Petrographic and geochemical comparisons between the lower

spe458-13 1st pgs page 255

255

The Geological Society of AmericaSpecial Paper 458

2009

Petrographic and geochemical comparisons between the lower crystalline basement-derived section and the granite megablock

and amphibolite megablock of the Eyreville B core, Chesapeake Bay impact structure, USA

Gabrielle N. Townsend*Roger L. Gibson

Impact Cratering Research Group, School of Geosciences, University of the Witwatersrand, P.O. WITS, Johannesburg 2050, South Africa

J. Wright Horton Jr.U.S. Geological Survey, 926A National Center, Reston, Virginia 20192, USA

Wolf Uwe ReimoldMuseum for Natural History–Leibniz Institute at the Humboldt University, Invalidenstrasse 43, 10115 Berlin, Germany,

and Impact Cratering Research Group, School of Geosciences, University of the Witwatersrand, PO WITS, Johannesburg 2050, South Africa

Ralf-Thomas SchmittMuseum for Natural History–Leibniz Institute at the Humboldt University, Invalidenstrasse 43, 10115 Berlin, Germany

Katerina BartosovaDepartment of Geological Sciences, University of Vienna, Althanstrasse 14, Vienna A-1090, Austria

ABSTRACT

The Eyreville B core from the Chesapeake Bay impact structure, Virginia, USA, contains a lower basement-derived section (1551.19 m to 1766.32 m deep) and two megablocks of dominantly (1) amphibolite (1376.38 m to 1389.35 m deep) and (2) granite (1095.74 m to 1371.11 m deep), which are separated by an impac-tite succession. Metasedimentary rocks (muscovite-quartz-plagioclase-biotite-graphite ± fi brolite ± garnet ± tourmaline ± pyrite ± rutile ± pyrrhotite mica schist, hornblende-plagioclase-epidote-biotite-K-feldspar-quartz-titanite-calcite amphibolite, and vesuvianite-plagioclase-quartz-epidote calc-silicate rock) are domi-nant in the upper part of the lower basement-derived section, and they are intruded

*[email protected]

Townsend, G.N., Gibson, R.L., Horton, J.W., Jr., Reimold, W.U., Schmitt, R.-T., and Bartosova, K., 2009, Petrographic and geochemical comparisons between the lower crystalline basement-derived section and the granite megablock and amphibolite megablock of the Eyreville B core, Chesapeake Bay impact structure, USA, in Gohn, G.S., Koeberl, C., Miller, K.G., and Reimold, W.U., eds., The ICDP-USGS Deep Drilling Project in the Chesapeake Bay Impact Structure: Results from the Eyreville Core Holes: Geological Society of America Special Paper 458, p. 255–XXX, doi: 10.1130/2009.2458(13). For permission to copy, contact [email protected]. ©2009 The Geological Society of America. All rights reserved.

256 Townsend et al.


INTRODUCTION

The International Continental Scientifi c Drilling Program (ICDP)–U.S. Geological Survey (USGS) Eyreville core holes were drilled in 2005 and 2006 in the Chesapeake Bay impact structure on Eyreville Farm, Delmarva Peninsula, on the eastern coast of Virginia, USA (Fig. 1). Three core holes at this drill site were designated Eyreville A, B, and C, and this study is based on Eyreville B, which recovered a nearly continuous section of cores from 738 m to 1766 m depth (Gohn et al., 2006, 2008). Geo-physical studies and drilling have revealed the structure, located near the mouth of Chesapeake Bay, to be ~85 km across, with an inner basin (30–38 km wide and 1–2 km deep) surrounded by an annular trough (21–31 km wide) (Fig. 1; e.g., Horton et al., 2005a, 2005c, 2008; Poag et al., 2004; Gohn et al., 2008). The crater excavated Early Cretaceous to Holocene Virginia coastal plain sediments that blanket the area and penetrated into the underlying Appalachian crystalline basement (Poag et al., 2004). Previously, only the Langley and Bayside drill cores on the inner western side of the annular trough had intersected crater base-ment (Horton et al., 2005a), and knowledge of other Appalachian basement target rocks of the crater was restricted to studies of the clast population in the limited impactite intersections obtained from these and other core holes around the crater margins (Hor-ton et al., 2005b; Poag et al., 2004).

The Eyreville B core intersected ~215 m of basement-derived crystalline rock between 1551.19 m and 1766.32 m (Fig. 2A). In addition, a 275-m-thick granite megablock (or slab; Horton et al., this volume, Chapter 2; Gibson et al., this volume) and a smaller underlying 13-m-thick amphibolite megablock located in the sandy deposits overlying the impactites (Fig. 2A) provide supplementary information about the nature of the basement rocks. Megablock is used here without a specifi ed size range, recognizing that other papers (e.g., Horton et al., this volume, Chapter 2) refer to the “granite slab” and “amphibolite block” using particle-size defi nitions of block (>4.1 m and <65.5 m) and

by pegmatitic to coarse-grained granite (K-feldspar-plagioclase-quartz-muscovite ± biotite ± garnet) that increases in volume proportion downward. The granite megablock contains both gneissic and weakly or nonfoliated biotite granite varieties (K-feldspar-quartz-plagioclase-biotite ± muscovite ± pyrite), with small schist xeno-liths consisting of biotite-plagioclase-quartz ± epidote ± amphibole.

The lower basement-derived section and both megablocks exhibit similar, middle- to upper-amphibolite-facies metamorphic grades that suggest they might represent parts of a single terrane. However, the mica schists in the lower basement-derived sequence and in the megablock xenoliths show differences in both mineralogy and whole-rock chemistry that suggest a more mafi c source for the xenoliths. Similarly, the mineralogy of the amphibolite in the lower basement-derived section and its asso-ciation with calc-silicate rock suggest a sedimentary protolith, whereas the bulk-rock and mineral chemistry of the megablock amphibolite indicate an igneous protolith. The lower basement-derived granite also shows bulk chemical and mineralogical dif-ferences from the megablock gneissic and biotite granites.

slab (>65.5 m and <1048.6 m) from Blair and McPherson (1999). The general stratigraphy of the Eyreville cores is described in Gohn et al. (2006, 2008), Horton et al. (this volume, Chapter 2), and Edwards et al. (this volume, Chapters 3 and 4).

This paper describes and compares the petrographic and chemical characteristics of the pre-impact crystalline rocks of the Eyreville B core, namely, the lower basement-derived sec-tion, the granite megablock, and the amphibolite megablock. Together with a companion paper (Gibson et al., this volume), which describes the metamorphic and structural features in the basement-derived section and amphibolite megablock, this paper catalogues a component of the crater basement that had not been recognized prior to the recovery of the Eyreville cores

77° 76°77°W 76°

37°N

38°

Figure 1. Map of the Chesapeake Bay impact structure showing the locality of the Eyreville core holes (from Gohn et al., 2006).

Petrographic and geochemical comparisons of the basement and megablock sections of the Eyreville B core 257


and discusses the implications for regional understanding of Appalachian geology.

PETROGRAPHY

Petrographic descriptions were obtained on a total suite of 91 samples, consisting of 73 samples collected by Gibson (RG series; Table 1), 16 collected by Reimold (W series), and 2 provided by Michael J. Kunk (MJK series). Of these, 49 samples are from the lower basement-derived section (28 mica schist, 7 amphibolite, 2 calc-silicate rock, 1 tourmalinite, 11 granite), 38 are from the granite megablock, and 4 are from the amphibolite megablock (Table 1). Contrary to Horton et al. (this volume, Chapter 2, who only observed shocked basement rock directly adjacent to a few injections of impact breccia) and unconfi rmed inferences by Glidewell et al. (2008), no evi-dence was found in any of these samples from either the lower

basement-derived rocks or the megablocks for unequivocal shock microdeformation features.

Lower Basement-Derived Section

The lower, crystalline basement-derived section of the Eyreville B core (1551.19–1766.32 m deep) consists of foli-ated metasedimentary rocks (including mica schist, amphibolite, and calc-silicate rock) and pegmatitic to coarse-grained gran-ite, with a distinct mylonitic zone between 1640 m and 1655 m depth (Fig. 2B). Several thin dikes of suevite or polymict impact breccia (Fig. 2B) occur from 1562 m to at least 1611 m depth (Reimold et al., 2007; Horton et al., this volume, Chapter 2). Mica schists dominate the upper part of the section. These con-tain several granite veins that become more abundant downward, and the lowermost ~75 m of the core consists almost entirely of granite (Fig. 2B).

/

0

1766

Compositeresults

Postimpact sediments

A B

Figure 2. (A) Eyreville B composite core log (after Gohn et al., 2006). (B) Core log of the lower basement-derived section from 1551.19 m to 1766.32 m depth of the Eyreville B core hole (after Horton et al., this volume, Chapter 2).

258 Townsend et al.


TA

BLE

1. S

AM

PLE

DE

NO

MIN

AT

ION

S A

ND

LIT

HO

LOG

Y W

ITH

CO

RR

ES

PO

ND

ING

DE

PT

H

Sam

ple

Dep

th (

m)

Dep

th (

ft)

Lith

olog

y S

ampl

e D

epth

(m

) D

epth

(ft)

Li

thol

ogy

Sam

ple

Dep

th (

m)

Dep

th (

ft)

Lith

olog

y M

JK18

17

16.5

6 56

31.7

5 M

ica

schi

st

RG

45

1670

.73

5481

.4

Mic

a sc

hist

R

G67

16

01.6

52

54.5

9 M

ica

schi

st

MJK

21

1671

.86

5485

.1

Mic

a sc

hist

R

G46

16

63.7

5 54

58.5

G

rani

te

RG

68

1597

.67

5241

.7

Mic

a sc

hist

R

G01

13

78.1

5 45

21.4

9 A

mph

ibol

ite

RG

47

1655

.29

5430

.74

Mic

a sc

hist

R

G69

15

90.3

1 52

17.5

5 M

ica

schi

st

RG

02

1384

.43

4542

.09

Am

phib

olite

R

G48

16

52.0

2 54

20.0

1 M

ica

schi

st

RG

70

1578

.83

5179

.89

Bio

tite

schi

st

RG

03

1387

.51

4552

.2

Am

phib

olite

R

G49

16

48.5

7 54

08.6

9 M

ica

schi

st

RG

71

1577

51

73.8

8 B

recc

ia

RG

04

1389

.32

4558

.14

Am

phib

olite

R

G50

16

46.0

1 54

00.3

C

alc-

silic

ate

RG

72

1569

.84

5150

.39

Mic

a sc

hist

R

G05

13

85.4

7 45

45.5

1 A

mph

ibol

ite

RG

51

1644

.58

5395

.6

Am

phib

olite

R

G73

15

62.2

2 51

25.3

9 B

recc

ia

RG

06

1387

.24

4551

.31

Am

phib

olite

R

G52

16

42.3

2 53

88.1

9 M

ica

schi

st

RG

74

1559

.51

5116

.5

Bre

ccia

R

G31

17

22.9

1 56

52.5

9 G

rani

te

RG

53

1641

.81

5386

.52

Mic

a sc

hist

R

G75

15

48.9

6 50

81.8

9 B

recc

ia

RG

32

1722

.09

5649

.9

Gra

nite

R

G54

16

38.9

7 53

77.2

M

ica

schi

st

RG

76

1545

.26

5069

.75

Mic

a sc

hist

R

G33

17

19.7

7 56

42.2

9 G

rani

te

RG

55

1637

.63

5372

.8

Mic

a sc

hist

R

G77

15

44.0

7 50

65.8

5 B

recc

ia

RG

34

1717

.47

5634

.74

Mic

a sc

hist

R

G56

16

35.3

7 53

65.3

9 M

ica

schi

st

RG

78

1539

.26

5050

.07

Mic

a sc

hist

R

G35

17

03.7

3 55

89.6

7 G

rani

te

RG

57

1627

.33

5339

.01

Tou

rmal

inite

R

G15

0 13

59.8

44

61.3

G

rani

te g

neis

s R

G36

16

93.2

2 55

55.1

8 G

rani

te

RG

58

1625

.96

5334

.51

Mic

a sc

hist

R

G15

1 13

51.8

2 44

35.1

G

rani

te g

neis

s R

G37

16

91.9

9 55

51.1

5 G

rani

te

RG

59

1622

.28

5322

.44

Gra

nite

R

G15

2 13

51.3

6 44

33.6

G

rani

te g

neis

s R

G38

16

93.0

4 55

54.5

9 M

ica

schi

st

RG

60

1616

.28

5302

.76

Mic

a sc

hist

R

G15

3 13

42.9

8 44

06.1

G

rani

te g

neis

s R

G39

16

88.5

9 55

39.9

9 M

ica

schi

st

RG

61

1611

.93

5288

.48

Gra

nite

R

G15

4 13

15.6

7 43

16.5

G

rani

te g

neis

s R

G40

16

87.1

55

35.1

G

rani

te

RG

62

1609

.77

5281

.4

Bre

ccia

R

G15

5 13

14.2

1 43

11.7

G

rani

te g

neis

s R

G41

16

79.3

55

09.5

1 M

ica

schi

st

RG

63

1607

.3

5273

.29

Bre

ccia

R

G15

6 12

99

4261

.8

Gra

nite

R

G42

16

77.9

5 55

05.0

9 M

ica

schi

st

RG

64

1604

.83

5265

.19

Mic

a sc

hist

R

G15

7 12

90.8

6 42

35.1

G

rani

te

RG

43

1675

.33

5496

.49

Mic

a sc

hist

R

G65

16

03.4

6 52

60.7

B

recc

ia

RG

158

1256

.23

4121

.5

Gra

nite

R

G44

16

72.1

3 54

85.9

9 M

ica

schi

st

RG

66

1604

.39

5263

.75

Gra

nite

R

G15

9 12

51.3

6 41

05.5

G

rani

te g

neis

s S

ampl

e D

epth

(m

) D

epth

(ft)

