225
Louisiana State University LSU Digital Commons LSU Historical Dissertations and eses Graduate School 1990 Petrogenesis and Provenance of Epiclastic Volcanic Cobbles From the Cretaceous Woodbine Formation, Southwest Arkansas. Christina Lee Livesey Louisiana State University and Agricultural & Mechanical College Follow this and additional works at: hps://digitalcommons.lsu.edu/gradschool_disstheses is Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and eses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. Recommended Citation Livesey, Christina Lee, "Petrogenesis and Provenance of Epiclastic Volcanic Cobbles From the Cretaceous Woodbine Formation, Southwest Arkansas." (1990). LSU Historical Dissertations and eses. 4933. hps://digitalcommons.lsu.edu/gradschool_disstheses/4933

Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

  • Upload
    others

  • View
    3

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Louisiana State UniversityLSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

1990

Petrogenesis and Provenance of Epiclastic VolcanicCobbles From the Cretaceous WoodbineFormation, Southwest Arkansas.Christina Lee LiveseyLouisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion inLSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please [email protected].

Recommended CitationLivesey, Christina Lee, "Petrogenesis and Provenance of Epiclastic Volcanic Cobbles From the Cretaceous Woodbine Formation,Southwest Arkansas." (1990). LSU Historical Dissertations and Theses. 4933.https://digitalcommons.lsu.edu/gradschool_disstheses/4933

Page 2: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

INFORMATION TO USERS

T he m ost advanced technology has been used to p h o to g rap h and reproduce this manuscript from the microfilm m aster. U M I films the

text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type o f com puter printer.

The quality o f th is reproduction is dependent upon the quality o f the copy subm itted . B roken or indistinct print, colored o r p o o r quality illustrations and photographs, print bleedthrough, substandard margins, and im proper alignment can adversely affect reproduction.

In the unlikely event tha t th e au thor did not send U M I a com plete

m anuscript and there are missing pages, these will be noted. Also, if unauthorized copyright m aterial had to be removed, a note will indicate the deletion.

O versize m aterials (e.g., maps, drawings, charts) are rep roduced by sectioning the original, beginning a t the upper left-hand co rner and continuing from left to right in equal sections with small overlaps. Each o rig ina l is also p h o to g rap h e d in one exposure and is in c lu d ed in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. H igher quality 6" x 9" b lack and white

photographic prints are available for any photographs o r illustrations appearing in this copy for an additional charge. Contact U M I directly to order.

UMIUniversity M icrofilms International

A Bell & H ow ell Information C om p an y 3 0 0 North Z e e b R oad , Ann Arbor, Ml 4 8 1 0 6 -1 3 4 6 USA

3 1 3 /7 6 1 -4 7 0 0 8 0 0 /5 2 1 -0 6 0 0

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

Page 3: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

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

Page 4: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Order Number 9104152

Petrogenesis and provenance of epiclastic volcanic cobbles from the Cretaceous Woodbine Formation, southwest Arkansas

Livesey, C hristina Lee, Ph.D .

The Louisiana State University and Agricultural and Mechanical Col., 1990

UMI300 N. Zeeb Rd.Ann Arbor, MI 48106

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

Page 5: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

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

Page 6: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

from the Cretaceous Woodbine Formation,

Southwest Arkansas

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and

Agricultural and Mechanical College in partial fulfillment of the

requirements for the degree of

Doctor o f Philosophy

m

The Department of Geology and Geophysics

by

Christina Lee LiveseyB.S., University o f Iowa, 1983

M.S., University o f Kansas, 1988

May, 1990

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

Page 7: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

A cknow ledgm ents

I'd like to thank my major advisor, Gary Byerly, for introducing me to the Upper

Cretaceous igneous rocks of the northern Gulf of Mexico, and for filling in some enormous

gaps in my background in the field of igneous petrology; Gary proved to be an outstanding

scientist, a fine person, and a good friend to me.

I'd like to thank my minor advisor, Ajoy Baksi, for providing valuable

geochronologic data to the project and for providing many interesting discussions.

I'd like to acknowledge my remaining committee members, Roy Dokka, Stan

Williams, and Darrell Henry for support and council throughout this project. Each, in his

own way rendered help and advice to me at times when it was needed. Professor Dokka

aided with mineral separations and helped me improve my basic mapping skills at the LSU

field camp. Professor Williams taught me a little bit about volcanology and a lot about the

politics of life. Professor Henry helped me understand and utilize the electron microprobe,

and provided many interesting discussions about a wide variety of subjects within geology.

I'd like to thank Dr. Steve Nelson at Tulane University for very generously

providing XRF data for the Woodbine volcanic cobbles.

Many good friends helped me throughout this project. I'd like to acknowledge

Megan Jones, Kart Kempter, and Kathy McManus for support and friendship along the

way.

My parents, Robert and Margaret Livesey, provided love and fnendship; I

recognize that my accomplishments are, in part, a reflection of the support they have

always given me. My brother, Todd, generously spent ten, cold days in the field with me,

and has always been one of my best fiiends.

My husband, Steve Jones, helped me to realize what life is really all about.

This dissertation was funded by a GSA research grant to Christina Livesey, and by

fellowship monies provided by Shell Oil, and by EXXON.

I I

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

Page 8: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

A cknow ledgm ents........................................................................................................... ii

Table o f C ontents..........................................................................................................iii

L ist o f T ables..................................................................................................................vi

L ist o f F igures...................................................... v ii

A b stra c t................................................................................................................................. ix

Chapter 1. Introduction...........................................................................................................1

1. Introduction and Objectives.................................................................................. 1

2. Geologic S etting ............................................................................................4

A. W oodbine Form ation........................................................................4

B. Cretaceous Igneous Rocks of the Gulf of Mexico B asin ...................... 10

Chapter 2. Methods............................................................................................................... 16

1. Sample Collection...................................................................................................16

2. Petrology.................................................................................................................18

3. Volcanic Whole Rock Geochemistry........................................................ 18

4. Mineral and Melt Inclusion Compositions................................................18

5. Geochronology......................................................................................................19

6. Petrogenesis.................................................................................................. 19

a) Major and Minor Element Modeling.............................................. 19

b) Trace Element Modeling...........................................................................20

Chapter 3. Petrology, Whole Rock Chemistry, and Mineral Chemistry..........................21

1. Petrology.................................................................................................................21

2. Whole Rock Chemistry......................................................................................... 27

A. Whole rock major element chemistry.............................................27

B. Whole rock trace element chemistry........................................................ 41

C. Whole rock chemistry and mineralogy...................................................41

3. Mineral chemistry..................................................................................................44

A. Clinopyroxenes....................................................................................... 44

111

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

Page 9: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

B. Feldspars.................................................................................................. 49

C. A m phiboles........................................................................................... 57

E. Melt Inclusions.........................................................................................62

4. Geochronology.....................................................................................................67

Chapter 4. Summary............................................................................................................. 78

1. Lithologie and Mineral Data................................................................................ 78

2. Mineral/melt Partition Coefficients..................................................................... 81

3. Geochronologic Summary..................................................................................85

Chapter 5. Petrogenesis of the Volcanic Suite Preserved as Cobbles...............................89

1. Are the cobbles cogenetic?.................................................................................. 92

2 Minéralogie and Chemical Variation..................................................................... 96

A. Partial M elting....................................................................................99

B. Magma Mixing......................................................................................... 99

C. Fractional Crystallization.........................................................................100

i) Major Element Models.................................................................. 103

ii) Trace Element Models.................................................................. 122

D. Assimilation and Fractional Crystallization............................................127

3. Sources o f E rror...........................................................................................129

4. Petrogenetic Differentiation Summary................................................................. 132

Chapter 6. Provenance of Epiclastic Volcanic Cobbles......................................................136

1. Similarities between Woodbine volcanic rocks and the Benton Dikes..............136

2. Characteristics of lamprophyres........................................................................... 138

3. Origin o f lamprophyric magmas..................................................................140

4. Benton Dike/Woodbine volcanic petrogenesis.....................................142

5. Comparison with other lamprophyres................................................................. 146

6. Summary.................................................................................................................147

I V

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

Page 10: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Chapter 7. Tectonic Implications of K-Ar and 40Ar/39Ar Ages of Igneous Rocks

of the Northern Gulf of Mexico Basin................................................................................... 149

1. Introduction............................................................................................................149

2. A nalytical Techniques................................................................................ 150

3. Geochronologic Results........................................................................................151

4. D iscussion ........................................................................................................153

5. Timing of the Cretaceous Igneous Activity in the Northern Gulf of

Mexico B asin...............................................................................................................154

6. Middle Cretaceous Volcanic Episodes................................................................ 156

7. Provenance of Volcaniclastic Sediments o f the Northern Gulf Basin...............157

8. Summary................................................................................................................ 159

Chapter 8. Conclusions.......................................................................................................... 161

R eferen ces .............................................................................................................................165

A ppendix 1..........................................................................................................................177

A ppendix 2 ......................................................................................................................... 186

A ppendix 3 ..........................................................................................................................191

A ppendix 4 ......................................................................................................................... 194

A ppendix 5 ......................................................................................................................... 197

A ppendix 6 ..................................................................................................... 202

A ppendix 7 .........................................................................................................................205

A ppendix 8 .........................................................................................................................208

V ita ..........................................................................................................................................209

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

Page 11: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

LIST OF TABLES

Table 1. Whole rock major and trace element compositions.

Table 2. Selected Woodbine volcanic cobble clinopyroxene compositions.

Table 3. Selected Woodbine volcanic cobble feldspar compositions.

Table 4. Selected Woodbine volcanic cobble amphibole compositions.

Table 5. Selected Woodbine volcanic cobble sphene compositions.

Table 6. Detrital Sanidine melt inclusion compositions.

Table 7. Results of Argon geochronologic data.

Table 8. Summary of mineral compositions and volcanic rock types.

Table 9. Summary of major element modeling of segment 1.

Table 10. Summary of major element modeling of segment 2.

Table 11. Summary of major element modeling of segment 3.

Table 12. Calculation of bulk distribution coefficients.

Table 13. Summary of trace element models.

VI

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

Page 12: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

LIST OF FIGU RES

Figure 1. Map showing locations of Upper Cretaceous igneous rocks of the northern Gulf

of Mexico Basin.

Figure 2, Depositional Environments of the Woodbine Formation.

Figure 3, Generalized northern Gulf of Mexico Cretaceous stratigraphie relationships.

Figure 4. Summary of radiometric ages for Cretaceous igneous rocks of the northern Gulf

of Mexico Basin.

Figure 5. Woodbine volcanic cobble sample location map.

Figure 6. Photomicrograph of a typical graywacke of the Woodbine Formation.

Figure 7. Woodbine volcanic cobble photomicrographs.

Figure 8. Major and trace element variation diagrams for volcanic cobbles of the Woodbine

Formation.

Figure 9. Whole rock variation diagrams showing the segmented nature of the geochemical

variation of the volcanic rocks of the Woodbine Formation.

Figure 10. Variation in mineralogy and in Sr and MgO concentrations of volcanic cobbles

from the Woodbine Formation.

Figure 11. Selected clinopyroxene compositions plotted on a Wo-En-Fs ternary diagram.

Figure 12. Selected clinopyroxene compositions plotted on a Ti02-Na20-M n0 ternary

diagram.

Figure 13. Feldspar compositions plotted on an An-Ab-Or temploL

v ii

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

Page 13: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Figure 14. Photomicrographs of detrital sanidine melt inclusions.

Figure 15. Mechanisms for trapping melt inclusions.

Figure 16. Variation diagrams of sanidine melt inclusion compositions.

Figure 17. Variation diagrams of melt inclusion and whole rock compositions.

Figure 18. 40Ar/39Ar two-step spectra for subsurface samples.

Figure 19. Summary of radiometric ages for Cretaceous igneous rocks of the northern

Gulf of Mexico Basin.

Figure 20. Whole rock geochemistry of rocks from several Gulf Coast igneous centers.

Figure 21. Idealized whole rock variation diagrams.

Figure 22. Sr and Y vs. MgO variation diagrams and relevant partition coefficients.

Figure 23. Model residual liquid compositions after removal of single mineral phases.

Figure 24. Composition of modeled magma evolution for segment 1.

Figure 25. Composition of modeled magma evolution for segment 2.

Figure 26. Composition of modeled magma evolution for segment 3.

Figure 27. Comparison of radiometric ages of a trachyte preserved as a cobble in the

Woodbine Formation and other northern Gulf of Mexico Basin igneous rocks.

Figure 28. Schematic amphibole stability P-T diagram.

Figure 29. Summary of radiometric ages for Cretaceous igneous rocks o f the northern

Gulf of Mexico Basin

vi i i

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

Page 14: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

ABSTRACT

Whole rock and mineral compositions of epiclastic volcanic cobbles have been

used to reconstruct the evolution of the magmatic system that supplied this volcanic

material. Using techniques that have been applied to stratified volcanic deposits, the

chemical and mineralogical variation of volcanic material preserved as cobbles in

sedimentary rocks have been used to identify a cogenetic suite of igneous rocks and to

evaluate possible fractionation mechanisms.

Volcanic rocks, preserved as rounded cobbles in the upper Cretaceous Woodbine

Formation of SW Arkansas, range from mafic, pyroxene-rich phonolites to felsic,

crystal-poor phonolites. Felsic lithologies contain abundant sanidine phenocrysts and

lesser amounts of calcic pyroxene and hornblende; sphene is an important minor phase.

Intermediate lithologies contain subequal proportions of anorthoclase and pyroxene

phenocrysts. Mafic lithologies contain only pyroxene phenocrysts.

Discrimination diagrams of whole rock major and trace element concentrations

reveal smooth compositional gradients. MgO contents vary from 6.0 wt.% in the most

mafic rocks to 0.5 wt.% in felsic cobbles. K2O and Si02 are enriched in the felsic rocks;

K2O ranges from 3.1 to 8.0 wt.%, and Si02 ranges from 45.2 to 61.5 wt.% from mafic

to felsic rocks, respectively. Compatible trace elements (Ni, Cu) are enriched in the

mafic lithologies while incompatible elements (Zr, Rb) are enriched in the felsic samples.

Electron microprobe analyses o f phenocryst phases from different lithologies reveal the

following trends; relative to mafic rocks, felsic samples contain 1) felspars rich in Na, K,

Ba and depleted in Fe, Sr, and Ca, and 2) pyroxenes rich in Mn, Fe, Si, and depleted in

T i . Electron microprobe analysis of melt inclusions in sanidine show smooth

compositional gradients consistent with whole rock data.

Modeling of whole rock and mineral compositions indicate that these volcanic

lithologies preserved as cobbles are related to each other by fractional crystallization;

IX

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

Page 15: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

felsic lithologies represent the more differentiated level of a magma chamber while mafic

samples represent relatively primitive magmas. Melt inclusion compositions have been

used with whole rock data to trace a liquid line of descent of the evolving magma.

Potential igneous source rocks are present throughout the Gulf Coast.

Geochemical and geochronologic data from rocks of these various igneous centers have

been compared with similar data of the Woodbine volcanics; rock compositions of the

Woodbine volcanics compare favorably to the lamprophyric dike swarm associated with

Magnet Cove in central Arkansas.

Eight new radiometric ages have been determined for Cretaceous igneous rocks of

the northern Gulf of Mexico Basin. Well cuttings of igneous rocks from the Monroe Uplift

(west Mississippi, east Louisiana) and from the Jackson Dome (central Mississippi)

provided relatively small sample sizes of ~50 mg that were utilized for 40Ar/39Ar dating.

A two-step degassing technique was the compromise between the stepwise (7-10

increments), which requires a larger sample size, and the total fusion

techniques. The usefulness of this two-step technique was clearly demonstrated.

These ages, combined with earlier geochronologic work, suggest that there is a

relationship between the timing and the east-west location of magmatic activity. Magmatic

activity migrated along a curvilinear trend that begins at the Prairie Creek lamproite in SW

Arkansas (100-110 Ma) and extends northeast to Magnet Cove (95-105 Ma) and Granite

Mountain (-90 Ma) Complexes, and then extends southeast to volcanic rocks preserved on

the Monroe Uplift (-80 Ma) and Jackson Dome (-70 Ma). Temporal changes in the

intraplate stress field may have resulted in a migration of crustal extension over the period

of 110-65 Ma. These extensional stresses apparently resulted in melting at the base of the

lithosphere and subsequent time transgressive volcanism at the surface. Upper Cretaceous

global plate reorganizations may have been responsible for these Gulf Basin extensional

stresses.

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

Page 16: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Chapter 1. Introduction1. Introduction and Objectives

This study focuses on the use of epiclastic volcanic material in reconstructing the

petrogenesis of a volcanic suite and the provenance of the corresponding reworked volcanic

rocks. The objectives of this study are two-fold: 1) to demonstrate that epiclastic, volcanic,

sedimentary debris can be used to reconstruct the evolution of the magmatic system that

supplied the volcanic debris, and 2) to introduce a new approach to the question of

provenance of reworked volcanic material. Specifically, the study involved examination of

the reworked volcanic rocks now preserved as cobbles and sand grains within the Upper

Cretaceous Woodbine Formation in order to understand the evolution o f the magmatic body

that erupted to produce the volcanic rocks, and to identify which of the igneous complexes

of the northern Gulf of Mexico Basin was the source of the volcaniclastic material in the

Upper Cretaceous Woodbine Formation.

The Woodbine Formation of the northern Gulf of Mexico Basin was chosen as the

volcaniclastic-rich sedimentary unit to examine because of the wide variety of very fresh

volcanic lithologies that are preserved as sand grains and as cobbles within this formation.

The presence of local, distinct igneous complexes that are potential source areas for the

reworked volcanic material made this unit ideal for a provenance study. Although several

studies have examined the field and pétrographie characteristics of the volcanic material in

the Woodbine, this study used a geochemical approach in an attempt to more fully

understand the petrogenesis and provenance of these volcanic rocks.

Widespread volcanism occurred throughout the northern Gulf of Mexico Basin

during the Late Mesozoic (Byerly, 1989) (Figure 1). This igneous activity produced a

variety o f lithologies including felsic (trachytes, syenite, phonolites), mafic (basalts), and

ultramafic (lamprophyres, nephelinites, and lamproites) rock types. This widespread,

although volumetrically minor, igneous activity provided abundant source material for

sedimentary units such as the Woodbine Formation within the Gulf Basin.

1

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

Page 17: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Prairie Creek Lamproite

Woodbine

Conglomerate

Baicones Province

Magnet Cove _ extent oflamprophere dikes

/ Granite Mountain

wonroe Upliftn Dome

Mexico

100 200 Mies— I____________I

Figure 1. Map showing locations of Upper Cretaceous igneous rocks of the northern Gulf

of Mexico Basin.

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

Page 18: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

It has been demonstrated that the extrusive deposits of evolved magma chambers

commonly exhibit systematic mineralogical and chemical variations that reflect the pre­

emption state of the magma (Smith and Bailey, 1965; Hildreth, 1979; Womer and

Schmincke, 1984 a,b). Continuous chemical variations in stratified, undisturbed volcanic

deposits provide evidence that such deposits are cogenetic. Single emptive units can be

used to represent a pre-emptive chemical state of a magma chamber and successive emptive

units can be used to reconstmct the evolution of a chamber through time or space.

Using techniques that have been applied to undisturbed volcanic stiuta, whole rock

and mineralogical chemical variations of volcanic material preserved as detrital grains in

sedimentary rocks have been used to identify a cogenetic suite of volcanic rocks and to

evaluate fractionation mechanisms. These techniques involve identification of the most

primitive volcanic rock, which represents the most primitive melt, and modeling whether

the more differentiated samples are related to the primitive sample by different degrees of

mantle partial melting, fractional crystallization, magma-mixing, or assimilation and

fractional crystallization (AFC), etc. Petrogenetic modeling involves using whole rock

geochemistry of the volcanic rocks and electron microprobe analysis of phenocrysts. Since

a volcanic stratigraphy is lacking in a study involving reworked volcanic rocks, comparison

with well understood, stratified volcanic deposits of similar composition will allow a more

complete interpretation concerning the magmatic evolution of epiclastic volcanic material.

Stratified volcanic deposits, representing magmatic systems of similar composition to the

reworked material, record the eruption sequence and the mineralogical and geochemical

evolution of that system.

Pétrographie and geochemical analysis of the epiclastic volcanic material can also be

used to identify the provenance of the sedimentary material. Whereas previous provenance

studies have relied on identification of detrital minerals present and determination of the

relative proportions of those minerals to identify source areas, this study invokes a more

quantitative approach. Geochemical and mineralogical examination of potential igneous

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

Page 19: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

source rocks and the reworked volcanic material allows a more precise identification of the

provenance. Careful comparison of whole rock geochemical trends of the reworked

volcanic rocks and of local igneous rocks may identify which of the potential provenance

centers is petrologically related to the volcanic sedimentary debris.

Geochronologic data provide important information concerning the provenance of

the epiclastic material. Precise, radiometric age determinations of the epiclastic material and

the potential igneous sources eliminate any igneous centers younger than the sediments.

Direct comparison of radiometric ages of this volcanic debris and igneous centers may

narrow provenance choices.

Combining all of the information from the detrital epiclastic volcanic material

provides new insight concerning the Cretaceous igneous activity of the northern Gulf of

Mexico Basin. The tectonic significance of these igneous rocks is poorly understood.

Geochemical and geochronologic data from the volcanic rocks preserved as sediments will

be added to the body of data from the Gulf o f Mexico Basin igneous rocks; this combined

information will be used to form a better understanding of source and magmatic history of

the widespread Cretaceous, Gulf of Mexico Basin igneous rocks.

2. Geologic Setting

A. Woodbine Formation

The Upper Cretaceous Woodbine Formation is composed largely of terrigenous

material derived from weakly metamorphosed Paleozoic rocks of the Ouachita Mountains

(Figure 1). Rocks of the Woodbine Formation were first described by Hill (1888) who

regarded them as Eocene in age and gave the name Bingen sand to those rocks in Arkansas

now known as the Woodbine and Tokio formations; Hill (1901) gave the name Woodbine

to similar, Cretaceous rocks he described in northeast Texas. Miser and Purdue (1919)

recognized an unconformity within the Bingen sand in southwest Arkansas, and divided

this unit into the Woodbine and Tokio Formations. Hazzard (1939) proposed the name

Centerpoint volcanics to replace the name Woodbine Formation in southwest Arkansas,

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

Page 20: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

and considered these sediments to be stratigraphically equivalent to the younger Eagleford

shale of northeast Texas; Bailey and others (1945) concluded that the Centerpoint

Formation is likely correlative with the Woodbine Formation of east Texas. Since the

nomenclature of Hazzard (1939) confuses an already complicated stratigraphy, the

formation name Woodbine, as mapped by the United States Geological Survey (State Map

of Arkansas), is used here. The Woodbine Formation is now considered equivalent, at

least in part, to clastic rocks of the Tuscaloosa Formation of the central and eastern Gulf

Coast (Murray, 1961), and to the Pepper shale o f south-central and south Texas (Adkins

and Lozo, 1951; Murray, 1961).

In the U.S. Gulf Coast, the Cretaceous is commonly divided into three provincial

Series, the Coahuilan, Comanchean, and Gulfian (Murray, 1961). The Woodbine

Formation is the lowermost unit of the Gulfian Series and may mark the beginning of a

major transgression (Dane, 1929; Adkins and Lozo, 1951) or may reflect subsidence of the

Northeast Texas Basin that was followed by a transgression and deposition of the

Eagleford shale (Oliver, 1970). In Arkansas, the boundary between Gulfian and

Comanchean rocks is delineated by a pronounced unconformity which coincides with the

94 Ma sequence boundary of Haq and others (1988) (pers comm. J. Hazel, 1989). The

rocks of the Woodbine Formation are considered Cenomanian in age (Mancini, 1979).

The Woodbine Formation of Arkansas, Oklahoma, and Texas is lithologically

diverse; poorly sorted conglomerates, sandstones, siltstones and shales are interbedded

and are areally adjacent. During deposition of the Woodbine, sediments shed from the

Ouachita Mountains were deposited in nearshore environments of the Cretaceous Gulf

Coast (Dane, 1929; Cotera, 1956; Oliver, 1970; Funkhouser et al., 1980) (Figure 2). Dane

(1929) constructed a model of the environment o f deposition for the Woodbine sediments.

He interpreted rounded pebbles within the Woodbine as evidence that these sediments were

waterlain, and he thought that the presence of cross-beds indicated strong currents.

Sedimentary glauconite and local calcite cement of the Woodbine suggested to Dane (1929)

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

Page 21: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Ouachita Mountains

Monroe Uplift

Sabine Uplift

L

Figure 2. Depositional Environments of the Woodbine Formation.

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

Page 22: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

that the clays, sands, and gravels were deposited in a marine environment, whereas the

irregular unconformity at the base of the Woodbine and the presence of plant material

indicate subaerial or near-subaerial deposition. Dane (1929) thought that most of the

formation may have been deposited in a shallow marine environment, but that some of the

materials, most notable the red clays and coarse gravels of the northeastern Woodbine may

indicate subaerial deposition. Based on lithology, sedimentary structures, and fossil

distribution in outcrop and subsurface samples, Oliver (1970) refined Dane's 1929

interpretation of the Woodbine depositional environment; he identified three main

depositional systems in the Woodbine Formation; 1) a fluvial system in the northern and

eastern limits of the formation, 2) a deltaic system with channel mouth bar sands and

coastal barrier sands to the south and, 3) a shelf-strandplain system of shelf muds and

strandplain sands in the upper Woodbine (Figure 2).

The Woodbine Formation has been divided into two members in northeast Texas; the

lower Dexter and upper Lewisville members are delineated by lithology and sedimentary

structures (Murray, 1961). The Dexter member is composed of massive, porous sands

free of volcanic debris (Ifill, 1901) which are interpreted as fluvial deposits (Murray, 1961;

Oliver, 1970) whereas the Lewisville member is composed of calcareous sandstone and

shale that is locally rich in volcanic material (Hill, 1901; Cotera, 1956; Murray, 1961)

which was deposited along the Cretaceous shelf (Dane, 1929; Cotera, 1956; Oliver, 1970).

The Woodbine of Arkansas is different to that in Texas; in Arkansas, Bailey et al., (1945)

found no evidence of an unconformity within the Woodbine and therefore did not divide

the formation into members. The Woodbine in Arkansas may correlate to the Lewisville

(upper) member of the Woodbine in northeast Texas (Hazzard, 1939); both the Woodbine

in Arkansas and the Lewisville member of northeast Texas are noted for being rich in

volcaniclastic material.

The length of time represented by the unconformity at the base of the Woodbine

becomes greater to the east (Dane, 1929; Ross et al., 1928); in Texas, the formation lies

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

Page 23: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

8

conformably on beds of the Washita Series and is conformably overlain by rocks of the

Austin Formation. The upper Washita rocks present in east Texas are not present in

Arkansas (Dane, 1929) (Figure 3). In Arkansas, the Woodbine Formation is bounded

above and below by unconformities and thins to the east (Figure 3). The break between the

Comanchean and Gulfian Series involved a warping or tilting toward the west (Dane, 1929;

Belk, 1984,1986). Pre-Woodbine erosion bevelled the uplifted rocks and produced a

slight angular unconformity, not obvious in single outcrops, at the base of the Woodbine

(Figure 3) (Dane, 1929).

The fluvial deposits within the Woodbine crop out in southwest Arkansas,

southeast Oklahoma, and northeast Texas. The outcrop area is a 150 mile (241 km) long

belt extending from Pike County, Arkansas west through Howard and Sevier Counties

Arkansas, M^Curtain, Choctaw, and Bryan Counties, Oklahoma, and Red River, Lamar,

Fannin, and Grayson Counties Texas (Figure 1). The formation is 500-625 feet (152-190

m) thick in NE Texas and thins to the east (Figure 3); it is 250-350 feet (76-107 m) thick in

Arkansas (Ross et al., 1928). The formation consists largely of quartz sand in northeastern

Texas and in Bryan and Choctaw Counties in Oklahoma. However, epiclastic volcanic

material is present in the Woodbine of Texas, Oklahoma, Arkansas and increases

dramatically in concentration to the north and east (Cotera, 1956; Dane, 1929; Ross et al.,

1928). Reworked volcanic material was first recognized in Arkansas by Miser and Purdue

(1919) and in Texas by Stephenson (Dane, 1929). This volcanic debris was described in

detail by Ross et al. (1929) and by Belk (1984,1986). The Woodbine graywackes are

olive-gray, soft, poorly sorted, cross-bedded, and locally calcite cemented; volcanic rock

fragments of varying lithology are preserved in sand- to cobble-sized clasts within the

Woodbine. In addition, fresh, vitreous, colorless detrital sanidines are preserved as

individual sand grains within the graywackes. Of particular interest is the basal

conglomerate of the Woodbine, which is exposed along Mine Creek and Blue Bayou in

Howard County (southwest) Arkansas. The gravel bed has been described as 60 feet (18

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

Page 24: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

73CD■DOQ .C

gQ .

■DCD

C/)(/)

8■D

CD

3.3"CD

CD■DOQ .C

aO3■oo

CDQ .

OC

"OCD

C/)C/)

West East

15 “

II TertiaryMississippi Embayment sediments &

<55

AustinEaolefon Tokio/ Brownstown

Woodbine

Washita

Rgure 3. Generalized Cretaceous northern Gulf of Mexico Basin stratigraphie relationships.

CO

Page 25: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 0

m) thick near the town of Centerpoint, Arkansas (Miser and Purdue, 1919) and as thinning

westward until it is one foot (0.3 m) thick near the town of Lockesburg, Arkansas. (Dane,

1929). Present day exposure is extremely poor so that at most only a few meters of gravel

can be seen at any single outcrop. The best exposures of the gravel are found along Mine

Creek and Blue Bayou in Howard County, Arkansas. Volcanic rocks of varying lithology

are preserved as cobbles in this conglomeratic unit. The volcanic cobbles are well-rounded

and up to 20 cm in diameter, with weathered rinds that are easily removed exposing very

fresh kernels of rock.

B. Cretaceous Igneous Rocks of the Gulf of Mexico Basin

The specific igneous source of this volcaniclastic debris is uncertain. Abundant

potential igneous source rocks are present; diverse Upper Cretaceous, alkaline,

intermediate, mafic, and ultramafic igneous rocks are found at many localities along the

northern margin of the Gulf of Mexico (Byerly, 1989). The Cretaceous alkalic province

(Morris, 1987) includes the Magnet Cove ring dike intrusive complex. Granite Mountain

syenites, the Benton dike swarm (lamprophyres), and the Prairie Creek lamproite; other

individual igneous complexes include the Baicones province nephelinites, alkali basalts,

and phonolites, and volcanic rocks preserved on the Monroe Uplift and Jackson Dome

(Byerly, 1989) (Figure 4). Although widespread, these igneous rocks are volumetrically

minor; individual igneous centers are often quite small. Magnet Cove covers approximately

12 km^, Granite Mountain syenites covers approximately 12 km^, Baicones province

consists of as many as 200 volcanic centers covering 300 km^.

Igneous rocks preserved at these centers have been dated by many workers using a

variety of techniques including K-Ar, Rb-Sr, and fission track dating on whole rocks and

mineral separates (Figure 4). Granite Mountain samples yield ages ranging fiom 87-95 Ma

(Monis, 1987; Morris and Eby, 1986; Zartman, 1977; Zartman, 1967), Magnet Cove

rocks have been dated at 97-103 Ma (Zartman, 1967), Benton Dikes are reported as 99-100

Ma (Morris, 1987), the Prairie Creek lamproite is 97-106 Ma (Zartman, 1977), and the

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

Page 26: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

11

Baicones

PrairieCreek

MagnetCove

VIntrusive

BentonDike

Swarm

PotashSulfur

Springs

GraniteMountain

MonroeUplift

IHIIIIIIIW IIIM M IIIIi

JacksonDome

Ma 60 70 80 90 100 110

Ages from the literature (see text)

Ages determined as part of this study

Figure 4. Summary of radiometric ages for Cretaceous igneous rocks of the northern Gulf

of Mexico Basin.

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

Page 27: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

12

Balcones rocks yield a larger variety of ages from 69-98 Ma (Burke et al., 1969;

Baldwin and Adams, 1971; Barker et al., 1987). In general, a pulse of igneous activity

occurred at 80-l(X) Ma at many localities along the northern margin of the Gulf of Mexico;

activity in the Balcones province apparently extended over a longer time interval.

The tectonic and petrogenetic significance of these igneous centers remains unclear.

Although located along the structural margin of the Gulf of Mexico where a Pennsylvanian

compressional belt and a Permo-Triassic extensional zone are located, no tectonic activity is

directly associated with the Upper Cretaceous igneous rocks (Byerly, 1989). Despite a

lack of obvious tectonic activity, rejuvenation of the Precambrian Reelfoot aulacogen

occurred in the early Upper Cretaceous (Ervin and McGinnis, 1975) and the Sabine and

Monroe Uplifts became active at that time. Formation of the modem Mississippi

Embayment began during early Late Cretaceous time by reactivation of the Reelfoot Rift

(Ervin and McGinnis, 1975). Tensional stresses associated with downwarping may have

resulted in zones of weakness that were intruded by alkaline rocks (Ervin and McGinnis,

1975; Kidwell, 1951); rejuvenation of a rift is commonly associated with alkaline igneous

activity (Burke and Dewey, 1973). Alkaline rocks in Mississippi, Louisiana, Arkansas and

Texas cannot be directly correlated to downwarping of the Mississippi Embayment,

however, tensional stresses may have occurred throughout the Gulf Margin at that time;

specific tectonic mechanisms are not well understood at this time.

Although it is difticult to envisage the Gulf Coast passive margin as an area of

widespread volcanic activity, the modem Niger delta is an example of a modem passive

margin dominated by deltaic sedimentation that is associated with active volcanoes (Wright,

1985; Hunter and Davies, 1979). This volcanic zone consists of basalts, phonolites and

trachytes. The Benue trough is an area of basin subsidence; this relationship between

alkaline volcanism, basin subsidence and deltaic sedimentation makes the Benue trough an

example of a modem aulacogen (Hoffman et al., 1974). The Benue trough is also an

excellent analogy for volcanic periods of the Gulf Coastal Province; Niger deltaic sediments

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

Page 28: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 3

interfinger with volcaniclastics and lava flows near Mt. Cameroon on the Niger delta

(Hunter and Davies, 1979). The Mississippi Embayment lies in a reentrant o f a continental

margin and is associated with subsidence and alkaline volcanism; it is possible that this

embayment is an aulacogen.

The petrogenetic relationships of the widespread Gulf of Mexico igneous rocks are

also not fully understood. Moms (1987) examined the diverse Cretaceous alkaline igneous

rocks of Arkansas and concluded that crystal fractionation played a minor role in the

transition from mafic to syenitic intrusions and that separate mantle sources and potassic

metasomatism were more important factors in the evolution of these rocks. Seven major

Cretaceous complexes are known in Arkansas (Figure 1); each seems to be a discrete

magmatic system (Morris, 1987). The rocks of the diamondiferous Prairie Creek lamproite

(Figure 1) are the most mafic and undeniably mantle-derived (Morris, 1987). On the basis

of whole rock geochemistry, mineralogy, mineral compositions and intrusion shape, the

hypabyssal peridotite, breccia and tuff of the Prairie Creek complex are properly classified

as lamproites; the glassy groundmass, presence of diopside, high Zr concentrations (1000-

1500 ppm), high T i02 concentrations (3-7 wL % ), and champagne glass shape are

characteristics typical o f lamproites rather than kimberlites (Morris, 1987). Lamproites, in

general, are thought to be related to activated rifts (Morris, 1987). Reactivation of the

Reelfoot Rift may have triggered emplacement of the Prairie Creek lamproite.

The Perryville-Morrilton carbonatites of central Arkansas consist of a carbonate

matrix with phenociysts of phlogopite, amphibole, diopside and xenoliths and xenocrysts.

They are extremely Si-undersaturated and have highly variable compositions (Morris,

1987). Mixing of phlogopite, apatite, brecciated sandstone and shale indicates explosive

emplacement of these rocks although petrogenetic relationships are not yet understood

(Morris, 1987).

