ZO/(‘l

 

 

 

 

..L.I.BRARY

U

Univeisity

 

This is to certify that the
thesis entitled

OLIVINE WEATHERING TEXTURES IN SERPENTINIZED
DUNITE, WEBSTER-ADDIE ULTRAMAFIC BODY, NORTH
CAROLINA

presented by

DANIEL R. SNYDER

has been accepted towards fulﬁllment
of the requirements for the

Master of degree in Geological Sciences
Science

 

 

MW Wﬂ/

Major Professor’s Signature

?' ﬂow? 9-0/0

Date

 

MSU is an Aﬁ‘in'native Action/Equal Opportunity Employer

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProj/Acc&Pres/ClRC/DateDue.indd

 

OLIVINE WEATHERINC
WEBSTEl

in pa

 

 

OLIVINE WEATHERING TEXTURES IN SERPENTINIZED DUNITE,
WEBSTER-ADDIE ULTRAMAFIC BODY,
NORTH CAROLINA

By

Daniel R. Snyder

A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
MASTER OF SCIENCE
Geological Sciences

2010

ouerEWEATHERlNG
WtaerR-ADDtE U

Many experimental stt
cared out since the middle <
iere no studies of natural w.
lerend on highly corrosive s
Fares. and use gem-qualrt‘
ﬂattering however, must

ﬁltrate and drawing infer

Thin sections and g

nc-ieultramaﬁc body in Vl

ABSTRACT

OLIVINE WEATHERING TEXTURES IN SERPENTINIZED DUNITE,
WEBSTER-ADDIE ULTRAMAFIC BODY, NORTH CAROLINA

By
Daniel R. Snyder

Many experimental studies of silicate mineral weathering have been
carried out since the middle of the twentieth century, but until the past year there
were no studies of natural weathering textures of olivine. Laboratory methods
depend on highly corrosive solvents to achieve results within reasonable time
frames, and use gem-quality olivine to isolate measurable effects. Natural
weathering, however, must be analyzed by studying the morphology of end-state

materials and drawing inferences from what is observed.

Thin sections and grain mounts of tectonized olivines from the Webster-
Addie ultramaﬁc body in western North Carolina were studied using optical
microscopy and scanning electron microscopy. Most natural dissolution features,

or etch pits, can be seen as variations of conic sections.

Etch pit ﬁelds observed in nature share a number of general
characteristics with such features formed experimentally, including uniform size,
spacing, and orientation. These observations support the conclusions of many
studies of experimental kinetics that olivine dissolution is surface-reaction limited,
and that dissolution textures result from preferential etching at edge and screw

dislocations within the olivine.

COPYRIGHT BY
DANIEL R. SNYDER
2010

This thesis is

Dedication
This thesis is dedicated in commemoration of my late sister,
Shirley Gumpper,

artist and teacher,

1929-2010

lam most appreci
he by my major profess
other members of my c:
comments, suggestions
substantially. In additio

me to sit in on their co

Invaluable tecl‘

assisted with the scar

out the X-ray fluoresr

Funding for ti
orovided by NASA I
itiichael A. Velbel.

Acknowledgements

I am most appreciative of the generosity and patient guidance provided
me by my major professor, Dr. Michael A. Velbel. I would also like to thank the
other members of my committee, Dr. Kazuya Fujita and Dr. Duncan Sibley. Their
comments, suggestions, and editing of this thesis improved the ﬁnal copy
substantially. In addition, I would like to thank these three professors for inviting

me to sit in on their courses to refresh my knowledge in their specialty areas.

Invaluable technical support was provided by Ewa Danielwicz, who
assisted with the scanning electron microscopy, and by Anna Losiak, who carried

out the X-ray ﬂuorescence analysis.

Funding for the electron microscopy and X-ray ﬂuorescence analysis was
provided by NASA Mars Fundamental Research Program Grant NNGO5GL77G

(Michael A. Velbel, Principal Investigator).

LIST OF TABLES. .. . . ..
LIST OF FIGURES .....
KEY TO SYMBOLS AI

INTRODUCTION ......

PART I.
REVIEW OF LITERA'

Chapter 1. Pro
Chapter 2. Na
Chapter 3. Te
Chal3ler4. Ex

hapter 5. Im

PART ll.
THE STUDY AREA

Chapler 6_ G
Chapter 7. Pi

PART III.
MATERIALS AND '

Chapter 8. T
Chapter 9_ A

p‘er IV.
RESULTS

Cha

New

DLSr 1 1 ‘
Featt
Span
00”"
Form

TABLE OF CONTENTS

LIST OF TABLES ................................................................................... viii
LIST OF FIGURES .................................................................................. ix
KEY TO SYMBOLS AND ABBREVIATIONS ................................................. xv
INTRODUCTION ...................................................................................... 1
PART I.
REVIEW OF LITERATURE
Chapter 1. Products of Natural Weathering .......................................... 6
Chapter 2. Natural Textures ............................................................ 14
Chapter 3. Textures Formed Experimentally ....................................... 20
Chapter 4. Experimental Kinetics ..................................................... 27
Chapter 5. Implications for a Rate-Determining Mechanism ................... 35
PART II.
THE STUDY AREA
Chapter 6. Geologic and Geographic Setting ..................................... 37
Chapter 7. Previous Studies of the Webster-Addie Ultramaﬁc Body ...... 59
PART III.
MATERIALS AND METHODS
Chapter 8. The Sample Suite .......................................................... 63
Chapter 9. Analytical Methods ........................................................ 65
PART IV.
RESULTS
Chapter 10. Results of X-Ray Fluorescence ...................................... 70
Chapter 11. Results of Microscopy
Features Visible at SEM and Optical Scales ............................. 76
Spatial Distribution of Etch Features ..................................... 109
Comparison of Observed Natural Features with Textures
Formed Experimentally .............................................................. 137
vi

PART V. DISCUSSION ..
PART VI. CONCLUSION

APPENDIX................._

REFERENCES CITED.

PART V. DISCUSSION ........................................................................ 155

PART VI. CONCLUSION ...................................................................... 169
APPENDIX ....................................................................................... 171
REFERENCES CITED .......................................................................... 175

vii

Table 1. (Appendix) M
Analyzed..................

LIST OF TABLES

Table 1. (Appendix) Major and Minor Element Composition of Samples
Analyzed .............................................................................................. 1 71

viii

Figure t. Schematic n
America .....................

Figure 2. Ultramafic t
Figure 3. Location of

boundaries, nearby Cl
Park

Figure 4. Highly genei
area

Figure 5. Ob

quue aeria
south........
Figure 6. Outcrop of dl
Flgure 7 eathered
Addie, N rth Carolin
Figure TB. per
IOUgh polish .........
fIiIure SEM image
exture of olivi

LIST OF FIGURES

Figure 1. Schematic map of ultramaﬁc bodies in eastern North
America ................................................................................................................ 38

Figure 2. Ultramaﬁc bodies in the North Carolina Blue Ridge ............................. 39
Figure 3. Location of the Webster-Addie ultramaﬁc body in relation to state
boundaries, nearby cities and towns, and Great Smoky Mountains National
Park ...................................................................................................................... 40

Figure 4. Highly generalized geologic map of the Webster-Addie-Sylva
area ...................................................................................................................... 43

Figure 5. Oblique aerial image of Webster-Addie-Sylva area, viewed from

south .................................................................................................................... 44
Figure 6. Outcrop of dunite near Webster, North Carolina, showing laminated
fabric .................................................................................................................... 46
Figure 7A. Weathered dunite specimen collected at abandoned olivine quarry,
Addie, North Carolina ............................................................................... 47
Figure 78. Same specimen as in Figure 7A, above, cut with diamond saw and
rough polished ........................................................................................ 47
Figure 8. SEM image of fractured surface of a grain of dunite showing granular
texture of olivine and uniform grain size .............................................................. 50
Figure 9. Foliation in dunite, Addie, North Carolina ............................................. 51

Figure 10. SEM image of part of a polished thin section showing olivine grains
(light gray) separated by serpentine layers (darker gray). which also ﬁll many

intra-mineral fractures .............................................................................. 52
Figure 11A. Weathered dunite rind (top) on serpentine (bottom) ........................ 54
Figure 113. Same specimen as above, tilted to show surface of cut .................. 54
Figure 12. “Soapstone” coating on dunite ............................................................ 55

Figure 13. Dunite compo
image) and sheared talc
birdoiimage)................_

Figure 14. Section of Sylr
lor samples used in this 1

Figure 15. Change in We
Weathering....................

Figure 16. Change in M
Weathering ...................

Figure 17. A well-devel
Figure 18. A ﬁeld of elc
Figure 19. A field of sir

Figtlre 20 Hi
, - gh—ma it
Its. tapering to g

F
d

harms

Nure 21. Linear etci

lemond‘ ShaIDEd pits

ggufe 22A Linear ET
P359~I<>base Cone

surface of the thin se

I:Igur
22A..E..2.25_SC“9"‘3

Figure Cross
pr’IS (indI ' ‘S'
‘Vlne ted by an
time
Ditsiyndl grosses
D1”lite by at1
our

With; U and 23L
F pI’Tatmiq
, Lillie 2

Riteeﬁi 3E 3F

Figure 13. Dunite composed of polygonal olivine grains (upper two-thirds of
image) and sheared talc clots with crushed and elongated olivine grains (lower
third of image) ...................................................................................................... 56

Figure 14. Section of Sylva North USGS 7.5’ quadrangle showing collection sites
for samples used in this study .............................................................................. 64

Figure 15. Change in Weight Percent MgO by Relative Degree of

Weathering ........................................................................................................... 72
Figure 16. Change in Magnesium Number by Relative Degree of

Weathering ........................................................................................................... 75
Figure 17. A well-developed ﬁeld of diamond-shaped etch pits ........................... 77
Figure 18. A ﬁeld of elongated diamond-shaped etch pits ................................... 78
Figure 19. A ﬁeld of sinuous etch pits .................................................................. 79

Figure 20. High-magniﬁcation SEM image of elongated diamond-shaped etch
pits, tapering to narrow points and aligned in parallel .......................................... 80

Figure 21. Linear etch pits with serrated edges, apparently formed by coalescing
diamond- shaped pits ........................................................................................... 81

Figure 22A. Linear etch pits with serrated edges, composed of coalescing pairs
of base-to-base cones exposed as coalescing diamond-shaped outlines at the
surface of the thin section .................................................................................... 82

Figure 223. Schematic diagrams of linear etch pits C, D, and E in Figure
22A ....................................................................................................................... 82

Figure 23A. Cross-section of a serpentine layer (S) with pyramidal or conical etch
pits (indicated by arrows) protruding from the margin of the serpentine layer into
olivine ................................................................................................................... 83

Figure 23B. Cross-section of a serpentine layer with pyramidal or conical etch
pits (indicated by arrows) protruding from the margin of the serpentine layer into
olivine ................................................................................................................... 83

Figure 230 and 23D. Illustration of possible formation of diamond-shaped etch pit
outlines by pyramidal etching (A) or by conical etching (B) ................................. 85

Figure 23E, 23F, and 236. Diamond-shaped etch pits with apparent bi-conical
three-dimensional form ........................................................................................ 85

 

Figure 24. A field of d
etch pits in an iron-oi

Figure 25. Conical eI
surface.......

Figure 26. Nested c

Figure 27. Conical I

..................

Figure 28. A field c

FlEIUre 29. A conic
olivine _

PIS...

Figure 31 Com
Conical etch pus

FIgUle 3'2 A TIEI
rOws

Figure 33. pa,

FIgUre 34A

Arra
Of an OII‘JIne

QT
FIguTe 3s
ﬁaﬂened Q0

Figure 24. A ﬁeld of diamond-shaped etch pits (right) grading into rectangular
etch pits in an iron-oxide-rich zone (left) .............................................................. 86

Figure 25. Conical etch pits creating elliptical outlines at an exposed grain

surface ................................................................................................................. 88
Figure 26. Nested conical etch pits ...................................................................... 89
Figure 27. Conical etch pits developed obliquely to the exposed

surface ................................................................................................................. 90
Figure 28. A ﬁeld of shallow-angle conical etch pits ............................................ 91

Figure 29. A conical etch pit protruding from a serpentine layer into
olivine ................................................................................................................... 92

Figure 30. An etch pit ﬁeld containing many “mouth-shaped” etch
pits ....................................................................................................................... 93

Figure 31. Compound “mouth-shaped” etch pits showing formation by coalescing
conical etch pits ................................................................................................... 94

Figure 32. A ﬁeld of shallow-angle conical etch pits aligned in parallel
rows ..................................................................................................................... 95

Figure 33. Parallel rows of very-shallow-angle ﬂuted etch pits ............................ 96

Figure 34. Arrays of shallow-angle ﬂuted etch pits on the cleanly-broken surface
of an olivine grain ................................................................................................. 97

Figure 35. A small ﬁeld of triangular etch pits, most of which appear to be
ﬂattened conic sections ........................................................................................ 98

Figure 36. A ﬁeld of sharply-etched triangular etch pits ....................................... 99

Figure 37. A complex shape composed of intersecting conical etch
pits ..................................................................................................................... 1 00

Figure 38. A series of aligned conical etch pits, framed by remnants of the
enclosing serpentine layer ................................................................................. 101

Figure 39. Straight, tubular or prismatic features protruding into olivine from the
margin of a serpentine layer .............................................................................. 102

xi

Figure 40. Linear etch pit
needles of a replacement
diierent mineral ..............

 

Figure 41. A low-relief rei
zone

Figure 42. A ﬁeld of low-
iron-oxiderich zone .......

Figure 43. Low-relief etc
between two serpentine

Figure 44. “Filigree” par

subsurface texture (up;

Figure 45. Low-magnifi
SylvadU, showing distr
WARE)...”

~......,‘

Figure 40. Linear etch pits with secondary formations of globular pits, or possibly
needles of a replacement mineral with globular accretions of the same or a
different mineral ................................................................................................. 103

Figure 41. A low-relief rectilinear etch pit pattern in an iron-oxide-rich
zone ................................................................................................................... 1 05

Figure 42. A ﬁeld of low-relief diamond-shaped and rectangular etch pits in an
iron-oxide-rich zone ........................................................................................... 106

Figure 43. Low-relief etch pit patterns, with corrugated subsurface structures,
between two serpentine-ﬁlled fractures in an iron-oxide rich zone .................... 107

Figure 44. “Filigree” pattern (lower right) and low-relief etch pits with corrugated
subsurface texture (upper center) in a heavily iron-oxide stained
zone ................................................................................................................... 1 08

Figure 45. Low-magniﬁcation SEM image of a small portion of thin section
Sylva6U, showing distribution of prominent etch pit clusters (outlined in
white) ................................................................................................................. 1 10

Figure 46. Scan of thin section of sample Sylva5C with dot symbols showing etch
pit clusters .......................................................................................................... 111

Figure 47. A ﬁeld of diamond-shaped etch pits at an intersection of two
serpentine-ﬁlled fractures ................................................................................... 1 12

Figure 48. Etch pits forming under a layer of serpentine and etch pits on the
surface of an olivine grain from which the surface was cleanly broken away
during sample preparation (upper three-quarters of image) .............................. 113
Figure 49. Etch pits developed along a serpentine layer (S) ............................. 114

Figure 50. Etch pits developed along a thick serpentine layer (top) and a narrow
fracture (lower center) with only a thin ﬁlling, possibly an alteration

product ............................................................................................................... 1 15
Figure 51. Three ﬁelds of scattered etch pits formed along unﬁlled fractures

(indicated by arrows) .......................................................................................... 117
Figure 52. Etch pits (dark spots) developed along curving fractures ................. 118
Figure 53. Evenly spaced, parallel etch pits at micron scale ............................. 119

xii

Figure 54. A ﬁeld of unrfo
removal of an enclosing 1
preparation .....................

Figure 55. Unifonn-size 1

Figure 56. A ﬁeld of etct
(top center) to smaller, E

Figure 57. A ﬁeld of inle
Figure 58. A large field

Figure 59. An etch pit f
Proportions, and degre
as a component throug

Figure 60. Linear arrai

oIrrirle...........

:Nure 61. Line

ar arra
aCtures.......

"Ouo.

F'
14:? 62. Parallel lin

a material.

Flsure 63 A”

Tamra... eys 0f '

u
‘0
~
.
I
O.
o
u
h
'-
n

Figure 54. A ﬁeld of uniform-size, evenly spaced, parallel etch pits exposed by
removal of an enclosing serpentine layer (remnant at left) in thin section
preparation... ..................................................................................................... 120
Figure 55. Uniform-size shallow-angle ﬂuted etch pits ...................................... 121

Figure 56. A ﬁeld of etch pits grading from larger, compact diamond-shaped pits

(top center) to smaller, elongated pits ................................................................ 122
Figure 57. A ﬁeld of intermixed large and small etch pits .................................. 123
Figure 58. A large ﬁeld of roughly triangular etch pits ........................................ 124

Figure 59. An etch pit ﬁeld enclosed by fractures, with pits varying in size,
proportions, and degree of coalescence, but retaining the basic diamond shape
as a component throughout the ﬁeld .................................................................. 125

Figure 60. Linear arrays of diamond-shaped etch pits viewed through
olivine ................................................................................................................. 127

Figure 61. Linear arrays of diamond-shaped etch pits along parallel
fractures ............................................................................................................. 128

Figure 62. Parallel linear arrays of diamond-shaped etch pits, apparently ﬁlled
with a dark material ............................................................................................ 129

Figure 63. Arrays of evenly-spaced diamond-shaped etch pits following curved
fractures ............................................................................................................. 130

Figure 64. Diamond-shaped etch pits lying along a curved fracture, intersected
by arrays on three smaller, parallel, straight fractures ....................................... 131

Figure 65. Straight, parallel features, possibly tubular or prismatic etch pits, or
possibly a replacement mineral, extending from serpentine layers into
olivine ................................................................................................................. 1 32

Figure 66. A large ﬁeld of straight, parallel etch features, or possibly needles of a
replacement mineral, protruding into olivine ...................................................... 133

Figure 67. Fairly advanced dissolution of olivine (left) ....................................... 134
Figure 68. Partially corroded etch pit ﬁeld (right) and apparent site of former etch
ﬁeld (center) removed by corrosion (note channel through

center) ................................................................................................................ 1 35

Figure 69. A well-deﬁned ﬁeld of triangular etch pits ......................................... 138

xiii

 
 
    

Figure ill. Experimentally
airis

Figureil. SEM image ott
Figure 72. Natural etch pit.

Figure 73. SEM image of I
oarallelto c-axis...............

Figure 74. Apparent conii
corners.................

Figure 75. Arrows indica

Figure 76. Wedge-shapi
layer into olivine ............

Figure 77. Cross-sectio
protruding into olivine ii

Figure 78. Polished an
pgs wrth pointed ends
Figure 79. Etc
Debdotite).....

h on her

Hsure a
. 3.
DINING...“ Rectangul

Figure 70. Experimentally etched surface of an olivine crystal parallel to the b-

axis ..................................................................................................................... 1 39
Figure 71. SEM image of forsterite etched by acid (pH2) .................................. 141
Figure 72. Natural etch pits in the form of truncated triangles ........................... 142

Figure 73. SEM image of forsterite etched with acid (pH1 or pH2), surface
parallel to c-axis ................................................................................................. 143

Figure 74. Apparent conical etch pits evolved into rectangular pits with rounded
corners ............................................................................................................... 144

Figure 75. Arrows indicate conical etch pits ....................................................... 145

Figure 76. Wedge-shaped etch pits protruding from the margin of a serpentine
layer into olivine ................................................................................................. 146

Figure 77. Cross-section of a serpentine layer with conical or pyramidal etch pits
protruding into olivine from both boundary surfaces .......................................... 147

Figure 78. Polished and etched surface of gem—quality olivine, showing oval etch
pits with pointed ends (a) and elongated pits with diamond-shaped centers
(b) ....................................................................................................................... 149

Figure 79. Etch pit ﬁelds in experimentally strained gem quality olivine (Red Sea
peridotite) ........................................................................................................... 150

Figure 80A and 808. Similar etch pits on experimental and natural
surfaces ............................................................................................................. 151

Figure 81A and 81 B. Similar etch pits on experimental and natural
surfaces ............................................................................................................. 1 52

Figure 82. Etching sequence of olivine from Horn and Muerette’s historic etching
experiment ......................................................................................................... 1 54

Figure 83. Rectangular etch pits formed on the (010) surface of stressed
olivine ................................................................................................................. 154

xiv

Purpose
The purpose I

(observed at optical
scales) formed duri
compositionally uri
following hypothes
conform to mecha

According

dissolution kinetii

CWSIaIIOQI'aphiQ;

surface layers, c

IBerner, 1978; \

mQTDI'IOIOQ 1031 f
some experime

AIlerriativeiyg a

Introduction

Purpose

The purpose of this thesis is to characterize weathering textures
(observed at optical petrographic and scanning electron microscope [SEM]
scales) formed during natural weathering of olivine in a large dunite body with
compositionally uniform olivine, and to interpret this information to test the
following hypothesis: Morphological features of naturally weathered olivine
conform to mechanisms inferred from experimentally determined kinetics.

According to the prevailing interpretation of experimental olivine
dissolution kinetics, textures formed by olivine weathering will be dominated by
crystallographically controlled etch pits, because, in the absence of protective
surface layers, dissolution kinetics of silicate minerals are interface limited
(Bemer, 1978; Velbel, 2009). Crystallographic control may result in different
morphological features aligned with the respective crystal axes, as was found in
some experimental studies (Wegner and Christie, 1976; Awad et al., 2000).
Alternatively, all etch features might be manifestations of the same cross-
sectional shape, and apparent variations in the micromorphology of etch pits and
channels may be the result of the coalescing of pits and of differences in the
angles of intersection of planes in olivine grains with the thin section plane.

The hypothesis presupposes that etch pits and channels in olivine are the
result of post-metasomatic weathering, and not a metasomatic phenomenon
(Velbel, 2009). If this is the case, etching features should be more intensively

developed in samples taken in or near the weathered rind of olivine rock than in

those from relatively ire
Rolls. 1998). A further c
density of etch pits. shr

veins and layers which

Background

Other specific i

many researchers. Tl

weathering prod ucts

dissolution, but, unti
natural olivine dissol

Ills study is to deter

those from relatively fresh parts of the rock (Delvigne et al.,1979; Bland and
Rolls, 1998). A further corollary is that dissolution of olivine, as evidenced by the
density of etch pits, should take place more readily along the edges of serpentine

veins and layers which act as paths for the movement of water (Velbel, 2009).

Background

Other speciﬁc major silicate mineral groups have been addressed by
many researchers. There have also been a number of studies on alteration and
weathering products of olivine, and several experimental studies of olivine
dissolution, but , until very recently, no systematic morphological test of the
natural olivine dissolution mechanism had been done. The primary objective of
this study is to determine the extent to which the rate-determining mechanism
inferred from morphological features of naturally weathered olivine (etch pits,
channels, wedges, etc.) conforms to the mechanism inferred from experimentally
determined kinetics.

Overviews of observations and implications of alteration textures for
reaction mechanisms have been published for rock-forrning silicates as a whole
(Velbel, 1993a; Vlﬁlson, 2004). Berner (1978) described the two main types of
mineral dissolution rate-controlling mechanisms as transport and surface
reaction. Transport mechanisms include advection of reactants in solution,
diffusion of reactants through a stagnant liquid, and diffusion through a solid -
possibly a protective layer of precipitate. Berner (1978) also reviewed several
experimental approaches to determining which type obtained for a given mineral

under given conditions. Of these, the simplest method is "microscopic

 

 

examination of the SUFI
( 1978) explained that "
smooth surfaces... be
rapid that selective eti
reactioncontrolled di:
dissolution of the cry:
and shows crystallog
Pits..." (Berner, 197%

In presenting
kinetics, Velbel (20C
limited kinetics from
eicierirnental condii
exFti‘eriments. Only I

I!

2U“), Could read”,

1"? 9X1

3 .

9“ a SITTIlIar
dISIInct'

examination of the surface morphology of partially dissolved crystals." Berner
(1978) explained that "crystals dissolved by transport control should exhibit
smooth surfaces... because ion detachment over the entire crystal surface is so
rapid that selective etching does not occur." Conversely, "During surface-
reaction-controlled dissolution ion detachment is so slow that selective
dissolution of the crystal surface occurs," and "a surface results that is angular
and shows crystallographically controlled and usually geometrically regular etch
pits. . . " (Berner, 1978).

In presenting a set of laboratory and homework exercises on geochemical
kinetics, Velbel (2004) listed "four tests that can be used to distinguish transport-
lim'rted kinetics from interface limited kinetics... in any individual set of
experimental conditions." Of these, three required measurement in controlled
experiments. Only one, "the surface morphology of the reacting crystal" (Velbel,
2004), could readily be applied to observation of naturally weathered minerals.

