Petrogenesis and Provenance of Epiclastic Volcanic Cobbles ...
Transcript of 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
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.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
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.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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
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.
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.
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.
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.
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.
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.
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.
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).
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
3 6
IS
iù
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.
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.
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.
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.
4 0
4000
3000 -
I«
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.
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.
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.
■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
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.
■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
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.
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.
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.
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.
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.
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>
a§
IîII1coden
I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
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
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.
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.
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.
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.
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.
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
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
.«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.
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.
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.
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.
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
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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
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.
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.
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.
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
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.
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.
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.
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.
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
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
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.
A ppendix 1 . C linopyroxene Com positions
177
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
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.
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.
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.
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.
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.
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.
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.
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.
Appendix 2. Intermediate Feldspar Compositions
186
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
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.
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.
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.
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.
Appendix 3. Felslc Feldspar Compositions
191
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
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.
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.
Appendix 4. Pyroclastic Feldspar Compositions
194
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
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.
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.
Appendix 5. Detrital Sanidine Compositions
197
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
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.
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.
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.
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.
Appendix 6. Amphibole Compositions
202
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
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.
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.
Appendix 7. Sphene Compositions
205
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
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.
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.
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.
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.
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.