hViE;I.J RETURNING MATERIA£§z
Place in book drop to

LIBRARIES remove this checkout from

.-__ your record. FINES will

 

 

be charged if book is
returned after the date
stamped below.

 

  

52 OM USE o,

 

D
"ma?

 

 

THE PETROGENESIS AND TECTONIC SETTING
OF THE EGYPTIAN ALKALINE COMPLEXES

By
PhiTip Clarence de Gruyter

A THESIS
Submitted to . 1
Michigan State University
in partiai fuifiliment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Geoiogy
1983

- Np.

/3‘/‘- ‘

ABSTRACT

THE PETROGENESIS AND TECTONIC SETTING
OF THE EGYPTIAN ALKALINE COMPLEXES

BY

Philip Clarence de Gruyter

The Egyptian alkaline complexes are similar in compo-
sition to many alkaline suites in other areas, and are con-
sidered to represent highly differentiated, mantle-derived
alkaline basaltic magmas. Low-pressure crystal fractiona-
tion is considered to have produced the initial bulk salic
compositions; magma zoning in the remaining halogen-vola-
tile-rich alkaline melt is considered to account for the
radial arrangement of silica-undersaturated and oversatur-
ated rock-types and the associated isotopic (Lutz, 1979)
and chemical gradients.

The Nubian shield provided a stable crustal setting
which allowed the long-term emplacement and similar quies-
cent differentiation of these melts along the major Pan-
African lineaments throughout portions of the Phanerozoic
eon: 554, 404, 351, 224, 140, and 90 m.y. B.P. These melts
are considered to have been generated in the asthenosphere
in response to shear—heating between the lithosphere and

the mesosphere during periods of major plate accelerations.

Dedicated to
Jacqueline and

Matthew Robert Allan

ii

ACKNOWLEDGMENTS

I thank Tom for his patience, help and friendship--
particularly considering the time and distance associated
with the following de Gruyter space. I regret the delay
and the freeze frame of friends I left behind. Those frames
never come back. I do remember one particular frame of a
different sort that turned up in a slide presentation of
mine (who's chest was that again?). T2 and Mongobongowhere-
didyougetyourcongo-congolas? (This is the only type of
acknowledgments anyone expected from me.)

So Tom...with tipped Stellas, here's to Eastern Desert
days and nights, Aswan gozas, the infamous ascent of Nigrub
El Fogani, the House that Jack Built, Ocum's Razor (heh-heh)
and those power spots that persist in Desert rock and sand.

I wished for one more quarter's worth of more leisurely
talking, writing and enjoying. Mongo and I would have surely
walked at least once across the expanse of the mantle.

To T2 who helped me by example more than he realized
(although I learn slowly). Thanks for friendship and for
finding a woman who will finally make something out of you.

I want to extend special thanks across the many miles

to Ahmed Abdallah Abdel Megid for Desert assistance, smiles

and a open friendliness to a "hkawaga".

iii

 

I'd like to thank John for his tailor-fitting in the
computer and calculation realm. Thanks to Duncan for sedi-
mentary inspiration, encouragement, and I can't forget the
cosmic inspiration as well. Thanks to Bill specifically for
a seminar early in my time at MSU which got me enthused.
Thanks also for his help in improving the following.

Thanks does not cover the patience and hard work that
Jackie has given me: sample sawing and grinding, help with
diagrams, enormous patience in my on/off writing for this
thesis and attempts at articles, and most of all for listen-
ing to my ramblings, and keeping the warmth in our home--
always. Asequim Acundua.

If there is light in this, it is solely due to His

faithfulness to me when I was not to Him.

iv

(\l

TABLE OF CONTENTS

List of Tables
List of Figures

General Introduction

PART I: THE GEOCHEMISTRY AND PETROGENESIS OF
THE EGYPTIAN ALKALINE COMPLEXES
Introduction

Chapter 1: The General Setting and Nature of
the Egyptian Alkaline Complexes

Chapter 2: Petrogenetic Models for Primitive
Alkaline Magmas

1. Origin of Primitive Alkaline Magmas

2. The Differentiation of Primitive Alkaline
Magmas

Chapter 3: The Geochemistry of the Egyptian
Alkaline Complexes

1. Major Elements
A. Major Element Variation Diagrams

2. Trace Elements
A. Rare Earth Elements
B. Other Trace Elements

Chapter 4: The Petrogenesis of Silica-Under-
saturated/Oversaturated Alkaline Complexes

Introduction '

Section I: Metasomatism

Section II: Fractional Crystallization

Introduction

1. Comparison of the Chemistry of the Egyptian
Alkaline Complexes to Other Alkaline Suites

2. Fractional Crystallization in Relation to the

V

Page
ix

x-xiv

13
13

16

21
22
25
33
33
55

59
59
60
64
64

66

Q‘V

§\v

Section

Trace Element Distributions of the Egyptian
Alkaline Complexes

A. Compatible Trace Elements--Gabal Abu
Khruq

B. Incompatible Trace Elements--Gabal Abu
Khruq

C. The Relationship Between Incompatible
and Compatible Trace Elements--Gabal Abu
Khruq

D. Crystal Fractionation Modelling

III: The Silica-Undersaturated/Oversatur-
ated Problem

Introduction

1. The Common Occurrence of Silica-Undersaturated/

Oversaturated Associations

2. The Silica-Undersaturated/Oversaturated Problem

in the Egyptian Alkaline Complexes

A. Major Element Trends in Relation to the
Silica-Undersaturated/Oversaturated Problem
B. Incompatible Trace Element Trends in
Relation to the Silica-Undersaturated/
Oversaturated Problem

C. The Apparent Order of Crystallization
in Relation to the Silica-Undersaturated/
Oversaturated Problem

D. REE Depleted/Enriched Stages in the
Development of Silica-Undersaturated/Over-
saturated Complexes

3. Silica-Undersaturated/Oversaturated Models

4. A Quantitative Approach to Assessing the Nature

of the REE Distributions

Section IV: Thermogravitational Diffusion and

Fluid Effects

Introduction

1. Nature of the Mechanism

vi

75

75

79

85

88

92
92

93

102

103

106

106

110

116

122

143

143
144

2. Evidence in Support of Simultaneous/Overlapping
Development of the Major Rock-Types

3. A Preliminary Thermogravitational Diffusion
Model for Gabal Abu Khruq

PART II: THE TECTONIC SETTING OF THE EGYPTIAN
ALKALINE COMPLEXES

Introduction

Chapter 5: Alkaline Magmatism and Plate Tectonics

Introduction

1. Characteristics of Alkaline Magmatism

2. The Diverse Tectonic Settings of Alkaline
Magmatism

Chapter 6: Models for the Production of Alka—
line Melts

1. Mantle Plumes and Global Mantle Disturbances

2. Lithospheric Flexures

3. Shear Heating

Chapter 7: A Tectonic Model for the Egyptian
Alkaline Province

1. The Tectonic Development of the Eastern Desert
and Adjacent Areas

2. The Episodes of Alkaline Magmatism in Egypt
A. Thermal Setting
B. Tectonic Setting

Appendices

Appendix I: Description of Four Egyptian Alkaline
Complexes

1. Gabal El Naga

2. Gabal Nigrub El Fogani

3. Gabal El Kahfa

4. Gabal Abu Khruq
A. General Petrography of the Rock-Types
at Abu Khruq
B. Summary of the Mineralogy of Abu Khruq

vii

146

150

163
163

164
164
168

172

179
179
182
182

191

191
201
205
206

213
213
215
217
220

220
233

Appendix II: A Crystal Fractionation Model for Gabal

WDWNH

Abu Khruq Based on the Trace Elements
Choice of Parent Melt
First Fractionation Stage
Second Fractionation Stage
Third Fractionation Stage

Fourth Fractionation Stage

Appendix III: Net-Vein Complex at Nigrub E1

Fogani: REE Patterns

Bibliography

viii

241
245
246
247
249
250

252
253

Table

Table

Table

Table

Table

Table

Table

Table

LIST OF TABLES

1: Rare Earth Element and Thorium Data
for the Egyptian Alkaline Complexes

l-A: Average REE and Thorium Concentra-
tions for the Major Rock Types at
Gabal Abu Khruq

2: Other Trace Elements for the Egyptian
Alkaline Complexes

3: First Approximation of Fractional
Crystallization Based on Incompatible
Trace Elements

4: Calculated Relative Bulk Distribution
Coefficients for Gabal Abu Khruq

5: Reciprocal Depletion/Enrichment Rela-
tionships Between the Silica-Undersat-
urated and Oversaturated Rock Series at
G. Abu Khruq

6: Changes in Plate Motion Associated with
Episodes of Alkaline Magmatism

A II 1: A More Detailed Fractional Crystal-
lization Model for Gabal Abu Khruq

ix

42

56-57

89

135

139

207

242-244

 

LIST OF FIGURES

Page

Figure 1: Some of the major alkaline complexes in

the Eastern Desert of Egypt.
Figure 2: Geologic map of Gabal Abu Khruq. 10
Figure 3: Silica-alkali diagrams for several

Egyptian alkaline complexes showing

progressive changes in alkalinity and

silica-saturation with time. 24
Figure 4: AFM ternary diagram of data from Egyp-

tian alkaline complexes containing only

silica-oversaturated rock-types. 26
Figure 5: AFM ternary diagram of data from Egyp-

tian alkaline complexes containing both

silica-undersaturated and oversaturated

rock-types. 27
Figure 6: Calcic-alkali ternary diagram for the

data from the Egyptian alkaline complexes. 28
Figure 7: Alkali-lime index diagram for the data

from the Egyptian alkaline complexes. 29
Figure 8: Major element variation diagrams for

the data from Gabal Abu Khruq. 32
Figure 9: Normalized REE patterns for the gabbros

from Gabal Abu Khruq. 44
Figure 10: Normalized REE patterns for the low-

REE syenites from Gabal Abu Khruq. 45-46
Figure 11: Normalized REE patterns for the high-

REE rock-types from Gabal Abu Khruq. 48-50
Figure 12: Normalized REE patterns for the dike

rocks. 51
Figure 13: Normalized REE patterns for xenoliths. 52

X

 

 

Figure 14: Normalized REE patterns of three

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

samples of country rock which surrounds
Gabal Abu Khruq.

15: SiO2 versus Sm for the rock series at
G. Abu Khruq.

16: AFM ternary diagrams for several well-
known oceanic alkaline suites which are
similar to those of the Egyptian alkaline
complexes, and similar to trends expected
for fractional crystallization.

17: Composite calcic-alkali ternary dia-
gram of several well-known oceanic alka-
line suites which are similar to those
of the Egyptian alkaline complexes, and
similar to trends expected for fractional
crystallization.

18: Silica-alkali diagrams for a selection
of alkaline suites from continental,
oceanic, extensional, compressional, and
intraplate tectonic settings.

19: Log thorium versus log cerium diagrams
for an alkaline suite from the Azores and
G. Abu Khruq, Egypt.

20: Ce/Sm versus Ce diagrams comparing the
REE enrichment and fractionation trends of
well-known oceanic alkaline suites and that
of G. Abu Khruq, Egypt.

21: Cobalt versus thorium (index of crystal-
linity) for G. Abu Khruq.

22: Trace elements with feldspar affinities
versus thorium (index of crystallinity) for
G. Abu Khruq (Sr and Ba).

23: Frequency diagram of thorium concentra-
tions from four Egyptian alkaline complexes.

24: Normalized REE plot of the major

xi

63

63

67

67

70-71

72

74

78

78

80

up“

uh.

\I:

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

rock-types at G. Abu Khruq.

25: Ce/Sm.versus Ce diagrams for some of
the Egyptian alkaline complexes.

26: Sr versus Rb for the rock series at
G. Abu Khruq.

27: Summary of the silica-alkali trends.

28: Summary of the types of Ce/Sm versus
Ce trends which this study noted to
regularly occur in many alkaline suites
throughout the world, including the
Egyptian alkaline complexes.

29: Silica-alkali diagram for the Pajara
alkaline ring complex in the Canary
Islands (Fuerteventura).

30: Normalized REE plot comparing similar
rock-types from Fuerteventura, Canary
Islands and G. Abu Khruq, Egypt.

31: Normative ne and qtz versus MgO
(index of crystallinity) for the rock
series at G. Abu Khruq.

32: REE versus thorium diagrams for G.
Abu Khruq.

33: Normalized REE plot showing the sim-
ilarity between the more mafic, less
incompatible-rich syenites of the sili-
ca-undersaturated and oversaturated rock
series.

34: REE concentration levels (represented
by Ce) in the major rock-types of G. Abu
Khruq arranged according to their appar-
ent orders of intrusion/extrusion.

35: REE depletion in the last phase of
intrusive activity which Balashov (1972)
found to typically occur in many alka-

line complexes.

xii

82

83

87
95

97

100

100

104

108

109

112

115

not

 

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

36: Trace element and MgO trends which are
considered to indicate low-pressure crys-
tal fractionation.

37: Log REE versus Log thorium for G.

E1 Naga.

38: Log REE versus Log thorium for G.
Nigrub E1 Fogani.

39: Log REE versus Log thorium for G.

El Kahfa.

40: Ce/Sm versus Ce for G. E1 Kahfa
and G. Abu Khruq, showing similarity
between these two Egyptian alkaline
complexes.

41: Log REE versus Log thorium for G.

Abu Khruq.

42: A schematic illustration of a
thermogravitational diffusion mechan-
ism in shallow alkaline magma chambers.

43: REE fractionations versus REE enrich—
ments for averages of the major rock-
types at G. Abu Khruq in terms of LREE/
MREE (Ce/Sm) and LREE/HREE (Ce/Lu).

44: Map of the major alkaline complexes
in the Eastern Desert of Egypt, and
the general tectonic style of Afro-Arabia.

45: Map of the general tectonic config-
uration of Africa.

46: Map showing palinspastic reconstruc-
tion of the Red Sea area and the distri-
bution of the Pan-African ultramafic
zones (Ophiolites).

47: Map showing paleopositions of the
major continents during the Triassic,
about 220:20 m.y. B.P.

xiii

120

125

127

130

131

133-134

152

160

167

193

195

200

l‘

Figure 48: A schematic illustration of the devel-

opment of Afro-Arabia. 203
Figure 49: Model of alkaline melt production via

shear heating in the asthenosphere in

response to major shifts in the motion

of the African plate and their ascent

along the Pan-African fractures. 204
Figure A III 1: Normalized REE plot of the mafic

and salic rock-types collected from a

net-vein complex with apparent liquid/

liquid contacts from G. Nigrub El Fogani. 252

xiv

I

IN

(I I
In

A.
'4

GENERAL INTRODUCTION

 

This thesis presents research pertaining to the geo-
chemistry and tectonic setting of alkaline complexes located
in the Eastern Desert of Egypt. The text is divided into
two parts: Part I deals with the geochemistry and petro-
genesis--primari1y based on the rare earth elements and
thorium; Part II discusses the tectonic setting of alkaline
magmatism in general, and relates the Egyptian alkaline
province to the tectonic development of Afro-Arabia.

The field work was conducted in the spring of 1978.
Approximately 290 samples were collected from the four
alkaline complexes visited by this study: Gabal Abu Khrug,
Gabal E1 Kahfa, Gabal El Naga,_and Gabal Nigrub El Fogani.
The radiometric dates for these complexes were not known at
the time of the field work, but were later supplied by
workers at the University of Pennsylvania (see Lutz, 1979;
Serencsits et al., 1981).

About 200 thin sections were prepared and studied. Two
hundred and eleven samples were analyzed for rare earth
elements (REE) and thorium. Forty-five samples were analyzed
out of house for major elements and 15 trace elements. This
study also used major element data from El Ramly et al.,
1969a, 1969b, and 1971, as well as Rb-Sr data collected by

Lutz, 1979.

PART I

THE GEOCHEMISTRY AND PETROGENESIS

OF THE EGYPTIAN ALKALINE COMPLEXES

Introduction

 

The first objective of this study is to investigate
the petrogenesis of some of the alkaline complexes in the
Eastern Desert of Egypt. Chapter One introduces the general
setting of the Egyptian alkaline province, with a brief
description of the petrography and general structure of the
complex which was selected for more detailed study (Gabal
Abu Khruq). A more complete description of the four alkaline
complexes sampled by this study is presented in Appendix I.
The second chapter consists of a discussion of the models
proposed for the origin of alkaline magmas. The third
chapter presents the chemical data on the Egyptian alkaline
complexes. The fourth chapter involves a discussion of
the petrogenesis of these alkaline complexes, focussing on
Gabal Abu Khruq. This involves the REE geochemistry of
these alkaline complexes, as well as a comparison to alka-

line suites in other areas.

V\
n

f‘.‘

3‘;-

I
1

I.

'1'
In

on
n«

\

(“i

P!-

CHAPTER 1

 

THE GENERAL SETTING AND NATURE OF THE

EGYPTIAN ALKALINE COMPLEXES

The Egyptian alkaline province is located between the
Red Sea and the Nile River, predominantly in the southern
portion of the Egyptian Eastern Desert, Figure 1. This
area is part of the exposed crystalline basement of the
Nubian shield, consisting of Precambrian amphibolitic and
granitic gneisses, metavolcanics, metasediments, and an
abundance of Pan-African salic intrusions. Considering the
profusion of alkaline magmatic centers in Egypt, as well
as Afro-Arabia in general, this area may represent one of
the largest alkaline provinces in the world (Vail, 1970,
1976, 1978; E1 Ramly et al., 1971; Baker et al., 1973;
Gass, 1977; Almond, 1977; Shimron, 1975; Hahn et al., 1976).

The Egyptian alkaline complexes are located along the
major lineaments of the Nubian shield (Garson and Krs, 1976:
El Ramly et al., 1969a, 1969b, 1971; Lutz, 1979). (See
Part II). The radiometric dates of these complexes range
over a period of more than 450 m.y., from 554 to 89 m.y.
B.P. (Serencsits et al., 1981; Hashad and E1 Reedy, 1979;
Lutz, 1979). In spite of their ages spanning most of the
Phanerozoic eon, they are chemically and mineralogically
similar to each other as well as numerous alkaline suites
in other areas. The strontium and oxygen isotOpic data

from ten alkaline complexes are consistent with their

3

 
 

6
RED‘

    

   
  

mum: cl mm ,
.mutu
on ma .. I f

.c.u:um

 

        

 

 

N...“ Wm

. man. Couple:

Figure 1.
Some of the major alkaline complexes in the Eastern Desert

of Egypt (after E1 Ramly et al., 1971).

* Refer to Figures 44 and 45 for tectonic map and rela-
tionship of the study area to the African continent.

5

origin as mantle-derived melts with no significant crustal
contamination (Lutz, 1979; e.g. initial Sr87/86=.703).

The Egyptian alkaline complexes may be generally
described as excavated, volcanic structures with a wide
variety of sodic-alkaline plutonic and effusive rock-types.
The radiometric dates and structural relationships of these
different rock-types indicate that each complex represents
a single discrete magmatic event and the rock—types do not
represent separate and/or unrelated magmatic events. The
complexes sampled by this study (Gabal El Naga, Gabal
Nigrub E1 Fogani, Gabal El Kahfa, and Gabal Abu Khruq) are
dominated by sodic, miarolitic, silica-undersaturated and
oversaturated syenites which are compositionally near the
agpaitic-miaskitic boundary. Other rock-types include
gabbros, alkaline trachytes and rhyolites, and a wide
variety of dike rocks. These four alkaline complexes
represent part of the two youngest episodes of alkaline
magmatism in Egypt which are known to have formed major
alkaline complexes. The older episode (G. E1 Naga and G.
Nigrub E1 Fogani) occurred during the Late Jurassic time;
the younger episode (G. E1 Kahfa and G. Abu Khruq) occurred
during the Late Cretaceous time (Serencsits et al., 1981;
Lutz, 1979). All four complexes contain both silica-under-
saturated and oversaturated rock—types. The older complexes

which were not sampled by this study, tend to consist of

only silica-oversaturated rock-types (Lutz, 1979).

6

All four complexes of this study are well-developed
ring structures, rather than isometric masses or poorly-
developed ring structures. Consistently, the silica-over-
saturated syenites form the outer portions of the complexes,
while the silica-undersaturated syenites are found in the
central portions. The remains of the overlying volcanic
cones tend to consist of evolved, siliceous and altered
flows. Based on limited exposures and comparisons to similar
alkaline complexes in other areas,gabbros and ultramafic
cumulates are inferred to form an underlying substratum to
these salic ring structures. Throughout these alkaline com-
plexes there are examples of gradational boundaries between
some of the coarse—grained rock-types with their correspond-
ing chemical and mineralogic gradients. There are also many
examples in which the boundaries between rock-types are
obscured by faults, alteration zones, dikes, erosion, or
colluvium. Overall, however, there is a scarcity of sharp
contacts between the coarse-grained rock-types. As will be
discussed, these complexes may represent a magma (or pulses
of a magma) which contained a radial compositional gradient
(i.e. liquid zoning). Such an approach is further supported
by similar alkaline complexes in other areas which display
more intact, unfaulted gradational zonation (e.g. Pankhurst
et al., 1976).

Gabal Abu Khruq represents the largest alkaline complex
of this study. It is the most studied alkaline complex in

Egypt due to its economic potential, degree of preservation,

Milo

{ti

7
size, and wide range of rock-types (E1 Ramly et al., 1969a,
1971; Serencsits et al., 1981; Lutz, 1979). This complex is
typical for many of the Egyptian alkaline complexes, and
serves as an example of the type of structure and mineralogy
of these alkaline ring complexes.

G. Abu Khruq is a sodic silica-undersaturated/oversatu-
rated, gabbro-syenite-volcanic alkaline complex, Figure 2.
The major rock—types are 1) olivine-rich cumulates (wehrlite),
2) labradorite-rich gabbros, 3) pulaskite syenites (silica-
undersaturated, small proportion of modal nepheline), 4)
alkaline syenites (silica-oversaturated, little or no modal
quartz), 5) quartz syenites (5-25% modal quartzL 6) alkaline
trachytes and rhyolites (silica-oversaturated),7) foyaite
syenites (strongly silica-undersaturated, nepheline and
sodalite-rich). The overall mineralogy of these rock-types is
characterized by a predominance of euhedral laths (If alkali
feldspar in the syenites and volcanics, and labradorite in
the gabbros. This is the "agpaitic order of crystalliza-
tion" common for many alkaline suites (see Kogarko et al.,
1978). Pyroxene is the dominant ferromagnesian mineral
(aegerine-augite in the syenites and volcanics; diOpside in
the gabbros). Nepheline and sodalite are abundant in the
central silica-undersaturated syenites (foyaites): quartz
(both primary and secondary) becomes progressively more
abundant in the outer perimeters of the complex. The
syenites are dominantly agpaitic, but also range into the

miaskitic compositions (see Sdrensen, 1974, p. 6).

$4

'1' '1.

i?

M'

(J

(I)

(I)

(I;

U)

8

E1 Ramly et al. (1969a) suggest that the original size
of the volcanic cone at Abu Khruq was about 9 kilometers in
diameter and 1.5 kilometers in height. The present complex
consists of a large discontinuous outer ring and an inner
ring with poorly-defined stock-like bodies of syenite. Two
major fault systems dissect the complex: one runs NNW
through the western portion of the complex, the other runs
approximately NW-SE through the inner and outer rings of
the complex, Figure 2. These fault systems are inferred to
be part of the major lineaments of the Nubian shield (see
Part II). Conical faults are inferred to exist in the low-
lying, flat wadis between the rings. These local collapse
surfaces probably occurred during and after the latter
stages of crystallization of the melt. The surrounding
country rock is made up of amphibole schists and gneisses,
with a large granitic body directly against the eastern side

of the complex.

The increasing silica content from the center of the
complex to the margin, is one of the most salient features
of Abu Khruq, as well as the other complexes visited. The
central foyaites are highly silica-deficient; the syenites
surrounding the foyaites tend to be slightly silica-under-
saturated (pulaskites). The southeastern portion of the
outer ring is also composed of pulaskites and foyaitic
syenites. The slightly silica-oversaturated syenites
(alkaline syenites) tend to be found in the outer portions

of the inner ring and portions of the outer ring. Other

Figure 2.
Geologic map of Gabal Abu Khruq (revised from El Ramly

et. al., 1969a and 1971). Numbers denote sample locations.
Sample numbers with prime (') symbol are the "Eg-" series;

sample numbers without the prime symbol are the "A-" series.

.771

K. "MIKE, 7’

.."",-v

'7: ‘7‘7 *1 "I A". III;

---—- __f—_.. ,fi._.. _",.__ ' _ _',

 

 

 

 

 

‘ \
’\/",~, \
’l ’\."|/

(\AI'I
\I‘
1 .

 

 

 

 

O

4.
+

Wadi colluvium

Ga bbros

Pulaski“: 3 Alkaline Spain:

Quart: Snnim

 

 

I.
A

L
7 <
( Al

 

 

 

 

 

 

 

 

 

Foyoim Fou It:

Scale: “$0,000

Voice-tic: .—_—_—q
I all.

Figure 2.
Amphibolc which I ”in"

Cream onus:

0"
In!

A.

III

:9!
mu

0 .

ho.

c:-
P:

'l’

‘M
5w

9 0.
up

['1

C“

.‘t.’

11

than the southeastern portion of the outer ring, the quartz
syenites form the bulk of the outer ring with modal quartz
increasing toward the perimeters. There are gradational
boundaries between the foyaites and pulaskites, between the
alkaline syenites and quartz syenites, and although further
sampling should be done, slightly silica-undersaturated and
slightly silica-oversaturated samples were collected from
apparently the same syenite bodies. There is also a lack
of clear, sharp contacts between the syenites and the under-
lying gabbros at Abu Khruq as well as some of the other
<comp1exes visited. Remnants of the volcanic cone cap parts
<3f both the outer and inner rings; a highly altered chimney
facies is exposed on the Abu Khruq peak in the central ring.
.Almost the entire volume of these volcanics appear to be
siliceous, autometasomatized flows.

It is likely that Abu Khruq and at least some of the
crther Egyptian alkaline complexes are underlain by a
sailica-undersaturated ultramafic substratum (olivine and
<11in0pyroxene cumulate phases). Although further investi-
gation is needed (drilling or geOphysical detection), the
existence of such a substratum is indicated by one sample
(If an olivine—rich rock-type at Abu Khruq (wehrlite),
Cilivine—clinOpyroxene xenoliths in possible diatremes at
El Kahfa, and a comparison to similar‘alkaline complexes
in other areas (Upton, 1974; Bridgewater and Harry, 1968;

Gastesi, 1969; Munoz, 1969; Ludden, 1977; Fernandez, 1980).

M'-
bu“

t»
In

'5

R:-
but.

‘0.
din

‘ I:
o...

'v.

 

 

 

12

Due to the gradational boundaries, the partially
concealed substratum, and the faulted pattern of the complex,
a simple order of crystallization/intrusion is rather
obscure. E1 Ramly et al. (1969a) suggest that the order
of occurrence at Abu Khruq was: volcanics, gabbros, alkaline
syenites (which include both silica-undersaturated and
oversaturated varieties), quartz syenites, and finally,
foyaites (in order of older to younger). Such an order
cannot be explained by a simple crystal fractionation
:model. Although this order may not be correct, the appear-
ance of such field relationships is important to note.

For a more complete description of the four alkaline

complexes sampled by this study, refer to Appendix I.

‘V'I

 

I
o...
r...

n,
‘u‘
I.‘

:I.

‘.'l

‘4

 

A.‘
n
5.“

 

 

CHAPTER 2

 

PETROGENETIC MODELS FOR PRIMITIVE ALKALINE MAGMAS

1. Origin of Primitive Alkaline Magmas

The primitive initial strontium ratios (average .703)
and the chemical characteristics of the Egyptian alkaline
complexes are consistent with a mantle origin (see Lutz,
1979). Alkaline basaltic melts, compared to tholeiitic
types, are characteristically enriched in alkalies, minor,
and incompatible trace elements. This enrichment is a
fundamental problem in explaining the origin of primitive
alkaline magmas. Primitive alkaline basalts and their
differentiates from a wide variety of crustal settings have
similar compositional trends (Bowen, 1928; Schwarzer and
rangers, 1974; Tilley, 1958; McBirney and Aoki, 1968; Kay
aand.Gast, 1973; Kogarko, 1977; Miyashiro, 1978; Frey et al.,
11978; Allegre, et al., 1981).

McBirney and Aoki (1968) emphasized the similarities
txetween the alkaline oceanic suite of Tahiti and the contin-
errtal alkaline suites of the Monteregian Hills in Canada.
Kay and Gast (1973) showed that alkaline basaltic rock-types
Often have similar major element, REE, Sr, and Ba patterns.
Miyashiro (1978) reviewed the major element and trace
element characteristics of alkaline volcanics and noted that
similar types occur in all types of crustal settings. In
an integrated trace element and major element model on the

PrOduction of basalts ranging from tholeiites to highly

13

 

 

14

alkaline basaltic varieties, Frey et a1. (1978) noted the
similarity of many alkaline basalts within different tectonic
settings--as Opposed to the variability of the tholeiitic
varieties. Schwarzer and Rogers (1974) compiled a large
number of major element data on alkaline olivine basalts and
their differentiates.‘ They concluded that there were no
systematic differences in the major element characteristics
of alkaline suites from continental, oceanic, and island-arc
tectonic settings. This consistency in the chemistry of
alkaline suites may represent a fairly regular alkaline
Inagmatic process on a global basis. It is likely to involve
a similar source under both continental and oceanic crust
and/or a similar mechanism of enrichment (see Allegre, et al.,
1981).

Experimental and natural data has been used to demon-
strate that primitive alkaline magmas are generated by
small degrees of partial melting of mantle material (Gast,
.1968; Allegre et al., 1977; Frey et al., 1978) at greater
depths than the source regions for tholeiitic magmas (Green
and Ringwood, 1967; Kushiro et al., 1968; Green, 1970;
Vlilkinson, 1974; Frey and Green, 1974; Maaloe and Aoki,
11977). Although small degrees of partial melting of mantle
uuaterial (e.g. 2-15%, Frey et al., 1978) can account for a
Certain amount of their enriched character, petrogenetic
m<3de11ing has caused workers to conclude that the source
regions must also be initially enriched in certain

components (e.g. Kay and Gast, 1973; Shimuzu and Arculus,

 

 

15

1975; Sun and Hanson, 1975; Frey et al., 1978). These
enriched zones in the mantle must be eXplained in order to
truly understand the origin of alkaline magmas.

Recent work has implemented volatile processes within
the mantle in order to account for this enrichment. The
role of Volatiles in highly-evolved alkaline magmas has
been emphasized for a long time (Symth, 1913; Kogarko, 1974).
However, their presence and activity in the mantle is of
current debate and has direct bearing on the origin of
alkaline magmas in the more recent models. One approach
invokes volatiles bound in the mantle's crystallized phases
(e.g. Eggler, 1976; Newton and Sharp, 1975; Wyllie, 1979),
and another approach suggests that they exist as vapor
;phases (e.g. Mysen and Holloway, 1977; Mysen, 1977b).
JRegardless of its form of residence, once heating begins,
the result is some type of metasomatic process. In a com-
;prehensive model on basalt petrogenesis, Frey et a1. (1978)
Ilse a melt or fluid that is possibly supercritically enriched
smith CO2 and H20 to selectively transport the elements which
are most strongly incompatible with the mantle phases into
cxancentrations within the low-velocity zone and the base of
tame lithosphere. This, Frey et a1. (1978) contend, will
lead to the formation of melts with the observed enriched
Characteristics of alkaline magmas.

However, this enrichment mechanism is still uncertain.
(A detailed investigation into the degree of consistency in

the enrichment characteristics of alkaline rocks from

 

16

different areas of the world would aide in determining
whether a regular, reproducible enrichment mechanism seems
likely, or whether several and/or erratic mechanisms cause
different levels of enrichment. This may have great impor-
tance in many alkaline suites--including the Egyptian
alkaline complexes. It is clear that volatile phases exert
a considerable amount of influence on the composition of
the mafic, undifferentiated alkaline melts, and this influ-
ence would tend to become even more critical as the melt
evolves and becomes more concentrated in volatile and alkali
components.

The role of partial melting of a similar, enriched
Inantle source region provides a regular process for the
.initial generation of alkaline basaltic magmas. However,
that only are mafic alkaline rock-types often similar through-
<Jut the world, but given the initial degree of silica-
saturation and alkalinity, their differentiates commonly
follow similar and very specific compositional trends (e.g.

Schwarzer and Rogers, .1974) .

'2. The Differentiation of Primitive Alkaline Magmas

The majority of the exposed rock-types in the Egyptian
alkaline province are highly-differentiated salic varieties.
The predominance of salic varieties is characteristic of
cOntinental alkaline provinces, and is the subject of
c‘<>II.'.~:iderab1e debate. Four fundamental models have been
Proposed for the origin of salic alkaline magmas. One of

these involves a crustal source and is unlikely to yield

 

 

 

 

17

alkaline magmas with primitive characteristics similar to
the Egyptian suites (crustal anatexis and metasomatism:
Bailey, 1964, 1970, 1974; Williams, 1970; MacDonald et al.,
1970). The three remaining models are:

1. Extraction of salic magmas directly from the mantle
(Wright, 1970; Martin, 1974).

2. Fractional crystallization of mantle—derived basaltic
magmas (Bowen, 1937, 1945; Saggerson, 1970; Weaver et al.,
1972; Barberi et al., 1975; Baker et al., 1977; Baker and
Henage, 1977).

3. Liquid fractionation:

a) Liquid immiscibility (Philpotts, 1971, 1972, 1974,

1976; Freestone, 1978; Eby, 1979).

b) Thermogravitational diffusion or magma zoning

(Hildreth, 1977, 1979, 1981; Kogarko et al., 1974;

Shaw, 1974; Shaw H.R., et al., 1976).

The presence of mafic end-members in the Egyptian
alkaline province, and the continuous compositional trends
between the mafic and salic rock-types may be used to dis-
ccnint the model involving the direct extraction of salic

alJcaline magmas from the mantle for the origin of the

E3957ptian alkaline complexes.

As will be discussed in the following chapter, there
5&3 considerable amount of evidence that fractional crystal-
lization of a mantleederived alkaline basaltic parent magma
was the major mechanism responsible for the development of

the salic compositions of the Egyptian alkaline suites.

18

This is based on isotOpic data (Lutz, 1979), field relation-
ships, chemical trends, and a comparison of these trends

to well-studied alkaline suites in which fractional crystal-
lization has been demonstrated.

Since the time of Bowen (1937), fractional crystalliza-
tion in alkaline magmas has often been suggested as the
dominant process of differentiation. Later work by Coombs
(1963), Kuno (1968), and Coombs and Wilkinson (1969)
demonstrated that the compositional variation of many
alkaline suites can be explained by the regular crystal
fractionation of an alkaline basaltic melt. Trace element
data has become widely used for petrogenetic modelling, and
many alkaline suites have been shown to be dominated by
fractional crystallization (e.g. Zielinski and Frey, 1970;
Sigurdsson et al., 1973; Zielinski, 1975; Price and Taylor,
1973; Sun and Hanson, 1976; Rock, 1978; Baker et al., 1977;
Baker and Henage, 1977; White et al., 1979). Regular
fractional crystallization of alkaline basaltic magmas
derived from mantle sources of similar composition and/or
similar enrichment mechanisms, might account for much of
the similarity in the compositional variation of the salic
alkaline rock-types in different parts of the world.

Experimental data and natural studies indicate that
alkaline melt systems in particular, demonstrate chemical
characteristics well-suited for immiscible liquid relation-
ships. However, this process is presently difficult to

detect and verify in slow-cooling magma systems. Although

 

19

this study does not discount the Operation of this mechanism
in the development of the Egyptian alkaline complexes, there
is little data to substantiate it as the dominant process

of differentiation. There is some evidence that indicates
that it may have played a part in attendant mechanisms.
There are net-vein complexes at G. Nigrub El Fogani, for
example, which contain apparent liquid-liquid contacts
between mafic and salic rock-types. The REE relationships
between these rock-types are consistent with some of the
experimental results of liquid immiscible systems (see
Appendix III; Watson, 1975, 1976; Watson and Natlund, 1977;
Ryerson and Hess, 1978).

Although fractional crystallization of an alkaline
basaltic parent melt is considered to have provided the
develOpment of the bulk salic compositions in the Egyptian
alkaline complexes, it cannot account for the relationship
between the different types of salic rocks. The most
evident relationship which cannot be explained by simple
crystal fractionation is the presence of both silica-under-
saturated and oversaturated salic rock-types within many
of the Egyptian alkaline complexes. This study considers
that thermogravitational diffusion and its attendant fluid
effects are responsible for the development of the silica-
oversaturated salic compositions from a silicaeundersaturated
melt system. This thermogravitational diffusion model is
evaluated in the following section and is supported by

field relationships and chemical data. In the next sections

20

it will be argued that the origin of the Egyptian alkaline
complexes consisted of efficient crystal fractionation of
a primitive alkaline basaltic parent magma with the gradual

develOpment of a thermogravitational diffusion mechanism.

CHAPTER 3

 

THE GEOCHEMISTRY OF THE EGYPTIAN ALKALINE COMPLEXES

Introduction

 

This chapter will present major element and trace
element data on the Egyptian alkaline complexes. This
study sampled four of the complexes--all of which are
younger complexes composed of both silica-undersaturated
and oversaturated rock-types. Approximately 290 samples
were collected from the four alkaline complexes visited by
this study. About 200 thin sections were prepared; 145
of these were from G. Abu Khruq. Two hundred and eleven
samples were analyzed for the REE and Th; 100 of these were
from G. Abu Khruq. In addition, 45 samples were analyzed
out of house for major elements and 15 trace elements
(Barringer Magenta Limited); 22 of these samples were from
G. Abu Khruq. This study also uses major element and
chlorine data collected by El Ramly et al. (1969a, 1969b,
1971), and the Rb-Sr data collected by Lutz (1979).

The rare earth elements and thorium were analyzed via
non-destructive gamma-ray spectrometry using instrumental
neutron activation analysis (INAA) as described by Gordon
et a1. (1968). An Ortec MCA 6240A coaxial Ge—(Li) low
energy photon detector was used, with a resolution of 2.08
Kev (FWHM) at 1.33 MeV and a 14.6% efficiency. The samples
were analyzed with about a 5-10% analytical precision.

The USGS and Canadian (Centres for Mineral and Energy

21

22

Technology) standards were used. Values for the concentra-
tions of the standards were obtained from Abbey (1977),
Flannagan (1973) and Duffield et a1. (1977). The most
significant problem in the REE data may be that of Yb. As
can be seen in the normalized REE diagrams, Yb concentra-
tions tend to be anomalous. This may be due to 1) analytical
problems of which peak interference in certain rock-types
may be responsible, or 2) a real Yb anamoly which may be
characteristic of these types of salic alkaline rocks. Very
similar REE distributions with this anamolous Yb pattern
were noted in similar alkaline rock-types from other areas

(e.g. Canary Islands, De Paepe et al., 1971).

1) Major Elements .

A standard way of demonstrating the alkalinity and
course of differentiation of alkaline suites is with the
use of silica-alkali diagrams (MacDonald and Katsura, 1964;
see Miyashiro, 1978). Figure 3 demonstrates the alkaline
nature of the Egyptian alkaline complexes with reference to
their respective ages. These trends are typical of many
alkaline suites in which crystal fractionation is thought
to have been a dominant mechanism of differentiation
(MacDonald, 1974). It is important to point out the changes
in alkalinity and silica—saturation with the ages of the
alkaline complexes. The older complexes are composed of
only silica-oversaturated rock-types of moderate levels of
alkalinity. The younger complexes, on the other hand, are

dominanted by silica-undersaturated rock—types,_but also

23

Figure 3.

Silica-alkali diagrams for several Egyptian alkaline com-
plexes showing progressive changes in alkalinity and
silica-saturation with time. Curved lines represent the
boundaries between sub-alkalic and alkalic rocks (after
Miyashiro, 1978). The chemical data are from this study
and El Ramly et al., 1969a, 1969b, 1971; the radiometric

dates are from Serencsits et al., 1981 and Lutz, 1979.

 

  

 

 

 

Q
3.
8*
b
:30
3‘
.-
3
8
235
K 09
Cif
N
c:
z "‘
”Iiflglllfll
8 ‘o‘upx0000: m
$0.00 00.00 00.00 00.00 00.00
5102 HT PCT
3.
3
:1
0-
:3:
2‘
h-
3
a 5 ° °
2 ° 0: “E 3
‘ 0
O 0‘
N
C
2 3
fl mm,
$0.00 00.00 70.00 00.00 00.00

 

5102 HT PCT -- EL NRGR

Figure 3.

24

N920 0 K20 HT PCT

22,71)!

 

:4)
‘00 .00 70.00 00.00 $00 70.0:

$102 HT PCT -- EL GEZIRR

 

 

8.
8
84
0—
:38
:-
.—
:x
O (D
2:;
x a:
0
*.8
:2“
N
cr
2 8
w mmy
“50.00 00.00 €0.00 10.00 70.00

5102 HT PCT --HISHBEH

3.
:1
3.
8‘
b
U
c.§
:-
p—
3 o
98.
N:-
x S
O o Q
q 0 O
a.
g a
z 0
. a
O

 

41.14:”?
‘0.’ .I o. 0“ I .D

5102 NT PCT -- RBU KHRUO

 

 

3.
a!
3.
I‘
h
:3!
2‘
o—
: E
8
Eis o 0% 9
x I o
O
o 8 @
ca;
N
I
z ' O
m
J 139 y
\.00 00.00 70.00 70.00 00' .00

5102 HT PCT --EL.FOGRNI

 

 

3.
8
g.
.—
U 0
O. 9
:<
.—
3
532
x - D (D B 0
06° 0
* s 909 6
o O
N A
(I 00
2 8
.‘1
w 01...,
$0.00 00.00 00.00 10.00 00' .00

5102 HT PCT -- EL KRHFR

El UNOERSRTURRTED
0 OVERSRTURRTED
A VOLCRNICS

AP”
Uh!

‘ .
i

q. .
‘00‘

:A‘
.9:

"I!
It."

0'-

.
\hh
8‘.

|‘
I~0.

‘0

6.

5“

[u

H—
(I)

Q

a“-

i

25

contain well-differentiated silica-oversaturated rock
members. It appears that there has been a steady increase
in alkalinity and an increasing predominance of silica-
undersaturation through time.

Figures 4 and 5 show the general differentiation trends
for the silica-oversaturated complexes and the silica-
undersaturated complexes, respectively, in terms of AFM
ternary diagrams. The older silica-oversaturated complexes
form a more pronounced iron enrichment trend than the alkali-
enriched, younger silica-undersaturated/oversaturated com-
plexes. However, both trends are similar to those of other
alkaline suites evolving by fractional crystallization (see
Schwarzer and Rogers, 1974 and Coombs, 1963).

Figure 6 demonstrates the calcic-alkali trend for the
Egyptian alkaline complexes in terms of the KZO-NaZO-Cao
ternary diagram. The CaO-rich plagioclase-dominated trend
leading to the high pr0portion of alkali-rich rock members
is characteristic of the fractional crystallization trend
of alkaline suites in general.

Figure 7 shows the CaO/Na20+K20 ratios versus SiO2
for many of the Egyptian alkaline complexes. This demon-
strates that the general alkali-lime index (ALI) is about
51-52--within the range of typical, moderately alkaline

suites (e.g. Petro, Vogel and Wilband, 1979).

A. Major Element Variation Diagrams - Gabal Abu Khruq

The changes in the concentrations of the individual

26

20

30

GO

 

O OVERSHTURRTED
A SRTURFITED

Figure 4.

AFM ternary diagram of data from Egyptian alkaline com-
plexes containing only silica-oversaturated rock-types
(data from El Ramly et al., 1971).

27

 

 

O UNDERSRTURFTTED
A OVERSRTURRTED
+ SRTURRTEU

Figure 5.
AFM ternary diagram of data from Egyptian alkaline com-

plexes containing both silica-undersaturated and over-
saturated rock-types (data from this study and E1 Ramly
et al., 1969a, 1969b, 1971).

28

A

 

 

O A C
r co
A OTHER COHPLEXES
Figure 6.

Calcic-alkali ternary diagram for the data from the
Egyptian alkaline complexes (G. Abu Khruq is denoted
by octagons).

29

3‘. 20

E1 980 KHRJO
1 \ 0 OTHER E0. COHF'LEXES

2.00

2:00
(3

N920 + CRU/Kzn
100
0
I3

20
G
V
G

e
O
3:
O7 m
0 O
O
00

(D 0 e 0 o O
%\O % O 0 o
T __. _—o
0 fig 0
L» fig 3 .
0 00 $.00 00.00

4 - 0 .00 00.00 €5.00 73.00 7k.00 00.00

5102 NT PCT-E0. COMPLEXES

0.40
l

 

 

 

Figure 7.
Alkali-lime index diagram for the data from the Egyptian

alkaline complexes. Alkali-lime index (ALI) is read as
the SiO2 value at the intersection of the data trend and
the line where Na20 + CaO/KZO is equal to one. The ALI
from this diagram is about 51-52.

30

major elements throughout the alkaline rock series at G.
Abu Khruq are shown in Figure 8. Even though these diagrams
provide only very general information on the chemical trends
of the major elements at Abu Khruq, they are useful in
illustrating the highly-evolved nature of the majority of
the exposed rock-types, as well as the relationship between
the silica-undersaturated rock-types (gabbros, pulaskites,
and foyaites), and the silica-oversaturated rock-types
(alkaline syenites, quartz syenites, and volcanics).

The major element oxides which are incorporated into

the more mafic rocks (TiOZ, FeO CaO, MnO, and P205)

total’
display depletion trends which are consistent with crystal
fractionation. The major element oxides which become more
important in the salic stages of differentiation (SiOZ,
A1203, Na20, and K20) display more complicated trends.
These trends consist of diverging patterns in which the
silica-undersaturated syenites are enriched in A1203 and
Nazo relative to the silica-oversaturated rock-types which

are comparatively enriched in SiO2 and FeO ,as well as

total'
lower NaZO/KZO ratios. The sharp inflections in these
trends are considered to reflect the crystallization of the
sodic-alumina phases in the salic stages of differentiation
(alkali feldspar, nepheline, and sodalite). The volcanics
(silica-oversaturated) form well-defined trends which plot
along concentration ranges similar to the silica-over-

saturated syenites. The greater scatter in the syenite

trends (both silica-oversaturated and undersaturated) is

31

Figure 8.

Major element variation diagrams for the data from Gabal
Abu Khruq (data are from this study and E1 Ramly et al.,
1969a). M90 is used as the index of crystallinity.

Solid squares = gabbros (always found on the right-hand

side of the variation diagrams).

Open squares = silica-undersaturated syenites (pulaskites

and foyaites).
Hexagons = silica-oversaturated syenites (alkaline syenites).

Triangles = volcanics.

 

32

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8- 8.
31“ 2‘ .
P— .—
.. mg
a 3" D $00 0. 8" a]
o— “ Q5 E) l '— I
3‘, 3‘3BEP
N ;1 m . m z & D 0
C) --1 E]
S . 2 Q5050 .
(D . C: SIM
E13.00 1200 0200 f. 00 I00 0'. 00 0'. or 7"0:00 1'. 00 2'. 00 3'. 00 a'. 00 0'. 00 0'. 00
M00 HT PCT - RBU KHRUO MOO HT PCT - RBU KHRUO
2‘. . a!)
:3 - :3 D
.. 3 8 a: :8. . -
D— & o O . ‘ A O
3 o P' 8
C38 3 E) I 3 8 o
w
(\I ' I C’ "
c S U I
Z I
8. 8
$700 {.00 {.00 0200 7.00 0'. 00 0'.oc ‘b .00 :2 00 f . 00 0200 a'. 00 0200 0'. 00
M00 HT PCT - RBU KHRUO MOO NT PCT — RBU KHRUO
8 8.
2'1 . ’— 9‘.‘
E o o m
T I m I
L) 8 [D .— g A .
0. .31 3 .. ‘ A
A I
E c E A 3115 0
9 0— 8
D '0" o o c' B
8 fig D I ’-
8. - 8 8.
$700 i200 7.00 $.00 (.00 7.00 T000 (A ‘0100 I200 {.00 0200 0200 (00 (mo
H00 HT PCT - RBU KHRUO M00 HT PCT - RBU KHRUO
"' i-
3 E. 8
A
I
'5 g; I
N 8’4 0 Sup
0 N g D f 0 [D
: 00 m . ' z - .
8m 8.
‘b . 00 3200 0200 {.00 0200 $.00 0'.00 '5700 1'00 £00 0200 0'. 00 £00 7.00
H00 HT PCT - RBU KHRUO H00 NT PCT - RBU KHRUQ
3.
.—
U I
0. z
b 0
3 I
s 5‘ 8 '
I!)
N
“'zgiEL I
.00 R00 0200 0200 0200 £00 0200

Figure 8.

"GO HT PCT - RBU KHRUO

33

attributed to the individual unique compositions often

found in coarse-grained samples. The foyaites (strongly
silica-undersaturated) formed the last stage of crystalliza-
tion at Abu Khruq and plot as a dense cluster of similar
compositions. The large separation between the gabbros

and the more mafic (silica-undersaturated) syenites reflects
the presence of large quantities of alkali feldspar in the
syenites, and the cumulate, labradorite-rich nature of the
gabbros (especially rich in CaO and A1203). The variable

FeO and TiO2 concentrations of the more mafic syenites

total
and gabbros are considered to reflect the larger prOportions
of the iron-titanium oxides in individual samples. The
gabbros and mafic syenites are rich in P205 because they

contain most of the apatite observed in these types of

alkaline suites.

2) Trace Elements
The trace element data consists of predominantly the
REE and thorium concentrations from the four silica-under-
saturated/oversaturated complexes sampled by this study.
There is a limited amount of compatible and other incompatible

trace element data, as well as Rb—Sr data from Lutz (1979).

A. Rare Earth Elements

Table 1 is a compilation of the rare earth element and
thorium data determined in this study for the four alkaline
complexes investigated. Figures 9513 display the normalized

REE patterns for the rocks from G. Abu Khruq.

Table l.

34

Rare Earth Element and Thorium Data
for the Egyptian Alkaline Complexes:
G. Abu Khruq; G. 81 Kahfa:

G. El Naga; G. Nigrub El Fogani.

'Firat value in ppm.
'Second value is chondrite normalized for the REE.

