121
811
THS

 

LIBRARY

Michigan Sta “:6
University

 

This is to certify that the

thesis entitled
MAJOR ELEMENT CHEMISTRY AND TECTONIC

SETTING 0F PLUTONIC ROCK SUITES

presented by

WILLIAM L. PETRO

has been accepted towards fulﬁllment
of the requirements for

M . S . degree in Geo I 03y

 

 

Date 8-25-77

 

 

MAJOR ELEMENT CHEMISTRY AND TECTONIC
SETTING OF PLUTONIC ROCK SUITES
BY

William L. Petro

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

College of Natural Science
Department of Geology

1977

61

C73

3 . E
121
r .-'
. , ,

ABSTRACT

MAJOR ELEMENT CHEMISTRY AND TECTONIC
SETTING 0F PLUTONIC ROCK SUITES

BY
William L. Petro

Plutonic rock suites from compressional and exten-
sional plate margins differ in major element chemistry and
igneous petrogenesis. Comparisons among type examples
of plutonic suites generated at the two types of plate
margins show that extensional suites are more alkali
enriched with respect to calcium and iron enriched with
respect to magnesium than are compressional suites.
Extensional suites have bimodal distributions of differ-
entiation index and composition of normative plagioclase,
whereas compressional suites have unimodal distributions.
These differences may be related to extensive crustal
melting and/or magma mixing in the generation of compres-
sional suites, and limited melting and mixing in exten-
sional suites. Extensional suites may also undergo
melting and fractionation at lower water pressures.
Comparison of possible compressional and extensional

plutonic suites with criteria developed from the

William L. Petro

comparison of type examples suggests that these criteria

may be useful in the determination of tectonic setting.

ACKNOWLEDGMENTS

I would like to thank Dr. T.A. Vogel for sugges-
tion of the problem, and for his patience, support, and
constructive criticism during the deve10pment of the
thesis.

I would like to thank Dr. J.T. Wilband, Dr. D.F.
Sibley, and Dr. F.W. Cambray for their constructive
criticism of the thesis. I would also like to thank
Dr. J.T. Wilband for his considerable assistance with
the computer programming.

John Ansted, Steve Mattson, Doug Padden, Mike
Price, and Dave Westjohn provided many helpful discussions.

I would also like to thank Nancy, Brian, Mom, and

Dad for their support.

ii

TABLE OF CONTENTS

Page
ACKNOWLEDGMENTS . . . . . . . . . . . . ii
LIST OF TABLES O O O O O O O O O O O 0 iv
LIST OF FIGURES . . . . . . . . . . . . V
Chapter
I. INTRODUCTION . . . . . . . . . . 1
Previous Work . . . . . . . . . . 3
Characteristics of Compressional and
Extensional Plutonic Suites . . . . . 4
II. CHEMICAL CHARACTERISTICS OF
PLUTONIC SUITES . . . . . . . . . 7
Approach and Methodology . . . . . . 7
Data 0 O O I O O O O O O 0 O I 10
III. DISCUSSION OF PETROGENESIS OF
PLUTONIC SUITES . . . . . . . . . 27
Compressional Suites . . . . . . . . 27
Extensional Suites . . . . . . . . 3O
Suites That Depart From Type Examples . . 34
IV. CONCLUSIONS . . . . . . . . . . . 36
REFERENCES 0 O O O I O O O O O O O O 39

iii

LIST OF TABLES

Table Page
1. List of Suites . . . . . . . . . . 8
2. ALI Values for Plutonic Suites . . . . . l6
3. Comparison of Means of Variables for

SiO2 Interval 70.00-75.00% . . . . . 23

iv

Figure

1.

LIST OF FIGURES

Page

Location map of suites (modified from
Sawkins et a1., 1974). Suite abbrevia-
tions are listed in Table l . . . . . . 9

Relative frequency distributions of DI:

(a-c) type compressional suites, (d-f)

type extensional suites, (g-i) possible
compressional suites, (j-l) possible exten-

sional suites. DI is the differentiation

index (Thornton and Tuttle, 1960), and is

given in class intervals of five. F is the
relative frequency in percent. Suite
abbreviations are listed in Table l . . . 12

' Relative frequency distributions of PL:

(a-c) type compressional suites, (d-f)

type extensional suites, (g-i) possible
compressional suites, (j-l) possible exten-

sional suites. PL is the composition of the
normative plagioclase in percent anorthite,

and is given in class intervals of five. F

is the relative frequency in percent.

Suite abbreviations are listed in Table l . l4

AFM ternary diagrams: (a) type compres-

sional suites, (b) type extensional

suites, (c) possible compressional

suites, (d) possible extensional suites . . 18

Relative frequency distributions of

alumina saturation types: (a) peralumi-

nous, (b) metaluminous, (c) peralkaline.

F is the relative frequency in percent.

Suite abbreviations are listed in Table l . 21

CHAPTER I

INTRODUCTION

It has long been recognized that igneous rock
suites can be divided into two tectonic types, orogenic
and anorogenic (see Earth, 1962, for review). Orogenic
suites are associated with compressional tectonics, and
commonly occur in fold belts. Anorogenic suites occur
in areas of extensional tectonics, such as rift zones,
or in areas of little or no tectonic activity.

In plate tectonics models, orogenic suites are
generated in subduction zones at two types of compres-
sional plate boundaries (Dickinson, 1971): ocean-ocean
plate boundaries where oceanic lithosphere is subducted
beneath oceanic lithosphere; and ocean—continent plate
boundaries where oceanic lithosphere is subducted beneath
continental lithosphere. A third type of compressional
plate boundary is a continent-continent plate boundary.
Due to the low density of continental lithosphere, sub-
duction is restricted. Magmatism associated with con-
tinental collision, though discussed in general (Naylor,
1971; Gilluly, 1971; Dewey and Burke, 1973), has not

been adequately evaluated due to the lack of unequivocal

or easily accessible examples of collision generated
suites.

Extensional (spreading) plate boundaries generate
anorogenic suites, which are therefore common at newly
formed continental margins or rift zones. Mantle plume
activity may also generate anorogenic suites.

The relationship between tectonic setting and
chemistry of volcanic rock suites has been well estab-
lished (Martin and Piwinskii, 1972; Christiansen and
Lipman, 1972). Also, plutonic suites have been compared
with volcanic suites of known tectonic setting, thus
allowing inferences to be made about the tectonic setting
of the plutonic suites (Strong and Minatidis, 1975).
However, there is a need to establish the direct relation-
ships between tectonic setting and plutonic rock chemistry.
The purpose of this paper is to document the chemical
differences between plutonic rock suites associated with
compressional and extensional plate boundaries, and to
attempt to explain the differences in terms of current
petrogenetic theories.