The lamprophyres of the Benton Dike Swarm (Figure 1) are lithologically diverse;

some workers suggest fractional crystallization of a hydrous basaltic magma as the origin

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

Page 29: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 4

of these Si-undersaturated rocks (Van Buren, 1968; Robison, 1977; Steele and Robison,

1976). Morris (1987) concludes that multiple sources or magma mixing must be

considered to fully understand the origin of these rocks. Robison (1977) interpreted the

lithologically diverse lamprophyric-gabbroic dikes of the Benton Dike Swarm as

comagmadc; he proposed that partial melting in the mantle and fractional crystallization of a

hydrous, basaltic, Si-undersaturated magma enriched in alkalies and volatiles in the lower

crust as the likely explanations for the origin of these dikes.

The Magnet Cove complex (Figure 1) is a classic alkalic, Si-undersaturated, ring

dike complex with lithologies ranging from mafic ijolites and jacupirangites to felsic

trachytes and phonolites. Erickson and Blade (1963) proposed periodic intrusion of a

single fractionating magma; Morris (1987) conclude ' «iat a more complicated model,

involving liquid immiscibility and separate magma sources, are needed to fully understand

the petrogenetic relations at Magnet Cove.

The V Intrusive (Figure 1) consists of at least 40 cross cutting dikes that range in

composition from mafic microijolites to felsic phonolites and syenites; mineral

compositions of rocks from the V intrusion are similar to those of Magnet Cove (Morris,

1987). Owens (1968) proposed a petrogenetic model for these rocks similar to the

Erickson and Blade (1963) model for Magnet Cove; sequential intrusions fiom a

fiactionating magma produced the variety of rocks at the V intrusive. The Potash Sulfur

Springs Complex (Figure 1) is a smaller ring dike complex similar to Magnet Cove;

lithologies range from ijolites to syenites. Geochemical trends from this complex are

consistent with derivation of these lithologies ftom a single fiactionating magma (Morris,

1987). Howard (1974) suggested that the syenites were intruded first as part of a

fractionated cupolas at the top of a alkali basaltic magma chamber. This magma produced

later lamprophyres and ijolites.

Syenites found at Granite Mountain (Figure 1) contain rare olivine, clinopyroxene,

amphibole and abundant feldspar. Major element data fiom these Si-saturated rocks

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

Page 30: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

15

indicate that they are altered or metasomatized (Morris, 1987). Trace element data,

combined with initial Sr isotopic ratios, indicate that these syenites have an increasing

crustal component as they become more hydrous and potassic. Thus, the rocks at Granite

Mountain do not appear to be related by simple fractionation but have undergone processes

that mobilized LIL's and added hydrous crustal material to the parent magma (Morris,

1987).

Upper Cretaceous alkaline rocks are also present in central Texas (Figure 1).

Barker et al. (1987) studied the bimodal (nephelinite-phonolite), undersaturated suite in the

Balcones province, and conclude that different degrees of mantle partial fusion and

firactional crystallization explain the chemical and minéralogie variation of those rocks.

They suggest this type of bimodal magmatism may typify areas of stable crust.

This extensive igneous activity provided abundant source material for sedimentary

units such as the Woodbine formation within the Gulf of Mexico. Many of the Gulf Coast

igneous centers (Magnet Cove, Granite Mountain, V intrusive. Potash Sulfur Springs)

preserve only intrusive rocks; volcanic equivalents are long since eroded and removed.

Thus, it is important to emphasize that the volcanic rocks preserved as cobbles within the

Woodbine formation are remarkably fresh and actually record volcanic lithologies not

preserved at any known Gulf Coast Cretaceous igneous center. A thorough study of the

volcanic rocks of this terrain must include examination of the reworked volcanic material in

the sediments.

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

Page 31: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Chapter 2. Methods

1. Sample Collection.

All samples were collected during four field trips to Arkansas, Texas, and

Oklahoma in October, 1986, December, 1986, April, 1987, and October, 1988. Two

types of samples were collected from the Woodbine formation. Feldspathic lithic

graywackes were collected from eight localities in SW Arkansas, NE Texas, and SB

Oklahoma (Figure 5). Some of the outcrops described by Ross et al. (1928) are still good

sites for collecting samples; the Mine Creek location, the Blue Bayou location, and several

sites including Arthurs Bluff, Lamar County and north of the Silver City Feny (now

bridge) along the Red River expose outcrops of Woodbine. Most of the road cuts

described in the 1961 and 1970 Field Trip Guidebooks of the Shreveport Geological

Society (Boatner, 1961; Durham, 1970) are overgrown and are useless sites for collecting.

Outcrop exposure within the Woodbine Formation is poor; the best collecting sites are

found along creek beds and in ditches and road cuts. Rocks which were described by

Spooner (1926) as Woodbine equivalents (Blossom sand) and which crop out along the

western rim of the Prothro Salt Dome in Bienville Parish, Louisiana were also collected.

Exposure along the rim of the Prothro Salt Dome is extremely poor and samples were

collected from small sand patches on the ground.

Approximately 60 volcanic lithic cobbles were collected from conglomeratic

localities along Mine Creek (T9S-R27W-S2) and Blue Bayou (behind the Blue Bayou

Church in T9S-R28W-S23) in Howard County (southwest) Arkansas. No systematic

stratigraphie relationship of cobble lithology was observed; metamorphic, sedimentary,

and volcanic cobbles of varying lithology were found apparently randomly distributed

throughout the conglomeratic unit. Volcanic rocks preserved as cobbles include rocks

resulting from effusive eruptions (trachytic textures) and pyroclastic eruptions. All cobbles

were collected directly from Woodbine outcrops with a rock hammer, modem stream

1 6

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

Page 32: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

CD■DOQ.C

gQ.

■DCD

C/)C/)

8■D( O '

3.3"CD

CD■DOQ.CaO3"Oo

McCunain CouniyHoward Couniy Murfreesbofo

NuhviUcj g

fdàbcl*^

Red River County

10 miles

CDO.

"OCD

(/)(/) Figure 5. Generalized map showing locations of outcrops of Upper Cretaceous Woodbine

Formation (After Ross et al., 1928).

Page 33: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 8

cobbles, though likely plucked fiom the Woodbine, and those cobbles reworked into the

overlying Tertiary sediments were not collected.

2. Petrology.

Impregnated thin sections of the graywackes were prepared for pétrographie

analysis. Thin sections of fresh volcanic cobbles and sanidine grain mounts were prepared

for pétrographie and electron microprobe analysis. Sanidine separates were easily obtained

fiom all the sandstones by disaggregating hand samples in a Na acetate buffer solution and

hand picking fiosh detrital sanidines. The sanidines were mounted in epoxy and prepared

for electron microprobe analyses.

3. Volcanic Whole Rock Geochemistry.

XRF geochemical analysis of major, minor, and trace elements for volcanic rocks

preserved as cobbles were generously provided by Dr. Stephen Nelson at Tulane

University. Elemental analyses were performed using a Siemens SRS 200 X-ray

fluorescence spectrometer with Cr and Mo anode X-ray tubes. Preliminary preparation of

volcanic cobbles for XRF analysis included the trimming weathered cobble rinds. The

samples were then ground in a small jaw crusher and powdered using a ceramic disc mill.

Undiluted powders were pressed into pellets in aluminum cups; no binder was used.

These pellets were used for major and trace element analyses. Calibration procedure are

described in Nelson and Livieres (1986). Pyroclastic cobbles were not included in whole

rock chemical analysis due to the abundance of lithic fragments and the highly altered

matrix; no severely altered cobbles of any lithology were used in this study.

4. Mineral and Melt Inclusion Compositions.

All mineral compositions and all melt inclusion compositions, except MgO, were

determined by wavelength dispersive spectroscopy (WDS) on a JEOL-733 electron

microprobe operating at 15 kv and 10-15 nA for mineral analyses and 5 nA for glass

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

Page 34: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 9

inclusion analysis in order to avoid damaging the inclusions. A defocussed (10m) beam

was used for all mineral and melt inclusion analyses. Standard electron microprobe

procedures were employed; polished thin sections and grain mounts were carbon coated

before analysis. MgO concentrations in melt inclusions were very low (< 0.5 w t %) and

were determined with energy dispersive spectroscopy (EDS) in order to avoid the long

residence time under the electron beam necessary to acquire WDS results.

5. Geochronology.

All geochronologic data for this study were generously provided by Dr. Ajoy Baksi

o f Louisiana State University. A K-Ar date was determined for a trachyte preserved as a

cobble in the Woodbine conglomerate. '^A r/^^A r total fusion and A rP ^ A r two-step

incremental heating ages were determined for rocks from several Cretaceous, Gulf Coast

igneous complexes. Ages from the literature for these, and other, igneous complexes,

which are potential source areas for the reworked volcanic material, were also used.

6 . Petrogenesis.

Petrogenetic modeling of major elements and of trace elements for the Woodbine

volcanics was conducted using a number of qualitative and quantitative methods.

Qualitative analysis of whole rock and mineral compositions was used to begin evaluating

whether partial mantle melting, fractional crystallization, magma-mixing, or AFC, could be

responsible for the chemical variation observed in volcanic rocks preserved as cobbles in

the Woodbine Formation. In order to verify and refine results obtained from qualitative

models, quantitative modeling was conducted. Two distinct modeling techniques, one

involving major element modeling and the other trace element modeling, were employed,

a) Major and Minor Element Modeling

Fractionation calculations of major and minor elements (810%, Ti02, AI2O3, FeO,

MnO, MgO, CaO, Na20, K2O, P2O5) were performed in the spreadsheet program Excel.

These fractional crystallization, mass balance calculations involve subtraction of mineral

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

Page 35: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

2 0

compositions fix>m primitive volcanic rock compositions in order to derive the composition

of a theoretical, normalized residual liquid. Modeled residual liquid compositions were

compared to compositions of differentiated volcanic rocks of the Woodbine suite to test

fractionation schemes.

b) Trace Element Modeling

Trace element modeling was performed using the Rayleigh fractionation Law:

C L / C O = p ( D - l )

whereCL = concentration of trace element in the residual liquid

CO = concentration of trace element in the source melt F = fraction of liquid remaining (F = 1 = all melt; F = 0 = all solid)

D = bulk distribution coefficient

Rayleigh fractionation assumes surface equilibrium during crystallization.

Partition coefficients were calculated, when possible, from rock and mineral

compositions determined as part of this study; other partition coefficients were collected

from the literature. Calculated trace element enrichment factors were compared to

enrichment factors of differentiated to primitive volcanic rocks of the Woodbine suite to test

fractionation schemes.

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

Page 36: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Chapter 3, Petrology, Whole Rock Chemistry, and M ineral Chemistry

1. Petrology.

Pétrographie analysis of 20 thin sections of the feldspathic lithic graywackes of the

Woodbine Formation shows that they are very poorly sorted and consist of rounded to

subrounded grains o f volcanic rock fragments, fresh sanidine crystals, quartz (single

domain, multi-domain, and stretched metamorphic), and chert fragments enclosed in a clay

matrix and calcite cement (Figure 6). The volcanic rock fragments (VRF’s) are of varying

lithology ranging from relatively fresh trachytic felsic fragments to more altered mafic

fragments (Figure 6). Montmorillonite clay rims are common around grains. Alteration of

VRF's resulted in an early primary clay cement; calcite cement (Figure 6) was indirectly

derived from alteration of the plagioclase rich groundmass of the VRFs (Belk, 1984).

An examination of the variation of graywacke composition revealed no obvious,

systematic change in grain size or mineral assemblage along formation strike. Graywacke

grain size is extremely variable at the outcrop; lenses, several centimeters to several meters,

o f fine and coarse material are commonly interbedded. Grain size, not sample location,

appears to be the dominant variable in determining graywacke composition within this

study area. Coarser-grained sandstones contain more volcanic lithic fragments and

sanidine grains, whereas finer-grained graywackes contain more quartz.

Pétrographie analysis of 50 thin sections of volcanic cobbles revealed four main

lithologies; mafic phonolites, intermediate trachybasalts, felsic alkali trachytes, and

pyroclastic welded tuffs are found preserved as cobbles. Again, no relationship between

cobble lithology and stratigraphie position within the conglomerate was discerned.

The four volcanic lithologies preserved as cobbles within the Woodbine are described as: 1)

Mafic rocks contain abundant, large (1 - 7 mm), euhedral clinopyroxene (titanaugite)

phenocrysts which are often strongly zoned and pleochroic in shades of pink (Figure 7a);

apatite is an important minor phase as inclusions within pyroxene. Plagioclase, when

21

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

Page 37: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

22

Figure 6 . Photomicrograph of a typical graywacke from the Woodbine Formation.

Volcanic lithic fragments (a), detrital sanidine grains (b), chert fragments (c), and calcite

cement (d), are present.

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

Page 38: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

23

r,

Figure 7. Photomicrographs of volcanic cobbles from the Woodbine Formation. Mafic

volcanic rocks (7a), clinopyroxene-rich intermediate volcanic rocks (7b), plagioclase-rich

intermediate volcanic rocks (7c), felsic volcanic rocks (7d), reaction rim around hornblende

phenociyst in a felsic rock and pyroclastic rocks (7f) are preserved in the conglomerates of

the Woodbine Formation.

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

Page 39: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

24

1

Figure 7 (cont.). Photomicrographs of volcanic cobbles from the Woodbine Formation.

Mafic volcanic rocks (7a), clinopyroxene-rich intermediate volcanic rocks (7b),

plagioclase-iich intermediate volcanic rocks (7c), felsic volcanic rocks (7d), reaction rim

around hornblende phenocryst in a felsic rock and pyroclastic rocks (7f) are preserved in

the conglomerates of the Woodbine Formation.

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

Page 40: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

25

%

& 5 m m e

Figure 7 (cont,). Photomicrographs of volcanic cobbles from the Woodbine Formation.

Mafic volcanic rocks (7a), clinopyroxene-rich intermediate volcanic rocks (7b),

plagioclase-rich intermediate volcanic rocks (7c), felsic volcanic rocks (7d), reaction rims

around hornblende phenocryst in a felsic rock (7e) and pyroclastic rocks (7f) are preserved

in the conglomerates of the Woodbine Formation.

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

Page 41: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

2 6

present, occurs only as microlites in the matrix and, along with pyroxene, magnetite, and

apatite, constitutes the groundmass (Figure 7a). Hornblende is a minor phenocryst phase

in one very fresh mafic sample (WDB-86-2A); severely altered phenocrysts in other mafic

samples are interpreted as amphibole on the basis of morphology.

2) Intermediate lithologies are characterized by subequal proportions of

subhedral plagioclase and eu-subhedral clinopyroxene (titanaugite) phenocrysts. Both

phases are commonly zoned. Severely altered phenocrysts in intermediate samples are

interpreted as amphibole on the basis of morphology. Intermediate lithologies can be

subdivided into clinopyroxene-rich and plagioclase-rich samples, dependent on the relative

proportion of these two phases (Figure 7b and c).

3). Felsic samples are dominated by complexly zoned, euhedral phenocrysts of

alkali feldspars with anorthoclase cores and sanidine rims, although clinopyroxene and

hornblende are also relatively abundant (Figure 7d). Hornblende phenocrysts commonly

exhibit magmatic reaction rims (Figure 7e); these reaction rims are distinct even when all

other phenocryst phases appear fresh and unaltered. The alteration rims are composed of

Fe-oxide and clinopyroxene. Euhedral sphene is an important minor phase.

4) The fourth rock type preserved as cobbles within the Woodbine

conglomerate represent a more explosive type volcanism; pyroclastic rocks are

characterized by broken alkali feldspar and hornblende phenocrysts, volcanic xenoliths,

and squashed pumice fragments in a chaotic and highly altered fine-grained matrix (Figure

7f).

Total phenocryst content does not vary systematically between lithologies or

individual samples although variations do occur. Two of the mafic cobbles (WDB-87-1M

and WDB-88-1M) have very high phenocryst proportions relative to all other rocks. These

samples may have accumulated crystals through crystal settling.

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

Page 42: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

2 7

2. Whole Rock Chemistry.

A. Whole rock major element chemistry.

Whole rock geochemical analyses are reported in Table 1. Substantial variation in

major and trace element concentrations are present in the suite of Woodbine volcanic

cobbles. MgO content of the volcanic cobbles ranges from 5.5 wt. % to 0.85 wt. %.

Calculated differentiation index, D.I., (normative quartz + orthoclase + albite + nepheline +

kalsilite + leucite) for these rocks range from 62 in the mafic samples to 92 in the most

felsic; Mg # ((Mg / Mg+Fe)*100) ranges from 50 in the most primitive samples to 18 in the

most differentiated. A lack of volcanic stratigraphy prohibited discrimination of early vs.

late eruptive units or even whether the volcanic rocks are cogenetic. Variation diagrams

have been constructed by plotting MgO along the X-axis as the concentration of this

variable is known to be a function of liquidus temperature; magmatic fractionation is also

temperature dependent. Each point on the variation diagrams represents an analysis of a

single volcanic cobble.

Most major elements vary systematically with MgO concentration (Figure 8).

Major elements show considerable variation throughout the collection of volcanic cobbles

(S i02 = 44 - 61 wt. %; Na20 = 1.5 - 6.1 wt. %; K2O = 3.0 - 7.7 wt. %; CaO = 1.3 -

8.9 wL %) with the most felsic samples being enriched in alkalis, Si02, and AI2O3 . All

'mafic' oxides (MgO, FeO, CaO, Ti02, P2O5 , and MnO) are depleted in the felsic samples

and compositional gradients are smooth and regular (Figure 8).

The only element that does not show a systematic variation with MgO is Na2 0

(Figure 8). The Na20 variation diagram reveals considerable scatter. Na is recognized as

a very mobile element and this scatter may reflect some post-crystallization alteration.

However, the alteration explanation is difficult to reconcile with the fact that samples

which are lithologically similar have very similar Na2 0 concentrations. For example, 4

samples, 86-2E, 87-1 A, 87-lD, 87-IF, appear identical in hand sample and in thin section;

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

Page 43: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

2 8

“ . s S S o Ç o S f S o o Ç Ç S ? c j S ^ i 2 ^ S ? c 3 5 ^ o 5 • o)CD ^ O) 40 o CM CO cvi i n d s o t - c M t- t- t- co

00 CLlO 1“

Oe^fi!?:coocooocMa>inu><oo>CMo>CMor^o>CMo>ooco; : . £ c q c û C M T - T - r ^ 7 - i r ) O O l O « > C O C M IO O IO M C O C M M O C M

’■"‘ Si

u.-sooi^wcooooTi-cMr»2Tj-cviinT-r>ocooocvno^o oî‘ooi-;eoocor^ooOeocM'^h«r^ ^co t - i^ t - t - .' cnoT-oooco i r ï iO ' t fog '^ J2 ’“ ’"ÇM

CM

. S c o o j c o c o i - c o t - t - c o c o ' ^ t j - c m c o i ^ c o c o c o T - r ^ o c o

Ui-gOr r^O'M-CMOJOOO'M-WCOeoCM'M-flOOCOr'OOM-CMtD cM~'M;eqtO'M;i-;coT «r«;COO>cocMCM<or«. i^cot-noco j, 0)0 T-o(oomtoiOTfo™ ^ <oO) CM

^-*N>CO^r^OlrtOT—CM'»“ T-wï->w'^UJww# i^w ( q o q o O ) ( M ^ ; 0 ( O I A I O O O ^ O O C O C M C O O1- (H ®? “? S'

^ « CM ^ o> o T j - h . i n e o 1 - g 5 J 2

a^ç3coe<5CMo>CM<or.o>co£ir»incMi-mocM'<»'<oo>cooo Ï 5 c 0 0 ( O CO m t ; 0 > o o o T - «0 CM CM < 0 0 0 O ) T - 0 0 CO CO CO < o

g ^ CM

ui^î-cocoomoo<Dr>eo^2 CMioeo<Dh*ocM<oeoo>N^

« s 8 f e 5 8 s i 8 5 Ç S ; 5 ! a - s g g S ° S i ! : : 5 ! 8 S ? ü

Z , ^ % ( ^ 7 Z O ( ^ O C 0 ( 0 C M O m T f ( 0 G ) ' e C 0 0 0 ^ r ^ O ( 0 ( M ( 0^.£cvjT^iqiocMcqo>*^rvo>«Mcoco’ co»^T-w<oincMO)<o• *5t5 m M \ f A W n i —2 n i n i 00 ^ CVI 1“i m d c o d d d ^ c o T ^ c o d ® ' ^ CO m

^ r - c o o c o ^ < o o O ’ co ‘M*r oo1 r* n\ irt im n\ n\ n\ ai n\ n%CM

a , , o>oioinco<OTfcoeocMio<ococM'if<or*ocMinT-cMtn ^.Sr^'>-;r^cMt-cocMi-r*co®. OTcoiçjT-co'i'^iniO'^o)^ ^ ■ iô m CO CO CO d o> cvi d g » E ^ ---- “

o •>- CO

u.(ï3:^S°C:‘P>^!i:®’ '^®oœ'^o>CMOcj>T-ocMco'0-co T-œ^'^®î*'l’r®?‘N‘ **l^“î ' ^ 0'J’“ '<>0OT-'i-OOCOCOCO J.COinCMlOQO'^dM-cÔ'r^^^ oo M• M i n C M C D O O ' M - O O ' M - C OS e '^<, , o<Dcoo>inocM-»-T-o>wo>ooior»'M'or^o>coo>Tf't- î3,H«i^o'!cO'i-;inoocMO<oocM'0-f>»cocMinr»<oioeoco'M- ^ « ç CM d d d i r i cô cvi cô d g '*■ f - co j

S S S ? ? § , ° S 8 S 2 N > - « ê 5 ô > 5 5 i 2 S

I

!u8

I•i

;

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

Page 44: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

2 9

o ,»Soo2 <5 SCMooooo>2Joot^^<oo)0 '^oo«Doeot - . 9 ‘o r ^ c o o o o > c O ' i - c o o o > c o c v i ' » " ^ c v i l o t - o

• i 2 o o t - C O o o T - < 0 (Ô o 2 < O i - T - CM 1- g JÎ S Æ ' O c m o cm

CL „t-cocoTj-cMcocDO'^r^o (Doit-r^oooirjoJcoi-cOT- T-.2 or«.ioir-;T^oj(OT-;»noo>mcMf^co«o lOr- y- m r~^ J - o CO o o 1 - lO «ci o 2 «D t - CM «D

CO , , < g î P Ç 2 ' > - ® < ï > ' M - ç O C O C O J > ‘O t O C M U Î U ) 0 < O C O I > . < O I ^ O i - . 9 > c ? t ^ c o T - ; C 3 r > . i n c > i o o > ^ T - c M C M c o o i o t - t - c m c ^ c o ^ i 2 ' - O T - e o o o ' « - m < ô o 2 * c * « o t - cm c m h -

; _a j <o CM■ ici <ô o g

-1 ,.SSr"C05:9?!2“5CMO)r^w"<i'ooo>o>oocMoo<oioin• — LO T“ T— CM 1—

o CO o o 0 0)00 CM ■M-

CO

00 „io«otnincM<oin<Docou>or^oin'M-of^i c ^ . ü c n r ^ o o q i - c o c o c o c n o ' M o o T - o o t o o o l o •

^ Æ «D o CO o o «o «D 6 g ™

" ü S S 5 S S § § S S S S ! 8 5 ! S i ^ S Ç ° S ? : ° ; :• « c i o T - C Ô c i C) T - « D «d o Z < 0 y - 1- CM

<D (D CM °

If) O) O CO

CM r - CM CM

^ . . S I 2 Ç O < 0 5 Ç M I O < ï ) l O O O o t O < O C O C O O C O « D O < 0 < D U Î ^ . 9 “ ? q « q q o i o c M O ) i - < o i o c M C M i f > o ) i o t - t - o ) oJ i 2 T - » - CM o 1 - 1 - 'M'<d o 2 T - 1- CM O)5 j » « 0 CM o

u. .^ocooi-g)CMor*tot^cj' ,-ocoo«nocO'>-i-u)co e M . 2 C M r ~ « i f ) o c M r ^ c o o ) 0 ’-. 1- T - 1 - 0 0 0 c o T- cM o o )

• Æ q o t - c o o T ^ T - ' c ô r > . ' < 3 g “ ’ - m - t - , - t -g j ) ( 0 CM °

ü S § S S è S i 5 i § 5 S 5 S S g § S i ° J 5 î ^ § S 5 g 5 S• C0 m ^ r v i m n ^ n i « < : c n n a > I O O r CM C M r C MNCOOrCM-TCOOg

_j fcSP^r^oomiOMtocoiomcMCMCMcmoocMocorooo ^ . 5 o o c M q r » T - T M f ) N c o r O i o c o L f ) c o r co i - c o p > . i ^ g r o c o c i r C M T r c c i c i g ' n c m t - w

(9 ■P.'2 5:'î'C)'<(f|v.O)OCM<0 <2 COCMT-O^OI^OI^COO)lO ^ . S l O C M W T - T - I O « O C M C O i - U ) T r c O O > ^ 0 O ) t - T - I ^ O ) cà rO-M-OrCM-Mr«ciog“ï 0>t-t- t- M

OQ ■Srinqoioo)<oo)T-'MTv^in^T-r.oooT-CM'f-ooo) ^ . 2 q « f ) i n r ~ o o ) T - ' o - f > C M ‘«) co c M to o c o co co cm co t - co • 0 ) 1 t - O ^ O t - ' c Ô c Ô U Î C ) ® ' ^ C M r r -s q;« CM

Z ■S«oooo<oooooin'M-(^ioint^'»i-T-inoeooi-i>^coco ^ . E ' M ; i q q c o T - T - < o ^ o c o ‘o ^CMoo«oo) t - co cm oo r co

co

2 9 2 ? | | S s 8 S L w . c wc o t - < u . S 5 0 2 a g o . H N > c Q g : N O > 0 2 Z . j m

g

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

Page 45: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

3 0

I

I

60 -

50 -

TT T1

T T2

T T3

T "T5

I

4 603

2

1

0

Wt. % MgO

Figure 8 . Major and trace element variation diagrams for volcanic cobbles of the Woodbine

Formation. Typical l a error bars are shown for a felsic sample.

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

Page 46: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

31

I

§

22 -

2 0 -

18 -

1 6 -

1 0 -

wt. % MgO

Figure 8 (cont). Major and trace element variation diagrams for volcanic cobbles of the

Woodbine Formation. Typical l a error bars are shown for a felsic sample.

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

Page 47: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

3 2

t

St

0.2

0. 1 -

0.0 T T"2

T T3

T T T5

T4

T10 6

10

8

6

4

2

0

Wt. % MgO

Figure 8 (cont.). Major and trace element variation diagrams for volcanic cobbles of the

Woodbine Formation. Typical l a error bars are shown for a felsic sample.

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

Page 48: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

33

i

e

7

6

5

4

3

28 u

wt. % MgO

Figure 8 (cont.). Major and trace element variation diagrams for volcanic cobbles of the

Woodbine Formation. Typical l a error bars are shown for a felsic sample.

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

Page 49: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

3 4

IN

1.4

1.2 -

1.0 -

0.8 -

0.6 -

0.4-

0.2 -

0.0

5 5 0 -

4 5 0 -

350

wt. % MgO

Figure 8 (cont). Major and trace element variation diagrams for volcanic cobbles of the

Woodbine Formation. Typical lo error bars are shown for a felsic sample.

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

Page 50: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

3 5

4 0 -

'S 3 0 -

1

2 0 -

10 -

2000

I3 1000 -

wt. % MgO

Figure 8 (cont.). Major and trace element variation diagrams for volcanic cobbles of the

Woodbine Formation. Typical l o error bars are shown for a felsic sample.

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

Page 51: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

3 6

IS

200

100 -

60

50 _

40 _

30 _

20 _

0 1

Wt. % MgO

Figure 8 (cont). Major and trace element variation diagrams for volcanic cobbles of the

Woodbine Formation. Typical l a error bars are shown for a felsic sample.

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

Page 52: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

3 7

i>

500

4 0 0 .

3 0 0 .

200 .

100.

0 2 31 4 5 6

wt. % MgO

Figure 8 (cont). Major and trace element variation diagrams for volcanic cobbles of the

Woodbine Formation. Typical lo error bars are shown for a felsic sample.

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

Page 53: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

3 8

I3

I

100

80 -

6 0 -

40 -

2 0 -

6 0 -

4 0 -

2 0 -

wt. % MgO

Figure 8 (cont). Major and trace element variation diagrams for volcanic cobbles of the

Woodbine Formation. Typical lo error bars are shown for a felsic sample.

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

Page 54: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

3 9

IZ

300

200 -

□ □

100 -

140

120 -

100 -

80 -

Wt. % MgO

Figure 8 (cont.). Major and trace element variation diagrams for volcanic cobbles of the

Woodbine Formation. Typical l a error bars are shown for a felsic sample.

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

Page 55: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

4 0

4000

3000 -

2000 -

10000 1 2 5 63 4

Wt. % MgO

Figure 8 (cont.). Major and trace element variation diagrams for volcanic cobbles of the

Woodbine Formation. Typical l a error bars are shown for a felsic sample.

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

Page 56: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

41

these samples have N a20 = 5.38,5.18,5.42, 5.34 respectively (Table 1, Figure 8). It is

unlikely that weathering and alteration could have affected these samples so consistently,

unless the alteration occurred before the lava flow was eroded.

Major and trace element vs. MgO variation diagrams reveal individual linear

segments which constitute the overall systematic variation (Figure 9). The CaO variation

diagram reveals a single segment relating constantly decreasing CaO with decreasing MgO

(Figure 8). By comparison, the Nb and Zr variation diagrams are each composed of 3

discreet segments (Figure 9).

B. Whole rock trace element chemistry.

Compatible elements such as Ni,Cr, V, and Cu show systematic changes in

concentration relative to wt. % MgO; mafic samples are enriched in these elements and

compositional gradients from mafic to felsic rocks are smooth (Figure 8). The variation

diagrams o f some of these elements can be divided into segments similar to the major

element segments shown above; Ni for example exhibits changes in gradient at MgO= 3.5,

and 1.5 although all segments indicate Ni depletion with decreasing MgO concentration.

Incompatible elements such as Rb, N b , and Zr (Figure 8) are enriched in the most felsic

samples and changes in compositional gradient are similar to the segments shown earlier.

The semi-compatible elements Sr, Ba, La, and Y (Figure 8) exhibit alternate enrichment

and depletion segments.

C. Whole rock chemistry and mineralogy.

The changes in segment slope on major and trace element variation diagrams

correspond to minéralogie changes in the volcanic rocks that make up the segments. To

illustrate this correlation between sample mineralogy and major/trace element variation, the

Sr vs. MgO diagram is shown in Figure 10. The smooth curve can be subdivided into four

segments that reflect sample lithology. The first segment is defined by samples 86-2A on

one end and 88- IE on the other. All of the samples in this segment are identified as the

mafic rocks described above. The rocks that make up this segment are dominated by

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

Page 57: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

4 2

I

IA

300

200 -

□ Q

100-

Q

550 -

450 -

350

wt. % MgO

Figure 9. Whole rock variation diagrams showing the segmented nature of the geochemical

variation of volcanic rocks from the Woodbine Formation.

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

Page 58: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

■ooÛ.cgQ.

■oCD

8■ov<ë '

3CD

p.3"CD

CD■oOQ .CaO3■DO

3000

2000

Sr(ppm)

1000 -

0

Plag'richIntermediatesamples

plag dominates +cpx

Cpx-rlchIntermediate samples

cpx dominates +plag

Mafic samples Cpx dominates

+apt,mag

Felsic samples _ g j Alkali-feldspar dominates

+sph,hb,cpx

-1----1---1--- 1---1---1--- 1---1--- 1--- 1--- 1— I---r-0 1

CDQ.

■DCD

C/)C/)

wt. % MgO

Figure 10. Variation in mineralogy and in Sr and MgO concentrations of the volcanic cobbles from the Woodbine Formation.

CO

Page 59: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

4 4

euhedral clinopyroxene phenocrysts. The second segment is defined by endmember

samples 88- IE and 87-lD. These samples are all clinopyroxene-dominated intermediate

Ethologies which contain small plagioclase phenocrysts in approximately equal or subequal

proportions with clinopyroxene. The third segment is defined by endmember samples 87-

ID and 88-IG. These samples are dominated by plagioclase phenocrysts; clinopyroxene is

common but less abundant than plagioclase. The fourth segment is defined by samples 88-

10 and 86-2F and are dominated by alkali feldspar phenocrysts; sphene, clinopyroxene

and hornblende are important minor phases. This relationship between sample mineralogy

and chemistry also exists for major elements. Major element segments on variations

diagrams also correspond to changes in sample mineral assemblage.

3. Mineral chemistry.

A. Clinopyroxenes.

Selected clinopyroxene analyses are reported in Table 2 and are plotted in Figures

11 and 12. A complete list of all clinopyroxene analyses may be found in Appendix 1.

Clinopyroxenes are present in all lithologies and compose a diverse and complex

group in this suite of rocks. Clinopyroxenes in mafic rocks are Mg-rich and Fe-poor

relative to both intermediate and felsic lithologies (Table 2). Mafic clinopyroxenes are

strongly zoned; cores are enriched in Mg and Cr, and are depleted in Ti, Fe, Na, Al, and

Mn relative to rims. Clinopyroxenes from intermediate rocks are rich in Ti (up to 4.0 wt.

% Ti02) and Al; intermediate clinopyroxenes are also zoned although much less strongly

than mafic phenocrysts. Clinopyroxenes firom felsic rocks are relatively rich in Fe and Mn

(Table 2).

The basic stoichiometric formula for clinopyroxenes is recognized as:

X i Y i Z2 06 where: X= Ca, Na, Mg, F e + 2

Y = Mg, F e + 2 , Mn, Al, Fe+3, Cr, Ti

Z = Si, Al

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

Page 60: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

■o1 sû .

■oCD

eneno'305CD

8■Ov<ë '

13CD

Cp.

CD■OOQ .CO

■oo

CDQ .

■DCD

M afic In te r m e d ia te Fe s ic P y ro c la s t lc

WDB-86-2A WDB-88-1K WDB-88-1E WDB-87-1A WDB-87-10 WDB-87-1E WDB-88-1G

core dm core rim core rim core rim core rim core rim core rim

S I 0 2 47.93 46.53 46.09 43.16 46.12 44.89 46.03 46.88 49.72 50.29 50.50 50.81 48.60 46.50

AI2 O 3 6.36 7.30 7.57 9.85 7.05 7.93 8.16 7.28 4.26 3.58 3.32 3.00 4.29 5.70

F e O 7.73 8.41 7.60 8.14 7.71 7.69 7.38 7.06 9.87 9.79 9.56 9.34 8.77 9.25

M gO 12.82 12.19 12.23 1 0 .8 8 12.49 12.57 11.40 11.94 11.61 11.55 11.41 11.37 12.71 11.75

C aO 22.26 22.41 2 2 .1 2 2 1 .8 6 21.77 21.81 21.83 21.93 23.09 22.91 21.91 21.87 2 2 .0 2 21.74

N 8 2 0 0.47 0.45 0.65 0.71 0.79 0.72 0.82 0.71 0 .8 8 1 .0 0 0.90 0.90 0.90 0.87

K2 O 0 .0 0 .01 0 .01 0 .01 0.01 0.01 0.01 0 0.01 0 .0 2 0 .0 0 .0 0 0.01

TIO 2 2.70 3.22 2.72 3.79 2.85 3.26 3.44 3.13 1.73 1.36 1.32 1 .2 0 1.69 2.26

M nO 0 .21 0 .21 0 .21 0.18 0.25 0.26 0.21 0 .2 0 0.73 0.87 0.71 0.74 0.49 0.48

C r2 0 3 0 .01 0 .0 0 0 0 0 0 0 0 .0 0.01 0 .0 0 .0 0.03 0

T o ta l 100.51 100.74 99.2 98.58 99.04 99.14 99.27 99.15 101.96 101.37 99.66 99.26 99.50 98.56

enen

Table 2. Selected Clinopyroxene Compositions

en

Page 61: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

4 6

Wo

All lithologies

En Fs

Figure 11. Selected clinopyroxene compositions plotted on a Wo-En-Fs ternary diagram.

