v—“fl w—w—i—w wiv

 

 

SEDIMENTS DERIVED FROM THE WEATHER‘ING‘ ,

OF A PRIMORDIAL ANORTHOSITIC CRUST

Thesis for the Degreeof M. S.
MICHIGAN STATE UNIVERSITY
THOMAS ALVIN TEST

1976‘ ‘

 

 

‘gliflmw 49¢

’. W" a, . _
I AIME-2121’

I

y.

  
  
    
 

O

_ W
. "'51 ‘= New
L.” d... \Ifi“

L ”V ' I

i

r. .-h-
., .~
«I - o‘.‘¢'

b“ a

0‘3“

 

-- f x“ ”I'D"
, z

’1‘,“ “Y. Q .mv‘ :
94¢“, .. -‘.{ .', L 44—1;-

  

q
-' v’l' *“4- ‘
tr {‘ H . .' "' ”u- ’ '

..

 

6702/.4/

ABSTRACT

SEDIMENTS DERIVED FROM THE WEATHERING
OF A PRIMORDIAL ANORTHOSITIC CRUST

by

Thomas Alvin Test

The sedimentary suite that would be derived from the weathering of
a primordial anorthositic crust is estimated using the mass balance equa-
tion: Primordial Crust + Primitive Sea = Sediments + Seawater. The de-
rived suite contains approximately 55% potassium-free claystones, 23%
limestone, 20% aluminum-rich sediment and one to two percent each evap-
orites and carbonate-rich iron formation. This suite does not resemble
modern sediments, nor does it resemble ancient sediments of appropriate
age and tectonic setting to represent the weathering products of a prim-
ordial continental crust. Lithologies of these oldest rocks of sediment-

ary origin suggest a primordial crust of more granitic composition.

SEDIMENTS DERIVED FROM THE WEATHERINC
OF A PRIMORDIAL ANORTHOSITIC CRUST

By

Thomas Alvin Test

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

ACKNOWLEDGMENTS

I am grateful to Duncan Sibley, my thesis committee chairman and
advisor, for suggesting this problem and then allowing me to develop it
myself. I thank him for his guidance and assistance when I asked for it,
and for the independence I had when I wanted it. I am also grateful to
Tom Vogel and John Wilband for their guidance and constructive criticism
of the manuscript, and I thank John for helping me learn to communicate
with the computer.

I am especially grateful to my wife Sandy, for her understanding
and support, as I know I was quite hard to live with while tied up with
this project, and to my daughter Jennifer, who didn't understand, but
was patient none the less. I also thank the rest of my family, and my
wife's family, for their support of my educational goals at a time in

life when most men have already completed their formal educations.

ii

Acknowledgments . . . . .
List of Tables . . . .

Introduction . . . . .

The Chemical Mass Balance Test

Choice of Materials for the Mass Balance Calculations

Results of Calculations .

TABLE OF CONTENTS

Effects of Sedimentary Recycling . . .

Discussion of Results of Calculations

Comparison of Sediment Derived from Primordial
Anorthositic Crust by Mass Balance with Ancient
sedimentary ROCkS O O O O O O O C O C O O C O O O O 0

Effects of a Primordial Anorthositic Crust on
Sedimentary Rock Compositions through Geologic Time

Conclusions . . . . . . .
References . . . . . . . .

Appendix:
with an Anorthositic Crust

iii

Sedimentary Suites that Do Not Balance

Page
ii

iv

. 17

25

28

40

42

46

48

55

Table

LIST OF TABLES

Component preportions of sedimentary rock types used in

mass balance calculations . . . . . . . . . r . . . . . . .
Compositions of anorthosites and Lunar crustal materials . .

Calculation of maximum neutralizing capacity of

anorthositic primordial crust . . . . . . . . . . . . . . .

Component proportions of primordial anorthositic crust plus
primitive acid sea mixtures used in chemical mass balance

caICUIationS O O O O O O O O O O O O O O O O O O O I O O O O

5-—10. Results of sedimentary chemical mass balance calculations

11.

12.

13.

14.

15.

16.

17.

Component proportions of granodiorite plus primitive acid
sea mixtures used in analytical sedimentary chemical mass

balance calculations . . . . . . . . . . . . . . . . . . . .

Component prOportions of andesite plus primitive acid sea
mixtures used in analytical sedimentary chemical mass

balance calculations . . . . . . . . . . . . . . . . .

Component proportions of tholeiitic basalt plus primitive
acid sea mixtures used in analytical sedimentary chemical

mass balance calculations . . . . . . . . . . . . . . . . .

Results of sedimentary chemical mass balance calculation,

granodiorite crust . . . . . . . . . . . . . . . . . . . . .

Results of sedimentary chemical mass balance calculation,

andesite crust . . . . . . . . . . . . . . . . . . . . . . .

Results of sedimentary chemical mass balance calculation,

tholeiitic basalt crust . . . . . . . . . . . . . . . . . .

Estimate of mass and composition of new igneous material
that must be added to anorthosite-derived sediments to

achieve average Precambrian sedimentary composition . . . .

iv

Page

12

15

16

19-24

34

35

36

37

38

39

43

INTRODUCTION

The early history of the earth's crustal development persists as
an unresolved question in geology. Evidence of the nature and evolution
of the early crust is infrequent and complex, so that models describing
a primordial crust are as varied as they are numerous. One of these
models is examined here.
The Model

The model investigated in this study is one of an anorthositic

primordial crust, as proposed by D. M. Shaw (1975).

"Following accretion, core separation and ensuing cool-
ing must have been completed in about 0.7 By (earth age 4.55
By) in order for rocks dated at 3.7-3.9 By (Greenland, Minn-
esota) to develop and be preserved.

"This early history is divided into Prearchean and

Procoarchean stages. The first comprised the major earth

 

differentiation into core, peridotitic lower mantle, and
basalt upper mantle: this stage terminated in the develop-
ment of a thin worldwide anorthositic crust. It was followed
by condensation of hot acid rains, draining to form shallow
intracontinental seas.

"Mantle convection aided rupture of crustal flakes
along thermal buckles. Archean greenstone belts appear to
represent weld sutures, building wider and thicker continental
masses, complemented by wider and deeper basins, leading to

present-day plate tectonics.’

The early differentiation of the earth and the formation of an ig-
neous global primordial crust is strongly suggested by interplanetary
analogy, while the direct terrestrial evidence, mostly isotopic, remains
ambiguous (Lowman, 1976).

An anorthositic composition for the primordial crust of the earth
is suggested by analogy to lunar crustal composition. Windley (1970)

I
has proposed that certain layered calcic anorthosites in very early
Archean high-grade metamorphic terrains, which are chemically quite sim-
ilar to lunar highland anorthosites, may be tectonically emplaced reworked
fragments of a primordial continental crust. No ancient igneous rocks
have been located, however, that can be convincingly identified as rem-
nants of the earth's first crust.

Other lines of evidence indicate that the composition of the
earth's first crust was probably not anorthositic. As older and older
igneous rocks are identified, they are found to be both widespread in
cratonic areas and granitic in composition (Hart and Goldich, 1975).

This suggests the primordial material might be better characterized as
acidic or intermediate in composition. The oldest yet known igneous
rocks, in Greenland, range in composition from granite to quartz diorite
(McGregor and Bridgwater, 1973; and Bridgwater et a1, 1974). These in-
trude yet older rocks of sedimentary origin, which may provide better
clues as to the composition of the primordial crust.

A Sedimentological Approach

The setting provided by the model suggests that the question of the
composition of the primordial crust may be investigated from a sedimento-
logical approach, in lieu of direct evidence from remnants. The inter-

action of shallow intracontinental acid seas with a bare igneous crust

would result in rapid weathering until the seas were neutralized to a
chemistry approximating present seawater. The weathered material would
be deposited as sediments of types in approximate chemical equilibrium
with the neutralized seas. If the primordial crust was not totally des-
troyed by subduction, some portion of these first continental sediments
could be preserved, and would indicate the nature of their primordial
source. They would at least impart something of their chemical nature
to subsequent sedimentary rocks.

The oldest known rdcks on earth are 3700 to 3900 million year old
metasediments on the Greenland craton (Moorbath et a1, 1973). According
to the timetable of Shaw's model, these are of sufficient age to have
been derived from the weathering of the first igneous crust, and their
existence suggests a sedimentological approach is reasonable.

The approach used in this examination of the suggested anorthositic
primordial composition is one of chemical mass balance between the sedi-
ments and their source rocks. In principle, it is an application of the
law of conservation of mass. When a mass of igneous rock of specified
composition is weathered to produce an assortment of sedimentary rocks,
the chemical composition of the total mass of sediments must be the same
as the composition of the source material, less the constituents that,
have remained in solution.

When the chemical composition of calcic anorthosites is compared
to estimates of the average composition of sedimentary rocks on earth,
or to estimates of the earth's average crustal composition, some quali-
tative differences are apparent. The ratio of 810 to Al O in anortho-

2 2 3
sites is on the order of 1.6 to 1, while this ratio is approximately

4.3 to 1 for averages of either sedimentary rocks or the crust as a
whole. In addition to this substantial excess of aluminum with respect
to silicon when compared to average sediments, anorthosites have a sub-
stantial deficiency of potassium and an excess of calcium. These dis-
crepancies are confirmed by the mass balance calculations in this study.
The chemical mass balance calculations for this study consider the
weathering of an anorthositic crust by primitive acid seas to produce
common sedimentary rock types. The calculations include the chemical
compositions of the initial and final seas, in addition to those of the
rocks, in order to include in the mass balance those constituents from
the weathered rocks that remain dissolved in the seas, and those comp-
onents now in sediments that would have originated with the volatiles

that formed the primitive seas. The balance equation is as follows:

Primitive Sea + Primordial Crust = b1 Sediment type 1 +

b2 Sediment type 2 + . . . + bk Seawater.

The coefficients for the sedimentary rock types derived from this calcu-
lation provide an estimate of the proportions of the sedimentary rock
types that would result from the weathering of the specified primordial
crustal material. This is a suite of sedimentary lithologies that may
be compared to ancient sedimentary suites to evaluate the specified
crustal material as their source.

This investigation also considers the effect of sedimentary re-
cycling on the sediment derived from a primordial crust. Studies of ages
and amounts of exposed sedimentary rocks indicate that a substantial
proportion of the sediments deposited at any time are derived from prev-

ious sedimentary rocks, and that the older a sedimentary rock, the less

remains of the mass of material originally deposited (Carrels and MacKen-
zie, 1971; Garrels et a1, 1972; and Blatt and Jones, 1975). This obser-
vation has some implications important to the present study.