This research is concerned only with the major
elements for reasons of economy and availability of
data. Pearce and Cann (1973), Miyashiro (1975), and
Strong and Minatidis (1975) have shown that trace ele-

ments are potential indicators of tectonic setting.

Previous Work

 

Martin and Piwinskii (1972) first quantitatively
evaluated the relationship between the tectonic setting
and chemistry of igneous rock suites. They distinguished
orogenic and anorogenic volcanic suites on the basis of
the frequency distribution on the differentiation index,
DI (DI = normative Q + Or + Ab + Ne + Lc + Kp; Thornton
and Tuttle, 1960). They showed that the DI frequency
distributions of orogenic suites are unimodal with an
intermediate mode, whereas those of anorogenic suites are
bimodal with acidic-basic modes. They used AFM diagrams
to show that anorogenic suites have stronger iron enrich—
ment trends than do orogenic suites. They also showed
that anorogenic suites contain more compositional scatter
in variation diagrams of weight percent oxides versus DI.

Christiansen and Lipman (1972) used the alkali-
1ime index (ALI) to distinguish the two tectonic types of
volcanic suites in the western U.S.A. The ALI (after
Peacock, 1931) is a single value that characterizes a
suite based on the variation of CaO/Nazo + K20 versus
8102. The ALI is the value of SiO2 for which CaO/Nazo +
K20 is equal to 1.00. They noted that orogenic volcanic
suites have higher ALI values than do anorogenic volcanic
suites. They also noted that the volcanic suites changed
from unimodal (intermediate) to bimodal (acidic-basic),

and related this to the change in tectonic setting that

occurred when compression changed to extension about 30
million years ago.

Strong and Minatidis (1975) used the ALI to com-
pare a plutonic suite with the volcanic suites of Lipman
and Christiansen (1972) to establish a tectonic setting
for the plutonic suite. Their use of AFM diagrams was
limited by the fact that they did not compare their data
with plutonic suites from extension zones.

Johnston et a1. (1976) have attempted to dis—
tinguish granitic rocks from two types of tectonic set-
tings. They concluded that extensional suites have
alumina undersaturation, high FeO, and low MgO and CaO,
whereas compressional suites have alumina oversaturation,
low FeO, and high MgO and CaO.

Characteristics of Compressional
and’ExtensionaI PlutonIc Suites

General structural and compositional differences
between compressional and extensional plutonic suites can
be determined by examining Mesozoic-Cenozoic occurrences
of plutonic rock suites of known tectonic setting.

Mesozoic-Cenozoic compressional suites are long,
linear belts parallel to plate margins. Regional meta-
morphism is often associated in space and time with the
forceful emplacement of the plutons. The suites are
predominately intermediate in composition (quartz

diorite-tonalite—granodiorite-quartz monzonite) with

relatively minor amounts of ultrabasic, basic, acidic,
and alkalic rocks. However, the sequence in a suite
associated with a subduction zone is generally basic
rocks first, with the intermediate rocks becoming more
significant with time, and acidic rocks late (Larsen,
1948; Jakes and White, 1971, 1972; Presnall and Bateman,
1973).

Mesozoic-Cenozoic extensional suites often occur
as local, spot-like complexes. Block faulting, cauldron
subsidence, ring dikes, and cone sheets in subvolcanic
complexes are common. If regional metamorphism is
present, the emplacement of the plutonic suites follows
the metamorphic event. End-member compositions (basic,
acidic, and alkalic) predominate in extensional suites.
Ultrabasic rocks are more common than in compressional
suites. In quartz oversaturated rocks, intermediate
compositions are relatively minor, except locally. Inter-
mediate silica compositions may be significant as quartz
undersaturated rocks. Coexisting acidic-basic magmas
are relatively common, often as net-veined complexes.
Carbonatites may be locally prominent in some extensional
environments (MacIntyre, 1977).

Complications in the distinction of tectonic
settings arise in those areas where both compression
and extension occur in close proximity (time and/or

space). Bussell et al. (1976) have described ring

complexes at high crustal levels in the Peruvian Coastal
Batholith, which is a compressional suite associated with
subduction (Hamilton, 1969a). The nature of ring com-
plexes requires extension, suggesting that both compres-

. sion and extension have occurred in the same environment.
Back-arc spreading in island arc complexes (Oxburgh and
Turcotte, 1971; Karig, 1971) may produce a similar type

of problem.

CHAPTER II

CHEMICAL CHARACTERISTICS

OF PLUTONIC SUITES

Approach and Methodology

Published analyses of plutonic rocks were
selected to evaluate chemical criteria for the distinction
of extensional and compressional suites. The major source
of data is a computer based set of igneous rock analyses
compiled by Eastern Washington State College (Mutschler,
Rougon, and Lavin, 1976a,b).

"Type" examples of plutonic rock suites were
selected on the basis of published work on tectonic set-
ting. The Sierra Nevada, Southern California, and
Aleutian-Alaska Batholiths were selected to represent
type examples of compressional suites (Hamilton, l969a,b;
Dickinson, 1970; Reed and Lanphere, 1973; Lanphere and
Reed, 1973). Iceland, East Greenland (Tertiary), and the
British Isles (Tertiary) were selected to represent type
examples of extensional suites (Morgan, 1971, 1972;
Brooks, 1973a,b; Einarsson, 1973).

Other suites were also investigated to evaluate
the use of chemical criteria in determining the tectonic
setting of plutonic suites. Possible compressional

7

suites selected were the Boulder Batholith (Kistler,
1974), the Hercynian granites of the British Isles
(Riding, 1974), and the South Mountain Batholith
(McKenzie and Clarke, 1975). Possible extensional
suites selected were the White Mt. Magma Series (Morgan,
1971), the Younger Granites of Nigeria (Bowden, 1970),
and the Keweenawan plutonic rocks (Chase and Gilmer,
1973; Sims, 1976).

Table 1 lists the age, total number of samples,
and abbreviation for each of the suites. Figure 1

shows the locations of the suites.

TABLE l.--List of Suites.