I. Gabal Abu Khruq

 

La

Gabbros
A-l 22.61
(68.51)
A-Sb 42.74
(129.51)
A-Sc 38.09
(115.42)
A-76 7.11
(altered) (21.54)
A-82 48.95
(148.34)
A-83 11.02
(33.38)
A-91a 18.82
(57.01)
A-9lb 19.93
(60.38)
A-103 23.73
(71.90)

Ce

55.77
(63.37)

84.31
(95.80)

80.21
(91.14)

25.85
(29.38)

89.61
(101.83)

34.86
(39.62)

47.36
(53.81)

32.76
(37.23)

48.42
(55.02)

Sm

7.28
(40.20)

9.34
(51.62)

9.08
(50.18)

1.84
(10.19)

11.37
(62.83)

3.46
(19.10)

4.11
(22.67)

4.73
(26.15)

5.25
(28.99)

Eu

3.79
(54.87)

6.35
(92.03)

5.98
(86.72)

2.10
(30.50)

6.55
(94.87)

1.73
(25.04)

2.61
(37.81)

2.87
(41.53)

2.68
(38.79)

Tb

1.32
(28.09)

1.23
(26.22)

0.97
(20.70)

0.38
(8.14)

1.57
(33.33)

0.66
(13.95)

0.54
(11.52)

0.66
(14.00)

0.56
(11.89)

Yb

3.40
(17.01)

2.43
(12.15)

2.87
(14.34)

0.49

30‘3
(17.14)

1.49
(7.47)

1.26
(6.31)

1.47
(7.32)

1.26
(6.28)

Silica-Undersaturated Syenites(F.-foyaite; P.-pulaakite)

A-7 75.29
(228.15)

A-9 82.08
(248.72)

A-lOa 77.08
(233.58)

A-ll 86.19
(261.17)

A-16 47.30
(143.34)

A-19a 132.14
(400.44)

A-20 58.00
(175.75)

A-Zl 72.37
(219.31)

A-30 49.24
(149.21)

A-32a 125.67
(380.82)

A-32c 105.51
(319.74)

A-35 110.40
(334.54)

A-45 23.64
(71.63)

5-50 53.84
(163.17)

A-Sl 43.60
(132.13)

A-56a 113.82
(344.91)

131.41
(149.33)

153.38
(174.29)

149.86
(170.29)

142.41
(161.83)

114.40
(130.00)

199.69
(226.92)

91.43
(103.89)

149.82
(170.25)

88.35
(100.40)

220.87
(250.99)

172.00
(195.45)

176.79
(200.90)

60.62
(68.89)

96.53
(109.69)

71.27
(80.99)

221.69
(251.92)

8.12
(44.87)

11.29
(62.40)

9.76
(53.95)

13.92
(76.91)

11.23
(62.04)

12.06
(66.66)

7.72
(42.65)

10.37
(57.30)

7.31
(40.40)

13.91
(76.86)

13.42
(74.16)

15.90
(87.87)

4.95
(27.35)

6.91
(38.15)

6.14
(33.93)

15.18
(83.85)

0.65
(9.39)

0.75
(10.88)

0.88
(12.83)

3.69
(53.44)

2.97
(42.98)

0.70
(10.15)

0.67
(9.78)

0.82
(11.86)

0.66
(9.58)

1.06
(15.36)

1.24
(18.00)

1.97
(28.56)

0.51
(7.33)

0.68
(9.93)

0.78
(11.25)

1.03
(14.95)

1.09
(23.25)

1.57
(33.31)

1.65
(35.02)

1.51

2.22
(47.14)

1.38
(29.39)

0.87
(18.42)

1.43
(30.50)

1.18
(25.03)

1.93
(41.12)

1.79
(38.13)

2.06
(43.80)

0.94
(19.94)

0.73
(15.57)

0.73
(15.48)

1.78
(37.98)

4.96
(24.79)

10.22
(51.12)

5.98
(29.88)

6.10

8.86
(44.32)

8.31
(41.53)

4.58
(22.90)

4.36
(21.82)

2.14
(10.72)

10.64
(53.18)

6.58
(32.90)

4.72
(23.61)

2.06
(10.28)

4.00
(19.98)

3.27
(16.34)

8.24
(41.19)

Lu

0.55
(16.15)

0.29
(8.54)

0.47
(13.68)

0.14
(4.18)

0.44
(12.99)

0.19
(5.69)

0.22
(6.34)

0.17
(4.92)

0.30
(8.85)

0.51
(15.00)

0.94
(27.74)

0.65
(19.12)

0.50
(14.60)

0.72
(21.04)

0.82
(24.13)

0.50
(14.81)

00“
(13.08)

0.40
(11.66)

0.92
(27.09)

1.00
(29.30)

0.67
(19.74)

0.38
(11.09)

0.44
(13.07)

0.28
(8.17)

0.78
(23.02)

Th

3.71

2.54

3.15

1.83

15.50

23.47

15.50

9.21

10.65

38.00

10.41

10.75

8.54

37.29

32.53

18.47

6.32

11.44

5.60

24.15

Hi:
H):
H):

Hit

)3"
Pg.

6.
8‘ ,:

Table 1 (cont'd).

A-56b
A-59a
A-S9b
A-60a
A-62
A-65
A—66a
A-86
A-87
A-88

39-22

La

131.67
(399.01)

85.69
(259.67)

75.73
(229.47)

44.29
(134.21)

84.83
(257.05)

68.32
(207.03)

80.31
(243.37)

107.80
(324.50)

90.39
(275.23)

92.95
(281.66)

91.48
(277.22)

C.

243.54
(276.75)

144.02
(163.66)

133.63
(151.85)

62.33
(70.83)

153.81
(174.78)

124.63
(141.63)

141.33
(160.60)

187.83
(213.33)

175.34
(199.24)

129.25
(146.87)

138.42
(157.29)

Sm

16.36
(90.37)

9.52
(52.57)

8.33
(46.05)

4.70
(25.93)

10.06
(55.60)

10.15
(56.05)

12.72
(70.26)

13.04
(72.05)

12.59
(69.56)

12.55
(69.34)

11.26
(62.23)

Eu

1.12
(16.20)

1.72
(24.86)

1.62
(23.43)

3.29
(47.72)

0.90
(13.00)

0.99
(14.37)

1.47
(21.32)

1.36
(19.65)

0.87
(12.59)

0.97
(13.99)

1.59
(23.04)

:35

Th

1.96
(41.78)

1.26
(26.89)

0.87
(18.49)

0.62
(13.15)

1.35
(28.72)

1.83
(38.89)

1.66
(35.42)

1.80
(38.37)

1.64
(34.86)

1.60
(33.95)

1.65
(35.10)

Silica-Oversaturated Syenites (A.5.-a1kaline syenite;

A-66b
A-69
A—70
A-72
A-75
A-77
A—79a
A-79b
A-81
1-85
59-5
39-23
89—25
39-27
39-32
3-44
A992a
A-93
A-96a

A-96c

79.11
(239.74)

47.77
(144.74)

88.29
(267.56)

78.53
(237.98)

71.33
(216.17)

60.19
(182.40)

46.03
(139.48)

154.17
(467.19)

129.33
(391.92)

82.20
(249.09)

98.35
(298.03)

77.17
(233.85)

86.97
(263.54)

53.94
(163.46)

31.13
(94.35)

62.90
(109.60)

51.45
(155.90)

32.35
(98.04)

147.12
(445.83)

217.15
(658.02)

141.10
(160.34)

83.17
(94.51)

170.61
(193.88)

158.61

142.98
(162.48)

120.07
(136.45)

91.33
(103.78)
275.78
(313.39)

233.43
(265.26)

158.19
(179.76)

153.34
(174.25)

147.23
(167.31)

137.96
(156.77)

124.34
(141.30)

56.03
(63.67)

177.95
(202.21)

89.21
(101.38)

122.45
(139.15)

260.77
(296.33)

343.14
(389.93)

14.43
(79.75)

10.72
(59.20)

14.57
(80.52)

14.03
(77.51)

13.30
(73.49)

9.50
(52.51)

9.70
(53.60)

24.94
(137.81)

20.20
(111.59)

17.06
(94.28)

12.62
(69.74)

14.64
(80.88)

10.90
(60.23)

11.45
(63.28)

8.34
(46.06)

19.20
(106.07)

10.17
(56.18)

6.99
(38.64)

26.74
(147.76)

33.72
(186.30)

2.98
(43.26)

3.97
(57.54)

1.96
(28.45)

1.83
(26.57)

2.00
(28.98)

1.86
(26.98)

3.77
(54.62)

4.52
(65.51)

3.70
(53.66)

4.70
(68.13)

1.33
(19.22)

3.62
(52.42)

1.49
(21.56)

1.69
(24.43)

4.97
(72.10)

4.43
(64.14)

2.91
(42.18)

0.68
(9.93)

5.13
(74.32)

5.86
(84.93)

2.67
(56.85)

1.46
(31.16)

1.92
(40.91)

2.31
(49.06)

1.94
(41.36)

1.49
(31.60)

1.03
(21.96)

3.50
(74.57)

3.02
(64.21)

4.26
(90.55)

1.40
(29.78)

1.55
(33.04)

1.10
(23.41)

2.76
(58.70)

0.84
(17.80)

5.16
(109.81)

1.47
(31.27)

1.61
(34.18)

4.07
(86.67)

3.76
(80.03)

Yb

8.85
(44.26)

4.35
(21.76)

5.87
(29.35)

2.11
(10.53)

5.88
(29.38)

4.08
(20.38)

3.84
(19.22)

7.37
(36.83)

6.87
(29.57)

6.52
(32.62)

4.64
(23.21)

0.5.Iquart2 syenite)

6.73
(33.66)

3.51
(17.56)

7.47
(37.34)

9.64
(48.18)

8.64
(43.18)

5.98
(29.88)

4.86
(24.30)

17.47
(87.35)

11.50
(57.49)

11.70
(58.48)

2.37
(11.83)

2.50
(12.49)

2.50
(12.48)

9.81

2.03
(10.16)

13.36
(66.78)

4.41
(22.05)

6.74
(33.72)

16.53
(82.63)

17.13
(85.66)

Lu Th

0.86 23.13 p,
(25.34)

0.45 12.11 p,
(13.20)

0.55 19.71 p,
(16.05)

0.28 8.70 p,
(8.08)

0.52 13.28 p.
(15.39)

0.64 14.73 P.
(18.70)

0.52 8.49 9.
(15.38)

0.51 21.07 F,
(15.09)

0.69 16.58 F.
(19.99)

0.73 17.55 F.
(21.34)

0.59 14.81 P.
(17.48)

1.01 11.48 A.S
(29.61)

0.73 7053 ADS
(21.46)

1.01 14.79 0.5.
(29.59)

0.98 16.04 0.5.
(28.93)

0.86 14.46 0.5
(25.17)

0.73 11.73 0.5.
(21.53)

0.59 5.46 A.5
(17.29)

1.86 27.72 0.5.
(54.64)

1.21 8.72 0.5.
(35.67)

1.14 15.74 0.5.
(33.52)

0.79 12.31 A.5.
(23.14)

0.56 7.77 0.5.
(16.36)

0.71 8.75 3.5.
(21.02)

0.76 15.30 A.5.
(22.48)

0.39 4.70 8.5.
(11.52)

1.04 16.86 0.5.
(30.46)

0.73 7.89 8.5.
(21.50)

00.1 110“ 008'
(23.76).,,v.ry altered

1.58 20.17 0.5.
(46.50)

1.57 44.69 0.5.

(46.12)

Table 1 (cont'd).

8-98

A-102

La
106.88
(323.88)

99.10
(300.30)

CG

182.06
(206.88)

167.98
(190.89)

Sm

18.32
(101.23)

18.95
(104.72)

Volcanic: (Rhyolites/Trachytes)

A-37 121.91
(369.43)
A-4la 92.05
(278.93)
A-42 111.14
(336.78)
A-84 147.55
(447.13)
A-90 135.50
(410.60)
A-95 70.65
(214.08)
A-97a 134.45
(407.43)
A-99 100.15
(303.47)
89-14 98.53
(298.53)
Eg-16 179.27
(543.25)
Dike Rocks

224.14
(254.71)

179.73
(204.24)

207.94
(236.30)

256.11
(291.03)

197.71
(224.67)

124.59
(141.58)

254.71
(289.44)

178.60
(202.96)

184.85
(210.05)

271.78
(308.84)

Silica-Underaaturated:

A-6

A-19b

8-24

A-28

A-49

A-55a

A-63

8-68

36.26
(109.87)

73.67
(223.24)

82.50
(250.00)

89.10
(269.99)

105.77
(320.52)

17.91
(54.26)

67.55
(204.71)

49.52
(150.06)

63.04
(71.69)

119.47
(135.76)

135.59
(154.08)

168.23
(191.17)

202.74
(230.39)

21.26
(24.16)

112.79
(128.17)

91.47
(103.94)

Silica-Oversaturated:

8-2

8983b

A-91c

A-92b

94.22
(285.52)

93.00
(281.82)

78.20

(236.98)

100.65

(305.01)

176.13
(200.15)

155.93
(177.19)

128.50
(146.02)

178.88
(203.27)

18.75
(103.57)

17.07
(94.29)

18.67
(103.15)

25.37
(140.17)

21.24
(117.34)

12.44
(68.73)

20.98

(115.92)

16.75
(92.54)

17.20
(95.04)

28.09
(155.20)

5.26
(29.06)

7.86
(43.41)

9.63
(53.19)

10.59
(58.49)

12.15
(67.11)

2.04
(11.28)

7.55
(41.74)

8.48
(46.85)

15.71
(86.79)

14.12
(77.98)

13.93
(76.96)

18.80
(103.87)

Eu

3.83
(55.47)

4.96
(71.85)

3.71
(53.71)

3.98
(57.64)

3.76
(54.51)

5.38
(77.91)

3.78
(54.75)

4.58
(66.44)

2.66
(38.49)

3.63
(52.58)

3.17
(45.96)

6.04
(87.54)

0.77
(11.11)

0.83
(11.97)

1.06
(15.42)

1.28
(18.52)

0.82
(11.87)

0.45
(6.45)

0.53
(7.66)

2.33
(33.75)

2.40
(34.75)

1.18
(17.13)

2.59
(37.55)

1.63
(23.57)

36

Tb

2.78
(59.12)

2.37
(50.46)

2.40
(51.02)

2.08
(44.24)

2.28
(48.49)

3.63
(77.20)

2.51
(53.46)

1.29
(27.45)

2.90
(61.61)

2.18
(46.35)

2.02
(43.04)

3.68
(78.38)

0.69
(14.74)

0.73
(15.50)

0.92
(19.52)

10‘?
(31.29)

1.18
(25.04)

0.38
(8.04)

0.93
(19.81)

1.27
(26.94)

2.22
(47.13)

2.32
(49.43)

2.59
(55.03)

3.13
(66.51)

Yb

10.64
(53.21)

9.70
(48.52)

11.15
(55.76)

8.85
(44.25)

9.53

14.64
(73.20)

11.20
(56.02)

5.14
(25.72)

13.79
(68.93)

8.64
(43.18)

2.98
(14.88)

3.59
(17.93)

2.95
(14.77)

3.69
(18.47)

4.46
(22.28)

6.45
(32.24)

6.35
(31.77)

1.39
(6.94)

3.29
(16.44)

4.87
(24.35)

8.57
(42.83)

5.73
(28.65)

5.50

7.84
(39.16)

Lu

1.01
(29.69)

1.23
(36.10)

1.07
(31.43)

0.84
(24.70)

1.02
(30.01)

1.48
(43.45)

1.08
(31.77)

0.46
(13.57)

1.26
(37.20)

0.88
(25.95)

1.04
(30.53)

1.28
(37.62)

0.32
(9.19)

0.45
(13.38)

0.53
(15.57)

0.62
(18.19)

0.67
(19.85)

0.30
(8.76)

0.57
(16.83)

0.49
(14.49)

0.83
(24.50)

009‘
(27.56)

0.87
(25.55)

1.11
(32.55)

Th
17.59 0.5.

9.92 0.5.

17.21
12.81
14.44
18.14
19.34
6.88...very altered
25.29
13.55
12.60

19.10

5.66...pegmatite
9.27

11.67

20.86

18.86
7.23...pegmatite

13.47

5.49

Table 1 (cont'd).

Xenoliths

A-10b

A-33a

A-33b

8-34

A-48

be

91.48

(277.22)

24.64
(74.65)

45.69

(138.47)

18.15
(55.00)

78.99

(239.35)

Ce

166.79

(189.53)

36.79
(41.81)

81.32
(92.41)

43.22
(49.12)

153.40

(174.32)

Sm

11.63
(64.25)

3.32
(18.31)

5.95
(32.86)

3.24
(17.91)

10.71
(59.16)

137

Eu

0.66
(9.62)

0.16
(2.35)

0.20
(2.94)

0.18
(2.60)

2.09
(30.26)

Wehrlite--Olivine-rich Ultramafic Cumulate

 

A-94 10.93 35.58
(33.12) (40.43)
Countrx Rock
A-57 8.54 26.23
(25.88) (29.81)
A-64 9.18 15.06
(27.83) (17.12)
A-67 2.81 10.68
(8.50) (12.14)
A-73 82.19 137.17
(249.07) (155.87)
11. Gabal El Kahfa
Gabbros
x-1 40.40 66.88
(122.43) (75.99)
K-9 60.66 106.39
(183.83) (120.90)
R-lOc 57.81 99.55
(175.17) (113.13)
K-lla 6.60 22.38
(altered) (20.01) (25.43)
K-llb 8.38 29.11
(altered) (25.38) (33.08)
K-37a 53.26 79.65
(161.39) (90.51)
K-37b 51.59 81.35
(156.34) (92.45)
R-39 40.41 71.91
(122.44) (81.71)
Bxenites
[-3 57.62 93.30
(174.61) (106.02)
K-4 54.91 101.17
(166.41) (114.97)
K-Sa 106.96 181.60
(324.13) (206.37)
K-Sb 56.19 94.37
(170.28) (107.24)
(155.94) (92.25)
R-7 68.68 116.17

(208.13) (132.01)

2.10
(11.63)

3.16
(17.46)

1.06
(5.83)

1.13
(6.26)

12.52
(69.19)

8.20
(45.30)

10.24
(56.58)

11.25
(62.15)

2.93
(16.17)

4.81
(26.60)

9.02
(49.86)

8.05
(44.48)

7.26
(40.10)

8.07
(44.61)

9.67
(53.41)

16.26

9.60
(53.04)

9.21
(50.90)

10.50
(58.04)

0.54
(7.89)

0.65
(9.40)

0.69
(9.93)

0.82
(11.84)

1.90
(27.56)

2.97
(41.62)

4.95
(71.73)

4.45
(64.50)

1.05
(15.16)

1.43
(20.67)

3.88
(56.18)

2.28
(33.05)

2.83
(41.08)

2.44
(35.37)

3.18
(46.08)

2.97
(40.39)

3.53
(51.22)

3.93
(56.98)

2.54
(36.88)

Tb

1.20
(25.48)

0.55
(11.65)

0.85
(18.12)

0.72
(15.27)

1.18
(25.09)

0.46
(9.79)

0.37
(7.97)

0.41
(8.74)

0.13
(2.78)

1.98
(42.21)

1.04
(22.15)

1.24
(26.38)

1.05
(22.30)

1.13
(23.95)

0.83
(17.60)

1.47
(31.26)

1.74
(37.06)

1.47
(31.26)

0.86
(18.20)

1.10
(23.43)

1.84
(39.14)

1.05
(22.34)

0.87
(18.56)

1.44
(30.73)

Yb

4.91
(24.53)

4.45
(22.25)

10.28
(51.38)

8.67
(43.37)

4.50
(22.52)

0.61
(3.07)

1.34
(6.68)

1.18
(5.90)

1.07
(5.37)

9.24
(46.21)

0.06
(0.30)

3.11
(15.53)

3.66
(18.28)

2.57
(12.86)

1.78
(8.91)

3.28
(16.39)

1.94
(9.72)

2.32
(11.61)

1.79
(8.96)

2.27
(11.35)

2.56
(12.80)

1.86
(9.32)

2.46
(12.28)

2.38
(11.91)

Lu

0.51
(14.94)

0.60
(17.62)

0.97
(28.59)

0.77
(22.72)

0.50
(14.74)

0.19
(5.67)

0.17
(5.01)

0.12
(3.49)

0.31
(9.24)

0.93
(27.23)

0.01
(0.22)

0.54
(16.01)

0.76
(22.47)

0.51
(14.98)

0.67
(19.72)

0.30
(8.88)

0.28
(8.09)

0.37
(10.91)

0.37
(10.78)

0.32
(9.42)

0.69
(20.23)

0.49
(14.54)

0.32
(9.50)

0.81
(23.68)

Th

1.48
(aerpentized)

2.76
(granite gneiss)

0.00
(altered amph. gneiss)

0.00
(amph. gneiss)

13.80
(hornfels
xenolith)

2.90
6.77
7.21
0.64
0.90
6.39
9.32

6.01

9.00
7.77
10.31

7.88

Table 1 (cont'd).

[-13

[-14

[-15

[-16b

[-17

[-24

[-25

[-26

[-27

[-28

[-29

[-32a

[-32b

[-32c

[-33

[-34

[-35

[-36

[-38

[-43c

[-44

[-52

[-56a

[-60

La
87.26
(264.41)

71.25
(215.92)

80.00
(242.42)

73.40
(222.41

51.54
(156.19)

61.96
(187.76)

79.58
(241.14)

60.42
(183.10)

80.05
(242.56)

33.65
(101.98)

117.91
(357.30)

101.47
(307.49)

30.56
(92.59)

27.09
(82.10)

60.48
(183.28)

33.61
(101.84)

23.17
(70.20)

19.08
(57.81)

76.49
(231.79)

76.47
(231.74)

77.25
(234.10)

83.04
(251.64)

79.74
(241.65)

98.18
(297.51)

26.12
(79.16)

15.49
(46.92)

79.81
(241.86)

30.13
(91.30)

Ce
132.40

(150.46)

107.44

(122.09)

124.22

(141.16)

117.46

(133.48)

82.04
(93.23)

97.14

(110.38)

126.72

(144.00)

89.56

(101.77)

128.39

(145.90)

52.96
(60.18)

191.66

(217.80)

147.62

(167.75)

54.80
(62.28)

41.08
(46.69)

109.86

(124.84)

66.36
(75.41)

35.38
(40.20)

33.60
(38.18)

129.10

(146.70)

131.07

(148.94)

132.05

(150.06)

124.97

(142.01)

132.39

(150.44)

143.05

(162.56)

37.88
(43.05)

35.79
(40.67)

112.71

(128.08)

55.39
(62.94)

5m
10.09
(55.74)

8.36
(46.17)

11.35
(62.69)

10.68
(59.01)

9.38
(51.82)

6.67
(36.88)

11.79
(65.16)

8.01
(44.27)

13.32
(73.60)

6.45
(35.63)

18.77

(103.69)

13.84
(76.49)

5.76
(31.85)

3.61
(19.94)

10.93
(60.36)

6.20
(34.27)

2.92
(16.14)

3.36
(18.55)

14.42
(79.65)

11.88
(65.66)

12.26
(67.72)

10.43
(57.61)

13.17
(72.74)

15.26
(84.30)

4.24
(23.42)

5.15
(28.45)

9.93
(54.86)

6.67
(36.86)

in
2.35
(34.03)

3.03
(43.97)

3.34
(48.47)

3.29
(47.63)

3.29
(47.75)

1.72
(24.97)

2.48
(35.96)

0.71
(10.36)

3.09
(44.81)

1.11
(16.08)

2.93
(42.44)

3.71
(53.81)

1.64
(23.79)

0.54
(7.79)

2.09
(30.34)

4.98
(72.20)

0.34
(4.93)

0.35
(5.08)

4.67
(67.63)

4.44
(64.38)

3.91
(56.72)

4.61
(66.81)

3.41
(49.43)

3.94
(57.12)

0.78
(11.34)

1.16
(16.80)

3.57
(51.68)

1.32
(19.14)

38

Th

1.60
(34.03)

0.94
(19.93)

0.99
(21.11)

0.92
(19.60)

0.97
(20.62)

0.83
(17.77)

1.50
(31.89)

0.88
(18.75)

1.80
(38.21)

0.50
(10.61)

2.51
(53.34)

2.47
(52.61)

0.95
(20.30)

0.50
(10.69)

2.21
(47.11)

1.39
(29.54)

0.31
(6.50)

0.56
(11.82)

2.65
(56.30)

1.97
(41.84)

2.75
(58.44)

1.75
(37.32)

1.10
(23.51)

2.67
(56.82)

1.00
(21.28)

1.13
(24.05)

1.87
(39.80)

1.21
(25.70)

Yb

1.95
(9.74)

3.06
(15.29)

3.41
(17.07)

3.20
(16.01)

4.08
(20.40)

1.52
(7.60)

1.69

1.44
(7.20)

2.28
(11.42)

1.96
(9.82)

2.68
(13.41)

5.77
(28.85)

2.13
(10.63)

1.71
(8.57)

4.55
(22.74)

1.39
(4.23)

1.53
(7.64)

1.01
(5.07)

5.48
(27.39)

4.40
(21.99)

4.97
(24.84)

3.11
(15.55)

4.51
(22.53)

4.21
(21.04)

2.18
(10.91)

2.91
(14.55)

1.57
(7.85)

4.00
(19.98)

Lu

0.41
(11.99)

0.32
(9.31)

0.43
(12.60)

0.25
(7.34)

0.45
(13.23)

0.20
(5.82)

0.64
(18.69)

0.56
(16.40)

0.74
(21.91)

0.27
(7.86)

0.98
(28.69)

0.74
(21.64)

0.37
(10.88)

0.35
(10.19)

0.71
(20.92)

0.85
(9.20)

0.35
(10.17)

0.26
(7.65)

0.72
(21.10)

0.67
(19.80)

0.43
(12.60)

0.36
(10.63)

0.51
(14.97)

0.64
(18.71)

0.21
(6.28)

0.31

0.34
(9.95)

0.45
(13.14)

Th

13.41

4.71

6.10

10.91

4.81

Table 1 (cont'd).

La
Volcanics
[-18a 58.15
(176.21)
[-19 82.27
(249.32)
x-59b 73.16
(221.71)
K-62 3.52
(10.68)
Dikes
K-ab 59.76
(181.09)
K-22 104.36
(316.23)
K-40 125.37
(379.91)
x-45 72.83
(220.71)
K-47a 66.57
(201.72)
K-47b 119.88
(363.26)
x-so 114.56
(347.15)
R-Sla 86.66
(262.60)
K-Slb 7.51
(22.76)
K-55a 167.57
(507.80)
K-55b 93.42
(283.08)
[-61 113.29
(343.29)
[-63b 25.49
(77.25)

Countrx Rock

[-12b

[-46

[-48a

[-59a

x-63a

3.88
(11.77)

252.75
(765.91)

9.14
(27.69)

12.95
(39.25)

12.44
(37.70)

Ce

95.31
(108.31)

129.18
(146.79)

104.67
(118.95)

16.36
(18.59)

75.30
(85.57)

150.46
(170.98)

182.23
(207.07)

115.43
(131.17)

102.42
(116.38)

187.67
(213.26)

168.66
(191.66)

138.44
(157.32)

14.06
(15.97)

220.34
(250.39)

123.16
(139.95)

183.41
(208.42)

42.85
(48.69)

18.38
(20.88)

406.41
(461.82)

21.19
(24.08)

29.11
(33.08)

21.10
(23.98)

9.83
(54.33)

10.76
(59.46)

12.35
(68.24)

2.11
(11.68)

6.57
(36.28)

12.28
(67.87)

9.77
(53.96)

9.89
(54.65)

11.46
(63.33)

16.19

17.27
(95.42)

15.16
(83.75)

3.15
(17.38)

14.06
(77.70)

9.19
(50.80)

14.82
(81.90)

5.29
(29.24)

2.17
(11.99)

31.34
(173.13)

2.62
(14.50)

5.66
(31.25)

3.60
(19.90)

39

Eu

3.23
(46.80)

3.36
(48.65)

4.03
(58.44)

0.91
(13.15)

1.03
(14.92)

2.50
(36.22)

0.54
(7.82)

1.67
(24.23)

3.74
(54.18)

4.24
(61.41)

2.83
(41.08)

1.98
(28.68)

1.31
(18.99)

2.35
(34.12)

2.32
(33.59)

1.64
(23.82)

1.23
(17.89)

0.91
(13.20)

4.56
(66.13)

0.69
(9.96)

1.41
(20.42)

1.28
(18.57)

Tb

1.21
(25.71)

1.26
(26.91)

1.82
(38.70)

0.43
(9.08)

0.55
(11.63)

1.47
(31.21)

2.88
(61.23)

2.44
(52.01)

1.88
(39.97)

2.28
(48.51)

3.86
(82.10)

2.62
(55.70)

0.14
(3.08)

2.83
(60.16)

1.80
(38.39)

3.66
(77.86)

1.00
(21.30)

0.08
(1.72)

2.47
(52.56)

0.52
(10.98)

1.03

0.78
(16.64)

Yb

4.20
(21.02)

3.18
(15.91)

4.08
(20.39)

0.72
(15.29)

2.54
(12.69)

2.14
(10.72)

6.52
(32.58)

8.20
(40.98)

4.80
(23.99)

9.10
(45.51)

7.25
(36.25)

5.32
(26.58)

3.27
(16.33)

4.47
(22.37)

3.44
(17.19)

7.76
(38.80)

2.28
(11.41)

2.21
(11.07)

6.09
(30.44)

0.47
(2.33)

1.76
(8.82)

0.54
(2.64)

Lu

0.48
(14.08)

0.30
(8.71)

0.37
(10.92)

0.38
(11.10)

0.74
(21.68)

0.56
(16.36)

0.75
(22.05)

0.94
(27.70)

0.67
(19.58)

0.85
(25.01)

1.07
(31.42)

0.60
(17.68)

0.33
(9.84)

0.48
(14.22)

OO‘I
(11.96)

0.75
(22.19)

0.43
(12.61)

0.31

0.89
(26.31)

0.22
(6.49)

0.38
(11.09)

0.19
(5.50)

13.05

40.80

19.29

16.22

17.61

59.61

17.52

12.53

3I‘7

(fenitized)

40

Table 1 (cont'd).

III. Gabal £1 Naga

 

La Ce Sm Eu Tb Yb Lu Th

Sxenites

N-Sb 76.78 125.93 10.95 4.79 1.34 2.77 0.49 10.08
(232.66) (143.10) (60.50) (69.39) (28.52) (13.85) (14.35)

N-Sc 74.73 124.51 10.63 4.44 1.21 2.71 0.47 9.76
(226.46) (141.49) (58.75) (64.29) (25.72) (13.53) (13.95)

N-7a 66.18 120.99 8.13 3.85 0.94 4.00 0.64 11.88
(200.54) (137.49) (44.94) (55.83) (20.00) (20.02) (18.97)

N-lo 54.77 93.88 5.60 5.88 0.72 2.25 0.37 5.10
(165.96) (106.69) (30.93) (85.18) (15.37) (11.24) (11.03)

N-14 71.64 128.63 10.49 3.35 1.10 3.77 0.54 10.16
(217.08) (146.17) (57.94) (48.54) (23.50) (18.86) (15.91)

N-l7 98.99 173.26 15.18 4.50 2.18 5.06 0.81 15.74
(299.97) (196.88) (83.89) (65.26) (46.38) (25.31) (23.82)

N-22 57.88 121.35 9.11 4.21 0.78 2.73 0.57 5.47
(175.39) (137.90) (50.36) (61.02) (16.57) (13.66) (16.70)

N-23 42.06 89.99 5.31 4.84 0.52 3.00 0.48 5.77
(127.47) (102.26) (29.36) (70.11) (10.99) (15.01) (14.05)

N-25 114.92 200.43 12.49 3.59 1.43 6.23 1.01 17.19
(348.24) (227.76) (69.02) (52.06) (30.48) (31.15) (29.57)

N-32 79.58 134.94 11.81 5.35 1.39 2.36 0.53 4.41
(241.15) (153.34) (65.24) (77.49) (29.59) (11.81) (15.46)

N-33 56.71 111.59 5.76 6.68 0.83 3.70 0.55 7.59
(171.86) (126.81) (31.84) (96.76) (17.76) (18.49) (16.07)

N-34 51.24 92.58 5.41 6.83 0.61 3.14 0.43 6.62
(155.26) (105.20) (29.87) (98.99) (13.04) (15.72) (12.60)

N-36 68.03 139.98 6.96 3.80 0.80 2.84 0.56 12.82
(206.14) (159.07) (38.46) (55.11) (17.09) (14.20) (16.34)

N-40 48.55 98.91 5.71 5.39 0.79 2.06 0.58 7.34
(147.13) (112.39) (31.52) (78.10) (16.83) (10.28) (17.17)

N-47 66.07 118.01 8.45 6.47 1.19 2.40 0.50 10.87
(200.22) (134.10) (46.70) (93.77) (25.22) (12.02) (14.82)

N-48a 44.23 97.02 6.78 1.33 0.54 2.51 0.53 3.80
(134.03) (110.25) (37.46) (19.30) (11.59) (12.57) (15.56)

N-Sl 65.54 142.92 9.19 4.17 0.83 3.41 0.60 7.03
(198.60) (162.41) (50.80) (60.43) (17.61) (17.06) (17.68)

N-52 69.33 122.47 7.50 4.48 0.74 1.89 0.56 5.59
(210.09) (139.17) (41.46) (64.88) (15.76) (9.46) (16.38)

Volcanic:

N-la 87.25 134.69 11.70 4.44 1.57 3.10 0.65 11.49
(264.39) (153.06) (64.66) (64.39) (33.47) (15.51) (19.20)

N-4 111.03 234.90 11.09 1.51 1.16 6.12 1.11 19.18
(336.45) (266.93) (61.29) (21.83) (24.78) (30.61) (32.78)

8-27 80.90 143.88 12.06 3.82 1.54 3.96 0.56 10.82
(245.14) (163.50) (66.64) (55.41) (32.77) (19.81) (16.58)

8-28 80.39 131.79 11.99 4.51 1.39 2.89 0.47 11.00
(243.62) (149.77) (61.80) (65.36) (29.55) (14.43) (13.75)

N-30 93.63 153.38 11.47 3.16 1.47 4.35 0.70 12.77
(283.74) (174.29) (63.39) (45.80) (31.25) (21.73) (20.62)

Dike:

l-20 89.45 145.98 12.70 3.58 1.52 3.78 0.62 9.48
(271.05) (165.88) (70.14) (51.95) (32.28) (18.92) (18.17)

l-38 53.33 109.02 7.63 1.90 0.78 2.46 0.42 4.30
(161.61) (123.88) (42.13) (27.57) (16.61) (12.31) (12.44)

8-39 185.75 341.65 17.98 1.20 2.00 10.78 1.67 33.45
(562.88) (388.23) (99.36) (17.44) (42.57) (53.91) (49.26)

Conntrx Rock

3-45 44.17 94.34 7.11 1.54 0.83 2.47 0.58 8.47
(133.85) (107.20) (39.28) (22.38) (17.74) (12.37) (17.03)

Table 1 (cont'd).

IV. Gabal Nigrub £1 Fogani

Sxenitea

P-1

Volcanic

P-17

m

F-7

52.12
(157.94)

28.49
(86.35)

56.33
(170.69)

51.09
(154.83)

50.27
(152.33)

91.22
(276.41)

48.51
(147.00)

93.91
(284.57)

77.46
(234.73)

43.97
(133.24)

88.53
(268.26)

21.25
(64.41)

93.34
(282.86)

56.96
(172.62)

94.44
(286.19)

10.47
(31.73)

Countrx Rock

P-4

P-30a

8.23
(24.95)

21.24
(64.37)

Ce

113.61
(129.11)

72.46
(82.34)

114.46
(130.07)

114.64
(130.27)

105.74
(120.16)

162.82
(185.02)

94.80
(107.73)

161.72
(183.77)

170.66
(193.93)

83.95
(95.39)

179.12
(203.55)

40.47
(45.98)

196.46
(223.25)

109.56
(124.50)

206.25
(234.38)

30.41
(34.56)

33.55
(38.13)

49.80
(56.59)

Sm

8.79
(48.55)

3.71
(20.47)

9.67
(53.40)

7.32
(40.43)

6.30
(34.83)

15.33
(84.70)

6.09
(33.66)

17.48
(96.55)

9.93
(54.89)

7.20
(39.79)

10.09
(55.73)

2.03
(11.20)

9.74
(53.81)

9.88
(54.60)

14.34
(79.23)

3.99
(22.06)

1.71
(9.44)

1.90
(10.51)

in

3.23
(46.78)

0.78
(11.36)

3.33
(48.19)

0.93
(13.44)

1.30
(18.84)

4.44
(64.33)

1.49
(21.65)

5.76
(83.50)

1.62
(23.52)

9.85
(142.69)

1.36
(19.72)

0.21
(3.01)

1.26
(18.23)

5.15
(74.64)

2.64
(38.33)

0.23
(3.39)

0.61
(8.84)

0.62
(8.94)

41

Tb

0.94
(19.99)

0.27
(5.71)

1.08
(22.91)

0.92
(19.55)

0.60
(12.66)

1.45
(30.90)

0.63
(13.40)

1.76
(37.36)

0.94
(20.07)

0.91
(19.29)

1.14
(24.17)

0.35
(7.43)

0.91
(19.37)

1.08
(22.88)

2.25
(47.79)

0.50
(10.56)

0.18
(3.83)

0.26
(5.61)

Yb

3.19
(15.97)

2.28
(11.39)

3.27
(16.36)

2.14
(10.72)

2.69
(13.45)

6.02
(30.12)

2.83
(14.15)

4.51
(22.57)

2.85
(14.27)

0.94
(4.72)

5.72
(28.59)

4.83
(24.14)

3.93
(19.63)

5.21
(26.06)

9.72
(48.58)

2.66
(13.32)

0.72
(3.59)

0.51
(2.53)

Lu

0.53
(15.62)

0.36
(10.44)

0.50
(14.79)

0.49
(14.52)

0.48
(14.21)

0.52
(15.40)

0.37
(10.87)

0.38
(11.05)

0.71
(20.78)

0.30
(8.76)

0.98
(28.74)

0.86
(25.35)

0.56
(16.46)

0.92
(27.08)

1.70
(50.02)

0.57
(16.77)

0.17
(5.06)

0.20
(5.83)

Th

7.28

9.

5.

6

19.

5.

7

3.

20.

17.

13.

96

56

55

.70

58 (net-vein
mixed rock)

03 (net-vein
leucosyenite)

87 (net-vein
melanosyenite)

.70

93

03

.49

36

.79

95

12.21

42

Average REE and Thorium Concentrations for the

Major Rock Types at Gabal Abu Khruq

Rock Type Ce Sm Lu Th
Foyaites 159.47 11.23 0.64 19.78
n=18

Pulaskites 115.01 9.90 0.49 10.37
n=7

Quartz 195.29 18.61 1.12 17.40
Syenites

n=l3

Alkaline 115.99 11.62 0.74 10.01
Syenites

n=6

Volcanics 210.04 19.65 1.06 16.94
n=9

Gabbros 55.46 6.07 0.31 2.18
n=9

Gabbros* 38.33 4.03 0.20 1.16
n=2

* Two outcrops of gabbro represent more
melanocratic varieties.

43

Figure 9.

Normalized REE patterns for the gabbros from Gabal Abu
Khruq. Note large positive Eu anomalies and overall sim-
ilar patterns except for one highly altered sample,
A-76.

a) More mafic, lower REE samples.

b) All other gabbro samples.

44

 

 

 

 

D
I
4
4 m oases: a-eaz
< 6 0888‘: 8-51;
X 885553 8-515
.0
U":
4..) 4
g 4
c 1
c
a d
2
E J
a
4
4
1
4
4
a LG CE 5") EU TB YE LL‘
0

 

 

 

55 $7 58 $5 60 6: 62 63 E4 55 as 6‘. ea 59 vs 7: 72
RTOEIC NUflBER

 

“C
;
1
q
4 B one“: 5-:
6 one”: n-se
‘ 0 one": 5-5:
- ' A cases: as:
.3: \ - x unease n-e:
U 4 -
‘ Y ”8860 G-)::
I. -e
E 1 ‘.\,/5\
4 J
a 4
S
g 4

(3"

A ALLA.

 

J

 

 

e L8 CE 5?) EU ‘8 Y8 LU
'5??? Eu 39 5'0 5'1 5'2 5'3 34 is is £7 5': 5'9 7'0 7'1 12
RTOHIC NUHBER

 

Figure 9.

45

Figure 10.

Normalized REE patterns for the low-REE syenites from

Gabal Abu Khruq.

a) Silica-undersaturated syenites which contain the lowest
REE concentrations.

b) Silica-undersaturated syenites with slightly higher
REE concentrations.

c) Silica-oversaturated syenites with REE concentrations
comparable to the silica-undersaturated syenites in
diagram (b).

d) Pegmatoidal silica-undersaturated samples.

46

 

(M1160 8W!

{'3'

 

A A ‘ALA‘

A

LR CE SH EU T8 Y8 LU

 

08

 

 

ss 67 {a £9 £0 61 62 £3 64 £575 67 is 6970 'IT: '72
nTonIc NUMBER

3 b

 

A

lo!
A A A A.

A

 

MI. (188 ROW!

(3"

A LALAJA

'9 L8 CE 5H EU TB YB LU
56757 in 69 6071 5'27: Tn 615 is 67 60 69 20 W72
RTOHIC NUNBER

 

 

 

Figure 10.

47

 

[Cr-32

0
I!)
a
0
x

{gaunt I280 me

A LALLAA

J

0 L8 CE SH EU T8

 

 

YB LU

 

 

RTOHIC NUHBER

 

A

A

a n-S
‘ o a-ssn

(a

A LAAAAA

a

(0

 

 

A AAAAA

0 L8 CE 5H EU TB

 

YB LU

 

 

Figure 10 (cont'd)

66 67 58 6'9 6'0 61 82 63 64 66 68 6'7 88 89 7o 7: 72

55 67 68 69 60 61 62 63 64 65 66 67 68 59 7'0 71 72

d

48

 

- AAAAA‘

. .10'

 

All

a

 

uopnntlzto naunonucc

A #AALLAAAktl

L8 CE SH EU T8 YE LU

 

 

.

O

—56 6'7 6'6 69 6'0 61 6'2 6'3 64 66 66 67 66 65 70 7; 72
a RTOHIC NUHBER

 

ALAAAJ

L

(0'

(280 ME

I LAAAA

 

-.(°

LLLAA

1

1: L8 CE 6n EU T8 Y8 LU
b —66 6'7 60 69 6'07: 6'2 5'3 64 6'6 6'6 67 6'6 6'9 7'0 7:1 72

 

 

 

Figure 11.

Normalized REE patterns for the high-REE rock-types

from Gabal Abu Khruq.

a) Highly silica-undersaturated foyaitic syenites.
Note overall consistent pattern between indivi-
dual samples.

b) Extremely sodalite-rich foyaites.

49

Figure 11 (cont'd.)

c) Quartz syenites. Note overall consistent pattern
between individual samples and the higher concen-
trations relative to the foyaites.

d) Quartz syenites which do not display the marked
negative Eu anomalies common to most of the highly
silica-oversaturated samples. These samples may
represent more mafic syenites with large amounts
of secondary quartz.

e) Volcanic silica-oversaturated samples. Note sim-

ilarity to the quartz syenites.

50

xuntax u=awc U
«r 2. an 0.. c.. h. a. a. up. n.- u.. .p. 9.. 0.. ... 9.. .—

3 2 2 3 z... 3 S o.

 

. 6.0681 3 0.936

11.00. nun-049

 

 

 

 

 

 

33748004088 0311188

 

 

 

 

 

 

. \OII
K
‘
0
eaten
$2.22 25E m o. 5952 25E
b _.b om MW 0.. N... mm M. I. m.. NW _.w 0.. mm 0.. FM ow Np Z. or mm on r. mm mm cw m. N. —c on an a... BM aw
o o
3 E E B :m 3 S . 3 E m: B :m 3 5
MM H...
J
u
m
m
u
m
m
u
n 6|
m. a.
0‘

 

 

iou

51

 

 

 

 

 

 

 

‘ 1
4
.0
u":
U 4
E :
g.
2 4
a 4
r:
E 1
E.
g:
1
4
J
. (.8 CE 571 EU TB Y8 LU
O
66 6'7 6'6 69 6071 62 63 6'4 66 66 67 68 69 70 71 72
a mom: NUMBER
“D
1 7
‘ I
“s
U 4
E I
g 4
a ‘
3
5'3.
7
4
8
ea L9 CE 5" EU T8 YB LU

 

 

 

b 66 6'7 6'8 6'9 6'0 61n§20H613c 64"}:BE: 6'7 6'6 6'9 7'0 7': 72
Figure 12.
Normalized REE patterns for the dike rocks.
a) Silica-undersaturated dike samples. Note simi-
1arity to the foyaitic syenites.
b) Silica-oversaturated dike samples. Note simi-
larities to the quartz syenites.

52

 

D
1

j m R-IOB

A .-

* s + 22::

' "3 0 9-34

u 0 9-48
EL
U n

‘L‘

BHHLIZED naunonuc

N0
1'0

 

Q
A

A I

LR CE SM EU TB YB LU

 

2 I I j fl I I I 1 ‘I I I 1 I I I
56 5'7 58 59 60 61 62 63 64 65 66 6'7 68 69 70 71 72
RTOHIC NUMBER

 

 

Figure 13.

Normalized REE patterns for xenoliths. Although
some samples display very large negative Eu ano-
malies, they are otherwise similar to the silica-
undersaturated dike rocks.

53

Several features should be noted in these normalized
REE plots:
1. There is an overall sub-parallel relationship between
the different varieties of rock-types.
2. The REE patterns are generally very similar for each
rock-type, i.e. the REE distributions are not erratic.
3. The REE concentrations of the gabbros are less than
those for the syenites and volcanics.
4. There is a wide range of Eu characteristics for this
entire alkaline suite; from large positive Eu anomalies in
the gabbros, to positive, flat, and negative Eu patterns in
the pulaskites and alkaline syenites, to small negative Eu
anomalies in the quartz syenites and volcanics, and finally,
very large negative Eu anomalies in the foyaites.
5. The REE patterns of the pulaskites and alkaline syenites
are very similar; both rock-types display a transition in
Eu anomalies. The average absolute REE concentrations are
slightly higher in the alkaline syenites than in the pula-
skites.
6. The REE patterns of the quartz syenites and the volcan-
ics are very similar.
7. The volcanics, which erupted early in the history of
Abu Khruq, contain the highest average REE concentrations.
8. The foyaites, which were the last major stage of
crystallization at Abu Khruq, contain average REE concen-
trations which are lower than the average REE concentrations

of any of the other syenite and volcanic rock-types.

54

9. Careful inspection of the REE data reveals that some
samples which appear to be highly differentiated, contain
lower concentrations of REE than some samples which appear
to be less differentiated (i.e. in thin section or with
respect to other elements). This can be seen especially
within the foyaites. Figure lOd illustrates an extreme
example of this apparent REE-enrichment reversal for highly-
evolved, pegmatoidal silica-undersaturated samples which
according to field relationships, crystallized in the final
stages of the develOpment of Abu Khruq.

A number of dikes were analyzed for REE because of
their abundance, their possible metasomatic affects on the
host rock within the complex, and the possibility that some
would provide examples of a parent magma (e.g. see Upton,
1974; Larsen and Steenfelt, 1974). Although there are many
varieties of dikes (see El Ramly et al., 1969a), Figure 12
simply divides them into silicaeundersaturated and over-
saturated types. They are similar to the silica-undersatur-
ated syenites, and quartz syenites/volcanics, respectively.
They do not appear to have unusual REE concentrations to
indicate associated metasomatic activity, nor do their
compositions suggest a large mafic component.

Xenoliths have been shown to reveal important features
about alkaline magmas (see DeLong et al., 1975; Ludden,
1977; Rock and Scoon, 1976). They occur in especially
large quantities in silica-undersaturated suites, often

coming from great depths in the mantle (Wilkinson, 1974).

55
At Abu Khruq, only the silica-undersaturated syenites
contain significant amounts of xenoliths (see Appendix I).
Petrographically, apart from some samples containing large
amounts of Fe-rich biotite, they appear similar to the
silicaeundersaturated dikes, and therefore, consist of
cognate material rather than mantle or deep-crustal xeno—
liths. Their REE patterns in Figure 13 are similar to the
silica-undersaturated syenites, although they display

larger negative Eu anomalies.

B. Other Trace Elements

Table 2 presents a limited number of analyses for
compatible and other incompatible trace elements. The Rb
and some of the Sr data is from Lutz (1979). The important
features to be noted about these trace element distributions
are:
l. The concentrations of the compatible trace elements
with known ferromagnesian affinities (Ni, Cr, Co) are very
low in the majority of the syenites and volcanics. The
concentrations of the incompatible trace elements for these
same rock-types are moderately high compared to the more
mafic rock-types.
2. The Sr concentrations in the gabbros are very high;
the Sr content is low in all other rock-types.
3. The Ba concentrations in a few syenites are very high.
These samples appear more mafic in thin section, and contain

higher concentrations of elements with mafic affinities.

The Ba content is essentially zero in the foyaites.

5(5

ono.o muaouuao
ooouo><

ooo.c uuwohom .
omnuu>< cow o ouwxoumaanmn

oso.o ouacoam ocaauzdmz ano.o ouwmaomnmu
omuuu>< a.~.c unwouuao-ad

 

mmuwcoxm couousummuopcauQOAme

.voms aw unauuaocoaoc a..~o uo aaaom Hm .Usuzx 92¢ .0
u0u Acmomav .Hc no aaemm Hm EOuu mama mcauono

 

 

 

 

 

 

Nv.~m v~.m~ mwlm om.o¢ mm.oa Hhum
h.mv~ on.o noun o.-~ ah.nN Chum
m.h°H mh.h Nolm mu.o¢ nm.na mwum

hm.v¢ hm.oa Honm «.mwd v~.n~ mmnm

mv.vm av.~n anlm ~.na~ hm.oN swim

vo.av o.o°~ onlm a.v~H Hm.vv omnm

vo.oh vo.mH oNIm mm.mv mv.hh mNIN
nv.wa vm.o~ mmlm h~.oh H.n- mum
mo.hm mm.h@ «mum hm.ca vw.HH mum
no.cw Ac.o~ ANum n.h- hm.m omlm
ma.va m.hhd adum h.ncn om.mv MIN
vv.¢o ma.od dim N.v0H hm.NA Num
caucusuomuw>o pouausuamuopco "mouwcomm
av.vh om.c~ coin Ha.ma ~.oww volm
oc.ap hn.cm swim om.- N.voa Oman
00.5w mu.Hn wonm ~¢.o~ o.o~m hum
unaccoao> .nounboo

nu um nu um

Usunx and .o uOu .mnaavuuaq scum sumo Adina

 

 

 

 

oNH aaa eves av m.oa .m:<
.muuaseso soauumca>aao oauasnuuas. uuaauaoz

o n.n mm cue own ovn¢

. a.» o «v ..~n noana

A800“ umO: Ufluflufluflnuflvflblflnvwnmmv azuwHOCUX

o n.» a ca~ v.9. «-4
Aconcaguauuo>o-nuauam. oxaa

as m.H~ a and o.o~ oauou

m n.m a can m.- vauuu

a a.ov a o.~ n.o~ can:

m o.ov a can n.~n odauc
NUwfiflnudOsr

.m.< ma a.HH a coma onm «n-6m
.m.< a“ m.a e on m.~n m~amu
.m.o mv o.m ma o.v v..m ahna
.m.o a ”.5 ma on ~.n~ ~p-¢

.wua:o>n «cwaaxau I.m.<

 

“wuwco>u nuuusv «.m.o. auuficoaw caucusuomuw>0uoow~wm

 

 

 

.m m m.¢ a can «NH aoonfl
.m AH o.c Ha omm awn ooou<
.m mv o.m mm onv ham ommnt
.m o 5.5 mm Av cow noml<
.h m m.~ 9 av n.m~ awn‘
.m m o.o 5 av a.~v o~n¢
.m «H a.ov Ha ocv n.mo on-‘
.m n o.o~ h Av o.h~ nomu<
.ouwxunasd um “ouwa>0u um. nouwcoam wounusuanuoocauanuum
on v.vm an ova sea aamu‘
mv ~.o~ mm ova con uncut
nounnno
mausx 38¢ dunno .H

00 no “2 on an

.enn :« ooucoooun cud

mucmfiumo nozuo “an .a :w uncuuauacoucou Unauo~£o
.«cooOh an bound: .6

“omoz «m .o “canon Hm .u ”vaunx 5&4 .o nauseanlou
0:33.: .3323 as» nan 3:233 00...? :55

.~ wanna

557

va

00

no

wz

 

 

dv w.m~ hack
cucuoao>

Av o.m~ mwnm

oa w.~m Nana
arc and south
mauwcoxw

acmmOh Am bouqu dunno .>H

an um

 

 

 

 

 

as o.n a omow man «an:
UMCHUHO>
Ha a.» a com. an man:
a ~.c a onaa use n~uz
m a.c m o~a~ can ~H-z
m a.o a cam. avn can:
0 o.o a omaa ooh 65:2
mmmmmmxm
mummflmmlmmmmm .HHH
m~ ans ma o.m~ can «ens
mm o.m~ an omm pom aaux
OHGGOHO>
MN a..~ .H cmo Ham menu:
cm m.om mm can mom ~mux
ow n.~m mm oh. aom vaux
cw o.o~ m~ cma can onvux
5H m.o m oucu mad aqua
o «.5 a on a.- an-»
a h.n o ow~a «on can:
a o.¢ m av o.o« v~-u
mm a.a~ an cNm can noacx
as m.- mm can ~c~ n~-x
a
on m..~ cm con man anus
«m o.o~ co oma mom 0.:
QOHQDUU
mucus Hm Hanan .HH

co to a2 on an

.Av.u:oo. ~ Unnaa

58

4. Sample A-94 represents an isolated outcrOp of an
ultramafic, serpentinized rock-type which is considered to
be part of the early deve10pment of Abu Khruq.