In two studies using SEM and X-ray photoelectron spectroscopy (XPS)
Berner and Holdren (1977, 1979) concluded that "surfaces of feldspars which
have undergone appreciable dissolution... are rough, pitted and angular." and
that "Surface irregularity comes about because dissolution occurs selectively at
surﬁcial sites of excess energy, such as dislocations" (Berner and Holdren,
1979). In a similar study of pyroxenes and amphiboles, Berner et al. (1980)
"found a distinctive type of etch pitting, analogous to that in feldspars, which
appears to be characteristic of chain-type silicates that have undergone

dissolution under earth surface conditions."

Addressing '
dissolution, Berner
of naturally weathe
from soils. They fc
appreciably thick y
pyroxene and am
crystallographica'
as well as that of

Velbel (1g
Protective Surtac
Olivine. How EVe
Possible Can di d
Ve'bel (:

Addressing the issue of a rate-determining mechanism in chain silicate
dissolution, Berner and Schott (1982) carried out a study (using XPS and SEM)
of naturally weathered and acid-etched pyroxene and amphibole grains taken
from soils. They found that there was "no evidence for the existence of an
appreciably thick protective layer of altered composition on the surface of
pyroxene and amphiboles undergoing weathering", and that "formation of
crystallographically controlled etch pits characterizes the weathering of feldspars
as well as that of pyroxenes and amphiboles".

Velbel (1993) calculated molar-volume relationships indicating that
protective surface layers, at least for some common products, cannot form from
olivine. However, since the calculated scenarios did not include smectite, a
possible candidate mineral in this context, the results were not all-inclusive.

Velbel (2007) reviewed the extensive literature on weathering of rock-
forrning minerals in general, and pyroxenes and amphiboles in particular, with
emphasis on the kinetics of dissolution as they apply to the natural weathering of
heavy minerals in the sedimentary cycle, as well as the resulting surface
textures. The review included a discussion of "relevant insights from studies of
other minerals". Using the term "pyriboles" to mean "pyroxenes and amphiboles",
Velbel noted that "experimental dissolution of pyriboles at temperature and pH
conditions similar to the conditions of natural weathering does not create familiar
etch features", but that "numerous experiments using strong acids at higher-than-
ambient temperatures or extreme pHs produce etch pits resembling those widely

observed on naturally weathered pyriboles." (Velbel, 2007).

Studies were al
garnets - Velbel, 1984
studies of clay replace
etal.,1972, Smith et
SEM studies of natur

Two recent w
the same question a
examine dissolutior

Ridge in North Carc

Studies were also made of some less-common rock-forrning silicates (e.g.,
garnets - Velbel, 1984, Velbel et al., 2007; staurolite - Velbel et al., 1996), and
studies of clay replacement of olivine are common (Delvigne et al., 1979, Nahon
et al., 1972, Smith et al., 1987, Banﬂeld et al., 1990), but similar petrographic and
SEM studies of naturally weathered olivine were lacking until very recently.

Two recent works by Velbel and Ranck (2008) and Velbel (2009) address
the same question as the present paper. Velbel and Ranck (2008) used SEM to
examine dissolution features in two tectonized dunites from the southern Blue
Ridge in North Carolina, and concluded that the samples examined were
corroded by conical etch pits. Velbel (2009) extended the study to include olivine
phenocrysts from Hawaiian basalts as well as additional tectonized dunites,
including one from the Webster-Addie ultramaﬁc body, and conﬁrmed the
ﬁndings of Velbel and Ranck (2008).

The present study examines in detail the olivine rocks of the Webster-
Addie body, and presents examples from optical microscopy as well as additional
examples from SEM of thin sections and grain mounts.

In the following chapters, previous studies of olivine alteration and
weathering are reviewed according to the following sequence: products of natural
weathering, textures formed by natural weathering, textures formed
experimentally, experimentally-determined kinetics, and implications for a rate-

detennining mechanism.

 

Part I. Revie

That close
appeals I0 have
the mid- 1 9") ce

hwy TaIIIng a

Part I. Review of Literature on Olivine Alteration and Weathering

Chapter 1.

Products of Natural Weathering

That observation that olivine is one of the most easily altered of minerals
appears to have been generally accepted by geologists at least as long ago as
the mid-19th century. Dana (1849), in his report on the American exploration
expeditions of 1838 through 1842, stated that in basalt on the island of Tahiti,
olivine was the ﬁrst mineral to decompose, “becoming at ﬁrst iridescent and

ﬁnally falling away to a soft, pulverulent, ochreous yellow or brown powder”.

At the onset of the thirty years (1870-1900) that have been called the
golden age of microscopic petrography (Miyashiro, 1973), Zirkel (1876)
described the olivine in a basalt specimen thus: "its larger crystals altered along
the borders and cracks and its smaller ﬁlled with a brownish-red, somewhat
ﬁbrous substance, which is, without doubt, of a serpentineous character this
phenomenon of decomposition is evident in all the olivines of our American
basalts...", and further stated "Olivine is the substance in all rocks ﬁrst falling a
victim to alteration. . . Zirkel, apparently like most geologists of his time, did not
differentiate between alteration and decomposition. Serpentine had been used as
a decorative building stone since antiquity (Pliny, Natural History, XXXVI), and
Zirkel seemingly uses the term “serpentineous” in the sense of a common

adjective.

Several differe
thelirst edition of his
- antigorite, lizard ite
11837) mentioned “p
occurring “at Lizzarc
Edition of the Manor
Precious” varieties.
presumably what is

“Chrysotile' to the li.

”875), IIITIIIEd by II

"3“ iricl
Consist a
. imi
”Dona I
T)
of ty-

Several different varieties of serpentine were described by Dana (1837) in
the ﬁrst edition of his System of Mineralogy, but the three polymorphs now known
— antigorite, lizardite, and chrysotile - had not yet been named, although Dana
(1837) mentioned “precious serpentine” and “common serpentine”, the latter
occurring “at Lizzard’s Point, in Cornwall, and many other places”. By the Fifth
Edition of the Manual of Mineralogy, Dana (1869) was still listing “common” and
“precious” varieties, but had combined these in a class as "ordinary serpentine”,
presumably what is now known as lizardite, and had added “antigorite” and
“chrysotile” to the list of described varieties (Dana, 1867). Six years later, Zirkel
(1876), limited by the tools and knowledge available at the time, and thus
identifying any assemblage of ﬁne-grained materials occupying fractures in
olivine as "serpentine", might well have mis-identiﬁed iddingsite or other common

products replacing olivine.

Near the end of the same thirty-year period, Merrill (1897), lamented that
"Since the introduction of the microscope into petrographic work, there has
been very little time devoted to the study of rocks in a weathered condition." He
then undertook to remedy this by writing what was probably the ﬁrst
comprehensive treatment of the subject of rock weathering. This text, (Merrill,
1897) which included many bulk chemical analyses done by Merrill himself,
consisted mainly of a review of the observations of dozens of other workers. The
importance of this work is evidenced by the fact that it is still being cited a century

after it appeared (White and Brantley, 1995; Taylor and Eggleton, 2001).

Mei
extremely

weathen'n

Merrill (1897) noted that “geologists and petrologists as a rule have been
extremely careless in their use of such terms as alteration, decomposition, and

weathering [italics his] ”. He provided his own deﬁnition:

“The term weathering, as here used, is applied only to those
superﬁcial changes in a rock mass brought about through atmospheric
agencies, and resulting in a more or less complete destruction of the rock
as a geological body It does not include those deeper-seated changes —
changes taking place below the zone of oxidation and which result mainly
in hydration and the production, it may be, of new mineral species but
during which the rock mass as a whole retains its individuality and
geological identity It seems best to limit the terms weathering and
decomposition to processes involving the destruction of the rock mass as
a geological body, and to designate the purely mineralogical deeper-
seated changes as alteration, which may or may not be due to

hydrometamorphism [italics his] ”. (Merrill, 1897)

With regard to olivine, Merrill addressed only alteration: "Olivine is subject
to extensive alteration, becoming changed by hydration into serpentine or talcose
and chloritic products... The serpentineous alteration takes place along the
irregular curvilinear lines of fracture, and under favorable conditions continues

until the transformation is complete."

At about this same time, lddings (1892) encountered an unknown olivine

alteration product, now known as iddingsite. He wrote, "the resultant mineral from

its optical properties
individual, with para
year, Lawson (1893
substance in ldding-
carried out an exter
concluding that it w
intrusions. They als
(Ross and Shannoi
Ross and Shannon
”93W not a Weath

the lava, but subse

(Edwards 1938).

S”“l1957)
thorough SIUdy Of .

iddingsite does It

meﬁTSI SIUdy Of Id

its optical properties is evidently not a confused aggregate, but a crystallographic
individual, with parallel orientation of all its parts..." (lddings, 1892). The next
year, Lawson (1893), who found similar occurrences in California, named the
substance in lddings' honor. Several years later, Ross and Shannon (1925)
carried out an extensive study of iddingsite with optical and chemical methods,
concluding that it was the result of deuteric processes in lavas and shallow
intrusions. They also compiled an explicit chemical formula for the substance
(Ross and Shannon, 1925). Edwards (1938) generally conﬁrmed the ﬁndings of
Ross and Shannon, stating that the iddingsite in the exposures he studied "is
clearly not a weathering produc and "was formed during the consolidation of
the lava, but subsequent to its extrusion, and after it had come to rest."

(Edwards, 1938).

Sun (1957) maintained that, contrary to Ross and Shannon's ﬁndings, "A
thorough study of the available data on iddingsite leads to the conclusion that
iddingsite does not have a deﬁnite chemical composition or optical properties". In
the ﬁrst study of iddingsite using x-ray powder diffraction, he concluded that
"goethite is the only crystalline phase and that the other substances shown by
chemical analysis to occur in iddingsite are largely amorphous". Further, Sun
cautiously allowed that alteration "may or may not be continued in the weathering

process".

Wilshire (1958) discussed the molecular characteristics of several
alteration products of olivine and orthopyroxene, which he investigated with x-ray

powder diffraction. He found that, in addition to goethite, a green substance

7

9

which he characterized
llilshire suggests that.

(commonly containing .

Hay (1959) inve
"many crystals of oIivi
Contrary to the results
olivine crystals of unv

overall weathering of

Using single—<
determine the reaso
subsequent researc

goethrte and a "Iaye

°PIICaI proneriies o<

‘ .

which he characterized as a mixed-layer smectite-chlorite is commonly formed.
Vlﬁlshire suggests that, in addition to deuteric alteration, "oxidized varieties

(commonly containing goethite) may also form by weathering of green varieties"

Hay (1959) investigated late Pleistocene ash deposits and noted that,
"many crystals of olivine are partially or completely replaced by iddingsite".
Contrary to the results of previous studies, he found that "The latter is absent in
olivine crystals of unweathered deposits, and its thickness is proportional to the

overall weathering of the ash; hence it is a product of weathering" (Hay, 1959).

Using single-crystal x-ray diffraction, Brown and Stephen (1959) sought to
determine the reason for the optical homogeneity noted by lddings (1892) and
subsequent researchers. They designated the phases comprising iddingsite as
goethite and a "layer lattice silicate", and they concluded that the "homogeneity in
optical properties occurs because the small crystals of both components are
strictly oriented throughout a single grain. The parallel alignment arises from the
nature of the alteration, the products of which inherit, goethite completely and the
layer lattice silicate partly, the oxygen framework of the original olivine" (Brown

and Stephen, 1959).

Gay and LeMaitre (1961) used methods similar to those of Brown and
Stephen (1959), but concluded that iddingsite could not be regarded "as a simple
sub-microscopic intergrowth of two or more well-characterized minerals". Instead,
they found that '"lddingsitization' is a continuous transformation in the solid

state". Although it "may be possible at any stage to recognize embryonic

1O

  
  
  

structural arrangement.
structures the alteret
arrangement...”. These
iron-rich, aqueous and a
pressures up to 1000 a
that "When alteration ri

were of the serpentine

Sheppard (196
partially iddingsitized
deuteric process. Hos

Sections of basaltic a

associated with all of

rims. suggest to us ti

and they Concluded ‘

simi - -
lar mposrtion a

structural arrangements, some of which approximate to normal ordered mineral
structures the altered olivine is at all times a disordered irregular
arrangement. ..". These workers also attempted to synthesize iddingsite from
iron-rich, aqueous and acid solutions, and at temperatures up to 600° C and
pressures up to 1000 atmospheres. They were, however, unsuccessful, ﬁnding
that "When alteration rims were produced, x-ray examination showed that they

were of the serpentine type."

Sheppard (1962) claimed that formation of fresh olivine overgrowths on
partially iddingsitized olivine phenocrysts proved that iddingsitization was a
deuteric process. However, Baker and Haggerty (1968) in their study of 500 thin
sections of basaltic and associated lavas, argued that "certain properties
associated with all of the 'iddingsites' studied, including those with fresh olivine
rims, suggest to us that the 'iddingsite' forms at low temperature in all cases",
and they concluded that "Weathering of olivine may produce pseudomorphs with

similar composition and properties to deuteric forms of 'iddingsite'.”

Delvigne et al. (1979) produced a comprehensive study of olivine
weathering forms and processes, in which they described the common products
of olivine weathering and alteration, including serpentine, iddingsite, chlorite,
nontron'rte, and iron hydroxides. Descriptions addressed occurrence,
composition, optical characteristics, and micromorphology. The textures
associated with these replacement products were illustrated with detailed line
drawings. Bowlingite/saponite and chlorophaeite were also described, but without

illustration. This work provided the basis for Delvigne's (1998) color atlas of

11

mineral alteration form

for alteration phenome

Using transmis
formation of iddingsiti
olivine crystals and d

product was called "

similarto brucite. Ms

formation of the 10-

Chemical analysis,

alteration assemble

Smith et al.

demonstrate that E

WEathering.

Banﬁeld et
weathering of olivi
olivine oxidation. 4
products. They di<
The third epis

Ode

mineral alteration forms, in which he introduced a genetic classiﬁcation scheme

for alteration phenomena, including its own Greek-based descriptive lexicon.

Using transmission electron microscopy (TEM), Eggleton (1984) studied
formation of iddingsite "composed of saponite and goethite" in unweathered
olivine crystals and described a two-stage alteration process. The ﬁrst-fanned

product was called "metastable phase M" by Eggleton because its structure was

similar to brucite, MgOHz. The second stage in the alteration process was "the

formation of the 10-A layer silicate in intimate association with phase M". Final
chemical analysis, however, showed "only 10-A layer silicate and goethite in the

alteration assemblage."

Smith et al. (1987) used TEM and scanning electron microscopy (SEM) to
demonstrate that Eggleton's model applied as well to iddingsites formed by

weathering.

Banﬁeld et al. (1990) used TEM to study both the alteration and
weathering of olivine in basaltic andesites. They identiﬁed three episodes of
olivine oxidation, each characterized by a distinct assemblage of alteration
products. They did not identify a separate weathering stage, stating only that
"The third episode of alteration involves the destruction of olivine by low-
temperature hydrothermal alteration and weathering." The same locale was
discussed by Banﬁeld et al. (1991) in terms of weathering of olivine, and other

primary minerals, as characterized by TEM of lake sediments.

12

In summary, mc
products of olivine four.
iscomposed of the iron
silicate. probably a sme
result of deuteric alterat

studies, Hay (1959) am

to weathering.

In summary, most of the researchers who investigated the alteration
products of olivine found that iddingsite, which commonly pseudomorphs olivine,
is composed of the iron oxyhydroxide goethite in a physical mixture with a layer
silicate, probably a smectite. The majority concluded that iddingsite was the
result of deuteric alteration. Some workers were equivocal, and only two of the
studies, Hay (1959) and Smith et al. (1987), attributed iddingsitization exclusively

to weathering.

13

Morton and Ha
as the ﬁrst to discuss
“the presence of corr
Hallsworth. 1999). H
published at least as
etched sapphires frc
crystallographic me;

(Prat 1897, p. 424'

Hay(1959)c
volcanic ash. He f0

pitting to sha
II)

it) cres

e Successive Stag
Ingle'ssion of iii/eggI

mineral Component

Chapter 2

Natural Textures

Morton and Hallsworth (1999) credit Edelman and Doeglas (1932, 1934)
as the ﬁrst to discuss mineral alteration features and relict structures, including
"the presence of corrosion textures on heavy mineral grain surfaces" (Morton and
Hallsworth, 1999). However, descriptions of etching ﬁgures on minerals were
published at least as early as the 1890's, when Pratt (1897) reported on naturally
etched sapphires from Montana, which were "etched to such a degree that no
crystallographic measurements were possible on the reﬂecting goniometer."

(Prat, 1897, p. 424).

Hay (1959) described brieﬂy the etching of olivine and other minerals in
volcanic ash. He found that "Etching of the olivine crystals ranges from ﬁne
pitting to sharp crests parallel to (010)." Brown and Haggerty (1967) described
the successive stages of low-temperature alteration of an olivine grain by the
ingression of wedges ("teeth") of green phyllosilicates, and mapped the various

mineral components of a typical phyllosilicate wedge.

Stoops et al. (1979) developed a classiﬁcation scheme for mineral
alteration textures in individual mineral grains, based on six different
characteristic textures and ﬁve degrees of alteration. The study of olivine
weathering forms by Delvigne et al. (1979) contained detailed descriptions of the

textures of common alteration products including iddingsite, chlorite,

14

   
   

bowlingitelsaponite. ch
micromorphology, they
which are not usually vi
distinguished during thi
of such structures. All i
respective alteration p

not addressed.

Nahon et al. (1
tephroite in bauxite. 8
illustrating the texture

stages, but did not dt

In his TEM stt

bowlingite/saponite, chlorophaeite, nontronite, and iron hydroxides. In terms of
micromorphology, they stated that "the crystallographic and cleavage patterns,
which are not usually visible in fresh olivines, can be exceptionally well
distinguished during the ﬁrst stages of alteration", and then described ﬁve types
of such structures. All of these structures, however, were composed of the
respective alteration products, while corrosion textures in the olivine itself were

not addressed.

Nahon et al. (1982) described in detail the weathering of forsterite and
tephroite in bauxite, and provided SEM photographs and structural formulas
illustrating the textures typical of Mg-Fe-Mn smectites at successive weathering

stages, but did not describe textures in the olivines themselves.

In his TEM study of iddingsite rims on olivine, Eggleton (1984) identiﬁed
changes in the microtexture of olivine in basalt, resulting from the ﬁrst stage of
olivine alteration. First, 50-A diameter needle-shaped domains are formed,
which then change to a metastable hexagonal phase. Solution channels are
opened by this reaction, and these channels are then partially ﬁlled with laths of
smectite. Although the emphasis this study was on the reaction products, it
described dissolution textures in olivine itself, albeit at sub-microscopic scale.
However, the study was carried out on olivine that had been altered at
temperatures above ambient, and water was thought to have been derived from
the reaction between the olivine and glass in the enclosing basalt. In a later
article referring to the same study, Eggleton (1986) described the alteration as

“comparable to weathering”, and provided a schematic diagram of the various

15

 

phases and the disso

dislocation arr aY$~

Smith et al. (1
ingression of planar I
products". These Chi
parallel to (001) and
altered olivine grains
points of the edges I
ofphyllosilicate wen
relationships betwei
iddingsite formation

(1987) focused on t

Using TEM t

phases and the dissolution channels, which were thought to emanate from

dislocation arrays.

Smith et al. (1987) found that "alteration of olivine proceeds via the
ingression of planar ﬁssures or channels ﬁlled with microcrystalline alteration
products". These channels, which are visible with an optical microscope, "are
parallel to (001) and are responsible for the typical lamellar structure in partly
altered olivine grains. In detail, the channels are wedge-shaped, with the sharp
points of the edges projecting into the unaltered olivine." On TEM images, strips
of phyllosilicate were seen to bridge the channels. TEM images also showed the
relationships between goethite crystals and smectite layers in the later stages of
iddingsite formation. Like most of the earlier workers, however, Smith et al.

( 1987) focused on the replacement products.

Using TEM to study natural olivine alteration, Banﬁeld et al. (1990),
interpreted three stages of alteration including a low-temperature episode
characterized which "involves the destruction of olivine by low—temperature
alteration and weathering." They describe nanometer-scale "elongate etch pits
and channels in the margins of fresh olivine crystals" that "contain semi-oriented
bands of smectite." These "textures resemble those reported by Eggleton (1984)

and Smith et al. (1987)."

A century after Merrill (1897), Delvigne (1998) lamented, in the foreword to
his monumental book on the micromorphology of mineral alteration (including

weathering), that many Earth scientists have, "over the last ﬁfteen years

16

neglected fund amer
their samples at the
the study of a few tl
inclination to procei
five hundred pages
interpretive text. TI
olivines. are QIOUp
mineral forms, cal
describing alteron
complememary (1'
Within the alimon—

deals only With th

etch IEXiUres in c

neglected fundamental descriptive work in the ﬁeld", and that "The observation of
their samples at the scale offered by the optical microscope, is often restricted to
the study of a few thin sections. . . in fact, young scientists have a strong
inclination to proceed directly to the ultramicroscopic scales." Following this are
ﬁve hundred pages, profusely illustrated with six hundred photomicrographs and
interpretive text. The minerals illustrated, about twenty percent of which are
olivines, are grouped according Delvigne's Greek-based nomenclature of altered
mineral forms, called alteromorphs. Also included are several chapters
describing alteromorphs "on the basis of microtextural criteria, i.e. the
complementary distribution of the secondary products and the residual pores
within the alteromorphs." Delvigne ( 1998), however, like most previous authors,
deals only with the morphology of alteration products, without illustrating unﬁlled

etch textures in olivine.

Velbel and Ranck (2008) described the micromorphology of weathering
features in tectonized olivine from the southern Blue Ridge of North Carolina.
The study dealt entirely with corrosion textures, and not with alteration products
or processes. These workers observed that aqueous weathering features in the
samples studied were crystallographically controlled, and were either conical
(funnel-shaped) pits, or were compound shapes made up of intersecting conical
pits. They suggested that the apices of the cones were at dislocations in the
crystal structure, and the etch pits develop as their apices follow the dislocation
array, with corrosion spreading progressively outward in circular patterns from

the axes of the cones. Velbel and Ranck (2008) further noted that, while some

17

etching might I
near weatherei

formed by wea

Velbel I
form of a comp
by SEM image
which such a 5
Samples from
the ﬁndings of
either 000103]
Ifacture sUr-fa C
diamond-Shag
associated Wi‘

Portions OI ro<

etching might have taken place during serpentinization, etch pit size increased
near weathered surfaces, and the largest etch pits (>10 pm in length) were

formed by weathering, not by pre-weathering alteration.

Velbel (2009) expanded on the work of Velbel and Ranck (2008) in the
form of a comprehensive study of natural weathering textures of olivine illustrated
by SEM images. Noting that olivine was the only major silicate mineral group for
which such a study was lacking, Velbel (2009) reported on weathering textures in
samples from olivine-phyric basalts and from tectonized dunites, and conﬁrmed
the ﬁndings of the earlier work by Velbel and Ranck (2008). All etch pits are
either conical depressions intersected at various angles by thin section or grain
fracture surfaces, or are composed of cones joined base-to-base forming
diamond-shaped cross-sections on exposed surfaces. Olivine etch pits are
associated with fractures and olivine-serpentine interfaces, and occur only in
portions of rocks exposed in outcrops. All etch pits are attributable to post-
serpentinization weathering, and “can be explained as being related to ingress of
low-temperature Earth-surface aqueous solutions to the site of olivine etching”.
Comparing natural weathering textures with those formed experimentally, Velbel
(2009) reported that the only experimentally formed etch pits that were similar to
pits formed by natural weathering of olivine were those created by etching with

highly acidic reagents.

Thus, while there were many studies in the Twentieth Century that
touched on weathering textures tangentially, in connection with natural alteration

of olivine to iddingsite and other alteration or weathering products, it was not until

18

very recently. with if
detailed descriptions

themselves were pL

very recently, with the works by Velbel and Ranck (2008) and Velbel (2009), that
detailed descriptions and illustrations of the olivine weathering textures

themselves were published.

19

 

Although til
an iron meteorite i
is credited with be
1894). According
study the dissoluti
of matter were spl

could account for

Another Br

known for formula

Chapter 3

Textures Formed Experimentally

Although the Austrian A. Widmannstatten had etched a polished section of
an iron meteorite in 1808 (Honess, 1927), the British chemist J. F. Daniell (1816)
is credited with being "the ﬁrst observer of the etching of crystals" (Baumhauer,
1894). According to Greene and Burke (1978), Daniell used the etch method to
study the dissolution of crystal faces in order to show "that the ultimate particles
of matter were spheres and that the various modes of stacking these spheres

could account for the diverse conﬁgurations of crystals."

Another British scientist, the physicist D. Brewster (1837, 1853), best
known for formulating the rule governing polarization of light by reﬂection
(Brewster's Law), "gave the ﬁrst report on light-ﬁgures" (Baumhauer, 1894) made

by reﬂected light on crystal faces of minerals roughened by chemical etching.