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

Page 62: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

4 7

Ti02

_{_ Mafic rocks

□ Intermediate rocks

Q Felsic rocks

N a20 MnO

Figure 12. Selected clinopyroxene compositions plotted on a Ti02-Na20-M n0 ternary

diagram.

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

Page 63: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

4 8

The sum of the cations (X, Y, Z) is equal to 4. Stoichiometric calculations for

clinopyroxenes of the Woodbine volcanic rocks can be characterized by the following

samples:

# i o n s m fo r m u la

87-lE (felsic) 87-lD fintermed) 86-2A fmafic)Si 1.87 1.71 1.87Al 0.16 0.37 0.18Fe 0.3 0.24 0.21

Mg 0.64 0.66 0.77Ca 0.91 0.88 0.89Na 0.11 0.10 0.04K 0 0 0Ti 0.04 0.09 0.05

Mn 0.03 0.01 0.01Cr 0 0 0

sum 4.07 4.06 4.01

Two types of clinopyroxene can be distinguished petrographically and chemically.

Type I pyroxenes are smaller (0.5 -1 .0 mm), subhedral, tan phenocrysts present in felsic

and plagioclase-rich intermediate rocks. The type U clinopyroxenes are characterized as

large (1 -7 mm) euhedral crystals, strongly zoned and pleochroic in shades of pink with

abundant apatite inclusions; type U pyroxenes are present in mafic and pyroxene-rich

intermediate lithologies. A Wo-En-Fs ternary plot of these pyroxenes does not discriminate

these two types (Figure 11); all pyroxenes are diopsides according to the pyroxene

nomenclature of the International Mineralogical Association (Morimoto, 1988).

The two types can be distinguished on a Ti0 2 -Na2 0 -Mn0 ternary plot (Figure 12).

Pink, type II clinopyroxenes are Ti- and Al-rich, and are relatively poor in Na. Type I, tan

clinopyroxenes are rich in Mn (up to 0.98 wt. % MnO) and Ti-poor. Mafic and pyroxene

rich intermediate samples contain pyroxenes that overlap in the type II field on the Ti02-

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

Page 64: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

4 9

Na20-MnO diagram; clinopyroxenes from felsic rocks plot in the type I field. Pyroxenes

in plagioclase-rich intermediate samples plot between more mafic and more felsic samples.

Despite lithologie similarities between mafic, and all intermediate rocks, the pyroxenes

from plagioclase intermediate rocks have been grouped with type I pyroxenes because they

plot along the felsic rock pyroxene trend (Figure 12).

These two types of clinopyroxene are very similar to clinopyroxenes from the

Laacher See phonolitic suite (Womer and Schmincke, 1984). The unreworked,

continuous, Laacher See suite exhibits variation in clinopyroxene chemistry and color

almost identical to that seen here. The Ti0 2 *Na2 0 -Mn0 variation diagrams for pyroxenes

in rocks from the Laacher See suite and the Woodbine rocks are essentially identical;

Laacher See type I pyroxenes are rich in Ti02 and MgO and are abundant in the more

highly differentiated rocks, while type II dominate the more primitive phonolites.

B. Feldspars.

Selected feldspar compositions are reported in Table 3 and all analyses are plotted in

Figure 13. A complete list of feldspar analyses may be found in Appendix 2 ,3 ,4 , and 5.

Feldspar phenocrysts are restricted to intermediate, felsic, and pyroclastic

lithologies; feldspar phenocrysts are not present in mafic samples. Feldspar composition

discriminates samples of different lithologies. Plagioclase phenocrysts in intermediate

lithologies are not strongly zoned; compositions range An=60-40 (Table 3a and Figure

13a). Anorthoclase crystals in felsic rocks are strongly and complexly zoned; cores are rich

in Ca, Na, and Sr, while rims are typically rich in K and Ba (up to 3.0 wt. % BaO) (Table

3b and Figure 13b). Thin Ab-iich overgrowths over Ba-rich zones are often present.

Some anorthoclase crystals have an irregular, patchy zoning. Feldspars in pyroclastic

rocks are unzoned and enriched in orthoclase relative to the intermediate and felsic feldspars

(Table 3c and Figure 13c). Pyroclastic feldspars are also Ba-rich; BaO ranges from 0-1.3

wt. % .

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

Page 65: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

50

00CO

ciIf)

00O)s

ffio5

(Do>Nin

if)0>CDU)

CDOs?cviLO

R00CNJ

COco%

s

0000

(Oo

U )

o

o(O

in

coco

cod

CM

d

oo

CM

d

inT o o

(Ocos

co%

mco

<Dco

CD 0>CN4

(D T-

Oco CMO 21

o

(O T -

CMCO

%

Ss

d

Tfco % cou> s

s

s

end

oCM

CM co

; I oo

O OO) cos U I I Ow

co I I

g00O)

lO

c/3

IîI

II00

cR

I

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

Page 66: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

51

rvh»(O

CMCO T - ( 0 r* VCM o N CO CO V

d d d d

y— 0> So

O CM O

n■M-

O)V

O)

u>

c \i

^ mCM (Od T-'

cocos

ooodco

codCNJ

toO%

od

No q

inV CO

d

inCO

NO

CM

O)

in

O) CMq qcô T-

in coV lOd d

CMcoS

gz

codCM

qdCM

oCM

CM CM CO CD CMq o h . CO

d d d d CM

CM ■O’q o q q CM qd d U ) d d

CO CM

O r

Irr<o

c o

g

SIo

qVCM

h .CO g

cd

O) T - in o> Si CDO)

d

co qh* r

NT* N(O (O

CMO

5 )

cooCM

<

O

0)

O

o>

E

O(0

O I I 9 S(0 m

CMO

00a >

g

O)

g

g

CDqa>

IîII1coden

I

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

Page 67: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

73CD■OOQ .C

gQ .

■OCD

C/)(go"Z5

=TCD

83ë '=T

3CD

CD■OOÛ .CaO3■OO

W D B -88-2C W D B -86-5d e tr ita l

s a n id in e

W D B -86-3d e tr i ta l

s a n id in e

W DB-8 6 - 6d e tr i ta l

s a n id in e

S I0 2 64.20 64.52 64.47 63.05 64.82 65.60AI2 O 3 19.08 19.59 19.19 19.25 19.28 18.72

F eO 0 .2 2 0.19 0.16 0.19 0.17 0.16

C aO 0.69 0 .6 8 0.55 0.60 0.57 0 .2 0

N 3 2 0 3.71 4.14 4.11 2.98 3.82 2.85K2 O 10.81 10.45 10.83 11.98 11.08 12.98

S rO 0 .0 0 0 .0 0 0 .0 0 0.26 0.07 0 .01

B aO 0 .0 0 0.24 0 .0 0 0.07 0 .1 0 0.01

T o ta l 98.71 99.81 99.31 98.44 99.97 100.56

Table 3c. Selected Pyroclastic Feldspars and Detrital Sanidine Compositions

CDQ .

■DCD

C/)en

OlfO

Page 68: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

5 3

An

TJpx-rich

'Ids-rich

Ab

A. Feldspar compositions of intermediate volcanic lithologies.

An

f p cores

rims

Ab

B. Feldspar compositions of felsic volcanic lithologies.

Figure 13. Feldspar compositions plotted on an An-Ab-Or temploL

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

Page 69: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

5 4

An

Ab Or

C. Feldspar compositions of pyroclastic volcanic lithologies.

An

Ab Or

D. Compositions of detrital sanidines..

Figure 13 (cont). Feldspar compositions plotted on an An-Ab-Or templot.

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

Page 70: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

5 5

Sanidine crystals preserved as individual sand grains within the Woodbine graywackes

were also petrographically and chemically examined. These clear, colorless, vitreous

crystals are very fresh. Electron microprobe analysis of these sanidines reveals that they

are similar to sanidines preserved in the pyroclastic cobbles (Table 3c and Figure 13c, d).

Sanidines from sandstones in SW Arkansas, NE Texas, and SE Oklahoma are Fe- and Na-

poor, whereas sanidines from equivalent

sandstones exposed around the rim of the Prothro salt dome in northern Louisiana are

relatively Fe- and Na-rich (Table 3c).

The basic stoichiometric formula for plagioclase feldspars is recognized as:

Na[AlSi3]08 - Ca[Al2Si2]0 g

The Si + Al cation site is equal to 4 and the Na + Ca + K site is equal to 1.

Plagioclase normally contains a small amount of the orthoclase component, KAlSigOg.

Ti'*'^, Fe will substitute for Al; Fe^^, Mg, Sr, Mn substitute for Ca. Stoichiometric

calculations for feldspars of the Woodbine volcanic rocks can be characterized by the

following samples:

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

Page 71: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

5 6

Intermediate and Felsic Volcanic Rock Feldspars U io n s in fo rm u la

WDB-88-1E WDB-87-1A WDB-87-1B WDB-87-1Cflnt) I M (fçlsiç), Ifelâcl

Si 2.500 2.413 2.679 2.708Al 1.499 1.572 1.315 1.278Fe 0.011 0.013 0.013 0.012Mg 0.000 0.000 0.000 0.000Ca 0.490 0.560 0.288 0.251Na 0.436 0.404 0.608 0.707K 0.056 0.037 0.075 0.059Sr 0.000 0.018 0.016 0.012Ba 0.004 0.004 0.011 0.009Tx 0.000 0.000 0.000 0.000

Si, Al site 3.999 3.985 3.994 3.986Ca, Na site 0.997 1.036 1.010 1.050

The basic stoichiometric formula for alkali feldspars is recognized as:

Na[AlSi3]0 g - K[AlSi3]0 g

The Si + Al cation site is equal to 4 and the Na + K site is equal to 1. Alkali

feldspars normally contain a small amount of the anorthoclase component, CaAl2Si2 0 g.

The amount of CaAl2Si2 0 g varies from usually less than 5 % for OrlOO AbO - OrSO Ab50

and then tends to increase slightly in the more Na-rich members (Deer et al., 1966). Ba

substitutes for K, Ti"*" , Fe will substitute for Al, and Fe"*" , Mg, Sr, Mn substitute for Ca.

Stoichiometric calculations for feldspars of the Woodbine pyroclastic rocks and in the

detrital sanidines preserved in the graywackes can be characterized by the following

samples:

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

Page 72: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

5 7

Pyroclastic Volcanic Rock Feldspars and Detrital Sanidines # io n s in fo rm u la

WDB-88-2C WDB-88-2C WDB-86-3 WDB-86-7

(detrit. san) (detrit. san)Si 2.959 2.927 2.956 2.971Al 1.043 1.080 1.036 1.018Ca 0.035 0.041 0.028 0.022

Na 0.355 0.389 0.338 0.491K 0.612 0.556 0.645 0.495Sr 0.000 0.000 0.002 0.000Ba 0.000 0.013 0.002 0.000

Fe 0.005 0.007 0.006 0.012

Si, Al site 4.002 4.007 3.995 3.991K, Na site 1.007 1.006 1.020 1.030

Although considerable compositional variation exists for feldspars from each

lithology, there is a strong correlation between volcanic rock type and feldspar chemistry.

Feldspar chemistry is interpreted as being a reflection of magmatic composition and does

not appear to have been affected by post-magmatic (alteration) processes.

C. Amphiboles.

Selected amphibole analyses are presented in Table 4. A complete list of amphibole

analyses may be found in Appendix 6.

Amphibole phenocrysts are present in all lithologies but are subordinate to

clinopyroxene phenocrysts in mafic and intermediate rocks; as the amount of clinopyroxene

decreases in the felsic and pyroclastic rocks, amphibole is present in subequal proportions

with clinopyroxene. Amphiboles in WDB-86-2A, a mafic sample, are fresh; amphiboles in

other mafic and intermediate rocks are strongly altered (Figure 7). In some samples, 88-

IK and 87-lD for example, severely altered phenocrysts have been identified as amphibole

on the basis of morphology and by the presence of brown pleochroism seen in a few

crystals. Amphibole phenocrysts in feldspar dominate intermediate rocks, and are often

very fresh with thick reaction rims in felsic samples (Figure 7e). These rims are now

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

Page 73: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

CD■DOQ .C

8Q .

■DCD

C/)Wo"303CD

8■D( O '3"

13CD

3.3"CD

CD■DOQ .C

aO3■DO

CDQ .

Mafic Fe Sic P v ro c la s tic

WDB-86-2A WDB-86-2B WDB-87-10 WDB-88-2C

S I0 2 40.01 39.51 40.08 39.94 40.21 39.62 39.8 39 .76

AI2 O 3 13.50 13.30 12.07 12.18 11.47 1 2 .1 2 12.46 12.34

F eO 11.21 11.30 14.79 14.79 15.99 15.88 15.46 16.16

M gO 13.22 13.09 11.57 11.43 1 1 .0 0 10.91 10.63 10.51

C aO 12.51 12.58 11.79 11.93 11.98 12.01 11.85 11.82

N 320 2.11 2.08 2.69 2.65 2.60 2.69 2.29 2.34

K2 O 1.62 1.63 1.67 1.62 1.45 1.51 1.91 1.95

TIO2 5.22 5.05 4.33 4.19 4.28 4.49 3.81 3.65

MnO 0.61 0.79 1.17 1.38 1.28 1.28 0.57 0.59

F 0.33 0.34 0 0.63 0.99 0.54 0.52 0.47

Cl 0.04 0.04 0.06 0.05 0.06 0.06 0.05 0.06

■DCD

(/)(/) Table 4. Selected Amphibole Compositions

en00

Page 74: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

5 9

composed of Fe-oxides and appear to be magmatic reaction rims rather than post-

crystallization alteration rims. Other phenociysts in these rocks do not exhibit these

reaction rims.

The basic stoichiometric formula for amphiboles is recognized as;

X2-3 Y5 Zg 022(0H )2 where: X = C a,N a,K

Y = Mg, Fe+2, Mn, Al, Fe+3, Cr, Ti

Z = Si,Al

Stoichiometric calculations for amphiboles of the Woodbine volcanic rocks can be

characterized by the following samples:

8&2A# io n s in fo rm u la

86-2B S7-1C 88-2CSi 6.112 6.213 6.390 6.325Al 2.425 2.205 2.148 2.334Fe 1.462 1.917 2.125 2.054Mg 3.018 2.673 2.605 2.518Ca 2.085 1.958 2.040 2.018Na 0.624 0.808 0.801 0.706K 0.322 0.330 0.294 0.387Ti 0.587 0.505 0.511 0.455

Mn 0.103 0.154 0.172 0.077F 0.166 0.000 0.498 0.261a 0.010 0.016 0.016 0.013

Based on high Ti and Na contents these amphiboles could be classified as

kaersutites (Hawthorne, 1981).

Amphiboles from the Woodbine volcanic compare well with amphiboles of similar

lithologies from other igneous centers. The Laacher See amphiboles are characterized by

high FeO (10 - 23 wt.% FeO), MnO (0 - 2.9 wt% MnO), and variable Na20 and Ti02

(2.2 - 5.9 wt.% Ti02) contents (Womer and Schmincke, 1984a). Amphiboles from the

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

Page 75: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

6 0

more differentiated Woodbine volcanic rocks are richer in FeO and MnO, and are poorer in

TiO% and MgO, as is typical for more evolved rocks.

D. Sphenes.

Selected sphene analysis are reported in Table 5. A complete list of sphene

analyses may be found in Appendix 7.

Euhedral, large (0.4 mm) honey colored sphenes are common in felsic and

pyroclastic rocks; sphene is not present in mafic and intermediate lithologies. Their

composition varies within a narrow range. Very high ZrO% (0.5 -1.5 wt. % ) and REE

concentrations are typical of sphenes in these samples. Low totals (97-98 wL %) may

indicate the presence of other trace elements that were not analyzed such as Y, Sr, and Nb.

The basic stoichiometric formula for sphene is as follows:

CaTiSiOs

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

Page 76: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

61

Fe Sic In te rm e d

WDB-86-2B WDB-87-10 WDB-87-1D

S I0 2 29.45 29.71 29.2 29.4 29.48

AI2 0 3 1.47 1.64 1.57 1.41 1.61

T I0 2 33.7 33.29 34.31 35.35 33.84

Z r0 2 1.61 1.94 1.09 0.53 1 .8 6

P 2 O 5 0.1 0.01 0.05 0 0.03

F e O 2.26 2.4 2.22 1.86 2.32

M gO 0.05 0.03 0.03 0.02 0.03

M nO 0.16 0.15 0.16 0.17 0.13

N 3 2 0 0.05 0.05 0.08 0.1 0.04

K2 O 0 0.01 0 0 0.01

C aO 26.44 26.66 26.38 26.19 26.43

L 3 2 0 3 0.36 0.33 0.38 0.41 0.29

0 8 2 0 3 1.76 1.6 1.66 2.16 1.55

to ta l 96.55 97.05 96.33 96.56 96.86

Table 5. Selected Sphene Compositions

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

Page 77: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

6 2

Stoichiometric calculations for sphenes of the Woodbine volcanic rocks can be characterized by the following samples:

# io n s in fo rm u la (5 o x y g e n s )

• 86-2B 87-1C 87-lDSi 1.013 1.004 1.005Ti 0.888 0.907 0.910Zr 0.016 0.010 0.012

P 0.001 0.003 0.003Fe 0.066 0.053 0.050Mg 0.002 0.002 0.003A1 0.066 0.066 0.062Ca 0.974 0.971 0.972Mn 0.004 0.005 0.006Na 0.003 0.004 0.004K 0.000 0.000 0.000La 0.003 0.004 0.003Ce 0.006 0.008 0.006

Si site 1.013 1.004 1.005Ca site 1.008 1.005 1.007Tisite 1.022 1.028 1.024

E. Melt Inclusions.

Remarkably fresh melt inclusions (now glass) are preserved in detrital sanidines of

the Woodbine graywackes (Figure 14). Melt inclusion compositions are reported in Table

6 .

A correct interpretation of the origin of fluid inclusions is essential to any useful

conclusions derived from them. There are many ways a fluid may become trapped in a

crystal, and fluid composition may be altered by post-entrapment processes (Roedder,

1984, p. 12). When crystals grow in a fluid medium of any kind, growth irregularities may

result in the trapping of small amounts of fluid within the solid; fluids entrapped this way

are called primary fluid inclusions (Figure 15a, b, c). Fluid may be entrapped after crystal

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

Page 78: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

63

iiiita

« pb

Figure 14. Photomicrographs of detrital sanidine melt inclusions. Primary melt inclusions

(14a) and secondary melt inclusions (14b) are present in detrital sanidines.

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

Page 79: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

6 4

ino

s

sin00COin

F:In

<o

gCMCO

g

5

CM

00in

CDCO

g

CO

g00

00

ino05

05CO

CD O

d 05CO T -

N00

in05

M 00in T -

CM in00

in ^

in

CMCM

g g

gCM

T - CM

oo

inCM

COCMCM

% CMcno

COCM

COCM

in

CM

O) 1- M

g CM O

f - CD

in

in

g

g

g

CM

T " CO

ino

g

CM COOo o

g COcn g

CO COCMCO

2 1- r r

g

2T - in

CM in

CM 00^ CO

CD 00

T - ^

CO o

V- fsl

05COCO

gCO

gCO

CO

CO

COCO

2CO

gCO

in(Ô

g

00o

o

ec>o

COCO

CO

CO

2

2

2

2

CM

o

COCMd

CMoÔ )

COoCM

<

oou .

o(0Ü

oCMCOz

oCMi£.

2 S O(0 m I- u

in

2

COd

COd

COd

COd

COd

COd

COd

COd

COd

COd

COd

go g

I o

NCM00

O T -

gCM O

R

COCO

O CO

CM 0>CO cn CO o CM T-

C/)I

}I3

Ivd

I

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

Page 80: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

6 5

0A. B.

D.

Figure 15. Mechanisms for trapping melt inclusions (after Roedder, 1984). Figure A

represents rapid dendritic growth followed by solid growth, B represents fracture of the

surface o f a growing crystal followed by solid growth, c represents subparallel growth of

crystal blocks, and D represents fracture of a crystal after growth.

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

Page 81: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

6 6

growth along microfractures; these inclusions are known as secondary inclusions (Figure

15d). There are many criteria for determining the nature of fluid entrapment are abundant,

but no single one is absolute. Size, orientation, number of inclusions per grain, and

association with fractures or cleavage are important parameters to consider when

determining the origin of a melt inclusion. Large (relative to the size o f the host crystal),

equant, random (3 dimensional orientation), single or isolated inclusions are good

candidates for primary inclusions. Inclusions that are part of linear or planar groups

outlining healed fractures or cleavage planes are possibly of secondary origin (Roedder,

1984).

Physical and chemical changes may occur within inclusions after trapping.

Although a single homogeneous fluid is often trapped at high temperature, multiple phases

are sometimes present at room temperature. Crystallization on the walls of the inclusion

leads to formation of daughter crystal of the host phase itself; in the case of primaiy

inclusions, the fluid is certainly saturated with respect to the host phase at the time of

entrapment Careful examination of the boundary between the host mineral and inclusion

can sometimes reveal daughter crystals. Volume changes due to shrinkage upon cooling

cause the fluid to split into two phases, liquid and vapor. Extensive post-entrapment

crystallization may result in enlargement of the vapor bubble; precipitation of daughter

crystals will cause an additional volume change so that vapor bubbles appear larger than

normal (Roedder, 1984).

The Woodbine detrital sanidine melt inclusions are large (up to 100 urn in

diameter), often unaltered, and are easily observed petrographically (Figure 14). Two

types of melt inclusions described by Roedder (1984) have been identified: 1) the primary

melt inclusions formed as melt was trapped during crystal growth (Figure 14a), and 2)

secondary inclusions formed as melt was trapped along microfractures after crystal growth

(Figure 14b). Both primary and secondary melt inclusions can be used to determine a

liquid line of descent for this evolving magmatic system. Inclusions that were identified as

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

Page 82: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

6 7

primary inclusions are relatively Fe- and Ti-rich (FeO=2.8, TiO2=0.9), whereas the Fe-Ti-

poor inclusions (FeO=2.0, TiO2=0.3) are secondary inclusions (Figure 16). In addition,

primary melt inclusions are relatively Cl-poor (Cl=0.2) while secondary inclusions are Cl-

rich (Cl=0.5) (Figure 16).

Consistently low totals (93 - 95 wt.%) for the melt inclusion compositional data

reflects the volatile component that cannot be analyzed on the microprobe. Volatiles most

likely present in these inclusions are H2O, and CO2.

Normalized electron microprobe data for melt inclusions have been plotted on

whole rock variation diagrams (Figure 17). These variation diagrams have been

constructed with Si02 plotted along the X-axis as MgO concentrations in the melt

inclusions are very low (MgO = 0.3 wt. %); the MgO concentrations were determined by

energy dispersive spectroscopy (EDS). The whole rock data for the most felsic samples

and the melt inclusion data are very similar. For some elements, Si02, CaO, K2O, SrO,

the results are virtually identical. For some elements, Ti02, FeO, MgO, the melt inclusion

data is consistent with a more fractionated liquid than the whole rock composition; in other

words, the melt inclusion composition lies along the trend of the segment defined by the

whole rock compositions but overlaps and extends beyond the segment end (Figure 17).

4. Geochronology.

Eight new radiometric ages were determined for Cretaceous igneous rocks of the

Gulf of Mexico Basin as part of this study. These new data are summarized in Table 7 and

in Figure 18.

A K-Ar whole rock analysis of a trachyte now preserved as a cobble in the

Woodbine Formation yielded an age of 98 ± 2 Ma.

A total fusion ^A r/39A r date on biotite from a syenite from the Granite Mountain

complex (Figure 1) yielded an age of 89.5 ± 0.7 Ma. A total fusion ^A r/39A r date on

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

Page 83: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

6 8

0.9 ■

0.8 •

0.7 ■oH 0.6 •

« 0.5 •

0.4 ■

03 •

0.2 •

U

i

62-

60 B

(S

O ■ ■ ■53 58- "

“ " B

i 56-

54-

52-

B

1 1 1 1 ^ i 1 1 1

2.0 2.2 2.4 2.6 2.8

Wt. % FeO

Figure 16. Variation diagrams of sanidine melt inclusion compositions.

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

Page 84: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

69

I

§%

a Ti02W DB cobble » 1Î0 2 melt inclusion

24

22 □ A1203WDB cobble » AI203 melt inclusion

20♦ ■*

18

16

14

Wt. % S i02

Figure 17. Variation diagrams of melt inclusion and whole rock compositions.

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

Page 85: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

7 0

g

I

wt. % Si02

12

10

8

6

4

2

0

6

FeO WDB cobble FeO meit inclusion

Figure 17 (cont.). Melt inclusion and whole rock compositions.

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

Page 86: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

71

8

I

10

8 0 CaO WDB cobble

* CaO melt inclusion

6

4

2

0

8

□ Na20 WDB cobble ♦ Na20 melt inclusion

Wt. % Si02

Figure 17 (cont.). Melt inclusion and whole rock compositions.

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

Page 87: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

7 2

a K 20 WDB cobble

* K 20 melt inclusione

4 -

2000

Q Sr WDB cobble

*■ Sr melt inclusiona,

CO1000 .

wt. % S i02

Figure 17 (cont.). Melt inclusion and whole rock compositions.

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

Page 88: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

7 3

^^Atk ^A r/-*^A r ^•’ArA^'^Ar ^^Ar/^^Ar Atmospheric Age Ca/K(10~^ cm^STP)_____________________________ contamination (%) (m.y.)________

NG B l-1 whole-rock, trachybasalt, K = 1 .4 % \ J = 0.002469 2.32 74.77 0.1871 1.182 73.9

13.83 24.44 0.0189 3.007 22.0Total Gas Age = 83.4 ± 1 .0 , Preferred Age = 83 ± 1 M a^.

NGB3-1 whole-rock trachyte, K = 2.3%, J = 0.0024671.186 46.01 0.0919 1.230 58.9

21.09 19.83 0.00415 1.060 5.8Total Gas Age = 81.2 ± 0.8 M a, Preferred Age = 81 ± 1 Ma,.

NGB7-1 whole-rock microsyenite, K = 2.6%, J = 0.002469 12.42 18.87 0.0107 0.184 16.712.81 18.23 0.0109 0.985 173

Total Gas Age = 67.2 ± 0.8 M a, Preferred Age = 67 ± 2 Ma.

P-1 (3180 ) whole-rock basalt, K = 0.82%, J = 0.002460 1.423 30.85 0.0488 1.077 46.56.65 24.91 0.0239 1.181 28.0

Total Gas Age = 76.7 ± 1 .0 Ma, Preferred Age = 78 ± 1 Ma.

P-l(3530') whole-rock basalt, K = 0.6s%, J = 0.002460 1.542 33.61 0.0640 1.465 56.04.803 22.93 0.0170 1.218 2 1 3

Total Gas Age = 74.9 ± 1.1 Ma, R eferred Age = 78 ± 1 M a

Magnet Cove Biotite, J = 0.006500. Total Fusion 92.8 10.12 0.00643 0.0135 18.8

Granite Mountain Biotite, J = 0.(X)6500. Total Fusion 66.2 10.13 0.00776 0.1306 2 2 3

85.1 ± 3 .2 2 2.1783.1 ± 0 .6 5.51

82.4 ± 5 .8 2 3 5 81.1 ± 0 .3 1.06

68.6 ± 0 3 0.34 65.9 ± 0 3 1.80

71.7 ± 2 .1 1.9777.8 ± 0 .7 2.16

6 4 3 ± 2 .8 2.68 7 8 3 ± 0 .6 2.23

93.6 ± 0 .6 0.028

8 9 3 ± 0 .7 0.267

Woodbine volcanoclastic cobble, K-Ar dating.K = 3.97%, 40at* = 1344 X 1 0 '? cm^STP/g, Atmos. = 1.48%, Date = 9 7 3 ± 1 3 Ma.

40at/39 Ar data corrected for decay of isotopes since neutron irradiation.IK coments o f whole-rocks estimated to be correct to ±10% (see text).2 All errors listed at l o level, internal precision for step ages and including an uncertainty o f 0.5% for the age o f the monitor sample, for preferred, total gas, and total fusion ages.

Table 7, Summary of geochronologic data.

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

Page 89: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

7 4

NGB3-1

I 70 -

60

90

It

I80

70

Preferred Age = 81 ± 1 Ma

Total Gas Age = 81.2 ± 0.8 Ma

-,-----1— 1— r -2 0 4 0

— I— 6 0

—r~ 8 0

Cumulative %39Ar

A.

NGBl-1

1 0 0

Preferred Age = 83 ± 1 Ma

Total Gas Age = 83.4 ± 1 .0 Ma

4 0 6 0 8 0 1 0 0

Cumulative %39Ar

B.

Figure 18. 40Ar/39Ar two-step spectra for subsurface samples. Graphs A,B,C, and D,

are for samples from the Monroe Uplift; graph E is for a sample from the Jackson Dome.

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

Page 90: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

.«j

II

90

G O ­

TO -

GO

90

II

80 -

70 -

60

P-1 (3180')

Preferred Age = 78 ± 1 Ma

Total Gas Age = 76.7 ± 1.0 MaT

2 0T 4 0

T60

T8 0 1 0 0

Cumulative %39Ar

C.

P-1 (3530')

Preferred Age 78 ± 1 M a Total Gas Age = 74.9 ± 1.1 Ma

T2 0

T4 0

T6 0

T8 0 1 0 0

Cumulative %39Ar

D.

7 5

Figure 18 (cont.). 40Ar/39Ar two-step spectra for subsurface samples. Graphs A,B,C,

and D, are for samples from the Monroe Uplift; graph E is for a sample from the Jackson

Dome.

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

Page 91: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

7 6

90

II!

80

70

NGB7-1

Preferred Age = 6 7 ± 2 Ma

Total Gas Age = 67.2 ± 0.8 Ma

4 0

Cumulative %39Ar

E.

1 0 0

Figure 18 (cont.). 40Ar/39Ar two-step spectra for subsurface samples. Graphs A,B,C,

and D, are for samples from the Monroe Uplift; graph E is for a sample from the Jackson

Dome.

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

Page 92: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

7 7

biotite fiom a jacupirangite sample from the Magnet Cove complex (Figure 1) yielded an

age of 93.6 ± 0.6 Ma.

Well cuttings from four samples on the Monroe Uplift and a single subsurface

sample from the Jackson Dome (Figure 1) were acquired; only small sample sizes of

approximately 50 mg were available for A r/^^A r analysis. The four subsurface samples

of trachytes and trachybasalts fix>m the Monroe Uplift yielded preferred, high temperature

40Ar/39Ar ages o f 75.6 ±0.6 Ma, 76.1 ± 0.4 Ma, 80.2 ± 0.04 Ma, and 79.6 ± 0.04 Ma

(Figure 18) (Table 7). A single sample of microsyenite from the Jackson Dome yielded a

high temperature age of 64.6 ± 0.8 Ma (Figure 18) (Table 7).

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

Page 93: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Chapter 4, Summary

1. Lithologie and Mineral Data.

Lithologie and minéralogie data are summarized in Table 8. Clinopyroxene and

feldspar eompositions vary systematieally throughout the voleanie suite. These two

minerals oeeur as distinet phases in eaeh lithology; these phases are easily distinguished by

ehemieal eomposition. Eaeh lithology has distinet mineral eompositions but the overall

variation is systematie and gradational; for example, the eomposition o f intermediate

feldspars grades into felsie whieh grades into pyroelastie (Figure 13). The eomposition of

these feldspar phenoerysts do not appear to be the result of alteration; they are magmatie

eompositions. Very little overlap of phenoeryst eomposition oeeurs; intermediate

elinopyroxene and feldspar eompositions are not found in mafie, felsie, or pyroelastie

roeks.

Melt inelusions are present only in detrital sanidine; sanidine is abundant only in

felsie and pyroelastie roeks. Therefore, the melt entrapped in sanidine must represent

magma that produeed felsie and pyroelastie roeks. Comparison of felsie whole roek

eompositions and melt inelusion data (Figure 17) shows that the two are highly eonsistant;

the most fraetionated whole roek compositions are similar to melt inelusion eompositions.

This supports the idea that the voleanie eobble whole roek eompositions do represent

magmatie eompositions reeording the magmatie history of the suite.

A study involving the use o f epielastie voleanie material to make inferenees

eoneeming the magma at the igneous souree must make use of previous work on

lithologieally and geoehemieally similar roeks. The laek of voleano-stratigraphie

information on age relationships of the various lithologies of the Woodbine eobbles makes

eomparison of the epielastie material to similar, stratified and well-understood sequenees

essential. The Quaternary Laaeher See voleanie deposits (W. Germany) range from highly

evolved, volatile-rieh and erystal-poor phonolite at the base to a mafie, erystal-rieh

78

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

Page 94: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

7)CD■OOQ.C

gQ .

CD

C/)C/)

8"O( O '

3.3"CD

CD"OOQ .C

aO3"OO

Waf.i.g

C px

P la g

S a n

H bld

S p h

A p t

Type 2 Mg, Cr-rich

In te rm e d ia te

Cpx - rich

Type 2 Al, 71 - rich

An 50 - 60

X (altered)

Plag-rich

Type 1

An 3 8 -4 5

X (altered)

F e ls ic

Type 1 Fe, Mn - rich

Cores = An 20 ■ 40

Rims = Or 30- 55; Ba-rich

X

X

X

P v ro c la s tic

Or = 50 - 65 Ba-rich

X

X

D etrita lS a n id in e s

O r=40 - 75 Ba-rich

CDQ .

Table 8. Summary of Mineral Compositions and Rock Type.

■DCD

(/)(/)

CD

Page 95: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

8 0

phonolite at the top (Womer and Schmincke, 1984 a,b). Although these rocks are

composed largely of tephra, they are mineralogically and geoehemieally very similar to the

Woodbine epielastie voleanie roeks. The Laaeher See sequence is dominated by the

minerals sanidine, plagioelase elinopyroxene, amphibole, sphene, hauyne, Ti-magnetite,

apatite, and phlogopite.

The geochemical similarities are also striking. Whole roek major element

eompositions for the Laaeher See suite range as follows from mafie to felsie samples: MgO

= 6.9-0.1 wt. %; S i02 = 49.9-54.4 wt. % ; CaO = 9.0-0.7 wt. %; K%0 = 4.3-5.6 wt. %;

T i02 = 2.1-0.1 wt. % ; AI2O3 = 14.7-21.0 wt. %; Na2 0 = 2.8-11.7 wt. %; P2O5 = 0.4-

0.03 wt. % ; FeO = 1 .2 - 2 3 wt. %. These eompositions are, with the exception o f Na2 0 ,

similar to the Woodbine voleanie whole roek data (Table 1).

The Crater Lake volcanic flows are well-stratified and show regular changes in

ehemieal eomposition and minéralogie mode (Ritchie, 1980). A single eruptive sequence,

with no significant temporal gaps, produeed lithologies beginning with daeitie pumice and

ending with basaltic scoria. Whole rock eompositions ranged from early to late eruptions

as follows: SiO2=70 - 55 wt %, Al203=15 -17 wt %, MgO=<l - 3 wt %, FeO=2 - 5 wt

%, CaO=2 - 5 wt %, Na20=6 - 4 wt %, K20= 3 -15 wt %. Later deposits are dominated

by mafic minerals (olivine, elinopyroxene, orthopyroxene) relative to the earlier deposits

(Ritchie, 1980). Although the Woodbine roeks are of a very different chemistry, more

alkaline and Si-undersaturated compared to the Crater Lake volcanic flows, similar

compositional trends are recognized. Whole rock eomposition within the Crater Lake

volcanic flows range from the first erupted to later deposits as follows: A1203 = 2 0 -15

wt.%, MgO = 0.2 -1 5 w t .%, CaO = 1 - 9 wt.%, K 20 = 5 - 4 wt.%, T i02 = 0 .1 -2

wt.%.

Trace element data from the two suites are also similar. Trace element

concentrations from the Crater Lake volcanic flows are: Ni = 98-4 ppm; Cu =44-20 ppm;

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

Page 96: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

81

Z n = 75-309 ppm; Rb = 81-72 ppm; Sr = 781-5 ppm; Y = 26-43 ppm; Mb = 68-413 ppm;

Ba = 1288-20 ppm.

