STRATIGRAPHY, STRUCTURE, AND METAMORPHISM IN THE MONADNOCK QUADRANGLE, NEW HAMPSHIRE

BY PETER J. THOMPSON

CONTRIBUTION NO. 58

DEPARTMENT OF GEOLOGY 8c GEOGRAPHY

UNIVERSITY OF MASSACHUSETTS

AMHERSTt MASSACHUSETTS

STRATIGRAPHY, STRUCTURE, AND METAMORPHISM

IN THE MONADNOCK QUADRANGLE,

NEW HAMPSHIRE

by

Peter J. Thompson

Contribution No. 58

Department of Geology and Geography

University of Massachusetts

Amherst, Massachusetts

August, 1985

-~~=======-- ------

© Peter James Thompson

All Rights Reserved

This research was supported in part by

grants from the National Science Foundation:

Grants EAR-79-15246

EAR-84-10370

(to Peter Robinson)

Grant EAR-81-16197

(to Peter Robinson and J.M. Rhodes)

ii

iii

TABLE OF CONTENTS

ABSTRACT. . . 1

INTRODUCTION. 3

Location and Physiography. 3 Purpose. 5 Previous Work. . 6 Methods of Study 9 Acknowledgments. 10

STRATIGRAPHY. . . . . 11

SWANZEY GNEISS, AMMONOOSUC VOLCANICS AND PARTRIDGE FORMATION 11 CLOUGH QUARTZITE AND FITCH FORMATION 14 RANGELEY FORMATION . . . . . . . . . 16

Description and Distribution of Rock Types 16 Sulfidic schist and gneiss. 17 Conglomerates . . . . . . . . . 20 Calc-silicate granulite . . . . 24 Gray schist and gneiss, granulite and augen schist. 24

Thickness. . . . . . 25 Age and Correlation. 28 Derivation . . . . . 29

PERRY MOUNTAIN FORMATION 30 Description and Distribution of Rock Types 30 Thickness. . . . . . 34 Age and Correlation. 34 Derivation . . . . . 35

FRANCESTOWN FORMATION. 35 Description and Distribution of Rock Types Thickness ..... . Age and Correlation. Derivation .....

WARNER FORMATION . . . Description and Distribution of Rock Types Thickness ..... . Age and Correlation. Derivation .....

LITTLETON FORMATION .. Description and Distribution of Rock Types Thickness ..... . Age and Correlation. Derivation

INTRUSIVE ROCKS

INTRODUCTION KINSMAN GRANITE. SPAULDING TONALITE AND RELATED ROCKS

35 38 38 39 40 40 44 44 45 45 45 53 53 55

56

56 58 62

FITZWILLIAM GRANITE. MICRODIORITE DIKES .

Mineralogy . . . . . Field Descriptions Contact Relations.

PEGMATITE. . . . TOURMALINE VEINS DIABASE DIKE . .

STRUCTURAL GEOLOGY.

INTRODUCTION . . DESCRIPTION OF MINOR STRUCTURAL FEATURES

Planar Features. Bedding . Foliation .. Mylonitic foliation Crenulation cleavage. Joints. . . . . . .

Linear Features ..... . Mineral lineations. . Intersection and crenulation lineations Minor folds . . . . . . . .

GEOMETRICAL ANALYSIS OF STRUCTURAL DATA. Structural Data in Subareas. . Construction of Cross Sections

PHASES ONE AND TWO: FOLD NAPPES AND THRUST NAPPES Introduction ... Tectonic Levels .. Monadnock Syncline Folds on Mt. Monadnock Other Map-scale Nappe-stage Folds.

Howe Reservoir syncline . Dublin Pond syncline ..... Gilson Pond anticline and Meade Brook syncline. Nappe-stage folds in the Derby Hill window. .

Nappe-stage Thrust Faults .......... . Brennan Hill fault ........... . Chesham Pond fault and Derby Hill window. Thorndike Pond fault zone . .

PHASES THREE AND FOUR: BACKFOLDING ... Introduction . . . . . . . . . . . . . Intermediate Stage Folds, Mt. Monadnock.

Asymmetric folds ... Boudinage . . . . . . . . . . . . . Thoreau Bog syncline. . . . . . . .

Intermediate Stage Folds, Poole Reservoir Area Intermediate Stage Folds and Mylonitization, Gilson Pond Intermediate Stage Folds, Southeast of Thorndike Pond. Beech Hill Anticline . . . . . . . . Intermediate Stage Folds in the Troy Area ....

iv

66 69 69 69 73 73

74 74

75

75 75 75 75 98 98 98 98 99 99 99 99

100 100 100 100 100 100 102 103 108 108 108 108 111 111 111 113 115 116 116 116 116 118 118 121

Area. 125 125 126 127

Intermediate or Dome Stage Folds in the Derby Hill Area. Summary of Backfolding Episode

PHASE FIVE: DOMING. Introduction . . . Marlboro Syncline. Rindge Area ... Late Open Folds, Mt. Monadnock Cobb Hill. . Discussion . . . . . . . .

LATE PALEOZOIC INTRUSIONS AND DEFORMATION. MESOZOIC FAULTING. . . . . . . . . . . . . SUMMARY AND REGIONAL STRUCTURAL IMPLICATIONS .

METAMORPHISM. .

INTRODUCTION PELITIC ROCKS.

Aluminum Silicate Polymorphs Metamorphic Zones.

Zone II . Zone III. Zone IV Zone V .. Zone VI .

Assemblages in Sulfidic Schists. Evidence for a Retrograde Episode. Garnet Zoning in Zone III.

Assemblage (1) .... . Assemblage (2) .... . Coticule garnet zoning.

Temperature Estimates. MK-432. . . . MK-629 ....

Pressure Estimates CALC-SILICATE ROCKS.

Mineralogy and Chemography of MND-8-74 Disequilibrium Textures and Reactions.

CORRELATION OF METAMORPHISM AND DEFORMATION. Age of Prograde Metamorphism . . . . . Age of Retrograde Metamorphism and "Permian Disturbance"

CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE RESEARCH

REFERENCES CITED.

APPENDIX 1.

APPENDIX 2.

v

128 128 130 130 131 131 131 132 132 134

• 134 136

138

138 138 138 140 140 144 145 146

. 147 14 7 148 150 150 157 160 161 161 164 165 166 166 170 174 174 176

177

179

189

190

vi

TABLES

la. Estimated modes of the Rangeley Formation: sulfidic schist

lb. lc.

ld.

2. 3. 4. Sa. Sb. 6a. 6b. 7. 8. 9. 10. 11. 12. 13. 14.

15. 16.

Figure 1. 2.

3. 4.

5.

6. 7. 8. 9. 10.

11.

12. 13. 14.

and gneiss . . • • . . . . . . . . . . . . . . • Conglomerate localities in the Rangeley Formation. . Estimated modes of the Rangeley Formation: conglomerates

and calc-silicate pods . . . . . • • . . . . Estimated modes of the Rangeley Formation: gray schist and

gneiss . . . . . . . . . . . . . • · · Important exposures of the Perry Mountain Formation. Estimated modes of the Perry Mountain Formation. Estimated modes of the Francestown Formation . . .

18 21

22

26 31 32 36 41 42 46 48 60 64 67 70 76

Estimated modes of the lower part of the Warner Formation. Estimated modes of the upper part of the Warner Formation. Estimated modes of the lower part of the Littleton Formation Estimated modes of the upper part of the Littleton Formation Estimated modes of the Kinsman Granite . . . . . . . . . Estimated modes of the Spaulding Tonalite and related rocks. Estimated modes of the Fitzwilliam Granite . . . . . . Estimated modes of microdiorite dikes. . . . ..•... Summary of structural history in the Monadnock quadrangle. Electron microprobe analyses of garnet from schists. . . . Estimated T from garnet-biotite and cordierite-garnet pairs. Electon microprobe analyses from the core of calc-silicate

• 152 162

granulite pod MND-8-74 . ... List of station codes and numbers.

168 189

List of mineral abbreviations. • 189

ILLUSTRATIONS

Index map of study area and adjacent parts of New England. Stratigraphic correlation diagram across the Silurian

tectonic hinge . . . . . . . . . . . . . . . • . Stratigraphic column, central Monadnock quadrangle . Correlation chart for Silurian-Devonian rocks of New Hampshire,

showing fossil control . . . . . . . . . . . . . . . • . . Outcrop sketches of penecontemporaneous sedimentary structures,

Mt. Monadnock ...•............... Measured sections of the "Seven Quartzites", Littleton Fm. Streckeisen plot of representative intrusive rocks . Outcrop sketch of Spaulding-Kinsman contact, MK-482 .. Map of mafic dikes on Mt. Monadnock. . . . . . ... Summary maps of structural features, Monadnock quadrangle: lOa. Axial traces of major folds and faults ...... . lOb. Subareas, average foliation and sillimanite lineations Equal area diagrams of structural data for the

Monadnock quadrangle . . . . . . . Equal area diagrams of structural data for each subarea .• Simplified geologic map, showing names of major structures Outcrop sketch of nappe-stage folds in Seven Quartzites ..

Page 4

7 12

15

52 54 57 63 72

77 78

79 82

101 104

Figure 15.

16. 17.

18.

ILLUSTRATIONS (CONT'D)

Outcrop sketch of "Billings fold", nappe-stage syncline near Mt. Monadnock summit ........ .

Geologic map of summit area of Mt. Monadnock . Proposed schematic cross section after nappe-stage

deformation, showing proposed thrust faults .. Geologic map of area south of Hurricane Hill, showing

alternate fault interpretation rather than folds Geologic map of south part of Derby Hill window. Geologic map of north part of Derby Hill window. Outcrop sketch of asymmetric backfold ..... . Outcrop sketches of boudinaged schist beds . . . Outcrop sketch of NE-plunging backfold (MK-799).

vii

Page

105 107

109

19. 20. 21. 22. 23. 24. 25. 26.

Geologic map of Poole Reservoir area, Monadnock State Park Outcrop sketch of disharmonic folds in Perry Mountain Fm. Interference pattern of south-plunging backfolds and

110 112 114 117 119 120 122 123

27. 28.

29. 30. 31. 32.

33. 34. 35. 36.

37.

38. 39. 40.

41. 42.

Plate

dome-stage folds south of Troy . . . . . . . Proposed schematic backfolding sequence. . . . . Comparison of equal area diagrams of sillimanite and mica

lineations from two structural domains . . . Equal area diagram of quartz veins, west of Rindge Metamorphic map of Monadnock quadrangle ..... . P-T trajectories ................. . Chemographic representation of mineral assemblages,

Zones II-VI, and the AKFM tetrahedron ..... . Contoured chemical map of zoned garnets, MK-432 .. . Zoning trend for MK-432 garnets on Fe-Mn-Mg ternary plot Contoured chemical map of zoned garnet, MK-629 .... Zoning trends for MK-629 and MK-6 garnets of Fe-Mn-Mg

ternary diagrams . . . . . . . . . . . . . . . . . Bustamite and clinopyroxene compositions in the system

CaSi03-FeSi03-MnSi03-MgSi03 ........... . Mineral compositions, MND-8-74, on ternary diagrams .. Disequilibrium textures and proposed reactions, MND-8-74 T-Xc02 equilibrium curves for calc-silicate assemblages.

Townships and station codes, Monadnock quadrangle. Fence diagram, Mt. Monadnock summit ....... .

1. Bedrock Geologic Map, Monadnock Quadrangle, in four quadrants: SE, SW, NE, NW ........ .

2. Planar Structural Features, Monadnock Quadrangle. 3. Linear Structural Features, Monadnock Quadrangle. 4. Geologic Map and Key to Plates ......... . 5. Geologic Cross Sections, A-A', B-B', C-C', D-D', E-E'

127 129

133 135 139 141

142 151 156 158

159

167

171 172 173

190 191

IN POCKET IN POCKET IN POCKET IN POCKET IN POCKET

1

ABSTRACT

The layered rocks of the Monadnock quadrangle, New Hampshire, have been mapped based on a stratigraphic sequence which correlates well with the sequence recently described by Lyons and Hatch in central New Hampshire and earlier by Moench in western Maine. Three distinctive Silurian units separate the Silurian Rangeley Formation from the Devonian Littleton Formation. In order from oldest to youngest these are: the Perry Mountain Formation, thinly bedded schist and white quartzite; the Francestown Formation, mainly sulfidic calc-silicate granulite with subordinate sulfidic schist; and the Warner Formation, consisting of a lower clean calc-silicate granulite and an upper feld­spathic granulite. The Rangeley Formation includes sulfidic to gray­weathering gritty schist with calc-silicate pods, granulite beds, quartz­pebble conglomerate lenses near the top, and granulite-matrix conglom­erate horizons near the lowest exposed part. The Littleton Formation consists of gray-weathering schists and quartzites, with the proportion of quartzite increasing upwards. A group of seven distinctively spaced quartzite beds is folded into isoclinal folds, and it is this folded sequence that forms the resistant summit of Mt. Monadnock.

The Silurian is represented by a much thinner sequence of rocks along the west edge of the quadrangle, where the Clough Quartzite and local Fitch Formation, probably correlative with the Silurian units described above, overlie Ordovician rocks of the Keene dome. The quad­rangle thus straddles a Silurian "tectonic hinge", which may have behaved as a zone of weakness during Acadian deformation.

Intrusive rocks include the pre- or syn-tectonic (Devonian) Kinsman Granite, syn-tectonic (Devonian) Spaulding Tonalite and related rocks, and post-tectonic (?Mississippian) Fitzwilliam Granite. Granite dikes and microdiorite dikes probably coeval with the Fitzwilliam cut all generations of folds on Mt. Monadnock.

Five phases of Acadian deformation have affected the rocks of the quadrangle. Fold nappes and then thrust faults transported rocks of the ~errimack trough (Rangeley through Littleton Formations) westward across the tectonic hinge onto the thinner Bronson Hill sequence. The rocks were then folded back toward the east in two complicated phases which included mylonitization along the short limb of a major backfold, the Beech Hill anticline. The final phase produced folds related to the rise of the Keene dome at the west edge of the quadrangle. There is evidence for several periods of movement along the four-kilometer-wide Thorndike Pond fault zone, including nappe-stage ductile thrusting, east­directed shear during the backfolding, late Paleozoic shear in the Fitzwilliam Granite, and Mesozoic normal faulting and silicification. A single occurrence of a presumably Mesozoic diabase dike was found in a float block of Kinsman Granite.

2

The dominant foliation lies parallel to axial planes of nappe-stage folds, and the peak metamorphism probably occurred during the early backfolding. Assemblages in pelitic schists range from Zone II in the west to Zone VI near the Kinsman Granite in the northeast, but Zones III (sillimanite-biotite-garnet-muscovite) and IV (sillimanite-biotite-garnet­muscovite-K-feldspar) predominate in most of the quadrangle. Garnet and biotite compositions in Zone III yield peak temperature estimates of 635-670° and evidence of re-equilibration during cooling and unloading. A pressure of 6.3 kbar is indicated by garnet composition in equilibrium with cordierite and sillimanite. A zoned calc-silicate pod from Zone III contains bustamite in the core, as well as a variety of calc-silicate minerals with interesting disequilibrium textural relationships.

" •• its summit is a bald rock; on some parts of it are large piles of broken rocks ••• and plumbago in large quantities."

-Jeremy Belknap, 1792

INTRODUCTION

Location and Physiography

The Monadnock quadrangle is located in southwestern New Hampshire approximately twelve miles east of the Connecticut River and two miles north of the Massachusetts state line (Figure 1, No. 13). The prin­ciple towns in the quadrangle are Jaffrey, Dublin, Harrisville, Nelson, Marlboro, Troy, and Fitzwilliam (Appendix 1, Figure 41; Plates 1 and 4). Mt. Monadnock dominates the topography, its treeless summit rising to 965 m (3165 ft.) near the center of the quadrangle. Three watersheds meet at a point on the ridge extending south from the mountain: the Ashuelot River drains the area to the west, the Contoo­cook River the area to the east, and tributaries of the Millers River the area to the south. Water in the Contoocook eventually joins the Merrimack River, while that in the other two empties into the Connec­ticut River.

Numerous small lakes and ponds dot the region, many of them arti­ficially dammed so as to maintain higher water levels. In the south there are numerous bogs and swamps, and the lakes are shallow. In the north the lakes are mostly deeper and apparently occupy bedrock basins.

The railroad bed from Hancock to Keene (Plates 1 NW and NE), abandoned in 1938, affords some good bedrock exposure, as does the more recently abandoned Boston and Maine line which parallels Rt. 12 (Plates 1 SW and SE). Four east-west highways cross the quadrangle: Rt. 12 from Winchendon, Massachusetts, through Fitzwilliam toward Keene; Rt. 124 from New Ipswich through Jaffrey to Marlboro; Rt. 101 from Peterborough through Dublin and Marlboro toward Keene; and Rt. 9 from Hillsboro diagonally across the northwest corner of the quad­rangle. Rts. 202 and 137 run north-south through Jaffrey in the eastern part of the quadrangle. Many secondary roads and old logging roads form a network which provides good access to most areas.

Most of the region is heavily forested by either mixed hardwoods and conifers, by stands of pine, or by fir and spruce. A few dairy farms are still operating, but many pastures are growing up to junipers and saplings. Juniper thickets, notably on Gap Mountain and Bigelow Hill, are nearly impenetrable, and recent logging in some areas makes for rough going. However, most outcrops are accessible, and the summit of Mt. Monadnock provides approximately one half square mile of continuous outcrop. The "tree line" is at about 723 m (2700 ft.), artificially lower than would be normal for these latitudes due

3

~

Fig. 1. Index map of areas and quadrangles mentioned in text.

1.

2.

3.

4.

5.

6. 7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Littleton-Moosilauke (Billings , 19 3 7)

Mt. Washington (Billings, 1941)

Rumney SE (Malinconico, 1982)

Holderness (Englund, 1976)

Sunapee septum (Dean, 1977)

Mt. Kearsarge Alton-Berwick area

(Eusden et al., 1984) Bellows Falls

(Kruger, 1946) Lovewell Mtn. (Heald, 1950); Gilsum-Marlow area

(Chamberlain, 1985) Hillsboro

(Nielson, 1974) Concord

(G. Duke, 1984) Keene-Brattleboro

(Moore, 1949) Monadnock

(Fowler-Billings, 1949) Peterborough

(Greene, 1970; E. Duke, 1984) Vernon-Northfield area

(Elbert, 1984) Orange area

(Robinson, 1963) Tully body

(Pike, 1968) Ashburnham-Ashby area

(Peterson, 1984) Brooks Village breccia

(Morton, 1985) Barre

(Tucker, 1977) Ware

(Field, 1975) Amherst block

(Jasaitis, 1983)

1«-1 l!b.J

r:.J L:::J

•

4

White Mountain series

New Hampshire series

dome gneisses

to numerous fires in the early 1800's. The north slopes are densely covered by young spruce trees where the forest is recovering from extensive damage in the 1938 hurricane.

5

Landforms in the quadrangle are directly related to the underlying geology (Plate 4). Mt. Monadnock itself is held up by quartzite beds in the Littleton Formation, which are folded back on themselves to form a thick resistant sequence. Irregular hills with elevations from 365 to 610 m and relatively thin glacial cover predominate in the northern third of the quadrangle, where the rocks are poorly bedded gneisses of the Rangeley Formation west of Harrisville, and Kinsman Granite in the northeast. The two branches of Minnewawa Brook follow east-west trending layered rocks between Dublin Pond and Marlboro village, the south branch in a broad alluvial valley and the north branch cutting gorges into bedrock.

The remainder of the quadrangle consists of lower topography with moderate relief and a north-northeast grain parallel to the layering of the metamorphic rocks, interrupted by still lower areas underlain by intrusive igneous rocks. The Kinsman Granite, alone among the intrusive rocks, forms more resistant hills and ridges, even in the narrow bodies where the rock is strongly sheared (Plate 1 SE). The ridge culminating in Little Monadnock Mountain (Plate 1 SW) is parallel to a syncline extending south from Troy. Gap Mountain is underlain by inclusions of metamorphic rocks surrounded by Spaulding Tonalite. Roughly parallel to the western edge of the quadrangle, a line of west-facing cliffs at about 300 m elevation marks the location of the Clough Quartzite, separating the Monadnock area from the broad valley and irregular hills to the west which are underlain by gneisses of the Keene dome.

Drumlinoid hills of 335-365 m elevation and large areas of glacial outwash and till obscure the bedrock in the east-central and southeast parts of the quadrangle. A large ridge (elevation 365 m) made up of Kinsman boulders obscures the south margin of the Cardigan pluton from ~Lake Skatutakee to the area north of Bonds Corner, east of Dublin. Fowler-Billings (1949b) suggested this represents a moraine deposit. The other important area of no outcrop extends from Beech Hill in Dublin westward to north of Chesham.

Purpose

The main purpose of this thesis is to present the results of mapping of the layered rocks in the Monadnock quadrangle, New Hamp­shire, based on field work during the summers of 1981, 1982 and 1983. The area was suggested to me by Peter Robinson for several reasons. It is a key area for stratigraphic correlations between central Massachusetts, where Robinson and his students have been mapping (Robinson, 1967; Pike, 1968; Field, 1975; Tucker, 1977; Peterson, 1984; Elbert, 1984), and central New Hampshire where John Lyons and his Dartmouth students have been mapping (Nielson, 1974; Englund,

'.:

6

1974; Nelson, 1975; Lyons, 1979; Malinconico, 1982; E. Duke, 1984; G. Duke, 1984) (Figure 1). Secondly, the Monadnock stratigraphy lies between the Bronson Hill anticlinorium, with its Paleozoic sequence originally defined in the Littleton quadrangle by Billings (1937), and the Merrimack synclinorium, with its stratigraphic sequence which has recently been correlated with rocks in Maine (Hatchet al., 1983). The quadrangle straddles a Silurian "tectonic hinge"between the anti­clinorium, where the Silurian rocks consist of a thin shelf sequence, and the synclinorium, where the Silurian rocks thicken to form a clastic wedge up to 2300 m thick (Hatchet al., 1983) (Figure 2). Thirdly, in Massachusetts the map pattern consists of narrow parallel units striking nearly north-south, apparently the root zone for major west-verging nappes (Robinson and Hall, 1980). North of the state line the structural grain trends more to the northeast, and the plunge of the fold axes apparently steepens so that numerous fold hinges intersect the earth's surface. Thus, the Monadnock quadrangle is potentially an area in which to observe the large backfolds which post-date the nappes, and which are deformed by folds related to the rise of gneiss domes in the Bronson Hill anticlinorium (Thompson et al., 1968).

" •• the upland between Treves and the Rhine is one of the best examples qf an uplifted peneplain • • • • Here and there it is still surmounted by low, linear, eminences ••• These I would call 'monadnocks', taking the name from a typical residual mountain which surmounts the uplifted peneplain of New England in southwestern New Hampshire."

-Davis, 1896, p.192.

Previous Work

Hitchcock (1877) distinguished four main groups of layered rocks in the Monadnock quadrangle: ferruginous schists, fibrolite schists, pyritiferous schists, and the Montalban series. From his location descriptions it is clear that the first three correspond respectively to Rangeley, Littleton and Francestown Formations of the present study, although his map leaves out some of the locations described in his text. He described Mt. Monadnock itself as a "double synclinal". Davis (1896) cited Monadnock as the type locality for resistant moun­tains rising above the general erosion level, thereby coining the term "monadnock" for such isolated peaks worldwide.

Perry (1904) concentrated his mapping on the immediate area around the mountain and recognized three units: gray garnet-biotite­sericite-fibrolite schist with pseudomorphs after andalusite (Littleton Formation of this report), gray quartzose mica schist with biotite and hornblende (Warner Formation), and a very rusty fissile quartzose schist (Francestown and in part Rangeley Formations). He recognized the gradational nature of the Littleton/Warner contact. Perry mapped the orientation of foliation, and showed on his map the

w KEENE

DOME

Littleton

II

Tectonic Hinge"

MONADNOCK

SEQUENCE

upper Littleton

schist and quartzite ~kHill

----- ----lower Littleton

schist

PETERBOROUGH

(E. DUKE, 1984)

Littleton

CENTRAL N.H ..

(HATCH ET AL., 1983)

upper, thick-bedded

-----Littleton----­lower, thin-bedded

Madrid upper

0 rn < 0 z )> z

• \~:~~~··'·,,~:~, : , , w,; x • • ,,;, , , ~~:'': ·~"· ... .. . . . . . "' ~"'' cg .

gran~lite~ ~------ upPer, rusty matnx cgl. ......._ ---Rangel ........_ ey_

? ---- lower, gray - - ---- Rangeley ------___ ? __ _ 7 __

E

Fig. 2. Stratigraphic diagram showing proposed correlations across the tectonic hinge , after Hatchet al. (1983), to include the Monadnock and Peterborough sections. Black arrows show change in sediment source direction from Silurian to Devonian. See text for details of correlation for each unit, and Figure 4 for fossil control.

-....J

8

large fold which dominates the mountain's summit. He attributed this deformation to plutons pushing aside the metamorphic rocks. The contacts of plutons are irregular, with abundant offshoots parallel to foliation in the schist, leading Perry to conclude that a large batholith may underlie the entire area. He recognized two phases of folding, citing the recumbent fold near the summit as evidence for an early deformation at depths where rock flowage occurred, and what he thought was a systematic relationship between joints and the map-scale syncline as evidence for a later, more brittle, deformation at less depth, accompanying the granite intrusion.

The first detailed geologic map of the quadrangle was produced by Fowler-Billings (1949a). For the first time the various plutons were differentiated into members of Billings' (1937) plutonic series, and their contacts approximated. The present study does not add much to the geology of the plutons, but instead concentrates on the layered rocks. Most of the pluton contacts on Plate 1 are much as Fowler­Billings mapped them. Some structural data from her map (1949a) are included on Plate 2 in plutons and in the Keene dome, but people interested in more detail, and the location of quarries and mines, should refer to her map.

With the exception of thin horizons of Ammonoosuc Vocanics, Partridge Formation and Clough Quartzite along the eastern edge of the Swanzey (Keene) dome, at the west edge of the quadrangle, Fowler­Billings included all the layered rocks in the Littleton Formation. She subdivided the Littleton into four members: a lower schist member (Rangeley and Perry Mountain Formations in the present study), a rusty quartzite member including actinolite granulite and biotite schist (Francestown and Warner Formations), a middle schist member (Littleton Formation) and an upper member of rusty-weathering schist and gneiss (Rangeley Formation). The rusty quartzite serves as a marker horizon and defines a large map-scale syncline on Fowler-Billings' map, as well as smaller folds near Thorndike Pond and Hurricane Hill. She did not attempt to separate various fold phases beyond suggesting that large recumbent folds near the summit formed early, possibly as soft sediment features. She correctly observed that the joints are a rela­tively late feature. One isolated area of rusty quartzite near Derby Hill surrounded by the schists and gneisses of the upper schist member was mentioned by Fowler-Billings as warranting more work (1949b, p.31).

Some unpublished field notes by Peter Robinson and J.B. Thompson, Jr. (1966) from the area along the edge of the Keene dome south from Rt. 12, and along Grassy Hill, have been incorporated in the present study. Their outcrop locations and structural data are included on Plates 1, 2, 3, and Figure 12, Subarea 6. Thompson et al. (1968) mapped a contact between the Littleton Formation which lies east of the Keene dome and the sulfidic, rusty-weathering schists farther east. They interpreted the sulfidic rocks as Partridge Formation in the overturned limb of the Fall Mountain nappe. The axial trace of a

nappe-stage syncline would lie somewhere within the narrow belt of Littleton. The present study interprets the contact between the gray and rusty rocks differently.

9

Nelson (1975) made a detailed study of the area between the south­east foot of Monadnock and Thorndike Pond. He separated the rock types in Fowler-Billings' rusty quartzite member and correlated them with Francestown and Warner Formations, following Nielson's (1974) nomenclature in the Hillsboro quadrangle. Graded beds show that the biotite granulites and clean calc-silicates (Warner) are intermediate in age between the rusty calc-silicates (Francestown) and gray schists (Littleton). Nielson (1974) had proposed correlating the Francestown and Warner with the Smalls Falls and Madrid Formations in Maine (Moench and Baudette, 1970), which are Silurian. Therefore the rocks below the Francestown had to be still older, and Nelson tentatively called them Ordovician Partridge Formation. Nelson used this strati­graphy to map out a complicated interference pattern of folds; he proposed that a large recumbent fold with a steeply plunging, northwest-trending axis is refolded about isoclinal folds with axes that plunge gently northeast. The structural interpretation in this thesis is somewhat different.

Methods of Study

Field work in 1981 was started in the 7 1/2' Winchendon, Massa­chusetts, quadrangle (1:25,000, U.S.G.S., 1971), following the Coys Hill Granite and adjacent lay~red rocks northward into the SE quadrant of the 15' Monadnock quadrangle (1:62,500, U.S.G.s., 1949). It soon became apparent that the geology in areas of abundant outcrop could not be successfully mapped at the smaller scale, so pace and compass maps were made at 1:3,000 and 1:6,000, and only isolated outcrops were plotted directly on the u.s.G.S. base map. I also began exploring Mt. Monadnock itself, plotting stations both on aerial photographs (1:20,000, U.S.D.A., 1975) and on a 1:32,200 map published by the Appalachian Mountain Club (1972). In 1982 I gained access to pre­liminary 1:25,000 U.S.G.S. maps with six meter contour intervals, and all subsequent mapping was done at this scale with the exception of work on the summit of Monadnock and in selected areas of complex geology.

The New Hampshire portion of the Winchendon quadrangle was not included in this study, unlike that of Fowler-Billings (1949). Two areas of layered rocks in the Monadnock quadrangle were poorly covered, and need more work: the southwestern corner south of latitude 42° 46', and the entire town of Nelson.

Stations were numbered chronologically, with a two letter prefix for the township in which they occur (Appendix 1, Table 15 and Figure 41). Sample and thin section numbers correspond to the station numbers. In cases of multiple samples from one station, they were designated -A, -B, etc. Two exceptions to this system include the

!

main mafic dike on Mt. Monadnock, which has the same number (MK-54) throughout its length, and the MK (Jaffrey) series, in which the stations on Mt. Monadnock were unfortunately not numbered in exact chronological order with reference to stations elsewhere in the township.

10

Samples were selected for thin sections on the basis of weathering freshness, representative rock type, and geographical distribution. 110 thin sections were examined using a petrographic transmitted light microscope. Those for microprobe analysis and reflected light micro­scopy were polished with 0.3 micron grit. Modal percentages were estimated using density charts, and point counts on the basis of 2000 points per slide were made on several of the more homogeneous sections for comparison. Anorthite contents of plagioclase were estimated using the Michel-Levy method and Figure 17-3 of Jones and Bloss (1980). Detailed electron microprobe analyses were made during 1984 on the three spectrometer ETEC automated electron microprobe at the University of Massachusetts Department of Geology, using a 15 kV accelerating potential and a 0.03 microamp beam current. Data was corrected by the Bence and Albee (1968) procedure.

Acknowledgments

This paper is submitted in partial fulfillment of the requirements for the Ph.D. degree in Geology at the University of Massachusetts, Amherst. Peter Robinson suggested the project and served as my advisor. I wish to thank him for his enthusiastic guidance through all stages of the project, from my initial introduction to the regional geology, to analysis and presentation of the results. I am especially indebted to Peter for help in the interpretation of garnet zoning and in preparation of the colored plates. John Lyons kindly showed me the central New Hampshire stratigraphy, and loaned me two of Carl Nelson's samples for microprobe study. Katharine Fowler-Billings made her field maps and thin sections available to me. Discussions with Page Chamberlain, Virginia Peterson, and Edward Duke, who were all mapping in adjacent quadrangles, provided insight and incentive, as did field and office sessions with Norman Hatch, John Lyons, Bob Moench, Gene Boudette, Jim Thompson, David Elbert, Spike Berry, MaryAnn Malinconico, Peter Morton, Jeff Josephson, and many other people. Peter Robinson, Leo Hall, Don Wise, Charles Dickinson, and Thelma Thompson reviewed the manuscript and made many helpful suggestions. Thelma Thompson and Rachel Wing were much appreciated field assistants. Stephen Field helped me with opaque mineral identi­fication, and David Elbert, Kurt Hollocher, and David Leonard helped me with microprobe procedure. Harold Robinson colored preliminary versions of Plates 1 and 5, and helped prepare the final plates. The hospitality of Mary Emerson and all my friends at Boulderidge made the field seasons doubly enjoyable.

Support for field and laboratory work and preparation of publi­cations for this project was provided by grants from the National

Science Foundation, Earth Sciences Division, as follows: from the Geology Program EAR-79-15246 (to Robinson); from the Petrogenesis Program EAR-81-16197 (to Robinson and J.M. Rhodes); and jointly from the Crustal Structure and Tectonics Program and the Petrogenesis Program EAR-84-10370 (to Robinson).

"At length as the craft was cast to one side, and ran ranging along with the White Whale's flank, he seemed strangely oblivious of its advance--as the whale some­times will--and Ahab was fairly within the smoky mountain mist, which thrown off from the whale's spout, curled round his great, Monadnock hump ••• "

-Herman Melville, 1851

STRATIGRAPHY

11

In the present study Nelson's (1975) work is extended into the rest of the Monadnock quadrangle, and rocks below the Francestown Formation are subdivided into thinly bedded quartzites and schists of the Perry Mountain Formation and a variety of rock types assigned to the Rangeley Formation (Figure 3), largely on the basis of correlation with similar rocks in central New Hampshire (Hatchet al., 1983). An attempt is also made to tie in with the regional geology, especially current work by E. Duke (1984) in the Peterborough quadrangle, Chamberlain (1984) in the Gilsum-Marlow area, and Robinson (1963; 1967; Thompson et al., 1968) in Massachusetts.

For each of the major units in this study, a description of rock types, mappable subunits, distribution and thicknesses is presented, followed by a discussion of nomenclature, regional correlation and postulated derivation. Although thickness estimates were made in areas where structural repetition is not apparent, it should be empha­sized that the high probability of either structural attenuation or structural thickening makes these estimates useful only in a quali­tative way.

I have not made any detailed study of the units along the east edge of the Keene dome where the Ammonoosuc Volcanics, Partridge Formation, Clough Quartzite and, locally, Fitch Formation overlie the gneisses of the dome. Therefore the descriptions of these units are condensed, and based mainly on previously published observations.

SWANZEY GNEISS, AMMONOOSUC VOLCANICS AND PARTRIDGE FORMATION

Fowler-Billings (1949a) and Moore (1949) described the gneiss of the Swanzey dome (now generally called the Keene dome) as an intrusive granodiorite to quartz diorite pluton of the Oliverian magma series. Thompson et al. (1968) argued that the rocks exposed in the Keene dome as well as-other domes in the Bronson Hill anticlinorium were older than the cover rocks rather than intrusive into them. Robinson et al. (1979) described in some detail the roadcut on Rt. 12 in East Swanzey

z <t

z 0 > w 0

I

thickness 1

(meters) 1..:-:::-::;:-~.::_

~· ~=-.;: :::: -. >600

=i~~.i:0

?E~t=~ 650

Fig. 3. Stratigraphic Column.

upper part of LITTLETON FORMATION gray schist and quartzite

lower part of LITTLETON FORMATION gray schist and sparse quartzite

12

~-- = ;---upper part of WARNER FORMATION ______

37_- co -\ J gray granulite

/ ~~~~~~!~~~.----lower part of WARNER FORMATION

z <t

a: ::> _J

(/')

45 89 ~ clean calc-silicate granulite 62 ~FRANCESTOWN FORMATION

900

>600

-;:_~•.~.::_- · ~ sulfidic calc-silicate granulite . . . . ~- PERRY MOUNTAIN FORMATION

~- _.,.

thinly bedded schist and white quartzite

upper part of RANGELEY FORMATION (C?) gritty schist with quartz pebble lenses and calc-silicate pods

middle part of RANGELEY FORMATION (B?) sulfidic schist with calc-silicate pods

lower part of RANGELEY FORMATION (A?) gray feldspathic schist and granulite with calc-silicate pods and granulite­matrix conglomerate

13

(on the boundary between Plate 1 NW and Plate 1 SW) which exposes the dome gneiss and the overlying units. The predominantly coarse-graLned biotite-plagioclase gneisses contain hornblende amphibolite layers and boudins. Earlier workers had interpreted the amphibolite as Ammonoosuc Volcanics intruded by the felsic rocks. Conspicuous mag­netite porphyroblasts are common in some of the felsic gneisses. Although most now agree that the contact between the dome gneiss and the Ammonoosuc is not intrusive, there remains a controversy as to whether it represents an unconformity or, as Naylor (1969) suggested, a change in volcanic chemistry.

The Ammonoosuc Volcanics, which include a lower mafic unit and an upper felsic unit, represent the remains of a Middle Ordovician island arc (Leo, 1980; Schumacher, 1983). Some horizons in the mafic unit contain anthophyllite, gedrite or cummingtonite (Robinson and Jaffe, 1969a and 1969b). At the Rt. 12 roadcut there are approximately 25m of the formation (Robinson et al., 1979). Correlative rocks in Maine contain Middle Ordovician brachiopods (Boucot, 1961). The Swanzey Gneiss and Ammonoosuc Volcanics were probably joined to the North American continent during the Taconian orogeny (Hall and Robinson, 1980).

The rusty-weathering schists of the Partridge Formation consist of quartz, muscovite, biotite, plagioclase, graphite and sulfides, with or without garnet and sillimanite (Fowler-Billings, 1949a). The Partridge is locally absent, probably due to faulting, as for example in Forbush Brook north of Rt. 12. However, it is exposed in enough places above the Ammonoosuc to suggest it was formerly continuous along the edge of the Keene dome. At the Rt. 12 roadcut it is about 18m thick (Robinson et al., 1979). Correlative rocks in Maine contain Middle Ordovician-graptolites (Harwood and Berry, 1967).

Rusty schist dominates the southeast corner of the Monadhock quadrangle. At least some of these rocks are on strike with belts of sulfidic schist which extend south to Brimfield, Massachusetts, and which contain amphibolites and small ultramafic bodies, and appear to belong to the Partridge Formation. This correlation is strengthened by the 440 m.y. Hedgehog Hill gneiss which intrudes the Brimfield Group in Connecticut (Pease and Barosh, 1981). One outcrop of amphi­bolite occurs in a small roadcut on Rt. 12 in the Winchendon quad­rangle, 100 m east of the Sip Pond belt of Kinsman Granite. No amphi­bolites have yet been found in the Monadnock quadrangle apart from the Ammonoosuc, and lacking other reliable criteria for distinguishing the Partridge from rusty rocks of the Rangeley Formation, a contact is difficult to map. More detailed mapping in the Winchendon qudrangle may clarify the situation, but for now no Partridge is shown on Plate 1 SE. In theory, the rusty Rangeley should be reddish-weathering, more quartzose, less aluminous and perhaps should contain more

14

abundant calc-silicate pods (Norman Hatch, pers. comm., 1983). The rusty weathering tends to be yellowish or greenish in the Partridge, although very yellow-weathering rocks also occur in what I have mapped as Rangeley in Troy and Roxbury.

CLOUGH QUARTZITE AND FITCH FORMATION

The Clough Quartzite includes metamorphosed orthoquartzite and quartz-pebble conglomerate, micaceous quartzite, and quartz-garnet­muscovite schist (Fowler-Billings, 1949a). On Fowler-Billings' map the Clough is shown as discontinuous lenses above the Partridge Forma­tion. However, since the Clough is present everywhere that exposure is adequate, a continuous thin layer (25 m thick at the Rt. 12 roadcut, Robinson et al., 1979) is shown on my Plate 1. In two of her five measured sections across the edge of the Keene dome, Fowler­Billings reported pegmatite at the position where the Clough should be. Pegmatite intrudes the Clough as well as other units at the Rt. 12 cut, and in fact is common at many places along the edge of the Keene dome, obscuring contact relations and confusing thickness estimates. However, these problems are minor when we consider that major tectonic thinning has probably reduced the original thickness of the dome cover sequence by as much as 90% on this part of the Keene dome (Robinson, 1963). The Clough 9uartzite is well established as late Llandoverian (Figure 4), based on numerous fossil localities elsewhere in New Hampshire (Boucot and Thompson, 1963). The clean, coarse nature of the quartzite, and the fact that in some places it lies directly on the Ammonoosuc Volcanics or even on the dome gneiss (Thompson et al., 1968), indicate its lower contact is probably an unconformity.

Probable Fitch Formation was found north and east of Mt. Huggins, north of Rt. 12 (Plate 1 NW). These rocks include well bedded calc­silicate granulite and biotite-feldspar-quartz granulite. They occur in outcrops east of the Clough Quartzite on the steep south-facing

~ slope of a 246+ m knob north of Mt. Huggins, on a narrow 240+ m ridge northeast of Mt. Huggins, and in the swampy headwaters of a west­flowing brook east of Mt. Huggins. The Fitch is Pridolian (Late Silurian) based on conodonts near Littleton, New Hampshire (Harris et al., 1983). The Fitch is typically discontinuous elsewhere in the anticlinorium (Thompson et al., 1968), perhaps due to local non-depo­sition or erosion prior to deposition of the Lower Devonian Littleton Formation. In various places the Littleton lies directly on each of the older units, including dome gneiss (Robinson, 1963; Hatchet al., in press). Thus, apparently the base of the Littleton also represents an unconformity, at least where the underlying units are missing.

Bronson Hill anticlinorium

Littleton! __

,, Monadnock

Littleton

Fitch3 ______ 1_ _ _¥arner ------------

5 Clough ----

Francestown

Perry Mountain

northwestern Maine

central Maine

fossil ranges

~:: ~~~~~<:~ ~ .:::.-.:: ]-:-: ::.-_:-:.:::._:. -_-_-_-_-: ~-_.:

Carrabassett Carrabassett

__ )i~<!£~<!- -- -'--~~1_1_..!3!'.9.9~-- ----I I I I I I

Smalls Falls 41 I I I Parkman Hill-----1 : ! I I

I

I I I I I I

------...l I I

Perry Mountai~ unnamed unit . 4

Sangerville -

age

Ems ian

3941 t::l t%j

Siegenian <: 0

401 z H

Gedinnian ~ ·408

Pridolian

414-

Ludlovian

421J en H t""

I ~

Wenlockian H

~

'·28-

Llandoverian u ~~;.I~;--l-;;n~~l.-y6 == = ~===~ ~ ~~ -----~~ r----:_ ~-D IIi

I I I Waterville 'j1-------~ I

~- ·-Vassalboro 7-- --u----+ 438 _.__--l

l:ireenvale Cove louimby

8 lnixvi 11 P Partridge ORDOVICIAN

Fig. 4. Correlation chart (after Hatchet al., 1983), showing age ranges of fossil assemblages in units believed to be equivalent to the Monadnock stratigraphy. 1. Boucot and Arndt (1960); Boucot and Rumble (1980) 2. Boucot (1969) 3. Harris et al. (1983) 4. Pankiwskyj et al. (1976) 5. Boucot and Thompson (1963) 6. Moench and Baudette (1970)~.1Dsberg (1980) 8. Harwood-;ndlBerry (1967). .....

