<%BANNER%>

Geology and Petrogenesis of Lavas From An Overlapping Spreading Center

Permanent Link: http://ufdc.ufl.edu/UFE0041974/00001

Material Information

Title: Geology and Petrogenesis of Lavas From An Overlapping Spreading Center 9 degrees North East Pacific Rise
Physical Description: 1 online resource (181 p.)
Language: english
Creator: Wanless, Virginia
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: assimilation, basalt, dacite, mid, morb, ocean, ridge, submarine
Geological Sciences -- Dissertations, Academic -- UF
Genre: Geology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In contrast to relatively homogeneous mid-ocean ridge basalt (MORB) compositions typically erupted on fast-spreading oceanic ridges, a wide range of rock types from basalts to dacites have been recovered at overlapping spreading centers (OSC). This study focuses on the petrogenesis of lavas erupted at the 9degreeN OSC on the East Pacific Rise in order to better understand the complex magmatic plumbing system beneath a ridge discontinuity. Lavas that span the entire compositional range observed on the global mid-ocean ridge (MOR) system, including basalts, ferrobasalts, FeTi basalts, basaltic andesites, andesites and dacites have erupted along the eastern, propagating limb of the OSC. Major and trace element analyses, radiogenic (Pb, Sr, Nd) and oxygen isotopic ratios, volatile contents (Cl, H2O, CO2) and geochemical modeling are used to determine the petrogenesis of MORB and genetically related high-silica magmas. The formation of high-silica dacites on MOR remains a petrologic enigma despite eruption on several different ridges. They are characterized by elevated U, Th, Zr, and Hf; relatively low Nb and Ta; and Al2O3 and K2O concentrations that are higher than expected from fractional crystallization. Additionally, high Cl and H2O concentrations and relatively low ?18O values in dacitic glasses require contamination from a seawater-altered component. Extensive petrologic modeling of MOR dacites suggests that fractional crystallization of a MORB parent combined with partial melting and assimilation of altered ocean crust can generate magmas with geochemical signatures consistent with MOR dacites. This suggests that crustal assimilation is a much more important process on ridges than previously thought and may be significant in the generation of evolved MORB in general. Petrologic models indicate that ferrobasalts and FeTi basalts erupting at the OSC can be explained by low-pressure fractional crystallization of a primitive MORB parent; however, both fractional crystallization and magma mixing produce intermediate compositions. Geochemical analyses suggest that there are two distinct populations of andesites erupted at the OSC. Andesites with high-P2O5 are the most evolved MOR compositions produced through fractional crystallization. In contrast, low-P2O5 andesites and basaltic andesites appear to have formed primarily through mixing of ferrobasaltic and dacitic magmas.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Virginia Wanless.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Perfit, Michael R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-02-28

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041974:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041974/00001

Material Information

Title: Geology and Petrogenesis of Lavas From An Overlapping Spreading Center 9 degrees North East Pacific Rise
Physical Description: 1 online resource (181 p.)
Language: english
Creator: Wanless, Virginia
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: assimilation, basalt, dacite, mid, morb, ocean, ridge, submarine
Geological Sciences -- Dissertations, Academic -- UF
Genre: Geology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In contrast to relatively homogeneous mid-ocean ridge basalt (MORB) compositions typically erupted on fast-spreading oceanic ridges, a wide range of rock types from basalts to dacites have been recovered at overlapping spreading centers (OSC). This study focuses on the petrogenesis of lavas erupted at the 9degreeN OSC on the East Pacific Rise in order to better understand the complex magmatic plumbing system beneath a ridge discontinuity. Lavas that span the entire compositional range observed on the global mid-ocean ridge (MOR) system, including basalts, ferrobasalts, FeTi basalts, basaltic andesites, andesites and dacites have erupted along the eastern, propagating limb of the OSC. Major and trace element analyses, radiogenic (Pb, Sr, Nd) and oxygen isotopic ratios, volatile contents (Cl, H2O, CO2) and geochemical modeling are used to determine the petrogenesis of MORB and genetically related high-silica magmas. The formation of high-silica dacites on MOR remains a petrologic enigma despite eruption on several different ridges. They are characterized by elevated U, Th, Zr, and Hf; relatively low Nb and Ta; and Al2O3 and K2O concentrations that are higher than expected from fractional crystallization. Additionally, high Cl and H2O concentrations and relatively low ?18O values in dacitic glasses require contamination from a seawater-altered component. Extensive petrologic modeling of MOR dacites suggests that fractional crystallization of a MORB parent combined with partial melting and assimilation of altered ocean crust can generate magmas with geochemical signatures consistent with MOR dacites. This suggests that crustal assimilation is a much more important process on ridges than previously thought and may be significant in the generation of evolved MORB in general. Petrologic models indicate that ferrobasalts and FeTi basalts erupting at the OSC can be explained by low-pressure fractional crystallization of a primitive MORB parent; however, both fractional crystallization and magma mixing produce intermediate compositions. Geochemical analyses suggest that there are two distinct populations of andesites erupted at the OSC. Andesites with high-P2O5 are the most evolved MOR compositions produced through fractional crystallization. In contrast, low-P2O5 andesites and basaltic andesites appear to have formed primarily through mixing of ferrobasaltic and dacitic magmas.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Virginia Wanless.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Perfit, Michael R.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-02-28

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041974:00001


This item has the following downloads:


Full Text





GEOLOGY AND PETROGENESIS OF LAVAS FROM AN OVERLAPPING
SPREADING CENTER: 9N EAST PACIFIC RISE




















By

V. DORSEY WANLESS


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2010









3-4 CI/K20 versus H20/K20 for glasses from the 9N OSC ................. .......... 99

3-5 180 versus MgO for glasses from the 9N OSC.............. ........... ............ 100

3-6 C02-H20 vapor saturation diagram ................. ....................................... 101

4-1 Bathymetric map of the northern EPR..... ................... ..... ............ 146

4-2 Bathymetric map of the 9N OSC with the location of rock samples ................. 147

4-3 Bathymetric map of the 9N OSC with the melt sills....... .......................... 148

4-4 Side scan sonar mosaic from data collected on the MEDUSA2007 cruise....... 149

4-5 FeO versus MgO for glasses collected from the east limb of the 9N OSC ...... 150

4-6 Major element variations versus MgO (wt%) for glasses collected from the
east limb of the 9N OSC ........ .................... .......... ...... ..................... 151

4-7 P205/TiO2 versus MgO (wt%) for east limb glasses ..................... ................ 152

4-8 Trace element concentrations versus Zr for glasses.................... ................ 153

4-9 Incompatible trace element ratios versus Zr for glasses erupted at the OSC. 154

4-10 Radiogenic isotope ratios showing the variation in sources............................... 155

4-11 Major element concentrations and ratios versus MgO comparing the east and
west lim b of the O SC ........... .. ................................ ................ 157

4-12 Trace element concentrations versus Zr comparing east and west limb........... 159

4-13 Primitive mantle normalized diagram showing variations in andesites and
basaltic andesites erupted at the OSC. ............. ......................................... 160


















V






4,
E f*'l


*', ',

'0

assimilation of altered
material


0 1 2 3 4 5
MgO (wt%)


* Basalts
A Bas. And.
*Andesite
* Dacites
o JdF data
A GSC data

A


A
A 0
Mantle d018

0 o
0


6 7 8 9


Figure 3-5. 6180 versus MgO for glasses from the 9N OSC. The MOR dacites,
andesites and several basaltic andesites lie below calculated fractional
crystallization trends (black dashed line). The lower 6180 values are
consistent with assimilation of altered oceanic crust, which has lower 6180
values due to high-temperature hydrothermal circulation. 6180 values for
lavas from the Juan de Fuca ridge and Galapagos Spreading Center are
shown for comparison.































100


7

































Figure 3-6. H20 versus C02 for 9N OSC glasses. Superimposed on this diagram are
C02-H20 vapor saturation curves for 200 to 800 bars based on models by
Dixon et al. (1995a, b). Black dashed lines show a general magma
degassing trends during ascent. The gray band represents an approximate
depth of the top of the imaged melt lens (Kent et al., 2000). The pressure at
the seafloor is shown as a dotted line. Most of the basalts are in equilibrium
with pressures consistent with the top of the imaged melt lens (~550 bars).
The dacites, andesites, and two basaltic andesites have completely degassed
C02 and may have also lost H20 prior to or during eruption. H20 and CO2
concentrations in one basaltic andesite lie between the high-silica lavas and
the basaltic lavas, which may indicate mixing (red dashed line).


basalts
250 '' dacites
&A bas.And.
200. *seafloor
S200 melt lens
150 mixing
- 5 degassing



so ,9rs

0
0 0.5 1 1.5 2 2.5
HZO (wt%)









CHAPTER 4
CRUSTAL DIFFERENTIATION AND SOURCE VARIATIONS AT THE 9N
OVERLAPPING SPREADING CENTER; EAST PACIFIC RISE

Introduction

Mid-ocean ridges (MOR) can be divided into a series of segments or

discontinuities that range in length from tens of meters to hundreds of kilometers

(Sempere & Macdonald, 1986; Macdonald et al., 1988). Overlapping spreading centers

(OSC) are second-order discontinuities that form between widely spaced first-order

transform faults on fast to intermediate spreading ridges (Macdonald & Fox, 1983;

Sempere & Macdonald, 1986; Carbotte & Macdonald, 1992). Both first and second

order discontinuities delineate physical and geochemical segmentation of the ridge that

reflect sub-ridge processes, such as variations in degrees of mantle melting and/or

separate crustal magma reservoirs (e.g. Macdonald et al., 1988).

Lavas erupted along fast to intermediate spreading centers, such as the northern

East Pacific Rise (EPR), may produce a range of basaltic lavas (e.g. Batiza & Niu,

1992), but they only rarely erupt compositions with MgO concentrations <5 wt%. This

relatively limited compositional diversity compared to other tectonic settings is

commonly attributed to shallow-level fractional crystallization of primitive magmas within

an axial magma chamber that is buffered by relatively frequent recharge with more

primitive mantle melts (Klein, 2005). Additionally, geochemical variations in mid-ocean

ridge basalt (MORB) may result from variable mantle melting parameters and/or mantle

sources (Klein & Langmuir, 1987; Langmuir et al., 1992).

Lavas erupted at ridge segment ends, such as OSC's, can have a broad range of

compositions compared to the relatively limited basaltic compositions erupted from

magmatically robust segment centers (Christie & Sinton, 1981; Langmuir et al., 1986;


102









Wanless et al., accepted). This variability can be attributed to lower magma supply and

cooler crust at the end of ridge segments (cold edge effect), which cause increased

magmatic fractionation prior to eruption (Christie & Sinton, 1981; Perfit et al., 1983;

Sinton et al., 1983; Perfit & Chadwick, 1998; Rubin & Sinton, 2007). Although crystal

fractionation is undoubtedly a primary process in magma differentiation at MOR, recent

geochemical studies show that highly evolved incompatible trace element

concentrations (Wanless et al., accepted) and low oxygen isotope ratios (Wanless et al.,

submitted) in MOR dacites require partial melting and assimilation of oceanic crust. The

extent to which these processes contribute to the chemistry of more mafic magmas on

MOR remains poorly constrained.

Lavas erupted at segment ends also preserve geochemical signatures that can be

ascribed to mantle source variations. A greater proportion of enriched mid-ocean ridge

basalt (E-MORB) erupted at ridge segment ends compared to segment centers on the

northern East Pacific Rise (EPR) may represent decrease in the amount of melt feeding

the ridge axis in these regions (Christie & Sinton, 1981; Sinton et al., 1983; Langmuir et

al., 1986). Despite evidence for variations in mantle source between segments, the

spatial distribution of E-MORB lavas at OSC is not well constrained.

The present study combines geophysical, geochemical, and bathymetric data to

examine the magmatic plumbing system at 90N OSC on the EPR (Figure 4-1). We use

major and trace element data and isotopic ratios to explore the relative roles of crystal

fractionation, assimilation and magma mixing within the shallow crust beneath the 9N

OSC. Trace element data and isotopic ratios are used to examine how variations in

mantle sources and magma supply can contribute to the distribution of compositions


103









observed on both limbs of the OSC. These analyses are compared to compositions of

lavas erupted from segment centers to the north (9-10N) and south (8037'N) of the

OSC to explore the extent of these variations beneath the EPR.

Background, Tectonic Setting and Geology of the 90N OSC

Overlapping Spreading Centers

At OSCs the ridge axis splits into two overlapping, curvilinear, sub-parallel axes or

"limbs" that may offset the ridge by up to 15 km (Macdonald & Fox, 1983; Macdonald et

al., 1988). The ratio of offset width to overlap length is approximately 1:3 (Macdonald et

al, 1988) and the inward curving limbs surround an elongate basin (Macdonald & Fox,

1983). The limbs migrate sub-parallel to the overall ridge strike with one limb

propagating and the other dying (Hey et al, 1980; Sinton etal., 1983; Pollard & Sydin,

1984). Consequently, the propagating limb migrates into older and colder ocean crust

and the receding limb gradually becomes amagmatic. As the OSC migrates with time, it

produces offsets in bathymetry and magnetic signature of the ocean crust (Hey et al.,

1977; Carbotte & Macdonald, 1992). Average OSC migration rates can be calculated

using the fossil V-shaped bathymetric scars developed in the wake of the propagating

OSC that are left either by linking of one ridge axis with the other during propagation,

which leads to decapitation of the ridge tip, or by repeated self-decapitation along a

single limb of the OSC (Macdonald et al., 1988).

Tectonic Setting and Previous Studies of 90N OSC

The 9N OSC is located between the Clipperton and Siqueiros transform faults

(Figure 4-1) and is one of eight 2nd order discontinuities on the northern EPR

(Macdonald & Fox, 1983). It consists of two north-south trending ridges that overlap by

-27 km and offset the ridge by ~8 km (Sempere & Macdonald, 1986). The 9N OSC is


104









divided into three main sections: the eastern propagating limb, the western receding

limb, and an overlap basin separating the two limbs (Figure 4-2). The eastern limb can

be further divided into the east limb ridge, the east limb tip and the northern inter-limb

region, also called the northwest flank (Figure 4-2), following nomenclature in Nunnery

et al., (2008). Based on geologic, magnetic and bathymetric data, the eastern limb has

been propagating south at a rate of ~42 km/Myr, the western limb has receded

(Macdonald & Fox, 1988; Carbotte & Macdonald, 1992).

The 9N OSC is one of the largest and most extensively studied 2nd order

discontinuities on the global MOR system. It has been the focus of several geophysical

studies (Detrick et al., 1987; Harding et al., 1993; Kent et al., 1993; Kent et al., 2000;

Bazin et al., 2001; Dunn et al., 2001; Tong et al., 2002), including the first multi-channel

seismic 3-D survey of a MOR (Kent et al., 2000) and a 3-D seismic refraction

study(Dunn et al., 2001). Collectively, these studies reveal the presence of a shallow

melt lens beneath each of the limbs and a widening of the eastern lens below the inter-

limb region north of the overlap basin (Figure 4-3; Kent et al., 2000). The melt lens

beneath the western, receding limb is narrow and shows no discernable variation in

depth along axis, but the melt lens beneath the eastern, propagating limb is variable in

width and depth (Kent et al., 2000). Beneath the southern portion of the east limb the

melt lens is narrower and deeper than the rest of the eastern ridge axis, plunging ~500

m over ~6 km (Kent et al., 2000). It also cuts across the tectonic seafloor fabric (White

et al., 2009). To the north, the melt lens widens to more than 4 km (Figure 4-3) and is

displaced slightly off axis to the west into the inter-limb region (Kent et al., 2000; Tong et

al., 2002). Tomographic studies reveal a low velocity zone beneath the entire OSC at ~


105









9 km depth near the mantle-crust transition (Toomey et al, 2007). This zone extends

~8 km to either side of the ridge axis (Dunn et al, 2001).

Recently, the 9N OSC was the focus of the MEDUSA2007 research cruise

(AT15-17), which carried out detailed mapping and extensive sampling of the region

using the ROV Jason2, DSL-120A side-scan system, and the WHOI TowCam (Fornari,

2003). This expedition acquired >10,000 photographs of the seafloor in combination

with the most complete and well-constrained lava sampling of any OSC (White et al.,

2009; Wanless et al., accepted). Prior to this study, limited lava sampling of the region

during the CHEPR cruise indicated that both high-silica and E-MORB lavas existed in

addition to N-MORB lavas (Langmuir et al., 1986).

9N OSC Geology

The northern portion of the eastern ridge is characterized by an axial summit

trough (AST) ~0.9 km across and ~50 m high (Figure 4-4). The AST walls gradually

diminish in height to the south, eventually disappearing by ~9o05' N. This topographic

change is accompanied by a gradual shift from a volcanically dominated seafloor fabric

in the north to a highly faulted and tectonized fabric observed at the southern tip that

also correlates with narrowing of the imaged melt lens, from ~4 km wide to <1 km

(Figure 4-3), and a near-absence of high-silica lavas (Figure 4-2).

Video, still photographs, and side scan collected during the 2007 cruise suggest

that over 80% of the lavas erupted within the mapped region are pillow lavas (White et

al., 2009), a morphology that is observed across all regions of the OSC, but dominates

the seafloor in the inter-limb region and overlap basin. In comparison, >80% of lavas

erupted on the EPR north of the OSC are lobate and sheet flows, and <20% are pillows

(Kurras et al., 2000; White et al., 2002; Soule et al., 2005; Fundis et al., 2010). Pillow


106









flow morphology at the OSC appears to be controlled by low effusion rates (<0.1 m3/s

at a viscosity of 100 Pa s; White et al, 2009). The youngest lavas on the northern

portion of the eastern limb, based on glassy surfaces, lack of sediment cover, and well

developed ornamentation, are confined to the neo-volcanic zone (Nunnery et al., 2008),

a narrow region of focused magmatism where zero-age lavas erupt (Perfit & Chadwick,

1998). On the southern tip of the eastern limb, the neo-volcanic zone is ill-defined;

however, the youngest lavas are primarily located along the western margin of the

eastern limb (Nunnery et al., 2008). We now describe the geology of each region of the

OSC in detail.

Despite the overwhelming abundance of pillow lavas formed at the OSC, lavas

erupted on the east limb ridge are morphologically quite diverse (sheet, hackly and

lobate flows, and small and large pillow lavas) and appear to correlate, at least to a first

degree, with composition. High-silica lavas erupted within the AST form a ~10 m high,

linear, pillow mound composed of atypically large pillow lavas. This pillow mound is

surrounded primarily by lobate lavas and to a lesser degree, sheet flows. Several large

areas of collapse, with drainback features, were observed within the AST and are

associated with basaltic sheet flows that surround the high-silica pillow mounds. The

lobate and sheet lavas are predominantly ferrobasaltic in composition; however, several

FeTi basalts were also recovered within the AST. Basaltic andesites also erupted within

the AST and form lobate flows and pillow mounds. Lavas sampled from both sides of

the AST walls at the edges of the ridge axis are primarily andesites and dacites and

form large, elongate to bulbous pillows. The only active hydrothermal vent at the OSC

was observed within the AST on the east limb axis within an andesitic to dacitic pillow









mound. A faulted and tectonized fabric dominates the seafloor on the southern tip of

the eastern limb. In places it is covered by elongate fresh pillow mounds that are cut by

fabric-parallel fissures. Lobate flows are sparse in this region.

Lavas erupted over the northern portion of the overlap basin and inter-limb region

primarily consist of pillow lavas. A linear pillow mound, trending approximately N-S,

defines the outer edge of this region and lies over the westernmost extents of the wide

melt lens (Kent et al., 2000). Lavas comprising this mound are younger than expected

(based on thin sediment cover and the presence of glassy buds) for their distance from

the neo-volcanic zone, suggesting that this region is the site of off-axis volcanism

(Nunnery et al., 2008; White et al., 2009). This conclusion is supported by excess 230Th

measured in several of these samples, which indicate eruption ages of <8,000 ka

(Waters, pers comm.). The southern overlap basin is primarily composed of bulbous

pillow mound fields, based on backscatter images (White et al., 2009), but photographic

and sample coverage is sparse.

The western limb of the OSC differs from the eastern limb in being primarily

comprised of sheet and lobate flows, with fewer pillow lavas (White et al., 2009) and

extensive areas of lava collapse and pillars with drain back features. Similar features

are common within AST's on many other regions of the northern EPR. Extinct vent

fields were also observed, but no active hydrothermal venting.

8037' N EPR Deval

The 8037'N deviation in axial linearity (deval) is located south of the 9N OSC and

approximately 40 km north of the Siqueiros transform fault (Figure 4-1; Langmuir, 1986).

It is the southern extension of the western limb of the 9N OSC. The 8037'N deval is

structurally similar to a small OSC and, therefore, may have a similar magmatic


108









plumbing system. This deval was dredged during the CHEPR cruise, which recovered

several E-MORB lavas and a high-silica andesite (Langmuir et al., 1986). Additional

investigations of the area were conducted during a response cruise in 2003 (Zierenberg,

2007 pers comm.) to determine if seismic events recorded on hydrophones in March of

2001 were indicators of new volcanic activity (Bohnenstiehl et al., 2003). This cruise

included several ALVIN dives; however, there was no evidence of a recent eruption.

Lava morphology in the region includes sheet flows, lobates and pillow lavas. During

these dives nine samples were recovered, including several glassy E-MORBS and one

glassy andesite. Data from these samples are discussed below.

Geochemical Methods

Over 280 rock samples were collected from the 9N OSC during the

MEDUSA2007 cruise. Of these, 275 have glassy outer rims, from which glass was

handpicked and analyzed on a JOEL 8900 electron microprobe for major and minor

element concentrations at the USGS facility in Denver, CO. Eight to ten points were

analyzed per sample. The probe diameter was routinely 20 pm, to minimize sodium

loss, with an accelerating voltage of 15 keV and a beam current of 20nA. Several

USGS minerals were used as calibration standards and secondary normalizations

involved the JdF-D2 glass "standard" (Reynolds, 1995), ALV 2392-9 (in-house

standard), and dacite glass GSC (USGS standard) to account for instrumental drift

(Smith et al., 2001). Chlorine and sulfur concentrations, as well as high-precision

potassium values were also determined on nine samples, using 200-second peak/100

second background counting times. Major element concentrations for samples from the

837'N deval on the EPR were determined by microprobe at UC Davis following

methods described in Schiffman et al., (2010).


109









LIST OF ABBREVIATIONS
AFC assimilation fractional crystallization

Cpx clinopyroxene

EPR East Pacific Rise

GSC Galapagos Spreading Center

HREE heavy rare earth elements

Ilm Ilmenite

JdFR Juan de Fuca Ridge

LLD liquid line of descent

MOR mid-ocean ridge

MORB mid-ocean ridge basalt

N-MORB normal mid-ocean ridge basalt

ODP Ocean Drilling Program

01 olivine

OSC overlapping spreading center

Plag plagioclase

QFM quartz-fayalite-magnetite

REE rare earth elements

ROV remotely operated vehicle

Sp spinel









A representative subset of fresh glasses were handpicked, cleaned in a dilute

acid, and dissolved for trace element and isotope analyses following methods described

in Goss et al., (2010). 73 samples from the 9N OSC and seven samples from the 8037

N deval were analyzed for high precision trace elements on an Element2 Inductively

Coupled Plasma Mass Spectrometer (ICP-MS) at the University of Florida. External

calibration was done to quantify results using a combination of internal (ENDV -

Endeavour and ALV 2392-9) and USGS (AGV-1, BIR-1, BHVO-1, BCR-2 and STM-1)

rock standards. High precision Pb, Sr, and Nd isotopic abundances on 37 samples were

determined using the Nu- Plasma multi-collector ICP-MS at the University of Florida

(Wanless et al., accepted). For detailed descriptions of sample preparation, dissolution

procedures, standards, and statistical data, see Goss et al., (2010) and Kamenov et al.,

(2007).

Geochemical Results

Lavas erupted at the 9N OSC display a large range of compositions on both the

east and west limb (Figure 4-2). Below, we discuss the geochemical variability on each

limb by rock type to better address the petrogenesis of the OSC lavas.

Geochemistry of East Limb Lavas

The east limb of the 9N OSC has produced basalts, ferrobasalts, FeTi basalts,

basaltic andesites, low-P205 and high-P205 andesites, and dacites (Figures 4-2, 4-5).

These lavas cover a wide compositional range and we, therefore, subdivide our results

below by rock type. Major and trace element data from the east limb are present in

Table 4-1 and 4-2, respectively. Radiogenic isotope ratios for the east limb lavas are

presented in Table 2-2.


110









Basalts

Basalts consists of all lavas with <52 wt% SiO2 and include ferrobasalts and FeTi

basalts. In comparison to more mafic basalts from the heavily sampled 9050 bulls-eye

site on the northern EPR, basalts from the east limb of the OSC are more evolved on

average (Table 4-1), with MgO concentrations ranging from 6.16 to 7.36 wt%, high FeO

(10.83 to 13.50 wt%), and TiO2(1.70 to 2.83 wt%) contents (Figures 4-5, 4-6). All lavas

are N-type MORB with variable K20 (0.11 to 0.27 wt%), P205 (0.12 to 0.27 wt%) and Cl

concentrations (0.004 to 0.05 wt%). P205/TiO2 ratios range from 0.08 to 0.13 (Figure 4-

7) and CI/K20 ratios from 0.03 to 0.29 and are thus comparable to other MOR

ferrobasalts (e.g. Michael & Cornell, 1998). Incompatible trace element concentrations

are relatively high compared to MORB from the northern EPR but ratios are relatively

constant (Figures 4-8, 4-9) despite eruption over wide geographic region at the OSC

and are comparable to other ferrobasalts erupted on MOR. Rare earth elements (REE)

patterns are remarkably similar in the ferrobasalts, with an average LaN/YbN ratio of 0.77

+ 0.05 (Figure 4-9). High field strength element (HFSE) ratios are also relatively limited

with Zr/Nb of 40 2.8, and U/Nb of 0.03 0.001. Compatible trace element

concentrations in the ferrobasalts are variable (Cr = 9 to 166 ppm; Ni = 32 to 71 ppm),

but show positive correlations with MgO.

Basaltic andesites/Iow-P205 andesites

Basaltic andesites (SiO2 >52 and <57 wt%SiO2) and low-P205 andesites from the

east limb have highly variable major and trace element compositions (Table 4-1). MgO

in the basaltic andesites ranges from 1.5 to 6.5 wt% and FeO ranges from 8.27 to 13.64

wt% (Figure 4-5). Minor elements are also highly variable (Figure 4-6; P205 = 0.18 to

0.58 wt%; K20 = 0.19 to 0.99 wt%; and Cl = 0.04 to 0.5 wt%). P205/TiO2 (average of









0.15 0.04) and P205/K20 (average of 0.79 0.28) ratios span the range between the

high-silica and ferrobasalt end-members (Figure 4-7). Ni and Cr concentrations ranging

from 6 to 50 ppm and 5 to 117 ppm, respectively (Figure 4-8) are generally, though not

exclusively, lower than in the ferrobasalts. REE and HFSE concentrations are variable,

but have relatively constant ratios with average LaN/YbN ratios = 1.08 0.12, Zr/Nb = 47

8.3, and U/Nb ratios20.03 (Figure 4 -9). The basaltic andesites lie along well-defined

trends toward high-silica lavas.

High-P20s andesites

Low MgO and high FeO, TiO2, P205, and K20 characterize the high-P205

andesites and distinguish them from low-P205 andesites discussed above. P205 ranges

from 0.54 to 0.78 wt%, which is up to 0.49 wt% greater than the average low-P205

andesite (Figure 4-6). Consequently, these lavas have higher P205/TiO2 (0.29 to 0.39)

and P205/K20 (0.86 to 1.47) ratios compared to other lavas on the east limb (Figure 4-

7). They have high average Cl concentrations of 0.3 ppm (Figure 4-6) compared to

MOR basalts. Most incompatible trace element concentrations are high in these lavas

compared to ferrobasalts and basaltic andesites, but similar to the low-P205 andesites

(Figure 4-8). The high-P205 andesites have an average Zr ~ 700 ppm, La ~22 ppm, and

Yb -15 ppm with LaN/YbN ratios of 1.15 0.11, and Zr/Nb ratios of 47 3.5, (Figure 4-

9). Distinctive features are the lack of negative Nb and Ta anomalies and slightly lower

U/Nb ratios (<0.03) when compared to the low-P205 andesites and some basaltic

andesites.

Dacites

The geochemistry of the MOR dacites is discussed in detail in Wanless et al.,

[accepted] and are, therefore, only briefly discussed here. They have SiO2


112









concentrations up to 67.5 wt% and MgO as low as 0.67 wt%. P205 concentrations range

from 0.15 to 0.25 wt% (Figure 4-6). K20 and Cl concentrations average 1.17 and 0.63

wt% respectively. P205/TiO2 and P205/K20 are also low (Figure 4-7). Highly

incompatible trace element concentrations are elevated, with average Ba, U, and La

concentrations of 62 ppm, 0.8 ppm and 28 ppm respectively (Figure 4-8). They also

have high Zr (734 ppm to 1050 ppm) and Hf (18.6 ppm to 24.9 ppm) concentrations, but

low Nb (13 ppm to 16.5 ppm), Ta (0.8 ppm to 1.4 ppm) and U/Nb ratios >0.03 (Figure 4-

9). The dacites have average LaN/YbN ratios of 1.26 0.10, Zr/Nb = 58 4.5, and U/Nb

ratios 0.04.

Isotopic compositions of East Limb lavas

In contrast to the variability noted in the major and trace element compositions of

east limb, the radiogenic isotopic ratios are relatively uniform. They have 208Pb/204Pb

and 206Pb/204Pb ranging from 37.642 to 37.699 and 18.235 to 18.294 respectively

(Figure 4-10 Oa). 87Sr/86Sr ranges from 0.70243 to 0.70258 and 143Nd/144Nd ratios range

from 0.51314 to 0.513120 (Figure 4-10b), which is similar to lavas erupted on the EPR

to the north near 9050'N (Sims et al., 2002; Sims et al., 2003; Wanless et al., accepted).

West Limb Lavas

Fifty lava samples (Table 4-3, 4-4) were collected from the west limb of the 9N

OSC of which 47 are basaltic, 2 are basaltic andesites and one is andesitic in

composition (Figure 4-11 a). Fourty-five samples have typical N-MORB K/Ti ratios (K/Ti

= K20/TiO2*100 <1.3; Langmuir et al., 1986; Sinton et al., 1991; Reynolds et al., 1992;

Perfit et al., 1994) and five have higher ratios typical of E-MORB (K/Ti>0.13; Figure 4-

11 b). Compared to the east limb basalts, the average west limb basalt is slightly more

primitive, with MgO concentrations ranging from 6.42 to 8.72 wt% (Figure 4-11 a) and


113









K20 and P205 concentrations from 0.07 to 0.17 wt% and 0.13 to 0.26 wt%, respectively

(Figure 4-11a). Lavas dredged in this region during the CHEPR cruise had similar

compositional variations, including E- and N-MORB (Langmuir et al., 1986). West limb

N-MORB lavas have LaN/YbN ratios of 0.72 0.04 and Zr/Nb of 37 to 45 (Figure 4-12).

Highly incompatible elements and HFSE have limited ranges in concentration, for

instance, Ba ranges from 3.4 ppm to 9.2 ppm, U from 0.04 ppm to 0.07 ppm, and Nb

from 1.55 ppm to 3.22 ppm (Figure 4-12). The west limb E-MORB lava has LaN/YbN

ratios of 2.8 and Zr/Nb ratios of 9.7. The single andesite has major element

concentrations (Figure 4-11 a) and trace element abundances (Figure 4-13) similar to

the high-P205 andesites on the east limb (Figure 4-11 a). It has a SiO2 concentration of

55.95 wt% and MgO of 2.84 wt%, with high P205 (0.73 wt%) and K20 (0.67 wt%).

Isotopically, the west limb lavas have slightly more radiogenic values compared to

the east limb lavas (Figure 4-10; Table 4-5). The N-MORB lavas have 208Pb/204Pb and
206pb/204pb ratios ranging from 37.737 to 37.846 and 18.253 to 18.369, while the E-

MORB lava has higher lead isotope values (208Pb/204Pb = 38.022 and 206Pb/204Pb =

18.590) (Figure 4-10a). N-MORB 87Sr/86Sr values ranges from 0.70249 to 0.70265 with

an average of 0.70256 (Figure 4-10b) whereas the E-MORB 87Sr/86Sr ratio is also more

radiogenic (0.70282). 143Nd/144Nd ratios of the N-MORB lava range from 0.51314 to

0.51315, while the E-MORB lava has a ratio of 0.51305. The andesite has isotopic

ratios intermediate between the west limb E-MORB and N-MORB values.

8037' EPR Lavas

Nine samples collected from the 8037 N deval range in composition from basalt to

andesite, with MgO and SiO2 concentrations from 7.29 to 3.08 wt % and 50.47 to 57.14

wt%, respectively (Table 4-6). P205 concentrations range from 0.3 to 0.5 wt.%. Two of


114









the samples are E-MORB with K/Ti values of 0.19 and 0.15. REE and HFSE

abundances are variable with average LaN/YbN ratios of 1.33 + 0.17, Zr/Nb of 20.92 to

31.84, and U/Nb of 0.02 to 0.04.

Radiogenic isotope ratios were measured in seven samples collected from 8037'N

(Table 4-7). Compared to basalts collected from the 9050N region of the EPR (Sims et

al., 2002; Sims et al., 2003) and the 9N OSC, these samples have, on average, more

radiogenic Pb (Figure 4-10a) and 87Sr/86Sr ratios (0.70255 to 0.70267) and less

radiogenic 143Nd/144Nd ratios (0.51312 to 0.51318; Figure 4-10b). Consistent with their

elevated incompatible element ratios, the samples have Sr, Nd and Pb isotopes that

plot between the field typical EPR N-MORB and more radiogenic E-MORB (Figure 4-

10).

Discussion

Shallow-level Processes Involved in the Petrogenesis of Ferrobasalts, FeTi
Basalts and Basaltic Andesites at the 90N OSC

Shallow level differentiation of MOR magma is primarily controlled by fractional

crystallization and mixing of different MORB melts (Clague & Bunch, 1976; Bryan &

Moore, 1977; Byerly, 1980). These processes are often difficult to discern from each

other on geochemical variation diagrams because mixing can produce compositions

that lie along liquid lines of descent resulting from fractional crystallization. However, our

geochemical results and petrographic studies suggest that both mixing and fractional

crystallization are important at the OSC, particularly in the generation of basaltic

andesites.

The geochemistry of most of the basaltic lavas erupted at the 9N OSC, including,

ferrobasalts, FeTi basalts, and many basaltic andesites, can primarily be explained


115









through low-pressure fractional crystallization of typical N-MORB parental magmas

(Figure 4-5). In general, the compositions of the 9N lavas are slightly more evolved

than the average MORB erupted along the northern EPR (Perfit et al., 1994; Goss et al.,

2010). Based on MELTS calculations (Ghiorso & Sack, 1995), ferrobasalts and FeTi

basalts can be generated by ~30% and ~55% fractional crystallization (respectively) of

ol + plag + cpx at pressures of 1 kbar from a relatively primitive MORB magma (2392-9

from the 1991 EPR eruption at 9050'N; Figure 4-6). These results are consistent with

previous studies of EPR lavas covering large geographic areas that collectively suggest

fractional crystallization is the dominant or exclusive process effecting the composition

of MOR magmas (e.g. Batiza & Niu, 1992). In contrast, many of the basaltic andesites

erupted at the 9N OSC lie off of typical fractional crystallization trends.

OSC lavas are predominantly aphyric, but some lavas contain sparse phenocrysts

and microphenocrysts of olivine, clinopyroxene and plagioclase that are consistent with

shallow- level fractional crystallization of a typical MORB (Grove et al., 1992).

Petrographic studies, however, suggest some lavas have had more complex

petrogenetic histories (Zaino, 2009). Although some plagioclase phenocrysts in east

limb ferrobasalts and basaltic andesites exhibit normal zoning from core to rim, others

show reverse zoning or no zoning at all. Commonly, all three zoning patterns occur in a

single sample. This suggests that each phenocryst has undergone a distinct magmatic

history prior eruption on the seafloor and supports the role of mixing in the petrogenesis

of these lavas. Resorbed rims on olivine phenocrysts in east limb basalts coupled with

more magnesian compositions that should be in equilibrium with co-existing glasses are

also indicative of magma mixing (Zaino, 2009).


116









Many of the low MgO OSC lavas have elemental concentrations that lie off of

calculated liquid lines of descent and are inconsistent with crystal fractionation alone.

This is particularly apparent in the concentrations of FeO, TiO2, and P205 in many

basaltic andesites (Figure 4-6). These elements are typically enriched in MORB

systems during fractional crystallization until melts become saturated with Fe-oxides (~5

wt % MgO) and apatite (~1 wt% MgO). However, many of the evolved OSC lavas have

relatively low concentrations of these elements and appear to lie along straight lines that

suggest mixing between high-silica and ferrobasaltic magmas. Similar linear trends are

observed in incompatible and compatible trace element concentrations (i.e. Cr and Ni)

versus Zr and MgO (not shown). Similarly, trace element ratios (e.g. U/Nb) of basaltic

andesites do not plot along fractional crystallization trends but lie along mixing lines

between evolved basalts and high-silica lavas (Figures 4-8, 4-9). The high-silica mixing

end-member is likely produced during partial melting of the oceanic crust or assimilation

and fractional crystallization (AFC) processes that can produce MOR dacites (Wanless

et al., accepted). Using an OSC dacite composition as an end-member, our mass-

balance calculations confirm that many of the basaltic andesites can be produced by

bulk mixing of 25% dacitic melt with 75% ferrobasaltic melt. Similar processes can

explain the low-P205 andesites (see discussion below).

This scenario is consistent with the petrogenetic models of dacite formation on

OSC, which requires an episodic magma supply at the propagating limb of the OSC

(Wanless et al., accepted). In this case, a basaltic magma is injected into the OSC and

undergoes variable degrees of crystal fractionation to produce ferro- and FeTi basalt

compositions. These magmas mix to varying degrees with pre-existing high-silica melts









beneath the ridge axis to produce a range of compositions, including basaltic andesites

and andesites (Figure 4-5).

Fractional crystallization versus assimilation in the petrogenesis of andesites

Although fractional crystallization and magma mixing appear to be the dominant

processes involved in the differentiation of magmas at the 9N OSC, incompatible trace

element concentrations suggest that the formation of highly evolved lavas (i.e., dacites)

on ridges requires partial melting and assimilation of altered ocean crust (Wanless et

al., accepted). Here, we assess the roles of fractional crystallization, assimilation and

magma mixing in the formation of intermediate compositions (andesites) on MOR.

There are two distinct geochemical populations of andesites at the OSC, which likely

require different petrogenetic histories (Figure 4-5).

Both populations of andesites (high-P205 and low-P205 andesites) have similar

SiO2 and MgO concentrations, but different TiO2, FeO, A1203, P205, and trace element

concentrations at a given MgO (Figures 4-5, 4-6, and 4-8). Petrologic modeling of

major element variations indicates that high-P205 andesite magmas formed primarily

through extensive crystal fractionation of basaltic magmas (Figure 4-6). MELTS

calculations (Ghiorso and Sack, 1995) suggest that this process involved up to 75%

fractional crystallization of ol + plag + cpx and minor amounts of Fe-oxide of a

ferrobasaltic parent (265-43) at 1 kbar pressure. Oxygen fugacity at the QFM-1 buffer is

required to delay the onset of Fe-oxide crystallization, and produce the elevated FeO

and TiO2 concentrations observed (Figure 4-5). Elevated P205 concentrations require

that apatite crystallization has not occurred during the evolution of the melt (Figure 4-6).

This is consistent with MELTS calculations and is supported by apatite saturation

calculations for these compositions (Watson, 1979), which suggest the high liquidus


118









temperature (>1000C) of these melt compositions inhibits apatite saturation. Trace

element ratios also suggest extensive fractional crystallization is required to explain

incompatible trace element concentrations and ratios (e.g. low U/Nb ratios). Rayleigh

fractionation calculations (Figure 4-8) suggest that ~85% fractional crystallization is

required to produce the high-P205 incompatible trace element abundances, compared

to the 75% calculated using MELTS for the major elements (Figure 4-6).

In contrast, low-P205 andesites have low FeO, TiO2, and P205 concentrations and

high U/Nb ratios (Figures 4-6, 4-9). The low FeO and TiO2 and high U/Th ratios are a

result of crystal fractionation of Fe-oxides and the low P205 is a results of apatite

crystallization. These lavas are spatially related to high-silica dacites erupted on-axis

and are likely formed from mixing of high-silica melts and ferrobasaltic magmas. The

high-silica melts are a result of AFC processes that include crystallization of Fe-oxides

and apatite (Wanless et al., accepted). This is similar to the petrogenesis of the basaltic

andesites, but in opposite proportions. Bulk mixing calculations suggest mixes of up to

25% ferrobasalt and as low as 75% dacitic melt will produce compositions similar to the

low-P205 andesites.

Formation of west limb lavas by fractional crystallization

The west limb of the OSC has been receding at a rate approximately equal to the

propagation of the east limb (Carbotte & Macdonald, 1992). This must reflect a

progressive decrease in magma supply on the western limb tip and, therefore, a

decrease in the heat supply and magma recharge, which could lead to greater degrees

of crystal fractionation (e.g. Christie & Sinton, 1981). Under these circumstances, one

might expect the eruption of a range of compositions, including high-silica lavas.


119









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

GEOLOGY AND PETROGENESIS OF LAVAS ON AN OVERLAPPING SPREADING
CENTER: 9N EAST PACIFIC RISE

By

V. Dorsey Wanless

August 2010

Chair: Michael Perfit
Major: Geology

In contrast to relatively homogeneous mid-ocean ridge basalt (MORB)

compositions typically erupted on fast-spreading oceanic ridges, a wide range of rock

types from basalts to dacites have been recovered at overlapping spreading centers

(OSC). This study focuses on the petrogenesis of lavas erupted at the 9N OSC on the

East Pacific Rise in order to better understand the complex magmatic plumbing system

beneath a ridge discontinuity. Lavas that span the entire compositional range observed

on the global mid-ocean ridge (MOR) system, including basalts, ferrobasalts, FeTi

basalts, basaltic andesites, andesites and dacites have erupted along the eastern,

propagating limb of the OSC. Major and trace element analyses, radiogenic (Pb, Sr, Nd)

and oxygen isotopic ratios, volatile contents (CI, H20, C002) and geochemical modeling

are used to determine the petrogenesis of MORB and genetically related high-silica

magmas.

The formation of high-silica dacites on MOR remains a petrologic enigma despite

eruption on several different ridges. They are characterized by elevated U, Th, Zr, and

Hf; relatively low Nb and Ta; and A1203 and K20 concentrations that are higher than

expected from fractional crystallization. Additionally, high Cl and H20 concentrations









However, the lavas erupted on the west limb encompass a narrower compositional

range than that observed on the eastern, propagating limb. With the exception of a

single andesitic lava, lavas erupted on the west limb can be explained by up to -25%

fractional crystallization of either an E- or N-MORB parent with ~8.5 wt% MgO (Figure

4-11). Trace element patterns are consistent with fractional crystallization as the

primary shallow-level differentiation process (Figure 4-12); however, some complex

zoning patterns in plagioclase phenocrysts (Zaino, 2009) are consistent with a

combination of crystallization and magma mixing.

The single west limb andesite recovered appears to have formed through

extensive fractional crystallization. Petrologic modeling suggest that ~75% fractional

crystallization can explain the major and trace element compositions observed in the

west limb andesite. When compared to the wide range of east limb andesites, the west

limb andesite has trace element patterns similar to the high-P205 andesites, which are

geochemically dominated by fractional crystallization (Figure 4-13). The west limb

andesite lacks negative Nb and Ta anomalies observed in the low-P205 andesites, has

a small negative K anomaly, and elevated phosphorus contents and P205/TiO2 ratios

(Figures 4-11, 4-14). Assimilation does not appear to play a significant role in the

formation of the west limb andesite, which may be due to cooler crustal conditions on

the dying limb.

Where Are the South Tip and West Limb Dacites?

A near-absence of high-silica lavas appears to be a characteristic of both the

southern tip of the eastern limb and the western limb as a whole (Figure 4-2), in contrast

to the abundant dacites erupted over the northern east limb. This lack of high-silica

samples is puzzling because the tips of both the eastern and western limbs should


120









experience low magma supply rates, which would promote greater degrees of fractional

crystallization and more evolved magma compositions (Christie & Sinton, 1981). We

argue that the subtle difference in the thermal conditions in the oceanic crust play a

major role in controlling the composition of erupted lavas.

Large amounts of heat, in part from the latent heat of crystallization and in part

from replenishment by an episodic magma supply, in the volcanically active region of

the propagating eastern limb enhances melting and assimilation of ocean crust, while

allowing for extensive fractional crystallization (Wanless et al., accepted). In contrast,

low magma recharge rates and cooler conditions in the tectonically controlled southern

east limb tip and dying west limb may result in extensive fractional crystallization, but

melting and assimilation are inhibited. The presence of an unusually large and

extensive melt lens below the northern portion of the eastern ridge axis is consistent

with higher crustal temperatures and greater magmatic activity that results in partial

melting of the oceanic crust. Additionally, the northern portion of the east limb may

experience higher rates of magma recharge than the starved eastern tip and the dying

western limb. This leads to the formation of high-silica lavas through AFC processes

(Wanless et al., accepted). In contrast, the narrower, deeper melt lenses in the

southern tip of the east limb and dying west limb (Kent et al., 2000) provide less heat to

the system, resulting in cooler crust, which may inhibit melting, assimilation and the

formation of dacitic lavas. These observations are supported by compositions of the

two andesite lavas erupted on the southern tip and the west limb (Figure 4-13), which

have incompatible trace element patterns similar to andesites produced primarily by

fractional crystallization with little evidence of assimilation. Another possibility is that









melting and assimilation may occur in these regions, but that the melts are

volumetrically too small to erupt, perhaps resulting in the formation of small

plagiogranite intrusions and veins within the crust rather than erupted lavas.

Composition of the Melt Lens Beneath the East Limb

Seismic evidence indicates that the shallow crustal melt lens widens to more than

4 km beneath the northern portion of the east limb (Figure 4-3) but it is centered

beneath the inter-limb region and not below the neo-volcanic zone (Kent et al., 2000;

Tong et al., 2002). Tomographic studies reveal a low velocity zone beneath the entire

OSC at ~ 9 km depth, near the mantle-crust transition (Toomey et al., 2007). Despite

this evidence for a regionally robust magmatic system, there is scant evidence for the

eruption of primitive lavas anywhere within the OSC, but, enigmatically, the lavas

erupted have evolved compositions. This suggests that even in areas with large,

imaged melt bodies, magmas that originate within the shallow mantle have complicated

and extended differentiation histories in the crust and are unlikely to erupt in pristine

condition.

It is difficult to relate lavas erupted within the inter-limb region to the currently

imaged melt lens, however, they do appear to be younger than expected, suggesting

that they did not erupt on-axis (Nunnery et al., 2008). Additionally, the observation that

lavas erupted within the inter-limb region (directly above the melt lens) are almost

exclusively composed of ferrobasalts (60 of 67 samples) with very uniform compositions

(Figures 4-6, 4-8) suggests that these reflect the relatively evolved nature of the shallow

melt lens.

The slight variability in major and trace element compositions of the inter-limb

ferrobasalts can be modeled by 20% to 30% fractional crystallization of a MORB


122









parental composition (MgO = 7 wt%). Many basalts erupted within the neo-volcanic

zone on the eastern limb are also ferrobasaltic and have incompatible trace element

concentrations that are similar to inter-limb ferrobasalts, despite the proximity of high-

silica lavas. Assuming that the these lavas are directly related to the current melt lens,

this suggests that the 4 km wide melt lens is primarily composed of ferrobasaltic magma

and that beneath the east limb neo-volcanic zone it has mixed with high-silica melts,

creating a wide range of compositions on the eastern edge of the melt lens.

Presumably a large volume of primitive mantle-derived basalts must have been

fractionally crystallized at crustal and possibly upper mantle depths to result in a large,

evolved melt lens. Crystallization of a primitive magma to form a 4 km wide melt lens

with ferrobasaltic composition would release significant amounts of heat to the

surrounding region. It is the latent heat of crystallization that provides the extra heat to

cause partial melting and assimilation of the crust leading to the formation of high-silica

magmas along the eastern limb of the OSC (Wanless et al., accepted).

E-MORB Distribution at 90N OSC

Both E-MORB and N-MORB lavas were recovered on the western limb of the OSC

in this study as well as during the CHEPR cruise [Langmuiret al., 1986]. In contrast,

only N-MORB lavas have been recovered from the east limb. Incompatible element

enrichment corresponds with more radiogenic Sr and Pb isotopes and less radiogenic

Nd isotopes in the west limb E-MORB lavas (Figure 4-10).

The sub-ridge upper mantle is generally thought to be relatively incompatible

element-depleted and isotopically homogeneous; however, there are well-documented

cases of small-scale (veined) heterogeneities in the MOR mantle source. Proximal and

roughly coeval eruptions of E-MORB and N-MORB at MOR's suggest that the sub-ridge


123









mantle varies in elemental and isotopic composition over small spatial/temporal scales

[e.g. Hart et al., 1973; Sun et al., 1975; White and Schilling, 1978]. Formation of E-

MORB magma is often interpreted as a result of overall lower degrees of mantle melting

of a source comprised of a greater compliment of incompatible enriched material

compared to N-MORB magma. More recently, 2 stage models involving "normal"

amounts of melting of enriched components embedded in a largely depleted sub-ridge

mantle have had success in explaining both elemental and isotopic systematics in E-

type MORB [Donnelly et al., 2004]. The enriched component is thought to be

volumetrically smaller than normal mantle and may consist of enriched veins in a

depleted mantle [e.g. Hanson, 1977]. While E-MORB may be an important component

in ridge magmas, it may be diluted or overwhelmed by the N-MORB signature during

more robust magmatic activity. Consequently, eruption of E-MORB compositions have

been linked to diminished magmatic activity along ridge segments [Reynolds and

Langmuir, 1997, Waters et al. in press].

The eruption of E-MORB lavas only on the western limb of the OSC suggests a

magmatic plumbing system that either allows for the preservation of enriched melts from

the mantle on one limb compared to the other or that the mantle sources supplying each

limb are different. The nonsymmetrical distribution of parental MORB types may result

from differences in magma supply to the two limbs. As spreading shifts from one axis to

the other at an OSC, the magma supply from the mantle will diminish on the dying limb

and progressively increase on the propagating limb. This allows for the preservation of

enriched melts on the dying limb in contrast to the magmatically robust propagating

limb.


124









9N OSC as a Division in Mantle Components

N-MORB lavas from the western limb of the OSC and at 8037'N have measurably

different isotope ratios compared to N-MORB lavas erupted on the east limb (Figure 4-

10). The east and west limb N-MORB lavas have similar incompatible trace element

ratios (i.e. LaN/YbN) but west limb N-MORB lavas have higher 208Pb/204Pb and

207Pb/204Pb ratios (Figure 10a) and lower sNd (Figure 10b) compared to east limb N-

MORB. These compositions cannot be explained by simple 2 component mixing of an

enriched source with a typical N-MORB component from the northern EPR (Figure 4-

10). These isotopic differences suggest that different mantle sources are feeding the

two limbs and that the OSC acts as a division between these sources.

The west limb E-MORB lavas have isotopic compositions similar to E-MORB lavas

erupted at the intersection of the EPR with the Siqueiros Transform fault to the south (~

8N). The 8037' lavas, which lie geographically between the Siqueiros Transform and

the 9N OSC, also have more radiogenic Sr and Pb than the east limb lavas (Figure 4-

10). This suggests that the mantle source below the EPR from the dying west limb to

the Siqueiros Transform fault is generally more enriched than the mantle beneath the

east limb of the OSC extending up to the Clipperton Transform [Sims et al., 2002; Sims

et al., 2003]. This is consistent with observations that the leading limb of the EPR may

tap a slightly more "enriched" mantle than the trailing limb [Carbotte et al., 2004]. While

it appears to be more enriched overall, simple 2 component mixing of depleted mantle

with an enriched end-member similar to the Siqueiros E-MORB cannot explain the

range of isotopic ratios erupted on this segment of the EPR.


125









Conclusions

The 9N OSC is one of the most extensively studied 2nd order discontinuities on

the global MOR system and one with the most completely imaged crustal melt lens. It

has erupted a wide compositions ranging from basalts to dacites. Much of the

compositional variability can be ascribed to low-pressure (~1 kbar) fractional

crystallization of N-MORB magma or mixing of ferrobasaltic and high-silica magmas.

The west limb magmas are slightly less evolved than those on the east limb.

Andesitic compositions on the east limb can be divided into two different groups

(high- and low-P2Os) with different major and trace element characteristics. The high-

P205 andesites are produced dominantly by extensive fractional crystallization (~75%).

Low-P205 andesites are produced through extensive mixing of high-silica dacitic and

ferrobasaltic magma. Magma mixing also explains the compositions of many of the

basaltic andesites erupted on the east limb that are not consistent with calculated

MORB fractional crystallization trends.

The near-absence of high-silica lavas on the southern tip of the east limb and the

entire west limb compared to the northern east ridge axis suggests a different tectono-

magmatic environment in these settings. We believe that this is due to the cooler

temperatures of the ocean crust in these regions as a consequence of decreased

magmatic input. Dacitic lavas are produced from the combination of assimilation and

fractional crystallization in regions where the latent heat of crystallization provides

enough heat to partially melt the surrounding wall rock. The cooler crust at the dying

western limb and the southern ridge tip may allow for extensive fractional crystallization,

however, it is either not enough to heat and melt the surrounding wall rock or these

partial (anatectic?) melts are not erupted.


126









The distribution of evolved lavas and E-MORB lavas across the OSC is not

symmetric, suggesting that the 2nd order discontinuity represents a boundary??

division? in the magmatic plumbing system of the EPR. E-MORB lavas are only

observed on the dying western limb and overall the lavas are less evolved. We suggest

that the lower magma supply at the west limb allows for the preservation and eruption of

E-MORB compositions, whereas the more robust magmatic system on the propagating

east limb overwhelms this signature.

N-MORB lavas on the west limb have more radiogenic Pb and Sr and less

radiogenic Nd compared to east limb N-MORB lavas. Lavas erupted to the south of the

OSC (837'N) also have more radiogenic Pb and Sr isotope ratios. This suggests a

slightly different mantle is feeding this section of the EPR and that the OSC provides a

fundamental division between mantle sources beneath the ridge axis.










Table 4-1. East limb major element data 9N OSC
sample Rock Type SiO2 TiO2 A1203 FeO Mn MgO CaO Na20 K20 P205 Cl S Total
WC-07 FeTi 49.28 2.60 12.57 14.08 0.24 5.71 9.24 3.21 0.20 0.28 97.42
WC-06 FeTi 49.73 2.49 12.69 14.02 0.24 5.67 9.35 3.28 0.19 0.27 97.94
WC-08 ferrobasalt 49.99 1.75 13.80 11.11 0.21 7.27 10.95 2.90 0.15 0.16 98.29
WC-01 ferrobasalt 50.07 1.83 13.69 11.36 0.21 7.20 10.94 2.94 0.14 0.18 98.57
265-43 ferrobasalt 50.53 1.92 13.88 11.56 0.21 6.98 11.14 2.86 0.13 0.19 0.01 0.25 99.67
WC-02 ferrobasalt 50.13 1.80 13.73 11.31 0.21 7.26 11.04 2.94 0.13 0.20 98.75
265-98 ferrobasalt 50.24 1.94 13.82 11.75 0.21 7.17 10.86 2.86 0.12 0.19 99.16
266-14 ferrobasalt 50.45 2.14 13.48 12.38 0.22 6.69 10.75 3.11 0.16 0.22 99.60
264-08 FeTi 50.10 2.68 12.73 14.06 0.26 5.69 9.58 3.27 0.21 0.28 0.07 0.32 99.24
266-15 ferrobasalt 50.51 2.38 13.09 13.30 0.23 6.27 10.11 3.27 0.18 0.26 99.61
266-51 ferrobasalt 50.55 1.94 13.85 11.59 0.21 7.33 10.97 2.86 0.13 0.19 0.01 99.62
267-14 ferrobasalt 50.55 1.92 14.04 11.57 0.21 6.96 11.05 2.86 0.13 0.21 99.51
265-20 ferrobasalt 50.68 1.85 13.92 11.31 0.22 7.21 11.34 2.95 0.13 0.18 0.01 0.22 100.0
266-48 ferrobasalt 50.57 1.92 14.09 11.56 0.21 7.34 11.00 3.03 0.13 0.19 100.0
266-38 FeTi 50.58 2.00 13.60 12.35 0.23 6.63 10.73 3.15 0.17 0.20 99.63
267-13 ferrobasalt 50.59 1.85 14.08 11.34 0.21 7.11 11.21 2.82 0.12 0.21 99.54
267-03 FeTi 50.60 2.13 13.58 12.49 0.23 6.45 10.15 3.05 0.16 0.26 99.10
264-23 ferrobasalt 50.61 1.87 13.89 11.29 0.22 7.28 11.14 2.92 0.13 0.18 99.52
266-16 ferrobasalt 50.61 2.02 13.67 12.10 0.23 6.87 10.75 3.15 0.16 0.22 99.79
265-44 ferrobasalt 50.62 1.93 14.02 11.72 0.22 6.86 11.11 2.81 0.13 0.19 99.62
266-40 ferrobasalt 50.63 1.77 13.96 11.52 0.20 7.24 11.25 2.96 0.14 0.16 99.84
266-17 ferrobasalt 50.63 1.87 14.00 11.31 0.20 7.24 11.43 3.01 0.14 0.19 100.0
265-99 ferrobasalt 50.65 1.84 13.91 11.42 0.20 7.36 11.00 2.90 0.13 0.18 99.57
266-34 ferrobasalt 50.65 1.84 14.10 11.30 0.21 7.25 11.08 2.90 0.11 0.18 99.63
265-87 ferrobasalt 50.65 1.98 13.84 11.94 0.23 6.84 11.06 2.77 0.14 0.20 99.64
266-32 ferrobasalt 50.67 1.85 14.01 11.38 0.22 7.27 11.13 2.92 0.12 0.18 99.74
265-96 ferrobasalt 50.67 1.83 14.00 11.43 0.19 7.35 11.01 2.85 0.13 0.18 99.65
267-05 FeTi 50.68 1.99 13.69 12.00 0.22 6.39 10.31 3.06 0.16 0.25 98.76
264-05 ferrobasalt 50.69 1.83 13.79 11.42 0.22 7.07 11.14 2.95 0.15 0.18 99.44
267-12 FeTi 50.70 2.20 13.58 12.75 0.22 6.38 10.40 3.03 0.16 0.25 99.67
266-39 FeTi 50.71 2.04 13.62 12.42 0.24 6.57 10.54 3.21 0.19 0.20 99.74
267-08 FeTi 50.71 1.99 13.80 12.03 0.21 6.79 10.77 2.95 0.14 0.22 99.62
265-19 ferrobasalt 50.62 1.83 14.02 11.23 0.21 7.19 11.24 2.93 0.12 0.18 0.01 0.24 99.83
266-33 ferrobasalt 50.73 1.84 14.00 11.35 0.22 7.29 11.13 2.88 0.12 0.20 99.75
266-13 ferrobasalt 50.74 2.29 13.21 12.95 0.23 6.34 10.54 3.21 0.17 0.25 99.93


128





129


Table 4-1. Continued
sample Rock Type SiO2 TiO2
266-52 ferrobasalt 50.74 1.93
265-14 ferrobasalt 50.75 1.83
WC-05 ferrobasalt 50.75 2.00
266-02 FeTi 50.76 2.08
266-08 ferrobasalt 50.76 1.98
265-86 ferrobasalt 50.76 1.99
267-09 ferrobasalt 50.76 1.95
265-97 ferrobasalt 50.77 2.02
265-82 ferrobasalt 50.77 1.95
265-92 ferrobasalt 50.77 1.98
265-10 ferrobasalt 50.78 1.86
265-52 ferrobasalt 50.78 1.93
265-15 ferrobasalt 50.79 1.86
267-11 FeTi 50.79 2.20
266-19 ferrobasalt 50.79 1.86
266-12 ferrobasalt 50.80 2.15
265-68 ferrobasalt 50.81 1.81
265-53 ferrobasalt 50.81 1.96
266-18 ferrobasalt 50.82 2.01
266-37 FeTi 50.82 2.03
265-62 ferrobasalt 50.82 1.95
266-11 ferrobasalt 50.82 2.23
267-07 ferrobasalt 50.82 1.96
265-13 ferrobasalt 50.83 1.85
265-11 ferrobasalt 50.84 1.80
267-01 FeTi 50.84 2.03
265-45 ferrobasalt 50.84 1.94
265-88 FeTi 50.85 2.02
266-09 ferrobasalt 50.85 1.93
265-17 FeTi 50.68 2.09
265-119 ferrobasalt 50.86 2.00
265-73 ferrobasalt 50.86 1.96
267-02 ferrobasalt 50.87 1.91
265-16 ferrobasalt 50.87 1.84
265-04 ferrobasalt 50.87 1.84


A1203 FeO
13.92 11.63
13.92 11.20
13.44 11.53
13.63 12.45
13.88 11.74
13.83 11.87
13.90 11.99
13.78 12.02
13.81 11.70
13.91 11.77
13.94 11.33
14.01 11.77
13.90 11.25
13.66 12.69
13.73 11.87
13.43 12.67
14.02 11.17
14.05 11.83
13.70 12.08
13.57 12.26
13.93 11.72
13.30 12.93
13.94 11.84
14.01 11.22
14.00 11.11
13.65 12.39
14.12 11.80
13.88 11.99
14.00 11.43
13.45 12.61
13.88 11.77
13.98 11.79
14.10 11.45
14.13 11.23
14.01 11.21


Mn MgO CaO Na20 K20 P205
0.21 7.30 10.95 2.89 0.13 0.18
0.21 7.16 11.19 2.97 0.12 0.18
0.21 6.11 9.69 3.35 0.26 0.27
0.22 6.68 10.51 3.11 0.12 0.21
0.22 7.12 10.92 2.86 0.13 0.20
0.22 6.79 11.08 2.72 0.14 0.20
0.22 6.90 10.82 2.95 0.14 0.24
0.22 7.03 10.87 2.61 0.13 0.21
0.22 6.93 11.21 2.69 0.14 0.19
0.22 7.08 10.99 2.93 0.13 0.20
0.22 7.13 11.23 2.99 0.13 0.18
0.22 6.99 11.18 2.92 0.13 0.20
0.21 7.14 11.24 2.99 0.13 0.17
0.22 6.44 10.32 2.95 0.16 0.25
0.23 6.88 11.03 3.09 0.14 0.20
0.24 6.51 10.66 3.27 0.18 0.22
0.21 7.29 11.51 2.79 0.13 0.17
0.22 6.85 11.17 2.92 0.13 0.19
0.22 6.91 10.88 3.14 0.16 0.22
0.22 6.73 10.58 3.02 0.15 0.20
0.22 6.92 11.19 2.90 0.13 0.19
0.24 6.43 10.34 3.26 0.18 0.27
0.20 6.96 10.86 2.94 0.14 0.22
0.22 7.21 11.28 2.93 0.12 0.18
0.21 7.23 11.29 2.95 0.12 0.17
0.21 6.34 10.18 3.07 0.16 0.23
0.21 6.87 11.21 2.89 0.13 0.19
0.22 6.71 11.06 2.85 0.15 0.21
0.21 7.10 11.08 2.84 0.12 0.19
0.23 6.48 10.75 3.14 0.13 0.20
0.22 7.12 10.86 2.90 0.13 0.19
0.22 6.97 11.20 2.83 0.13 0.20
0.20 6.73 10.73 2.71 0.15 0.24
0.22 7.22 11.22 3.01 0.12 0.17
0.21 7.15 11.38 3.00 0.13 0.18


Cl S Total
99.87
99.54
97.62
99.77
99.81
99.59
99.87
99.65
0.01 99.60
99.97
99.78
100.1
99.67
99.68
99.83
100.1
0.01 99.92
100.1
100.1
99.59
99.97
100.0
99.89
99.83
99.72
99.11
100.2
0.02 99.94
99.76
0.00 0.27 100.0
99.94
100.1
99.09
100.0
99.97









and relatively low 6180 values in dacitic glasses require contamination from a seawater-

altered component. Extensive petrologic modeling of MOR dacites suggests that

fractional crystallization of a MORB parent combined with partial melting and

assimilation of altered ocean crust can generate magmas with geochemical signatures

consistent with MOR dacites. This suggests that crustal assimilation is a much more

important process on ridges than previously thought and may be significant in the

generation of evolved MORB in general.

Petrologic models indicate that ferrobasalts and FeTi basalts erupting at the OSC

can be explained by low-pressure fractional crystallization of a primitive MORB parent;

however, both fractional crystallization and magma mixing produce intermediate

compositions. Geochemical analyses suggest that there are two distinct populations of

andesites erupted at the OSC. Andesites with high-P205 are the most evolved MOR

compositions produced through fractional crystallization. In contrast, low-P205 andesites

and basaltic andesites appear to have formed primarily through mixing of ferrobasaltic

and dacitic magmas.










Table 4-1. Continued


130


sample Rock Type
267-06 ferrobasalt
266-26 FeTi
265-79 ferrobasalt
266-04 FeTi
265-72 ferrobasalt
266-22 FeTi
265-39 ferrobasalt
264-06 ferrobasalt
265-46 ferrobasalt
266-23 FeTi
266-24 FeTi
265-08 ferrobasalt
265-18 ferrobasalt
265-09 ferrobasalt
266-27 FeTi
265-89 ferrobasalt
266-30 ferrobasalt
265-41 ferrobasalt
264-04 ferrobasalt
266-28 FeTi
266-35 FeTi
266-07 FeTi
267-10 ferrobasalt
265-05 ferrobasalt
265-78 ferrobasalt
264-07 ferrobasalt
267-15 ferrobasalt
266-29 FeTi
266-01 FeTi
265-71 ferrobasalt
265-80 ferrobasalt
265-74 ferrobasalt
265-02 ferrobasalt
266-36 ferrobasalt
266-21 ferrobasalt


SiO2 TiO2 A1203
50.88 1.97 14.02
50.88 2.13 13.57
50.89 1.89 13.99
50.89 2.36 13.25
50.89 1.93 14.05
50.90 1.92 13.87
50.90 1.95 13.80
50.90 1.83 13.75
50.91 1.94 14.09
50.91 1.89 13.89
50.91 1.90 13.87
50.92 1.86 13.94
50.75 1.82 14.01
50.93 1.85 13.96
50.94 2.13 13.59
50.94 1.88 14.01
50.95 1.88 13.78
50.96 1.92 14.14
50.54 1.84 13.66
50.97 2.15 13.52
50.97 1.91 13.74
50.97 2.36 13.15
50.97 1.97 13.99
50.98 1.88 13.88
50.98 1.88 14.01
50.98 1.82 13.84
50.98 2.00 13.80
50.99 2.19 13.32
50.99 2.00 13.51
51.00 1.89 13.96
51.01 1.92 13.95
51.01 1.99 13.93
51.04 1.89 13.98
51.05 1.89 13.83
51.06 1.85 13.95


FeO Mn MgO CaO
11.91 0.22 6.91 10.84
12.82 0.22 6.34 10.30
11.75 0.22 6.92 11.20
13.41 0.23 6.18 9.94
11.76 0.23 6.98 11.13
12.05 0.21 6.79 10.73
11.57 0.21 7.13 11.19
11.37 0.22 7.06 11.14
11.88 0.22 6.80 11.20
12.00 0.20 6.80 10.80
12.04 0.22 6.81 10.70
11.25 0.21 7.21 11.37
11.31 0.21 7.21 11.26
11.28 0.22 7.20 11.28
12.83 0.23 6.32 10.19
11.57 0.22 6.95 11.31
11.90 0.21 7.02 10.73
11.47 0.22 7.10 11.31
11.53 0.21 6.97 11.06
12.86 0.23 6.34 10.27
12.13 0.22 6.91 10.59
13.50 0.25 6.16 9.72
11.97 0.21 6.92 10.81
11.38 0.22 6.99 11.34
11.70 0.21 6.97 11.18
11.40 0.21 7.08 11.19
11.81 0.21 6.97 10.84
12.88 0.23 6.39 10.15
12.39 0.23 6.76 10.71
11.68 0.22 6.96 11.19
11.70 0.22 6.98 11.24
11.92 0.22 6.81 11.05
11.39 0.22 6.99 11.37
11.88 0.22 7.04 10.66
11.93 0.22 6.94 10.80


Na20
2.84
3.06
2.86
3.13
2.85
2.97
2.95
2.96
2.99
2.97
2.98
2.96
2.97
2.97
3.11
2.76
2.97
2.84
2.98
3.07
3.00
3.18
2.95
3.02
2.86
2.98
2.86
3.07
3.04
2.82
2.87
2.90
3.03
2.97
2.96


K20
0.14
0.15
0.14
0.17
0.13
0.13
0.13
0.15
0.14
0.13
0.13
0.12
0.12
0.13
0.14
0.14
0.12
0.13
0.15
0.15
0.12
0.18
0.14
0.13
0.14
0.14
0.14
0.15
0.12
0.14
0.13
0.14
0.12
0.13
0.13


P205
0.22
0.23
0.19
0.23
0.20
0.21
0.18
0.17
0.20
0.21
0.22
0.16
0.18
0.18
0.25
0.18
0.19
0.19
0.18
0.25
0.20
0.25
0.23
0.18
0.19
0.18
0.21
0.24
0.17
0.19
0.19
0.20
0.18
0.20
0.19


Cl S Total
99.94
99.69
100.0
99.79
100.1
99.79
100.0
99.56
100.4
99.80
99.79
100.0
0.01 0.22 100.1
99.99
99.74
99.94
99.76
100.3
0.01 0.25 99.39
99.80
99.80
99.71
100.2
99.99
100.1
99.80
99.82
99.61
99.91
100.0
100.2
100.2
100.2
99.87
100.0





Table 4-1.
sample
265-81
265-22
266-25
266-41
264-03
266-10
266-03
265-03
265-76
265-06
266-31
265-07
265-38
265-30
265-29
265-121
265-26
265-51
265-21
265-36
265-35
265-28
265-33
265-37
265-104
265-27
265-34
265-107
265-93
265-23
265-01
265-32
265-115
265-12
265-31


Continued
Rock Type
FeTi
ferrobasalt
FeTi
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
ferrobasalt
FeTi
ferrobasalt
FeTi
ferrobasalt
FeTi


SiO2
51.06
51.06
51.06
51.06
51.07
51.07
51.07
51.08
51.08
51.09
51.10
51.11
51.13
51.13
51.17
51.17
51.17
51.17
51.18
51.18
51.21
51.23
51.23
51.23
51.25
51.27
51.30
51.30
51.32
51.32
51.36
51.36
51.38
51.41
51.62


TiO2
2.00
1.86
1.98
1.86
1.84
1.93
1.77
1.89
1.89
1.85
1.88
1.90
1.89
1.87
1.88
1.89
1.94
1.92
1.83
1.93
1.94
1.79
1.97
1.93
1.90
1.79
1.99
1.92
1.97
1.88
2.03
1.96
2.17
1.92
2.01


A1203
14.13
14.05
13.68
14.09
13.89
13.95
14.24
13.89
13.98
13.89
13.78
13.98
13.91
14.10
14.04
14.05
13.73
14.19
14.04
13.90
13.78
14.12
13.93
13.95
13.79
14.12
13.93
13.83
13.90
13.78
13.82
13.89
13.34
13.75
13.89


FeO
11.93
11.35
12.26
11.30
11.34
11.50
10.83
11.40
11.61
11.44
11.90
11.47
11.34
11.45
11.44
11.44
12.03
11.66
11.33
11.42
11.57
11.19
11.48
11.36
11.65
11.13
11.52
11.67
11.82
11.92
11.89
11.53
12.72
11.45
11.72


Mn
0.22
0.22
0.22
0.22
0.21
0.21
0.21
0.22
0.22
0.21
0.22
0.22
0.22
0.22
0.22
0.21
0.22
0.22
0.21
0.22
0.22
0.21
0.22
0.22
0.22
0.21
0.22
0.21
0.22
0.22
0.23
0.22
0.23
0.22
0.23


MgO
7.00
7.15
6.57
6.84
7.07
6.97
7.04
7.02
6.98
7.00
7.09
6.96
7.09
6.91
6.88
7.12
6.84
7.11
7.20
7.10
6.95
7.18
6.83
7.11
6.90
7.10
6.92
6.85
6.84
6.92
6.50
6.90
6.16
6.60
6.85


CaO
11.24
11.37
10.61
10.74
11.19
10.97
10.86
11.27
11.25
11.32
10.72
11.24
11.11
11.32
11.27
10.93
11.09
11.27
11.37
11.12
11.14
11.37
11.16
11.15
10.59
11.47
11.11
10.51
10.62
10.98
10.71
11.11
9.77
10.69
11.07


Na20
2.39
3.01
2.98
3.22
2.98
2.89
3.06
3.03
2.84
3.00
2.96
3.02
3.00
3.01
3.04
2.91
3.08
2.93
2.98
2.96
2.99
2.99
3.02
3.00
3.00
2.99
3.01
3.07
3.02
3.02
3.21
3.00
3.24
3.11
3.05


K20
0.13
0.12
0.14
0.21
0.15
0.14
0.16
0.13
0.14
0.13
0.13
0.13
0.15
0.15
0.15
0.14
0.16
0.13
0.13
0.14
0.15
0.15
0.15
0.14
0.16
0.14
0.15
0.16
0.15
0.13
0.17
0.15
0.19
0.19
0.15


P205
0.19
0.18
0.22
0.18
0.17
0.21
0.21
0.19
0.19
0.17
0.18
0.19
0.19
0.18
0.18
0.19
0.19
0.19
0.17
0.19
0.20
0.18
0.18
0.19
0.22
0.17
0.19
0.23
0.21
0.19
0.23
0.18
0.25
0.20
0.20


Cl S Total
100.3
100.4
99.72
99.71
99.90
99.84
99.44
100.1
100.2
100.1
99.95
100.2
100.0
100.3
100.2
100.1
100.5
100.8
100.4
100.2
100.2
100.4
100.2
100.3
0.05 99.69
100.4
100.3
99.76
100.1
100.4
100.2
100.3
99.45
99.54
100.8










Table 4-1. Continued
Rock
sample Type SiO2 TiO2
265-110 ferrobasalt 51.65 1.90
265-114 FeTi 51.66 2.18
265-111 FeTi 51.78 2.17
265-112 FeTi 51.79 1.97
266-06 ferrobasalt 51.80 1.70
265-113 FeTi 51.92 2.17
WC-04 FeTi 51.94 2.00
265-61 ferrobasalt 51.96 1.81
basaltic
264-18 andesite 52.15 1.77
basaltic
266-45 andesite 52.24 1.79
basaltic
264-11 andesite 52.36 1.72
basaltic
264-17 andesite 52.61 1.77
basaltic
265-105 andesite 52.77 1.90
basaltic
265-116 andesite 52.86 1.84
basaltic
265-49 andesite 52.81 1.78
basaltic
265-60 andesite 53.08 1.74
basaltic
264-21 andesite 53.15 1.82
basaltic
265-122 andesite 53.28 2.06
basaltic
265-106 andesite 53.33 1.94
basaltic
264-13 andesite 53.41 1.68
basaltic
265-118 andesite 53.42 1.86


A1203 FeO Mn
13.81 11.69 0.20
13.39 12.83 0.24
13.38 12.65 0.22
13.67 12.00 0.22
14.21 10.87 0.20
13.40 12.82 0.23
12.96 12.63 0.25
14.06 11.26 0.21

14.04 10.89 0.20

13.86 11.28 0.22

13.79 10.75 0.20

14.12 10.95 0.20

13.33 12.06 0.22

13.77 11.64 0.21

14.00 11.09 0.21

14.06 10.93 0.21

14.00 11.09 0.21

13.40 12.27 0.23

13.26 12.35 0.24

14.44 10.43 0.20

13.57 11.90 0.21


MgO CaO Na20
6.71 10.41 3.07


5.88
5.73
6.41
6.72
5.93
4.88
6.43

6.05

6.46

6.55

6.23

5.58

5.73

5.98

5.93

5.82

5.36

5.11

5.76

5.16


9.53 3.29
9.40 3.35
10.11 3.12
10.59 3.05
9.48 3.28
8.59 3.57
10.62 3.05

9.94 3.27

10.18 3.33

10.47 3.13

10.07 3.27

9.26 3.44

9.38 3.37

10.03 3.23

10.12 3.26

9.69 3.41

8.84 3.52

8.71 3.67

9.53 3.42

8.85 3.54


K20 P205 Cl S Total


0.17 0.23


99.84


0.22 0.27 99.48
0.23 0.30 99.21
0.19 0.24 99.72
0.18 0.14 99.47
0.22 0.28 99.72
0.33 0.44 97.59
0.19 0.18 99.76