Li

thol

ogy

Sam

ple

Dep

th (

m)

Dep

th (

ft)

Lith

olog

y R

G16

0 12

48.9

2 40

97.5

G

rani

te g

neis

s W

-037

11

61.3

38

10.0

4 G

rani

te g

neis

s R

G16

1 12

45.9

40

87.6

G

rani

te

W-0

38

1166

.6

3827

.43

Gra

nite

R

G16

2 12

33.7

1 40

47.6

G

rani

te

W-0

39

1175

.1

3855

.31

Gra

nite

gne

iss

RG

163

1212

.86

3979

.2

Gra

nite

gne

iss

W-0

41

1233

.1

4045

.6

Gra

nite

R

G16

4 11

47.5

1 37

64.8

G

rani

te g

neis

s W

-042

12

52

4107

.61

Gra

nite

gne

iss

RG

165

1126

.51

3695

.9

Gra

nite

gne

iss

W-0

43

1263

.7

4146

G

rani

te g

neis

s R

G16

6 10

96.5

9 35

97.7

5 G

rani

te g

neis

s W

-044

12

89.5

42

30.6

4 G

rani

te

RG

167

1655

.4

5431

.1

Gra

nite

W

-045

13

12.6

43

06.4

3 G

rani

te

RG

168

1654

.91

5429

.5

Qua

rtz

vein

W

-046

13

28.8

43

59.5

8 G

rani

te

RG

169

1687

.34

5535

.9

Mic

a sc

hist

W

-047

13

36.4

43

84.5

1 G

rani

te

RG

170

1204

.63

3952

.2

Gra

nite

gne

iss

W-0

48

1360

44

61.9

4 G

rani

te

RG

171

1190

.73

3906

.6

Gra

nite

gne

iss

W-0

49

1367

.9

4487

.86

Gra

nite

R

G17

2 11

75.5

4 38

56.7

5 G

rani

te

W-0

51

1377

.33

4518

.8

Am

phib

olite

R

G17

3 11

77.2

6 38

62.4

G

rani

te g

neis

s W

-052

13

82.9

7 45

37.3

A

mph

ibol

ite

RG

174

1159

.76

3805

G

rani

te

N

ote:

The

dep

th o

f eac

h lit

holo

gic

sam

ple

in th

e E

yrev

ille

core

s is

rec

orde

d ei

ther

in fe

et a

s bo

xed

(ft*

), w

hich

is th

e pr

imar

y re

cord

, or

in r

evis

ed m

eter

s co

mpo

site

dep

th (

rmcd

),

the

com

puta

tiona

lly a

djus

ted

valu

e in

met

ers

rega

rded

to b

e th

e be

st e

stim

ate

of th

e tr

ue d

epth

, as

expl

aine

d in

Edw

ards

et

al.

(thi

s vo

lum

e, C

hapt

ers

3 an

d 4)

, and

Hor

ton

et a

l. (t

his

volu

me,

Cha

pter

2).

Thi

s pa

per

sim

ilarly

use

s th

e gi

ven

core

de

pths

in fe

et a

s bo

xed

(ft*

) an

d in

met

ers

that

are

co

mpu

tatio

nally

adj

uste

d (r

mcd

).

RG

175

1108

.12

3635

.55

Chl

orite

frac

ture

R

G17

6 13

82.1

2 45

34.5

A

mph

ibol

ite

RG

177

1386

.6

4549

.2

Am

phib

olite

R

G17

8 16

41.6

5 53

86

Chl

orite

frac

ture

R

G17

9 16

45.3

4 53

98.1

C

alc-

silic

ate

W-0

35

1105

.9

3628

.28

Gra

nite

W

-036

11

23.8

36

87.0

1 G

rani

te



Mica SchistsThe mica schists are located between 1560.24 m (5118.9 ft)

and 1717.66 m (5635.4 ft). They vary in color from black to silver-gray, are fi ne-grained (average grain size ≤1 mm), foliated, locally porphyroblastic, and contain highly deformed relict sedi-mentary layering and quartz and granite veins. The mica schists consist of muscovite (20–40 vol%), plagioclase (20 vol%), biotite (10–15 vol%), quartz (15–30 vol%), and graphite (5–10 vol%), with accessory sillimanite (fi brolite), garnet, tourmaline, rutile, pyrite, chalcopyrite, and pyrrhotite. Qualitatively, graphite shows a general increase in concentration with depth.

Microscopically, the mica schists make up millimeter-scale mica and quartz-plagioclase foliae that defi ne the folia-tion, which is locally folded (Fig. 3A; see Gibson et al. [this volume] for microstructural analysis). Muscovite is fi ne-grained (average 0.5–1 mm), subidioblastic, and platy, and it deter-mines the dominant foliation with biotite. The grains indicate variable levels of internal strain caused by localized mylonitic deformation (Gibson et al., this volume), including kink banding and undulose extinction. Muscovite also occurs as lath-shaped grains in the graphite-rich cores of plagioclase porphyroblasts. Sericite is a common alteration product of feldspars in the mica

(a)

Fibr

Pl

Mu

2 mm0.5 mm

SilGr

Ms

Cal

A (b)

Grt

Bt 2 mm

(b)

Grt

Bt

2mm5 cm

Bt

Grt

Qtz

B

Am-Ep-Ttn

Bt-Am

Qtz veins

4 cm

C

4 cm

Ves

Ep

Qtz

Cal

D

2.5 mm

Tur

E

GrtPl

Kfs

Bt

2 mm

F

Figure 3. Photomicrographs and hand specimen photographs of rock types in the basement-derived section of the Eyreville B core: (A) Photomicrograph of folds that deform the foliation in mica schist, sample RG60 (see Table 1 for sample depth); plane-polarized light (PPL). (B) Photomicrograph of a skeletal garnet in mica schist, sample RG53; PPL. (C) Amphibo-lite, showing the amphibole + epidote + titanite and biotite + amphibole mineral banding; sample RG51. (D) Calc-silicate rock showing calcite-quartz-epidote veining, sample RG50. (E) Photomicrograph of the tourmalinite, sample RG54; PPL. (F) Fractured garnet in granite, sample RG46; crossed polarizers (XP). Abbreviations (after Siivola and Schmid, 2007): Sil—sillimanite; Pl—plagioclase; Ms—muscovite; Grt—garnet; Bt—biotite; Qtz—quartz; Am—amphibole; Cal—calcite; Ves—vesuvianite; Ep—epidote; Kfs—potassium feldspar; Ttn—titanite.

260 Townsend et al.


schists, occurring together with epidote and calcite. Biotite is fi ne-grained (average ~0.6 mm), subidioblastic, and commonly intergrown with muscovite. It is commonly kinked and altered to chlorite in the vicinity of mylonitic deformation bands and fracture and breccia zones.

Sillimanite occurs in knotted fi brolitic masses up to several millimeters long. Garnet forms skeletal to poikiloblastic por-phyroblasts up to 2 mm in diameter that contain inclusions of biotite, quartz, and rutile. Grains are commonly skeletal, highly fractured, and variably altered to chlorite (Fig. 3B). Plagioclase occurs both in the matrix and as texturally zoned poikiloblastic porphyroblasts up to 2 mm in diameter that contain quartz, tour-maline, muscovite, graphite, and rutile inclusions and have expe-rienced variable amounts of dissolution against the mica foliation (see Gibson et al., this volume). Quartz forms granoblastic to lensoid grains (0.2–1 mm) exhibiting variable amounts of undu-lose extinction and grain-boundary migration recrystallization, depending on the intensity of strain.

Graphite occurs in both disseminated form and as coarser, irregular masses within the mica schist, as well as in the form of inclusions in plagioclase porphyroblasts. It is found in breccia matrices and in fractures in the lower parts of the core.

Tourmaline is fi ne-grained (0.5 mm) and idioblastic to subi-dioblastic. Pyrite is very fi ne-grained (0.1 mm) and occurs as blebs, stringers, or idioblastic crystals, and locally as inclusions in plagioclase porphyroblasts; it is also associated with chlorite and calcite along veins and fractures. Rutile is very fi ne-grained (≤0.1 mm), idioblastic to rounded, and occurs in trace amounts in pyrrhotite, which occurs as individual blebs (0.2 mm) or as aggregates (up to 3.5 mm), and is xenoblastic. Epidote forms a very fi ne-grained (≤0.3 mm) retrograde product in plagioclase adjacent to calcite-fi lled fractures. Chlorite replacing the peak metamorphic assemblage is particularly abundant adjacent to calcite veins and fractures, and in breccias.

Biotite SchistToward the top of the lower basement-derived section,

from 1578.74 m (5179.6 ft) to 1578.96 m (5180.3 ft) depth, there is a 22-cm-thick layer of quartz-plagioclase-biotite schist. It contains 35 vol% quartz, 35 vol% plagioclase, 25 vol% bio-tite, 5 vol% graphite, and accessory rutile and traces of musco-vite. The rock has a distinctive lensoid microstructure—biotite wraps around fi ne-grained lensoid quartz-plagioclase aggre-gates that exhibit internal strain and dynamic recrystallization features. Plagioclase is locally subidioblastic with oscillatory zoning. Biotite grains range up to 2.5 mm in length and exhibit strong kinking and bent cleavage. They contain fi ne-grained graphite inclusions. Graphite also occurs as coarser aggre-gates and seams parallel to the foliation and in crosscutting fractures. Gibson et al. (2007) proposed that very fi ne-grained white mica aggregates in parts of the rock may represent pseu-domorphs after highly poikiloblastic cordierite porphyroblasts; however, no evidence of cordierite has been obtained through microprobe analysis.

AmphiboliteA 58-cm-thick banded amphibolite layer is located from

1644.58 m (5395.6 ft) to 1645.16 m (5397.5 ft) depth in the upper part of the mylonite zone (see Gibson et al., this volume, Fig. 3B therein). This amphibolite is fi ne to medium grained and has millimeter- to centimeter-scale dark-gray to light-gray banding parallel to a foliation defi ned by aligned amphibole grains and aggregates (Fig. 3C). The mineral texture is subidioblastic with seriate-polygonal mineral aggregates. The mineralogy is highly variable within individual bands, some of which contain up to 50 vol% epidote; however, amphibole and plagioclase together constitute more than 50 vol% of the rock, with quartz, biotite, potassium feldspar (K-feldspar), calcite, epidote, and titanite (sphene) occurring in variable proportions in different layers. Amphibole crystals reach up to 2.5 mm in length and range from idioblastic to highly irregular, poikiloblastic grains con-taining inclusions of quartz, plagioclase, epidote, and titanite. They are locally pseudomorphed by uralite and—together with biotite—to chlorite in the vicinity of fractures and calcite veins. Biotite typically occurs as irregular interstitial grains between quartz and feldspar and, locally, as irregular seams parallel to the banding. Titanite (≤0.5 mm) and epidote (≤0.8 mm) commonly display idioblastic crystal shapes and are part of the prograde peak assemblage. Large, irregular quartz grains, locally with evidence of dynamic recrystallization, together with relatively coarse plagioclase-quartz myrmekite, suggest a hydrothermal component added during mylonitization. Disrupted calcite veins may be related to this event or to brecciation that may postdate mylonitization (see discussion in Gibson et al., this volume).