 

 

 

Suite

Code* N# Name Age
ARA 138 Aleutian-Alaska Batholith M-T
BIT 164 British Isles Tertiary T
BLB 106 Boulder Batholith (Montana) M
COR 058 Hercynian Granites (British Isles) P
EGL 061 East Greenland Tertiary T
ISL 104 Iceland C
KEW‘ 086 Keweenawan Plutonic Rocks Pe
NIG 089 Younger Granites of Nigeria M
PSM 074 South Mt. Batholith (Nova Scotia) P
SCB 087 Southern California Batholith M
SRN 302 Sierra Nevada Batholith M
WMT 094 White Mt. Magma Series M

Ages: C Cenozoic T = Tertiary M = Mesozoic

P = Paleozoic P6 = Precambrian
* Abbreviations from Mutschler et al. (1976b)

# Number of plutonic rocks

.H manna :ﬂ woumﬂa mum mcoﬂumw>onnnm .
muasm .Avhma ..Hm no mcﬂx3mm Eouw cmﬂwwwoev mmuwsm mo mmE :OHumUOAII.H ounmﬁm

 

 

 

 

 

 

  

22a
56:.

       

0.30
2.5:. III

21w

F 2-... — A}:

:3:~E< r
mam A; a. .

 
  

5:33.

 

v.2:
9.14:6

 

 

10

222:

Data were taken as weight percent oxides and
recalculated to 100 percent (anhydrous). Normative
minerals were calculated using a modified C.I.P.W. norm
computer program. Several petrochemical indicators were
also calculated, including the differentiation index (DI),
alkali-lime index (ALI), and types of alumina saturation.

Several types of variation diagrams were used
in an attempt to distinguish the two types of suites.
Simple variation diagrams of weight percent oxides versus
Si02, DI, or Larsen Index (Larsen Index = 1/3 SiO2 +
K20 - CaO - FeO - MgO; Larsen, 1948) do not show clear
differences between the two types of suites.

Histograms were used to illustrate the relative
frequency distributions of the DI (Figures 2a-l) and
the composition of the normative plagioclase, PL (Figures
3a-1). Comparison of the type compressional suites
(Figures 2a—c, 3a-c) with the type extensional suites
(Figures 2d-f, 3d—f) shows that the compressional suites
have unimodal distributions with intermediate modes
whereas the extensional suites have bimodal distribu-
tions with acidic-basic modes. Other types of relative
frequency distributions were examined (for example,
8102, Larsen Index), but DI and PL are preferred because

they indicate more clearly the differences between the

two types of suites.

11

Figure 2.--Re1ative frequency distributions of DI: (a-c)
type compressional suites, (d-f) type exten-
sional suites, (g-i) possible compressional
suites, (j-l) possible extensional suites. DI
is the differentiation index (Thornton and
Tuttle, 1960), and is given in class intervals
of five. F is the relative frequency in per-
cent. Suite abbreviations are listed in Tablelu

13

Figure 3.--Relative frequency distributions of PL: (a-c)
type compressional suites, (d-f) type exten-
sional suites, (g-i) possible compressional
suites, (j-l) possible extensional suites.

PL is the composition of the normative plagio-
clase in percent anorthite, and is given in
class intervals of five. F is the relative
frequency in percent. Suite abbreviations are
listed in Table l.

l4

 

 

 

 

 

 

 

15

The alkali-lime index (ALI) is a useful classi-
fication of suites. The ALI values were determined in two
ways. Smooth curves of CaO/Nazo + K20 versus SiO2 were
hand-drawn, and a visual estimate of the value of SiO2
for which the curve intercepts a ratio value of 1.00 was
made. Also, third order "best fit" curves were calcu-
lated using SPSS subprogram Regression (Nie et al., 1975).
Table 2 presents the ALI values for the suites. Third
order equations were used to balance goodness of fit and
simplicity. The data were edited for the use of samples
with CaO/Nazo + K20 less than 4.00, because the region
of interest is for values near 1.00, and inclusion of
higher values compresses the scale so that investigation
of the desired area is difficult. Values for the Hercy-
nian and South Mt. suites were not reported because of
lack of apprOpriate compositional range that produces
geologically unreasonable ALI values (less than 0 or
greater than 100) or r2 values that indicate poor fit
(r2 less than .5). Comparison of the type examples
shows that compressional suites have ALI values that
are high (60-64), whereas extensional suites have
values that are low (50-56). This indicates that
extensional suites are more alkali enriched with respect
to calcium than are compressional suites.

AFM ternary diagrams also distinguish the two

tectonic types of suites (Figures 4a-d). Comparison of

16

TABLE 2.--ALI Values for Plutonic Suites.

 

 

Suite Published Visual Third Order Mult. r2
Code N* Value # Estimate Calculated for calc.
ARA 135 - 61 61.6 .82
BIT 105 - 56 55.0 .91
BLB 105 58 58 59.1 .91
COR $ - — — -
EGL 043 - 50 50.9 .62
ISL 094 - 55 56.0 .97
KEW 076 - 52 50.8 .57
NIG 088 - 52 49.7 .85
PSM $ - - - -
SCB 078 64 64 63.5 .94
SRN 291 60,63@ 62 61.4 .88
WMT 093 - 53 51.6 .71

 

* Data edited for use of CaO/NaZO + K20 of less than
4.000 (see accompanying text)

# From Kistler (1974)
$ ALI not determined (see accompanying text)
@ 60 for the central SRN

63 for the western SRN
the type examples shows that extensional suites (Figure
4b) have stronger iron enrichment trends than do com-
pressional suites (Figure 4a). The extensional trend
tends to be closer to and more nearly parallel to the AF
side at compositions approaching the A apex, whereas the
compressional trend tends to be more nearly perpendicular
to the FM side for the entire trend. Considerable over-
lap does occur.

Degrees of alumina saturation are determined by

comparisons of the molecular proportions of A12 3,

17

Figure 4.--AFM ternary diagrams: (a) type compressional
suites, (b) type extensional suites, (c) pos-
sible compressional suites, (d) possible
extensional suites.

18

 

 

{III .1 J III.< .|.I4III4III4I3

 

as .\
no“ 1r
m .m P
E F.
A
.‘L
.,/ s. “L
.4 E
l/ \
/

LL

 

 

om... mm mm #:opnm

19

Na 0 + K O, and Ca0 + NaZO + K20 (Shand, 1927; Carmichael,

2 2
Turner, and Verhoogen, 1974). Three main alumina satura-
tion types may be considered. Peraluminous rocks have
A1203 greater than Ca0 + Nazo + K20 (molecular). Meta-
luminous rocks have A1203 greater than NaZO + K20, but
less than Ca0 + Nazo + K20. Peralkaline rocks have

Al 0 less than Na 0 + K O. Peraluminous rocks are

2 3 2 2
characterized by corundum in the norm, whereas metalumi-
nous rocks are characterized by anorthite in the norm.
Peralkaline rocks may have acmite, sodium silicate, or
potassium silicate in the norm. Figure 5 presents the
relative frequency distributions of the alumina satura-
'tion types of the suites. Comparison of type examples
shows that only extensional suites have peralkaline rocks
(Figure 5c). Compressional suites tend to have higher
frequencies of peraluminous rocks (Figure 5a), whereas
extensional suites tend to have slightly higher fre-
quencies of metaluminous rocks (Fibure 5b). The varia-
tion in alumina saturation appears to be more due to
variations in CaO and NaZO + K20 than in A1203. A1203
versus DI or SiO2 variation diagrams show that both
types of suites have similar A1203 (also compare Table
3).

