0017 0.0071 0.47 0.0208 0.0131 0.0136 0.9 0.0205 0.8 0.0119 0.0065 0.3 mad 35.106 0.0176 34.8789 37.4042 31.4711 36.9616 37.0046 36.9449 mad mad 29.2075 0.0054 36.3062 32.4609 30.0603 34.9946 37.5831 34.4742 35.2261 35.5177 35.6152 34.9016 37.1028 31.5164 37.6263 37.0013 3 5 .1 2 3 6 mws 35.5357 om mws om om mws 33.482230.9495 37.4191 27.5875 33.1403 33.0826 33.0169 3 1 .2 1 8 8 27.5943 29.2129 27.5917 30.0811 33.488530.9541 36.3079 32.468 33.4893 35.0082 33.087235.0564 34.4873 33.0221 37.6036 33.145433.0844 35.6271 35.2326 sample kinsman g-2 (a) g-2 (a) w-2 (b) gsp-2 (a) rgm-l(a) gsp-2 (b) gsp-2 (c) gsp-2 (d) rgm-l(b) agv-2 (a) dnc-1 dnc-1 (a) agv-2 (b) bhvo-2 (a) bhvo-2 (b) ballfield metased ballfield gneiss 2 ballfiled intrusive spaulding spaulding sargent wise mine qtz (a) wise mineqtz (b) sample (standards) sample (standards) granite n. n. granite peak village Appendix B continued weathered rind n. peak village

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10.509 10.504 10.5089 10.5085 10.5484 10.5123 10.5485 mmv mmvw/flux mmv mmvw/flux 2.2073 10.5096 3.0374 9.0245 3.51013.5118 10.5149 10.516 3.50883.5059 10.5475 (%) 1.3 3.5007 10.5049 3.8 3.5068 0.4 0.3 0.850.17 3.5041 3.5079 10.503 0.130.13 3.503 3.508 0.890.87 3.5038 3.5..?0.11 0.22 10.5231 10.6065 3.5141 10.5036 0.18 0.27 0.190.19 0.190.31 3.516 3.5077 3.4338 10.5039 11.96 3.5004 10.5072 loi (%) loi 0.0068 0.4734 0.0106 0.0345 0.0071 0.0106 0.0076 0.0074 0.0106 0.0129 36.982536.9738 0.0341 0.0511 34.752837.1233 0.1501 0.0358 34.000137.0335 0.0033 0.009 37.4878 37.012131.0323 37.5951 0.0073 0.0182 mws mad loss mws mad loss 37.102 37.0949 38.968637.0166 37.0249 0.036537.4111 37.4043 0.92 37.441241.0892 36.9678 41.0786 37.013434.9029 37.1591 37.008439.0022 0.005 38.9677 37.0425 34.0034 35.218237.4954 39.0078 35.2076 37.0194 31.0429 39.0004 37.6133 35.2475 35.2346

om om 33.084 39.0051 33.0162 33.0803 33.4825 27.5893 33.4828 33.0196 33.0844 30.9519 33.1429 35.0547 33.0199 33.1427 31.2206 33.4852 35.0541 30.9526

GSP-2 sample sample NY-Qtz A12Si05 C2 213 1C2 213 2 30.9522 C2 213 7 C2 213 9 33.0884 C2 213 3 C2 213 4 C2 213 5 C2 213 6 C2 213 8 Bg n. road C2 213 10 sqd firehousesqd wise mine qtz ballfield gneiss 1 ballfield metased mgc 3 nations rent Appendix B continued Bethlehem gneiss n. road

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.