Calc-Silicate RockA strongly brecciated, dark to light green–brown calc-silicate

rock is spatially associated with the amphibolite from 1645.77 m (5399.5 ft) to 1646.44 m (5401.7 ft) depth (Fig. 3D; see also fi gure 3B in Gibson et al., this volume). Although Gibson et al. (2007) described it as an epidosite, the epidote belongs to a ret-rograde assemblage that locally constitutes up to 50 vol% of the rock. The peak metamorphic assemblage consists of vesuvianite (70 vol%), plagioclase (15 vol%), and quartz (15 vol%). Rare irregular relics of an isotropic mineral may be garnet. Vesuvianite is very coarse-grained (up to 10 mm) and forms an interlocking aggregate enclosing plagioclase and quartz, which form irregular grains up to 1.5 mm across.

The brecciation seen in hand specimen extends to the microscale, with intense fracturing of the vesuvianite assem-blage. Fractures vary from large and irregular to intersecting sets of closely spaced parallel fractures. Most are fi lled with epidote. The larger and more irregular fractures contain a polymineralic epidote-calcite-quartz fi ll. The retrograde paragenesis consists of epidote (40 vol%), calcite (25 vol%), quartz (20 vol%), chlorite (10 vol%), and amphibole (5 vol%). Epidote grains locally reach 1.5 mm in length and are subidioblastic where not constrained by fracture margins; calcite is medium-grained (2 mm) and xeno-blastic, whereas quartz is very fi ne-grained (0.1 mm). Chlorite is



fi ne-grained and occurs in irregular masses up to 1 mm across. Aggregates of very fi ne-grained (0.04 mm), acicular, pale-green amphibole, inferred to be actinolite, occur along a single fracture.

TourmaliniteTourmaline-rich assemblages occur in two forms. The more

common form represents dense, locally monomineralic, aggre-gates up to several centimeters in width adjacent to quartz and quartz-feldspar lenses that are interpreted as boudinaged quartz and granite veins (e.g., as seen at 1627 m [5338 ft] depth). Tourma-line also occurs within these veins, where it ranges up to 5 mm in length. Grains are locally bent or broken. The second occurrence is a tourmalinite (Fig. 3E) made up of bands with up to 80 vol% tourmaline at 1639 m depth (5377 ft; sample RG54). Plagioclase, quartz, pyrite, graphite, and rutile constitute the remainder of the assemblage. Given its close proximity to the mylonite zone, which contains several granite veins, the tourmalinite may have formed by fl uid escape from the granite veins; alternatively, tourmaline metasomatism may have accompanied mylonitization.

GraniteGranite in the lower basement-derived section of the

core occurs as pegmatitic to coarse-grained veins in the mica schist but increases in volume downward until it is the domi-nant lithology below 1690 m (Fig. 2B). The granite varies from pink/orange to white. The pink/orange variety is dominated by K-feldspar megacrysts; the orange coloration indicates feldspar sericitization. The white variety is dominated by plagioclase and quartz. The granite has an idiomorphic to hypidiomorphic, inequigranular-polygonal texture. In addition to K-feldspar (20–40 vol%), plagioclase (30–35 vol%), and quartz (20–25 vol%), there are variable amounts of muscovite (0–13 vol%), biotite (0–5 vol%), and garnet (0–2 vol%).

Potassium feldspar is coarse-grained (averages 8 mm, but grain size may increase to 40 mm where the granite is pegmatitic), idiomorphic to subidiomorphic, perthitic, poikilitic (inclusions of muscovite, plagioclase, and quartz), and displays patchy micro-cline twinning. Plagioclase is fi ne- to coarse-grained (0.4–6 mm), subidiomorphic to idiomorphic, and locally poikilitic (inclusions of muscovite and quartz). Quartz and feldspars display signs of internal deformation and dynamic recrystallization related to mylonitic deformation and cataclasis (see Gibson et al., this vol-ume). Quartz is fi ne-grained (0.1–0.6 mm) and occurs as an inter-stitial phase between feldspars. Garnet is fi ne- to medium-grained (0.8–2 mm), idiomorphic to subidiomorphic, and poikilitic, with inclusions of quartz and plagioclase (Fig. 3F). It typically displays fractures. In contrast to the granites in the megablock, muscovite (0.4–0.8 mm) is more common than biotite (0.2–0.4 mm) and locally ranges up to several centimeters across in pegmatites. Chlorite is very fi ne-grained (≤1 mm) and replaces biotite and gar-net in the vicinity of calcite-fi lled fractures and cataclasite zones. The alkali feldspar and plagioclase are locally extensively altered to sericite and saussurite, respectively, with calcite replacing twin lamellae in plagioclase in several samples.

Mylonite ZoneThe mylonitic deformation in the basement-derived rocks is

described in Gibson et al. (this volume); however, the 15-m-thick mylonite zone located between ~1640 m and 1655 m depth is also lithologically signifi cant because it contains a variety of distinctive rock types. In addition to the amphibolite and calc-silicate rock described previously, it includes alternating dark-gray to light-brown and light-pink, very fi ne-grained (<1 mm) layers. Microscopic analysis confi rms that these represent mylonitized mica schists and granite (Fig. 4A). In general, most of the mica schist samples consist of plagioclase (20–30 vol%), quartz (10–20 vol%), muscovite (20–30 vol%), biotite/chlorite (0–20 vol%), graphite (0–10 vol%), fi brolite (0–5 vol%), garnet (0–2 vol%), and pyrite (0–2 vol%). Most granite samples consist of K-feldspar (20–30 vol%), plagioclase (20–30 vol%), quartz (30–40 vol%), muscovite (10 vol%), biotite (0–8 vol%), and gar-net (2 vol%). The light-brown mica schists are more retrogres-sively altered than the dark-gray variety, with biotite replaced by chlorite and muscovite. A more comprehensive description of the microstructures in the mylonite zone is presented in Gibson et al. (this volume).

PETROGRAPHY OF THE GRANITE AND AMPHIBOLITE MEGABLOCKS

The two lithic blocks that occur in the gravelly sand units above the impactite succession in the Eyreville B core (Fig. 5) are largely composed of granite and amphibolite, respectively.

Granite Megablock

The granite megablock is located between 1095.74 m (3595.0 ft) and 1371.11 m (4498.4 ft) depth. It contains two major rock types: gneissic biotite granite that occurs mainly in the upper portion, and biotite granite that is interleaved with the gneiss and dominates the lower portion. The gneiss is intruded by the granite with generally sharp or lit-par-lit contacts with no evidence of chill margins (Fig. 4B); in rare cases, the contacts appear gradational. A third, minor, component of this mega-block is biotite-plagioclase-amphibole schist xenoliths in the granite, which are also found locally in the gneiss. Horton et al. (this volume, Chapter 2) distinguish a fourth component, altered red granite, at the basal contact. A 1–2 cm zone of similar fi ne-grained red granite fl anking thin quartz veins ~15 cm from the top of the megablock suggests that this alteration is related to fl uid movement along pre-impact fractures in the biotite granite.

Gneissic Biotite GraniteThe gneiss is light -gray to pink, fi ne- to coarse-grained

(1–2.5 mm), and displays millimeter- to centimeter-scale banding. It constitutes ~29% of the total granite megablock core intersection (Fig. 5). The gneiss consists of K-feldspar (25–35 vol%), quartz (25 vol%), plagioclase (25 vol%), biotite (10–15 vol%), pyrite (0–5 vol%), and magnetite (0–5 vol%),

262 Townsend et al.


and it has a hypidiomorphic, inequigranular-interlobate texture. The color variation is a product of varying biotite and K-feldspar contents. The K-feldspar is hypidiomorphic, poikilitic, and phenocrystic (up to 2.5 mm), and it exhibits fl ame perthite and microcline twinning. Inclusions in the phenocrysts are quartz, idiomorphic to hypidiomorphic plagioclase, and biotite. The matrix consists of medium-grained quartz and plagioclase (2 mm). Biotite is very fi ne-grained (0.1 mm), hypidiomorphic, locally retrogressed to chlorite, and defi nes the foliation. Pyrite and magnetite are fi ne-grained (1–2 mm) and idiomorphic, and magnetite may be surrounded by diffusion haloes (up to 2.5 mm).

Biotite GraniteThe biotite granite is pink to gray-white, medium- to coarse-

grained (5–10 mm), and locally pegmatitic. The colors refl ect variations in the K-feldspar and biotite contents. The granite is generally massive but locally contains a weak foliation defi ned

by alkali feldspar and/or biotite that is most visible adjacent to the gneiss. It constitutes ~70% of the total granite megablock. The granite has a hypidiomorphic, inequigranular-interlobate mineral texture.

The granite consists of K-feldspar (20–50 vol%), quartz (20–35 vol%), plagioclase (15–30 vol%), biotite (2–10 vol%), muscovite (2 vol%), magnetite (0–3 vol%), and pyrite (0–2 vol%). The K-feldspar is locally megacrystic (up to 15 mm), idiomor-phic to hypidiomorphic, poikilitic, and exhibits fl ame perthite, microcline twinning, and myrmekitic margins. Inclusions in the megacrysts are quartz, plagioclase, and biotite. Quartz is fi ne-grained (up to 2 mm), subhedral, and exhibits undulose extinc-tion and locally sutured contacts. Plagioclase is medium-grained (2 mm) and hypidiomorphic. Biotite is medium-grained (up to 2 mm), and muscovite is very fi ne-grained (0.3 mm) and most commonly occurs pseudomorphing biotite, together with chlo-rite. The absence of evidence of signifi cant crystal plastic fl ow in

2 mm

Bt

Grt

Pl

Qtz

A

2.5 cm

Granite

Gneiss

ContactB

2.5 cm

Bt schist

Granite

C

2.5 cm

Am

Am

D

Figure 4. (A) Photomicrograph showing the highly foliated nature of the mylonite zone in the lower basement-derived section, RG40a; crossed polarizers (XP). (B) Core photograph of the megablock gneissic biotite granite and biotite granite contact. (C) Core photograph of the schist xenoliths within the biotite granite megablock. (D) Core photograph of mega-block amphibolite, with amphibole glomerocrysts. For abbreviations, see Figure 3 caption.



the matrix suggests that the alkali feldspar and biotite alignment is a magmatic foliation. Pyrite (1 mm) and magnetite (1 mm) are hypidiomorphic. Diffusion haloes (up to 2 mm) are found around magnetite only in close proximity to the biotite gneiss.

The red granite at the base of the megablock (Horton et al., this volume, Chapter 2) is characterized by comprehensive feld-spar alteration to submicroscopic aggregates.

SchistSeveral small (usually <10 cm, but up to 24 cm at 1354 m

[4441.0 ft] to 1354 m [4442.7ft] depth), black to dark-green schist xenoliths comprising a very fi ne- to medium-grained (0.2–2 mm) biotite-plagioclase-quartz-epidote ± amphibole assemblage occur mostly within the biotite granite (Fig. 4C); however, rare xeno-liths occur also within the gneiss. The schist consists of biotite

0

1766

Compositeresults

Postimpact sediments

1096

Figure 5. Core log of the granite and amphibolite megablocks from depth 1095.74 m to depth 1389.71 m of the Eyreville B core hole (after Gohn et al., 2006; Horton et al., 2007, this volume, Chapter 2).

264 Townsend et al.


(35 vol%), plagioclase (35–40 vol%), quartz (10–18 vol%), pyrite (2–10 vol%), epidote (2–10 vol%), and amphibole (0–10 vol%).