2. Mineral/melt Partition Coefficients

The importance of realistic partition coefficients in trace element modeling cannot be

overemphasized. Trace elements can be used, even for weathered and metamorphosed

volcanic rocks, to interpret petrogenetic processes (Pearce and Nony, 1979). In order to

model a particular magmatic system, distribution coefficients for that range of magmatic

compositions must be available. Petrogenetic modeling of trace elements quantifies the

extent to which an element is incorporated into the minerals which are crystallizing from a

magma. This concept is expressed by a partition coefficient. Km . which is the ratio of the

concentration (by weight) of an element, M, in a mineral to the concentration in the melt, so

that

Km = [M] mineral / [M] magma

Data from this study has been used to determine partition coefficients for minerals

of the feldspar group, clinopyroxene, hornblende, and sphene. These coefficients will be

helpful to other workers because good partition coefficients for alkaline. Si-saturated or -

undersaturated rocks are lacking. Trace element modeling of alkaline rocks must rely on

partiton coefficients collected from basalts, andésites, andrhyolites which may or not be

accurate for magmatic systems with different bulk chemistries. Partition coefficients

calculated as part of this study are intended to be added to other similar studies in order to

construct a collection of good partition coeffcients for a variety of magmatic conditions.

All partition coefficients were calculated by using electron microprobe data to

determine elemental compositions in the mineral phases, and whole rock geochemical data

to estimate magma compositions. The electron microprobe is the ideal tool to determine

precise mineral compositions and, therefore, is the ideal tool to determine partition

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

Page 97: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

8 2

coefficients. Trace element compositions within a single mineral phase, in a single sample,

can be extremely variable; individual crystals may be strongly zoned with respect to trace

elements. Therefore, a weighted average of the trace element concentrations was used in

coefficient calculations. This weighted average was based on the approximate volume of

phenoeryst with a specific trace element abundances. For example, when calculating an

average Ba concentration, a thin rim of Ba-poor felspar would not be heavily weighted,

while the thick core of Ba-rich feldspar would be more heavily weighted,

a. Feldspar Partition Coeffcients

Distribution coefficients were calculated for Ba, Sr, and Fe for the variety of

feldspars present in the Woodbine suite. This study documents how the behavior of these

trace elements changes with respect to the magma compositions and feldspar type. The

degree of trace element incorporation into feldspars is an important parameter to understand

when attempting trace element modeling. Feldspar partition coefficients of the Woodbine

volcanics are summarized, as:

Ba Sr Fe

plagioelase 0.78 3.5 0.06

anorthoclase 8.0 6.0 0.08

sanidine 6.1 3.2 - -

The Woodbine partition coefficients for plagioelase from mafic rocks are slightly

larger than reported partition coefficients in plagioelase from basaltic rocks (Henderson,

1980). Partition coefficients reported in Henderson (1980) can be summarized as average

KSSplag = 0.23 (the range of reported coefficients = 0.05 - 0.59), and KSrpi^g =1.8 (the

range of reported coefficients = 1.3 - 2.9). Woodbine partition coefficients for alkali

feldspars compare well with coefficients reported for alkali feldspars in Henderson (1980);

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

Page 98: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

83

in dacitic and rhyolitic rocks average KBaalkfld = 6.6 (the range of reported coefficients =

2.7 -12,9) and = 9.4 (the range of reported coefficients = 3.6 - 26).

Experimental work of Guo and Green (1989) has yielded some important

information concerning the effect that temperature, pressure, and composition have on the

distribution o f B ain alkali feldspars in trachytic liquids. In summary they find that 1)

KB^alkfd decreases with increasing temperature, 2) decreases with increasing

pressure, 3) K®^alkfd decreases for anhydrous melt up to 4.2 wt. % H2O, and 4) an

overall decrease in with increasing albite content, although this relationship is not

well defined in the O rio - Or^Q range (Guo and G reen, 1989). Although their

experiments yielded a range of partition coefficients for varying temperatures, pressures,

and melt compositions, the actual values are consistent with Ba coefficients determined as

part o f this study. At 10 kb pressure, varies from 8.4 at 900°C to 6.6 at

1000°C; at 25 kb, coefficents vary fijom 4.3 at 900°C to 2.0 at 1000°C. Partition

coefficients for Ba in alkali feldspars of the Woodbine volcanic rocks are in the range of 5 -

10.

The differences in Ba partition coeffcients within the alkali felspars is not simply a

function o f pressure or temperature differences. If temperature zonation within a magma

chamber were responsible for the difference in Ba partiton coeffcient in the anorthite and

sanidine, the anorthite (from a more mafic, higher temperature rock) would have a smaller

Ba partition coefficient than the sanidine; the experimental work of Guo and Green (1989)

demonstrated that increasing temperature should lower the partitoning o f Ba into the

feldspar. These partition coefficients may instead reflect the changing chemical

composition of the entire sytem with time,

b. Clinopyroxene Partition Coefficients

In clinopyroxnes, Mn, Ti, Ni, and Cr substitute into the Y site, replacing Fe and

Mg. Cr is an important minor constituent in Mg-rich augites; in contrast, Mn is an

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

Page 99: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

8 4

important minor constituent in the ferroaugites (Deer et al., 1966). Clinopyroxene partition

coefficients determined as part of this study are summarized as:

Mn E Ni Q r

cpx: mafic 1.3 0.9 1.1 13.7

cpx: intermediate 1.7 1.6 — —

cpx: felsic 6.7 1.7 — —

The partition coefficients calculated for the mafic Woodbine rocks compare well

with reported Mn, Ni, Cr coefficients for basalts. Partition coefficients reported in

Henderson (1980) can be summarized as average KMn^p^ = 0.9 (range = 0.6 -1.3),

KTicpx = 0.8, KNiqjx = 2.6 (range = 1.4 - 4.4), and KCr^px = 8.4 (range = 4.7 - 20);

partition coefficients for Mn and Ti in clinopyroxenes of intermediate and felsic samples are

not reported. Pearce and Norry (1979) report KTigpx = 0.3 in mafic rocks, KTiq,x = 0.4

in intermediate rocks, and KTi^px = 0.7 in felsic rocks. Woodbine coefficients for Ti in

clinopyroxene are larger than those reported by Pearce and Nony (1979), but are in

agreement with coefficients for mafic rocks reported by Henderson, (1980).

c. Amphibole Partition Coeffcients

The amphibole structure admits extensive substitution. Mn and Ti substitute for Mg

and Fe. A summary of partition coefficients for hornblende of the Woodbine volcanic

rocks follows:

hornblende: mafic

hornblende: felsic

Mq

4.7

13.3

E

1.9

5.8

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

Page 100: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

8 5

No partition coeffcients for Mn or Ti for mafic, intermediate, or felsic rocks were

reported in Henderson (1980). Pearce and Nony (1979) report Ti partition coefficients in

hornblende as KTi^yi = 1.5 in mafic rocks, KTi^y^ = 3.0 in intermediate rocks, and as

KTihbl = 7.0 in felsic rocks. Partition coefficients fom this study are consistent with those

of Pearce and Norry (1979). Partition coefficients for Mn in hornblende were not found in

the literature.

e. Sphene Partition Coeffcients

Sphene, CaTiSiOj, can be an important minor phase in felsic and alkaline systems

(Morris, 1987; Womer and Schmincke, 1984 a,b; Hildreth, 1979), and often contains a

large number of chemical substituents. Alkaline earth and alkaline elements, especially, Sr,

Ba and Na, as well as REE, Th, U, and Y commonly substitute for Ca in the sphene

structure. Al, Fe, Zr, Cr, and Mg substitute for Ti; A1 substitutes for Si. Trace element

substitutions in sphene can have a strong influence on magmatic compositions, and again,

the degree of trace element incorporation into sphene is an important parameter to

understand when attempting trace element modeling. Sphene partition coefficients for the

Woodbine felsic volcanics are summarized as:

U a

sphene 28 9.6

3. Geochronologic Summary

The new radiometric ages determined as part of this study were intended to evaluate

some Gulf Coast Cretaceous igneous rock ages that have been reported in the literature, and

to place some age constraints on previously undated igneous centers. A comprehensive

summary of the ages of the centers may be found in Figure 19.

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

Page 101: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

8 6

Balcones

PrairieCreek

MagnetCove

VIntrusive

BentonDike

Swarm

PotashSuifur

Springs

GraniteMountain

mmKmmmm

MonroeUpiift

JacksonDome

Ma 60 70 80 90 100 110

Ages from the literature (see text)

Ages determined as part of this study

Figure 19. Summary of geochronologic data for Cretaceous igneous rocks of the northern

Gulf of Mexico Basin

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

Page 102: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

8 7

Two isotopic ages have been previously reported for volcanic rocks from the

Monroe Uplift Sundeen and Cook (1977) report a K-Ar date on biotite from a biotite

analcimite of 91 ± 3 Ma, and a K-Ar whole rock age of a phonolite of 78 ± 3 Ma. The four

subsurface samples analyzed as part of this study yielded 2-step ages of 75.6

±0.6 Ma, 76.1 ± 0.4 Ma, 80.2 ± 0.04 Ma, and 79.6 ± 0.04 Ma. These new data are

consistent with the ages, particularily the younger age, reported by Sundeen and Cook

(1977) (Figure ). Monroe (1954) interprets the volcanism that is preserved on and

apparently accompanied the formation of the Jackson Dome as the result of Cenomanian

volcanoes that were eroded until Santonian time (97 - 84 Ma). A single sample analyzed as

part of this study yielded and age of 64.6 ± 0.8 Ma. This date is significantly younger than

the age constraints presented in Monroe (1954). The new data can be integrated with the

old in two ways: 1) the volcanism may have extended over a relatively long time interval so

that both sets of data are interpreted as meaningful, or 2) the interpretation of Monroe

(1954) may have been in error so that the new data is interpreted as the correct age of the

volcanism. The A r/^^A r date of the igneous rock of the Jackson Dome certainly

represents an age of a period of igneous activity for that tectonic area; with the present

available data, it is impossible to determine whether the igneous activity extended over a

longer time interval.

The ^^Ar - 39Ar age determined for the Magnet Cove complex, 92.9 ± 0.6 Ma,

compares well with the reported K - A r biotite age of 95 ± 5 Ma of Zartman et al. (1967).

Other geochronological data, including fission tracks in apatites (Scharon and Hsu, 1969)

and a K - A t whole rock date of a Magnet Cove trachtye (Baldwin and Adams, 1971), are

slightly older than, yet still consistant with, the new ^^Ar - 39Ar age. The Ar - Ar age

determined for the Granite Mountain syenite, 89.5 ± 0.7 Ma, compares well with reported

ages of 87 ± 4 and 9 1 + 5 Ma (K- Ar on biotite) and 86 + 3 Ma (Rb - Sr on biotite) of

Zartman (1967).

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

Page 103: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

8 8

The K - A r whole rock age of 98 ± 2 Ma determined on a trachyte preserved as a

cobble within the Woodbine Formation compares well with stratigraphie age relationships

of the sedimentary unit itself; although stratigraphie relationships alone do not provide a

narrow age bracket for the age of Woodbine deposition, the sediments are restricted to early

Late Cretaceous (97 - 91 Ma) (Ross et al., 1928). It appears that the volcanism that

produced the Woodbine extrusive rocks only slightly predates the deposition of the

reworked cobbles.

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

Page 104: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Chapter 5. Petrogenesis o f the Volcanic Suite Preserved as Cobiles

A petrogenetic understanding of the chemical and mineralogical relationships and of

the evolution of the epiclastic extrusive rocks of the Woodbine formation involves:

determination of whether the various volcanic lithologies preserved as detrital material are

cogenetic and explanation of the mineralogical and chemical variation of the different

lithologies.

Differing degrees of partial melting of the mantle, ftactional crystallization,

assimilation, and magma-mixing are processes which can explain the formation of a wide

variety of rocks from a single parental material. Igneous systems are extremely

complicated and several of these processes may be important during the magmatic evolution

of any one system. Because differences in chamber composition, size, temperature, depth,

and life-span can affect fractionation history, no one model of magma chamber evolution

may be applied to all systems. Despite these complex, inter-related variables, mechanisms

of magmatic evolution are being qualitatively and quantitatively examined and are becoming

better understood and evaluated.

Examination of stratified, volcanic deposits has proven useful in understanding the

evolution of magmatic systems. It has been demonstrated that the extrusive deposits of

evolved magma chambers commonly exhibit systematic mineralopcal and chemical

variations that reflect pre-emptive compositional stratification of the magma bodies

(Hildreth, 1979; Ritchie, 1980; Womer and Schmincke a,b, 1984; Fino et al., 1986;

Barberi et al., 1988; Hooper, 1988). Continuous chemical variation in stratified,

undisturbed volcanic deposits provides evidence that such deposits are cogenetic (Womer

and Schmincke, 1984a,b). Single eruptive units can be used to represent a pre-emptive

chemical state of a magma chamber and successive emptive units can be used to reconstmct

the evolution of a chamber through time or space. Compositional stratification of volcanic

deposits may reflect 1) inverted, pre-emption, magma chamber zonation in which case

8 9

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

Page 105: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

9 0

lowest deposits (from the top of the chamber) are enriched in incompatible and alkaline

elements and volatiles, whereas upper deposits (derived from the bottom of the chamber)

are more mafic (Womer and Schmincke, 1984; Fino et al., 1986; Barberi et al., 1988;

Hooper, 1988), or 2) periodic tapping of an evolving chamber in which case the volcanic

deposits range from mafic at the bottom to more differentiated rocks at the top.

Whole rock major and trace element concentration gradients and mineral

composition variations within stratified volcanic suites have been used by many workers to

evaluate possible fractionation mechanisms within the pre-emptive magma chamber, crystal

liquid fractionation, magma mixing, and assimilation-fractional crystallization (AFC) are

several mechanisms that can be supported or discredited by analysis of undisturbed

volcanic sequences. Perhaps the best known study involving examination of stratified

volcanic deposits and inferences concerning magma processes is that of Hildreth (1979).

The 0.7 Ma Bishop Tuff exhibits field, minéralogie, and compositional evidence that have

been interpreted as the result of continuous tapping of a thermally and chemically zoned

siliceous magma; these minéralogie and chemical gradients existed in the magma prior to

phenocryst precipitation and developed by the combined effects of convective circulation

intemal diffusion, complexation, and wall-rock interchange (Hildreth, 1979). Examination

of 8 major, stratified eruptive units showed that the eruptive sequence reflects a simple and

systematic tapping of progressively hotter and deeper levels in the magma chamber,

early/late deposits (top of the chamber/bottom of the chamber) enrichment factors were

determined by comparison of elemental concentrations in the lower and upper volcanic

deposits. Ba, Sr, Eu and Mg were extremely depleted in the magma chamber's roof zone;

U, Nb, Sc, Mn, HREE, and Y were enriched in this upper chamber zone (Hildreth, 1979).

Once documented, the origin of this magmatic zonation was investigated. Crystal settling,

assimilation, liquid immiscibility, progressive partial melting, and convection-driven

thermogravitational diffusion were considered as processes that may have been responsible

for the compositional zonation. After analysis of abundant whole rock chemical data and

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

Page 106: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

91

phenocryst compositions Hildreth was able to exclude crystal fractionation, partial melting,

assimilation, immiscibility and magma mixing as processes responsible for the zonation;

the data support liquid state diffusional processes as the zonation mechanism of this high-

level rhyolitic chamber (Hildreth, 1979).

Quantitative mass balance least squares analysis of whole rock and mineral

compositions of volcanic strata can be applied to a number of pétrographie problems

including magma mixing and liquid Une of descent calculations. Very simply, these models

solve petrologic equations of the type

basalt = andésite + olivine + clinopyroxene ( 1 )

andésite = basalt+ rhyolite (2)

biotite + sülimanite + quartz = garnet + K-spar + water. (3)

Comparison of the calculated solution to the actual rock suite allows evaluation of

firactionation mechanisms. Even in cases of complex magmatic history, modeling of

geochemical data can be used to evaluate fractionation processes. A study o f this sort was

conducted by Barbieii et. al (1988). Modeling of major, trace, and Sr isotopic data on Si-

undersaturated lavas from the Vico volcano. Central Italy indicated that fractional

crystallization was an important evolutionary process and that AFC and magma mixing

were also important processes. Specifically, the early erupted trachytes of this suite were

derived from mafic lavas by AFC and the late magmas, already fractionated by crystal

precipitation, were intruded by primitive liquids forming a hybrid magma (Barbieri, et. al,

1988).

A study involving stratified. Si-undersaturated, volcanic rocks of the Laacher See

tephra sequence (West Germany) revealed strong compositional zonation interpreted as the

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

Page 107: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

9 2

result of continuous eruption of a zoned magma chamber (Womer and Schmincke, 1984 a,

b). Highly evolved rocks from the stratigraphically lower extrusive deposits were

interpreted as early erupted material &om the top of the zoned chamber, more primitive

rocks from the bottom of the chamber were erupted last and were deposited on top of more

evolved rocks. Mass balance fractionation calculations were employed to test the

hypothesis that fractional crystallization was responsible for the chemical variation found

within the Laacher See suite. Model liquid compositions that are similar to actual rock

compositions support crystal fractionation as an important process in the generation of tis

suite of rocks (Womer and Schmincke, 1984 b). Mafic phonolites preserved in the upper

deposits were modeled as parent material for more evolved felsic samples; removal of

varying proportions o f feldspars, clinopyroxenes, amphiboles, magnetite, apatite and

sphene resulted in liquids very similar in composition to the zoned extrusive sequence.

Mafic mineral phases became less important during fractionation compared to sanidine in

highly evolved rocks of the middle and lower deposits.

1. Are the cobbles cogenetic?

A petrogenetic model for volcanic rocks preserved as cobbles must first address

whether or not the cobbles are cogenetic. Systematic changes in mineralogy throughout

the volcanic suite indicates a cogenetic relationship. A continuous gradation in changes in

modal mineralogies exists within the Woodbine volcanics. Clinopyroxene-rich rocks,

clinopyroxene-rich rocks with minor plagioclase, plagioclase-rich rocks with minor

clinopyroxene, and alkali-feldspar-rich rocks with minor clinopyroxene are all present in

the Woodbine suite. This gradation supports a cogenetic origin for the set of volcanic

rocks; reworked volcanic rocks from various igneous suites might exhibit more lithologie

variation.

Continuous chemical and mineralogical zonation of the Laacher See volcanic

deposits is interpreted as the result of the eruption of a single zoned magma chamber.

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

Page 108: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

93

Chemical and mineralogical gradients in the deposits indicate that a continuous

differentiation series existed within the magma body prior to eruption with an increasing

degree of differentiation toward the top. Similar chemical and mineralogical variation

within Woodbine volcanic rocks indicates that the variation may be the result of

differentiation of a single, cogenetic magma system.

Coherent patterns on major and trace element variation diagrams (Figure 7) support

the hypothesis that the Woodbine volcanic cobbles were derived from a single evolving

magmatic center, if the cobbles were derived from a number of unrelated volcanic sources,

there would be considerably more scatter on these diagrams (Womer and Schmicke, 1984;

Hyndman, 1972). In order to further evaluate this hypothesis, variation diagrams

comparing Woodbine volcanic rock chemistry and other Upper Cretaceous Gulf coast

igneous rocks were prepared. Major and trace element data from Magnet Cove (Erickson

and Blade, 1963), Granite Mountain (Morris, 1987), Balcones Province (Spencer, 1969),

and Benton Dike Swarm (Morris, 1987; Robison, 1977) were plotted along with data from

the Woodbine volcanic rocks (Figure 20). Comparison of intrusive and extrusive rocks

can be misleading. Plutons are extremely complex; parameters such as grain size, color,

and composition can be extremely variable over short distances within a pluton. The

composition of plutons are dramatically affected by late stage processes such as cumulate

and pegmatite formation, and aqueous fluid movement Despite these complexities, these

diagrams quickly reveal how much chemical variation exists between these suites. Some

elements are more useful than others in discriminating individual centers; Ti02 vs. MgO

shows some overlap between the suites, although Magnet Cove has a distinctly higher

T i02 signature and Balcones rocks have a lower T102 trend (Figure 20). Lamprophyric

dikes of the Benton Dike Swarm show a Ti02 trend similar to, and consistent with more

primitive Woodbine lithologies. FeO also reveals obvious variation in concentration with

Balcones and Magnet Cove rocks having significantly higher FeO concentrations than

Woodbine volcanics (Figure 20). MnO also shows considerable variation between the

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

Page 109: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

9 4

§

20 4 6 8 10

n —Woodbine

-Magnet Cove

àtanite Mountain

— •— Balcones

— a — Benton Dike

MgO

O0) 1 0 -

20 4 6 8 10

■,H -Woodbine

- 4— -Magnet Cove

-a— Granite Mountain

. . .4 Balcones

— n— Benton Dike

MgO

Figure 20. Whole rock geochemistry of rocks from several Gulf Coast igneous centers.

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

Page 110: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

95

i

§

8

6

4

2

00 2 4 6 108

MgO

MgO

30 4- 0 2 4 6 8 10

a -'Woodbine

-Magnet Cove

Branite Mountain

» - Balcones

. . . I n -Benton Dike

-B— Woodbine

-Magnet Cove

-a— Granite Mountain

fr Balcones

B Benton Dike

Figure 20 (cont.). Whole rock geochemistry of rocks from several Gulf Coast

igneous centers.

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

Page 111: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

9 6

suites with Magnet Cove, Granite Mountain, and Balcones rocks having distinctly higher

MnO concentrations than Woodbine volcanic rocks (Figure 20). K2O vs MgO reveals

considerable overlap between Balcones rocks and the Woodbine volcanic rocks, however.

Magnet Cove and Granite Mountain rocks have distinctly lower K2O trends (Figure 20).

Every element examined indicates that Benton dike lamprophyres and primitive Woodbine

rocks form coherent trends on variation diagrams.

If the Woodbine volcanic cobbles were derived from a number of Gulf Coast

Cretaceous igneous centers, variation diagrams would show a wider scatter of data as

opposed to the reasonably well constrained trends that are observed.

2 Minéralogie and Chemical Variation

DiflFering degrees of partial melting of a mantle source, magma mixing, crystal

fiactionation, and assimilation and fractional crystallization (AFC) are processes which

could be responsible for the variation observed within the Woodbine volcanic rocks.

General patterns on variation diagrams along with certain trace element parameters can be

used to identify the process most likely responsible for the Woodbine variation. For

example, primary mantle melts have trace element compositions of Ni = 350 ppm and Cr =

900 ppm. Estimated mantle compositions for Ni and Cr are 2400 ppm and 2700 ppm,

respectively, and bulk distribution coefficients for peridotite are D ^i = 7 and D cr = 3; Ni

and Cr concentrations for primary mantle melts, over a wide range of partial melting, can

be determined (BVSP, 1981; Hooper, 1988). Mafic igneous rocks with these trace element

compositions are good candidates for primary mantle melts. Rocks with lower Ni and Cr

concentrations must represent more fractionated magmas. Mg # (Mg / (Mg+Fe)* 100) can

also be used to help identify a primary mantle melt High Mg #s, values of 60-70, are

typical of primary mantle melts. Lower Mg #s represent more fractionated melts; the

smaller the Mg#, the more differentiated the melt.

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

Page 112: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

9 7

General petrogenetic interpretations can be made from examination of variation

diagrams (Cox et al., 1979). Examination of the patterns, or trends, on variation diagrams

can yield early hypotheses regarding the petrogenetic process responsible for the observed

variation (Figure 21). Smooth, continuous curves of any type on variation diagrams

suggest that the trends are the result of differentiation of a single magma (Womer and

Schmicke, 1984 a,b). Closer examination of the curves may indicate the process

responsible for the differentiation. Straight line variation may simply represent a mixing

curve; this simple linear variation may be the result of the mixing of two end members

(Figure 21a). The two end members may both be liquid, as in magma mixing, or one end

member may be solid as in assimilation of a single type of country rock. Simple linear

variation may also be the result of partial melting variations, or of fractional crystallization;

linear variation is often difficult to interpret because several different magmatic process

could result in this chemical trend. Continuous trends that are not simple straight lines, for

example trends that are composed of curved segments, parallel segments, or linear

segments with distinct kinks, indicate that processes other than mixing are responsible for

the variation (Figure 21b,c,d). Multiple sources, multiple differentiation processes, and

individually evolving magmatic systems could result in parallel segments. Fractional

crystallization of a single magmatic body results in smooth variation lines that represent a

liquid line of descent for that magma; the type and assemblage of mineral phases

precipitating out of a magma will determine the whole rock geochemical variation patterns.

A curved liquid line of descent may be the result of precipitation of a solid solution series or

changing proportions of two precipitating phases (Figure 21c). Kinks in the trends of

these variation diagrams indicate the addition of a new phase (or the loss o f a phase) to the

precipitating assemblage (Figure 2 Id). As mafic minerals crystallize in a magma chamber,

the residual liquid commonly evolves to a more fractionated magma that is enriched in

alkalis and SiO% and depleted in FeO, MgO, CaO, and TiO%. A change in the precipitating

mineral assemblage will cause the magma to evolve chemically along a different path and

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

Page 113: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

CD■DOQ.C

gQ.

■DCD

C/)C/)

8■o

CD

■*i+.

- magma mixing -assimilation -partial melting -fractional xtal

-multiple sources -multiple differentiation

processes

■ " + +

3.3"CD

A. SitTtple linear varaition B. Multiple linear trends

CD■DOQ.

C/)C/)

ca Fractional crystallization with Fractional crystallization withO3 -xtal of solid solution phase -change in xtal. assembladgc"O ^ -changing phase proporttions -new phase begins pet.O3"cr

-loss of pet. phase

1—kCDÛ.§ «—k3" %O

%

C. Curved variation D. Linear trend with bend

Rgure21. Idealized whole rock variation diagrams. See text for discussion.

COCO

Page 114: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

99

thereby cause a kink in the resulting variation diagram. A scatter of points at the mafic end

of the curve is common, and is often due to crystal accumulation (Cox et al., 1979).

Each of the four differentiation processes, partial melting, magma mixing, fractional

crystallization, and AFC will be evaluated as potential causes of the Woodbine minéralogie

and geochemical variation. An examination of the principles described above will be

applied to the Woodbine volcanic suite.

A. Partial Melting.

Low Mg #'s (31-50), low Cr concentrations (0-70 ppm), low Ni concentrations (0-

60 ppm), and high La (80 -140 ppm), and high Nb concentrations (90 - 230 ppm) argue

against the hypothesis that any of the Woodbine volcanic rocks represent primary mantle

melts. The most primitive rocks of this suite could not be in equilibrium with a mantle

assemblage. Even the most primitive Woodbine cobble, WDB-86-2A (Mg# = 49, Ni = 63

ppm, Cr = 69 ppm), must have undergone substantial differentiation, either by

ferromagnesian fractionation or by crustal contamination, before eruption. Therefore, the

variation observed within the Woodbine suite is not the result of different degrees of partial

melting of a mantle source.

B. Magma Mixing

Some of the Woodbine variation diagrams exhibit simple "straight line" variation.

CaO, FeO, Ti02, and Nb variation diagrams (Figure 8) are, at a first approximation,

curves made up of a single segment or a single straight line. These diagrams indicate that

magma mixing may be the process responsible for this chemical variation. Mixing of 2 end

member magmas, one felsic and one mafic, could result in a variety of volcanic rocks of all

intermediate compositions. This process could produce a suite of igneous rocks with

substantial geochemical and mineralogical variation. The hypothesis that magmatic mixing

is the cause of the chemical variation of the Woodbine volcanics is not supported by other

important geochemical and pétrographie data. Major elements such as AI2O3, MnO, K2O,

P2O5 , and many trace elements including Zr, Y, Sr, Ni, La, and Ba do not exhibit simple

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

Page 115: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 0 0

linear variation, and therefore do not support the magma mixing hypothesis. Mixing of

two end member liquids could not produce the segmented variation of these elements. No

pétrographie evidence in favor of magma mbcing, such as chemically distinct inclusions

with quenched or chilled rims (Bloomfield and Axculus, 1989) has been identified.

C. Fractional Crystallization.

Variation diagrams of volcanic rocks of the WoodbineFormation can be best

described as smooth curves with distinct kinks. These kinks define the segments identified

on the variation diagrams. A working hypothesis to explain the major element variation of

this suite is that fractional crystallization of a single magma resulted in the variety of rocks

present in this suite, and that kinks in the variation trends reflect changes in the precipitating

mineral assemblage.

Sample lithology, variation diagrams, and realistic partition coefficients support the

idea that a single evolving magma is responsible for the minéralogie and chemical variation

observed. The variation o f Sr and Y vs. MgO is composed of four distinct segments

(Figure 22). Partition coefficients are presented to the right of Figure 22. K^^piag was

determined by microprobe and whole rock data collected as part of this study; other

partition coefficients are firom the literature. Qualitative examination of the variation

diagrams is the first step in determining whether crystal fractionation is a viable process for

this suite of rocks. Along the first segment of the Sr variation diagram, Sr is increasing

with deceasing MgO; plagioclase (KSr=3.5) could not have been a major precipitating

phase as these rocks crystallized (Figure 22). Such a trend is consistent with dominant

clinopyroxene (KSr=0) fractionation; volcanic cobbles that make up this part of the trend

are dominated by clinopyroxene phenocrysts. Significant fractionation of clinopyroxene

fiom the most primitive Woodbine sample, WDB-86-2A, could qualitatively explain the

variation in Sr and Mg concentrations observed within this segment and is consistent with

the mineral assemblage present in these rocks.

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

Page 116: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

101

2000

Sr (ppm)

1000 ■

= 0.0cpx

=3.5plag

K * ' = 0amph

MgO

40-

2 0 -

10-

YK =1.5

cpxY

K , =0.1 plagY

= 0.0KsparY

K =2.5 amph

MgO

Figure 22. Sr and Y vs. MgO variation diagrams and relevant partition coefficients.

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

Page 117: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 02

Similar observations can be made along the other 3 segments. The second and third

segments show a slight and moderate decrease in Sr with decreasing MgO. These patterns

are consistent with subequal proportions of clinopyroxene and plagioclase precipitation

with the second segment dominated by clinopyroxene and the third segment dominated by

plagioclase fractionation. Plagioclase preferentially incorporates Sr into its structure and

depletes the residual melt in Sr. Again, the samples that make up these two segments have

mineral assemblages that are consistent with interpretations made using the variation along

these segments and these partition coefficients. The fourth segment shows a strong

depletion in Sr with decreasing MgO; this is consistent with significant fractionation of

plagioclase or sanidine and compares well with anorthoclase as the dominant phenocryst in

samples from this segment

Similar qualitative interpretations can be made by examining the Y variation diagram

(Figure 22). Again, a curve of four distinct segments comprises the total variation. The Y

variation, however, is opposite to that o f Sr. The first segment relates a strong depletion

of Y with decreasing MgO. The second and third segments reveal a constant Y

concentration and strong increase in Y, respectively. The fourth segment exhibits Y

depletion. Qualitatively, the mineralogy of this suite of rocks and partition coefficients

explain these variation patterns. Y is compatible in clinopyroxene (K^cpx = 1.5 - 4.0) and

amphibole (K^amph = 2.5 - 6.0) and incompatible in feldspars (K^fldsp = 0.06 - 0.1).

The strong Y depletion in the first segment (Figure 22) reflects the dominance of ferro­

magnesian minerals in the fractionating assemblage. The change in slope in the second and

third segments reflects the increasing importance of feldspar precipitation; as feldspar

becomes more abundant, Y is increasingly excluded from the crystal fraction and becomes

enriched in the melt. The depletion of Y in the foiirth segment may reflect incorporation of

Y into sphene; sphene becomes an important phase in rocks of the fourth segment.

Qualitative assessment of trace element variation supports the working hypothesis that all of

the volcanic cobbles of varying lithology may be related to each other by crystal

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

Page 118: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 0 3

fractionation, and the changes in slope of line segments on the variation diagrams may

represent changes in the precipitating mineral assemblage.

i) Major Element Models

Modeling of major, minor, and trace elements was employed in order to further test

the hypothesis that crystal fractionation is the cause of the variation. These calculations can

test the viability of fractional crystallization as the dominant differentiation process

(Womer and Schmicke, 1984 b). The concept of crystal fractionation can be summarized

by:

magma(parent) - (x) minerali - (y) mineralg - (z) minerals =magma(diffemtiate)-

where (x), (y), and (z) are the proportions of the minerals that must be removed to derive

the residual magma. Compositions of the primitive and residual magmas are estimated with

whole rock compositions.

Modeling of crystal fractionation can be easily understood and visualized with

variation diagrams. Figure 23 illustrates how precipitation of various minerals from an

initial primitive magma, represented by rock sample WDB-86-2A, would affect the

composition of the residual liquid. Individual mineral arrows on each variation diagram in

Figure 23 represent the path that the residual liquid would take upon removal of that

mineral. The lengths of the arrows qualitatively represent the magnitude o f the effect that

phase has during fractionation; the length of the plagioclase arrow on each variation

diagram represents ~30% plagioclase precipitation, the length of the clinopyroxene

represents ~30% precipitation, the length of the hornblende arrow represents ~9%

precipitation, the length of the Ti-magnetite arrow represents ~9% precipitation, and the

length of the apatite arrow represents ~5% precipitation. A fixed amount of crystallization

o f a phase has varying effects on the different elements within the residual liquid; for

example, precipitation of Ti-magnetite strongly influences the Ti02 and FeO concentrations

of the residual magma and the Ti-magnetite arrow is relatively long on the Ti02 and FeO

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

Page 119: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 0 4

g(O

oF

60 -

Ti-mag

apthbld50 -

cpxp l ^

2 4 60 8

MgO5

4 plag

apt3 cpx

hbld2

Ti-mag00 @ 0

1

020 4 6 8

MgO

Figure 23. Model residual liquid compositions after removal of single mineral phases. See

text for discussion.

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

Page 120: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 0 5

§

22 -

*b H20 -

Ti-mag18 .

cpxapt16 -

hbli

14 . plag

0 2 4 6 8

MgO

O

14 - plagcpx12 -

hbld10 -

Ti-mag

0 2 4 6 8

MgO

Figure 23 (cont.). Model residual liquid compositions after removal of single mineral

phases. See text for discussion.

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

Page 121: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 0 6

Oc

§

0.3

0.2 -

0.1

0.0

Cpx□ Ti-maga

m a apthbldB

—T"2

“I"4

*T“6

MgO10 Ti-mag

hbl8

apt6

4

cpx2

02 4 60 8

MgO

Figure 23 (cont.). Model residual liquid compositions after removal of single mineral

phases. See text for discussion.

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

Page 122: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 0 7

i

I

8

7

6

5

cpxTi-mag4

hbl, B p t34

MgO

□ D Ti-ma<cpxbld plag

apt

0 42 6 8

MgO

Figure 23 (cont.). Model residual liquid compositions after removal of single mineral

phases. See text for discussion.

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

Page 123: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 0 8

variation diagrams (Figure 23 b,d). Removal of the same 9% Ti-magnetite does not

strongly affect the Si02 composition of the residual melt.

The effect that precipitation of individual phases has on the composition of the

residual liquid can be seen in Figure 23. Removal of clinopyroxene (MgO = 11.71, CaO =

22, SiOz = 44.57, T i02 = 3.07) from an initial primitive magma (MgO = 5.50, CaO = 8.8,

S i02 = 49.30, T i02 = 2.76) would result in the residual liquid becoming depleted in MgO

and CaO without a large change in Si02 or T i02 (Figure 23 a,b,f). Precipitation of

plagioclase (Si02 = 53, FeO = 0.4, MnO = 0, T i02 = 0, K2O = 0.7, AI2O3 = 28, CaO =

11) results in a residual liquid that is enriched in MgO, Ti02, FeO, MnO, and K2O, and

depleted in Si02, AI2O3, and CaO. Removal of an indivddual phases tends to "drive" the

liquid away from the composition of that phase.