5. There are large differences between some samples in
terms of certain trace elements, and very little differences
in terms of other trace elements for those same samples.
This is considered to be due to the cumulate nature of some
individual samples, and perhaps due to the unknown affect
which mechanisms other than crystal fractionation may have

exerted on the trace element distributions.

CHAPTER 4

THE PETROGENESIS OF SILICA-UNDERSATURATED/

OVERSATURATED ALKALINE COMPLEXES

Introduction

 

This chapter will discuss the interpretation of the
chemical data on the Egyptian alkaline complexes in general,
but with more emphasis on the four silica-undersaturated/
oversaturated associations sampled by this study. In view
of the general chemical similarities between these alkaline
complexes, Gabal Abu Khruq is considered to serve as an
example of the type of processes Operative in the evolution
of many of the Egyptian alkaline complexes. The obvious
differences with respect to some of the other complexes will
be discussed, and these will serve as examples of the types
of variations possible.

The Egyptian alkaline suites are composed of predomin-
antly salic, low-pressure alkaline rock-types which have
many characteristics of a volatile-rich residual liquid.
The presence of typical residual mineral phases, high
chlorine concentrations, and widespread metasomatic/hydro-
thermal features give some indication of the low-temperature,
halogen-rich nature of these alkaline magmas. This is
supported by the projection of the compositions of the
syenites into the thermal minimum of the Q-Ne-Ks residua
system (Lutz, 1979, pp. 122-123). These types of melt
systems have often been associated with many complex,

59

60
erratic, and poorly-defined mechanisms. Further, the
majority of the rocks in many of the Egyptian alkaline
complexes are coarse-grained, and unlike volcanic rocks,
they often cannot be considered to represent the composition
of the melt at the time of their crystallization. Rather,
they may represent an unknown combination of crystals and
trapped liquid (e.g. Irving, 1979). These features compli-
cate a petrogenetic solution and cause scatter in the com-
positional trends of the chemical data for the Egyptian

alkaline suites.
Section I

Metasomatism

 

The chemistry of many salic alkaline suites have been
reported to be strongly affected by metasomatic fluids,
expecially in the final stages of crystallization (Mineyev,
1963; Ganzeyeva and Ganzeyeva, 1975; Aleksiyev, 1970;
Bowden and Whitley, 1974; Mitchell and Brunfelt, 1975;

Eby, 1975; Borodin and Pavlenko, 1974; Barber, 1974; Rock,
1976; Martin et al., 1978). Before the chemistry of the
Egyptian alkaline suites can be interpreted as primary
petrogenetic indicators, the extent of metasomatism for
each complex should be addressed.

Gabal Nigrub El Fogani contains signs of extensive
alteration (see Appendix I). Much of it appears to be
autometasomatism and may relate to the presence of COZ-rich

phases during the latter part of the crystallization of the

61

melt and the formation of carbonatite dikes. There is sub-
stantial evidence for alteration due to migrating fluids

at Gabal El Kahfa. Rather explosive events are likely to
account for the large cataclastic zones within the complex,
as well as the fused and altered appearance of many of the
rocks within the complex and the surrounding gabbros,
volcanics, and amphibole schists. Gabal El Naga shows the
least amount of alteration, both within and outside the
complex (contrary to the observations of El Ramly et al.,
1969b, see Appendix I). The country rock surrounding Gabal
Abu Khruq contains little signs of widespread fenitization.
Figure 14 shows the normalized REE distributions of three
samples of country rock; one from the granite gneiss and
two from the amphibole gneiss. Their REE concentrations
(as well as Th concentrations, see Chapter 3) are low and
not unlike the patterns of similar metamorphic rocks (refer
to Shaw, D.M., et al., 1976; McClennen et al., 1979; Nance
and Taylor, 1977). Petrographically, only A-64 shows signs
of extensive alteration and its REE distribution is not
significantly different than A-67 which is otherwise similar
in thin section. A-57 is a particularly useful sample
because it was collected from the granite gneiss-alkaline
complex contact. It appears to contain its original quartz
and have an overall fresh appearance in thin section. This
would not support extensive metasomatism at Abu Khruq, nor
a silica-transport mechanism from the surrounding country

rock to the alkaline complex (see Lutz, 1979, pp.233-242).

62

Figure 14.

Normalized REE patterns of three samples of country
rock which surrounds Gabal Abu Khruq. In general,
these REE patterns, similar to their petrographic
appearances, do not indicate extensive fenitization
or alteration due to the adjacent alkaline magmatic
activity. These REE patterns are similar to those
of equivalent metamorphic rocks in other areas (see
text).

Figure 15.

8102 versus Sm for the rock series at G. Abu Khruq.
The positive relationship between SiO2 and Sm is con-
sistent with fractional crystallization rather than
silica-leaching and associated REE-enrichment via
fenitizatization (see Martin et al., 1978).

63

 

“O
;
‘ A GRANITE GNEXSS
.. m RHPHIBOLE ONEISS
E; o nn9Hl60LE ONEISS
U I
E :
g 4
a 1
U
2.
59..
1
4
4
.D LR CE SH EU TB

 

 

 

 

970?) I C NUHBER

Figure 14.

39.06

2} .66

29.60

«.6. 0

SH PPH

 

66 67 66 69 60 61 62 63 64 36 66 67 66 39 7o 72 72

 

 

6.66 63.66 63.66 than “.66 3.66 a.”

5102 HT PCT - RBU KHRUD

8
a

Figure 15.

than

64

Silicate rocks derived by fractional crystallization
should show a sympathetic increase between SiO2 and REE.

A negative correspondence might indicate fenitization and

the transport of silica (see Martin et al., 1978). Figure
15 demonstrates a positive relationship. This trend does

not support extensive post-magmatic fenitization, nor the

formation of the silica-undersaturated rocks via fenitiza-
tion of originally silica-undersaturated rocks.

A general indication of the absence of large amounts
of fenitization at G. Abu Khruq (as well as El Naga) is the
overall sub-parallel normalized REE patterns for the entire
rock series.

To summarize, there is evidence of strong alteration
at G. El Kahfa, and late-stage metasomatic activity at G.
Nigrub E1 Fogani. There is little evidence for widespread
metasomatic fluids in disequilibrium with the alkaline melt

systems at G. El Naga and G. Abu Khruq.

Section II

 

Fractional Crystallization

 

Introduction

 

The gabbro-syenite association has been noted in
alkaline complexes throughout the world (e.g. Sorensen (ed.),
1974). Since Bowen's (1945) postulation of the ”plagioclase
effect" many workers have drawn upon the fractionation of
large amounts of plagioclase (found in the gabbros) from

an originally basaltic magma to explain the development of

65

peralkaline residual melts which crystallize and form the
syenites and alkaline volcanics (see Upton, 1974; MacDonald,
1974; Edgar, 1974). In a preliminary way, the field rela-
tionships between the syenites and gabbros, and the general
major element and trace element compositions of the rock
series in the Egyptian alkaline complexes are qualitatively
consistent with the involvement of crystal fractionation.
There are two general ways in which fractional crystal-
lization may be shown to have occurred in the evolution of
the Egyptian alkaline complexes. One approach entails a
comparison of the chemistry of the Egyptian alkaline com-
plexes with that of alkaline suites in other areas which
have frequently been demonstrated to have evolved largely
through the process of crystal fractionation. The other
approach qualitatively assesses the chemical trends of the
Egyptian alkaline complexes in light of the trends expected
for crystal fractionation. The paucity of highly mafic
end-members in the four alkaline complexes of this study
does not allow a rigorous petrogenetic solution. This lack
of mafic end-members is considered to be due to the effi-
cient differentiation of mantle-derived, volatile-rich
alkaline melts, resulting in the formation of ultramafic
cumulate strata mostly below the exposed level of the
alkaline complexes. Plutonic/volcanic alkaline complexes
similar to those found in Egypt frequently contain evidence
of associated lower units of gabbro and ultramafic strati-

form complexes (e.g. Munoz, 1969; Gastesi, 1969; Upton,

66

1974; Bridgewater and Harry, 1968; Fernanadez, 1980).

1) Comparison of the Chemistry of the Egyptian Alkaline
Complexes to Other Alkaline Suites

The worldwide similarity of many primitive alkaline
suites was discussed in Chapter 3. Many of the composi-
tional trends of these suites have been frequently inter-
preted to be the result of crystal fractionation. Due to
the lack of representation of mafic end-members in the
Egyptian alkaline complexes, it is difficult to reconstruct
the early differentiation stages during which crystal
fractionation is likely to have been an important process.
However, by comparing the compositional trends of the
Egyptian alkaline complexes to these alkaline suites in
other areas, it is possible to qualitatively note the
conformity of the Egyptian data to the trends expected for
crystal fractionation.

Figures 16-20 illustrate the similarity between the
chemical trends of numerous alkaline suites--many of which
have been interpreted to be the result of crystal fraction-
ation. Similar chemical data was plotted for a large
number of alkaline suites from different areas of the world.
In addition, Dr. R. Schwarzer (Rice University) supplied
this study with unpublished AFM and silica-alkali compila—
tions of alkaline suites formed in different tectonic
settings which supplemented the data presented by Schwarzer

and Rogers (1974). These compilations further substantiate

67

Figure 16.
AFM ternary diagrams for several well-known oceanic
alkaline suites which are similar to those of the
Egyptian alkaline complexes, and similar to trends
expected for fractional crystallization. Data com-
piled from Bishop and Wooley, 1973; Goldich et al.,
1975; and Schmincke, 1973.
*Refer to Figures 4 and 5 in Chapter 3 for AFM ter-

nary diagrams of the Egyptian alkaline complexes.

Figure 17.
Composite calcic-alkali ternary diagram of several

well-known oceanic alkaline suites which are similar
to those of the Egyptian alkaline complexes, and
similar to trends expected for fractional crystal-
lization. Data from.Schmincke, 1973.

*Refer to Figure 6 in Chapter 3 for calcic-alkali
ternary diagram of the Egyptian alkaline complexes.

 

68

 

 

‘ ASCENSION

' ‘ IZORES '

 

 

 

 

 

 

ROSSSL

 

 

 

 

 

' 61. HELENA

Figure 16.

" ‘msm on CUNIIA "

 

 

N020 Wt.” COO
0C EANIC ALK ANNE SUITE 5

Figure 17.

rUA PUMAFOUESAS BL.

69

Figure 18.

Silica—alkali diagrams for a selection of alkaline
suites from continental, oceanic, extensional, com-
pressional, and intraplate tectonic settings. These
trends are similar, and there is no apparent syste-
matic difference according to crustal setting. See
text for further explanation. Data compiled from:

1. Tristan da Cunha, Atlantic Ocean,

2. Ross Island, Antartic Ocean, Goldich et al., 1975.

3. Gough Island, Atlantic Ocean, Le Maitre, 1962.

4. Tahiti, Pacific Ocean, McBirney and Aoki, 1968.

5. Ua Pu, Marquesas Island, Pacific Ocean, Bishop and Woolley, 1973.

6. Reunion Island, Indian Ocean, Upton and Wadsworth, 1972.

7. Hanish-Zukur, Red Sea, Gass et al., 1973.

8. Jebel al Abyad, Saudi Arabia, Baker et al., 1973.

9. Kenya Rift, Africa, Saggerson, 1970; see also Lippard, 1973;
King, 1965; King and Chapman, 1972.

10. Monchique, Portugal, Rock, 1978.

ll. Sintra, Portugal, Sparks and Wadge, 1975.

12. Cuttingsville, Canada, Laurent and Pierson, 1973.

*See also Figure 29 for another silica-alkali trend from
the Canary Islands which is very similar to that of G.
Abu Khruq in Egypt.

**Refer to Figure 3 for the silica-alkali diagrams of the
Egyptian alkaline complexes.

 

70

 

 

 

 

 

 

S 3 3
3.” x. :q
S 8 8
a. a! 2.
6— 0— S
“c “a s
. m ,
a a m 5 m 9
o— ' >— D r—
3 3 3
° 5 8 5
o P D - g D
N 2 _ N b N “
x 0 x D x
" 4»
* s 6 8. t.
o v o o f Qts o f
N N N
c a a d! c
z 8.15 z 2% Z 6.
6. B." '5"
aI'D-(m ID.OO “1.70 70-09 III-DC 53.00 ID-OD It-DC 70-Dt It.f0 °‘C-0€ ICJ‘C I5.“ 7: f: 02.0:

5102 HT PCT - TRISTRK

 

8102 HT PCT - RUSS ISL.

 

 

 

8; s

:7 :-

8; a.

2' 2‘
6— .—
L.) c.)

8 8
m 9 a E a
o— o—
: D 3 m
D 8. m D 8. m U D
N 1“ o N 3‘.“ E]
x D x m

D

*6 Us” *6 o
D ‘4 D D ‘4 U 8
s s D
2 g 2 8 ‘

.7 'H 9‘

E! 6
.0
56.66 66.66 66.66 76.66 66.66 °66 66 66.66 66.66 76.66 66.66

5102 NT PCT - TRHITI

Figure 18.

5102 HT PCT -UR PU

E'J NE-NORMRTIVE
(D QTZ-NORMFITIVE
A SIOZ—SFITURFITED

5102 HT PCT - DOUGH ISL.

NRZO + K20 HT PCT

‘1
Q

341
56 .66 “do €6.66 76.66 66.66

5102 NT PCT —REUNION

 

 

N920 + K20 HT PCT

 

 

a 2.
a z
a s.
It 8‘
.— I—
; LJ
3. E 8. m
25‘ 27
.. h-
3 3
3 c>8 o
2:" N 27 W (\I
x E I
8 5’ 8 m
0‘ D o I“ D
(\J N
a (gm a
8;. z 84 2
C C ‘
S“ 6
$6.66 66.66 66.66 76.66 66.66 ”66.66 66.66 66.66 76.66 66.66

71

 

 

5102 HT PCT-HRNISH-ZUKUR 5102 HT PCT -RL RBYRD

".00

I

II.”

ano + K20 ur Pcr
O.”

b

 

aflfiL

(32

(0
6—6

G

1"
~00

66.66 66.66 710.00 I'0.0C

HT PCT - HONCHIQUE

Figure 18 (cont'd).

 

8
8
8"
8
21
A
8
s 6%
‘ngrvr 9‘:
34 E) 96 0.
8 fin
£7 qfizib
J.
”66.66 66.66 66.66 76.66 66.66

 

5102 NT PCT -KENYR

 

 

 

 

5102 HT PCT - SINTRR

[I] NE-NORMRTIVE
O OTZ-NORMHTIVE
A SIOZ-SRTURRTED

8 8
8. a:
21 84
-— o—
i 8. E a
_ P h,9
3 3 m
— 3 mg c,8 s3
— :57 % :r A
2’ "’6 2 LB 68
I 8 U 0 'P 8 3 O
N 2
.§ 6% 2.8 O
.70 ‘ O"
y) g)
546.66 66.66 66.66 76.66 66.66 56.66 66.66 66.66 76.66 66.66
CUTTINGSVILLE

72

.Amhma

..Hm um mafic: "mmuoN< Eoum name name .mmcwuuwm Hmumouo
ucmummmwp >uw> Eouw mpcmuu unmEmHm momuu o3» wmmcu cw
mmaon ucmscowucw pcm mmmcmu coflumuucmocoo umawcwm
wuoz .ua>mm .Usucx and .0 can mmuoN¢ 6:» Beam muwsm

mcwamxam cm MOM mfimummwp Sowumo 00H msmum> Edwuocu won

 

 

 

 

.mH whomflm
H mmmo~¢ u 2mm I» earn: 3mm I tam :h
o S m: S O" o_ a: S
N PEPPb b P— T—bphh» b b P0 ~>~h>bb h ~>AI I N ngb PL" bbbhth b no bh-bbbh b ~P' I
no nu
I 0 T o
L1 m...
013 01.3
. 3 . 3
a
1 d a a v d
a n d as u d
9 AU nU
Ban 8 .. Z ”U + u 2.
a . _h( 6 .
T ‘ .6
nu nu
E E

73

Figure 20.

Ce/Sm versus Ce diagrams comparing the REE enrichment
and fractionation trends of well-known oceanic alkaline
suites and that of G. Abu Khruq, Egypt. Though G. Abu
Khruq demonstrates more scatter as a single suite, the
concentration ranges and enrichment slopes are very
similar. The larger amount of scatter is attributed to
the plutonic nature of G. Abu Khruq. Data on the oceanic

suites is compiled from:

Azores, White et al., 1979.
Reunion, Ludden, 1978; Zielinski, 1975.
Gough, Zielinski and Frey, 1970.

Grenada, Arculus, 1976.

U'luwaH

Ross, Sun and Hanson, 1976.

 

74

 

 

3.
'1
8 x 61 626666
6:7 o REUNION
A OOUGR
8 + GRENROR
' x x R055
3‘ x
8.
z- X
z 0
6. 3. x x
n. '“ x
I:
w 8.
\ :7 x
u.)
6.: £1 x m
+
8 )3 a m X
M
' >13 “7% c? D
s “x Q; :6
. D
0‘ i
19
6. ‘
Q I .
$.66 66.66 766.66 766.66 666.66 666.66 666.66 666.66 600.00 466.66
CE PPM - OCERNIC SUITES
8.
”T
g D GRBBROS
g. o UNDERSMURRTED SYENXTES
0 OVERSRTURRTED SYENITES
8 + VOLCRNICS
£4 x ovensmunmso BIKES
0 UNDERSRTURBTED BIKES
8.
:4
t s,
g; 8'
z' .
(D 8 D o
I: .' “00¢” o 0
0
u m o .00 oo 0
:1 :0 g h. . o * +
'- m o 19‘. ‘* ‘ ‘
O D 0293 t + t... A
a“ I!) g x “
O I!)
m A
'5.
8.

 

 

5.66 66.66 766.66 666.66 666.66 666.66 666.66 666.66 {66.66 666.66
P. CE PPH - RBU KHRUO
J~gure 20.

Fifi

 

75

their observation of the worldwide compositional regulari-

ties of alkaline suites. By referring to the diagrams

presented here and in Chapter 2, one can note the similarity
of these chemical trends to those of the Egyptian alkaline

complexes. This similarity exists on both the major element

and trace element levels, and appears to persist between
alkaline occurrences found in different crustal settings.
For the data on the Azores alkaline suite in particular,

(1979) concluded that the dominant mechanism

The

White et al .

<3f differentiation was fractional crystallization.
strong similarity of the REE and Th data between the Azores
and Abu Khruq in terms of both enrichment lepe and concen-
‘tration ranges, may indicate that fractional crystallization

was an important mechanism in the melt at Abu Khruq also,

Figure 19 .

2) Fractional Crystallization in Relation to the Trace
Eleunent Distributions of the Egyptian Alkaline Complexes
Thegeneral depletion trends of the compatible trace
elements and the general enrichment trends of the incom-
Patible trace elements in the four Egyptian alkaline com-
Plan-(es of this study are similar to those expected for the
differentiation of an alkaline magma in which crystal

fractionation was an important mechanism.

1" Compatible Trace Elements -- Gabal Abu Khruq
Ni, Cr, and Co are strongly incorporated into the

mafic mineral phases. Ni is most strongly partitioned into

 

76

the structure of olivine, Cr into pyroxene and Fe-oxides,

and Co into olivine and pyroxene, but to a much lesser

extent than Ni or Cr. The distribution of Co with respect

to Th (used as the magma crystallinity index) for Abu Khruq

is shown in Figure 21. The distributions for Ni and Cr

are similar, but show an even sharper depletion (see Chapter

3). The lack of exposed mafic rock-types at Abu Khruq does

not provide well-defined trends for these compatible trace
elements with ferromagnesian affinities.
Sr and Ba are especially sensitive to feldspar frac-

tionation. Sr serves as an indication of the fractionation

of particularly plagioclase; Ba is strongly partitioned
into the structure of alkali feldspar (e.g. Hanson, 1978;

Baker et al., 1977) . Variation diagrams are shown for Sr

and Ba in Figure 22. The Sr trend demonstrates the effi-

cient fractionation of plagioclase into the gabbros with
the resulting severe Sr depletion in the majority of the

e"“Posed rock-types. Ba is interpreted to show a curve

Which is typical for a rock series in which alkali feldspar
be<:c>mes the latter dominant fractionating phase and depletes

the remaining melt in Ba. Note the inferred similarity to

the Gregory Rift alkaline volcanic series shown in the
inSet of Figure 22b, where Zr is used as the index of
crYstallinity.

The five compatible trace elements presented are all
interpreted to show the general trends expected for the

CI:YS.ta1 fractionation of an alkaline basaltic magma. The

77

Figure 21.

Cobalt versus thorium (index of crystallinity) for G. Abu
Khruq. It is interpreted to illustrate the efficient frac-
tionation of the ferromagnesian minerals early in the evo-
lution of the melt at G. Abu Khurq.

Symbols are the same in this figure and Figure 22.

OL-crx cun.

aflBBROS

UNDERSHTURRTED SYENITES
OVERSRTURRTED SYENITES
VOLCRNICS

X+bGG

Figure 22.

Trace elements with feldspar affinities versus thorium

(index of crystallinity) for G. Abu Khruq.

a) Strontium versus thorium; interpreted to illustrate
the efficient fractionation of plagioclase. See Figure
26 for a similar presentation with a crystal fraction-
ation path superimposed.

b) Barium versus thorium; interpreted to illustrate effi-
cient fractionation of alkali feldspar during the inter-
mediate stages in the evolution of the melt at G. Abu
Khruq. The inset diagram is from Baker et a1. (1977) of
data from the alkaline volcanics of the Gregory Rift. It
serves to illustrate with less scatter than the intrusive
Egyptian data, a similar type of Ba pattern (using Zr as
the index of crystallinity). As is common in the litera-
ture on alkaline suites, this pattern is attributed to
the fractionation of alkali feldspar during the intermed-
iate stages of fractional crystallization.

Symbols are the same as in Figure 21.

78

C0 PF"
69 66

6.10

0 +

g : .“é '4‘ ’k A
"6.66 6166 I .66 r t .66 66.66 56.66 36.66

TH PPM - RBU KHRUO

 

 

 

Figure 21.

 

2.” 166.66 I’D.”
(3

fl"

SR
66

a
. '*‘§x «J

a $.66 6*.66 66.66 T66 .66 66.66 F6666
TH PPH — RBU KHRUQ

 

 

 

 

 

 

 

Figure 22.

79

lack of mafic rocks at Abu Khruq and the number of analyses
does not allow a more rigorous treatment of these compatible
trace element distributions. However, based on the order

of crystallization of the major phases in a basaltic magma
and the compatibility of these trace elements for these
individual phases, these depletions are interpreted to be
the result of efficient crystal fractionation of especially
olivine, diopside, labradorite, and alkali feldspar--a11

observed in the mineralogy of the rocks at Abu Khruq.

B. Incompatible Trace Elements -- The Four Alkaline
Complexes of this Study

Alkaline suites are characteristically bimodal in the
distribution of their major elements (e.g. Petro, Vogel and
Wilband, 1979). As is evident in many of the major element
diagrams, the Egyptian alkaline complexes appear to be
well—represented by the salic and some mafic compositions,
with few intermediate compositions. Figure 23 is a fre-
quency distribution of Th, a strongly incompatible trace
element, for 188 samples from the four alkaline complexes
sampled by this study. Although there is a sampling bias
due to the method of collection, there is no hint of a
bimodal pattern which coincides with the maficesalic bimodal
distribution of the major elements. The lack of such a
compositional gap is consistent with fractional crystalliza-
tion. It may be inferred that the bimodal distribution of

the major elements is simply a function of efficient crystal

80

 

 

 

 

 

 

 

 

Th pom

Figure 23.

Frequency diagram of thorium concentrations from four
Egyptian alkaline complexes (188 analyses from G. El
Naga, G. Nigrub E1 Fogani, G. El Kahfa, and G. Abu Khruq).
There does not appear to be a bimodal thorium distri-
bution (incompatible trace element) which coincides with
a mafic versus salic bimodal distribution as noted

in the major element diagrams for these and many other
alkaline suites in other areas. Frequency plots of
individual alkaline complexes are similar to this
composite diagram.

81

fractionation and the formation of distinct intervals of
crystallization of the major phases (see Clague, 1978;
Wood, 1978).

A major problem in solving the petrogenesis of many
alkaline complexes is the origin of the gabbros which are
typically associated with the syenites and alkaline volcan-
ics. Upton (1974) suggested that the REE data might be
used to test whether these gabbroic bodies can be derived
by extensive plagioclase crystal fractionation from a
basaltic magma. Although this study's sampling does not
include any plausible parental basaltic representatives,
the overall sub-parallel REE-enrichment and Opposing Eu
anomalies of the gabbros and the syenites/volcanics are
typical of plagioclase separation (Eu-rich) and a residual
Eu-depleted liquid.

The sub-parallel nature of the normalized REE plots
for the entire suite of major rock-types at Abu Khruq is
illustrated in Figure 24. Assuming a mantle source region
containing garnet, such a gradual sub-parallel enrichment of
REE is unlike that expected for a partial melting relation-
ship (which tends to produce large LREE/HREE changes), and
similar to that eXpected for the fractional crystallization
of the dominant minerals observed in the rock-types at Abu
Khruq (clinOpyroxene and feldspars).

Figure 25 shows Ce versus Ce/Sm for the four Egyptian
alkaline complexes of this study. The REE enrichment (Ce)

versus the REE fractionation (Ce/Sm) trends of Abu Khruq,

82

16’

GRBBRO

HLKRLINE SYENITES
0T2 SYENITES
VOLCRNIC

NE SYENITES
FDYHITES

‘0 G)-<-O D E!

.10

A

\V/\
\. -6

Ill

A

 

A

 

 

l"3'3"!“ I ZED QBUNDRNCE

j A l

L

9 LR CE 8" EU TB YB LU

D j T fl 1 1r 1 1 —I 1 1 1 1 1 1 T
56 S7 58 59 60 61 62 63 64 65 66 67 68 69 7D '71 '72
RTOHIC NUMBER

 

 

 

Figure 24.

Normalized REE plot of the major rock-types at
G. Abu Khruq. Note the overall sub-parallel REE—
enrichment pattern which supports a comagmatic
relationship for this rock suite. In view of

the mineralogy of these rocks (predominantly
feldspars and clinopyroxenes), and the lack

of minerals which cause large REE fractiona-
tions, this sub-parallel pattern is consis-

tent with a fractional crystallization relation-
ship rather than one involving partial melting.

83

Figure 25.

Ce/Sm versus Ce diagrams for some of the Egyptian alkaline

complexes. These types of diagrams represent REE-fractiona-

tion versus REE-enrichment. Arrows are intended to show the
general directions of enrichment/depletion via the fraction-
ation of the dominant mineral phases.

a) G. El Naga: this pattern is similar to that in which
amphibole is an important crystal fractionation phase.
The arrow is arbitrarily drawn to illustrate the effect
of fractionating amphibole as a single phase.

*
Where D = 1.44 and D = 3.21
Ce Sm

b) G. El Kahfa: this pattern is typical for the fractiona-
tion of clinopyroxene and plagioclase. The arrow is arbi-
trarily drawn to illustrate the effect of fractionating
70% plagioclase and 30% clinopyroxene.

*

Where DCe= 0.27 and DSm= 0.13.... for plagioclase,
*

and DCe= 0.50 and DSm= 1.67.... for clinopyroxene,

to give the following bulk distribution coefficients:
DCe= 0.34 and DSm= 0.59.... for 70% plagioclase
and 30% clinopyroxene.
c) G. Abu Khruq: this pattern, similar to G. El Kahfa, is
typical for the fractionation of clinopyroxene and pla-
gioclase. The arrow is arbitrarily drawn to illustrate
the effect of fractionating 70% plagioclase and 30%
clinOpyroxene. The distribution coefficients are the

same as those used for G. El Kahfa in (b).

 

* Distribution coefficients are taken from Sun and Hanson
(1976) and Arth and Hanson (1976).

84

0

03:3. :0: .. an... mu

 

 

 

 

 

 

 

.61. 6.1. 6.1. 6.! 6.3. 6.1. 6.! 63F 6.6km... .mN mufimflh
aux-o 3.33—3:; 0 a
3;... 823.33... x
3.63.: o M
3:35 333.33». 6
3:35 8.82339... 0
no: 0 r6
a
C
.-
«J
6 n.
. m...
6 u
u H
m
.8
Q .. w
”:25. d .. :5... mu .1 mac: .5 u 5... mu
I6! I.‘ 8.61 I6! 8.61 a... 8.1. 8.1. Ii If... a 8.1. 8.]. .661 8.61 86! 86! 8.3. 8.1. 8.1
a
7.. a
a ..
a
:-
a ....
a a a
I «14. B a...“
nu / a I T /
. w W a w
a u a 6 6 a
w H a o H
.u v.”
c a
J. v.
p“ a
n T

 

 

.—

85

El Kahfa, and El Naga are similar to those expected for the
crystal fractionation of the observed dominant minerals;
that of Nigrub E1 Fogani is difficult to evaluate due to the
scattered nature of its predominantly altered rock-types.
The Ce versus Ce/Sm patterns for Abu Khruq and E1 Kahfa are
typical of clinopyroxene and plagioclase crystal fractiona-
tion; that of E1 Naga is characteristic of amphibole crystal
fractionation (refer to Arth and Hanson, 1976; Tanaka and
Nishizaqa, 1975; Arth and Barker, 1976; Irving, 1978; Sun
and Hanson, 1976).
C. The Relationship between Incompatible and Compatible
Trace Elements -- Gabal Abu Khruq

Figure 26 shows Sr, a compatible trace element in
feldspar-rich rocks, versus Rb, an incompatible trace ele-
ment, for all the major rock-types from G. Abu Khruq (Rb-Sr
data from Lutz, 1979). As can be seen by the superimposed
fractional crystallization and partial melting curves,
fractional crystallization can account for the sharp deple-
tion of Sr, whereas the partial melting curve cannot. These
curves have been drawn based on partition coefficients
which appear to be reasonable for Abu Khruq, as well as
being similar to average values often used in the litera-
ture (refer to Allegre et al., 1977; Frey et al., 1978;

Baker et al., 1977). Reasonable changes in the relative

bulk partition coefficients will not alter these relation-

ships.

86

Figure 26.

Sr versus Rb for the rock series at G. Abu Khruq.

The rapid depletion of Sr, a compatible trace ele-

ment, cannot be explained by a partial melting

relationship. As can be seen by the superimposed
partial melting and fractional crystallization

curves, the data is more closely approximated

by the fractional crystallization relationship.

Data from Lutz (1979).

a) This diagram shows only the syenites and volcanics
from G. Abu Khruq. The gabbros are excluded because
they are considered to have a cumulate origin. Cal-
culated partial melting and fractional crystal-
lization curves are based on:

BSr = 3.0 (a compatible trace element for alkali

feldspar and plagioclase-rich rock suites).

Bab = 0.1 (a known strongly incompatible trace element).

b) This diagram includes the gabbros. If the gabbros
did not have a cumulative origin, the Sr depletion
is more pronounced, and the partial melting rela-
tionship deviates from the data trend even more
than in diagram (a).

RB PPM

80.00

RB PPM

00

40.

120.00

87

8
:3
a-
8 ID UNDERSRTURFITED SYENITES
E. (D OVERSRTURRTED SYENITES
A VOLCRNICS
8
52'.

l

J

 

 

 

46.66 66.66 166.66 166.66 266.66 266.66

SR PPM - HBU KHRUO

MBBROS

UNDERSRTURHTED SYENITES
OVERSRTURRTED SYENITES
VOLCRNlCS

TINOURHES

X+DGB

m 6?

 

 

66.66 66.66 66.69 66.66 65.66 666.66

SR PPM -'Hbu KHRUO Figure 26,

88

D. Crystal Fractionation Modelling (See Appendix II for a
more detailed model.)

Detailed trial-and-error crystal fractionation modelling
in terms of the compatible and incompatible trace elements
was attempted for the data from Abu Khruq. Minerals used
in these fractionation models included olivine, clinOpyrox-
ene, plagioclase, Fe-oxides, alkali feldspar, apatite, and
nepheline (nepheline appears to have partition coefficients
similar to alkali feldspar, Eby, 1975). The calculations
were performed on both calculated relative bulk partition
coefficients (after the method outlined by Allegre et al.,
1977) and published partition coefficients. The choice of
the parent magma's composition was approximated by noting
the chemical characteristics of the Egyptian suites and
comparing them to many other similar alkaline suites which
contain more mafic end—members. Although the use of 5
compatible trace elements (Ni, Cr, Co, Sr and Ba) and 4
incompatible trace elements (Ce, Sm, Lu and Th) served to
constrain the models, the lack of an exposed mafic sequence
of rocks at Abu Khruq did not allow a unique, rigorous
solution. Consistently however, these models yielded very
large amounts of crystal fractionation: about 2/33 of a
moderately alkaline basaltic parent (similar to the average
Hawaiian alkaline basalt) needed to be crystallized in
order to produce the gabbros as cumulates, and greater than
90% of the original parent needed to crystallize in order to

produce the bulk compositions of the syenites and volcanics.

89

Table 3. First Approximation of Fractional Crystallization

Based on Incompatible Trace Elements
(See Appendix I for a more detailed presentation).

*

Element C0 C1 . 01 D F

Ce 44.00 184.97 188.27 0.46 0.07

Sm 5.53 16.38 16.50 0.57 0.08

Lu 0.23 0.91 0.94 0.51 0.06

Th 2.90 18.52 18.04 0.23 0.09

Co concentration of element in postulated alkaline
basaltic parent magma.

Cl concentration of element in calculated remaining
liquid after crystal fractionation.

01 concentration of element in the observed rocks.

5 bulk distribution coefficient for respective
elements. Derived by Allegre et a1.'s (1977)
method which is explained in Chapter 4, Sec-
tion III, 4. They compare closely to Allegre
et a1.'s (1977) coefficients in another alka-
line suite.

F fraction of original magma remaining after

crystal fractionation and which contains the

concentrations of the elements listed in Cl'

the observed concentrations are based on an
average of 40 analyses.

Alkaline Basaltic Parent Magma

crystal fractionation

Bulk Salic Composition (Average
of Foyaites, Quartz Syenites and
Volcanics).

90

The incompatible trace elements are not as affected by
the early stages of crystallization as the compatible trace
elements. Ignoring the early enrichment of the incompatible
trace elements during the initial mafic stages of crystal-
lization simply results in a more conservative fractionation
model for only the incompatible trace elements. By comparing
the compositional trends of the moderately alkaline, silica-
undersaturated rock series at Abu Khruq to those of similar
alkaline suites with more mafic members, a moderately alka-
line basaltic parent melt similar to an average Hawaiian
alkaline basalt was chosen (Schilling and Winchester, 1969;
Kay and Gast, 1973; Frey et al., 1978). The bulk partition
coefficients for the REEs were calculated via the method
outlined by Allegre et a1. (1977) (see pp.123-24). The bulk
partition coefficient for Th was derived by taking an average
of the values used by Baker et al.(1977) for the early mafic
and late salic stages of differentiation in the Gregory Rift
alkaline volcanics. The "Observed" concentrations of the
derived liquid are a result of averaging 40 analyses of
syenites and volcanics (an estimated bulk salic composition).
The derived F values in this scheme all indicate that greater
than 90% of the original alkaline basaltic parent must
crystallize before the estimated bulk salic composition is
attained.

Crystal fractionation models of other salic alkaline
complexes have resulted in similar extreme amounts of

crystal fractionation. Engell (1973), for example,

91

demonstrated that if fractional crystallization accounts
for the evolved rock-types of the Ilimaussaq alkaline com—
plex in Greenland, about 75-95% of the original augite
syenitic magma must have crystallized before the last
sodalite-rich foyaite stage was reached. This is the same
rock-type which is thought to have ended the magmatic
activity at Abu Khruq. If this amount must be fractionated
from an augite syenitic magma to produce such rock-types,
then even a greater proportion must be separated from an
alkaline basaltic parent magma.

The relatively small volume of the Egyptian alkaline
complexes does not create as great a problem in explaining
reasonable amounts of the original basaltic magma undergoing
fractional crystallization as in the more voluminous alka-
line occurrences such as the East African Rift volcanics
(see Baker et al., 1977; Baker and Henage, 1977). If the
volume of the evolved rocks at Abu Khruq (syenites and
volcanics) is approximated by a cone with the maximum
diameter of Abu Khruq (8 km) and an estimated thickness of
1.5 km (based on a reconstruction by El Ramly Et al., 1969a),
and if these evolved rocks represent the final 10% of the
original alkaline basaltic parent melt, then abOut 754
cubic kilometers of alkaline basaltic magma is needed for
the initial parent melt volume. This would mean that a
simplified cylindrical magma chamber about twice the diameter
of the preSent complex (16 km) would have to be about 3.75

km in height. The magnitude of these dimensions are

92

considered small enough to allow the possibility of in situ
fractionation of a primitive alkaline basaltic magma at

shallow crustal depths below the complex.

Section III

 

The Silica-Undersaturated/Qversaturated Problem

 

Introduction

 

One of the major problems in petrology involves rock
associations which contain both silica-undersaturated and
silicaeoversaturated rock-types. The critical plane of
silica-saturation in the basaltic tetrahedron (Yoder and
Tilley, 1962) appears to effectively control the direction
of fractionation so that the evolution of the liquid is
strongly dependent on the initial composition of the parent
(see Coombs, 1963). This prevents the transition of a
basaltic magma from silica-undersaturated to silica-over—
saturated compositions, and vice versa, via low-pressure
crystal fractionation. This feature can be observed in the
Q—Ne-Ks system of salic rock compositions in which the
alkali feldspar join forms a thermal barrier at low pressures
so that quartz-normative and nepheline-normative liquids are
prevented from passing from one side to the other via
crystal fractionation (Bowen, 1937). This problem is pertin-
ent to the development of many of the alkaline complexes
in the Eastern Desert of Egypt, as well as many other

alkaline suites throughout the world.

93

1) The Common Occurrence of Silica-Undersaturated/Over-
saturated Associations

There are many examples of silica-undersaturated/over-
saturated alkaline suites of different ages and different
crustal settings. Similar to the overall worldwide chemical
regularity of many alkaline suites (e.g. Schwarzer and
Rogers, 1974), silica-undersaturated/oversaturated associa-
tions tend to form regular trends in terms of both major
elements and trace elements. Figure 27 shows a summary of
the silica-alkali trends based on this study, as well as
work by Schwarzer and Rogers (1974), Schwarzer (personal
communication, 1979), Upton (1974), Miyashiro (1978) and
others. The accompanying diagram from Upton (1974) shows
a similar regular silica—undersaturated/oversaturated trend
in the Gardar alkaline province of Greenland. The course
of these trends appears to be strongly dependent on the
initial composition of the parent magma (see Coombs, 1963).
There are three fundamental trends; the Egyptian alkaline
complexes sampled by this study fall into Trend 2 types.

Trend l: The first trend shows the differentiation
pattern in silica-oversaturated alkaline suites in which a
silica-oversaturated or hypersthene-normative mafic parent
gives rise to slightly alkaline rock-types, ending with
highly—evolved quartz-rich differentiates.

Trend 2: The second trend consists of a slightly
silica-undersaturated parent giving rise to more alkaline

rock-types than Trend 1, and in many suites of this type,

94

Figure 27 .

Summary of silica-alkali trends.

a) Summary of the silica-alkali trends noted for
the Egyptian alkaline complexes and alkaline
suites in other areas. See text for explanation.

b) A compilation of the silica-alkali trends noted
in the Gardar alkaline province of Greenland
(after Upton, 1974). This diagram demonstrates
a similar regular occurrence of silica-under-
saturated/oversaturated trends.

95

ALKALIC

20.00

1

18.00
A

N820 + K20 NT PCT
2

 

" THOLEIITIC

3
6.66 66.66 66.66 76.66 66.66

5102 NT PCT - GENERRL TRENDS

 

‘0 .66

 

 

 

a
AFTER 6610611974) , o"

F— 25. ' CJLBVV
O
a.

N
'—
3 l5
3“")
O s
“29‘

l l0

 

 

 

 

:6 . 66 66 60 76
8:02 WT PCT-GAFDAR-GREENLAND

Figure 27 . b

96

the differentiates are both silica-undersaturated and
silica-oversaturated (see Miyashiro's (1978) "straddle
type").

Trend 3: The third trend begins with a highly silica-
undersaturated parent and leads to the develOpment of very
silica-undersaturated, highly alkaline differentiates, and
in some cases, some silica-oversaturated rock-types.

This study noted that there is widespread occurrence
of alkaline suites with an initially silica-undersaturated
parent giving rise to both silica-undersaturated and over-
saturated differentiates. However, similar unambiguous
examples of an initially silica-oversaturated mafic parent
rgiving rise to both varieties was not noted in this study.
In some cases of si1ica-undersaturated/oversaturated assoc-
iations, the most mafic rock-type was silica-oversaturated,
but these were fairly well-differentiated rocks and not
basaltic representatives (e.g. Pankhurst et al., 1976).

It is possible that these salic complexes are underlain by
silica-undersaturated, mafic rock-types.

Figure 28 shows a summary of these silica-undersaturated/
oversaturated trends noted in the Egyptian and other alka-
line suites in terms of Ce versus Ce/Sm. The consistency of
these silica-undersaturated and oversaturated trends in
terms of the major elements, and even on the trace element
level, is considered to reflect a regular mechanism whereby

the silica-saturation transition is accomplished.

97

66.66

CE/SM PPM
69."

If.”

 

 

 

#66 66.66 666.66 666.66 36.66 36.66 666.66 666.66 666.66 666.66
CE PPM - GENERAL TRENDS

Figure 28.

Summary of the types of Ce/Sm versus Ce trends which
this study noted to regularly occur in many alkaline
suites throughout the world, including the Egyptian
alkaline complexes.

98

The similarity between silica—undersaturated/over—
saturated associations is particularly striking upon compar-
ison of plutonic alkaline complexes, such as the Egyptian
complexes of this study. There is a remarkable similarity
between the Pajara alkaline ring complex on the island of
Fuerteventura, Canary Islands, and Abu Khruq in Egypt. This
similarity can be seen in the major and trace element chem-
istry, structure, mineralogy, and even the "secondary"
alteration features (Munoz, 1969; De Paepe et al., 1971;
see also Fuster et al., 1968; Gastesi, 1969). The Pajara
alkaline ring complex contains the same type of radial
gradient in varying silica content, from nepheline syenites
in the center, to quartz-rich syenites on the outer peri-
meters. Figure 29 shows the silica-alkali diagram for the
syenitic and cumulate rock-types (gabbros and ultramafics)
of the Pajara ring complex. Figure 30 demonstrates the
similarity between the same rock-types from Abu Khruq and
the alkaline rock-types of Fuerteventura, in terms of
normalized REE plots.

The recently discovered alkaline complexes of the
Velasco province in Brazil are very similar to the Egyptian
alkaline complexes. They contain the same types of silica-
undersaturated and oversaturated rocks, with a similar
radial arrangement (see Darbyshire and Fletcher, 1979).

The alkaline province in southern Greenland is similar
to the Egyptian province in general structural setting, as

well as in the petrography and chemistry of its alkaline

99

Figure 29.

Silica-alkali diagram for the Pajara alkaline ring
complex in the Canary Islands (Fuerteventura). This
trend, as well as many features of this Canary Island
complex, is very similar to the silica-alkali diagrams
of the Egyptian alkaline complexes--G. Abu Khruq in
particular (Figure 3, Chapter 3). Pajara data from
Mufioz (1969) and Gastesi (1969).

Figure .30.
Normalized REE plot comparing similar rock-types from

Fuerteventura, Canary Islands and G. Abu Khruq, Egypt.
Note the close resemblance, even on an incompatible
trace element level. Fuerteventura data from De Paepe

et al. (1971).

100

3.
8. 61 66666667666766
3‘ o 6V66667666766

A VOLCHNICS

76.66

.6653

01 66
B3

N820 + K20 HT PCT
7; 66

 
    

6.00
l

RING COMPLEX
8f

‘6 .66 66.66 66.66 76.66 66.66

5102 HT PCT-CRNRRY IS.

 

Figure 29.

ABU KHRUG&CANARY ISL.
"a REE COMPARISON

 

 

 

5 ~— c-"I'MmaMaven-66c». am:
7 --..-. . .
1 "3‘“ W M clubs. Only m
‘ O
1 ‘ .'~
\I .....
7 . .....
\‘ ...... o. .
U ..Q. Q
o ‘1
w": ..
6;: .
a i
u
N
"‘ 1
E
E:
"I
4 *
7 ‘ ...........
7 ‘1’ 6666667666766 Val"
‘ X WTMTIHVIJEO
. LR CE 8" EU ya LU
o In

 

 

66 67 66 6'9 £6757 6'2 63 34 6'6 66 67 66 69 7D 71 72
' RTOMIC NUMBER

Figure 30.

101

complexes (e.g. Upton, 1974). The Kangerdlugssuaq
intrusion in this alkaline province may prove to be very
important for understanding the petrogenesis of these
silica-undersaturated/oversaturated alkaline complexes. It
consists of the same rock-types as Abu Khruq, but rather
than having a conical faulted ring structure, it is an
intact, largely unfaulted intrusion containing only grada-
tional petrologic boundaries, and a continual gradient of
Si02, Nazo and K20 in a radial fashion (Pankhurst et al.,
1976). Thus, without any sharp discontinuities, the inner
silica-undersaturated foyaites grade into slightly silica-
undersaturated syenites withtq>to 5% feldspathoids; these
grade into slightly silica-oversaturated syenites with less
than 2% modal quartz, and finally, these syenites give way
to quartz syenites with up to 10-15% modal quartz. A
similar radial gradient appears to exist at Abu Khruq and
many other plutonic alkaline complexes, but in a disrupted,
faulted pattern (see Lutz, 1979).

It is also important to point out that there are plu-
tonic alkaline complexes with similar rock-types, but
different apparent orders of crystallization. Tilley (1958)
noted that alkaline complexes often are composed of silica-
undersaturated centers surrounded by silica-oversaturated
perimeters, or alternatively, silica-oversaturated centers
surrounded by silica-undersaturated perimeters. The rela-
tionship between the syenites and gabbros also appear to

vary in apparent order of intrusion (e.g. Eby, 1979). These

102

different field relationships between very similar rock-
types are puzzling in terms of a regular petrogenetic model.
This study considers that any proposed model on the differ-
entiation of one type of intrusion must also allow for
these other variations in apparent order of emplacement and

crystallization.

2) The Silica-Undersaturated/Oversaturated Problem in the
Egyptian Alkaline Complexes

As was shown in the previous section, there is a con-
siderable amount of data which may be interpreted to indi-
cate that crystal fractionation was an important process
in the evolution of the Egyptian alkaline complexes. How-
ever, the relationships between the salic rock-types present
four main observations which cannot be explained by simple
crystal fractionation:
l. the field relationships of the silica-undersaturated
and oversaturated rock-types,
2. the REE-depleted and REE-enriched stages of develOpment
in the silica-undersaturated and oversaturated rock-types,
respectively,.
3. several chemical trends which are consistent with the
simultaneous low-pressure development of the silica-under-
saturated and oversaturated rock series,
4. the incompatible trace element relationships which
indicate that the REE in the silica-oversaturated series

behaved as though they were more mobile than the REE in

103

the silica-undersaturated series, and further, that the
HREE in both series have been partitioned as though they
were unusually mobile relative to the other REE.

Each of these observations deviate from that exPected
for crystal fractionation, and is considered to be related
to the comagmatic deve10pment of the silica-undersaturated
and oversaturated rock-types. The mechanism of this
deve10pment must be compatible with the data supporting
crystal fractionation. The regular occurrence of these
types of alkaline associations in the Eastern Desert of
Egypt, as well as many other parts of the world, is consid-
ered to be indicative of a fairly consistent process leading
to the evolution of silica-undersaturated/oversaturated
alkaline associations. This fairly consistent process is
considered to involve a thermogravitational diffusion pro—
cess and its attendant fluid effects. The following pages
will review the nature of the silica-undersaturated/over-
saturated probelm in the Egyptian alkaline complexes, and
the mechanisms which have been proposed in the literature

to eXplain silica-undersaturated/oversaturated associations.

A. Major Element Trends in Relation to the Silica-Under-
saturated/Oversaturated Problem

Figure 31 displays the silica-undersaturated/over-
saturated problem in terms of normative nepheline and quartz
with reSpect to Mgo for the rock series at Abu Khruq. This

shows that both the silica-undersaturated and oversaturated

104

a
..
a.»
“a:
m-
M
0
“V

 

 

 

n " 9 9
81
2° n
2.
I!
30.6
1 i is i i 347
MGOWTPCT
Figure 31.

Normative ne and qtz versus Mgo (index of crystal-
linity) for the rock series at G. Abu Khruq. Note

the progressively diverging salic trends of silica-
undersaturation and oversaturation from the silica-
undersaturated mafic rock-types. g=gabbro: ncsilica-
undersaturated syenites; ossilica-oversaturated syen-
ites; v=volcanics.

105

trends converge toward the more mafic silica-undersaturated
compositions. The silica-undersaturated rock-types cover
the entire range of MgO concentrations; the silica-over-
saturated rock-types cover only the less mafic compositions.
The most mafic syenites of both silica—saturation levels
were collected from apparently continuous rock bodies.

The major element variation diagrams in Chapter 3
demonstrated the highly-evolved, salic nature of the majority
of the rocks at Abu Khruq. Although more samples of mafic
syenites need to be analyzed for major elements, some
general conclusions may be derived from these major element
trends.

The CaO,_Ti02, and P205 vairations are consistent with
fractional crystallization of a mafic parent melt. These
element oxides are included in the early crystal phases:
calCic plagioclase, calcic pyroxene, titanium oxides, and
apatite. Due to the predominance of salic compositions in
the exposed rock-types at Abu Khruq, these element oxides
are not expected to show much variation. The Si02,A1203,

Na 0 and FeO

0,_K variations cannot be interpreted as

2 2 total
a result of simple crystal fractionation of a single magma.
These element oxides all have bifurcating trends corres-
ponding to the silica-undersaturated and oversaturated

series, similar to the ne/qtz versus MgO diagram of Figure

31.

106

B. Incompatible Trace Element Trends in Relation to the
Silica-Undersaturated/Oversaturated Problem

Figure 32 clearly illustrates the separation of the
silica-undersaturated and oversaturated trends at Abu Khruq.
It becomes apparent that the silica-undersaturated and over—
saturated trends begin from similar concentration ratios and
progressively diverge along separate directions. This
divergence becomes more noticeable for the HREE. An indica-
tion of the initial similarity between the less-evolved
silica-undersaturated and oversaturated rock-types may be
seen in the normalized REE plots of Figure 33. The more
evolved rock-types, however, gain increasingly greater
difference in terms of silica—saturation levels and REE
concentrations. For a given thorium concentration, the
silica-undersaturated rocks are depleted in REE, and espec-
ially HREE, relative to the silica-oversaturated rocks.
The silica-undersaturated rocks, however, attain similar
high Th concentrations, and much higher REE fractionation,
i.e. higher Ce/Sm and Ce/Lu ratios. The relative REE
depletion in the silica-undersaturated series and the
development of two separate REE trends cannot be explained

by simple crystal fractionation.