The biotite is very fi ne- to medium-grained (0.2–2 mm) and idioblastic to subidioblastic. Plagioclase in the matrix is very fi ne grained (0.2 mm) and subidioblastic to xenoblastic, but it forms porphyroblastic grains up to 1 mm in length that overgrow the foliation. Pyrite is fi ne-grained (1 mm) and usually occurs as subidioblastic aggregates. Epidote is fi ne-grained (0.8 mm), subi-dioblastic, and poikiloblastic. Amphibole, where present, occurs as rare radiating or sheaf-like aggregates up to 1.5 mm in diam-eter. The contacts between the schist and granite are commonly highly irregular, and coarse plagioclase aggregates are locally found adjacent to the granite contacts.

Amphibolite Megablock

The amphibolite megablock lies in sand between the granite megablock and impactite succession, from 1376.38 m (4515.7 ft) to 1389.71 m (4559.4 ft) depth (Horton et al., this volume, Chap-ter 2). It is black to dark-gray to dark-green, fi ne- to medium-grained (1–4 mm), relatively homogeneous in composition, and locally foliated (Fig. 4D). Horton et al. (this volume, Chapter 2) describe a metagabbroic to metadioritic component with a rel-ict igneous texture between 1386.93 m and 1387.54 m depth, which correlates with our sample RG03. All our other samples have an idioblastic to subidioblastic, inequigranular-interlobate metamorphic texture. The amphibolite consists of amphibole (40–50 vol%), plagioclase, (20–25 vol%), biotite (10–20 vol%), quartz (10 vol%), epidote (0–5 vol%), chalcopyrite (0–2 vol%), pyrrhotite (0–2 vol%), pyrite (<1 vol%), magnetite (<1 vol%), and titanite (<1 vol %).

Amphibole is commonly idioblastic, relatively unaltered, and locally forms glomerocrysts up to 40 mm in size (Fig. 4D) that contain very fi ne-grained (0.1 mm) quartz and plagioclase inclusions. Plagioclase is fi ne-grained (0.6 mm), subidioblastic, and locally altered to calcite and epidote. Biotite is very fi ne-grained (0.4 mm), subhedral, and altered to chlorite; the bio-tite tends to increase in volume percent toward the base of the megablock. Quartz is very fi ne-grained (0.1 mm), subhedral, and exhibits undulose extinction. Epidote is very fi ne-grained (0.1 mm) and xenoblastic. The amphibolite contains quartz and calcite veins and chlorite-fi lled fractures. Chalcopyrite, pyrrho-tite, pyrite, magnetite, and titanite are very fi ne-grained (0.3 mm) and subidioblastic.

GEOCHEMISTRY

Bulk-rock X-ray fl uorescence (XRF) chemical analyses were obtained on 73 samples, 42 of which were analyzed at the Humboldt University in Berlin (RG and W series) and 31 at the University of Vienna (CB6 series). Analytical geochemistry for major and trace elements was obtained on these 73 samples, of which 43 were from the basement-derived section (18 mica schist, 28 granite), 26 were from the granite megablock, and

4 were from the amphibolite megablock. Further details about the analytical procedure and the XRF results are recorded in Table A1 of Schmitt et al. (this volume). Mineral chemical anal-yses were obtained from 21 samples, of which 7 were from the basement-derived mica schist, 1 was from the basement-derived amphibolite, 1 was from the basement-derived calc-silicate, 4 were from the basement-derived granite, 6 were from the gran-ite megablock, and 2 were from the amphibolite megablock. Table 2 details the mineral chemistry of selected minerals within the analyzed samples.

Mineral Chemistry

Mineral chemical analyses were obtained using a Cameca SX 100 electron microprobe at the University of Pretoria (South Africa), with a 20 kV acceleration potential, a 20 nA beam cur-rent, and a defocused beam size of 10 µm. Counting times were 20 s on peak position and 10 s on each background position. The following X-ray lines, spectrometer crystals, and standards (in brackets) were used: SiKα, TAP (KP garnet); CaKα, PET (KP garnet); AlKα, TAP (KP garnet); MgKα, TAP (diopside); FeKα, LLIF (KP garnet); MnKα, LLIF (rhodonite); TiKα, PET (rutile); BaLa, LLIF (barite); KKα, PET (KP hornblende); and NaKα, LTAP (KP hornblende). The AX activity-composition calcula-tion program for rock-forming minerals (Holland and Powell, 2000) was used in the recalculation of the data.

Elemental ratios are defi ned by the following equations: X

Mg = Mg/(Mg + Fe2+); X

Mn = Mn/(Mn + Fe2+ + Mg + Ca); X

Ca =

Ca/(Ca + Fe2+ + Mg + Mn); An = 100Ca/(Ca + Na); Or = 100K/(K + Na); Ps = 100Fe3+/(Fe3+ + Al).

Garnet within mica schist of the lower basement-derived section has the following ranges: X

Mg = 0.13–0.16, X

Mn = 0.10–

0.42, and XCa

= 0.05–0.08. Most garnets show high spessartine content (X

Mn > 0.25) and relatively fl at compositional profi les,

but a garnet in sample RG53 (1641.8 m; 5386.5 ft) displays a spessartine-rich marginal zone and a spessartine-poorer, almandine-richer core. X

Mg in this garnet slightly increases in

the core but decreases in the marginal zone. We interpret this zonation as prograde growth zoning in the core and retrograde diffusional zoning in the marginal zone related to garnet break-down. This is supported by geothermometric calculations (Gib-son et al., this volume).

Garnet within granite of the lower basement-derived section is less magnesian and calcic, and generally more spessartine-rich, than the mica schist garnets (X

Mg = 0.01–0.08, X

Mn = 0.23–0.37,

XCa

= 0.01–0.05) and has fl at compositional profi les. Considering the textural differences (the granite garnets are less poikilitic and commonly idioblastic), we conclude that the garnet in the granite crystallized from the melt, rather than being a xenocrystic phase inherited from the mica schists.

Biotite is ubiquitous in all of the rock types studied. Its XMg

ranges in the mica schists between 0.42 and 0.69, and it contains Alvi = 1.64–1.87, K = 0.75–0.91, and Ti = 0.12–0.19 per formula unit (11 oxygens), with values increasing slightly with increasing



TABLE 2. REPRESENTATIVE ELECTRON MICROPROBE ANALYSES

Sample: RG51 W51 W52 RG70 RG51 RG40a W35 W36 W45 RG70 Mineral: Am Am Am Bt Bt Bt Bt Bt Bt Chl Lithology: B-A MB-A MB-A B-MS MB-A B-G MB-G MB-G MB-G B-MS (wt%) SiO2 30.1 43.7 51.43 39.34 36.86 34.83 34.95 35.38 36.16 38.07 TiO2 35.03 0.77 0.24 1.53 2.51 2.23 3.39 2.76 2.8 1.76 Al2O3 2.96 13.66 4.96 16.17 15.77 18.8 17.66 17.21 17.44 16.05 Cr2O3 0 0 0 0 0 0 0 0 0 0 Fe2O3 1.09 2.47 2.15 2.06 0 0 0 0 0 0 FeO 0 11.99 14.41 10.49 18.12 25.78 21.95 22.79 21.99 13.1 MnO 0.09 0.23 0.4 0.35 0.3 0.58 0.51 0.3 0.42 0.37 MgO 0.02 11.01 16.31 17.64 11.89 4.25 8.05 7.83 8.61 17.13 CaO 29.08 11.53 6.83 0 0 0.02 0 0 0.04 0.05 Na2O 0.02 1.64 0.61 0.2 0.1 0.1 0.08 0.09 0.12 0.17 K2O 0.01 0.34 0.07 8.51 9.88 8.36 9.9 9.68 9.58 7.2 Totals 98.29 97.09 97.2 96.29 95.44 94.96 96.5 96.05 97.17 93.91 Oxygens 23 23 23 11 11 11 11 11 11 14 (cations) Si 4.596 6.431 7.436 2.831 2.798 2.735 2.681 2.729 2.739 3.582 Ti 4.022 0.085 0.026 0.083 0.143 0.132 0.196 0.16 0.16 0.125 Al 0.533 2.37 0.845 1.372 1.411 1.74 1.597 1.565 1.557 1.781 Cr 0 0 0 0 0 0 0 0 0 0 Fe3+ 0.125 0.273 0.234 0.111 0 0 0 0 0 0 Fe2+ 0 1.475 1.743 0.631 1.15 1.693 1.408 1.47 1.393 1.031 Mn 0.012 0.029 0.049 0.021 0.019 0.039 0.033 0.02 0.027 0.029 Mg 0.005 2.415 3.514 1.892 1.345 0.497 0.92 0.9 0.972 2.402 Ca 4.757 1.818 1.058 0 0 0.002 0 0 0.003 0.005 Na 0.006 0.468 0.171 0.028 0.015 0.015 0.012 0.013 0.018 0.031 K 0.002 0.064 0.013 0.781 0.957 0.837 0.969 0.953 0.926 0.864 Sum 14.095 15.52 15.167 7.75 7.839 7.69 7.816 7.812 7.795 9.851

(Continued )

TABLE 2. REPRESENTATIVE ELECTRON MICROPROBE ANALYSES (Continued )

Sample: W51 RG46 W35 RG50 RG51 RG48 RG53 RG32 RG40 RG40a Mineral: Chl Chl Chl Ep Ep Grt Grt Grt Grt Grt Lithology: MB-A B-G MB-G B-CS B-A B-MS B-MS B-G B-G B-G (wt%) SiO2 36.66 36.81 31.54 37.58 38.15 40.6 37.28 36.18 36.22 36.18 TiO2 1.96 0.88 2.22 0.07 0.05 0.07 0 0.05 0.01 0.08 Al2O3 16.19 26.56 17.01 25.55 25.22 19.61 21.07 20.87 21.09 21.14 Cr2O3 0 0 0 0 0 0 0 0 0 0 Fe2O3 0 0 0 0 0 0 0.08 0.35 0.54 0.65 FeO 15.71 19.91 24.44 9.71 10.93 20.63 31.23 27.51 27.27 29.51 MnO 0.06 0.41 0.47 0.18 0.13 15.05 4.77 14.14 13.41 10.44 MgO 13.98 2.83 10.47 0.04 0.03 1.88 3.17 0.27 0.83 0.81 CaO 0.04 0.03 0.73 24.12 23.43 1.96 2.2 0.67 0.7 1.19 Na2O 0.35 0.1 0.14 0.01 0.01 0.04 0.01 0.03 0.02 0.06 K2O 7.2 2.83 4.72 0 0.01 0 0 0 0 0 Totals 92.16 90.36 91.74 97.27 97.96 99.84 99.8 100.03 100.03 100 Oxygens 14 14 14 12.5 12.5 12 12 12 12 12 (cations) Si 3.553 3.226 2.974 2.982 3.228 2.999 2.976 2.966 2.96 Ti 0.143 0.064 0.171 0.004 0.003 0.004 0 0.003 0.001 0.005 Al 1.857 3.022 2.051 2.384 2.324 1.838 1.998 2.024 2.036 2.039 Cr 0 0 0 0 0 0 0 0 0 0 Fe3+ 0 0 0 0.527 0.707 0 0.005 0.022 0.033 0.04 Fe2+ 1.278 1.607 2.091 0.116 0.007 1.372 2.101 1.892 1.868 2.019 Mn 0.005 0.034 0.041 0.012 0.009 1.014 0.325 0.985 0.93 0.724 Mg 2.027 0.407 1.596 0.005 0.003 0.223 0.38 0.033 0.101 0.099 Ca 0.004 0.003 0.08 2.045 1.962 0.167 0.19 0.059 0.061 0.104 Na 0.066 0.019 0.028 0.002 0.002 0.006 0.002 0.005 0.003 0.01 K 0.894 0.348 0.616 0 0.001 0 0 0 0 0 Sum 9.842 9.056 9.9 8.068 8.001 7.852 8 8 8 8

(Continued )