V1

~

16

RANGELEY FORMATION

Description and Distribution of Rock Types

Rocks assigned to the Rangeley Formation in the Monadnock quad­rangle include rusty-weathering sulfidic schists, gray-weathering schists, quartz-feldspar-biotite granulites, a variety of conglomerate horizons and lenses, and a variety of calc-silicate granulite horizons and lenses. Approximately 40% of the quadrangle is underlain by the Rangeley, distributed in three main areas (Plate 4): an area sur­rounding Little Monadnock Mountain in the southwestern part of the quadrangle and extending north to Marlboro, a northern area west of the Cardigan pluton and north of the Chesham Pond fault, and an elon­gate northeast-trending area east of the Thorndike Pond fault zone. In addition, the formation is well exposed below the Perry Mountain Formation in the folds southeast of Mt. Monadnock and surrounding the Spaulding Hill pluton. In the northern third of the quadrangle the rocks have gneissic textures, but the same rock types occur. In these gneisses strong foliation and bedding are hard to discern in most outcrops, except where calc-silicate pods mark primary layering.

The mapping of rusty- and gray-weathering rocks within the Rangeley Formation has not shown any clear internal stratigraphy; the situation seems to be more complicated than simply an upper rusty part and lower gray part, as suggested by Hatchet al. (1983) for central New Hampshire. Because of the heterogeneity of the unit, consistent internal stratigraphy should probably not be expected over wide areas. In the Monadnock quadrangle west of Troy, three informal parts have tentatively been identified (Figure 3 and Plate 1 SW).

17

The lower part is predominantly gray-weathering schist with calc­silicate pods and local granulite and granulite-matrix conglomerate. The middle part consists of a thick sequence of predominantly sulfidic schist with calc-silicate pods. The upper part consists of inter­bedded gray and sulfidic schists with gritty horizons, calc-silicate pods, and local lenses of quartz-pebble conglomerate. Bedding is . generally well defined in the upper part and beds up to 15 em thick with "slow grading" are present. Further detailed mapping with particular attention to the topping direction of graded beds is needed to verify the proposed scheme. Unfortunately graded beds are abundant only in the upper, gritty member. Both because of the uncertain division into three parts, and for the sake of clarity, the rock types in the Rangeley are described below without regard to the proposed parts.

Sulfidic schist and gneiss. Reddish-rusty, quartz-biotite­muscovite-plagioclase schist (Table 1a), with gritty horizons and calc-silicate pods, is the most common rock type in the Rangeley For­mation of the Monadnock quadrangle. The plagioclase ranges from An23 to An55 • Sillimanite, garnet, orthoclase, and retrograde staurolite and Mg-chlorite may or may not be present, and zircon, ilmenite, graphite, apatite, allanite, and tourmaline are accessory. The rusty weathering is due to the presence of iron sulfide minerals, which in thin section MK-564 were identified as pyrrhotite, marcasite after pyrrhotite, and minor chalcopyrite. Although the reddish color is characteristic, yellow and orange colors are also found, as well as gray-weathering beds interlayered with the rusty ones. The cause of the color variation in the rusty schists is uncertain.

Many outcrops of rusty schist in the Rangeley Formation contain gritty horizons. What were apparently sand grains and small pebbles ranging in size from 0.5 to 4 mm have been recrystallized. Sample MK-1061A, for example, is a very gritty rock with abundant 1 to 2 mm domains of quartz, and a few as large as 1 em. Thin section study of ~hese domains shows that they consist of grains finer than 1 mm. Relict pebble boundaries are not visible in thin section. Graded beds are locally well exposed in outcrop. Beds from 3 to 15 em thick show a gradual decrease in grain size, or "slow" grading. The locations of some outcrops with graded beds are shown on Plate 2 by dots above the strike and dip symbols.

,.,

Table 1a. Estimated modes of the Rangeley Formation: sulfidic schist and gneiss.

MK MK MK TR MK HV DB RI HV 564 1027 1061A 21 _l2.L 52 165 100 36

Quartz 28 32 47 52 24 52 32 25 32

Plagioclase 15 7 13 10 22 6 3 10 15 (An40) (An42) (An41) (An23) (An55) (An34)(Anl7-29)(An30) (An26)

Micro cline 4

Muscovite 11 32 15 2 5 11 34 54 29

Biotite 11 23 12 25 46 12 7

Garnet 2 8 4 4 3

Sillimanite 25 X 3 10 5 X

Staurolite 1 .,

Chlorite X X 3 23 7 10 (Retrograde) (Mg) * (Mg) (Mg) (Fe) (Fe-Mg) (Mg)

Opa<{Ues 3 2 1 2 2 3 4 3 Graphite 2 X X X X X Ilmenite 2 X X Pyrrhotite X X Chalcopyrite X Marcasite 6 X Undifferent'd X X X X X X X

Apatite X X X X

Zircon X X X X 1 X X X X

Tourmaline tr X

X less than 1%, but more or less ubiquitous tr trace

* chlorite compositions estimated optically ..... 00

List of Samples in Table 1a.

ffiZ-564 Gray medium-grained graphitic sillimanite schist. Weathers red rusty.

50m W of Stony Brook Farm Rd., 88m SW of Thorndike Pond, Jaffrey.

MK-1027 Gray massive gneiss with 2 em garnet porphyroblasts. Rusty­weathering.

Fresh rock blasted from Wallace driveway E of Thorndike Pond Rd., due E of rocky peninsula in Thorndike Pond, Jaffrey.

19

MK-1061A Gritty, massive, coarse quartzose schist, with up to 2 em garnet porphyroblasts and lenses several em long of sillimanite schist. Garnets have clear rims around "sugary"-textured cores which are full of inclusions: quartz, micas, tourmaline, opaques.

900m E of Thorndike Pond, 300m S of Dublin-Jaffrey town line, W of contact with Kinsman granite.

TR-21 Gray medium- to fine-grained foliated to granular schist. Weathers rusty and outcrop contains calc-silicate pods.

Rt. 12 roadcut 3.5 km W of Troy village

MK-192 Well foliated gray- to brown-weathering medium-grained granulite. Gray schist beds and calc-silicate pods in same outcrop.

350m S of Rt. 124, 425m SE of Jaffrey Center Fire Station.

HV-52 Massive, medium-grained blue-gray retrograded gneiss, with dark biotite-rich patches and 1-2 mm garnet porphyroblasts. Weathers rusty. Local calc-silicate pods.

Cellar hole W of S\v corner of Silver Lake, Harrisville. Exposure later covered by backfill; weathered outcrops E of house are extensive to the lake shore.

DB-165 Greenish-gray to rusty fine-grained well foliated retrograded ~ schist, with 2-5 mm garnets which vary from fresh to completely

replaced by chlorite. Roadcut contains sparse calc-silicate beds. N side of long roadcut, Rt. 101, E edge of Monadnock quadrangle.

RI-100 Rusty-weathering, medium-grained retrograded schist with abundant graphite. Superficially resembles "whiteschist".

Rt. 202 roadcut 2 km N of West Rindge.

HV-36 Coarse, massive blue-gray retrograded gneiss with quartz-mus­covite "augen".

Jeep road between Blood Hill and Cobb Hill, elev. 507m, Harrisville.

Conglomerates. Two distinct types of conglomerate occur in the Rangeley Formation: quartz-pebble conglomerate, and granulite- matrix conglomerate. Occurrences are shown by a separate color on Plate 4. We will return to the significance of these horizons in a later dis­cussion of regional correlations. Localities are listed in Table 1b and estimated modes in Table 1c.

Quartz pebble conglomerate lenses and beds were found in several localities, most of which are within 250 m of a contact with younger formations. They are generally enclosed in rusty rocks, although at MB-81 and TR-143 they are associated with gray schist. Station TR-65 is the best exposed of the quartz-pebble conglomerates. The bed is about 50 em thick and of indeterminate lateral extent. Pebbles are flattened in the plane of foliation at a low angle to the basal contact with the underlying rusty micaceous quartzite. The pebbles consist almost entirely of vein quartz and white quartzite, with rare calc-silicate granulite clasts which give the outcrop a pitted surface where they have weathered out. Although it is a grain-supported conglomerate, there is a calc-silicate matrix which includes diopside, garnet, clinozoisite, actinolite, sphene and calcic plagioclase. Sample MK-1051 also has a calc-silicate matrix.

The granulite-matrix conglomerate occurs in a lens which is at most 100 m thick, and which is exposed below the west-facing cliffs ENE of Mt. Huggins (SZ-16 and SZ-23), in a brook north of Page Hill (MB-297), and on the east-facing slope of Page Hill. A similar but probably separate horizon is exposed approximately 400 m to the east. The granulite-matrix conglomerates are much more feldspathic and biotitic than the quartz-pebble conglomerates, and they are associated with gray schist and granulite. The clasts are similar to the matrix in mineralogy (see Table 1c, SZ-23), consisting mainly of quartz, plagioclase, biotite, and garnet. There are some vein quartz clasts,

: and others that may have been intrusive igneous rocks. The clasts have been flattened and elongated to shapes up to 10 X 3 X 1 em, so that in cross sections parallel to their long axes they resemble thin granulite beds.

20

21

Table lb. List of conglomerate localities in the Rangeley Formation.

TR-65 366+m hill N of High Street, 875m W of Troy village green. In contact with rusty gritty massive schist, about 125m below Perry Mountain Formation. See Table lc.

TR-130 In brook, 400m S of TR-65. 15cm-thick pebble horizon in rusty sillimanite schist.

TR-143 1600m NW of TR-65, elev. 416m, lOOm E of \.Jest Hill Rd., Troy. Gray sillimanite schist, quartzite, and quartz pebble conglomerate lenses 15cm thick. Rusty rocks with calc-silicate pods 20m to W.

TR-106 On power line, 600m NE of TR-143, elev. 372m, W of gully. Quartz pebble boudin in bedded quartzite and rusty schist.

MK-701A West of Stony Brook Farm Road 850m SW of Thorndike Pond. Horizon with 50mm quartz grains in gritty rusty schist.

MK-1051 See Table lc.

RX-147 \vest of Derby Hill window, W of Willard Hill, elev. 398m in N-S brook. Sparse 1.5cm quartz pebbles in somewhat rusty schist.

HV-63A Float block 6m H of Perry Mountain Formation, elev. 462m, Derby Hill. 0.5-lcm quartz pebbles and a few garnet schist clasts in matrix-poor conglomerate.

MB-290 425m WSW of Glen Brook Pond, knobby outcrop. Quartz pebble conglomerate boudin in red rusty schist with sharply bounded quartzite beds, some pitted.

HB-81 450m NNE from E end of Clapp Pond, in woods west of field. Garnet-sillimanite schist with quartzite beds and quartz pebble conglomerate horizons. Pebbles elongated down-dip toward NW.

SZ-23 See Table lc.

SZ-16 500m E of SZ-23, 200m E of top of cliff. Granulite matrix conglomerate with granulite and quartzite clasts, interbedded with gray schist and feldspathic granulite.

MB-303 Small ledge in brook, 750m SE of SZ-23, similar rock to SZ-23.

Mll-295 Dip-slope ledge on E side of Page Hill. Granulite and con­glomerate intruded by pegmatite. Strong pebble lineation.

MB-326 Poor outcrop south of Rt. 124, 250m W of Troy Rd., Marlboro. Quartz pebble horizons, quartzite and schist intruded by pegmatite. About 80m W of outcrops with graded beds topping to the west.

... Table 1c. Estimated modes of the Rangeley Formation: conglomerates and calc-silicate pods.

Conglomerate Calc-Silicate Pods

MK sz sz TR MK MK DB HV RX TR RX MB 1051 23 23A 65 294 294A 90 44 54 93 5 75A

Quartz 62 36 78 72 38 28 49 34 43 48 14 44

Plagioclase 15 36 11 6 32 18 27 39 36 39 40 2 (An51)(An37+)(An37+)(An70+) (An73) (An73)(An65+)(An67+) (An67+) (An75) (An64+)

Diopside 20 6 28 11 5 14

Actinolite 1 16 1 3 13 1

Clinozoisite 3 19 13

Ferrian Zoisite 2 tr

Garnet 2 2 3 4 6 3 7 10 4 20 tr

Biotite 26 8 -,

10 6 3

Huscovite X 4 14 2 19

Chlorite (Fe-Mg) 5

Sphene 1 2 2 1 2 X 7 5

Opaques tr X X 2 tr 1 1 2 2 1 2 Graphite X X X X X X X X X X X Pyrrhotite X Undifferent'd X X X X X X X X X

Apatite X X X X X 1 2 1

Zircon X X X X X X

Tourmaline X X

Calcite 3

X less than 1%, but more or less ubiquitous N

tr trace N

List of Samples in Table lc.

MK-1051 Massive, medium-grained conglomerate with O.Scm quartz pebbles in a weathered calc-silicate matrix. Diopside in hand sample.

Elev. 363m, E-facing ledges W of Frost Pond, Jaffrey.

SZ-23 Gray to brownish granulite-matrix conglomerate with quartzite and vein quartz "stretched" pebbles. SZ-23 is the matrix, SZ-23A is a quartzite clast. Clasts most apparent on weathered surface.

Elev. 264m, W of tall cliffs, N of Page Hill, Swanzey.

23

TR-65 Gray to white quartz pebble conglomerate, with pebbles flattened in foliation, up to Scm long. Vein quartz, quartzite, and garnet schist clasts.

366+m hill N of High Street, 875m W of Troy village green.

MK-294 Fine-grained, white mottled with green and interlayered with pink calc-silicate granulite. Pod from rusty schist.

300m N of Rt. 124, 1400m E of Cummings Pond, about lOOm E of Perry Mountain Formation, Jaffrey. MK-294A is the pink layer.

DB-90 Fine-grained, mottled, pale gray calc-silicate granulite with dark green chlorite and biotite, and a 2mm garnet horizon. Pod from quartzose rusty schist. Weathers light tan to pink.

E side of 364m hill E of Rt. 137, 750m N of Jaffrey-Dublin town line.

HV-44 Fine-grained, mottled, pale green calc-silicate granulite with dark green actinolite up to 4mm long. Center of pod from rusty gneiss.

Small roadcut NW of Childs Bog, Harrisville.

RX-54 Medium-grained dark gray calc-silicate granulite, with up to 4mm garnets and patches of biotite. Pod :rom rusty schist.

40m W of S end, Woodward Pond, Roxbury.

TR-93 Fine-grained dark gray calc-silicate granulite with up to 2mm t vitreous quartz grains. Pod in rusty schist. Ridge NW from Troy, elev. 390m.

RX-5 Medium-grained pale gray calc-silicate granulite mottled with actinolite, diopside and garnet, from pod in gray augen schist (RX-2, Table ld). Weathers tan.

Elongate ridge 200m SE of Hardy Hill, Roxbury.

MB-75A Magnetic fine-grained dark gray calc-silicate granulite with darker porphyroblasts. Sulfides visible in hand specimen. Weathers with a punky brown crust.

Float block in brook, north of NE end, Clapp Pond, Marlboro.

24

Calc-silicate granulite. Lenses, pods, or "footballs" (Field, 1975, p.28) of calc-silicate granulite are common in both rusty and gray rocks of the Rangeley Formation. In some areas at least one pod can be found in every outcrop, but this is not true everywhere. They range in size from 10 to 100 em and although most are indeed foot­ball-sized, few if any approximate feotballs in shape. Flattened loaves of bread might be a more apt description, with the shortest axis perpendicular to bedding in the enclosing rocks. There is commonly a very hard, flinty, homogeneous core surrounded by a rim of less resistant feldspathic granulite that weathers to form a shallow depression. The cores are white, to light green, to shades of gray and black, spotted by green and red. They are composed of quartz, calcic plagioclase, and various amounts of diopside, grossular garnet, clinozoisite, ferrian zoisite, biotite, calcite, sphene, and muscovite alteration of plagioclase (Table 1c). Apatite, zircon, and opaque minerals are accessory.

Locally there are continuous beds of calc-silicate granulite, most notably at station TR-54, a large roadcut on Rt. 12 5.5 km west of Troy, where calc-silicate beds up to 30 em thick are interlayered with rusty schists. Elsewhere thinner beds can be found, but discontinuous lenses are much more common. Where several pods are aligned the question arises whether they were formed as discrete bodies, perhaps as concretions, or whether once-continuous beds have been disrupted by boudinage. The rims and rounded contours tend to favor a concretion­ary or~g~n. Where interlayered calc-silicate beds and schists have obviously undergone boudinage, for example in the Francestown Formation at HV-54 near Derby Hill, the calc-silicates have squared­off ends and there are scar folds in the schist. Guthrie and Burnham (1985) suggested that some pods in the Rangeley may have formed as rip-up clasts or blocks.

Gray schist and gneiss, granulite, and augen schist. Not all the schists of the Rangeley Formation show rusty weathering. Many outcrops are monotonous gray rocks (Table 1d), and an attempt has been

~ made to map out the larger areas of gray-weathering rocks. It should be emphasized, however, that many of these rocks are also interbedded with the rusty schist. A comparison of modes from the two groups (Tables 1a and 1d) shows very little difference in average mineralogy. Many gray schists in the Rangeley greatly resemble rocks of the Littleton Formation. Although the number of samples is not large enough to draw statistically meaningful conclusions, the plagioclase content of the Rangeley gray schists (range 0-14 modal %, average 4%) seems to be greater than in Littleton gray schists (range 0-10%, average 1.7%). Anorthite content in plagioclase from gray Rangeley schists ranges from An13 to AnJ6· Locally the schist grades into quartz-plagioclase-biotite granulite. Granulite horizons are most common in the western part of the quadrangle, in the lower part of the formation. Platy minerals (biotite, muscovite and retrograde chlorite) seem to be more abundant in the Littleton. However, these subtle differences are not good field criteria for mapping. A more

25

reliable hallmark of the Rangeley gray schists is the presence of calc- silicate granulite pods similar to those in the rusty schists. By contrast, only the lowermost 30 m or so of the Littleton Formation in the Monadnock quadrangle contains calc-silicate pods. It may be significant that the schists near the base of the Littleton are also more plagioclase-rich (Table 6, MK-604 and HV-135).

On Cobb Hill north of Lake Skatutakee (Plate 1 NE) there are some exposures of extremely aluminous gray schist with as much as 53 modal % sillimanite (Table 1d, HV-41). Fowler-Billings (1944b) described this occurrence in some detail. The sillimanite-rich schist is about 15 m thick, with coarse sillimanite pseudomorphs after andalusite averaging 1.5 em in length. Her samples K-116A and K-117 from Cobb Hill also contain cordierite and alkali feldspar.

In some parts of the Rangeley Formation, but especially in the gray-weathering schist and gneiss, segregations of quartz and muscovite, with or without feldspar and sillimanite, form augen. Some of these, for example FZ-30 in Table 1d, probably represent pockets of recrystallized pegmatitic melt. Others, such as RX-2A, may have replaced K-feldspar porphyroblasts. Heald (1950) observed orthoclase porphyroblasts surrounded by shells of quartz and muscovite in the Lovewell Mountain quadrangle. The augen textures may have been accentuated by mylonitization during deformation. When I did most of the field work, I overlooked the importance of trying to distinguish between augen formed by these three processes. Those which appear to be pseudomorphs are most common in the northern third of the quad­rangle, but it is uncertain whether they are restricted to this area. Northwest of Derby Hill there is a small gravel pit in saprolite that consists of deeply weathered augen schist which somehow escaped removal by the glaciers (location shown on Figure 20).

Thickness

It is impossible to estimate accurately the total thickness of the Rangeley Formation in the Monadnock quadrangle, because the location of the lower contact is highly uncertain. If there were no strati­graphic repetitions in the area west of Troy, the maximum thickness would be approximately 2825 m. However, graded beds topping west in outcrops south of Rt. 124 approximately 3 km southeast of Marlboro village make this assumption suspect. There is apparently a syncline with rocks belonging to the upper part of the Rangeley in its center. West of this syncline there are about 900 m of the middle part of the formation, and 600 m exposed of the lower part, below which there is a proposed fault. The upper part of the Rangeley is about 600 m thick at High Street in Troy. The maximum exposed Rangeley, taking into account the syncline, would be about 2100 m. In Maine the Rangeley Formation thickens abruptly eastwards from its westernmost outcrops to a maximum of about 2700 m and then gradually thins again toward the southeast, primarily by loss of coarse clastics (Moench, 1970).

,,..

Table ld. Estimated modes of the Rangeley Formation: gray schist and gneiss.

(Retrograded)

RX RX FZ RX RX HV K K K NL NL MK DB DB 2 2A 30 186 ~ 41 116A 117 102 2 29 417 130 97

Quartz 33 83 63 47 55 4 X X X 36 33 46 51 45

Plagioclase 3 7 3 13 5 5 8 14? * tr

Orthoclase 2 (An23)(Anl5?)(An34) X X X (An36)(An26)(An40)(Anl3)

Muscovite 11 15 8 16 9 12 X 11 24 30 15 30

Biotite 31 1 17 24 11 12 X X X 19 5 8 7 11

Garnet 6 1 5 17 X X X 12 6 2 tr 8

Sillimanite 12 5 9 6 53 X X X 5 2 X 1

Cordierite X X X 10?*

Staurolite (Retro.) 1

Chlorite 1 X 7 23 2 tr 2 (Retrograde) (Fe) (Mg) (Mg) (Fe) (Mg) (Mg-Fe)

Opaques 2 X X X X 2 X X X 4 3 2 2 Graphite X X X X X Ilmenite X

I X X 3

Undifferent'd X X X X X X X X X X

Apatite X X X X

Zircon X X X X X X X X X X X X X X

Tourmaline X

Calcite X * Strongly altered: some alteration resembles that typical of cordierite. X less than 1%, but more or less ubiquitous, except for K-116A, K-117, K-102, where

X denotes presenc~ since mode was not estimated. N 0\

27

List of Samples in Table ld.

RX-2 Massive, medium-grained gray sillimanite schist, with 2 mm garnets and 2-5 em quartz-muscovite "augen" (RX-2A). Gritty 2 mm quartz and 1 em sillimanite form nubbles. Calc-silicate pod (RX-5) is in the same outcrop.

Elongate ridge 200m SE of Hardy Hill, 1075m SSE from Woodward Pond, Roxbury.

FZ-30 Well foliated medium-grained quartzose sillimanite schist with strong fibrolite and mica lineations, gray with muscovite spangles in foliation plane. 1 em-thick quartz-feldspar segregations.

40m due W from Little Monadnock summit, Fitzwilliam.

RX-186 Strongly foliated medium-grained black to gray biotitic schist with 0.5 mm quartz clasts and 3 em quartz-feldspar segregations.

Elev. 277m, due E 425m from Roxbury Town Hall, !375m S of Otter Brook Dam.

RX-189 Gritty, medium-grained, foliated gray sillimanite schist with quartz clasts up to 0.5 em. Outcrop contains quartz-feldspar segre­gations and calc-silicate pods.

Elev. 326m, 1,.1 side of 345m knob 825m SSE from Otter Brook Dam.

HV-41 Massive,coarse, gray,unusually sillimanite-rich schist, with up to 1 em garnet porphyroblasts and up to 3 em bundles of sillimanite, probably pseudomorphs after andalusite.

Cobb Hill, elev. 573m, E side of western knob, Harrisville.

K-116A Approximately same location as HV-41.

K-117 Cobb Hill, west of K-116A.

K-102 Garnet-rich schist. S of Cobb Hill, elev. 566m.

NL-2 Gray, massive, medium to coarse gneiss, with 2 mm quartz grits and quartz-feldspar segregations. Calc-silicate pods in same outcrop.

t 490m hilltop 150m S of road that leads SW from Nelson to Thunder Hill.

NL-29 Massive, medium-grained gray gneiss with 4-6 em quartz-feldspar­muscovite segregations. Outcrop contains calc-silicate pods.

Elev. 475m, whaleback outcrop W of road SW of Tolman Pond, Nelson.

MK-417 Medium-grained gray feldspathic schist, interlayered with gray granulite in outcrop.

Ledge, E edge of swampy pond 750m SSE of Poole Reservoir, Jaffrey.

DB-130 Coarse blue-gray foliated muscovite-biotite gneiss. Brook N of Hud Pond, elev. 306m, 750m N of Rt. 101, Dublin.

DB-97 Coarse gray schist with muscovite after sillimanite and garnets up to 1. 5 em.

W-facing ledges, 450m E of Thorndike Pond Road, 75m S of marsh.

28

Age and Correlation

In central and southern New Hampshire, Hatchet al. (1983) have proposed that monotonous gray pelitic schist and the-overlying, slightly rusty-weathering pelitic schist with local calc-silicate boudins and graded beds correspond to distal equivalents of the Rangeley Formation in Maine south of the type area. In Maine the Rangeley Formation contains a variety of clastic rock types, with polymictic conglomerates at the base (Rangeley A), rusty metamorphosed shale with graded interbeds of quartzite and two conglomeratic layers (Rangeley B), and an upper pelitic member with quartz granule and peb­ble conglomerate beds which locally contain abundant calc-silicate minerals (Rangeley C) (Moench and Boudette, 1970). Moench and Boudette base a Silurian age (late Llandovery) for the Rangeley on fossils in calcareous quartz conglomerate at Blanchard Ponds, Maine, which does not occur in the type section, but is correlated with part of Rangeley C.

In the Monadnock quadrangle the lower gray schist with granulite and granulite-matrix conglomerate may correspond roughly with Rangeley A, the uppermost schist containing quartz-pebble lenses with Rangeley c, and the intervening sulfidic rocks with Rangeley B. Boone et al. (1970) demonstrated evidence for a sediment supply to the west-or-­north for the Merrimack synclinorium during the Silurian. The presence of conglomerate horizons in the Monadnock and Gilsum-Marlow areas suggests a more proximal site of deposition than for Rangeley equivalents in central New Hampshire, or in the Peterborough quad­rangle where no conglomerates have been reported (E. Duke, 1984) (Figure 2). Duke mapped a lower gray schist unit and an upper rusty schist unit which he suggested correlate respectively with Rangeley B and C. He has locally mapped out a volcanoclastic member (Haunted Lake Member) at the top of the Rangeley, and a banded calcareous granulite horizon toward the base of the rusty unit. A thorough discussion of several alternative stratigraphic interpretations was presented by Peterson (1984) for the same rocks to the south near the

t Massachusetts border.

In the Ware and Barre quadrangles, Massachusetts, some of the schists mapped as Partridge and Littleton Formations by Field (1975) and Tucker (1977) may be Rangeley equivalents (Robinson et al., 1982a). The Lyon Road anticlinal belt in particular contains rocks lithically identical to the red rusty Rangeley of the Monadnock area. Rusty units containing amphibolites probably should retain assignment to the Partridge Formation. Gray schist overlying them, particularly where there is graded bedding (Tucker, 1977), might then correlate with the lower part of the Rangeley. In light of the present study it seems doubtful that the Littleton Formation would lie unconformably on the Partridge Formation so far east of the Bronson Hill anticlinorium, if the intervening Silurian formations thicken toward the east, as believed. Most of the rocks immediately west of the Hardwick pluton are on strike with rocks assigned to the Rangeley Formation in the

29

Monadnock quadrangle.

Chamberlain (1985) has found a variety of conglomerates in the Gilsum-Marlow area, including massive clean quartzite, granulite-matrix polymictic conglomerate, and quartz-pebble lenses and beds. Some of these had previously been mapped as Clough Quartzite (Heald, 1950; Trask and Thompson, 1967; Dean, 1977) but others are new discoveries. It is becoming apparent that there is more than one conglomerate horizon in the stratigraphy, and the question arises as to how the Clough Quartzite and the Rangeley Formation correlate. Both the Clough and the rocks at Blanchard Ponds which are correlated with Rangeley C contain late Llandovery fossils (Figure 4), but the fossils at Blanchard Ponds only date the Rangeley C unit, with no direct control on the age of the lower parts of the formation. If, as traditionally viewed (Osberg et al., 1968; Hatchet al., 1983), the Rangeley represents an eastwardly thickening facies of the Clough, then in the Monadnock area the Rangeley has been structurally trans­ported into proximity with the Clough. On the other hand, the Clough might lie at the base of a thick sequence of Rangeley-equivalent rocks, including what is interpreted as Littleton Formation along the east edge of the Keene dome (Plate 4), and for some reason the basal unit of the Rangeley contains clean quartzite at this latitude rather than the polymictic conglomerate of Rangeley A in Maine. I prefer the former interpretation, as represented in Figure 2. According to this scenario, lower units of the Rangeley would have been deposited in pre-Clough time, which along the Bronson Hill axis was a period of non-deposition. This interpretation necessitates a nappe-stage west­directed overthrust in the Monadnock quadrangle between the Bronson Hill sequence and the Rangeley Formation.

West of the Connecticut Valley border fault there are three areas of probable Rangeley Formation: the top of Fall Mountain, Walpole, New Hampshire; rocks above the Ashuelot pluton northwest of Winchester, New Hampshire; and the "Amherst block" between the Hartford and Deerfield Mesozoic basins in Massachusetts. In the last area, Jasaitis (1983) mapped out gray and rusty pelitic units as well as lenses of conglomerate, some of which have a granulite matrix. These three areas represent rocks which were transported far to the west during the nappe stage of deformation.

Derivation

The Rangeley Formation formed as a clastic wedge along the basin margin, proabably derived from rapid erosion of the former volcanic island arc and Taconian Mountains to the west. The Taconic uncon­formity, so well documented along the axis of the Bronson Hill anticlinorium, may be locally lacking in Maine where Upper Ordovician rocks grade upward into the Silurian (Pavlides et al., 1968; Boone et al., 1970, p.14; Hatchet al., 1983, p.758). Sediments were -­apparently being shed more or less continuously into the Merrimack trough as the sea gradually transgressed toward the west during the

30

Silurian. The immature, polymictic rocks of the lower part of the Rangeley indicate rapid sedimentation, whereas the cleaner quartz-rich rocks near the top represent the transition to a more deeply weathered source area with more extensive reworking of sediments. Deformation of clasts and recrystallization of grain boundaries makes any analysis of rounding characteristics impossible in the conglomerates of the Monadnock quadrangle.

PERRY MOUNTAIN FORMATION

Description and Distribution of Rock Types

In the Monadnock quadrangle an interval of thinly bedded (2-5 em) clean quartzite and gray to slightly rusty schist occurs in most places between the Rangeley and Francestown Formations. Quartzite­schist contacts are sharp, and graded beds are sparse. Where graded beds are present the topping directions are commonly hard to read. Because the Perry Mountain Formation has not previously been recognized in the quadrangle, the most important exposures are listed in Table 2. Some outcrops in the upper part of the Rangeley Formation resemble the Perry Mountain, and indeed the contact may be gradational. Following Hatch et al. ( 1983), the term "Perry Mountain" is restricted to sharply bedded clean quartzite and schist. East of the Thorndike Pond fault zone, little if any Perry Mountain Formation can be recognized at the appropriate position.

The quartzites consist of quartz, muscovite, biotite, garnet, and opaques, with or without sillimanite, plagioclase, retrograde chlorite, and accessory apatite, zircon, and tourmaline (Table 3). The schists are gray to somewhat rusty-weathering, quartz-biotite­plagioclase-garnet rocks (Table 3, DB-289A and MK-342) with or without sillimanite, muscovite, orthoclase, and accessory opaques and zircon. Locally there are large sillimanite pseudomorphs after andalusite.

Thin garnet-quartz granulite beds and lenses are locally present t in the Perry Mountain Formation. One such coticule bed is exposed on

Hurricane Hill in an outcrop of sillimanite-rich schist (DB-289A). The mode estimated from thin section is 70% quartz, 28% garnet, and the remainder biotite, sillimanite and opaques (DB-289B). John Lyons and Norman Hatch (pers. comm., 1983) and Eusden et al. (1984) have also observed coticule in the Perry Mountain Formation.

A different sort of coticule occurs within quartzite beds as 1-3 em lenses which weather out to form pits. Similar pitted quart­zites also occur in the upper part of the Rangeley. The best exposure of pitted quartzite is at MK-382 (Table 2, Location B) where it occurs in nearly vertical beds two meters from the Francestown Formation. One unweathered lens observed in thin section is composed of garnet, quartz, biotite, apatite and opaques (Table 3, MK-382A). The quartzite immediately around the lens is more quartz-rich than the surrounding quartzite. Both the rim and the lens contain 5% apatite.

Table 2. Important exposures of the Perry Mountain Formation.

A In complicated folds south of Poole Reservoir at Monadnock State Park, especially E of foot trail E of Meade Brook, Jaffrey (see Figure 24).

B W of Dole Brook 300m W of Whites Pond, N of Rt. 124 3 km W of Jaffrey Center (MK-382W). Contact with Francestown Formation is folded. Fossil-like pits in quartzite bed.

C At "Porcupine Ledges" along X-C ski trail N of Rt. 124, 1.9 km W of Jaffrey Center (}~-309).

D N of the old Troy- Dublin Road, 1.9 km E of Gleason Brook bridge (DB-190, DB-192).

E On SE ridge of Hurricane Hill 400m due W from Dark Pond, Dublin (DB-289 and others) (see Figure 18).

F On a knoll on the E shore of Howe Reservoir, Dublin, on south limb of "Howe Reservoir syncline" (DB-283). Less well exposed on north limb.

G N of the Old Chesham Road 1. 4 to 1. 8 km E of its junction with Rt. 101, Marlboro, where there is a small graphite prospect (MB-208).

H At Derby Hill in Harrisville, W of chimneys in clearing (HV-55-57) and also N of summer house on NE part of hill, where there are coticule beds and minor folds (HV-130 and -134) (see Figure 20).

I On the ridge NW of Willard Hill (RX-18) and in the swamps W of Willard Hill (RX-77-78) (see Figure 19).

J On a small ridge at elevation 470m S of the E-W power lines which cross Little Monadnock ridge 3.3 km S of Troy (TR-217).

K Several poorly exposed outcrops N of High Street 600m W of Troy.

31

'"'

Table 3. Estimated modes of the Perry Mountain Formation.

Quartzite Coticule I Schist

MK MK MK MK MK MK MB DB DB MK 587 677 356 382W 382A 382B 165 289B 289A 342

Quartz 77 73 64 78 10 81 31 70 43 58

Plagioclase tr 3 8 (An27)

Orthoclase I 5 I

Muscovite 9 15 18 2 16

Biotite 8 6 4 4 7 2 1 tr 22 14

Garnet 1 3 4 12 73 10 53 29 4 1

Sillimanite 2 X 22

Chlorite 1 2 8 (Retrograde) (Mg) (Fe+Mg)(Fe)

Opaques 2 1 1 2 5 2 2 X I

1 3 Graphite X X X X Pyrite X Magnetite X

I X

Undifferent' d X X X X X X

Apatite X 5 5

Zircon X X X I X X

Tourmaline X X

Allanite

I X

Grunerite 13

w N

List of Samples in Table 3.

MK-587 Gray fine-grained quartzite. Ark Brook, 700m NE of State Park headquarters, elev. 408m, Jaffrey.

MK-677 Gray well foliated micaceous quartzite. Woods S of Harling Trail, E of Ark Brook and Hinkley Trail, Jaffrey.

ffiZ-356 Well foliated, fine-grained gray to brown micaceous quartzite. Chlorite and biotite intergrown. Garnet concentrated in layers.

300m N of Rt. 124, 1400m E of Cummings Pond, east of ridge held up by Francestown Formation, Jaffrey.

MK-382W Pale gray fine-grained quartzite with pitted horizons, inter-bedded with gray schist. In contact with Francestown Formation to

the east (MK-382). Three modes from one thin section are listed in Table 3 as follows: MK-382\v, quartzite; MK-382A unweathered coticule from pit; MK-382B rim around the pit, about 3mm thick.

Dole Brook, 305m W of Whites Pond, north of former "Mountain Shade Inn" (Nelson, 1975), Rt. 124, 3 km W of Jaffrey Center.

MB-165 Fine-grained, layered pink to gray to green quartz-garnet­grunerite granulite, from a boudin in well bedded gray quartzite and schist.

325m SE of Horse Hill Road-Richardson Road junction, Marlboro, on small N-S ridge north of stone wall.

DB-289B 2 em layer of fine-grained coticule in sample DB-289A.

DB-289A Mottled coarse schist with biotite-fibrolite clumps. 120m SE of Hurricane Hill summit, on ridgeline, 400m due W from

Dark Pond, Dublin.

Mk-342 Gray schist with 5mm garnets, well foliated, quartz and feldspar somewhat segregated into streaks.

400m E of Meade Brook, 400m S of Poole Reservoir, Jaffrey.

33

34

Some of the pits resemble the molds of brachiopod fossils. Similar rocks in the Clough Quartzite at Hetty Brook, Croydon Mountain, New Hampshire, turned out to yield diagnostic fossils. There, "lenticular masses of garnet simulate the shapes of fossils" (Boucot and Thompson, 1963). Samples from MK-382 have been sent to Arthur Boucot for study.

A third type of coticule was found in the Perry Mountain Formation on a small ridge east of Horse Hill Road, Marlboro, as an 80 X 20 em boudin consisting of garnet, quartz, grunerite, and opaques (MB-165).

Thickness

The Perry Mountain Formation is relatively thin in the Monadnock quadrangle. Some maximum estimates are 61 m on Old Chesham Road ridge; 53 m north of High Street in Troy; and 73 m from an area north of Rt. 124, 245 m southeast of Porcupine Ledges. In the folded rocks south of Poole Reservoir, a maximum of 18 m was estimated. The formation is absent in most of the area east of the Thorndike Pond fault zone. The exposed section of Perry Mountain Formation in the type area is 370m thick (Boone, 1973). Hatchet al. (1983) estimated a thickness of 500 min central New Hampshire, while E. Duke (1984) estimated 120 m in the Peterborough quadrangle. Although tectonic thinning cannot be ruled out as a factor, the greatly reduced thickness in the Monadnock quadrangle may indicate a deposition site much closer to the source area (Figure 2).

Age and Correlation

The Perry Mountain Formation in Maine, as revised by Osberg et al. (1968), consists of cyclically interbedded white quartzite and light­gray micaceous metamorphosed shale. Cross laminations and convolute laminations are common in the quartzite. No fossils have been found in the Perry Mountain, but it lies between units believed to be Silurian (Figure 4). The Sangerville Formation is at least partially correlative and contains late Llandovery to middle Wenlock or early

~ Ludlow graptolites (Pankiwskyj et al., 1976).

Correlative rocks elsewhere in New Hampshire include the Roundtop Quartzite (Englund, 1976), the upper part of the Crotched Mountain Formation in the Hillsboro quadrangle (Nielson, 1981) and, according to E. Duke's 1984 remapping of the Peterborough quadrangle, part of the Crotched Mountain Member of the Littleton (Greene, 1970). The rocks on Crotched Mountain itself are Rangeley Formation. The Perry Mountain in the Peterborough quadrangle is not only thicker than at Monadnock, but is also more feldspathic and contains minor calc­silicate pods toward the base. The formation is also present locally along the west edge of the Ashuelot pluton (David Elbert, pers. comm., 1984). No rocks lithically similar to the Perry Mountain have yet been recognized in Massachusetts, or in the Gilsum-Marlow area.

Derivation

The Perry Mountain Formation in the type area is conformable and gradational with the underlying Rangeley Formation (Boone et al., 1970), and also thickens eastwards, indicating a similar depositional setting. The thin bedding and clean quartzites suggest a maturing of the erosional cycle in the source area, with more extensive chemical weathering. If the Clough Quartzite and the Perry Mountain are roughly correlative, the change from Rangeley to Perry Mountain may represent the point at which rocks of the former volcanic arc became eroded down enough to allow sediments from the more distant Taconian mountains to reach the Merrimack basin. The Clough, Perry Mountain, and Sangerville might represent respectively shelf, slope and rise deposits.

FRANCESTOWN FORMATION

Description and Distribution of Rock Types

35

Extremely rusty-weathering, blocky calc-silicate granulite and rusty-weathering graphitic schist make this the most distinctive unit in the Monadnock stratigraphy. Graphite and sulfides together compose up to 10% of the rock. The best place to observe the Francestown Formation is at Monadnock State Park, in the campground and south of Poole Reservoir. It is less well exposed along the northwest side of the Monadnock syncline due to glacial cover, and it crops out discon­tinuously around the syncline south of Troy. Although graded beds are rare, the topping direction of the sequence is well established, so that where the Francestown is in association with the overlying Warner Formation, as it is around Dublin Pond and Howe Reservoir, the topping direction is certain. Both the upper and lower contacts are sharp. Fine graded laminae can be seen in some thin sections. The formation is well exposed along the Old Chesham Road ridge in Marlboro, at Derby Hill west of Silver Lake, and west of Willard Hill in the southeast corner of Roxbury. At Derby Hill and in several infolds of Francestown surrounded by Rangeley east of Thorndike Pond, schist predominates over calc-silicate granulite, whereas the reverse is true elsewhere in the quadrangle. There are some interesting cemented till outcrops along Ainsworth Brook, west of Poole Reservoir, developed from pieces of Francestown held together by iron sulfides.

The calc-silicate granulites are composed of quartz, calcic plagioclase, graphite and iron sulfide minerals with or without sphene, actinolite, diopside, zoisite, microcline, minor muscovite, Mg-biotite, and Mg-chlorite, and accessory apatite, zircon, tourma­line, and allanite (Table 4). The iron sulfides include pyrite, pyrrhotite and secondary marcasite apparently replacing pyrrhotite. Many outcrops contain enough pyrrhotite to deflect a compass needle. Weathering of the sulfides results in red, orange, yellow, brown and black outcrops. The rocks are hard and well jointed, breaking into brick-sized fragments. The unweathered calc-silicates may be massive

.... Table 4. Estimated modes of the Francestown Formation.

Calc-silicate Granulite Schist

MK MK MK MK HK DB TR TR MK W< RX 502 414 502A 701B 726 78 50 127 11.L 351 _R_

Quartz 16 45 10 32 38 45 20 41 35 49 36 Plagioclase 26 40 70 19 24 25 18 25 2 32

(An66+)(An73)(An66+)(An71)(An62)(An70+)(An61)(An63+) (An25?) (An67)

Micro cline 16

Muscovite 3 43 32 23

Hg-rich Biotite 9 tr 18 tr tr 5 5 Chlorite 2 12 2 tr

(Retrograde) (Mg) (Mg-Fe) (Mg) (Fe)

Actinolite 47 7 35 27 3 17 2

Ferrian Zoisite 7* 6

Diopside 14 20

Sphene 4 2 10 3 2 2 2 --

Rutile

I 2 1

Opaques 7 8 3 5 8 7 5 3 8 10 3 Ilmenite X X Graphite X X X X X X X X X X X Pyrrhotite X X X X X X Pyrite/Marcasite X X X X X X Undiff. other X X X X X X X

Apatite X X X

Zircon X X X tr X

Tourmaline X

Allanite X I X w 0'1

*uncertain identification

List of Samples in Table 4.

MK-414 Very rusty fine-grained dark gray calc-silicate granulite, strongly magnetic, delicately laminated: some layers richer in muscovite, others in quartz, others in pyrrhotite, one layer up to 5% sphene.

Ainsworth Brook, elev. 433m, S of Parker Trail, Mt. Monadnock.

37

MK-502 Rusty fine-grained dark gray calc-silicate granulite, strongly magnetic. Actinolite and sulfides visible in hand sample. MK-502A is a lmm sphene-rich layer.

Meade Brook below Poole Reservoir, elev. 390m, at contact with Warner Formation, Monadnock State Park.