Major element modeling of any suite of volcanic rocks involves mass balance

calculations that determine the composition of theoretical residual liquids. Modeling the

effects of multiple phase precipitation proportionally combines all of the mineral path

arrows of Figure 23 into a single path of theoretical residual liquid composition (Figure 24,

25,26). The correct fractionating proportion of each phase is chosen so that the theoretical

residual liquid is chemically similar to the suite of differentiated rocks. A good fit for the

modeled residual liquid composition and the fractionated rock composition supports the

hypothesis that fractional crystallization is the process responsible for the chemical variation

of the suite. This "best fit" approach results in a set of theoretical fractionating mineral

proportions; these proportions must be checked against the mineral assemblage present in

the rocks. A modeled composition of a residual liquid is of no use if it is derived by a

modeled mineral suite that is geologically inconsistent with the volcanic rocks being

modeled. On the other hand, if the mineral proportions determined by the best fit mass

balance calculations are in agreement with the observed mineral proportions, then the

support for the model is strengthened. Igneous systems, however, are extremely

complicated; there are uncertainties about the depth and pressure of the crystallizing magma

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

Page 124: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 0 9

chamber and therefore, uncertainties about the fractionating assemblage itself (Hooper,

1988). High pressure phases may have influenced residual liquid compositions; at low

pressures, these phases may have become unstable and are no longer present.

Crystal fractionation of Woodbine volcanic rocks was modeled in a series of steps

corresponding to changes in segment slope on the whole rock variation diagrams (Figure 9,

10); it was shown earlier that these changes in slope may represent changes in the

precipitating mineral assemblage. Four segments have been identified within the overall

Woodbine chemical variation (Figure 10). The rock samples that compose the first,

second, and third segments are consistent on all variation diagrams; each of those segment

boundaries are defined by the same samples for all elements, and the placement of samples

within the segments is consistent for all elements. The fourth segment, composed of the

most fractionated rocks, does not exhibit the same consistency. The fourth segment is

composed of the same samples on all of the variation diagrams, but these samples are not

systematically arranged within the fourth segment of all the elemental variation diagrams;

the samples that define the segment boundaries are not the same for all elements. In

addition, many major and minor variation diagrams (Si02, Ti02, FeO, CaO, P2O5)

(Figure 8) do not show any chemical variation within the fourth segment. Because of these

patterns, the fourth segment was not subjected to major and trace element modeling

techniques; the fourth segment will be examined separately.

Quantitative examination of the differentiation process in which crystals are

removed from a liquid was described by Wright and Doherty (1970) as:

X (parent) • S xjyi = ynX(differentiate), where

X = chemically analyzed constituents of a rock

xi = wL % of constituent X in mineral i

yi = wL % of mineral i in the rock

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

Page 125: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 1 0

In order to model the crystal fractionation process, the above equation was set-up

for each oxide used to describe the suite of rocks and minerals. The composition of the

most primitive magma was estimated with the most primitive rock (WDB-86-2A); by

subtracting specific amounts of mineral compositions, a theoretical residual liquid was

determined. An important parameter, F, is defined as the amount of liquid remaining after

fractionation; F = 1 when the system is all liquid, and F = 0 when the system is all solid.

Using the notation of Wright and Doherty,

F = 100 - X yn.

The endpoints of the first segment are defined by the samples WDB-86-2A and

WDB-88-1E. For each element, minerals observed as phenocrysts in mafic samples, using

mineral compositions determined by the microprobe, were subtracted from the primitive

sample:

SEGMENT 1: WDB-86-2A - (w) clinopyroxene -(x) amphibole - (y) Ti-

magnetite - (z) apatite = theoretical residual magma

where (w), (x), (y), and (z) are proportions of minerals removed.

Various proportions of different phenocryst phases were removed from the

primitive rock in an attempt to derive a theoretical residual magma which is similar in

composition to the composition of the more differentiated rock (WDB-88-1E). Ten major

and minor elements have been modeled in the Woodbine system. A solution, consisting of

phenocryst proportions which derives a theoretical residual liquid that compares best with

the differentiated rock, was determined for segment 1 by trial and error. This best fit

solution for segment 1 is summarized in Table 9. These ten elements have been modeled

together; one solution of a set of mineral proportions has been applied to all elements. Each

element, however, is graphically represented on an individual variation diagram (Figure

24).

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

Page 126: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

111

D )(0E.JLH

J Z

a .En

îo

I<o

+UJT “

É(ON

<200

oo

m

O

COoCÔO)

oodo

oCOO)

pCOO)

■M-COdm

CMCOO)

oodo

qdo

inO<N0 .

sd d

ood

ood

ood

ood

sd

COq

oCM

od

5CÔ

opd

opd

s sd

sCO

on

oCM

CMCM

RCM

Kd

opd

oCM

CMi n

CO

CM%CM

z

o(0O

CMOOCO

O)cnCO

COfs.dCM

COpd

00i nc i

COCO

COCOd d

OoS

oi nU )

( 0i n

opd

%CM

O)oCO

ood

CM ow■M"

oc

ind

i nd

CD

d

Ood

oCOd

ood

COT-d

CO

d

s

od>u_

«d

sd

CMCOCO

9

iS:

oCO COCOd

CO g

toOw<

8i n

si n

CO%

op

00M

s

sd

hd

CMOP

Ko i

0 0c i

ooi n

Oin

o<pd

CMCMCM

r -T—cvi

CMOtn

o«0 )

COCMdi n

i nCOi n

o

dindCO

ins

TtinCMin

qi n

1CO

dCOd

i nC»d

o<pd CON

d< < OJ CM

to toCO CO

Eoc

Xao

O)(0E1 -

£aE(0

a.0)•a«■ooE

IIU - u j _

^ ECO o OO c

It/3

11

IOQo\

I

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

Page 127: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 1 2

i

8

OQ6

4

2

00 1 2 3 4 5 6

MgO

O

10 .

0 1 2 3 4 5 6

MgOFigure 24. Composition of modeled magma evolution for segment 1.

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

Page 128: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

113

CO

§<

24

22

20

18

16

140 1 2 3 4 5 6

MgO

§

10

8

6

4

2

00 2 4 51 3 6

MgO

Figure 24 (cont). Composition of modeled magma evolution for segment 1.

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

Page 129: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 1 4

For the first segment, the modeled composition for the elements SiOi, TiOi, AI2O3, FeO,

MgO, CaO, Na20, and K2O are within 5% of the most differentiated rock of that segment

(Table 9). The modeled residual liquid composition for these elements compares extremely

well with the hypothesized fractionated rock. The remaining element MnO and P2O5

model reasonably well, but do have greater than 5% variation between modeled and actual

concentrations. The elements Si02, Ti02, FeO, MnO, and in particular Na2 0 reveal some

degree o f scattering along this segment (Figure 8). Since Na2 0 exhibits a great deal of

scatter throughout the variation diagram (Figure 8), it will be considered separately at the

end of this section. The first segment was modeled with 88-IE as the most fractionated

sample, however, the general differentiation trends and model trends can be compared in

Figure 24. Examination of these trends can give a better feel for how well the model

results fit the actual data than the model summary in Table 9. All 10 elements, even MnO

and P2O5 , have model trends that compare well with the actual chemical variation of the

rock suite. The model fractionation path for segment 1 and the actual variation of mafic

Woodbine volcanic rocks are compatible.

Identical techniques were used to model the next segment:

SEGMENT 2: WDB-88-1E - (v) clinopyroxene -(x) amphibole - (w) feldspar -

(y) Ti-magnetite - (z) apatite = theoretical residual magma

Again, the theoretical residual magma was compared to the most differentiated

sample of the segment (Table 10).

The second segment model results can be found in Table 10 and these results are

presented graphically in Figure 25. Sample 88-IE has been defined as the primitive end of

this trend whereas four rocks, 86-2E, 87-1 A, 87-ID, 87-lF define the more fractionated

end. These four samples appear similar petrographically and in hand sample, and cluster

together on almost every variation diagram. For modeling purposes, 87-ID was used as

the fractionated end (model results were 'best fit' to 87-ID), although any of these four

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

Page 130: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

CD■DOQ.C

gQ.

■DCD

C/)Wo'303CD

8■D( O '3"

13CD

"nc3.3"CD

CD■DOQ.CaO3■DO

CDQ.

■DCD

(100) 88-1E = (85) 87-ID + (6) am ph + (4) cpx + (3) fidsp + (2) Ti-mag + (1) ap t

% xtal S I0 2 T I0 2 A I2 0 3 F e O M nO MqO C aO N a 2 0 K 2 0 P 2 0 5 total

8 8 -1 E 48.41 2.06 18.73 7.30 0.15 3.98 6.36 2.77 4.08 1.01 94.858 8 -1 Enorm 51.04 2.17 19.75 7.70 0.16 4.20 6.71 2.92 4.30 1.06 10 0 .0 0

c p x 0.04 48.75 2.14 7.81 8.44 0 .1 0 13.00 21.73 0 .8 6 0 .0 0 0 .0 0 102.83

fId sp 0.03 52.51 0 .0 0 28.78 0.36 0 .0 0 0 .0 0 11.36 4.52 0 .6 8 0 .0 0 98.21

am p h 0.06 39.51 5.05 13.30 11.30 0.61 13.22 12.58 2.08 1.63 0 .0 0 99.28

Ti-m ag 0 .0 2 0 .1 0 15.42 1.39 78.43 0.33 2.29 0.06 0 .0 0 0 .0 0 0 .0 0 98.02

A p a tite 0.01 0 .0 0 0 .0 0 0 .0 0 0 .21 0.52 0.54 52.14 0 .0 0 0 .0 0 40.98 94.39

F = m o d e l r e s u l t s

0 .8 5

53.29 1.81 20.96 6.39 0.13 3.36 5.13 3.10 4.94 0.89 1 0 0 .0 0

8 7 - 1 D 50.39 1 .8 6 20.60 6.27 0.15 3.24 5.23 5.42 4.79 0.85 98.80

(/)(/)

Table 10. Summary of major element modeling for segment 2.

tn

Page 131: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 1 6

§

10

8

6

4

2

00 51 2 3 4 6

MgO

O

8 W

6

4

2

00 1 2 53 4 6

MgOFigure 25. Composition of modeled magma evolution for segment 2.

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

Page 132: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 1 7

Oi2î

EO1 0 .

0 1 3 4 52 6

MgO

Oâ!

1.4

0.8 .

0.6 .

0.4 .

0.2 .QQ

0.00 1 2 3 4 5 6

MgOFigure 25 (cont). Composition of modeled magma evolution for segment 2.

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

Page 133: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 1 8

samples could have been chosen. Model results for the elements S1O2 , Ti02, AI2O3, FeO,

MgO, CaO, K2O, and P2O5 compare well, within 5%, with the actual fractionated rock

composition. Na20 and MnO modeled results are greater than 5% off of rock

compositions, however, the MnO modeled fractionation path compares well with the

second segment MnO variation diagram. Samples at the differentiated end of this segment

on the MnO variation diagram do not cluster as well as most elements; the four samples that

define this end of the segment have MnO concentrations of 0.12,0.13,0.14,0.15 so that

the modeled concentration of MnO = 0.13 falls within this spread of actual MnO data. AU

of the elements, except Na2 0 , have model fractionation paths very similar to the actual data

variation. Again, Na20 will be discussed at the end of this section.

SEGMENT 3: WDB-87-1D - (v) clinopyroxene -(x) amphibole - (w) feldspar -

(y) Ti-magnetite - (z) apatite = theoretical residual magma

The third segment is defined by the sample 87-lD at the primitive end and by sample 88-

1G at the differentiated end. Model results for the oxides FeO, Ti02, and K2O compare

very weU, within 5%, with the fractionated sample composition (Table 11 and Figure 26);

the oxides S1O2 , MgO and CaO model to within 10% of the composition of the

differentiated rock. The oxides CaO, MnO, and P2O5 have very low concentrations at the

most fractionated end of the third segment. The relatively high percentage errors between

model and actual concentrations of these elements is a function of their low abundances.

For example, model P2O5 = 0.19 whereas actual 88-IG P2O5 = 0.16; this smaU absolute

concentration difference yields a 19% error but is actuaUy a very good fit Examination of

the variation diagrams of Figure 26 shows that model results compare very weU with

sample trends for the third segment The oxides AI2O3 and perhaps Si02 indicate the

model is not completely adequate. Na2 0 , which has proved to be a problem element, wiU

be examined later.

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

Page 134: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

CD"OOQ .C

gQ .

■DCD

C/)Wo"3O

8■DCQ-3"

i3CD

3.3"CD

CD■DOQ .C

aO3■DO

CDQ .

■DCD

(100) 87-1D = (65) 88-1G + (3) cpx + (14) am ph + (15) fidsp + (1.8) ap t + (1.5) Ti-mag

% xtal S I 0 2 T I0 2 A I2 0 3 F e O M nO M gO C aO N a 2 0 K 2 0 P 2 0 5 to ta l

8 7 1 D 50.39 1.86 20.60 6.27 0.15 3.24 5.23 5.42 4.79 0.85 98.808 7 1 Dnorm 51.00 1.88 20.85 6.35 0.15 3.28 5.29 5.49 4.85 0.86 100.00

c p x 0.030 47.76 3.02 7.38 7.62 0.24 12.14 21.93 0.91 0.02 0.00 101.02

A pt 0.018 0.00 0.00 0.00 0.21 0.52 0.54 52.40 0.00 0.00 40.98 94.65

f id s p 0.150 60.01 0.00 22.81 0.31 0.00 0.00 3.58 8.19 2.32 0.00 97.22

Tl m ag 0.015 0.10 19.42 1.39 74.43 0.33 2.29 0.06 0.00 0.00 0.00 98.02

am p h 0.140 39.51 5.05 13.50 15.00 0.42 13.09 12.58 2.08 1.63 0.00 102.94

F = m o d e l

r e s u lt s

0 .6 5

54.07 1.22 23.61 4.40 0.11 1.60 2.15 6.08 6.59 0.19 100.00

88 - 1 G 58.54 1.24 20.54 4.15 0.14 1.57 2.69 4.20 6.32 0.16 99.55

3C/)wo"

Table 11. Summary of major element modeling for segment 3.

CD

Page 135: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

120

§

10

8

6

4

2

00 1 4 52 3 6

MgO

O

da10 -

0 1 2 53 4 6

MgOFigure 26. Composition of modeled magma evolution for segment 3.

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

Page 136: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

121

i

8

a Qtf6

4

2

00 1 2 4 53 6

oÛ.

MgO

0.8 -

0.6 .0.4 -

0.2 .

0.00 1 2 3 4 5 6

MgOFigure 26 (cont). Composition of modeled magma evolution for segment 3.

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

Page 137: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 2 2

ii) Trace Element Models

In order to test the results of major element fractionation models, trace element

models based on the Rayleigh fractionation law were constructed. The importance of

certain trace elements, such as Ti, Zr, Nb, and Y, in petrogenetic modeling has been

emphasized by several authors (Pearce and Norry, 1979; Hildreth, 1980; Hooper, 1988,

Womer and Schmicke, 1984(b)). Elements with a high field strength (charge/radius ratio)

are not easily transported in aqueous fluids and therefore are often unaffected during

metasomatic alteration (Pearce and Noiry, 1979). These elements also show systematic

variation during magmatic differentiation; in fact, trace elements may exhibit strong

concentration gradients reflecting petrogenetic process when the major elements remain

constant (Hildreth, 1980). For these reasons, Ti, Zr, Nb, and Y can be used to study

weathered volcanic rocks whose major element chemistry has been too severely altered for

petrogenetic interpretation. Pétrographie work and whole rock geochemical analyses

indicate that the Woodbine volcanic rocks are not seriously altered. However, chemical

weathering and alteration must always be seriously considered in a study involving

epiclastic volcanic rocks. In this study, trace elements were used in conjunction with major

element models. A firactionation model that is supported by both major and trace element

calculations, yields strong support for that hypothesized differentiation process.

Five trace elements were examined and modeled. Petrogenetic modeling of trace

elements quantifies the extent to which an element is incorporated into the minerals which

are crystallizing from a magma. This concept is expressed by a partition coefficient, Km .

which is the ratio of the concentration (by weight)of an element, M, in a mineral to the

concentration in the melL

The value of Km is a function of magmatic composition, temperature and pressure.

In order to assess the behavior of a trace element when considering all the minerals

fractionating from a magmatic system, a bulk distribution coefficient must be calculated.

Since each segment on the Woodbine variation diagrams represents a slightly different

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

Page 138: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 2 3

magmatic system, in that different proportions of different minerals are precipitating along

each segment, bulk distribution coefficients for each segment had to be determined

separately. Bulk distribution coefficients for each of the five trace elements, for each

segment, were calculated as

Kbulk = a Km Q + P Km R + Y Km S

where Q, R, S represent three minerals crystallizing together in the modal normalized

proportions cc, P, Y (a + p + Y = 1). These bulk partition coefficients are summarized in

Table 12.

The partition coefficients of Sr and Ba into feldspar were determined from

compositional data collected as part of this study; other partition coefficients were collected

from the literature (Pearce and Norry, 1979; Henderson, 1982). Since the partition

coefficients firom the literature were determined for basalts and andésites, they most likely

do not fully describe the behavior of these elements in the alkaline Woodbine system. They

do, however, represent a good estimation of the theoretical behavior of these elements in

this system.

Using the solution of the major element models for F and to determine Kbulk,

model trace element enrichment factors were calculated by the Rayleigh Fractionation Law:

Cl / C o = F (Kbulk - 1)

where Cl = concentration of trace element in residual liquid,

Cq = concentration of trace element in original, primative liquid,

F = fraction of liquid remaining after mineral crystallization,

Kbulk = bulk distribution coefficient.

Model Nb, Sr, & , and Ba enrichment factors (concentration of element in

differentiated sample / concentration of element in the primitive sample) were determined

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

Page 139: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

CD" O

OQ .C

gQ .

■DCD

C/)C/) 1. Minerai Partition C oefficients3o cpx hb ld m ag p lag3 K Nb (mafic rocks) 0.1 0.8 0.4 0.01

8■D K Nb (intermediate rocks) 0.3 1.3 1 0.03

( O '3" K Zr (intermediate rocks) 0.25 1.4 0.2 0.03

i3CD K Sr (intermediate rocks) 0 0 0 5

"nc K V (intermediate rocks) 1.5 2.5 0.5 0.063.3"CD

CD■DOQ .CaO3"OO

CDQ .

■DCD

C/)C/)

NOTE: ail Nb, Zr, and Y partition coefficients are from Pearce and Norry (1979)

K Sr in cpx, hbld, mag are from Henderson (1982)

K Sr in plag is from this study

2. Bulk D istributuion C oefficientsD Nb (segment 1) = (.8 M ) + (.03 * .8) + (.16 * .4) = .20 D Zr (segment 1) = (.74 * .25) + (.05 * 1.4) + (.21 * .2) = .30 D Sr (segment 1 ) = (.74 * 0) + (.05 * 0) + (.21 * 0) = 0 D Ba (segment 1) = (.74 * .07) + (.05 * .31) + (.21 * .07) = .08

D Nb (segment 2 ) = ( . 2 6 M ) ( . 2 * . 0 1 ) + ( . 3 9 * . 8 ) + ( .1 * . 4 ) + ( .0 5 * 0 ) = . 3 8

D Zr (segment 2 ) = ( . 2 6 * . 2 5 ) + ( . 2 * . 0 3 ) + ( . 3 9 * 1 . 4 ) + ( .1 * . 2 ) + ( . 0 5 * 0 ) = . 6 4

D Sr (segment 2 ) = ( . 2 6 * 0 ) + ( . 2 * 3 . 5 ) + ( . 3 9 * 0 ) + (.1 * 0 ) + ( . 0 5 * 0 ) = . 7

D Ba (segment 2 ) = ( . 2 6 * . 0 7 ) + ( . 2 * . 7 8 ) + ( . 3 9 * . 3 1 ) + ( .1 * . 0 7 ) + ( . 0 5 * 0 ) = . 3

D Nb (segment 3 ) = ( .1 1 * . 3 ) + ( .4 1 * . 0 3 ) + ( . 0 5 * 0 ) + ( . 3 9 * 1 . 3 ) + ( . 0 4 * 1 . 0 ) = . 5 9

D Zr (segment 3 ) = ( .1 1 * . 2 5 ) + ( .4 1 * . 3 ) + ( . 0 5 * 0 ) + ( . 3 9 * 1 . 3 ) + ( . 0 4 * . 2 ) = . 5 9

D Sr (segment 3 ) = ( .1 1 * . 3 ) + ( .4 1 * 6 . 0 ) + ( . 0 5 * 0 ) + ( . 3 9 * 0 ) + ( . 0 4 * 0 ) = 2 . 5

D Ba (segment 3 ) = ( .1 1 * . 0 7 ) + ( .4 1 * 2 . 0 ) + ( . 0 5 * 0 ) + ( . 3 9 * . 3 1 ) + ( . 0 4 * . 0 7 ) = . 9 5

Table 12. Calculation of Bulk Distribution Coefficientsro4

Page 140: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 2 5

for each fractionation segment. Enrichment factors indicate the extent to which an element

is becoming enriched, or depleted, in the more differentiated magmas. Enrichment factors.

C l / Co. that are greater than 1 indicate that the element is becoming enriched in the

residual liquid; an enrichment factor of less than 1 indicates that the element is preferentially

incoiporated into the solid phase during crystal precipitation and is therefore depleted in the

more differentiated magma. Solutions from the major element models for each segment

were used in the trace element models; F, the fraction of liquid remaining (F = 1 when the

system is all melt, and F = 0 when the system is all solid) was determined for each segment

from major element models. Bulk distribution coefficients were calculated using the

mineral assemblages of major element models.

These model and actual enrichment factors are summarized in Table 13. The

"actual" enrichment factors were compared to "model" ratios. Overall, the model results

for Nb, 2Ï, Sr, and Ba compare very well with enrichment factors present within the actual

Woodbine variation segments. When considering all of these four trace elements and all

three modeled segments, the models account for 86% of the variation observed in the

Woodbine volcanics. It is, however, more useful to examine each of the segments

separately.

Modeled and actual enrichment factors for segment 1 compare the poorest of any of

the segments. Along segment 1, the elements Nb, Sr, and Ba are strongly enriched in the

more differentiated liquids (Table 13); the model enrichment factors indicate that these

elements do become enriched, but not to as great an extent as the actual compositions

indicate (Table 13). The hypothesis of fractional crystallization is supported by the fact that

model and actual enrichment factors are greater than 1; errors in magnitude of the

enrichment factors may indicate that some process other than fractional crystallization may

be responsible for some of the chemical variation observed in these rocks.

Trace element models for segments 2 and 3 compare very well with actual rock

compositions (Table 13). Along the second segment, 2 elements, Nb and Zr are slightly

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

Page 141: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 2 6

SEGM ENT 1 862A-881E

SEGMENT 2 881E-871D

SEGM ENT 3 871D-881G

N bactual Cl/Co

89 - 134 ppm 1.5

134-176 ppm 1.3

176 - 207 ppm 1.2

model Q/Co 1.2 1.1 1.2

Z ractual C]/Co

429 - 402 ppm 0.9

402 - 448 ppm 1.1

448 - 548 ppm 1.2

model Q/Cb 1.1 1.1 1.2

S r

actual Cl/Co

775 - 1683 ppm 2.2

1683 -1645 ppm 1.0

1645 - 991 ppm 0.6

model Q/Cb 1.3 1.0 0.5

B aactual Q /C o

1441-23681.6

2368 - 2155 0.9

2155- 1864 0.9

model Cl/Co 1.2 1.1 1.0

Table 13. Summary of Trace Element Models.

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

Page 142: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 2 7

enriched in the more differentiated rocks. Model enrichment factors for Nb and Sr are in

agreement with the direction and with this degree of enrichment. Sr and Ba have flat

variation trends along this segment, indicating that these elements remained at a constant

concentration during this part of the magmatic differentiation. Model results also indicate

little or no enrichment or depletion in these samples (Table 13).

Rocks that compose the third segment exhibit significant enrichment in the elements

Nb and Zr with increasing differentiation. Model results for these elements agree extremely

well with actual enrichment factors. Sr exhibits significant depletion along along this

segment; model results for the element Sr are in excellent agreement with the actual degree

of depletion seen in these rocks. Ba exhibits a relatively flat enrichment trend along the

third segment and again, model results compare very well with this trend of little or nor

change in Ba concentration during this stage of fractionation fractionation.

D. Assimilation and Fractional Crystallization

Assimilation and fractional crystallization (AFC) combines the effect of fiactional

crystallization and assimilation of wall rock. Several geochemical parameters can be used

to identify assimilation. Assimilation results in 1) extreme enrichment of incompatible

elements, and 2) compatible elements not necessarily declining to 0 , but instead declining to

a value controlled by wall rock concentrations.

The role of assimilation in the fractionation of the Woodbine suite of rocks is not an

obvious one. Compatible trace elements, such as Ni, Cr and Cu are present in very low

concentrations (< 10 ppm) in the most differentiated samples; this may indicate that the

rocks were not affected by assimilation or that the assimilating material had very low

concentrations of these compatible elements. At the most fiactionated end of the Woodbine

volcanic geochemical trends, an increase in the scatter of certain incompatible elements such

as La, and Ba, is observed; part of this scatter is a reflection of very high Ba and La

concentrations of a few samples. Melt inclusion compositional data from sanidines from

highly differentiated rocks indicates that the high and low Ba concentrations are magmatic

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

Page 143: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 2 8

(Figure 17); the extreme Ba concentrations are not the result of crystal settling or post-

crystallization alteration. Furthermore, the lack of systematic geochemical patterns in

relation to the volcanic samples implies that a process other than fractional crystallization

affected the differentiation of these samples. Assimilation of crustal material could be that

process. Local incorporation of varying amounts of crustal material could explain the

variable Ba and La concentrations in the most differentiated Woodbine rocks. Crustal

rocks with high concentrations of Ba and La, and low concentrations of Ni, Cr, and Cu,

may have become assimilated, in small amounts and on a local scale, into the fractionating

Woodbine magma.

This interpretation is reinforced by a similar conclusion based on recent, Si-

undersatured rocks in Central Italy. A study of the phonolites and trachytes from Vico

volcano revealed extreme variation and general enrichment of trace element compositions in

rocks with a similar degree of evolution (Barbieri et al., 1988). Rocks with similar MgO,

Si02, and CaO contents exhibit variable and, in general, enriched concentrations of

incompatible trace elements (Barbieri et al., 1988). Major and trace element calculations

indicate that most of the differentiated rocks could be derived from the more primitive by

fractional crystallization of clinopyroxene, leucite, plagioclase and/or sanidine, biotite,

oxides and apatite; this fractionation is supported by pétrographie work (Barbieri et al.,

1988). However, the large variation of trace element abundances within rocks with a

similar major element composition argues against the derivation of all the Vico suite from a

single parent magma through simple crystal fractionation (Barbieri et al., 1988). Instead,

extreme fractional crystallization combined with small amounts of magma mixing with a

primitive melt resulted in a hybrid magma with enriched incompatible element

concentrations and whole rock compositions similar to less fractionated rocks (Barbieri et

al., 1988).

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

Page 144: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 2 9

3. Sources of Error

Any qualitative or quantitative analysis of volcanic material must evaluate the

potential sources of error inherent in these models. The three main sources of error that can

affect model results are: 1) errors associated with the whole rock geochemical data, 2)

errors associated with mineral compositional data, and 3) errors associated with partition

coefficients.

Whole rock geochemical errors can be divided into analytical errors, alteration of

volcanic rocks, and non-representative samples prepared for analysis. Errors associated

with XRF analytical work are believed to be negligible; in order to test the accuracy of the

analysis, a USGS rock standard, STM-1, a nepheline syenite firom Table Mountain,

Oregon, was prepared and analyzed as an unknown with the Woodbine volcanics. No

major discrepancies between the XRF analysis and the reported composition for the

standard were found;

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

Page 145: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 3 0

STM -1 USGS R eported Analysis

STM -1 XRF A nalysis (this

s tu d y )

Si02 59.54 59.60

AI2O3 18.61 18.50FeO 4.95 4.65MgO 0.10 0.09CaO 1.16 1.10

NaiO 8.97 8.90K2O 4.24 4.30Ti02 0.13 0.15P2O5 0.16 0.16

Ba 484 590Cr 0 <10La 188 170Nb 298 270Ni 0 <5Sr 767 669Y 57 51Zr 1230 1234

Errors in precision for whole rock, XRF, major element compositional data are

estimated at 1-3% and XRF trace element precision errors are estimated at 10-20% by Steve

Nelson at Tulane University.

Errors associated with alteration of volcanic rocks must be seriously considered

when working with epiclastic volcanic material. Whole rock geochemical samples were

carefully chosen as the freshest portions of the freshest cobbles. The linear trends on

variation diagrams of the whole rock data indicate that alteration has not significantly

affected these samples; weathering and alteration would not affect the cobbles in a

systematic way, so significant alteration of the cobbles would result in a high degree of

scattering on the variation diagrams. However, some variation diagrams exhibit a

scattering of data points. The entire Na20 variation diagram is composed of scattered data

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

Page 146: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

131

points, and certain segments of the Si02, MnO, Zr, Zn, La diagrams exhibit scattering.

Chemical alteration could be the cause of some of this scatter. The third error associated

with whole rock geochemistry, non-representative samples, is a potentially serious one.

Normal whole rock sample preparation involves crushing and powdering a volume of rock

large enough to insure that the final powder represents the whole rock. The Woodbine

volcanic cobbles prepared for whole rock analysis range firom 3-8" in diameter. It is

possible that a single, large phenocryst could have been crushed during sample preparation

so that the resulting powder would have a higher proportion of that mineral than the

volcanic rock as a whole. Using small sample sizes allows the possibility of analyzing

non-representative samples. Care was taken to avoid inclusion of single large phenocrysts

in the XRF samples, but when the amount of available rock is small, this problem is a

difficult one to prevent completely.

Errors associated with mineral compositional data are two-fold; these are analytical

errors and the problem of estimating an average crystal composition for strongly zoned and

variable minerals. Analytical errors of the electron microprobe are easily identified; mineral

standards are run as unknowns every 10-20 analysis to determine the accuracy and

precision of the analysis. Stoichiometric consideration of the data evaluate the reliability of

the analysis. Any imreliable analysis were not included as part of this study. Based on

replicate analyses, precision errors associated with electron microprobe data are estimated at

1-5% for major elements and 10-20% for trace elemtents. Once the microprobe data is

collected, the average composition of a mineral phase for an entire segment must be

estimated to complete major element models. For example, clinopyroxene compositions

along the first segment are variable; individual crystals are strongly zoned and mineral

compositions vary throughout the segment. An ideal clinopyroxene composition for the

entire segment was determined. These ideal compositions for each phase, for each

segment, are a good estimation of the average phase composition, but some errors must be

associated with these compositions.

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

Page 147: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 3 2

Errors associated with partition coefficients reflect the fact that the extent to which a

trace element is incorporated into a precipitating mineral is a function of temperature,

pressure, and magmatic composition. Partition coefficients from the literature were derived

for specific magmatic systems (Pearce and Norry, 1979; Henderson, 1982). In order to

minimize these errors, care was taken to choose values from the literature that were derived

for magmatic systems similar to the Woodbine system; however, good partition coefficients

for alkaline rocks are not available. Mineral / melt partition coefficients determined for

basalts, andésites, and rhyolites were used in this study. The extent to which the use of

these coefficients introduces error into this study is not known. The use of different, still

valid, partition coefficients would produce slightly different results for the trace element

models, however, it is believed that these differences would not significantly change any

geologic interpretations.

4. Petrogenetic Dififerentiation Summary

Magmatic systems are extremely complicated. As magmas rise, cool and solidify, a

wide variety of processes can simultaneously act to influence the chemical composition of

the system. As these hot liquids rise, they may incorporate a wide variety of wall rocks by

assimilation. As the magma cools, precipitating crystals may sink or rise, and effectively

be removed from the system, so that the total composition of the system is changed.

Minerals which precipitate at high pressures may be unstable and be resorbed into the

magma at low pressures and thereby complicate the chemical evolution of the system.

Overall chemical compostion of the magma, particularily volatile composition, strongly

influences trace element behavior during crystal precipitation. It is therefore extremely

difficult to model these systems precisely; determining the compositions and proportions of

the variety of assimilated wall rocks, determining the types and amounts of crystal settling,

and estimating trace element behavior with accurate partition coefficients are all difficult to

do well. Examination of ancient, reworked volcanic debris introduces further variables,

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

Page 148: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 3 3

including alteration and sampling volume problems, that increase the difficulty in modeling

the data. Thus, a theoretical model that can describe 85 - 95% of the observed variation in

a suite of volcanic rocks can be considered a good model.

The major and trace element crystal fractionation models presented in this study

explain approximately 85 - 95% of the chemical variation observed in the Woodbine suite

o f volcanic rocks. The major element models produce results that are consistent with the

observed phenocryst assemblages for each of the modeled segments; only minerals present

in these rocks were used as part of the modeled fractionating assemblage, and the model

mineral proportions and overall mineral to melt ratios are consistant with those observed in

the rocks. Trace element models are in agreement with the results of major element

models. These factors combine to support the hypothesis that fractional crystallization of a

primitive mafic magma was the dominant petrogenetic process that resulted in the chemical

and mineralogical variety of the Woodbine volcanics.

The pyroclastic rocks preserved as cobbles within the Woodbine Formation may

represent a late stage, highly fractionated magma that was explosively erupted from the

chamber that produced the effusive lavas described above. The textures of these rocks,

including broken, angular phenocrysts, abundant lithic fragments, and squashed pumice

fragments in a chaotic and highly altered matrix, suggest that the eruption of these rocks

was explosive in nature. It was not possible, because of their highly altered nature, to

determine whole rock compositions of these rocks, and it was therefore not possible to

quantitatively assess whether the pyroclastic rocks are related to the effusive volcanic rocks

by the process of fractional crystallization. However, since the pyroclastic rocks are found

in the same sedimentary unit as volcanic rocks that are interpreted as cogenetic, they were

likely derived from the same volcano, and are interpreted as the highly fractionated rocks of

the fractionating magma chamber.

Fractionation of the Woodbine magmas was dominated by pyroxene, magnetite,

amphibole, feldspar, apatite, and sphene. Cumulate rocks of this fractionating sequence

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

Page 149: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 3 4

would be rich in the denser minerals of this precipitating assemblage. Rocks that are rich in

pyroxene and magnetite are present within the Magnet Cove Complex in central Arkansas.

Jacupirangite, a magnetite pyroxeneite, constitutes about 10% of the exposed igneous rocks

at Magnet Cove (Erickson and Blade, 1963). Pyroxene is the most abundant phase and

constitues at least 50% of the rock; magnetite is always abundant Apatite, biotite, and

sphene are always present (Erickson and Blade, 1963). The pyroxene crystals, salite, are

up to 10mm long and appear to have formed from an early crystal mush (Erickson and

Blade, 1963). Sphene pyroxenites are found in association with the jacupirangites; they are

composed of approximately 67% pyroxene, 8% sphene, and 4% apatite. Although these

cumulate rocks may not be the direct result of crystal settling during fractionation of the

Woodbine magmas, they must be very similar to Woodbine cumulate rocks that, if they

exist, have not been sampled. Erickson and Blade (1963) interpreted the jacupirangites to

be the result of a marie residue and volatile constituents that accumulate at the bottom of the

chamber after precipitating feldspar and nepheline float to the top.

It must be noted that the Woodbine model residuals are larger than model residuals

fijom some similar studies involving recent, stratified volcanic deposits (Womer and

Schmincke, 1984 b; Barberi et al.,1981; Barbieri et al., 1988). Elemental residuals of

riractionation models and comparison of actual rock compositions fiom these recent rocks

are typically less than 0.5 w l % . All of these studies have examined volcanic, stratified

rocks which are less than 1 Ma. The larger Woodbine residuals (Tables 11,12,13) are

believed to be a consequence of the age (Cretaceous) and nature (epiclastic) of the rocks of

this study. Improper (too small) sample size and alteration result in errors associated with

the whole rock chemical analyses and therefore a larger discrepancy between model and

whole rock results. There is no satisfactory method of deciding whether a model fit is

good or bad; intuition is the standard method for evaluating model fit (Le Maitre, 1981).

Since these major and trace element models explain 85 - 95% of the observed chemical

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

Page 150: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 3 5

variation, fractional crystallization is interpreted as the dominant process responsible for the

mineralogical and chemical variation of the suite of Woodbine volcanics.