The inﬂuence of Brewster's description of light ﬁgures was felt mainly in
Germany, where A. Leydolt (1854) repeated Brewster's experiments, and
popularized the etch method among German scientists (Baumhauer, 1894). Over
the next four decades, a number of German researchers used the etch method to
study a variety of minerals. Honess (1927) lists works by a dozen scientists.
According to Honess (1927), "The general content of these papers comprises a
theoretical study of the etch ﬁgure in connection with extensive experimental

work, involving the etching of many of the commoner minerals." One of these

20

experimenters was
difference in etch r

of chemical reager

Probably th
comprehensive te;
Research" by H. E
obtained on nine I
PhotornicrographS
thorOUgh and unt
tears, duﬁng Whi

author bearing 0‘

At the em
undertook "A Cc

pyroxenes" in v

experimenters was Becke (1890), who "appears to be the ﬁrst who reported a
difference in etch rates along different directions on crystal surfaces by the action

of chemical reagents." (Sangwal, 1987).

Probably the most signiﬁcant of the late-nineteenth-century works was the
comprehensive textbook "Results of the Etch Method in Crystallographic
Research" by H. Baumhauer (1894). This 131-page book describes etch ﬁgures
obtained on nine representative minerals, and includes a folio of 48
photomicrographs. Honess ( 1927) describes Baumhauer’s book as "a very
thorough and untiring study of the etch method covering a period of over twenty
years, during which there appeared thirty separate publications by the same

author bearing on this line of investigation."

At the end of the nineteenth century, the American R. A. Daly (1899)
undertook "A comparative Study of Etch-ﬁgures. The Amphiboles and
Pyroxenes" in which he described etch features produced on more than 60
samples of amphiboles. The 67-page article was illustrated with 30
photomicrographs and 40 line drawings, and included extensive notes on
solvents and methods. However, with only ﬁve samples of pyroxenes, Daly
(1899) acknowledged that "This is one part of the investigation which the author

regrets to leave at an especially incomplete stage".

In the early years of the twentieth century, "a lively interes " was "shown in
the etching of crystals" (Honess, 1927). However, with regard to deﬁnitive works

on study of etch ﬁgures, Honess (1927) wrote that Baumhauer’s (1894) book, "as

21

 

far as is known to tI
available today." It i
only in German. Ho
ﬁgures, one is impri
investigation. Very I
this field until the la

the X-ray analysis c

In 1927 Hon
Figures on Crystal:
"Baumhauers wort
Honess Provided c
as well as the diffs

apophyllite, 10p 82

far as is known to the present writer, is the only complete book on etch ﬁgures
available today." It should also be noted that Baumhauer's book was available
only in German. Honess remarked further that "In reviewing the literature of etch
ﬁgures, one is impressed by the leadership of the Germans in this line of
investigation. Very few English or American mineralogists have been attracted to
this ﬁeld until the last few years, which mark the discovery and development of

the X-ray analysis of crystals." (Honess, 1927).

In 1927 Honess' own book, "The Nature, Origin, and Interpretation of Etch
Figures on Crystals", was published. Honess (1927) acknowledged that
"Baumhauer's work has been freely consulted in the preparation of this book".
Honess provided detailed information the attributes of various chemical etchants,
as well as the differing responses of ten minerals (including the silicates

apophyllite, topaz, and tourmaline) to these solvents.

Almost sixty years later, Wegner and Christie (1985b) wrote that "works by
Daly ( 1899) and Honess (1927), describing etch ﬁgures on crystals, contain a
wealth of information on etchants and their use. Because many of the 'etch
ﬁgures' in crystals are localized at the emergence of dislocations, much of this
classical work is undoubtedly applicable to production of dislocation etch pits on

silicate mineral crystals." (Wegner and Christie, 1985).

In the 1930's and 1940's "with the advent of x-rays and their
determination of the crystal structure, the etching technique suffered a setback"

(Sangwal, 1987). However, in the late 1940's, the results of several studies by

22

 

   
  

physicists "showed th
screw dislocations." (t
papers on the ﬁeld of
which time "the etching

characterization of stri

What was repc
Christie, 1974) was c
ﬁnd a suitable etchar
The results indicated
which repriented the

caused the deveIOpi

physicists "showed that etch pits could serve to locate the sites of edge and
screw dislocations." (Sangwal, 1987). Then, during the 1950's, the number of
papers on the ﬁeld of etching increased from only one in 1950 to 109 in 1961, by
which time "the etching technique had become an established tool for the

characterization of structural defects." (Sangwal, 1987).

What was reportedly the ﬁrst etching experiment with olivine (Wegner and
Christie, 1974) was carried out by Horn and Maurette (1967). Their aim was "to
ﬁnd a suitable etchant for heavy-ion-tracks in olivine." (Horn and Maurette, 1967).
The results indicated that, with certain etchants, elongated etch pits developed
which re-oriented themselves as they grew larger, whereas other etchants

caused the development of elongated pits that grew in a constant direction.

In the mid-1970's, laboratory studies involving dissolution of olivine and
other magnesium silicates (Luce et al., 1972; Choi et al., 1976; and Grandstaff,
1977) showed dissolution rates decreased with time. This was generally
interpreted as "resulting from diffusion control of the reaction rate, with diffusion
occurring through a reprecipitated layer or a residual leached layer." (Grandstaff,

1978).

However, Berner and Holdren (1977, 1979) concluded, after studying
both naturally weathered and acid-etched feldspars using SEM and X-ray photo-
electron spectroscopy (XPS), that "the ﬁne-grained material remaining on the
surface of feldspars does not constitute a continuous, tightly-adhering, low-water-

content layer." Based on interpretation of "several hundred photographs of both

23

naturally weathered a

piece together a disso
initial stage in which s
development of deep,

etch pits to produce a

Berner et at. i
and amphiboles USII
of weathered and a
Berner and Holdre
selective etching a
and that “pyroxen
leaction and not

Protective Ia‘i Er ‘

naturally weathered and artiﬁcially HF-leached feldspar grains" they were "able to
piece together a dissolution sequence". This sequence is characterized by an
initial stage in which shallow, almond-shaped etch pits are formed, followed by
development of deep, square-shaped etch pits, and ﬁnally by the coalescing of

etch pits to produce a "rough, pitted, and furrowed surface."

Berner et al. (1980) and Berner and Schott ( 1982) investigated pyroxenes
and amphiboles using an approach similar to that of Berner and Holdren's study
of weathered and acid-etched feldspars. They drew similar inferences to those of
Berner and Holdren, ﬁnding that "silicate weathering, in general, takes place by
selective etching and not by general attack on the surface." (Berner et al., 1980),
and that "pyroxene and amphibole weathering is controlled by surface chemical
reaction and not by diffusion either through aqueous solution or through a

protective layer on the mineral surface."

Grandstaff (1978) carried out a study of olivine similar to that of Berner
and Holdren's (1977) experiments with feldspars. After being partially dissolved
in acid, some areas of grain surfaces had been highly etched, but others showed
only minor pitting. Grandstaff interpreted this as showing that olivine grains are
"anisotropic with respect to dissolution: that dissolution occurs at much greater
rates on some surfaces than on others". It appeared that pitted surfaces
developed "in areas of intersecting cleavage planes or lattice dislocations. If the
material bordering the dislocations is preferentially removed, perhaps due to high
surface free energy, material between the intersecting planes may remain as

isolated pinnacles". Further, Grandstaff (1978) found that "SEM

24

mycrophotographs I

or leached layers."

Kirby and W1
study chemically etr
in the Earth’s mantli
dislocations on etch
that "(010) arrays w
[100] and [001]", an
the twist type.“ They
Co"Nosed of arrays
SITorig tendency for

microphotographs show that most surfaces are not covered by thick precipitated

or leached layers."

Kirby and Wegner ( 1978) used transmission electron microscopy (TEM) to
study chemically etched surfaces of olivines as a means of estimating ﬂow stress
in the Earth's mantle. They took it as axiomatic that "The points of emergence of
dislocations on etched surfaces are marked by etch pits." Speciﬁcally, they found
that "(010) arrays were composed of dense networks of dislocations parallel to
[100] and [001]", and "that the dislocations are screws and that the arrays are of
the twist type." They also found that (100) subgrain boundaries are principally
composed of arrays of dislocations parallel to [010]." In general, 'There is a very
strong tendency for dislocation etch pits in the crystals under study to align
themselves in arrays parallel to the traces of the pinacoidal planes. The driving
potential for these alignments is the elastic strain energy of the dislocation
arrays; certain conﬁgurations are favored because they minimize the boundary

energy."

Seyama et al. (1996) used XPS and SEM to study the surface alteration of
forsterite and fayalite. Although the "XPS result offers direct evidence for the
formation of a leached surface layer" in fayalite, this only "gives information about
changes in the average surface composition", and "the SEM observations of
olivine grains showed that olivine dissolution proceeded heterogeneously",
resulting in "a harshly etched structure for fayalite and a locally etched structure

for forsterite."

25

 

Awad et al. (1
identiﬁed different fc
crystal axes. Cubes
1 or pH 2, at temper
pH and temperature
found that "The shag
crystallographic direi
dominated by "rectai
Sides Of the pits were
baxis formed "bro ac
walls", While i.

diamor

direCtlon of the C~axi

Awad et al. (2000) examined acid-etched olivine cubes with SEM and
identiﬁed different forms and dimensions of etch pits corresponding to different
crystal axes. Cubes of forsteritic olivine were etched in diluted sulfuric acid at pH
1 or pH 2, at temperatures ranging from 23° C to 90°, and for durations scaled to
pH and temperature. The olivine cubes were 1 mm on a side. Awad et al. (2000)
found that "The shape and geometry of the etch pits are characteristic for each
crystallographic direction." Cube faces "in the direction of the a-axis" were
dominated by "rectangular-shaped etch pits with V-shaped bottoms", and "the
sides of the pits were frequently covered by incipient pits." Dissolution down the
b-axis formed "broad, ﬂat-bottomed rectangular etch pits with steeply sloping
walls", while "diamond-shaped etch pits are characteristic for the cube face in the

direction of the c-axis.".

In general, experimental studies of olivine textures have been carried out
either to advance techniques of deﬁning dislocations, or to resolve questions of
dissolution kinetics. Such studies were begun rather late, in the 1960’s, 150
years after Daniell’s etching experiments with calcite in 1816. The most recent
study reviewed, that by Awad et al. (2000), provides the most detailed description
to date of the morphology of olivine etch pits in relation to crystallographic axes.
Illustrations from several of those papers reviewed above that deal with olivine
textures are reproduced in Chapter 15: “Comparison of Observed Natural

Features with Textures Developed Experimentally.”

26

 

Early experii
were R. E. Rogers
(1877) in Germany
readiness with whit
decomposing and <
fact we believe no

found that it was "t1

1897).

Many or the
IEXTUI'es"‘ w

Chapter 4

Experimental Kinetics

Early experimenters with dissolution of silicate minerals in water and acids
were R. E. Rogers and W. B. Rogers (1848) in the United States and R. Miiller
(1877) in Germany. Rogers and Rogers (1848) noted that "The comparative
readiness with which the magnesian and calceo-magnesian silicates yield to the
decomposing and dissolving action of carbonated and even simple water, is a
fact we believe no less important than it is true." With respect to olivine, Miiller
found that it was "the most readily attacked of all the silicates tested" (Merrill,

1897).

Many of the works reviewed in the previous chapter, under "experimental
textures", were undertaken to address dissolution kinetics. Grandstaff ( 1978)
sought to identify a mechanism that would explain why olivine dissolution rates
decreased with time, as was determined for other silicate minerals by previous
workers (Luce et al., 1972, Choi et al., 1976; and Grandstaff, 1977). Grandstaff
(1978) found that the results "suggest that dissolution is controlled by rates of
surface reactions" and that "although the speciﬁc surface area of olivine
increased during the dissolution experiments due to dissolution along planar
discontinuities, the rate of dissolution decreased as this rapidly dissolving
material was removed. The onset of linear dissolution may occur when general
surface dissolution comes to dominate the overall rate of reaction." Thus
Grandstaff ( 1978), in concluding that surface reaction was the rate controlling

27

step in

Holdrei

fayalite

the cou
precipit

Precipit

the amt

ox‘lgen

step in the dissolution of olivine, conﬁrmed the similar ﬁndings of Berner and

Holdren (1977) with regard to dissolution of feldspar.

Siever and Woodford (1979) carried out dissolution experiments with
fayalite and other maﬁc minerals, excluding forsterite. They found that "During

the course of experiments using oxygen atmospheres, a reddish-brown

precipitate, presumably Fe(OH)3, formed after the ﬁrst few days. A similar

precipitate formed much more slowly in air." Although they "could not determine

the amount of precipitate", they concluded that "dissolution in the presence of

oxygen is slowed by the annoring effect of the Fe(OH)3 ..."

Following Berner's (1978) reasoning that "during surface reaction-
controlled dissolution, ion detachment is so slow that selective dissolution of the
crystal surface occurs", Berner and Holdren (1979) concluded from their study of
natural and acid-etched feldspars that "dissolution occurs by the development of
etch pits at points of excess energy". This means that "attack on the surface is
selective [italics theirs]. If diffusion through a continuous surface layer were
rate limiting, concentrations of dissolved species should build up to saturation at
the interface and become laterally uniform". This, in turn, means that "under
diffusion control, surface attack should be non-selective and, as a result, general
rounding of feldspar crystals should occur. Instead a highly angular surface,
resulting from etch pit formation and coalescence, is formed which is

characteristic of surface reaction controlled dissolution."

28

Be
pyroxene
during we

diffusion

llv

forsteritic

values, v
dissolutii

WOQEIIU:
ComDosi

Alssolutt
Conclude
amOllnt l
hydroxid
r93301an
ICFSleme
IXiaring l

Berner and Schott ( 1982) came to the same conclusion after treating
pyroxenes and amphiboles in acid: "that pyroxene and amphibole dissolution
during weathering is a surface reaction-controlled process and not limited by bulk

diffusion either in solution or through a protective layer.”

Wogelius and Walter (1991) measured the dissolution rates of two

forsteritic olivines (F0100 and F091) in aqueous solutions in a wide range of pH

values, with and without 002 and organic ligands. They derived an olivine
dissolution rate equation for the low C02 pressure range. Following this,

Wogelius and Walther (1992), working with olivines of widely different

compositions (forsterite F091 and fayalite F06). again experimented with

dissolution in liquids of different pH levels. With respect to fayalite, they
concluded that, "apparently, after an extended period of time, a signiﬁcant
amount of Fe on the mineral surface becomes oxidized or enough fern-oxide or -
hydroxide has precipitated on the surface to decrease the number of rapidly
reacting surface sites and thereby decrease the dissolution rate." In the case of
forsterite, they stated equivocally that "the precipitation of secondary Fe(lll)-
bearing phases (or in situ oxidation of surface Fe) may eventually interfere with

the dissolution reaction."

Using data on stochiometries and molar volumes of minerals, Velbel
(1993a) showed that most major rock-forming silicates, including olivines, cannot
form protective surface layers during well-leached, oxidizing conditions because

the volume of the product formed is less than the volume of reactant consumed.

29

 

COrnDreh

Later, Vt
orthosilic
and faya

and varii

techniqt
namn
of forste
in silica,
a magn.
ItStitched
Rte late

tetrahec

found if
BdISSOIU
POnCIUd
COmblnf

8‘3qu is

One Unit.

RI

I

Later, Velbel ( 1999) explained variations in dissolution rates among
orthosilicates, including the signiﬁcantly different dissolution rates of forsteritic
and fayalitic olivines, in terms of the strengths of bonds between oxygen anions

and various cations.

Pokrovsky and Schott (2000) used several spectroscopic and chemical
techniques to "characterize the surface chemistry of forsterite in aqueous solution
as a function of solution composition and pH." They determined that "the surface
of forsterite reacted in solutions at pH < 9 is depleted in magnesium and enriched
in silica. This silica-rich layer is formed by the exchange of two hydrogen ions for
a magnesium atom at the forsterite surface." However, "the thickness of this Mg-
leached layer does not exceed 1-2 unit cells, i.e., 10-20 A." They proposed that
the rate limiting step is "decomposition of protonated polymerized silica

tetrahedra linked to Mg atoms deeper in the structure."

In their experiments dissolving olivine cubes in acid, Awad et al. (2000)
found that "the cubes visibly shortened in the direction of the b-axis", and that
"dissolution down the a-axis is signiﬁcantly slower than down the c-axis." They
concluded that "the dissolution process probably involves surface reactions
combined with near-surface volume-type diffusion of protons into the olivine
structure", assuming that "protons readily diffuse 5 to 10 A (a distance of about

one unit-cell length) into the structure."

Rosso and Rimstidt (2000) sought to contribute to the "establishment of

comprehensive [olivine dissolution] rate laws based upon well understood

30

 

Ro
mmpmhe
reaction r
collecting
activated
tefmsh
associate
adsorbei
“1098 Itc
bonds"
eiperim.

reaCIlOry

C

LI” 91 al
m9Ciel n
DH Iprot
ICands)
by initial

Rosso and Rimstidt (2000) sought to contribute to the "establishment of
comprehensive [olivine dissolution] rate laws based upon well understood
reaction mechanisms" and to minimize the "effect of random errors by
collecting a large amount of data." Based on their results, they "postulate that the
activated complex for forsterite dissolution involves a hydronium ion adsorbed to
the forsterite surface in such a way that two of the three hydrogen atoms are
associated with two Si-O-Mg bridging oxygen atoms. In this conﬁguration, the
adsorbed hydrogen ion dissociates to form H-O-Si bonds as water molecules
move from the solution to hydrate the two Mg atoms, breaking the Mg-O-Si
bonds." They conclude that the "high resolution results" produced by their
experimental approach "create signiﬁcant constraints on any model of the

reaction mechanism for olivine dissolution." (Rosso and Rimstidt, 2000).

Combining chemical experimentation with quantum-mechanical modeling,
Liu et al. (2006) sought to replicate the "four important phenomena that a rate
model must explain." These are 1) increasing speed of dissolution with declining
pH (proton-promoted dissolution), 2) increased dissolution in the presence of
ligands (ligand-promoted dissolution), 3) production of a silica-enriched surface
by initial (incongruent) dissolution, and 4) correlation of olivine dissolution rates
with rates of water exchange with corresponding M2+ cations. These researchers
developed "large cluster models More than 200 atoms are in each model. The
bottom layer of atoms in each model was anchored at crystallographic

locations".

The models developed by Liu et al. (2006) showed that

31

o 325

evt
Ms
SL

th

de

#01

The p
Ira/planet’s l2.
olivine dissoluti

wide range of h

that

"In all cases, the bonds break between Mg and O atoms leaving the

O atom with its afﬁxed proton attached to the adjacent silica group. As the

Mg-O bonds break, water molecules coordinate with the Mg2+ ion until it
eventually dissociates from the surface as M92+(H20)5. The departing
Mg2+ ion carries away the positive charge that had been deposited on the
surface by the protonation steps. The 8104 units remain tightly linked to

2
the bulk structure after the departure of the Mg + from the surface layer.”

“The silica enriched surface predicted by our ab initio quantum
mechanical model appears to be relatively stable suggesting that the rate
determining step for the overall dissolution process is the release of silica

from the surface.”

The possibility that there had been liquid water on Mars, at some time in
that planet's history, provided the impetus for Stopar et al. (2006) to apply the
olivine dissolution rates determined by Wogelius and Walther (1991, 1992) to a
wide range of hypothetical conditions. The logical basis for this approach was

that

"If water was ever abundant for a signiﬁcant time, then we might

expect much of the surface to be chemically altered.”

“By studying the dissolution of primary minerals under simulated
Martian conditions, we can begin to quantify the extent and duration of

32

aqueous
olivine be

amounts

Paramel

lradius 0.01 on
Results were g

Illlermediate v.

Olson a
the duration 0
dissolution Sty
besari with a
With pH belwi
Importam rate
the SIlltllat ar
However, the
grain lifegme

i‘rreiimes Can

aqueous alteration. In particular, olivine is notorious for altering quickly in
olivine bearing basalts and is, therefore, a sensitive indicator of moderate

amounts of aqueous alteration."

Parameters included olivine composition (F0100 to Foo), particle size

(radius 0.01 cm to 0.1 cm), temperature (-50°C to 100°C), and pH (2 to 12).
Results were graphically presented for different parameter pairs, allowing

intermediate values to be determined by interpolation.

Olson and Rimstidt (2007), also interested in the implications of olivine for
the duration of liquid water on Mars, generalized "data from eight olivine
dissolution studies" and used them to construct a mineral lifetime diagram. They
began with a "baseline model for the lifetime of a 1 mm grain in dilute solutions
with pH between 0 and 12 at 298 K" then evaluated "how changing other
important rate controlling parameters affects this lifetime." They acknowledged
the similar analysis of olivine lifetimes carried out by Stopar et al. (2006).
However, they maintained that "our mineral lifetime diagram shows how olivine
grain lifetimes relate to rate-controlling variables so that new estimates of grain

lifetimes can be made as our knowledge of Martian surface conditions evolves."

Studies of experimental kinetics in the 1970’s and 1980’s were mainly
focused on determining whether dissolution was transport-controlled or surface-
interaction controlled. Most of these studies concluded that, in forsteritic olivine,
the rate-deterrnining step was surface interaction, while there was some

evidence that in fayaltic olivine transport might be the rate-determining step due

33

mmhmwm
studies of kinet
carrying out ne
experiments. it
molecular leve

further reseatt

to the formation of a protective surface layer. In the 1990’s and since then,
studies of kinetics have evolved toward more analytical works. Rather than
carrying out new physical experiments, these studies have used data from earlier
experiments, either to explain speciﬁc aspects of olivine dissolution at the
molecular level, or to generalize from existing data to provide new tools for

further research.

The li
composition
textures in o.
cleavage pla
al.,1987), bL
Ullder the re;
did not arise
decreased W1

OtherWise IOTI

Natun
Very re‘Cently
descnbed aht
IeCIOTIIZed du
Ullifoymly OVey

Shamlygeﬁne

Chapter 5

Implications for a Rate-Determining Mechanism.

The literature on products of natural weathering focused on the mineral
composition of alteration products and the processes of their formation. Alteration
textures in olivine were found to be aligned to the crystallographic axes and
cleavage planes in weathered olivine (Hay, 1959, Delvigne et al. 1979, Smith et
al., 1987), but little attention was given to the surface textures of the olivine itself,
under the replacement minerals. The question of a rate-determining mechanism
did not arise until laboratory experiments showed that dissolution rates
decreased with time, suggesting that a protective coating was deposited or

othenlvise formed on the surface of the olivine.

Natural dissolution textures of olivine did not receive close attention until
very recently. The papers by Velbel and Ranck (2008) and Velbel (2009)
described and interpreted natural weathering microtextures of olivine in both
tectonized dunites and phenocrysts in basalts. Dissolution does not take place
uniformly over a surface, creating round forms, but instead creates arrays of
sharply-deﬁned etch pits. This suggests that dissolution attacks dislocations
intersecting the olivine surface, and the etch pits deﬁne the sites of the
dislocations. Thus, as shown by Berner (1978), the dissolution process is

interface reaction controlled.

35

Experimentally produced textures have generally shown alignment of etch
pit arrays with each other and with crystallographic axes. This has been
interpreted by most researchers as indicating that interface reaction is the rate-
controlling mechanism, and that dissolution proceeds more rapidly along
dislocations parallel to crystallographic axes because this is where excess

surface energy is greatest (Kirby and Wegner, 1978; Awad et al., 2000).

Experiments involving dissolution kinetics further strengthened the view
that interface reaction is the rate controlling mechanism in dissolution of many
common rock-forming silicates, including forsteritic olivine. Protective surface
layers were not formed (Pokrovsky and Schott, 2000), and Mg ions were carried
away in solution (Rosso and Rimstidt, 200; Liu et al., 2006). This is consistent
with experimentally determined product/reactant molar volume ratios of <1,

which prevent the formation of thick protective surface layers (Velbel, 1993).

Thus, in reviewing the results of studies of natural weathering textures,
textures formed experimentally, and experimental kinetics, it appears that the
debate over a rate controlling mechanism for silicate weathering has been
resolved in favor of interface reaction rates. However, there is still much that can
be learned about the patterns of etch pits and solution channels themselves,
even with optical microscopy. This is particularly true in the case of serpentinized
peridotites or dunites, where networks of serpentine veins and serpentine layers
between olivine grains have likely provided channels for penetration of ﬂuids. In
such cases, the micromorphology of etch pits and solution channels can provide

insights into the relationship between dissolution and serpentinization.