}~-701B Rusty fine-grained gray calc-silicate granulite. Not magnetic. 6lm NW of Stony Brook Farm Road, 880m SW of Thorndike Pond, Jaffrey.

MK-726 Rusty fine-grained dark gray calc-silicate granulite, with network of pyrrhotite veinlets, strongly magnetic. From a pod in graphitic whiteschist. Have~ Pines State Park, on 382m hill, 1020m NNE of Jaffrey Center.

DB-78 Rusty fine-grained dark calc-silicate granulite in a small fold hinge. Biotite and graphite have moderately developed preferred orientation parallel to axial plane. Strongly magnetic.

250m N from Trowbridge house, on bridle path toward Hurricane Hill, W of SH corner of Dublin Pond, Dublin.

TR-50 Very rusty actinolite-bearing light gray calc-silicate granu­lite, not magnetic.

Brandy Brook, small falls 200m upstream from Ashuelot River, Troy.

TR-127 Rusty light gray to green calc-silicate granulite, not magnetic. Interlayered with graphitic whiteschist.

Brook south of High Street 750m WSW from Troy village green.

HK-125 !<' ine-grained rusty-weathering dark gray schist, not magnetic. Jeep road 1200m due E of Jock Page Hill, Jaffrey.

MK-351 Medium-grained deeply rusty-weathering granular schist. White micas and graphite prominent in unweathered rock. Weakly magnetic.

Field behind Adullam (old seminary) on Old Dublin Road, Jaffrey.

RX-97 Rusty-weathering medium-fine-grained white schist with graphite. Along old wood road 900m SW of Willard Hill, Roxbury.

38

or finely laminated, and dark gray to nearly white.

Schists of the Francestown Formation weather deeply to a rusty brown-orange-yellow rind surrounding a white interio~ flecked with graphite. Unweathered schist is hard to find, but it is dark gray. The schists contain quartz, muscovite, Mg-biotite, sillimanite, graphite, iron sulfides, and secondary Mg-chlorite, with or without plagioclase, rutile, sphene and zircon. Neither the schists nor the calc-silicates contain black mica, and this is one of the chief criteria for distinguishing isolated outcrops from otherwise similar rusty schists in the Rangeley Formation. The sulfur content is appar­ently so high that nearly all the iron was consumed during metamor­phism to form pyrrhotite, leaving little for the silicate minerals Tracy et al., 1976a; Robinson et al., 1982a). The sphene shows red-brown to pale yellow pleochroism in thin section.

Thickness

Maximum estimates are 137 m thick in Dole Brook at elevation 451 m, east of the old toll road on Mt. Monadnock; 53 m thick north of High Street, Troy; and 76 m thick south of Rt. 124, 3 km west of Jaf­frey Center. The thickness of the Francestown is greatly varied where the rocks are complexly folded, such as around Poole Reservoir.

The Francestown and correlative rocks, as with the other Silurian units, thicken eastwards away from the Bronson Hill anticlinorium (Figure 2). Field (1975) estimated 15m thickness for correlative rocks in the Ware quadrangle, Massachusetts. E. Duke (1984) estimated 10-50 min the Peterborough quadrangle, Englund (1974) 425 min central New Hampshire, Hatchet al. (1983) 30-300 min northeastern New Hampshire and 825 m at Smalls-Falls, Maine. In the Bronson Hill anticlinorium the Fitch Formation, believed to correlate with both the Francestown and Warner Formations, was estimated at 120-150 m thick in the Littleton area (Billings, 1937) and 0-250 m in the Orange, Massa­chusetts, area (Hatchet al., in press).

Age and Correlation

The Francestown was defined as a member of the Devonian Littleton Formation by Greene (1970), and although the units do not connect on the ground, he suggested a correlation with the Rusty Quartzite Member of the Littleton in the Monadnock quadrangle (Fowler-Billings, 1949a). Nielson (1974) changed the unit to formation status, and suggested a correlation with the Silurian Smalls Falls Formation of Maine, which may be a more distal facies with much more pelite and less calc-silicate granulite. At the type locality of the Smalls Falls Formation non-calcareous quartzite is interbedded with schist, but calc-silicate rocks occur elsewhere near the top of the formation (Hatchet al., 1983). Fossils are not known in either the Francestown or Smalls Falls Formations. Graptolites in the correlative Parkman Hill Formation, central Maine, are middle Wenlockian to early

39

Ludlovian (Pankiwskyj et al., 1976).

Field (1975) mapped a rusty calc-silicate horizon in central Massachusetts, west of the Coys Hill Granite, and identified this as identical to the Francestown of the Monadnock area, and to rusty calc-silicate rocks at Gee Mill in the Sunapee septum where Dean (1977) mapped them as Fitch Formation. The regional importance of the Gee Mill occurrence will be addressed later.

White sulfidic schist (Spsq on the Massachusetts state map, Zen et al., 1983) is exposed about 3 km east of the sulfidic calc-silicate-­horizon, along what is shown on the state map as the contact between Partridge and Littleton Formations and between Partridge and Paxton Formations (Field, 1975; Zen et al., 1983). Berry (1985) has extended this belt of Smalls Falls equivalent rocks into Connecticut. Tucker (1977) reported feldspathic and micaceous quartzites associated with the white schist. The Francestown east of the Thorndike Pond fault zone and at Derby Hill, as well as that in the Pack Monadnock Range in the Peterborough quadrangle (E. Duke, 1984), has a schist to granulite ratio more like that of the white schist in Massachusetts and the Smalls Falls Formation in Maine, than that of the typical Francestown. Assuming that the Smalls Falls is a more distal facies originally deposited east of the Francestown, then the Smalls Falls-like rocks in the Monadnock quadrangle may have been tectonically transported westward to their present position. I think that the name "Francestown Formation" should be retained for the Smalls Falls correlative rocks with a high proportion of calc-silicate granulites, to emphasize the facies differences.

Boone (1973) reported a 30-50 m transition from Perry Mountain to Smalls Falls Formations in the Little Bigelow Mountain area, Maine. The contact in the Monadnock quadrangle is abrupt, and the Perry Mountain is locally missing. Whether the absence is due to local erosion or non-deposition is unknown, but it is most apparent below the Smalls Falls-like rocks east of the Thorndike Pond fault zone. A po~sibility exists that the rusty white schists in this area are stratigraphically lower than the Francestown, and occur as one or more horizons within the Rangeley Formation. The main argument against this interpretation is that lithically identical rocks occur above very well bedded Perry Mountain Formation at Derby Hill. Furthermore, Hatchet al. (1983) reported that the Perry Mountain is locally missing below Francestown calc-silicate rocks in the Pinkham Notch area.

Derivation

The strongly sulfidic and graphitic Francestown Formation was deposited as calcareous mud in an environment with restricted oxygen supply. The apparent lack of fossil remains, and the presence of fine laminae which would have been destroyed by bioturbation, might indicate truly anoxic conditions (Williams and Rickards, 1984). The

40

Merrimack trough may have become cut off from circulation with the open ocean. Alternatively, anoxic conditions may have resulted from a local poorly oxidized mass of water, as described by Williams and Rickards (1984) in modern open oceans. The water depth had to be less than the carbonate compensation depth, perhaps analogous to the modern Black Sea, where thinly laminated carbonate-rich and organic-rich muds are being deposited (Reineck and Singh, 1980, p.487-500).

WARNER FORMATION

Description and Distribution of Rock Types

The Warner Formation is exposed in all the areas described for the Francestown. The most accessible exposure~ are in Monadnock State Park northeast of Gilson Pond and downstream from Poole Reservoir. Although they have not been mapped separately, two parts can be distinguished.

The lower part of the Warner Formation consists of thinly bedded (0.5 - 5 em) green, pink, gray or white calc-silicate granulites. The color variation is due to differences in the relative proportions of actinolite, diopside, garnet, clinozoisite, ferrian zoisite, sphene, biotite and calcite (Table 5a). Quartz is present in all rocks, and zircon, tourmaline, allanite and opaques are accessory. A pitted pink horizon (sample MB-8A) contains calcite, idocrase, and an unknown, colorless, prismatic mineral associated closely with garnet. This unknown mineral is mainly in the cores of garnet, but also in the matrix. The garnet rims are free of this mineral. Rare schist beds with graded bedding occur in the lower part of the Warner, for example HV-7 (Table 5a), and near the base of the formation west of Poole Reservoir.

In the field the lower part of the Warner forms distinctive slabby outcrops, or smooth water-worn surfaces in brooks. Several brooks follow this more easily eroded unit, particularly Meade Brook south of

~ Poole Reservoir, the brook draining Gilson Pond, Minnewawa Brook in Eliza Adams Gorge and Gleason Brook northwest of Mt. Monadnock.

The upper part of the Warner Formation consists of fine-grained quartz-biotite-plagioclase granulite, with minor muscovite and sphene, and accessory zircon, garnet, apatite, rutile, tourmaline and opaques (Table 5b). Sphene is pleochroic red to yellow in both the granulites and the calc-silicates. The plagioclase is andesine. Outcrops are thickly bedded, massive, with smooth purplish-gray "salt and pepper" surfaces. These granulites superficially resemble the fine-grained tonalites and microdiorite dikes found in the quadrangle, so care must be taken to distinguish them. They are also similar to the granulite beds in the Rangeley Formation. Calc-silicate pods with mineralogy similar to the lower part of the Warner are common in the upper part. The pods are generally zoned with a weathered-out depression around a very resistant core. Nelson (1975) reported wollastonite from one of

.... Table Sa. Estimated modes of the lower part of the Warner Formation.

HK HK MB NH/MND HV 04D 004 8 7-74 7

green white gray pink pink white white pink green pink pink green white (schist)

Quartz 12 66 50 73 45 83 73 34 41 40 70 20 49 39

Plagioclase 15 5 tr 31 15 39 (An70+) (An35) (An66+)(An66+)(An53)

Micro cline 16 20

Clinozoisite 30 8 15 7 10

Ferrian Zoisite 12 10 27 5 X 1

Garnet 15 36 4 42 25 20 3 2

Diopside 15 20 4 4 6 5 6 27 20 10 25 1

Actinolite 49 25

Sphene 3 2 tr 2 5 10 1 2

Calcite 3 5

Idocrase? tr

Unknown mineral 15

Biotite 5 1 14

Muscovite 3

Opaques X X X X X X 2 2 (mainly graphite)

Apatite . X

Zircon X X X

Tourmaline X

Allanite X X ~ ......

... Table Sb. Estimated modes of the upper part of the Warner Formation.

Granulite and Schist Calc-silicate Pods

MK DB DB MK MK MK NH/MND 8-74-1 MK-759B 04B 8 9 638 629 759A core trans. rim core rim

Quartz 40 68 25 41 46 51 Quartz 30 X 44 60 45

Plagioclase 30 6 25 7 16 20 Plagioclase tr X 24 18 (An39)(An40)(An37)(An75)(An55)(An76) (An64) (An76)

Muscovite 3 2 20 7 Clinozoisite X 4 29 12

Biotite 25 21 10 28 19 28 Zoisite 5 X

Cordierite 5 Diopside 20 X 4

Garnet X 10 5 8 Grossular 8 X 2 tr 5

Sillimanite 6 4 Actinolite 21 4 18

Staurolite (Retrograde) 2 Calcite 18 X 2

Chlorite (Retrograde) lO(Mg) Bustamite 18 X

Sphene 2 2 X Sphene 1 X 3 3 2

Opaques tr X X 1 1 Opaques X X X X X

Graphite X X X 2 X X Graphite X X X X X

Ilmenite 1 X X Pyrrhotite X Pyrite tr

Apatite X X X X X Apatite X

Zircon X X X X X X Zircon X X

Tourmaline X X X Biotite " A

Rutile X Allanite tr

,c.. N

List of Samples in Table 5.

MK-04D Fine-grained greenish-gray granulite with 2 mrn actinolite. W of old toll road, Mt. Monadnock, elev. 497rn, Jaffrey.

MK-004 Fine-grained banded calc-silicate granulite with compositional layering on a scale of 1 to 10 rnrn. Slabby-weathering outcrop.

W of old toll road, Mt. Monadnock, elev. 497rn, Jaffrey.

43

MB-8 Fine-grained bedded calc-silicate, with pink, pitted, dark brown­weathering horizon.

South bank, Gleason Brook, 262m W of Dublin-Marlboro to~ line.

NH/HHD-7-74 Well bedded calc-silicate collected by Carl Nelson. Heade Brook, below Poole Reservoir, Jaffrey.

HV-7 Medium-grained gray feldspathic garnet schist, cut by mafic dike. Eliza Adams Gorge, N bank, 160m downstream from Howe Reservoir Darn,

Harrisville.

HR-04B Fine-grained purplish-gray granulite, massive in outc!"op but with foliation defined by biotite. ,

W of old toll road, ~1t. Monadnock, elev. 504rn, Jaffrey.

DB-8 Fine-grained light-gray granulite. Foliation defined by biotite. iJE of Dublin Pond, at junction of Rt. 101 and Old Harrisville Road.

DD-9 Fine-grained gray feldspathic schist. NE of DB-8 on steep south-facing slope, Dublin.

~rrC-638 Medium-grained gray biotitic schist with 1 ern sillimanite patches and smaller garnets. Interbedded with granulite.

t

Ledges on W bank of Stony Brook, elev. 402rn, 875rn SH of Gilson Pond, Jaffrey.

}~-629 Medium-grained gray granular schist with 0.5-1 ern garnet porphyro­blasts, which weather as bumps on the outcrop.

91rn SW of SW corner of Poole Reservoir, Monadnock State Park, Jaffrey.

l~C-759A Fine-grained biotite-quartz-feldspar granulite. Birchtoft Trail, elev. 414rn, H of Gilson Pond, Jaffrey.

NH/}fi~D-8-74 Zoned calc-silicate pod collected by Carl Nelson. Area east of outlet to Gilson Pond, Jaffrey.

l~C-759B Calc-silicate pod enclosed by granulite sample MK-759A. The thin section includes the transition zone between the two, which is listed in the table as the 11 rirn", and which weathers more readily than the core.

44

these pods, which is actually bustamite (see section on metamorphism). A comparison of estimated modes from Rangeley and Warner calc-silicate pods shows somewhat higher plagioclase content in the Rangeley pods, but otherwise a similar range of mineral assemblages.

Schist is interbedded with the granulite, the proportion of schist increasing toward the top of the formation. The topping direction of the sequence is confirmed by graded beds at several localities. The contact with the overlying Littleton Formation is gradational; calc-silicate pods and granulite beds persist into the lower part of the Littleton. For mapping purposes, the uppermost continuous granulite bed is taken as the upper contact of the Warner. Schists within the Warner and lowermost Littleton are more feldspathic than those higher in the Littleton. Horizons bearing 0.5 to 2 em garnets are common, and the garnets result in characteristic bumpy-weathered surfaces. The cordierite-bearing sample MK-629, described in detail in the section on metamorphism, came from one of these feldspathic "big-garnet schists".

Thickness

A section was measured west of the old toll road on Mt. Monadnock between elevations 490 and 520 m. Allowing for a three-meter granite sill, the lower part is 45 m thick, and the upper part is 37 m thick. In Dole Brook east of the toll road, a total thickness of 90 m was estimated for the Warner, but the upper contact is poorly exposed. North of High Street in Troy the Warner is at least 30 but not more than 68 m thick. This is very thin compared to maximum estimates of 450 min central New Hampshire (Hatchet al., 1983). In the Peterborough quadrangle the Warner thins southeastward and is locally absent (E. Duke, 1984).

Age and Correlation

Nielson (1974) established the Warner Formation as calc-silicate ~ granulite and biotite schist stratigraphically above the Francestown

in the Mt. Kearsarge and Hillsboro quadrangles. Neither the Warner nor the correlative Madrid Formation of Maine has yielded fossils, but Hatchet al. (1983) present a strong argument for correlation with the Fitch Formation to the west, which bears conodonts of Pridolian age (Harris et al., 1983). The Fitch is overlain by type Littleton Forma­tion, and both the Warner and Madrid are overlain by Littleton-like rocks. The age of the Warner thus hinges on the correctness of the Littleton correlations (Figure 4).

The upper part of the Warner in central New Hampshire includes some pervasively rusty-weathering sillimanite-biotite schist with calc-silicate pods, which John Lyons informally calls the Andover Member (pers. comm., 1982). Warner schists in the Monadnock quad­rangle are generally gray-weathering.

45

The Warner probably correlates with part of the Paxton Formation in central Massachusetts (Robinson, 1981; Hatchet al., 1983). Part of the Paxton lies above the Smalls Falls rock type8; but other gray granulites mapped as Paxton apparently lie lower in the section (Henry Berry, pers. comm., 1984), and may be equivalent in part to the Vassalboro Formation of Maine, which was formerly thought to be an eastern facies of the Madrid (Osberg et al., 1968; Osberg, 1980).

Derivation

The lower part of the Warner Formation was deposited as calcareous mud similar to the Francestown, but lacking the graphite and sulfides. Fossils in the Fitch reflect a hi gh-energy, near shore environment (Harris et al., 1983), and the Warner is presumably a deeper water facies of the Fitch. The transition to the upper part of the Warner might be attributable to an influx of detritus, perhaps in the form of volcanic ash (McKerrow and Ziegler, 1972). Volcanism was going on during the upper Silurian along the present-day Maine coast, and might have provided a source of ash (Brookins et al., 1973). In north­central Maine, volcanics are interlayered with clastic rocks and minor limestone on the northwest flank of the Weeksboro-Lunksoos Lake anticline (Neuman and Rankin, 1980, p.92) . The Seboomook Formation overlies these volcanics in a relationship similar to that of the Littleton overlying the Warner and Madrid Formations, with a gradually increasing proportion of pelitic sediments.

LITTLETON FORMATION

Description and Distribution of Rock Types

The Littleton Formation in the Monadnock quadrangle consists pre­dominantly of gray-weathering pelitic schist and micaceous quartzite.

The schists consist of quartz and muscovite with or without biotite, staurolite, garnet, sillimanite, plagioclase, K-feldspar, graphite and other opaques, retrograde chlorite, chloritoid and staurolite, and accessory tourmaline, zircon and apatite (Table 6). Pseudomorphs of sillimanite after andalusite are common in some parts of the quadrangle, depending on the metamorphic history, but no relict andalusite was found in thin section. Perry (1904) observed that some pseudomorphs preserve chiastolite cross-shaped inclusion patterns. He also noted that pseudomorphs which have been further replaced by muscovite tend to weather out as pits in the rock, in contrast to the raised lumps where sillimanite remains. Robinson (in Hatchet al., 1983) has dubbed similar pseudomorphs in Massachusetts "andalumps".

Table 6a. Estimated modes of the lower part of the Littleton Formation.

Zone II Zone III Zone III (Retrograded)

sz MK MK MB TR TR RX MK HV MK DB DB 27 432 604 172 11 20 59 371 162 216 10 69

Quartz 41 17 32 41 22 39 36 31 25 15 22 40

Plagioclase 6 1 4 3 2 2 (An31)

K-feldspar

Cordierite? X

Biotite 16 24 23 15 25 21 Zl 10 11 tr

Muscovite 21 27 13 28 8 28 13 30 37 54 41 39

Garnet 3 16 5 2 3 10 7 8 4 7

Sillimanite tr 13 24 9 37 7 17 19 12 1 tr

Chlorite 1 . tr 2 17 10 12 (Retrograde) (Mg) (Fe) (Fe-Mg)(Fe) (Fe) (Fe)

Chloritoid 9 19

Staurolite 10 tr

Opaques 2 2 2 3* 3 1 2 2 2 Graphite 1 X X X X 2 X X X X 3 X Ilmenite 1 X X X X X Hematite (weath.) X X Undifferent'd X X X X X X X

Apatite X X X X X

Zircon X X X X X X X X X X

Tourmaline X 2 X tr 2 tr X

Allanite X

Sphene X *includes pyrite inclusions in andalump

DB 171

24

tr

60

11 (Fe)

2 2

1

Zone IV

RX 172

15

2

18

25

1

8

28

1 X

X

X

~ 0'1

List of Samples in Table 6a.

SZ-27 Medium-grained gray to slightly rusty-weathering feldspathic schist. Staurolite visible with a hand lens. Slightly retrograded.

Small outcrop E of roadcut N side Rt. 12, 200m E of junction with Mill Road, East Swanzey.

MK-432 Well foliated gray garnet schist. Outcrop contains granulite bed and contact with Warner Formation.

Meade Brook, 27m S from Poole Reservoir dam, Monadnock State Park.

MK-604 Medium-coarse gray schist with sillimanite patches and garnets up to 1 em. 68m upstream from nearest Warner outcrop.

Meade Brook, 500m upstream from Poole Reservoir, elev. 457m, Jaffrey.

MB-172 Medium-grained gray schist with up to 3 em fibrolite bundles and 2 mm garnets. Very rough-textured outcrop.

Glen Brook, elev. 309m, in small gorge, Marlboro.

TR-11 Medium-grained massive gray schist with up to 4 em sillimanite pseudomorphs after andalusite with well preserved chiastolite structure,and 2 nun garnets. Very rough "boot-grabbing" outcrop.

NW ridge of Gap Mountain, SW of little gap at 488m, Troy.

TR-20 Coarse gray micaceous schist with strong foliation cut by crenulation cleavage.

Sewer excavation, Monadnock St ., 40m N of Rt. 12, Troy.

~X-59 Medium-coarse gray schist with abundant 1 em sillimanite bundles. E of jeep road 200m E of Woodward Pond,350m N of jeep road junction,

Roxbury.

MK-371 Well foliated fine-grained gray schist with sillimanite clumps and garnets up to 5nun. Slightly retrograded.

180m S of Rt. 124, 360m E of Cummings Pond, Jaffrey.

HV-162 Hedium-grained gray to slightly rusty schist with up to 3 em sillimanite and 4 mm garnet. Rough outcrop. Slightly retrograded.

47

Small knob, elev. 408m, Nl5E 400m from 376m hill, BOOm E of Chesham Pond.

tOC-216 Well foliated gray muscovite schist with 5 mm green chlorite t patches and unaltered garnet. Strongly retrograded.

Bald Rock, elev. 798m, SW ridge of Mt. Monadnock, Jaffrey.

DB-10 Fine- to medium-grained gray-green schist with chloritoid and elongate quartz-muscovite streaks. Strongly retro~raded.

Beech Hill, elev. 57lm, southernmost of three knobs

1 em knobs of Locally rusty.

on ridge, Dublin.

DB-69 Massive, medium-grained gray schist. Strongly retrograded. What appears to be sillimanite in hand sample has been replaced.

W of Rt. 137, 200m SW of sharp bend in Windmill Hill Road, Dublin.

DB-171 Somewhat rusty-weathering medium-grained gray schist. Retrograded. Hountain Brook, N side of Mt. Monadnock, elev. 436m at second old

dam site upstream from farm, Dublin.

~X-172 Hedium-grained gray schist with 5 mm sillimanite and abundant 3 mm garnet.

64m N of jeep road, 700m E of Woodward Pond dam, elev. 453m.

..• Table 6b. Estimated modes of the upper part of the Littleton Formation.

Quartzite Schist Retrograded Schist Coticule Misc.

MK DB HV MK MK DB DB DB DB MK HV DB DB DB DB 873 255 135 210 210A 114 196 253 17 6 135A 196A 55 220 236

Quartz 65 82 35 43 10 25 47 35 24 48 61 45 22 58 37 Plagioclase tr 10 tr 4 1 tr 10 39

(An43) (An24) (An25) (An14)(An64+)

Biotite 12 3 17 11 3 10 12 2 tr 3 7 X 3 14

Muscovite 15 5 15 35 55 35 33 48 44 5 4 16

Garnet 1 4 7 tr 5 tr 3 12 43 30 50 62 tr 7

Sillimanite 2 13 tr 7 13 X

Hornblende 9

Chlorite 1 . 3 5 2 5 .5 9 17 2 5 tr 5 (Retrograde) (Fe) (Mg) (Fe) (Fe) (Fe) (Mg)(Fe-Mg) (Fe) (Mg) (Mg-Fe)(Fe) (Mg)

Chloritoid (Retr.) tr 3 9 6

Staurolite (Retr.) tr 1 6

Opaques 1 2 1 1 3 2 3 1* 2 X 2* 2 Graphite X 2 X X X X X 2 Ilmenite X X 4 Undifferent'd X X X X X X X X X

Apatite X X X X 3 X

Zircon X X X X X X X X X X 2 X

Tourmaline 2 1 3 tr 2 tr tr

Allanite X

Sphene X X

*Includes chalcopyrite and pyrrhotite inclusions in garnet ~ 00

49

List of Samples in Table 6b.

MK-873 Light gray micaceous quartzite with poorly defined laminations. Interbedded with schist in outcrop.

S of Pumpelly Trail, E of Red Spot Trail, elev. 869m, }{t. Monadnock.

DB-255 Medium-fine-grained gray quartzite from the base of a 60 em graded bed.

Ledges N of Col. Sewell house, S end Beech Hill, elev. 520m, Dublin.

HV-135 Medium-grained gray schist with 5 mm coticule bed (HV-135A). Float block, NW end of Beech Hill ridge, elev. 455m, SSW of Harrisville.

MK-210 Medium-grained gray schist with biotites across the foliation, unaltered garnet, pitted weathering (chlorite-chloritoid), and retrograded "andalumps" (MK-210A) up to 8 em long.

Ledges S of White Cross Trail, elev. 786m, known as "The Caves", Mt. Monadnock.

DB-114 Fine-grained gray-green schist with 1-3 em-long sillimanite after andalusite and garnets up to 5 mm. Retrograded.

Snow Brook, elev. 468m, 2120m NW of Thorndike Pond, Dublin.

DB-196 Coarse-grained gray schist with up to 5 em altered sillimanite pseudomorphs after andalusite and an 8 mm coticule bed (DB-169A). Retrograded.

N side of Mt. Monadnock, elev. 533m, upper branch of Gleason Brook above Dublin Trail.

DB-253 Rhythmically bedded gray schist and quartzite with conspicuous 1 em chlorite patches and unaltered garnets up to 1 em. Retrograded.

Float block on steep S-facing slope 620m E of Old Chesham Road, S of Harrisville-Dublin town line.

DB-17 Rhythmically bedded gray schist and quartzite with 1.5-2 em graded beds. Sparse garnets up to 1 em. Retrograded.

320m SW from where Monument Road crosses Dublin-Harrisville town t line, E of Beech Hill, Dublin.

MK-6 Fine-grained pink granulite (coticule), 2.8 em-thick lens in schist. "Billings fold", elev. 945m, on cliffS of Smith Summit Trail, Mt.

Monadnock.

DB-55 Dense fine-grained coticule pod with pitted core, 4 X 8 X ? em. Float block, Pumpelly Ridge, elev. 747m, 600m NE from Cascade Link Trail.

DB-220 Fine-grained gray granulite with 1-5 mm quartz-muscovite nubbles. From a granulite bed several meters thick in schist.

NW end of boggy tarn, elev. 522m, Pumpelly Ridge, 380m S of Oak Hill.

DB-236 Fine-grained gray calc-silicate pod in gray schist, associated with granulite beds.

Pumpelly Ridge, elev. 510m, near head of small gully on SE side, 500m S of Oak Hill, Dublin.

The quartzites contain quartz, muscovite, biotite, garnet and opaques, with or without sillimanite, plagioclase, retrograde chlo­rite, and accessory tourmaline, zircon and apatite (MK-873 and DB-255).

The Littleton Formation occupies a narrow belt extending south from Troy village along the east side of Little Monadnock, and an irregular area centered on Bigelow Hill northeast of Troy (Plate 1 SW). A much broader belt extends from Mt. Monadnock northeast to beyond Rt. 101 in Dublin, and then wraps around Beech Hill to form a westerly striking belt which continues nearly to Marlboro village.

50

The Littleton is also found west and southwest of Derby Hill, in a very narrow belt extending south from Poole Reservoir, and in another belt extending southwest from Thorndike Pond. The belt of gray schist east of the Keene dome between the Clough Quartzite and Rangeley Formation is also assigned to the Littleton.

The proportion of quartzite to schist increases in strati­graphically higher parts of the formation, so that two parts can be informally distinguished. These may or may not correspond to the lower and upper parts of the Littleton of Hatchet al. (1983), and I have chosen not to divide the Littleton of the Monadnock area into formal members. An approximate contact between the two parts is shown on Plates 1 and 4.

The lower part of the Littleton consists mostly of very thick­bedded schist. Widely spaced, thin quartzite beds (less than 10 em thick) locally define bedding. Equally thin garnet-biotite-rich horizons which weather rusty brown are present in several locations. Pale pink, fine-grained garnet-quartz granulite lenses and layers up to 3 em thick are fairly common in both the lower and upper parts. These coticules, the rusty horizons, and quartzite beds are the main clues to original layering in what is otherwise rather monotonous gray-weathering, aluminous schist. Generally more than one quartzite bed is not observed in a single outcrop. However, in some areas of

~ good exposure, especially along the southwestern ridge of Mt. Monadnock, quartzite beds are from one to five meters apart. The presence of graded bedding is certain only where the thin sandy beds grade up into schist. The thick schists may have originally shown a gradual decrease in grain size upwards, but this is obscured by the metamorphic minerals. Andalumps are distributed throughout the schists, but in some cases seem especially abundant near the contact below quartzite beds. The graded laminae which Hatchet al. (1983) described as the hallmark of the lower part of the Littleton, are not prevalent in the lower part in the Monadnock area.

Feldspathic granulite and rare calc-silicate pods occur mainly in the basal 30 m of the formation. The main exception occurs about 575 m stratigraphically above the Warner contact in an overturned section at Oak Hill, on the north end of Pumpelly Ridge. The granulite there (Table 6b, DB-220) is quite similar to granulites in the upper part of

51

the Warner, and calc-silicate beds and pods (DB-236) are also present. These rocks lie approximately between the lower and upper parts of the Littleton, and may have regional significance (see "Age and Correlation", below).

The upper part of the Littleton Formation contains more abundant quartzite than the lower part, but is otherwise similar. It is best exposed on the summit of Mt. Monadnock, on Pumpelly Ridge, and around the village of Dublin. The increase in quartzite is gradual, so that the mapping out of an exact contact is difficult. The quartzite beds are progressively thicker and more abundant upward in the section. Thickness of quartzite beds ranges from a few centimeters to 40 em, and they are separated by anywhere from 5 to 100 em thick schist beds. The quartzite to schist ratio is generally less than one except in a distinctive set of seven quartzite beds, described in more detail below, where the ratio exceeds one. Many of the beds show excellent graded bedding. Some quartzites are in sharp contact with schist at both top and bottom, but this is less common than slow grading. What at first appears to be cross bedding in the quartzites is actually a faintly defined foliation, which can be traced into the schists. Weathered-out ellipsoidal chlorite and chloritoid leave pits that are locally parallel to this weak foliation.

Some structural features may have been produced penecontemporan­eously with deposition. High angle faults locally disturb bedding over a distance of less than a meter, with beds above and below undisturbed. These faults appear to have developed during compaction and pre-date the pervasive foliation. Along the White Arrow Trail on the south side of Mt. Monadnock, at an elevation of 829 m, there is a vertical wall with what appears to be large scale cross bedding (Figure 5a). It is more likely a fault related to sedimentary slumping. Folds at the top of the same wall are cut off by a knife­sharp surface, interpreted as a slump fault. Another fault of this sort is exposed at elevation 913 m east of the White Arrow Trail (Figure 5b). These faults are discrete planes without vein fillings or breccia. The bases of some of the quartzite beds have irregular lobes extending into the schist which might have been load casts. Other structures that might be interpreted as sand dikes and other dewatering structures (Shizuo Yoshida, pers. comm., 1984) have origins that are more open to debate. Tectonic deformation, metamorphism, and quartz veins greatly confuse the interpretation.

Coticule horizons are common in the upper part of the Littleton, and are especially well exposed on Mt. Monadnock. Some are associated with quartzite beds but others are completely enclosed in schist. In cross section they average one to two em thick, and they pinch and swell along bedding planes. At elevation 902 m the White Cross Trail crosses an overturned bedding plane with several oval coticule lenses on its surface. The lenses average 20 X 30 em and resemble concre­tionary masses. Coticules reported in the literature (Clifford, 1960; Fermor, 1909; Huntington, 1975; Karamata et al., 1970; Kramm, 1976;

•"'

~~

Sa.

Sb.