The rate of sedimentary recycling will determine the probability
of finding a very ancient sedimentary rock. Estimates of recycling rates
predict that sedimentary rocks of near primordial age may still exist,
but would be rare.

Differential rates of recycling, whereby more soluble types of
sedimentary rocks are less well preserved, are are weathered and redep-
osited more frequently, would alter the proportions of rock types in
sediments of a given age. It may be inferred that older sedimentary
rocks may have originally had greater proportions of the more soluble
lithologies, such as evaporites and carbonates, than are represented by
their modern exposures. The effects of differential recycling are gener-
ally predictable.

Sedimentary recycling will also impart some of the chemical charac-
ter of earlier sedimentary rocks to subsequent sediments. This effect is

evaluated in this study.

THE CHEMICAL MASS BALANCE TEST

The sedimentary suites that would be derived from the weathering of
a primordial anorthositic crust are estimated with a chemical mass balance
equation as follows:

Primitive acid seas + Primordial Crust = b Sediment type 1 +

1

b2 Sediment type 2 + . . . + bk Seawater.

The primordial crust, primitive sea, sedimentary rock types, and final

seawater are characterized by the mass proportions of their chemical com-
ponents, so that there is a linear balance equation for each component
considered. A least squares regression procedure is used to estimate

the b coefficients for the sedimentary rock types from the set of linear
balance equations, so that the masses of the components on each side of
the equation are balanced. This estimate can be compared to ancient
sedimentary suites that may represent the weathering products of the pro-
posed primordial crust.

Method of calculation.

The method used to calculate this mass balance is the least squares
approximation method, using high speed computers, developed by Bryan et
al (1968) for estimating proportions in petrographic mixing equations.
This method has been applied to sedimentary chemical mass balance equat-
ions by Sibley and Vogel (1976).

The computer program is a multiple regression procedure whereby
the b values, the least squares estimates of the regression coefficients,
in this case the coefficients for the sedimentary rock types, are ob-
tained from a set of equations X'Xb = X'Y by matrix manipulation such
that b = (X'X)‘1(X'Y). The method requires that the system analysed has
a number of independent equations (number of components characterizing a
rock type), n, equal to or greater than the number of parameters to be
estimated (number of b coefficients to be generated), k. This study
meets this requirement for all estimates calculated.

The program places an additional constraint on the system in the
least squares solution of the predicted value of a constituent in the
parent, or source rock, 1, from the regression equation:

Y = A + b1X1 +b2X2 + . . . + kak.

This constraint consists of forcing the equation through the origin, so
that the intercept A is zero. This is achieved by eliminating the first
row and column of the X'X matrix and the first terms of the b and X'Y
column vectors to form reduced matrices.

While predictions of small numbers are more error prone using re-
duced matrices than using nonreduced matrices, and care is required in
the interpretation of the significance of small numbers produced, a test
of the method by comparison with regressions calculated using nonreduced-
matrices (Sibley and Wilband, 1977) indicates that the Bryan et al method
is as good a predictor of b values (regression coefficients) as more con-
ventional regression techniques so long as the values of the sums of
squares of residuals (SSR) are low. Forcing the equations through the
origin does not significantly alter the b values for those regressions
providing a good fit, whether derived on reduced or nonreduced matrices.
The geological ambiguity implied by nonzero intercepts is avoided when

the Bryan et al method is used.

CHOICE OF MATERIALS FOR THE MASS BALANCE CALCULATIONS
Sedimentary rock types.

Where possible, the compositions for the sedimentary rock types
were obtained from widely available published data, with a preference
for data previously used in chemical mass balance calculations. The chosen
pelagic red clay was selected for its low carbonate content to represent
pre-Mesozoic pelagic sediments. Clarke's (1924) average shale, average
Paleozoic shale, and average Cenozoic-Mesozoic shale were selected in or-

der to independently evaluate the balance with chemically different shale

types. Averages of Grout's (1933) analyses of Knife Lake lutites and
graywackes were selected to determine possible balance differences with
Archean sediments; these are well described geologically and are among
the oldest on the North American shield. Smectite analyses were obtained
from Grim (1968), to represent low potassium clays. Iron formation comp-
ositions were selected from Archean types to provide a range of variation
of composition.

The chemical compositions for the materials used in the balance
calculations are recalculated so that all iron species appear as grams
elemental iron per 100 grams rock. Similarly, all carbon species are re-
Mass balance can be

calculated as grams CO , and sulfur species as SO

2 3'
thus calculated for these elements without concern for the evolution of
their oxidation states through geological time as a result of biological
and chemical processes. Compositions are not recalculated to 100%. The
tabulated component proportions represent the number of grams of the
component in the tabulated form that would be derived frOm all forms of
the material in 100 grams of rock.

Elements such as manganese and phosphorous, which occur in very
small proportions in all the materials used in the balance calculations,
were not used in the calculations. Their abundances are so small that
their contributions to the predicted coefficients are insignificant rel-
ative to the effect of the major constituents.

Component proportions for the sedimentary rock types used in the
mass balance calculations are listed in Table 1.

Final seawater.

The final seawater composition used for the mass balance calcul-

ations is that of modern seawater. Some investigators have estimated

8

00~00 .wxum0ov m0~£m o0ouom0wm owmuw>m “:0

00000 .Cm0xmusev uwumzmmm cuovos u<mm A¢No0

00000 .50000 mwu0couucoa usom mo wwm0w>m "9202 .0000000 m0m:m o0oNOmwz-o0ouoamo wwmum>m "20:0
00000 .E0uuv 000000000 03» mo owmum>m 0040 A¢~00 .mxum0ov @0050 owmuo>m 0><mm

00000 080000 mmu0:0000uoeuaoe 0 0o mwmuo>m “9202 00000 .uaouuv mwxoms
00000 .00000 0cm 000x03 0mv 0000 000 o0ww0m0 000m ->muw oxm0 0000M ammsou< mounu 0o mwmuw>w ”0M30
00000 .usouov Am000 .cnom0uuomv oxomshmuw mwmum>w "30

000050 mxm0 000cx cmwzo0< mwunu mo mmmuo>w "0x10 am000 .c5000uuwmv chumucmm wwwum>w "mm
--- 00.0 --- 00.0 00.0 00.0 0.0 0.0 0.0 0.0 0.0 0.0 0000
00.0 00.0 --- 00.0 00.0 00.0 0.0 0.0 0.0 00.0 0.0 0.0 000
00.0 --- --- --- --- 00.0 0.0 0.0 0.0 00.0 00.0 0.0 000
00.0 -u- -n -n -n -- n- -- -u- un- nn- -- 00
00.0 --- --- --- 0.0 --- 0.0 0.0 0.0 0.0 00.0 0.0 000
0.00 00.00 00.00 00.00 0.0 00.0 0.0 0.0 0.0 0.0 0.0 0.0 00:
00.0 00.0 -- 00.0 0~.0 00.~ 0.0 0.0 0.0 0.0 o.~ «.0 0002
00.0 00.0 mm.0 00.0 00.0 0.~ 0.0 0.0 0.0 0.0 0.0 0.0 000
00.0 00.0 0.00 00.0 00.0 0.0 m.~ 0.~ 0.0 0.0 0.~ «.0 0m:
-- 00.00 00.0 00.~ 0.0 0.0 0.0 m.0 0.0 0.0 0.0 0.0 mm
--- 00.0 0.00 00.00 00.00 0.00 0.00 0.00 0.00 0.00 0.00 0.0 00000
--- 0.00 0.00 00.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0000
<00 Hzoz m<m 0202 0mm 0xzm :0 20:0 ><rm 0x30 30 mm pamcoasoo

.mco0um0so0ao oocw0wn 0008 :0 0mm: moaxu xoou xumucmE00mm mo 00000000000 uanoQEoo .0 00000

xoou no00-Esc0850m 0wo0umnuoazs uncug<
0400 .0000 .00030000 0000 000000 00<x

.wxpm0o m0- .m000 .uumzmuwv xoou 000000;:0

00000
":20

A0000 .mw0hmmv xoou ou0Eo0oc 0000
A0000 .mxum0uv 000000500 mwmuo>m 000
00000 .000000 0000
-um00 00000: .xoon 000muOmmca0Enou00wawpm uu<9

00.0
mm.o
0N.o
No.0
00.00
00.0
00.0
00.~0
0.0
00.0
00.0
00.0
00

van 0000 .mxu0aov m0Ewow0EmB .uuoconmu0umawwe
00000.00a200v 00000000 0nmmwz .uuwso-ou0uwawae

00.0
0.0

00.0
0.0
0.0m

00.0
0.00
U<H

00000 .0030000 00000 .000: 00> 0:0 0c0>000

uo0uum00 uc00maso .a00um8000 c000 oumaonumo

0000
00000 .00 00 000000
”0002
0000:
00000 .umnamv uo0uum0c

o0nwwoo .co00050om co00 wu0uwcwme 00o<z

uum00 me0Eocmz .500umE000 c000 mu0um8m£

-n 00.0
0~.o -n-
00.00 n.0N
00.0 -u

0.0 0.0

0.m 0.m

0.00 m.mm
0N.o 0m.0

0.00 0.-

9H0 00¢:

.00000..wwsmhv u00u

":0:
00.0 --- 0000
00.0 --- 000
00.0 00.0 000
--- --- 00
0.0 0.0 000
00.0 --- 00:
--- --- 0002
.0.0. 0.0 000
0.0 0.0 00:
0.00 0.00 00
0.0 0.0 00000
0.00 0.00 0000
00¢: 2m: uanOQEoU

0m0n0ucoo .0 w0nme

10

that the relative proportions of some of the metal ions in seawater may
have changed through geologic time, and that its total salinity may have
been different by a small factor. These differences would have been
greater in the early Precambrian than in the Phanerozoic (Ronov, 1968;
and Lafon and MacKenzie, 1971). The choice of an alternate to modern
seawater as a final seawater composition in the mass balance calculations,
however, would only slightly alter the results. The mass of the rock-
forming components in any reasonable estimate of seawater composition is
small compared to their mass in the sediments. Use of an alternative
composition for the final seawater to achieve a better balance would at
best transfer observed discrepancies in the balance to a probable error
in the estimate of the composition of the alternative. As the differences
between the composition of modern seawater and estimates of earlier sea-
water compositions are approximately the same as the uncertainties in
those estimates, it is preferable for the purposes of this study to use
the known composition of modern seawater for the final seawater composi—
tion in the mass balance calculations. This choice prevents mass balance
discrepancies from being hidden in the probable error in the estimate of
the final seawater composition.

The component proportions of modern seawater are listed in Table
1.
Anorthositic crust.

The compOsition for the primordial crust is estimated from analyses
of anorthosites that have been suggested to be remnants of a primordial
anorthositic crust (Windley, 1970). The composition used in the mass

balance is an average anorthosite composition in column D, Table 2.