36

Part II. The Study Area

Chapter 6
Geologic and Geographic Setting

Throughout the Blue Ridge, the Piedmont, New England and the northern
Appalachians, from northern Alabama to Newfoundland — a distance of 3000
kilometers - there are hundreds of bodies of ultramaﬁc rocks (Figure 1).
According to Cronin (1983), there are 170 such ultramaﬁc bodies in the North
Carolina Blue Ridge alone. These are distributed more or less evenly in
elongated clusters trending southwest-northeast (Figure 2). The rocks in these
bodies include peridotites, pyroxenites, gabbros, amphibolites, and diorites
(Miller, 1951). The Webster-Addie ultramaﬁc body, (also referred to as the
Webster-Addie ultramaﬁc "ring" or "complex") is one of the largest of the

southern Appalachian ultramaﬁc bodies.

The study area is in far western North Carolina, a few tens of kilometers
south of Great Smoky Mountains National Park (Figure 3). The nearest cities are
Asheville, North Carolina, 75 kilometers to the northeast and Knoxville,
Tennessee, 95 kilometers to the northwest. The town of Sylva lies adjacent to the
study area on the west. The entire area of outcrops of the Webster-Addie
ultramaﬁc body, as well as the subsurface extent of the body inferred from

outcrops of secondary minerals (Pratt and Lewis, 1905; Cronin, 1983), falls within

37

 

. l f Webster-Addie ultramaﬁc body

a. -' f y
o o . [I 1" 040
o I/ " o
/
o o ’ o_,,.--' 4"
. ’ P” 9170'
05:01,
00 o 500 1000

lllllllllll
kilometers

0 Ultramaﬁc rocks (schematic)

Figure 1. Schematic map of ultramaﬁc bodies in eastern North
America. Open circles represent general locations of one or more
ultramaﬁc bodies. Modiﬁed from Miller (1951 ).

38

  
 
 
 
   
    

Webster-Addie
ultramaﬁc body

84° _

36°+ 7:21}: '

  

50
I I

 

100
I

kilometers

Figure 2. Ultramaﬁc bodies in the North Carolina Blue Ridge.
Modiﬁed from Larrabee (1966) and Cronin (1983).

39

 

 

’ a
. I
Knoxville Great Smoky Mountains ” | ’
National Park I ’
I ‘-I
l
Temmooeee "_.,_...;«.:_-I No C.
..".. ’ ....
- --.,_,.-' I". '- Asheville
a . .- I
: ’ﬂih' .: -_. H.-
5., ... .-
Io'ouu... . .- .. ...-l...
' . Sylva Webster-Addie
' ultramafic body
a _ ’
’ Franklin I "V - - "'
a r ,
I——l-—--—----T’
Greenville
’ I
,2 / 2, .
lei eemgie , s3, (9:,
s
\

510 1 ('30 ‘

r-O
I
|

 

kilometers \

 

 

 

Figure 3. Location of the Webster-Addie ultramaﬁc body in relation to
state boundaries, nearby cities and towns, and Great Smoky Mountains
National Park.

40

the southern half of the Sylva North USGS topographic quadrangle map and the

northern half of the Sylva South quadrangle.

The topography of the Webster-Addie-Sylva area is dominated by a series
of ridges trending southwest-to-northeast, more or less parallel to the trend of the
Blue Ridge mountains and the Appalachian chain in general. Areas along the
main rivers and creeks lie at about 625 to 650 meters above sea level, while the
higher ridge-tops reach elevations of 1200 to 1250 meters, so overall relative
relief in the area is about 600 meters. The ridges have been dissected by a
dense network of ﬁrst- and second- order streams, giving the area a rugged
surface. Slopes of more than 40 percent (calculated from USGS topographic
maps) are common along the ﬂanks of the ridges.

Annual precipitation at Sylva averages more than 1300 mm. (www.city-
data.com), supporting a dense forest cover. The largest stream is the
Tuckaseegee River, a tributary of the Little Tennessee. Flowing westward, the
Tuckaseegee crosses the Webster-Addie ultramaﬁc body at two places near the
hamlet of Webster.

The Webster-Addie body is a domed sheet of ultramaﬁc rock, with the top
of the dome removed. The ultramaﬁc rocks are enclosed by gneisses of the
Tallulah Falls formation, for which the protoliths were Pre-Cambrian sediments.
(Misra and Keller, 1978; Quinn, 1991). Rocks of the Tallulah Falls formation
elsewhere in North Carolina and in Georgia have been dated at 1.0 to 1.3 Ga

from detrital zircon data (Bream et al., 2004).

41

The ultramaﬁc body crops out in the form of a discontinuous ellipse, which
passes through the hamlets of Webster and Addie, after which the body is
named (Figure 4). The long diameter of the ellipse is almost 10 km, and the
short diameter about 6 kilometers. The northeastern half of the body lies in the
rugged terrain of the western edge of the heavily wooded Great Balsam
Mountains, while the southwestern half extends through an area of lower relief
with more roads, buildings, and cleared land. (Figure 5).

The maximum breadth of the outcrop area, reached at the southwestern
and northeastern extremes (at Webster and Addie, respectively), is
approximately 500 meters (Lewis, 1896; Miller, 1953). This broadening of the
outcrop area at the ends of the ellipse suggests that the body was emplaced as a
phacolith, possibly in a zone of weakness in the foliation of the country rock
(Pratt and Lewis, 1905, Madison, 1968). The broad exposure at Addie pinches
out toward the south, breaking into several small lenses at the eastern edge of
the outcrop area (Ryan et al., 2005), and then becoming continuous again toward
Webster. For a distance of approximately three kilometers on the northwestern
edge of the ellipse (east of Sylva), the outcrop area disappears completely, but
contiguity of the body is inferred from thin outcrops of talc schist (Pratt and Lewis,
1905; Cronin, 1983).

The ultramaﬁc rocks consist mainly of dunite, with lenses of harzburgite
and pyroxenites (Madison, 1968; Ryan et al., 2005). Accessory chromite occurs
throughout the dunite. A large lens-shaped mass of websterite - a diopside—

bronzite pyroxenite - approximately 750 meters long and 125 meters wide at its

42

83° 15’

83° 10' 83° 7’ 30"
l

 

35’ 25"

35° 20’-

 

    

N

Sample sites

1 l I I

kilometers

 

 

Ultramaﬁc rocks . . .
- (Webster-Addie ultramaﬁc body) Mass've amph'b°"te

M Ionite zone
of Otto formation I y

tff Tallulah Falls formation Broken lines indicate
gs Great Smoky group inferred locations.

Figure 4. Highly generalized geologic map of the Webster-Addie-
Sylva area, North Carolina. Modiﬁed from Quinn (1991). For detailed
information on sample sites, see Chapter 8.

43

. .-.w ....
infra‘u

.Ew “(mun-...: ...wﬂ dull

 

 

3‘7» 3.4

. . o
-J O.

 

Figure 5. Oblique aerial image of Webster-Addie-Sylva area, viewed from
south. Outcrop area of Webster-Addie ultramafic body is outlined in white.
Dashed lines indicate inferred extent of body. Modified from Miller (1951),
Cronin (1983), Quinn (1991), and Ryan et al. (2005). Digital terrain model,
enhanced by sunlight algorithm: Google Earth. Land cover data: U.S. Farm

Service Agency.

 

WebSIB

tau

3 Cfc
Ilenti

stair

sh

 

widest point (Madison, 1968), crops out in the middle of the ultramaﬁc body near

Webster. This is the type locality for websterite.

Although most ultramaﬁc bodies in the southern Appalachians show only a
faint compositional banding apparent mainly in the alignment of chromite grains,
the Webster-Addie body is an exception, being marked by a conspicuous
laminated appearance on weathered surfaces (Figure 6), caused mainly by

layers of talc and chlorite (Misra and Keller, 1978).

When weathered, dunite outcrops show a characteristic yellowish-brown
to reddish-brown color. Figure 7A illustrates a specimen of dunite from an
abandoned quarry south of Addie. The color is yellowish-brown with rust-colored
patches. Rubbing the weathered surface of this rock with the ﬁnger abrades
sand-sized grains of olivine. The dark spots visible at the top and right are green
olivine grains exposed by the crumbling away of the weathered rind. Figure 7B is
a cross-section of the same specimen. The weathered rind is about one
centimeter thick, and contains dark grains of chromite and a few larger relict
grains of olivine. The darker area in the center is transparent olive-green olivine

showing ﬁne-grained foliation.

The rocks of the ultramaﬁc body are everywhere conformable with the
enclosing gneisses and amphibolites (Quinn, 1991). Olivine in the dunite is
highly magnesian, ranging between 85 and 95 mol percent forsterite, and
averaging about F090 (Madison, 1968; Cronin, 1983; Quinn, 1991). Olivine in the

harzburgite is less magnesian, in some cases as low as F073 (Madison, 1968).

45

a'

““3““?! new. ~-
.,.‘.:-,,.,_.~ ,. 5.. . , 2w .. _
to .‘ .\ “ V'ng" . '5“:

43" "" riar-r-~-'«

t. .. . t

' ' 1 vii}:

‘ F' 'i-f '1' I
. ,

1 ' I

.
U
0 i
z . r .
.1 . . . ‘
. , -
f . ' u‘,‘ ‘ .
’ ‘ ' :I u I
. r - . r
.. . n 1
a I
‘ I
. .
‘ -
I t
' , . ‘v
‘ . I
.
A ' .

 

Figure 6. Outcrop of dunite near Webster, North Carolina,

showing laminated fabric. Man’s hat (center) for scale. Photo: Pratt
and Lewis (1905).

46

 

Figure 7A. Weathered dunite specimen collected at abandoned
olivine quarry, Addie, North Carolina. Scale in centimeters.

 

Figure 7B. Same specimen as in Figure 7A, above, cut with
diamond saw and rough polished. Dark area in center is fresh
olivine. Darker spots are chromite grains. Light area around edges
is weathered rind. Note fine-grained foliation oriented from lower
left to upper right. Scale in centimeters.

47

 

There is no evidence of thermal metamorphism at the contacts between
the dunite and the country rocks. Miller (1951) noted a 15-centimeter- to one-
meter-wide zone of vermiculite where the dunite-gneiss contact is exposed,
which would have obliterated any contact metamorphic zone, had one existed.
Madison (1968) reported an alteration zone of 25 centimeters to 1.5 meters
consisting of talc, chlorite, and tremolite between the country rock and the
ultramaﬁc body. This suggests that the Webster-Addie body, like other ultramaﬁc
bodies in the Blue Ridge belt, was tectonically emplaced in a solid state at
comparatively low temperatures (Bowen and Tuttle, 1949; Misra and Keller,

1978)

Several writers have suggested that the Webster- Addie ultramaﬁc body is
either an obducted ophiolite or a mantle diapir (Madison, 1968; Greenberg, 1976;
Cronin, 1983), without favoring one origin over the other. According to Misra and
Keller (1978), no ophiolite had deﬁnitely been identiﬁed in the southern
Appalachians. However, Swanson et al. (2005) consider all alpine ultramaﬁc
bodies as ophiolites. While cautioning that the history of Blue Ridge ultramaﬁc
rocks is still unresolved, they use chromite core composition data to suggest that
the protolith of the Webster-Addie body was a mantle peridotite from a
suprasubduction zone setting. Ryan et al. (2005) ﬁnd that bulk chemical data
point to a residual mantle origin for the Webster-Addie body, and hypothesize
that it represents an ancient subduction-related accretionary block emplaced at a

convergent plate boundary.

48

 

0109

gra

gra

so

an

 

The time of emplacement is generally put before or during the Taconian
orogeny in Ordovician time (Pratt and Lewis, 1905; Madison, 1968; Misra and
Keller, 1978). At least two metamorphic events and as many as six folding
episodes have affected the rocks in the study area since the body was emplaced
(Hatcher, 1978). The dunites in the body have undergone extensive straining,
and reached temperatures high enough for recrystallization at least once. Thus,
these ultramaﬁcs, enclosed in schists and gneisses, are themselves

metamorphic rocks.

Recrystallized dunite, when relatively undisturbed, is characterized by
equant grains in close-packed polygonal texture with ~120° triple junctions at
grain boundaries (Figure 8). In the Webster-Addie body, undeforrned olivine
grains range in size from 0.1 to 1 millimeter. However, tectonic events have
imparted a noticeable foliation to areas within the dunite, which is often apparent
in thin section. In Figure 9, crushed olivine grains have become elongated and
fragmented, and there is a preferred orientation of grains. The crushing of olivine
grains has also promoted fractures to allow ﬂuid circulation and the formation of
hydrous phases (Swanson, 2001), as evidenced by large talc clots between

some of the olivine grains.

Serpentinization has affected almost all peridotites in the body, and varies
from a few percent to almost 50 percent modal abundance at the thin section
scale (Madison, 1968). Typically, thin (microscopic) serpentine layers separate
and enclose most olivine grains, and occupy many intra-grain fractures (Figure

10). A specimen cut from a relatively thick layer of serpentine is shown in

49

 

.“ _ . A . ' “ ‘ \. ‘
‘wa’d' ’. ' T’ I
"A ll, .. .

S , . ._ ‘
. 4‘.~.' T
f
,.
t
I
V

I
N

500 m

 

1;
. $
. .

Figure 8. SEM image of fractured surface of a grain of dunite
showing granular texture of olivine and uniform grain size. Sample
Sylva5, SEI, 40x.

50

 

. ﬁt- xxx-.... . .
Figure 9. Foliaton in dunite, Addie, North Carolina. Distended and
fractured olivine grains are oriented from lower left to upper right. Individual
grains are outlined in white based on interference colors. IrregularIy-shaped
areas between grains are mainly talc (tlc). Serpentine fills intra—grain fractures
and forms thin layers between grains (mostly covered by white outlines).

Sample Sleec, XPL, 40x

51

 

 

a t ‘ ,7... 1.3. g. . . _
. 7. . _..' r’,_ r' _ ": . ‘ y \ ‘ " -. . _ . "_. - ‘4 ' ‘r
. Y ’ ... 4 . . ~ ~ . _ . o < .‘ .
'-'~;.:- .-r-.~ ~. . -. .t‘ . u _. -
‘ .. .‘ J 0" fr _. , . . _ \ ~ ‘ , _ . .7 _.
..- r * . '. . - i , ., , I'd y,

- r6

Figure 10. SEM image of part of a polished thin section showing olivine
grains (light gray) separated by serpentine layers (darker gray), which also ﬁll
many infra-mineral fractures. More massive darker gray areas are talc.
Sample Sylva6, BEI, 100x.

 

_ ...
l

52

Figures 11A and 11B. Bowen and Tuttle (1949) postulated that serpentinization
of a dunite mass undergoing strong thrusting is likely to take place
simultaneously with the solid-state emplacement of the dunite, and Madison
(1968) concludes, based on geologic evidence, that this is the most probable

origin of most of the serpentine in the Webster-Addie body.

Talc is also a common secondary mineral, and, often together with
anthophyllite, forms “soapstone” shear planes within the dunite and between the
ultramaﬁc body and the country rocks (Figure 12). Several authors (Pratt and
Lewis, 1905; Madison, 1968; Cronin, 1983) have relied on the presence of talc
ﬂoat at the land surface to infer the location of the dunite-gneiss contact, and
Pratt and Lewis (1905) mapped the inferred continuation of the ultramaﬁc body
on the western side of the ellipse, where no surface outcrop exists, as “talc
schist” . Petrographic evidence has shown that talc formed after serpentine in the
Webster-Addie body, and independently of serpentinization (Madison, 1968).
Distended talc clots can be seen in thin sections, intermixed with crushed or
elongated olivine grains, but at the same time bordering directly on areas of
undisturbed olivine (Figure 13).

Together with talc and anthophyllite, a number of other secondary hydrous
minerals are found in or around the dunite and harzburgite. The commonest of
these are chlorite, tremolite, and vermiculite. These minerals are thought to have
formed mainly during the Acadian orogeny at about 350 Ma . At that time,
granitoid intrusions were emplaced throughout the eastern Blue Ridge (Butler,

1 972; Hatcher, 1987). In the Webster-Addie body as elsewhere, dikes and

53

 

Figure 11A. Weathered dunite rind (top) on serpentine
(bottom). Specimen is resting on surface of cut, shown in
Figure 11 B, below. Sample Sylva7. Scale in centimeters.

. Xi'an. ~“t .‘

" "‘

 

Figure 11 B. Same specimen as above, tilted to show
surface of out. Top 1/4 of surface is dunite, lower 3/4 is
serpentine. Scale in centimeters.

 

Figure 12. “Soapstone” coating on dunite. Light areas are blue-
green talc-anthophyllite, dark material is reddish-brown dunite.
Abandoned olivine quarry, Addie, North Carolina. Coin is 2.4 cm
in diameter. Photo: Warner and Yurkovich (2005).

55

gin
n-...v n'm-& 'l‘ﬂ'.‘

Figure 13. Dunite composed of polygonal olivine grains (upper two-thirds of
image) and sheared talc clots with crushed and elongated olivine grains (lower
third of image). Thin section of sample from road cut adjacent to abandoned
olvine quarry, Addie, North Carolina. XPL, macrophotograph.

 

56

pegmatite bodies
which reacted wit
phases (Swanso:
Ross et at
tittieWebster-F
“T

soil is da

a friable

teet thic

that let:
merges
Surface

tar tntc
Contrc

and ti

Clom

(fresh
QUlnn
COlor,

appare

pegmatite bodies associated with these intrusions provided metasomatic ﬂuids
which reacted with the ultramafics to form a variety of hydrous metamorphic
phases (Swanson, 2001; Ryan et al., 2005).
Ross et al. (1928) gave the following description of the weathered dunite
in the Webster-Addie ultramaﬁc body :
“The mantle of weathered material is not usually thick. The surface
soil is dark reddish brown and supports a meager vegetation. Below this is
a friable dark-red residual material that is commonly not more than 8 to 12
feet thick this grades downward into a light yellow porous material
that retains much of the structure of dunite, and this in turn gradually
merges into fresh dunite, or partially serpentinized dunite. The upper
surface of the fresh dunite is irregular, and rounded masses locally project
far into the weathered material. The rapidity of weathering seems to be
controlled largely by the ease with which water can penetrate the mass,

and the more porous parts of the dunite weather fastest.”

Cronin (1983) reported that, in exposures in the Webster-Addie body,

"ultramaﬁc rocks exhibit a dark brown weathered to light green
(fresh) appearance, depending on the degree of weathering". According to
Quinn (1991), the dunite in the body "weathers to a deep metallic brown
color. Weathered surfaces highlight lamination in dunite that is less

apparent in fresh exposures" and further, "Dunite is commonly broken

57

 

 

along lamin

|
setpentine.‘

along lamination-parallel joint surfaces that are covered with talc and

serpentine."

58

Chapter 7
Previous Studies of the Webster-Addie Ultramaﬁc Body

The ultramaﬁc bodies of western North Carolina were known as early as
the mid-nineteenth century, and were described by Shepard (1872), Kerr (1875),
and Genth (1881). Julien (1882) investigated the "dunyte-beds of North Carolina"
and interpreted them as "a mechanical accumulation in the form of ancient olivine
sand." He listed four common modes of "indigenous alteration" (chalcedonic,
homblendic, talcose, and ophiolitic) and described the ophiolitic mode as "a
common process, well shown near Bakersville, Webster, etc." in which "the
olivine has suffered alteration in exactly the same way as that in the chrysolitic

lavas of numerous foreign localities."

The Webster-Addie ultramaﬁc body has been studied by several
researchers, beginning with Lewis (1896), who was the ﬁrst to describe its ring-
like shape. Lewis’ map remains the most detailed geologic map of most of the
Webster-Addie body. Lewis' report also included a map of the Appalachian
crystalline belt from Alabama to Maine which showed the distribution of

peridotites and corundum deposits.

Pratt and Lewis (1905), who expanded on Lewis' (1896) report, included
the Webster-Addie body in a broader (440-page) study of peridotites in western
North Carolina, which also contained 60 photomicrographs of representative thin

sections, and added lines to Lewis' 1896 map showing directions of foliation in

59

the enclosing gneiss. They described the Webster locality as "the largest
peridotite area in the state, and in many respects the most remarkable one in the

Appalachian region".

Ross et al. (1928) studied the relationship between the occurrence of
nickel and the weathering of dunite near Webster, and determined that "nickel as
it was set free from the weathering dunite has been fixed in the vermiculite

bearing veins by a process of base exchange." (Ross et al., 1928).

Hunter (1941), Hunter et al. (1942), and Murdock and Hunter (1946)
published reports on economic minerals of the Webster-Addie area (forsterite
olivine, chromite, and vermiculite, respectively) as parts of statewide mineral

bulletins.

Gwinn (1950) studied the structure and petrology of the Webster-Addie
body and interpreted it as a ring-dike. At about the same time, Miller (1951)
mapped the outcrop in detail and described its component minerals,
concentrating on the alteration of chromite. In the main, Miller veriﬁed Lewis'
(1896) map. He concluded that the body originated as a sheet-like intrusion,
which was later domed by regional deformation. Johnson (1958) carried out a
ground-based magnetometer survey in an attempt to conﬁrm the extension of the
body underground where there were no surface outcrops. As evident from the
map accompanying the report, there was no discernible correlation between the
magnetometer results and either the surface exposures or the inferred lineament

of the ultramaﬁc body.

60

Disagreements over the origin and chronology of the Webster-Addie “ring”
prompted Madison (1968) to carry out the most complete study of the petrology
of the body to that time. Madison’s report included data on mineral composition
of 50 samples from four sites. Two of the sites were at the same locations as
those from which the samples used in the present study were collected. Madison
(1968) also used paleomagnetic data (based on remanent magnetization
measured in 12 oriented samples) to refute those who interpreted the body as a
ring-dike, and to support the view that it was a sheet-like mass that was
deformed after emplacement. Though numerous, Madison’s samples were
collected from short traverses — two at sites south of Addie, one on the southwest
side of the outcrop area at Webster, and a fourth traverse consisting of only four
samples at Cane Creek on the southeast side of the outcrop area. Condie and
Madison (1969) used the same sample data to conclude that the extensive

serpentinization of the Webster-Addie body was a post-emplacement process.

Greenberg (1976) investigated petrofabrics from "several samples of
texturally different dunite and harzburgite". He concluded that the results "support
the pre- to syn—deformational diapiric intrusion of a peridotitic mass", followed by

"events which cross-folded it into a domal conﬁguration." (Greenberg, 1976).

Cronin (1983) sought to understand the petrogenetic processes leading to
the formation of the complex. Based on the analysis of 165 samples, he
concluded that it might “represent the basal portion of an obducted ophiolite

sequence or an intruded ultramaﬁc diapir”. Cronin’s sampling was limited to the

61

same

com;

Addi
the l
recs
ma;
we:
25f
Ad
Th

35

Uli

Wc

same areas as Madison’s had been, and his ﬁnished report included modal

composition data for only 15 samples.

Until the 1990’s, geologic mapping of the local area including the Webster-
Addie ultramaﬁc complex (the area surrounding the town of Sylva, and including
the hamlets of Webster and Addie) had been completed only at the
reconnaissance level. Quinn (1991) carried out detailed 1:24,000-scale geologic
mapping of the area in connection with mapping the boundary between the
western and eastern Blue Ridge. Quinn’s map covered an area of approximately
250 square kilometers and, while agreeing with previous maps of the Webster-
Addie ultramaﬁc body, provided new information on the surrounding formations.
This information has been reﬂected in all subsequent geologic maps of the area,

as well as in geologic maps of the southern Blue Ridge in general.

Recently, Ryan et al. (2005) worked intensively in the Chestnut Gap area
on the eastern edge of the Webster-Addie body and mapped an area
approximately 1.5 kilometers (E-W) by 3 kilometers (N-S). They found that the
ultramaﬁc rocks in this area are more dismembered than recognized by previous

workers, and that most of the dunite has been altered to serpentine.

62

Part III. Materials and Methods

Chapter 8
The Sample Suite

Forty-ﬁve hand samples of dunites, or rocks that appeared to be dunites,
were collected in 2005 at four sites near Addie, North Carolina, by Dr. Michael
Velbel of Michigan State University. Two of the sites were adjacent, inactive
dunite quarries approximately one kilometer south of Addie, North Carolina. Two
sample series were collected at this location. As indicated in Figure 14, the
“CGSZ-3” samples were collected at the northern quarry site, and the “Sleec”
samples at the southern site. The remaining two sites were at road cuts on
opposite sides of the T-intersection of US. Highway 23, also known as the Great
Smoky Mountains Expressway, and State Route 1709, also known at this point
as Blanton Branch Road. The “Syl1709” samples were collected on the north
side of the Highway 23, and the “Sylva” series on the southeast corner of the

intersection.

63

 

  
 
 
 
  
      
    
   

. Soon?

) U 3 ,1
.I
,
. o
.. - a”
-' II.
. _ , -. a “5&5“ . > ._
j ' \. ,f:'0 "' "‘l *‘
f’ I I “ ' 3-....54 . o
’0 '..
' l
u ' , c
U
. .‘ . _ ' I .
._ . . _ ._>‘L,v . /
t x ‘ ‘ \. ‘00 /
’ . t/

 
 

. \ «5";‘78'69 ‘\
~ l'-,t é " ‘ ‘

( \IWﬂNi’I/Y‘t/azm

~Syl1709 series; t ‘

-‘ ‘ \'§r//” (If?