C. The Apparent Order of Crystallization in Relation to
the Silica-Undersaturated/Oversaturated Problem
The apparent order of emplacement and crystallization

of the different rock-types in the Egyptian alkaline

107

Figure 32.
REE versus thorium diagrams for G. Abu Khruq. Note

the progressive separation between the silica-under-
saturated and oversaturated trends with increasing
Th concentrations (index of crystallinity). The most
Th-enriched silica-undersaturated rocks (foyaites)
are REE-depleted relative to the most Th-enriched
silica-oversaturated rocks (quartz syenites and

volcanics).

LR PPM

66 666

YB PP"

108

 

8.
0
a.
3. :6 6666666
.3. 0 66666661666166 6166116:
A 6V66661666166 SYEMTES
g + V6L666166 A
9
24
+
8.
§~ .. .
8 A +* (3* é:
o + p O C‘
21 ‘ c
3” c
450 g 0
‘5 35
3‘ %£%

 

fi

 

 

 

 

6.66 26.66 313.00 (2.6: 63.6: 6'6.
TH PPM - REU KHRUO
3.
:6
E1 MBBROS
8, 0 66666661666166 61661166
9‘ A 6V66661666166 61661166
+ VOLCRNTCS
8 A
8‘
+5
5 + .
+ +
8 ‘5 3}
9 T o 95
A:‘@ (D 0 as
D 0
. . 9.6“ o
0‘ 0% 0
$6»
30
5.66 16.66 66.66 66.66 66.66 €6.66 €6.66
TH PPM - FIBU KHRUO
8.
3.
3‘
I ‘ A
9; A
«6
6 . *
.2.
t" o o
t A
! °*‘ 1: o
.d ‘00
+9.00 0 o
1 ‘ 0
' 69333 +
‘0‘
I
~ofl a.» 6.66 3.66 3.66 3.66 66.66

TH PPH - RBU KHRUD

Figure 32.

CE 666

TB PPM

LU PP"

 

 

 

 

 

 

3.
Sn
6 61 6666666
§q 0 66666661666166 SYENITES
A OVERSRTURRTED SYENTTES
g + VOLCRNICS ‘
O
#4
a A
2 +‘ +
z. ‘ O
+ + O C
8 +
67. ”0‘00 e
8 ;£000
2: 6 If“
3.
51.00 111.66 :6 .66 ab .66 36 .66 66 . 66 6‘6 .66
TH PPM - RBU KHRUQ
8.
“ 16 6666666
0 UNOERSRTURRTEO SYENITES
g A A OVERSRTURRTED SYENITES
°‘ + voLcmucs
8 ‘ ‘
' A
4* A
8
'd ‘
a “ +
8 3 ++
‘ o
" 6°61 o 0
m “=6 0 ° 0
.1 0%” q” o
O
3.
$166 a.” 76.66 35.66 66.66 3.66 66.66
TH PPM - RBU KHRUQ
2
N'-
5. 63 6666666
0 UNUERSRTURRT'ED SYENTTES
O OVERSRTURRTED SYENITES
g + voLC66Ics
2. M T *
‘8‘ o
0 o
3. 0 0
.‘ é%% o
‘Wflp
5.6666416 66.66 66.66 $.66 €6.66

jTH PPH - FIBU KHRUQ

109

 

d“

‘U

4\

666L1266 666666666 2
A ‘0
1,

.\

LA

 

V
‘ 9
35L 9 “
i 9
co LR CE 511 EU TB YB LU

 

 

 

— I j 1 I I T T 1 7 T T I j 1
56 5'7 58 59 60 61 62 63 54 65 66 6'7 68 £9 70 '71 72
RTOMIC NUMBER

Figure 33.
Normalized REE plot showing the similarity between

the more mafic, less incompatible-rich syenites of
the silica-undersaturated and oversaturated rock series.
Diamonds = silica-undersaturated syenites
Triangles = silica-oversaturated syenites.

llO

complexes is extremely difficult to incorporate into a
petrogenetic model. For example, the order of intrusion/
extrusion at G. Abu Khruq according to E1 Ramly et a1o(l969a)
was:

The volcanics (silica-oversaturated) initiated the mag-
matic activity, followed by the emplacement of the gabbros
(silica-undersaturated), then the alkaline syenites (which
included both silica-undersaturated and oversaturated ‘
varieties in El Ramly et a1.'s rock classification), then a
slightly later emplacement of the quartz syenites (silica-
oversaturated), and finally ending with the intrusion of
the foyaites (strongly silica-undersaturated). This appar-
ent alternation of silica-saturation levels cannot be
eXplained by a simple crystal fractionation model. Thus,
either this interpretation of the field relations is
incorrect, and/or a more complicated petrogenetic solution

than simple crystal fractionation exists.

D. REE Depleted/Enriched Stages in the Development of
Silica-Undersaturated/Oversaturated Complexes

In an igneous sequence derived by crystal fractiona-
tion, the initial stages of eruption are expected to be the
least REE-enriched, and the final stages of crystallization
are exPected to be the most REE-enriched. Figure 34 shows
the average Ce concentrations (representative of total REE)
for the major rock-types at Abu Khruq in their apparent

order of crystallization. As can be seen, the siliceous

111

Figure 34.

REE concentration levels (represented by Ce) in the
major rock-types of G. Abu Khruq arranged according

to their apparent orders of intrusion/extrusion. Sim-
ilar to the pattern noted by Balashov (1972) in Figure
33, the last rocks to crystallize at G. Abu Khruq (foy-
aites) are relatively depleted in the REE with respect
to the previous stages of crystallization. Similar to
the enriched nature of peralkaline flows early in the
magmatic history of an area, the REE-enriched volcanic
cone formed early in the development of G. Abu Khruq.
These depletion and enrichment levels with respect to
. the apparent order of intrusion/extrusion are not con-
sistent with crystal fractionation.

Ce ppm

112

  

   

 

:80 V
VOLCAIICS
800 't- ”In 81061115
FOYAITES
050'"
PULAIKITEO | ALIALK MINES

no u-
66 it «mums

6 t : 6 4 *6

 

Apparent Stages of Crystallization
(Early----------------~ -------------- Late)

Figure 34.

113

volcanics represent the most REE-enriched rock-type at Abu
Khruq in spite of their early formation in the deve10pment
of the complex. The presence of such evolved, REE-enriched
effusives before the crystallization of the more mafic
rock-types lower in REE, cannot be eXplained by fractional
crystallization. The deve10pment of such siliceous REE-
enriched volcanic or ignimbrite flows early in the magmatic
history of an area is not unique to Abu Khruq. This pattern
has been frequently noted in peralkaline volcanic areas

(Hildreth, 1981; see Bulletin Vo1canologique, Special Issue -

 

Peralkaline Rocks, v. 38, #3).

The foyaites at Abu Khruq present a similar problem in
terms of the expected REE distribution fOr a melt evolving
via crystal fractionation. Similar to their relationship
in many alkaline complexes, these highly silica-undersatur-
ated rocks form the last stage of crystallization. Despite
their highly-evolved major element chemistry and their
extreme depletion in compatible trace elements, they contain
lower abundances of REE than the previously formed silica—
oversaturated volcanics and quartz syenites, Figure 34.
Balashov (1972) demonstrated that intrusive alkaline com-
plexes typically end their magmatic activity with the forma-
tion of rock-types relatively depleted in REE with respect
to the previous stages of crystallization, Figure 35. The
regular occurrence of this depleted REE pattern in the
final stages of crystallization of alkaline intrusions

demands attention. According to the observations of this

114

Figure 35.
REE depletion in the last phase of intrusive activity

which Balashov (1972) found to typically occur in
many alkaline complexes.

115

 

1000f:

3

l / \
LU
Ll%nfl—
I :
w ..

«J REE DEPLETION IN LAST INTRUSIVE PHASES

. IN ALKALINE MASSIFS
2011
After Balashov,(1972).

 

 

 

Phase I 2 3 4 5

Figure 35.

116

study on si1ica-undersaturated/oversaturates suites, there
appears to be a consistent enriched versus depleted REE
relationship between the silica-oversaturated and under-
saturated sequences, respectively. These relationships
must be taken into account in any petrogenetic model on the

evolution of these types of alkaline melts.

3) Silica-Undersaturated/Oversaturated Models
There are several models which have been proposed to
account for the deve10pment of silica—undersaturated and
oversaturated rock—types in close proximity. Four of these
models are rejected for the Egyptian alkaline complexes
based on the primitive isotOpic data (Lutz, 1979), the
radiometric age-dating of the rock-types within individual
complexes (Serencsits et al., 1981; Lutz, 1979), the nature
of the chemical trends, and the magnitude of the change
required in the silica-saturation levels for the Egyptian
alkaline complexes.
1. Crustal contamination models.
2. Resorption mechanisms (see Luth, 1976; MacDonald,
1974; Edgar, 1974; Gittins, 1979).
3. Fractionation of silica-deficient phases (see Oftedahl,
1948; Muir and Smith, 1956; MacDonald, 1974)
4. Liquid immiscibility (see discussion in Chapter 2).
There are three remaining possibilities for the origin

of the types of silica-undersaturated/oversaturated complexes

117

observed in Egypt. Two of theSe involve high preSsure
fractionation, and the other invokes thermogravitational
diffusion and fluid effects. High pressure fractionation
is discounted; thermogravitational diffusion and attendant
fluid effects are considered to be responsible for the
deve10pment of a silica-oversaturated rock series from a
silica-undersaturated magma.
5. High pressure fractionation

Experimental work has demonstrated that at high pres-
sures, the thermal barrier between silica-deficient and
silica-excess systems is lowered (Schairer and Bowen, 1935;
Tuttle and Bowen, 1958). This allows for two suggestions
on the origin of silica-undersaturated/oversaturated
associations which involve high pressure fractionation.
a. The first possibility involves the coincidence of a
silica-undersaturated magma and a silica-oversaturated
magma in the same localized area, either commingling or in
discrete episodes. This involves the emplacement of separ-
ate batches of melt from a common magma at very great
depths, or separate batches of partial melt from a source
region at very great depths. The regular occurrence of
similar silica-undersaturated/oversaturated alkaline
complexes in Egypt as well as other areas, the general lack
of sharp contacts between the major rock-types, and the
nature of some of the chemical trends, all serve to dis-

count this type of simultaneous two-magma model.

118

b. The second possibility involving high pressure frac-
tionation makes use of high vapor pressures within shallow
magma chambers in order to explain the transition of silica-
saturation levels in many alkaline associations. The
thermal barrier between silica-undersaturation and oversatur-
ation may break down at pressures greater than 10 kilobars
(Morse, 1969, 1970). As a result, some petrogenetic models
on silica-undersaturated/oversaturated associations have
implemented models of very high vapor pressures in the

magma chambers (e.g. Pankhurst el al., 1976). Alkaline
magmas tend to have high COZ/HZO ratios (e.g. Mysen, 1976),
low H20 solubilities (Kogarko, Burnham and Shettle, 1978),
and large amounts of halogens (e.g. Bailey and MacDonald,
1975; Baker, et al., 1977). The effect of these volatiles
on silicate magma systems is uncertain, and more experi-
mental work is needed in order to fully evaluate their
influences in alkaline magmatic systems. However, it is
unlikely that such high vapor pressures may be attained

in shallow crustal magma chambers. If vapor pressures build
up enough to lower the thermal barrier at shallow depths,
the vapor separation and circulation cannot lower the
temperature of the magma very much or its effect on lowering
the thermal barrier will be counteracted (see Pankhurst

et al., 1976). More importantly, even if the thermal
barrier can be lowered sufficiently, a chemical mechanism

is still needed to cause the magma to change its composi-

tion toward the new silica—saturation levels.

119

Figure 36 supports plagioclase fractionation which
can occur at pressures equivalent to depths above about
25-30 kilometers (plagioclase stability limit, Green and
Ringwood, 1967) with a rapid decrease in oxygen fugacity
(close to the QFM buffer, Weill and Drake, 1973). (See
Thorpe et al., 1977; Baker et al., 1977; and Kay, 1978 for
interpretation of these types of diagrams.) Thus, these
trends are consistent with low-pressure fractionation for
the rock series at Abu Khruq.

EXperimental data indicates that the agpaitic order
of crystallization (the dominant felsic minerals crystal-
lize earlier than the dominant ferromagnesian minerals)
which occurs in the Egyptian and other similar alkaline

complexes, cannot occur at P O greater than 100-200

H
atmospheres (about 0.2-0.3 kiiobars) (Kogarko, Burnham and
Shettle, 1978).

The field relationships including gradational, unre-
acted petrologic boundaries, the miarolitic textures in all
the rock-types, and the shallow nature of the collapsed
sub-volcanic ring structures (e.g. Bahat, 1980), all support
low-pressure differentiation.

The evidence against high pressure fractionation and
the fact that low pressure crystal fractionation cannot
account for the development of both silica-undersaturated

and oversaturated trends,_1eads to the conclusion that

another process other than fractional crystallization must

120

 

EU/SH N
1..

 

I!)
‘6‘.» 416.66 $.66 $6.66 €66.66 366.66 636.66 666.66

SR/SM PPM - RBU KHRUO

 

3.
D \
1‘ m m
2 D
x 8.
U) -‘
\
D
u . 0
mm

 

" f
O

3.}! 8
$166 6166 7.66 6166 8'.” 7.66 $.66 6 .66

HBO NT PCT - RBU KHRUO

Figure 36.
moeelemtanduptraxdswhicharemideredtofldicate
low—pressure crystal fractionation (see Thorpe et al., 1977;
Baker et al., 1977a, Kay, 1978). The diagrams are based on the
plagioclasestabilitylimitandtheaffinityvhichthesetraoe
ailments display for plagioclase.

121

be responsible for the origin of these silica-undersaturated/
oversaturated complexes.

6. Thermogravitational diffusion and attendant fluid
effects.

This is the final possibility which is considered for
the origin of silica-undersaturated/oversaturated complexes
such as those in Egypt. Several mechanisms have been
suggested to be responsible for the deve10pment of liquid
fractionation in alkaline magmas. Desilication of an
oversaturated magma via volatile transfer was suggested by
Currie (1970) in order to explain the silica-transition of
some of the alkaline complexes in Canada. Highly differen-
tiated, halogen-rich alkaline melt systems of long-term
crustal residence appear to provide the ideal conditions
for a fairly stable thermogravitational process similar
to those suggested for peralkaline, siliceous effusive rock
suites (Hildreth, 1977, 1979, 1981; Saggerson, 1970). There
is evidence that excess silica and other components will
tend to accumulate in the roof and outer portions of certain
types of volatile-rich magma chambers (e.g. Hildreth, 1977,
1979, 1981). Although structural and magmatic events may
disturb the radial chemical gradients which are set up by
convective circulation, Soret diffusion, and the movement
of discrete vapor phases through the magma, it is possible
that near-equilibrium mass transfer of certain components
in these alkaline systems may occur along similar composi-

tional trends as those directed by the overall

122

differentiation of the magmatic system. More specifically,
this may involve the gradual change from solid-liquid
equilibria to a gathering influence of solid-liquid-vapor
equilibria. This approach to the silica-undersaturation/
oversaturation problem is considered to be the most consis-
tent with the available data on the Egyptian alkaline

complexes.

4) A Quantitative Approach to Assessing the Nature of
the REE Distributions

Use of the REES as petrogenetic indicators in evolved
alkaline melts is of special concern because these systems
contain large amounts of alkalies and volatiles which may
have a strong effect on both the primary REE distribution
in the crystallized phases, as well as the alteration of
these distributions due to possible metasomatic effects
from the increasing prOportions of these components in the
residual liquids (Flynn and Burnham, 1978; Mysen, 1976,
1977a, 1977b; Mysen et al., 1975; Kogarko, 1974, 1977;
Kogarko et al., 1978; Balashov and Krigman, 1975; Wendlandt
and Harrison, 1979; Wyllie, 1979; Hards, 1976).

A preliminary precaution on assessing the petrogenetic
validity of the REE is to make note of any known REE-com-
plexing minerals which formed early in the crystallization
history of the rock suite. Two REE-rich minerals which are
common in the Egyptian alkaline rocks are apatite and

magnitite (see Schock, 1979). By comparing the P205 and

123

FeO concentrations to the REE distribution in these

total
alkaline rock series, it was concluded that neither mineral
exerted a significant control on the overall REE distribu-
tions. No other unusually REE-enriched minerals have been
observed in the Egyptian alkaline complexes (see also Lutz,
1979, p. 84; Fryer and Edgar, 1977; Miller and Mittlefehldt,
1982).

As was presented in the previous pages, the REE dis-
tributions of the rocks at Abu Khruq appear to bear a
relationship to the deve10pment of the silica-undersaturated
and oversaturated series. If the REE reflect their primary
magmatic distributions, it is possible that a detailed
study of their patterns may yield evidence of the mechan-
ism(s) which caused the development of the silica-under-
saturated and oversaturated rock series. In the last few
years,_a quantitative approach has been develoPed in order
to more critically evaluate the nature of the trace element
distributions. This approach was initiated by Anderson and
Greenland (1969), and then more fully developed by Treuil
and Varet (1973), and then by Allegre et al.(1977),

Minster et al (1977), and Minster et al.(1978) in a three
part series (see also Allegre and Minster, 1978; Hanson,
1978).

The result of this quantitative approach is to arrive
at some calculated bulk partition coefficients for suites
of trace elements in relation to a well-established

"hygromagmatoPhile" or highly incompatible reference trace

124

element (see Treuil and Varet, 1973; Allegre et al., 1977;
Wood et al., 1979). For an explanation of this method, the
reader is referred to Allegre et al. (1977).

Thorium was chosen for the highly incompatible refer-
ence trace element. Several factors make Th an excellent
reference trace element for the Egyptian alkaline complexes:
it was analyzed with high precision (INAA), it very rarely
has a preferred mineral phase, there is lack of large
amounts of radioactive mineralization in the Egyptian alka—
line complexes (Hussein and Hassan, 1973), and Th is one of
the most immobile trace elements (Adam and Gasparini, 1970).

Figures 37, 38, 40 and 41 show the log Th-log REE plots
involved in the method outlined by Allegre et al. (1977)
for the four Egyptian alkaline complexes of this study.

The relative bulk partition coefficients for the REE from

these plots have been calculated only for Abu Khruq.

Gabal E1 Naga

The log Th-log REE plots for El Naga show an extremely
well—defined trend, Figure 37. This complex contains the
least amount of evidence for widespread autometasomatism or
alteration of the four alkaline complexes sampled by this
study. E1 Naga is the only complex of this study in which
amphibole is the dominant ferromagnesian mineral. In spite
of these differences, the general nature of its geochemistry
is similar to the other Egyptian alkaline complexes. The

differences in terms of alteration, ferromagnesian phases,

125

ma
[3

 

 

 

 

1
.4
N '1 D
0 Law
= = dd.
m .
0... d
c__ .
"’ D
'“E
o i
C)
I I I‘VIIII I T1IIYI‘I'1 I I IIITYII
-1 0 1 2
10 10 10 10
TH PPM - EL NRGR
N
CD
o—l: m
' m
z—- m
0.. C) m
m ““1 oma:fl%
5 E d:;;9m
J
D
O
F‘ :‘1 I T‘TIII' 0' j Y IIIVUI {I I FITTTII 2
10 10 10 10
TH PPM — EL NRGR
Figure 37.

CE PPM

LU PPM

 

 

 

 

('0
C)
F? m
E m
“' ‘ J
O 185%.
C)
D .-
O
l Tvrrilll I 1111111] 1 I IIHIII
- 0 1 2
10 10 10 10
TH PPH - EL NRGH
N
D
'43
1
J
O.
"E
;
.J
° 0
CD 61
F? m
1 ”Ghana
'7 - m
D
H 1 1111161] 1 IVIIIII' TWUYWII
-l 0 1 2
10 10 10 10

log REE versus Log thorium for G. El Naga.

TH PPH - EL NRGR

126

and the well—defined REE trends may all be explained as
the result of the retention of the volatile phases in the

magma at El Naga. P is thought to be the dominant

vapor
control on whether amphibole or pyroxene lies on the

liquidus (Ernst, 1968; Ferguson, 1978a, 1978b; Stephenson,
1972; Larson, 1976). The retention of the volatile phases
in the magma at El Naga would have caused the crystalliza-

tion of amphibole rather than pyroxene, and would have

avoided the alteration caused by otherwise migrating fluids.

Gabal Nigrub El Fogani

At the present level of exposure, Nigrub E1 Fogani
contains predominantly high—evolved, altered rock-types,
including carbonatite dikes. The presence of carbonatites
at this complex attests to the importance of COZ—rich
phases in the deve10pment of this alkaline melt. Experi-
mental work indicates that CO2 and carbonate complexing of
the REE may have a significant effect on the final REE
distributions of rocks crystallizing in COz-rich systems
(Wendlandt and Harrison, 1979; Balashov and Krigman, 1975).
Large-scale, erratic effects on the REE distributions of
other carbonatite-rich alkaline complexes have been
attributed to this COZ-complexing (ema. Eby, 1975). A
similar effect may be reflected in the scattered REE

trends of Nigrub E1 Fogani, Figure 38.

127

 

 

 

 

 

 

F) m
C) D
E! v—l‘: m
I
. m
N N
C) CD mm “b
m . m m
1:thmi E E m m
m m # 9 m
.. m m w__ .
C) U o
m D "E
o o ‘
O O
fl—r T T T TTUTII l T VITIIII T I V TTTIT] "T l I I IIYIII 1 fl TiTT‘r‘r‘ f I T TflUT'
- 0 1 2 -l 0 1 2
10 10 10 10 10 10 10 10
TH PPM - EL FOGRNI TH PPM - EL FOGRNI
N N
D CD
'4‘: m
‘ o
‘ m
C) m
FIT—j D Em fig 0...
2 mm 2:, m
: m " o m
‘ mm o B
‘ 6190”!)
- m a slum
m _
O , a 6:”
O CD
j T TWTTIY] I T TTTYTTI I I 1 TTTTTI "I Y TWTUVII' T j I 1111'! I I I ITUYII
-1 0 l 2 -l O l 2
10 10 10 10 10 10 10 10

TH PPM - EL FOGRNI TH PPM - EL FOGRNI

Figure 38.
log REE versus log thorium for G. Nigrub E1 Fogani.

128

Gabal El Kahfa

In the log Th-log REE plots for El Kahfa, the trends
become increasingly less linear for the heavier REE, Figure
39. This "distortion" of the REE trends is considered to
reflect the effective presence of volatile-rich phases
which mobilized especially the HREE.

This complex is similar to Abu Khruq in age, mineralogy,
and LREE distribution. Figure 40 shows the similarity
between the two complexes in terms of Ce/Sm versus Ce. Both
trends--as discussed earlier (p.81)—-are similar to that
expected for the fractionation of predominantly pyroxene
and feldspar phases. However, El Kahfa lacks the large
amounts of sodalite (chlorine-rich) and nepheline-rich
syenites which occur at Abu Khruq. El Kahfa contains
large cataclastic zones, plugs of granulated rock contain—
ing inclusions of olivine and diopsidic pyroxene, along
with partially fused feldspar fragments. A large proportion
of the exposed rocks at El Kahfa have a similar fused and
altered appearance in thin section. The widespread deve10p-
ment of these features was not noted in the other complexes,
and suggest that El Kahfa's evolution involved large
quantities of fluid phases which preceded the emplacement
of the late stage dikes.

Based on the similarities between El Kahfa and Abu
Khruq, it is assumed that the alkaline melt at El Kahfa
contained high concentrations of chlorine. Depending on

the silica activity of a melt, the alternative to the

129

Figure 39.

IogREEversuslogthorimnforG. ElKahfa. Notethepro-
greesively mre anomalous REE concentrations in the heavier
REE.

130

 

 

 

 

 

 

 

 

 

 

 

m ("'1
C) Q
Hi PIE
2 2 m
d m -
cu ‘ N .
O O 0%
H: m v—JT. ~
5 = ‘° 0 a E = m m w
°‘ 1 ”a: m‘fi ‘5 °' - m %” mm
s— - 3- - . a... n
O m [D m D
H: D ..
E '5 mm 5
1 m 1
-« m -
o ‘ c: ‘
O C)
H -T11 ITTIW' 7 T ITUITTI l‘fi I 1111]] fl .1! I TIUIUUI 0‘ 1 [I'FTII 1T jTU‘T'II
10 10010 10210 10‘J 10 102
N TH PPM - EL KRHFR N TH PPM - EL KRHFR
(D O
HE c—l’g
: I
-1 [D "‘ d
C)
.. m "3
E" ‘9 ”35% z :
m 3 gang: E I ‘9'”
U) '_‘ (g: m +—° 0 Elm
- m D m C) no m;g%§
- E) F? m D D ID
- % 069 m E @500
. D . m
c: 7 ..
C) D E!
F. T I I IIIIII 0' 1 TITITI] 1| l TTTTTTI :11 7 III I 01 I Y TUUTTI If T I YTTTTI 2
10010 10:10 10010 10
N16 TH PPM - EL KFIHFR ND TH PPM - EL KRHFR
C)
69%!
z 6:
t "95, t
0
“3°C 00 ”a 3°C
I!
El
T
C) D
H—l ll fiIU'UUl a 1 TVII‘TI 1" Y I'VT‘I' 2__' '1‘ I‘V'Ull 0' U ‘UfiITYUTl 1' I VYITYTI
10100” 10 10210 100 10 10:2
TH PPM - EL KRHFR TH PPM - EL KRHFH

Figure 39.

131

.mwxwfiwcoo gag 5% 95 0mg c6953 muwggm
g .g :2 .w new mmcmm am .0 “Ow mu msmug ca\wo

 
 

 

 

. ov «Mama
oaxzx no: u zen mu
6 I. I. I. 8. I. .r mercy. d - can. no

r 8.61 0.65 0.61 66.61 66.61 8.1,. 0.3. 0.1. 8%- I...
w

.m
a .m

. w m a m a“.
Q .9 O X a O
6 6 I ‘80 6 I0 9 Iain“. z.
6 0 fl 9 D
o + 6 6 «i 0 906 F Mm. a grantees a

o 0 a I ..

O O I
O 6 a u M. a ..
u 1..
a

a a

nun T
3....— Sugafiug 6 ...u u
.u;_o cusses—coco». x a .6
«aliaso m

3:55 8:62.333 6 a
3:3: 33.3.8395 o .- .u
nelni a w

a
c a
u. u
a a

 

Hdd LIB/33

132

crystallization of Cl-rich minerals such as sodalite, is
the exsolution of NaCl—rich fluids (Koster van Groos and
Wyllie, 1969; Roedder and Coombs, 1967). The lack of the
crystallization of Cl-rich sodalite at El Kahfa is consid-
ered to have resulted in the release, perhaps eXplosively,
of large quantities of Cl—rich fluid phases from the magma.
The experimental work of Flynn and Burnham (1978) indicates
that Cl is perhaps the most efficient REE-complexing agent.
Due to lanthinide contraction, Cl complexes the REE with
greater efficiency as a function of decreasing ionic radii.
These effects are interpreted to be reflected in the log
Th-log REE plots of E1 Kahfa. The extensive signs of
alteration, the lack of sodalite-rich residual phases, the
increased mobilization of the HREE are all attributed to
the widespread activity of Cl-rich fluids in the deve10pment
of E1 Kahfa. This complex may represent an extreme of a
Cl-related process which may have occurred in other com-
plexes, such as Abu Khruq, and may be indicative of the
importance of the volatile components in these types of

salic alkaline complexes.

Gabal Abu Khruq

Figure 41 demonstrates the log Th-log REE trends
for the alkaline suite at Abu Khruq. The relative bulk
partition coefficients for the REE of Abu Khruq have been
calculated using linear regression analysis to obtain the

best-fit line for the data, Table 4. These caluclated

LR PPM

SM PPM

 

133

 

 

 

 

 

 

Log REE versus Log thorium for G. Abu Khruq.

OVERSRTURRTED BIKES

m m
D C)
I j .
N ‘ + ‘ N 4 ‘ {5:32
C) . 6 c: g
- o
: £13, g5?“ E : Mm
.. 5 EL _ U m
_. ~ 0 main 0 “4.. 4 D
c: m U C)
H: Fr:
3 a a
q .1
. I
o 4 o .
C) CD
fl T ITTTTTI fl 7 T TTTTT' T T TTTTTI] q——( 1 1 111111! —T T Ijfijtt‘l j’ fittilll
-1 0 1 2 - 1 2
10 10 10 10 10 10 10 10
TH PPM - RBU KHRUO TH PPM - RBU KHRUO
N N ’
O O
—*2
1
I . ..
"A O
‘ “A?
.. ‘5 lib
CD a m"¢+% E ‘ + ‘$t ‘
-—r' A*%y ‘L m ‘2’
“ mm 69 o m m E! “,7: +
: m 6'66!) Do 9
. 0 Lu 0 .
. o m Q90 0 ‘ as
.. m m
:6 ° £0 0 0
u o o
. o _
D I
C) D
T T T TTTTTI T T lrmq T T [1111]} "I T T TTIIITI 1* T TUTTTTr TfiEYTITIIl
-l 0 1 2 '1 0 1 2
10 10 10 10 10 10 10 10
TH PPM - RBU KHRUO TH PPM - RBU KHRUQ
D ORBBROS
0 UNDERBHTURRTED BYENITES
' O OVERSRTURRTED OVENTTES
Figure 41' + voLcnuxcs
X
0

UNDERSRTURHTEO 0 I MES

TB PPM

134

 

 

 

 

N N
O D
v—fE HE
E 3
1 1
— q _ A
C CD
a "‘i
Z : E: 2
: A #4.. ‘ Q- :
X & D
o " . o " A
CD gm «$00 o -J O at 49+ a
v-d'j D ‘bfl 0 Fri x 8%
.1 I!) [D I! D Q I a
: E, : so.”
. . :5! o
T 1 T 4 0
CD c3 gs
9" fi 1 TTTTWI 0| 1 7 ITYWI 11 I T YTTTT] 2 F1 11 T T Imtl Di 1 TTTTTTI If I 1 TTIYT] 2
10 10 10 10 10 10 10 10

TH PPM - RBU KHRUO TH PPM - RBU KHRUQ

1.02

l llllll

YB PPM
l l lllllAOl
GE
(9 O
k
Hg
.99.. :

 

 

$+
o m 603‘
O 0 Duo
"3
I
: U
T -
D
_ T T T'TT‘TI T T TTTTTT' fl T I TrrT—rl
-l 0 l 2
10 10 10 10

TH PPM - HBU KHRUQ

Figure 41 (cont'd).

135

Table 4. Calculated Relative Bulk Distribution Coefficients
for Gabal Abu Khruq

Correlation Standard

D Slope Coefficient Deviation Intercept
La .36 .64 .87 .13 +1.19
Ce .46 .54 .86 .12 +1.55
Sm .57 .43 .71 .16 +0.61
Tb .59 .41 .64 .18 -0.24
Yb .36 .64 .76 .20 +0.06
Lu .51 .49 .78 .14 -0.70

 

Method of calculating these bulk distribution coefficients
is after Allegre et a1. (1977).

 

D = bulk distribution coefficient.

136

values indicate that the REE other than Eu, are behaving

as incompatible trace elements. The relative bulk parti-
tion coefficients for the LREE (La, Ce, and Sm) demonstrate
steadily increasing values, as eXpected for their normal
behavior within a melt undergoing fractional crystalliza-
tion. These values are very reasonable considering the
mineralogy of the major rock-types, and compared to
published partition coefficients (see Allegre et al.,

1977 for comparison). The relative bulk partition coeffi-
cients for the HREE, however, are not consistent with

their normal behavior in a melt undergoing fractional
crystallization. These values indicate that the HREE were
behaving anomalously incompatibly.* Although the calculated
relative bulk partition coefficient for Lu is only slightly
less than Sm, the importance of these numbers is that

given the progression of bulk partition coefficients from
the LREE to the MREE--va1ues which are entirely reasonable
for normal crystal fractionation--Lu would be expected to
have a much greater value than the one observed. This
apparent mobility of the HREE at Abu Khruq is similar to
the extreme HREE mobilization at El Kahfa, although not as
pronounced. It is possible that a Cl-rich fluid system

similar to the one postulated for the development of

 

*The extremely low relative bulk partition coefficient for
Yb may reflect an analytical problem, or alternatively, an
extremely anomalous Yb behavior in alkaline melt systems.

137

El Kahfa, also existed at Abu Khruq, but due to the crystal-
lization of Cl-rich sodalite in the foyaites, was much less
pervasive.

Crystal fractionation may have been an important
process of differentiation in the Egyptian alkaline melts,
but as the volatile phases became progressively more abun-
dant with crystallization of the early solid phases, crystal-
liquid equilibria may have been gradually transformed to
crystal-liquid-vapor equilibria. With the present data and
modelling techniques, it is not possible to quantitatively
assess such a multi-phase system. The chemical trends
which were formed as a result of crystal fractionation
would have begun to deviate from their expected patterns
according to these vapor-fluid influences. In order to
gain insight into these deviations from the expected
compositional trends of crystal fractionation, calculations
using the Rayleigh relationship for normal crystal fraction-
ation were performed for the silica-undersaturated and
oversaturated series, respectively.

The objective of these calculations is to investigate
chemical relationships via the Rayleigh equation, and
thereby, perhaps observe the nature of the deviations from
normal crystal fractionation. Rather than using the
standard ”F" notation common to crystal fractionation
calculations (amount of parent melt volume remaining), it
is more apprOpriate to consider the analagous value in

these calculations as assumed indirect measures of

138

enrichment, or "E". Average Ce, Sm, Lu and Th concentra-
tions for individual rock-types were used in these
calculations. The foyaites were calculated as derivatives
of a pulaskite liquid (silica-undersaturated series);

the quartz syenites were calculated as derivatives of an
alkaline syenite liquid. The calculated relative bulk
partition coefficients were used for the REE and an arbi-
trary bulk partition coefficient of 0.10 was used for Th
(taken from Baker et a1. (1977) for the Gregory Rift
alkaline volcanics). Due to the objective of this scheme,
the "E" values will be allowed to vary in order to yield
the observed concentrations of the end-members. Table 5
summarizes the results of these calculations. It is clear
from the widely varying E values that the process of
differentiation between the mafic syenites and the salic
end-members of each series did not consist of simple crystal
fractionation.

First, it should be noted that the E values of the REE
for the silica-undersaturated series are larger than those
for the silica-oversaturated series. This reflects the
greater REE enrichment in the silica-oversaturated rock
series. In Spite of the foyaites having much lower REE
concentrations than the quartz syenites, their average Th
concentration is slightly higher than that of the quartz
syenites. This results in a slightly smaller E value for
Th in the silica-undersaturated series. There is also a

strong reciprocal relationship between the amount of

139

.uxou own .ouuv no aau nouosonnuaaua use>uo>ca cocume <6

 

mach.ou Hovw.o Ho.o mm.o :4
HHhm.c comm.o om.o v1.0 Em
«vov.a cmmo.o «h.o oN.o mu

 

mmfluflm fiUUOhsuNQHU>OIQUMHMm .HH

wouo.on mnvn.o mn.o ~o.o :4
adoo.c mavm.o on.° ch.c Em
owam.a Hvo¢.c hv.c mm.o mu

 

NOwHQm UfluflhfluflnHOUCDlQUwHMm .H

unwaveuuooo
sawucaouuou macaw o

unwoumucH
.Aaowuuu xoou coco uOu >Hououu

-aon “662665 .eeae. .Ho 66 ouoo-¢ 6:6 6«> cauussuaou.
nuauom xoom caucusuunuo>o can pouuusuoauooc3|60«~wm one
uOu poundsoaau macawofiuuoou nodusbeuunuo xasn o>nuoaom

.uxou 00. .h anon: ou anomanaco .uouoau us0i:o«u:o I a
.auc0u0«uuuoo sawusnwuuawc xnsa I a

.uxuOu vo>uoubo use :«
.pesvwa maucwaldu wouoHsUHau :«

.vunvwn

.uunhunsn mu «0
.mmuanncn o no
.mouxaocu on no
.monauocu n no

ucuuaq cw

omouo><++
umuuu><+
om6u6><66
monuo><6

 

QflUQHSHQQNO>O +

vm.o 0A.
no.0 Hm.
mm.o hm.
am.o cc.

m o

mwuecm>m uuuuso

ucoEOHO uo cauuauucuucou 6H0

usuflcao no scuucuucoucoo 6H0

acuEUHo no cOwucuucoosoo uou

nc.d £3
Ho.d 5A
ho.H Em
no.9 uU

 

pounuauoauoccauoanwm "aoa~o> u no Esm

o cv.n~ nv.nu Ho.ou as
0 -.~ ~H.H vh.o :4
o Hm.m~ no.o~ ~w.- Em
o a~.mm~ am.mo~ m¢.m~n no
Anouwcoam .muuecoam
uuuoso. wcwdaxat.
++~o do +00

 

. nouwcuam 6:332: "magnum caucusuomuosloozfim .HH

mv.o on.
mm.o Am.
mh.o hm.
mm.o ow.

m m

o oe.a~ sp.aa sn.o~ as
c vo.o .o.o a..o an
o n~.- o~.~H oo.m 5m
o a..am~ no.om~ Ho.mnn ou
.mouensom. Aaouwxuausa.
CIHC HU .00

 

mmuan>om . moat—mound "magnum kuuusunnuapcauaozwm .n

.dwzchMuoaou :mucaaux on» mean: unawusneuunqo
be can mum "cans: ant .o as nuwuom zoom vouch:
nuuuuo>o vca couausuunuoccanaouuqm on» cuoauon
magnuCOwuuqou u:62:0aucm\20euuqdoo ”cucuawoou .m Ounce

140

enrichment necessary for the two series. The summation of
the E values for each trace element results in a number
near 1.00. In the standard crystal fractionation equations
this represents 100% volume of the original parent melt; in
these calculations it is used as the "original enrichment
level" of the parent melt (see TableES). Therefore, with
respect to these calculations via the Rayleigh relationship,
the unusual REE-depletion in the foyaites is roughly equal
to the REE-enrichment in the quartz syenites. This may be
taken as an indication that the REE were preferentially
partitioned into the silica-oversaturated system, causing a
corresponding depletion in the silicaeundersaturated system.
Based on the homogeniety of the REE distributions within
each rock-type, it is likely that this occurred before
crystallization of both the silica-oversaturated rock series
and the silica-undersaturated rock series. A mechanism
other than crystal fractionation must be called upon, one in
which the REE components were transported in liquid state
from the silica-undersaturated portions of the melt to the
silica-oversaturated volumes. Unlike the E values of the
REE, the E value of thorium in the silica-undersaturated
series is smaller than that of the silica-oversaturated
series. The relative difference of the E values for Th,
however, is much less than those for the REE. This may
reflect the ineffectiveness of the transport mechanism with

respect to thorium, perhaps due to its extreme immobility.

141

These relationships will be discussed further in the next
section.

The relative bulk partition coefficients were calcu-
lated separately for the silicaéundersaturated series and
the silica-oversaturated series. These values are given on
the right-hand side of Table 5. The relative bulk parti-
tion coefficients for the silica-oversaturated series are
smaller than those for the silica-undersaturated series.
The apparent greater incompatibility of the REE in the
silica—oversaturated system may be interpreted to indicate
that the REE were allowed to become more concentrated in
the final silica-oversaturated liquids. The lower incom-
patibility of the REE in the silica-undersaturated liquid
and the inferred directions of REE-transport in this liquid
zoned system did not allow as great an enrichment in the

final silica-undersaturated liquids.

Summary of the Log Th-Log REE Plots

The log Th-log REE trends for the four Egyptian
alkaline complexes are similar to each other in terms of
their general degrees and directions of enrichment. The
nature of especially the LREE are similar to those eXpected
for an alkaline melt evolving by fractional crystallization.
The HREE trends for E1 Kahfa and Abu Khruq clearly reflect
a process(es) other than simple crystal fractionation. A
large control on the more detailed distributions of the

REE is considered to be exerted by the volatile and

142

halogen-rich phases which beCOme volumetrically more
important in the evolved, salic stages of the alkaline
melts. Part of the nature and eXtent of the changes brought
about due to these components depends on 1) their composi-
tion (i.e. COz-rich, Cl-rich), and 2) the extent to which
they are eventually released from the magmatic system.

This helps to determine the crystallization of amphibole

(E1 Naga) versus pyroxene (El Kahfa and Abu Khruq), the
extent of alterations and autometasomatism,_and the crystal—
lization of residual minerals such as sodalite. Although
many volatile species may have been important in the
evolution of these alkaline melts, the obvious abundant
presence of chlorine at Abu Khruq forms the basis for an
initial qualitative assessment of the halogen and volatile
affects. Widespread activity of Cl-rich fluid phases may
explain the extensive signs of alteration and unusual HREE
mobility (as predicted by experimental work). The importance
of the volatile phases has always been emphasized in the
petrogenesis of especially salic alkaline melts, but only
recently has there been a growing body of quantitative work
done on the specific effects of individual volatile compon-
ents. Although this lack of understanding does not allow
elaborate quantitative models, it is becoming increasingly
evident that the halogens and CO2 exert strong controls on
the differentiation of magmas, and especially highlyfevolved
melt systems rich in these components (e.g. Hildreth, 1977,

1979, 1981; Konnerup-Madsen and Rose-Hansen, 1982).

Section IV

 

Thermogravitational Diffusion and Fluid Effects

 

Introduction

 

Magmatic processes other than fractional crystalliza-
tion, partial melting, and magma mixing are difficult to
unambiguously test using the geochemical data and methods
available. There is considerable evidence to indicate that
the quiescent shallow magma chambers of the Egyptian alka-
line complexes allowed efficient crystal fractionation of
an alkaline basaltic parent melt. However, there are
regular mineralogic and chemical features found in many of
these alkaline complexes which cannot be explained by
crystal fractionation. Considering the large degree of
crystal fractionation necessary to produce the bulk salic
compositions observed in these alkaline complexes, it is
likely that the progressive enrichment of the volatile and
incompatible components had an important role in the differ-
entiation of espcially the salic end-members.

This study considers that a thermogravitational
diffusion mechanism is responsible for major features in
the Egyptian alkaline complexes--ones which cannot be
explained by crystal fractionation. Recently, Hildreth
(1977, 1979, 1981) has addressed the processes of thermo—
diffusion and volatile complexing in silicic magmas with
the intent to demonstrate quantitatively that compositional

gradients within liquid zoned melts have extremely

143

144

important implications for the evolution of especially
rhyolitic and grantic associations. This study considers
that alkaline silica-undersaturated magmas contain

chemical characteristics which may provide Optimum condi-
tions for the mechanisms which lead to magma zoning.
Although these mechanisms may result in different chemical
and mineralogic gradients in alkaline systems compared to
rhyolitic systems due to unique chemical and thermal charac-
teristics, alkaline si1ica-undersaturated/oversaturated
systems such as the Egyptian ones may provide some of the
most convincing evidence of liquid zoned magmas. The most
important feature which may be explained by thermogravita-
tional diffusion is the regular transition of volatile-rich
alkaline magmas from silica-undersaturated compositions to
oversaturated compositions. Such a mechanism may have
relevance to many primitive alkaline associations in Egypt

as well as other portions of the world.

1) Nature of the Mechanism

Little geochemical work has been done on quantitatively
assessing the mechanism by which a magma becomes zoned.
Peralkaline volcanic suites have often been explained in
terms of a liquid zonation rather than normal crystal
fractionation (Bryan, 1966; Saggerson, 1970; Bailey and
MacDonald, 1975; see Bulletin Volcanolggique, Special
Issue - Peralkaline Rocks, v. 38, #3; Hildreth,_198l).

Many early volcanic and ignimbrite flows which initiate

145

the magmatic activity of an area are more siliceous and
incompatible-rich than the eruptions which follow. The
composition of the different eruptions on a single volcanic
edifice sometimes appear to vary according to the level of
the eruption with respect to the volcanic cone. These
unusual compositional patterns have often been attributed
to the liquid distillation of a single magma in which the
lighter and more volatile components are thought to
accumulate near the tOp of the magma chamber, thereby
establishing a stratified column of magma. The composi-
tional gradients of numerous rhyolitic flows including
extreme trace element depletions or enrichments, cannot be
explained by crystal fractionation. Many of the chemical,
isotOpic, and field relationships, however, can be incor-
porated into a liquid zoned model (e.g. Hildreth, 1981).

Liquid zoned models have also been proposed for
plutonic complexes (e.g. Currie, 1970; see Hildreth, 1981).
The gradational zonations of the Kangerdlugssuaq silica-
undersaturated/oversaturated alkaline complex in Greenland
may represent an unfaulted, typeéexample of liquid zonation
in plutonic alkaline complexes (the complex is described
by Pankhurst et al., 1976).

87/86and del O18 isotopic gradients

The presence of Sr
in igneous suites may often be indicative of large-scale
mass transport in magmas (see Hildreth, 1979, 1981). Such
transport may be more feasible in alkaline magmas in

comparison to other types of melts. This is due to the

146

polymerized state of alkaline melts because of the low
concentrations of silica and high proportions of volatile
and halogen components (e.g. Baker et al., 1977; Mustart,
1972). Long-term, low-pressure crustal residence is ideal
for the development of efficient crystal fractionation and
a progressive enrichment in the volatile, halogen, alkali,
and incompatible components. This further decreases the
polymerization of the melts. With a slow quiescent evolu-
tion, temperature and density gradients are likely to
develOp. Thus, a combination of convective circulation,
Soret diffusion, and volatile complexing/diffusion all
aide in forming a thermogravitational diffusion process
which leads to the development of a liquid zoned magma.
Such zoning may be capable of maintaining long-term liquid

interfaces (Hildreth, 1977, 1979, 1981).

2) Evidence in Support of Simultaneous/Overlapping
Development of the Major Rock-Types

There are numerous factors which may be interpreted to
indicate a simultaneous or overlapping development of the
major rock-types at Abu Khruq. Such a deve10pment is
considered to support a liquid zoned magma.
1. The field relationships of the rock-types at Abu Khruq
and many other similar alkaline complexes, are often inter-
preted to indicate that there was an alternating crystalli-
zation sequence of silica-undersaturated and oversaturated

melts (e.g. E1 Ramly et al., 1969a). Explanations of such

147

alternations of silica-saturation levels may involve
separate batches of melt of contrasting silica-saturations.
This is unlikely, however, based on the gradational mineral-
ogic and chemical relationships at Abu Khruq, and the
frequent occurrence of similar alkaline complexes throughout
the Eastern Desert as well as in many other areas of the
world. An alternative model to such oscillating sequences
is the simultaneous deve10pment of both silica—undersaturated
and oversaturated compositions in a liquid zoned magma.

2. There are gradational boundaries between the pulaskites
and foyaites (silica-undersaturated varieties), between
alkaline syenites and quartz syenites (silica-oversaturated
varieties), and between pulaskites and alkaline syenites
(slightly silica-undersaturated and oversaturated varieties).
Such gradational boundaries are even more evident in less
faulted alkaline complexes (e.g. Kangerdlugssuaq, Greenland,
Pankhurst et al., 1976). A similar gradational relationship
may also explain the apparent lack of clear gabbro-syenite
contacts.

3. The compositions of the different varieties of syenites
may be used to support their simultaneous or overlapping
deve10pment. The trace element characteristics of the
pulaskites and alkaline syenites (slightly silica-under-
saturated and oversaturated) are very similar. The
gradations toward the foyaites and quartz syenites (strongly
silica-undersaturated and oversaturated) demonstrate

progressively diverging trace element characteristics. This

148

gradual chemical divergence is considered to reflect the
general radial chemical gradient from the inner portions
of the complex to the outer portions.

4. The composition of the volcanic cone is very siliceous
and incompatible-rich. Due to the early formation of the
cone, this is considered to indicate that a portion of the
melt at Abu Khruq was highly differentiated and silica—
oversaturated even in relatively early stages of the
alkaline magmatic activity.

5. The major element and trace element concentrations of
the quartz syenites are essentially identical to the
volcanics. This is considered to indicate that a portion
of the melt at Abu Khruq_maintained a similar silica-over-
saturated and highly differentiated compositions-some of
which was erupted early in the deve10pment of Abu Khruq, and
the remainder crystallizing later as plutonic counterparts.
Due to the detailed chemical similarity of the quartz
syenites to the volcanics (e.g. the REE), it is likely

that their process of differentiation was similar.

6. The general lack of evidence of large—scale reactions
between the different varieties of rock-types at Abu Khruq
is considered to indicate that the deve10pment of these
different compositions occurred under quasi-equilibrium
conditions within a single magmatic system. The absence

of reactions between the syenites of contrasting silica-
saturation levels may be used to support a near-equilibrium

transfer of silica throughout the melt. The presence of

149

secondary quartz in the quartz syenites, and the lack of
secondary quartz in the immediately underlying gabbros, for
example, is difficult to eXplain unless this alkaline magma-
tic system contained specific directions of silica transport.
7. The radial chemical gradient across the complex of Abu
Khruq is an extremely important observation. Lutz (1979)
discusses the presence of a radial isotopic gradient at

87/86 and del 018. These gradients

Abu Khruq in terms of Sr
are unlikely to represent crustal interactions because the
isotOpic ratios are within mantle-derived values (Lutz,
1979). An alternative explanation is that such isotopic
gradients reflect the mass transport of material within

the magma (Hildreth, 1979, 1981). Lutz (1979) also noted

a strong radial gradient in terms of SiOz, Na/K, and Sr.
Due to the somewhat asymmetric and faulted structure of
Abu Khruq, these apparent radial gradients need to be
studied more closely by detailed radial sampling and analy-
sis. Even so, the general radial arrangement of the major
rock-types and the radial distribution of silica are clear.
Due to this radial geometry and the presence of gradational
boundaries, it is likely that the chemical characteristics
noted for each rock-type also reflect this radial relation—
ship. Accordingly, this would mean a radial enrichment of
the REE from the inner portions of the complex to the
outer portions. This is similar to the REE gradient noted

by Hildreth (1977, 1979) for the liquid zoning model of the

BishOp Tuff.

150

3) A Preliminary Thermogravitational Diffusion Model for
Gabal Abu Khruq

Although there are numerous parameters which are
unknown in this model, there are many features which are
qualitatively consistent with a thermogravitational diffu-
sion mechanism for the differentiation of the salic rock-
types at Abu Khruq.

The thick, stable crust of the Nubian shield provided
a quiescent tectonic setting for the emplacement and
differentiation of the Egyptian alkaline melts. This
allowed the regular development of efficient crystal
fractionation, stable thermal and density gradients, and a
progressive enrichment of the alkali, volatile, and halogen
components of the melt. This type of magmatic system
provides the Optimum conditions necessary for the deve10p-
ment of a thermogravitational diffusion mechanism due to
the low-polymerized nature of such an enriched melt, and
the enhanced diffusivities of cations and complexed species
(see Hildreth, 1979). It is prOposed that thermodiffusion
and volatile compleXing caused mass transfer of Si, K in
preference to Na, Sr, Ba, the REE, and many other components.
These constituents were fluxed from the inner portions of
the magma to the roof and outer portions of the chamber.
This is depicted in the schematic diagram of Figure 42.

It is rather uncertain as to the details of how and
when a thermogravitational diffusion process might have

been established within the sodic alkaline melt at Abu

151

annrz42.