An attempt was made to evaluate the mean

differences of variables for samples in a particular

interval of SiO2 (70.00-75.00 weight percent). This

20

Figure 5.--Relative frequency distributions of alumina
saturation types: (a) peraluminous, (b)
metaluminous, (c) peralkaline. F is the
relative frequency in percent. Suite abbre-
viations are listed in Table 1.

 

 

21

1M

9III

3]

SJ

IO)

“8
SISI
$193

will
IIIIS
83$

VIIV

 

 

 

22

interval was selected because it is useful in the analy-
sis of granites, and in some areas, granites are the
predominant rock type. There is a need to distinguish
granites from compressional and extensional environments.
Tukey's test (Snedecor and Cochran, 1974) was used to
test the equality of means for several variables for
two type suites from each tectonic setting. Twenty
samples were randomly selected from the appropriate
interval for each of the four suites (exact results
require equal sample sizes). Table 3 presents the
results of the test. Compressional suites have higher
CaO, and extensional suites have higher DI and total
alkalies. Extensional suites are more alkali enriched
with respect to calcium and iron enriched with respect
to magnesium than are compressional suites, as shown by
the differences in the two ratios in Table 3. Because
the DI includes albite and orthoclase, but not anorthite,
the difference in DI may also be related to the differ-
ences in CaO and Nazo + K20. These results are consis-
tent with those previously discussed (ALI, AFM) in
showing alkali and iron enrichment in extensional suites.
Comparisons of the possible compressional and
extensional suites may be useful in determining how
well the suites fit their postulated settings, and how

useful the chemical criteria (established by comparison

23

TABLE 3.--Comparison of Means of Variables for SiOz Interval
70.00-75.00%.

 

 

BIT ISL ARA SCB

DI 88.87 a 91.97 a 83.10 b 83.34 b **
A1203 13.34 a 13.23 a 14.90 b 13.95 a
Fe203 1.23 b 1.90 a 0.73 b 1.14 b
FeO 1.97 a,c 0.79 d 1.13 c 1.56 a,d
MgO 0.31 a,b 0.22 a 0.52 b 0.54 b
CaO 1.24 a 0.86 a 2.40 b 2.22 b **
NaZO 3.79 a,b 4.35 a 4.09 a,b 3.53 b
K20 4.78 a 4.08 a,b 2.62 c 3.41 b
TiO2 0.29 a 0.26 a 0.25 a 0.26 a
NaZO + K20 8.57 a 8.40 a 6.68 b 6.91 b **
CaO/NaZO + K20 0.152 a 0.104 a 0.382 b 0.342 b **
FeO*/FeO* + MgO 0.907 a 0.928 a 0.791 b 0.829 b **

 

Suites with the same letter show no significant difference for that
variable at a = .05 with Tukey's test.

* FeO = total iron as FeO

** variable suitable for distinction of compressional and exten-
sional suites

of type suites) are in unambiguously determining the tec-

tonic setting of plutonic rock suites.

A comparison of the relative frequency distri-
loutions of DI and PL of the possible compressional
suites shows that the Boulder Batholith (Figures 29, 3g)
:resemblesthe type examples with an intermediate, uni—
Inodal distribution, whereas the Hercynian suite (Figures
2h, 3h) and the South Mt. Batholith (Figures 2i, 3i) have

unimodal distributions that are skewed toward the acidic

24

end-members. Comparison of the possible extensional
suites shows that the Keweenawan suite (Figures 2j, 3j)
is similar to the type examples in having a bimodal
distribution, whereas the Nigerian suite (Figures 2k, 3k)
and the White Mt. Magma Series (Figures 21, 31) have
acidic modes that predominate over the basic "modes" of
the suites such that the distributions appear unimodal
with acidic distributions. The White Mt. and Nigerian
suites differ from the Hercynian and South Mt. suites in
that the latter suites lack basic rocks.

Comparison of ALI values for "possible" suites
(Table 2) shows that the Keweenawan, White Mt., and
Nigerian suites are consistent with extension in having
ALI in the 50-56 range. The Boulder Batholith has an
ALI (58) that is intermediate between type examples of
compressional and extensional suites. ALI values were
not reported for the Hercynian or South Mt. Batholith
suites because these suites lack the appropriate
compositional range.

Comparison of AFM ternary diagrams for the pos-
sible compressional and extensional suites (Figures 40
and 4d, respectively) shows that the suites are consis-
tent with their postulated tectonic setting. Although
the Keweenawan trend does not fall as close to the AF
side as the type extensional suites, it does define a

strong iron enrichment trend.

25

Comparison of alumina saturation frequency distri-
butions for the possible compressional and extensional
suites with those of the type examples (Figures 5a-c)
shows that the Keweenawan, Nigerian, and White Mt. suites
are consistent with extension in having peralkaline rocks,
and relatively low frequencies of peraluminous rocks.

The Hercynian and South Mt. suites are consistent with
compression in lacking peralkaline rocks, and having high
frequencies of peraluminous rocks for the Hercynian and
South Mt. suites are higher than for type suites. The
Boulder Batholith is anomalous in resembling extensional
suites in having low frequencies of peraluminous rocks,
but resembling compressional suites in lacking peralkaline
rocks. Metaluminous frequencies do not appear to be
useful in that considerable overlap of the possible
suites suggests that both types of environments have
about the same distributions of metaluminous rocks.

Considering the data as a whole, the type examples
show clear differences between compressional and exten-
sional suites. Application of the criteria to the other
suites produces many exceptions. However, the
Keweenawan suite is consistent with extension. The
Younger Granites of Nigeria and the White Mt. Magma
Series are consistent with extension, except for the
relative lack of basic rocks. The Boulder Batholith is

consistent with compression, except for enrichment in

26

alkalies that causes it to fall in anomalous positions
for ALI and alumina saturation. The Hercynian granites
of the British Isles and the South Mountain Batholith are
consistent with compression except for the skewness of
the frequency distributions of DI and PL to the acidic

end-members, and higher frequencies of peraluminous rocks.