11

Analyses of composite materials considered representative of the Lunar
highlands (Lowman, 1976) are tabulated for comparison, as the analogy
with the Lunar primordial crust is implicit in the anorthositic crustal
model. The choice of ancient terrestrial anorthosites for the mass bal-
ance calculations rather than lunar materials is in recognition of the
probable differences in the overall chemical compositions of the earth

and the Moon.

Table 2. Compositions of anorthosites and Lunar crustal materials

Component A B C D E F G
SiO2 46.77 46.64 46.86 46.76 45.2 45.1 47.9
TiO2 0.33 0.15 0.03 0.17 1.0 0.6 1.7
A1203 27.43 25.53 33.05 28.67 21.7 26.7 17.6
Fe203 1.59 0.70 0.46 0.92 --- 0.0 0.0
FeO 3.17 4.41 0.48 2.69 7.9 5.5 10.4
MnO 0.03 0.05 0.01 0.03 --- 0.1 0.1
MgO 2.86 6.81 0.42 3.36 9.7 5.9 9.2
CaO 14.74 13.84 16.53 15.04 13.1 15.4 11.2
NaZO 1.93 1.05 1.99 1.66 0.4 0.5 0.7
K20 0.25 0.00 0.04 0.10 0.2 0.1 0.6

A: homogenous hornblende anorthosite, southern West Greenland (Windley,
1969)

B: layered hypersthene anorthosite, southern West Greenland (Windley,
1969)

C: anorthosite, Sittamundi complex, India (Subramanian, 1956)

D: average of A, B, and C

E: average composition of three massif (Lunar highland) derived categor-
ies of sample, Apollo 17 (Lowman, 1976)

F: average of six soils from several Apollo 16 stations (Lowman, 1976)
G: average of six Frau Mauro soils, taken as representative of Lunar

highland compositions with respect to major elements (Lowman, 1976)

12

Primitive acid sea.

The composition of the primitive ocean is derived from a weather-

ing model developed by Lafon and MacKenzie (1971) for the primordial

earth under environmental conditions as suggested by Shaw, 1. e. weath-

ering a primordial crust by acid seas derived from the excess volatiles

released by early differentiation of the cruat. Their estimate of excess

volatiles is from the chemical mass balance work of Horn and Adams (1966).

Estimate of Mass of Excess Volatiles

H20 16,700 x
C as C02 1,100 x
S 31 x
N 39 x
Cl 560 x

H, Br, B, As, F 16 x

1020grams
1020grams
lOZOgrams
1020grams
1020grams

1020grams

(Lafon and MacKensie, 1971, after Horn and Adams, 1966)

By assuming the volatiles are entirely dissolved in the aqueous

phase, and considering only the major reactive species, Lafon and Mac-

Kenzie derived the following primitive acid sea, with a total mass of

18,460 x 1020grams:

Primitive Acid Sea

Dissolved Concentration
Species (Moles/kg H20)
H2C03 1.51
HCl 0.946
H28 0.058

(Lafon and MacKenzie, 1971)

The assumption of total dissolution is not entirely accurate, but is jus-

tified for a balance carried to neutralization. C0 will continue to be

2

13

dissolved at saturation levels until its mass is distributed between the
sediments and the dissolved aqueous species, with atmospheric CO2 levels
approximating today's.

The initial pH of the model acid sea would be approximately zero,
with H+ contributed primarily by HCl, which dissociates completely at the
estimated concentration. H2C03 and H28 do not significantly dissociate
at this pH. They would remain in solution as uncharged species until the
seas were neutralized to a pH of approximately 4. Their dissociation will
maintain the pH at this level until neutralization reduces their total
concentrations to approximately their present levels in seawater.

The total mass of the major reactive species is applied to the mass
balance calculations. The mass of these volatiles in the atmosphere is
insignificant compared to their mass in the hydrosphere and the sediment-
ary pile (Poldervaart, 1955), so that the calculations provide a good
approximation of their distribution after neutralization of the primitive
acid seas. With this application, the total acidity of the primitive sea
is 2.57 equivalents per kilogram. This provides, for a primitive sea
mass of 18,460 x 1020 grams, 47.5 x 1020 equivalents of acid to be neut-
ralized by the primordial crust.

Proportions of primitive acid sea and primordial crust.

The minimum mass of anorthositic crust required to neutralize the
primitive acid seas can be approximated if it is assumed that all of the
metals in the rock react with the seas. The equation for calculating the

neutralizing capacity per gram of crust, for each component oxide, is as

tollows:
Number of equivalents of metal ion =

(Mass of oxidegper gram crust)(Charge on ion)(Gravimetric factor)
Atomic weight of metal

14

For a crustal composition from column D, Table 2, the neutralizing capac-

ity of the primordial crust is 0.057 equivalents per gram (Table 3, below).

Table 3. Calculation of maximum neutralizing capacity of

anorthositic primordial crust, per gram of crust.

3102 = .4676g = 0.03113 equiv. 51::

Ti02 = .0017g = 0.00009 equiv. Ti}+
A1203 = .2867g = 0.01687 equiv. A13+
FeZO3 = .0092g = 0.00035 equiv. Fe3+

FeO = .0269g = 0.00112 equiv. Fe

MgO = .0336g = 0.00168 equiv. Mg2+

Ca0 = .1504g = 0.00536 equiv. Ca2+

Na20 = .0166g = 0.00054 equiv. Nif

K20 = .0010g = 0.00002 equiv. K

Total = 1.00g = 0.05714 equiv. metal ions

830 x 1020 grams of crust are thus required to neutralize the 47.5 x 1020

equivalents of acid in 18,460 x 1020 grams of primitive sea. This is a
ratio of approximately 96 percent sea to 4 percent crust.

It is not realistic to assume such an extreme neutralizing capacity
for the crust, as even at the low initial pH level, metals such as silicon
and aluminum do not go into solution as ions. Only the more soluble metals
can be expected to react to neutralize the acid, so that the amount of
crust required for neutralization may be greater than the estimated 830 x
1020 grams by at least an order of magnitude.

Mass balance for each trial suite of sedimentary rock types is thus
calculated for a range of primordial anorthositic crust : primitive acid
sea mass ratios from 96 : 4 (770 x 1020grams crust + 18,460 x 1020 grams
sea) to 60 : 40 (12,310 x 1020 grams crust + 18,460 x 1020 grams sea).

The results of the mass balance calculations, particularly the balance

15

of the volatiles, approximate the effective point of neutralization
within this range. Component proportions for mixtures of primordial

anorthositic crust and primitive acid sea are listed in Table 4.

Table 4. Component proportions of primordial anorthositic crust
plus primitive acid sea mixtures used in chemical mass balance

calculations.

Component CR4 CR10 CR15 CR20 CR25 CR30 CR35 CR40
SiO2 1.87 4.68 7.02 9.36 11.70 14.04 16.38 18.72
A1203 1.15 2.87 4.31 5.74 7.18 8.61 10.05 11.48
Fe 0.11 0.27 0.41 0.54 0.68 0.81 0.95 1.08
MgO 0.14 0.34 0.51 0.68 0.85 1.02 1.19 1.36
CaO 0.60 1.50 2.25 3.00 ,3.75 4.50 5.25 6.00
NaZO 0.07 0.17 0.26 0.34 0.43 0.51 0.60 0.68
H20 86.87 81.44 76.92 72.39 67.87 63.34 58.82 54.29
C02 6.38 5.99 5.65 5.32 4.99 4.66 4.32 3.99
Cl 2.91 2.73 2.58 2.42 2.27 2.12 1.97 1.82
SO3 0.40 0.38 0.36 0.34 0.32 0.29 0.27 0.25
K20 0.004 0.01 0.015 0.02 0.025 0.03 0.035 0.04
Ti02 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08

CR4: 4% crust (770 x 1020g crust + 18460 x 1020g sea)

CR10: 10% crust (2050 x 1020g crust + 18460 x 1020g sea)
CR15: 15% crust (3260 x 1020g crust + 18460 x 1020g sea)
CR20: 20% crust (4620 x 1020g crust + 18460 x 1020g sea)
CR25: 25% crust (6150 x 1020g crust + 18460 x 1020g sea)
CR30: 30% crust (7910 x 1020g crust + 18460 x 1020g sea)

CR35: 35% crust (9940 x 1020g .rust + 18460 x 1020g sea)

CR40: 40% crust (12310 x 1020g crust + 18460 x 1020g sea)

16

RESULTS OF CALCULATIONS

The first objective of the mass balance calculations was to find a
suite of sedimentary rock types that satisfied the mass balance equation:
Primordial crust + Primitive acid sea = b1 Sediment1 +

b2 Sediment2 + . . . + bk seawater
such that the residuals of the component oxides were reasonably small and
the b coefficients for the sedimentary rock types were positive within
normal error (one standard deviation), so that the equation balanced
well with all of the sediment types on the products side of the equation.
A value of 2.0 or less for the sums of squares of residuals (SSR) was
taken as a preliminary indicator of a good balance. An examination of
individual component residuals, however, is necessary in the interpret-
ation of the significance of the results of the regressions.

The second stage was to improve the balance by trying alternative
sediments of similar types, ideally including Archean sedimentary rock
compositions in the suites. The ultimate goal was to find a suite of
sedimentary rock types that might be reasonably compared to observed
suites of early Archean sediments and metasediments.

0f forty-seven calculated suites, six approximated a reasonable
chemical mass balance with the modeled primordial anorhtositic crust.
These balanced at primordial crust to primitive acid sea mass proportions
of approximately 30 : 70.

No suites balanced that included a sandstone or a graywacke, as

there is insufficient SiO in the anorthositic weathering parent, with

2

respect to A120 for a sandstone to balance on the products side of the

3’

equation. Suites containing clays or shales, and no sandstones, balanced

17

only if the suite also contained a hypothetical aluminum-rich rock. The
anorthositic weathering parent has an excess of alumina with respect to
the other components, when compared to clays and shales, so that an
aluminum-rich rock must be included as.a product if the equation is to
balance.

The results of the calculations for the six sedimentary suites that
approximated a reasonable chemical mass balance are listed in Tables 5

through 10, below.

When the proportions of the sedimentary rock types for the six
tabulated calculations are recalculated excluding seawater, and averaged,

the composite sedimentary suite looks something like this:

Shale, or clay,
representing mudstones: 55%

Carbonate-rich iron

formation: 1 to 2%
Limestone: 23%
Evaporites: 1 to 2%
Aluminum-rich rock: 20%

The suite is additionally characterized as being low in sodium and quite
low in potassium, as the available mass of these components in the anorth-
ositic crust is insufficient to meet the requirements for them in the est-

imated sedimentary suite.