266 Townsend et al.



Sample: RG46 RG70 RG51 W35 W45 RG68 RG70 RG40 W35 W45 Mineral: Grt Kfs Kfs Kfs Kfs Mu Mu Mu Mu Mu Lithology: B-G B-MS B-A MB-Gn MB-G B-MS B-MS B-G MB-Gn MB-G (wt%) SiO2 36.41 40.3 63.73 63.18 64.32 45.97 31.61 48.07 28.75 45.92 TiO2 0 1.6 0.02 0.03 0.01 1.02 1.51 0.25 1.1 2.6 Al2O3 20.73 16.83 18.87 18.99 19.04 34.59 19.45 36.39 18.85 19.13 Cr2O3 0 0 0 0 0 0 0 0 0 0 Fe2O3 0 13.56 0.09 0 0 0 17.67 0 19.86 12.24 FeO 28.7 0 0 0 0 0.76 6.81 1.37 7.66 4.72 MnO 12.41 0.31 0 0.01 0.01 0.02 0.36 0.03 0.57 0.22 MgO 0.58 17.6 0 0.01 0.01 1.11 12.19 0.61 11.33 5.75 CaO 0.85 0.03 0 0.04 0.02 0 0.06 0 0.13 0.96 Na2O 0.02 0.25 0.83 0.83 1.15 0.65 0.09 0.4 0.14 5.69 K2O 0 7.26 15.82 15.93 14.89 9.73 3.86 9.87 3.25 1.99 Totals 99.7 97.75 99.37 99.04 99.46 93.86 93.61 97 91.64 99.22 Oxygens 12 8 8 8 8 11 11 11 11 11 (cations) Si 2.997 2.022 2.967 2.955 2.975 3.087 2.372 3.117 2.242 3.082 Ti 0 0.06 0.001 0.001 0 0.052 0.085 0.012 0.065 0.131 Al 2.012 0.995 1.036 1.047 1.038 2.739 1.72 2.782 1.733 1.514 Cr 0 0 0 0 0 0 0 0 0 0 Fe3+ 0 0.512 0.003 0 0 0 0.997 0 1.166 0.618 Fe2+ 1.976 0 0 0 0 0.043 0.427 0.074 0.5 0.265 Mn 0.865 0.013 0 0 0 0.001 0.023 0.002 0.038 0.013 Mg 0.071 1.316 0 0.001 0.001 0.111 1.363 0.059 1.317 0.575 Ca 0.075 0.002 0 0.002 0.001 0 0.005 0 0.011 0.069 Na 0.003 0.024 0.075 0.075 0.103 0.085 0.013 0.05 0.021 0.74 K 0 0.465 0.94 0.951 0.879 0.834 0.369 0.817 0.323 0.17 Sum 7.999 5.409 5.021 5.034 4.997 6.951 7.376 6.914 7.416 7.177

(Continued )


Sample: RG68 RG70 RG51 W51 W52 RG40 W35 W36 RG50 Mineral: Pl Pl Pl Pl Pl Pl Pl Pl Ves Lithology: B-MS B-MS B-A MB-A MB-A B-G MB-Gn MB-Gn B-CS (wt%) SiO2 64.37 59.46 56.68 56.28 57.19 64.4 61.38 60.99 36.96 TiO2 0 0 0.04 0.04 0.01 0 0 0 1.08 Al2O3 22.16 25.57 27.28 27.02 26.69 22.31 24.34 24.5 17.67 Cr2O3 0 0 0 0 0 0 0 0 0 Fe2O3 0.02 0.14 0.18 0.19 0.04 0 0.03 0.09 0 FeO 0 0 0 0 0 0 0 0 3.52 MnO 0 0.03 0 0 0.01 0.01 0.01 0.02 0.07 MgO 0 0 0 0.01 0.01 0 0 0 1.71 CaO 3.13 6.89 8.99 9.63 8.86 2.96 5.44 5.6 36.05 Na2O 9.77 7.87 6.68 6.14 6.52 10.09 8.65 8.55 0.14 K2O 0.1 0.11 0.16 0.06 0.04 0.13 0.19 0.23 0 Totals 99.55 100.07 100.01 99.37 99.37 99.9 100.04 99.98 97.23 Oxygens 8 8 8 8 8 8 8 8 53 (cations) Si 2.848 2.651 2.546 2.545 2.577 2.842 2.725 2.713 19.08 Ti 0 0 0.001 0.001 0 0 0 0 0.42 Al 1.156 1.344 1.445 1.44 1.418 1.161 1.274 1.285 10.75 Cr 0 0 0 0 0 0 0 0 0.00 Fe3+ 0.001 0.005 0.006 0.006 0.002 0 0.001 0.003 0.00 Fe2+ 0 0 0 0 0 0 0 0 1.52 Mn 0 0.001 0 0 0 0 0 0.001 0.03 Mg 0 0 0 0.001 0.001 0 0 0 1.32 Ca 0.148 0.329 0.433 0.467 0.428 0.14 0.259 0.267 19.94 Na 0.838 0.68 0.582 0.538 0.57 0.863 0.745 0.737 0.14 K 0.006 0.006 0.009 0.003 0.002 0.007 0.011 0.013 0.00 Sum 4.996 5.017 5.022 5.002 4.998 5.013 5.015 5.019 53.196

Note: Data in weight percent. Abbreviations used: B-A—basement-derived amphibolite; MB-A—megablock amphibolite; B-MS—basement-derived mica schist; B-G—basement-derived granite; MB-G—megablock granite; MB-Gn—megablock granite gneiss; Am—amphibole; Bt—biotite; Chl—chlorite; Ep—epidote; Grt—garnet; Kfs—potassium feldspar; Mu—muscovite; Pl—plagioclase feldspar; Ves—vesuvianite.



XMg

. The most highly magnesian biotites (XMg

= 0.62–0.75) are found in the quartz-plagioclase-biotite schist at 1578.74 m to 1578.96 m (5179.6 ft to 5180.3 ft) depth, and we speculate that it may be a metavolcanic or metavolcaniclastic rock, rather than a metapelite. Biotite in the megablock amphibolite has higher X

Mg

and Alvi (XMg

= 0.7; Alvi = 1.50–1.51) than biotite in the amphibo-lite from the lower basement-derived section (X

Mg = 0.53–0.56;

Alvi = 1.37–1.52). However, K and Ti values show the reverse relationship, being generally higher in the biotite of the lower basement-derived amphibolite (K = 0.96–0.97; Ti = 0.13–0.17, versus K = 0.78; Ti = 0.08 in the megablock). The K-rich nature of the lower basement-derived amphibolite is consistent with the presence of K-feldspar in the rock.

The XMg

of biotite in the granite of the lower basement-derived section (X

Mg = 0.20–0.25) is lower than that in the mega-

block granite (XMg

= 0.34–0.46) and megablock granite gneiss (X

Mg = 0.25–0.27). Biotite K in the lower basement-derived gran-

ite (K = 0.78–0.93) shows a slightly narrower range than in the megablock granite (K = 0.80–1.02) and granite gneiss (K = 0.78–0.95). Ti values are similar throughout the basement-derived and megablock granites, showing a slightly narrower range in biotite in the lower basement-derived granite (Ti = 0.08–0.17), in com-parison to the megablock granite (Ti = 0.05–0.23) and gneissic granite (Ti = 0.15–0.21). Higher Al values for biotite in the lower basement-derived granite (Alvi = 1.69–1.85) occur in comparison to those in the megablock granite (Al = 1.40–1.69) and gneissic granite (Al = 1.59–1.66).

Chlorite in mica schists of the lower basement-derived sec-tion has formed through retrogression of biotite and garnet. It displays a wide range of X

Mg values (0.25–0.72), but most values

lie in the range XMg

= 0.40–0.44. Chlorite associated with biotite schist of the lower basement-derived section has the highest X

Mg

values (0.5–0.7), and most occur within the range XMg

= 0.64–0.70. This is consistent with the Mg-rich nature of the biotite in the assemblage.

Chlorite in the megablock amphibolite has higher XMg

values (0.59–0.62) than the basement amphibolite (X

Mg = 0.32–0.40),

whereas chlorite in the granite of the lower basement-derived section and the megablock has the lowest values of X

Mg (0.20

and 0.20–0.43, respectively). The chlorite within the megablock gneissic granite is slightly more magnesian (X

Mg = 0.25–0.28),

whereas the aluminum content in the chlorite is higher in gran-ite in the lower basement-derived section (Alvi = 3.02) compared to the content in the megablock granite (Alvi = 2.05–2.2) and gneissic granite (Alvi = 2.51–2.79).

Plagioclase in mica schists of the lower basement-derived section has a broad compositional range between albite and andesine (An

1–34; Fig. 6A), and most analyses lie in the ranges

An2–7; 13–17; 30–33

. Slight compositional zoning is observed in por-phyroblasts, where textural zoning is prevalent. Plagioclase in the amphibolites is more calcic than that in the mica schists, and it var-ies from andesine (An

39–43) in the lower basement-derived section

to andesine to labradorite (An35–54

) in the amphibolite megablock. Thin albitic rims (An

0–9) are also found in the basement-derived

A

B

50

60

70

80

90

1000

10

20

30

40

50

Na

0.45

0.5

0.55

0.6

0.65

0.7

0.75

0.8

6 6.1 6.2 6.3 6.4 6.5 6.6 6.7

Si

Na

+ K

Pargasite Edenite

Tschermakite Hornblende

CaK

Figure 6. Mineral compositions show-ing (A) K-Na-Ca abundances for pla-gioclase, and (B) compositions in terms of Na + K versus Si abundances for am-phibole (after Deer et al., 1992). Sym-bols: closed black circles—megablock biotite granite, closed gray circles—megablock biotite gneissic granite, open circles—granite of lower basement-derived section, closed black squares—megablock amphibolite, closed black triangles—megablock biotite schist, open triangles—mica schist of lower basement-derived section.

268 Townsend et al.


section. Strong reverse zonation (An35–54

) occurs in the amphib-olite megablock. The granite of the lower basement-derived section contains plagioclase with An

5–16 (albite to oligoclase),

whereas the granite megablock shows a much wider range, from albitic to andesinitic (An

2–9, An

12–37), again consistent with the

greater range of lithologies in the megablock.Potassium feldspar compositions range from Or

77 to Or

96

in the megablock granite and gneiss, and from Or73

to Or96

in the lower basement-derived granite. K-feldspar in the lower basement-derived section ranges from Or

77 to Or

100 in the

amphibolite, Or83

to Or92

in the mica schist, and Or95

to Or96

in the quartz-plagioclase-biotite schist.

Amphibole in the amphibolite in the lower basement-derived section is pargasitic in composition ([Na + K] = 0.58–0.75 atoms per formula unit [apfu] for 23 oxygens), in comparison to the megablock amphibole (Fig. 6B), which is pargasitic to horn-blendic ([Na + K] = 0.48–0.59 apfu; Leake, 1978). Amphibole in the megablock is slightly more magnesian (X

Mg = 0.60–0.67) than

that in the lower basement-derived section (XMg

= 0.48–0.60).White mica is dominantly muscovite with <10 mol% para-

gonite, except in a few rare cases where higher Na and Ca con-tents occur (Fig. 7A). Muscovite ranges from X

Mg = 0.18 to

XMg

= 0.89 in the mica schist, mostly between XMg

= 0.45 and X

Mg = 0.80, and Si:Al = 1.44:1 (Fig. 7A). Muscovite observed

within the lower basement-derived section granites is the least Mg-rich (X

Mg = 0.22–0.72; most lie in the range X

Mg = 0.26–

0.57), which is consistent with the XMg

data for the ferromagne-sian minerals (Fig. 7B). The Si:Al value is 1.96:1.

The megablock granite and gneissic granite contain mus-covite with X

Mg = 0.31–0.73 and X

Mg = 0.43, respectively. The

Si:Al ratios for the megablock granite (1.37:1) and granite gneiss (1.73:1) are slightly lower than in the basement-derived granites. Muscovite in the lower basement-derived granite shows K:Na = 2.57:1, whereas the value for muscovite in the megablock granite is much higher (K:Na = 22.34:1). The mega-block granite gneiss shows K:Na = 1.20:1. Alvi in muscovite in the lower basement-derived granite apparently increases slightly with increasing X

Mg.

Epidote compositions in the lower basement-derived amphibolite show a narrower range of pistacite contents (Ps

19–24) than in the lower basement-derived calc-silicate rocks

(Ps5–24

), although most values lie in the range Ps15–24

. Slight zoning in the lower basement-derived amphibolite and strong zoning in the calc-silicate rocks occur. Higher pistacite con-tents are characteristic of epidote that forms part of the peak assemblage, whereas epidote characterizes the calcite-quartz-chlorite- epidote fractures.