0.26 0.20 98.77

0.27 0.17 99.80

0.24 0.19 99.40

0.25 0.20 99.66

0.27 0.32 99.13

0.26 0.28 99.33

0.27 0.20 0.04 0.21 99.84

0.27 0.19 99.80

0.29 0.22 99.71

0.30 0.23 99.50

0.32 0.39 0.19 99.31

0.33 0.20 99.40

0.31 0.35 99.16


132










Table 4-1. Continued
Rock
sample Type SiO2 TiO2 A1203 FeO Mn
basaltic
264-19 andesite 53.51 1.77 13.90 10.83 0.21
basaltic
265-48 andesite 53.85 2.36 13.06 13.33 0.24
basaltic
264-22 andesite 53.58 1.75 13.93 10.90 0.20
265- basaltic
108 andesite 53.77 1.95 13.28 12.60 0.23
basaltic
265-24 andesite 54.07 2.35 12.90 13.64 0.25
basaltic
265-50 andesite 54.15 2.14 13.41 12.19 0.22
basaltic
266-43 andesite 54.14 1.90 13.56 12.05 0.24
basaltic
265-59 andesite 54.24 2.15 13.42 12.27 0.23
basaltic
265-58 andesite 54.25 2.12 13.56 12.13 0.22
265- basaltic
120 andesite 54.33 2.06 12.78 13.10 0.24
basaltic
266-62 andesite 54.48 1.53 14.24 9.87 0.18
basaltic
265-55 andesite 54.55 2.10 13.45 12.19 0.22
basaltic
264-20 andesite 54.59 1.71 13.95 10.68 0.21
265- basaltic
123 andesite 54.62 1.91 13.46 11.67 0.22
basaltic
265-56 andesite 54.98 2.13 13.68 12.13 0.23
265- basaltic
124 andesite 55.06 1.69 13.55 10.78 0.19
basaltic
265-54 andesite 55.14 2.04 13.65 11.75 0.22
basaltic
264-10 andesite 55.00 1.62 13.34 10.23 0.19


MgO CaO Na20

5.72 9.52 3.44

4.44 8.36 3.57

5.80 9.62 3.45

5.08 8.80 3.63

3.71 7.61 3.75

4.54 8.45 3.59

4.42 8.16 4.03

4.61 8.52 3.53

4.56 8.52 3.57

4.22 7.87 3.72

5.33 8.89 3.46

4.51 8.38 3.62

5.10 8.84 3.60

4.93 8.40 3.61

4.42 8.36 3.63

5.03 8.44 3.54

4.20 8.11 3.64

5.27 8.76 3.56


K20 P205 Cl S Total

0.31 0.20 99.41

0.35 0.26 0.05 0.28 100.2

0.31 0.20 99.75

0.32 0.38 0.20 100.0

0.50 0.58 99.36

0.40 0.28 0.09 0.22 99.67

0.41 0.32 99.22

0.40 0.28 99.65

0.39 0.27 99.59

0.38 0.52 99.23

0.41 0.19 98.58

0.41 0.28 99.71

0.39 0.21 99.28

0.36 0.22 99.37

0.42 0.28 100.2

0.40 0.20 98.88

0.45 0.27 99.47

0.43 0.21 0.12 0.20 98.93


133










Table 4-1. Continued
Rock
sample Type SiO2 TiO2
basaltic
265-102 andesite 55.31 2.22
basaltic
WC-03 andesite 55.46 2.02
basaltic
265-126 andesite 55.48 1.66
basaltic
264-12 andesite 55.51 1.64
basaltic
265-125 andesite 55.59 1.66
basaltic
265-109 andesite 55.66 1.81
basaltic
265-77 andesite 55.74 2.13
basaltic
265-117 andesite 55.77 1.81
basaltic
265-101 andesite 55.90 2.14
basaltic
264-16 andesite 56.02 2.06
basaltic
265-57 andesite 56.29 2.12
basaltic
265-103 andesite 56.50 2.01
basaltic
265-91 andesite 56.83 2.10
266-61 andesite 57.47 1.70
266-59 andesite 57.57 1.58
265-90 andesite 58.08 1.91
265-100 andesite 58.09 1.76
266-05 andesite 58.40 1.86
266-55 andesite 59.52 1.65
266-54 andesite 59.65 1.72
265-69 andesite 61.02 1.72
265-25 andesite 61.17 1.48
264-14 andesite 61.75 1.30


A1203 FeO Mn

12.51 14.05 0.27

12.02 13.80 0.27

13.97 10.46 0.18

13.35 10.18 0.20

13.63 10.73 0.19

12.83 12.31 0.23

13.52 11.98 0.21

12.90 12.26 0.23

12.29 13.95 0.26

13.21 12.23 0.25

12.59 12.75 0.25

12.33 13.74 0.26

12.31 14.23 0.28
14.07 10.35 0.19
13.16 10.41 0.20
12.43 13.69 0.27
12.64 12.68 0.24
13.18 11.14 0.21
13.14 10.83 0.20
13.22 11.25 0.21
13.62 9.97 0.18
13.63 9.84 0.19
13.46 8.71 0.17


MgO CaO Na20 K20 P205 Cl S


3.34

2.50

4.78

4.98

4.89

3.90

3.95

3.90

3.02

4.37

3.69

2.74

2.09
3.13
3.66
1.74
1.89
3.04
2.13
2.28
1.91
1.92
2.47


7.14

6.33

8.37

8.47

8.30

7.47

7.80

7.45

6.76

8.10

7.46

6.45

6.24
6.57
6.71
5.76
5.51
6.29
5.30
5.60
5.38
5.03
5.54


3.85

4.10

3.67

3.65

3.64

3.76

3.61

3.75

4.03

2.40

3.59

3.77

3.45
3.76
3.60
3.51
3.82
3.77
3.86
3.91
3.89
4.28
3.94


0.44 0.65

0.56 0.74

0.41 0.19

0.47 0.21

0.41 0.20

0.45 0.43

0.48 0.31

0.43 0.44

0.49 0.65

0.45 0.37

0.52 0.40

0.52 0.65 0.30

0.60 0.78 0.31
0.63 0.24
0.61 0.23
0.66 0.74 0.34
0.63 0.54
0.66 0.30
0.77 0.42
0.75 0.43 0.42
0.80 0.27 0.20
0.99 0.43
0.83 0.22


Total

99.79

97.80

99.18

98.66

99.22

98.85

99.73

98.95

99.50

99.46

99.64

98.96

98.91
98.13
97.74
98.80
97.80
98.85
97.82
99.02
98.76
98.96
98.39










Table 4-1. Continued
Rock


sample Type
266-49 andesite
266-57 andesite
266-56 andesite
265-66 andesite
266-58 dacite
265-65 dacite
265-64 dacite
265-67 dacite
266-50 dacite
266-53 dacite
265-84 dacite
265-63 dacite
266-47 dacite
265-85 dacite
266-46 dacite
265-94 dacite
264-09 dacite
265-70 dacite
265-42 dacite
265-83 dacite
265-95 dacite
265-40 dacite


SiO2 TiO2 A1203 FeO Mn
62.34 1.27 13.15 8.64 0.15
62.47 1.30 13.16 9.20 0.16
62.47 1.35 13.25 9.24 0.17
62.81 1.43 13.13 9.05 0.18
63.01 1.10 13.06 8.43 0.16
63.79 1.26 13.25 8.14 0.15
64.04 1.28 13.12 8.27 0.16
64.10 1.34 13.33 8.49 0.16
64.26 1.07 13.17 8.08 0.14
64.28 1.06 13.31 8.06 0.14
64.39 1.13 13.17 8.18 0.15
64.43 1.29 13.26 8.22 0.15
64.53 0.99 13.18 7.74 0.14
65.01 1.06 13.13 7.99 0.16
65.03 0.94 12.90 7.17 0.15
65.22 0.97 13.04 7.90 0.14
65.76 0.89 13.15 7.03 0.13
66.26 0.87 13.20 7.17 0.14
66.46 0.94 13.04 7.92 0.16
67.46 0.76 13.27 6.68 0.13
67.46 0.77 13.10 6.47 0.12


MgO
2.19
1.59
1.58
1.99
1.75
1.34
1.60
1.49
1.27
1.12
1.23
1.29
1.02
1.18
1.41
1.13
1.06
0.80
0.89
0.67
0.94


CaO Na20
4.86 4.18
4.37 4.11
4.45 3.95
4.98 3.86
4.34 3.63
4.21 3.84
4.45 3.46
4.41 3.93
3.78 4.23
3.73 4.16
3.92 3.41
4.21 3.71
3.53 4.94
3.78 3.67
3.71 4.76
3.54 4.29
3.48 4.24
3.23 4.08
3.50 3.99
2.98 3.88
3.01 4.43


K20 P205 Cl S Total
0.91 0.21 97.90
0.98 0.29 0.51 97.64
0.96 0.33 97.74


0.89 0.24 0.23
0.96 0.26
0.97 0.22
0.97 0.20 0.24
0.95 0.23
1.10 0.24
1.09 0.25 0.65
1.19 0.22
0.99 0.21
1.22 0.23


98.55
96.69
97.17
97.55
98.42
97.35
97.20
96.98
97.76
97.54


1.22 0.20 0.64 97.41
1.19 0.17 97.43
1.14 0.23 97.61
1.21 0.20 0.58 0.06 97.78
1.33 0.19 0.70 97.27
1.20 0.21 0.51 0.07 98.91
1.37 0.16 0.67 97.37
1.21 0.15 97.67


135










Table 4-2. Trace element data 9N OSC
sample 265-43 264-08 266-33 267-09 265-88 265-72 266-22 265-18 264-04 266-28 266-07 265-78 267-15 266-01


7.78
42
347
108
41
54
59
94
18
1.14
120
44
126
3.08
0.01
8.37
4.54
14.16
2.34
12.6
4.42
1.48
5.78
1.08
7.22
1.52
4.39
0.67
4.43
0.68
3.44
0.21
0.39
0.18
0.08


9.99
42
450
17.97
43
34
60
116
21
2.12
126
59
180
5.46
0.03
17.07
6.81
20.51
3.32
17.3
5.99
1.94
7.74
1.45
9.69
2.05
5.94
0.91
6.01
0.93
4.83
0.37
0.61
0.34
0.13


7.52 8.23
41 42
321 354
155 87.16
40 41
48 46
61 60
88 96
18 19
1.20 1.50
114 113
39 44
125 140
2.91 3.65
0.02 0.02
8.25 11.12
4.13 4.79
13.18 14.97
2.29 2.56
12.1 13.5
4.16 4.69
1.45 1.58
5.53 6.10
1.02 1.14
6.62 7.36
1.43 1.56
4.04 4.47
0.62 0.68
3.97 4.35
0.60 0.66
3.20 3.58
0.20 0.25
0.60 0.59
0.18 0.23
0.08 0.09


7.97 7.74 8.24 7.48
41 42 43 43
351 350 359 338
68.56 96.21 50.29 166
42 42 43 42
47 54 38 53
60 60 61 64
97 96 99 95
18 18 19 18
1.25 1.09 1.05 1.00
124 117 120 126
44 44 45 42
133 126 130 122
3.46 3.07 3.03 2.76
0.01 0.01 0.01 0.01
10.31 8.44 7.85 6.87
4.94 4.47 4.61 4.27
14.97 13.90 14.38 13.40
2.46 2.31 2.38 2.21
13.2 12.5 12.8 11.9
4.48 4.37 4.45 4.15
1.52 1.46 1.52 1.40
5.92 5.75 5.87 5.46
1.11 1.07 1.11 1.02
7.32 7.09 7.31 6.74
1.53 1.51 1.55 1.44
4.48 4.34 4.51 4.15
0.68 0.66 0.69 0.63
4.47 4.33 4.55 4.16
0.68 0.67 0.70 0.64
3.59 3.40 3.50 3.23
0.23 0.21 0.21 0.19
0.45 0.40 0.41 0.38
0.20 0.18 0.18 0.16
0.08 0.07 0.07 0.07


136


8.47
53
406
119
51
72
76
103
24
1.64
146
45
142
4.01
0.02
13.20
4.98
15.49
2.61
13.9
4.78
1.67
6.18
1.15
7.46
1.57
4.55
0.68
4.41
0.68
3.59
0.28
0.59
0.22
0.09


9.21 7.70 7.77 7.94 5.99
40 36 43 41 33
357 337 353 330 273
40.69 9.33 84.45 138 40.62
41 38 43 40 34
32 47 52 51 38
55 50 64 58 51
100 93 95 93 79
20 17 18 18 14
1.50 1.55 1.22 1.37 0.98
111 98 125 108 89
43 42 43 42 32
157 139 124 136 95
3.87 3.84 3.19 3.42 2.20
0.02 0.02 0.02 0.02 0.02
10.35 11.73 9.49 9.99 6.46
5.40 4.84 4.64 4.50 3.10
17.22 14.95 14.06 14.13 9.93
2.95 2.47 2.31 2.44 1.76
15.5 13.1 12.5 13.0 9.5
5.48 4.49 4.35 4.47 3.32
1.77 1.49 1.47 1.49 1.18
7.03 5.78 5.71 5.82 4.42
1.30 1.07 1.07 1.07 0.82
8.41 6.92 7.04 7.04 5.34
1.82 1.46 1.49 1.50 1.14
5.18 4.27 4.33 4.27 3.26
0.80 0.64 0.66 0.66 0.50
5.15 4.17 4.37 4.19 3.19
0.76 0.63 0.66 0.64 0.48
4.04 3.49 3.37 3.46 2.50
0.26 0.27 0.22 0.22 0.16
0.87 0.58 0.42 0.77 0.42
0.25 0.23 0.20 0.21 0.13
0.10 0.09 0.08 0.08 0.06










Table 4-2. Continued


sample
Li
Sc
V
Cr
Co
Ni
Cu
Zn
Ga
Rb
Sr
Y
Zr
Nb
Cs
Ba
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
Hf
Ta
Pb
Th
U


266-10
8.51
46
347
150
44
57
65
99
21
1.51
121
47
156
3.87
0.02
11.08
5.17
16.46
2.77
14.7
5.00
1.66
6.40
1.19
7.71
1.61
4.68
0.70
4.55
0.70
3.90
0.27
0.68
0.22
0.09


265-35
7.85
42
342
127
41
57
62
96
18
1.18
119
45
130
3.15
0.02
8.69
4.66
14.43
2.38
12.7
4.47
1.48
5.82
1.10
7.24
1.54
4.44
0.68
4.45
0.69
3.54
0.21
0.46
0.19
0.08


265-31
7.91
42
342
112.
41
54
59
94
18
1.25
118
44
131
3.21
0.02
8.75
4.71
14.56
2.40
12.7
4.44
1.48
5.83
1.09
7.24
1.54
4.45
0.68
4.45
0.68
3.52
0.22
0.42
0.20
0.08


265-113
11.01
38
323
32.12
39
34
51
103
21
2.23
111
61
229
5.67
0.03
15.60
7.63
23.69
3.92
20.1
6.68
1.99
8.52
1.58
10.13
2.20
6.26
0.96
6.17
0.95
5.72
0.36
0.95
0.40
0.15


265-24
15.32
27
196
12.25
27
16
31
118
23
4.70
99
95
468
10.33
0.06
29.41
14.86
45.03
7.01
35.1
10.74
2.75
12.94
2.36
15.14
3.19
9.36
1.44
9.27
1.42
11.27
1.50
1.63
0.84
0.31


265-50
28.86
13
73
4.07
13
7
20
100
29
12.5
69
138
967
14.82
0.12
58.22
28.00
79.99
11.64
52.4
15.64
2.96
17.82
3.41
22.37
4.79
14.29
2.30
15.29
2.29
23.87
0.89
3.54
2.53
0.84


265-125
14.31
31
247
50.08
31
29
46
97
23
4.22
96
72
380
6.33
0.05
24.31
11.10
32.73
5.08
24.6
7.54
1.98
9.21
1.72
11.31
2.39
7.09
1.12
7.23
1.10
9.42
0.45
1.73
0.74
0.28


265-109
19.24
29
176
29.19
27
23
35
126
27
4.42
104
110
557
10.95
0.05
28.72
15.75
48.54
7.76
38.9
12.07
3.15
14.75
2.70
17.61
3.71
10.97
1.69
10.88
1.66
13.27
0.76
2.55
0.78
0.30


264-16
18.52
33
230
44.46
30
25
34
129
24
4.60
113
101
371
8.89
0.05
29.81
13.61
41.17
6.42
30.9
10.37
2.68
12.89
2.44
16.19
3.44
10.16
1.58
10.43
1.62
10.10
0.61
1.65
0.76
0.30


265-57
20.73
31
217
31.96
29
18
27
137
26
5.34
109
116
425
10.06
0.07
32.73
15.76
47.56
7.34
35.3
11.90
2.98
14.65
2.78
18.52
3.93
11.60
1.81
11.91
1.84
11.55
0.68
1.89
0.88
0.35


265-103
22.27
26
129
6.40
23
10
27
144
30
5.30
103
135
671
13.51
0.06
33.81
19.57
60.43
9.64
48.6
14.99
3.78
18.33
3.34
21.76
4.52
13.34
2.05
13.19
2.07
15.95
0.95
3.28
0.96
0.36


265-91
23.57
25
93
2.31
21
6
23
147
31
5.74
106
144
724
14.52
0.06
36.06
21.43
66.14
10.54
52.9
16.18
4.10
19.83
3.57
23.38
4.89
14.24
2.20
14.01
2.14
17.25
1.03
3.76
1.05
0.39


265-90
28.20
24
70
4.82
20
6
22
174
30
6.34
114
170
680
15.81
0.08
42.35
23.72
73.19
11.55
55.8
18.61
4.57
22.78
4.26
27.74
5.88
17.37
2.68
17.70
2.73
17.78
1.03
2.41
1.15
0.46










Table 4-2. Continued
sample 265-100 266-05 265-69 265-25 264-14 266-56 265-66 265-65 265-64 265-67 266-53 265-84
Li 27.14 20.40 26.34 28.11 26.38 27.40 36.88 31.68 33.61 31.78 30.17 28.59
Sc 24 26 22 19 24 17 26 17 20 18 15 15
V 100 216 180 87 160 93 184 122 140 121 61 102
Cr 9.51 9.77 16.05 9.15 40.69 5.34 15.33 12.94 12.27 12.40 3.81 1.75
Co 21 26 21 17 21 15 23 15 17 16 13 14
Ni 10 15 11 9 22 7 15 9 10 9 6 7
Cu 26 33 22 26 29 22 24 17 19 18 18 21
Zn 150 107 112 124 114 106 143 110 124 122 109 100
Ga 33 26 28 28 26 29 42 30 35 28 29 29
Rb 6.51 7.01 7.35 10.8 7.79 10.7 11.6 9.1 10.5 9.6 12.8 12.7
Sr 112 92 88 97 97 81 114 76 90 89 86 70
Y 153 105 114 154 124 133 164 132 148 142 157 133
Zr 721 645 605 881 542 901 945 735 842 622 856 968
Nb 16.24 10.90 11.39 17.36 11.40 15.17 16.79 12.98 14.81 12.97 15.52 14.61
Cs 0.07 0.08 0.08 0.14 0.10 0.11 0.12 0.11 0.12 0.12 0.14 0.13
Ba 41.77 38.30 40.22 58.57 43.51 53.60 62.57 50.53 57.07 52.86 64.61 59.79
La 23.67 17.91 18.93 28.81 19.39 25.36 28.95 23.25 26.31 23.43 28.53 27.27
Ce 72.67 52.40 55.75 84.42 56.99 73.59 84.22 67.48 76.53 67.97 82.17 77.81
Pr 11.30 7.89 8.40 12.25 8.47 10.86 12.53 10.06 11.37 10.05 11.87 11.21
Nd 56.3 37.2 39.7 54.9 39.1 49.6 58.3 46.4 52.8 45.9 53.7 50.9
Sm 17.27 11.43 12.26 16.73 12.38 14.91 17.07 14.09 15.37 14.62 16.25 14.32
Eu 4.20 2.60 2.81 3.31 2.75 3.07 3.78 3.00 3.35 3.12 3.18 2.80
Gd 20.71 13.66 14.69 19.55 14.82 17.58 20.38 16.83 18.13 17.43 19.00 16.58
Tb 3.79 2.53 2.76 3.68 2.86 3.25 3.83 3.19 3.43 3.36 3.65 3.09
Dy 24.73 16.76 18.27 24.22 19.28 21.34 25.32 21.08 22.81 22.85 24.21 20.31
Ho 5.24 3.60 3.98 5.17 4.14 4.60 5.42 4.58 4.88 4.92 5.19 4.33
Er 15.41 10.63 11.66 15.49 12.44 13.59 16.36 13.55 14.65 14.86 15.79 13.16
Tm 2.38 1.68 1.86 2.49 1.98 2.18 2.58 2.19 2.33 2.37 2.50 2.10
Yb 15.34 10.90 12.12 16.50 13.15 14.05 16.83 14.25 15.20 15.66 16.74 13.66
Lu 2.36 1.67 1.85 2.64 2.04 2.16 2.55 2.15 2.31 2.44 2.56 2.10
Hf 17.09 15.73 15.48 22.31 14.75 21.69 23.08 18.69 20.95 17.54 22.28 23.14
Ta 1.14 0.66 0.69 1.13 0.80 0.91 1.23 0.81 1.10 0.92 1.03 1.39
Pb 3.08 2.02 2.96 3.07 2.67 3.21 6.24 3.33 5.05 3.41 3.15 4.89
Th 1.17 1.31 1.35 2.31 1.28 2.08 2.00 1.74 1.82 1.64 2.29 2.28
U 0.45 0.46 0.46 0.89 0.51 0.70 0.71 0.59 0.65 0.64 0.86 0.82


138










Table 4-2. Continued
sample 265-63 265-85 265-94 264-09 265-70 265-42 265-83 265-95 265-40
Li 30.23 31.42 30.77 26.68 33.56 31.62 32.45 31.37 30.55
Sc 17 14 13 12 12 14 11 10 12
V 121 73 63 46 45 58 32 52 51
Cr 9.79 4.65 1.49 3.82 3.70 3.42 3.40 3.01 4.17
Co 15 12 11 10 10 11 8 8 10
Ni 8 7 5 6 5 6 5 5 5
Cu 17 19 17 16 16 15 14 15 17
Zn 113 108 105 89 106 119 103 98 103
Ga 31 28 30 28 29 30 29 30 30
Rb 9.5 13.8 12.4 12.9 15.0 13.7 15.5 12.5 12.4
Sr 76 81 68 78 78 83 76 61 73
Y 132 154 146 151 160 160 159 145 146
Zr 745 872 1050 824 934 816 922 985 1013
Nb 13.18 15.27 16.15 14.89 15.90 16.36 15.57 16.53 16.60
Cs 0.10 0.15 0.13 0.13 0.17 0.16 0.17 0.13 0.13
Ba 49.78 68.07 60.04 65.65 72.69 70.01 76.40 62.17 59.70
La 23.53 29.07 29.10 28.98 30.89 29.03 30.69 29.16 29.47
Ce 68.11 82.46 83.93 83.89 88.15 82.95 87.16 83.65 85.00
Pr 10.16 11.78 12.15 11.97 12.49 11.98 12.32 12.02 12.44
Nd 47.3 52.1 55.2 52.7 55.0 53.5 54.0 54.6 56.6
Sm 13.77 16.01 15.74 15.92 16.67 16.69 16.46 15.64 16.89
Eu 2.99 3.01 3.02 2.95 3.09 3.39 3.05 2.89 3.32
Gd 16.37 18.51 18.17 18.54 19.47 19.69 18.90 17.66 19.68
Tb 3.10 3.55 3.39 3.55 3.69 3.76 3.64 3.35 3.67
Dy 20.53 23.80 22.36 23.80 25.06 25.37 24.51 22.13 23.89
Ho 4.40 5.12 4.77 5.12 5.38 5.47 5.27 4.74 5.20
Er 13.19 15.50 14.37 15.66 16.42 16.48 16.18 14.61 15.35
Tm 2.10 2.49 2.29 2.52 2.63 2.64 2.62 2.32 2.49
Yb 13.54 16.69 14.81 16.84 17.54 17.56 17.54 15.12 16.19
Lu 2.06 2.58 2.30 2.61 2.73 2.75 2.72 2.32 2.42
Hf 18.62 22.96 24.87 22.68 24.80 22.24 24.75 24.51 24.97
Ta 0.99 1.05 1.18 1.04 1.10 1.12 1.08 1.23 1.02
Pb 5.82 3.59 5.75 2.76 3.84 3.47 3.80 4.12 3.64
Th 1.65 2.43 2.35 2.59 2.68 2.36 2.80 2.36 2.49
U 0.59 0.92 0.84 0.98 1.04 0.91 1.05 0.86 0.84


139









CHAPTER 1
INTRODUCTION

Mid-ocean ridges (MOR) are comprised of a series of segments that can be

subdivided at a variety of scales, ranging from tens of meters to hundreds of kilometers

between 1st order discontinuities marked by transform faults (Sempere and Macdonald,

1986; Macdonald et al., 1988). Overlapping spreading centers (OSC) are 2nd order

discontinuities that form between widely spaced transform faults on fast to intermediate

spreading ridges (Macdonald and Fox, 1983; Sempere and Macdonald, 1986, Carbotte

and Macdonald, 1992). These offsets provide both a physical and a geochemical

segmentation of the ridge, which may result from variations in mantle melting and/or

separation of crustal magma reservoirs between segments (e.g. Macdonald et al.,

1988).

Lavas erupted along fast to intermediate spreading centers, such as the northern

East Pacific Rise (EPR), may produce a range of basaltic lavas (e.g., Batiza and Niu,

1992), but they rarely erupt compositions with MgO concentrations <5 wt%. This

relatively limited compositional diversity compared to other tectonic settings is

commonly attributed to shallow-level fractional crystallization of primitive magmas within

an axial magma chamber buffered by relatively frequent recharge with more primitive

melts (Klein, 2005). Additionally, geochemical variations in MORB may result from

variable mantle melting parameters and/or mantle sources (Klein & Langmuir, 1987;

Langmuir et al., 1992).

In contrast, lavas erupted at ridge segment ends, such as an OSC, can have

highly variable compositions compared to a relatively limited range of basaltic

compositions erupted from magmatically robust segment centers (e.g. Christie and










Table 4-3. West limb maior element data 9N OSC


140


sample Rock Type
267-16 ferrobasalt
267-17 FeTi
267-18 ferrobasalt
267-19 ferrobasalt
267-20 FeTi
267-21 ferrobasalt
267-22 ferrobasalt
267-23 basaltic andesite
267-24 ferrobasalt
267-25 FeTi
267-26 ferrobasalt
267-27 ferrobasalt
267-29 ferrobasalt
267-30 basaltic andesite
267-32 ferrobasalt
267-33 ferrobasalt
267-34 ferrobasalt
267-35 FeTi
267-37 ferrobasalt
267-38 ferrobasalt
267-39 ferrobasalt
267-40 ferrobasalt
267-41 basaltic andesite
267-42 ferrobasalt
267-43 ferrobasalt
267-44 ferrobasalt
267-45 ferrobasalt
267-46 ferrobasalt
267-47 ferrobasalt
267-48 ferrobasalt
267-50 ferrobasalt
267-51 ferrobasalt
267-52 ferrobasalt
267-53 ferrobasalt


SiO2 TiO2
50.76 1.18
51.07 2.12
51.15 1.71
50.97 1.83
50.78 2.16
51.23 1.72
51.02 1.80
55.95 2.04
50.74 1.81
50.90 2.06
50.70 1.86
50.69 1.87
51.68 1.61
52.33 1.92
50.57 1.75
50.72 1.78
50.67 1.80
50.49 2.10
50.79 1.88
50.83 1.58
50.78 1.78
50.88 1.76
52.40 3.06
50.69 1.76
50.74 1.76
50.86 1.76
50.77 1.76
50.78 1.78
50.33 1.83
49.95 1.65
49.76 1.58
50.64 1.80
50.20 1.95
50.67 1.57


A1203
14.81
13.54
14.29
13.97
13.45
14.26
14.15
13.02
13.96
13.49
13.73
13.44
13.62
13.28
14.05
14.04
14.00
13.72
13.76
14.28
14.05
14.18
12.67
14.11
14.16
14.06
13.88
14.01
14.12
14.43
14.83
14.11
13.70
14.28


FeO MnO MgO
9.62 0.19 8.72
12.18 0.22 6.74
10.63 0.18 7.70
11.13 0.20 7.40
12.43 0.22 6.66
10.64 0.20 7.77
10.87 0.20 7.52
11.98 0.21 2.84
10.78 0.21 7.64
12.42 0.23 7.29
11.57 0.22 7.21
11.44 0.22 7.12
11.20 0.21 7.00
12.09 0.23 5.43
10.61 0.20 7.76
10.72 0.20 7.71
10.72 0.21 7.73
12.00 0.23 6.88
11.61 0.21 7.26
10.23 0.19 7.67
10.80 0.20 7.75
10.76 0.21 7.82
14.13 0.24 4.21
10.74 0.20 7.81
10.78 0.20 7.77
10.90 0.21 7.54
10.84 0.20 7.71
10.88 0.20 7.63
10.24 0.20 7.28
10.66 0.21 8.01
10.57 0.20 8.31
10.86 0.20 7.55
11.64 0.21 7.08
10.24 0.20 7.84


CaO Na20
12.51 2.06
10.70 2.98
11.18 2.70
10.95 2.78
10.47 2.93
11.14 2.74
10.85 2.84
6.59 3.70
10.85 3.00
10.84 2.59
10.77 2.87
10.60 2.82
10.45 2.80
9.06 3.25
10.89 2.99
10.89 2.99
10.85 2.98
10.83 3.19
11.03 2.90
11.76 2.98
11.05 3.00
10.88 3.00
7.83 3.77
10.99 3.02
10.97 3.03
11.11 2.99
11.01 2.95
11.04 2.98
11.38 2.89
11.73 2.62
11.69 2.63
11.08 2.96
10.97 3.12
11.78 2.73


K20
0.07
0.11
0.10
0.10
0.12
0.10
0.11
0.67
0.12
0.15
0.16
0.17
0.22
0.37
0.12
0.12
0.12
0.14
0.14
0.08
0.12
0.12
0.64
0.12
0.12
0.12
0.12
0.12
0.32
0.09
0.09
0.12
0.12
0.11


P205
0.13
0.23
0.17
0.20
0.23
0.18
0.21
0.73
0.21
0.24
0.22
0.25
0.19
0.48
0.20
0.21
0.18
0.22
0.21
0.14
0.20
0.21
0.46
0.20
0.17
0.19
0.19
0.21
0.27
0.21
0.17
0.21
0.22
0.17


Total
100.1
99.89
99.81
99.53
99.44
99.97
99.57
97.73
99.32
100.2
99.31
98.61
98.97
98.45
99.14
99.39
99.25
99.81
99.78
99.73
99.72
99.83
99.41
99.64
99.70
99.74
99.44
99.63
98.87
99.54
99.82
99.54
99.23
99.58










Table 4-3. Continued
sample Rock Type
267-54 ferrobasalt
267-55 ferrobasalt
267-56 FeTi
267-57 ferrobasalt
267-58 ferrobasalt
267-59 FeTi
267-60 FeTi
267-61 ferrobasalt
267-62 ferrobasalt
267-63 ferrobasalt
267-64 ferrobasalt
267-65 ferrobasalt
267-66 ferrobasalt
267-67 FeTi
267-68 ferrobasalt
267-69 FeTi


SiO2 TiO2 A1203 FeO MnO
49.98 1.71 14.19 10.68 0.20
50.30 1.54 14.48 10.02 0.20
50.18 2.22 13.05 13.11 0.24
49.82 1.36 14.93 9.35 0.18
49.98 1.82 14.25 10.24 0.21
49.67 2.00 14.29 10.64 0.20
50.14 2.12 13.13 12.54 0.23
49.46 1.56 14.79 10.50 0.20
50.70 1.35 14.64 9.44 0.18
50.13 1.43 13.96 9.89 0.20
49.72 2.40 14.67 10.76 0.19
50.97 1.72 14.00 10.91 0.21
49.91 2.11 13.27 12.34 0.22
50.70 2.05 13.44 12.03 0.23
50.66 2.02 13.58 11.94 0.21
50.80 2.04 13.55 12.01 0.21


MgO
7.65
7.98
6.42
8.35
7.22
6.99
6.73
8.17
8.34
8.10
6.46
7.50
6.76
6.87
6.95
6.97


CaO Na20 K20 P205 Total
11.52 2.91 0.11 0.18 99.13
11.83 2.71 0.11 0.17 99.33
10.06 3.09 0.15 0.26 98.78
12.37 2.55 0.14 0.16 99.20
11.52 2.88 0.32 0.27 98.70
11.16 3.10 0.36 0.26 98.67
10.35 2.97 0.14 0.24 98.61
11.62 2.62 0.09 0.18 99.20
11.98 2.46 0.08 0.14 99.30
12.11 2.41 0.09 0.16 98.47
10.21 3.19 0.59 0.41 98.59
11.05 2.66 0.12 0.18 99.33
10.15 2.87 0.13 0.23 97.99
10.27 2.91 0.12 0.21 98.84
10.33 2.86 0.12 0.22 98.89
10.35 2.88 0.12 0.21 99.14










Table 4-4. West limb trace element data 90N OSC
267- 267- 267- 267- 267- 267- 267- 267-
sample 16 18 23 62 63 64 68 69
Li 5.3 6.9 19.4 5.4 5.6 7.9 8.2 7.7
Sc 40 41 26 40 39 41 42 41
V 275 312 148 278 266 345 368 350
Cr 368 147 24 359 339 240 80 71
Co 43 40 21 40 38 43 43 42
Ni 117 64 15 72 66 113 48 47
Cu 146 64 27 78 73 66 56 56
Zn 73 86 117 75 73 96 100 96
Ga 15 17 25 15 15 25 19 19
Rb 0.52 1.02 8.91 0.89 0.98 11.13 1.01 1.15
Sr 87 109 102 96 96 327 116 106
Y 27 37 127 29 29 43 47 43
Zr 62 112 669 82 82 208 133 134
Nb 1.55 2.48 19.92 2.00 2.16 21.36 3.22 3.13
Cs 0.01 0.02 0.11 0.01 0.02 0.12 0.01 0.02
Ba 3.39 7.24 70.75 7.07 7.31 131.32 8.56 9.24
La 2.38 3.65 23.42 2.86 2.90 15.43 4.80 4.52
Ce 7.21 11.68 66.80 8.95 9.02 36.58 14.93 14.46
Pr 1.21 2.08 10.13 1.54 1.59 4.96 2.47 2.47
Nd 6.98 11.05 48.44 8.39 8.63 22.53 13.42 13.23
Sm 2.44 3.86 14.98 2.95 3.03 5.96 4.72 4.58
Eu 0.89 1.34 3.55 1.07 1.11 2.00 1.60 1.55
Gd 3.37 5.08 18.03 3.93 4.07 6.87 6.15 5.89
Tb 0.65 0.93 3.28 0.73 0.75 1.17 1.16 1.10
Dy 4.26 6.08 21.11 4.74 4.90 7.27 7.67 7.14
Ho 0.93 1.30 4.50 1.01 1.06 1.48 1.61 1.50
Er 2.68 3.69 12.96 2.90 2.99 4.18 4.68 4.36
Tm 0.41 0.56 2.02 0.44 0.46 0.62 0.71 0.66
Yb 2.68 3.60 12.92 2.83 2.93 3.93 4.68 4.23
Lu 0.41 0.55 1.98 0.43 0.44 0.60 0.72 0.65
Hf 1.74 2.86 16.00 2.09 2.17 4.70 3.56 3.41
Ta 0.10 0.17 1.11 0.14 0.15 1.37 0.22 0.22
Pb 0.13 0.52 2.02 0.24 0.41 1.86 0.44 0.39
Th 0.09 0.15 1.85 0.12 0.14 1.42 0.18 0.18
U 0.04 0.06 0.54 0.05 0.06 0.42 0.07 0.07


142










Table 4-5. West limb isotopic data 9N OSC
sample 267-18 267-23 267-62 267-63 267-64 267-69
208Pb/204Pb 37.736 37.846 37.748 37.737 38.022 37.738
2 sigma error 0.0018 0.0019 0.0020 0.0022 0.0017 0.0015
207Pb/204Pb 15.478 15.495 15.480 15.476 15.530 15.476
2 sigma error 0.0007 0.0007 0.0008 0.0008 0.0006 0.0006
206Pb/204Pb 18.253 18.368 18.258 18.256 18.590 18.263
2 sigma error 0.0008 0.0008 0.0010 0.0009 0.0006 0.0007
208Pb/206Pb 2.0674 2.0603 2.0674 2.0671 2.04529 2.0664
2 sigma error 0.00004 0.00003 0.00003 0.00003 0.00003 0.00003
207Pb/206Pb 0.8480 0.8435 0.8478 0.847 0.8354 0.8474
2 sigma error 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001
87Sr/86Sr 0.70249 0.70265 0.70254 0.70259 0.70282 0.70254
2 sigma error 0.00001 0.00001 0.00001 0.00002 0.00002 0.00002
143Nd/144Nd 0.513152 0.513139 0.513149 0.51315 0.513049 0.513144
2 sigma error 0.000007 0.000006 0.000004 0.000007 0.000004 0.000004
Eps Nd 10.0 9.8 10.0 10.0 8.0 9.8


143










Table 4-6. Major and trace element data from 8037'N EPR
3925-0 3925-2 3925-3 3926-1 3926-3 3926-4 3926-5 3927-1 3927-2
SiO2 51.97 50.7 50.64 50.59 55 57.14 50.73 50.47 50.74
TiO2 1.72 1.63 1.63 1.72 1.87 1.64 1.6 1.82 1.73
A1203 14.53 14.7 14.63 14.68 14.17 14.14 14.56 14.37 14.49
FeO 10.43 10.49 10.31 10.31 10.95 10.15 10.38 10.94 10.63
MnO 0.20 0.21 0.18 0.19 0.18 0.17 0.17 0.19 0.19
MgO 6.25 7.29 7.24 7.21 3.85 3.08 7.26 6.83 7.07
CaO 10.42 11.59 11.69 11.59 7.36 6.5 11.56 11.1 11.41
Na20 3.19 2.84 2.83 2.94 4.05 4.25 2.86 3 2.9
K20 0.33 0.16 0.16 0.25 0.69 0.87 0.17 0.2 0.18
P205 0.31 0.29 0.29 0.33 0.46 0.48 0.29 0.32 0.31
S02 0.24 0.26 0.29 0.25 0.22 0.18 0.26 0.28 0.26
Total 99.61 100.2 99.93 100.1 98.83 98.61 99.86 99.55 99.93
Li 9.08 7.64 6.03 17.72 6.24 6.57 6.53
Sc 40.38 39.55 42.32 25.02 42.47 41.93 41.19
V 309. 288 315 152 314 331 315
Cr 114 179 185 43.99 168 89.65 138
Co 39.47 38.39 41.32 23.96 41.38 40.96 39.94
Ni 48.26 49.41 53.77 17.92 51.88 46.03 48.29
Cu 65.95 71.34 76.86 41.13 76.57 71.19 70.79
Zn 90.55 82.35 81.56 110.08 112.00 86.75 82.43
Ga 17.75 16.36 15.78 21.32 15.84 16.52 15.81
Rb 4.92 3.67 2.08 14.11 1.95 3.00 2.40
Sr 140 127 131 119 129 148. 134
Y 53.42 45.73 35.37 117.23 35.66 38.69 36.61
Zr 211 171 110 550 125 122 112
Nb 7.47 5.72 4.19 19.41 3.94 5.85 4.73
Cs 0.06 0.05 0.03 0.16 0.03 0.04 0.03
Ba 38.11 28.50 20.10 99.52 18.91 30.14 23.83
La 8.91 7.22 4.66 23.07 4.47 5.91 5.02
Ce 24.28 19.91 13.20 61.72 12.77 16.24 14.11
Pr 3.69 3.04 2.12 8.97 2.05 2.51 2.25
Nd 17.93 15.03 10.88 41.80 10.72 12.67 11.54
Sm 5.66 4.85 3.69 12.61 3.63 4.14 3.85
Eu 1.71 1.50 1.33 3.01 1.29 1.45 1.38
Gd 6.97 6.09 4.81 14.87 4.69 5.25 4.96
Tb 1.31 1.13 0.89 2.74 0.87 0.96 0.92
Dy 8.44 7.38 5.75 17.82 5.68 6.29 5.98
Ho 1.79 1.57 1.21 3.79 1.21 1.32 1.25
Er 5.22 4.58 3.49 11.08 3.48 3.81 3.61
Tm 0.80 0.70 0.53 1.75 0.52 0.57 0.54
Yb 5.23 4.51 3.43 11.48 3.37 3.69 3.48
Lu 0.80 0.69 0.52 1.76 0.52 0.56 0.53
Hf 5.37 4.48 2.91 13.78 3.23 3.15 2.95
Ta 0.46 0.36 0.26 1.16 0.25 0.37 0.30
Pb 0.87 0.77 0.59 2.16 1.50 0.68 0.61
Th 0.75 0.59 0.30 2.15 0.29 0.42 0.34
U 0.26 0.21 0.10 0.74 0.10 0.14 0.12


144










Table 4-7. Radiogenic isotope ratios 8037'N EPR
Sample 3925-RO 3925-R2 3925-R3 3926_R4 3926-R5 3927-R1 3927-R2
208Pb/204Pb 37.957 37.938 37.976 37.911 37.965 37.992 37.983
2 sigma error 0.0020 0.0019 0.0036 0.0020 0.0017 0.0017 0.0023
207Pb/204Pb 15.510 15.508 15.513 15.505 15.514 15.515 15.517
2 sigma error 0.0008 0.0008 0.0015 0.0008 0.0007 0.0006 0.0008
206Pb/204Pb 18.468 18.445 18.484 18.437 18.463 18.4960 18.482
2 sigma error 0.0008 0.0008 0.0016 0.0008 0.0008 0.0007 0.0009
208Pb/206Pb 2.0553 2.0568 2.0546 2.0562 2.0562 2.0541 2.0551
2 sigma error 0.00004 0.00004 0.00005 0.00003 0.00003 0.00003 0.00005
207Pb/206Pb 0.83984 0.84077 0.83928 0.84093 0.84025 0.83881 0.83955
2 sigma error 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001
87Sr/86Sr 0.70262 0.70267 0.70257 0.70265 0.70258 0.70265 0.70260
2 sigma error 0.00001 0.00001 0.00002 0.00001 0.00002 0.00001 0.00001
143Nd/144Nd 0.51314 0.51315 0.51312 0.51313 0.51314 0.51314 0.51316
2 sigma error 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001
Eps Nd 9.7 9.9 9.5 9.6 9.8 9.9 10.2


145













































Figure 4-1. Bathymetric map of the northern EPR, including the location of 9N, 8037,
the Clipperton Transform and the Siqueiros Transform. (Data from
GeoMapApp; Carbotte et al., 2004). View looking north.












146











IT'ImW 104 120iV


Rock Type
* FeTi Basalt
* Andesite
o Basalt
* Basaltic Andesite
* Dacite
50 m counlour


Low 2164 m I -,


)i'\





I4II




-. Bi m. o
I ,


,. -A 0 ..e.
-, 'r- Ia I "





























MEDUSA2007 cruise, with warmer colors representing higher silica contents.
Colored outlines delineate regions discussed in the text.
104H h"" 10!:I" 0 70t 0 1'' 14 5r' 141'" 0"1'"W 141"T 0 10W 1 4130
Fgr4-.Btyeri ma fte9NOCwt 0 cnor.Crlsso h
loaino oksmlscletd sn h ao2RVdrn h













10418'W


104'16'W 10414'W 104"12'W
I A k


104*10'W


Anomalously
z Wide Melt Lens Z
ipDrirprimate -
Sxtei of Dacites

...i .




I




'.. ci t,'
1 Ponqinq N









3 4

J f<




104"1 8W 104*16"W 104614'W 104'i2'W 104*10'W


Figure 4-3. Bathymetric map of the 9oN OSC with 50m contours. The melt sills
underlying the east and west limbs of the OSC are shaded in gray (Kent et
al., 2000). Black bar represents the approximate extent of dacites.


148










104"19Q0"W 104I'18 W 104"17"W 10416'0"W 104"15'0"W 10414'0"W 104"13'0W 10412lCW 104110W 104'10(Y"W
I I I I I I I I I I


Ledgend
Rock Type
* FeTi Basalt
* Andesite
o Ferrobasalt
* Basaltic Andesite
* Dacite
Sidescan
High


N










Axial
Summit
Trough

































0 0.51 2 3 4
Kilometers


II w M M I I
10419'0W 1 ia 18'V 104-17-W im0o W 10415.0'W 104'140,W M 13'O W 104'12'0 10411'O"W W 10o0"W


Figure 4-4. Side scan sonar mosaic from data collected on the MEDUSA2007 cruise
using DSL-120A (White et al., 2009). Circles show the location of rock
samples collected using the Jason2 ROV during the MEDUSA2007 cruise,
with warmer colors representing higher silica contents. Yellow bar represents
the approximate extent of the axial summit trough (AST) in the region. South
of the AST, the neo-volcanic zone is difficult to identify.



149









Sinton, 1981; Langmuir et al., 1986). These rock types can range from typical MORB

lavas to iron-enriched FeTi basalts, andesites, and high-silica dacites. This

geochemical variability is commonly attributed to lower magma supply and cooler crust

(cold edge effect) at the end of ridge segments, which causes increased magmatic

fractionation prior to eruption (Christie and Sinton, 1981; Sinton et al., 1983; Perfit et al.,

1983; Perfit and Chadwick, 1998; Rubin and Sinton, 2007). While crystal fractionation

is undoubtedly a primary process in magma differentiation at MOR, it may not be the

only process involved in the petrogenesis of evolved lavas on MOR.

Elevated Cl concentrations in many MORB suggest that partial melting and

assimilation of seawater-altered material may be an important, but often overlooked,

process in MOR magmatism (Michael and Schilling, 1989; Michael and Cornell, 1998).

Evidence of these processes at MOR is supported by textural observations in ophiolites,

which show melting of overlying crustal material at the top of the magma chamber

(Coogan et al., 2003). Additionally, experimental results suggest that hydrous partial

melting of altered ocean crust can produce high-silica plagiogranite veins, which are a

small (<2%), but ubiquitous part of the ocean crust (Koepke et al., 2004; Koepke et al.,

2007). Despite the clear evidence of assimilation in these lavas, many models for the

magmatic plumbing system at MORs do not include partial melting and assimilation.

Dacites have been sampled from several different spreading centers; including,

the East Pacific Rise (EPR), the Juan de Fuca Ridge (JdFR), and the Galapagos

Spreading Center (GSC); however, there is no consensus on how these high-silica

lavas form on MOR. To answer these questions, we undertook a 35-day research cruise

(AT15-17) in the Spring of 2007 to the 9N OSC, during which we surveyed 200 sq. km.













High-P205

14





10


Basaltic
8 Andesites/


6 / Dacites Low-P205 Mixing
Andesites + QFM-1 F.C.
OFM F.C.

4 _L
0 1 2 3 4 5 6 7 8
MgO (wt%)


Figure 4-5. FeO versus MgO for glasses collected from the east limb of the 90N OSC.
Shaded regions discriminate between different rock types, which are
dominated by different petrologic processes. Black lines with x's represent
various bulk mixes of high-silica dacites and lower silica end-members. Red
and blue lines with crosses show liquid-lines of descent (calculated using
MELTS; Ghiorso and Sack, 1995) of two different oxygen fugacities (blue =
QFM-1 and red = QFM). The majority of lavas erupted at the OSC can be
explained by fractional crystallization or mixing various proportions of a high-
silica and basaltic end-member.


150










3.5 15

3- 14

2 51
N 1 .5 0- f

-1 2






S. 0.6



0 basalts
A 4 ,D
10 +-A








High P205
N dacites
F.C. QFM 0.6



c 0.4


F.C. QFM0.6F.
1 -- F.C. QFM-1 .







S2 4 8 2 4 6 8
MgO (wt%) MgO (wt%)

Figure 4-6. Major element variations versus MgO (wt%) for glasses collected from the
east limb of the 9oN OSC. Compositions are compared to two fractional
crystallization trends (calculated using MELTS; Ghiorso and Sack, 1995) with
the same parental composition but different oxygen fugacities. Black lines
with x's represent bulk mixing trends. The majority of east limb lavas can be
explained by either by fractional crystallization or by mixing of an evolved and
dacitic end-members.



















S0.25 \

&7\ -






0.05
0 1 2 3 4 5 6 7 8
MgO (wt%)

Figure 4-7. P205/TiO2 versus MgO (wt%) for east limb glasses. The high-P205 andesites
lie along calculated fractional crystallization trends, while the low-P205
andesites and dacites have lower P205/TiO2 ratios for a given MgO. Many of
the basaltic andesites can be explained by mixing of various end-members.


152









































0 200 400 600 800
Zr (ppm)


400 600 800
Zr (ppm)


Figure 4-8. Trace element concentrations versus Zr for glasses collected from the east
limb of the OSC. Not all concentrations can be explained by fractional
crystallization alone. Mixing of high-silica lavas with various ferrobasalts can
explain a wide range of compositions erupting at the OSC.


153


1000 1200




















































600 800 1000 1200
Zr


200 400 600
Zr


800 1000 12


Figure 4-9. Incompatible trace element ratios versus Zr for glasses erupted at the OSC.
Low U/Nb ratios in high-P205 andesites can be explained by fractional
crystallization, however, Zr/Nb ratios indicate another process, such as mixing
or assimilation, must be involved.


154


B.
S------











8


E 50
ci,
N 45
40

35
30

25



2.2

2

1.8

.0 1.6

S1.4

1.2

1


Em

$*










C.


200 400


n --- --- -- --- --- --




















A

: +

+ +
+ *


+ EPR
* East Limb
* West Limb -
8-3
0 8"37
A Siquerous


18.2 18.3 18.4
206Pb/204Pb


+
++ +
-f 4


18.2 18.3 18.4
206Pb/204Pb


18.5 18.6 18.7


18.5 18.6 18.7


Figure 4-10. Radiogenic isotope ratios showing the variation in sources along the
northern EPR, from 9050 N to the Siquerous Transform Fault. A) Pb/Pb data
of lavas collected south of the OSC, generally have more radiogenic
signatures than lavas collected to the north of the OSC. N-MORB from both
the east and west limb are similar to other N-MORB lavas erupted along the
EPR but the west limb are slightly more radiogenic. The west limb has also
erupted E-MORB lavas. B) Epsilon Nd versus Sr isotopes, showing the
variation in sources erupting at the EPR.


155


38.1


38


S37.9

N
L 378


CL 37.7


18.1


1553
15.52

15.51

155

15.49
15-18

15.47

15.46

15 45


I*l


---i


37/


0
0000




















-0

z+
,-


7.5

7 A
0.7024 0.7025 0.7026 0.7027
87Sr/86Sr


0.7028 0.7029 0.703


Figure 4-10. Continued.


156












3.5
3


25
C\ 1.5 0 "

1 A.
0.5
0
0 2 4 6
MgO (wt%)

14

12

10






2
0 2 4 6
MgO (wt%)


2 4 6
MgO (wt%)


17

16

15

S14
0
Q 13

12

11
C


2 4 6
MgO (wt%)


8 10


8 10


MgO (wt%)


8 10 0 2 4 6
MgO (wt%)


8 10


Figure 4-11. Major element concentrations and ratios versus MgO comparing the east
and west limb of the OSC. Samples collected from the EPR north of the OSC
(black crosses) and during the CHEPR cruise near the OSC and 837'(green
symbols) are shown for comparison. A) Major element variations versus
MgO. The east limb of the OSC has erupted a much wider compositional
range compared to the west limb but the west limb is, on average, more
primitive. Red line with crosses represents a calculated fractional
crystallization trend using a primitive EPR lava as a parent. B) Major element
ratios versus MgO. The west limb has erupted basaltic compositions with
higher K20/TiO2 than the east limb, consistent with E-MORB compositions.


50 a


EU. .. .:i
^*^-^ysU
f **


,P EPR
d% East Axis
West Axis
E o o c CHEPR

U .














West Axis
0.6 \ CHEPR


0 a
0.2
2 S, II a. t
0
0 2 4 6 8 10
MgO (wt%)

0.4
0.35
^ 0.3
7 U0.25 %, 0 .





0 2 4 6 8 10
MgO (wt%)
B

Figure 4-11. Continued.
0.15 io ..


0.05,---------^ -- ^ --
0 2 4 6 8 10
MgO (wt%)
B

Figure 4-11. Continued.


158











16

14

12

10
8
-j
6

4



0.4


E 0.3

0.2
D

0.1

400


300

a
a200


100


0


A











C




e
o











E 0 0
0% 0
0*








CC,
0



Oe






-c








i 3


100


300 400


100 200
Zr (ppm)


300 400


Figure 4-12. Trace element concentrations versus Zr comparing east and west limb
basalts and basaltic andesites. The west limb lavas are more primitive
relative to the east limb lavas.


159


0 100 200
Zr (ppm)


1.5



1



S0.5






80


B






0*





D East Axis
West Axis
0 EPR Lavas




S




0o
*
*






c o

*









with DSL-120A for side scan sonar, mapped and sampled with the ROV Jason II (~7000

digital photographs, recovery of >280 rock samples, sampling of hydrothermal vent

waters and biota); and took photographs with the WHOI TowCam (~10,000 digital

photographs, CTD and MAPR data), which constitute one of the most detailed data

sets from an OSC. The compositions of lavas recovered from the OSC exhibit

remarkable diversity, ranging from basalt to dacite: 33% of OSC lavas have SiO2 > 52

wt.%, compared to ~3% for ocean ridge basalts worldwide. This study utilizes this data

to explore the roles of crystal fractionation, partial melting and assimilation in the

petrogenesis of high-silica lavas on MOR. In addition, geochemical data are used to

determine which processes are involved in the formation of the range of compositions

(basalts, ferrobasalts, FeTi basalts, basaltic andesites, andesites and dacites) erupted

at the 9N OSC to better understand the anatomy of a 2nd order MOR discontinuity.










Andesite Comparison


RbBaTh U NbTa LaCe SrNdZr HfSmEu Dy Y YbLu


Figure 4-13. Primitive mantle normalized diagram showing variations in andesites and
basaltic andesites erupted at the OSC. Red line is a MORB from the east
limb. Blue lines are the high-P205 andesites from the east limb. Green line
is the andesite erupted on the west limb. Gray lines are andesites and
basaltic andesites erupted on the east limb. The west limb andesite and the
high-P205 andesite lack the distinct negative Nb, Ta anomaly and do not
have as high U and Th compared to the other east limb lavas.


160


r nn P









CHAPTER 5
CONCLUSIONS

The majority of eruptions at spreading centers produce basalts with relatively

limited chemical variability; however, compositions ranging from basalts to dacites have

been sampled at ridge segment ends. We have documented the eruption of high -silica

lavas on the propagating eastern limb of the 9N overlapping spreading center (OSC)

on the East Pacific Rise. The dacites, which have erupted on several other ridges,

appear to represent an end-member composition that shows similar major element

trends and incompatible trace element enrichments, suggesting similar processes

controlled their petrogenesis.

The formation of highly evolved lavas on MOR requires a combination of partial

melting, assimilation and crystal fractionation. The highly enriched incompatible trace

element signatures cannot be produced through crystal fractionation alone and appears

to require partial melting of altered ocean crust. EC-AFC modeling suggests significant

amounts (>75%) crystallization of a MORB parent magma and modest amounts (5-

20%) of assimilation of hydrothermally altered ocean crust can produce geochemical

signatures consistent with dacite compositions. The AFC process explains trace

element abundances in high-silica lavas and accounts for several major and minor

element concentrations (i.e. A1203, K20 and CI).

The formation of dacitic lavas on MOR appears to require a unique tectono-

magmatic setting, where episodic magma supply allows for extensive crystal

fractionation, partial melting and assimilation of altered crustal material. These

conditions are met in regions of ridge propagation, such as OSC and propagating ridge

tips, where down axis diking allows for episodic injection of magma into older, altered









ocean crust. Here, the magma undergoes extensive crystallization without repeated

replenishment, creating enough latent heat of crystallization to melt and assimilate

surrounding wall rock.

Variations in volatile concentrations and 5180 in 9N OSC lavas also suggest that

the OSC magmas have experienced assimilation during their petrogenesis, with the

most extreme signatures observed in high-silica andesites and dacites and little

evidence in basaltic lavas. H20 concentrations are up to two times higher in dacitic

lavas compared to calculated fractional crystallization trends, whereas Cl has excesses

of seven to ten times predicted values. 6180 values are on average ~1%o lower than

ratios expected from fractional crystallization of ferromagnesian silicates and Fe-Ti

oxide phases, consistent with assimilation of an additional component or components.

The source of the excess H20 and Cl and low 6180 values is partially melted,

hydrothermally altered oceanic crust. Vapor saturation pressures calculated from H20-

CO2 data suggest that assimilation most likely occurs at the top of the melt lens, which

at the 9N OSC, corresponds approximately to the base of the sheeted dikes.

The distribution of evolved lavas and E-MORB lavas across the OSC is not

symmetric, suggesting that the 2nd order discontinuity represents a division in the

magmatic plumbing system of the EPR. E-MORB lavas are only observed on the dying

western limb and overall the lavas are less evolved than the adjacent eastern limb. We

suggest that the lower magma supply at the west limb allows for the preservation and

eruption of E-MORB compositions, whereas the more robust magmatic system on the

east limb overwhelms this signature. N-MORB lavas on the west limb have more

radiogenic Pb and Sr and less radiogenic Nd compared to east limb N-MORB lavas.


162









Lavas erupted to the south of the OSC (8037'N) also have more radiogenic Pb and Sr

isotope ratios. This suggests a slightly different mantle is feeding this section of the

EPR and that this large OSC provides a fundamental division between mantle sources

beneath the ridge axis.


163









LIST OF REFERENCES


Agrinier, P., Hekinian, R., Bideau, D. & Javoy, M. (1995). O and H stable isotope
compositions of oceanic crust and upper mantle rocks exposed in the Hess Deep
near the Galapagos Triple Junction. Earth and Planetary Science Letters 136,
183-196.

Alt, J., Honnorez, J., Laverne, C. & Emmerman, R. (1986). Hydrothermal Alteration of a
1 km Section Through the Upper Oceanic Crust, Deep Sea Drilling Project Hole
504B: Mineralogy, Chemistry, and Evolution of Seawater-Basalt Interations.
Journal of Geophysical Research 91, 10309-10335.

Alt, J., Laverne, C., Vanko, D. A., Tartarotti, P., Teagle, D. A. H., Bach, W., Zuleger, E.,
Erzinger, J., Honnorez, J., Pezard, P., Becker, K., Salisbury, M. H. & Wilkens, R.
H. (1996). Hydrothermal Alteration of a Section of Upper Oceanic Crust in the
Eastern Equatorial Pacific: A synthesis of results from site 504 (DSDP Legs 69,
70, and 83, and ODP Legs 111, 137, 140, and 148). Proceedings of the Ocean
Drilling Program, Scientific Results 148, 417-434.

Alt, J. & Teagle, D. A. H. (eds.) (2000). Hydrothermal alteration and fluid fluxes in
ophiolites and oceanic crust. Boulder, CO: Geological Society of America

Alt, J. C. & Teagle, D. A. H. (2003). Hydrothermal alteration of upper oceanic crust
formed at a fast-spreading ridge: mineral, chemical, and isotopic evidence from
ODP Site 801. Chemical Geology 201, 191-211.

Anderson, A. T., Clayton, R. N. & Mayeda, T. K. (1971). Oxygen isotope thermometry of
mafic igneous rocks. Journal of Geophysical Research 85, 715-729.

Aumento, F. (1969). Diorites from the mid-Atlantic ridge at 450N. Science 165, 1112-
1113.

Bachmann, O. & Bergantz, G. W. (2004). On the Origin of Crystal-poor Rhyolites:
Extracted from Batholithic Crystal Mushes. Journal of Petrology 45, 1565-1582.

Barker, A. K., Coogan, L. A. & Gillis, K. M. (2008). Strontium isotope constraints on fluid
flow in the sheeted dike complex of fast spreading crust: Pervasive fluid flow at
Pito Deep. Geochemistry Geophysics Geosystems 9, 1019.

Batiza, R. & Niu, Y. (1992). Petrology and Magma Chamber Processes at the East
Pacific Rise ~ 9030'N. Journal of Geophysical Research 97, 6779-6797.

Bazin, S., Harding, A. J., Kent, G. M., Orcutt, J. A., Tong, C. H., Pye, J. W., Singh, S.
C., Barton, P. J., Sinha, M. C., White, R. S., Hobbs, R. W. & Van Avendonk, H. J.
A. (2001). Three-dimensional shallow crustal emplacement at the 903' N
overlapping spreading center on the East Pacific Rise: Correlations between


164









magnetization and tomographic images. Journal of Geophysical Research 106,
16,101-116,117.

Beard, J. S. & Lofgren, G. E. (1991). Dehydration melting and water-saturated melting
of basaltic and andesitic greenstones and amphibolites at 1, 3, and 6.9 kb.
Journal of Petrology 32, 365-401.

Beccaluva, L., Chinchilla-Chaves, A. L., Coltorti, M., Giunta, G., Siena, F. & Vaccaro, C.
(1999). Petrological and structural significance of the Santa Elena-Nicoya
ophiolitic complex in Costa Rica and geodynamic implications. Contributions to
Mineralogy and Petrology 11, 1091-1107.

Bedard, J. H., Hebert, R. & Berclaz Alain, V., Veronika (eds.) (2000). Syntexis and the
genesis of lower oceanic crust. Boulder, CO: Geological Society of America

Berndt, M. E. & Seyfreid, W. E. J. (1990). Boron, bromine, and other trace elements as
clues to the fate of chlorine in mid-ocean ridge vent fluids. Geochemica et
Cosmochimica Acta 54, 2235-2245.

Bohnenstiehl, D. R., Tolstoy, M., Fox, C. G., Dziak, R. P., Chapp, E., Fowler, M., Haxel,
J., Fisher, C., Van Hilst, C., Laird, R., Collier, R., Cowen, J. P., Lilley, M., Simons,
K., Carbotte, S. M., Reynolds, J., R. & Langmuir, C. H. (2003). Anomalous
Seismic Activity at 8037-42'N on the East Pacific Rise: Hydroacoustic Detection
and Site Investigation. Ridge 2000 Newsletter 1, 18-20.

Bohrson, W. A. & Reid, M. R. (1997). Genesis of Silicic Peralkaline Volcanic Rocks in
an Ocean Island Setting by Crustal Melting and Open-system Processes:
Socorro Island, Mexico. Journal of Petrology 38, 1137-1166.

Bohrson, W. A. & Reid, M. R. (1998). Genesis of Evolved Ocean Island Magmas by
Deep- and Shallow-Level Basement Recycling, Socorro Island, Mexico:
Constraints from Th and other Isotope Signatures. Journal of Petrology 39, 995-
1008.

Bohrson, W. A. & Spera, F. (2001). Energy-Constrained Open-System Magmatic
Processes II: Application of Energy Constrained Assimilation-Fractional
Crystallization (EC-AFC) Model to Magmatic Systems. Journal of Petrology 42,
1019-1041.

Bowen, N. L. (1928). The evolution of igneous rocks: Princeton University Press.

Brophy, J. G. (2009). La-SiO2 and Yb-SiO2 systematics in mid-ocean ridge magmas:
implications for the origin of oceanic plagiogranite. Contributions to Mineralogy
and Petrology 158, 99-111.


165









Bryan, W. B. & Moore, J. G. (1977). Compositional variations of young basalts in the
Mid-Atlantic Ridge rift valley near 36049' N. Geological Society of America
Bulletin 88, 556-570.

Byerly, G. R. (1980). The nature of differentiation trends in some volcanic rocks from the
Galapagos Spreading Center. Journal of Geophysics 85, 3797-3810.

Byerly, G. R. & Melson, W. G. (1976). Rhyodacites, andesites, ferro-basalts and ocean
tholeiites from the Galapagos spreading center. Earth and Planetary Science
Letters 30, 215-221.

Canales, P. J., Detrick, R. S., Carbotte, S. M., Kent, G. M., Diebold, J. B., Harding, A. J.,
Babcock, J., Nedimovie, M. R. & van Ark, E. (2005). Upper crustal structure and
axial topography at intermediate spreading ridges: Seismic constraints from the
southern Juan de Fuca Ridge. Journal of Geophysical Research 110, 1-27.

Carbotte, S. M., Arko, R., Chayes, D. N., Haxby, W., Lehnert, K., O'Hara, S., Ryan, W.
B. F., Weissel, R. A., Shipley, T., Gahagan, L., Johnson, K. T. M. & Shank, T. M.
(2004a). New Integrated Data Management System for Ridge2000 and
MARGINS Research. EOS, Transactions American Geophysical Union 85(51),
553.

Carbotte, S. M. & Macdonald, K. C. (1992). East Pacific Rise 80-10030'N: Evolution of
ridge segments and discontinuities from SeaMARC II and three-dimensional
magnetic studies. Journal of Geophysical Research 97, 6959-6982.

Carbotte, S. M., Small, C. & Donnelly, K. (2004b). The influence of ridge migration on
the magmatic segmentation of mid-ocean ridges. Nature 429, 743-746.

Casey, J. F. (ed.) (1997). Comparison of major- and trace-element geochemistry of
abyssal peridotites and mafic plutonic rocks with basalts from the MARK region
of the mid-Atlantic ridge. College Station: Ocean Drilling Program.

Chadwick, J., Perfit, M. R., Ridley, W. I., Jonasson, I. R., Kamenov, G., Chadwick, W.
W. J., Embley, R. W., le Roux, P. J. & Smith, M. C. (2005). Magmatic effects of
the Cobb hot spot on the Juan de Fuca Ridge. Journal of Geophysical Research
110, 1-16.

Christie, D. M. & Sinton, J. M. (1981). Evolution of abyssal lavas along propagating
segments of the Galapagos Spreading Center. Earth and Planetary Science
Letters 56, 321.

Clague, D. A. & Bunch. (1976). Formation of ferrobasalt at East Pacific mid ocean
spreading centers. Journal of Geophysical Research 81, 4247-4256.


166









Coleman, R. G. & Donato, M. M. (1979). Oceanic plagiogranite revisited: Elsevier,
Amsterdam.

Coogan, L. A. (2003). Contaminating the lower crust in the Oman ophiolite. Geology 31,
1065-1068.

Coogan, L. A., Gillis, K. M., MacLeod, C. J., Thompson, G. M. & Hekinian, R. (2002).
Petrology and geochemistry of the lower ocean crust formed at the East Pacific
Rise and exposed at Hess Deep: A synthesis and new results. Geochemistry
Geophysics Geosystems 3, 1-30.

Coogan, L. A., Banks, G. J., Gillis, K. M., MacLeod, C. J. & Pearce, J. A. (2003a).
Hidden melting signatures recorded in the Troodos ophiolite plutonic suite:
evidence for widespread generation of depleted melts and intra-crustal melt
aggregation. Contributions to Mineralogy and Petrology 144, 484-505.

Coogan, L. A., Mitchell, N. C. & O'Hara, M. J. (2003b). Roof assimilation at fast
spreading ridges: An investigation combining geochemical, and field evidence.
Journal of Geophysical Research 108, 1-14.

Cotsonika, L. (2006). Petrogenesis of Andesites and Dacites from the Southern Juan
De Fuca Ridge. Department of Geological Sciences. Gainesville: University of
Florida, 188.

Danyushevsky, L. V. (2001). The effect of small amounts of H20 on crystallization of
mid-ocean ridge and backarc basin magmas. Journal of Volcanology and
Geothermal Research 110, 265-280.

De Paolo, D. J. (1981). Trace element and isotopic effects of combined wallrock
assimilation and fractional crystallization. Earth and Planetary Science Letters
53, 189-202.

Delacour, A., Fruh-Green, G. L., Frank, M., Gutjahr, M. & Kelley, D. S. (2008). Sr- and
Nd-isotope geochemistry of the Atlantis Massif (300N, MAR): Implications for fluid
fluxes and lithospheric heterogeneity. Chemical Geology 254, 19-35.

Detrick, R. S., Buhl, P., Vera, E., Mutter, J., Orcutt, J. A., Madsen, J. & Brocher, T.
(1987). Multichannel seismic imaging of a crustal magma chamber along the
East Pacific Rise. Nature 326, 35-41.

Dick, J. B., Natland, J. H., Alt, J., Bach, W., Bideau, D., Gee, J. S., Haggas, S.,
Hertogen, J. G. H., Hirth, G., Holm, P. M., Ildefonse, B., Iturrino, G. J., John, B.
E., Kelley, D. S., Kikawa, E., Kingdon, A., le Roux, P. J., Maeda, J., Meyer, P. S.,
Miller, D. J., Naslund, H. R., Niu, Y., Robinson, P. T., Snow, J., Stephen, R. A.,
Trimby, P. W., Worm, H. U. & Yoshinobu, A. (2000). A long in situ section of the









lower ocean crust: results of ODP Leg 176 drilling at the Southwest Indian Ridge.
Earth and Planetary Science Letters 179, 31-51.

Dixon, J. E. & Pan, V. (1995). Determination of the molar absorptivity of dissolved
carbonate in basanitic glass. American Mineralogist 80, 1339-1342.

Dixon, J. E., Stolper, E. M. & Delaney, J. R. (1988). Infrared spectroscopic
measurements of CO2 and H20 in Juan de Fuca Ridge basaltic glasses. Earth
and Planetary Science Letters 90, 87-104.

Dixon, J. E., Stolper, E. M. & Holloway, J. R. (1995a). An Experimental Study of Water
and Carbon Dioxide Solubilities in Mid-Ocean Ridge Basaltic Liquids. Part I:
Calibration and Solubility Models. Journal of Petrology 36, 1607-1631.

Dixon, J. E., Stolper, E. M. & Holloway, J. R. (1995b). An Experimental Study of Water
and Carbon Dioxide Solubilities in Mid-Ocean Ridge Basaltic Liquids. Part II:
Applications to Degassing. Journal of Petrology 36, 1633-1646.

Donelly, K. E., Goldstein, S. J., Langmuir, C. H. & Spiegelman, M. (2004). Origin of
enriched ocean ridge basalt and implications for mantle dynamics. Earth and
Planetary Science Letters 226, 347-366.

Dunn, R. A., Toomey, D. R., Detrick, R. S. & Wilcock, W. S. D. (2001). Continuous
Mantle Melt Supply beneath an Overlapping Spreading Center on the East
Pacific Rise. Science 291, 1955-1958.

Eiler, J. M. (2001). Oxygen Isotope Variations of Basaltic Lavas and Upper Mantle
Rocks. Reviews in Mineraology and Geochemistry 43, 319-364.

Eiler, J. M., Farley, K. A., Valley, J., Hofmann, A. W. & Stolper, E. M. (1996). Oxygen
isotope constraints on the sources of Hawaiian volcanism. Earth and Planetary
Science Letters 144, 453-468.

Embley, R. W., Chadwick, W. W. J., Clague, D. A. & Stakes, D. S. (1999). 1998
eruption of Axial Volcano: Multibeam anomalies and seafloor observations.
Geophysical Research Letters 26, 3425-3428.

Embley, R. W., Chadwick, W. W. J., Perfit, M. R. & Baker, E. T. (1991). Geology of the
northern Cleft segment, Juan de Fuca Ridge: Recent lava flows, sea-floor
spreading, and the formation of megaplumes. Geology 19, 771-775.

Embley, R. W., Jonasson, I. R., Perfit, M. R., Franklin, J. M., Tivey, M. A., Malahoff, A.,
Smith, M. F. & Francis, T. J. G. (1988). Submersible investigations of an extinct
hydrothermal system on the Galapagos Ridge: Sulfide mound, stockwork zone,
and differentiated lavas. Canadian Mineralogy 26, 1-35.


168









Embley, R. W. & Wilson, D. (1992). Morphology of the Blanco Transform Fault zone-NE
Pacific: Implications for its tectonic evolution. Marine Geophysical Research 14,
25-45.

Fornari, D. J. (2003). A new deep-sea towed digital camera and multi-rock coring
system. EOS, Transactions American Geophysical Union 84, 69-76.

Fornari, D. J., Perfit, M. R., Malahoff, A. & Embley, R. W. (1983). Geochemical studies
of abyssal lavas recovered by DSRV Alvin from the Eastern Galapagos Rift, Inca
Transform, and Ecuador Rift 1. Major element variation in natural glasses and
spatial distribution of lavas. Journal of Geophysical Research 88, 10159-10529.

Fundis, A., Soule, A. S., Fornari, D. J. & Perfit, M. R. (2010). Paving the seafloor:
volcanic emplacement processes during the 2005-06 eruption at teh fast-
spreading East Pacific Rise, 9050'N. Geochemistry Geophysics Geosystems in
press.

Garcia, M. O., Muenow, D. W., Aggrey, K. E. & O'Neil, J. R. (1998). Major element,
volatile and stable isotope geochemistry of Hawaiian submarine tholeiitic
glasses. Journal of Geophysical Research 94, 10,525-510,538.

Gautason, B. & Muehlenbach, K. (1998). Oxygen isotopic fluxes associated with high-
temperature processes in the rift zones of Iceland. Chemical Geology 145, 275-
286.

Ghiorso, M. S. & Sack, R. O. (1995). Chemical Mass Transfer in Magmatic Processes.
IV. A Revised and Internally Consistent Thermodynamic Model for the
Interpolation and Extrapolation of Liquid-Solid Equilibria in Magmatic Systems at
Elevated Temperatures and Pressures. Contributions to Mineralogy and
Petrology 119, 197-212.

Gillis, K. M. (2008). The roof of an axial magma chamber: A hornfelsic heat exchanger.
Geology 36, 299-302.

Gillis, K. M. & Coogan, L. A. (2002). Anatectic Migmatites from the Roof of an Ocean
Ridge Magma Chamber. Journal of Petrology 43, 2075-2095.

Gillis, K. M., Coogan, L. A. & Chaussidon, M. (2003). Volatile element (B, CI, F)
behaviour in the roof of an axial magma chamber from the East Pacific Rise.
Earth and Planetary Science Letters 213, 447-462.

Gillis, K. M., Muehlenbach, K., Stewart, M., Gleeson, T. & Karson, J. A. (2001). Fluid
flow patterns in fast spreading East Pacific Rise crust exposed at Hess Deep.
Journal of Geophysical Research 106, 26311-26329.


169









CHAPTER 2
DACITE PETROGENESIS ON MID-OCEAN RIDGES: EVIDENCE FOR OCEANIC
CRUSTAL MELTING AND ASSIMILATION

Abstract

While the majority of eruptions at oceanic spreading centers produce lavas with

relatively homogeneous mid-ocean ridge basalt (MORB) compositions, the formation of

tholeiitic andesites and dacites at mid-ocean ridges (MOR) is a petrologic enigma.

Eruptions of MOR high-silica lavas are typically associated with ridge discontinuities and

have produced regionally significant volumes of lava. Andesites and dacites have been

observed and sampled at several different locations along the global MOR system;

including propagating ridge tips at ridge-transform intersections on the Juan de Fuca

Ridge and eastern Galapagos spreading center, and at the 9N overlapping spreading

center on the East Pacific Rise. Despite the formation of these lavas at various different

ridges, MOR dacites show remarkably similar major element trends and incompatible

trace element enrichments, suggesting that similar processes are controlling their

chemistry. Although most geochemical variability in MOR basalts is consistent with low-

pressure fractional crystallization of various mantle-derived parental melts, our

geochemical data from MOR dacitic glasses suggest that contamination from a

seawater-altered component is important in their petrogenesis. MOR dacites are

characterized by elevated U, Th, Zr, and Hf, low Nb and Ta concentrations relative to

the rare earth elements (REE) and A1203, K20, and Cl concentrations that are higher

than expected from low-pressure fractional crystallization alone. Petrologic modeling of

MOR dacites suggests that partial melting and assimilation are both integral to their

petrogenesis. Extreme fractional crystallization of a MORB parent combined with partial

melting and assimilation of amphibole-bearing altered crust produces a magma with









Gillis, K. M. & Roberts, M. D. (1999). Cracking at the magma-hydrothermal transition:
Evidence from the Troodos ophiolite, Cyprus. Earth and Planetary Science
Letters 169, 227-244.

Goss, A., Perfit, M., Ridley, W. I., Rubin, K. H., Kamenov, G., Soule, A. S., Fundis, A. &
Fornari, D. J. (2010). Geochemistry of lavas from the 2005-2006 eruption at the
East Pacific Rise, 9046'N-9056'N: Implications for ridge crest plumbing and
decadal changes in magma chamber compositions. Geochemistry Geophysics
Geosystems 11, 1-35.

Grimes, C. B., Ushikubo, T., John, B. E. & Valley, J. (2010a). Uniformly mantle-like
5180 in zircons from oceanic plagioranites and gabbros. Contributions to
Mineralogy and Petrology in press, 1-21.