Examination of Miocene, minette dikes in NW Colorado by Leat et al. (1988)

yielded fractional crystallization whole rock models with residuals similar to the Woodbine

models; some elemental residuals from the Leat study exceed 1 wt.%. Study of young (< 1

Ma) volcanic deposits results in several advantages over the analysis of ancient and

reworked volcanic rocks; recent deposits are less likely to be affected by weathering and

chemical alteration, and layered volcanic deposits allow large, unbiased sample volumes.

The limitations of the study of epiclastic, volcanic material may result in larger residuals in

major and trace element models of the data. Examination of this material may also result,

however, in the understanding of older, previously inaccessible igneous terranes.

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

Page 151: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Chapter 6. Provenance o f Epiclastic Volcanic Cobbles

Ross et al. (1928) concluded the source of the Woodbine volcaniclastic material to

be buried igneous centers or pipes in Howard County Arkansas. However, Belk (1986)

noted that geophysical data has failed to reveal any such buried centers. On the basis of

pétrographie data, Belk (1986) speculated Magnet Cove as a likely source for this volcanic

material. Examination of variation diagrams composed of whole rock data from several

Cretaceous Gulf Basin igneous centers suggests that a relationship exists between the

Woodbine volcanic rocks and the lamprophyric dikes of the Benton Dike Swarm (Figure

20). For example, the FeO vs. MgO variation diagram (Figure 20) suggests that rocks of

the Magnet Cove and Balcones centers have significantly higher FeO concentrations and

were evolving along distinctly different chemical trends when compared to the Woodbine

suite. The MnO vs. MgO variation diagram reveals that the most differentiated rocks from

the Granite Mountain, Balcones and Magnet Cove complexes are enriched in MnO relative

to the Woodbine volcanics. The K%0 variation diagram suggests that Balcones, Magnet

Cove, and Granite Mountain rocks are depleted in K2O relative to the Woodbine volcanics.

Without exception, rocks o f the Benton Dikes Swarm seem to be part of the Woodbine

chemical evolutionary trend, with the lamprophyric dike chemistry very similar to whole

rock chemistry of the most mafic Woodbine volcanic rocks.

1. Similarities between Woodbine volcanic rocks and the Benton Dikes

The minéralogie and geochemical similarities between mafic Woodbine volcanic

rocks and the differentiated Benton lamprophyres are impressive. The mineral assemblage

composed of dominant clinopyroxene with lesser amounts of hornblende, biotite, and late

felsic minerals in the groundmass is almost identical in the two sets of rocks.

Clinopyroxenes from the dikes are described by Morris (1987) as zoned fiom salitic (Ca-

rich augite) cores to titanaugite rims. The clinopyroxenes from the mafic Woodbine rocks

1 3 6

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

Page 152: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

137

are strongly zoned Ca-rich augite and titanaugites. Amphiboles from both suites are

kaersutite (Ti-rich hornblendes), with approximately 5 wL % Ti02. Amphiboles in the

Woodbine volcanic rocks commonly show signs of resorption; the amphibole phenocrysts

are corroded, ambayed and are altered to a mixture of Fe-oxides and clinopyroxenes.

Amphibole phenocysts in the lamprophyric dikes sometimes show the effects of resorption

(Valdovinos, 1968).

Whole rock chemistry from the two suites compares favorably (Figure 20). For

every element examined, continuity exists between the chemical trends defined by the

Benton lamprophyres and the trends of the Woodbine volcanic rocks. For example, the

chemical evolutionary trends exhibited on the K2O vs. MgO, the MnO vs. MgO, the Ti02

vs. MgO, and the FeO vs. MgO for the Woodbine volcanic rocks and the lamprophyric

dikes are completely congruous.

Radiometric ages of the Woodbine volcanics and the Benton Dikes also compare

favorably (Figure 27). A whole rock K-Ar age, determined as part of this study, of a

trachyte preserved as a cobble in the Woodbine conglomerate yielded an age of 98 ± 2 Ma.

Fission track ages for the Benton Dike Swarm have been determined for 2 samples; sphene

from a sannaite yielded a date o f 100 Ma, and apatite from a camptonite yielded a date of 99

Ma (Eby, written communication to Morris, 1987).

Whole rock geochemical data indicates that Granite Mountain syenites are a possible

intrusive equivalent for the Woodbine volcanics. Chemical trends on many of the whole

rock variation diagrams indicate that the Granite Mountain and Woodbine suites could be

genetically related; the Si02, Ti02, and FeO variation diagrams (Figure 20a,b,c) indicate

that the most felsic Woodbine volcanic rocks are chemically very similar to the Granite

Mountain syenites. However, other elements, such as MnO, K20, and Ba (Figure

20d,e,f) indicate that the Granite Mountain syenites and the Woodbine volcanics are

chemically distinct and are not genetically related to each other. Furthermore,

geochronologic evidence does not support the hypothesis that the syenites are genetically

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

Page 153: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 3 8

related to the Woodbine volcanics. A total fusion ^ A r - 39Ar date determined on a biotite

from a Granite Mountain syenite as part of this study yielded and age of 89.5 ± 7 Ma; this

date corresponds well with geochronologic data on the same rocks from other studies

(Figure 27) (Morris, 1987). The age of the syenites based on K - Ar on biotite, and fission

track ages of both sphene and apatite is 87 - 89 Ma; however, Morris (1987) calculated a

fission track age of 95 Ma on apatites fr o m the most primitive rock at Granite Mountain, an

olivine syenite. A whole rock K -Ar age determined an a Woodbine trachyte yielded an age

of 98 ± 2 Ma. Although there is some overlap between these two age spectra, the bulk of

the igneous activity at Granite Mountain predated the volcanism that resulted in the

Woodbine volcanic rocks. Based on certain chemical parameters and on geochronologic

data, it seems unlikely that the syenites of Granite Mountain are genetically related to the

volcanic rocks preserved in the Woodbine Formation. Lithologie, minéralogie,

geochemical, and geochronologic data support the lamprophyres of the Benton Dike

Swarm as the provenance of the epiclastic volcanic cobbles of the Woodbine Formation.

2 . Characteristics of lamprophyres

Lamprophyric dikes occur throughout much of the central and eastern Ouachitas.

They are particularly abundant within a 350 km^ area north of the town of Benton,

Arkansas. The dikes are generally l-2m wide and vary from 100m to 2km in mapped

length (Montis, 1987).

Lamprophyres are dark colored, fine-grained, poiphyritic rocks, dominated by

euhedral mafic phenocrysts. Phenocrysts commonly include biotite, hornblende, augite or

titanaugite, and olivine.

In many instances, lamprophyres contain two generations of mafic phenocrysts,

one of which is an earlier, altered phase and another of later, fiesh crystals of the same

mineral. Although the groundmass of a typical lamprophyre is dark in hand sample, it

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

Page 154: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 3 9

Balcones

PrairieCreek

MagnetCove

VIntrusive

BentonDike

Swarm

PotashSulfur

Springs

GraniteMountain

MonroeUplift

JacksonDome

Ma 60 70 80 90 100 110

Ages from the literature (see text)

Ages determined as part of this study

Age of Woodbine trachytic cobble

Figure 27. Comparison of radiometric ages of a trachyte preserved as a cobble in the

Woodbine Formation and other northern Gulf of Mexico Basin igneous rocks.

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

Page 155: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

140

usually consists of felsic minerals including plagioclase, sanidine, and nepheline and

secondary minerals such as zeolites, calcite, and talc (Hyndman, 1972).

Chemically, lamprophyres are similar to alkaline olivine basalts and related rocks;

lamprophyres are low in Si02, and high in Na2 0 , K2O, Ti02, SrO, BaO, and P2O5 ,

Exposure of lamprophyric dikes in central Arkansas is poor, felsic varieties weather

especially quickly (Robison, 1977; Morris, 1987). The dikes vary in lithology and have

been given a variety of confusing names including ouachitatites, minettes (biotite,

orthoclase lamprophyres), monchiquites (analcite,homblende, augite lamprophyres),

camptonites (hornblende, augite, plagioclase lamprophyres), and sannaites. A simpler and

more descriptive classification scheme identifies these rocks as olivine lamprophyres,

pyroxene lamprophyres, pyroxene, amphibole lamprophyres, olivine lamprophyres, and

feldspathoidal gabbros. Olivine is present in some samples of the Benton Dike Swarm, but

is always less abundant than clinopyroxene (Morris, 1987). Amphibole is abundant in

some lamprophyres, and biotite occurs in some samples (Robison, 1977).

3. Origin of lamprophyric magmas

Hypotheses concerning the origin of lamprophyres usually begin with an alkaline,

basaltic magma that is modified in some way. The modification may come in the form of

crustal contamination, fi:actional crystallization, assimilation, or resorption of biotite or

hornblende.

Berger (1923) and Bowen (1928) suggested that resorption of ferromagnesian

minerals such as biotite and hornblende into a mafic magma would result in the liquid

becoming enriched in alkalis and volatiles with simultaneous precipitation of augite and

olivine.

Recent ideas concerning the origin of alkali basalts and related rocks, including

lamprophyres, center on the concept of mantle metasomatism. Depleted mantle peridotite is

not a viable source of alkali basalts (Wilshire, 1987; Barreiro and Cooper, 1987;

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

Page 156: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

141

Hawkesworth, et al., 1984; Wass and Rogers, 1980; Menzies and Murthy, 1980). Mantle

metasomatism is a complex process, with many unresolved details; the source of the

metasomatizing agents, the relationship between metasomatism and magmatism, and the

role of fractionation remain unanswered questions. Mantle metasomatism may involve

several processes, but the most common explanation for metasomatism is the infiltration of

fluids into mantle peridotites. These fluids enrich the peridotite in incompatible trace

elements (LREE, Ba, Sr) (Menzies et al., 1985; Wilshire, 1987; Wass and Rogers, 1980;

Hawkesworth, et al., 1979) or in Fe, Ti, and A1 (Wilshire, 1987; Wilshire and Trask,

1971; Menzies et al., 1985; Wilshire et al., 1980; Irving, 1980). Many sources for these

fluids have been proposed including mantle partial melts (Barreiro and Cooper, 1987;

Wilshire, 1987), continental crust through subduction (Hoffman and White, 1982;

Hawksworth et al., 1984), and migration of trace elements in hydrous or carbonic fluids

from deep in the mantle (Wass and Rogers, 1980; Wilshire et al., 1980).

Examination of peridotite xenoliths in alkaline basalts has revealed the existence of

amphibole segregations or veins within the peridotite (Nielson and NoUer, 1987; Wilshire,

1987; Wass and Rogers, 1980; Menzie and Murthy, 1980; Best, 1974; Wilshire and Trask,

1971). These secondary minerals in the mantle peridotite are generally thought to be

caused by infiltration of deep mantle fluids (Wass and Rogers, 1980; Best, 1974; Wilshire

and Trask, 1971) or by interaction of the peridotite with mafic melts (Nielson and Noller,

1987; Barreiro and Cooper, 1987; Wilshire, 1987). The role of hydrous silicate phases,

most notably amphibole, phlogopite, and apatite, is apparently very important in the

process of mantle metasomatic genesis of alkaline magmas. These mantle fluids or

magmas that act as the metasomatizing agents are thought to have a wide range in

compositions (Best, 1974) but are thought to precipitate amphibole once the fluids rise

above the lower limit of amphibole stability (60 -100 km in the uppermost mantle) (Best,

1974). The relationship between the metasomatizing event and partial melting resulting in

alkaline magmatism is not known. Some authors conclude that the metasomatizing event

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

Page 157: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 4 2

contributes to and triggers the formation of the mafic alkaline magmas (Hawksworth et al.,

1979), whereas others conclude that the two events are unrelated (Barreiro and Cooper,

1987; Wilshire and Trask, 1971).

Recent ideas concerning mantle metasomatism and alkaline magmas are not

dissimilar to the ideas proposed by Berger (1923) and by Bowen (1928). They ascribed

mafic alkaline volcanism as the result of addition of amphibole phenocrysts in a typical

mantle derived magma. Although the details of the process are poorly understood, the

addition of a hydrous silicate phase, such as hornblende, that is rich in incompatible

elements to the mantle or mantle partial melts to produce alkaline mafic magmas is generally

agreed upon.

4. Benton Dike/Woodbine volcanic petrogenesis

Woodbine volcanic rocks contain significant amounts of Ti-rich hornblende

(kaersutite). Hornblende phenocrysts are noteworthy in these rocks in that they show

signs of resorption; the amphibole phenocrysts are corroded, embayed, and are altered to a

mixture of Fe-oxides and clinopyroxene (Figure 7). The interpretation of these amphiboles

is not straightforward; the amphiboles may be phenocrysts or xenocrysts. They may be of

primary mantle origin (xenoliths, incorporated into the alkaline magma as foreign material),

they may represent a cognate phase of the mafic melt that crystallized under high pressures

and predate crystallization of the other phenocryst phases, or they may have crystallized

simultaneously with the other mineral present in the volcanic rocks. However, the reaction

rims must indicate that the amphiboles were out of equilibrium with the latest magma.

Woodbine amphiboles may be a reflection of a metasomatic event that enriched the

mantle in incompatible elements. This enriched peridotite underwent partial melting to form

the alkaline mafic magmas that gave rise to the lamprophyres. Trace element abundance

and isotopic relations suggest that the mafic rocks of the Benton Dike Swarm represent a

mantle derived, relatively unfractionated magma (Morris, 1987). 87/86 Sr ratios for biotite

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

Page 158: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 4 3

firam a fresh monchiquite is 0.7043, Cr = 100 - 230 ppm, LREE abundance are 100 - 200

times chondritic abundances (Morris, 1987).

An association of mantle xenoliths containing amphiboles and isolated amphiboles

in the volcanic rocks may indicate that amphiboles are mantle xenocrysts (Best, 1974). The

Woodbine amphiboles are not likely candidates for accidental mantle xenocrysts; no

peridotite xenoliths are present in the Woodbine volcanic rocks. Furthermore, major

element modeling of the fractionation of the primitive Woodbine volcanic rocks indicates

that hornblende was an important fractionating phase of the mafic magma. Precipitation of

the amphiboles along with clinopyroxene, Ti-magnetite, feldspars, apatite, and sphene

resulted in the variety of igneous rocks present in the Benton Dike and in the Woodbine

volcanic cobbles; fractionation models suggest that all of these phases precipitated

simultaneously.

The slightly corroded and altered nature of the hornblende in the Woodbine

volcanics in rocks where the other phenocryst phases are relatively quite fresh, indicates

that they did not precipitate at rock solidification pressures. Woodbine amphiboles are the

result of relatively high pressure fractionation of the lamprophyric melt. Hornblende

phenocrysts must have precipitated from the lamprophyric mafic melt at some depth; as the

magma rose to shallower depths and experienced a loss of volatiles, the hornblende became

unstable and began to be resorbed back into the melt. Other phenocryst phases precipitated

with the amphiboles at the higher pressure but were not as strongly affected by the

depressurization and devolatilization as were the amphiboles. This slight resorption of

hornblende indicates that the mantle partial melt moved into the amphibole stability field,

crystal fractionation occurred, and then the melt moved outside the amphibole stability

field. Examination of a generalized P-T diagram which shows the stability of amphibole

reveals that this scenario is geologically reasonable (Figure 28). After rising and cooling, a

hydrous, mantle partial melt naturally enters the amphibole stability field. Fractionation of

the magma at this stage results in amphibole crystals in the magma. Adiabatic rising of the

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

Page 159: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 4 4

Mantle partial melt

Melt nses and cools

Crystal fractionation

Lamprophyric dikes

Figure 28. Schematic amphibole stability P-T diagram. After Gilbert et al. (1982).

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

Page 160: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 4 5

magma after amphibole fractionation results in the amphibole incongruently melting to

produce liquids plus crystalline phases such as pyroxenes and spinel after leaving the

amphibole stability field (Gilbert et al., 1982). The specific pressures and temperatures of

the fractionation of the Benton /Woodbine magmas are impossible to determine. Although

the shape of the amphibole stability field is well-established, the position of this curve

relative to the X (temperature) and Y (pressure) axes (Figure 28) is extremely dependent on

the volatile composition of the melt (Gilbert et al., 1982). The point where the amphibole-

in curve and the rock solidus curve intersect may vary from 0.5 - 8 kbar as PH20 faUs

relative to Ptotal (Gilbert et al., 1982). Woodbine amphiboles are the result of fractionation

of the lamprophyric melt at pressures greater than the dike intrusion pressures. Other

phenocryst phases precipitated with the amphiboles at the higher pressure but were not as

strongly affected by the depressurization and devolatilization.

This scenario explains the pétrographie and geochemical data of the Woodbine

volcanics; the overall petrogenetic picture for the Woodbine volcanic rocks can be

summarized as follows. A metasomatic event enriched the mantle peridotite in incompatible

elements. Partial melting of the enriched mantle gave rise to mafic magmas. The

lamprophyric magma rose and cooled; ôactional crystallization of clinopyroxene,

hornblende, Ti-magnetite, apatite, plagioclase, anorthoclase, and sphene resulted in a

variety of volcanic lithologies. This fractionating system gave rise to the intrusion of a

series of dikes and sills in central Arkansas; some of these magmas were extruded onto the

surface while the remainder of the melt was preserved as intrasive, shallow dikes. Periodic

tapping of the evolving magmatic system resulted in a variety of dike lithologies ranging

from ultramafic primitive dikes to more differentiated, felsic dikes. Rapid devolatilization

of the melt as the magma rose up into the dikes resulted in the partial resorbtion of

hornblende phenocrysts. The extrusive effusive and pyroclastic rocks at the surface were

subjected to weathering and erosion. The volcanic lithic fragments were carried by

Woodbine steams and were deposited in southwestern Arkansas, northeastern Texas, and

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

Page 161: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 4 6

southeastern Oklahoma. Belk et al.(1986) considered that east-west drainage patterns

along valleys formed by differential erosion of softer Ouachita facies, flanked by ridges of

more resistant units, were cut by rejuvinated streams that became incised across the

Ouachita front. Woodbine volcanic conglomerates would have developed where streams

cut through ridges. The present Mine Creek/Blue Bayou area must have been such a place.

As these streams flowed out of the Ouachitas, they flowed across a gravel beach which

marks the base of the Woodbine Formation in Arkansas (Belk et al., 1986).

Lamprophyric dikes occur throughout much of the central and eastern Ouachitas;

they are abundant within a 350 km^ area north of Benton, Arkansas. There seems to be no

sytematic distribution of dike compositions (Robison, 1977; Morris, 1987). Exposure of

the dikes is poor and unless covered by excavation or strem erosion, they are nearly

impossible to find (Morris, 1987). Dikes extend to within approximately 50 km of the

Woodbine conglomerates. The specific source of the Woodbine volcaniclastic material,

within the dike swarm, is impossible to determine.

5. Comparison with other lamprophyres

A recent study of the alkaline lamprophyres and related rocks of South Island, New

Zealand (Barreiro and Cooper, 1987) concludes that the wide variety of tinguaite

(phonolite), trachyte and carbonatite dikes evolved by fractional ciystallization of

amphibole, titanbiotite and apatite of the lamprophyres. As a result, the more differentiated

rocks are enriched in Na, K, Rb, Zr, and Nb, but are depleted in P and Ti relative to the

lamprophyres. The similarities in the differentiation process, proposed fractionation

assemblage, and resulting geochemical trends of the New Zealand lamprophyres and of the

Woodbine volcanic suite are obvious, and the field association of the South Island trachytes

and lamprophyres lends support to the hypothesis that these same lithologies found in the

Cretaceous Gulf Basin, although no longer associated in the field, are petrogenetically

related.

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

Page 162: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 4 7

In another study, it was determined that chemical and isotopic data from Miocene

trachydacite and minette (biotite-lamprophyre) sills in NW Colorado (Leat, et al., 1988) are

consistent with the derivation of a silicic trachydacite magma by fractional crystallization of

the lamprophyre. Again, as in the Barreiro and Cooper (1987) study, the field association

of mafic, lamprophyric dikes and felsic hypabyssal, or volcanic, rocks has been interpreted

as a genetic relationship; fractional crystallization of lamprophyric magmas resulted in

igneous rocks of granitic to syenitic composition.

6 . Summary

In summaiy, mineralogical, geochemical and geochronolgic support the conclusion

that the source of the reworked volcanic cobbles now preserved in the Woodbine

Formation of southwest Arkansas is the lamprophyric dikes of central Arkansas.

Lamprophyric dikes occur throughout much of central Arkansas. Identification of the

specific dikes that represent the intrusive equivalents of the extrusive Woodbine rocks is

not possible. The magmatic history and interrelationships of the various lithologies within

the dike swarm are not yet completelt resolved (Morris, 1987), however, previous work

(Van Buren, 1968; Robison, 1977; Steele and Robison, 1976) suggests that fractional

crystallization of a hydrous basaltic magma resulted in the variety of dike lithologies.

Previous provenance studies have relied on identification of detrital minerals present

and determination of the relative proportions of those minerals to identify source areas; this

study invokes a more quantitative approach. Geochemical and mineralogical examination

of potential igneous source rocks and the reworked volcanic material allows a more precise

identification of the provenance. Careful comparison of whole rock geochemical trends of

the reworked volcanic rocks and o f local igneous rocks can identify which of the potential

provenance centers is petrologically related to the volcanic sedimentary debris.

Geochronologic data provides important information concerning the provenance of the

epiclastic material. Precise, radiometric age determinations of the epiclastic material and the

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

Page 163: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

14 8

potential igneous sources have eliminated some centers; any igneous centers younger than

the sediments are eliminated as possible sources for the epiclastic material. Direct

comparison of radiometric ages of this volcanic debris and igneous centers can be used to

narrow provenance choices.

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

Page 164: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Chapter 7. Tectonic Implications o f K-Ar and ^OAr/^^Ar Ages o f Igneous Rocks o f the Northern Gulf o f Mexico Basin

1. Introduction

The upper Cretaceous igneous rocks of the northern Gulf of Mexico Basin are the

result of a poorly understood magmatic event; despite decades of exploration for oil and gas

in this region, the tectonic, petrogenetic, and geochronologic relationships of these

widespread and diverse igneous rocks are not well documented. These igneous rocks

range in lithology from the olivine nephelinite - alkali basalt - phonolite series of the

Balcones Province (South-Central Texas) to the lamproite of central Arkansas, to the

lamprophyre-trachyte-syenite-carbonatite series of Magnet Cove, Granite Mountain and

other alkaline. Si-undersaturated complexes (central Arkansas), to the alkali basalts and

trachytes of the Monroe Uplift and Jackson Dome (eastern Mississippi) (Figure 1) (Byerly,

1989; Morris, 1987). In an attempt to understand the petrogenetic and tectonic

relationships of these widespread, and yet volumetrically minor igneous rocks, a program

of geochronologic investigation has been initiated at LSU. The new radiometric data,

combined with data from other workers in the Gulf Basin, is used to construct a

chronologic picture of the Cretaceous igneous activity of the northern Gulf Basin.

Correlations between the timing of magmatic activity and either geographic location or

lithology could yield insight into the petrogenetic and tectonic relationships of the igneous

complexes. The radiometric ages are also used to aid in the interpretation of the provenance

of epiclastic volcanic lithic fragments now preserved in sedimentary rocks in the Gulf

Coastal sequence.

The new radiometric dates reported here are from 5 localities in the northern Gulf

Basin (Figure 1). Samples from the Magnet Cove intrusive complex (central Arkansas),

the Granite Mountain syenites (central Arkansas), the Monroe Uplift (Mississippi,

1 4 9

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

Page 165: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 50

Louisiana, Arkansas), the Jackson Dome (eastern Mississippi), and a sample of a trachyte

preserved as a cobble in the Woodbine Formation (southwest Arkansas) have been studied.

2. Analytical Techniques

These igneous rocks have been dated by three techniques; the trachytic Woodbine

cobble has been dated by the K-Ar method, the Magnet Cove and Granite Mountain

samples have been dated by the total fusion ^0Ar/39Ar method, and the subsurface samples

from the Monroe Uplift and from the Jackson Dome have been dated by a two-step

^ A r / ^ ^ A r incremental heating technique. The K-Ar method (Dalrymple and Lanphere,

1969) involves determination of the and ^Ar concentrations from splits of a single

sample. The dating technique (Merrihue and Turner, 1966) involves neutron

irradiation of the sample which converts part of the into 39Ar, the samples are then

heating in a vacuum line so that the argon is released from the sample for mass

spectrometric analysis. The heating may be limited to a single step, the total fusion

method, or may involve several steps of increasing temperature, the incremental heating

technique. The total fusion technique is advantageous over the K-Ar method in that it

yields higher precision ages and permits use of smaller subsamples; the incremental heating

technique is advantageous over the total fusion method in that it often yields information

concerning post-crystallization Ar loss as well as the presence of excess ^^Ar*.

A typical stepwise heating analysis involves isotopic analysis of Ar gas at 7 -10

temperatures (Merrihue and Turner,1966; Lanphere and Dalrymple, 1971,1976; Dalrymple

and Lanphere,1974; Berger and York, 1981; Baksi et al., 1987a); here we utilize a different

approach. Subsurface samples of ~200 mg from drillcore cuttings were available from

igneous rocks from the Jackson Dome and Monroe Uplift. The small sample sizes

precluded 7-10 step gas extractions as each step would release too small a quantity of gas

for precise and accurate mass spectrometric analysis. After sample irradiation, the whole

rocks were placed in tantalum cylinders in a high-vacuum extraction line and baked out for

~16 hours at ~200°C. Heating was achieved by passing a high current at low voltage

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

Page 166: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

151

through the cylinder using a modified version of a Bieri et al. (1966) type furnace. A two-

step heating procedure was followed with estimated temperatures of ~600°C and ~1300®C.

The gas samples were cleaned with a Zr-Ti alloy followed by a SAES getter, prior to static

analysis on a modified MS 10 mass spectrometer (Archibald et al., 1983). The advantage

of the two-step over the total fusion is that 1) it may be possible to determine if the sample

has been disturbed with respect to the K-Ar system, 2) it may be possible to determine an

age that approaches the original cooling age of rocks that have experienced partial Ar loss,

and 3) a high precision "age" should be obtained on the fusion step, by removal of most of

the atmospheric argon within the sample in the low temperature step.

The total fusion work was carried out at USGS, Menlo Park, with SB-3 Biotite as

the monitor, following standard techniques (Dalrymple & Lanphere, 1974). The two-step

work was carried out at Queen's University, utilizing Fish Canyon Tuff Biotite

as the monitor. All ^OAr/^^Ar ages reported herein, are relative to 162.9 Ma for SB-3

Biotite. The K-Ar analysis was carried out at Queen's University, using standard

procedures (Baksi et al., 1987b). All results have been (re) calculated to the decay

constants, isotopic abundances recommended by Steiger and Jager (1977).

3. Geochronologic Results

All analytical results are reported in Table 7; all sample descriptions and locations

are reported in Appendix 8. Four whole rock samples from the Monroe Uplift (Figure 1)

were analyzed with the two-step incremental heating OAr/^^Ar technique described

above. These two-step age spectra are presented in Figure 18. Sample NGB3-1, a fine­

grained trachyte, from a well near Washington, Mississippi, yielded a preferred, high

temperature age, of 81 ± 1 Ma (Figure 18). In this case, the total gas age and the preferred

age are essentially identical, with most of the atmospheric argon released in the low

temperature step; the whole-rock sample would appear to be relatively undisturbed.

Sample NGBl-1, a fine-grained trachybasalt from a well near Sharkey, Mississippi,

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

Page 167: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 5 2

yielded a preferred, high temperature age of 83 ± 1 Ma (Figure 18). Its release pattern is

relatively similar to that of NGB3-1.

Samples P-1 (3180') and P-1 (3530’) are from a single well in Madison Parish,

Louisiana on the Monroe Uplift (Figure 1). These two basaltic rocks demonstrate the

utility of the two-step heating method. The total gas ages for these two rocks, 76.7 ±1 .0

and 74.9 ±1.1 Ma respectively, (Figure 18) are not consistent with the stratigraphie

position o f these two volcanic rocks. The preferred (high temperature) ages of 78 ± 1 Ma,

are in excellent agreement. In each case, the low temperature step taps the more altered

portion of the rocks revealing the low age, and displays a large amount of atmospheric

argon. The second, high temperature analysis, yielded a high precision age that better

approximates the age of crystallization of each rock.

A single sample, NGB7-1, from a well near Rankin, Mississippi, on the Jackson

Dome (Figure 1), was analyzed using this two-step technique. In this case, the low

temperature step yielded a slightly older date. This medium to fine-grained microsyenite

yielded a preferred age of 67 ± 2 Ma (Figure 18) - essentially the average of the two steps.

Two mineral separates from samples of igneous outcrops in central Aikansas were

analyzed by the total fusion ^^Ar/^^Ar method. A biotite separate firom a syenite of the

Granite Mountain Complex (Figure 1) gave an age of 89.5 ± 0.7 Ma. A biotite separate

from a jacupirangite from the Magnet Cove Complex (Figure 1) yielded an age of 93.6 ±

0.6 Ma.

A trachyte, now preserved as a cobble within the Upper Cretaceous Woodbine

Formation in Southwest Arkansas (Figure 1) yielded a K-Ar date of 98 ± 2 Ma. All

relevant analytical data are presented in Table 7. K (and Ca) contents of the whole-rock

specimens were calculated from the measured amount of ^Aix and the 37ArcV^^ArK

values; we estimate a precision of ± 10% in these values, the large errors stemming

primarily from the uncertainty in the sample weights of the individual specimens.

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

Page 168: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 5 3

4. Discussion

Ages for northern Gulf Basin igneous rocks determined as part of this study and

ages reported in the literature are summarized in Figure 29. Our ages for rocks of the

Monroe Uplift agree well with Sundeen and Cook’s (1977) K-Ar whole rock date of 80 ±

3 Ma on a phonolite fiom the Monroe Uplift, but are of higher precision. Their K-Ar date

on a biotite separate from a biotite analcimite of 94 ± 3 Ma (Sundeen and Cook, 1977) may

not be a crystallization age. This sample is suspect for several reasons; 1) The K content of

this biotite (~6% K) is low on stoichiometric grounds; biotites from Magnet Cove contain

over 7 % K (Zartman et al., 1967): 2) the analcimite "age" is out of stratigraphie sequence

with the other K-Ar result from the same drillhole. The phonolite age, 80 + 3 Ma, is

supported as a crystallization age by our four results (Figure 29), which fall in

the range 78-83 Ma.

Our result from a microsyenite from the Jackson Dome is not entirely satisfactoiy,

as the low temperature step yields a slightly higher apparent age than the fusion step. We

estimate the age of crystallization to be 67 ± 2 Ma. Merrill (1983) reported K-Ar dates of

73.5 + 2.8 and 74.8 + 2.8 Ma for two samples from wells in Hinds and Rankin Counties

respectively. These results are in permissive agreement, as our microsyenite specimen was

recovered from a slightly higher level in the well in Rankin County than the one dated by

Merrill (1983). Volcanism at Jackson Dome apparently occurred over a few million years

at ~70 Ma.

Our biotite specimen from the Granite Mountain syenite contained numerous, small,

apatite inclusions, as reflected by the Ca/K ratio of the sample (Table 7). Nevertheless, our

total fusion age of 89.5 ± 0.7 Ma is in excellent agreement with Zartman et als

(1967) K-Ar and Rb-Sr results on the Little Rock syenite which average 89.5 ±1 .6 Ma; the

Little Rock syenite and the Granite Mountain syenite are recognized as two exposures of a

single intrusive complex. Our biotite separate from the Magnet Cove Complex showed

minor chloritization; this may explain why our total fusion ^A r/^^A r age of 93.6 ± 0.6 Ma

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

Page 169: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 5 4

is somewhat younger than the weighted K-Ar and Rb-Sr results (99 ± 2 Ma) on biotite

from the same complex (Zartman et al., 1967) (Figure 29).

5. Timing of the Cretaceous Igneous Activity in the Northern Gulf of Mexico Basin

Igneous activity during the Upper Cretaceous extended from south-central Texas

(Balcones Province) to eastern Mississippi (Jackson Dome). The rocks vary in lithology

from ultramafic lamproites and lamprophyres, to mafic alkali basalts and nephelinites, to

felsic trachytes and syenites. Any relationship between the timing of magmatism and either

geographic location or rock composition could yield important information concerning the

genetic interpretations on all of these complexes.

No strong correlation exists between sample lithology and age. While the most

primitive rocks, those of the Prairie Creek lamproite (Morris, 1987), are amongst the

oldest, the ultramafic rocks of the Magnet Cove Complex and of the Balcones Province

represent a wide temporal range of magmatic activity. The most differentiated rocks of the

trend, syenites of Granite Mountain and Magnet Cove and trachytes from the Monroe

Uplift, also represent a range of ages. No chronologic trend in magmatic composition is

evident at this time. Within single intrusive complexes, not enough high precision

geochronologic data exists to determine the relationships between mafic and felsic rocks; at

Magnet Cove, outcrop relationships suggest that the trachytes were emplaced before the

mafic rocks (Erickson and Blade, 1963).

Zartman and Howard (1987) propose that the intrusion sequence of the presumed

genetically related rocks in central Arkansas began in the southwest and migrated to the

northeast over a time span of 15 myr. They illustrate the geographic and geochronologic

relationship of the rocks in central Arkansas, with the Prairie Creek lamproite as the oldest

igneous complex and Granite Mountain syenites as amongst the youngest (Zartman and

Howard, 1987). We believe that the igneous rocks of central Arkansas must be placed in

the larger northern Gulf of Mexico Basin framework. The Upper Cretaceous igneous

rocks extend from south-central Texas to eastern Mississippi, and any correlations between

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

Page 170: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 5 5

the timing of magmatic activity and geographic location must examine rocks of the entire

northern Gulf Basin; the rocks of central Arkansas are a part of a larger igneous trend

(Figure 1). We have prepared a diagram similar to Figure 2 of Zartman and Howard

(1987) which includes rocks of the Balcones Province and volcanic rocks in northeast

Louisiana and eastern Mississippi (Figure 29). No simple age/location correlation exists;

the westernmost rocks (Balcones Province) and the easternmost rocks (Jackson Dome)

were erupted contemporaneously. However, when rocks of the Balcones Province are

excluded, a strong correlation exists between the timing of magmatic activity at an igneous

complex and the relative east-west location of the complex (Figure 29).

Compositions in the Balcones Province are significantly different than those found

elsewhere in the Late Cretaceous Gulf of Mexico Basin; the Balcones rocks are composed

of a higher proportion mafîc and ultramafic compositions compared to the dominantly

phonolitic and trachytic compositions of central Arkansas and eastern Mississippi (Byerly,

1989). The new radiometric dates reported here basically support the conclusion of

2^artman and Howard (1987) that magmatic activity migrated from the west to the east.

However, these igneous complexes are not strictly located along an east-west transect; the

igneous centers define an arc which begins at Prairie Creek and extends northeast to

Magnet Cove and Granite Mountain and then southeast to volcanic rocks on the Monroe

Uplift and Jackson Dome (Figure 1), Zartman and Howard (1987) very tentatively suggest

a hot spot or other mantle instability that controlled the location of magmatic activity.

Relative plate motion for the period of 65-110 Ma determined by Engebretson et al.

(1985) indicate that the curvilinear magmatic trend of the northern Gulf of Mexico Basin

was not controlled by movement of the North American plate over a hot spot. The model of

Engebretson et al. (1985) for the displacement history of North America for the period 65-

110 Ma shows that the plate was moving to the northwest relative to fixed hot spots. The

motion could explain the temporal magmatic trend along the Granite Mountain-Monroe

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

Page 171: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

156

Uplift-Jackson Dome segment, although it cannot explain the trend along the Prairie Creek-

Magnet Cove-Granite Mountain segment.