CHAPTER III

DISCUSSION OF PETROGENESIS

OF PLUTONIC SUITES

Compressional Suites

Three regions of melting are present in subduction
zones: the subducted oceanic lithosphere, the overlying
mantle wedge, and the overlying crustal complex (oceanic
or continental).

Melting of the upper portion of the subducted
oceanic lithosphere has been proposed by many workers
(for example, Fitton, 1971; Boettcher, 1973; Marsh and
(Zarmichael, 1974; Younker and Vogel, 1976), but melting
rnay be limited by the heat sink effect of endothermic
dehydration reactions (Anderson et a1. , 1976) . Fric-
txional heating (Turcotte and Schubert, 1973) appears to

hm; the most significant source of heat for dehydration
<31: melting. Wyllie et a1. (1976) have reviewed experi-
nmerrtal evidence on the compositions of magmas derived
by melting of the downgoing slab, and have concluded
that the magmas are of intermediate composition, but
are: not primary andesite or tonalite. Subducted sedi-
ments, if present, may also melt to produce granitic
magnnas (Huang and Wyllie, 1973).

27

28

The mantle wedge overlying the downgoing slab may
melt in response to two main processes: (1) water
derived from the dehydration of the downgoing slab lowers
the solidus temperatures of the mantle material (Wyllie,
1971), and (2) water and/or magma derived from the slab
reduces the density of the mantle material which pro-
duces diapiric uprise and melting in the rising diapirs
by adiabatic decompression (Ringwood, 1975). The composi-
tion of the liquids produced depends, in part, upon the
water content. Higher water contents drive the
composition of the liquids toward more intermediate
compositions (Ringwood, 1975; Yoder, 1976: Wyllie et al.,
1976).

The overlying crustal complex may be melted by
the heat transported by the rising mantle derived magmas
(Younker and Vogel, 1976). Melting of most crustal rocks
[Iroduces magmas that are generally granitic in composi-
trion. Granites represent the minimum melting composi-
tion, whereas melting at temperatures higher than the

minimum produces more intermediate compositions (Tuttle
and Bowen, 1958; Winkler, 1976; Brown, 1970, 1973;
Fyfe, 1973; Presnall and Bateman, 1973).

Alternative heat sources for melting are decay
0f5.radiogenic heat elements and conduction from the
hOtter underlying mantle, but Younker (1974) has shown

that only large amounts of mantle derived magmas can

29

supply the heat necessary for significant amounts of
crustal melting. Other factors to be considered are the
lowering of solidus temperatures by non-hydrostatic
stress or introduction of volatiles from below. The
former has yet to be quantitatively evaluated (Yoder,
1976). The latter is a possibility in light of the
dehydration of the downgoing slab in subduction zones,
and needs further evaluation.

Two alternatives that account for the predominately
intermediate composition and unimodal distribution of
plutonic suites generated at compressional plate margins
are:

l) Crustal derived magmas mix in varying propor-
tions with mantle derived magmas to produce
magmas of intermediate composition (Younker
and Vogel, 1976). Magma mixing has been
demonstrated to be a viable petrogenetic
process by Eichelberger (1975) and
Anderson (1976). Mixing may be promoted
by deformation.

2) Compositionally intermediate magmas are
derived by the melting of crustal rocks
at temperatures higher than the minimum
melting temperatures. Equilibrium fusion
events of this type followed by fractional

crystallization of the magmas have been

30

proposed by Presnall and Bateman (1973)

to account for the spectrum of rock types

observed in the Sierra Nevada Batholith.
Both of these processes may operate, but both are
dependent upon large amounts of heat for crustal melting,
which according to Younker (1974) can only be supplied

by the flux of mantle derived magmas.

Extensional Suite

 

Gravitational instabilities in the mantle result
in the diapiric uprise of hot mantle material. When
rising mantle diapirs impinge upon the lower crust,
rifting may begin (Morgan, 1971, 1972; Burke and Wilson,
1976). Melting in the rising mantle material produces
predominately basaltic liquids (Ringwood, 1975).

The composition of basaltic liquids produced by
melting is controlled by pressure, volatiles, and the
degree of melting. Alkalic liquids are produced at
(greater depths (higher pressures) than are tholeiitic
liquids (Ringwood, 1975; Yoder, 1976). Melting in the
presence of high water pressures drives the composition
toward more intermediate compositions (Ringwood, 1975;
Yoder, 1976; Wyllie et al., 1976), whereas melting in
the presence of high CO2 pressures drives the composi-
tion of the liquids toward more alkalic, quartz under-

saturated compositions (Yoder, 1976; Wyllie and Huang,

31

1976). Small degrees of melting produce alkalic liquids,
whereas moderate degrees of melting tend to produce
tholeiitic liquids (Ringwood, 1975; Yoder, 1976).

Compressional suites have tholeiitic to andesitic
basalts, whereas extensional suites have alkalic to
tholeiitic basalts, suggesting that the extensional
suites have mantle derived liquids produced by smaller
degrees of melting at higher total and CO2 pressures
than those associated with compressional suites.

Fractional crystallization of basaltic magmas may
also be important, and some of the alkalic basaltic
rocks in extensional zones may represent liquids modified
by the high pressure fractionation of orthopyroxenes,
which produces alkalic, quartz undersaturated basaltic
liquids (Ringwood, 1975; Yoder, 1976). These alkalic
basalts may be fractionated at crustal levels to produce
syenitic rocks.

If a sufficient flux of magma intrudes the over-
lying crust, melting of the crustal rocks will produce
granitic liquids (Younker, 1974). The differences of
rock type distributions between compressional and exten-
sional suites may be due to either greater degrees of
magma mixing or greater degrees of crustal melting at
temperatures higher than the minimum melting temperature
in compressional areas. Coexisting acidic and basic

magmas appear to have been common in extensional areas,

32

whereas evidence for mixing of the magmas in minor.

The lack of mixing may be related to the tectonics of
the areas, because in extensional areas, the magmas are
passively emplaced. However, in compressional areas,
deformation may tend to mix the magmas.

A consequence of the quiet tectonic environment
is that in some of the extensional zones, fractionation
of basic liquids may occur, producing layered gabbroic
complexes. Wager and Brown (1967) have reviewed the
nature of layered gabbroic complexes. Strong iron
enrichment trends in these complexes is evidence for
fractionation at low water pressures. Only minor amounts
of granitic liquids are produced by the fractionation
of these magmas. In extensional areas where granitic
rocks are abundant, they are most likely produced by
the melting of crustal rocks, and not by the fractiona-
tion of basic magmas (for example, Dunham and Thompson,
1967).