\ ... ,\0(\

\0 \U

~~~:. -·~

Fig. S. Soft-sediment slump faults in upright sections of the upper part of the Littleton Formation. Foliation cuts across these faults.

Sa. Station MK-2: vertical cliff along the White Arrow Trail at elevation 829 m. Inset shows fold near the leading edge of a second minor slump fault at the top of the cliff.

Sb. Station MK-4: East of the White Arrow Trail, elevation 913 m. Note that foliation post-dates the slump fault.

l.n N

53

Schiller and Taylor, 1965) show a wide range in garnet chemistry, but many are spessartine-rich (33.8 to 92 weight % spessartine molecule). Coticule from near the summit of Monadnock (Table 6b, MK-6) was studied with the electron microprobe. Garnet in the coticule contains 73.7% almandine, 13.4% spessartine, 9.2% pyrope, and 3-7% grossular.

A distinctive set of seven quartzite beds separated by schist beds, and having a total thickness of two to three meters, has been very useful in tracing out the pattern of tectonic folds on Mt. Monadnock. These light gray quartzites form regular stripes across the outcrop surfaces. A comparison of the thicknesses of the seven quartzite beds in four widely separated localities on the mountain (Figure 6) shows that the first, third, and sixth from the bottom are relatively thinner. The fourth locally grades "slowly" into the overlying schist, whereas the others tend to have sharp contacts at both top and bottom. It might be valuable to use this set of beds as a starting point for a detailed description of bedding characteristics in adjacent beds, and eventually develop an internal stratigraphy for the upper part of the Littleton Formation. The amount of exposure would certainly allow this, and a more complete understanding of the detailed stratigraphy would help refine the structural interpretation.

The stratigraphically highest rocks exposed in the upper part of the Littleton have rhythmically graded beds 5 to 10 em thick, with a quartzite to schist ratio of about one to two. They are best exposed along Pumpelly Ridge, but also occur on the hills south of Dublin village.

Thickness

The lower part of the Littleton Formation has as approximate thickness of 575 m, estimated in the overturned section northwest from Oak Hill toward Dublin Pond. On the upright section along Meade Brook, on the southeast slope of Monadnock, it is from 600 to 800 m thick, depending on how much section is repeated by isoclinal folds. T~ere is at least a comparable thickness of the upper part of the formation, but the top of the formation is not exposed. The highest rocks in the section lie on the northern slopes of the mountain, where a late syncline intersects the nappe-stage Monadnock syncline (see Figure 13).

Age and Correlation

The Littleton Formation was first described by Billings (1937) in northern New Hampshire to include black slate and gray sandstone with subordinate greenstone and soda-rhyolite volcanic conglomerate, and their higher grade metamorphic equivalents. Both the type Littleton (Billings and Cleaves, 1934; Boucot and Arndt, 1960) and the Seboomook Formation (Boucot, 1969) contain Lower Devonian fossils. The Little­ton of the Monadnock quadrangle is correlated with the type Littleton on the basis of similarities to the Seboomook in both rock type and

SEVEN QUARTZITES

80 11

quartzite

schist

80"J-2m

MK 27 MK 36 MK 859 MK8

Fig. 6. Comparison of measured sectio~s of a distinctive set of seven quartzite beds in the upper part of the Littleton Formation from four different locations on Mt. Monadnock, shown on Figure 16.

54

its stratigraphic position. Neither the Seboomook nor the Monadnock Littleton contains volcanics. Lyons (1979) recognized an upper unit with excellent graded bedding and called it the Kearsarge Member. This probably correlates with the upper part of the Littleton on Mt. Monadnock. Hatchet al. (1983) suggested correlating the lower and upper units in New Hampshire respectively with the Carrabassett and Seboomook Formations in Maine, which are separated by discontinuous marble, calc-silicate, and granulite beds of the Hildreths Formation (Moench, 1971). Malinconico (1982) found laminated calcareous quartzite in the Rumney quadrangle, in approximately the correct position to be correlative with the Hildreths, but this unit has not been recognized elsewhere in New Hampshire. The granulite and calc­silicate rocks at Oak Hill in the Monadnock quadrangle might also correlate with the Hildreths.

Derivation

55

The sedimentary precursors of the metamorphic rocks that make up the Littleton Formation consisted of mudstone and fine-grained quartzose sandstone. Hall et al. (1976) showed that the Seboomook Formation was deposited largely by turbidity currents in a base-of­slope environment, with a source area to the east. Together, the Seboomook and Matagamon Sandstone constitute an upward-coarsening, westerly-prograding "flysch basin-margin delta system" (Hall et al., 1976). Prodelta sloge facies grade upward into delta front facies. They estimated a N80 W down-slope direction from the orientation of flutes on the bases of sandstone beds, which is consistent with the orientation of fluxoturbidite channels, and with a slope direction estimated from the separation angle of slump fold axes plotted on an equal area diagram (Hall, 1973). No such study has been done on the Littleton Formation, but it has generally been assumed that the Littleton was also derived from the east (L. Hall and Robinson, 1982). However, the southeast-directed premetamorphic down-to-basin faults and folds documented by Moench (1970) involve both Siliurian and Devonian rocks. Therefore it seems likely that the Merrimack trough Peceived sediments from both east and west during the Early Devonian. A thorough examination of slump features on Mt. Monadnock might yield pertinent information.

Graded beds in the lower Littleton are separated by thick pelitic portions, suggesting that turbidites may be interbedded with "normal" pelagic sediments. Higher in the section, entire turbidite cycles as described by Bouma (1962) might be present, but the fine laminae and ripple cross-laminae of Bouma's (b) and (c) intervals have been obli­terated during metamorphism. The increased number and thickness of quartzite beds upwards suggests an increasingly proximal site of deposition through time.

It might be argued that the thick mass of the upper part of the Littleton localized at Mt. Monadnock, although tectonically thickened, represents a base-of-slope submarine fan of some sort. Mt. Kearsarge,

60 km to the north, might be another such fan. However, the bedding in the upper part of the Littleton is continuous over hundreds of meters and very few features have been found that might have been distributary channels. The local presence of upper Littleton is more likely due to the level of stratigraphy preserved in cores of nappe-stage synclines.

56

The origin of quartz-garnet granulite (coticule) lenses within the Littleton is uncertain. The fact that coticule occurs enclosed in schist as well as in quartzite makes a detrital garnet origin unlikely. The model preferred here involves the chemical precipi­tation of silica with iron and manganese oxides as Mn-rich chert nodules, followed by rapid burial and reduction. This model would be somewhat different from the Mn oxide nodule origin suggested by Clifford (1960) on the basis of the trace element assemblage in coticule from the Partridge Formation at Mill Hollow, New Hampshire. The Mn-oxide nodules on the modern ocean floor contain manganese as trivalent and tetravalent species. However, the reduction of Mn species in chert nodules may have taken place by a process similar to that reported by Burdige and Gieskes (1983). They proposed a diagen­etic origin for Mn oxide nodules based on phases observed in shallow marine drill holes. The interplay of Eh, pH, Mn2+ in pore fluids, sedimentation rates, and oxidatiop of organic matter results in a redox boundary which migrates gradually upwards, with the reduced species below the boundary. Rapid burial, such as by turbidites, might periodically bury a Mn2+ -rich zone too deeply for complete re-equilibration. It seems that Mn-rich chert nodules might be reduced in the same way.

Kramm (1976) argued for a volcaniclastic or~g~n of coticule in the Ardennes, Belgium, whereby Mn2+ would be present at the outset, incorporated in the structure of montmorillonite clays, and would stay in the divalent state through diagenesis and metamorphism. Coticules occur in a variety of geologic settings, the other important one being

! associations with metamorphosed volcanics (e.g. the Hawley Formation, Emerson, 1898). Analyses of trace element assemblages might shed some light on coticule genesis and help distinguish between various types.

INTRUSIVE ROCKS

INTRODUCTION

Fowler-Billings (1949a) distinguished three intrusive igneous rock units in the Monadnock quadrangle, each part of the New Hampshire plutonic series. In order of decreasing age they are the Kinsman "quartz monzonite", Spaulding "quartz diorite", and "Concord granite". According to the lUGS classification scheme (Streckeisen, 1973), the Kinsman is mostly granite and the Spaulding is mainly tonalite (Figure 7), so these rock names are used here. The name "Concord" was applied to granite plutons which may not all be the same age (Lyons, 1979), so the name "Fitzwilliam Granite" is substituted for the late two-mica

'"'

o Mfg

A. Dst

• Dkg

• Mmd

• •

Q

•

D

GRANITE

A.

c • (jl

c

• DIORITE, QTZ- SYENITE I QTZ- MONZONITE

KL----L. _ _ _ _ _j_ ____ _L __ ~-=--=s=~;;ABBRO

Fig. 7. Streckeisen (1973) plot of intrusive rocks, recalculated in terms of quartz, K-feldspar, and ~lagioclase, from modes in Tables 7-9. Open squares -Fitzwilliam Granite; open triangles - Spaulding Tonalite and related rocks; closed circles - Kinsman Granite. Some are based on modes published by Fowler­Billings (1949a), Duke (1984) and Shearer (1983).

Vl "'--

granites. Estimated modes presented in the tables include some from Fowler-Billings (1949a), E. Duke (1984) from the Peterborough quad­rangle, and Shearer's (1983) samples from the Monadnock quadrangle. Shearer's modes are the most reliable, because he stained his thin sections with sodium cobaltinitrite and counted 2000 points per section. Modes are plotted in Figure 7 in terms of quartz, plagio­clase, and alkali feldspar.

58

Inasmuch as the present study does not concern itself with the interiors of plutons, any changes from previous mapping involve either minor changes in the position of contacts or changes in the assignment of isolated bodies of rock to the three units.

Fowler-Billings (1949a) reported a "biotite schist dike" east of the summit of Mt. Monadnock at elevation 2940 ft. (896 m), which she believed was a metasedimentary dike similar to those she had described in the Mt. Washington area (Fowler-Billings, 1944a). I have located four other biotitic mafic dikes, and believe they are igneous micro­diorite dikes related to the Fitzwilliam Granite. A float block containing diabase was found north of Lake Skatutakee. The dikes, as well as pegmatites and tourmaline veins, are discussed in separate sections following the more important intrusive rocks.

KINSMAN GRANITE

The Kinsman is a coarse-grained peraluminous granite with micro­cline or plagioclase megacrysts commonly 2-5 em long. The groundmass consists of plagioclase, quartz, biotite, and muscovite, with or without garnet, sillimanite, and accessory sphene, zircon, apatite, allanite, graphite, ilmenite, and secondary chlorite and epidote (Table 7). The Kinsman crops out over most of the northeast corner of the quadrangle, where it is part of the Cardigan pluton (Fowler­Billings, 1949a). The ratio of megacrysts to groundmass is varied. A systematic study of the distribution of megacrysts within the Cardigan pluton might aid in our understanding of the Kinsman and its relation-

t ship with the Bethlehem Gneiss.

At least three belts of Kinsman Granite trend southwest from the Cardigan pluton toward the main belt of Coys Hill Granite in Massachu­setts. Outcrop control is terrible for the two wider belts, but the much narrower, westernmost belt, 5-15 m wide, is more convincingly continuous, lending credence to the continuity of the others as they are shown on Plate 1 SE. The westernmost belt is cut off by the Fitz­william pluton, the middle belt follows roughly the eastern edge of that pluton to join Kinsman outcrops at Sip Pond (see Fowler-Billings, 1949a, Plate 1) and the eastern belt projects to the north toward Kinsman outcrops in the Peterborough quadrangle (E. Duke, 1984) and to the south toward outcrops at Damon Reservoir in the Winchendon quad­rangle. The northernmost exposure of lithically identical Coys Hill Granite occurs near a railroad cut in the southwest corner of the Winchendon quadrangle, on strike 10.5 km south of the Sip Pond area.

59

In the cut there are two belts of Coys Hill Granite separated by Francestown and Littleton, which have been interpreted as lying in a fold hinge, partly on the basis of minor folds (Robinson, unpub. data; Zen et al., 1983). Farther south there is one nearly continuous belt of Coys Hill extending to within 10 km of the Connecticut border (Field, 1975; Tucker, 1977; Zen et al., 1983).

In the Monadnock quadrangle the Kinsman in the narrow belts is strongly foliated and the feldspar megacrysts have tapered, sheared outlines. Biotite, flattened quartz grains and the megacrysts show a strong preferred orientation parallel to foliation in the country rocks. Inclusions of schist lying parallel to the plane of foliation are common. A deeply weathered, coarse garnet-biotite horizon occurs within the Kinsman southwest of Gilmore Pond. It may represent a "restite" similar to one described by Clark (1972) in the Kinsman near Bradford, New Hampshire.

Mylonite occurs near the contacts of Kinsman in the narrow belts in several places, for example at the outlet of Mud Pond, Dublin (Table 7, DB-133); west of "The Ark", Jaffrey Center (MK-484); and east of Thorndike Pond (MK-1029). A mylonite covers the dip-slope surface of a large Kinsman outcrop behind a house east of Rt. 137, 2.5 km north of Jaffrey (MK-1010). The Kinsman outcrops west of The Ark, and strongly sheared rocks that were probably Kinsman east of Gilson Pond, lie west of the westernmost continuous belt. The age of the faulting that produced the mylonites is uncertain (see structure section), but they may serve as evidence that the three belts of Kinsman were once parts of a single body that was sliced by faults to produce the repeated map pattern. Alternatively, a single body might be repeated by early folds. There is no direct evidence to support folding, but this may be due to the poor exposure. Yet another inter­pretation would be that of parallel sills extending southwest from the Cardigan pluton, but this is not meant to imply any connotation of "feeder dikes".

The origin of the Kinsman Granite has long been a matter of debate. The current consensus favors an igneous plutonic origin, as well as an igneous origin for the garnets as phenocrysts (Clark, 1972; Lyons et al., 1973; Barreiro and Aleinikoff, 1985). Barreiro and Aleinikoff reported an age from Sm-Nd whole rock and garnet data at 413 ~ 5 m.y. Pb-Pb data on zircons indicate an inherited Proterozoic age which they suggested may reflect a Chain Lakes-type basement under southwestern New Hampshire. A Rb-Sr whole-rock age had previously been reported at 402 + 19 m.y. (Lyons and Livingston, 1977; revised, Lyons et al., 1982). -There is no conclusive evidence that the Kinsman intrudes Devonian rocks in the Monadnock quadrangle, but it apparently does intrude bona fide Littleton Formation in the Rumney quadrangle (Malinconico,-,g82-)-.--There is enough uncertainty in the absolute age for the base of the Devonian, that 413 m.y. may still be Devonian (Lyons and Livingston, 1977). Furthermore, a Silurian age would imply that the Kinsman is pre-Acadian, which seems unlikely.

....

Table 7. Estimated modes of the Kinsman Granite. Modes for "K" samples from Fowler-Billings (1949a) and those for "D" samples from E. Duke (1984) in the Peterborough quadrangle.

Extension of Coys Hill pluton Cardigan pluton Haunted Lake Gilson Pond

HK K K K D D MK 199 70 111 113 72 18 1169

Quartz 48 10 14 20 25 24 36

Plagioclase 34 40 40 10 22 18 12 (An38) (Oligoclase-Andesine) (An15) (An31) (An38)

Micro cline X* 35 20 56 28 43 16

Biotite 14 10 15 5 7 5 12

Muscovite 2 4 X 8 11 3 21

Garnet X 10 X 2

Sillimanite X

Cordierite X

Chlorite X 4 4 (Retrograde)

Epidote X X X _{Retrograde) Opaques 1 X X X 1 X

Ilmenite X X Graphite X X Undifferent'd X X X X X

Apatite X X X 0.5 X X

Zircon X X X X 0.5 X X

Sphene X X X X

Allanite X X

*small patches inside plagioclase megacrysts 0\ 0

61

List of Samples in Table 7.

MK-199 Coarse, strongly foliated, dark gray porphyritic "granite" with up to 2 em feldspar phenocrysts.

Southwest edge of swamp, 1100m NW of Hodge Pond, 900m SE of power lines, Jaffrey.

K-70 Porphyritic quartz monzonite; phenocrysts of potash feldspar 4 em long. Large outcrop with mylonitic zones.

E of Rt. 137, 2.7 km N of Jaffrey (=MK-1010).

K-111 Porphyritic "granite"; phenocrysts of potash feldspar 5 X 1.5 em make up 20% of rock; garnet grain size is 0.5 em.

On road ~E of Jacquith Brook, 300m SE of 338m crossroads, Hancock.

K-113 Porphyritic granite. Halfway between Moose Brook and Hosley Brook, 350m H of edge of

quadrangle on W side of knob, 396m altitude, Hancock.

D-72 Porphyritic granite. 1 km NW of Greenfield, Peterborough quadrangle.

D-18 Sheared porphyritic granite. Southeast of Haunted Lake, Francestown, Peterborough quadrangle.

MK-1169 Very fine-grained mylonite with 5 mm to 2 em feldspar porphyro­clasts with recrystallized "tails" and sparse 5 mm garnets.

350m W of Gilson Pond, elev. 428m, Jaffrey.

62

SPAULDING TONALITE AND RELATED ROCKS

Fowler-Billings (1949a) named this unit after Spaulding Hill west of Dublin. It includes a variety of rock types similar to those in the Hardwick pluton of Massachusetts, which reaches the Monadnock quadrangle southwest of Fitzwilliam. Other bodies of Spaulding may have been continuous with the Hardwick pluton prior to intrusion of the Fitzwilliam Granite. No effort has been made here to characterize mineralogically the various Spaulding bodies. Shearer (1983) described four rock types in the Hardwick pluton: biotite tonalite and hornblende-biotite tonalite in the interior, biotite-muscovite tonalite locally within one kilometer of the pluton's margins, and biotite-garnet tonalite still closer to the margins. These rock types as well as small bodies of strongly foliated granite, quartz gabbro, and gabbro have also been assigned to the Spaulding in the Monadnock quadrangle. Duke (1984) considered the Peterborough granite pluton, part of which extends into the Monadnock quadrangle northeast of Jaffrey, as related to the Spaulding. I have included it with the Fitzwilliam Granite.

The tonalites contain plagioclase (An 27 _49 ), quartz and biotite, with or without K-feldspar, muscovite, garnet, hornblende and secondary chlorite, with accessory sphene, zircon, apatite, ilmenite, allanite, calcite, anatase, tourmaline and clinozoisite (Table 8).

The plutons at Gap Mountain and south of Mt. Monadnock may once have been continuous with the Hardwick pluton. The Spaulding Hill pluton is separate, but a former connection through the area now occupied by granite north of Bigelow Hill cannot be ruled out. Outcrops of unusually garnet-rich tonalite (MB-64) which are between Littleton schist and two-mica granite north of Rt. 124, may be remnants of such a connection. There are large inclusions of strati­fied rocks on both summits of Gap Mountain, surrounded by tonalite. They are apparently more resistant to weathering than the tonalite.

Lyons and Livingston (1977; revised, Lyons et al., 1982) reported a late Early Devonian Rb-Sr whole-rock isochron age of 393 ~ 5 m.y. for Spaulding Tonalite in central New Hampshire. Garnet-bearing tonalite and hornblende gabbro-diorite samples from the same plutons have much lower initial Sr 87 /Sr 86 ratios and do not plot on the 393 m.y. isochron, suggesting to the authors a mixed mantle-crustal derivation for the unit as a whole. Shearer (1983) has written exten­sively on this topic.

Small elongate tonalite plutons and isolated outcrops are abundant parallel to the Kinsman belts southeast of Mt. Monadnock. Both units are strongly foliated and locally mylonitic, but west of The Ark (Plate 1 SE), at map scale, the tonalite appears to cut across a small body of Kinsman. A contact between Spaulding and Kinsman is well exposed in an outcrop under power lines which lead to the State Park,

63

but superimposed metamorphic foliation obscures any original intrusive relations at this scale (Figure 8).

-----

feldspar megacryst

Fig. 8. Spaulding (Dst)-Kinsrnan(Dkg) contact, MK-482. Contact is sharp except below hammer handle, where there is apparently an irregu­larity in the original contact. Mylonitic foliation is oriented N45°E, 580ffiv in both units. It is impossible to determine age relations at this outcrop.

The tonalite bodies southeast of Mt. Monadnock are texturally different from rocks in the larger tonalite plutons, and are informally referred to as the "Gilson Pond type". Foliation is much more strongly developed, feldspars are typically aligned and sheared, and quartz grains are ellipsoidal, producing a rock that resembles a small-scale version of textures in foliated Kinsman. The feldspar megacrysts are rarely larger than 2 em long, however, and plagioclase greatly exceeds K-feldspar in abundance. East of Gilson Pond, sheared examples of both the Spaulding and Kinsman are present, and they can easily be confused. The oblong pluton of tonalite west of Gilson Pond is less strongly foliated, but still contains the 1-2 em feldspar phenocrysts. This distinctive type of Spaulding also occurs in small plutons around Dublin Pond. There are numerous isolated bodies of more typical tonalite, too small to show on Plate 1, especially in the Rangeley Formation. South of Thorndike Pond the tonalite contains xenoliths of calc-silicate granulite and schist. Locally tonalite appears to grade into rocks of the Rangeley, as though it formed through in situ melting of the schist, and the calc-silicate pods were left intact.

... Table 8. Estimated modes of the Spaulding Tonalite and related rocks. Modes for "K" samples from Fowler-Billings (1949a) and those for "SM" samples from Shearer (1983).

- - pluton t~pe

Spaulding Gap Mtn. Hardwick "Gilson Pond" Gabb

DB MB SM TR MB K SM MK MK RD MK MK 162 29W 1-1 14 64 32 5 753 1139 5 195A 703

Quartz 44 28 19 24 18 10 21 37 34 15 X 14

Plagioclase 42 20 38 35 32 67 43 45 47 23 48 25 (An34)(An27)(An29) (An41) (An49) (01- . (An45 (An34) (An40) (An65)(An67)(An66)

And) -26) K-feldspar 27 2 X 5 X

Biotite 12 15 37 33 19 20 27 16 18 25 5 13

Muscovite 1 1 4 1 X X 4

Hornblende 29 38 35

Garnet 11

Sillimanite 5

Staurolite X

Chlorite X X 5

Opaques X 1 X X X 1 X 1 X 2 4 3 Ilmenite X X X X X X X X X Pyrite X tr X X X Undifferent' d X X X

Sphene 1 6 X 6 1 X 1 X 4 X

Anatase X

Apatite X 1 X 2 1 X X X X 2 X 1

Zircon X X X X X X X X X

Allanite X X X X X X X

Calcite X 1

Tourmaline X X

0\ +:-

List of Samples in Table 8.

DB-162 Strongly foliated, medium-coarse-grained, medium-gray biotite tonalite.

Roadcut S side of Rt. 101 near Howe Reservoir, 5 km W of Dublin.

MB-29W Medium-grained, weakly foliated, medium-gray granite, in contact with darker tonalite (SM1-1).

N side of Rt. 124 550m W of Old Dublin Road, Marlboro.

SM1-1 Coarse, weakly foliated, light-gray tonalite. E end of same roadcut as MB-29W.

TR-14 Medium-coarse, weakly foliated, medium-gray biotite tonalite. Elev. 442m on N side of Gap Mountain, along Metacornet-Honadnock

Trail, Troy.

MB-64 Atypical medium-coarse, poorly foliated, gray tonalite with abundant 3rnrn garnets and rniuor sillimanite and staurolite. Probably contaminated by incorporation of schist.

N of Rt. 124 300m E of Monadnock Drive, Marlboro.

K-32 Medium-grained dark gray quartz diorite. Quarry 650m E of Fitzwilliam Depot, S of railroad bed.

SM-5 Medium, weakly foliated gray tonalite. Quarry 200m S of railroad bed, 1.6 krn W of Fitzwilliam.

MK-753 Medium-grained rnylonitized gray biotite tonalite with 2-5 rnrn feldspar megacrysts.

65

75rn E of SE corner, Thorndike Pond, behind house E of road, Jaffrey.

MK-1139 Medium-grained rnylonitized gray to brownish biotite tonalite with 3 rnrn to 1 ern tapered feldspar megacrysts.

1.4 krn NNE of Jock Page Hill, 400m E of logging road junction, elev. 360m on E-facing slope.

RD-5 Fine-grained, dark gray biotite-hornblende quartz gabbro. Concordant sill.

Elev. 495rn toward Wend of E-W ridge, 1750m W of Little Monadnock ridge, Richmond.

MK-195A Coarse-grained, dark greenish-gray hornblende-biotite gabbro. Power lines 875m SW of Jaffrey Center, where they turn from SW to ESE.

MK-703 ~1ediurn-grained weakly foliated dark greenish-gray hornblende­biotite quartz gabbro.

N of stonewall, edge of beaver swamp, Stony Brook, upstream from Kinsman outcrops. 500m SW of Thorndike Pond, Jaffrey.

!

66

A sill of fine-grained hornblende-biotite quartz gabbro (RD-5) crops out on the east-west hill west of Little Monadnock. A similar sill at elevation 405 m on the east slope of Little Monadnock contains sphene which is conspicuous in hand sample. One outcrop of medium­grained hornblende quartz gabbro (MK-703) occurs near the east contact of the Kinsman belt south of Thorndike Pond. Several coarser-grained hornblende gabbro outcrops (MK-195A) also lie east of the Kinsman, south of Rt. 124.

FITZWILLIAM GRANITE

The youngest plutons in the quadrangle are peraluminous two-mica granite. They are massive to weakly foliated, and they cut across all folds and earlier intrusions. They are thus clearly post-tectonic. In detail the contacts are irregular, with many apophyses intruding the country rocks parallel to foliation. The most strongly foliated Fitzwilliam outcrops occur near silicified zones presumed to be Mesozoic.

There are five major plutons of Fitzwilliam Granite: the Troy Quarry pluton west of Mt. Monadnock, the Fitzwilliam pluton, the pluton south of Marlboro village, the Babbidge Reservoir pluton north of Marlboro, and an elongate pluton along the west edge of the quad­rangle in Keene. Most of these occur in areas of low topography where outcrop is generally poor and large rounded granite boulders are common. The granite was apparently deeply weathered prior to glaciation. The unit was extensively quarried in the nineteenth century, but no quarries are currently in operation, except for crushed stone production from waste rock in Marlboro. The weakly foliated granite northeast of Jaffrey (MK-1010, Table 9) may have Spaulding rather than Fitzwilliam affinities (E. Duke, 1984), but is shown as Fitzwilliam on Plate 1 SE. There are numerous smaller plutons, dikes and sills of two-mica granite, especially in the western and southern parts of the quadrangle, which are not shown on the plates.

Estimated modes are presented in Table 9, partly from previously published data. Shearer (1983) showed that the Fitzwilliam pluton is more heterogeneous than it would appear from hand samples alone, and includes granite, quartz syenite, and quartz monzonite. The granite consists of plagioclase (An 18_34 ) and microcline in roughly equal amounts, quartz, muscovite, and biotite, with secondary chlorite and accessory zircon, apatite, opaques, sphene and tourmaline.

Lyons and Livingston (1977; revised, Lyons et al., 1982) estimated a Mississippian Rb-Sr whole-rock isochron age for Concord granite in the Sunapee pluton of 326 + 3 m.y. Hayward (1983) estimated Rb-Sr whole-rock isochron ages of 349 m.y. for the Sunapee pluton and 383 m.y. for the Fitzwilliam pluton. Hayward's age determination for the Fitzwilliam seems too old. The Fitzwilliam cuts across dome-stage structural features, and yet the Prescott and Belchertown plutons of

Quartz

Table 9. Estimated modes of the Fitzwilliam Granite. Modes for "SM" samples from Shearer (1983).

M lb ar oro 1 ZW1 1am F't '11' area w f 0

pluton pluton Mt. Monadnock MB SM SM MK MK MK

_]_Q_ _3_ 11 828 1123 41B

42 31 18 31 26 37

Plagioclase 27 29 39 28 20 38

N of

(An28) (An28) (An30-22) (An34)(An34)(An22-26)

Microcline 21 31 36 29 35 18

Biotite 8 5 7 6 11 5

Muscovite 2 5 2 3 8 1

Opaques X X X 1 X X Ilmenite X X X X Pyrite X X Undifferent'd X X

Sphene 2 X X

Tourmaline X X

Chlorite X X X

67

Jaffrey

MK 1012

42

7 (An29)

35

3

12

X X

1

List of Samples in Table 9.

MB-30 Medium to fine-grained, non-foliated, light-gray granite. Harlboro Quarry, E of Rt. 124, 1. 75 km S of junction with Rt. 101.

SM-3 Coarse, non-foliated, light-gray granite. N side of Rt. 12, roadcut 1.8 km N of Fitzwilliam, across from

landfill.

SM-11 Coarse, non-foliated, light-gray "granite". S of Rt. 119, 1.3 km W of Fitzwilliam Depot.

MK-828 t1edium-coarse, weakly foliated, light-gray granite. Dike parallel to mafic dike, elev. 628m, Mossy Brook, Jaffrey.

MK-1123 Hedium-fine, light-gray granite. 465m knob, 125m W of Fassetts Brook, W of Mt. Monadnock, Jaffrey.

68

MK-41B Fine-grained brownish-weathering granite with xenoliths of gray schist. In association with black garnet-bearing mafic dike.

In gulch E of Marlboro Trail, elev. 704m, Mt. Monadnock, Jaffrey.

MK-1012 Medium-grained, weakly foliated, light-gray granite. 2.7 km N of Jaffrey, 750m E of Rt. 137, roadcut in housing development.

Massachusetts, dated respectively at 377 ~ 20 m.y. (Naylor, 1970) and 380 ~ 5 m.y. (Ashwal et al., 1979), are very strongly deformed by the dome stage of deformation.

MICRODIORITE DIKES

69

Five biotitic mafic dikes have been found. In the field they resemble lamprophyres, in the sense that they are mafic dikes con­taining only mafic phenocrysts. Mineralogically (Table 10) they are biotite-rich microdiorite, or perhaps kersantite, which is a lampro­phyre in which biotite and plagioclase together compose about 75% of the rock (Tr6ger, 1935). Some geochemical analyses would aid in understanding the unusual mineralogy of these dikes. Kersantites have about 3.7 wt. % K2o and 51.8% Si02 (Metais and Chayes, 1963).

Mineralogy

Estimated modes for the microdiorite dikes are presented in Table 10. Due to the very fine-grained texture of these rocks, the modes are not very accurate. The typical dike rocks are tan-weathering, dark gray to black, fine-grained, biotite-rich microdiorites, with 2-5 mm clots of biotite that give the weathered rock a spotted appearance. Sparse 2-15 mm garnets form knobs in ~ome portions of the dikes. In thin section the garnets have irregular, corroded outlines, and refractive indices for a garnet from sample MK-54A are greater than 1.76, suggesting almandine garnet. The garnets are probably xeno­crysts from the Littleton Formation. The dike groundmass consists of biotite and plagioclase, with or without green hornblende, K-feldspar, quartz, and accessory ilmenite, sphene, zircon, and allanite. Most samples show a foliation defined by biotite, although relict ophitic texture is also preserved (MK-54N). Some biotite clots surround horn­blende, and ilmenite is commonly rimmed by sphene.

Field Descriptions

One of the dikes (HV-7) is vertical and only 15 em wide. It trends N37° E, cutting across the Warner Formation in Eliza Adams Gorge (Plate 1 NW). The others are all on Mt. Monadnock (Figure 9). MK-69A is the "biotite schist dike" east of the summit described by Fowler-Billings (1949a, p.1271) as a metamorphosed sedimentary dike. This dike is different from the others in that it appears to have intruded along an irregular shear zone that offsets bedding in the Littleton. It is about 20 em thick, but pinches and swells and at one point forks into two branches. However, its mineralogy is similar to that of other dikes in the quadrangle and an igneous origin is likely.

The most prominent dike (MK-54) is about 1.8 m thick, nearly vertical, and trends N54° E to N33° E across the summit of Mt. Monadnock (Figure 9). It crosses the summit ridge on the Marlboro­Dublin Trail about 115m north of the summit, following a bush-filled lineament that shows up clearly on the air photo. The rock weathers

, .. Table 10. Estimated modes of microdiorite dikes.

,, MAIN DIKE contaminated"

,..... 1<: IZ 1<: I::S: I~ 10 lrz.. I~

~~ ~~ ~~ ~~ ~~ ~~ ~~ ~~ - -- -- -- -- -- -- -- --' 37 X 23

...., 8 42 30 26

I 38 52 43 )-35 54 20 33 17 (An45) (An46)(Anl4)(An26)

23 20

Biotite 41 20 22 30 57 20 7 10 33

Hornblende 20 K 2 13

Muscovite 5 3 18

Garnet xenocrysts X X X

Ilmenite 3 3 5 3 6 2 2 2 1

Sphene 3 X 2 4

Apatite X 1 X X X X 1 X X

Zircon X 1 1 1 X X X X X

Allanite X

Pis tacite

\()

I

~ -

1

I

(Al

27

17

X

5

1

X

X

X

...., 0

List of Specimens in Table 10.

HV-7 Very fine-grained black mafic dike, weathering brown with darker plates of biotite; vertically cross-cuts garnet schist.

Eliza Adams Gorge, N bank, 160m downstream from Howe Reservoir Dam, Harrisville.

MK-69A Fine-grained, foliated, dark gray mafic dike. White Cross Trail, elev. 893m, SE of Monadnock summit, Jaffrey.

MK-54N Dense, fine-grained black hornblende-bearing mafic dike, weathering brown with darker 2mm clumps of biotite and sparse 5-lOmm garnet xenocrysts.

North of Monadnock summit, main dike, elev. 875m, Jaffrey.

MK-54A Fine-grained black mafic dike, weathering brown with darker biotite clots and sparse garnet xenocrysts.

SW of Monadnock summit, main dike, elev. 928m, east of Smith Summit Trail in one meter wide notch, Jaffrey.

71

}0C-54W Very fine-grained black mafic dike with 2-5mm clots of biotite. SW of Monadnock summit, main dike, ' elev. 914m, Jaffrey.

l1K-54E Fine-grained medium dark gray mafic dike with 2-4mm clots of biotite and up to 5mm long Lathes of feldspar.

SW of Monte Rosa, in Mossy Brook, elev. 613m, Jaffrey.

MK-54D Fine-grained light gray granitic dike, weathering brownish-gray. W of Monte Roca, elev. 67lm, Jaffrey.

MK-54F Fine-grained light gray to brown granitic dike mottled by 3mm biotite clots and sparse 2cm muscovite xenocrysts (after andalumps?).

SW of Monte Rosa, S of Mossy Brook, near trail, elev. 640m, Jaffrey.

MK-41B Fine-grained brownish-weathering granite with xenoliths of gray schist which contain 2-3mm muscovites concentrated along contact. In association with black mafic dike with garnet xenocrysts.

In gulch E of Marlboro Trail, elev. 704m, Jaffrey. Mafic dike crosses the trail at the top of the gulch.

MK-1116 Fine-grained dark gray mafic dike with 2-3mm clots of biotite and up to 5mm feldspar lathes; in contact with black mafic dike.

Marian Trail, W side of Mt. Monadnock, elev. 65lm, just S of large W-facing open rocks, in woods, Jaffrey.

72

Fig. 9. Mafic dikes on Mt. Monadnock (shaded), showing sample locations. Dotted line represents the Seven Quartzite beds in the Littleton Forma­tion; random dashes represent Fitzwilliam Granite. The mafic dikes and granite are probably contemporaneous. Contour interval 30 meters.

to smooth rounded brownish-gray rubble and can be observed in place only here and there. It has been traced from elevation 838 m on the north side of the mountain to 570 m on the south side for a length of about 1.8 km. Mossy Brook follows the dike from elevation 642 m to 570 m (Figure 9). Locally there are two parallel mafic dikes and there is also a vertical mafic dike trending N70° W. The dikes clearly cut all phases of Acadian folds exposed on Mt. Monadnock.

73

Two other mafic dikes crop out along the west slope. One is a typical black fine-grained dike where it crosses the Marlboro Trail at elevation 704 m, but it gives way to granite (Table 9, MK-41B) in a gulch east of the trail, and both die out at elevation 680 m. Another dike, MK-1116, crosses the Marion Trail at elevation 651 m. It crops out over a width of about nine meters, and its extent along strike is not yet known. It was initially hoped that some of these dikes could be traced west to the granite pluton, but glacial cover is too extensive to permit this.

Contact Relations

Granite dikes also parallel the NE-trending mafic dikes, and, at elevation 604 m where Mossy Brook Trail descends south away from the brook, a contact between granite and . the main mafic dike (MK-54G) is exposed west of a small waterfall over andalump schist. The dike has an extremely fine-grained chill zone next to the granite. Downstream, at elevation 497 m, there is a large float block of granite with inclusions of mafic dike rock. The two rock types thus seem to be coeval. Upstream from MK-54G, at elevation 640 m, a granite sill intrudes the Littleton at a small saddle in the ridge east of the brook. Where this granite reaches the brook (Table 9, MK-828), the "mafic" dike is a fine-grained light brown dike with biotite clots which approaches granite in composition (Table 10, MK-54F). Down­stream, the dike is an intermediate dark gray rock (MK-54E). It appears that the mafic dike material was somehow contaminated by the granitic material where the two dikes cross. The Marion Trail dike

· also contains some portions of intermediate composition. In con­clusion, the mafic microdiorite dikes appear to be about the same age as the Fitzwilliam Granite (?Mississippian), and yet they were weakly metamorphosed, perhaps in the "Permian disturbance" (see concluding section in metamorphism chapter).

PEGMATITE

Pegmatite is common in the Monadnock quadrangle, both as foliated bodies concordant to regional foliation and as undeformed bodies. Foliated and sheared pegmatite in the Gilson Pond area, for example, is deformed by the backfold stage of deformation, whereas large masses of non-foliated pegmatite hold up the ridges of The Pinnacle and Bald Hill in Roxbury (Plate 1 NW). The pegmatites consist primarily of quartz, plagioclase, perthitic K-feldspar, and muscovite, with or without biotite, tourmaline, garnet and other accessory minerals.

Fine examples of plumose muscovite were found on The Pinnacle. There are distinctive tourmaline- and garnet-bearing aplite and pegmatite dikes on Mt. Monadnock which locally contain masses of fine-grained, pale green muscovite, which appear to be pseudomorphs after silli­manite.

74

Pegmatitic material is common as lenses and stringers in many of the schistose rocks, apparently produced in situ as melt pockets during metamorphism. Some of these may have coalesced to form larger masses, but other pegmatites probably were intruded from deeper sources associated with plutons. Fowler-Billings (1949a) showed the location of the major pegmatites in the Monadnock quadrangle by small red crosses on her map. They are especially common intruding the Rangeley Formation in the towns of Roxbury and Sullivan, but pegmatite also intrudes the Swanzey Gneiss at Mt. Huggins, and intrudes Little­ton, Warner and Francestown Formations in the Marlboro syncline 800 m due east of Marlboro village. Cameron et al. (1954), in their discus­sion of the "Keene pegmatite district", reported that pegmatites intrude Bethlehem Gneiss, Kinsman Granite, and "Concord" Granite, but they most commonly intrude the metamorphic rocks. These authors agreed with the conclusions of Fowler-Lunn and Kingsley (1937), that in the Grafton district each of the units in the New Hampshire plutonic series gave rise to pegmat~tes. Although they favored the "Concord" as a source for the larger pegmatite bodies in the Keene district, they did not draw any decisive conclusions.

TOURMALINE VEINS

Thin, very fine-grained, black tourmaline veins are common throughout the quadrangle. Cross-cutting relations and association with non-foliated pegmatites indicate a post-tectonic age. A block of microdiorite float was found at MK-54A in which the dike rock is cut by a tourmaline vein, on each side of which is a 2 em thick pegmatite layer. Tourmaline veins have not been noted cutting the granite.

! They are especially prominent on Mt. Monadnock, where tourmaline has replaced aluminous minerals in rocks adjacent to the veins. Silli­manite-muscovite pseudomorphs seem especially susceptible to this replacement. The source of boron may have been hydrothermal, or somehow concentrated from the schists themselves. A geochemical comparison of rocks far from tourmaline veins, to rocks progressively nearer the veins, might help clarify this. If the boron came from the schists it seems there ought to be a boron-depleted zone between the tourmalinized zone· and the schists farther away.

DIABASE DIKE

A block of float was found in a brook north of Lake Skatutakee, at elevation 380 m, 600 m east of Harrisville village, made up of diabase cutting Kinsman Granite. The diabase is dark, fine-grained, with sparse feldspar phenocrysts. It is probably Mesozoic.

"The rockwork is interesting and grand; --the clean cleavage, the wonderful slabs, the quartz dikes, the rock torrents in some parts ••• 11

-Ralph Waldo Emerson, 1866

STRUCTURAL GEOLOGY

INTRODUCTION

75

Five phases of Acadian deformation have affected the rocks of the Monadnock quadrangle. The earliest, isoclinal folds, are believed to be related to the huge west-verging nappes proposed by Thompson et al. (1968). West-verging ductile thrust faults then developed and cut-­across the axial surfaces of the earlier fold nappes. The fold nappes and thrust faults are deformed by two phases of folds and backthrusts related to a complicated "backfolding" episode, and by folds related to the rise of gneiss domes. A summary of the structural history is presented in Table 11. Figure 10a shows the axial traces of the major folds and their relative ages. Figure 10b shows average foliation and lineation orientations in 36 subareas in the quadrangle.

The nappes and thrusts transported hot rocks onto relatively cooler rocks, setting up temperature and pressure gradients which led to peak metamorphic conditions closely following the nappe stage. The dominant foliation is parallel to axial planes of nappe-stage folds. Reactions proposed to explain garnet zoning (see metamorphism section) imply a declining pressure just beyond the peak of metamorphism as a result of uplift. However, thermal equilibrium was apparently not attained throughout the pile of nappes, resulting in an inverted meta­morphic sequence, with lower grade rocks at the lowest structural level next to the Keene dome. The boundaries between Zone II and III assemblages and between Zones III and IV (explained in detail in the metamorphism section) roughly follow the nappe-stage thrust faults, ~but there are no sharp discontinuities in metamorphic grade across the faults. A large backfold (Beech Hill anticline) deforms the Zone III-Zone IV boundary. The rocks were in the stability field of silli­manite during backfolding and doming.

DESCRIPTION OF MINOR STRUCTURAL FEATURES

Equal area diagrams summarizing data for some of the minor struc­tural features from the whole quadrangle are presented in Figure 11.

Planar Features

Bedding. Bedding is present at some scale in all the metamor­phosed sedimentary units. It is most easily seen in the Perry Mountain, Francestown, Warner, and upper part of the Littleton Formations. The lower part of the Littleton is commonly thickly bedded, but with a little effort thin quartzite beds can be found in

Age

Mesozoic

Late Paleozoic

?Mississippian

Devonian

(Acadian orogeny)

Ordovician to Devonian

?Proterozoic Z to ?Ordovician

,...,

Table 11. Summary of structural history in the Monadnock quadrangle.

~

245-144 Extensional faulting; silicified zones; diabase dike.

326-245 Continued local shearing; ?Permian metamorphic disturbance?

326 Intrusion of Fitzwilliam Granite plutons and microdiorite dikes.

<380

393

Late open folds

DOMING

LATE BACKFOLDING

NW-trending, steep AP's; local crenulation cleavage.

Various trends, steep AP's; crenulation cleavage; strong linear fabric swirl

NE-trending fold axes, inclined AP's; linear fabric?

EARLY BACKFOLDING W-over~E verging reclined folds; local mylonitization and mylonitic foliation; some linear fabric; peak of metamorphism?

Intrusion of Spaulding Series plutons.

THRUST NAPPES E-over-W verging ductile thrust faults.

FOLD NAPPES E-over-~: verging isoclinal folds; pervasive foliation; quartz lineations?

413-402 Intrusion of Kinsman Granite.

480-415 Deposition of volcanic and sedimentary rocks.

600-480 Genesis of rocks forming cores of later gneiss domes.

'-I 0'1

Fig. 10. Summary of structural features in Monadnock quadrangle.

I j I I I

41

l I

? /

N

1 SCALE

0 2mi.

0 2 3km.

77

Fig. lOa. Axial traces of major folds and faults, showing phases of deformation: (1) nappes, (2) thrusts,(3) early backfolds, (4) late back­folds, (5) dome-stage and other late folds, (M) Mesozoic faults. Base of the Littleton Formation (dashed line) is shown locally for reference. Names of structural features are shown in Figure 13.

Fig. lOb. Subareas with average dominant foliation orientations and average sillimanite lineations. Major plutons are in gray. Base of the Littleton Formation (solid line) is locally shown for reference.

78

Fig. 11. Equal area diagrams summarizing structural features. Data is from the entire quadrangle unless indicated otherwise. Planar features are on plots to left; linear features to right.

PLANAR FEATURES

A. Poles to axial planes of 13 nappe-stage folds

B. Poles to axial planes of 60 intermediate age folds

C. Poles to axial planes of 15 late open folds on Mt. Monadnock

D. Poles to axial planes of 15 late folds east of the Keene dome

E. Poles to 7 mafic dikes

G. Poles to 46 tourmaline veins.

LINEAR FEATURES

A'. 17 nappe-stage fold axes

B'. 127 intermediate age fold axes

C'. 25 fold axes of late open folds on Mt. Monadnock

D'. 29 fold axes of late folds east of the Keene dome

F. 224 sillimanite and 7 andalump lineations

H. 101 quartz lineations

79

• •

•

• • •

• •

• • •

•

+

•

+

~. • • • , .. .. ..

•

•

• •

•

• ••

Early folds

•