We believe that intraplate stresses, similar to those proposed by Solomon and Sleep

(1974) and by Pilger (1982), may be responsible for the time transgressive nature of this

volcanism. Intraplate magmatism may reflect melting at the base of the lithosphere that is a

result of pressure reduction due to extensional stresses within the plate (Solomon and

Sleep, 1974). Zoback and Zoback (1980) determined that a uniform extensional stress

field exists within the sediments of the Gulf Coast Province; the state of stress within the

bedrock underlying Gulf Coast sediments remains unknown. Although this extension

caused by active normal faulting may simply reflect sediment loading and not the state of

lithospheric stress, the faulting is consistent with a state of crustal stress that is expected at

a passive continental margin (Bott and Dean, 1972; Zoback and Zoback, 1980). Temporal

changes in the intraplate stress field, affecting pre-existing lineaments of the northern Gulf

Basin, may have resulted in a migration of crustal extension over the period of 65 -110

Ma. This extensional stress may have resulted in melting at the base of the lithosphere and

subsequent time transgressive volcanism at the surface.

6. Middle Cretaceous Volcanic Episodes

Evidence for global synchronism and episodic increases of volcanic activity at hot

spots has been documented by Vogt (1972,1979); global episodic increases of Cenozoic

volcanic activity at island arc volcanoes is evidenced by ash frequency in DSDP boreholes

in the Pacific (Kennett et al., 1977) and in the Indian and Atlantic Oceans (Kennett and

Thundell, 1975). Analysis of DSDP cores from the western Pacific Ocean identify two

major episodes of Cretaceous igneous activity; these cores demonstrate that volcanic

activity did not proceed at a uniform pace throughout the Cretaceous (Rea and Vallier,

1983). The two periods of increased activity are at 65 - 80 m.y. and at 95 -110 m.y.;

between these pulses, volcanic activity was not as extensive (Rea and Vallier, 1983). The

episodic Cretaceous volcanism may have been globally synchronized (Vogt, 1989). It is

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

Page 172: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 5 7

possible that the periods of increased Cretaceous igneous activity are also reflected in the

igneous rocks of the northern Gulf of Mexico Basin; episodic pulses of increased activity

recognized in the Pacific Ocean may also be reflected in continental intra-plate volcanism.

The volcanic rocks of the Balcones Province, the Monroe Uplift, and the Jackson Dome

were erupted during the 65 - 80 m.y. episode of Rea and Vallier (1983) (Figure 29).

Igneous rocks of the Prairie Creek lamproite and the Magnet Cove complex, as well as

other related alkaline bodies such as the Potash Sulfur Springs complex (Zartman and

Howard, 1987), lamprophyres of the Benton Dike Swarm (Morris, 1987), and the V

intrusive (Owens, 1968; Morris, 1987) reflect magmatism that occurred during the 95 -110

m.y. episode of Rea and Vallier (1983) (Figure 29). Only the extensive syenites of Granite

Mountain are not part of either of these two episodes and are instead part of the period of

waning magmatic activity, 95 - 80 m.y., of Rea and Vallier (1983) (Figure 29).

Interpretation of our data alone would not lend itself to the hypothesis of episodic

Cretaceous igneous activity; our data does however, fit nicely into the patterns of increased

volcanic activity described elsewhere (Rea and Vallier, 1983). Between these pulses of

activity, less extensive magmatic activity did occur in the Pacific Basin; the age of the

Granite Mountain syenites does not preclude the hypothesis of episodic activity within the

Gulf Basin.

The specific mechanism responsible for these periods of increased activity is not

well understood. Kennett et al. (1977) propose that rapid seafloor spreading and increased

plate motions may be related to increased volcanic activity in the Cenozoic, but they also

point out that the volcanic activity must amplify the effects of these pulses in spreading

rates.

7. Provenance of Volcaniclastic Sediments of the Northern Gulf Basin

The Upper Cretaceous Woodbine Formation of Southwest Arkansas is rich in

reworked volcanic material. Rounded sand to cobble sized volcanic lithic fragments of

varying composition are preserved in this unit. The volcanic rocks preserved as cobbles

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

Page 173: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 5 8

Balcones

PrairieCreek

.y w w n ii iw

MagnetCove

^ S m o o m m w

CranteMountain mmmumm

MonroeUplift V ' # : ' s ' '

tiw iM imnwi

JacksonDome

Ma 60 70 80 90 100 110

Ages determined as part of this study

Ages from the literature (see text)

Periods of increased volcanic activity (Rea and Vailier, 1983)

Figure 29. Summary of radiometric ages for Cretaceous igneous rocks of the northern

Gulf o f Mexico Basin; see text for discussion.

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

Page 174: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 5 9

range in lithology from mafic, clinopyroxene-rich rocks to intermediate alkali basalts to

felsic trachytes (Livesey and Byerly, 1989). The source o f this volcanic debris, although

long speculated upon (Ross et al., 1928; Belk et al.,1986), remains uncertain. Information

concerning the age of these volcanic cobbles has enabled us to eliminate some of the Gulf

Coast igneous centers as potential sources for the reworked volcanic material. A K-Ar date

of 98 ± 2 Ma on a trachyte preserved as a cobble in the Woodbine Formation indicates that

volcanic rocks of the Balcones Province (65-75 Ma), Granite Mountain (~90 Ma), Monroe

Uplift (~80 Ma), and Jackson Dome (~70 Ma) were not the source for the volcaniclastic

material of the Woodbine Formation. The most likely source of this material is the Magnet

Cove Complex and related lamprophyric dikes, 95-105 Ma, (Morris, 1987) of central

Arkansas (Livesey and Byerly, 1989).

8 . Summary

New radiometric ages of igneous rocks of the northern Gulf of Mexico Basin

indicate that when the volcaitic rocks of the Balcones Province are excluded, a strong

correlation between the timing and eastern location of magmatic activity exists. The oldest

complex, the Prairie Creek lamproite (95-110 Ma), is the westernmost o f these igneous

centers. Rocks of the Magnet Cove complex, the Granite Mountain complex, the Monroe

Uplift and the Jackson Dome are respectively located further east and are progressively

younger with the volcanic rocks of the Jackson Dome yielding an estimated age of 70 Ma.

This chronologic trend was first recognized by Zartman and Howard (1987) in rocks

located in central Arkansas; we extend this trend into northeastern Louisiana and eastern

Mississippi along an arc that runs northeast from Prairie Creek lamproite to Granite

Mountain and then southeast to the Monroe Uplift and Jackson Dome (Figure 1).

Temporal changes in the intraplate stress field may have resulted in a migration of crustal

extension over the period of 65 -110 Ma. This extensional stress may have resulted in

melting at the base of the lithosphere and subsequent time transgressive volcanism at the

surface.

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

Page 175: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 6 0

The ages determined in this study apparently fit into two global episodes of

increased igneous activity; periods of increased activity recognized in the Pacific Ocean

correspond to increased activity in the northern Gulf of Mexico. Byerly (1989) suggested

that global plate reorganizations were responsible for igneous activity along the northern

rim of the Gulf of Mexico Basin; worldwide Upper Cretaceous changes in plate motions

may have resulted in 1) episodic increases of volcanic activity in the Pacific Basin, and 2)

minor extensional stresses along the northern Gulf basin that triggered magmatic activity at

that time.

These new ages have yielded insight into the problem of provenance for the

epiclastic volcanic cobbles in the Woodbine Formation of southwest Arkansas.

Geochronolgic relationships indicate that Magnet Cove and related alkaline rocks (Benton

Dike Swarm, Potash Sulfur Springs Complex, and V Intrusive) are the most likely sources

for the Woodbine volcaniclastics.

Finally, the usefulness of the two-step ^^Ar/^^Ar technique has been demonstrated

when small subsamples are available for dating. This technique is a useful compromise

between the 7-10 step method, which requires larger sample and more laboratory time, and

the total fusion method, which may yield precise data, but no information concerning the

possible presence of excess argon or post-ciystallization loss of ^Ar*.

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

Page 176: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Chapter 8. Conclusions

1) The most general conclusion of this study is that epiclastic volcanic material

preserved in sedimentary rocks can be used to 1) determine the petrogenesis and

provenance for a suite of igneous rocks, 2) aid in a study of regional geology, and 3) add

insight into the general understanding of magmatic evolution. Volcanic rocks are

particularly susceptible to weathering and erosion; the ability to evaluate petrogenetic

processes by examining epiclastic material is of obvious usefulness in older terranes. This

study has demonstrated that techniques used in interpreting stratified volcanic rocks can be

applied to reworked volcanic rocks that are preserved as sediments.

2) Fresh volcanic rocks are preserved as cobbles within conglomerates of the

Upper Cretaceous Woodbine Formation. These volcanic rocks are lithologically and

geochemically extremely variable; they range from mafic, clinopyroxene-rich rocks (MgO =

5.50, S i02 = 49.30, K2O = 3.01), to intermediate rocks characterized by subequal

proportions of clinopyroxene and plagioclase (MgO = 3.24, S i02 = 50.39, K2O = 4.79),

to felsic rocks dominated by strongly zoned anorthoclase phenocrysts (MgO = 0.92, Si02

= 60.60, K2O = 6.30). Pyroclastic rocks preserved as cobbles are also present.

Examination of the minéralogie and geochemical trends of these rocks suggests that they

are cogenetic; they were derived from a single evolving magmatic system.

3) Qualitative and quantitative analysis of major and trace element variation within

the Woodbine suite of cobbles was intended to evaluate different fractionation processes

that could be responsible for this variation. Differing degrees of partial melting, magma-

mixing, fractional crystallization and AFC are differentiation processes considered as

possible mechanisms responsible for Woodbine variation. Mass balance fractionation

calculations suggest fractional crystallization of a primitive mafic magma as the process

responsible for this suite of rocks. Kinks in smooth trends on variation diagrams occur at

161

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

Page 177: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 6 2

discrete chemical points and are interpreted as the result of changes in the fractionating

mineral assembledge. Early precipitation of titanaugite and magnetite resulted in the first

fractionation segment which is characterized by decreasing MgO, FeO, CaO, Ti02, and

increasing AI2O3 and K2O. Precipitation of varying proportions of clinopyroxene,

plagioclase, hornblende and apatite resulted in the second and third fractionation segments.

The fourth segment represents a significant change in the fractionating assemblege to

anorthoclase, hornblende, clinopyroxene, and sphene.

This suite of volcanic rocks represents evolution of an magma chamber. However,

these different lithologies may be related to each other through time or space. These

chemical variations may represent pre-emptive magma chamber zonation or the periodic

tapping of an evolving chamber. Magma chamber zonation with coexisting layers of mafic,

intermediate, and felsic melts would result in some crystal settling. Throughout the entire

Laacher See sequence, all types of phenocrysts are found in varying proportions; crystal

settling within the chamber resulted in a mixing of phenocrysts (Womer and Schmicke,

1984). No such mixing is observed within the Woodbine suite. Mafic samples do not

contain any felsic-type pyroxenes or feldspars; sphene is abundant in felsic samples and is

completely absent in mafic. Crystal settling of dense minerals such as sphene and

pyroxene did not occur. It seems likely then that the Woodbine suite is related by periodic

tapping of an evolving chamber, mafic lithologies empted first, then intermediate samples,

and finally felsic lavas.

4) The source of the Woodbine epiclastic volcanic material has been long

speculated upon. Comparison of whole rock, geochronologic and mineral composition

data of the Woodbine volcanics and the potential provenance centers in the Gulf Coast

suggest that the source of the volcanic debris was the Benton Dike Swarm. These

lamprophyric dikes are found throughout much of the Arkansas alkalic province and are

particularily abundant near Benton, Arkansas. These dikes range from felsic phonolites to

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

Page 178: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 6 3

felspathoidal gabbros to mafic melamonchiquites. Several lines of evidence lead to the

conclusion that the Woodbine volcanic rocks are related to the Benton lamprophyres.

Geochronologic data for the two suites are sympathetic. A K-Ar analysis for a Woodbine

trachyte yielded an age of 98± 2 Ma which is consistant with Benton dike ages of 99 Ma

and 100 Ma (Morris, 1987). Granite Mountain syenites, Jackson Dome volcanics, Monroe

Uplift volcanics, and Balcones volcanics all yield ages inconsistant with the Woodbine

volcanic rocks. Whole rock chemical data from the Benton dikes and Woodbine volcanic

rocks are compatible. Variation diagrams suggest that the two rock suites could be related

to each other with the Benton dikes representing the more mafic and the Woodbine volcanic

rocks the more differentiated ends of a single evolving magmatic system. Mineralogically

and texturally the Woodbine volcanic rocks appear related to lamprophyres. The most

primitive Woodbine rocks are characterized by an abundance of euhedral mafic minerals

which is diagnostic of lamprophyres.

5) Radiometric ages of igneous rocks of the northern Gulf of Mexico Basin

indicate that when the volcanic rocks of the Balcones Province are excluded, a strong

correlation between the timing and eastern location of magmatic activity exists. The oldest

complex, the Prairie Creek lamproite (95-110 Ma), is the westernmost of these igneous

centers. Rocks o f the Magnet Cove complex, the Granite Mountain complex, the Monroe

Uplift and the Jackson Dome are respectively located further east and are progressively

younger with the volcaitic rocks of the Jackson Dome yielding an estimated age of 70 Ma.

This chronologic trend was first recognized by Zartman and Howard (1987) in rocks

located in central Arkansas; we extend this trend into northeastern Louisiana and eastern

Mississippi along an arc that runs northeast from Prairie Creek lamproite to Granite

Mountain and then southeast to the Monroe Uplift and Jackson Dome. Temporal changes

in the intraplate stress field may have resulted in a migration of crustal extension over the

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

Page 179: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 6 4

period of 65 -110 Ma. This extensional stress may have resulted in melting at the base of

the lithosphere and subsequent time transgressive volcanism at the surface.

These ages determined in this study apparently fit into two global episodes of

increased igneous activity; periods of increased activity recognized in the Pacific Ocean

correspond to increased activity in the northern Gulf of Mexico. Byerly (1989) suggested

that global plate reorganizations were responsible for igneous activity along the northern

rim of the Gulf of Mexico Basin; worldwide Upper Cretaceous changes in plate motions

may have resulted in 1) episodic increases of volcanic activity in the Pacific Basin, and 2)

minor extensional stresses along the northern Gulf basin that triggered magmatic activity at

that time.

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

Page 180: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

REFER EN C ES

Adkins, W. S. and Lozo, P. E. J., (1951). Stratigraphy of the Woodbine and Eagleford,

Waco Area, Texas. Fondren Seriez 4:101-164.

Archibald, D.A., Glover,J.K., Price,R.A. and Farrar,E., (1983). Geochronology and

tectonic implications of magmatism and metamorphism in the southern Kootenay

arc and neighbouringregions of southeastern British Columbia: Canadian Journal of

Earth Sciences, v. 20, p. 1891-1913.

Bailey, T. L., Evan, F. G., and Adkins, W. S., (1945). Revision of the stratigraphy of

part of the Cretaceous in the Tyler Basin, northeast Texas. A A P G B u lle tin 29(2):

170-186.

Baksi, A. K., Archibald, D. A., Sarkar, S. N., and Saha, A. K., (1987(a)). 40Ar-39Ar

incremental heating study of mineral separates &om the early Archean east Indian

Craton; implications for the thermal history of a section o f the Singbhum Granite

h a th o lo i ih ic C o m p lc x . C a n a d ia n J o u r n a l o f E a r th S c ie n c e s 24:1985-1993.

Baksi, A.K., Barman, T.R., Paul, D.K. and Farrar, E., (1987(b)). Widespread Early-

Cretaceous flood basalt volcanism in eastern India: geochemical data from the

Rajmahal/Bengal/Sylhet Traps: Chemical Geology, v. 63, p. 133-141.

Baldwin, O. D. and Adams, J. A. B., (1971). K^^/at^O aggg of the alkalic igneous rocks

of the Balcones fault trend of Texas. T e x a s J o u r n a l o f S c ie n c e 22:223-231.

Barberi, P., Bizouard, H., Clocchiatti, R., Metrich, N., Santacroce, R., and Sbrana A.,

(1981). The Somma-Vesuvius Magma Chamber: a Petrological and Mineralogical

Approach. B u lle t in o f V o lc a n o lo g y 44: 295-315.

Barbieri, M., Peccerillo, A., Poli, G., and Tolomeo, L., (1988). Major, trace element and

isotopic composition of lavas from Vico volcano (Central Italy) and their evolution

on an open system. C o n tr ib u tio n s to M in e r a lo g y a n d P e tr o lo g y 99:485-497.

Barker, D. S., Mitchell, R. H., and McKay, D., (1987). Late Cretaceous nephelinite to

phonolite magmatism in the Balcones province, Texas, in Morris, E.M., and

1 6 5

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

Page 181: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 6 6

Fastens, J.D., eds.. Mantle metasomatism and alakline magmatism, G S A S p e c ia l

P a p e r 215: 293-304.

Barreiro, B. A. and Cooper, A. P., (1987). A Sr, Nd, and Pb isotopic study of alkaline

lamprophyres and related rocks from Westland and Otago, South Island, New

Zealand, ia Morris, E.M., and Fastens, J.D., eds.. Mantle metasomatism and

alakline magmatism, G S A S p e c ia l P a p e r 215:115-125.

Basalt Volcanism Study Project (1981). Basaltic Volcanism on the Terrestrial Planets. New

York, Pergamon Press, 1286p.

Belk, J. K., (1984). A pétrographie analysis of the volcaniclastic Woodbine Formation

(Upper Cretaceous) of Southwestern Arkansas. Stephen F. Austin University,

Nachadoches, Texas (Masters Thesis), 93p.

Belk, J. K., Ledger, E. B., and Crocker, M. C., (1986). Petrography of the volcaniclastic

Woodbine Formation, Southwest Arkansas. T ra n sa c tio n s o f th e G u l f C o a s t

A s s o c ia tio n G e o lo g ic a l S o c ie ty 36:391-400.

Berger, G. W. and York, D., (1981). Geothermometry from 40Ar-39Ar dating

experiments. G e o c h im ic a e t C o sm o c h im ic a A c ta 45:795-811.

Berger, P. J., (1923). Der chemismus der lamprophyre, in P. Niggli. G e s te in s u n d

M in e ra lp ro v in ze n , I , B e r l in : 217-574.

Best, M. G., (1974). Mantle-derived amphibole within inclusions in alkalic-basaltic lavas.

J o u r n a l o f G e o p h y s ic a l R e se a rc h 79: 2107-2113.

Bieri,J.H., Dymond,J.D. and Koide,M., (1966). A high temperature ultra high vacuum

furnace for gas extraction from geologic samples: Earth and Planetary Science

Letters, v. 1, p. 359-360.

Bloomfield, A. L. and Arculus, R. J., (1989). Magma mixing in the San Fransisco

volcanic field, AZ. C o n tib u tio n s to M in e r a lo g y a n d P e tro lo g y 102:429-453.

Boatner, P.L., Jr., ed., (1961). Cretaceous of Southwest Arkansas, Guidebook to the

21 St Annual Spring Field Trip of the Shreveport Geological Society, 89p.

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

Page 182: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 6 7

Bott, M, H. P. and Dean, D. S., (1972). Stress systems at young continental margins.

N a tu r e . 2 3 5 : 23-25.

Bowen, N. L., (1928). The Evolution of the Igneous Rocks. New York, Dover

Publications, 251p.

Burke, K. and Dewey, J. F., (1973). Plume generated triple junctions: Key indicators in

applying plate tectonics to old rocks. J o u r n a l o f G e o lo g y 81:406-433.

Burke, W. H., Otto, J. B., and Denison, R. E., (1969). Potassium-Argon dating of

basaltic rocks. J o u r n a l o f G e o p h y s ic a l R e s e a r c h 74:1082-1086.

Byerly, G. R., (1989). Igneous Activity in the Gulf of Mexico Basin. D e c a d e o f N o r th

A m e r ic a n G e o lo g y , G e o lo g ic a l S o c ie ty o f A m e r ic a J : in press.

Cotera, A. S., (1956). Petrology of the Cretaceous Woodbine sand in Northeast Texas.

University of Texas, Austin (Masters Thesis), 189p.

Dalrymple,G.B., and Lanphere, M.A., (1969). Potassium-Argon Dating, W.H. Freeman

and Sons, San Francisco, 258 p.

Dalrymple, G. B. and Lanphere, M. A., (1974). 40Ar/39Ar age spectra of some

undisturbed terrestrial samples. G e o c h im ic a e t C o s m o c h im ic a A c ta 38:715-738.

Dane, C. H., (1929). Upper Cretaceous of Southwest Arkansas. A r k a n s a s G e o lo g ic a l

S u r v e y B u l le t in 1, 215p.

Deer, W. A., Howie, R. A., and Zussman, J., (1966). An Introduction the Rock Forming

Minerals. Longman Group Limited, 528p.

Durham, C.O., ed.,(1970). Guidebook to the 24^i Annual Spring Field Trip of the

Shreveport Geological Society, 37p.

Engebretson, D.C., Cox, A., Gordon, R.G., (1985). Relative motions between oceanic

and continental plates in the Pacific Basin: Geological Society of America Special

Paper 206, p. 1-59.

Erickson, R. L. and Blade, L. V., (1963). Geochemistry and petrology of the alkalic

igenous rocks at Magnet Cove, Arkansas. U S G S P ro fe s s io n a l P a p e r 425; 95 p.

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

Page 183: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 6 8

Ervin, C. P. and McGinnis, L. D., (1975). Reelfoot Rift: Reactivated precursor to the

Mississippi Embayment. G5A 5«//erw 86 : 1287-1295.

Eskola, P., (1954). Ein Lamprophyrgana in Helsinki und die Lamprophyrprobleme.

T s c h e r m a k m in . u . p e r t . M itt . Bd. 4: 329-337.

Pino, M., La Volpe, L., Peccerillo, A.,Piccaireta, G., and Poli, G., (1986). Petrogenesis

of Monte Vulture volcano (Italy): inferences from mineral chemistry, major and

trace element data. Contributions to Mineralogy and Petrology 92:135-145.

Flanigan, F. J., (1976). Descriptions and analysis o f eight new USGS rock standards.

U S G S P r o fe s s io n a l P a p e r 840: 1-10.

Funkhouser, L. W., B. F.X. and Humpris, C. C. J., (1980). The deep Tuscaloosa gas

trend of South Louisiana. O il a n d G a s J o u r n a l 78:96-101.

Gilbert, M.C., Helz, R.T., Popp, R.K., and Spear, F.S., (1982). Experimental studies of

amphibole stability, ia, Veblen, D.R. and Ribbe, P.H., eds.. Amphiboles:

Petrology and experimental phase relations, in, Ribbe, P.H., Series ed.. Reviews

in Mineralogy, Mineralogical Society of America, volume 98:229-354.

Guo, J. and Green, T. H., (1989). Barium partitioning between alkali feldspar and silicate

liquid at high temperatuer and pressure. C o n tr ib u tio n s to M in e ra lo g y a n d P e tro lo g y

102: 328-335.

Haq, B. U., Hardenbol, J., and Vail, P. R. (1988). Mesozoic and Cenozoic

chronostratigraphy and cycles of sea level change. S E P M S p e c ia l P u b lic a tio n 42:

71-108.

Hawksworth, C. J., Norry, M. J., Roddick, J. C., and Vollmer, R., (1979).

143Nd/144Nd and 87si-/86sr ratios from the Azores and their significance in LIL-

element enriched mantle. N a tu r e 280:28-31.

Hawksworth, C. J., van Rogers, N. W., Calstem, P. W. C., and Menzies, M. A.,

(1984). Mantle enrichment processes. N a tu r e 311:331-335.

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

Page 184: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 6 9

Hawthorne, F.C., (1981). Crystal chemistry of the amphiboles, in, Veblen, D.R., ed..

Amphiboles and other hydrous pyriboles-Mineralogy, in, Ribbe, P.H., Series ed..

Reviews in Mineralogy, Mineralogical Society of America, volume 9A: 1-102.

Hazzard, R. T., éd., (1939). The Centerpoint volcanics of Southwest Arkansas, a facies of

the Eagleford of northeast Texas. G u id e b o o k , 14^^ A n n . F ie ld tr ip , S h r e v e p o r t

G e o lo g ic a l S o c ie ty : 133-151.

Henderson, P., (1982). Inorganic Geochemistry. Oxford: Pergamon Press, 353p.

Hildreth, W., (1979). The Bishop Tuff: Evidence for the origin of compositional zonation

in silicic magma chambers. G S A S p e c ia l P a p e r 180:43-75.

Hill, R. T., (1888). The Neozoic geology of southwest Arkansas. A r k a n s a s G e o lo g ic a l

S u rv e y A n n u a l R e p o r t f o r 1 8 8 8 2: 56-58.

Hill, R. T., (1901). The geography and geology of the Black and Grand Praries o f Texas.

U S G S 2 I s t A n n . R e p o r t , part 7, 666 p.

Hoffman, A. W. and White, W. M., (1982). Mantle plumes from ancient oceanic crust.

E a r th a n d P la n e ta ry S c ie n c e L e tte r s 57:421-436.

Hoffman, P., Dewey, F., and Burke, K., (1974). Aulacogens and their genetic

relationship to geosynclines with a Proterozoic example from the Great Slave Lake,

Canada. Modem and Ancient Geosynclinal sedimentation, in. ' R. H. Dott Jr. and

R. H. Shaver, eds., SEPM Special Publication 19: 38-55.

Hooper, P. R., (1988). Crystal firactionatiopn and recharge (RFC) in the American Bar

Flows of the Imnaha basalt, Colunbia River Basalt Group. J o u r n a l o f P e tro lo g y

2 9 : 1097-1118.

Howard, J. M., (1974). Transition element geochemistry and petrography of the Potash

Sulfur Springs intrusive complex. Garland Co., Arkansas. University of Arkansas,

Fayetteville (M.S. Thesis) 118p.

Hull, J. P. D., (1925). Prothro Salt Dome, Bienville Parish, La. A A P G B u lle t in 9: 904-

906.

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

Page 185: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 70

Hunter, B. E. and Davies, D. K., (1979). Distribution of volcanic sediments in the Gulf

Coastal Province-Significance to petroleum geology. T ra n sa c tio n s o f th e G u l f

C o a s t A s s o c ia tio n o f G e o lo g ic a l S o c ie ty 2 9 : 147-155.

Hyndman, D. W., (1972). Petrology of igenous and metamorphic rocks. International

Series in the Earth and Planetary Sciences. McGraw-Hill Book Co., 533p.

Irving, A. J., (1980). Petrology and geochemistry of composite ultramafic xenoliths in

alkali basalts and implications for magmatic processes within the upper mantle.

A m e r ic a n J o u r n a l o f S c ie n c e 280A: 389-426.

Kennett, J. P., McBimey, A. R., and Thunell, R. C., (1977). Episodes of Cenozoic

volcanism in the circum-Pacific region. Jo u r n a l o f V o lca n o lo g ica l a n d G eo th e rm era l

R e s e a r c h 2:145-163.

Kennett, J. P. and Thunell, R. C., (1975). Global increase in Quaternary explosive

volcanism. S c ie n c e 187: 497-503.

Kidwell, A. L., (1951). Mesozoic igenous activity in the northern Gulf Coastal Plain.

T ra n sa c tio n s o f th e G u l f C o a s t A s s o c ia tio n o f G e o lo g ic a l S o c ie ty , A n n u a l

M e e tin g , p . 182-199.

Lanphere, M. A. and Dalrymple, G. B., (1971). A test of the 40Ar/39Ar age spectra

technique on some terrestrial materials. E a r th a n d P la n e ta ry S c ie n c e L e tte r s 12:

359-372.

Lanphere, M. A. and Dalrymple, G. B., (1976). Identification o f excess 40Ar by the

40Ar/39Ar age spectrum technique. E a r th a n d P la n e ta ry S c ie n c e l e t t e r s 32:141 -

148.

Leat, P. T., Thompson, R. N., Morrison, M. A., Hendry, G. L., and Dickin, A. P.,

(1988). Silicic magmas derived by fractional crystallization from Miocene minette,

Elkhead Mountains, Colorado. M in e ra lo g ic a l M a g a z in e 52:577-585.

LeMaitre, R. W., (1981). Genmix- A generalized petrological mixing model program.

C o m p u te r s a n d G e o sc ie n c e 7: 229-247.

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

Page 186: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

171

Livesey, C.L., and Byerly, G.R., (1989). Petrogenesis and source of epiclastic volcanic

cobbles from the Cretaceous Woodbine Formation, SW Arkansas: Geological

Society of America Abstracts with Programs, p. A200.

Lozo, P. E., (1948). Stratigraphie relationships of the outcrop Woodbine sand of

Northeast Texas. GSA Bu/ferin 59: 1337-1338.

Mancini, E. A., (1979). Late Albian and early Cenomanian Grayson Ammonite

biostratigraphy in Northe-Central Texas./oMr/ifl/o/Ba/eonro/ogy 53:1013-1022.

Menzies, M. A., Kempton, P. D., and Dungan, M. A., (1985). Interaction of continental

lithosphere and asthenospheric melts below the Geronimo volcanic field, Arizona,

U.S.A. J o u r n a l o f P e tr o lo g y 26: 287-302.

Menzies, M. A. and Murthy, V. R., (1980). Alkaline basalts on Nunivak Island, Alaska.

E a r th a n d P la n e ta ry S c ie n c e L e tte r s 46:323-334.

Menzies, M. A. and Murthy, V. R., (1980). Mantle metasomatism as a precursor to the

genesis of alkaline magmas; isotopic evidence. A m e r ic a n J o u r n a l o f S c ie n c e

280A: 622-638.

Merrihue, C. and Turner, G., (1966). Potassium-Argon dating by activation with fast

nuetrons. J o u r n a l o f G e o p h y s ic a l R e s e a r c h 71: 2852-2857.

Merrill, R. K., (1983). Source of the volcanic precursor to Upper Cretaceous bentonite in

Monroe County, Mississippi. M is s is s ip p i G e o lo g y 3:1-6.

Miser, H. D. and Purdue, A. H., (1919). Gravel deposits of the DeQueen and Caddo Gap

quadrangles. U S G S B u lle t in 690:

Miser, H. D. and Ross, C. S., (1925). Volcanic rocks in the Upper Cretaceous of SW

Arkansas and SE Oklahoma. Amer/cfl/z/oMrna/o/Sczence 5(9): 113-126.

Morimoto., (1988). Subcommittee of pyroxenes, commission on new minerals and mineral

names, Intemetional Minéralogie Association. M in e ra lo g y a n d P e tro lo g y 39:55-

76.

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

Page 187: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 7 2

Morris, E. M., (1987). The Cretaceous Arkansas alkalic province: A summary of petrology

and geochemistry, in Morris, E.M., and Pasteris, J.D., eds.. Mantle metasomatism

and alakline magmatism, G S A S p e c ia l P a p e r 215:217-233.

Morris, E. M. and Eby, G. N., (1986). Petrologic and age relations in Granite Mountain

syenites. G S A A b s tr a c ts w ith P r o g r a m s 18:256.

Murray, G. E., (1961). Geology of the Atlantic and Gulf Province of North America.

Harper and Brothers, 692p.

Nelson, S. A. and Livieres, R. A., (1986). Contemporaneous calc-alkaline and alkaline

volcanism at Sanganguey Volcano, Nayarit, Mexico. G S A B u lle t in 97:798-808.

Nielson, J. E. and Noller, J. S., (1987). Processes of mantle metasomatism; constraints

from observations of composite peridotite xenoliths, in Morris, E.M., and Pasteris,

J.D., eds.. Mantle metasomatism and alakline magmatism, G S A S p e c ia l P a p e r

215: 61-76.

Oftedahl, C., (1957). Studies on the igneous complex of the Oslo region, XV. N o r s k e

V id e n sk -A k a d . O s lo S k r . M a t .- N a tu r v . K l. N o 2 .

Olafsson, M. and Eggler, D. H., (1983). Phase relations of amphibole, amphibole-

carbonate, and phlogopite-carbonate peridotite; petrologic constraints on the

asthenosphere. E a r th a n d P la n e ta ry S c ie n c e L e tte r s 64:305-315.

Oliver, W. B., (1970). Depositional systems in the Woodbine Formation (Upper

Cretaceous), northeast Texas. University of Texas, Austin (Masters Thesis), 62p.

Owens, D. R., (1968). Bedrock geology of the "V" intrusive. Garland County, Arkansas.

University of Arkansas, Fayetteville, ( M.S. Thesis), 96p.

Oxburgh, E. R., (1964). Petrologic evidence for the presence of amphibole in the upper

mantle and its petrogenetic and geophysical implications. G e o lo g ica l M a g a z in e

101: 1-19.

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

Page 188: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 7 3

Pearce, J. A. and Norry, M. J., (1979). Petrogenetic implications of Ti, Zr, Y, Nb

variations in volcanic rocks. C o n tr ib r ib u tio n s to M in e ra lo g y a n d P e tro lo g y 69; 33-

47.

Pilger, R. H., Jr., (1982). The origin of hotspot traces: Evidence from eastern Australia.

J o u r n a l o f G e o p h y s ic a l R e se a rc h 87:1825-1834.

Rea, D. and Vallier, T., (1983). Two Cretaceous volcanic episodes in the western Pacific

Ocean. G S A B u lle t in 94: 1430-1437.

Ritchie, G. L., (1980). Divergent magmas at Crater Lake: products of fractional

crystallization and vertical zoning in a shallow, water undersaturated chamber.

J o u r n a l o fV o lc a n o lo g ic a l a n d G e o th e rm a l R e se a rc h 7: 373-386.

Robison, E. C., (1977). Geochemistry of lamprophyric rocks o f the eastern Ouachita

Mountains, Arkansas. University of Arkansas, Fayetteville (M.S. Thesis), 147p.

Roedder, E., (1984). Fluid Inclusions, in, Ribbe, P., Series ed.. Reviews in Mineralolgy,

volume 12, Mineralogical Society of America, 646p.

Ross, C. S., (1955). Provenance of pyroclastic materials. G S A B u lle t in 66 : 427-434.

Ross, C. S., Miser, H. D., and Stephenson, L. W., (1928). Water laid volcanic rocks of

early Upper cretaceous age in southwestern Arkansas, southeastern Oklahom, and

northeast Texas. U S G S P r o fe s s io n a l P a p e r 154F: 175-202.

Sack, R. O., Walker, D., and Carmichael, I. S. E., (1987). Experimental petrology of

alkalic lavas: constraints on cotectics of multiple saturation in natural basaltic

C o n tib u tio n s to M in e ra lo g y a n d P e tr o lo g y 96: 1-23.

Scott, G., (1926). The Woodbine sand of Texas interpreted as a regressive phenomenon.

A A P G B u l le t in 10: 613-624.

Solomon, S. C. and Sleep, N. H., (1974). Some simple physical modles for absolute plate

motions. J o u r n a l o f G e o p h y s ic a l R e a s e a r c h 79:2557-2567.

Spencer, A. B., (1969). Alakline igneous rocks of the Balcones Province, Texas. J o u r n a l

o f P e tr o lo g y 10: 272-306.

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

Page 189: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 7 4

Spooner, W. C., (1926). Interior salt domes of Louisiana. A A P G B u lle t in 10: 217-292.

Steele, K. F. and Robison, E. C., (1976). Chemical relationships of lamprophyre, central

Arkansas. E O S A m e r ic a n G e o p h y s ic a l U n io n T ra n sa c tio n s 57:1018.

Steiger, R.H. and Jager, E., (1977). Subcommission on geochronology: convention on the

use of decay constants in geo- and cosmo- chronology: Earth and Planetary Science

Letters, v. 36, p. 359-362.

Sundeen, D. A. and Cook, P. L., (1977). K-Ar dates from Upper Cretaceous volcanic

rocks in the subsurface of west-central Mississippi. G S A B u lle tin 88:1144-1146.

Valdovinos, D. L., (1968). Petrography of some lamprophyres of the eastern Ouachita

Mountains of Arkansas. University of Arkansas, Fayetteville (Masters Thesis),

146p.

Van Buren, W. M., (1968). Geochemistry of syenite in Arkansas. University of Arkansas,

Fayetteville (Masters Thesis), lOOp.

Vogt, P. R., (1972). Evidence for global synchronism in mantle plume convection and

possible significance for geology. N a tu r e 240: 338-342.

Vogt, P. R., (1979). Global magmatic episodes: New evidence and impplications for the

steady-state Mid-Oceanic Ridge. G e o lo g y 7:93-98.

Vogt, P. R., (1989). Volcanogenic upwelling of anoxic, nutrient-rich water: A possible

factor in carbonate-bank/reef demise and benthic faunal extinctions? G SA B u lle tin

101: 1225-1245.