Vesuvianite compositions are relatively uniform through-out the sample despite the coarse grain size, and calcium con-tent varies from 19.78 to 20.22 apfu for 53 oxygens. The Ti con-tent ranges from 0.39 to 0.54 apfu and is accompanied by low Na (≤0.17 apfu), consistent with Groat et al.’s (1992) prediction that Na is less than 0.2 apfu in vesuvianite where Ti values are less than 0.75 apfu.

0 10 20 30 40 50 60 70 80 90 1000

10

20

30

40

50

60

70

80

90

1000

10

20

30

40

50

60

70

80

90

100

Na

K

A

Ca

0 10 20 30 40 50 60 70 80 90 1000

10

20

30

40

50

60

70

80

90

1000

10

20

30

40

50

60

70

80

90

100

Fe

Mg

B

Ti

Figure 7. (A) K-Na-Ca and (B) Ti-Fe-Mg abundances for white mica. Symbols: closed black circles—megablock biotite granite, closed gray circles—megablock biotite gneissic granite, open circles—granite of lower basement-derived section, closed black squares—megablock amphibolite, closed black triangles—megablock biotite schist, open triangles—mica schist of lower basement-derived section.

Bulk-Rock Chemistry

Seventy of the eighty-four samples studied were selected for bulk-rock chemical analysis using X-ray fl uorescence (XRF): 18 samples from the metasedimentary succession, 25 from granite of the lower basement-derived section, 4 from the amphibolite megablock, and 23 from the granite megablock, including 3 samples from the biotite-schist xenoliths in this megablock. Rep-resentative analyses of the RG (1–6, 38–40, 44, 53, 54, 61, 64,



68, 70, and 76), W, and CB6 series are presented in Table A1 of Schmitt et al. (this volume). The chemical analyses for the RG, W, and CB6 series were obtained with a SIEMENS SRS 3000 at the Berlin Museum für Naturkunde. Methods of sample preparation and instrumental analysis are detailed in Schmitt et al. (2004). The data are summarized in Harker diagrams (Fig. 8). Further discussion of the chemical results is provided by Schmitt et al. (this volume).

Trace-element analyses were also performed on the RG, W, and CB6 series using XRF methods at the Museum of Natural History (Mineralogy), Humboldt University of Ber-lin, Germany. Schmitt et al. (this volume) also report INAA (instrumental neutron activation analysis) data for the W and CB6 series, obtained at the Department of Lithospheric Research, Centre for Earth Sciences, University of Vienna, Austria. In this paper, comparisons in trace elements for the

0

2

4

6

8

10

12

0

2

4

6

8

10

12

14

16

0

5

10

15

20

25

30

0

2

4

6

8

0

5

10

15

20

6

8

10

12

14

16

18

20

22

24

26

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

30 40 50 60 70 80 90

30 40 50 60 70 80 90

30 40 50 60 70 80 90 30 40 50 60 70 80 90

30 40 50 60 70 80 90

30 40 50 60 70 80 90

30 40 50 60 70 80 90

Fe2O3

Al2O3

MgO

CaO

K2O

Na2O

TiO2

SiO2

SiO2

Figure 8. Harker variation diagrams showing major-element bulk-rock com-positions (all data in wt%) versus silica content for the lower basement-derived and megablock lithologies. Symbols: closed black circles—megablock biotite granite; closed gray circles—megablock biotite granite gneiss; open circles—lower basement-derived granite; closed black squares—megablock amphibo-lite; closed black triangles—megablock biotite schist; open triangles—lower basement-derived mica schist; closed in-verted triangles—late Paleozoic granitic samples from the Raleigh metamorphic belt and the Eastern Slate Belt (Coler et al., 1997).

270 Townsend et al.


lower basement-derived and megablock granites were based on the XRF data.

Mica SchistsThe lower basement-derived mica schists show a broad

range in SiO2 contents from 40 to 68 wt%, consistent with the

presence of fi ne-scale layering and highly disrupted quartz veins. Only Fe

2O

3 shows a negative correlation with SiO

2. It is

unclear whether the poor correlation of CaO, Na2O, K

2O, and

Al2O

3 refl ects hydrothermal alteration. Although every attempt

was made to avoid veins and breccias, the rocks show extensive hydrothermal alteration both in the altered metamorphic assem-blages and in the abundant calcite and quartz veining. Figure 8 shows that schist xenoliths in the granite megablock are chemi-cally distinct from mica schists in the lower basement-derived section, being signifi cantly enriched in TiO

2, MgO, CaO, and

Fe2O

3, and depleted in Al

2O

3 and SiO

2 (49–56 wt %). This is con-

sistent with the differences in mineralogy described previously. Na

2O/K

2O ratios of the lower basement-derived mica schist and

the megablock schist lie between 0.1 and 0.8, with an average of 0.4, less than half the average pelitic ratio, according to Taylor and McLennan (1985).

Biotite schist sample RG70, which contains unusually mag-nesian biotite and lacks signifi cant muscovite, has a high MgO content (4.6 wt%) compared to all other rocks analyzed and is depleted in Al

2O

3 relative to the other mica schists. It also shows

elevated Na2O and, to a lesser extent, CaO values, which are

consistent with the high plagioclase content. The Na2O/K

2O

ratio for the biotite schist sample RG70 is 0.6, which is only slightly less than the average pelitic Na

2O/K

2O ratio due to its

enrichment in MgO.The lowest SiO

2 content (40.4 wt%) is found in the tourma-

linite (sample RG54), which also has elevated Al2O

3 and Fe

2O

3

values, but which is depleted in MgO and K2O relative to the

mica schists, refl ecting the absence of micas.

Amphibolites and Calc-Silicate RockWith the exception of the tourmalinite, which we interpret as

a highly metasomatized rock, the megablock amphibolite sam-ples show the lowest SiO

2 contents (Fig. 8), although one sample

contains >60 wt% SiO2, which probably refl ects quartz veining.

The megablock amphibolite is generally signifi cantly enriched in Fe

2O

3 and CaO relative to the mica schists but shows less

enrichment of Fe2O

3 and TiO

2. It is depleted in K

2O and Al

2O

3

but shows similar Na2O contents. There is a comparatively closer

correlation between the amphibolite and the schist xenoliths from the granite megablock, which reinforces Schmitt et al.’s (this vol-ume) suggestion that the schist xenoliths form a distinct suite separate from the mica schists.

The calc-silicate rock, described as epidosite by Schmitt et al. (this volume) after Gibson et al. (2007), has the highest CaO content of all rocks analyzed (14.3–25.6 wt%), although quartz veining is refl ected in a single sample (SiO

2 > 60 wt%). The calc-

silicate rock also shows the lowest Na2O values (0.11–0.37 wt%)

and low K2O values (<0.23 wt%). The high CaO content, coupled

with the relatively high Al2O

3 (7.2–14.8 wt%) and Fe

2O

3 (5.65–

7.26 wt%) values, is consistent with marly sediments (Pettijohn, 1957). We conclude that the calc-silicate rock is a metamarl.

Granitic Rock TypesThe lower basement-derived granite and pegmatites

and the megablock gneissic granite and granite show strong compositional overlap (Fig. 8). Considering the diffi culty in obtaining representative chemical analyses from small core samples of pegmatites, it is not surprising that the basement-derived granite and pegmatite show the largest range, includ-ing two samples with over 80 wt% SiO

2. The almost complete

absence of opaque or ferromagnesian minerals—with the exception of rare biotite and garnet—in the basement-derived granite and pegmatite is consistent with the TiO

2-, Fe

2O

3-, and

MgO-poor nature of these samples and distinguishes them from the megablock granites (Fig. 8). The megablock gneissic granite has slightly higher TiO

2, Fe

2O

3, and MgO values than

the biotite granite. It also displays the highest K2O content of

the three granite types. The average CaO content of the lower basement-derived granite is lower, and the Na

2O content is

slightly higher, than that of the megablock granites. The lower basement-derived granite shows Na

2O > K

2O, while both

megablock granites show Na2O < K

2O.

The megablock granite and granite gneiss are peraluminous (Al

2O

3/[CaO + Na

2O + K

2O] molar ratios are 1.02–1.14 and

1.03–1.06, respectively), whereas the lower basement-derived granite is slightly metaluminous to peraluminous in composi-tion (Al

2O

3/[CaO + Na

2O + K

2O] molar ratio is 0.69–1.7). The

wide ranges in both SiO2 content and alumina saturation index

(ASI) in the lower basement-derived granite most likely refl ect the strongly pegmatitic nature of this granite. A special effort was made to obtain ASI values for fi ner-grained granite samples, such as those in the mylonite zone. These indicate a narrower range that is peraluminous in character (Al

2O

3/[CaO + Na

2O + K

2O]

molar ratios are 1.36–1.59).

DISCUSSION

The mineral assemblages observed in the lower basement-derived metasedimentary rocks, as well as geothermobaromet-ric results for the lower basement-derived mica schist, indicate middle- to upper-amphibolite-facies (600–670 °C) peak meta-morphic conditions (Gibson et al., this volume). This is in contrast to the predominantly greenschist-facies crystalline clast popula-tion retrieved from previous drill cores into the Chesapeake Bay impact structure and included in the impactite sequence in the Eyreville B core (see Horton et al., this volume, Chapter 2; Gib-son et al., this volume, and references therein), and these results indicate that the Appalachian basement sampled by the Chesa-peake Bay impact crater was composed of more than one terrane. A more comprehensive discussion of this problem is provided by Gibson et al. (this volume).



Provenance of Metasedimentary Rocks

Hand-specimen, mineralogical, and chemical features of the mica schists in the lower basement-derived section support their derivation from pelitic protoliths. Based on the bulk-rock chemistry, Schmitt et al. (this volume) proposed a shale to Fe-rich shale pro-tolith, where the wide spread in SiO

2 contents likely refl ects a pre-

dominantly mafi c to intermediate source for the original sediments, with a minor felsic component. Using the K

2O/Na

2O versus SiO

2

discrimination diagram for sandstone-mudstone suites (Roser and Korsch, 1986), the mica schists from the lower basement-derived section and megablock xenoliths plot primarily into the fi eld for passive-margin sediments (Fig. 9A), although the latter plot close to the active continental margin fi eld. Provenance discrimination (Fig. 9B) for the same rocks was determined for sandstone-mudstone suites using discriminant function 2 (0.445TiO

2 + 0.07Al

2O

3 –

0.25Fe2O

3(total) – 1.142MgO + 0.438CaO + 1.475Na

2O + 1.426K

2O

– 6.861) versus discriminant function 1 (−1.773TiO2 + 0.607Al

2O

3

+ 0.76Fe2O

3(total) – 1.5 MgO + 0.616CaO + 0.509Na

2O – 1.224K

2O

– 9.09), after Roser and Korsch (1988). According to this scheme, the megablock xenoliths have a quartzose sedimentary (QSP) sig-nature, whereas the lower basement-derived mica schists refl ect either a quartzose sedimentary (QSP) or felsic igneous (FIP) signa-ture. This differs from the fi ndings of Schmitt et al. (this volume), who favor a less siliceous source.

Four outliers are observed in Figure 8A, of which samples RG54 and RG64 plot into the island-arc fi eld, and samples RG70 and CB6–141 plot into the active continental margin fi eld.

Four outliers are also observed in Figure 8B; samples RG38 and RG54 plot into the MIP (mafi c igneous province) fi eld, sam-ple RG64 plots into the IIP (intermediate igneous province) fi eld, and sample CB6–141 plots far away from all other samples into the QSP fi eld. RG38 has a high Fe

2O

3 content compared to the

other mica schist due to its alteration by iron-rich fl uids in the form of pyrite. Sample RG54 is the tourmalinite found in the lower basement-derived mica schists, which is depleted in SiO

2

compared to the mica schist. The mineralogy of the mica schist sample RG64 is enriched in plagioclase, producing an enrich-ment in Na. Both RG70 and CB6–141 are enriched in MgO, thus plotting away from the other mica schists.

Although every effort was made to avoid calcite or quartz veins in samples during bulk-rock analysis, the widespread alter-ation documented in hand specimen and petrographic analysis of the rocks from the lower basement-derived section require cau-tion in inferring protolith compositions from the bulk chemical data (Schmitt et al., this volume).