I ntermedlate

symbols as above

• •• •

• • •

• •

• • • • . . '

Late folds Mt. Monadnock

symbols as above

•

•

• •• • •

• • •

• •

+

• •

• • • • • • • . :. ... , t •• • •• • . .. ....

• • • • • •• . . . . . ...... ~ .. . . .. ... . •• • • • •• • • •• • +

•

• • • . ·.: .... . . •

• • • •

• • • • • • •

• •

• •••

+

•

• •

• •

•

•

•

' :

•

• ••• • •

• •

•

• • •

•

80

• •

• •• .. .. • •

• •

•

• •

Late folds Keene dome

poles to axial

planes

/

all sillimanite

ond onoolump lineations

\

all quartz

lineations

+

. . . • • ••

0 ,.·. . .. ~ ·w . ~·.. . . ... . ..

. t: '2-!b •• ' • : c! ~~·: • . ... .. ~=: ..... . ,. . . ., ... •.. : .·· .

• . . ..

.. . • . .

. • . . . . .. ·--~ .... ..

. . ... t . . .. . . .... . . . · ... . . +

..

. , .

.. •

81

•

• • •

. . . .

Figure 12. Equal area plots of planar and linear features for the subareas shown in Figure 11.

PLAl.~AR FEATURES (on left)

0 poles to bedding • poles to dominant foliation A poles to secondary foliation )( poles to axial planes of folds 0 poles to fault planes

Numbers of measurements·

82

LINEAR FEATURES (on right)

+ sillimanite and mica lineations ¢ andalump lineations • quartz lineations x feldspar lineations ~ intersection lineations A crenulation lineations o early fold axes • intermediate fold axes • late fold axes

PLANAR FEATURES LINEAR FEATURES Mineral Fold Axes Lineations ., ., ., c:: ., Gl .,

c:: ~ N Gl Gl Gl 0 c:: .... Gl ~ c:: c:: .... ~ 0 cO c:: .... c:: c:: cO ., ~ .... ~ ~ ~

C::..lll 0 0 ~ ~ C::IO § Sol u .... "tl ~ ., Gl ... cO C>O ~ ~ "" "" m~ ., Gl ., Gl c::

Gl ;; ~ ~ c:: .... .... N p. ., ~ e Gl Sol ~ ., ., ...... .... ~:c ~ .... ., Sol =' >. ..Ill cO s "tl ., "tl ~ ~ ., ~ ...... cO Sol "tl Gl c:: ~ Gl Gl u .0 oc::.o "tl ~ ~ ~ =' ~-a "tl ., ~ .... Gl Sol .... .... ~

=' Sol cO '::1 Gl 0 0 ~ ., ~c:: ~ & Gl c:: Sol cO c:: :i ~

tn ""~ ., IQ Is< Is< Is< tnOS Is< H u ~ H tn

1 The Pinnacle 5 49 14 2(out) 7 33 1 2 1 2(in) Derby Hill 25 30 1 5 3 2 6 3(out) 17 61 6 4 1 2 3(in) Willard Hill 57 21 4 3 5 1 12 4 Page Hill 22 56 3 17 4 2 1 6 2 5 Marlborough 8 45 1 1 7 3 2 3 6 Brennan Hill 12 96 2 3 29 3 3 1 8 2 7 Glen Brook 17 15 2 5 3 1 1 2 8 Cooper Hill 23 1 2 9 Minnewawa Brook 16 38 3 9 4 2 6 1

10 Meetinghouse Pond ll 39 20 3 3 2 ll Shaker Brook 13 69 6 30 1 1 5 1 7 3 12 Little Monadnock 68 164 1 8 75 2 3 4 1 9 5 13 Howe Reservoir 33 13 4 2 2 4 2 14 Seaver Pond 6 13 1 2 1 15 Harrisville 10 28 2 16 N base, Monadnock 9 31 1 1 8 1 17 Hurricane Hill 14 7 2 1 4 18 Beech Hill 16 17 1 4 2 2 19 Dublin 21 33 8 3 20 NE of Thorndike Pd. 9 30 3 2 2 1 6 21 Hud Pond 3 24 1 2 22 Monte Rosa 42 56 1 2 9 1 2 1 2 3 23 Monadnock summit* 266 100 6 20 30 3 12 5 6 15 1 24 Pumpelly Ridge 41 43 3 6 20 4 1 3 5 2 4 25 Bigelow Hill 5 42 3 1 2 1 26 Parker Trail 31 56 1 10 1 1 27 Poole Reservoir* 148 149 6 26 16 45 4 2 1 42 7 28 Gilson Pond 17 51 1 1 14 2 1 2 10 29 SE of Thorndike Pd. 1 45 1 2 1 30 E of Thorndike Pd. 4 21 4 2 1 31 Cwnmings Pond 13 29 1 2 1 2 2 32 Jaffrey Center 14 88 1 1 12 6 3 12 33 Jaffrey 11 1 1 34 Gilmore Pond 12 26 1 4 2 35 E of Pearly Pond 1 6 3 36 Rindge 2 23 4 11 2 4

* No.23 includes MK-6; No.27 includes MK-447 and MK-454.

• •. I •• ... . • . 0 + • • .

' . • • • •

•

•

•

+ •

0

0

•

• •

• 0

+

• • • • •

•

0

• 0

•

• 0

0

0

•

•

0

• . . •

0 .. 0

\ . . . • .

0 •

0

_.. .x . 0

0 . oO

• 0 0

• • • 0

2 outside window

2

i nside window

83

0

X

•

• •

•

0

"

0

0 i •

0 0.

•

0

0

+

.. • • •

+

.. • 0

,. oCI:l

0

•

•

+

•

•

•

0

.., .. .. . • •• • 0 • •

• • • 0 •• --t ... • ••• o •

0 <8' • •o ••

0 0 0

0 0 0

0 0

• 0

0 • 0 0

o o o oo'\ cii•0° o o . 0 0

0

0 •

0

0 00

0 oo 0 0

• 0

• 0

0

• 0

•

•

0 00 0

0 0 .. 0 0 ~ oX

o o R o 0"-

~ • 0'b dlt~-. o o • o o• o 0

• •• I) 0 o. 0 ..

• • ... oo 0 •

o"" 0 0

.o. • • 0

outside window

3 inside

window

inside window

•

•

•

• + +

•

•

• • .. • + •

+ •

•

... •

+

• +

=-.~ +. • .+ • • • •

•

•

+· +

•

•

84

+

+

• 0

0 0

• • • ... i .

• 0 • "b . . + • eo •.· • : .. . 0.

oo . . . • • 0 •••

0 0

. . :.

• 0 . . . .. . . . .. • • •

.. • •

0 • •

•

.· . .

¥.~:~.

0 .

0 +

• • 0 :. • 0 • ~

oo••z~_. #• . ... . .. • ··~·:.: .. •• ••• ~0 •

o .•o o o • • o ~ .. ,.. . ·+

\o •

85

4

+ •• +e+

+ + ... • • ..

.. + +

+ •

0

+ +

5

•

+ +

+

• + +

I> ., +

,.1.

•

6

86

'I(

0 + 0 0

0 0

• • • • •

0

• • •

+ • • •

• • • • •

• • •

9

• • • • • + + • + +

.. + + "\. + •

+ + +

• •

•

0 • ... ~ a.· .... '

• 00 0 0

• •• ')(')(

0

0 •

• oe . ~ .. .,. • •

0 • 0 •

•

• •• •

0

••

• +

0

0

+· 0

0

0

'I<

0

0

."

. 0

.

10

II

12

0

0

87

+ •

\+ • ••

+ + + !~ + .. , •

+ •

+ -a+ +

·.~ + •• + +

+ \. ~. + + •• .. +-4:

+ 6. + A+

0

• ~')( cP

0 • ~

•

• •

•

..

0

0

• 0

• eo

0 0

oo'6 0

0 0 0 0 )(

0 )(

+ •

6 0 •• Q.

0

•

•

+ 0 0

• • 0

• 0

•

•

• • • • •

•

0

0 •

• •

88

13

0

14

15

0

89

16

+

+ + + 0 +

+

+ + • • • • 0

• •·\ 0

• 0 0 •

• t 0 • • • • •• • ••

~ • • • • • 0

17

• • •

•

+ + • • •

0 • ')( 0

0

0

• oo Ooo

0 0

18 c.:

• • 0

• 0 0 + 0 • •

• 0 •

• • •

0 • 0 • 0 0 • • 0

•o 0 ')( •

0

90

19

20

21

0

• •

0

0

0 0

• •

0

•

• • •

0

•• •

oOO oooo • o .• • : ""+ oo o •'o•o • ~0 0 0 • • • • •• 0 •

f)o • ••\ • • o # o o • • oo8 .i • •••

o• o • :o •

• • • • • • •

22

23 poles to bedding

0 0

0

0

~0+ 8

8>000

e ~o oo 0

0 oo 0 0

0 0

+

0

c9

0 0

% 0

0

0

0

OCP 0 0

0 0

I

24

symbols as above

+

• ...

••

+ -It> ...

• •• + +

• •

"t4' • • .,

~+ + •+ +

+to +

+ •

+

91

•

• • . )(,. .. .. . . . . + . ....,:·. . ..

• 'l( ...... . • • •• 4ele• .. .

6 .,X ... -.. ... • • A e e ~~ \. #e e • • .r.· -'1\c. • • •• )(

-:~'l( :··

)C

• • • 'l( .6.

)(

• 6

• + A • • !.

jJI!~'. • •• • ~ ... .

• • • ••

23 poles to fold

foliation and aKia I planes

•

•

/

24 symbols

as above

MK-6 random

andolumps on "-.....

bedding

/

•

o.,

0

+

0

+

, :I, .... ~ .

+

• Ill •

• • ' • ·\·· .. . .. • • .. .. . 0

•

.. . -\. ,.

• 0

.-.;.

•

• •• r • •

• • ••• ••

• •

92

25

•

+ •

•• 0 • 0

• I 0 • • • • • • • •

+ -t + ... ++ ...

0 +

+ ...

0

• • • 0 +

0 •

o. 0

0

• oq,

• +

0 ~-·· • • ' .

0 0 ••••• o , .. : • 0 • • o• ~ o o•l • •

• ~· •

00 • ~

tJia 0 •

27 t

0 0

0 0 + •

0 00

0 q, cf'CX)

0

0

+ •

0 0

+

0

~ • •

0 0

0

• ··+ •

+ Joo

+ rl' -+

0

tJ+e i .. •

"' od'

.·.t.• ~ .

0

Bog • • •

• ~0

, ... -:, •

0

00

+

0

~~Q) 0

•

00~ Ot:t~

• •

a, 0 0

•

0 0

•

Jl~ ago 00

0

0 0

0 0 ~ oo 0

lJo

0 GO

0 '0

•

•

0 0

+

27 poles to

fol!ation and axial planes

ll •

¥-11

/

• • .,. .. •

• . ; .. • •••• 'I( !I

~ • ,. o!l•." '~ ......

. : -'!·'1!;~~~~ ........ :-t :::. :~t. - " ..•. :.· ...

• •

MK-447

0 0 0

0 ~

oo"o

0 0 ox X~

0

+ X oo

0

)(

oO

0 )(

)(

X

MK-454

+

0 0)( 0

0 0 0

0 0

'I( ... 0 oo

94

• • • •• • A

•• • . • • i

' . ~· A

• .....

• • .. •• • • .. • .. • •• + • • • • ' .. • •

• • •

• • I' .. + ,. .. . • • ol •

•• • • I ••

• •• , . +

• + • • • • • •

0 ••• • • •r• • • •• •• • .. • • • • •

+ • • •• ..

• ••• •

0

•

28

29

•

• •

30

0 • • o:· •

,..,. ..... --;--~ .. ~ / ,' : • .I / \ / ..... ___ ......

~ •••••••

• . . + ........ "

+

95

+

• • • • • . , • •

•

31

32

+

•

33

+

96

+ • ~:Ji + t• • .... . • • •••

+ •

+

0

+

0 • • • • • •• • •

+ • 0 • •

• • • • • 0 .,.,

:~~~· . •

0

•

• • • fo •

0

0

• t>

• •

• • •

34

35

... +

+

36

+ • • • •

•

+

+

97

98

many outcrops. The quality of bedding in the Rangeley varies. The more massive schists are frustrating, especially where they contain abundant quartzo-feldspathic lenses and veinlets, in that bedding is not preserved and only the calc-silicate pods offer any clue as to bedding orientation. This is a major problem in the area north of the Chesham Pond fault, where the schists rarely have good foliation let alone bedding. Other parts of the Rangeley, for example the first 50 m below the Perry Mountain, are well bedded. Graded beds have been found in all units, and the topping direction from one unit to the next is certain. Locations of important outcrops where topping directions can be observed are shown on Plate 2.

Foliation. A pervasive foliation in the schists and most of the quartzites is defined by platy minerals. This foliation is nearly parallel to bedding in most outcrops. It lies parallel to the axial planes of tight isoclinal folds on Mt. Monadnock which are thought to have formed during the nappe stage of folding. Andalusite pseudo­morphs generally lie within the bedding planes, forming spectacular "turkey track" patterns on many surfaces. Some quartzo-feldspathic segregations and veins lie in the foliation while others cut across it. Because the Kinsman Granite is believed, for several independent reasons, to have intruded before or during the formation of the nappes, the oldest foliation in t~e Kinsman is believed to be coeval with the pervasive foliation in the schists, but this has not been proven. In central Massachusetts, Tucker (1977) concluded that the feldspar phenocrysts in the Coys Hill Granite predate the earliest tectonic foliation.

Mylonitic foliation. Both the Kinsman and the Spaulding bodies southeast of Mt. Monadnock, as well as some pegmatites, show various degrees of mylonitization. Quartz grains are flattened and feldspars are tapered with crushed margins. Because the Spaulding cuts across a nappe-stage anticline north of Thorndike Pond, it is clearly yo~nger than the nappe stage and the mylonitization must have taken place still later, probably during backfolding. In some mylonitized Kinsman the older nappe-stage foliation is weakly preserved. Most of the schists in adjacent outcrops do not show two foliations.

Crenulation cleavage. Both the late backfolds and the dome stage folds locally have an associated crenulation cleavage. This feature ranges from a weakly defined alignment of crenulation fold limbs to a well developed spaced cleavage in which platy minerals define cleavage planes that are two to four centimeters apart. Crenulation cleavage is best developed on the short limbs of outcrop-scale folds.

Joints. No systematic study of joint sets was made during this project, although Mt. Monadnock itself offers an ideal place to study brittle features because of the excellent exposure. Most of the large surfaces visible from a distance are joints rather than foliation or bedding. Many are coated by tourmaline and/or quartz. A plot of poles to 46 tourmaline veins from throughout the quadrangle (Figure

99

11 G) shows that most strike roughly east-west and dip steeply south. A systematic study of tourmaline vein orientations relative to joint sets might shed some light on their origin and relative age. They are younger than the microdiorite dikes. The most prominent lineaments on an air photo of the mountain trend N30°E, and the main microdiorite dike follows one of these directions. These relationships suggest that at least some of the joints had formed by the time of the post­tectonic (Fitzwilliam) intrusions. Other grominent joint sets visible on air photos include N25°W, N40°W and N70 E. Joints approximately parallel to the topographic surface are probably related to unloading. In the layered rocks unloading joints are best d~veloped where bedding dips gently, as on the west side of Mt. Monadnock. Here prominent joints are parallel to bedding. There are also gently dipping joints in the plutons. A photo of such sheeting joints in the granite quarry at Marlboro was used in Billings' text on structural geology (1954, Plate XIII, p.122).

Linear Features

Mineral lineations. Fine-grained sillimanite, biotit~ and musco­vite after sillimanite commonly show a strongly preferred orientation within planes of foliation or bedding. The platy mineral lineations may represent the lines of intersect~on with some other planar feature. Elongate quartz and feldspar grains form lineations in some of the rocks, and quartz rods are common both on quartzite beds and quartz vein surfaces, and as stretched pebbles in the conglomerates. Andalusite pseudomorphs show a preferred orientation mainly in the area of late fold hinges. Equal area plots summarizing data from the entire quadrangle are shown for sillimanite and quartz lineations in Figure 11 F and H.

Intersection and crenulation lineations. The lines of inter­section of bedding on foliation planes can be seen as compositionally different layers. The intersection of foliation with cleavage commonly appears as a crinkle or crenulation lineation.

Minor folds. Folds were described in the field in terms of fold axis plunge, axial plane orientation, rotation sense, tightness, and a record of what planes are deformed. The folds range in scale from tiny crenulations seen only with a hand lens to folds with amplitudes in tens of meters. Axial plane and fold axis data are presented for the entire quadrangle in Figure 11. It was not always possible to distinguish relative ages of isolated minor folds in the field. Isoclinal folds with axial planes parallel to the pervasive foliation were assigned to the nappe stage (Figure 11 A). Open folds with steep axial planes associated with crenulation cleavage, especially prevalent in the western part of the quadrangle, were assigned to the dome stage (Figure 11 D). Upright folds on Mt. Monadnock, similar in style to the dome-stage folds but perhaps younger, are shown on a separate equal area diagram (Figure 11 C). That leaves a large number of folds that deform foliation, but whose relative ages are unclear

100

beyond "post-nappe stage". They are grouped together as "intermediate folds" (Figure 11 B).

GEOMETRICAL ANALYSIS OF STRUCTURAL DATA

Structural Data in Subareas

The quadrangle has been divided into 36 subareas (Figure 10b) mainly for the purpose of presenting structural data. Equal area diagrams for each subarea are shown in Figure 12. Some subareas merely present data from isolated groups ot outcrops, while others were chosen to portray specific major structures, such as the Beech Hill anticline (Subarea 18) or the two limbs of the Thoreau Bog syncline (Subareas 23 and 24). Many subarea boundaries had to be arbitrarily drawn through areas with a continuum of data variation, for example between Subareas 6 and 12, where the dip of foliation gradually steepens from west to east. The average strike and dip symbols and average sillimanite lineation directions in Figure 10b were visually estimated from the subarea plots of Figure 12. They are meant to give a general idea of the overall structure and should not be taken too literally.

Construction of Cross Sections

Five east-west cross sections were constructed (Plate 5) at a scale of 1:50,000 and with no vertical exaggeration. Average dips of contacts were projected into the cross sections from nearby outcrops. The plunge of major fold axes were projected from areas of known trend and plunge, and adjusted according to known changes in plunge of minor folds and lineations at the earth's surface. In most cases there were few constraints on these adjustments, so the cross sections are highly interpretative.

PHASES ONE AND TWO: FOLD NAPPES AND THRUST NAPPES

t Introduction

The interpretation here is tentative and rests heavily on previously published ideas of the regional geology. However, there is strong evidence, first, that the large isoclinal folds on Mt. Monadnock are of the same age as larger isoclinal nappe-stage folds, and second, that the nappes have been cut by major west-directed ductile thrust faults. The assumption that the stratigraphic syncline referred to below as the "Monadnock syncline" formed during the nappe stage, may be incorrect. If so, much of the early structural history proposed below will require revision.

Tectonic Levels

Three tectonic levels are present in the Monadnock quadrangle, apparently separated by early thrust faults (the Chesham Pond fault

Gl E 0 -o Gl c:

"' Q)

~

·· F itzwilliam

· pluton

101

N

I SCALE

2mi.

0 2 3 km.

Fig. 13. Simplified geologic map of the Monadnock quadrangle showing the three structural levels separated by faults. and various structural features mentioned in the text. in part outlined by the contact at the base of the Littleton Formation.

102

and the Brennan Hill fault, Figure 13). These levels contain: (1) an autochthonous, upright sequence overlying the Keene dome, consisting of Ammonoosuc Volcanics, Partridge Formation, Clough Quartzite, local Fitch Formation, and Littleton Formation, (2) the folded Monadnock sequence consisting of the units shown in the strati­graphic column of Figure 3, intruded by Spaulding Tonalite, and (3) a "gneissified" sequence consisting mainly of Rangeley Formation and Kinsman Granite. Gravity studies show that the Cardigan pluton forms a 2-3 km thick, subhorizontal sheet-like mass (Nielson et al., 1976). Partly because the Kinsman and related Bethlehem Gneiss-are-exposed above the Bernardston and Skitchewaug nappes to the west, they are believed to have intruded prior to formation of the nappes (Thompson et al., 1968). The Kinsman is apparently cut by the Chesham Pond fault. Evidence in the Monadnock quadrangle indicates the Spaulding Tonalite is younger than the nappes. We will return to a discussion of the thrust faults which separate the tectonic levels after presenting evidence for the fold nappes, which are older.

Monadnock Syncline

The Monadnock syncline is separated into two parts by intrusions. Southwest of the intrusions, the syncline is relatively narrow, with good symmetry across its axial trace. It has Littleton Formation in its core, and extremely thin Perry Mountain, Francestown, and Warner Formations on each limb, probably thinned tectonically. An isoclinal fold in the Littleton Formation on the east slope of Little Monadnock Mountain plunges 54° south. It is believed to be a nappe-stage fold because the pervasive foliation is parallel to its axial plane. However, the plunge of the Monadnock syncline on a regional scale must be toward the north. More work is needed to determine how the south end of the Monadnock syncline relates to the Tully body of Monson Gneiss (Pike, 1968), and to the belt of sulfidic rocks mapped by Fitzgerald (1960) east of the Tully body, in Massachusetts.

The Monadnock syncline north of the intrusions is wider, and over-t turned to the southeast. Although the stratigraphy is grossly sym­

metrical across it, the presently upright southeast limb is much thicker than the presently overturned northwest limb. Prior to backfolding, these topping directions are believed to have been oppo­site to what they are today. The southeast limb includes a thick section of the upper part of the Littleton Formation, at least in part thickened tectonically. The southeast limb is folded by the younger northeast-trending Thoreau Bog syncline, but the northwest limb is not affected by it. This younger fold is apparently disharmonic, dying out rapidly to the north. The axial trace of the Monadnock syncline crosses Pumpelly Ridge near a gulch at 728 m elevation. North of this gulch graded beds are overturned, whereas most of the beds to the south are upright. Isoclinal folds along Pumpelly Ridge south of the gulch are overturned toward the northeast, consistent with the sense of other nappe-stage folds on the southeast limb, discussed below.

103

Folds on Mt. Monadnock

Nappe-stage folds are abundantly exposed on Mt. Monadnock. They consist of tight isoclinal folds with amplitudes far exceeding their wavelengths, and axial planes parallel to the pervasive foliation. Amplitudes are generally greater than five meters. Very few smaller nappe-stage folds were observed. The lack of small-scale folds may explain why nappe-stage folds were seldom observed in other parts of the quadrangle where outcrops are smaller. Most of the nappe-stage fold data in Figure 11 came from Mt. Monadnock.itself. Nappe-stage fold axes plunge in various directions due to rotation around younger folds. This is apparent in the wider spread of early fold axis orien­tations compared to intermediate axes in Figure 12, Subarea 23. Figure 14 is a sketch of the Seven Quartzites folded by nappe-stage isoclinal folds, about ten meters east of the Smith Summit Trail at elevation 823 m (MK-1104). The amplitude to wavelength ratio is approximately five to one. These are similar folds, in which beds are thicker in fold noses than on the limbs. Pervasive foliation in the schist, and ellipsoidal pits due to weathered-out chlorite in the quartzite, lie parallel to the axial planes of the folds. Data from a smaller nappe-stage fold exposed nearby (MK-1105) are plotted on an equal area diagram in Figure 14. The fold axis plunges N87°E at 26°.

The most dramatic exposure of a nappe-stage fold is on a seven meter west-facing cliff 150m west of the summit (MK-6). This fold was pictured in the frontispiece of the first edition of Billings' Structural Geology (1942), and so has been nicknamed "the Billings fold". It is a recumbent downward-facing syncline with fold axis plunging N58°E at 32°. A sketch of this fold, and data plotted on an equal area diagram, are shown in Figure 15. The foliation is parallel to the axial plane with strike N16°W, and dip 36°NE. A large bedding surface at the base of the cliff contains abundant andalusite pseudomorphs which appear to be randomly oriented in the foliation plane. A plot of 96 "andalumps" measured within one square meter of that plane shows no pronounced preferred orientation (Figure 12, MK-6, ~ith Subarea 23), but perhaps a slight maximum toward the NNE. This suggests the andalusite crystals formed under relatively static conditions, possibly later than the nappe-stage folding. Examples of andalumps forming a strong lineation are rare, and a special symbol is used for these on other diagrams in Figure 12.

The Billings fold lies structurally within the uppermost of three southeast-opening isoclinal synclines defined by the Seven Quartzites (Figure 16). The fact that they now open southeastward is a function of later folding, during which the originally overturned limb of the nappe, with the isoclinal folds as "parasites", was backfolded to its present stratigraphically upright position. The isoclinal pattern shown in Figure 16 was mapped out by following the Seven Quartzites back and forth across the mountain and paying close attention to the topping directions of graded beds. One of the best places to observe the folds is at elevation 914 m near the Smith Summit Trail, where the

i•'t

25m

3m{ zr.r=·_.:::_._::::··/ ·;.; ~· .·k~~- .-~·:·: .~:: ~

NW

foliation parallel to

axial plane

Fig. 14. Schematic view of Seven Quartzites folded by nappe-stage folds on a surface dipping 30 toward viewer. Fold axes trend nearly perpen­dicular to this plane. Equal area diagram shows four beds and the axial plane from a smaller nappe­stage fold near MK-1104, with fold axis plunging N87°E at 26•.

SE

N

...... 0 ~

...

I

Fig. 15. View to the east of "Billings fold", 150 meters west of Monadnock summit, on cliff trending Nmv. Quartzose horizons define bedding. Six measurements from a single bed plotted on the equal area diagram give a nappe-stage fold axis of N58°E, 32°NE, plunging nearly perpendicular to cliff. Dashed line on diagram is foliation in schist.

N

...... 0 l/1

106

Seven Quartzites close in an anticlinal hinge just north of the mafic dike. The Seven Quartzites can be followed in each limb north across the mountain as they get farther and farther apart. The beds to the southeast are upright while those to the northwest are overturned. An isoclinal fold hinge can be observed in schists between the two limbs where they cross the Marlboro-Dublin Trail, but outcrops are sparse to the north in the dense fir thickets on the north side of the mountain. The synclinal hinge structurally below this anticline is exposed near the Marlboro-Dublin Trail 30m to the northwest (MK-55).

Another interesting area where the nappe-stage folds are exposed is along the White Arrow Trail from elevations 836 to 910 m. East of the trail, on a large southeast-slanting surface, one can walk around the hinge of a recumbent anticline closing southeast (Figure 16, MK-2A). The axial plane foliation is well developed in the schist, and is refracted through the quartzite beds, forming a fan-shaped array around the fold hinge. The syncline structurally above the anticline at MK-2A is exposed on the adjacent cliff, although the ledges have slid down the mountain and thus are not exactly in place. The rotation sense of this fold pair (east-over-west), represented by an S-shaped line on Figure 16, indicates they are on the presently overturned limb of a larger scale anticline. Climbing up the trail one soon crosses the Seven Quartzites where they are doubled in the nose of this larger anticline. Above this hinge, east of the trail, there are isoclinal folds with west-over-east rotation sense, on the upright limb below the Billings fold. The rotation sense of these minor folds is shown schematically as a Z-shaped line in Figure 16.

There may be a nappe-stage fault below the Seven Quartzites at MK-1104. Graded beds are upright across almost continuous exposure from MK-1104 (elevation 810 m) down to the Black Precipice, a 10m cliff at elevation 795 m, where the Seven Quartzites reappear, also upright. I am reasonably convinced that the quartzites in the two places represent the same stratigraphic section. The first column on the left in Figure 6 (MK-27) was measured on strike southeast from MK-1104, while the second (MK-36) was measured at the Black Precipice. If they are the same, and if graded beds are upright between the two, the only possible explanation is a fault.

The axial traces of the isoclinal folds west of the summit, if they could be followed east through the rhythmically bedded schists and quartzites above the Seven Quartzites, should be exposed east of Pumpelly Ridge. Indeed there are isoclinal folds exposed along the Cascade Link Trail at about elevation 738 m (east of Figure 16). If one could follow the axial traces in the other direction, one should eventually be able to see the upper/lower Littleton contact and the Littleton/Warner contact folded by the isoclinal folds. Unfortunately that area has been intruded by the Troy Quarry pluton at the present level of erosion (see Plate 5, Cross Section C-C').

Billings fold

Smith

··.o ·.~

·. ~. ··.~

·1. Halfway

House I site

.!.....t... A Bald Rock

0 5 0 0

0

/ 0~

~0

1000 1500 2000 2500ft.

250 500 750m.

107

./

./

Fig. 16. Summit area of Mt. Monadnock, showing isoclinal folds marked by the Seven Quartzites. Station numbers refer to locations mentioned in the text. Bedding symbols with dot toward present tops.

Other Map-scale Nappe-stage Folds

Besides the Monadnock syncline and the smaller-scale isoclinal folds on Mt. Monadnock, some of the other map-scale folds on Plate

108

are interpreted as belonging to the nappe stage on the basis of their isoclinal shapes and senses of rotation. Figure 17 shows the schematic configuration of nappes and thrusts (discussed below) prior to backfolding. It can be seen in Figure 17 that the Howe Reservoir and Dublin Pond synclines are each paired with anticlines that close toward the west. They are on the originally upright limb of the Monadnock syncline. Smaller folds with the same rotation sense occur at Mountain Brook at the north base of the mountain (Plate 1 NE) and south of Gleason Brook in the northwest corner of Jaffrey (Plate 1 SW). The Gilson Pond anticline has the opposite sense of rotation and is on the originally overturned limb Qf the Monadnock syncline. The isoclinal folds defined by the Seven Quartzites, described above, are also on this limb. Although it would appear from Figure 17 that the Monadnock syncline represents a major nappe-stage closure, in a regional context it too may be on the limb of a still larger-scale nappe. This is important in that it could explain the presence of units younger than Rangeley Formation which are exposed in synclines associated with Kinsman Granite far to the south in Massachusetts (Field, 1975; Robinson et al., 1982a).

Howe Reservoir syncline. Although the actual hinge of the Howe Reservoir syncline is not exposed; the stratigraphic symmetry and iso­clinal nature of the fold is clearly demonstrated by an area of outcrops east of Howe Reservoir. Bedding and foliation strike WNW and dip moderately NE. The Warner Formation is exposed in the core of the fold, with Francestown and Perry Mountain on either limb. A mylonitic zone in the Warner suggests some local shearing in the axial region of the fold. The mylonite is deformed by a minor asymmetric backfold.

Dublin Pond syncline. The structure west of Dublin Pond is complicated and the outcrop is not good enough to determine whether repetitions in the stratigraphic sequence are due to folding or faulting. Plate 1 NE shows a nappe-stage syncline-anticline pair east

~ of Hurricane Hill which is deformed by the younger Beech Hill anti­cline. From Hurricane Hill southeast to the base of Mt. Monadnock the Francestown Formation is repeated four times, but in each case the sequence tops toward the southeast (Figure 18). The Warner cannot actually be followed around the fold noses from one belt to the next. Thus the repetitions could be due to faults rather than folds. The nappe-stage anticline north of Dublin Pond is based largely on two areas of float, one consisting of Rangeley in the core of the fold, and one of Littleton on the southwest limb.

Gilson Pond anticline and Meade Brook syncline. Another important nappe-stage fold in the Monadnock quadrangle is the Gilson Pond anticline. This was mapped by Nelson (1975). In its core is a 1.5 km long belt of Perry Mountain Formation, which locally is less than 100 m wide. Several brooks run transversely across the fold and the stratigraphic symmetry is best exposed in outcrops along "Ark Brook"

w

Level ( 3)

.c:._ Dkg

Sr

Level (2)

Fig. 17. Schematic configuration in cross-section of nappe-stage deformation, showing three tectonic levels separated by faults. Only dome gneiss, Littleton and Rangeley Formations, and Kinsman Granite are shown for clarity.

E

...... 0

"'

N

1

\.I ........ . /- \ I "\ I < ·,·.

-; ./ -- i" ·. ,,_, .... ,, · .. I \ -/ I .

-;1 ''--- - ""· , - I 1- I -, -': _,1~-.:: I_....\ /I-: ,-- . _...-\ ' I I:._ \ - _..... / .:

- I. I \ I- I '_, .· - I . \- _..... '--I I ..

"\ / I I_... ".:.- '· .· .....- \ ..... r I,.··

- I...,_\<. l I- I.·· - / .· -I.··

110

Dublin

Pond

76~

0 1/2 I km.

Fig. 18. Alternate interpretation for area south of Hurricane Hill, Dublin, at same scale as Plate 1 NE. Francestown Formation (north of stippled unit) and Warner Formation (stippled) are repeated in four places, each with tops toward the south (arrows). Heavy lines with triangles represent possible faults. Bedding and foliation symbols same as on Plate 2. Bedding symbols with dots represent overturned beds.

north of the Poole Memorial Road to the State Park. Both the Warner and Francestown are much thicker on the northwest limb of the anticline than on the southeast. There are isoclinal digitations of Francestown into the Warner along the west limb of the anticline, apparently nappe-stage structures.

111

Meade Brook, which drains Poole Reservoir, approximately follows the Warner Formation along the axial trace of a tight syncline, the Meade Brook syncline. This fold and associated nappe-stage folds are deformed by backfolds, discussed below (see Figure 24). The Littleton Formation appears in this syncline in a small area south of the reservoir, and again farther south (outside the area of Figure 24), continuing south as far as the Fitzwilliam pluton (Plate 1 SE). The Meade Brook syncline is believed to be comparable in scale to the Gilson Pond anticline. The axial traces of two smaller nappe-stage anticlines are shown on Figure 24. One axial trace passes through the digitation of Perry Mountain Formation southeast of the reservoir, and then is folded back and forth through the large area of Francestown north of the reservoir. The anticline west of the brook also has a core of Perry Hountain Formation.

The syncline east of the Gilson Pond anticline is incomplete, cut off by the Thorndike Pond fault zone. At Gilson Pond the axial trace of the anticline is cut by mylonitic tonalite of the "Gilson Pond type". If this tonalite is of the same age as the Spaulding Tonalite, this indicates a post-nappe age for the Spaulding. The anticline and the mylonitic tonalite are deformed by a younger NE-trending fold. Nelson (1975) correctly observed that early folds with steep NW-trending fold axes are deformed by the younger NE-trending folds. However, most of the NW-trending fold axes do not belong to the nappe stage. This is explained more fully in the section on backfolding.

Nappe-stage folds in the Derby Hill window. There are tight isoclinal folds in the south part of the window, exposed in a series of outcrops in the swampy area west of Willard Hill (Figure 19). The axial traces of these folds are cut by faults (described below) and folded by still younger folds. The isoclinal folds may have occupied the same structural position relative to the Monadnock syncline as the Howe Reservoir syncline.

Nappe-stage Thrust Faults

Brennan Hill fault. The nature of the contact between levels (1) and (2) (Figures 13 and 17) is subject to debate. The interpretation favored in this thesis is that a nappe-stage thrust fault separates Littleton Formation to the west from overlying gray-weathering Rangeley Formation to the east. The contact is difficult to map, and the name "Brennan Hill fault" is not meant to imply that it is any better exposed on Brennan Hill than elsewhere. The rocks to the west (Littleton) are sillimanite-rich, locally staurolite-bearing, monotonous gray-weathering schists. Those to the east ("lower part of

67~

Sra

) (

Sro

r I J~ ( Sra

'

112

0 soo'

9 lOOm ~------''

N

l

Fig. 19. South part of Derby Hill window. Formations are abbreviated as on Plate 1. Sra is augen schist assigned to the Rangeley, which may in part be metamorphosed mylonite.

113

the Rangeley") are more feldspathic, with abundant quartz-feldspar knots. In general they have a less uniform character, but they also weather gray. These same distinctions were discussed by Robinson (1963) between what he mapped as Littleton to the west, and the "Gray Schist Member of the Partridge Formation" to the east. A fault interpretation is strengthened by the rocks mapped as Fitch Formation exposed above the Clough Quartzite near Mt. Huggins (Plate 1 NW).

Alternatively, as discussed earlier, the Littleton of level (1) may not be Littleton at all, but rather may be Rangeley. The rocks east of the Keene dome would thus constitute a depositional sequence with Rangeley overlying Clough, in which case both levels (1) and (2) would be autochthonous. The arguments favoring this interpretation stem from observations to the north by J.B. Thompson, Jr., and Page Chamberlain (pers. comm., 1983). On the Unity dome and in the Marlow-Gilsum area, there are two or more conglomerate horizons with gray Littleton-like schist between them, apparently representing the transition from Clough Quartzite to the Rangeley Formation. The "Clough-Rangeley" rocks thicken from one nappe level to the next as one goes structurally higher, or toward what would have been the Merrimack trough prior to deformation. As mentioned elsewhere in this thesis, it seems odd that the clean quartz conglomerates of the Clough would be the basal member here rather than the polymictic conglomerate which is at the base of the type Rangeley section in Maine.

Chesham Pond fault and Derby Hill window. The Chesham Pond fault separates levels (2) and (3) and is much better documented than the contact between levels (1) and (2). The proposed configuration prior to later deformation is shown in Figure 17. At several places north of Minnewawa Brook and east of Chesham Pond the contact between undoubted Littleton Formation in level (2) and the physically overlying Rangeley Formation in level (3) can be approached to within a few meters. The normally intervening units are thus missing or extremely attenuated. They are exposed, perhaps in fault slices, along Horse Hill Road northeast of Marlboro, and locally along the west margin of the Derby Hill window (Plate 1 NW). Recrystallized mylonite has not been identified for certain, but the texture of the schist above the contact could have been produced by mylonitization of pre-existing quartz-feldspar segregations. The Chesham Pond fault cuts across the axial trace of the Monadnock syncline, in which the youngest rocks belong to the upper part of the Littleton Formation. Toward the northwest the fault cuts stratigraphically downward, approaching the base of the Littleton, and eventually cuts into the Rangeley. The point where this actually happens is intruded by the Babbidge Reservoir pluton (Figure 13). The upright "Monadnock sequence" may reappear to the northwest at Gee Mill. There is no control on the transport direction of the thrust, but if we assume that it was toward the present-day west, one possibility is that this cutting down across stratigraphy is due to a local lateral ramp (Butler, 1982), that descends toward the north.

'!"

Sro ~ § 39 Q: conglomerate /

. .

tl fsr I I lj /

cJ /

I Sr /

;4f' 301 I

I /

'1' / Spm ,.... l .. ·

/

/ I

Sf r

~~ I 44

I Sr I

/.,----. __ /

I /../ I (~ I I " \ " I

\ ......

I Sw '-

'-- ./ 0( ~

!I I I

e;, -Q

0 500ft.

0 lOOm.

Fig. 20. A portion of the north part of Derby Hill window. Abbreviations as on Plate 1.

114

N

1

ll5

The Derby Hill window consists of an area of folded rocks of the Monadnock sequence surrounded by augen schists and rusty gneisses of the Rangeley Formation. Initially it was thought the rocks in the window (Figures 19 and 20) were e~posed by a NNE-trending anticline that folds the Chesham Pond thrust surface. However, there is no symmetry within the window rocks across the proposed fold axial trace to support such a hypothesis. The alternative presented in this thesis is that the east margin of the window is an east-directed backthrust. Cross Section A-A', Plate 5, passes through the northern part of the window, and shows the proposed backthrust displacing the earlier west-directed thrust. It should be emphasized that the angle of dip on the faults is unknown. All the rocks are refolded by NW-trending late folds (dome-stage or younger).

The proposed backthrust cuts the axial traces of nappe-stage isoclinal folds. At the south end of the window a confusing group of outcrops may represent fault-bounded slivers of Littleton and Francestown surrounded by Rangeley. A contact between Francestown and the augen schist is well exposed there. The nappe-stage folds inside the window may have been on the upright limb of the Monadnock syncline, in a position similar to that of the Howe Reservoir syncline (Figure 17), and the Chesham Pond thrust has cut down close to them (compare Cross Sections B-B' and A-fo').

Along the west side of the window and at Horse Hill Road, however, the stratigraphic sequence locally faces downward from the Chesham Pond fault, with the intervening units between Rangeley and Littleton present. This can be seen on Plate 1 NW where the Warner and Francestown crop out locally along the fault east of Woodward Pond. These areas may represent fault-bounded slices carried along the fault from the overturned limb of the Monadnock syncline itself, or from the overturned limb of a subsidiary nappe-stage fold on the upright limb, as shown in the cross sections.

Thorndike Pond fault zone. The contact between levels (2) and (3) ~ is less easily identified east of Mt. Monadnock. North of Thorndike

Pond (Plate 1 NE) the Littleton Formation lies to the west of a narrow belt of Kinsman Granite with the Rangeley to the east. Southwest of Gilmore Pond (Plate 1 SE) a wider belt of Kinsman separates Rangeley on the east from an eastward-facing sequence on the west (only Francestown and Warner are exposed; the Littleton here is specu­lative). One interpretation is that the fault splays into several branches which are responsible for the repeated belts of Kinsman. Alternatively, these repetitions may be due to younger faulting. The latter interpretation is shown on Plates 1 and 5 and will be discussed further in later sections. In some places the nappe-stage fault may have been reactivated during backfolding and again during Mesozoic normal faulting. Because of the apparent repeated episodes of faulting over a broad area, the name "Thorndike Pond fault zone" is adopted for the four-kilometer-wide zone between the elongate mylonitic Spaulding Tonalite bodies at Whites Pond and Windmill Hill

116

on the west, and the widest belt of Kinsman on the east (Figure 13).

PHASES THREE AND FOUR: BACKFOLDING

Introduction

There is a wide range of folds which deform the dominant foliation in the Monadnock quadrangle. In several outcrops nappe-stage isoclinal folds are deformed by younger folds, but there are very few instances of intersecting younger folds, so that little success has been made in sorting out the folds of intermediate age. Open folds with steep axial planes and associated crenulation cleavage are obviously younger and present fewer problems. They are presented separately in Figures 11 and 12 as late or dome-stage folds.

The intermediate age folds are assigned to what Robinson (1979) referred to as the backfold stage of deformation. Axial surfaces of the nappes were "backfolded on a grand scale by east-directed folds" (Robinson, 1979, p.126). Mapping in central Massachusetts indicates that this episode was a complicated, multi-stage process involving recumbent folds, longitudinal flowage of some gneissic basement and mylonitization that occurred late in the episode. Intermediate age structures in the Monadnock area will be described from various parts of the quadrangle, and then an attempt will be made to present a possible sequence of events during backfolding. Regional correlation will be reserved until after the description of all phases of deformation in the quadrangle.

Intermediate Stage Folds, Mt. Monadnock

Asymmetric folds. West-over-east asymmetric folds are very commonly exposed on Mt. Monadnock. They deform the dominant foliation as well as bedding. The short limbs are generally steep to slightly overturned. The maximum amplitudes are on the order of one meter. Where exposure is sufficiently good, it can be seen that these folds

~ usually die out over about 10 meters. Figure 21 shows the lower part of an asymmetric fold north of MK-1104. A pegmatite vein is deformed by the fold, and a late quartz vein cuts across it at a low angle. The axial plane is oriented approximately N61°W, 57°NE, and the fold axis plunges 31° in a N67°E direction. Folds such as this occur all across the west limb of the Thoreau Bog syncline (Figures 13 and 16), where they have a northwest-over-southeast rotation sense. One excellent exposure of asymmetric folds deforming cyclic graded beds, stratigraphically above the Seven Quartzites, occurs on a nearly hori­zontal outcrop about 10 m north of the Billings fold, northwest of the mafic dike. On the east limb of the Thoreau Bog syncline, on Pumpelly Ridge, asymmetric folds also have a west-over-east sense, with axes plunging toward the northwest. They thus appear to predate the Thoreau Bog syncline. On Pumpelly Ridge care must be taken not to confuse them with the late upright folds, as both sets have axes plunging northwest.

0 2m.

N

N

117

s

Fig. 21. Profile view to east of asymmetric fold which deforms foliation (MK-1107), at elevation 825m, west of Monadnock summit. Equal area diagram shows six bedding measurements taken at corresponding points in sketch. Dashed line on diagram is the approximate axial plane. Fold axis plunges N67°E at 31°.

118

Boudinage. In some places where closely spaced quartzite beds are folded by the asymmetric folds, a curious sort of boudinage of the schist beds seems to have occurred, allowing the quartzite to flow and form a connection between adjacent schist beds (Figure 22). Dr. Shizuo Yoshida (pers. comm., 1984) suggested that some of these were sand dikes formed prior to lithification, but several factors favor a tectonic origin. Foliation in the quartzite beds is deformed and curves into the boudin necks from above and below. The boudin neck lines lie parallel to asymmetric fold axes, and in some instances vein quartz or pegmatitic material has collected along the neck lines. In some outcrops apophyses of quartzite project from isolated quartzite beds into the schist. Where the beds are upright and the quartzite projects downward these can look like load casts, but in other places they also project upwards from upright quartzite beds. Here, too, the axial planes of such structures lie approximately parallel to the axial planes of nearby asymmetric folds. If the schist beds are indeed boudinaged, this indicates some rather unusual competence contrasts between the schist and quartzite. Perhaps andalusite crystals strengthened the schist, or the quartzites were enough wetter than the schists to allow them to behave in a more ductile manner than the schist during the same stage of deformation.

Thoreau Bog syncline. The large anticline-syncline pair which dominates the foliaton pattern on Mt. Monadnock is responsible for the topographic shape of the mountain. I have named the syncline for a tarn-like bog located just north of Pumpelly Ridge near the axial region of the syncline, known as Thoreau Bog (Figure 16). Thoreau described the bog in his visits to Monadnock in the 1850's. Actually there are two bogs, one in each limb of the fold, and they are elongated parallel to foliation. The way in which the foliation pattern controls topography in this hinge region is especially striking on the air photo from which Figure 16 was made. Data from the two limbs is presented on Figure 12, Subareas 23 and 24. Bedding and foliation are shown separately for these subareas for the sake of clarity. It is obvious that both bedding and the dominant foliation