Grimes, C. B., Ushikubo, T. & Valley, J. W. (2010b). Low 5180 (Zrc) in plagiogranites at
Oman: evidence for remelting. Geochemica et Cosmochimica Acta 74, A355.

Grove, T. L., Kinzler, R. J. & Bryan, W. B. (1992). Fractionation of mid-ocean ridge
basalts. Geophysical Monograph 71, 281-310.

Haase, K. M., Stroncik, N. A., Hekinian, R. & Stoffers, P. (2005). Nb-depleted andesites
from the Pacific-Antarctic Rise as analogs for early continental crust. Geology 33,
921-924.

Hacker, B. R. (1990). Amphibolite-facies to granulite-facies reactions in experimentally
deformed, unpowered amphibolite. American Mineralogist 75, 1349-1361.

Hanson, G. N. (1977). Geochemical evolution of the suboceanic mantle. Journal of
Geological Society of London 134, 235-260.

Harding, A. J., Kent, A. J. R. & Orcutt, J. A. (1993). A multichannel seismic investigation
of upper crustal structure at 9N on the East Pacific Rise: Implications for crustal
accretion. Journal of Geophysical Research 98, 13925-13944.

Hart, S. R., Schilling, J. G. & Powell, J. L. (1973). Basalts from Iceland and along
Reykjanes Ridge -- Sr isotope geochemistry. Nature 246, 104-107.

Hey, R. N., Duennebier, F. K. & Morgan, J. W. (1980). Propagating rifts on mid ocean
ridges. Journal of Geophysical Research 85, 3647-3658.

Hey, R. N., Johnson, G. L. & Lowrie, A. (1977). Recent tectonic evolution of the
Galapagos area and plate motions in the East Pacific. Geological Society of
America Bulletin 88, 1385-1403.


170









Johannes, W. & Koepke, J. (2001). Incomplete reaction of plagioclase in experimental
dehydration melting of amphibolite. Australian Journal of Earth Science 48, 581-
590.

Johnson, E. R., Wallace, P., Delgado Granados, H., Manea, V. C., Kent, A. J. R.,
Bindeman, I. N. & Donegan, C. S. (2009). Subduction-rlated Volatile Recycling
and Magma Generation beneath Central Mexico: Insights from Melt Inclusions,
Oxygen Isotopes and Geodynamic Models. Journal of Petrology 50, 1729-1764.

Juster, T. C., Grove, T. L. & Perfit, M. R. (1989). Experimental Constraints on the
Generation of FeTi Basalts, Andesites, and Rhyodacites at the Galapagos
Spreading Center, 850W and 950W. Journal of Geophysical Research 94, 9251-
9274.

Kamenov, G., Saunders, A. D. & Hames, W. E. (2007). Mafic Magmas as Sources for
Gold in Middle-Miocene Epithermal Deposits of Northern Great Basin, USA:
Evidence from Pb Isotopic Compositions of Native Gold. Economic Geology 102,
1191-1195.

Karsten & Delaney, J. R. (1989). Hot spot-ridge crest convergence in the northeast
Pacific. Journal of Geophysical Research 94, 700-712.

Kelley, D. S. & Delaney, J. R. (1987). Two-phase separation and fracturing in mid-
ocean ridge gabbros at temperatures greater than 7000C. Earth and Planetary
Science Letters 83, 53-66.

Kent, A. J. R., Harding, A. J. & Orcutt, J. A. (1993). Distribution of magma beneath the
East Pacific Rise near 9003'N overlapping spreading center from forward
modeling of CDP data. Journal of Geophysical Research 98, 13971-13995.

Kent, A. J. R., Norman, M. D., Hutcheon, I. D. & Stolper, E. M. (1999). Assimilation of
seawater-derived components in an oceanic volcano: evidence from matrix
glasses and glass inclusions from Loihi seamount, Hawaii. Chemical Geology
156, 299-319.

Kent, A. J. R., Peate, D. W., Newman, S., Stolper, E. M. & Pearce, J. A. (2002).
Chlorine in submarine glasses from the Lau Basin: seawater contamination and
constraints on the composition of slab-derived fluids. Earth and Planetary
Science Letters 202, 361-377.

Kent, G. M., Singh, S. C., Harding, A. J., Sinha, M. C., Orcutt, J. A., Barton, P. J., White,
R. S., Bazin, S., Hobbs, R. W., Tong, C. H. & Pye, J. W. (2000). Evidence from
three-dimensional seismic reflectivity images for enhanced melt supply beneath
mid-ocean ridge discontinuities. Nature 406, 614-618.









Klein, E. M. (ed.) (2005). Geochemistry of the Igneous Oceanic Crust. Oxford: Elsevier-
Pergamon.

Klein, E. M. & Langmuir, C. H. (1987). Global correlations of ocean ridge basalt
chemistry. Journal of Geophysical Research 92, 8089-8115.

Koepke, J., Berndt, J., Feig, S. T. & Holtz, F. (2007). The formation of SiO2-rich melts
within the deep oceanic crust by hydrous partial melting of gabbros.
Contributions to Mineralogy and Petrology 153, 67-84.

Koepke, J., Feig, S. T., Snow, J. & Freise, M. (2004). Petrogenesis of oceanic
plagiogranites by partial melting of gabbros: an experimental study. Contributions
to Mineralogy and Petrology 146, 414-432.

Kurras, G. J., Fornari, D. J., Edwards, M. F., Perfit, M. R. & Smith, M. C. (2000).
Volcanic morphology of the East Pacific Rise crest 9 49'-52'N: Implications for
volcanic emplacement processes at fast-spreading mid-ocean ridges. Marine
Geophysical Research 21, 23-41.

Kvassnes, A. J. S. & Grove, T. L. (2008). How partial melts of mafic lower crust affect
ascending magmas. Contributions to Mineralogy and Petrology 156, 49-71.

Langmuir, C. H. (1989). Geochemical consequences of in situ crystallization. Nature
340, 199-205.

Langmuir, C. H., Bender, J. F. & Batiza, R. (1986). Petrological and tectonic
segmentation of the East Pacific Rise, 5030'-14030'N. Nature 322, 422-429.

Langmuir, C. H., Klein, E. M. & Plank, T. (1992). Petrological Systematics of Mid-Ocean
Ridge Basalts: Constraints on Melt Generation Beneath Ocean Ridges.
Geophysical Monograph 71 Mantle Flow and Melt Generation at Mid-Ocean
Ridges, 183-280.

le Roux, P. J., Shirey, S. B., Hauri, E. H., Perfit, M. R. & Bender, J. F. (2006). The
effects of variable sources, processes and contaminants on the composition of
northern EPR MORB (8-10N and 12-14N): Evidence from volatiles (H20, CO2, S)
and Halogens (F, CI). Earth and Planetary Science Letters 251, 209-231.

Macdonald, K. C. & Fox, P. J. (1983). Overlapping spreading centers: new accretion
geometry on the East Pacific Rise. Nature 302, 55-58.

Macdonald, K. C. & Fox, P. J. (1988). Earth and Planetary Science Letters 88, 119.


172









Macdonald, K. C., Fox, P. J., Perram, L. J., Eisen, M. F., Haymon, R. M., Miller, S. P.,
Carbotte, S. M., Cormier, M.-H. & Shor, A. N. (1988). A new view of the mid-
ocean ridge from the behaviour of ridge-axis discontinuities. Nature 335, 217-
225.

Macdonald, K. C., Scheirer, D. S. & Carbotte, S. M. (1991). Mid-Ocean Ridges:
Discontinuities, Segments, and Giant Cracks. Science 253, 986-994.

Maclennan, J. (2008). The Supply of Heat to Mid-Ocean Ridges by Crystallization and
Cooling of Mantle Melts. Geophysical Monograph 178, 45-73.

Mandeville, C. W., Webster, J. D., Rutherford, M. J., Taylor, B. E., Timbal, A. & Faure,
K. (2002). Determination of molar absorptivities for infrared absorption bands of
H20 in andesitic glass. American Mineralogist 87, 813-821.

Matsuhisa, Y., Matsubaya, O. & Sakai, H. (1973). Oxygen Isotope Variations in
Magmatic Differentiation Processes of the Volcanic Rocks in Japan.
Contributions to Mineralogy and Petrology 39, 277-288.

McBirney, A. R. (ed.) (1993). Differentiated rocks of the Galapagos hotspot: Geological
Society of London Special Publications

Michael, P. & Cornell, W. C. (1998). Influence of spreading rate and magma supply on
crystallization and assimilation beneath mid-ocean ridges: evidence from chlorine
and major element chemistry of mid-ocean ridge basalts. Journal of Geophysical
Research 103, 18325-18356.

Michael, P. & Schilling, J. G. (1989). Chlorine in mid-ocean ridge magmas: Evidence for
assimilation of seawater-influenced components. Geochimica Cosmochemica
Acta 53, 3131-3143.

Michard, A. & Albarede, F. (1986). The REE content of some hydrothermluids.
Chemical Geology 55, 51-60.

Muehlenbach, K. & Byerly, G. R. (1982). 180-Enrichment of Silicic Magmas Caused by
Crystal Fractionation at the Galapagos Spreading Center. Contributions to
Mineralogy and Petrology 79, 76-79.

Muehlenbach, K. & Clayton, R. N. (1972). Oxygen isotope studies of fresh and
weathered submarine basalts. Canadian Journal of Earth Science 2, 172-184.

Nakamura, K., Kato, Y., Tamaki, K. & Ishii, T. (2007). Geochemistry of hydrothermally
altered basaltic rocks from the Southwest Indian Ridge near the Rodriquez Triple
Junction. Marine Geology 239, 125-141.


173









Natland, J. H., Langmuir, C. H., Bender, J. F., Batiza, R. & Hopson, C. (1986).
Petrologic systematics in the vicinity of the 9 degrees N nontransform offset, East
Pacific Rise. EOS, Transactions American Geophysical Union 67, 1254.

Natland, J. H. & MacDougall, J. D. (1986). Parental abyssal tholeiites and alkali basalts
at the East Pacific Rise near 9 degrees N and the Siqueiros fracture zone. EOS,
Transactions American Geophysical Union 67, 410-411.

Newman, S. & Lowenstern, J. B. (2002). VOLATILECALC: a silicate melt-H20-CO2
solution model written in Visual Basic for excel. Computers and Geosciences 28,
597-604.

Nicholson, H., Condomines, M., Fitton, J. G., Fallick, A. E., Gronvold, K. & Rogers, G.
(1991). Geochemical and Isotopic Evidence for Crustal Assimilation Beneath
Krafla, Iceland. Journal of Petrology 32, 1005-1020.

Niu, Y., Gilmore, T., Mackie, S., Greig, A. & Bach, W. (2002). Mineral chemistry, whole-
rock compositions, and petrogenesis of Leg 176 gabbros: data and discussion.
College Station: Ocean Drilling Program.

Nunnery, A., Klein, E. M., Perfit, M., White, R. S., Mason, J. L. & Zaino, A. J. (2008).
Correlation of Seafloor Surface Features and Underlying Melt Bodies at the 9 N
Overlapping Spreading Center, East Pacific Rise. EOS, Transactions American
Geophysical Union 89(53), Fall Meet. Suppl., Abst V53D-03.

O'Hara, M. J. (1977). Geochemical evolution during fractional crystallization of a
periodically refilled magma chamber. Nature 266, 503-507.

O'Nions, R. K. & Gronvold, K. (1973). Petrogenetic relationships of acid and basic rocks
in iceland: Sr-isotopes and reare-earth elements in late and postglacial volcanics.
Earth and Planetary Science Letters 19, 397-409.

Pedersen, R. B. & Malpas, J. (1984). The Origin of oceanic plagiogranites from the
Karmoy ophiolite, western Norway. Contributions to Mineralogy and Petrology
88, 36-52.

Perfit, M., Ridley, W. I. & Cotsonika, L. (in prep). Juan de Fuca Dacites.

Perfit, M. R. & Chadwick, W. W. J. (1998). Magmatism at Mid-Ocean Ridges:
Constraints from Volcanological and Geochemical Investigations. Geophysical
Monograph 106 Faulting and Magmatism at Mid-Ocean Ridges, 59-115.

Perfit, M. R. & Fornari, D. J. (1983). Geochemical studies of abyssal lavas recovered by
DSRV Alvin from Eastern Galapagos Rift, Inca Transform, and Ecuador Rift 2.
Phase chemistry and crystallization history. Journal of Geophysical Research 88,
10530-10550.


174









Perfit, M. R., Fornari, D. J., Malahoff, A. & Embley, R. W. (1983). Geochemical Studies
of Abyssal Lavas Recovered by DSRV Alvin From Eastern Galapagos Rift, Inca
Transform, and Ecuador Rift 3. Trace Element Abundances and Petrogenesis.
Journal of Geophysical Research 88, 10,551-510,572.

Perfit, M. R., Fornari, D. J., Smith, M. C., Bender, J. F., Langmuir, C. H. & Hayman, N.
W. (1994). Small-scale spatial and temporal variations in mid-icean ridge crest
magmatic processes. Geology 22, 375-379.

Perfit, M. R., Ridley, W. I. & Jonasson, I. R. (1999). Geologic, petrologic and
geochemical relationships between magmatism and massive sulfide
mineralization along the eastern Galapagos Spreading Center. Reviews in
economic geology 8, 75-99.

Perfit, M. R., Schmidt, A. K., Ridley, W. I., Rubin, K. H. & Valley, J. (2007). Petrogenesis
of dacites from the southern Juan de Fuca Ridge. Goldschmidt Conf Abstracts,
Geochimica et Cosmochimica Acta 72, A736.

Pollard, D. D. & Sydin, A. (1984). Propagation and Linkage of Oceanic Ridge
Segments. Journal of Geophysical Research 89, 10,017-010,028.

Pollock, M. A., Klein, E. M., Karson, J. A. & Tivey, M. A. (2005). Temporal and spatial
variability in the composition of lavas exposed along the Western Blanco
Transform Fault. Geochemistry Geophysics Geosystems 6, 1-17.

Rapp, R. P., Watson, E. B. & Miller, C. F. (1991). Partial melting of amphibolite/eclogite
and the origin of Archean trondhjemites and tonalites. Precambrian Research 51,
1-25.

Regelous, M., Niu, Y., Went, J. I., Batiza, R., Greig, A. & Collerson, K. D. (1999).
Variations in geochemistry of magmatism on the East Pacific Rise at 10030' N
since 800 ka. Earth and Planetary Science Letters 168, 45-63.

Reynolds, J., R. (1995). Segment-scale systematic of mid-ocean Ridge magmatism and
geochemistry. Palisades: Columbia University, 483.

Reynolds, J., R. & Langmuir, C. H. (1997). Petrological systematics of the Mid-Atlantic
Ridge south of Kane: Implications for ocean crust formation. Journal of
Geophysical Research 102, 14915-14946.

Reynolds, J., R., Langmuir, C. H., Bender, J. F., Kastens, K. A. & Ryan, W. B. F. (1992).
Spatial and temporal variability in the geochemistry of basalts from the East
Pacific Rise. Nature 359, 493-499.


175









Rubin, K. H. & Sinton, J. M. (2007). Inferences on mid-ocean ridge thermal and
magmatic structure from MORB compositions. Earth and Planetary Science
Letters 260, 257-276.

Saal, A. E., Hauri, E. H., Langmuir, C. H. & Perfit, M. (2002). Vapour undersaturation in
primitive mid-ocean-ridge basalt and the volatile content of Earth's upper mantle.
Nature 419, 451-455.

Schiffman, P., Zierenber, R., Chadwick, W. W. J., Clague, D. A. & Lowenstern, J. B.
(2010). Contanination of basaltic lava by seawater: Evidence found in a lava
pillar from Axial Seamount, Juan de Fuca Ridge. Geochemistry Geophysics
Geosystems 11, 1-12.

Schmitt, A. K., Perfit, M., Rubin, K. H., Stockli, D. F., Smith, M. C., Cotsonika, L.,
Zellmer, G. F., Ridley, W. I. & Lovera, O. M. (submitted). Zircon in Evolved Mid-
Ocean Ridge Lavas Constrains Fast Cooling Rates of Oceanic Lower Crust.
Nature.

Sempere, J.-C. & Macdonald, K. C. (1986). Deep-Tow Studies Of The Overlapping
Spreading Centers at 903' N On The East Pacific Rise. Tectonics 5, 881-900.

Sempre, J.-C. & Macdonald, K. C. (1986). Deep-Tow Studies Of The Overlapping
Spreading Centers at 903' N On The East Pacific Rise. Tectonics 5, 881-900.

Sigurdsson, H. (1977). Generation of Icelandic rhyolites by melting of plagiogranites in
the oceanic layer. Nature 269, 25-28.

Sigurdsson, H. & Sparks, R. S. J. (1981). Petrology of rhyolitic and mixed magma ejecta
from the 1875 eruption of Askja, Iceland. Journal of Petrology 22, 41-84.

Sims, K. W. W., Blichert-Toft, J., Fornari, D. J., Perfit, M. R., Goldstein, S. J., Johnson,
P., De Paolo, D. J. & Michael, P. (2003). Abberant Youth: Chemical and isotopic
constraints on the young off-axis lavas of the East Pacific Rise. Geochemistry
Geophysics Geosystems 4, 8621.

Sims, K. W. W., Goldstein, S. J., Blichert-Toft, J., Perfit, M. R., Kelemen, P. B., Fornari,
D. J., Michael, P., Murrell, M. T., Hart, S. R., DePaolo, D. J., Layne, G., Ball, L.,
Jull, M. & Bender, J. F. (2002). Chemical and isotopic constraints on the
generation and transport of magma beneath the East Pacific Rise. Geochimica et
Cosmochimica Acta 66, 3481-3504.

Singh, S. C., Harding, A. J., Kent, G. M., Sinha, M. C., Combier, V., Bazin, S., Tong, C.
H., Pye, J. W., Barton, P. J., Hobbs, R. W., White, R. S. & Orcutt, J. A. (2006).
Seismic reflection images of the Moho underlying melt sills at the East Pacific
Rise. Nature 442, 1-4.


176









Sinton, J. M., Ford, L. L., Chappell, B. & McCulloch, M. T. (2003). Magma Genesis and
Mantle Heterogeneity in the Manus Back-Arc Basin, Paupa New Guinea. Journal
of Petrology 44, 159-195.

Sinton, J. M., Smaglik, S. M. & Mahoney, J. J. (1991). Magmatic processes at superfast
spreading mid-ocean ridges: glass compositional variations along the East
Pacific Rise 130-230S. Journal of Geophysical Research 96, 6133-6155.

Sinton, J. M., Wilson, D., Christie, D. M., Hey, R. N. & Delaney, J. R. (1983).
Petrological consequences of rift propagation on oceanic spreading ridges. Earth
and Planetary Science Letters 62, 193-207.

Smith, M. C., Perfit, M. R., Fornari, D. J., Ridley, W. I., Edwards, M. H., Kurras, G. J. &
Von Damm, K. L. (2001). Magmatic processes and segmentation at a fast
spreading mid-ocean ridge: Detailed investigation of an axial discontinuity on the
East Pacific Rise crest at 9037'N. Geochemistry Geophysics Geosystems 2, 1-
32.

Smith, M. C., Perfit, M. R. & Jonasson, I. R. (1994). Petrology and geochemsitry of
basalts from the Southern Juan de Fuca Ridge: Controls on the spatial and
temporal evolution of mid-ocean ridge basalts. Journal of Geophysical Research
99, 4787-4812.

Soule, A. S., Fornari, D. J., Perfit, M. R., Tivey, M. A., Ridley, W. I. & Schouten, H.
(2005). Channelized lava flows at the East Pacific Rise crest 9-10 N: The
importance of off-axis lava transport in developing the architecture of young
oceanic crust. Geochemistry Geophysics Geosystems 6, 1-28.

Sparks, J. W. (1995). Geochemistry of the Lower Sheeted Dike Complex, Hole 504B,
Leg 140. Proceedings of the Ocean Drilling Program, Scientific Results 137/140,
81-97.

Stakes, D. S., Perfit, M. R., Tivey, M. A., Caress, D. W., Ramirez, T. M. & Maher, N.
(2006). The Cleft revealed: Geologic, magnetic, and morphologic evidence for
construction of upper oceanic crust along the southern Juan de Fuca Ridge.
Geochemistry Geophysics Geosystems 7, 1-32.

Sun, S. & McDonough, W. F. (eds.) (1989). Chemical and Isotopic Systematics of
oceanic basalts: implications for Mantle Composition and Processes: Geological
Society of London, Special Publications

Sun, S. S., Tatsumoto, M. & Schilling, J. G. (1975). Mantle Plume Mixing Along the
Reykjanes Ridge Axis: Lead Isotopic Evidence. Science 190, 143-147.









Sun, W. D., Binns, R. A., Fan, A. C., Kamenetsky, V. S., Wysoczanski, R., Wei, G. J.,
Hu, Y. H. & Arculus, R. J. (2007). Chlorine in submarine volcanic glasses from
the eastern Manus basin. Geochemica et Cosmochimica Acta 71, 1542-1552.

Taylor, H. P. J. (1968). The oxygen isotope geochemistry of igneous rocks.
Contributions to Mineralogy and Petrology 19, 1-71.

Taylor, H. P. J. (1980). The effects of assimilation of country rocks by magmas on
18Q/160 and 87Sr/86Sr systematics in igneous rocks. Earth and Planetary Science
Letters 47, 243.

Thy, P., Lesher, C. E. & Mayfield, J. D. (1999). 9. Low-Pressure Melting Studies of
Basalts and Basaltic Andesites From The Southeast Greenland Continental
Margin And The Origin Of Dacites At Site 917. Proceedings of the Ocean Drilling
Program, Scientific Results 163, 95-112.

Tong, C. H., Pye, J. W., Barton, P. J., White, R. S., Sinha, M. C., Singh, S. C., Hobbs,
R. W., Bazin, S., Harding, A. J., Kent, G. M. & Orcutt, J. A. (2002). Asymmetric
melt sills and upper crustal construction beneath overlapping ridge segments:
Implications for the development of melt sills and ridge crests. Geology 30, 83-
86.

Toomey, D. R., Jousselin, D., Dunn, R. A., Wilcock, W. S. D. & Detrick, R. S. (2007).
Skew of mantle upwelling beneath the East Pacific Rise governs segmentation.
Nature 446, 409-414.

Unni, C. K. & Schilling, J. G. (1978). Cl and Br degassing by volcanism along the
Reykjanes Ridge and Iceland. Nature 272, 19-23.

Wanless, D., Perfit, M., Ridley, W. I. & Klein, E. M. (2008). Origin of Dacites on Mid-
Ocean Ridges. EOS, Transactions American Geophysical Union 89(53), Fall
Meeting Supplementary, Abstract V41 B-2084.

Wanless, D., Perfit, M., Ridley, W. I., Klein, E. M., Wallace, P., Valley, J. & Grimes, C.
(submitted). Role of Assimilation in the Petrogenesis of Lavas on Mid-Ocean
Ridges inferred from Cl, H20, CO2 and Oxygen Isotope Analyses.

Wanless, D., Perfit, M., Ridley, W. I., Klein, E. M. & White, R. S. (in prep). 9N
Overlapping Spreading Center: Magma Genesis and Ridge Segmentation.
Geochemistry Geophysics Geosystems.

Wanless, D., Perfit, M. R., Ridley, W. I., Wallace, P., Valley, J., Grimes, C. & Klein, E.
M. (2009). Geochemical Evidence for Crustal Assimilation at Mid-Ocean Ridges
Using Major and Trace Elements, Volatiles and Oxygen Isotopes. EOS,
Transactions American Geophysical Union, Fall Meeting Supplementary,
Abstract V51A-1665.


178









Wanless, V. D., Perfit, M., Ridley, W. I. & Klein, E. M. (accepted). Dacite Petrogenesis
on Mid-Ocean Ridges: Evidence for Crustal Melting and Assimilation. Journal of
Petrology Accepted.

Watson, E. B. (1979). Apatite saturation in basic to intermediate magmas. Geophysical
Research Letters 6, 937-940.

Webster, J. D., Kinzler, R. J. & Mathez, E. A. (1999). Chloride and water solubility in
basalts and andesite melts and implications for magmatic degassing.
Geochemica et Cosmochimica Acta 63, 729-738.

West, M., Menke, W., Tolstoy, M., Webb, S. & Sohn, R. (2001). Magma storage
beneath Axial volcano on the Juan de Fuca mid-ocean ridge. Nature 413, 833-
836.

White, S. M., Haymon, R. M., Fornari, D. J., Perfit, M. R. & Macdonald, K. C. (2002).
Correlation between tectonic and volcanic segmentation at fast-spreading ridges:
evidence from the distribution of volcanic structures and lava flow morphology
along East Pacific Rise at 9 10N. Journal of Geophysical Research 107.

White, S. M., Mason, J. L., Macdonald, K. C., Perfit, M. R., Wanless, V. D. & Klein, E.
M. (2009). Significance of widespread low effusion rate eruptions over the past
two million years for delivery of magma to the overlapping spreading centers at
9N East Pacific Rise. Earth and Planetary Science Letters 280, 175-184.

White, W. M. & Schilling, J. G. (1978). The nature and origin of geochemical variation in
Mid-Atlantic Ridge basalts from the Central North Atlantic. Geochemica et
Cosmochimica Acta 42, 1501-1516.

Wilson, D. S., Teagle, D. A. H., Alt, J., Banerjee, N. R., Umino, S., Miyashita, S., Acton,
G. D., Anma, R., Barr, S. R., Belghoul, A., Carlut, J., Christie, D. M., Coggon, R.
M., Cooper, K. M., Cordier, C., Crispini, L., Durand, S. R., Einaudi, F., Galli, L.,
Gao, Y., Geldmacher, J., Gilbert, L. A., Hayman, N. W., Herrero-Bervera, E.,
Hirano, N., Holter, S., Ingle, S., Jiang, S., Kalberkamp, U., Kerneklian, M.,
Koepke, J., Laverne, C., Vasquez, H. L. L., Maclennan, J., Morgan, S., Neo, N.,
Nichols, H. J., Park, S.-H., Reichow, M. K., Sakuyama, T., Sano, T., Sandwell,
R., Scheibner, B., Smith-Duque, C. E., Swift, S. A., Tartarottie, P., Tikku, A. A.,
Tominaga, M., Veloso, E. A., Yamasaki, T., Yamazaki, S. & Ziegler, C. (2006).
Drilling to Gabbro in Intact Ocean Crust. Science 312, 1016-1020.

Wolf, M. B. & Wyllie, P. J. (1994). Dehydration-melting of amphibolite at 10 kbar: the
effects of temperature and time. Contributions to Mineralogy and Petrology 115,
369-383.


179









geochemical signatures consistent with MOR dacites. This supports the hypothesis that

crustal assimilation is an important process in the formation of highly evolved MOR

lavas and may be significant in the generation of evolved MORB in general.

Additionally, these processes are likely to be more common in regions of episodic

magma supply and enhanced magma-crust interaction such as at the ends of ridge

segments.

Introduction

Fast to intermediate oceanic spreading centers typically erupt geochemically

diverse basaltic lavas (e.g. Klein, 2005); however, a much more extensive range of lava

compositions, including ferrobasalts and FeTi basalts as well as rarer high-silica

andesites and dacites have been recovered from several different ridges (Perfit et al.,

1983; Langmuir et al., 1986; Natland et al., 1986; Natland & MacDougall, 1986;

Regelous et al., 1999; Smith et al., 2001). These great variations in compositions are

commonly attributed to low magma supply and/or cooler crust at ridge segment ends, or

the cold edge effect, which promote greater differentiation of magmas prior to eruption

(Christie & Sinton, 1981; Perfit et al., 1983; Sinton et al., 1983; Perfit & Chadwick, 1998;

Rubin & Sinton, 2007).

The formation of highly evolved, silicic magmas in non-ridge settings (e.g. ocean

islands, arc volcanoes, and continental interiors) have been attributed to several

different processes, including crystal fractionation, partial melting of overlying crust,

and/or assimilation of crustal material into an evolving magma chamber. On mid-ocean

ridges (MOR), many studies have documented the dominant role crystal fractionation

plays in magma differentiation (e.g. Clague & Bunch, 1976; Bryan & Moore, 1977;

Byerly, 1980) whereby extensive crystallization of olivine, plagioclase, pyroxene and Fe-









Zaino, A. J. (2009). Petrology and Mineral Chemistry of the 9003'N Overlapping
Spreading Center, East Pacific Rise. Department of Earth and Ocean Sciences.
Durham: Duke Univeristy, 60.


180









BIOGRAPHICAL SKETCH

Virginia Dorsey Wanless was born in Denver, Colorado but grew up in Topeka,

Kansas. She earned her Bachelor of Arts in geology from Colgate University in 2001

and her Master of Science from the Department of Geology and Geophysics at the

University of Hawai'i in 2005. Interspersed with her education she worked at the

Hawaiian Volcano Observatory (HVO), FUGRO Sea Floor Surveys, and as a side scan

sonar analyst for the Hawaiian Research Mapping Group (HMRG) and the REMUS

6000 team at Woods Hole Oceanographic Institution (WHOI). After completion of her

Doctor of Philosophy in the summer of 2010, she began a postdoctoral fellowship at

WHOI.









Ti oxides leads to the generation of highly evolved melts enriched in SiO2 and depleted

in MgO, FeO and TiO2 (e.g. Juster et al., 1989). However, while crystal fractionation is

undoubtedly a primary process involved in the differentiation of most MORB magmas, it

may not be the only mechanism involved in the formation of high-silica MOR andesites

and dacites.

Partial melting (or anatexis) of basaltic crustal material may produce evolved

compositions, particularly in settings where magma-rock interactions are likely, such as

the top of an axial magma chamber (e.g. Coogan et al., 2003b; Gillis, 2008). This

process has been suggest as the origin for high-silica lavas erupted on many ocean

islands (Iceland; O'Nions & Gronvold, 1973; Sigurdsson & Sparks, 1981; Galapagos

Islands; McBirney, 1993; Socorro Island; Bohrson & Reid, 1997; Bohrson & Reid, 1998)

and may explain the formation of high-silica lavas in back arc settings, most recently

along the Lau Spreading Center (e.g. Kent et al., 2002) and Manus Basin (Sinton et al.,

2003); although the presence of a subduction zone makes this tectonic setting much

more complicated. Evidence from ophiolites suggests that the top of the axial magma

chamber on MOR is a dynamic boundary where magmas may interact with and melt

different layers of crustal material; including both gabbros and sheeted dikes (Coogan et

al., 2003b). Recent experimental evidence suggests that partial melting of hydrous

gabbroic rock similar to that in the lower ocean crust can form silicic compositions

(Koepke et al., 2004; Kvassnes & Grove, 2008) and may explain the presence of highly

evolved plagiogranite veins in the ocean crust (Koepke et al., 2004; Brophy, 2009).

Other studies indicate that low degrees of dehydration partial melting of altered basalt,

similar in composition to dikes of the upper ocean crust, can produce dacitic melts

































2010 V. Dorsey Wanless









(Beard & Lofgren, 1991). These experimental studies suggest that oceanic crust will

begin to melt at temperatures as low as 8500 to 900C and <10% melting of the crust

will yield dacitic or tonalitic melts (Beard & Lofgren, 1991; Koepke et al., 2004;

Kvassnes & Grove, 2008). Kvassnes and Grove (2008) state that mineral pairs

(plagioclase-olivine and plagioclase-augite) similar to oceanic gabbros from the lower

crust will melt quickly and easily at temperatures similar to that of primitive MOR

magmas (1220-1330C). All of these studies indicate that high level partial melting of

ocean crust can produce high-silica melts on MOR, but the role that this process may

play in the formation of voluminous extrusive silicic lavas on the seafloor has not yet

been assessed.

The compositional variability observed in arc and continental volcanics is

commonly ascribed to the associated processes of assimilation and fractional

crystallization (AFC; e.g. Bowen, 1928; De Paolo, 1981) but similar processes may

also occur where thickened oceanic crust leads to magma-crust interaction, for

instance, within Icelandic volcanoes (e.g. Nicholson et al., 1991). On smaller scales,

the combined effects of these processes have been observed in ophiolites, where sub-

axial intrusive magmas have been in contact with and have partially melted the

overlying sheeted dikes (Gillis & Coogan, 2002; Coogan, 2003; Gillis, 2008). AFC

processes have also been invoked to explain high Cl concentrations observed in some

MORB magmas (Michael & Schilling, 1989; Michael & Cornell, 1998; le Roux et al.,

2006). During this process, a magma undergoes crystal fractionation, and the resultant

latent heat of crystallization provides the heat needed to partially melt the surrounding

wall rock. These melts are then assimilated into, and homogenized with, the









fractionating magma reservoir. AFC processes can produce a wide range of rock types

depending on the initial composition of the intruding magma, the degree of crystal

fractionation, the initial wall rock composition, and the amount of melting and

assimilation.

High-silica compositions are found throughout the ocean crust and are commonly

observed as intrusive or plutonic material. As mentioned above, plagiogranites veins

are a ubiquitous component of the ocean crust and have been observed in ophiolites

(e.g. Pedersen & Malpas, 1984), drill cores from the ocean crust (Casey, 1997; e.g.

Dick et al., 2000; Wilson et al., 2006), and as xenoliths in Icelandic lavas (Sigurdsson,

1977). The origin of these veins remains unclear but two main hypotheses are: 1) partial

melting of gabbroic crust (e.g. Koepke et al., 2004; Koepke et al., 2007; Nunnery et al.,

2008) and 2) extreme crystal fractionation of tholeiitic magmas(Coleman & Donato,

1979; Beccaluva et al., 1999; Niu et al., 2002). There are also many examples of

evolved plutonic rocks from slower spreading centers (e.g. Mid-Atlantic Ridge,

Aumento, 1969), which may suggest AFC or partial melting processes are occurring on

much smaller scales, deeper in the ocean crust.

In this study we examine the geochemistry of high-silica lavas from three different

MOR, including the East Pacific Rise, Juan de Fuca Ridge, and Galapagos Spreading

Center (Figure 2-1) and show they have remarkably similar major and trace element

compositions (Figure 2-2), suggesting that similar sources and processes control their

petrogenesis. More specifically, we examine the roles that crystal fractionation, partial

melting, and AFC may have played in the formation of an exceptional suite of high-silica

lavas from the 90N overlapping spreading center (OSC) on the East Pacific Rise, and









evaluate if these results apply generally to the formation of high-silica lavas on other

MOR. We focus on the petrogenesis of dacites at the 90N OSC because it is the most

complete and geologically well-constrained data set available but descriptions of the

geologic settings of high-silica lavas in the other environments are important in order to

ascertain the role tectono-magmatic settings may have on their petrogenesis.

Geologic and Tectonic Setting

The MOR system is over 70,000 km long (Macdonald et al., 1991) and the crust

formed at these ridges is overwhelmingly basaltic in nature. High-silica lavas, however,

have erupted on several fast- and intermediate-spreading ridges and are commonly

associated with specific tectonic settings; including propagating ridge tips (Christie &

Sinton, 1981; Fornari et al., 1983; Perfit & Fornari, 1983); OSC (Christie & Sinton, 1981;

Perfit et al., 1983; Sinton et al., 1983; Bazin et al., 2001); regions of ridge-hotspot

interaction (Chadwick et al., 2005; Haase et al., 2005) and at 10030 N on the East

Pacific Rise near the ridge-transform intersection (Regelous et al., 1999). Below, we

describe the geologic setting of the 9N OSC and the three other ridges (East Pacific

Rise, Juan de Fuca Ridge, and Galapagos Spreading Center) where highly evolved

lavas have erupted, to elucidate the relationship of magmatism to different MOR

tectono-magmatic environments.

90N East Pacific Rise Overlapping Spreading Center

The 90N OSC (Figure 2-1 a) is located on the East Pacific Rise between the

Clipperton and Siqueiros transform faults. It consists of two north-south trending ridges

that overlap by ~27 km and partly enclose a large overlap basin (Sempere &

Macdonald, 1986) The limbs are separated by ~8 km (Singh et al., 2006). The eastern









limb is propagating to the south into older crust at a rate of ~42 km Myr1 (Carbotte &

Macdonald, 1992).

The 90N OSC has been the focus of several geophysical studies (Detrick et al.,

1987; Harding et al., 1993; Kent et al, 1993; Kent et al., 2000; Bazin et al., 2001; Dunn

et al., 2001; Tong et al., 2002); including the first MCS 3D survey of a MOR (Kent et al.,

2000) and a 3D seismic refraction study (Dunn et al., 2001). These studies resulted in

the first 3D image of a subsurface magma chamber along a MOR, which showed a

shallow melt lens lies beneath both limbs of the OSC with an anomalously large melt

lens in the interlimb region, north of the overlap basin. This suggests that the region

currently has an unusually high magma supply rate for a ridge segment end (Kent et al.,

2000).

High-silica andesites and dacites were recovered from the eastern limb during the

Medusa2007 cruise (AT15-17) using the ROV Jason2 (Wanless et al., 2008; White et

al., 2009). Several high-silica lavas were also recovered from this area during dredging

operations in the late 1980's (Langmuir et al., 1986). The siliceous lavas are primarily

confined to the northern section of the neo-volcanic zone on the eastern, propagating

limb, along the eastern edge of the melt lens (Figure 2-3). Morphologically, the dacites

form large individual bulbous to elongate pillows that can be several meters in diameter

(Figure 2-4). The pillows are highly striated and have a coarse bread crust surface

texture. Typically, the pillows are stacked into mounds, which can be several meters

high or constructional domes. Dacites largely occur in two regions: as a nearly linear

pillow mound in the center of the east limb neo-volcanic zone and large, elongate pillow









lavas on the flanks of the axial graben (Figure 2-3). Their axial and near-axial positions,

low sediment cover and unaltered nature suggest they are relatively young.

Juan de Fuca Ridge Propagating Ridge Tip and Axial Seamount

The Cleft segment is the southernmost segment of the Juan de Fuca Ridge

(Figure 2-1b). It terminates at 44027'N at a ridge-transform intersection, where it

intersects and overlaps the Blanco Transform Zone (Embley et al, 1991; Embley &

Wilson, 1992; Smith et al., 1994). This intersection is characterized by a series of

curved ridges that overshoot the Blanco Transform Zone onto the older Pacific plate

(Stakes et al., 2006).

High-silica andesites and dacites comprise two small constructional domes on the

Pacific plate, where the axial ridge intersects, and is believed to propagate past the

Blanco Transform Zone into the older ocean crust (~6.3 ma) that was created at the

Gorda Ridge (Embley & Wilson, 1992; Stakes et al., 2006). The domes are -20 to 30 m

high and 200 to 500 m in diameter and were sampled using rock core and the ROV

Tiburon during research cruises in 2000 and 2002 (Cotsonika, 2006). High-resolution

bathymetric maps show there are numerous other constructional domes in the region

but they have not yet been sampled, although we surmise that they are also composed

of high-silica lavas. Rare andesites have also been recovered within the axis and along

the bounding faults of the southern Cleft segment (Stakes et al., 2006).

Seismic studies of the southern Juan de Fuca Ridge indicate the presence of an

axial magma chamber beneath most of the Cleft segment (Canales et al., 2005).

However, an axial magma chamber reflector is absent south of 44038'N where the high

silica lavas were recovered, suggesting the presence of small melt volumes resulting

from weak melt supply to the ridge-transform intersection. Zircon thermochronology and









U-series data indicate that the dacites erupted less than 30 ka ago (Schmitt et al.,

submitted).

The Axial segment of the Juan de Fuca Ridge is a second order ridge segment

that currently overlies the Cobb hot spot, which has produced a chain of seamounts

trending NW away from the ridge axis (Chadwick et al, 2005). The Juan de Fuca Ridge

is migrating NW at a rate of 3.1 cm/yr and has been situated above the Cobb hot spot

for the last ~0.2 to 0.7 Myr, creating a large on-axis seamount, known as Axial

seamount (Karsten & Delaney, 1989).

Axial seamount is the largest feature on this segment of the Juan de Fuca Ridge

and has a large summit caldera (Embley et al., 1999) underlain by a large seismically-

imaged axial magma chamber (West et al., 2001). It has two prominent rift zones,

extending to the north and south, which create bathymetric highs. Extensive sampling

of the main edifice shows that it is composed of moderately evolved and slightly

enriched MORB (Chadwick et al., 2005). The rift zones have linear ridges that appear

to accommodate extensive diking from the main caldera system (Chadwick et al., 2005).

Rare high-silica andesites sampled by three rock cores are located east of the northern

rift zone and may be associated with dike propagation from the main axial magma

chamber into older ridge crust.

Galapagos Spreading Center- Extinct OSC or Propagating Ridge Tip?

High-silica lavas were sampled at the eastern end of the Galapagos Spreading

Center at ~850 W. The area was extensively studied though dredging and Alvin

exploration in the early 1980s (Fornari et al., 1983; Perfit & Fornari, 1983; Perfit et al.,

1983; Embley et al., 1988). The evolved lavas erupted within the axial valley and along









the axis-bounding faults of the Galapagos Spreading Center approximately 20 km east

of the ridge-transform intersection with the Inca transform fault (Figure 2-1c).

Bathymetric data reveal two curved ridges surrounding a depression within this region,

which has been interpreted as an old, small, extinct OSC or deviation from axial linearity

(Perfit et al., 1983; Embley et al., 1988) that has been rifted away from the neo-volcanic

zone. Most of the evolved lavas (~63% SiO2) at the Galapagos Spreading Center were

found off-axis along the bounding faults associated with the southern portion of the

extinct OSC.

Petrography

The 9N OSC dacites are glassy and predominantly aphyric, with sparse

microphenocrysts and very rare, small phenocrysts of plagioclase and clinopyroxene.

The small phenocrysts of plagioclase commonly have resorbed edges and sieve

textures. Several of the samples contain small clots of basaltic xenoliths comprised of

subophitic plagioclase and clinopyroxene surrounded by dacitic or andesitic glass.

Geochemical Methods

Glass chips from the outer rims of 18 dacites collected at the 90N OSC were

analyzed on a JEOL 8900 Electron Microprobe for major element concentrations at the

USGS in Denver, Colorado (Table 2-1). Eight to ten individual points were analyzed per

sample. USGS mineral standards were used to calibrate the microprobe and secondary

normalizations were done to account for instrument drift using the JdF-D2 glass

"standard" (Reynolds, 1995), University of Florida in-house standard ALV 2392-9 from

the East Pacific Rise (Smith etal., 2001) and USGS standard dacite glass GSC (for

more detail on methods see (Smith et al., 2001)). The probe diameter during glass

analyses was 20 pm, and an accelerating voltage of 15keV and a beam current of 20nA









were used. High-precision chlorine and potassium concentrations were also

determined by microprobe on seven of the dacites using 200-second peak/100 second

background counting times.

Small glass fragments (10-50 mm) were handpicked, avoiding microphenocrysts

and alteration, cleaned, and dissolved for trace element and isotope analyses following

methods described in (Goss et al., 2010). Fourteen dacites from the 90N OSC were

analyzed at medium resolution for trace element concentrations on a high precision

Element2 Inductively Coupled Plasma Mass Spectrometer (ICP-MS) at the University of

Florida (Table 2-1). Radiogenic isotope ratios (Pb, Sr, and Nd) were determined for 10

dacites using a Nu-Plasma multi-collector ICP-MS at the University of Florida Center for

Isotope Geoscience (Table 2-2). For a detailed description of sample preparation,

dissolution procedures, standards, and errors, see (Kamenov et al., 2007; Goss et al.,

2010; Goss et al., in prep). External calibration was done to quantify results using a

combination of in-house basalts (ENDV Endeavour and ALV 2392-9) and USGS

(AGV-1, BIR-1, BHVO-1, BCR-2 and STM-1) rock standards (Kamenov et al., 2007;

Goss et al., 2010).

Geochemical Results

Major Element Results

Major element compositions of the 9N OSC high-silica andesites and dacites are

presented (along with the trace element abundances) in Table 2-1. The major element

geochemistry of the 90N OSC lavas is similar to Juan de Fuca Ridge and Galapagos

Spreading Center lavas (Figure 2-2). Here, only data from the 9N OSC is discussed in

detail but it is important to note that the major and trace element contents and elemental









trends in high-silica lavas from all three ridges are similar. New analyses of some of the

high-silica samples previously analyzed and discussed by Perfit et al., (1983; 1999) and

representative samples from the Juan de Fuca Ridge are presented in Supplementary

Data. All high-silica samples from the 90N OSC appear unweathered with minimal

amounts of Fe-Mn oxide coating and are essentially aphyric.

9N OSC tholeiitic andesites through dacite samples exhibit increasing SiO2 with

decreasing FeO, TiO2, and MgO (Figures 2-2 and 2-5) with the most differentiated

dacites having ~67 wt% SiO2 and <1 wt% MgO. A1203 concentrations in the dacites

(12.9 to 13.3 wt%), however, do not show a large decrease compared to the OSC

basalts (Figure 2-5). The dacites have high incompatible major element concentrations

(K20> 0.90 wt % and Na20 > 3.4 wt %; Figure 2-5) but low P205 concentrations (< 0.26

wt%; Figure 2-5) compared to basalts. Chlorine concentrations in the dacites range

from 0.24 to 0.70 wt% compared to < 0.01 to 0.04 wt% in the OSC basalts (Figure 2-6).

Trace Element Results

90N OSC dacites are enriched in incompatible trace elements compared to 90N

OSC basalts (Figure 2-7), the latter having compositions typical of normal, incompatible

trace element-depleted mid-ocean ridge basalts (N-MORB) from the northern East

Pacific Rise. For example, Zr and Hf concentrations in the dacites range from 622 to

1050 ppm and 18 to 25 ppm, respectively (Table 2-1). The dacites also contain high

concentrations of Rb, Ba, U and Th but relatively low Sr and Eu contents. Compared to

East Pacific Rise N-MORB the dacites have relatively flat REE patterns (Figure 2-2). On

mantle-normalized diagrams the dacites have positive Zr, Hf, U, and Th anomalies, and

negative Nb and Ta anomalies (Figure 2-2). Consequently, the dacites also have









slightly lower Nb/La and higher Zr/Dy ratios compared to OSC basalts (Figure 2-8) and

Ce/Yb and Nd/Y ratios increase from basalt to dacites (Figure 2-8). Compatible trace

elements are low in the dacites, with Ni concentrations ranging from 9.8 to 4.9. ppm

(Figure 2-7) and Cr concentrations from 13 to 1 ppm. Most incompatible trace elements

have negative correlations with MgO; however, compatible elements (i.e. Ni and Cr) and

Nb/La are positively correlated. U/Nb, Nd/Y and Ce/Yb ratios are negatively correlated

with MgO in the dacites.

Isotopic Data

The 90N OSC dacites have very limited ranges of Pb, Sr and Nd isotopic

compositions, which lie within the general field of East Pacific Rise MORB lavas (Table

2-2; Figure 2-9). 87Sr/86Sr ratios range from 0.70246 to 0.70258, with an average of

0.70250. These values are well within the range of N-MORB East Pacific Rise lavas

from 9-100 N (Sims et al., 2002; Sims et al., 2003; Goss et al., 2010; Goss et al., in

prep) and similar to N-MORB lavas from 90N OSC. 14Nd/144Nd ratios are also similar to

East Pacific Rise N-MORB and range from 0.513140 (SNd = 9.8) to 0.513196 (SNd =

10.9). Pb isotopes ratios for the 90N OSC dacites are indistinguishable from 90N OSC

basalts and other East Pacific Rise lavas, having averages of 206Pb/204Pb = 18.268,

207pb/204Pb = 15.473, and 208Pb/204Pb = 37.679. Pb isotopes for 9 N dacites form a

tight cluster in the center of the field defined for other 9 N lavas (Figure 2-9).

Petrogenetic Models For High-Silica Lavas

We now examine the results of various models of fractional crystallization, partial

melting and assimilation and compare the results with the geochemical data described

above to evaluate their relevance to the formation of MOR dacites. Specifically, we


























To my family and friends









focus on physically reasonable models that are consistent with the highly differentiated

major element concentrations, high concentrations of incompatible elements, distinct

trace element patterns, and N-MORB-like isotopic signatures. The models must also be

able to explain relatively high U, Th, Zr and Hf, and low Nb and Ta as well as the flat

REE patterns. Additionally, markedly high CI, K (and high CI/K), A1203, and low P205

must be accounted for in successful petrogenetic schemes.

Crystal Fractionation

Several petrologic models, including MELTS thermodynamic modeling (Ghiorso &

Sack, 1995), Rayleigh crystal fractionation, crystal-melt segregation, and in-situ

crystallization are investigated here to determine if various processes of crystal

fractionation can account for major and trace element compositions of MOR dacites.

Rayleigh fractional crystallization

The program MELTS (Ghiorso & Sack, 1995) provides a useful framework to

evaluate if major element compositions of MOR dacites can be produced by crystal

fractionation. Petrologic modeling was also carried out using the program PETROLOG

(Danyushevsky, 2001) with similar results for high-silica lavas, although the programs

generate somewhat different results for intermediate compositions. Several 9N N-

MORB's were used as starting parental melt compositions (Table 2-3) to determine if a

moderately evolved magma could partially crystallize to produce a dacite. These

included a slightly evolved MORB (265-113), a ferrobasalt (265-43), and a FeTi basalt

(264-08). Pressures for each MELTS run were set at 1 kbar to simulate an approximate

minimum depth of crystallization in the shallow oceanic crust, the oxygen fugacity was

set at the quartz-fayalite-magnetite, and the H20 concentrations varied from 0.2 to 0.35

wt% depending on the parent melt composition. Liquid lines of descent were also









calculated for higher pressures (up to 5 kbar) to simulate depths of crystallization within

the nascent layer 3 and the shallow mantle. However, the liquid lines of descent

converge on similar end-member compositions at low MgO and high SiO2.

Liquids with MgO and SiO2 compositions similar, though not identical to 90N OSC

dacites can be produced by ~75-85 % crystal fractionation of a ferrobasaltic parent.

Results predict a crystallization sequence of 01, followed by 01 + Plag, 01 + Plag + Cpx,

Plag + Cpx + Sp (titanomagnetite), and in some models, late stage crystallization of

apatite. No pigeonite or orthopyroxene crystallization is predicted, in contrast to

experimental results (Juster et al., 1989) but pigeonite is observed in some MOR

andesites and dacites. The calculations suggest temperatures of <980C are reached

when residual liquids attain compositions similar to dacites. The models are in

agreement with anhydrous experimental results that indicate residual dacitic liquids form

after -87% at temperatures of ~1050C (Juster et al., 1989).

For several major elements (TiO2, FeO, and SiO2), compositions similar to 90N

OSC dacites are obtained through crystal fractionation of a MORB magma; the total

amount of crystallization varies slightly depending on whether the starting composition

was a moderately evolved basalt, ferrobasalt, or FeTi basalt (Figure 2-5; maximum of

~85 % crystallization). In contrast, calculated abundances of K20, P205, A1203, and Cl

do not match the dacite end-member composition using any of the parents; modeled

residual liquids have higher P205 by factors of 5-10, lower K20 by factors of 1.5-2.5,

lower A1203 by factors of 1.4-1.5 and lower Cl by factors of 10 to 12 (Figure 2-5 and 2-

6). Although MELTS does not predict apatite saturation in andesitic liquids the

decreasing P205 contents in some andesites and very low values in dacites strongly









suggest apatite crystallization. Juster et al., (1989) calculated apatite saturation would

occur at approximately 0.7 wt% P205 in Galapagos Spreading Center andesites.

Additional tests of Rayleigh fractionation include modeling the behavior of trace

elements using as input parameters the degree of crystallization and mineral modal

proportions determined from the MELTS modeling. Basaltic partition coefficients are

used in the trace element modeling up to ~ 57% Si02, and andesitic partition

coefficients are used for > 57% Si02 (Table 2-4). The Rayleigh fractionation equation

(CI/Co=F^(D-1)) is used to simulate a continuously evolving magma chamber in which

phenocrysts are immediately separated from the liquid. The starting composition was a

ferrobasalt from the 9N OSC (265-43).

As shown in Figure 2-7, the trace element concentrations observed in the dacites

cannot be reproduced using Rayleigh fractionation (Figure 2-7) with the constraints

imposed by major element variations. This model reproduces some dacite incompatible

element compositions but does not reproduce the observed enrichments in most of the

incompatible elements. Although the most incompatible elements (Rb, Ba, Th, U) show

the greatest difference between the observed and calculated compositions, even less

incompatible elements (Nb, Zr, Y, Hf) require > 90% crystal fractionation. For instance,

maximum calculated Zr and Nb concentrations are 705 ppm and 13 ppm, respectively,

compared to an average of 870 and 15 ppm in the dacites. In addition, the UN/NbN is

predicted by modeling to be <1, whereas the measured dacite values are >1 and the

modeling does not reproduce the high ZrN/DyN and CeN/YbN and low NbN/LaN ratios in

the 90N OSC lavas (Figure 2-8). The middle to heavy REE concentrations (i.e., Nd, Sm,

Eu, Dy, Yb, and Lu) can only be generated by > 90% crystal fractionation. Regardless,









such extreme degrees of crystallization are inconsistent with major element model

calculations.

Summarily, the calculated liquid lines of descent do not provide a good fit to the

observed major and minor compositions of the high-silica lavas, and trace element

models parameterized from the MELTS calculations do not reproduce measured trace

element abundances or trace element ratios. Thus, we conclude that extensive low-

pressure crystal fractionation is unlikely to be the sole mechanism to explain the

genesis of the high-silica lavas at the 90N OSC.

Crystal-melt segregation model

Bachmann & Bergantz, (2004) suggest that intermediate liquids (andesites and

dacites) will separate from crystals (via filter pressing) when a magma has undergone

>40-50 volume percent crystallization. The segregated melt, which is more evolved

than the original melt, will once again crystallize until it reaches 40-50% phenocrysts,

when again the new evolved melt will separate from the phenocrysts. While this

segregation model applies strictly to systems of intermediate compositions (Bachmann

& Bergantz, 2004), here we evaluate whether a basaltic magma can evolve

geochemically, through a series of segregation events, to form compositions similar to

the MOR dacites.

To simulate these conditions a 90N OSC ferrobasalt was allowed to undergo

equilibrium crystallization to andesitic compositions, using MELTS thermodynamic

calculations (Ghiorso & Sack, 1995) and starting conditions described for Rayleigh

fractionation models. At this composition, the liquid separates from the phenocrysts

(Bachmann and Bergantz, 2004), creating a new parent melt. This parent composition









becomes the new starting concentration (an andesite) for the next run, which

subsequently crystallizes 50% by volume. This process was repeated (3 times) until

MgO and SiO2 concentrations similar to the 90N OSC dacites were obtained (Figure 2-

10). This occurred after three segregation events or 87.5 wt% crystallization, however,

as was noted during the MELTS calculations, several dacitic major and minor element

concentrations (i.e. A1203, K20 and P205) could not be produced. In addition, the

calculated incompatible trace element abundances produced using this process were

also lower than those observed in the MOR dacites (Figure 2-10).

In situ crystallization calculations

A different approach to magma crystallization is in situ crystallization, where

phenocrysts do not separate from interstitial melt until a small remaining volume of

liquid is pressed from the crystallizing mush and mixed with the main body of melt (e.g.

Langmuir, 1989; Reynolds & Langmuir, 1997; Pollock et al., 2005). This process

assumes that crystallization occurs along a temperature gradient within a solidification

front (or boundary layer) and interstitial melt evolves independently from the main melt

body. The mixing of interstitial melt back into the main magma body, gradually causes

bulk increases in incompatible element abundances and changes in trace element

ratios (Langmuir, 1989). This process will cause an increase in highly incompatible

elements compared to Rayleigh crystal fractionation, because these elements are

continually returned to the residual magma body (Langmuir, 1989).

In situ crystallization was evaluated following Reynolds and Langmuir (1997), with

a starting composition of a 90N OSC ferrobasalt (265-43) using the same partition

coefficients as for Rayleigh fractional crystallization. Crystallizing phases include 01,









Plag, Cpx, and eventually, Fe-oxides. In the modeled system, the boundary layer is

always 5% of the liquid magma chamber volume and the boundary layer crystallizes

until 35% interstitial liquid remains (Reynolds & Langmuir, 1997). All of the residual

liquid mixes back into the magma chamber during each iteration of boundary layer

crystallization. As the magma chamber continues to crystallize by this mechanism, the

boundary layer moves inward leaving restite crystals behind and the volume of the

magma body decreases. Theoretically, this process will continue to modify the liquid

magma chamber composition until an infinitesimally small amount of melt is left.

After 85% in situ crystallization (Figure 2-10) calculated major and incompatible

trace element concentrations do not match those measured in the MOR dacites. This

processes can only account for Zr concentrations in the melt that are less than 3 times

the original concentration, reaching values of 320 ppm. Even > 95% in situ

crystallization cannot reproduce the enriched trace element signatures of the MOR

dacites. Although in situ crystallization does enrich the residual melt in incompatible

elements compared to Rayleigh crystal fractionation, the failure to reproduce the

observed enrichments in MOR dacites suggests that the latter cannot result from this

process either.

Partial Melting (Anatexis)

Experimental studies (Beard & Lofgren, 1991; Koepke et al., 2004) suggest that

<10% partial melting of altered oceanic crust will produce melt with major element

concentrations consistent with MOR dacite compositions. Of particular note are the high

SiO2, A1203 and K20 and low FeO, TiO2, and P205 concentrations produced by partial

melting of amphibolite facies or greenschist facies minerals because these are also the

chemical characteristics of MOR dacites. Basaltic rocks are known to undergo partial or









complete alteration and recrystallization on-axis due to pervasive high temperature

hydrothermal alteration (Alt et al, 1986; Gillis & Roberts, 1999). Such altered rocks are

an attractive starting composition for anatexis because their solidus temperatures are

much lower than fresh MORB.

Altered oceanic crust can have a wide range of trace element concentrations,

depending on the degree of alteration (Alt et al., 1986). To better evaluate the

composition of wall rock involved in the formation of dacites on the 9N OSC, we use

the batch melting equation (CI=Co/(Dbulk[1 F] + F) and andesitic partition coefficients

(Table 2-4) to solve for a range of possible parental wall rock compositions (Co) and

then compare these results to compositions of fresh and altered MORB. Trace element

patterns generated from the calculations suggest altered basalt provides a better fit than

fresh MORB as a source (wall rock) composition for the 9N OSC dacites (Figure 2-11).

Consequently, we model partial melting of an altered MOR basalt (Nakamura et al.,

2007) to determine if this process could produce the geochemical characteristics

observed in the 90N OSC lavas (Figure 2-11).

Altered basaltic wall rock is melted using two different modal mineralogies,

including one with amphibole (Haase et al., 2005) and one without (Koepke et al.,

2004). Calculated trace element concentrations resulting from 1 to 15% partial melting

of ocean crust are shown in Figures 2-7, 2-8, and 2-12. Partial melting in the absence

of amphibole (19% 01, 30% Cpx, 50% Plag, 1% Ilm) can reproduce some, but not all, of

the trace element enrichments observed in the 90N OSC dacites (Figure 2-12a). In

particular, the HREE in the 90N OSC dacites are higher than the calculated

abundances. The concentrations derived from partial melting of altered crust with









amphibole (20% Cpx, 25% Opx, 49% Plag, 5% Amph, 1% Fe-Oxide; based on modal

proportions from (Haase et al., 2005)) are much closer to the concentrations of

incompatible elements in the 90N OSC dacites, suggesting that amphibole is an

important component in the melting source rock (Figure 2-12b).

The best-fit model relies on <10% partial melting of amphibole-bearing altered

oceanic crust to produce incompatible trace element compositions comparable to those

in 90N OSC dacites. In particular, the important characteristics of this model are melts

with positive Zr and Hf anomalies, negative Nb and Ta anomalies on mantle normalized

diagrams (Figure 2-12), relatively high U and Th concentrations, and high UN/NbN and

CeN/YbN ratios (Figure 2-8).

Melting of altered basalt can also produce elevated Cl concentrations similar to

those observed in the MOR dacites (Figure 2-6). Although Cl partition coefficients are

poorly constrained, we use estimates to model Cl partitioning during melting (Gillis et

al., 2003). Using the median Cl concentration from altered basalts in ODP hole 504B

(350 ppm) as a starting composition and modal proportions described above, we

calculate a range of possible Cl enrichment for 1-15% melting to be from 0.2 wt% to >

1.0 wt%. This spans the range of Cl concentrations observed in MOR dacites (Figure 2-

6).

Assimilation Fractional Crystallization

The Energy Constrained Assimilation Fractional Crystallization (EC-AFC)

formulation of Bohrson & Spera (2001) is used to assess the role that these associated

processes play in the formation of MOR dacites. The amount of crystallization required

to produce enough heat to melt the surrounding crust is calculated and in turn, this









produces a specific mass of melt of a specific composition. Several physical parameters

are required as inputs to EC-AFC calculations (Table 2-5). These include the liquidus of

the magma (~1200C based on results of MELTS modeling of OSC lavas), the

temperature and solidus of the wall rock, and the temperature of equilibrium between

the wall rock and the magma. The initial magma composition is assumed to be N-

MORB and the assimilant is amphibole-bearing altered ocean crust with the modal

composition described above.

The wall rock may span a range of temperatures (800C to 400C) depending on

the age of the ocean crust (Maclennan, 2008). Higher initial wall rock temperatures

allow melting to begin earlier in the evolution of the magma reservoir because less

additional heat is required to raise the wall rock above its solidus temperature, i.e.,

smaller amounts of crystallization are required to initiate melting and assimilation.

However, this lowers the overall amount of incompatible trace element enrichment in

the resulting magma because the initial magma is less chemically evolved during

assimilation and larger masses of anatectic melt may be produced (Figure 2-13). For

instance, crust with an initial temperature of 800C will begin melting after 50 to 55%

crystallization and the resultant magma has a maximum of ~25 ppm La (Figure 2-13).

Antithetically, lowering the wall rock temperature decreases the total mass of wall rock

assimilated while increasing the amount of crystallization needed to initiate melting.

Consequently, this causes an increase in the overall incompatible element

concentration possible in the melt. Thus, crust with an initial temperature of 50C

requires >85% crystallization to begin melting, but results in concentrations of ~35 ppm









La in the magma (Figure 2-13). Based on these competing processes, the best-fit wall

rock temperature to generate the 90N OSC dacites is between 650 and 7200C.

The local solidus, as described by Bohrson and Spera (2001), is the solidus of the

assimilant, in this case, amphibolitized basalt. Several experimental studies have

examined dehydration melting of altered oceanic crust and amphibolites (Hacker, 1990;

Rapp et al., 1991; Wolf & Wyllie, 1994; Johannes & Koepke, 2001) but few studies were

performed under conditions comparable to those expected at MOR (Beard & Lofgren,

1991). These experiments determined that the solidus temperatures of altered basalts

are between 8500C and 900C (Beard & Lofgren, 1991). Gillis & Coogan (2002)

discuss the effects of melting altered crust at the roof of an axial magma chamber and

suggest a solidus temperature of 875C. Ti-in-zircon thermometry (90534C at TiO2

activity of 0.320.02 estimated from coexisting Fe-Ti oxides) from phenocrysts in the

Juan de Fuca Ridge dacites is broadly consistent with zircon saturation thermometry

(average of 824 150C)) and Fe-Ti oxide temperatures (~830C Schmitt et al., in prep).

Based on these combined results, 875C was used as the local solidus temperature for

AFC calculations.

The final thermal input parameter is the temperature of equilibrium, which is

defined as the final equilibrium temperature of the magma and wall rock. Generally, this

temperature should correspond to the temperature of the erupted lava. Although the

temperature of the erupted 90N OSC dacites is uncertain, the temperatures of other

dacitic magmas have been estimated. The temperature of crystallization of Galapagos

Spreading Center andesitic magma was calculate to be as low as ~910 to 940C based

on coexisting titanomagnetite and ilmenite grains (Perfit et al., 1983) and experimental









ACKNOWLEDGMENTS

I would like to thank my advisor (Michael Perfit), collaborators (W. lan Ridley,

Emily Klein, Scott White, Paul Wallace, John Valley, and Craig Grimes) and supervisory

committee (Paul Mueller, Ray Russo, George Kamenov, and David Richardson) for

their mentoring and encouragement throughout this study. Additionally, I would like to

thank the Captain, officers and crew of the R/V Atlantis for all their help during cruise

AT15-17, the MEDUSA2007 Science party (including S. White, K. Von Damm, D.

Fornari, A. Soule, S. Carmichael, K. Sims, A. Fundis, A. Zaino, J. Mason, J. O'Brien, C.

Waters, F. Mansfield, K. Neely, J. Laliberte, E. Goehring, and L. Preston) for their

diligence in collecting data and samples for this study. I thank the ROV Jason II

shipboard and shore-based operations group for their assistance in collecting these

data and HMRG for processing all DSL-120 side scan and bathymetry data collected

during this cruise. Discussions with S. White and A. Goss are gratefully acknowledged.

Thanks to G. Kamenov and the UF Center for Isotope Geosciences for laboratory

assistance and to the Department of Geological Sciences staff for all of their help.

Finally, I thank my friends and family for all their support over the years. This research

was supported by the National Science Foundation (grants OCE-0527075 to MRP and

OCE-0526120 to EMK).









partial melts resulting from melting of oceanic gabbros showed dacitic melts at

temperatures of approximately 900C (Koepke et al., 2004). Consequently, we use

900C as the input equilibrium temperature. The partition coefficients are the same as

those used for partial melting and fractional crystallization models. Basaltic partition

coefficients are used for the fractionating magma, while andesitic bulk Kd values were

used for the assimilant in the absence of a comprehensive dataset of dacite Kd values.

Results of EC-AFC calculations suggest that combinations of 73 to 85% crystal

fractionation of a basaltic magma and assimilation of 5 to 20% by mass of partially

melted wall rock produces melts that have trace element compositions consistent with

the 90N OSC dacites. In the best-fit model, melting and assimilation begins after 68%

crystallization and a further 5-17% crystallization occurs as the wall rock melt is

assimilated.

Consequently, EC-AFC trace element calculations suggest that many of the

incompatible trace element concentrations and ratios observed in the 90N OSC dacites

can be explained through this combination of processes (Figures 2-7, 2- 8, 2-14). Of

particular importance are negative Nb and Ta anomalies (relative to La), an increase in

Zr and Hf concentrations (relative to HREE), relatively flat mantle-normalized HREE

patterns, and ratios of light to heavy REE and middle to heavy REE that are similar to

those observed in MOR dacites (Figure 2-14). For instance, Zr concentrations in the

AFC models are 852 ppm and Nb concentrations are 16 ppm compared to an average

9N OSC dacite concentrations of 870 and 15 respectively. Although the overall fit of

the model data to the observed data is encouraging, the model values for Ba, Th, U and

Hf are slightly under-enriched (Figure 2-14). However, it should be pointed out that









some of the input parameters to these calculations, such as the actual degree of

alteration (and hence its composition) of the crustal assimilant, and the temperature of

the surrounding wall rock, are not well constrained and are very likely not constants.

Discussion

Petrogenesis of High Silica Lavas

Extreme crystal fractionation, partial melting of crustal material, and/or AFC

processes have been proposed as explanations for the formation of highly silicic

compositions in continental interiors, arc and ocean island settings, but only a few

studies have focused on the formation of high-silica lavas at MOR (Byerly & Melson,

1976; Perfit et al., 1983; Juster et al., 1989; Haase et al., 2005). The petrogenetic

calculations presented above demonstrate that crystal fractionation alone is not a viable

mechanism for the formation of high-silica MOR lavas, despite using a range of starting

compositions and several different end-member models (Figures 2-5, 2-7, 2-8). Instead,

results emphasize the importance of partial melting and assimilation of altered material

in the formation dacites on MOR.

Geochemical evidence of partial melting

Will crustal anatexis create geochemical signatures similar to those observed in

MOR dacites? Based on elemental systematics (e.g. CI, U/Nb; Figures 2-6 and 2-7)

and partial melting calculations presented, it appears that partial melts of altered

oceanic crust may be involved in the generation of the MOR dacites. Partial melts of

hydrothermally altered crust produces distinct signatures compared to those of

unaltered oceanic crust, due to changes in mineralogy and bulk composition during

hydrothermal circulation. Hydrothermal circulation in layer 2B or the top of layer 3 may

cause alteration to greenschist or amphibolite assemblages, where Ca-rich plagioclase









is replaced by sodic plagioclase and pyroxenes form rims or overgrowths of amphibole

(Alt et al, 1986; Coogan et al., 2003b). The degree to which this occurs depends on

temperature, water/rock ratios and fluid chemistry. To explain the geochemical

signatures in the MOR dacites, our melting assemblage must include amphibole.

Melting of amphibole-bearing assemblages, a common component in altered layer

2B (Alt et al., 1986; Coogan, 2003; Coogan et al., 2003b), can explain the anomalously

high A1203 concentrations observed in the oceanic dacites (Figure 2-5). Dehydration

partial melting experiments, where water exists only as hydrous phases within the rock,

provide a better fit to MOR dacite compositions than hydrous partial melting results

(Koepke et al., 2004). Dehydration melting experiments produce a range of A1203

concentrations (Beard & Lofgren, 1991) that are similar to or higher than MOR dacites

(Figure 2-5). Comparatively high Na20 concentrations in the dacites may result from

the melting of albitic plagioclase. Elevated Na20 concentrations are not observed in all

experimental results of Beard and Lofgren (1991) but appear to be a function of the

degree of albitization of the starting material. Variable P205 concentrations are also

observed in the experimental melts suggesting that P205 contents are very low in the

source rock or that it is a residual phase in the melting residue. Similar conclusions can

be applied to the MOR dacites, which have low phosphorous contents (Figure 2-5);

however, this may also be a function of apatite crystallization during AFC processes

(see next section). Additionally, low FeO and TiO2 suggest that Fe-oxides are not a

primary melting component in the source rock. This is consistent with our proposed

source rock comprised of olivine, plagioclase, cpx, amphibole, Fe-oxides.









High Cl concentrations in MOR dacites also support the role of partial melting of

rocks altered by seawater-derived fluids (Michael & Schilling, 1989; Michael & Cornell,

1998; Coogan et al, 2003b; Gillis et al., 2003). Although Cl behaves incompatibly

during crystallization many MOR lavas show over-enrichments compared to other

elements with similar compatibilities. For example, after ~85 % fractional crystallization

of a MORB parent (with 0.01 wt% CI), there is less than a ten-fold enrichment in CI,

resulting in concentrations of ~0.07 wt%, compared to an order of magnitude more

(~0.7 wt% CI) in the dacites. Analyses of altered basalt from sheeted dikes in drill holes

show that Cl concentrations span a range from 49-650 ppm (Sparks, 1995). Partial

melting (1-15%) of an amphibole-bearing wall rock (with 350 ppm CI) results in anatectic

melts with 0.9 to 0.3 wt% Cl covering the range observed in dacites (Figure 2-6).

Hydrothermal alteration and metamorphism are known to cause increased

concentration of some trace elements, including U, Th, Rb, and Ba, as well as Cl (e.g.

Alt & Teagle, 2003). Observed positive anomalies of some highly incompatible

elements (e.g. U and Th) relative to other incompatible elements with similar distribution

coefficients are consistent with partial melting of hydrothermally altered ocean crust

(Figure 2-12). Partial melting may also explain some of the anomalies in the high field

strength elements whose concentrations are not affected by alteration or metamorphism

but can be fractionated due to mineralogic effects. For example, the relatively low

abundances of Nb and Ta (moderately compatible during melting) and the relatively

high Zr (highly incompatible) concentrations in the high-silica lavas are a consequence

of partial melting of altered crustal material (Figure 2-12). Additionally, melting and AFC









models above point to the importance of amphibole in the melting assemblage to

explain the HREE (compare Figure 2-12a and 2-12b).

The need for crystallization, assimilation and altered crust in dacite petrogenesis

We propose that the partial melting and assimilation of oceanic crust plays a

significant role in the formation high-silica MOR lavas; however, the we stress that most

of the heat required to melt the wall rock comes from extensive fractional crystallization.

Coogan et al., (2003b) show that the latent heat of crystallization from the formation of 4

km thick gabbro sequence provides enough energy to heat ~1.3 km of overlying crust

from 450 to 11500C, which promotes partial melting. It is the assimilation of these partial

melts into a fractionally crystallizing magma reservoir that produces the highly evolved

melts with enriched incompatible trace element signatures (e.g. De Paolo, 1981; Bedard

et al., 2000).

Major element compositions of MOR dacites often lie between experimental partial

melts of altered basalt (Beard & Lofgren, 1991) and liquids produced by moderate to

large extents of crystal fractionation (Figure 2-5). The major element compositions of

magmas produced by AFC may therefore, lie between partial melts of altered oceanic

crust and fractionated basaltic magmas. This is particularly apparent in A1203 and may

explain why very few lavas at the 9N OSC (including ferrobasalts and basaltic

andesites) lie on the calculated liquid lines of descent (Figure 2-5).

Results from EC-AFC calculations (Bohrson & Spera, 2001) confirm that

assimilation of anatectic melts into a residual fractionated magma can explain a wide

range of trace element concentrations in MOR dacites (Figure 2-7, 2-8, 2-14). The

best-fit model for 90N OSC dacite compositions requires significant crystal fractionation









(73-85 wt%) of a ferrobasalt parental magma in combination with 10 to 25% (by mass)

anatectic melt, which provides the additional incompatible element enrichments

observed.

Our models also indicate that in order to explain the relative enrichments in Rb,

Ba, Th and U concentrations present in the MOR dacites assimilation of low degree

partial melts of hydrothermally altered oceanic basalt are required. Relatively low Nb

and Ta concentrations are a consequence of both the removal of phenocrysts during

late-stage crystal fractionation and residual iron-titanium oxides in the partially melted

wall rock. Elevated Zr and Hf concentrations result from little to no zircon crystallization

in the fractionating magma and/or no residual zircon in the melting assemblage. It is

important to note that the extreme Cl enrichments in MOR dacites require a seawater

component that can be derived by AFC process and that Cl over-enrichments observed

in many MORB have been explained either by small amounts of assimilation of either

hydrothermally altered ocean crust or Cl-rich brines stored in the crust (Michael &

Schilling, 1989; Michael & Cornell, 1998; Perfit etal., 1999; Coogan etal., 2003a; le

Roux et al., 2006).

Isotopic signature of assimilation

Given that the 9N OSC dacites have radiogenic isotopes similar to those in

spatially related basalts (Figure 2-9) what effects might assimilation, particularly of

altered crust, have on derivative melts? AFC processes can change radiogenic isotopic

ratios if the assimilant has relatively high concentrations of the element in question and

significantly different isotopic ratios than the original magma reservoir (e.g. Taylor,

1980; De Paolo, 1981). In general however, the effect on the resultant isotopes is less

dramatic in AFC processes compared to partial melting because AFC processes create









mixtures of altered and fresh material, while melting alone will retain the isotopic

signature of the altered crust.

Assimilation of altered oceanic crust may increase Sr isotope ratios depending on

the amount of assimilation and extents of fluid/rock interaction (e.g. Alt & Teagle, 2003).

Altered ocean crust can have a range of Sr concentrations (from less than to greater

than typical MORB compositions) depending on the type/degree of alteration (Alt &

Teagle, 2003). Assuming 80% fractional crystallization (which will not change the

isotopic ratios) and assimilation of 10 mass percent wall rock, we can calculate the

isotopic composition of the resulting melt using a ratio of 2:1 fractionatedd melt to

anatectic melt). Using reasonable values for the isotopic ratios of altered sheeted dike

lavas (0.7028; average of basal dikes from Pito Deep; Hess Deep and Hole 504B;

Barker et al., 2008) and initial MORB parental magma (0.7025) and their respective Sr

concentrations (100 ppm and 120 ppm; a typical altered East Pacific Rise

concentration) mass balance calculations indicate assimilation of altered crust will

cause an increase of ~0.0001 in the 87Sr/86Sr ratio of the final magma. EC-AFC

modeling of Sr isotopes produces similar results but requires less crystal fractionation

(73%) and a ratio of crystallization to assimilation of 0.07 to produce the most

radiogenic 87Sr/86Sr signatures observed in the MOR dacites (0.70258). Therefore, this

process has the potential to slightly affect the Sr isotope ratios in the dacitic magma, but

will not result in ratios as elevated those generated directly from partial melting of

altered oceanic crust (~0.7028). Additionally, slightly elevated Sr isotope ratios in high-

silica lavas from the Galapagos Spreading Center are consistent with AFC processes

(Perfit et al., 1999). In comparison, Nd isotopes are unaffected during fractional









crystallization and are relatively immobile during hydrothermal alteration (Michard &

Albarede, 1986; Delacour et al., 2008). Nd isotopes from the OSC dacites are similar to

basalts from the region.

9 N OSC dacites form a tight cluster in Pb isotopic composition compared to 9N

OSC basalts (Figure 2-9). Pb isotopes are not significantly affected by hydrothermal

alteration provided sediments (which are not abundant in this environment) are not

involved in the alteration process. We suggest that the similarity in isotopic ratios in the

dacites compared to the basalts represents an overall averaging of isotopic values from

basalts in the region due to melting and assimilating a range of different MORB

compositions at the base of the sheeted dike layer.