18

Table 5.

(potassium and titanium excluded from calculation).

Rock Type

Paleozoic shale

Gogebic magnetite iron formation

Limestone
Dolomite

Seawater

Aluminum-rich rock

Component
SiO2
A120
Fe
MgO
CaO
Na 0
H 0
C0
C1

SO

3

SSR = 1.2140

Y. Est.

14.03
8.61
0.82
1.03
4.51
1.16

63.35
4.65
1.23
0.31

30% crust, 70% sea

Y. Obs.
14.04
8.61
0.81
1.02
4.50
0.51
63.34
4.66
2.12
0.29

Variable

Coefficient

0.2287
-0.0095
0.1165
-0.0258
0.6326
0.0955

Residual
-0.0124
0.0000
0.0119
0.0089
0.0080
0.6482
0.0079
-0.0091
-0.8904
0.0240

19

Results of sedimentary chemical mass balance calculation

Standard

Deviation

0.0136
0.0156
0.0307
0.0305
0.0056
0.0118

% Residual

-0.09
0.00
1.47
0.87
0.18

127.10
0.01
-0.20
~42.00
8.28

Table 6. Results of sedimentary chemical mass balance calculation

(potassium and titanium excluded from calculation)

Variable Standard
Rock Type Coefficient Deviation
Paleozoic shale 0.2272 0.0120
Temiscamie magnetite iron formation -0.0103 0.0173
Limestone 0.1154 0.0173
Dolomite -0.0223 0.0318
Seawater 0.6327 0.0056
Aluminum-rich rock 0.0959 0.0117
Component Y. Est. Y. Obs. Residual % Residual
3102 14.03 14.04 -0.0130 -0.09
A1203 8.61 8.61 0.0000 0.00
Fe 0.84 0.81 0.0267 3.30
MgO 1.07 1.02 0.0499 4.89
CaO 4.54 4.50 0.0351 0.78
NaZO 1.16 0.51 0.6467 126.80
H20 63.35 63.34 0.0078 0.01
C02 4.62 4.66 -0.0448 -0.96
Cl 1.23 2.12 -0.8903 -42.00
SO3 0.31 0.29 0.0230 7.93
SSR = 1.2181

30% crust, 70% sea

20

Table 7. Results of sedimentary chemical mass balance calculation

(potassium and titanium excluded from calculation)

Variable
Rock Type Coefficient
Cenozoic-Mesozoic shale 0.2476
Temiscamie magnetite iron formation -0.0057
Limestone 0.0817
Dolomite -0.0116
Seawater 0.6297
Aluminum-rich rock 0.1026
Component Y. Est. Y. Obs. Residual
SiO2 14.00 14.04 -0.0374
A1203 8.61 8.61 0.0000
Fe 0.90 0.81 0.0850
MgO 1.19 1.02 0.1655
CaO 4.62 4.50 0.1172
N820 1.37 0.51 0.8590
H20 63.34 63.34 0.0044
C02 4.51 4.66 -0.1489
Cl 1.22 2.12 -0.8968
SO3 0.36 0.29 0.0749
SSR = 1.6198

30% crust, 70% sea

21

Standard

Deviation

0.0151
0.0198
0.0362
0.0369
0.0065
0.0133

% Residual

-0.27

0.00
10.49
16.23

2.60
168.43

0.01
-3.20
-42.30
25.83

Table 8. Results of sedimentary chemical mass balance calculation.

Variable Standard
Rock Type Coefficient Deviation
Paleozoic shale 0.2257 0.0216
Temiscamie magnetite iron formation -0.0090 0.0146
Limestone 0.1167 0.0433
Dolomite -0.0236 0.0421
Halite 0.0031 0.0087
Anhydrite -0.0008 0.0117
Seawater 0.6327 0.0071
Aluminum-rich rock 0.0963 0.0146
Component Y. Est. Y. Obs. Residual % Residual
SiO2 13.97 14.04 -0.0705 -0.50
A1203 8.61 8.61 0.0000 0.00
Fe 0.87 0.81 0.0605 7.47
MgO 1.06 1.02 0.0355 3.48
CaO 4.53 4.50 0.0253 0.56
N820 1.32 0.51 0.8067' 158.18
H20 63.34 63.34 0.0015 0.00
C02 4.63 4.66 -0.0319 -0.68
Cl 1.41 2.12 -0.7055 -33.28
SO3 0.27 0.29 -0.0181 -6.24
K20 0.88 0.03 0.8527 2842.33
TiO2 0.19 0.06 0.1275 212.50
SSR = 1.9037

30% crust, 70% sea

22

Table 9. Results of sedimentary chemical mass balance calculation.
Variable Standard

Rock Type Coefficient Deviation

Montmorillonite 0.3038 0.0133

Temiscamie magnetite iron formation 0.0068 0.0167

Limestone 0.1009 0.0120

Halite 0.0056 0.0073

Anhydrite 0.0043 0.0092

Seawater 0.5368 0.0064

Aluminum-rich rock 0.0805 0.0167

Component Y. Est. Y. Obs. Residual Residual

SiO2 16.32 16.38 -0.0646 -0.39

A1203 10.05 10.05 0.0000 0.00

Fe 0.90 0.95 -0.0502 -5.28

MgO 2.07 1.19 0.8797 73.92

CaO 5.07 5.25 -0.1814 ~3.46

NaZO 1.28 0.60 0.6795 113.25

H20 58.82 58.82 -0.0005 0.00

C02 4.35 4.32 0.0251 0.58

Cl 1.38 1.97 -0.5945 -30.18

SO3 0.40 0.27 0.1296 48.00

K20 0.14 0.04 0.1074 268.50

TiO2 0.04 0.07 -0.0305 -43.57

SSR = 1.6586

35% crust, 65% sea

23

Table 10. Results of sedimentary chemical mass balance calculation.
Variable Standard

Rock Type Coefficient Deviation

Montmorillonite 0.2956 0.0262

Carbonate iron formation 0.0126 0.0269

Limestone 0.0997 0.0130

Halite 0.0056 0.0072

Anhydrite 0.0049 0.0094

Seawater 0.5383 0.0077

Aluminum-rich rock 0.0837 0.0154

Component Y. Est. Y. Obs. Residual Residual

SiO2 16.32 16.38 -0.0606 -0.37

A1203 10.50 10.50 0.0000 0.00

Fe 0.89 0.95 -0.0556 -5.05

MgO 2.04 1.19 0.8523 71.62

CaO 5.02 5.25 -0.2318 -4.42

N320 1.28 0.60 0.6762 112.70

H20 58.82 58.82 -0.0005 0.00

C02 4.40 4.32 0.0814 1.88

Cl 1.38 1.97 -0.5916 -30.03

SO3 0.44 0.27 0.1657 61.37

K20 0.14 0.04 0.1048 262.00

TiO2 0.04 0.07 -0.0315 -45.00

SSR = 1.6402

35% crust, 65% sea (Best volatile balance at 32% crust)

24

The approximate mass of these sediments can be calculated from the
masses of the starting materials. The mass of the primitive acid sea is
fixed at 18,460 x 1020 grams by the estimate of excess volatiles. The
necessary amount of anorthositic crust for neutralization is added to
the system to obtain the appropriate crust to sea mass ratio, providing
a fixed total mass for a given mixture of primordial crust and primitive
acid sea. By subtracting the derived proportion of final seawater from
the total mass for each of the six tabulated calculations above, the mass
of the derived sediments can be obtained. The mean for the six tabulated
sedimentary suites is 10,850 x 1020 grams. This value is 30% to 70%
less than estimates of the total mass of sediments and sedimentary rocks
(Garrels and Mackenzie, 1971; Carrels et a1, 1972; Horn and Adams, 1966;
and Li, 1972). The calculated mass of the final neutralized sea is
16,200 x 1020 grams, which is within the range of values of estimates of
the total mass of the hydrosphere (Mason, 1966; Poldervaart, 1955; Ronov,

1968; and Carrels and MacKenzie, 1971).

EFFECTS OF SEDIMENTARY RECYCLING

The amount of the 10,850 x 1020 grams of sediment derived from the
weathering of a primordial anorthositic crust that may remain at the pres-
ent may be estimated with the aid of sedimentary recycling models. The
timetable for Shaw's crustal evolution model places the deposition of
these sediments in a period between 3900 million years before present and
the beginning of the development of greenstone belts, which date to ap-
proximately 3500 million years before present. Sedimentary recycling
processes thus will have been operating on these primitive sediments for

a mean of 3700 million years.

25

Estimates of the total mass of sediments and sedimentary rocks
range from 14,000 x 1020 grams to 32,000 x 1020 grams. The most frequently
derived estimates are on the order of 24,000 x 1020 grams. This inter-
mediate figure is used for estimates of the proportion of the total sedi-
mentary mass represented by the surviving portion of the primitive sedi-
ments derived from the weathering of a primordial crust.

Estimates of sedimentary recycling rates vary. Rates from the re-
cycling models of Carrels et al (1972) and Blatt and Jones (1975) provide
upper and lower limits, respectively, for the expected preservation of
the derived sedimentary suite.

Blatt and Jones estimate a sediment half-age of 130 million years
from data on the areal extent of exposures of Paleozoic and younger sed-
imentary rocks. This rate provides a survival fraction of 2.7 x 10"9
for sediments 3700 million years old, or 2.9 x 1015 grams preserved from
the derived mass of of 10,850 x 1020grams. This is a fraction of 1.2 x
10"9 of the present sedimentary mass, a little more than one millionth.
If the Blatt and Jones recycling rate is thus extrapolated 3700 million
years into the past, the probability of finding a sedimentary rock of
this age would be remote.

Carrels et al have estimated different recycling rates for dif-
ferent sedimentary rock types. This reflects differences in resistance
to erosion due to differences in solubility. Their recycling rates are
based on maximum thickness data for sedimentary rocks, including data on
exposures of Precambrian sediments and metasediments.

This model assigns a half-age of 600 million years for sandstones
and shales; the survival fraction for 3700 million year old rocks is

0.01. The sandstone and shale recycling rate is assumed to also apply

26

to ironstones. Carbonates are assigned a half-age of 300 million years,
for a 3700 million year survival fraction of 1.9 x 10-4. Evaporites are
assigned a half-age of 200 million years, for a 3700 million year sur-
vival fraction of 2.7 x 10'6.