~ were deformed by the Thoreau Bog syncline. A 3-D view can be visualized by assembling the fence diagram in Appendix 1, Figure 42. Sillimanite and mica lineations trend NW-SE on both limbs, indicating that they are of a different age than the fold, but whether they are older or younger is not certain.

The asymmetry of this syncline has the same west-over-east sense as the minor folds described above. Because the northwest limb of the Monadnock syncline is not deformed by the Thoreau Bog syncline, I had originally thought there might be a fault across the lower north slope of Mt. Monadnock. However, if the syncline dies out in amplitude in the same way that the minor asymmetric folds do, a fault is not necessary. This follows an extension of "Pumpelly's rule" (Pumpelly et al., 1894, p.158), which seems appropriate since Pumpelly Ridge is also named after Raphael Pumpelly. Pumpelly's rule states that axes and axial surfaces of minor folds in an area are congruent with those

~

~~. "7 . ..;._~==. ===. ;=-,-"~.,....._.,~~--~~ - . : ....

Fig. 22. Two views of boudinaged schist beds. Quartzite beds are stippled. Upper view is part of the Seven Quartzite sec­tion, toward east, elev. 860m on the White Arrow Trail. Pits are weathered-out chlorite, with long dimension lying parallel to foliation. Note minor asymmetric folds. Lower view shows boudins in schist layers above the Seven Quartzites, elev. 893m, south of the Marlboro Trail, viewed toward the northeast.

119

120

Pond NO<Ih Pookl Mooo:,:::,dlike

Thoreau Bog

~ « --"l_p_ 21 y~---~--=z:::::::::::-""'---=::-l ~ ~ ~-~/T,·~

Fig. 23. View towards the east of NE­plunging backfold, along Pumpelly Trail 200m east of Monadnock summit (MK-799). Hammer handle is approximately parallel to the azimuth of the fold axis, which plunges N33°E at 29°. Lines on bedding surface are crenulations. Fine sillimanite and quartz lineations lie parallel to them. Symbols on the equal area diagram are the same as in Figure 12.

N

121

of the major fold structures of the same phase of deformation.

Along the Pumpelly Trail 200 m east of the summit, at elevation 920 m, there is an intermediate age syncline which plunges northeast (MK-799). A one-by-three meter block of rock from inside the syncline has slid down the slope, exposing a curved bedding surface that one can walk around on. Figure 23 is a sketch of this fold with data plotted on an equal area diagram. Bedding measurements define a fold axis plunging N33°E at 29°. Sillimanite lineations and coarse crenu­lations trend nearly N-S, obliquely across the fold, which suggests they date from an unrelated phase of deformation. They define a great circle on the equal area diagram, and are presumablyyounger than the fold. The axial planes of the crenulations are roughly parallel to those of late open folds observed in nearby outcrops.

Intermediate Stage Folds, Poole Reservoir Area

The map pattern in Monadnock State Park near Poole Reservoir is complicated (Figure 24). Despite excellent exposure, the structural history is poorly understood. The following analysis is presented as a possible explanation. Structural data from Subarea 27, which includes Poole Reservoir, is presented in Figure 12 with bedding, mineral lineations, foliation, and fpld axes shown in separate plots due to the abundance of data. Details from single outcrops (MK-447 and MK-454) are discussed as examples of the backfold stage.

Most planar features around Poole Reservoir dip northwest except around the areas of fold hinges (Plate 2). The spread in both planar and linear data is due to the interference of at least three sets of folds: nappe-stage folds, early backfolds plunging northwest, and later backfolds plunging northeast. The relative ages between these backfolds and those on Mt. Monadnock are unknown, but the following discussion will attempt to show they are younger, mainly on the basis of relationships to sillimanite lineations.

The axial surfaces of nappe-stage folds around Poole Reservoir are deformed by backfolds with a range of plunge directions. The style of deformation in outcrop differs greatly from one formation to the next. Backfolds in the lower part of the Warner Formation appear to have formed very plastically, with tight hinges and similar shapes. Some calc-silicate layers are extremely attenuated on the fold limbs. The Francestown, with its more massive bedding habit, behaved in a less ductile fashion. Folds which approach concentric shapes are common. The interbedded schists and quartzites of the Perry Mountain, by contrast, favored the formation of disharmonic folds due to slip within schist horizons.

An outcrop in the digitation of Perry Mountain Formation east of Meade Brook (MK-447) has been studied in detail (Figure 25). It serves as an example of the style of backfolds in the Perry Mountain. Minor fold axes and quartz lineations plunge steeply west. The

./ I

---/ I

I

Sfr

I I I

mineral lineations

/ quartz

/ mica

/ sillimanite

122

N

0 lOOm. ~-___J

0 500ft.

Fig. 24. Poole Reservoir area, Monadnock State Park. Axial traces of nappe-stage folds (dotted) are folded by backfolds (solid). Folds at MK-447 and MK-454 described in text. Samples from MK-432 and MK-629 described in section on metamorphism. Abbreviations as on Plate 1; "f1" = nappe-stage minor folds. Perry Mountain Fm. stippled.

N

1 Plan view

I ft .

...

lm .

quartz vein

N

·, •• !'. + ·I'.,. . . ... • . ..

0 0 0 0

0 i!! .. .., oo

...

0

co,.. .. ~ 0

"oo

0

" 0 0

0 ...

"

Fig. 25 Outcrop sketch showing disharmonic folds in the Perry Mountain Formation (MK-447). Equal area diagram shows data from this outcrop. Open circles -poles to bedding; X's - poles to axial planes; closed circles - fold axes; squares -quartz lineations. The outcrop occurs on the short limb of a larger fold.

....... N w

124

lineations are probably of the same age as the fold axes, but it is possible that some were rotated into colinearity from some previous (nappe-stage?) orientation. No nappe-stage fold closures were found in this outcrop. At first sight the outcrop pattern appears to be the result of interfering sets of folds. On the equal area diagram the fold axes form a nearly coaxial cluster, and the poles to axial planes lie on the same great circle as the poles to bedding. Such a pattern could indeed be produced by two sets of coaxial folds. However, close inspection of the outcrop sketch shows that some of the folds are box folds, and the pattern is probably a result of disharmonic folding due to the strong ductility contrast between the thinly bedded schists and quartzites.

In the southern part of Figure 24, the rotation sense of NW-trending folds is dominantly counterclockwise. MK-447 is on the short limb of one such fold.

A detailed study was also made at MK-454, where the contact between Francestown and Perry Mountain Formations is deformed by a NE-trending fold. As one walks around the nose of this fold, the sense of rotation of minor folds changes from clockwise west of the fold hinge, to counterclockwise on the east. Quartz lineations on many bedding surfaces plunge 21° to 70° toward the northeast, roughly parallel to the fold axes.

Equal area plots of data from MK-454 and MK-447 are included in Figure 12 for comparison with data from all of Subarea 27. It can be seen that the distribution of data for the whole subarea is a combi­nation of the NW- and NE-trending structures. Age relations are not clear in the Poole Reservoir area, but in the Gilson Pond area 2.5 km to the northeast (see below), NE-plunging lineations appear to be younger. The fact that the quartz lineations show a greater spread than the sillimanite and mica lineations suggests the quartz lineations are older, or that they were not parallel to begin with. Some may be inherited sedimentary features. Although the amount of

~ plunge varies greatly for the sillimanite and mica, the direction of plunge is mostly to the NNW. Whether the sillimanite and mica linea­tions are related to the NW-trending backfolds, or represent a still younger phase of deformation, is uncertain. It is important to note that sillimanite and mica lineations trend toward the northwest all across the eastern and central portions of the quadrangle, regardless of the orientation of backfolds (Plate 3). They may be related to the transport direction whereas quartz lineations are more tightly constrained to be parallel to folds in bedding.

125

Intermediate Stage Folds and Mylonitization, Gilson Pond Area

Some important structural relationships are shown by the rocks around Gilson Pond (Plate 1 SE). The Gilson Pond (nappe-stage) anticline, which at this point has a core of Francestown Formation, is cut off by the Spaulding Tonalite ("Gilson Pond type"). This may be an intrusive relationship or a fault relationship, or both. There are also tight isoclinal folds along the Francestown-Warner contact which are cut off by the tonalite. The tonalite has been mylonitized to varying degrees. Small bodies of Kinsman Granite, and a pegmatite dike 50 to 100 em thick and about one kilometer long, are also strongly sheared. Tails on sheared feldspar grains in the tonalite suggest a west-side-up sense of shear.

All the rocks, stratified and intrusive, bear NW- to WSW-plunging mineral lineations (Figure 12, Subarea 28). In the Warner Formation the NW-plunging lineations are pervasive but faint. A stronger set of NE-plunging mineral lineations lies parallel to NE-plunging folds with west-over-east movement sense. The northeast lineations are especially well developed in the Warner and Francestown, but are also present in the Spaulding, mainly as faint biotite lineations. Along the southeast shore of Gilson Pond, both the Warner and the Spaulding are deformed by a large NE-plunging fold. The shoreline, concave to the north, approximately follows the shape of the synclinal part of this fold. Poles to bedding and foliation in Figure 12, Subarea 28, show a spread due to folding around both NW- and NE-trending fold axes. The younger NE-plunging linear structures are shown encircled by a dashed line.

The large body of Spaulding Tonalite west of Gilson Pond has a mylonitic texture, but has a less concordant shape than the Spaulding bodies east of the pond. It cuts across the axial trace of the nappe-stage digitation to the south, but its contacts have not been mapped in enough detail to ascertain whether it is deformed by the backfolds.

In summary, there is evidence that the Spaulding Tonalite intruded across nappe-stage folds, underwent mylonitization during a west-side-up phase of deformation, and then was deformed by NE-plunging, west-over-east backfolds. The Spaulding also bears NW-plunging lineations which may or may not be related to the NW-plunging folds in the metamorphosed sedimentary rocks.

Intermediate Stage Folds, Southeast of Thorndike Pond

East of a belt of Kinsman Granite, a nappe-stage syncline with a core of Francestown Formation is deformed by folds with axial planes dipping northwest (Plate 1 SE; Figure 12, Subarea 29). Mineral lineations plunging steeply northwest were observed in several outcrops, and the folds are interpreted to be early backfolds similar to those near Poole Reservoir. Minor folds at the scale of the

126

outcrop are rare, although there are some minor upright warps that probably have little effect on the map pattern. The map pattern west of the Kinsman is very different. A belt of Francestown Formation strikes parallel to the Kinsman for hundreds of meters. Several outcrops of mylonitic schist occur within the first 30 m east of the Kinsman. It is difficult to tell what the protolith may have been. Two to five millimeter feldspar grains lie par'allel to a steeply SE-dipping foliation. A younger mylonitic foliation dips 70° to the northwest, and the feldspars are sheared out with tails that indicate a west-over-east sense of shear. The younger foliation contains steeply plunging biotite lineations. On Plate 1 SE a fault is drawn through this area, with teeth marks indicating thrust motion toward the east. The same line to the north and south is shown as a Mesozoic normal fault, and this is meant to imply that thrusts which developed during the backfold stage were later reactivated during the Mesozoic. There is no evidence right at Thorndike Pond for normal movement, but there may be some parallel normal faults which are not exposed.

Beech Hill Anticline

The strike of bedding and foliaton in Subareas 16 and 19 (Figure 10) is dominantly to the northeast, with dips to the northwest. In Subarea 18, however, the strike ~s progressively more to the north until, on Beech Hill, it turns quite abruptly to the WNW, with dips to the north. This then defines the Beech Hill anticline (Figures 10a and 13). West of Beech Hill there is a large area of no outcrop, but still farther west in Subarea 13 the dominant strike is roughly E-W, and then in Subareas 8 and 9 it is again dominantly to the northeast. The bull's eye symbol on Figure 12, Subarea 18, is a beta maximum of bedding and foliation intersections from the whole subarea, plunging N46°W at 48°. The axis of the Beech Hill anticline estimated from measurements taken just around Beech Hill plunges N57°W at 38°. Regardless of the precise direction, the point is that this fold is overturned toward the southeast, with a fold axis oriented in the same direction as many of the minor folds and mineral lineations in the central part of the quadrangle. On the overturned limb of the Beech Hill anticline, between Beech Hill and Dublin Pond, there are minor folds in the Warner Formation which plunge northwest and have a counterclockwise rotation sense, consistent with the major anticline. Sillimanite and quartz lineations also plunge northwest.

The Beech Hill anticline at first appears to be paired with the Monadnock syncline because both are overturned toward the southeast. However, my present interpretation is that the Monadnock syncline is an older structure on the southeast limb of the Beech Hill anticline. Thus, the anticline is a much larger-scale fold than is at first apparent, deforming the Kinsman Granite as well as the metamorphosed fault between levels (2) and (3). This fault is the Chesham Pond fault west of Dublin, and it is folded by the Beech Hill anticline around to the southeast and south, where its trace is lost in the Thorndike Pond fault zone. The overturned southeast limb of the Beech

12 7

Hill anticline continues south across central Massachusetts. It is presumably a short limb in the backfold system. Deep seismic reflection profiling to the north (Ando et al., 1984) suggests that the steeply dipping structures do not continue very far below the earth's surface.

In the area west of Beech Hill, the Spaulding Hill pluton appears to have domed the overlying rocks around it (Figure 10b). Perhaps some of the Spaulding Tonalite plutons moved tectonically late in the backfolding episode, in the same way that the streamlined bodies of Monson Gneiss basement in Massachusetts are thought to have moved (Robinson, 1963; Fitzgerald, 1960; Pike, 1968). Perry (1904) may not have been so far off the mark when he observed that the plutons around Mt. Monadnock appear to have pushed aside the foliation in the country rocks, although he attributed the deformation to the time of intrusion. If the NW-trending linear fabric formed at the same time as the Beech Hill anticline and other NW-trending backfolds, it may record the transport direction during backfolding. As we will see in the discussion of the dome stage, however, late open folds have approximately the same orientation, and some of the mineral lineations may have formed then instead.

Intermediate Stage Folds in the Troy Area

Two sets of folds which deform foliation are exposed in an outcrop of Rangeley Formation on the west side of a residential street, 1.7 km SSE of Troy village (TR-174, Figure 26). The older fold has a west-over-east sense of rotation, an axial plane which strikes N17°E and dips 78°SE, and a fold axis plunging S09°W at 34°. It deforms a still older foliation. The younger folds are more open, clearly deform the first, and have an associated crenulation cleavage which strikes N48°E and dips 60°SE. Their rotation sense is neutral to east-over-west, and they plunge S23°W at about 36°.

crenulation cleavage

0 lm

N

1 Fig. 26. Interference pattern formed by two sets of south-plunging folds in the Rangeley Formation (TR-174), south of Troy. Expla­nation in text.

128

The younger folds are presumably dome-stage folds. This outcrop illustrates how difficult it is to assign a relative age to folds in the southwestern part of the quadrangle, except where folds of different ages intersect one another. The linear fabrics which developed during backfolding and doming both plunge south (Figure 12, Subareas 6 and 12). The situation is similar to that in the central part of the quadrangle (discussed below), where both linear fabrics plunge northwest.

Intermediate or Dome Stage Folds in the· Derby Hill Area

The axial traces of nappe-stage folds and the proposed thrust faults in the Derby Hill window are deformed by a variety of cross-folds, all of which are shown in Figures 10a and 12 as intermediate in age. Some of them could belong to the dome stage. The proposed backthrust along the east side of the window may have formed late in the nappe stage, or it may belong to the backfolding phase. If the latter is true, then the cross-folds must be late backfolds or younger.

The equal area diagrams (Figure 12, Subareas 2 and 3) are divided to show data separately for structures outside and inside the window. The plots show that there is litt~e difference in orientation inside and outside the window. Thus the events responsible for the distribu­tion of planar and linear orientations mainly post-date the nappes and the Chesham Pond fault. There is a cluster of fold axes and mineral lineations which plunge northwest. The axial planes to these NW-trending folds dip moderately to steeply north. Some outcrops contain W-plunging folds with upright axial planes. In addition, in the northern part of the window around Derby Hill, there are some S-plunging folds. The relative ages of these various folds are uncertain. Nappe-stage foliation is deformed by all three.

Summary of the Backfolding Episode

The correlation of intermediate stage folds from one area to another presents some problems. The simplest solution is to extend the age relations observed in the Gilson Pond area to the rest of the central part of the quadrangle. Folds with NW-trending axes formed early, perhaps contemporaneous with local mylonitization, and were followed by NE-trending folds. The early backfolds would thus include the Beech Hill anticline with its related minor folds, the asymmetric folds on Mt. Monadnock, and the NW-trending folds southeast of the mountain. According to this solution, the Thoreau Bog syncline later rotated some of the folds of this early stage away from their north­westerly orientations. However, the minor folds south of Beech Hill and those on Mt. Monadnock have opposite senses of rotation, and if they are the same age, this would require a backfold syncline between the two areas.

a.

NW

b.

C. ..__ __ _

NW

/

--

--­Chesham

.---..-...-

129

SE

SE

. c\\(\e --Q(\\1 --~ ...-

~ ~

~ SE

Fig. 27. Schematic cross-sections to show proposed backfolding sequence. a. Asymmetric folds deform nappe-stage structures. b. Thoreau Bog syn­cline. c. Earlier structures rotated into short limb of the Beech Hill anticline. Heavy arrows show regional shear sense. Small pole shows topping direction, and X is shown as a reference point.

130

An alternative solution would have the asymmetric folds on Mt. Monadnock as the earliest manifestation of backfolding, as shown in Figure 27a schematically superimposed on the nappe-stage folds in the Seven Quartzites. These folds were then carried along passively as the Thoreau Bog syncline formed (Figure 27b). We noted earlier that sillimanite lineations seem to be independent of the Thoreau Bog syncline. Unless the syncline formed by passive flow, which is doubtful in view of its disharmonic form, these lineations are apparently younger. If the lineations are related in time to the Beech Hill anticline, then it too must be younger than the Thoreau Bog syncline. By incorporating the resulting pattern into the short limb of the Beech Hill anticline, we arrive at the schematic cross section in Figure 27c. To the east the short limb has undergone mylonitiza­tion, especially in the area of the earlier nappe-stage thrust fault. Further compression resulted in the late NE-trending folds such as those southeast of the mountain.

The timing of the sillimanite lineations remains a problem. Throughout most of the eastern and northern portions of the quad­rangle, fine sillimanite and mica lineations plunge down-dip toward the northwest, or, in cases where surfaces dip east, toward the south­east (Plate 3). Whether or not they are of the same age as the back­folds is a question we will discu~s during the discussion of the dome stage.

PHASE FIVE: DOMING

Introduction

The Keene dome is one of the many gneiss domes along the Bronson Hill anticlinorium, which, due to their relatively low density, rose gravitationally late in the Acadian orogeny (Thompson et al., 1968). These authors describe the Keene dome as having mushroom-shaped margins, with a northeast lobe overturned northward, and a highly compressed conical shape at the southern end. The south end is superimposed on an earlier recumbent anticline and syncline, over­turned toward the southeast, which do not appear at the surface in the Monadnock quadrangle. Foliation and bedding next to the east edge of the Keene dome dip moderately to the east (Figure 12, Subareas 4, 5 and 6; Plate 2). Farther east dips steepen to vertical and beyond (Subareas 10, 11, 12). This is interpreted as an effect of doming (Sections D-D' and E-E', Plate 5). Features believed to belong to the dome stage are described below from various parts of the quadrangle, followed by a general discussion.

There are abundant minor structures east of the Keene dome which are obviously related to the dome. They are plotted on Figure 11 D. Upright folds with an associated crenulation cleavage are abundant up to four kilometers away from the dome. North of the latitude where Rt. 12 crosses the edge of the dome, minor folds plunge gently north (Figure 12, Subarea 4). At Rt. 12 they are nearly horizontal. To the

131

south, most dome-stage folds plunge south (Subarea 6). The rotation sense nearest the dome ranges from east-over-west to more or less neutral. This is apparent from the map pattern of contacts in the area of Page Hill, north of Rt. 12 (Plate 1 NW). East of the Marlboro syncline's axial trace, minor dome-stage folds have the opposite rotation sense.

Marlboro Syncline

The domes of the Bronson Hill anticlinorium are characteristically separated by tight synclines in the mantling metamorphic rocks (Thompson et al., 1968). Although there is no well defined dome to the east, the Marlboro syncline is believed to be a dome-stage feature (Figure 10a). East of Marlboro it is well defined, where coinci­dentally its axial trace is roughly parallel to the zone in which bedding and foliation are vertical. Subarea 7 on Figure 12 illustrates the axial region of the fold, which plunges northeast at that point. To the south of the Marlboro pluton, the syncline follows a belt of the upper part of the Rangeley Formation. The fold axis, like those of the minor folds, presumably flattens out and eventually plunges south. There are graded beds with tops facing west along its east limb where it crosses Rt. 124. South of there, the syncline's position is uncertain due to the current poor understanding of strati­graphy within the Rangeley. Close attention to the rotation sense of minor folds should help locate the axial trace. More work is needed to determine the relations of the Marlboro syncline to the Tully dome in Massachusetts, if indeed it extends that far south. North of Marlboro the syncline's trace is also uncertain. It strikes toward the area of a high positive Bouguer gravity anomaly which is centered on the Rangeley Formation west of the Cardigan pluton (Nielson et al., 1976). Peter Robinson (pers. comm., 1984) has suggested that this-­anomaly might be due to a deep synclinal septum of metamorphosed sedi­mentary rocks which would be more dense than the surrounding plutons and domes.

! Rindge Area

Minor folds in the southeast corner of the Monadnock quadrangle are similar in orientation and style to folds near the Keene dome. They are upright, plunge gently north and south, and are associated with a crenulation cleavage (Figure 12, Subarea 36). The crenulation forms an intersection lineation on bedding and foliation, and silli­manite lineations are approximately parallel to this. This fabric is similar to the dome-stage fabric that pervades central Massachusetts, both in orientation and in amount of plunge (Robinson, 1979, Fig. 5-7).

Late Open Folds, Mt. Monadnock

Late symmetrical folds trending northwest are exposed in several areas on Mt. Monadnock. They have steep axial planes which are paral-

132

lel to crenulation cleavage in schist beds. They range in wavelength from 10 em to 10 m. Some outcrops, where minor late folds are not obvious, have surfaces which are broadly warped about NW-trending axes. A plot of structural data from these folds is shown in Figure 11 C. The great spread in fold axis orientations is due to the fact that the surfaces which are folded had a wide range of previous orientations. Along Pumpelly Ridge the folds plunge moderately to the northwest, whereas west of Mt. Monadnock they are nearly horizontal. The broad NNW-trending anticline centered on Monte Rosa is a large fold of the same phase (Figures 10a, 13, and Plate 1 SE). These late open folds may post-date the dome stage.

Cobb Hill

The contact between the Rangeley Formation and the Cardigan pluton is folded by an open NW-trending fold across Cobb Hill north of Lake Skatutakee (Figures 10a, 13, and Plate 1 NE). The Kinsman Granite lies under the Rangeley south of Cobb Hill, and relationships are unclear to the north. The fold deforms an older S-shape in the contact, which may be a backfold, or may have been an irregularity in the original intrusive contact.

Discussion

The problem of correlating dome-stage folds from one part of the quadrangle to another is similar to the problem encountered with the backfold stage, and is tied to the important question of timing and significance of the mineral lineations. Plate 3 shows how the linea­tion pattern changes from a dominantly northwest trend in the central and northern parts of the quadrangle, to a dominantly south trend in the southern part. Figure 28 shows equal area plots of mica and sillimanite lineations from these two large domains.

Two interrelated questions need to be answered: 1) Did the silli­manite and mica lineations form during backfolding, or doming, or

t both? 2) Do the NW-plunging lineations and the S-plunging lineations date from the same stage of deforma~ion?

We know from the Gilson Pond area that sillimanite and mica linea­tions lie parallel to the axes of two different sets of folds, one plunging northwest and the other northeast. Perhaps a more diligent search in the areas with nearly coaxial backfolds and dome-stage folds would reveal cases where two nearly parallel sillimanite lineations occur together in the same outcrop. So far this has not been observed, at least not on the same surface. The significance of two slightly different lineation orientations in the same outcrop may not have been recognized in the field. Subareas 10 and 11 (Figure 12) are in the critical region where lineations seem to fan from northwest to south. There is quite a mixture of orientations in the area immedi­ately south and southeast of the Marlboro pluton. However, as yet no outcrops have been found with intersecting sillimanite lineations,

!

N

. . . .. . . .. . . ~ . ·. i. .

• ·.,·j,:::. :-. .. .. . . . . . . ·.: .. . . : ·.. ·,: ' ··. . . '. .:. : ... : ~·. . :· -·· . . . • ·:.:-a: •• •

~. ·? .· .• ....

N

.. . .. . .. : .... t'·: :. . . . .... . . . .

.• :··· .. :..:i~ . :··· .. • • . • · .. .::;.: :i ••• • . . .. .. . . -~ .. ,·· ....

·.J •• • ••• • •• . . . .. . . . . . ·: ; . :. '

. +

. .

. .

. . ..

. .

Fig. 28. Sillimanite and mica lineations from two domains: left, 171 from the southern part of the quadrangle (Subareas 5, 6, 10, 11, 12, 25, 35 and 36); right, 208 from the central and northern part (Subareas 1-4, 7-9, 13-24 and 26-34).

133

although in a few cases adjacent outcrops have mineral lineations and crenulation lineations at an angle to each other.

One interpretation would be that all the NW-plunging lineations formed during backfolding, and all ~he S-plunging lineations formed during doming, or vice versa. Alternatively, they all may have formed during one phase, and the pattern on Plate 3 may represent a swirl in the direction of transport. A third possibility is that the linea­tions record a swirl in transport direction that was operative during both backfolding and doming.

To the southwest in Massachusetts, east of the Warwick dome, the map pattern of dome-stage lineations and fold axes forms a distinct swirl (Robinson, 1979). The trace of the swirl trends approximately N-S. West of this trace the lineations plunge north, across the trace they plunge due east, and to the east they swing around to the south. This S-plunging domain is continuous with the S-plunging domain in the Monadnock quadrangle.

The swirl of lineations in the Monadnock quadrangle, if that is what it truly represents, may in part record the transport direction of the Spaulding Hill pluton in the core of the Beech Hill anticline, as it moved north and then up toward the southeast. The Gap Mountain and Cummings Pond plutons (Figure 13), which may be the north end of the Hardwick pluton, may also have moved longitudinally. The long axis of the Tully body, which is a zeppelin-shaped body of Monson Gneiss plunging south (Pike, 1968), projects toward the southwest corner of the quadrangle, west of the Monadnock syncline. The Tully body and the Spaulding Hill pluton apparently lie in the same axial surface. The late backfolding and doming may not have been too remote from each other in time (Robinson, 1967), and I envision a scenario in which plutons moved longitudinally in response to E-W compression just as the domes began to rise. We will return to this idea in our

134

discussion of the regional context.

LATE PALEOZOIC INTRUSIONS AND DEFORMATION

The Fitzwilliam Granite plutons and the microdiorite dikes cut across all the structures described above. As discussed in the section on intrusive rocks, the Fitzwilliam is probably Mississippian. Fowler-Billings (1949a, p.1269) described the granite as "slightly foliated in many exposures". Some of this foliation may be igneous flow texture, but the more strongly developed foliation parallel to the Thorndike Pond fault zone suggests some late Paleozoic deformation along this zone. The microdiorite dikes are also weakly foliated. The possibility of late Paleozoic metamorphism is discussed in the metamorphism section.

MESOZOIC FAULTING

Silicified zones, a breccia zone, and minor faults with slicken­sides are evidence for extensional faulting. They are probably of the same age as the Connecticut Valley border fault, which passes about four kilometers west of the Monadnock quadrangle (Moore, 1949; Thompson et al., 1968), and similar zones in adjacent quadrangles (Robinson-,-1963; Fitzgerald, 1960; Pike, 1968; Greene, 1970; Peterson, 1984; E. Duke, 1984).

Silicified zones occur at five localities in the quadrangle, shown on Plates 1 and 4 in black. Three are in the southern part of the quadrangle, on strike with the Thorndike Pond fault zone. Fowler­Billings (1949a) described two of these, as well as two more en echelon to the south in the Winchendon quadrangle. The zones-consist of very fine-grained pink to gray silicified rock, cut by a network of quartz veinlets. Because they are more resistant to weathering, they locally form sharp linear ridges. The silicified zone northwest of Pearly Lake strikes N30°E. Kinsman Granite crops out 150 m to the west, and strongly foliated Fitzwilliam Granite 250 m to the east. A

t newly found silicified locality lies west of the same belt of Kinsman. The other silicified zones lie within the Fitzwilliam pluton. Poles to 10 quartz veins measured in an outcrop 200 m east of Rt. 12, 250 m north of the quadrangle boundary (FZ-59), suggest a minimum compres­sion direction oriented about N81°E, 40° (Figure 29). The foliation in the granite dips steeply west. More work is needed to see if this foliation is most strongly developed near the silicified zones. If so, it may record late Paleozoic deformation, suggesting a long history of movement along this zone.

In roadcuts along Rt. 202 south of West Rindge, there are some vuggy quartz veins and small faults. These roadcuts are about halfway between the silicified zone at FZ-59, described above, and the inferred extension of the Spofford Gap fault to the east (E. Duke, 1984). The poles to some of the faults and the directions of the slickenlines are included in Figure 12, Subarea 36. Steps on the

N

135

Fig. 29. Equal area diagram of ten quartz veins in sili­cified zone at station FZ-59. Solid circles are poles to veins. Dashed line is foli­ation in granite. Beta inter­section probably represents a common slip direction during Mesozoic extension.

slickensided surfaces indicate normal movement in all cases, regard­less of direction of dip, with a very minor right-lateral component.

136

A well developed network of quartz veins also cuts rocks of the Littleton Formation at elevation 375m on an east-facing slope, 1.3 km S30°W of Bonds Corner in Dublin (Plate 1 NE). The dominant set of veins strikes N31°E and dips 73°SE. The schists are strongly altered, green, chloritic rocks, but the alteration probably predates the quartz veins, as it is very widespread in this part of the quadrangle. Kinsman Granite crops out down the slope 100 m to the east. Although the country rock is not silicified to the same extent as in the zones on strike to the south, this occurrence is believed to be related to the same Mesozoic fault system.

A fifth, spatially unrelated, silicified fault breccia occurs in the Littleton Formation west of the Spaulding Hill pluton (MB-40). The silicified zone does not crop out, but float can be found for about 20 m along a N25°E trend between two big NW-facing ledges of schist, south of the old railroad bed 75 m east from where it crosses Minnewawa Brook. The zone is about three meters wide.

SUMMARY AND REGIONAL STRUCTURAL IMPLICATIONS

In summary, rocks of the Merrimack trough were deformed by fold­nappes and thrust-nappes directed toward the west, followed by back­folds and doming. The Brennan Hill fault transported rocks of the "Monadnock sequence" over the thinner, autochthonous sequence of the Bronson Hill anticlinorium. The thrust is interpreted here as a ductile thrust near the root zone of the Bernardston nappe. The Ches­ham Pond thrust carried the Kinsman Granite and Rangeley Formation westward over the Monadnock sequence, cutting across the nappe-stage Monadnock syncline between the Fall Mountain nappe and lower nappes. A major backfold, the Beech Hill anticline, deformed the Chesham Pond thrust, and it is in the core of this anticline that the nappe-stage syncline is exposed.

A similar interpretation can be extended northward. The "Kear­sarge-Central Maine synclinorium" (Lyons et al., 1982), along which the New Hampshire sequence is exposed beneath a sheet of Kinsman Granite, is proably the same nappe-stage syncline exposed because of a younger backfold anticline. According to this model, the Fall Mountain nappe and the Kinsman Granite must be rooted east of the Kearsarge-Central Maine synclinorium. Unfortunately, the latter name is confusing. It refers to one of several stratigraphic synclines, which are nappe-stage features with Littleton Formation in their centers. I propose calling it simply the Kearsarge syncline, with the understanding that it is a nappe-stage syncline equivalent to the Monadnock syncline, and that it may extend as far north as central Maine. The younger structural anticline should be given a different

~

137

name, perhaps using the name Beech Hill anticline from the Monadnock area. The Littleton Formation at Mt.Wachusett, Massachusetts (Tucker and Robinson, 1976-1977), in the Peterborough quadrangle (E. Duke, 1984) and in the Alton-Berwick area (Eusden et al., 1984) apparently belong to the next higher nappe-stage syncline, above the Fall Mountain level.

As mapping has proceeded during the past twenty years some dis­cussion has centered around the exact lGcation of the axial surface of the Merrimack synclinorium. Because the rocks of the Merrimack trough were involved in nappes which were later crumpled by backfolds, there is no clearly defined structural synclinorium between the Bronson Hill anticlinorium and the Massabesic terrane (Lyons et al., 1982). However, the term "Merrimack synclinorium" can still be applied to the broad, complexly folded area of Silurian and Devonian rocks.

The Monadnock sequence does not appear at Fall Mountain because the thrust fault under the Fall Mountain nappe has cut out, or beheaded, the Monadnock syncline. One place where these rocks may reappear is at Gee Mill, where rocks very similar to those at Monadnock are exposed in an anticline flanked on either side by Kinsman Granite and Bethlehem Gneiss (Chamberlain, 1984). Chamberlain has mapped a continuation of the Ch~sham Pond fault into the Lovewell Mountain quadrangle (Chamberlain, in progress), but the location of the Brennan Hill fault north of the Monadnock quadrangle is less certain. Another place where the Chesham Pond fault may be exposed lies west of the Connecticut Valley border fault. There, the Ashuelot pluton of Kinsman Granite is exposed in the hanging wall of the Meso­zoic fault, in the Fall Mountain nappe level (Thompson et al., 1968). Inverted units of the Monadnock sequence appear locally along the west edge of the pluton and it appears that a thrust fault may separate them from the rocks of the Bernardston nappe below (David Elbert, pers. comm., 1984).

The Brennan Hill fault probably extends south into the area west of the Tully body in Massachusetts, and more work in that area may improve our understanding of the thrust-nappe root zone. The Thorndike Pond fault zone probably also extends south into Massachu­setts, but because of backfold-stage mylonitization such as that at Brooks Village (Morton, 1984), the thrust-nappe root zone will be harder to locate.

"Can thy style-discerning eye The hidden-working builder spy, Who builds, yet makes no chips, no din, With hammer soft as snowflake's flight;. "

-Ralph Waldo Emerson, 1847, from the poem "Monadnoc"

METAMORPHISM

INTRODUCTION

138

Regional metamorphism has affected all the layered rocks, as well as some of the intrusive rocks, in the Monadnock quadrangle. The quadrangle lies between two of the highest grade Acadian metamorphic areas in New England, the central Massachusetts metamorphic high (Tracy et al., 1976b; Robinson et al., 1982b; Zen et al., 1983), and a smaller-area in south central New Hampshire (Chamberlain and Lyons, 1983). The peak of metamorphism produced predominantly "upper sillimanite" zone mineral assemblages in the pelitic schists: Zones III (sil-mus-gar-biot) and IV (sil-mus-gar-biot-ksp) of Tracy, 1975. (Mineral abbreviations are explained in Appendix 1, Table 16.) Zone II (sil-mus-st-gar-biot), Zone V (sil-gar-biot-ksp), and Zone VI (sil-gar-biot-ksp-crd) assemblage~ are locally present, and extensive retrograde processes have affected the rocks in much of the quadrangle. Calc-silicate rocks contain diopside and grossular garnet, indicating metamorphic conditions consistent with sillimanite-bearing pelitic assemblages (Thompson and Norton, 1968). Amphibolite facies assemblages in the Ammonoosuc Volcanics (Fowler­Billings, 1949a) are also consistent. I have done no further work on the mafic rocks.

PELITIC ROCKS Aluminum Silicate Polymorphs

Thompson and Norton (1968) and Chamberlain and Lyons (1983) showed the trace of a fossil isobaric surface crossing southwestern New Hamp­shire along the east edge of the Monadnock quadrangle. This line marks the approximate former position of the triple point between the three aluminum silicate polymorphs, kyanite, andalusite, and silli­manite. Kyanite was transformed to sillimanite west of the line, and andalusite was replaced by sillimanite to the east. However, silli­manite polymorphs after andalusite ("andalumps", Robinson, in Hatch et al., 1983) are abundant in much of the quadrangle (Figure 30). Therefore the triple point trace must lie farther west than the position shown by previous authors. Neither relict andalusite nor relict kyanite has been found in the Monadnock quadrangle. Edward Duke (1984) reported several samples from the Peterborough quadrangle which contain both andalusite and sillimanite. Kyanite, locally with fibrolitic overgrowths, has been found in the Ammonoosuc Volcanics at the south end of the Keene dome (Robinson, 1963; Robinson et al., 1982b). Sillimanite pseudomorphs after kyanite are apparently a rare

m

\ \ \ \ li\

" '\ \ n\

I \ I

n I I I I I I I I

TIJ

0' I

I I

I I

m (N)

I'il , I :~--, r• -~Derby Hill , I'il'' ,' window

' f" I _...... ' , ..... I UYJ

Til\\- / I I J (

N : 11 lll __ ,......"'ii( \'21.

,~I/........ --~ I,~ -.....;

~ ,..-11 m '- 0- I

"I( ~1,.

;.oJ'

m

m

ill

m

z f' I ~

m II II II II ,, N II

m m I 1/ II

m (ID)

Oil)

~

( Jl \ \.../

•* m i·" "'*

...

139

KEY

Roman numerals-Zones, explained in text.

( ) strongly retro­

m

*

om* (IDJ*

(ID)

graded rocks. * chloritoid • retr. staur. • Zone III

cordierite­bearing assem­blage.

1 2mi .

0 2 3km

Fig. 30. Metamorphic zones based on assemblages observed in thin sec­tion, or staurolite in hand sample (Zone II). Includes data from Chamberlain (1981). Plutons outlined for reference.

140

phenomenon (Tracy and Robinson, 1980). Thus the trace of the fossil isobar can only be located approximately as a line along the west edge of the area that contains andalumps. My approximation is shown as a heavy double line in Figure 30. Although there are nubbles of fibrolite up to about one centimeter long west of t nis line, they are smaller than the andalumps, and lack the microscopic mosaic texture of subparallel sillimanite prisms described by Rosenfeld (1969) in a sample from Gap Mountain.

Chamberlain and Lyons (1983) recognized a three-stage sequence of metamorphism in southwestern New Hampshire: (1) early production of andalusite-bearing assemblages, (2) peak metamorphism producing silli­manite-bearing assemblages of Zones III through VI, and (3) local retrogression. Tracy and Robinson (1980) and Robinson et al. (1982b) proposed P-T trajectories to show the different paths taken through time by rocks in the Bronson Hill anticlinorium and in the Merrimack synclinorium in central Massachusetts (Figure 31). During the progress of metamorphism all the rocks experienced increases in pressure and temperature, but those in the Merrimack synclinorium followed a path on the low pressure side of the aluminum silicate triple point while rocks farther west stayed on the high pressure side. Stages (1) and (2) of Chamberlain and Lyons thus reflect the trajectory from the andalusite field into the sillimanite field. The model which Tracy and Robinson proposed to explain the different trajectories, the trajectory proposed for the Monadnock area, and the timing relative to deformational history, are discussed in later sections of this chapter.

Metamorphic Zones

Mineral assemblages representing Zones II, III, and IV of Tracy (1975) are the most important in the Monadnock quadrangle. They reflect a general increase in metamorphic grade from west to east (Figure 30).

Zone II. The prograde Zone II assemblage, sillimanite-muscovite­staurolite-garnet-biotite-quartz, is present mainly in the belt of Littleton Formation east of the Keene dome (for example, Table 6a, SZ-27). Staurolite grains attain lengths of 5 mm, and show typical seive textures with abundant inclusions. Sillimanite is present as tiny prisms included within quartz, muscovite, and garnet, or as masses of fibrolite. This assemblage can be portrayed in a projection of mineral phases from muscovite (Thompson, 1957) onto the AFM plane of the AKFM four component tetrahedron (Figures 32a and 32h). Staurolite also occurs as tiny prisms in several samples from else­where in the quadrangle, but these are apparently of retrograde origin and are discussed later.

The sillimanite-staurolite-biotite field (Figure 32a) would shift progressively toward more Fe-rich compositions as the prograde reaction,

...

6

p kb

3

500

I I

I

600

T °C

I I

I

700

Fig. 31. P-T trajectory for rocks southeast of Mt. Monadnock, compared to trajectories for Bronson Hill anticlinorium and Merrimack synclinorium rocks in central Hassachusetts (after Robinson et al., 1982b). Aluminum silicate triple point from Holdmvay (1971). Staurolite-out reaction from Dutrow and Holdaway (1983).

141

A

Zone II +mus

• qtz

. . ·.· .. . . . ·.: ~ .

Fig. 32a. Muscovite projection, showing Zone II assemblages, using data from Hall, 1970, and Tracy et al., 1976b, sample 36Y.

Zone m +Sil + qtz

Zone m

ksp

+mus + qtz

142

Fig. 32b. Muscovite projection, showing Zone III assemblages, using data from MK-432 (Table 12).

Zone ISr

gar

+mus + qtz