Relatively low oxygen isotope ratios observed in MOR dacites from the Galapagos

Spreading Center (Perfit et al., 1999) and the 9 N OSC (Wanless et al., 2009; Wanless

et al., in prep-a) also support these conclusions. Fresh MORB will have mantle oxygen

isotope values (~5.5), however, seawater alteration (seawater 6180 = 0) will decrease

this ratio (Gillis et al., 2001), while fractional crystallization of Fe-oxides, and to a lesser

extent olivine and pyroxene, will cause an increase (Matsuhisa et al., 1973). Therefore,

partial melting and assimilation of altered basalt should produce melts with lower

oxygen isotope ratios than predicted by fractional crystallization calculations

(Muehlenbach & Clayton, 1972). MOR dacites have oxygen isotope ratios similar to

MORB values (Perfit et al., 1999; Perfit et al., 2007; Wanless et al., 2009; Wanless et

al., in prep-a), suggesting that fractional crystallization alone cannot explain the

formation of dacites on MOR. Taylor (1968) suggests that to a first approximation, the

effect of assimilation on oxygen isotopes can be determined using mass balance.









Assuming an evolved magma has a 6180 ratio of 6.8 (largely due to fractionation of

silicates and iron oxides) and an assimilant has a 6180 ratio of 3.5 (due to seawater

alteration), the resultant oxygen isotope ratio of the AFC magma would be ~6. This

value is similar to those observed in the MOR dacites and is less than predicted by

fractional crystallization alone.

AFC Processes and Tectonic Setting

The remarkable geochemical similarity of dacites erupted at the three different

MOR discussed here indicates similar processes are controlling their petrogenesis

(Figure 2-2) and we propose these processes are linked to tectono-magmatic settings.

Andesites and dacites have erupted on several ridges, often in regions of propagation,

such as propagating ridge tips and OSC (Christie & Sinton, 1981; Perfit et al., 1983;

Sinton et al., 1983; Sinton et al., 1991). High-silica lavas have also been found along

the Pacific-Antarctic Rise (Haase et al., 2005), at Axial Seamount on the Juan de Fuca

Ridge (Chadwick et al., 2005) adjacent to a large axial magma chamber, and at the end

of a first and second-order ridge segments on the Northern East Pacific Rise (~ 8037' N

and 10030'N (Langmuir et al., 1986). Collectively, these lavas erupted on intermediate

to fast spreading ridges, in settings where magma reservoirs have the potential to

undergo extensive fractional crystallization and interact with colder, and variably altered

crust.

A key result of this study is that the geochemical signatures of MOR dacites

require assimilation of partial melts of hydrothermally altered crust into an extremely

fractionated magma (that provides the heat needed to melt the oceanic crust). The

extensive amounts of fractional crystallization required suggest episodic or sporadic









magma supply to magma reservoirs, which may not be characteristic of more "steady-

state" ridge environments. These requirements are met at the ends of ridge segments,

where magma reservoirs may have a sporadic magma supply. In these regions,

magmas are considered to be fed intermittently to the ridge tip through dike propagation

from a more robust central region (Christie & Sinton, 1981). Between diking events the

ridge tip magma supply is cut off, allowing for increased extents of crystal fractionation

and interaction of the melt with older, altered crust. This may increase the likelihood of

eruption of high-silica lavas through AFC processes. This is not to suggest that AFC

processes do not occur in "steady-state" ridge environments but that the high-silica

melts may not be preserved or erupted in these regions (see section on Effects of

Assimilation on Typical MORB).

Model for Formation of MOR Dacites

Based on the similarity of composition of high-silica lavas from three MOR,

petrologic modeling calculations applied to dacites at the 90N OSC, and published

experimental results, we suggest that MOR dacites form under specific conditions that

include: 1. A tectono-magmatic setting in which magma injection is episodic, allowing for

extensive crystal fractionation; 2. The presence of altered crust, which facilitates

geochemical enrichments observed in the MOR dacites. Based on these two

requirements, the tectonic setting and available geophysical information, we propose

the following model for dacite formation on MOR (Figure 2-15):

1. Injection of basaltic magma by lateral dike propagation. Formation of axial magma
reservoirs.

2. Magma supply to the region is cut off or reduced, allowing for extensive fractional
crystallization of the magma reservoir.









TABLE OF CONTENTS

page

ACKNOW LEDG M ENTS ........... ..... ....... ... ... ....... .. .. ......... ......... ............. ..

LIST OF TABLES ........................................ .................. 8

LIST OF FIGURES ............... ........ ............ ........ .............. .............9

ABSTRACT. ............. ......... ....... ......... .... ......... 12

CHAPTER

1 INTRODUCTION ............. ...... ............. ...... ........ ..... ......... 14

2 DACITE PETROGENESIS ON MID-OCEAN RIDGES: EVIDENCE FOR
OCEANIC CRUSTAL MELTING AND ASSIMILATION ......................... ............... 17

A abstract ................. .................................................................. 17
Introduction .............. ....... ......... ...... ..........18
G eologic and T ectonic S getting .............. .... .. .................................................... 22
9N East Pacific Rise Overlapping Spreading Center................. ................ 22
Juan de Fuca Ridge Propagating Ridge Tip and Axial Seamount.....................24
Galapagos Spreading Center- Extinct OSC or Propagating Ridge Tip? ..........25
P e tro g ra p h y ...................... .. ............. .. ..................................................2 6
G eochem ical M methods ........ ..... ........ ..................... ......... ..... ...... 26
Geochemical Results ................................ ................ 27
M ajor E lem ent R results .......................... .... ......... ........ .............. 27
Trace Elem ent Results ........ .................................... ....... ......... .... 28
Iso to p ic D a ta ...................... .. .............. .. .............................. 2 9
Petrogenetic Models For High-Silica Lavas ...................................................... 29
C crystal Fractionation .................................................................. ......... ........ ...... 30
Rayleigh fractional crystallization..................... .................... 30
Crystal-melt segregation model ................................ .. .............. 33
In situ crystallization calculations ......... ..... ..... ........ .. .............. 34
Partial M elting (A natexis) .......... .. ........ ................. .... .......... ................ 35
Assimilation Fractional Crystallization ...... .................. .................. 37
D discussion ....................... ......... .. ..... ......... ....................4 1
Petrogenesis of High Silica Lavas ................ ...... .... .... ................ 41
Geochemical evidence of partial melting .................... ..... .................. 41
The need for crystallization, assimilation and altered crust in dacite
petrogenesis .......... .... .... ............. ......... ... ....... ......... 44
Isotopic signature of assim ilation............... ...... ....... .. .............. 45
AFC Processes and Tectonic Setting ........... .... ........ ...... ....... .......... 48
Model for Formation of MOR Dacites............................... .. .............. 49
Relationship of M elt Lens to Dacites at 9 N................................... ..... ........... 50









3. During extensive fractional crystallization the released latent heat of crystallization
initially heats then partially melts the surrounding altered wall rock, which might be
layer 2b dikes or high level altered gabbros.

4. The anatectic melts are assimilated into the fractionating magma body. The AFC
melts may be of dacitic composition depending on crustal temperatures, extent of
fractional crystallization, amount of anatectic melt, and efficiency of assimilation.

This model may account for the formation of highly evolved magmas at OSC,

propagating ridge tips, ridge-transform intersections and along dikes associated with the

down-rift volcanism on Axial seamount. This situation may also be analogous to Krafla

volcano in Iceland, where the imaged melt lens is thought to be primarily composed of

iron-rich basalts but high-silica lavas are associated with the edges of the caldera rim,

where increased magma-rock interactions may be likely (Nicholson et al., 1991).

Relationship of Melt Lens to Dacites at 90N

Dacitic lavas at 9N OSC erupted on-axis, over the eastern edge of the large,

seismically imaged melt lens (Kent et al., 2000; Dunn et al., 2001). Despite the eruption

of young, fresh high-silica lavas in the neo-volcanic zone, the underlying melt lens is not

assumed to be dacitic in composition. The composition of the basalts overlying the axial

magma chamber suggests it has undergone a moderate amount of crystallization (to

ferrobasalts). This suggests that the melt lens is composed primarily of basaltic magma

that has mixed to varying degrees with a highly evolved end-member on axis (Wanless

et al., in prep-b). There is also evidence of relatively young off-axis basaltic volcanism

over the main body of the imaged melt lens (north of the overlap basin; (Nunnery et al.,

2008).

The presence of a large, seismically imaged melt lens at 9N OSC does not

contradict the episodic magma supply requirement for dacite formation. Instead, it may

enable and enhance AFC processes in the region. AFC modeling suggests that









extensive fractional crystallization is required to produce dacitic compositions, which

suggests low magma supply. Somewhat antithetically, the 9N region has an

anomalously large basaltic axial magma chamber suggesting the current melt lens may

only be indirectly related to the dacites. We envision that the large melt lens is acting as

a mobilizer for the eruption of the dacites that formed in isolated magma pockets or sills

on the eastern edge of the axial magma chamber. Additionally, small-scale local mixing

of highly evolved compositions with the moderately evolved basaltic melt lens may

account for the range of compositions erupting at the OSC (Figure 2-15).

Effects of Assimilation on Typical MORB Compositions

The formation of dacite compositions on MOR requires assimilation of anatectic

melts into residual fractionated magmas, however, AFC processes may also explain

slightly elevated incompatible elements observed in MORB lavas from all sections of

MOR. The geochemical signatures of anatectic melts may be subtle in less evolved

magmas, however, elevated CI, A1203, and K20 are common (Michael & Schilling, 1989;

Michael & Cornell, 1998; Perfit et al., 1999; Coogan et al., 2003a; le Roux et al., 2006).

At more magmatically active ridge sections, wall rock may have a higher initial

temperature, which would allow for melting and assimilation to begin earlier in the

evolution of the magma body and require less fractional crystallization to occur prior to

melting (Figure 2-13). For instance, at 8000C assimilation begins after only 53%

fractional crystallization compared to 68% crystallization for 7200C crust. Assimilation of

anatectic melts into a magma reservoir that has undergone less crystallization produces

less evolved compositions and therefore, cannot produce MOR dacites. It does,

however, increase the incompatible element abundances in the melt phase and may









explain the commonly noted anomalous incompatible element enrichments, low FeO

and elevated A1203 concentrations in some "normal" MORB lavas.

Conclusions

The majority of eruptions at spreading centers produce basalts with relatively

limited chemical variability; however, high-silica lavas have been sampled at several

ridges. Eruptions of andesites and dacites are typically associated with ridge

discontinuities and produce significant volumes of lava at a local scale. Limited

amounts of these lavas have been sampled at the southern terminus of the Juan de

Fuca Ridge, along the eastern Galapagos spreading center, at 8037'N and off-axis at

~10030'N on the East Pacific Rise. We have documented more voluminous eruptions of

high -silica lavas including highly evolved dacite on the propagating eastern limb of the

9N overlapping spreading center (OSC) on the East Pacific Rise. Collectively, the

dacites appear to represent an end-member composition that shows similar major

element trends and incompatible trace element enrichments, suggesting similar

processes controlled their petrogenesis.

The formation of highly evolved lavas on MOR requires a combination of partial

melting, assimilation and crystal fractionation. The highly enriched incompatible trace

element signatures cannot be produced through crystal fractionation alone and appears

to require partial melting of altered ocean crust. EC-AFC modeling suggests significant

amounts (>75%) crystallization of a MORB parent magma and modest amounts (5-

20%) of assimilation of hydrothermally altered ocean crust can produce geochemical

signatures consistent with dacite compositions. The AFC process explains trace

element abundances in high-silica lavas and accounts for several major and minor

element concentrations (i.e. A1203, K20 and CI).









An important constraint provided by AFC calculations is the temperature of the

assimilant. Varying the wall rock temperature can change the amount of

crystallization/assimilation that occurs and the overall enrichments observed in

incompatible trace element concentrations. The formation of dacites at the 9N OSC

requires temperatures of surrounding crust to be 650 720 oC, which requires >68%

fractional crystallization ferrobasalt magma before melting can begin. While this amount

of crystallization is unlikely in regions of high/constant magma supply, the surrounding

wall rock in typical ridge settings may be much warmer than at the ends of ridge

segments, allowing for assimilation at much lower percent of crystallization. At wall

rock temperatures of 800 oC, calculations suggest that assimilation begins after ~53

wt% fractional crystallization. This suggests that while conditions are not appropriate for

the petrogenesis of dacites at typical ridge settings, assimilation of crustal material may

be common but geochemically cryptic.

The formation of high-silica lavas on MOR appears to require a unique tectono-

magmatic setting, where episodic magma supply allows for extensive crystal

fractionation, partial melting and assimilation. These conditions are met in regions of

ridge propagation, such as OSC and propagating ridge tips, where diking allows for

episodic injection of magma into older, altered ocean crust. Here, the magma

undergoes extreme crystallization without repeated replenishment, creating enough

latent heat of crystallization to melt and assimilate surrounding wall rock.

Acknowledgements

We thank the Captain, officers and crew of the RN Atlantis for all their help

during cruise AT15-17, the MEDUSA2007 Science party (including White, S., Von

Damm, K., Fornari, D., Soule, A., Carmichael, S., Sims, K., Zaino, A., Fundis, A.,









Mason, J., O'Brien, J., Waters, C., Mansfield, F., Neely, K., Laliberte, J., Goehring, E.,

and Preston, L.) for their diligence in collecting data and samples for this study. We

thank the Jason II shipboard and shore-based operations group for their assistance in

collecting these data and HMRG for processing all DSL-120 sidescan and bathymetry

collected during this cruise. Discussions with S. White and A. Goss are gratefully

acknowledged and contributed to this research. Thanks to G. Kamenov and the UF

Center for Isotope Geoscience for laboratory assistance. This research was supported

by the National Science Foundation [grants OCE-0527075 to MRP and OCE-0526120

to EMK].










Table 2-1. Dacite major and trace element data
Sample 266- 265- 265- 265- 266- 266- 265- 265- 266- 265- 266- 265-
# 58 65 64 67 50 53 84 63 47 85 46 94
SiO2 63.0 63.8 64.0 64.1 64.3 64.3 64.4 64.4 64.5 65.0 65.0 65.2
TiO2 1.10 1.26 1.28 1.34 1.07 1.06 1.13 1.29 0.99 1.06 0.94 0.97
A1203 13.1 13.2 13.1 13.3 13.2 13.3 13.2 13.3 13.2 13.1 12.9 13.0
FeO 8.43 8.14 8.27 8.49 8.08 8.06 8.18 8.22 7.74 7.99 7.17 7.90
MnO 0.16 0.15 0.16 0.16 0.14 0.14 0.15 0.15 0.14 0.16 0.15 0.14
MgO 1.75 1.34 1.60 1.49 1.27 1.12 1.23 1.29 1.02 1.18 1.41 1.13
CaO 4.34 4.21 4.45 4.41 3.78 3.73 3.92 4.21 3.53 3.78 3.71 3.54
Na20 3.63 3.84 3.46 3.93 4.23 4.16 3.41 3.71 4.94 3.67 4.76 4.29
K20 0.96 0.97 0.97 0.95 1.10 1.09 1.19 0.99 1.22 1.22 1.19 1.14
P205 0.26 0.22 0.20 0.23 0.24 0.25 0.22 0.21 0.23 0.20 0.17 0.23
CI 0.24 0.65 0.64
Total 96.69 97.17 97.55 98.42 97.35 97.20 96.98 97.76 97.54 97.41 97.43 97.61
Trace Elements (ppm)
Li 32 34 32 30 29 30 31 31
Sc 17 20 18 15 15 17 14 13
V 122 140 121 61 102 121 73 63
Cr 13 12 12 4 2 10 4.6 1
Co 15 17 16 13 14 15 12 11
Ni 9.0 9.8 9.2 5.8 7.0 8.4 6.6 4.9
Cu 17 19 18 18 21 17 19 17
Zn 110 124 122 109 100 113 108 105
Ga 30 35 28 29 29 31 28 30
Cs 0.11 0.12 0.12 0.14 0.13 0.10 0.15 0.13
Rb 9.1 10 10 13 13 9.5 14 12
Ba 51 57 53 65 60 50 68 60
Th 1.7 1.8 1.6 2.3 2.3 1.7 2.4 2.3
U 0.59 0.65 0.64 0.86 0.82 0.59 0.92 0.84
Nb 13 15 13 16 15 13 15 16
Ta 0.81 1.1 0.92 1.0 1.4 0.99 1.1 1.2
La 23 26 23 29 27 24 29 29
Ce 67 77 68 82 78 68 82 84
Pr 10 11 10 12 11 10 12 12
Sr 76 90 89 86 70 76 81 68
Nd 46 53 46 54 51 47 52 55
Zr 735 842 622 856 968 745 872 1050
Hf 19 21 18 22 23 19 23 25
Sm 14 15 15 16 14 14 16 16
Eu 3.0 3.3 3.1 3.2 2.8 3.0 3.0 3.0
Tb 3.2 3.4 3.4 3.6 3.1 3.1 3.6 3.4
Dy 21 23 23 24 20 21 24 22
Y 132 148 142 157 133 132 154 146
Ho 4.6 4.9 4.9 5.2 4.3 4.4 5.1 4.8
Er 14 15 15 16 13 13 16 14
Yb 14 15 16 17 14 14 17 15
Lu 2.1 2.3 2.4 2.6 2.1 2.1 2.6 2.3
Tm 2.2 2.3 2.4 2.5 2.1 2.1 2.5 2.3
Gd 17 18 17 19 17 16 19 18
Pb 3.3 5.0 3.4 3.2 4.9 5.8 3.6 5.7










Table 2-1. Continued
Sample 264- 265- 265- 265- 265- 265-
# 09 70 42 83 95 40
SiO2 65.8 66.3 66.5 67.5 67.5
TiO2 0.89 0.87 0.94 0.76 0.77
A1203 13.2 13.2 13.0 13.3 13.1
FeO 7.03 7.17 7.92 6.68 6.47
MnO 0.13 0.14 0.16 0.13 0.12
MgO 1.06 0.80 0.89 0.67 0.94
CaO 3.48 3.23 3.50 2.98 3.01
Na20 4.24 4.08 3.99 3.88 4.43
K20 1.21 1.33 1.20 1.37 1.21
P205 0.20 0.19 0.21 0.16 0.15
CI 0.58 0.70 0.51 0.67
Total 97.78 97.27 98.91 97.37 97.67


Trace Elements (ppm)
Li 27 34 32
Sc 12 12 14
V 46 45 58
Cr 4 4 3
Co 10 10 11
Ni 5.7 5.0 5.9
Cu 16 16 15
Zn 89 106 119
Ga 28 29 30
Cs 0.13 0.17 0.16
Rb 13 15 14
Ba 66 73 70
Th 2.6 2.7 2.4
U 0.98 1.04 0.91
Nb 15 16 16
Ta 1.0 1.1 1.1
La 29 31 29
Ce 84 88 83
Pr 12 12 12
Sr 78 78 83
Nd 53 55 53
Zr 824 934 816
Hf 23 25 22
Sm 16 17 17
Eu 3.0 3.1 3.4
Tb 3.6 3.7 3.8
Dy 24 25 25
Y 151 160 160
Ho 5.1 5.4 5.5
Er 16 16 16
Yb 17 18 18
Lu 2.6 2.7 2.7
Tm 2.5 2.6 2.6
Gd 19 19 20
Pb 2.8 3.8 3.5


32 31 31
11 10 12
32 52 51
3 3 4
8 8 10
4.7 5.4 5.3
14 15 17
103 98 103
29 30 30
0.17 0.13 0.13
15 12 12
76 62 60
2.8 2.4 2.5
1.05 0.86 0.84
16 17 17
1.1 1.2 1.0
31 29 29
87 84 85
12 12 12
76 61 73
54 55 57
922 985 1013
25 25 25
16 16 17
3.1 2.9 3.3
3.6 3.3 3.7
25 22 24
159 145 146
5.3 4.7 5.2
16 15 15
18 15 16
2.7 2.3 2.4
2.6 2.3 2.5
19 18 20
3.8 4.1 3.6













Table 2-2. Radiogenic isotopes for 9N OSC lavas
208Pb/ 207Pb/ 206Pb/ 208Pb/
Sample 204Pb 20 error 204Pb 20 error 204Pb 20 error 206Pb 20 error
E. Limb Basalts
264-04 37.699 1.74E-03 15.476 7.89E-04 18.279 8.93E-04 2.062 2.87E-05
265-18 37.677 1.94E-03 15.472 7.33E-04 18.275 7.80E-04 2.062 3.56E-05
265-35 37.664 1.60E-03 15.470 5.59E-04 18.250 7.06E-04 2.064 3.08E-05
265-43 37.661 1.51E-03 15.467 5.45E-04 18.249 6.11E-04 2.064 3.17E-05
265-113 37.695 1.67E-03 15.477 6.33E-04 18.275 6.83E-04 2.063 2.45E-05
266-01 37.642 1.95E-03 15.469 7.30E-04 18.235 9.22E-04 2.064 2.17E-05
266-33 37.683 2.19E-03 15.474 8.55E-04 18.277 8.56E-04 2.062 2.76E-05
265-05 37.687 1.96E-03 15.478 6.26E-04 18.294 7.24E-04 2.060 3.51E-05
E. Limb Basaltic Andesites
265-24 37.683 1.68E-03 15.473 5.87E-04 18.264 6.10E-04 2.063 3.34E-05
265-56 37.681 1.49E-03 15.474 5.67E-04 18.270 6.18E-04 2.062 3.06E-05
265-91 37.683 1.40E-03 15.475 5.25E-04 18.266 5.05E-04 2.063 3.09E-05
265-103 37.674 2.06E-03 15.472 7.56E-04 18.261 8.79E-04 2.063 2.74E-05
265-109 37.673 1.79E-03 15.471 6.69E-04 18.259 7.04E-04 2.063 2.64E-05
265-125 37.672 1.90E-03 15.471 7.32E-04 18.265 7.80E-04 2.063 2.57E-05
264-20 37.690 1.54E-03 15.476 6.14E-04 18.269 5.94E-04 2.063 3.47E-05
265-49 37.689 1.60E-03 15.474 5.77E-04 18.268 6.22E-04 2.063 3.25E-05
E. Limb Andesites
264-14 37.673 2.08E-03 15.472 9.65E-04 18.262 1.11E-03 2.063 5.32E-05
265-25 37.661 1.69E-03 15.469 6.99E-04 18.251 7.47E-04 2.064 2.44E-05
265-90 37.675 1.52E-03 15.472 6.01E-04 18.262 6.60E-04 2.063 2.75E-05
265-100 37.680 1.46E-03 15.474 5.81E-04 18.265 6.24E-04 2.063 2.68E-05
266-54 37.688 1.69E-03 15.475 6.59E-04 18.272 7.73E-04 2.063 3.29E-05
E. Limb Dacites
264-09 37.676 1.78E-03 15.472 7.14E-04 18.268 8.10E-04 2.062 2.88E-05
265-40 37.674 2.05E-03 15.471 8.24E-04 18.265 8.86E-04 2.063 3.58E-05
265-42 37.678 1.72E-03 15.472 6.53E-04 18.270 7.40E-04 2.062 3.40E-05
265-64 37.682 2.51E-03 15.476 9.76E-04 18.267 1.13E-03 2.063 3.32E-05
265-70 37.676 1.96E-03 15.471 7.58E-04 18.266 7.97E-04 2.063 3.64E-05
265-83 37.681 1.51E-03 15.473 5.94E-04 18.271 6.65E-04 2.062 2.49E-05
265-84 37.689 1.67E-03 15.476 6.55E-04 18.274 7.49E-04 2.062 2.59E-05
265-85 37.677 2.05E-03 15.471 7.79E-04 18.267 8.31E-04 2.063 3.02E-05
265-95 37.680 1.67E-03 15.475 6.95E-04 18.263 8.27E-04 2.063 3.80E-05
266-53 37.678 1.32E-03 15.472 5.01E-04 18.268 5.10E-04 2.063 3.16E-05
2 sigma error reflects in-run machine error. Long term reproducibility estimates are: "'Sr/ bSr =
0.00003, 143Nd/144Nd = 0.000018, 26b/204Pb = 0.0034 (205 ppm), 207Pb/204Pb = 0.0028 (184
ppm), 208Pb/204Pb = 0.0086 (234 ppm)
*Unknowns were normalized to an 141Nd/144Nd value of 0.511215 0.000007 for JNdi-1, which is
reported by Tanaka et al. (2000) relative to a La Jolla 143Nd/144Nd value of 0.511858 (Lugmair and
Carlson, 1978).













Table 2-2. Continued

Sample 8Sr/86Sr 20 error 143Nd/144Nd* 20 error Eps Nd
E. Limb Basalts
264-04 0.70250 0.000011 0.513163 0.000011 10.2
265-18 0.70249 0.000018 0.513190 0.000013 10.8
265-35 0.70247 0.000012 0.513164 0.000005 10.3
265-43 0.70250 0.000021 0.513154 0.000009 10.1
265-113 0.70249 0.000016 0.513172 0.000005 10.4
266-01 0.70246 0.000015 0.513158 0.000007 10.1
266-33 0.70244 0.000013 0.513160 0.000005 10.2
265-05 0.70243 0.513191 0.000008 10.8
E. Limb Basaltic Andesites
265-24 0.70256 0.000013 0.513187 0.000005 10.7
265-56 0.70253 0.000012 0.513187 0.000004 10.7
265-91 0.70253 0.000012 0.513179 0.000004 10.6
265-103 0.70245 0.000012 0.513162 0.000004 10.2
265-109 0.70249 0.000016 0.513145 0.000007 9.9
265-125 0.70246 0.000045 0.513154 0.000006 10.1
264-20 0.70244 0.513179 0.000004 10.6
265-49 0.70244 0.513196 0.000007 10.9
E. Limb Andesites
264-14 0.70243 0.000014 0.513153 0.000005 10.0
265-25 0.70254 0.000018 0.513152 0.000005 10.0
265-90 0.70250 0.000011 0.513159 0.000006 10.2
265-100 0.70249 0.000012 0.513179 0.000004 10.6
266-54 0.70246 0.513193 0.000005 10.8
E. Limb Dacites
264-09 0.70254 0.000019 0.513171 0.000008 10.4
265-40 0.70247 0.000011 0.513149 0.000008 10.0
265-42 0.70246 0.000015 0.513140 0.000006 9.8
265-64 0.70258 0.000015 0.513185 0.000007 10.7
265-70 0.70248 0.000024 0.513141 0.000005 9.8
265-83 0.70253 0.000015 0.513147 0.000007 9.9
265-84 0.70251 0.000011 0.513179 0.000007 10.6
265-85 0.70248 0.000019 0.513148 0.000005 9.9
265-95 0.70247 0.000019 0.513165 0.000006 10.3
266-53 0.70247 0.000015 0.513175 0.000010 10.5













Table 2-3. Starting compositions for modeling
Sample # 264-08 265-43 265-113
SiO2 50.1 50.5 51.9
TiO2 2.68 1.92 2.17
A1203 12.7 13.9 13.4
FeO 14.1 11.6 12.8
MnO 0.26 0.21 0.23
MgO 5.69 6.98 5.93
CaO 9.58 11.14 9.48
Na20 3.27 2.86 3.28
K20 0.21 0.13 0.22
P205 0.28 0.19 0.28
C1 0.07 0.01 0.02
Total 99.24 99.67 99.72
Trace Elements (ppm)
Li 10 8 11
Sc 42 42 38
V 450 347 323
Cr 18 109 32
Co 43 41 39
Ni 34 54 34
Cu 60 59 51
Zn 116 94 103
Ga 21 18 21
Cs 0.03 0.01 0.03
Rb 2.1 1.1 2.2
Ba 17 8 16
Th 0.34 0.18 0.40
U 0.13 0.08 0.15
Nb 5.5 3.1 5.7
Ta 0.37 0.21 0.36
La 6.8 4.5 7.6
Ce 21 14 24
Pr 3.3 2.3 3.9
Sr 126 120 111
Nd 17 13 20
Zr 180 126 229
Hf 4.8 3.4 5.7
Sm 6.0 4.4 6.7
Eu 1.9 1.5 2.0
Tb 1.5 1.1 1.6
Dy 10 7 10
Y 59 44 61
Ho 2.0 1.5 2.2
Er 5.9 4.4 6.3
Yb 6.0 4.4 6.2
Lu 0.93 0.68 0.95
Tm 0.91 0.67 0.96
Gd 7.7 5.8 8.5
Pb 0.61 0.39 0.95









Effects of Assimilation on Typical MORB Compositions ................ ................51
C conclusions ..................................................... ................. ...... ... ... 52
A know led g e m e nts ................ ........................................................... 53

3 ROLE OF ASSIMILATION IN THE PETROGENESIS OF LAVAS ON MID-
OCEAN RIDGES INFERRED FROM CI, H20, CO2 AND OXYGEN ISOTOPE
VARIATIONS .............. ................................................... 79

In tro d u c tio n ............. .. ............... ................. ............................................. 7 9
Geologic Setting ................ ...................... 81
A analytical M methods and R results ................................................................................. 83
Major and Trace Elements ............... .. ......... .................. 83
V volatile Elem ents .......................... ................. ................. 84
Oxygen Isotope Analyses and Results ................................. .. .................... 85
D discuss io n ......................... ........................................ ...................................... 8 5
Magma Crystallization versus Assimilation-Fractionation-Crystallization
Processes ........... .... .... .............................. ...................... 85
Evidence for Assimilation from Oxygen Isotopes ......... .................................. 87
CO2 and H20 Degassing, Magma Ascent Rates and Depth of Assimilant........ 89
S source of A ssim ilant ....... .................................................. ...... ........ 91
O ceanic Plagiogranites .......... ................................... .... ..................... 92
C o n c lu s io n s ............. ................. ................. ............................................... 9 3

4 CRUSTAL DIFFERENTIATION AND SOURCE VARIATIONS AT THE 9N
OVERLAPPING SPREADING CENTER; EAST PACIFIC RISE...................... 102

Introduction ................ ....... .......... ......... ............. ............ 102
Background, Tectonic Setting and Geology of the 90N OSC............................... 104
O overlapping Spreading C enters ......................................... ........................... 104
Tectonic Setting and Previous Studies of 9N OSC..................................... 104
9N OSC Geology ....................... ......... ... ................ 106
8037' N EPR Deval.... ............................................................ 108
G eochem ical M methods .......................................... .. ..... .. .......... .. 109
Geochemical Results ...... .................. ................ 110
Geochemistry of East Limb Lavas................... ....... .... .... ............ 110
B a s a lts .................. ...... ..... ....................................... ............ 1 1 1
Basaltic andesites/low-P205 andesites................... .............. ................ 111
H ig h-P 20 5 andesites .... ...................................................... ........ ...... .... 112
Dacites ............................................................................. ...................................112
Isotopic com positions of East Lim b lavas.................................................. 113
W est Lim b Lavas ........... ....................... ....... .. .................. ................ 113
8037' EPR Lavas ...................... ............... .................................114
D discussion ......... .. .. ........... .. ............ ......................................................115
Shallow-level Processes Involved in the Petrogenesis of Ferrobasalts, FeTi
Basalts and Basaltic Andesites at the 9N OSC .......................................... 115
Fractional crystallization versus assimilation in the petrogenesis of
andesites ........ ................ .. ...................................... 118


6











Table 2-4. Partition coefficients for Rayleigh fractional crystallization, partial melting and AFC calculations


Andesite Partition Coefficients


Basalt Partition Coettic iient.


Element
Rb
Ba
Th
U
Nb
Ta
La
Ce
Sr
Nd
Zr
Hf
Sm
Eu
Dy
Y
Yb
Lu


Primary
Reference
Secondary
Reference


Olivine
0.01
0.01
0.01
0
0.00017
0.00002
0.00006
0.00006
0.00217
0.00015
0.00450
0.00370
0.00044
0.00056
0.00250
0.00380
0.05600
0


Zanetti et
al., 2004
Gill,
1979


Klein et Dunn and
al., 2000 Sen, 1994
Rollinson,
Gill, 1979 1993


CPX
0.02
0.02
0.01
0
0.005
0.014
0.062
0.116
0.08
0.33
0.14
0.21
0.41
0.57
0.94
0.9
0.63
0.605


Plag
0.025
0.155
0.19
0.34
0.033
0.11
0.082
0.072
2.7
0.045
0.0009
0
0.033
0.55
0.034
0.01
0.014
0.039


Apatite
0.001
0.12
1.28
1.4
0.0011
0.003
11.4
12.9
4.3
32.8
0.042
0.014
16.1
25.5
34.8
7.1
15.4
3.92
Prowatke
and
Klemme,
2006
Fujimaki,
1986


Ilmenite
0.034
0.00034
0.00055
0.0082
2
1.7
0.000029
0.000054
0
0.00048
0.29
0.38
0.00059
0.009
0.01
0.0045
0.17
0.084

Zack and
Brumm,
1998
Rollinson,
1993


Amphibole
0.04
0.1
0.15
0.008
0.28
0.27
0.027
0.0293
0.28
0.0325
0.26
0.43
0.024
0.0498
0.0136
0.0196
0.102
0


Bottazzi et
al., 1999
Haase et
al., 2005


Olivine
0.0003
0.00001
7.00E-06
9.00E-06
0.00005
0
0.0002
7.00E-05
0.00004
0.0003
0.001
0.0029
0.0009
0.0005
0.0027
0.0082
0.024
0.016

Halliday
et al.,
1995
Rollinson,
1993


CPX
0.0004
0.0003
0.0021
0.001
0.0089
0.013
0.054
0.086
0.091
0.19
0.26
0.33
0.27
0.43
0.44
0.47
0.43
0.56


Halliday et
al., 1995
Rollinson,
1993


Plag
0.056
1.45
0.13
0.051
0.045
0.066
0.13
0.11
1.4
0.066
0.048
0.051
0.054
0.65
0.024
0.013
0.0079
0.06


Dunn and
Sen., 1994
Rollinson,
1993


Imenite
0.034
0.00034
0.00055
0.0082
2
1.7
0.000029
0.000054
0
0.00048
0.29
0.38
0.00059
0.009
0.01
0.0045
0.17
0.084

Zack and
Brumm,
1998
Rollinson,
1993











Table 2-5. AFC modeling parameters
Parameter Abbreviation Value Units
Magma Liquidus Temp tim 1200 deg C
Magma Temp tmo 1200 deg C
Assm. Liquidus Temp tla 1100 deg C
Country Rx Temp tao 720 deg C
Solidus Temp ts 875 deg C


Magma Spec. Heat

Assm. Spec. Heat

Crsyt. Enthal
Fusion Enthal
Equilibration Temp


cpm

cpa

Hcry
Hfus
Teq


J/Kg
1484 K
J/Kg
1388 K


396000
354000
900


J/Kg
J/Kg
deg C

































A

Figure 2-1. Bathymetric maps showing the tectonic setting of the MOR dacites (data
from GeoMapApp; Carbotte et al, 2004). Boxes show the general locations
of high-silica lavas on each ridge. Dacites are commonly associated with the
ends of ridge segments, such as overlapping spreading centers (OSC) and
propagating ridge tips at ridge transform intersections. A) 9N OSC on the
East Pacific Rise (EPR) B) Propagating Ridge Tip on the Juan de Fuca Ridge
(JdFR) and Axial seamount and C) possible OSC near the Propagating Ridge
Tip on the Galapagos Spreading Center (GSC).



















1 .i
.4 ... .. .



... r i .. .. ::: ,: iii" "" ;. ,,
.. .". ... ..........

... .. .... ... .... .... .. .... ..:I .. "
i::


.." .: .........
.. ... ..P :...... i ..... i <. '
.' .... ...'.:.
Ii' d : ..i::,:' i:+ ... .. .. ..;.

..:iid ..... ...." "....... ...


.... .. -
.k ......... .. "
............, .. ,.. .. .. ::.. : ,, ,: ,
H.4 .'
r 17-










Fi.r 2-1 .C...inuedq
I 4;' ly
In 32w 114 x n 20 X 26W 10 24
"* ~'';1 B
Figur 2-1 Coninue




























C

Figure 2-1. Continued


















i 10 -




S1 A GSC Dacite
.E Axial Andesite
C JdF Dacite
SOSC Dacite
avg. EPR MORB
0.1
RbBaTh U NbTaLaCe Pr SrNdZr HfSmEuDy Y YbLu

3 75
B C JF daTa
5 7 OS- GC cala
2.5 A70 AxiI cata
05O aala


|60 2
NN
0 1.5 50

m| EPR MORB ^ ----
EPR B 50 EPRMORB

0.5 4 45
0 2 4 6 8 0 2 4 6 8
MgO (wt%) MgO (wt%)





Figure 2-2. Comparison of major and trace element compositions in MOR high-silica
andesites and dacites from the East Pacific Rise, Juan de Fuca Ridge, and
Galapagos Spreading Center. MOR dacitic lavas have similar major and trace
element compositions, while andesites have more variable compositions that
lie between dacites and highly evolved MORB. A) Mantle-normalized
diagram showing similarities in trace element compositions between dacites
from three different ridges and an andesite from Axial Seamount. Average
composition for N-MORB from the 9 17' 10N segment of the East Pacific
Rise is shown for comparison. MOR dacites are characterized by low Nb and
Ta and high U, Th, Zr and Hf relative to elements of similar incompatibilities.
B) and C) Major element plots showing the range of compositions of MOR
dacites compared to East Pacific Rise MORB. Gray field shows the range of
compositions from >1200 analyses of MORB glasses from the East Pacific
Rise north of the 9N OSC (from PetDB, Sims et al., 2002, 2003 and Perfit
unpublished data).











0,W 10410U'"W


9N OSC '
Rock Type
S dacite -

asalic an desie i

FeTI O
o ba,11 O li
Qi
C-P0
1 00.


A
o'


IA -




I. i'






rI
~, ,'





rj ', '




'i 'i o. ,




CI


0




e0 C
", .1, *




Oi

Lo



r.-, ,


MI
,,I') t ,;


o.
r- c o,, 'i .
o *, ', i


II


; I I ;
I i .
'' -. 4..




J ", '


I* I
* 6 ,i- i' ."*









II I :I '


-I- I'(.- -


Figure 2-3. Bathymetric map of the 9N OSC showing locations of samples collected
during the MEDUSA2007 cruise. Samples are divided into rock types based
on silica content (dacite >62wt% Si02; andesite 57-62 wt % Si02; basaltic
andesite 52-57 wt% Si02; basalts <52wt% Si02 and FeTi basalts <52wt%
SiO2 and >12 wt% FeO). Dacites are primarily found on-axis on the eastern
limb of the OSC. 50 m contour intervals are shown.


I'
*1











I
(. I





i
(t '-


! r


IM 1"%~l-


1..1 If 11 *k


1. U 1 -(0 1







































Figure 2-4. Photographs of MOR high-silica lavas from A) and B) 9N OSC East Pacific
Rise, C) Galapagos Spreading Center and D) Juan de Fuca Ridge.
Morphologically, dacites typically form blocky angular flows and large
elongate pillow lavas with roughly corrugated or striated surfaces.












^ OSCLavas OSC Lavas
Is 12
o0 8 -o1 O-
,,- < 10 -,
4. B
IA B
-- FC parent
15 FC parent
1.5 FC parent
Sk B&L PMelts
I W OSC Lavas 1 OSC dacites
0 / 0 OSC Lavas
0.5 0.5
*- C D






4.5 0fF + 5 A.
0 2 2o


35 + 2 4 6 8Lavas
MgO (wt%) MgO (wt%)


Figure 2-5. Major element variations versus MgO (wt.%) for dacites from the 9N OSC
on the East Pacific Rise (filled squares). Gray field represents all lavas
collected in 2007 from the 9N OSC. Dacites are compared to three low-
pressure fractional crystallization trends (calculated using MELTS; Ghiorso &
Sack, 1995) using parental compositions from OSC basalts (see text for
modeling parameters and details). Not all dacite major element variations
can be explained by fractional crystallization alone (e.g. A1203, K20, and
P205). Experimental compositions from partial melting of altered basalt (Beard
and Lofgren, 1991) are shown for comparison (*).










I 1-15% partial
I-4-- melts
0.7- m-ts FC parents
I FC parent
'- FC parent
0.6 I OSC dacites

0.5


s 0.4 I
range of starting ma-
S|) I trial for partial melt-
0.3 ing (median value of
350 ppm CI)
0.2 -
OSC Lavas

0.1 85% .
S,800b 70
+ 70% -f_ __________
0
0 .....---,--1-, -- -= -r -, -1-
0 1 2 3 4 5 6 7 8
MgO (wt%)

Figure 2-6. Variation diagram showing Cl (wt%) versus MgO (wt%) for OSC lavas.
Superimposed are three liquid lines of descent (calculated using Melts;
Ghiorso & Sack, 1995), showing that the maximum amount of Cl enrichment
due to extensive fractional crystallization cannot produce the high Cl
concentrations in the OSC dacites. The dashed rectangle represents the
range of compositions that can be produced through 1-15% partial melting of
an altered basalt with 350 ppm Cl (purple star). 350 ppm Cl is the median
value of Cl analyzed in sheeted dikes from ODP Hole 504B with minimum
and maximum values (49 and 650 ppm) shown with an error bar. Partition
coefficients for Cl are from Gillis et al., (2003). The range of MgO values for
partial melts were taken from experimental partial melts of less than 15%
(Beard & Lofgren, 1991).









Formation of west limb lavas by fractional crystallization ........... .......... 119
W here Are the South Tip and W est Limb Dacites?..................................... 120
Composition of the Melt Lens Beneath the East Limb................ ..... ........ 122
E-MORB Distribution at 9N OSC ........... ......... ......... ................123
9N OSC as a Division in Mantle Components............ .... .................... 125
C conclusions ........... ... ............ ................. .. .......... ............. 126

5 CONCLUSIONS.......................................... 161

LIST O F REFERENC ES ............. ..... .......... .. ........ .. ..... ............ .. 164

B IO G R A P H IC A L S K E T C H .................................................... ...................................... 18 1







































7













50 -v- AFC .
Dacite -"
..... 2 .5 -
40 --PMelt 5% .2.5 I
Modell 2 2%
S30 -- Model2i L 1.5 5%






C D
12 2% A FC
10-05



10- 200


E 0.8 i% 150
0.6 0 CL
0.6 AFC > 100 .-"
0.4
50
0,2 50 5 8%FC
,,. 85% F.C.
E 2% F
120 70

1000 5% 60
*50-
80 E 40
S a. 40
30
40 -
840 F.-- -.^" 'AFC 20
20 10 N5% F.ac...., I I

0 500 1000 1500 0 500 1000 1500
Zr (ppm) Zr (ppm)





Figure 2-7. Trace element variations versus Zr (ppm) in 9N OSC lavas. Superimposed
on the diagrams are calculated trends for fractional crystallization (model 1
and 2 using the Rayleigh fractionation equation), 1 to 15 % partial melting
(assuming batch melting), and AFC simulations (EC-AFC; Bohrson and
Spera, 2001). See text for model parameters. Kinks in models represent
changes in crystallizing phases.














































1.2 1.4
CeNYbN


Figure 2-8. Normalized trace element ratio diagrams showing the range of dacite
compositions at the 9N OSC. Fractional crystallization, partial melting and
AFC trends are shown as in Figure 2-4. Tick marks on trend lines for partial
melts are in 1% increments, while fractional crystallization tick marks
represent 10% intervals. Concentrations are mantle-normalized (Sun &
McDonough, 1989).





71














37.95-

37.90-

S 37.85-


a-
o
. 37.80
c.
00
N 37.75-

37,70

37.65-
18.25


X OSC N-MORB
0 Dacites


18.30


18.35


18.40


N-MORB EPR


11.0
10.8 x
10.6 x
10.4
10.2 x
10.0 -
9.8 -- a
9.6
9.4
9.2
9.0
0.7024 0.70245


X OSC N-MORB
U Dacite


0.7025 0.70255

87Sr/86Sr


0.7026


0.70265 0.7027


Figure 2-9. Radiogenic isotope ratios of 9N OSC lavas. A) Pb-isotope ratios showing
9N OSC dacites with 208Pb/204Pb and 206Pb/204Pb ratios similar to OSC N-
MORB basalts and northern East Pacific Rise N-MORB (gray field; data from
(Sims et al., 2002; Sims et al., 2003; Goss et al, 2010). B) Nd and Sr
isotopes with 9N OSC lavas that are similar to EPR and OSC N-MORB.









4 40
A C
3.5 35




o2 G7 20
oo 0 \
F- 1.5 1 15

1 OSC Lavas 10 OSC Lavas
0.5 5
BII 1D
B D
1.2 B L+ insitu
crystal-mell 200
.- segregation
1 m dacites

0.8 OSC Lava5 150
S0.8 I
v 0.6> 100
0.4
50
0.2 50
OSC Lavas
I----I----I--* -- 0 ---- ---- *---
0 2 4 6 8 10 0 500 1000 1500
MgO (wt%) Zr (ppm)


Figure 2-10. Elemental variation diagrams showing 9N OSC dacites versus calculated
liquid lines of descent produced during two alternative types of crystallization.
Crosses show the evolution of a melt during in situ crystallization (Langmuir,
1989). Circles show the liquid line of descent of a magma that undergoes
50% fractional crystallization and is then separated (via filter pressing) from
the phenocrysts (melt-segregation model). The resulting magma undergoes
an additional 50% crystallization, until it once again separates from the
crystals following the model of Bachman and Bergantz (2004). Kinks in
models represent changes in crystallizing phases.











Source Models


OSC Dacite


range of source
material


Average EPR MORB
Altered Basalt 1
Alatered Basalt 2


U.1,
Rb Ba Th U Nb Ta La Ce Sr Nd Zr Hf Sm Eu Dy Y Yb Lu

Figure 2-11. Partial melting model showing a range of possible parents (gray lines) that
could produce the 9N OSC dacites (bold red line) from 1-15% partial melting.
Sheeted dikes and the upper parts the gabbroic layer may be composed of a
range of compositions depending on the composition of the starting material
and degree of alteration. Possible parental compositions were calculated
using the batch melting equation and solving for the initial parent composition
(Co). Three possible wall rock compositions are superimposed on the
calculated parental range. Altered basalt 1 (squares; (Nakamura et al., 2007)
provides the best match to the calculated source compositions and is used as
the parent rock in subsequent partial melting and AFC models.


" '














a 100



E
S10



E


I1

0.1


B


Amphibole-bearing assemblage


I t I I I


Rb Ba Th U


I I I I I I I I I


Nb Ta La Ce Sr Nd Zr


Hf Sm Eu Dy Y


Figure 2-12. Mantle-normalized diagrams showing results of 1-15% partial melting of an
altered basalt (see Figure 2-11) using two different modal minerologies.
Partial melts were calculated using the batch melting equation. See text for
more detail. A) Partial melting results using a non-amphibole-bearing gabbro
assemblage (19% 01, 30% Cpx, 50% Plag, 1% lIm). B) Partial melting results
using an amphibole-bearing gabbro (20% Cpx, 25% Opx, 49% Plag, 5%
Amph, 1% Fe-Oxide; Haase et al., 2005). The amphibole-bearing gabbro
provides the best fit to the 9N OSC dacites.


I I I I I I I I


1-15% P.M


100


IYb
Yb Lu


I(IIIIII



















S15
10



0 0.1 0.2 0,3 0.4 0.5 0.6 0.7 0.8 0.9 1
Mass of crystallization (Mc)
40 r T T T
B


30Mas off A i~ -La Concentration of Dacites

b s e u fc25l A +++i at o isw e Tev
1E A +++
S 2 ++ XXxxx
1 ++ xX.





0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Mass of Assimilation(Ma)/Mass of crystallization (Mc)

Figure 2-13. Diagram showing the calculated effects of varying wall rock temperature on
incompatible trace element composition (La) during AFC. A) Higher wall rock
temperatures cause earlier onset of assimilation compared to lower initial
temperatures. Higher initial temperatures produces lower overall
incompatible element abundances compared to lower wall rock temperatures
because the amount of fractional crystallization is lower. The average La
concentration of the 90N OSC dacites is -28ppm, which can be produced by
assimilating partial melts of a wall rock at starting temperatures of 650 to
720C. B) Plot showing the ratio of assimilation to crystallization (Ma*/Mc) for
various temperatures of wall rock. Lower ratios produce higher incompatible
trace element concentrations in the hybrid melts.












S100 Assimilation
OSC Dacite and Fractional
SCrystallization






Fractional
t Crystallization





1m


Rb Ba Th U Nb Ta La Ce Sr Nd Zr Hf Sm Eu Dv Y Yb Lu

Figure 2-14. Mantle-normalized trace element diagram showing results of the best-fit
AFC model (thin black lines). This model requires a total of 73-85% fractional
crystallization in combination with 5-20% assimilation of partially melted wall
rock to produce trace element compositions similar to the 9SN OSC dacites
(bold red line). Fractional crystallization is the dominant process until 68% of
the magma has crystallized. This is followed by 5-20% assimilation of partial
melts in conjunction with an additional 5-17% crystallization.


I I I I I I I I I I I I










Middle Ridge Segment


Hvdrothermal alteration


End of Ridge Segment: OSC or
PRT


Melting of wall rock


B \L\


'I
PI?^I~ ,
-^
f e knichdI


j ~Aband dd
'" I I .. ii-. c-- l.i\_g '
.x ,, ."/

i ,, ., i -.-


Melting and assimilation


ll r_ ,
- \ Hybr dDaciti \
-~~~~ ~ ~ -c 2 /l-'~ii-


Figure 2-15. Cartoon showing a possible scenario for dacite formation on MOR. A)
Injection of basaltic magma into a ridge segment end through dike
propagation. B) Magma supply rates diminish at segment end, abandoning
pockets of magma and allowing for extensive fractional crystallization. The
latent heat of crystallization begins to heat up and partially melt the
hydrothermally altered wall rock. C) Partial melts of wall rock are assimilated
or mixed into the evolving magma chamber, resulting in dacitic magmas.


Lavas





\ 1\ \
- / %


A N









CHAPTER 3
ROLE OF ASSIMILATION IN THE PETROGENESIS OF LAVAS ON MID-OCEAN
RIDGES INFERRED FROM CL, H20, CO2 AND OXYGEN ISOTOPE VARIATIONS

Introduction

Crustal assimilation has been proposed as an important process in the

petrogenesis of mid-ocean ridge (MOR) magmas (e.g., O'Hara, 1977; Michael &

Schilling, 1989; Michael & Cornell, 1998), but it is largely ignored as a primary igneous

process in ridge settings for several reasons. First, contamination is commonly

overlooked because most geochemical variations in MOR lavas can be readily

explained by fractional crystallization, variations in mantle melting parameters, or

differences in mantle source compositions. Second, the low volume of assimilated melt

compared to the more voluminous mid-ocean ridge basalt (MORB) magmas makes

assimilation a difficult process to identify in the erupted lavas. Third, the magma and

wall rock may have similar major and trace element compositions, resulting in liquids

(assimilants) that are geochemically difficult to discriminate from typical MORB lavas.

However, variable degrees of hydrothermal alteration of basaltic crust can produce

significant changes in fluid mobile element concentrations and isotopic ratios depending

on the water/rock ratios (e.g., Alt & Teagle, 2000). Therefore, components that are

particularly sensitive to seawater interaction, such as CI, U, H20, and oxygen isotopes,

can be used to determine the extent to which crustal assimilation is involved in MOR

magmatism.

Contamination of MOR magmas by a seawater-altered component was first

proposed based on excess Cl in MORB glasses (Michael & Schilling, 1989; Michael &

Cornell, 1998). The elevated Cl concentrations compared to elements of similar

incompatibility (K, Nb, Ti) in fresh MORB glass cannot be explained by post-eruption









LIST OF TABLES

Table page

2-1 Dacite major and trace element data............ .... ......... ............................ ......... 55

2-2 Radiogenic isotopes for 9N OSC lavas...................................... ............. 57

2-3 Starting compositions for geochemical modeling....................... .............. .. 59

2-4 Partition coefficients for Rayleigh fractional crystallization, partial melting and
A FC calculations ............... .................................... ... .... ...... .......... 60

2-5 A FC m odeling param eters ......................................... .................... ................ 61

3-1 G eochem ical data ................ ..................... ................................ 95

4-1 East limb major element data 9N OSC ........... ......... .. .. ................ 128

4-2 Trace element data 9N OSC ....... .......... .............. .................. 136

4-3 W est limb major element data 9N OSC........ ........... ...................... 140

4-4 W est limb trace element data 9N OSC ........... ......... .. .. ................ 142

4-5 W est limb isotopic data 9N OSC ....... ........ ........... ......... ...... ......... 143

4-6 Major and trace element data from 8037'N EPR ................ ................ 144

4-7 Radiogenic isotope ratios 8037'N EPR ...................................... .................... 145









alteration or fractional crystallization and are instead attributed to assimilation of a

seawater-derived component such as saline brines or altered ocean crust (Michael &

Schilling, 1989; Michael & Cornell, 1998). Consequently, Cl over-enrichment has been

identified in many submarine settings, including MOR (Perfit et al., 1999; Coogan et al.,

2003b; le Roux et al, 2006; Wanless et al., 2010; Wanless et al., accepted), back-arc

basins (Kent et al., 2002; Sun et al., 2007) and ocean islands (e.g., Kent et al., 1999).

Despite the clear evidence of assimilation in these lavas, many models for the

magmatic plumbing system at MORs continue to ignore this process.

An alternate approach to identifying crustal contamination on MORs is through

analyses of oxygen isotope ratios. Low oxygen isotope ratios relative to mantle values

are observed in many lavas from Icelandic volcanoes (e.g., Gautason & Muehlenbach,

1998). Lower than mantle values have also been observed in melt inclusions from

Hawaiian glasses and are attributed to assimilation of altered Pacific crust (Eiler et al.,

1996; Garcia et al., 1998). Decreases in 6180 relative to primary mantle-like values are

attributed to the temperature dependent fractionation of oxygen isotopes between

seawater and mineral phases in ocean crust during high-temperature hydrothermal

alteration (Muehlenbach & Clayton, 1972; Alt et al., 1996). Therefore, assimilation of

crust altered at high-temperatures near MORs should lower the oxygen isotope ratios of

the resulting magma. Unfortunately, this signal may be hard to identify in many MORB

magmas because the mass of assimilant relative to parent magma is generally not

sufficient to appreciably change the oxygen isotope ratios. However, it may be an

important discriminator in magmas that have undergone significant assimilation.









Here, we examine CI, H20, and C02 concentrations and oxygen isotope ratios

from a suite of lavas collected at the 9N overlapping spreading center (OSC) on the

East Pacific Rise (EPR) to place constraints on the role of crustal assimilation at MORs.

This suite includes a nearly continuous range of compositions from basalts to dacites,

including one of the most evolved lava compositions sampled on a MOR (>67 wt%

SiO2). Major and trace element data indicate that assimilation of altered ocean crust is a

critical process for the formation of MOR dacites (Wanless et al., accepted) and as

such, it should be reflected in the CI, H20, and C02 concentrations and oxygen isotope

ratios of these lavas. Additionally, we use geochemical results to better define the

source and depth of assimilation processes beneath MORs and use C02 concentrations

to provide information on degassing of magmas during extensive fractional

crystallization and the ascent rates of high-silica magmas.

Geologic Setting

The 9N OSC is located between the Clipperton and Siqueiros transform faults on

the EPR (Figure 3-1). It is a second order ridge discontinuity consisting of limbs that

overlap by ~27 km and offset the ridge axis ~8 km from east to west (Sempere &

Macdonald, 1986). The OSC has been migrating southward down the ridge axis at a

rate of approximately 42 km/Myr as the eastern limb propagates into older crust and the

western limb recedes or dies (Macdonald & Fox, 1983; Carbotte & Macdonald, 1992).

The 9N OSC is one of the most extensively studied OSC on the MOR system. It

has been the focus of several geophysical studies (Detrick et al., 1987; Harding et al.,

1993; Kent etal., 1993; Kent etal., 2000; Bazin etal., 2001; Dunn etal., 2001; Tong et

al., 2002), which have produced the first 3D multi-channel seismic survey of a mid-

ocean ridge (Kent et al., 2000) and a 3D seismic refraction study (Dunn et al., 2001).









These studies reveal a shallow melt lens beneath both limbs of the OSC and in the

interlimb region north of overlap basin (Kent et al, 2000). The western, receding limb

melt lens is narrow and shows no significant variation in depth along axis (Kent et al.,

2000), while the melt lens beneath the eastern, propagating limb shows variations in

both width and depth. Beneath the southern portion of the east limb the melt lens is

narrow and significantly deeper than the rest of the eastern ridge axis, plunging ~500 m

southward over ~6 km (Kent et al., 2000). North of the overlap basin, the melt lens is

anomalously wide (> 4 km) and is not centered directly below the ridge axis, instead

extending from the axis ~4 km to the west (Kent et al., 2000; Tong et al., 2002).

Although the depth of the lens varies along axis, the top of the melt lens appears to

follow the base of the sheeted dikes, at approximately 1.5-2 km beneath the seafloor

(Kent et al., 2000; Tong et al., 2002).

The first lava sampling in this region occurred during the CHEPR dredging and

wax coring cruise that recovered several high-silica lavas, along with basalts and FeTi

basalts (Langmuir et al., 1986). More recently, the 9N OSC was the focus of the

MEDUSA2007 research cruise (AT15-17), which completed detailed mapping using the

DSL-120A side-scan system (White et al., 2009), and the WHOI TowCam (Fornari,

2003) and extensive sampling using the ROV Jason2, (Wanless et al., accepted;

Wanless et al., in prep-b) of the region. Results of this cruise (White et al., 2009;

Wanless et al., accepted; Wanless et al., in prep-b) revealed a range of rock types from

basalts to dacites, but the high-silica lavas are confined to the eastern propagating limb

of the OSC (Wanless et al., accepted). Additionally, the basalts erupted at the OSC are









dominantly ferrobasalts, in contrast to the dominantly MORB lavas erupted from the

9015 to 100 N section of the EPR (Perfit etal, in prep).

Analytical Methods and Results

Major and Trace Elements

During the MEDUSA2007 research cruise, 275 glassy samples were collected

from the 9N OSC. Methods, standards and results from all major and trace element

analyses are discussed in detail in Wanless et al., (accepted) and Wanless et al., (in

prep.). Major elements were analyzed on a JEOL 8900 Electron Microprobe at the

USGS in Denver, Colorado. High-precision Cl and K20 concentrations were determined

using 200-second peak/100 second background counting times. Samples were

analyzed for trace element concentrations on an Element2 Inductively Coupled Plasma

Mass Spectrometer (ICP-MS) at the University of Florida. Concentration data and ratios

used in this paper are listed in Table 3-1.

Here, we use the general term "basalt and basaltic" to include basalts sensu

strict, ferrobasalts and FeTi basalts. The intermediate lavas include both basaltic

andesites and andesites, with Si02 concentrations that range from 52-57 wt% and 57-

62 wt%, respectively. Dacites have >62 wt% Si02. Cl concentrations range from 0.01 to

0.07 wt% in the basalts and are generally higher in the basaltic andesites (0.01 to 0.31

wt %), andesites (0.20 to 0.42 wt %) and dacites (0.23 to 0.70; Figure 3-2). K20

concentrations range from 0.13 to 1.37 wt % and generally increase with increasing

Si02 content. Basalts range from 0.13 to 0.21 wt%, basaltic andesites range from 0.26

to 0.60 wt%, andesites range from 0.63 to 0.83 wt% and dacites range from 0.89 to 1.37

wt% (Figure 3-2). CI/K20 ratios range from 0.04 to 0.60 (Figures 3-3 and 3-4).









Volatile Elements

A representative subset of 20 samples (covering the range of rock types) from the

9N OSC was selected for volatile analyses (Table 3-1). Several glass chips were

handpicked from each sample, avoiding alteration and microphenocrysts. Samples were

analyzed for H20 and CO2 concentrations by Fourier transform infrared (FTIR)

spectroscopy at the University of Oregon (Johnson et al., 2009). Water concentrations

were calculated either using the fundamental OH- stretching vibration at 3559 cm-1 or

from the average of the two molecular water peaks (1630 cm-1 and 5200 cm-') and the

4500 cm-' OH- peak. An absorption coefficient of 63 L/mol cm was used for the 3550

cm-' peak (Dixon et al., 1995b; Dixon et al., 1995a), and absorption coefficients for the

near-IR peaks were calculated based on major element concentrations following

methods in Mandeville et al., (2002). CO2 concentrations were measured using the

carbonate peaks at 1515 and 1430 cm-', using background subtraction procedures

described in Johnson et al., (2009) and absorption coefficients calculated from Dixon &

Pan (1995).

Volatile concentrations are highly variable in the OSC lavas, but are consistent

with major element trends (Table 3-1). H20 concentrations range from 0.23 to 0.39 wt%

in basaltic lavas and from 0.24 to 1.56 in basaltic andesite samples (Figure 3-2).

Andesites have H20 concentrations ranging from 0.99 to 1.50 wt% and dacites range

from 1.53 to 2.35 wt%. CO2 concentrations in basalts range from 131 to 256 ppm

(Figure 3-2). Two of the basaltic andesites have CO2 concentrations (232 and 184

ppm) but otherwise all others had CO2 concentrations below the detection limits of ~25

ppm. H20/Ce ratios, involving elements of similar magmatic incompatibility, generally

increase with increasing silica and range from 0.014 to 0.024 (Figure 3-3), while









H20/K20 ratios vary from 0.095 to 2.59, with the highest ratios observed in the basaltic

andesites (Figure 3-4).

Oxygen Isotope Analyses and Results

Oxygen isotope ratios (6180, per mil notation) were determined on 26 fresh,

microphenocrysts-free glass chips that cover the range of rock types (Table 3-1) at the

C02-laser-fluorination laboratory at the University of Wisconsin, Madison, following

methods described in Valley et al., (1995). Aliquots of 2.4-3.2 mg were treated with BrF5

overnight, and then individually heated with a CO2 laser in the presence of BrFs.

Measurements were standardized with 4-5 analyses of UWG-2 garnet standard per day

(5180 = 5.8%o; Valley et al. 1995), and are reported in standard 5-notation relative to

Standard Mean Ocean Water (SMOW). Reproducibility of the standard during each

session was better than 0.15%o (2SD). The 6180 values range from 5.31 to 6.19 %o in

the OSC lavas with an average of 5.79 %o (Figure 3-5). The basalts and basaltic

andesites have similar oxygen isotope ratios, ranging from 5.51 to 5.79 %o and 5.31 to

5.92 %o respectively. The andesites and dacites have variable "180 (5.38 to 6.19 %o),

with a mean of 5.86 %o.

Discussion

Magma Crystallization versus Assi milation-Fractionation-Crystallization
Processes

Liquid lines of descent (LLD's) were calculated using the MELTS thermodynamic

modeling program to simulate fractional crystallization at fO2 = QFM and P = 1 kbar

(Ghiorso & Sack, 1995). These results suggest that fractional crystallization alone

cannot account for the high Cl and H20 concentrations observed in the MOR dacites,

andesites or basaltic andesites (Figure 3-2). H20 concentrations in the dacites are as











much as two times greater than model predictions. Similarly, Cl concentrations

observed in the dacites are more than ten times greater than model predictions (Figure

3-2). H20/Ce ratios, which should not change over a wide range of fractional

crystallization, are generally higher than expected in the MOR dacites, having almost

two times higher ratios than the basalts, whereas basaltic andesites and andesites

show a wider range of ratios (Figure 3-2a).

The high Cl concentrations, however, are similar to those produced by 1-15%

partial melting of altered basaltic crust (Wanless et al., accepted). While both H20 and

Cl exhibit over-enrichments compared to calculated fractional crystallization trends

(Figure 3-2), the Cl over-enrichment is much greater than that of H20. The difference

between these enrichment factors may be due to the higher Cl contents in the altered

crustal source and, to a lesser extent, variable amounts of H20 degassing during

magma ascent and eruption on the seafloor (see degassing section below). In contrast

to H20, Cl remains soluble in most basaltic and andesitic lavas at eruption depths

greater than 700 meters below sea level (Unni & Schilling, 1978; Webster et al., 1999)

and therefore, does not degas.

Le Roux et al., (2006) used CI/Nb ratios to assess the role of assimilation in

MORB magmas because Cl and Nb have similar partition coefficients in basaltic

systems. Cl is enriched in seawater-altered crust while Nb remains immobile during

seawater alteration, which allows for discrimination between the effects of fractional

crystallization and assimilation. However, the advanced fractional crystallization in the

OSC lavas results in precipitation of Fe-Ti oxides in which Nb is a compatible element,

so the CI/Nb ratio has limited applicability in the OSC lavas. Here, we use CI/K20









(Figure 3-3) because these elements also have broadly similar incompatibilities over a

wide range of crystallization (e.g., Kent et al., 1999) and this ratio has been used in

several studies to identify crustal contamination (e.g., Michael & Cornell, 1998; Kent et

al., 1999). K20 concentrations are not affected by high temperature alteration in the

lower sheeted dikes and upper gabbros (Alt et al., 1996). Many of the MOR dacites

have CI/K20 ratios that are greater than five times the ratios observed in the spatially

related basalts (Figure 3-3). Additionally, many andesites and basaltic andesites

erupted at the OSC have CI/K20 ratios three to seven times higher than the basalts

erupted in the region and higher than mantle values (0.065) predicted by Michael and

Cornell (1998), suggesting they have also been contaminated by crustal material.

These observations clearly suggest the operation of assimilation in the formation of

basaltic andesites, andesites and dacites on MOR.

Evidence for Assimilation from Oxygen Isotopes

Crystallization of Fe-Ti oxides leads to a significant increase in oxygen isotope

ratios of evolving magmas (Taylor, 1968) because these phases and to a lesser extent

Fe-Mg silicate phases preferentially incorporate 160 relative to 180 compared to the

remaining melt (Taylor, 1968; Anderson et al., 1971; Muehlenbach & Byerly, 1982).

During crystallization of MORB magma there is a slight increase in the 8180 as the

magma crystallizes ferromagnesian silicates, followed by a dramatic increase when Fe-

Ti oxides precipitate (Matsuhisa et al., 1973). Fractional crystallization can result in an

increase of 6180 values by 1-1.5%o (e.g., Muehlenbach & Byerly, 1982).

Using modal mineral proportions calculated from MELTS (Ghiorso & Sack, 1995)

and oxygen isotope fractionation factors from Bindeman et al., (2008), we have









calculated the LLD for 6180 during fractional crystallization of a MORB parental magma

(Figure 3-5). These calculations suggest that if the MOR dacites formed through

fractional crystallization the 6180 should be ~6.8%o; however, the measured ratios are

~1%o lower (dacite average = 5.86%o), supporting the conclusion that the dacites could

not form from fractional crystallization alone. Similar observations have been made for

dacites erupted at the Galapagos Spreading Center (6180 = 3.9-6.2%o; Perfit et al.,

1999).

The temperature-dependent fractionation of oxygen isotopes between seawater

and mineral phases causes decreases in 6180 of ocean crust relative to primary mantle-

like values during high-temperature hydrothermal alteration, but increases 6180 during

low-T alteration (<200-250C; Muehlenbach & Clayton, 1972; Alt et al., 1996).

Additionally, as seawater migrates through and reacts with the crust at temperatures <

200-2500C, the 6180 of that water evolves to lower values. This results in an overall

decrease in 6180 with depth in the crust. Profiles of oxygen isotope ratios in exposed

sections of ocean crust (Hess Deep; Agrinier etal., 1995) and drill holes (e.g., ODP Site

504; Alt et al., 1996) show that 180 values are higher than unaltered MORB

compositions (~5.6 +- 0.2 %o; Eiler, 2001) in the upper volcanic section of the crust, but

decrease to lower than the mantle values in the lower sheeted dikes and gabbros (for

profiles see Alt & Teagle, 2000). Therefore, if assimilation occurs at the base of the

sheeted dikes or in the upper gabbros, it should lower the oxygen isotope ratios of the

resulting magma.

High temperature hydrothermal alteration of the ocean crust results in an observed

decrease in the 6180 of the sheeted dike and gabbro layers to ~ 4%o (Alt et al., 1996)









compared to mantle values of 5.6%o (e.g., Alt etal., 1986; Eiler, 2001). Therefore,

partial melting of this material will produce melts with oxygen isotope ratios less than

mantle-like values, whereas assimilation of this material into fractionating magma will

produce magmas with oxygen isotope ratios between altered basalt and that expected

from fractional crystallization alone. The l180 value of the MOR andesites and dacites

range from 5.38 to 6.19%o, which is much more variable than the spatially related

basalts and lower than calculated fractional crystallization trends (Figure 3-5).

Taylor (1968) suggested that to a first approximation, the effect of assimilation on

oxygen isotope ratios can be determined using mass balance equations. Using results

from Energy Constrained-AFC petrologic modeling calculations (Bohrson & Spera,

2001), the ratio of fractionating magma to assimilant in the dacites is 2:1(Wanless et al.,

accepted). Assuming the evolved magma has a 6180 of 6.8 (largely due to fractionation

of silicates and iron oxides) and a crustal assimilant with a c180 of 4 (due to high

temperature seawater alteration; Alt et al., 1996), the resultant oxygen isotope ratio of

the AFC magma would be ~5.9. This value is similar to the average of ratios observed

in the MOR dacites and is less than predicted by fractional crystallization alone.

CO2 and H20 Degassing, Magma Ascent Rates and Depth of Assimilant

Although the geochemical data are consistent with crustal assimilation, these

signatures do not constrain where this contamination occurs. The depth of equilibration

of vapor-saturated melts can be calculated using the H20 and C02 solubility model of

Dixon et al., (1995a,b). This model is based on experimental results at pressures and

temperatures similar to MOR magmatic conditions. A potential complication of this

modeling is that C02 and H20 may undergo variable degassing during ascent from the

magma chamber to the seafloor, which is caused by slow diffusion of C02 to the









LIST OF FIGURES


Figure page

2-1 Bathymetric maps showing the tectonic setting of the MOR dacites .................62

2-2 Comparison of major and trace element compositions in MOR high-silica
andesites and dacites ............. .... ...................... ........ ...... .. .. .......... ..... 65

2-3 Bathymetric map of the 9N OSC showing locations of samples collected
during the M EDUSA2007 cruise...................................................................... 66

2-4 Photographs of MOR high-silica lavas ....... ............................................ ... ............ 67

2-5 Major element variations versus MgO (wt.%) for dacites from the 9N OSC
o n th e E ast P ac ific R is e .............. ........................ .......................... .... ..... 6 8

2-6 Variation diagram showing Cl (wt%) versus MgO (wt%) for OSC lavas. ..........69

2-7 Trace element variations versus Zr (ppm) in 9N OSC lavas............................. 70

2-8 Norm alized trace elem ent ratio diagram s................................ ...................... 71

2-9 Radiogenic isotope ratios of 9N OSC lavas............................. ..... ........... 72

2-10 Elemental variation diagrams showing in situ crystallization and melt-
segregation m models ........ ............. ......... ......................... ................ 73

2-11 Partial melting model showing a range of possible parents ............................. 74

2-12 Mantle-normalized diagrams showing results of 1-15% partial melting of an
altered basalt ......... .. ......... ..........................................................75

2-13 Diagram showing the calculated effects of varying wall rock temperature on
incompatible trace element composition (La) during AFC .............. ................ 76

2-14 Mantle-normalized trace element diagram showing results of the best-fit AFC
m od e l .................. ................ ........................................... 77

2-15 Cartoon showing a possible scenario for dacite formation on MOR ................78

3-1 Bathymetric map of the East Pacific Rise showing the location of the 9N
OSC, the Clipperton and Siqueiros transform faults ....................... ................ 96

3-2 H20 (wt%), Cl (wt%), and CO2 (ppm) versus MgO (wt%) for glasses from
the 9N O S C ............. .................... ................ ....... .. .... ............ 97

3-3 H20/Ce and CI/K20 ratios versus MgO (wt%) for glasses from the 9N OSC.....98









nearest gas bubble during ascent. However, the rapid ascent rates in MORB magmas

and quenching of the lavas at the seafloor will trap the volatiles and allow for very little

degassing (e.g., Dixon et al., 1988). Therefore, these model calculations can provide

minimum depths of equilibration of the melts prior to eruption. Using this model, we can

calculate the equilibrium pressure of a vapor-saturated melt (i.e., magma chamber

depth) of a given composition prior to eruption on the seafloor using the VOLATILECALC

program (Newman & Lowenstern, 2002).

H20 and C02 concentrations in the OSC lavas suggest a range of equilibration

pressures with a maximum of ~550 bars. Most of the basaltic lavas have equilibration

pressures that approximately equate to the top of the imaged melt lens (~1.5 km or 450

bars; Kent et al., 2000), suggesting little to no degassing during ascent and eruption on

the seafloor. In contrast, MOR andesites and dacites have completely degassed C02,

and some may have also lost H20 (Figure 3-6). The evolved lavas have equilibrium

pressures approximately equal to that of the seafloor (~250 bars), suggesting slower

ascent rates and higher degrees of degassing prior to eruption.

The vapor saturation calculations support the hypothesis that the depth of

contamination on MOR occurs at the top of the axial magma chamber (le Roux et al.,

2006). Textural observations in ophiolites show melting and assimilation occurs at the

roof of the magma chamber, which may migrate within the ocean crust but often

coincide with the base of the sheeted dikes (Coogan et al., 2002; Gillis & Coogan, 2002;

Gillis, 2008).

The hydrostatic pressure at the seafloor results in extensive C02 degassing from a

C02-saturated basaltic melt, however, quenched lavas supersaturated with C02 are









common in submarine settings (Dixon et al., 1988; Dixon et al., 1995b; Dixon et al.,

1995a; Saal et al., 2002; le Roux et al., 2006). This has been attributed to rapid ascent

rates of basaltic magmas, which do not allow for complete vapor exsolution. In contrast,

slow ascent rates or pooling of isolated magma batches at shallow depths could lead to

magma degassing (Dixon et al., 1988). A key factor that affects the ascent and effusion

rate of magma is viscosity. The andesites and dacites erupted on the seafloor have

higher viscosities than MORB magmas, which may allow for significant bubble

nucleation and growth prior to eruption. Magma degassing during ascent or eruption of

the high-silica dacites is supported by large elongate vesicles observed in hand

samples and low CO2 contents in the glasses.

Source of Assimilant

Despite growing evidence of crustal contamination on MORs, the source of the

assimilant and depth of the process are poorly understood. Glass compositions suggest

that assimilation of brines is responsible for the Cl contamination in some EPR MORB

(le Roux et al., 2006); however, assimilation of partially melted, altered crust may also

result in anomalous Cl enrichments (e.g., Michael & Schilling, 1989). Partial melting or

thermal breakdown of CI-bearing amphibole in altered crust may result in elevated Cl

concentrations in the resulting magma (Michael & Schilling, 1989) and has been

suggested as a possible source of Cl enrichment in Galapagos Spreading Center

andesites and dacites (Perfit et al., 1999) and in dacites from the 9N OSC (Wanless et

al., accepted).

As mentioned above, CI/K20 and H20/K20 ratios provide a means to discriminate

between sources of contamination on MOR because of the variable concentrations of

these elements in possible assimilants (Kent et al, 1999). Potential contaminants on









MOR include altered basaltic crust (CI = 0.1 wt%, H20 = 5 wt%, K20 = 1 wt%), seawater

(CI = 1.935 wt%, H20 = 97.5 wt%, K20 = 0.04 wt%); 15% NaCI brine (CI = 9.9 wt%, H20

= 85 wt%, K20 = 0.25 wt%); and 50% NaCI brine (CI = 30.3 wt%, H20 = 50 wt%, K20 =

0.25 wt%; see Kent et al., 1999 and references therein). Fluids enriched with as much

as 50% NaCI have been observed in melt inclusions in several MOR gabbros (Kelley &

Delaney, 1987). These brines form from high temperature phase separation of

seawater during hydrothermal circulation (e.g., Berndt & Seyfreid, 1990) and may be

trapped along grain boundaries or pore spaces within the altered crust (Michael &

Schilling, 1989).

Bulk mixing of any of these contaminants with OSC basalt cannot produce the

observed compositions of the MOR dacites (Figure 3-4). Instead, most dacites have

lower H20/K20 ratios and higher CI/K20 ratios compared to the possible assimilants.

However, if we assume that the dacites formed from AFC processes, then assimilation

of low-degree partial melts (5-10%) of altered basalt into a fractionating MORB magma

should produce compositions similar to the OSC dacites. This process can explain the

elevated CI/K20 ratios observed in the MOR dacites. The higher H20/K20 ratios

observed in several MOR dacites (Figure 3-4) suggest that the magmas may have also

interacted with small volumes of a saline brine. For instance, mixing of ~0.2 wt% of a

50% NaCI brine can produce H20/K20 concentrations similar to the MOR dacites and

for this small amount of brine, the change in 5180 of the magma would be negligible

(<<0.01).