These survival fractions leave 59.7 x 1020 grams of shale, 1.63 x
1020 grams of iron formation, 0.474 x 1020 grams of limestone and 4.4 x
1016 grams of evaporite from the 10,850 x 1020 grams of sediment derived
from the primordial crust. As aluminum oxides are less soluble than sil-
ica, it is reasonable to assume the sandstone and shale recycling rate
would apply to the aluminum-rich rock, so the amount preserved would be
21.7 x 1020 grams. The total amount of sediment preserved would be 83.5
x 1020 grams, a fraction of 7.7 x 10'3 of the total sedimentary mass,
approximately one one-hundred-thirtieth. This fraction suggests the
probability of locating a remnant of sedimentary rock of this age is
quite good.

The differential recycling model of Carrels et al has an important
effect on the proportions of rock types after recycling processes have
acted for 3700 million years. After differential recycling for this

length of time at the recycling rates above, the sedimentary suite derived

from a primordial anorthositic crust looks like this:

Shale, or clay,

representing mudstones: 71.5%
Carbonate-rich

iron formation: 2.0%
Limestone: 0.6%
Evaporites: Nil
Aluminum-rich rock 26.0%

27

As an aluminum-rich rock would be as persistant as a shale, if the
primordial crust were anorthositic we could expect to find, in ancient
sedimentary rocks, a mass of aluminum-rich rock of approximately one-
third the mass of shales of the same age. This trend should continue in
sedimentary rocks of younger age age until enough silica is added to the
sedimentary system from the erosion of younger, more siliceous igneous
rocks for the aluminum-rich rocks to be Changed to clays and aluminosili-
cates by weathering and resedimentation, or by metamorphism. Within the
framework of Shaw's model of crustal evolution, these changes could begin
with the onset of greenstone belt activity; with the addition of sediment-
ary material derived from the weathering of new igneous rocks, the chemi-
cal character of the sedimentary mass would begin to change, and increas-
ingly younger sedimentary rocks would look less like the material derived

from the weathering of an anorthosite.

DISCUSSION OF RESULTS OF CALCULATIONS

The six tabulated estimates of the proportions of sedimentary rock
types calculated from the mass balance equation provide an approximation
of the proportions of sediment types that might result from the weathering
of an anorthositic primordial crust by acid seas condensed from the earth's
excess volatiles. The approximation does not resemble estimates of the
proportions of sedimentary rock types presently in the crust.

The most apparent difference between the derived sedimentary suite
and common sedimentary rocks is the high ratio of alumina to silica in
the suite derived from the anorthosite. An anorthositic crust is so poor
in silica with respect to alumina that a sandstone cannot be derived from

it with a mass balance calculation. In order to balance the equation

28

with a clay or shale as a product, an aluminum-rich rock must be included
in the product sedimentary suite; the anorthosite is silica deficient in
comparison to clays and shales. 0f common sedimentary minerals, only
kaolinite minerals'and aluminum hydroxides are sufficiently rich in alum-
inum to balance with the anorthosite.

In order to analyse the magnitude of this and other mass balance
discrepancies, a limited number of calculations were made with other
igneous rock types (granodiorite, andesite, and tholeiitic basalt from
Hyndman, 1972) as the primordial crust. These were balanced against a
sedimentary suite resembling the ones that balanced best with the anorth-
osite. No attempt was made to find the sedimentary suites that balanced
best with these alternative crustal materials, as the purpose in their
calculation was to aid in the evaluation of residual imbalances that oc-
curred in calculations made for the anorthositic crust. The data for
these calculations is listed in Tables 11 through 13, and the results of
the calculations are in Tables 14 through 16, below.

These additional calculations showed that a material as rich in
silica as a granodiorite is needed as a weathering parent if the crustal
material is to balance with a sedimentary suite containing a quartz sand-
stone. Balances calculated with andesite and tholeiitic basalt as prim-
ordial crust produced positive coefficients for aluminum-rich rock in
suites containing a shale and no sandstone. The aluminum excess in these
materials, however, is not as great as that in the anorthosite.

The second major residual balance discrepancy is in potassium.
Estimates using shales as the mudstones in the sedimentary suites showed

a potassium deficiency on the order of 3000% of the amount available in

29

the anorthositic weathering parent. Estimates using low-potassium clays
to represent mudstones showed potassium deficiencies on the order of
300%.

The andesite and tholeiitic basalt alternatives showed potassium
deficiencies an order of magnitude smaller; these materials might balance
well with low-potassium clays. The granodiorite alternative produced
potassium residuals small enough to suggest that it would balance well
with respect to potassium when balanced against a suite containing a
shale and a sandstone, i. e. a suite more apprOpriate to its composition.
These results indicate that the potassium balance discrepancy is not a
result of the approach to the problem, it is rather the result of a real
and serious potassium deficiency in the anorthositic weathering parent.

A third residual imbalance occurs with respect to sodium, and it is
consistently related to an imbalance with respect to chlorine. The sod-
ium residuals indicate a deficiency of approximately 100% to 150% of the
amount of sodium available in the anorthosite. There is a consistent
chlorine surplus-of approximately 30% to 40% of the amount of chlorine in
the model primitive acid sea.

Two factors may be working together to produce these balance dis-
crepancies. The anorthositic weathering parent is certainly deficient in
sodium compared to the prOportions of this element in sedimentary rocks.
Also, the estimate of the amount of chlorine in the earth's excess vola-
tiles may be slightly high, as noted by Lafon and Mackenzie (1971). In
the balance of the components, chlorine occurs on the products side of
the equation in significant amounts only in seawater and halite; both of
these require sodium in an approximate 3 : 4 mass ratio to the amount of

chlorine. The amount of seawater estimated is mostly determined by the

30

H20 mass, and the mass of chlorine not balanced in seawater must be bal-
anced, accompanied by sodium, in halite rock. A sodium deficiency in

the anorthositic weathering parent will restrict the amount of halite est-
imated, leaving a high negative value for the chlorine residual, which
appears as a surplus in the primitive acid sea. Conversely, an actual
chlorine surplus, resulting from a high value for the mass of chlorine

in Horn and Adams' estimate of crustal mass balance, will accentuate any
sodium deficiency by producing greater amounts of halite in the mass bal-
ance calculations.

Trends of residuals with increasing proportions of primordial crust
to primitive acid sea demonstrate this reciprocal relationship. With an
increasing proportion of crust, absolute values of the chlorine and sodium
, and SO

residuals converge toward zero. The other volatiles, H O, CO

2 3’

have near zero residuals at a common ratio of primordial crust to acid

2

sea, a ratio at which the values of the residuals for iron, magnesium,
calcium, silica and aluminum also converge at zero. This proportion is
in the range 30% to 35% crust for all of the good balances. The converg-
ence of sodium and chlorine residuals, however, is at much higher crust
to sea ratios.

The resolution of this imbalance is complicated by the uncertainty
as to the composition of the seas in the early Precambrian. If, as Ronov
(1968) suggests, the early oceans had higher Mg2+: Na+ and Caz+: Na+ ratios
than modern seawater, the chlorine in the early neutralized oceans may
have been electrically balanced by these ions, and these would be replaced
by Na+ as it was introduced by the weathering of later igneous rocks.

As noted in the discussion of the choice of a final seawater for these

mass balance calculations, there is enough uncertainty as to the early

31

composition of seawater to preclude the choice of some alternative to
modern seawater as the final seawater composition.

The mass balance calculations using alternative primordial crustal
compositions were made primarily to evaluate this sodium-chlorine imbal-
ance. ’Granodiorite and andesite, which contain approximately five times
as much sodium as the anorthositic crust, balanced with much smaller
chlorine and sodium residuals. This confirms that much of the imbalance
is due to a sodium deficiency in the anorthosite. The chlorine balance,
however, with these materials still requires a greater proportion of crust
to sea than required for the balance of the other volatiles and the major
metal oxides, suggesting that the sodium deficiency is accentuated by a
high estimate of chlorine in the excess volatiles.

The calculations reported in Tables 7, 9 and 10 suggest imbalances
of 503. These apparent imbalances result from the method of reporting
data. The best balances for these three calculations occurred at crust
to sea proportions intermediate to ratios for which balances were actually
calculated and results tabulated. The SO3 residuals diverged far enough

at the nearest data point to the best balance point to compute as high

percentages of the small mass of SO in the system. Graphs of residual

3
values plotted against crust to sea prOportions for all computed balances
demonstrate that linear interpolation of residual values between calcul-

ated data points is justified. Such interpolation of values of SO3 resi-

duals for these three suites indicates that the SO3 residuals converge to
zero at the same crust to sea mass ratio the other well balanced compon-
ents converge to zero.

The titanium imbalances indicated do not appear to be the result of

an interpolation difficulty. These imbalances are consistent over the

32

full range of crust to sea ratios for the balance calculations in which
they occur. Their geological significance is questionable, because the
titanium proportions in all of the rock types are quite small, and they
contribute little to the least squares estimate of mass balance. The
titanium residuals have both positive and negative values, indicating
titanium deficiencies and surpluses, respectively, for balances with

quite similar sedimentary suites.

33

Table 11. Component proportions of granodiorite plus primitive acid sea

mixtures used in analytical sedimentary chemical mass balance calculations.

Component GD4 GD10 GD15 CD20 GD25 G030 CD35 GD40
8102 2.68 6.69 10.03 13.38 16.72 20.06 23.41 26.75
A1203 0.63 1.57 2.35 3.13 3.92 4.70 5.48 6.26
Fe 0.12 0.29 0.44 0.59 0.74 0.88 1.03 1.18
MgO 0.06 0.16 0.24 0.31 0.39 0.47 0.55 0.60
CaO 0.14 0.36 0.53 0.71 0.89 1.07 1.25 1.42
NaZO 0.15 0.38 0.58 0.77 0.96 1.15 1.34 1.54
H20 86.90 81.51 77.01 72.52 68.03 63.54 59.05 54.55
C02 6.38 5.99 5.65 5.32 4.99 4.66 4.32 3.99
Cl 2.91 2.73 2.58 2.42 2.27 2.12 1.97 1.82
SO3 0.40 0.38 0.36 0.34 0.32 0.29 0.27 0.25
K20 0.12 0.31 0.46 0.61 0.77 0.92 1.07 1.23
TiO2 0.02 0.06 0.09 0.11 0.14 0.17 0.20 0.23

GD4: 4% granodiorite, 96% sea

GD10: 10% granodiorite, 90% sea
GD15: 15% granodiorite, 85% sea
GD20: 20% granodiorite, 80% sea
GD25: 25% granodiorite, 75% sea
GD30: 30% granodiorite, 70% sea
GD35: 35% granodiorite, 65% sea
GD40: 40% granodiorite, 60% sea

34

Table 12.

mixtures used in analytical sedimentary chemical mass balance calculations.