F ~--""g'-o-r-ne"'"t'--------c~r~d---...--J>M .::·:.;.:,·.<·:.::··::·-::::·:-:·:-:-:=.::.:: bio·t: ::>:: ::,:'::-:·."·:·::.:::.:.:;:·:':=:·::.-: :_: . . . . . . . . . · ,L. ·· ......... . ... .

Fig. 32c. Sillimanite projection, showing Zone III assemblages from both MK-432 and MK-629 (Table 12).

ksp

Fig. 32d. Huscovite projection, showing Zone IV assemblages, using data from Tracy et al., 1976b, sample 871.

Fig. 32. Chemographic representation of mineral assemblages in Zones II-IV, based on microprobe data from central Massachusetts (32a, d-g) and from this paper (32b and c). AKFM tetrahedron from which the ternary diagrams are constructed is shown in 32h.

Zone nz tkSp +Qtz

Fig. 32e. K-feldspar projection, showing same assemblages as in 32d (Tracy et al., 1976b, 871).

Zonelll tksp + qtz

Fig. 32g. K-feldspar projection, showing Zone VI assemblages, using data from Tracy et al., 1976b. sample FW154.

Fig. 32 (continued)

M

Zone ll: + ksp + qtz

143

Fig. 32f. K-feldspar projection, showing Zone V assemblages, using data from Tracy et al., 1976b, sample 067D. ·

AKFM tetrahedron

F

M

Fig. 32h. AKFM tetrahedron from which projections in 32a-g were constructed. A = Al

2o

3, K = (K,Na)

Al02 , F = FeO + MnO, M = MgO. Quartz is in excess and H20 is perfectly mobile.

144

[st + mus + qtz = Fe-richer st + biot + sil + H2o] (1)

proceeded. Staurolite would finally be consumed by the reaction:

[st + mus + qtz = biot + sil +gar+ H2o]. (2)

The apparent coincidence of the staurolite-out isograd and the Little­ton-Rangeley contact east of the Keene dome may not accurately reflect a change in metamorphic conditions, but may be due to the absence of bulk compositions appropriate for the formation of staurolite in the Rangeley Formation. If the isograd does indeed follow the contact, which I have interpreted as a nappe-stage fault, then the fault may have juxtaposed rocks of different grade after the assemblages had reached equilibrium. However, since there are no metamorphic grades missing, the assemblages may have resulted after the faulting had set up the necessary thermodynamic gradients.

The staurolite-out reactions are temperature dependent, and lie on the high-temperature side of the aluminum silicate triple point. Thus in rocks at pressures near the triple point, sillimanite would be the polymorph involved in staurolite-destroying reactions (Figure 31). At higher pressures, kyanite would be the stable polymorph. Kyanite and andalusite had presumably been replaced by the time staurolite was consumed in most of the Monadnock quadrangle.

Zone III. The Zone III assemblage, sillimanite-muscovite-garnet­biotite-quartz ~ plagioclase, is the most widespread in the quad­rangle, and occurs in most of the pelitic schists in the Rangeley, Perry Mountain, and Littleton Formations (Figure 32b). In rocks where retrogression has destroyed most of the sillimanite (e.g. MK-216 and TR-20 in Table 6a), inclusions in quartz and primary muscovite attest to its former presence. Both fibrolitic and prismatic sillimanite are present, with the prismatic proportion apparently increasing toward the northeast.

Two samples of schist were selected for detailed analyses of minerals using the electron microprobe on polished thin sections. One of these, MK-432, is a typical gray-weathering schist of the Littleton Formation with a Zone III mineral assemblage of muscovite, biotite, quartz, garnet, sillimanite and plagioclase, and accessory graphite, ilmenite, apatite, and zircon. This is shown in muscovite projection in Figure 32b, and as assemblage (1) in sillimanite projection in Figure 32c, using mineral compositions from the microprobe analyses. Assemblage (2) in Figure 32c represents MK-629, a cordierite-bearing schist from an outcrop approximately 120 m from MK-432 (locations shown in Figure 24), in the upper part of the Warner Formation. MK-629 consists of quartz, biotite, plagioclase (An55 ), garnet, cordierite, and sillimanite, with accessory graphite, ilmenite, apatite and zircon. It contains neither muscovite nor K-feldspar, and thus occupies a different chemical space in Figure 32c than MK-432, but with tie line arrangements consistent with an identical P-T

145

history. Garnet zoning and estimates of metamorphic conditions from these two rocks are discussed in a later section. The outlines of garnets in both tend to be ragged, suggesting that they were involved in garnet-consuming reactions at some point. Garnet would be consumed by one such reaction in the more typical schist:

[gar + mus = sil + biot + qtz]. (3)

A rise in temperature alone, a drop in pressure, or a rise in tempera­ture with very gradual increase in pressure, would tend to drive this reaction to the right. It could proceed to the left only so long as pressure increased at a sufficiently high rate. Garnet may have been consumed in the less typical schist by the pressure-sensitive reaction:

[gar+ sil + qtz = crd]. (4)

Some sillimanite might also have been produced in Zone III as musco­vite was depleted in paragonite content according to the reaction:

[Na-bearing mus + ab + qtz = K-richer mus + sil + K-richer ab +H2o]. (5)

Zone IV. The appearance of K-f~ldspar by the reaction

[mus + ab + qtz = ksp + sil + H2o] (6)

marks the orthoclase or "second sillimanite" isograd, forming the assemblage: sillimanite-muscovite-garnet-biotite-K-feldspar­plagioclase (Figures 32d and 32e). Some Zone IV assemblages were found in rocks north of the Chesham Pond fault, and from inside the Derby Hill window. Textures suggest that the rocks attained condi­tions favorable for Zone IV assemblages relatively early in the metamorphic history. There are roundish zones of quartz and feldspar surrounded by confused masses of biotite, sillimanite, retrograde Mg-chlorite, and muscovite, some of which is almost certainly retrograde. This matrix lacks any strong foliation. Garnets in many cases are surrounded by biotite and sillimanite. There are also roundish masses of biotite and sillimanite which appear to have replaced some mineral, perhaps garnet or cordierite. In the thin sections which contain orthoclase there are commonly small myrmekitic intergrowths of plagioclase and quartz adjacent to the K-feldspar. Biotite and quartz also occur locally in myrmekitic intergrowths. The massive, gneissic texture may have developed while the retrograde reactions were in progress, erasing any previous strong foliation that may have been present. At Otter Brook Dam (KN-2, Figure 30), pristine orthoclase augen cut across a pervasive foliation. Heald (1950) reported small inclusions of gneiss inside feldspars, with foliation parallel to that in the matrix.

Many of the gray schists in the Rangeley Formation north of the Chesham Pond fault contain augen averaging four centimeters in

146

diameter. In thin section these are seen to be dominantly quartz and muscovite (Table 1d, RX-2A). They may represent orthoclase porphyro­blasts which were retrograded by reaction (6) running toward the left. Heald (1950, p.75) reported a large area west of the Cardigan pluton in which "all stages in the transition from orthoclase to aggregates of muscovite and quartz" can be observed. He showed on his map that secondary muscovite occurs throughout most of this area, except for a fringe in the west near the orthoclase "isograd", where orthoclase occurs with primary muscovite only. More field work and thin section work are needed in the Monadnock quadrangle to define the silli­manite-orthoclase isograd. Identifying the first traces of K-feldspar in pelitic schists near the isograd is commonly difficult.

The orthoclase augen at Otter Brook Dam, which lies west of the Chesham Pond fault, indicate the isograd does not exactly follow the fault. A comparison of Figures 13 and 30 shows that farther east it is approximately parallel to the fault. Zone IV assemblages were not found in the Monadnock quadrangle in the supposed level (3) area east of the Thorndike Pond fault zone, but farther east, an isolated locality was reported by Chamberlain and Lyons (1983) in the south­western part of the Peterborough quadrangle (Figure 30).

It is important to distingui~h augen that were once K-feldspar porphyroblasts from quartz-feldspar segregations (e.g. FZ-30, Table 1d), which are distributed much more widely, and probably represent local melt pockets. Tracy (1978) interpreted partial melting and muscovite dehydration as nearly simultaneous processes in central Massachusetts. The fact that quartz-feldspar segregations seem to be more widespread than augen after K-feldspar, occurring in level (2) as well as in level (3), suggests that local melting may have occurred earlier than the muscovite dehydration reaction (6). At least some of the quartz-feldspar segregations are deformed by what I have interpreted to be the earliest backfolds on Mt. Monadnock.

Some garnets in the area east of the Thorndike Pond fault zone (e.g. in MK-1061A, Table 1d) have inclusion-rich cores surrounded by clear rims. The rims may represent prograde overgrowths formed perhaps during the prograde reaction,

[biot + sil + qtz = ksp + gar + H2o], (7)

around garnets that had been partially resorbed in earlier, lower-grade reactions. It would be interesting to probe some of these garnets and compare zoning patterns to those in Zone III garnets.

Zone V. The final destruction of muscovite results in the typical Zone V assemblage, sillimanite-garnet-biotite-K-feldspar (Figure 32f). Because of the problems in differentiating prograde muscovite from secondary muscovite, I am certain of only one Zone V assemblage in the thin sections studied. This occurs in a rather peculiar coticule­bearing schist from Hurricane Hill in the Perry Mountain Formation

147

(DB-289A, Table 3).

Near the K-feldspar there are some small patches of green-brown biotite and sillimanite, which may be evidence for reaction (7) in reverse. There are also clumps of red-brown biotite in 5 mm patches, and garnets embayed by biotite and sillimanite. Some of these patches might have replaced cordierite.

Zone VI. Cordierite may appear in normal pelitic schists through the reaction:

[biot + sil + qtz = gar + crd + ksp +H2o]. (8)

There are several occurrences of the Zone VI assemblage, sillimanite­garnet-biotite-K-feldspar-cordierite (Figure 32g) in the Monadnock quadrangle. Two of these are Fowler-Billings' samples K-116A and K-117 (Table 1d), both from Cobb Hill, where the western contact of the Cardigan pluton jogs to the east and back again. The cordierite has yellow alteration along cracks, yellow pleochroic haloes around zircon, and is surrounded by nests of sillimanite prisms and fine­grained red-brown biotite. The K-feldspar has a perthitic texture. Both these samples lack muscovite. K-102, also from Cobb Hill, was collected from the same sillimanite-rich unit as my sample HV-41, and although it, too, contains retrograde muscovite and chlorite, it contains K-feldspar and cordierite as well. K-106, from along the contact to the southwest, is similar to K-102. Sample DB-130 may contain some cordierite, but there is no yellow alteration or haloes to distinguish it from plagioclase. The presence of cordierite mainly along the Kinsman contact suggests a localized contact metamorphic effect. This does not necessarily contradict Chamberlain and Lyons' (1983) conclusion that the pattern of peak regional metamorphism cuts across the Cardigan pluton.

Assemblages in Sulfidic Schists

Mineral assemblages in the sulfidic schists of the Rangeley t Formation are similar to those in the gray-weathering rocks, with the

addition of pyrrhotite (Table 1a). The sulfidic mica schists of the Francestown, however, are distinctive in that they are so sulfidic that most of the iron is taken up by the sulfides (Table 4). As a consequence, garnet is absent and biotite is a pale brown to white Mg-rich variety. No chemical study was made for Francestown rocks in the Monadnock quadrangle, but Field (1975) reported pure end-member Mg-cordierite, and biotite with only .04% FeO, from correlative rocks in Zone VI in central Massachusetts. Robinson et al. (1982b) discussed the effect of sulfides on pelitic schist-assemblages in detail, with particular attention to the role of volatile components. In bulk compositions rich in sulfur, rutile is the stable Ti-bearing phase rather than ilmenite. Biotite in equilibrium with graphite and pyrite becomes more and more Mg-rich as the following reaction proceeds:

148

[py + biot + gr = po + Mg-richer biot + sil + ksp +H2o+ co2]. (9)

The Francestown schists from the Monadnock quadrangle contain rutile, pyrrhotite, and a non-magnetic sulfide which is probably secondary marcasite derived from pyrrhotite (Steven Haggerty, pers. comm., 1982). Probably none contains the pyrite that has been reported from sulfidic rocks of extreme composition at several localities in central Massachusetts (Robinson et al., 1982b).

Evidence for a Retrograde Episode

Retrograde metamorphic assemblages in the Monadnock quadrangle are of two sorts. The first sort is apparent in nearly all areas, and it involves the partial or total replacement of sillimanite by muscovite, and of K-feldspar by muscovite and quartz. The second sort occurs mainly in the area northeast of Mt. Monadnock, and represents more advanced, lower-temperature re-hydration. In the latter, Fe-chlorite has partially replaced garnet, and secondary muscovite and chlorite are major components of the rock. Biotite is absent in many of these rocks (Table 1a, DB-165; Table 6, MK-216, DB-10, DB-69, DB-171, MK-210, and DB-17). Hollocher (1981) studied the detailed chemistry and textural relations of retrograde minerals in the Littleton Formation near New Salem, Massachusetts. Hollocher's methods could be applied to explain some of the apparently complex retrograde assemblages of the Monadnock area. Only the most salient features and problems will be discussed here.

As mentioned earlier, the coarse gneissic textures of the Zone IV rocks north of Chesham Pond fault seem to be at least in part due to widespread retrograde metamorphism. It would be interesting to study the quartz-muscovite augen in some detail to see if there is chemical evidence to support the textural evidence cited by Heald (1950) for replacement of K-feldspar porphyroblasts. A possible reaction might be

[ksp + sil + H2o = mus + qtz], (10)

but the role of Na in this would have to be explored. Heald noted that plagioclase is involved in some cases.

Sillimanite is partially to completely replaced in much of the quadrangle, mainly by muscovite. Hollocher (1981, p.178-179) proposed the following reaction, balanced according to his microprobe analyses,

149

for the destruction of sillimanite:

[45 biot + 51 sil + 9 qtz + 86 H2o = 41 mus + 22 chl + 4 ilm]. (11)

The chlorite produced is much more Mg-rich relative to the remaining biotite. Although no microprobe work was done on Monadnock retrograde rocks, it was noted that the chlorite intergrown with biotite and secondary muscovite has gray to brown interference colors, indicating a Mg/Fe ratio greater than one. Chlorite associated with garnet, by contrast, shows the anomalous blue interference colors characteristic of Fe-rich chlorite. Some rocks contain both Fe- and Mg-chlorite, implying disequilibrium, which is a different situation from that reported by Hollocher (1981). He proposed three reactions for the destruction of garnet, which would vary according to the garnet zoning. These reactions are essentially the simple hydration reaction,

[gt +H2o = chl + qtz], (12)

with more or less ilmenite and sphene involved depending on the calcium content of the garnet, and minor amounts of biotite and muscovite. These low-temperature garnet-consuming reactions involved extensive re-hydration, and took place only locally in the Monadnock area, in contrast to the more subtle and higher-temperature retrograde ion exchange reactions, which may have resulted in changes in garnet rim compositions. These are discussed in a later section.

Many of the rocks on Mt. Monadnock and the area to the northeast contain chloritoid in addition to Fe-chlorite (Figure 30). The chloritoid appears as dark green clots in hand sample, which commonly weather out to form roundish pits. Many of the chloritoid-bearing rocks also contain unaltered garnet and tiny (<0.5 mm) staurolites. A thin section from sample MK-210 (Table 6b, MK-210A), displays an andalump that contains relict patches of sillimanite that all have the same optic orientation, surrounded by an intergrowth of fibrolite and muscovite. Staurolite and chloritoid cut across the other minerals. Plagioclase and tourmaline are more abundant within the andalump than in the matrix. The staurolite, chlorite, chloritoid, and muscovite are probably all retrograde reaction products from the breakdown of biotite and sillimanite.

Outcrops on Beech Hill contain large clumps of chloritoid. In thin section (Table 6a, DB-10) grains up to 10 mm long have ragged boundaries and appear to be in disequilibrium, while much smaller grains cut across the older (nappe-stage?) fabric. There were apparently two periods of chloritoid growth, the first perhaps by replacement of a relatively Fe-rich cordierite and the second by a reaction involving sheet silicates. Since there is no biotite left in the rock, the second chloritoid-producing reaction might be similar to ones proposed by Hollocher (1981) involving garnet, chlorite and muscovite, such as:

[gar+ H2o = chl + ctd], (13)

[mus + chl = more phengitic mus + ctd], (14)

and [ctd +gar+ H2o = Mn-richer ctd + chl]. (15)

The final destruction of biotite in severely retrograded rocks may have resulted mainly in the production of chlorite, along with cela­donite-richer muscovite and ilmenite (Hollocher, 1981):

[mus + biot + qtz +H2o = mus + chl + ilm]. (16)

Garnet Zoning in Zone III

150

Detailed microprobe analyses of garnets in thin sections MK-432, MK-6 and MK-629 revealed interesting zoning patterns which, if they can be correctly interpreted relative to reactions involving garnet, should theoretically provide information about metamorphic conditions based on Fe and Mg fractionation between biotite and garnet and between cordierite and garnet. Chemical zoning in garnets is believed to be due to garnet's slow ionic diffusion rate, and thus should reflect changes in metamorphic conditions through time. Repre­sentative mineral analyses are presented in Table 12. The garnet compositions were recalculated in terms of molecular proportions of pyrope (Mg-Al garnet), almandine (Fe-Al garnet), spessartine (Mn-Al garnet), and grossular (Ca-Al garnet). All the iron is assumed to be FeO.

Assemblage (1). MK-432 is a typical pelitic schist with the Zone III assemblage mus-biot-qtz-gar-sil, with minor amounts of plagioclase, graphite, ilmenite, and accessory minerals (Table 6a). Two garnets were probed in a thin section from sample MK-432 (Figure 33). The thin section cuts approximately through the center of garnet X, but only through a portion near the rim of garnet Y. The garnets are zoned, and because of the orientation of the sections, garnet X shows the full range of compositions from core to rim, whereas variation in garnet Y is much more restricted. Although there are only a few data points near the biotite-free rim of garnet X, the central part of the cut through garnet Y is believed to also represent a biotite-free rim. Pyrope increases toward the rim from about 14 to about 17%, except where the garnet touches biotite, and there pyrope decreases abruptly down to as little as 9%. Almandine is fairly constant except toward rims touching biotite, where it rises from about 76% up to as much as 79%. Spessartine follows a pattern inverse to that of pyrope, decreasing toward the rim from 5 to 3%, except next to biotite, where it increases abruptly to as much as 8%. Grossular varies from 2.8 to 3-7% with a much less clear pattern, and so was omitted from Figure 33. The lower values of grossular tend to be toward the rim.

··~

c ____ ..._

........

((. 0~~:-y.~> ) . '• • , & . /

' 15 ......... ...._ __ _

Pyrope

Mg

Almandine

Fe

0 I 2mm

Spessartine

Mn

Fig. 33. Contoured values of atomic percent of Mg, Fe, and Mn in garnets that some contours are omitted near the rims for clarity of presentation. 2.8 to 3.7 with no clear pattern. Dots show microprobe analysis points. indicate biotite.

from MK-432. Note Ca varies from

Par::tllel lines

"'"" Vl

"'""

152

Table 12. Electron microprobe analyses from minerals in Zone III schists and coticule.

Garnet Sample No. tfK-423 rim next rim next Location ~ core rim rim to biotite to biotite Analysis No. Bw Ik' Ag A a Bl Bv

Si02 36.16 39.03 38.19 36.57 36.97 36.20 Al 2o3 21.77 20.18 20.04 20.43 22.09 22.55 MgO 3.67 3.50 4.20 3.86 2.63 2.20 FeO 34.89 35.17 34.95 35.63 35.76 34.96 MnO 2.43 2.56 1. 34 1.49 3.64 3.59 CaO 1. 22 1. 21 1. 22 1.15 1. 13 1.23 Sum 100. 14 101.65 99.94 99.13 102.22 100.73

Structural Formulae (based on 12 oxygens)

Si 2.915 3.084 3.059 2.979 2.935 2.912 Al 2.069 1.880 1. 893 1. 959 2.068 2.139 Mg .441 .413 .502 .470 .311 .264 Fe 2.354 2.325 2.343 2.430 2.376 2.353 Mn .166 .172 .091 .103 .245 .245 Ca .106 .103 .105 .100 .097 .106 Total 8.051 7. 977 7.993 8.041 8.032 8.019

% Pyrope 14.4 13.7 16.5 15.2 10.3 8.9 % Amandine 76.7 77.2 77.0 78.3 78.4 79.3 % Spessartine 5.4 5.7 3.0 3.3 8.1 8.2 % Grossular 3.5 3.4 3.5 3.2 3.2 3.6 Total 100.0 100.0 100.0 100.0 100.0 100.0

Mg/ (Hg+Fe+Mn) . 149 .142 .171 .157 .106 .092

t Hg/(Hg+Fe) .158 .151 .176 .162 .116 .101

153

Table 12. (cont'd)

Garnet Sample No. MK-629 MK-6

rim rim Location core near crd near biot core rim Analysis No. A-C A-a A-s A4-l A4-w

Si02 36.16 34.61 36.56 38.25 37.37 Al203 21.39 22.77 21.92 21.75 21.94 MgO 6.13 4.56 3.97 2.19 1. 73 FeO 32.35 32.87 34.20 30.16 32.67 UnO 2.09 1. 98 1.86 6.06 5.35 CaO 1.38 1. 73 2.01 1.06 1. 55 Sum 99.50 98.52 100.52 99.47 100.61

Structural Formulae (based on 12 oxygens)

Si 2.900 2.818 2.921 3.065 2.997 Al 2.023 2.187 2.066 2.056 2.074 Mg . 733 .554 .473 .262 .206 Fe 2.171 2.240 2.287 2.022 2.192 Mn .142 .137 .126 .411 .363 Ca .119 . 151 .173 . .091 .133 Total 8.088 8.087 8.046 7.907 7.965

% Pyrope 23.2 18.0 15.5 9.4 7.1 % Almandine 68.6 72.7 74.8 72.6 75.7 % Spessartine 4.5 4.4 4.1 14.8 12.5 % Grossular 3.7 4.9 5.6 3.2 4.7 Total 100.0 100.0 100.0 100.0 100.0

Mg/(Mg+Fe+Mn).241 .182 .164 .097 .075

Mg/(Mg+Fe) .253 .198 .171 .115 .086 ...

15Lf

Table 12. (cont'd)

Biotite Sample No. MK-432 HK.-629

near near Location matrix inclusions garnet garnet matrix Analysis No. C-3a B-2b B-1a NB-6 A-f A-b

Si02 33.41 37.85 36.37 35.95 36.80 37.13 Ti02 2.34 1.84 3.03 2.14 1.61 2.02 Al 2o3 21.03 21.01 21.19 21.59 19.83 19.36 cr2o3 .06 .08 .12 .02 .13 .11 FeO 20.05 15.38 15.40 19.47 16.88 17.43 MnO .08 .02 .01 .05 .04 .06 MgO 9.46 12.31 11.79 10.03 13.39 12.81 ZnO .14 .11 .07 .04 0 0 Na2o .05 .32 .23 .15 .17 .06 K2o 9.22 8.71 9.62 9.29 7.81 8.28 Sum 95.84 97.63 97.83 98.73 96.66 97.26 H2o 4.16 2.37 2.17 1. 27 3.34 2.74

100.00 100.00 100.00 100.00 100.00 100.00

Structural Formulae (based on 11 oxygens) A-site K .895 .800 .890 .865 .729 . 771 Na .008 .045 .032 .021 .024 .008 Sum .903 .845 .922 .886 .753 .779 IV-site Si 2.542 2. 723 2.635 2.623 2.689 2.709 Al 1.458 1.277 1.365 1.377 1.311 1.291 Sum 4.000 4.000 4.000 4.000 4.000 4.000 VI-site Al .429 .505 .446 .481 .398 .375 Fe 1. 276 .926 .934 1.189 1.032 1.064 Mn .005 .001 .001 .003 .003 .004

! Mg 1.073 1.320 1. 274 1.092 1.459 1.394 Zn .008 .006 .004 .002 0 0 Cr .003 .005 .007 .001 .008 .007 Ti .134 .099 .165 .117 .088 .111 Sum 2.929 2.862 2.831 2.885 2.988 2.955 Total 7.831 7.707 7.753 7. 771 7.741 7.734

Mg/(Mg+Fe) .457 .588 .577 .479 .586 .567

155

Table 12. (cont'd)

Sample No. MK-629 MK-629 Mineral Cordierite Cordierite Plagioclase Location core near garnet Analysis No. A-f A-e C-b

Si02 48.68 46.64 Si02 54.79 Al2o3 32.18 31.13 Al2o3 29.59 Tio 2 0 0 BaO .38 MgO 9.19 9.44 FeO .02 FeO 7.53 6.81 CaO 10.17 MnO .07 .08 Na2o 4.63 CaO 0 0 K2o 0 Na20 .18 .09 Total 99.58 K2o 0 0 Total 100.21 99.77

Structural Formulae (18 oxygens) (8 oxygens)

Si 5.020 4.989 IV-site Al 3.914 3.927 Si 2.469 Mg 1. 414 1.506 Al 1.573 Fe .650 .609 Sum 4.042 Mn .006 .007 A-site Na .036 .018 Ba .007 Total 11.040 11.056 Fe .001

Ca .491 Mg/ (Hg+Fe) .685 .712 Na .405

Sum .904 Total 4.946

Ab.452An.548

t.:

I Mn

10

Fe

rim next to biotite

10

MK-432 garnet

20

Mg-+

156

30

Fig. 34a. Mn-Fe-Mg composition trend for garnet from MK-432. Black circles from core to "normal" rim; open circles from rim adjacent to biotite.

I Mn

~L_----~~----~~----~------~~~~~-----10 20 30 40 biotite 50

Mg-+

Fig. 34b. Compositions of zoned garnet and coexisting biotite in MK-432. Arrow points from core toward rim. Biotite compositions have a small composition range, with slightly more Mg next to garnet.

157

The composition trend from core to rim can be seen in Figure 34 which is a portion of an MnO-FeO-MgO ternary plot. Each point corre­sponds to a point on the contour map of the garnets in Figure 33. One possible interpretation for this trend involves two stages of zoning. The decrease in MnO and increase in MgO/(MgO + FeO) toward the rim may have resulted from garnet growth during the prograde continuous reaction,

[ st + biot + qtz = mus +gar+ H2o ], (17)

for which T Mg > T Fe > T Mn (Tracy et al. , 197 6 b) • After staurolite was consumed, garnet growth stopped and the garnet was locally resorbed according to the reaction,

[ mus +gar = biot + sil + qtz ], (3)

for which T Mn > T Fe > T Mg. This reaction took place mainly where the garnet was in contact with muscovite and biotite, resulting in the zoning trend toward higher MnO and lower MgO/(MgO + FeO) especially next to what is now biotite. Tiny sillimanite prisms are locally present in the adjacent garnet rim. The localized reversed zoning next to biotite resembles somewhat that observed in garnets in Zone VI in Massachusetts which has been attributed to local retrograde cation exchange (Richardson, 1975; Tracy et al., 1976b). However, the MnO enrichment in MK-432 garnets is much more extreme than in those garnets, and ion exchange alone could not produce such large MnO enrichment.

Figure 34b shows a simplified version of the garnet zoning path. The sharp hook in the path represents the transition from Zone II to Zone III assemblage conditions. The ragged edges of many garnets in Zone III rocks may be . added evidence for a stage of garnet resorption. The different biotite compositions shown in Figure 34b may not be significant; biotite grains in the matrix have an average of 45.7 MgO/(MgO + FeO) (range 44.7 to 47.0 in 14 points analysed), compared to 46.7 (range 45.0 to 48.7 in 17 points) in grains adjacent to garnet.

Assemblage (2). MK-629 is a K-poor schist from an outcrop 120m from MK-432. It contains the assemblage quartz-biotite- plagio­clase-garnet-cordierite-sillimanite, but no muscovite or K-feldspar. Garnet, biotite, cordierite, and other minerals in a thin section from MK-629 were analysed with the electron microprobe (Table 12). A garnet was studied in detail, as well as adjacent cordierite and biotite (Figure 35). The zoning in the garnet suggests that it was once two grains which have coalesced. Pyrope decreases toward the rim from 23% in the core to about 18% at contacts with cordierite and 15.5% at contacts with biotite. There is really only one small length of biotite-free rim along the lower right side. Almandine increases toward the rim from 68% to 73% and 75%. Spessartine decreases toward the rim, but then locally increases slightly at the very rim next to biotite. Grossular increases slightly from 3.2% to 5.7% toward the

'"'

Pyrope Almandine Spessartine

Mg Fe Mn

0 I 2mm

Fig. 35. Contours of atomic percent of Mg, Fe, and Mn in garnets from MK-629. from 4.0 to 5.7 with no clear pattern. Dots show microprobe analysis points. lines indicate biotite.

Ca varies Parallel

f-' I.J1 00

1 Mn

20

I Mn

MK-6 garnet

10

MK-629 garnet

~ ..... ~~ rim~ core

FeL_------------~1L0--------------~2~o~------------~3o~---­Mg---+

Fig. 36a. Mn-Fe-Mg composition trends for garnets from MK-629 (K-poor schist) and MK-6 (coticule). Black circles from cores; open circles from rims.

159

10~--------~~--------~----------~----------~---------

10

MK-629 garnet

20 30 40 Mg~

50 biotite 60 cordierite 70

Fig. 36b. Compositions from zoned garnet and coexisting cordierite and biotite in MK-629. Arrow points from core to rim. Biotite adjacent to garnet, Mg/(Mg +Fe) = .581, is slightly more Mg-rich than that in matrix, Mg/(Mg +Fe) = .565. Biotite next to cordierite is slightly more Fe-rich, Mg/(Mg +Fe) = .535.

160

rim.

The composition trend for MK-629 (Figure 36) is markedly different from the MK-432 trend. The MgO/(MgO + FeO) ratio decreases continu­ously, and spessartine shows a distinct slight decline with decreasing XMg• This is coupled with a slight increase in grossular content.

The cordierite grain in Figure 35 shows very little chemical vari­ation. For 11 points analysed, MgO/(MgO + FeO) ranges from 0.675 to 0.718. Biotite compositions also show very little variation, the MgO/ (MgO + FeO) ratios averaging 0.565 in the matrix with sillimanite and quartz. Ti per 11 oxygens ranges from 0.088 to 0.139, though better analyses have Ti below 0.115. MgO increases slightly to 0.581 next to the garnet, and decreases slightly to 0.535 next to cordierite, suggesting the possibility of local late retrograde ion exchanges. More analyses are needed to substantiate this data and test its significance. However, this does not explain the overall composition trend from the core outwards in the garnet.

Because the bulk composition of MK-629 is too K-poor to contain muscovite or K-feldspar, the reactions proposed for zoning in MK-432, or in the garnets studied by Tracy et al. (1976b), do not apply. The garnet-consuming reaction,

[ sil +gar+ qtz = crd ], (4)

would produce a slightly flatter composition trend on the MnO-FeO-MgO plot than a reaction involving biotite, because the cordierite is more Mg-rich than the biotite (Figure 36b). However, some other reaction must have been involved to allow the spessartine content to remain so constant rather than rising as is true of most garnet-consuming reactions. One possibility is that FeO was provided to the garnet by a reaction that consumed ilmenite, which is the only phase in this rock richer in FeO than garnet. The Ti02 would have to be taken up by biotite, in reactions such as

[3 Mg-Fe biot + 3 ilm + 5 sil = 3 Ti-biot + 2 gar + 5 qtz], (18)

or [3 Mg-Fe biot + 3 ilm + 9 sil = 3 Ti-biot + 3 crd]. (19)

The amounts of garnet and cordierite would depend on the progress of reaction. 4. A second possibility is that CaO was provided to the garnet by the reaction,

[3 an =gross+ 2 sil + qtz]. (20)

These possibilities are discussed further below.

Coticule garnet zoning. The composition trend for a garnet in sample MK-6 (Table 6b) is also shown in Figure 36a, and two repre­sentative analyses are in Table 12. The decrease in spessartine from

161

core to rim is characteristic of garnet growth zoning, but the trend of decreasing pyrope and increasing almandine toward the rim has not been identified in other rocks of the region. The garnets probably formed relatively early from chlorite or other sheet silicates, or possibly from carbonates. There are now small amounts of biotite and chlorite in the rock, but the exact garnet-forming reaction is unknown. The present composition of the tiny amount of biotite is certainly a result of retrograde reaction.

Temperature Estimates

Metamorphic temperatures were estimated from the distribution of Fe and Mg between garnet and biotite, and between garnet and cordierite, based on the calibrations of Thompson (1976) and Ferry and Spear (1978). A problem in estimating temperatures of metamorphism from distribution coefficients of mineral pairs is to decide which pairs, if any, represent pairs that formed in equilibrium during any stage of metamorphism. Zoned garnets provide information about changes in composition through time, but P-T evaluation can be difficult if the zoning resulted from slow diffusion during garnet-consuming reactions. Only mineral pairs that truly equilibrated during the metamorphic history will give "true" tempera­ture estimates. Temperatures and pseudotemperatures estimated from various mineral pair compositions are presented in Table 13. Values for ln KD are listed, where KD is the distribution coefficient equal to

(XGar Fe

XBiot) I (XGar . ~iot) -~g 11g -r'e '

or the analogous ratio for garnet and cordierite, followed by temperatures from both Thompson's (1976) and Ferry and Spear's (1978) calibration curves. Only those based on Thompson will be quoted in the text. The Ferry and Spear estimates should be used to compare

~ with temperatures on Chamberlain and Lyons' 1983 map. In the -discussion below, an effort is made to decide which estimates are reasonable, and which must be pseudotemperatures.

Sample MK-432. Figures 33 and 34 show that the garnet in this rock has two distinct zoning trends, one toward biotite-free portions of the rim, and a different trend toward the rim where it is in contact with biotite. In Figure 34a, there is a range in Mg/Fe ratios for each of the three groups labeled "core", "rim", and "rim next to biotite". Two garnet compositions from each of these groups have been selected for use in making temperature estimates. These six garnet compositions are given in Table 12.

Four biotite analyses are presented in Table 12. The average matrix biotites have Mg/(Mg +Fe) between 0.45 and 0.47, although they range from 0.447 to 0.487. Ti per 11 oxygens ranges from 0.098 to 0.138. Analysis NB-6, Table 12, is from a more Mg-rich biotite in

162

Table 13. Estimated temperatures from garnet-biotite and garnet-cordierite geothermometry.

T°C T°C (Ferry

(Thompson, and Spear, Sample MK-432 XMg ln KD-- 1976) 1978)

Mg-rich Garnet Core Bw .158 1.50 630 665 Matrix Biotite c C3a .457

Fe-Hn-rich Garnet Core Be' .151 1.55 610 630 Matrix Biotite C3a .457

Fe-Hn-rich Garnet Core Be' .151 2.08 485 465 Biotite Inclusion 2b .588

Mg-rich Garnet Rim Ag .176 1. 37 670 710 Matrix Biotite C3a .457

Fe-rich Garnet Rim Aa .162 1. 47 635 670 Matrix Biotite C3a .457

Fe-rich Garnet Rim A a .162 1. 95 515 500 Biotite Inclusion 1a .577

Garnet Rim against Biotite Bl .116 1. 86 530 525 Matrix Biotite C3a .457

Garnet Rim against Biotite Bv .101 2.01 500 480 Matrix Biotite C3a .457

Garnet Rim against Biotite Bv .101 2.10 480 455 Extreme Biotite NB-6 .479

SamEle HK-629

t Garnet Core A-C .253 1.36 670 715 Matrix Biotite A-b .567

Garnet Core A-C .253 1.86 715 Cordierite Core A-f .685

Garnet Rim Near Cordierite A-a .198 1. 65 580 595 Matrix Biotite A....;b .567

Garnet Rim near Cordierite A-a .198 2.16 620 Cordierite Core A-f .685

Extreme Garnet Rim near Biot. A-s .171 1.85 535 530 Matrix Biotite A-b .567

Garnet Rim next to Cordierite A-a .198 2.30 580 Cordierite near Garnet A-e .712

163

close proximity to garnet. The other two in the table are from tiny biotite inclusions within garnet X (Figure 33). Inclusion 2b is in the garnet core, and inclusion 1a is fairly close to the biotite-free rim. The latter is the most Ti-rich biotite yet found in sample MK-432.

Temperature estimates in Table 13 were made by comparing various garnet-biotite pairs. The two garnet core analyses (Bw and Be) compared with matrix biotite (C3a) give estimates of 630 and 610°C. These are certainly pseudotemperatures, because the matrix biotite has changed from what it was when the garnet core formed.

The garnet core Be' is compared with biotite inclusion 2b as well, giving an estimate of 485°C. This is probably the result of local retrograde Mg-Fe ion exchange between the inclusion and the surrounding host garnet. The Ti content of the inclusion, however, has probably not changed since the biotite was included, provided no Ti-bearing phases were available for re-equilibration. The Ti content of .099 Ti per 11 oxygens is typical of metamorphic Zone I in central Massachusetts (Robinson et al., 1982b, Fig. 22).

The two biotite-free garnet rim analyses (Ag and Aa) compared with matrix biotite (C3a) give temperature estimates of 670 and 635°C. These may not be pseudotemperatures. Although the garnet rims have a different composition where they to~ch biotite, these rims represent only a tiny proportion of the 16% garnet in the mode (Table 6a) available to react with a much larger amount of matrix biotite (24% of the mode). Thus the Mg/(Mg +Fe) in the matrix biotite may not have changed much from what was once in equilibrium with the biotite-free garnet rims. By contrast, biotite inclusion 1a has a Mg/(Mg + Fe) of about 0.577, because it has exchanged Fe for Mg with the enclosing garnet. This biotite inclusion compared with garnet rim Aa gives a temperature of 515°C. The high Ti per 11 oxygens of 0.165 in the inclusion, however, is typical of ilmenite-saturated biotites in Zones III and IV pelitic schists of central Massachusetts (Robinson et al., 1982b, Fig. 22).

Two garnet rim analyses next to biotite (Bl and Bv) were compared with matrix biotite C-3a, giving temperature estimates of 530 and 500°C. These are probably "true" temperatures of the last equilibrium between biotite and garnet rims. They are much lower than the 625-660°C temperature estimates for Zone III in central Massachusetts (Tracy et al., 1976b). The rea c tion responsible for the reversed zoning in garnet next to biotite was probably the continuous retrograde reaction

[gar+ mus = biot + sil + qtz], (3) which consumed garnet. The equilibrium was apparently maintained by diffusion through the sheet silicates rather than through a pervasive metamorphic fluid, or one would expect the reversed zoning in all parts of the garnet rim. Comparing garnet rim Bv with the Mg-enriched biotite NB-6 gives a still lower temperature of 485°C. This may be the temperature at

164

which very local ion exchange took place as the diffusion rate through biotite became too sluggish for reaction 3 to continue.

Sample MK-629. As described above, garnet zoning in this rock has one distinct trend (Figure 36). Three garnet analyses were selected for use in making temperature estimates, one from the core (A-C, Table 12), one from the rim near cordierite (A-a), and one from the rim near biotite (A-s). Matrix biotite is represented by analysis A-b, and a more Mg-rich analysis from near garnet is given for comparison. Cordierite analysis A-f represents the core, and A-e is a single analysis near garnet.

The question of temperature estimates for MK-629 is a difficult one. In the previous section it was suggested that the zoning trend in Figure 36 might be controlled by a reaction involving the breakdown of ilmenite, whereby the biotite would become enriched in Ti, and Fe from ilmenite and biotite could go into garnet and cordierite. Against this is the lack of any obvious Ti enrichment in biotite, though one might argue this effect was swamped by the large amount of biotite present. If this ilmenite reaction were the key to the peculiar garnet zoning in MK-629, then one must question why it was not also important in other rocks such as MK-432.

A more promising solution lie's in the fact that the HK-629 garnet shows about a 2% increase in grossular content in the range where pyrope falls from 24 to 16% and spessartine falls by about 1%. The reaction that controls the grossular content of garnet in this assem­blage is

[3 an = gross + 2 sil + qtz] (20)

An increase of P at constant T, or a decrease of T at constant P, would cause an increase of grossular content (Newton, 1983). This reaction can be combined with reaction 4,

[2 alm-pyr + 4 sil + 5 qtz = 3 crd],

which would favor garnet growth with increasing P at constant T (Thompson, 1976), to yield

[2 alm-pyr + 3 qtz + 6 an = 23 gross + 3 crd]. (21)

This implies for each amount of almandine-pyrope consumed an equal amount of grossular is produced. In going from garnet core to garnet rim in MK-629 almandine-pyrope falls from 91.8 to 90.3%, a drop of 1.5%; thus grossular should rise from 3.8 to 5.3%. The actual rise is 5.7%. It seems that garnet consumption from the cordierite-producing part of the reaction is equalled or even outweighed by garnet production from the anorthite-consuming part of the reaction, and that this accounts for the slight decline in spessartine content with falling XMg (Figure 36).

~

165

With the above discussion in mind, T estimates for garnet-biotite and garnet-cordierite pairs have been listed in Table 13. In the first two pairs, garnet core composition is compared with matrix biotite and cordierite core compositions yielding estimates of 670 and 715°C. If the interpretation of garnet zoning were based entirely on the phenomenon of retrograde ion exchange, then these values might represent true conditions of peak metamorphism. At least the cordierite temperature, however, is way outside the realm of possibility for Zone III conditions. If the interpretation of garnet zoning is based mainly on continuous almandine-pyrope- consuming reactions that decrease the XMg of biotite and cordierite, then the 670 and 715°C must be considered pseudotemperatures.

In the second two pairs an intermediate garnet that is in contact with cordierite is compared with matrix biotite and core cordierite, yielding temperatures of 580 and 620°C. The second temperature could be construed as a true temperature from cordierite grown from garnet in an almandine-pyrope-consuming reaction, and is consistent with inferred Zone III conditions.

The third two pairs must surely represent some form of retrograde re-equilibration. The extreme garnet composition with matrix biotite yields 535°C. The intermediate garnet rim and the cordierite immedi­ately in contact with it yield 580°C.

Pressure Estimates

Tracy et al., 1976b, Fig. 6) presented a calibration of reaction 4 ([sil + gar-+-qtz = crd]) in the Mg-Mn-Fe system for the purposes of estimating pressures of metamorphism. Disregarding difficulties with regard to the effect of H2o in cordierite (Newton, 1983), it can be used to make comparisons with other pressure estimates from central New England. In an assemblage with all four phases, the pressure may be estimated from a garnet composition and a temperature estimate. If the assemblage lacks cordierite, the pressure will only be a minimum estimate (Tracy et al., 1976b).

For sample MK-432 and an estimated temperature of 640°C, garnets Ag and Aa give estimated minimum pressures of 5.7 and 5.6 kbar respec­tively. For sample MK-629, where T estimates are more uncertain, cordierite is present, but it is not known which cordierite composi­tion was in equilibrium with which garnet composition. The most magnesian garnet composition and a temperature of 640°C gives an estimated pressure of 6.3 kbar. The garnet rim composition (A-z) in contact with cordierite, and the temperature estimated from core cordierite composition (A-f), gives a pressure of 6.1 kbar. These estimates seem reasonably consistent with the concept that garnet zoning took place under conditions of falling temperature, but because of the accepted H20 content of cordierite in the Thompson calibration, do not require falling pressure.

166

The high anorthite content of the plagioclase in sample MK-629 and the zoning of grossular content of garnet suggest a possibility of applying the garnet-plagioclase barometer of Ghent et al. (1976) based on the end member reaction

[an = sil +gross + qtz]. (20)

Application of the Ghent et al. (1976) equation as modified by Ghent (1977) for the garnet core-with mole fraction grossular of 0.038 at 640°C yields 3.7 kbar and for the garnet with XGross of 0.057 at 580°C yields 4.4 kbar.

Finally, for MK-432, using matrix biotite composition XMg = 0.45 and the biotite-free rim garnet composition XMg = 0.16 on the garnet and biotite isopleth diagram of Spear and Selverstone (1984, Fig. 1) yields an intersection at 650°C and 5.8 kbar, ignoring effects of spessartine and grossular in the garnet.

CALC-SILICATE ROCKS

Calc-silicate pods and beds are widespread in the quadrangle, such that assemblages could be studied through the full range of meta­morphic grades seen in the pelitic rocks. Such a study is beyond the scope of this thesis. What follqws is a discussion of calc-silicate assemblages studied in a single zoned calc-silicate pod from the Warner Formation near Gilson Pond (MND-8-74), originally collected by Carl Nelson.

Mineralogy and Chemography of MND-8-74

Nelson (1975) reported wollastonite in the core of a calc-silicate pod, and concluded its presence indicated relatively low C02 activity in the fluid phase. This would not be surprising, considering that the relative proportion of carbonate was originally small, and that large quantities of H20 were produced during metamorphism in the surrounding schists.

A new thin section was cut from Nelson's sample and microprobe analyses were done (1) to confirm the presence of wollastonite and (2) to place some chemical constraints on the reactions that produced the assemblage. Representative microprobe analyses are presented in Table 14.

There is a good probablility that the pyroxenoid is actually bustamite. The composition is approximately Ca. 82 Fe. 09 Mn. 09 Si03 , which is less calcic than expected for wollastonite, and plots in the bustamite field shown by Brown et al. (1980) for upper amphibolite facies conditions (Figure 37). -optical properties are similar for wollastonite and bustamite. Both are optically negative; for wollas­tonite, 2V=39, r>v, and R.I. = 1.620-1.634; for bustamite, 2V:45, r<v, and R.I. = 1.662-1.676 (Winchell and Winchell, 1951). Their powder

167

k===================~~fa MgSi03 opx FeSi0 3

Fig. 37. Pyroxenoid and clinopyroxene compositions in MND-8-74, plotted in the system CaSi0

3-FeSi0

3-MnSi0

3-MgSi0

3 after Brown et al.

(1980). Mineral abbreviations: wo- wollastonite; bu- bustamite; cpx - clinopyroxene; rh - rhodonite; pxmn - pyroxmangite; opx -orthopyroxene; fa - fayalite. Tie lines (dotted) are schematic.