Oceanic Plagiogranites

Plagiogranites veins are a volumetrically small but ubiquitous component of the

ocean crust and have been observed in ophiolites (e.g. Pedersen & Malpas, 1984), drill









cores from the ocean crust (e.g. Casey, 1997; Dick et al., 2000), and as xenoliths in

Icelandic lavas (Sigurdsson, 1977). There are also many examples of evolved plutonic

rocks from slower spreading centers (e.g., Mid-Atlantic Ridge; Aumento, 1969), which

may suggest AFC or partial melting processes are occurring on much smaller scales,

deeper in the ocean crust. The origin of these veins remains unclear but two main

hypotheses are: 1) partial melting of gabbroic crust (e.g. Koepke et al., 2004; Koepke et

al., 2007) and 2) extreme crystal fractionation of tholeiitic magmas (Beccaluva et al.,

1999, Coleman & Donato, 1979, Niu etal., 2002).

The composition of oceanic plagiogranites varies considerably (Koepke et al.,

2007), making a direct petrologic comparison between the MOR dacites and the silicic

veins difficult. The mantle-like "180 values of zircons collected from plagiogranite veins

in the gabbroic crust at the mid-Atlantic ridge (~5.2 0.2 %o) suggest little to no

seawater contamination in the evolved melt at these depths (Grimes et al., 2010a), but

lower 6180 values in plagiogranite veins from the Oman ophiolite (average of 4.6 0.6

%o) are thought to represent remelting of altered ocean crust (Grimes et al., 2010b).

These studies suggest that high-silica lavas can form in a variety of different MOR

settings and may have a range of different compositions but that melting is an important

process in their petrogenesis.

Conclusions

Variations in volatile concentrations and 5680 in 9N OSC lavas suggest that these

magmas have experienced assimilation during their petrogenesis, with the most

extreme signatures observed in high-silica andesites and dacites and little evidence in

basaltic lavas. H20 concentrations are up to two times higher in dacitic lavas compared

to calculated fractional crystallization trends, whereas Cl has excesses of seven to ten









times predicted values. 6180 values are on average ~1%o lower than ratios expected

from fractional crystallization of ferromagnesian silicates and Fe-Ti oxide phases,

consistent with assimilation of an additional component or components.

The source of the excess H20 and Cl and low 6180 values is partially melted,

hydrothermally altered oceanic crust, but may also include small volumes of saline

brines produced during two-phase separation of high-temperature hydrothermal fluids.

Vapor saturation pressures calculated from H20-C02 data suggest that assimilation

most likely occurs at the top of the melt lens, which at the 9N OSC, corresponds

approximately to the base of the sheeted dikes.

Basaltic lavas were supersaturated with C02 at their eruption depths suggesting

fast ascent rates. In contrast, high-silica lavas were completely degassed C02 and

variably degassed H20 prior to quenching on the seafloor. This suggests slower ascent

rates and/or lower effusion rates for the high-silica lavas, which is consistent with their

higher viscosities and the presence of large elongate vesicles.










Table 3-1. Geochemical data


Cl
0.07
0.01
0.01
0.02
0.01
0.05
0.01
0.19
0.30
0.31


H20
0.39
0.25
0.24
0.26
0.23
0.31
0.24
0.65
1.24
1.56


C02
231
224
256
131
218
219
232
184
0
0


Table 3-1. Geochemical data


H20/
K20
1.82
1.87
1.74
1.72
1.82
1.91
0.95
2.02
2.38
2.59


H20/
Ce
0.019
0.018
0.019
0.017
0.017
0.018
0.018
0.019
0.020
0.024


Sample
264-08
265-43
265-82
265-88
266-51
265-104
264-18
265-106
265-103
265-91
265-125
265-90
265-100
266-54
264-14
265-69
265-63
265-64
265-65
265-66
265-42
265-67
265-70
265-83
265-84
265-85
265-94
265-95
266-53
266-57
264-09


Si02
50.45
50.12
50.77
50.85
50.55
51.25
52.15
53.33
56.50
56.83
55.59
58.08
58.09
59.65
61.75
61.02
64.43
64.04
63.79
62.81
66.92
64.10
66.26
67.46
64.39
65.01
65.22
67.46
64.28
62.47
65.97


TiO2
2.70
1.92
1.95
2.02
1.94
1.90
1.77
1.94
2.01
2.10
1.66
1.91
1.76
1.72
1.30
1.72
1.29
1.28
1.26
1.43
0.94
1.34
0.87
0.76
1.13
1.06
0.97
0.77
1.06
1.30
0.90


A1203
12.86
13.86
13.81
13.88
13.85
13.79
14.04
13.26
12.33
12.31
13.63
12.43
12.64
13.22
13.46
13.62
13.26
13.12
13.25
13.13
13.09
13.33
13.20
13.27
13.17
13.13
13.04
13.10
13.31
13.16
13.19


FeO
14.07
11.57
11.70
11.99
11.59
11.65
10.89
12.35
13.74
14.23
10.73
13.69
12.68
11.25
8.71
9.97
8.22
8.27
8.14
9.05
8.04
8.49
7.17
6.68
8.18
7.99
7.90
6.47
8.06
9.20
7.02


MgO
5.69
6.83
6.93
6.71
7.33
6.90
6.05
5.11
2.74
2.09
4.89
1.74
1.89
2.28
2.47
1.91
1.29
1.60
1.34
1.99
0.86
1.49
0.80
0.67
1.23
1.18
1.13
0.94
1.12
1.59
1.05


CaO
9.51
11.13
11.21
11.06
10.97
10.59
9.94
8.71
6.45
6.24
8.30
5.76
5.51
5.60
5.54
5.38
4.21
4.45
4.21
4.98
3.49
4.41
3.23
2.98
3.92
3.78
3.54
3.01
3.73
4.37
3.47


Na20
3.33
2.81
2.69
2.85
2.86
3.00
3.27
3.67
3.77
3.45
3.64
3.51
3.82
3.91
3.94
3.89
3.71
3.46
3.84
3.86
0.83
3.93
4.08
3.88
3.41
3.67
4.29
4.43
4.16
4.11
4.32


K20
0.21
0.13
0.14
0.15
0.13
0.16
0.26
0.32
0.52
0.60
0.41
0.66
0.63
0.75
0.83
0.80
0.99
0.97
0.97
0.89
1.17
0.95
1.33
1.37
1.19
1.22
1.14
1.21
1.09
0.98
1.20


P205
0.28
0.19
0.19
0.21
0.19
0.22
0.20
0.39
0.65
0.78
0.20
0.74
0.54
0.43
0.22
0.27
0.21
0.20
0.22
0.24
0.21
0.23
0.19
0.16
0.22
0.20
0.23
0.15
0.25
0.29
0.21


1.81 0


0.70 2.35 0
0.67 1.53 0

0.64 1.90 0


0.66
0.51
0.58


1.74 0
2.08 0


0.34 0.99 0

0.42 1.50 0

0.20 1.44 0

0.24 1.73 0


d180 Ce
5.79 20.5
14.2
5.51 13.0
5.71 15.0
13.3
16.9
5.72 13.4
5.77 33.9
5.73 60.4
5.31 66.1
5.92 32.7
73.2
5.56 72.7
5.83 67.9
6.06 57.0
55.8
5.57 68.1
6.19 76.5
5.86 67.5
5.84 84.2
5.87 82.9
5.92 68.0
6.08 88.1
5.94 87.2
5.73 77.8
5.95 82.5
5.90 83.9
6.07 83.6
5.38 82.2
5.65 72.4
5.73 83.9


CI/K20
0.35
0.10
0.07
0.13
0.09
0.29
0.04
0.59
0.58
0.52

0.51

0.56

0.25

0.25

0.26
0.44

0.53
0.49

0.52


0.60
0.52
0.48


1.50 0.014

2.01 0.022

1.80 0.026

1.79 0.023

2.04 0.022


1.76 0.027
1.12 0.018

1.55 0.023


1.60 0.021
2.13 0.029


0.23
0.51












































Figure 3-1. Bathymetric map of the East Pacific Rise showing the location of the 9N
OSC, the Clipperton and Siqueiros transform faults.













96











1.6 2.5
1.4 A 2.0 1-85% Fractional B
1.2 1-85% Fractional Crystallization
Crystallization Crysta




0.28-
r 0.1
0.8 x / A.

0.2 -S*X,,
0.8
07 C basalt D
0.6 250 A a ns SsA *A
0.6 200 andesite
0.s 5- 1-85% Fractional 0 dacite A
0.4 Crystallization 150 -- F.C.
0.3 A A 100-
0.2 A
0.01 1 50
0.0 I.. .... ::: : l'. O a E A
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
MgO (wt %) MgO (wt %)


Figure 3-2. H20 (wt%), Cl (wt%), and C02 (ppm) versus MgO (wt%) for glasses from
the 9N OSC. Black crosses indicate the calculated fractional crystallization
trend using MELTS (Ghiorso and Sack, 1995). All dacitic lavas and several
basaltic andesites lie above the calculated trend, indicating another
processes is involved in their petrogenesis.















Ut


\ Assimilation


A


0.030

0.026

0.022

0.018

0.014

0.010


MgO (wt %)


*A



-"^



Fractional
Crystallization


Assimilation
N


A


MgO (wt %)


Figure 3-3. H20/Ce and CI/K20 ratios versus MgO (wt%) for glasses from the 9N
OSC. Generalized trends for assimilation and fractional crystallization are
shown as dashed lines. In general, the andesites, dacites and most of the
basaltic andesites have higher incompatible element ratios than the basaltic
lavas, suggesting that they cannot result from fractional crystallization alone.
Instead, they are consistent with assimilation of an altered basalt.


4 ---------.-----------
Fractional
Crystallization


* basalt
A basaltic
andesite
* andesite
* dacite


0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0


* basalt
A basaltic
andesite
* andesite
* dacite















068 5% melt of
altered basalt
0.7

06 A basalt
10% melt of A A basaltic andesite
altered basalt 0.2% + andesite
0.5 dacite
-)-Alt Basalt
0-14 50% Brine
v 0.4
S- 15% Brine
Seawater
S5% melt Alt bas
S10% melt alt bas
02
2 MELTS
fractional -
0.1 crystallization -
altered basalt

0.0
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
H2i0K20




Figure 3-4. CI/K20 versus H20/K20 for glasses from the 9N OSC. Lines representing
mixing of 6 possible assimilants with an OSC basalt are shown. Mixing end-
members include 5 and 10% partial melts of an altered basalt, an altered
basalt, a 50% and 15% NaCI brine, and seawater (see Kent et al., 1999 for
references). See text for mixing end-member concentrations. A combination
of partial melting of an altered basalt and <0.2 wt% of a 50% NaCI brine with
a basaltic end-member can explain the formation of high-silica lavas on the
OSC.





PAGE 1

1 GEOLOGY AND PETROGENESIS OF LAVAS FROM AN OVERLAPPING SPREADING CENTER: 9N EAST PACIFIC RISE By V. DORSEY WANLESS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF TH E REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 V. Dorsey Wanless

PAGE 3

3 To my family and friends

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my advisor (Michael Perfit), collaborators (W. Ian Ridley, Emily Klein, S cott White, Paul Wallace, John Valley, and Craig Grimes) and supervisory committee (Paul Mueller, Ray Russo, George Kamenov, and David Richardson) for their mentoring and encouragement throughout this study. Additionally, I would like to thank the Captain, officers and crew of the R/V Atlantis for all their help during cruise AT15-17, the MEDUSA2007 Science party (including S. White, K. Von Damm, D. Fornari, A. Soule, S. Carmichael, K. Sims, A. Fundis, A. Zaino, J. Mason, J. OBrien, C. Waters, F. Mansfiel d, K. Neely, J. Laliberte, E. Goehring, and L. Preston) for their diligence in collecting data and samples for this study. I thank the ROV Jason II shipboard and shorebased operations group for their assistance in collecting these data and HMRG for proce ssing all DSL120 side scan and bathymetry data collected during this cruise. Discussions with S. White and A. Goss are gratefully acknowledged. Thanks to G. Kamenov and the UF Center for Isotope Geosciences for laboratory assistance and to the Department of Geological Sciences staff for all of their help. Finally, I thank my friends and family for all their support over the years. This research was supported by the National Science Foundation (grants OCE -0527075 to MRP and OCE -0526120 to EMK).

PAGE 5

5 TABLE OF CONTENTS ACKNOWLEDGMENTS ...................................................................................................... 4 page LIST OF TABLES ................................................................................................................ 8 LIST OF FIGURES .............................................................................................................. 9 ABSTRACT ........................................................................................................................ 12 C H APT ER 1 INTRODUCTION ........................................................................................................ 14 2 DACITE PETROGENESIS ON MID -OCEAN RIDGES: EVIDENCE FOR OCEANIC CRUSTAL MELTING AND ASSIMILATION ............................................ 17 Abstract ....................................................................................................................... 17 Introduction ................................................................................................................. 18 Geologic and Tectonic Setting ................................................................................... 22 9 N East Pacific Rise Overlapping Spreading Center ..................................... 22 J uan de Fuca Ridge Propagating Ridge Tip and Axial Seamount ..................... 24 Galpagos Spreading Center Extinct OSC or Propagating Ridge Tip? ........... 25 Petrography ................................................................................................................ 26 Geochemical Methods ................................................................................................ 26 Geochemical Results .................................................................................................. 27 Major Element Results ......................................................................................... 27 Trace Element Results ......................................................................................... 28 Isotopic Data ........................................................................................................ 29 Petrogenetic Models For High-Silica Lavas ............................................................... 29 Crystal Fra ctionation ............................................................................................ 30 Rayleigh fractional crystallization .................................................................. 30 Crystal melt segregation model .................................................................... 33 In situ crystallization calculations .................................................................. 34 Partial Melting ( Anatexis) ..................................................................................... 35 Assimilation Fractional Crystallization ................................................................. 37 Discussion ................................................................................................................... 41 Petrogenesis of High Silica Lavas ....................................................................... 41 Geochemical evidenc e of partial melting ...................................................... 41 The need for crystallization, assimilation and altered crust in dacite petrogenesis ............................................................................................... 44 Isotopic signature of assimilation .................................................................. 45 AFC Processes and Tectonic Setting ................................................................. 48 Model for Formation of MOR Dacites .................................................................. 49 Relationship of Melt Lens to Dacites at 9N ........................................................ 50

PAGE 6

6 Effects of Assimilation on Typical MORB Compositions .................................... 51 Conclusions ................................................................................................................ 52 Acknowledgements ..................................................................................................... 53 3 ROLE OF ASSIMILATION IN THE PETROGENESIS OF LAVAS ON MID OCEAN RIDGES INFERRED FROM Cl, H2O, CO2 AND OXYGEN ISOTOPE VARIATIONS .............................................................................................................. 79 Introduction ................................................................................................................. 79 Geologic Setting ......................................................................................................... 81 Analytical Methods and Results ................................................................................. 83 Major and Trace Elements ................................................................................... 83 Volatile Elements ................................................................................................. 84 Oxygen Isotope Analyses and Results ............................................................... 85 Discussion ................................................................................................................... 85 Magma Crystallization versus AssimilationFractionation-Crystallization Processes ......................................................................................................... 85 Evidence for Assimilation from Oxygen Isotopes ............................................... 87 CO2 and H2O Degassing, Magma Ascent Rates and Depth of Assimilant ........ 89 Source of Assimilant ............................................................................................ 91 Oceanic Plagiogranites ........................................................................................ 92 Conclusions ................................................................................................................ 93 4 CRUSTAL DIFFERENTIAION AND SOURCE VARIATIONS AT THE 9N OVERLAPPING SPREADING CENTER; EAST PACIFIC RISE ............................ 102 Introduction ............................................................................................................... 102 Background, Tectonic Setting and Geology of the 9oN OSC .................................. 104 Overlapping Spreading Centers ........................................................................ 104 Tectonic Setting and Previous Studies of 9N OSC ......................................... 104 9N O SC Geology .............................................................................................. 106 837 N EPR Deval ............................................................................................. 108 Geochemical Methods .............................................................................................. 109 Geochemical Results ................................................................................................ 110 Geochemistry of East Limb Lavas ..................................................................... 110 Basalts ......................................................................................................... 111 Basaltic andesites/low -P2O5 andesites ....................................................... 111 High -P2O5 andesites .................................................................................... 112 Dacites ......................................................................................................... 112 Isotopic compositions of East Limb lavas ................................................... 113 West Limb Lavas ................................................................................................ 113 837 EPR Lavas ................................................................................................ 114 Discussion ................................................................................................................. 115 Shallow -level Processes Involved in the Petrogenesis of Ferrobasalts, FeTi Basalts and Basaltic Andesites at the 9N OSC ........................................... 115 Fractional crystallization versus assimilation in the petrogenesis of andesites .................................................................................................. 118

PAGE 7

7 Formation of west limb lavas by fractional crystallization .......................... 119 Where Are the South Tip and West Limb Dacites? .......................................... 120 Composition of the Melt Lens Beneath the East Limb ...................................... 122 E-MORB Distribution at 9N OSC ..................................................................... 123 9N OSC as a Division in Mantle Components ................................................. 125 Conclusions .............................................................................................................. 126 5 CONCLUSIONS ........................................................................................................ 161 LIST OF REFERENCES ................................................................................................. 164 BIOGRAPHICAL SKETCH .............................................................................................. 181

PAGE 8

8 LIST OF TABLES Table page 2 -1 Dacite major and trace element data..................................................................... 55 2 -2 Radiogenic isotopes for 9N OSC lavas ................................................................ 57 2 -3 St arting compositions for geochemical modeling .................................................. 59 2 -4 Partition coefficients for Rayleigh fractional crystallization, partial melting and AFC calculations .................................................................................................... 60 2 -5 AFC modeling parameters ..................................................................................... 61 3 -1 Geochemical data .................................................................................................. 95 4 -1 East limb major element data 9N OSC .............................................................. 128 4 -2 Trace element data 9N OSC .............................................................................. 136 4 -3 West limb major element data 9N OSC ............................................................. 140 4 -4 West limb trace element data 9N OSC .............................................................. 142 4 -5 West limb isotopic data 9N OSC ........................................................................ 143 4 -6 Major and trace element data from 837'N EPR ................................................. 144 4 -7 Radiogenic isotope ratios 837'N EPR ................................................................ 145

PAGE 9

9 LIST OF FIGURES Figure page 2 -1 Bathymetric maps showing the tectonic setting of the MOR dacites ................... 62 2 -2 Comparison of major and trace element compositions in MOR high-silica andesites and dacites ............................................................................................ 6 5 2 -3 Bathymetric map of the 9N OSC showing locations of samples collected during the MEDUSA2007 cruise ............................................................................ 66 2 -4 Photographs of MOR high-silica lavas .................................................................. 67 2 -5 Major element variations versus MgO (wt.%) for dacites from the 9N OSC on the East Pacific Rise ......................................................................................... 68 2 -6 Var iation diagram showing Cl (wt%) versus MgO (wt%) for OSC lavas. ........... 69 2 -7 Trace element variations ver sus Zr (ppm) in 9N OSC lavas ............................... 70 2 -8 Normalized trace element ratio diagrams .............................................................. 71 2 -9 Radiogenic isotope ratios of 9N OSC lavas ......................................................... 72 2 -10 Elemental variation diagrams showing in situ crystallization and melt segregation model s ................................................................................................ 73 2 -11 Partial melting model showing a range of possible parents ................................. 74 2 -12 Mantlenormalized diagrams showing results of 115% partial melting of an altered basalt .......................................................................................................... 75 2 -13 Diagram showing the calculated effects of varying wall rock temperature on incom patible trace element composition (La) during AFC .................................... 76 2 -14 Mantlenormalized trace element diagram showing results of the best -fit AFC model ...................................................................................................................... 77 2 -15 Cartoon showing a possible scenario for dacite formation on MOR .................... 78 3 -1 Bathymetric map of the East Pacific Rise showing the location of the 9N OSC, the Clipperton and Siqu eiros transform faults. ............................................ 96 3 -2 H2O (wt%), Cl (wt%), and CO2 (ppm) versus MgO (wt%) for glasses from the 9N OSC. ........................................................................................................ 97 3 -3 H2O/C e and Cl/K2O ratios versus MgO (wt%) for glasses from the 9 N OSC ..... 98

PAGE 10

10 3 -4 Cl/K2O versus H2O/K2O for glasses from the 9N OSC ....................................... 99 3 -5 18O versus MgO for glasses from the 9N OSC ................................................ 100 3 -6 CO2 H2O vapor saturation diagram .................................................................... 101 4 -1 Bathymetric map of the northern EPR. ................................................................ 146 4 -2 Bathymetric map of the 9N OSC with the location of rock samples ................. 147 4 -3 Bathymetric map of the 9N OSC with t he m elt sills ........................................... 148 4 -4 Side scan sonar mosaic from data collected on the MEDUSA2007 cruise ....... 149 4 -5 FeO versus MgO for glasses collect ed from the east limb of the 9N OSC ...... 150 4 -6 Major element variations versus MgO (wt%) for glasses collected from the east limb of the 9N OSC ..................................................................................... 151 4 -7 P2O5/TiO2 versus MgO (wt%) for east limb glasses ........................................... 152 4 -8 Trace element concentrations versus Zr for glasses .......................................... 153 4 -9 Incompatible trace element ratios versus Zr for glasses erupted at the OSC. 154 4 -10 Radiogenic isotope ratios showing the variation in sources ............................... 155 4 -11 Major element concentrations and ratios versus MgO comparing the east and west limb of the OSC ........................................................................................... 157 4 -12 Trace element concentrations versus Zr comparing east and west limb ........... 159 4 -13 Primitive mantle normalized diagram showing variations in andesites and basaltic andesites erupted at the OSC. ............................................................. 160

PAGE 11

11 LIST OF ABBREV IATIONS AFC assimilation fractional crystallization Cpx clinopyroxene EPR East Pacific Rise GSC Galapagos Spreading Center HREE heavy rare earth elements Ilm Ilmenite JdFR Juan de Fuca Ridge LLD liquid line of descent MOR mid ocean ridge MORB mid ocean ridge basalt N -MORB normal mid ocean ridge basalt ODP Ocean Drilling Program Ol olivine OSC overlapping spreading center Plag plagioclase QFM quartz -fayalite magnetite REE rare earth elements ROV remotely operated vehicle Sp spinel

PAGE 12

12 Abstract of Dissertation P resented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy GEOLOGY AND PETROGENESIS OF LAVAS ON AN OVERLAPPING SPREADING CENTER: 9N EAST PACIFIC RISE By V. Dorsey Wanless August 2010 Chair: Michael Perfit Major: Geology In contrast to relatively homogeneous mid ocean ridge basalt (MOR B ) compositions typically erupted on fast -spreading oceanic ridges a wide range of rock types from basalts to dacites have been recovered at overlapping spreading centers (OSC). This study focuses on the petrogenesis of lavas erupted at the 9N OSC on the Ea st Pacific Rise in order to better understand the complex magmatic plumbing system beneath a ridge discontinuity. La vas that span the entire compositional range observed on the global midocean ridge (MOR) system, including basalts, ferrobasalts, FeTi basalts, basaltic andesites, andesites and dacites have erupted along the eastern, propagating limb of the OSC. M ajor and trace element analyses, radiogenic (Pb, Sr, Nd) and oxygen isotopic ratios volatile content s (Cl, H2O, CO2) and geochemical modeling are used to determine the petrogenesis of MORB and genetically related high -silica magmas. The formation of high-silica d acites on MOR remains a petrologic enigma despite eruption on several different ridges. They are characterize d by elevated U, Th, Zr, and Hf; relatively low Nb and Ta ; and Al2O3 and K2O concentrations that are higher than expected from fractional crystallization. Additionally, high Cl and H2O concentrations

PAGE 13

13 and relatively low 18O values in dacitic glasses require contamination fr om a seawater altered component. Extensive p etrologic modeling of MOR dacites suggests that fractional crystallization of a M OR B parent combined with partial melting and assimilation of altered ocean crust can generate magmas with geochemical signatures consistent with MOR dacites. This suggests that crustal assimilation is a much more important process on ridges than previously thought and may be significant in the generation of evolved MORB in general. Petrologic models indicate that ferrobasalts and FeTi basalts erupting at the OSC can be explained by low -pressure fractional crystallization of a primitive MORB parent; however, both fractional crystallization and magma mixing produce intermediate compositions. Geochemical analyses suggest that there are two distinct populations of andesites erupted at the OSC. Andesites with high-P2O5 are the most evolved MOR compositions produced through fractional crystallization. In contrast, low -P2O5 andesites and basaltic andesites appear to have formed primarily through mixing of ferrobasalt ic and dacitic magmas.

PAGE 14

14 CHAPTER 1 INTRODUCTION Mid ocean ridges (MOR) are comprised of a series of segments that can be subdivided at a variety of scales, ranging from tens of meters to hundreds of kilometers between 1st order discontinuities marked by transform faults (Sempere and Macdonald, 1986; Macdonald et al., 1988). Overlapping spreading cent ers (OSC) are 2nd order discontinuities that form between widely spaced transform faults on fast to intermediate spreading ridges (Macdonald and Fox, 1983; Sempere and Macdonald, 1986, Carbotte and Macdonald, 1992). These offsets provide both a physical and a geochemical segmentation of the ridge, which may result from variations in mantle melting and/or separation of crustal magma re servoirs between segments (e.g. Macdonald et al., 1988). Lavas erupted along fast to intermediate spreading centers, such a s the northern East Pacific Rise (EPR), may produce a range of basaltic lavas (e.g., Batiza and Niu, 1992) but they rarely erupt compositions with MgO concentrations <5 wt%. This relatively limited compositional diversity compared to other tectonic setti ngs is commonly attributed to shallow -level fractional crystallization of primitive magmas within an axial magma chamber buffered by relatively frequent recharge with more primitive melts (Klein, 2005). Additionally, geochemical variations in MORB may resu lt from variable mantle melting parameters and/or mantle sources ( Klein & Langmuir, 1987; Langmuir et al., 1992). In contrast, l avas erupted at ridg e segment ends, such as an OSC, can have highly variable compositions compared to a relatively limited rang e of basaltic compositions erupted from magmatical ly robust segment centers (e.g. Christie and

PAGE 15

15 Sinton, 1981; Langmuir et al., 1986). These rock types can range from typical MORB lavas to ironenriched FeTi basalts, andesites, and high-silica dacites. This geochemical variability is commonly attributed to lower magma supply and cooler crust (cold edge effect) at the end of ridge segments, which causes increased magmatic fractionation prior to eruption (Christie and Si nton, 1981; Sinton et al., 1983; Perfit et al., 1983; Perfit and Chadwick, 1998; Rubin and S inton, 2007) W hile crystal fractionation is undoubtedly a primary process in magma differentiation at MOR, it may not be the only process involved in the petrogenesis of evolved lavas on MOR. Elevated Cl concentrations in many MORB suggest that partial melting and assimilation of seawater altered material may be an important, but often overlooked, process in MOR magmatism (Michael and Schilling, 1989; Michael and Cornell, 1998). Evidence of these processes at MOR is supported by textural observations in ophiolites, which show melting of overlying crustal material at the top of the magma chamber (Coogan et al., 2003). Additionally, experimental results suggest that hydrous partial melting of altered ocean crust can produce high-silica plagiogranite veins, which are a small (<2%) but ubiquitous part of the ocean crust (Koepke et al., 2004; Koepke et al., 2007). Despite the clear evidence of assimilation in these lavas, many models for the magmatic plumbing system at MORs do not include partial melting and assimilation. Dacites have been sampled from several different spreading centers; including, the East Pacific Rise (EPR), the Juan de Fuca Ridge (JdFR), and the Ga lapagos Spreading Center (GSC); however, there is no consensus on how these high-silica lavas form on MOR. To answer these questions, we undertook a 35day research cruise (AT15 -17) in the Spring of 2007 to the 9N OSC, during which we surveyed 200 sq. km.

PAGE 16

16 with DSL 120A for side scan sonar mapped and sampled with the ROV Jason II (~7000 digital photographs, recovery of >280 rock samples, sampling of hydrothermal vent waters and biota); and took photographs with the WHOI TowCam (~10,000 digital photographs, CTD and MAPR data), which constitiute one of the most detailed data sets from an OSC. The compositions of lavas recovered from the OSC exhibit remarkable diversity, ranging from basalt to dacite: 33% of OSC lavas have SiO2 > 52 wt.%, compared to ~3% for ocean ridge basalts worldwide. This study utilizes this data to explore the roles of crystal fractionation, partial melting and assimilation in the petrogenesis of high-silica lavas on MOR. In addition, geochemical data are used to determine which process es are involved in the formation of the range of compositions (basalts, ferrobasalts, FeTi basalts, basaltic andesites, andesites and dacites) erupted at the 9N OSC to better understand the anatomy of a 2nd order MOR discontinuity.

PAGE 17

17 CHAPTER 2 DACITE PETROGENESIS ON MID -OCEAN RIDGES: EVIDENCE FOR OCEANIC CRUSTAL MELTING AND ASSIMILATION Abstract While the majority of eruptions at oceanic spreading centers produce lavas with relatively homogeneous midocean ridge basalt (MORB) compositions, the formation of tholeiitic andesites and dacites at midocean ridges (MOR) is a petrologic enigma. Eruptions of MOR high-silica lavas are typically associated with ridge discontinuities and have produced regionally significant volumes of lava. Andesites and dacites h ave been observed and sampled at several different locations along the global MOR system; including propagating ridge tips at ridge-transform intersections on the Juan de Fuca Ridge and eastern Galpagos spreading center, and at the 9N overlapping spreadi ng center on the East Pacific Rise. Despite the formation of these lavas at various different ridges, MOR dacites show remarkably similar major element trends and incompatible trace element enrichments, suggesting that similar processes are controlling th eir chemistry. Although most geochemical variability in MOR basalts is consistent with low pressure fractional crystallization of various mantle-derived parental melts, our geochemical data from MOR dacitic glasses suggest that contamination from a seawateraltered component is important in their petrogenesis. MOR dacites are characterized by elevated U, Th, Zr, and Hf, low Nb and Ta concentrations relative to the rare earth elements (REE) and Al2O3, K2O, and Cl concentrations that are higher than expected from low -pressure fractional crystallization alone. Petrologic modeling of MOR dacites suggests that partial melting and assimilation are both integral to their petrogenesis. Extreme fractional crystallization of a MORB parent combined with partial melting and assimilation of amphibole-bearing altered crust produces a magma with

PAGE 18

18 geochemical signatures consistent with MOR dacites. This supports the hypothesis that crustal assimilation is an important process in the formation of highly evolved MOR lavas and m ay be significant in the generation of evolved MORB in general. Additionally, these processes are likely to be more common in regions of episodic magma supply and enhanced magma-crust interaction such as at the ends of ridge segments. Introduction Fast t o intermediate oceanic spreading centers typically erupt geochemically diverse basaltic lavas (e.g. Klein, 2005); however, a much more extensive range of lava compositions, including ferrobasalts and FeTi basalts as well as rarer high-silica andesites and dacites have been recovered from several different ridges (Perfit et al., 1983; Langmuir et al., 1986; Natland et al., 1986; Natland & MacDougall, 1986; Regelous et al., 1999; Smith et al., 2001) These great varia tions in compositions are commonly attributed to low magma supply and/or cooler crust at ridge segment ends, or the cold edge effect, which promote greater differentiation of magmas prior to eruption (Christie & Sint on, 1981; Perfit et al., 1983; Sinton et al., 1983; Perfit & Chadwick, 1998; Rubin & Sinton, 2007) The formation of highly evolved, silicic magmas in nonridge settings (e.g. ocean islands, arc volcanoes, and continental interiors) have been attributed to several different processes, including crystal fractionation, partial melting of overlying crust, and/or assimilation of crustal material into an evolving magma chamber. On midocean ridges (MOR), many studies have documented the dominant role crystal fractionation plays in magma differentiation (e.g. Clague & Bunch, 1976; Bryan & Moore, 1977; Byerly, 1980) whereby extensive crystallization of olivine, plagioclase, pyroxene and Fe-

PAGE 19

19 Ti oxides leads to the generation of highly evolved melts enriched in SiO2 and depleted in MgO, FeO and TiO2 (e.g. Juster et al., 1989) However, while crystal fractionation is undoubtedly a primary process involved in the di fferentiation of most MORB magmas, it may not be the only mechanism involved in the formation of high-silica MOR andesites and dacites. Partial melting (or anatexis) of basaltic crustal material may produce evolved compositions, particularly in settings w here magma-rock interactions are likely, such as the top of an axial magma chamber (e.g. Coogan et al., 2003b; Gillis, 2008) This process has been suggest as the origin for high-silica lavas erupted on many ocean islands (Iceland; O'Nions & Gronvold, 1973; Sigurdsson & Sparks, 1981; Galapagos Islands; McBirney, 1993; Socorro Island; Bohrson & Reid, 1997; Bohrson & Reid, 1998) and may explain the formation of high -silica lavas in back arc settings, most recently along the Lau Spreading Center (e.g. Kent et al., 2002) and Manus Basin (Sinton et al., 2003) ; although the presence of a subduction zone makes this tectonic setting much more complicated. Evidence from ophiolites suggests that the top of the axial magma chamber on MOR is a dynamic boundary where magmas may interact with and melt different layers of crustal material; including both gabbros and sheeted dikes (Coogan et al., 2003b) Recent experimental evidence suggests that partial melting of hydrous gabbroic rock similar to that in the lower ocean crust can form silicic compositions (Koepke et al., 2004; Kvassnes & Grove, 2008) and may explain the presence of highly evolved plagiogranite veins in the ocean crust (Koepke et al., 2004; Brophy, 2009) Other studies indicate that low degrees of dehydration partial melting of altered basalt, similar in composition to dikes of the upper ocean crust, can produce dacitic melts

PAGE 20

20 (Beard & Lofgren, 1991) These experimental studies suggest that oceanic crust will begin to melt at temperatures as low as 850 to 900C and <10% melting of the crust will yield dacitic or tonalitic melts (Beard & Lofgren, 1991; Koepke et al., 2004; Kvassnes & Grove, 2008) Kvassnes and Grove (2008) state that mineral pairs (plagioclase olivine and plagioclase augite) similar to oceanic gabbros from the lower crust will melt quickly and easily at temperatures similar to that of primitive MOR magmas (12201330C). All of these studies indicate that high level partial melting o f ocean crust can produce high-silica melts on MOR, but the role that this process may play in the formation of voluminous extrusive silicic lavas on the seafloor has not yet been assessed. The compositional variability observed in arc and continental volc anics is commonly ascribed to the associated processes of assimilation and fractional crystallization (AFC; e.g. Bowen, 1928; De Paolo, 1981) but similar processes may also occur where thickened oceanic crust leads to magma-crust interaction, for instance, within Icelandic volcan oes (e.g. Nicholson et al., 1991) On smaller scales, the combined effects of these processes have been observed in ophiolites, where subaxial intrusive magmas have been in contact with and have partially melted the overlying sheeted dikes (Gillis & Coogan, 2002; Coogan, 2003; Gillis, 2008) AFC processes have also been invoked to explain high Cl concentrations observed in some MORB magmas (Michael & Schilling, 1989; Michael & Cornell, 1998; le Roux et al., 2006) During this process, a magma undergoes crystal fractionation, and the resultant latent heat of crystallization provides the heat needed to partially melt the surrounding wall rock. These melts are then assimilated into, and hom ogenized with, the

PAGE 21

21 fractionating magma reservoir. AFC processes can produce a wide range of rock types depending on the initial composition of the intruding magma, the degree of crystal fractionation, the initial wall rock composition, and the amount of melting and assimilation. High -silica compositions are found throughout the ocean crust and are commonly observed as intrusive or plutonic material. As mentioned above, plagiogranites veins are a ubiquitous component of the ocean crust and have been obser ved in ophiolites (e.g. Pedersen & Malpas, 1984) drill cores from the ocean crust (Casey, 1997; e.g. Dick et al., 2000; Wilson et al., 2006) and as xenoliths in Icelandic lavas (Sigurdsson, 1977) The origin of these veins remains unclear but two main hypotheses are: 1) partial melting of gabbroic crust (e.g. Koepke et al., 2004; Koepke et al., 2007; Nunnery et al., 2008) and 2) extreme crystal fractionation of tholeiitic magmas (Coleman & Donato, 1979; Beccaluva et al., 1999; Niu et al., 2002) There are also many examples of evolved plutonic rocks from slower spreading centers (e.g. Mid -Atlantic Ridge, Aumento, 1969) which may suggest AFC or partial melting processes are occurring on much smaller scales, deeper in the ocean crust. In this study we examine the geochemistry of high-silica lavas from three different MOR, including the East Pacific Rise, Juan de Fuca Ridge, and Galapagos Spreading Center ( Figure 2-1) and show they have remarkably similar major and trace element compositions ( Figure 22), suggesting that similar sources and processes control their petrogenesis. More specifically, we exami ne the roles that crystal fractionation, partial melting, and AFC may have played in the formation of an exceptional suite of high-silica lavas from the 9 N overlapping spreading center (OSC) on the East Pacific Rise, and

PAGE 22

22 evaluate if these results apply generally to the formation of high -silica lavas on other MOR. We focus on the petrogenesis of dacites at the 9 N OSC because it is the most complete and geologically well -constrained data set available but descriptions of the geologic settings of high-silic a lavas in the other environments are important in order to ascertain the role tectono magmatic settings may have on their petrogenesis. Geologic and Tectonic Setting The MOR system is over 70,000 km long (Macdonald et al. 1991) and the crust formed at these ridges is overwhelmingly ba saltic in nature. High-silica lavas, however, have erupted on several fast and intermediate -spreading ridges and are commonly associated with specific tectonic settings; including propagating ridge tips (Christie & Sinton, 1981; Fornari et al. 1983; Perfit & Fornari, 1983) ; OSC (Christie & Sinton, 1981; Perfit et al. 1983; Sinton et al. 1983; Bazin et al. 2001) ; regions of ridge-hotspot interaction (Chadwick et al. 2005; Haase et al. 2005) and at 1030 N on the East Pacific Rise near the ridge-transform intersection (Regelous et al. 1999) Below, we describe the geologic setting of the 9N OSC and the three ot her ridges (East Pacific Rise, Juan de Fuca Ridge, and Galapagos Spreading Center) where highly evolved lavas have erupted, to elucidate the relationship of magmatism to different MOR tectono magmatic environments. 9 N East Pacific Rise Overlapping Sprea ding Center The 9 N OSC ( Figure 2 -1a) is located on the East Pacific Rise between the Clipperton and Siqueiros transform faults. It consists of two north-south trending ridges that overlap by ~27 km and partly enclose a large overlap basin (Sempere & Macdonald, 1986) The limbs are separated by ~8 km (Singh et al. 2006) The eastern

PAGE 23

23 limb is propagating to the south into older crust at a rate of ~42 km Myr1 (Carbotte & Macdonald, 1992) The 9 N OSC has been the focus of several geophysical studies (Detrick et al. 1987; Harding et al. 1993; Kent et al. 1993; Kent et al. 2000; Bazin et al. 2001; Dunn et al. 2001; Tong et al. 2002) ; including the first MCS 3D survey of a MOR (Kent et al. 2000) and a 3D seismic refraction study (Dunn et al. 2001) These studies resulted in t he first 3D image of a subsurface magma chamber along a MOR, which showed a shallow melt lens lies beneath both limbs of the OSC with an anomalously large melt lens in the interlimb region, north of the overlap basin. This suggests that the region currentl y has an unusually high magma supply rate for a ridge segment end (Kent et al. 2000) High -silica andesites and dacites were recovered from the eastern limb during the Medusa2007 cruise (AT1517) using the ROV Jason2 (Wanless et al. 2 008; White et al. 2009) Several high-silica lavas were also recovered from this area during dredging operations in the late 1980s (Langmuir et al. 1986) The siliceous lavas are primarily confined to the northern section of the neovolcanic zone on the eastern, propagating limb, along the eastern edge of the melt lens ( Figure 2 -3). Morphologically, the dacites form large individual bulbous to elongate pillows that can be several meters in diameter (Figure 24). The pillows are highly striated and have a coarse bread crust surface texture. Typically, the pillows are stacked into m ounds, which can be several meters high or constructional domes. Dacit es largely occur in two regions: as a nearly linear pillow mound in the center of the east limb neo volcanic zone and large, elongate pillow

PAGE 24

24 lavas on the flanks of the axial graben ( Figu re 2 3). Their axial and near axial positions, low sediment cover and unaltered nature suggest they are relatively young. Juan de Fuca Ridge Propagating Ridge Tip and Axial Seamount The Cleft segment is the southernmost segment of the Juan de Fuca Ridge (Figure 21b). It terminates at 4427N at a ridge transform intersection, where it intersects and overlaps the Blanco Transform Zone (Embley et al. 1991; Embley & Wilson, 1992; Smith et al. 1994) This interse ction is characterized by a series of curved ridges that overshoot the Blanco Transform Zone onto the older Pacific plate (Stakes et al. 2006) High -silica andesites and dacites comprise two small constructional domes on the Pacific plate, where the axial ridge intersec ts, and is believed to propagate past the Blanco Transform Zone into the older ocean crust (~6.3 ma) that was created at the Gorda Ridge (Embley & Wilson, 1992; Stakes et al. 2006) The domes are ~20 to 30 m high and 200 to 500 m in diameter and were sampled using rock core and the ROV Tiburon during research cruises in 2000 and 2002 (Cotsonika, 2006) High-reso lution bathymetric maps show there are numerous other constructional domes in the region but they have not yet been sampled, although we surmise that they are also composed of high-silica lavas. Rare andesites have also been recovered within the axis and along the bounding faults of the southern Cleft segment (Stakes et al. 2006) Seismic studies of the southern Juan de Fuca Ridge indicate the presence of an axial magma chamber beneath most of the Cleft segment (Canales et al. 2005) However, an axial magma chamber reflector is absent south of 4438N where the high silica lavas were recovered, suggesting the presence of small melt volumes resulting from weak melt supply to the ridge-transform intersection. Zircon thermochronology and

PAGE 25

25 U -series data indicate that the dacites erupted less than 30 ka ago (Schmitt et al. submitted) The Axial segment of the Juan de Fuca Ridge is a second order ridge segment that currently overlies the Cobb hot spot, which has produced a chain of seamounts trending NW away from the ridge axis (Chadwick et al. 2005) The Juan de Fuca Ridge is migrating NW at a rate of 3.1 cm/yr and has been situated above the Cobb hot spot for the last ~0.2 to 0.7 Myr, creating a large onaxis seamount, known as Axial seamount (Karsten & Delaney, 1989) Axial seamount is the largest feature on this segment of the Juan de Fuca Ridge and has a large summit caldera (Embley et al. 1999) underlain by a large seismically imaged axial magma chamber (West et al. 2001) It has two prominent rift zones, extending to the north and south, which create bathymetric highs. Extensive sampling of the m ain edifice shows that it is composed of moderately evolved and slightly enriched MORB (Chadwick et al. 2005) The rift zones have linear ridges that appear to accommodate extensive diking from the main caldera system (Chadwick et al. 2005) Rare high -silica andesites sampled by three rock cores are located east of the northern rift zone and may be associated with dike propagation from the main axial magma chamber into older ridge crust. Galpagos Spreading Center Extinct OSC or Propagating Ridge Tip? High -silica lavas were sampled at the eastern end of the Galpagos Spreading Center at ~85 W. The area was extensively studied though dredging and Alvin exploration in the early 1980s (Fornari et al. 1983; Perfit & Fornari, 1983; Perfit et al. 1983; Embley et al. 1988) The evolved lav as erupted within the axial valley and along

PAGE 26

26 the axis -bounding faults of the Galpagos Spreading Center approximately 20 km east of the ridge -transform intersection with the Inca transform fault ( Figure 21c). Bathymetric data reveal two curved ridges sur rounding a depression within this region, which has been interpreted as an old, small, extinct OSC or deviation from axial linearity (Perfit et al. 1983; Embley et al. 1988) that has been rifted away from the neovolcanic zone. Most of the evolved lavas (~63% SiO2) at the Galpagos Spreading Center were found off -axis along the bounding faults associated with the southern portion of the extinct OSC. Petrography The 9N OSC dacites are glassy and predominantly aphy ric, with sparse microphenocrysts and very rare, small phenocrysts of plagioclase and clinopyroxene. The small phenocrysts of plagioclase commonly have resorbed edges and sieve textures. Several of the samples contain small clots of basaltic xenoliths comprised of subophitic plagioclase and clinopyroxene surrounded by dacitic or andesitic glass. Geochemical Methods Glass chips from the outer rims of 18 dacites collected at the 9 N OSC were analyzed on a JEOL 8900 Electron Microprobe for major element con centrations at the USGS in Denver, Colorado (Table 2 -1). Eight to ten individual points were analyzed per sample. USGS mineral standards were used to calibrate the microprobe and secondary normalizations were done to account for instrument drift using th e JdF -D2 glass standard (Reynolds, 1995) University of Florida inhouse standard ALV 23929 from the East Pacific Rise (Smith et al. 2001) and USGS standard dacite glass GSC (for more detail on methods see (Smith et al. 2001) ). The probe diameter during glass analyses was

PAGE 27

27 were used. Highprecision chlorine and potassium concentrations were also determined by microprobe on seven of the dacites using 200-second peak/100 second background counting times. Small glass fragments (1050 mm) were handpicked, avoiding microphenocrysts and alteration, cleaned, and dissolved for trace element and isotope analyses following methods described in (Goss et al. 2010) Fourteen dacites from the 9 N OSC were analyzed at medium resolution for trace element concentrations on a high precision Element2 Inductively Coupled Plasma Mass Spectrometer (ICP -MS) at the University of Florida (Table 2 1). Radiogenic isotope ratios (Pb, Sr, and Nd) were determined for 10 dacites using a NuPlasma multi -collector ICP MS at the Univers ity of Florida Center for Isotope Geoscience (Table 2 2). For a detailed description of sample preparation, dissolution procedures, standards, and errors, see (Kamenov et al. 2007; Goss et al. 2010; Goss et al. i n prep) External calibration was done to quantify results using a combination of inhouse basalts (ENDV Endeavour and ALV 2392 9) and USGS (AGV-1, BIR 1, BHVO -1, BCR 2 and STM -1) rock standards (Kamenov et al. 2007; Goss et al. 2010 ). Geochemical Results Major Element Results Major element compositions of the 9N OSC high-s ilica andesites and dacites are presented (along with the trace element abundances) in Table 2 1. The major element geochemistry of the 9 N OSC lavas is sim ilar to Juan de Fuca Ridge and Galpagos Spreading Center lavas ( Figure 2 2). Here, only data from the 9N OSC is discussed in detail but it is important to note that the major and trace element contents and elemental

PAGE 28

28 trends in high-silica lavas from all three ridges are similar. New analyses of some of the high-silica samples previously analyzed and discussed by Perfit et al., (1983; 1999) and representative samples from the Juan de Fuca Ridge ar e presented in Supplementary Data. All high -silica samples from the 9 N OSC appear unweathered with minimal amounts of Fe Mn oxide coating and are essentially aphyric. 9N OSC tholeiitic andesites through dacite samples exhibit increasing SiO2 with decre asing FeO, TiO2, and MgO ( Figures 2 -2 and 2 5) with the most differentiated dacites having ~67 wt% SiO2 and <1 wt% MgO. Al2O3 concentrations in the dacites (12.9 to 13.3 wt%), however, do not show a large decrease compared to the OSC basalts ( Figure 2 5). The dacites have high incompatible major element concentrations (K2O> 0.90 wt % and Na2O > 3.4 wt %; Figure 2 5) but low P2O5 concentrations (< 0.26 wt%; Figure 2-5) compared to basalts. Chlorine concentrations in the dacites range f rom 0.24 to 0.70 wt% compared to < 0.01 to 0.04 wt% in the OSC basalts ( Figure 2 -6). Trace Element Results 9 N OSC dacites are enriched in incompatible trace elements compared to 9 N OSC basalts ( Figure 2-7), the latter having compositions typical of normal, incompatible tra ce element depleted midocean ridge basalts (N -MORB) from the northern East Pacific Rise. For example, Zr and Hf concentrations in the dacites range from 622 to 1050 ppm and 18 to 25 ppm, respectively (Table 2 1). The dacites also contain high concentrati ons of Rb, Ba, U and Th but relatively low Sr and Eu contents. Compared to East Pacific Rise N MORB the dacites have relatively flat REE patterns ( Figure 2-2). On mantlenormalized diagrams the dacites have positive Zr, Hf, U, and Th anomalies, and negativ e Nb and Ta anomalies ( Figure 2-2). Consequently, the dacites also have

PAGE 29

29 slightly lower Nb/La and higher Zr/Dy ratios compared to OSC basalts ( Figure 28) and Ce/Yb and Nd/Y ratios increase from basalt to dacites ( Figure 2 -8). Compatible trace elements are low in the dacites, with Ni concentrations ranging from 9.8 to 4.9. ppm (Figure 2 7) and Cr concentrations from 13 to 1 ppm. Most incompatible trace elements have negative correlations with MgO; however, compatible elements (i.e. Ni and Cr) and Nb/La are positively correlated. U/Nb, Nd/Y and Ce/Yb ratios are negatively correlated with MgO in the dacites. Isotopic Data The 9 N OSC dacites have very limited ranges of Pb, Sr and Nd isotopic compositions, which lie within the general field of East Pacific R ise MORB lavas (Table 2 -2; Figure 2 9). 87Sr/86Sr ratios range from 0.70246 to 0.70258, with an average of 0.70250. These values are well within the range of N -MORB East Pacific Rise lavas from 9 -10 N (Sims et al. 2002; Sims et al. 2003; Goss et al. 2010; Goss et al. in prep) and similar to N -MORB lavas from 9 N OSC. 143Nd/144Nd ratios are also similar to East Pacific Rise N MORB and range from 0.513140 ( Nd = 9.8) to 0.513196 ( Nd = 10.9). Pb isotopes rati os for the 9 N OSC dacites are indistinguishable from 9 N OSC basalts and other East Pacific Rise lavas, having averages of 206Pb/204Pb = 18.268, 207Pb/204Pb = 15.473, and 208Pb/204Pb = 37.679. Pb isotopes for 9 N dacites form a tight cluster in the cent er of the field defined for other 9 N lavas ( Figure 2 9). Petrogenetic Models For High Silica Lavas We now examine the results of various models of fractional crystallization, partial melting and assimilation and compare the results with the geochemical d ata described above to evaluate their relevance to the formation of MOR dacites. Specifically, we

PAGE 30

30 focus on physically reasonable models that are consistent with the highly differentiated major element concentrations, high concentrations of incompatible el ements, distinct trace element patterns, and N MORB like isotopic signatures. The models must also be able to explain relatively high U, Th, Zr and Hf, and low Nb and Ta as well as the flat REE patterns. Additionally, markedly high Cl, K (and high Cl/K), Al2O3, and low P2O5 must be accounted for in successful petrogenetic schemes. Crystal Fractionation Several petrologic models, including MELTS thermodynamic modeling (Ghiorso & Sack, 1995) Rayleigh crystal fractionation, crystal melt segregation, and in -situ c rystallization are investigated here to determine if various processes of crystal fractionation can account for major and trace element compositions of MOR dacites. Rayleigh fractional c rystallization The program MELTS (Ghiorso & Sack, 1995) provides a useful fram ework to evaluate if major element compositions of MOR dacites can be produced by crystal fractionation. Petrologic modeling was also carried out using the program PETROLOG (Danyushevsky, 2001) with similar results for high -silica lavas although the programs generate somewhat different results for intermediate compositions. Several 9N N MORBs were used as starting parental melt compositions (Table 2 3) to determine if a moderately evolved magma could partially crystallize to produce a dacite. These included a slightly evolved MORB (265 -113), a ferrobasalt (26543), and a FeTi basalt (264-08). Pressures for each MELTS run were set at 1 kbar to simulate an approximate minimum depth of crystallization in the shallow oceanic crust, the ox ygen fugacity was set at the quartz -fayalitemagnetite, and the H2O concentrations varied from 0.2 to 0.35 wt% depending on the parent melt composition. Liquid lines of descent were also

PAGE 31

31 calculated for higher pressures (up to 5 kbar) to simulate depths of crystallization within the nascent layer 3 and the shallow mantle. However, the liquid lines of descent converge on similar end member compositions at low MgO and high SiO2. Liquids with MgO and SiO2 compositions similar, though not identical to 9 N OSC dacites can be produced by ~75 -85 % crystal fractionation of a ferrobasaltic parent. Results predict a crystallization sequence of Ol, followed by Ol + Plag, Ol + Plag + Cpx, Plag + Cpx + Sp (titanomagnetite), and in some models, late stage crystallizati on of apatite. No pigeonite or orthopyroxene crystallization is predicted, in contrast to experimental results (Juster et al. 1989) but pigeonite is observed in some MOR andesites and dacites. The calculations su ggest temperatures of <980C are reached when residual liquids attain compositions similar to dacites. The models are in agreement with anhydrous experimental results that indicate residual dacitic liquids form after ~87% at temperatures of ~1050C (Juster et al. 1989) For several major elements (TiO2, FeO, and SiO2), compositions similar to 9 N OSC dacites are obtained through crystal fractionation of a MORB magma; the total amount of crystallization varies sligh tly depending on whether the starting composition was a moderately evolved basalt, ferrobasalt, or FeTi basalt ( Figure 2-5; maximum of ~85 % crystallization). In contrast, calculated abundances of K2O, P2O5, Al2O3, and Cl do not match the dacite endmember composition using any of the parents; modeled residual liquids have higher P2O5 by factors of 5 -10, lower K2O by factors of 1.5 2.5, lower Al2O3 by factors of 1.4 -1.5 and lower Cl by factors of 10 to 12 ( Figure 25 and 2 6). Although MELTS does not predi ct apatite saturation in andesitic liquids the decreasing P2O5 contents in some andesites and very low values in dacites strongly

PAGE 32

32 suggest apatite crystallization. Juster et al., (1989) calculated apatite saturation would occur at approximately 0.7 wt% P2O5 in Galpagos Spreading Center andesites. Additional tests of Rayleigh fractionation include modeling the behavior of trace elements using as input parameters the degree of crystallization and mineral modal proportions determined from the MELTS modeling. Basaltic partition coefficients are used in the trace element modeling up to ~ 57% SiO2, and andesitic partition coefficients are used for > 57% SiO2 (Table 2 4). The Rayleigh fractionation equation (Cl/Co=F^(D 1)) is used to simulate a continuously evolving magma chamber in which phenocrysts are immediately separated from the liquid. The starting composition was a ferrobasalt from the 9N OSC (265-43). As shown in Figure 27, the trace element concentrations observed in the dacites cannot be reproduced using Rayleigh fractionation ( Figure 2-7) with the constraints imposed by major element variations. This model reproduces some dacite incompatible element compositions but does not reproduce the observed enrichments in most of the incompatible elements. A lthough the most incompatible elements (Rb, Ba, Th, U) show the greatest difference between the observed and calculated compositions, even less incompatible elements (Nb, Zr, Y, Hf) require > 90% crystal fractionation. For instance, maximum calculated Zr and Nb concentrations are 705 ppm and 13 ppm, respectively, compared to an average of 870 and 15 ppm in the dacites. In addition, the UN/NbN is predicted by modeling to be <1, whereas the measured dacite values are >1 and the modeling does not reproduce th e high ZrN/DyN and CeN/YbN and low NbN/LaN ratios in the 9 N OSC lavas ( Figure 2-8). The middle to heavy REE concentrations (i.e., Nd, Sm, Eu, Dy, Yb, and Lu) can only be generated by > 90% crystal fractionation. Regardless,

PAGE 33

33 such extreme degrees of crystal lization are inconsistent with major element model calculations. Summarily, the calculated liquid lines of descent do not provide a good fit to the observed major and minor compositions of the high-silica lavas, and trace element models parameterized from the MELTS calculations do not reproduce measured trace element abundances or trace element ratios. Thus, we conclude that extensive low pressure crystal fractionation is unlikely to be the sole mechanism to explain the genesis of the high -silica lavas at the 9oN OSC. Crystal melt s egregati on m odel Bachmann & Bergantz, (2004) suggest that intermediate liquids (andesites and dacites) will separate from crystals (via filter pressing) when a magma has undergone >40 -50 volume percent crystallization. The segregated melt, which is more evolved than the original melt, will once again crystallize until it reaches 40-50% phenocrysts, when again the new evolved melt will separate from the phenocrysts. While this segregation model applies strictly to systems of intermediate compositions (Bachmann & Bergantz, 2004) here we evaluate whether a basaltic magma can evolve geochemically, through a series of segregation events, to form compositions similar to the MOR dacites. To simulate these conditions a 9 N OSC ferrobasalt was allowed to undergo equilibrium cryst allization to andesitic compositions, using MELTS thermodynamic calculations (Ghiorso & Sack, 1995) and starting conditions described for Rayleigh fractionation models. At this composition, the liquid separates from the phenocrysts (Bachmann and Bergantz, 2004), creating a new parent melt. This parent composition

PAGE 34

34 becomes the new starting concentration (an andesite) for the next run, which subsequently crystallizes 50% by volume. This process was repeated (3 times) until MgO and SiO2 concentrations similar to the 9 N OSC dacites were obtained ( Figure 2 10). This occurred after three segregation events or 87.5 wt% crystallization, however, as was noted during the MELTS calculations, several dacitic major and minor element concentrations (i.e. Al2O3, K2O and P2O5) could not be produced. In addition, the calculated incompatible trace element abundances produced using this process were also lower than those observed in the MOR dacites ( Figure 2-10). In situ crystallization c alculations A different approach to magma c rystallization is in situ crystallization, where phenocrysts do not separate from interstitial melt until a small remaining volume of liquid is pressed from the crystallizing mush and mixed with the main body of melt (e.g. Langmuir, 1989; Reynolds & Langmuir, 1997; Pollock et al. 2005) This process assumes that crystallization occurs along a temperature gradient within a solidification front (or boundary layer) and interstitial melt evolves independently from the main melt body. The mixing of interstitial melt back into the main magma body, gradually causes bulk increases in incompatible element abundances and changes in trace element ratios (Langmuir, 1989) This process will cause an increase in highly incompatible elements compared to Rayleigh crystal fractionation, because these elements are continually returned to the residual magma body (Langmuir, 1989) I n situ crystallization was evaluated following Reynolds and Langmuir (1997), with a starting composition of a 9 N OSC ferrobasalt (26543) using the same partition coefficients as for Rayleigh fractional crystallization. Crystallizing phases include Ol,

PAGE 35

35 Plag, Cpx, and eventually, Fe oxides. In the modeled system, the boundary layer is always 5% of the liquid magma chamber volume and the bound ary layer crystallizes until 35% interstitial liquid remains (Reynolds & Langmuir, 1997) All of the residual liquid mixes back into the magma chamber during each iteration of boundary layer crystallization. As the magma chamber continues to crystallize by this mechanism, the boundary lay er moves inward leaving restite crystals behind and the volume of the magma body decreases. Theoretically, this process will continue to modify the liquid magma chamber composition until an infinitesimally small amount of melt is left. After 85% in situ crystallization ( Figure 2-10) calculated major and incompatible trace element concentrations do not match those measured in the MOR dacites. This processes can only account for Zr concentrations in the melt that are less than 3 times the original concentr ation, reaching values of 320 ppm. Even > 95% in situ crystallization cannot reproduce the enriched trace element signatures of the MOR dacites. Although in situ crystallization does enrich the residual melt in incompatible elements compared to Rayleigh crystal fractionation, the failure to reproduce the observed enrichments in MOR dacites suggests that the latter cannot result from this process either. Partial Melting ( A natexis) Experimental studies (Beard & Lofgren, 1991; Koepke et al. 2004) suggest that <10% par tial melting of altered oceanic crust will produce melt with major element concentrations consistent with MOR dacite compositions. Of particular note are the high SiO2, Al2O3 and K2O and low FeO, TiO2, and P2O5 concentrations produced by partial melting of amphibolite facies or greenschist facies minerals because these are also the chemical characteristics of MOR dacites. Basaltic rocks are known to undergo partial or

PAGE 36

36 complete alteration and recrystallization onaxis due to pervasive high temperature hydro thermal alteration (Alt et al. 1986; Gillis & Roberts, 1999) Such alter ed rocks are an attractive starting composition for anatexis because their solidus temperatures are much lower than fresh MORB. Altered oceanic crust can have a wide range of trace element concentrations, depending on the degree of alteration (Alt et al ., 1986). To better evaluate the composition of wall rock involved in the formation of dacites on the 9N OSC, we use the batch melting equation (Cl=Co/(Dbulk [1 F] + F) and andesitic partition coefficients (Table 2 4) to solve for a range of possible p arental wall rock compositions (Co) and then compare these results to compositions of fresh and altered MORB. Trace element patterns generated from the calculations suggest altered basalt provides a better fit than fresh MORB as a source (wall rock) compos ition for the 9N OSC dacites ( Figure 211). Consequently, we model partial melting of an altered MOR basalt (Nakamura et al. 2007) to determine if this process could produce the geochemical characteristics observed in the 9 N OSC lavas ( Figure 211). Altered basaltic wall rock is melted using two d ifferent modal mineralogies, including one with amphibole (Haase et al. 2005) and one without (Koepke et al. 2004) Calculated trace element concentrations resulting from 1 to 15% partial melting of ocean crust are shown in Figures 2 -7, 2 -8, and 2 12. Partial melting in the absence of amphibole (19% Ol, 30% Cpx, 50% Plag, 1% Ilm) can reproduce some, but not all, of the trace element enrichments observed in the 9 N OSC dacites ( Figure 2 12a). In p articular, the HREE in the 9 N OSC dacites are higher than the calculated abundances. The concentrations derived from partial melting of altered crust with

PAGE 37

37 amphibole (20% Cpx, 25% Opx, 49% Plag, 5% Amph, 1% Fe -Oxide; based on modal proportions from (Haase et al. 2005) ) are much closer to the conce ntrations of incompatible elements in the 9 N OSC dacites, suggesting that amphibole is an important component in the melting source rock ( Figure 2-12b). The best -fit model relies on <10% partial melting of amphibole bearing altered oceanic crust to prod uce incompatible trace element compositions comparable to those in 9 N OSC dacites. In particular, the important characteristics of this model are melts with positive Zr and Hf anomalies, negative Nb and Ta anomalies on mantle normalized diagrams ( Figure 2 -12), relatively high U and Th concentrations, and high UN/NbN and CeN/YbN ratios ( Figure 2 8). Melting of altered basalt can also produce elevated Cl concentrations similar to those observed in the MOR dacites ( Figure 26). Although Cl partition coeff icients are poorly constrained, we use estimates to model Cl partitioning during melting (Gillis et al., 2003). Using the median Cl concentration from altered basalts in ODP hole 504B (350 ppm) as a starting composition and modal proportions described abov e, we calculate a range of possible Cl enrichment for 115% melting to be from 0.2 wt% to > 1.0 wt%. This spans the range of Cl concentrations observed in MOR dacites ( Figure 26). Assimilation Fractional Crystallization The Energy Constrained Assimila tion Fractional Crystallization (EC -AFC) formulation of Bohrson & Spera (2001 ) is used to assess the role that these associated processes play in the formation of MOR dacites. The amount of crystallization required to produce enough heat to melt the surrounding crust is calculated and in turn, this

PAGE 38

38 produces a specific mass of mel t of a specific composition. Several physical parameters are required as inputs to EC -AFC calculations (Table 2 -5). These include the liquidus of the magma (~1200C based on results of MELTS modeling of OSC lavas), the temperature and solidus of the wall r ock, and the temperature of equilibrium between the wall rock and the magma. The initial magma composition is assumed to be N MORB and the assimilant is amphibole bearing altered ocean crust with the modal composition described above. The wall rock may span a range of temperatures (800C to 40 C) depending on the age of the ocean crust (Maclennan, 2008) Higher initial wall rock temperatures allow melting to begin earlier in the evolution of the magma r eservoir because less additional heat is required to raise the wall rock above its solidus temperature, i.e., smaller amounts of crystallization are required to initiate melting and assimilation. However, this lowers the overall amount of incompatible tra ce element enrichment in the resulting magma because the initial magma is less chemically evolved during assimilation and larger masses of anatectic melt may be produced ( Figure 2 13). For instance, crust with an initial temperature of 800C will begin melting after 50 to 55% crystallization and the resultant magma has a maximum of ~25 ppm La ( Figure 2-13). Antithetically, lowering the wall rock temperature decreases the total mass of wall rock assimilated while increasing the amount of crystallization needed to initiate melting. Consequently, this causes an increase in the overall incompatible element concentration possible in the melt. Thus, crust with an initial temperature of 50C requires >85% crystallization to begin melting, but results in concentr ations of ~35 ppm

PAGE 39

39 La in the magma ( Figure 2 13). Based on these competing processes, the best -fit wall rock temperature to generate the 9 N OSC dacites is between 650 and 720 C. The local solidus, as described by Bohrson and Spera (2001), is the solidus o f the assimilant, in this case, amphibolitized basalt. Several experimental studies have examined dehydration melting of altered oceanic crust and amphibolites (Hacker, 1990; Rapp et al. 1991; Wolf & Wyllie, 1994; Johannes & Koepke, 2001) but few studies were performed under conditions comparable to those expected at MOR (Beard & Lofgren, 1991) These experiments determined that the solidus temperatures of altered basalts are between 850oC and 900C (Beard & Lofgren, 1991) Gillis & Coogan (2002) discuss the effects of melting altered crust at the roof of an axial magma chamber and suggest a solidus temperature of 875C. Ti -in zircon thermometry (90534C at TiO2 activity of 0.320.02 estimated from coexisting Fe-Ti oxides) from phenocrysts in the Juan de Fuca Ridge dacites is broadly consistent with zircon saturation thermometry (average of 824 15C)) and Fe-T i oxide temperatures (~830C Schmitt et al. in prep) Based on these combined results, 875C was used as the local solidus temperature for AFC calculations. The final thermal input parameter is the temperature of equilibrium, which is defined as the final equilibrium temperature of the magma and wall rock. Generally, this temperature should correspond to the temperature of the erupted lava. Although the temperature of the erupted 9 N OSC dacites is uncertain, the t emperatures of other dacitic magmas have been estimated. The temperature of crystallization of Galapagos Spreading Center andesitic magma was calculate to be as low as ~910 to 940C based on coexisting titanomagnetite and ilmenite grains (Perfit et al. 1983) and experimental

PAGE 40

40 partial melts resulting from melting of oceanic gabbros showed dacitic melts at tem peratures of approximately 900C (Koepke et al. 2004) Consequently, we use 900C as the input equilibrium temperature. The partition coefficients are th e same as those used for partial melting and fractional crystallization models. Basaltic partition coefficients are used for the fractionating magma, while andesitic bulk Kd values were used for the assimilant in the absence of a comprehensive dataset of dacite Kd values. Results of EC -AFC calculations suggest that combinations of 73 to 85% crystal fractionation of a basaltic magma and assimilation of 5 to 20% by mass of partially melted wall rock produces melts that have trace element compositions consist ent with the 9 N OSC dacites. In the best -fit model, melting and assimilation begins after 68% crystallization and a further 517% crystallization occurs as the wall rock melt is assimilated. Consequently, EC -AFC trace element calculations suggest that many of the incompatible trace element concentrations and ratios observed in the 9 N OSC dacites can be explained through this combination of processes ( Figures 2 7, 2 8, 2 -14). Of particular importance are negative Nb and Ta anomalies (relative to La), an increase in Zr and Hf concentrations (relative to HREE), relatively flat mantle-normalized HREE patterns, and ratios of light to heavy REE and middle to heavy REE that are similar to those observed in MOR dacites ( Figure 2-14). For instance, Zr concentrations in the AFC models are 852 ppm and Nb concentrations are 16 ppm compared to an average 9N OSC dacite concentrations of 870 and 15 respectively. Although the overall fit of the model data to the observed data is encouraging, the model values for Ba, Th, U and Hf are slightly under enriched ( Figure 2 14). However, it should be pointed out that

PAGE 41

41 some of the input parameters to these calculations, such as the actual degree of alteration (and hence its composition) of the crustal assimilant, and the temper ature of the surrounding wall rock, are not well constrained and are very likely not constants. Discussion Petrogenesis of High Silica Lavas Extreme crystal fractionation, partial melting of crustal material, and/or AFC processes have been proposed as expl anations for the formation of highly silicic compositions in continental interiors, arc and ocean island settings, but only a few studies have focused on the formation of high-silica lavas at MOR (Byerly & Melson, 19 76; Perfit et al. 1983; Juster et al. 1989; Haase et al. 2005) The petrogenetic calculations presented above demonstrate that crystal fractionation alone is not a viable mechanism for the formation of high -silica MOR lavas, despite using a range of st arting compositions and several different end-member models ( Figures 2 5, 2 -7, 2 -8). Instead, results emphasize the importance of partial melting and assimilation of altered material in the formation dacites on MOR. Geochemical evidence of partial m elting Will crustal anatexis create geochemical signatures similar to those observed in MOR dacites? Based on elemental systematics (e.g. Cl, U/Nb; Figures 2 6 and 2 7) and partial melting calculations presented, it appears that partial melts of altered oceanic crust may be involved in the generation of the MOR dacites. Partial melts of hydrothermally altered crust produces distinct signatures compared to those of unaltered oceanic crust, due to changes in mineralogy and bulk composition during hydrothermal circ ulation. Hydrothermal circulation in layer 2B or the top of layer 3 may cause alteration to greenschist or amphibolite assemblages, where Ca-rich plagioclase

PAGE 42

42 is replaced by sodic plagioclase and pyroxenes form rims or overgrowths of amphibole (Alt et al. 1986; Coogan et al. 2003b) The degree to which this occurs depends on temperature, water/rock ratios and fluid chemistry. To explain the geochemical signatures in the MOR dacites, our melting assemblage must incl ude amphibole. Melting of amphibole-bearing assemblages, a common component in altered layer 2B (Alt et al. 1986; Coogan, 2003; Coogan et al. 2003b) can explain the anomalously high Al2O3 concentrations observed in the oceanic dacites ( Figure 25). Dehydration partial melting experiments, where water exists only as hydrous phases within the rock, provide a better fit to MOR dacite compositions than hydrous partial melting results (Koepke et al. 2004) Dehydration melting experiments produce a range of Al2O3 concentrations (Beard & Lofgren, 1991) that are similar to or higher than MOR dacites (Figure 25). Comparatively high Na2O concentrations in the dacites may result from the melting of albitic plagioclase. Elevated Na2O concentrations are not observed in all experimental results of Beard and Lofgren (1991) but appear to be a function of the degree of albitization of the starting material. Variable P2O5 concentrations are also observ ed in the experimental melts suggesting that P2O5 contents are very low in the source rock or that it is a residual phase in the melting residue. Similar conclusions can be applied to the MOR dacites, which have low phosphorous contents ( Figure 25) ; however, this may also be a function of apatite crystallization during AFC processes (see next section) Additionally, low FeO and TiO2 suggest that Fe oxides are not a primary melting component in the source rock. This is consistent with our proposed source rock comprised of olivine, plagioclase, cpx, amphibole, Fe oxides.

PAGE 43

43 High Cl concentrations in MOR dacites also support the role of partial melting of rocks altered by seawater -derived fluids (Michael & Schilling, 1989; Michael & Cornell, 1998; Coogan et al. 2003b; Gillis et al. 2003) Although Cl behaves incompatibly during crystallization many MOR lavas show over enrichments compared to other elements with similar compatibilities. For example, after ~85 % frac tional crystallization of a MORB parent (with 0.01 wt% Cl), there is less than a ten-fold enrichment in Cl, resulting in concentrations of ~0.07 wt%, compared to an order of magnitude more (~0.7 wt% Cl) in the dacites. Analyses of altered basalt from sheet ed dikes in drill holes show that Cl concentrations span a range from 49 -650 ppm (Sparks, 1995) Partial melting (115%) of an amphibole-bearing wall rock (with 350 ppm Cl) results in anatectic melts with 0.9 to 0.3 wt% Cl covering the range observed in dacites ( Figure 2-6). Hydrothermal alteration and metamorphism are known to cause increased concentration of some trace elemen ts, including U, Th, Rb, and Ba, as well as Cl (e.g. Alt & Teagle, 2003) Observed positive anomalies of some highly incompatible elements (e.g. U and Th) relative to other incompatible elements with similar distribution coefficients are consistent with partial melting of hydrothermally altered ocean crust (Figure 212). Partial melting may also explain some of the anomalies in the high field strength elements whose concentrations are not affected by alteration or metamorphism but can be fractionated due to mineralogic effects. For example, the relatively low abundances of Nb and Ta (moderately compatible during melting) and the relatively high Zr (highly incompatible) concentrations in the high-silica lavas are a consequence of partial melting of altered crustal material ( Figure 2-12). Additionally, melting and AFC

PAGE 44

44 models above point to the importance of amphibole in the melting assemblage to explain the HREE (compare Figure 2-12a and 2 12b). The need for crystallization, assimilation and altered c rust in dacite p etrogenesis We propose that the partial melting and assimilation of oceanic crust plays a significant role in the formation high-silica MOR lavas; however, the we stress that most of the heat required to melt the wall rock comes from extensive fractional crystal lization. Coogan et al., (2003b) show that the latent heat of crystallization from the formation of 4 km thick gabbro sequence provides enough energy to heat ~1.3 km of overlying crust from 450 to 1150C, which promotes partial melting. It is the assimilation of these partial melts into a fractionally crystallizing magma reservoir that produces the highly evolved melts with enriched incompatible trace element signatures (e.g. De Paolo, 1981; Bdard et al. 2000) M ajor element compositions of MOR dacites often lie between experimental partial melts of altered basalt (Beard & Lofgren, 1991) and liquids produced by moderate to large extents of crystal fractionation ( Figure 2 5). The major element compositions of magmas produced by AFC may therefore, lie between partial melts of altered ocean ic crust and fractionated basaltic magmas. This is particularly apparent in Al2O3 and may explain why very few lavas at the 9N OSC (including ferrobasalts and basaltic andesites) lie on the calculated liquid lines of descent ( Figure 25). Results from EC -AFC calculations (Bohrson & Spera, 2001) confirm that assimilation of anatectic melts into a residual fractionated magma can explain a wide range of trace element concentrations in MOR dacites ( Figure 27, 2 8, 2 -14). The best -fit model for 9 N OSC dacite compositions requires significant crystal fractionation

PAGE 45

45 (7385 wt%) of a ferrobasalt parental magma in combination with 10 to 25% (by mass) anatectic melt, which provides the additional incompatible element enrichments observed. Our models also indicate that in order to explain the relative enrichments in Rb, Ba, Th and U concentrations presen t in the MOR dacites assimilation of low degree partial melts of hydrothermally altered oceanic basalt are required. Relatively low Nb and Ta concentrations are a consequence of both the removal of phenocrysts during late -stage crystal fractionation and r esidual iron -titanium oxides in the partially melted wall rock. Elevated Zr and Hf concentrations result from little to no zircon crystallization in the fractionating magma and/or no residual zircon in the melting assemblage. It is important to note that the extreme Cl enrichments in MOR dacites require a seawater component that can be derived by AFC process and that Cl over enrichments observed in many MORB have been explained either by small amounts of assimilation of either hydrothermally altered ocean crust or Cl -rich brines stored in the crust (Michael & Schilling, 1989; Michael & Cornell, 1998; Perfit et al. 1999; Coogan et al. 2003a; le Roux et al. 2006) Isotopic signature of a ssimilation Given that the 9N OSC dacites have radiogenic isotopes similar to those in spatially related basalts ( Figure 29) what effects might assimilation, particularly of altered crust, have on derivative melts? AFC processes can change radiogenic isotopic ratios if the assimil ant has relatively high concentrations of the element in question and significantly different isotopic ratios than the original magma reservoir (e.g. Taylor, 1980; De Paolo, 1981) In general however, the effect on the resultant isotopes is less dramatic in AFC processes compared to partial melting because AFC processes create

PAGE 46

46 mixtures of altered and fresh material, while melting alone will retain the isotopic signature of the altered crust. Assimilation of altered oceanic crust may increase Sr isotope ratios depending on the amount of assimilation and extents of fluid/rock interaction (e.g. Alt & Teagle, 2003) Altered ocean crust can have a range of Sr concentrations (from less than to greater than typical MORB compositions) depending on the type/degree of alteration (Alt & Teagle, 2003) Assuming 80% fractional crystallization (which will not change the isotopic ratios) and assimilation of 10 mass percent wall rock, we can calculate the isotopic composition of the resulting melt using a ratio of 2:1 (fractionated melt to anatectic melt). Using reasonable values for the isotopic ratios of altered sheeted dike lavas (0.7028; average of basal dikes from Pito Deep; Hess Deep and Hole 504B; Barker et al. 2008) and initial MORB parental magma (0.7025) and their respective Sr concentrations (100 ppm and 120 ppm; a typical altered East Pacific Rise concentration) mass balance calcula tions indicate assimilation of altered crust will cause an increase of ~0.0001 in the 87Sr/86Sr ratio of the final magma. EC -AFC modeling of Sr isotopes produces similar results but requires less crystal fractionation (73%) and a ratio of crystallization to assimilation of 0.07 to produce the most radiogenic 87Sr/86Sr signatures observed in the MOR dacites (0.70258). Therefore, this process has the potential to slightly affect the Sr isotope ratios in the dacitic magma, but will not result in ratios as el evated those generated directly from partial melting of altered oceanic crust (~0.7028). Additionally, slightly elevated Sr isotope ratios in highsilica lavas from the Galapagos Spreading Center are consistent with AFC processes (Perfit et al. 1999) In comparison, Nd isotopes are unaffected during fractional

PAGE 47

47 crystallization and are relatively immobile during hydrothermal alteration (Michard & Albarde, 1986; Delacour et al. 2 008) Nd isotopes from the OSC dacites are similar to basalts from the region. 9 N OSC dacites form a tight cluster in Pb isotopic composition compared to 9N OSC basalts ( Figure 2-9). Pb isotopes are not significantly affected by hydrothermal alterati on provided sediments (which are not abundant in this environment) are not involved in the alteration process. We suggest that the similarity in isotopic ratios in the dacites compared to the basalts represents an overall averaging of isotopic values from basalts in the region due to melting and assimilating a range of different MORB compositions at the base of the sheeted dike layer. Relatively low oxygen isotope ratios observed in MOR dacites from the Galapagos Spreading Center (Perfit et al. 1999) and the 9 N OSC (Wanless et al. 2009; Wanless et al. in prepa) also support these conclusions. Fresh MORB will have mantle oxygen isotope values (~5.5), however, seawater alteration (seawater 18O = 0) will decrease this ratio (Gillis et al. 2001) while fractional crystallization of Fe oxides, and to a lesser extent olivine and pyroxene, will cause an increase (Matsuhisa et al., 1973). Therefore, partial melting and assimilation of altered basalt should produce melts with lower oxygen isotope ratios than predicted by fractional crystallization calculations (Muehlenba ch & Clayton, 1972) MOR dacites have oxygen isotope ratios similar to MORB values (Perfit et al. 1999; Perfit et al. 2007; Wanless et al. 2009; Wanless et al. in prepa) suggesting that fractional crystalliza tion alone cannot explain the formation of dacites on MOR. Taylor (1968) suggests that to a first approximation, the effect of assimilation on oxygen isotopes can be determined using mass balance.