Component proportions of andesite plus primitive acid sea

Component AD4 AD10 AD15 ADZO AD25 AD3O AD35 AD4O
SiO2 2.17 5.42 8.13 10.84 13.55 16.26 18.97 21.68
A120 0.69 1.72 2.58 3.43 4.29 5.15 6.01 6.87
Fe 0.27 0.67 1.01 1.34 1.68 2.01 2.35 2.68
MgO 0.17 0.44 0.65 0.87 1.09 1.31 1.53 1.74
CaO 0.32 0.79 1.19 1.58 1.98 2.38 2.77 3.17
NaZO 0.15 0.37 0.55 0.73 0.92 1.10 1.28 1.47
H20 86.90 81.53 77.05 72.56 68.08 63.60 59.12 54.64
C02 6.38 5.99 5.65 5.32 4.99 4.66 4.32 3.99
Cl 2.91 2.73 2.58 2.42 2.27 2.12 1.97 1.82
SO3 0.40 0.38 0.36 0.34 0.32 0.29 0.27 0.25
K20 0.04 0.11 0.17 0.22 0.28 0.33 0.39 0.44
TiO2 0.05 0.13 0.20 0.26 0.33 0.39 0.46 0.52
AD4: 4% andesite, 96% sea

AD10: 10% andesite, 90% sea
AD15: 15% andesite, 85% sea
AD20: 20% andesite, 80% sea
AD25: 25% andesite, 75% sea
AD30: 30% andesite, 70% sea
AD35: 35% andesite, 65% sea
AD40: 40% andesite, 60% sea

35

Table

13.

Component proportions of tholeiitic basalt plus primitive acid

sea mixtures used in analytical sedimentary chemical mass balance

calculations.

Component

SiO2

A1203
Fe
MgO
CaO
Na 0
H O
C0

Cl

80

K O

T10

T84:

T810:
T815:
T820:
T825:
T830:
T835:
T840:

T84
2.03
0.56
0.36
0.25
0.42
0.09
86.91
6.38
2.91
0.40
0.03
0.08

T810

OHOOHU‘I

U1

.08
.41
.91
.63
.04
.22
81.
.99
.73
.38

53

0.20

4% basalt, 96% sea

10%
15%
20%
25%
30%
35%
40%

basalt,
basalt,
basalt,
basalt,
basalt,
basalt,
basalt,

90%
85%
80%
75%
70%
65%
60%

sea

sea

sea

sea

sea

sea

sea

T815
7.62
2.11
1.36
0.95
1.56
0.33
77.05
5.65
2.58
0.36
0.12
0.30

T820
10.17
2.81
1.81
1.27
2.08
0.45
72.57
5.32
2.42
0.34
0.16
0.41

36

T825
12.71
3.52
2.27
1.59
2.61
0.56
68.10
4.99
2.27
0.32
0.21
0.51

T830
15.25
4.22
2.72
1.90
3.13

63.62

0.25

T835
17.79
4.92
3.17
2.22
3.65
0.78
59.14
4.32
1.97
0.27
0.29
0.81

T840
20.33
5.63
3.62
2.54
4.17
0.89
54.66
3.99
1.82
0.25
0.33
0.91

Table 14. Results of sedimentary chemical mass balance calculation,

granodiorite crust.

Variable Standard
Rock Type Coefficient Deviation
Paleozoic shale 0.4389 0.0118
Limestone 0.0452 0.0431
Dolomite -0.0121 0.0407
Halite 0.0101 0.0087
Anhydrite -0.0084 0.0116
Seawater 0.5338 0.0070
Aluminum-rich rock -0.0195 0.0143
Component Y. Est. Y. Obs. Residual % Residual
SiO2 26.65 26.75 -0.1034 -0.39
A1203 6.26 6.26 0.0000 0.00
Fe 2.24 1.18 1.0641 90.18
MgO 1.23 0.63 0.6022 95.59
CaO 1.87 1.42 0.4463 31.43
NaZO 1.75 1.54 0.2069 13.44
H20 54.55 54.55 -0.0005 0.00
C02 3.42 3.99 -0.5700 -14.29
Cl 1.64 1.82 -0.1811 -9.95
SO3 -0.07 0.25 -0.3190 -127.60
K20 1.62 1.23 0.3920 31.87
TiO2 0.35 0.23 0.1238 53.83
SSR = 2.3761

40% granodiorite, 60% sea

37

Table 15. Results of sedimentary chemical mass balance calculation,

andesite crust.

Variable Standard
Rock Type Coefficient Deviation
Paleozoic shale 0.3117 0.0090
Limestone 0.0384 0.0331
Dolomite 0.0245 0.0312
Halite 0.0088 0.0067
Anhydrite -0.0017 0.0089
Seawater 0.5863 0.0054
Aluminum-rich rock 0.0173 0.0110
Component Y. Est. Y. Obs. Residual % Residual
8102 18.98 18.97 0.0115 0.06
A1203 6.01 6.01 0.0000 0.00
Fe 1.63 2.35 -0.7227 -30.75
MgO 1.64 1.53 0.1150 7.52
CaO 2.80 2.77 0.0343 1.24
NaZO 1.63 1.28 0.3493 27.29
H20 59.12 59.12 0.0002 0.00
C02 4.26 4.32 -0.0578 -1.34
Cl 1.66 1.97 -0.3058 -15.52
SO3 0.25 0.27 -0.0245 -9.07
K20 1.16 0.39 0.7745 198.59
TiO2 0.25 0.46 -0.2083 -45.28
SSR = 1.3995

35% andesite, 65% sea

38

Table 16.
tholeiitic basalt crust.

Results of sedimentary chemical mass balance calculation,

Variable Standard
Rock Type Coefficient Deviation
Paleozoic shale 0.2515 0.0143
Limestone 0.0267 0.0523
Dolomite 0.0500 0.0494
Halite 0.0042 0.0106
Anhydrite 0.0002 0.0141
Seawater 0.6350 0.0085
Aluminum rich rock 0.0015 0.0173
Component Y. Est. Y. Obs. Residual Residual
SiO2 15.31 15.25 0.0647 0.42
A1203 4.22 4.22 0.0000 0.00
Fe 1.34 2.72 -1.3803 -50.75
MgO 1.95 1.90 0.0456 2.40
CaO 3.09 3.13 -0.0443 -1.42
NaZO 1.40 0.67 0.7287 108.76
H20 63.62 63.62 0.0013 0.00
C02 4.70 4.66 0.0355 0.76
Cl 1.48 2.12 -0.6372 -30.06
S03 0.32 0.29 0.0316 10.90
K20 0.95 0.25 0.6961 278.44
TiO2 0.20 0.61 -0.4072 -66.75
SSR = 3.5031

30% tholeiitic basalt, 70% sea

39

COMPARISON OF SEDIMENT DERIVED FROM PRIMORDIAL ANORTHOSITIC CRUST BY
MASS BALANCE WITH ANCIENT SEDIMENTARY ROCKS

The oldest known terrestrial rocks are the metasedimentary Isua
supracrustals of southern West Greenland. Iron formation from this
succession has been dated at 3760 f 70 million years (Moorbath et al, 1973).
This date has been interpreted as the age of metamorphism of the iron
formation, and the geochronologists suggest the depositional age may be
as great as 3900 million years.

The Isua supracrustals have been interpreted as a shallow-water
shelf facies (Bridgwater et al, 1973), with elongate masses of ultrabasic
rocks between sedimentary successidns suggestive of submarine lavas (Brid-
gwater and McGregor, 1974). The bulk of the succession consists of carb-
onate-rich quartzofeldspathic metasediments, quartzite-magnetite and
siderite-magnetite iron formation, carbonate-bearing quartzites, and silica-
rich carbonate-bearing garnet-biotite-muscovite schists. Conglomeratic
units contain boulders, cobbles and pebbles of muscovite-rich quartzo-
feldspathic rocks, for which preliminary analyses show high-potassium
granitic compositions, in a fine-grained carbOnate-rich matrix; and units
of quartzite pebbles in garnet-biotite-muscovite schists (Bridgwater and
McGregor, 1974; and Allaart, 1975).

The Isua metasediments compare favorably to the sedimentary suite
derived from the mass balance calculations only with respect to the
dominance of pelitic rocks (shales and clays, or schists), and the pres-
ence of significant amounts of iron formation. Closer comparison only
emphasizes the differences.

The schist mineralogies in the Isua supracrustals indicate a high

potassium content, and these could not be derived from the weathering of

40

the modeled primordial anorthositic crust. This discrepancy is empha-
sized by the presence of potassium-rich granitic cobbles and boulders in
Isua conglomerates; these suggest derivation from the weathering of a
granitic crust rather than one of anorthositic composition.

The Isua sediments are remarkably silica rich. The schists are
quartz-bearing, and the ironstones are notably quartz rich. Quartzites
are volumetrically significant in the succession. The modeled anortho-
sitic crust is quartz deficient with respect to shales, and cannot be ex-
pected to produce quartz sandstones. There is no evidence in the Isua
succession of an aluminum-rich rock that would be volumetrically signif-
icant in a sedimentary suite derived from the weathering of a primordial
anorthositic crust.

The preliminary mapping teams who have examined the Isua sediments
suggest that they are a shallow-water continental shelf facies, which is
reasonably interpretable as deposition in a shallow intracontinental sea.
The tectonic setting, as well as age, of the succession is thus appropri-
ate for sediments weathered from the primordial crust. The succession
does not lithologically resemble the suite predicted by chemical mass
balance calculations for the weathering of an anorthositic primordial
crust. A granitic composition is more strongly indicated for the pri-
mordial crust.

The other known rocks of sedimentary origin sufficiently old to
represent the weathering products of a primordial crust are the sedi-
ments in the 3300 to 3400 million year old Swaziland Sequence of South
Africa (Hurley et al, 1972; and Allsopp et a1, 1968 and 1974). The
characteristic associations of these oldest sediments on the African
craton indicate that their origin is related to the development of green-

41

stone belts, and they cannot be compared to the sedimentary products
predicted by a mass balance calculation for the weathering of the first

crust.

EFFECTS OF A PRIMORDIAL ANORTHOSITIC CRUST ON SEDIMENTARY ROCK COMPO-
SITIONS THROUGH GEOLOGIC TIME.

Sedimentary recycling rates predict that a maximum of one percent
of the material derived from the weathering of a primordial crust would
be preserved as first cycle sedimentary rocks. The other 99 percent will
be recycled to become part of younger sediments, and a distinctive anorth-
ositic composition may be expected to impart a chemical imprint to the
sedimentary mass. This imprint will be stronger the closer in age sub-
sequent cycles of sedimentation are to the first cycle, and the waning
of this chemical imprint should appear as marked trends in the chemistry
of sedimentary rocks through geologic time.

These expected chemical trends can be approximated by comparing
the composition of the original anorthositic material to the composition
of the average Precambrian sediment, and calculating the mass and compos-
ition of the new igneous material that must be added to achieve the ob-
served average chemistry.