A schematic illustration of a thermogravitational
diffusion mechanism in shallow alkaline magma cham-
bers. Note the radial volatile-flux pattern and

the development of a liquid-zoned magma chamber.

Such zoning may not have contributed to the for-
mation of the gabbros early in the history of the
development of the melt at G. Abu Khruq. It is likely
that crystal fractionation was especially effective

in the early development of these alkaline magmas, and
it may be soley responsible for the formation of

the gabbros. Crystal fractionation may have continued
to operate in conjunction with the thermogravitational
diffusion mechanism. Continued fluxing toward the outer
crystallized margins of the complex resulted in the
deposition of secondary quartz and possibly other
components. This type of mechanism is considered to

be responsible for the radial chemical and isotopic
gradients in these complexes, and the overall develop-
ment of the silica-undersaturated/oversaturated asso-

ciations.

It should be noted that development of ring fractures
and pulses of magma from different volumes in these
types of liquid—zoned magma chambers may result in

a variety of configurations in these alkaline complexes.

152

  

p

> v -

. Volcanics
J

——~ ~«—— -— 1 Gabbros

Ultramafics

 

Figure 42.

153

Khruq. It is possible that its effectiveness was not
Operative until after the crystal fractionation Of the
gabbroic phases, resulting in the compositional drive
toward peralkalinity via the "plagioclase effect", as well
as an enrichment in the volatile and incompatible compon-
ents. It is also uncertain as to the importance of
fractional crystallization in the final stages of differ-
entiation. It is possible that a combination of thermo-
gravitational diffusion and crystal fractionation occurred
throughout most Of the evolution of these salic rock-types
at Abu Khruq and other similar alkaline complexes.

The early formation Of the evolved volcanic cone at
Abu Khruq is attributed to magma zonation. It is likely
that thermodiffusion and volatile complexing had already
exerted a profound influence on the melt at Abu Khruq with
the build-up of a silica-rich, volatile-rich cupola which
erupted and formed the siliceous, REE-rich alkaline volcanics
with widespread signs Of autometasomatism. Such an origin
for these volcanics may explain why almost their entire
volume consists Of silica-oversaturated compositions in
spite Of resting on, and being intruded by silica-under—
saturated plutonic rocks. If such a liquid zoned mechanism
accounts for the nature Of the volcanics, it is not unreason-
able that the same process affected other portions Of the
melt. This is, in part, supported by the very similar
compositions of the quartz syenites tO the volcanics and

their position in the outer portions of the complex.

154

Once the outer margins of the complex began to solidify,
silica and probably other components would continue to be
deposited. This would eXplain the presence of secondary
quartz in the syenites which form the outer perimeters of
the complex. Because solidification is likely to have
preceded gradually inward,_the outermost quartz syenites
would have received the largest amounts of secondary quartz.
This is the relationship Observed in the field and in thin
sections.

The alkaline syenites (slightly silica-oversaturated)
and the pulaskites (slightly silica-undersaturated) form
the intermediate compositions of the complex. They may
represent the initial stages of the thermogravitational
diffusion process. Although much of their volume may have
been partially crystalline, they may have played a part as
the medium for thermodiffusion and volatile-fluxing.

The central foyaites crystallized last and represent
the final residual—hydrothermal system, characterized by
the metasomatic textural features common to many plutonic
alkaline complexes (e.g. Kogarko et al., 1978). These
rocks represent the de-silicated, transport-depleted portion
of an originally silica-undersaturated melt. They may have
formed a hot fractionating column within the central
portions of the melt, yielding the silica and other
components necessary for the formation of the silicaeover-
saturated perimeters of the magma. It is likely that this

inner portion of the magma was involved in a complex

155

interaction Of many proceSses including crystallization,
re-melting, mixing, and vapor-fluxing. Similar to the
suggestions by Luth and Tuttle (1953) and the agpaitic
alkaline magmatic model prOposed by Kogarko, Burnham and
Shettle (1978), the discrete vapor phases are not lost
outside the magma system, but rather are slowly re-dissolved
back into the residual melt as temperatures decrease. The
final stages of salic magmatism would then involve a
continuous transformation from a magmatic to a hydrothermal
system. The gradual deve10pment of these fluid effects

and their genetic relationship to the magmatic system, may
allow these alterations to occur under quasi-equilibrium
conditions (e.g. albitization, sericitization, cancriniti-
zation, feldspar-clouding, and secondary quartz). This may
help to eXplain the lack of severe scatter in the chemical
data in many alkaline complexes in spite of the extent of
the alterations. The final stages of salic alkaline melts
may be characterized by a regular association of residual-
hydrothermal phases and low-temperature alterations. Such
an approach possibly Offers some insight into the abundance
of the "secondary alterations" which have been noted to
occur in an almost identical pattern in many alkaline
complexes throughOut the world. Given similar initial
compositions and an uninterrupted differentiation sequence,
these fluid affeCts may be nearly as regular as the crystal-

liquid equilibria of crystal fractionation.

156

Roofward migration Of certain chemical species does
not require a discrete vapor phase (Kennedy, 1955; Shaw,
1974). Although the major elements may not have been
significantly affected, volatile, isotOpic, minor and trace
element compositional gradients may have develOped early
in the history of the alkaline melt at Abu Khruq. As the
melt cooled, volatile complexing and the strong thermal
and density gradients may have begun to exert an increasing
influence on the differentiation of the magma. Although it
is uncertain as to the timing and extent of vapor phase
separation, it is possible that halogen-rich and COz-rich
aqueous fluids began to flux through the magma and greatly
enhance the deve10pment of a liquid zoned magma.

The polymerization Of a melt is perhaps one of the
most important factors in thermodiffusion and volatile
transfer. Thus, an important component in liquid zoned
systems is thought to be the halogens because they are
highly soluble in peralkaline liquids, cause de-polymeriza-
tion of the melt, and allow greater mobility of the cations
(Mustart, 1972; Baker et al., 1977; Hildreth, 1979, 1981).
It is clear that chlorine was an important component in
the magma at Abu Khruq (crystallization of sodalite; Cl
analyses by El Ramly et al., 1969a). Based on experimental
and natural studies, it is likely that Cl plays an important
role in the thermogravitational diffusion mechanism
(Hildreth, 1979, 1981; Mustart,_1972). As temperatures

declined in the sodic alkaline melt at Abu Khruq, there

157

would have been a tendency to exsolve a NaCl-rich fluid
and/or eventually crystallize sodalite (see Koster van
Groos and Wyllie, 1969; Roedder and Coombs, 1967). Experi-
mental data indicates that Si, K, Na, the REE, and other
components may become mobile in the presence of chlorine
and other commonly associated volatile species such as the
hydroxides (MacKenzie, 1960; Orville, 1963, 1972; Anderson
and Burnham, 1967; Flynn and Burnham, 1978). Orville (1972)
demonstrated that silica and potassium may be more mobile
than sodium in the presence of a Cl—rich phase. Chlorine
is also perhaps one of the most efficient REE-complexing
agents (Flynn and Burnham,_l978). Similar to other
REE-complexing agents, Cl complexes the REE with increasing
efficiency in the heavier elements of the series due to

the lanthinide contraction (Flynn and Burnham, 1978). Due
to the high concentration of C1 in the volatile flux, and
the complexing characteristics Of Cl as predicted‘ by
experimental work, the REE, silica, potassium in preference
to sodium, and other components, were preferentially
carried toward the roof and outer portions Of the magma.
This resulted in the REE-rich silica-oversaturated syenites
and volcanics in the outer and erupted portions of the
magma. Due to chlorine's preference for the heavier REE,
the HREE were more abundant in the melt than normally
expected in silicate systems. This is reflected in the
unusually small relative bulk partition coefficient of Lu

(refer to Table‘4).

158

As crystallization preceded inward, only a residual
melt and the fluxing fluid phases remained as the final
stages Of the magma. Similar to the alkaline magmatic
model prOposed by Kogarko, Burnham and Shettle (1978), it
is suggested that these REE-depleted fluid phases gradually
re-dissolved back into this residual melt. This final
melt-hydrothermal system is represented by the foyaites,
which are strongly depleted in Si, K, the REE, Sr, and Ba
(around zero ppm), and relatively enriched in other compon-
ents such as Al, Na, and Th. The Rayleigh fractionation
scheme performed in the previous section for the silica-
undersaturated and oversaturated series, respectively, may
provide some additional insight into the reciprocal
enriched versus depleted pattern between the outer silica-
oversaturated rocks, and the inner silica-undersaturated
rocks. Assuming an exponential relationship in the distri—
bution of the REE and Th, it appears that the degree of
depletion in these final foyaitic rocks is roughly equal to
the degree of enrichment in the outer quartz syenites.

This is interpreted to support the transport of REE from
the central portions of the melt, toward the outer portions
of the magma. Unlike the REE, thorium is interpreted to
have been less affected by the transport mechanism--perhaps
due to its greater incompatibility. Even so, its slight
relative enrichment in the final crystallizing stage of the
foyaites, is reciprocal to the slight relative depletion in

the quartz syenites.

159

In spite of the overall REE depletion in the foyaites,
the incorporation of Cl into the crystal structure of
sodalite caused the remaining REE to be strongly incorpor-
ated into this final crystallizing stage because of its
unusual REE-complexing characteristics. This is reflected
in the high relative bulk partition coefficients for the
REE as calculated for only the silica-undersaturated series
(refer to Table 5). In addition to the relative depletion
of the REE in the foyaites, the presence of large amounts
of chlorine caused an increase in the relative abundance Of
the HREE compared to the LREE due to Cl's greater prefers
ence of the heavier elements of this series. This resulted
in unusually low Ce/Lu ratios in the foyaites, and their
"flattened" normalized REE patterns (see Figure 11). Thus,
the inferred importance of C1 in these sodic alkaline
melts is interpreted to be the cause of the decrease in the
Ce/Lu ratios between the pulaskites and the foyaites, see
Figure 43, (rather than an increase as would be expected
in a case of crystal fractionation).

Although the roofward enrichment of silica and the
LREE at Abu Khruq is similar to the pattern noted by
Hildreth (1977, 1979, 1981) in his study of the BishOp Tuff,
many of the other elements display Opposite enrichment
directions--including Eu, the HREE, Th, and Na/K. Further
study of these types of sodic alkaline systems--as Oppossed
to the silica-rich potassic rhyolite systems such as the

BishOp Tuff-fmay reveal the reasons for such differences.

160

"alaseiié 6 gfibflgfldnQfi.nfimmQHUEdd&_Qfiucflflfibo.fifid§fixugflanH$E<Mwflfiau

«LHMMHHHUHHHH H ugaaamfifllqwmQREUQoHQRHEOmfiumflOz.Aoflhfi:"mamhaflqmcm
366. 6.62 + . 1889 magmas mo 9:3 5 was. an... .o um 8%qu .83..
oEHHHMHHH m mfiumommfifihgmhhd”wtafififiafiwmgmmBMH§.QEfiU§HMHHfiumam

.3 8ng

czar: 3mm I :LL mu czar: arm n tum mu

 

 

 

8.83 8.11 8.6!. 8.3. 8.8 836 8.w 8...... 8.3a 8.3. 8.3. 8.6.. 8.9. 8.?
w m
w m
u o
\ \x. .. \ \x ..
\.\ N.J \.\.\ M
\ 0‘
6\ e mmu 66
«I
m d \ w
_d \
Mn
n l\ h
I
fi m
.w M
.m

 

Ndd N9/33

161

Differences in the thermal and density structures Of melts,
differences in the degree and nature of polymerization, the
available volatile complexing, Soret coefficients, and
many other factors may cause distinct gradients within
different chemical systems. Models involving liquid

zoning must take into account not only these differences in
liquid structure and composition, but also the time factor
involved in plutonic versus volcanic associations. The
re-dissolution of the volatile-flux phases in the plutonic
case, for example, may result in more complicated differen-
tiation sequences and chemical patterns. It is also
important to point out that many events may occur in the
history of any specific alkaline magma so that unique
geometries and compositions may be produced. This may
involve structural changes in the surrounding crust, addi-
tional magmatic activity, or individual pulses of magma
from different volumes in the zoned magma. Alkaline
complexes with similar rock-types, but different apparent
orders of intrusion/crystallization may be explained by
any combination of pulses Of magma from different volumes
or levels of the liquid zoned magma.

The evolution of alkaline melts via fractional crystal-
lization, thermogravitational diffusion, and the gradual
transformation of the residual volatile-rich melt toward a
hydrothermal system, may represent a regular sequence of
processes involving crystal-liquid equilibria as well as

crystal-liquid-vapor equilibria, all set within the thermal

162

and density structure of the magmatic system. Thus, given
similar initial alkaline-volatile "packages" from the
mantle, and a stable magma chamber, these alkaline melts
may follow a fairly consistent course of differentiation so

that similar alkaline complexes are frequently generated.

PART II

THE TECTONIC SETTING OF THE EGYPTIAN ALKALINE COMPLEXES

Introduction

 

The second objective of this study is to investigate
the tectonic setting of the alkaline magmatism in Egypt.
In spite of the similar chemical patterns of many primitive
alkaline suites throughout the world, there is no well
defined, unified model on the tectonic processes associated
with alkaline magmatism. Thus, in order to approach this
problem, the tectonic settings of other alkaline provinces
are used as comparisons, the general tectonic deve10pment
of Afro-Arabia is discussed, and common features are com-
pared tO the Egyptian alkaline province. A tectonic
process is prOposed for alkaline magmatism, and evaluated
in terms of the common features noted for alkaline magmatism
in general, and more specifically, for the timing-and

relationship of events in the Egyptian alkaline province.

163

CHAPTER 5

 

ALKALINE MAGMATISM AND PLATE TECTONICS

Introduction

 

The tectonic setting of alkaline magmatism has long
been a major problem in petrology (Harker, 1896, 1909).
Alkaline rock-types have been noted to occur in areas of
relatively little crustal activity and low heat flow
(Bowen, 1937; Tilley, 1958). This type of setting has
been classically referred to as "anorogenic". This term
implies no fundamental crustal process, and describes the
state Of most of earth's surface throughout much Of its
history. In spite of this, the term "anorogenic" Often is
used to describe the tectonic setting of alkaline magmatism
(e.g. MacIntyre, 1977).

With the deve10pment of the theory of plate tectonics,
a substantial amount of work has been devoted to character-
izing magmas according to their tectonic setting (e.g. Barth,
1962; Martin and Piwinskii, 1972; Sugisaki, 1976; Pearce
et al., 1977; Petro, VOgel and Wilband, 1979). This type
of approach led to generalized conclusions in which calc—
alkaline magmas are most frequently produced at compressional
plate boundaries, and alkaline magmas are most frequently
produced at extensional plate boundaries. This may appear
to be a better description for the tectonic setting of
alkaline magmatism. Many models on the origin Of alkaline

magmatism have been based on the rift/dome setting,_and

164

165

this has developed to such an extent that some alkaline
provinces with no apparent rift association are sometimes
presumed to be simply "pre-rift" or "failed rift" areas
(e.g. Dostal et al., 1979; Currie, 1970; Upton, 1974;

Upton and Blundell, 1978; Kisvarsanyi, 1980; Darbyshire and
Fletcher, 1979; Ressetar et al., 1981; Strong and Minatidis,
1975; Sykes, 1978; Kumparapeli, 1970; Falvey, 1974). How-
ever, such models On alkaline magmatism ignore a large
number of alkaline assemblages formed in tectonic settings
which cannot be described as rifts.

This strong correlation Often found in the literature
between alkaline magmatism and rifting would seem especially
applicable to the Egyptian alkaline province in view of
its proximity to the Red Sea, Figure 44. However, age-
dating of the Egyptian alkaline complexes demonstrates that
this similar type of alkaline magmatism occurred sporadi-
cally over a period of more than 450 m.y., from 554 to 89
m.y. B.P. (Serencsits et al., 1981; Lutz,_l979; Hashad and
El Reedy, 1979), see Figure 44. If one ascribes to the
Red Sea rift/dome model for the generation of the younger
complexes (see Ressetar et al., 1981), it is not unreasonable
to expect similar tectonic events for the origin Of the
older complexes. However, there is presently little
eVidence for rifting in the immediate Red Sea area before
the doming and strike—slip deformation of the Late
Cretaceous, and the well—deVelOped extensional activity of

the Eocene (Garson and Krs,_l976; Youssef, 1968; Swartz and

166

Figure 44.

Map of the major alkaline complexes in the Eastern
Desert Of Egypt, and the general tectonic style of
Afro-Arabia.

. Wadi Dib, 554 m.y. B.P.

. Zargat Naam, 404 m.y. B.P.

. Tarbtie (North), 351 m.y. B.P.

E1 Gezira, 229 m.y. B.P.

E1 Naga, 145 m.y. B.P.

El Mishbeh, 142 m.y. B.P.

. Nigrub El Tahtani, 140 m.y. B.P.

. Nigrub El Fogani, 139 m.y. B.P.

WQGUwaT-J

9. Mansouri, 132 m.y. B.P.
10. El Kahfa, 91 m.y. B.P.
ll. Abu Khruq, 89 m.y. B.P.

Dates from Serencsits et a1. (1981) and Hashad
and El Reedy (1979). Map and tectonic lineaments
after Carson and Krs (1976).

167

 

36°

«‘9 Noid \‘

4_ UBIAN
- 9 SHIELD

Major Alkaline Complex
,Eastern Desert, Egypt
8.

.General Tectonic Style
of Afro-Arabia

b
k

 

J l

\ ronscurrenl‘ \\_
Fault System‘
"?

 

/

I j

38" 4T
6 Alkaline Complexes
“s Shear Zones

\ Fracturesllineaments‘
(Et.€_" )
263

q

ARABIAN
SHIELD

33 26:
' /

'3.

RED
SEA

 

 

Figure 44.

168

Arden, 1960; Burek, 1969; Abdel-Gawad, 1969; Hancock and
Kadhi, 1978; Brown, 1970; Coleman, 1974; Pilger and Rosler,
1976; Le Pichon and Francheteau, 1978). It is unlikely
that each of these episodes of alkaline magmatism in Egypt
occurred with rifts similar to the early Red Sea activity
throughout most of the Phanerozoic eon.

It is prOposed here that these multiple episodes of
similar alkaline magmatism coincide with periods of tectonic
reactivation during major changes in plate motion and
inferred periods of thermal disequilibrium between the
lithosphere and the mesosphere. These changes in plate
motion have involved not only rifting/extension, but com-

pression, and large-scale strike-slip motion.

1) Characteristics of Alkaline Magmatism

The Egyptian and many other alkaline provinces are
similar in chemistry, general structural setting, and
long-term reoccurrence of alkaline magmatism (see Part I).
Thus, support for the prOposed tectonic setting of the
Egyptian alkaline province consists of comparisons to other
alkaline occurrences as well as several approaches which
corroborate the coincidence of the Egyptian episodes of
alkaline magmatism with periods of regional structural
re-adjustment and major shifts in plate motion.

Based on a review of the Egyptian and numerous other
alkaline provinces, several features are characteristic Of

many occurrences of alkaline magmatism:

169

1. Similar alkaline magmatism occurs in oceanic and
continental crust.

2. Similar alkaline magmatism occurs in regions dominated
by either extension or compression (Schwarzer and Rogers,
1974; Frey et al., 1978; McBirney and Aoki, 1968; Miyashiro,
1978; Baker et al.,,l977).

3. Alkaline magmas tend to occur away from the most
thermally and mechanically active areas (Bowen, 1937;
Tilley, 1958; Earth, 1962; Sorensen, 1974; Miyashiro,
1978).

4. This alkaline magmatism tends to occur along major
crustal fractures, Often with regional strike-slip deforma-
tion (Sorensen, 1974a; Sykes, 1978; Garson and Krs, 1976;
De Long et al., 1975; Hawkins and Natland, 1975).

5. Episodes Of alkaline magmatism can often be shown to
coincide with major shifts in plate motion (Boyd, 1973;
Marsh, 1973; Opdyke and MacDonald, 1973; van Breemen and
Bowden, 1973; Foland and Paul, 1977; Herz,_l977; Sykes,
1978), or they may occur along zones Of sustained strike-
slip motion (see Hawkins and Natland, 1975; De Long et al.,
1975; Arculus et al., 1977).

6. Alkaline magmatism Often reoccurs over very long
periods of time along the same set of major fractures in
areas which are not disturbed by high thermal regimes
(Zartman, 1977; Foland and Paul, 1977; Foland and Loiselle,
1981; Sykes, 1978; van Breemen et al., 1975; Currie,_l970;

Doig and Barton, 1968; Kumarapeli and Saull, 1966;

170

S¢rensen, 1974;, ButakOva, 1974; Blaxland et al., 1978;

Lutz, 1979; Serencsits et al., 1981).

Crustal Thermal Regime and Structural Setting

Alkaline, and especially highly silica-undersaturated
alkaline magmas, tend to occur away from the thermal high
of active, tholeiitic—dominated mid-oceanic ridges and
the normal convergent zones of island arcs (Miyashiro,
1978; DeLong et al., 1975). This same feature has been
noted in analagous continental rift areas in which alkaline
magmatism tends to occur on the flanks of the rift structure,
away from the most thermally active areas (Wright, 1965,
1970; Le Bas, 1971; Baker et al., 1972) and in the foreland
of convergent zones (van Breemen et al., 1979; Butakova,
1974). Schwarzer and Rogers (1974) stressed that the
worldwide similarity of the major element chemistry of
alkaline stuies in different crustal settings reflects a
mantle process deep enough in the mantle to be unaffected
by the chemical and thermal differences between mantle
regions under the various tectonic settings.

Alkaline magmatism has been frequently noted to be
associated with major crustal fractures and lineaments
(e.g. S¢rensen, 1974, p. 145; Sykes, 1978; DeLong et al.,
1975; Hawkins and Natland, 1975; Garson and Krs, 1976).
Although the exact nature of some of these lineaments is
obscure, many prove to be major fault systems, frequently

with significant strike-slip components (see Sykes, 1978;

171

DeLong et al., 1975). Some lineaments may represent crustal
fractures, possibly annealed after several episodes of
movement. Based on deep-sea dredgings and petrogenetic
work (Thompson and Melson, 1972; Stebbins and Thompson,
1978), oceanic alkaline magmatism may be the result of
leaky transform faults away from the active mid-oceanic
thermal highs. These faults may provide access to a deeper
source region in the mantle than the more voluminous
tholeiitic variety of magmatism (Shibata, Thompson and
Frey, 1979). There are numerous examples of alkaline
assemblages associated with continental lineaments, and
Often these are on trend with transform faults in the
adjacent oceanic plates (Foland and Faul, 1977; Ballard and
Uchupi, 1972; Fletcher, Sbar and Sykes, 1974; Sykes, 1978;
MacFarlane and Ridley, 1949; Marsh, 1973; Dingle and
Scrutton, 1974; Rabinowitz, 1976; Cornelissen and Verwoerd,
1975; Fodor and Husler, 1977; Herz, 1977; Fitton and Hughes,
1977; Thompson and Melson, 1972; Stebbin and Thompson,

1978; DunlOp and Fitton, 1979; Booth et al., 1978;

Shibata, Thompson and Frey, 1979). Even in areas which do
not display obvious crustal Offset, major episodes of

of alkaline magmatism have been explained in terms of
fracture mechanics (e.g. Hawaiian Seamount chain; Jackson,
Shaw and Bargar, 1975; Turcotte and Oxburgh, 1978; Shaw,
1980; Shaw et al., 1980; e.g. Easter—Sala y Gomez and Juan
Fernandez island-seamount chains; Pilger and Handschumacher,

1981).

172

2) The Diverse Tectonic Settings Of Alkaline Magmatism

In order to formulate a better concept of the possible
tectonic conditions under which alkaline magmatism may
occur, a review was made of many alkaline occurrences.

Some examples of this review are briefly described and
referred to in the following pages with the intent of
isolating consistent tectonic aspects associated with the
alkaline magmatism in these different settings.

The Atlantic alkaline provinces lend some good examples
of alkaline magmatism before, during, and after major
periods of continental rifting. The White Mountain Magma
Series (Foland and Paul, 1977; Foland and Loiselle, 1981),
the Monteregian Hills (Kumparapeli, 1970; Foland and Paul,
1977), the NigeruNigerian alkaline province (van Breeman
and Bowden, 1973; van Breeman et al., 1975), the Angolan
and South Afraican alkaline provinces (Marsh, 1973; Neill,
1973; Sykes, 1978), the Brazilian alkaline occurrences
(Amaral et al., 1967; Marsh, 1973; Neill, 1973; Opdyke and
MacDonald, 1973; Herz, 1977; Sykes, 1978), and the Bolivian
Velasco alkaline province (Darbyshire and Fletcher, 1979)
are all clear examples of continental alkaline provinces
which contain complexes emplaced during the major openings
of the Atlantic Ocean. Many of the above workers have
noted that these alkaline complexes tend to lie along
major lineaments thOught to have been activated during
rifting. The major rifting activity and accompanying

alkaline magmatism in the South Atlantic occurred around

173

140 m.y. B.P.; in the central Atlantic similar rifting and
alkaline magmatism occurred earlier, around 180 m.y. B.P.
(e.g. Sykes, 1978).‘

Initially it was often assumed that all the alkaline
complexes in these Atlantic alkaline provinces evolved in
the early rifting activity (see van Breemen and Bowden,
1973). The on—going work in the age-dating of alkaline
provinces is revealing that isolated portions of especially
the Older continental areas are commonly the sites of
long-term reoccurrences of alkaline magmatism. Similar to
the long age span of the Egyptian alkaline province, the
ages of the alkaline complexes in many provinces Span
several hundred million years (see Zartman, 1977; Blaxland
et al., 1978; Butakova, 1974; Currie, 1970; Doig and Barton,
1968; Kumarapeli and Saull, 1966; Sykes, 1978; Herz, 1977;
Marsh, 1973; Neill, 1973; Opdyke and MacDonald, 1973;
Amaral et al., 1967; van Breemen and Bowden, 1973; van
Breeman et al.,_l975; Foland and Faulr 1977; Foland and
Loiselle, 1981). This means that the alkaline magmatism
in some Of these provinces occurred in apparent episodes
before, during, and after rifting (see especially Foland
and Paul, 1977; van Breemen et al., 1975; Sykes, 1978).

To date, the episodes of alkaline magmatism preceding
continental rifting in theSe Atlantic alkaline provinces
have not been adequately explained in terms of tectonic
events. The two major episodes of alkaline magmatism which

occurred after the continental rifting of the South and

174

Central Atlantic ocean basins, respectively, have been
correlated with major shifts in the spreading of the
Atlantic. Outbreak of alkaline magmatism in the South
Atlantic alkaline provinces appears to coincide with a
sudden shift in spreading of the South Atlantic around
80-90 m.y. B.P. during the Late Cretaceous (Marsh, 1973;
Opdyke and MacDonald, 1973; Herz, 1977; Sykes, 1978).
Similarly, an episode of alkaline magmatism in the pro-
vinces surrounding the Central Atlantic appear to coincide
with a shift in the spreading of the Central Atlantic
around 120 m.y. B.P. (e.g. Poland and Paul, 1977; Sykes,
1978).

Besides the separation in time from rifting, it is
also important to note the emplacement Of alkaline magmas
along major lineaments far inland from the Atlantic contin—
ental margins during these periods of rifting and shifts in
spreading. The recently discovered Velasco alkaline pro-
vince in Bolivia intruded major lineaments in the Amazon
craton at the same time the initial rifting Of the South
Atlantic was taking place 1600 kilometers away (circum 140
m.y. B.P,; Darbyshire and Fletcher, 1979). The Monteregian
Hills alkaline compleXes were emplaced during the prOposed
shift in the spreading of the Central Atlantic, well inland
from the then mature rifted continental margins (Foland and
Paul, 1977).

There are numerous examples of alkaline magmatism

occurring within regional compreSsional tectonic regimes

175

(Schwarzer and Rogers, 1974; DeLong et al., 1975; Hawkins
and Natland, 1975; Barth, 1962; Payne and Shaw, 1967;
Appleyard, 1974; Barker, 1969, 1974; Sorensen, 1974, p. 146;
Bowden, 1974; Marsh and Carmichael, 1974; Arculus, 1976;
Arculus et al., 1977; Smith et al., 1977; Johnson et al.,
1976; Bevier et al., 1979; Robin and Tournon, 1978; Miller,
1977; Miyashiro, 1978; Heming, 1979; Lutz, 1979; van Breemen
et al., 1979).

The Samoan island chain represents one of the largest
single occurrences of alkaline magmatism in the world.
Silica-undersaturated alkaline volcanics erupted simul-
taneously along a 500 kilometer lineament which Hawkins and
Natland (1975) interpret to be a fracture propagated by the
subduction of the sharply flexed Pacific plate descending
into the Tonga trench.

The island of Grenada, located at the southern end
of the Lesser Antilles island arc, contains a highly silica-
undersaturated alkaline suite (Arculus, 1976; Sigurdsson
et al., 1973). There is seismic (Molnar and Sykes, 1969)
and structural evidence (e.g. Alberding, 1957; Malfait and
Dinkelman, 1972) indicating that this area is dominated by
large-scale strike-slip faulting between the Caribbean
and South American plates. DeLong et a1. (1975) concluded
that such strike-slip tectonics are genetically related
to the presence of the alkaline magmatism rather than the
subduction proceSs itself. They made similar interpreta-

tions for eight other alkaline occurrences in oceanic

176

subduction zones. They concluded that these alkaline
suites occurred either 1) "along or near the edge Of a
plate where hinge-faulting is suggested...or 2) when a
fracture or other linear feature roughly perpendicular to
the arc is being subducted" (DeLong et al., 1975, p. 731;
see also Marsh and Carmichael, 1974).

Appleyard (1974) and Payne and Shaw (1967) have
reviewed a "syn-tectonic" alkaline suite which appears to
have been emplaced along strike—slip faults during the
Grenville orogeny in Ontario, Canada. Paleomagnetic
interpretation supports this style of strike-slip deforma-
tion with large-scale dextral shear between Gondwana and
Laurentia (Morel and Irving, 1978).

There are some important alkaline suites which are
associated with the Caledonian compressional tectonics.
In the British Isles two major centers of alkaline magma-
tism are Loch Borrolan and Loch Ailsh. It has been con-
cluded by several workers that they were emplaced in an
episode of thrusting associated with the closing Of the
Iapetus Ocean (most recently by van Breemen et al., 1979).
A similar tectonic setting may have existed for the
Caledonian alkaline assemblages of northern Norway (Sturt
and Ramsay, 1965; Appleyard,_l974),and Newfoundland
(O'Driscoll and Strong, 1979). Each of these Caledonian
alkaline magmatic centers occurred in regions that were
dominated by Oblique convergence and large-scale strikes

slip deformation during the Caledonian compressional event

177

(Phillips et al., 1976; Mitchell and McKerrow, 1975).
Northern Norway, in particular, represents a transform
boundary which, according to recent paleomagnetic evidence,
was the site of extremely large amounts Of strike-slip
movement during both the Opening and closing of the Iapetus
Ocean (Kjode et al., 1978).

One of the largest and best-studied alkaline provinces
in the world is the Gardar alkaline province in southern
Greenland. This area was dominated by Proterozoic strike-
slip tectonics (Bak et al., 1975; Grocott, 1979; Watterson,
1978; Upton, 1974; Stephenson, 1976). The dates of the
alkaline complexes correlate with or slightly preceding
the Grenville orogeny (Blaxland et al., 1978; van Breemen
and Upton, 1972). Upton (1974) and others have suggested
that the alkaline magmatism is actually associated with a
Proterozoic rift system which was active during or slightly
before the Grenville Orogeny (see Upton and Blundell, 1978).
van Breemen and Upton (1972) simply state that the alkaline
magmatism occurred in the foreland of the Grenville
orogeny. The work Of Bak et al. (l975),_Watterson (1978)
and Grocott (1979) on the shear deformation of Greenland
appears to favor compressional tectonics during the same
time as the alkaline magmatism.

Compressional forces may be more effective in fractur-
ing the crust than extensional ones. Oblique compressional
forces may be highly effective in forming or reactivating

major zones of weakness in the crust (e.g. Benioff, 1962;

178

Fitch, 1972; Molnar and Tapponier, 1975; Mitchell and
McKerrow, 1975; Holcombe, 1977; Tapponnier and Molnar, 1976,
1979; Watterson,_l978). Such fault systems may have great
importance in the initial formation Of intraplate disloca-
tions, crustal response to latter deformation, (see
Wetterson, 1978) and localization of alkaline magmatism.

A major point in this study is the Observation that
the alkaline magmatic process is versatile with respect to
many specific types of crustal activity. The reoccurring
nature of alkaline magmatism in a given region does not
support a model for the origin of these melts which is
dependent on any single type of tectonic event such as
rifting or subduction. Similar types of alkaline magmatism
occur in a wide variety of crustal settings, including
oceanic and continental crust undergoing extension or
compression. The sum of the conditions which may be
observed in the diverse crustal settings of alkaline magma-
tims, and the long-term reoccurrence Of alkaline magmatism
in old cratonic areas, may provide some constraints on
a model for the origin of many alkaline occurrences. It is
possible that alkaline magmatism repreSents a fairly regular
.geodynamic process which is not constrained to any single

type of tectonic setting.

CHAPTER 6

 

MODELS FOR THE PRODUCTION OF ALKALINE MELTS

There are fundamentally four models on the initial
production Of mantle melts which may be applicable to the
_generation of primitive alkaline melts:

l) Localized mantle plumes,_
2) Global mantle disturbances,
3) Lithospheric flexure,

4) Shear heating.

The first two approaches suggest that the mantle is the
active system which creates melts and causes major struc-
tural changes in the overlying lithosphere via thermal
or chemical disturbances (e.g. Morgan, 1971, 1979; Wilson,
1963; Anderson, 1975; Vogt, 1972, 1979, 1981; Burke and
Dewey, 1973; Crough, 1978; Smith, 1979; Thiessen et al.,
1979). The last two approaches suggest that the litho-
sphere plays a major active role in the creation of mantle
melts rather than being the passive system (e.g. Schubert
et al., 1976; Schubert and Yuen, 1978; Froidevaux and
Souriau, 1978; Bailey, 1977; Sykes, 1978; de Gruyter and
VOgel, 1981; Feigenson and Spera, 1980, 1981; Shaw, 1973,

1980; Shaw et al., 1980).

l) Mantle Plumes and Global Mantle Disturbances
The first two approaches are closely related and are

the source of current debate (Gass et al., 1978; Vogt, 1981;

179

180

Pollack et al., 1981). One approach proposes that out—
breaks Of magmatism are the result of the localization of
individual mantle plumes, plate thickness and plate velocity
(Gass et al., 1978; Pollack et al., 1981). The second
approach calls upon similar mantle plumes, but on a simul-
taneous, global scale (Vogt, 1972, 1979, 1981).

Although it is often difficult to prove or disprove
the existence of mantle plumes in the geologic history of
many areas, there are two features of the Egyptian and
other alkaline provinces which make it difficult to impose
any of the current plume models for the production of these
alkaline melts. The ages of the Egyptian alkaline com-
plexes--similar to the wide age distributions of many
alkaline provinces--5pan a period of over 450 m.y. The
complexes do not display a regular geometric pattern with
respect to their ages, similar to those presented as evi-
dence for the presence Of mantle plumes in other areas. It
is unlikely that mantle plumes may remain fixed to the
lithosphere for the length of time over which alkaline
magmatism reoccurred. It is also unlikely that mantle
plumes may account for the irregular geometric patterns.

This type Of longnterm,.non-systematic spatial and/or
temporal distribution of alkaline magmatism has been used
as evidence against a mantle plume origin in many areas:
the White Mountain Magma Series, the Monteregian Hills
(Foland and Faul, 1977), several African and Brazilian

alkaline provinces (Marsh,_l973; Sykes, 1978), the Samoan

181

island chain (Hawkins and Natland, 1975), other alkaline
occurrences in subduction zones (DeLong et al., 1975), and
the Egyptian alkaline province by other authors (Serencsits
et al., 1981; Lutz, 1979). The simultaneous eruption of
alkaline volcanics along a 500 kilometer lineanment in the
Samoan island chain is a case in which any current plume
model may be rejected (Hawkins and Natland, 1975). The
synchronous emplacement of alkaline magmas in widely
separated areas is difficult to account for by the local-
ized mantle plume hypothesis. A good example of this is
the synchronous emplacement of alkaline magmas in several
areas along portions of the mature Atlantic margins, well
after the opening of the ocean basins.

It is important to point out that even the best
examples of the mantle plume model and linear age progres-
sion have also been eXplained in terms of fracture mechanics
(e.g. the Hawaiian seamount chain: Jackson, Shaw and Bargar,
1975; Shaw and Jackson, 1973; Turcotte and Oxburgh, 1978;
Shaw, 1980; Shaw et al., 1980).

Thus, if mantle plumes consistently account for
alkaline magmatism, such models must become more complex
in order to explain the spatial and temporal distribution
of many alkaline igneous centers in rift areas, subduction
zones, collisional fronts, continental margins of mature
oceans, and intraplate areas. The global synchronism Of
mantle plume convection, as suggested by Vogt (1972, 1979,

1981), may provide such an alternative.

182

This study suggests an approach other than a planetary "hot

spot" system.

2) Lithospheric Flexures

This approach also utilizes mantle upwellings, but
the cause Of the mantle disturbance lies in the motion of
the lithOSphere (e.g. Bailey, 1977). Bailey (1970, 1974)
suggested that crustal arching of a rigid anisotrOpic plate
by mechanical compression of surrounding plates, could
cause partial melting via decompression in the underlying
mantle. Volatile—fluxing would be an important result,
thus explaining the enriched alkali—incompatible trace
element composition of alkaline magmas being emplaced in
these areas. This is a rather novel idea in that rifting
would begin by compression. van Breemen, Aftalion and
Johnson (1979) suggest that this approach may explain the
presence of the Scottish compressional alkaline complexes
(e.g. Loch Borrolan). Although this model cannot be dis-
counted in certain areas, there are many alkaline suites
which are not associated with crustal arching nor signs of

immediately preceding compression.

3) Shear Heating

With the increasing number of advances in the under-
standing of creep prOperties of Earth materials and their
application to mantle and lithosphere rheology, it is
becoming evident that rather than the mantle always Operat-

ing as the active system, deformation caused by the

183

movement of the lithosphere is of great importance in the
production Of magmas (Shaw, 1963, 1965, 1969, 1973; Shaw

and Jackson, 1973; Jackson et al., 1972; Jackson and Shaw,
1975; Jackson et al., 1975; Shaw, 1980; Shaw et al., 1980;

Feigenson and Spera, 1981; see Phil. Trans. Roy. Soc. Lon.,

 

v. 288-A, pp. 383—646, 1978; Phys. Earth Plan. Int. Special

 

Issue, v. 17, no. 1, 1978). The current models on the
mantle emphasize the slow, solid-state convection of the
mesosphere (e.g. Weeterman, 1970; Weeterman and Weeterman,
1975; Elsasser et al., 1979), which is probably moving on
the order of mm/yr (Minster and Jordan, 1978). The litho-
sphere, On the other hand, moves at relatively rapid rates,
on the order of cm/yr., and often undergoes abrupt changes
in acceleration. It is unlikely that the mantle, with the
currently prOposed rheology, could provide the forces
necessary to consistently initiate these rapid changes.
Many Of the abrupt changes in the motion of the lithosphere
may be cuased by the geometry of the plates and the sub-
duction-collision mechanism. If this is the case, it is
the mantle which must adjust to these rapid changes in
lithospheric motion. This is accomplished within the
asthenosphere, a ductile layer (perhaps around 70 to 400
kilometers in depth, Weeterman, 1970; Stocker and Ashby,
1973) Of relatively low viscosity which forms a zone of
mechanical decoupling between the rigid lithosphere and

the mesosphere.

184

It has been noted for some time that the flowage of
material results in the dissipation of energy and the
involvement of shear heating (Zener, 1948; Kennedy, 1963).
There has been considerable theoretical work on shear
heating, and recently, it has been applied to large-scale
thermo-mechanical modelling of Earth's lithospheric and
mantle systems (e.g. Schubert and Yuen, 1978; Melosh, 1976,
Yuen and Schubert, l977,_l979; Anderson and Perkins, 1974;
Melosh and Ebel, 1979; Froidevaux and Souriau, 1978;
Feigenson and Spera, 1980, 1981). It is possible that
shear heating in the Earth as a result of changes in the
motion of the lithospheric plates may provide a heat
source necessary for the generation of mantle melts.

Indirect measurement of the rheology and shear stress
at the base Of the lithosphere shows that large litho-
spheric plates may have forces caused by basal shear stress
that are comparable to the forces acting at the major plate
boundaries (Melosh, 1977). This results in the concentra-
tion of large amounts Of shear in the asthenosphere and the
production of heat due to viscous dissipation. The regular
translation of the lithosphere may thus result in the
production of melts in the asthenosphere. However, based
on indirect viscosity measurements (e.g. Drury, 1978) and
the apparent balance between plate motion and the resistive
forces (JacOby, 1970), it is unlikely that this melt pro-
duction due to the constant translation Of the plates is

large enough to be a regular souce for crustal magmatism.

185

Studies on the thermal and mechanical considerations of

the lithosphere-mantle systems using different assumptions
and variables indicate that under a given rate of plate
motion and other conditions, the thermal balance between
the lithosphere and the mantle must move toward thermal
steady-state conditions in order to avoid an eventual
thermal runaway (Anderson and Perkins, 1974; Melosh,_l976;
Schubert and Yuen 1978; Yuen and Schubert, 1979; Melosh and
Ebel, 1979). However, plate accelerations cause an increase
of shear stress on the base of the lithospere and in the
asthenosphere, and may result in increased heat dissipation
and deep thermal anomalies (Froidevaux and Souriau, 1978;
Melosh and Ebel, 1979). This causes a decrease in the
viscosity of this decoupling zone which may lead to addi-
tional increases in shear stress and plate velocity so that
heat dissipation may again increase and effectively lead

to a thermally unstable state (thermal feedback: Grundfest,
1963; Shaw, 1969; Schubert and Yuen, 1978; Melosh, 1976;
Yuen and Schubert, 1977, 1979; Froidevaux and Souriau, 1978;
Melosh and Ebel, 1979). It is possible that under these
conditions, significant amounts of asthenospheric melts are
generated in order to aide in returning the lithosphere-
mantle system to new thermal steady-state conditions. These
partial melts may be produced in sufficient quantities to
become a significant form of crustal magmatism (e.g. Shaw,.
1969; Shaw and Jackson, 1973; Jackson and Shaw, 1975;

FroideVaux and Souriau, 1978; Shaw, 1980; Shaw, et al.,.

186

1980; Feigenson and Spera, 1981). Although it is possible
that melts arise from the lithosphere/asthenosphere inter-
face (see Shaw, l973),_the common lack of coincidence
between the low-velocity/seismic attenuated zones and

this interface (Froidevaux and Souriau, 1978) indicates
that melt production probably occurs at different levels
(Shankland, 1977) in a potentially unstable,_thermally
active "boundary zone" of the general lower lithosphere-
asthenosphere regions (Koide and Bhattacharji, 1977).

This boundary zone may be important in terms of providing
a fairly continuous mantle source region for the production
of magmas of different degrees Of alkalinity and silica—
depletion.

High pressure experimental work indicates that alkaline,
silica-poor melts may be generated under large amounts of
pressure, pressures similar to those expected for the
asthenosphere (Green and Ringwood, 1967; Kushiro, 1968;
Green, 1970). Trace element modeling Of mantle compositions
and alkaline rock-types indicates that alkaline melts are
the result of small degrees of partial melting Of mantle
material (Gast, 1968; Kay and Gast, 1973; Frey et al.,
1978). The presence of small amounts of melt in the astheno-
sphere has been suggested based on geOphysical evidence of
low-velocity and seismic attenuated zones (Nur, 1971).
Electrical conductivity modeling supports the presence of
small amounts of partial melt within the lower lithosphere

and upper mantle (Shankland, 1977; Drury, 1978). Laboratory

187

(Unger, 1967) and field (Boudier and Nicolas, 1972;
Padovani and Carter, 1977) evidence suggests that this
partial melt exists as well-interconnected, low melt
fractions.

The recent thermo~mechanical modelling which evaluates
the theoretical aspects of shear heating in the astheno-
sphere and lower lithosphere, combined with the above
geOphysical and petrologic interpretations, appears to
provide a reasonable basis for a model on the production of
primitive alkaline magmas. Unlike the theoretical problems
encountered in accounting for the large thermal anomalies
necessary for the generation of silica-rich, non-alkaline
melts via shear heating (e.g. Yuen and Schubert, 1979;
Froidevaux and Souriau, 1978), alkaline melts require only
relatively small thermal anomalies as effective heat sources.
The shear heating model is consistent with all the common
characteristics noted for primitive alkaline magmatism
(p.168). Such a model would concur with the correlation
between alkaline magmatism and major shifts in plate
motion (e.g. Marsh, 1973; Opdyke and MacDonald, 1973, van
Breemen and Bowden, 1972; Foland and Paul, 1977; Herz, 1977;
Sykes, 1977; Hawkins and Natland, 1975; DeLong et al., 1975;
Arculus et al., 1977; Shaw, 1980; Shaw et al., 1980;

McHone, 1981). The common association between alkaline
magmatism and major crustal fractures may be a result of
the accompanying crustal failure under such plate motion,

thus, providing a more feasible mechanism for the ascent

188

of the small degrees of asthenospheric melts. The viscous
dissipation and the resulting generation of alkaline melts
in the asthenosphere, as well as the initiation or reacti-
vation of the overlying crustal fractures, may be responses
to regional extension, compression, or transform plate
movement. Such an approach provides a versatile process
for the generation of alkaline magmas which requires only
shifts in plate motion rather than specific types of tec-
tonic settings. This may explain the common long-term
reoccurrences of alkaline magmatism in many alkaline pro-
vinces. Intraplate fractures with dominant strike-slip
motion maintain the relatively low thermal regime estab-
lished by the underlying deep/low melt—fraction source
region. The lower incidence of alkaline magmatism in areas
containing high levels of mechanical and thermal activity--
such as is common near subduction zones—~may be a result of
the dominant melt source region being too shallow, and
generating degrees of partial melting which are too large
to produce an alkaline character. In areas of gradually
changing or fluctuating thermal/lithospheric activity, the
corresponding changes in the depth and degree of melting of
the shear zones in the melt source regions may produce the
ranges of alkalinity and silica-depletion Observed in alka-
line provinces. These zones Of shearing in the mantle may
provide a suitable locus for the metasomatic fluids and

enriched mantle material which appear necessary for the

189

generation of primitive alkaline melts (e.g. Frey et al.,
1978; see Chapter 2).

Factors which are likely to have a bearing on the
incidence of alkaline magmatism are the degree of thermal
disequilibrium (magnitude of the shift in the motion of
the plates and localization of possible concentrated shear
dissipation), the ease of ascent of the alkaline melts
(presence/activation Of major fractures), and the involve-
ment of shallow, larger degrees of non-alkaline melts which
would effectively Obscure the alkaline character of the
deeper mantle melts (i.e. mixing). Ascent of the astheno-
Spheric melts in the mantle is not addressed by this study,
but is discussed by some workers (e.g. Turcotte and Ahern,
1978; Koide and Bhattacharji, 1977; Anderson, 1981).

Perhaps one of the most important aspects of alkaline
magmatism which can be explained by shear heating model
is the tendency for alkaline magmatism to occur simul-
taneously in widely separated areas of the world. Although
more detailed work is needed on the age dating Of alkaline
complexes and their tectonic significance, there are a
large number of alkaline occurrences which tend to cluster
around similar dates (e.g. see Sykes, 1978; Darbyshire and
Fletcher, 1979; Foland and Faul, 1977; Marsh, 1973).
Commonly, these clusters of dates correspond to alkaline
magmatism which occur in areas affected by the same episode
of plate acceleration, but which may be separated by very

large distances (see p. 174). The Egyptian and Atlantic

190

alkaline provinces are conSidered to be examples of the
simultaneous incidence of alkaline magmatism in widely

separated areas which were affected by the same shifts in

plate motion.

CHAPTER 7

 

A TECTONIC MODEL FOR THE EGYPTIAN ALKALINE PROVINCE

l) The Tectonic DevelOpment of the Eastern Desert and
Adjacent Areas

The African continent ingeneral is composed of three
major stable, Old, and largely thermally inactive cratons
surrounded by younger, more active crust which is thought
to have been formed in the Pan-African tectono-thermal
event, Figure 45 (see Kennedy, 1964; Clifford, 1970; Gass,
1970). The exact meaning Of the "event" is rather Obscure,
occurring throughout portions of Africa from about 1100 to
500 m.y. B.P. (Gass, 1977). It was characterized by
sequences of small amounts of folding, sedimentation,
greenschist metamorphism, and a predominance of strike-slip
faulting. Hurley (1972) perhaps made the initial suggestion
that Africa may represent a landmass similar to one that
might result if the Pacific Ocean were to close completely
and leave a few Old cratons surrounded by compressed island
arc material and occassional oceanic crust. Detailed
descriptions of the geologic deve10pment of Afro-Arabia
can be reviewed in the following articles: Kennedy (1964),
Clifford (1070), E1 Shazly (1977),_Youssef (1968), Gass
(1977), Engel et al.,(l979, 1980), Greenwood et al. (1976),
Fleck et a1. (1976), Stern (1979), Dixon (1979, 1981),
Greenberg (1978), Rogers et al.(1978); Marzouki and Fyfe
(1977), Neary et a1. (1976), Naseef and Gass (1977), Gass

191

192

Frmnxa45.

Map Of the general tectonic configuration of Africa:
Three Precambrian cratons surrounded by younger,
largely Pan-African crust; stippled pattern repre-
sents Cenozoic magmatism diagonal lines indicate
Hercynian deformation; "X" marks location of the
Egyptian alkaline province within Pan-African

crust (after Thorpe and Smith, 1974).

193

 

 

 

“\\

 

 

After Thorpe & Smith, (1974).

Figure 45.

194

et al. (1978), Bakor et a1. (1976), Garson and Krs (1976),
Garson and Shalaby (1975), KrOner (1977a, b, 1979), Kroner
et a1. (1979), Vail (1976), Swartz and Arden (l960),_Said
(1962), Brown and Jackson (1960), Brown (1970), Brown and
Coleman (1972), Al Shanti and Mitchell (1976), Shackleton
(1973), Thorpe and Smith (1974), Burek (1969), Abdel-Gawad
(1969), Nur and Ben-Avraham (1978), Hancock and Kadhi
(1978), Le Pichon and Francheteau (1978), Richardson and
Harrison (1976), Pilger and ROsler (1976), Dewey et al.
(1973), Moore (1979), SengOr (1979), SengOr et al. (1980,
1982), Bergougnan and Fourquin (1982), Laubscher and
Bernoulli (1977), Al-Shanti (1980), Hadley and Schmidt,
1978).

The majority of the Afro-Arabian crust was formed
during the Pan-African event by a cratonization process
involving Oblique compression and accretion of oceanic
and island arc systems, shallow-water sedimentation, and
intense plutonism (Engel et al., 1980; Gass, 1977). One
of the most Obvious indications of this cratonization is
the presence of four or five ultramafic Ophiolite zones
which cross the Eastern Desert in a NW-SE direction, Figure
46 (see Neary et al., 1976; Bakor et al., 1976; Gass, 1977;
Stern, 1979). This Pan-African geologic record in the
Eastern Desert consists of a tholeiitic substratum with
overlying immature volcanogenic metasediments and calc-
alkaline volcanics, all of which were intruded by very

deep-seated, primitive quartz diorites, quartz monzonites,

 

 
 
 
  

.0 ,. ..-.-.-I=I=€{€3'? ULTRA MA FIC

000000000000
O 0 O O O O O O O O O O .
O O O O O 0.. 0.. O 0.. O... 0... 0.. 0000000

 

 

 

Figure 46.
Map showing palinspastic reconstruction of the Red

Sea area and the distribution of the Pan-African
ultramafic zones (Ophiolites), (after Neary et al.,
1976).