PAGE 48

48 Assuming an evolved magma has a 18O ratio of 6.8 (largely due to fractionation of silicates and iron oxides) and an assimilant has a 18O ratio of 3.5 (due to seawater alteration), the resultant oxygen isotope ratio of the AFC magma would be ~6. This value is similar to those observed in the MOR dacites and is less than predicted by fractional crystallization alone. AFC Processes and Tectonic Setting The remarkable geochemical similarity of dacites erupted at the three different MOR discussed here indicates similar processes are controlling their petrogenesis (F igure 22) and we propose these processes are linked to tectonomagmatic settings. Andesites and dacites have erupted on several ridges, often in regions of propagation, such as propagating ridge tips and OSC (Chris tie & Sinton, 1981; Perfit et al. 1983; Sinton et al. 1983; Sinton et al. 1991) High-silica lavas have also been found along the Pacific -Antarctic Rise (Haase et al. 2005) at Axial Seamount on the Juan de Fuca Ridge (Chadwick et al. 2005) adjacent to a large axial magma chamber, and at the end of a first and secondorder ridg e segments on the N orthern East Pacific Rise (~ 837 N and 1030N (Langmuir et al. 1986) Collectively, these lavas erupted on intermediate to fast spreading ridges, in settings where magma reservoirs have the potential to undergo extensive fractional crystallization and interact with colder, and variably altered crust. A key resul t of this study is that the geochemical signatures of MOR dacites require assimilation of partial melts of hydrothermally altered crust into an extremely fractionated magma (that provides the heat needed to melt the oceanic crust). The extensive amounts o f fractional crystallization required suggest episodic or sporadic

PAGE 49

49 magma supply to magma reservoirs, which may not be characteristic of more steady state ridge environments. These requirements are met at the ends of ridge segments, where magma reservoir s may have a sporadic magma supply. In these regions, magmas are considered to be fed intermittently to the ridge tip through dike propagation from a more robust central region (Christie & Sinton, 1981) Between diking events the ridge tip magma supply is cut off, allowing for increased extents of crystal fractionat ion and interaction of the melt with older, altered crust. This may increase the likelihood of eruption of high -silica lavas through AFC processes. This is not to suggest that AFC processes do not occur in steady -state ridge environments but that the hi gh -silica melts may not be preserved or erupted in these regions (see section on Effects of Assimilation on Typical MORB). Model for Formation of MOR Dacites Based on the similarity of composition of high -silica lavas from three MOR, petrologic modeling calculations applied to dacites at the 9oN OSC, and published experimental results, we suggest that MOR dacites form under specific conditions that include: 1. A tectonomagmatic setting in which magma injection is episodic, allowing for extensive crystal fr actionation; 2. The presence of altered crust, which facilitates geochemical enrichments observed in the MOR dacites. Based on these two requirements, the tectonic setting and available geophysical information, we propose the following model for dacite fo rmation on MOR ( Figure 2-15): 1 Injection of basaltic magma by lateral dike propagation. Formation of axial magma reservoirs. 2 Magma supply to the region is cut off or reduced, allowing for extensive fractional crystallization of the magma reservoir.

PAGE 50

50 3 During extensive fractional crystallization the released latent heat of crystallization initially heats then partially melts the surrounding altered wall rock, which might be layer 2b dikes or high level altered gabbros. 4 The anatectic melts are assimilated into t he fractionating magma body. The AFC melts may be of dacitic composition depending on crustal temperatures, extent of fractional crystallization, amount of anatectic melt, and efficiency of assimilation. This model may account for the formation of highly evolved magmas at OSC propagating ridge tips, ridge transform intersections and along dikes associated with the down-rift volcanism on Axial seamount. This situation may also be analogous to Krafla volcano in Iceland, where the imaged melt lens is thought to be primarily composed of iron-rich basalts but high -silica lavas are associated with the edges of the caldera rim, where increased magma-rock interactions may be likely (Nicholson et al. 1991) Relationship of Melt Lens to Dacites at 9N Dacitic lavas at 9N OSC erupted on axis, over th e eastern edge of the large, seismically imaged melt lens (Kent et al. 2000; Dunn et al. 2001) Despite the eruption of young, fresh high -silica lavas in the neovolcanic zone, the underlying melt lens is not ass umed to be dacitic in composition. The composition of the basalts overlying the axial magma chamber suggests it has undergone a moderate amount of crystallization (to ferrobasalts). This suggests that the melt lens is composed primarily of basaltic magma t hat has mixed to varying degrees with a highly evolved endmember on axis (Wanless et al. in prep-b) There is also evidence of relatively young off -axis basaltic volcanism over the main body of the imaged melt lens (north of the overlap basin; (Nunnery et a l. 2008) The presence of a large, seismically imaged melt lens at 9N OSC does not contradict the episodic magma supply requirement for dacite formation. Instead, it may enable and enhance AFC processes in the region. AFC modeling suggests that

PAGE 51

51 extens ive fractional crystallization is required to produce dacitic compositions, which suggests low magma supply. Somewhat antithetically, the 9N region has an anomalously large basaltic axial magma chamber suggesting the current melt lens may only be indirec tly related to the dacites. We envision that the large melt lens is acting as a mobilizer for the eruption of the dacites that formed in isolated magma pockets or sills on the eastern edge of the axial magma chamber. Additionally, small -scale local mixin g of highly evolved compositions with the moderately evolved basaltic melt lens may account for the range of compositions erupting at the OSC ( Figure 215). Effects of Assimilation on Typical MORB Compositions The formation of dacite compositions on MOR r equires assimilation of anatectic melts into residual fractionated magmas, however, AFC processes may also explain slightly elevated incompatible elements observed in MORB lavas from all sections of MOR. The geochemical signatures of anatectic melts may b e subtle in less evolved magmas, however, elevated Cl, Al2O3, and K2O are common (Michael & Schilling, 1989; Michael & Cornell, 1998; Perfit et al. 1999; Coogan et al. 2003a; le Roux et al. 2006) At more magmat ically active ridge sections, wall rock may have a higher initial temperature, which would allow for melting and assimilation to begin earlier in the evolution of the magma body and require less fractional crystallization to occur prior to melting ( Figure 2 13). For instance, at 800C assimilation begins after only 53% fractional crystallization compared to 68% crystallization for 720C crust. Assimilation of anatectic melts into a magma reservoir that has undergone less crystallization produces less evol ved compositions and therefore, cannot produce MOR dacites. It does, however, increase the incompatible element abundances in the melt phase and may

PAGE 52

52 explain the commonly noted anomalous incompatible element enrichments, low FeO and elevated Al2O3 concentr ations in some normal MORB lavas. Conclusions The majority of eruptions at spreading centers produce basalts with relatively limited chemical variability; however, high-silica lavas have been sampled at several ridges. Eruptions of andesites and dacites are typically associated with ridge discontinuities and produce significant volumes of lava at a local scale. Limited amounts of these lavas have been sampled at the southern terminus of the Juan de Fuca Ridge, along the eastern Galpagos spreading center, at 837N and off axis at ~1030N on the East Pacific Rise. We have documented more voluminous eruptions of high silica lavas including highly evolved dacite on the propagating eastern limb of the 9N overlapping spreading center (OSC) on the East Pa cific Rise. Collectively, the dacites appear to represent an endmember composition that shows similar major element trends and incompatible trace element enrichments, suggesting similar processes controlled their petrogenesis. The formation of highly evolved lavas on MOR requires a combination of partial melting, assimilation and crystal fractionation. The highly enriched incompatible trace element signatures cannot be produced through crystal fractionation alone and appears to require partial melting of altered ocean crust. EC AFC modeling suggests significant amounts (>75%) crystallization of a MORB parent magma and modest amounts (520%) of assimilation of hydrothermally altered ocean crust can produce geochemical signatures consistent with dacite com positions. The AFC process explains trace element abundances in high-silica lavas and accounts for several major and minor element concentrations (i.e. Al2O3, K2O and Cl).

PAGE 53

53 An important constraint provided by AFC calculations is the temperature of the assim ilant. Varying the wall rock temperature can change the amount of crystallization/assimilation that occurs and the overall enrichments observed in incompatible trace element concentrations. The formation of dacites at the 9N OSC requires temperatures of surrounding crust to be 650 720 C, which requires >68% fractional crystallization ferrobasalt magma before melting can begin. While this amount of crystallization is unlikely in regions of high/constant magma supply, the surrounding wall rock in typic al ridge settings may be much warmer than at the ends of ridge segments, allowing for assimilation at much lower percents of crystallization. At wall rock temperatures of 800 C, calculations suggest that assimilation begins after ~53 wt% fractional cryst allization. This suggests that while conditions are not appropriate for the petrogenesis of dacites at typical ridge settings, assimilation of crustal material may be common but geochemically cryptic. The formation of high-silica lavas on MOR appears t o require a unique tectono magmatic setting, where episodic magma supply allows for extensive crystal fractionation, partial melting and assimilation. These conditions are met in regions of ridge propagation, such as OSC and propagating ridge tips, where diking allows for episodic injection of magma into older, altered ocean crust. Here, the magma undergoes extreme crystallization without repeated replenishment, creating enough latent heat of crystallization to melt and assimilate surrounding wall rock. Acknowledgements We thank the Captain, officers and crew of the R/V Atlantis for all their help during cruise AT15-17, the MEDUSA2007 Science party (including White, S., Von Damm, K., Fornari, D., Soule, A., Carmichael, S., Sims, K., Zaino, A., Fundis, A.

PAGE 54

54 Mason, J., OBrien, J., Waters, C., Mansfield, F., Neely, K., Laliberte, J., Goehring, E., and Preston, L. ) for their diligence in collecting data and samples for this study. We thank the Jason II shipboard and shore based operations group for their ass istance in collecting these data and HMRG for processing all DSL120 sidescan and bathymetry collected during this cruise. Discussions with S. White and A. Goss are gratefully acknowledged and contributed to this research. Thanks to G. Kamenov and the UF Center for Isotope Geoscience for laboratory assistance. This research was supported by the National Science Foundation [grants OCE 0527075 to MRP and OCE -0526120 to EMK].

PAGE 55

55 Table 2 1. Dacite major and trace element data Sample # 266 58 265 65 265 64 265 67 266 50 266 53 265 84 265 63 266 47 265 85 266 46 265 94 SiO2 63.0 63.8 64.0 64.1 64.3 64.3 64.4 64.4 64.5 65.0 65.0 65.2 TiO2 1.10 1.26 1.28 1.34 1.07 1.06 1.13 1.29 0.99 1.06 0.94 0.97 Al2O3 13.1 13.2 13.1 13.3 13.2 13.3 13.2 13.3 13.2 13.1 12 .9 13.0 FeO 8.43 8.14 8.27 8.49 8.08 8.06 8.18 8.22 7.74 7.99 7.17 7.90 MnO 0.16 0.15 0.16 0.16 0.14 0.14 0.15 0.15 0.14 0.16 0.15 0.14 MgO 1.75 1.34 1.60 1.49 1.27 1.12 1.23 1.29 1.02 1.18 1.41 1.13 CaO 4.34 4.21 4.45 4.41 3.78 3.73 3.92 4 .21 3.53 3.78 3.71 3.54 Na2O 3.63 3.84 3.46 3.93 4.23 4.16 3.41 3.71 4.94 3.67 4.76 4.29 K2O 0.96 0.97 0.97 0.95 1.10 1.09 1.19 0.99 1.22 1.22 1.19 1.14 P2O5 0.26 0.22 0.20 0.23 0.24 0.25 0.22 0.21 0.23 0.20 0.17 0.23 Cl 0.24 0.65 0.64 Total 96.69 97.17 97.55 98.42 97.35 97.20 96.98 97.76 97.54 97.41 97.43 97.61 Trace Elements (ppm) Li 32 34 32 30 29 30 31 31 Sc 17 20 18 15 15 17 14 13 V 122 140 121 61 102 121 73 63 Cr 13 12 12 4 2 10 4.6 1 Co 15 17 16 13 14 15 12 11 Ni 9.0 9.8 9.2 5.8 7.0 8.4 6.6 4.9 Cu 17 19 18 18 21 17 19 17 Zn 110 124 122 109 100 113 108 105 Ga 30 35 28 29 29 31 28 30 Cs 0.11 0.12 0.12 0.14 0.13 0.10 0.15 0.13 Rb 9.1 10 10 13 13 9.5 14 12 Ba 51 57 53 65 60 50 68 60 Th 1.7 1.8 1.6 2.3 2.3 1.7 2.4 2.3 U 0.59 0.65 0.64 0.86 0.82 0.59 0.92 0.84 Nb 13 15 13 16 15 13 15 16 Ta 0.81 1.1 0.92 1.0 1.4 0.99 1.1 1.2 La 23 26 23 29 27 24 29 29 Ce 67 77 68 82 78 68 82 84 Pr 10 11 10 12 1 1 10 12 12 Sr 76 90 89 86 70 76 81 68 Nd 46 53 46 54 51 47 52 55 Zr 735 842 622 856 968 745 872 1050 Hf 19 21 18 22 23 19 23 25 Sm 14 15 15 16 14 14 16 16 Eu 3.0 3.3 3.1 3.2 2.8 3.0 3.0 3.0 Tb 3.2 3.4 3.4 3.6 3.1 3.1 3.6 3.4 Dy 21 23 23 24 20 21 24 22 Y 132 148 142 157 133 132 154 146 Ho 4.6 4.9 4.9 5.2 4.3 4.4 5.1 4.8 Er 14 15 15 16 13 13 16 14 Yb 14 15 16 17 14 14 17 15 Lu 2.1 2.3 2.4 2.6 2.1 2.1 2.6 2.3 Tm 2.2 2.3 2.4 2.5 2.1 2.1 2.5 2.3 Gd 17 18 17 19 17 16 19 18 Pb 3.3 5.0 3.4 3.2 4.9 5.8 3.6 5.7

PAGE 56

56 Table 2 1. Continued Sample # 264 09 265 70 265 42 265 83 265 95 265 40 SiO2 65.8 66.3 66.5 67.5 67.5 TiO2 0.89 0.87 0.94 0.76 0.77 Al2O3 13.2 13.2 13.0 13.3 13.1 FeO 7.03 7.17 7.92 6.68 6.47 MnO 0.13 0.14 0.16 0.13 0.12 MgO 1.06 0.80 0.89 0.67 0.94 CaO 3.48 3.23 3.50 2.98 3.01 Na2O 4.24 4.08 3.99 3.88 4.43 K2O 1.21 1.33 1.20 1.37 1.21 P2O5 0.20 0.19 0.21 0.16 0.15 Cl 0.58 0.70 0. 51 0.67 Total 97.78 97.27 98.91 97.37 97.67 Trace Elements (ppm) Li 27 34 32 32 31 31 Sc 12 12 14 11 10 12 V 46 45 58 32 52 51 Cr 4 4 3 3 3 4 Co 10 10 11 8 8 10 Ni 5.7 5.0 5.9 4.7 5.4 5.3 Cu 16 16 15 14 15 17 Zn 89 106 119 103 98 103 Ga 28 29 30 29 30 30 Cs 0.13 0.17 0.16 0.17 0.13 0.13 Rb 13 15 14 15 12 12 Ba 66 73 70 76 62 60 Th 2.6 2.7 2.4 2.8 2.4 2.5 U 0.98 1.04 0.91 1.05 0.86 0.84 Nb 15 16 16 16 17 17 Ta 1.0 1.1 1.1 1.1 1.2 1.0 La 29 31 29 31 29 29 Ce 84 88 83 87 84 85 Pr 12 12 12 12 12 12 Sr 78 78 83 76 61 73 Nd 53 55 53 54 55 57 Zr 824 934 816 922 985 1013 Hf 23 25 22 25 25 25 Sm 16 17 17 16 16 17 Eu 3.0 3.1 3.4 3.1 2.9 3.3 Tb 3.6 3.7 3.8 3.6 3.3 3.7 Dy 24 25 25 25 22 24 Y 151 160 160 159 145 146 Ho 5.1 5.4 5.5 5.3 4.7 5.2 Er 16 16 16 16 15 15 Yb 17 18 18 18 15 16 Lu 2.6 2.7 2.7 2.7 2.3 2.4 Tm 2.5 2.6 2.6 2.6 2.3 2.5 Gd 19 19 20 19 18 20 Pb 2.8 3.8 3.5 3.8 4.1 3.6

PAGE 57

57 Table 2 2. Radiogenic i sotopes for 9N OSC l avas Sample 208Pb/ 204Pb 207Pb/ 204Pb error 206Pb/ 204Pb 208Pb/ 206Pb E. Limb Basalts 264 04 37.699 1.74E 03 15.476 7.89E 04 18.279 8.93E 04 2.062 2.87E 05 265 18 37.677 1.94E 03 15.472 7.33E 04 18.275 7.80E 04 2.062 3.56E 05 265 35 37.664 1.60E 03 15.470 5.59E 04 18.250 7.06E 04 2.064 3.08E 05 265 43 37.661 1.51E 03 15.467 5.45E 04 18.249 6.11E 04 2.064 3.17E 05 265 113 37.695 1.67E 03 15.477 6.33E 04 18.275 6.83E 04 2.063 2.45E 05 266 01 37.642 1.95E 03 15.469 7.30E 04 18.235 9.22E 04 2.064 2.17E 05 266 33 37.683 2. 19E 03 15.474 8.55E 04 18.277 8.56E 04 2.062 2.76E 05 265 05 37.687 1.96E 03 15.478 6.26E 04 18.294 7.24E 04 2.060 3.51E 05 E. Limb Basaltic Andesites 265 24 37.683 1.68E 03 15.473 5.87E 04 18.264 6.10E 04 2.063 3.34E 05 265 56 37.681 1.49E 03 15.474 5 .67E 04 18.270 6.18E 04 2.062 3.06E 05 265 91 37.683 1.40E 03 15.475 5.25E 04 18.266 5.05E 04 2.063 3.09E 05 265 103 37.674 2.06E 03 15.472 7.56E 04 18.261 8.79E 04 2.063 2.74E 05 265 109 37.673 1.79E 03 15.471 6.69E 04 18.259 7.04E 04 2.063 2.64E 05 2 65 125 37.672 1.90E 03 15.471 7.32E 04 18.265 7.80E 04 2.063 2.57E 05 264 20 37.690 1.54E 03 15.476 6.14E 04 18.269 5.94E 04 2.063 3.47E 05 265 49 37.689 1.60E 03 15.474 5.77E 04 18.268 6.22E 04 2.063 3.25E 05 E. Limb Andesites 264 14 37.673 2.08E 03 1 5.472 9.65E 04 18.262 1.11E 03 2.063 5.32E 05 265 25 37.661 1.69E 03 15.469 6.99E 04 18.251 7.47E 04 2.064 2.44E 05 265 90 37.675 1.52E 03 15.472 6.01E 04 18.262 6.60E 04 2.063 2.75E 05 265 100 37.680 1.46E 03 15.474 5.81E 04 18.265 6.24E 04 2.063 2.68E 05 266 54 37.688 1.69E 03 15.475 6.59E 04 18.272 7.73E 04 2.063 3.29E 05 E. Limb Dacites 264 09 37.676 1.78E 03 15.472 7.14E 04 18.268 8.10E 04 2.062 2.88E 05 265 40 37.674 2.05E 03 15.471 8.24E 04 18.265 8.86E 04 2.063 3.58E 05 265 42 37.678 1.72E 0 3 15.472 6.53E 04 18.270 7.40E 04 2.062 3.40E 05 265 64 37.682 2.51E 03 15.476 9.76E 04 18.267 1.13E 03 2.063 3.32E 05 265 70 37.676 1.96E 03 15.471 7.58E 04 18.266 7.97E 04 2.063 3.64E 05 265 83 37.681 1.51E 03 15.473 5.94E 04 18.271 6.65E 04 2.062 2.4 9E 05 265 84 37.689 1.67E 03 15.476 6.55E 04 18.274 7.49E 04 2.062 2.59E 05 265 85 37.677 2.05E 03 15.471 7.79E 04 18.267 8.31E 04 2.063 3.02E 05 265 95 37.680 1.67E 03 15.475 6.95E 04 18.263 8.27E 04 2.063 3.80E 05 266 53 37.678 1.32E 03 15.472 5.01E 04 18.268 5.10E 04 2.063 3.16E 05 2 sigma error reflects in run machine error. Long term reproducibility estimates are: 87 Sr/ 86 Sr = 0.00003, 143Nd/144Nd = 0.000018, 206Pb/204Pb = 0.0034 (205 ppm), 207Pb/204Pb = 0.0028 (184 ppm), 208 Pb/ 204 Pb = 0.00 86 (234 ppm) *Unknowns were normalized to an 143 Nd/ 144 Nd value of 0.511215 0.000007 for JNdi 1, which is reported by Tanaka et al. (2000) relative to a La Jolla 143Nd/144Nd value of 0.511858 (Lugmair and Carlson, 1978).

PAGE 58

58 Table 2 2. Continued Sample 87Sr /86Sr 143Nd/144Nd* Eps Nd E. Limb Basalts 264 04 0.70250 0.000011 0.513163 0.000011 10.2 265 18 0.70249 0.000018 0.513190 0.000013 10.8 265 35 0.70247 0.000012 0.513164 0.000005 10.3 265 43 0.70250 0.000021 0.513154 0.000009 1 0.1 265 113 0.70249 0.000016 0.513172 0.000005 10.4 266 01 0.70246 0.000015 0.513158 0.000007 10.1 266 33 0.70244 0.000013 0.513160 0.000005 10.2 265 05 0.70243 0.513191 0.000008 10.8 E. Limb Basaltic Andesites 265 24 0.70256 0.000013 0.513187 0.000 005 10.7 265 56 0.70253 0.000012 0.513187 0.000004 10.7 265 91 0.70253 0.000012 0.513179 0.000004 10.6 265 103 0.70245 0.000012 0.513162 0.000004 10.2 265 109 0.70249 0.000016 0.513145 0.000007 9.9 265 125 0.70246 0.000045 0.513154 0.000006 10.1 264 20 0.70244 0.513179 0.000004 10.6 265 49 0.70244 0.513196 0.000007 10.9 E. Limb Andesites 264 14 0.70243 0.000014 0.513153 0.000005 10.0 265 25 0.70254 0.000018 0.513152 0.000005 10.0 265 90 0.70250 0.000011 0.513159 0.000006 10.2 265 100 0.70249 0 .000012 0.513179 0.000004 10.6 266 54 0.70246 0.513193 0.000005 10.8 E. Limb Dacites 264 09 0.70254 0.000019 0.513171 0.000008 10.4 265 40 0.70247 0.000011 0.513149 0.000008 10.0 265 42 0.70246 0.000015 0.513140 0.000006 9.8 265 64 0.70258 0.000015 0.513185 0.000007 10.7 265 70 0.70248 0.000024 0.513141 0.000005 9.8 265 83 0.70253 0.000015 0.513147 0.000007 9.9 265 84 0.70251 0.000011 0.513179 0.000007 10.6 265 85 0.70248 0.000019 0.513148 0.000005 9.9 265 95 0.70247 0.000019 0.513165 0.000006 1 0.3 266 53 0.70247 0.000015 0.513175 0.000010 10.5

PAGE 59

59 Table 2 3. Starting compositions for modeling Sample # 264 08 265 43 265 113 SiO2 50.1 50.5 51.9 TiO2 2.68 1.92 2.17 Al2O3 12.7 13.9 13.4 FeO 14.1 11.6 12.8 MnO 0.26 0.21 0.23 MgO 5.69 6.98 5.93 CaO 9.58 11.14 9.48 Na2O 3.27 2.86 3.28 K2O 0.21 0.13 0.22 P2O5 0.28 0.19 0.28 Cl 0.07 0.01 0.02 Total 99.24 99.67 99.72 Trace Elements (ppm) Li 10 8 11 Sc 42 42 38 V 450 347 323 Cr 18 109 32 Co 43 41 39 Ni 34 54 34 Cu 60 59 51 Zn 116 94 103 Ga 21 18 21 Cs 0.03 0.01 0.03 Rb 2.1 1.1 2.2 Ba 17 8 16 Th 0.34 0.18 0.40 U 0.13 0.08 0.15 Nb 5.5 3.1 5.7 Ta 0.37 0.21 0.36 La 6.8 4.5 7.6 Ce 21 14 24 Pr 3.3 2.3 3.9 Sr 126 120 111 Nd 17 13 20 Zr 180 126 229 Hf 4.8 3.4 5.7 Sm 6.0 4.4 6.7 Eu 1.9 1.5 2.0 Tb 1.5 1.1 1.6 Dy 10 7 10 Y 59 44 61 Ho 2.0 1.5 2.2 Er 5.9 4.4 6.3 Yb 6.0 4.4 6.2 Lu 0.93 0.68 0.95 Tm 0.91 0.67 0.96 Gd 7.7 5.8 8.5 Pb 0.61 0.39 0.95

PAGE 60

60 Table 2 4. Partition coefficients for Rayleigh f ractional crystallization, partial melting and AFC calculations Andesite Partition Coefficients Basalt Partition Coefficients Element Olivine CPX Plag Apatite Ilmenite Amphibole Olivine CPX Plag Ilmenite Rb 0.01 0.02 0.025 0.001 0.034 0.04 0.0003 0.0004 0.056 0.034 Ba 0.01 0.02 0.155 0.12 0.00034 0.1 0.00001 0.0003 1.45 0.00034 Th 0.01 0.01 0.19 1.28 0.00055 0.15 7.00E 06 0.0021 0.13 0.00055 U 0 0 0.34 1.4 0.0082 0.008 9.00E 06 0.001 0.051 0.0082 Nb 0.00017 0.005 0.033 0.0011 2 0.28 0.0000 5 0.0089 0.045 2 Ta 0.00002 0.014 0.11 0.003 1.7 0.27 0 0.013 0.066 1.7 La 0.00006 0.062 0.082 11.4 0.000029 0.027 0.0002 0.054 0.13 0.000029 Ce 0.00006 0.116 0.072 12.9 0.000054 0.0293 7.00E 05 0.086 0.11 0.000054 Sr 0.00217 0.08 2.7 4.3 0 0.28 0.00004 0.091 1.4 0 Nd 0.00015 0.33 0.045 32.8 0.00048 0.0325 0.0003 0.19 0.066 0.00048 Zr 0.00450 0.14 0.0009 0.042 0.29 0.26 0.001 0.26 0.048 0.29 Hf 0.00370 0.21 0 0.014 0.38 0.43 0.0029 0.33 0.051 0.38 Sm 0.00044 0.41 0.033 16.1 0.00059 0.024 0.0 009 0.27 0.054 0.00059 Eu 0.00056 0.57 0.55 25.5 0.009 0.0498 0.0005 0.43 0.65 0.009 Dy 0.00250 0.94 0.034 34.8 0.01 0.0136 0.0027 0.44 0.024 0.01 Y 0.00380 0.9 0.01 7.1 0.0045 0.0196 0.0082 0.47 0.013 0.0045 Yb 0.05600 0.63 0.014 15.4 0.17 0.102 0.024 0.43 0.0079 0.17 Lu 0 0.605 0.039 3.92 0.084 0 0.016 0.56 0.06 0.084 Primary Reference Zanetti et al., 2004 Klein et al., 2000 Dunn and Sen, 1994 Prowatke and Klemme, 2006 Zack and Brumm, 1998 Bottazzi et al., 1999 Halliday et al., 1995 Halliday et al., 1995 Dunn and Sen., 1994 Zack and Brumm, 1998 Secondary Reference Gill, 1979 Gill, 1979 Rollinson, 1993 Fujimaki, 1986 Rollinson, 1993 Haase et al., 2005 Rollinson, 1993 Rollinson, 1993 Rollinson, 1993 Rollinson, 1993

PAGE 61

61 Table 2 5. AFC m odeling p arameters Parameter Abbreviation Value Units Magma Liquidus Temp tlm 1200 deg C Magma Temp tmo 1200 deg C Assm. Liquidus Temp tla 1100 deg C Country Rx Temp tao 720 deg C Solidus Temp ts 875 deg C Magma Spec. Heat cpm 1484 J/Kg K Assm. Spec. Heat cp a 1388 J/Kg K Crsyt. Enthal Hcry 396000 J/Kg Fusion Enthal Hfus 354000 J/Kg Equilibration Temp Teq 900 deg C

PAGE 62

62 A Figure 21. Bathymetric maps showing the tect onic setting of the MOR dacites (data from GeoMapApp; Carbotte et al. 2004) Boxe s show the general locations of high-silica lavas on each ridge. Dacites are commonly associated with the ends of ridge segments, such as overlapping spreading centers (OSC) and propagating ridge tips at ridge transform intersections. A) 9N OSC on the Ea st Pacific Rise (EPR) B) Propagating Ridge Tip on the Juan de Fuca Ridge (JdFR) and Axial seamount and C) possible OSC near the Propagating Ridge Tip on the Galapagos Spreading Center (GSC).

PAGE 63

63 B Figure 21. Continued

PAGE 64

64 C Figure 21. Continued

PAGE 65

65 Figure 22. Comparison of major and trace element compositions in MOR high-silica andesites and dacites from the East Pacific Rise Juan de Fuca Ridge, and Galapagos Spreading Center MOR dacitic lavas have similar major and trace element compositions, while andes ites have more variable compositions that lie between dac ites and highly evolved MORB. A) Mantle -normalized diagram showing similarities in trace element compositions between dacites from three different ridges and an andesite from Axial Seamount. Averag e composition for N MORB from the 9 17 10N segment of the East Pacific Rise is shown for comparison. MOR dacites are characterized by low Nb and Ta and high U, Th, Zr and Hf relative to elements of similar incompatibilities. B) and C) Major element plots showing the range of compositions of MOR dacites compared to East Pacific Rise MORB. Gray field shows the range of compositions from >1200 analyses of MORB glasses from the East Pacific Rise north of the 9N OSC (from PetDB, Sims et al., 2002, 2003 and Perfit unpublished data).

PAGE 66

66 Figure 23. Bathymetric map of the 9N OSC showing locations of samples collected during the MEDUSA 2007 cruise. Samples are divided into rock types based on silica content (dacite >62wt% SiO2; andesite 57 62 wt % SiO2; basaltic andesite 525 7 wt% SiO2; basalts <52wt% SiO2 and FeTi basalts <52wt% SiO2 and >12 wt% FeO). Dacites are primarily found on axis on the eastern limb of the OSC. 50 m contour intervals are shown.

PAGE 67

67 Figure 24. Photographs of MOR high-silica lavas from A) and B) 9N OSC East Pacific Rise C) Galapagos Spreading Center and D) Juan de Fuca Ridge Morphologically, dacites typically form blocky angular flows and large elongate pillow lavas with roughly corrugated or striated surfaces.

PAGE 68

68 Figure 25. Major element variations versus MgO (wt.%) for dacites from the 9N OSC on the East Pacific Rise (filled squares). Gray field represents all lavas collected in 2007 from the 9N OSC. Dacites are compared to three low pressure fractional crystallization trends (calculated using MELTS; Ghiorso & Sack, 1995) using parental compositions from OSC basalts (see text for modeling parameters and details) Not all dacite major element variations can be explained by fractional crystallization alone (e. g. Al2O3, K2O, and P2O5). Experimental compositions from partial melting of altered basalt (Beard and Lofgren, 1991) are shown for comparison (*). A B C D E F

PAGE 69

69 Figure 26. Variation diagram showing Cl (wt%) versus MgO (wt%) for OSC lavas. Superimposed are three liquid lines of descent (calculated using Melts; Ghiorso & Sack, 1995), showing that the maximum amount of Cl enrichment due to extensive fractional crystallization cannot produce the high Cl concentrations in the OSC dacites. The dashed rectangle represents the range o f compositions that can be produced through 1-15% partial melting of an altered basalt with 350 ppm Cl (purple star). 350 ppm Cl is the median value of Cl analyzed in sheeted dikes from ODP Hole 504B with minimum and maximum values (49 and 650 ppm) shown with an error bar. Partition coefficients for Cl are from Gillis et al., (2003). The range of MgO values for partial melts were taken from experimental partial melts of less than 15% (Beard & Lofgren, 1991).

PAGE 70

70 Figure 27. Trace element variations versus Zr (ppm) in 9N OSC lavas. Superimposed on the diagrams are calculated trends for fractional crystallization ( model 1 and 2 using the Rayleigh fractionation equation), 1 to 15 % partial melting (assuming batch melting), and AFC simulations (EC -AFC; Bohrs on and Spera, 2001). See text for model parameters. Kinks in models represent changes in crystallizing phases.

PAGE 71

71 Figure 28. Normalized trace element ratio diagrams showing the range of dacite compositions at the 9N OSC. Fractional crystallization, par tial melting and AFC trends are shown as in Figure 2-4. Tick marks on trend lines for partial melts are in 1% increments while fractional crystallization tick marks represent 10% intervals. Concentrations are mantle-normalized (Sun & McDonough, 1989)

PAGE 72

72 Figure 29. Radiogenic isotope rat ios of 9N OSC lavas. A) Pb -isotope ratios showing 9N OSC dacites with 208Pb/204Pb and 206Pb/204Pb ratios similar to OSC N MORB basalts and northern East Pacific Rise N MORB (gray field; data from (Sims et al. 2002; Sims et al. 2003; Goss et al. 2010) B) Nd and Sr isotopes with 9N OSC lavas that are similar to EPR and OSC N MORB.

PAGE 73

73 Figure 210. Elemental variation diagrams showing 9N OSC dacites versus calculated liquid lines of descent produced during two alternative types of crystallization. C rosses show the evolution of a melt during in situ crystallization (Langmuir, 1989). C ircles show the liquid line of descent of a magma that undergoes 50% fractional crystallization and is then separated (via filter pressing) from the phenocrysts (melt -segregation model) The resulting magma undergoes an additional 50% crystallization, until it once again separates from the crystals following the model of Bachman and Bergantz ( 2004) Kinks in models represent change s in crystallizing phases.

PAGE 74

74 Figure 211. Partial melting model showing a range of possible parents (gray lines) that could produce the 9N OSC dacites ( bold red line) from 1 15% partial melting. Sheeted dikes and the upper parts the gabbroic layer may b e composed of a range of compositions depending on the composition of the starting material and degree of alteration. Possible parental compositions were calculated using the batch melting equation and solving for the initial parent composition (Co). Thr ee possible wall rock compositions are superimposed on the calculated parental range. Altered basalt 1 (squares ; (Nakamura et al. 2007) provides the best match to the calculated source compositions and is used as the parent rock in subsequent partial melting and AFC models.

PAGE 75

75 Figure 212. Mantlenor malized diagrams showing results of 115% partial melting of an altered basalt (see Figure 2-11) using two different modal minerologies. Partial melts were calculated using the batch melting equation. See text for more detail. A) Partial melting results using a non amphibole bearing gabbro assemblage (19% Ol 30% Cpx, 50% Plag, 1% Ilm). B) Partial melting results using an amphibole bearing gabbro (20% Cpx, 25% Opx, 49% Plag, 5% Amph, 1% Fe-Oxide; Haase et al., 2005). The amphibole-bearing gabbro provide s the best fit to the 9N OSC dacites.

PAGE 76

76 Figure 213. Diagram showing the calculated effects of varying wall rock temperature on incompatible trace element composition (La) during AFC A) Higher wall rock temperatures cause earlier onset of assimilation compared to lower initial temperatures. Higher initial temperatures produces lower overall incompatible element abundances compared to lower wall rock temperatures because the amount of fractional crystallization is lower. The average La concentration o f the 9N OSC dacites is ~28ppm, which can be produced by assimilating partial melts of a wall rock at starting temperatures of 650 to 720C. B) Plot showing the ratio of assimilation to crystallization (Ma*/Mc) for various temperatures of wall rock. Low er ratios produce higher incompatible trace element concentrations in the hybrid melts.

PAGE 77

77 Figure 214. Mantlenormalized trace element diagram showing results of the best -fit AFC model (thin black lines) This model requires a total of 73-85% fractional crystallization in combination with 5-20% assimilation of partially melted wall rock to produce trace element compositions similar to the 9N OSC dacites (bold red line). Fractional crystallization is the dominant process until 68% of the magma has crystallized. This is followed by 5 -20% assimilation of partial melts in conjunction with an additional 517% crystallization.

PAGE 78

78 Figure 215. Cartoon showing a possible scenario for dacite formation on MOR. A) Injection of basaltic magma into a ridge segment end through dike propagation. B) Magma supply rates diminish at segment end, abandoning pockets of magma and allowing for extensive fractional crystallization. The latent heat of crystallization begins to heat up and partially melt the hydrothermally altered wall rock. C) Partial melts of wall rock are assimilated or mixed into the evolving magma chamber, resulting in dacitic magmas.

PAGE 79

79 CHAPTER 3 ROLE OF ASSIMILATION IN THE PETROGENESIS OF LAVAS ON MID -OCEAN RIDGES INFERRED FROM CL, H2O, CO2 AND OXY GEN I SOTOPE VARIATIONS Introduction Crustal assimilation has been proposed as an important process in the petrogenesis of mid ocean ridge (MOR) magmas (e.g., O'Hara, 1977; Michael & Schilling, 1989; Michael & Cornell, 1998) but it is largely ignored as a primary igneous process in ridge settings for several reasons. First, contamination is commonly overlooked because most geochemical variations in MOR lavas can be readily explained by fractional crystallization, variati ons in mantle melting parameters, or differences in mantle source compos itions. Second, the low volume of assimilated melt compared to the more voluminous midocean ridge basalt (MORB) magma s makes assimilation a difficult process to identify in the erupt ed lavas. Third, the magma and wall rock may have similar major and trace element compositions, resulting in liquids (assimilants) that are geochemically difficult to discriminate from typical MORB lavas. However, variable degrees of hydrothermal alterati on of basaltic crust can produce significant changes in fluid mobile element concentrations and isotopic ratios depending on the water/rock ratios (e.g., Alt & Teagle, 2000) Therefore, components that are particularly sensitive to seawater interaction, such as Cl, U, H2O, and oxygen isotopes, can be used to determine the extent to which crustal assimilation is involved in MOR magmatism. Contamination of MOR magmas by a seawater altered component was first proposed based on excess Cl in MORB glasses (Michael & Schilling, 1989; Mich ael & Cornell, 1998) The elevated Cl concentrations compared to elements of similar incompatibility (K, Nb, Ti) in fresh MORB glass cannot be explained by post eruption

PAGE 80

80 alteration or fractional crystallization and are instead attributed to assimilation of a seawater derived component such as saline brines or altered ocean crust (Michael & Schilling, 1989; Michael & Cornell, 1998) Consequently, Cl over enrichment has been identified in many submarine settings, including MOR (Perfit et al. 1999; Coogan et al. 2003b; le Roux et al. 2006; Wanless et al. 2010; Wanless et al. accepted) back arc basins (Kent et al. 2002; S un et al. 2007) and ocean islands (e.g., Kent et al. 1999) Despite the clear evidence of assimilation in these lavas, many models for the magmatic plumbing system at MORs continue to ignore this process. An alternate approach to identifying crust al contamination on MORs is through analyses of oxygen isotope ratios. Low oxygen isotope ratios relative to mantle values are observed in many lavas from Icelandic volcanoes (e.g., Gautason & Muehlenbach, 1998) Lower than mantle values have also been observed in melt inclusions from Hawaiian glasses and are attributed to assimilation of altered Pacific crust (Eiler et al. 1996; Garcia et al. 1998) Decreases in 18O relative to primary mantle like values are attributed to the temperature dependent fractionation of oxygen isotopes between seawater and mineral phases in ocean crust during hightemperature hydrothermal alteration (Muehlenbach & Clayton, 1972; Alt et al. 1996) Therefore, assimilation of crust altered at high -temperatures near MORs should lower the oxygen isotope ratios of the resulting magma. Unfortunately, this signal may be hard to identify in many MORB magmas because the mass of assimilant relative to parent magma is generally not sufficient to appreciably change the oxygen isotope rat ios. However, it may be an important discriminator in magmas that have undergone significant assimilation.

PAGE 81

81 Here, we examine Cl, H2O, and CO2 concentrations and oxygen isotope ratios from a suite of lavas collected at the 9N overlapping spreading center (O SC) on the East Pacific Rise (EPR) to place constraints on the role of crustal assimilation at MORs. This suite includes a nearly continuous range of compositions from basalts to dacites, including one of the most evolved lava compositions sampled on a MO R (>67 wt% SiO2). Major and trace element data indicate that assimilation of altered ocean crust is a critical process for the formation of MOR dacites (Wanless et al. accepted) and as such, it should be reflected in the Cl, H2O, and CO2 concentrations and oxygen isotope ratios of these lavas. A dditi onally, we use geochemical results to better define the source and depth of assimilation processes beneath MORs and use CO2 concentrations to provide information on degassing of magmas during extensive fractional crystallization and the ascent rates of hig h -silica magmas. Geologic Setting The 9N OSC is located between the Clipperton and Siqueiros transform faults on the EPR (Figure 3-1) It is a second order ridge discontinuity consisting of limbs that overlap by ~27 km and offset the ridge axis ~8 km fr om east to west (Semp e re & Macdonald, 1986) The OSC has been m igrating southward down the ridge axis at a rate of approximately 42 km/Myr as the eastern limb propagates into older crust and the western limb recedes or dies (Macdonald & Fox, 1983; Carbotte & Macdonald, 1992) The 9N OSC is one of the most extensively studied OSC on the MOR system. It has been the focus of several geophysical studies (Detrick et al. 1987; Harding et al. 1993; Kent et al. 1993; Kent et al. 2000; Bazin et al. 2001; Dunn et al. 2001; Tong et al. 2002) which have produced the first 3D multi -channel seismic survey of a midocean ridge (Kent et al. 2000) and a 3D seismic refraction study (Dunn et al. 2001)

PAGE 82

82 These studies reveal a shallow melt lens beneath both limbs of the OSC and in the interlimb region north of overlap basin (Kent et al. 2000) The western, receding limb melt lens is narrow and shows no significant variation in depth along axis (Kent et al. 2000) while the melt lens beneath the eastern, propagating limb shows variations in both width and depth. Beneath the souther n portion of the east limb the melt lens is narrow and significantly deeper than the rest of the eastern ridge axis, plunging ~500 m southward over ~6 km (Kent et al. 2000) North of the overlap basin, the melt lens is anomalously wide (> 4 km) and is not centered directly b elow the ridge axis, instead extending from the axis ~4 km to the west (Kent et al. 2000; Tong et al. 2002) Although the depth of the lens varies along axis, the top of the melt lens appears to follow the base o f the sheeted dikes, at approximately 1.5-2 km beneath the seafloor (Kent et al. 2000; Tong et al. 2002) The first lava sampling in this region occurred during the CHEPR dredging and wax coring cruise that recovered several high-silica lavas, along with basalts and FeTi basalts (Langmuir et al. 1986) More recently, the 9N OSC was the focus of the MEDUSA2007 research cruise (AT15 -17), which completed detailed mapping using the DSL -120A side -scan system (White et al. 2009) and the WHOI TowCam (Fornari, 2003) and extensive sampling using the ROV Jason2 (Wanless et al. accepted; Wanless et al. in prep-b) of the region. Results of this cruise (White et al. 2009; Wanless et al. accepted; Wanless et al in prepb) revealed a range of rock types from basalts to dacites but the high-silica lavas are confined to the eastern propagating limb of the OSC (Wanless et al. accepted) Additionally, the basalts erupted at the OSC are

PAGE 83

83 dominantly ferrobasalts in contrast to the dominantly MORB lavas erupted from the 915 to 10 N section of the EPR (Perfit et al. in prep) Analytical Methods and Results Major and Trace Elements During the MEDUSA2007 research cruise, 275 glassy samples were collected from the 9N OSC. Methods, standards and results from all major and trace element analyses are discussed in detail in Wanless et al., (accepted) and Wanless et al., (in prep.). Major elements were analyzed on a JEOL 8900 Electron Microprobe at the USGS in Denver, Colorado. Highp recision Cl and K2O concentrations were determined using 200 -second peak/100 second background counting times. Samples were analyzed for trace element concentrations on an Element2 Inductively Coupled Plasma Mass Spectrometer (ICP MS) at the University of Florida. Concentration data and ratios used in this paper are listed in Table 3 1. Here, we use the general term basalt and basaltic to include basalts sensu stricto, ferrobasalts and FeTi basalts. The intermediate lavas include both basaltic andesit es and andesites, with SiO2 concentrations that range from 52 57 wt% and 5762 wt%, respectively. Dacites have >62 wt% SiO2. Cl concentrations range from 0.01 to 0.07 wt% in the basalts and are generally higher in the basaltic andesites (0.01 to 0.31 wt % ), andesites (0.20 to 0.42 wt %) and dacites (0.23 to 0.70; Figure 3 -2). K2O concentrations range from 0.13 to 1.37 wt % and generally increase with increasing SiO2 content. Basalts range from 0.13 to 0.21 wt%, basaltic andesites range from 0.26 to 0.60 w t%, andesites range from 0.63 to 0.83 wt% and dacites range from 0.89 to 1.37 wt% ( Figure 3 -2). Cl/K2O ratio s range from 0.04 to 0.60 (Figures 3 -3 and 3 -4).

PAGE 84

84 Volatile Elements A representative subset of 20 samples (covering the range of rock types) from the 9N OSC was selected for volatile analyses (Table 3 -1). Several glass chips were handpicked from each sample, avoiding alteration and microphenocrysts. Samples were analyzed for H2O and CO2 concentrations by Fourier transform infrared (FTIR) spectroscopy at the University of Oregon (Johnson et al. 2009) Water concentrations were calculated either using the fundamental OH stretching vibration at 3559 cm1 or from the average of the two m olecular water peaks (1630 cm1 and 5200 cm1) and the 4500 cm1 OHpeak. An absorption coefficient of 63 L/mol cm was used for the 3550 cm1 peak (Dixon et al. 1995b; Dixon et al. 1995a) and absorption coefficients for the near -IR peaks were calculated based on major element concentrations following methods in Mandeville et al., (2002) CO2 co ncentrations were measured using the carbonate peaks at 1515 and 1430 cm1, using background subtraction procedures described in Johnson et al., (2009) and absorption coeffic ients calculated from Dixon & Pan (1995) Volatile concentrations are highly variable in the OSC lavas but are consistent with major element trends (Table 3 1). H2O co ncentrations range from 0.23 to 0.39 wt% in basaltic lavas and from 0.24 to 1.56 in basaltic andesite samples ( Figure 32). Andesites have H2O concentrations ranging from 0.99 to 1.50 wt% and dacites range from 1.53 to 2.35 wt%. CO2 concentrations in bas alts range from 131 to 256 ppm (Figure 32). Two of the basaltic andesites have CO2 concentrations (232 and 184 ppm) but otherwise all others had CO2 concentrations below the detection limits of ~25 ppm. H2O/Ce ratios, involving elements of similar magmatic incompatibility, generally increase with increasing silica and range from 0.014 to 0.024 ( Figure 3 -3), while

PAGE 85

85 H2O/K2O ratios vary from 0.095 to 2.59, with the highest ratios observed in the basaltic andesites ( Figure 3 4). Oxygen Isotope Analyses and Re sults Oxygen isotope ratios ( 18O, per mil notation) were determined on 26 fresh, microphenocrysts-free glass chips that cover the range of rock types (Table 3 1) at the CO2-laser -fluorination laboratory at the University of Wisconsin, Madison, following m ethods described in Valley et al. (1995). Aliquots of 2.4 3.2 mg were treated with BrF5 overnight, and then individually heated with a CO2 laser in the presence of BrF5. Measurements were standardized with 45 analyses of UWG -2 garnet standard per day 18 notation relative to Standard Mean Ocean Water (SMOW). Reproducibility of the standard during each session was better than 0.15 (2SD). 18O values range from 5.31 to 6.19 in the OSC lavas with an average of 5.79 ( Figure 3-5). The basalts and basaltic andesites have similar oxygen isotope ratios, ranging from 5.51 to 5.79 and 5.31 to 5.92 respect 18O (5.38 to 6. 19 ), with a mean of 5.86 Discussion Magma C rystallization versus Assimilation -Fractionation-Crystallization P rocesses Liquid lines of descent (LLDs) were calculated using the MELTS th ermodynamic modeling program to simulate fractional crystallization at f O2 = QFM and P = 1 kbar (Ghiorso & Sack, 1995) These results suggest that fractional crystallization alone cannot account for the high Cl and H2O concentrations observed in the MOR dacites, a ndesites or basaltic andesites ( Figure 3 2). H2O concentrations in the dacites are as

PAGE 86

86 much as two times greater than model predictions. Similarly, Cl concentrations observed in the dacites are more than ten times greater than model predictions ( Figure 3 -2 ). H2O/Ce ratios, which should not change over a wide range of fractional crystallization, are generally higher than expected in the MOR dacites, having almost two times higher ratios than the basalts, whereas basaltic andesites and andesites show a wider range of ratios ( Figure 3 -2a). The high Cl concentrations, however, are similar to those produced by 1 -15% partial melting of altered basalt ic crust (Wanless et al. accepted) While both H2O and Cl exhibit over enrichments compared to calculated fractional crystallization trends (Figure 3 2), the Cl over enrichment is much greater than that of H2O. The difference between these enrichment factors may be due to the higher Cl contents in the altered crustal source and, to a lesser extent, variable amounts of H2O degassing during magma ascent and eruption on the seafloor (see degassing section below). In contrast to H2O, Cl remains soluble in most basaltic and andesitic lavas at eruption depths greater than 700 meters below sea level (Unni & Schilling, 1978; Webster et al. 1999) and therefore, does not degas. Le Roux et al., (2006) used Cl/Nb ratios to assess the role of assimilation in MORB magmas because Cl and Nb have similar partition coefficients in basaltic systems. Cl is enriched in seawater altered crust while Nb remains immobile during seawater alteration, which allows for discrimination between the effects of fractional crystallization and assimilation. However, the advanced f ractional crystallization in the OSC lavas results in precipitation of Fe -Ti oxides in which Nb is a compatible element, so the Cl/Nb ratio has limited applicability in the OSC lavas. Here, we use Cl/K2O

PAGE 87

87 (Figure 3 3) because these elements also have broadly similar incompatibilities over a wide range of crystallization (e.g., Kent et al. 1999) and this ratio has been used in several studies to identify crustal contamination (e.g., Michael & Cornell, 1998; Kent et al. 1999) K2O concentrations are not affected by high temperature alteration in the lower sheeted dikes and upper gabbros (Alt et al., 1996). Many of the MOR dacites have Cl/K2O ratios that are greater than five times the ratios observed in the spatially related basalts ( Figure 3 3). Additionally, many andesites and basaltic andesites e rupted at the OSC have Cl/K2O ratios three to seven times higher than the basalts erupted in the region and higher than mantle values (0.065) predicted by Michael and Cornell (1998), suggesting they have also been contaminated by crustal material. These observations clearly suggest the operation of assimilation in the formation of basaltic andesites andesites and dacites on MOR. Evidence for Assimilation from Oxygen Isotopes Crystallization of Fe-Ti oxides leads to a significant increase in oxygen isotope ratios of evolving magmas (Taylor, 1968) because these phases and to a lesser extent Fe Mg silicate phases preferentially incorporate 16O relative to 18O compared to the remaining melt (Taylor, 1968; Anderson et al. 1971; Muehlenbach & Byerly, 1982) During crystallization of MORB magma there is a sligh 18O as the magma crystallizes ferromagnesian silicates, followed by a dramatic increase when FeTi oxides precipitate (Matsuhisa et al. 1973) Fractional crystallization can result in an increase of 18O values by 1 1.5 (e.g., Muehlenbach & Byerly, 1982) Using modal mineral proportions calculated from MELTS (Ghiorso & Sack, 1995) and oxygen isotope fractionation factors from Bindeman et al., (2008), we have

PAGE 88

88 calcul 18O during fractional crystallization of a MORB parental magma (Figure 35). These calculations suggest that if the MOR dacites formed through 18O should be ~6.8; however, the measured ratios are ~1 lowe r (dacite average = 5.86), supporting the conclusion that the dacites could not form from fractional crystallization alone. Similar observations have been made for dacites erupted at the Galapagos Spreading Center (18O = 3.9 6.2; Perfit et al. 1999) The temperature dependent fractionation of oxygen isotopes between seawater and mineral phases causes decreases in 18O of ocean crust relative to primary mantlelike values during high -temperature hydrothermal alte ration, but increases 18O during low -T alteration (<200250C; Muehlenbach & Clayton, 1972; Alt et al. 1996) Additionally, as seawater migrates through and reacts with the crust at temperatures < 200-250C, the 18O of that water evolves to lower values. This results in an overall decrease in 18O with depth in the crust. Profiles of oxygen isotope ratios in exposed sections of ocean crust (Hess Deep; Agrinier et al. 1995) and drill holes (e.g., ODP Site 504; Alt et al. 1996) show that 18O values are higher than unaltered MORB compositions (~5.6 + 0.2 ; Eiler, 2001) in the upper volcanic section of the crust but decrease to lower than the mantle values in the lower sheeted dikes and gabbros (for profiles see Alt & Teagle, 2000) Therefore, if assimilation occurs at the base of the sheeted dikes or in the upper gabbros, it should lower the oxygen isotope rat ios of the resulting magma. High temperature hydrothermal alteration of the ocean crust results in an observed decrease in the 18O of the sheeted dike and gabbro layers to ~ 4 (Alt et al., 1996)

PAGE 89

89 compared to mantle values of 5.6 (e.g., Alt et al. 1986; Eiler, 2001) Therefore, partial melting of this material will produce melts with oxygen isotope ratios less than mantlelike values, whereas assimilation of this material into fractionating magma will produce magmas with oxygen isotope ratios between altered basalt and that expected 18O value of the MOR andesites and dacites range from 5.38 to 6.19, which is much more variable than t he spatially related basalts and lower than calculated fractional crystallization trends ( Figure 3 5). Taylor (1968) suggested that to a first approximation, the effect of assimilation on oxygen isotope ratios can be determined using mass balance equations. Using results from Energy Constrained -AFC petrologic modeling calculations (Bohrson & Sper a, 2001) the ratio of fractionating magma to assimilant in the dacites is 2:1(Wanless et al. accepted) Assuming the evolved magma has a 18O of 6.8 (largely due to fractionation 18O of 4 (due to high temperature seawater alteration; Alt et al., 1996), the resultant oxygen isotope ratio of the AFC magma would be ~5.9. This valu e is similar to the average of ratios observed in the MOR dacites and is less than predicted by fractional crystallization alone. CO2 and H2O Degassing, Magma Ascent Rates and Depth of Assimilant Although the geochemical data are consistent with crustal assimilation these signatures do not constrain where this contamination occurs. The depth of equilibration of vapor -saturated melts can be calculated using the H2O and CO2 solubility model of Dixon et al., (1995a,b). This model is based on experimental results at pressures and temperatures similar to MOR magmatic conditions. A potential complication of this modeling is that CO2 and H2O may undergo variable degassing during ascent from the magma chamber to the seafloor, which is caused by slow diffusion o f CO2 to the

PAGE 90

90 nearest gas bubble during ascent. However, the rapid ascent rates in MORB magmas and quenching of the lavas at the seafloor will trap the volatiles and allow for very little degassing (e.g., Dixon et al. 1988) Therefore, these model calculations can provide minimum depths of equilibration of the melts prior to eruption. Using this model, we can calculate the equilibrium pressure of a vapor -saturated melt (i.e., magma chamber depth) of a given composition prior to eruption on the seafloor using the VOLATILECALC program (Newman & Lowenstern, 2002) H2O and CO2 concentrations in the OSC lavas suggest a range of equilibration pressures with a maximum of ~550 bars. Most of the basaltic lavas have equilibr ation pressures that approximately equate to the top of the imaged melt lens (~1.5 km or 450 bars; Kent et al. 2000) suggesting little to no degassing during ascent and eruption on the seafloor. In contrast, MOR andesites and dacites have completely degassed CO2, and some may have also lost H2O ( Figure 3 6). The evolved lavas have equilibrium pressures approximately equal to that of the seafloor (~250 bars), suggesting slower ascent rates and higher degrees of degassing prior to erupti on. The vapor saturation calculations support the hypothesis that the depth of contamination on MOR occurs at the top of the axial magma chamber (le Roux et al. 2006) T extural observations in ophiolites show melting and assimilation occurs at the roof of the magma chamber, which may migrate within the ocean crust but often coincide with the base of th e sheeted dikes (Coogan et al. 2002; Gillis & Coogan, 2002; Gillis, 2008) The hydrostatic pressure at the seafloor result s in extensive CO2 degassing from a CO2-saturated basaltic melt, however, quenched lavas supersaturated with CO2 are

PAGE 91

91 common in submarine settings (Dixon et al. 1988; Dixon et al. 1995b; Dixon et al. 1995a; Saal et al. 2002; le Roux et al. 2006) This has been attributed to rapid ascent rates of basal tic magmas, which do not allow for complete vapor exsolution. In contrast, slow ascent rates or pooling of isolated magma batches at shallow depths could lead to magma degassing (Dixon et al. 1988) A key factor that affects the ascent and effusion rate of magma is viscosity. The andesites and dacites erupted on the seafloor have higher viscosities than MORB magmas, which may allow for significant bubble nucleation and growth prior to eruption. Magma degassing during ascent or eruption of the high -silica dacites is supported by large elongate vesicles observed in hand samples and low CO2 contents in the glasses. Source of Assimilant Despite growing evidence of crustal contamination on MORs, the source of the assimilant and depth of the process are poorly understood. Glass compositions suggest that assimilation of brines is responsible for the Cl contamination in some EPR MORB (le Roux et al. 2006) ; however, ass imilation of partially melted, altered crust may also result in anomalous Cl enrichments (e.g., Michael & Schilling, 1989) Partial melting or thermal breakdown of Cl bearing amphibole in altered crust may result in elevated Cl concentrations in the resulting magma (Michael & Schilling, 1989) and has been suggested as a possible source of Cl enrichment in Galapagos Spreading Center andesites and dacites (Perfit et al. 1999) and in dacites from the 9N OSC (Wanless et al. accepted) As mentioned above, Cl/K2O and H2O/K2O ratios provide a means to discriminate between sources of contamination on MOR because of the variable concentrations of these elements in possible assimilants (Kent et al. 1999) Potential contaminants on

PAGE 92

92 MOR include altered basaltic crust (Cl = 0.1 wt%, H2O = 5 wt%, K2O = 1 wt%), seawater (Cl = 1.935 wt%, H2O = 97.5 wt%, K2O = 0.04 wt%); 15% NaCl brine (Cl = 9.9 wt%, H2O = 85 wt%, K2O = 0.25 wt%); and 50% NaCl brine (Cl = 30.3 wt%, H2O = 50 wt%, K2O = 0.25 wt%; see Kent et al., 1999 and references therein). Fluids enriched with as much as 50% NaCl have been observed in melt inclusions in several MOR gabbros (Kelley & Delaney, 1987) These brines form from high temperature phase separation of seawater during hydrothermal circulation (e.g., Berndt & Seyfreid, 1990) and may be trapped along grain boundaries or pore spaces within the altered crust (Michael & Schilling, 1989) Bulk mixing of any of these contaminants with OSC basalt cannot produce the observed compositions of the MOR dac ites ( Figure 3 4). Instead, most dacites have lower H2O/K2O ratios and higher Cl/K2O ratios compared to the possible assimilants. However, if we assume that the dacites formed from AFC processes, then assimilation of low degree partial melts (5-10%) of a ltered basalt into a fractionating MORB magma should produce compositions similar to the OSC dacites. This process can explain the elevated Cl/K2O ratios observed in the MOR dacites. The higher H2O/K2O ratios observed in several MOR dacites ( Figure 3 -4) s uggest that the magmas may have also interacted with small volumes of a saline brine. For instance, mixing of ~0.2 wt% of a 50% NaCl brine can produce H2O/K2O concentrations similar to the MOR dacites and for this small amount of brine, the change in 18O of the magma would be negligible (<<0.01). Oceanic Plagiogranites Plagiogranites veins are a volumetrically small but ubiquitous component of the ocean crust and have been observed in ophiolites (e.g. Pedersen & Malpas, 1984) drill

PAGE 93

93 cores from the ocean crust (e.g. Casey, 1997; Dick et al. 2000) and as x enoliths in Icelandic lavas (Sigurdsson, 1977) There are also many examples of evolved plutonic rocks from slower spreading centers (e.g., Mid -Atlantic Ridge; Aumento, 1969) which may suggest AFC or partial melting processes are occurring on much smaller scales, deeper in the ocean crust. The origin of these veins remains unclear but two main hypotheses are: 1) partial melting of gabbroic crust (e.g. Koepke et al. 2004; Koepke et al. 2007) and 2) extreme crystal fractionation of tholeiitic magmas (Beccaluva et al. 1999, Coleman & Donato, 1979, Niu et al. 2002 ). The composition of oceanic plagiogranites varies considerably (Koepke et al. 2007) making a direct petrologic comparison between the MOR dacites and the silicic veins difficult. The mantle18O values of zircons collected from plagiogranite veins in the gabbroic crust at the mid-Atlantic ridge (~5.2 0.2 ) suggest little to no s eawater contamination in the evolved melt at these depths (Grimes et al. 2010a) but 18O values in plagiogranite veins from the Oman ophiolite (average of 4.6 0.6 ) are thought to represent remelting of altered ocean crust (Grimes et al. 2010b) These studies suggest that high -silica lavas can form in a variety of different MOR settings and may have a range of different compos itions but that melting is an important process in their petrogenesis. Conclusions V ariations in v olatile concentrations and 18O in 9N OSC lavas suggest that these magmas have experienced assimilation during their petrogenesis, with the most extreme signatures observed in high-silica andesites and dacites and little evidence in basaltic lavas. H2O concentrations are up to two times higher in dacitic lavas compared to calculated fractional crystallization trends, whereas Cl has excesses of seven to ten

PAGE 94

94 times predicted values. 18O values are on average ~1 lower than ratios expected from fractional crystallization of ferromagnes ian silicates and Fe -Ti oxide phases, consistent with assimilation of an additional component or components. The source of the excess H2O and Cl and low 18O values is partially melted, hydrothermally altered oceanic crust, but may also include small volum es of saline brines produced during twophase separation of hightemperature hydrothermal fluids. Vapor saturation pressures calculated from H2O -CO2 data suggest that assimilation most likely occurs at the top of the melt lens, which at the 9N OSC, corresponds approximately to the base of the sheeted dikes. Basaltic lavas were supersaturated with CO2 at their eruption depths suggesting fast ascent rates. In contrast, high -silica lavas were completely degassed CO2 and variably degassed H2O prior to quenching on the seafloor. This suggests slower ascent rates and/or lower effusion rates for the high-silica lavas, which is consistent with their higher viscosities and the presence of large elongate vesicles.

PAGE 95

95 Table 3 1. Geochemical d ata Sample SiO 2 T iO 2 Al 2 O 3 FeO MgO CaO Na 2 O K 2 O P 2 O 5 Cl H 2 O CO 2 d 18 O Ce Cl/K 2 O H2O/ K 2 O H2O/ Ce 264 08 50.45 2.70 12.86 14.07 5.69 9.51 3.33 0.21 0.28 0.07 0.39 231 5.79 20.5 0.35 1.82 0.019 265 43 50.12 1.92 13.86 11.57 6.83 11.13 2.81 0.1 3 0.19 0.01 0.25 224 14.2 0.10 1.87 0.018 265 82 50.77 1.95 13.81 11.70 6.93 11.21 2.69 0.14 0.19 0.01 0.24 256 5.51 13.0 0.07 1.74 0.019 265 88 50.85 2.02 13.88 11.99 6.71 11.06 2.85 0.15 0.21 0.02 0.26 131 5.71 15.0 0.13 1.72 0.017 266 51 50.55 1.94 13.85 11.59 7.33 10.97 2.86 0.13 0.19 0.01 0.23 218 13.3 0.09 1.82 0.017 265 104 51.25 1.90 13.79 11.65 6.90 10.59 3.00 0.16 0.22 0.05 0.31 219 16.9 0.29 1.91 0.018 264 18 52.15 1.77 14.04 10.89 6.05 9.94 3.27 0.26 0.20 0.01 0.24 232 5.72 13.4 0.04 0.9 5 0.018 265 106 53.33 1.94 13.26 12.35 5.11 8.71 3.67 0.32 0.39 0.19 0.65 184 5.77 33.9 0.59 2.02 0.019 265 103 56.50 2.01 12.33 13.74 2.74 6.45 3.77 0.52 0.65 0.30 1.24 0 5.73 60.4 0.58 2.38 0.020 265 91 56.83 2.10 12.31 14.23 2.09 6.24 3.45 0.60 0.78 0.31 1.56 0 5.31 66.1 0.52 2.59 0.024 265 125 55.59 1.66 13.63 10.73 4.89 8.30 3.64 0.41 0.20 5.92 32.7 265 90 58.08 1.91 12.43 13.69 1.74 5.76 3.51 0.66 0.74 0.34 0.99 0 73.2 0.51 1.50 0.014 265 100 58.09 1.76 12.64 12.68 1.89 5.51 3.82 0.63 0.5 4 5.56 72.7 266 54 59.65 1.72 13.22 11.25 2.28 5.60 3.91 0.75 0.43 0.42 1.50 0 5.83 67.9 0.56 2.01 0.022 264 14 61.75 1.30 13.46 8.71 2.47 5.54 3.94 0.83 0.22 6.06 57.0 265 69 61.02 1.72 13.62 9.97 1.91 5.38 3.89 0.80 0.27 0.20 1.44 0 55.8 0.25 1.80 0.026 265 63 64.43 1.29 13.26 8.22 1.29 4.21 3.71 0.99 0.21 5.57 68.1 265 64 64.04 1.28 13.12 8.27 1.60 4.45 3.46 0.97 0.20 0.24 1.73 0 6.19 76.5 0.25 1.79 0.023 265 65 63.79 1.26 13.25 8.14 1.34 4.21 3.84 0.97 0.22 5.86 67.5 265 66 62.81 1.43 13.13 9.05 1.99 4.98 3.86 0.89 0.24 0.23 1.81 0 5.84 84.2 0.26 2.04 0.022 265 42 66.92 0.94 13.09 8.04 0.86 3.49 0.83 1.17 0.21 0.51 5.87 82.9 0.44 265 67 64.10 1.34 13.33 8.49 1.49 4.41 3.93 0.95 0.23 5.92 68.0 265 70 66.26 0.87 13.20 7.17 0.80 3.23 4.08 1.33 0.19 0.70 2.35 0 6.08 88.1 0.53 1.76 0.027 265 83 67.46 0.76 13.27 6.68 0.67 2.98 3.88 1.37 0.16 0.67 1.53 0 5.94 87.2 0.49 1.12 0.018 265 84 64.39 1.13 13.17 8.18 1.23 3.92 3.41 1.19 0.22 5.73 77.8 265 85 65.01 1.0 6 13.13 7.99 1.18 3.78 3.67 1.22 0.20 0.64 1.90 0 5.95 82.5 0.52 1.55 0.023 265 94 65.22 0.97 13.04 7.90 1.13 3.54 4.29 1.14 0.23 5.90 83.9 265 95 67.46 0.77 13.10 6.47 0.94 3.01 4.43 1.21 0.15 6.07 83.6 266 53 64.28 1.06 13.31 8.06 1.12 3.7 3 4.16 1.09 0.25 0.66 1.74 0 5.38 82.2 0.60 1.60 0.021 266 57 62.47 1.30 13.16 9.20 1.59 4.37 4.11 0.98 0.29 0.51 2.08 0 5.65 72.4 0.52 2.13 0.029 264 09 65.97 0.90 13.19 7.02 1.05 3.47 4.32 1.20 0.21 0.58 5.73 83.9 0.48

PAGE 96

96 Figure 31. Bathyme tric map of the East Pacific Rise showing the location of the 9N OSC, the Clipperton and Siqueiros transform faults.

PAGE 97

97 Figure 32. H2O (wt%), Cl (wt%), and CO2 (ppm) versus MgO (wt%) for glasses from the 9N OSC. Black crosses indicate the calcula ted fractional crystallization trend using MELTS (Ghiorso and Sack, 1995). All dacitic lavas and several basaltic andesites lie above the calculated trend, indicating another processes is involved in their petrogenesis. A B C D

PAGE 98

98 Figure 33. H2O /Ce and Cl/K2O ratios versus MgO (wt%) for gl asses from the 9N OSC. Generalized trends for assimilation and fractional crystallization are shown as dashed lines. In general, the andesites, dacites and most of the basaltic andesites have higher incompatible element ratios than the basaltic lavas, suggesting that they cannot result from fractional crystallization alone. Instead, they are consistent with assimilation of an altered basalt A B

PAGE 99

99 Figure 34. Cl/K2O versus H2O/K2O for glasses from the 9N OSC. Lines repre senting mixing of 6 possible assimilants with a n OSC basalt are shown. Mixing endmembers include 5 and 10% partial melts of an altered basalt, an altered basalt, a 50% and 15 % NaCl brine, and seawater (see Kent et al., 1999 for references). See text for mixing endmember concentrations. A combination of partial melt ing of an altered basalt and <0.2 wt% of a 50% NaCl brine with a basaltic endmember can explain the formation of high-silica lavas on the OSC.

PAGE 100

100 Figure 35. 18O versus MgO for glasses fro m the 9N OSC. The MOR dacites andesites and several basaltic andesites lie below calculated fractional crystallization trends (black dashed line). The lower 18O values are consistent with assimilation of altered oceanic crust which has lower 18O val ues due to high -temperature hydrothermal circulation 18O values for lavas from the Juan de Fuca ridge and Galapagos Spreading Center are shown for comparison.

PAGE 101

101 Figure 36. H2O versus CO2 for 9N OSC glasses. Superimposed on this diagram are CO2 H2O v apor saturation curves for 200 to 800 bars based on models by Dixon et al. (1995a, b). Black dashed lines show a general magma degassing trends during ascent. The gray band represents an approximate depth of the top of the imaged melt lens (Kent et al., 2000). The pressure at the seafloor is shown as a dotted line. Most of the basalts are in equilibrium with pressures consistent with the top of the imaged melt lens (~550 bars) The dacites, andesites, and two basaltic andesites have completely degassed CO2 and may have also lost H2O prior to or during eruption. H2O and CO2 concentrations in one basaltic andesite lie between the high-silica lavas and the basaltic lavas, which may indicate mixing (red dashed line).