-" I /.H:\. -\ —» “'1 1‘)
Z‘Sylva series

\

\“\ *QCUL mu ‘ _\.: . .. _
s .
\ sex

(

 
 
 
     

a \\_ :>’~,__4/'
.- “ --. ,-

  
    
  
 
 

      
    
  

“ a (“nd .-. \
. . ﬂ ‘ . . , \ ._/%:’) \
\ \ - \.\ g2 , \
» '0' \«J l@{/¢‘\Kf\ L3 ‘(w \:\{ If“ \f
,. I i: \\\\M ’ "'ﬁ.‘ 500 m ‘5
"Inf/W .x‘f?’ .1 . M/ & 3 C\ A . \ \ Nun l pey2v<¥
Figure 14. Section of Sylva North USGS 75’ quadrangle showing
collection sites for samples used in this study. Dashed outline
indicates extent of ultramaﬁc outcrop area. After Quinn (1991) and
Ryan et al. (2005).

 

       
 

 

Chapter 9
Analytical Methods

Sample Preparation.

Two or more microscope thin sections were prepared from each hand
sample — one covered thin section for optical microscopy and one uncovered and

polished thin section for examination with a scanning electron microscope (SEM).

Thin sections were made from hand samples from the four sites, as

follows:

CGSZ-3 series 20 thin sections
Sleec series 8 “ “
Syl1709 series 6 “ “

Sylva series 14

In addition to the thin sections, grain mounts were prepared from a

selected subset of hand samples.

X-Ray Fluorescence Analysis and ICP-MS.

Six hand samples, most of which showed well-deﬁned fresh cores and
weathered rinds, were selected for X-ray ﬂuorescence (XRF) analysis. Individual
samples were subdivided by visual inspection into “fresh” and “most weathered”
sections, and, on some samples, sections with intermediate degrees of

weathering were identiﬁed. The terms “fresh”, “weathered”, and “most

65

weathered” refer only to relative degrees of weathering within each individual
sample, and are based on the subjective judgment of the analyst (Losiak,
2008a). Likewise, “core” and “rind” refer, respectively, to the least weathered and
most weathered parts of the hand sample, and do not necessarily indicate the

ﬁeld position of the section analyzed.

Preparation followed the conventional method of pulverizing the
fragments, adding ﬂux to the powder and fusing it into discs, and analyzing the
disks using an XRF apparatus and an lCP-MS. Core, rind, and intermediate
samples were analyzed for major, minor and trace element composition. The

resulting sample set consisted of the following:

CGS 2-3#1 8 samples (2 suites of 4 samples each, fresh through
weathered)

CGS 2-3#2 5 samples(2 fresh, 2 weathered, 1 composite)

Sleec 5 samples (2 core, 2 intermediate, 1 rind)
Syl1709 4 samples (2 pairs of 2 samples each)
SylvaS 2 samples (core and rind)

Sylva6 1 sample

Sylva7 1 sample

Optical Microscopy.

Because etch features are too small to be seen by the unaided eye or with
a hand lens, the transmitted-light optical microscope is not only a necessary tool,

but also offers the choice of plane-polarized or cross-polarized light, a range of

66

diaphragm openings to control visual relief in the image, and sequential three-

dimensional viewing through the 30-micron thin section by varying the focus.

Using a Pentax e-110 digital camera with 6—megapixel resolution,
photomicrographs were made of etch features in the thin sections. Lower
magniﬁcation images (objective lenses 2x, 4x, and 10x) were ﬁrst made, and
were used to eliminate from further consideration those thin sections that
contained substantial amounts of pyroxenes, talc, and amphiboles. Also
eliminated were those thin sections containing most metamorphic minerals from

the gneisses and schists enclosing the ultramaﬁc body.

The remaining eighteen thin sections, containing mostly olivine, secondary
serpentine, and accessory chromite, were then photographed in greater detail. At
least three strips of sequential, overlapping photomicrographs were made of
each thin section, using both plane-polarized and cross-polarized light. Next,
these images were inspected to locate examples of olivine dissolution (etch)
features, and representative images were acquired at higher magniﬁcations
(objective lenses 20x and 40x). In many cases, photomicrograph series were
made using sequential focusing, and digital mosaics were compiled from these

images to provide more extensive views of etch features.

Morphological information derived from these images includes the form,
number, and spatial patterns of etch features, as well as the spatial and inferred

genetic relationship between olivine etch features, serpentine layers, and iron-

67

oxide mineral occurrences. In all, approximately 8,000 photomicrographs were

produced using optical microscopy and conventional digital photography.

Scanning Electron Microscopy (SEM) of Thin Sections.

The beneﬁts of SEM, compared to optical microscopy, are higher
magnifying power, higher resolution, and the ability to discriminate between
minerals based on the atomic numbers of component elements, which result in
differences in brightness levels. Disadvantages are the limited time an individual
researcher can use the microscope, lack of color imaging, and restriction of

imaging capability to only the outer or upper surface of a specimen.

Over an eleven-month period, small portions of eighteen thin sections
were examined and photographed at the Michigan State University Center for
Advanced Microscopy, using a JEOL 6400 scanning electron microscope. The
highly-polished thin sections were prepared for SEM imaging by coating them
with carbon on the exposed side to insure uniform, continuous electrical

conductivity.

Surface topography was imaged using secondary electron microscopy,
and a ﬁrst approximation of compositional differences was obtained using
backscattered electron (BSE) imaging, which also provided higher contrast

between etch features and olivine. In total, 1100 images were made.

68

The geometry and spatial ordering of groups of micron-scale features, as

well as numerous larger-scale features, were observed, recorded, and classiﬁed .

Scanning Electron Microscopy of Grain Mounts.

One advantage of grain-mount imaging, compared to thin-section imaging,
is that the exposed surfaces have much greater relief, as do the etch features
found on these surfaces. Another advantage is that grain-mount preparation
frequently splits the sample along boundaries between olivine grains and
serpentine layers, usually stripping off the serpentine and exposing etch features
in the olivine. A disadvantage is that the irregular surface of fractured grain
mounts reﬂects BSE signals in all directions, and thus BSE imaging is not usable

with grain mounts.

More than 300 images were made of etch features on mineral grains
taken from eight hand specimens. The eight grains were glued to aluminum
cylinders and coated with gold for conductivity. All grain-mount images were
made using secondary electron imaging. As with the thin-section SEM images,
the grain-mount images were printed on paper and recorded on compact digital

disks.

69

Part IV. Results

Chapter 10
Selected Results Of Composition Analysis (XRF)

XRF analysis provided detailed data on the major and minor elements in
the samples tested. Results for two parameters: weight percent MgO and
Magnesium number, are presented below. Complete results of the analysis are

included in the Appendix.

Weight Percent M90 in “Fresh” and “Most Weathered” Samples

As indicated in Chapter 9, “Analytical Methods”, the terms “fresh” and
“most weathered” refer to the visually determined extremes in a given sample.
Because no quantitative data were produced, there is no rational basis for
assigning numerical values to the degree of weathering. Therefore, in Figures 15
and 16, the minima and maxima on the horizontal axes are schematically
designated as “Fresh” and “Most Weathered”, providing a conceptual framework

for visualizing the variation of composition with degree of weathering.

Because there is no basis for scaling the degree of weathering, and
because there was no consistency in the number of intermediate samples
analyzed, samples representing intermediate degrees of weathering are not

shown on Figure 15 and 16. To provide consistency in the graphic presentation,

70

only data from the extremes - the freshest and the most weathered samples -
are included. However, in each of the four suites that included intermediate
samples, weight percent MgO decreased consistently from the fresh to the most

weathered samples.

Figure 15 illustrates changes in weight percent MgO by relative degree of
weathering for the ﬁrst nine suites and pairs listed. Weight percent MgO
averaged 43.92 percent for the eight fresh samples analyzed (both Syl1709 pairs
used the same fresh sample), ranging from 42.24 percent to 47.95 percent (a
range of 7.51), with a standard deviation of 1.77. However, Sylva5, with a weight
percent MgO of 47.95, a much higher percentage than that of the second highest
sample, was a distinct outlier in this data set. Vtﬁthout Sylva5, the range of the
remaining seven samples was only 2.36 (42.24 to 44.60), and the standard

deviation was only 0.75.

As illustrated in Figure 15, all of the fresh/weathered pairs, with the
exception of Sylva5, showed a decrease in weight percent MgO from fresh to
weathered, as had been found in previous studies (Hotz, 1964, Trescases, 1975,
Golightly, 1979). The mean decrease between fresh and weathered samples was

1.8 percentage points, excluding Sylva5 from the calculation.

The greatest decreases were in the two Syl1709 sample pairs, 4.86 and
5.6 percentage points. It should be noted, however, that these were both based
on a comparison with the same fresh sample. The relatively large decreases can

be explained by the high percentages of CaO in the two weathered Syl1709

71

Figure 15. Change in Weight Percent MgO
by Relative Degree of Weathering

 

 

 

 

 

49

 

47

 

 

Sample
.- .— ... SylvaS
====== SleecA
45 -------- SleecB

“ .. . . .- CGSZ-SalHA
_— CGSZ-3#1B
. - .. .. .. CGSZ-S#2C&E
....... CGSZ-3#2A&B
----.. Syl1709A
—- -— Syl1709B

 

 

 

 

Weight % MgO

 

 

 

 

 

 

 

 

 

 

 

 

 

 

37 i
Fresh Most

Weathered

 

 

 

 

72

 

 

samples — 1.49 percent for Syl1709A (D) and 2.16 percent for Syl1709B(C).
These were the highest levels of CaO for any of the samples, fresh or weathered,
and suggest the presence of clinopyroxene. If this were the case, the relatively
higher persistence of pyroxenes compared to olivine could result in marginally
higher levels of Ca, and proportionally lower levels of Mg, in the weathered

samples.

The two Sleec results were calculated using the same weathered
sample. This is apparent from the convergence of lines for the respective sample

pairs in Figure 15.

In the case of Sylva5, which showed a very slight increase (0.4
percentage points) in percent weight MgO from the fresh core to the weathered
rind, two factors should be taken into account. First, the loss in ignition reported
for the core sample was 1.23 percent, compared to only 0.27 percent for the rind.

This difference of almost one percentage point was enough to account for the
relative increase (based on a smaller divisor) in M90, SiOz, and Fe203, all of
which increased slightly between the two samples. Of the major element oxides

reported, only Al203 decreased (from 0.32 to 0.30 percent), while TiOz, MnO,

and CaO were unchanged. Neither Na20 and K20 were detected in either

sample. Secondly, Losiak (2008a) noted that even the freshest part of the Sylva5
hand sample looked weathered. Subsequently, etch pit clusters in a thin section
from the core of this hand sample were mapped in detail (See Chapter 11), and

were found to be concentrated in a roughly circular area occupying the center of

73

the thin section, with very few etch pits outside this area. This suggests that the
fracture network characterizing the center of the core of this hand sample
provided greater access to corroding ﬂuids, which carried off more than the

expected amount of Mg in solution.

Magnesium Number in “Fresh” and “Most Weathered” samples.

Magnesium number, calculated as: mol Mg/(mol Mg + mol Fe), varied in a
similar way as did weight percent MgO. Magnesium numbers for the fresh
samples ranged from 0.942 to 0.952, a difference of only 0.010. The mean
magnesium number was 0.947, and the standard deviation was 0.004. Although
Sylva5 had the highest magnesium number, it was not extremely high compared

to the other samples.

As shown in Figure 16, all sampled pairs showed a decline in magnesium
number from fresh to weathered. The average decrease was .0067. As in the
case of the results for percent weight MgO, the greatest decreases were in the
Syl1709 sample pairs, owing probably to the presence of pyroxenes, as
discussed above. SylvaS showed only a miniscule decrease of .001 between

core and rind, also due to reasons discussed above.

74

Figure 16. Change in Magnesium Number
by Relative Degree of Weathering

 

Magnesium Number (mol Mg/ [mol Mg + mol Fe])

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.954
0952 - ... - -—'-—-"-"'""=T'_
\:
0'95 .' Sample
\K‘ . ° - - - SY'V85
0348 == SleecA
-------- SylFlecB
0.946
— - . -— CGSZ-3#1A
— CGSZ-SMB
0.944
- - - - - CGSZ-SIIFZC8tE
0.942 - ------- CGSZ-3#2A&B
----- Syl1709A
0.94 -— -— Syl17098
0.938
0.935
0.934 '
0.932 as”--- -—-- ~ --- ~ . .._.__.
Fresh - Mosr
Weathered

 

 

 

 

 

 

 

 

75

 

7‘!

u,

(1)

IT!

Chapter 11
Features Visible at Optical and SEM Scales.

Form of Individual Etch Pits.

The most commonly occurring etch pit shape is that of a diamond or
lozenge. Diamond-shaped etch pits appear to be formed as dissolution proceeds
along and outward from a line of weakness in the olivine, such as an edge or
screw dislocation. Such pits may be either compact (Figure 17) or elongated
(Figure 18), and often distorted to appear somewhat sinuous (Figure 19). On an
exposed surface, elongated pits usually tend to taper to narrow points at one or
both ends (Figure 20). In many cases, elongated pits with serrated edges can be
seen to be composites of coalescing diamond-shaped pits (Figure 21). Where
the walls of serrated edges are visible, they may show conical indentations,
indicating that the diamond shapes apparent at the surface are cross sections of

pairs of cones (Figures 22A and 22B).

When viewed from the side through transparent olivine with the optical
microscope, individual pits often appear at ﬁrst glance to be pyramidal in three
dimensions (Figure 23A and 23B). They seem to be faceted and pointed, like
pyramids. However, on close inspection there is usually only one facet edge
visible, where there should be three or four. As observed by Velbel (2009),
diamond-shaped etch pits could also be cross-sections of cones joined base-to-
base. Pits formed in either way would be manifested as diamond-shaped in plan

view. Pyramidal pits would, of course, be diamond-shaped in cross section, as

76

 

 

y

20 um

 

Figure 17. Awell-developed ﬁeld of diamond-shaped etch pits. Sample Sylva5C,
PPL, 400x.

77

”a 7' ‘ ‘1‘ '2! '4: . ‘
g . \ -d y. l
. . 7 0%. _ 7
~ ' 'w’ ' -‘ . 0' w I’
. “ﬁn." 3- ' i I I
g \

.4'

 

Figure 18. A ﬁeld of elongated diamond-shaped etch pits. Sample Syl1709,
PPL, digital mosaic, 400x.

78

syf'v

3‘ Mara” .z‘u ' '\ 4*
Figure 19. Afield of sinuous etch pits (left). Sample Syl1709, PPL, digital
mosaic, 400x.

79

 

I

 

 

   

,_,’ I a,"
v s. . ”g!" _ .t, 4*. Y.
Y' '1 - :h‘ J H u
«3‘ Ni" wt “‘1‘ ' ‘ l.- ” 1 m
”a: . a ' I W“ . u

 

Figure 20. High-magnification SEM image of elongated diamond-shaped etch
pits, tapering to narrow points and aligned in parallel. Sample Sylva7, fractured
grain, SEI, 25,000x

80

 

Figure 21. Linear etch pits with serrated edges, apparently formed by
coalescing diamond-shaped pits. Geometry of diamond-shaped pits is unclear
because pit bottoms and walls are largely obscured by secondary minerals.
Sample Sylva5U, polished thin section, SEI, 2500x.

81

FIgUre
pairs 01
Figure ;

FW62;
abOVe. 3.
Lines ittdii
lonngdm

l0 each 0”
JW’ielltatl

  

 

 

A.1‘£.‘l‘

Figure 22A. Linear etch pits with serrated edges, composed of coalescing
pairs of base- to- base cones exposed as coalescing diamond- -shaped outlines
at the surface of the thin section. See schematic diagrams C, D, and E in
Figure 228, below. Sample SylvaSC, PPL, 400x.

 

 

 

 

 

Figure 22B. Schematic diagrams of linear etch pits C, D, and E in Figure 22A,
above. Solid lines indicate edges of paired cones joined base-to-base, broken
lines indicate inferred join of cone bases. Cone axes are aligned with deeper
longitudinal channels in pit centers, which likely mark fractures or dislocations.
Cone bases are perpendicular to these channels, and are thus subparallel

to each other where channels are relatively straight, as in C and D, but vary
in orientation where channels are curved, as in E.

82

 

Figure 23A (above). Cross-section of a serpentine layer (S) with pyramidal or
conical etch pits (indicated by arrows) protruding from the margin of the
serpentine layer into olivine. Sample CGSZ-3#1, PPL, 400x.

Figure 23B (below). Cross-section of a serpentine layer with pyramidal or
conical etch pits protruding from the margin of the serpentine layer into olivine.
Same symbols as above. Sample SylvaSC, PPL, 400x.

 

83

illustrated in Figure 23C. Bi-conical pits, or pits formed by pairs of cones joined
base-to-base, can result in diamond-shaped pit openings as shown in Figure
23D. In such cases, the apparent “point” of the pyramid would be the edge of the
cone base, and the single facet edge would be the joined base lines of the two

cones.

Unambiguous pyramidal pits would be point-bottomed and four-faceted,
like the dent made by the point of a steel nail. No such pits were observed in this
study, but unambiguous bi-conical diamond-shaped pits were observed.

Examples are shown in Figures 23E, enlarged in Figure 23F and Figure 23G.

Velbel (2009) showed several SEM images of diamond-shaped etch pits
in polished thinsections. Diamond-shaped etch pits have also been imaged at
smaller scales in TEM studies. Eggleton (1986) showed a TEM image with
isolated diamond-shaped etch pits approximately 20 nanometers in length, and
Banﬁeld et al. (1990), also using TEM, showed olivine surfaces with sawtooth
borders made up of coalescing 10- to 15-nanometer-long partial diamond
shapes. While these examples show pits that are two to three orders of
magnitude smaller than those visible in SEM or optical microscopy, the geometry

of the pits is identical at all scales.

Rectangular etch pits, other than diamond-shaped, also occur but are
uncommon, and are typically found in the more iron-oxide rich zones of etch pit

ﬁelds (Figure 24).

 

 

 

Figures 230 and 23 D. Illustration of possible formation of diamond- shaped
etch pit outlnes by pyramidal etching (C) or by conical etching (D). Shaded area
represents etch pit opening at surface of olivine. Figure 13-6D adapted from

Velbel (2009).

 

1
O .

Figures 23 E, 23 F, and 23G. Diamond-shaped etch pits with apparent

bi-conical three-dimensional form. Note horizontal bisectors across pits in inset
F, bi-conical shape of large pit in inset G. Large pit is 25 pm in length. Scale of
insets F and and G is four times larger than scale of E. Sample SylvaSC, PPL,

100x (E), 400x (F and G)

85

 

Figure 24. A ﬁeld of diamond-shaped etch pits (right) grading into rectangular

etch pits in an iron-oxide—rich zone (left). Sample CGSZ-3#1, PPL, digital mosaic,

400x.

86

Conical, or funnel-shaped etch pits occur widely, and are most evident in
SEM images of olivine chips, which give a three-dimensional presentation
(Figures 25 and 26). They are also found, but less frequently, on SEM images of
thin sections (Figures 27 and 28). Occasionally, a well-formed conical etch pit

can be found at the optical microscope scale (Figure 29).

Velbel and Ranck (2008) and Velbel (2009) found that almost all etch pit
types are formed by joining or coalescing conical etch pits - diamond-shaped
pits, for example are composites of conical pits joined at their bases and exposed
in cross-sections out along their axes, as illustrated in Figure 23G. In the present
study, other cone-based composite etch pit types were identiﬁed. One
composite type is the “mouth-shaped” etch pit (Figures 30 and 31), that
resembles the lips drawn in cartoon characters’ faces. Where the surface of
olivine intersects the thin section surface at a shallow angle, these “mouth”
shapes become ﬂattened into low-angle ﬂuted etch pits (Figures 32 and 33).
Similar shapes are also revealed on olivine grains, where a fracture surface cuts
the grain at a shallow angle (Figure 34). Triangular etch pits can also be seen as
cross-sections of conical or wedge-shaped pits (Figures 35 and 36). Where
conical etch pits have developed from different directions, as along different
crystallographic axes, complex three-dimensional, curvilinear dissolution features

may be formed (Figures 37 and 38).

A somewhat unusual type of etch feature is that of long, straight pits,
usually found as tubes or prisms (Figure 39), often with secondary etch pits

developed along their sides (Figure 40). It is not clear whether these are actually

87

 

 

In

tical outlines at an exposed gra

l etch pits creating ellip

Ica

Figure 25. Con

Sample

ine layers at left and right.

surface. Remnants of enclosing serpent

Sylva5, SEI,

5000x.

88

 

 

 

rface,

l etch pits. Sample Sylva5, exposed grain su
89

Ica

Figure 26. Nested con

SEI, 6500x.

   

Figure 27. Conical etch pits developed obliquely to the exposed surface.
Sample Sylva6U, polished thin section, BEI, 3500x.

90

 

 

 

 

ngre 28. A ﬁeld of shallow-angle conical etch pits. Sample Sylva5U, polished
thin section, BEI, 2500x.

91

 

_ I 'V g
_ _ j at. “s 10 pm

Figure 29. A conical etch pit protruding from a serpentine layer into olivine.
This large, symmetrical pit appears to be partially ﬁlled with an alteration product.
Small etch pits, possibly also conical, protrude from both surfaces of the
serpentine layer. Sample CGSZ-3#2, PPL, 400x.

92

 

 

Figure 30. An etch pit field containing many “mouth-shaped”etch pits.
Sample SYL1709, polished thin section, BEI, 500x.

93

 

 

 

Figure 31. Compound “mouth-shaped” etch pits showing formation by
coalescing conical pits. Sample Sylva6, fractured grain surface, SEI, 5000x.

 

Figure 32. A ﬁeld of shallow-angle etch pits aligned in parallel rows. Most of
the pits appear to be conic sections with “lips”, and some obvious conical pits
are also present. Sample Sylva6U, polished thin section, BEI, 2500x.

95

   

 

    

‘ k ,.
. ' -\_I
I". " . ‘4‘ h
’. . _. , r I‘
. -. ,
. a ‘4
- .. 4 . ,
l 9
. ‘ n I
. 4 a .
.‘ I
e ‘ k
.
O
‘J
‘ r.- I ' N1
no. i

Figure 33. Parallel rows of very-shallow-angle fluted etch pits. Fluted pattern

appears to be made up of flattened conic sections, and a few triangular etch
pits (or, stated differently, triangular cross-sections through cone-shaped etch
pits) are visible at left center. Sample Sylva6U, polished thin section, SEI, 1000x.

 

 

 

Figure 34. Arrays of shallow-angle fluted etch pits on the cleanly-broken
surface of an olivine grain. A thin remnant of the enclosing serpentine is visible
at upper left. Lighter area at right is adjacent face of grain. Sample Sylva5,

8 El, 1 000x.

 

-? y 3
5‘ ' A: —. —

Figure 35. A small field of triangular etch pits, most of which appear to be
flattened conic sections. Sample Sylva6U, polished thin section, BEl, 2500x.

98

 

 

Figure- 36. A field of sharply-etched triangular pits. Sample Sylva5, exposed
surface of fractured grain, SEI, 5000x.

99

 

Figure 37. A complex shape composed of intersecting conical etch pits,
exposed through a break in the enclosing layer of serpentine. Sample Sylva5,
fractured grain surface, SEI, 2500x.

100

 

. ‘ VI. .
a . - - .v
' , a. ‘ ‘
.M t
l , ' '- ‘ - A V,
T ' ‘ \ ‘ " f '
x; 3.1- . ‘ . .' , .-
v a}. - . . - .
‘ f, ‘p 7 » 1 _ .‘ ...;
n .‘ .' ‘ . e. ~ 0

wc.

. - . a? - . ’. .. ' ‘ ' N‘
~. .‘ H in . _ ‘ I .‘ In 1 . " . e. . " . M -,
—-. W ,_' .3... .‘ ., w"~ r, ’
r ‘- .‘i 0'. im, .' - t . . . ~ .' >-‘ ‘ T .'
. . . - . .

 

.

Figure 38. A series of aligned conical etch pits, framed by remnants of the
enlosing serpentine layer. Sample Sylva5, fractured grain surface, SEI, 1500x.

101

 

*3. '
aw

'i
is

l

. _ I
I " R. an“?

_—I
20pm

Figure 39. Straight, tubular or prismatic features protruding into olivine
from the margin of a serpentine layer. As can be seen from variations in focal
clarity, these are not surface cracks, but are needle-like features within the
olivine. Note the broadening of the dark feature (in the center of the group)
where it meets the serpentine layer. Sample Sylva5C, PPL, 400x.