The equation for this calculation is, for each component:

(1-X)(proportion of component in anorthositic crust) +
(X)(proportion of component in rock to be added) =

proportion of component in average Precambrian sediment.
The composition of the average Precambrian sediment used in this calcula-
tion is the estimate by Reilly and Shaw (1967) for the sediments of the

Canadian shield. Results of the calculations for values of X from 0.5 to

42

0.8, representing ratios of primitive sediment to new sedimentary material

of 1 : 1 to 1 : 4, are tabulated below:

Table 17. Estimate of mass and composition of new igneous material
that must be added to anorthosite-derived sediments to achieve average

Precambrian sedimentary composition.

Proportion ' Pr0portion in material to
in average be added to achieve average
Proportion in Precambrian Precambrian composition for
Component anorthosite sediment indicated value of X

0.50 0.67 0.75 0.80

8102 47.5 66.5 85.5 76.0 72.8 71.3
.41203 29.1 15.3 1.5 8.4 10.7 11.9
Fe 2.8 5.6 8.4 7.0 6.5 6.3
MgO 3.4 3.0 2.6 2.8 2.9 2.9
CaO 15.3 4.9 -5.5 -o.3 1.4 2.3
Na20 1.7 3.1 4.5 3.8 3.6 3.5
K20 0.1 1.7 3.3 2.5 2.2 2.1
T102 0.2 0.6 1.0 0.8 0.7 0.7

The composition of the average Precambrian sediment can be achieved
if the sedimentary material derived from the primordial anorthositic
CIWJSt is mixed with three to four times its mass of new material of ap-
ptwapriate composition. Based on the estimated mass of the sediment de-
rinved from anorthosite with the mass balance equations, the total mass of
setiiments after this addition is slightly greater than the highest esti-
"Eites of the total mass of sediments and sedimentary rocks in the crust
(Charrels and MacKenzie, 1971; and Li, 1972). The 1 : 3 (X = 0.75) and
1 : 4 (X = 0.80) mixing ratios thus represent the maximum amount of new
nlaterial it would be reasonable to add to the sedimentary mass.

At these mixing ratios, the composition of the material that must

43

be added to the anorthosite-derived sediment has a silica to alumina mass
ratio within the range of that of rhyolites, rhyodacites, and quartz
monzonites. These, however, are generally low in magnesium particularly,
and iron, to an extent, with respect to the required compositions, and
somewhat rich in alkalis. The appropriate compositions are also approached
by quartz diorites, which are somewhat deficient in alkalis. Ultramafic
rocks have favorable silica to alumina ratios, and small proportions of
these could provide the necessary amounts of magnesium and iron. It is
thus possible to bridge the compositional gap between an anorthositic
primordial crust and average Precambrian sediments with a mixture of
silica-rich alkali volcanic or plutonic rocks and ultramafic rocks.

This change in the chemical character would have had to occur
between 3500 million years before present, when greenstone belt volcanic
activity commenced to add new material to the sedimentary mass, and the
end of the Precambrian, at which time the character of the sediments had
changed to the average composition estimated by Reilly and Shaw.

Definite geochemical trends can be predicted for sedimentary rocks
within the Precambrian if this change has occurred. The ratio of A1203
to SiOz, and the proportion of CaO in Precambrian sediments should decrease

with decreasing age. The pr0portions of Fe, Na 0, and K 0 should increase

2 2
with decreasing age, as should the ratios K O : Na 0, MgO : C80, and

2 2
Fe : MgO. If the primordial crust were anorthositic, these chemical
trends should be quite marked in Precambrian sediments.
Two problems hinder the evaluation of these trends. The first is
the problem of determining the pre-metamorphic nature of most Precambrian
rocks. It is usually quite difficult to distinguish metavolcanic rocks

from metasedimentary rocks, and for rocks that have experienced very high

44

degrees of metamorphism, it may be impossible to distinguish rocks of
volcanic or sedimentary pre-metamorphic character from metamorphosed plut-
onic rocks. These distinctions are often made on the basis of the chem-
istry of the metamorphic rocks, so that metasediments substantially
different from typical modern sediments may not be identified as such,

and these chemically "different" metasediments are the ones that would be
needed to identify chemical trends. The most extensive compilations of
chemical data for Precambrian rocks do not make a distinction between
metasediments and metavolcanics (Ronov and Migdisov, 1971; Ronov, 1972;
and Shaw et al, 1967).

The second problem is that of determining ages of Precambrian
sediments and metasediments for which geochemical data are available.
Compilations of data for Canadian shield Precambrian sediments (Reilly
and Shaw, 1967) make no age distinctions, while estimates of Canadian
shield geochemistry that distinguish between Archean and Proterozoic
rocks rocks do not distinguish these rocks by pre-metamorphic character.
This bimodal time distinction, even when made, is not sufficiently sensi-
tive to evaluate the chemical trends we seek. In each of the two time
periods, chemical data is averaged over a range of ages in excess of 1.5
billion years. The chemical changes for which we seek to evaluate trends
may be completed entirely within one of these time periods.

The data of Ronov and Migdisov are compiled for four time periods
within the Precambrian: Azoic (older than 3500 million years), Archeozoic
(3500 to 2700 million years before present), and two subdivisions of
Proterozoic (2700 to 1400 million years before present, and 1400 to 600
million years before present). These time distinctions are also not

sufficiently sensitive to identify the chemical trends that need to be

45

evaluated, due to the length of the Archeozoic and early Proterozoic
divisions.

The data of Ronov and Migdisov can be roughly compared to the pre-
dicted trends. Their data show a decrease in the proportion of K20 in
clays and shales with greater age, as predicted. The data also indicate
a decrease in the ratio of A1203 to SiO2 in clays and shales with greater
age, and an increase in the amount of iron in clays and shales; both of
these trends are counter to those predicted for an anorthositic primord-
ial crust. These trends are based on geochemical data for sedimentary
and volcanic rocks in the Archeozoic and the two divisions of the Proter-
ozoic. The trends do not support an anorthositic crustal model, but as
noted above, the data are not sensitive enough to accurately evaluate the
predicted trends.

Ronov (1972) has published trends for the chemical character of
the continental erosion surfaces through geologic time. These cannot be
used to evaluate the trends predicted for the anorthositic model, as they
are based only in part on observed chemical data. The trends, particul-

arly those for the early Precambrian, are derived instead from Ronov's

own model of crustal evolution.

CONCLUSIONS

Application of a chemical mass balance equation to a modeled anorth-
ositic crust produces a predicted sedimentary suite of 55% shale or clay,
I to 2% carbonate-rich iron formation, 23% limestone, 1 to 2% evapor-
ites, and 20% aluminum-rich rock. This suite is unlike modern propor-
tions of sedimentary rock types, and should be recognizable if it occurs
as an ancient deposit.

46

Lithologies of the only sedimentary or metasedimentary rocks yet
discovered of apprOpriate age and tectonic setting to represent the sedi-
ments weathered from a world-wide primordial crust, the earliest sediments
in Greenland, indicate a more siliceous and more alkali primordial compo-
sition than the anorthosite modeled by Shaw.

The suggested anorthositic primordial crust is sufficiently dis-
tinct in chemical character that it would have a marked effect on the
chemistry of early sediments as it is weathered and recycled. The geo-
chemical trends predicted for the evolution of the sedimentary mass
from this distinctive initial composition to the average composition of
Precambrian sediments are not consistently observed in ancient sediments.
The data on the chemistry of Precambrian sediments are not, however,
sufficiently sensitive to accurately identify the predicted trends.

This study indicates that the choice of an anorthositic compos-
ition for a primordial world-wide crust predicts a suite of sedimentary
rock types for the earliest sediments that is not consistent with obser-
vations of the oldest known continentally derived and continentally de-
posited sediments; these indicate a more granitic primordial crustal
composition. If the predicted suite does occur, or if it has imparted
a distinctive chemical character to ancient sediments through recycling,
the sediments that would substantiate an anorthositic primordial compo-

sition have not been located, or have not been properly identified.

47

REFERENCES

REFERENCES

Allaart, J. H., 1975, Field mapping of the pre-3760 Myr. old supra-
crustal rocks of the Isua area, southern West Greenland, Grfin.

Geol. Unders. Rappt. Nr. 75, pp. 53-56

Allsopp, H. L., Ulrych, T. J., and Nicolaysen, L. 0., 1968, Dating some
significant events in the history of the Swaziland system by the

Rb-Sr isochron method, Canad. Jour. Earth Sci., v. 5, pp. 605-619

Allsopp, H. L., Roberts, H. R., and Schreiner, G. D. L., 1974, Rb-Sr age
measurments on various Swaziland granites, Jour. Geophys. Res.,

v. 67, p. 5307

Bayley, W. S., 1904, The Menominee iron-bearing district of Michigan,

U. S. Geol. Surv. Monograph 46

Blatt, H. and Jones, R., 1975, Proportions of exposed igneous, meta-
morphic and sedimentary rocks, Geol. Soc. Amer. 8ull., v. 86,

pp. 1085-1088

Bridgwater, D., Watson, J., and Windley, 8. F., 1973, The Archean craton
of the North Atlantic region, Phil. Trans. Royal Soc. London, Ser.
A, v. 273, pp. 493-512

Bridgwater, D., McGregor, V. R., and Myers, J. S., 1974, A horizontal
tectonic regime in the Archean of Greenland and its implications
for early crustal thickening, Precamb. Res., v. 1, pp. 179-197
Discussion and reply, Chadwick, B. and Coe, K., 1975, Precamb.

Res., v. 2, p. 397
Bridgwater, D., and McGregor, V. R., 1974, Field work on the very early

Precambrian rocks of the Isua area, southern West Greenland, Grdn.