168

Table 14. Electron microprobe analyses from the core of calc-silicate granulite pod NH-MND-8-74.

Bustamite Calcite Garnet (3 pts) (4 pts) (1 pt) (5 pts) (1 pt)

Si02 50.50 CaO 58.55 62.12 Sio2 40.01 38.76

Al2o3 .03 MnO .20 .18 Al2o3 19.27 19.06

Na2o . 05 MgO .03 .04 FeO 6.30 6.37

K20 0 FeO 0 0 CaO 30.54 32.82

CaO 38.61 Sum 58.78 62.34 MnO 4.34 3.84

FeO 5.45 MgO • 07 • OS

MnO 5.21 Sum 100.53 100.90

MgO . 18

Sum 100.03

Structural Formulae

(3 oxygens) (1 oxygen) (12 oxygens)

Si .998 Ca .996 .997 Si 3.062 2.951 . Al .001 Mn . 003 .002 Al-site

Sum . 999 Mg .001 .001 Al 1. 739 1. 712

Mg . 005 Fe 0 0 Fe3+ .261 .288 Fe2+ .090 Total 1.000 1.000 Sum 2.000 2.000

Mn . 087 M2+ . -s1te

Ga . 818 Ca 2.506 2.678

Na .002 Fe2+ .143 .118

Sum 1.002 Mn .281 .248 t

Total 2.001 Mg . 008 .005

Sum 2.938 3.049

Total 8.000 8.000

Andradite 13.0 14.4 Grossular 72.2 73.4 Almandine 4.9 3.9 Spessartine 9.6 8.1 Pyrope 0.3 0.2 Total 100.0 100.0

169

Table 14. (cont' d)

ClinoEyroxene Zoisite Clinozoisite Plagioclase

(3 pts) (2 pts) (1 pt) (1 pt) (1 pt) Sio2 50.62 50.88 Si02 39.84 39.74 Si02 44.97 Ti02 .OS .06 Al2o3 31.49 31.17 Al2o3 34.33 Al2o3 .25 . 39 FeO . 99 1. 39 CaO 20.23 FeO 17.88 18.26 MgO 0 0 BaO n.a. MnO 3.88 3.67 CaO 24.96 25.43 ua2o .34 MgO 3.84 3.69 MnO .07 .10 K2o n.a. CaO 22.94 22.90 Sum 97.35 97.83 Sum 99.87 Na2o .10 .17 Sum 100.66 100.02

Structural Formulae

(6 oxygens) (All Fe as Fe2o

3)

(12.5 oxygens) (8 oxygens) Si-site Si-site

IV-site Si 2.033 2.020 Si 3.055 3.045

Si 2.083 Al . 012 .018 Sum 3.055 3.045

Al 1.876 Ti . 001 .002 Al-site

Sum 3.959 Swn 2.046 2.040 Al 2.000 2.000 A-site

M-2 site Sum 2.000 2.000 3+ . CaO 1.005 Ca . 967 .975 Fe -s1te

i~a2o .030 Na .008 .013 Al . 347 .817

! Fe3+ Sum 1.035 Sum . 975 . 988 .064 .089

Total 4.994 M-1 site Mg 0 0 Fe2+ .588 .606 Sum .911 .906 Mn .129 .123 Ca-site An97Ab3

Mg .225 .219 Ca 2.052 2.089 Sum . 942 . 948 Mn .004 .006 Total 3.963 3.976 Sum 2.056 2.095

Total 8.022 8.046

Fe/Fe+Mg .022 .031 Fe3+/Fe3++Al

Di27Hd73 Di27Hd73 .022 .031

170

X-ray patterns are too similar for easy differentiation (Brown et al., 1980). In the following discussion the pyroxenoid will be called-­bustamite, largely on the basis of the composition and the dispersion r<v. The other ternary diagrams in Figure 37 show the coexisting pyroxenoid-pyroxene pair in MND-8-74 from various vantage points in the system CaSi03-MgSi03-MnSi03-FeSi03• Because the clinopyroxene is relatively poor in diopside component, the relations are best observed on the Mg-free face projection and in the schematic 3-D tetrahedron.

Estimated modes for MND-8-74 are listed in Table 5b, showing the assemblages present in the core, transition zone, and rim. The core is mostly quartz, calcite and bustamite, whereas the rim is mostly quartz, calcic plagioclase, and actinolite. Grossular garnet, zoisite, clinozoisite, and sphene make up the other important minerals present. The electron microprobe compositions for zoisite, garnet, bustamite and pyroxene are shown in Figure 38 on two different ternary plots. The garnet appears within the three phase field zo-bu-px on the ACFm plot, but because of Fe3+ the garnet and zoisite compositions are not in the same composition plane as bustamite and pyroxene, as seen in Figure 38b.

Disequilibrium Textures and Reactions

Disequilibrium textures wer~ observed among the minerals in the core, especially in the form of grossular rims around zoisite (Figure 39), providing evidence for the reactions:

[2 zo + 5 cc + 3 qz = 3 gross +

[2 zo + 3 bu + 2 cc = 3 gross +

H2o+ 5 C02],

H2o+ 2 C02],

(22)

(23)

and [2 zo + 5 bu = 3 gross + 2 qz + 1 H2o]. (24)

Because the garnet lies to the right of the zo-bu tie line in Figure 38a, reaction 23 is not supported by the probe data. The bustamite

~ locally separates quartz and calcite and appears to have formed earlier than the grossular by a reaction of the form:

[cc + qz = wo + C02]. (25)

Although pyroxene is also separated from zoisite by grossular, pyroxene's role in the reactions is difficult to assess. Some possible reactions involving pyroxene, balanced according to microprobe compositions, are listed below and can be visualized on Figure 38a:

[px + 3 bu + 2 zo = 3 gross + 2 qz +H2o], (26)

[px + cc + qz = 3 bu + C02], (27)

[px + 2 cc + zo + qz = gross + co2 + H2o]. (28)

Fig. 38. Mineral compositions in MND-8-74 as determined from micro­probe analyses (Table 14). Zoisite and garnet do not actually lie

171

in the plane of the upper ternary diagram because of ferric iron, as shown in lower diagram. Abbreviations: an - anorthite; zo - zoisite; gt - garnet; cc - calcite; bu - bustamite; px - pyroxene.

[on+ cc + qz - gr + co2] (29)

cc

cc

[zo + CC+ qz

~o+ bu = gr+qz +H2o] (24)

~o+bU+CC = gr. H20+ co2](23)

172

[ cc + qz bu + co2] (25)

qz

[cc + px + qz = bu + co2 J (27)

Fig. 39. Portions of core mineral assemblages as seen in thin section of calc-silicate pod MND-8-74. Stippled pattern is grossular garnet. Other minerals as follows: qz -quartz; px- pyroxene; zo- zoisite; cc - calcite; an - anorthite; bu- bustamite; sph - sphene; zr - zircon. Textures suggest the reactions indicated.

t T

0

I() (\J

173

~ Fig. 40. (after Kerrick, 1970) Schematic equilibrium curves for calc-silicate assemblages at low co2 partial pressures. Reactions are numbered as in the text and in Figure 39. A, B, and C are isobaric invariant points. Mineral abbreviations same as in other figures.

174

The reactions involving end-member wollastonite, quartz, calcite, clinozoisite, grossular, COz and HzO define an invariant point on a T-XCOz diagram at constant pressure, shown schematically as point "C" in Figure 40 (after Kerrick, 1970). Compositions in MND-8-74, except for calcite, are so far from the end-member compositions on which Figure 40 is based, that it is only qualitatively relevant.

As temperature increased, the pyroxenoid-producing reaction 25 probably buffered the system up to the point C where grossular became stable. The effects of Mn, Mg and Fe on the position of the invariant points have not been explored experimentally, but we can predict certain trends. The curve for reaction 25 would shift toward lower temperatures for a given XCOz with added impurities in the wollas­tonite, and presumably the analogous reaction involving bustamite would behave similarly. The effects on reactions involving garnet and epidote are less clear. Kerrick (1970) states that, because the distribution of Fe3+ is such that Fe3+ in grossularite exceeds that in zoisite, reaction 28 will shift toward higher temperatures with increased Fe3+ in the system. However, he suggests that the analogue to reaction 24 would shift in the same direction, which is not true if Fe~toss > Fe~6· The plot of grossular and zoisite compositions in Figure 38b makes this clear.

Plagioclase is present in sparse amounts in the core assemblage, and textures (Figure 39) indicate the reaction:

[an + 2 cc + qz = gross + 2 C02]. (29)

No evidence was observed for reaction 30 (Figure 40), nor any of the others which involve plagioclase. Why anorthite should be locally present rather than zoisite is unclear, but it occurs adjacent to a large patch of calcite where perhaps the co2 activity was locally higher. We know that the fluid composition varied greatly from the core, with its carbonate-bearing assemblage, to the rim, where hydrous phases become increasingly abundant, but evidence for reaction 29 in the core suggests minor variations in the fluid composition over less than one millimeter.

CORRELATION OF METAMORPHISM AND DEFORMATION

Age of Prograde Metamorphism

The peak of metamorphism occurred sometime during the Acadian orogeny, but various geologists have different ideas about its timing relative to the phases of deformation. The fact that andalusite pseudomorphs occur mainly in the higher tectonic levels of the nappe pile has led to a general consensus that sillimanite did not make its appearance until after the nappes' emplacement (Thompson et al., 1968; Tracy and Robinson, 1980; Chamberlain and Lyons, 1983; Spear et al., 1983). Kyanite formed at lower levels in the nappe pile, which in present geographical terms translates to "west of the andalumps",

175

except where higher tectonic levels are preserved in the cores of later synformal structures. The rocks in the upper levels were hotter, perhaps in part due to the proximity of the Bethlehem and Kinsman plutons, and at lower pressure, than those deeper in the pile. Although it is the Fall Mountain nappe level that contains the andalumps in the Bronson Hill anticlinorium (Thompson et al., 1968; Spear et al., 1983), in the Monadnock quadrangle the fossil isobar cuts obliquely through the rocks which were in the nappe-stage syncline below level (3), i.e. below the Fall Mountain level (compare Figures 13 and 30). This is especially apparent in the area of Little Monadnock Mountain and Troy. The nappe-stage deformation probably did not juxtapose previously andalusite-bearing rocks on top of kyanite­bearing rocks, but rather set up the conditions necessary for the growth of these minerals. The kyanite-andalusite boundary was probably migrating upwards through the rocks as temperature and pressure conditions changed, and the fossil isobar observed today captured its position where sillimanite became stable.

The boundaries between Zones II, III, and IV (Figure 30) seem to follow approximately the boundaries between levels (1), (2), and (3) (Figure 13). At first sight this might seem to suggest that the peak condition mineral assemblages preceded or coincided with the thrust­nappes. However, the same pattern could have resulted if those rocks which were initially hotter (i.e. in level 3) remained hotter than the rest as rocks in all levels experienced increases in temperature and pressure during early backfolding. The fact that K-feldspar augen cut across foliation at Otter Creek Dam supports this.

Rocks in different parts of the orogen experienced different P-T conditions depending on their locations in the backfolds and later doming. Robinson and Tracy (1980) proposed that rocks in the Merri­mack synclinorium underwent a continued pressure increase after the peak temperature (Figure 31). In contrast, the rocks around the domes rose, and were subjected to less pressure even as the temperature continued to rise. The P-T estimates from garnet zoning studies

! southeast of Mt. Monadnock suggest a trajectory intermediate between the two.

On a smaller scale, Chamberlain (in progress, 1985) has studied the detailed textural relations and mineral compositions at what he interprets to be four cross-fold intersections in the Gilsum-Marlow area. In brief, his early anticline-late anticline intersection shows a gradual decrease in temperature, the early anticline-late syncline intersection shows a temperature decrease followed by increase, the early syncline-late anticline pair shows early temperature increase followed by decrease, and the syncline-syncline crossing shows a continuous increase. This sort of approach could probably be applied in the Monadnock area, but care must be taken not to interpret structures on the basis of the metamorphic results. Chamberlain's "early anticline and syncline", for example, may reflect the west to east increase in grade across the level (2)-level (3) boundary as

176

defined in the Monadnock area. His early syncline is continuous with the Marlboro syncline.

In the Monadnock quadrangle Zone IV assemblages are mainly confined to rocks in level (3), except for the rocks in the Derby Hill window. The rocks within the window were apparently subjected to similar conditions as the rocks outside the window, which supports my suspicion that Zone IV conditions post-date the nappe-stage struc­tures. Although later structures brought the rocks in the window back toward the surface, the axial trace of the Marlboro syncline trends north-south through the area west of the window. This might be a good area in which to do a detailed metamorphic study of the sort done by Chamberlain, although it would be difficult to find samples of the appropriate bulk composition which have escaped retrogression.

The peak of metamorphism certainly preceded dome-stage defor­mation, because quartz-feldspar segregations which presumably formed at approximately peak conditions (Tracy, 1978) are deformed by dome­stage folds. However, the rocks remained hot long enough to allow the formation of the pervasive sillimanite lineations in the dome stage.

Age of Retrograde Metamorphism and the "Permian Disturbance"

The cause of the severe retrogression in the area north and east of Mt. Monadnock is uncertain. There does not seem to be any obvious spatial relationship with late plutons or the mafic dikes.

Krueger and Reesman (1971) reported that K-Ar radiometric "ages" determined from muscovite in ten samples of schist from Mt. Monadnock appear to have been disturbed during the Permian. Zartman et al. (1970) showed the Monadnock area as lying approximately on the-­boundary between an area to the west where K-Ar ages of micas (350-260 m.y.) were partially reset during the Permian, and an area where K-Ar mica ages (260-200 m.y.) were completely reset. They discussed whether the ages were reset during gradual cooling from peak Acadian conditions, or whether the disturbance was related to the Alleghenian orogeny. The "ages" of coexisting biotite at Monadnock are consis­tently older than for muscovite, and Krueger and Reesman estimated a 390 m.y. age for the Acadian metamorphism based on a model whereby Ar leakage from muscovite controls Ar leakage from biotite. The mafic dikes on Monadnock cut all stages of deformation and thus are presumably younger than the Acadian metamorphism, and yet the biotites form a weak foliation and ilmenites are rimmed by sphene. It would be interesting to compare K-Ar ages from biotite in the mafic dikes with those from the Littleton Formation. Since the dikes were not present during peak Acadian metamorphism, ages on biotites from the dikes and muscovite from inclusions might shed some light on the so-called Permian disturbance. More detailed sampling of schists from throughout the quadrangle might show whether or not the zones with severe re-hydration have any relationship to zones of possible late Paleozoic deformation.

177

CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE RESEARCH

The important contributions of this thesis are as follows: (1) Stratigraphic units, which are correlated with those of

central New Hampshire, and indirectly with northwestern Maine, have been defined and their distribution has been mapped. Only the middle member of Fowler-Billings' (1949a) Littleton Formation is retained as Littleton. All the other metamorphosed sedimentary rocks are older.

(2) Belts of Kinsman Granite across the southeast part of the quadrangle are interpreted as a connection between the Cardigan pluton and the Coys Hill Granite, which is repeated in the map pattern by faults. The Spaulding Tonalite cuts the oldest folds whereas the Kinsman preceded them. The Fitzwilliam Granite and several previously unreported microdiorite dikes are post-Acadian.

(3) Major structural features have been defined, and an attempt has been made to determine their relative ages. The Monadnock syn­cline closes south of Troy, and is believed to be a nappe-stage recum­bent syncline now completely reoriented. Large isoclinal folds and the pervasive foliation on Mt. Monadnock are also believed to have formed in the nappe stage. These folds are responsible for the thick pile of quartzite beds which hold up the summit. West-directed, ductile thrust faults cut the nappe-stage folds, and separate the region into three tectonic levels. The Brennan Hill fault separates the Monadnock stratigraphy from the underlying autochthonous rocks on the Keene dome. The Chesham Pond fault and the Thorndike Pond fault zone separate the Monadnock stratigraphy from an overlying sheet of chiefly Rangeley Formation and Kinsman Granite. Augen schists and gneisses, believed to represent recrystallized mylonitic rocks, are found along the thrusts. The nappes and thrusts were deformed by backthrusts and backfolds, chief of which is the Beech Hill anticline, which overturned the Monadnock syncline back toward the southeast. The Thorndike Pond fault zone was reactivated during backfolding, with west-over-east motion. The final phases of Acadian deformation produced folds related to the movement of tonalite plutons, and to the rise of gneiss domes to the west. In the Mesozoic, the Thorndike Pond fault zone was again active, this time with normal faulting down to the west.

(4) Metamorphic assemblages show an increase in grade from Zone II in the autochthonous rocks, to Zone IV in the gneisses above the Chesham Pond fault. An extensive area of Zone III rocks is charac­terized by sillimanite pseudomorphs after andalusite. Garnet zoning patterns in Zone III pelitic schists suggest a P-T history intermediate between those of the Bronson Hill anticlinorium and the Merrimack synclinorium as outlined by Tracy et al. (1976). Retrograde metamorphism has locally produced chlorite, staurolite, and chlori­toid, the last particularly in rocks formerly rich in sillimanite.

Further work is needed in the following areas: (1) Detailed mapping within the Rangeley Formation may yield an

internal stratigraphy beyond the tentative suggestions in this thesis. The southwest portion of the quadrangle has the best potential for such work, and more mapping is needed from there south to the Tully body of Monson Gneiss in Massachusetts. There are also areas in

Roxbury, Nelson, and the southeasternmost corner of the quadrangle where I have not mapped.

178

(2) More could probably be done to define a detailed stratigraphy in the upper part of the Littleton Formation, starting with the Seven Quartzites.

(3) A systematic study of metamorphic assemblages in pelitic schists would be valuable, to refine the metamorphic map presented here and to aid the attempt at relating metamorphic history to struc­tural development. Zoned garnets from various structural positions and metamorphic zones in the quadrangle could be compared.

(4) The geochemistry of the microdiorite dikes and of the coticules should be studied.

179

REFERENCES CITED

Ando, c: .J., Czuchra, B.L., Klemperer, S.L., Brown, L.D., Cheadle, M.J., Cook, F.A., Oliver, J.E., Kaufman, s., Walsh, T., Thompson, J. B., Jr., Lyons, J.B., and Rosenfeld, J.L., 1984, Crustal profile of mountain belt: COCORP deep seismic reflection profiling in New England Appalachians and implications for architecture of conver-gent mountain chains: American Association of Petroleum Geologists Bulletin, v.68, p.819-837.

Ashwal, L.D., Leo, G.W., Robinson, Peter, Zartman, R.E., and Hall, D.J., 1979, The Belchertown quartz monzodiorite pluton, west-central Massachusetts, a syntectonic Acadian intrusion: American Journal of Science, v.279, p.936-969.

Barreiro, Barbara, and Aleinikoff, J.N., 1985, Sm-Nd and U-Pb isotopic relationships in the Kinsman Quartz Monzonite, New Hampshire: Geological Society of America Abstracts with Programs, v.17, p.3.

Bence, A.E., and Albee, A.L., 1968, Empirical correction factors for the electron microanalysis of silicates and oxides: Journal of Geology, v.76, p.382-403.

Berry, H.N., IV, 1985, The Silurian Smalls Falls Formation in south­central Massachusetts and adjacent Connecticut: Geological Society of America Abstracts with Programs, v.17, p.4.

Billings, M.P., 1937, Regional metamorphism of the Littleton­Moosilauke area, New Hampshire: Geological Society of America Bulletin, v.48, p.463-566. , 1941, Structure and metamorphism in the Mount Washington area, New Hampshire: Geological Society of America Bulletin, v.52, p.863-936. , 1942 and 1954, Structural geology: Prentice Hall, Englewood

----Cliffs, N.J., 514 p. Billings, M.P., and Cleaves, A.B., 1934, Paleontology of the Littleton

area, New Hampshire: American Journal of Science, v.28, 5th series, p.412-438.

Boone, G.M., 1973, Metamorphism, stratigraphy, petrology and struc­tural geology of the Little Bigelow Mountain map area, western Maine: Maine Geological Survey Bulletin 24, 136 p.

Boone, G.M., Baudette, E.L., and Moench, R.H., 1970, Bedrock geology of the Rangeley Lakes-Dead River basin region, western Maine: in Boone, G.M., ed., New England Intercollegiate Geological Conference Guidebook, 62nd Annual Meeting, p.1-24.

Boucot, A.J., 1961, Stratigraphy of the Moose River synclinorium, Maine: U.S. Geological Survey Bulletin 1111-E, 188 p. , 1969, Geology of the Moose River and Roach River synclinoria, northwestern Maine: Maine Geological Survey Bulletin 21, 117 p.

Boucot, A.J., and Arndt, Robert, 1960, Fossils of the Littleton Formation (Lower Devonian) of New Hampshire: U.S. Geological Survey Professional Paper 334-B, p.41-51.

Boucot, A.J., and Rumble, Douglas, III, 1980, Regionally metamorphosed (high sillimanite zone, granulite facies) Early Devonian fossils from the Littleton Formation of New Hampshire: Journal of Paleontology, v.54, p.188-195.

Boucot, A.J., and Thompson, J.B., Jr., 1963, Metamorphosed Silurian brachiopods from New Hampshire: Geological Society of America Bulletin, v.74, p.1313-1334.

Bouma, A.H., 1962, Sedimentology of some flysch deposits: Elsevier Scientific Publishing Company, Amsterdam, 168 p.

180

Brookins, D.G., Berdan, J.M., and Stewart, D.B., 1973, Isotopic and paleontological evidence for correlating three volcanic sequences in the Maine coastal volcanic belt: Geological Society of America Bulletin, v.84, p.1619-1628.

Brown, P.E., Essene, E.J., and Peacor, D.R., 1980, Phase relations inferred from field data for Mn pyroxenes and pyroxenoids: Contributions to Mineralogy and Petrology, v.74, p.417-425.

Burdige, D.J., and Gieskes, J.M., 1983, A pore water/solid phase diagenetic model for manganese in marine sediments: American Journal of Science, v.283, p.29-47.

Butler, R.W.H., 1982, The terminology of structures in thrust belts: Journal of Structural Geology, v.4, p.239-245.

Cameron, E.N., Larrabee, D.M., McNair, A.H., Page, J.J., Stewart, G.W., and Shainin, V.E., 1954, Pegmatite investigations, 1942-1945, New England: U.S. Geological Survey Professional Paper 255.

Chamberlain, C.P., 1981, Metamorphic zonation in south-central New Hampshire: M.A. thesis, Dar~mouth College, Hanover, N.H., 191 p.

_____ , 1985, Petrologic evidence for repeated folding of isotherms during regional metamorphism: Ph.D. thesis in preparation, Harvard University, Cambridge, Mass.

Chamberlain, C.P., and Lyons, J.B., 1983, Pressure, temperature and metamorphic zonation studies of pelitic schists in the Merrimack synclinorium, south-central New Hampshire: American Mineralogist, v.68, p.530-540.

Clark, R.G., 1972, Petrologic and structural aspects of the Kinsman Quartz Monzonite and some related rocks: Ph.D. thesis, Dartmouth College, Hanover, N.H., 225 p.

Clifford, T.N., 1960, Spessartine and magnesium biotite in coticule­bearing rocks from Mill Hollow, Alstead Township, New Hampshire, U.S.A., a contribution to the petrology of metamorp~osed mangan­iferous sediments: Neues Jahrbuch fUr Mineralogie, Abhandlungen, v.94, p.1369-1400.

Davis, W.M., 1896, The Seine, the Meuse, and the Moselle: National Geographic Magazine, v.7, p.189-202 and 228-238.

Dean, C.S., 1977, Structure and stratigraphy of the Sunapee septum, southwestern New Hampshire: Ph.D. thesis, Harvard University, Cambridge, Mass., 248 p.

Duke, E.F., 1984, Stratigraphy, structure, and petrology of the Peterborough 15-minute quadrangle, New Hampshire, Part I: Ph.D. thesis, Dartmouth College, Hanover, N.H., 168 p.

Duke, G.I., 1984, Structure and petrology of the eastern Concord quadrangle, New Hampshire: M.S. thesis, Dartmouth College, Hanover, N.H., 115 p.

181

Dutrow, B.L., and Holdaway, M.J., 1983, Upper stability of staurolite and quartz at low pressures: Geological Society of America Abstracts with Programs, v.15, p.563.

Elbert, D.C., 1984, Stratigraphic and structural reinterpretation in the Bernardston-Northfield area, north-central Massachusetts: Geological Society of America Abstracts with Programs, v.16, p.14.

Emerson, B.K., 1898, Geology of Old Hampshire County, Massachusetts, comprising Franklin and Hampden counties: U.S. Geological Survey Monograph 29, 790 p.

Englund, E.J., 1976, The bedrock geology of the Holderness qudrangle, New Hampshire: New Hampshire Department of Resources and Eco­nomic Development Bulletin 7, 90 p.

Eusden, J.D., Bothner, W.A., Hussey, A.M., and Laird, J., 1984, Silurian and Devonian rocks in the Alton and Berwick quadrangles, New Hampshire and Maine: in Hanson, L.S., ed., New England Intercollegiate Geological-conference Guidebook, 76th Annual Meeting, p.325-351.

Fermor, L.L., 1909, The manganese-ore deposits of India: Geological Survey of India Memoir, v.37.

Ferry, J.M., and Spear, F.S., 1978, Experimental calibration of the partitioning of Fe and Mg between biotite and garnet: Contribu­tions to Mineralogy and Petrology, v.66, p.113-117.

Field, M.T., 1975, Bedrock geology Of the Ware area, central Massachu­setts: Contribution No.22 (Ph.D. thesis), Department of Geology and Geography, University of Massachusetts, Amherst, 186 p.

Fitzgerald, K., 1960, Geology of the western half of the Royalston quadrangle, Massachusetts: M.S. thesis, University of Massachu­setts, Amherst, 120 p.

Fowler-Billings, Katharine, 1944a, Igneous and metasedimentary dikes of the Mt. Washington area, New Hampshire: Geological Society of America Bulletin, v.55, p.1255-1278. , 1944b, Sillimanite deposits in the Monadnock quadrangle, New

-----Hampshire Mineral Resource Survey, Part 8: New Hampshire Development Commission, Concord, 15 p.

! , 1949a, Geology of the Monadnock region of New Hampshire: ----Geological Society of America Bulletin, v.60, p.1249-1280.

, 1949b, Geology of the Monadnock quadrangle, New Hampshire: New Hampshire Department of Resources and Economic Development, Concord, 48 p.

Fowler-Lunn, Katharine, and Kingsley, L., 1937, Geology of the Cardigan quadrangle, New Hampshire: Geological Society of America Bulletin, v.48, p.1363-1386.

Greene, R.C., 1970, The geology of the Peterborough quadrangle, New Hampshire: New Hampshire Department of Resources and Economic Development, Concord, 88 p.

Guthrie, G.D., and Burnham, c.w., 1985, Petrology and origin of calc­silicate bodies from the Rangeley Formation, New Hampshire: Geological Society of America Abstracts with Programs, v.17, p.22.

Hall, B.A., 1973, Slump folds and the determination of paleoslope: Geological Society of America Abstracts with Programs, v.5, p.648.

Hall, B.A., Pollock, S.G., and Dolan, K.M., 1976, Lower Devonian Seboomook Formation and Matagamon Sandstone, northern Maine: a flysch basin-margin delta complex: in Page, L.R., ed., Contributions to the stratigraphy of New England: Geological Society of America Memoir 148.

182

Hall, D.J., 1970, Compositional variations in biotites and garnets from kyanite and sillimanite zone mica schists, Orange area, Massachusetts and New Hampshire: Contribution No.4 (M.S. thesis), Department of Geology and Geography, University of Massachusetts, Amherst, 110 p.

Hall, L.M., and Robinson, Peter, 1982, Stratigraphic-tectonic subdivisions of southern New England: in St. Julien, P., and B~land, J., eds., Major structural zones and faults of the northern Appalachians: Geological Association of Canada Special Paper 24, p.15-41.

Harris, A.G., Hatch, N.L., Jr., and Dutro, J.T., Jr., 1983, Late Silurian conodonts update the metamorphosed Fitch Formation, Littleton area, New Hampshire: American Journal of Science, v.283, p.722-738.

Harwood, D.S., and Berry, W.B.N., 1967, Fossiliferous lower Paleozoic rocks in the Cupsuptic quadrangle, west-central Maine: U.S. Geological Survey Professional Paper 575-D, p.D16-D23.

Hatch, N.L.,Jr., Moench, R.H., and Lyons, J.B., 1983, Silurian-Lower Devonian stratigraphy of eastern and south-central New Hampshire: extensions from western Maine: American Journal of Science, v.283, p.739-761.

Hatch, N.L., Jr., Robinson, Peter, and Stanley, R.S., in press, Stratigraphy of the Connecticut Valley belt: U.S. Geological Survey Professional Paper, Chapter G, 97 p.

Hayward, J.A., 1983, Rb-Sr geochronology and the evolution of some peraluminous (two-mica) granites in New Hampshire: Geological Society of America Abstracts with Programs v.16, p.146.

! Heald, M.T., 1950, Structure and petrology of the Lovewell Mountain quadrangle, New Hampshire: Geological Society of America Bulletin, v.61, p.43-89.

Hitchcock, C.H., 1877, Geology of New Hampshire, Part II: Strati­graphical geology, and atlas: Concord, 684 p.

Holdaway, M.J., 1971, Stability of andalusite and the aluminum silicate phase diagram: American Journal of Science, v.271, p.97-131.

Hollocher, K.T., 1981, Retrograde metamorphism of the Lower Devonian Littleton Formation in the New Salem area, west-central Massachu­setts: Contribution No.37 (M.S. thesis), Department of Geology and Geography, University of Massachusetts, Amherst, 268 p.

Huntington, J.C., 1975, Mineralogy and petrology of metamorphosed iron-rich beds in the Lower Devonian Littleton Formation, Orange area, Massachusetts: Contribution No.19 (M.S. thesis), Department of Geology and Geography, University of Massachusetts,

~

Amherst, 106 p. Jasaitis, R., 1983, Geology of pre-Mesozoic bedrock of the Amherst

area, west-central Massachusetts: Contribution No.45 (M.S. tnesis), Department of Geology and Geography, University of Massachusetts, Amherst, 98 p.

Jones, N.W., and Bloss, F.D., 1980, Laboratory manual for optical mineralogy: Burgess Publishing Company, Minneapolis, Minn.

Karamata, V.S., Keesmann, I., and Okrusch, M., 1970, Ein Paragonit­fUhrender Granatquartzit in Raum Brezovica, Slldserbien: Neues Jahrbuch fllr Mineralogie, Monatshefte 1970, v.1, p.1-19.

183

Kerrick, D.M., 1970, Contact metamorphism in some areas of the Sierra Nevada, California: Geological Society of America Bulletin, v.81, p.2913-2938.

Kramm, Ulrich, 1976, The coticule rocks (spessartine quartzites) of the Venn-Stavelot massif, Ardennes, a volcanoclastic metasedi­ment?: Contributions to Mineralogy and Petrology, v.56, p.135-155. .

Krueger, H.W., and Reesman, R.H., 1971, Radiometric determination of Late Paleozoic erosion rate in New Hampshire and the age of meta­morphism of the Littleton Formation: Geological Society of America Abstracts with Programs, v.3, p.627-628.

Kruger, F.K., 1946, Structure and metamorphism of the Bellows Falls quadrangle of New Hampshire: qeological Society of America Bulletin, v.57, p.161-206.

Leo, G.W., 1980, Petrologic constraints on an island-arc origin of Oliverian domes and Ammonoosuc Volcanics, Abstract in Wones, D.R., ed., Proceedings, IGCP Project 27, The Caledonides in the U.S.A.: Department of Geological Sciences, Virginia Polytechnical Institute and State University, Blacksburg, Virginia, Memoir 2, p.A18.

Lyons, J.B., 1979, Stratigraphy, structure and plutonism east of the Bronson Hill anticlinorium, New Hampshire: in Skehan, J.W., S.J., and Osberg, P.H., eds., Contributions, IGCP Project 27, The Caledonides in the U.S.A.: Weston Observatory, Department of Geology and Geophysics, Boston College, Weston, Mass., p.73-92.

Lyons, J.B., Boudette, E.L., and Aleinikoff, J.N., 1982, The Avalon and Gander zones in central and eastern New England: in St. Julien, P., and B~land, J., eds., Major structural zones and faults of the northern Appalachians: Geological Association of Canada Special Paper 24, p.43-66.

Lyons, J.B., and Livingston, D.E., 1977, Rb-Sr age of the New Hampshire Plutonic Series: Geological Society of America Bulletin, v.88, p.1808-1812.

Lyons, J.B., Norwick, Stephen, and Lambine, Janna, 1973, Mantled garnet in plutonic rocks: Geological Society of America Program with Abstracts, v.5, p.190.

Malinconico, M.L., 1982, The stratigraphy and structure of the south­eastern Rumney 15' quadrangle, New Hampshire: M.A. thesis, Dartmouth College, Hanover, N.H., 234 p.

McKerrow, w.s., and Ziegler, A.M., 1972, Silurian paleogeographic development of the proto-Atlantic Ocean: 24th International Geologic Congress, Montreal, Section 6, p.4-10.

184

Metais, D., and Chayes, F., 1963, Varieties of lamprophyre: Yearbook, Carnegie Institution, Washington, D.C., 1962, p.156-157.

Moench, R.H., 1970, Premetamorphic down-to-basin faulting, folding, and tectonic dewatering, Rangeley area, western Maine: Geological Society of America Bulletin, v.81, p.1463-1496.

_____ , 1971, Geologic map of the Rangeley and Phillips quadrangles, Franklin and Oxford Counties, Maine: U.S. Geological Survey Miscellaneous Investigations Map I-605.

Moench, R.H., and Baudette, E.L., 1970, Stratigraphy of the northwest limb of the Merrimack synclinorium in the Kennebago Lake, Rangeley, and Phillips quadrangles, western Maine: in Boone, G.M., ed., New England Intercollegiate Geological Conference Guidebook, 62nd Annual Meeting, p.A-1, 1-25.

Moore, G.E., Jr., 1949, Structure and metamorphism of the Keene­Brattleboro area, New Hampshire and Vermont: Geological Society of America Bulletin, v.60, p.1613-1669.

Morton, P.S., 1984, Tectonic breccia of tonalite and granite in an Acadian shear zone, east side of the Hardwick pluton, Brooks Village, north-central Massachusetts: Geological Society of America Abstracts with Progra~s, v.16, p.52.

Naylor, R.S., 1969, Age and origin ' of the Oliverian domes, central­western New Hampshire: Geological Society of America Bulletin, v.8o, p.405-427.

Naylor, R.S., 1970, Radiometric dating of pennine-type nappes in the northern Appalachians: Eclogae Geologicae Helvetiae, v.63, p.230.

Nelson, C.E., 1975, Structure and petrology of the Thorndike Pond area, southwestern New Hampshire: M.A. thesis, Dartmouth College, Hanover, N.H., 75 p.

Neuman, R.B., and Rankin, D.W., 1980, Bedrock geology of the Shin Pond-Traveler Mountain region: in Roy, D.C., and Naylor, R.S., eds., New England Intercollegiate Geological Conference, 72nd Annual Meeting Guidebook, p.86-97.

Nielson, D.L., 1974, The structure and petrology of the Hillsboro quadrangle, New Hampshire: Ph.D. thesis, Dartmouth College, N.H., 254 p.

Nielson, D.L., 1981, The bedrock geology of the Hillsboro quadrangle, New Hampshire: New Hampshire Department of Resources and Economic Development, Concord, 76 p.

Nielson, D.L., Lyons, J.B., Englund, E.J., and Borns, D.J., 1976, Gravity models and mode of emplacement of the New Hampshire Plutonic Series: in Lyons, P.C., and Brownlow, A.H., eds., Studies in New England geology: Geological Society of America Memoir 146, p.301-318.

Osberg, P.H., 1980, Stratigraphic and structural relations in the turbidite sequence of south-central Maine: in Roy, D.C., and Naylor, R.S., eds., New England Intercollegiate Geological Conference, 72nd Annual Meeting Guidebook, p.278-296.

185

Osberg, P.H., Moench, R.H., and Jeffrey, Warner, 1968, Stratigraphy of the Merrimack synclinorium in west-central Maine: in Zen, E-an, White, W.S., Hadley, J.B., and Thompson, J.B., Jr., eds., Studies of Appalachian geology, northern and maritime: Interscience Publishers, New York, p.241-253.

Pankiwskyj, K.A., Ludman, Allan, Griffin, J.R., and Berry, W.B.N., 1976, Stratigraphic reltions on the southeast limb of the Merrimack synclinorium in central and west-central Maine: in Lyons, P.C., and Brownlow, A.H., eds., Studies in New England geology: Geological Society of America Memoir 146, p.263-280.

Pavlides, Louis, Boucot, A.J., and Skidmore, W.B., 1968, Stratigraphic evidence for the Taconic orogeny in the northern Appalachians: in Zen, E-an, White, w.s., Hadley, J.B., and Thompson, J.B., Jr., eds., Studies of Appalachian geology, northern and maritime: Interscience Publishers, New York, p.61-82.

Pease, M.H., Jr., and Barosh, P.J., 1981, Distribution and structural significance of the Oakdale Formation in northeastern Connec­ticut: in Boothroyd, J.C., and Hermes, O.D., eds., New England Intercollegiate Geological Conference, 73rd Annual Meeting Guidebook, p.17-34.

Perry, J.H., 1904, Geology of Monadnock Mountain, New Hampshire: Journal of Geology, v.12, p.1-14.

Peterson, V.L., 1984, Structure and stratigraphy of bedrock in the Ashburnham-Ashby area, north-central Massachusetts: Contribution No.47 (M.S. thesis), Department of Geology and Geography, University of Massachusetts, Amherst, 182 p.

Pike, T.M., 1968, Structural geology of the Tully body of Monson Gneiss: M.S. thesis, University of Massachusetts, Amherst, 48 p.

Pumpelly, Raphael, Wolf, J.E., and Dale, T.N., 1894, Geology of the Green Mountains in Massachusetts: u.s. Geological Survey Monograph 23, 206 p.

Reineck, H.E., and Singh, I.B., 1980, Depositional sedimentary environments: Springer-Verlag, New York, p.84-94.

Richardson, S.M., 1974, Cation exchange reactions and metamorphism of high-grade pelites in central Massachusetts: Geological Society of America Abstracts with Programs, v.6, p.1059.

Robinson, Peter, 1963, Gneiss domes of the Orange area, Massachusetts and New Hampshire: Ph.D. thesis, Harvard University, Cambridge, Mass., 253 p.

_____ , 1967, Gneiss domes and recumbent folds of the Orange area, west-central Massachusetts: in Robinson, Peter, and Drake, D.P., eds., New England Intercollegiate Geological Conference, 59th Annual Meeting Guidebook, p.17-47.

_____ , 1979, Bronson Hill anticlinorium and Merrimack synclinorium in central Massachusetts: in Skehan, J.W., S.J., and Osberg, P.H., eds., Contributions, IGCP Project 27, The Caledonides in the U.S.A.: Weston Observatory, Department of Geology and Geophysics, Boston College, Weston, Mass., p.126-174.

_____ , 1981, Siluro-Devonian stratigraphy of the Merrimack synclinorium, central Massachusetts--Review based on correlations in Maine: Geological Society of America Abstracts with Programs,

v.13, p.172. Robinson, Peter, and Field, M.T., and Tucker, R.D., 1982a,

Stratigraphy and structure of the Ware-Barre area, central Massachusetts: in Joesten, R., and Quarrier, S.S., eds., New England Intercollegiate Geological Conference, 74th Annual Meeting Guidebook, p.341-373.

Robinson, Peter, and Hall, L.M., 1980, Tectonic synthesis of southern New England: in Wones, D.R., ed., Proceedings, IGCP Project 27, The Caledonides in the U.S.A.: Department of Geological Sciences, Virginia Polytechnical Institute and State University, Blacksburg, Va., Memoir 2, p.73-82.

Robinson, Peter, Tracy, R.J., Hollocher, K.T., and Dietsch, C.W., 1982b, High grade Acadian regional metamorphism in south-central Massachusetts: in Joesten, R., and Quarrier, S.S., eds., New England Intercollegiate Geological Conference, 74th Annual Meeting Guidebook, p.289-339.

Robinson, Peter, and Jaffe, H.W., 1969a, Aluminous enclaves in gedrite-cordierite gneiss from southwestern New Hampshire: American Journal of Science, v.267, p.389-421.

_____ , 1969b, Chemographic exploration of amphibole assemblages from central Massachusetts and southwestern New Hampshire: Mineralogical Society of America, Special Paper 2, p.251-274.

Robinson, Peter, Thompson, J.B., Jr., and Rosenfeld, J.L., 1979, Nappes, gneiss domes, and regional metamorphism in western New Hampshire and central Massachusetts: in Skehan, J.W., S.J., and Osberg, P.H., eds., Contributions, IGCP Project 27, The Caledonides in the U.S.A.: Weston Obsrvatory, Department of Geology and Geophysics, Boston College, Weston, Mass., p.93-125.

Rosenfeld, J.L., 1969, Stress effects around quartz inclusions in almandine and the piezothermometry of coexisting aluminum silicates: American Journal of Science, v.267, p.317-351.

Schiller, E.A., and Taylor, F.C., 1965, Spessartine-quartz rocks (coticules) from Nova Scotia: American Mineralogist v.50, p.1477-1481.

Schumacher, J.L., 1983, Stratigraphic, geochemical, and petrologic

186

t studies of the Ammonoosuc Volcanics, north-central Massachusetts, and southwestern New Hampshire: Ph.D. thesis, University of Massachusetts, Amherst, 237 p.

Shearer, C.K., 1983, Petrography, mineral chemistry, and geochemistry of the Hardwick Tonalite and associated igneous rocks, central Massachusetts: Ph.D. thesis, University of Massachusetts, Amherst, 265 p.

Spear, F.S., Hickmott, D., and Selverstone, J., 1983, Petrologic evidence for a metamorphic discontinuity--Bellows Falls area,

New Hampshire and Vermont: EOS (American Geophysical Union Transactions), v.64, p.878.

Streckeisen, A.L., 1973, Plutonic rocks--classification and nomen­clature recommended by the I.U.G.S. subcommission on the systematics of igneous rocks: Geotimes, v.18, p.26-30.

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

Thompson, J.B., Jr., 1957, The graphical analysis of mineral assemblages in pelitic schists: American Mineralogist, v.42, p.842-858.

187

Thompson, J.B., Jr., and Norton, S.A., 1968, Paleozoic regional meta­morphism in New England and adjacent areas: in Zen, E-an, White, w.s., Hadley, J.B., and Thompson, J.B., Jr., eds., Studies of Appalachian geology, northern and maritime: Interscience Publishers, New York, p.319-327.

Thompson, J.B., Jr., Robinson, Peter, Clifford, T.N., and Trask, J.J., 1968: in Zen, E-an, White, W.S., Hadley, J.B., and Thompson, J.B., Jr., eds., Studies of Appalachian geology, northern and maritime: Interscience Publishers, New York, p.203-218.

Tracy, R.J., 1975, High-grade metamorphic reactions and partial melting in pelitic schists, Quabbin Reservoir area, Massachusetts: Ph.D. thesis, University of Massachusetts, Amherst, 127 p.

_____ , 1978, High grade metamorphic reactions and partial melting in pelitic schist, west-central Massachusetts: American Journal of Science, v.278, p.150-178.

Tracy, R.J., and Robinson, Peter, ~980, Evolution of metamorphic belts: information from detailed studies: in Wones, D.R., ed., Proceedings, IGCP Project 27, The Caledonides in the U.S.A.: Department of Geological Sciences, Virginia Polytechnical Institute and State University, Blacksburg, Va., Memoir 2, p.189-195.

Tracy, R.J., Robinson, Peter, and Field, M.T., 1976a, Phase relations of natural pure magnesian cordierite and biotite from pyrite­sillimanite- orthoclase schist, central Massachusetts: EOS (American Geophysical Union Transactions), v.57, p.338-339.

Tracy, R.J., Robinson, Peter, and Thompson, A.B., 1976b, Garnet composition and zoning in the determination of temperature and pressure of metamorphism, central Massachusetts: American Mineralogist, v.61, p.762-775.

Trask, J.J., and Thompson, J.B., Jr., 1967, Stratigraphy and structure of the Skitchewaug nappe in the Bernardston area, Massachusetts and adjacent New Hampshire and Vermont: in Robinson, Peter, and Drake, D.P. eds., New England Intercollegiate Geological Conference, 59th Annual Meeting Guidebook, p.129-142.

Tr6ger, W.E., 1935, Spezielle Petrographie der Eruptivgesteine: Deutschen Mineralogischen Gesellschaft, Berlin.

Tucker, R.D., 1977, Bedrock geology of the Barre area, central Massa­chusetts: Contribution No.30 (M.S. thesis), Department of Geo­logy and Geography, University of Massachusetts, Amherst, 132 p.

Tucker, R.D., and Robinson, Peter, 1976-1977, unpublished reconnais­sance data, Gardner and Wachusett Mountain quadrangles.

Williams, S.H., and Rickards, R.B., 1984, Paleoecology of graptolitic black shales: in Bruton, D.L., ed., Aspects of the Ordovician System: Paleontological Contributions from the University of

188

Oslo, Norway, no.295, p.159-166. Winchell, A.N., and Winchell, H., 1951, Elements of optical mineralogy

4th edition, Part II, Description of minerals: Wiley and Sons, New York, 551 p.

Zartman, R.E., Hurley, P.M., Krueger, H.W., and Giletti, B.J., 1970, A Permian disturbance of K-Ar radiometric ages in New England: its occurrence and cause: Geological Society of America Bulletin v.81, p.3359-3373.

Zen, E-an, Goldsmith, Richard, Ratcliffe, N.M., Robinson, Peter, and Stanley, R.S., 1983, Bedrock geological map of Massachusetts: U.S. Geological Survey.

189

APPENDIX 1.

Table 15. List of station codes and numbers.

Code Towns hiE Numbers

DB Dublin 1-327; 01-04 FZ Fitzwilliam 1-67 HK Hancock 1-12 HV Harrisville 1-179 KN Keene 1-32 MB Marlboro 1-346 MK Jaffrey 1-957; 997-1188; 001-013 NL Nelson 1-31 RD Richmond 1-26 RI Rindge 101-128 RX Roxbury 1-187 sv Sullivan 1-23 sz Swanzey 1-63 TR Troy 1-344

Table 16. List of mineral abbreviations.

Abbreviation Mineral Abbreviation Mineral

ab albite hd hedenbergite act actinolite ky kyanite alm almandine ksp K-feldspar and andradite mt magnetite an anorthite mus muscovite

'.: ap apatite or orthoclase biot biotite opx orthopyroxene bu bustamite py pyrite cc calcite px pyroxene chl chlorite qtz; qz quartz

crd cordierite rh rhodonite cpx clinopyroxene sil sillimanite ctd chloritoid spess spessartine czo; cz clinozoisite sph sphene di diopside st staurolite

fa fayalite wo wollastonite gar; gt garnet zr zircon gross; gr grossular zo zoisite gph graphite

' ' ; 'fk

Sullivan

KN

Swanzey

sz

Richmond

RD

Roxbury

RX

Marlboro

MB

Troy

TR

Fitzwilliam

FZ

APPENDIX 2.

Nelson

NL

Harrisville

HV

Dublin

DB

• Mt. Monadnock

Jaffrey

MK

Rindge

Rl

Hancock

HK

0

0 2

Fig. 41. Townships in the Monadnock quadrangle, and codes used for stations (see Table 15).

190

N

~

I 2 mi.

3 km.

191

. : :·.:~=.'f: . ·. ·.·:

:-,.,..:-:--,-----,_,..,._...,...,.._ ...,_-;-. .. . ·.· . ;}F { .... ··:· .. :_::-:· .. ::·~<:.' ·~:··: ·>-::.· .. · ·:::~.-:. · ..... .

w

Fig. 42. Two piece fence diagram to cut out and assemble of Mt. Monadnock summit. Dotted pattern represents rocks younger than Seven Quartzites. Instructions: (1) Trace all lines onto a sheet of paper or light cardboard. (2) Tra~e mirror image on reverse side of paper, using light table. (3) Cut out the two sections, and cut along dashed lines. (4) Fit together with points of compass properly arranged. (5) View from NE, NW, SW, and SE to visualize structures in three dimensions .

. . ... .... : : ~ :· : . :.. . . . . . . . . . . . . ..... . ~.: . .

. .. . . ·.·.~ · - · . · .· .. · · -: ·.-:::· ·

. : : . . . .. ·. : .. .. ~ ... : .

E

S N