102

 

Figure 40. Linear etch pits with secondary formations of globular pits,
or possibly needles of a replacement mineral with globular accretions of
the same or a different mineral. Sample Sylva5C, PPL, 400x.

103

etch features or if they are needles of a replacement mineral forming along

dislocations within olivine.

Where iron-oxide mineralization is concentrated, etch pits of any shape
are usually expressed in low relief (Figures 41 and 42). Frequently, a corrugated

subsurface pattern is exposed inside such pits (Figures 43 and 44).

104

 

Figure 41. Low-relief rectilinear etch pit pattern in an iron-oxide rich zone.
Note also “filigree” or “embossed” texture at bottom center. Sample SylvaSC,
PPL, 400x.

105

u. ' V1.
5‘

" .‘b‘ “R a.» - 3.
‘ (jg-g, \‘ ’ I ’ D I.

- "h

 

Figure 42. A field of low-relief diamond-shaped and rectangular etch pits in
an iron-oxide rich zone. Sample CGSZ-3#1, PPL, digital mosaic, 400x.

106

 

Figure 43. Low-relief etch pit patterns, with corrugated subsurface structures,
between two thin serpentine-filled fractures in an iron-oxide rich zone. Entire
pitted area is stained by iron oxide minerals. Surrounding olivine at borders of

image is clear. Note “filigree” pattern at left, where staining is heaviest. Sample
SylvaSC, PPL, 400x.

107

 

#

20pm
Figure 44. “Filigree” pattern (lower right) and low-relief etch pits with

corrugated subsurface texture (upper center) in a heavily iron-oxide stained
zone. Sample Sylva5C, PPL, 400x.

108

Spatial Distribution Of Etch Features

General Extent and Distribution of Etch Pits. Etch pits do not occur
uniformly throughout the sample set. Many of the thin sections and grains
examined show little or no etch pit development. On those thin sections where
etch pits occur, etch pit ﬁelds or clusters are seldom contiguous, and there are
usually large areas with no etch pits between the clusters. This can be seen in
Figure 45, which illustrates a representative area on thin section Sylva6U, a
sample characterized by an unusually dense occurrence of etch pit ﬁelds. At the
thin section scale, distribution may be roughly uniform or may be concentrated in
one area of the thin section, as shown in Figure 46, a generalized map of thin

section clusters on sample Sylva5C.

Almost all of the etch pit ﬁelds observed occur adjacent to fractures in or
between olivine grains. Etch pits are more common adjacent to fractures ﬁlled
with serpentine (Figures 47 through 50). The apparent width of these serpentine-
ﬂlled fractures varies from a few microns to several tens of microns, depending
not only on the thickness of the serpentine layer, but also on the angle at which

the thin section surface intersects the layer. In the samples analyzed for this

s‘tl-ldy, the actual thickness of serpentine layers seldom exceeds 50 microns.

Men viewed in cross-section, etch pit depth and width appear to be roughly
proportional to the thickness of the serpentine layer. Variation of etch pit density

and degree of coalescence with thickness of serpentine layers was not assessed

109

 

. ¥ . j..- v’ ’ . . g * ‘ ' ' ,
Figure 45. Low-magnification SEM image of a small portion of thin
section Sylva6U, showing distribution of prominent etch pit clusters (outlined
in white). Light gray areas are olivine, dark gray linear features are serpentine
layers, massive dark gray areas are talc. Large black area at upper right is a
void, probably caused by plucking of a mineral grain during thin section
preparation. Smaller, irregular black shapes are likely dust or pollen particles.
BEI, nominal magniﬁcation 100x, actual magnification of printed image
~190x target size. Printed image resolution 200 pixels per inch, image size
1000 by 1200 pixels (Same as original SEM image).

110

 

Figure 46. Scan of thin section of Sample SylvaSC with dot symbols
showing etch pit clusters. Number of etch pits and extent of cluster varies.
Gray shading at left and lower right indicates plucked areas (voids).

111

 

Figure 47. A field of diamond—shaped etch pits at an intersection of two
serpentine-filled fractures. Small dark spots are dark green spinel grains.
Sample CG82-3#1, PPL, digital mosaic, 400x.

112

,I'N'; 1.

av, , ;;;

     

      

   

  

     

   

Figure 48. Etch pits forming under a layer of serpentine (b ttom and etch

pits on the surface of an olivine grain from which the serpentine layer was
cleanly broken away during sample preparation (upper three-quarters of image).
Note open space between serpentine and olivine. Sample Sylva5, SEI, 1000x.

113

 

 

100m

Figure 49. Etch pits developed along a serpentine layer (S). Sample Sylva6U,
polished thin section, SEI, 2500x.

114

 

 

Figure 50. Etch pits developed along a thick serpentine layer (top) and a
narrow fracture (lower center) with only a thin filling, possibly an alteration
product. The thicker layer is divided in two by a central parting, common in
serpentine layers, and also has channels along its upper and lower surfaces.
The uppermost channel (indicated by arrows) has been largely filled by an
alteration product. Sample Sylva6U, polished thin section, BEI, 750x.

115

because actual serpentine layer thickness cannot be determined in oblique or

plan views.

Less common, but still numerous, etch pit ﬁelds are also found adjacent to
narrow, apparently empty fractures (Figures 51 and 52). Very few etch pit ﬁelds
were observed where there were no visible fractures present, and even these
may have been exposed by the cutting of the thin section in such a way as to

intersect the pits but miss the fracture.

Spatial Anangement of Etch Pits Within Fields. Vtﬁthin an etch pit ﬁeld, pits are
typically spaced at generally even intervals, and, if elongated, are generally
aligned parallel to one another. Figures 53 and 54 illustrate representative

examples.

In some cases, pits are roughly the same size (Figure 55), while in other
cases pit size varies from one part of a ﬁeld to another (Figure 56). In the latter
case, the larger pits may be more accessible to ﬂuids penetrating via fractures.
Less commonly, small and large pits are intermixed (Figure 57). In such cases

differences in pit size may be due to strain patterns within the crystal lattice.

Etch pit shapes are usually similar throughout a ﬁeld (Figure 58) and
a'though the appearance of etch pits may vary in terms of length-to-width ratio or
deQree of etch pit coalescence (Figure 59), the basic etch pit type hardly ever

Varies from one part of a ﬁeld to another. _

116

 

Figure 51. Three fields of scattered etch pits formed along unfilled fractures
(indicated by arrows). Sample Syl1709, PPL, 400x.

117

 

Figure 52. Etch pits (dark spots) developed along curving fractures. Sample
CGSZ-3#2, PPL, digital mosaic, 400x.

118

 

 

'5
,r
. v ,
1 I g- ' i 1';
. h ‘ ‘ I .I'V i
u! .‘ v“,
7‘79 5
r .
x r... c
a ‘ ’7”’cr
1 ‘2' ‘
|
1 ¢ .
’.
7' .
2 pm

 

Figure 53. Evenly-spaced, parallel etch pits at micron scale. Sample Sylva7,
fractured grain surface, SEI, 10,000x.

119

 

 

'4' “. ‘ - ,1. _
' _ I 5
I ' - ‘ - O J , -.‘ -. r ,-
. .2 ‘ _ . .
‘1’ . -" ' ‘ ' .' ' I
e - - ..
V ‘ 7 7 ~ ‘ , v r I I '
‘ u . ' ‘W
-‘ . . -: .
s . . 5’
m 4 . 'I‘ . .
. " ' - ‘5 u
> " . . 4 I. ‘ '__ -_g ¢ _ (n -,
i , .
‘ . - ~ I. I - ’1'
. . ‘ I n , i-v.
i ’ H: ‘ t I y 4".
) _ ' .. N
~« . . l 10 pm
8 V I ..
I . . .1
l ' ' ‘

Figure 54. Afield of uniform-size, evenly-spaced, parallel etch pits
exposed by removal of an enclosing serpentine layer (remnant at left) in
thin section preparation. Sample Sylva5U, polished thin section, SEI, 2000x.

120

 

/

 

Figure 55. Uniform-size shallow-angle fluted etch pits. Sample Sylva6U,
polished thin section, SEI, 1000x.

121

 

Figure 56. A field of etch pits grading from larger, compact diamond-shaped
pits (top center) to smaller, elongated pits (left and bottom). Sample SylvaGC,
PPL, digital mosaic, 400x.

122

10 um

 

Figure 57. A field of intermixed large and small etch pits. Sample Sylva5U,
polished thin section, BEI, 2000x.

123

 

 

v

4'

‘23 ",'~',‘ ”.1

Figure 58. A large ﬁeld of roughly triangular etch pits. Sample SylvaSC, PPL,
digital mosaic, 400x.

124

 

 

200m
- .. \P‘
Figure 59. An etch pit field enclosed by fractures, with pits varying in size,
proportions, and degree of coalescence, but retaining the basic diamond shape
as a component throughout the field. Sample Sylva5C, PPL, digital mosaic, 400x.

125

In addition to apparent “two-dimensional” ﬁelds, a common type of spatial
arrangement is the “one-dimensional” linear array, in which similarly shaped etch
pits, roughly uniform in size, occur at approximately even intervals along a
straight fracture or dislocation path (Figures 60 through 62). A variant of this form
is the less common curved array (Figures 63 and 64). Etch pits in such arrays
are almost always diamond-shaped, although it is possible that the diamond

l
shapes are, in many cases, cross-sections of paired cones joined at their bases. 1
l

Straight prismatic or tubular etch features occur in small clusters of
parallel linear voids, which can easily be mistaken for surface fractures at low
magniﬁcation, but can often be seen in the optical microscope to lie at different
depths from one another, some coming into focus and others going out of focus
as the ﬁne focal adjustment knob is turned. Also, as shown in Figure 65, these
features usually appear to be tubular, or at least to have a hollow core which is
often partially ﬁlled with what appears to be a dark, opaque alteration product.
Figure 66 is an oblique view of a large group of such features, perhaps with their
bases truncated in this thin section. The microscope’s focus is adjusted to show

the ends nearest the camera in sharp detail, leaving the opposite ends blurred.

Few cases are found where etch pits in olivine have coalesced to the
degree that only a few remnants of the former surfaces remain. A case of fairly
advanced dissolution is shown in Figure 67. Beyond this stage, continued
dissolution will likely remove enough material that the surface is no longer

recognizable as an etch pit ﬁeld, as in Figure 68. Such areas will also be

126

 

 

Figure 60. Linear arrays of diamond-shaped etch pits viewed through olivine.
Sample sylvaSC, PPL, 400x.

127

 

 

, -.....f

..f ’ it

I i 2

(I

:4

it
3,

5 pm

Figure 61. Linear arrays of diamond-shaped etch pits along parallel
fractures. Sample Sleec, polished thin section, BEI, 5000x.

128

   
 

...-i: f" j I
. . (”'13. 10 um

. ._ ' - ﬂ”; . M: .-'f._-. .A" * g. r .-
Figure 62. Parallel linear arrays of diamond-shaped etch pits, apparently
filled with a dark mineral. Very small arrays can be seen at lower left (black dots)
and upper left (white dots). Large triangular pit at lower left appears to be a
flattened cone. Sample Sylva4C, PPL, 400x.

   

129

    

  

" .7. 10 pm ,“\
Figure 63. Arrays of evenly-spaced diamond-shaped etch pits following
curved fractures. Sample Sylva4C, PPL, 400x.

. r

130

 

Figure 64. Diamond- -shaped etch pits lying along a curved fracture, inter-
sected by arrays on three smaller, parallel, straight fractures. Sample Sleec,
polished thin section, BEI, 2000x.

131

 

 

ngre 65. Straight, parallel features (center), possibly tubular or prismatic
etch channels, or possibly a replacement mineral, extending from serpentine
layers into olivine. Sample Sylvasc, PPL, 400x.

132

 

Figure 66. A large field of straight, parallel etch features, or possibly needles
of a replacement mineral, protruding into olivine. Possibly the serpentine layer
from which they are protruding was removed by thin section preparation (a
remnant is at the top of the image). Only the top (near) halves of the features
are in focus, with the bottom halves blurred and faded. Sample Syl1709, PPL,
400x.

133

  

    

I, .‘O
'-‘W to ’4».

 

Figure 67. Fairly advanced dissolution of olivine (left). Small section of etch
pit field at far left has almost disappeared, leaving only the cracked surface of
an alteration product. Sample Syl1709, polished thin section, SEI, 1500x.

134

 

 

 

Figure 68. Partially corroded etch pit field (right) and apparent site of former
etch pit ﬁeld (center) removed by corrosion (note channel through center).
Sample Sylva6U, polished thin section, SEI, 1000x.

135

 

especially susceptible to abrasion, and thus may be removed during thin section

preparation.

 

136

Comparison of Observed Natural Features with Textures Developed

Experimentally.

General Similarities. In their gross morphology, natural weathering
textures of olivine resemble textures created experimentally. Shape, size,
orientation, spacing, and alignment are similar within a single etch pit ﬁeld in both
types of textures. Figure 69, an SEM image of a naturally weathered olivine
surface, illustrates this generalization. The pits in the ﬁeld all have a deﬁnite,
repeated geometric shape. They are all roughly the same size - within one order
of magnitude, and approximately the same depth. All pits are oriented such that
their apices point in the same directions. The pits are approximately evenly
spaced and show a subtle alignment along an upper left to lower right direction.
These statements also describe the general characteristics of most, if not all,
experimentally created corrosion textures of olivine. For example, Figure 70
(Awad et al., 2000), an SEM image of an acid-etched olivine surface, displays
essentially the same general characteristics as the naturally weathered surface in

Figure 69, even though there is little resemblance in the details.

From the above observations it may be inferred that the same principles
apply in both cases, namely that surface reaction is the rate-determining step

and that etch pit geometry is crystallographically controlled.

137

 

 

 

Figure 69. Awell-defined field of triangular etch pits.
fractured grain surface, SEI, 1000x.

138

 

Sample Sylva5,

 

.—

   

Figure 70. Experimentally etched surface of an olivine crystal parallel to the
b-axis. Alter Awad et al. (2000).

 

139

Speciﬁc Similarities. There are few examples reported of experimentally formed
etch pit shapes that are similar to those formed naturally. Grandstaff (1978)
found that sharply-deﬁned diamond-shaped etch pits were formed by etching on
some olivine surfaces. These pits, illustrated in an SEM image, were

approximately 2 microns in length, and coalesced to form serrated channels.

Pokrovsky and Schott (2000) illustrated an etched olivine surface with
rudimentary shallow-angle conic sections (Figure 71) comparable to those found
in the present study (Figure 72). Awad et al. (2000) found that elongated
diamond-shaped etch pits with curving sides developed along a surface

paralleling the c-axis of etched forsterite (Figure 73).

Wegner and Christie (1976) generated, among other results, a ﬁeld of pits
with apparent conical centers and rounded rectangular edges (Figure 74). This
approaches the same pit shapes observed in some SEM images acquired in the
present study (Figure 75). Other cases, observed with the optical microscope,
suggest that some etch pits may begin as cones and then evolve into diamond
shapes. In Figure 76, thin white trails or “leaders” are visible extending from the
points of wedge-shaped (elongated diamond-shaped) etch pits, creating a small
conical or funnel shaped protrusion at the apex. In Figure 77, the margin of a
serpentine layer is lined with apparently faceted etch pits terminating in a

rounded tip, in many cases with a very thin dark “leader” extending from them.

140

 

 

 

   

‘r- :3 .1, », -. air _' , '
. :‘f‘. ‘r’ .‘ ., .4 A ' v l7».'. .‘ f
- .

Figure 71. SEM image of forsterite etched by acid (pH 2). Note etch pits
in the form of triangles or truncated conic sections, center and upper right.
After Pokrovsky and Schott (2000).

r e ”Ti-:3" as}

 

141

 

 

 

Figure 72. Etch pits in the form of truncated triangles. Compare to Figure 71.
Sample Sylva1, fractured grain surface, SEI, 2500x.

 

142

 

20 pm

Figure 73. SEM Image of forsterlte etched with acid (pH 1 or pH 2),
suface parallel to c-axis. Alter Awad et al. (2000).

 

 

143

 

   

Figure 74. Apparent conical etch pits evolved into rectangular pits
with rounded comers. After Wegner and Christie (1976).

 

144

 

 

\ * ' . . 6
Figure 75. Arrows indicate conical etch pits. Large
pit at left and pit at top right have evolved into sub-
rectangular pits with rounded corners. Compare to
Figure 74. Sample Syl1709, polished thin section,
BEI, 1200x.

145

 

   

 

 

A} ‘3, , ,mﬁl ‘ a" .E‘ ’ g _. ‘3? hgwmm
Figure 76. Wedge-shaped etch pits protruding from the margin of a

serpentine layer into olivine. Note white “leaders” extending from pits.
Sample Sylva50, PPL, 400x.

146

 

 

 

   

20 pm

0 - . K
Figure 77. Cross-section of a serpentine layer with conical or pyramidal

etch pits protruding into olivine from both boundary surfaces. Sample SylvaSC,
PPL, 400x.

 

147

Wegner and Christie (1976) also show ﬁelds of elongated etch pits (Figure
78) that resemble closely the elongated pits occurring commonly in naturally
weathered olivine. Earlier, Wegner and Christie (1974) showed examples of
linear etch pits in experimentally deformed olivine (Figure 79) that resemble the
linear etch pits found throughout the Webster-Addie samples. This suggests that
more similarities might be found if future researchers were to use tectonized

olivine in etching experiments.

 

Elongated boat-shaped etch pits produced by Wegner and Christie (1974),
shown in Figure 80A, closely resemble natural etch pits found on a fractured
grain of Webster-Addie olivine, Figure 803. The naturally weathered surface,
however, is a unique example. No other boat-shaped etch pits were found in the
present study, and this was also the only example found of point-bottomed

shapes.

Similarly, although there is a close resemblance between the V-shaped
etch pits produced by Awad et al. (2000) and those found on a fractured grain
from a sample used in the present study (Figures 81A and 813), this was also a
one-of-a-kind find. It should be noted, however, that only 12 hours of SEM

observation time was devoted to examination of grains, compared to 60 hours of

 

SEM time examining polished thin sections. It is probable that more examples of

these surfaces would be found if additional time were spent looking for them.

148

 

Figure. 78. Polished and etched surface of gem-quality
olivine, showing oval etch pits with pointed ends (a) and

elongated pits with diamond-shaped centers (b). After Wegner
and Christie (1974).

149

 

 

 

   

Figure 79. Etch pit fields in experimentally strained gem-quality olivine
(”Red Sea peridotite”). After Wegner and Christie (1974).

 

150

 

, . . .
I ‘u \ 3 ; _ ’II‘I'. . f
. . ' I =5. 44;]?

Figures 80A and BOB. Similar etch pits on experimental and natural surfaces.
Left (A): defects brought out by etching F085 olivine, locality unknown, after
Wegner and Christie (1974). Right (B): natural etch pits in fractured surface of
olivine grain, sample Sylva2, SEI, 2500x.

151

 

 

 

Hi

 

 

5pm

‘—

 

 

Figures 81A and 818. Similar etch pits on experimental and natural surfaces.
Left (A): etched surface parallel to a-axis of gem-quality forsteritic olivine (F091)
from San Carlos, AZ, after Awad et al.(2000). Right (B): natural etch pits in
fractured surface of olivine grain, sample Sylva1, SEI, 5000x.

152

 

Speciﬁc Differences. The examples given above notwithstanding, etch pit shapes
in most experimentally formed textures, at least in forsteritic olivine, do not have
exact counterparts in the naturally weathered specimens examined in this study.
Rectangular pits with sharp internal angles were found by Horn and Muerette
(1964), the ﬁrst researchers to etch olivine (Figure 82). Young (1969) found
similar pits (Figure 83). Wegner and Christie (1976) produced ﬂat-bottomed,
elongated hexagonal pits, and Kirby and Wegner (1978) etched gem-quality
olivine to produce straight rows of uniform, point-bottomed pyramidal pits with

rectangular outlines.

According to Velbel (2009), the only cases of olivine textures formed
experimentally that resemble those formed in nature are those resulting from

treatment with extremely acidic etchants.

153

 

 

 

90 Min
: [010] 20 pm

 

 

 

Figure 82. Etching sequence of olivine from Horn and Muerette’s historic
etching experiment. Note that V-bottomed rectangular pits start out elongated
in a right-left direction, then become square, and then elongated in an up-down
direction. After Horn and Muerette (1967).

 

 

w‘ - .. -- .. ' $.32-

Figure 83. Rectangular etch pits formed on the (010)
surface of stressed olivine. Maximum principal stress
was left-right. After Young (1969).

154

 

Part V. Discussion
Diamond-shaped etch pits are commonest form, and are composites of cones.

Diamond-shaped pits occur as individual lozenges and as linear pits
composed of aligned coalescing diamond-shapes, which eventually coalesce
further to leave corroded surfaces. Diamond-shaped outlines could conceivably
be formed as cross-sections of pyramids, or as cross-sections of cones joined
base-to-base, as described in Chapter 12. If these were formed as cross-
sections of pyramids, they would be point-bottomed with four discernible facets
when seen in plan view, and would also display four facets when seen from the
side. However, no point-bottomed, four-faceted pits were encountered in any of
the optical and SEM images made, and side views of etch pits in optical images
typically showed only two areas separated by a single dividing line, as would be

the case if the pits were composites of conical forms.
Most Etch Pits are Composites of Conic Sections.

Most of the etch pit shapes illustrated in previous sections can be seen as
cones, conic sections, or joined or coalescing conic sections, as indicated in text
references to the respective images. In recent articles, Velbel and Ranck (2008)
and Velbel (2009) concluded that all etch pits found in the olivine samples they

analyzed were forms of conic sections.

155

 

 

 

Simple Cones.

Conical etch pits, as shown in Figures 25 through 29, are of course the
clearest example of this form, and cannot reasonably be questioned. Elliptical
etch pits, many of which can be seen to taper inward (Figure 25), are simply
cones intersected at an angle by the thin section or fracture surface. Viewed in
the optical microscope, the inside surfaces of elliptical etch pits are not usually
visible, but proﬁles of etch pits protruding from the margins of serpentine layers
into olivine often appear to be conical (Figures 23A and 233), or at least to have
conical tips (Figure 77). Occasionally, a straight or sinuous “leader” is seen

protruding from the tip into the olivine as illustrated in Figure 76.

Wedge-shaped ancll Triangular Etch Pits.

Wedge-shaped etch pits that occur along the edges of olivine grains are
easily explained as halves of diamond-shaped pits in which the opposing half has
been corroded away. Isolated triangular etch pits are more somewhat more
problematic, as in Figures 35 and 36. Here, conical pits appear to have
developed in one direction only, leaving the material on the other side of the
cone’s base intact. The biconic shapes, however, continue beneath the exposed

surface, leaving only part of the etch pit exposed.

156

 

 

 

Shallow-angle Fluted Etch Pits

Shallow-angle ﬂuted etch pits usually display conic sections in the form of
ﬂattened cones with converging edges. In Figures 32 and 33, cone-shaped
depressions are visible as darker shadow areas, separated by lighter-colored
ridges or ﬂutes. The rows of coalescing cones are separated by narrow remnants
of the olivine grain, which have been corroded along the edges by the expanding

pits.

 

“Mouth-shaped” Etch Pits

Distinctly “mouth-shaped” etch pits are more difﬁcult to explain in terms of
conic sections. However, close inspection, especially of SEM images of three-
dimensional olivine chips, shows that ﬂattened cones developing at different

densities in opposite directions are the components of these shapes. (Figure 31)

Straight, Needle-Like Features.

The straight-sided features (Figure 39, 40, 65, and 66), whether tubular or

 

prismatic in cross-section, cannot be explained as variations of conic sections.
Cronin (1983) observed (without illustration) that some olivine grains in the dunite
of the Webster-Addie body possess abundant parallel fractures, although not
necessarily at grain boundaries. He also noted that, in some cases, these

fractures are occupied by needles of anthophyllite. However, the features

157

 

 

described in the previous chapters appear to be hollow tubes or prisms,
sometimes containing disconnected bits of dark matter. The most likely
candidates for anthophyllite needles are shown in Figures 39, where the central
feature has a feathery appearance, and Figure 65, in which the two linear
features on the left of the image are shorter, thicker, and have a more crystalline
appearance than the rod-like features in other images. However, if these are
elongate needles of anthophyllite or a similar mineral, they would not be

considered etch pits, but rather a form of mineralization.

Rectangular Etch Pits.

Rectangular etch pits, other than diamond-shaped, cannot be explained
as conic sections. Where they are found together with diamond-shaped pits
(Figure 42), the most common type of occurrence, they might be considered a
distorted type of paired half-cone cross-sections. Fields made up entirely of
rectangular etch pits (Figure 41) are rare, and usually exhibit low relief and iron-
oxide staining, suggesting that they may be composite (corrosion/deposition)

features related to iron oxide mineralization.