The production of syenitic rocks by the fractiona-
tion of basaltic magmas, combined with the generation
of granitic rocks by melting of crustal rocks, accounts
for the coexistence of quartz oversaturated and under-
saturated rocks in some complexes (Chapman, 1976).
Chapman (1976) also suggested that minor mixing of the
syenitic and granitic magmas could produce small amounts
of alkali granites and quartz syenites in the same

complexes.

33

The granitic rocks in extensional zones are iron
and alkali enriched when compared with granitic rocks in
compressional zones. Smaller degrees of melting at
lower water pressures in extensional zones may account
for these differences. Brown and Fyfe (1970) have shown
that the initial granitic magmas formed by fusion of
crustal rocks associated with muscovite breakdown at
relatively low temperatures are low in MgO. Bowden
(1970) suggests that the alkali enriched rocks in
Nigeria are the initial crustal melting products, whereas
less alkali enriched rocks are products of more extensive
melting.

If extensional granitic rocks represent more
limited crustal melting than do compressional granitic
rocks, there may be a major thermal difference between
the two types of plate margins. The analysis of
Younker (1974) shows that higher heat flux is required
for more extensive melting, and that the heat flux may
be related to the flux of mantle derived magmas through
the crust. Compressional suites may therefore represent
suites generated in areas of higher heat flux of mantle
derived magmas, or flux over more extended periods of
time. Extensional suites may have a shorter time of
residence over the region of high heat flux, and after
the crust moves off the area of rifting (and high heat

flow), magma generation is greatly reduced.

34

Suites That Depart From
Type Examples

 

 

The comparison of the possible compressional and
extensional suites with the type examples has shown that
certain inconsistencies exist for the possible suites.
These inconsistencies suggest that the petrogenesis of
these suites departs from the petrogenesis of the type
examples.

The Boulder Batholith is consistent with type
compressional suites except that is slightly more
alkalic. This may be due to the variation in the source
rock that was melted to produce the suite (Kistler,
1974). Alternatively, the difference may be related to
the depth of the subduction zone, in a manner analogous
to that suggested by Dickinson and Hatherton (1967) for
the variation in K20. The Boulder Batholith is further
to the east ("inland") of the postulated plate margin
than the other western U.S.A. batholiths.

The Hercynian granites and the South Mt. Batholith
are consistent with type compressional suites except for
the predominance of silica rich and alumina oversaturated
rocks. Two alternatives may be considered to account for
these differences. First, Paleozoic subduction may
have been fundamentally different from Mesozoic-Cenozoic
subduction. Second, the nature of the differences

suggests a larger crustal component in their petrogenesis.

35

McKenzie and Clarke (1975) describe the South Mt.
Batholith as post—tectonic, but generated in response to
the Acadian orogeny. This batholith may have formed by
predominately crustal fusion caused by crustal thicken-
ing and heating due to continental collision. Both of
these hypotheses remain to be tested.

The White Mt. Magma Series and the Younger Granites
of Nigeria are consistent with type extensional suites
except that the acidic modes predominate over the basic
modes in the relative frequency distributions of DI and
PL. In the case of the White Mt. Magma Series, the
basic rocks may lie concealed at depth (Chapman, 1976).
The Nigerian suite has no appreciable amount of basic
rocks associated with it (Wright, 1970). The lack of
basic rocks in these suites suggests that mantle derived
magmas may have solidified at the mantle-crust interface
where the granitic magmas were generated. The rise of
these granitic magmas produced predominately acidic
suites. An alternative is that the granitic magmas
were generated by heat flow from the mantle without the
emplacement of mantle derived magmas. Younker (1974)
concluded that heat flow by conduction without penetra-
tion by mantle derived magmas is not a likely source of

crustal melting.

CHAPTER IV

CONCLUSIONS

The major element chemistry of plutonic suites

indicates some trends that may be used to distinguish

compressional and extensional plutonic suites. The

chemical criteria that most clearly show the differ-

ences between the two tectonic types of suites are:

l)

2)

3)

Frequency distributions of DI and PL show
that unimodal (intermediate) distributions
are characteristic of compressional suites,
whereas bimodal (acidic-basic) distribu-
tions are characteristic of extensional
suites. Unimodal acidic distributions

may be ambiguous.

Alkali-lime index (ALI) values for exten-
sional suites are in the range 50-56,
whereas ALI values for compressional
suites are in the range 60-64. The
intermediate range (56-60) may be ambi-
guous.

AFM ternary diagrams show that exten-
sional suites have stronger iron enrich-
ment trends than do compressional suites.

36

37

4) Peralkaline rocks are characteristic of
extensional suites, whereas compressional
suites tend to have higher frequencies of
peraluminous rocks. Metaluminous rocks
are common in both suites. Differences
in alumina saturation appear to be more
controlled by CaO and Nazo + K20 than by
A1203.7

The use of single criteria to determine tectonic
setting may give ambiguous results. Several criteria
should be used, as consistent results from several
indicators assure the best estimate of the tectonic
setting of a plutonic rock suite. Conflicting results
from different criteria suggest that the petrogenesis
of the suite is different from those suites selected as
type examples. Thus, these criteria may also be valuable
in determining areas of unique petrogenesis.

Ambiguous results may also occur because of the
lack of proper sampling. Frequency diagrams can only
work when the total areal distribution is evaluated.
Variation diagrams are successful only if the complete
compositional spectrum is sampled.

There are several petrogenetic differences
between extensional and compressional suites. Com-
pressional suites have more intermediate rock types due

to more extensive melting of the continental crust (or

38

arc complex in island arcs). Extensional suites have
acidic—basic associations due to limited melting and
mixing. Insome areas, extensional plutonic suites

appear unimodal due to the lack of rise of mantle derived
magmas from the mantle-crust boundary, where granitic
magmas are generated to eventually form predominately
acid complexes.

Extensional suites are more alkalic than compres-
sional suites due to the depth of origin of the basic
magmas and the small degrees of melting of the crust to
form acidic magmas. Syenitic rocks are most likely
formed by fractionation of the basic magmas, whereas
granitic rocks are most likely formed by melting of
crustal rocks. Iron differences are probably due to
the higher water pressures and degrees of melting in
compressional areas, which produce less iron enrich-
ment than in extensional areas, where water pressures

and degrees of melting are lower.

REFERENCES

39

REFERENCES

Anderson, A.T. "Magma mixing: Petrological Process and
Volcanological Tool," J. Volc. Geotherm. Res., 1,
3-33, 1976.

 

Anderson, R.N., Uyeda, S., and Miyashiro, A. "Geophysical
and Geochemical Constraints at Converging Plate
Boundaries - Part I: Dehydration in the Down-
going Slab," Geophys. J. Roy. Astron. Soc., 44,
333-357, 1976. -_

 

Barth, T.F.W. Theoretical Petrology, Second Edition,
416 pp., John Wiley & Sons, New York, 1962.