Wass, S. Y. and Rogers, N. W., (1980). Mantle metasomatism - precursor to continental

alkaline volcanism. GeocWmica et Cosmoc/ii/nicfl Acta 44:1811-1823.

Wilshire, H. G., (1987). A model of mantle metasomatism, in Morris, E.M., and Pasteris,

J.D., eds.. Mantle metasomatism and alakline magmatism, G SA S p e c ia l P a p e r

215: 47-60.

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

Page 190: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 7 5

Wilshire, H. G., Nielson-Pike, J. E., Meyer, C. E., and Schwarzman, E. L., (1980).

Amphibole-rich veins in Iherzolite xenoliths, Dish Hill and Dead Man Lake,

California. A m e r ic a n J o u r n a l o f S c ie n c e 280A; 576-593.

Wilshire, H. G. and Trask, N. J., (1971). Structural and Textural relationships of

amphibole and phlogopite in peridotite inclusions. Dish Hill, California. A m eric a n

M in e r a lo g is t 56:240-255.

Womer, G. and Schmincke, H. U., (1984 (a)). Minéralogie and chemical zonation of the

Laacher See Tephra sequence (East Eifel, W. Germany). J o u r n a l o f P e tro lo g y 25:

805-835.

Womer, G. and Schmincke, H. U., (1984 (b)). Petrogenesis of the zoned Laacher See

tephra. J o u r n a l o f P e tr o lo g y 25: 836-851.

Wright, T. L. and Doherty, P. C., (1970). A linear programming and least squares

computer method for solving petrologic mixing problems. G S A B u lle t in 81:1995-

2008.

Zartman, R. E., (1977). Geochronology of some alkalic rock provinces in eastern and

central United States. A n n u a l R e v ie w E a r th a n d P la n e ta ry S c ie n c e L e tte r s 5:257-

286.

Zartman, R. E., Brock, M. R., Heyl, A. V., and Thomas, H. H., (1967). K-Ar and Rb-Sr

ages of some alkalic intrusive rocks from central and eastem U.S. A m e r ic a n

J o u r n a l o f S c ie n c e 265: 848-870.

Zartman, R. E. and Howard, J. M., (1985). U-Th-Pb ages of large zircon crystals from

the Potash Sulfur Springs igneous complex. Garland County, Arkansas. G SA

A b s tr a c ts w ith P r o g r a m s 17:198.

Zartman, R. E. and Howard, J. M., (1987). Uranium-lead age of large zircon crystals

from the Potash Sulfur Springs igneous complex. Garland County, Arkansas, in

Morris, E.M., and Pasteris, J.D., eds.. Mantle metasomatism and alakline

magmatism, G S A S p e c ia l P a p e r 215: 35-240.

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

Page 191: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 7 6

Zoback, M. L. and Zoback, M., (1980). State of stress in the conterminous United States.

J o u r n a l o f G e o p h y s ic a l R e s e a r c h 85: 6113-6156.

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

Page 192: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

A ppendix 1 . C linopyroxene Com positions

177

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

Page 193: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

17 8

gooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

p s îS i ïS s iI s î î i i i ï i l iS S sI î iI s s I s s i î i l l i s s s I l î i s i l iS l i l j s

— — —

p a s ’ s S a i a s S S i î î a i S S a a i s S S I S I s i s î î I s s i l l s H S I s i S S î S l i J S S i ; ^

psasjSsisaassisssSsiiiîiisiisîSîssijssisassasssaisssiissssa

i a s a a s j a î S j s j i i i i a a i s î s s s S s s s s s a s i s a s a a a s s s s a s a î j s s S s s s i ^

liliyiiiiiiiîlliîitiMfîiliiiiyillliyîi9oooo99ooooo9^9oo9ooooooooooooooooooooooooooooooooooooooooooo

î a=sa5ii2!55i!j55i5fsa53|||5is!!iiâilli!iiil^

! -S!S5l5l3Sss55-55S5lsSS!i|sill|iilil|!!i!5l|!!j|ll^

i sillilli=illi i i=l5lsail5S3l§35j§l3Sllll i

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

Page 194: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 7 9

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

lsîiîS!îili lS!slSis5sisi!5îîSii l l l l î lS!s5!îll!SISssÎ2SÎ;Slll i^P£

: ...............................................................................................5 ......................................................%

|3S35g:5::3s:;:;35;;;;3§!:!353§!!3S3::Ss333s:s^

j s s ;S 3 3 g s 5 g 3 S S 3 g !3 H 3 S 3 3 3 3 g S 3 g s g 3 S 5 5 S 3 S S 5 s S s 5 5 s g 5 5 g s s ^

j3 3 3 3 :g 3 g 3 S 3 3 3 3 S 3 3 !3 s S 3 !3 3 S 3 s S § S sg s : !5 :3 S S 3 3 g 3 3 3 S § 5 s S g 5 S 5 ^

|3 3 3 3 3 g s g : 5 5 : 5 3 5 ! 3 5 5 : 5 : 5 5 5 5 3 2 5 5 5 5 5 ! 3 S : 3 3 S 5 3 3 3 3 3

iiliiiiiiliiiiiiiiiiiiiliiiiiilliiiliiiiiisiiiiiiiii

8 | l l l l 5 l i l S = l i S = M 5 5 5 | H i = 5 i s i = l 3 5 | i j | ^

| = ! g : 2 5 î S : ! 3 5 5 = 5 ! i l i 5 ! I S 5 i 5 3 2 | 5 i s l ! ^ 5 S ! l i

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

Page 195: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 8 0

i - s l ' - l i l i l s iS is l îH isS i i iS l l l is i l iSISSssISIsIs i»

l s i i i S î î H 5 s S s : i ! l s i l ! S 5 ; 5 l 5 | I S i s i î ! l i i H : 3 S I I S : | i ! 5 I S s s Î 5 l : | S s

l l s S l î s î i i ! S : s ! 5 5 ! : ! ! 5 s ü l s Î ! ; s ^ : ; ; ; s l ^ H i 3 l ! 5 î î s l : ; 5 S 3 l ; ! 5 i 5 | 5

ssHîîîsiiSlîISSîsSsIsljsHSIsISssSJISssiSsIsîsSsSIsîjsSsssIS!

s S S S S s s I s l I s S s i l s s s i s s I s S s i s l l i s s i ï J s J i l S S I s I s i s s I ' S s S i l S »

S i S s s s s s I H S l S s s s s s S s s s s s S s s s s I s î î s s j j j S l l î s s s s s j s S s ^

: H î 5 : ? 5 5 : d= 5 ! :5 :5 = 5 5 := = = = = 5 ï 5 ! 5 S 5 5 ; s5 ! 5 : * ! 5 ! 5 ! 5 : 5 : î î 555:=5555

liïîilslilîiiiiliiiiiillillliiyiyiliisiiii^iiliiillï

S | ! i l l » J ! 3 l ! ! 2 5 5 j : 5 5 i s i r = 5 S ! 5 E i 5 i l i i ! ! | ! Î ! 2 S 5 S ^

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

Page 196: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

181

M§00000000000 0 0 0 0 0 0 0 0 0 0 0 0 0

gOOdooooooooooooooooooood

o p o o o o p o p o o o »_oooooooooooo<10000^0000000

p

goooooooooooooooooooooooc

sg-SrZrrSrSSgSîS-SSîSïïSZSS-Si*®d®d®®d®dddddddddddddd®d

Sg28»S333SSSSSSS*gSSSS»253■sdd^ddddddddddd^ddddd^ddda£;SK^sssS'TB^;:s%^;:si:2s:R:3:"dd^dddd^d^ddd^dddddddddda

— dooddd^ddddddddddddddddd

odooodddddddddddddddddddd

| = gS 2»5"2“ S::SSSRRSi!^SSKJ;2B

2 " t : s s S " !3 2 : 3 z s :3 z 3 3 % :5 : : :BA A " • « * ' ' ' > _' _■ _» _* .* « «

Ainci ^ (M ^ ^ V o — "» «m m«®^ooo®999oo®ooo999o99o9o g o o o o o o o o o o o o o

i ldooo^ddddddddddd^d^ddddd

Oe^dP<;*»o*“ r » 9 « o o ( o o o o h » ^ f ^ * * o o r < * o oP ^^t'icicici'^cicincieicicicincincicicicici

S o o 9 9 9 o o 9 9 9 9 o o 9 9 o 9 o 9 o o o 9 9 o o o o o o o o o o o o o o

9S^ëSSSoSô3KS;:SSggg3S»SSoxd^edddddddd^ddddd«*ddd^d«~oSSSSSSSi^SRSSSgaSSgSiSSSgS

) = 5 2 3 a 3 : : : 3 ; | 5 ! ! s : : ! : 5 | ; !

Q3SS%32:KS=ssz:s :z2Ss%»s:KdddddddddNNM:dNNdNNNdd'~d^:

gî2SSS::SS~2SSSSSgSS9«,!?S$!5::g^.«moNOf'.d2p*dddddddd2vddddd

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

Page 197: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 8 2

oooo^ooôoôooôoôoooodoooooooooooboooooooooooooo^oodoooôooooooo

J s I p s s s s S s i ^ S s s s p p a s s s R s S s p s s s s s s S s s s s S s S s S s s S s i s s s s B S s s pooooooooooo^odo^odeddddddd^d^ddddddd^dddd^d^d^dddedddddeooo^d

e cOOoooôooooddodoaooododdooôôdoôdo«-«-'^dodôddoôoo’ôdv*^doôod'~. od*~oo'> «or*h»«Dro io<o»iof«iAKr«>iAiA4ear«404or«<AM«» Kior«i«iAf«>«ia«04or*h« <

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

Page 198: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 8 3

^ S l l l 2 5 5 ! 2 2 l s ! 5 i 2 i i l 2 2 l i 2 5 5 2 l ! l ! l i l 2 l 2 2 ; 2 2 l f l 2 i 2 l 2 5 2 2 2 2 2 2

fâ2 : :2 :5S S = : :3S 3 : :3SS 23 :z252% SR K S :35?28Sa2as :3zR 5% % :35s«z% S 2:28o^ooo^ooeo^ooooooooooooooodooooooooodoo^odooo^ooooeooo^oeto^ooo

M 5 g :^ ^ ^ ^ g M g ^ S g « S 5 g S 5 g 3 ; : 5 S a 3 S = S S S S = = : = = ; g s : 8 : ; B 2 3 2 = : = 3 g : : 2 3 gooooooooo^eo^oooooddddddoedd^^ddddeoddddddddddeddddddddddddodd

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

Page 199: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 8 4

edooeoo^dooo^oododooooooooooooooooooooododoooooo^odooooooooooo

^S225S2S22222SS22222226 5S2S222222S2S22225b2S52252552S52522222sS §ê 8 S S 8 S S 8 S 8 S 3 8 3 3 S S 3 S S 3 8 8 8 3 8 3 8 S 3 8 B S S S B 3 S 3 3 S S S 8 8 S S S S 8 8 S S S S S B S S S S S S

< a K o

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

Page 200: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

185

0 00000 ^ 000 ^ 0 0 0 '

.5S:SK;2R32: :3SS:S%Sg22SS5fr>p>r»rjp>oor>oc»r>r»ooooono.r»nr>o

ooeoooodooooooooodd^oooo

. . . . ..-lOfotBforHxroror r ooopxooiDXKo g O O O O O O O O O O O O O O O O O O O O O O O O

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

Page 201: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Appendix 2. Intermediate Feldspar Compositions

186

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

Page 202: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 8 7

lliliiilliiiilliliiiiillillillilillliiilliiliillllililflllliiil

lîliiiiliilllililiilliîlililliiilllllliilliiîllllillliilliîlill

liliiiiyiislllilllllliilslillliilliliisillllilliillllllill»

I î î l ! l ! ! l l l i 3 î î l ! i l l ! l l s ! î î l l i l î i î î ! ! ! l ! î i î î i ! î î ! î ! l l ! î î l î i 5 i l l

J 5 i s i l s 3 i ” 5 s 2 j i 3 ; 3 5 ! s s 3 2 3 2 5 s 5 s î s i s S 5 î î 5 l 5 5 i 3 l i î ; 5 s ! i

i ! l i s 2 i 3 î 3 2 3 î ; i S i ! 5 î ! 3 l î i ! S 5 l i i î s ; i S l î l 2 s i S 5 s î î l 5 î 5 ! l î 2 Î ! s j $ 3 î î

I3î;!3!î;;3:;2ïsl533!s223:3f!2s;l:355535:;î5:3$!î333:3j5s5!3î:5

} ! is l ! l î ! i l i ! s is l l2 2 l l î i s l l î i î î i i l ! ! î ! î î i î s s l l l i î ! ! î ! ! î i i ! î î2 î i

S S s s j a S s S s s J s s s î I s S s j j S î S s ï î j s s i l s j i s s î s J i î j s s s j s S I s j s s s î s s

i l i l i l i i l l I i l l l l l l l i s I l i s i i i s s i i i i i i i i i i i s i i H

< o lU

I 1

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

Page 203: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 8 8

III

I s i£bd

grp2SS

m

i l l

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

Page 204: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

189

^ s s s s s s s s a s a a o o ô s o â o o s o o s o i s o i i o s s s s i i s i s s s i i s i i i i s i s o ^a

SS52?999^ooooeoooooooooooooooooooooeoeoooooooooooooooooooooeeoo S S 5 5 5 9 9 ? ^ ^ o c > o o o o e o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o e o o o o o « • p p o o p o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o e o o o o o o p e o o o o o o d o o o o o o o d o o d o o o o o o o o o o o o o e o o o o o d d o o o o o o d d o o o o o o d o d d o o o o

| s s | s s s s s s s s s s & 5 s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s â s s s s s s s s s^ p p p p p o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o oâoooooddddddddddddddddddddddddddddddddddddddddddddddddddddddddd

l = =:ssss5S:: : :SSS2SSS8S:2SS8SSS8gSS=3Sg5S:SS3S5SS8SSSS2S5SSSSSSSSp p p e p p p e o o o o o o o o o o o o e o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o oj looooooeddddddddddddddddddddddddddddddddddddddddddedddddddddddd

! s s s ^ s : s : : g s : s s 2 : s s s 8 s s s s s 3 S S N 3 8 : § g g s s s s 3 s : § g S s ^)Cdooooodddddddddddddddddddddddddddddddddddddddddddddddddddddddo

issl i l l s s s s s l l s s s s l s s s s s i s i i i l i l s l i i l s i i i s i s i s i i s i l l l s s i s s s m i i ;

I I I

gsss333 :33z33S5%gK SSK R3; :S9 ;s% ss :aK :ss5S822s ;s2S : : :8S :S3=z% :2SS3^oeoooeoooddddddddddddddddddddddddddddddddddddoddddodddddddoooo

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

Page 205: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 9 0

Hi

A 6 6

I s s

l i ?jbd

!!!

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

Page 206: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Appendix 3. Felslc Feldspar Compositions

191

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

Page 207: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 9 2

îiiiiilliliiillllliliilliliililiilillillliiiiiilillliliiilliill

lillilliillliililillliililliliîîiiliiiiiliilillliiilililiililll

îiililiiiiiîliiiiiliiliHIiHIÎiiiiiliiiiiiiiiisi

i ! l îS i l l i lSSIî l îSI ! ! ! f l sS l l s l î !5 îS l î l î sS ! l l î î î l î sS i s i î î l s l iS î l l

I S ; j $ 3 3 5 ; ! i 3 s î i s ! l i 5 J 3 ; 5 ” 5 î a 5 5 j l : i i ! s i 5 3 i 3 j 5 Î S i l S I ! s ? 3 s s l j ^

35533!!::S5jis!jij5!!!:!5i5:s:S535H5î53::H5SilS!!!^

i S : 2 : S 5 5 s ! 3 S 2 l ! 3 S 3 5 : 2 3 5 ! 2 s 2 : ; i 3 s 3 3 ! g S 3 5 5 ” ! J 5 ! ; S 5 5 S 3 3 ^

8 3 i ! ! ! : ! l S j 3 S s : 3 S ï : î l 3 5 5 3 S 5 l s 5 î : ? î § 2 : 3 5 3 3 ! ; S 3 ! S S 5 3 § ! S : ! 3 : î 3 i 5 S s

î i i î l î s l l i s î 3 l s S î î l $ î l i î l î 3 i î î i î î i î i î S î l i l i S 3 l l î î l l l l l 3 î ! l i l ! î f

S i S s s s S i s s H s s s I S s n î s s S S s M s s S S S l S s s J I J S ’ S s S S s s S j s î s I s s S s s s s S s

IlsIllilIlillliilliiillîliPllllilsMljlilsllH

iîiiliillilllliliMliiilliiliilllliiliilliSiiyiiiliym m o

il il il

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

Page 208: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

19 3

î i s I s s s s i i i s s s s s i s s s I i P g g i l g i i i i s i s s g i l i g

_ 9 O 9 0 40 ^ W A 40 O a # 9 40 40 #0 C» — w vw v# w* ■■>« V %#» v« w w w# w# w •'« w w# w «v mw w w «rf« ««« ^ «h# **# w w w w w OJ «■»^n^oooononnnonnnnnnnnnnnnonnoCTnoricinnt'innnnnnAnnnnnffinnmnnrimmeinnnn

S s S 5 9 9 9 9 9 9 9 o o < ^ < > 0 9 o o o A o e e o e o o e o o o o e o e o o o o o o o e e o o o o o e o e e e o e o e o e o oo S S S S S 5 S 5 9 S 9 ^ ^ ^ o S ^ ^ ^ ^ o o o e o o o o o o o o o e e o o o o o o o o o o o o o o o o o o o o o o o o o o o~ R ^ ^ ^ ^ ^ P P P P ^ P R 0 O ^ p o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o e o o o o o o o o opooooooooooeoooooooooooooeooooociooooooooooooooooooooooooooooood

l i s g s i s i g i s P i s s i g i s s P l i i l l i l l I s s i s i i g l i l S s i p i i i g ^j o o o o o o o o o e d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d

e(DM*>^oaoM^ie^«M4O«OQ«0tat«««r«a»«(A4D«ouiM<»«DOu>«tttr<»oiw*n«>^«>«iAr>»Mr»MCMa^e«nflkoevaaiO4O A^ ^^o^e ^oo<**oooooooveoooo^oooooo« ' OCONN^' ~«*«>«**««*«»^^oovc<»*«~* ' * ' ^^ovvoo**o . ppppoopooooooooooooooooooooeooooeooooooooooooooooooooooooooooo dooooddoddeddddoodo'deddddedddddeddddddddddddddddpddpdddeddddddd

l s P n 3 i g i S 5 = , 5 p S 5 5 | S a | g | S S 5 S S S S g | g | | g g | R S = R = p S S S = 5^j t foooooooooddddddoddddddddddddddddddddddddddddddddddddoooooooddd

C«4eo#-(0* o*NNeyN«m(Wio#w» — w»o*»o»iûN*-*40 40N* «ow»mc»*N«o#«oa#p*Mmoo«*M*»m — oivtmioaS3SgSg83gg33SSSSSSSSg2358SSS8SaS8=8g::8SgZS8=:5Sg:gggggg5g3g3SAdddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd

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

Page 209: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Appendix 4. Pyroclastic Feldspar Compositions

194

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

Page 210: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 9 5

ÇeoooNeeeeoeoûoooeoooo o o o e o o o o o o o o o o o e o o o o*700000000000000000000(3Soooodoooooe>ooodooooo

:çur«>«ovakMaNoa»r>>r)ou)oo(QinN|C>p}n^<vmoc>nnn^n^o^n^nnSoooooooooooooooooeoo

i S S ! ___________________________________________________ ___

*o»mo«o*mmmMvv*AvONmoiA*nMoootooaooooaiockokotooooc*g)Moicic<dMoicic<jciMcjcioiojcioic«Moi

:%88832s;s looaaooiddei

; s s a s 5 soedddddddddddoddddde

o»ow»>»« ««»oooo»i/»»»ooooowpc<«e<oojop« of>»ooo«';eioMoopdddd^d«~ddddddd«*ddddd

S

o o o ^ ^ ^ o o o o c

— or.r)inmo«oOP$«o*-oo^o N^^ppppWNpvh»ppp«) — OOpjdddddddddddddddddddd

=“ SS5“ s2S~KsS5Ss5gS5j j 2 S 2 ^ ® 2 « S S o i o i o i 2 » ® “ ° “ ° '

• ooi<r«0'*ci«onro«ou>«inr<»tA<««r®ui pppnpppppontociiAc^ooNnngoaoddddddddddddddddd

pMr»or»«peiA«r»«ecvr ioo««irtwppppppppp*»pvppmpvppnO<O(PO<O<OOV><O<DIO<OÔÔtO<OM>4O<0tf>a

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

Page 211: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 9 6

_ *-• * # #'# W «— w «'1 ▼ W %V %% V, M# 1g999999999999P999999<

_ooaa«»Ov«»aoaiAaoooaeoo o o C» Ok o o o Ok 0» p p p p p o p p p o p

#SBS2 s : a s B S : s s s = s 2 S8 s^oooooooooooooooooooo^oooooodoooooodoooooo

»? e 2 p in n «08 O O O ^ CM «•

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

Page 212: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Appendix 5. Detrital Sanidine Compositions

197

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

Page 213: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 98

Ijliiliiilfiliiiilliiiillliiiillilliililililllliililililfilliii

l l l l i l l i i l l l i l l i l i i i î i l l i i î i l l î l l l l l l i i l l l l i i i i l l f l l l i i i l i l i l l l

I lîiisSî;;:IS2:l-ils»i” SSÎiîliiîli!| |ïi!

S SïSSÎsSs^ssSlSsîSssSîHSîâssîSsâssîSsssssssssssIâssSîsss

I i s i s i l l i l i i l i i l i l i s i S i i i s i i l i i i i l i i i

8 illiiiilisllîisiiilisiisliliiiiiiillisiîiirliiy

i i i I II

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

Page 214: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

1 99

I

l î î M

i l s i l

sS«i2 «i

§32%:X « • » fs

83252

85235

g SS 3 2 «2222

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

Page 215: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

2 0 0

s i s g i i i s s S i i s i e i s s s s g i s g g i g i g ^ i»«o r>Ar»oinu)cycsir>»M«oiAr«>iAors<M»><>»otA«<MaiA<s«cvcMCMC4CM<<><4n(2Nne>(o«Mvu>r«inintAtAin«i>intn«D<or)'vooooooooooooooooooooooooooooooc

3a

i l g g g g g ô S S S g g S S S g S g S S g g S S g S S S g S S g S ô S g S S S g S S S S S S S S S S S S S S S S S S S S S S— SPPR^^PR^poeooooopopoooooooooooooeoooooopppopppppppopc -------— - - -> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (

l i i i i i i i i i i i i i s i i i i g p i i i i i i i i i i s i s i i i i s i i i i s s i i i i i i i i i i i i i i i i* o o o o o o o o e o o o o d o o o o o o o o o o o o d o < >oooooooooooooooooooe<

S%gS5gSSSggSg8ga5S=SSaSgSg5S§gSSSgg%3SSSSS8Sg3gggSSgS2aSSSgS:S5^pppppppppooppooooooooooooooooooooooooooooooooooooooooooooooooj | * * o o o o o o o o o d d o o d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d dS

B*0<»U»04IO*«OC-_,_ — - _ _o ^ * " T T * 0 T ^ * ^ * A * A « o u > r « « » « o < o « i A ( p « > r * r > i r « > v > f « * < 0 r « v « o t A < p i A « 0 « e « i A i o r > » < D < o < D r < » r « < D « ô « o « 4 o < à ( ô ù ) i A r < » i n i o û i w t r » r « n n) C ^ o o Q o d o o d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d

! ; 3 g 2 g 8 = s 3 s s g g g g g : s g g : g g 3 s ; : a : g g s a s : 5 2 5 t % g g 2 : : g : 3 g g g % : 3 = :l ^ o o o o o o o o o o o o o o d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d o c

IS3SSS> P» M CM <0 «0» d d d d d

C 2oor)ooOf»*no^«»-a»oa#2S3SSgSSS2g3ggg " d d d d d d d d d diggggs> o o o o o » d d d d d

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

Page 216: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

201

jSSSS

l i i i ip d d b d

s i i i i! d d d d

A d d d d

8 N 0 Ol M•> lA tn lo wj d d d d

i a s s sflSSSS

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

Page 217: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Appendix 6. Amphibole Compositions

202

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

Page 218: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

2 0 3

8ineno ^oh> «« (oooiA«aioiA««««fla«o<M«o«vu)<e(D(oaio«A<iaE5SZS2%S%SSZSSS;S5S%3SS55S:993:gS:;S%SSS3S^f MWN(MN(\IWN(WNNNC\i(\it\iWNWa#WMmmW(\iN(VW(y<WNWWN(y(WNW(

|U»»5S«S»«à5555SSoSS5S;f?oi = = Soo==^

^ MCMMClOIMMMCIMCIMMNNMCMCVCIciCMMMCioiMOiMMNcicjMNMNMOif»

§ # : ; | ; g 5 5 ! ; | 5 5 ; ; : ! ; : ; 3 3 3 s 3 s s : : ^

d ! : ! : ! ! : 3 5 : : s : ! 3 : 3 ! 3 : 5 # g : s ! ! : ! 3 : ! 3 3 ^

“• “ 3 i33S5i2SSS5SSSS2223SS3235i3S5i25S252i

S =0:2::2: : : : :22:!5: : :2ss225:23^

o < ? * o S « î 2 S î C : r 2 o ! « T « ? r « ? « î 3 2 § î g 5 S S 5 S S S S * Ç a « , ^ Ç « > c

I 5 3 [ t3 z s : z 2 2 2 2 : : s % = : 2 s%2s = s: : : = : ! :2 s:tt>w>«0<o«(>«o«04Ov>«>4O<0r«<o(p«e<D«o(O9a*ci0tta»c

3 33zz;:S!;ss:;$ss3::;ssS5;:::S8S!;i23;2:.3K$gg3%MMo i Mo j c j o i < >i c j c< i c i i Mc <* Ne i c j c i o i Mo i MC<j e < i c < i Mo j c i i c i Mo i M* ' * o i MMMo i o j K

3 55 : :

— " JgwgJOkW

* * ^ * S 5 î » S î t « w S o ‘* ! t o S S < 5 w S ^ c 5 * ^ S S n ^ t î S * S S ’ o « ‘* ï o 2 5 2 2 2 î 2 2 2 2 2 2 2 2 2 : î 2 ï £ 2 î ï Z = £2SSî2î2î2£î2î2-ïî2:-î2d

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

Page 219: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

2 0 4

i lE<on*>nn n«>n«o<ODr>4onn«on«0r)r>oono«D«>ai(D«D«r)O<Dnn«*«r>o

^ : : : : : : : : : : : : s : : 5 3 : : : s : : : : : : : : : : : : : : : : : 5

;S;;sRR:qmS = 5SS5ss%%s«:: :%s%:%aas ;% s3:a :g■oooooooodooeooooodoooooooooooooooooddoo

8 -e » «a §$21 = = ;:;: = = = = %:: =j ' ^dddddddddddddd

;SSS28Rg3S5S2S%S2S;%S3ooooeoooooec ooooooooooooo

Sgujçf rtof-nort®*!| S Î 5 S S ? Î Î 5 5 !

p*’ooooooeoed<

n<»ojtAor r>«c4«oszs23tg!;q3ooooooooo

«r»iA««40«ott>*«-«<oiA04r»«o3S3!999338!:!;538SS83geedodeooodooooddddo

l2om«r)n>*r»r«>oor»r«>mh>u>ov-«aiOa«oc<f».tnnnoroain nak«)e nr« 82222SGf25222< C33*^^*^MmM M*»o«oo**a**o»oo»m mmim ;xr>onor>nnon<<»nnnnc»ne«n«in«t«iort«t^o«»vin«t«««) o* «* « « « « E^ooooddddddddddddeddeddededddddddedddddd

IgSSKRRNKZKSKKKKRKRRSZZSSKKKgSKgZSgSKggZ:[ " ■ o o o o d d o d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d *

<«OM«sMr>ar»Qici«a«iDiA «>o0<ftMio« iA«MttO<9O<9oniD(0r«>h> •Ma cv0 a orou>r»«ioai r«>«aioiA«B«>vMN«*Mnv<vN«0aiair<»iAr '^oi0iaoooooooooooeooa»ai»9«oi^^^ci«ie49étieicicie<icic4t>it>ici^^ i^ôo tain aar»o*«r«avorou>r»«ioai«

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

Page 220: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

Appendix 7. Sphene Compositions

205

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

Page 221: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

2 0 6

Su>«o<Mnnr»<Meka«>tn«>a>«i(Oinu>OMOM(*>aoNiAr««-(no<o*o<o«04ûiA<o«AiAinuiMtnu><OM(n«ototBm4omtd<oin«0(Oin< - 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ^ 0 0 0 0 0 0 03 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

emoimmo — ovimNioooovoomoo — oimvof»—ON o^*^0v0^oakooooooooooooc400o00a»ai — OOOOOOOOOOOOOOOOOOOOOIAOOOOOOO— — #-ooo—'00^"^d^^*"*-^o*»*«^^oo

•—r«>a>NNinro.NonacM*ONC<JNr<>nr> — N<no«->toOo«u) 5Now> v>r<»iAinMo ^o« ^«otooon- r>mtnoooo 9wWV<dcDIA<l>ldlbu)tA(AtAW>inkOIAlA<DIAindlA«IA«WtAU) >-00000000000000000000000000000

AN — OmONOWOO— »OimOONf»«V"»00*-OOf»OV^ o o d d d d o d d d d d d o Q d d d d o d d d d d d d d d

8 !..................................................................j d d d d d d d d d d d d d d d d d d d d d d d d d d d d d

io^ONOMtAoooonienNniAN^^ONOvno—tnt» voN«o*nw»o*0r".m* #;- omrx-;- f*«pmov(wvoooy «otAwi«etoto<oiOiod(Otdd<oddu>dtdd<ô***dd(du>«oddOCSICJOCXCMOOOOMCMOCXCMCINOMONNOOJOOOCMN

»000000-*cij00»”0^*“»“»-0000"^»“-*00^0»“i O O O O O O O O Q O O O O O O O O e o O O O O O O O O O O• d d d d d d d d d d d d d d d d d d d d d d d d d d d d d

ppppooooopp«^oooo«^ooc>opoeoopoo% d d d d d d d d d d d d d d d o d d d d d - » d d d d d d d

SkAN«otAiAn iAnooMonooooowif«>«oifloo#)oeoooooooooooooeoooooooooooooooo

> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ^ 0 0 0 0 0 0 0

l S S g g : : S 2 S S S S ; S 2 2 S 5 S ; S 5 S S g S K

ooooooooOooo^oo^^— >^^oo^0 . O O O O O O O O O O O O O O O O O O O O O O O O O O O O O

î a s s s s i s g s s s s a s s a s ï g ï g î s g g g a s s R% o o o ^ # » d ^ o d d d d d d d d d d d d d d d d d d d d d

S * T * * * 2 2 S S m S S o S 2 2 • * S S p g • 2 S S» * S pP 3 Z2 : 2 0 3 gow; oou;w$w;iAww$u*0 'imwimoim*nim

22v(S?v2!n22oM3oow2amM2ZRo!SS2o3oas sa s ss s s i s s s s s i s s i s i ss s is i sagss is ss s

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

Page 222: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

2 0 7

— 50«tft^ r».5«t«» ooi»rt»rto» ou»cM«or»u»»o« VS95r^9<'*590"*99*>ooooiaooooode»aioi^oaaqooooooaaaaoooaaooaoooaaoo

2m**m«OT0?*AN«ooaa«*oamoo^*wvu»N^aN — aaaoaaoooaoooowaooaoaa 9'>°®9«>'>‘l<»<»°o!“. <=°°“ °°=><’ “>‘»o. o°ooi<»

8ioi N. <ar<»tnr««oh>«or>u>(D«r»acar»iAr«.o(o»«>0iou>aoooeeooooooooooeoooeoeoooooooo*>qqoqooooooooooooooooooooooooo^oodooododoooooooooooooooooooo

gSSSS32S23S23g33SSSS8SS238S8S8~qqqqqeqqqqqqqqqqqqqoqooeqooooJJddddddddddddddddddddddddddddd

iSSSSSSSffSgSSSSl^SSgSRgSRSSSSSg^«aqqaaqqaqaqaqaaqqaaar^aaaaaaaA o o o o o o o o o e d o o e d d o o d d d d d d d d d d d

go o o o o o o v o o o o o e o e o o o o o o o o e o o o o ooooeoooeoooooooooooooooooooo ^qqoqoooqooooooooooeoooooooooo i^eododddoddddddddddddddddddddd

qeooooooeooooooaoooooMooeooeooqoqooooooQooooooooooooooooooj o o d o d d d d d d d d d d d d d d d d d d d d e d d d d

;0®990P90<»99®900099«>«»®«*9®«?9090rooooooooooooooooooooooooooooo

I8 : § 8 0 S S 8 3 3 S S 3 S 8 3 S 3 8 3 3 S S S 8 S S S 3^^ddddddddddddddddddddddddddddd

iqqpqqoopqooooooooooooooooooooQ^oooooddoodddddddddddddddddddd

S s : 2 Sa2 S12S:SSSS2 2 SgSSSgSSS!SSS2•-qqqqooopoooooooooooooeooooooo^oooodddoddddddddddddddddddddd

Iggllissssssisssssgssslsllsgslp o o o o d d d o e d d d d d d d d d d d d d d d d d d d d

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

Page 223: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

APPENDIX 8. SAM PLE DESCRIPTION S AND LOCATIONS.

Sample NGBl-1 is a relatively fresh, fine-grained trachybasalt; dominant

phenocrysts include plagioclase (Angg) and titanaugite. Magnetite and biotite are present as

small phenocrysts and in the groundmass. The sample was prepared from well cuttings

taken at a depth of 5040' (1536m), within the Travis Peak Formation, from the British

American #3 Houston Estates well located in T11N-R7W-S15, on the east side of the

Monroe Uplift, in Sharkey County, MS.

Sample NGB3-1 is a relatively fresh, fine-grained trachyte; plagioclase phenocrysts

dominate the rock. Minor biotite is the only ferromagnesian mineral. The sample was

prepared from core material from a depth of 4365' (1330m), within Lower Cretaceous

rocks, from the Murphy-Sun, Mayhall #1 well located in T16N-R7W-S21, in Washington

County, MS.

Sample NGB7-1 is relatively fresh, medium to fine-grained, holocrystalline,

microsyenite; plagioclase dominates the sample. Acicular clinopyroxene, minor biotite,

sphene, and apatite are also present. The sample was prepared from core material from a

depth o f 4024' (1227m), within the Cotton Valley Formation, from the Gulf Oil Company

#1 Hamilton well located in T5N-R2E-S4, on the Jackson Dome, in Rankin County, MS.

Samples P-1 (3180') and P-1 (3530') are relatively fresh basaltic rocks. The

samples were prepared from well cuttings taken at depths of 969m and 1076m respectively,

from the Parham Kathan Johnson #1 well located in T18N-R10E-S35, on the Monroe

Uplift, in Madison Parish, LA.

208

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

Page 224: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

VTTA

Christina Lee Livesey was bora to Robert and Margaret Livesey, in Elgin Illinois,

on June 12,1961. She graduated from Oak Park River Forest High School in Oak Park,

Illinois, in 1979. She received a B.S. with Honors, in Geology, from the University of

Iowa in 1983, and an M.S. in Geology from the University of Kansas in 1988. She

currently resides in Houston, Texas with her husband, Steve, and her two dogs. Spot and

Othello. She is currently employed by Chevron Oil Company as an exploration geologist.

2 0 9

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

Page 225: Petrogenesis and Provenance of Epiclastic Volcanic Cobbles

DOCTORAL EXAMINATION AND DISSERTATION REPORT

Candidate: C hristina Lee L ivesey

Major Field: Geology

Title of Dissenation: P etrogenesis and Provenance o f E p ic la st ic Volcanic Cobbles from the Cretaceous Woodbine Formation, SW Arkansas

Approved:

____M a jo r f jp ro fe s s ^ an d C n a lrm a

%D e an of th e G ra d u a te School

EXAMINING COMMITTEE:

Date of Examination:

February 1, 1990

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