Surface and Subsurface Features Of Iron-Oxide Rich Zones

Features typical of iron-oxide rich zones, such as low-relief pits of various

shapes (Figures 41 and 42), “ﬁligree” patterns and subsurface corrugations

158

 

 

(Figures 41 and 13-44), are apparently the result of excess iron being set free by
serpentinization. Serpentine minerals are able to take up only very small
amounts of Fe. Deer et al. (1992) generalize that, “when serpentines are formed
in peridotitic rocks, most of the iron present in the original olivine or pyroxene is
incorporated in magnetite or hematite and does not enter the serpentine
structure.” Accumulation of iron oxide minerals appears to smooth over etch pit
ﬁelds and to create layers or encrustations of iron oxyhydroxides on olivine
surfaces. The relative abundance of rectangular etch pits in such zones may be
the result of post-mineralization etching, in which a different crystallographic

structure is reﬂected in the shape of the etch pits.

Solution Channels Adjacent to Serpentine- Filled Fractures.

As stated previously, almost all etch pit ﬁelds are found adjacent to one or
more serpentine layers (Figures 23A, 238, 47, 49, 50, and 77). In some cases,
especially in SEM images of grain mounts, the ﬁeld is visible but the enclosing
serpentine layer has been broken away in grain mount preparation (Figures 48
and 54). A smaller, but still signiﬁcant number of etch pit ﬁelds lie adjacent to

empty fractures (Figures 51 and 52).

In many cases, a channel providing for ingress of ﬂuids and removal of
solutes is visible between the serpentine layer and olivine (Figure 48). In some
cases however, such open channels can simply be open spaces caused by the

dehydration and shrinkage of the serpentine, as is likely the case in Figure 48. As

159

 

seen in this image, there is no evidence of ﬁlling by an alteration product in the
channel and, except for the openings of the adjacent etch pits, there is no
evidence of channel ﬂow. Although there is probably a connection between this
olivine-serpentine interface and the adjacent etch pits, it was likely due to a slow

transfer of ﬂuids over a long time through a much narrower space.

Correlation Of Etch Pit Size And Density With Proximity To Weathering Surface —
Not Determined.

The nature of the sample suite did not permit evaluation of the size or
density of etch pits in relation to the weathering surface. Many thin sections did
not contain recognizable exposed areas, and fractured grains were extracted

randomly from hand samples.

Correlation Of Etch Pit Size And Density With Chemical Weathering lndioes -

Not Determined.

Studies of mineral corrosion frequently use various chemical weathering
indices to compare morphology to degree of weathering. This was not done in
the present study for two reasons: First, although surrogate weathering indices
for olivine phenocrysts in basalt can be calculated using bulk chemistry of the
host rock, these indices cannot be used for olivine in ultramaﬁc rocks because
they rely on measures of alkalis and alumina, which are only minor constituents

of such rocks and are usually distributed unevenly (Velbel, 2009). Secondly, as

160

 

 

illustrated in Figures 45 and 46, etch pit distribution is itself heterogeneous, which
precludes the identiﬁcation of portions of a specimen with uniform density or size

of etch pits, large enough to be sampled for chemical analysis.

Comparison with Textures Formed Experimentally.

General Similarities between Naturally Weathered and Experimentally

Corroded Olivines.

As discussed in a previous section, natural weathering textures of olivine
resemble textures created experimentally in many ways, including uniformity of
size, spacing, and orientation within a given etch pit ﬁeld. There were few cases
of close similarity between the details of etch pit shape observed in this study
and those produced in laboratory experiments. These cases are described in

Chapter 15.

Differences between Natural and Experimental Conditions.

Natural weathering of olivine takes place over long periods of time at near
Earth-surface conditions. Laboratory experiments usually take place over
comparatively short periods of time (several orders of magnitude shorter than
natural weathering). Also, because experimentation covering natural time spans
is impossible, dissolution and etching experimenters compensate by the use of

strong reagents, usually acidic solutions at pH 1 or pH 2, and experiments

161

usually take place at temperatures higher than those typical of Earth-surface

conditions.

Furthermore, laboratory researchers typically use gem-quality or other
high-grade olivine. For example, Wegner and Christie (1974) used fresh “Red
Sea peridotite”, presumably from the high-grade body at St. John’s Island, Egypt,
and Awad et al. (2000) experimented on gem-quality San Carlos olivine, well
known as probably the highest quality olivine occurring naturally within the United

States.

In order to constrain the results, laboratory experiments have generally
been carried out on olivine by itself, isolated from other minerals with which it

typically occurs - pyroxenes, amphiboles, and of course, serpentines.

In contrast, the dunites of the Webster-Addie ultramaﬁc body are
dissected by a dense microfracture network, are heavily serpentinized, possibly
having experienced several episodes of serpentinization, and show signs of
straining such as intra-grain microfractures and slippage along fracture planes.
Thus, the present study addresses weathering of tectonized olivine occurring

with a variety of other minerals.

As pointed out by Casey et al. ( 1993), dissolution experiments commonly
disperse a small amount of mineral into a large volume of agitated solution,
whereas TEM studies (Eggleton,1984; Smith et al.,1987; Banﬁeld et al., 1991a,

b) showed that olivine weathering under natural conditions often proceeds in very

162

small (~10A) clay-ﬁlled channels where mineral chemistry is affected by

interaction between the closely spaced electrostatic layers.

A further difference, emphasized by Velbel (2009), is that olivine
dissolution experiments are typically set up with reference to crystal axes,
whereas natural conditions expose grains in an inﬁnite variety of surface

orientations.

Thus, it is not surprising that, while the general spatial characteristics of
natural weathering textures and those formed experimentally are similar, they
differ markedly in detail, primarily in the shapes of etch features. Experimental
textures are generally more angular and have higher local relief than natural
textures. Point-bottom and V-bottom etch pits are common in textures produced
experimentally, whereas only three examples of such textures were found in the

present study — all on fractured grain surfaces.

Variations Within The Sample Suite.

There was wide variation in the occurrence of etch pits throughout the
sample suite. Among the thin sections examined with the optical microscope, a
few samples were particularly rich in etch pits, many had scattered etch pits, and
on several thin sections no etch pits were observed, although they might have
been missed because visual scanning was only partial on most thin sections.

Polished thin sections showed a similar pattern.

163

Because time and budget were limited, and the ﬁrst 60 hours of SEM
imaging was done using polished thin sections, only four 3-hour SEM sessions
were available for studying fractured olivine grains. These sessions were used to
image grains from the “Sylva” series, Sylva1 through Sylva7. Therefore, no

generalization can be made for the entire sample suite.

Comparison Of Results Of Optical Microscopy With SEM Images of Polished
Thin Sections And Fractured Grain Surfaces.

Images made using optical microscopy are constrained by the resolution
of the optical microscope, and are therefore of limited use in determining the
shapes of small, individual etch pits. The main advantage of optical microscopy is
the ability to see through olivine, serpentine, and etch pit fields, and thus gain a
three-dimensional, if relatively low-resolution, view of the spatial arrangement of
etch features. A signiﬁcant practical advantage is that optical microscopes are
available without cost, scheduling, or technical assistance, and an unlimited
number of images can be made with a digital camera. Most of the optical
microscope images presented in this study were acquired after the SEM phase

had ended.

SEM images of polished thin sections in SEI mode provide high-resolution
information on etch pit topography, showing clearly, for example, the serrations
on the edges of linear etch pits composed of coalescing diamond-shaped pits, as
well as the contours of etch pit sides and bottoms revealed by differences in
shading. BEI images allow differentiation between olivine and alteration products,

164

and also show topography, sometimes more clearly than SEI. The main
disadvantage of SEM imagery of polished thin sections is that it only records
surface phenomena, and does not detect features beneath the surface, in the
bulk of the thin section. As an example, most etch pit ﬁelds visible with optical
microscopy, including very extensive ﬁelds, are invisible to SEM. Features are
visible only where they fortuitously intersect the surface of the thin section. For
this reason, a good deal of time must be spent scanning large areas of the thin

section in order to ﬁnd features that are worth imaging.

While SEM study of fractured grains also is limited to the surface of the
specimen, experience in the present study suggests that fractured grains are
more fertile in numbers of etch features, and in variety of etch pit types, than are
polished thin sections. Fractured grains are particularly useful for imaging three-
dimensional and composite shapes. A disadvantage of fractured grains is that
the process of fracturing the grain is necessarily somewhat haphazard, and
favors exposure in areas of weakness, with the result that a disproportionate
number of exposed surfaces are the interfaces between olivine and serpentine.
This is evident, in Figure 8, an SEM image of a fractured grain in which almost all
of the fracture surfaces can be seen to follow grain boundaries, outlining the

intact olivine grains.

In summary, each of the three methods has its strengths, optical
microscopy for breadth and depth, SEM of thin sections for etch pit detail and
composition, and SEM of grain mounts for viewing high relief fractured surfaces.

Using all three in a balanced approach should give optimal results.

165

A Conceptual Mechanism For Conical Etch Pit Formation.

The formation of conical etch pit appears to proceed like the spread of a
drop of oil into a stack of sheets of paper towel. In the interest of clarity, a single
cone will be modeled. For a bi-conical etch pit, the single cone model can simply

be mirrored.

In this model, the axis of the cone begins at the point where the drop of oil
ﬁrst touches the topmost sheet. The spread of oil from this point represents the
dissolution of olivine. In natural weathering of olivine, this point would likely be

the intersection of a microfracture with an edge or screw dislocation.

In this model the oil is assumed to spread more rapidly across the surface
of the sheet than downward into the stack, resulting in a cone with a radius that is
greater than the length of its axis. Horizontal spread is assumed to take place at
a uniform rate in all directions, deﬁning the circular base of the cone. However,
some images of wedge-shaped etch pits and shallow-angle ﬂuted etch pits
suggest that these may be composed of “ﬂattened” cones with oval or elliptical
bases. This can easily be accommodated by varying the rate of spread in
hypothetical “x” and “y” directions, with all intermediate directions expressed as

vectors of “x” and “y”.

As the drop percolates downward at an assumed uniform rate, it deﬁnes
the axis of the cone, and oil spreads outward horizontally at the same rate as on

the surface of the top sheet. When the oil is completely absorbed, the cone’s

166

apex at the lowest point the drop has reached, and the straight sides of the cone

are deﬁned by the loci of points on the circumference of the stacked circles.

If two such cones are connected base-to-base, a cross section through
their common axis will deﬁne a diamond shape, with the long dimension of the

diamond lying in a horizontal direction.

A Question Of Etch Pit Spacing.

Many researchers have established that etch pits result from preferential
etching of edge and screw dislocations within olivine. If, as has been inferred,
etch pits develop at the intersections of dislocations with exposed surfaces, and
dissolution is surface-reaction limited, does this mean that there are no
dislocations in the spaces between etch pits? Is the spacing of dislocations

naturally so convenient for low-magniﬁcation optical and SEM imaging?

It is evident from the images presented in this paper that there are ﬁelds of
uniformly spaced large etch pits, easily visible by optical microscopy, with spaces
between the pits that are apparently lacking in etch pits but which, at SEM
scales, are found to contain numerous smaller etch pits. Figure 20 illustrates
such a combination of large pits that would easily be seen with an optical
microscope and small pits that are below the 0.2-um limit of optical resolution. At
greater resolution, TEM images show etch pits and channels too small to be

seen in SEM imagery. Presumably, increasing resolution with more powerful

167

imaging methods, like atomic force microscopy, would reveal even smaller etch

features.

This raises the following question: Is the a process at work in olivine
dissolution that is akin to the process crystal formation, where some minerals
crystallize more readily, generating greater numbers of crystals, while others
grow more rapidly, forming fewer, larger crystals? Or is there a process like
Ostwald ripening, where larger crystals grow larger while smaller crystals
stagnate and eventually become absorbed by the larger ones. Clearly, etch
features that are already formed are more likely to collect and channel ﬂuids,
thus increasing dissolution. However, this does not explain the more or less
uniform spacing of etch pits, or of rows of etch pits. Perhaps the answer to this

question is trivial. If not, it deserves further research.

168

Part VI. Conclusion

The hypothesis being tested - that natural weathering textures of olivine
conform to kinetics determined experimentally - has not been disproved. There
are speciﬁc examples of similar textures found on both naturally and artiﬁcially

etched types of surfaces.

Equally important, if not more so, there is a wide range of general
similarities, such as uniform size of etch pits within a ﬁeld, similar etch pit forms
within a ﬁeld, and uniform orientation, implying that reﬁnements of experimental
etching procedures could result in an even greater number of closely similar

speciﬁc features.

Roughly uniform spacing and subtle alignment of etch pits in naturally
weathered olivine, as well as in experimentally etched olivine, with very little
corrosion between pits, afﬁrrns that dissolution is anisotropic and surface reaction
is the rate-limiting factor. This in turn suggests that etch pits occur where edge or
screw dislocations intersect the exposed surface, creating nodes of relatively

higher free energy that are more susceptible to corrosion.

The commonest etch pit shape is that of the diamond or lozenge.
Diamond-shaped pits occur singly and in rows of coalescing pits forming linear

features with serrated edges. Diamond-shaped pits are cross-sections of pairs of

169

conical pits joined base-to-base. Almost all other etch pit forms can also be

described in terms of conic sections.

Etch pits are distributed heterogeneously, and are concentrated in ﬁelds or
clusters adjacent to fractures that permitted ingress to ﬂuids. Most etch pit ﬁelds
are adjacent to serpentine-ﬁlled fractures, and linear arrays of etch pits are found

on unﬁlled fractures.

Due to the heterogeneity of etch pit distribution in the samples studied, it
was not possible to determine if etch pit size and density were correlated with
proximity to surfaces exposed to weathering. To study this relationship, oriented
proﬁles from bedrock outcrops should be used, with thin sections and grain

mounts referenced to locations within the proﬁles.

Each of the three methods used in the study — optical microscopy of thin
sections, SEM of polished thin sections, and SEM of grain mounts - has its
strengths and weaknesses. The three methods complement each other in a
study such as the present one. In reviewing the results, it appears that the study
would have beneﬁtted from a greater effort directed toward SEM of grain mounts

from the entire range of specimens within the sample suite.

170

 

APPENDIX

171

 

Table 1. Major, Minor, and Trace Element Composition

of Samples Analyzed by X-ray Fluorescence and lCP—MS*

Sample Si02 (%) MgO (%) Fe203 1102 (%) Al203 (%)
(%)

CGS 2-3#1 A4 42.43 43.13 9.67 0.02 0.71

CGS 2-3 #1 A1 42.02 42.8 11.32 0.02 0.75

CGS 2-3 #1 B4 42.06 43.19 9.61 0.02 0.64

CGS 2-3 #1 B1 42.35 42.68 10.66 0.02 0.77

CGS 2-3 #2 A 41.71 42.24 10.26 0.02 0.72

fresh

CGS 2-3 #2 B 42.03 41.7 11.35 0.02 0.75

weathered

CGS 2—3 # 2 C 41.09 42.81 11.13 0.02 0.69

CGS 2-3 #2 E 41.82 41.68 11.53 0.02 0.76

weathered

CGS 2-3 #2 D 41.37 39.59 13.02 0.02 0.83

weathered

Sleec A 43.05 44.6 9.35 0.01 0.48

fresh

Sleec A most 43.7 43.42 10 0.01 0.56

weathered

Sleec B 43.03 43.73 9.06 0.01 0.62

fresh

Sleec B most 43.7 43.42 10 0.01 0.56

weathered

Syl1709 A fresh 42.83 43.67 8.83 0.02 0.7

Syl1709 A (D) 42.32 38.81 10.64 0.05 1.68

weathered

Syl1709 B (C) 43.87 38.07 9.62 0.05 1.43

weathered

Sylva5 core 40.31 47.95 9.42 0.01 0.32

Sylva5 rind 40.6 48.35 9.72 0.01 0.3

*With only two exceptions, the following elements were not detected at levels
greater than 2 PPM in any of the samples listed: Rb, Sr, Y, Zr, Nb, Ba, and
La. The exceptions were one case of 5 PPM Sr and one case of 16 PPM La.

172

Table 1. (Continued) Major, Minor, and Trace Element Composition of
Samples Analyzed by X-ray Fluorescence and lCP-MS

Sample MnO(%) CaO Na20(%) K20 P205(%) Totals
(%) (%)

CGS 2-3#1 A4 0.13 0.48 0 0.01 0 96.58

CGS 2-3 #1 0.15 0.58 0 0 0 97.64

A1

CGS 2-3 #1 0.14 0.59 0 0 0 96.25

B4

CGS 2-3 #1 0.14 1.07 0 0 0 97.69

B1

CGS 2-3 #2 A 0.14 0.8 0 0.01 0 95.9

fresh

CGS 2-3 #2 B 0.14 0.92 0 0.01 0 96.92

weathered

CGS 2-3 # 2 C 0.15 0.68 0 0 0 96.57

CGS 2-3 #2 E 0.15 1.05 0 0 0 97.01

weathered

CGS 2-3 #2 D 0.17 1.24 0 0 0.01 96.25

weathered

Sleec A 0.14 0.27 0 0 0 97.9

fresh

Sleec A most 0.14 0.47 0 0.01 0.02 98.33

weathered

Sleec B 0.13 0.56 0 0 0 97.14

fresh

Sleec B most 0.14 0.47 0 0.01 0.02 98.33

weathered

Syl1709 A 0.12 0.47 0 0.02 0 96.66

fresh

Syl1709 A (D) 0.11 1.49 0 0.01 0 95.11

weathered

Syl1709 B (C) 0.11 2.16 0.04 0.01 0 95.36

weathered

Sylva5 core 0.12 0.03 0 0 0 98.16

Sylva5 rind 0.12 0.03 0 0 0 99.13

173

Table 1. (Continued) Major, Minor, and Trace Element Composition

of Samples Analyzed by X-ray Fluorescence and lCP-MS

Sample

CGS 2-3#1 A4
CGS 2-3 #1 A1
CGS 2-3 #1 B4
CGS 2-3 #1 B1

CGS 2-3 #2 A
fresh

CGS 2-3 #2 B
weathered
CGS 2-3 # 2 C

CGS 2-3 #2 E
weathered
CGS 2-3 #2 D
weathered
Sleec A
fresh

Sleec A most
weathered
Sleec B
fresh

Sleec B most
weathered

Syl1709 A fresh

Syl1709 A (D)
weathered

syl 1709 B (C)
weathered
Sylva5 core

Sylva5 rind

LOl (%)
2.95
1.77
3.27
1.7
3.55
2.51
2.82
2.42
3.12
1.63
1.1
2.4
1.1
2.83
3.83
3.97
1.23

0.27

Cr (PPM)

2466
3265
2480
3507
3170
2971
3698
3060
3615
2180
2848
2244
2848
2120
6659
2876
2999

2950

Ni (PPM)

2310
2545
2237
2555
2170
2612
2314
2474
2508
2361
2665
2220
2665
2989
3766
3583
2948

3001

174

Cu (PPM) Zn (PPM)

0

4

090

16

17

N

ONO

65

46
56
48
60
55
52
63
53
55
53
63
44
63
38
76
45
45

46

 

 

REFERENCES CITED

175

 

References Cited

Awad, A., A.F. Koster van Groos, and S. Guggenheim, 2000, Forsteritic olivine:
Effect of crystallographic direction on dissolution kinetics. Geochimica et
Cosmochimica Acta, v. 64, pp. 1765-1772.

Baker, Ian and Stephen E. Haggerty, 1967, The alteration of olivine in basaltic
and associated lavas, Part II: lnterrnediate and low temperature alteration.
Contributions to Mineralogy and Petrology, v. 16, pp. 258-273.

Banﬁeld, J. F ., D. R. Veblen, and B. F. Jones, 1990, Transmission electron
microscopy of subsolidus oxidation and weathering of olivine. Contributions to
Mineralogy and Petrology, v. 106, 110-123.

Banﬁeld, J. F., B. F. Jones, and D. R. Veblen, 1991, An AEM-TEM study of
weathering and diagenesis, Albeit Lake, Oregon: I. Weathering reactions in
the volcanics. Geochimica et Cosmochimica Acta, v. 55, pp. 2781-2793.

Banﬁeld, J. F, G. G. Ferruzzi, W. H. Casey, and H. R. Westrich, 1995, HRTEM
study comparing naturally and experimentally weathered pyroxenoids.
Geochimica et Cosmochimica Acta, v. 59, pp. 19-31.

Baumhauer, H. A., The Results of the Etch Method in Crystallographic Research
(Die Resultate der Aetzmethode in der krystallographischen Forschung, in
German), Leipzig, 1894, 137 pp.

Becke, F., 1890, Tscherrnaks Mineralogische und Petrologische Mitteilungen, v.
11, p. 349, cited in Sangwal, 1987, p. viii)

Berger, Suzette, Deborah Cochrane, Kyla Simons, Ivan Savov, J. G. Ryan, and
V. L. Peterson, 2001, Insights from rare earth elements into the genesis of the
Buck Creek complex, Clay County, NC. Southeastern Geology, v. 40, n. 3,
September, 2001, pp. 201-212.

Berner, R. A., 1978, Rate control of mineral dissolution under earth surface
conditions. American Journal of Science, v. 278, pp. 1235-1252.

Berner, RA, and Holdren, GR, 1977, Mechanism of feldspar weathering: some
observational evidence. Geology, v. 5, pp. 369-372.

176

 

Berner, R.A., and Holdren, GR, 1979, Mechanism of feldspar weathering: ll.
Observations of feldspars from soils. Geochimica et Cosmochimica Acta, v.
43, pp. 1173-1186.

Berner, R.A., and Schott, J., 1982, Mechanism of pyroxene and amphibole
weathering - ll. Observations of soil grains. American Journal of Science, v.
282, pp. 1214-1231.

Berner, R.A., Sjoberg, E.L., Velbel, MA, and Krom, MD, 1980, Dissolution of
pyroxenes and amphiboles during weathering. Science, v. 207, pp. 1205-
1206.

Bowen, N. L., and O. F. Tuttle, 1949, The system MgO—SiOz-HZO. Geological
Society of America Bulletin, v. 60, pp. 439-460.

Brantley, Susan L., Stephanie R. Crane, David A. Crerar, Roland Hellman, and
Robert Stallard, 1986, Dissolution at dislocation etch pits in quartz.
Geochimica et Cosmochimica Acta, v. 50, pp. 2349-2361.

Bream, Brendan R., Robert D. Hatcher, Jr., Calvin. F. Miller, and Paul. D.
Fullagar, 2004, Detrital zircon ages and Nd isotopic data from the southern
Appalachian crystalline core, Georgia, South Carolina, North Carolina, and
Tennessee: New provenance constraints for part of the Laurentian margin, in
Tollo, Richard T., Louise Corriveau, James M. McLelland, and Mervin J.
Bartholomew, eds, Proterozoic Tectonic Evolution of the Grenville Orogen in
North America, Geological Society of America Memoir No. 197, Boulder,
Colorado, Geological Society of America, pp. 459-475.

Brewster, D., 1837, 1853, On the optical ﬁgures produced by the disintegrated
surfaces of crystals, originally in the Transactions of the Royal Society of
Edinburgh, v. 14, having been read February 6, 1837. Re-issued in the
London, Edinburgh and Dublin Philosophical Magazine, Series. 4. v. 5. No.
29. Jan. 1853, pp. 16—28.

Brown, G. and l. Stephen, 1959, A structural study of iddingsite from New South
Wales, Australia. The American Mineralogist, v. 44, pp. 251-260.

Butler, J. R., 1972, Age of Paleozoic regional metamorphism in the Carolinas,
Georgia, and Tennessee southern Appalachians. American Journal of
Science, v. 272, pp. 319-333.

Cailliere, S., and S. Henin, 1949, The properties of saponite (bowlingite). Clay
Minerals, v. 1, pp. 138-144.

177

 

Carpenter, J. R., and D. W. Phyfer, 1969, Proposed origin of the "alpine-type"
ultramaﬁcs of the Appalachians (discussion paper). Abstracts with programs,
Geological Society of America Annual Meeting, 1969, part 7, pp. 261-263.

Carpenter, J. R., and D. W. Phyfer, 1975, Olivine compositions from southern
Appalachian ultramaﬁcs. Southeastern Geology, 1975, pp. 169-171

Carpenter, R. H., 1970, Metamorphic history of the Blue Ridge province of
Tennessee and North Carolina. Geological Society of America Bulletin, v. 81,
pp. 749-762.

Carter, N. L. and H. G. Avé Lallemant, 1970, High temperature ﬂow of dunite and
peridotite. Geological Society of America Bulletin, v. 81, pp. 2181-2202.

Casey W.H., J. F. Banﬁeld, H. R. Westrich, and L. McLaughlin, 1993, What do
dissolution experiments tell us about natural weathering? Chemical Geology,
v. 105, pp. 1-15.