196

and LIL-enriched granites (Dixon, 1979; Greenberg, 1978;
Boulos et al., 1979). This conversion from an oceanic
setting to a thickened continental crust appears to have
occurred from about 880 to 550 m.y. B.P. (Engel et al.,
1980).

The structural deformation of this oblique compression
and accretion was characteristic of the Pan-African tec-
tonics throughout many parts of Africa, with the lack of
large magnitudes of compressional deformation, and the
predominance of strike-slip faulting. Oblique compressional
tectonics have been shown to generate extremely effective
strike-slip fault systems in both ancient and recent
crustal settings (e.g. Benioff, 1962; Fitch, 1972;
Mitchell and McKerrow, 1975; Phillips et al., 1976;
Arthaud and Matte, 1977; Bak et al., 1975; Tapponnier
and Molnar, 1976. 1979; Molnar and Tapponnier, 1975; Nur
and Ben-Avraham, 1978; Watterson, 1978). Watterson (1978)
emphasized the importance of these primary compressional
shear zones which instill a strong anisotropic structural
grain in the crust and provide preferential planes Of
weakness in subsequent tectonic activity. This style of
deformation in the Pan-African culminated in major strike—
slip dislocations in the Arabian shield during the late
Proterozoic and the early Phanerozoic times (Brown and
Jackson, 1960; Moore, 1979). Brown (1972) indicates that
up to 240 kilometers of lateral displacement occurred

along the Najd transcurrent fault system which Fleck et a1.

197

(1976) estimate occurred over a period of about 50 m.y.

at around 550 m.y. B.P. (see Figure 44). The magnitude of
this fault is comparable to the Great Glen fault of Scotland
and other major transcurrent fault systems (see Moore,
1979). Though more work is needed on the structural history
of the Eastern Desert, a palinspastic reconstruction of the
present Red Sea area makes it likely that the Najd fault
system continues into the northern portion Of the Eastern
Desert (see Moore, 1979).

Strike-slip deformation did not only occur in the
early history Of Afro-Arabia.' It was important throughout
the Phanerozoic up until even recent times (Nur and Ben-
Avraham, 1978; Brown, 1970; Swartz and Arden, 1960; Youssef,
1968; Abdel-Gawad, 1969; Burek, 1969; Garson and Krs, 1976;
Moore, 1979; Hancock and Kadhi,.l978). Garson and Krs
(1976), for example, emphasize the large-scale strike-slip
faulting during the Late Cretaceous along the Precambrian
lineaments (e.g. 100 kilometers Of lateral displacement).
Paleomagnetic evidence does not contradict such lateral
displacement. On the contrary, such offsets may allow a
better solution for the Opening Of the Red Sea (Le Pichon
and Francheteau, 1978; Richardson and Harrison, 1976).

Recent work is beginning to decipher some of the
complex history of the eastern Mediterranean and Afro-
Arabian tectonics (e.g. SengOr, 1979; Hsfi et al., 1977;

Hsfi, 1977; Laubscher and Bernoulli, 1977; Dewey and SengOr,

1979; Dewey et al., 1973; Nur and Ben-Avraham, 1978). The

198

pattern which is emerging from such work is the occurrence
of large-scale strike-slip tectonics induced by oblique
convergence and irregular continental margins (Dewey and
SengOr, 1979; Hsfi, 1977; Nur and Ben-Avraham, 1978). A
consistent result of this style of plate interaction
appears to be the "splintering" of the edges of the con—
verging plates, back-arc spreading, and the formation of
micrOplates. Thus, both compressional and extensional
features are Often recorded in close proximity in space and
time (e.g. Hsfi, 1977; SengOr, 1979; Laubscher and Bernoulli,
1977; Stocklin, 1968). A continuation of this intricate
compression-transform-extension tectonics may be seen in
the Zagros-Levantine—Red Sea activity of the Tertiary (e.g.
Nur and Ben-Avraham, 1978) and that of the Aegean region in
more recent times (Dewey and SengOr, 1979).

Although much of the detailed structural and tectonic
history of Afro-Arabia is unknown, there is some evidence
which may support this general style of tectonics since at
least the end of the Paleozoic era (SengOr, 1979). Since
this time, the African plate has been moving generally
northward and rotating counter—clockwise (e.g. Laubscher
and Bernoulli, 1977). This general movement continued
throughout the spreading-Of the Atlantic and Indian Oceans,
until the relatively recent Red Sea extension. This resulted
in a dominant compressional stress in northeast Africa and
Arabia during much of the Phanerozoic (Dewey et al., 1973).

An important mechanism in this compreSSion is considered to

199

be the Oblique convergence of Africa and Eurasia with the
closing of the Tethyan ocean basins and the corresponding
Opening of the Atlantic and Indian Oceans, Figure 47. This
oblique convergence may have been accentuated by the irregu-
lar continental margin of the Arabian Promontory. The
oblique compression and subduction of the Tethyan oceanic
crust may have resulted in the reactivation and further
propagation of the fractures and strike-slip faults within
the Arabian Promontory and the adjoining Nubian Shield.
Gflmaexistence of the major Pan-African dislocations in Afro-
Arabia coupled with these types of regional forces may

help to explain the persistence of strike-slip tectonics
throughout much of the history of Afro-Arabia.

A major point to note in relation to the small amount
of data available on the tectonic development Of the
Eastern Desert is the importance Of active intraplate zones
Of weakness. All the Egyptian alkaline complexes appear
to lie along the major lineaments of the Nubian shield,
Figure 44 (El Ramly et al., 1969a, 1969b, 1971; Garson and
Krs, 1976; Lutz, 1979; Serencsits et al., 1981). The
long—term reoccurrence of alkaline magmatism along these
lineaments and the indications Of multiple reactivations
of these planes of weakness throughout the Phanerozoic,
may be indicative Of an intimate relationship between the
alkaline magmatic process and this style of intraplate

deformation.

200

 

  
   

5
J -ntf’
\ fx'ap‘ R-’ $

‘ ‘ PAlEo-TETHYS

a.--”

 

 

 

Figure 47.

Map showing paleopositions of the major continents
during the Triassic, about 220120 m.y. B.P. Note that
the opening of the Atlantic will be accompanied by a
correspOnding closure Of the Paleo-Tethys. Heavy black
line represents the east-west Mesozoic break-up of
Pangea; barbed lines represent Mesozoic subduction
zones: "*" marks the alkaline provinces of Egypt and

the White Mountain Magma Series.
(After Laubscher and Bernoulli, 1977).

201

2) The Episodes of Alkaline Magmatism in Egypt

Figure 48 illustrates a schematic interpretation of
the development of Afro-Arabia and the emplacement of
alkaline melts along the major fractures in the Nubian
shield. These fractures are considered to have been
initiated during the formation of this crust due to the
oblique compressional tectonics of the Pan-African event.
The complex interaction of the Tethyan, Eurasian and
African plates caused the continued reactivation of these
fractures. The relative rotation of the African plate
and the irregular geometry of the leading continental
margin of the Arabian Promontory both may have induced
continued oblique compressional tectonics and a mechanism
by which these reactivations occurred. The episodes of
similar alkaline magmatism in the Eastern Desert are con-
sidered to have coincided with major shifts in plate motion.
It is prOposed that these alkaline melts were products of
shear heating between adjacent volumes of the asthenosphere
as the slow, solid-state convection of the mesosphere
adjusted to the differential motion of the overlying plates,
Figure 49. Their generation reflected the thermal dis-
equilibrium between the lithosphere and the mesosphere
during these plate accelerations. Thus, shifts in the
motion of these plates caused the generation of small
amounts of alkaline melt in the asthenosphere, and also
aided in their accessibility to the surface by promoting

the reactivation Of the fracture system throughout the

202

Figure 48.

A schematic illustration of the development Of Afro-

Arabia:

a) Oblique subduction-accretion, formation of Pan-
African Afro-Arabian crust, the envelopment of
oceanic material (Ophiolites), the development
of strike-slip fault systems (splaying etc; e.g.
Najd transcurrent fault system), and a fracture
pattern; emplacement of alkaline magma near the
end of this Pan-African event along the Najd
shear zone (Wadi Dib, 554 m.y. B.P.).

b) Periods of reactivation and strike-slip faulting
during major shifts in plate motion throughout
most of the Phanerozoic eon (dominated by plate-
wide African-Eurasian convergence), and the
emplacement of alkaline magma along these zones
of weakness--especially at their intersections.

c) Late Cretaceous strike-slip deformation and the
emplacement of alkaline magma during the intense
convergence in northeastern Arabia, Iran, and
Anatolia, followed by doming (not shown), mantle
up-welling, and rifting along intensely weakened

shear zones (Red Sea).

 

 

 

 

\
\

 

l/

0‘7"" m... “land are: -_ __.__/\
8 vs\ ‘ /\\

 

 

 

 

 

.llll'.

Bi
)1 \
«IS 1

 

 

 

LOBLIQUE SUBDUCTION-ACCRETION.
~880'550llll

0 Alkaline Magmatism

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

b ‘ q 0
‘\
I
E
2. BLOCK FAULTING / SHEARING.
~ 550m -Late Cretaceous‘
C o— —o I
_J r, ’\ 0— 7 -O t I \‘I’ D

 

 

 

 

 

 

 

Figure 48.

 

3.8TRIKE-SLIP PRINTING. IIPJING (Domingnot Ml

 

 

204

 

 

0 Alkaline Magmalosm

;)\1f:.k / :.

'u' IV."-

   
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A
3
3 AFIICA
3
¢
/4¢%{"hssr' . ‘ n
:\ ‘ l \
=LDM' \ ’5’ g." . .... t 6.. r . “ - 0
ENE ‘ E /....40 066a ‘ \
\ 'f 6'... I - ‘ Q ‘ ‘
I \ -
t if: I
I ’ -
s
\n
F
M E S O S P H E H E
Sana Stale Convecuon
Figure 49.

Model of alkaline melt production via shear heating

in the asthenosphere in response to major shifts in
the motion of the African plate and their ascent

along the Pan-African fractures. Reactivations Of
these fractures and the ease of ascent of the alkaline
melts may have been especially favorable in Afro-Ara-
bia due to shearing and strike-slip faulting between
the Arabian Promontory and the Nubian shield. Such
deformation may have been largely the result of the
oblique convergence between the African and Eurasian
plates (Oblique subduction-transform motion-extension
of the numerous Tethyan ocean basins), as well as the
irregular continental margin of the Arabian Promon-
tory. Dotted area outlines a general region of con-
centrated shear dissipation and melt release. No scale
intended.

205

Phaneronic eon. This reOccurring activation and intrusion
of these fractures may have led to a significant weakening
within the plate and the eventual formation of the recent
Red Sea plate boundary at an oblique angle with respect to
the Pan-African Ophiolite suture zones, Figure 48c (and

refer to Figure 46).

A. Thermal Setting

The early emplacement of alkaline magma at Wadi Dib
(554 m.y. B.P.) in the last compressional stages of the
Pan-African subduction-accretion process may reflect the
relatively low thermal gradient characteristic of the Pan-
African tectono-thermal event (Greenwood et al., 1976;
Gass, 1977), and the lack of a continent-continent collision
in the formation of Afro-Arabia (Kroner et al., 1979).
Throughout the following numerous reoccurrences of alkaline
magmatism in the Eastern Desert of Egypt, there appears to
have been an overall change in the degree of alkalinity and
silica-saturation, see Figure 3. The older complexes
are consistently silica—oversaturated and relatively low
in alkalis, whereas the younger complexes are richer in
alkalis and contain silica-undersaturated mafic and salic
rock-types as well as silica-oversaturated salic rock-types.
This change in the compositions of the alkaline melts over
most of the Phanerozoic eon is interpreted to be a function
of the decreasing thermal gradient after the formation of

the AfrosArabian Pan-African crust. With the general

206

decline in the thermal gradient and a gradual thickening
of the lithosphere, it is suggested that the localization
of the effective zones of shear heating during subsequent
periods of plate acceleration would migrate to progressively
deeper levels and contain smaller degrees of melting.
According to high pressure eXperimental work (e.g. Green,
1970), this would lead to more silica-undersaturated alka-
line melts. This longeterm chemical trend throughout the
evolution of the Egyptian alkaline province is interpreted
to further substantiate that the mechanism responsible

for these melts was not restricted to any single type of
tectonic event. Rather, it is prOposed that the Egyptian
alkaline melts were generated as a result of a similar
deep shear heating mechanism in response to any type of
major shift in plate motion, within an overall tectonic
regime of increasing stability and decreasing thermal

gradient.

B. Tectonic Setting

There are four general approaches which support the
coincidence of the episodes of alkaline magmatism with
periods of regional structural readjustment and major
shifts in plate motion:
1. limited information on the structural history of Afro-
Arabia,_
2. Africa's polar wandering path which may be interpreted

as an indication of its motion since Pan-African times,

Table 6,

Table (5. Changes in plate motion associated

22(3'7

episodes of alkaline magmatism.

Mioda of my: 29
aauganagx

van-African event of Iaetern Deaert,

Igypt: iooo-sso -.y.t.p."11. Large
ooale tranecurrent faulting (lajd

fault ”eta. active 3.0-530 m.y.].P.

toward end of 'ao-Afrioan event or
'lijaa Orogenic eycio')""3'1‘.

Initial break-up of Pengeelsleeaing
to extenaion o! I. A-erica-Atrica/
I. Aeerioa. Overall owreeaion and
abduction of 'aleo-hethye north of
Afro-Arabia: poaaihle hack-arc type
”reading of 'Ci-erian' land one
tree It portion of Gonhanaunclud-
ing Afro-Arabia) and formation of

the loo-Tethys“.
Large~acale dextral ehear (ain-

inn of 3500 in) between Laureeia
and Goodwena17.

initial rifting o! the South Atlan-
tic: Africa-S. America and Ienue
Trough ertenaion 7. Correeponding
eonreeaion in Afro-Arabia and
Tethyan 6min. atrike-elip fault-
ing. cloeely followed Dheirgtenaion
between iron and Arabia ' and
pooaihle eeparation of the Anato-

lian plate 1m Afro-Arabia”.

Iejor ehitt in epreaaing of the
South Atlantic and aaaociated
eeoono eejor epieode of alkaline
eepetiae 00.90a.y.l.t.n'u.
urgeoecale atrihe-elip faulting
in Afro-Arabia," 2. ioteoee ou-
lreeaion in iran and Turkey with
eeaociated cphioixzoa"'1"”"‘.
who.“ of hen ofliolitea
in It Arabia’s

lee lea “ing and eateoeioo”'”.

l. Iriden and caaa, 1974.
2. Vail et al., 1077.

3. hith, l!7l.

6. laii, 1071.

3. Diet: and Iolaeo. ”70.
0. Late, 1070.

7. lerenoeita at al., 1001.
I. laehad and Il body, 1979.
3. teen. 1777.

lo. legel et al., 1070.

11. legal et al.. 1900.

12. loore, l9”.

Perioda of logic m
m}; M Mlltive notion
lend__e_rin_g _ig between Africa
211615;: See bevel2 %' '17
yr ‘Tlfil— ‘
000-060
“0
410-390
020-220 350
330-320
200
22°_1.5 250-20017
140 105-00
(Ilia-Cret. to
Late Cret.)
llo-do ’5 .0-00
(Late Cret.
00 to tarly
locene)
00—0 30
(lerly
hoeee)
l3. fleet et al., l376. 35.
10. Iron. 1072. 2‘.
l5. Cea- et al.. 1970. 37.
ls. leogsr. 1979.
17. floral and lrving. 1900.
ll. Dewey et al., ”73. so.
1!. Alavi. 1000.
20. Leuhaeher one Ieraoulli, 1977.
21. lar-h, l’73.
22. Syhea. 1070. 30.
23. Mylo and leonooald, 1073.
20. van Ireueo and Iowdeo, 1073. u,

 

with

351

229 . 223. 221

145,102,100
139.132

OLD!

recent alkali
eopetiem
aeaociated

“£15 ”a.”
opening .

81 my et al., 1971.
!oueee!, l0“.

Iwarta and Arden. 1960.
20. Derek. 3000.

2!. Abdel-Genoa. 1009.
Ieneooh ed Kadhi, 1970.
ll. Brown. 1070.

32. Iaraoe eel In. 1070.
1:. ani. 1911.

Diner-Ire- eel ilhen, 1077.
3!. Ma et el.. 1073.
Dilger eat ‘aler. 107C.

37. Le Pichon eel Francheteau, ”7|.

m It al., 12.1.

208

3. motion of the adjacent oceanic plates which may be
applied to the past 180 m.y. B.P., Table 6,
4. global sea level curves, Table 6. Unfortunately, only
the second-order, smoothed curves have been published (Vail
et al., 1977). The detailed global sea level curves are
held as prOprietary information by Exxon Production Research
Company. With the exception of the earliest episode of
alkaline magmatism in Egypt (554 m.y. B.P.) where the
curve is supported by very little data, all the episodes of
alkaline magmatism correspond to rapid changes in the curve.
The episodes of alkaline magmatism at l40, 351 and 404 m.y.
B.P. coincide exactly with major inflections in the global
sea level curve. The episodes at 90 and 224 m.y. B.P.
correSpond to periods of numerous rapid changes leading
toward a major peak in marine transgression. Obviously,
more study is needed on the use and significance of these
curves, however, one of the most important mechanisms
for large changes in sea level is rapid tectonic movement.
Assuming that these rapid changes in the curve correspond
to large-scale tectonism, then these observations are
consistent with a model in which alkaline magmatism tends
to occur during major plate accelerations.

The first known alkaline magma in Egypt (Wadi Dib,
554 m.y. B.P.) intruded the northeastern portion of the
Nubian shield during the final oblique compressional
stages of the Pan-African event in the Eastern Desert. It

is suggested that this alkaline magma was emplaced within

209

the large-scale strike-slip tectonics of the Egyptian
portion of the Najd transcurrent fault system. This fault
system was undergoing about 240 kilometers of lateral
displacement during the same time as the emplacement of
this early alkaline magma (see Moore, 1979) Figure 44.

Two alkaline complexes dated at 404 and 351 m.y.

B.P. have uncertain tectonic events associated with their
emplacement, but similar to the alkaline magmatic episodes
that follow, they coincide with periods of rapid change in
the inferred ocean levels around the world, Table 6. This
correspondence is taken as an indication of abrupt changes
in the tectonic pattern of the p1ates--changes which
affected the ocean basin volumes and coincided with the
outbreak of alkaline magmatism.

The 224 m.y. episode of alkaline magmatism occurs at
the end of the second rapid-shift in the motion of the
African plate, Table 6. The general period of time marked
the breakup of Pangea (Gass et al., 1978), with extremely
large amounts of dextral shearing between Laurasia and
Gondwana (a minimum of 3500 kilometers over a period of
time from 250 to 200 m.y. B.P., Morel and Irving, 1981;
see also Arthaud and Matte, 1977). Swanson (1982) considers
large-scale dextral shearing as an early transform history
for the initiation of the central Atlantic rift (see also
Cherkis et a1. 1973). It is in this period of time that
the Paleo-Tethys began closing with the proposed back-arc

spreading of the Neo-Tethys (forming the "Cimmerian"

210

landmass, Sengbr, 1979; see also Stocklin, 1965; Tekeli,
1981). The Levantine transform boundary along the Arabian
Promontory may have also been activated during this period
of rapid plate movement (Dewey et al., 1973).

The 140 m.y. and 90 m.y. episodes of alkaline magmatism
in Egypt coincide with the Jurassic and Cretaceous shifts
in the relative motion between Africa and Europe, Table 6.
These shifts have been inferred from the conjugate plate
motion involved in the spreading of the Atlantic and the
closing of the Tethyan domain (Dietz and Holden, 1970;
Smith,_l97l,‘Hsfi, 1971). The late Jurassic episode occurred
during the final closing and collision of the Paleo-Tethys
(Sengor, 1979), with the corresponding Opening of the
South Atlantic (Pitman and Talwani, 1972). The Late
Cretaceous episode occurred during the hairpin turn in the
African polar wandering curve, Table 6, and a major shift
in the spreading of the South Atlantic (Sykes, 1978). This
was accomplished by corresponding compressional events in
the Tethyan area. This involved vigorous strike-slip
faulting in Afro-Arabia during the Late Cretaceous (Garson
and Krs, 1976; Youssef, 1968; Swartz and Arden, 1960; Burek,
1969; AbdeleGawad, 1969; Hancock and Kadhi, 1978; Brown,,
1970), the obduction of the Oman Ophiolites on the north-
eastern margin of Arabia (Glennie et al., 1972) as well as
the Late Cretaceous Ophiolites in Anatolia (Pinar-Erdem
and Ilhan, 1977), and the intense compressional deformation

in Iran (Hsu, 1977; Alavi, 1980; Laubscher and Bernoulli,

211

1977; Dewey et al., 1973; Stocklin, 1968). Again, there
is a correspondence between these episodes of alkaline
magmatism and the sharp infleCtions in the prOposed global
sea level curves.

The coincidence of alkaline magmatism in widely
separated areas is well illustrated in the case of the
Egyptian and Atlantic alkaline provinces. The 224 m.y.
episode of alkaline magmatism in Egypt coincides with the
first known major episode of alkaline magmatism in the White
Mountain Magma Series (Foland and Paul, 1977) along the
crustal lineaments within the (still) adjoining North
American continent. Figure 47 gives an indication of the
relation of this North American tectonic lineament with
Africa, possibly along the embryonic New England Seamount
Fracture (Dewey et al., 1972; Foland and Faul, 1977), the
South Atlas shear zone (Arthaud and Matte, 1977; Dewey et
al., 1973), and the major plate boundaries between Africa
and the Tethys (Dewey et al., 1973). These tectonic
boundaries appear to have been the focal point of the
tremendous strike-slip plate movement between Laurasia and
Gondwana during the same period of time as this alkaline
magmatism (Morel and Irving, 1981: a minimum of 3500 kilo-
meters of dextral displacement) and may represent an early
strike-slip stage of the Atlantic (Swanson, 1982).

The 140 m.y. and 90 m.y. episodes of alkaline magma—
tism in Egypt coincide with the two major episodes of

alkaline magmatism occurring in the areas surrounding the

212

South Atlantic. Alkaline magmatism around 140 m.y. B.P.

in Brazil, Uruguay, Bolivia, and western and southern Africa
has been related to the initial rifting of Africa from South
America (Marsh, 1973; Herz, 1977; Sykes, 1978; Neill, 1973;
Darbyshire and Fletcher, 1979; Amaral, et al., 1967).

A second episode of alkaline magmatism in many of these
areas around 80-90 m.y. B.P. has been correlated with a
major shift in the spreading of the South Atlantic during
the Late Cretaceous (Marsh, 1973; Opdyke and MacDonald,
1973; Herz, 1977; Sykes, 1978; Boyd, 1973).

Thus, periods of rapid convergence and associated
strike-slip motion with the closing of the Tethyan basins
coincide with conjugate stages of Opening and transform
motion in the deve10pment of the Atlantic. Considering
this reciprocal plate motion, the synchronous emplacement
of these alkaline melts is not likely to be the result of
unrelated mantle disturbances. It is prOposed that shear
heating in the asthenosphere due to these major changes
in plate motion is a more likely explanation. Depending on
the presence and activation of major fractures, the local-
ization of possible concentrated shear dissapation in the
underlying asthenosphere, and the lack of interference or
mixing from larger degrees of non-alkaline melts, alkaline
magmatism may occur simultaneously in several portions of
the world in reSponse to the accelerative motions of the

plates.

APPENDICES

APPENDIX I

APPENDIX I

DESCRIPTION OF FOUR EGTPTIAN ALKALINE COMPLEXES

l) Gabal El Naga: Latitude 22°42'00"N, Longitude

 

34°28'25"E; Elevation: 827 meters; Height above

surrounding wadi: 300 meters; Maximum Diameter: 4

kilometers; Age: 142 m.y. B.P. (Lutz 1979).

Gabal El Naga is the most unique of the four complexes
in this study. It alone contains amphibole as the dominant
ferromagnesian mineral.* It also lacks the extremes of
quartz—rich and highly silica—undersaturated rock-types seen
in the other complexes. Even so, the more silica-under—
saturated intrusive rock-types occur in the central portions
of the complex. The difference in silica-saturation levels
and the predicted chemical gradient would simply be less
dramatic than in the other complexes. Unlike the other
complexes, the majority of the rocks at El Naga have a less
altered appearance. This is especially noticeable for the
volcanics, which are fresh at El Naga and extremely altered

in most of the flows at the other compleXes.
G. El Naga is an eXtremely well—formed ring structure

made up of an outer and inner ring, with the more stock-like

 

*El Ramly et al., 1969b, identify the dominant ferromag—
nesian mineral as aegerine-augite.

213

214

syenite bodies centered about the inner wadi. The country—
rock is dominantly gneissic country rock, often with
porphyroblastic textures. This study found that the outer
ring consists of predominantly syenite rather than the
gneissic country rock described by El Ramly et al. (1969b).
The large zones of fenitization mapped by El Ramly et al.
(1969b) were identified as syenites by this study. On the
contrary, apart from narrow zones of alteration coinciding
with fault zones and a rather altered condition in the
syenites of the lower portions of the outer ring, El Naga
contains much less evidence of widespread metasomatism than
the other complexes visited.

The outer ring consists of silica-oversaturated syenites:
hard, partially silicified, fine crystalline textures in
the upper part of the ring; softer, coarse crystalline and
yellow-stained syenites in the lower part of the ring. The
outer ring is dissected by hydrothermally altered fault
zones which El Ramly et al. (1969b) found to extend into
sheared, granulated zones in the porphyroblastic gneissic
country-rock. The inner ring contains three major intrusive
rock-types capped by older, silica-oversaturated trachy-
basalts. The oldest intrusive rock (cross-cutting relation-
ships) is a rather enigmatic variety not seen at the other
alkaline complexes, and called a "hybrid" rock by El Ramly
et al. (1969b). ‘This is a silica-oversaturated fine to
medium-grained syenodiorite, which has a fused,_recrystal—

lized appearance in thin-section. It may be related to the

215

early magmatic phase accompanying the eruption of the
trachybasalts. The remaining two types of intrusive rock
in the inner ring are silica—undersaturated and are much
more abundant than the small zones of "hybrid" or syeno-
diorite. The older variety (cross-cutting relationships)
consists of extremely coarsesgrained alkali feldspar laths,
euhedral amphibole, large biotite books, and nepheline
(called umptekite by El Ramly et al. 1969b). The later
variety is more silica-undersaturated, contains more
nepheline, and some interstitial analcime or sodalite.

The lack of widespread metasomatism, the crystalliza-
tions of amphibole rather than pyroxene, and lack of a
strong silica-undersaturated to oversaturated gradient
from the inner to outer portions of the complex may all be
related to the retention of this magma's vapor phases and
the lack of an effective transport medium for silica and

other components.

2) Gabal Nigrub El Fogani: Latitude 25°51'29"N, Longitude

 

34°56'49"E; Elevation: 1078 meters; Height above

surrounding wadi: 500 meters; Maximum Diameter: 6

kilometers; Age: 139 m.y. B.P. (Lutz 1979).

Gabal Nigrub El Fogani is the most intensely altered
complex of the four complexes sampled. It alone contains
carbonatite dikes and net vein complexes. It also contains
the largest amount of volcanic rocks.

This is a highly eccentric ring structure consisting

216
of an incomplete outer ring which is fragmented in the
southern portion, and a large central intrusive stock in
the northeastern part of the complex, next to the eastern
side of the outer ring. The surrounding country-rock is
granitic and amphibole gneiss. Extremely quartz-rich
syenites (alkaline granites) compose the outer and tOp
portion of the outer ring. From the textures, much of the
quartz appears to be secondary in origin. The inner side
of the outer ring consists of silica-saturated or slightly
undersaturated syenites which E1 Ramly et al. (1969b)
call leucocratic essexites. This same rock-type abuts the
central stock on the eastern side of the complex. The
central stock is made of pink, intensely-altered nepheline
syenites. This alteration has contains ferrugination and
liebertinized nepheline (adds to the pink coloration) to
the extent of reducing some outcr0p lepes to a pink
powder. From rather obscure crystal relationships in
thin-section, much of this central stock may have contained
a large amount of sodalite which has since been destroyed.
These central syenites appear to have crystallized last,
and since the surrounding earlier rock-types do not display
similar intense alteration, it is inferred that most of
this alteration of the nepheline syenites was caused by
autometasomatism. This high abundance of fluid phases,
carbonatite dikes,_voluminous effusive rocks, and the
inferred efficient silica-transport toward the outer margins

of the complex all may be related to the effective"

217

separation of this magma's vapor phases via discrete

separation and volatile transfer.

3) Gabal El Kahfa: Latitude 2508'18"N, Longitude

 

34°38'55"E; Elevation 1018 meters; Height above

surrounding wadi: 120 meters; Maximum Diameter: 4

kilometers; Age: 91 m.y. F.P. (Lutz 1979).

Gabal El Kahfa contains evidence of explosive magmatic
activity and widespread alteration which is different than
that found at Nigrub El Fogani. El Kahfa is very similar
to Gabal Abu Khruq in mineralogy, and representation of
rock-types. However, it does not have large amounts of
foyaites (sodalite-nepheline syenites).

This complex consists of a wide horseshoeeshaped
outer ring which Opens to the south. Between these southern
terminations of the outer ring lies the central stock. The
country-rock is amphibole schist which is Often highly
altered around the complex. Volcanics which are inferred
to be comagmatic, and gabbros also occur around the ring
structure. Both rock-types are intensely altered. (The
highly altered gabbros were included as "epidiorites" or
country rock by El Ramly et al., 1969b.) Once again, the
outermost syenites are the most silica-oversaturated, the
syenites on the inside of the outer ring (although very
altered) are near silica-saturation, and the central stock

contains highly silica—undersaturated syenites.

218

The outer ring consists of predominantly altered
quartz syenites with fused textures, brown-stained alkali
feldspar laths with vermicular crystal boundaries, corroded
aegerine-augite subhedral crystals with associated Opaque
alteration products, and commonly blastic textures with
the graphic quartz. The inner portion of the outer ring
is a wide cataclastic zone made up Of alkaline syenite
with noticeably less quartz and altered essexite gabbros.
These cataclastic zones consist of extremely broken-up
rock--from large boulder-size blocks to fine granulated
rock materia1--zones Of shearing and hydrothermal alteration.
This physical and chemical alteration occurred before the
emplacement of the late-stage syenite dikes which cross-
cut these cataclastic zones.

The inner stock consists of a small prOportion of
nepheline syenite. These nepheline syenites are surrounded
by a predominance of essexite gabbros which are similar
to the highly altered gabbros outisde the ring structure,
although the inner gabbros are less mafic. Considering
this close association in the central stock between the
gabbros and nepheline syenites, a special attempt was made
to distinguish the contacts. Either these rock boundaries
are Obscure for some unidentified reason, or the contacts
are gradational.

Within the central stock are several plugs which E1
Ramly et al. (1969b) refer to as "ore essexite diabases".

These rock bodies have sheared and altered contacts with

219

the surrounding gabbros. They consist of fractured and
granulated mafic material which are considered to represent
possible diatremes of comagmatic ultramafic cululates from
the deep portions of the complex's substratum. The

matrix is recrystallized or blastic alkali feldspar and
plagioclase, scattered throughOut with fractured euhedral
pale-colored or colorless pyroxene (diopsidic), subhedral
Olivine, and biotite. Close by these possible xenolithic
plugs are volcanic vent rocks (agglomerates and breccias)
forcibly intruded (unusual "squirted" appearance) into the
surrounding coarse—grained rock-types. There are cavities
in these volcanics which contain free-surface quartz
crystals. Some of the volcanic rocks include very mafic
compositions. The exact structural relationship of these
remnant flows to the surrounding intrusive rocks and
xenolithic bodies is uncertain. However, field relation-
ships allow the possibility that the olivine-diOpside xeno-
lith "plugs" and the cataclastic zones occurred in a single
violent event, before the termination of the magmatic
activity at El Kahfa. They may represent an explosive
episode involving a build—up and release of magmatic vapor
phases which aided in breaking up the syenites and gabbros,
forcibly intruding the xenolithic cumulate material from
an inferred ultramafic substratum, as well as causing wide—
spread chemical alteration within and outside of the
complex. 'This intense alteration is reflected in the

chemical data on El Kahfa.

220

4) Gabal Abu Khruq

 

A. General Petrography of the Rock-Types at Abu Khruq

Wehrlite

 

This study is not aware of any previous reports of this
diOpside-olivine rock from any Of the Egyptian alkaline
complexes. The sample was collected from a small outcrOp
on the outisde of the northeastern portion Of the outer
ring, refer to Figure 2. The structure of this outcrOp
does not suggest that it is a dike. It may be a xenolithic
mass brought up from the lower portions of the complex, or
it may represent a partially exposed, in situ outcrop of
an ultramafic substratum below Abu Khruq.

This rock-type displays a cumulate texture, it is
highly serpentinized, and is composed Of predominatly
euhedral diOpsidic-pyroxene crystals (about 60-70%, smaller
amounts Of fractured euhedral foresterite-olivine crystals
(about 10-15%), small tabs of interstitial plagioclase
(about 10%), some scattered nepheline crystals (less than
5%), and a widespread serpentinized alteration/groundmass
material. The presence of the nepheline crystals suggests
the comagmatic relationship of this ultramafic rock-type,
and the existence of an early silica-undersaturated alka-
line magma (the REE contents of this rock—type are uneXpect-
edly high). Drilling or geOphysical techniques may be
required to investigate the possibility of ultramafic
cumulate layers below Abu Khruq and other Egyptian alkaline

compleXes.

221

Gabbros

The gabbros represent the most common basic rocketype
at Abu Khruq. Chemical data indicates a comagmatic rela-
tionship as plagioclase-rich cumulates of an intermediate
syenitic liquid.

The eXposures of gabbro both outside and inside the
ring-structure, and their apparent lack of conformity to
the conical (ring) faults, makes them structurally differ-
ent than all the syenites. El Ramly et al. (1969a) suggest
that this relationship indicates the gabbros were emplaced
before the syenites and the deve10pment Of the ring-struc-
ture. They also indicate that there is evidence for the
emplacement of the gabbros after the formation of the
volcanic cone. This sequence would have great implications
on the petrogenesis Of the magma at Abu Khruq.

There are few areas where the gabbros show sharp
contacts with the other rock-types, and these often can
be shown to be dikes rather than the syenite stocks. The
major rock boundaries Often are obscured by colluvium. ring
faults, and/or are gradational with respect to the syenites.
(As at El Kahfa, considerable time was spent attempting to
locate sharp contacts between the gabbros and syenites
without success.)

The gabbros display coarse, ophitic, slightly
miarolitic textures, and are composed of mostly labradorite

laths (An ,a second generation of smaller albitized

66—52”
rectangular plagioclase crystals (total plagioclase:

222

50-65%), and diopsidic augite (20-30%). The labradorite
has welledeveloped albite twinning, occasional percline and
albite twinning, often non-complex zoning, and frequent
thin adcumulus albite overgrowths. The pyroxene occurs
interstitially with respect to the labradorite; the flesh-
colored subhedral crystals often form large continuous
"chains" of pyroxene around the plagioclase laths. Olivine
is rare and always has a reaction rim of pyroxeneiamphibole.
There are very small amounts of orthoclase, occurring as
small interstitial tabs in some of the gabbro speciments.

A common mineral which can represent up to 8% Of the rock
is the Fe-rich biotite (stilpnomelane). This occurs as
large irregular, red sheets, most Often in association

with alteration products of amphibole and Opaques. An
important minor mineral to note in the gabbros is nepheline.
It occurs as small hexagonal and square euhedral crystals
within and around the pyroxene. Thus, nepheline proceeds
the crystallization Of pyroxene. Apatite represents
another important mineral phase as it can amount to almost
5% of the gabbros. The gabbro crystallization phase may
represent a significant proportion of the P205 in the
original melt. It occurs mostly as poikilitic euhedral
crystals within the pyroxene, or immediately adjacent to
the pyroxene_grain boundaries. There are scattered
acceSsory minerals, the most common of which are large
subhedralgrains of sphene, and small spinel euhedral

crystals.

223

Although the gabbros, similar to all the major rock-
types at Abu Khruq, contain certain amounts of alteration,
the fundamental mineralogic and chemical character of
these rocks has been preserved. The feldspars are commonly
partially altered to cancrinite and sericite. Similar to
the syenites, albitization is widespread, and in the
apparent extreme cases, the grain boundaries of the plagio-
clase laths become wavy or vermicular. The pyroxenes Often
alter to amphibole, especially uralite and associated
Fe—oxides. The Fe-rich biotite has been occassionally
converted to chlorite or epidote, but the most common is
its conversion to Opaque Fe—oxides. Zeolites are present

in the interstices of the more altered samples.

Volcanics

 

All the volcanics sampled from Abu Khruq have trachytic
or rhyolitic compositions; there do not appear to be any
basaltic flows. They are predominantly silica-oversaturated
(one analyses from El Ramly et al., 1969a is slightly
silica-undersaturated, 1.84 normative ne), and most samples
contain quartz. The volcanic samples are typically altered,
and based on the consistency and nature of this alteration,
it is likely that autometasomatism was largely responsible.
This cause for alteration is not unusual in many other
peralkaline and alkaline volcanics.

Generally, the volcanics consist of 10-30% phenocrysts.

These are dominated by large, partially resorbed alkali

224

feldspar phenocrysts with irregular, corroded boundaries
which are commonly albitized and leached into the matrix
material. These alkali feldspars display faint Carlsbad
twinning, and are otherwise structureless. Often feldspar
clouding and brown-staining has produced essentially Opaque
alkali feldspars; otherwise they have undulatory extinction.
Occasionally, alkali feldspar also occurs as xenoliths of
partially resorbed crystal aggregates. No other xenolithic
components were noted in the volcanics. A much smaller
generation of albitized alkali feldspar with very diffuse
grain boundaries is scattered throughout the matrix.

The next dominant type of phenocryst is green pleo—
chroic aegerine or blue pleochrOic riebeckite. Both
varieties of ferromagnesian minerals occur in the rhyolites
and trachytes, but not together in a single sample. Some
rhyolitic samples contain virtually no pyroxene or amphibole.
Both aegerine and riebeckite tend to occur as small tabs
with occasional larger fragmented grains which appear to
have given rise to the smaller tabs before crystallization
of the matrix. These ferromagnesian minerals commonly have
reacted to form anhedral Opaques and red-orange alteration
products. Small amounts of the ubiquitous red Fe-rich
biotite is commonly associated with this alteration.

There are small grains Of anhedral quartz in the
rhyolitic samples. Extinction is sharp in these quartz
"blebs", but they often contain a large amount of fine,

unidentified inclusions.

225

The matrix is usually cloudy, with indistinguishable
mineralogies. Typically the matrix displays undulatory
extinction, with a fused and altered appearance. Some
samples show a glassy matrix with fine alkali feldspar,
undulatory quartz, and a large number of zeolites.

It is difficult to make consistent distinctions between
the trachytes and rhyolites. Often there appears to be
examples of each type from apparently single flows. The
trachytes frequently display well-developed flow textures;
the rhyolites tend to be more altered and Often contain
brecciated textures.

There are five main areas in the complex which are
capped by large amounts Of effusive rocks. The peak area
of Abu Khruq contains a larger prOportion of rhyolites with
vent, agglomerate and breccia facies. The large volcanic
cap on the north side of the inner ring contains a larger
proportion of trachytes. All contacts between the volcanics
and syenites demonstrate an older age for the volcanics.
These contacts form as altered, pegmatoidal boundaries for
the syenites, and leached silicified boundaries for the

volcanics.

Syenites

 

A strict petrographic division of the syenites is
difficult because clear sharp contacts are uncommon and
many syenite sampleSdisplay a gradation Of the character-

istics noted for ObViouSly distinct varieties. The extent

226

of this problem was not completely reCOgnized in the field,
and thus, caused a less effective sampling method. Future
sampling should include systematic traverses across the
syenitic bodies--especially in a radial fashion with
respect to the ring structure. The majority of the contacts
at Abu Khruq (as well as the other complexes visited) are
obscured by the conical faults, the intervening wadis,
colluvium, and modification of the rock due to sub-solidus
and/or secondary affects. The presence of secondary quartz
is a major problem in this respect. Though more study is
needed, it is likely that many of the contacts between the
different varieties Of syenite are gradational. This is
obvious for the boundaries between the different varieties
of silica-oversaturated syenites, and undersaturated
syenites respectively, but this gradational relationship
also appears to exist between some of these undersaturated
and oversaturated types. Such a physical relationship

is fundamental to a petrogenetic interpretation.

This study uses the term alkaline syenite for the

 

silica-oversaturated syenites with little or no modal
quartz. The term quartz syenite is used for syenite with
greater than about 5% modal quartz. Silica-undersaturated
syenites without large amounts of nepheline (less than
about 10%) and without sodalite or possible primary analcime

are called pulaskites. The nepheline and sodalite-rich

 

syenites are classified as foyaites.

227

Alkaline Syenites

 

The alkaline syenites are characterized by miarolitic,
coarse-grained textures with a predominance of perthitic
alkali feldspar laths (GS—80%) and a variety Of interstial
pyoxene (lo-15%). The alkali feldspar is usually albitized
with perthitic textures, more widespread replacement, and
small amounts of adcumulus growth. Feldspar clouding and
brown-staining, Often in grid or linear patterns, are
superimposed on the perthetic textures. Sericitization is
common throughout the alkali feldspars. The pyroxene is
dominantly interstitial aegerine-augite with the typical
pleochroic green character (lo-15%). It is typically
miarolitic, with poikilitic or closely associated euhedral
apatite (3-5%), euhedral Opaques (Fe-Ti oxides), and Fe-
rich biotite. The interior of some Of the aegerine-augite
grains zone toward progressively paler green, non-pleochroic
cores which probably represent more augite-diopside composi-
tions. There are small amounts of pleochroic brown-blue
amphibole and needle-shaped, secondary uralitee(less than 5%).

The uralite forms with the ubiquitous amphibole-Opaque-Fe-

rich biotite alteration assemblage which is common through
all the major rock-types at Abu Khruq. One feature which
aides in distinguishing the more mafic alkaline syenites
from the other silica-oversaturated syenites is the presence
of small amounts of oligoclase (5-10%), in the form Of

small euhedral tabs commonly enclosed by the alkali feld-

spars. Thesefrequently have been replaced by cancrinite.

228

The exact location and extent of the alkaline syenites
is uncertain because of the gradational relationships with
the surrounding syenites. In a general way, the alkaline
syenites appear to predominate around the outer portions of
the inner ring (surrounding the foyaites), and the inner
portions of the outer ring (rimming the inside of the
quartz syenites). However, the possibility of secondary
quartz in originally quartz—absent alkaline syenites must
be considered. More importantly, a feature which was
ignored by El Ramly et al. (1969a, 1969b, 1971) is the close
association of the silica-oversaturated syenites with the
silica-undersaturated syenites. Both levels Of silica-
saturation were collected from apparently continuous rock
bodies. Though this relationship needs to be studied
further in the field, the gradational boundaries between
silica-oversaturated syenites (alkaline syenites) and silica—
undersaturated syenites (pulaskites) has serious implications
on a petrogenetic model, and needs to be closely documented.
Such a gradational relationship between similar rock-types
is clearly displayed in the Kangerdlugssuaq alkaline

complex in East Greenland (Pankhurst et a1. 1976).

 

Quartz Syenites

The mineralogy and alteration Of the quartz syenites
are very similar to the alkaline syenites other than the
presence of large amounts of quartz (5-25%). The quartz

occurs in two dominant habits: one is a polycrystalline,

229

interstitial type which does not appear to replace other
mineral phases; the second is a graphic, myrmekitic type
which forms huge Optically continuous grains and can be
Observed to replace alkali feldspar laths and pyroxene (a
minor variety of quartz can be found in the miarolitic
cavities of the pyroxene subhedral crystals). The propor-
tion of quartz in some specimens approaches 25%, leaving

a "skeleton" of fused alkali feldspar laths in which the
grain boundaries as well as the Carlsbad twinning become
vermicular or wavy in outline. The blastic quartz-rich
textures were verified to contain large amounts of secondary
quartz in some samples using cathodeluminescence. The
prOportion of aegerine-augite and apatite is less in the
quartz syenites compared to the alkaline syenites. However,
the same type of ferromagnesian alteration products are
present. Large zones of ferruginated quartz syenites are
visible at Abu Khruq-—especia11y in the outer ring. Some
samples from these zones have been converted to an Opaque,

Fe-rich material, surrounded only by unaltered quartz.

Pulaskites

 

The pulaskites are silica-undersaturated syenites
which contain about 70-80% perthitic alkali feldspar,
around 15-20% green pleochroic aegerineeaugite, about 10%
or less feldspathoids (dominantly nepheline, with rare
sodalite and some secondary analcime), and the remainder

of the rock consisting of euhedral Fe—Ti oxides, red

230

Fe-rich biotite, and accessory zircon and apatite crystals.
There is common sericitization and cancrinitization in the
feldspars; nepheline is much less affected by alteration,
but occasionally is converted to liebernite (pink colora-
tion) and cancrinite. The texture of the pulaskites is
similar to that of the alkaline syenites; the ferromagnesian
alterations are also similar. Nepheline occurs as small
poikilitic euhedral crystals and anhedral "blebs" within
the alkali feldspar, as well as the euhedral crystals in
the interstices. There is a small amount of Olioclase
(5-10%) in the pulaskites. The order of crystallization
appears to be oligoclase—alkali feldspar-nepheline-pyroxene,
with the crystalline of nepheline overlapping with that of
the alkali feldspar laths. The pyroxene, as is typical of
the entire alkaline rock series, is strictly interstitial

with respect to the salic mineral phases.

Foyaites

 

These highly silica—undersaturated, coarse-grained
syenites found in the center of the ring structure, were
the last major phase of crystallization at Abu Khruq. The
dominant mineral is highly albitized, miarolitic alkali
feldspar (SO—70%) which is usually altered in the same
fashion as the alkali feldspar in the other varieties of
syenite. Nepheline accounts for about 20-30% of the rock
in the form of hexagonal or cubic crystal habits. The
majority of the nepheline is unaltered, unlike the alkali

feldspar. Some nepheline, however, has been converted to

231

cancrinite, and more rarely, to analcime. The majority of
the euhedral isotropic mineral in the foyaites is sodalite
(based on preliminary microprobe data), and not analcime as
indicated by El Ramly et al. (1969a). Analcime is observed
to be a replacement mineral of sodalite, and rarely, nephe-
line. It also fills in fractures and cavities in the
foyaites. Sodalite and analcime occupy about 5-15% of the
rock. Pyroxene is again the typical green pleochroic
aegerine-augite variety (5-10%), Often slightly zoned, with
occasional poikilitic inclusions of apatite and nepheline;
it is again associated with the previously described ferro-
magnesian alterations. There is much less apatite and Fe-
Ti euhedral oxides in the foyaites than in the pulaskites;
apatite tends to occur as second generation (?) slender
needles visible only at high magnification. No plagioclase
was identified in these syenites. Accessory zircon (less
than 1%) occurs as small scattered crystals.

El Ramly et al. (1969a) make a textural distinction
between foyaites and ditroites in these sodalite-nepheline
syenites. Though there is no significant chemical differ-
ence, the foyaites display a coarse trachytic texture, and
the ditroites show a finer granular texture. The ditroites
tend to be found structurally higher than the foyaites, and
they contain a much higher proportion of xenoliths. The
sodalite-nepheline syenites are the only rocks to contain
significant amounts of xenoliths. These cognate xenoliths

appear to be very similar petrographically and chemically

232

to some of the syenitic dikes. They can be generally
described as finesgrained nepheline-alkali feldspar micro-
syenites with higher prOpOrtions Of aegerine-augite (needle
crystal habit), and red Fe-rich biotite. The xenoliths
which were analyzed from the ditroites were more mafic than
the host rock; those from the foyaites were very similar to
the host rock. Thus, the foyaites are considered to
represent deeper portions Of a magma which allowed slower
cooling, more chemical equilibration with the cognate
xenoliths, and the deve10pment of coarser, flow textures.
The ditroites, on the other hand, probably represent the
upper pulses of the same magma which accumulated the
majority of the xenoliths from the wall-rock, ascended, and
cooled more rapidly than the lower foyaitic portions. Thus,
less chemical equilibration was achieved between the magma
and xenoliths, and fine granular textures resulted.

In all textures, the sodalite-nepheline syenites
follow the same order of crystallization: alkali feldspar
overlapping with nepheline, sodalite, and pyroxene.

The foyaites can be found only in the central, stock-
like portion of the inner ring. In the outer portions of
these syenite bodies, there are large zones of ferrugina-
tion and shearing. Two mine shafts tunneled well into the
northeastern portion of the foyaites demonstrate how exten-
sive this alteration is. Feéoxidation, calcification,
albitization, liebernitization of the nepheline, analcime

replacement of nepheline, and corrosion of the pyroxenes

233

are all widespread. Fluorite and other secondary minerals
are especially common throughOut these altered syenites.
The structural relationships between the foyaites and
other types of syenite are again Obscure. This is commonly
due to the ferruginated zones and intervening wadis. It
is the observation of this study,_however, that they dis-
play a gradational bOundary with the pulaskites, and thus,
indirectly grade into the silica-oversaturated alkaline
syenites. Once again, such a relationship between the
silica-undersaturated and oversaturated syenites has

important implications on a petrogenetic interpretation.

There are a wide variety Of dikes at Abu Khruq which
were sampled and studied in case they represented an
important igneous phase of activity (e.g. early mafic
phases), or had caused a strong affect on their host rocks
(e.g. metasomatic). There is no evidence for them repre-
senting an extremely large volume of the magma, nor that
they had a large affect on the host rock-types. Refer to
E1 Ramly et al.(1969a) for a description of the different

dike rocks.

B. Summary of the Mineralogy of Abu Khruq

Similar to the other Egyptian alkaline complexes, and
many alkaline suites in general, the order of crystalliza—
tion in the rocks at Abu Khruq begin with the crystalliza-

tion of the salic minerals, and are followed by the

234

crystallization Of the ferromagnesian minerals (agpaitic
order of crystallization, e-g. Kogarko et al., 1978).
Only in the wehrlite (pyroxene—Olivine cumulate) is there
an indication Of the ferromagnesian minerals crystallizing
before the salic minerals. ’Even in this rock-type, however,
nepheline forms in euhedral crystal habit. In the gabbros
labradorite is the primary cumulus phase; in all the
syenites alkali feldspar is the dominant primary cumulus
phase; in the volcanics alkali feldspar is the major pheno-
cryst.

Overall, the crystallization of the major minerals
at Abu Khruq (as well as the other complexes) appears to
have occurred in very discrete intervals of composition.
Evidence of an ultramafic substratum similar to the
werhlite collected suggests that Olivine and calcic-
pyroxene precipitated into dense layers deep within the
complex; the gabbros represent very plagioclase-rich
accumulations in the lower part of the exposed complex.
Because no rock—type was observed to have sub-equal pro-
portions of plagioclase and alkali feldspar, it appears
that there was a rapid transition to the crystallization of

alkali feldspar which dominates throughout the syenites.

Olivine
Olivine is found in large amounts in only the pyroxene-
olivine cumulate (wehrlite). In the only sample collected,

it appears to be foresterite-rich. The only other

235

occurrences of olivine at Abu Khruq are small reacted
grains in some of the gabbros. It appears that by the
time most of the visible rockétypes were crystallizing,

olivine had moved off the liquidus.