 

Boettcher, A.L. "Volcanism and Orogenic Belts - The
Origin of Andesites," Tectonophysics, 11,
223-240, 1973.

 

Bowden, P. "Origin of the Younger Granites of Northern
Nigeria," Contrib. Mineral. Petrol., 25, 153-162,

1970.

Brooks, C.K. "Tertiary of Greenland - A Volcanic and
Plutonic Record of Continental Break-up," in
Arctic Geology, edited by M.G. Pitcher, Amer.
Assoc. Petrol. Geol. Mem., 19, 150-160, 1973a.

 

 

Brooks, C.K. "Rifting and Doming in Southern East
Greenland," Nature Phys. Sci., 244, 23-25, 1973b.

 

Brown, G.C. "A Comment on the Role of Water in the
Partial Fusion of Crustal Rock," Earth Planet.
SCio Lett., 9’ 355-358, 1970.

 

Brown, G.C. "Evolution of Granite Magmas at Destructive
Plate Marginsﬂ'Nature Phys. Sci., 241, 26-28,
1973. ’

 

Brown, G.C., and Fyfe, W.S. "The Production of Granitic
Melts During Ultrametamorphism," Contrib.
Mineral. Petrol., 28, 310-318, 1970.

 

Burke, K.C., and Wilson, J.T. "Hot Spots on the Earth's
Surface," Sci. Amer., 235, 46-57, 1976.

 

40

41

Bussell, M.A., Pitcher, W.S., and Wilson, P.A. "Ring
Complexes of the Peruvian Coastal Batholith: A
Long-standing Subvolcanic Regimef'Can. J. Earth
833., 18, 1020-1030, 1976.

 

Carmichael, I.S.E., Turner, F.J., and Verhoogen, J.
Igneous PGtrOIOQXJ 739 pp., McGraw-Hill, New
York, 1974.

 

Chapman, C.A. "Structural Evolution of the White Moun-
tain Magma Series," Geol. Soc. Amer. Mem., 146,
281-300, 1976.

Chase, C.G., and Gilmer, T.H. "Precambrian Plate Tec-
tonics: The Midcontinent Gravity High," Earth
Planet. Sci. Lett., 81, 70-78, 1973.

 

Christiansen, R.L., and Lipman, P.W. "Cenozoic Volcanism
and Plate-tectonic Evolution of the Western
United States. II. Late Cenozoic," Phil.
Trans. Roy. Soc. London, Ser. A, 271, 249-284,
1972.

Dewey, J.F., and Burke, K.A. "Tibetan, Variscan, and
Precambrian Basement Reactivation: Products of
Continental Collisionf'J. Geol., 8;, 683-692,
1973.

Dickinson, W.R. "Relations of Andesites, Granites, and
Derivative Sandstones to Arc-trench Tectonics,"
Rev. Geophys. Space Phys., 8, 813-860, 1970.

Dickinson, W.R. "Plate Tectonic Models for Orogeny at
Continental Margins/'Nature, 232, 41-42, 1971.

Dickinson, W.R., and Hatherton, T. "Andesitic Volcanism
and Seismicity Around the Pacific," Science, 157,
801-803, 1967.

Dunham, A.C., and Thompson, R.N. "The Origin of Granitic
Magmas - Skye and Rhum," Geol. Soc, Australia J.,
13, 339-343, 1967.

Eichelberger, J.C. "Origin of Andesite and Dacite:
Evidence of Mixing at Glass Mountain in Cali-
fornia and at Other Circum-Pacific Volcanoes,"
Geol. Soc. Amer. Bull., 88, 1381-1391, 1975.

 

Einarsson, T. "Geology of Iceland," in Arctic Geology,
edited by M.G. Pitcher," Amer. Assoc. Petrol.
6901. Mem., L2, 171-175, 19730

 

42

Fitton, J.G. "The Generation of Magmas in Island Arcs,"
Earth. Planet. Sci. Lett., ll, 63-67, 1971.

 

Fyfe, W.S. "The Generation of Batholiths," TectonOphysics,
ll, 273-283, 1973.

 

Gilluly, J. "Plate Tectonics and Magmatic Evolution,"
Geol. Soc. Amer. Bull., 88, 2383-2396, 1971.

 

Hamilton, W. "The Volcanic Central Andes - A Modern
Model for the Cretaceous Batholiths and Tectonics
of Western North America," in Proceedings of the
Andesite Conference, edited by A.R. McBirney,
Oregon Dept. GeBIOgy and Mineral. Bull., 65, 175-
184, l969a. ‘—

 

 

 

Hamilton, W. "Mesozoic California and the Underflow of
the Pacific Mantle," Geol. Soc. Amer. Bull.,
80, 2409-2430, 1969b.

 

Huang, W.L., and Wyllie, P.J. "Melting Relations of
Muscovite-Granite to 35 kbar as a Model for
Fusion of Metamorphosed Subducted Oceanic
Sediments," Contrib. Mineral. Petrol., 42,
1-14, 1973. '—

 

Jakes, P., and White, A.J. "Composition of Island Arcs
and Continental Growth," Earth Planet. Sci.
Lett., 18, 224-230, 1971.

 

Jakes, P., and White, A.J.R. "Major and Trace Element
Abundances in Volcanic Rocks of Orogenic Areas,"
Geol. Soc. Amer. Bull., 83, 29-39, 1972.

Johnston, J.S., Mutschler, F.E., and Barr, D. "Granites
and Granites - A Statistical Investigation of
Paleo Stress Gauges," Geol. Soc. Amer. Abstr.
Programs, 8, 943, 1976.

 

Karig, D.E. "Origin and Development of Marginal Basins
in the Western Pacificﬂ'J. Geophys. Res., 18,
2542-2561, 1971.

 

Kistler, R.W. "Phanerozoic Batholiths in Western North
America: A Summary of Some Recent Work on
Variations in Time, Space, Chemistry, and Iso-
topic Compositions," Ann. Rev. Earth Planet.
8gi,, 8, 403-418, 1974.

43

Lanphere, M.A., and Reed, B.L. "Timing of Mesozoic and
Cenozoic Plutonic Events in Circum-Pacific North
America," Geol. Soc. Amer. Bull., 88, 3773-3782,
1973.

 

Larsen, R.S., Jr. "Batholith and Associated Rocks of
Corona, Elsinore, and San Luis Rey Quadrangles,
Southern Californiaﬂ'Geol. Soc. Amer. Mem., 29,
182 pp., 1948. ‘—

 

MacIntyre, R.M. "Anorogenic Magmatism, Plate Motion,
and Atlantic Evolution," J. Geol. Soc. London,
133, 375-384, 1977.