Comparisons of the metasedimentary rocks in the lower basement-derived section and the biotite schist xenoliths in the megablock are restricted because of the small xenolith size and limited material available. The felsic gneiss found as blocks within the impactite sequence shows a moderate similarity in terms of major elements to the mica schists from the lower basement-derived section (Schmitt et al., this volume). Our study shows, however, that the lower basement-derived section includes

SiO2

log

(K2O

/Na 2O

)

0.1

1

10

Island-arc

Active continental margin

Passive margin

Discriminant function1

Dis

crim

inan

t fun

ctio

n 2

-10-10 -5 0 5 10 15

-8

-6

-4

-2

0

2

4

QSP

FIP

IIP

MIP

35 40 45 50 55 60 65 70

A

B

Figure 9. (A) SiO2 (wt%) versus K

2O/Na

2O (wt%) indicating the

tectonic environment, and (B) provenance signature of sandstone-mudstone suites using major elements (after Roser and Korsch, 1986, 1988). QSP—quartzose sedimentary provenance; FIP—felsic igneous provenance; IIP—intermediate igneous prove-nance; MIP—mafi c igneous provenance. Symbols: closed black triangles—megablock biotite schist; open triangles—basement mica schist.

one highly magnesian sample of quartz-plagioclase-biotite schist (RG70) that bears some mineralogical and compositional resem-blance to the felsic gneiss. The origin of these muscovite-poor schistose to gneissose lithologies warrants further petrographic and chemical analysis. We tentatively suggest, on the basis of the very limited presently available evidence (Gibson et al., this vol-ume; Schmitt et al., this volume; this study), that the felsic gneiss may be derived from the same amphibolite-facies metasedimen-tary sequence as the lower basement-derived schists.

Provenance of Amphibolites

The amphibolites in the megablock and the lower basement-derived section differ considerably both mineralogically and chemically. The lower basement-derived amphibolite is banded

272 Townsend et al.


with strongly varying mineralogy, contains a pargasitic amphi-bole and comparatively more Na-rich plagioclase as well as alkali feldspar, and is closely associated with calc-silicate rock. This amphibolite, consequently, is interpreted to be a metamarl of sedimentary origin (Pettijohn, 1957). The megablock amphib-olite is mineralogically more homogeneous, contains pargasitic to hornblendic amphibole and more calcic plagioclase, and its major- and trace-element chemistry is consistent with a tholeiitic basaltic protolith (Schmitt et al., this volume). If Horton et al.’s (this volume, Chapter 2) and Gibson et al.’s (this volume) obser-vation of relict gabbroic textures is correct, then it seems likely that some or all of this amphibolite represents part of a metamor-phosed gabbroic intrusion.

Both amphibolites contain middle- to upper-amphibolite-facies metamorphic assemblages (Gibson et al., this volume). While this may suggest that they formed part of a single met-amorphic terrane, evidence of biotite schist xenoliths in the gneissic granite of the megablock may indicate similar metamor-phic grades in different parts of the target basement. Modeling of the impact event (Gohn et al., 2008) suggests that the megablocks were derived from further afi eld than the parautochthonous lower basement-derived section, which raises the possibility that they could come from a different tectonostratigraphic terrane.

Granite Petrogenesis

A U-Pb sensitive high-resolution ion microprobe (SHRIMP) study by Horton et al. (this volume, Chapter 14) has shown that the gneissic granite in the megablock is considerably older (615 ± 7 Ma) than the biotite granite (254 ± 3 Ma) and is within error coeval with the Neoproterozoic Langley Granite (Horton et al., 2005b). Ca. 240–250 Ma 40Ar-39Ar mineral ages have been obtained from the basement-derived section and are interpreted as evidence that the amphibolite-facies metamorphism and granite in the lower basement-derived section were coeval with the biotite granite intru-sion (Horton et al., this volume, Chapter 14; M.J. Kunk, 2009, per-sonal commun.). There is little evidence of exactly contemporane-ous granitoid magmatism and metamorphism regionally; Samson et al. (1995) and Coler et al. (1997) documented ages for granitic plutons from 264 Ma to 326 Ma and 285 Ma to 345 Ma, respec-tively, which places the Eyreville B biotite granite and metamor-phism among the youngest events yet recorded for the Alleghanian orogeny and requires slight downward revision of the 270–330 Ma age assigned by Rankin (1994) to this orogeny.

All three of the granite types intersected in the Eyreville B core show slightly metaluminous to peraluminous alkali indices, indicating I-type sources (Chappell and White, 1974; White and Chappell, 1983). The granite samples show a signifi cant over-lap for both the granite megablock and the lower basement-derived granite on Rb versus Nb + Y (Fig. 10A) and Nb versus Y (Fig. 10B) discrimination diagrams (Pearce et al., 1984). In the Rb versus (Y + Nb) plot (Fig. 10A), most samples plot in the within-plate granite fi eld, although close to the syncollisional granite fi eld, and only a few samples fall into the volcanic-arc

granite fi eld. In the Nb versus Y diagram (Fig. 10B), the mega-block granite plots primarily in the volcanic-arc granite + syncol-lisional granite fi elds, whereas the gneissic granite plots primarily in the within-plate granite fi eld close to the ocean-ridge granite fi eld. The lower basement-derived granite samples are more dis-persed, plotting mainly in the within-plate granite fi eld and, to a lesser extent, into the VAG + syn-COLG (syncollisional granite) fi eld. Schmitt et al. (this volume), on the other hand, deduced a syn-COLG setting for the megablock granites from the Rb versus (Yb + Ta) plot. On the basis of Rb versus (Yb + Ta), Ta versus Yb, and rare earth element (REE) contents, they note that the Langley Granite is most closely related to the gneissic granite, with which it shares a similar Neoproterozoic age (Horton et al., this vol-ume, Chapter 14). Given the present scatter of data over several fi elds and subtle shifts between different plots in the Pearce et al.

Y + Nb

10 100 1000

Rb

10

100

1000

syn-COLG

WPG

VAGORG

Y

10 100 1000

Nb

1

10

100

1000

VAG & syn-COLG

ORG

WPG

A

B

Figure 10. (A) Rb vs. (Y + Nb) and (B) Nb vs. Y discriminant dia-grams for granites, indicating different tectonic settings (after Pearce et al., 1984). VAG—volcanic-arc granite; syn-COLG—syncollisional granite; WPG—within-plate granite; ORG—ocean-ridge granite. Symbols: closed black circles—megablock biotite granite; closed gray circles—megablock biotite granite gneiss; open circles— basement granite.



(1984) diagrams, it is as yet not possible to determine the exact setting of granite genesis for the Eyreville B granites within the regional geological evolution, which includes possible continen-tal and volcanic-arc sources and an orogenic event.

Figure 11 shows the granite sample compositions from the Eyreville B core relative to the cotectic curves for granitic melts from Johannes and Holtz (1990). The large dispersion seen in data from the lower basement-derived granite samples may refl ect a combination of the diffi culty of sampling pegmatite and the highly evolved nature of granite pegmatites. Two fi ner-grained granite samples (RG40 and RG61) from the mylonite zone were analyzed in an attempt to evaluate more representative bulk compositions. However, both samples were signifi cantly enriched in quartz relative to the other granite samples, and they plot well away from the cotectic lines. Given the petrographic and chemical evidence obtained from other samples in and near the mylonite zone, this enrichment in silica most likely refl ects hydrothermal fl uids associated with mylonitization. Because of the strong dynamic recrystallization in the mylonites, it cannot be determined how much silica might have been introduced during the retrograde event.

Less scatter is seen in Figure 11 for the megablock granite and gneissic granite samples, which cluster around the 2–5 kbar

cotectic curves. This clustering suggests that both the megablock granite and gneissic granite formed at middle- to shallow-crustal levels (2–5 kbar), in an H

2O-undersaturated system. The samples

from the Raleigh metamorphic belt and Eastern Slate Belt gran-ites display a tighter cluster, but a signifi cant portion of samples lie between the 5 and 10 kbar cotectic lines. The Raleigh meta-morphic belt granites and gneisses include biotite ± hornblende ± muscovite ± garnet calc-alkaline granites, and the Eastern Slate Belt granites include biotite ± hornblende granites, which suggest a different protolith (Coler et al., 1997). Although the defi nitive trend seen in the granites from the Raleigh metamorphic belt and Eastern Slate Belt contrasts with the granites from the Eyreville B and Langley cores (Fig. 11), considerable overlap is still observed.

CONCLUSIONS

Petrographic and chemical analyses of the crystalline basement-derived rocks of the lowermost part of the Eyreville B drill core indicate an association of amphibolite-facies pelitic mica schists with subsidiary muscovite-poor or muscovite-absent biotite schists and metamarls (amphibolite and calc-silicate rock) intruded by pegmatitic granite. Small schist xenoliths within biotite granite and gneissic biotite granite, and an amphibolite derived from a mafi c igneous protolith that occur in the mega-blocks, show a similar peak metamorphic grade to the lower section. Together with similar intrusive and metamorphic ages obtained from U-Pb and Ar-Ar age data, this may or may not indicate that the megablocks were derived from the same terrane as the rocks in the lower section.

Tectonic discrimination diagrams suggest that the schists were formed on a passive continental margin and are primarily of quartz sedimentary provenance, with a felsic to intermediate igneous component; however, signifi cant alteration of the meta-morphic rocks requires some caution in interpreting bulk-rock chemical data. The granite in the lower basement-derived section is more leucocratic than the megablock granite and gneiss, and it contains small amounts of garnet and biotite, with muscovite >> biotite, whereas the megablock contains biotite >> musco-vite. All are peraluminous and I-type, but they differ from I-type biotite granites in the Raleigh belt and Eastern Slate Belt in their lack of hornblende.

ACKNOWLEDGMENTS

The International Continental Scientifi c Drilling Program (ICDP), U.S. Geological Survey (USGS), and National Aero-nautics and Space Administration (NASA) provided the funding for the Chesapeake Bay Deep Drilling Project. Funding for the work by Townsend and Gibson was provided by the National Research Foundation of South Africa (GUN 2074407). Peter Gräser, Kathrin Krahn, Kirsten Born, Graham Beneke, and Zane Roux are thanked for technical support, and David Pow-ars provided the core box photographs. R.G. Luedke, M.J. Kunk, and A.R. Bobyarchick are thanked for their reviews of

Or

Qtz

00 20 40 60 80 100

20

40

60

80

100

Ab

0

20

40

60

80

100

2 kbar

5 kbar

10 kbar

Figure 11. Qtz-Ab-Or ternary diagram with cotectic curves in an H

2O-undersaturated system from Johannes and Holtz (1991). Sam-

ples from granitic plutons in the Raleigh metamorphic belt and East-ern Slate Belt (Coler et al., 1997) have also been included for com-parison. Dashed line represents the water-undersaturated curve for a

H2O = 0.8 crustal fl uids in equilibrium with graphite (Johannes and

Holtz, 1991; Ebadi and Johannes, 1991). Black circles—megablock biotite granite; gray circles—megablock gneissic biotite granite; open circles—lower basement-derived granite; stars—fi ne-grained lower basement-derived granite; closed inverted triangles—granitic samples from the Raleigh metamorphic belt and the Eastern Slate Belt (from Coler et al., 1997); open inverted triangles—granitic sam-ples from the Langley Granite (from Horton et al., 2005b).

274 Townsend et al.


earlier drafts of this manuscript. Any use of trade, product, or fi rm names is for descriptive purposes only and does not imply endorsement by the U.S. government. This is Impact Cratering Research Group contribution 112.

REFERENCES CITED

Blair, T.C., and McPherson, J.G., 1999, Grain-size and textural classifi cation of coarse sedimentary particles: Journal of Sedimentary Research, v. 69, p. 6–19.

Chappell, B.W., and White, A.J.R., 1974, Two contrasting granite types: Pacifi c Geology, v. 8, p. 173–174.

Coler, D.G., Samson, S.D., and Speer, J.A., 1997, Nd and Sr isotopic constraints on the source of Alleghanian granites in the Raleigh metamorphic belt and Eastern Slate Belt, southern Appalachians, U.S.A.: Chemical Petrology, v. 134, p. 257–275.

Deer, W.A., Howie, R.A., and Zussman, J., 1992, An Introduction to the Rock-Forming Minerals (2nd edition): Essex, Longman Scientifi c and Techni-cal, 696 p.