Casey W. H. and C. Ludwig, 1993, Silicate mineral dissolution as a ligand-
exchange reaction. In Chemical Weathering Rates of Silicate Minerals (eds.
A. F. White and S. L. Brantley). v. 31. pp. 87—117. Mineralogical Society of
America.

Chen, Y., and S. L. Brantley, 2000, Dissolution of forsteritic olivine at 65°C and
2<pH<5. Chemical Geology, v. 165, pp. 267-281.

Choi, l., S. G. Malghan, and R. W. Smith, 1976, The dissolution kinetics of
ﬁbrous amphibole minerals in water. In Proceedings, International
Symposium on Water-rock Interaction. (Eds. J. Cadek and T. Paces, pp. 395-
403.

Clark, A. H., T. H. Pearce, P. L. Roeder, and I. Wolfson, 1986, Oscillatory zoning
and other microstructures in magmatic olivine and augite: Nomarski
interference contrast observations on etched polished surfaces. The
American Mineralogist, v. 71, pp. 734-741.

Condie, Kent 0., and James A. Madison. 1969, Compositional and volume
changes accompanying progressive serpentinization of dunites from the
Webster-Addie ultramaﬁc body, North Carolina. The American Mineralogist, v.
54, July-August 1969, pp. 1173-1179

Cronin, T. P., 1983, Petrogenesis of the Webster-Addie ultramaﬁc body, Jackson

County, North Carolina, unpublished Master’s thesis, University of
Tennessee, Knoxville, 112pp.

178

 

Daly, R. A., 1899, A comparative study of etch ﬁgures. The amphiboles and
pyroxenes. Proceedings of the American Academy of Arts and Sciences, v.
34, n. 15, PP. 374-437.

Dana, JD, 1837, A System of Mineralogy, New Haven, Durrie and Peck, 613 pp.

Dana, JD, 1849, Report of the United States Exploring Expedition During the
Years 1838, 1839, 1840, 1841, and 1842, Volume 10: Geology. Philadelphia,
C. Sherman, 756 pp..

Dana, JD, 1869, A System of Mineralogy, Fifth Edition, New York, Wiley and
Sons, 893 pp.

Daniell, J. F., 1816, On some phenomena attending the process of solution and
their application to the laws of crystallization. Quarterly Journal of Science, v.
1, 186, pp. 24-49.

Deer, W. A., R. A. Howie, and J. Zussman, 1992, An Introduction to the Rock-
forrning Minerals (2nd Ed.)

Delvigne, J., 1998, Atlas of Micromorphology of Mineral Alteration and
Weathering, The Canadian Mineralogist, Special Publication 3, 495 pp.

Delvigne, J., E.B.A. Bisdom, J. Sleeman, and G. Stoops, 1979, Olivines, their
pseudomorphs and secondary products. Pédologie, v. 29, pp. 247-309.

Doughty, D., Loehn, C., Allison, B., Newby, J. and 2001 REU Program
Participants Yurkovich, 8., Ryan, J., Peterson, V., Burr, J., and Kruse, S.,
2002, Field and geochemical comparison of the Webster-Addie and Balsam
Gap ultramaﬁc bodies, southern Blue Ridge, North Carolina. 2002 Abstracts
with Programs, Geological Society of America Southeastem/North Central
Section Meeting, v. 34, p. A-82.

Edelman, C. H., and Doeglas, D. J., 1932, Reliktstrukturen detritischer pyroxene
und amphibole. Tscherrnaks Mineral. Petrogr. Mitteilungen, v. 42, pp. 482-
490.

Edelman, C. H., and Doeglas, D. J., 1934. Uber umwandlungserscheinungen an
detritischen staurolith und anderen mineralien. Tscherrnaks Mineral. Petrogr.
Mitteilungen, v. 44, pp. 225-234.

Edwards, A. B., 1938, The formation of iddingsite. American Mineralogist, v. 23,
pp.277—281

Eggleton, R. A., 1984, Formation of iddingsite rims on olivine: a transmission
electron microscope study. Clays and Clay Minerals, v. 32, n. 1, pp. 1-11.

179

Eggleton, R. A., 1986, The relation between crystal structure and silicate
weathering rates. In Colman, S. M., and D. P. Dethier (eds), Rates of
Chemical Weathering of Rocks and Minerals, London, Academic Press, 603

pp-

Eggleton, R. A., and J. N. Boland. 1982,Weathering of enstatite to talc through a
series of transitional phases. Clays and Clay Minerals, v. 30, n. 1, pp. 11-20.

Fawoett, J. J., Alteration products of olivine and pyroxene in basalt lavas from the
Isle of Mull. The Mineralogical Magazine, v. 35, pp. 55-68

Gay, P., and R. W. LeMaitre, 1961, Some observations on "iddingsite". The
American Mineralogist, v. 46, pp. 92-1 11.

Genth, F. A., 1881, The minerals and mineral localities of North Carolina, being
Chapter 1 of the Second Volume of the Geology of North Carolina, Raleigh,
North Carolina Geological Survey.

Goldich, SS, 1938, A study in rock weathering. Journal of Geology, v. 46, pp.
17-58.

Golightly, J. P, 1979a, Nickel'rferous laterites: a general description, in Evans, D.
J. l., Shoemaker, R. S., and Veltman, H., eds., lntemational Laten'te
Symposium. New York, American Institute of Mining, Metallurgical, and
Petroleum Engineers, pp. 3-23.

Grandstaff, DE, 1978, Changes in surface area and morphology and the
mechanism of forsterite dissolution. Geochimica et Cosmochimica Acta, v. 42,
pp. 1899-1901.

Greenberg, J. K., 1976, The alpine ultramaﬁc problem in the southern
Appalachians: The Webster-Addie dunite. Geological Society of America
Abstracts with Programs, v. 8, p. 185.

Greene, J. C., and J. G. Burke, 1978, The Science of minerals in the age of
Jefferson, Transactions of the American Philosophical Society, v. 68, pt. 4,
pp. 1-1 13.

Gwinn, G. R., 1950, Dunite ring dike at Webster, North Carolina (abstract).
Geological Society of America Bulletin, v. 61, pp. 1567-1568.

Hatcher, R. D., Jr., 1978, Tectonics of the western Piedmont and Blue Ridge,

southern Appalachians: Review and speculation. American Journal of
Science, v. 278, pp. 276-304.

180

Hatcher, R. 0., Jr., 1987, Tectonics of the southern and central Appalachian
internides. Annual Reviews of Earth and Planetary Science, v. 15, pp. 337-
362.

Hatcher, R. D., Jr., A. J. Merschat, and J. R. Thigpen, 2005, Blue Ridge primer,
in Hatcher, R. D., Jr., and Merschat, A. J., eds., Blue Ridge Geology
Geotraverse East of the Great Smoky Mountains National Park, Western
North Carolina. North Carolina Geological Survey, Carolina Geological
Society Annual Field Trip Guidebook, 2005, pp. 1-24.

Hay, R.L., 1959, Origin and weathering of late Pleistocene ash deposits of St.
Vincent, B.W.l. Journal of Geology, v. 67, pp. 65-87.

Hemley, J. J., J. W. Montoya, D. R. Shaw, and R. W. Luce, 1977, Mineral
equilibria in the MgO—SiOz-HZO system - ll. Talc-antigorite-forsterite—
anthophyllite-enstatite stability relations and some geological implications in
the system. American Journal of Science, 277, pp. 353-383.

Honess, Arthur, 1921, The etching ﬁgures of topaz. The American Mineralogist,
v. 6, n. 4, pp. 71-77.

Honess, Arthur, 1927, The Nature, Origin, and Interpretation of the Etch Figures
on Crystals. New York, Wiley & Sons, 171 pp.

Horn, P., and M. Maurette, 1967, Ein Beitrag zum Atzverhalten des Olivins (in
German with English abstract). Neues Jahrbuch ft‘ir Mineralogie, Monatshefte,
v. 6, pp. 170-178.

Hotz, P., 1964, Nickeliferous Iaterites in southwest Oregon and northwest
California. Economic Geology, v. 59, no. 3, p. 356-396.

Hunter, C. E., 1941, Forsterite—olivine deposits of North Carolina, N. C., Dept. of
Cons. and Dev., Div. of. Min. Res. Bulletin, v. 41, 56 pp.

Hunter, C. E., Murdock, T. G., and MacCarthy, G. R., 1942, Chromite deposits of
North Carolina, N. C., Dept. of Cons. and Dev., Div. of. Min. Res. Bulletin, v.
42, 39 pp.

lddings, J. P., 1892, Microscopical Petrography of the Eruptive Rocks of the
Eureka District, Nevada U. S. Geological Survey Monograph No. 20,
Appendix B, pp. 337-406.

Johnson, A. C., 1958, A magnetometer survey of the Webster-Addie ultramaﬁc
ring, Jackson County, North Carolina, unpublished Master’s thesis, University
of Michigan, Ann Arbor, Michigan.

181

Julien, A. A., 1882, The dunyte-beds of North Carolina, Proceedings of the
Boston Society of Natural History, v. 22, pp. 141-149.

Kerr, W. C., 1875, The Geology of North Carolina, Raleigh.

Kirby, S. H., and Wegner, M. W., 1978, Dislocation substructure of mantle-
derived olivine as revealed by selective chemical etching and transmission
electron microscopy. Physics and Chemistry of Minerals, v. 3, pp. 309-330.

Klein, Cornelius, and Cornelius S. Hurlbut, Jr., 1993, Manual of Mineralogy (21st
Ed.)

Krauskopf, Konrad B., and Dennis K. Bird, 1995, Introduction to Geochemistry
(3rd Ed.)

Larrabee, David M., 1966, Map Showing Distribution of Ultramaﬁc and Intrusive
Maﬁc Rocks from Northern New Jersey to Eastern Alabama, U. S. Geological
Survey, Miscellaneous Publication Series l-476.

Lasaga, AC, 1984, Chemical kinetics of water-rock interactions, Journal of
Geophysical Research, v. 89, no. 6, pp. 4009-4025.

Lasaga, A.C, J. M. Soler, J. Ganor, T. E. Burch, and K. L. Nagy, 1994, Chemical
weathering rate laws and global geochemical cycles. Geochimica et
Cosmochimica Acta, v. 58, pp.2361-2386.

Lewis, J. V., 1896, Corundum and the basic magnesian rocks of western North
Carolina, Bulletin No. 11, North Carolina Geological Survey, Winston, 62 pp.

Leydolt, A., 1854, Uber eine neue Methode, die Strukturen und
Zusammensetzung der Krystalls untersuchen mit besonderer
Berticksichtigung der Varietaten des rhomboédrischen Quarzes, cited in
Honess (1927), p. 2.

Liu, Y., A. A. Olsen, and J. D. Rimstidt, 2006, Mechanism for the dissolution of
olivine series minerals in acidic solutions. The American Mineralogist, v. 91,
pp. 455-458.

Long, L. E., J. L. Kulp, and F. D. Eckelmann, 1959, Chronology of major
metamorphic events in the southeastern United States. American Journal of
Science, v. 257, pp. 585-603.

Losiak, A., 2008a, Methods description, unpublished manuscript, 8pp.

Losiak, A., 2008b, XRF_Dunites_080315_plots_indices.xls, unpublished Excel
spreadsheet.

182

Madison, James A., 1968, Petrology and Geochemistry of the Webster-Addie
Ultramaﬁc Body, Jackson County, North Carolina, unpublished Doctoral
dissertation, Washington University, St. Louis, 139pp.

Merrill, George P., 1897, A Treatise on Rocks, Rock Weathering, and Soils, New
York, Macmillan, 412 pp.

Miller, Roswell, 1951, The Structure and Petrology of the Webster-Addie
ultramaﬁc ring, Jackson County, North Carolina, unpublished doctoral
dissertation, Princeton University

Miller, Roswell, 1953, The Webster-Addie ultramaﬁc ring, Jackson County, North
Carolina, and secondary alteration of its chromite. The American Mineralogist,
v. 38, pp. 1134-1147.

Misra, Kula C., and Fred B. Keller, 1978, Ultramaﬁc bodies in the southern
Appalachians: a review. American Journal of Science, v. 278, pp. 389-418.

Miyashiro, Akiho, 1973, Metamorphism and Metamorphic Belts, New York,
Halsted Press.

Morton, A. C., and Hallsworth, C. R., 1999, Processes controlling the
composition of heavy minerals in sandstones. Sedimentary Geology, v. 124,
pp. 3-29.

Muller, R., 1877, Untersuchen ﬁber die Entwirklung des kohlensaurehalﬁgen
Wassers auf einige Mineralien und Gesteine, Tscherrnaks Mineralogishe
Mitteilungen, 1877, p. 25, cited in Merrill (1897), p. 192.

Murdock, T. G., and C. E. Hunter, 1946, The Verrniculite deposits of North
Carolina. N. C. Department of Conservation and Development, Division of
Mineral Resources Bulletin, v. 50, 45 pp.

Nahon, 0., Colin, F., and Tardy, Y., 1982, Formation and distribution of Mg, Fe,
Mn-smectites in the ﬁrst stages of weathering of forsterite and tephroite. Clay
Minerals, v. 17, pp. 339-348.

Nesbitt, H. Wayne, and O. P. Bn'cker ,1978, Low temperature alteration
processes affecting ultramaﬁc bodies. Geochimica et Cosmochimica Acta,
1978, v. 42, pp. 403-409.

Olsen, A. A. and J. D. Rimstidt, 2007, Using a mineral lifetime diagram to
evaluate the persistence of olivine on Mars. The American Mineralogist, v. 92,
pp. 598-602.

183

Page, N. J., 1967, Serpentinization considered as a constant volume
metasomatic process: a discussion. The American Mineralogist, v. 52, pp.
545-549.

Patir‘io, L.C., M. A. Velbel, J. R. Price, and J. A. Wade, 2003, Trace element
mobility during spheroidal weathering of basalts and andesites in Hawaii and
Guatemala. In: Anderson, SP. and A. E. Blum, (editors), Controls on
Chemical Weathering. Chemical Geology, v. 202, pp. 343-364.

Pliny (The Elder), Natural History, Books XXXVI and XXXVII, translated by D. E.
Eicholz, 1962. Loeb Classical Library, Cambridge, Harvard University Press,
344 pp.

Pokrovsky, OR, and Schott, J., 2000. Kinetics and mechanism of forsterite
dissolution at 25°C and pH from 1 to 12. Geochimica et Cosmochimica Acta,
v. 64, pp. 3313-3325.

Pratt, J. H., 1897, On the crystallography of the Montana sapphires, American
Journal of Science, series 4, issue 24 (Dec. 1897), pp. 424-429.

Pratt, J. H., and J. V. Lewis, 1905, Corundum and the Peridotites of Western
North Carolina, NC. Geological Survey, v. 1, 47 pp.

Quinn, M. J., 1991, Two lithotectonic boundaries in western North Carolina:
Geologic interpretation of a region surrounding Sylva, Jackson County,
unpublished Master's thesis, University of Tennessee, Knoxville, 226pp.

Rogers, W. B., and R. E. Rogers, 1848, On the decomposition and partial
solution of minerals, rocks, &c., by pure water, and water charged with
carbonic acid, American Journal of Science and Arts, series 5, issue 15 (May,
1848), pp. 401-406.

Ross, C. S. and E. V. Shannon, 1925, The origin, occurrence, composition, and
physical properties of the mineral iddingsite, Item No. 2579, Proceedings of
the U. S. National Museum, v. 67, Art. 7, pp. 1-19, pls. 1-2.

Ross, C. S., E. V. Shannon, and F. A. Gonyer, 1928, The nickel silicates at
Webster, N. C.. Economic Geology, v. 23, pp. 63-74.

Ross, C. S., E. V. Shannon, and F. A. Gonyer, 1929, Is chromite always a
magmatic segregation product? Economic Geology, v. 24, pp. 641—645.

Ross, C. S., M. D. Foster, and A. T. Myers, 1954, Origin of dunites and of olivine-

rich inclusions in basaltic rocks. The American Mineralogist, v. 39, pp.693—
737.

184

Ros

Rya

Sam

'53

Rosso, J. J., and J. D. Rimstidt, 2000, A high resolution study of forsterite
dissolution rates. Geochimica et Cosmochimica Acta, v. 64, n. 5 pp. 797-811.

Ryan, Jeffrey 0., Steven Yurkovich, Virginia Peterson, Jonathan Burr, and Sarah
Kruse, 2005 Geology and petrogenesis of maﬁc and ultramaﬁc rocks of the
Willets-Addie area, central Blue Ridge, NC, in Hatcher, R. 0., Jr., and
Merschat, A. J., eds., Blue Ridge Geology Geotraverse East of the Great
Smoky Mountains National Park, Western North Carolina. North Carolina
Geological Survey, Carolina Geological Society Annual Field Trip Guidebook,
2005, pp.91-98.

Sangwal, K., 1987, Etching of Crystals: Theory, Experiment, and Application,
Amsterdam, North-Holland, 487 pp.

Santelli, C.M., S. A. Welch, H. R. Westrich, and J. F. Banﬁeld, 2001. The effect of
Fe-oxidizing bacteria on Fe-silicate mineral dissolution: Chemical Geology, v.
180, PP. 99-1 15.

Seyama, H., M. Soma, and A. Tanaka, 1996. Surface characterization of acid-
leached olivines by X-ray photoelectron spectroscopy. Chemical Geology, v.
129, pp. 209-216.

Shepard, C. U., 1872, On the corundum region of North Carolina and Georgia,
with descriptions of two gigantic crystals of that species. American Journal of
Sciences and Arts, Aug 1872, series 4, issue 20, pp. 109-117.

Sheppard, R. A., 1962, lddingsitization and recurrent crystallization of olivine in
basalts from the Simcoe Mountains, Washington. American Journal of
Science, v. 260, pp. 67-74.

Siever, R., and N. Woodford, Dissolution kinetics and the weathering of maﬁc
minerals. Geochimica et Cosmochimica Acta, v. 43, pp.717-724.

Smith, K.L., Milnes, AR, and Eggleton, R.A., 1987. Weathering of basalt:
Formation of iddingsite. Clays and Clay Minerals, v. 35, pp. 418-428.

Stoops, G., H.-J, Altemiiller, E. B. A. Bisdom, J. Delvigne, V. V. Dobrovolsky, E.
A. Fitzpatrick, G. Paneque, and J. Sleeman, 1979, Guidelines for the
description of mineral alterations in soil morphology. Pédologie, v. 29, n. 1,
pp. 121-135.

Stopar, J. D., G. J. Taylor, V. E. Hamilton, and L. Browning, 2006, Kinetic model

of olivine dissolution and extent of aqueous alteration on Mars. Geochimica et
Cosmochimica Acta, v. 70, pp. 6136-6152.

185

Su

Sw

Tr

Sun, Ming-shan, 1957, The nature of iddingsite in some basaltic rocks of New
Mexico. The American Mineralogist, v. 42, pp. 525-533.

Swanson, S. E., 2001, Ultramaﬁc rocks of the Spruce Pine area, western North
Carolina: A sensitive guide to ﬂuid migration and metamorphism.
Southeastern Geology, v. 40, no. 3, pp. 163-182.

Swanson, S. E., L. A. Raymond, R. D. Warner, J. G. Ryan, S. P. Yurkovich, and
V. l. Peterson, 2005, Petrotectonics of maﬁc and ultramaﬁc rocks in Blue
Ridge terranes of western North Carolina and northern Georgia, in Hatcher,
R. B, Jr., and Merschat, A. J., eds., Blue Ridge Geology Geotraverse East of
the Great Smoky Mountains National Park, Western North Carolina. North
Carolina Geological Survey, Carolina Geological Society Annual Field Trip
Guidebook, 2005, pp. 99-112.

Trescases, Jean-Jacques, 1975, L’évolution Géochimique Supergene des
Roches Ultrabasiques en Zone Tropicale, Paris, O.R.S.T.O.M., Memoir No.
78, 259 pp.

Velbel, MA, 1984. Natural weathering mechanisms of almandine garnet.
Geology, v. 12, pp. 631-634.

Velbel, MA, 1986. Inﬂuence of surface area, surface characteristics, and
solution composition on feldspar weathering rates. In: Davis, J.A., and Hayes,
K.F. (editors), Geochemical Processes at Mineral Surfaces, American
Chemical Society, Symposium Series, No. 323, pp. 615-634.

Velbel, M.A., 1993a. Formation of protective surface layers during silicate-
mineral weathering under well-leached, oxidizing conditions. American
Mineralogist, v. 78, pp. 408-417.

Velbel, M. A. 1993b, Constancy of silicate-mineral weathering-rate ratios
between natural and experimental weathering. Chemical Geology, v.105, pp.
89-99.

Velbel, MA, 1999. Bond strength and the relative weathering rates of simple
orthosilicates. American Journal of Science, v. 299 (Berner volume), pp. 679-
696.

Velbel, M. A. 2004, Laboratory and homework exercises in the geochemical
kinetics of mineral-water interaction: rate law, Arrhenius activation energy,
and the rate-determining step in the dissolution of halite. Journal of
Geoscience Education, v. 52, no. 1, pp. 52-59.

Velbel, MA, 2007. Surface Textures and Dissolution Processes of Heavy
Minerals in the Sedimentary Cycle: Examples from Pyroxenes and

186

Amphiboles. In Mange, M., and Wright, D., eds, Heavy Minerals in Use.
Developments in Sedimentology v. 58, pp. 113-150.

Velbel, MA, 2009. Dissolution of olivine during natural weathering, Geochimica
et Cosmochimica Acta, v. 73, pp. 6098-61 13.

Velbel, M.A., McGuire, J.T., and Madden, AS, 2007. Scanning electron
microscopy of garnet from southern Michigan soils: Etching rates and
inheritance of pre—glacial and pre-pedogenic grain-surface textures. In Mange,
M., and Wright, D., eds, Heavy Minerals in Use. Developments in
Sedimentology v. 58, pp. 413-432. I

Velbel, MA, and J. M. Ranck, 2009. Etch pits on naturally altered olivine from
dunites of the Appalachian Blue Ridge Mountains, North Carolina, USA,
Mineralogical Magazine, vol. 72(1), pp. 149-152. I.

 

Warner, R. D., and C. W. Helper, 2005, Petrology of the Dark Ridge
metaultramaﬁc body, North Carolina, in Hatcher, R. D., Jr., and Merschat, A.
J., eds, Blue Ridge Geology Geotraverse East of the Great Smoky
Mountains National Park, Western North Carolina. North Carolina Geological
Survey, Carolina Geological Society Annual Field Trip Guidebook, 2005,
pp.99-1 12.

Warner, R. D., and S. P. Yurkovich, 2005, Field trip stop description in Hatcher,
R. 0., Jr., and Merschat, A. J., eds, Blue Ridge Geology Geotraverse East of
the Great Smoky Mountains National Park, Western North Carolina. North
Carolina Geological Survey, Carolina Geological Society Annual Field Trip
Guidebook, 2005, pp.129-130.

Wegner, M. W., and J. M. Christie, 1974, Preferential chemical etching of
terrestrial and lunar olivines, Contributions to Mineralogy and Petrology, v. 43,
pp. 195—212.

Wegner, M. W., and J. M. Christie, 1976, Chemical etching of dislocations in
forsterite. Contributions to Mineralogy and Petrology, v. 59, pp. 131-140.

Wegner, M. W., and J. M. Christie, 1985a, Chemical etching of amphiboles and
pyroxenes. Physics and Chemistry of Minerals, v. 12, pp. 86-89.

Wegner, M. W., and J. M. Christie, 1985b, General chemical etchants for
microstructures and defects in silicates. Physics and Chemistry of Minerals, v.
12, pp. 90-92.

Welch, SA, and J. F. Banﬁeld, 2002. Modiﬁcation of olivine surface morphology

and reactivity by microbial activity during chemical weathering. Geochimica et
Cosmochimica Acta, v. 66, pp. 213-221.

187

 

 

Wilshire, H G.,1958, Alteration of olivine and orthopyroxene in basic lavas and
shallow intrusions. American Mineralogist, vol.43, no.1-2, pp.120-147.

leon, M. J., 2004, Weathering of primary rock-forming minerals: processes,
products, and rates. Clay Minerals, v. 39, pp. 233-266.

Wogelius, R. A., and Walther, J. V., 1991, Olivine dissolution at 25°C: Effects of

C02 and organic acids. Geochimica et Cosmochimica Acta, v. 55, Issue 4,
pp. 943-954.

 

Wogelius, R. A., and Walther, J. V., 1992, Olivine kinetics at near surface 1'
conditions. Chemical Geology, v. 97, pp. 101-1 12.

Young, C. III, 1969, Dislocations in the deformation of olivine. American Journal
of Science, v. 267, pp. 841-852. R

Zirkel, Ferdinand, 1876, Microscopical Petrography: Report of the Geological
Exploration of the Fortieth Parallel, V. 4, Professional Papers of the Engineer
Department, U. 8., Army, No. 18, Washington, US. Government Printing
Ofﬁce.

 

188

 

llllllllllll '

70

. .