PAGE 102

102 CHAPTER 4 CRUSTAL DIFFERENTIAION AND SO URCE VARIATIONS AT THE 9N OVERLAPPING SPREADING CENTER; EAST PACIFIC RISE Introduction Mid ocean ridges (MOR) can be divided into a series of segments or discontinuities that range in length from tens of meters to hundreds of kilometers (Sempere & Macdonald, 1986; Macdonald et al. 1988) Overlapping spreading centers (OSC) are second order discontinuities that form between widely spaced first order transform faults on fast to intermediate spreading ridges (Macdonald & Fox, 1983; Sempere & Macdonald, 1986; Carbotte & Macdonald, 1992) Both first and second order discontinuities delineate physical and geochemical segmentation of the ridge that reflect sub -ridge processes, such as variations in degrees of mantle melting and/or separate crustal magma reservoirs (e.g. Macdonald e t al. 1988) Lavas erupted along fast to intermediate spreading centers, such as the northern East Pacific Rise (EPR), may produce a range of basaltic lavas (e.g. Batiza & Niu, 1992) but they only rarely erupt compositions with MgO concentrations <5 wt%. This relatively limited compositional diversi ty compared to other tectonic settings is commonly attributed to shallow -level fractional crystallization of primitive magmas within an axial magma chamber that is buffered by relatively frequent recharge with more primitive mantle melts (Klein, 2005) Additionally, geochemical variations in mid ocean ridge basalt (MORB) may result from variable mantle melting parameters and/or mantle sources (Klein & Langmuir, 1987; Langmuir et al. 1992) L avas erupted at ridge segment ends, such as OSC's, can have a broad range of compositions compared to the relatively limited basaltic compositions erupted from magmatically robust segment centers (Christie & Sinton, 1981; Langmuir et al. 1986;

PAGE 103

103 Wanless et al. accepted) This variability can be attributed to lower magma supply an d cooler crust at the end of ridge segments (cold edge effect), which cause increased magmatic fractionation prior to eruption (Christie & Sinton, 1981; Perfit et al. 1983; Sinton et al. 1983; Perfit & Chadwick, 19 98; Rubin & Sinton, 2007) Although crystal fractionation is undoubtedly a primary process in magma differentiation at MOR, recent geochemical studies show that highly evolved incompatible trace element concentrations (Wanless et al. accepted) and low oxygen isotope ratios (Wanless et al. submitted) in MOR dacites require partial melting and assimilation of oceanic crust. The extent to which these processes contribute to the chemistry of more mafic magmas on MOR remains poorly constrained. Lavas erupted at segment ends also preserve geochemical s ignatures that can be ascribed to mantle source variations. A greater proportion of enriched mid ocean ridge basalt (E -MORB) erupted at ridge segment ends compared to segment centers on the northern East Pacific Rise (EPR) may represent decrease in the am ount of melt feeding the ridge axis in these regions (Christie & Sinton, 1981; Sinton et al. 1983; Langmuir et al. 1986) Despite evidence for variations in mantle source between segments, the spatial distributio n of E MORB lavas at OSC is not well constrained. The present study combines geophysical, geochemical, and bathymetric data to examine the magmatic plumbing system at 9N OSC on the EPR (Figure 4-1). We use major and trace element data and isotopic rati os to explore the relative roles of crystal fractionation, assimilation and magma mixing within the shallow crust beneath the 9N OSC. Trace element data and isotopic ratios are used to examine how variations in mantle sources and magma supply can contribute to the distribution of compositions

PAGE 104

104 observed on both limbs of the OSC. These analyses are compared to compositions of lavas erupted from segment centers to the north (9 -10N) and south (837N) of the OSC to explore the extent of these variations beneath the EPR. Background, Tectonic Setting and Geology of the 9oN OSC Overlapping Spreading Centers At OSC s the ridge axis splits into two overlapping, curvilinear, sub-parallel axes or limbs that may offset the ridge by up to 15 km (Macdonald & Fox, 1983; Macdonald et al. 1988) The ratio of offset width to overlap length is approximately 1:3 (Macdonald et al. 1988) and the inward curving limbs surround an elongate basin (Macdonald & Fox, 1983) The limbs migrate sub -parallel to the overall ridge strike with one limb propagating and the other dying (Hey et al. 1980; Sinton et al. 1983; Pollard & Sydin, 1984) Consequently, the propagating limb migrates into older and colder ocean crust and the receding limb gradually becomes amagmatic. As the OSC migrates with time, it produces offsets in bathymetry and magnetic signature of the ocean crust (Hey et al. 1977; Carbotte & Macdonald, 1992) Average OSC migration rates can be calculated using the fossil V -shaped bathymetric scars developed in the wake of the propagating OSC that are left either by linking of one ridge axis with the other during propagation, which leads to decapitation of the ridge tip, or by repeated self -decapitation along a single limb of the OSC (Macdonald et al. 1988) Tectonic Setting and Previous Studies of 9N OSC The 9N OSC is located between the Clipperton and Siqueiros transform faults (Figure 41) and is one of eight 2nd order discontinuities on the northern EPR (Macdonald & Fox, 1983) It consists of two north -south trending ridges that overlap by ~27 km and offset the ridge by ~8 km (Sempere & Macdonald, 1986) The 9N OSC is

PAGE 105

105 divided into three main sections: the eastern propagating limb, the western receding limb, and an overlap basin separating the two limbs (Figure 42). The eastern limb can be further divided into the east limb ridge, the east limb tip and the northern inter limb region also called the northwest flank (Figure 42), fo llowing nomenclature in Nunnery et al., (2008). Based on geologic, magnetic and bathymetric data, the eastern limb has been propagating south at a rate of ~42 km/Myr, the western limb has receded (Macdonald & Fox, 1988; Carbotte & Macdonald, 1992) The 9N OSC is one of the largest and most extensively studied 2nd order discontinuities on the global MOR system. It has been the focus of several geophysical studies (Detrick et al. 1987; Harding et al. 1993; Kent et al. 1993; Kent et al. 2000; Bazin et al. 2001; Dunn et al. 2001; Tong et al. 2002) including the first multi -channel seismic 3 -D survey of a MOR (Kent et al. 2000) and a 3-D seismic refraction study (Dunn et al. 2001) Collectively, these studies reveal the presence o f a shallow melt lens beneath each of the limbs and a widening of the eastern lens below the inter limb region north of the overlap basin (Figure 4 3; Kent et al. 2000) The melt lens beneath the western, receding limb is narrow and shows no di scernable variation in depth along axis, but the melt lens beneath the eastern, propagating limb is variable in width and depth (Kent et al. 2000) Beneath the southern portion of the east limb the melt lens is narrower and deeper than the rest of the eastern ridge axis, pl unging ~500 m over ~6 km (Kent et al. 2000) It also cuts a cross the tectonic seafloor fabric (White et al. 2009) To the north, the melt lens widens to more than 4 km (Figure 43) and is displaced slightly off axis to the west into the inter -limb region (Kent et al. 2000; Tong et al. 2002) Tomographic studies reveal a low velocity zone beneath the entire OSC at ~

PAGE 106

106 9 km depth near the mantle-crust transition (Toomey et al. 2007) This zone extends ~8 km to either side of the ridge axis (Dunn et al. 2001) Recently, the 9N OSC was the foc us of the MEDUSA2007 research cruise (AT15 -17), which carried out detailed mapping and extensive sampling of the region using the ROV Jason2 DSL -120A side -scan system, and the WHOI TowCam (Fornari, 2003) This expedition acquired >10,000 photogr aphs of the seafloor in combination with the most complete and well -constrained lava sampling of any OSC (White et al. 2009; Wanless et al. accepted) Prior to this study, limited lava sampling of the region durin g the CHEPR cruise indicated that both high-silica and E MORB lavas existed in addition to N MORB lavas (Langmuir et al. 1986) 9N OSC Geology The northern portion of the eastern ridge is characterized by an axial summit trough (AST) ~0.9 km across and ~50 m high (Figure 4-4). The AST walls gradually diminish in height to the south, eventually disappearing by ~905 N. This topographic change is accompanied by a gradual shift from a volcanically dominated seafloor fabric in the north to a highly faulted and tectonized fabric observed at the southern tip that also correlates with narr owing of the imaged melt lens, from ~4 km wide to <1 km (Figure 43) and a near absence of high -silica lavas (Figure 42). Video, still photographs, and side scan collected during the 2007 cruise suggest that over 80% of the lavas erupted within the mapp ed region are pillow lavas (White et al. 2009) a morphology that is observed across all regions of the OSC but dominates the seafloor in the inter limb region and overl ap basin. In comparison, >80% of lavas erupted on the EPR north of the OSC are lobate and sheet flows, and <20% are pillows (Kurras et al. 2000; White et al. 2002; Soule et al. 2005; Fundis et al. 2010) Pillow

PAGE 107

107 flow morphology at the OSC appears to be controlled by low effusion rates (<0.1 m3/s at a viscosity of 100 Pa s; White et al., 2009). The youngest lavas on the northern portion of the eastern limb, based on glassy surfaces, lack of sediment cover, and well developed ornamentation, are confined to the neovolcanic zone (Nunnery et al. 2008) a narrow region of focused magmatism where zero age lavas erupt (Perfit & Chadwick, 1998) On the southern tip of the eastern limb, the neovolcanic zone is ill -defined; however, the youngest lavas are primarily l ocated along the western margin of the eastern limb (Nunnery et al. 2008) We now describe the geology of each region of the OSC in detail. Despite the overwhelming abundance of pillow lavas formed at the OSC, lavas erupted on the east limb ridge are morphologically quite diverse (sheet, hackly and lobate flows, and small and large pillow lavas) and appear to correlate, at least to a first degree, with composition. High-silica lavas erupted within the AST form a ~10 m high, linear, pillow mound composed of atypically large pillow lavas. This pil low mound is surrounded primarily by lobate lavas and to a lesser degree, sheet flows. Several large areas of collapse, with drainback features, were observed within the AST and are associated with basaltic sheet flows that surround the high-silica pillow mounds. The lobate and sheet lavas are predominantly ferrobasaltic in composition; however, several FeTi basalts were also recovered within the AST. Basaltic andesites also erupted within the AST and form lobate flows and pillow mounds. Lavas sampled fr om both sides of the AST walls at the edges of the ridge axis are primarily andesites and dacites and form large elongate to bulbous pillows. The only active hydrothermal vent at the OSC was observed within the AST on the east limb axis within an andesit ic to dacitic pillow

PAGE 108

108 mound. A faulted and tectonized fabric dominates the seafloor on the southern tip of the eastern limb. In places it is covered by elongate fresh pillow mounds that are cut by fabric -parallel fissures. Lobate flows are sparse in this r egion. Lavas erupted over the northern portion of the overlap basin and inter -limb region primarily consist of pillow lavas. A linear pillow mound, trending approximately N -S, defines the outer edge of this region and lies over the westernmost extents of the wide melt lens (Kent et al. 2000) Lavas comprising this mound are younger than expected (based on thin sediment cover and the presence of glassy buds) for their distance from the neovolcanic zone, suggesting that this region is the site of off axis volcanism (Nunnery et al. 2008; White et al. 2009) This conclusion is supported by excess 230Th measured in several of these samples, which indicate eruption ages of <8,000 ka (Waters, pers comm.). The southern overlap basin is primarily composed of bulbous pillow mound fields, based on backscatter images (White et al. 2009) but photographic and sample coverage is sparse. The western limb of th e OSC differs from the eastern limb in bei ng primarily comprised of sheet and lobate flows, with fewer pillow lavas (White et al. 2009) and extensive areas of lava collapse and pillars with drain back features. Similar features are common within ASTs on many other regions of the northern EPR. Extinct vent fields were also observed, but no active hydrothermal venting. 837 N EPR Deval The 837N deviation in axial linearity (deval) is located south of the 9N OSC and approximately 40 km north of the Siqueiros transform fault (Figure 4 1; Langmuir, 1986). It is the southern extension of the western limb of the 9N OSC. The 837N deval is structurally similar to a small OSC and, therefore, may have a similar magmatic

PAGE 109

109 plumbing system. This deval was dredged during the CHEPR cruise, which recovered several E MORB lavas and a high-silica andesite (Langmuir et al. 1986) Additional investigations of the area were conducted during a response cruise in 2003 (Zierenberg, 2007 pers comm.) to determine if sei smic events recorded on hydrophones in March of 2001 were indicators of new volcanic activity (Bohnenstiehl et al. 2003) This cruise included several ALVIN dives; however, there was no evidence of a recent eruption. Lava morphology in the region includes sheet flows, lobates and pillow lavas. During th ese dives nine samples were recovered, including several glassy E -MORBS and one glassy andesite. Data from these samples are discussed below. Geochemical Methods Over 280 rock samples were collected from the 9N OSC during the MEDUSA2007 cruise. Of the se, 275 have glassy outer rims, from which glass was handpicked and analyzed on a JOEL 8900 electron microprobe for major and minor element concentrations at the USGS facility in Denver, CO. Eight to ten points were analyzed per sample. The probe diamete loss, with an accelerating voltage of 15 keV and a beam current of 20nA. Several USGS minerals were used as calibration standards and secondary normalizations involved the JdF -D2 glass standard (Reynolds, 1995) ALV 2392-9 (inhouse standard), and dacite glass GSC (USGS standard) to account for instrument al drift (Smith et al. 2001) Chlorine and sulfur concentrations, as well as highprecision potassium values were also determined on nine samples, using 200 second peak/100 second background counting times. Major element concentrations for samples from the 837N deval on the EPR were determined by microprobe at UC Davis following methods described in Schiffman et al. (2010)

PAGE 110

110 A representative subset of fresh glasses were handpicked, cleaned in a dilute acid, and dissolved for tra ce element and isotope analyses following methods described in Goss et al ., (2010) 73 samples from the 9N OSC and seven samples from the 837 N deval were analyzed for high precision trace elements on an Element2 Inductively Coupled Plasma Mass Spectrometer (ICP MS) at the University of F lorida. External calibration was done to quantify results using a combination of internal (ENDV Endeavour and ALV 2392-9) and USGS (AGV -1, BIR -1, BHVO 1, BCR -2 and STM 1) rock standards. High precision Pb, Sr, and Nd isotopic abundances on 37 samples were determined using the NuPlasma multi -collector ICP MS at the University of Florida (Wanless et al. accepted) For detailed descriptions of sample preparation, dissolution procedures, standards, and statistical data, see Goss et al ., (2010) and Kamenov et al ., (2007) Geochemical Results Lavas erupted at the 9N OSC display a large range of compositions on both the east and west limb (Figure 4-2). Below, we discuss the geochemical variability on each limb by rock type to better address the petrogenesis of the OSC lavas. Geochemistry of East Limb Lavas The east limb of the 9N OSC has produced basalts, ferrobasalts, FeTi basalts, basaltic andesites, low -P2O5 and high-P2O5 andesites, an d dacites (Figures 4-2, 45). These lavas cover a wide compositional range and we, therefore, subdivide our results below by rock type. M ajor and trace element data from the east limb are present in Table 4 1 and 4 -2, respectively. Radiogenic isotope rati os for the east limb lavas are presented in Table 2-2.

PAGE 111

111 Basalts Basalts consists of all lavas with <52 wt% SiO2 and include ferrobasalts and FeTi basalts. In comparison to more mafic basalts from the heavily sampled 950 bulls eye site on the northern EPR basalts from the east limb of the OSC are more evolved on average (Table 4 1), with MgO concentrations ranging from 6.16 to 7.36 wt%, high FeO (10.83 to 13.50 wt%), and TiO2 (1.70 to 2.83 wt%) contents (Figures 4-5, 46). All lavas are N -type MORB with variable K2O (0.11 to 0.27 wt%), P2O5 (0.12 to 0.27 wt%) and Cl concentrations (0.004 to 0.05 wt%). P2O5/TiO2 ratios range from 0.08 to 0.13 (Figure 47) and Cl/K2O ratios from 0.03 to 0.29 and are thus comparable to other MOR ferrobasalts (e.g. Michael & Cornell, 1998) Incompatible trace element concentrations are relatively high compared to MORB from the northern EPR but ratios are relatively constant (Figures 48, 49) despite eruption over wide geographic region at the OSC and are comparable to ot her ferrobasalts erupted on MOR. Rare earth elements (REE) patterns are remarkably similar in the ferrobasalts, with an average LaN/YbN ratio of 0.77 0.05 (Figure 49). High field strength element (HFSE) ratios are also relatively limited with Zr/Nb of 40 2.8, and U/Nb of 0.03 0.001. Compatible trace element concentrations in the ferrobasalts are variable (Cr = 9 to 166 ppm; Ni = 32 to 71 ppm) but show positive correlations with MgO. Basaltic andesites/low P2O5 a ndesites Basaltic andesites (SiO2 >5 2 and <57 wt%SiO2) and low -P2O5 andesites from the east limb have highly variable major and trace element compositions (Table 4 1). MgO in the basaltic andesites ranges from 1.5 to 6.5 wt% and FeO ranges from 8.27 to 13.64 wt% (Figure 4-5). Minor element s are also highly variable (Figure 46; P2O5 = 0.18 to 0.58 wt%; K2O = 0.19 to 0.99 wt%; and Cl = 0.04 to 0.5 wt%). P2O5/TiO2 (average of

PAGE 112

112 0.15 0.04) and P2O5/K2O (average of 0.79 0.28) ratios span the range between the high-silica and ferrobasalt endmembers (Figure 47). Ni and Cr concentrations ranging from 6 to 50 ppm and 5 to 117 ppm, respectively (Figure 4-8) are generally, though not exclusively, lower than in the ferrobasalts. REE and HFSE concentrations are variable, but have relatively constant ratios with average LaN/YbN ratios = 1.08 0.12, Zr/Nb = 47 8.3, and U/Nb ratios 9). The basaltic andesites lie along well defined trends toward high-silica lavas. High P2O5 a ndesites Low MgO and high FeO, TiO2, P2O5, and K2O charact erize the highP2O5 andesites and distinguish them from low -P2O5 andesites discussed above. P2O5 ranges from 0.54 to 0.78 wt%, which is up to 0.49 wt% greater than the average low -P2O5 andesite (Figure 4-6). Consequently, these lavas have higher P2O5/TiO2 (0.29 to 0.39) and P2O5/K2O (0.86 to 1.47) ratios compared to other lavas on the east limb (Figure 47). They have high average Cl concentrations of 0.3 ppm (Figure 4 6) compared to MOR basalts. Most incompatible trace element concentrations are high i n these lavas compared to ferrobasalts and basaltic andesites but similar to the low -P2O5 andesites (Figure 48). The highP2O5 andesites have an average Zr ~ 700 ppm, La ~22 ppm, and Yb ~15 ppm with LaN/YbN ratios of 1.15 0.11, and Zr/Nb ratios of 47 3.5, (Figure 49). Distinctive features are the lack of negative Nb and Ta anomalies and slightly lower U/Nb ratios (<0.03) when compared to the low -P2O5 andesites and some basaltic andesites. Dacites The geochemistry of the MOR dacites is discussed i n detail in Wanless et al., [accepted] and are therefore, only briefly discussed here. They have SiO2

PAGE 113

113 concentrations up to 67.5 wt% and MgO as low as 0.67 wt%. P2O5 concentrations range from 0.15 to 0.25 wt% (Figure 4-6). K2O and Cl concentrations avera ge 1.17 and 0.63 wt% respectively. P2O5/TiO2 and P2O5/K2O are also low (Figure 47). Highly incompatible trace element concentrations are elevated, with average Ba, U, and La concentrations of 62 ppm, 0.8 ppm and 28 ppm respectively (Figure 4-8). They also have high Zr (734 ppm to 1050 ppm) and Hf (18.6 ppm to 24.9 ppm) concentrations, but low Nb (13 ppm to 16.5 ppm), Ta (0.8 ppm to 1.4 ppm) and U/Nb ratios >0.03 (Figure 4 9). The dacites have average LaN/YbN ratios of 1.26 0.10, Zr/Nb = 58 4.5, and U/N b ratios Isotopic compositions of East Limb l avas In contrast to the variability noted in the major and trace element compositions of east limb, the radiogenic isotopic ratios are relatively uniform. They have 208Pb/204Pb and 206Pb/204Pb ranging fr om 37.642 to 37.699 and 18.235 to 18.294 respectively (Figure 410a). 87Sr/86Sr ranges from 0.70243 to 0.70258 and 143Nd/144Nd ratios range from 0.51314 to 0.513120 (Figure 410b), which is similar to lavas erupted on the EPR to the north near 950N (Sims et al. 2002; Sims et al. 2003; Wanless et al. accepted) West Limb Lavas Fifty lava samples (Table 4 3, 44) were collected from the west limb of the 9N OSC of which 47 are basaltic, 2 are basaltic andesites and one is andesitic in composition (Figure 4 -11a). Fourty -five samples have typical N MORB K/Ti ratios (K/Ti = K2O/TiO2*100 <1.3; Langmuir et al. 1986; Sinton et al. 1991; Reynolds et al. 1992; Perfit et al. 199 4) and five have higher ratios typical of E -MORB (K/Ti>0.13; Figure 411b). Compared to the east limb basalts, the average west limb basalt is slightly more primitive, with MgO concentrations ranging from 6.42 to 8.72 wt% (Figure 4-11a) and

PAGE 114

11 4 K2O and P2O5 concentrations from 0.07 to 0.17 wt% and 0.13 to 0.26 wt%, respectively (Figure 411a). Lavas dredged in this region during the CHEPR cruise had similar compositional variations, including E and N -MORB (Langmuir et al. 1986) West limb N -MORB lavas have LaN/YbN ratios of 0.72 0.04 and Zr/Nb of 37 to 45 (Figure 4-12). Highly incomp atible elements and HFSE have limited ranges in concentration, for instance, Ba ranges from 3.4 ppm to 9.2 ppm, U from 0.04 ppm to 0.07 ppm, and Nb from 1.55 ppm to 3.22 ppm (Figure 4-12). The west limb E -MORB lava has LaN/YbN ratios of 2.8 and Zr/Nb ratio s of 9.7. The single andesite has major element concentrations (Figure 411a) and trace element abundances (Figure 4-13) similar to the high -P2O5 andesites on the east limb (Figure 4-11a). It has a SiO2 concentration of 55.95 wt% and MgO of 2.84 wt%, with high P2O5 (0.73 wt%) and K2O (0.67 wt%). Isotopically, the west limb lavas have slightly more radiogenic values compared to the east limb lavas (Figure 4-10; Table 4 -5). The N MORB lavas have 208Pb/204Pb and 206Pb/204Pb ratios ranging from 37.737 to 37. 846 and 18.253 to 18.369, while the E MORB lava has higher lead isotope values (208Pb/204Pb = 38.022 and 206Pb/204Pb = 18.590) (Figure 4-10a). N MORB 87Sr/86Sr values ranges from 0.70249 to 0.70265 with an average of 0.70256 (Figure 410b) whereas the E MO RB 87Sr/86Sr ratio is also more radiogenic (0.70282). 143Nd/144Nd ratios of the N -MORB lava range from 0.51314 to 0.51315, while the E MORB lava has a ratio of 0.51305. The andesite has isotopic ratios intermediate between the west limb E MORB and N MORB values. 837 EPR Lavas Nine samples collected from the 837 N deval range in composition from basalt to andesite, with MgO and SiO2 concentrations from 7.29 to 3.08 wt % and 50.47 to 57.14 wt%, respectively (Table 4 6). P2O5 concentrations range from 0.3 to 0.5 wt.%. Two of

PAGE 115

115 the samples are E -MORB with K/Ti values of 0.19 and 0.15. REE and HFSE abundances are variable with average LaN/YbN ratios of 1.33 0.17, Zr/Nb of 20.92 to 31.84, and U/Nb of 0.02 to 0.04. Radiogenic isotope ratios were measured in seven samples collected from 837N (Table 47). Compared to basalts collected from the 950N region of the EPR (Sims et al. 2002; Sims et al. 2003) and the 9N OSC, these samples have, on average, more radiogenic Pb (Figure 4 -10a) and 87Sr/86Sr ratios (0.70255 to 0.70267) and less radiogenic 143Nd/144Nd ratios (0.51312 to 0.51318; Figure 4-10b). Consistent with their elevated incompatible element ratios, the samples have Sr, Nd and Pb isotopes that plot between th e field typical EPR N MORB and more radiogenic E MORB (Figure 410). Discussion Shallow -level Processes Involved in the Petrogenesis of Ferrobasalts, FeTi Basalts and Basaltic A ndesites at the 9N OSC Shallow level differentiation of MOR magma is primari ly controlled by fractional crystallization and mixing of different MORB melts (Clague & Bunch, 1976; Bryan & Moore, 1977; Byerly, 1980) These processes are often difficult to discern from each other on geochemical variation diagrams because mixing can produce compositions that lie along liquid lines of descent resulting from fractional crystallization. However, our geochemical results and petrographic studies suggest that both mixing and fractional crystallization are important at the OSC, particularly in the generation of basaltic andesites. The geochemistry of most of the basaltic lavas erupted at the 9N OSC, including, ferrobasalts, FeTi basalts, and many basaltic andesites, can primarily be explained

PAGE 116

116 through low pressure fractional crystallization of typical N MORB parental magmas (Figure 45). In general, the compositions of the 9N lavas are slightly more evolved than the average MORB erupted along the northern EPR (Per fit et al. 1994; Goss et al. 2010) Based on MELTS calculations (Ghiorso & Sack, 1995) ferrobasalts and FeTi basalts can be generated by ~30% and ~55% fractional crystallization (respectively) of ol + plag + cpx at pressures of 1 kbar from a relatively prim itive MORB magma (23929 from the 1991 EPR eruption at 9o50N; Figure 4-6). These results are consistent with previous studies of EPR lavas covering large geographic areas that collectively suggest fractional crystallization is the dominant or exclusive process effecting the composition of MOR magmas (e.g. Batiza & Niu, 1992) In contrast, many of the basaltic andesites erupted at the 9N OSC lie off of typical fractional crystallization trends. OSC lavas are predominantly aphyric, but some lavas contain sparse phenocrysts and microphenocrysts of olivin e, clinopyroxene and plagioclase that are consistent with shallow level fractional crystallization of a typical MORB (Grove et al. 1992) Petrographic studies, however, suggest some lavas have had more complex petrogenetic histories (Zaino, 2009) Although some plagioclase phenocrysts in east limb ferrobasalts and basaltic andesites exhibit normal zoning from core to rim, others show reverse zoning or no zoning at all. Commonly, all three zoning patterns occur in a single sample. This suggests that each phenocryst has undergone a distinct magmatic history prior eruption on the seafloor and supports the role of mixing in the petrogenesis of these lavas. Resorbed ri ms on olivine phenocrysts in east limb basalts coupled with more magnesian compositions that should be in equilibrium with co existing glasses are also indicative of magma mixing (Zaino, 2009)

PAGE 117

117 Many of the low MgO OSC lavas have elemental concentrations that lie off of calculated liquid lines of descent and are i nconsistent with crystal fractionation alone. This is particularly apparent in the concentrations of FeO, TiO2, and P2O5 in many basaltic andesites (Figure 4-6). These elements are typically enriched in MORB systems during fractional crystallization unti l melts become saturated with Feoxides (~5 wt % MgO) and apatite (~1 wt% MgO). However, many of the evolved OSC lavas have relatively low concentrations of these elements and appear to lie along straight lines that suggest mixing between high -silica and ferrobasaltic magmas. Similar linear trends are observed in incompatible and compatible trace element concentrations (i.e. Cr and Ni) versus Zr and MgO (not shown). Similarly, trace element ratios (e.g. U/Nb) of basaltic andesites do not plot along fract ional crystallization trends but lie along mixing lines between evolved basalts and high-silica lavas (Figures 4 -8, 4 9). The high -silica mixing end-member is likely produced during partial melting of the oceanic crust or assimilation and fractional cryst allization (AFC) processes that can produce MOR dacites (Wanless et al., accepted). Using an OSC dacite composition as an endmember, our mass balance calculations confirm that many of the basaltic andesites can be produced by bulk mixing of 25% dacitic m elt with 75% ferrobasaltic melt. Similar processes can explain the low -P2O5 andesites (see discussion below). This scenario is consistent with the petrogenetic models of dacite formation on OSC, which requires an episodic magma supply at the propagating li mb of the OSC (Wanless et al., accepted). In this case, a basaltic magma is injected into the OSC and undergoes variable degrees of crystal fractionation to produce ferro and FeTi basalt compositions. These magmas mix to varying degrees with preexistin g high-silica melts

PAGE 118

118 beneath the ridge axis to produce a range of compositions, including basaltic andesites and andesites (Figure 45). Fractional c rystallization versus a ssimilation in the p etrogenesis of a ndesites Although fractional crystallization and magma mixing appear to be the dominant processes involved in the differentiation of magmas at the 9N OSC, incompatible trace element concentrations suggest that the formation of highly evolved lavas (i.e., dacites) on ridges requires partial melting and assimilation of altered ocean crust (Wanless et al. accepted) Here, we assess the roles of fractional crystallization, assimilation and magma mixing in the formation of intermediate compositions (andesites) on MOR. There are two distinct geochemical populations of andesites at the OSC, which likely require different petrogenetic histories (Figure 45). Both populations of andesites (high-P2O5 and low -P2O5 andesites) have similar SiO2 and MgO concentrations but different TiO2, FeO, Al2O3, P2O5, and trace element concentrations at a given MgO (Figures 4 5, 46, and 4-8). Petrologic modeling of major element variations indicates that high-P2O5 andesite magmas formed primarily through extensive crystal fractionation of basaltic magmas (Figure 4 6). MELTS calculations (Ghiorso and Sack, 1995) suggest t hat this process involved up to 75% fractional crystallization of ol + plag + cpx and minor amounts of Fe oxide of a ferrobasaltic parent (26543) at 1kbar pressure. Oxygen fugacity at the QFM 1 buffer is required to delay the onset of Fe oxide crystalliz ation, and produce the elevated FeO and TiO2 concentrations observed (Figure 45). Elevated P2O5 concentrations require that apatite crystallization has not occurred during the evolution of the melt (Figure 46). This is consistent with MELTS calculations and is supported by apatite saturation calculations for these compositions (Watson, 1979) which suggest the high liquidus

PAGE 119

119 temperature (>1000C) of these melt compositions inhibits apatite saturation. Trace element ratios also suggest extensive fractional c rystallization is required to explain incompatible trace element concentrations and ratios (e.g. low U/Nb ratios). Rayleigh fractionation calculations (Figure 4-8) suggest that ~85% fractional crystallization is required to produce the high-P2O5 incompatib le trace element abundances, compared to the 75% calculated using MELTS for the major elements (Figure 4 6). In contrast, low -P2O5 andesites have low FeO, TiO2, and P2O5 concentrations and high U/Nb ratios (Figures 46, 49). The low FeO and TiO2 and hig h U/Th ratios are a result of crystal fractionation of Fe oxides and the low P2O5 is a results of apatite crystallization. These lavas are spatially related to high-silica dacites erupted onaxis and are likely formed from mixing of high-silica melts and ferrobasaltic magmas. The high-silica melts are a result of AFC processes that include crystallization of Feoxides and apatite (Wanless et al. accepted) This is similar to the petrogenesis of the basaltic andesites but in opposite proportions. Bulk mixing calculations suggest mixes of up to 25% ferrobasalt and as low as 75% dacitic melt will produce compositions similar to the low -P2O5 andesites. Formation of w est l imb l avas by fractional c rystallization The west limb of the OSC has been receding at a rate approximately equal to the propagation of the east limb (Carbotte & Macdonald, 1992) This must reflect a progressive decreas e in magma supply on the western limb tip and, therefore, a decrease in the heat supply and magma recharge, which could lead to greater degrees of crystal fractionation (e.g. Christie & Sinton, 1981) Under these circumstances, one might expect the eruption of a range of compositions, including high silica lavas.

PAGE 120

120 However, the lavas erupted on the west limb encompass a narrower compositional range than that observed on the eastern, propagating limb. With the exception of a single andesitic lava, lavas erupted on the west limb can be explained by up to ~25% fractional crystallization of either an E or N MORB parent with ~8.5 wt% MgO (Figure 4 -11). Trace element patterns are consistent with fractional crystallization as the primary shallow -level differentiation process (Figure 412); however, some complex zoning patterns in plagioclase phenocrysts (Zaino, 2009) are consistent with a combination of crystallization and magma mixing. The single west limb andesite recovered appears to have formed through extensive fractional crystallization. Petrologic modeling suggest that ~75% fractional crystallization c an explain the major and trace element compositions observed in the west limb andesite. When compared to the wide range of east limb andesites, the west limb andesite has trace element patterns similar to the high-P2O5 andesites, which are geochemically dominated by fractional crystallization (Figure 413). The west limb andesite lacks negative Nb and Ta anomalies observed in the low -P2O5 andesites, has a small negative K anomaly, and elevated phosphorus contents and P2O5/TiO2 ratios (Figures 4 11, 414). Assimilation does not appear to play a significant role in the formation of the west limb andesite, which may be due to cooler crustal conditions on the dying limb. Where Are the South Tip and West Limb Dacites? A near absence of high-silica lavas appear s to be a characteristic of both the southern tip of the eastern limb and the western limb as a whole (Figure 42), in contrast to the abundant dacites erupted over the northern east limb. This lack of high-silica samples is puzzling because the tips of b oth the eastern and western limbs should

PAGE 121

121 experience low magma supply rates, which would promote greater degrees of fractional crystallization and more evolved magma compositions (Christie & Sinton, 1981) We argue that the subtle difference in the thermal conditions in the oceanic crust play a major role in controlling the composition of erupted lavas. Large amounts of heat, in part from the latent heat of crystallization and in part from replenishment by an episodic magma supply, in the volcanically active region of the propagating eastern limb enhances melting and assimilation of ocean crust while allowing for extensive fractional crystallization (Wanless et al., accepted). In contrast, low magma recharge rates and cooler conditions in the tectonically controlled southern east limb tip and dying west limb may res ult in extensive fractional crystallization but melting and assimilation are inhibited. The presence of an unusually large and extensive melt lens below the northern portion of the eastern ridge axis is consistent with higher crustal temperatures and greater magmatic activity that results in partial melting of the oceanic crust. Additionally, the northern portion of the east limb may experience higher rates of magma recharge than the starved eastern tip and the dying western limb. This leads to the form ation of high-silica lavas through AFC processes (Wanless et al. accepted) In contrast, the narrower, deeper melt lenses in the southern tip of the east limb and dying west limb (Kent et al. 2000) provide less heat to the system, resulting in cooler crust, which may inhibit melting, assimilation and the formation of d acitic lavas. These observations are supported by compositions of the two andesite lavas erupted on the southern tip and the west limb (Figure 413), which have incompatible trace element patterns similar to andesites produced primarily by fractional crys tallization with little evidence of assimilation. Another possibility is that

PAGE 122

122 melting and assimilation may occur in these regions, but that the melts are volumetrically too small to erupt, perhaps resulting in the formation of small plagiogranite intrusions and veins within the crust rather than erupted lavas. Composition of the Melt Lens Beneath the East Limb Seismic evidence indicates that the shallow crustal melt lens widens to more than 4 km beneath the northern portion of the east limb (Figure 43) but it is centered beneath the inter limb region and not below the neo volcanic zone (Kent et al. 2000; Tong et al. 2002) Tomographic studies reveal a low velocity zone beneath the entire OSC at ~ 9 km depth, near the mantle -crust transition (Toomey et al. 2007) Despite this evidence for a regionally robust magmatic system the re is scant evidence for the eruption of primitive lavas anywhere within the OSC but enigmatically, the lavas erupted have evolved compositions. This suggests that even in areas with large, imaged melt bodies, magmas that originate within the shallow mantle have complicated and extended differentiation histories in the crust and are unlikely to erupt in pristine condition It is difficult to relate lavas erupted within the inter limb region to the currently imaged melt lens, however, they do appear to be younger than expected, suggesting that they did not erupt onaxis (Nunnery et al. 2008) Additionally, the observation that lavas erupted within the inter limb region (directly above the melt lens) are almost exclusively composed of ferrobasalts (60 of 67 samples) with very uniform compositions (Figures 4 6, 4-8) suggests that these reflect the relatively evolved nature of the shallow melt lens. The slight variability in major and trace element compositions of the inter -limb ferrobasalts can be modeled by 20% to 30% fractional crystallization o f a MORB

PAGE 123

123 parental composition (MgO = 7 wt%). Many basalts erupted within the neovolcanic zone on the eastern limb are also ferrobasaltic and have incompatible trace element concentrations that are similar to inter -limb ferrobasalts, despite the proximity of highsilica lavas. Assuming that the these lavas are directly related to the current melt lens, this suggests that the 4 km wide melt lens is primarily composed of ferrobasaltic magma and that beneath the east limb neovolcanic zone it has mixed with hi gh-silica melts, creating a wide range of compositions on the eastern edge of the melt lens. Presumably a large volume of primitive mantle-derived basalts must have been fractionally crystallized at crustal and possibly upper mantle depths to result in a large, evolved melt lens. Crystallization of a primitive magma to form a 4 km wide melt lens with ferrobasaltic composition would release significant amounts of heat to the surrounding region. It is the latent heat of crystallization that provides the extra heat to cause partial melting and assimilation of the crust leading to the formation of high-silica magmas along the eastern limb of the OSC (Wanless et al. accepted) E-MORB Distribution at 9N OSC Both E MORB and N MORB lavas were recovered on the western limb of the OSC in this study as wel l as during the CHEPR cruise [ Langmuir et al. 1986] In contrast only N -MORB lavas have been recovered from the east limb. Incompatible element enrichment corresponds with more radiogenic Sr and Pb isotopes and less radiogenic Nd isotopes in the west limb E -MORB lavas (Figure 410) The sub -ridge upper mantle is generally thought to b e relatively incompatible element -depleted and isotopically homogeneous; however, there are well -documented cases of small -scale (veined) heterogeneities in the MOR mantle source. Proximal and roughly coeval eruptions of E MORB and N MORB at MORs suggest that the sub -ridge

PAGE 124

124 mantle varies in elemental and isotopic composition over small spatial/temporal scales [e.g. Hart et al. 1973; Sun et al. 1975; White and Schilling 1978] Formation of EMORB magma is often i nterpreted as a result of overall lower degrees of mantle melting of a source comprised of a greater compl iment of incompatible enriched material compared to N -MORB magma. More recently, 2 stage models involving normal amounts of melting of enriched comp onents embedded in a largely depleted subridge mantle have had success in explaining b oth elemental and isotopic systematics in E type MORB [ Don n elly et al. 2004] The enriched component is thought to be volumetrically smaller than normal mantle and may consist of enriched veins in a depleted mantle [e.g. Hanson 1977] While E MORB may be an important component in ridge magmas, it may be diluted or overwhelmed by the N MORB signature during more robust magmatic activity Consequently, eruption of E -MORB compositions have been linked to diminished magmatic activity along ridge segments [ Reynolds and Langmuir 1997 Waters et al. in press ] The eruption of E MORB lavas only on the western limb of t he OSC suggests a magmatic plumbing system that either allows for the preservation of enriched melts from the mantle on one limb compared to the other or that the mantle sources supplying each limb are different. The nonsymmetrical distribution of parental MORB types may result from differences in magma supply to the two limbs. As spreading shifts from one axis to the other at an OSC the magma supply from the mantle will diminish on the dying limb and progressively increase on the propagating limb. This allows for the preservation of enriched melts on the dying limb in contrast to the magmatically robust propagating limb

PAGE 125

125 9N OSC as a D ivision in M antle C omponents N -MORB lavas from the western limb of the OSC and at 837N have measurably different isotope ratios compared to N MORB lavas erupted on the east limb (Figure 4 10). The east and west limb N MORB lavas have similar incompatible trace element ratios (i.e. LaN/YbN) but west limb N MORB lavas have higher 208Pb/204Pb and 207Pb/204Pb ratios (Figur e 10a) and lower Nd (Figure 10b) compared to east limb N MORB. These compositions cannot be explained by simple 2 component mixing of an enriched source with a typical N -MORB component from the northern EPR ( F igure 4 10). These isotopic differences suggest that different mantle sources are feeding the two limbs and that the OSC acts as a division between these sources. The west limb E MORB lavas have isotopic compositions similar to E -MORB lavas erupted at the intersection of the EPR with the Siqueiros Transform fault to the south (~ 8N). The 837 lavas, which lie geographically between the Siqueiros Transform and the 9N OSC, also have more radiogenic Sr and Pb than the east limb lavas ( F igure 4 10). This suggests that the mantle source below the EPR from the dying west limb to the Siqueiros Transform fault is generally more enriched than the mantle beneath the east limb of the OSC extending up to the Clipperton Transform [ Sims et al. 2002; Sims et al. 2003] This is consistent with observations that the leading limb of the EPR may tap a slightly more enriched mantle than the trailing limb [ Carbotte et al. 2004] While it appears to be more enriched overall, simple 2 component mixing of depleted mantle with an enriched end-member similar to the Siqueiros E MORB cannot explain the range of isotopic ratios erupted on this segment of the EPR.

PAGE 126

126 Conclusions The 9N OSC is one of the most extensively studied 2nd order discontinuities on the global MOR system and one with the most completely imaged crustal melt lens. It has erupted a wide compositions ranging from basalts to dacites. Much of the compositional variability can be ascribed to low -pressure (~1 kbar) fractional crystallization of N -MORB magma or mixing of ferrobasaltic and highsilica magmas. The west limb magmas are slightly less evolved than those on the east limb. Andesitic compositions on the east limb can be divided into two different groups (highand low P2O5) with different major and trace element characteristics. The highP2O5 andesites are produce d dominantly by extensive fractional crystallization (~75%). Low -P2O5 andesites are produced through extensive mixing of high-silica dacitic and ferrobasaltic magma. Magma mixing also explains the compositions of many of the basaltic andesites erupted on the east limb that are not consistent with calculated MORB fractional crystallization trends. The near absence of high-silica lavas on the southern tip of the east limb and the entire west limb compared to the northern east ridge axis suggests a different tectono magmatic environment in these settings. We believe that this is due to the cooler temperatures of the ocean crust in these regions as a consequence of decreased magmatic input. Dacitic lavas are produced from the combination of assimilation and f ractional crystallization in regions where the latent heat of crystallization provides enough heat to partially melt the surrounding wall rock. The cooler crust at the dying western limb and the southern ridge tip may allow for extensive fractional crystallization, however, it is either not enough to heat and melt the surrounding wall rock or these partial (anatectic?) melts are not erupted.

PAGE 127

127 The distribution of evolved lavas and E -MORB lavas across the OSC is not symmetric, suggesting that the 2nd order discontinuity represents a boundary?? division? in the magmatic plumbing system of the EPR. E MORB lavas are only observed on the dying western limb and overall the lavas are less evolved. We suggest that the lower magma supply at the west limb allows for the preservation and eruption of E-MORB compositions, whereas the more robust magmatic system on the propagating east limb overwhelms this signature. N -MORB lavas on the west limb have more radiogenic Pb and Sr and less radiogenic Nd compared to east li mb N MORB lavas. Lavas erupted to the south of the OSC (837N) also have more radiogenic Pb and Sr isotope ratios. This suggests a slightly different mantle is feeding this section of the EPR and that the OSC provides a fundamental division between mant le sources beneath the ridge axis.

PAGE 128

128 Table 4 1. East limb major element data 9N OSC sample Rock Type SiO2 TiO2 Al2O3 FeO Mn MgO CaO Na2O K2O P2O5 Cl S Total WC 07 FeTi 49.28 2.60 12.57 14.08 0.24 5.71 9.24 3.21 0.20 0.28 97.42 WC 06 FeTi 49.73 2.49 12.69 14.02 0.24 5.67 9.35 3.28 0.19 0.27 97.94 WC 08 ferrobasalt 49.99 1.75 13.80 11.11 0.21 7.27 10.95 2.90 0.15 0.16 98.29 WC 01 ferrobasalt 50.07 1.83 13.69 11.36 0.21 7.20 10.94 2.94 0.14 0.18 98.57 265 43 ferrobasalt 50.53 1.92 13.88 11.56 0.21 6.98 11.14 2.86 0.13 0.19 0.01 0.25 99.67 WC 02 ferrobasalt 50.13 1.80 13.73 11.31 0.21 7.26 11.04 2.94 0.13 0.20 98.75 265 98 ferrobasalt 50.24 1.94 13.82 11.75 0.21 7.17 10.86 2.86 0.12 0.19 99.16 266 14 ferrobasalt 50.45 2.14 13.48 12 .38 0.22 6.69 10.75 3.11 0.16 0.22 99.60 264 08 FeTi 50.10 2.68 12.73 14.06 0.26 5.69 9.58 3.27 0.21 0.28 0.07 0.32 99.24 266 15 ferrobasalt 50.51 2.38 13.09 13.30 0.23 6.27 10.11 3.27 0.18 0.26 99.61 266 51 ferrobasalt 50.55 1.94 13.85 11.59 0.21 7.33 10.97 2.86 0.13 0.19 0.01 99.62 267 14 ferrobasalt 50.55 1.92 14.04 11.57 0.21 6.96 11.05 2.86 0.13 0.21 99.51 265 20 ferrobasalt 50.68 1.85 13.92 11.31 0.22 7.21 11.34 2.95 0.13 0.18 0.01 0.22 100.0 266 48 ferrobasalt 50.57 1.92 14.09 11.56 0. 21 7.34 11.00 3.03 0.13 0.19 100.0 266 38 FeTi 50.58 2.00 13.60 12.35 0.23 6.63 10.73 3.15 0.17 0.20 99.63 267 13 ferrobasalt 50.59 1.85 14.08 11.34 0.21 7.11 11.21 2.82 0.12 0.21 99.54 267 03 FeTi 50.60 2.13 13.58 12.49 0.23 6.45 10.15 3.05 0.16 0.26 99.10 264 23 ferrobasalt 50.61 1.87 13.89 11.29 0.22 7.28 11.14 2.92 0.13 0.18 99.52 266 16 ferrobasalt 50.61 2.02 13.67 12.10 0.23 6.87 10.75 3.15 0.16 0.22 99.79 265 44 ferrobasalt 50.62 1.93 14.02 11.72 0.22 6.86 11.11 2.81 0.13 0.19 99. 62 266 40 ferrobasalt 50.63 1.77 13.96 11.52 0.20 7.24 11.25 2.96 0.14 0.16 99.84 266 17 ferrobasalt 50.63 1.87 14.00 11.31 0.20 7.24 11.43 3.01 0.14 0.19 100.0 265 99 ferrobasalt 50.65 1.84 13.91 11.42 0.20 7.36 11.00 2.90 0.13 0.18 99.57 266 34 ferrobasalt 50.65 1.84 14.10 11.30 0.21 7.25 11.08 2.90 0.11 0.18 99.63 265 87 ferrobasalt 50.65 1.98 13.84 11.94 0.23 6.84 11.06 2.77 0.14 0.20 99.64 266 32 ferrobasalt 50.67 1.85 14.01 11.38 0.22 7.27 11.13 2.92 0.12 0.18 99.74 265 96 ferrobasa lt 50.67 1.83 14.00 11.43 0.19 7.35 11.01 2.85 0.13 0.18 99.65 267 05 FeTi 50.68 1.99 13.69 12.00 0.22 6.39 10.31 3.06 0.16 0.25 98.76 264 05 ferrobasalt 50.69 1.83 13.79 11.42 0.22 7.07 11.14 2.95 0.15 0.18 99.44 267 12 FeTi 50.70 2.20 13.58 12.7 5 0.22 6.38 10.40 3.03 0.16 0.25 99.67 266 39 FeTi 50.71 2.04 13.62 12.42 0.24 6.57 10.54 3.21 0.19 0.20 99.74 267 08 FeTi 50.71 1.99 13.80 12.03 0.21 6.79 10.77 2.95 0.14 0.22 99.62 265 19 ferrobasalt 50.62 1.83 14.02 11.23 0.21 7.19 11.24 2.93 0 .12 0.18 0.01 0.24 99.83 266 33 ferrobasalt 50.73 1.84 14.00 11.35 0.22 7.29 11.13 2.88 0.12 0.20 99.75 266 13 ferrobasalt 50.74 2.29 13.21 12.95 0.23 6.34 10.54 3.21 0.17 0.25 99.93

PAGE 129

129 Table 4 1. Continued sample Rock Type SiO2 TiO2 Al2O3 FeO Mn MgO CaO Na2O K2O P2O5 Cl S Total 266 52 ferrobasalt 50.74 1.93 13.92 11.63 0.21 7.30 10.95 2.89 0.13 0.18 99.87 265 14 ferrobasalt 50.75 1.83 13.92 11.20 0.21 7.16 11.19 2.97 0.12 0.18 99.54 WC 05 ferrobasalt 50.75 2.00 13.44 11.53 0.21 6.11 9. 69 3.35 0.26 0.27 97.62 266 02 FeTi 50.76 2.08 13.63 12.45 0.22 6.68 10.51 3.11 0.12 0.21 99.77 266 08 ferrobasalt 50.76 1.98 13.88 11.74 0.22 7.12 10.92 2.86 0.13 0.20 99.81 265 86 ferrobasalt 50.76 1.99 13.83 11.87 0.22 6.79 11.08 2.72 0.14 0.20 99.59 267 09 ferrobasalt 50.76 1.95 13.90 11.99 0.22 6.90 10.82 2.95 0.14 0.24 99.87 265 97 ferrobasalt 50.77 2.02 13.78 12.02 0.22 7.03 10.87 2.61 0.13 0.21 99.65 265 82 ferrobasalt 50.77 1.95 13.81 11.70 0.22 6.93 11.21 2.69 0.14 0.19 0.01 99. 60 265 92 ferrobasalt 50.77 1.98 13.91 11.77 0.22 7.08 10.99 2.93 0.13 0.20 99.97 265 10 ferrobasalt 50.78 1.86 13.94 11.33 0.22 7.13 11.23 2.99 0.13 0.18 99.78 265 52 ferrobasalt 50.78 1.93 14.01 11.77 0.22 6.99 11.18 2.92 0.13 0.20 100.1 265 15 ferrobasalt 50.79 1.86 13.90 11.25 0.21 7.14 11.24 2.99 0.13 0.17 99.67 267 11 FeTi 50.79 2.20 13.66 12.69 0.22 6.44 10.32 2.95 0.16 0.25 99.68 266 19 ferrobasalt 50.79 1.86 13.73 11.87 0.23 6.88 11.03 3.09 0.14 0.20 99.83 266 12 ferrobasalt 50.8 0 2.15 13.43 12.67 0.24 6.51 10.66 3.27 0.18 0.22 100.1 265 68 ferrobasalt 50.81 1.81 14.02 11.17 0.21 7.29 11.51 2.79 0.13 0.17 0.01 99.92 265 53 ferrobasalt 50.81 1.96 14.05 11.83 0.22 6.85 11.17 2.92 0.13 0.19 100.1 266 18 ferrobasalt 50.82 2.01 13.70 12.08 0.22 6.91 10.88 3.14 0.16 0.22 100.1 266 37 FeTi 50.82 2.03 13.57 12.26 0.22 6.73 10.58 3.02 0.15 0.20 99.59 265 62 ferrobasalt 50.82 1.95 13.93 11.72 0.22 6.92 11.19 2.90 0.13 0.19 99.97 266 11 ferrobasalt 50.82 2.23 13.30 12.93 0.24 6.43 10.34 3.26 0.18 0.27 100.0 267 07 ferrobasalt 50.82 1.96 13.94 11.84 0.20 6.96 10.86 2.94 0.14 0.22 99.89 265 13 ferrobasalt 50.83 1.85 14.01 11.22 0.22 7.21 11.28 2.93 0.12 0.18 99.83 265 11 ferrobasalt 50.84 1.80 14.00 11.11 0.21 7.23 11.2 9 2.95 0.12 0.17 99.72 267 01 FeTi 50.84 2.03 13.65 12.39 0.21 6.34 10.18 3.07 0.16 0.23 99.11 265 45 ferrobasalt 50.84 1.94 14.12 11.80 0.21 6.87 11.21 2.89 0.13 0.19 100.2 265 88 FeTi 50.85 2.02 13.88 11.99 0.22 6.71 11.06 2.85 0.15 0.21 0.02 9 9.94 266 09 ferrobasalt 50.85 1.93 14.00 11.43 0.21 7.10 11.08 2.84 0.12 0.19 99.76 265 17 FeTi 50.68 2.09 13.45 12.61 0.23 6.48 10.75 3.14 0.13 0.20 0.00 0.27 100.0 265 119 ferrobasalt 50.86 2.00 13.88 11.77 0.22 7.12 10.86 2.90 0.13 0.19 99.94 26 5 73 ferrobasalt 50.86 1.96 13.98 11.79 0.22 6.97 11.20 2.83 0.13 0.20 100.1 267 02 ferrobasalt 50.87 1.91 14.10 11.45 0.20 6.73 10.73 2.71 0.15 0.24 99.09 265 16 ferrobasalt 50.87 1.84 14.13 11.23 0.22 7.22 11.22 3.01 0.12 0.17 100.0 265 04 ferro basalt 50.87 1.84 14.01 11.21 0.21 7.15 11.38 3.00 0.13 0.18 99.97

PAGE 130

130 Table 4 1. Continued sample Rock Type SiO2 TiO2 Al2O3 FeO Mn MgO CaO Na2O K2O P2O5 Cl S Total 267 06 ferrobasalt 50.88 1.97 14.02 11.91 0.22 6.91 10.84 2.84 0.14 0.22 99.94 2 66 26 FeTi 50.88 2.13 13.57 12.82 0.22 6.34 10.30 3.06 0.15 0.23 99.69 265 79 ferrobasalt 50.89 1.89 13.99 11.75 0.22 6.92 11.20 2.86 0.14 0.19 100.0 266 04 FeTi 50.89 2.36 13.25 13.41 0.23 6.18 9.94 3.13 0.17 0.23 99.79 265 72 ferrobasalt 50.89 1.93 14.05 11.76 0.23 6.98 11.13 2.85 0.13 0.20 100.1 266 22 FeTi 50.90 1.92 13.87 12.05 0.21 6.79 10.73 2.97 0.13 0.21 99.79 265 39 ferrobasalt 50.90 1.95 13.80 11.57 0.21 7.13 11.19 2.95 0.13 0.18 100.0 264 06 ferrobasalt 50.90 1.83 13.75 11.37 0.22 7.06 11.14 2.96 0.15 0.17 99.56 265 46 ferrobasalt 50.91 1.94 14.09 11.88 0.22 6.80 11.20 2.99 0.14 0.20 100.4 266 23 FeTi 50.91 1.89 13.89 12.00 0.20 6.80 10.80 2.97 0.13 0.21 99.80 266 24 FeTi 50.91 1.90 13.87 12.04 0.22 6.81 10.70 2.98 0. 13 0.22 99.79 265 08 ferrobasalt 50.92 1.86 13.94 11.25 0.21 7.21 11.37 2.96 0.12 0.16 100.0 265 18 ferrobasalt 50.75 1.82 14.01 11.31 0.21 7.21 11.26 2.97 0.12 0.18 0.01 0.22 100.1 265 09 ferrobasalt 50.93 1.85 13.96 11.28 0.22 7.20 11.28 2.97 0.13 0.18 99.99 266 27 FeTi 50.94 2.13 13.59 12.83 0.23 6.32 10.19 3.11 0.14 0.25 99.74 265 89 ferrobasalt 50.94 1.88 14.01 11.57 0.22 6.95 11.31 2.76 0.14 0.18 99.94 266 30 ferrobasalt 50.95 1.88 13.78 11.90 0.21 7.02 10.73 2.97 0.12 0.19 99.76 26 5 41 ferrobasalt 50.96 1.92 14.14 11.47 0.22 7.10 11.31 2.84 0.13 0.19 100.3 264 04 ferrobasalt 50.54 1.84 13.66 11.53 0.21 6.97 11.06 2.98 0.15 0.18 0.01 0.25 99.39 266 28 FeTi 50.97 2.15 13.52 12.86 0.23 6.34 10.27 3.07 0.15 0.25 99.80 266 35 FeTi 50.97 1.91 13.74 12.13 0.22 6.91 10.59 3.00 0.12 0.20 99.80 266 07 FeTi 50.97 2.36 13.15 13.50 0.25 6.16 9.72 3.18 0.18 0.25 99.71 267 10 ferrobasalt 50.97 1.97 13.99 11.97 0.21 6.92 10.81 2.95 0.14 0.23 100.2 265 05 ferrobasalt 50.98 1.88 13.8 8 11.38 0.22 6.99 11.34 3.02 0.13 0.18 99.99 265 78 ferrobasalt 50.98 1.88 14.01 11.70 0.21 6.97 11.18 2.86 0.14 0.19 100.1 264 07 ferrobasalt 50.98 1.82 13.84 11.40 0.21 7.08 11.19 2.98 0.14 0.18 99.80 267 15 ferrobasalt 50.98 2.00 13.80 11.81 0. 21 6.97 10.84 2.86 0.14 0.21 99.82 266 29 FeTi 50.99 2.19 13.32 12.88 0.23 6.39 10.15 3.07 0.15 0.24 99.61 266 01 FeTi 50.99 2.00 13.51 12.39 0.23 6.76 10.71 3.04 0.12 0.17 99.91 265 71 ferrobasalt 51.00 1.89 13.96 11.68 0.22 6.96 11.19 2.82 0.14 0.19 100.0 265 80 ferrobasalt 51.01 1.92 13.95 11.70 0.22 6.98 11.24 2.87 0.13 0.19 100.2 265 74 ferrobasalt 51.01 1.99 13.93 11.92 0.22 6.81 11.05 2.90 0.14 0.20 100.2 265 02 ferrobasalt 51.04 1.89 13.98 11.39 0.22 6.99 11.37 3.03 0.12 0.18 100 .2 266 36 ferrobasalt 51.05 1.89 13.83 11.88 0.22 7.04 10.66 2.97 0.13 0.20 99.87 266 21 ferrobasalt 51.06 1.85 13.95 11.93 0.22 6.94 10.80 2.96 0.13 0.19 100.0

PAGE 131

131 Table 4 1. Continued sample Rock Type SiO2 TiO2 Al2O3 FeO Mn MgO CaO Na2O K2O P2 O5 Cl S Total 265 81 FeTi 51.06 2.00 14.13 11.93 0.22 7.00 11.24 2.39 0.13 0.19 100.3 265 22 ferrobasalt 51.06 1.86 14.05 11.35 0.22 7.15 11.37 3.01 0.12 0.18 100.4 266 25 FeTi 51.06 1.98 13.68 12.26 0.22 6.57 10.61 2.98 0.14 0.22 99.72 266 41 fe rrobasalt 51.06 1.86 14.09 11.30 0.22 6.84 10.74 3.22 0.21 0.18 99.71 264 03 ferrobasalt 51.07 1.84 13.89 11.34 0.21 7.07 11.19 2.98 0.15 0.17 99.90 266 10 ferrobasalt 51.07 1.93 13.95 11.50 0.21 6.97 10.97 2.89 0.14 0.21 99.84 266 03 ferrobasalt 51.07 1.77 14.24 10.83 0.21 7.04 10.86 3.06 0.16 0.21 99.44 265 03 ferrobasalt 51.08 1.89 13.89 11.40 0.22 7.02 11.27 3.03 0.13 0.19 100.1 265 76 ferrobasalt 51.08 1.89 13.98 11.61 0.22 6.98 11.25 2.84 0.14 0.19 100.2 265 06 ferrobasalt 51.09 1.85 13.89 11.44 0.21 7.00 11.32 3.00 0.13 0.17 100.1 266 31 ferrobasalt 51.10 1.88 13.78 11.90 0.22 7.09 10.72 2.96 0.13 0.18 99.95 265 07 ferrobasalt 51.11 1.90 13.98 11.47 0.22 6.96 11.24 3.02 0.13 0.19 100.2 265 38 ferrobasalt 51.13 1.89 13.91 11. 34 0.22 7.09 11.11 3.00 0.15 0.19 100.0 265 30 ferrobasalt 51.13 1.87 14.10 11.45 0.22 6.91 11.32 3.01 0.15 0.18 100.3 265 29 ferrobasalt 51.17 1.88 14.04 11.44 0.22 6.88 11.27 3.04 0.15 0.18 100.2 265 121 ferrobasalt 51.17 1.89 14.05 11.44 0.21 7 .12 10.93 2.91 0.14 0.19 100.1 265 26 ferrobasalt 51.17 1.94 13.73 12.03 0.22 6.84 11.09 3.08 0.16 0.19 100.5 265 51 ferrobasalt 51.17 1.92 14.19 11.66 0.22 7.11 11.27 2.93 0.13 0.19 100.8 265 21 ferrobasalt 51.18 1.83 14.04 11.33 0.21 7.20 11.37 2.98 0.13 0.17 100.4 265 36 ferrobasalt 51.18 1.93 13.90 11.42 0.22 7.10 11.12 2.96 0.14 0.19 100.2 265 35 ferrobasalt 51.21 1.94 13.78 11.57 0.22 6.95 11.14 2.99 0.15 0.20 100.2 265 28 ferrobasalt 51.23 1.79 14.12 11.19 0.21 7.18 11.37 2.99 0.15 0.18 100.4 265 33 ferrobasalt 51.23 1.97 13.93 11.48 0.22 6.83 11.16 3.02 0.15 0.18 100.2 265 37 ferrobasalt 51.23 1.93 13.95 11.36 0.22 7.11 11.15 3.00 0.14 0.19 100.3 265 104 ferrobasalt 51.25 1.90 13.79 11.65 0.22 6.90 10.59 3.00 0.16 0.22 0.05 99.69 265 27 ferrobasalt 51.27 1.79 14.12 11.13 0.21 7.10 11.47 2.99 0.14 0.17 100.4 265 34 ferrobasalt 51.30 1.99 13.93 11.52 0.22 6.92 11.11 3.01 0.15 0.19 100.3 265 107 ferrobasalt 51.30 1.92 13.83 11.67 0.21 6.85 10.51 3.07 0.16 0.23 99.76 265 93 ferrobasalt 51.32 1.97 13.90 11.82 0.22 6.84 10.62 3.02 0.15 0.21 100.1 265 23 ferrobasalt 51.32 1.88 13.78 11.92 0.22 6.92 10.98 3.02 0.13 0.19 100.4 265 01 FeTi 51.36 2.03 13.82 11.89 0.23 6.50 10.71 3.21 0.17 0.23 100.2 265 32 ferrobasal t 51.36 1.96 13.89 11.53 0.22 6.90 11.11 3.00 0.15 0.18 100.3 265 115 FeTi 51.38 2.17 13.34 12.72 0.23 6.16 9.77 3.24 0.19 0.25 99.45 265 12 ferrobasalt 51.41 1.92 13.75 11.45 0.22 6.60 10.69 3.11 0.19 0.20 99.54 265 31 FeTi 51.62 2.01 13.89 11. 72 0.23 6.85 11.07 3.05 0.15 0.20 100.8

PAGE 132

132 Table 4 1. Continued sample Rock Type SiO2 TiO2 Al2O3 FeO Mn MgO CaO Na2O K2O P2O5 Cl S Total 265 110 ferrobasalt 51.65 1.90 13.81 11.69 0.20 6.71 10.41 3.07 0.17 0.23 99.84 265 114 FeTi 51.66 2.18 13.39 12.83 0.24 5.88 9.53 3.29 0.22 0.27 99.48 265 111 FeTi 51.78 2.17 13.38 12.65 0.22 5.73 9.40 3.35 0.23 0.30 99.21 265 112 FeTi 51.79 1.97 13.67 12.00 0.22 6.41 10.11 3.12 0.19 0.24 99.72 266 06 ferrobasalt 51.80 1.70 14.21 10.87 0.20 6.72 10.59 3.05 0.18 0.14 99.47 265 113 FeTi 51.92 2.17 13.40 12.82 0.23 5.93 9.48 3.28 0.22 0.28 99.72 WC 04 FeTi 51.94 2.00 12.96 12.63 0.25 4.88 8.59 3.57 0.33 0.44 97.59 265 61 ferrobasalt 51.96 1.81 14.06 11.26 0.21 6.43 10.62 3.05 0.19 0.18 99.76 264 18 basaltic andesite 52.15 1.77 14.04 10.89 0.20 6.05 9.94 3.27 0.26 0.20 98.77 266 45 basaltic andesite 52.24 1.79 13.86 11.28 0.22 6.46 10.18 3.33 0.27 0.17 99.80 264 11 basaltic andesite 52.36 1.72 13.79 10.75 0.20 6.55 10.47 3.13 0.24 0.19 99.40 264 17 basaltic andesite 52.61 1.77 14.12 10.95 0.20 6.23 10.07 3.27 0.25 0.20 99.66 265 105 basaltic andesite 52.77 1.90 13.33 12.06 0.22 5.58 9.26 3.44 0.27 0.32 99.13 265 116 basaltic andesite 52.86 1.84 1 3.77 11.64 0.21 5.73 9.38 3.37 0.26 0.28 99.33 265 49 basaltic andesite 52.81 1.78 14.00 11.09 0.21 5.98 10.03 3.23 0.27 0.20 0.04 0.21 99.84 265 60 basaltic andesite 53.08 1.74 14.06 10.93 0.21 5.93 10.12 3.26 0.27 0.19 99.80 264 21 basaltic ande site 53.15 1.82 14.00 11.09 0.21 5.82 9.69 3.41 0.29 0.22 99.71 265 122 basaltic andesite 53.28 2.06 13.40 12.27 0.23 5.36 8.84 3.52 0.30 0.23 99.50 265 106 basaltic andesite 53.33 1.94 13.26 12.35 0.24 5.11 8.71 3.67 0.32 0.39 0.19 99.31 264 13 basaltic andesite 53.41 1.68 14.44 10.43 0.20 5.76 9.53 3.42 0.33 0.20 99.40 265 118 basaltic andesite 53.42 1.86 13.57 11.90 0.21 5.16 8.85 3.54 0.31 0.35 99.16

PAGE 133

133 Table 4 1. Continued sample Rock Type SiO2 TiO2 Al2O3 FeO Mn MgO CaO Na2 O K2O P2O5 Cl S Total 264 19 basaltic andesite 53.51 1.77 13.90 10.83 0.21 5.72 9.52 3.44 0.31 0.20 99.41 265 48 basaltic andesite 53.85 2.36 13.06 13.33 0.24 4.44 8.36 3.57 0.35 0.26 0.05 0.28 100.2 264 22 basaltic andesite 53.58 1.75 13.93 10.90 0. 20 5.80 9.62 3.45 0.31 0.20 99.75 265 108 basaltic andesite 53.77 1.95 13.28 12.60 0.23 5.08 8.80 3.63 0.32 0.38 0.20 100.0 265 24 basaltic andesite 54.07 2.35 12.90 13.64 0.25 3.71 7.61 3.75 0.50 0.58 99.36 265 50 basaltic andesite 54.15 2.14 13.4 1 12.19 0.22 4.54 8.45 3.59 0.40 0.28 0.09 0.22 99.67 266 43 basaltic andesite 54.14 1.90 13.56 12.05 0.24 4.42 8.16 4.03 0.41 0.32 99.22 265 59 basaltic andesite 54.24 2.15 13.42 12.27 0.23 4.61 8.52 3.53 0.40 0.28 99.65 265 58 basaltic andesite 54.25 2.12 13.56 12.13 0.22 4.56 8.52 3.57 0.39 0.27 99.59 265 120 basaltic andesite 54.33 2.06 12.78 13.10 0.24 4.22 7.87 3.72 0.38 0.52 99.23 266 62 basaltic andesite 54.48 1.53 14.24 9.87 0.18 5.33 8.89 3.46 0.41 0.19 98.58 265 55 basaltic ande site 54.55 2.10 13.45 12.19 0.22 4.51 8.38 3.62 0.41 0.28 99.71 264 20 basaltic andesite 54.59 1.71 13.95 10.68 0.21 5.10 8.84 3.60 0.39 0.21 99.28 265 123 basaltic andesite 54.62 1.91 13.46 11.67 0.22 4.93 8.40 3.61 0.36 0.22 99.37 265 56 basalti c andesite 54.98 2.13 13.68 12.13 0.23 4.42 8.36 3.63 0.42 0.28 100.2 265 124 basaltic andesite 55.06 1.69 13.55 10.78 0.19 5.03 8.44 3.54 0.40 0.20 98.88 265 54 basaltic andesite 55.14 2.04 13.65 11.75 0.22 4.20 8.11 3.64 0.45 0.27 99.47 264 10 b asaltic andesite 55.00 1.62 13.34 10.23 0.19 5.27 8.76 3.56 0.43 0.21 0.12 0.20 98.93

PAGE 134

134 Table 4 1. Continued sample Rock Type SiO2 TiO2 Al2O3 FeO Mn MgO CaO Na2O K2O P2O5 Cl S Total 265 102 basaltic andesite 55.31 2.22 12.51 14.05 0.27 3.34 7.14 3.8 5 0.44 0.65 99.79 WC 03 basaltic andesite 55.46 2.02 12.02 13.80 0.27 2.50 6.33 4.10 0.56 0.74 97.80 265 126 basaltic andesite 55.48 1.66 13.97 10.46 0.18 4.78 8.37 3.67 0.41 0.19 99.18 264 12 basaltic andesite 55.51 1.64 13.35 10.18 0.20 4.98 8.4 7 3.65 0.47 0.21 98.66 265 125 basaltic andesite 55.59 1.66 13.63 10.73 0.19 4.89 8.30 3.64 0.41 0.20 99.22 265 109 basaltic andesite 55.66 1.81 12.83 12.31 0.23 3.90 7.47 3.76 0.45 0.43 98.85 265 77 basaltic andesite 55.74 2.13 13.52 11.98 0.21 3 .95 7.80 3.61 0.48 0.31 99.73 265 117 basaltic andesite 55.77 1.81 12.90 12.26 0.23 3.90 7.45 3.75 0.43 0.44 98.95 265 101 basaltic andesite 55.90 2.14 12.29 13.95 0.26 3.02 6.76 4.03 0.49 0.65 99.50 264 16 basaltic andesite 56.02 2.06 13.21 12.23 0.25 4.37 8.10 2.40 0.45 0.37 99.46 265 57 basaltic andesite 56.29 2.12 12.59 12.75 0.25 3.69 7.46 3.59 0.52 0.40 99.64 265 103 basaltic andesite 56.50 2.01 12.33 13.74 0.26 2.74 6.45 3.77 0.52 0.65 0.30 98.96 265 91 basaltic andesite 56.83 2.10 1 2.31 14.23 0.28 2.09 6.24 3.45 0.60 0.78 0.31 98.91 266 61 andesite 57.47 1.70 14.07 10.35 0.19 3.13 6.57 3.76 0.63 0.24 98.13 266 59 andesite 57.57 1.58 13.16 10.41 0.20 3.66 6.71 3.60 0.61 0.23 97.74 265 90 andesite 58.08 1.91 12.43 13.69 0.27 1. 74 5.76 3.51 0.66 0.74 0.34 98.80 265 100 andesite 58.09 1.76 12.64 12.68 0.24 1.89 5.51 3.82 0.63 0.54 97.80 266 05 andesite 58.40 1.86 13.18 11.14 0.21 3.04 6.29 3.77 0.66 0.30 98.85 266 55 andesite 59.52 1.65 13.14 10.83 0.20 2.13 5.30 3.86 0.77 0.42 97.82 266 54 andesite 59.65 1.72 13.22 11.25 0.21 2.28 5.60 3.91 0.75 0.43 0.42 99.02 265 69 andesite 61.02 1.72 13.62 9.97 0.18 1.91 5.38 3.89 0.80 0.27 0.20 98.76 265 25 andesite 61.17 1.48 13.63 9.84 0.19 1.92 5.03 4.28 0.99 0.43 98.9 6 264 14 andesite 61.75 1.30 13.46 8.71 0.17 2.47 5.54 3.94 0.83 0.22 98.39

PAGE 135

135 Table 4 1. Continued sample Rock Type SiO2 TiO2 Al2O3 FeO Mn MgO CaO Na2O K2O P2O5 Cl S Total 266 49 andesite 62.34 1.27 13.15 8.64 0.15 2.19 4.86 4.18 0.91 0.21 97 .90 266 57 andesite 62.47 1.30 13.16 9.20 0.16 1.59 4.37 4.11 0.98 0.29 0.51 97.64 266 56 andesite 62.47 1.35 13.25 9.24 0.17 1.58 4.45 3.95 0.96 0.33 97.74 265 66 andesite 62.81 1.43 13.13 9.05 0.18 1.99 4.98 3.86 0.89 0.24 0.23 98.55 266 58 dacit e 63.01 1.10 13.06 8.43 0.16 1.75 4.34 3.63 0.96 0.26 96.69 265 65 dacite 63.79 1.26 13.25 8.14 0.15 1.34 4.21 3.84 0.97 0.22 97.17 265 64 dacite 64.04 1.28 13.12 8.27 0.16 1.60 4.45 3.46 0.97 0.20 0.24 97.55 265 67 dacite 64.10 1.34 13.33 8.49 0.1 6 1.49 4.41 3.93 0.95 0.23 98.42 266 50 dacite 64.26 1.07 13.17 8.08 0.14 1.27 3.78 4.23 1.10 0.24 97.35 266 53 dacite 64.28 1.06 13.31 8.06 0.14 1.12 3.73 4.16 1.09 0.25 0.65 97.20 265 84 dacite 64.39 1.13 13.17 8.18 0.15 1.23 3.92 3.41 1.19 0.22 96.98 265 63 dacite 64.43 1.29 13.26 8.22 0.15 1.29 4.21 3.71 0.99 0.21 97.76 266 47 dacite 64.53 0.99 13.18 7.74 0.14 1.02 3.53 4.94 1.22 0.23 97.54 265 85 dacite 65.01 1.06 13.13 7.99 0.16 1.18 3.78 3.67 1.22 0.20 0.64 97.41 266 46 dacite 65.0 3 0.94 12.90 7.17 0.15 1.41 3.71 4.76 1.19 0.17 97.43 265 94 dacite 65.22 0.97 13.04 7.90 0.14 1.13 3.54 4.29 1.14 0.23 97.61 264 09 dacite 65.76 0.89 13.15 7.03 0.13 1.06 3.48 4.24 1.21 0.20 0.58 0.06 97.78 265 70 dacite 66.26 0.87 13.20 7.17 0.14 0.80 3.23 4.08 1.33 0.19 0.70 97.27 265 42 dacite 66.46 0.94 13.04 7.92 0.16 0.89 3.50 3.99 1.20 0.21 0.51 0.07 98.91 265 83 dacite 67.46 0.76 13.27 6.68 0.13 0.67 2.98 3.88 1.37 0.16 0.67 97.37 265 95 dacite 67.46 0.77 13.10 6.47 0.12 0.94 3.01 4.43 1.21 0.15 97.67 265 40 dacite

PAGE 136

136 Table 4 2. Trace element data 9N OSC sample 265 43 264 08 266 33 267 09 265 88 265 72 266 22 265 18 264 04 266 28 266 07 265 78 267 15 266 01 Li 7.78 9.99 7.52 8.23 7.97 7.74 8.24 7.48 8.47 9 .21 7.70 7.77 7.94 5.99 Sc 42 42 41 42 41 42 43 43 53 40 36 43 41 33 V 347 450 321 354 351 350 359 338 406 357 337 353 330 273 Cr 108 17.97 155 87.16 68.56 96.21 50.29 166 119 40.69 9.33 84.45 138 40.62 Co 41 43 40 41 42 42 43 42 51 41 38 43 40 34 Ni 54 34 48 46 47 54 38 53 72 32 47 52 51 38 Cu 59 60 61 60 60 60 61 64 76 55 50 64 58 51 Zn 94 116 88 96 97 96 99 95 103 100 93 95 93 79 Ga 18 21 18 19 18 18 19 18 24 20 17 18 18 14 Rb 1.14 2.12 1.20 1.50 1.25 1.09 1.05 1.00 1.64 1.50 1.55 1.22 1.37 0.98 Sr 120 126 114 113 124 117 120 126 146 111 98 125 108 89 Y 44 59 39 44 44 44 45 42 45 43 42 43 42 32 Zr 126 180 125 140 133 126 130 122 142 157 139 124 136 95 Nb 3.08 5.46 2.91 3.65 3.46 3.07 3.03 2.76 4.01 3.87 3.84 3.19 3.42 2.20 Cs 0.01 0.03 0.02 0.02 0.01 0.01 0.01 0.01 0.02 0.02 0.02 0.02 0.02 0.02 Ba 8.37 17.07 8.25 11.12 10.31 8.44 7.85 6.87 13.20 10.35 11.73 9.49 9.99 6.46 La 4.54 6.81 4.13 4.79 4.94 4.47 4.61 4.27 4.98 5.40 4.84 4.64 4.50 3.10 Ce 14.16 20.51 13.18 14.97 14.97 13.90 14.38 1 3.40 15.49 17.22 14.95 14.06 14.13 9.93 Pr 2.34 3.32 2.29 2.56 2.46 2.31 2.38 2.21 2.61 2.95 2.47 2.31 2.44 1.76 Nd 12.6 17.3 12.1 13.5 13.2 12.5 12.8 11.9 13.9 15.5 13.1 12.5 13.0 9.5 Sm 4.42 5.99 4.16 4.69 4.48 4.37 4.45 4.15 4.78 5.48 4.49 4.35 4.47 3.32 Eu 1.48 1.94 1.45 1.58 1.52 1.46 1.52 1.40 1.67 1.77 1.49 1.47 1.49 1.18 Gd 5.78 7.74 5.53 6.10 5.92 5.75 5.87 5.46 6.18 7.03 5.78 5.71 5.82 4.42 Tb 1.08 1.45 1.02 1.14 1.11 1.07 1.11 1.02 1.15 1.30 1.07 1.07 1.07 0.82 Dy 7.22 9.69 6.62 7.36 7.32 7.09 7.31 6.74 7.46 8.41 6.92 7.04 7.04 5.34 Ho 1.52 2.05 1.43 1.56 1.53 1.51 1.55 1.44 1.57 1.82 1.46 1.49 1.50 1.14 Er 4.39 5.94 4.04 4.47 4.48 4.34 4.51 4.15 4.55 5.18 4.27 4.33 4.27 3.26 Tm 0.67 0.91 0.62 0.68 0.68 0.66 0.69 0.63 0.68 0.80 0.64 0.66 0.66 0.50 Yb 4.43 6.01 3.97 4.35 4.47 4.33 4.55 4.16 4.41 5.15 4.17 4.37 4.19 3.19 Lu 0.68 0.93 0.60 0.66 0.68 0.67 0.70 0.64 0.68 0.76 0.63 0.66 0.64 0.48 Hf 3.44 4.83 3.20 3.58 3.59 3.40 3.50 3.23 3.59 4.04 3.49 3.37 3.46 2.50 Ta 0.21 0.37 0.20 0.25 0.23 0.21 0.21 0.19 0.28 0.26 0.27 0.22 0.22 0.16 Pb 0.39 0.61 0.60 0.59 0.45 0.40 0.41 0.38 0.59 0.87 0.58 0.42 0.77 0.42 Th 0.18 0.34 0.18 0.23 0.20 0.18 0.18 0.16 0.22 0.25 0.23 0.20 0.21 0.13 U 0.08 0.13 0.08 0.09 0.08 0.07 0.07 0.07 0.09 0.10 0.09 0.08 0.08 0.06

PAGE 137

137 Table 4 2. Continued sample 266 10 265 35 265 31 265 113 265 24 265 50 265 125 265 109 264 16 265 57 265 103 265 91 265 90 Li 8.51 7.85 7.91 11.01 15.32 28.86 14.31 19.24 18.52 20.73 22.27 23.57 28.20 Sc 46 42 42 38 27 13 31 29 33 31 2 6 25 24 V 347 342 342 323 196 73 247 176 230 217 129 93 70 Cr 150 127 112. 32.12 12.25 4.07 50.08 29.19 44.46 31.96 6.40 2.31 4.82 Co 44 41 41 39 27 13 31 27 30 29 23 21 20 Ni 57 57 54 34 16 7 29 23 25 18 10 6 6 Cu 65 62 59 51 31 20 46 35 34 27 27 23 22 Zn 99 96 94 103 118 100 97 126 129 137 144 147 174 Ga 21 18 18 21 23 29 23 27 24 26 30 31 30 Rb 1.51 1.18 1.25 2.23 4.70 12.5 4.22 4.42 4.60 5.34 5.30 5.74 6.34 Sr 121 119 118 111 99 69 96 104 113 109 103 106 114 Y 47 45 44 61 95 138 72 110 101 116 135 144 170 Zr 156 130 131 229 468 967 380 557 371 425 671 724 680 Nb 3.87 3.15 3.21 5.67 10.33 14.82 6.33 10.95 8.89 10.06 13.51 14.52 15.81 Cs 0.02 0.02 0.02 0.03 0.06 0.12 0.05 0.05 0.05 0.07 0.06 0.06 0.08 Ba 11.08 8.69 8.75 15.60 29.41 58.22 24.3 1 28.72 29.81 32.73 33.81 36.06 42.35 La 5.17 4.66 4.71 7.63 14.86 28.00 11.10 15.75 13.61 15.76 19.57 21.43 23.72 Ce 16.46 14.43 14.56 23.69 45.03 79.99 32.73 48.54 41.17 47.56 60.43 66.14 73.19 Pr 2.77 2.38 2.40 3.92 7.01 11.64 5.08 7.76 6.42 7.34 9.6 4 10.54 11.55 Nd 14.7 12.7 12.7 20.1 35.1 52.4 24.6 38.9 30.9 35.3 48.6 52.9 55.8 Sm 5.00 4.47 4.44 6.68 10.74 15.64 7.54 12.07 10.37 11.90 14.99 16.18 18.61 Eu 1.66 1.48 1.48 1.99 2.75 2.96 1.98 3.15 2.68 2.98 3.78 4.10 4.57 Gd 6.40 5.82 5.83 8.52 12. 94 17.82 9.21 14.75 12.89 14.65 18.33 19.83 22.78 Tb 1.19 1.10 1.09 1.58 2.36 3.41 1.72 2.70 2.44 2.78 3.34 3.57 4.26 Dy 7.71 7.24 7.24 10.13 15.14 22.37 11.31 17.61 16.19 18.52 21.76 23.38 27.74 Ho 1.61 1.54 1.54 2.20 3.19 4.79 2.39 3.71 3.44 3.93 4.52 4.89 5.88 Er 4.68 4.44 4.45 6.26 9.36 14.29 7.09 10.97 10.16 11.60 13.34 14.24 17.37 Tm 0.70 0.68 0.68 0.96 1.44 2.30 1.12 1.69 1.58 1.81 2.05 2.20 2.68 Yb 4.55 4.45 4.45 6.17 9.27 15.29 7.23 10.88 10.43 11.91 13.19 14.01 17.70 Lu 0.70 0.69 0.68 0.95 1.42 2.29 1.10 1.66 1.62 1.84 2.07 2.14 2.73 Hf 3.90 3.54 3.52 5.72 11.27 23.87 9.42 13.27 10.10 11.55 15.95 17.25 17.78 Ta 0.27 0.21 0.22 0.36 1.50 0.89 0.45 0.76 0.61 0.68 0.95 1.03 1.03 Pb 0.68 0.46 0.42 0.95 1.63 3.54 1.73 2.55 1.65 1.89 3.28 3.76 2 .41 Th 0.22 0.19 0.20 0.40 0.84 2.53 0.74 0.78 0.76 0.88 0.96 1.05 1.15 U 0.09 0.08 0.08 0.15 0.31 0.84 0.28 0.30 0.30 0.35 0.36 0.39 0.46