Pyroxene

The pyroxene in the wehrlite sample appears to be
diOpside. The gabbros contain a flesh-colored, non-pleo-
chroic diopsidic—augite (see El Ramly et al., 1969a). The
most typical variety of pyroxene for the majority of the
exposed rocks at Abu Khruq is aegerine-augite. This is a
green pleochroic, Often slightly zoned, miarolitic, subhe-
dral mineral phase. It occurs interstially throughout all
the syenites. In much smaller proportions, there are also
other varieties of pyroxene, such as a pleochroic blue-green
type which is easily confused with the minor amounts of

amphibole in the syenites.

Amphibole

 

Amphibole rarely occurs over 2-3% of any of the rocks
at Abu Khruq. The most common form is uralite: a green pleo-
chroic variety which forms as an alteration. product from
pyroxene as clusters of fine needle—shaped crystals. It is
usually associated with the Opaque Fe-oxides and red Fe-rich
biotite which also commonly surround reacted pyroxene.
Other minor varieties of amphibole include a brown, anhedral
alteration product of pyroxene; there is also some kaersu-

tite (blue-green variety) and some minor arfvedsonite.

236

Plagioclase

 

Large prOportions of labradorite (An ) occur in

66-52
the gabbros as coarse-grained laths with well-develOped
albite twinning. Other than this, only very small amounts

of oligoclase were identified in the other rock-types at

Abu Khruq.

Alkali Feldspar

 

The initial composition of the alkali feldspars in the
various rock-types at Abu Khruq is difficult to determine
due to the widespread albitization and other sub-solidus or
secondary reactions (sericitization, cancrinitization,
feldspar clouding, dissolution). In many cases, it appears
to be orthoclase laths in the syenites, with evident Carls-
bad twinning, and albite exsolution textures. Some of the
rhyolites contain sanidine phenocrysts, and the gabbros

have very small amounts of orthoclase.

Feldspathoids

The dominant feldspathoid is nepheline. It appears
in the ultramafic wehrlite sample, as well as in the gabbros,
as small euhedral crystals. It is absent from all the
volcanic samples (nor is there leucite in the volcanic
rocks). It occurs as a cumulus phase in the silica-under-
saturated syenites (pulaskites and foyaites), following
only alkali feldspar in crystallization order.

The other common feldspathoid at Abu Khruq is sodalite.

Based on preliminary micrOprobe data, the large crystals

237

of isotrOpic material in the foyaites are not alacime (El
Ramly et al., 1969a), but sodalite. Analcime, however, was
Observed to replace some of the sodalite, more rarely, the
nepheline, and also fill in fractures and cavities in the

foyaites and pulaskites.

Quartz

Quartz occurs in two dominant habits: one is a poly-
crystalline, interstitial type which does not appear to
replace other mineral phases; the second type is a graphic,
myrmekitic type which forms very large, Optically continu-
ous grains, and can be Observed to replace alkali feldspar
laths and pyroxene crystals. This last type of quartz
involves a secondary origin as verified by cathode-lumine-
scence. Due to the presence of secondary quartz, it is
difficult to determine if certain samples may not have
originally contained quartz; In spite of this possibility,
all intrusive samples with quartz are referred to as "quartz
syenites". Although this classification may include some
alkaline syenites and some slightly silica-undersaturated
syenites, it is important to note that no quartz was
observed in samples of gabbro or samples with large amounts
of nepheline. Thus, either the second generation of quartz
was deposited under near-equilibrium conditions in rocks of
compatible mineralogies, or the original character of the
rocks were radically altered along with the deposition of

quartz. This second possibility is not considered likely.

238

The prOportion of quartz in some quartz syenites
approaches 25%. Such quartz-rich rocks are essentially
"skeletons" of fused alkali feldspar laths and a few green
pyroxene grains surrounded by blastic quartz textures.
Small inclusions of silica are common in these syenites, as

well as in some of the volcanic samples.

 

Fe-rich Biotite (Stilpnomelane)

The most ubiquitous minor mineral in all the rocks at
Abu Khruq is a red, occasionally brown biotite variety
which may be stilpnomelane. Based on the varying degree
of coloration, it contains large differences in Fe content.
In many cases, it becomes very dark red to Opaque, and
therefore, is difficult to separate from the commonly
associated Fe—oxides. It most often occurs in typical
pyroxene-Fe-oxidetamphibole alteration assemblages in all
the intrusive rock-types. More rarely, it occurs inter-
stially with alkali feldspar as large subhedral flakes.
It was observed in essentially all dike rocks, in the
gabbros, and all syenites, but does not occur in the
volcanics. 'The reason for this consistent appearance is
uncertain, but it would appear to indicate a general
characteristic for the entire magmatic series, one involving

a hydrated, Fe-excess reaction.

Fe-Oxides

 

Fe-oxides occur in two textural varieties: one as

euhedral cumulate varieties, the other as anhedral

 

239

alteration products from the corrosion of the pyroxenes

and cumulate Fe-oxides. ’The euhedral crystals are Fe-Ti
varieties (micrOprobe data and see E1 Ramly et al. (1969a),
and are most abundant in the gabbros and more mafic
syenites-~usually in association with pyroxene (often
poikilitically). The anhedral Opaques are most often

found in the pyroxene-Fe-Ti-oxide-stilpnomelanelamphibole
reaction associations. No dense bands of cumulate Fe-Oxides
were Observed at Abu Khruq, as sometimes occur in other

alkaline complexes.

Apatite

Apatite is an important accessory mineral in alkaline
suites--especially in studies using REE since it represents
one of the most common REE-enriched minerals. Apatite,
similar to the Fe-oxides, is sometimes found in dense
cumulate bands in alkaline complexes. This was not observed
at Abu Khruq. It occurs in two dominant forms: one is a
large euhedral variety found within and beside the pyroxenes
of especially the gabbros and more mafic syenites; the
other type of apatite occurs as a second generation of fine
needles seen only at high magnification throughout the

majority of the rocks at Abu Khruq.

Other Minerals

 

Calcite, fluorite, and perhaps other low-temperature
minerals occur in all the syenites——especially miarolitic

cavities of the silica—oversaturated syenites. These types

240

Of minerals were not identified in the gabbros. Zeolites,
cancrinite, sericite, liebernite, chlorite, and epidote
were observed in small amounts as alteration minerals of
all the major rock-types.

Minor minerals originally crystallized in the major

rock-types include sphene, spinel, and zircon.

APPENDIX II

APPENDIX II

A CRYSTAL FRACTIONATION MODEL FOR GABAL ABU KHRUQ

BASED ON THE TRACE ELEMENTS

The objective of this crystal fractionation model is
to present several stages Of crystal fractionation in a
manner which is reasonable for the observed rock-types at
Abu Khruq, and also taking into account the nature of
similar alkaline complexes in other areas. Due to the lack
of well-exposed mafic rock-types at Abu Khruq, and the lack
of a representative of a likely parent melt, this model is
not well constrained. ReView of the crystal fractionation
models in the literature on similar alkaline suites aided
in making certain decisions on the format of the model,
particularly in the early mafic stages. The following
fractionation scheme represents one possible sequence of
crystal fractionations which is consistent with the
observed trace element distributions of the major rock-types
found at Abu Khruq. An attempt was made in the model to
use partition coefficients which 1) are reasonable for an
alkaline suite,_and 2) yield a more conservative fractiona-
tion model (i.e. smaller amounts of crystal fractionation).
AlthOugh the detailed results of this scheme may be altered
with the insertion of different partition coefficients, and

241

242

.nouodnabo no«uuoc«>«~0 cod + oouousado nowuuocoxou>n
awn + nounbom o- + .aouao>0u oco anacooHO> .aouw
ucoaa uuuasu ecu mo noduwuonsoo gash ocu cu nouwsaa
.apaocuooooHOu + unmounou waoxuo uo huucocAEOOOun
ocwuuwecoo nucocoaeoo ouoaoesu «mm + cwovwa eastwuou as
"mummmmm
.noaoca mowuwdaounxuo acocaeoo on» on
unaduou raw: aanfiuooeoocw oca>onob zanonoun who can acOau
nouucoocoo 30H noon on who aocu ousouon autos powuwanEAn
aura an uuzuuou omuuwuu no: oua oo on» .uo .«z unsu muoz.
.nouwo»0u ozu oco
nuacouao> any .aouwcoao uuuosv on» uo ooouo>e
neon a an poucoaouaou ow nods: cone Houoauo
o aw ouch .oooua has» no aouoaoeno suaaucw
on» now) cOAquOOnao OnOHo cw noduouwauounauo
mcaouuco on «teen noun: :3 .HO. canvas anon
nwuou nowuuaopwocuooaoaou .LOAuuuonooHON “aoxa<

Acusvuu aquacoxuv omoua maca>oun EOuu and!

 

"mmwum coduocofluuoum nunsom

.aouauoeso auguuoca>auo noo«>euo ecu + aouonoasu
auguuoooxOuan aoo«>oun ecu + anont poo enou< nounneo O»

hasasa. ..uo. manganese goauuouoauoamuaa .ua + .Ho. anon .«n

 

nauasaom

.Aomnumm too «out: .aouuceaa ouuouOOeel
.ounaoosuuuca. unawaa announce + .aam-¢ o:-

onon< .OunnoO. .uu. nouoaoeso zuuuuoaoHOO«ooHo

. oooua 2525.3 loan .50:

 

"omoum caduocOAuooum pouch
.uooua asofi>oum uo ouousasu sown

ocw>wao + .u0. mouoasfiso nodunocoxouam own + «H0. uaoa awn

 

"magnum:
.AmoHaaoa :« vaucoaounou
uocv pascan Hosofioom + mouenoaso sawuuocoxou>m

mooua unaw>ouo Eouu ado:

 

u OWUUW COHUGCOMuUGHh 0:000“

.vo-¢ oaasaa
on unaasam .Luo. uncaaaao nonhumna>3Ho .oH + .30. have .oa

 

«nuasaom
.AmOHQEou ca paucoaououu Doc. cusvwn «anodeox +
Leona .ouwaunozo ouoaoeso suauuoca>w~o Oauoaouuaa

nevus acouon Owunoeon ocanox~<

 

humoum cOwuo:O«uueuh ueuah

 

 

mucoEon oooua vanquomaou

.cOwuocOwuoouu uo oooaua uouuoa on» uOu noes

ouo oucoEOAo ooouu maoauodeoo on» «nevus on» no huou
Ina: c0auoc0auoouu aauoo us» ca noduoCOHuoouu Houa>uu
uo uuouoo«oc« no con: man aucofiono ououu wunwuooaou one

.uauzx =n¢ dunno how Home:
:OAuouwauoumauo Macawuooum ufluwouwn Ono: 1 .d HM ‘ wanna

243

.mm one an .ce .mum ecu mmoauu
cOuuoc0«uoeuu dosau ecu ceezuen >uceumwucoo ecu euOZc

.Ameuoassso cOwuaecw>aao aoa one .cuauuecexouam own

.uwoucceo onm nood>eun ecu +V neuuo>0u use unucoo

uHO> .meuwcean nuuosv ecu uo cOauwnooEOO xaon ecu Ou

unausaa .mOuocuomapHeu + uoamoaeu «Hexao acucecMEOpeun
mo ocauaaucoo oucecOoEOO euoasasu «mm + vanqua aespwmeu uh
"wwwmmmm

.ueu«u>0u one

.mOacooHO> .ueuwceaa uuuosv ecu uo eoeue>e casc
o an oeuceaeuoeu ea one .eeecoe ucelene eoeuu
eucuuooeoo ecu ca no eooua case nouaauu elea
ecu aw once .eooua awcu no aeuouslao ouuelcu
ecu cu“: caduoquuao encuo cu cauuonuuneuehuo
ocuomnco an euecu caucz cu can. uwsvun «ecu
uuueu co«u-~oo«ocuooaoneu .cOuunuoaaoaeu «Hoxuc

eoeua asow>eun loam uuet

 

"umeum cOwuocOAuooum cuusOh

.mouoaasso oaounnam .vo + .xmp-< .awa-<

.ommumm .oooat .onac .neuuceaa OuuouOanl .euoaoeaueucu

ecu >c oeuceoeuneu. Ado. uaefi Owuuceaa euouoeaueucw o~n

.Aueuuceaa OwuouOOneE .euouoeaueucw. vanqun
costumer Ouuwceam + .eudeoa ecu mo ensue: cwvoenu
unwcoaummx .oenwcuucenuea Ou ecu uOu oeucsooooco.

neueasaso conunecexouxo oco eca>w~o + “odout one

unmac .Ouccom. .AuO. neuousfiso co«u:eae~o°«ooam

 

"eunsaex

. . . Samoa uceuea equaoaoc 093321

”emoum cOwuocOMuoeum ouuca Ou unuum

.esecon uceeeae euouu eacwuomsbo ecu no emoum cauu
nocOMuooum ounce ecu ou oconaeuuoo cauc) eeuouslsu Owouc
scam ecu uo :OMuosoOuo ecu cud: acumen uanaeu «o«u«c« ecu
usc ..020cou useseue eoouu eacwuooaou ecu cu ea eflea.
efiooe uceuoo Owuqomoc ecuaoxco co cu“) acumen auceseae

oomuu eacauonfioocu ecu onw>HO>cu eEecoa cOAuocOwuoouu och

 

 

auceEeHm eoeufi eacauomaoucn

.xo.ucoo. a an a edges

244

.cOuuo:O«uoouu
Houaxuo ueuue ocucueeeu asbefl Hocwuwuo uo COwuoouu

.uceMOwuueOo cOuuscuuuauv xuoc

.cOAuocOHu

Ioouu Houaauu neuuo pusqwm ocwcwofieu ecu uceueudeu
cauch exuou oe>uemco ecu cw ucefieae uo cauuouuceocou

.cOAuocOwuoouu Heuoauo ueuue vasUwH
pe>uuep peuonsuaoo ecu cw ucefleae no cOuuouuceocOo

.neooco euoaseso uo dosOHeeu ecu uceeeuoeu
coach axoou ce>uemco ecu cw uceEeHe uo cOwuouuceocou

.cOAuocOauoouu Heuaauo ueuuo eueasaso no Hosoameu
oe>wueo oeuoasOHeu ecu cu uceaeae mo cOauouuceocOo

.o«aquu uceuon ecu ca uceEene no cOuuouuceocoo

ID

 

.meuaexou

one newco0ao> .aeuucexa uuuesv uo neaaaoco ov uo eoeue>< c

 

mm.o avo.o~ no.- II om.h ac.» MN.
Hm.o avm.o «0.6 II Nv.o cv.o MN.
hm.o aom.od mm.@n II oo.mu no.m mm.
no.0 ah~.ocu oo.coH II am.~m ov.ao MN.

:9
3A
Em
00

 

.uHeE uceuoa Hecwmwuo no no.» "mmoum cOuuocOMuueum cuusou

 

.oNMImmqmws.o~m.eoo.wHI< .meuacexa euowerueucu uo emoue>< +
.oaowc bce onmId .mOucnmnlmO emoue><Ib

o~.o +mv.h oo.a aou.d Ha.o om.~ «n.

Hm.c +mm.o cv.c cow.o HN.O nN.o an.

hm.o +~o.m no.a «mo.v m~.n nm.m mm.

wv.o +oo.~o ov.n¢ ann.wn vv.hm oc.vv an.
AQUEOBUHN

£9
34
Em
00

eoeua eucuuoosoocuc "eooum cauuocOAuoouh chuck ou umuwm

 

.memxwoca on v «o emoue><++
.nobmnv sung poo mpsun amcu aouwlaee>~oce wmlmo eooue><+

Hm.v n m II nN hflcu MN. on

++
om.~ +~n no II om van MN. um

 

cuaefi uceuea acououuo uo no.. ”eueum cOHuo:O«uoouh cuusOh

 

.cmhocv nusq um oeuoeaaoo ouoo Eouu oesao> eumeee necOMuuove

Ozu amHn eaoIc .onmI< nemococmneoHo> eooue>o um amman

w~.o coda sac“ ova vow ban on. an
no.« awn can amen new one on. um
on.H vd mm an nv an on. 00
vv.n ou «N pm or or 90. no

cm.n on 5 am vw mm on. «2

 

cudee uceuon accumuuo «O «n.. "mmwum acuueCOwuoeum chuck

o~.o II haw II no ovv on. an
o~.o II @Ho II won vvm ow. um
on.H II an II Hm ow co. 00
vv.n II or II can AFN on. HO
om.n II mm II on Ha on. «2

 

cuaea uceuum documauo no vm.. "emoum coduoco«uueum ocooem

 

o~.o II ovv a we oov cm. on
o~.o II vvm he «on com oa. um
Ho.n II we cud med co cm. 00
vv.n II Abu one «no can oo. no
oo.nu II am ovon vmmu omn om. «z
"mmouw cauuocOMuueum uauum

9 ac Ho uo h.o cu m

 

deco: couuchAuooum Houexuu ecu «O aueueflouoe one eunsaez

.Av.u=00~ A HH ¢ wandb

245

to a limited extent, with the use of another reasonable
parent melt, the general conclusions are not greatly
changed. Refer to Arth and Hanson (1976), Zielinski (1975),
Sun and Hanson (1976), Allegre et a1. (1977), Frey et a1.
(1978), and Baker et a1. (1977) for the range of published
partition coefficients used in this crystal fractionation
model.

The major conclusions of the crystal fractionation
calculations are:
1. Crystal fractionation was an important process of
differentiation in the rocks at G. Abu Khruq.
2. Large amounts of efficient crystal fractionation
occurred (about 2/35 removal of the parent melt for the
stage after the separation of the gabbroic phases, and
about 90% removal of the parent melt in order to achieve
the bulk compositions of the most evolved rock-types).
3. It is likely that the parent melt of G. Abu Khruq was
a moderately alkaline basaltic magma.
4. The distributions Of the compatible and incompatible
trace elements agree reasonably well so that a fairly

unique solution for crystal fractionation is possible.

1) Choice of Parent Melt

The choice of a primitive parent melt for the alkaline
rock suite at G. Abu Khruq was based on:
1. The relatively low concentrations Of incompatible trace

elements at Abu Khruq and the other Egyptian alkaline

246

complexes visited eliminated the possibility of a nephe-
linitic, basanitic, or other extremely enriched alkaline
basaltic parent. The most likely parent is a moderately-
enriched alkaline basalt.
2. The most mafic rock-types at Abu Khruq are silica-
undersaturated, and therefore, the parent was probably
silica-undersaturated.
3. The mafic rock—types of El Gezira, Egypt (El Ramly
et al., 1971), Jebel Al Abyad, Saudi Arabia (Baker et al.,
1973), Canary Islands (Munoz,_l969), Reunion Island (Upton
and Wadsworth, 1972; Zielinski, 1973), eastern Australia
(Frey et al., 1978), and Hawaii (Schilling and Winchester,
1969; Kay and Gast, 1973; Frey et al., 1978) all aided in
selecting trace element compositions which are remarkably
similar throughout the world for low-REE, fairly undiffer-
entiated alkaline basalts.

It should be noted that the REE distributions for the
theoretical parent of Abu Khruq is essentially that of an
average Hawaiian alkali olivine basalt (Schilling and

Winchester, 1969; Kay and Gast, 1973; Frey et al., 1978).

2) First Fractionation Stage

The first fractionation stage approximates the olivine-
rich separation stage Observed in sample A994, a olivine
(90%) clinOpyroxene (10%) ultramafic cumulate (wehrlite).

5&1 I-15(°9°) + 3(.101 = 13.8'

D

Cr p2.15(.90) + 15(.10) = 3-44

247

DCo = 3.88(.90) + l.21(.10) = 3.61

U
ll

Sr 0.2 ... Sr is an incompatible trace element in this
stage, and this bulk partition coefficient
is reasonable compared to values used in the
literature for similar alkaline mafic or
ultramafic separations (e.g. Allegre et al.,
1977).

D
II
C
H

Ba is an incompatible trace element in this
stage, and this bulk partition coefficient is
reasonable compared to values used in the
literature for similar alkaline mafic or
ultramafic separations. Compared to that used
by Frey et al. (1978) it is larger and thus
will reSult in a more conservative model.

Ba

The F value of .lO'for this first fractionation stage
was chosen based on the observed trace element distribution
of sample A-94. About 10% olivine separation is not unrea-
sonable for the earliest fractionations of an alkaline
basaltic melt. About 15-25% total Olivine fractionation is
likely for alkaline basaltic suites which achieve trachytic
compositions (e.g. Zielinski, 1975; Sun and Hanson, 1976;
Baker et al., 1977). The remaining olivine separation

will be accomodated in the following fractionation stage.

3) Second Fractionation Stage

The second fractionation stage approximates pyroxene-
rich separations which were not Observed at Abu Khruq.
Although such cumulate rock-types were not Observed, it is
a geologically reasonable step between the preceding Olivine—
rich separations and next plagioclase-rich cumulate stage
(gabbros). This stage does not seriously effect the

general magnitudes of the results in this crystal

248

fractionation model. 'Pyroxenites are commonly obServed

in similar alkaline complexes with deeper exposure (e.g.

see Gastesi, 1969; MunOz, 1969). The presence of such
rock—types may be indicated by the cognate xenoliths which
were found in the diatreme-like features observed at G.

El Kahfa (see Appendix I). In view of the fact that this
study is the first to report the presence of the olivine-
rich cumulate, it is not unreasonable to expect that further
investigation of these alkaline complexes will reveal more

ultramafic rock-types in their underlying structures.

DNi = 3.5 ... relative bulk partition coefficient; this

value is reasonable for the mineralogy of
the predicted pyroxenites; it is similar to
the relative bulk partition coefficient used
by Allegre et al. (1977) for the mafic
fractionations of an alkaline suite.

C
II

3.44 ... unchanged from the first fractionation stage;
the relative bulk partition coefficient was
uncertain due to the scatter in the Cr dis-
tributions. This scatter may be largely due
to the unique amounts of Fe-Ti oxides in
individual samples. Due to the extremely
strong incorporation of Cr into the Fe-Ti
oxides, and the pyroxene-rich nature of this
fractionation stage, this rather high bulk
partition coefficient was retained for this
second fractionation stage.

Cr

D
II
H
w

relative bulk partition coefficient; com-
ments on the Ni value of this stage also
apply to Co.

CO

DS = 0.2 ... unchanged from first fractionation stage due
r to similar conditions.

U
(I
O
H

'unchanged from first fractionation stage due

Ba to similar conditions.

249

The F value of .60 for this stage is somewhat arbitrary.
It considers the amount of ferromagnesian separations
necessary before the following more salic rock-types may be
derived (see the following fractionations based on Sr and
Ba), and it also takes into account the amount of ferro-
magnesian separations which are typically removed in other
studies for alkaline suites (eug. Baker et al., 1977;

Zielinski, 1975).

4) Third Fractionation Stage
The third fractionation stage approximates the plagio-
clase and pyroxene-rich separations similar to the observed

gabbros (60% plagioclase, and 40% clinOpyroxene).

Ni = 3.5 ... unchanged

D
II

3.44 ... unchanged

Cr
DCo = 1.3 ... unchanged
DSr = 2.59(.60) + 0.20(.40) = 1.63 ... average values from
published partition coefficients.
BBa = 0.36(.60) + 0.l(.40) = 0.26 ... average values from

published partition coefficients

The F value for this third fractionation stage is a
result of approximating the Observed concentrations in the
gabbros and the intermediate syenites. At this stage

about 68% of the parent melt has crystallized.

250

5) Fourth Fractionation Stage‘

The compatible trace elements with strong ferromagne—
sian affinities are considered to start behaving incompa-
tibly, and their low concentrations are below any feasible

resolution for this crystal fractionation model.

DSr

3.87(.75) + 0.15(.15) = 2.98 ... where the syenites
and volcanics are approximated by 75%
alkali feldspar and feldspathoids, and 15%
clinOpyroxene (a less mafic aegerine-
augite variety).

DBa

6.12(.75) + 0.l(.15) = 4.61 ... same comment as for Sr.

These two trace elements are the most difficult to model
because their bulk partition coefficients change from highly
incompatible ones in the early ferromagnesian crystalliza-
tion stages to highly compatible ones in the intermediate
crystallization stages. Thus, these trace elements become
very depleted in the final stages of crystallization.

The fractionation scheme for the REE and Th is much
more straight-forward because their bulk partition coef-
ficients can be approximated by nearly constant values
throughout the evolution Of the entire melt. These values
are relative bulk partition coefficients for the REE
(derived by the method outlined by Allegre et al., 1977).
The bulk partition coefficient of Th is arbitrarily chosen
as 0.10 for the gabbroic fractionation stage. This value
is in complete agreement with similar values used in the
literature (see for example, Frey et al., 1978). This

extremely small value is likely to increase slightly in

251

the final salic stages of differentiation. Thus, an
arbitrary value Of 0.35 was chosen with reference to the
eXperimental data reviewed by Baker et al. (1977).

In this very late stage of fractionation, it is likely
that there was not a complete separation between the resi-
dual liquid and the crystallizing phases. Thus, these
final alkali feldspar and feldspathoidal-rich cumulates
were probably distributed throughout the final crystal-
lizing rock-types.

It should be noted that the bulk compositions of the
final salic rock-types were used (averages). As was dis-
cussed in the text, crystal fractionation cannot account
for the relationship between the various varieties of salic
rock-types. In spite of the few rigid constraints in this
model, there is good agreement between these trace elements
in terms of very similar F values. Numerous variations of
crystal fractionation models were performed, and consis-
tently this same general type of result was concluded: in
order to achieve the bulk salic compositions of the syenites
and volcanics, about 90% of an alkaline basaltic parent

melt must be removed.

APPENDIX I II

APPENDIX III

NET-VEIN COMPLEX AT NIGRUB EL FOGANI: REE PATTERNS

.
O

 

d
4
J

4

4

J

(I! FELSIC ROCK - NET VEIN
O HRFIC ROCK - NET VEIN

1°

A LLLAL

A

 

 

nopnnLIzso nauuonucc

A A. A LAAA‘

. LR CE SH EU TB YB LU

O

 

 

 

as 37 £8 £9 30 8182838486 36 £7 5'8 39 70 3172
HTOHIC NUHBER
Figure A III 1. Normalized REE plot of the mafic and salic
rock-types collected from a net-vein complex with apparent
liquid/liquid contacts from G. Nigrub El Fogani. Similar to
experimental data on mafic/salic liquid immiscible systems,
these mafic and salic members show that the REE are richer
in the mafic phase than in the salic phase. A crystal frac-
tionation relationship would require the opposite. See
Chapter 2, part B.

BIBLIOGRAPHY

BIBLIOGRAPHY

Abbey, 3., 1977, Studies in "standard samples" for use in
the general analysis of silicate rocks and minerals,
Part 5, edition of "usable" numbers, Can., Geol. Surv.,
Pap. NO. 77+34, 31 p.

Abdel-Gawad, M., 1969, Geological structures of the Red Sea
area inferred from satellite pictures, in get Brines
and Recent Heavy Metal Deposits in the Red Sea,
Degens, E.T., and Ross, D.A., eds., Springer-Verlag,
pp. 25-37.

Adams, J.A.S., Gasparini, P., 1970, Gamma-Ra S ectrometr
of Rocks, Methods in Geochemistry and Geopfiysics, 51g,
ElseV1er, 295 p.

 

Alavi, Mehdi, 1980, Tectonostratigraphic evolution of the
Zagrosides of Iran, Geology, v. 8, pp. 144-149.

Alberding, H., 1957, Application of principles of wrench—
fault tectonics of Moody and Hill to northern South
America, Geol. Soc. Am. Bull., 68, 785-790.

 

Aleksiyev, E.I., 1970, Genetic significance of the REE in
the younger granites of northern Nigeria and the
Cameroons, Geochem. Intl., v. 7, pp. 127-132.

 

Allegre, C.J., Dupre', E., Lambert, B., Richard P., 1981,
The subcontinental versus suboceanic debate, I. lead-
neodymium-strontium isotopes in primary alkali basalts
from a shield area: The Ahaggar volcanic suite,

Earth Planet. Sci. Lett., v. 52, 85—92.

 

Allegre, C.J., Minster, J.F., 1978, Quantitative models of
trace element behavior in magmatic processes, Earth
P1311611. SCie Latte" Ve 38’ ppe 1-25.

 

Allegre, C.J., Treuil, M., Minster, J.F., Minster, B.,
Albarde, F., 1977, Systematic use of trace elements in
igneous processes, Part 1. Fractional crystallization
processes in volcanic suites, Contrib. Mineral Petrol.
v. 60, pp. 59-75.

Almond, D.C., 1977, The Sabaloka igneous complex, Sudan,
Phil. Trans. R. Soc. Lond., 287-A, 596-633.

253

254

Al-Shanti, A.M.S., ed., 1980, Evolution and Mineralization
of the Arabian-Nubian Shield, v. 3, Institute of
Applied Geology. King Abdulaziz University, Jeddah,
Kingdom of Saudi Arabia, Pergamon Press, 164 pp.

 

Al-Shanti, A.M.S., Mitchell, A.H.G., 1976, Late Precambrian
subduction and collision in the Al Amar-Idsas region,
Arabian Shield, Kingdom of Saudi Arabia, Tectongphysics,
v. 30, p. 47.

 

Amaral, G., 1967, Potassium-argon ages of alkaline rocks
from southern Brazil, Geochim. Cosmochim. Acta, 31,
117-142.

 

Anderson, A.T., Greenland, L.P., 1969, Phosphorus fractiona-
tion diagram as a quantitative indicator of crystalliza-
tion differentiation of basaltic liquids. Geochim.
Cosmochim. Acta, 33, 493-505.

 

 

Anderson, D.L., 1975, Chemical plumes in the mantle, Geol.
SOCe Amere Bu11e’ 86' 1593-1600e

 

Anderson, D.L., 1981, Rise of deep diapirs, Geology, 9, 7-9.

Anderson, G.M., Burnham, C.W., 1967, Reactions of quartz
and corundum with aqueous chloride and hydroxide
solutions at high temperatures and pressures, Am.
Jour. Sci, 265, 12-27.

 

Anderson, O.L., Perkins, P.C., 1974, Runaway temperatures in
the asthenosphere resulting from viscous heating, J.
Geophys. Res., 79, 2136-2138.

 

Appleyard, E.C., 1974, Syn-orogenic igenous alkaline rocks
of eastern Ontario and northern NOrway, Lithos, 7,
147-169.

Arculus, R.J., 1976, The geology and geochemistry of the
alkali, andesite association of Grenada, Lesser Antilles
island arc, Geol. Soc. Amer. Bull., 87, 612—624.

Arculus, R.J., Delong, S.E., Kay, R.W., Brooks, C., Sun,
8.8., 1977, The alkalic rock suite of Bogoslof Island,
Eastern Aleutian arc, Alaska, J. Geol., 85, 177—186.

Arth, J.G., Barker, F., 1976, REE partitioning between horn-
blende and dacitic liquid and implications for the
genesis of trondhjemitic-tonalitic magmas, Geology, 4,
534—536.

Arth, J.G., Hanson, G.N., 1976, Geochemistry and origin of
the early Precambrian crust of NE Minnesota,'Geochim.
cosmOChime ACtap 39' 325-362e

 

255

Arthand, F., Matte, P., 1977, Late Paleozoic strike—slip
faulting in southern and northern Africa: result of
a right-lateral shear zone between the Appalachains and
the Urals, Geol. Soc. Amer. Bull., 88, 1305-1320.

Bahat, D., 1980, A Hertzian quasi-oval fracture model for
ring-complexes, J. Geol., 88, 271-284.

Bailey, D.K., 1964, Crustal warping - a possible tectonic
control of alkaline magnatism, J. Ge0phys. Res., 69,
1103-1111.

 

Bailey, D.K., 1970, Vblatile flux, heat focussing and the
generation of magma, Geol. J. Spec. Iss., 2, 177-186.

 

Bailey, D.K., 1974, Continental rifting and alkaline magma-
tism, in The Alkaline Rocks, SOrensen, H.,ed., Wiley-
Interscience, 148-159.

Bailey, D.K., 1976, Application of experiments to alkaline
rocks, in The Eyglution of the Crystalline Rgcks,
Bailey, D.K., MacDonald, R., eds., Academic Press,
419-469. '

 

Bailey, D.K., 1977, Lithosphere control of continental rift
magmatism, J. Geol. Soc. Lon., 133, 103-106.

 

Bailey, D.K., MacDonald, R., 1975, Fluorine and chlorine in
peralkaline liquids and the need for magma generation
in an open system, Min. Nagy, 40, 405—14.

 

Bak, I.K., Grocott, J., Korstgard, J.A., Nash, D.F.,
S¢rensen, K., Watterson, J., 1975, Tectonic implica-
tions of Precambrian shear belts in western Greenland,
Nature, 254, 566-569.

Baker, B.H., Goles, G.G., Leeman, W.P., Lindstrom, M.M.,
1977,_Geochemistry and petrogenesis of a basalt—
benmoreite-trachyte suite from the southern part of the
Gregory Rift, Kenya, Contrib. Mineral. Petrol., 64,
303-332.

 

Baker, B.H., Henage, L.F., 1977, Compositional changes
during crystallization of some peralkaline silicic
lavas of the Kenya Rift Valley, J. VOlc. Geotherm. Res.,
2, 17-28.

Baker, B.H., Mohr, P.A., Williams, L.A.J., 1972, Geology of
the eastern rift system of Africa, Geol. Soc. Am. Spec.

Papers, 136, 67 pp.

Baker, P.E., Brosset, R., Gass, I.G., Neary, C.R., 1973,
Jebel a1 Abyad: a recent alkalic volcanic complex in
western Saudi Arabia, Lithos, 6, 291-314.

256

Baker, P.E., Gass, I.G., Harris, P.G., LeMaitre, R.W., 1964,
The volcanological report of the Royal Society Expedi—
tion to Tristan da Cunha, 1962, Phil. Trans. R. Soc.
Lond., 256-A, 439-576.

 

Baker, A.R., Gass, I.G., Neary, C.R., 1976, Jabal Al Wask,
northwest Saudi Arabia: an Eocambrian back-arc
Ophiolite, Earth Planet. Sci. Lett., 30, 1-9.

 

Balashov, Yu A., 1972, Partition of REE during fractional
crystallization of magmatic melts, Geochem. Intl., 9,
320-335.

 

Balashov, Yu A., Krigman, L.D., 1975, The effects of alka-
linity and volatiles on REE separation in magmatic
systems, Geochem. Intl., 12, 165-170.

 

Ball, E., 1980, An example of very consistent brittle defor-
mation over a wide intracontinental area: the late
Pan-African fracture system of the Tuareg and Nigerian
shield, Tectonophysics, 61, 363-379.

 

Ballard, R.D., Uchupi, E., 1972, Carboniferous and Triassic
rifting: a preliminary outline of the tectonic history
of the Gulf of Maine, Geol. Soc. Amer. Bull., 83,
2285-2302.

 

Barber, C., 1974, The geochemistry of carbonatites and
related rocks from two carbonatite complexes, south
Nyanza, Kenya, Lithos, 7, 53-63.

Barberi, F., Ferrara, G., Santacroce, R., Treuil, M., Varet,
J., 1975, A transitional basalt-pantellerite sequence
of fractional crystallization, the Boina centre (Afar
Rift, Ethiopia), J. Petrol., 16, 22—56.

 

Barker, D.S., 1969, North American feldspathoidal rocks in
space and time, Geol. Soc. Amer. Bull., 80, 2369-2372.

Barker, D. S., 1974, Alkaline Rocks of North America, in The
Alkaline Rocks, S¢rensen, H., ed., Wiley-Interscience,

 

Barth, T.F.W., 1962, Theoretical Petrology, Wiley & Sons,
Inc., N.Y., 197-217.

Benioff, H., 1962, Movements on major transcurrent faults,
in Implications of Continental Drift on the Earth
Sciences, Internat. GeOphys. Series, Runcorn, S. K., ed.,
v. 3, 103- 134. .

 

257

Bergougnan, H., Fourquin, C., 1982, Discussion to Remnants of
a pre-Late Jurassic ocean in northern Turkey: Fragments
of Permian-Triassic Paleo-Tethys?, Geol. Soc. Am. Bull.,
93, 929-932.

 

Bevier, M.L., Armstrong, R.L., Souther, J.G., 1979, Miocene
peralkaline volcanism in west-central British Columbia
- its temporal and plate—tectonic setting, Geology, 7,
389-392.

Bishop, A.C., WOolley, A.R., 1973, Basalt-trachyte-phonolite
series from UaPu, Marquesas Island, Contr. Minerals

 

Blaxland, A.B., van Breemen, O., Emeleus, C.H., Anderson,
J.G., 1978, Age and origin of the major syenite centers
in the Gardar province of south Greenland: Rb-Sr
studies, Geol. Soc. Amer. Bull., 89, 231-244.

 

Booth, B., Croasdale, R., Walker, G.P.L., 1978, A quantita—
tive study of five thousand years of volcanism on Sao
Miguel, Azores, Phil. Trans. R. Soc. Lond., 288—A,

‘271-319.

 

Borodin, L.S., Pavlenko, A.S., 1974, The role of metasomatic
processes in the formation of alkaline rocks, in The
Alkaline Rocks, Sorensen, H., ed., 515-534.

 

Boudier, F., Nicolas, A., 1972, Fusion partielle gabbroique
dans la lherzolite de Lanzo (Alpes piemontaires), Bull.
Suisse Min. Petrogr., 52, 39-56.

 

Boulos, F.K., Morgan, P., Hennin, S.E., El Sayed, A.A.,
Melik, Y.S., 1979, Abstract in Proceeding of the 5th
Conference on African Geology, Cairo, Egypt, 40.

 

 

Bowden, P., 1974, Oversaturated alkaline rocks, in The
Alkaline Rocks, Sdrensen, H., ed., Wiley-Interscience,
109-123.

 

Bowden, P., Whitley, J.E., 1974, REE patterns in peralkaline
and associated granites, Lithos, 7, 15—21.

Bowen, N.L., 1928, The Evolution of the Igneous Rocks,
Princeton University Press, Princeton, N.J.

Bowen, N.L., 1937, Recent high-temperature research on sili-
cates and its significance in igneous geology, Am. J.
SCie ' ser. 5' 33' l—Zle

Bowen, N.L., 1945, Phase equilibria bearing on theorigin
and differentiation of alkaline rocks, Am. J. Sci.,
243 A (Daly vol.) 75-89.

258

Boyd, F.R., 1973, Origin of the lherzolite nodules in
kimberlites of northern LeSothO, paper presented at
lst Int'l. Kimberlite Conference, Int'l. Assoc.
Geochim. Cosmochim. et. al., Univ. Cape Town, South
Africa, (ref. from Hargraves and Onstott, 1980).

Briden, J.C., Gass, I.G., 1974, Plate movement and continen-
tal magmatism, Nature, 248, 650—653.

Brown, G.F., 1970, Eastern margin of the Red Sea and the
coastal structures in Saudi Arabia, Phil. Trans. R.
Soc. Lond., 276—A,_75-87.

 

 

Brown, G.F., 1972, Tectonic map of the Arabian Peninsula:
Saudi Arabian Dir. Gen. Mineral Resources Arabian
Peninsula Map AF-2, scale l:4,000,000.

Brown, G.F., Coleman, R.G.,_l972, The tectonic framework
of the Arabian peninsula, XXIV Int. Geol. Congr., 3,
300—305.

 

Brown, G. R., Jackson, R. D., 1960, The Arabian Shield, Int.
Geol. Cong. XXI, 9, 69— 77.

 

Bridgewater, D., Harry, W. T., 1968, Anorthosite xenoliths
and plagioclase megacrysts in Precambrian intrusions
of south Greenland, Bull Gnénlands Geol. Unders.,
77, 243 pp. (also Meddr. Gronland,Bd.185,NO. 2).

 

 

Bryan W.B., 1966, History and mechanism of eruption of
sodarhyolite and alkali basalt, Socorro Island,
Mexico, Bull. VOlc., 29, 453—479.

 

Bulletin VOlcanologique, Special Issue, Peralkaline Rocks,
1974, 38, #3.

 

Burek, P.J., 1969, Structural effects of seafloor-spread-
ing in the Gulf of Aden and the Red Sea on the
Arabian Shield, in Hot Brine and Recent Heavy Metal
Deposits in the Red Sea, Degens, E. T., Ross, D. A.,
eds.,‘59-70.

 

 

 

Burke, K., Dewey, J. F., l973,P1ume-generated triple
junctions: key indicators in applying plate tectonics
to old rocks, J. Geol.,_81, 406n433.

Butakova, E. L., 1974, Regional distribution and tectonic
relations of the alkaline rocks of Siberia, in The
Alkaline Rocks, SOrensen, H., ed., Wiley—Interscience,
172-189.

 

259

Cherkis, N.Z., Fleming, H.S., Massingill, J.V., 1973, Is the
Gibbs fracture zone a westward projection of the
Hercynian front into North America?,Nature Phys. Sci.,
245, 113-115.

 

Clague, D.A., 1978, The oceanic basalt-trachyte association:
an explanation of the Daly Gap, J. Geol., 86, 739—743.

Clifford, T.N., 1970, The structural framework of Africa, in
African Magmatism and Tectonics, Clifford, T.N., Gass,
I.G., eds., Hafner, 1—26.

Coombs, D.S., 1963, Trends and affinities of basaltic magmas
and pyroxenes as illustrated on the diopside-olivine-
silica diagram, Min. Soc. Am. Spec. Pap., 1, 227—50.

 

Coombs, D.S., Wilkinson, J.F.G., 1969, Lineages and fraction-
ation trends in undersaturated volcanic rocks from the
East Otago volcanic province (New Zealand) and related
rocks, J. Petrol., 10, 440-501.

 

Coleman, R. G., 1974, Geologic background of the Red Sea,
Initial Report DSDP, XXIII, 813- 819.

 

Cornelissen, A. K., Verwoerd, W. J., 1975, The bushmanland
kimberlites and related rocks, Phys. Chem. Earth 9,
71- 80.

 

Crough, S.T., 1978, Thermal origin of mid-plate hot spot
swells, Geophys. J.R. astr. Soc., 55,,451-469.

 

Currie, K.L., 1970, An hypothesis on the origin of alkaline
rocks suggested by the tectonic setting of the
Monteregian Hills, Can. Min., 10, 411-420. ‘

 

Darbyshire, D.P.F., Fletcher, C.J.N., 1979, A Mesozoic alka-
line province in eastern Bolivia, Geology, 7, 545-548.

deGruyter, P., VOgel, T .A., 1981, A model for the origin
of the alkaline complexes of Egypt, Nature, 291,
571- 574.

Delong, S.E., Hodges, F.N., Arculus, R.J., 1975, Ultramafic
and mafic inclusions, Kanaga Island, Alaska, and the
occurrence of alkaline rocks in island arcs, J. Geol.,,
83, 721-736.

De Paepe, P., Gijbels, R., Hertogen, J., 1971, Rare earths
and some other trace elements in an igneous rock suite
from Fuerteventura (Canary Islands), Bull. Soc. Belg_
Ge'ol. Paleont., Hydrol., 80, 31- 50. ‘77

 

260

Dewey, J.E., Pitman, W.C., III, Ryan, W.B.F., Bonin, J.,
1973, Plate tectonics and the evolution of the Alpine
system, Geol. Soc. Am. Bull., 84, 3137-3180.

 

Dewey, J.E., Sengor, A.M.C., 1979, Aegean and surrounding
regions: complex multiplate and continuous tectonics
in a convergent zone, Geol. Soc. Am. Bull., 90, 84-92.

 

Dietz, R.S., Holden, J.C., 1970, Reconstructions of Pangea:
break-up and dispersion of continents, Permian to
recent, J. GeOphys. Res., 75, 4939-4956.

 

Dingle, R.V., Scrutton, R.A., 1974, Continental break-up and
the development of post-Paleozoic sedimentary basins
around southern Africa, Geol. Soc. Am. Bull., 85,
1467-1474.

 

Dixon, T.H., 1979, The evolution of continental crust in the
Late Precambrian Egyptian Shield, Ph.D. dissertation,
Univ. of California, San Diego, 231 p.

Dixon, T.H., 1981, Age and chemical characteristics of some
pre-Pan-African rocks in the Egyptian Shield,
Precambrian Res., 14, 119-133.

 

Doig, R., Barton, J.M., 1968, Ages of carbonatites and other
alkaline rocks in Quebec, Can. J. Earth Sci., 5,
1401-1407.

 

Dostal, J., Caby, R., Dupuy, C., 1979, Metamorphosed alka-
line intrusions and dyke complexes within the Pan-
African Belt of Western Hoggar (Algeria): geology and
geochemistry, Precambrian Res., 10, l-20.

 

Drury, N.J., 1978, Partial melt in the asthenosphere:
evidence from electrical conductivity data, Phys.
Earth Planet. Inter., l7, P16-P20.

 

Duffield, J., Gilmore, G.R., 1979, An Optimum method for the
determination of rare earth elements by neutron activa-
tion analysis, J. Radioanal. Chem., 48, 139-145.

 

Dunlop, H.M., Fitton, J.G., 1979, A K-Ar and Sr-isotOpic
study of the volcanic rocks of the Island of Principe,
West Africa - Evidence for mantle heterogeneity beneath
the Gulf of Guinea, Contrib. Mineral. Petrol., 71,
125-131.

Eby, G.N.,.l973, Scandium geochemistry of the Oka Carbona-
tite complex, Oka, Quebec, Am. Min., 58, 819—825.

 

261

Eby, G.N., 1975, Abundance and distribution of the REE and Y
in the rocks and minerals of the Oka Carbonatite com-
plex, Quebec, Geochim. Cosmochim. Acta., 39, 597-620.

 

Eby, G.N., 1979, Mount Johnson, Quebec - an example of
silicate-liquid immiscibility?, Geology, 7, 491-494.

Edgar, A.D., 1974, Experimental studies, in The Alkaline
Rocks, Sorensen, H., ed., Wiley-Interscience, 355-389.

 

Eggler, D.H., 1974, Does C02 cause partial melting in the
low-velocity layer of the mantle?, Papers from the
GeOphysical Lab., Carnegie Inst., No. 1684, Geology,
2, 69-72.

El Ramly, M.F., Dereniuk, N.E., Budnov, V.I., Armanious,
L.K., Hayek, G.G., 1969a, A petrological study on the
central part of Gabal Abu Khrug ring complex, Geol.
Survey Egypt Paper, no. 51, 53. p.

 

E1 Ramly, M.F., Armanious, L.K., Budanov, V.I., Dereniuk,
N.E., 1969b, The three ring complexes of Gabal E1
Kahfa, Gabal Nigrub El Fogani, and Gabal El Naga,
Geol. Survey Egypt Paper, no. 52, 39 p.

 

El Ramly, M.F., Budanov, V.I., Hussein, A.A.A., 1971, The
alkaline rocks of south-eastern Egypt, Geol. Survey
Egypt Paper, no. 53, 111 p.

 

 

Elsasser, W.M., Olson, P., Marsh, B.D., 1979, The depth of
mantle convection, J. GeOphys. Res., 84, 147-155.

 

El Shazly, E.M., 1977, The geology of the Egyptian region,
in The Ocean Basins and Margins: The Eastern
Mediterranean, 4A, Nairn, A.E.M., Kanes, W.H., Stehli,
E.C., eds., Plenum Press, 379-444.

Engel, A.E.J., Dixon, T.H., Stern, R., 1979, Age and Evolu-
tion of the Afro-Arabian crust, Abstract in Proceeding
of the 5th Conference on African Geology, Cairo, Egypt,
35-36.

Engel, A.E.J., Dixon, T.H., Stern, R.J., 1980, Late
Precambrian evolution of Afro-Arabian crust from ocean
arc to craton, Geol. Soc. Am. Bull., Part I, 91,
699-706.

Engell,J}, 1973, A closed system crystal fractionation model
for the agpaitic Ilimaussag intrusion, S. Greenland,
with special reference to the lujavrites, Geol. Soc.
Denmark, 22, 334-362.

262

Ernst, W.G., 1968, Amphiboles,g£ystal Chemistry Phase
Relations and Occurrence, Springer-Verlag, N.y.,
125 p.

 

Falvey, D.A., 1974, The development of continental margins
in plate tectonic theory, Austr. Petrol. Exp. Assoc.
Jouro' 95-1060

 

Feigenson, M.D., Spera, F.J., 1980, Melt production by
viscous dissipation: role of heat advection by magma
transport, Geophys, Res. Lett., 7, 145-148.

 

Feigenson, M.D., Spera, F.J., 1981, Dynamical model for
temporal variation in magma type and eruption interval
at Kohala volcano, Hawaii, GeolOgy, 9, 531-533.

Ferguson, A.K., 1978a, The crystallization of pyroxenes and
amphiboles in some alkaline rocks and the presence of a
pyroxene compositional gap, Contrib.'Minera1. Petrol.,
67, 11-15.

 

Ferguson, A.K., 1978b, The occurrence of ramsayite, titan-
lavenite and a fluorine—rich eucolite in a nepheline—
syenite inclusion from Tenerife, Canary Islands,
Contrib. Mineral. Petrol., 66, 15-20.

 

Fernandez, L.A., 1980, Geology and petrology of the Nordeste
volcanic complex, Sao Miguel, Azores: Summary, Geol.
SOC. Am. Bull., 91' 675-6800

 

Fitch, T.J., 1972, Plate convergence, transcurrent faults,
and internal deformation adjacent to SE Asia and the
W. Pacific, J. Geophys. Res., 77, 4432-4462.

 

Fitton, J.G., Hughes, D.J., 1977, Petrochemistry of the
volcanic rocks of the island of Principe, Gulf of
Guinea, COntrib. Mineral. Petrol., 64, 257-272.

 

Flanagan, F.J., 1973, 1972 values for international geo-
chemical reference standards, Geochim. Cosmochim. Acta,
37, 1189-1200.

 

Fleck, R.J., Coleman, R.G., Cornwall, H.R., Greenwood, W.R.,
Hadley, D.G., Schmidt, D.L., Prinz, w.c., Ratte', J.C.,
1976, Geochronology of the Arabian Shield, western
Saudi Arabia: K-Ar results, Geolé‘Soc.'Am. Bull., 87,
9-21. -

Fletcher, J.P., Sbar, M.L., Sykes, L.R., 1974, Seismic zones
and travel time anomalies in eastern NOrth America
related to fracture zones active in the early opening
of the Atlantic, Am. Geophys. Union Trans., 55, 447.

263

Flynn, R.T., Burnham, C.W., 1978, An experimental determina-
tion of rare earth partition coefficients between a
chloride containing vapor phase and silicate melts,
Geochim. Cosmochim. Acta, 42, 685-701.

 

Fodor, R.V., Husler, J.W., 1977, Petrology of volcanic rocks
from an aseismic rise: implications for the origin of
the Rio Grande rise, S. Atlantic Ocean, Earth Planet.
Sci. Lett., 35, 225-233.

 

 

Foland, K.A., Faul, H., 1977, Ages of the White Mountain
intrusives - New Hampshire, Vermont, and Maine, U.S.A.,
Am. J. Sci., 277, 888—904.

 

Foland, K.A., Loiselle, M.C., 1981, Oliverian syenites of
the Pliny region, northern New Hampshire, Geol. Soc.
Amer. Bull., 92, 179-188.

 

 

Freestone, I.C., 1978, Liquid immiscibility in alkali-rich
magmas, Chem. Geology, 23, 115-123.

 

Frey, P.A., Green, D.H., 1974, The mineralogy, geochemistry,
and origin of lherzolite inclusions in Victorian
basanites, Geochim. Cosmochim. Acta, 38, 1023—1059.

 

Frey, F.A., Green, D.H., Roy, S.D., 1978, Integrated models
of basalt petrogenesis: a study of quartz tholeiites
to olivine melilitites from southeastern Australia
utilizing geochemical and experimental petrological
data, J. Petrol., 19, 463—513.