Geol. Unders. Rappt. Nr. 65, pp. 49-54

49

Bryan, W. 8., Finger, L. W., and Chayes, F., 1968, Estimating propor-
tions in petrographic mixing equations by least squares approxima-

tion, Science, v. 163, pp. 926-927

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

Clarke, F. W., 1915, Analyses of rocks and minerals from the laboratory
of the United States Geological Survey, 1880-1914, U. S. Geol.
Surv. Bull. 591

Clarke, F. W., 1924, The Data of Geochemistry, U. S. Geol. Surv. Bull.
770

El Wakeel, S. K., and Riley, J. P., 1961, Chemical and mineralogical
studies of deep-sea sediments, Geochim. Cosmochim. Acta, v. 25,

pp. 110-146

Carrels, R. M., and MacKenzie, F. T., 1971, Evolution of Sedimentary

 

Rocks, New York, Norton

Carrels, R. M., MacKenzie, F. T., and Siever, R., 1972, Sedimentary cyc-
ling in relation to the history of the continents and oceans, in
Robertson, E. C. (Ed.), Nature of the Solid Earth, New York,
McGraw-Hill, pp. 93-121

 

Goodwin, A. M., 1956, Facies relations in the Gunflint iron formation,

Econ. Geol., v. 51, pp. 565-595

Grim, R. E., 1968, Clay Mineralogy, 2nd. Ed., New York, McGraw-Hill

 

Grout, F. F., 1933, Contact metamorphism of the slates of Minnesota by
granite and gabbro magmas, Geol. Soc. Amer. Bull., v. 44, pp. 989-
1040

50

Gruner, J. W., 1946, Mineralogy and Geology of the Mesabi Range, St. Paul,

Minn., Iron Range Resources and Rehabilitation Commission

Hart, S. R., and Goldich, S. S., 1975, Most ancient known rocks may be
found in all earth's Precambrian shields, Geotimes, v. 20, nr. 3,

p. 22

Horn, M. K., and Adams, J. A. S., 1966, Computer derived geochemical
balances and element abundances, Geochim. Cosmochim. Acta, v. 30,

pp. 279-297

Huber, N. K., 1959, Some aspects of the origin of the Ironwood iron form-

ation of Michigan and Wisconsin, Econ. Geol., v. 54, pp. 82-118

 

Hurley, P. M., Pinson, W. H., Nagy, 8., and Teska, T. M., 1972, Ancient
age of the Middle Marker Horizon, Onverwacht Group, Swaziland

Sequence, South Africa, Earth Planet. Sci. Ltrs., v. 14, p. 360

Hyndman, D. W., 1972, Petrology 2£_lgpeous and Metamorphic Rocks, New
York, McGraw-Hill

 

Irving, R. D., and Van Hise, C. R., 1892, The Penokee iron-bearing
series of Michigan and Wisconsin, 0. S. Geol. Surv. Monograph
19

 

James, H. L., 1954, Sedimentary facies of iron formation, Econ. Geol.,

v. 49, pp. 235-291

James, H. L., 1966, Chemistry of the iron-rich sedimentary rocks, in
Fleischer, M. (Ed.), Data of Geochemistry, 6th Ed., U. S. Geol.
Surv. Prof. Paper 440W

Jones, R. L., 1973, Determination of amounts of igneous, metamorphic and
sedimentary rocks exposed on the earth's land surface, Master's

Thesis, University of Oklahoma

51

Lafon, G. M., and MacKenzie, F. T., 1971, Early evolution of the oceans-

a weathering model, Amer. Assoc. Petrol. Geol. Bull., v. 55, p. 348

Leith, C. K., 1903, The Mesabi iron-bearing district of Minnesota, U. S.

Geol. Surv. Monograph 43

Li, Y., 1972, Geochemical mass balance among lithisphere, hydrosphere

and atmosphere, Amer. Jour. Sci., v. 272, pp. 119-137

Lowman, P. D. Jr., 1976, Crustal evolution in silicate planets; impli-

cations for the origin of continents, Jour. Geol., v. 84, p. 1

Mason, 8., 1966, Principles 2f Geochemistry, 3rd. Ed., New York, Wiley

 

McGregor, V. R., and Bridgwater, D., 1973, Field mapping of the Pre-
cambrian basement in the Godthaabsfjord district, southern West

Greenland, Grdn. Geol. Unders. Rappt. Nr. 55, pp. 29-32

Moorbath, S., O'Nions, R. K., and Pankhurst, R. J., 1973, An early Arch-
ean age for the Isua iron-formation, Nature, Phys. Sci., v. 245,

pp. 138-139

Pettijohn, F. J., 1963, Chemical composition of sandstones-- excluding
carbonate and volcanic sands, in Fleischer, M. (Ed.), Data of

Geochemistry, 6th. Ed., U. S. Geol. Surv. Prof. Paper 440S

Poldervaart, A., 1955, Chemistry of the earth's crust, in Poldervaart,
A. (Ed.), Crust of the earth, Geol. Soc. Amer. Spec. Paper 62,
pp. 119-144

Quirke, T. T., Goldich, S. S., and Krueger, H. W., 1960, Composition and

age of the Temiscamie iron formation, Mistassini Territory, Quebec,

Canada, Econ. Geol., v. 55, pp. 311-326

52

Quirke, T. T. Jr., 1961, Geology of the Temiscamie iron formation, Lake
Albanel iron range, Mistassini Territory, Quebec, Canada, Econ.

Geol., v. 56, pp. 299-320

Reilly, c. A., and Shaw, D. M., 1967, An estimate of the composition of
part of the Canadian Shield in northwestern Ontario, Canad. Jour.

Earth Sci., v. 4, p. 25

Ronov, A. 8., 1968, Probable changes in the composition of seawater dur-

ing the course of geological time, Sedimentology, v. 10, pp. 25-43

Ronov, A. 8., and Migdisov, A. A., 1971, Geochemical history of the crys-
talline basement and the sedimentary cover of the Russian and

North American platforms, Sedimentology, v. 16, pp. 137-185

Ronov, A. 8., 1972, Evolution of rock composition and geochemical pro-
cesses in the sedimentary shell of the earth, Sedimentology, v. 19,

pp. 157-172

Shaw, D. M., Reilly, G. A., Muysson, J. R., Pattenden, G. E., and Camp-
bell, F. E., 1967, An estimate of the chemical composition of the
Canadian Precambrian shield, Canad. Jour. Earth Sci., v. 4,

p. 829

Shaw, D. M., 1975, Development of the early continental crust. Part 2,
the Prearchean and Protoarchean stages, Geol. Soc. Amer. Abstr.

and Prog., v. 7, p. 857 (Abstr.)

Sibley, D. F., and Vogel, T. A., 1976, Chemical mass balance of the
earth's crust: the calcium dilemma (2) and the role of pelagic

sediment, Science, v. 192, pp. 551-553

Sibley, D. F. and Wilband, J. T., 1977, Chemical balance of the earth's

crust, Geochim. Cosmochim. Acta (In Press)

Stewart, F. H., 1963, Marine evaporites, in Fleischer, M. (Ed.), Data of

53

Geochemistry, 6th Ed., U. S. Geol. Surv. Prof. Paper 440Y

Subramanian, A. P., 1956, Mineralogy and petrology of the Sittampundi
complex, Salem district, Madras State, India, Geol. Soc. Amer.

Bull., v. 67, pp. 317-390

Turekian, K. K., 1969, The oceans, streams and atmosphere, in Wedepohl,
K. H. (Ed.), Handbook g£_Geochemistry, New York, Springer-Verlag,
v. 1, pp. 297-310

Windley, 8. F., 1969, Anorthosites of southern West Greenland, in Kay, M.
(Ed.), North Atlantic: Geology and continental drift, a symposium,

papers, Amer. Assoc. Petrol. Geol. Memoir 12, pp. 899-915

Windley, B. F., 1970, Anorthosites in the early crust of the earth and on
the Moon, Nature, v. 226, pp. 333-335

54

APPENDIX

APPENDIX.

Sedimentary suites that do not balance with an anorthositic crust.

All mass balance calculations were made with sedimentary suites
that included one or more carbonate rocks, and one or more silica or
alumina-silica rocks, i.e. shales, graywackes, clays, or sandstones.

Some calculations were made with suites that excluded evaporites. A com-
parison of the results of these calculations with the results of calcu-
lations for the same suites with evaporites included showed that the
evaporites had little effect on the determination of coefficients for the
sedimentary rock types; their presence or absence was noticable only in
the values of sodium, chlorine, and sulfate residuals.

A number of calculations were made with potassium and titanium ex-
cluded from the component balances. Comparison of the results of these
calculations with the results of recalculations of the same suites with
these two components included in the balance demonstrated that the in-
clusion or exclusion of potassium and titanium had little effect on the
estimation of the rock-type coefficients.

Seawater was included in all calculations.

No sedimentary suites that included a sandstone balanced satis-
factorally with an anorthositic crust. For suites that included both a
sandstone and a clay or shale, with one exception, the mass balance esti-
mates produced negative coefficients for the sandstone, in the range
-0.20 to -0.30, depending on the other rock types in the suite. With
respect to the mass balance equation, this means that balance was ach-
ieved by having the sandstone on the reactants or source side of the

equation rather than on the products side of the equation. The exception

56

was a suite containing a sandstone, a graywacke, and a shale. This suite
balanced with graywacke coefficients in the range of -1.48 to -0.846.
Suites that included ironstones with a sandstone and a shale produced
negative coefficients for the ironstones as well as for the sandstone.

Suites that included a graywacke and a shale, without a quartz
sandstone, balanced with negative graywacke coefficients. Balances cal-
culated with Archean graywacke and Archean shale were less satisfactory,
as the coefficients estimated for the graywacke were even more negative
than those estimated for modern graywackes. Ironstones included in shale
plus graywacke suites also had negative coefficients.

Suites calculated with shales, and with sandstones, ironstones and
graywackes excluded, generated positive coefficients for all rock types.
These balances were unsatisfactory due to high residual values, with best
values of SSR in the range 11.8 to 24.1. The results for these calcula-

tions showed high negative A120 residuals, indicating that the anortho-

3
site was rich in aluminum compared to the estimated sedimentary suite,
and high positive SiO2 residuals, which indicated a silica deficiency in
the anorthosite. When graywackes were substituted for shales in these
suites, the adverse alumina and silica residuals were accentuated.

Addition of an ironstone to suites containing a shale as the only
other silica rock generated negative ironstone coefficients, but silica
residuals were reduced.

The first reasonable mass balances were produced by adding a hypo-
thetical aluminum-rich rock to the sedimentary suites, representing an

aluminum ore of a hydroxide type. This technique did not work equally

well with all sedimentary suites. Those that included a sandstone, a

57

shale, and alumina rock generally produced negative shale coefficients.
Suites that included a shale or clay and alumina rock, and excluded a
sandstone produced positive coefficients for both the shale and the alum-
ina rock, with SSR values below 2.0. The alumina rock technique did not
work for either Archean shales or Archean graywackes, or modern gray-
wackes, or any combination of these rock types.

Ironstones in these first reasonable balances had positive or neg-
ative coefficients depending on the Choice of the type of ironstone and
the clay or shale. Dolomite usually had negative coefficients, but not
always in excess of the standard deviation for the dolomite coefficient.

Sedimentary suites with successful mass balances included a clay or
shale, alumina rock, an ironstone, limestone with or without dolomite,

. evaporites, and seawater. The results of the successful mass balance

calculations are tabulated in Tables 5 through 10.

58

MICHIGAN STATE UNIVERSITY LIBRARIES

II III ”III |

3 1293 03175 0965

 

— 7'"— ’7‘»