 

Marsh, B.D., and Carmichael, I.S.E. "Benioff Zone Magma-
tism," J. Geophys. Res., 18, 1196-1206, 1974.

 

Martin, R.F., and Piwinskii, A.J. "Magmatism and Tec-
tonic Settings," J. Geophys. Res., 81, 4966-
4975, 1972.

 

McKenzie, C.B., and Clarke, D.E. "Petrology of the South
Mountain Batholith, Nova Scotia," Can. J. Earth
Sci., 88, 1209-1218, 1975.

 

Miyashiro, A. "Volcanic Rock Series and Tectonic Set-
ting," Ann. Rev. Earth Planet. Sci., 8, 251-
269, 1975.

 

Morgan, W.J. "Convection Plumes in the Lower Mantle,"
Nature, 230, 42-43, 1971.

 

Morgan, W.J. "Deep Mantle Convection Plumes and Plate
Motions," Amer. Assoc. Petrol. Geol. Bull.,
88, 203-213, 1972.

 

Mutschler, F.E., Rougon, D.J., and Lavin, O.P. "Petros--
A Data Bank of Major-Element Chemical Analyses of
Igneous Rocks for Research and Teachingﬂ'ggm-
puters and Geosciences, 8, 51-57, 1976a.

 

Mutschler, F.E., Rougon, D.J., and Lavin, O.P. "Petros:
A Data Bank of Major-Element Chemical Analyses of
Igneous Rocks for Research and Teaching (version
2.1)R'Dept. Geol., Eastern Washington State
College, Cheney, Washington, 1976b.

Naylor, R.S. "Acadian Orogeny: An Abrupt and Brief
Event," Science, 172, 558-560, 1971.

44

Nie, N.H., Hull, C.B., Jenkins, J.G., Steinbrenner, K.,
and Bent, D.H. SPSS: Statistical Package for
the Social Sciences, Second Edition, 675 pp.,
McGraw-Hill? New York, 1975.

 

 

Oxburgh, E.R., and Turcotte, D.L. "Origin of Paired
Metamorphic Belts and Crustal Dilation in Island
Arc Regions," J. GeOphys. Res., 18, 1315-1327,
1971.

 

Peacock, M.A. "Classification of Igneous Rock Series,"
90 G801o' 3—9-’ 54-67, 19310

Pearce, J.A., and Cann, J.R. "Tectonic Setting of Basic
Volcanic Rocks Determined Using Trace Element
Analyses," Earth Planet. Sci. Lett., 88, 290-
300, 1973.

 

Presnall, D.C., and Bateman, P.C. "Fusion Relations in
the System NaAlSi3Og-CaA128i203-KAlSi308-Si02-H20
and Generation of Granitic Magmas in the Sierra
Nevada Batholithﬂ'Geol. Soc. Amer. Bull., 84,
3181-3203, 1973. '—

Reed, B.L., and Lanphere, M.A. "Alaska-Aleutian Range
Batholith: Geochronologyr Chemistry, and Rela-
tion to Circum-Pacific Plutonism," Geol. Soc.
Amer. Bull., 88, 2583-2610, 1973.

 

 

Riding, R. "Models of the Hercynian Foldbeltﬂ'Earth
Planet. Sci. Lett., 88, 125-135, 1974.

 

 

Ringwood, A.E. Composition and Petrology of the Earth's
Mantle, 617 pp., McGraw-HiIl, New Yofk, 1975.

Sawkins, F.J., Chase, C.G., Darby, D.C., and Rapp, G.,
Jr. The Evolving Earth: A Text in Physical
Geology, 477—pp., MacMillan, New York, I974.

Shand, S.J. The Ergptive Rocks, 488 pp., Wiley, New
York, 1927.

 

Sims, P.K. "Precambrian Tectonics and Mineral Deposits,
Lake Superior Region," Econ. Geol., 18, 1092-
1127, 1976.

 

Snedecor, G.W., and Cochran, W.G. Statistical Methods,
Sixth Edition, 593 pp., Iowa State Univ. Press,
Ames, 1974.

45

Strong, D.F., and Minatidis, D.G. "Geochemistry and
Tectonic Setting of the Late Precambrian Holyrood
Plutonic Series of Eastern Newfoundlandﬂ'Lithos,
8, 283-295, 1975.

Thornton, C.P., and Tuttle, O.F. "Chemistry of Igneous
Rocks: I. Differentiation Indexﬂ'Amer. J. Sci.,
258, 664-684, 1960.

 

Turcotte, D.L., and Schubert, G. "Frictional Heating of
the Descending Lithosphere," J. Geophys. Res.,
18, 5876-5886, 1973.

Tuttle, O.F., and Bowen, N.L. "Origin of Granite in
Light of Experimental Studies in the System
NaAlSi3Og-KAlSi3O8-Si02-H20," Geol. Soc. Amer.
888., 18, 153 pp., 1958.

Wager, L.R., and Brown, G.M. Layered Igpeous Rocks,
588 pp., Freeman, San Francisco, 1967.

 

Winkler, H.G.F. Petrogenesis of Metamorphic Rocks,
Fourth Edition, 334 pp., Springer-Verlag, New
York, 1976.

Wright, J.B. 92: "Origin of the Younger Granites of
Northern Nigeria," by P. Bowden, Contrib.
Mineral. Petrol., 88, 89-90, 1970.

 

Wyllie, P.J. "Role of Water in Magma Generation and
Initiation of Diapiric Uprise in the Mantle,"

 

Wyllie, P.J., and Huang, W.L. "High C02 Solubilities in
Mantle Magmas," Geology, 8, 21-24, 1976.

Wyllie, P.J., Huang, W.L., Stern, C.R., and Maal¢e, S.
"Grantic Magmas: Possible and Impossible Sources,
Water Contents, and Crystallization Sequences,"
Can. J. Earth Sci., 88, 1007-1019, 1976.

 

Yoder, H.S., Jr. Generation of Basaltic Magma, 265 pp.,
National Academy of Sciences, Washington, D.C.,
1976.

Younker, L.W. "Role of Mantle Derived Magmas in the
Production of Crustal Melts," Ph.D. thesis, 72 pp.,
Michigan State Univ., E. Lansing, MI, 1974.

Younker, L.W., and Vogel, T.A. "Plutonism and Plate
Dynamics: The Origin of Circum-Pacific Batho-
liths," Can. Minera1., 88, 238-244, 1976.