 

Froidevaux, C., Souriau, M., 1978, Thermo-mechanical models
of lithosphere and asthemosphere: can a change in plate
velocity induce magmatic activity?, Phil. Trans. R. Soc.
Lond., 288A, 387—392.

Fryer, B.J., Edgar, A.D., 1977, Significance of REE distri-
butions in coexisting minerals of peralkaline under-
saturated rocks, Contrib- Mineral Petrol., 61, 35—48.

 

deter, J.M., Gastesi, P., Cendrero, A., Ibarrola, E.,
Lopez Ruiz, J., 1968, Fuerteventura, Instituto "Lucas
Mallada", Consejo Superior de Investigaciones
Cientificas, 239 p. "”

Ganzeyeva, L.V., Ganzeyev, A.A., 1975, Distribution of the
rare earth elements in the alkalic apodiabase and
apogranite albitites of the Russian Platform, Geochem.
Int., 12, 47-52.

Garson, M.S., Krs, M., 1976, Geophysical and geological evi-
dence of the relationship of Red Sea transverse tec-
tonics to ancient fractures, Geol. Soc. Am. Bull., 87,
169-181.

 

l

264

Garson, M.S., Shalaby, I.M., 1975, Precambrian—Lower
Paleozoic plate tectonics and metallogenesis in the
Red Sea region, _ymp3 on Metallogeny and Plate'
Tectonics: Geol. Assoc. CanadaZMineral Assoc. Canada
Mtg. 1974, St. JohnTs, Newfoundland.

 

 

 

Gass, I.G., 1977, The evolution of the Pan-African crystal-
line basement in NE Africa and Arabia, J. Geol. Soc.
10nd. ' 134’ 129-1380

 

Gass, I.G., Chapman, D.S., Pollack, N., Thorpe, R.S., 1978,
Geological and geophysical parameters of mid-plate
volcanism, Phil. Trans. R. Soc. Lond., 288-A, 581-597.

 

Gass, I.G., Mallick, D.I.J., Cox, R.G., 1973, Volcanic
islands of the Red Sea, J. Geol. Soc. Lond., 129,
275-310.

 

Gast, P.W., 1968, Trace element fractionation and the origin
of tholeiitic and alkaline magma types, Geochim.
Cosmochim. Acta, 32, 1057—1086.

 

 

Gastesi, P., 1969, Petrology of the ultramafic and basic
rocks of Betancuria Massif, Fuerteventura Island
(Canarian Archipelago), Bull. VOlcan., 38, 1008-1037.

 

Gittins, J., 1979, The feldspathoidal alkaline rocks, The
Evolution of the Igneous Rocks, Yoder, H. 8., ed.,*
Pr1nceton Univ. Press, 351— 390.

 

Glennie, K.W., Boeuf, M.G.A., Hughes Clark, M.W., Moody-
Stuart, M., Pilaar, W.F.H., Reinhardt, B.M., 1973,.
Late Cretaceous nappes in Oman Mountains and their
geologic evolution, Am. Assoc. Petroleum Geologists
Bull., 57/1, 5- 27.

 

Goldich, S.S., Treves, S.E., Suhr, U.H., Stuckless, J.S..
1975, Geochemistry of the Cenozoic Volcanic rocks of
Ross Island and vicinity, Antartica, J. Geology, 83,
415-435.

Gordon, G.E., Randle, Kquoles, G.G., Corliss, J.B., Beeson,
M.H., Oxley, 5.8., 1968, Instrumental activiation
analysis of standard rocks with highereSOlution gamma-
ray detectors, Geochim. Cbsmochim. Acta, 32, 369—396.

Green, D.H., 1970, A review of experimental evidence on the
origin of basaltic and nephelinitic magmas, Phys.
Earth Planet. Inter., 3, 221—235.

 

Green, D.H., Ringwood, A.B., 1967, The genesis of basaltic
magmas, Contrib. Mineral Petrol., 15, 103-190.

265

Greenberg, J.K., 1978, Geochemistry, petrology and tectonic
origin of the Egyptian Younger Granites, Ph.D. disser—
tation, Univ. of North Carolina, Chapel Hill, 134 p.

Greenwood, W.R., Hadley, D.G., Anderson, R.E., Fleck, R.J.,
Schmidt, D.L., 1976, Late Proterozoic cratonization
in southwestern Saudi Arabia, Phil. Trans. R. Soc.
Lond., 280-A, 517-527.

 

Grocott, J., 1979, Shape fabrics and superimposed simple
shear strain in a Precambrian shear belt, W. Greenland,
J. Geol. Soc. Lond., 136, 471-488.

 

Grundfest, I.J., 1963, Thermal feedback in liquid flow:
plane shear at constant stress, Trans. Soc. Rheol., 7,
195-207.

 

Hadley, D.G., Schmidt, D.L., 1978, Proterozic sedimentary
rocks and basins of the Arabian shield and their
evolution, Precambrian Res., 5, A23-A24.

 

Hahn, G.A., Raynolds, G.H., WOod, R.A., 1976, The geology
of the Angareb Ring Dike Complex, NW Ethiopia, Bull.
VblcanolOgique, 40, 1-10.

 

Hancock, P.L., Kadhi, A., 1978, Analysis of megasc0pic
fractures in the Dhruma—Nisah segment of the central
Arabian graben system, J. Geol. Soc. Lond., 135,,
339-347.

 

Hanson, G.N., 1978, The application of trace elements to the
petrogenesis of igneous rocks of granitic composition,
Earth Planet. Sci. Lett., 38, 26-43.

 

Hards, N., 1976, Distribution of elements between fluid phase
and silicate melt phase of granites and nepheline
syenites, Nat. Envir. Res; Council, Publ. D 6, 88-90.

Hargraves, R.B., Onstott, T.C., 1980, Paleomagnetic results
from some southern African Kimberlites, and their
tectonic significance,'J.‘Geophys. Res., 85, 3587—3596.

Harker, A., 1896, The natural history of igneous rocks. I.
Their geographical and chronological distribution, Sci.
Progress, 6, 12-33.

 

Harker, A., 1909, The Natural-History of Igneous Rocks,
Macmillan, N.Y.

Hashad, A.H., E1 Reedy, M.W.M., 1979, Geochronology of the
anorOgenic alkalic rocks, South Eastern, Egypt,
Abstracts 5th Conference on African Geolbgy, Cairo,
Egypt, 44-45.

266

Hawkins, J.W., Natland, J.H., 1975, Nephelinites and basa-
nites of the Samoan linear volcanic chain: their
possible tectonic significance, Earth Planet! Sci.
Lett., 24, 427-439.

 

Heming, R.F., 1979, Undersaturated lavas from Ambittle
Island, Papua New Guinea, Lithos, 12, 173-186.

Herz, N., 1977, Timing of spreading in the South Atlantic:
information from Brazilian alkalic rocks, Geol. Soc.
Am. Bull., 88, 101-112.

 

 

Hildreth, W., 1977, The magma chamber of the Bishop Tuff:
Gradients in temperature, pressure, and composition,
Ph.D. thesis, Univ. California, Berkeley, 328 p.

Hildreth, W., 1979, The Bishop Tuff: Evidence for the origin
of compositional zonation in silicic magma chambers,
Geol. Soc. Am. Spec. ng,, 180, 43-75. ’

 

Hildreth, W., 1981, Gradients in silicic magma chambers:
implications for lithOSpheric magmatism, J. Geophys.
Res., 86, 10153-192.

 

Holcombe, C. J., 1977, How rigid are the lithospheric plates?
Fault and shear rotations in southeast Asia, J. Geol.
Soc. Lond., 134, 325- 342.

 

 

Hsu, R.J., 1971, Origin of the Alps and western Mediter-
ranean, Nature, 233, 44-48.

Hsu, K. J., 1977, Tectonic evolution of the Mediterranean
Basins, in The Ocean Basins and Margins: The Eastern
Mediterranean, 4A, Nairn A. E. M., Kanes, W. H., Stehli,
F. G., eds., Plenum Press, 29-75.

Hsu, R.J., Nachev, I.K., Vuchev,_V.T., l977, Geologic evolu—
tion of Bulgaria in light of plate tectonics,
Tectonophys., 40, 245-256.

 

Hurley, P.M., 1972, Can the subduction process of mountain
building be extended to Pan—African and similar oro—
genic belts?, Earth Planet. Sci. Lett.,_15, 305-314.

Hussein, A.A.A., Hassan, M.A., 1973, Further contribution
to geology and geochemistry of some Egyptian ring
complexes, Annals of the Geological Survey of Egypt,
v. III, 167:175.

Irving, A. J., 1978, A review of experimental studies of
crystal/liquid trace element partitioning. Geo chim.
Cosmochim. Acta, 42, 743-770.

 

267

Irving, T.N., 1979, Infiltration metasomatism, adcumulus
growth, and secondary differentiation in the Muskox
Intrusion, Annual Report of-the Gegphysical Lab.,
Carnegie Institution, no. 1755,4743-751.

Jackson, B.D., Shaw, H.R., 1975, Stress fields in central
portions of the Pacific plate: delineated in time by
linear volcanic chains, J. Geophys. Res., 80, 1861—
1874.

 

Jackson, E.D., Shaw, H.R., Bargar, K.E., 1975, Calculated
geochronology and stress field orientations along the
Hawaiian chain, Earth Planet. Sci. Lett., 26, 145—155.

 

Jackson, B.D., Silver, Eli A., Dalrymple, G.B., 1972,
Hawaiian - Emperor chain and its relation to Cenozoic
circum-Pacific tectonics, Geol. Soc. Am. Bull., 83,
601-617.

 

Jacoby, W.R., 1970, Instability in the upper mantle and
global plate movements, J. Geophys. Res., 75,
5671-5680.

 

Johnson, R.W., Wallace, D.A., Ellis, D.J., 1976,
Feldspathoid-bearing potassic rocks and associated
types from volcanic islands off the coast of New
Ireland Papua New Guinea: a preliminary account of
geology and petrology, in volcanism in Australia,
Johnson, R.W., ed., Elsevier, Amsterdam.

Kay, R.W., 1978, Trace elements in ocean ridge basalts,
Earth Planet. Sci. Lett., 38, 95-116.

 

Kay, R.W., Gast, P.W., 1973, The rare earth content and
origin of alkali-rich basalts, J. Geol.,_81, 653-682.

Kennedy, W.Q., 1965, The influence of basement structure on
the evolution of the coastal (Mesozoic and Tertiary)
basins of Africa: in Salt Basins Around Africa, Ion,
D.C. (ed.), Lond. Inst. Petroleum, 7416.

 

Kennedy, A.J., 1963, Process of Creep and Fatigue in Metals,
N.Y., John Wiley & Sons.

Kennedy, G.C., 1955, Some aspects of the role of water in
rock melts, Geol. Soc. Am. Spec. Pap., 62, 489—504.

King, B.C., 1965, Petrogenesis of the alkaline igneous rock
suites of the volcanic and intrusive centers of Eastern
Uganda, J. Petrol., 6, 67—100.

 

King, B.C., Chapman, G.R., 1972, Vblcanism of the Kenya rift
valley, Phil. Trans, R. Soc. Lond., 271~A,_185~208.

268

Kisvarsanyi, E.B., 1980, Granitic ring complexes and Pre-
cambrian hot spot activity in the St. Francois terrane,
Midcontinent region, United States, Geology, 8, 43-47.

Kjdde, J., Storetvedt, K.M., Roberts, D., Gidskehaug, A.,
1978, Paleomagnetic evidence for large-scale dextral
movement along the Trollfjord—Komagelvfault, Finnmark, N.
Norway, Phys. Earth-P1anet.‘Inter., 16, 132-144.

 

Koide, H., Bhattacharji, S., 1977,_A model for partial melt-
ing of upper mantle rocks by local instability and its

correlation with deep-seated fractures, Magma Genesis, OR
Dept. of Geol. and Mineral Indus., Bulletin , 3- 80.

Kogarko, L.N., 1974, Role of volatiles, in The Alkaline
Rocks, S¢rensen, H., ed., Wiley-Interscience, 474-487.

 

Kogarko, L.N., 1977, General regularities of differentiation
of magmas oversaturated with alkalis;- Geochem., Int.,
14' 9-250

 

Kogarko, L.N., Burnham, C., Shettle, D., 1978, Water regime
in alkalic magmas, Geochem. Int., 14, 1-8.

 

Kogarko, L.N., Ryabchikov, I.D., S¢rensen, H., 1974, Liquid
fractionation, in The Alkaline Rocks, Sorensen, H., ed.,
Wiley-Interscience, 488-500.

 

Konnerup-Madsen, J., Rose-Hansen, J., 1982, VOlatiles asso-
ciated with alkaline igneous rift activity: fluid
inclusions in the Ilimaussag intrusion and the Gardar
granitic complexes (South Greenland), Chem. GeolOgy,
37, 79-93.

 

Koster van Groos, A.F., Wyllie, P.J., 1969, Melting relation-
ships in the system NaAlSi303-NaCl-H20 at one kilobar
pressure with petrological applications, J. Geol., 77,
581-605.

Krbner, A., l977a, The Precambrian geotectonic evolution of
Africa: plate accretion versus plate destruction,
Precambrian Res., 4, 163-213.

 

Kroner, A., l977b, Precambrian mobile belts of southern and
eastern Africa - ancient sutures of sites of ensialic
mobility? A case for crustal evolution towards plate
tectonics, Tectonophys., 40, 101-135.

 

Kroner, A., 1979, Pan-African plate tectonics and its reper-
cussions on the crust of northeast Africa, Geol.
Rundschau, 68, 565-583.

 

Kroner, A., Roobol, M.J., Ramsay, C.R., Jackson, N.J., 1979,
Pan-African ages of some gneissic-rocks in the Saudi
Arabian Shield, J. Geol.-Soc. Lond., 136, 455-461.

 

269

Kumparapeli, P.S., 1970, Monteregian alkalic magmatism
and the St. Lawrence rift system in Space and time,
Can. Mineral., 10, 421-431.

 

Kumparapeli, P.S., Saull, V.A., 1966, The St. Lawrence
valley system: a North American equivalent of the
East African rift valley system, Can. J. Earth Sci.,
3, 639-658.

 

Kuno, H., 1968, Differentiation of basalt magmas, in Basalts:

The Poldervaart Treatise on Rocks of'Basaltic Composi-

tion, vol. 2} Hess, H.H., Poldervaart, A., edS.,

Kushiro, I., 1968, Composition of magmas formed by partial
zone melting of the Earth's upper mantle, J. Geophys.
Res., 73, 619-634.

 

 

Larson, L.M., 1976, Clinopyroxenes and coexisting mafic
minerals from the alkaline Ilimaussag intrusion, S.
Greenland, J. Petrol., 17, 258-290.

 

Larson, L.M., Steenfelt, 1974, Alkali loss and retention in
an iron-rich peralkaline phonOlite dyke from the Gardar
province, 8. Greenland, Lithos, 7, 81-90.

Laubscher, H., Bernoulli, D., 1977, Mediterranean and Tethys
in The Ocean Basins and Margins: The Eastern Medi-
terranean, 4A, Nairn, A.E.M.,Kanes, W.H., Stehli, F.G.,
eds.,gPlenum, N.Y., 1-28.

 

 

Laurent, R., Pierson, T.C., 1973, Petrology of alkaline rocks
from Cuttingsville and the Shelburne Peninsula, Vermont,
Can. J. Earth Sci., 10, 1244-1256.

 

Le Bas, M.J., 1971, Peralkaline volcanism, crustal swelling,
and rifting, Nature Phys. Sci., 230, 85-87.

 

LeMaitre, R.W., 1962, PetrolOgy of volcanic rocks, Gough
Island,South Atlantic, Geol. Soc. Am. Bull., 73,
1309-1340.

Le Pichon, K., Francheteau, J., 1978, A plate-tectonic
analysis of the Red Sea - Gulf of Aden, Tectono-
physics, 46, 369-406.

Lippard, S.J., 1973, The petrology of phonolites from the
Kenya rift, Lithos, 6, 217-234.

Ludden, J.N., 1977, The mineral chemistry and origin of
xenoliths from the lavas of Anjouan, Comores.
Archipelago, Western Indian Ocean, Contrib. Mineral
'Petrol., 64, 91-107.

d

270

Ludden,J.N., 1978, Magmatic evolution of the basaltic
shield volcanoes of Reunion Island, J. Vblc. Geotherm.
Res., 4, 171-198. '

 

Luth, W.C., 1976, Experimental petrology: igneous rocks, in
The Evolutign of the Crystalline Rocks, Bailey, D.K.,
MacDonald, R., eds., Academic Press, N.Y., 333-417.

 

Luth, W.C., Tuttle, O.F., 1963, Eternally heated cold-seal
pressure vessels for use to 10,000 bars and 750°C,
American Mineralogist, 11, 1401-1409.

 

Lutz, T.M., 1979, Strontium and oxygen isotope relations and
geochemistry of the Abu Khrug Complex, Egypt: implica-
tions for petrogenesis of the alkaline rocks of the
Eastern Desert, Egypt, Unpublished Ph.D. dissertation,
Univ. of Pa.

Maalde, S., Aoki, K., 1977, The major element composition of
the upper mantle from the composition of lherzolites,
Contrib. Mineral. Petrol., 63, 161-173.

 

MacDonald, R., 1974, The role of fractional crystallization
in the formation of the alkaline rocks, in The Alkaline
Rocks, Sdrensen, H., ed., Wiley-Intersciences,i442-459.

 

MacDonald, R., Bailey, D.K., Sutherland, D.S., 1970, Over-
saturated peralkaline glassy trachytes from Kenya,
J. Petrol., 11, 507-517.

 

MacDonald, G.A., Katsura, T., 1964, Chemical composition of
Hawaiian lavas, J. Petrol,, 5, 82-133.

 

MacFarlane, D.J., Ridley, W.I., 1969, An interpretation of
gravity data for Lanzarote, Canary Islands, Earth
Planet. Sci. Lett., 6, 431-436.

 

MacIntyre, R.M., 1977, Anorogenic magmatism, plate motion
and Atlantic evolution, J. Geol. Soc. Lond., 133,
375-384. '

MacKenzie, W.S., 1960, Review of some contributions of exper-
imental studies to petrology, Geol. J.,_2, 369-386.

 

Malfait, E.T., Dinkelman, M.G., 1972, Circum-Caribbean
tectonic and igneous activity and the evolution of
the Caribbean Plate, Geol. Soc. Am. Bull., 83, 251-272.

Marsh, B.D., Carmichael, I.S.E., 1974, Benioff zone magma-
tism, J. Geophys, Res.,_79, 1196-1206.

 

Marsh, J.S., 1973, Relationships between transform directions
and alkaline igneous rock lineaments in Africa and South
America, Earth P1anet.-Sci.~Lett., 18, 317-323.

271

Martin R.F., 1974, Role of water in Panellerite genesis,
Bulletin VblcanOIOgique, 38, #3 (Special Issue -
Peralkaline Rocks), 666-679.

 

Martin, R.F., Piwinskii, A.J., 1972, Magmatism and tectonic
setting, J. Geophys. Res., 77, 4966-4975.

 

Martin, R.F., Whitley, J.E., WOoley, A.R., 1978, An investi-
gation of REE mobility: fenitized quartzites, Borrolan
complex, NW Scotland, Contrib. Mineral Petrol., 66,
69-73. ' ' ‘

 

Marzouki, F., Fyfe, W.S., l977, Pan-African plates: addi-
tional evidence from igneous events in Saudi Arabia,
Contrib. Mineral. Petrol., 60, 219-224.

 

McBirney, A.R., Aoki, K., 1968, Petrology of the island of
Tahiti, Geol. Soc. Am. Mem., 116, 523-556. ’

 

McClennen, B.M., Fryer, B.J., Young, G.M., 1979, REE in
Huronian (Lower Proterozoic) sedimentary rocks:
composition and evolution of the post-Kenoran upper
crust, Geochim. Cosmochim. Acta, 43, 375-388.

 

McHone, J.G., 1981, Comment and Reply on Mesozoic hot.
spot epeirogeny in eastern North America, Geology,
9, 341-342.

Melosh, H.J., 1976, Plate motion and thermal instability in
the asthenosphere, Tectonophys., 35, 363-390.

 

Melosh, H.J., 1977, Shear stress on the base of a litho-
spheric plate, Pure Appl. Geophys., 115, 429-439.

 

Melosh, H.J., Ebel, J., 1979, A simple model for thermal
instability in the asthenosphere, Geophys. J.R.
AStro SOC. " 59’ 419—4360

 

Miller, C.F., 1977, Early alkalic plutonism in the calc-
alkalic batholith belt of California, Geolo , 5,
685-688.

Miller, C.F., Mittlefehldt, D.W., 1982, Depletion of light
rare-earth elements in felsicmagmas, GeOIOgy, 10,
129-133. '

Mineyev, D.A., 1963, Geochemical differentiation of the

 

Minster, J.B., Jordan, T.H., 1978, Present-day plate
motions, J. Geophys, Res., 83, 5331-5354.

 

272

Minster, J. F., Allegre, C. J., 1978, Systematic use of trace
elements in igneous processes, Part III, Inverse
problem of batch melting in volcanic suites. Contrib.
Mineral Petrol., 68, 37-52.

 

Minster, J.E., Minster, J.B., Allegre, C.J., Treuil, M.,
1977, Systematic use of trace elements in igneous
processes, Part II. Inverse problem of the fractional
crystallization process in volcanic suites, Contrib.
Mineral. Petrol., 61, 49- 77.

 

 

Mitchell, A.H.G., McKerrow, W.S., 1975, Analogous evolution
of the Burma orogen and the Scottish Caledonides,
Geol. Soc. Am. Bull., 86, 305-315.

 

Mitchell, R.H., Brunfelt, A.O., 1975, REE geochemistry of
the Pen Alkaline complex, Norway, Contrib. Mineral.
Petrol., 52, 247-259.

 

Miyashiro, A., 1978, Nature of alkalic rock series, Contrib.
Mineral. Petrol., 66, 91-104.

 

Molnar, P., Sykes, L.R., 1969, Tectonics of the Caribbean
and Middle America regions from focal mechanisms and
seismicity, Geol. SOC. Am. Bull., 80, 1639-1684.

 

Molnar, P., Tapponnier, P., 1975, Cenozoic tectonics of
Asia: effects of a continental collision, Science,
189, 419-426.

Moore, J. Mc., 1979, Tectonics of the Najd transcurrent
fault.system, Saudi Arabia, J. Geol. Soc. Lond., 136,
441-454.

 

Morel, P., Irving, E., 1978, Tentative paleocontinental maps
for the Early Phanerozoic and Proterozoic, J. Geol.,
86, 535- 561.

 

Morgan, W.J., 1971, Convection plumes in the lower mantle,_
Nature, 230, 42-43.

Morgan, W.J., 1979, HotSpot tracks and the opening of the
Atlantic and Indian Oceans, in The Sea, vol. 7,
Emiliani, C., ed., Interscience, N. Y., 443-487.

Mbrse, S. A., 1969, Syenites, Ann. Rept. Dir. Geophys. Lab.,
Washington, 67, 112- 120.

 

Morse, S. A., 1970, Alkali feldspars with water at 5 kb
pressure, J. Petrol., 11, 221- 251.

 

Muir, I.D., Smith, J.V.,'l956, Crystallization of feldspars
in larvikite, Zeit.‘Krist., 107, 182-195.

 

273

Munoz, M., 1969, Ring complexes of Pajara in Fuerteventura,
Bull. VOlcanol., 33, 840-861.

 

Mustart, D.A., 1972, Phase relations in the paralkaline
portion of the system NaZO-A1203-Si02—H20, Ph.D.
dissertation, Stanford Univ.

Mysen, B.O., 1976, The role of volatiles in silicate melts:
solubility of carbon dioxide and water in feldspar,
pyroxene, and feldspathoid melts to 30 Kb and 16250C,
Am. J. Sci., 276, 969-996.

 

Mysen, B.O., l977a, The solubility of H20 and C02 under
predicted magma genesis conditions and some petro-
logical and geophysical implications, Rev. GeOphys.
Space Phys., 15, 351-361.

 

 

Mysen, B.O., 1977b, Solubility of volatiles in silicate
melts under the pressure and temperature conditions
of partial melting in the upper mantle, in Ma ma
Genesis, Oregon Dept. Geol. Mineral. Ind. Bu 1., 96,
I-14.

Mysen, B.O., Arculus, R.J., Eggler, D.H., 1975, Solubility
of carbon dioxide in melts of natural nephelinite,
tholeiite and andesite composition, Contrib. Mineral
Petrol., 53, 227- 239.

 

Mysen, B.O., Holloway, J.R., 1977, Experimental determina-
tion of rare earth fractionation patterns in partial
melts from peridotite in the upper mantle, Earth
Planet. Sci. Lett., 34, 231-237.

 

Nance, W.B., Taylor, S.R., l977, REE patterns and crustal
evolution - II. Archean sedimentary rocks from
Kalgoorlie, Australia, Geochim. Cosmochim. Acta, 41,
225-31.

Naseef, A.O., Gass, I.G., l977, Granitic and metamorphic
rocks of the Taif area, western Saudi Arabia, Geol. Soc.
Am. Bull., 88, 1721-1730.

Neary, C. R., Gass, I. G., Cavanagh, B. J., 1976, Granitic
association of northeastern Sudan, Geol. Soc. Am. Bull.,
87, 1501- 1512.

Neill, W.M., 1973, Possible continental rifting in Brazil
and Angola related to the opening of the South Atlantic,
Nature 'Phys. Sci., 245, 104-107.

 

Newton, R. C., Sharp, w. C., 1975, Stability of forsterite +
C02 and its bearing on the role of C02 in the mantle,
Earth Planet. Sci. Lett., 26, 239- 244.

274

Nur, A., 1971, Viscous phase in rocks and the LVZ,'g;
Ge0phys. Res., 76, 1270-1277.

 

Nur, A., Ben-Avraham, 2., 1978, The eastern Mediterranean
and Levant: tectonics of continental collision,
Tectongpgys., 46, 297-311.

 

O'Driscoll, C.F., Strong, D.F., 1979, Geology and geo-
chemistry of L. Precambrian volcanics and intrusive
rocks of SW Avalon Zone, Newfoundland, Precambrian Res.,
8, 19-48.

 

Oftedahl, C., 1948, Studies on the igneous rock complex of
the Oslo region 9: The feldspars, Skr. Norske
Vidensk Akad., Mat.—naturv., K1, 3.

 

 

Orville, P.M., 1963, Alkali ion exchange between vapor and
feldspar phases, Am. J. Sci., 261, 201-237.

 

Orville, P.M., 1972, Plagioclase cation exchange equilibrium
with aqueous chloride solution: results at 7000C and
2000 bars in the presence of quartz, Am. J. Sci., 272,
234—272.

 

Opdyke, N.D., MacDonald, W.D., 1973, Paleomagnetism of the
Late Cretaceous Pocos de Caldas alkaline complex,
southern Brazil, Earth Planet. Sci.-Lett., 18, 37-44.

 

Padovani, E.R., Carter, J.L., l977, Nonequilibrium fusion due
to decompression and thermal effects in crustal xeno-
liths, Magma Genesis, State of Oregon Dept. Geol. and
Mineral Indus., Bulletin 96, 43-58.

 

Pankhurst, R.J., Beckinsale, R.D., Brooks, C.R., 1976,
Strontium and oxygen isotope evidence relating to the
petrogenesis of the Kangerdlugssuag alkaline intrusion,,
East Greenland, Contrib. Mineral. Petrol., 54, 17-42.

Payne, J.G., Shaw, D.M., 1967, K-Rb relations in the Blue
Mountain nepheline syenite, Earth Planet. Sci. Lett.,,
2, 290-292.

Pearce, J.A., Gorman, B.E., Birkett, T.C., 1977, The
relationship between major element chemistry and
tectonic environment of basic and intermediate volcanic
rocks, Earth-Planet. Sci. Lett., 36, 121-132.

Petro, W.L., ngel, T.A., , Wilband, J.T., 1979, Major-
element chemistry of plutonic rock suites from compres~
sional and extensional plate boundaries,_Chem. Geology,
26, 217-235.

 

275

Phillips, W.E.A., Stillman, C.J., Murphy, T., 1976, A
Caledonian plate tectonic model, J. Geol. Soc. Lond.,
132, 579-609.

 

Philpotts, A.R., 1971, Immiscibility between feldspathic
and gabbroic magmas, Nature, 229, 107-109.

Philpotts, A.R., 1972, Density, surface tension, and vis—
cosity of the immiscible phase in a basic. alkaline
magma, Lithos, 5, 1—18.

Philpotts, A.R., 1974, The Monteregian Province, in The
Alkaline Rocks, Sdrensen, H., ed., Wiley—Intersc1ence,
293-310.

 

Philpotts, A.R., 1976, Silicate liquid immiscibility: its
probable extent and petrogenetic significance, Am. J.
Sci., 276, 1147—1177.

Phil. Trans. R. SOC. Lond., 1978, 288-A, 383-646.

 

Phys. Earth Planet. Inter., 1978, Sp. Issue, 17, no. 1.

 

Pilger, A., Rosler, A., 1976, Temporal relationships in the
tectonic evolution of Afar Depression (Ethiopia) and
the adjacent Afro-Arabian rift system, in Proceedin s
of an Intl.-Symposium on the Afar Region and Relateg
Rift PrOblems, Bad Bergzabern, Afar between continental
and oceanic rifting, v. II, Pilger, A., Rosler, A.
(éds.),'l-10.

 

 

 

Pilger, R.H., Handschumacher, D.W., 1981, The fixed hotSpot
hypothesis and origin of the Easter-Sala y Gomez-Nazca
trace, Geol. Soc. Am. Bull., 92, 437—446.

 

Pinar-Erdem, N., Ilhan, E., 1977, Outlines of the stra-
tigraphy and tectonics of Turkey: with notes on the
geology of Cyprus, in The Ocean Basins and Mar ins:

The Eastern Mediterranean, v. 4A, Nairn, A.E.M., Kanes,
W.H., Stéhli,F.G., eds., Plenum Press, 379-444.

 

 

Pitman, W.C. III, Talwani, M., 1972, Seafloor spreading in
the North Atlantic, Geol.'Soc. Am. Bull., 83, 619—643.

 

Pollack, H.N., Gass, I.G., Thorpe, R.S., Chapman, D.S.,
1981, On the vulnerability of Lithospheric plates to
Mid-plate VOlcanism: Reply to comments by P.R. Vogt,
J. Geophys. Res., 86, 961—966.

 

Price, R.C., Taylor, S.R., 1973, The geochemistry of the
Dunedin vo1cano, East Otago, New Zealand: REE,
Contrib. Mineral. Petrol., 40, 195—205.

 

 

276

Proceeding of the 5th Conference on African Geology,
Cairo, Egypt, 1979.

Rabinowitz, P.D., 1976, Geophysical study of the con-
tinental margin of southern Africa, Geol. Soc. Amer.
Bull., 87, 1643-1653.

Reid, J.B., Frey, P.A., 1971, REE distribution in lherzolite
and garnet pyroxenite xenoliths and the constitution
of the upper mantle, J. Geophys. Res., 76,1184-1196.

Ressetar, R., Nairn, A.E.M., Monrad, J.R., 1981, Two phases
of Cretaceous - Tertiary magmatism in the Eastern
Desert of Egypt: Paleomagnetic, chemical and K-Ar
evidence, Tectonophys., 73, 169-193.

 

Richardson, E.S., Harrison, C.G.A., 1976, Opening of the
Red Sea with two poles of rotation, Earth Planet. Sci.
Lett., 30, 135-142.

 

Robin, C., Tournon, J., 1978, Spatial relations of andesitic
and alkaline provinces in Mexico and Central America,
Can. J. Earth Sci., 15, 1633-1641.

 

Rock, N.M.S., 1976, Fenitization around the Monchique alka-
line complex, Portugal, Lithos, 9, 263-279.

Rock, N.M.S., 1978, Petroloqy and petrogenesis of the
Monchique alkaline complex, Southern Portugal, g;

Rock, N.M.S., Scoon, J.B., 1976, Petrogenetic significance
of some new xenolithic alkaline rocks from East Africa,
Mineral. Mag:, 40, 611-624.

 

Roedder, E., Coombs, D.S., 1967, Immiscibility in granitic
melts, indicated by fluid inclusions in ejected
granitic blocks from Ascension Island, J. Petrol., 8,
417-451.

Rogers, J.J.W., Ghuma, M.A., Nagy, R.M., Greenberg, J.R.,
Fullager, P.D., 1978, Plutonism in Pan-African belts
and the geological evolution of northeastern Africa,
Earth Planet. Sci. Lett., 39, 109-117.

 

Ryerson, F.J., Hess, P.C., 1978, Implications of liquid-
liquid distribution coefficients to mineral-liquid
partitioning, Geochim. Cosmochim. Acta., 42, 921-932.

Said, 4., 1962, The Geology of Egypt, Elsevier, 377 p.

Saggerson, B.P., 1970, The structural control and genesis of
alkaline rocks in Central Kenya, Bull. volcano, 34,
38-76.

277

Schairer, J.E., Bowen, N.L., 1935, Preliminary report on
equilibrium relations between feldspars, and silica,
Trans. Am. Geophys. Un., 16th annual meeting, 325—328.

Schilling, J.G., Winchester, J.W., 1969, Rare earth contribu-
tion to the origin of Hawaiian lavas, Contrib. Mineral.
PEtIOle' 23' 22-37.

Schmincke, R.V., 1973, Magmatic evolution and tectonic
regime in the Canary, Madeira, and Azores Island
groups, Geol. Soc. Am. Bull., 84, 633-648.

Schock, B.H., 1979, Distribution of REE and other trace
elements in magnetites, Chem. Geology, 26, 119-134.

 

Schubert, G., Froidevaux, C., Yuen, D.A., 1976, Oceanic
lithosphere and asthenosphere: thermal and mechanical
structure, J. Geophys. Res., 81, 3525-3540.

 

Schubert, G., Yuen, D.A., 1978, Shear heating instability
in the Earth's upper mantle, Tectonophys., 50,,
197-205.

Schwarzer, R.R., Rogers, J.J.W., 1974, A worldwide compari-
son of alkali olivine basalts and their differentiation
trends, Earth Planet. Sci. Lett., 23, 286-296.

 

 

Sengor, A.M.C., 1979, Mid-Mesozoic closure of Permo-Triassic
Tethys and its implications, Nature, 279, 590-593.

Sengor, A.M.C., Yilmaz, Y., Ketin, I.,_1980, Remnants of a
pre-Late Jurassic ocean in northern Turkey: Fragments
of Permian-Triassic Paleo-Tethys?, Geol. Soc. Am. Bull.,
91, 599-609.

Sengor, A.M.C., Yilmaz, Y., Ketin, I., 1982, Reply to
Discussion by Bergougnan and Fourquin, Geol. Soc. Am.
Bull., 93' 932-936. v—fi

Serencsits, C. McC., Paul, H., Foland, K.A., Hussein, A.A.,
Lutz, T.M., 1981, Alkaline ring complexes in Egypt:
their ages and relationship in time, J. Geophys. Res.,
86, 3009-3013. :7'

Shackleton, R.M., 1973, Correlation of structures across
Precambrian orogenic belts in Africa, in Implications
of Continental Drift to the Earth Sciences, Tarling,
*D.H., Runcorn, S:fi., eds., vol. 2, 109141095.

 

Shankland, T.J., 1977, Partial melt in regions of high elec-
trical conductivity, in Mggma Genesis, Oregon Dept of
Geol. and Mineral Indus., ulletin 96, 215-218.

 

 

278

Shaw, D.M., Dostal, J., Keays, R.R., 1976, Additional esti-
mates of continental surface Precambrian Shield
composition in Canada, Geochim. Cosmochim. Acta, 40,
73-83.

Shaw, H.R., 1963, Obsidian-H20 viscosities at 1000 and
2000 bars in the temperature range 7000-9000C, J;
Geophys. Res., 68, 6337-43.

 

Shaw, H.R., 1965, Comments on viscosity, crystal settling
and convection in granitic magmas, Am. J. Sci.,
263, 120-152.

Shaw, H.R., 1969, Rheology of basalt in the melting range,
J. Petrol., 10, 510-535.

 

Shaw, H.R., 1973, Mantle convection and volcanic periodicity
in the Pacific; evidence from Hawaii, Geol. Soc. Am.
Bull., 84, 1505-1526.

 

Shaw, H.R., 1974, Diffusion of BzO in granitic liquids:
Part I., Experimental data; Part II, Mass transfer in
magma chambers in Geochemigal Transport and Kinetics,
Hoffman, A.W., Giletti, B.J., Yoder, H.S., Yund, §.A.,
eds., Carnegie Institution of Washington Publ., 634,
139-170.

Shaw, H.R., 1980, The fracture mechanisms of magma transport
from the mantle to the surface, in Physics of Magmatic
Processes, Hargraves, R.B., ed., Princeton univerSIty
Press, Princeton, N.J., 201-232.

Shaw, H.R., Jackson, E.D., 1973, Linear island chains in the
Pacific: result of thermal plumes or gravitational
anchors?, J. Geophys. Res., 78, 8634-8652.

Shaw, H.R., Jackson, E.D., Bargar, R.E., 1980, Volcanic
periodicity along the Hawaiian-Emperor chain, in
Jackson volume, Irving, A.J. et. al., eds., Am. J.
Sci., 280-A, Part 2, 667-708.

Shaw, R.R., Smith, R.L., Hildreth, W., 1976, Thermogravita-
tional mechanisms for chemical variations in zoned
magma chambers, Geol. Soc. Am. Abs. Progi, v. 8, 1102.

Shibata, T., Thompson, G., Frey, P.A., 1979, Tholeitic and
alkali basalts from the.Mid-Atlantic ridge at 43°N,
Contrib. Mineral. Petrol.,-70, 127-141.

279

Shimron, A.B., 1975, PetrOgenesis of the Tarr albitite -
carbonatite complex, Sinai Peninsula, Min. Mag., 40,
13-24 0

Shimuzu, N., Arculus, R.S., 1975, REE concentrations in a
suite of basanitoids and alkali olivine basalts, from
Grenada, Lesser Antilles, COntrib. Minera1.'Petrol.,
50, 231-240.

 

Sigurdsson, H., Tomblin, J.E., Brown, G.M., Holland, J.G.,
Arculus,_R.J., 1973, Strongly undersaturated magmas
in the Lesser Antilles island arc, Earth Planet. Sci.
Lett., 18, 285-295. 7'7' ‘7‘

 

Smith, A.G., 1971, Alpine deformation and the oceanic areas
of the Tethys, Mediterranean and Atlantic, Geol. Soc.
Am. Bull., 82, 2039—2070.

 

 

Smith, I.E.M., Chappell, B.W., ward, G.R., Freeman, R.S.,
1977, Peralkaline rhyolites associated with andesitic
arcs of the SW Pacific, Earth Planet. Sci. Lett., 37,
230-236.

 

Smith, P.J., 1979, Hotspots and African mantle,’Na‘turez
281, 13.

sorensen, H., ed., 1974, The Alkaline Rocks, Wiley-
Interscience, 622 p.

 

Sparks, R.S.J., Wadge, G., 1975, Geological and geochemical
studies of the Sintra alkaline igneous complex,
Portugal, Bull. Vblcanologique, 39, 385—406.

 

Stebbins, J., Thompson, G., 1978, The nature and petro-
genesis of intra-ocean plate alkaline eruptive and
plutonic rocks: King's Trough north-east Atlantic,
J. VOlcanol. Geotherm. Res., 4, 333-361.

 

Stephenson, D., 1972, Alkali Clinopyroxenes from nepheline
syenites of South Qorog centre, South Greenland,
LithOSe ' 5' 187-201.

Stephenson, D., 1976, A simple-shear model for the ductile
deformation of high-level intrusions in South Green-
land, J. Geol. Soc. Lond., 132, 307—318.

 

Stern, R.J., 1979, Late Precambrian Ensimatic volcanism in
the Central Eastern Desert of Egypt, Ph.D. dissertation,
Univ. of California, San Diego, 210 p.

Stocker, R.L., Ashby, M.F., 1973, On the rheology of the
upper mantle, Rev. Geophys.’Space’Phys., 11, 391-426.

 

280

Stocklin, J., 1968, Structural history and tectonics of
Iran: a review, Am. Assoc. Petroleum Geologists,
1229-1258.

 

Strong, D.F., Minatidis, D.G., 1975, Geochemistry and
tectonic setting of the late Precambrian Holyrood
plutonic series of eastern Newfoundland, Lithos, 8,
283-295.

Sturt, B.A., Ramsay, D.M., 1965, The alkaline complex of
the Breivikbotn area, S¢r¢y, northern Norway, Norz
Geol. Unders., 231, 1-142.

 

Sugisaki, R., 1976, Chemical characteristics of volcanic
rocks: relation to plate movements, Lithos, 9, 17-30.

Sun, S.S., Hanson, G.N., 1975, Origin of Ross Island basani-
toids and limitations upon the heterogeniety of mantle
sources for alkali basalts and nephelinites, Contrib.
Mineral. Petrol., 52, 77-106.

Sun, S.S., Hanson, G.N., 1976., REE evidence for differen-
tiation of McMurdo volcanics, Ross Island, Antartica,
Contrib. Mineral. Petrol., 54, 139-155.

 

Swanson, M.T., 1982, Preliminary model for an early trans—
form history in central Atlantic rifting, Geology,
10, 317—320.

Swartz, D.H., Arden, D.D., 1960, Geological history of Red

Sea area, Am. Assoc. Petroleum Geologists, 44, 1621-
1637.

 

Sykes, L.R., 1978, Intraplate seismicity, reactivation of
pre-existing zones of weakness, alkaline magmatism,
and other tectonism post-dating continental fragmenta-
tion, Rev. GeOphys. Space Phys., 16, 621-688.

 

Symth, C.H., 1913, Composition of the alkaline rocks and
its significance as to their origin, Am. J. Sci., 36,
1-36.

 

Tanaka, T., Nishizawa, 0., 1975, Partitioning of REE, Ba,
and Sr between crystal and liquid phases for a natural
silicate system at 20 Kb pressure, Geochem. J., 9,
161-166.

 

Tapponier, P., Molnar, P., 1976L Slip line field theory and
large-scale continental tectonics, Nature, 264,
319-324.

Tapponnier, P., Molnar, P., 1979, Active faulting and
Cenozoic tectonics of the Tien Shan, Mongolia, and
Baykal regions, J. Geophys. Res., 84, 3425-3459.

 

281

Tekeli, 0., 1981, subduction complex of pre-Jurassic age,
northern Anatolia, Turkey, Geology, 9, 68-72.

Thiessen, R., Burke, K., Kidd, W.S.F., 1979, African hot-
spots and their relation to the underlying mantle,
Geology, 7, 263-266.

Thompson, G., Melson,_W.G., 1972, The petrology of oceanic
crust across fracture zones in the Atlantic Ocean:
evidence of a new kind of sea—floor spreading, J. Geol.,
80, 526-538.

 

Thorpe, R.S., Potts, P.J., Sarre, M.B., 1977, Rare earth
evidence concerning the origin of granites of the Isle
of Skye, NW Scotland, Earth Planet. Sci. Lett., 36,
111-120.

 

Thorpe, R.S., Smith, K., 1974, Distribution of Cenozoic
volcanism in Africa, Earth Planet. Sci. Lett., 22,.
91-95.

 

Tilley, C.E., 1958, Problems of alkali rock genesis, Quart.
J. Geol. Soc., 113, 323-359.

 

Treuil, M., Varet, J., 1973, Criteres volcanologiques,
petrologiques et geochimiques de la genese et de la
differentiation des magmas basaltiques: example de
l'Afar, Bull. Soc. Geol. Fr., 15, 506-540.

 

Turcotte, D.L., Ahern, J.L., 1978, A porous flow model for
magma migration in the asthenosphere, J. Geophys. Res.,
83, 767-772.

 

Turcotte, D.L., Oxburgh, E.R., 1978, Intra-plate volcanism,
Phil. Tran. R. Soc. Lond., 288-A, 561-579.

 

Tuttle, C.F., Bowen, N.L., 1958, Origin of granite in light
of experimental studies in the system NaAlSi308-
KAlSi308-Si02—H20, Geol. Soc. Am. Mem., no. 74.

Unger, J.D., 1967, Melting of granite under effective
confining pressure, Unpubl. M.Sc. thesis, Mass. Inst.
Tech.

Upton, B.G.J., 1974,_The alkaline province of Southwest
Greenland, in The Alkaline Rocks, Sdrensen, H.,_ed.,
Wiley-Interscience,_221-238.

Upton, B.G.J., Blundell,_D.J., 1978, The Gardar igneous
province: evidence for Proteronic Continental rifting,
in PetrOIOgy and GeOchemistry_of Continental‘Rifts,
Neuman, E.R., Ramberg,_eds., ReideI}_Dordrecht,,163-l72.

282

Upton, B.G.J., Wadsworth, W.J., 1972, Aspects of magmatic
evolution on Reunion Island, Phil. Trans. R. Soc.
Lond., 27l-A, 105-130.

 

Vail, J.R., 1970, Tectonic control of dykes and related
irruptive rocks in eastern Africa, in African Magma—
tism and Tectonics, Clifford, T.N., Gass, I.G., eds.,
Oliver and Boyd, Edinburugh, 337-354.

 

Vail, J.R., 1976, Location and geochronology of igneous
ring-complexes and related rocks in northeast Africa,
Geol. Jb., 20, 97-114. ‘

 

Vail, J.R., Kuron, J.L., 1978, High level igneous emplace—
ments in the Red Sea hills, Sudan, Geol. Rundschau.,
67, 521—530.

 

Vail, P.R., Mitchum, R.M., Thompson, S., 1977, Seismic
stratigraphy and global changes of sea level, Part 4:
fglobal cycles of relative changes of sea level, in
Seismic Stratigraphy - applications to hydrocarbon
exploration, Am.7ASSoc. Petroleum Geologists, Memoir
26, 83—97.

 

 

van Breemen, 0., Aftalion, M., Johnson, M.R.W., 1979, Age of
the Loch Borrolan complex, Assynt, and late movements
along the Moine Thrust zone, J. Geol. Soc. Lond., 136,
489—495.

 

van Breemen, 0., and Bowden, P., 1973, Sequential age trends
for some Nigerian Mesozoic granites, Nature Phys. Sci.,
242, 9-11. ‘1

 

van Breemen, 0., Hutchinson, J., Bowden, P., 1975, Age and
origin of the Nigerian Mesozdic granites: A RbnSr
isotopic study, Cbntrib. Mineral. Petrol., 50, 157—172.

 

van Breemen, 0., Upton, B.G.J., 1972, The age of some Gardar
intrusive complexes, South Greenland, Geol.-Soc. Am.
Bull., 83, 3381—3390.

 

VOgt, P.R., 1972, Evidence for global synchronism in mantle
plume convection, and possible significance for geology,
Nature, 240, 338—342.

VOgt, P.R., 1979, Global magmatic episodes: new evidence and
implications for the steady-state mid-oceanic ridge,

VOgt, P.R., 1981, On the applicability of thermal conduction
models to mid—plate volcanism: comments on a paper by
Gass et al., J. Geoph. Res., 86, 950-960.

 

 

283

Watson, E. B., 1975, Partitioning of trace and minor elements
between coexisting silicate liquids; Carnegie Inst.
Washington Year Book, 74, 500- 504.

 

 

Watson, E.B., 1976, Two liquid partition coefficients:
experimental dataand geochemical implications,
Contrib. Mineral. Petrol., 56, 119-134.

 

watson, E. B., Naslund, H. R., 1977, The effect of pressure
on liquid immiscibility in the system Kzo-FeO-A1203-
SiOz-COZ, Carnegie Inst., washington Yearbook, 76,
410—414.

Watterson, J., 1978, Proterozoic intraplate deformation in
the light of Southeast Asian neotectonics, Nature, 273,
636-640.

Weaver, S.D., Sceal, J.S.C., Gibson, I.L., 1972, Trace—
element data relevant to the origin of trachytic and
pantelleritic lavas in the East African rift system,
Contrib. Mineral. Petrol., 36, 181-194.

 

Weeterman, J., 1970, The creep strength of Earth's mantle,
Rev. Geophys. Space Phys., 8, 145-168.

Weeterman, J., Weeterman, J.R., 1975, High temperature
creep of rock and mantle viscosity, Ann. Rev. Earth
Planet Sci., 3, 293-315.

 

 

weill, D.F., Drake, M.J., 1973, European anomaly in plagio-
clase feldspar: experimental results and semiquantita-
tive model, Science, 180, 1059-60.

Wendlandt, R.F., Harrison, W.J., 1979, Rare earth partition-
ing between immiscible carbonate and silicate liquids
and 002 vapor: results and implications for the
formation of light rare earth-enriched rocks, Contrib.
Mineral. Petrol., 69, 409-419.

 

 

White, W. M., Tapia, M. D. M., Schilling, J. G., 1979, The
petrology and geochemistry of the Azores Island,
Contrib. Mineral. Petrol., 69, #3, 201-214.

 

Wilkinson, J.E.G., 1974, The mineralogy and petrography of
alkali basaltic rocks, in The Alkaline Rocks, Sbrensen,
H...(ed.), Wiley-Interscience, 67—95.

 

Williams, L.A.J., 1970, The volcanics of the Gregory Rift
Valley, East Africa,'Bull.‘Vo1canol., 34, 439-465.

 

Wilson, J.T., 1963, Evidence from islands on the spreading
of ocean floors, Nature, 197, 536-538.

284

Wright, J.B., 1965, Petrographic sub-provinces in the
Tertiary to recent volCanics of Kenya, Geol.'Mag.,
102, 541—557.

 

Wright, J.B., 1970, Distribution of volcanic rocks about
mid-ocean ridges and the Kenya rift valley, Geol. Mag.,

 

WOod, D.A., 1978, Major and trace element variations in the
Tertiary Lavas of Eastern Iceland and their signifi-
cance with respect to the Iceland geochemical anomaly,
J. Petrol., 19, 393—436.

 

Wbod, D.A., Tarney, J., varet, J., Saunders, A.D., Bougault,
H., Joron, J.L., Treuil, M., Cann, J.R., 1979, Geo-
chemistry of basalts drilled in the N. Atlantic by IPOD
Leg 49: implications for mantle heterogeneity, Earth
Planet. Sci. Lettg, 42, 77—97.

Wyllie, P.J., 1979, Magmas and volatile components, Am.
Mineral., 64, 469-500.

 

Yoder, H.S., Tilley, C.E., 1962, Origin of basalt magmas:
experimental study of natural and synthetic rock
systems, J. Petrol., 3, 342—532.

 

Youssef, M.I., 1968, Structural pattern of Egypt and its
interpretation, Am.'Assocs Petroleum GeOIOQists, 52,
601-614.

Yuen, D.A., Schubert, G., 1977, AsthenOSpheric shear flow:
thermally stable or unstable?‘ Geophys.'Res.‘Lett., 4,
503-506.

 

Yuen, D.A., Schubert, G., 1979, The stability of frictionally
heated upper mantle shear flows, Geophys.'J. R. astr.
'ISOCo' 57’ 189'2070

 

zartman, R.E., 1977, Geochronology of some alkalic rock
provinces in eastern and central United States, Ann.
‘Rev. Earth Planet. Sci., 5, 257—286.

 

Zener, C., 1948, Elasticity and Anelasticity of Metals,.
Chicago, Univ. of7ChicagoPress. T

 

Zielinski, R.A., 1975, Trace element evaluation of a suite
of rocks from Reunion Island, Indian Ocean, Geochim.
Cosmochim. Acta, 39, 7l3é734.

 

 

Zielinski, R.A., Frey, P.A., 1970, Gough Island: evaluation
of a fractional crystallization model, Contrib. Mineral.