Edwards, L.E., Powars, D.S., Gohn, G.S., and Dypvik, H., 2009, this volume, Chapter 3, Geologic columns for the ICDP-USGS Eyreville A and B cores, Chesapeake Bay impact structure: Sediment-clast breccias, 1096 to 444 m depth, in Gohn, G.S., Koeberl, C., Miller, K.G., and Reimold, W.U., eds., The ICDP-USGS Deep Drilling Project in the Chesapeake Bay Impact Structure: Results from the Eyreville Core Holes: Geological Society of America Special Paper 458, doi: 10.1130/2009.2458(03).

Edwards, L.E., Powars, D.S., Browning, J.V., McLaughlin, P.P., Miller, K.G., Self-Trail, J.M., Kulpecz, A.A., and Elbra, T., 2009, this volume, Chap-ter 4, Geologic columns for the ICDP-USGS Eyreville A and C cores, Chesapeake Bay impact structure: Postimpact sediments, 444 to 0 m depth, in Gohn, G.S., Koeberl, C., Miller, K.G., and Reimold, W.U., eds., The ICDP-USGS Deep Drilling Project in the Chesapeake Bay Impact Structure: Results from the Eyreville Core Holes: Geological Society of America Special Paper 458, doi: 10.1130/2009.2458(04).

Gibson, R.L., Townsend, G.N., Horton, J.W., Jr., and Reimold, W.U., 2007, Pre-impact tectonothermal evolution of the crystalline target in the ICDP-USGS Eyreville-B core, Chesapeake Bay Impact Structure: Geological Society of America Abstracts with Programs, v. 39, no. 6, p. 451.

Gibson, R.L., Townsend, G.N., Horton, J.W., Jr., and Reimold, W.U., 2009, Pre-impact tectonothermal evolution of the crystalline basement-derived rocks in the ICDP-USGS Eyreville B core, Chesapeake Bay impact struc-ture, in Gohn, G.S., Koeberl, C., Miller, K.G., and Reimold, W.U., eds., The ICDP-USGS Deep Drilling Project in the Chesapeake Bay Impact Structure: Results from the Eyreville Core Holes: Geological Society of America Special Paper 458, doi: 10.1130/2009.2458(12).

Glidewell, J., Harris, R.S., King, D.T., Jr., and Petruny, L.W., 2008, Stratigra-phy, petrology, and shock petrography based on well logs and selected samples from the Eyreville drill core: Chesapeake Bay impact struc-ture, Virginia, in 39th Lunar and Planetary Science Conference: http://www.lpi.usra.edu/meetings/lpsc2008/pdf/2438.pdf (accessed November 2008), abstract 2438, 2 p.

Gohn, G.S., Koeberl, C., Miller, K.G., Reimold, W.U., Cockell, C.S., Horton, J.W., Jr., Sanford, W.E., and Voytek, M.A., 2006, Chesapeake Bay impact structure drilled: Eos (Transactions, American Geophysical Union), v. 8735, p. 349–355.

Gohn, G.S., Koeberl, C., Miller, K.G., Reimold, W.U., Browning, J.V., Cock-ell, C.S., Horton, J.W., Jr., Kenkmann, T., Kulpecz, A.A., Powars, D.S., Sanford, W.E., and Voytek, M.A., 2008, Deep drilling into the Chesa-peake Bay impact structure: Science, v. 320, p. 1740–1745, doi: 10.1126/science.1158708.

Gohn, G.S., Powars, D.S., Dypvik, H., and Edwards, L.E., 2009, this volume, Rock-avalanche and ocean-resurge deposits in the late Eocene Chesapeake Bay impact structure: Evidence from the ICDP-USGS Eyreville cores, Virginia, USA, in Gohn, G.S., Koeberl, C., Miller, K.G., and Reimold, W.U., eds., The ICDP-USGS Deep Drilling Project in the Chesapeake Bay Impact Structure: Results from the Eyreville Core Holes: Geological Society of America Special Paper 458, doi: 10.1130/2009.2458(26).

Groat, L.A., Hawthorne, F.C., and Ercit, T.C., 1992, The chemistry of vesuvian-ite: Canadian Mineralogist, v. 30, p. 19–48.

Holland, T.J.B., and Powell, R., 2000, AX: A Program to Calculate Mineral End-Members from Chemical Analyses (Usually from an Electron Micro-probe): Cambridge, Cambridge University Press, 11 p.

Horton, J.W., Jr., Aleinikoff, J.N., Kunk, M.J., Gohn, G.S., Edwards, L.E., Self-Trail, J.M., Powars, D.S., and Izett, G.S., 2005a, Recent research on the Chesapeake Bay impact structure, USA: Impact debris and reworked ejecta, in Kenkmann, T., Hörz, F., and Deutsch, A., eds., Large Mete-orite Impacts III: Geological Society of America Special Paper 384, p. 147–170.

Horton, J.W., Jr., Aleinikoff, J.N., Kunk, M.J., Naeser, C.W., and Naeser, N.D., 2005b, Petrography, structure, age, and thermal history of granitic coastal plain basement in the Chesapeake Bay impact structure, USGS-NASA Langley core, Hampton, Virginia, chap. B, in Horton, J.W., Jr., Pow-ars, D.S., and Gohn, G.S., eds., Studies of the Chesapeake Bay Impact Structure—The USGS-NASA Langley Corehole, Hampton, Virginia, and Related Coreholes and Geophysical Surveys: U.S. Geological Survey Professional Paper 1688, p. B1–B29.

Horton, J.W., Jr., Powars, D.S., and Gohn, G.S., 2005c, Studies of the Chesa-peake Bay impact structure—Introduction and discussion, chap. A, in Horton, J.W., Jr., Powars, D.S., and Gohn, G.S., eds., Studies of the Ches-apeake Bay Impact Structure—The USGS-NASA Langley Corehole, Hampton, Virginia, and Related Coreholes and Geophysical Surveys: U.S. Geological Survey Professional Paper 1688, p. A1–A24.

Horton, J.W., Jr., Gohn, G.S., Gibson, R.L., Reimold, W.U., and Edwards, L.E., 2007, Geologic column of the ICDP-USGS Eyreville-B core Chesapeake Bay impact structure: Breccias, blocks, and crystalline rocks, 1095–1766 m depth: Geological Society of America Abstracts with Programs, v. 39, no. 6, p. 314.

Horton, J.W., Jr., Gohn, G.S., Powars, D.S., and Edwards, L.E., 2008, Origin and emplacement of impactites in the Chesapeake Bay impact structure, Virginia, USA, in Evans, K., Horton, J.W., Jr., King, D.T., Jr., and Mor-row, J.R., eds., The Sedimentary Record of Meteorite Impacts: Geological Society of America Special Paper 437, p. 73–97.

Horton, J.W., Jr., Gibson, R.L., Reimold, W.U., Wittmann, A., Gohn, G.S., and Edwards, L.E., 2009, this volume, Chapter 2, Geologic columns for the ICDP-USGS Eyreville B core, Chesapeake Bay impact structure: Impac-tites and crystalline rocks, 1766 to 1096 m, in Gohn, G.S., Koeberl, C., Miller, K.G., and Reimold, W.U., eds., The ICDP-USGS Deep Drilling Project in the Chesapeake Bay Impact Structure: Results from the Eyre-ville Core Holes: Geological Society of America Special Paper 458, doi: 10.1130/2009.2458(02).

Horton, J.W., Jr., Kunk, M.J., Belkin, H.E., Aleinikoff, J.N., Jackson, J.C., and Chou, I-M., 2009, Chapter 14, Evolution of crystalline target rocks and impactites in the Chesapeake Bay impact structure, ICDP-USGS Eyre-ville B core, in Gohn, G.S., Koeberl, C., Miller, K.G., and Reimold, W.U., eds., The ICDP-USGS Deep Drilling Project in the Chesapeake Bay Impact Structure: Results from the Eyreville Core Holes: Geological Society of America Special Paper 458, doi: 10.1130/2009.2458(14).

Johannes, W., and Holtz, F., 1990, Formation and composition of H2O-

undersaturated granitic melts, in Ashworth, J.R., and Brown, M., eds., High-Temperature Metamorphism and Crustal Anatexis: The Mineralogi-cal Society Series, v. 2, p. 87–104.

Leake, B.E., 1978, Nomenclature of amphiboles: Mineralogical Magazine, v. 42, p. 533–563, doi: 10.1180/minmag.1978.042.324.21.

Pearce, J.A., Harris, N.B.W., and Tindle, A.G., 1984, Trace element discrimina-tion diagrams for the tectonic interpretation of granitic rocks: Journal of Petrology, v. 25, p. 956–983.

Pettijohn, F.J., 1957, Sedimentary Rocks (2nd edition): New York, Harper,718 p.

Poag, C.W., Koeberl, C., and Reimold, W.U., 2004, The Chesapeake Bay Impact Crater—Geology and Geophysics of a Late Eocene Submarine Impact Structure: New York, Springer-Verlag, 522 p. plus CD-ROM.

Rankin, D.W., 1994, Continental margin of the eastern United States; past and present, in Speed, R.C., ed., Phanerozoic Evolution of North American Continent-Ocean Transitions: Boulder, Colorado, Geological Society of America, Decade of North American Geology, Continent-Ocean Transect Volume, p. 129–218.

Reimold, W.U., Kenkmann, T., Gibson, R.L., Bartosova, K., Schmitt, R.-T., Hecht, L., Koeberl, C., and Horton, J.W., Jr., 2007, Dike breccias in the basement-derived section of ICDP-USGS drill core Eyreville B, Chesa-peake Bay impact structure: Geological Society of America Abstracts with Programs, v. 39, no. 6, p. 451.



Roser, B.P., and Korsch, R.J., 1986, Determination of tectonic setting of sandstone–mudstone suites using SiO

2 content and K

2O/Na

2O ratio: The

Journal of Geology, v. 94, p. 635–650.Roser, B.P., and Korsch, R.J., 1988, Provenance signatures of sandstone– mudstone

suites determined using discriminant function analysis of major-element data: Chemical Geology, v. 67, p. 119–139, doi: 10.1016/0009-2541(88)90010-1.

Samson, S.D., Coler, D.G., and Speer, J.A., 1995, Geochemical and Nd-Sr-Pb isotopic composition of Alleghanian granites of the southern Appa-lachians: Origin, tectonic setting, and source characterization: Earth and Planetary Science Letters, v. 134, p. 359–376.

Schmitt, R.T., Wittmann, A., and Stöffl er, D., 2004, Geochemistry of drill core samples from Yaxcopoil-1, Chicxulub impact crater, Mexico: Meteoritics & Planetary Science, v. 39, no. 6, p. 979–1001.

Schmitt, R.T., Bartosova, K., Reimold, W.U., Mader, D., Wittmann, A., Koe-berl, C., and Gibson, R.L., 2009, this volume, Geochemistry of impac-tites and crystalline basement-derived lithologies from the ICDP-USGS Eyreville A and B drill cores, Chesapeake Bay impact structure, Virginia/

USA, in Gohn, G.S., Koeberl, C., Miller, K.G., and Reimold, W.U., eds., The ICDP-USGS Deep Drilling Project in the Chesapeake Bay Impact Structure: Results from the Eyreville Core Holes: Geological Society of America Special Paper 458, doi: 10.1130/2009.2458(22).

Siivola, J., and Schmid, R., 2007, List of mineral abbreviations, in Fettes, D., and Desmons, J., eds., Metamorphic Rocks; A glossary of Terms, Recom-mendations of the International Union of Geological Sciences Subcom-mission on the Systematics of Metamorphic Rocks: Cape Town, Cam-bridge University Press, p. 93–110 (also online at http://www.bgs.ac.uk/scmr/docs/papers/paper_12.pdf).

Taylor, S.R., and McLennan, S., 1985, The Continental Crust: Its Composition and Evolution: Oxford, Blackwell, 312 p.

White, A.J.R., and Chappell, B.W., 1983, Granitoid types and their distribu-tion in the Lachlan fold belt, southeastern Australia, in Roddick, J.A., ed., Circum-Pacifi c Plutonic Terranes: Geological Society of America Mem-oir 159, p. 21–34.

MANUSCRIPT ACCEPTED BY THE SOCIETY 3 MARCH 2009

Printed in the USA

Documents

Petrographic and geochemical comparisons between the lower