PAGE 138

138 Table 4 2. Continued sample 265 100 266 05 265 69 265 25 264 14 266 56 265 66 265 65 265 64 265 67 266 53 265 8 4 Li 27.14 20.40 26.34 28.11 26.38 27.40 36.88 31.68 33.61 31.78 30.17 28.59 Sc 24 26 22 19 24 17 26 17 20 18 15 15 V 100 216 180 87 160 93 184 122 140 121 61 102 Cr 9.51 9.77 16.05 9.15 40.69 5.34 15.33 12.94 12.27 12.40 3.81 1.75 Co 21 26 21 17 21 1 5 23 15 17 16 13 14 Ni 10 15 11 9 22 7 15 9 10 9 6 7 Cu 26 33 22 26 29 22 24 17 19 18 18 21 Zn 150 107 112 124 114 106 143 110 124 122 109 100 Ga 33 26 28 28 26 29 42 30 35 28 29 29 Rb 6.51 7.01 7.35 10.8 7.79 10.7 11.6 9.1 10.5 9.6 12.8 12.7 Sr 112 92 88 97 97 81 114 76 90 89 86 70 Y 153 105 114 154 124 133 164 132 148 142 157 133 Zr 721 645 605 881 542 901 945 735 842 622 856 968 Nb 16.24 10.90 11.39 17.36 11.40 15.17 16.79 12.98 14.81 12.97 15.52 14.61 Cs 0.07 0.08 0.08 0.14 0.10 0.11 0.12 0.11 0.12 0.12 0.14 0.13 Ba 41.77 38.30 40.22 58.57 43.51 53.60 62.57 50.53 57.07 52.86 64.61 59.79 La 23.67 17.91 18.93 28.81 19.39 25.36 28.95 23.25 26.31 23.43 28.53 27.27 Ce 72.67 52.40 55.75 84.42 56.99 73.59 84.22 67.48 76.53 67.97 82.17 77.81 Pr 11. 30 7.89 8.40 12.25 8.47 10.86 12.53 10.06 11.37 10.05 11.87 11.21 Nd 56.3 37.2 39.7 54.9 39.1 49.6 58.3 46.4 52.8 45.9 53.7 50.9 Sm 17.27 11.43 12.26 16.73 12.38 14.91 17.07 14.09 15.37 14.62 16.25 14.32 Eu 4.20 2.60 2.81 3.31 2.75 3.07 3.78 3.00 3.35 3 .12 3.18 2.80 Gd 20.71 13.66 14.69 19.55 14.82 17.58 20.38 16.83 18.13 17.43 19.00 16.58 Tb 3.79 2.53 2.76 3.68 2.86 3.25 3.83 3.19 3.43 3.36 3.65 3.09 Dy 24.73 16.76 18.27 24.22 19.28 21.34 25.32 21.08 22.81 22.85 24.21 20.31 Ho 5.24 3.60 3.98 5.17 4. 14 4.60 5.42 4.58 4.88 4.92 5.19 4.33 Er 15.41 10.63 11.66 15.49 12.44 13.59 16.36 13.55 14.65 14.86 15.79 13.16 Tm 2.38 1.68 1.86 2.49 1.98 2.18 2.58 2.19 2.33 2.37 2.50 2.10 Yb 15.34 10.90 12.12 16.50 13.15 14.05 16.83 14.25 15.20 15.66 16.74 13.66 L u 2.36 1.67 1.85 2.64 2.04 2.16 2.55 2.15 2.31 2.44 2.56 2.10 Hf 17.09 15.73 15.48 22.31 14.75 21.69 23.08 18.69 20.95 17.54 22.28 23.14 Ta 1.14 0.66 0.69 1.13 0.80 0.91 1.23 0.81 1.10 0.92 1.03 1.39 Pb 3.08 2.02 2.96 3.07 2.67 3.21 6.24 3.33 5.05 3.41 3.15 4.89 Th 1.17 1.31 1.35 2.31 1.28 2.08 2.00 1.74 1.82 1.64 2.29 2.28 U 0.45 0.46 0.46 0.89 0.51 0.70 0.71 0.59 0.65 0.64 0.86 0.82

PAGE 139

139 Table 4 2. Continued sample 265 63 265 85 265 94 264 09 265 70 265 42 265 83 265 95 265 40 Li 30.23 31.42 30.77 26. 68 33.56 31.62 32.45 31.37 30.55 Sc 17 14 13 12 12 14 11 10 12 V 121 73 63 46 45 58 32 52 51 Cr 9.79 4.65 1.49 3.82 3.70 3.42 3.40 3.01 4.17 Co 15 12 11 10 10 11 8 8 10 Ni 8 7 5 6 5 6 5 5 5 Cu 17 19 17 16 16 15 14 15 17 Zn 113 108 105 89 106 119 103 98 103 Ga 31 28 30 28 29 30 29 30 30 Rb 9.5 13.8 12.4 12.9 15.0 13.7 15.5 12.5 12.4 Sr 76 81 68 78 78 83 76 61 73 Y 132 154 146 151 160 160 159 145 146 Zr 745 872 1050 824 934 816 922 985 1013 Nb 13.18 15.27 16.15 14.89 15.90 16.36 15.57 16.53 16.60 Cs 0.10 0.15 0.13 0.13 0.17 0.16 0.17 0.13 0.13 Ba 49.78 68.07 60.04 65.65 72.69 70.01 76.40 62.17 59.70 La 23.53 29.07 29.10 28.98 30.89 29.03 30.69 29.16 29.47 Ce 68.11 82.46 83.93 83.89 88.15 82.95 87.16 83.65 85.00 Pr 10.16 11.78 12.15 11.97 12.4 9 11.98 12.32 12.02 12.44 Nd 47.3 52.1 55.2 52.7 55.0 53.5 54.0 54.6 56.6 Sm 13.77 16.01 15.74 15.92 16.67 16.69 16.46 15.64 16.89 Eu 2.99 3.01 3.02 2.95 3.09 3.39 3.05 2.89 3.32 Gd 16.37 18.51 18.17 18.54 19.47 19.69 18.90 17.66 19.68 Tb 3.10 3.55 3. 39 3.55 3.69 3.76 3.64 3.35 3.67 Dy 20.53 23.80 22.36 23.80 25.06 25.37 24.51 22.13 23.89 Ho 4.40 5.12 4.77 5.12 5.38 5.47 5.27 4.74 5.20 Er 13.19 15.50 14.37 15.66 16.42 16.48 16.18 14.61 15.35 Tm 2.10 2.49 2.29 2.52 2.63 2.64 2.62 2.32 2.49 Yb 13.54 16.69 14.81 16.84 17.54 17.56 17.54 15.12 16.19 Lu 2.06 2.58 2.30 2.61 2.73 2.75 2.72 2.32 2.42 Hf 18.62 22.96 24.87 22.68 24.80 22.24 24.75 24.51 24.97 Ta 0.99 1.05 1.18 1.04 1.10 1.12 1.08 1.23 1.02 Pb 5.82 3.59 5.75 2.76 3.84 3.47 3.80 4.12 3.64 T h 1.65 2.43 2.35 2.59 2.68 2.36 2.80 2.36 2.49 U 0.59 0.92 0.84 0.98 1.04 0.91 1.05 0.86 0.84

PAGE 140

140 Table 4 3. West limb major element data 9N OSC sample Rock Type SiO2 TiO2 Al2O3 FeO MnO MgO CaO Na2O K2O P2O5 Total 267 16 ferroba salt 50.76 1.18 14.81 9.62 0.19 8.72 12.51 2.06 0.07 0.13 100.1 267 17 FeTi 51.07 2.12 13.54 12.18 0.22 6.74 10.70 2.98 0.11 0.23 99.89 267 18 ferrobasalt 51.15 1.71 14.29 10.63 0.18 7.70 11.18 2.70 0.10 0.17 99.81 267 19 ferrobasalt 50.97 1.83 13.97 11.13 0.20 7.40 10.95 2.78 0.10 0.20 99.53 267 20 FeTi 50.78 2.16 13.45 12.43 0.22 6.66 10.47 2.93 0.12 0.23 99.44 267 21 ferrobasalt 51.23 1.72 14.26 10.64 0.20 7.77 11.14 2.74 0.10 0.18 99.97 267 22 ferrobasalt 51.02 1.80 14.15 10.87 0.20 7.52 10.85 2 .84 0.11 0.21 99.57 267 23 basaltic andesite 55.95 2.04 13.02 11.98 0.21 2.84 6.59 3.70 0.67 0.73 97.73 267 24 ferrobasalt 50.74 1.81 13.96 10.78 0.21 7.64 10.85 3.00 0.12 0.21 99.32 267 25 FeTi 50.90 2.06 13.49 12.42 0.23 7.29 10.84 2.59 0.15 0.24 10 0.2 267 26 ferrobasalt 50.70 1.86 13.73 11.57 0.22 7.21 10.77 2.87 0.16 0.22 99.31 267 27 ferrobasalt 50.69 1.87 13.44 11.44 0.22 7.12 10.60 2.82 0.17 0.25 98.61 267 29 ferrobasalt 51.68 1.61 13.62 11.20 0.21 7.00 10.45 2.80 0.22 0.19 98.97 267 30 basa ltic andesite 52.33 1.92 13.28 12.09 0.23 5.43 9.06 3.25 0.37 0.48 98.45 267 32 ferrobasalt 50.57 1.75 14.05 10.61 0.20 7.76 10.89 2.99 0.12 0.20 99.14 267 33 ferrobasalt 50.72 1.78 14.04 10.72 0.20 7.71 10.89 2.99 0.12 0.21 99.39 267 34 ferrobasalt 5 0.67 1.80 14.00 10.72 0.21 7.73 10.85 2.98 0.12 0.18 99.25 267 35 FeTi 50.49 2.10 13.72 12.00 0.23 6.88 10.83 3.19 0.14 0.22 99.81 267 37 ferrobasalt 50.79 1.88 13.76 11.61 0.21 7.26 11.03 2.90 0.14 0.21 99.78 267 38 ferrobasalt 50.83 1.58 14.28 10.23 0 .19 7.67 11.76 2.98 0.08 0.14 99.73 267 39 ferrobasalt 50.78 1.78 14.05 10.80 0.20 7.75 11.05 3.00 0.12 0.20 99.72 267 40 ferrobasalt 50.88 1.76 14.18 10.76 0.21 7.82 10.88 3.00 0.12 0.21 99.83 267 41 basaltic andesite 52.40 3.06 12.67 14.13 0.24 4.21 7.83 3.77 0.64 0.46 99.41 267 42 ferrobasalt 50.69 1.76 14.11 10.74 0.20 7.81 10.99 3.02 0.12 0.20 99.64 267 43 ferrobasalt 50.74 1.76 14.16 10.78 0.20 7.77 10.97 3.03 0.12 0.17 99.70 267 44 ferrobasalt 50.86 1.76 14.06 10.90 0.21 7.54 11.11 2.99 0.12 0.19 99.74 267 45 ferrobasalt 50.77 1.76 13.88 10.84 0.20 7.71 11.01 2.95 0.12 0.19 99.44 267 46 ferrobasalt 50.78 1.78 14.01 10.88 0.20 7.63 11.04 2.98 0.12 0.21 99.63 267 47 ferrobasalt 50.33 1.83 14.12 10.24 0.20 7.28 11.38 2.89 0.32 0.27 98.87 267 48 ferrobasalt 49.95 1.65 14.43 10.66 0.21 8.01 11.73 2.62 0.09 0.21 99.54 267 50 ferrobasalt 49.76 1.58 14.83 10.57 0.20 8.31 11.69 2.63 0.09 0.17 99.82 267 51 ferrobasalt 50.64 1.80 14.11 10.86 0.20 7.55 11.08 2.96 0.12 0.21 99.54 267 52 ferrobasalt 5 0.20 1.95 13.70 11.64 0.21 7.08 10.97 3.12 0.12 0.22 99.23 267 53 ferrobasalt 50.67 1.57 14.28 10.24 0.20 7.84 11.78 2.73 0.11 0.17 99.58

PAGE 141

141 Table 4 3. Continued sample Rock Type SiO2 TiO2 Al2O3 FeO MnO MgO CaO Na2O K2O P2O5 Total 267 54 ferrobasalt 49.98 1.71 14.19 10.68 0.20 7.65 11.52 2.91 0.11 0.18 99.13 267 55 ferrobasalt 50.30 1.54 14.48 10.02 0.20 7.98 11.83 2.71 0.11 0.17 99.33 267 56 FeTi 50.18 2.22 13.05 13.11 0.24 6.42 10.06 3.09 0.15 0.26 98.78 267 57 ferrobasalt 4 9.82 1.36 14.93 9.35 0.18 8.35 12.37 2.55 0.14 0.16 99.20 267 58 ferrobasalt 49.98 1.82 14.25 10.24 0.21 7.22 11.52 2.88 0.32 0.27 98.70 267 59 FeTi 49.67 2.00 14.29 10.64 0.20 6.99 11.16 3.10 0.36 0.26 98.67 267 60 FeTi 50.14 2.12 13.13 12.54 0.23 6. 73 10.35 2.97 0.14 0.24 98.61 267 61 ferrobasalt 49.46 1.56 14.79 10.50 0.20 8.17 11.62 2.62 0.09 0.18 99.20 267 62 ferrobasalt 50.70 1.35 14.64 9.44 0.18 8.34 11.98 2.46 0.08 0.14 99.30 267 63 ferrobasalt 50.13 1.43 13.96 9.89 0.20 8.10 12.11 2.41 0.09 0.16 98.47 267 64 ferrobasalt 49.72 2.40 14.67 10.76 0.19 6.46 10.21 3.19 0.59 0.41 98.59 267 65 ferrobasalt 50.97 1.72 14.00 10.91 0.21 7.50 11.05 2.66 0.12 0.18 99.33 267 66 ferrobasalt 49.91 2.11 13.27 12.34 0.22 6.76 10.15 2.87 0.13 0.23 97.99 267 67 FeTi 50.70 2.05 13.44 12.03 0.23 6.87 10.27 2.91 0.12 0.21 98.84 267 68 ferrobasalt 50.66 2.02 13.58 11.94 0.21 6.95 10.33 2.86 0.12 0.22 98.89 267 69 FeTi 50.80 2.04 13.55 12.01 0.21 6.97 10.35 2.88 0.12 0.21 99.14

PAGE 142

142 Table 4 4. West limb trace element data 9N OSC sample 267 16 267 18 267 23 267 62 267 63 267 64 267 68 267 69 Li 5.3 6.9 19.4 5.4 5.6 7.9 8.2 7.7 Sc 40 41 26 40 39 41 42 41 V 275 312 148 278 266 345 368 350 Cr 368 147 24 359 339 240 80 71 Co 43 40 21 40 38 43 43 42 Ni 117 64 15 72 66 113 48 47 Cu 146 64 27 78 73 66 56 56 Zn 73 86 117 75 73 96 100 96 Ga 15 17 25 15 15 25 19 19 Rb 0.52 1.02 8.91 0.89 0.98 11.13 1.01 1.15 Sr 87 109 102 96 96 327 116 106 Y 27 37 127 29 29 43 47 43 Zr 62 112 669 82 82 208 133 134 Nb 1.55 2 .48 19.92 2.00 2.16 21.36 3.22 3.13 Cs 0.01 0.02 0.11 0.01 0.02 0.12 0.01 0.02 Ba 3.39 7.24 70.75 7.07 7.31 131.32 8.56 9.24 La 2.38 3.65 23.42 2.86 2.90 15.43 4.80 4.52 Ce 7.21 11.68 66.80 8.95 9.02 36.58 14.93 14.46 Pr 1.21 2.08 10.13 1.54 1.59 4.96 2.47 2.47 Nd 6.98 11.05 48.44 8.39 8.63 22.53 13.42 13.23 Sm 2.44 3.86 14.98 2.95 3.03 5.96 4.72 4.58 Eu 0.89 1.34 3.55 1.07 1.11 2.00 1.60 1.55 Gd 3.37 5.08 18.03 3.93 4.07 6.87 6.15 5.89 Tb 0.65 0.93 3.28 0.73 0.75 1.17 1.16 1.10 Dy 4.26 6.08 21.11 4.74 4.90 7.27 7.67 7.14 Ho 0.93 1.30 4.50 1.01 1.06 1.48 1.61 1.50 Er 2.68 3.69 12.96 2.90 2.99 4.18 4.68 4.36 Tm 0.41 0.56 2.02 0.44 0.46 0.62 0.71 0.66 Yb 2.68 3.60 12.92 2.83 2.93 3.93 4.68 4.23 Lu 0.41 0.55 1.98 0.43 0.44 0.60 0.72 0.65 Hf 1.7 4 2.86 16.00 2.09 2.17 4.70 3.56 3.41 Ta 0.10 0.17 1.11 0.14 0.15 1.37 0.22 0.22 Pb 0.13 0.52 2.02 0.24 0.41 1.86 0.44 0.39 Th 0.09 0.15 1.85 0.12 0.14 1.42 0.18 0.18 U 0.04 0.06 0.54 0.05 0.06 0.42 0.07 0.07

PAGE 143

143 Table 4 5. West limb isotopic d ata 9N OSC sample 267 18 267 23 267 62 267 63 267 64 267 69 208Pb/204Pb 37.736 37.846 37.748 37.737 38.022 37.738 2 sigma error 0.0018 0.0019 0.0020 0.0022 0.0017 0.0015 207Pb/204Pb 15.478 15.495 15.480 15.476 15.530 15.476 2 sigma error 0.0007 0.000 7 0.0008 0.0008 0.0006 0.0006 206Pb/204Pb 18.25 3 18.368 18.258 18.256 18.590 18.263 2 sigma error 0.0008 0.0008 0.0010 0.0009 0.0006 0.0007 208Pb/206Pb 2.0674 2.0603 2.0674 2.0671 2.04529 2.0664 2 sigma error 0.00004 0.00003 0.00003 0.00003 0.00003 0.0 0003 207Pb/206Pb 0.8480 0.843 5 0.8478 0.847 0.8354 0.8474 2 sigma error 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 87Sr/86Sr 0.70249 0.70265 0.70254 0.70259 0.70282 0.70254 2 sigma error 0.00001 0.00001 0.00001 0.00002 0.00002 0.00002 143Nd/144Nd 0.51315 2 0.513139 0.513149 0.51315 0.513049 0.51314 4 2 sigma error 0.000007 0.000006 0.000004 0.000007 0.000004 0.000004 Eps Nd 10.0 9.8 10.0 10.0 8.0 9.8

PAGE 144

144 Table 4 6. Major and trace element data from 837'N EPR 3925 0 39 25 2 3925 3 3926 1 3926 3 3926 4 3926 5 3927 1 3927 2 SiO2 51.97 50.7 50.64 50.59 55 57.14 50.73 50.47 50.74 TiO2 1.72 1.63 1.63 1.72 1.87 1.64 1.6 1.82 1.73 Al2O3 14.53 14.7 14.63 14.68 14.17 14.14 14.56 14.37 14.49 FeO 10.43 10.49 10.31 10.31 10.95 1 0.15 10.38 10.94 10.63 MnO 0.2 0 0.21 0.18 0.19 0.18 0.17 0.17 0.19 0.19 MgO 6.25 7.29 7.24 7.21 3.85 3.08 7.26 6.83 7.07 CaO 10.42 11.59 11.69 11.59 7.36 6.5 11.56 11.1 11.41 Na2O 3.19 2.84 2.83 2.94 4.05 4.25 2.86 3 2.9 K2O 0.33 0.16 0.16 0.25 0.69 0 .87 0.17 0.2 0.18 P2O5 0.31 0.29 0.29 0.33 0.46 0.48 0.29 0.32 0.31 SO2 0.2 4 0.26 0.2 9 0.2 5 0.22 0.1 8 0.2 6 0.28 0.26 Total 99.61 100.2 99.93 100.1 98.83 98.61 99.86 99.55 99.93 Li 9.08 7.64 6.03 17.72 6.24 6.57 6.53 Sc 40.38 39.55 42.32 25.02 42.4 7 41.93 41.19 V 309. 288 315 15 2 314 331 315 Cr 114 179 185 43.99 168 89.65 138 Co 39.47 38.39 41.32 23.96 41.38 40.96 39.94 Ni 48.26 49.41 53.77 17.92 51.88 46.03 48.29 Cu 65.95 71.34 76.86 41.13 76.57 71.19 70.79 Zn 90.55 82.35 81.56 11 0.08 112.00 86.75 82.43 Ga 17.75 16.36 15.78 21.32 15.84 16.52 15.81 Rb 4.92 3.67 2.08 14.11 1.95 3.00 2.40 Sr 140 127 131 119 129 148. 134 Y 53.42 45.73 35.37 117.23 35.66 38.69 36.61 Zr 211 171 110 550 125 122 112 Nb 7.47 5.72 4.19 19.41 3.94 5.85 4.73 Cs 0.06 0.05 0.03 0.16 0.03 0.04 0.03 Ba 38.11 28.50 20.10 99.52 18.91 30.14 23.83 La 8.91 7.22 4.66 23.07 4.47 5.91 5.02 Ce 24.28 19.91 13.20 61.72 12.77 16.24 14.11 Pr 3.69 3.04 2.12 8.97 2.05 2.51 2.25 Nd 17.93 15.03 10. 88 41.80 10.72 12.67 11.54 Sm 5.66 4.85 3.69 12.61 3.63 4.14 3.85 Eu 1.71 1.50 1.33 3.01 1.29 1.45 1.38 Gd 6.97 6.09 4.81 14.87 4.69 5.25 4.96 Tb 1.31 1.13 0.89 2.74 0.87 0.96 0.92 Dy 8.44 7.38 5.75 17.82 5.68 6.29 5.98 Ho 1.79 1.57 1.21 3.79 1.21 1.32 1.25 Er 5.22 4.58 3.49 11.08 3.48 3.81 3.61 Tm 0.80 0.70 0.53 1.75 0.52 0.57 0.54 Yb 5.23 4.51 3.43 11.48 3.37 3.69 3.48 Lu 0.80 0.69 0.52 1.76 0.52 0.56 0.53 Hf 5.37 4.48 2.91 13.78 3.23 3.15 2.95 Ta 0.46 0.36 0.26 1.16 0.25 0.37 0.30 Pb 0.87 0.77 0.59 2.16 1.50 0.68 0.61 Th 0.75 0.59 0.30 2.15 0.29 0.42 0.34 U 0.26 0.21 0.10 0.74 0.10 0.14 0.12

PAGE 145

145 Table 4 7. Radiogenic isotope ratios 837'N EPR Sample 3925 R0 3925 R2 3925 R3 3926_R4 3926 R5 3927 R1 3927 R2 208 Pb/204Pb 37.957 37.938 37.976 37.911 37.965 37.992 37.983 2 sigma error 0.0020 0.0019 0.0036 0.0020 0.0017 0.0017 0.0023 207Pb/204Pb 15.510 15.508 15.513 15.505 15.514 15.515 15.517 2 sigma error 0.0008 0.0008 0.0015 0.0008 0.0007 0.0006 0.0008 206Pb/2 04Pb 18.468 18.445 18.484 18.437 18.463 18.4960 18.482 2 sigma error 0.0008 0.0008 0.0016 0.0008 0.0008 0.0007 0.0009 208Pb/206Pb 2.0553 2.0568 2.0546 2.0562 2.0562 2.0541 2.0551 2 sigma error 0.00004 0.00004 0.00005 0.00003 0.00003 0.00003 0.00005 207 Pb/206Pb 0.83984 0.84077 0.83928 0.84093 0.84025 0.83881 0.83955 2 sigma error 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 87Sr/86Sr 0.70262 0.70267 0.70257 0.70265 0.70258 0.70265 0.70260 2 sigma error 0.00001 0.00001 0.00002 0.00001 0.0000 2 0.00001 0.00001 143Nd/144Nd 0.51314 0.51315 0.51312 0.51313 0.51314 0.51314 0.51316 2 sigma error 0.0000 1 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 Eps Nd 9.7 9.9 9.5 9.6 9.8 9.9 10.2

PAGE 146

146 Figure 4 1. Bathymetric map of the northern EPR, including the location of 9N, 837, the Clipperton Transform and the Siqueiros Transform. (Data from GeoMapApp; Carbotte et al., 2004). View looking north.

PAGE 147

147 Figure 4 2 Bathymetric map of the 9N OSC with 50m contours. Circles show the location of rock samp les collected using the Jason2 ROV during the MEDUSA2007 cruise, with warmer colors representing higher silica contents. Colored outlines delineate regions discussed in the text.

PAGE 148

148 Figure 4 3. Bathymetric map of the 9N OSC with 50m contours. The melt si lls underlying the east and west limbs of the OSC are shaded in gray (Kent et al., 2000). Black bar represents the approximate extent of dacites

PAGE 149

149 Figure 4 4. Side scan sonar mosaic from data collected on the MEDUSA2007 cruise using DSL-120A (White et al., 2009). Circles show the location of rock samples collected using the Jason2 ROV during the MEDUSA2007 cruise, with warmer colors representing higher silica contents. Yellow bar represents the approximate extent of the axial summit trough (AST) in the region. South of the AST, the neovolcanic zone is difficult to identify.

PAGE 150

150 Figure 4 5 FeO versus MgO for glasses collected from the east limb of the 9N OSC. Shaded regions discriminate between different rock types, which are dominated by different pet rologic processes. Black lines with xs represent various bulk mixes of high -silica dacites and lower silica end -members. Red and blue lines with crosses show liquid-lines of descent (calculated using MELTS; Ghiorso and Sack, 1995) of two different oxygen fugacities (blue = QFM 1 and red = QFM). The majority of lavas erupted at the OSC can be explained by fractional crystallization or mixing various proportions of a highsilica and basaltic endmember.

PAGE 151

151 Figure 4 6. Major element variations versus MgO (wt%) for glasses collected from the east limb of the 9N OSC. Compositions are compared to two fractional crystallization trends (calculated using MELTS; Ghiorso and Sack, 1995) with the same parental composition but different oxygen fugacities. B lack lines with xs represent bulk mixing trends. The majority of east limb lavas can be explained by either by fractional crystallization or by mixing of an evolved and dacitic endmembers. A B. C. D. E. F.

PAGE 152

152 Figure 4 7. P2O5/TiO2 versus MgO (wt%) for east limb glasses. The high -P2O5 andesites lie along calculated fractional crystallization trends, while the low -P2O5 andesites and dacites have lower P2O5/TiO2 ratios for a given MgO. Many of the basaltic andesites can be explained by mixing of various end -members.

PAGE 153

153 Figure 4 8. Trace element concentrations versus Zr for glasses collected from the east limb of the OSC. Not all concentrations can be explained by fractional crystallization alone. Mixing of high-silica lavas with various ferrobasalts can explain a wid e range of compositions erupting at the OSC. A B. C. D. E. F.

PAGE 154

154 Figure 4 9 Incompatible trace element ratios versus Zr for glasses erupted at the OSC. Low U/Nb ratios in high -P2O5 andesites can be explained by fractional crystallization, however, Zr/Nb ratios indicate another process, such as mixing or assimilation, must be involved A B. C. D.

PAGE 155

155 A Figure 4 10 Radiogenic isotope ratios showing the variation in sources along the northern EPR, from 950 N to the Siquerous Transform Fault. A) Pb/Pb data of lavas collected south o f the OSC, generally have more radiogenic signatures than lavas collected to the north of the OSC. N -MORB from both the east and west limb are similar to other N MORB lavas erupted along the EPR but the west limb are slightly more radiogenic. The west li mb has also erupted E -MORB lavas. B) Epsilon Nd versus Sr isotopes, showing the variation in sources erupting at the EPR.

PAGE 156

156 B Figure 410. Continued.

PAGE 157

157 A Figure 4 11 Major element concentrations and ratios versus MgO comparing the east and west limb of th e OSC. Samples collected from the EPR north of the OSC (black crosses) and during the CHEPR cruise near the OSC and 837(green symbols) are shown for comparison. A) Major element variations versus MgO. The east limb of the OSC has erupted a much wider compositional range compared to the west limb but the west limb is, on average, more primitive. Red line with crosses represents a calculated fractional crystallization trend using a primitive EPR lava as a parent. B) Major element ratios versus MgO. The west limb has erupted basaltic compositions with higher K2O/TiO2 than the east limb, consistent with E MORB compositions.

PAGE 158

158 B Figure 411. Continued.

PAGE 159

159 Figure 4 12. Trace element concentrations versus Zr comparing east and west limb basalts and b asaltic andesites. The west limb lavas are more primitive relative to the east limb lavas. A B C D E F

PAGE 160

160 Figure 4 13 Primitive mantle normalized diagram showing variations in andesites and basaltic andesites erupted at the OSC. Red line is a MORB from the east lim b. Blue lines are the high -P2O5 andesites from the east limb. Green line is the andesite erupted on the west limb. Gray lines are andesites and basaltic andesites erupted on the east limb. The west limb andesite and the high-P2O5 andesite lack the dist inct negative Nb, Ta anomaly and do not have as high U and Th compared to the other east limb lavas.

PAGE 161

161 CHAPTER 5 CONCLUSIONS The majority of eruptions at spreading centers produce basalts with relatively limited chemical variability; however, compositio ns ranging from basalts to dacites have been sampled at ridge segment ends We have documented the eruption of high silica lavas on the propagating eastern limb of the 9N overlapping spreading center (OSC) on the East Pacific Rise. The dacites which ha ve erupted on several other ridges, appear to represent an endmember composition that shows similar major element trends and incompatible trace element enrichments, suggesting similar processes controlled their petrogenesis. The formation of highly evolved lavas on MOR requires a combination of partial melting, assimilation and crystal fractionation. The highly enriched incompatible trace element signatures cannot be produced through crystal fractionation alone and appears to require partial melting of al tered ocean crust. EC AFC modeling suggests significant amounts (>75%) crystallization of a MORB parent magma and modest amounts (520%) of assimilation of hydrothermally altered ocean crust can produce geochemical signatures consistent with dacite compos itions. The AFC process explains trace element abundances in high-silica lavas and accounts for several major and minor element concentrations (i.e. Al2O3, K2O and Cl). The formation of dacitic lavas on MOR appears to require a unique tectonomagmatic sett ing, where episodic magma supply allows for extensive crystal fractionation, partial melting and assimilation of altered crustal material These conditions are met in regions of ridge propagation, such as OSC and propagating ridge tips, where down axis di king allows for episodic injection of magma into older, altered

PAGE 162

162 ocean crust. Here, the magma undergoes extensive crystallization without repeated replenishment, creating enough latent heat of crystallization to melt and assimilate surrounding wall rock. V ariations in v olatile concentrations and 18O in 9N OSC lavas also suggest that the OSC magmas have experienced assimilation during their petrogenesis, with the most extreme signatures observed in high-silica andesites and dacites and little evidence in basaltic lavas. H2O concentrations are up to two times higher in dacitic lavas compared to calculated fractional crystallization trends, whereas Cl has excesses of seven to ten times predicted values. 18O values are on average ~1 lower than ratios expected from fractional crystallization of ferr omagnesian silicates and Fe-Ti oxide phases, consistent with assimilation of an additional component or components. The source of the excess H2O and Cl and low 18O values is partially melted, hydrothermally altered oceanic crust. Vapor saturation pressures calculated from H2O CO2 data suggest that assimilation most likely occurs at the top of the melt lens, which at the 9N OSC, corresponds approximately to the base of the sheeted dikes. The distribution of evolved lavas and E -MORB lavas across the OSC i s not symmetric, suggesting that the 2nd order discontinuity represents a division in the magmatic plumbing system of the EPR. E MORB lavas are only observed on the dying western limb and overall the lavas are less evolved than the adjacent eastern limb. We suggest that the lower magma supply at the west limb allows for the preservation and eruption of E MORB compositions, whereas the more robust magmatic system on the east limb overwhelms this signature. N -MORB lavas on the west limb have more radiogenic Pb and Sr and less radiogenic Nd compared to east limb N -MORB lavas.

PAGE 163

163 Lavas erupted to the south of the OSC (837N) also have more radiogenic Pb and Sr isotope ratios. This suggests a slightly different mantle is feeding this section of the EPR and that this large OSC provides a fundamental division between mantle sources beneath the ridge axis.

PAGE 164

164 LIST OF REFERENCES Agrinier, P., Hekinian, R., Bideau, D. & Javoy, M. (1995). O and H stable isotope compositions of oceanic crust and upper mantle rocks exposed in the Hess Deep near the Galapagos Triple Junction. Earth and Planetary Science Letters 136, 183-196. Alt, J., Honnorez, J., Laverne, C. & Emmerman, R. (1986). Hydrothermal Alteration of a 1 km Section Through the Upper Oceanic Crus t, Deep Sea Drilling Project Hole 504B: Mineralogy, Chemistry, and Evolution of Seawater -Basalt Interations. Journal of Geophysical Research 91, 10309 -10335. Alt, J., Laverne, C., Vanko, D. A., Tartarotti, P., Teagle, D. A. H., Bach, W., Zuleger, E., Erzi nger, J., Honnorez, J., Pezard, P., Becker, K., Salisbury, M. H. & Wilkens, R. H. (1996). Hydrothermal Alteration of a Section of Upper Oceanic Crust in the Eastern Equatorial Pacific: A synthesis of results from site 504 (DSDP Legs 69, 70, and 83, and ODP Legs 111, 137, 140, and 148). Proceedings of the Ocean Drilling Program, Scientific Results 148, 417434. Alt, J. & Teagle, D. A. H. (eds.) (2000). Hydrothermal alteration and fluid fluxes in ophiolites and oceanic crust Boulder, CO: Geological Society of America Alt, J. C. & Teagle, D. A. H. (2003). Hydrothermal alteration of upper oceanic crust formed at a fast -spreading ridge: mineral, chemical, and isotopic evidence from ODP Site 801. Chemical Geology 201, 191 -211. Anderson, A. T., Clayton, R. N. & Mayeda, T. K. (1971). Oxygen isotope thermometry of mafic igneous rocks. Journal of Geophysical Research 85, 715 -729. Aumento, F. (1969). Diorites from the mid-Atlantic ridge at 45N. Science 165, 11121113. Bachmann, O. & Bergantz, G. W. (2004). On the Origin of Crystal poor Rhyolites: Extracted from Batholithic Crystal Mushes. Journal of Petrology 45, 1565 -1582. Barker, A. K., Coogan, L. A. & Gillis, K. M. (2008). Strontium isotope constraints on fluid flow in the sheeted dike complex of fast spreading crust: Pervasive fluid flow at Pito Deep. Geochemistry Geophysics Geosystems 9, 1019. Batiza, R. & Niu, Y. (1992). Petrology and Magma Chamber Processes at the East Pacific Rise ~ 930'N. Journal of Geophysical Research 97, 67796797. Bazin, S., Hardin g, A. J., Kent, G. M., Orcutt, J. A., Tong, C. H., Pye, J. W., Singh, S. C., Barton, P. J., Sinha, M. C., White, R. S., Hobbs, R. W. & Van Avendonk, H. J. A. (2001). Three-dimensional shallow crustal emplacement at the 903' N overlapping spreading center on the East Pacific Rise: Correlations between

PAGE 165

165 magnetization and tomographic images. Journal of Geophysical Research 106, 16,101116,117. Beard, J. S. & Lofgren, G. E. (1991). Dehydration melting and water saturated melting of basaltic and andesitic greens tones and amphibolites at 1, 3, and 6.9 kb. Journal of Petrology 32, 365401. Beccaluva, L., Chinchilla -Chaves, A. L., Coltorti, M., Giunta, G., Siena, F. & Vaccaro, C. (1999). Petrological and structural significance of the Santa Elena-Nicoya ophiolitic complex in Costa Rica and geodynamic implications. Contributions to Mineralogy and Petrology 11, 1091 1107. Bdard, J. H., Hbert, R. & Berclaz Alain, V., Veronika (eds.) (2000). Syntexis and the genesis of lower oceanic crust Boulder, CO: Geological Soc iety of America Berndt, M. E. & Seyfreid, W. E. J. (1990). Boron, bromine, and other trace elements as clues to the fate of chlorine in mid ocean ridge vent fluids. Geochemica et Cosmochimica Acta 54, 2235 -2245. Bohnenstiehl, D. R., Tolstoy, M., Fox, C. G., Dziak, R. P., Chapp, E., Fowler, M., Haxel, J., Fisher, C., Van Hilst, C., Laird, R., Collier, R., Cowen, J. P., Lilley, M., Simons, K., Carbotte, S. M., Reynolds, J., R. & Langmuir, C. H. (2003). Anomalous Seismic Activity at 837 -42'N on the East Pac ific Rise: Hydroacoustic Detection and Site Investigation. Ridge 2000 Newsletter 1, 18-20. Bohrson, W. A. & Reid, M. R. (1997). Genesis of Silicic Peralkaline Volcanic Rocks in an Ocean Island Setting by Crustal Melting and Open-system Processes: Socorro Island, Mexico. Journal of Petrology 38, 1137 1166. Bohrson, W. A. & Reid, M. R. (1998). Genesis of Evolved Ocean Island Magmas by Deepand Shallow -Level Basement Recycling, Socorro Island, Mexico: Constraints from Th and other Isotope Signatures. Journa l of Petrology 39, 9951008. Bohrson, W. A. & Spera, F. (2001). Energy -Constrained OpenSystem Magmatic Processes II: Application of Energy Constrained Assimilation-Fractional Crystallization (EC -AFC) Model to Magmatic Systems. Journal of Petrology 42, 10 191041. Bowen, N. L. (1928). The evolution of igneous rocks : Princeton University Press. Brophy, J. G. (2009). La -SiO2 and Yb-SiO2 systematics in mid ocean ridge magmas: implications for the origin of oceanic plagiogranite. Contributions to Mineralogy a nd Petrology 158, 99111.

PAGE 166

166 Bryan, W. B. & Moore, J. G. (1977). Compositional variations of young basalts in the Mid -Atlantic Ridge rift valley near 3649' N. Geological Society of America Bulletin 88, 556-570. Byerly, G. R. (1980). The nature of different iation trends in some volcanic rocks from the Galapagos Spreading Center. Journal of Geophysics 85, 3797 3810. Byerly, G. R. & Melson, W. G. (1976). Rhyodacites, andesites, ferrobasalts and ocean tholeiites from the Galapagos spreading center. Earth and Planetary Science Letters 30, 215221. Canales, P. J., Detrick, R. S., Carbotte, S. M., Kent, G. M., Diebold, J. B., Harding, A. J., Babcock, J., Nedimovie, M. R. & van Ark, E. (2005). Upper crustal structure and axial topography at intermediate spreading ridges: Seismic constraints from the southern Juan de Fuca Ridge. Journal of Geophysical Research 110, 1 -27. Carbotte, S. M., Arko, R., Chayes, D. N., Haxby, W., Lehnert, K., O'Hara, S., Ryan, W. B. F., Weissel, R. A., Shipley, T., Gahagan, L., Johnson, K. T. M. & Shank, T. M. (2004a). New Integrated Data Management System for Ridge2000 and MARGINS Research. EOS, Transactions American Geophysical Union 85(51), 553. Carbotte, S. M. & Macdonald, K. C. (1992). East Pacific Rise 8 -1030'N: Evolution of ridg e segments and discontinuities from SeaMARC II and three-dimensional magnetic studies. Journal of Geophysical Research 97, 6959 -6982. Carbotte, S. M., Small, C. & Donnelly, K. (2004b). The influence of ridge migration on the magmatic segmentation of mid o cean ridges. Nature 429, 743746. Casey, J. F. (ed.) (1997). Comparison of major and traceelement geochemistry of abyssal peridotites and mafic plutonic rocks with basalts from the MARK region of the midAtlantic ridge. College Station: Ocean Drilling P rogram Chadwick, J., Perfit, M. R., Ridley, W. I., Jonasson, I. R., Kamenov, G., Chadwick, W. W. J., Embley, R. W., le Roux, P. J. & Smith, M. C. (2005). Magmatic effects of the Cobb hot spot on the Juan de Fuca Ridge. Journal of Geophysical Research 110, 1 -16. Christie, D. M. & Sinton, J. M. (1981). Evolution of abyssal lavas along propagating segments of the Galapagos Spreading Center. Earth and Planetary Science Letters 56, 321. Clague, D. A. & Bunch. (1976). Formation of ferrobasalt at East Pacific mid ocean spreading centers. Journal of Geophysical Research 81, 42474256.

PAGE 167

167 Coleman, R. G. & Donato, M. M. (1979). Oceanic plagiogranite revisited : Elsevier, Amsterdam. Coogan, L. A. (2003). Contaminating the lower crust in the Oman ophiolite. Geology 31, 10651068. Coogan, L. A., Gillis, K. M., MacLeod, C. J., Thompson, G. M. & Hekinian, R. (2002). Petrology and geochemistry of the lower ocean crust formed at the East Pacific Rise and exposed at Hess Deep: A synthesis and new results. Geochemistry Geophysics Geosystems 3, 1 -30. Coogan, L. A., Banks, G. J., Gillis, K. M., MacLeod, C. J. & Pearce, J. A. (2003a). Hidden melting signatures recorded in the Troodos ophiolite plutonic suite: evidence for widespread generation of depleted melts and intra -crusta l melt aggregation. Contributions to Mineralogy and Petrology 144, 484505. Coogan, L. A., Mitchell, N. C. & O'Hara, M. J. (2003b). Roof assimilation at fast spreading ridges: An investigation combining geochemical, and field evidence. Journal of Geophysi cal Research 108, 1-14. Cotsonika, L. (2006). Petrogenesis of Andesites and Dacites from the Southern Juan De Fuca Ridge. Department of Geological Sciences Gainesville: University of Florida, 188. Danyushevsky, L. V. (2001). The effect of small amounts of H2O on crystallization of mid ocean ridge and backarc basin magmas. Journal of Volcanology and Geothermal Research 110, 265280. De Paolo, D. J. (1981). Trace element and isotopic effects of combined wallrock assimilation and fractional crystallizatio n. Earth and Planetary Science Letters 53, 189202. Delacour, A., Fruh -Green, G. L., Frank, M., Gutjahr, M. & Kelley, D. S. (2008). Sr and Nd -isotope geochemistry of the Atlantis Massif (30N, MAR): Implications for fluid fluxes and lithospheric heterogeneity. Chemical Geology 254, 1935. Detrick, R. S., Buhl, P., Vera, E., Mutter, J., Orcutt, J. A., Madsen, J. & Brocher, T. (1987). Multichannel seismic imaging of a crustal magma chamber along the East Pacific Rise. Nature 326, 3541. Dick, J. B., Natla nd, J. H., Alt, J., Bach, W., Bideau, D., Gee, J. S., Haggas, S., Hertogen, J. G. H., Hirth, G., Holm, P. M., Ildefonse, B., Iturrino, G. J., John, B. E., Kelley, D. S., Kikawa, E., Kingdon, A., le Roux, P. J., Maeda, J., Meyer, P. S., Miller, D. J., Naslu nd, H. R., Niu, Y., Robinson, P. T., Snow, J., Stephen, R. A., Trimby, P. W., Worm, H. U. & Yoshinobu, A. (2000). A long in situ section of the

PAGE 168

168 lower ocean crust: results of ODP Leg 176 drilling at the Southwest Indian Ridge. Earth and Planetary Science Letters 179, 31 51. Dixon, J. E. & Pan, V. (1995). Determination of the molar absorptivity of dissolved carbonate in basanitic glass. American Mineralogist 80, 1339 -1342. Dixon, J. E., Stolper, E. M. & Delaney, J. R. (1988). Infrared spectroscopic measurem ents of CO2 and H2O in Juan de Fuca Ridge basaltic glasses. Earth and Planetary Science Letters 90, 87 -104. Dixon, J. E., Stolper, E. M. & Holloway, J. R. (1995a). An Experimental Study of Water and Carbon Dioxide Solubilities in Mid-Ocean Ridge Basaltic Liquids. Part I: Calibration and Solubility Models. Journal of Petrology 36, 1607 -1631. Dixon, J. E., Stolper, E. M. & Holloway, J. R. (1995b). An Experimental Study of Water and Carbon Dioxide Solubilities in Mid-Ocean Ridge Basaltic Liquids. Part II: Ap plications to Degassing. Journal of Petrology 36, 1633 -1646. Donelly, K. E., Goldstein, S. J., Langmuir, C. H. & Spiegelman, M. (2004). Origin of enriched ocean ridge basalt and implications for mantle dynamics. Earth and Planetary Science Letters 226, 34 7 -366. Dunn, R. A., Toomey, D. R., Detrick, R. S. & Wilcock, W. S. D. (2001). Continuous Mantle Melt Supply beneath an Overlapping Spreading Center on the East Pacific Rise. Science 291, 1955-1958. Eiler, J. M. (2001). Oxygen Isotope Variations of Basalt ic Lavas and Upper Mantle Rocks. Reviews in Mineraology and Geochemistry 43, 319364. Eiler, J. M., Farley, K. A., Valley, J., Hofmann, A. W. & Stolper, E. M. (1996). Oxygen isotope constraints on the sources of Hawaiian volcanism. Earth and Planetary Sci ence Letters 144, 453-468. Embley, R. W., Chadwick, W. W. J., Clague, D. A. & Stakes, D. S. (1999). 1998 eruption of Axial Volcano: Multibeam anomalies and seafloor observations. Geophysical Research Letters 26, 3425 -3428. Embley, R. W., Chadwick, W. W. J., Perfit, M. R. & Baker, E. T. (1991). Geology of the northern Cleft segment, Juan de Fuca Ridge: Recent lava flows, seafloor spreading, and the formation of megaplumes. Geology 19, 771 -775. Embley, R. W., Jonasson, I. R., Perfit, M. R., Franklin, J. M ., Tivey, M. A., Malahoff, A., Smith, M. F. & Francis, T. J. G. (1988). Submersible investigations of an extinct hydrothermal system on the Galapagos Ridge: Sulfide mound, stockwork zone, and differentiated lavas. Canadian Mineralogy 26, 1 -35.

PAGE 169

169 Embley, R. W. & Wilson, D. (1992). Morphology of the Blanco Transform Fault zone -NE Pacific: Implications for its tectonic evolution. Marine Geophysical Research 14, 2545. Fornari, D. J. (2003). A new deep -sea towed digital camera and multi -rock coring system. EOS, Transactions American Geophysical Union 84, 69-76. Fornari, D. J., Perfit, M. R., Malahoff, A. & Embley, R. W. (1983). Geochemical studies of abyssal lavas recovered by DSRV Alvin from the Eastern Galapagos Rift, Inca Transform, and Ecuador Rift 1. Major element variation in natural glasses and spatial distribution of lavas. Journal of Geophysical Research 88, 1015910529. Fundis, A., Soule, A. S., Fornari, D. J. & Perfit, M. R. (2010). Paving the seafloor: volcanic emplacement processes during the 2005-06 eruption at teh fast spreading East Pacific Rise, 950'N. Geochemistry Geophysics Geosystems in press. Garcia, M. O., Muenow, D. W., Aggrey, K. E. & O'Neil, J. R. (1998). Major element, volatile and stable isotope geochemistry of Hawaiian submarine tholeiitic glasses. Journal of Geophysical Research 94, 10,525510,538. Gautason, B. & Muehlenbach, K. (1998). Oxygen isotopic fluxes associated with high temperature processes in the rift zones of Iceland. Chemical Geology 145, 275286. Ghiorso, M. S. & Sa ck, R. O. (1995). Chemical Mass Transfer in Magmatic Processes. IV. A Revised and Internally Consistent Thermodynamic Model for the Interpolation and Extrapolation of LiquidSolid Equilibria in Magmatic Systems at Elevated Temperatures and Pressures. Contr ibutions to Mineralogy and Petrology 119, 197-212. Gillis, K. M. (2008). The roof of an axial magma chamber: A hornfelsic heat exchanger. Geology 36, 299-302. Gillis, K. M. & Coogan, L. A. (2002). Anatectic Migmatites from the Roof of an Ocean Ridge Magm a Chamber. Journal of Petrology 43, 2075-2095. Gillis, K. M., Coogan, L. A. & Chaussidon, M. (2003). Volatile element (B, Cl, F) behaviour in the roof of an axial magma chamber from the East Pacific Rise. Earth and Planetary Science Letters 213, 447 -462. Gillis, K. M., Muehlenbach, K., Stewart, M., Gleeson, T. & Karson, J. A. (2001). Fluid flow patterns in fast spreading East Pacific Rise crust exposed at Hess Deep. Journal of Geophysical Research 106, 2631126329.

PAGE 170

170 Gillis, K. M. & Roberts, M. D. (1999). Cracking at the magma-hydrothermal transition: Evidence from the Troodos ophiolite, Cyprus. Earth and Planetary Science Letters 169, 227244. Goss, A., Perfit, M., Ridley, W. I., Rubin, K. H., Kamenov, G., Soule, A. S., Fundis, A. & Fornari, D. J. (2010). Geochemistry of lavas from the 20052006 eruption at the East Pacific Rise, 946'N 956'N: Implications for ridge crest plumbing and decadal changes in magma chamber compositions. Geochemistry Geophysics Geosystems 11, 1-35. Grimes, C. B., Ushikubo, T., John, B. E. & Valley, J. (2010a). Uniformly mantle like Contributions to Mineralogy and Petrology in press, 121. at Oman: evidence for remelting. Geochemica et Cosmochimica Acta 74, A355. Grove, T. L., Kinzler, R. J. & Bryan, W. B. (1992). Fractionation of midocean ridge basalts. Geophysical Monograph 71, 281-310. Haase, K. M., Stroncik, N. A., Hkinian, R. & Sto ffers, P. (2005). Nbdepleted andesites from the Pacific -Antarctic Rise as analogs for early continental crust. Geology 33, 921-924. Hacker, B. R. (1990). Amphibolite-facies to granulite-facies reactions in experimentally deformed, unpowered amphibolite. American Mineralogist 75, 13491361. Hanson, G. N. (1977). Geochemical evolution of the suboceanic mantle. Journal of Geological Society of London 134, 235 -260. Harding, A. J., Kent, A. J. R. & Orcutt, J. A. (1993). A multichannel seismic investigation o f upper crustal structure at 9N on the East Pacific Rise: Implications for crustal accretion. Journal of Geophysical Research 98, 1392513944. Hart, S. R., Schilling, J. G. & Powell, J. L. (1973). Basalts from Iceland and along Reykjanes Ridge -Sr isot ope geochemistry. Nature 246, 104-107. Hey, R. N., Duennebier, F. K. & Morgan, J. W. (1980). Propagating rifts on mid ocean ridges. Journal of Geophysical Research 85, 36473658. Hey, R. N., Johnson, G. L. & Lowrie, A. (1977). Recent tectonic evolution o f the Galapagos area and plate motions in the East Pacific. Geological Society of America Bulletin 88, 1385-1403.

PAGE 171

171 Johannes, W. & Koepke, J. (2001). Incomplete reaction of plagioclase in experimental dehydration melting of amphibolite. Australian Journal o f Earth Science 48, 581590. Johnson, E. R., Wallace, P., Delgado Granados, H., Manea, V. C., Kent, A. J. R., Bindeman, I. N. & Donegan, C. S. (2009). Subduction-rlated Volatile Recycling and Magma Generation beneath Central Mexico: Insights from Melt Inc lusions, Oxygen Isotopes and Geodynamic Models. Journal of Petrology 50, 17291764. Juster, T. C., Grove, T. L. & Perfit, M. R. (1989). Experimental Constraints on the Generation of FeTi Basalts, Andesites, and Rhyodacites at the Galapagos Spreading Cente r, 85W and 95W. Journal of Geophysical Research 94, 9251 9274. Kamenov, G., Saunders, A. D. & Hames, W. E. (2007). Mafic Magmas as Sources for Gold in Middle Miocene Epithermal Deposits of Northern Great Basin, USA: Evidence from Pb Isotopic Compositions of Native Gold. Economic Geology 102, 11911195. Karsten & Delaney, J. R. (1989). Hot spot -ridge crest convergence in the northeast Pacific. Journal of Geophysical Research 94, 700712. Kelley, D. S. & Delaney, J. R. (1987). Two-phase separation and fr acturing in mid ocean ridge gabbros at temperatures greater than 700C. Earth and Planetary Science Letters 83, 53-66. Kent, A. J. R., Harding, A. J. & Orcutt, J. A. (1993). Distribution of magma beneath the East Pacific Rise near 903'N overlapping spreading center from forward modeling of CDP data. Journal of Geophysical Research 98, 1397113995. Kent, A. J. R., Norman, M. D., Hutcheon, I. D. & Stolper, E. M. (1999). Assimilation of seawater derived components in an oceanic volcano: evidence from matrix glasses and glass inclusions from Loihi seamount, Hawaii. Chemical Geology 156, 299-319. Kent, A. J. R., Peate, D. W., Newman, S., Stolper, E. M. & Pearce, J. A. (2002). Chlorine in submarine glasses from the Lau Basin: seawater contamination and constraints on the composition of slab-derived fluids. Earth and Planetary Science Letters 202, 361-377. Kent, G. M., Singh, S. C., Harding, A. J., Sinha, M. C., Orcutt, J. A., Barton, P. J., White, R. S., Bazin, S., Hobbs, R. W., Tong, C. H. & Pye, J. W. (2000) Evidence from threedimensional seismic reflectivity images for enhanced melt supply beneath mid ocean ridge discontinuities. Nature 406, 614618.

PAGE 172

172 Klein, E. M. (ed.) (2005). Geochemistry of the Igneous Oceanic Crust Oxford: Elsevier Pergamon Klein, E M. & Langmuir, C. H. (1987). Global correlations of ocean ridge basalt chemistry. Journal of Geophysical Research 92, 8089 8115. Koepke, J., Berndt, J., Feig, S. T. & Holtz, F. (2007). The formation of SiO2-rich melts within the deep oceanic crust by hy drous partial melting of gabbros. Contributions to Mineralogy and Petrology 153, 6784. Koepke, J., Feig, S. T., Snow, J. & Freise, M. (2004). Petrogenesis of oceanic plagiogranites by partial melting of gabbros: an experimental study. Contributions to Mi neralogy and Petrology 146, 414432. Kurras, G. J., Fornari, D. J., Edwards, M. F., Perfit, M. R. & Smith, M. C. (2000). Volcanic morphology of the East Pacific Rise crest 9 49' 52'N: Implications for volcanic emplacement processes at fast -spreading mido cean ridges. Marine Geophysical Research 21, 2341. Kvassnes, A. J. S. & Grove, T. L. (2008). How partial melts of mafic lower crust affect ascending magmas. Contributions to Mineralogy and Petrology 156, 4971. Langmuir, C. H. (1989). Geochemical consequences of in situ crystallization. Nature 340, 199-205. Langmuir, C. H., Bender, J. F. & Batiza, R. (1986). Petrological and tectonic segmentation of the East Pacific Rise, 530' -1430'N. Nature 322, 422 -429. Langmuir, C. H., Klein, E. M. & Plank, T. (1992). Petrological Systematics of Mid-Ocean Ridge Basalts: Constraints on Melt Generation Beneath Ocean Ridges. Geophysical Monograph 71 Mantle Flow and Melt Generation at MidOcean Ridges, 183 280. le Roux, P. J., Shirey, S. B., Hauri, E. H., Perfit, M. R & Bender, J. F. (2006). The effects of variable sources, processes and contaminants on the composition of northern EPR MORB (8 10N and 12 -14N): Evidence from volatiles (H20, CO2, S) and Halogens (F, Cl). Earth and Planetary Science Letters 251, 209 231. Macdonald, K. C. & Fox, P. J. (1983). Overlapping spreading centers: new accretion geometry on the East Pacific Rise. Nature 302, 5558. Macdonald, K. C. & Fox, P. J. (1988). Earth and Planetary Science Letters 88, 119.

PAGE 173

173 Macdonald, K. C., Fox, P. J., Per ram, L. J., Eisen, M. F., Haymon, R. M., Miller, S. P., Carbotte, S. M., Cormier, M.-H. & Shor, A. N. (1988). A new view of the mid ocean ridge from the behaviour of ridge axis discontinuities. Nature 335, 217225. Macdonald, K. C., Scheirer, D. S. & Carb otte, S. M. (1991). Mid-Ocean Ridges: Discontinuities, Segments, and Giant Cracks. Science 253, 986 -994. Maclennan, J. (2008). The Supply of Heat to Mid -Ocean Ridges by Crystallization and Cooling of Mantle Melts. Geophysical Monograph 178, 45-73. Mandev ille, C. W., Webster, J. D., Rutherford, M. J., Taylor, B. E., Timbal, A. & Faure, K. (2002). Determination of molar absorptivities for infrared absorption bands of H2O in andesitic glass. American Mineralogist 87, 813821. Matsuhisa, Y., Matsubaya, O. & Sakai, H. (1973). Oxygen Isotope Variations in Magmatic Differentiation Processes of the Volcanic Rocks in Japan. Contributions to Mineralogy and Petrology 39, 277288. McBirney, A. R. (ed.) (1993). Differentiated rocks of the Galapagos hotspot : Geological Society of London Special Publications Michael, P. & Cornell, W. C. (1998). Influence of spreading rate and magma supply on crystallization and assimilation beneath midocean ridges: evidence from chlorine and major element chemistry of midocean ridge basalts. Journal of Geophysical Research 103, 18325 -18356. Michael, P. & Schilling, J. G. (1989). Chlorine in mid ocean ridge magmas: Evidence for assimilation of seawater -influenced components. Geochimica Cosmochemica Acta 53, 3131 -3143. Michard, A. & A lbarde, F. (1986). The REE content of some hydrothermal Chemical Geology 55, 5160. Muehlenbach, K. & Byerly, G. R. (1982). 18O -Enrichment of Silicic Magmas Caused by Crystal Fractionation at the Galapagos Spreading Center. Contributions to Minera logy and Petrology 79, 76 -79. Muehlenbach, K. & Clayton, R. N. (1972). Oxygen isotope studies of fresh and weathered submarine basalts. Canadian Journal of Earth Science 2, 172-184. Nakamura, K., Kato, Y., Tamaki, K. & Ishii, T. (2007). Geochemistry of h ydrothermally altered basaltic rocks from the Southwest Indian Ridge near the Rodriquez Triple Junction. Marine Geology 239, 125141.

PAGE 174

174 Natland, J. H., Langmuir, C. H., Bender, J. F., Batiza, R. & Hopson, C. (1986). Petrologic systematics in the vicinity of the 9 degrees N nontransform offset, East Pacific Rise. EOS, Transactions American Geophysical Union 67, 1254. Natland, J. H. & MacDougall, J. D. (1986). Parental abyssal tholeiites and alkali basalts at the East Pacific Rise near 9 degrees N and the Siq ueiros fracture zone. EOS, Transactions American Geophysical Union 67, 410411. Newman, S. & Lowenstern, J. B. (2002). VOLATILECALC: a silicate melt H2O CO2 solution model written in Visual Basic for excel. Computers and Geosciences 28, 597-604. Nicholso n, H., Condomines, M., Fitton, J. G., Fallick, A. E., Gronvold, K. & Rogers, G. (1991). Geochemical and Isotopic Evidence for Crustal Assimilation Beneath Krafla, Iceland. Journal of Petrology 32, 1005 1020. Niu, Y., Gilmore, T., Mackie, S., Greig, A. & B ach, W. (2002). Mineral chemistry, wholerock compositions, and petrogenesis of Leg 176 gabbros: data and discussion. College Station: Ocean Drilling Program. Nunnery, A., Klein, E. M., Perfit, M., White, R. S., Mason, J. L. & Zaino, A. J. (2008). Correla tion of Seafloor Surface Features and Underlying Melt Bodies at the 9 N Overlapping Spreading Center, East Pacific Rise. EOS, Transactions American Geophysical Union 89(53), Fall Meet. Suppl., Abst V53D -03. O'Hara, M. J. (1977). Geochemical evolution dur ing fractional crystallization of a periodically refilled magma chamber. Nature 266, 503507. O'Nions, R. K. & Gronvold, K. (1973). Petrogenetic relationships of acid and basic rocks in iceland: Sr -isotopes and reareearth elements in late and postglacial volcanics. Earth and Planetary Science Letters 19, 397409. Pedersen, R. B. & Malpas, J. (1984). The Origin of oceanic plagiogranites from the Karmoy ophiolite, western Norway. Contributions to Mineralogy and Petrology 88, 36-52. Perfit, M., Ridley, W. I. & Cotsonika, L. (in prep). Juan de Fuca Dacites. Perfit, M. R. & Chadwick, W. W. J. (1998). Magmatism at Mid -Ocean Ridges: Constraints from Volcanological and Geochemical Investigations. Geophysical Monograph 106 Faulting and Magmatism at Mid-Ocean Ri dges, 59 -115. Perfit, M. R. & Fornari, D. J. (1983). Geochemical studies of abyssal lavas recovered by DSRV Alvin from Eastern Galapagos Rift, Inca Transform, and Ecuador Rift 2. Phase chemistry and crystallization history. Journal of Geophysical Research 88, 10530-10550.

PAGE 175

175 Perfit, M. R., Fornari, D. J., Malahoff, A. & Embley, R. W. (1983). Geochemical Studies of Abyssal Lavas Recovered by DSRV Alvin From Eastern Galapagos Rift, Inca Transform, and Ecuador Rift 3. Trace Element Abundances and Petrogenesis. Journal of Geophysical Research 88, 10,551-510,572. Perfit, M. R., Fornari, D. J., Smith, M. C., Bender, J. F., Langmuir, C. H. & Hayman, N. W. (1994). Small -scale spatial and temporal variations in mid-icean ridge crest magmatic processes. Geology 22, 37 5 -379. Perfit, M. R., Ridley, W. I. & Jonasson, I. R. (1999). Geologic, petrologic and geochemical relationships between magmatism and massive sulfide mineralization along the eastern Galapagos Spreading Center. Reviews in economic geology 8, 75 99. Perf it, M. R., Schmidt, A. K., Ridley, W. I., Rubin, K. H. & Valley, J. (2007). Petrogenesis of dacites from the southern Juan de Fuca Ridge. Goldschmidt Conf. Abstracts, Geochimica et Cosmochimica Acta 72, A736. Pollard, D. D. & Sydin, A. (1984). Propagation and Linkage of Oceanic Ridge Segments. Journal of Geophysical Research 89, 10,017010,028. Pollock, M. A., Klein, E. M., Karson, J. A. & Tivey, M. A. (2005). Temporal and spatial variability in the composition of lavas exposed along the Western Blanco Tr ansform Fault. Geochemistry Geophysics Geosystems 6, 117. Rapp, R. P., Watson, E. B. & Miller, C. F. (1991). Partial melting of amphibolite/eclogite and the origin of Archean trondhjemites and tonalites. Precambrian Research 51, 1 -25. Regelous, M., Niu, Y., Went, J. I., Batiza, R., Greig, A. & Collerson, K. D. (1999). Variations in geochemistry of magmatism on the East Pacific Rise at 1030' N since 800 ka. Earth and Planetary Science Letters 168, 45-63. Reynolds, J., R. (1995). Segment -scale systematic of mid ocean Ridge magmatism and geochemistry. Palisades: Columbia University, 483. Reynolds, J., R. & Langmuir, C. H. (1997). Petrological systematics of the Mid-Atlantic Ridge south of Kane: Implications for ocean crust formation. Journal of Geophysica l Research 102, 1491514946. Reynolds, J., R., Langmuir, C. H., Bender, J. F., Kastens, K. A. & Ryan, W. B. F. (1992). Spatial and temporal variability in the geochemistry of basalts from the East Pacific Rise. Nature 359, 493 -499.

PAGE 176

176 Rubin, K. H. & Sinton, J. M. (2007). Inferences on midocean ridge thermal and magmatic structure from MORB compositions. Earth and Planetary Science Letters 260, 257276. Saal, A. E., Hauri, E. H., Langmuir, C. H. & Perfit, M. (2002). Vapour undersaturation in primitive mid o cean -ridge basalt and the volatile content of Earth's upper mantle. Nature 419, 451455. Schiffman, P., Zierenber, R., Chadwick, W. W. J., Clague, D. A. & Lowenstern, J. B. (2010). Contanination of basaltic lava by seawater: Evidence found in a lava pilla r from Axial Seamount, Juan de Fuca Ridge. Geochemistry Geophysics Geosystems 11, 1-12. Schmitt, A. K., Perfit, M., Rubin, K. H., Stockli, D. F., Smith, M. C., Cotsonika, L., Zellmer, G. F., Ridley, W. I. & Lovera, O. M. (submitted). Zircon in Evolved MidOcean Ridge Lavas Constrains Fast Cooling Rates of Oceanic Lower Crust. Nature. Sempere, J. -C. & Macdonald, K. C. (1986). Deep-Tow Studies Of The Overlapping Spreading Centers at 903' N On The East Pacific Rise. Tectonics 5, 881-900. Sempre, J. -C. & Mac donald, K. C. (1986). Deep -Tow Studies Of The Overlapping Spreading Centers at 903' N On The East Pacific Rise. Tectonics 5, 881-900. Sigurdsson, H. (1977). Generation of Icelandic rhyolites by melting of plagiogranites in the oceanic layer. Nature 269, 25 28. Sigurdsson, H. & Sparks, R. S. J. (1981). Petrology of rhyolitic and mixed magma ejecta from the 1875 eruption of Askja, Iceland. Journal of Petrology 22, 41 84. Sims, K. W. W., Blichert-Toft, J., Fornari, D. J., Perfit, M. R., Goldstein, S. J., Jo hnson, P., De Paolo, D. J. & Michael, P. (2003). Abberant Youth: Chemical and isotopic constraints on the young off axis lavas of the East Pacific Rise. Geochemistry Geophysics Geosystems 4, 8621. Sims, K. W. W., Goldstein, S. J., Blichert-Toft, J., Perfi t, M. R., Kelemen, P. B., Fornari, D. J., Michael, P., Murrell, M. T., Hart, S. R., DePaolo, D. J., Layne, G., Ball, L., Jull, M. & Bender, J. F. (2002). Chemical and isotopic constraints on the generation and transport of magma beneath the East Pacific Ri se. Geochimica et Cosmochimica Acta 66, 3481 -3504. Singh, S. C., Harding, A. J., Kent, G. M., Sinha, M. C., Combier, V., Bazin, S., Tong, C. H., Pye, J. W., Barton, P. J., Hobbs, R. W., White, R. S. & Orcutt, J. A. (2006). Seismic reflection images of the Moho underlying melt sills at the East Pacific Rise. Nature 442, 14.

PAGE 177

177 Sinton, J. M., Ford, L. L., Chappell, B. & McCulloch, M. T. (2003). Magma Genesis and Mantle Heterogeneity in the Manus Back Arc Basin, Paupa New Guinea. Journal of Petrology 44, 1591 95. Sinton, J. M., Smaglik, S. M. & Mahoney, J. J. (1991). Magmatic processes at superfast spreading midocean ridges: glass compositional variations along the East Pacific Rise 13 -23S. Journal of Geophysical Research 96, 6133 -6155. Sinton, J. M., Wils on, D., Christie, D. M., Hey, R. N. & Delaney, J. R. (1983). Petrological consequences of rift propagation on oceanic spreading ridges. Earth and Planetary Science Letters 62, 193207. Smith, M. C., Perfit, M. R., Fornari, D. J., Ridley, W. I., Edwards, M H., Kurras, G. J. & Von Damm, K. L. (2001). Magmatic processes and segmentation at a fast spreading midocean ridge: Detailed investigation of an axial discontinuity on the East Pacific Rise crest at 937'N. Geochemistry Geophysics Geosystems 2, 132. Smith, M. C., Perfit, M. R. & Jonasson, I. R. (1994). Petrology and geochemsitry of basalts from the Southern Juan de Fuca Ridge: Controls on the spatial and temporal evolution of midocean ridge basalts. Journal of Geophysical Research 99, 4787-4812. Soul e, A. S., Fornari, D. J., Perfit, M. R., Tivey, M. A., Ridley, W. I. & Schouten, H. (2005). Channelized lava flows at the East Pacific Rise crest 910 N: The importance of off axis lava transport in developing the architecture of young oceanic crust. Geoch emistry Geophysics Geosystems 6, 128. Sparks, J. W. (1995). Geochemistry of the Lower Sheeted Dike Complex, Hole 504B, Leg 140. Proceedings of the Ocean Drilling Program, Scientific Results 137/140, 8197. Stakes, D. S., Perfit, M. R., Tivey, M. A., Car ess, D. W., Ramirez, T. M. & Maher, N. (2006). The Cleft revealed: Geologic, magnetic, and morphologic evidence for construction of upper oceanic crust along the southern Juan de Fuca Ridge. Geochemistry Geophysics Geosystems 7, 132. Sun, S. & McDonough, W. F. (eds.) (1989). Chemical and Isotopic Systematics of oceanic basalts: implications for Mantle Composition and Processes : Geological Society of London, Special Publications Sun, S. S., Tatsumoto, M. & Schilling, J. G. (1975). Mantle Plume Mixing Along the Reykjanes Ridge Axis: Lead Isotopic Evidence. Science 190, 143 147.

PAGE 178

178 Sun, W. D., Binns, R. A., Fan, A. C., Kamenetsky, V. S., Wysoczanski, R., Wei, G. J., Hu, Y. H. & Arculus, R. J. (2007). Chlorine in submarine volcanic glasses from the eastern Manus basin. Geochemica et Cosmochimica Acta 71, 15421552. Taylor, H. P. J. (1968). The oxygen isotope geochemistry of igneous rocks. Contributions to Mineralogy and Petrology 19, 171. Taylor, H. P. J. (1980). The effects of assimilation of country rocks b y magmas on 18O/16O and 87Sr/86Sr systematics in igneous rocks. Earth and Planetary Science Letters 47, 243. Thy, P., Lesher, C. E. & Mayfield, J. D. (1999). 9. Low -Pressure Melting Studies of Basalts and Basaltic Andesites From The Southeast Greenland Continental Margin And The Origin Of Dacites At Site 917. Proceedings of the Ocean Drilling Program, Scientific Results 163, 95-112. Tong, C. H., Pye, J. W., Barton, P. J., White, R. S., Sinha, M. C., Singh, S. C., Hobbs, R. W., Bazin, S., Harding, A. J., K ent, G. M. & Orcutt, J. A. (2002). Asymmetric melt sills and upper crustal construction beneath overlapping ridge segments: Implications for the development of melt sills and ridge crests. Geology 30, 8386. Toomey, D. R., Jousselin, D., Dunn, R. A., Wilc ock, W. S. D. & Detrick, R. S. (2007). Skew of mantle upwelling beneath the East Pacific Rise governs segmentation. Nature 446, 409414. Unni, C. K. & Schilling, J. G. (1978). Cl and Br degassing by volcanism along the Reykjanes Ridge and Iceland. Nature 272, 19-23. Wanless, D., Perfit, M., Ridley, W. I. & Klein, E. M. (2008). Origin of Dacites on Mid Ocean Ridges. EOS, Transactions American Geophysical Union 89(53), Fall Meeting Supplementary, Abstract V41B -2084. Wanless, D., Perfit, M., Ridley, W. I., Klein, E. M., Wallace, P., Valley, J. & Grimes, C. (submitted). Role of Assimilation in the Petrogenesis of Lavas on Mid -Ocean Ridges inferred from Cl, H2O, CO2 and Oxygen Isotope Analyses. Wanless, D., Perfit, M., Ridley, W. I., Klein, E. M. & White, R. S. (in prep). 9N Overlapping Spreading Center: Magma Genesis and Ridge Segmentation. Geochemistry Geophysics Geosystems. Wanless, D., Perfit, M. R., Ridley, W. I., Wallace, P., Valley, J., Grimes, C. & Klein, E. M. (2009). Geochemical Evidence for Crustal Assimilation at Mid-Ocean Ridges Using Major and Trace Elements, Volatiles and Oxygen Isotopes. EOS, Transactions American Geophysical Union, Fall Meeting Supplementary, Abstract V51A 1665.

PAGE 179

179 Wanless, V. D., Perfit, M., Ridley, W. I. & Klein, E. M. (accep ted). Dacite Petrogenesis on Mid-Ocean Ridges: Evidence for Crustal Melting and Assimilation. Journal of Petrology Accepted. Watson, E. B. (1979). Apatite saturation in basic to intermediate magmas. Geophysical Research Letters 6, 937-940. Webster, J. D. Kinzler, R. J. & Mathez, E. A. (1999). Chloride and water solubility in basalts and andesite melts and implications for magmatic degassing. Geochemica et Cosmochimica Acta 63, 729738. West, M., Menke, W., Tolstoy, M., Webb, S. & Sohn, R. (2001). Magma storage beneath Axial volcano on the Juan de Fuca midocean ridge. Nature 413, 833836. White, S. M., Haymon, R. M., Fornari, D. J., Perfit, M. R. & Macdonald, K. C. (2002). Correlation between tectonic and volcanic segmentation at fast -spreading ridges: evidence from the distribution of volcanic structures and lava flow morphology along East Pacific Rise at 9 10N. Journal of Geophysical Research 107. White, S. M., Mason, J. L., Macdonald, K. C., Perfit, M. R., Wanless, V. D. & Klein, E. M. (2009). Si gnificance of widespread low effusion rate eruptions over the past two million years for delivery of magma to the overlapping spreading centers at 9N East Pacific Rise. Earth and Planetary Science Letters 280, 175184. White, W. M. & Schilling, J. G. (19 78). The nature and origin of geochemical variation in Mid -Atlantic Ridge basalts from the Central North Atlantic. Geochemica et Cosmochimica Acta 42, 1501 -1516. Wilson, D. S., Teagle, D. A. H., Alt, J., Banerjee, N. R., Umino, S., Miyashita, S., Acton, G D., Anma, R., Barr, S. R., Belghoul, A., Carlut, J., Christie, D. M., Coggon, R. M., Cooper, K. M., Cordier, C., Crispini, L., Durand, S. R., Einaudi, F., Galli, L., Gao, Y., Geldmacher, J., Gilbert, L. A., Hayman, N. W., Herrero Bervera, E., Hirano, N., Holter, S., Ingle, S., Jiang, S., Kalberkamp, U., Kerneklian, M., Koepke, J., Laverne, C., Vasquez, H. L. L., Maclennan, J., Morgan, S., Neo, N., Nichols, H. J., Park, S.H., Reichow, M. K., Sakuyama, T., Sano, T., Sandwell, R., Scheibner, B., Smith -Duque C. E., Swift, S. A., Tartarottie, P., Tikku, A. A., Tominaga, M., Veloso, E. A., Yamasaki, T., Yamazaki, S. & Ziegler, C. (2006). Drilling to Gabbro in Intact Ocean Crust. Science 312, 10161020. Wolf, M. B. & Wyllie, P. J. (1994). Dehydration-melting o f amphibolite at 10 kbar: the effects of temperature and time. Contributions to Mineralogy and Petrology 115, 369-383.

PAGE 180

180 Zaino, A. J. (2009). Petrology and Mineral Chemistry of the 903'N Overlapping Spreading Center, East Pacific Rise. Department of Earth and Ocean Sciences Durham: Duke Univeristy, 60.

PAGE 181

181 BIOGRAPHICAL SKETCH Virginia Dorsey Wanless was born in Denver, Colorado but grew up in Topeka, Kansas. She earned her Bachelor of Arts in geology from Colgate University in 2001 and her Master of Science from the Department of Geology and Geophysics at the University of Hawai`i in 2005. Interspersed with her education she worked at the Hawaiian Volcano Observatory (HVO), FUGRO Sea Floor Surveys, and as a side scan sonar analyst for the Hawaiian Researc h Mapping Group (HMRG) and the REMUS 6000 team at Woods Hole Oceanographic Institution (WHOI). After completion of her Doctor of Philosophy in the summer of 2010, she began a postdoctoral fellowship at WHOI.